UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The diversity and biosynthesis of plant cuticular waxes Busta, Lucas Howard 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_november_busta_lucas.pdf [ 66.62MB ]
Metadata
JSON: 24-1.0313404.json
JSON-LD: 24-1.0313404-ld.json
RDF/XML (Pretty): 24-1.0313404-rdf.xml
RDF/JSON: 24-1.0313404-rdf.json
Turtle: 24-1.0313404-turtle.txt
N-Triples: 24-1.0313404-rdf-ntriples.txt
Original Record: 24-1.0313404-source.json
Full Text
24-1.0313404-fulltext.txt
Citation
24-1.0313404.ris

Full Text

THE DIVERSITY AND BIOSYNTHESIS OF PLANT CUTICULARWAXESbyLucas Howard BustaB.Sc., The University of Minnesota - Duluth, 2011A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THEREQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe Faculty of Graduate and Postdoctoral Studies(Chemistry)The University of British Columbia(Vancouver)August 2016© Lucas Howard Busta, 2016AbstractPlants coat themselves in a cuticle to hinder transpiration across the vast surfaceareas they require for photosynthesis. The cuticle is made of cutin, a polyester, andcuticular waxes, aliphatic compounds that form the water barrier.Cuticles of model plants had been major targets for cuticular wax research, andknowledge of their surfaces is relatively advanced. However, this knowledge still doesnot answer crucial questions about relationships between wax structure and function.Limited studies of non-model species had provided glimpses of a much greater waxchemical structural diversity than that present on model plants. These also had hintedthat diverse wax coverages and compositions exist on the surfaces of different speciesand different plant organs. Before relationships between structure and function canbe established, the major dimensions of wax diversity must be described in moredetail.To contribute, I aimed to describe wax structural and biological diversity in somemodel and non-model plant species. I examined the structural diversity of wax com-pound aliphatic tails and determined that branched compounds on Arabidopsis leavesare iso-branched and that certain wax biosynthesis enzymes can exhibit bias towardsor against branched substrates. I also studied functional group diversity by perform-ing a comprehensive literature search to codify our knowledge of wax compounds withsecondary functions.I furthered our knowledge of wax coverage and composition on diverse biologicalsurfaces by determining the structures of novel wax compounds from the moss Fu-naria hygrometrica and profiling the waxes from multiple F. hygrometrica surfaces toreveal that these moss cuticles have some similarities to those of flowering plants. Bystudying developing Arabidopsis leaves I found that their wax composition, but notcoverage, is dynamic with time, pointing to functional optimization and synchronousiiAbstractcell expansion and wax production.This work highlights the importance of chain length specificity in wax biosynthesis,though the mechanisms by which such is achieved are unclear. It also confirms thatsecondary functional groups on wax molecules are installed by a variety of processes,that these are connected with biosynthetic chain length specificity and that bothlikely influence the physical and water barrier properties of the cuticle.iiiPrefaceI designed and performed all experiments, measurements, data processing, and,with Reinhard Jetter, the writing that led to the contents of Chapter 2.I performed all literature searching, cataloging, figure preparation, and, with Rein-hard Jetter, the writing of the text in Chapter 3.Jessica Budke grew and extracted waxes from the moss surfaces described inChapters 4 and 5. I performed all GC analyses, syntheses, data processing, andplotting. Together Jessica Budke, Reinhard Jetter and I wrote the contents of thesechapters. Both have been published:• Lucas Busta, Jessica M. Budke, Reinhard Jetter. Identification of β-hydroxyfatty acid esters and primary,secondary alkanediol esters in cuticular waxes ofthe moss Funaria hygrometrica. Phytochemistry, 122:38-49, 2016.• Lucas Busta, Jessica M. Budke, Reinhard Jetter. The moss Funaria hygro-metrica has cuticular wax similar to vascular plants, with distinct compositionon leafy gametophyte, calyptra, and sporophyte capsule surfaces. Annals ofBotany, in press.The morphological data in Chapter 6 were collected by Yan Cao. She also grewthe plants, prepared the wax samples, and set up the GC runs that led to the datain Figures 6.3 and 6.4. Reinhard Jetter and I performed the other steps leading tothe figures containing morphological and chemical data including data extraction,processing, normalization, and plotting. Reinhard Jetter designed and I carried outthe steps leading to the chemical data in Figure 6.6. Daniela Hegebarth performed allgene expression measurements and prepared the figure and supplementary tables thatcontain the gene expression data and associated information. Edward Kroc designedivPrefaceand performed all the statistical tests. Together Daniela Hegebarth, Edward Kroc,Reinhard Jetter and I wrote the contents of the chapter.vTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiiList of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviiAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xx1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 The plant cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Plant cuticle research . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.2.1 Cuticular wax diversity . . . . . . . . . . . . . . . . . . . . . . 71.3 Goals of this work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Branched compounds from Arabidopsis thaliana cuticular waxes 132.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202.2.1 Structure of branched aliphatic wax compounds on Arabidopsissurfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202.2.2 Accumulation of branched wax compounds in wax biosynthesismutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34viTable of Contents2.3.1 Generation of branched chain wax precursors . . . . . . . . . . 352.3.2 Modification of branched wax compounds by wax biosynthesisenzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392.5 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402.5.1 Plant materials, wax extraction, and GC analysis . . . . . . . 402.5.2 Transformation of TLC-purified primary alcohols . . . . . . . 402.5.3 Synthesis of authentic standards . . . . . . . . . . . . . . . . . 412.6 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . 443 The diversity and biosynthesis of specialty compounds in plantcuticular waxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463.2 Occurrence and diversity of specialty wax compounds . . . . . . . . . 513.2.1 Specialty compounds with a single secondary oxygen functionalgroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563.2.2 Specialty compounds with two secondary functional groups . . 613.2.3 Specialty compounds with three or more secondary functionalgroups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 663.3 Biosynthesis of specialty wax compounds . . . . . . . . . . . . . . . . 663.3.1 Specialty compound classes with secondary functional groupson carbons with mixed parity . . . . . . . . . . . . . . . . . . 673.3.2 Specialty compounds with functional groups on carbons of sin-gle parity near the chain terminus . . . . . . . . . . . . . . . . 693.3.3 Compound classes with a single secondary functional groupnear the middle of the chain . . . . . . . . . . . . . . . . . . . 713.3.4 Compound classes with two secondary functional groups on car-bons of single parity near the middle of the chain . . . . . . . 753.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 823.5 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . 86viiTable of Contents4 Identification of β-hydroxy fatty acid esters and pri-mary,secondary-alkanediol esters in cuticular waxes of themoss Funaria hygrometrica . . . . . . . . . . . . . . . . . . . . . . . . 874.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 874.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 914.2.1 Identification and quantification of unknown series A . . . . . 914.2.2 Identification and relative quantification of unknown series B . 964.2.3 Identification of other bifunctional wax components . . . . . . 1014.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1014.3.1 Potential pathway leading to β-hydroxy fatty acid (FA) esters 1024.3.2 Potential pathway leading to 1,X-alkanediol esters . . . . . . . 1054.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1084.5 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1094.5.1 Growth conditions, wax extraction, purification, and transes-terification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1094.5.2 Derivatization and gas chromatography - mass spectrometry(GC-MS) analysis . . . . . . . . . . . . . . . . . . . . . . . . . 1104.5.3 Synthesis of authentic standards . . . . . . . . . . . . . . . . . 1104.6 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . 1135 The moss Funaria hygrometrica has cuticular wax similar to vas-cular plants, with distinct composition on leafy gametophyte, ca-lyptra, and sporophyte capsule surfaces . . . . . . . . . . . . . . . . 1155.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1155.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1185.2.1 Total wax amounts, compound class and chain length distribu-tions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1185.2.2 Ester isomer distributions . . . . . . . . . . . . . . . . . . . . 1225.2.3 Hydroxy ester isomer distributions . . . . . . . . . . . . . . . 1255.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1295.3.1 Cuticular wax coverage on Funaria hygrometrica . . . . . . . . 1305.3.2 Cuticular wax constituents common to all surfaces of Funariahygrometrica . . . . . . . . . . . . . . . . . . . . . . . . . . . 131viiiTable of Contents5.3.3 Distinguishing features in the wax mixtures of the three F. hy-grometrica structures . . . . . . . . . . . . . . . . . . . . . . . 1335.3.4 Isomer patterns in different wax ester classes on the three Fu-naria hygrometrica structures . . . . . . . . . . . . . . . . . . 1355.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1365.5 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1385.5.1 Moss growth conditions . . . . . . . . . . . . . . . . . . . . . 1385.5.2 Surface area measurement and wax extraction . . . . . . . . . 1395.5.3 Wax derivatization and gas chromatography (GC) conditions . 1395.5.4 Wax quantification and ester analysis . . . . . . . . . . . . . . 1405.5.5 Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . 1405.6 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . 1406 Changes in cuticular wax coverage and composition on develop-ing Arabidopsis leaves are influenced by wax biosynthesis geneexpression levels and trichome density . . . . . . . . . . . . . . . . . 1426.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1426.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1456.2.1 Morphological changes on developing leaves . . . . . . . . . . 1456.2.2 Cuticular waxes from whole Arabidopsis leaves of different age 1476.2.3 Regional distribution of wax on Arabidopsis leaves . . . . . . 1516.2.4 Expression of wax biosynthesis genes during leaf development 1546.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1576.3.1 Pavement cell age effects on wax composition . . . . . . . . . 1586.3.2 Leaf expansion effects on wax coverage . . . . . . . . . . . . . 1616.3.3 Epidermal cell type effects on wax composition . . . . . . . . 1626.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1646.5 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1656.5.1 Plant material, growth conditions, and leaf harvesting . . . . . 1656.5.2 Leaf morphological analysis . . . . . . . . . . . . . . . . . . . 1656.5.3 Wax sample preparation and GC analysis . . . . . . . . . . . 1666.5.4 RNA extraction and gene expression analysis by quantitativeRT-PCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167ixTable of Contents6.5.5 Adjustment for multiple comparisons . . . . . . . . . . . . . . 1686.6 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . 1687 Conclusions and future directions . . . . . . . . . . . . . . . . . . . . 1707.1 Substrate and product chain length profile specificity in wax biosynthesis1727.2 Installation of secondary functional groups in specialty wax compoundbiosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1747.3 Influence of chain length distributions and secondary functional groupson the physical properties of wax mixtures . . . . . . . . . . . . . . . 175Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178A Supplementary data for chapter 2 . . . . . . . . . . . . . . . . . . . . 208B Supplementary data for chapter 3 . . . . . . . . . . . . . . . . . . . . 223C Supplementary data for chapter 4 . . . . . . . . . . . . . . . . . . . . 227D Supplementary data for chapter 5 . . . . . . . . . . . . . . . . . . . . 234E Supplementary data for chapter 6 . . . . . . . . . . . . . . . . . . . . 256xList of TablesA.1 Wax composition on flowers of wild-type and mutant Arabidopsis lines. 213A.2 Wax composition on flowers of wild-type and mutant Arabidopsis lines. 215A.3 Wax composition on leaves of wild-type and mutant Arabidopsis lines. 218A.4 Wax composition on leaves of wild-type and mutant Arabidopsis lines. 220B.1 Species with specialty wax compounds. . . . . . . . . . . . . . . . . . 223D.1 Wax coverage on three F. hygrometrica organs. . . . . . . . . . . . . 235D.2 Amount of esterified fatty acids in each alkyl ester homolog on theleafy gametophyte of F. hygrometrica. . . . . . . . . . . . . . . . . . . 237D.3 Amount of esterified alcohols in each alkyl ester homolog on the leafygametophyte of F. hygrometrica. . . . . . . . . . . . . . . . . . . . . . 239D.4 Amount of esterified fatty acids in each alkyl ester homolog on thegametophyte calyptra of F. hygrometrica. . . . . . . . . . . . . . . . . 240D.5 Amount of esterified alcohols in each alkyl ester homolog on the game-tophyte calyptra of F. hygrometrica. . . . . . . . . . . . . . . . . . . . 241D.6 Amount of esterified fatty acids in each alkyl ester homolog on thesporophyte capsule of F. hygrometrica. . . . . . . . . . . . . . . . . . 244D.7 Amount of esterified alcohols in each alkyl ester homolog on the sporo-phyte capsule of F. hygrometrica. . . . . . . . . . . . . . . . . . . . . 246D.8 Amount of esterified β-hydroxy fatty acids in each β-hydroxy esterhomolog on the leafy gametophyte of F. hygrometrica. . . . . . . . . . 248D.9 Amount of esterified alcohols in each β-hydroxy ester homolog on theleafy gametophyte of F. hygrometrica. . . . . . . . . . . . . . . . . . . 249D.10 Amount of esterified fatty acids in each β-hydroxy ester homolog onthe gametophyte calyptra of F. hygrometrica. . . . . . . . . . . . . . 251xiList of TablesD.11 Amount of esterified alcohols in each β-hydroxy ester homolog on thegametophyte calyptra of F. hygrometrica. . . . . . . . . . . . . . . . . 253E.1 Wax composition on gl1 eighth leaves. . . . . . . . . . . . . . . . . . 259E.2 Wax composition on wild-type eighth leaves. . . . . . . . . . . . . . . 260E.3 Wax composition on bases and tips of wild-type eighth leaves. . . . . 261E.4 Details of chapter-wide statistical testing . . . . . . . . . . . . . . . . 265E.5 Statistical analysis of the gene expression data . . . . . . . . . . . . . 266xiiList of Figures1.1 Plant cuticle layer structure. . . . . . . . . . . . . . . . . . . . . . . . 21.2 Plant cuticle biogenesis. . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Biosynthesis of cuticular wax compounds. . . . . . . . . . . . . . . . 62.1 Biosynthesis of wax alkanes and alcohols using different starter molecules. 152.2 A. thaliana branched alkane structure elucidation. . . . . . . . . . . . 212.3 A. thaliana branched alcohol structure elucidation. . . . . . . . . . . 232.4 Reduction of A. thaliana primary alcohols. . . . . . . . . . . . . . . . 252.5 Synthesis of C24 aldehydes with and without 2-methyl branching. . . 272.6 Oxidation of A. thaliana primary alcohols. . . . . . . . . . . . . . . . 292.7 Relative abundance of n- and iso-branched alcohols and alkanes inArabidopsis flower and leaf waxes. . . . . . . . . . . . . . . . . . . . . 302.8 Differences in flower wax composition between Arabidopsis mutantsand corresponding wild type. . . . . . . . . . . . . . . . . . . . . . . . 312.9 Differences in leaf wax composition between Arabidopsis mutants andcorresponding wild type. . . . . . . . . . . . . . . . . . . . . . . . . . 332.10 Biosynthesis of n- and iso-branched wax compounds. . . . . . . . . . 383.1 Structure and biosynthesis of ubiquitous wax compounds. . . . . . . . 473.2 Tabulation of major ubiquitous wax compounds from A. thaliana andB. oleracea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483.3 Catalog of example specialty wax compounds. . . . . . . . . . . . . . 533.4 Color coding and nomenclature for specialty wax compounds. . . . . 543.5 Catalog of specialty wax compounds with one secondary functionalgroup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57xiiiList of Figures3.6 Catalog of specialty wax compounds with two secondary functionalgroups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 623.7 Catalog of specialty compounds with three secondary functional groups. 653.8 Biosynthesis of specialty wax compounds with functional groups oncarbons of mixed parity. . . . . . . . . . . . . . . . . . . . . . . . . . 683.9 Biosynthesis of specialty compounds by intercept modification. . . . . 703.10 Biosynthesis of specialty wax compounds by intercept reentry. . . . . 733.11 Biosynthesis of specialty wax compounds by intercept condensation. . 763.12 Competition for elongation pathway intermediates and products. . . . 793.13 Relative abundance of products derived from elongation intermediateintercept. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 813.14 Chain lengths of specialty compounds relative to co-occurring ubiqui-tous compounds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 833.15 Distribution of specialty wax compounds across plant orders. . . . . . 854.1 Identification of unknown series A found in the Funaria hygrometricacalyptra and leafy gametophyte waxes. . . . . . . . . . . . . . . . . . 934.2 Synthesis of a C46 β-hydroxy FA ester (docosyl 3-hydroxytetracosanoate). 954.3 Quantification of hydroxy ester homologs in Funaria hygrometrica waxes. 964.4 Identification of unknown series B found in the Funaria hygrometricasporophyte capsule wax. . . . . . . . . . . . . . . . . . . . . . . . . . 974.5 Synthesis of 7-hydroxytriacontyl palmitate. . . . . . . . . . . . . . . . 994.6 Relative abundances of homologous diol esters in Funaria hygrometricasporophyte capsule wax. . . . . . . . . . . . . . . . . . . . . . . . . . 1004.7 Biosynthesis model for the hydroxy esters found in Funaria hygromet-rica waxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1075.1 Wax coverage on three major structures of Funaria hygrometrica. . . 1195.2 Relative abundance of wax compounds on three major structures ofFunaria hygrometrica. . . . . . . . . . . . . . . . . . . . . . . . . . . 1205.3 Isomer distribution within the FA alkyl esters in Funaria hygrometricawaxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123xivList of Figures5.4 Relative abundance of FA alkyl ester-bound fatty acids and alcoholsin Funaria hygrometrica waxes. . . . . . . . . . . . . . . . . . . . . . 1265.5 Relative abundance of ester-bound β-hydroxy fatty acids and alcoholsin Funaria hygrometrica waxes. . . . . . . . . . . . . . . . . . . . . . 1275.6 Isomer distribution within the β-hydroxy fatty acid esters in Funariahygrometrica waxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1285.7 Isomer distribution within diol esters in Funaria hygrometrica sporo-phyte wax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1305.8 Comparison of ester-bound and free alkyl and acyl chains in Funariahygrometrica waxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1376.1 Surface area and epidermal cell numbers on developing wild-type Ara-bidopsis eighth leaves. . . . . . . . . . . . . . . . . . . . . . . . . . . 1466.2 Wax coverage on developing Arabidopsis gl1 and wild-type eighthrosette leaves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1486.3 Wax composition on developing Arabidopsis gl1 mutant eighth rosetteleaves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1506.4 Wax composition on developing Arabidopsis wild-type eighth rosetteleaves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1526.5 Pavement cell size on segments of developing Arabidopsis wild-typeeighth rosette leaves. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1546.6 Wax composition on the base and tip sections of wild-type Arabidopsiseighth rosette leaves at 13 days of age. . . . . . . . . . . . . . . . . . 1556.7 Expression of wax biosynthesis genes in developing wild-type Arabidop-sis eighth leaves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156A.1 Mass spectrum of compound tetracosanal (2.1) . . . . . . . . . . . . 209A.2 Mass spectrum of 2-methyloctacos-4-ene (2.3) . . . . . . . . . . . . . 209A.3 Mass spectrum of 1,22-docosandiol (2.5) . . . . . . . . . . . . . . . . 210A.4 Mass spectrum of docosandial (2.6) . . . . . . . . . . . . . . . . . . . 210A.5 Mass spectrum of (((29-methyltriaconta-3,25-dien-1-yl)oxy)methyl)benzene (2.8) . . . . . . . . . . . . . . . . . . . . . . . 211A.6 Mass spectrum of docosanal (2.10) . . . . . . . . . . . . . . . . . . . 211xvList of FiguresA.7 Mass spectrum of compound tricosan-2-ol (2.11) . . . . . . . . . . . 212A.8 Mass spectrum of 2-bromotricosane (2.12) . . . . . . . . . . . . . . . 212C.1 Relative quantification of methanolysis products obtained from TLCpurification of unknown series A. . . . . . . . . . . . . . . . . . . . . 228C.2 Potential paths to the fragments observed in the mass spectra of C46β-hydroxy fatty acid esters. . . . . . . . . . . . . . . . . . . . . . . . 228C.3 The mass spectrum of docosyl 2-bromoacetate (3.2) and potentialsources of observed characteristic fragments. . . . . . . . . . . . . . . 229C.4 Mass spectra and possible fragmentation of aliphatic aldehydes . . . . 230C.5 Potential paths to the fragments observed in the mass spectra of C46β-hydroxy fatty acid esters. . . . . . . . . . . . . . . . . . . . . . . . 231C.6 Mass spectrum and fragmentation of 2-((6-bromohexyl)oxy)tetrahydro-2H-pyran (3.7) . . . . . . . . . . . . . . . 231C.7 The mass spectrum of tetracosanol (3.10) and potential sources ofobserved major fragments. . . . . . . . . . . . . . . . . . . . . . . . . 232C.8 Identification of octacosane-1,3-diol in the leafy gametophyte and ga-metophyte calyptra waxes of F. hygrometrica. . . . . . . . . . . . . . 232C.9 Identification of triacontane-1,7-diol in the sporophyte capsule of F.hygrometrica. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233E.1 Morphological data for developing wild-type Arabidopsis eighth leaves. 256E.2 Sampling scheme for studying cell size distributions across sections ofArabidopsis wild-type leaves of different ages. . . . . . . . . . . . . . 257E.3 Calculation of wax production rates. . . . . . . . . . . . . . . . . . . 258E.4 Calculated trichome wax composition. . . . . . . . . . . . . . . . . . 262E.5 Calculated trichome wax composition. . . . . . . . . . . . . . . . . . 263E.6 List of primer sequences used in the qRT-PCR analysis. . . . . . . . . 264xviList of AbbreviationsCC silica gel column chromatographyCHCl3 chloroformCoA coenzyme ACH2Cl2 dichloromethaneDMAP dimethylaminopyridineECR enoylacyl-CoA reductaseEI electron impactER endoplasmic reticulumEt2O diethyl etherFA fatty acidFAE fatty acid elongaseFAR fatty acyl-CoA reductaseFID flame ionization detectionGC gas chromatographyGC-FID gas chromatography - flame ionization detectionxviiList of AbbreviationsGC-MS gas chromatography - mass spectrometryHBr hydrobromic acidHCD hydroxyacyl-CoA dehydrataseHCl hydrocloric acidKCR ketoacyl-CoA reductaseKCS ketoacyl-CoA synthaseLACS long-chain acyl-CoA synthaseLCFA long-chain fatty acidMAH midchain alkane hydroxylaseMeOH methanolMS mass spectrometryN2 nitrogenNaOH sodium hydroxideNMR nuclear magnetic resonancePBr3 phosphorus tribromidePCC phyridinium chlorochromatePKS polyketide synthaseRT room temperatureSEM scanning electron microscopyTCN total carbon numberTEM transmission electron microscopyTFA trifluoroacetic acidxviiiList of AbbreviationsTHF tetrahydrofuranTLC thin layer chromatographyTMS trimethyl silylUV ultra-violetVLC very-long-chainxixAcknowledgementsThe assistance of S. Komar and E. Magat with extraction of F. hygrometricaleafy gametophytes is recognized with appreciation, as is the technical assistanceand advice of Chen Peng on the diol ester synthesis documented in Chapter 4. Ialso wish to draw attention to the morphometric analyses carried out by Yan Cao,the RT-PCR measurements skillfully made by Daniela Hegebarth, and the statisticalanalyses performed by Edward Kroc that are presented as part of Chapter 6. I thankthese individuals for their patience while working with me.I wish to acknowledge the members of my supervisory committee: Dr. RaymondAndersen, Dr. Ed Grant, and Dr. Robin Turner, each of whom has graciously setaside time to meet with me, helped improve my work, and challenged me as a youngscientist.I also want to thank Jessica Budke, who has helped me mature as a scientistboth academically and personally.Finally, I offer my sincere and eternal gratitude to both my supervisor ReinhardJetter and my wife Erica. They have aided me in diverse and uncountable ways in thelast five years regardless of whether it was easy or difficult for them to do so. Withouttheir assistance, guidance, and company this work would not have been completed.xxChapter 1Introduction1.1 The plant cuticlePlants are a key ecosystem component that enable animal life by capturing theenergy of the sun through photosynthesis and storing that energy in chemical form.As sessile organisms, plants lack the ability to respond to external stresses throughmovement, but they have developed elaborate chemical mechanisms that enable themto survive in aquatic and terrestrial environments. On land, all organisms face chal-lenges related to the acquisition, retention, and use of water. Accordingly, plantsuse a chemical strategy to solve this problem; their aerial surfaces are covered witha lipophilic layer that facilitates internal water transfer and thus absorption of waterthrough the roots, and deters water loss across the vast surfaces that plants mustcreate with which to utilize incident sunlight for photosynthesis.The protective hydrophobic coating on plant aerial surfaces is called the cuticle.It is a layered structure with two components that is built on the exterior of the cellwall carbohydrates (Fig. 1.1). A layer of the structural polyester cutin is linked tothe cell wall and serves as a molecular scaffold within and on top of which cuticularwaxes accumulate as a film and sometimes in crystalline form. Experimental evidenceindicates that cutin influences the biomechanical properties of the cuticle and thatthe waxes provide transpiration control [187, 92]. Thus, both are crucial functionalcomponents of plants and a fundamental understanding of the biochemical processesleading to these structures is warranted.11. IntroductionCuticular wax [SEALANT]Cutin [SCAFFOLD]Cell wall carbohydratesFigure 1.1: Plant cuticle layer structure. Cutin (orange) and cuticular wax(green) comprise the cuticle. Cutin is deposited on top of the cell wall carbohydrates(purple). Within and on top of the cutin scaffold the cuticular wax accumulates andseals the surface to hinder transpirational water loss. Cuticular waxes are present asa wax film (green shading) and sometimes in crystalline form (green rectangles at thetop of the diagram).1.2 Plant cuticle researchThe use of analytical and organic chemistry techniques in combination with molec-ular biology and genetics had been demonstrated to be effective means by which tostudy the fundamentals of the plant cuticle. Isolation of wax and cutin compoundswith extraction and (in the case of cutin) depolymerization had yielded componentmixtures that could be separated with thin layer chromatography (TLC) and gas chro-matography (GC). For cutin and wax compound quantification GC had frequentlybeen coupled with flame ionization detection (FID), while identification of separatedcuticle compounds had typically been accomplished with electron impact (EI)-massspectrometry (MS).To elucidate structures of unknown cuticle compounds, derivatization reactionshad been used to great effect. For example, transamination of keto groups, theconversion of hydroxyl groups into trimethyl silyl (TMS) ethers, or the methanolysis ofester linkages had been shown to provide the information necessary to unambiguouslyassign structures to unknown cuticle compounds [102, 140]. In some cases candidate21. Introductionstructures had been hypothesized from the mass spectra of unknown compounds andstandards had then been synthesized to confirm the hypotheses [170]. Due to thevery low abundance of cuticle compounds and the complex mixtures in which theyare found NMR techniques are seldom used to investigate their structures.Complementary to the chemical techniques used to identify the molecular speciesthat comprise the cuticle are the molecular biological and genetic techniques that hadfrequently been used to generate mutant plants or create in vitro systems with whichto investigate the roles of individual genes or enzymes.Since these biological investigations are facilitated by the use of plant species withsequenced genomes, short life cycles, and small space requirements, the plant speciesthat had been studied most frequently were model species. In particular, Arabidop-sis thaliana had been studied by combining the described chemical and biologicaltechniques, which had led to a relatively detailed understanding of cutin and waxbiosynthesis in this species.In Arabidopsis, both cutin and waxes are products derived from C16 and C18 fattyacids (called long-chain fatty acids (LCFAs)) [252]. In the plastid organelles, LCFAsare biosynthesized de novo via acyl-ACP intermediates (Fig. 1.2). Through variousP450-catalyzed oxidation reactions LCFAs are converted into the variety of hydroxyfatty acids, dihydroxy fatty acids, dioic fatty acids, and hydroxy dioic fatty acids(collectively (di)hydroxy (dioic) acids) that comprise the monomers used in cutinpolymerization. For cuticular wax biosynthesis LCFAs are transferred to the endo-plasmic reticulum where they are converted into long-chain fatty acyl-CoAs. Theseare substrates for the elongation pathway in which fatty acid elongase (FAE) com-plexes generate very-long-chain (VLC) (> C20) acyl-CoAs [185]. In turn, these VLCacyl-CoAs are the substrates for the modification pathways: the acyl reduction path-way, the decarbonylation pathway, and the fatty acid-forming pathway that generatewax compounds.Many of the enzymes on the elongation and modification pathways in A. thalianahad been characterized. Four reactions comprise the elongation pathway, first a sub-strate acyl-CoA enters a FAE enzyme complex where it is condensed with malonyl-CoA by a ketoacyl-CoA synthase (KCS) enzyme to produce a β-ketoacyl-CoA(Fig. 1.3) [212]. This β-functional compound is then substrate for the next enzyme31.IntroductionPlastidsCytosolFatty Acid De Novo SynthesisEndoplasmic ReticulumPlasma MembraneCuticle = cutin + waxFatty acyl-ACPLCFAsLong chain acyl-CoAsVLC acyl-CoAsVLC estersVLC alcoholsVLC aldehydesVLC alkanesFatty Acid ElongationAcyl reduction pathwayDecarbonylation pathwayCutin MatrixCell Wall MatrixIntracuticular WaxEpicuticular Wax FilmEpicuticular Wax CrystalsVLC fatty acidsFatty acid-forming pathwayFigure 1.2: Plant cuticle biogenesis. Free C16 fatty acids are biosynthesized de novo in the plastids (green chamber)and are then transferred to the endoplasmic reticulum (yellow chamber) and converted into C16 fatty acyl-CoAs. Thefatty acid elongation pathway uses these to generate very-long-chain (VLC) acyl-CoAs. These VLC acyl-CoAs are usedin the acyl reduction, decarbonylation, and fatty acid-forming pathways to form VLC alcohols and esters, VLC aldehydesand alkanes, and VLC fatty acids, respectively. These wax compounds are transported outside the plasma membrane(dark blue) and cell wall (blue matrix) and are incorporated into the cuticle.41. Introductionin the complex, a ketoacyl-CoA reductase (KCR) which reduces the β-ketoacyl-CoA,generating a β-hydroxyacyl-CoA [17]. This compound is passed to a hydroxyacyl-CoAdehydratase (HCD) that catalyzes the elimination reaction that leads to an unsatu-rated enoylacyl-CoA [7]. In turn, the enoylacyl-CoA is reduced by an enoylacyl-CoAreductase (ECR) enzyme to generate a saturated acyl-CoA two carbons longer thanthat which entered the condensation enzyme [256]. This sequence of reactions canhappen repeatedly such that an acyl-CoA emerging from the last of the four reac-tions can be acted upon again by a KCS enzyme, thereby initiating another roundof elongation in which the substrate will pass through each enzyme of the FAE com-plex in turn. The number of rounds of elongation that each acyl-CoA can undergo iscontrolled by the KCS, which determines the chain length of the products the FAEcomplex can generate [141].Many KCS genes, each with different chain length specificities, are present inthe A. thaliana genome [104]. When more than one of these is expressed simul-taneously, multiple KCS enzymes are present, giving rise to multiple types of FAEcomplexes that are differentiated by the KCS they contain. Each of these distinctFAE complexes is part of the elongation pathway. Since the KCS is the primarychain length-determining part of each FAE, then each of these complexes generatesa pool of acyl-CoAs with different chain length distributions [141, 213]. Since thesubstrates for elongation have even total carbon numbers (TCNs), and KCS action ineach elongation round adds a C2 unit, product acyl-CoAs have predominantly evenTCNs. Thus, the elongation pathway generates a homologous series of VLC acyl-CoAs with even TCNs with a chain length profile that is the sum of the individualprofiles generated by each FAE complex.The VLC acyl-CoAs, or wax precursors, generated by the elongation pathwayare substrates for the acyl-CoA modification pathways. One modification pathway,the decarbonylation pathway, consists of a reductase enzyme (RED) that generatesaldehydes and an aldehyde decarbonylase enzyme (AD) that can produce alkanes(Fig. 1.3) [24]. Another, the acyl reduction pathway, consists of a fatty acyl-CoAreductase enzyme (FAR) that generates primary alcohols [182] that are then availableto a wax ester synthase enzyme (WS) for esterification with long-chain acyl-CoAs toproduce alkyl esters [124]. Fatty acids are found as a major components in Arabidopsis51. IntroductionOSCoAOSCoAOOSCoAOHOSCoAOSCoAKCSKCRHCDECRADWSOOHOOREDFARESTOOHC18 fatty acyl-CoAModification pathway products(ubiquitous wax compounds)Elongation pathway products(wax precursors)Elongation intermediatesOSCoARe-entry into elongation cycleEntry into elongation cyclennnnnnnnnnnnFigure 1.3: Biosynthesis of cuticular wax compounds. Fatty acyl-CoAs enterinto fatty acid elongation (magenta pathway) where they are elongated and then serveas wax precursors (dashed box). These precursors can reenter the elongation cycle togenerate longer wax precursors, or can be passed to the modification pathways (bluepathways) to produce wax compounds (compounds with grey background). Solidarrows represent enzymatic steps and dashed arrows indicate transfer of compoundsbetween enzymes and/or enzyme complexes.leaf wax mixtures, so there is likely also a fatty acid-forming acyl-CoA modificationpathway, though an enzyme that might catalyze fatty acid production, such as anesterase (EST), has not yet been identified.From the homologous series of acyl-CoAs generated by the elongation pathway,the modification pathways produce different classes of wax compounds that are alsopresent as homologous series. Each of these compound classes has a defining ho-molog distribution, for example, a distribution of alcohols ranging from C26 to C34that peaks in abundance at the C28 homolog, or a Gaussian distribution of alkanesranging from C27 to C35 that peaks at the C31 homolog. The exact characteristics ofthese distributions are determined by the chain length-specifying enzymes of the FAE61. Introductioncomplexes, primarily the KCSs. Since every round of elongation adds two carbons tosubstrate acyl-CoAs with even TCNs, wax compounds are generally present in ho-mologous series whose members differ in length by two carbons and have even TCNs.However, since the activity of the aldehyde decarbonylase enyzme generates alkanesvia the removal of a carbon atom from homologous substrate aldehydes, membersof homologous series of alkanes also differ in length by two carbons, but have oddTCNs. Thus, each compound class can be described according to the distributionof homologs it contains, the parity (even or odd) of the TCNs of its constituents,and the oxidation state of the compounds’ head groups (if present). Together, thedescriptions of each compound class describe the composition of a wax mixture froma given surface.1.2.1 Cuticular wax diversityDespite our relatively advanced understanding of wax compound structure andbiosynthesis in Arabidopsis, crucial questions remain unanswered about how chemi-cal composition determines the physical properties of the cuticular wax mixture andthus the characteristics of the transpiration barrier. Before these questions can beaddressed, a thorough understanding of the major dimensions of cuticular wax diver-sity is needed. In particular, the diversity in the chemical structures and diversityin the coverage and composition of cuticular waxes on different biological surfaceswill enable comparisons that will move us closer to a fundamental understanding ofthe relationships between structure and function in the plant cuticle. Some work hasbeen done that begins to outline the bounds of cuticular wax structural and biologicaldiversity. This section will describe this work and how it has advanced our under-standing but will also highlight its limits and certain questions that remain about thestructural diversity of plant cuticular wax compounds and the biological variabilityin wax coverage and composition.Wax compound structural diversitySince wax compounds contain aliphatic tails and oxygen-containing functionalgroups, structural diversity may be exhibited in the aliphatic tail (e.g., TCN, numberof unsaturations, aliphatic branches, etc.), or in the functional group(s) (e.g., oxida-71. Introductiontion state, position on the aliphatic tail, number of functional groups, etc.). Thoughthe compound classes that had been most commonly reported from cuticular waxesare homologous series of monofunctional compounds with linear aliphatic chains [99],there are a few reports that had noted the presence of wax compounds with modifi-cations in their aliphatic tails such as branches or unsaturations [10, 240]. Some ofthese had been found in high abundance, indicating that they likely play an importantrole in defining the physical properties of the wax layer and thus the functionalityof the cuticular wax mixture. Microscale derivatization and mass spectrometry hadbeen combined to determine the structures of some of these compounds [10], andhypotheses for how the branches are introduced into the aliphatic chains of wax pre-cursors had been developed based on experimentation [114, 240]. However, it is notclear what wax biosynthesis machinery is responsible for converting branched waxprecursors into branched wax compounds and if it is related to that which producesunbranched compounds. To resolve this uncertainty the analysis of a model specieswith characterized biosynthetic pathways and branched wax compounds is necessary.Authors had noted highly abundant Arabidopsis wax compounds that bearbranches in their aliphatic chains [28, 37, 176]. However, the structure and biosyn-thesis of these compounds had remained unclear, probably because of the difficultiesassociated with elucidating the structure of compounds that are members of a com-plex mixture that are difficult to purify and obtain in even milligram quantities.Accordingly, study of Arabidopsis branched alcohols can further our understandingof not only wax compound aliphatic tail structural diversity but also of how thesecompounds might be biosynthesized.Structural diversity may also be exhibited by the functional groups of waxmolecules. Fatty acids, alcohols, wax esters, aldehydes, and alkanes are ubiquitouswax compounds, comprising the wax mixtures of nearly all plant species investi-gated to date [99], however, published wax profiles from diverse species also revealthat, in addition to accumulating the ubiquitous wax compounds, some plants ac-cumulate specialty wax compounds that contain secondary functional groups. Forexample, many gymnosperm species accumulate considerable amounts of secondaryalcohols and diols in addition to ubiquitous wax compounds [63, 77, 246], ketones,ketoalcohols, and ketoaldehydes accumulate on the surfaces of the fern Osmunda re-81. Introductiongalis [103], and β-diketones on the surfaces of wheat and barley [93, 216]. Thus,while these species all produce ubiquitous wax compounds, the true diversity of cu-ticular wax composition is much greater than that produced by the ubiquitous waxcompound-producing pathways elucidated from Arabidopsis leaves. Efforts to catalogthe specialty wax compounds that create this further diversity had been made [99],and had indicated that in addition to the dimensions of structural diversity exhibitedby ubiquitous wax compounds described above (aliphatic tail diversity, head groupoxidation state) the presence of secondary functional groups on wax compounds incertain species provide further diversity in the form of isomerism and secondary func-tional group oxidation state. However, no attempt has been made to comprehensivelycatalog these compounds, nor has a systematic analysis of possible biosynthetic stepsleading to these compounds been performed. Thus, while there are relatively fewgaps in our knowledge of ubiquitous wax compound biosynthesis, our understandingof both the structural diversity and biosynthesis of specialty wax compounds is stillin nascent stages.Biological variability in wax coverage and compositionDiverse cuticular wax mixtures are found on different biological surfaces. For ex-ample, highly abundant esters on the dicot Camelina sativa [173] indicate that boththe reduction and esterification steps of the acyl reduction pathway are probablyhighly active in this species, while abundant alcohols on the monocot Hordeum vul-gare indicate an active reduction step, but no esterification. In contrast, the highlyabundant VLC alkanes on the surfaces of the monocot Aloe arborescens indicate ahighly active decarbonylation pathway [170]. This demonstrates that there is waxcompositional variability between species of different lineage. However, while thereare many reports on the wax coverage and composition of angiosperm species, thereare relatively few detailing that of gymnosperms, ferns, and mosses [31]. These presentmainly microscopic data and reveal the presence of characteristic surface ornamenta-tions [113, 32, 183], but chemical information is largely missing. Accordingly, whetherspecies in these plant groups have wax compositions and coverages similar to thoseof vascular plants is unclear, and our knowledge of the biological variability of waxcoverage and composition on the species within these groups is limited. Targeted91. Introductionanalyses of these species will advance our fundamental understanding of the cuti-cle, and, as these groups are taxonomically divergent from well-studied angiospermspecies, their wax profiles will be crucial to understanding the evolution of the plantcuticle.Aerial plant organs are covered with waxes throughout their development. Sinceplant organs are pressured by different stressors according to their age, it seems likelythat wax coverage and composition may be optimized for these stresses in an age-dependent manner. Indeed, previous reports had indicated that wax coverages andcompositions vary between plant surfaces of different age [6, 217, 184, 75, 235]. How-ever, while some of these reports provide detailed chemical information, it is notlinked with the physiology and genetics of the developing organ to move closer toanswering larger biological questions about relationships between structure and func-tion. To do so again requires the tools available in model systems. In an attempt tobetter understand how wax coverage and composition relate to the development ofplant surfaces, an investigation of wax on developing Arabidopsis stems had been at-tempted [204]. However, since stems expand very rapidly it was not possible to carryout measurements with sufficient resolution to accomplish these goals. Accordingly,our understanding of how cuticular wax deposition relates to plant surface develop-ment in Arabidopsis remains unclear, begging alternative means of investigation ofthis dimension of biological variability and how it is related to the developmentalbiology of the plant.Thus, the literature had so far demonstrated that cuticular waxes exhibit struc-tural diversity in their aliphatic tails and functional group(s) and biological variabilitybetween surfaces of different species, surfaces of different plant organs, and organ sur-faces at different ages. Limited evidence had also suggested that different epidermalcell types may coat themselves with distinct wax mixtures [71]. All this diversityindicates that surfaces are optimized for different environmental conditions and phys-iological functions. A full understanding of the major dimensions of this diversity is aprerequisite for understanding the relationships between cuticular wax structure andfunction.101. Introduction1.3 Goals of this workGiven our current knowledge of cuticular wax compound structure, biosynthe-sis, and diversity as outlined above, many questions still remain in the field. Forexample, how do chain length distributions affect the water barrier formed by cutic-ular waxes? What are the specific functions of the different wax compound classes?How are wax compositions optimized for different environmental conditions? Howdoes the optimization of wax composition relate to plant development? What kindsof compositional or coverage optimizations characterize surfaces with different func-tions? What are the structures of plant specialty wax compounds? How are theybiosynthesized? Do their functions differ from those of ubiquitous wax compounds?These questions will be answered in time by contributions from many investigators.To assist this effort, I studied the wax mixtures of A. thaliana and those of the mossspecies Funaria hygrometrica, as well as the structures of specialty wax compoundswith following questions. 1) What are the structures of the branched wax compoundson A. thaliana, how are these biosynthesized, and how does this biosynthesis relatedto that of unbranched compounds? 2) What are the structure of the specialty waxcompounds that have been reported thus far and, based on their structures, howmight they be biosynthesized? 3) What is the coverage and composition of the waxmixture on a moss species, and how does this compare with those of well-studiedangiosperm species? 4) How do the coverage and composition of the A. thaliana waxmixture relate to leaf development?To answer these questions I combined organic and analytical chemistry techniques.In Chapter 2 I present a determination of the structure of A. thaliana branched waxcompounds via mass spectral analysis, microscale derivatization, and the synthesis ofauthentic standards. I also quantified branched and unbranched compounds on wild-type and mutant plants to advance our understanding of the biosynthesis of bothbranched and unbranched wax compounds. In Chapter 3 I present a comprehensiveliterature survey that systematically documents the diversity of specialty wax com-pounds that had been reported in literature. I then use their structures and relativeabundances to outline several possible ways in which they might be biosynthesized.In Chapter 4 I document the structural elucidation of novel wax compounds fromthe surface of the moss Funaria hygrometrica via mass spectral analysis and organic111. Introductionsynthesis. Chapter 5 complements this structure elucidation with a comprehensiveanalysis of the waxes from each aerial structure of F. hygrometrica plants1 to providea picture of wax coverage and composition on a member of a plant group that hadseldom been investigated. Finally, in Chapter 6 I present two orthogonal analyses ofthe changes in wax coverage and composition on developing Arabidopsis leaves thatare complemented with a gene expression analysis2 to demonstrate how cuticular waxcoverage and composition changes significantly3 throughout leaf development4.1grown and extracted by Jessica Budke2performed by Daniela Hegebarth3statistical analyses performed by Edward Kroc4leaf morphometrics performed by Yan Cao12Chapter 2Branched compounds fromArabidopsis thaliana cuticularwaxes2.1 IntroductionAerial plant surfaces are covered with a layered lipophilic structure called thecuticle that prevents transpirational water loss, facilitates internal water transfer,and is a first line of defense against herbivory and pathogen infection [252]. Thecuticle is built on top of epidermal cell polysaccharide cell walls and consists of twocomponents: cutin and cuticular wax. Cutin is a polyester matrix that providesstructural support and influences the biomechanical properties of the cuticle [86, 50].Cuticular waxes are mixtures of very-long-chain (VLC) fatty acid derivatives foundimbedded in and deposited on top of the cutin matrix where they self-assemble andself-heal to seal the surface against water loss [92, 252, 111, 12, 112]. Typical waxcompounds are homologous series of unbranched aldehydes, alkanes, primary alcohols,wax esters, and fatty acids.Cuticular wax compounds are formed via dedicated biosynthetic pathways thatuse long-chain fatty acids (LCFAs) as substrates. These substrates are biosynthesizedde novo in epidermal cell plastids via the condensation of an acetyl-CoA startermolecule (a C2 compound) with malonyl-ACP and subsequent reduction of the β-132. Branched compounds from A. thaliana waxescarbon to generate butanoyl-ACP (a C4 chain). This product is again condensedwith malonyl-ACP and reduced to form hexanoyl-ACP (a C6 chain). By repeatingsuch condensation and reduction cycles the plastidial protein machinery generatesC16 fatty acids (called LCFAs) that serve as precursors for membrane and storagelipids, but also for extracellular lipids such as cutin, suberin, and waxes.LCFAs are delivered to the endoplasmic reticulum (ER) where they are convertedinto acyl-CoAs, substrates for wax biosynthesis. Wax biosynthesis consists of elon-gation and modification pathways. Analyses of model plant species, Arabidopsisthaliana in particular, have been instrumental in identifying the enzymes involvedin the elongation of LCFAs. In the elongation pathway, a condensation and reductioncycle very similar to that operating in LCFA de novo biosynthesis is carried out bya set of enzyme complexes called fatty acid elongases (FAEs). In these a ketoacyl-CoA synthase (KCS) enzyme first condenses substrate acyl-CoAs with malonyl-CoAto produce a β-ketoacyl-CoA. These β-functional compounds are then reduced bya ketoacyl-CoA reductase (KCR) to generate β-hydroxyacyl-CoAs. An eliminationreaction catalyzed by a hydroxyacyl-CoA dehydratase (HCD) then generates enoyl-CoAs (α,β-unsaturated acyl-CoAs) that are then substrates for an enoylacyl-CoAreductase (ECR) to produce saturated acyl-CoAs with two carbons more than theoriginal KCS substrate. Acyl-CoAs can repeatedly enter a FAE complex and becomevery-long-chain acyl-CoAs ranging from C22 to C38. Several FAE complexes exist inArabidopsis, very likely in close proximity within the epidermal cell ER, and all sharethe same KCR, HCD and ECR enzymes but differ in their KCS. It is well establishedthat the KCSs have different substrate and product chain length specificities, for ex-ample, the Arabidopsis KCS enzyme encoded by the CER6 1 gene had been foundto elongate acyl-CoAs primarily to C28 [213]. Thus, the FAEs containing differentKCSs elongate LCFA-derived acyl-CoAs to different lengths, and, depending on therelative amounts and activities of the different FAEs, one or more pools of acyl-CoAwax precursors with characteristic chain length distributions are generated.Acyl-CoA elongation in Arabidopsis is also influenced by proteins belonging toa subfamily of the BAHD2. These proteins, CER2, CER26 and CER26-like, seem1CER, from the latin esceriferum, meaning “without wax”2named after the founding members of the enzyme family: benzylalcohol O-acetyltransferase, anthocyanin O-hydroxycinnamoyltransferase, anthranilate N-hydroxy-142.BranchedcompoundsfromA.thalianawaxesOSCoAOSCoAB)C)DEARFAEFAEOSCoAD)DEARFAEtotal carbonnumber parityoddevenevenoddevenoddDEARnOHOnOHOnOHOstarter molecule LCFA wax compoundsOSCoALCFAbiosynthA)FAEoddevenDEARnnOHOwax compound modification pathwaysOSACPOHOOSCoAOHOnn nOHwax precursorelongation pathwayLCFA de novobiosynthesisOSACPOHOOSCoAOHOOSACPOHOOSCoAOHOOSACPOHOOSCoAOHOn nn nOHn nn nOHn nn nOHn nOSCoAn nOSCoAn nOSCoAn nOSCoAacyl-CoA wax precursorsLCFA de novo biosynthesis Wax biosynthesisLCFAbiosynthLCFAbiosynthLCFAbiosynthFigure 2.1: (Continued on the following page.)152. Branched compounds from A. thaliana waxesFigure 2.1: Biosynthesis of wax alkanes and alcohols using different startermolecules. Biosynthesis of wax compounds is composed of (left to right) the initia-tion of LCFA de novo biosynthesis with a starter molecule, wax precursor elongationwith malonyl-CoA, and wax compound modification. This yields wax compoundswith characteristic total carbon number (TCN) and parity. A) wax biosynthesis ini-tiated with an acetyl-CoA starter yields unbranched alkanes with odd TCNs andunbranched alcohols with even TCNs. B) wax biosynthesis initiated with a valine-derived starter yields iso-alkanes with odd TCNs and iso-alcohols with even TCNs.C) wax biosynthesis initiated with a leucine-derived starter yields iso-alkanes witheven TCNs and iso-alcohols with odd TNCs. D) wax biosynthesis initiated with anisoleucine-derived starter yields anteiso-alkanes with even TCNs and anteiso-alcoholswith odd TCNs. Abbreviations: DE = decarbonylation pathway, AR = acyl reduc-tion pathwayto modulate the specificity of the CER6 condensing enzyme. Using heterologousexpression in yeast, it had been observed that, in the presence of CER2, the productprofile of CER6 shifts towards C30 [84], whereas the presence of CER26 enabled CER6to elongate up to C34 [160]. However, the biochemical mechanism by which the CER2-like proteins influence CER6 is still unclear [82]. Nevertheless, these proteins playimportant roles in influencing the characteristic chain length distributions of the acyl-CoA wax precursor pool.The elongation pathway generates a pool of homologous VLC acyl-CoAs, generallyranging from C22 to C38, that can serve as substrates for the modification pathways.These pathways are also in the ER, and modify wax precursor head groups to gen-erate wax compounds. One modification pathway, the alkane pathway, contains areduction step, yielding aldehydes with even total carbon numbers (TCNs), and adecarbonylation step, generating alkanes with odd TCNs (Fig. 2.1A, decarbonylationpathway, DE). Mutant analyses and heterologous expression in yeast had establishedthat the CER3 and CER1 enzymes catalyze these reactions in Arabidopsis, respec-tively [24]. In a second modification pathway, the acyl reduction pathway, acyl-CoAscan be reduced to alcohols with even TCNs (Fig. 2.1A, acyl reduction pathway, AR).In Arabidopsis this process is carried out in a single enzymatic step catalyzed bythe fatty acyl-CoA reductase CER4 [182]. The resulting alcohols are then availablefor esterification with fatty acyl-CoAs by a wax ester synthase enzyme to producecinnamoyl/benzoyltransferase, deacetylvindoline 4-O-acetyltransferase162. Branched compounds from A. thaliana waxeswax esters, a processes catalyzed by the enzyme WSD1 in Arabidopsis [124]. Finally,based on the presence of free VLC fatty acids in almost all plant wax mixtures, theexistence of a fatty-acid forming pathway seems likely, though no gene candidate forsuch has been identified. To summarize, acyl-CoA starter molecules are convertedinto wax molecules via three distinct processes, first, plastidial condensation andreduction generates LCFAs, then further ER-based elongation generates VLC acyl-CoA wax precursors, and finally the modification pathways convert VLC acyl-CoAwax precursors in the into the wax compounds that accumulate in and on the plantcuticle and form the transpiration barrier.A wax compounds head group polarity and the aliphatic tail structure both affectthe physical properties and thus the biological functions of the cuticular wax mix-ture, similar to the effects of head group and aliphatic tail structure in bilayer lipids.In particular, chain length distributions and unsaturations or branches in the carbonbackbone affect the packing of hydrocarbons and therefore influence the temperature-dependent fluidity of the waxes, analogous to bilayers [22, 195]. However, while chainlength variability among plant wax mixtures had been well documented, unsaturatedor branched wax compounds had, by comparison, scarcely been reported. It is notclear if such carbon backbone variability is indeed rare in plant cuticular waxes, ifthese compounds had been overlooked, or if they could not be identified in previ-ous routine wax analyses. The few reports that had documented such compoundsfound, for example, branched alkanes on Solanum species, Nicotiana benthamina,and Camelina sativa [207, 123, 143, 173], branched fatty acids and alcohols on Bras-sica species [10, 88], and alkenes, cyclopropanes, and branched alkanes on Hordeumvulgare [240]. These studies had indicated that the most commonly encountered mod-ification to wax compound aliphatic tails are branches, and that many more speciesprobably have cuticular wax compounds containing branched aliphatic tails, at leastin low concentrations, that may well affect specific cuticle functions.An evaluation of the role of branched wax compounds in cuticle function will re-quire a detailed understanding of the structure and biosynthesis of these compounds.First, chemical analyses are needed to elucidate exact branch geometries, second, bio-chemical information is needed to shed light on how branches are introduced into thehydrocarbon skeleton, and finally, the biosynthetic pathways that convert branched172. Branched compounds from A. thaliana waxeswax precursors into branched wax compounds will need to be determined.Mass spectrometry, retention time indices, and microscale derivatization hadbeen demonstrated to be effective tools for elucidating the structures of branchedwax compounds. Using these methods, branched alkanes had been reportedfrom a variety of plant species with characteristic branching structures and cor-responding TCNs. For example, both iso-alkanes with odd TCNs only and/oranteiso-alkanes with even TCNs only had been found on Nicotiana benthamiana[143, 191, 197, 244], several Brassica species [10, 114, 154, 208], several Solanumspecies [92, 116, 145, 198, 207, 236], and Camelina sativa [173]. Using the same meth-ods, alkanes with in-chain methyl branches and cyclopropyl rings and even TNCs onlyhad been identified on Hordeum vulgare [240]. Finally, iso-branched primary alcoholsand fatty acids with even TCNs only had been reported from Brassica oleracea [10].The structures of these alcohols and acids were determined by reducing each to alka-nes and analyzing the resulting mass spectra. In summary, methods for determiningexact branch geometries had been established.Based on the branch positions and TCNs of the branched wax compounds thathad been found, several hypotheses for branch installation had been proposed. Fore-most is the incorporation of branched amino acid-derived starters into the de novobiosynthesis of LCFAs, thus leading to wax precursors with methyl branches nearthe chain terminus [114, 154]. In this hypothesis, the branched C4 structure of 2-methylpropanoyl-CoA, generated from valine, could serve as starter for elongationthat could lead to a wax precursor bearing a sub-terminal methyl branch at the farend of the chain, specifically an iso-LCFA or ω−2 methyl-branched LCFA with aneven TCN (Fig. 2.1B). Similarly, leucine could be converted into a 3-methylbutanoyl-CoA starter molecule that could then be used to generate an iso-LCFA with an oddTCN (Fig. 2.1C), and an isoleucine-derived starter could lead to an anteiso-LCFAwith an odd TCN (Fig. 2.1D). It had been demonstrated that such amino acid-derivedstarters could be used by the fatty acid synthesizing machinery in Brassica oleracea togenerate branched LCFAs [114]. Hypotheses for the installation of branches in plantcuticular wax compounds bearing branches not near the chain terminus had alsobeen proposed, for example, the incorporation of a methyl malonyl-CoA extender inLCFA biosynthesis or wax precursor elongation, by which one or more in-chain methyl182. Branched compounds from A. thaliana waxesbranches could be installed during LCFA biosynthesis or wax precursor elongation,though it had been acknowledged that this process could also occur via methylationof an in-chain double bond [240]. Thus, plausible methods had been put forward forbranch installation and thus generation of branched wax precursors.Branched acyl-CoA wax precursors must be converted into branched wax com-pounds by wax biosynthesis enzymes before they can accumulate on the plant sur-face. Branched wax compounds had been quantified on surfaces of wax biosynthesismutant lines of Brassica oleracea and Hordeum vulgare. These data had revealed thatthe ratio of branched compounds to unbranched compounds can be altered in certainmutant lines, suggesting that some enzymes may have bias for or against branchedsubstrates, and that compounds with branched backbones may be synthesized bydifferent enzymes than those generating unbranched compounds [10, 240]. However,because these analyses were performed in species where wax biosynthesis pathwaysare not well characterized, details about which enzymes may exhibit bias or how suchpathways might overlap with unbranched biosynthesis pathways remain unclear.To date, the species in which wax biosynthesis pathways are best characterizedis Arabidopsis thaliana. Iso-alkanes had been reported from Arabidopsis surfaces,as had branched alcohols, though the alcohols structures are not known. Thus, tofurther investigate the biosynthesis of branched wax compounds and move closer tounderstanding their roles in cuticular wax mixtures, this study was undertaken withthe goals of 1) determining the structure of branched wax compounds on Arabidopsissurfaces, and 2) evaluating which wax biosynthesis genes might be responsible fortheir biosynthesis. Accordingly, waxes from Arabidopsis stems, siliques, flowers, andleaves were extracted and analyzed with gas chromatography - mass spectrometry(GC-MS) and gas chromatography - flame ionization detection (GC-FID). Candidatebranched wax compound classes were analyzed in detail and authentic standards weresynthesized to confirm their structures. Finally, the relative abundances of branchedwax compounds on the surfaces of wax biosynthesis mutant lines were measured todetermine what biosynthesis genes might participate the generation of branched waxcompounds.192. Branched compounds from A. thaliana waxes2.2 ResultsThe objectives of this investigation were to determine the structure of branchedcompounds in Arabidopsis thaliana cuticular waxes and to test the involvement ofdiverse wax biosynthesis genes involved in their production. Homologous series ofbranched wax compounds identified on wild-type Arabidopsis surfaces were purifiedwith thin layer chromatography (TLC) and converted into structurally informativederivatives (2.2.1). The abundances of branched wax compounds on surfaces of Ara-bidopsis mutant lines were then determined (2.2.2).2.2.1 Structure of branched aliphatic wax compounds onArabidopsis surfacesTo search for branched wax compounds, wax samples from wild-type Arabidopsisstem, silique, flower and leaf surfaces were screened. Besides the typical, unbranchedwax constituents, several unknown compounds were detected with GC-MS. Based onsimilar MS fragmentation patterns and GC elution intervals, these were assigned totwo homologous series, designated A and B, which were both present in substantialquantities in flower and leaf waxes, respectively. Series A was also identified in tracequantities in stem wax. Neither series was present in silique wax.To investigate series A further, large quantities of flower wax were extracted andfractionated by TLC. Compounds in A co-eluted with the unbranched (n-)alkanespresent in the flower wax, indicating that compounds in A had very low polarity andsuggesting hydrocarbon structures devoid of functional groups. Each compound in Aeluted ca. 0.7 min. before an n-alkane under the GC conditions used here (Fig. 2.2A,top panel), supporting the hypothesis that compounds in A were branched alkaneisomers. The mass spectra of all compounds in A were characteristic of alkanes, withseries of highly abundant alkyl fragments m/z 57, 71, 85, etc. (Fig. 2.2B, top panel).These mass spectra were further characterized by diagnostic fragments M-15 and M-43 but no fragment M-29, indicative of a methyl branch on a penultimate carbon andthus iso-alkane structures.To validate the structure assignment for compounds in A, an authentic C29 iso-alkane standard was synthesized. For this, tetracosanol was first converted into tetra-202. Branched compounds from A. thaliana waxesFig.	2:		 	OHC20H41 DMPOC20H41C20H41H2, Pd(C)C20H412.12.42.2CHCl3DCM PPhPhPh2.3Figure 2.2: A. thaliana branched alkane structure elucidation. A) Gas chro-matogram of TLC-purified flower wax alkanes (top panel), gas chromatogram of thesynthetic C29 iso-alkane standard (bottom panel). B) Mass spectrum of the branchedC29 alkane from the TLC-purified flower wax (top panel), mass spectrum of the C29iso-alkane standard (bottom panel). C) Synthesis of a C29 iso-alkane standard.cosanal (2.1), and the latter coupled with an isoamyl triphenylphosphorous ylide (2.2,Fig. 2.2C). The resulting C29 iso-alkene (2.3) was then reduced with hydrogen in thepresence of a palladium catalyst to yield a C29 iso-alkane (2.4). The GC retention212. Branched compounds from A. thaliana waxesbehaviour of this synthetic standard (Fig. 2.2A, bottom panel) matched that of onehomolog in series A, and the mass spectra of the standard (Fig. 2.2B, bottom panel)and natural product were identical. These findings confirmed that compounds inseries A were C27, C29, C30, and C31 iso-alkanes, and ruled out any alternative struc-tures with in-chain methyl branches or double-branching near both chain termini.To elucidate the structures of the compounds in series B, leaf wax was iso-lated and fractionated with TLC prior to GC-MS analysis. Compounds in B co-migrated with the unbranched (n-)primary alcohols during TLC separation, indi-cating similar polarity and thus probably isomeric primary alcohol structures. Totest this hypothesis, an aliquot of the TLC fraction was first derivatized with bis-N,O-trimethylsilyltrifluoroacetamide to transform hydroxyl groups into trimethyl silyl(TMS) ethers. GC-MS analysis showed that each homolog of series B eluted ca. 0.7min. before an n-primary alcohol homolog (Fig. 2.3A, top panel), suggesting thatB might be a series of branched primary alcohols. The mass spectra of the TMSderivatives of compounds in B were identical to those of n-primary alcohols, withcharacteristic fragments m/z 73 [Si(CH3)3]+ and 103 [CH2OSi(CH3)3]+ indicatingthe presence of a TMS-derivatized primary hydroxyl group (Fig. 2.2B, top panel).Chain-length-specific M-15 fragments matched between each unbranched alcohol ho-molog and the compound in B eluting directly before it, further confirming that thecompound classes were isomeric. Together, the data indicated that compounds in Bwere primary alcohols with branched hydrocarbon backbones.To test whether compounds in B were branched primary alcohols, a representativeC30 iso-alcohol was synthesized. To this end, docosanedioic acid was converted intwo redox steps to docosanedial (2.6), and the resulting dialdehyde was coupledsimultaneously with a (3-benzyloxypropyl)triphenyl phosphorous ylide (2.7) and anisoamyl triphenyl phosphorous ylide (2.2) to produce a benzyl-protected C30 iso-alkene (2.8, Fig. 2.3C). This intermediate product was then reduced and deprotectedby stirring with palladium on carbon under a hydrogen atmosphere, to yield a C30iso-alcohol (2.9). The retention time of the synthetic C30 iso-alcohol (Fig. 2.3A,bottom panel) matched that of one homolog in series B, and the mass spectrumof the synthetic compound (Fig. 2.3B, bottom panel) was identical to that of thenatural product. This further confirmed the structure of the compounds in B as222. Branched compounds from A. thaliana waxesFig.	3:	Elucidation	of	A.	thaliana	branched	alcohol	structure.		 	OOHOOH9LAHHOOH9DMPOO99H2, Pd(C)HO2.5 2.62.82.9THFCHCl3O PPhPhPhO12PPhPhPhCH2Cl22.72.2Figure 2.3: (Continued on the following page.)232. Branched compounds from A. thaliana waxesFigure 2.3: A. thaliana branched alcohol structure elucidation. A) Gas chro-matogram of TLC-purified leaf wax alcohols (top panel), gas chromatogram of thesynthetic C30 iso-alcohol standard (bottom panel). B) Mass spectrum of the branchedC30 alcohol from the TLC-purified leaf wax (top panel), mass spectrum of the C30iso-alcohol standard (bottom panel). C) Synthesis of a C30 iso-alcohol standard.branched primary alcohols. However, it should be noted that similar skeletal isomers,for example, 2-methyl primary alcohols, would likely have very similar GC behavior,and therefore the branching geometry of compounds B could therefore not be assignedbased on GC retention behavior alone.The mass spectrum of the iso-branched C30 primary alcohol standard (Fig. 2.3B,bottom panel) matched that of corresponding natural product (Fig. 2.3B, top panel),in accordance with the notion that the homolog of B was iso-branched C30 alco-hol. However, both mass spectra were also identical to those of the accompanyingunbranched alcohol and an authentic unbranched alcohol standard. Thus, the massspectra of the putative branched alcohols in series B did not provide any informationabout their branch structure and, in contrast to the branched alkanes, the structureof the branched primary alcohols could not be unambiguously assigned based on GC-MS analysis of the TMS derivatives alone. To fully elucidate the structure of thecompounds in B, they were transformed into two other derivatives whose mass spec-tra promised to provide information on the nature and position of the hydrocarbonbackbone branch.In one experiment, the primary alcohols of series B were transformed into alkanesin order to assess the number of branches and locate the branch position(s) relative tothe chain termini. For this, a method similar to that of Baker and Holloway was used[10]. An aliquot of the TLC-isolated Arabidopsis leaf primary alcohols were treatedwith phosphorus tribromide (PBr3) to convert them into the corresponding alkylbromides, which were then reduced to alkanes with lithium aluminum hydride (LAH,Fig. 2.4A). GC-MS analysis of the reaction products revealed a series of nine homologs(Fig. 2.4B) with relative abundances similar to those of the original n-primary alcoholsin the fraction and characteristic mass spectra (Fig. 2.4C, top panel), allowing theseries to be assigned as n-alkanes ranging from C26 to C34. A second series was foundthat contained five homologous compounds in relative abundances similar to those of242. Branched compounds from A. thaliana waxesFig.	4:	Reduction	of	A.	thaliana	primary	alcohols.		 	??PBr3CH2Cl2LAHTHFOHOH????BrBr??????nnnnnn5771859MW = 4225771859407379MW = 422Figure 2.4: Reduction of A. thaliana primary alcohols. A) The mixture of n-and branched alcohols from Arabidopsis leaves were reacted with phosphorous tribro-mide (PBr3) and then reduced with lithium aluminum hydride (LAH) to transformthem into alkanes. B) Gas chromatogram of the reduction reaction products. C) EImass spectrum of the C30 n-alkane (top) and EI mass spectrum of the first homologin series B after reduction (bottom).252. Branched compounds from A. thaliana waxesseries B in the untreated TLC fraction. The compounds had GC elution times 0.7min. less than those of the accompanying n-alkanes (Fig. 2.4B), similar to the iso-alkanes detected in Arabidopsis flower waxes, and also mass spectra matching thoseof iso-alkanes (Fig. 2.4C, bottom panel). Considered together, the evidence suggestedthat this was a series of iso-alkanes with a methyl branch on C-2, and that this seriesarose from the reduction of the compounds in B. Therefore, the primary alcohols ofseries B contained one methyl branch adjacent to a chain terminus. However, basedon the reduction reaction products alone, it could not be distinguished whether themethyl branch was located on the alpha or ω−2 carbon of the primary alcoholsin series B, and thus whether iso-branched or 2-methyl branched primary alcoholscomprised that series.To determine whether compounds in B had 2-methyl or omega-2-methyl branches,they had to be transformed into further derivatives with diagnostic MS fragmentationpatterns. For this second experiment, it seemed reasonable to convert the alcoholsinto aldehydes, as the latter generates characteristic MS fragments that include oneor more carbons adjacent to the carbonyl function whose mass would probably beincreased by the presence of a methyl branch on the alpha-carbon. To verify this de-duction, a C24 2-methyl aldehyde was synthesized. For this, docosanol was oxidizedto docosanal (2.10), the aldehyde was converted with methyl magnesium bromideinto tricosan-2-ol (2.11), and this alcohol was transformed with PBr3 into the corre-sponding alkyl bromide (2.12), then into a Grignard reagent that was treated withanhydrous dimethylformamide to generate 2-methyl-tricosanal (2.13, Fig. 2.5A). Thebase peak in the mass spectrum of this 2-methyl aldehyde was m/z 58 (Fig. 2.5B, toppanel), whereas the corresponding unbranched C24 aldehyde, which had been synthe-sized as an intermediate leading to the C29 iso-alkane (Fig. 2.2C) did not exhibit thisfragment (Fig. 2.5B, bottom panel), confirming m/z 58 as diagnostic for 2-methylbranching in VLC aldehydes.In order to exploit the diagnostic MS fragment for the presence or absence of2-methyl branching in VLC aldehydes, the primary alcohols in series B had to beconverted into the corresponding aldehydes. An aliquot of the TLC fraction was dis-solved in dry dichloromethane (CH2Cl2) and oxidized with Dess-Martin periodinane(DMP, Fig. 2.6A). GC-MS analysis of the reaction products revealed two homolo-262. Branched compounds from A. thaliana waxesFig.	5:	Aldehyde	synthesis	C19H39 OH C19H39 OC19H39 OHC19H39 BrC19H39ODMPTHFMeMgBrCH2Cl2PBr3THF 1. Mg2. DMF2.102.112.122.13CH2Cl2Figure 2.5: Synthesis of C24 aldehydes with and without 2-methyl branching.A) Synthesis of a C24 2-methyl aldehyde. B) EI mass spectrum of the synthetic C242-methyl aldehyde (top), EI mass spectrum of the C24 n-aldehyde (bottom), obtainedas an intermediate in the synthesis of the C29 iso-alkane (Fig. 2.2C).gous series. The first had nine members with relative abundances similar to those ofthe n-primary alcohols in the untreated TLC fraction (Fig. 2.6B) and mass spectrawith characteristic fragments m/z 82 and 96, but no peak at m/z 58 (Fig. 2.6C, toppanel). The compounds in this homologous series were thus identified as aldehydeswithout 2-methyl branching, were recognized as derivatives of the unbranched pri-272. Branched compounds from A. thaliana waxesmary alcohols in the original fraction, and were thus assigned as n-aldehydes rangingC26 to C34. A second homologous series was detected by GC-MS, comprising fourcompounds eluting ca. 0.7 min. before homologs of the first series and a chain lengthprofile identical to that of series B. The mass spectra of compounds in this second se-ries also exhibited prominent fragments m/z 82 and 96, revealing that they were alsoaldehydes that did not have 2-methyl branching (Fig. 2.6C, lower panel), eliminatingthe possibility that the compounds in B were 2-methyl alcohols.In summary, our experiments characterized compounds in B as a series of ho-mologs that contained a primary alcohol function and an iso-alkane backbone, but no2-methyl branching. It was thus concluded that compounds in B were iso-branchedprimary alcohols, and Arabidopsis flower and leaf waxes both contain homologousseries of iso-alcohols with mainly even total carbon numbers ranging from C30 to C34.2.2.2 Accumulation of branched wax compounds in waxbiosynthesis mutantsMany wax biosynthesis mutant phenotypes have been reported as part of pub-lished Arabidopsis gene characterizations. However, the abundances of branchedcompounds are seldom included in these analyses. To establish which wax biosynthe-sis genes might be involved in the production of branched compounds, the coverageand composition of both unbranched and branched compounds on flowers and leavesof Arabidopsis wild type and cer1, cer3, cer4, cer6, cer2, cer26, and cer16 mutantswere determined.Cuticular waxes extracted from mature Columbia0 (Col0) and Landsberg (Ler)flowers and leaves were separated with gas chromatography, identified with massspectrometry, and quantified with flame ionization detection. Flower waxes fromboth ecotypes contained ubiquitous compound classes with typical parity: fatty acids(C24 - C30), n-alcohols (C22 - C32), alkyl esters (C36 - C54), and aldehydes (C26 -C30) all with even total carbon numbers, and n-alkanes (C25 - C33, Tables A.1 andA.2) with odd total carbon numbers. Flower samples from both Col0 and Ler alsocontained iso-alkanes (C27, C29, C30, and C31), and traces of iso-alcohols (C28 andC30). In both ecotypes the major n-alkane homolog in flower wax was C29 (30-40%of the total wax, Fig 2.7A), and while the iso-alkanes in Col0 were relatively evenly282. Branched compounds from A. thaliana waxesFig.	6:	Oxidation	of	A.	thaliana	primary	alcohols.		 	??DMPCH2Cl2OHOH????OO??nnnnFigure 2.6: Oxidation of A. thaliana primary alcohols. A) The mixture ofn- and branched alcohols from Arabidopsis leaves were reacted with Dess-Martinperiodinane to transform them into aldehydes. B) Gas chromatogram of the reductionreaction products. C) EI mass spectrum of the C30 n-aldehyde (top) and EI massspectrum of the first homolog in oxidized series B (bottom).distributed between C29 and C31 (ca. 4-5% each), in Landsberg the C31 homolog wasnearly twice as abund t as the C29 homolog (10% vs. 5%).Leaf wax from both ecotypes also contained ubiquitous compound classes of typ-ical parity: fatty acids (C22 - C34), n-alcohols (C22 - C34), alkyl esters (C36 - C48),292. Branched compounds from A. thaliana waxesFigure 2.7: Relative abundance of n- and iso-branched alcohols and alkanesin Arabidopsis flower and leaf waxes. A) Relative abundance of alcohol andalkane compounds in Col0 (black) and Ler (grey) flowers. B) Relative abundanceof alcohol and alkane compounds in Col0 (black) and Ler (grey) leaves. Error barsindicate the standard deviation of five independent samples.aldehydes (C26 - C34) all with even total carbon numbers, and n-alkanes (C25 - C37)with odd total carbon numbers (Tables A.3 and A.4). Leaf samples also containedboth iso-alkanes (C29 and C31) and iso-alcohols (C28, C30, C31, C32, and C34). Onleaves from both ecotypes the n-alkanes were present mainly as C29, C31, and C33homologs (ca. 10%, 20%, and 7% of the total wax, respectively), and the two iso-alkanes were present in roughly equal amounts (ca. 1%, Fig. 2.7B). The n-alcoholswere dominated by the C28 homolog (ca. 4%), though the C26, C30, and C32 ho-mologs were also present in substantial amounts (ca. 2% each). The most abundantiso-alcohol was the C32 homolog (ca. 5%) followed by the C30 homolog (ca. 4%).Overall, iso-branched compounds on both flowers and leaves were present as pre-dominantly C32/C31 homologs followed by C30/C29 homologs, regardless of the chainlength distribution of the corresponding unbranched compound class.The flower and leaf waxes of the wax biosynthesis mutants were analyzed in thesame way as the wild-type waxes. The difference in the relative abundance of each302. Branched compounds from A. thaliana waxesFigure 2.8: (Continued on the following page.)312. Branched compounds from A. thaliana waxesFigure 2.8: Differences in flower wax composition between Arabidopsis mu-tants and corresponding wild type. Bar heights indicate the difference betweenthe relative abundance of each alcohol and alkane compound on each mutant and thecorresponding wild type. Error bars indicate the standard deviation of the differencebetween five independent samples of each mutant and wild type. Grey boxes high-light groups of compounds with decreased abundance. Asterisks indicate significantdifferences determined using the Benjamini and Hochberg false discovery rate (FDR)controlling procedure (q=0.01).compound on wild-type and mutant surfaces was calculated and used to assess theeffects of each mutation on the biosynthesis of unbranched and branched wax com-pounds. The flowers of cer1 and cer3 exhibited significant reductions in n-alkaneabundance accompanied by small reductions in iso-alkane abundance (Fig. 2.8). Mu-tation of the acyl reduction pathway gene CER4 caused decreases in n-alcohols with30 or fewer carbons and a small decrease in C30 iso-alcohol abundance. Mutationof the CER6 locus caused reductions in the relative abundance of both n- and iso-compounds with more than 28 carbons. Mutation of the elongation machinery com-ponent CER2 led to a decrease in the abundance of unbranched compounds withmore than 28 carbons, and an increase in the abundance of iso-alkanes. Flower waxfrom cer26 was indistinguishable from that of the wild type. Finally, cer16 exhib-ited a slight increase in the abundance of C29 alkane, and was completely devoid ofiso-alkanes.In leaves, mutation of either of the decarbonylation pathway genes CER1 or CER3caused decreases in the abundance of n-alkanes, and cer3 also exhibited decreases inprimary alcohol abundance (both n- and iso-, Fig. 2.9). Leaves of cer4 had slightlymore C32 iso-alkane than the wild type. Leaves of cer6 exhibited massive reductionsin nearly all compounds with more than 26 carbons, including n- and iso-alcoholsand the n-alkanes. Mutation at the CER2 locus caused a small decrease in iso-alkaneabundance but increased abundance of C32 iso-alcohol. The cer26 line exhibited areduction in the abundance of n-alkanes and n-alcohols with more than 30 carbons.Finally, cer16 leaves exhibited significant reductions in the levels of C28 and C30branched alcohols.To summarize, mutation of the genes encoding for in the decarbonylation pathwayenzymes (CER1 or CER3 ) led to elimination of leaf and flower iso-alkanes, while322. Branched compounds from A. thaliana waxesFigure 2.9: (Continued on the following page.)332. Branched compounds from A. thaliana waxesFigure 2.9: Differences in leaf wax composition between Arabidopsis mu-tants and corresponding wild type. Bar heights indicate the difference betweenthe relative abundance of each alcohol and alkane compound on each mutant and thecorresponding wild type. Error bars indicate the standard deviation of the differencebetween five independent samples of each mutant and wild type. Grey boxes highlightgroups of compounds with decreased abundance. Asterisks indicate significant dif-ferences determined using the Benjamini and Hochberg (FDR) controlling procedure(q=0.01).deactivation of the acyl reduction pathway (CER4 ) led to an increase in leaf iso-alcohols. Mutation of CER6 eliminated most branched compounds on both organs,and elimination of elongation components CER2 or CER26 caused relatively littlechange in branched compound abundance on flowers and increases in the levels ofbranched alcohols on leaves. Mutation of CER16 reduced the levels of branchedcompounds on both organs.2.3 DiscussionThe objectives of this work were to determine the structure of branched com-pounds in Arabidopsis thaliana cuticular waxes and to quantify them in wax biosyn-thesis mutant lines. Branched alkanes and alcohols were found in flower and leafwaxes, and microscale derivatization combined with organic synthesis demonstratedthat these compounds contained iso-branched aliphatic chains. In wild type, thebranched alkanes and alcohols both had chain length profiles peaking at C31 and C32,respectively, regardless of the homolog distributions of accompanying unbranchedcompounds. The relative abundances of branched wax compounds on the cer6, cer3,cer1, and cer16 mutant lines were significantly affected, but the relative abundancesof branched compounds on cer2 and cer26 were not affected. Here these findingswill be discussed in the context of the generation of branched-chain wax precursors(2.3.1) and their modification by wax biosynthesis enzymes (2.3.2).342. Branched compounds from A. thaliana waxes2.3.1 Generation of branched chain wax precursorsUnbranched alkane and alcohol wax compounds are derived from unbranched long-chain fatty acids (LCFAs) that in turn arise from an (unbranched) acetyl-CoA startermolecule (Fig. 2.1A). By this process, the alkanes generated by the decarbonyla-tion pathway have odd total carbon numbers (TCNs) and the alcohols generatedby the acyl reduction pathway have even TCNs. It had been demonstrated thatbranched LCFAs are biosynthesized from branched starter molecules [117], and ithad been hypothesized that these branched LCFAs serve as precursors for branchedwax compound biosynthesis [114, 125]. For example, a valine-derived starter moleculewould lead to iso-branched LCFAs and subsequently iso-branched alkanes with oddTCNs and iso-branched alcohols with even TCNs (Fig. 2.1B), while a leucine-derivedstarter would lead to iso-alkanes with even TCNs and iso-alcohols with odd TCNs(Fig. 2.1C). By the same mechanism, an isoleucine-derived starter would lead toanteiso-branched alkanes with even TCNs and anteiso-branched alcohols with oddTCNs. (Fig. 2.1D).Both iso-alkanes with odd TCNs and anteiso-alkanes with even TCNs had beenreported from other species, for example, from Nicotiana benthamiana [143, 191, 197,244], several Brassica species [10, 114, 154, 208], several Solanum species [92, 116,145, 198, 207, 236], Hordeum vulgare [240], and Camelina sativa [173]. These patternsindicate that, in these species, LCFAs arise from both valine- and isoleucine-derivedstarters and are probably incorporated into the biosynthesis of wax precursors thatin turn give rise to branched wax alkanes. Further investigations may reveal if thesame occurs for leucine-derived starter molecules in plant epidermal cell metabolism.In this study each Arabidopsis wax sample was searched for all possible branchedalkane isomers, but only iso-alkanes with odd TCNs were detected. This is identical toprevious analyses of Arabidopsis waxes [28, 160], though here additional support forthe iso-alkane structure assignment is provided with authentic standard coelution.These data indicate one of two possibilities for Arabidopsis, either 1) only valinederived-starters are incorporated into the de novo synthesis of LCFA or 2) multipletypes of branched starters are used in Arabidopsis de novo LCFA biosynthesis but onlyLCFAs arising from valine-derived starters are used in the biosynthesis of branchedalkane wax compounds.352. Branched compounds from A. thaliana waxesIn this work, Arabidopsis surfaces were also exhaustively searched for possiblebranched alcohol isomers. Only iso-branched alcohols with even TCNs were found.This is consistent with previous investigations, which found branched alcohol waxcompounds on, for example, Nicotiana benthamiana [236], Brassica napus [208], andArabidopsis thaliana [28, 37, 160, 176]. On these species, branched alcohols hadeven TCNs and in some cases had been reported as iso-alcohols, though no struc-tural evidence for such assignment had been provided. Here, conclusive evidenceis presented so that Arabidopsis branched alcohols may now be unambiguously as-signed as iso-alcohols with even TCNs. Based on these structural characteristics itseems that Arabidopsis iso-alcohols, just like Arabidopsis iso-alkanes, are generatedfrom LCFAs arising from valine-derived starters, even though leucine- and isoleucine-derived starters are available in Arabidopsis metabolism [209]. This suggests thatthese two branched wax compound classes are biosynthetically related, just as un-branched wax alkanes and unbranched wax alcohols are related.2.3.2 Modification of branched wax compounds by waxbiosynthesis enzymesWild-type Arabidopsis plants (Columbia and Landsberg ecotypes) contained rel-atively highly abundant iso-alkanes in their flower waxes and relatively highly abun-dant iso-alcohols in their leaf waxes. The homolog distributions of these iso-branchedcompound classes peaked at C31 and C32, respectively, indicating that majority ofiso-branched wax precursors are C32. In contrast, the homolog distributions of un-branched alkanes and alcohols either did not peak at C32, or were found with muchbroader homolog profiles, indicating that the majority of unbranched wax precursorsare generated with chain lengths other than C32. This difference strongly suggeststhat both iso-branched compound classes are biosynthesized by the same chain-lengthspecifying machinery and that this machinery is in part or whole different from thatwhich generates unbranched wax compound precursors.The major compound classes in flower wax were n- and iso-alkanes. The relativeabundances of n-alkanes on this organ were significantly reduced in the cer1, cer3,cer2, and cer6 mutant lines. Flower wax profiles had not yet been reported forArabidopsis, but the flower wax profiles of these mutants are consistent with those362. Branched compounds from A. thaliana waxespreviously reported for their stems [24, 84, 95, 160], and they support the currentmodel for n-alkane biosynthesis [24, 82]. In this model CER6 is the ketoacyl-CoAsynthase (KCS) that, assisted by CER2, elongates wax precursors to C30, C32, or C34which then serve as substrate for CER3 and CER1 with which to generate n-alkanes(Fig. 2.10A, red pathway).In flower wax, iso-alkanes were present as C29 and mainly C31 homologs. Theirrelative abundances were reduced on the cer6, cer3, and cer1 mutant lines, just aswere the relative abundances of n-alkanes. This suggests that the condensing activityof CER6 and the decarbonylation activity of CER3 and CER1 are necessary for iso-alkane production, just like n-alkane production, and that therefore the biosyntheticpathway to iso-alkanes is quite similar to that leading to n-alkanes (Fig. 2.10A, bluepathway). Unlike n-alkanes, however, the relative abundances of iso-alkanes were sig-nificantly lower on the cer16 mutant line and were unaffected on the cer2 line. CER6requires an additional protein element to generate compounds with more than 28carbons, and in n-alkane biosynthesis this protein is CER2 [84, 82]. The mutant datapresented here suggest that CER16 may be the additional protein element requiredby CER6 to elongate iso-acyl-CoA precursors past C28 (Fig. 2.10A, blue pathway).Although leaf waxes also contained substantial amounts of n-alkanes, the majorbranched compounds present on this organ were iso-alcohols (C30 and mainly C32),which were accompanied by n-alcohols (present in a broad distribution with a subtlepeak at C28). Here, the relative abundances of n-alcohols were significantly affectedon the cer3, cer6, and cer26 lines. This is consistent with literature reports of mutantleaf waxes [95], and partially supports the current model for n-alcohol biosynthesisin which CER6 is the KCS that, assisted by CER26, elongates wax precursors tochain lengths between C26 and C34 that then serve as substrate for the acyl reductionpathway (Fig. 2.10B, red pathway). The current model for n-alcohol biosynthesis isbased primarily on analyses of Arabidopsis stems, where CER4 had been identifiedas the reducing enzyme on the acyl reduction pathway [182]. However, based on theabsence of n-alcohols in the cer3 mutant wax profile reported here, it seems thatCER3 participates in the generation of n-alcohols in leaf wax biosynthesis. SinceCER3 does contain a reductase domain [24], it seems plausible that this enzyme maybe responsible for the reduction reaction necessary that generates n-alcohols from372. Branched compounds from A. thaliana waxesC26 iso-CoAC28 iso-CoAC30 iso-CoAC32 iso-CoA6A) alkane biosynthesis in wild-type flowersC26 n-CoAC28 n-CoAC30 n-CoAfrom fatty acid elongationC29 iso-alkaneC31 iso-alkane3 C29 n-alkane166 16166 6 23 13 1C26 iso-CoAC28 iso-CoAC30 iso-CoAC32 iso-CoA6B) alcohol biosynthesis in wild-type leavesC26 n-CoAC28 n-CoAC30 n-CoAfrom fatty acid elongationC30 iso-alcoholC32 iso-alcohol3 C30 n-alcohol66 16166 6 26333 C28 n-alcohol3 C26 n-alcoholC32 n-CoA3 C32 n-alcohol6 26C34 n-CoA3 C34 n-alcohol6 26Figure 2.10: Biosynthesis of n- and iso-branched wax compounds. Mutantwax phenotypes and the known roles of characterized wax biosynthesis genes suggestwax biosynthesis pathways for wax alkanes and alcohols. Here the intermediatesand products on these pathways are connected with arrows that indicate enzymaticsteps and boxed numbers that indicate the CER gene(s) that seem to participatein those steps. A) n-alkane biosynthesis is carried out by CER6, CER2, CER3 andCER1 (red pathway), mutant data presented here suggests that CER16 may playbe an auxiliary protein in the biosynthesis of iso-alkanes (blue pathway). B) n-alcohol biosynthesis in leaves is carried out by CER6, CER26, and (as suggested bymutant data presented here) CER3 (red pathway). The same data suggest that thebiosynthesis of iso-alcohols is carried out by the same machinery except that CER16may be the auxiliary protein required instead of CER26 (blue pathway).382. Branched compounds from A. thaliana waxeswax precursors (Fig. 2.10B, red pathway).In leaf wax, iso-alcohols were present as C30 and mainly C32 homologs. Therelative abundances of these compounds were affected in the cer3 and cer6 mutantlines, just as were the abundances of the n-alcohols, suggesting that the condensingactivity of CER6 and the potential reducing activity of CER3 are needed for iso-alcohol formation and that therefore the biosynthesis of both these alcohol classesmay be quite similar. However, unlike the n-alcohols, iso-alcohol abundance wasreduced on the cer16 mutant, but was not reduced in the cer26 mutant. Again,CER6 requires an auxiliary protein element to elongate wax precursors beyond C28which, in leaves, is CER26 [160, 82]. Based this information and the mutant waxphenotypes reported here, it seems that the additional protein element required forelongating iso-branched wax precursors beyond C28 may not be CER26, but insteadCER16 (Fig. 2.10B).2.4 ConclusionsThe structure elucidations and mutant wax profile analyses presented here leadto two main conclusions. First, in Arabidopsis thaliana branched chain LCFAs aris-ing from valine-derived starter molecules are incorporated into the biosynthesis ofwax precursors and eventually the production of branched wax compounds, and thatbranched LCFAs arising from other branched starter molecules are either not presentin the ER, or are selected against by the wax precursor-generating machinery. Second,the same wax biosynthesis proteins responsible for the generation of unbranched waxcompounds handle branched wax precursors to generate branched wax compounds,except for CER2 and CER26, which do not have roles in branched wax compoundbiosynthesis. Instead, the uncharacterized CER16 gene may encode the additionalprotein element necessary for CER6 to elongate branched wax precursors beyond C28.In addition, the data indicate that CER16 may impart the branched wax precursorelongation machinery with specificity such that C32 branched wax precursors are thepredominant branched substrates for the modification pathways. While speculation,the possible roles of CER16 in branched wax compound biosynthesis and in particularits potential biochemical mechanism of action present excellent avenues for future re-392. Branched compounds from A. thaliana waxessearch. Finally, while this work demonstrates that at least one dedicated gene existsfor the production of branched wax compounds and that such compounds are deliber-ate additions to Arabidopsis wax mixtures, the specific function of the branched waxcompounds and how they might contribute to the physical properties of the cuticularwax mixture remains unclear.2.5 Experimental2.5.1 Plant materials, wax extraction, and GC analysisSeeds of Arabidopsis wild-type Columbia0 plants as well as the correspondingcer1, cer2, cer3, cer4, cer6, and cer26 mutants were cold treated and germinated onagar plates as previously described [37]. Seeds from Arabidopsis wild-type Lands-berg plants and the corresponding cer16 mutant were germinated in the same way.Seedlings of all plant lines were transferred to soil once they had developed rosetteleaves and then cultivated in a growth chamber as previously described [37].Wax samples were prepared from Arabidopsis stems, leaves, siliques, and flowersand were analyzed with gas chromatography - flame ionization detection (GC-FID)and gas chromatography - mass spectrometry (GC-MS) to determine wax compositionas described previously [37], identical to as is described in detail in Chapter 5.2.5.2 Transformation of TLC-purified primary alcoholsPrimary alcohols were purified with thin layer chromatography. A plate of 0.25mm thickness (Analtech) was loaded with crude waxes obtained via chloroform ex-traction and was then developed with CHCl3:EtOH (99:1). The bands were scratchedout, extracted with CHCl3, and a small aliquot of each was analyzed with GC-MS.Other aliquots of the primary alcohol-containing fraction were then transferred toother 2 mL vials and the solvent was evaporated. The alcohols in one of these smallvials were dissolved in dry CH2Cl2 (1 mL) and PBr3 (2 drops) was added. The mix-ture was allowed to sit overnight, then saturated KBr solution (1 mL) was added, themixture was vortexed, then the organic layer was removed, the solvent evaporated,and and aliquot of the residue derivatized with pyridine and BSTFA and then an-402. Branched compounds from A. thaliana waxesalyzed with GC-MS. The remainder of the residue was dissolved in dry THF, thenlithium aluminum hydride (LAH, 1 mg) was added. The mixture was allowed to reactovernight, then 1 mL of water was added, the mixture was extracted with hexanes,the hexane solvent was evaporated and the residue was treated with pyridine andBSTFA and analyzed with GC-MS.A second aliquot of the TLC-purified primary alcohols were dissolved in dryCH2Cl2 to which was added Dess-Martin Periodinane (1 mg). This mixture wasallowed to sit and react overnight, then water (1 mL) was added, the mixture wasvortexed, the organic layer was removed, the solvent evaporated, the residue deriva-tized with pyridine and BSTFA, and analyzed with GC-MS.2.5.3 Synthesis of authentic standardsTetracosanal (2.1)Tetracosanol (50 mg, 1 eq.), was dissolved in freshly distilled CH2Cl2 (50 mL)to which was added Dess-Martin Periodinane (DMP, 89 mg, 1.5 eq.) and one dropof trifluoroacetic acid (TFA). The mixture was allowed to stir until thin layer chro-matography (TLC) analysis (CHCl3 mobile phase) indicated the consumption of thestarting material, then the reaction was diluted with Et2O (100 mL), washed withNaOH (2x 30 mL, 1.3 M), dried with Na2SO4, and concentrated in vacuo to yieldcrude tetracosanal that was further purified with silica gel column chromatography(CC) (CHCl3 mobile phase). MS analysis of the product revealed no traces of startingmaterial or products of over oxidation, but some docosanal was present, presumablyderived from docosanol impurity in the starting material. The solid from the product-containing fractions was used directly in the next reaction.2-methyloctacos-4-ene (2.3)(3-Benzyloxypropyl)triphenyl phosphonium bromide (64 mg, 1.1 eq) was sus-pended in freshly distilled tetrahydrofuran (THF) (25 mL) to which was added n-butyllithium (0.09 mL, 1.6 M, 1 eq.) to generate the corresponding phosphorous ylide(2.2). After 15 min. of stirring, tetracosanal that had been dried under high vacuumfor two hours was dissolved in THF (10 mL) and added dropwise. After one hr. stir-412. Branched compounds from A. thaliana waxesring the mixture was diluted with Et2O (50 mL), washed with H2O (2x 25 mL), andthe organic layer was dried with Na2SO4 and concentrated in vacuo to yield crude2.3 which was purified with a short flash chromatography column (hexane mobilephase). The product of this separation was pure 2.3 by GC-MS analysis.2-methyloctacosane (2.4)2-methyloctacos-4-ene (2.3) was dissolved in chloroform (CHCl3) (40 mL), towhich was added palladium on carbon (Pd(C), 10 mg). The mixture was placedunder a H2 atmosphere, stirred vigorously for 12 hrs, then diluted with CHCl3 (80mL), filtered through Celite, and concentrated in vacuo to yield crude 2.4, whichwas purified with TLC (hexane mobile phase). The TLC purified product containedsome unreacted starting material that was easily distinguished from the product byboth retention time and mass spectral fragmentation patterns.1,22-docosanediol (2.5)Docosanedioic acid (200 mg, 1 eq.) was added to a suspension of lithium alu-minum hydride (LAH, 51 mg, 2.5 eq.) in dry THF which was then stirred for twohours until TLC (ethyl acetate mobile phase) indicated consumption of the startingmaterial. A Feiser workup was used to remove aluminum byproducts and isolate or-ganic compounds in Et2O, which was then dried with Na2SO4 and removed in vacuoto yield 2.5, which was judged to be pure by GC-MS analysis.Docosanedial (2.6)Docosane-1,22-diol (2.5) was dissolved in freshly distilled CH2Cl2 (50 mL) towhich DMP (619 mg, 2.5 eq.) and two drops of TFA were added. The reactionwas allowed to stir until TLC (CHCl3 mobile phase) indicated consumption of thestarting material, then the reaction was diluted with diethyl ether (Et2O) (100 mL),and washed with sodium hydroxide (NaOH) (2x 50 mL, 1.3M). The organic layer wasdried with Na2SO4 and concentrated in vacuo. The crude product was purified withCC (CHCl3 mobile phase) and product-containing fractions were pooled to yield 2.6,which was judged to be pure by GC-MS analysis and contained no traces of startingmaterial or over-oxidation products.422. Branched compounds from A. thaliana waxes(((29-methyltriaconta-3,25-dien-1-yl)oxy)methyl)benzene (2.8)Isoamyl triphenyl phosphonium bromide (188 mg, 1.3 eq.) and (3-Benzyloxypropyl)triphenylphosphonium bromide (160 mg, 1.3 eq.) were suspendedin dry THF (50 mL) to which was added n-butyllithium (0.46 mL, 1.6 M, 2.5 eq.) toyield the corresponding phosphorous ylide (2.7). After 15 min. stirring, docosane-dial (2.6, 100 mg, 1 eq.) dissolved in dry THF (10 mL) was added dropwise and thereaction was stirred for two hours. Then the mixture was diluted with Et2O (100mL), washed with H2O, dried with Na2SO4, and then concentrated in vacuo. GC-MSanalysis of the crude product revealed two homo addition products (iso + dialdehyde+ iso; benzyl + dialdehyde + benzyl) and one hetero addition product (iso + di-aldehyde + benzyl) in approximately a 1:4:5 ratio. The hetero addition product waspurified with CC (CHCl3 mobile phase) and product containing fractions were pooledto yield 2.8, which was was then used directly in the next step.29-methyltriacontan-1-ol (2.9)To 2.8 in CHCl3 was added Pd(C) (20 mg). The mixture was placed undera H2 environment, stirred vigorously for 12 hrs, then diluted with Et2O (50 mL),filtered through Celite, then concentrated in vacuo, purified with TLC, analyzedwith GC-MS, and found to contain a C30 iso-alcohol and an n-C28 alcohol arisingfrom impurities in the isoamyl triphenyl phosphonium bromide reagent.Docosanal (2.10)This compound was prepared in the same way as tetracosanal, described above,using docosanol as a starting material.Tricosan-2-ol (2.11)Docosanal (2.10) was dissolved in freshly distilled THF, capped with a septum,and cooled to 0◦C. Methyl magnesium bromide was via syringe and the mixture wasallowed to stir for 30 minutes. After this time Et2O (20 mL) was added and themixture was washed with water (2x 20mL) and brine (1x 15 mL). The organic layerwas dried with Na2SO4 and concentrated in vacuo to yield 2.11 in high purity by432. Branched compounds from A. thaliana waxesGC-MS analysis. No compound corresponding to the addition of multiple methylunits was observed.2-bromotricosane (2.12)Tricosan-2-ol (2.11) was dissolved in freshly distilled dichloromethane (CH2Cl2)and capped with a septum. Phosphorus tribromide (PBr3) was added via syringeand the reaction was allowed to stir overnight at room temperature (RT). The nextday the reaction was added to an ice cold saturated solution of KBr (100 mL). Thismixture was extracted with CHCl3 (3x 20mL). The combined extracts were dried withNa2SO4 and concentrated in vacuo to yield crude 2.11, which was further purifiedwith CC (CHCl3 mobile phase) to yield pure 2.11.2-methyltricosanal (2.13)Mg powder was activated by stirred vigorously under N2 overnight as recom-mended in the literature [11] . Then it was suspended in freshly distilled THF (10mL). To this was added 2-bromotricosane (2.12) and the mixture was then slowlyheated and refluxed for 30 minutes. Then dimethylformamide that had been driedover activated molecular sieves was added dropwise via syringe. The mixture wasallowed to reflux for 2 hours, then it was cooled and poured over an ice cold HClsolution (1 M, 30 mL). This mixture was allowed to warm to room temperature andwas stirred overnight. Next, the mixture was extracted with Et2O (3x 25 mL), thecombined extracts were dried with Na2SO4 and concentrated in vacuo. The crudeproduct was purified with TLC (CHCl3 mobile phase), product-containing fractionswere pooled and analyzed with GC-MS to acquire characterization information forpure 2.13.2.6 Supplementary dataSupplementary data for this chapter can be found in Appendix AFig. A.1 Mass spectrum of tetracosanal (2.1)442. Branched compounds from A. thaliana waxesFig. A.2 Mass spectrum of 2-methyloctacos-4-ene (2.3)Fig. A.3 Mass spectrum of 1,22-docosandiol (2.5)Fig. A.4 Mass spectrum of docosandial (2.6)Fig. A.5 Mass spectrum of (((29-methyltriaconta-3,25-dien-1-yl)oxy)methyl)benzene (2.8)Fig. A.6 Mass spectrum of docosanal (2.10)Fig. A.7 Mass spectrum of tricosan-2-ol (2.11)Fig. A.8 Mass spectrum of 2-bromotricosane (2.12)Table A.3 Wax composition on leaves of wild-type and mutant Arabidopsis lines.Table A.4 Wax composition on leaves of wild-type and mutant Arabidopsis lines.Table A.1 Wax composition on flowers of wild-type and mutant Arabidopsis lines.Table A.2 Wax composition on flowers of wild-type and mutant Arabidopsis lines.45Chapter 3The diversity and biosynthesis ofspecialty compounds in plantcuticular waxes3.1 IntroductionAerial plant surfaces are covered with cuticular wax. Certain wax componentsare ubiquitous and are found on almost all plant species as mixtures of very-long-chain (VLC) linear aliphatic compounds. These are often present as homologousseries of alkanes, primary alcohols, aldehydes, fatty acids, and alkyl esters. Thus, twodimensions of structural variability are present in ubiquitous wax molecules: totalcarbon number (TCN) and terminal carbon oxidation state (R).Ubiquitous wax compound mixtures are complex, and comparing wax profilesfrom various species facilitates the determination of wax compound biosynthesis. Forexample, the major ubiquitous wax compounds from leaves of Arabidopsis thaliana[176] and Brassica oleracea [194] are homologous series of fatty acids, aldehydes, andalcohols with even TCNs, and alkanes with odd TCNs (Fig. 3.1A). These compoundscan be entered into a table that is defined by the two dimensions of ubiquitous waxcompound structural variability (TCN and R, Fig. 3.2). This table reveals patternsof co-occurrence and makes possible comparisons of structural characteristics withinand between the various homologous series which in turn facilitate hypotheses about463.DiversityandbiosynthesisofspecialtywaxcompoundsOSCoAOSCoAOOSCoAOHOSCoAOSCoAKCSKCRHCDECRADWSOOHOOREDFARESTOOHFrom fatty acid de novo biosynthesis initiationModification pathway products(ubiquitous wax compounds)Elongation pathway products(wax precursors)Elongation intermediatesO OSACPKARHADEARKASOSACPOH OSACPOSACPOSACPCoAACPB)OHOOHOC26H51OHOC26H51 C22H45OHOHOHC22H45C22H45OOC24H49C24H49C24H49C24H49A)C22H45OHC26H51OHOC26H51C22H45OHOHOHC22H45C22H45OOC24H49C24H49C24H49C24H49C26H51OHOWax compounds in Arabidopsis thaliana (At)Wax compounds in Brassica oleracea (Bo)Re-entry into elongation cycleEntry into elongation cycleReentry into cyclennnnnnnnnnnnnnnnFigure 3.1: Structure and biosynthesis of ubiquitous wax compounds. A) Structures of major ubiquitous waxcompounds from Arabidopsis thaliana (At) and Brassica oleracea (Bo). B) Biosynthesis of ubiquitous wax compounds.De novo fatty acid biosynthesis (green pathway), fatty acid elongation (magenta pathway) produce wax precursors (very-long-chain acyl-CoAs, dashed box), which wax modification pathways (blue) process into ubiquitous wax compounds(boxed compounds). The grey background indicates that all these compounds have been reported from plant waxes.Abbreviations: KAS = ketoacyl-ACP synthase, KAR = ketoacyl-ACP reductase, HAD = hydroxyacyl-ACP dehydratase,EAR = enoylacyl-ACP reductase, KCS = ketoacyl-CoA synthase, KCR = ketoacyl-CoA reductase, HCD = hydroxyacyl-CoA dehydratase, ECR = enoylacyl-CoA reductase, RED = reductase, AD = aldehyde decarbonylase, EST = esterase,FAR = fatty acyl reductase, WS = wax ester synthase.473. Diversity and biosynthesis of specialty wax compounds        TCN → R ↓C22 C23 C24 C25 C26 C27 C28 C29 C30 C31 C32 C33 C34 C35 C36-CH3At, BoAt, Bo-CH2OH AtAt, BoAt, BoAt, Bo-CHO At, BoAt, Bo-COOH At, Bo At At AtFigure 3.2: Tabulation of major ubiquitous wax compounds from A.thaliana and B. oleracea. Columns indicate the total carbon number (TCN)and rows indicate the terminal carbon oxidation state (R) of a possible ubiquitouswax compound. The entry of “At” (A. thaliana) or “Bo” (B. oleracea) in a cellindicates the presence of that compound in that species.the biosynthesis of these compounds.In A. thaliana and B. oleracea, the TCNs of adjacent homologs in each homol-ogous series of ubiquitous wax compounds all differ by two, most have even parity,and the functional groups these compounds bear are located exclusively on termi-nal carbons (Fig. 3.2). Such structures are highly reminiscent of long-chain fattyacids (LCFAs), which are dominated by C14, C16, and C18 homologs. In LCFA denovo biosynthesis fatty acyl-ACPs are condensed with malonyl-ACP to generate 3-ketoacyl-ACPs that are then reduced to saturated fatty acyl-ACPs (Fig. 3.1B, greenpathway), thus increasing the TCN of the original fatty acyl-ACP molecule by two.Based on the similarities between the structures of LCFAs and ubiquitous wax com-pounds, the simplest hypothesis for the biosynthesis of ubiquitous wax compoundaliphatic chains is via chemical transformations identical to those in fatty acid (FA)de novo biosynthesis.Published molecular genetics investigations confirm hypotheses about the biosyn-thesis of ubiquitous wax compound aliphatic chains derived above via analyses oftabulated wax profiles. Specifically, C18 fatty acyl-ACPs from de novo biosynthesisin the plastids are transferred to the endoplasmic reticulum where they are convertedto acyl-CoAs and enter a fatty acid elongase (FAE). In this enzyme complex, acyl-483. Diversity and biosynthesis of specialty wax compoundsCoAs are condensed with malonyl-CoA by a ketoacyl-CoA synthase (KCS) to form3-ketoacyl-CoAs (Fig. 3.1B). These are then substrate for a ketoacyl-CoA reductase(KCR) that produces 3-hydroxyacyl-CoAs. These are then used by a 3-hydroxyacyl-CoA dehydratase (HCD) in an elimination reaction that produces enoylacyl-CoAsthat in turn are reduced by an enoylacyl-CoA reductase (ECR) to produce saturatedfatty acyl-CoAs with two carbons more than the original KCS substrates. Via thiselongation cycle (Fig. 3.1B, magenta pathway), acyl-CoAs can repeatedly be extendedsuch that C18 fatty acyl-CoAs reach to between C22 and C38 or more, where they serveas wax precursors.In both A. thaliana and B. oleracea, series of aldehydes and alkanes co-occur withsimilar homolog distributions that are offset by one carbon such that an aldehydedistribution with a peak at Cn is accompanied by an alkane distribution of similarshape with a peak at Cn−1 (Fig. 3.1A). This suggests that alkanes might be biosyn-thesized from wax precursor-derived aldehydes via head group removal. The homologdistributions of primary alcohols and fatty acids (FAs) differ from one another andfrom those of aldehydes and alkanes, an observation that had led to the hypothesisthat they are derived from wax precursors via two additional pathways [43]. Finally,the co-occurrence of esters and primary alcohols and the match between chain lengthdistributions of alkyl ester-bound primary alcohols and free primary alcohols in bothspecies had suggested that alkyl esters likely arise from esterification of free primaryalcohols [118, 194].Molecular genetics experiments in model plant species largely confirm the biosyn-thetic relationships that had been described or were derived above via structuralanalyses. Specifically, in an decarbonylation pathway a reductase (RED) convertsfatty acyl-CoA precursors into aldehydes that are then available to an aldehyde de-carbonylase (AD) for the production of alkanes [24] (Fig. 3.1B, blue pathway). In theparallel acyl reduction pathway a fatty acyl-CoA reductase (FAR) generates primaryalcohols [182] that are then available to a wax ester synthase (WS) for the assemblyof alkyl esters [124]. Although the homolog distributions of ubiquitous wax com-pounds strongly suggest that a separate fatty acid-forming pathway also exists, noenzymes involved in this process have been identified, though it seems likely that a(thio)esterase (EST) may catalyze such reactions. Thus, analyses of wax profiles tab-493. Diversity and biosynthesis of specialty wax compoundsulated according to structural characteristics have utility not only in hypothesizingaliphatic chain biosynthesis, but also in predicting the biosynthesis of the differentterminal functional groups.In summary, the biosynthesis of ubiquitous wax compounds can be hypothe-sized by tabulating the wax profiles of multiple species, looking for patterns of co-occurrence, and comparing structural characteristics within and between homologousseries. Such hypotheses are confirmed by molecular genetics investigations in modelspecies and demonstrate that carbon chain formation by fatty acid elongation pre-cedes head group modification. The FAE generates fatty acyl-CoAs with even TCNsto serve as precursors with which the modification pathways then produce the ubiq-uitous wax compounds that are found on the surfaces of almost all plant species.However, species of diverse lineage in the dicots, monocots, gymnosperms, ferns,and mosses bear wax compounds other than the ubiquitous compounds describedabove. These specialty compounds can be isoprenoid in nature or bear branches intheir hydrocarbon chains, but are often found as homologous series of VLC linearaliphatics with one or more secondary oxygen functional groups instead of or in ad-dition to a terminal functional group. Thus, these specialty wax compounds includesecondary functional group position and oxidation state as additional dimensions ofstructural variability and have substantially greater structural diversity than ubiq-uitous wax compounds. Others have attempted to describe this diversity [99], butnone have done so comprehensively, recently, or using systematic searching tools andquantitative information.The installation of the secondary functional groups that set specialty compoundsapart from ubiquitous compounds may occur before, during, or after the biosynthesisof the other portions of the molecule. Based on structural analyses of specialty waxcompounds from one or a limited number of species of similar lineage some authorshad put forward possible biosynthetic mechanisms leading to these compounds. Forexample, the biosynthesis of ketones from ubiquitous alkanes by oxygenation [114]the action of PKS enzymes in the biosynthesis of β-diketones in monocots [189, 239],the elongation of intercepted ubiquitous elongation pathway intermediates for theproduction of compounds with regioselectively placed secondary functional groupsin ferns and mosses [39, 97], and the head-to-head condensation of acyl compounds503. Diversity and biosynthesis of specialty wax compoundsto produce ketones in the dicots [97]. Thus it had been postulated that secondaryfunctional groups might be installed before or after aliphatic chain and head groupbiosynthesis by oxygenation, or during aliphatic chain and head group biosynthesisby the modification of KCS- or PKS-generated intermediates or by head-to-headcondensation. However, only the oxygenation mechanism had been validated whenmolecular biological investigation implicated a P450 enzyme in specialty compoundwax biosynthesis in Arabidopsis [72], and none had been considered in the context ofa near-comprehensive specialty wax compound database.The published models for specialty wax compound biosynthesis described abovehad also suggested that ubiquitous and specialty wax compounds may be derivedfrom the same or similar biosynthetic machinery. Since ubiquitous and specialty waxcompounds had thus far been found to always co-occur and share some structuralcharacteristics, it is possible to examine potential biosynthetic overlap through struc-tural comparisons of these two groups of molecules. However, no large-scale, com-prehensive comparisons of this kind had been performed, and how such comparisonsmay influence biosynthesis hypotheses is unclear.Thus, the goals of this chapter were to generate a catalog of the specialty com-pounds reported thus far (3.2), and to assess co-occurrence and the structural charac-teristics of these compounds to improve our understanding of their biosynthesis andhow such may relate to the biosynthesis of ubiquitous wax compounds (3.3).3.2 Occurrence and diversity of specialty waxcompoundsTo assess the diversity of specialty wax compounds reported thus far, CASSciFinder was used to search for reports of mono, bi, tri, and tetrafunctional lin-ear aliphatic compounds with 17 to 52 carbons. The database was queried usingmolecular formulas, and to narrow the scope of returned references only compoundsfrom biological sources were considered. This produced a preliminary list of ca. 350references. Some of these articles were excluded as they described compounds fromanimals, insects, or bacteria (usually < C20), or the contents of geological samples.Others described VLC compounds with unsaturation(s) in the carbon chain, but no513. Diversity and biosynthesis of specialty wax compoundssecondary oxygen functional groups and were thus also excluded. This left ca. 120references that together described the surface wax of ca. 90 specialty wax compound-bearing plant species. A list of these species is provided as a supplement (Table B.1).To present and organize the specialty compounds (data) such that co-occurrenceand structural characteristics might be evaluated, the data were organized into one ofthree sets of tables, each set containing compounds with one, two, or three secondaryfunctional groups, respectively. Each table in each set was defined by the dimensionsof structural variability exhibited by specialty wax compounds. Thus, columns ineach table indicated the TCN of each compound and were grouped according to theoxidation state(s) of the compounds secondary functional group(s). Rows in eachtable indicated the carbon numbers on which the functional groups were found andwere grouped by terminal carbon oxidation state. For example, if a species (Genusspecies = “Gs”) contained a homologous series of C24, C26, C28, and C30 3-ketoacids,an isomeric mixture of C24 ketoacids with 1,6-; 1,7-; 1,8-; and 1,9-geometry, and asecond isomeric mixture of C28 ketoacids with 1,7-; 1,9-; 1,11-; and 1,13-geometry, thetable for specialty wax compounds with one secondary functional group would receivetwelve total entries of “Gs” in squares corresponding to the structural characteristicsof these compounds (Fig. 3.3).To facilitate the comparison making process, homologous series or isomeric mix-tures of specialty compounds were color coded according to the isomer characteristicsexhibited by their secondary functional groups. For this, a color coding and nomen-clature system was devised based on the structures of series and mixtures of specialtycompounds found in the SciFinder searches. Series and mixtures of specialty com-pounds were first categorized according to the parity of the carbons on which sec-ondary functional groups were found. Series and mixtures whose members secondaryfunctional groups appeared on both even- and odd-numbered carbons (EO), onlyeven-numbered carbons (E), or only odd-numbered carbons (O) were given distinctcolor and letter assignments (Fig. 3.4A). Each of these categories were further sub-divided according to whether their members secondary functional groups appearednear the terminus of the molecule (C-X, X ≤ 3, (a)), or in the central portion of themolecule (C-X, X > 3, (b)). Homologous series or isomeric mixtures whose membershad more than one functional group were described by combining the nomenclature523. Diversity and biosynthesis of specialty wax compounds       TCN→       C-X ↓ C18 C20 C22 C24 C26 C28 C30 C32 C341,21,3 Gs Gs Gs Gs1,41,51,6 Gs1,7 Gs Gs1,8 Gs1,9 Gs Gs1,101,11 Gs1,121,13 Gsoxofatty	acidsC12H25 COOHC12H25 COOHC12H25 COOHC12H25 COOHOOOOC12H25 COOHCOOHCOOHCOOHOOOOC12H25C12H25C12H25COOHCOOHCOOHCOOHC12H25C12H25C12H25C12H25OOOOFigure 3.3: Catalog of example specialty wax compounds. A species initials(e.g. Genus species = “Gs”) are entered once for each specialty wax compound foundin that species and in the cell corresponding to the structural characteristics of thatspecialty wax compound. Columns indicate the total carbon number (TCN) andare grouped by secondary functional group oxidation state. Rows indicate functionalgroup positions (C-X) and are grouped by terminal carbon oxidation state. Entriesare color coded to highlight members of the same homologous series or mixture ofisomers. The structures shown here are theoretical examples chosen as an example.533.DiversityandbiosynthesisofspecialtywaxcompoundsA)	function	position	-> C-n,	n	≤	3 C-n,	n	>	3 C-n,	n	≤	3 C-n,	n	>	3 C-n,	n	≤	3 C-n,	n	>	3EO(a) EO(b) O(a) O(b) E(a) E(b)B)EO(b)EO(b) E(b)E(b) O(b)O(b) EO(b)E(b) EO(b)E(b)E(b) O(b)O(a)function	parity	->Mixed	parity Single	parityeven	and	odd only	evenonly	oddROHRROHOHRRRRRRRRRROHOHOHOHRROHOHROHRROHOHRRROHOHOHRRROHOHOHOHOHOHOHOHOHOHOHOHOHOHOHOHOHOHRRROHOHOHOHOHOHROH OHROH OHROH OHOHOHOHRRROHOHOHOHOHOHRROHOHRROHOHROHROHRROHOHRRROHOHOHRRROHOHOHOHOHOHROH OHROH OHROH OHOHOHOHRRROHOHOHOHOHOHROHRROHOHRRROHOHOHRRROHOH OHOHOHOHOHOHOHOHOHOHnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnFigure 3.4: (Continued on the following page.)543. Diversity and biosynthesis of specialty wax compoundsFigure 3.4: Color coding and nomenclature for specialty wax compounds.Using the structures of specialty wax compounds reported in literature a scheme fordistinguishing homologous series or isomeric mixtures of specialty wax compoundsfrom each other was developed. A) Homologous series or isomeric mixtures of spe-cialty compounds were first distinguished according to whether their functional groupswere found on carbons with mixed or single parity (functions on even- and odd-numbercarbons, e.g. C-14, C-15, C-16 (EO); functions on only odd-numbered carbons; e.g.C-7, C-9, C-11 (O); or functions on only even-numbered carbons, e.g., C-4, C-6, C-8(E)). Next, series or mixtures were distinguished according to whether their mem-bers secondary functional groups were present near the terminus (C-X, X ≤ 3 (a))or towards the middle (C-X, X > 3 (b)) of the aliphatic carbon chain. Colors wereassigned to each category distinguished by these criteria, and theoretical exampleisomers (on one side of each dotted line) and homologs (homologs on opposite sidesof dotted lines) are shown for each category. B) Nomenclature was assigned to se-ries or mixtures of specialty compounds whose members contained more than onesecondary functional group by combining the nomenclature for single secondary func-tional groups presented in A). For series or mixtures where the additional secondaryfunctional group was a repeat of the first, new colors were assigned. For series ormixtures where the additional secondary functional group(s) were different from thefirst, colors for the two distinct types of secondary functional groups were mixed. R= COOR’, COOH, CHO, CH2OH, CH3.and coloration defined for single functional groups (Fig. 3.4B). Thus, in the exampleof the ketoacids from Genus species, each table entry received coloration to distin-guish the three different series or mixtures, and the three would be referred to ascompounds with O(a), EO(b), and O(b) secondary functional groups, respectively.Using the criteria described above, the mixtures of specialty wax compounds re-ported in the articles found via the SciFinder searches were tabulated and color coded.Here, the tabulated homologous series and isomeric mixtures of specialty compoundswill be briefly described beginning with those that contained one (3.2.1), and thentwo secondary functional groups (3.2.2), and finally three or more (3.2.3) secondaryfunctional groups.553. Diversity and biosynthesis of specialty wax compounds3.2.1 Specialty compounds with a single secondary oxygenfunctional groupHomologous series of specialty compounds have been reported whose membershave one secondary oxo or hydroxyl function on C-2 (an even number) of their car-bon chains, thus E(a) functional groups. Members of these classes had predominantlyodd TCNs between C17 and C37 and have been reported as surface wax componentson flowering plants such as Aloe arborescens [170], Arabidopsis thaliana [69], Cynomo-rium songaricum [257], Laurus nobilis [65], Phragmites australis and Juncus effusus[128], Prunus domestica [130], Solanum tuberosum [207], and in esterified form onSorghum bicolor [241] and the gymnosperm Pinus radiata [62] (Fig. 3.5A). Thesecompounds have also been found in many plant essential oils (C17 and C19) (Fig. 3.5A,Many O), as aphrodisiac pheromones on the wings of Pieris butterflies (C29) [253], asnest identity signals from bumblebees (C29) [181], and from many other insect sources(Fig. 3.5A, Many I).Specialty wax compound series and mixtures have also been found whose membershad a single secondary oxo or hydroxyl group on secondary carbons other than C-2,and instead on even- and odd-numbered carbons, for example a mixture of nonacosan-6-one, nonacosan-7-one, nonacosan-8-one, nonacosan-9-one, and nonacosan-10-one,thus EO(b) functional groups. Members of these compound classes typically had oddTCNs spanning C23 to C33 and have been reported as homologous series and/or isomermixtures in the surface waxes of the flowering plants Arabidopsis thaliana [81, 136,248], Berberis aquifolium [23], several Brassica species1 [55, 121, 153, 194], Clematisvitalba [228], Eunymus latifolius [229], Foeniculum vulgare [144], several Fragaria spp.[8], Fumaria parviflora [150], Hibiscus rosasinensis [196], Hymenocallis littoralis [1],Malus domestica [232], Pisum sativum [127], Rosa virgo [142], and Solanum tuberosum[207] (Fig. 3.5A). Compounds with these structural characteristics have also beenfound on butterfly (C27 C31) [253] and wasp wings (C23) [90].Compounds in still other series or mixtures of monofunctional specialty wax com-pounds bear a secondary oxo or hydroxyl group located almost exclusively on even-numbered carbons, for example a mixture of hentriacontan-8-one, hentriacontan-1entered into the table as BR, with two capital letters to denote the finding of correspondingcompounds in multiple species563.Diversityandbiosynthesisofspecialtywaxcompounds		 	         TCN→       C-X ↓ C17 C19 C21 C23 C25 C27 C29 C31 C33 C35 C37 C17 C19 C21 C23 C25 C27 C29 C31 C33 C35 C37At, SO At, SO Aa, At, SOPd, Dm Bo-I, PI1-I Sb*3 Ac1, Pt Ac1 Hm Hm4 Ia Ac1 RO RO RO PI2, RO RO5 Sc Sm La* RO PI1-I, Md1, RO RO RO RO6 Ns PI1-I Ci Hh-I* RO PI1-I, RO PI1-I, RO RO RO7 PI1-I  Hh-I* PI1-I, Er, Pd PI1-I, Pd NnPI1-I Md1, PI1-I, Er, PdSO Af9Hr + many OCU PI1-I, RO Hh-I*PI2, PI1-I, Md1, Er, PdPI1-I, Md1, Er, Pd FR, ErPI1-I PI1-I PI1-I, Md1, Er, PdPI1-I, Md1, Ba, PI2, PI2*, ErSO, Or, EN, FpRO, Or, Af, Fp Fv, El, CvPo1, Nn, Pa2, Tb, Pb, Ec3, PA, Pa1, Js11 PI1-I PI1-I, Md1, Ba  FR, PI1-I, Er FR, PI1-IBR FR, Ps2  FR, PI1-IRO, AfTbBR, AtHrBR, At BR, At FR, Ps215 BR, PI1-I, At At BR, At, Ps2 At, Ps2 Ps2SO, Hl, VaAt, Po2, Ps2EU-O, As, Ap, GbVa17A)SOSOSO AfAt, Ps2BR, At, Ps2HlROROOrAtTh, Po2At  FR, PI1-I, Ps2Ps2At, Ps2BR, AtOr, EN PI1-I, FR, Eroxo hydroxylAt, SO many OLn + many OPI2PI3PI1-I, Er, PdCs3*Ps2 AfR = CH32 many Omany ISO AtSOSOSc1316 SOFc-I* Hh-I*SOSO, Or SO, Or81214Aa Aa, SO Aa, At, SO Aa, At At10 CU, Tp Ns SO, OrSO, Ormany OBR, PI1-I, FRFigure 3.5: (Continued on the following page.)573.Diversityandbiosynthesisofspecialtywaxcompounds         TCN→       C-X ↓ C18 C20 C22 C24 C26 C28 C30 C32 C34 C36 C38 C18 C20 C22 C24 C26 C28 C30 C32 C34 C36 C381,2 Cb Cb CbCb Cb CbFh* Rc, Fh* Rc, Fh*1,41,5 Tb Fh, Tb Tb Tb Tb1,61,7 Or Tg* Fh1.81,9 Or Or TR* Fh Mg21,10 TR*1,11 Or, Or* Or, Or* Or, Or* TR* Or Mg2, Or, Af(*) Mg2, Or Mg21,12 TR* Ps21,13 Or Or Or Ps2 Af(*), Mg21,14 Cm2-A Ps21,15 Or Hm Mp Ps2 NC-A Af(*)1,16 Ps21,17 Ps2 Af(*)1,18 Ln* Tg*1,3 PA, Rc PA, Rc Rc Rc1,4 La1,5 Tb Tb Tb Tb Tb Tb Tb1,7 PA1,9 PA1,11 Or Or PA1,13 Or Or1,15 OrCb Cb CbMg1 Mg1 Mg11,3 Cb Cb Cb Aa Aa Aa1,5 Cm1* Cm1* Cm1*  Cm1*1,71,9 Al Am-O Sd-S**1,11 Am-O Am-O NC-A1,13 Sd-S**1,15 Cs2 Sd-S** NC-A1,17 NC-AAaSA, TR1,11 Af1,13 Af1,15 Sd-S** Af1,17 AfPs1 Rc, Fh* Rc, FhR	=	CHOR	=	CH2OH1,3Wt, Cs1-A, Mr, Mb, Pp**Aa AaR	=	COOH1,2hydroxylB) Aa AaoxoR	=	COOMe 1,3 At At AtFigure 3.5: (Continued on the following page.)583. Diversity and biosynthesis of specialty wax compounds						Fig.	5:	Catalog	of	specialty	wax	compounds	with	one	secondary	functional	group.	A	species’	initials	(e.g.	Genus	species	=	“Gs”)	are	entered	once	for	each	specialty	wax	compound	found	in	that	species	and	in	the	cell	corresponding	to	the	structural	characteristics	of	that	specialty	wax	compound.	Columns	indicate	the	 total	 carbon	 number	 (TCN)	 and	 are	 grouped	 by	 secondary	 functional	 group	 oxidation	 state.	 Rows	indicate	functional	group	positions	(C-f)	and	are	grouped	by	terminal	carbon	oxidation	state.	Entries	are	color	coded	to	highlight	members	of	the	same	homologous	series	or	mixture	of	isomers.	A)	compounds	with	 CH3	 terminal	 carbons,	 B)	 compounds	 with	 oxygen-bound	 terminal	 carbons.	 C)	 Color	 coding	 and	nomenclature.	Compounds	marked	with	an	asterisk	were	reported	in	esterified	form	and	those	with	two	asterisks	 had	 a	 methoxy	 secondary	 functional	 group.	 Note	 that	 row	 numbering	 is	 not	 necessarily	continuous.	         TCN→       C-f ↓ C17 C19 C21 C23 C25 C27 C29 C31 C33 C35 C37 C17 C19 C21 C23 C25 C27 C29 C31 C33 C35 C37At, SO At, SO Aa, At, SOPd, Dm Bo-I, PI1-I Sb*3 Ac1, Pt Ac1 Hm Hm4 Ia Ac1 RO RO RO PI2, RO RO5 Sc Sm La* RO PI1-I, Md1, RO RO RO RO6 Ns PI1-I Ci Hh-I* RO PI1-I, RO PI1-I, RO RO RO7 PI1-I  Hh-I* PI1-I, Er, Pd PI1-I, Pd NnPI1-I Md1, PI1-I, Er, PdSO Af9Hr + many OCU PI1-I, RO Hh-I*PI2, PI1-I, Md1, Er, PdPI1-I, Md1, Er, Pd FR, ErPI1-I PI1-I PI1-I, Md1, Er, PdPI1-I, Md1, Ba, PI2, PI2*, ErSO, Or, EN, FpRO, Or, Af, Fp Fv, El, CvPo1, Nn, Pa2, Tb, Pb, Ec3, PA, Pa1, Js11 PI1-I PI1-I, Md1, Ba  FR, PI1-I, Er FR, PI1-IBR FR, Ps2  FR, PI1-IRO, AfTbBR, AtHrBR, At BR, At FR, Ps215 BR, PI1-I, At At BR, At, Ps2 At, Ps2 Ps2SO, Hl, VaAt, Po2, Ps2EU-O, As, Ap, GbVa17A)SOSOSO AfAt, Ps2BR, At, Ps2HlROROOrAtTh, Po2At  FR, PI1-I, Ps2Ps2At, Ps2BR, AtOr, EN PI1-I, FR, Eroxo hydroxylAt, SO many OLn + many OPI2PI3PI1-I, Er, PdCs3*Ps2 AfR = CH32 many Omany ISO AtSOSOSc1316 SOFc-I* Hh-I*SOSO, Or SO, Or81214Aa Aa, SO Aa, At, SO Aa, At At10 CU, Tp Ns SO, OrSO, Ormany OBR, PI1-I, FR         TCN→       C-f ↓ C18 C20 C22 C24 C26 C28 C30 C32 C34 C36 C38 C18 C20 C22 C24 C26 C28 C30 C32 C34 C36 C381,2 Cb Cb CbCb Cb CbFh* Rc, Fh* Rc, Fh*1,41,5 Tb Fh, Tb Tb Tb Tb1,61,7 Or Tg* Fh1.81,9 Or Or TR* Fh Mg21,10 TR*1,11 Or, Or* Or, Or* Or, Or* TR* Or Mg2, Or, Af(*) Mg2, Or Mg21,12 TR* Ps21,13 Or Or Or Ps2 Af(*), Mg21,14 Cm2-A Ps21,15 Or Hm Mp Ps2 NC-A Af(*)1,16 Ps21,17 Ps2 Af(*)1,18 Ln* Tg*1,3 PA, Rc PA, Rc Rc Rc1,4 La1,5 Tb Tb Tb Tb Tb Tb Tb1,7 PA1,9 PA1,11 Or Or PA1,13 Or Or1,15 OrCb Cb CbMg1 Mg1 Mg11,3 Cb Cb Cb Aa Aa Aa1,5 Cm1* Cm1* Cm1*  Cm1*1,71,9 Al Am-O Sd-S**1,11 Am-O Am-O NC-A1,13 Sd-S**1,15 Cs2 Sd-S** NC-A1,17 NC-AAaSA, TR1,11 Af1,13 Af1,15 Sd-S** Af1,17 AfPs1 Rc, Fh* Rc, FhR	=	CHOR	=	CH2OH1,3Wt, Cs1-A, Mr, Mb, Pp**Aa AaR	=	COOH1,2hydroxylB) C)Aa AaoxoR	=	COOMe 1,3 At At AtEO(a) EO(b) O(a) O(b) E(a) E(b)EO(b)EO(b) E(b)E(b) O(b)O(b) EO(b)E(b) EO(b)E(b)E(b) O(b)O(a)Figure 3.5: Catalog of specialty wax compounds with one secondary func-tional group. A species initials (e.g., Genus species = “Gs”) are entered once foreach specialty wax compound found in that species and in the cell corresponding tothe structural characteristics of that specialty wax compound. Columns indicate thetotal carbon number (TCN) a d are grouped by secondary functional group oxidationstate. Rows indicate functional group positions (C-X) and are grouped by terminalcarbon oxidation state. Entries are color coded to highlight members of the samehomologous series or mixture of isomers. A) compounds with CH3 terminal carbons,B) compounds with oxygen-bound terminal carbons. C) Color coding and nomen-clature. Compounds marked with an asterisk were reported in esterified form andthose with two asterisks had a methoxy secondary functional group. Note that rownumbering is not necessarily continuous. A full list of the species names and theircorresponding two-letter codes is provided in Table B.1.10-one, hentriacontan-12-one, hentriacontan-14-one, and hentriacontan-16-one, thusE(b) functional groups. Members of this class usually had odd TCNs between C27and C35, and have been found on Rosa canina [36], and Solanum tuberosum [207],and the fern Osmunda regalis [102] (Fig. 3.5A).In some cases compounds with secondary functional groups on even-numberedcarbons were found, but no accompanying homologs or isomers were reported tooccur with them, such as nonacosan-10-ol from Hymenocallis littoralis [1], Nelumbonucifera [56], Prunus avium [162], Rosa canina [36], the gymnosperms Picea omorika[156], several Pinus species [77], and the ferns Azolla filliculoides [30] and Osmundaregalis [102]. Other examples include the symmetrical ketones and secondary alcoholsfound with mainly odd TCNs (C17 to C31) on surfaces of Allium ampeloprasum [131],Annona squamosa [192], Evolvulus alsinoides [199], Ginkgo biloba, Magnoflia gran-diflora, and Liriodendron tulipifera [74], Hibiscus rosasinensis [196], Hymenocallislittoralis [1], Piper betle [52], Pisum sativum [127], Platanus orientale [47], Thesiumhumile [67], and Tridax procumbens [233] (Fig. 3.5A).Many investigators have encountered wax molecules with one terminal and one593. Diversity and biosynthesis of specialty wax compoundssecondary function. These compounds exhibited vast functional group diversity owingto the numerous oxidation states that a terminal oxygen group may occupy. Fattyacids, methyl esters, primary alcohols, and aldehydes with secondary methoxy, oxo,and hydroxyl groups have all been reported. Some of these were found as mixturesof compounds with 1,2 and 1,3 functional group geometry, thus EO(a) functionalgroups, as have been reported as diols from Cosmos bipinnatus [38] (Fig. 3.5B).More frequently compounds with one terminal and one secondary functional grouphave been reported as homologous series of compounds with 1,3 functional groupgeometry, thus O(a) functional groups, with almost exclusively even TCNs rangingfrom C20 to C32. These have been reported from Aloe arborescens [170], Cosmosbipinnatus [35], Funaria hygrometrica [39], several Papaver species [100], Pegolettiasenegalensis [27], and several Salvia species [2] (Fig. 3.5B). The seed oil of severalWrightia tinctoria (C18) [146], the volatiles of Minuartia recurva (C18) [105], volatilesof Melicocca bijuga fruit (C18) [164], and the microalga Chlorella sorokiniana (C18)[250] also have been found to contain these compounds.Other mixtures of specialty compounds with one terminal function and one sec-ondary function had geometries such as 1,5-; 1,7-; 1,9-; etc., thus secondary functionalgroups on odd-numbered carbons greater than C-3, thus O(b) functional groups.These compounds almost always had even TCNs between C24 and C36, and werefound on surfaces of Cerinthe minor [101], Myricaria germanica [97], several Papaverspecies [100], the gymnosperm Taxus baccata [247], the ferns Osmunda regalis [102]and Azolla filiculoides [30], and the moss Funaria hygrometrica [39] (Fig. 3.5B). Theywere also found in Chlamydomonas monoica (C24) [26], Argemone mexicana seed oil(C28 and C30) [76], species of Schizymenia red algae (C22 and C24) [13], and speciesof Nannochloropsis microalga (C32 and C34) [66].In two cases compounds with one terminal function also contained one secondaryfunction found on both even- and odd-numbered carbons when considering all themembers of the mixture together, taking on geometries such as 1,11-; 1,12-; 1,13-;1,14; etc., thus EO(b) functional groups. These compounds have been reported asC28 diols from Pisum sativum [245], and as esterified C26 diols from Triticum aestivum[223] (Fig. 3.5B).Thus, the structural diversity of monofunctional specialty wax compounds is con-603. Diversity and biosynthesis of specialty wax compoundssiderable. Three sub-groups have been encountered most frequently. First amongthese are C27 and C29 14- and 15-secondary alcohols and ketones, 10-nonacosanol and10-nonacosanone, and hentriacontan-16-one. These have been found in diverse plantspecies including angiosperms, gymnosperms, and ferns. Notably, many of these formepicuticular wax crystals [112].3.2.2 Specialty compounds with two secondary functionalgroupsMembers of some specialty wax compound mixtures had two secondary oxo orhydroxyl groups that were either adjacent or separated by one carbon (alternating),thus EO(b) functional groups. These compounds usually had odd TCNs betweenC27 and C31 and have been found on Arabidopsis thaliana [248], several Brassica spp.[88, 194], as well as on the wings of Pieris butterflies (C27 - C31) [253] (Fig. 3.6A).Other mixtures of specialty compounds were found that had members with twosecondary oxo or hydroxyl groups always separated by one carbon (alternating) andalmost exclusively on even-numbered carbons of the chain, thus two E(b) functionalgroups that together are referred to here as E(b)E(b) functional groups. Thesecompounds spanned a wide range of mainly odd TCNs from C19 to C37 and havebeen found Buxus sempervirens [49, 138], Carthamus tinctorius [4, 3], Chrysanthe-mum segetum [138], Eucalyptus globulus [158, 201, 202], Myricaria germanica [97],and several Rhododendron species [58] (Fig. 3.6A). Many reports indicate that grassspecies also frequently include these compounds in their waxes, including Agropyronspecies [221, 219], Andropogon species [225], Eragrostis curvula [220], Elymus species[226], Hordeum vulgare [93, 139], Leymus arenarius [138], Panicum virgatum [226],Triticum species [41, 215, 216], and Secale cereale and Triticale hexaploide [224], aswell as in sunflower pollen (C19 - C31) [189, 227].There are also some reports of mixtures of bifunctional compounds whose memberscontained one secondary functional group always found on the same, even-numberedcarbon, and another secondary functional group in a position that varied greatly be-tween isomers, being found on both even- and odd-numbered carbons when examiningall members together. Therefore, these compounds had mixed E(b)EO(b) functionalgroups. These were found as C29 diols in Cupressus species [165], Encephalartos [159],613.DiversityandbiosynthesisofspecialtywaxcompoundsC19 C21 C23 C27 C29 C31 C33 C35 C37 C19 C21 C23 C25 C27 C29 C31 C33C35C37 C19 C21 C23 C25 C27 C29 C31 C33 C35 C372,	37,	8 PI1-I PI1-I7,	9 PI1-I PI1-I8,	9 PI1-I PI1-I8,	10 PI1-I PI1-I9,	10 PI1-I PI1-I PI1-I9,	11 PI1-I PI1-I10,	11 PI1-I PI1-I PI1-I10,	12 PI1-I11,	12 PI1-I11,	1312,	13 Bv BR12,	14 PI1-I13,	14 BR13,	15 PI1-I At, BR14,	15 At At, BR14,	16 At At, BR2,	4    4,	6 Ha Ha Ha Ha Ha Ha Ha Ha Ha Ha Ha6,	8 Ct Ct Ct Ct Ct Ct Ct Cs2 Ha Ha Ha Ha BsAN, Hv, Ec2Ec1, PvEU, AN, Hv, Ec2, RHHl, AG, Tr, TR, Fo, Va16,	18 EU EU, Hv EUA)Cm2-AMg2Ct14,	16R = CH3alternating C-X 8,	1010,	1212,	14 Mg2Mg2Md2-I, ANMmNtC25Hv, ANRH, Hadihydroxy hydroxyketo diketoBsHa, Nt, AN, RHAG HaNt,  HaCtMg2, CtMg2, CtMg2Mg2, Ct           TCN→     C-X ↓NtMg2Mg2Mg2alternating and adjacent C-XBsNt, RHMg2Mg2Mg2, Ct HaMg2, CtFigure 3.6: (Continued on the following page.)623. Diversity and biosynthesis of specialty wax compounds	 			 	EO(a) EO(b) O(a) O(b) E(a) E(b)EO(b)EO(b) E(b)E(b) O(b)O(b) EO(b)E(b) EO(b)E(b)E(b) O(b)O(a)hydroxyketo diketo        TCN →   C-X ↓ C21 C23 C25 C27 C31 C33 C35 C37 C29 C293, 104, 10 Nn6, 107, 10 Nn, Tf8, 109, 10 Ha Ha Ha10, 11 Or Or10, 12 Or10, 1310, 1410, 1510, 162, 12 Mg23, 12 Mg24, 12 Mg25, 12 Mg26, 12 Mg27, 12 Mg28, 12 Mg29, 1210, 1211, 1212, 13 Mg212, 14 Mg212, 15 Mg212, 16 Mg2 Mg212, 17 Mg212, 18 Mg2C20 C22 C24 C34 C36 C38MA-O MA-O MA-OHaB)R = CH3dihydroxyPI2, Nn, TfPI2, Nn5, 10           TCN →     C-X ↓15,173,7dihydroxyC29DmHa, Mg2, Pa1, NnHa, PI2, Pa1, EN, Js, Pa1, NnPI2, Ha, EN, Pa1, Js, Co, NnOrOrHa, NnHa, TfMg2PI2, EN, Pa1, JsPI2R=COOHC)D)diketoFigure 3.6: (Continued on the following page.)633. Diversity and biosynthesis of specialty wax compoundsFigure 3.6: Catalog of specialty wax compounds with two secondary func-tional groups. A species initials (e.g., Genus species = “Gs”) are entered oncefor each specialty wax compound found in that species and in the cell correspondingto the structural characteristics of that specialty wax compound. Columns indicatethe total carbon number (TCN) and are grouped by secondary functional group ox-idation state. Rows indicate functional group positions (C-X) and are grouped byterminal carbon oxidation state. Entries are color coded to highlight members of thesame homologous series or mixture of isomers. A) Homologous series or mixtures ofisomers with CH3 terminal carbons are grouped according to whether their memberssecondary functional groups were found on alternating and adjacent carbons (alter-nating and adjacent C-X) or only on alternating carbons (alternating C-X). B) Ho-mologous series or isomer mixtures with CH3 terminal carbons and functional groupsin varying positions. C) Homologous series or isomer mixtures with oxygen-boundterminal carbons. D) Color coding and nomenclature. Compounds marked with anasterisk were reported in esterified form. Note that row numbering is not necessarilycontinuous. A full list of the species names and their corresponding two-letter codesis provided in Table B.1.Juniperus scopulorum [222], Myricaria germanica [97], Nelumbo nucifera and Tropae-olum majus [110], Picea abies [161], several Pinus species [63, 77], and Taxus baccata[246], and as diols, ketols, and diketones on Osmunda regalis [102] (Fig. 3.6B). Sim-ilar C31 diols have also been reported from Myricaria germanica [97]. In some casesmixtures of compounds bearing E(b)EO(b) functional groups had to be carefullydistinguished from being those with E(b)E(b) functional groups (e.g., Osmundaregalis diols, ketols, diketones), however, in each of these cases the mixtures wereunambiguously classified by considering all member homologs and isomers togetherin the context of accompanying compounds. Further examples of VLC aliphatic com-pounds with two secondary functional groups can be found outside plant waxes. AC36 diketoacid with secondary keto groups on odd-numbered carbons of backboneswith even TCNs spanning C26 to C28 ((E(b)E(b))) have been reported from He-lianthus annuus pollen [189] (Fig. 3.6C). In addition, dihydroxy acids with hydroxylgroups on odd-numbered carbons of backbones with even TCNs between C20 and C24(O(b)O(a)) were encountered the zygospore cell walls of Chlamydomonas monoica[26], the floral oils of Malpighia species [190], and the floral oils collected by bees [174](Fig. 3.6C).Overall, specialty wax compounds with two secondary functional groups have643. Diversity and biosynthesis of specialty wax compoundsC29 C31 C33 C35 C29 C31 C33 C35 C29 C31 C33 C35AN, AN, Fo Ec17 - O AG8 - OH TR9 - OH TR10 - O AG25 - OH Ec2, TR, AG Ec225 - O AG26 - OH AG14,16-diketo 16,18-hydroxyketo12,14-diketo        TCN →           C-X ↓R = CH35 - OHFigure 3.7: Catalog of specialty compounds with three secondary functionalgroups. A species initials (e.g., Genus species = “Gs”) are entered once for eachspecialty wax compound found in that species and in the cell corresponding to thestructural characteristics of that specialty wax compound. Columns indicate thetotal carbon number (TCN) and are grouped by paired secondary functional groupoxidation states and positions. Rows indicate third functional group position (C-X)and oxidation state and are grouped by terminal carbon oxidation state. Entrieswere color coded to highlight members of the same homologous series or mixtureof isomers. Colored cells in this figure indicate compounds with E(b)E(b)EO(b)functional groups. Note that row numbering is not necessarily continuous. A full listof the species names and their corresponding two-letter codes is provided in Table B.1.been found in a variety forms and on many species. The most frequently reportedamong these are the α-ketols from the Brassicas and the β-diketones. Both of thesecompound classes have also been shown to participate in the formation of epicuticularwax crystals that successfully create superhydrophobic surfaces, which may in partexplain why they are found on the surfaces of diverse plant species [112].653. Diversity and biosynthesis of specialty wax compounds3.2.3 Specialty compounds with three or more secondaryfunctional groupsLastly, there have been some reports of specialty wax compounds that bore threeor more functional groups. Hydroxy- and oxo-β-diketones have been found, usuallyalongside β-diketones, with the β-keto groups almost always on odd-numbered car-bons and with the hydroxyl or oxo group in various positions, including even- andodd-numbered carbons of the backbone, and thus had two E(b) functional groups andone EO(b) group (together E(b)E(b)EO(b)). These compounds have been foundwith odd TCNs C31 and C33 on Agropyron species [221, 219, 218], several Andro-pogon spp. [225], Eucalyptus [158], Festuca ovina [226], and seeds of Hordeum vulgare[85] (Fig. 3.7). Roots of Corallocarpus epicaeus [220] also bore these compounds. Astructurally related tetrafunctional compound, 4-hydroxy-25-oxohentricontane-14,16-dione, has been reported from Agropyron elongatum [221].Overall, not very many examples of specialty wax compounds with three or morefunctional groups have been encountered so far. However, hydroxy β-diketones havebeen reported many times and the structural diversity within that compound class isrelatively well documented.3.3 Biosynthesis of specialty wax compoundsThe second goal of this chapter was to assess the structural characteristics ofcooccurring specialty compounds to determine how they might be biosynthesized, andhow such hypotheses and processes might relate to or be influenced by ubiquitous waxcompound biosynthesis. The occurrence and structural characteristics of ubiquitouswax compounds revealed biosynthetic connections between ubiquitous classes andshed light on the biosynthesis pathways that lead to them (Fig. 3.1). Here, theoccurrence and structural characteristics of the various specialty wax compoundsencountered thus far will be analyzed similarly, beginning with homologous series orisomer mixtures whose compounds have secondary functional groups on carbons ofmixed parity (EO, 3.3.1), then those bearing single secondary functional groups oncarbons of single parity near the chain terminus ((a), 3.3.2), followed by those witha single secondary functional groups on carbons of single parity near the middle of663. Diversity and biosynthesis of specialty wax compoundsthe carbon skeleton ((b), 3.3.3), and finally those that have two secondary functionalgroups on carbons of single parity near the middle of the carbon skeleton (3.3.4).3.3.1 Specialty compound classes with secondary functionalgroups on carbons with mixed paritySpecies that produced compounds with EO(b)EO(b) functional groups (α- andβ-diols, ketols, and diketones, as found on Arabidopsis thaliana [136], Brassica napusand Brassica oleracea [55], and Osmunda regalis [97]), also produced compounds bear-ing a single EO(b) functional group (mid-chain secondary alcohols and ketones, cf.Fig. 3.5A and Fig. 3.6A). Such cooccurrence indicates that the processes by which thefunctional groups these compounds bear are installed are probably related and thatneither process is highly regioselective. Genetic tools have been used to make a de-tailed investigation of the biosynthesis of compounds with EO(b) and EO(b)EO(b)functional groups from Arabidopsis thaliana. A P450 enzyme called MAH1 was iden-tified that produces isomeric mixtures of secondary alcohols and ketones (compoundswith EO(b) groups) from ubiquitous alkanes and can also perform further oxida-tion reactions to generate α- and β-diols and ketols (compounds with EO(b)EO(b)groups) [72, 248] (Fig. 3.8A). These studies confirm the relatively poor regioselectiv-ity of a P450 enzyme known to be involved in specialty wax compound biosynthesis,and demonstrate that these enzymes may repeatedly oxidize aliphatic chains, eitheron single or multiple carbons.Since P450s are found in virtually all plant genomes it is reasonable to hypothe-size the involvement of these enzymes in the installation of all functional groups that,within a series or isomer mixture, are found on carbons of mixed parity (both even-and odd-numbered carbons). For example, the functional groups that, between ho-mologs and isomers, are present carbons of mixed parity on the diols, ketols, hydroxy-,and oxo-diketones in Tables 3.6B and 3.7 (summarized in Fig. 3.8B and C, orangefunctional groups) are probably installed by a P450 enzyme, as had been speculatedby other authors [63]. Similarly, the secondary functional groups present on 1,2- and1,3-diols may also be installed by a P450 enzyme as they appear on both even- andodd-numbered carbons, though this could also happen via hydration of an α,β doublebond (Fig. 3.8D, yellow functional groups). Though it is conceivable that some of673.DiversityandbiosynthesisofspecialtywaxcompoundsOHOHOHOHOHOH OHOHOHOHOHOHOHP450OHP450P450P450 ?OHOHOHOHOHEO(a)EO(b)EO(b)EO(b)P450OHOHP450OHOH OHOH OHOH OHOHOHOHOHOHOHOHOHOH OHOH OHOHOHOHOHE(b)EO(b)E(b)E(b)EO(b)OHOHOH OHOHOHOHOHP450OHP450OH OH OHA)D)B)C)OHOHOHOHOH + H2OnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnFigure 3.8: Biosynthesis of specialty wax compounds with functional groups on carbons of mixed parity.Biosynthesis of homologous series or isomeric mixtures of very-long-chain specialty wax compounds whose memberssecondary functional groups are on carbons of mixed parity (even- and odd-numbered carbons) has been demonstratedto be carried out by a P450 enzyme in Arabidopsis. In other species such enzymes may also install functional groupson specialty compounds with similar structural characteristics, including A) other secondary alcohol, ketones, or diols,ketols, or diketones with functional groups on carbons of mixed parity in the center of the molecule (EO(b) andEO(b)EO(b)), and the functional groups found on carbons of mixed parity in the presence of B) one (E(b)EO(b)) orC) two (E(b)E(b)EO(b)) additional functional groups found on carbons with single parity. D) Biosynthesis of seriesor mixtures of compounds with secondary functional groups on carbons of mixed parity near the chain terminus mayproceed via P450 oxidation (left), or possibly by hydration of an α,β double bond (right).683. Diversity and biosynthesis of specialty wax compoundsthese (Brassica β-ketols in particular) might be derived via head-to-head condensa-tion of two long-chain substrates, their co-occurrence with α-ketols and secondaryalcohols and ketones with EO(b) groups suggests that this is unlikely.Thus secondary functional group installation by P450 enzymes is likely commonto multiple types of specialty wax compounds. Furthermore, the diversity of putativeP450 products summarized here indicates that, while these enzymes do not act withstrict regioselectivity, they are able to discriminate between the central and terminalportions of their substrates’ aliphatic chains.3.3.2 Specialty compounds with functional groups on car-bons of single parity near the chain terminusSpecialty compounds with O(a) secondary functional groups had even TCNs,strict 1,3-geometry, and were not accompanied by 1,2- or 1,4-isomers. This indicatedthat their secondary functions are probably not products of P450 enzyme activity.However, compounds with O(a) secondary functional groups bear strong resemblanceto the 1,3-functional intermediates of the elongation pathway, prompting the hypoth-esis that these compounds may be derived from elongation intermediates [170].Molecular genetic investigations in Solanum lycopersicum had revealed that thehydrolysis of C12, C14, and C16 fatty acid biosynthesis intermediates, acyl-ACPs, isenzymatically catalyzed by a thioesterase, MKS1, to produce 3-keto fatty acids [254].This suggests that a similar hydrolytic fatty acid-forming pathway might use elonga-tion pathway intermediates, acyl-CoAs, to generate compounds with O(a) functionalgroups (Fig. 3.9). Indeed, thus far the literature had documented observations of 1,3-diols in Cosmos bipinnatus [38], 3-hydroxy fatty acids in Aloe arborescens [170], and3-hydroxy aldehydes in some Papaver species [100], suggesting that both hydroxyacyl-and ketoacyl-CoA elongation intermediates might serve as substrates for both the acylreduction and fatty acid-forming pathways, as well as the first step of the decarbony-lation pathway.Specialty compounds with E(a) functional groups (2-alcohols and 2-ketones) hadfunctional groups C-2 only, suggesting that these were also not installed by P450enzymes. Compounds of this type had odd TCNs and frequently co-occurred withcompounds bearing O(a) functional groups, suggesting that they may be derived via693. Diversity and biosynthesis of specialty wax compoundsOSCoAOSCoAOOSCoAOHOSCoAOSCoAKCSKCRHCDECRADREDFARESTOHOHOHOH OHOOOHO(a)From fatty acid de novo biosynthesisDEE(a)Re-entry into elongation cycleIntercept modificationnnnnnnnnnFigure 3.9: Biosynthesis of specialty compounds by intercept modification.Biosynthesis of homologous series or isomeric mixtures of very-long-chain specialtywax compounds whose members secondary functional groups are on carbons withsingle parity (only even- or only odd-numbered carbons) has been hypothesized tooccur via the intercept and modification of elongation intermediates (bracketed com-pounds). Reduction or hydrolysis of these intermediates could lead to 1,3-functionalcompounds (O(a)), while reduction and decarbonylation or hydrolysis and decar-boxylation could lead to 2-functional compounds (E(a)), all of which have beenreported from plant waxes (indicated by grey background). Abbreviations: KCS =ketoacyl-CoA synthase, KCR = ketoacyl-CoA reductase, HCD = hydroxyacyl-CoAdehydratase, ECR = enoylacyl-CoA reductase, RED = reductase, AD = aldehydedecarbonylase, EST = esterase, FAR = fatty acyl reductase, DE = decarboxylase.decarboxylation or decarbonylation (head group removal) of putative O(a) precur-sors, as other authors had speculated [170].Interestingly, further study in the Solanum model had revealed another enzyme,MKS2, that catalyzes decarboxylation of 3-ketoacids (compounds with O(a) groups)to produce 2-ketones with odd TCNs (compounds with E(a) groups) [254] (DE,Fig. 3.9). This suggests that a similar decarboxylation enzyme may exist in otherspecies by which other compounds with O(a) groups serve as precursors to com-703. Diversity and biosynthesis of specialty wax compoundspounds with E(a) groups and one less carbon. Alternatively, since it seems pos-sible that 1,3-functional compounds with even TCNs can serve as substrates forthe first step of the decarbonylation pathway as described above, another path-way to 2-functional compounds with odd TCNs could be via the decarbonylationof 3-functional aldehydes (AD, Fig. 3.9). Such reactions could be catalyzed by theenzyme that forms ubiquitous alkanes from aldehydes. The searches performed hereindicate the presence of both 2-ketones and 2-alcohols in the surface wax of a varietyof species (Table B.1), again suggesting that modification pathways, in this case adecarboxylase enzyme and/or the complete decarbonylation pathway can have accessto both hydroxyacyl- and ketoacyl-CoA elongation intermediates.In summary, hypotheses about specialty wax compound biosynthesis that couldbe drawn from the structural characteristics of compounds with O(a) and E(a) func-tional groups are supported by molecular studies of the Solanum model. These pointto the relevance of both hydroxyacyl- and ketoacyl-CoA elongation intermediate inter-cepts to the biosynthesis of specialty wax compounds. Accordingly, compounds withO(a) or E(a) functional groups may serve as biomarkers for elongation intermediateintercept events.3.3.3 Compound classes with a single secondary functionalgroup near the middle of the chainMany series or mixtures of specialty wax compounds had been encountered withmembers that bore secondary functional groups only on odd-numbered carbons nearthe middle of the chain (regiospecific secondary alcohols and ketones, O(b) func-tional groups). This suggested that these functions were probably installed by ahighly regioselective process. Furthermore, in some cases, these series or mixtureswere found alongside series or mixtures of compounds with O(a) or E(a) functionalgroups (Table B.1), suggesting that the intercept of elongation intermediates may oc-cur in species that are generating compounds with O(b) groups, and that thereforethe intercepted elongation intermediates might themselves be intermediates in thebiosynthesis of compounds with O(b) functional groups.Based on the structural characteristics described above some authors have hypoth-esized that intercepted elongation intermediates (Fig. 3.10A, compounds bound by713. Diversity and biosynthesis of specialty wax compoundscurly brackets) might reenter the FAE or similar elongation machinery so that theirC-3 functional group is retained and shifted to C-5 of the product molecule after oneround of elongation [39, 97, 247] (Fig. 3.10). Any subsequent elongation of this samekind would separate the retained functional group from the head group by an addi-tional two carbons, producing compounds with 1,7-; 1,9-; 1,11-; etc., functional groupgeometry. This process could thus generate product fatty acyl-CoA compounds with asecondary functional group that appeared on odd-numbered carbons only (Fig. 3.10A,dashed box). By passing through the modification pathways used for the generationof ubiquitous wax compounds, these product fatty acyl-CoAs could be converted intoaldehydes, fatty acids, and/or alcohols with secondary functional groups present onodd-numbered carbons only (Fig. 3.10A).In some species compounds with O(b) functional groups co-occur with compoundsbearing E(b) functional groups that are identical except for their head group carbon,suggesting that these compounds might be biosynthetically related by a head groupremoving step. Such compounds had been found in, for example, Osmunda regalis as1,11-C30 diols accompanied by 10-nonacosanol. Similar examples had been found inMyricaria germanica waxes and in waxes from several Papaver species (cf. Fig. 3.5Band A). Since the TCNs of these co-occurring compounds differ by one, it seemsreasonable that intercepted elongation intermediates that reenter a FAE (as hypoth-esized above for the formation of compounds with O(b) functional groups) may alsopass through the decarbonylation step of the decarbonylation pathway and lose aterminal carbon to produce compounds with functional groups present on exclusivelyeven-numbered carbons (compounds with E(b) functional groups).In species where this co-occurrence does not occur, or where the compounds withE(b) functional groups are unaccompanied by homologs or isomers, an alternativebiosynthetic mechanism could be the head-to-head condensation of fatty acyl com-pounds (Fig. 3.10B). Of course, it is possible that compounds with E(b) functionalgroups that are unaccompanied by seemingly related compounds with O(b) func-tional groups may still biosynthesized by intercept-reentry processes. Such uncer-tainty may be clarified by detailed analysis of more species or by molecular geneticand biochemical investigations of genes that might be involved in any of these pro-cesses.723.DiversityandbiosynthesisofspecialtywaxcompoundsOSCoAOSCoAOOSCoAOHOSCoAOSCoAKCSKCRHCDECROOH OSCoAKCRHCDECRKCRHCDECROH OSCoAKCSKCSOSCoAOHOHOH OSCoAOH OSCoAOH OSCoAOSCoAOH OOSCoAOH OHOSCoAOHOSCoAOHADREDFARESTOHOHOHOHE(b)O(b)ADREDFARESTOHOHOHOHOOOHOHOHOHOHOHOR-CORR'OR'R, R' = OH, SACP, or SCoAOB)A) From fatty acid de novo biosynthesisDEO(b)O(a)+E(b)OOOHOHE(b)E(a)Re-entry into elongation cycleRe-entry into elongation cycleRe-entry into elongation cycleIntercept reentryInterceptIntercept modificationREDFARESTnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn nFigure 3.10: (Continued on the following page.)733. Diversity and biosynthesis of specialty wax compoundsFigure 3.10: Biosynthesis of specialty wax compounds by intercept reentry.A) Biosynthesis of homologous series or isomeric mixtures of very-long-chain spe-cialty wax compounds whose members secondary functional group is on carbons ofsingle parity (only even- or only odd-numbered carbons) and with a carbon numbergreater that three (C-X, X > 3) has been hypothesized to occur via the intercept ofelongation intermediates (bracketed compounds), followed by subsequent reentry intofurther elongation cycles. The products of these further elongation cycles (compoundsin dashed boxes) may be available to the modification (blue) pathways to generatecompounds with secondary functional groups on only odd-numbered carbons (O(b)),or via the decarbonylation pathway, products with secondary functional groups ononly even-numbered carbons (E(b)), all of which have been reported from plantwaxes (indicated by grey background). Elongation intermediates arising from thereentry of an intercepted intermediate (square bracketed compounds) could also beintercepted and passed to modification enzymes, resulting in 1,3-compounds with anadditional secondary functional group on only odd-numbered carbons (O(b)O(a))or, via the decarbonylation pathway, 2-functional compounds with an additional sec-ondary functional group on even-numbered carbons (E(b)E(a)). B) Biosynthesis ofvery-long-chain ketones (E(b)) by head-to-head condensation. Abbreviations: KCS= ketoacyl-CoA synthase, KCR = ketoacyl-CoA reductase, HCD = hydroxyacyl-CoAdehydratase, ECR = enoylacyl-CoA reductase, RED = reductase, AD = aldehyde de-carbonylase, EST = esterase, FAR = fatty acyl reductase, DE = decarboxylase.Since the intercept of elongation intermediates from a normal fatty acid elongationcycle has been invoked to explain the biosynthesis of specialty wax compounds, it maybe further hypothesized that intercept of elongation intermediates with additional sec-ondary functional groups (Fig. 3.10A, compounds bound by straight brackets) mightalso occur. Such an event would lead to trifunctional compounds with secondaryfunctional groups on C-3 and one other secondary, odd-numbered carbon (thus com-pounds with O(b)O(a) functional groups). While such compounds have been foundin plant floral oils (Fig. 3.6C), only detailed investigations of more species will revealif these compounds are components of cuticular waxes.743. Diversity and biosynthesis of specialty wax compounds3.3.4 Compound classes with two secondary functionalgroups on carbons of single parity near the middle ofthe chainSeries or mixtures of specialty compounds bearing two secondary functional groupson carbons of single parity near the middle of the carbon chain (E(b)E(b) func-tional groups in particular) had been encountered in the wax mixtures from manyplant species. These had functional groups on carbons with single parity, suggestingfunctional group installation by a highly regioselective mechanism. In a few casesthese compounds had been found alongside compounds with O(a) and/or E(a) func-tional groups (Table B.1), suggesting that the intercept of elongation intermediatesmight be relevant to the biosynthesis of compounds with E(b)E(b) functional groups.Overall, these compounds are structurally similar to those bearing one E(b) func-tional group, except that they possess two functional groups with such characteristics.Based on these observations it seems possible that compounds with E(b)E(b) func-tional groups might be biosynthesized by a mechanism quite similar to that leadingto compounds with a single E(b) functional group.Based on the structural characteristics and similarities described above, authorshad hypothesized that a plastidial fatty acid de novo biosynthesis intermediate 3-functional-ACP (Fig. 3.11A, curly-bracketed compounds) might be further condensedwith another equivalent of malonyl-CoA by a PKS enzyme to produce a tri-functionalsubstrate for the elongation pathway [97, 239, 189]. Further rounds of elongationwould result in diketoacyl-CoA products (Fig. 3.11A, dashed box). These productsmay be available to the modification pathways that could then generate compoundswith O(b)O(b) functional groups, and, by the action of the decarbonylation stepof the decarbonylation pathway, compounds with E(b)E(b) functional groups. Analternative mechanism for the biosynthesis of compounds with E(b)E(b) functionalgroups and odd TCNs (β-diketones in particular) is the head-to-head condensationof an intercepted 3-functional fatty acid with a simple VLC fatty acyl-CoA or fattyacid (Fig. 3.11B). This mechanism is also supported by the presence of compoundswith E(a) and O(a) functional groups (potential elongation intercept markers).In a few cases, β-diketones (Fig. 3.6A) had been reported alongside compoundswith E(a) and O(a) functional groups (potential elongation intercept markers), but753.DiversityandbiosynthesisofspecialtywaxcompoundsOSACPOSACPOOSACPOHOSACPOSACPKASKARHADEAROOH OSCoAKCRHCDECRKCRHCDECROH OSCoAKCSKCSOHOH OSCoAOHOHOH OSCoAOHOH OSCoAOHOH OSCoAOHOHOSCoAOHOH OOSCoAOHOH OHOSCoAOHOHOSCoAOHOHADOHOHOHOHO(b)O(b)E(b)E(b)OHOHOHOHADREDFARESTOHOHOHOHOHOOHOOHOHOHOHREDFARESTOHOHOHOHOR-CORR'OR'R, R' = OH, SACP, or SCoAOH OH OB)A)From fatty acid de novo biosynthesisPKS x2DE+OOOHOHE(b)E(b)Re-entry into elongation cycleRe-entry into elongation cycleRe-entry into cycleInterceptTransfer to condensation enzymeIntercept modificationREDFARESTcondensation withmalonyl-CoA (PKS)nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnFigure 3.11: (Continued on the following page.)763. Diversity and biosynthesis of specialty wax compoundsFigure 3.11: Biosynthesis of specialty wax compounds by intercept conden-sation. A) Biosynthesis of homologous series or isomeric mixtures of very-long-chainwax compounds whose members secondary functional groups are on carbons withsingle parity (only even- or only odd-numbered carbons) and with carbon numbersgreater that three (C-X, X > 3) has been hypothesized to occur via the interceptof fatty acid biosynthesis intermediates (bracketed compounds), followed by conden-sation with malonyl-CoA by a PKS enzyme, and subsequent reentry into furtherelongation cycles. The products of these further elongation cycles (compounds indashed boxes) may be available to the modification (blue) pathways to generatecompounds with two secondary functional groups on only odd-numbered carbons(O(b)O(b)), which have been reported form plant waxes as fatty acids (indicatedby grey background), or via the alkane forming pathway, products with secondaryfunctional groups on only even-numbered carbons (E(b)E(b)). Elongation interme-diates arising from intercepted fatty acid biosynthesis intermediates by the methoddescribed above (square bracketed compounds) could also be intercepted and passedto modification enzymes, resulting in 1,3-compounds with two additional secondaryfunctional groups on only odd-numbered carbons (O(b)O(a)) or, via the decarbony-lation pathway, 2-functional compounds with two additional secondary functionalgroups on even-numbered carbons (E(b)E(a)). B) Biosynthesis of very-long-chaindiketones and ketols (E(b)E(b)) by head-to-head condensation. Abbreviations: KCS= ketoacyl-CoA synthase, KCR = ketoacyl-CoA reductase, HCD = hydroxyacyl-CoAdehydratase, ECR = enoylacyl-CoA reductase, RED = reductase, AD = aldehyde de-carbonylase, EST = esterase, FAR = fatty acyl reductase, DE = decarboxylase.this characteristic is shared by both the potential PKS and head-to-head mechanismsas described above. There had been one report describing the co-occurrence of com-pounds bearing E(b)E(b) and O(b)O(b) functional groups, potentially providingsome support for the PKS mechanism, however, these compounds were reported fromonly one species, and in a pollen extract (Helianthus annus [189]). Thus, it will beinteresting to see if further examples of co-occurring compounds bearing E(b)E(b)and O(b)O(b) functional groups will be found, or if molecular genetic experimentscan provide more conclusive support for one mechanism or the other.As detailed above, elongation intermediate intercept is an important process in thebiosynthesis of specialty compounds whose functional groups are found on carbons ofsingle parity (all non-P450 specialty compounds). Example specialty compound pro-files highlight interesting characteristics of likely intercept processes. For example, byconsidering the profiles of 2-alcohols (mainly C31), and 3-hydroxy acids (mainly C28)773. Diversity and biosynthesis of specialty wax compoundsin Aloe arborescens [170], it becomes apparent that intercepted elongation intermedi-ates may preferentially enter different modification pathways depending on the chainlength at which they are intercepted. This suggests that other enzymes in addition tohydrolases may compete for elongation intermediates, and that these enzymes mayexhibit their own chain length specificity. Moreover, this means that competition forelongation intermediates may exist between the decarbonylation pathway, the fatty-acid forming pathway, the acyl reduction pathway, further elongation within the sameFAE, and further elongation in a different FAE (Fig. 3.12), thus five-way competi-tion. Saturated acyl-CoA elongation products (Fig. 3.12, dotted box) have completedundergoing elongation within one FAE, thus there may be four-way competition forthese products.The chain lengths and steps at which elongation intermediate intercept occurredwere further analyzed by calculating the chain length and oxidation state of theintercept precursors that potentially led to each specialty compound with functionalgroups on carbons of single parity (all non-P450 specialty compounds). For example,by intercept and reentry-based biosynthesis a C30 1,7-diol would arise from a C263-hydroxy intercept precursor that underwent two further rounds of elongation andwas then passed to the acyl reduction pathway (Fig. 3.6). This same calculation wasrepeated for each non-P450 specialty compound. Then, the relative abundance of eachof these specialty compounds within its compound class was plotted according to itsintercept precursor’s chain length and oxidation state. This exercise revealed thatprecursors to these specialty compounds were generally intercepted at chain lengthsgreater than C18 (some C20, mainly C22, some C24, C26, and C28, Fig. 3.13A), whichare chain lengths that are present only in the endoplasmic reticulum. In contrast,precursors to specialty compounds with two non-P450 secondary functional groupswere predominantly C14 and C16, chain lengths that are only present in the plastids.This indicates that, though quite similar at first glance, the potential pathways leadingto compounds with E(o) functional groups and the pathway leading to compoundswith E(o)E(o) functional groups may begin in different subcellular compartments.Next, the relative abundance of each non-P450 specialty compound within its totalwax mixture was plotted as a function of its intercept precursors chain length andoxidation state. This revealed that even though elongation intermediate intercept783. Diversity and biosynthesis of specialty wax compoundsOSCoAOOSCoAOHOSCoAOSCoAKCSKCRHCDECRREDFARESTREDFARESTIntercept reentryIntercept modificationReentry intoelongation cyclennnnFigure 3.12: Competition for elongation pathway intermediates and prod-ucts. Based on the profiles of specialty wax compounds reported thus far, competitionfor fatty acid elongation intermediates (bracketed compounds) may exist between theHCD enzyme, the first enzyme on each modification pathway (RED, EST, FAR),and an intercept reentry process (dashed magenta arrow), thus five-way competition.Similarly, competition for elongation products (compounds in dashed boxes) may ex-ist between the first enzyme on each modification pathway (RED, EST, FAR) and areentry into further elongation cycle process (dashed magenta arrow), thus four-waycompetition. Abbreviations: KCS = ketoacyl-CoA synthase, KCR = ketoacyl-CoAreductase, HCD = hydroxyacyl-CoA dehydratase, ECR = enoylacyl-CoA reductase,RED = reductase, EST = esterase, FAR = fatty acyl reductase.793. Diversity and biosynthesis of specialty wax compoundsprobably occurs at chain lengths ranging from C10 to C36 (Fig. 3.13A), only C14, C16,and C22 intercepts seem to be used to make products that accumulate in substantialamounts (Fig. 3.13B). This indicates that intercept events at some chain lengths mayarise as mistakes or leakages from the elongation pathway, while those at C14, C16, andC22 are likely tightly controlled, deliberate processes in which elongation intermediateintercept entry into a new elongation cycle, possibly in a new FAE complex, wins outover other competing processes.In the intercept-reentry biosynthesis hypothesis for non-P450 specialty com-pounds, intercepted elongation intermediates reenter the FAE and undergo furtherelongation before they are passed to the modification pathways. However, whetherthe machinery that carries out these further elongation rounds is identical to thatwhich handles ubiquitous wax compounds is uncertain. Since the elongation ma-chinery is known to control the chain length profile of the products it creates, thisuncertainty can be addressed using the data presented here by comparing the chainlengths of non-P450 specialty compounds with the chain length profiles of cooccurringubiquitous wax compounds. For example, if the most prominent homolog in a seriesof O(b) diols was C32, but the most prominent homolog in a series of co-occurringubiquitous primary alcohols was C28, it could be supposed that these two compoundclasses may have been generated by different sets of elongation machinery.Accordingly, the difference between the chain length of each non-P450 specialtycompound the chain length of the most abundant, co-occurring ubiquitous compoundwith the same head group was calculated. Then, the relative abundance of each spe-cialty compound within its total wax mixture was plotted according to this differ-ence and the chain length of its intercept precursor. This revealed that the highlyabundant specialty compounds potentially derived from C14/C16 intercepts (mainlyβ-diketones) and C22 intercepts (mainly 10-nonacosanol) are not necessarily elon-gated to the same chain lengths as co-occurring corresponding ubiquitous compoundswith the same head group (Fig. 3.14A). One interpretation of this result is that thesespecialty classes are produced by elongation machinery distinct from that which gen-erates ubiquitous wax compounds, perhaps by a separate elongase that is dedicatedto specialty wax compound formation. Alternatively, this may suggest that theseclasses are not biosynthesized by an intercept-reentry mechanism and that perhaps803.Diversityandbiosynthesisofspecialtywaxcompounds	 32	Fig.	 13:	 Relative	 abundance	 of	 products	 derived	 from	 elongation	 intermediate	 intercept.	 Each	circle	represents	one	specialty	compound	that	had	been	 found	as	a	member	of	a	homologous	series	or	mixture	of	isomers	for	which	abundance	information	had	been	reported.	Here,	each	specialty	compound	is	plotted	according	to	the	oxidation	state	of	its	secondary	functional	group(s)	(y-axis)	and	as	a	function	of	 the	 chain	 length	 of	 its	 intercept	 precursor	 (x-axis).	 Color	 denotes	 the	 isomer	 characteristics	 of	 the	secondary	 functional	 group(s),	 and	 the	 border	 of	 each	 circle	 indicates	 the	 terminal	 carbon	 oxidation	state.	A)	Circle	size	represents	the	relative	abundance	of	each	specialty	compound	in	its	compound	class,	B)	circle	size	represents	the	relative	abundance	of	each	specialty	compound	in	the	total	wax	mixture	in	which	it	was	found.		 	5/8/2016 localhost:8000/RA CC 2.htmlhttp://localhost:8000/RA%20CC%202.html 1/110 12 14 16 18 20 22 24 26 28 30 32 34 36Chain length of intercept precursorOHOOxidation stateE(a) and O(a)E(b) and O(b)E(b)E(b) and O(b)O(b)aldehyde, alkanealcohol, esteracid, methyl esterA) 5/8/2016 localhost:8000/RA CC 2.htmlhttp://localhost:8000/RA%20CC%202.html 1/110 12 14 16 18 20 22 24 26 28 30 32 34 36Chain length of intercept precursorOHOOxidation stateE(a) and O(a)E(b) and O(b)E(b)E(b) and O(b)O(b)aldehyde, alkanealcohol, esteracid, methyl esterB) Figure 3.13: Relative abundance of products derived from elongation intermediate intercept. Each circlerepresents one specialty compound that had been found as a member of a homologous series or mixture of isomers forwhich abundance information had been reported. Here, each specialty compound is plotted according to the oxidationstate of its secondary functional group(s) (y-axis) and as a function of the chain length of its intercept precursor (x-axis).Color denotes the isomer characteristics f the s c ndary functional group(s), and the border of each circle indicatesthe terminal carbon oxidation state. A) Circle size represents the relative abu dance of each specialty compound in itscompound class, B) circle size represents the relative abundance of each specialty compound in the total wax mixture inwhich it was found.813. Diversity and biosynthesis of specialty wax compoundsthe alternate head-to-head mechanism should receive further consideration.The relative abundance of each non-P450 specialty compound within its totalwax mixture was then plotted according to its chain length and the chain length ofits intercept precursor. This revealed that compounds derived from a C22 intercept(mainly 10-nonacosanol) are all elongated to C30 prior to modification (Fig. 3.14B),suggesting that a dedicated C30-specific FAE may be responsible for this process. Incontrast, specialty compounds potentially derived from C14 and C16 intercepts (mainlyβ-diketones) are, if biosynthesized by an intercept/reentry mechanism, elongated toC30, C32, and C34, suggesting that the FAE responsible for such elongation either has abroad product profile, a product profile that is species-dependent, that C14 interceptsmay serve as substrates for an FAE with C34 product specificity and C16 interceptsfor one with C32 specificity, or that the β-diketone products are biosynthesized bya different mechanism entirely. Overall, this evidence reenforces the idea that β-diketone biosynthesis by a head-to-head mechanism should not be overlooked.3.4 ConclusionsThis literature survey reveals that specialty cuticular wax compounds are all char-acterized by a number of structural variables including TCN parity, head group oxida-tion state, secondary functional group oxidation state(s), and isomer characteristicsof the secondary functional group(s). These compounds were catalogued accordingto these parameters to facilitate pattern recognition. The isomer characteristic of thesecondary functional group(s) proved to exhibit patterns particularly useful in hy-pothesizing the biosynthetic mechanisms leading to these compounds. It seems highlylikely that series or mixtures of specialty compounds whose members have functionalgroups on carbons of mixed parity (even- and odd-numbered carbons) have theirsecondary functional groups installed by P450 enzymes, as precedence for such hadbeen set by investigation of an Arabidopsis P450. In contrast, series or mixtures withfunctional groups on carbons of single parity (functional groups on only even- or onlyodd-numbered carbons) seem quite likely to be biosynthesized via the interceptionof elongation intermediates that then reenter the elongation enzyme complex(es) andare then modified into compounds with regioselectively installed functional groups823. Diversity and biosynthesis of specialty wax compounds	 33																																				Fig.	 14:	 Chain	 lengths	 of	 specialty	 compounds	 relative	 to	 co-occurring	 ubiquitous	 compounds.	Each	circle	represents	one	specialty	compound	that	had	been	found	as	a	member	of	a	homologous	series	or	mixture	of	isomers	for	which	specialty	compound	and	co-occurring	ubiquitous	compound	abundance	information	had	also	been	reported.	Here,	each	specialty	compound	is	plotted	as	a	function	of	the	chain	length	 of	 its	 intercept	 precursor	 (x-axis).	 Color	 denotes	 the	 isomer	 characteristics	 of	 the	 secondary	functional	group(s),	and	the	border	of	each	circle	indicates	the	terminal	carbon	oxidation	state.	Circle	size	represents	 the	 relative	 abundance	of	 each	 compound	 in	 the	 total	wax	mixture	 in	which	 it	 is	 found.	A)	Specialty	compounds	plotted	according	to	the	difference	between	their	chain	length	and	the	chain	length	of	 the	most	abundant	ubiquitous	compound	with	 the	same	terminal	carbon	oxidation	state	 (y-axis).	B)	Specialty	compounds	plotted	according	to	their	chain	length	(y-axis)			 	5/8/2016 localhost:8000/A1 TW.htmlhttp://localhost:8000/A1%20TW.html 1/110 12 14 16 18 20 22 24 26 28 30Chain length of intercept precursor-8-6-4-202468Corresponding ubiq. chain length ­ specialty compound chain lengthE(a) and O(a)E(b) and O(b)E(b)E(b) and O(b)O(b)aldehyde, alkanealcohol, esteracid, methyl ester5/8/2016 localhost:8000/RA TW 2.htmlhttp://localhost:8000/RA%20TW%202.html 1/110 12 14 16 18 20 22 24 26 28 30Chain length of intercept precursor24262830323436Product Chain LengthE(a) and O(a)E(b) and O(b)E(b)E(b) and O(b)O(b)aldehyde, alkanealcohol, esteracid, methyl esterA) B) Figure 3.14: (Continued on the following page.)833. Diversity and biosynthesis of specialty wax compoundsFigure 3.14: Chain lengths of specialty compounds relative to co-occurringubiquitous compounds. Each circle represents one specialty compound that hadbeen found as a member of a homologous series or mixture of isomers for which spe-cialty compound and co-occurring ubiquitous compound abundance information hadalso been reported. Here, each specialty compound is plotted as a function of thechain length of its intercept precursor (x-axis). Color denotes the isomer characteris-tics of the secondary functional group(s), and the border of each circle indicates theterminal carbon oxidation state. Circle size represents the relative abundance of eachcompound in the total wax mixture in which it is found. A) Specialty compoundsplotted according to the difference between their chain length and the chain lengthof the most abundant ubiquitous compound with the same terminal carbon oxida-tion state (y-axis). B) Specialty compounds plotted according to their chain length(y-axis).near the middle of the carbon skeleton. Through analyses of specialty wax compoundcooccurrence and structural characteristics it is difficult to predict how β-diketonesmight be biosynthesized. Though the prevailing hypothesis in the literature, a com-bined intercept/reentry and PKS mechanism, is supported by the collected data here,an analysis of the relationships between specialty and co-occurring ubiquitous com-pounds does not strongly support this hypothesis and indicates that a head-to-headcondensation mechanism should not yet be ruled out.Finally, this work reveals that specialty wax compounds had been identified on orin at least one species of nearly all plant phylogenetic groups (Fig. 3.15). This suggeststhat the machinery required to make specialty wax compounds was probably presentin a common ancestor of all plant species. Furthermore, multiple investigations inalgae had found the presence of compounds with the same structural characteristicsas specialty wax compounds. This indicates that the machinery required for specialtycompound biosynthesis may have been present in plant predecessors even before thecolonization of land, and that such machinery may have been recruited from otherareas of metabolism to assist in the formation of the plant cuticle.843. Diversity and biosynthesis of specialty wax compoundsPlant	Order EO(a)EO(b)EO(b)EO(b)O(a)E(a)O(b)E(b)E(b)EO(b)O(b)O(b)E(b)E(b)E(b)E(b)EO(b)O(b)O(a)Ferns OsmundalesAlismatalesPinalesMagnoliidaeSalvinialesGinkgoalesAlgae ChlorellalesBryophytes FunarialesEustigmatalesPolytrichalesLauralesMagnolialesGymnospermsCycadalesAsparagalesMonocots ArecalesPoalesRanunculalesZingiberalesProtealesBuxalesMyrtalesCelastralesDicotsFabalesRosalesMalpighialesBrassicalesMalvalesEricalesSapindalesSolanalesApialesAsteralesFigure 3.15: Distribution of specialty wax compounds across plant orders.Each plant order in which specialty compounds have been found is listed on the leftside of the matrix, and the isomer characteristic of the classes discussed in this workare listed across the top of the matrix. Each filled cell indicates the presence of oneor more specialty compounds with that isomer characteristic in one or more speciesin that order.22Phylogeny created according to: Stevens, P. F. (2001 onwards). Angiosperm Phylogeny Website.Version 12, July 2012 [and more or less continuously updated since].853. Diversity and biosynthesis of specialty wax compounds3.5 Supplementary dataSupplementary data for this chapter can be found in Appendix BTable B.1: Species with specialty cuticular wax compounds.86Chapter 4Identification of β-hydroxy fattyacid esters andprimary,secondary-alkanediolesters in cuticular waxes of themoss Funaria hygrometrica4.1 IntroductionThe evolutionary transition of plant life from aquatic to terrestrial environmentspresented great challenges, desiccation being principal among them. One structuralfeature that helps plants reduce transpiration across their vast aerial surfaces is thecuticle, an extra-cellular membrane covering epidermal cells [252]. Plant cuticles con-sist of the fatty acid (FA) polyester cutin that provides a structural framework, andembedded waxes that seal the surface [50]. Plant cuticular waxes are complex mix-tures of very-long-chain (VLC) FA derivatives that can be divided into two categories:those that occur ubiquitously in the wax of almost all plant species, and specialtycompounds that are encountered only in the wax of certain plant taxa [99]. Theubiquitous wax constituents are mostly VLC aliphatics with one terminal functionalgroup such as fatty acids, aldehydes, and primary alcohols, but may also include874. Hydroxy esters from Funaria hygrometricaalkanes (no functional groups) or dimers formed by ester linkages between fatty acidsand wax alcohols [185]. Specialty wax compounds are mostly also fatty acid-derived,but may contain more than one functional group including in-chain functionalities,thus creating much greater structural diversity.Our current understanding of wax biosynthesis is primarily based on moleculargenetic and biochemical characterizations of model plants, most prominently Ara-bidopsis thaliana, the wax mixtures of which comprise mostly ubiquitously occurringcompounds [95, 172]. In the wax biosynthetic pathways leading to these wax con-stituents, two stages can be distinguished: elongation and diversification. First, evencarbon numbered long-chain fatty acyl-CoAs originating from plastidial FA de novobiosynthesis are elongated by the addition of C2 units derived from malonyl-CoA.The fatty acid elongase (FAE) multi-enzyme complex accomplishes the transforma-tion by catalyzing four sequential reactions. Initially a ketoacyl-CoA synthase (KCS)condenses the fatty acyl-CoA and malonyl-CoA substrates to form a β-ketoacyl in-termediate [104], then a ketoacyl-CoA reductase (KCR) reduces the secondary oxogroup to a hydroxyl function [17]. Next, the β-hydroxyl group is eliminated by a de-hydratase hydroxyacyl-CoA dehydratase (HCD), and the resulting α,β-double bondis finally saturated by an enoylacyl-CoA reductase (ECR) to form a fatty acyl-CoAtwo carbons longer than the original KCS substrate [7, 256]. Repeated FAE cycleslead to acyl-CoAs with chain lengths ranging from C24 to C34.In the second stage of ubiquitous wax compound biosynthesis, elongated VLCacyl-CoAs are diversified into various derivatives through head group modifications.These take place in a chain-length-specific manner such that the full range or onlya selection of the chain lengths in the acyl-CoA precursor pool is found in eachof the final wax products. Matching chain length profiles have been reported forthe free and esterified n-alkanols in various plant species [118, 173], indicating thatthe esters and free alkanols are biosynthetically related. Indeed, it had been shownthat in Arabidopsis a fatty acyl-CoA reductase (FAR) generates VLC n-alkanolsthat are then linked with fatty acyl-CoAs by a wax ester synthase [124], and thatboth the alcohol intermediates and the alkyl ester end products are exported to thecuticle. Further chain length comparisons across species suggested that a separatebiosynthetic pathway leads to aldehydes and alkanes. Molecular genetic evidence,884. Hydroxy esters from Funaria hygrometricaagain from Arabidopsis, has confirmed that reduction of VLC acyl-CoAs truly leadsto even-carbon-numbered aldehydes and further decarbonylation to correspondingodd carbon-numbered alkanes [24, 42].While the biosynthesis of the ubiquitous wax compounds is well understood, theprocesses leading to specialty compounds have received far less attention. Excep-tions to this are the secondary alcohols, diols, ketones, and ketols found as majorconstituents of Brassicaceae waxes [122, 255]. Again, homolog profiles suggest abiosynthetic relationship: that all these specialty compound classes are derived fromthe ubiquitous alkanes. Further detailed isomer analysis of Arabidopsis stem waxconstituents with secondary functional groups indicates that hydroxylation on threecarbon atoms near the center of alkane precursor molecules likely leads to the sec-ondary alcohols, and repeat hydroxylation or further oxidation to the ketones, diols,and ketols [248]. Molecular genetic investigations confirmed this hypothesis, and aP450 monooxygenase with midchain alkane hydroxylase (MAH) activity was shownto be involved in the process by oxidizing any carbon from C-13 to C-15 of the C29alkane chain to produce a mixture of isomers whose secondary functional groups areon adjacent carbons [72].Many other wax compounds with one or more secondary oxygen-containing func-tional groups that do not seem to be biosynthetically related to the secondary func-tional compounds of the Brassicaceae have been described as major wax constituentsin several plant taxa. For example, the secondary alcohol 10-nonacosanol is a veryprominent wax component of various gymnosperm and angiosperm taxa, as well asmultiple Pogonatum moss species [14, 100, 151]. In contrast to the isomer mixturesof Arabidopsis secondary alcohols, this compound usually appears as a single iso-mer, and while some species also contained trace amounts of 8- or 12-nonacosanol,the 9- or 11-isomers were not detected. Similarly, ketones had been found in someplant waxes, most prominently on the fern Osmunda regalis [103], and also in theform of β-diketones in the Poaceae [15]. Based on their isomer patterns, these waxconstituents with secondary functionalities have long been suspected to originatefrom processes other than P450 hydroxylation, possibly through activity of polyke-tide synthase (PKS) enzymes [239]. In essence, their secondary functional groups werehypothesized to originate during chain elongation, as remnants of β-functionalities894. Hydroxy esters from Funaria hygrometricaintroduced through Claisen condensation reactions, thus explaining their exclusivepresence on particular carbons. Diverse biochemical experiments have since con-firmed this hypothesis for β-diketones, but have not provided information about theenzymes and genes involved [238]. Similar biochemical and molecular genetic evidenceis also lacking for other specialty compounds.In the absence of enzymological or genetic information, the biosynthetic machinerybehind the formation of specialty wax constituents is best assessed by further detailedanalyses of diverse compounds with secondary functionalities. In particular, thosewith oxygen functionalities on both a primary and a secondary carbon are particularlyinformative, for example 5-hydroxy aldehydes and 1,5-alkanediols in Taxus baccataand 5-hydroxy acids in Cerinthe minor [101, 247], 11-keto alcohols and aldehydesin Osmunda regalis [102], 1,7-; 1,9-; and 1,11-diols in Papaver alpinum [100], and3-hydroxy fatty acids in Aloe arborescens [170]. The primary functionalities thereinadded substantial chemical diversity in the form of compound classes with terminalcarboxylic acid, ester, aldehyde, or alcohol group. Comparisons between the homologand isomer patterns of these bifunctional compound classes, where they were co-occurring, thus proved to be particularly informative for narrowing down possible,novel biosynthetic pathways.In order to further our understanding of wax biosynthesis, novel compounds withboth primary and secondary functional groups have to be identified, which necessi-tates analyses of waxes from diverse plant lineages. It is important that both vascularand non-vascular plants are included in this survey, especially since specialty com-pounds have been found as prominent members of the waxes of angiosperms, gym-nosperms, ferns, and mosses alike. Most notably, waxes from several Polytrichalesmosses comprised large percentages of compounds with secondary functional groups,including 10-nonacosanol [151]. Waxes of Andreaea, Pogonatum, Syntrichia, andPhyscomitrella moss species contained ubiquitous wax compounds including fattyacids, alkanols, alkyl esters, aldehydes, and alkanes [31, 78, 249]. However, bifunc-tional compound classes from moss waxes have yet to be identified. The goal of thepresent study was to provide further moss wax analyses and to identify bifunctionalcompounds if possible.Recently, microscopic examination of multiple structures of the moss Funaria hy-904. Hydroxy esters from Funaria hygrometricagrometrica revealed that the calyptra, a maternal protective structure of mosses, hasa cuticle [32]. Interest in the wax of a cuticle that had not been previously an-alyzed prompted a preliminary analysis, which indicated the presence of unknowncompounds that potentially contained two functional groups. As candidates for com-pounds that could expand our understanding of the diversity and potentially thebiosynthesis of specialty wax compounds, structural elucidation of these compoundswas the focus of this work. Accordingly, waxes were extracted from the surfaces ofthe three main structures of F. hygrometrica, the maternal leafy gametophyte, theoffspring sporophyte capsule, and the maternal calyptra. The wax mixtures wereseparated with thin layer chromatography (TLC) where necessary, and analyzed astrimethyl silyl (TMS) derivatives with GC-MS. Authentic standards were then syn-thesized to confirm the structures of the unknown compounds.4.2 ResultsThe aim of the present work was to identify novel compounds in the cuticularwaxes covering various F. hygrometrica structures. Extracts from the surfaces ofthe calyptra, leafy gametophyte, and sporophyte capsule yielded typical wax mix-tures composed of VLC fatty acids, n-alkanols, alkyl esters, aldehydes, and alkanes,together with two prominent groups of novel compounds. One was found in the ca-lyptra and leafy gametophyte waxes (designated compound class A), and the otherin the sporophyte capsule samples (compound class B). Mass spectroscopic analysisand organic synthesis were used to determine the structures of the compounds in eachgroup.4.2.1 Identification and quantification of unknown series ATo isolate compound class A, surface wax was extracted from several hundred leafygametophytes and separated into three fractions using TLC. An aliquot of each frac-tion was reacted with bis-N,O-trimethylsilyltrifluoroacetamide to form TMS deriva-tives of groups with exchangeable protons, and the resulting mixtures were analyzedwith GC-MS. Compound class A was found in a fraction (Rf 0.39) well separatedfrom all other wax components and running between n-alcohols and wax esters. Ini-914. Hydroxy esters from Funaria hygrometricatial gas chromatography - mass spectrometry (GC-MS) analysis revealed that fractionA contained seven compounds (Figure 4.1A) that shared MS characteristics. Theirfragmentation patterns included some fingerprint ions that were identical between allseven compounds and some with differences of m/z 28 between them. Based on thisMS behavior and the evenly spaced gas chromatography (GC) peaks, fraction A wasrecognized as a series of seven homologous compounds.All members of unknown series A had high GC retention times and containedfragments larger than m/z 700+ in their mass spectra, both reminiscent of alkylesters formed by monofunctional wax alcohols and acids. To assess whether fractionA contained similar ester-linked dimers, and to determine the number and nature ofthe functional groups involved, the remainder of the TLC fraction was subjected totransesterification with BF3 and methanol (MeOH) followed by TMS derivatization.GC-MS analysis showed that the reaction mixture contained eight major compounds,which were identified as the methyl ester TMS ether derivatives of C20, C22, C24, andC26 β-hydroxy fatty acids, and as TMS ethers of C20, C22, C24, and C26 n-alkanols.The C24 β-hydroxy FA methyl ester and C22 n-alkanol were found to predominate(Figure C.1). Thus, the initial analysis suggested that the compounds in series Awere VLC β-hydroxy fatty acid alkyl esters.Next, the mass spectra of the TMS derivatives of the compounds in A wereexamined to further confirm their potential β-hydroxy FA ester structures. Allseven spectra exhibited (i) prominent fragments m/z 73 [Si(CH3)3]+ indicative ofa TMS-derivatized hydroxyl group, (ii) series of fragments m/z 57, 71, 85, 99, etc.([CnHn+1]+) suggesting at least one aliphatic tail, and (iii) fragments m/z 145 and 161,likely due to combined α-fragmentation and rearrangement, indicating a β-hydroxyacid moiety (Figure C.2).Each of the seven mass spectra exhibited further characteristic fragments, in-cluding ions [M-15]+, possibly due to methyl loss, and [M-90]+, potentially due toHOSi(CH3)3 elimination from the parent molecules (m/z 750 and 675, respectively,for the C46 compound in Figure 4.1C)1. These fragments differed by m/z 28 between1Due to the high number of carbon atoms in wax molecules and thus the high probability of13C and/or 2D incorporation, large fragments such as [CnH2n+1OTMS-15]+ are expected to haveaverage m/z (3n x 0.01) units more than the nominal mass. The standard resolution of the MS dataacquisition system used here rounds m/z values to the nearest integer and will thus report m/z 750for the above ion.924. Hydroxy esters from Funaria hygrometricaFigure 4.1: (Continued on the following page.)934. Hydroxy esters from Funaria hygrometricaFigure 4.1: Identification of unknown series A found in the Funaria hygro-metrica calyptra and leafy gametophyte waxes. A) Selected ion chromatogram(m/z 161) of fraction A isolated from the leafy gametophyte extract using TLC. B)Selected ion chromatogram (m/z 161) of synthetic C46 β-hydroxy FA ester (docosyl3-hydroxytetracosanoate (3.4), obtained from docosanol (3.1), see Fig. 4.2). C) EImass spectrum of the most abundant homolog (C46) in A). D) EI mass spectrum ofthe synthetic C46 β-hydroxy fatty acid ester shown in B). E) Potential routes to themajor fragments observed in the mass spectra of the C46 β-hydroxy fatty acid ester.Additional fragmentation mechanisms and ions resulting from the fragmentation ofother C46 β-hydroxy acid isomers are described in Figure C.5.homologs. A third fragment, likely arising from McLafferty rearrangement with dou-ble hydrogen transfer (r2H, Figure 4.1E), as had been reported for other VLC longchain esters [230], was observed as a small peak in each spectrum (m/z 457 for theC46 compound). Finally, two α-fragments generated about either side of the (TMS-derivatized) hydroxyl group indicated the presence of one alkyl tail and a carboxylhead group (m/z 397 and 469 for the C46 compound), which confirmed the β-hydroxyacid ester structure. The major isomer of the C46 compound was thus tentatively iden-tified as the C22 alkyl ester of C24 β-hydroxy acid (docosyl 3-hydroxytetracosanoate).It should be noted that the α-fragments of the β-hydroxyl function were flankedby pairs of homologous ions (m/z 369 and 497, as well as m/z 341 and 525 for theC46 ester, (Figure C.2), suggesting the presence of isomeric β-hydroxy acid esterswith shorter alcohol and correspondingly longer β-hydroxy acid moieties. The pres-ence of such ester isomers was confirmed by corresponding rearrangement fragments(m/z 385, 413, 441 for the C46 homolog, (Figure C.2), present in approximately thesame ratios as the parent ions. The relative abundances of these ions did not varyacross individual ester homolog peaks, indicating that the ester isomers were notGC-separated under the conditions used here.To confirm the structure assignment for fraction A, an authentic standard of do-cosyl 3-hydroxytetracosanoate was synthesized (Figure 4.2). To this end, docosanol(3.1) was transformed into both docosyl 2-bromoacetate (3.2) and docosanal (3.3)via reaction with bromoacetyl chloride and anhydrous oxidation with PCC, respec-tively. Mass spectroscopic characterization of these intermediates is provided as theappendix. A Reformatsky reaction was then used to unite 3.2 and 3.3 to obtain do-944. Hydroxy esters from Funaria hygrometricaOClBrOOBrC21H43C21H43HOC21H43OPCCDCMZn3.1 3.23.3OO C21H43C20H41OH3.4pyridineTHFOOZnC21H43BrOO C21H43C20H41OBrZnHClH2OFigure 4.2: Synthesis of a C46 β-hydroxy FA ester (docosyl 3-hydroxytetracosanoate). Docosanol (3.1) was converted into docosanal (3.3) aswell as bromoacetate ester 3.2 and the corresponding organo-zinc reagent, which wasthen combined with 3.3 to form docosyl 3-hydroxytetracosanoate (3.4). Compoundsin parentheses were not isolated, but used directly in the following step. Yields andprocedural details are in the Experimental section.cosyl 3-hydroxytetracosanoate (3.4). The final product was isolated by preparativeTLC and found to co-migrate on analytical TLC with the natural product. A TMSderivative of the TLC-purified synthetic standard had mass spectral and GC retentionbehavior identical to those of one isomer of one homolog in series A, thus confirm-ing the identity of the latter. Examining all the evidence jointly, the compounds infraction A were identified as homologous β-hydroxy FA esters of the four even- andthree odd-numbered chain lengths ranging from C42 to C48 (Figure 4.1A-D).The relative amounts of the hydroxyester homologs were quantified in the waxmixtures from different moss organs by integrating appropriate selected ion chro-matograms (m/z 161). In the calyptra wax, the seven β-hydroxy FA esters witheven-numbered chain lengths between C40 and C52 were found in a bimodal distri-bution with peaks at C42 and C50 (Figure 4.3). In the total wax mixture from leafygametophytes, 13 β-hydroxy fatty acid ester homologs ranging from C40 to C52 weredetected, thus even more than in the corresponding TLC fraction (Figure 4.1A). Thedistribution of these compounds in the leafy gametophyte wax was approximatelynormal with a peak at C46.954. Hydroxy esters from Funaria hygrometricaFigure 4.3: Quantification of hydroxy ester homologs in Funaria hygromet-rica waxes. Relative abundances of homologous β-hydroxy FA esters in the ca-lyptra (black) and leafy gametophyte (grey) were determined by integrating selectedion chromatograms (m/z 161). Values are means of three independent parallels, witherror bars indicating standard deviation.4.2.2 Identification and relative quantification of unknownseries BThe next objective was to identify the constituents of compound class B. Sincerelatively little sporophyte capsule wax was available, the structure elucidation hadto be carried out directly from the total wax mixture without prior TLC separation.However, the TMS derivatives of series B obtained without pre-separation proved tohave MS characteristics that enabled a tentative structure assignment. Late-eluting,regularly spaced GC peaks (Figure 4.4A) and multiple homologous MS fragmentsagain indicated that all compounds in B belonged to one homologous series.The mass spectra of the TMS derivatives of all homologs in series B had four char-acteristics in common (Figure 4.4C): (i) prominent fragments m/z 73 ([Si(CH3)3]+),but not m/z 147 ([(CH3)3SiOSi(CH3)2]+), indicating the presence of only one hy-droxyl functionality, (ii) alkyl fragments m/z 57, 71, etc. [CnH2n+1]+, suggesting thepresence of at least one unfunctionalized hydrocarbon terminus, (iii) a pair of frag-964. Hydroxy esters from Funaria hygrometricaFigure 4.4: (Continued on the following page.)974. Hydroxy esters from Funaria hygrometricaFigure 4.4: Identification of unknown series B found in the Funaria hy-grometrica sporophyte capsule wax. A) Selected ion chromatogram (m/z 425)of the total wax mixture extracted from the capsules, depicting homologous seriesB. B) Selected ion chromatogram (m/z 425) of the synthetic C46 1,7-diol ester (7-hydroxytriacontyl palmitate (3.13), obtained from 1,6-hexane diol (3.5), Figure 4.5).C) EI mass spectrum of the C46 homolog of series B. D) EI mass spectrum of thesynthetic C46 1,7-diol ester shown in B). E) Potential routes to the major fragmentsobserved in the mass spectra of the C46 1,7-diol ester. Additional fragmentationmechanisms and ions resulting from the fragmentation of other C46 1,X-diol esterisomers are described in Figure C.6.ments m/z 239 and 257 interpreted as [CH3(CH2)14CO]+ and [CH3(CH2)14C(OH)2]+,respectively, and hence pointing to the presence of a C16 acid moiety, and (iv)a pair of fragments m/z 313 ([CH3(CH2)14COOSi(CH2)2]+, (Figure C.5) and 329([CH3(CH2)14C(OSi(CH3)3)OH]+, Figure C.5) suggesting a C16 acid ester containinga (TMS-derivatized) hydroxyl function in its alkyl moiety [180].Considering all the mass spectral evidence thus far, the compounds in B weresurmised to be FA esters of alkanediols and hence positional isomers of the com-pounds in series A. In accordance with this, the two most abundant fragments inthe spectrum of each (TMS-derivatized) compound in B could be interpreted asα-fragments generated on either side of the secondary hydroxyl group located inthe alkyl chain of a wax ester. For example, the fragments m/z 441 and 425 (forthe homolog eluting at ca. 54 min) were explained as [CH3(CH2)22CHOSi(CH3)3]+and [CH3(CH2)14COO(CH2)6CHOSi(CH3)3]+, respectively, and therefore pointed toa 1,7-constellation of one primary and one secondary hydroxyl function in the alkylmoiety of the ester (Figure 4.4E). Satellite fragments of each of these α-fragmentswere present in the spectrum from each homolog (m/z 413 and 469, as well as m/z397 and 453, respectively, for the homolog shown in (Figure 4.4C). These could beexplained as α-fragments from co-eluting isomers in which the hydroxyl group wasfive or nine carbons away from the ester group. With all major fragments in thespectra thus explained, compounds in B were tentatively assigned to a homologousseries of 1,X-diol ester isomers (X = 5, 7, or 9).In order to confirm the 1,X-diol ester structure of compounds in B, the synthe-sis of a representative of this compound class was attempted. Considering the two984. Hydroxy esters from Funaria hygrometricaOHHOHBrBrOH BrOOBrMgOOC22H45OHC22H45OHHOOOC22H45OOHC22H45OC14H29C22H45OOHOHOHC22H45tolueneDHP, PPTSDCMMgTHFLAHDCMPCCDCMHClH2OCH3(CH2)14COCl, DMAPDCM3.63.103.73.113.53.93.13 3.12MgBr3.8Figure 4.5: Synthesis of 7-hydroxytriacontyl palmitate. Tetracosanal (3.11),as obtained from tetracosanoic acid through two redox steps, was added to the Grig-nard reagent 3.8 obtained from THP-protected, brominated 1,6-hexanediol (3.5).An acidic workup afforded 1,7-tricontanediol (3.12), which was then esterified withpalmitoyl chloride to form 7-hydroxytriacontyl palmitate (3.12). Compounds inparentheses were not isolated, but used directly in the following step. Yields andprocedural details are given in the Experimental section.functional groups in the target molecule, a synthetic scheme was designed wherethe secondary hydroxyl functionality was generated first, via Grignard reaction, andthe ester group second, using an acyl chloride (Figure 4.5). One hydroxyl group of1,6-hexanediol (3.5) was substituted with bromine (3.6), the other protected as atetrahydropyranyl ether, and then the resulting compound (3.7) was added to mag-nesium to afford the Grignard reagent 3.8. In parallel, tetracosanoic acid (3.9) wasconverted in two redox steps via the corresponding alcohol (3.10) into tetracosanal(3.11). Finally, 3.8 and 3.11 were combined to afford 1,7-triacontanediol (3.12).Information regarding mass and 1HNMR spectroscopic characterization of the inter-mediates in this synthesis is provided (Figures C.6 and C.7).An aliquot of 3.12 was further reacted with dimethylaminopyridine (DMAP) andpalmitoyl chloride to produce the target compound 3.12. Selective acylation of theprimary hydroxyl group was achieved by using limiting amounts of DMAP and palmi-toyl chloride. The diol ester product of this reaction was purified by TLC, transformedinto the corresponding TMS derivative, and characterized by GC-MS. Both the re-tention behavior and MS characteristics of 3.12 matched those of one isomer in themiddle homolog of series B (Figure 4.4A-D). This match of properties thus confirmedthe 1,X-diol ester structure of one major compound in B. Interestingly, closer inspec-994. Hydroxy esters from Funaria hygrometricaFigure 4.6: Relative abundances of homologous diol esters in Funaria hy-grometrica sporophyte capsule wax. Amounts are reported as the average of n= 3 replicate samples quantified by integrating the selected ion chromatograms (m/z425). Error bars denote standard deviation.tion established that the tetracosanoic acid starting material for synthesis of 3.12contained a minor impurity in the form of docosanoic acid, which resulted in smallamounts of homolog byproducts in all steps and finally led to 7-hydroxyoctacosylpalmitate observed as a small peak in the selected ion chromatogram of the syn-thetic product (Figure 4.4B). The mass spectrum of its TMS derivative showed allthe characteristics expected according to the above interpretation of diol ester MSfragmentations, thus providing further support for the homologous nature of the se-ries.Overall, a series of 1,X-diol esters with even-numbered chain lengths C40-C52 andodd-numbered homologs C45 and C47 was identified (Figure 4.4A). The abundance ofthe satellite α-fragments described above was used to determine the approximate ratioof esterified 1,5-; 1,7-; and 1,9-diol isomers (secondary hydroxyl position isomers) tobe 1:4:1, which was roughly the same for all homologs. Finally, the relative amount ofeach diol ester homolog was quantified by integrating the selected ion chromatogramfor m/z 425 (Figure 4.4A). Their distribution was approximately normal and the C46homolog had the highest abundance (Figure 4.6).1004. Hydroxy esters from Funaria hygrometrica4.2.3 Identification of other bifunctional wax componentsBy detailed examination of the GC-MS traces, two diols were also identifiedin the wax from F. hygrometrica. First, in the calyptra and leafy gametophytewaxes, 1,3-octacosanediol was found based on previous reports of C22 and C26 VLC1,3-diols [234, 38]. The mass spectrum of this compound displayed characteristicα-fragments and peaks at m/z 73 ([Si(CH3)3]+), 103 ([CH2OSi(CH3)3]+), and 147([(CH3)3SiOSi(CH3)2]+) (Figure C.8). Second, 1,7-triacontanediol (3.12) was identi-fied in the wax of the sporophyte capsule. Its mass spectrum also contained fragmentswith m/z 73, 103, and 147, characteristic α-fragments, a peak at m/z 584 ([M-CH3]+),and the products of single elimination of [HOSi(CH3)3] from both the intact molecule(giving rise to [CH3(CH2)22CHOSi(CH3)3(CH2)4CHCH2], m/z 509) and the bifunc-tional α-fragment (observed as [CHOSi(CH3)3(CH2)4CHCH2], m/z 185). Confirma-tion of this structural assignment was provided by a 1,7-triacontanediol standard(Figure C.9), obtained as an intermediate in the synthesis of 7-hydroxytriacontylpalmitate. Further searches using similar strategies failed to provide evidence for thepresence of other 1,X-diols.Finally, the wax mixtures from the leafy gametophyte and calyptra were searchedfor traces of the β-hydroxy acid esters, and the sporophyte capsule wax for the 1,X-diolesters. No evidence was found for presence of the novel compounds outside the mossstructures where they had originally been identified. Also absent from all the waxsamples was any indication of structural variants of series B, for example, a diol esterin which the secondary hydroxyl group was esterified instead of the terminal hydroxylgroup. The possibility of bis-esterified 1,X-diol esters (where both the terminal andmidchain hydroxyl groups had been esterified with a fatty acid) cannot be excludedbecause it is likely that these molecules would escape detection, as they would be toolarge to elute from the GC within the 77-minute run time programmed here.4.3 DiscussionThis chapter presents the structure elucidation of two novel classes of wax com-pounds from three surfaces of the moss Funaria hygrometrica. Waxes from the leafygametophyte and calyptra contained β-hydroxy fatty acid (FA) esters, while those1014. Hydroxy esters from Funaria hygrometricaon the sporophyte capsule comprised 1,X-alkanediol esters instead. Both compoundclasses had several structural features in common, including the ester linkage betweentwo unbranched, fully saturated hydrocarbon chains bearing terminal carboxyl or hy-droxyl groups. Most notably, all the ester structures further contained one additionalsecondary hydroxyl function, albeit in varying positions. Overall, both compoundclasses may thus be described as ester-linked dimers of one mono- and one bifunc-tional very-long-chain (VLC) compound.While wax constituents with primary functionality typically occur in simple ho-mologous series, those with secondary groups frequently exhibit additional regioi-somerism. Accordingly, the bifunctional monomers contained within the two esterclasses, combining a primary and a secondary functional group, had the potentialfor homology as well as isomerism. Interestingly, only a few characteristic homologsand isomers were found for each bifunctional ester monomer, all of them with a sec-ondary hydroxyl group, an odd number of methylene units between the primary andsecondary functionalities, and predominantly even-numbered overall carbon chainlength. Both types of bifunctional monomers hence had a strict 1,X-constellationof functionalities, where X was an odd number. These commonalities indicate thatboth bifunctional monomers co-occurring in the same plant species may be biosyn-thetically related and, based on the specific isomer and homolog distributions, thattheir common biosynthetic pathway may be hypothesized. This should aid and en-able future biochemical investigations into the machinery involved in the productionof these specialty compounds. Towards this end, discussed here are possible reactionsteps leading to β-hydroxy FA esters and alkanediol esters.4.3.1 Potential pathway leading to β-hydroxy FA estersThe F. hygrometrica β-hydroxy FA esters had predominantly even-numbered totalcarbon chain lengths ranging from C40 to C52, resulting from combination of acyl andalkyl portions each between 20 and 26 carbons in length. Furthermore, the additionalsecondary hydroxyl group was found exclusively on C-3 of the acyl moiety. Thesecharacteristics suggest that the β-hydroxy functionality is biosynthesized either byintroducing the hydroxyl group into a pre-existing fatty acyl chain or as a by-productof chain elongation.1024. Hydroxy esters from Funaria hygrometricaTwo scenarios may be distinguished for introduction of a hydroxyl group into apre-existing fatty acyl chain. In one, free or esterified acyl chains may be hydroxylatedby a P450-dependent enzyme similar to the Arabidopsis mid-chain alkane hydroxylase(MAH1). However, MAH1 has only limited regioselectivity, repeatedly oxidizing upto three adjacent methylene units [72, 248]. Therefore, a P450 enzyme clearly distinctfrom MAH1 would have to be invoked for formation of the strict 1,3-geometry of func-tional groups observed in the F. hygrometrica β-hydroxy FA esters. Alternatively, the3-hydroxyl group might also stem from mitochondrial fatty acid β-oxidation. Whilethis would explain the specific location of the functional group, it would likely neces-sitate transport of intermediates from mitochondria to the endoplasmic reticulum asthe compartment for wax ester biosynthesis and origin of wax export [25]. Neither ofthe two scenarios involving hydroxylation of a pre-existing fatty acyl chain is able toaccount for the chain length distribution of the F. hygrometrica β-hydroxyacyls.To better explain the homolog specificity of β-hydroxyacyl formation, two morescenarios can be envisioned, with introduction of the hydroxyl group as a by-productof chain elongation. For one, a polyketide synthase (PKS) enzyme could use VLCacyl-CoAs as starters for condensation with one C2 unit from a malonyl-CoA to formβ-ketoacyls. Most plant PKSs use aromatic starters, but some have been shown toalso accept aliphatic substrates, albeit usually for multiple condensation rounds. Ac-cordingly, PKSs have been invoked in biosynthesis of certain wax constituents in otherplant species, mostly to form β-keto compounds [239], and thus a F. hygrometricaPKS generating VLC β-ketoacyls may be hypothesized. However, it would likely haveto be associated with a reductase to transform the ketoacyls into the correspondingβ-hydroxy compounds. There is no precedence for a PKS/reductase complex withactivity towards VLC substrates.Finally, the β-hydroxyacyl compounds could be diverted from normal wax biosyn-thesis, where they would occur as intermediates during FA elongation. In the fattyacid elongase (FAE) complex, a ketoacyl-CoA synthase (KCS) and ketoacyl-CoA re-ductase (KCR) generate a β-hydroxyacyl intermediate that is usually channeled tothe dehydratase and enoylacyl-CoA reductase (ECR) components of the complex fortransformation into the elongated acyl-CoA [83]. However, it is feasible that the β-hydroxyacyl intermediate may be either released from the FAE complex or actively in-1034. Hydroxy esters from Funaria hygrometricatercepted by an additional enzyme. A specific thioesterase intercepting intermediatesbetween the KCR and the dehydratase of the FAE may be implicated in formation ofthe F. hygrometrica β-hydroxyacyls. Precedence exists for similar thioesterases, how-ever with specificity for β-keto intermediates (intercepted after condensation ratherthan reduction) of medium-chain fatty acyl-ACPs (of the fatty acid synthase complexrather than the FAE) during the biosynthesis of methyl ketones in tomato [254].The predominance of the C24 β-hydroxy fatty acyl moiety in the F. hygrometricaesters suggests that potential intercept/release from the FAE might occur mainly dur-ing the elongation round from C22 to C24 acyl-CoA (Figure 4.7, first dashed arrow).The presence of the other less abundant β-hydroxyacyl chain lengths in the ester prod-ucts may then be explained by intercept/release occurring to a lesser extent duringthe adjacent rounds of elongation, i.e., those that produce C22 and C26 compounds.Most interestingly, the current results on homolog distribution of F. hygrometricaβ-hydroxyacyls match those of previous reports on 1,3-bifunctional wax compoundsin diverse higher plants. For example, C24 and C26 1,3-diols and hydroxyaldehydeswere found in Ricinus communis [234], C22 and C24 1,3-diols have been identified inPapaver orientale [100], and C20, C22 and C24 1,3-diols (accompanied by 1,2-diols)in Cosmos bipinnatus [38]. Overall, this points to the importance of the elongationround from C22 to C24 acyl-CoAs for the formation of 1,3 bifunctional compounds.However, a recent study showed similar β-hydroxy fatty acyl products in the wax ofAloe arborescens, but with chain lengths around C28, and therefore thought to ariseduring elongation from C26 to C28 acyl-CoA [170].In contrast with previously characterized 1,3-bifunctional wax compounds, thoseof F. hygrometrica were esterified, suggesting involvement of a wax ester synthasesimilar to those involved in formation of esters lacking secondary hydroxyl functions.It is well established that these enzymes use fatty acyl-CoAs rather than free acidsas substrates [124]. Consequently, a free β-hydroxy FA generated by a thioesteraseupon interception at the FAE would have to be re-activated into the coenzyme A(CoA) ester, likely by the action of a long-chain acyl-CoA synthase (LACS) enzyme.Alternatively, a passively released β-hydroxyacyl-CoA might be immediately availablefor esterification. In this context, it should be noted that no free β-hydroxy fatty acidswere found in the waxes from various moss surfaces investigated here, however this1044. Hydroxy esters from Funaria hygrometricacannot be interpreted as evidence for either intermediate interception or release.While VLC β-hydroxy FA alkyl esters have not been previously identified inplants, C24 C32 β-hydroxy fatty acid methyl esters were recently identified in Aloearborescens cuticular waxes [170]. The hypothesized biosynthetic route to the acylportion of these compounds was similar to that outlined above. However, the latterportion of biosynthesis is likely quite different between the methyl and alkyl esters.In the absence of methanol (MeOH) in biological contexts, methyl esters are typicallyformed by methylation of the carboxyl oxygen with S-adenosyl methionine [48, 149].This is in contrast with the known biosynthesis of VLC alkyl esters, and likely by ex-tension β-hydroxy acid esters, which occurs via acyl transfer onto a hydroxyl oxygen.4.3.2 Potential pathway leading to 1,X-alkanediol estersThe second series of F. hygrometrica wax esters contained characteristic 1,X-alkanediols. When compared with the β-hydroxyacyl monomers, these diol monomershad a larger number of methylene units between the two functional groups (1,5-; 1,7-; and 1,9-geometry), longer overall chain lengths (mostly C30) and a hydroxyl asterminal functional group. Relatively little chain length and isomer variability wasobserved in the diol moieties of the esters, with C30 1,7-diol largely dominating. Thus,the chain length of the acyl portion was the sole determining factor of the total esterchain length, which ranged from C40 to C52 and was strongly dominated by even-numbered homologs.The secondary hydroxyl function of the 1,X-alkanediols could be generatedthrough P450 hydroxylase activity, β-oxidation, PKS-based elongation, or FAEintermediate intercept/release, similar to the mechanisms discussed above for β-hydroxyacyls. However, regioselectivity and literature precedence again favor abiosynthetic sequence that begins with intercept/release of a β-hydroxyacyl interme-diate of elongation by the FAE complex. The two pathways to VLC diol esters andβ-hydroxy FA esters thus could share the β-hydroxyacyl-CoA (and/or free β-hydroxyfatty acids) as intermediate(s), but they diverge beyond this point (Figure 4.7). Whileβ-hydroxy fatty acid ester biosynthesis might require only one more step (acyl trans-fer), the hypothesized pathway to diol esters is more complex.Formation of the predominant C30 1,7-diol structure may be explained by inter-1054. Hydroxy esters from Funaria hygrometricacept/release of the β-hydroxyacyl intermediate of the elongation round leading fromC24 to C26 acyl-CoA. The resulting C26 β-hydroxy fatty acyl-CoA could then re-enteran FAE for two more elongation rounds to yield C30 7-hydroxyacyl-CoA (Figure 4.7,bold arrows). Similarly, other β-hydroxyacyl intermediates might be intercepted inthe elongation rounds leading to C28 and C24 acyl-CoA, which upon being elongatedby one or three rounds, respectively, would generate C30 5- and 9-hydroxyacyl-CoAs.The co-occurrence of esterified and free 1,X-alkanediols in the same moss struc-tures strongly suggests that both compound classes are biosynthetically related, mostlikely with the free diols serving as precursors for esterification. This implies that the5-, 7-, and 9-hydroxyacyl-CoA intermediates first have their carboxyl head groupsreduced, leading to 1,5-; 1,7-; and 1,9-C30 diols, respectively. Such a reduction couldpotentially be accomplished by a fatty acyl reductase fatty acyl-CoA reductase (FAR)enzyme similar to those involved in acyl reduction to ubiquitous wax alkanols. Theresulting free 1,5-; 1,7-; and 1,9-diols could then be exported to the cuticle, or couldbe available for esterification. It should be noted that similar free 1,7-diols had beendescribed as cuticular wax components in various Papaver species [100], in homologand isomer mixtures similar to those found for F. hygrometrica, albeit in free (notesterified) form. Other 1,X-diols have been reported from the fern Azolla filliculoidesand several algae species, though were found as 1,13- through 1,19-diols (inclusive)[137, 200, 237], indicating that the biosynthesis of these compounds may be differentfrom the products described here.While a fraction of the F. hygrometrica 1,X-alkanediols are exported to the cuticlewithout further modification, the majority were observed to be esterified with VLCacids. It seems likely that the necessary acyl transfer may be catalyzed by a waxester synthase similar to those involved in normal wax ester biosynthesis, and alsoto the enzyme invoked in β-hydroxy FA ester biosynthesis above. Similar diol estershave been found, for example, in the stem bark of Tectona grandis as a mixture of 7-hydroxyoctacosyl decanoate, 18-hydroxyhexacosyl decanoate, and 20-hydroxyeicosyllinolenate [107], in Azolla filliculoides ferns as 1,13-; 1,15-; and 1,17-C48 alkanediolesters [200], and in Laurus nobilis fruits as a 1,19-C42 alkanediol ester [65]. Pea weevilinsects also produce a VLC diol ester (22-hydroxydocosyl hydroxypropanoate) thatserves as a mitogen [157]. Finally, esters of C30 and C32 1,11- and 1,13-hydroxyketones1064.HydroxyestersfromFunariahygrometricaSOCoASOCoAKCSKCRHCDECROHSOCoASOCoAKCSKCRHCDECRSOCoAOHSOCoASOCoAKCSKCRHCDECRSOCoASOCoAKCSKCRHCDECROHOHSOCoAC20H41elongationto C24C22C20H41C20H41C20H41C20H41C20H41C20H41C20H41C20H41C20H41OHROOHC20H41OHOHC20H41FARSOCoAC20H41OHWSOOC20H41OHCnHn+1OOHC20H41OCnHn+1Normal elongationHydroxy ester biosynthesis via intercept-elongationWSelongationto C26elongationto C28elongationto C30intercept / releaseintercept / releasere-entryFigure 4.7: (Continued on the following page.)1074. Hydroxy esters from Funaria hygrometricaFigure 4.7: Biosynthesis model for the hydroxy esters found in Funaria hy-grometrica waxes. Normal arrows indicate reactions in the biosynthesis of normalVLC FA elongation. Dotted arrows indicate the hypothesized pathway to β-hydroxyfatty acid esters via a wax ester synthase (WS). Bold arrows indicate the proposedbiosynthesis of the 1,7-diol esters via a fatty acyl reductase FAR and WS. Elongationis carried out by four enzymes, the names of which have been abbreviated: KCS ,ketoacyl-CoA reductase KCR, hydroxyacyl-CoA dehydratase HCD, and enoylacyl-CoA reductase ECR.have been identified in wax from Osmunda regalis fronds [102]. Nevertheless, VLC1,X-alkanediol esters have not been described before as components of cuticular waxes.4.4 ConclusionsIn summary, two novel compound classes were identified in the waxes of the mossFunaria hygrometrica, both with characteristic homolog and isomer patterns. Basedon their common chemical features, it is proposed that the bifunctional monomersoriginate through intercept and/or release of β-hydroxyacyl intermediates of the FAEcomplex followed by further modification. On the one hand, the β-hydroxyacyl-CoAintermediates from the FAE may be esterified with wax alkanols, or they may re-enter into the FAE cycle following which subsequent FAR reduction may lead to 1,5-;1,7-; and 1,9-diols. These may then be esterified with fatty acids (Figure 4.7). Theshared biosynthetic elements may be fine-tuned in each moss structure to generatethe respective isomer and homolog distributions of the compounds identified.The proposed biosynthetic pathways explain the observed similarities and differ-ences between the two novel natural product classes and require very few new ormodified enzymatic steps. It seems plausible that either a special FAE allowing re-lease of β-hydroxyacyls or recruitment of a thioesterase could generate the necessaryprecursors. Furthermore, slight modifications in the substrate specificities of certainknown wax biosynthetic enzymes could then account for the biosynthesis of the dis-tinct homolog distributions of these two major classes of wax compounds in the moss.However, it should be emphasized that, based on the current chemical data alone,these biosynthetic pathways can be tentatively laid out but not proven. Importantquestions arising from this discussion include the involvement of a thioesterase in-1084. Hydroxy esters from Funaria hygrometricateracting with (one of) the FAE complex(es) in F. hygrometrica, and the alternativepathways invoking PKS or P450 enzymes for the formation of secondary functionalgroups in the 1,X-bifunctional compounds found here. Together, the implied modelspresent directions for future research.4.5 Experimental4.5.1 Growth conditions, wax extraction, purification, andtransesterificationCollections of F. hygrometrica were made from four Connecticut populations withdeveloping sporophytes (CONN Budke #142, #144, #145; Goffinet #9027). Sporesfrom several sporophytes per site were used to establish laboratory populations foreach locality. Laboratory populations were grown, cold treated to stimulate repro-duction, and further cultivated to produce sporophytes as described previously [32].Each wax sample contained the cuticular wax from twenty square centimeters ofplant surface area (74 leafy gametophytes, 465 calyptra, or 235 sporophyte capsules).Three samples of each moss structures cuticular waxes were prepared. To create asample, each structure was chemically extracted by rinsing twice in 4 mL chloroform(CHCl3) (Aldrich), for 30 s each time. Cut ends of sporophyte capsules and leafygametophytes were held out of the CHCl3 to avoid extraction of internal lipids. Theextracts from both rinses were pooled, filtered, and n-tetracosane (5 µg) was addedas an internal standard. The CHCl3 was evaporated by air-drying overnight prior toanalysis.Isolation of individual compound classes, where necessary and/or possible, wasaccomplished by loading a wax extract or reaction mixture onto a thin layer chro-matography (TLC) plate of 0.25 mm thickness (Analtech), which was then developedwith CHCl3:EtOH (99:1). The plates were stained with primuline (Aldrich) dissolvedin acetone, and bands were visualized under ultra-violet (UV) light (365 nm). Bandswere excised, adsorbed compounds extracted from silica using CHCl3, and the solventevaporated. Transesterification of resulting mixtures was carried out by dissolving in100 µl of 14% BF3 in MeOH (Aldrich), incubating at 70◦C for 18 hours, then evapo-rating excess reagent under a stream of N2.1094. Hydroxy esters from Funaria hygrometrica4.5.2 Derivatization and gas chromatography - mass spec-trometry (GC-MS) analysisDerivatization was accomplished by dissolving each sample in 10 µl of both pyri-dine (Aldrich) and bis-N,O-trimethylsilyltrifluoroacetamide (Aldrich), reacting at70◦C for 45 min, evaporating excess reagents at 60◦C under a stream of N2, andfinally dissolving in CHCl3 (10 µl).Wax analyses were carried out on a 6890N Network gas chromatography (GC)(Agilent) equipped with a HP-1 capillary column (Agilent, length 30 m, i.d. 320 µm,1 µm film thickness). 1 µl of each sample was injected on-column into a flow of Heprogrammed for constant flow at 1.4 mL/min. The GC oven was held at 50◦C for 2min, followed by a 40◦C/min ramp to 200◦C, held at 200◦C for 2 min, increased by3◦C/min to 320◦C, and held at 320◦C for 30 min. Analytes were detected with anAgilent 5793N Mass Selective Detector (EI 70 eV; m/z 50800, 1 scan s−1).4.5.3 Synthesis of authentic standardsDry solvents, when required, were dried over activated molecular sieves for 24hours. Other materials were used as received from the supplier. Silica gel columnchromatography (CC) was performed using Silicycle SilicaFlash G60 (60-200 µm size,70-230 mesh). 1H-NMR spectra were obtained using a Bruker Avance 300 MHznuclear magnetic resonance (NMR) spectrometer (CDCl3, 25◦C). TLC comparisonof synthetic and biological compounds was performed using analytical silica gel 60TLC plates on aluminum (Merck) developed in CHCl3:EtOH (99:1).Docosyl 2-bromoacetate (3.2)Docosanol (200 mg, 1 eq.) was dissolved in dry dichloromethane (CH2Cl2), thenbromoacetyl chloride (113 mg, 60 µl, 1.2 eq.) and pyridine (30 µl, 0.6 eq.) were added.When TLC (developed in toluene) indicated alcohol consumption, the reaction waswashed with hydrocloric acid (HCl) (1M), and the mixture was loaded onto a silicacolumn. Eluting with hexane:toluene (2:1) yielded docosyl 2-bromoacetate, whichwas used in the next step without further purification (50 mg, 20% crude yield).1104. Hydroxy esters from Funaria hygrometricaDocosanal (3.3)Docosanol (200 mg, 1 eq.) was dissolved in dry CH2Cl2 (20 mL), and phyridiniumchlorochromate (PCC) (130 mg, 1 eq.) was added. The mixture was stirred at roomtemperature (RT) overnight, after which the solvent was evaporated, the remainingsolid dissolved in hexane:CHCl3 (2 mL, 1:1) and purified by CC using hexane:CHCl3(1:1) to afford docosanal (139 mg, 70% yield).Docosyl 3-hydroxytetracosanoate (3.4)Docosyl 2-bromoacetate (50 mg, 3.2, 1 eq.), docosanal (29 mg, 3.3, 0.7 eq.), zincpowder (13 mg, 1.5 eq.), and a small chip of iodine were dissolved in dry tetrahydrofu-ran (THF) (20 mL). Heating was carried out until reflux began and the temperaturewas held overnight at 70◦C. When the consumption of the aldehyde was confirmedby TLC (developed in toluene) the reaction mixture was washed with HCl (1M) andextracted with Et2O (x3). The combined extracts were concentrated and loaded ontoa preparative TLC plate alongside a small amount of unknown series B as purifiedfrom the leafy gametophyte wax extract. The plate was developed in toluene andstained with primuline. One band from the reaction mixture co-migrated with thenatural product, which was then excised and the CHCl3 extract was analyzed withGC-MS. Yield was not recorded.1-Bromohexanol (3.6)To a stirred solution of 1,6-hexanediol (6.0 g, 1 eq.) in toluene, hydrobromicacid (HBr) (7 mL, 9 M, 1.3 equivalents) was added and allowed to react at 80◦C for20 hours. Then the reaction mixture was extracted with Et2O (x3), the combinedextracts dried over Na2SO4, filtered, and the solvent was removed under reducedpressure. The resulting liquid was purified by CC using hexane:Et2O (1:1) as mobilephase. From this, 1-bromohexanol was obtained (7.073 g, 73% yield). 1H-NMR (300MHz, CDCl3): δ 3.63 (2H, t, J=6.6 Hz) and 3.40 (2H, t, J=6.6 Hz) from CH2Br andHOCH2,2 1.86 (2H, q, J=7.2, CH2CH2Br)3, 1.20-1.60 (6H, m, aliphatic CH).2resonances from these protons are predicted by ChemDraw to overlap3chemical shifts assigned based on prediction by ChemDraw1114. Hydroxy esters from Funaria hygrometrica2-((6-Bromohexyl)oxy)tetrahydro-2H-pyran (3.7)1-bromohexanol (7.0 g, 1 eq.) was dissolved in CH2Cl2 and cooled with an icebath, then dihydropyran (5.3 mL, 1.5 eq.) and pyridinium p-toluenesulfonate (50mg, 0.05 eq.) were added. After stirring overnight, the mixture was washed withH2O (x2), once with brine, dried with Na2SO4, filtered, and the solvent was removedunder reduced pressure. The resulting solid was dissolved in hexane:Et2O (5 mL,20:1), loaded onto a silica column, and eluted using hexane:Et2O (20:1). Productcontaining fractions were combined and the solvent removed to afford compound 3.7(6.442 g, 61% yield).Tetracosanol (3.10)A cold slurry of lithium aluminum hydride (LAH, 110 mg, 3 eq.) in THF underN2 was prepared. Tetracosanoic acid (400 mg, 1 eq.) was dissolved in dry THFand added dropwise to the slurry through a septum using a syringe. The reactionwas stirred overnight at RT and then quenched with H2O, the mixture was washedwith dilute HCl and extracted with Et2O (x3), the combined extracts were driedwith Na2SO4, filtered, and the solvent was evaporated under reduced pressure. Thisafforded tetracosanol (0.3162 g, 82% yield).Tetracosanal (3.11)Tetracosanol (300 mg, 9, 1 eq.) and pyridinium chlorochromate (215 mg, 1.3 eq.)were dissolved in 50 mL dry DCM. The mixture was gently heated until the solidsdissolved and stirred overnight at room temperature. Then the solvent was removedunder reduced pressure, the remaining solid was dissolved in CHCl3 (5 mL), thesolution was filtered through a short silica column, stirred with activated charcoal,and then filtered through another short silica column to yield tetracosanal (206 mg,69% yield).1,7-Triacontanediol (3.12)Mg powder (30 mg, 2 eq.) and a chip of iodine were suspended in dry THF (2mL). Dry tetracosanal (200 mg, 10, 1.3 eq.) and dry compound 3.7 (188 mg, 1 eq.)1124. Hydroxy esters from Funaria hygrometricawere each dissolved in dry THF (5 mL). Compound 3.7 was added dropwise at 70◦Cand the mixture was refluxed for 3 hours; 3.11 was then added dropwise and thereaction cooled and stirred overnight. The reaction was then quenched with H2O,the mixture acidified with HCl and extracted with Et2O (x3), the combined extractswashed with saturated NaCl, dried with Na2SO4, filtered, and the solvent evaporatedunder reduced pressure. A small portion of the resulting solid was loaded onto a TLCplate and developed with CHCl3, from which approximately 10 mg of the desiredproduct, 1,7-triacontanediol, were recovered and analyzed with GC-MS. Yield wasnot recorded.7-Hydroxytriacontyl palmitate (3.12)1,7-triacontanediol (2 mg) was dissolved in dry CH2Cl2 (50 µl), to which wasadded dimethylaminopyridine (DMAP) (1 mg) and palmitoyl chloride (0.1 µl). Afterone hour stirring, the entire mixture was loaded onto an analytical TLC plate and de-veloped in CHCl3:EtOH (99:1). GC-MS analysis of the three resulting bands revealedthey contained palmitic acid, DMAP, and 7-hydroxytriacontyl palmitate. Yield wasnot recorded.4.6 Supplementary dataSupplementary data for this chapter can be found in Appendix CC.1 Relative quantification of methanolysis products obtained from TLC purifi-cation of unknown series A.C.2 A) Potential paths to the fragments observed in the mass spectra of C46β-hydroxy fatty acid esters.C.3 The mass spectrum of docosyl 2-bromoacetate (3.2) and potential sources ofobserved characteristic fragments.C.4 A) The mass spectrum and structure of docosanal (3.3). B) The mass1134. Hydroxy esters from Funaria hygrometricaspectrum of tetracosanal (3.11). C) Potential paths to observed major fragments.C.5 A) Potential paths to the fragments observed in the mass spectra of C46β-hydroxy fatty acid esters.C.6 A) The mass spectrum and structure of 2-((6-bromohexyl)oxy)tetrahydro-2H-pyran (3.7). B) Potential paths to observed major fragments.C.7 The mass spectrum of tetracosanol (3.10), and potential sources of observedmajor fragments.C.8 Identification of octacosane-1,3-diol in the leafy gametophyte and gameto-phyte calyptra waxes of F. hygrometrica.C.9 Identification of triacontane-1,7-diol in the sporophyte capsule of F. hygro-metrica.114Chapter 5The moss Funaria hygrometricahas cuticular wax similar tovascular plants, with distinctcomposition on leafy gametophyte,calyptra, and sporophyte capsulesurfaces5.1 IntroductionPrimary, aboveground plant surfaces are covered with a layered, waxy cuticlethat decreases water loss across their vast surfaces [179]. The cuticle consists of twocomponents, cutin and waxes, both biosynthesized by epidermal cells [252]. Cutinis a hydroxy acid- and dicarboxylic acid-rich polyester matrix deposited outside theepidermal secondary cell walls [86]. Within and on top of this scaffold is the cu-ticular wax, a mixture of very-long-chain (VLC) aliphatic compounds that usuallyincludes alcohols, wax esters, aldehydes, alkanes, and fatty acid (FA)s [99]. It isthe accumulated wax that forms the transpiration barrier and seals the plant surface[187, 80, 92], and thus was a key innovation to provide protection in non-aqueous1155. Cuticular waxes of Funaria hygrometricaenvironments during the colonization of dry land [53]. Determining the similaritiesbetween the waxes of distantly related plant lineages, such as mosses and vascularplants, is the first step toward envisioning the cuticle of their most recent commonancestor.Chemical analyses of several moss species have revealed that at least some oftheir surfaces are coated with waxes that constitute ca. 0.1% of the moss dry weight[79, 249, 31], and micro-relief typical of crystalline wax projections has been detectedon many species by scanning electron microscopy (SEM) [167, 152, 113]. However,due to the small size and complex geometry of moss structures, the amount of waxthey accumulate per surface area has not been determined as is commonly done forplants with large, relatively flat leaves. Accordingly, how much wax per unit areathese early diverging plants can amass compared to well-studied vascular plants isunknown.Moss surfaces can be coated with any or all of the ubiquitous wax compoundclasses found on vascular plants [79, 249, 31], and in some cases with specialtywax compounds [78, 151, 39]. For example, leafy gametophytes of Pogonatumurnigerum are covered with mainly aldehydes, while leafy gametophytes of Pogo-natum aloides and Andreaea rupesteris bore principally wax esters and FAs [79],those of Physcomitrella patens yielded mainly alcohols [31], and those of Syntrichiacaninervis alkanes [249]. In contrast, sporophytes of Polytrichales species were cov-ered mainly with the specialty wax compound 10-nonacosanol [151]. Thus, diversewax compositions exist among the moss cuticles that have been analyzed, however,they are few in number and waxes from both the offspring and maternal structuresof a single species have not been analyzed. Accordingly, it is unclear to what degreethe waxes on different surfaces of a single moss species may vary in composition andoverall amount.The presence of all five ubiquitous compound classes in moss waxes suggests thatmoss wax biosynthesis may be similar to that of well-studied vascular plants. Inthese, wax biosynthesis begins with the elongation of long-chain fatty acyl-CoAs toproduce a range of VLC acyl-CoAs [185]. The decarbonylation pathway then producesaldehydes with corresponding chain length ranges and proceeds to alkanes through theremoval of the terminal aldehyde head group carbon [24]. In parallel, VLC acyl-CoAs1165. Cuticular waxes of Funaria hygrometricapass through the acyl reduction pathway to form alcohols as well as wax esters, whichare wax dimers linking FAs and alcohols [182, 124]. Free FAs are also a ubiquitouswax compound class, however their biosynthesis is not fully understood. To clarifythe extent to which moss wax biosynthesis is similar to that of higher plants and toaid future biochemical investigations, detailed wax analyses of more moss species arerequired.Mosses have two multicellular phases that are both exposed to the surround-ing environment and are covered by a cuticle. The haploid leafy gametophyte islargely dominant and often lacks internal water conduction, instead relying on wa-ter uptake from the surrounding environment directly through the cuticle and cellwalls, and is often desiccation-tolerant [168]. In contrast, the desiccation-resistantdiploid sporophyte relies on the leafy gametophyte for water as it grows and elevatesan undifferentiated group of cells above the protection of the boundary layer on astalk. Meanwhile, a cap of maternal gametophyte tissue called the calyptra coversthe immature sporophyte apex during early development, functioning to protect thedehydration-sensitive cells underneath prior to and during capsule formation, meio-sis, and spore production [34]. At maturity, the calyptra falls off, and the spores arereleased and dispersed.Funaria hygrometrica is a well-studied model for moss biology [64, 147, 193, 129],however, it was long unclear if the leafy gametophyte and protective calyptra boasteda cuticle. Transmission electron microscopy transmission electron microscopy (TEM)investigation revealed the presence of a characteristic lipophilic coating on both struc-tures, and that on the calyptra was thicker than that of either the leafy gametophyteor the sporophyte capsule [32]. Further analyses determined that during early sporo-phyte development the calyptra cuticle is fully formed and enhances capsule devel-opment and spore production by protecting the young sporophyte from dehydration,thus functioning similarly to cuticles on vascular plants [33, 34]. In contrast, the leafygametophyte of F. hygrometrica is likely desiccation-tolerant and can absorb watervia its cuticle, similar to the gametophytes of many other moss species. Thus, thegametophyte calyptra and leafy gametophyte of F. hygrometrica play distinct rolesin the life cycle, and the ecophysiological functions of their cuticles may differ greatly.However, the exact water transport properties of these cuticles are unclear, and com-1175. Cuticular waxes of Funaria hygrometricaprehensive analyses of the cuticular waxes covering the F. hygrometrica structures,including the sporophyte, are lacking. I discovered the presence of novel β-hydroxyFA esters and diol esters in the gametophyte and sporophyte waxes of F. hygrometrica[39], but this needs to be complemented with comprehensive, quantitative informationon all wax constituents.To address the issues outlined above, presented here is a comprehensive analy-sis of the coverage and composition of the waxes on each main aerial surface of F.hygrometrica. In particular, I aimed to determine 1) the wax coverage on the leafygametophyte, calyptra and sporophyte of F. hygrometrica, and 2) the compositionaldifferences between the three moss structures. Overall, such chemical data may en-able comparisons of the wax biosynthesis machinery between different moss structuresand that of vascular plants.5.2 ResultsThis study provides a comprehensive analysis of the wax mixtures covering theleafy gametophyte, gametophyte calyptra, and sporophyte capsule surfaces of Funariahygrometrica. Triplicates of each moss structure were sampled by surface extractionand investigated first with gas chromatography - mass spectrometry (GC-MS) toidentify individual homologs of various compound classes, and then by gas chro-matography - flame ionization detection (GC-FID) to quantify them. However, posi-tional isomers could not be separated under the gas chromatography (GC) conditionsemployed here, so their distributions were assessed using further GC-MS analyses.Based on the relative abundances of characteristic MS fragments within each GCpeak, isomer distributions were calculated for all wax ester classes.5.2.1 Total wax amounts, compound class and chain lengthdistributionsThe F. hygrometrica leafy gametophyte, calyptra, and sporophyte capsule surfaceswere covered by 0.94 ± 0.13 µg/cm2, 2.0 ± 0.2 µg/cm2, and 0.44 ± 0.10 µg/cm2 oftotal wax, respectively (Fig. 5.1). Within each of the three wax mixtures, compoundswith an additive coverage of 0.83 µg/cm2 (88%), 1.8 µg/cm2 (92%), and 0.32 µg/cm21185. Cuticular waxes of Funaria hygrometricaFigure 5.1: Wax coverage on three major structures of Funaria hygromet-rica. The total amount of wax from each structure is expressed in terms of wax mass(µg) per surface area extracted (cm2). Bar heights and error bars represent the meanand standard deviation of three independent samples, respectively. Letters indicatesignificant differences based on Tukey post-hoc tests (p < 0.01) and ANOVA (F2,6 =111.3, p < 0.001).(75%) were identified. The dry weight of the leafy gametophyte material was alsodetermined, revealing that the (GC-determined) wax amount represented 0.07% ofthe materials dry weight.On the leafy gametophyte, FA alkyl esters (67% of the total wax) and β-hydroxyFA alkyl esters (17%) were the most abundant compound classes, accompanied byFAs (1%), alcohols (4%), and diols (0.3%) (Table D.1). On the calyptra, the FA alkylesters (alkyl esters) were the major compound class (85%), with minor admixtures ofFAs (1%), alcohols (2%), diols (0.07%), β-hydroxy FA alkyl esters (4%), and alkanes(0.3%). The sporophyte capsule wax harbored mainly alkyl esters (40%), togetherwith FA esters of primary, primary,secondary alkanediols (28%), alkanes (6%), FAs(0.2%), alcohols (3%), diols (0.6%), and aldehydes (0.6%).Many of the compounds identified on the leafy gametophyte were present as ho-mologous series. The detected FAs had even carbon numbers ranging from C20 toC26, with relative abundances increasing towards the longer chain lengths (Fig. 5.2A).The homologous series of alcohols had even carbon numbers that ranged from C22 toC30, with a bimodal distribution peaking at C22 and C28. FA alkyl esters had witheven and odd carbon numbers between C36 and C54, and a relatively broad maximumaround C44, C46, and C48. Similarly, the β-hydroxy FA alkyl esters had even carbon1195.CuticularwaxesofFunariahygrometricaFigure 5.2: (Continued on the following page.)1205. Cuticular waxes of Funaria hygrometricaFigure 5.2: Relative abundance of wax compounds on three major structuresof Funaria hygrometrica. The amount of each wax compound is expressed as apercent of the total amount of wax extracted from the leafy gametophyte (A), thegametophyte calyptra (B), and the sporophyte capsule (C). Labels on the x-axisindicate the compound class and carbon number of each compound. Bar heights anderror bars represent the mean and standard deviation of three independent samples,respectively. Abbreviations: FAs = fatty acids, FA alkyl esters = Fatty acid alkylesters, 28 Diol = octacosane-1,3-diol, 30 Diol = triacontane-1,7-diol, β-OH FA esters= β-hydroxy fatty acid esters, Unk.= the sum of unidentified GC peaks.numbers ranging from C38 to C52, and with homologs C42 to C48 dominating. Onlyone diol, octacosane-1,3-diol, was detected in the leafy gametophyte wax.In the calyptra wax, largely the same individual compounds were detected as in theleafy gametophyte wax, leading to fairly similar overall chain length ranges within thecompound classes from both moss structures (Fig. 5.2B). In addition, the calyptra waxcontained β-hydroxy FA alkyl esters with the C42 and C44 homologs dominating, anda single diol, octacosane-1,3-diol. However, in contrast to the qualitative similarities,several of the chain length distributions within the calyptra differed quantitativelyfrom those in the leafy gametophyte wax. Shorter FA homologs had higher relativeabundances in the calyptra wax. Both the alcohols and the alkyl esters had bimodaldistributions, peaking at C22 and C30, and at C42/44 and C50/52, respectively.The sporophyte capsule yielded FAs with even carbon numbers between C20 andC24, all with relatively similar abundances (Fig. 5.2C). The alcohols with even carbonnumbers C22 to C30 were detected in a bimodal distribution, thus similar to those inthe calyptra wax. Sporophyte capsule wax had FA alkyl esters with both even andodd carbon numbers that ranged from C36 to C52 and were dominated by the C46homolog. In addition, FA esters of primary,secondary alkanediols (diol esters) witheven carbon numbers C42 to C52 were present in similar abundances, accompanied bysmall amounts of the C40 homolog. Finally, C28 and (predominantly) C30 aldehydeswere found, as well as C27 and (predominantly) C29 alkanes and triacontane-1,7-diol.1215. Cuticular waxes of Funaria hygrometrica5.2.2 Ester isomer distributionsUnlike other wax constituents, esters are dimers, formed by linking FA and alco-hol monomers. Esters of a single overall chain length can result from several differentcombinations of FA and alcohol homologs (e.g., C46 esters can be made up of C20acid + C26 alcohol, C22 acid + C24 alcohol, or other combinations), giving rise toisomerism within each ester homolog in a series. Therefore, the next stage of theanalysis was directed toward quantifying both the acid and alcohol chain length pro-files within individual ester homologs of substantial abundance, and the distributionof all esterified acids and alcohols across all ester homologs.In the wax of the leafy gametophyte, the prevalent C46, C48, and C50 ester ho-mologs all contained mainly C24 FA linked to C22, C24, and C26 alcohol, respectively(Tables D.2 and D.3, Fig. 5.3A and B). The accompanying shorter ester homologsconsisted of somewhat shorter FAs and alcohols, namely C20 acid and C20 alcohol,C20 acid and C22 alcohol, as well as C22 acid and C22 alcohol. Similarly, a longer(C52) ester homolog was formed by combination of somewhat longer building blocks,C26 acid and C26 alcohol. Finally, the esters with extreme chain lengths (C38 andC54) were composed of multiple, more evenly distributed acids and alcohols, includ-ing relatively short acids and alcohols (C18 to C22) in the C38 esters and extremelylong acids and alcohols (up to C32) in the C54 esters. Overall, the different alkyl es-ter homologs in the leafy gametophyte wax had fairly similar acid and alcohol chainlengths, thus positioning the ester functional group near the center of the overallbackbone structure.The alkyl esters in the calyptra wax may be grouped into two series of homologsdistinguished by their isomer patterns. The first group consisted of the C38 to C44 es-ter homologs, all characterized by having the same predominant alcohol (C22) linkedto varying acids C16 to C22 (Tables D.4 and D.5, Fig. 5.3C and D). In the other group,the C50 to C54 ester homologs each contained predominantly C30 alcohol, bound toC20 to C24 acids. The two groups of esters, distinguishable by their isomer compo-sitions, were also reflected in the bimodal distribution of ester homologs (compareFig. 5.2B). The isomer compositions of the remaining ester homologs (C46 and C48)were intermediate between the two dominating ester series.The alkyl esters of the sporophyte capsule could also be divided into two groups of1225.CuticularwaxesofFunariahygrometricaFigure 5.3: (Continued on the following page.)1235. Cuticular waxes of Funaria hygrometricaFigure 5.3: Isomer distribution within the FA alkyl esters in Funaria hy-grometrica waxes. Relative amounts of fatty acids (A, C, E) and alcohols (B, D,F) are expressed as a percent of the total amount of fatty acids or alcohols in eachFA alkyl ester homolog, respectively. The amount and chain length of the fatty acidsbound in a single FA alkyl ester can be read in the height and color or the bars abovethe carbon number corresponding to that FA alkyl ester. FA alkyl esters with vary-ing carbon numbers are plotted for the leafy gametophyte (A, B), the gametophytecalyptra (C, D), and the sporophyte capsule (E, F). Bar heights and error bars rep-resent the mean and standard deviation of three independent samples, respectively.Bars with rising black hashes represent the abundance of an ester-bound fatty acid.Bars with rising bold hashes indicate the abundance of an ester-bound alcohol. Barcolors denote fatty acid and alcohol chain lengths as indicated in the legend. In afew cases, alkyl esters of extreme chain length (C36 or C54) were above the limit ofhomolog quantification, but were too low in abundance to analyze for their isomercomposition.homologs. One group of esters (C50 and C52) was formed mainly by C30 alcohol boundto different acids (C20 and C22 acids, respectively) (Tables D.6 and D.7 Fig. 5.3Eand F). This group of esters thus had isomer compositions similar to those in thecalyptra wax. Conversely, another group of esters (C38 and C40) was formed mainlyby one acid (C16), linked to different alcohols (C22 and C24 alcohols, respectively).Neighboring ester homologs (C36 and C42) also contained substantial amounts of C16acid, together with acids of other chain lengths. These two ester groups intersectedat the most abundant ester homolog (C46), which consisted mainly of C16 acid andC30 alcohol.Finally, the overall chain length profiles of alkyl ester-bound FAs and alcohols ineach moss structure were determined by summing the amounts of constituent acids oralcohols with the same carbon number across all ester homologs. This revealed thatthe leafy gametophyte had alkyl ester-bound FAs with mainly even carbon numbersspanning C16 to C28, within which C20 to C24 were the most abundant (Fig. 5.4A).These acids were esterified predominantly with C22 to C26 alcohols (Fig. 5.4B). Onthe calyptra, mostly even carbon-numbered acid homologs ranging from C16 to C24were observed, with C20 and C22 predominating (Fig. 5.4C). In this moss structure,the corresponding ester alcohols had a distinct bimodal distribution, peaking at C22and C30 (Fig. 5.4D). The sporophyte capsule esters contained FAs ranging from C141245. Cuticular waxes of Funaria hygrometricato C30, in a bimodal distribution peaking at C16 and C22, and alcohols with a bimodaldistribution peaking at C22 and C30 (Fig. 5.4E and F).5.2.3 Hydroxy ester isomer distributionsβ-hydroxy FA esters may also occur as isomer mixtures, arising from the combina-tion of β-hydroxy FAs and alcohols of varying chain lengths. Again using diagnosticMS fragments, the isomer composition of each β-hydroxy FA ester homolog fromthe wax of each moss structure was determined. The prominent C42, C44, and C46β-hydroxy FA esters from the leafy gametophyte contained C20, C22, and C24 β-hydroxy FAs, respectively, mainly in combination with C22 alcohol (Tables D.8 andD.9, Fig. 5.6A and B). Longer β-hydroxy FA esters contained largely C24 β-hydroxyFA and alcohols ranging from C22 to C28. Thus, the isomer composition of the β-hydroxy FA esters resembled those of corresponding alkyl ester homologs in the leafygametophyte wax.On the calyptra, the prominent C42 and C44 β-hydroxy FA esters were formedby C20 and C22 β-hydroxy FAs together with C22 alcohol (Tables D.10 and D.11,Fig. 5.6C and D). Longer β-hydroxy FA esters contained mostly C20 β-hydroxy FA,esterified with particularly long alcohols ranging from C26 to C32.Next, the total profiles of esterified β-hydroxy FAs and β-hydroxy FA-bound alco-hols were determined using the same process described for the alkyl esters. Esterifiedβ-hydroxy FAs on the leafy gametophyte and the calyptra had mainly even carbonnumbers from C16 to C30. On the leafy gametophyte, C20, C22, and C24 β-hydroxyFAs were most abundant, while on the calyptra the C20 homolog was most common(Fig. 5.5A and C). The accompanying esterified alcohols had even carbon numbersfrom C14 to C36, with a normal distribution peaking at C22 on the leafy gametophyteand a bimodal distribution peaking at both C22 and C30 on the calyptra (Fig. 5.5Band D).Fatty acid esters of primary,secondary alkanediols (diol esters) not only exhibit thepotential for isomerism via combination of various FA and alkanediol chain lengths,but also through variability in the position of the secondary hydroxyl group on the diolmoiety. Inspection of the MS fragmentation patterns of the sporophyte capsule diolesters revealed that they contained such positional isomers. In particular, diagnostic1255. Cuticular waxes of Funaria hygrometricaFigure 5.4: Relative abundance of FA alkyl ester-bound fatty acids andalcohols in Funaria hygrometrica waxes. Relative amounts of fatty acids (A,C, E) and alcohols (B, D, F) are expressed as a percent of the total amount of fattyacids or alcohols from all FA alkyl ester homologs in the wax mixture from a singlestructure, respectively. FA alkyl esters with varying carbon numbers are plotted forthe leafy gametophyte (A and B), the gametophyte calyptra (C and D), and thesporophyte capsule (E and F). Bar heights and error bars represent the mean andstandard deviation of three replicate samples, respectively. Bars with rising blackhashes represent the abundance of an ester-bound fatty acid. Bars with rising boldhashes indicate the abundance of an ester-bound alcohol.1265. Cuticular waxes of Funaria hygrometricaFigure 5.5: Relative abundance of ester-bound β-hydroxy fatty acids andalcohols in Funaria hygrometrica waxes. Relative amounts of β-hydroxy FAs(A, C) and alcohols (B, D) are expressed as a percent of the total amount of FAsor alcohols from all the β-hydroxy FA ester homologs in the wax mixture from asingle structure, respectively. β-hydroxy FA esters with varying carbon numbersare plotted for the leafy gametophyte (A and B) and the gametophyte calyptra (Cand D). Bar heights and error bars indicate the mean and standard deviation ofthree independent samples, respectively. Bars with falling black hashing indicate theabundance of an ester-bound β-hydroxy fatty acid. Bars with rising bold hashingindicate the abundance of an ester-bound alcohol.1275.CuticularwaxesofFunariahygrometricaFigure 5.6: (Continued on the following page.)1285. Cuticular waxes of Funaria hygrometricaFigure 5.6: Isomer distribution within the β-hydroxy fatty acid esters inFunaria hygrometrica waxes. Relative amounts of β-hydroxy fatty acids (A, C)and alcohols (B, D) are expressed as a percent of the total amount of β-hydroxy FAsor alcohols in each β-hydroxy FA ester homolog, respectively. The amount and chainlength of the fatty acids bound in a single β-hydroxy FA ester can be read in the heightand color or the bars above the carbon number corresponding to that β-hydroxy FAester. Esters with varying carbon numbers are plotted for the leafy gametophyte (Aand B) and the gametophyte calyptra (C and D). Bar heights and error bars representthe mean and standard deviation of three independent samples, respectively. Barswith falling black hashes represent the abundance of an ester-bound β-hydroxy fattyacid. Bars with rising bold hashes indicate the abundance of an ester-bound alcohol.Bar colors denote specific chain lengths as indicated in the legend. Abbreviations:β-OH fatty acid (ester) = β-hydroxy fatty acid (ester).α-fragments indicated that the secondary hydroxyl function was located mainly onC-7 of the diol moiety, but with admixtures of corresponding C-5 and C-9 isomers ineach ester homolog (Fig. 5.7A). Across all homologs, the average ratio of esterified1,5-; 1,7-; and 1,9-diols was approximately 1:4:1 (Fig. 5.7B). It should be noted thatthe diol esters did not exhibit additional isomerism due to combinations of FA andalkanediol chain lengths, as did the other ester classes described above. Instead, allthe diol esters contained only C30 diols, esterified with varying FA homologs.5.3 DiscussionThe present study aimed to provide a comprehensive analysis of cuticular waxeson the aerial surfaces of the moss Funaria hygrometrica. I found that (i) the leafygametophyte, gametophyte calyptra, and sporophyte capsule were all covered by cu-ticular waxes, with amounts in the low end of the range reported for vascular plants(ca. 1 µg/cm2), that (ii) the waxes on all three moss structures contained large por-tions of alkyl esters with minor admixtures of other ubiquitous compound classes,that (iii) each moss structure was distinguished by its own characteristic compoundclass(es) and/or chain length profile(s), and that (iv) ester isomer patterns differedbetween ester classes and moss structures. In the following I will discuss each of thesefour major findings.1295. Cuticular waxes of Funaria hygrometricaFigure 5.7: Isomer distribution within diol esters in Funaria hygrometricasporophyte wax. For diol esters with varying carbon numbers, relative amounts ofdiol isomers are expressed as a percent of the total amount of diols in each diol esterhomolog (A), or of the total amounts of diols in all diol ester homologs (B). Barswith falling bold hashing represent the abundance of ester-bound diol isomers. Barheights and error bars represent the mean and standard deviation of three independentsamples, respectively. Bar colors denote specific chain lengths as indicated in thelegend.5.3.1 Cuticular wax coverage on Funaria hygrometricaTo put wax amounts into context with cuticle structure and function, and toenable comparison with vascular plant species, they are preferably measured as cov-erages (i.e., in units of µg per cm2 surface area). However, the complex geometry ofmosses makes measuring their surface area particularly challenging and has impededdetermination of their wax coverages thus far. In this study, the surface area of eachmoss structure to be extracted was first measured by counting pixels in a photographof dissected and flattened representative structures. In parallel, the geometricallysimplest structures, the calyptrae, were approximated with geometric shapes to cal-culate their surface areas. The results from the geometric method were within 5% ofthe values from the pixel counting approach, validating the latter.Based on the approximations for the surface areas of the moss structures, the leafygametophyte, calyptra, and sporophyte capsule of F. hygrometrica had wax coveragesof 0.5 to 2.0 µg/cm2. These findings enable direct comparisons with species of otherlineages for the first time. Leaves of seedless vascular plants such as the fern Osmundaregalis have a wax coverage of approximately 20 µg/cm2 [97], while those of other1305. Cuticular waxes of Funaria hygrometricaferns range from 5 to 15 µg/cm2 (Li and Jetter, unpublished). Wax coverage on seedplant surfaces also varies widely, for example reaching 20-30 µg/cm2 for needles ofgymnosperms such as Taxus baccata [246] and typically ranging from 1 to 30 µg/cm2on angiosperm leaves [172, 206, 251, 155]. Thus, the wax amounts on F. hygrometricasurfaces fall within the low end of the range of wax coverages on vascular plants.Prior to this study, total wax amounts on mosses had only been reported for leafygametophytes, and as wax amounts per plant dry weight rather than per surface area.To enable comparisons with these previous reports, the dry weight of F. hygrometricaleafy gametophyte material was determined in parallel with wax extraction. Usingthe gas chromatography (GC) results, the wax represented approximately 0.07% ofthe leafy gametophyte dry weight. This finding is similar to other moss leafy gameto-phytes, where wax amounts ranged from 0.02% dry weight for Syntrichia caninervis[249] and Physcomitrella patens [31] to 0.05%, 0.08%, and 0.12% for Andreaea ru-pestris, Pogonatum aloides, and Pogonatum urnigerum [79], and 5% for Saelaniaglaucescens [78].Finally, the wax coverages of the three F. hygrometrica structures can also becompared with previous reports on relative thicknesses of their respective cuticles.transmission electron microscopy (TEM) investigations showed the calyptra to havethe thickest cuticle of the three moss structures, more than twice that of the othertwo structures, when considering all lipophilic layers of the cuticle [32]. The chemicalresults presented here parallel the anatomical observations, revealing wax coveragesroughly proportional to apparent cuticle thicknesses. However, it must be noted thatmoss cuticles, like those of vascular plants, consist of both wax and cutin [31]. Hence,TEM data may reflect cutin amounts more than those of waxes, and can therefore beexpected to correlate only loosely with wax amounts.5.3.2 Cuticular wax constituents common to all surfaces ofFunaria hygrometricaThe wax mixtures from the leafy gametophyte, calyptra, and sporophyte capsuleall contained FAs, alcohols, and fatty acid (FA) alkyl esters, and the sporophytecapsule and calyptra also bore aldehydes and alkanes. Chain lengths ranged fromC20 to C30 for monomeric compounds, and from C36 to C54 for ester-linked, dimeric1315. Cuticular waxes of Funaria hygrometricacompounds. The same compound classes and chain lengths have been reported forthe cuticular wax mixtures of other mosses [79, 249, 31] as well as many vascular plantspecies, indicating that mosses can regularly have a full complement of the ubiquitouswax compounds. It is interesting to note that, therefore, entire wax-forming pathwayssimilar or identical to those in vascular plants are generally operational in mosses, anda full set of wax biosynthesis gene orthologs is likely present. It seems plausible thata common ancestor of mosses and vascular plants had the ubiquitous wax compoundbiosynthesis machinery in place, and that further differences in relative amounts ofwax products between various plant lineages would have resulted from differentialregulation of the common wax biosynthesis genes.In particular, the wax mixtures on the leafy gametophyte, calyptra, and sporo-phyte capsule of Funaria hygrometrica contained high percentages of alkyl esters.Interestingly, comparable alkyl ester amounts had been reported before for only rel-atively few plant species, including the leafy gametophytes of the mosses Andreaearupesteris and Pogonatum aloides [79], the leaves of vascular plants such as Camelinasativa [173], Copernicia cerifera [115], Quercus ilex [134], and the seed oil of Sim-mondsia chinensis [21]. However, it must be noted that some GC configurations andsettings may bias against alkyl esters due to their high molecular weights. For thisreason they may have been underestimated or entirely overlooked in some previousplant wax analyses.Alcohols and alkyl esters, the products of the acyl reduction biosynthetic pathway,accounted for more than 90% of the wax mass on each of the three F. hygrometricastructures. Similarly high amounts of acyl reduction products have been reportedfor diverse plant species, albeit mostly in the form of free (rather than esterified)alcohols, for example the mosses Physcomitrella patens [31], Andreaea rupesteris andPogonatum aloides [79], the ferns Osmunda regalis [97] and Pteridium aquilinum [9],the cycad genus Encephalartos [159], and the grass genera Triticum and Agropyron[217, 221].It is of note that both acyl reduction and decarbonylation pathway-dominatedwaxes have been found in mosses. For example, Pogonatum urnigerum leafy game-tophyte wax is rich in aldehydes, while Andreaea rupesteris and Pogonatum aloidesleafy gametophytes bear mostly alkyl esters [79], and Pogonatum belangeri and Pog-1325. Cuticular waxes of Funaria hygrometricaonatum urnigerum sporophytes are mostly covered with secondary alcohols [151].Overall, this shows that the balance of substrate flux between wax biosynthesis path-ways is set differently by each moss species, and even varies between closely relatedspecies, as in Pogonatum.5.3.3 Distinguishing features in the wax mixtures of thethree F. hygrometrica structuresThe parallel investigation of three F. hygrometrica structures enables compar-isons between their characteristic wax compositions. For example, the sporophytecapsule wax contained large quantities of alkanediol esters, accompanied by a smallpercentage of corresponding free diols. In sharp contrast, the same or similar di-ols were not detected in the wax mixtures on the gametophyte structures, and theymay thus serve as specific markers for the sporophyte capsule. It remains to be seenwhether comparable alkanediol esters occur in other plant species, now that their gaschromatography - mass spectrometry (GC-MS) characteristics are available for usein future wax analyses [39].The sporophyte capsule wax also contained substantial amounts of alkanes andtraces of aldehydes, while only minute amounts of alkanes were on the calyptra andneither was on the leafy gametophyte. Overall, the sporophyte capsule had a uniquewax composition with only partial resemblance to the wax mixtures on the game-tophyte moss structures. Conversely, the wax of the F. hygrometrica gametophytestructures comprised characteristic β-hydroxy FA esters that were absent from sporo-phyte capsule wax (or at best present in trace amounts below the detection limit).The β-hydroxy FA esters may therefore be regarded as unique to the gametophytestructures in this moss species.The F. hygrometrica calyptra and leafy gametophyte waxes were further distin-guished from each other by the presence and absence, respectively, of C29 alkane. Thecalyptra physically touches the sporophyte capsule during development, potentiallyleading to contamination of the calyptra wax preparation with minor amounts ofsporophyte capsule wax material. However, the absence of diol esters (as sporophytecapsule wax markers) in the calyptra samples rules out cross-contamination. Thus,the accumulation of alkanes distinguished the calyptra from the leafy gametophyte.1335. Cuticular waxes of Funaria hygrometricaThis suggests that the decarbonylation biosynthesis pathway is indeed active in onlyone of the gametophyte structures, the calyptra, and otherwise in the sporophytecapsule. Overall, the different F. hygrometrica structures all have distinct compoundclass compositions. This suggests that the balance of substrate flux between waxbiosynthesis pathways is set differently in each moss structure. This may reflect thedifferences in their ability to absorb water through their cuticle and to prevent waterloss.Lastly, the overall chain length distributions in the wax of each moss structurecan be assessed by examining the homolog profiles across all compound classes, in-cluding esterified compounds. The leafy gametophyte wax contained predominantlyC24 and C26 compounds, while the calyptra and sporophyte capsule waxes also boresubstantial amounts of C30 compounds. The major chain lengths involved in bothcases are reminiscent of those predominating in certain angiosperm waxes, suggestingthat the biosynthetic machineries determining chain length profiles may be similarin mosses and vascular plants. In Arabidopsis thaliana, the ketoacyl-CoA synthase(KCS) enzyme CER6 and the putative BAHD-family protein CER2 are known to,together, be important for chain length elongation from C24 and C26 towards C30[141, 84]. Elongation beyond C30 is facilitated by the CER2-like enzyme CER26.Several candidate CER6 orthologs have been annotated in the Physcomitrella patensgenome [119], but their putative roles in FA elongation have not been tested to date.In contrast, no CER2 -like genes (including CER26 ) were annotated in the P. patensgenome, and sequence similarity-based searches revealed putative BAHD proteins dis-tantly related to CER2 but no direct ortholog. It may thus be speculated that F.hygrometrica has an ortholog of CER6 and perhaps CER2, and that one (or both) ofthese genes is (are) differentially regulated between the leafy gametophyte (very littleexpression), the calyptra (low expression), and the sporophyte capsule (high expres-sion). Based on the absence of monomeric wax compounds with more than thirtycarbons, it seems likely that this moss species does not express a CER26 ortholog.Nevertheless, the wax biosynthesis machinery in both mosses and vascular plants haslikely been inherited from a common ancestor. This parallels other observations ofcompounds produced in both mosses and vascular plants by related genes, for exam-ple callose, where homologous biosynthesis genes have been identified in Arabidopsis1345. Cuticular waxes of Funaria hygrometricaand Physcomitrella [91, 188].5.3.4 Isomer patterns in different wax ester classes on thethree Funaria hygrometrica structuresWax esters are formed from the combination of two (very-)long-chain compounds,a fatty acyl component (bearing a carboxyl group) and an alkyl component (bearinga hydroxyl group). The possible incorporation of acyl and alkyl moieties with diversechain lengths (and maybe additional hydroxyl functions) into the broad array of F.hygrometrica esters warranted detailed analysis. Using the isomer data presentedhere, the chain length profiles within the diverse ester classes were explored to gaininsights into their biosynthesis.The chain lengths of wax ester-bound alcohols on the three F. hygrometrica struc-tures varied from C14 to C36 (Fig. 5.4 and Fig. 5.5). The overall range of free alcoholsin the total wax mixture of all F. hygrometrica structures matched those of respectiveesterified alcohols (Fig. 5.2), indicating that free alcohols may serve as substrates foralkyl ester formation. Previous analyses of wax ester composition in angiospermshave led to similar observations and conclusions [118, 173], and feeding experimentsfurther confirmed that the pool of wax alcohols serves as substrate for wax esterbiosynthesis in Arabidopsis [124].FA alkyl esters consist of FAs and alcohols. The chain length range of ester-boundFAs was C16 to C30 (Fig. 5.4), within which falls the chain length range of free FAs inthe total wax mixture. Thus, a wide range of acyl precursor chain lengths appears toserve as substrates for ester formation in F. hygrometrica. In conclusion, the overallchain length ranges of both esterified acyls and alkyls indicate that ester moietiesare directly recruited from respective wax precursor pools. This finding is in sharpcontrast to Arabidopsis thaliana, where the chain length profile of ester-bound FAspeaks sharply at C16 [118], likely reflecting predominance of this chain length in thesubstrate pool available to the ester synthase involved.While the qualitative information on ester isomer ranges reveals possible substratepools, the quantitative acyl and alkyl chain length profiles may further inform aboutsubstrate specificities of the ester synthase(s) involved. The F. hygrometrica leafygametophyte wax contained FA alkyl esters and β-hydroxy FA esters, both made up1355. Cuticular waxes of Funaria hygrometricaalmost exclusively of C20 to C24 acyl (Fig. 5.4A and Fig. 5.5A) and C22 to C26 alkylmoieties (Fig. 5.4B and Fig. 5.5B). However, the free FA and alcohol pools accom-panying these esters in the leafy gametophyte wax contained substantial amounts ofC26 FA, and C28 and C30 alcohols (Fig. 5.2). This indicates that in FA alkyl esterformation the C26 FA is selected against, while in both FA and β-hydroxy FA alkylester formation C28 and C30 alcohols are selected against (Fig. 5.8). The very similarsubstrate biases in the biosynthesis of both ester classes suggest that they are formedby the same enzyme, or by two enzymes with very similar characteristics.On the calyptra, FA esters and β-hydroxy FA alkyl esters were again both present.Here, ester-bound acyl moieties were mainly C18, C20, C22, and C24 in both esterclasses (Fig. 5.4C and Fig. 5.5C), while the ester alkyls had consistent, bimodaldistributions peaking at C22 and C30. The accompanying FAs and alcohols (Fig. 5.2)had homolog profiles very similar to those of ester acyls and alcohols (Fig. 5.8),suggesting that the ester-forming enzyme(s) in this moss structure does (do) notexhibit strong substrate chain length specificity.Finally, the F. hygrometrica sporophyte capsule wax contained both FA alkylesters and FA diol esters. The acyl moieties in the FA alkyl esters were mainly C16,C20, C22, and C24 (Fig. 5.4E), and the FA diol esters predominantly C16, C18, and C20acyls (calculated from the total carbon number of the diol ester by subtracting thecarbon number of the sole diol homolog, C30). Conversely, mainly C30 and some C22alkyl units were incorporated into the esters of the sporophyte capsule (Fig. 5.4F).The ester isomer profiles thus match those of free FAs and alcohols as well as diols inthe same wax mixture, except for the additional esterification of C16 acyl components.Overall, it seems that the ester-forming enzyme(s) in this moss structure have accessto substantial amounts of C16 fatty acyl-CoA, and that they may exhibit a preferencefor this substrate, unlike those in the other two structures.5.4 ConclusionsThe detailed analyses of wax esters on three unique F. hygrometrica structuressuggest that incorporation of FAs and alcohols into wax esters proceeds with charac-teristic bias against or preference for different substrate chain lengths in each moss1365.CuticularwaxesofFunariahygrometricaFigure 5.8: (Continued on the following page.)1375. Cuticular waxes of Funaria hygrometricaFigure 5.8: Comparison of ester-bound and free alkyl and acyl chains inFunaria hygrometrica waxes. Relative amounts of total ester-bound acyl chainsare plotted alongside their free acid counterparts (A, C, E), and relative amountsof total ester-bound alkyl moieties are plotted alongside corresponding free alcohols(B, D, F) for the leafy gametophyte (A, B), the gametophyte calyptra (C, D), andmature sporophyte (E, F). Black bars represent the abundance of free compounds asa percent of their compound class in the wax mixture from each structure, and hashedbars indicate the relative abundance of each esterified component in its compoundclass in the wax mixture from each structure. Bars with rising black hashes representthe abundance of an ester-bound fatty acid. Bars with rising bold hashes indicate theabundance of an ester-bound alcohol. Bar heights and error bars indicate the averageand standard deviation of three independent measurements, respectively.structure. These disparate patterns indicate that different enzymes may be active ineach. Nevertheless, wax coverage on all three moss structures were similar to thosefound on vascular plants. Additional compositional comparisons suggest that thewax biosynthesis machinery in members of both lineages may be inherited from acommon ancestor, but fine-tuned in each species and even between organs/structuresin a single species, presumably to optimize function. Further studies may illuminatewhether and how the different wax coverages and compositions on the three mossstructures affect their surface properties and ecophysiological functions. High waxcoverage on the calyptra may provide more durable protection for the developingsporophyte capsule, which is unable to survive even mildly desiccating conditionsduring early development without the protective calyptra cuticle [34].5.5 Experimental5.5.1 Moss growth conditionsSpores from four F. hygrometrica populations with developing sporophytes werecollected in Connecticut (CONN Budke #142, #144, #145; Goffinet #9027) andused to establish laboratory leafy gametophyte populations. These were grown forat least 4 months at room temperature (RT) with 16 hrs of light per day. Leafygametophytes were treated as previously described to produce sporophytes (Budke etal., 2011).1385. Cuticular waxes of Funaria hygrometrica5.5.2 Surface area measurement and wax extractionSurface areas of F. hygrometrica leafy gametophytes, gametophyte calyptrae, andsporophyte capsules were determined by averaging the surface areas of five individualsof each structure. Structures were dissected with a scalpel, flattened on a microscopeslide with water, and photographed. Surface areas of the moss structures were de-termined from these images with Adobe Photoshop CS3 (Adobe Systems) by pixelcounting and comparing with a photograph of a ruler taken at the same magnification.Surface areas were also determined for the calyptrae by approximating the rostrumwith a cylinder (surface area = height ∗ width ∗ 2pi and the inflated base with thefrustum of a cone (surface area = pi ∗ (r1 + r2) ∗√(h2 + (r1 − r2)2). Using the pixelcounting method, the average surface areas were 27 mm2 per leafy gametophyte, 4.3mm2 per calyptra, and 8.5 mm2 per sporophyte capsule.To determine the wax coverage per unit mass, leafy gametophytes were allowedto dry at RT for at least 24 hours and weighed multiple times until they reached asteady state. The average weight for 74 leafy gametophytes was 11.3 ± 2.9 mg (mean± SD, throughout).Leafy gametophytes with a total surface area of 20 cm2 were isolated and dippedin two changes of chloroform (CHCl3) for 30 s each to extract cuticular waxes. Tetra-cosane (10 µg) was added as an internal standard and the solvent was evaporated.Samples of calyptra and sporophyte capsule waxes were prepared using the same pro-cedure. Cut ends of the leafy gametophyte and sporophyte stalk were held out of thechloroform to avoid internal lipid extraction.5.5.3 Wax derivatization and GC conditionsWax samples were dissolved in pyridine (10 µl, Sigma Aldrich), bis-N,O-trimethylsilyltrifluoroacetamide (10 µl, BSTFA, Sigma Aldrich) was added, and thesamples were incubated at 70◦C for 45 min. Excess derivatization reagents were thenevaporated under a stream of nitrogen (N2) gas, and the dry residue was dissolved inCHCl3 (20 µl).Derivatized samples were analyzed quantitatively with a 6890N GC (Agilent)equipped with an on-column injector and a HP-1 capillary column (Agilent, length30 m, i.d. 320 µm, 1 µm film thickness). An aliquot of each sample was injected1395. Cuticular waxes of Funaria hygrometricainto the machine with H2 flowing through the column (2 ml/min) and the oven setto 50◦C. After 2 min the temperature was raised at 40◦C/min to 200◦C, held for 2min, raised at 3◦C/min to 320◦C, and then held for 30 min. Analytes were detectedwith a flame ionization detector (Agilent) at 250◦C that burned H2 (30 ml/min) inair (200 ml/min). The flame was shaped with N2 (20 ml/min).The samples were analyzed qualitatively with a separate 6890N GC (Agilent)equipped with the same injector, column, and oven program, but a column flow of He(1.4 ml/min) and a 5793N Mass Selective Detector (EI, 70 eV, m/z 50-800, 1 scan/s).5.5.4 Wax quantification and ester analysisPeaks in the gas chromatography-flame ionization detector gas chromatography -flame ionization detection (GC-FID) chromatograms were integrated and their sumarea was used to determine the total wax load by comparing with that of the in-ternal standard. The identity of each GC-FID peak was determined using the in-formation from corresponding peaks in the gas chromatography-mass spectrometryGC-MS chromatograms. Finally, the amount of each wax component was determinedby comparing its GC-FID peak area with the area of the internal standard.5.5.5 Statistical analysisData were analyzed using R 3.2.0 (R Core Team 2015). An ANOVA was used toassess whether there were differences in coverage between moss structures. Followinga significant ANOVA, Tukey post-hoc tests (P < 0.01) were conducted to determinesignificant differences between pairs of moss structures.5.6 Supplementary dataSupplementary data for this chapter can be found in Appendix DD.1: Wax coverage on three F. hygrometrica organs.D.2: Amount of esterified fatty acids in each alkyl ester homolog on the leafy1405. Cuticular waxes of Funaria hygrometricagametophyte of F. hygrometrica.D.3: Amount of esterified alcohols in each alkyl ester homolog on the leafygametophyte of F. hygrometrica.D.4: Amount of esterified fatty acids in each alkyl ester homolog on thegametophyte calyptra of F. hygrometrica.D.5: Amount of esterified alcohols in each alkyl ester homolog on the gameto-phyte calyptra of F. hygrometrica.D.6: Amount of esterified fatty acids in each alkyl ester homolog on thesporophyte capsule of F. hygrometrica.D.7: Amount of esterified alcohols in each alkyl ester homolog on the sporophytecapsule of F. hygrometrica.D.8: Amount of esterified β-hydroxy fatty acids in each β-hydroxy fatty acidester homolog on the leafy gametophyte of F. hygrometrica.D.9: Amount of esterified alcohols in each β-hydroxy fatty acid ester homologon the leafy gametophyte of F. hygrometrica.D.10: Amount of esterified β-hydroxy fatty acids in each β-hydroxy fatty acidester homolog on the gametophyte calyptra of F. hygrometrica.D.11: Amount of esterified alcohols in each β-hydroxy fatty acid ester homologon the gametophyte calyptra of F. hygrometrica.141Chapter 6Changes in cuticular wax coverageand composition on developingArabidopsis leaves are influencedby wax biosynthesis geneexpression levels and trichomedensity6.1 IntroductionPlant organ development relies on the tightly controlled formation and expansionof various tissues based on limited resources of reduced carbon and nutrients. Toprotect precious new organs from adverse conditions, physical and chemical defensesare established early during development [61, 19, 211], and they must continuouslyexpand to remain effective over the course of organ growth. Therefore, developingorgans must continuously invest in both construction and protection of new structures[45, 258, 16, 87].A balanced use of resources in leaf construction and protection is particularlyimportant for epidermal cells because they form the leaf-environment interface. For1426. Arabidopsis leaf wax developmentexample, the rapidly expanding epidermis of growing leaves must constantly protectthe entire organ against physical damage, insect attack, and excessive water lossby transpiration. The plant epidermis is made up of pavement cells, guard cells,and trichomes. Where present, all three epidermal cell types will thus contribute tothe functions of the epidermis. Pavement cells, the most abundant epidermal celltype on all organ surfaces, are the major protective surface barrier [171]. Guardcells, present on many organ surfaces in smaller numbers than pavement cells, areimportant for regulating gas exchange and for protecting the surface around stomata[106]. Finally, trichomes have a variety of roles including ultra-violet (UV) protection,heat insulation, transpiration control, and insect deterrence [242]. To gain insight intothe various roles of epidermal cells, their development on leaves of various speciesincluding Arabidopsis had been studied in much detail [68, 120, 73].The three epidermal cell types, together comprising the organ-environment in-terface, are coated by a continuous lipophilic layer [185, 252]. This extracellularmembrane, termed the cuticle, consists of the polyester cutin and a mixture of very-long-chain (VLC) hydrophobic wax compounds. The latter, typically composed offatty acids, primary alcohols, wax esters, aldehydes, and alkanes, constitutes thetranspiration barrier [187, 80, 92]. The amount of wax per surface area (coverage)and the composition of the wax mixture vary between tissues, organs, and species[6, 217, 184, 75, 235, 98, 177, 29, 109, 231], suggesting characteristic adaptations tooptimize specific functions.Cuticular wax ontogeny had been investigated in some species, and in most caseshad been found to be dynamic with respect to time. For example, changes in waxcoverage and composition during development had been reported for Kalanchoe dai-gremontiana [231], Prunus laurocerasus [98], Coffea arabica [203], Malus domestica[29], Sesamum indicum [109], Hordeum vulgare [177], Triticum aestivum [217], Hed-era helix [235], and Fagus sylvatica [166]. Thus, the observed dynamics of cuticularwax suggest that coverages and compositions may not only be optimized for diversefunctions between organs, but also during discrete developmental stages.To further our understanding of cuticular wax coverage, composition, and functionon developing organs, integrated approaches combining organ morphometrics, waxchemical profiling, and gene expression analyses of a model species are required. Ac-1436. Arabidopsis leaf wax developmentcordingly, cell expansion rates, wax composition, and expression levels of wax biosyn-thesis genes had been investigated in bolting stems of Arabidopsis thaliana [204],the species for which wax biosynthesis pathways are currently best characterized. Itis well established that wax biosynthesis relies on plastid-derived fatty acyl-CoAs,which are converted by the fatty acid elongase (FAE) multi-enzyme complex to VLCacyl-CoAs with chain lengths between C22 and C38. The chain length distribution ofacyl-CoAs generated by the FAE is primarily determined by its ketoacyl-CoA syn-thase (KCS) component enzymes [141] and associated proteins CER2 and CER26[82]. The acyl-CoAs are then passed to either the acyl reduction pathway enzymesCER4 and WSD1 that generate primary alcohols and wax esters [182, 124], or thedecarbonylation pathway enzymes CER3 and CER1 that produce aldehydes and alka-nes, respectively [24]. Though fatty acids are major wax compounds on Arabidopsisleaves, their biosynthesis is not yet fully understood.Interestingly, expanding Arabidopsis stem sections had been found to expressmany wax biosynthesis genes more highly than sections that had completed expan-sion, pointing to transcriptional regulation of wax biosynthesis [204]. Nonetheless,neither wax coverage nor composition differed between the top, middle, and bottomsections of the stem, which had been sampled as proxies for organs of different age.However, Arabidopsis stems grow very rapidly, mainly by expanding in a very shortzone near the top of the stem, rendering potential morphological, chemical, or ge-netic gradients within this zone only visible via analyses with very high spatial andtemporal resolution.In contrast to bolting inflorescence stems, Arabidopsis rosette leaves develop muchmore slowly and may therefore be good candidates for identifying developmentalchanges in wax coverage and/or composition that can be linked to cell expansion andgene expression. Arabidopsis leaf waxes consist mainly of alkanes and primary alco-hols, together with fatty acids and aldehydes [95]. An early study on the dynamics ofArabidopsis leaf waxes found that whole rosettes from young plants have higher waxcoverage, higher percentages of fatty acids, and longer alcohols than whole rosettesfrom older plants [96]. Here, I sought to first corroborate these results with highertemporal resolution using leaves from only one nodal position and then integrate theseresults with epidermal cell development and gene expression data. Preliminary tests1446. Arabidopsis leaf wax developmenthad shown that, under the conditions used here, leaves on nodes one to seven grewto variable, relatively small sizes, while leaves on nodes eight and higher all grew toapproximately the same full size. Therefore, eighth rosette leaves of wild-type andtrichome-less (gl1 ) Arabidopsis plants were harvested every four days during devel-opment, and investigated using light and confocal microscopy, gas chromatographygas chromatography (GC) with flame ionization detection (FID) or coupled to massspectrometry (MS), and quantitative RT-PCR.6.2 ResultsThe goal of this study was to link changes in cuticular wax coverage and compo-sition on growing Arabidopsis leaves with epidermal cell development and potentialchanges in the expression of underlying wax biosynthesis genes. Based on preliminarystudies, the eighth rosette leaf was selected for investigation. First, leaf blade andepidermal cell surface areas were determined, together with the numbers of epider-mal cells present (6.2.1). Next, wax coverage and composition on leaves without andwith trichome cells were measured every four days during growth (6.2.2). To furtherunderstand local differences along developing leaves, pavement cell size and wax com-position on discrete leaf sections were measured (6.2.3). Finally, the expression levelsof wax biosynthesis genes were measured over the course of leaf development (6.2.4).6.2.1 Morphological changes on developing leavesLeaf surface areas and the numbers of epidermal cells were measured as a functionof leaf age. Under the growth conditions used here, leaf blades expanded steadily from10 mm2 at day five to 138 mm2 at day 21 (Fig. 6.1A), after which they did not change(Figure E.1). The length of the petiole increased from 1 mm at day five to 14 mm atday 21.The average surface area of pavement cells increased from ca. 490 µm2 to 3910 µm2between days five and 21 (Fig. 6.1A), and thus pavement cell expansion accounted forthe majority of macroscopic leaf growth. However, leaf expansion was also partiallydue to pavement cell division, which led the number of pavement cells to increase fromca. 20,500 per blade at day five to ca. 29,100 by day 13 and remain roughly constant1456. Arabidopsis leaf wax developmentFigure 6.1: Surface area and epidermal cell numbers on developing wild-typeArabidopsis eighth rosette leaves. A) Leaf surface area (black circles, left y-axis)and pavement cell surface area (white circles, right y-axis) measured between five and21 days of leaf age. B) Numbers of pavement cells (white squares, left y-axis), guardcells (light grey squares, left y-axis), and trichome cells (dark grey squares, righty-axis) measured between five and 21 days of leaf age. Point positions and errorbars indicate the mean and standard deviation of five independent measurements,respectively.1466. Arabidopsis leaf wax developmentthereafter (Fig. 6.1B). Concomitantly, the number of guard cell pairs increased fromday five (ca. 6,700 per blade) to day 13 (10,100) and then also stayed constant(Fig. 6.1B). The number of trichomes, ca. 75, was constant throughout eighth leafdevelopment, resulting in relatively high trichome densities on younger leaves and asteady decrease in trichome density as the leaf expanded (Fig. 6.1B).6.2.2 Cuticular waxes from whole Arabidopsis leaves of dif-ferent ageTo monitor leaf wax development, leaf surfaces were extracted every four daysduring growth, and the resulting leaf wax mixtures were analyzed with gas chro-matography - mass spectrometry (GC-MS) and GC-FID. To delineate effects arisingfrom shifts in the relative abundance of trichome cells over time, the waxes fromdeveloping leaves of both the trichome-less mutant gl1 (6.2.2) and wild-type plants(6.2.2) were investigated.Wax development on growing eighth rosette leaves of the trichome-lessArabidopsis mutant gl1Five-day-old gl1 leaves were covered with 0.62 ± 0.05 µg/cm2 extractable wax(Fig. 6.2, Table E.1) made up of fatty acids (34% of the overall wax mixture), alka-nes (26%), branched alcohols (13%), n-alcohols (7%), aldehydes (2%), alkenes (2%),and 16% of unidentified compounds (Fig. 6.3A). At 21 days of age, gl1 leaves bore0.64 ± 0.04 µg/cm2 wax, which was composed of alkanes (42%), fatty acids (19%),branched alcohols (11%), n-alcohols (7%), aldehydes (5%), and alkenes (2%), leaving14% unidentified. Thus, wax coverage remained constant over the course of gl1 leafdevelopment, and after day five the composition of the mixture gradually shifted,becoming significantly less fatty acid- and more alkane-dominated.Each of the chemical classes was present as a homologous series spanning chainlength ranges typical of cuticular compounds. Fatty acids were present with evencarbon numbers between C22 and C32, within which the C24 and C26 homologs werethe most prominent (Fig. 6.3B). The n-alcohols also had predominantly even carbonnumbers, but ranged from C26 to C34 with an approximately normal distribution.1476. Arabidopsis leaf wax development0.00.20.40.60.81.01.25 7 9 11 13 15 17 19 21Age of leaf eight (days)Wax coverage (µg/cm2 )gl1WTFigure 6.2: Wax coverage on developing Arabidopsis gl1 and wild-typeeighth rosette leaves. The total amount of wax from gl1 and wild type is ex-pressed as wax mass (µg) per surface area extracted (cm2). Eighth rosette leaveswere harvested every four days between five to 21 days of leaf age, and their totalwax coverages were measured with GC-FID. Lines connect the mean of the five inde-pendent measurements. The slope of a first order linear model fitted to the wild-typedata was significantly different from zero (p <0.0001).Three branched-chain alcohols with total carbon numbers C30, C32, and C34 wereidentified, with the C32 homolog being the most abundant. Four unbranched alde-hydes with even carbon numbers from C28 to C34 were detected in roughly equalamounts. n-Alkanes were found with odd carbon numbers ranging from C27 to C37,the C29, C31, and C33 homologs being the most abundant. Finally, C35 and C37n-alkenes were present in roughly equal amounts.Chain length shifts were observed during gl1 leaf development, and were mostpronounced within the fatty acid (FA) and alkane series. From day five to day 21,the relative abundance of C24 fatty acid decreased steadily from 13% to 3% of thetotal wax mixture (Fig. 6.3B). Simultaneously, relative amounts of C29 and C31 alkaneincreased steadily from ca. 7% to 18% and 9% to 18%, respectively, while the relativeabundance of C33 alkane decreased from 11% to 3%. Thus, a significant, combinedshift from acid- to alkane-dominance and from C24 to C29 and C31 compounds wasobserved over the course of development on gl1 leaves.1486. Arabidopsis leaf wax developmentLastly, the overall chain length profiles across all compound classes within thewax mixture on developing gl1 leaves were calculated. For this, the relative amountsof all compounds formed by modification of the same acyl-CoA precursor were addedtogether, e.g., C30 acid, C30 alcohols, C30 aldehyde, and C29 alkane. At five days ofage, gl1 leaf wax was made up of ca. 22% each C26 and C32 compounds, and ca.16% each C24, C30, and C34 compounds (Fig. 6.3C). At 21 days of age, the mixturecontained ca. 45% C32 compounds, 30% C30 compounds, and 26% C26 compounds,together with smaller amounts of C24 and C34 compounds. Relatively minor quantitiesof compounds with chain lengths C22, C28, C36, and C38 were present throughoutdevelopment. Overall, the most pronounced changes over time were observed asincreases in the relative abundance of C32 and C30 compounds and decreases in therelative abundance of C24, C26, and C34 compounds.Wax development on growing eighth rosette leaves of wild-type Arabidop-sisThe cuticular wax of wild-type eighth rosette leaves was harvested and analyzedin the same way and at the same points during development as described for gl1.Coverage on wild-type leaves decreased significantly between five days of age (1.06 ±0.02 µg/cm2) and 21 days of age (0.86 ± 0.06 µg/cm2) (Fig. 6.2, Table E.3).The wax on five-day-old wild-type eighth leaves contained alkanes (39%), fattyacids (21%), n-alcohols (9%), branched alcohols (6%), alkenes (7%), and aldehydes(3%), leaving 15% of the wax unidentified. Alkanes were the most prominent com-pound class throughout development of wild-type leaves, increasing significantly from39% to 49% of the total wax mixture between days five and 21 (Fig. 6.4A). Over thesame time interval, the relative abundance of fatty acids and alkenes decreased sig-nificantly from 20% to 12% and from 7% to 2%, respectively, while other compoundclasses exhibited relatively little fluctuation.Wild-type leaf wax contained homologous series of fatty acids (C22 - C32), n-alcohols (C26 - C34), branched alcohols (C30 - C34), and aldehydes (C28 - C34), all withpredominantly even carbon numbers, as well as alkanes (C27 - C37) and alkenes (C35- C37) with predominantly odd carbon numbers. Overall, the same compounds wereidentified in the leaf waxes of wild type and the gl1 mutant. Homolog distributions1496.ArabidopsisleafwaxdevelopmentA)Fatty acids n-Alcohols br.-Alcohols Aldehydes Alkanes Alkenes Unidentifieds* * *010203040505 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21Age (days)Rel. abundance  (% total wax)B)***** * *****0246810121416182022242622 24 26 28 30 32 26 28 30 32 34 30 32 34 28 30 32 34 27 29 31 33 35 37 35 37Rel. abundance  (% total wax) day 5day 9day 13day 17day 21Fatty Acids n-Alcohols br.-Alcohols Aldehydes Alkanes AlkenesC)Total C22 Total C24 Total C26 Total C28 Total C30 Total C32 Total C34 Total C36 Total C38* * * * * * *010203040505 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21Age (days)Rel. abundance  (% total wax)Figure 6.3: (Continued on the following page.)1506. Arabidopsis leaf wax developmentFigure 6.3: Wax composition on developing Arabidopsis gl1 mutant eighthrosette leaves. A) Relative abundance of each compound class as a function of leafage. B) Relative abundance of each wax compound as a function of leaf age. Labelson the x-axis indicate the carbon number and class of each identified compound. C)Relative abundance of each chain length as a function of leaf age. Odd-numberedhomologs of alkanes and alkenes are derived from even-numbered precursors with onecarbon more, and these compounds are therefore grouped with compounds derivedfrom respective precursors with even total carbon numbers. Bar heights and errorbars indicate the mean and standard deviation of five independent measurements,respectively. Lines connect the means of the five independent measurements made ateach time point. Asterisks indicate significant time-dependent changes derived froma permutation test of the ordinary least squares linear fit after study-wide adjustmentfor multiple comparisons.within these chain length ranges were also similar on both lines. Over time, therelative abundance of C24 FA decreased significantly from ca. 8% to 2% in wild-typewax, while C29 and C31 alkanes increased significantly from 5% to 17% and from 15%to 20%, respectively. In contrast, the relative abundance of the longer alkanes (C33,C35, and C37) as well as C35 and C37 alkenes decreased significantly over time.In terms of constituent chain lengths, wild-type eighth leaf wax was made up of29% C32 compounds, ca. 20% C34 compounds, 13% each C26 and C30 compounds, ca.9% each C24 and C36 compounds, and ca. 4% C38 compounds at day five (Fig. 6.4C).At 21 days of age, the mixture consisted of 38% C32, 28% C30, 13% C26, 11% C34compounds, ca. 2% each C24 and C36 compounds, and 1% C38 compounds. Thus,the relative abundance of C30 and C32 compounds increased significantly over time,while that of C24, C34, C36, and C38 compounds decreased significantly. Throughoutdevelopment C22 compounds contributed relatively minor amounts to the total waxmixture.6.2.3 Regional distribution of wax on Arabidopsis leavesTo test whether differences in wax composition also existed between discrete leafregions potentially differing in average cell age, leaves were sectioned (Fig. E.2), andthe size distribution of pavement cells on each section was investigated. Five-day-oldleaves exhibited a pavement cell size gradient along their longitudinal axes from 300± 50 µm2 at the base of the blade to 900 ± 200 µm2 at the tip (Fig. 6.5). A larger1516.ArabidopsisleafwaxdevelopmentAFatty acids n-Alcohols br.-Alcohols Aldehydes Alkanes Alkenes Unidentifieds* * *010203040505 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21Age (days)Rel. abundance  (% total wax)B******** ******* *0246810121416182022242622 24 26 28 30 32 26 28 30 32 34 30 32 34 28 30 32 34 27 29 31 33 35 37 35 37Rel. abundance  (% total wax) day 5day 9day 13day 17day 21Fatty Acids n-Alcohols br.-Alcohols Aldehydes Alkanes AlkenesCTotal C22 Total C24 Total C26 Total C28 Total C30 Total C32 Total C34 Total C36 Total C38* * * * * * *010203040505 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21 5 9 13 17 21Age (days)Rel. abundance  (% total wax)Figure 6.4: (Continued on the following page.)1526. Arabidopsis leaf wax developmentFigure 6.4: Wax composition on developing Arabidopsis wild-type eighthrosette leaves. A) Relative abundance of each compound class as a function of leafage. B) Relative abundance of each homolog within compound classes as a function ofleaf age. Labels on the x-axis indicate the carbon number and class of each identifiedcompound. C) Relative abundance of each chain length as a function of leaf age. Odd-numbered homologs of alkanes and alkenes are derived from even-numbered precursorswith one carbon more, and these compounds are therefore grouped with compoundsderived from respective precursors with even total carbon numbers. Lines connectthe means of the five independent measurements made at each time point. Asterisksindicate significant time-dependent changes derived from a permutation test of theordinary least squares linear fit after study-wide adjustment for multiple comparisons.size difference was observed on leaves at day nine, where pavement cells varied in sizefrom 1,000 µm2 at the base to 3,000 µm2 at the tip. Similar gradients were observedon leaves at days 13 and 17. In contrast, at day 21, pavement cells along the entireblade were similar in size, ca. 5,000 ± 1,000 µm2. Overall, the pavement cell sizedistributions indicated that pavement cells at the leaf tip initiated and completedexpansion before those at the leaf base. At 13 days of age leaves exhibited the largestdifference in pavement cell size and age between leaf base and tip, with cells at thebase most similar to those found on young leaves and cells at the tip most similar tothose found on mature leaves.To correlate pavement cell size and age with wax composition, 13-day-old leaveswere cut into three equal-sized pieces, and the wax compositions of the base andtip segments were determined as described above for whole-leaf wax analysis. Theidentity and span of the classes of homologous compounds identified on both the leafbases and tips were identical, containing fatty acids (C22 - C34), n-alcohols (C24 - C34),branched alcohols (C30 - C34), aldehydes (C26 - C34), alkanes (C27 - C37), and alkenes(C35 - C37), but the relative abundance of some of these compounds differed betweenthe leaf tips and bases. In particular, C24 and C26 acids made up 5% and 14% of thetotal wax on leaf bases, respectively, while on the tips they were represented only 2%and 8% (Fig. 6.6). Furthermore, leaf bases were covered with 9% and 13% C29 andC31 alkanes, respectively, while these compounds comprised 17% and 19% of the waxon leaf tips. The relative abundances of all other compounds were present in similaron leaf tips and bases.1536. Arabidopsis leaf wax developmentFigure 6.5: Pavement cell size on segments of developing Arabidopsis wild-type eighth rosette leaves. Pavement cell surface areas (µm2) are plotted as afunction of their position along the longitudinal axis of the leaf (mm, measured fromthe leaf blade base) for leaves between five and 21 days of age. Point positions and er-ror bars indicate the mean and standard deviation of five independent measurements,respectively.6.2.4 Expression of wax biosynthesis genes during leaf devel-opmentTo assess possible contributions of wax biosynthesis gene expression to fluctuationsin wax production, transcript levels were monitored at the same time points usedfor wax sampling. To account for variation in the amount of tissue harvested, theexpression levels of target genes had to be normalized. Preliminary experimentsshowed that, of the four reference genes Actin2, GAPDH, UBQ10, and UBC21, thelatter exhibited the most stable expression over all time points (data not shown) andwas therefore used to normalize expression levels of target genes. Based on their1546. Arabidopsis leaf wax development***** * ****02468101214161820222422 24 26 28 30 32 34 24 26 28 30 32 34 30 32 34 26 28 30 32 34 27 29 31 33 35 37 35 37Rel. abundance (% total wax)basetipFatty Acids n-Alcohols br.-Alcohols Aldehydes Alkanes AlkenesFigure 6.6: Wax composition on the base and tip sections of wild-typeArabidopsis eighth rosette leaves at 13 days of age. The relative abundanceof each homolog in each compound class is plotted as a percent of the total amountof extracted wax. Labels on the x-axis indicate the carbon number and class of eachidentified compound. Leaf base and tip samples were prepared by selecting eighthleaves at 13 days of age, cutting them into three equal segments, and independentlyextracting the base and tip segments. Bar heights and error bars indicate the meanand standard deviation of five independent measurements, respectively. Asterisksindicate significant mean differences between base and tip derived from a permutationtest after study-wide adjustment for multiple comparisons.function, the investigated genes were grouped as coding for 1) condensing enzymes ofelongation complexes (KCSs), 2) proteins associated with the elongation complex(es),and 3) head group-modifying enzymes.Among the KCS genes, CER6 and KCS8 were expressed at much higher levelsthan KCS1, KCS5, and KCS16 (Fig. 6.7A). Over the course of leaf development,KCS1 and CER6 expression increased significantly, expression of KCS8 decreasedsignificantly, and that of KCS5 and KCS16 exhibited a slight decrease between daysfive and 13. All the genes coding for proteins associated with elongation complex(es),CER10, CER8, CER2 and CER26, were expressed at relatively high and constantlevels throughout leaf development (Fig. 6.7B), excepting the slightly higher expres-sion of CER10 and CER8 at day five. Among the head group-modifying enzymes,CER3 and CER1 were expressed at intermediate levels and CER4 at very low levels(Fig. 6.7C). Expression of CER3 and CER1 varied relatively little over the course ofleaf development, with higher expression of CER1 at day five, and CER4 expressionincreased slightly throughout development.1556.Arabidopsisleafwaxdevelopment          CER1Age of leaf (days)5 9 13 17 21024681012*5 9 13 17 210.00.10.20.30.4Age of leaf (days)CER4Figure 7: Expression of wax biosynthesis genes in developing wild-type Arabidopsis eighth leaves. Expression levels were determined using qRT-PCR and normalized against the reference gene UBC21. A) Expression of selected KCS genes, encoding the FAE complex enzymes responsible for chain length control during wax precursor elongation. B) Expression of genes encoding proteins associated with the KCSs during elongation. C) Expression of genes involved in the conversion of wax precursors into wax compounds. Just as for the chemical data, changes in transcript levels as a function of leaf development between five and 21 days of age were analyzed with permutation tests on robust ordinary least squares method. Trends found to be significantly different from zero by this test after study-wide adjustment for multiple comparisons were flagged with an asterisk (*). Due to apparent inflation of gene expression measurements on day five, a sensitivity analysis was performed, and trends that were significant upon lightening or omission of day five data were flagged with a diamond (u). Details of statistical tests are presented in Appendix S1.  B C A CER25 9 13 17 21024681012Age of leaf (days)CER3Age of leaf (days)Gene expression (relative to UBC21)5 9 13 17 21024681012CER265 9 13 17 21024681012Age of leaf (days)5 9 13 17 210.00.51.01.52.0Age of leaf (days)Gene expression (relative to UBC21) KCS1*Age of leaf (days)5 9 13 17 210.00.51.01.52.0 KCS5*5 9 13 17 21024681012Age of leaf (days)CER6*5 9 13 17 21024681012Age of leaf (days)KCS8*5 9 13 17 210.00.51.01.52.0Age of leaf (days)KCS16*CER105 9 13 17 21024681012Age of leaf (days)Gene expression (relative to UBC21)*CER8Age of leaf (days)5 9 13 17 21024681012*Figure 6.7: (Continued on the following page.)1566. Arabidopsis leaf wax developmentFigure 6.7: Expression of wax biosynthesis genes in developing wild-typeArabidopsis eighth leaves. Expression levels were determined using qRT-PCRand normalized against the reference gene UBC21. A) Expression of selected KCSgenes, encoding the FAE complex enzymes responsible for chain length control duringwax precursor elongation. B) Expression of genes encoding proteins associated withthe KCSs during elongation. C) Expression of genes involved in the conversion ofwax precursors into wax compounds. Changes in transcript levels as a function of leafdevelopment between five and 21 days of age were analyzed using permutation testson the least squares slopes of the expression profiles. Asterisks indicate significanttrends based on these tests, after study-wide adjustment for false discoveries. Due toapparent inflation of gene expression measurements on day 5, a sensitivity analysiswas performed, the full details of which are available in Table E.4 and Table E.5.Trends that are significant with lightened or omitted day 5 data are flagged withspades.6.3 DiscussionOverall, our morphometric results first established that trichome density decreasedsteadily during leaf development, that leaf tip expansion preceded leaf base expan-sion, and that gradients in pavement cell maturity and size between leaf tips and baseswere maximized around 13 days of leaf age. Wax coverage on wild type decreased sig-nificantly with expansion, while gl1 had constant, lower coverage. An age-dependentshift from C24/C26 to C30/C32 compounds was accompanied by a relative decreasein fatty acid (FA) abundance and an increase in alkane abundance on both plantlines. Wild-type leaves also exhibited a simultaneous decrease in the relative abun-dance of C35+ compounds. qRT-PCR analyses revealed that head group-modifyingenzymes were expressed at fairly constant, albeit different levels. An increase in theexpression of KCS1 and CER6 was accompanied by a decrease in that of KCS5 andKCS16, whereas non-KCS elongation genes were expressed at roughly constant lev-els throughout leaf development. These findings can now be integrated to discusswax dynamics in the context of pavement cell age (6.3.1), leaf expansion (6.3.2), andepidermal cell composition (6.3.3).1576. Arabidopsis leaf wax development6.3.1 Pavement cell age effects on wax compositionArabidopsis leaf morphology was monitored to give context to the wax compositionand gene expression data. The data acquired here spanned leaf development froman early stage, defined by the onset of pavement cell expansion, to a late stage whenpavement cell expansion was largely complete. This occurred after 21 days of growth,similar to previous observations of Arabidopsis leaf six [70]. Like other studies ongrowing leaves from Arabidopsis and other species [205, 51, 148, 40, 54], I found thatpavement cells near the leaf tip begin to expand at a time when many cells near thebase were still dividing, and that tip cells finished expansion long before base cells.Thus, the position of a pavement cell along the leaf axis is related to cell age, with theyoungest cells being present at the leaf base. Accordingly, a comparison of leaf basesand tips (a spatial distribution) may serve as an orthogonal method of comparingyoung and old leaves (a temporal distribution). Under our growth conditions, thelargest difference in cell maturity between tip and base pavement cells was reachedat day 13, when tip cells had reached their final size and base cells had only begunto expand, nearly identical to what had been reported for leaf three [5].By both spatial and temporal comparisons, leaves of gl1 and wild type exhibiteddecreases in the relative abundance of C24/C26 compounds and increases in that ofC30/C32 compounds or their C29 and C31 alkane derivatives. A previous analysis ofArabidopsis leaves also found the relative abundance of C31 alkane to increase withage, albeit accompanied by a decrease in that of C29 alkane [96]. Furthermore, anincrease in wax compound chain lengths, mostly of alkanes, had been observed onthe developing leaves of the monocot Sorghum bicolor [6] and diverse dicots includingMalus domestica [29], Sesamum indicum [109], Coffea arabica [203], and Rhododen-dron fortunei [184]. In contrast, leaves of Kalanchoe daigremontiana [231] and severalTriticum species exhibited alkane shortening as they developed [217, 231].Chain length profiles of wax compounds are established by elongation in fatty acidelongase (FAE) complexes, within which the ketoacyl-CoA synthase (KCS) enzymesare known to exert control over product chain length distributions [59, 141]. Conse-quently, it seems plausible that differential expression of one or more KCS enzymesmay lead to the observed shift from C24/C26 to C30/C32 wax compounds. I foundthat the expression level of each KCS gene was different, as had been reported pre-1586. Arabidopsis leaf wax developmentviously [104, 108], and that in many cases these expression levels also varied over thecourse of leaf development. Expression of KCS5, KCS8, and KCS16 decreased withage, indicating that these are probably not involved in the increase of C30/C32 waxcompounds. In contrast, expression of KCS1 increased with leaf age, thus parallelingthe main chain length shift. However, heterologous expression of KCS1 yielded C20- C26 products [213] and, conversely, the kcs1 mutant had been found to be affectedmainly in the accumulation of C26 and C28 wax compounds [212], together suggest-ing that this enzyme is likely not involved in the observed chain length shift. Incontrast, CER6 was highly and increasingly expressed throughout leaf development.This KCS had produced C30 and C32 compounds when heterologously expressed withCER2 and CER26 [84, 160, 82], and the cer6 mutant had been found to accumulateC26 compounds [95, 60, 89]. Together with these literature data, our observationsnow strongly suggest that the turnover of C26 acyl-CoA precursors and the resultingaccumulation of C30/C32 products that characterizes Arabidopsis leaf development islargely controlled by the expression level of CER6.Since CER2 and the CER2-like proteins are also involved in chain elongation byallowing CER6 to produce compounds with 30 or more carbons [84, 160, 82], it seemedplausible that they may also play a role in the developmental chain length shift ob-served here. However, the expression levels of the CER2-like genes did not changeduring leaf development, making their involvement in the shift unlikely. Interestingly,their constant expression throughout also implies that there is an excess of CER2 andCER26 transcripts at the outset of pavement cell expansion, and that the abundanceof CER6 transcripts, rather than that of CER2(-likes), initially limits the productionof longer chain lengths. The expression of CER10, also part of the FAE complex, de-creased significantly, mostly during early leaf development. However, since CER10 isequally involved in all elongation rounds, it seems unlikely that the chemical changesare driven by differential expression of CER10. Similarly, CER8 also showed only aslight decrease in expression levels mainly during early development, making it un-likely that the encoded long-chain acyl-CoA synthase (LACS) enzyme likely formingacyl-CoAs from fatty acids [126] has a role in driving the chemical changes.In addition to the chain length shift, our temporal and spatial wax comparisonsalso revealed a compound class shift from fatty acid to alkanes. This trend is similar1596. Arabidopsis leaf wax developmentto a previous report on Arabidopsis leaf wax development [96], except for the higheroverall amounts of fatty acid found here. Many other species exhibit compound classshifts similar to what has been observed for Arabidopsis, including Coffea arabica[203], Prunus laurocerasus [98], and Malus domestica [29], though in some, such asSorghum bicolor, both fatty acids and alkanes continue to accumulate as leaves age[6].The head group modification pathways generate the different compound classes,and the expression levels of genes encoding respective modifying enzymes are there-fore prime candidates for instigators of the observed compound class shift. However,expression of CER3 and CER1, the genes involved in alkane production, did not in-crease during leaf development. The expression level of CER4, encoding the reductaseresponsible for production of primary alcohols, showed a slight increase over time, de-spite fairly constant wax alcohol amounts. Taken together, our data suggest that abalance of head group-modifying enzymes for the decarbonylation and acyl reductionpathways is maintained by relatively constant expression of all genes encoding them.Therefore, the drastic change in chemical composition from acids to alkanes cannotbe explained by differential expression of the genes coding for head group-modifyingenzymes.While it is possible that the shift from fatty acids to alkanes is the consequence ofthe down-regulation of a gene responsible for acid formation, no such enzyme activityhas been identified. Alternatively, the increase in alkane abundance could arise fromthe increase in CER6 expression if the precursors generated by CER6 are channeledpreferentially into the decarbonylation pathway instead of being equally available tothe acyl reduction, decarbonylation, and acid-forming pathways. Such associationbetween KCS enzymes and head group modification pathways could also explain theselective effect of kcs1 on the alcohol and acid compound classes [212].Finally, it should be noted that, in addition to the increase in the relative abun-dance of C30/C32, there was also a significant decrease in C34 products (mainly inthe form of C33 alkane) during development of both gl1 and wild-type leaves (com-pare Fig. 6.3B). This trend might be due to gradually decreasing expression of anunknown KCS with C34 product specificity, or to a steady increase in competitionbetween that KCS and the more abundant CER6. The latter explanation would1606. Arabidopsis leaf wax developmentimply that the products of the CER6-containing FAE, C30/C32 acyl-CoAs, preferen-tially serve as substrates for alkane formation (by CER3 and CER1) rather than forfurther elongation, thus providing further support for substrate channeling from theCER6-containing FAE into the decarbonylation pathway.6.3.2 Leaf expansion effects on wax coverageThe total wax coverage on mature wild-type leaves was ca. 0.9 µg/cm2, wellwithin the range of literature values (0.5 µg/cm2 to 1.8 µg/cm2) for this ecotype[28, 24, 84, 160]. I found that the glabrous leaves of gl1 had a roughly constantcoverage of ca. 0.6 µg/cm2 at all stages of development, approximately 65-70% ofthe macroscopic coverage of the corresponding wild type, which is consistent withpreviously reported ratios [175]. It should be noted that the wild type wax coveragespresented here, like those in previous reports, were calculated based on projected,macroscopic leaf surface area, thus ignoring the contribution of trichomes to the true,microscopic surface area.High trichome densities on young leaves and steady decreases in trichome densityduring leaf expansion had been reported for many species such as Bemisia tabasiand Arabidopsis [44, 135]. Furthermore, a decrease in macroscopic coverage withage/expansion was observed on leaves of trichome-bearing species such as Arabidop-sis [96], Malus domestica [29], Sesamum indicum [109], Hedera helix [235], and Fagussylvatica [75], but was not observed on leaves lacking trichomes, such as those ofKalanchoe daigremontiana [231]. Overall, these observations indicate that the de-crease in macroscopic coverage on developing leaves is probably due to a decrease intrichome density.Due to their complex geometry, the surface area of Arabidopsis leaf trichomesand their contribution to the total leaf surface area are difficult to assess accurately.Eighth leaves counted an average of 75 trichomes on eighth leaves regardless of theirage, a number roughly similar to previous reports and well within the range of countsthat had been reported for different rosette leaves [57]. Approximating a trichome asa cylinder (of 50 µm diameter and 150 µm height) topped with three cones (each 20µm across and 150 µm high; [132]), I estimate the sum surface area of all trichomes onthe blade to be 0.03 cm2. Consequently, trichomes may add ca. 15% and 1% to the1616. Arabidopsis leaf wax developmentprojected surface area of young and mature eighth leaves, respectively. By combiningthe trichome and pavement cell surfaces, the overall surface area of young and matureleaves must be corrected to 0.23 cm2 and 2.83 cm2, respectively. The wax coveragesof wild type leaves may accordingly be re-calculated, resulting in coverages of ca. 0.9µg/cm2 irrespective of age. Thus, wax coverages are roughly constant during growthof both wild type and gl1 leaves, as has been observed for growing Arabidopsis stems[204], together suggesting that wax production is synchronized with cell expansion.To maintain constant wax coverage on expanding leaf surfaces, the epidermis mustproduce wax at a particular rate. According to published expansion rates and cover-ages for Arabidopsis stem segments [204], the rapidly expanding stem tops producedwax at ca. 1.1 µg/day, while the middle of the stem produced 0.15 µg/day. Based onthe surface expansion of wild-type eighth leaves (0.16 cm2/day), I estimate a wax pro-duction of ca. 0.15 µg/day between days five and 21 (Table E.3). Similar calculationsmay be performed for specific epidermal cell types by first determining cell-specificwax coverage. Based on their approximated surface area (0.03 cm2), the differencein the absolute wax amounts on young and mature wild type leaves (0.05 µg), andthe time it takes for a trichome to expand fully (1.5 days), trichome cells shouldproduce wax at a rate of roughly 18 pg/hr/cell (Table E.3). For pavement and guardcells together, this rate is approximately 0.1 pg/hr/cell. Overall, these estimationsshow that rates of wax production vary considerably between cell types and organsto maintain constant coverage, suggesting that wax production is tightly linked to(cell) surface area expansion rates. The mechanisms controlling this synchrony at celland organ-specific levels are far from clear, and may comprise genetic, biochemical,and/or physiological feedback regulation.6.3.3 Epidermal cell type effects on wax compositionFluctuations in trichome density may not only affect total wax coverage but alsowax composition. I found a significant, time-dependent decrease in the relative abun-dance of C35 and C37 alkanes and alkenes on wild type leaves that was not observedon gl1 leaves (cf. Fig. 6.4 and Fig. 6.3), suggesting that these compounds may differin their contributions to the total wax loads covering pavement cells and trichomes.This notion can be further investigated by approximating a trichome-specific wax1626. Arabidopsis leaf wax developmentcomposition by subtracting the wax composition of (adaxial and abaxial) pavementand guard cells (i.e., the gl1 wax mixture) from the wax composition of (adaxial andabaxial) pavement, guard, and trichome cells (i.e., the wild type wax mixture). Ashad been reported by others, I did not observe any abaxial trichomes on wild typeleaves, the number of which vary with ecotype, leaf number, and light conditions[210]. Thus, this subtractive comparison reflects the adaxial trichome wax composi-tion, which I estimate to consist of n-alcohols (ca. 15% of calculated total trichomewax), alkanes (62%), and alkenes (ca. 15%), with each compound class dominatedby respective C32+ homologs (Fig. E.4, Fig. E.5).To evaluate chain length differences between wax mixtures on trichomes and(mainly) neighboring pavement cells, the calculated trichome wax composition canbe compared with literature reports on the wax coating the adaxial side of gl1 leaves.For example, while the most abundant alkanes on gl1 adaxial surfaces were C29 andC31 (ca. 24% and 50% of all alkanes, respectively) [37], the most abundant alkanes inthe calculated trichome wax composition were C31 and C33 (ca. 50% and 45% of allalkanes, respectively, Fig. E.4). However, the largest compositional differences wereobserved for C35+ alkanes, which comprised ca. 2% of the alkanes on gl1 adaxialsurfaces, but almost 20% of the calculated trichome wax composition (Fig. E.5). Theabundance of C35 and C37 alkenes, although not reported for gl1 adaxial surfaces, canbe compared between our gl1 wax results and the calculated trichome composition,where they represented 2% and ca. 14%, respectively. Together, these comparisonsstrongly suggest that the wax mixture produced by trichome cells differs from thatof neighboring pavement cells in that it is enriched in C33+ alkanes and alkenes.To identify gene candidates for the production of C33 - C37 compounds, KCS s mustagain be considered. However, although considerable progress has been made in thecharacterization of KCS s, none has been found that produces compounds longer thanC34, even with the aid of CER2-like proteins. The data presented here point to KCSsspecifically involved in trichome wax elongation. The decrease in trichome density wasparalleled by apparent decreases in the expression levels of KCS5, KCS8, and KCS16,genes that had also been found expressed preferentially in trichomes [133]. It shouldbe noted that, since our gene expression measurements convey the average expressionlevel in all leaf cells, it is possible that the relative expression level of these genes in1636. Arabidopsis leaf wax developmenttrichome cells is underrepresented in the data presented here. Consequently, KCS5,KCS8, and KCS16 may be considered as primary candidates for the production ofC35+ compounds in trichomes. KCS5 and CER6 share 88% amino acid identity [104],and the two have similar product profiles [213], suggesting that KCS5, like CER6,only produces up to C30/C32 precursors. KCS8 and KCS16 show 74% amino acididentity, are members of the same KCS subclass [104], and have highest transcriptabundance in leaves [108], though KCS16 is also highly expressed in siliques [104].Based on the currently available data, both these KCSs remain likely candidates forthe production of C35+ compounds, particularly in Arabidopsis leaf trichomes.It should be noted that the calculated trichome wax composition contained rela-tively low amounts of branched alcohols. This compound class constituted ca. 14%of the adaxial wax of gl1 leaves [37], but less than 1% of the calculated trichome waxmixture (Fig. E.5). Interestingly, while the chain lengths of n-alcohols on trichomesand pavement cells differ, as did the chain lengths of alkanes, branched alcohols donot follow this trend. These observations may indicate that branched alcohols areproduced by a biosynthesis pathway distinct from that which leads to unbranchedwax compound classes.Finally, it is possible that changes in the relative abundance of guard cells mayalso contribute to observed coverage and/or compositional shifts, especially since waxbiosynthesis gene expression within these cells can differ from that of neighboringpavement cells [71]. However, guard cells are much smaller than both pavement andtrichome cells, there were only half as many guard cells as pavement cells present, andthe ratio of guard cells to pavement cells remained constant throughout the periodof leaf development studied here (Fig. 6.1). These observations make it unlikely thatguard cells contribute to developmental changes in wax coverage and compositionreported here.6.4 ConclusionsIn this study, I monitored the development of Arabidopsis leaf surfaces using acombination of morphological, chemical, and gene expression measurements. Devel-opmental shifts in the relative abundance of wax compound classes and chain lengths1646. Arabidopsis leaf wax developmentfollowed the cell cycle arrest front, and could be attributed mainly to differential ex-pression of CER6. Furthermore, the elongation-associated enzymes CER2 and CER26seem to be present in relatively high abundance early in leaf development, awaitingincrease in CER6 protein availability to produce additional wax compounds of higherchain length. The mechanism by which alkanes become the dominant compound classas the leaf ages is unknown, however, it seems possible that CER6-derived wax pre-cursors may have a higher likelihood of entering the decarbonylation pathway thanother pathways. Based on these findings, it is interesting to speculate on possiblereasons for the synchrony between cell expansion and the shifts in wax composition,and how these may relate to the development of leaf structure and function. Cu-ticular wax affects the mechanical properties of the cuticle [214], and high alkaneabundance may be correlated with an effective transpiration barrier and high tem-perature tolerance [123, 155]. Thus, perhaps a FA-rich wax mixture is advantageousearly during leaf expansion, and a more alkane-rich wax mixture during late expansionand beyond. Though conjecture, these interpretations represent possible directionsfor future research.6.5 Experimental6.5.1 Plant material, growth conditions, and leaf harvestingArabidopsis thaliana wild type (Col-0) and gl1 (SALK 039478C) seeds were germi-nated on agar plates and then transferred to and grown in soil as described previously[118]. The sizes of eighth leaf blades and petioles were determined daily using a ruler.6.5.2 Leaf morphological analysisEighth leaves were harvested every two days between five and 21 days of agefor morphometrics. Five leaves were segmented (Fig. E.2), then each segment wasstained with propidium iodide (100 mg/ml, Sigma) and each segment was capturedwith confocal laser-scanning microscope photographs (Radiance 2000, Bio-Rad, ex.568 nm, em. 580-600 nm). Images were processed with ImageJ to obtain the totalarea and number of pavement and guard cells on each segment. The mean number1656. Arabidopsis leaf wax developmentand size of cells on each leaf were determined by averaging across segments. Anadditional five leaves were photographed under a light microscope, and the numberof trichomes was counted.6.5.3 Wax sample preparation and GC analysisEighth leaves were harvested every four days between five and 21 days of ageby cutting between blade and petiole. Leaf blades with a sum surface area of atleast 10 cm2 were used for a single wax sample, and five independent samples wereprepared at each time point. Wax samples were analyzed with GC-MS and GC-FIDas described previously [37]. To detect meaningful trends over time, R [169] was usedto conduct permutation tests on the ordinary least squares slopes of the data fromgl1 mutant and wild-type leaves. A total of 42 tests were performed using the gl1mutant data (see Fig. 6.3): one for each compound class (seven tests), one for therelative abundance of each homolog within the total wax mixture (26 tests), and onefor the relative abundance of the sum total of compounds derived from each precursorchain length (nine tests). The same tests were also performed on the data from wild-type leaves (see Fig. 6.4). Example R code used to conduct this analysis appears inTable E.4.For leaf tip and base analyses, 30 eighth leaves at 13 days of age were harvested,cut into thirds, and then the 30 tip sections and 30 base sections were divided into fourindependent collections each. A total of 29 permutation tests on the mean differencebetween the relative abundance of each homolog in each compound class of the baseand tip samples was performed to detect significant effects (see Fig. 6.6). ExampleR code used to conduct this analysis appears in Appendix S1. Prior to extraction,each independent set of leaves was photographed alongside a ruler, and ImageJ wasused to count leaf pixels to determine the surface areas extracted. Waxes from eachindependent set of leaves were then extracted and analyzed in the same way as thewhole leaf wax samples.1666. Arabidopsis leaf wax development6.5.4 RNA extraction and gene expression analysis by quan-titative RT-PCRFor gene expression analysis, eighth leaves were harvested every four days be-tween five and 21 days of age and immediately frozen in liquid nitrogen. An averageamount of 50 mg plant material was homogenized with Zirconia beads (2 nm diame-ter, BioSpec Products) at 4◦C using the Precellys-24 homogenizer (Bertin, 5500 rpm,2 x 25 s). Total RNA was extracted using the PureLink RNA mini kit (Invitrogen) asdescribed in the manufacturers protocol. On-column DNA digestions were performedusing PureLink DNase Set (Invitrogen) following the manufacturers protocol. The in-tegrity of extracted RNA and absence of genomic DNA was confirmed by agarose gelelectrophoresis (2%), and concentrations and purity were determined measuring UVspectra and 260/280 and 260/230 ratios using a NanoDrop 8000 Spectrophotome-ter (ThermoFischer Scientific). Samples with a 260/230 ratio between 1.7 and 2.2were selected as templates for qRT-PCR. For first-strand cDNA synthesis, 5 µg totalRNA and Oligo(dT)20 primers (Invitrogen) were used together with SuperScript Re-verse Trancriptase II (Invitrogen) following the manufacturers protocol, and resultingcDNA samples were stored at -20◦C.The expression level of wax biosynthesis genes was measured by quantitative RT-PCR using iQ SYBR Green Supermix (Bio-Rad) following the manufacturers pro-cedure and using gene-specific primers (Figure E.6). Hard-shell 96-well PCR plateswith thin walls were used (Bio-Rad) with the CFX Connect Real-time PCR DetectionSystem (Bio-Rad) under the following PCR conditions: one cycle at 95◦C for 3 min,followed by 40 cycles at 95◦C for 15 s, and at 60◦C for 30 s. A total amount of 50ng cDNA was used in a 10 µl reaction volume. Each sample was run in triplicatealongside a no template control, and five to seven independent measurements of eachgene’s expression level were at each time point. The primer efficiency was tested foreach primer pair (between 93% and 108%) and used for normalization. CFX ConnectReal-time PCR Detection System software (Bio-Rad) was used for data acquisition,and data were analyzed by adjusting the threshold cycles. Relative expression wascalculated using the Pfaffl method [163]. Four genes (GAPDH, UBC21, Actin2 andUBQ10 ) were tested as reference genes, and UBC was used for normalization todetermine relative gene expression.1676. Arabidopsis leaf wax developmentJust as for the chemical data, permutation tests were conducted on the leastsquares slopes of the expression profiles of the 12 genes were conducted to detect sig-nificant changes over time. To determine which effects remained robust to temperingof possible inflation of gene expression levels at day five, a sensitivity analysis of theseeffects was performed. The output of these statistical procedures is summarized inFig. 6.7, while R code and full analytical details appear in Table E.4.6.5.5 Adjustment for multiple comparisonsTo control the possible inflation of the rate of Type 1 error due to the manystatistical comparisons being made, the Benjamini-Hochberg method was used so thatthe study-wide expected proportion of falsely rejected null hypotheses was no morethan 1% [18]. All effects found statistically significant according to this procedurewere flagged with an asterisk.6.6 Supplementary dataSupplementary data for this chapter can be found in Appendix EFigure E.1 Morphological data for developing wild-type Arabidopsis eighthleaves.Figure E.2 Sampling scheme for studying cell size distributions across sectionsof Arabidopsis wild-type leaves of different ages.Figure E.3 Calculation of wax production rates.Table E.1 Wax composition on gl1 eighth leaves.Table E.2 Wax composition on wild-type eighth leaves.Table E.3 Wax composition on bases and tips of wild-type eighth leaves.1686. Arabidopsis leaf wax developmentFigure E.4 Calculated trichome wax composition.Figure E.5 Calculated trichome wax composition.Figure E.6 List of primer sequences used for qRT-PCR analysis.Table E.4 Permutation tests and statistical analysis.Table E.5 Statistical analysis of the gene expression data.169Chapter 7Conclusions and future directionsIn this work I present the results of my studies on the structural diversity and bi-ological variability of cuticular waxes. First, I investigated wax compound structuraldiversity. In Chapter 2, I made our knowledge of the model Arabidopsis cuticles morecomplete by investigating the structure and abundance of branched wax compounds.I found iso-alkanes and iso-alcohols on leaf and flower surfaces, respectively. Theirabundances on mutant plant surfaces were affected differently than were the abun-dances of unbranched compounds. For example, while the abundances of unbranchedcompounds were severely affected on cer2 and cer26, the abundances of branchedcompounds were not. The opposite was true of cer16, where branched compoundswere absent, and unbranched compounds were unaffected. In Chapter 3, I expandedand codified our knowledge of specialty wax compound structure and biosynthesis.These had been found with one, two, three, and in one case four secondary functionalgroups and have characteristic total carbon numbers (TCNs) and isomer distribu-tions. Specialty compounds with diverse structures occur on the surfaces of a widevariety of plant species that span diverse lineages of the plant kingdom and similarcompounds have been found in algae. In Chapter 4 I used organic synthesis andmass spectrometry to determine the structures of novel specialty wax compoundsfrom the moss Funaria hygrometrica. I also investigated the biological variabilityin wax coverage and composition. In Chapter 5 I determined that these β-hydroxyfatty acid esters and diol esters comprise a substantial portion of the wax covering F.hygrometrica surfaces and that exact coverages of these and other wax compounds1707. Conclusions and future directionsdiffer on each F. hygrometrica aerial surface, but are overall reminiscent of wax mix-tures found on vascular plants. Finally, in Chapter 6 I returned to the Arabidopsiscuticle model and performed a high-quality analysis of the changes in wax coverageand composition on developing Arabidopsis leaves. This revealed that wax coverageon leaf epidermal cells is probably constant during growth, though wax compositionchanges from being fatty acid- to alkane-dominated and from C24/C26 to C30/C32.These shifts paralleled the movement of the cell cycle arrest front and correlated withshifts in the expression of the CER6 KCS enzyme.Between Chapters 2, 3, 5, and 6, differences in wax composition were documentedamong different surfaces of different age, different epidermal cell types, different or-gans, and different species, confirming age, cell type, organ, and species levels asdimensions in which wax compositional diversity exists. These were often presentas chain length differences. Furthermore, differences in compound class compositioncould often be attributed to chain length-specifying enzymes and proteins, as thatof the fatty acid and alkane difference that might be attributed to CER6 expression(Chapter 6). Overall, these findings point to the importance of substrate and productchain length specificity in wax biosynthesis, and that the control of such specificitiesis a major way in which plants tune their surface compositions.Chapters 3 and 4 include substantial amounts of information about the isomercharacteristics of specialty compounds, which are particularly important characteris-tics when attempting to determine how secondary functional groups may be installedin specialty wax compound biosynthesis.Here therefore, the findings from the data chapters will be integrated to brieflydiscuss, offer speculation, and indicate directions for future research on substrateand product chain length specificity in wax biosynthesis (7.1) and the installationof secondary functional groups in specialty wax compound biosynthesis (7.2). Toconclude I will briefly discuss how both chain length distributions and the presence ofwax compounds with secondary functional groups might affect the physical properties,and thus functions, of cuticular wax mixtures (7.3).1717. Conclusions and future directions7.1 Substrate and product chain length profilespecificity in wax biosynthesisThis work has revealed that cuticular wax composition differs between surfacesat different developmental stages and surfaces of different cell types on Arabidopsis(Chapter 6), between surfaces of different organs on both Arabidopsis and Funaria hy-grometrica (Chapters 2 and 5), and between different species (Chapter 3). This workalso points to several mechanisms by which such differences might be achieved, includ-ing differential expression of the KCS genes CER6 and KCS16 or KCS8 to generatecompound class and chain length gradients in Arabidopsis (Chapter 6), CER16 toregulate differential accumulation of branched wax compounds in Arabidopsis (Chap-ter 2), and probably through regulation of elongation intermediate intercept processesto control specialty compound production in many species (Chapter 3). Among these,the expression of CER6 is particularly important for generating different wax com-positions, as exemplified by the analysis of cuticular waxes from Arabidopsis leavesat different developmental stages and surfaces of different F. hygrometrica structures(Chapters 6 and 5). In each of these, CER6 or a CER6 homolog shifts the chainlength and compound class profile of the wax mixture towards C30 and towards alkaneproduction, thus generating a combined chain length and compound class shift.Although information about the role of the CER6 enzyme in wax formation hadbeen available for more than 20 years [95, 60, 89], the mechanisms by which it exhibitschain length specificity, both in the products it produces and the substrates it selectsfor elongation, are unclear. A molecular caliper mechanism had been discovered ina similar condensing enzyme in yeast that effectively measures the chain length theproduct [46]. In this mechanism the enzyme aligns the aliphatic tail of the conden-sation reaction product along a transmembrane α-helix, and if the chain terminusreaches a cysteine residue at the far end of the helix then no further condensationreactions take place and the rounds of elongation cease. Site-directed mutagenesishad then demonstrated that shifting the terminal cysteine residue up the -alpha he-lix and closer to the active site yielded correspondingly shorter elongation products.However, whether this or a similar mechanism is used by CER6 is unclear, thoughthe idea is attractive. In contrast, hypotheses about how substrate specificity might1727. Conclusions and future directionsbe effected in CER6 are not prominent in the literature. It is difficult to envision aphysical mechanism such as a calipier that might preclude the entrance of acyl-CoAsubstrates that are “too short” to fit into the enzyme active site. An alternative isthat substrates below a certain chain length are not physically available to CER6 andare retained in other subcompartments of the endoplasmic reticulum (ER) membraneor are acted upon by other enzymes before reaching CER6.A further level of complexity is added to KCS chain length specificity by the effectsof the gene products of CER2 and CER26, which modulate CER6 chain length speci-ficity, as evidenced here with Arabidopsis mutant wax composition data (Chapter 2),and as had been demonstrated by others with heterologous expression [84, 160, 82].Based on the results of the mutant data I present it also seems that CER16 may havethe ability to influence the specificity of CER6 towards iso-branched substrates. Themechanism by which these specificity-modulating proteins interact with CER6 is en-tirely unknown. It could be possible that these auxiliary proteins bind to CER6 andinfluence the physical structure of its substrate binding pocket to induce specificity.Alternatively, these enzymes could increase the stability of the CER6 elongation com-plex, thus allowing CER6 to execute further rounds of elongation prior to productrelease. However, these possibilities are speculation and by no means represent allpossible mechanisms by which shifts in product specificity might arise. Thus, theelucidation of either the CER6 product chain length-specifiying mechanism or thatby which it interacts with specificity-modulating proteins represents an excellent di-rection for future research that will have broad impact. One of the first steps towardsthis goal will be the fine mapping of CER16 so that it may be cloned and used inexperiments requiring heterologous expression, thus determining if it is indeed similarto CER2 and CER26.C30 and C30-derived compounds and the alkane compound class were both highlyabundant in F. hygrometrica and in A. thaliana where these coincided with highexpression of CER6. This suggests that CER6 may also affect compound class distri-butions and thus there may be a mechanism by which the differential flux of substratethrough each modification pathway is influenced by this KCS enzyme. One particu-larly attractive mechanism for this is a potential physical association between CER6and the enzymes of the decarbonylation pathway, CER3 and CER1. Such associa-1737. Conclusions and future directionstion would explain the accumulation of C29 alkane on sporophyte moss tissues, whichotherwise have a highly active acyl reduction pathway, as well as the increase in therelative abundance of C29 and C31 alkane on Arabidopsis leaves as CER6 expres-sion increases but expression of modification pathway enzymes remains unchanged.Such association between the elongation pathway and modification pathway enzymeswould represent a mechanism by which plants could biosynthesize a wax mixtureof high purity, which is a prerequisite for wax crystal formation. This observationcould also be explained if modification enzymes were to exhibit strong chain lengthspecificity, an idea that has received relatively little testing. Some investigation intomodification enzyme chain length specificity has been done with acyl reduction en-zymes in Triticum aestivum, which exhibited only loose specificity [243], but none fordecarbonylation pathway enzymes. Thus, the mechanism by which CER6 influencescompound class distribution, whether it be through enzyme association, modificationpathway enzyme chain length specificity, or some other means, is unclear, and is adirection for future research.7.2 Installation of secondary functional groups inspecialty wax compound biosynthesisThis work confirms the wide-spread occurrence of wax compounds that bear sec-ondary functional groups, here referred to as specialty wax compounds. I cata-loged and expanded our knowledge about the diversity of the chemical structuresand biosynthesis of these compounds with experimental work in F. hygrometricaand through literature review (Chapters 4 and 3, respectively). From the moss Iidentified two new compound classes, β-hydroxy fatty acid esters and diol esters,whose monomers had secondary functional groups on odd-numbered carbons only,suggesting biosynthesis by an intercept/reentry mechanism. In the review, I identi-fied that two principal methods of biosynthesis are probably responsible for specialtywax compounds: P450 oxidation and intercept/reentry processes. The abundances ofthe specialty compounds I found in F. hygrometrica waxes and those that had beenreported in literature both suggest that, while some specialty compounds present intrace amounts may arise as leakages from the elongation pathway, many accumulate1747. Conclusions and future directionsand are even the major wax compound on the surface and are thus deliberate devia-tions from ubiquitous wax compound biosynthesis. While a P450 enzyme involved inspecialty wax compound biosynthesis has been characterized [72], no genes involvedin intercept/reentry biosynthesis of wax compounds have been identified, even thoughboth genetic evidence [186] and structural characteristics (Chapter 3) suggest thatgenes dedicated to these processes exist. Accordingly, such investigations representexcellent starting points for future work.Whether or not all specialty wax compound biosynthesis enzymes may be uniqueto specialty compound pathways, or whether ubiquitous compound biosynthesis en-zymes may be repurposed for the production of specialty compounds is also unclear.Such may depend on whether ubiquitous wax biosynthesis enzyme binding pocketscan tolerate substrates with additional secondary functional groups. The involve-ment of ubiquitous wax compound biosynthesis enzymes CER3, CER1, and CER6in the biosynthesis of branched wax compounds does indicate that these enzymes doat least tolerate substrates with iso-branched aliphatic chains, and thus they mayperhaps tolerate further variability in substrate structure in the form of secondaryfunctional groups. A specialty compound-rich species with corresponding mutantlines affected in ubiquitous wax compound biosynthesis would be a good startingplace from which to resolve such uncertainty, for example, a gymnosperm or grassspecies with a mutation in respective CER3 or CER1 homologs.7.3 Influence of chain length distributions and sec-ondary functional groups on the physical prop-erties of wax mixturesBoth specialty compounds and ubiquitous wax compounds had been found ondiverse species in crystalline form [112, 20, 178, 94, 110]. Furthermore, ubiquitouscompounds had always been found to accompany crystalline specialty compounds,though crystalline ubiquitous compounds have been found unaccompanied by eventrace amounts of specialty compounds. This indicates that certain plant species,though they can and do produce ubiquitous compounds, choose to synthesize mostly1757. Conclusions and future directionsspecialty compounds with which to form surface crystals. In turn, this suggests thatthe crystals formed by these two groups of compounds (specialty and ubiquitous) maydiffer in their physical properties and the functions they serve on plant surfaces.However, not all plant surfaces bear wax crystals; some are coated only with a waxfilm. Wax compounds in these mixtures are typically characterized by having muchbroader chain length distributions than those that comprise crystalline wax mixtures.Interestingly, different surfaces of the same species may differ in the crystallinity ofthe wax mixtures they bear, for example, Arabidopsis stems bear substantial amountsof secondary alcohol crystals, while leaves of the same plant are covered with a waxfilm. This suggests that each type of wax mixture (crystalline or film) may also servea different function and also indicates that, for some functions, high crystallinity maybe counterproductive. In this context, my observations of differences in the relativeabundance of iso-alcohols, C35 and C37 alkanes and alkenes, and low abundance spe-cialty compounds between waxy surfaces of different age, on different plant structuresor organs, or on different species (Chapters 6, 5, 3) indicate that doping the overallwax mixture with minor amount of certain compounds may be a means by whichplants fine-tune the crystallinity of their wax mixtures to optimize function.This idea will be supported by identifying genes potentially dedicated to the pro-duction of these minor components of the wax mixture and by determining if theexpression of such correlates with wax composition, thereby demonstrating that geneexpression is not only used to drive major changes in wax composition, as documentedin Chapter 6, but also to potentially provide fine control over the crystallinity or otherphysical characteristics of the wax mixture, and thus its function. This idea could befurther explored by measuring physical properties of wax mixtures or cuticles withand without these minor components. For example, the melting points of isolated,respective wax mixtures, or the permeabilities of respective, intact cuticles.Finally, my finding of both ubiquitous and specialty wax compounds in the mossF. hygrometrica and in the angiosperm A. thaliana in addition to my literature surveydemonstrate that, whether in wax crystal or wax film form, and whether specialty orubiquitous, wax compounds are found in plant species all across the plant kingdom.Furthermore, my literature survey indicates that chemical compounds with the samestructural characteristics as specialty wax compounds have been found in multiple1767. Conclusions and future directionsalgae species (Chapter 3). This suggests that specialty wax compound biosynthesismachinery and thus possibly ubiquitous wax compound biosynthesis machinery orhighly similar enzymes were present in plant ancestors before they colonized land.This has implications for our vision of cuticle evolution, indicating that cuticle formingmachinery was probably recruited from other, existing metabolic pathways as plantsmoved to dryer climes.177Bibliography[1] Amina H. Abou-Donia, Soad M. Toaima, Hala M. Hammoda, Eman Shawky,Eri Kinoshita, and Hiromitsu Takayama. Phytochemical and biological investi-gation of Hymenocallis littoralis Salisb. Chemistry and Biodiversity, 5(2):332–340, 2008.[2] G. Agar, A. Adiguzel, O. Baris, M. Gulluce, and F. Sahin. Phenotypic andgenetic variation of some Salvia species grown in eastern Anatolia region ofTurkey. Asian Journal of Chemistry, 20(5):3935–3944, 2008.[3] Toshihiro Akihisa, Akiko Nozaki, Youhei Inoue, Ken Yasikawa, YoshimasaKasahara, Shigeyasu Motohashi, Kunio Kumaki, Norio Tokutake, MichioTakido, and Toshitake Tamura. Alkane diols from flower petals of Carthamustinctorius . Phytochemistry, 45(4):725–728, 1997.[4] Toshihiro Akihisa, Hirotoshi Oinuma, Toshitake Tamura, Yoshimasa Kasahara,Kunio Kumaki, Ken Yasikawa, and Michio Takido. Erythro-hentriacontane-6,8-diol and 11 other alkane-6,8-diols from Carthamus tinctorius . Phytochemistry,36(1):105–108, 1994.[5] Megan Andriankaja, Stijn Dhondt, Stefanie De Bodt, Hannes Vanhaeren,Frederik Coppens, Liesbeth De Milde, Per Mu¨hlenbock, Aleksandra Skirycz,Nathalie Gonzalez, Gerrit T.S. Beemster, and Dirk Inze´. Exit from prolifera-tion during leaf development in Arabidopsis thaliana: a not-so-gradual process.Developmental Cell, 22(1):64–78, 2012.[6] D.S.J. Atkin and R. J. Hamilton. The changes with age in the epicuticular waxof sorghum bicolor. Journal of Natural Products, 45(6):697–703, 1950.178Bibliography[7] Lieˆn Bach, Louise V. Michaelson, Richard Haslam, Yannick Bellec, Lionel Gis-sot, Jessica Marion, Marco Da Costa, Jean-Pierre Boutin, Martine Miquel,Fre´de´rique Tellier, Frederic Domergue, Jonathan E. Markham, Frederic Beau-doin, Johnathan A. Napier, and Jean-Denis Faure. The very-long-chain hydroxyfatty acyl-CoA dehydratase PASTICCINO2 is essential and limiting for plantdevelopment. Proceedings of the National Academy of Sciences, 105(38):14727–31, sep 2008.[8] E A Baker and Grace M Hunt. Secondary alcohols from Fragaria leaf waxes.Phytochemistry, 18(1969):1059–1060, 1979.[9] Edward A. Baker and Robyn E. Gaskin. Composition of leaf epicuticular waxesof Pteridium sub-species. Phytochemistry, 26(10):2847–2848, 1987.[10] Edward A. Baker and Peter J. Holloway. Branched-chain constituents of Brus-sels sprout wax. Phytochemistry, 14:2463–2467, 1975.[11] Karen V Baker, John M Brown, Nigel Hughes, A Jerome Skarnulis, and AnnSexton. Mechanical activation of magnesium turnings for the preparation ofreactive grignard reagents. Journal of Organic Chemistry, 56:698–703, 1991.[12] Hendrik Bargel, Kerstin Koch, Zdenek Cerman, and Christoph Neinhuis. Struc-turefunction relationships of the plant cuticle and cuticular waxes a smartmaterial? Functional Plant Biology, 33(10):893, 2006.[13] Gilles Barnathan, Nathalie Bourgougnon, and Jean-Michel Kornprobst.Methoxy fatty acids isolated from the red alga, Schizymenia dubyi . Phyto-chemistry, 47(5):761–765, 1998.[14] W. Barthlott, C. Neinhuis, R. Jetter, T. Bourauel, and M. Riederer. Waterlily,poppy, or sycamore: On the systematic position of Nelumbo. Flora, 191(2):169–174, 1996.[15] Wilhelm Barthlott, Christoph Neinhuis, David Cutler, Friedrich Ditsch, IrisMeusel, Inge Theisen, and Hiltrud Wilhelmi. Classification and terminology ofplant epicuticular waxes. Botanical Journal of the Linnean Society, 126:237–260, 1998.179Bibliography[16] Fakhri A. Bazzaz, Nona R. Chiariello, Phyllis D. Coley, and Louis F. Pitelka. Al-locating resources to reproduction and defense. BioScience, 37(1):58–67, 2016.[17] Fre´de´ric Beaudoin, Xianzhong Wu, Fengling Li, Richard P. Haslam, Jonathan E.Markham, Huanquan Zheng, Johnathan A. Napier, and Ljerka Kunst. Func-tional characterization of the Arabidopsis β-ketoacyl-coenzyme A reductasecandidates of the fatty acid elongase. Plant Physiology, 150(3):1174–91, jul2009.[18] Yoav Benjamini and Yosef Hochberg. Controlling the false discovery rate: apractical and powerful approach to multiple testing. Journal of the Royal Sta-tistical Society. Series B, 57(1):289–300, 1995.[19] Richard N Bennett and Roger M Wallsgrove. Secondary metabolites in plantdefence mechanisms. New Phytologist, 127(4):617–633, 1994.[20] Martina J. Benz, Elena V. Gorb, and Stanislav N. Gorb. Diversity of theslippery zone microstructure in pitchers of nine carnivorous Nepenthes taxa.Arthropod-Plant Interactions, 6(1):147–158, 2012.[21] A. Benzioni and M. Forti. Jojoba. In G. Robbelen, R. K. Downey, and A. Ashri,editors, Oil Crops of the World, pages 448–461. McGraw-Hill, New York, 1989.[22] J. M. Berg, J. L. Tymoczko, and L. Stryer. There are three common types ofmembrane lipids. In Biochemistry, chapter 12.3. W. H. Freeman and Company,New York, 5th edition, 2002.[23] Virginia Seymour Berg. Physical, chemical and spectral properties of leaf sur-faces of Berberis aquifolium (Pursh.). Plant, Cell and Environment, 8(9):631–638, 1985.[24] Ame´lie Bernard, Fre´de´ric Domergue, Ste´phanie Pascal, Reinhard Jetter, Char-lotte Renne, Jean-Denis Faure, Richard P. Haslam, Johnathan A. Napier, Rene´Lessire, and Je´roˆme Joube`s. Reconstitution of plant alkane biosynthesis in yeastdemonstrates that Arabidopsis ECERIFERUM1 and ECERIFERUM3 are corecomponents of a very-long-chain alkane synthesis complex. The Plant Cell,24(7):1–14, jul 2012.180Bibliography[25] Ame´lie Bernard and Je´roˆme Joube`s. Arabidopsis cuticular waxes: advances insynthesis, export and regulation. Progress in Lipid Research, 52(1):110–29, oct2013.[26] Peter Blokker, Stefan Schouten, Jan W. De Leeuw, Jaap S. Sinninghe Damste´,and Herman van den Ende. Molecular structure of the resistant biopolymer inzygospore cell walls of Chlamydomonas monoica. Planta, 207(4):539–543, 1999.[27] Ferdinand Bohlmann, Jasmin Jakupovic, and Angelica Schuster. 8-hydroxypegolettiolide, a sesquiterpene lactone with a new carbon skeleton andfurther constituents from Pegolettia senegalensis . Phytochemistry, 22(7):1637–1644, 1983.[28] Brice Bourdenx, Ame´lie Bernard, Fre´de´ric Domergue, Ste´phanie Pascal, Aman-dine Le´ger, Dominique Roby, Marjorie Pervent, Denis Vile, Richard P. Haslam,Johnathan A. Napier, Rene´ Lessire, and Je´roˆme Joube`s. Overexpression ofArabidopsis ECERIFERUM1 promotes wax very-long-chain alkane biosynthe-sis and influences plant response to biotic and abiotic stresses. Plant Physiology,156(1):29–45, may 2011.[29] Katja Bringe, Christina F.A. Schumacher, Michaela Schmitz-Eiberger, UlrikeSteiner, and Erich-Christian Oerke. Ontogenetic variation in chemical and phys-ical characteristics of adaxial apple leaf surfaces. Phytochemistry, 67(2):161–70,jan 2006.[30] Paul Brouwer, Adrie van der Werf, Henriette Schluepmann, Gert-Jan Reichart,and Klaas G. J. Nierop. Lipid yield and composition of Azolla filiculoides andthe implications for biodiesel production. Bioenergy Research, 9(1):369–377,2015.[31] Gregory J. Buda, William J. Barnes, Eric A. Fich, Sungjin Park, Trevor H.Yeats, Lingxia Zhao, David S. Domozych, and Jocelyn K. C. Rose. An ATPbinding cassette transporter is required for cuticular wax deposition and desic-cation tolerance in the moss Physcomitrella patens. The Plant Cell, 25(10):4000–13, oct 2013.181Bibliography[32] Jessica M. Budke, Bernard Goffinet, and Cynthia S. Jones. A hundred-year-oldquestion: is the moss calyptra covered by a cuticle? A case study of Funariahygrometrica. Annals of Botany, 107:1279–1286, 2011.[33] Jessica M Budke, Bernard Goffinet, and Cynthia S Jones. The cuticle on thegametophyte calyptra matures before the sporophyte cuticle in the moss Fu-naria hygrometrica (Funariaceae). American Journal of Botany, 99(1):14–22,jan 2012.[34] Jessica M Budke, Bernard Goffinet, and Cynthia S Jones. Dehydration pro-tection provided by a maternal cuticle improves offspring fitness in the mossFunaria hygrometrica. Annals of Botany, 11(5):781–789, 2013.[35] Christopher Buschhaus, Dana Hager, and Reinhard Jetter. Wax layers on Cos-mos bipinnatus petals contribute unequally to total petal water resistance. PlantPhysiology, 167:80–88, 2015.[36] Christopher Buschhaus, Hubert Herz, and Reinhard Jetter. Chemical compo-sition of the epicuticular and intracuticular wax layers on adaxial sides of Rosacanina leaves. Annals of Botany, 100(7):1557–1564, 2007.[37] Christopher Buschhaus and Reinhard Jetter. Composition and physiologicalfunction of the wax layers coating Arabidopsis leaves: β-amyrin negatively af-fects the intracuticular water barrier. Plant Physiology, 160(2):1120–9, oct 2012.[38] Christopher Buschhaus, Chen Peng, and Reinhard Jetter. Very-long-chain 1,2-and 1,3-bifunctional compounds from the cuticular wax of Cosmos bipinnatuspetals. Phytochemistry, 91:249–256, 2013.[39] Lucas Busta, Jessica M. Budke, and Reinhard Jetter. Identification of β-hydroxy fatty acid esters and primary, secondary-alkanediol esters in cuticularwaxes of the moss Funaria hygrometrica. Phytochemistry, 121:38–49, 2016.[40] Mary E. Byrne. Networks in leaf development. Current Opinion in PlantBiology, 8(1):59–66, 2005.182Bibliography[41] D. Cervantes, S. D. Eigenbrode, H-J. Ding, and N. Bosque-Perez. Ovipositionresponses by Hessian fly, Mayetiola destructor, to wheats varying in surfacewaxes. Journal of Chemical Ecology, 28(1):193–210, 2002.[42] Xinbo Chen, S. Mark Goodwin, Virginia L. Boroff, Xionglun Liu, andMatthew A. Jenks. Cloning and characterization of the WAX2 gene of Ara-bidopsis involved in cuticle membrane and wax production. The Plant Cell,15(May):1170–1185, 2003.[43] A. C. Chibnall and S. H. Piper. The metabolism of plant and insect waxes.Biochemical Journal, 28(6):2209–2219, 1934.[44] Chang-Chi Chu, Thomas P. Freeman, James S. Buckner, Thomas J. Hen-neberry, Dennis R. Nelson, and Eric T. Natwick. Susceptibility of upland cottoncultivars to Bemisia tabaci Biotype B (Homoptera: Aleyrodidae) in relation toleaf age and trichome density. Annals of the Entomological Society of America,94(5):743–749, 2001.[45] Phyllis D. Coley, John P. Bryant, and Stuart F. III Chapin. Resource availabilityand plant antiherbivore defense. Science, 230(4728):895–899, 1985.[46] Vladimir Denic and Jonathan S Weissman. A molecular caliper mechanism fordetermining very long-chain fatty acid length. Cell, 130(4):663–77, aug 2007.[47] D. N. Dhar and R. C. Munjal. Chemical examination of the fruits of Platanusorientalis . Planta Medica, 29:75–76, 1976.[48] Jeroen S. Dickschat, Hilke Bruns, and Ramona Riclea. Novel fatty acid methylesters from the actinomycete Micromonospora aurantiaca. Beilstein Journal ofOrganic Chemistry, 7:1697–1712, 2011.[49] P. J. Dierickx. New β-diketones from Buxus sempervirens . Phytochemistry,12:1498–1499, 1973.[50] Eva Dominguez, Jose´ Alejandro Heredia-Guerrero, and Antonio Heredia. Thebiophysical design of plant cuticles: an overview. New Phytologist, 189(4):938–949, 2011.183Bibliography[51] P. M. Donnelly, D. Bonetta, H. Tsukaya, R. E. Dengler, and N. G. Dengler. Cellcycling and cell enlargement in developing leaves of Arabidopsis. DevelopmentalBiology, 215(2):407–19, 1999.[52] B.K. Dwivedi and B.K. Mehta. Chemical investigation of aliphatic compoundsof Piper betle (leaf stalk). J. Nat. Prod. Plant Resour., 1(2):18–24, 2011.[53] D. Edwards. Cells and tissues in the vegetative sporophytes of early land plants.New Phytologist, 125:225–247, 1993.[54] Idan Efroni, Yuval Eshed, and Eliezer Lifschitz. Morphogenesis of simple andcompound leaves: a critical review. Plant Cell, 22(4):1019–1032, 2010.[55] Sanford D. Eigenbrode and Satish K. Pillai. Neonate Plutella xylostella re-sponses to surface wax components of a resistant cabbage (Brassica oleracea).Journal of Chemical Ecology, 24(10):1611–1627, 1998.[56] Hans J. Ensikat, Petra Ditsche-Kuru, Christoph Neinhuis, and WilhelmBarthlott. Superhydrophobicity in perfection: the outstanding properties ofthe lotus leaf. Beilstein Journal of Nanotechnology, 2:152–161, 2011.[57] Jeffrey J. Esch, Margaret Chen, Mark Sanders, Matthew Hillestad, SampsonNdkium, Brian Idelkope, James Neizer, and M. David Marks. A contradictoryGLABRA3 allele helps define gene interactions controlling trichome develop-ment in Arabidopsis. Development, 130(24):5885–5894, 2003.[58] David Evans, A. Knights, and Antony L. Ritchie. β-diketones in Rhododendronwaxes. Phytochemistry, 14:2447–2451, 1975.[59] Eberhard Fehling and Kumar Mukherjee. Acyl-CoA elongase from a higherplant (Lunaria annua): metabolic intermediates of very-long-chain acyl-CoAproducts and substrate specificity. Biochemica et Biophysica Acta, 1082:239–246, 1991.[60] A. Fiebig, J. A. Mayfield, N. L. Miley, S. Chau, R. L. Fischer, and D. Preuss.Alterations in CER6, a gene identical to CUT1, differentially affect long-chainlipid content on the surface of pollen and stems. The Plant Cell, 12(10):2001–8,2000.184Bibliography[61] Gottfried Fraenkel. The raison d’etre of secondary plant substances. Science,129:1466 – 1470, 1959.[62] Robert A. Franich, Steven J. Goodin, and John K. Volkman. Alkyl esters fromPinus radiata foliage epicuticular wax. Phytochemsitry, 24(12):2949–2952, 1985.[63] Robert A. Franich, Ann P. Gowar, and John K. Volkman. Secondary diols ofPinus radiata needle epicuticular wax. Phytochemistry, 18(9):1563–1564, 1979.[64] Author J. C. French and D. J. Paolillo. On the role of the calyptra in permittingexpansion of capsules in the moss Funaria. The Bryologist, 78(4):438–446, 1975.[65] S. N. Garg, M. S. Siddiqui, and S. K. Agarwal. New fatty acid esters andhydroxy ketones from fruits of Laurus nobilis . Journal of Natural Products,55(9):1315–1319, 1992.[66] Franc¸ois Gelin, John K. Volkman, Jan W. de Leeuw, and Jaap S. SinningheDamste´. Mid-chain hydroxy long-chain fatty acids in microalgae from the genusNannochloropsis . Phytochemistry, 45(4):641–646, 1997.[67] Samir A. Gharbo, M.R.I. Saleh, and E.M.A. Shukry. Phytochemical investiga-tion of Thesium humile Vahl. Planta Medica, 17(3):124–125, 1962.[68] Beverley J. Glover. Differentiation in plant epidermal cells. Journal of experi-mental botany, 51(344):497–505, 2000.[69] S. Mark Goodwin, Aaron M. Rashotte, Musrur Rahman, Kenneth A. Feld-mann, and Matthew A. Jenks. Wax constituents on the inflorescence stemsof double eceriferum mutants in Arabidopsis reveal complex gene interactions.Phytochemistry, 66(7):771–80, apr 2005.[70] Christine Granier, Catherine Massonnet, Olivier Turc, Bertrand Muller, KarineChenu, and Franc¸ois Tardieu. Individual leaf development in Arabidop-sis thaliana: a stable thermal-time-based programme. Annals of Botany,89(5):595–604, 2002.185Bibliography[71] Julie E. Gray, Geoff H. Holroyd, Frederizue M. van der Lee, Ahmad R. Bahrami,Peter C. Sijmons, F. Ian Woodward, Wolfgang Schuch, and Alistair M. Hether-ington. The HIC signalling pathway links CO2 perception to stomatal develop-ment. Nature, 408(6813):713–716, 2000.[72] Stephen Greer, Miao Wen, David Bird, Xuemin Wu, Lacey Samuels, LjerkaKunst, and Reinhard Jetter. The cytochrome P450 enzyme CYP96A15 is themidchain alkane hydroxylase responsible for formation of secondary alcohols andketones in stem cuticular wax of Arabidopsis. Plant Physiology, 145(3):653–67,nov 2007.[73] Sonia Guimil and Christophe Dunand. Cell growth and differentiation in Ara-bidopsis epidermal cells. Journal of Experimental Botany, 58(14):3829–3840,2007.[74] P.-G. Gu¨lz, E. Mu¨ller, K. Schmitz, F. J. Marner, and S. Gu¨th. Chemical compo-sition and surface structures of epicuticular leaf waxes of Ginkgo biloba, Mag-nolia grandiflora and Liriodendron tulipifera. Zeitschrift fu¨r Naturforschung,47c:516–526, 1992.[75] P.-G. Gu¨lz, R. B. N. Prasad, and E. Mu¨ller. Surface structures and chemicalcomposition of epicuticular waxes during leaf development of Fagus sylvaticaL. Zeitschrift fu¨r Naturforschung, 47c:190–196, 1992.[76] Frank D. Gunstone, Janet A. Holliday, and Charles M. Scrimgeour. The long-chain oxo acids (argemonic acid) in Argemone mexicana seed oil. Chemistryand Physics of Lipids, 20:331–335, 1977.[77] M. S. Gu¨nthardt-Goerg. Epicuticular wax of needles of Pinus cembra, Pinussylvestris and Picea abies . European Journal of Forest Pathology, 16(7):400–408, 1986.[78] K. Haas. Surface wax of Andreaea and Pogonatum species. Phytochemistry,21(3):657–659, 1982.186Bibliography[79] K. Haas. The surface lipids of Saelania moss gametophytes: a comparison withcuticular wax of higher plants. In D.F. Cutler, K.L. Alvin, and C.E. Price,editors, The Plant Cuticle, pages 225–230. Academic Press, London, 1982.[80] K. Haas and J. Scho¨nherr. Composition of soluble cuticular lipids and waterpermeability of cuticular membreanes from Citrus leaves. Planta, 146:399–403,1979.[81] Abdelali Hannoufa, John McNevin, and Bertrand Lemieux. Epicuticular waxesof eceriferum mutants of Arabidopsis thaliana. Phytochemistry, 33(4):851–855,1993.[82] Tegan M. Haslam, Richard Haslam, Didier Thoraval, Ste´phanie Pascal, CamilleDelude, Fre´de´ric Domergue, Aurora Man˜as Ferna´ndez, Fre´de´ric Beaudoin,Johnathan A. Napier, Ljerka Kunst, and Je´roˆme Joube`s. CER2-LIKE pro-teins have unique biochemical and physiological functions in very-long-chainfatty acid elongation. Plant Physiology, 167:682–692, 2015.[83] Tegan M. Haslam and Ljerka Kunst. Extending the story of very-long-chainfatty acid elongation. Plant Science, 210:93–107, sep 2013.[84] Tegan M Haslam, Aurora Man˜as Ferna´ndez, Lifang Zhao, and Ljerka Kunst.Arabidopsis ECERIFERUM2 is a component of the fatty acid elongation ma-chinery required for fatty acid extension to exceptional lengths. Plant Physiol-ogy, 160(November):1164–1174, aug 2012.[85] Horacio Heinzen and Patrick Moyna. Surface waxes from seeds of Hordeumvulgare. Phytochemistry, 27(2):429–431, 1988.[86] Antonio Heredia. Biophysical and biochemical characteristics of cutin, a plantbarrier biopolymer. Biochimica et Biophysica Acta, 1620(1):1–7, mar 2003.[87] Daniel A. Herms and William J. Mattson. The dilemma of plants: to grow ordefend. The Quarterly Review of Biology, 67(3):283–335, 2016.[88] P. J. Holloway, G. A. Brown, E. A. Baker, and M. J. K. Macey. Chemicalcomposition and ultrastructure of the epicuticular wax in three lines of Brassicanapus (L). Chemistry and Physics of Lipids, 19:114–127, 1977.187Bibliography[89] Tanya S. Hooker, Anthony A. Millar, and Ljerka Kunst. Significance of theexpression of the CER6 condensing enzyme for cuticular wax production inArabidopsis. Plant Physiology, 129(August):1568–1580, 2002.[90] Ralph W. Howard and James E. Baker. Cuticular hydrocarbons and wax es-ters of the ectoparasitoid Habrobracon hebetor : ontogenetic, reproductive, andnutritional effects. Archives of Insect Biochemistry and Physiology, 53(1):1–18,2003.[91] Lijun Huang, Xiong Yan Chen, Yeonggil Rim, Xiao Han, Won Kyong Cho,Seon Won Kim, and Jae Yean Kim. Arabidopsis glucan synthase-like 10 func-tions in male gametogenesis. Journal of Plant Physiology, 166(4):344–352, 2009.[92] Tal Isaacson, Dylan K. Kosma, Antonio J. Matas, Gregory J. Buda, YonghuaHe, Bingwu Yu, Arika Pravitasari, James D. Batteas, Ruth E. Stark,Matthew A. Jenks, and Jocelyn K. C. Rose. Cutin deficiency in the tomato fruitcuticle consistently affects resistance to microbial infection and biomechanicalproperties, but not transpirational water loss. The Plant Journal, 60(2):363–377, 2009.[93] Larry L. Jackson. Beta-diketone, alcohol and hydrocarbons of barley surfacelipids. Phytochemistry, 10:487–490, 1971.[94] C. E. Jeffree, E. A. Baker, and P. J. Holloway. Ultrastructure and recrystalliza-tion of plant epicuticular waxes. New Phytologist, 75:539–549, 1975.[95] M. A. Jenks, H. A. Tuttle, S. D. Eigenbrode, and K. A. Feldmann. Leaf epi-cuticular waxes of the eceriferum mutants in Arabidopsis. Plant Physiology,108(1):369–377, may 1995.[96] Matthew A. Jenks, Hillary A. Tuttle, and Kenneth A. Feldmann. Changes inepicuticular waxes on wildtype and ESCERIFERUM mutants in Arabidopsisduring development. Phytochemistry, 42(1):29–34, 1996.[97] R Jetter. Long-chain alkanediols from Myricaria germanica leaf cuticular waxes.Phytochemistry, 55(2):169–76, sep 2000.188Bibliography[98] R Jetter and S Scha¨ffer. Chemical composition of the Prunus laurocerasus leafsurface. Dynamic changes of the epicuticular wax film during leaf development.Plant Physiology, 126(4):1725–37, aug 2001.[99] Reinhard Jetter, Ljerka Kunst, and A. Lacey Samuels. Composition of plantcuticular waxes. In M. Riederer and C. Mueller, editors, Biology of the PlantCuticle, pages 145–181. Blackwell, Oxford, 2006.[100] Reinhard Jetter and Markus Riederer. Cuticular waxes from the leaves and fruitcapsules of eight Papaveraceae species. Canadian Journal of Botany, 74(3):419–430, mar 1996.[101] Reinhard Jetter and Markus Riederer. Homologous long-chain δ-lactones in leafcuticular waxes of Cerinthe minor. Phytochemistry, 50:1359–1364, 1999.[102] Reinhard Jetter and Markus Riederer. Long-chain alkanediols, ketoaldehydes,ketoalcohols and ketoalkyl esters in the cuticular waxes of Osmunda regalisfronds. Phytochemistry, 52:907–915, 1999.[103] Reinhard Jetter and Markus Riederer. Composition of cuticular waxes on Os-munda regalis fronds. Journal of Chemical Ecology, 26(2):399 – 412, 2000.[104] Je´roˆme Joube`s, Sylvain Raffaele, Brice Bourdenx, Christel Garcia, JeannyLaroche-Traineau, Patrick Moreau, Fre´de´ric Domergue, and Rene´ Lessire. TheVLCFA elongase gene family in Arabidopsis thaliana: phylogenetic analysis, 3Dmodelling and expression profiling. Plant Molecular Biology, 67(5):547–66, jul2008.[105] Olga Jovanovic´, Niko Radulovic´, Radosav Palic´, and Bojan Zlatkovic´. Volatilesof Minuartia recurva (All.) Schinz et Thell. subsp. recurva (Caryophyllaceae)from Serbia. Journal of Essential Oil Research, 21(5):429–432, 2009.[106] Ellen V. Kearns and Sarah M. Assmann. The guard cell-environment connec-tion. Plant Physiology, 102:711–715, 1993.[107] Zeba Khan, M. Ali, and Priyanka Bagri. A new steroidal glycoside and fatty acidesters from the stem bark of Tectona grandis Linn. Natural Product Research,24(11):1059–68, jul 2010.189Bibliography[108] Juyoung Kim, Jin Hee Jung, Saet Buyl Lee, Young Sam Go, Hae Jin Kim,Rebecca Cahoon, Edgar B Cahoon, Jonathan E. Markham, and Mi ChungSuh. Arabidopsis 3-ketoacyl-CoA synthase 9 is involved in the synthesis oftetracosanoic acids as precursors of cuticular waxes, suberins, sphingolipids,and phospholipids. Plant Physiology, 162(2):567–80, apr 2013.[109] Myoung-Seok Kim, Kang-Bo Shim, Si-Hyung Park, and Kwan-Su Kim.Changes in cuticular waxes of developing leaves in sesame (Sesamum indicumL.). Journal of Crop Science and Biotechnology, 12(3):161–167, 2009.[110] K. Koch, W. Barthlott, S. Koch, a. Hommes, K. Wandelt, W. Mamdouh, S. De-Feyter, and P. Broekmann. Structural analysis of wheat wax (Triticum aes-tivum, c.v. ’Naturastar’ L.): From the molecular level to three dimensionalcrystals. Planta, 223(2):258–270, 2006.[111] Kerstin Koch, Bharat Bhushan, Hans-Ju¨rgen Ensikat, and Wilhelm Barthlott.Self-healing of voids in the wax coating on plant surfaces. Philosophical Trans-actions of the Royal Society A, 367(1894):1673–88, may 2009.[112] Kerstin Koch and Hans-Ju¨rgen Ensikat. The hydrophobic coatings of plantsurfaces: epicuticular wax crystals and their morphologies, crystallinity andmolecular self-assembly. Micron, 39(7):759–772, oct 2008.[113] Kerstin Koch, Jan-Peter Frahm, and Rossarin Pollawatn. The cuticle of theBuxbaumia viridis sporophyte. Flora, 204(1):34–39, jan 2009.[114] P E Kolattukudy. Species specificity in the biosynthesis of branched paraffinsin leaves. Plant Physiology, 43(9):1423–1429, 1968.[115] P. E. Kolattukudy. Chemistry and biochemistry of natural waxes. ElsevierScientific Pub. Co., Amsterdam, NY, 1976.[116] Dylan K Kosma, Jill a Nemacheck, Matthew a Jenks, and Christie E Williams.Changes in properties of wheat leaf cuticle during interactions with Hessian fly.The Plant Journal, 63(1):31–43, jul 2010.190Bibliography[117] A. B. Kroumova, Z. Xie, and G. J. Wagner. A pathway for the biosynthesisof straight and branched, odd- and even-length, medium-chain fatty acids inplants. Proceedings of the National Academy of Sciences of the United Statesof America, 91(24):11437–11441, 1994.[118] Christine Lai, Ljerka Kunst, and Reinhard Jetter. Composition of alkyl estersin the cuticular wax on inflorescence stems of Arabidopsis thaliana cer mutants.The Plant Journal, 50(2):189–96, apr 2007.[119] D. Lang, J. Eisinger, R. Reski, and S. A. Rensing. Representation and high-quality annotation of the Physcomitrella patens transcriptome demonstrates ahigh proportion of proteins involved in metabolism in mosses. Plant Biology,7(3):238–250, 2005.[120] John C. Larkin, Matt L. Brown, and John Schiefelbein. How do cells knowwhat they want to be when they grow up? Lessons from epidermal patterningin Arabidopsis. Annual Review of Plant Biology, 54(1):403–430, 2003.[121] J.L. Laseter, D.J. Weber, and J. Oro. Characterization of cabbage leaf lipids:n-alkanes, ketone, and fatty acids. Phytochemistry, 7:1005–1008, 1968.[122] Jeongyeo Lee, Kyungbong Yang, MiYe Lee, Sewon Kim, Jungeun Kim,Soohwan Lim, Gwan-Ho Kang, Sung Ran Min, Sun-Ju Kim, Sang Un Park,Young Soo Jang, Soon Sung Lim, and HyeRan Kim. Differentiated cuticu-lar wax content and expression patterns of cuticular wax biosynthetic genes inbloomed and bloomless broccoli (Brassica oleracea var. italica). Process Bio-chemistry, 50:456–462, 2015.[123] Jana Leide, Ulrich Hildebrandt, Kerstin Reussing, Markus Riederer, and GerdVogg. The developmental pattern of tomato fruit wax accumulation and itsimpact on cuticular transpiration barrier properties: effects of a deficiency in aβ-ketoacyl-coenzyme A synthase (LeCER6). Plant Physiology, 144(3):1667–79,2007.[124] Fengling Li, Xuemin Wu, Patricia Lam, David Bird, Huanquan Zheng, LaceySamuels, Reinhard Jetter, and Ljerka Kunst. Identification of the wax ester191Bibliographysynthase/acyl-coenzyme A: diacylglycerol acyltransferase WSD1 required forstem wax ester biosynthesis in Arabidopsis. Plant Physiology, 148(1):97–107,sep 2008.[125] Li Li, Piyada Thipyapong, David C. Breeden, and John C. Steffens. Overex-pression of a bacterial branched-chain alpha-keto acid dehydrogenase complexin Arabidopsis results in accumulation of branched-chain acyl-CoAs and alter-ation of free amino acid composition in seeds. Plant Science, 165(6):1213–1219,2003.[126] Shiyou Lu, Tao Song, Dylan K. Kosma, Eugene P. Parsons, Owen Rowland,and Matthew A. Jenks. Arabidopsis CER8 encodes LONG-CHAIN ACYL-COASYNTHETASE 1 (LACS1) that has overlapping functions with LACS2 in plantwax and cutin synthesis. The Plant Journal, 59(4):553–64, aug 2009.[127] MJK Macey. The effect of light on wax synthesis in leaves of Brassica oleracea.Phytochemistry, 9(1968):757–761, 1970.[128] Andre´ Macherius, Peter Kuschk, Claus Haertig, Monika Moeder, Natalia I.Shtemenko, Antonio Heredia Bayona, Jose´ A. Heredia Guerrero, and ManfredGey. Composition changes in the cuticular surface lipids of the helophytes,Phragmites australis and Juncus effusus as result of pollutant exposure. Envi-ronmental Science and Pollution Research, 18(5):727–33, jun 2011.[129] M. Magdy, O. Werner, S. F. McDaniel, B. Goffinet, and R. M. Ros. Genomicscanning using AFLP to detect loci under selection in the moss Funaria hygro-metrica along a climate gradient in the Sierra Nevada Mountains, Spain. PlantBiology, 18:280–288, 2015.[130] Azhar Mahmood, Rehana Ahmed, and Shaheen Kosar. Phytochemical screen-ing and biological activities of the oil components of Prunus domestica Linn.Journal of Saudi Chemical Society, 13(3):273–277, 2009.[131] C. G.-A. Maier and Dusty Post-Beittenmiller. Epicuticular wax on leek invitro developmental stages and seedlings under varied growth conditions. PlantScience, 134(1):53–67, 1998.192Bibliography[132] M. David Marks. Molecular genetic analysis of trichome development in Ara-bidopsis. Annual Review of Plant Physiology and Plant Molecular Biology,48(1):137–163, 1997.[133] M. David Marks, Jonathan P. Wenger, Edward Gilding, Ross Jilk, andRichard A. Dixon. Transcriptome analysis of Arabidopsis wild-type and gl3-sstsim trichomes identifies four additional genes required for trichome develop-ment. Molecular Plant, 2(4):803–822, 2009.[134] C. M C Martins, S. M M Mesquita, and W. L C Vaz. Cuticular waxes of theHolm (Quercus ilex L. subsp. ballota (Desf.) Samp.) and cork (Q. suber L.)oaks. Phytochemical Analysis, 10(1):1–5, 1999.[135] Rodney Mauricio. Ontogenetics of QTL: The genetic architecture of trichomedensity over time in Arabidopsis thaliana. Genetica, 123:75–85, 2005.[136] J. P. McNevin, W. Woodward, A. Hannoufa, K. A. Feldmann, and B. Lemieux.Isolation and characterization of eceriferum (cer) mutants induced by T-DNAinsertions in Arabidopsis thaliana. Genome, 36(3):610–8, jun 1993.[137] Laurence Me´janelle, Angels Sanchez-Gargallo, Ilhem Bentaleb, and Joan O.Grimalt. Long chain n-alkyl diols, hydroxy ketones and sterols in a marineeustigmatophyte, Nannochloropsis gaditana, and in Brachionus plicatilis feedingon the algae. Organic Geochemistry, 34(4):527–538, 2003.[138] I. Meusel, C. Neinhuis, C. Marksta¨dter, and W. Barthlott. Chemical compo-sition and recrystallization of epicuticular waxes: Coiled rodlets and tubules.Plant Biology, 2:462–470, 2000.[139] Jørn Dalgaard Mikkelsen. Structure and biosynthesis of β-diketones in barleyspike epicuticular wax. Carlsberg Research Communications, 44(3):133–147,1979.[140] Jorn Dalgaard Mikkelsen. Biosynthesis of esterified alkan-2-ols and B-diketonesin barley spike epicuticular wax: synthesis of radioactive intermediates. Carls-berg Research Communications, 49:391–416, 1984.193Bibliography[141] A Millar and L Kunst. Very-long-chain fatty acid biosynthesis is controlledthrough the expression and specificity of the condensing enzyme. The PlantJournal, 12(1):121–31, jul 1997.[142] K. Mladenova, B. Stoianova-Ivanova, C. M. Camaggi, and B. Stolanova-Ivanova.Composition of neutral components in flower of some decorative roses. Phyto-chemistry, 16(2):269–272, 1977.[143] J D Mold, R E Means, R K Stevens, and J M Ruth. The paraffin hydrocarbonsof wool wax. Normal, iso, anteiso, and other branched isomers. Biochemistry,3(9):1293–1295, 1964.[144] B. Muckensturm, D. Foechterlen, J. P. Reduron, P. Danton, and M. Hilden-brand. Phytochemical and chemotaxonomic studies of Foeniculum vulgare.Biochemical Systematics and Ecology, 25(4):353–358, 1997.[145] S. S. Nadakuduti, M. Pollard, D. K. Kosma, C. Allen, J. B. Ohlrogge, and C. S.Barry. Pleiotropic Phenotypes of the sticky peel Mutant Provide New Insightinto the Role of CUTIN DEFICIENT2 in Epidermal Cell Function in Tomato.Plant Physiology, 159(3):945–960, 2012.[146] M.A.H. Nagalakshmi and K. Sri Rama Murthy. Phytochemical profile of crudeseed oil of Wrightia tinctoria R. BR. and Wrightia arborea (DENNST.) MABB.by GC-MS. Internation Journal of Pharmaceutical Sciences Review and Re-search, 31(2):46–51, 2015.[147] Paula C. Nakosteen and Karen W. Hughes. Sexual life cycle of three species ofFunariaceae in culture. The Bryologist, 81(2):307–314, 1978.[148] Utpal Nath, Brian C. W. Crawford, Rosemary Carpenter, and Enrico Coen.Genetic control of surface curvature. Science, 299(5611):1404–1407, 2003.[149] Thorben Nawrath, Klaus Gerth, Rolf Mu¨ller, and Stefan Schulz. Volatile methylesters of medium chain length from the bacterium Chitinophaga Fx7914. Chem-istry and Biodiversity, 7:2228–2253, 2010.194Bibliography[150] Ishrat Naz, Saifullah, and M. R. Khan. Nematicidal activity of nonacosane-10-ol and 23a-homostigmast-5-en-3β-ol isolated from the roots of Fumaria parvi-flora (Fumariaceae). Journal of Agricultural and Food Chemistry, 61:5689–5695,2013.[151] C. Neinhuis and R. Jetter. Epicuticular wax of Polytrichales sporophytes. Jour-nal of Bryology, 18:399–406, 1995.[152] C. Neinhuis and R. Jetter. Ultrastructure and chemistry of epicuticular waxcrystals in Polytrichales sporophytes. Journal of Bryology, 18:339–406, 1995.[153] A.G. Netting and M.J.K. Macey. The composition of ketones and secondaryalcohols from Brassica oleracea waxes. Phytochemistry, 10(8):1917–1920, 1971.[154] A.G. Netting, M.J.K. Macey, and H.N. Barber. Chemical genetics of a sub-glaucous mutant of Brassica oleracea. Phytochemistry, 11(2):579–585, 1972.[155] Y. Ni, Y. J. Guo, J. Wang, R. E. Xia, X. Q. Wang, G. Ash, and J. N. Li.Responses of physiological indexes and leaf epicuticular waxes of Brassica napusto Sclerotinia sclerotiorum infection. Plant Pathology, 63(1):174–184, apr 2013.[156] Biljana Nikolic´, Vele Tesˇevic´, Iris orevic´, Marina Todosijevic´, Milka Jadranin,Srdjan Bojovic´, and Petar D. Marin. Variability of n-alkanes and nonacosan-10-ol in natural populations of Picea omorika. Chemistry & Biodiversity,10(3):473–83, mar 2013.[157] James E. Oliver, Robert P. Doss, R. Thomas Williamson, John R. Carney,and E. David DeVilbiss. Bruchinsmitogenic 3-(Hydroxypropanoyl) esters oflong chain diols from weevils of the Bruchidae. Tetrahedron, 56(39):7633–7641,2000.[158] Toshihiko Osawa and Mitsuo Namiki. Natural antioxidants isolated from Euca-lyptus leaf waxes. Journal of Agricultural and Food Chemistry, 36(4):673–676,1988.[159] Roy Osborne and Jan F. Stevens. Epicuticular waxes and glaucousness of En-cephalartos leaves. Phytochemistry, 42(5):1335–1339, 1996.195Bibliography[160] Ste´phanie Pascal, Ame´lie Bernard, Maud Sorel, Marjorie Pervent, Denis Vile,Richard P. Haslam, Johnathan A. Napier, Rene´ Lessire, Fre´de´ric Domergue,and Je´roˆme Joube`s. The Arabidopsis cer26 mutant, like the cer2 mutant, isspecifically affected in the very-long-chain fatty acid elongation process. ThePlant Journal, 73(5):733–46, feb 2013.[161] Kevin E. Percy, Sirkku Manninen, Karl-Heinz Ha¨berle, C. Heerdt, H. Werner,Gary W. Henderson, and Rainer Matyssek. Effect of 3 years’ free-air expo-sure to elevated ozone on mature Norway spruce (Picea abies (L.) Karst.) nee-dle epicuticular wax physicochemical characteristics. Environmental Pollution,157:1657–65, 2009.[162] Stefanie Peschel, Rochus Franke, Lukas Schreiber, and Moritz Knoche. Com-position of the cuticle of developing sweet cherry fruit. Phytochemistry,68(7):1017–25, apr 2007.[163] M W Pfaffl. A new mathematical model for relative quantification in real-timeRT-PCR. Nucleic Acids Research, 29(9):e45, 2001.[164] J. A. Pino, R. Marbot, and J. Aguero. Volatile components of geneps (Melicoccabijuga L.) fruit. Journal of Essential Oil Research, 14(4):247–248, 2002.[165] L. Piovetti, A. Yani, T. Combaut, and A. Diara. Waxes of Cupressus duprezianaand Cupressus sempervirens . Phytochemestry, 20(5):1135–1136, 1981.[166] R.B.N. Prasad and P Gu¨lz. Developmental and seasonal variations in the epicu-ticular waxes of beech leaves (Fagus sylvatica L.). Zeitschrift fu¨r NaturforschungC, 45(7-8):805–812, 1990.[167] M.C.F. Proctor. Surface wax on the leaves of some mosses. Journal of Bryology,10:531–538, 1979.[168] Michael C. F. Proctor, Melvin J. Oliver, Andrew J. Wood, Lloyd R. Stark,Natalie L. Cleavitt, and Brent D. Mishler. Desiccation-tolerance in bryophytes:a review. The Bryologist, 110(4):595–621, 2007.[169] R Core Team. R: A language and environment for statistical computing, 2015.196Bibliography[170] Radu C. Racovita, Chen Peng, Takayoshi Awakawa, Ikuro Abe, and ReinhardJetter. Very-long-chain 3-hydroxy fatty acids, 3-hydroxy fatty acid methyl estersand 2-alkanols from cuticular waxes of Aloe arborescens leaves. Phytochemistry,113:183–194, sep 2014.[171] Nicola A. Ramsay and Beverley J. Glover. MYB-bHLH-WD40 protein complexand the evolution of cellular diversity. Trends in Plant Science, 10(2):63–70,2005.[172] A. M. Rashotte, M. A. Jenks, and K. A. Feldmann. Cuticular waxes on ecer-iferum mutants of Arabidopsis thaliana. Phytochemistry, 57(1):115–23, may2001.[173] Fakhria M. Razeq, Dylan K. Kosma, Owen Rowland, and Isabel Molina. Ex-tracellular lipids of Camelina sativa: characterization of chloroform-extractablewaxes from aerial and subterranean surfaces. Phytochemistry, 106:188–96, oct2014.[174] Mariza G. Reis, Aparecida D. de Faria, Isabel Alves dos Santos, Maria doCarmo E. Amaral, and Anita J. Marsaioli. Byrsonic acid the clue to flo-ral mimicry involving oil-producing flowers and oil-collecting bees. Journal ofChemical Ecology, 33:1421–1429, 2007.[175] Eva E. Reisberg, Ulrich Hildebrandt, Markus Riederer, and Ute Hentschel.Phyllosphere bacterial communities of trichome-bearing and trichomeless Ara-bidopsis thaliana leaves. Antonie van Leeuwenhoek, 101(3):551–60, mar 2012.[176] Eva E. Reisberg, Ulrich Hildebrandt, Markus Riederer, and Ute Hentschel.Distinct phyllosphere bacterial communities on Arabidopsis wax mutant leaves.PloS one, 8(11):e78613, jan 2013.[177] Andrew Richardson, Rochus Franke, Gerhard Kerstiens, Mike Jarvis, LukasSchreiber, and Wieland Fricke. Cuticular wax deposition in growing barley(Hordeum vulgare) leaves commences in relation to the point of emergence ofepidermal cells from the sheaths of older leaves. Planta, 222(3):472–83, oct2005.197Bibliography[178] Michael Riedel, Anna Eichner, Harald Meimberg, and Reinhard Jetter. Chem-ical composition of epicuticular wax crystals on the slippery zone in pitchers offive Nepenthes species and hybrids. Planta, 225(6):1517–1534, 2007.[179] Markus Riederer. Introduction: Biology of the Plant Cuticle. In MarkusRiederer and Caroline Muller, editors, Biology of the Plant Cuticle, pages 1–10.Blackwell Publishing Ltd, Oxford, 2006.[180] J. F. Rontani and C. Aubert. Trimethylsilyl transfer during electron ionizationmass spectral fragmentation of some ω-hydroxycarboxylic and ω-dicarboxylicacid trimethylsilyl derivatives and the effect of chain length. Rapid Communi-cations in Mass Spectrometry, 18(17):1889–1895, 2004.[181] Ann-Marie Rottler, Stefan Schulz, and Manfred Ayasse. Wax lipids signal nestidentity in bumblebee colonies. Journal of Chemical Ecology, 39(1):67–75, 2013.[182] Owen Rowland, Huanquan Zheng, Shelley R. Hepworth, Patricia Lam, Rein-hard Jetter, and Ljerka Kunst. CER4 encodes an alcohol-forming fatty acyl-coenzyme A reductase involved in cuticular wax production in Arabidopsis.Plant Physiology, 142(3):866–77, nov 2006.[183] F D Sack and D J Paolillo. Stomatal Pore and Cuticle Formation in Funaria.Protoplasma, 116:1–13, 1983.[184] Inno Salasoo. Effect of leaf age on epicuticular wax alkanes in Rhododendron.Phytochemistry, 22(2):461–463, 1983.[185] Lacey Samuels, Ljerka Kunst, and Reinhard Jetter. Sealing plant surfaces:cuticular wax formation by epidermal cells. Annual Review of Plant Biology,59:683–707, jan 2008.[186] Lizette M Schneider, Nikolai M Adamski, Caspar Elo Christensen, David B Stu-art, Sonia Vautrin, Mats Hansson, Cristobal Uauy, and Penny Von Wettstein-knowles. The Cer-cqu gene cluster determines three key players in a β-diketonesynthase polyketide pathway synthesizing aliphatics in epicuticular waxes. Jour-nal Of Experimental Botany, 2016.198Bibliography[187] J. Scho¨nherr. Water permeability of isolated cuticular membranes: The effectof cuticular waxes on diffusion of water. Planta, 131(2):159–164, 1976.[188] Scott Schuette, Andrew J. Wood, Matt Geisler, Jane Geisler-Lee, RobertoLigrone, and Karen S. Renzaglia. Novel localization of callose in the sporesof Physcomitrella patens and phylogenomics of the callose synthase gene fam-ily. Annals of Botany, 103(5):749–756, 2009.[189] Stefan Schulz, Cristian Arsene, Marianne Tauber, and Jeremy N. McNeil. Com-position of lipids from sunflower pollen (Helianthus annuus). Phytochemistry,54(3):325–336, 2000.[190] Lars Seipold, Gu¨nter Gerlach, and Ludger Wessjohann. A new type of floral oilfrom Malpighia coccigera (Malpighiaceae) and chemical considerations on theevolution of oil flowers. Chemistry and Biodiversity, 1(10):1519–1528, 2004.[191] Ray F. Severson, Richard F. Arrendale, Orestes T. Chortyk, Albert W. Johnson,D. Michael Jackson, G. Richard Gwynn, James F. Chaplin, and Michael G.Stephenson. Quantitation of the major cuticular components from green leaf ofdifferent tobacco types. Journal of Agricultural and Food Chemistry, 32:566–570, 1984.[192] K. Shiva Shanker, S. Kanjilal, B. V S K Rao, K. Hara Kishore, S. Misra, andR. B N Prasad. Isolation and antimicrobial evaluation of isomeric hydroxyketones in leaf cuticular waxes of Annona squamosa. Phytochemical Analysis,18(1):7–12, 2007.[193] A. Johnathan Shaw. The genetic structure of sporophytic and gametophyticpopulations of the moss, Funaria hygrometrica Hedw. Evolution, 45(5):1260–1274, 1991.[194] T. Shepherd, G. W. Robertson, D. W. Griffiths, A. N. E. Bircht, and G. Duncan.Effects of environment on the composition of epicuticular wax from kale andswede. Phytochemistry, 40(2):407–417, 1995.[195] Tom Shepherd and D Wynne Griffiths. The effects of stress on plant cuticularwaxes. The New Phytologist, 171(3):469–99, jan 2006.199Bibliography[196] Anees A. Siddiqui, Sachin M. Wani, R. Rajesh, and V. Alagarsamy. Isola-tion and hypotensive activity of five new phytoconstituents from chloroformextract of flowers of Hibiscus rosasinensis Linn. Indian Journal of Chemistry,44(4):806–811, 2005.[197] Stephen P Slocombe, Ines Schauvinhold, Ryan P McQuinn, Katrin Besser,Nicholas a Welsby, Andrea Harper, Naveed Aziz, Yi Li, Tony R Larson, JamesGiovannoni, Richard a Dixon, and Pierre Broun. Transcriptomic and reversegenetic analyses of branched-chain fatty acid and acyl sugar production inSolanum pennellii and Nicotiana benthamiana. Plant physiology, 148(4):1830–46, dec 2008.[198] Anna Smirnova, Jana Leide, and Markus Riederer. Deficiency in a very-long-chain fatty acid β-ketoacyl-coenzyme a synthase of tomato impairs microgame-togenesis and causes floral organ fusion. Plant physiology, 161(January):196–209, 2013.[199] Md Sohail Akhtar, Raad A. Kaskoos, Showkat Rasool Mir, and Mohammed Ali.New chromone derivative from Evolvulus alsinoides Linn. aerial parts. Journalof Saudi Chemical Society, 13(2):191–194, 2009.[200] Eveline N. Speelman, Gert Jan Reichart, Jan W. de Leeuw, W. Irene C Rijpstra,and Jaap S. Sinninghe Damste´. Biomarker lipids of the freshwater fern Azollaand its fossil counterpart from the Eocene Arctic Ocean. Organic Geochemistry,40(5):628–637, 2009.[201] Martin J. Steinbauer, Noel W. Davies, Cyril Gaertner, and Sylvie Derridj. Epi-cuticular waxes and plant primary metabolites on the surfaces of juvenile Euca-lyptus globulus and E. nitens (Myrtaceae) leaves. Australian Journal of Botany,57(2008):474–485, 2009.[202] Martin J. Steinbauer, Florian P. Schiestl, and Noel W. Davies. Monoterpenesand epicuticular waxes help female autumn gum moth differentiate betweenwaxy and glossy Eucalyptus and leaves of different ages. Journal of ChemicalEcology, 30(6):1117–1142, 2004.200Bibliography[203] Hans Stocker and Long Ashton. Changes in the composition of coffee wax withdevelopment. Phytochemistry, 14:1919–1920, 1975.[204] Mi Chung Suh, A. Lacey Samuels, Reinhard Jetter, Ljerka Kunst, Mike Pollard,John Ohlrogge, and Fred Beisson. Cuticular lipid composition, surface struc-ture, and gene expression in Arabidopsis stem epidermis. Plant Physiology,139(December):1649–1665, 2005.[205] Anne W Sylvester, W. Zacheus Cande, and Michael Freeling. Division anddifferentiation during normal and liguleless-1 maize leaf development. Devel-opment, 110(3):985–1000, 1990.[206] Beata Szafranek, Karolina Chrapkowska, Maria Pawin´ska, and JanuszSzafranek. Analysis of leaf surface sesquiterpenes in potato varieties. Jour-nal of Agricultural and Food Chemistry, 53(8):2817–2822, may 2005.[207] Beata M. Szafranek and Elzbieta E. Synak. Cuticular waxes from potato(Solanum tuberosum) leaves. Phytochemistry, 67(1):80–90, 2006.[208] Erica E. Tassone, Alexander E. Lipka, Pernell Tomasi, Greg T. Lohrey, WeiQian, John M. Dyer, Michael A. Gore, and Matthew A. Jenks. Chemical vari-ation for leaf cuticular waxes and their levels revealed in a diverse panel ofBrassica napus L. Industrial Crops and Products, 79:77–83, 2016.[209] Nicolas L Taylor, Joshua L Heazlewood, David A Day, and A Harvey Millar.Lipoic acid-dependent oxidative catabolism of alpha-keto acids in mitochondriaprovides evidence for branched-chain amino acid catabolism in Arabidopsis 1.Plant physiology, 134(February):838–848, 2015.[210] Abby Telfer, Krista M. Bollman, and R. Scott Poethig. Phase change andthe regulation of trichome distribution in Arabidopsis thaliana. Development,124(3):645–654, 1997.[211] Donglan Tian, John Tooker, Michelle Peiffer, Seung Ho Chung, and Gary W.Felton. Role of trichomes in defense against herbivores: Comparison of her-bivore response to woolly and hairless trichome mutants in tomato (Solanumlycopersicum). Planta, 236(4):1053–1066, 2012.201Bibliography[212] James Todd, Dusty Post-Beittenmiller, and Jan G. Jaworski. KCS1 encodesa fatty acid elongase 3-ketoacyl-CoA synthase affecting wax biosynthesis inArabidopsis thaliana. The Plant Journal, 17(2):119–130, 1999.[213] Sandra Trenkamp, William Martin, and Klaus Tietjen. Specific and differen-tial inhibition of very-long-chain fatty acid elongases from Arabidopsis thalianaby different herbicides. Proceedings of the National Academy of Sciences,101(32):11903–11908, 2004.[214] Shuntaro Tsubaki, Kazuki Sugimura, Yoshikuni Teramoto, Keizo Yonemori,and Jun-ichi Azuma. Cuticular membrane of Fuyu persimmon fruit is strength-ened by triterpenoid nano-fillers. PloS One, 8(9):1–13, 2013.[215] a P Tulloch and L L Hoffman. Leaf wax of durum wheat. Phytochemistry,10(11393):0–6, 1971.[216] A. P. Tulloch and R. O. Weenink. Composition of the leaf wax of Little Clubwheat. Canadian Journal of Chemistry, 47(fraction 1):3119–3126, 1969.[217] Alexander P. Tulloch. Composition of leaf surface waxes of Triticum species:variation with age and tissue. Phytochemistry, 12(1972):2225–2232, 1973.[218] Alexander P Tulloch. EPICUTICULAR WAX OF AGROPYRON INTER-MEDIUM. Phytochemistry, 15(1):1153–1156, 1976.[219] Alexander P Tulloch. Epicuticular Wax of Agropyron Smithii Leaves. Phyto-chemistry, 15(1):1153–1156, 1976.[220] Alexander P. Tulloch. Epicuticular wax of Eragrostis curvula. Phytochemistry,21(3):661–664, 1982.[221] Alexander P. Tulloch. Epicuticular waxes from Agropyron dasystachyum,Agropyron riparium and Agropyron elongatum. Phytochemistry, 22(7):1605–1613, 1983.[222] Alexander P. Tulloch and Lothar Bergter. Epicuticular wax of Juniperus scop-ulorum. Phytochemistry, 20(12):2711–2716, 1981.202Bibliography[223] Alexander P. Tulloch and Leslie L. Hoffman. Leaf wax of Triticum aestivum.Phytochemistry, 12(9):2217–2223, 1973.[224] Alexander P. Tulloch and Leslie L. Hoffman. Epicuticular waxes of Secalecereale and Triticale hexaploide leaves. Phytochemistry, 13(11):2535–2540, 1974.[225] Alexander P. Tulloch and Leslie L. Hoffman. Epicuticular waxes of Andropogonhallii and A. scoparius . Phytochemistry, 18:267–271, 1979.[226] AP Tulloch and LL Hoffman. Epicuticular wax of Panicum virgatum. Phyto-chemistry, 19(18032):837–839, 1980.[227] M Ukiya, T Akihisa, H Tokuda, K Koike, J Takayasu, H Okuda, Y Kimura,T Nikaido, and H Nishino. Isolation, structural elucidation, and inhibitoryeffects of terpenoid and lipid constituents from sunflower pollen on Epstein-Barrvirus early antigen induced by tumor promoter, TPA. Journal of Agriculturaland Food Chemistry, 51(10):2949–2957, 2003.[228] AyHan Ulubelen. Constituents of the leaves and the stems of Clematis vitalba.Phytochemistry, 9(1954):233–234, 1970.[229] Ayhan Ulubelen and T. Baytop. Hydrocarbons and triterpenes of the leaves ofEuonymus latifolius . Phytochemistry, 12(1973):1824, 1973.[230] M. Vajdi, W. W. Nawar, and C. Merritt. GC/MS analysis of some long chainesters, ketones and propanediol diesters. Journal of the American Oil Chemist’sSociety, 58(2):106–110, 1981.[231] Clare Van Maarseveen, Hong Han, and Reinhard Jetter. Development of thecuticular wax during growth of Kalanchoe daigremontiana (Hamet et Perr. dela Bathie) leaves. Plant, Cell and Environment, 32(1):73–81, jan 2009.[232] Giancarlo Verardo, Emanuela Pagani, Paola Geatti, and Paolo Martinuzzi. Athorough study of the surface wax of apple fruits. Analytical and BioanalyticalChemistry, 376(5):659–667, 2003.[233] Ram K. Verma and Madan M. Gupta. Lipid constituents of Tridax procumbens.Phytochemistry, 27(2):459–463, 1988.203Bibliography[234] Cornelia P. Vermeer, Peter Nastold, and Reinhard Jetter. Homologous very-long-chain 1,3-alkanediols and 3-hydroxyaldehydes in leaf cuticular waxes ofRicinus communis L. Phytochemistry, 62(3):433–438, 2003.[235] M. A. Viougeas, R. Rohr, and A. Chamel. Structural changes and permeabilityof ivy (Hedera helix L.) leaf cuticles in relation to leaf development and afterselective chemical treatments. New Phytologist, 130(3):337–348, 1995.[236] Gerd Vogg, Stephanie Fischer, Jana Leide, Eyal Emmanuel, Reinhard Jetter,Avraham A. Levy, and Markus Riederer. Tomato fruit cuticular waxes andtheir effects on transpiration barrier properties: Functional characterization ofa mutant deficient in a very-long-chain fatty acid β-ketoacyl-CoA synthase.Journal of Experimental Botany, 55(401):1401–1410, 2004.[237] John K. Volkman, Stephanie M. Barrett, and Susan I. Blackburn. Eustig-matophyte microalgae are potential sources of C29 sterols, C22C28 n-alcoholsand C28C32 n-alkyl diols in freshwater environments. Organic Geochemistry,30(5):307–318, 1999.[238] Penny von Wettstein-Knowles. Waxes, cutin, and suberin. In T.S. Moore, edi-tor, Lipid Metabolism in Plants, pages 127–166. CRC Press, Inc,, Boca Raton,1993.[239] Penny von Wettstein-Knowles. Biosynthesis and genetics of waxes. In R. J.Hamilton, editor, Waxes: chemistry, molecular biology and functions, volume 6,pages 91–129. The Oily Press, Dundee, 1995.[240] Penny Von Wettstein-Knowles. Analyses of barley spike mutant waxes identifyalkenes, cyclopropanes and internally branched alkanes with dominating isomersat carbon 9. Plant Journal, 49(2):250–264, 2007.[241] Penny von Wettstein-Knowles, Jørn Dalgaard Mikkelsen, and J. Ø Mad-sen. Nonan-2-ol esters in sorghum leaf epicuticular wax and their collec-tion by preparative gas chromatography. Carlsberg Research Communications,49(7):611–618, 1984.204Bibliography[242] G. J. Wagner, E. Wang, and R. W. Shepherd. New approaches for studyingand exploiting an old protuberance, the plant trichome. Annals of Botany,93(1):3–11, 2004.[243] Meiling Wang, Yong Wang, Hongqi Wu, Jing Xu, Tingting Li, Daniela Hege-barth, Reinhard Jetter, Letian Chen, and Zhonghua Wang. Three TaFAR genesfunction in the biosynthesis of primary alcohols and the response to abioticstresses in Triticum aestivum. Scientific Reports, 6(April):25008, 2016.[244] John D. Weete, S. Venketeswaran, and John L. Laseter. Two populations ofaliphatic hydrocarbons of teratoma and habituated tissue cultures of tobacco.Phytochemistry, 10(1963):939–943, 1971.[245] Miao Wen, Jason Au, Franka Gniwotta, and Reinhard Jetter. Very-long-chainsecondary alcohols and alkanediols in cuticular waxes of Pisum sativum leaves.Phytochemistry, 67(22):2494–2502, 2006.[246] Miao Wen, Christopher Buschhaus, and Reinhard Jetter. Nanotubules on plantsurfaces: Chemical composition of epicuticular wax crystals on needles of Taxusbaccata L. Phytochemistry, 67(16):1808–1817, 2006.[247] Miao Wen and Reinhard Jetter. Very-long-chain hydroxyaldehydes from the cu-ticular wax of Taxus baccata needles. Phytochemistry, 68(20):2563–2569, 2007.[248] Miao Wen and Reinhard Jetter. Composition of secondary alcohols, ketones,alkanediols, and ketols in Arabidopsis thaliana cuticular waxes. Journal ofExperimental Botany, 60(6):1811–21, jan 2009.[249] Shu-Jun Xu, Ping-An Jiang, Zheng-Wu Wang, and Yan Wang. Crystal struc-tures and chemical composition of leaf surface wax depositions on the desertmoss Syntrichia caninervis . Biochemical Systematics and Ecology, 37(6):723–730, dec 2009.[250] Xuewei Yang, Xin Dai, Hui Guo, Shu Geng, and Gangyi Wang. Petrodiesel-likestraight chain alkane and fatty alcohol production by the microalga Chlorellasorokiniana. Bioresource technology, 136:126–30, 2013.205Bibliography[251] Trevor H. Yeats, Gregory J. Buda, Zhonghua Wang, Noam Chehanovsky,Leonie C. Moyle, Reinhard Jetter, Arthur A. Schaffer, and Jocelyn K. C. Rose.The fruit cuticles of wild tomato species exhibit architectural and chemical di-versity, providing a new model for studying the evolution of cuticle function.The Plant Journal, 69(4):655–666, feb 2012.[252] Trevor H. Yeats and Jocelyn K. C. Rose. The formation and function of plantcuticles. Plant Physiology, 163(1):5–20, 2013.[253] Selma Yildizhan, Joop van Loon, Anna Sramkova, Manfred Ayasse, ChristianArsene, Cindy ten Broeke, and Stefan Schulz. Aphrodisiac pheromones fromthe wings of the small cabbage white and large cabbage white butterflies, Pierisrapae and Pieris brassicae. ChemBioChem, 10:1666–1677, 2009.[254] Geng Yu, Thuong T. H. Nguyen, Yongxia Guo, Ines Schauvinhold, Michele E.Auldridge, Nazmul Bhuiyan, Imri Ben-Israel, Yoko Iijima, Eyal Fridman,Joseph P. Noel, and Eran Pichersky. Enzymatic functions of wild tomatomethylketone synthases 1 and 2. Plant Physiology, 154(1):67–77, 2010.[255] Xi Zhang, Zhiyong Liu, Ping Wang, Qiushi Wang, Shuo Yang, and Hui Feng.Fine mapping of BrWax1, a gene controlling cuticular wax biosynthesis in Chi-nese cabbage (Brassica rapa L. ssp. pekinensis). Molecular Breeding, 32:867–874, jul 2013.[256] Huanquan Zheng, Owen Rowland, and Ljerka Kunst. Disruptions of the Ara-bidopsis enoyl-CoA reductase gene reveal an essential role for very-long-chainfatty acid synthesis in cell expansion during plant morphogenesis. The PlantCell, 17:1467–1481, 2005.[257] Yu-Bi Zhou, Run-Rong Ye, Xue-Feng Lu, Peng-Cheng Lin, Shi-Bing Yang,Peng-Peng Yue, Chang-Xian Zhang, and Min Peng. GCMS analysis of liposol-uble constituents from the stems of Cynomorium songaricum. Journal of Phar-maceutical and Biomedical Analysis, 49(4):1097–1100, 2009.[258] Tobias Zust, Bindu Joseph, Kentaro K Shimizu, Daniel J. Kliebenstein, andLindsay A. Turnbull. Using knockout mutants to reveal the growth costs206Bibliographyof defensive traits. Proceedings of the Royal Society B: Biological Sciences,278(1718):2598–2603, 2011.207Appendix ASupplementary data for chapter 2208A. Supplementary data for chapter 2	O6MW = 352-H2Om/z 33457Figure A.1: A) Mass spectrum of tetracosanal (2.1) and B) possible pathsto major fragments observed	8MW = 40657 363Figure A.2: A) Mass spectrum of 2-methyloctacos-4-ene (2.3) and B) pos-sible paths to major fragments observed209A. Supplementary data for chapter 2	O6MW = 486-HOSi(CH3)3m/z 396OSiSi103471Figure A.3: A) Mass spectrum of 1,22-docosandiol (2.5) and B) possiblepaths to major fragments observed	O576MW = 338-H2Om/z 320OFigure A.4: A) Mass spectrum of dicosandial (2.6) and B) possible paths tomajor fragments observed210A. Supplementary data for chapter 2	8MW = 524 O91433Figure A.5: A) Mass spectrum of (((29-methyltriaconta-3,25-dien-1-yl)oxy)methyl)benzene (2.8) and B) possible paths to major fragmentsobserved	O576MW = 306-H2Om/z 278Figure A.6: A) Mass spectrum of docosanal (2.10) and B) possible paths tomajor fragments observed211A. Supplementary data for chapter 2	O57117Si397736MW = 412Figure A.7: A) Mass spectrum of tricosan-2-ol (2.11) and B) possible pathsto major fragments observed	Br57713236MW = 402Figure A.8: A) Mass spectrum of 2-bromotricosane (2.12) and B) possiblepaths to major fragments observed212A.Supplementarydataforchapter2Table A.1: Wax composition on flowers of wild-type and mutant Arabidopsis lines. Amounts are tabulated inµg/cm2 as the mean of five independent samples, n.d. = not detected.Compound Class Cn Col0 Ler cer1 cer2 cer3Fatty acids 22 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 024 0.017 ± 0.0018 0.018 ± 0.00088 0.026 ± 0.0048 0.017 ± 0.0013 0.0043 ± 0.001826 0.0088 ± 0.0007 0.014 ± 0.0026 0.018 ± 0.0023 0.052 ± 0.0056 0.0035 ± 0.001428 0.019 ± 0.0009 0.018 ± 0.0010 0.021 ± 0.0082 0.045 ± 0.0058 0.011 ± 0.002730 0.041 ± 0.0028 0.050 ± 0.0037 0.014 ± 0.0097 0 ± 0 0.0039 ± 0.002432 0 ± 0 0 ± 0 0 ± 0.0000 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0.086 ± 0.004 0.45 ± 0.017 0.079 ± 0.0099 0.11 ± 0.012 0.022 ± 0.0063n-Alcohols 22 0.0081 ± 0.00109 0.0129 ± 0.0004 0.0090 ± 0.00077 0.0054 ± 0.0017 0.0012 ± 0.0003723 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 024 0.0060 ± 0.00218 0.0092 ± 0.0005 0.0019 ± 9.9E-04 0.0036 ± 0.00037 0.0020 ± 0.0003025 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 026 0.049 ± 0.0029 0.044 ± 0.0041 0.039 ± 0.0046 0.31 ± 0.015 0.011 ± 0.003427 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 028 0.16 ± 0.0105 0.15 ± 0.0081 0.084 ± 0.012 0.363 ± 0.018 0.029 ± 0.008229 0.077 ± 0.0032 0.098 ± 0.0014 0.042 ± 0.015 0.029 ± 0.0035 0.018 ± 0.002930 0.089 ± 0.0045 0.13 ± 0.0047 0.14 ± 0.020 0.002 ± 0.0003 0.15 ± 0.04131 0 ± 0 0.000 ± 0 0.0000 ± 0 0.000 ± 0 0 ± 032 0.0044 ± 0.0017 0.004 ± 0.0006 0.0097 ± 0.00174 0.008 ± 0.0005 0.0016 ± 0.0005533 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0.39 ± 0.024 0.01 ± 0.002 0.32 ± 0.027 0.72 ± 0.034 0.21 ± 0.055Esters 36 0.0065 ± 0.0023 0.022 ± 0.00091 0 ± 0 0.0068 ± 1.2E-03 0 ± 038 0.019 ± 0.0015 0.080 ± 0.0070 0.018 ± 0.00253 0.040 ± 0.00552 0.012 ± 0.001640 0.059 ± 0.0055 0.17 ± 0.0086 0.064 ± 1.0E-02 0.29 ± 0.018 0.031 ± 0.006242 0.096 ± 8.1E-03 0.13 ± 0.0047 0.092 ± 0.017 0.27 ± 0.016 0.14 ± 0.03744 0.070 ± 0.0028 0.040 ± 0.0056 0.13 ± 0.031 0.074 ± 0.013 0.26 ± 0.06646 0.053 ± 2.6E-03 0.061 ± 0.0054 0.039 ± 0.018 0.062 ± 0.018 0.077 ± 0.02048 0.070 ± 0.0075 0.014 ± 0.0042 0.0054 ± 0.00076 0.086 ± 0.033 0.013 ± 0.001950 0.0099 ± 0.0073 0.012 ± 0.0053 0.0071 ± 0.00080 0.015 ± 0.0062 0 ± 052 0.0087 ± 0.0050 0.010 ± 0.0035 0 ± 0 0.011 ± 0.0031 0 ± 054 0.0080 ± 0.0030 0.0000 ± 0.0000 0 ± 0 0.0073 ± 0.0049 0 ± 0Total 0.40 ± 0.035 0.010 ± 0.0020 0.35 ± 0.070 0.86 ± 0.093 0.54 ± 0.13Aldehydes 26 0.014 ± 0.0049 0 ± 0 0 ± 0 0.016 ± 0.0037 0 ± 028 0.0064 ± 0.0026 0 ± 0 0 ± 0 0.015 ± 0.0031 0 ± 030 0.012 ± 0.0061 0.010 ± 0.0020 0.035 ± 0.00257 0 ± 0 0 ± 032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Continued on next page213A.Supplementarydataforchapter2Table A.1 – Continued from previous pageCompound Class Cn Col0 Ler cer1 cer2 cer3Total 0.032 ± 0.0082 3.1 ± 0.083 0.035 ± 0.0026 0.032 ± 0.0068 0 ± 0n-Alkanes 25 0.013 ± 0.0021 0.0022 ± 0.00028 0.0027 ± 0.00106 0.026 ± 0.0019 0 ± 026 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 027 0.12 ± 0.013 0.14 ± 0.0016 0.067 ± 0.00519 0.20 ± 0.014 0 ± 028 0.062 ± 0.0080 0 ± 0 0 ± 0 0.035 ± 0.0038 0 ± 029 3.8 ± 0.28 2.6 ± 0.087 0.73 ± 0.043 0.31 ± 0.029 0.11 ± 0.02830 0.089 ± 0.0091 0.070 ± 0.0051 0.037 ± 0.0022 0.029 ± 0.0036 0.0052 ± 0.001331 0.19 ± 0.015 0.20 ± 0.0023 0.085 ± 0.0058 0.070 ± 0.010 0.036 ± 0.008132 0.016 ± 0.0022 0.016 ± 0.00027 0 ± 0 0 ± 0 0 ± 033 0.025 ± 0.0061 0.017 ± 0.0011 0 ± 0 0.0019 ± 0.00072 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 035 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 036 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 037 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 4.3 ± 0.32 0.47 ± 0.019 0.93 ± 0.047 0.68 ± 0.060 0.15 ± 0.037Ketols 29 0.043 ± 0.0029 1.1 ± 0.065 0 ± 0 0 ± 0 0 ± 0Ketones 29 1.3 ± 0.077 0 ± 0 0 ± 0 0 ± 0 0.11 ± 0.0330 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0 ± 0 0.52 ± 0.033 0 ± 0 0 ± 0 0 ± 0Total 1.3 ± 0.077 0.051 ± 0.0020 0 ± 0 0 ± 0 0.11 ± 0.03Sec. alcohols 29 0.67 ± 0.043 1.09 ± 0.065 0 ± 0 0.0038 ± 0.00051 0.039 ± 0.01130 1.3 ± 0.077 0.051 ± 0.0020 0 ± 0 0 ± 0 0 ± 031 0.043 ± 0.0029 0.13 ± 0.0019 0 ± 0 0 ± 0 0 ± 0Total 0.67 ± 0.043 1.09 ± 0.065 0 ± 0 0.0038 ± 0.00051 0.039 ± 0.011iso-Alkanes 27 0.010 ± 0.0048 0.022 ± 0.0014 0 ± 0 0.0082 ± 0.0011 0 ± 028 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 029 0.32 ± 0.025 0.30 ± 0.012 0.11 ± 0.0093 0.16 ± 0.014 0.0063 ± 0.001330 0.088 ± 0.011 0.15 ± 0.0056 0.015 ± 0.00086 0.067 ± 0.0076 0 ± 031 0.26 ± 0.021 0 ± 0 0.050 ± 0.00315 0.18 ± 0.017 0 ± 032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0 ± 0 0.0097 ± 0.0018 0 ± 0 0 ± 0 0 ± 0Total 0.68 ± 0.056 0.52 ± 0.033 0.17 ± 0.011 0.42 ± 0.039 0.0063 ± 0.0013iso-Alcohols 28 0.0030 ± 0.00071 0 ± 0 0.0029 ± 0.0016 0.0119 ± 0.00077 0.0033 ± 0.0008529 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 030 0.0068 ± 0.0012 0 ± 0 0.030 ± 0.0014 0 ± 0 0.020 ± 0.004531 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0.0515 ± 0.0020 0 ± 0 0 ± 0 0 ± 0Continued on next page214A.Supplementarydataforchapter2Table A.1 – Continued from previous pageCompound Class Cn Col0 Ler cer1 cer2 cer3Total 0.010 ± 0.0013 0.542 ± 0.030 0.032 ± 0.0022 0.012 ± 0.0008 0.023 ± 0.0052Total unidentifieds 0.76 ± 0.073 0 ± 0 0.92 ± 0.16 0.74 ± 0.072 0.66 ± 0.095Total wax 8.9 ± 0.59 0 ± 0 3.1 ± 0.27 3.7 ± 0.29 2.0 ± 0.39Table A.2: Wax composition on flowers of wild-type and mutant Arabidopsis lines. Amounts are tabulated inµg/cm2 as the mean of five independent samples, n.d. = not detected.Compound Class Cn cer4 cer6 cer16 cer26 kcs1Fatty acids 22 0 ± 0 0.010 ± 0.00063 0 ± 0 0 ± 0 0 ± 024 0.015 ± 0.0010 0.026 ± 0.0016 0.019 ± 0.0053 0.021 ± 0.0022 0.023 ± 0.002226 0.0082 ± 0.0020 0.018 ± 0.0006 0.009 ± 0.0011 0.0088 ± 0.0044 0.015 ± 0.001728 0.017 ± 0.0026 0.0073 ± 0.00039 0.011 ± 0.0022 0.020 ± 0.0035 0.027 ± 0.001530 0.028 ± 0.0038 0 ± 0 0.023 ± 0.0046 0.053 ± 0.010 0.050 ± 0.002332 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0.068 ± 0.0048 0.062 ± 0.0015 0.063 ± 0.012 0.10 ± 0.018 0.11 ± 0.0058n-Alcohols 22 0.010 ± 0.0020 0.050 ± 0.0035 0.023 ± 0.0012 0.0065 ± 6.4E-04 0.0073 ± 0.001423 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 024 0.0027 ± 0.00027 0.0092 ± 0.00094 0.0054 ± 0.0012 0.0055 ± 0.00086 0.011 ± 0.001825 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 026 0.0041 ± 0.0013 0.047 ± 0.0026 0.044 ± 0.0063 0.065 ± 0.0026 0.066 ± 0.005527 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 028 0.020 ± 0.0014 0.018 ± 0.0010 0.108 ± 0.012 0.20 ± 0.0093 0.21 ± 0.006429 0.064 ± 0.0027 0.014 ± 0.00048 0.054 ± 0.0014 0.084 ± 0.0035 0.097 ± 0.003330 0.014 ± 0.0024 0.0042 ± 0.00020 0.090 ± 0.0076 0.11 ± 0.0068 0.12 ± 0.003031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 032 0 ± 0 0 ± 0 0.0070 ± 0.00077 0.0042 ± 0.0015 0.0043 ± 0.001733 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0.11 ± 0.0074 0.14 ± 0.0078 0.33 ± 0.025 0.48 ± 0.0221 0.51 ± 0.020Esters 36 0.0017 ± 0.0011 0.0097 ± 0.00063 0.0024 ± 0.00035 0.0099 ± 0.00050 0.0084 ± 0.0036338 0.0097 ± 0.0021 0.037 ± 0.0024 0.017 ± 0.0011 0.022 ± 0.0025 0.028 ± 0.002240 0.048 ± 0.0058 0.069 ± 0.0059 0.046 ± 0.0021 0.075 ± 0.0042 0.093 ± 0.001642 0.069 ± 0.0095 0.15 ± 0.014 0.15 ± 0.0074 0.12 ± 0.0018 0.21 ± 0.009344 0.12 ± 0.020 0.048 ± 0.0028 0.11 ± 0.0076 0.089 ± 0.0078 0.15 ± 0.003446 0.046 ± 0.0072 0.048 ± 0.0065 0.023 ± 0.0026 0.053 ± 0.015 0.051 ± 0.003048 0.089 ± 0.013 0.071 ± 0.0060 0.044 ± 0.0013 0.050 ± 0.019 0.14 ± 0.01250 0 ± 0 0.035 ± 0.0034 0.0073 ± 0.00067 0.0063 ± 0.0034 0.0074 ± 0.0025Continued on next page215A.Supplementarydataforchapter2Table A.2 – Continued from previous pageCompound Class Cn cer4 cer6 cer16 cer26 kcs152 0 ± 0 0 ± 0 0.0099 ± 2.0E-04 0.0075 ± 0.0019 0.0069 ± 0.0007354 0 ± 0 0 ± 0 0.0080 ± 2.2E-03 0.0062 ± 0.0025 0.0088 ± 0.00053Total 0.38 ± 0.055 0.47 ± 0.034 0.41 ± 0.017 0.43 ± 0.036 0.70 ± 0.016Aldehydes 26 0 ± 0 0 ± 0 0 ± 0 0.013 ± 0.00093 0.0055 ± 0.002128 0.0037 ± 0.0026 0 ± 0 0 ± 0 0.011 ± 0.0013 0.0078 ± 0.004530 0.0037 ± 0.0014 0 ± 0 0.0051 ± 0.00018 0.0093 ± 0.0029 0.019 ± 0.01432 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0.0073 ± 0.0019 0 ± 0 0.0051 ± 0.00018 0.033 ± 0.0036 0.032 ± 0.019n-Alkanes 25 0.012 ± 0.0041 0.025 ± 0.0022 0.0077 ± 0.0007 0.015 ± 0.0034 0.014 ± 0.002726 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 027 0.11 ± 0.0083 0.12 ± 0.0069 0.14 ± 0.0036 0.11 ± 0.010 0.12 ± 0.009728 0.057 ± 0.00049 0.034 ± 0.0036 0 ± 0 0.055 ± 0.0034 0.062 ± 9.6E-0329 3.8 ± 0.31 0.15 ± 0.0061 2.3 ± 0.13 4.0 ± 0.099 4.3 ± 0.08030 0.080 ± 0.0069 0.0068 ± 0.00036 0.10 ± 0.0032 0.086 ± 0.0035 0.089 ± 0.003331 0.19 ± 0.012 0.028 ± 0.00075 0.20 ± 0.0058 0.20 ± 0.0044 0.17 ± 0.004132 0.020 ± 0.0010 0 ± 0 0.0055 ± 0.00064 0.013 ± 0.0010 0.013 ± 0.001633 0.0071 ± 0.0012 0 ± 0 0.0057 ± 0.0023 0.026 ± 0.014 0.026 ± 0.01234 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 035 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0.000 ± 0.000036 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 037 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 4.2 ± 0.32 0.37 ± 0.017 2.8 ± 0.14 4.6 ± 0.11 4.8 ± 0.11Ketols 29 0.048 ± 0.0027 0 ± 0 0.047 ± 0.0059 0.042 ± 0.0020 0.047 ± 0.0035Ketones 29 1.3 ± 0.13 0 ± 0 0.54 ± 0.037 1.4 ± 0.065 1.7 ± 0.02930 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 1.3 ± 0.13 0 ± 0 0.54 ± 0.037 1.4 ± 0.065 1.7 ± 0.029Sec. alcohols 29 0.67 ± 0.071 0 ± 0 0.34 ± 0.023 0.76 ± 0.036 0.86 ± 0.01330 1.3 ± 0.13 0 ± 0 0.54 ± 0.037 1.4 ± 0.065 1.7 ± 0.02931 0.048 ± 0.0027 0 ± 0 0.047 ± 0.0059 0.042 ± 0.0020 0.047 ± 0.0035Total 0.67 ± 0.071 0 ± 0 0.34 ± 0.023 0.76 ± 0.036 0.86 ± 0.013iso-Alkanes 27 0.010 ± 0.0019 0.025 ± 0.0023 0.0014 ± 0.00020 0.0085 ± 0.0023 0.0060 ± 0.001228 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 029 0.40 ± 0.015 0.10 ± 0.0046 0.0077 ± 0.0067 0.28 ± 0.015 0.38 ± 0.02230 0.089 ± 0.0039 0 ± 0 0.0046 ± 0.0012 0.074 ± 0.0068 0.10 ± 0.007131 0.31 ± 0.012 0.033 ± 0.0012 0 ± 0 0.24 ± 0.013 0.35 ± 0.02032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Continued on next page216A.Supplementarydataforchapter2Table A.2 – Continued from previous pageCompound Class Cn cer4 cer6 cer16 cer26 kcs1Total 0.81 ± 0.029 0.16 ± 0.0078 0.014 ± 0.0077 0.60 ± 0.037 0.83 ± 0.0502iso-Alcohols 28 0 ± 0 0.0084 ± 0.00027 0 ± 0 0.0038 ± 0.00012 0.0055 ± 0.001729 0 ± 0 0.0026 ± 0.00045 0 ± 0 0 ± 0 0 ± 030 0 ± 0 0 ± 0 0 ± 0 0.0067 ± 0.00064 0.0076 ± 0.002031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0 ± 0 0.011 ± 0.00025 0 ± 0 0.011 ± 0.0006 0.013 ± 0.0028Total unidentifieds 0.66 ± 0.010 0.78 ± 0.064 0.91 ± 0.020 0.70 ± 0.062 0.94 ± 0.11Total wax 8.5 ± 0.57 2.1 ± 0.12 5.6 ± 0.24 9.3 ± 0.24 11 ± 0.24217A.Supplementarydataforchapter2Table A.3: Wax composition on leaves of wild-type and mutant Arabidopsis lines. Amounts are tabulated inµg/cm2 as the mean of five independent samples, n.d. = not detected.Compound Class Cn Col0 Ler cer1 cer2 cer3Fatty acids 22 0.00032 ± 0.00015 0.0019 ± 0.0010 0.0016 ± 0.00049 0.00028 ± 0.00011 0.0056 ± 0.003824 0.0056 ± 0.00073 0.018 ± 0.0039 0.0059 ± 0.0012 0.0075 ± 0.00089 0.0079 ± 0.001226 0.011 ± 0.0040 0.043 ± 0.0045 0.023 ± 0.0028 0.042 ± 0.0045 0.0411 ± 0.004128 0.0081 ± 0.0027 0.021 ± 0.0034 0.0040 ± 0.0011 0.012 ± 0.0016 0.018 ± 0.001230 0.0052 ± 0.0021 0.010 ± 0.0011 0.0034 ± 0.00075 0.015 ± 0.0013 0.0092 ± 0.001532 0.0064 ± 0.0020 0.022 ± 0.0057 0.0048 ± 0.0019 0.025 ± 0.0057 0.030 ± 0.004034 0.0056 ± 0.0014 0.0087 ± 0.0030 0 ± 0 0.032 ± 0.0057 0 ± 0Total 0.043 ± 0.012 0.12 ± 0.021 0.043 ± 0.0067 0.13 ± 0.014 0.11 ± 0.011n-Alcohols 22 0.0075 ± 0.00095 0.0049 ± 0.0033 0 ± 0 0.0022 ± 0.00064 0.0057 ± 0.003023 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 024 0.0037 ± 0.00069 0.0094 ± 0.0049 0.00012 ± 4.5E-05 0.00074 ± 0.00025 0.00059 ± 0.0002925 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 026 0.0068 ± 0.0028 0.012 ± 0.0017 0.0039 ± 0.00068 0.0085 ± 0.0021 0.017 ± 0.008727 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 028 0.018 ± 0.0064 0.021 ± 0.0017 0.0086 ± 0.00073 0.022 ± 0.0029 0.046 ± 0.02129 0.0074 ± 0.0016 0.0073 ± 0.0017 0 ± 0 0.0065 ± 0.0016 0 ± 030 0.0090 ± 0.0017 0.014 ± 0.0020 0.0062 ± 0.00077 0.014 ± 0.0012 0.0051 ± 0.002031 0.0025 ± 0.00050 0.012 ± 0.0030 0.0014 ± 0.00033 0.010 ± 0.0017 0 ± 032 0.0094 ± 0.0015 0.012 ± 0.0020 0.0055 ± 0.00082 0.017 ± 0.0017 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0.0073 ± 0.0017 0.0071 ± 0.0011 0.0029 ± 0.00031 0.012 ± 0.0014 0 ± 0Total 0.072 ± 0.016 0.10 ± 0.018 0.029 ± 0.0022 0.094 ± 0.0076 0.075 ± 0.031Esters 36 0.00035 ± 0.00010 0.00040 ± 0.00012 0 ± 0 0.00018 ± 8.3E-05 0 ± 038 0.00069 ± 0.00030 0.00064 ± 0.00039 0 ± 0 0.00025 ± 0.00013 0 ± 040 0.00062 ± 0.00017 0.0010 ± 0.00073 0.00018 ± 8.4E-05 0.00061 ± 0.00047 0.0022 ± 0.0003342 0.00050 ± 8.3E-05 0.00077 ± 0.00039 0.00054 ± 0.00024 0.00078 ± 0.00058 0.0017 ± 0.0006944 0.00050 ± 0.00013 0.00071 ± 0.00020 0.00038 ± 0.00011 0.00062 ± 0.00041 0.0012 ± 0.001246 0.00015 ± 5.8E-05 0.0012 ± 0.00039 0 ± 0 0.00079 ± 0.00035 0.00096 ± 0.0005548 0.00047 ± 0.00024 0.0013 ± 0.00052 0 ± 0 0 ± 0 0 ± 050 0 ± 0 0.0016 ± 0.00050 0 ± 0 0 ± 0 0 ± 052 0 ± 0 0.0016 ± 0.00017 0 ± 0 0 ± 0 0 ± 054 0 ± 0 0.0014 ± 0.00038 0 ± 0 0 ± 0 0 ± 0Total 0.0033 ± 0.00091 0.011 ± 0.0025 0.0011 ± 0.00033 0.0032 ± 0.0011 0.0060 ± 0.0017Aldehydes 26 0.0042 ± 0.00091 0.0034 ± 0.00040 0.0031 ± 0.00030 0.0025 ± 0.00056 0 ± 028 0.0067 ± 0.00079 0.0077 ± 0.0018 0.0066 ± 0.00044 0.0054 ± 0.00037 0 ± 030 0.011 ± 0.0015 0.0082 ± 0.00068 0.0063 ± 0.00062 0.0083 ± 0.0010 0 ± 032 0.020 ± 0.0026 0.0041 ± 0.00058 0.012 ± 0.0012 0.0034 ± 0.00031 0 ± 034 0.011 ± 0.0040 0.0044 ± 0.00066 0.0048 ± 0.00071 0.0074 ± 0.0016 0 ± 0Continued on next page218A.Supplementarydataforchapter2Table A.3 – Continued from previous pageCompound Class Cn Col0 Ler cer1 cer2 cer3Total 0.052 ± 0.0082 0.028 ± 0.0023 0.033 ± 0.0024 0.027 ± 0.0025 0 ± 0n-Alkanes 25 0 ± 0 0.0034 ± 0.0024 0 ± 0 0.0026 ± 0.00056 0.0020 ± 0.0001526 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 027 0.0034 ± 0.00082 0.0080 ± 0.0019 0 ± 0 0.0041 ± 0.0029 0.0028 ± 0.001728 0.00031 ± 0.00017 0.00015 ± 4.2E-05 0 ± 0 0.00046 ± 0.00011 0 ± 029 0.062 ± 0.013 0.083 ± 0.014 0.0034 ± 0.00060 0.053 ± 0.0057 0.010 ± 0.007130 0.0032 ± 0.00081 0.0051 ± 0.00096 0 ± 0 0.0061 ± 0.00033 0 ± 031 0.12 ± 0.026 0.15 ± 0.019 0.00036 ± 0.00031 0.11 ± 0.011 0.0020 ± 0.0007032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0.034 ± 0.0063 0.042 ± 0.0047 0 ± 0 0.032 ± 0.0045 0 ± 034 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 035 0.0019 ± 0.00023 0.0031 ± 0.00059 0 ± 0 0.0028 ± 0.00045 0 ± 036 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 037 0.00037 ± 9.4E-05 0.0028 ± 0.00081 0 ± 0 0.0025 ± 0.00037 0 ± 0Total 0.22 ± 0.046 0.30 ± 0.037 0.0038 ± 0.00080 0.21 ± 0.019 0.017 ± 0.0062Ketols 29 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Ketones 29 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 030 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Sec. alcohols 29 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 030 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0iso-Alkanes 27 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 028 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 029 0.0020 ± 0.00059 0.0055 ± 0.0028 0 ± 0 0 ± 0 0 ± 030 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0.0012 ± 0.00029 0.0075 ± 0.0043 0 ± 0 0 ± 0 0 ± 032 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0.0032 ± 0.00085 0.013 ± 0.0069 0 ± 0 0 ± 0 0 ± 0iso-Alcohols 28 0 ± 0 0.0021 ± 0.00014 0.00090 ± 0.00043 0.0013 ± 0.00026 0 ± 029 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 030 0.012 ± 0.0025 0.021 ± 0.0026 0.0049 ± 0.00069 0.014 ± 0.0011 0.00038 ± 0.0005131 0.0011 ± 3.0E-05 0.0046 ± 0.0016 0.0022 ± 0.00044 0.0020 ± 0.00031 0 ± 032 0.026 ± 0.0040 0.028 ± 0.0034 0.015 ± 0.0016 0.048 ± 0.0035 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0.0012 ± 0.00025 0.0012 ± 0.00029 0.0011 ± 0.00013 0.0021 ± 0.00038 0 ± 0Continued on next page219A.Supplementarydataforchapter2Table A.3 – Continued from previous pageCompound Class Cn Col0 Ler cer1 cer2 cer3Total 0.040 ± 0.0066 0.057 ± 0.0071 0.024 ± 0.0028 0.068 ± 0.0050 0.00038 ± 0.00051Total unidentifieds 0.11 ± 0.013 0.20 ± 0.053 0.085 ± 0.015 0.15 ± 0.012 0.23 ± 0.063Total wax 0.58 ± 0.11 0.90 ± 0.088 0.28 ± 0.027 0.73 ± 0.035 0.63 ± 0.14Table A.4: Wax composition on leaves of wild-type and mutant Arabidopsis lines. Amounts are tabulated inµg/cm2 as the mean of five independent samples, n.d. = not detected.Compound Class Cn cer4 cer6 cer16 cer26 kcs1Fatty acids 22 0.00099 ± 0.00040 0.0024 ± 0.00092 0.0018 ± 0.00056 0.00027 ± 0.00014 0.0029 ± 0.001324 0.0073 ± 0.0017 0.068 ± 0.015 0.020 ± 0.0017 0.0058 ± 0.00092 0.0074 ± 0.001126 0.035 ± 0.0073 0.069 ± 0.018 0.060 ± 0.0077 0.029 ± 0.0026 0.019 ± 0.003228 0.012 ± 0.0016 0.0044 ± 0.00067 0.030 ± 0.0027 0.023 ± 0.0016 0.012 ± 0.001430 0.015 ± 0.0053 0.00057 ± 0.00022 0.013 ± 0.0019 0.049 ± 0.0036 0.0066 ± 0.0007332 0.021 ± 0.0021 0 ± 0 0.028 ± 0.0028 0 ± 0 0.0031 ± 0.0004334 0.016 ± 0.0061 0 ± 0 0.013 ± 0.0028 0 ± 0 0.0030 ± 0.0025Total 0.11 ± 0.018 0.14 ± 0.035 0.17 ± 0.013 0.11 ± 0.0078 0.054 ± 0.0049n-Alcohols 22 0 ± 0 0.0037 ± 0.0037 0.0084 ± 0.0028 0.00014 ± 5.7E-05 0 ± 023 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 024 0 ± 0 0.0026 ± 0.00063 0.0066 ± 0.0024 0.0020 ± 0.00060 0.0010 ± 0.0003025 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 026 0.0018 ± 0.00049 0.022 ± 0.0081 0.013 ± 0.0031 0.0038 ± 0.00083 0.030 ± 0.01227 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 028 0.014 ± 0.0021 0.012 ± 0.0021 0.025 ± 0.0026 0.022 ± 0.0038 0.065 ± 0.02029 0.0084 ± 0.0016 0.0018 ± 0.00074 0.0084 ± 0.0012 0.0084 ± 0.0015 0.0054 ± 0.001030 0.018 ± 0.0016 0.0035 ± 0.00027 0.028 ± 0.0059 0.038 ± 0.0034 0.020 ± 0.003631 0.0088 ± 0.0015 0 ± 0 0.018 ± 0.0023 0.00585 ± 0.0011 0.0048 ± 0.001132 0.018 ± 0.0035 0.0025 ± 0.00052 0.018 ± 0.0035 0.0033 ± 0.0003 0.014 ± 0.001633 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0.012 ± 0.0031 0 ± 0 0.010 ± 0.0018 0 ± 0 0.010 ± 0.0013Total 0.080 ± 0.012 0.049 ± 0.015 0.14 ± 0.014 0.084 ± 0.0094 0.15 ± 0.037Esters 36 0 ± 0 0.0017 ± 0.00096 0.00072 ± 0.00013 0.00069 ± 0.00038 0.0017 ± 0.0003938 0.0013 ± 0.00073 0.00060 ± 0.00044 0.00066 ± 0.00036 0.00028 ± 0.00011 0.011 ± 0.006040 0.00067 ± 0.00049 0.0011 ± 0.00041 0.00085 ± 0.00021 0.00081 ± 0.00013 0.0097 ± 0.004842 0.0034 ± 0.0067 0.0010 ± 0.00036 0.0013 ± 0.00044 0.0012 ± 0.0018 0.0071 ± 0.003544 0.00059 ± 0.00023 0.0011 ± 0.00036 0.0008 ± 0.00014 0.00036 ± 0.000079 0.0029 ± 0.001046 0 ± 0 0 ± 0 0.0010 ± 0.00034 0.00031 ± 0.00011 0.0013 ± 0.0003448 0 ± 0 0 ± 0 0.0013 ± 0.00016 0 ± 0 0.0012 ± 0.0007250 0 ± 0 0 ± 0 0.00062 ± 0.00024 0 ± 0 0 ± 0Continued on next page220A.Supplementarydataforchapter2Table A.4 – Continued from previous pageCompound Class Cn cer4 cer6 cer16 cer26 kcs152 0 ± 0 0 ± 0 0.00028 ± 9.6E-05 0 ± 0 0 ± 054 0 ± 0 0 ± 0 0.00022 ± 7.1E-05 0 ± 0 0 ± 0Total 0.0060 ± 0.0081 0.0054 ± 0.0021 0.0076 ± 0.00086 0.0037 ± 0.0017 0.035 ± 0.011Aldehydes 26 0.00096 ± 0.0011 0 ± 0 0.0024 ± 0.00041 0.0018 ± 0.00026 0.0020 ± 0.001128 0.0031 ± 0.0015 0 ± 0 0.024 ± 0.0030 0.012 ± 0.0013 0.0044 ± 0.001130 0.0093 ± 0.0037 0 ± 0 0.010 ± 0.0013 0.077 ± 0.011 0.0051 ± 0.001632 0.0055 ± 0.0014 0 ± 0 0.0087 ± 0.00098 0.0028 ± 0.00035 0.0099 ± 0.0009834 0.0095 ± 0.0042 0 ± 0 0.0050 ± 0.00089 0 ± 0 0.013 ± 0.0042Total 0.028 ± 0.011 0 ± 0 0.051 ± 0.0022 0.093 ± 0.012 0.034 ± 0.0045n-Alkanes 25 0 ± 0 0.0040 ± 0.00067 0.0043 ± 0.0018 0.00077 ± 0.00035 0.00081 ± 0.0005626 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 027 0.0036 ± 0.0011 0.0040 ± 0.0014 0.0093 ± 0.00091 0.0040 ± 0.00059 0.0073 ± 0.0009328 0.00092 ± 0.00067 0 ± 0 0.0012 ± 0.00041 0 ± 0 0.00036 ± 9.8E-0529 0.12 ± 0.023 0.039 ± 0.010 0.12 ± 0.010 0.19 ± 0.018 0.16 ± 0.02530 0.0096 ± 0.0027 0.0024 ± 0.00051 0.0088 ± 0.00060 0.0056 ± 0.00070 0.011 ± 0.001231 0.21 ± 0.041 0.038 ± 0.0088 0.20 ± 0.015 0.047 ± 0.0016 0.26 ± 0.04932 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0.063 ± 0.015 0 ± 0 0.058 ± 0.0064 0 ± 0 0.099 ± 0.02534 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 035 0.0055 ± 0.0015 0 ± 0 0.0047 ± 0.00045 0 ± 0 0.010 ± 0.003636 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 037 0.0023 ± 0.0016 0 ± 0 0.0020 ± 0.00081 0 ± 0 0.0022 ± 0.00074Total 0.41 ± 0.083 0.087 ± 0.020 0.40 ± 0.027 0.25 ± 0.019 0.55 ± 0.097Ketols 29 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Ketones 29 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 030 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Sec. alcohols 29 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 030 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Total 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0iso-Alkanes 27 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 028 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 029 0.00059 ± 0.00047 0.0016 ± 0.00064 0.00347 ± 0.00064 0.00063 ± 0.00029 0.0021 ± 0.0005430 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 031 0.0033 ± 0.0013 0 ± 0 0.000648 ± 0.00016 0.0028 ± 0.0015 0.00093 ± 0.0002332 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0Continued on next page221A.Supplementarydataforchapter2Table A.4 – Continued from previous pageCompound Class Cn cer4 cer6 cer16 cer26 kcs1Total 0.0039 ± 0.0016 0.0016 ± 0.00064 0.004118 ± 0.00079 0.0034 ± 0.0015 0.0030 ± 0.00067iso-Alcohols 28 0 ± 0 0.0031 ± 0.00040 0 ± 0 0.0018 ± 0.00026 0.00079 ± 0.0001729 0 ± 0 0.0013 ± 0.00022 0 ± 0 0 ± 0 0 ± 030 0.027 ± 0.0032 0.0054 ± 0.00064 0.007464 ± 0.0031 0.024 ± 0.0024 0.024 ± 0.002431 0.0027 ± 0.0013 0 ± 0 0 ± 0 0.0011 ± 0.00023 0.0049 ± 0.0005532 0.066 ± 0.014 0.0010 ± 0.00017 0.010056 ± 0.0049 0.036 ± 0.0034 0.054 ± 0.007033 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 034 0.0030 ± 0.0015 0 ± 0 0.00145 ± 0.00053 0.00054 ± 0.00051 0.0071 ± 0.0052Total 0.098 ± 0.019 0.011 ± 0.0012 0.01897 ± 0.0083 0.063 ± 0.0053 0.091 ± 0.0084Total unidentifieds 0.23 ± 0.11 0.13 ± 0.010 0.176124 ± 0.013 0.12 ± 0.010 0.26 ± 0.071Total wax 1.0 ± 0.25 0.48 ± 0.083 1.024848 ± 0.052 0.77 ± 0.037 1.3 ± 0.14222Appendix BSupplementary data for chapter 3Table B.1: Species with specialty wax compounds. Each species is listed along-side its abbreviation used in the specialty compound catalog, the source of the spe-cialty compounds, and the order to which that species or group of species belongs.Species Abbv. Source Plant OrderAgropyron spp. AG cuticular wax PoalesAlbizia lebbeck Al cuticular wax FabalesAllium porrum Ap cuticular wax AsparagalesAloe arborescens Aa cuticular wax AsparagalesAndropogon spp. AN cuticular wax PoalesAnnona cherimolia Ac1 cuticular wax MagnolialesAnnona squamosa As cuticular wax MagnolialesArabidopsis thaliana At cuticular wax BrassicalesAzolla filiculoides Af cuticular wax SalvinialesBauhinia variegata Bv cuticular wax FabalesBerberis aquifolium Ba cuticular wax RanunculalesBerberis jaeschkeana Bj cuticular wax RanunculalesBrassica spp. BR cuticular wax BrassicalesBuxus sempervivens Bs cuticular wax BuxalesCanna indica Ci cuticular wax ZingiberalesCarthamus tinctorius Ct cuticular wax AsteralesContinued on next page223B. Supplementary data for chapter 3Table B.1 – Continued from previous pageSpecies Abbv. Source Plant OrderCerinthe minor Cm1 cuticular wax contestedClematis vitalba Cv cuticular wax RanunculalesCosmos bipinnatus Cb cuticular wax AsteralesCostus speciosus Cs2 cuticular wax ZingiberalesCupressus spp. CU cuticular wax PinalesCynomorium songaricum Cs3 cuticular wax contestedDuboisia myoporoides Dm cuticular wax SolanalesEgagrostis curvula Ec1 cuticular wax PoalesElymus cinereus Ec2 cuticular wax PoalesEncephalartos spp. EN cuticular wax CycadalesEschscholtzia californica Ec3 cuticular wax RanunculalesEucalyptus spp. EU cuticular wax MyrtalesEuonymus latifolius El cuticular wax CelastralesExochorda racemosa Er cuticular wax RosalesFestuca ovina Fo cuticular wax PoalesFoeniculum vulgare Fv cuticular wax ApialesFragaria spp. FR cuticular wax RosalesFumaria parviflora Fp cuticular wax RanunculalesFunaria hygrometrica Fh cuticular wax FunarialesGinkgo biloba Gb cuticular wax GinkgoalesHelianthus annuus Ha cuticular wax AsteralesHibiscus rosasinensis Hr cuticular wax MalvalesHordeum vulgare Hv cuticular wax PoalesHymenocallis littoralis Hl cuticular wax AsparagalesHyoscyamus muticus Hm cuticular wax SolanalesIndigofera aspalathoides Ia cuticular wax FabalesJuniperus scopulorum Js cuticular wax PinalesLaurus nobilis Ln cuticular wax LauralesLeucas aspera La cuticular wax LamialesMalus domestica Md1 cuticular wax RosalesContinued on next page224B. Supplementary data for chapter 3Table B.1 – Continued from previous pageSpecies Abbv. Source Plant OrderMelicocca bijugatus Mb fruit cuticular wax SapindalesMertensia maritima Mm cuticular wax contestedMinuartia recurva Mr cuticular wax CaryophyllalesMiscanthus x giganteus Mg1 cuticular wax PoalesMucuna pruriens Mp cuticular wax FabalesMyricaria germanica Mg2 cuticular wax CaryophyllalesNelumbo nucifera Nn cuticular wax ProtealesNicotiana tabacum Nt cuticular wax SolanalesNigella sativa Ns cuticular wax RanunculalesOsmunda regalis Or cuticular wax OsmundalesPanicum vergatum Pv cuticular wax PoalesPapaver spp PA cuticular wax RanunculalesPegolettia senegalensis Ps1 cuticular wax AsteralesPheonix theophrasti Pt cuticular wax ArecalesPicea abies Pa1 cuticular wax PinalesPicea omorika Po1 cuticular wax PinalesPinus spp. PI2 cuticular wax PinalesPiper spp. PI3 cuticular wax PiperalesPisum sativum Ps2 cuticular wax FabalesPlatanus orientalis Po2 cuticular wax ProtealesPogonatum belangeri Pb cuticular wax PolytrichalesPrunus avium Pa2 cuticular wax RosalesPrunus domestica Pd cuticular wax RosalesPrunus persica Pp fruit cuticular wax RosalesRhododendron spp. RH cuticular wax EricalesRicinus communis Rc cuticular wax MalpighialesRosa spp. RO cuticular wax RosalesSalvia spp. SA cuticular wax LamialesSilybum marianum Sm cuticular wax AsteralesSolanum spp. SO cuticular wax SolanalesContinued on next page225B. Supplementary data for chapter 3Table B.1 – Continued from previous pageSpecies Abbv. Source Plant OrderSorghum bicolor Sb cuticular wax PoalesSyzygium cumini Sc cuticular wax MyrtalesTaxus baccata Tb cuticular wax PinalesTectonca grandis Tg cuticular wax LamialesThalictrum flavum Tf cuticular wax RanunculalesThesium humile Th cuticular wax SantalalesTridax procumbens Tp cuticular wax AsteralesTriticale Tr cuticular wax PoalesTriticum spp. TR cuticular wax PoalesVaccinium ashei Va cuticular wax EricalesArgemone mexicana Am-O flower oil RanunculalesAzotobacter chroococcum Ac2-B bacteria -Bombus terrestris Bo-I insect -Chlamydomonas monoica Cm2-A algae -Chlorella sorokiniana Cs1-A algae -Fannia canicularis Fc-I insect -Habrobracon hebetor Hh-I insect -Malpighia spp. MA-O flower oils -Mayetiola descructor Md2-I insect -Nannochloropsis spp. NC-A algae -Pieris spp. PI1-I insect -Schizymenia dubyi Sd-S seaweed -Wrightia tinctoria Wt-O plant oil -226Appendix CSupplementary data for chapter 4227C. Supplementary data for chapter 4Figure C.1: Relative quantification of methanolysis products obtained fromTLC purification of unknown series A. The band was wide with a darkly stainedfront and more lightly stained tail. The substrate for methanolysis was obtained byscratching out only the front of the band.C20H41OSiOOC20H41397469 750C20H41O OOH C20H41C20H41OSiOHOm/z 441OSiOHOm/z 161C18H37OSiOOC22H45369497 750C18H37OSiOHOm/z 413C16H33OSiOOC24H49341525 750C16H33OSiOHOm/z 385C24 beta-OH acid + C22 alcohol C22 beta-OH acid + C24 alcoholSiH3CC18H37O OOH C22H45SiH3CC16H33O OOH C24H49SiH3CRO OOH C20H41SiH3CC20 beta-OH acid + C26 alcoholFigure C.2: Potential paths to the fragments observed in the mass spectraof C46 β-hydroxy fatty acid esters.228C. Supplementary data for chapter 4Figure C.3: The mass spectrum of docosyl 2-bromoacetate (3.2) and poten-tial sources of observed characteristic fragments.229C. Supplementary data for chapter 4Figure C.4: Mass spectra and possible fragmentation of aliphatic aldehydesA) The mass spectrum and structure of docosanal (3.3). B) The mass spectrum oftetracosanal (3.11). C) Potential paths to observed major fragments.230C. Supplementary data for chapter 4C30 1,5-diol + C16 CoA C30 1,9-diol + C16 CoAC12H25OC25H51O OSiC14H29OOOSiC21H4341345339746973750+2H=257239MW = 76473750+2H=257239MW = 764HO C11H23OSiO C11H23OSim/z 329 m/z 313Figure C.5: Potential paths to the fragments observed in the mass spectraof C46 β-hydroxy fatty acid esters.Figure C.6: Mass spectrum and fragmentation of 2-((6-bromohexyl)oxy)tetrahydro-2H-pyran (3.7) A) The mass spectrum andstructure of 2-((6-bromohexyl)oxy)tetrahydro-2H-pyran (3.7). B) Potential paths toobserved major fragments.231C. Supplementary data for chapter 4Figure C.7: The mass spectrum of tetracosanol (3.10) and potential sourcesof observed major fragments.Figure C.8: Identification of octacosane-1,3-diol in the leafy gametophyteand gametophyte calyptra waxes of F. hygrometrica.232C. Supplementary data for chapter 4Figure C.9: Identification of triacontane-1,7-diol in the sporophyte capsuleof F. hygrometrica. A) EI mass spectrum of the naturally occurring triacontane-1,7-diol. As this compound was found in trace amounts, the mass spectra obtained forthe natural product contains some background signal, present as small mass spectralpeaks that are not in the spectrum of the authentic standard. B) EI mass spectrumof the synthetic triacontane-1,7-diol. C) Potential sources of the major fragmentsobserved in the mass spectrum of the diol.233Appendix DSupplementary data for chapter 5234D.Supplementarydataforchapter5Table D.1: Wax coverage on three F. hygrometrica organs. Amounts are tabulated in µg/cm2 as the mean ofthree independent samples, n.d. = not detected.Compound Class Chain length Gametophyte Calyptra SporophyteFatty acids (FAs) 20 0.0012 ± 0.00010 0.0074 ± 0.0011 0.00030 ± 0.0001222 0.0017 ± 0.00047 0.010 ± 0.0029 0.00026 ± 0.0001224 0.0041 ± 0.0013 0.0027 ± 0.0013 0.00012 ± 0.00006626 0.0043 ± 0.0011 n.d. ± n.d. n.d. ± n.d.Total 0.011 ± 0.0030 0.020 ± 0.0053 0.00068 ± 0.00031Alcohols 22 0.013 ± 0.0012 0.014 ± 0.0025 0.0019 ± 0.0005424 0.0052 ± 0.00064 0.0031 ± 0.00066 0.00061 ± 0.0002126 0.0052 ± 0.00073 0.0018 ± 0.00019 0.00081 ± 0.0006328 0.0068 ± 0.00097 0.0081 ± 0.00071 0.0023 ± 0.0006930 0.0041 ± 0.00010 0.014 ± 0.00057 0.0049 ± 0.0026Total 0.034 ± 0.0036 0.041 ± 0.0046 0.011 ± 0.0047Diols 28 (1,3) 0.0025 ± 0.00082 0.0014 ± 0.00014 n.d. ± n.d.30 (1,7) n.d. ± n.d. n.d. ± n.d. 0.0020 ± 0.0011Alkyl esters 38 0.010 ± 0.0022 0.039 ± 0.0033 0.0067 ± 0.0007139 0.010 ± 0.0014 0.0063 ± 0.00044 0.0024 ± 0.001040 0.028 ± 0.0044 0.13 ± 0.0069 0.0088 ± 0.002641 0.0070 ± 0.00038 0.018 ± 0.00055 0.0012 ± 0.0001842 0.073 ± 0.016 0.34 ± 0.016 0.016 ± 0.001443 0.017 ± 0.0036 0.041 ± 0.0015 0.002 ± 0.00063Continued on next page235D.Supplementarydataforchapter5Table D.1 – Continued from previous pageCompound Class Chain length Gametophyte Calyptra Sporophyte44 0.12 ± 0.022 0.36 ± 0.019 0.024 ± 0.004445 0.013 ± 0.0025 0.030 ± 0.0031 0.0055 ± 0.002046 0.11 ± 0.018 0.13 ± 0.0095 0.040 ± 0.008747 0.010 ± 0.0016 0.027 ± 0.0030 0.0043 ± 0.002148 0.12 ± 0.017 0.13 ± 0.0082 0.018 ± 0.001249 0.011 ± 0.0027 0.022 ± 0.0031 0.0039 ± 0.0005850 0.074 ± 0.011 0.22 ± 0.024 0.019 ± 0.002351 0.0072 ± 0.00054 0.022 ± 0.0020 0.0036 ± 0.00352 0.019 ± 0.0029 0.17 ± 0.026 0.019 ± 0.0079Total 0.63 ± 0.11 1.7 ± 0.13 0.17 ± 0.039Esterified Diols 40 n.d. ± n.d. n.d. ± n.d. 0.0022 ± 0.001542 n.d. ± n.d. n.d. ± n.d. 0.018 ± 0.01644 n.d. ± n.d. n.d. ± n.d. 0.014 ± 0.008146 n.d. ± n.d. n.d. ± n.d. 0.028 ± 0.01448 n.d. ± n.d. n.d. ± n.d. 0.023 ± 0.007950 n.d. ± n.d. n.d. ± n.d. 0.028 ± 0.0094Total 0 ± 0 0 ± 0 0.11 ± 0.057β-OH FA esters 38 0.0054 ± 0.00065 0.0017 ± 0.00046 n.d. ± n.d.40 0.0063 ± 0.0013 0.0075 ± 0.00062 n.d. ± n.d.42 0.033 ± 0.0093 0.022 ± 0.00078 n.d. ± n.d.44 0.036 ± 0.0084 0.021 ± 0.0031 n.d. ± n.d.46 0.032 ± 0.0051 0.0075 ± 0.0029 n.d. ± n.d.48 0.026 ± 0.0057 0.0098 ± 0.0012 n.d. ± n.d.Continued on next page236D.Supplementarydataforchapter5Table D.1 – Continued from previous pageCompound Class Chain length Gametophyte Calyptra Sporophyte50 0.011 ± 0.00082 0.0091 ± 0.0021 n.d. ± n.d.Total 0.15 ± 0.031 0.079 ± 0.011 0 ± 0Aldehydes 28 n.d. ± n.d. n.d. ± n.d. 0.00033 ± 0.0002130 n.d. ± n.d. n.d. ± n.d. 0.0019 ± 0.0013Total 0 ± 0 0 ± 0 0.0022 ± 0.0015Alkanes 27 n.d. ± n.d. n.d. ± n.d. 0.0047 ± 0.002629 n.d. ± n.d. 0.0053 ± 0.0009 0.022 ± 0.013Total 0 ± 0 0.0053 ± 0.0009 0.027 ± 0.016Unidentified Wax 0.11 ± 0.0086 0.17 ± 0.028 0.097 ± 0.025Total Wax 0.94 ± 0.13 2.0 ± 0.15 0.44 ± 0.10Table D.2: Amount of esterified fatty acids in each alkyl ester homolog on the leafy gametophyte of F.hygrometrica. The carbon numbers of esterified fatty acids are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC160.0031 0.0011 0.0015 0.001 0.0007 0.00034 0.00011 0.0080± ± ± ± ± ± ± ±0.00080 0.00016 0.00034 0.00027 0.00030 0.00012 0.000032 0.0012Continued on next page237D.Supplementarydataforchapter5Table D.2 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC180.00076 0.0028 0.0028 0.0014 0.00068 0.00073 0.00034 0.000087 0.0093± ± ± ± ± ± ± ± ±0.00057 0.00090 0.00074 0.00027 0.00015 0.00019 0.00018 0.000028 0.0023C200.0019 0.0098 0.026 0.012 0.0056 0.0027 0.00065 0.00041 0.00010 0.060± ± ± ± ± ± ± ± ± ±0.00020 0.0014 0.0061 0.0024 0.0012 0.00016 0.0005 0.00010 0.00004 0.011C220.000833 0.0036 0.040 0.017 0.013 0.0024 0.00060 0.00021 0.080± ± ± ± ± ± ± ± ±0.00034 0.00053 0.0079 0.0021 0.0014 0.00032 0.00019 0.000036 0.013C240.0033 0.0053 0.033 0.035 0.026 0.0025 0.00040 0.11± ± ± ± ± ± ± ±0.00038 0.000414 0.0050 0.0049 0.0042 0.00039 0.00012 0.015C260.0011 0.0015 0.0097 0.0069 0.0058 0.00050 0.025± ± ± ± ± ± ±0.000048 0.00019 0.0015 0.00096 0.00075 0.00010 0.0034C280.00064 0.0010 0.0014 0.00040 0.00024 0.0037± ± ± ± ± ±0.00035 0.00070 0.00035 0.000091 0.000046 0.0015238D.Supplementarydataforchapter5Table D.3: Amount of esterified alcohols in each alkyl ester homolog on the leafy gametophyte of F.hygrometrica. The carbon numbers of esterified alcohols are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC180.0016 0.00066 0.0024 0.00074 0.00041 0.0051± ± ± ± ± ±0.00018 0.00027 0.00028 0.000033 0.00023 0.0016C200.00080 0.0094 0.0031 0.0043 0.0011 0.00073 0.019± ± ± ± ± ± ±0.00060 0.0013 0.00046 0.00034 0.00014 0.00049 0.0033C220.0040 0.0032 0.027 0.038 0.030 0.0080 0.0011 0.11± ± ± ± ± ± ± ±0.0010 0.0010 0.0064 0.0076 0.0044 0.0012 0.00027 0.020C240.0016 0.0035 0.014 0.017 0.033 0.0062 0.00034 0.076± ± ± ± ± ± ± ±0.00023 0.00093 0.0027 0.0022 0.0047 0.00086 0.000076 0.012C260.0023 0.0019 0.0069 0.015 0.027 0.0056 0.00020 0.058± ± ± ± ± ± ± ±0.00051 0.00036 0.0015 0.0015 0.0043 0.00073 0.000042 0.0087C280.0016 0.00099 0.0035 0.0029 0.0028 0.00048 0.012Continued on next page239D.Supplementarydataforchapter5Table D.3 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester Total± ± ± ± ± ± ±0.00044 0.00022 0.00021 0.00038 0.00043 0.00011 0.0016C300.0013 0.0011 0.00091 0.00078 0.00051 0.0046± ± ± ± ± ±0.00050 0.00029 0.00070 0.00024 0.00014 0.00074C320.00062 0.00056 0.00061 0.00028 0.0021± ± ± ± ±0.00021 0.00029 0.00014 0.000050 0.00062C340.00021 0.00015 0.00015 0.00040± ± ± ±0.000062 0.000048 0.000060 0.00020Table D.4: Amount of esterified fatty acids in each alkyl ester homolog on the gametophyte calyptra of F.hygrometrica. The carbon numbers of esterified fatty acids are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC160.01029 0.00292 0.00185 0.00353 0.00726 0.02586± ± ± ± ± ±0.00092 0.00009 0.00011 0.00008 0.00053 0.00145Continued on next page240D.Supplementarydataforchapter5Table D.4 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC180.00456 0.03331 0.00476 0.00231 0.00704 0.01115 0.06314± ± ± ± ± ± ±0.00055 0.00185 0.00029 0.00013 0.00098 0.00029 0.00377C200.00409 0.01905 0.13709 0.01549 0.00833 0.03418 0.05545 0.27369± ± ± ± ± ± ± ±0.00014 0.00104 0.00600 0.00080 0.00061 0.00268 0.00582 0.01529C220.00706 0.02182 0.16005 0.02006 0.00914 0.03838 0.07298 0.32950± ± ± ± ± ± ± ±0.00046 0.00160 0.00869 0.00101 0.00085 0.00459 0.01078 0.02725C240.00100 0.00475 0.00637 0.02044 0.00327 0.00211 0.00450 0.01003 0.05247± ± ± ± ± ± ± ± ±0.00011 0.00021 0.00029 0.00157 0.00029 0.00021 0.00085 0.00163 0.00166Table D.5: Amount of esterified alcohols in each alkyl ester homolog on the gametophyte calyptra of F.hygrometrica. The carbon numbers of esterified alcohols are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC160.00066 0.00066± ±0.00008 0.00008Continued on next page241D.Supplementarydataforchapter5Table D.5 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC180.00354 0.00561 0.00349 0.01264± ± ± ±0.00012 0.00036 0.00015 0.00059C200.00478 0.01820 0.01912 0.00515 0.04726± ± ± ± ±0.00057 0.00099 0.00140 0.00023 0.00316C220.01311 0.03824 0.14324 0.15346 0.01811 0.36616± ± ± ± ± ±0.00117 0.00212 0.00627 0.00833 0.00139 0.01829C240.00404 0.00594 0.01758 0.02089 0.00314 0.05158± ± ± ± ± ±0.00012 0.00036 0.00090 0.00105 0.00028 0.00231C260.00277 0.00311 0.01020 0.01027 0.00219 0.02853± ± ± ± ± ±0.00016 0.00018 0.00075 0.00096 0.00022 0.00207C280.00566 0.01016 0.04491 0.04628 0.00502 0.11204± ± ± ± ± ±0.00014 0.00142 0.00352 0.00554 0.00094 0.01107Continued on next page242D.Supplementarydataforchapter5Table D.5 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester C54 ester TotalC300.01242 0.01719 0.07784 0.09401 0.01194 0.21340± ± ± ± ± ±0.00090 0.00045 0.00817 0.01389 0.00194 0.02123243D.Supplementarydataforchapter5Table D.6: Amount of esterified fatty acids in each alkyl ester homolog on the sporophyte capsule of F.hygrometrica. The carbon numbers of esterified fatty acids are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C36 ester C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC140.00029 0.00021 0.00014 0.00034 0.00132 0.00012 0.00007 0.00248± ± ± ± ± ± ± ±0.00017 0.00013 0.00005 0.00005 0.00045 0.00004 0.00001 0.00057C160.00021 0.00251 0.00239 0.00264 0.00229 0.00939 0.00105 0.00008 0.00001 0.02056± ± ± ± ± ± ± ± ± ±0.00012 0.00021 0.00069 0.00031 0.00077 0.00127 0.00049 0.00010 0.00001 0.00062C180.00017 0.00011 0.00055 0.00042 0.00018 0.00032 0.00168 0.00013 0.00003 0.00358± ± ± ± ± ± ± ± ± ±0.00004 0.00001 0.00022 0.00005 0.00005 0.00012 0.00032 0.00008 0.00004 0.00042C200.00015 0.00026 0.00224 0.00071 0.00035 0.00100 0.00496 0.00032 0.00998± ± ± ± ± ± ± ± ±0.00007 0.00012 0.00072 0.00028 0.00011 0.00012 0.00063 0.00026 0.00048C220.00005 0.00053 0.00076 0.00551 0.00218 0.00094 0.00146 0.00608 0.01750± ± ± ± ± ± ± ± ±0.00004 0.00019 0.00024 0.00010 0.00114 0.00003 0.00013 0.00230 0.00329C240.00004 0.00008 0.00119 0.00071 0.00478 0.00266 0.00123 0.00135 0.01203Continued on next page244D.Supplementarydataforchapter5Table D.6 – Continued from previous pageC36 ester C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total± ± ± ± ± ± ± ± ±0.00000 0.00005 0.00100 0.00021 0.00194 0.00101 0.00048 0.00121 0.00556C260.00009 0.00007 0.00004 0.00016 0.00011 0.00053 0.00023 0.00017 0.00137± ± ± ± ± ± ± ± ±0.00002 0.00005 0.00002 0.00006 0.00006 0.00011 0.00013 0.00014 0.00019C280.00004 0.00007 0.00003 0.00007 0.00007 0.00034 0.00006 0.00069± ± ± ± ± ± ± ±0.00003 0.00004 0.00001 0.00002 0.00003 0.00012 0.00003 0.00017C300.00006 0.00003 0.00023 0.00021 0.00084 0.00137± ± ± ± ± ±0.00001 0.00002 0.00014 0.00013 0.00036 0.00061245D.Supplementarydataforchapter5Table D.7: Amount of esterified alcohols in each alkyl ester homolog on the sporophyte capsule of F.hygrometrica. The carbon numbers of esterified alcohols are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C36 ester C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC120.000041 0.000019 0.00005± ± ±0.00001 0.00002 0.00004C140.00002 0.00004 0.00004 0.00003 0.00012± ± ± ± ±0.00000 0.00003 0.00002 0.00001 0.00005C160.00004 0.00005 0.00003 0.00002 0.00002 0.00014± ± ± ± ± ±0.00003 0.00004 0.00001 0.00001 0.00001 0.00005C180.00017 0.00013 0.00042 0.00088 0.00011 0.00005 0.00014 0.00189± ± ± ± ± ± ± ±0.00004 0.00006 0.00015 0.00073 0.00004 0.00001 0.00008 0.00067C200.00025 0.00011 0.00024 0.00067 0.00057 0.00008 0.00005 0.00014 0.00212± ± ± ± ± ± ± ± ±0.00014 0.00001 0.00012 0.00021 0.00017 0.00005 0.00002 0.00008 0.00036C220.00041 0.00320 0.00063 0.00234 0.00529 0.00423 0.00043 0.00026 0.00061 0.01740Continued on next page246D.Supplementarydataforchapter5Table D.7 – Continued from previous pageC36 ester C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total± ± ± ± ± ± ± ± ± ±0.00024 0.00027 0.00025 0.00075 0.00010 0.00172 0.00009 0.00009 0.00026 0.00071C240.00032 0.00331 0.00052 0.00080 0.00227 0.00256 0.00021 0.00005 0.01005± ± ± ± ± ± ± ± ±0.00021 0.00096 0.00006 0.00032 0.00119 0.00097 0.00012 0.00002 0.00315C260.00024 0.00394 0.00025 0.00043 0.00105 0.00128 0.00017 0.00736± ± ± ± ± ± ± ±0.00009 0.00046 0.00007 0.00013 0.00003 0.00050 0.00013 0.00039C280.00062 0.00367 0.00046 0.00132 0.00176 0.00150 0.00933± ± ± ± ± ± ±0.00009 0.00124 0.00017 0.00016 0.00015 0.00134 0.00277C300.00254 0.01607 0.00259 0.00696 0.00783 0.03600± ± ± ± ± ±0.00087 0.00218 0.00050 0.00089 0.00297 0.00599C320.00025 0.00191 0.00021 0.00048 0.00285± ± ± ± ±0.00009 0.00089 0.00012 0.00039 0.00032C340.00015 0.00016 0.00005 0.00024± ± ± ±Continued on next page247D.Supplementarydataforchapter5Table D.7 – Continued from previous pageC36 ester C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total0.00002 0.00019 0.00008 0.00029Table D.8: Amount of esterified fatty acids in each β-hydroxy ester homolog on the leafy gametophyte ofF. hygrometrica. The carbon numbers of esterified β-hydroxy fatty acids are indicated in the top row, and the carbonnumbers of the esters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean ofthree independent samples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC160.00051 0.00043 0.00030 0.00013 0.00018 0.000065 0.0016± ± ± ± ± ± ±0.00040 0.00009 0.00008 0.00006 0.00004 0.00002 0.00032C180.00036 0.00072 0.00033 0.00021 0.00008 0.00169± ± ± ± ± ±0.00011 0.00040 0.00009 0.00002 0.00007 0.00050C200.00088 0.00922 0.00247 0.00117 0.00079 0.00025 0.00010 0.01488± ± ± ± ± ± ± ±0.00033 0.00335 0.00060 0.00018 0.00032 0.00014 0.00003 0.00426C220.00167 0.00359 0.01064 0.00456 0.00306 0.00100 0.00028 0.02480± ± ± ± ± ± ± ±0.00019 0.00090 0.00283 0.00039 0.00086 0.00013 0.00015 0.00446C240.00041 0.00260 0.00214 0.00707 0.00646 0.00291 0.00074 0.02234Continued on next page248D.Supplementarydataforchapter5Table D.8 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total± ± ± ± ± ± ± ±0.00020 0.00067 0.00034 0.00078 0.00151 0.00030 0.00022 0.00308C260.00069 0.00148 0.00078 0.00093 0.00082 0.00066 0.00038 0.00574± ± ± ± ± ± ± ±0.00020 0.00074 0.00037 0.00040 0.00067 0.00012 0.00027 0.00184C280.00225 0.00163 0.00028 0.00416± ± ± ±0.00083 0.00087 0.00019 0.00187C300.00039 0.00075 0.00137 0.00251± ± ± ±0.00012 0.00023 0.00032 0.00066Table D.9: Amount of esterified alcohols in each β-hydroxy ester homolog on the leafy gametophyte of F.hygrometrica. The carbon numbers of esterified alcohols are indicated in the top row, and the carbon numbers of theesters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of three independentsamples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC140.00038 0.00018 0.00056± ± ±0.00011 0.00006 0.00017Continued on next page249D.Supplementarydataforchapter5Table D.9 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC160.00027 0.00090 0.00128 0.00040 0.00286± ± ± ± ±0.00013 0.00045 0.00047 0.00012 0.00103C180.00132 0.00190 0.00053 0.00104 0.00082 0.00561± ± ± ± ± ±0.00015 0.00049 0.00026 0.00056 0.00019 0.00153C200.00084 0.00314 0.00173 0.00070 0.00020 0.00662± ± ± ± ± ±0.00032 0.00078 0.00028 0.00030 0.00013 0.00152C220.00963 0.01020 0.00627 0.00067 0.02677± ± ± ± ±0.00350 0.00271 0.00069 0.00055 0.00651C240.00280 0.00475 0.00621 0.00059 0.01435± ± ± ± ±0.00068 0.00041 0.00145 0.00011 0.00239C260.00077 0.00048 0.00143 0.00343 0.00302 0.00036 0.00949± ± ± ± ± ± ±0.00060 0.00015 0.00022 0.00097 0.00031 0.00026 0.00200C280.00069 0.00103 0.00104 0.00121 0.00083 0.00480Continued on next page250D.Supplementarydataforchapter5Table D.9 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total± ± ± ± ± ±0.00014 0.00057 0.00042 0.00016 0.00025 0.00077C300.00051 0.00051 0.00036 0.00037 0.00175± ± ± ± ±0.00014 0.00014 0.00019 0.00020 0.00043C320.00024 0.00034 0.00015 0.00073± ± ± ±0.00011 0.00004 0.00005 0.00019C340.00035 0.00014 0.00049± ± ±0.00007 0.00012 0.00017Table D.10: Amount of esterified fatty acids in each β-hydroxy ester homolog on the gametophyte calyptraof F. hygrometrica. The carbon numbers of esterified β-hydroxy fatty acids are indicated in the top row, and thecarbon numbers of the esters in which they were found in the left column. Amounts are tabulated in µg/cm2 as themean of three independent samples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC160.00025 0.00027 0.00044 0.00023 0.00002 0.00122± ± ± ± ± ±0.00019 0.00018 0.00024 0.00009 0.00001 0.00040Continued on next page251D.Supplementarydataforchapter5Table D.10 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC180.00134 0.00251 0.00160 0.00037 0.00165 0.00070 0.00010 0.00827± ± ± ± ± ± ± ±0.00095 0.00020 0.00047 0.00027 0.00031 0.00050 0.00007 0.00060C200.00180 0.00551 0.00250 0.00101 0.00095 0.00242 0.00138 0.01556± ± ± ± ± ± ± ±0.00069 0.00075 0.00061 0.00024 0.00014 0.00041 0.00050 0.00119C220.00052 0.00132 0.00402 0.00086 0.00038 0.00038 0.00100 0.00849± ± ± ± ± ± ± ±0.00053 0.00058 0.00164 0.00039 0.00022 0.00011 0.00038 0.00236C240.00023 0.00069 0.00160 0.00071 0.00093 0.00018 0.00005 0.00440± ± ± ± ± ± ± ±0.00017 0.00035 0.00056 0.00077 0.00034 0.00006 0.00004 0.00141C260.00038 0.00052 0.00006 0.00014 0.00016 0.00005 0.00130± ± ± ± ± ± ±0.00016 0.00047 0.00002 0.00014 0.00013 0.00005 0.00063C280.00039 0.00040 0.00007 0.00009 0.00002 0.00097± ± ± ± ± ±0.00013 0.00006 0.00010 0.00007 0.00002 0.00009C300.00004 0.00004Continued on next page252D.Supplementarydataforchapter5Table D.10 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total± ±0.00002 0.00002Table D.11: Amount of esterified alcohols in each β-hydroxy ester homolog on the gametophyte calyptraof F. hygrometrica. The carbon numbers of esterified alcohols are indicated in the top row, and the carbon numbersof the esters in which they were found in the left column. Amounts are tabulated in µg/cm2 as the mean of threeindependent samples.C38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC140.00020 0.00020± ±0.00006 0.00006C160.00015 0.00023 0.00023 0.00002 0.00063± ± ± ± ±0.00011 0.00010 0.00004 0.00001 0.00000C180.00041 0.00050 0.00035 0.00005 0.00131± ± ± ± ±0.00042 0.00026 0.00032 0.00006 0.00098C200.00172 0.00116 0.00129 0.00005 0.00006 0.00428± ± ± ± ± ±0.00066 0.00051 0.00045 0.00001 0.00005 0.00160Continued on next page253D.Supplementarydataforchapter5Table D.11 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester TotalC220.00153 0.00576 0.00386 0.00063 0.00011 0.00002 0.01191± ± ± ± ± ± ±0.00109 0.00078 0.00157 0.00068 0.00012 0.00001 0.00268C240.00313 0.00283 0.00089 0.00090 0.00014 0.00790± ± ± ± ± ±0.00026 0.00069 0.00041 0.00033 0.00011 0.00058C260.00037 0.00215 0.00124 0.00043 0.00019 0.00005 0.00442± ± ± ± ± ± ±0.00029 0.00063 0.00029 0.00024 0.00006 0.00004 0.00104C280.00044 0.00054 0.00125 0.00046 0.00006 0.00274± ± ± ± ± ±0.00030 0.00039 0.00018 0.00014 0.00004 0.00054C300.00076 0.00254 0.00339 0.00129 0.00798± ± ± ± ±0.00041 0.00048 0.00057 0.00049 0.00069C320.00043 0.00115 0.00206 0.00364± ± ± ±0.00016 0.00082 0.00075 0.00138C340.00004 0.00018 0.00021Continued on next page254D.Supplementarydataforchapter5Table D.11 – Continued from previous pageC38 ester C40 ester C42 ester C44 ester C46 ester C48 ester C50 ester C52 ester Total± ± ±0.00001 0.00013 0.00013255Appendix ESupplementary data for chapter 6Plant Leaf	8 Petiole Blade20 5 1 4 74 ± 3 6675 ± 734 20521 ± 1259 0.20 ± 0.02 487 ± 3022 7 2 5 75 ± 2 7502 ± 1339 22064 ± 2297 0.36 ± 0.02 809 ± 8424 9 4 7 77 ± 6 7549 ± 1176 25106 ± 2028 0.80 ± 0.06 1586 ± 12826 11 6 9 75 ± 3 8283 ± 1023 25180 ± 2911 0.98 ± 0.08 1958 ± 22628 13 7 12 78 ± 6 10070 ± 1214 29097 ± 3173 1.6 ± 0.1 2808 ± 30630 15 10 13 75 ± 4 9892 ± 858 30013 ± 2220 2.1 ± 0.06 3456 ± 25632 17 11 14 75 ± 5 10411 ± 994 30660 ± 1821 2.4 ± 0.1 3914 ± 23236 21 14 16 72 ± 6 10366 ± 1156 29602 ± 2681 2.8 ± 0.08 4646 ± 42140 25 17 18 70 ± 8 9885 ± 1242 30446 ± 2303 3.2 ± 0.1 5230 ± 3960.004 ## 0.00040.0072 ## 0.00040.016 ## 0.00120.0196 ## 0.00160.0328 ## 0.0020.0416 ## 0.00120.048 ## 0.00240.0552 ## 0.00160.0636 ## 0.002original	(mm2/blade)10 ± 118 ± 140 ± 349 ± 482 ± 5104 ± 3120 ± 6138 ± 4159 ± 5Surface	areaLeaf	(cm2) Pavement	cell	(μm2cell-1)	Age	(days) Length	(mm) Number	of	cells	(on	one	side	of	the	leaf)Trichome Guard	(pairs) PavementFigure E.1: Morphological data for developing wild-type Arabidopsis eighthleaves. Values are means ± standard deviation from five independent leaves. Leafsurface area is the macroscopic surface area of both sides of the blade, ignoring con-tributions from trichomes.256E. Supplementary data for chapter 65 days / 4 mm9 days / 7 mm13 days / 12 mm17 days / 14 mm21 days / 16 mmleaf base leaf tip1 mm1.75 mm2 mm2.3 mm2.6 mmFigure E.2: Sampling scheme for studying cell size distributions across sec-tions of Arabidopsis wild-type leaves of different ages. Leaf age (days) andblade length (mm) are shown at the left side of each oval. Dashed lines delineatesections of equal length for cell size counting. The length of each section along thelongitudinal leaf axis is denoted underneath each corresponding double-headed arrow.257E. Supplementary data for chapter 6*	(calculated)	from	Table	S1	or	Fig.	S2measured	values	=	green **	from	Tables	S2	and	Fig.	S2calculated	values	=	white ***	From	Suh	et	al.,	2005"answers"	=	yellow ****	Coverage	=	WT	leaf	(micro)	coverage	-	gl1	leaf	coverageday WT	leaf	(micro) gl1	leaf day WT	stem	top*** WT	stem	middle*** day Formulassurface	area	(cm^2)* 5 0.23 0.20 0 0.00 0.00 T1 Asurface	area	(cm^2)* 21 2.83 2.80 1 0.60 0.08 T2 Bwax	coverage	(ug/cm2)** 0.90 0.60 30.00 30.00 CNEW	surface	area	(cm^2) 2.60 2.60 0.60 0.08 =B-A=DNEW	wax	(ug) 2.34 1.56 18.00 2.40 =D*C=Ewax	production	RATE	(ug/day) 0.15 0.10 1.13 0.15 =E/(T2-T1)=Fwax	production	RATE	(ug/hr) 0.006 0.004 0.05 0.006 =F/24=Gday day day Formulassurface	area	(cm^2)* 0 13 T1 Asurface	area	(cm^2)* 1.5 21 T2 Bwax	coverage	(ug/cm2)** CNEW	surface	area	(cm^2) =B-A=DNEW	wax	(ug) =D*C=Ewax	production	RATE	(ug/day) =E/(T2-T1)=Fwax	production	RATE	(ug/hr) =F/24=Gnumber	of	cells** Hwax	production	RATE	per	cell	(pg/hr/cell) =G/H75 4000017.8 0.07*****	assuming	guard	cell	pair	surface	area	=	0.25	x	pavement	cell	area,	and	mature	WT	coverage	=	pavement	+	guard	cell	coverage0.03200 0.067500.00133 0.002811.60 0.900.03 0.600.04800 0.54000Leaf	trichomes**** Leaf	pavement	and	guard	cells*****0.00 1.780.03 2.99Figure E.3: Calculation of wax production rates. Wax accumulation was calcu-lated for expanding organs with constant wax coverage by determining the amountof new surface area created in a particular time frame then multiplying by the corre-sponding coverage and dividing by the time over which the expansion occurred (seeFormulas columns). Cell wax production rates were calculated in the same way exceptthat the number of cells was further used to normalize the accumulation rate.258E.Supplementarydataforchapter6Table E.1: Wax composition on gl1 eighth leaves. The coverage of each compound is presented in units of µg/cm2.days of age -> 5 9 13 17 21Acids 20 0.0032 ± 0.0010 0.0022 ± 0.00070 0.0035 ± 0.0017 0.0044 ± 0.0013 0.0000 ± 0.000022 0.0030 ± 0.00085 0.0021 ± 0.00079 0.0030 ± 0.0011 0.0006 ± 0.00039 0.0013 ± 0.0001924 0.081 ± 0.013 0.060 ± 0.011 0.046 ± 0.0041 0.040 ± 0.012 0.020 ± 0.001326 0.11 ± 0.017 0.12 ± 0.011 0.096 ± 0.011 0.044 ± 0.0085 0.083 ± 0.004928 0.0082 ± 0.0050 0.0079 ± 0.0018 0.012 ± 0.0045 0.014 ± 0.0032 0.011 ± 0.001230 0.0025 ± 0.0010 0.0027 ± 0.0010 0.0029 ± 0.0010 0.0058 ± 0.0015 0.0028 ± 0.001232 0.0029 ± 0.0020 0.0010 ± 0.00040 0.0020 ± 0.00062 0.0038 ± 0.0016 0.0018 ± 0.00074Alcohols 22 0.0014 ± 0.00028 0.0011 ± 0.00032 0.0016 ± 0.00051 0.0050 ± 0.0020 0.0000 ± 0.000026 0.0014 ± 0.0006 0.0013 ± 0.00036 0.0034 ± 0.00071 0.0077 ± 0.0012 0.0032 ± 0.0004628 0.0057 ± 0.0012 0.0089 ± 0.00088 0.016 ± 0.0015 0.021 ± 0.0029 0.013 ± 0.0