UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Secretion of plant cell wall polysaccharides by the Golgi apparatus in Arabidopsis thaliana seed coat… Young, Robin Elizabeth 2009

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

Item Metadata

Download

Media
24-ubc_2009_fall_young_robin.pdf [ 10.79MB ]
Metadata
JSON: 24-1.0067318.json
JSON-LD: 24-1.0067318-ld.json
RDF/XML (Pretty): 24-1.0067318-rdf.xml
RDF/JSON: 24-1.0067318-rdf.json
Turtle: 24-1.0067318-turtle.txt
N-Triples: 24-1.0067318-rdf-ntriples.txt
Original Record: 24-1.0067318-source.json
Full Text
24-1.0067318-fulltext.txt
Citation
24-1.0067318.ris

Full Text

SECRETION OF PLANT CELL WALL POLYSACCHARIDES BY THE GOLGI APPARATUS IN ARABIDOPSIS THALIANA SEED COAT CELLS  by  Robin Elizabeth Young M.Sc. Université de Montréal, 2003 B.Sc. McMaster University, 2000  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in  The Faculty of Graduate Studies (Botany)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2009   Robin Elizabeth Young, 2009  ABSTRACT The plant cell wall determines cell shape and is essential for plant growth during development. Pectin is an important component of cell walls and, like many wall polysaccharides, is synthesized in the Golgi apparatus and secreted by vesicles. In Arabidopsis thaliana seed coats, pectin-rich mucilage is secreted in a polarized fashion during a specific stage of development. How the Golgi apparatus in seed coat cells accommodates the large increase in pectin-rich mucilage provides a unique window into the cellular machinery that supports cell wall polysaccharide biosynthesis and secretion. Examination of seed coat cells, using cryo-fixation and transmission electron microscopy and electron tomography, showed that Golgi stacks undergo dramatic changes in structure during mucilage production. Initiation of mucilage biosynthesis also correlated with increased numbers of Golgi stacks per cell. To understand if these cellular changes were dependent on pectin biosynthesis, the cell structure of a reduced mucilage mutant, mum4, was studied by similar methods and revealed that, while the morphology of Golgi stacks was dependant on mucilage, the increased stack number was not. To determine what proportion of the scattered Golgi stacks were producing mucilage, immunogold labeling with the novel mucilage-specific antibody CCRC-M36 was used to detect the pectin cargo. The large percentage of labeled Golgi stacks found suggests that many stacks produce pectin synchronously, rather than a subset of specialist Golgi. To test if a pectin modifying enzyme, MUM2, is co-secreted with pectin, a tagged MUM2 was engineered and introduced into mum2 mutants, where it rescued the mutant phenotype. However, the tag was not detectable using antibodies in immunofluorescence. Although mucilage was secreted to the top of the cell, antibody label demonstrated that pectin-producing stacks were randomly distributed throughout the cytoplasm, indicating that the ii  destination of cargo has little effect on location of the Golgi stack producing it. The mechanism of targeting of vesicles with the domain of the plasma membrane exclusively at the mucilage pocket is unknown, although the correlation of a population of densely staining vesicles and abundant cortical microtubules in the cell cortex at the site of secretion was documented.  iii  TABLE OF CONTENTS ABSTRACT.............................................................................................................................. ii TABLE OF CONTENTS ........................................................................................................ iv LIST OF TABLES ................................................................................................................. vii LIST OF FIGURES............................................................................................................... viii LIST OF ABBREVIATIONS .................................................................................................. x ACKNOWLEDGEMENTS .................................................................................................. xiii 1  GENERAL INTRODUCTION .......................................................................................... 1 1.1 THE PLANT CELL WALL ................................................................................................... 1 1.1.1 Structure and synthesis of the plant cell wall ........................................................... 1 1.1.2 Studying polysaccharides within the context of the cell wall ................................... 5 1.2 THE SECRETION OF MATRIX POLYSACCHARIDES ............................................................... 7 1.2.1 The Golgi apparatus as a major site of polysaccharide production ........................... 7 1.2.2 The structure and function of the plant Golgi apparatus ........................................... 8 1.2.3 Post-Golgi traffic................................................................................................... 10 1.2.4 Vesicle targeting and the cytoskeleton................................................................... 13 1.3 POLARIZED SECRETION IN PLANTS ................................................................................. 15 1.4 THE ARABIDOPSIS SEED COAT ....................................................................................... 18 1.4.1 Polysaccharide components of Arabidopsis seed coat mucilage............................. 19 1.5 OBJECTIVES.................................................................................................................. 21 1.5.1 Question 1: How does the Golgi apparatus change as the level of matrix polysaccharide biosynthesis and secretion changes?.......................................................... 22 1.5.2 Questions 2: Are there individual Golgi stacks in the cell dedicated to producing specific matrix polysaccharides?....................................................................................... 23 1.5.3 Question 3: How does the plant cell ensure that carbohydrates produced in the Golgi apparatus arrive at a specific cell wall domain? Is the mobility of the Golgi stacks a mechanism of plant secretion? .......................................................................................... 23 1.5.4 Question 5: How does the distinct cytoskeletal arrangement observed in the seed coat epidermal cells affect secretion? ................................................................................ 24 1.5.5 Question 6: How does the seed coat accommodate the secretion of both mucilage and mucilage-modifying enzymes such as MUM2? .......................................................... 24  2  MATERIALS AND METHODS...................................................................................... 28 2.1 ARABIDOPSIS GROWTH AND FLOWER DEVELOPMENT STAGING ....................................... 28 2.1.1 Plant material and growth conditions..................................................................... 28 2.1.2 Staging of flower age ............................................................................................ 28 2.2 MICROSCOPIC TECHNIQUES ........................................................................................... 29 2.2.1 Ruthenium red staining ......................................................................................... 29 2.2.2 High-pressure freezing, freeze substitution and embedding ................................... 29 2.2.3 Electron tomography ............................................................................................. 30 2.3 QUANTITATIVE MORPHOLOGICAL TECHNIQUES .............................................................. 31 2.3.1 Golgi stack and electron-dense vesicle density analysis ......................................... 31 2.3.2 Statistical analysis ................................................................................................. 32 iv  2.4 IMMUNOLABELLING TECHNIQUES FOR MICROSCOPY ....................................................... 33 2.4.1 Primary antibodies ................................................................................................ 33 2.4.2 Intact whole seed and mucilage immunolabelling.................................................. 34 2.4.3 Immunofluorescence of thick sections................................................................... 35 2.4.4 Immunogold labeling for TEM.............................................................................. 36 2.4.5 Immunofluorescent labeling of whole mount, developing seeds for intracellular observations with confocal microscopy............................................................................. 37 2.5 CLONING AND EXPRESSION OF MUM2-HIS .................................................................... 38 2.5.1 Epitope tagging of MUM2..................................................................................... 38 2.5.2 Immunoblotting to test for 6xHis labeling ............................................................. 39 2.5.2.1 2.5.2.2 2.5.2.3 2.5.2.4  Yeast controls.......................................................................................................................................39 Leaf press immunoblot ........................................................................................................................40 Leaf extracts .........................................................................................................................................40 Dot blots ...............................................................................................................................................41  3 ULTRASTRUCTURE OF THE GOLGI APPARATUS IN THE ARABIDOPSIS SEED COAT...................................................................................................................................... 43 3.1 INTRODUCTION ............................................................................................................. 43 3.1.1 Previous and related work ..................................................................................... 45 3.1.2 Objectives ............................................................................................................. 46 3.2 RESULTS ...................................................................................................................... 46 3.2.1 The Arabidopsis seed coat and Golgi apparatus undergo distinct ultrastructural changes during development............................................................................................. 46 3.2.1.1 3.2.1.2 3.2.1.3  The seed coat and Golgi apparatus at 4 DPA.....................................................................................47 The seed coat and Golgi apparatus at 7 DPA.....................................................................................47 The seed coat and Golgi apparatus at 9 DPA.....................................................................................48  3.2.2 The density of stacks in the seed coat Golgi apparatus also changes during development ..................................................................................................................... 49 3.2.3 Decreases in mucilage production affects the morphology, but not the number of Golgi stacks in seed coat epidermal cells........................................................................... 51 3.3 DISCUSSION.................................................................................................................. 52 3.3.1 Prolific mucilage production correlates with distinct morphological changes in the Golgi and TGN................................................................................................................. 52 3.3.2 Biological control of Golgi stack proliferation....................................................... 54 4  LOCALIZATION OF POLYSACCHARIDES IN SEED COAT MUCILAGE............ 63 4.1 INTRODUCTION ............................................................................................................. 63 4.1.1 Objectives ............................................................................................................. 65 4.2 RESULTS ...................................................................................................................... 65 4.2.1 Screening of antibodies for reactivity to mucilage ................................................. 65 4.2.1.1 4.2.1.2  Antibody screening on hydrated mucilage of mature seeds .............................................................66 Antibody screening on sections of cryofixed developing seeds .......................................................68  4.2.2 Plant Golgi stacks produce mucilage synchronously during seed coat differentiation..... .............................................................................................................. 71 4.3 DISCUSSION.................................................................................................................. 72 4.3.1 Golgi stacks respond synchronously, and collectively represent a single Golgi apparatus .......................................................................................................................... 72 4.3.2 The composition of seed coat mucilage ................................................................. 73 4.3.3 Xyloglucan in the secondary cell wall ................................................................... 74 4.3.4 Tracking polysaccharide production with mucilage-specific antibodies ................. 75 v  5  POLARIZED SECRETION AND POST-GOLGI TRAFFIC........................................ 85 5.1 INTRODUCTION ............................................................................................................. 85 5.1.1 Related work ......................................................................................................... 87 5.1.2 Objectives ............................................................................................................. 87 5.2 RESULTS ...................................................................................................................... 88 5.2.1 Golgi stacks do not cluster near the site of secretion .............................................. 88 5.2.2 Vesicles with electron-dense cargo are observed in developing seed coat epidermal cells..... ............................................................................................................................. 89 5.2.3 Electron-dense vesicles also do not cluster near the site of secretion at 7 DPA ...... 91 5.2.4 Cortical microtubules line the mucilage pocket at 7 DPA, but not 9 DPA.............. 93 5.3 DISCUSSION.................................................................................................................. 97 5.3.1 Targeting mucilage to the correct domain of the mucilage pocket.......................... 97 5.3.2 Electron-dense vesicles ......................................................................................... 98 5.3.3 The role of microtubules in seed coat development ............................................. 100  6  SECRETION OF MUCILAGE AND MUCILAGE-MODIFYING ENZYMES ......... 115 6.1 INTRODUCTION ........................................................................................................... 115 6.1.1 Previous and related work ................................................................................... 116 6.1.2 Objectives ........................................................................................................... 117 6.2 RESULTS .................................................................................................................... 118 6.2.1 Production of MUM2-His ................................................................................... 119 6.2.2 Immunolabeling of MUM2-His and mum2-1....................................................... 121 6.2.2.1 MUM2-His immunolabeling.............................................................................................................121 6.2.2.2 Characterization of mum2-1..............................................................................................................124 DISCUSSION................................................................................................................ 125  6.3 6.3.1 Troubleshooting MUM2-His labeling in planta................................................... 126 6.3.2 Exploration of the mum2-1 phenotype................................................................. 128 7  CONCLUSIONS AND FUTURE DIRECTIONS ......................................................... 141  8  BIBLIOGRAPHY........................................................................................................... 146  vi  LIST OF TABLES TABLE 2-1: Decision making process for choice of statistical tests to use in each analysis.......33 TABLE 4-1: List of cell wall probes tested during preliminary screen for antibodies which react with Arabidopsis seed mucilage....................................................................................................68 TABLE 5-1: Results of statistical analysis of mean heights and widths of cryofixed, sectioned seed coat epidermal cells, with standard deviations......................................................................96 TABLE 6-1: Epitopes recognized by RGS-His, Penta-His and Tetra-His antibodies................123  vii  LIST OF FIGURES FIGURE 1-1: Schematic of the flow of traffic in the plant endomembrane system....................26 FIGURE 1-2: Schematic of Arabidopsis thaliana seed coat epidermal cell development..........27 FIGURE 2-1: Sequencing primers for confirmation of orientation of insert in MUM2 to add 6xHis tag........................................................................................................................................42 FIGURE 3-1: The stages of Arabidopsis seed coat development in cryofixed samples..............56 FIGURE 3-2: The morphology of individual Golgi stacks at the different stages of seed coat development...................................................................................................................................57 FIGURE 3-3: Tomographic reconstruction of a single Golgi stack during the mucilage-secreting stage of seed coat development.....................................................................................................58 FIGURE 3-4: Quantification of Golgi stacks visible in sections of seed coat epidermal cells during development.......................................................................................................................59 FIGURE 3-5: Comparison of wild-type and mum4 seed coat morphology at 7 DPA.................60 FIGURE 3-6: Comparison of the ultrastructure of wild-type and mum4 Golgi stacks................61 FIGURE 3-7: Quantification of changes in Golgi stacks in sections of wild-type and mum4 seed coat epidermal cells at 4 and 7 DPA..............................................................................................62 FIGURE 4-1: Antibody screen of mucilage from hydrated mature seeds...................................76 FIGURE 4-2: Distribution of epitopes recognized by anti-seed mucilage and anti-xyloglucan antibodies in developing Arabidopsis seeds..................................................................................78 FIGURE 4-3: Distribution of epitopes recognized by anti-pectins in developing Arabidopsis seeds...............................................................................................................................................79 FIGURE 4-4: Fluorescent labeling antibody controls..................................................................80 FIGURE 4-5: Details of seed mucilage (CCRC-M36) or xyloglucan labeling in 7 and 9 DPA seed coat cells................................................................................................................................81 FIGURE 4-6: Single epitope immunogold labeling of Golgi stacks............................................82 FIGURE 4-7: Double immunogold labeling of wild-type, 7 DPA seed coat epidermal cells.....84 FIGURE 5-1: Distribution of Golgi stacks in wild-type, 7 DPA seed coat epidermal cells......104 FIGURE 5-2: Electron-dense vesicles in 7 DPA seed coat epidermal cells..............................105 viii  FIGURE 5-3: Tomographic slice overlain with a 3-dimensional reconstruction modeled from the whole electron tomogram......................................................................................................106 FIGURE 5-4: Electron-dense vesicles near plasma membrane in columella of 7 and 9 DPA seed coat epidermal cells......................................................................................................................107 FIGURE 5-5: Double immunogold labeling of 7 DPA seed coat epidermal cells using CCRCM36 and anti-xyloglucan.............................................................................................................108 FIGURE 5-6: Immunogold labeling with antibodies to RabA4b in wild-type, 7 DPA seed coat epidermal cells.............................................................................................................................109 FIGURE 5-7: Distribution of electron-dense vesicles in different regions of wild-type, 7 DPA seed coat epidermal cells.............................................................................................................110 FIGURE 5-8: TEM images of cortical microtubule arrangement under the mucilage pocket at 7 and 9 DPA....................................................................................................................................111 FIGURE 5-9: Confocal images of microtubule bundle distribution in 7 and 9 DPA seed coat epidermal cells.............................................................................................................................113 FIGURE 5-10: Height and width of seed coat epidermal cells of different ages and mutant lines..............................................................................................................................................114 FIGURE 6-1: Testing of anti-epitope antibodies.......................................................................130 FIGURE 6-2: Cloning strategy for the production of 6xHis tagged MUM2.............................131 FIGURE 6-3: Schematic of origin and genotype of tissues in each generation.........................132 FIGURE 6-4: T2 seed phenotypes and seedling growth on selective media.............................133 FIGURE 6-5: T3 seed phenotypes and seedling growth on selective media.............................135 FIGURE 6-6: Immunoblotting of leaf and transformed yeast extracts to detect 6xHis epitope..........................................................................................................................................136 FIGURE 6-7: immunfluorescent labeling of cryofixed, resin embedded sections of 7 DPA seeds of T2 plants transformed with M2H vector.................................................................................138 FIGURE 6-8: Confocal images of developing T3 seeds at 7 DPA immunolabeled with the TetraHis antibody.................................................................................................................................139 FIGURE 6-9: Immunolabeling of wild-type and mum2-1 seeds at 9 DPA................................140  ix  LIST OF ABBREVIATIONS 6xHis:  Six times histidine, an epitope tag with six consecutive histidine residues.  35S:  Cauliflower mosaic virus promoter, that acts as a strong constitutive promoter in dicot plants.  ANOVA:  Analysis of variance.  AT:  Arabidopsis thaliana.  ATG:  Adenosine, thymine, guanine. Also known as the start codon.  AtRabA4b:  Arabidopsis thaliana version of RAB protein (or gene) A4b.  ATPase:  An enzyme that breaks down adenosine triphosphate (ATP).  BY-2:  Bright Yellow 2 cultivar of Nicotiana tabacum.  CBD-OG:  Cellulose binding domain, conjugated to Oregon Green.  CCRC:  Complex Carbohydrate Research Center (Athens, GA).  cDNA:  Complementary DNA.  cMyc:  Epitope tag derived from the cMyc gene (amino acid sequence: EQKLISEEDL).  COB:  COBRA. The name derives from the shape of the mutant roots.  DNA:  Deoxyribonucleic acid  DPA:  Days post-anthesis  EDTA:  Ethylene diamine tetraacetic acid  EGTA:  Ethylene glycol tetraacetic acid  EMS:  Ethyl methane sulfonate  ER:  Endoplasmic reticulum  GAUT1:  GALACTURONOSYL TRANSFERASE 1.  GFP:  Green fluorescent protein.  GPI:  Glycosylphosphatidylinositol.  GUS:  -glucuronidase. x  HGA:  Homogalacturonic acid.  IRX8:  IRREGULAR XYLEM 8.  LB:  Luria-Bertani medium.  M2H:  MUM2 gene or protein with a 6xHis tag attached.  MOR1:  MICROTUBULE ORGANIZING 1.  MUM2:  MUCILAGE MODIFIED 2.  MUM2-His:  MUM2 gene or protein with a 6xHis tag attached.  MUM4:  MUCILAGE MODIFIED 4.  NBT/BCIP:  Nitro-Blue Tetrazolium Chloride/ 5-Bromo-4-Chloro-3'-Indolyphosphate pToluidine Salt. Used for colorimetric localization of alkaline phosphatase.  NFDM:  Non-fat dry milk  PCR:  Polymerase Chain Reaction  PEM:  Buffer containing 50 mM PIPES buffer, 2 mM EGTA and 2 mM MgSO4.  PEMT:  PEM buffer with 0.05% Triton-X added.  PGA:  Polygalaturonic acid. Synonym of HGA.  PGM1:  PHOSPHOGLUCOMUTASE 1.  PIN1:  PIN FORMED 1. The name derives from the shape of the mutant inflorescence.  PIPES:  Piperazine-N,N-bis(2-ethanesulfonic acid).  pM2H:  Vector bearing the MUM2-His gene.  PMSF:  Phenylmethylsulphonyl fluoride.  pMUM2g:  Vector bearing the genomic version of MUM2.  QUA1:  QUASIMODO 1. The name derives from the shape and size of the mutant plant.  RAB:  Ras-like in rat brain. The largest group of proteins in the Ras superfamily of GTPases.  RG:  Rhamnogalacturonan. xi  RGI:  Rhamnogalacturonan I.  RGII:  Rhamnogalacturonan II.  RT-PCR:  Reverse-transcriptase PCR.  RGXT1:  RHAMNOGALACTURONAN XYLOSYLTRANSFERASE 1  RGXT2:  RHAMNOGALACTURONAN XYLOSYLTRANSFERASE 2.  SCAMP1:  SECRETORY CARRIER MEMBRANE PROTEIN 1 (Note: AtSCAMP1 refers to Arabidopsis thaliana SCAMP1 and OsSCAMP1 refers to Oryza sativa SCAMP1).  SNARE:  Soluble N-ethylmaleimide protein (SNAP) receptor.  ST-GFP:  Sialyl-transferase GFP.  T2, T3:  Transformed plant, generation #2 (or 3).  TBST:  Tris buffered saline with Tween 20.  TEM:  Transmission electron microscopy.  TGN:  Trans-Golgi network.  UBC:  The University of British Columbia.  UTR:  Untranslated region of DNA.  XG:  Xyloglucan.  -XG:  Anti-xyloglucan.  xii  ACKNOWLEDGEMENTS First and foremost, I would like to acknowledge my supervisors, George Haughn and Lacey Samuels, for their continuous support and guidance. I learned a great deal watching how each one of them approached science, and feel I have emerged from this as a better scientist for having had the chance to be a member of both of their labs. My committee members, Geoff Wasteneys and Linda Matsuuchi have been enthusiastic about my work, as well as offering insightful suggestions for improvement. Past and present members of the Haughn, Kunst, Samuels and Wasteneys labs have been generous with their suggestions, protocols and their time. Though all members have been extremely helpful in their own way, this is especially true of Gillian Dean, who was almost entirely responsible for teaching me the ways of molecular biology, Sarah McKim for helping me out when Gill left, Minako Kaneda for continuous support and teaching me about high pressure freezing, David Bird for endless discussions about statistics and molecular biology, Angel Shan for helping me with genotyping, jonathan Griffiths for helping with the yeast work and Heather McFarlane for putting up with me as a mentor, helping me with my project, and coauthoring 2 papers with me. The staff of the UBC Bioimaging facility have always been generous with their time and knowledge and I am grateful to them for that. Zachary Gergely and other members of the lab of Andrew Staehelin and the Boulder Lab for 3D Electron Microscopy trained me in electron tomography. Michael Hahn provided me with the anti-mucilage antibody that much of this work depends on and is a co-author on my Plant Cell paper. Tamara Western started the seed coat project as a member of George Haughn’s lab and is also a co-author on my Plant Cell paper. Finally, I would like to thank my friends, family and my partner, Miles Hunter. I could not have done this on my own, and so I thank you. xiii  1 GENERAL INTRODUCTION 1.1 The plant cell wall The plant cell wall is an integral component of the plant cell. It is a dynamic structure outside of the plasma membrane (an area known as the apoplast) that is crucial to many aspects of plant growth, development and in the maintenance of cell shape. The industrial uses of plant material are far too many to describe here, from the agriculture industry, to construction, to pulp and paper, to medicine. Pectin is a major component of plant cell walls and pectins are valuable to the food and cosmetic industries as gelling agents and food stabilizers (Willats et al., 2006). The production and transport of pectins and other matrix polysaccharides are the primary focus of this thesis.  1.1.1 Structure and synthesis of the plant cell wall When biologists began studying the plant cell, they viewed the wall as a static structure that should be considered separately from the rest of the cell. It wasn’t until the last 30 years that this view has been challenged and the cell wall is now thought of as a highly dynamic, complex structure that is integral to the proper functioning of the plant cell (Roberts, 2001). The spatial complexity of cell wall polymers has been a barrier to understanding cell wall architecture, since differentiated cell types can have highly variable subdomains of wall structure (Knox, 2008). The current generic model for primary cell wall structure describes a series of frameworks, which are crosslinked to provide maximum strength and support (Somerville et al., 2004; Cosgrove, 2005). The majority of the cell wall is made of complex polysaccharides, though structural and enzymatic proteins also have important roles to play.  1  The most important structural framework consists of a cellulose and hemicellulose network, which provides the strength necessary for the plant to withstand both tension and compression. Cellulose microfibrils consist of paracrystalline arrays of long chains of (14)D-glucose (Roelofsen, 1965; Reiter, 2002). Synthesis of these microfibrils requires complexes of cellulose synthases, known as rosettes, which are found in the plasma membrane (Kimura et al., 1999; Paredez et al., 2006; Persson et al., 2007a). In diffusely elongating cells, cellulose microfibrils wrap around the cell perpendicular to the direction of growth, restricting turgordriven growth to the long axis of the plant organ and determining the shape of plants. Cellulose microfibrils are crosslinked together by hemicelluloses, which coat the microfibrils in a tight, Hbonded association. In the dicot primary cell wall, the predominant hemicelluloses are xyloglucans, characterized by a linear backbone of (14)-D-glucose, with the most common type of side chain being -D-xylose residues attached in a (16) arrangement, which may be further modified by the addition of -L-fucose, -D-galactose and sometimes -L-arabinose (O'Neill and York, 2003). Unlike the plasma membrane-localized synthesis of cellulose, hemicelluloses, such as xyloglucan, have been shown to be synthesized inside the Golgi apparatus, and are then secreted outside the plasma membrane via vesicular transport (Moore et al., 1991). Together, cellulose and hemicelluloses such as xyloglucan are the major load bearing framework of the cell wall (Jarvis and McCann, 2000; Burgert, 2006). The hemicellulose-cellulose framework described above is embedded in a gel, composed mainly of pectic polysaccharides. Pectins are defined by their high galacturonic acid content, and are generally categorized into 3 different types, based on their specific composition: homogalacturonan, rhamnogalacturonan I and rhamnogalacturonan II (Ridley et al., 2001; Mohnen, 2008). Homogalacturonan (HGA) is a linear polymer of D-galacturonic acid residues that are linked together in an (14) linkage, with varying levels of methyl esterification 2  (Willats et al., 2001c; Mohnen, 2008). Rhamnogalacturonan I (RGI) has a backbone of repeating disaccharides, made of D-rhamnose and D-galacturonic acid linked together in a (14) linkage (Lau et al., 1985). The rhamnose-galacturonic acid disaccharides are connected together via a (12) linkage. Varying levels of branching are seen in RGI, with different types of side groups possible, including linear and branched (15)--L-arabinans and/ or (14)--Dgalactans linked to the rhamnose residues at the C4 position (O'Neill and York, 2003; Mohnen, 2008). Despite its name, rhamnogalacturonan II (RGII) has an HGA backbone (not RG) with four distinct, well characterized side chains that are attached to it (Perez et al., 2003; Mohnen, 2008). In the cell wall, RGII has been shown to form dimers that are cross-linked by a borate diester linkage (Kobayashi et al., 1996). There seems to be little overall variability in RGII structure between species, except in one of the four possible side chains (O'Neill and York, 2003; Perez et al., 2003; Mohnen, 2008). Although these compounds are often discussed as if they were three separate molecules, it is possible that this is not the case. It is likely that the different pectins are somehow joined together in the wall to form very large molecules with different domains, and yet the specific arrangement is unknown (Willats et al., 2006). The most widely accepted model is that of the backbones of the three molecules (HGA or RG, depending on the molecule) being covalently joined together in a linear fashion (Perez et al., 2000; Willats et al., 2006). However, an alternative model exists which places the long HGA chains as a side chain of RGI (Vincken et al., 2003). As is the case with the hemicelluloses, pectin biosynthesis has been localized to the Golgi apparatus (Moore and Staehelin, 1988; Moore et al., 1991; Zhang and Staehelin, 1992; Mouille et al., 2007). There is still much that is not known about the biosynthesis of pectin, but it is considered to proceed by the addition of nucleotide-sugars that have been imported from the 3  cytoplasm by transporters onto the growing polysaccharide by glycosyl transferases (Mohnen, 2008). Based on the number of different types of linkages in pectin, and the assumption that there is one enzyme for every linkage, as many as 67 different transferases including glycosyl, methyl and acetyl-transferases (Mohnen, 2008). Very few of these proteins have been characterized, though the rate of discovery is slowly increasing. One of the first genes discovered encoding a pectin biosynthetic enzyme was QUASIMODO1 (QUA1); plants bearing a mutation in this glycosyltransferase had a 25% decrease in galacturonic acid content, as well as deficiencies in cell-cell adhesion, a property conferred by the pectin-rich middle lamella (Bouton et al., 2002; Orfila et al., 2005). RGXT1 and RXGT2 are Golgi-localized xylosyltransferases thought to be specifically involved in the production of RGII (Egelund et al., 2006). GALACTURONOSYL TRANSFERASE 1 (GAUT1, Sterling et al., 2006) and IRREGULAR XYLEM 8 (IRX8, Persson et al., 2007b) are both members of a multigene family of glycosyl transferases (that also includes QUA1) and are both involved in the biosynthesis of HGA, albeit differently; GAUT1 is a galacturonyltransferase involved in the production of the backbone of HGA, whereas IRX8 is important in the production of a subset of HGA onto which -1,4-xylan is attached. In addition to these polysaccharide frameworks that make up the bulk of the primary cell wall, structural and enzymatic proteins are also present, and have an important role to play. Structural proteins such as a group of hydroxyproline rich glycoproteins, known as extensins, are non-enzymatic and found in very low amounts, less than 2% of the wall (Jamet et al., 2008) and yet mutations in these proteins can be embryo-lethal (Johnson et al., 2003; Cannon et al., 2008). Extensins are thought to self-assemble into a positively charged scaffold that may act as a template for the negatively charged pectic polysaccharides (Cannon et al., 2008). Several different types of enzymes, and the non-enzymatic expansins, are secreted to the apoplast, and 4  aid in the assembly and post-secretory modification of the cell wall (Cosgrove, 2003; Rose et al., 2003; Cosgrove, 2005). Expansins modify the non-covalent associations of the cellulosehemicellulose network, which allows the cell to expand in response to turgor pressure (Cosgrove, 2000). Xyloglucan endotransglucosylases/ hydrolases carry out several functions that involve xyloglucans, including both the loosening and strengthening of walls, and also the incorporation of new xyloglucans into the existing framework (Cosgrove, 2005; Popper, 2008). Pectin methylesterases remove methyl groups from linear portions of the pectin HGA, thereby allowing the formation of calcium cross-linked sections of pectin, which have been shown to increase rigidity of the wall (Micheli, 2001; Pelloux et al., 2007), and are especially important in the proper development of pollen tubes (Bosch et al., 2005). An interesting aspect of cell wall secretion is that pectin methylesterases, as well as other cell wall modifying enzymes, must travel through the Golgi apparatus on their route to the apoplast, alongside their polysaccharide substrates. Not much is known about how other enzymes inhibit function for transport, but pectin methylesterases have a leader sequence that is thought to be involved in targeting them to the correct domain of the plasma membrane, in addition to the regulation of enzyme activity during transit (Bosch et al., 2005; Dorokhov et al., 2006). This sequence is cleaved prior to secretion of the pectin methylesterase to the apoplast (Wolf et al., 2009). A hypothesis that describes cosecretion of pectin methylesterases and polysaccharides in the same vesicles exists (Micheli, 2001), but at this point no research has been done that can support or refute it.  1.1.2 Studying polysaccharides within the context of the cell wall Considering the complexity of the plant cell wall, in addition to high levels of variability of the components, both within and between species (Popper, 2008) it is no surprise that gaining insight into the architecture and biosynthesis of the plant cell wall has been difficult. Originally, work on cell wall composition was done mainly by chemical extraction of cell wall components 5  (Northcote and Pickett-Heaps, 1966). Although this did give insight into the general composition (Roelofsen, 1965), specifics were lacking. In addition, removal of components from the context of the cell wall did little to shed light on its architecture in vivo. Electron microscopy allowed for insight into cellulose microfibril orientation, but not other components of the cell wall (Roelofsen, 1965). Autoradiographic studies, using radioactive glucose, were used to decipher the biosynthetic pathway of cell wall materials, showing that pectins and hemicelluloses originate in the Golgi apparatus, while cellulose is produced at the plasma membrane (Northcote and Pickett-Heaps, 1966; Harris and Northcote, 1971). The development of antibodies to cell wall polysaccharides not only confirmed data obtained by autoradiography and other methods (Knox, 1997), but increased the rate of discovery and the level of specificity that could be achieved. With antibodies, both the composition and the linkages of the polysaccharide could be targeted. For example, two antibodies to the pectin homogalacturonan (HGA), JIM5 and JIM7, (Knox et al., 1990), are both specific for HGA but differentiate molecules with varying levels of methyl esterification. JIM5 recognizes HGA epitopes with low (up to 40%) amounts of methyl esterification. JIM7 reacts preferentially with HGA with 15% to 80% esterification, and cannot react with completely demethylated HGA (Willats et al., 2001c). Neither antibody shows any cross reactivity with RGI (Knox et al., 1990). The ability to differentiate these molecules by levels of esterification was proven to be important when Knox et al. (1990) showed that HGA molecules with different levels of esterification had different patterns of localization in several tissue types. This also implied that esterification of HGA was specifically regulated by the plant, an observation later confirmed by the discovery of the pectin methylesterases. Polysaccharide-specific antibodies continue to be developed, each with unique labeling patterns that shed light on cell wall composition. Paul Knox (University of Leeds, Leeds, UK) 6  maintains a supply service (Plant Probes: http://www.plantprobes.net/) which provides 25 antibodies to cell wall polysaccharides, 11 of which bind to different kinds of pectins, including 3 antibodies raised specifically against seed coat mucilage (Verhertbruggen et al., 2009). Michael Hahn’s laboratory at The Complex Carbohydrate Research Center (CCRC) in Athens, GA has also developed antibodies to cell wall polymers (Puhlmann et al., 1994) with 118 antibodies available on their website including 6 that were raised against Arabidopsis seed coat mucilage (Carbosource: http://www.ccrc.uga.edu/ ~carbosource/CSS_home.html).  1.2 The secretion of matrix polysaccharides 1.2.1 The Golgi apparatus as a major site of polysaccharide production As previously mentioned, pectin production was first localized to the Golgi apparatus through the use of autoradiography (Northcote and Pickett-Heaps, 1966; Harris and Northcote, 1971; Bolwell and Northcote, 1983). Moore et al. (1986) later used a polyclonal antibody to the pectin rhamnogalacturonan I (RGI) to confirm that pectins such as RGI could be localized to the Golgi apparatus and secretory vesicles (Moore and Staehelin, 1988). This antibody is believed to recognize the HGA-like backbone of RGI, and has been shown to cross react with HGA in vitro. Different combinations of antibodies (Moore et al., 1991) such as the anti-RGI described above, an antibody to the hemicellulose xyloglucan (Moore and Staehelin, 1988) and an antibody to the cell wall protein extensin (Stafstrom and Staehelin, 1988) were used to test the hypothesis that the plant Golgi apparatus produces multiple products in a single stack. Using immungold labeling it was shown that individual Golgi stacks were capable of simultaneously producing different complex polysaccharides and modifying proteins destined for the cell wall (Moore et al., 1991). In addition, the biosynthetic function of the different cisternae of the Golgi stack could be distinguished. Complex polysaccharides such as RGI could be labeled in all 7  compartments of the Golgi stack, whereas xyloglucan could only be labeled in the trans cisternae and the Trans Golgi Network (TGN) (Moore et al., 1991; Zhang and Staehelin, 1992). Interestingly, the focus of most antibody labeling experiments has been to identify specific subdomains of the cell wall (Knox et al., 1990; Lynch and Staehelin, 1992; Freshour et al., 1996) or to examine the organization of individual Golgi stacks (Moore and Staehelin, 1988; Moore et al., 1991; Staehelin et al., 1991; Zhang and Staehelin, 1992). There has been little work to determine how the entire Golgi population uses its multiple stacks to move specific cargo around the cell and to its destination in the cell wall.  1.2.2 The structure and function of the plant Golgi apparatus The plant Golgi apparatus is comprised of large numbers of individual Golgi stacks, which are widely distributed throughout the cytoplasm and are capable of streaming, a process mediated by the actin cytoskeleton (Griffing, 1991; Boevink et al., 1998; Nebenführ et al., 1999). Each individual Golgi stack is made of several flattened sacs, called cisternae. The cisternae each have discrete morphological and functional characteristics and, as such, a polarity can be assigned to the Golgi stack (Mollenhauer and Morre, 1994; Driouich and Staehelin, 1997). The cisterna of the cis face is the least electron dense when observed in transmission electron microscopy (TEM), and generally is observed to have the widest lumen, relative to the other cisternae. It is also the site of entry for proteins coming from the ER. Both the density of staining and the compression of the lumen increase as one moves through the stack towards the opposite, or trans side of the stack (Robinson and Kristen, 1982; Andreeva et al., 1998). Adjacent to the trans-most cisterna is a tubulo-vesicular network known as the trans Golgi Network (TGN). It is where products are sorted and packaged for transport to other organelles, or the plasma membrane (Moore et al., 1991). Despite these general characteristics, the specific morphology of the plant Golgi apparatus varies widely in different cells and tissues, and also as cells undergo 8  differentiation (Driouich and Staehelin, 1997; Winicur et al., 1998). As in all eukaryotes, the role of the Golgi apparatus is to modify, sort and package proteins that are destined for the lysosome, vacuole, or the plasma membrane. Additionally, the Golgi apparatus is the major site of production of matrix polysaccharides for the plant cell wall (Doblin et al., 2003). A positive correlation has been observed between specific Golgi stack ultrastructure and periods of intense polysaccharide secretion in land plants. The assumption is that Golgi stack ultrastructure is dependant on the intensity of polysaccharide biosynthesis, however this has not been tested. Although the severity of the changes in the Golgi stack varies from one plant system to another, the characteristics of Golgi stacks observed in cells involved in intense polysaccharide secretion include: a more prominent electron dense cis to trans polarity, increased swelling at the margins of the cisternae, a more prominent TGN, and many vesicles or vesicular clusters associated with the Golgi stack (Staehelin et al., 1991; Mollenhauer and Morre, 1994). This particular morphology has been documented in several different plant systems, including the cells of the developing root cap of maize (Dauwalder et al., 1969; Staehelin et al., 1990), during the formation of secondary xylem in pine (Samuels et al., 2002) and during the cell cycle of the meristematic cells of onion roots (Gonzalez-Reyes et al., 1988). Interestingly, this change is always observed to occur in all Golgi stacks synchronously (Cronshaw and Bouck, 1965; Winicur et al., 1998). Not only does the morphology of the individual stacks of the Golgi apparatus change during development, but the number of stacks in the cell has also been observed to change over time. During the cell cycle of meristematic cells, the number of Golgi stacks in the cell doubles just prior to the onset of mitosis (Garcia-Herdugo et al., 1988; Segui-Simarro and Staehelin, 2006). However, despite evidence that the number of Golgi stacks in a cell can be specifically modulated, the possibility of using the number of Golgi stacks in the cell as a way to 9  accommodate changes in secretion has not been studied in great detail. There are likely multiple reasons for this. First, in many of the plant systems used to study secretion it is difficult to compare Golgi stacks based on developmental stages. This is the case in the root cap; the developmental stage of the cells in the root cap is best described as a gradient, with very few clear morphological indicators to mark the transition from one developmental stage to the next, other than the structure of the Golgi apparatus itself (Dauwalder et al., 1969). With no clear demarcation of developmental stages, it is difficult to use TEM data to explore this aspect of Golgi apparatus function. Second, though GFP has been used to examine the total number of Golgi stacks in suspension cultures of BY-2 cells (Nebenführ et al., 1999), the use of this tool to look at differentiating cells in tissue is more complex. This is further confounded by the fact that the 35S promoter does not always express well in the epidermal cells that are so commonly used to study secretion, making localization of the GFP fluorescence problematic (Young et al., 2008). Third, and potentially less important, the fragmentation of the Golgi apparatus into multiple stacks is by no means a universal eukaryotic trait; it is possible that the importance of the capacity to modulate stack numbers has been overlooked in the past. Considering the very obvious morphological changes and the highly varied levels of secretion that the Arabidopsis seed coat undergoes during its development (Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000), it is well suited to the study of changes in Golgi stack number during development. The fact that mutants of Arabidopsis exist in which secretion in the seed coat has been altered, without any other obvious phenotype (eg. Western et al., 2001; Western et al., 2004; Dean et al., 2007) makes it an even more powerful tool.  1.2.3 Post-Golgi traffic The Golgi stack is not only responsible for the production of cell wall products; it is also a sorting centre for proteins produced in the ER that are destined for lytic and storage vacoules 10  and the endosomal system, in addition to the secretory pathway to the plasma membrane. The trans-Golgi network (TGN) is thought to be the main site within the Golgi apparatus that is responsible for sorting of products and packaging them into vesicles for transport to their final destination. A schematic of the plant secretory pathway can be seen in Figure 1-1. There is still much to learn about the exact mechanisms by which the TGN sorts cargo for delivery to different compartments in plants. By analogy to other eukaryotic systems, receptors recognizing specific peptide or carbohydrate molecules present within the structure of the cargo allow for incorporation into different types of vesicles that have different destinations (De Matteis and Luini, 2008). The presence of two unique and distinct types of vacuole, storage and lytic, suggests that sorting of cargo in plants could be even more complex than anticipated (Neuhaus and Rogers, 1998). However, it has been shown that plants manage at least some of this complexity through the use of different types of vesicles and sorting signals for different compartments. The default pathway for products in the Golgi apparatus is the plasma membrane and, as such, no specialized sorting sequence is thought to be required (Denecke et al., 1990; Hwang, 2008). Despite this, membrane-bound H+-ATPases in tobacco have been shown to require a sorting sequence in order to be properly targeted to the plasma membrane (Lefebvre et al., 2004). Surprisingly little is known about the vesicles destined for the plasma membrane, as no experimental procedure has been developed in plants that can reliably differentiate between free vesicles destined for the plasma membrane and those destined for different compartments in the cell (Battey et al., 1999; Rojo and Denecke, 2008). However, free vesicles of a similar size and electron density to swollen cisternal margins are often observed in highly secreting cell types, such as root cap cells (Staehelin et al., 1990). Since these vesicles are much larger than those derived via coat proteins such as clathrin, it is believed that clathrin is not involved in direct traffic to the plasma membrane for the most part (Moore et al., 1991; Sherrier and 11  Vandenbosch, 1994; Driouich and Staehelin, 1997). Some studies suggest that an intermediate compartment exists between the TGN and the plasma membrane, known as the partially coated reticulum (Pesacreta and Lucas, 1985; Mollenhauer et al., 1991) while others maintain that this structure is merely an extension of the TGN (Griffing, 1991; Staehelin and Moore, 1995). If the partially-coated reticulum is, indeed, its own compartment, none would dispute the idea that much of vesicle traffic to the plasma membrane bypasses this compartment entirely (Staehelin and Moore, 1995). Traffic to either the storage or lytic vacuole from the Golgi apparatus requires some kind of sorting sequence (Hwang, 2008; Rojo and Denecke, 2008). Transport to the lytic vacuole is most likely through a pre-vacuolar compartment, which may be analogous with the multi-vesicular body (Tse et al., 2004) via clathrin-coated vesicles (Kirsch et al., 1994). Alternatively, uncoated electron-dense vesicles bearing storage proteins could bypass the prevacuolar compartment and go directly to the storage vacuole (Hohl et al., 1996; Robinson and Hinz, 2005; Vitale and Hinz, 2005). No research has been done to examine whether uncoated vesicles, such as those involved in transport to the plasma membrane or the storage vacuole, require specific cellular machinery to form, such as coat proteins. Despite the fact that all membranes are theoretically capable of forming uncoated vesicles by spontaneous fusion and scission, the electrostatic forces involved are high (Kuzmin et al., 2001) and, as such, the chances of spontaneous formation or fusion of vesicles is slim (Esseling-Ozdoba et al., 2008a). The use of coat proteins and other molecular machinery by the cell to aid in overcoming these forces (Staehelin and Moore, 1995; Jürgens and Geldner, 2002; Jahn et al., 2003) has the added advantage of allowing for specificity. This, in turn, helps to ensure that vesicles fuse with only their target membranes (Blatt et al., 1999; Surpin and Raikhel, 2004). Two types of uncoated vesicle have been loosely defined based on morphological characteristics: those that have electron dense contents, and those that do not 12  (Driouich and Staehelin, 1997). Whether these two types of vesicles form two distinct populations involved in different post-Golgi pathways is unknown at this point. Electron-dense vesicles have been specifically implicated in the transport of proteins to the storage vacuole (Hohl et al., 1996; Robinson et al., 1997; Hillmer et al., 2001). On the other hand, electron-dense vesicles have also been observed in cells that are not likely to be involved in protein storage, such as cells undergoing xylogenesis (Cronshaw and Bouck, 1965) so it is unlikely that transport to the storage vacuole is the sole purpose of electron-dense vesicles.  1.2.4 Vesicle targeting and the cytoskeleton Vesicles must travel through the cytosol in order to reach their destination. In mammalian cells, long-range transport of vesicles is primarily via the microtubule network, which radiates outward from the centrosome (Cole and Lippincott-Schwartz, 1995; Bloom and Goldstein, 1998). Once at the target membrane, vesicle tethering, docking and fusion is highly regulated to ensure that cargo proteins do not end up in the wrong place (Bock et al., 2001). As many as 12 proteins could be involved in the docking and fusion of a single vesicle, including proteins such as SNAREs and the small monomeric GTPases known as RABs, with several orthologs of each protein in a single cell to ensure specificity (Bock et al., 2001). The major plasma membrane targeting complex in yeast and animals is the exocyst (Guo et al., 2000), an eight-protein complex that most likely forms a rod-like structure to facilitate docking of the incoming vesicles at the target membrane (Munson and Novick, 2006). In plant cells, microtubules do not radiate outwards from a central microtubuleorganizing center, but form a cortical interphase array adjacent to the plasma membrane (Wasteneys and Galway, 2003). Not surprisingly, Golgi stack movement seems to be largely independent of microtubules, as shown by the fact that Golgi stacks continue to move in the presence of microtubule inhibitors, but not actin inhibitors (Mollenhauer and Morre, 1976; 13  Nebenführ et al., 1999). Chemical disruption of both actin and microtubules did not affect transport between the ER and the Golgi in BY-2 cells (Saint-Jore et al., 2002) and in tobacco leaves (Brandizzi et al., 2002). These observations, as well as the observation that Golgi stacks often cluster near sites of intense secretion has been used as the premise for a hypothesis that states that localization of the Golgi stack is a major factor in post-Golgi traffic, and that vesicles travel to their destination from the Golgi stack by diffusion alone. Both Golgi stacks and vesicles are in motion within the cell, as a result of cytoplasmic streaming. Cytoplasmic streaming occurs via the acto-myosin network (Kachar and Reese, 1988; Shimmen and Yokota, 2004) but it is unclear whether the movement of vesicles within the cell is also directly dependant on this network. Mathematical modeling of the effects of cytoplasmic streaming on the rest of the cytosol indicates that vesicles and other unbound cellular contents could be carried along by hydrodynamic flow, and thus they may not require direct connections to the cytoskeleton (Houtman et al., 2007). Microinjection of synthetic vesicles into streaming cells has confirmed that vesicles can move through the cell in this fashion, but these studies cannot address whether true vesicles use the actin network for movement (Esseling-Ozdoba et al., 2008b). The role of vesicle docking and fusion machinery in plants in the constitutive transport of cell wall polysaccharides is an area of active research. Homologs of most, if not all of the components identified in yeast and mammalian systems exist in plants, including RABs and SNAREs (Jürgens and Geldner, 2002; Lipka et al., 2007; Nielsen et al., 2008), but a role for these in constitutive secretion has been difficult to identify. All eight subunits of the plasma membrane targeting complex, the exocyst, have been identified in Arabidopsis and shown to be important for targeted secretion in developing pollen tubes, but a role in diffuse growth has yet to be determined (Hala et al., 2008). Considering the mobility of the Golgi stacks within the cell, 14  and the fact that diffuse growth appears to require similar proteins and polysaccharides along entire faces of the cell, the protein machinery used in vesicle targeting in non-plant systems may be less important. On the other hand, the fact that plasma membrane targeting mutants have been so difficult to find for so long may indicate that such mutations are lethal, or that multiple redundant copies of targeting proteins exist, making it difficult to knock out functionality. Either way it may speak to the importance of the proteins or protein families involved.  1.3 Polarized secretion in plants In plants, cell growth and secretion are intimately linked, due to the requirement for cell wall polysaccharides originating from the endomembrane system during cell division and expansion. The growth of the cell in all directions equally, or isotropic growth, is common in the earliest stage of growth of the cell in the meristem. The cell quickly begins to expand in a directed fashion as it begins to differentiate. Directed growth, also known as anisotropic growth, is found in the majority of the cells of the plant to a varying degree (Schnepf, 1986; Martin et al., 2001). Growth is considered to be diffuse if the cell expands equally along a single axis, driven mainly by turgor pressure, with the directionality of growth attributed to the arrangement of microtubules and cellulose microfibrils more than the deposition of cell wall components produced in the Golgi apparatus (Kost et al., 1999; Smith and Oppenheimer, 2005). A more extreme case of polarized growth is that of tip growing cells. Tip growth is polarized growth that is found in highly specialized cells, such as root hairs (Sherrier and Vandenbosch, 1994) and pollen tubes (Mascarenhas, 1993; Geitmann and Emons, 2000). In this case, the secretion of cell wall compounds into the apoplast occurs only at the apex of the developing tube (Martin et al., 2001; Campanoni and Blatt, 2007). Much of the work that has been done to characterize directed secretion has been done on polarized tip growing cells (Hepler et al., 2001; Campanoni and Blatt, 2007). In both root hairs 15  and pollen tubes the area directly adjacent to the growing apex is enriched in secretory vesicles carrying cell wall polysaccharides (Mascarenhas, 1993; Sherrier and Vandenbosch, 1994). Golgi stacks are not observed in this zone, but are found clustered just behind it, along with microtubules, which do not enter the apical region of the cell. Actin filaments are found in bundles that become finer as one nears the extreme apex, where they are said to be absent (Geitmann and Emons, 2000; Geitmann et al., 2000; Ketelaar et al., 2003). However, single strands of actin are extremely difficult to image reliably, as they are below the resolution of light microscopy, and are highly sensitive to fixation (Segui-Simarro et al., 2004). Actin appears to play an important role in the delivery of cell wall products because actin depolymerizers disrupt secretion. However, its precise function remains unclear (Miller et al., 1999; Geitmann et al., 2000; Ketelaar et al., 2003; Bove et al., 2008). In pollen tubes, vesicles enter the apical region, whereas other organelles do not. Additionally, vesicles may enter and leave the apical area several times before fusing with the plasma membrane (Bove et al., 2008). Despite the fact that the actin network was not investigated in the Bove et al. (2008) study, it seems unlikely that a vesicle that was entirely dependent on actin for its motility would move through the apex of the pollen tube in this manner, as the actin filament could provide specific directionality to the movement of the vesicle, thereby making fusion more efficient. Inhibitor studies on root hairs also indicate that microtubules are important in order to maintain growth directionality (Bibikova et al., 1999; Ketelaar et al., 2003; Sakai et al., 2008) and also to organize actin filaments in the apex of the growing tip (Tominaga et al., 1997; Geitmann and Emons, 2000). Cell division is another well-studied example of polar secretion, though not of polar growth per se. Cytokinesis in plants involves the formation of a cell plate, which is produced and secreted by the Golgi apparatus (Samuels et al., 1995; Jürgens, 2005). In a similar fashion to root hairs and pollen tubes, Golgi stacks cluster near the site of cell plate formation, but are separated 16  from the cell plate by the cytoskeletal array known as the phragmoplast (Nebenführ et al., 2000; Segui-Simarro and Staehelin, 2006). Microtubules and actin filaments together form the phragmoplast (Jürgens, 2005). Vesicles are observed to attach to microtubules of the phragmoplast (Segui-Simarro et al., 2004), however vesicles can also move to the central region of the growing cell plate, even though microtubules remain at the periphery (Esseling-Ozdoba et al., 2008a). In diffusely growing cells, Golgi stacks within interphase cells are found streaming throughout the cytoplasm with the aid of the acto-myosin network (Kachar and Reese, 1988; Nebenführ et al., 1999), in a seemingly random pattern (Staehelin et al., 1990; Segui-Simarro and Staehelin, 2006) while microtubules form a cortical array that encircles the cell (Collings and Wasteneys, 2005). How the polysaccharides produced in these streaming Golgi stacks are deposited in a controlled manner around the cell is not known. There is virtually no experimental evidence concerning the control of vesicle deposition during secretion of cell wall polysaccharides in diffusely growing cells. In the case of proteins, the most well-studied cases of polar localization in non-tip growing cells are the plasma membrane-bound auxin transporters, which are found on a single face of the cell (Feraru and Friml, 2008). The polar localization of PIN1, an auxin efflux carrier (Gälweiler et al., 1998) was found to be maintained primarily by rapid cycling of the PIN1 protein through the endosome (Steinmann et al., 1999; Geldner et al., 2003). However, no research has conclusively determined whether the secretion of the PIN proteins by the Golgi apparatus is itself polar. The only protein that has been shown experimentally to be secreted in a polar fashion to date is COBRA (COB), a GPI-anchored protein that is secreted preferentially to the outer periclinal plasma membrane and cell wall of elongating root cells in a pattern coincident with the cortical microtubules (Roudier et al., 2005).  17  1.4 The Arabidopsis seed coat Arabidopsis thaliana has myxospermous seeds, which produce large amounts of pectinaceous mucilage that is secreted to the apoplast during a specific developmental stage and can be released upon hydration to form a gel-like capsule around the seed. This pectin-rich gel is thought to have a role in hydration and protection of the seed, as well as maintaining seed dormancy (Haughn and Chaudhury, 2005). The mucilage is targeted to a donut-shaped pocket outside the plasma membrane at the junction of the radial and outer tangential cell walls of the epidermal cells of the seed coat (Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000). Development of the Arabidopsis seed coat has been studied using conventional light and electron microscopy (Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000) and the major developmental stages have been characterized (Figure 1-2). Prior to fertilization, the outer cells of the ovule appear relatively unspecialized. Fertilization initiates seed coat development and, as such, is the theoretical starting point from which the stage of the seed coat is measured. However, the true moment of fertilization is virtually impossible to know with certainty, therefore the stages of seed coat development are counted from the earliest moment that fertilization could take place, namely the day that the anthers dehisce, and pollen is released. As such, the stage of development of the seed and seed coat is counted as days post-anthesis, or DPA. After fertilization, the developing seed grows rapidly for the first few days (Figure 1-2). Mucilage production is initiated around 6 or 7 DPA, and the newly synthesized mucilage components are deposited in the donut-shaped pocket between the plasma membrane and the primary cell wall (Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000). Coincident with the deposition of mucilage is the formation of a volcano-shaped cytoplasmic column 18  (Figure 1-2). By 9-10 DPA, mucilage production is completed, and a secondary cell wall is deposited around the cytoplasmic column (Figure 1-2). At maturity, the seed coat epidermal cells consist of a columella made exclusively of secondary cell wall components, with a donut shaped mucilage pocket below the primary cell wall on the outer side of the cell (Figure 1-2, Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000).  1.4.1 Polysaccharide components of Arabidopsis seed coat mucilage Previous work on the seed coat cells of Arabidopsis has revealed that there are a number of different cell wall polysaccharides present in seed coat mucilage. Historically, indirect methods, such as chemical stains and antibody labeling have been used to examine mucilage composition. Ruthenium Red staining of extruded mucilage indicated the presence of acidic pectic polysaccharides (Western et al., 2000; Willats et al., 2001a); Calcofluor staining indicated that cellulose was also present (Willats et al., 2001a). Analysis of monosaccharides in mucilage confirmed that pectin is likely a major component of mucilage, due to the presence of galacturonic acid and rhamnose (Western et al., 2000; Western et al., 2001). However, this information does not provide any insight into the relative amounts of different components, nor does it explain how these polysaccharides are arranged within the mucilage. To better understand the architecture of mucilage, a number of pectic antibodies have been used to label whole seeds, as well as high-pressure frozen and sectioned material. The antibodies JIM5 and JIM7 both labeled extruded mucilage, but differently: JIM5 (0-40% esterified HGA) labeled the inner layer of mucilage, whereas JIM7 (40-100% esterified HGA) labeled primarily the outer layer of mucilage, especially when pretreated by a cation chelator, such as EDTA (Willats et al., 2001a). However, Western et al. (2004) found that JIM5 and JIM7 did not label mucilage at all, unless seeds were subjected to extensive shaking prior to immunolabeling. They also showed that both extruded mucilage and the intact mucilage pocket 19  of 9 DPA seeds labels strongly with the polyclonal antibody to RGI described previously (Moore et al., 1986). Since the exact chemical composition of mucilage was unknown, these discrepancies were difficult to resolve. In 2007, a study was done in which the chemical composition of mucilage was extensively examined (Macquet et al., 2007a). In this study, a combination of antibody labeling and confocal microscopy, as well as various techniques for chemical and enzymatic analysis, were used to examine the general structure of mucilage. The main component of all layers of mucilage was RGI, which confirmed previous studies (Western et al., 2000; Western et al., 2001; Willats et al., 2001a; Usadel et al., 2004; Western et al., 2004). In addition, HGA and cellulose were also found in mucilage, but in much lower quantities, and mostly in the inner, adherent layer (Macquet et al., 2007a). The lower relative amount of HGA, as well as its location in the more tightly packed inner layer, could be part of the reason for the discrepancies in labeling mucilage with antibodies such as JIM5 and JIM7 that have been reported earlier (Willats et al., 2001a; Western et al., 2004). Regardless, despite the advantages of knowing which specific polysaccharides are in mucilage, this study does not examine how these components are arranged in relation to each other, and how such structure affects the hydrating properties of mucilage. Mutant analysis of plants with altered mucilage has also increased our understanding of mucilage biosynthesis and these findings can be extrapolated to primary cell wall biosynthesis generally. The MUCILAGE-MODIFIED4 (MUM4) gene encodes a putative rhamnose synthase, and the severe decrease in mucilage in the MUM4 mutant provides further evidence that rhamnose is an important component of mucilage (Usadel et al., 2004; Western et al., 2004). MUCILAGE-MODIFIED2 (MUM2) encodes a -galactosidase that appears to be involved in the modification of mucilage. Mutants that lack properly functioning MUM2 have mucilage that does not expand when hydrated (Dean et al., 2007; Macquet et al., 2007b). This work was 20  important, since it showed for the first time that the architecture of mucilage is integral to its proper function.  1.5 Objectives Pectin is a critical component of plant cell walls that is synthesized in the Golgi apparatus and secreted to the apoplast by vesicles, like many other wall polysaccharides. Pectin research has recently focused primarily on two main areas for the most part: identifying the biosynthetic enzymes involved in pectin production (Reviewed by Mohnen, 2008) and examining how different pectins are arranged within the architecture of the cell wall (Reviewed by Knox, 2008). However, there are many unanswered questions about the cellular processes involved in pectin production. The largest gap in understanding involves how the Golgi apparatus, both at the level of the individual Golgi stacks and as a collective that makes up the Golgi apparatus, supports the biosynthesis of pectin. Then, once it is synthesized, how are pectins and other compounds destined for the cell wall packaged into vesicles and secreted to specific domains of the apoplast? The research described in this thesis examines the cellular machinery supporting the biosynthesis and secretion of cell wall polysaccharides, specifically the Golgi apparatus, in the context of mucilage secretion during development of the Arabidopsis seed coat. To place pectin production within the context of the Golgi apparatus as well as to increase our understanding of the spatial relationship between Golgi stacks and its secreted product, the following questions were asked: (1) How does the Golgi apparatus change as the level of matrix polysaccharide biosynthesis and secretion changes? (2) Are there individual Golgi stacks in the cell dedicated to producing specific matrix polysaccharides? (3) How does the plant cell ensure that carbohydrates produced in the Golgi apparatus arrive at a specific cell wall domain? Is the mobility of the Golgi stacks a mechanism of plant secretion? 21  (4) How does the distinct cytoskeletal arrangement observed in the seed coat epidermal cells affect secretion? (5) How does the seed coat accommodate the secretion of both mucilage and mucilagemodifying enzymes such as MUM2? High resolution microscopic techniques, including high pressure freezing, freeze substitution and resin embedding, were used to address these questions, as these have been shown to be superior for the preservation of ultrastructure in plant cells (Kiss et al., 1990).  1.5.1 Question 1: How does the Golgi apparatus change as the level of matrix polysaccharide biosynthesis and secretion changes? The correlation between periods of intense secretion and changes in Golgi stack ultrastructure have been previously observed, including changes in the number of Golgi stacks within the cell. This leads to the question of whether ultrastructure and/ or the number of Golgi stacks is driven by the amount of product being synthesized by the Golgi apparatus. To examine Golgi ultrastructure in detail, transmission electron microscopy and electron tomography were used to document Golgi stack ultrastructure before, during and after mucilage production. To determine whether the number of Golgi stacks that make up the Golgi apparatus is also affected by the levels of secretion, Golgi stacks at each developmental stage were mapped. Comparisons of the ultrastructure and number of Golgi stacks in wild-type cells with those of the reduced mucilage mutant, mum4, allowed me to ask whether these cellular changes are dependent on polysaccharide biosynthesis, or instead are a result of the specific developmental program of seed coat epidermal cells. All of this work is included in the first results chapter, Chapter 3, entitled Ultrastructure of the Golgi Apparatus in the Arabidopsis Seed Coat.  22  1.5.2 Questions 2: Are there individual Golgi stacks in the cell dedicated to producing specific matrix polysaccharides? Previous studies have focused primarily on how matrix polysaccharides are distributed among the cisternae of individual Golgi stacks. As a result, the question of whether different Golgi stacks take on different roles during polysaccharide biosynthesis, thereby working in a coordinated fashion to fulfill the needs of the cell, has been largely ignored. To address this, the distribution of mucilage-specific epitopes within the Golgi apparatus, was examined by antibody labeling. The distribution of different matrix polysaccharides within the Golgi apparatus could be determined by tracking them during the same developmental stage, whereas tracking polysaccharides over time answered questions about how polysaccharide production changes over time, as the needs of the cell change from one developmental stage to another. This work is described in Chapter 4, entitled Localization of Polysaccharides in Seed Coat Mucilage. In addition, the results described in this chapter and Chapter 3 were published together in Plant Cell (Young et al., 2008, see page 43 for full citation).  1.5.3 Question 3: How does the plant cell ensure that carbohydrates produced in the Golgi apparatus arrive at a specific cell wall domain? Is the mobility of the Golgi stacks a mechanism of plant secretion? In plant cells involved in obvious polarized secretion such as root tips, pollen tubes, or cells undergoing division, Golgi stacks are often observed clustering near the site of deposition of product, leading to a hypothesis that clustering of Golgi stacks is a mechanism of concentrating secretory product to a single spatial domain. This hypothesis is tested in Chapter 5, entitled Polarized Secretion and Post-Golgi Traffic, by examining the distribution of 23  mucilage-containing Golgi stacks in the Arabidopsis seed coat epidermal cells. Additionally, a population of electron-dense vesicles observed in these cells was examined to determine whether it has a role in secretion. Some of this work was also included in the Plant Cell paper mentioned above (Young et al., 2008).  1.5.4 Question 5: How does the distinct cytoskeletal arrangement observed in the seed coat epidermal cells affect secretion? Microtubules are important for plant secretion (Wasteneys and Galway, 2003; Wasteneys and Ambrose, 2009), and yet their exact function is not entirely clear. Previous work had shown that 7 DPA seed coat cells had a distinct cortical array of microtubules lining the plasma membrane adjacent to the mucilage pocket. This led to a hypothesis that these microtubules were involved in targeting of mucilage products to the proper domain of the cell wall. The hypothesis was tested by Heather McFarlane, an undergraduate student under my supervision. Together we determined that organized microtubules were not required for targeted secretion and published the results of her study in Planta (McFarlane et al., 2008, for full citation see page 42). As no specific role could be determined for microtubules in secretion, other possibilities were examined. This work is also in Chapter 5.  1.5.5 Question 6: How does the seed coat accommodate the secretion of both mucilage and mucilage-modifying enzymes such as MUM2? The secretion of matrix polysaccharide modifying enzymes poses a interesting problem for plant cells, as both the enzyme and its potential substrate must both pass through the Golgi apparatus prior to secretion to the apoplast. Enzymes must be strictly regulated during secretion, in order to ensure that they do not act upon their substrates prematurely. MUM2 is a galactosidase, whose putative role is to modify mucilage so as to increase its hydration properties 24  (Dean et al., 2007). MUM2 is a secreted enzyme and, as such, it must travel through the Golgi apparatus alongside rhamnogalacturonan I (RGI), its predicted substrate. The distribution of MUM2 proteins relative to mucilage in the Golgi apparatus, as well as in secretory vesicles, would provide insight into how the plant cell manages the secretion of enzymes and their substrates. This work is described in the final chapter of results, Chapter 6, entitled Secretion of Mucilage and Mucilage-Modifying Enzymes.  25  FIGURE 1-1: Schematic of the flow of traffic in the plant endomembrane system. Proteins are translated into the ER, and sent to the Golgi apparatus. Polysaccharides are synthesized in the Golgi apparatus. All Golgi cargo is sorted into clathrin-coated or uncoated vesicles in the TransGolgi network, depending on its destination. Uncoated vesicles are involved in the transport to the plasma membrane, and the storage vacuole. Vesicles destined for the plasma membrane are likely to undergo maturation and/ or processing during transport. Clathrin-coated vesicles carry cargo destined for the lytic vacuole, via the pre-vacuolar compartment, which may be the same as the multivesicular body. Clathrin-coated vesicles are also involved in endocytosis, as well as transport to and from the partially coated reticulum (Based on a figure in Hawes et al., 1999).  26  FIGURE 1-2: Schematic of Arabidopsis thaliana seed coat epidermal cell development. a – amyloplast; v – vacuole; m – mucilage; 2cw – secondary cell wall.  27  2 MATERIALS AND METHODS This chapter provides general details of the methods that were used to obtain results described in Chapters 3, 4, 5 and 6. In some cases, specific aspects of methods were altered for troubleshooting, or to try different approaches for specific portions of the work described in this thesis. These details are described in the relevant sections of the Results.  2.1 Arabidopsis growth and flower development staging 2.1.1 Plant material and growth conditions Wild-type Arabidopsis thaliana was Columbia-2 (Col-2) ecotype (Lehle Seeds, Round Rock, TX). The mutant lines, mum4-1 (referred to as mum4) and mum2-1 have been described (Western et al., 2001; Western et al., 2004; Dean et al., 2007). Additional mutant lines used in this thesis include mor1-1 (Whittington et al., 2001) and pgm1 (Caspar et al., 1985). Seeds were grown on prepared soil mix (Sunshine 5 Professional Growing Mix, Sungro Horticulture Canada, Seba Beach, AB) that had been fertilized with AT medium (Haughn and Somerville, 1986) in growth chambers at 21°C under continuous light (90-120 μE m-2s-1 photosynthetically active radiation).  2.1.2 Staging of flower age As it is difficult to know the true moment of fertilization in living plants, the age of the developing seed is counted in days from the time that the anthers dehisce and pollen is released (days post-anthesis, DPA). This is the earliest possible date that fertilization can occur and is counted as 0 DPA (Western et al., 2000). Anther dehiscence can be tracked morphologically, as it occurs on the day that flowers opened but before the gynoecium lengthens beyond the stamens. The pedicels of flowers displaying this morphology were painted with non-toxic, water based 28  paint to mark them. Colour coding of paint allowed seeds to be collected from siliques that had were of the appropriate age (4, 7 or 9 DPA) on the same inflorescence stem. In addition to the flower staging method, a morphological examination of seed coat cells in sectioned material with a light microscope was also used to ensure that flower staging was consistent. Thick (0.5 μm) sections, stained with toluidine blue, were examined by light microscopy for intact seed coat cells at each stage of development and to ensure that embedded seeds had embryo development and seed coats with morphology that was consistent with previous results (Western et al., 2000). 4 DPA seed coat cells had no mucilage, with a large central vacuole. 7 DPA seed coat cells have small to large mucilage pockets visible, with the characteristic central cytoplasmic column in the centre of the cell. 9 DPA cells have secondary cell wall visible along the entire apical surface of the cell, below the mucilage pocket and above the central cytoplasmic column.  2.2 Microscopic techniques 2.2.1 Ruthenium red staining Mature seeds were placed in 1.5 ml microcenterfuge tubes containing an aqueous solution of 0.1% ruthenium red (Sigma-Aldrich Canada Ltd, Oakville, ON) for a minimum of 30 minutes on an orbital shaker at 150 rpm. The hydrated mucilage layer could then be examined by light or dissecting microscope.  2.2.2 High-pressure freezing, freeze substitution and embedding Four, 7 and 9 DPA seeds were excised from siliques, stabbed with an insect pin (in the case of 7 and 9 DPA seeds), and prepared for TEM by high-pressure freezing, freeze substitution and resin embedding according to Rensing et al. (2002). Samples were loaded into copper hats (Ted Pella, Redding, CA, USA) filled with 1-hexadecene and high-pressure frozen using a Bal29  Tec HPM 010 High Pressure Freezer (Balzers Instruments, Balzers, Leichtenstein). The hats were immediately transferred to frozen cryovials containing freeze substitution medium consisting of either 2% (w/v) osmium tetroxide in acetone with 8% (v/v) dimethoxypropane, for morphological assays, or 0.25% (v/v) glutaraldehyde, 0.1% (w/v) uranyl acetate in acetone with 8% (v/v) dimethoxypropane for immunolabeling assays. Freeze substitution was carried out for 4-6 days at -80°C by incubation in an acetone/dry ice slush, followed by transfer of the cryovials into a metal block pre-cooled to -20°C, which was warmed to room temperature over 20 hours to allow reaction of fixatives. The samples were removed from the sample holders, rinsed in anhydrous acetone several times and slowly infiltrated and embedded in either Spurr's resin (Spurr, 1969) for morphological examination or LR White (London Resin Company, Berkshire, UK) for immunological studies. Regardless of the resin used, infiltration was carried out initially by transferring a solution containing a single drop of resin/ ml of acetone to samples, and placed on a variable speed rotator for 1h at 3 rpm. This was continued until samples could be transferred into a solution containing 5 drops of resin/ ml of acetone. Samples were then transferred to 10% resin in acetone and left on the rotator overnight. On the second day, samples were transferred into 20%, 40% and 60% resin in acetone, with 2-3 hours between each transfer. Once the samples were in 60% resin they were left overnight. On the third day, samples were transferred into an 80% resin solution and left for 2-3 hours, followed by two 100% resin solutions. On the morning of the fourth day, samples were transferred to BEEM capsules (if embedding resin was Spurr's) or gelatin capsules (if embedding resin was LR White) and placed in a 50°C oven for a minimum of 16 hours.  2.2.3 Electron tomography Tomographic analysis of 7 DPA seed coat cells was done according to Donohoe et al. (2006), with some minor changes. Samples were cryofixed and embedded as described above for 30  morphological examination. Spurr’s embedded samples were sectioned 200nm thick on a Leica Ultracut T Ultramicrotome (Leica Microsystems GmbH, Wetzlar, Germany), and picked up on 0.75% (w/v) formvar coated copper/ rhodium slot grids (Ted Pella, Redding, CA, USA) and stained with 2% (w/v) uranyl acetate in 70% (v/v) methanol for 20-25 minutes, followed by Reynold’s lead citrate for 6-8 minutes. After staining, 15 nm unconjugated colloidal gold particles were added to both sides of the grid to be used as fiducial markers to align the series of tilted images. Tomograms were acquired at The Boulder Lab for 3D Electron Microscopy of Cells (University of Colorado, Boulder, CO) on an FEI Technai TF30 intermediate voltage EM (FEI, Hillsboro, OR, USA), operating at 300 KV. Tilt-series were acquired at 23 000x, from -60° to +60° at 1° intervals about two orthogonal axes (Mastronarde, 1997) using a Gatan Megascan 795 digital camera (Gatan, Pleasanton, CA, USA), giving a calculated pixel size of 1 nm. Dual-axis tomograms were generated using the Etomo software interface, a part of the IMOD software package (Kremer et al., 1996). Tomograms were displayed and analyzed with 3dmod, the graphics component of the IMOD software package. Golgi stacks, vesicles, and other structures were modeled manually, according to Donohoe et al. (2006).  2.3 Quantitative morphological techniques 2.3.1 Golgi stack and electron-dense vesicle density analysis To calculate the density of Golgi stacks (# Golgi stacks/μm2 cytosol) or electron-dense vesicles (# dense vesicles/m2 cytosol), thin sections were cut using a Leica Ultracut T Ultramicrotome (Leica Microsystems GmbH, Wetzlar, Germany), stained and examined using a Hitachi H-7600 Transmission Electron Microscope (Hitachi High-Technologies Canada, Toronto, ON). Two different stains were used to examine samples in TEM: 2% (w/v) uranyl 31  acetate in 70% (v/v) methanol for 10 minutes, followed by Reynold’s lead citrate for 2 minutes, or the polysaccharide stain, alkaline bismuth (0.05% NaOH, 0.02% sodium potassium tartrate, 0,01% bismuth subnitrate) for 45 minutes at 37°C (Park et al., 1987). These stains were used interchangeably, as they gave comparable results. However, mucilage and amyloplasts appear much darker using alkaline bismuth (compare figure 3-5, A and D to figure 3-5, B and C) The number of Golgi stacks (or electron-dense vesicles) visible in a given TEM section of a cell was counted, and images of the counted cells were collected. The cytosol was defined as all areas within the cytoplasm, except the nucleus, amyloplasts, and vacuole. The surface area of the cytosol was calculated using ImageJ (Rasband, W.S., U.S. N.I.H., Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij/, 1997-2006). The number of cells measured was chosen so that there were approximately equivalent amounts of cytosolic surface area measured at all three day stages. For Golgi stack density, this resulted in the examination of a total of 61 wild-type cells and 152 Golgi stacks at 4 DPA, 24 cells and 474 Golgi stacks at 7 DPA and 17 cells and 333 Golgi at 9 DPA. In the case of mum4, 46 cells and 115 Golgi stacks were examined at 4 DPA and 57 cells with 800 Golgi stacks at 7 DPA. For the survey of electron-dense vesicles, 10 cells and 761 vesicles were examined.  2.3.2 Statistical analysis Methodology for statistical analysis was determined on a case-by-case basis, according to Table 2-1. In all cases, data analysis was conducted with SPSS 13 software (SPSS, Inc., Chicago, IL). Throughout this thesis, the specific tests used in each case are listed, along with real and corrected p-values (Corrected using Bonferroni’s correction, in order to limit the chance of Type I errors). In all cases, results were deemed significant if p<0.05.  32  TABLE 2-1: Decision making process for choice of statistical tests to use in each analysis.1  Comparing 2 independent means: Comparing 2 dependent means: Comparing several independent means:  Kolgomov-Smirnov Test for Normality | | PASS FAIL   Parametric tests used: Non-Parametric tests used: T-test for independent means Mann-Whitney test (U)  T-test for repeated measures  Wilkoxon Signed Ranks test (Reported as a z-score)  Analysis of variance (ANOVA) Kruskal-Wallis test (H) Followed by post-hoc analysis Followed by Mann-Whitney test (Using Bonferroni’s correction) (Using Bonferroni’s correction)  1  In all cases, results were deemed significant if p<0.05. For the analysis of several means, Bonferoni’s correction was used (divide p-value by number of tests to be run) in order to avoid Type I errors.  2.4 Immunolabelling techniques for microscopy 2.4.1 Primary antibodies Previously described polysaccharide antibodies, which were used in this study, include JIM5 and JIM7 against methyl-esterified homogalacturonans (Knox et al., 1990), JIM8 (Pennell et al., 1991) and JIM13 (Knox et al., 1991) against arabinogalactan proteins, -RGI/PGA, a polyclonal raised against rhamnogalacturonan/ polygalacturonic acids and -XG, also polyclonal, raised against xyloglucan (Moore et al., 1986), PAM1 against unesterified homogalacturonans (Willats et al., 1999), LM5 against anti-(14)--D-galactan (Jones et al., 1997), LM6 against (15)--L-arabinan (Willats et al., 1998), LM7 against partially methylesterified homogalacturonan (Willats et al., 2001c), LM10 against xylans (McCartney et al., 2005), CCRC-M1 against fucosylated xyloglucan and CCRC-M2 and CCRC-M7 against RGI (Puhlmann et al., 1994). A previously published cellulose binding domain was also used, 33  conjugated to the fluorescent molecule Oregon Green (CBD-OG) (Jervis et al., 1997). Previously unpublished antibodies used include a set kindly donated by Michael Hahn of the Complex Carbohydrate Research Center (CCRC-M30, CCRC-M31, CCRC-M32, CCRC-M34, CCRCM36 and CCRC-M38), which were generated against extracted Arabidopsis seed mucilage (T. Bootten, Z. Popper, C. Deng, R. Jia, W.S. York, M.A. O’Neill, M.G. Hahn, in preparation). A further set of antibodies (CCRC-M48, CCRC-M54, CCRC-M57, CCRC-M58) was generated against tamarind seed xyloglucan (Z. Popper, T. Bootten, S. Tuomivaara, A.G. Swennes, R. Jia, W.S. York, and M.G. Hahn, in preparation). All of the unpublished antibodies are available from CarboSource (http://cell.ccrc.uga.edu/~carbosource/ CSS_home.html). Other primary antibodies include anti-tubulin  antibodies from Sigma-Aldrich Canada Ltd (Oakville, ON) and anti-epitope tag antibodies. Anti-epitope tag antibodies were from several commercially available sources. Four anti-His antibodies were tried, including three (RGS-His, Tetra-His and Penta-His) from Qiagen Canada (Mississauga, ON) and one (SigmaHis) from Sigma-Aldrich Canada Ltd (Oakville, ON). Antibodies to cMyc were produced by Molecular Probes (Invitrogen Corporation, Carlsbad, CA, USA). Unless otherwise indicated, all antibodies described in this thesis are monoclonal.  2.4.2 Intact whole seed and mucilage immunolabelling For whole seed immunolabeling, small batches (10-50) of unfixed whole seeds were placed in 1.5 ml microcenterfuge tubes. Seeds were incubated in 5% (w/v) non-fat dry milk (NFDM) in Tris-buffered saline/ 0.1% (v/v) Tween 20 (TBST) for 20 min with gentle agitation on an orbital shaker to hydrate mucilage and block any non-specific protein binding. Samples were then incubated at room temperature for 1 h with primary antibodies at 1:10 (v/v) dilutions in 1% (w/v) NFDM in TBST, and for 1 h in secondary antibodies at 1:100 (v/v) dilution. Secondary antibodies were anti-mouse, -rat or -rabbit (depending on the primary antibody) 34  conjugated to Alexafluor 594 (Invitrogen Corporation, Carlsbad, CA, USA). Negative control experiments were performed under the same conditions, with the omission of the primary antibody, or both the primary and secondary (to examine autofluorescence of whole seeds). Seeds were rinsed before and after antibody incubations by washing 5 times in TBST. Samples were mounted in dilute [1:125 (v/v)] India Ink and examined via epifluorescence using a Leica DMR light microscope (Leica Microsystems GmbH, Wetzlar, Germany).  2.4.3 Immunofluorescence of thick sections For examination of immunofluorescence of seed coat epidermal cells, 0.3-0.5 μm LR White sections were mounted on 10-well, Teflon-coated microscope slides. Non-specific protein binding was blocked by incubating slides in Coplin jars filled with of 5% (w/v) NFDM in TBST for 20 min. Samples were then incubated at room temperature for 1 h with primary antibodies at 1:20 (v/v) dilutions in 1% (w/v) NFDM in TBST, and for 1 h in secondary antibodies (Alexafluor 594, Invitrogen Corporation, Carlsbad, CA, USA) at 1:100 (v/v) dilution. Control experiments were performed to test specific binding of primary antibodies CCRC-M36 and XG by pre-incubating the antibodies with their specific purified antigen to block antibody binding sites. Antigen for CCRC-M36 was mucilage that had been extracted from wild-type seeds by vortexing whole seeds, and the antigen for anti-XG was 2 mg/ml of tamarind xyloglucan (Megazyme International Ireland, Ltd, Bray, Ireland). In both cases, pre-incubation of primary antibody with excess antigen completely blocked binding to the sections (Figure 4-4). Rinses were performed in Coplin jars with TBST before and after incubations. Samples were mounted in 90% (v/v) glycerol in water and examined via epifluorescence using a Leica DMR light microscope fitted with long pass filters (Leica Microsystems GmbH, Wetzlar, Germany).  35  2.4.4 Immunogold labeling for TEM LR White samples were cut into 70 nm sections and mounted on 200 mesh, fine bar, nickel grids (Ted Pella, Redding, CA, USA) that had been coated with formvar (0.3% (w/v) in 1,2-dichloroethane). Non-specific protein binding was blocked by floating grids on 10 l drops of 5% (w/v) NFDM in TBST on parafilm for 20 min. Excess solution was blotted off and grids were transferred to a drop of primary antibody, diluted 1:5 (CCRC-M36) or 1:20 (-XG) with 1% (w/v) NFDM in TBST, and incubated for 1 h at room temperature. After washing grids in three subsequent washes of TBST for 15 seconds each, grids were transferred to secondary antibodies, and diluted 1:100 with 1% (v/v) NFDM in TBST for 1 h. Secondary antibodies were goat anti-mouse IgG + IgM (for CCRC-M36) or goat anti-rabbit IgG (for -XG), conjugated to 10 nm colloidal gold (both from Ted Pella, Inc. Redding, CA, USA). Grids were washed again, followed by three washes of distilled water for 15 seconds each, and then post-stained with 2% (w/v) uranyl acetate for eight minutes, and Reynold’s lead citrate for two minutes. For the double label (i.e., CCRC-M36 and -XG), 15 nm colloidal gold conjugated to goat anti-mouse IgG + IgM was used to locate CCRC-M36. Primary or secondary antibodies were mixed together and treated as described above. To determine the % gold-labeled Golgi stacks (Chapter 4), Golgi stacks were counted manually on TEM images. Total counts were 31 cells and 287 Golgi stacks for 7 DPA cells labeled with CCRC-M36, 28 cells and 252 Golgi stacks for 7 DPA cells labeled with -XG, 23 cells and 287 Golgi stacks for 9 DPA cells labeled with CCRC-M36, and 20 cells and 282 Golgi stacks for 9 DPA cells labeled with -XG. For the double label, 57 Golgi stacks were examined in 5 different cells.  36  2.4.5 Immunofluorescent labeling of whole mount, developing seeds for intracellular observations with confocal microscopy Immunofluorescent imaging was done according to Collings and Wasteneys (2005), with slight modifications. Developing seeds at 7 and/ or 9 DPA were staged, dissected from siliques, scored with a razor blade to increase accessibility of inside tissues to fixatives and placed in 1.5 ml microcentrifuge tubes with fixative [0.5% glutaraldehyde, 1.5% formaldehyde in PEMT (0.05% Triton X-100 in 50 mM PIPES buffer, 2 mM EGTA and 2 mM MgSO4)] for 30 min to 1 h. Samples were then washed several times in PEMT, and transferred to wall digestion solution [0.1% pectinase, 0.1% cellulase (Sigma-Aldrich Canada Ltd, Oakville, ON)], 0.4 M sorbitol and 1% bovine serum albumin in PEM buffer (50 mM PIPES buffer, 2 mM EGTA and 2 mM MgSO4)] for 1.5 h. Samples were once again washed several times in PEMT, transferred to a cryoprotectant solution (0.25 M sorbitol in PEMT), squashed gently by applying pressure to seeds in micocentrifuge tubes with a mortar, and placed on dry ice for 5 min. Samples were removed from dry ice, mortars were removed and samples transferred to permeabilization buffer (1% Triton X-100 in phosphate buffered saline) for 2 h. After permeabilization, samples were washed in phosphate buffered saline, blocked for 30 min in blocking buffer (1% bovine serum albumin in phosphate buffered saline and 0.05 M glycine) prior to overnight incubation with primary antibody [anti-tubulin  at 1:1000 (v/v) or Tetra-His at 1:20 (v/v)] at 4°C on an orbital shaker at 150 rpm. The following morning, samples were washed in phosphate buffered saline and 0.05 M glycine several times, and incubated in secondary antibody conjugated to Alexafluor 488 (Invitrogen Corporation, Carlsbad, CA, USA) for 3 h at 37°C on an orbital shaker at 150 rpm. After secondary antibody incubation, samples were washed and mounted in Prolong Gold antifade mounting medium (Invitrogen Corporation, Carlsbad, CA, USA). Samples were imaged in spectral imaging mode using a Zeiss 510 Meta confocal laser scanning head mounted on an 37  Axiovert 200 M inverted microscope (Carl Zeiss Microimaging GmbH, Munich, Germany). Collected stacks were prepared for presentation in this thesis using the Volocity software package (Improvision, Inc. Waltham, MA, USA).  2.5 Cloning and expression of MUM2-His 2.5.1 Epitope tagging of MUM2 Genomic MUM2 was cloned previously by Dr. Gillian Dean into pCambia1200 (http://www.cambia.org/ daisy/cambia/home.html) to form the pMUM2g vector (Dean et al., 2007). A 6XHistidine (6xHis) tag was added to the 3' end by PCR. Due to the location of the unique restriction enzyme site nearest the 3' end (ApaI) the final 370 bp were amplified by PCR, with a forward primer upstream of the ApaI site (m2hisF: 5'-CCTAATGGAGATGGGCCC-3') and a reverse primer at the 3' end that coded for the histidine tag directly upstream of the stop codon, together with a second ApaI site and an XbaI restriction enzyme site for removal of duplicated 3’end of the gene  (mum2hisR: 5'-ATGGGCCCTCTAGATCAATGATGATGA  TGATGATGGGAGAATTGAGATTGAGCTT-3'). pMUM2g and the amplified insert were digested with ApaI (Invitrogen, Carlsbad, CA, USA). pMUM2g was gel purified using a QAIEX II Gel Extraction Kit (Qiagen Canada, Mississauga, ON) while the insert was purified using a PureLink PCR Purification Kit (Invitrogen, Carlsbad, CA, USA). The insert and vector were ligated using T4 DNA Ligase (Invitrogen, Carlsbad, CA, USA) to produce the vector, pM2H, and then transformed into DH5 chemically competent E. coli cells (Invitrogen, Carlsbad, CA, USA). Transformed colonies were selected on LB medium supplemented with chloramphenicol. Identity of colony and correct orientation of insert was confirmed by PCR, followed by sequencing of the insert to ensure that no errors had been introduced during PCR amplification (sequencing primer sequences are shown in Figure 2-1). Both pM2H and the original vector, 38  pMUM2g, were purified and transformed into Agrobacterium tumefaciens strain LBA4404 (Hoekema et al., 1983) by electroporation. mum2-1 plants were transformed by the floral dipping method (Clough and Bent, 1998), slightly modified by spraying transformation medium onto plants, using A. tumefaciens carrying either pM2H or pMUM2g as a control. Hygromycinresistant strains were identified by growth of T1 plants on AT medium supplemented with 25 μg/ml Hygromycin B (Invitrogen, Carlsbad, CA, USA).  2.5.2 Immunoblotting to test for 6xHis labeling 2.5.2.1 Yeast controls Strains of Pichia pastoris that carried a methanol-inducible pPIC9 vector with either the MUM2 coding sequence, or a negative control bearing the coding sequence for albumin, had been produced previously by Dr. Gill Dean (Dean et al., 2007) and were used as a control for all immunoblots in this thesis. Cultures of each line were transferred from YPD plates to Buffered Glycerol Complex Medium (BMGY) medium and grown at 30°C until optical density at 600 nm (OD600) was between 2 and 6. Liquid cultures were transferred to Buffered Methanol Complex medium (BMMY) medium for methanol induction and grown for 3 days at 30°C. Once methanol induction was complete, cultures were centrifuged to separate supernatant from cell pellet. Supernatants (called M2H-s or GS115-s) were flash-frozen and stored at -80°C until dot blots could be carried out. Crude cell extracts were prepared from cell pellets by resuspension of cells in Breaking Buffer (50 mMNaPO4, 1 mM PMSF, 1 mM EDTA and 5% glycerol. 100 l Breaking Buffer per 1 ml of original liquid cell culture). An equal volume of 0.5 mm acid washed glass beads (Sigma-Aldrich Canada Ltd, Oakville, ON) was added to cell suspensions and then suspensions were passed through 8 cycles of vortexing for 30 sec, followed by 30 sec on ice. Suspensions were centrifuged at maximum speed (4500 rpm) for 10 min to separate cell 39  debris and glass beads from cell lysates. Lysates (labeled M2H-p and GS115-p) were flashfrozen at -80°C to store until immunoblots were carried out. For immunoblots described below, approximately 3-5 l of each solution was applied to nitrocellulose.  2.5.2.2 Leaf press immunoblot This protocol was based very roughly on work described by Varner and Taylor (1989). Leaves were pressed onto nitrocellulose using the rounded end of a 1.5 ml microcentrifuge tube, with a layer of parafilm to protect samples. The nitrocellulose membrane was allowed to completely dry, and then blocked in blocking buffer (5% bovine serum albumin in TBST). Membranes were sealed in Glad Press-and-Seal bags with 10 ml of primary antibody in blocking buffer [Penta-His, 1:2000 (v/v)] and placed on an orbital shaker at room temperature for 1 h. After washing samples in TBST, secondary antibodies were applied in the same fashion as primary antibodies. Secondary antibodies were IgG + IgM anti-mouse antibodies conjugated to 10 nm gold (Ted Pella, Inc. Redding, CA, USA). Secondary antibodies were detected using Ted Pella’s silver enhancement kit for detection of gold particles (Ted Pella, Inc. Redding, CA, USA).  2.5.2.3 Leaf extracts For extraction of leaf cell contents for dot blotting, Arabidopsis leaves were crushed using a mortar and pestle for 30 sec. Tris-buffered saline (40 l) was added, and samples were frozen at -20°C in order to help disrupt cells. 3-5 l of each extract was added directly to nitrocellulose, followed by dot blotting procedure described below (Section 2.5.2.4).  40  2.5.2.4 Dot blots This protocol is based on two (Protocol 5 and 8) found in the QIAexpress Detection and Assay Handbook (Qiagen, Inc. October 2002). Strips of nitrocellulose were pre-wet in Trisbuffered saline, excess liquid was allowed to dry off and 2-5 l of each sample (P. pastoris and leaf extracts, described in Section 2.5.2.1 and 2.5.2.2, respectively) were applied directly to nitrocellulose. Nitrocellulose strips were blocked in 3% bovine serum albumin in Tris-buffered saline, washed twice in TBST for 10 min, and once in Tris-buffered saline for 10 min, followed by primary antibody incubation using either Penta-His, RGS-His or Tetra-His antibodies (Qiagen, Inc. Mississauga, ON. All raised in mouse, see description in Section 6.2.2.1 and Table 6-1) at 1:1000 (v/v) in Glad Press-and-Seal bags for 1 h. After washing again in TBST followed by Tris-buffered saline samples were incubated in secondary antibody in the same way as primary antibodies. Secondary antibodies were anti-mouse antibodies conjugated to alkaline phosphatase (Sigma-Aldrich Canada Ltd, Oakville, ON) and were used at a dilution of 1:1000 (v/v) in 3% bovine serum albumin in Tris-buffered saline. Chromogenic development of secondary antibody signal was done according to instructions in the NBT/BCIP kit (Roche Diagnostics, GmbH, Mannheim, Germany).  41  Name m2hisF mum2hisF mum2seqF mum2seq2F mum2seq3F F6  Sequence 5’-CCGAATGGAGATGGGCCC-3’ 5’-CCGATTGGAGATGGGCCCGT-3’ (Error highlighted. Should be A, as in m2hisF) 5’-CTCGCTTGGTACAAGGTATGC-3’ 5’-TGTGAAGGTGGGTTTGCTTGGAGA-3’ 5’-GTGTCACATGTAAAGAAACTTC-3’ 5’-GGACTCTGGAGCTTACATGG-3’  Legend: RED – stop codon in original gene sequence Blue – insert bearing 6xHis (6xHis is underlined) BLUE – inserted stop codon Green – Xba1 restriction enzyme site Purple – Apa1 restriction enzyme site Grey – original 3’UTR 5’-...gtaagacgtttactgatactagaaccctgtttattaagtgagtagatagaatcataacatttttttcaaccttaataataaaagGACTCTGGA GCTTACATGGAGAGGAGATCTTATGGACTAACCAAAGTACAAATTAGCTGTGGCGGGACAAAACCCA TCGATTTGAGTAGATCTCAATGGGGTTACTCGgtaatatgatgaataagcctttcattttatgagtgatggaagattgtgttgatac ataacctagtatgtatatgtgttgtgaagGTGGGTTTGCTTGGAGAGAAAGTTAGGTTATATCAATGGAAGAATTTAA ATAGAGTGAAGTGGAGTATGAATAAAGCTGGACTAATTAAGAACCGCCCCCTCGCTTGGTACAAGgtat gcctatttacatatgtaaataagtactatatgtgtaatttatgaatagtatcggtgacgaatgtgaatatgcagACAACGTTTGATGGGCCG AATGGAGATGGGCCCGTGGGCCTACACATGTCGAGTATGGGTAAAGGAGAAATTTGGGTGAATGGT GAAAGCATTGGTCGTTATTGGGTCTCGTTTCTCACCCCTGCTGGACAACCTTCTCAATCTATgtaagtgtc acacacacacctggaaacacaaaattatgtttgttgttactttttgatatggaattggatttggttaatgcagATATCATATTCCTCGGGCGT TCTTAAAACCATCTGGAAATTTATTGGTGGTATTTGAAGAGGAAGGTGGTGATCCTCTTGGGATATCT TTGAATACAATATCGGTTGTTGGTTCGAGTCAAGCTCAATCTCAATTCTCCCATCATCATCATCATCAT TGATCTAGAGGGCCCGTGGGCCTACACATGTCGAGTATGGGTAAAGGAGAAATTTGGGTGAATGGT GAAAGCATTGGTCGTTATTGGGTCTCGTTTCTCACCCCTGCTGGACAACCTTCTCAATCTATgtaagtgtc acacacacacctggaaacacaaaattatgtttgttgttactttttgatatggaattggatttggttaatgcagATATCATATTCCTCGGGCGT TCTTAAAACCATCTGGAAATTTATTGGTGGTATTTGAAGAGGAAGGTGGTGATCCTCTTGGGATATCT TTGAATACAATATCGGTTGTTGGTTCGAGTCAAGCTCAATCTCAATTCTCCTGAagattatttttttaaaaatggtttt tgggtaattgggggtgttttataaaccaaacaatacattatattgaaatgatacaatacacaaattgaatgaaatatacaggtttcataaatcttagtgtc acatgtaaagaaacttcagatatatcagtttcacaaatttcgtttaatttgctcaagtactcaaaatattcaaaagatgcaaaatactagaaaaaagag agatccaccgatcacgctcgtatcccaagtacgaagggtccagtctagagcctggctttgttggttcggcgaatccagttggctcttcaccgacacca aattggcaaacactttcaacgcgccgtttaattgttgaagcgatattcgttgtatatcacatgcgtatctgagaaagtgtggctcacgctccaatag...-3’  FIGURE 2-1: Sequencing primers for confirmation of orientation of insert in MUM2 to add 6xHis tag. Primers are identified in the table at the top of the page, and located in the gene sequence below. Sequence of MUM2 shown here begins at the start of intron 12 and ends just prior to the end of the 3’UTR.  42  3 ULTRASTRUCTURE OF THE GOLGI APPARATUS IN THE ARABIDOPSIS SEED COAT* 3.1 Introduction The secretory system of plants is integral to the synthesis and deposition of many cell wall components, including pectin (Doblin et al., 2003). The plant Golgi apparatus is comprised of large numbers of individual Golgi stacks, which are widely distributed throughout the cytoplasm and are capable of streaming, a process mediated by the actin cytoskeleton (Griffing, 1991; Boevink et al., 1998; Nebenführ et al., 1999). Each individual Golgi stack is made of several cisternae that are arranged in a polar fashion from cis to trans. Directly apposed to the trans-most cisternae is a tubulo-vesicular network known as the trans Golgi Network (TGN). In all eukaryotes, the role of the Golgi apparatus is to modify, sort and package proteins that are destined for the lysosome, vacuole, or plasma membrane. The Golgi apparatus of plants has an additional, and possibly more important role, as it is the major site of production of matrix polysaccharides for the plant cell wall (Doblin et al., 2003). Plant cells that are in a developmental stage that requires active secretion of cell-wall polysaccharides have been identified in electron microscopy by the particular morphology of their Golgi stacks. This morphology includes a more prominent cis to trans polarity, increased swelling at the margins of the cisternae, a more prominent TGN, and many vesicles or vesicular clusters associated with the Golgi stack (Staehelin et al., 1991; Mollenhauer and Morre, 1994). In addition to changes in the  A version of this chapter has been published: • Young, R.E., McFarlane, H.E., Hahn, M.G., Western, T.L., Haughn, G.W. and Samuels, A.L. (2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized secretion of pectin-rich mucilage. Plant Cell 20: 1623-1638. • McFarlane, H.E., Young, R.E., Wasteneys, G.O., and Samuels A.L. (2008). Cortical microtubules mark the mucilage domain of the plasma membrane in Arabidopsis seed coat cells. Planta 227: 1363-1375. 43  morphology of the Golgi apparatus during development, the number of Golgi stacks in the cell has also been observed to change over time (Garcia-Herdugo et al., 1988; Segui-Simarro and Staehelin, 2006). Unfortunately, factors affecting the number of Golgi stacks have not been studied in detail to date. It is possible that one of the main reasons for this is because a suitable experimental system, where polysaccharide production is strongly upregulated in a defined developmental context has not been available. The epidermal cells of developing Arabidopsis seed coats provide such an experimental system in which to characterize changes in Golgi stack ultrastructure and number. Developing Arabidopsis seeds grow rapidly for the first few days after fertilization. Mucilage production in the seed coat epidermis is initiated around 6 or 7 DPA, and the newly synthesized mucilage components are deposited in a donut-shaped pocket between the plasma membrane and the primary cell wall (Figure 1-2 and Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000). Coincident with the deposition of mucilage is the formation of a volcano-shaped cytoplasmic column. By about 10 DPA, mucilage production is completed and a secondary cell wall is deposited around the cytoplasmic column. At maturity the seed coat epidermal cells consist of a columella made exclusively of secondary cell wall components, with a donut shaped mucilage pocket below the primary cell wall on the outer side of the cell (Figure 1-2 and Beeckman et al., 2000; Western et al., 2000; Windsor et al., 2000). Mutations in the MUCILAGE MODIFIED 4 (MUM4) gene, encoding a putative rhamnose synthase, result in a significant decrease in seed coat mucilage. The morphology of the seed coat epidermal cells is also altered, including a significant reduction of the columella (Usadel et al., 2004; Western et al., 2004). As the epidermal cells of the Arabidopsis seed coat undergo distinct morphological changes during development, which coincide with very obvious and abrupt changes in the levels of polysaccharide secretion (Figure 1-2 and Beeckman et al., 2000; Western et al., 2000; Windsor et 44  al., 2000) and mutants affecting the production of mucilage are available, this cell type is well suited to the study of changes in Golgi stack number during development.  3.1.1 Previous and related work Prior to the work described in this thesis, some preliminary ultrastructural work had been done by light microscopy and TEM (Western et al., 2000, Diana Young, unpublished data). Starting from 6 DPA, the growing mucilage pocket could be seen at the outer edges of the cell, with the developing cytoplasmic column in the middle. Below the column, a large vacuole occupied the majority of the space. Golgi stacks were observed throughout the cells, but most prominently in the cytoplasmic column. Structures resembling cortical microtubules were seen lining the cytosolic side of the mucilage pocket, but not in adjacent areas where the plasma membrane remains pressed against the primary cell wall. Diana Young, as a member of Lacey Samuels’ lab, made many of the original observations of the cortical microtubules. Heather McFarlane (Samuels lab) expanded upon Diana’s work as an undergraduate under my supervision. In addition, Heather found that in the cytoplasm of the seed coat, there are unusual cytoskeletal bundles, which in longitudinal views show filamentous structures that resemble microtubules, but did not label with antibodies to - or -tubulin. When viewed in cross-section, the filaments appear hollow, although the 13 protofilament arrangement, characteristic of microtubules, is not evident (McFarlane et al., 2008). During this study, Heather McFarlane, now as a member of Tamara Western’s lab at McGill University, was involved with some of the data collection. Using a trans Golgi marker Sialyl Transferase conjugated to Green Fluorescent Protein (ST-GFP), driven by the constitutive 35S promoter (Boevink et al., 1998), Heather performed live-cell imaging experiments to provide further evidence of the changes in number of Golgi stacks during development. She 45  examined cells expressing ST-GFP at different stages of development, counted the Golgi stacks in those cells, and then sent the data to UBC for statistical analysis. This data was included in the publication of this work (Young et al., 2008) and is discussed in Section 3.2.2.  3.1.2 Objectives In this chapter I examine how the scattered stacks of the plant Golgi apparatus respond to a surge in polysaccharide production, both at the level of the individual stack and collectively. I focused on three particular stages of seed coat development: prior to the onset of mucilage production, at 4 days post-anthesis (DPA); during the stage of intense mucilage production, at 7 DPA; and after mucilage production, when the secondary cell wall of the columella is being secreted, at 9 DPA. In addition, I examined how the number of Golgi stacks within the seed coat epidermal cells changes during these three stages. Finally I compared the Golgi apparatus in wild-type seeds to those of mucilage-modified 4 (mum4), a mutant with decreased levels of mucilage in its seed coat epidermal cells.  3.2 Results 3.2.1 The Arabidopsis seed coat and Golgi apparatus undergo distinct ultrastructural changes during development An overview of seed coat epidermal cells during differentiation was undertaken to examine the organellar rearrangements that occur in the whole cell and to put polysaccharide production and secretion into this context. Arabidopsis seeds were excised from siliques and high pressure frozen before (at 4 DPA), during (at 7 DPA) and after (at 9 DPA) the mucilageproducing stages of seed coat development and the ultrastructure, specifically in the Golgi apparatus, was examined.  46  3.2.1.1 The seed coat and Golgi apparatus at 4 DPA Prior to the mucilage-secreting phase (4 DPA), epidermal cells had a large central vacuole, a nucleus typically located at the basolateral side of the cell, and amyloplasts (containing starch granules) that were either basal or apical (Figure 3-1, A-B). As expected, these cells had no obvious mucilage pocket development between the plasma membrane and the primary cell wall at the outer corners of the apical side of the cells. Golgi stacks were seen in the narrow cytosolic region surrounding the vacuole. Cisternae were long and relatively thin, and the trans Golgi network (TGN) was small, compared to other stages of development (Figure 3-2, AB). In many cases, the Golgi stack appeared cup-shaped, with the TGN on the concave surface (Figure 3-2A).  3.2.1.2 The seed coat and Golgi apparatus at 7 DPA During mucilage secretion (7 DPA), seed coat epidermal cells had well-developed, ringshaped mucilage pockets between the plasma membrane and primary cell wall, with the vacuole limited to the lower portion of the cell (Figure 3-1, C-D). The nucleus and starch granules were positioned in the cytoplasmic column, with the nucleus usually found higher in the column than the starch granules. Cortical microtubules were seen lining the membrane on the cytosolic side of the mucilage pocket. These microtubules were highly concentrated in area of the apical region apposed to the mucilage pocket, but not at the apex of the cytoplasmic column, or in any of the basolateral region of the cell (McFarlane et al., 2008). Vesicles were sometimes observed in close proximity to these microtubules on the cytoplasmic side, but never between the microtubules and the plasma membrane. Some of these vesicles had an electron dense cargo, and a particular morphology (This will be discussed in detail in Chapter 5). The Golgi stack morphology at 7 DPA (Figure 3-2, C-D) differed from 4 DPA in that the stacks had a shorter diameter than those observed at 4 DPA, and the cis-trans polarity across the stack was more 47  pronounced. The lumens of the cis cisternae were wider while the medial and trans cisternae had compressed lumens, like those at 4 DPA, except with swollen margins (compare Figure 3-2, C-D with 3-2, A-B). In addition, there appeared to be more TGN as well as more large vesicles associated with the TGN. Considering that a large, convoluted TGN might be difficult to capture in thin sections, as connections between portions of the TGN might be outside of the plane of section, electron tomography was used to examine the morphology of the 7 DPA Golgi stack in greater detail. Models that had been built from three different tomograms showed that much of what appeared to be free vesicles near the TGN in thin sections were interconnected vesicular clusters (Figure 3-3). During the intense secretion of mucilage that is characteristic of 7 DPA seed coat epidermal cells, the swollen vesicular regions of the TGN lacked obvious coats and were large, about 200 nm. This is morphologically distinct from the partially coated reticulum, an interconnected cluster of coated vesicles, with dilations that are approximately half the size of those described here (Pesacreta and Lucas, 1985). The clusters in this study appeared to be budding from flattened cisternal domains, suggesting that these sites represent the exit site of the Golgi stack (Figure 3-3C). The TGN interconnected vesicular clusters observed were immediately adjacent to the trans-most cisterna, in agreement with the results of Segui-Simarro et al. (2006) where all TGN and Golgi stacks in meristematic cells were tracked.  3.2.1.3 The seed coat and Golgi apparatus at 9 DPA After the secretion of mucilage ended (9 DPA), seed coat epidermal cells had a distinguishable layer of secondary cell wall lining the apical side of the cell between the mucilage pocket and the plasma membrane and along the top of the cytoplasmic column (Figure 3-1, E-F). The vacuole was still located at the basal side of the cell but appeared smaller in volume. Microtubules around the mucilage pocket were not obvious at this stage. Instead a large 48  number of vesicles, including some with electron-dense cargo described above, were often observed at the apex of the cytoplasmic column, sometimes with vesicle fusion also being observed (This will be discussed in more detail in Chapter 5) Golgi stacks were similar to those seen in 7 DPA cells, exhibiting short, compressed cisternae with swollen margins and extensive TGN (Figure 3-2, E-F).  3.2.2 The density of stacks in the seed coat Golgi apparatus also changes during development In order to determine whether the onset of mucilage production resulted in changes in the number of individual stacks in the Golgi apparatus, the number of Golgi stacks in 4, 7 and 9 DPA cells were quantified. This was done using a two-fold approach. First, high-resolution TEM of seed coat epidermal cells at each developmental stage was undertaken in order to count Golgi stacks in seed coat epidermal cells directly. When thin sections of seed coat epidermal cells were examined in TEM, an increase in the number of Golgi stacks was observed (Figure 3-4A). However, during the same period (between 4 DPA and 7 DPA) seed coat epidermal cells underwent dramatic cytosolic rearrangements (Figure 3-1 compare A-B with C-D), which resulted in an increase in the surface area of cytoplasm visible in thin sections between 4 and 7 DPA (Figure 3-4B). Thus, the surface area of cytoplasm available for quantification at 4 DPA was significantly less than at 7 DPA, which could have resulted in an under-representation of the true number of Golgi stacks present in 4 DPA cells. For this reason we determined the density of Golgi stacks (# Golgi stacks/m2) visible in our sections by dividing the absolute number of stacks by the cytoplasmic surface area (Figure 3-4C). Non-parametric, three-way analysis of Golgi stack density at 4, 7 and 9 DPA showed statistically significant differences (KruskalWallis test; H(2)= 66.94, p<0.05). Post-hoc analysis using the Mann-Whitney test (U) showed a significant increase in the density of Golgi stacks at 7 DPA, compared to 4 DPA (U=326.0, 49  p<0.025). The number of Golgi stacks/m2 approximately doubled between 4 DPA and 7 DPA (0.08 stacks/ um2 versus 0.14 stacks/ um2, respectively). In contrast, when the density of Golgi stacks in 7 and 9 DPA seed coat epidermal cells were compared, no significant differences were found (U=152.0, p>0.025). This demonstrates that an increase in the number of Golgi stacks found in seed coat epidermal cells is correlated with the increase in polysaccharide production at the onset of mucilage secretion. A second method that was used to quantify Golgi stacks was live imaging of Golgi stacks in whole seed coat epidermal cells, using the transgenic Arabidopsis plants bearing the trans Golgi marker Sialyl Transferase conjugated to Green Fluorescent Protein (ST-GFP), driven by the constitutive 35S promoter (Boevink et al., 1998). Data collection from confocal images was done by Heather McFarlane in Tamara Western's Lab at McGill University (Young et al., 2008), prior to being sent to UBC for analysis. In cells which exhibited fluorescence, punctate structures of the correct size and shape as Golgi stacks were observed (Young et al., 2008). The density of Golgi stacks approximately doubled from 4 DPA (24 ± 9 Golgi stacks/ cell, n=33 cells) to the mucilage secretion phase at 7 DPA (49 ± 14 Golgi stacks/ cell, n=42 cells). This was considered significant (Mann-Whitney U=65.0, p<0.025). Together, these two different techniques provide complementary information: analysis by TEM allows Golgi stacks to be positively identified and the cytosol is sampled in square microns, providing a statistical sampling of the Golgi stack density. The fluorescence data can directly report the number of ST-GFP positive puncta in a whole cell. Both techniques showed an approximate doubling of Golgi stacks that correlated with increased mucilage secretion.  50  3.2.3 Decreases in mucilage production affects the morphology, but not the number of Golgi stacks in seed coat epidermal cells. The data in the previous section demonstrates that changes in Golgi stack number and morphology occur throughout the cell during mucilage production, but do not determine whether the changes are in response to increases in the pectin product or due to the intrinsic developmental program that seed coat epidermal cells undergo. Plants with loss-of-function mutations in MUM4 have seed coat epidermal cells with significantly reduced amounts of mucilage (Usadel et al., 2004; Western et al., 2004). Ultrastructural analysis of mum4 cells at 7 DPA by TEM confirmed previous results (Figure 3-5). Comparison of the Golgi apparatus of the wild-type and mum4 seed coat epidermal cells at the mucilage producing stage (7 DPA) allowed us to determine what happens to Golgi stack structure and numbers when the surge of polysaccharide product is drastically reduced. As documented earlier (Figures 3-2 and 3-3), wild-type Golgi stacks that are producing mucilage have short cisternae with compressed lumens and swollen margins, as well as a convoluted TGN. In comparison, the Golgi cisternae of the stacks observed in mum4 cells at 7 DPA were longer, with more open lumens and thin, fenestrated margins (Figure 3-6). Strikingly, they had a less complex TGN, lacking the elaborate clusters of vesicles/swollen networks seen in 7 DPA wild-type cells. In this regard, the mum4 Golgi stacks examined at 7 DPA had cisternae that were more similar in appearance to those of wild-type cells at 4 DPA (Compare figure 3-2, A-B with 3-6). The similarities and differences in Golgi apparatus morphology between wild-type and mum4 at 7 DPA raised the question of whether the increased density (# Golgi stacks/m2) of Golgi stacks seen in wild-type cells between 4 and 7 DPA would also be present in mum4. If the production of mucilage drives Golgi stack proliferation, then comparison of 7 DPA wild-type and mum4 seed coat epidermal cells during intense mucilage production would be predicted to 51  reveal a lower density of Golgi stacks in the mum4 cells. In contrast, if Golgi stack proliferation is an intrinsic component of the development of the seed coat epidermal cells, then the mum4 plants would be expected to display the same increases in Golgi stack density as wild type during mucilage production. High-resolution electron microscopy was once again employed to examine differences in Golgi stack density between wild-type and mum4 seeds at 4 or 7 DPA (Figure 37). Despite the strong reduction in the amount of mucilage produced by these cells at 7 DPA, the density of Golgi stacks in mum4 cells at 7 DPA was not significantly different from the same developmental stage in the wild-type seed coat (Figure 3-7C) (Mann-Whitney U=601.5, p>0.05), indicating that even with a large reduction in the amount of secretory product in the mutant, Golgi stack proliferation occurred between 4 and 7 DPA.  3.3 Discussion 3.3.1 Prolific mucilage production correlates with distinct morphological changes in the Golgi and TGN In the present study, high levels of mucilage secretion were correlated with a specific Golgi stack morphology, which includes flattened cisternae with swollen, fenestrated margins and a complex TGN. From cis to trans, the cisternal lumen becomes increasingly flattened and the margins of the cisternae become increasingly swollen. This is consistent with what has been documented in other systems that are active in cell wall matrix production, such as the mucilageproducing root cap cells. As the meristematic cells differentiate into the specialized and highly secreting cells of the root cap, changes in the morphology of the Golgi apparatus, such as increasingly swollen cisternal margins and increased cis/ trans polarity reflect its increased importance in the production of the pectinaceous mucilage of the root cap (Craig and Staehelin, 1988; Staehelin et al., 1990). The correlation between morphology and secretion is further 52  strengthened by the fact that the mutant, mum4, is both unable to produce a high level of seed coat mucilage and lacks this characteristic Golgi stack morphology. A hypothesis has been put forward which attempts to explain Golgi stack morphology in the context of biosynthetic function. Based on freeze-fracture data, it has been postulated that the bulk of the biosynthetic enzymes are maintained within the flattened central portion of the Golgi stack, while the nascent polysaccharide chains are pushed outwards, to form the swollen margins of the cisternae, and concentrate the polysaccharide product for vesicle transport (Staehelin et al., 1990). If this model were correct, the decreased amount of mucilage that is being produced in mum4 would explain why the margins of cisternae do not appear swollen and the TGN is decreased. It would be interesting to examine if mum4 Golgi stacks show similar results to wild type in freeze fracture, which would imply that the amount of biosynthetic enzymes embedded within in the cisternae are similar. During mucilage production, mucilage is produced in large amounts by the Golgi apparatus and must be packaged at the trans-face of the Golgi stack. Tomography demonstrates that the most likely exit site for mucilage, the TGN, consists of interconnected vesicular clusters. That this exit site is a network, rather than free vesicles, suggests that there must be subsequent fission and/or budding steps to give rise to free secretory vesicles. The clustered vesicles of the TGN appear to mature into secretory vesicles that are electron dense when observed by TEM and which move to the cortical cytoplasm, where they can be observed closely associated with the abundant cortical microtubules below the mucilage pocket (McFarlane et al., 2008). This is unlike the peripheral root cap cells’ secretion, where very large and bulbous margins of the Golgi stacks produce equally large secretory vesicles that are observed to condense during their migration to the cell surface. This is believed to occur by a process of clathrin-coated membrane  53  retrieval (Mollenhauer et al., 1991; Mollenhauer and Morre, 1994). The post-Golgi traffic of mucilage-containing vesicles will be discussed in greater detail in Chapter 5.  3.3.2 Biological control of Golgi stack proliferation The up-regulation of mucilage synthesis in the seed coat epidermis occurs prior to 7 DPA, and is correlated with a dramatic increase in the number of Golgi stacks. In mum4, both the size and timing of the increase in Golgi stack density remains unchanged, even though the production of mucilage is dramatically reduced. Although polysaccharide secretion required for the formation of the secondary cell wall of the columella still occurs (at 9 DPA) it is not correlated in time with an increase in Golgi stack density. This could be due to the fact that the density of Golgi stacks is already high as a result of the production of mucilage during the previous stage and, as such, no more Golgi stacks are required to perform the new task of secreting the secondary wall that will form the columella. The fact that the timing of the proliferation of Golgi stacks in the mum4 mutant is identical to the wild type, despite a dramatic reduction in the secretory demands on the mum4 cell, implies that Golgi stack proliferation is not a direct consequence of polysaccharide production. Instead, Golgi stack proliferation may be independently activated by developmental signals during cell differentiation. Changes in either the density or the number of Golgi stacks have been shown to be associated with specific time periods of the cell cycle. Garcia-Herdugo et al. (1988) found that Golgi stack density was at a minimum during interphase in onion root meristems, with both the numerical density and the total volume of Golgi stacks increasing throughout mitosis. Segui-Simarro and Staehelin (2006) found that despite a relatively constant Golgi stack density during interphase, the number of Golgi stacks doubled in G2, and this was coincident with the doubling in size of the cell in preparation for mitosis. However, in each of  54  these studies, the nature of the developmental cue involved in expansion of Golgi stack numbers was not determined. There are two possible mechanisms for the formation of Golgi stacks in a given cell: fission of previously existing Golgi stacks, or the de novo formation of new Golgi stacks. In plants, several studies have reported Golgi stacks that have been caught undergoing division in a cis to trans direction, which is strong evidence for Golgi fission as the mostly likely mechanism (Bosabalidis, 1985; Craig and Staehelin, 1988; Hirose and Komamine, 1989; Langhans et al., 2007; Staehelin and Kang, 2008). The mechanism of Golgi stack doubling in seed coat epidermal cells has not been confirmed, though there is evidence that they also divide by Golgi fission (Mollenhauer and Morre, 1991; McFarlane et al., 2008).  55  FIGURE 3-1: The stages of Arabidopsis seed coat development in cryofixed samples. (A-B) 4 DPA cells are relatively undifferentiated, with a large central vacuole. (C-D) At 7 DPA, the characteristic mucilage pocket is visible, with a central cytoplasmic column. The vacuole is only found in the lower portion of the cell. (E-F) In 9 DPA cells, secondary cell wall is being deposited on the apical side of the cell. The formation of the columella is in its initial stages. a - amyloplast; 2cw - secondary cell wall; m - mucilage; n - nucleus; v - vacuole. Scale bars are 5m.  56  FIGURE 3-2: The morphology of individual Golgi stacks at the different stages of seed coat development. (A-B) 4 DPA Golgi stacks are long and relatively thin, with a small TGN. In some cases the Golgi stacks of 4 DPA cells appeared cup-shaped, with the TGN on the concave surface (A). (C-D) Golgi stacks at 7 DPA have a shorter diameter than those of 4 DPA, with a more pronounced cis-trans polarity. The cis cisternae lumen are wider, while the medial and trans cisternae have compressed lumen, but with swollen margins. The TGN appears more extensive as well. (E-F) At 9 DPA, Golgi stacks have a similar morphology to those at 7 DPA. c - cis-face of Golgi stack; t - trans face of Golgi stack; TGN - trans Golgi network; 1cw - primary cell wall; 2cw - secondary cell wall; m - mucilage. Scale bars are 250 nm.  57  FIGURE 3-3: Tomographic reconstruction of a single Golgi stack during the mucilage-secreting stage of seed coat development (7 DPA). (A-B) 2 nm slices from a tomogram, showing how the TGN appears as a series of vesicles. (C-D) En face (C) and (D) side view of a three-dimensional model of a Golgi stack, based on tomographic reconstruction shown in (A-B). The TGN (bright pink) is an extensive interconnected network of large vesicle-like structures. c - cis-face of Golgi stack; t - trans-face of Golgi stack; TGN - trans Golgi network. Scale bars are 250 nm.  58  FIGURE 3-4: Quantification of Golgi stacks visible in sections of seed coat epidermal cells during development. (A) Mean number of Golgi stacks visible in sections of seed coat epidermal cells at 4, 7 and 9 DPA (B) Mean surface area of cytosol in thin sections of seed coat epidermal cells at 4, 7 and 9 DPA. Cytosol was defined as all areas inside the cell walls, excluding the nucleus, vacuole and amyloplasts. (C) Average density of Golgi stacks (number of Golgi stacks/ um2 of cytosol) in thin sections of seed coat epidermal cells at 4, 7 and 9 DPA. For this analysis, three separate experiments were conducted, resulting in a total of 61 cells and 152 Golgi stacks at 4 DPA, 24 cells and 474 cells at 7 DPA, and 17 cells and 333 Golgi stacks at 9 DPA. Raw data were obtained by TEM. Error bars represent 95% confidence intervals (Figure originally published by Young et al., 2008. Copyright Amercian Society of Plant Biologists).  59  FIGURE 3-5: Comparison of wild-type and mum4 seed coat morphology at 7 DPA. (A) A wildtype, 7-DPA ceed coat cell. Note the large mucilage pocket and the central cytoplasmic column above the vacuole. (B-D) A mum4 seed coat epidermal cell at 7 DPA. Cells are flatter than wildtype cells, with very small mucilage pockets and virtually no central cytoplasmic column. Vacuoles are approximately the same size in both wild type and mum4 cells. Note that amyloplasts and mucilage appear darker in B and C as these were stained with alkaline bismuth, while contents of amyloplasts in A have fallen out of the section. a - amyloplast; m - mucilage, v - vacuole. Scale bars are 5 m.  60  FIGURE 3-6: Comparison of the ultrastructure of wild-type and mum4 Golgi stacks. (A) Wildtype, 7 DPA Golgi stack. (B-D) mum4 Golgi stacks at 7 DPA. Cisternae appear longer, with open lumens that are less swollen at the margins, and a less complex TGN. c - cis face of Golgi stack; t - trans face of Golgi stack; TGN - trans Golgi network; m - mucilage; mt mitochondrion. Scale bars are 250 nm.  61  FIGURE 3-7: Quantification of changes in Golgi stacks in sections of wild-type and mum4 seed coat epidermal cells at 4 and 7 days post-anthesis (DPA). (A) Comparison of the mean number of Golgi stacks visible in thin sections of wild-type and mum4 seed coat epidermal cells. (B) Mean surface area of cytosol in thin sections wild-type and mum4 seed coat epidermal cells. (C) Mean density (# Golgi stacks/ m2 cytosol) in thin sections of 7 DPA wild-type and mum4 seed coat epidermal cells. For this analysis, three separate experiments were conducted. For wild type, a total of 61 cells and 152 Golgi stacks at 4 DPA and 24 cells with 474 Golgi stacks at 7DPA were examined. In the case of mum4, 46 cells and 115 Golgi stacks were examined at 4 DPA and 57 cells with 800 Golgi stacks were examined at 7 DPA. Raw data were acquired by TEM. Error bars represent 95% confidence intervals (Figure originally published by Young et al., 2008. Copyright Amercian Society of Plant Biologists).  62  4 LOCALIZATION OF POLYSACCHARIDES IN SEED COAT MUCILAGE* 4.1 Introduction In the previous chapter, the ultrastructure of the Golgi apparatus was examined in detail as it responded to the secretory demands of seed coat development. Comparisons of Golgi stacks in wild type and mum4 seed coat epidermal cells showed that the number of Golgi stacks at a given stage was predetermined by its developmental stage, but the morphology was not. This implies that Golgi stack ultrastructure, but not the number of Golgi stacks found in the cell, is dependent on the amount of cargo that the Golgi stack is carrying at a given time. In this chapter I further explore the role of cargo in the functioning of the seed coat Golgi apparatus. As seed coat mucilage polysaccharides resemble those of the primary cell wall, the work in this chapter is relevant to polysaccharide products being synthesized and secreted by the plant Golgi apparatus generally. The cell wall is a very complex and dynamic structure, composed primarily of polysaccharides that fall into 3 major categories: cellulose, hemicellulose and pectin. The composition of cellulose is consistent across all plants, however within the other two categories, and especially in the case of pectins, the exact composition of a given polysaccharide and its side chains is highly variable (Ridley et al., 2001). In addition, once a given polysaccharide has been incorporated into the wall, it may be subjected to at least one, if not several modifications during  A version of this chapter has been published: • Young, R.E., McFarlane, H.E., Hahn, M.G., Western, T.L., Haughn, G.W. and Samuels, A.L. (2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized secretion of pectin-rich mucilage. Plant Cell 20: 1623-1638. 63  the lifetime of the cell (Willats et al., 2001b). The development of antibodies to cell wall polysaccharides was an important milestone in cell wall biology (Knox, 1997). With antibodies, both the composition and the linkages of the polysaccharide could be targeted in situ, which, in turn, has allowed biologists to learn much about the complexity of the architecture of the cell wall. Most antibody labeling experiments have been focused on the identification of specific subdomains in the cell wall (Knox et al., 1990; Lynch and Staehelin, 1992; Freshour et al., 1996) or on examining the organization of individual Golgi stacks (Moore and Staehelin, 1988; Moore et al., 1991; Staehelin et al., 1991; Zhang and Staehelin, 1992). Pectin production has been localized to the Golgi apparatus through the use of autoradiography (Northcote and PickettHeaps, 1966; Harris and Northcote, 1971; Bolwell and Northcote, 1983) as well as by antibody labeling (Moore and Staehelin, 1988; Moore et al., 1991). Immungold labeling of Golgi stacks with different antibodies showed that individual Golgi stacks were capable of simultaneously producing different complex polysaccharides and glycosylating proteins destined for secretion to the cell wall (Moore et al., 1991). There has been virtually no work on plant cells that examined the distribution of products across the entire population of Golgi stacks within a cell, in order to determine whether the plant cell uses the fragmented Golgi apparatus as a way to compartmentalize certain functions of the secretory pathway. Previous work on the seed coat epidermal cells of Arabidopsis has revealed that there are several different cell wall polysaccharides present in seed coat mucilage. A combination of chemical and immunological techniques has shown that mucilage is made primarily of the pectin RGI and extruded mucilage is comprised of layers which have different relative concentrations of RGI and other components, including HGA and cellulose (Western et al., 2000; Western et al., 2001; Willats et al., 2001a; Usadel et al., 2004; Macquet et al., 2007a). Mucilage is secreted 64  during a very distinct period of seed coat epidermal cell development, followed by the synthesis of the secondary cell wall of the columella. Considering that mucilage and the columella have very different properties, it is highly likely that their composition is also very different. As such, the epidermal cells of the seed coat are an ideal system in which to ask to ask whether a particular cargo, e.g. RGI, is synthesized by the Golgi apparatus as a whole or by specialist stacks.  4.1.1 Objectives The goal of the work described in this chapter was to determine how the seed coat Golgi apparatus as a whole manages the organization of cargo during mucilage production. To that end, an antibody screen was performed on unfixed, hydrated mucilage, and cryofixed, sectioned material in order to identify antibodies that were specific to cell wall polysaccharides from different cell wall regions. These included mucilage, which is secreted at 7 days post-anthesis (DPA) and the secondary cell wall, which is laid down starting at 9 DPA. I used these antibodies to examine the distribution of specific polysaccharide epitopes in the entire population of Golgi stacks visible in cryofixed and sectioned material via transmission electron microscopy (TEM). I also examined how the secretion of specific products changes over time, as the needs of the cell change as it passes from one developmental stage to another. Finally, I performed double labeling experiments to determine whether or not Golgi stacks carrying the different epitopes represented separate populations within the Golgi apparatus.  4.2 Results 4.2.1 Screening of antibodies for reactivity to mucilage The identification of antibodies that reacted strongly with mucilage in cryofixed, resinembedded and sectioned material was required to follow mucilage secretion within the cell. An 65  ideal probe would react strongly with the pectin of the mucilage, but only weakly or not at all with the pectins of the cell wall. Previous research indicated that antibodies that reacted with RGI would be the most likely candidates (Western et al., 2004), but a wide range of antibodies to pectins and other cell wall components were evaluated for their ability to bind mucilage. In addition, Michael Hahn’s group at the CCRC (Athens, GA) had produced several new antimucilage antibodies, which were provided to us for this study. Immunofluorescence of hydrated mucilage from whole mature Arabidopsis seeds was therefore performed. Antibodies that could successfully bind hydrated mucilage were then tested on resin-embedded sections of cryofixed seeds at different stages of development to see if mucilage could also be labeled within the developing seed coat epidermal cell.  4.2.1.1 Antibody screening on hydrated mucilage of mature seeds Initial screening of candidate antibodies was done on extruded mucilage from mature, hydrated seeds. The rationale was that hydrated mucilage is easily accessible for antibody labeling, as it surrounds the seeds and as such, did not require lengthy sample preparation prior to labeling; Seeds can be hydrated by gently shaking in an aqueous environment for a short time in conjunction with the blocking step, prior to primary antibody application. Since there is virtually no sample preparation required, a large number of antibodies can be screened in a relatively short period of time. Of 24 antibodies tested (See Table 4-1), 8 were observed to label extruded mucilage, whole or in part (Figure 4-1, asterisks). One of the strongest reactions was with the polyclonal anti-RGI (Figure 4-1A, asterisk, Moore et al., 1986). Anti-RGI labeling was expected, as it had been used previously to study mucilage (Western et al., 2004). Hydrated mucilage also labeled with CCRC-M36 (Figure 4-1C, asterisk), a monoclonal antibody raised against Arabidopsis seed coat mucilage and that binds strongly to RGIs from Arabidopsis, tomato, lettuce and soybean, as 66  well as to Arabidopsis and Sinapis (mustard) seed coat mucilages (T. Bootten, Z. Popper, A.G. Swennes, M.G. Hahn, in preparation). The CCRC-M36 antibody was one of 6 antibodies received from the CCRC. Interestingly, despite the fact that all 6 antibodies (CCRC-M30, CCRC-M31, CCRC-M32, CCRC-M34, CCRC-M36, CCRC-M38) were raised against seed coat mucilage, three of them showed no reactivity with extruded mucilage on hydrated seeds under the conditions used (Figure 4-1C, asterisks). In addition to CCRC-M36, CCRC-M30 labeled only the inner layer of mucilage, while CCRC-M38 labeled all visible layers of mucilage. JIM5 and JIM7 (Knox et al., 1990) both also labeled the inner layer of mucilage in this study, but differently (Figure 4-1A, asterisks). Whereas JIM5 brightly labeled the inner adherent layer, JIM7 only mildly labeled the outer portion of the inner adherent layer. Despite the fact that LM5 (Jones et al., 1997) did not appear to label hydrated mucilage, a filamentous structure attached to the top of the columella was observed that was strongly labeled by LM5 (Figure 4-1A). It should be noted that some of the bright spots observed (e.g. JIM7 and LM5, Figure 4-1A) are nonspecific, as proven by the fact that they can also be observed in the controls (Figure 4-1E). Interestingly, a polyclonal anti-xyloglucan showed strong reactivity with extruded mucilage as well (Figure 4-1B, asterisk, Moore and Staehelin, 1988). This result was surprising as pectin was considered to be the major component of mucilage at this stage. The presence of xyloglucan in mucilage implied that cellulose may also be present and, as such, a Cellulose Binding Domain (Jervis et al., 1997) conjugated to Oregon Green (CBD-OG, a kind gift from Dr. Douglas Kilburn, Emeritus Professor, Dept of Microbiology and Immunology, UBC) was used to test this hypothesis. Labeling of mucilage with CBD-OG showed that cellulose was indeed present in hydrated mucilage (Figure 4-1D, asterisk) confirming the results of Willats et al., (2001a) who detected the presence of cellulose by using calcofluor staining.  67  TABLE 4-1: List of cell wall probes tested during preliminary screen for antibodies which react to Arabidopsis seed mucilage. Antibodies Anti-pectin JIM5 JIM7 Anti-RGI/PsGA PAM1 LM5 LM7 CCRC-M2 CCRC-M7 Anti-xyloglucan CCRC-M1 CCRC-M48 CCRC-M54 CCRC-M57 CCRC-M58 Anti-XG Anti-seed mucilage CCRC-M30 CCRC-M31 CCRC-M32 CCRC-M34 CCRC-M36* CCRC-M38 Other CBD-OG JIM8 JIM13 LM10 Mac207  Antigen 0-40% esterified HGA 40-100% esterified HGA Multiple epitopes (polyclonal) in RGI and HGA Blockwise de-esterified HGA (14)--D-galactan (side chains of RGI) Non-blockwise de-esterified HGA RGI (unknown epitope) 6-linked -D-galactose oligomers that contain arabinose (RGI and arabinogalactan proteins). -L-fucosylated xyloglucan Non-fucosylated XG (unknown epitope)  Multiple epitopes (polyclonal) in XG. No cross reactivity with RGI Unknown epitopes (raised against Arabidopsis seed mucilage) *Note: CCRC-M36 has been partially characterized and binds to RGI-like molecules. See text (p. 68)for details. Bacterial cellulose binding domain conjugated to Oregon Green Arabinogalactan proteins Arabinogalactan proteins (-D-GlcA-(1,3)--D-GalA-(1,2)--L-Rha) Xylan component (feruloylated 1,4-linked -D-galactan) Arabinogalactan proteins (-GlcA-(1,3)--GalA-(1,2)--Rha)  Reference (Knox et al., 1990) (Knox et al., 1990) (Moore et al., 1986) (Willats et al., 1999) (Jones et al., 1997) (Willats et al., 2001c) (Puhlmann et al., 1994) (Puhlmann et al., 1994)  (Puhlmann et al., 1994) (Z. Popper, T. Bootten, S. Tuomivaara, A.G. Swennes, R. Jia, W.S. York, and M.G. Hahn, in preparation) (Moore et al., 1986)  (T. Bootten, Z. Popper, C. Deng, R. Jia, W.S. York, M.A. O’Neill, M.G. Hahn, in preparation)  (Jervis et al., 1997) (Pennell et al., 1991) (Knox et al., 1991) (McCartney et al., 2005) (Pennell et al., 1989)  4.2.1.2 Antibody screening on sections of cryofixed developing seeds The second stage of antibody screening was on cryofixed sections of developing seeds to test whether the epitopes could be detected in the developing seed coat epidermal cells after high-pressure freezing, freeze substitution and resin embedding. Samples were cryofixed at 4, 7 68  and 9 DPA, sectioned and labeled by immunofluorescence (Figure 4-2 and 4-3). Despite variable binding to hydrated mucilage, all of the CCRC antibodies raised against seed mucilage were tested on cryofixed samples, as sectioning of material may expose new epitopes (Figure 4-2). Only CCRC-M36 labeled the mucilage pocket in sections (Figure 4-2, J-L). Interestingly, CCRC-M30 (not shown) and CCRC-M38 (Figure 4-2, M-O), both of which showed obvious labeling in extruded mucilage (Figure 4-1C, asterisk), failed to label the mucilage pocket in sections. CCRC-M30 had no obvious reaction with any part of the seed at all three stages of development in sectioned material while CCRC-M38 labeled primary cell walls, but not the mucilage (Figure 4-2, M-O). The epitope of CCRC-M36 appears to reside on the unbranched backbone of RGI and does not bind strongly with arabinogalactans, xylans, xyloglucans, or methyl-esterified homogalacturonans (C. Deng, R. Jia, A. Albert, W.S. York, M.A. O’Neill, M.G. Hahn, in preparation). Pre-absorbing CCRC-M36 with extracted seed mucilage completely destroyed labeling pattern (Figure 4-4). At 4 DPA (Figure 4-2A), prior to the onset of mucilage production, CCRC-M36 showed no reactivity to any cell walls in cryofixed, sectioned seeds, including specifically the seed coat epidermal cells (Figure 4-2J). At 7 DPA (Figure 4-2B and 45A), at the height of mucilage production, the mucilage pockets of seed coat epidermal cells were brightly labeled, but the cell walls of these and other cell types were not (Figure 4-2K and 4-4C). Punctate structures were observed in the cytosol of 7 DPA seed coat epidermal cells (Figure 4-5C) but not at 9 DPA (Figure 4-5D), when polysaccharide production had shifted from the production of mucilage to the production of secondary cell wall (Figure 4-2C and 4-5B). Labeling of the mature mucilage pockets was strong in 9 DPA seed coat epidermal cells (Figure 4-2L and 4-5D). Of the rest of the antibodies, JIM5, JIM7, anti-RGI and anti-xyloglucan were chosen to examine in developing seed coats (Figure 4-2, P-R and Figure 4-3). LM5 was also chosen, due to 69  the intriguing labeling of the strands on top of the columella (Figure 4-3, J-L). Considering that anti-RGI had been used previously to label sectioned material (Western et al., 2004) and that it labeled hydrated mucilage well (Figure 4-1A, asterisk), this antibody was expected to label mucilage in sections of developing seeds. Surprisingly, this was not the case. Despite the fact that anti-RGI did label primary cell walls throughout the seed at all stages of development, no label was ever observed in the mucilage pocket (Figure 4-3, M-O). JIM5 and JIM7 both labeled the inner layer of hydrated mucilage under our conditions (Figure 4-1A, asterisks), but neither one of them labeled the mucilage pocket in sectioned material (Figure 4-3, D-I). JIM5 labeled very little of the seed in all stages, with the strongest label near the funicular end (Figure 4-3, DF). JIM7 labeled much of the primary cell wall in the seed, especially in the layers of the outer integument (Figure 4-3, G-I). Interestingly, LM5 appeared to have a strong label in the outer integument cells at 4 DPA, before the onset of mucilage production (Figure 4-3J), but this label disappeared as the seed matured. No label was observed in the mucilage pocket (Figure 4-3, KL). Anti-xyloglucan was raised against sycamore maple (Acer pseudoplatanus) xyloglucan and does not cross-react with pectic components, like RGI (Moore and Staehelin, 1988). Labeling with this antibody was destroyed by pre-absorption with tamarind xyloglucan (Figure 4-4) In addition to the labeling of developing secondary cell walls in the Arabidopsis seed coat, this antibody also labeled primary cell walls of the seed throughout development, and the mucilage pockets, but at a less intense level than CCRC-M36 (Figure 4-2P-R, and Figure 4-5). Of all the antibodies screened, CCRC-M36 and anti-xyloglucan were the only antibodies to show a strong label in the mucilage pocket. Thus CCRC-M36 and anti-xyloglucan were chosen to investigate changes in polysaccharide product by immunolabeling sections of seeds before, during and after mucilage production.  70  4.2.2 Plant Golgi stacks produce mucilage synchronously during seed coat differentiation The changes in the CCRC-M36 labeling patterns observed in seed coat epidermal cells by light microscopy indicated that a closer examination of the Golgi apparatus might yield new information about the regulation of the secretory pathway during seed coat development. Since CCRC-M36 labels only mucilage and not other primary cell wall components in this cell type, this antibody was used to identify the specific population of Golgi stacks that are involved in the production of mucilage during seed coat development.  To resolve the punctate structures  observed in fluorescence microscopy at 7 DPA (Figure 4-5C) and to quantify individual Golgi stacks producing mucilage, sections of 7 and 9 DPA seeds were immunolabeled with either CCRC-M36, or -XG, and detected with gold-conjugated secondary antibodies for examination by transmission electron microscopy (TEM) (Figure 4-6). CCRC-M36 labeled 70% of all Golgi stacks visible in 7 DPA seed-coat sections, most commonly labeling the medial and/ or trans cisternae, as well as the TGN. In contrast, 0% of Golgi stacks were labeled by CCRC-M36 in 9 DPA cells. These data indicate that the pectin epitope is indeed present in the Golgi apparatus during a short period of intense mucilage secretion but not at later stages during synthesis of the secondary cell wall. Thus, the short-term, specialized production of mucilage is reflected in the pattern of Golgi apparatus labeling. On the other hand, when -XG was used as the primary antibody, gold particles were seen associated with 16% of Golgi stacks in 7 DPA cells, and 22% of Golgi stacks in 9 DPA cells suggesting that xyloglucan is secreted during the synthesis of both the mucilage and columella. Despite the high percentage of Golgi stacks that label with CCRC-M36 at 7 DPA, it is conceivable that the epitopes recognized by the two different antibodies are being carried by different groups of Golgi stacks. To address this question, double labeling experiments, using 71  both CCRC-M36 and -XG, were performed on 7 DPA seed coat epidermal cells (Figure 4-7). Results of double labeling were consistent with single labeling assays; 77% of Golgi stacks labeled with CCRC-M36 in the double labeling experiment, whereas 11% of Golgi stacks labeled with -XG. Of the six Golgi stacks observed to have -XG label, only one of them did not also have CCRC-M36 label, suggesting that the different antibodies do not identify distinct groups of Golgi stacks within the cell.  4.3 Discussion 4.3.1 Golgi stacks respond synchronously, and collectively represent a single Golgi apparatus Antibodies have been used to examine how compartmentalization of the Golgi apparatus has a role in ensuring the proper polysaccharides are produced within a single Golgi apparatus (Moore et al., 1991; Lynch and Staehelin, 1992; Zhang and Staehelin, 1992; Staehelin and Moore, 1995). Moore et al., (1991) established that multiple products could, in fact, be produced in the same Golgi stack. However, the possibility that the different Golgi stacks in a single cell could make distinct products has not been carefully examined. The studies described in this chapter address this issue, through the use of antibodies that specifically label mucilage. The fact that the antibodies used also label products that are produced differentially throughout seed coat development, allows us to gain insight into how the plant manages the temporal demands on secretion within a single cell. Our study shows that the majority of the Golgi stacks are involved in mucilage production at 7DPA, based on the widespread distribution of the mucilage-specific RGI epitope, CCRC-M36. As well, less abundant polysaccharides that are also found in mucilage, such as xyloglucan, are produced in many of the same Golgi stacks as those bearing the mucilage72  specific RGI epitope. These data, in conjunction with the uniform morphological changes in the Golgi apparatus described in Chapter 3, support the idea that the multiple stacks of the Golgi apparatus respond synchronously to developmental cues, and act in unison to produce mucilage at 7 DPA.  4.3.2 The composition of seed coat mucilage In order to be able to use the secretion of seed coat mucilage as a system in which to examine polysaccharide biosynthesis, it is important to know its exact composition. Extruded mucilage has several layers, including an inner adherent layer and an outer layer that is much easier to separate from the seed (Western et al., 2000; Macquet et al., 2007a). Although the relative amounts of different polysaccharides in these layers appear to differ (Willats et al., 2001a; Macquet et al., 2007a; Young et al., 2008), the array of components involved remains relatively constant. In all 3 layers the primary component of seed mucilage is RGI. In addition, the inner adherent layer is also thought to be composed of more densely aggregated polysaccharides compared to the outer layers, and as such have comparatively higher concentrations of HGA and cellulose (Macquet et al., 2007a). The study presented here confirms the presence of RGI in mucilage, through the use of CCRC-M36, which recognizes the backbone of RGI-like polymers (C. Deng, R. Jia, A. Albert, W.S. York, M.A. O’Neill, M.G. Hahn, in preparation). Chemical analysis of mucilage, in addition to mutant analysis, has shown that the particular form of RGI found in mucilage is most likely to be relatively unbranched (Dean et al., 2007; Macquet et al., 2007b). It is possible that this particular form of RGI is important for enabling mucilage expansion upon hydration, since mutants such as mum2, in which some linkages observed in mucilage polysaccharides are altered, also have a decreased capacity for mucilage expansion (Dean et al., 2007; Macquet et al., 2007b) 73  This study also reveals strong xyloglucan labeling in extruded mucilage, as well as labeling of mucilage in the mucilage pocket of developing seed coats. Interestingly, Macquet et al. (2007a) only found trace amounts of xyloglucan in mucilage in their study. Since the presence of cellulose has been suggested by several different methods (Western et al., 2000; Willats et al., 2001a; Macquet et al., 2007a; Young et al., 2008), it is perhaps not surprising that hemicellulose is also present to form a cellulose-hemicellulose network. The fact that only 16% of Golgi stacks label with anti-xyloglucan, compared to 80% of stacks that label with the anti-RGI, CCRC-M36, suggests that that the percentage of XG in mucilage is relatively small.  4.3.3 Xyloglucan in the secondary cell wall Our data also demonstrates that XG is present in the columella, based on the antibody labeling of seed coat epidermal cells at 9 DPA. The columella is considered to be a secondary cell wall by definition, due to the fact that it is laid down after the completion of cell expansion. Xyloglucan is the hemicellulose most commonly associated with type I primary cell walls (Carpita and Gibeaut, 1993), whereas secondary cell walls in other Arabidopsis cell types such as xylem and interfascicular fibres, show reactivity with anti-xylans (such as LM10 and LM11) (McCartney et al., 2005; Persson et al., 2007b). In developing hybrid aspen (Populus tremula x P. tremuloides) secondary xylem, xyloglucans were localized with CCRC-M1 to the compound middle lamella region of the secondary cell wall of fibres and are believed to link the primary and secondary cell walls during secondary cell wall deposition (Bourquin et al., 2002). Therefore, the presence of XG throughout the secondary cell wall layer in the seed coat is intriguing.  74  4.3.4 Tracking  polysaccharide  production  with  mucilage-specific  antibodies I have tested a series of novel antibodies to mucilage, and shown that the CCRC-M36 antibody reacts only with mucilage in the developing seed, providing an invaluable tool for the analysis of seed coat mucilage secretion. Although there are other antibodies which label seed coat mucilage, including -XG (reported here) and PGA/RGI (Western et al., 2004), CCRCM36 has the advantage that in the Arabidopsis seed coat it is specific to mucilage. This fact has allowed the selective tracking of cargo that is being targeted to a specific cell wall domain. The work described in this chapter has been dependent on this aspect of CCRC-M36 labeling. Future experiments that are now possible with this antibody include examining how polysaccharides and enzymatic cargo destined for the apoplast are packaged and sorted by the Golgi apparatus.  75  76  FIGURE 4-1 (overleaf): Antibody screen of mucilage from hydrated mature seeds. All antibodies that label extruded mucilage have been marked with an asterisk (*) in the top righthand corner of the image. (A) Antibody screen using known anti-pectins. (B) Anti-xyluglucan antibody screen. Of these, only anti-xyloglucan (-XG) showed reactivity to mucilage. (C) Antibody screen using antibodies raised against extracted seed coat mucilage (T. Bootten, Z. Popper, C. Deng, R. Jia, W.S., York, m.a. O'Neill, M.G. Hahn, in preparation). (D) Other cell wall markers were also examined, including Cellulose Binding Domain conjugated to Oregon Green (CBD-OG), the arabinogalactan markers, JIM8, JIM13 and MAC207, and an antibody specific to xylans, LM10. (E) Controls. India ink counterstaining was used to determine the size of the mucilage halo, in comparison to the fluorescent labeling. Typical controls without primary antibody for Alexafluor 594 (red) and Alexafluor 488 (green) are also shown. Note that bright spots visible in some Alexafluor 594 labeling are non-specific (e.g. LM5), and considered to be an artifact. Scale bars are 100 m (Figure originally published by Young et al., 2008. Copyright American Society of Plant Biologists).  77  FIGURE 4-2: Distribution of epitopes recognized by anti-seed mucilage and anti-xylogucan antibodies in developing Arabidopsis seeds. (A-C) Toluidine blue-stained sections show seed anatomy at 4, 7 and 9 DPA. (D-O) Sections of 4, 7 and 9 DPA seeds fluorescently immunolabeled with different antibodies raised against seed mucilage. (D-F) CCRC-M32. (G-I) CCRC-M34. (J-L) CCRC-M36. (M-O) CCRC-M38. (P-R) Sections of 4, 7 and 9 DPA seeds immunolabeled with a polyclonal antibody raised against xyloglucan. Scale bars are 100 m.  78  FIGURE 4-3: Distribution of epitopes recognized by anti-pectins in developing Arabidopsis seeds. (A-C) Toluidine blue-stained sections show seed anatomy at 4, 7 and 9 DPA. (D-F) Sections of 4, 7 and 9 DPA seeds immunolabeled with JIM5, an antibody that recognizes low to moderately esterified HGA. (G-I) Sections of 4, 7 and 9 DPA seeds immunolabeled with JIM7, an antibody that recognizes moderate to highly esterified HGA. (J-L) Sections of 4, 7 and 9 DPA seeds immunolabeled with LM5, an antibody that recognizes galactan sidechains of RGI. (M-O) Sections of 4, 7 and 9 DPA seeds immunolabeled with anti-RGI, a polyconal antibody that recognizes RGI. Scale bars are 100 m.  79  FIGURE 4-4: Fluorescent labeling antibody controls. (A-B) Thick sections of wild-type, 7 days post-anthesis (DPA) seeds labeled with the anti-mucilage, CCRC-M36 (A) or anti-xyloglucan (-XG) (B). (C-D) When the primary antibody is omitted, non-specific reactivity of the secondary antibody is not observed. (E-F) When the primary antibody is preabsorbed with mucilage (E) or tamarind xyloglucan (F), non- specific reactivity is not observed. Scale bars represent 100m (Figure originally published by Young et al., 2008. Copyright American Society of Plant Biologists).  80  FIGURE 4-5: Details of seed mucilage (CCRC-M36) or xyloglucan (-XG) labeling in 7 and 9 DPA seed coats. (A-B) Toluidine blue-stained sections of cryofixed Arabidopsis seed coats at 7 and 9 DPA. (C-D) Seed coat epidermal cells at 7 and 9 DPA showing a strong CCRC-M36 immunofluorescent label in the mucilage pocket. At 7 DPA (C), punctate structures in the cytosol also label with CCRC-M36 (arrowheads). (E-F) Anti-xyloglucan (-XG) labels both the mucilage pocket and the primary cell wall. In addition, -XG labels the developing secondary cell wall of the columella (arrows) at 9 DPA (F). Wrinkles in section lead to the appearance of uneven labeling in (E). Scale bars are 15 m (Figure originally published by Young et al., 2008. Copyright American Society of Plant Biologists).  81  82  FIGURE 4-6 (overleaf): Single epitope immunogold labeling of Golgi stacks. (A) Immunogold labeling of wild-type, 7 DPA seed coat epidermal cells with the anti-mucilage antibody, CCRCM36, using 10 nm gold particles (squares). (B) Immunogold labeling of wild-type, 9 DPA seed coat epidermal cells with anti-xyloglucan (-XG), using 10 nm gold particles (circles). c - cis face of Golgi stack, m - mucilage, t - trans face of Golgi stack, TGN - trans Golgi network. Scale bars are 250 nm. (C) Percentage of Golgi stacks that immunolabel with 10 nm gold in thin sections of seed coat epidermal cells at 7 and 9 DPA. Sections were labeled with either CCRCM36 or -XG. This analysis is based on three separate experiments, which includes a total of 31 cells and 287 Golgi stacks for 7 DPA cells labeled with CCRC-M36, 28 cells and 252 Golgi stacks for 7 DPA cells labeled with -XG, 23 cells and 287 Golgi stacks for 9 DPA cells labeled with CCRC-M36, and 20 cells and 282 Golgi stacks for 9 DPA cells labeled with -XG. Error bars represent range of percentages obtained in replicates (Figure originally published by Young et al., 2008. Copyright American Society of Plant Biologists).  83  FIGURE 4-7: Double immunogold labeling of wild-type, 7 DPA seed coat epidermal cells. (A) Labeling density in seed coat mucilage within the mucilage pocket (see inset for location with respect to the seed coat epidermal cell) using CCRC-M36 (15 nm gold) and -XG (10 nm gold, circles). (B) Double immunogold labeling of Golgi stack using CCRC-M36 (15 nm gold, squares) and -XG (10 nm gold, circles). c - cis face of Golgi stack, t - trans face of Golgi stack, TGN - trans Golgi network. Scale bar is 250 nm (Figure originally published by Young et al., 2008. Copyright American Society of Plant Biologists).  84  5 POLARIZED SECRETION AND POST-GOLGI TRAFFIC* 5.1 Introduction Thus far in this thesis I have examined the Golgi apparatus in detail, by studying both the morphology and the cargo of Golgi stacks in seed coat epidermal cells throughout development. I have shown that Golgi stack morphology is correlated with the levels of polysaccharide being secreted, but that the number of stacks within the cell is not. I have also shown that during mucilage production, the majority of the Golgi stacks are producing mucilage and that more than one component of mucilage can be produced in the same Golgi stacks. In this chapter I will focus on how the polysaccharide products of the Golgi apparatus of these cells get transported and delivered to their correct destinations within the apoplast. Growth in plant cells requires the secretion of cell wall polysaccharides into an existing framework that allows for expansion. Two main types of cell growth are described in the literature: in diffuse growth, the secretion of new cell wall products by the Golgi apparatus is thought to occur along the entire wall equally (Martin et al., 2001; Smith, 2003). The second type of growth, tip growth, is highly polarized in that cell wall products are specifically secreted to a single domain of the wall, driving the growth in a specific direction (Martin et al., 2001; Smith and Oppenheimer, 2005; Campanoni and Blatt, 2007). Tip growth is observed in extremely specialized cells, such as pollen tubes and root hairs, whereas diffuse growth is observed in the majority of cells that form tissues and organs in the plant. Seed coat epidermal cells are  A version of parts of this chapter has been published: • Young, R.E., McFarlane, H.E., Hahn, M.G., Western, T.L., Haughn, G.W. and Samuels, A.L. (2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized secretion of pectin-rich mucilage. Plant Cell 20: 1623-1638. • McFarlane, H.E., Young, R.E., Wasteneys, G.O., and Samuels A.L. (2008). Cortical microtubules mark the mucilage domain of the plasma membrane in Arabidopsis seed coat cells. Planta 227: 1363-1375. 85  intriguing, as they grow diffusely during primary growth, and then they secrete mucilage and the columella in a polarized fashion. There are several routes cargo can take after it leaves the Golgi apparatus (Figure 1-1 shows the major post-Golgi pathways). However, traffic between the Golgi apparatus and plasma membrane is thought to be the default pathway of the plant secretory system. Very little is known about how the plant cell targets products to the plasma membrane, especially in the case of diffuse growth (Jürgens and Geldner, 2002). In many of the more extreme types of polarized secretion, such as in tip growth or the formation of the cell plate in mitosis, Golgi stacks are observed to cluster near the site of secretion and, as such, this has often been considered a mechanism for polarized secretion in plants (Jürgens, 2005; Smith and Oppenheimer, 2005). Yet polarized secretion has not been studied in detail in diffusely growing cells. Studies of vesicle traffic between the Golgi apparatus and the plasma membrane are also underrepresented in the literature, possibly due to the fact that vesicles bearing cargo destined for the plasma membrane are difficult to identify experimentally (Battey et al., 1999; Rojo and Denecke, 2008). Arabidopsis seed coat epidermal cells appear to undergo diffuse growth, based on the fact that they grow in all directions relatively equally until the onset of mucilage production (Western et al. 2000 and Section 5.2.4) and yet immunolabeling of mucilage discussed in Chapter 4 shows that mucilage products identified by CCRC-M36 only accumulate in the mucilage pocket, and not other areas of the apoplast. This implies that despite the fact that diffuse growth is taking place, specific domains of the plasma membrane are being targeted for the secretion of mucilage products. This chapter focuses on how mucilage products are specifically targeted to the mucilage pocket, and not other areas of the plasma membrane, by the secretory system of seed coat epidermal cells.  86  5.1.1 Related work Work done by Heather McFarlane (Samuels Lab) is directly related to the work described in this chapter, and forms the basis for some of the conclusions described here. I trained Heather to high-pressure freeze/freeze substitute 7 DPA Arabidopsis seed coats and she performed a TEM study of the cortical cytoskeleton during secretion. She quantified the arrangement of cortical microtubules in 7 DPA seed coat epidermal cells, and found a more than a 4-fold larger number of microtubules under the mucilage pocket, versus other areas of the plasma membrane (McFarlane et al., 2008). Additionally, she compared 7 DPA wild-type cells to those of the temperature sensitive mor1-1 mutant. At the restrictive temperature, the mor1-1 protein is not functional, and the resultant phenotype is short, disorganized microtubules in the cortical array (Whittington et al., 2001). Immunofluorescence labeling confirmed the disorganized state of microtubules in mor1-1 seed coat epidermal cells at the restrictive temperature but the concentration of microtubules below the mucilage pocket was not severely altered, nor was the targeting of mucilage-bearing vesicles to the mucilage pocket (McFarlane et al., 2008). Interestingly, despite the fact that mucilage appeared to be properly targeted to the mucilage pocket, mor1-1 seeds grown at restrictive temperature did not release mucilage properly 40% of the time, compared to wild-type, which had abnormal mucilage release only 20% of the time at the restrictive temperature. This was shown to be significant [2(6, n=526, 0.05)=67.325] (McFarlane et al., 2008).  5.1.2 Objectives In this chapter I will be addressing the question of how the polysaccharide products of the Golgi apparatus are transported and targeted to their destination in the mucilage pocket. I will be examining the role that the location of the Golgi stack within the cell plays in targeting, as well as the morphological characteristics and distribution of a population of post-Golgi structures 87  named ‘electron-dense vesicles,’ based on the fact that they are densely stained when observed in TEM. Finally I will look at whether the microtubule array observed below the mucilage pocket in 7 DPA cells plays a role in the targeted secretion of mucilage.  5.2 Results 5.2.1 Golgi stacks do not cluster near the site of secretion Since mucilage secretion is targeted to the outer tangential side of the cell, the mobile Golgi stacks producing the secretory product theoretically could cluster at the site of secretion, eliminating much of the need for long distance transport and targeting of vesicles to the apical domain of the plasma membrane. The distribution of Golgi stacks in the cell was compared at sites of mucilage deposition (apical domain, Figure 5-1A) and in the portions of the cell distal to the mucilage pockets (basolateral domain, Figure 5-1A). Golgi stack distribution was mapped, but there were no significant differences between the densities of Golgi stacks in the apical domain versus the basolateral domains at 7 DPA (compared using the non-parametric Wilcoxon Signed-Ranks tests for repeated measures; z = -1.975, p>0.05) (Figure 5-1B). Thus, the entire Golgi apparatus of seed coat epidermal cells is evenly distributed around the cell. This was confirmed by live cell imaging with ST-GFP as well, where an even distribution of punctate structures characteristic of GFP-tagged Golgi stacks were observed in seed coat epidermal cells at 4 and 7 DPA (Young et al., 2008). Although the entire population of mobile Golgi stacks does not cluster near the site of mucilage secretion, it is possible that there is a polarized distribution of mucilage-carrying Golgi stacks in the apical region. If this were true, it would suggest that there are distinct populations of Golgi stacks, i.e. ‘specialist’ Golgi stacks, which reside near the apex and produce large amounts of pectin. The distribution of the Golgi stacks labeled with CCRC-M36 was examined 88  using immunogold/ TEM, in order to determine whether more Golgi stacks carrying specific products were found in the areas where mucilage was incorporated. When the cellular locations of Golgi stacks containing the CCRC-M36 pectin-epitope or -XG were mapped using immunogold TEM, there were virtually no differences between the percentages of antibodygold-labeled Golgi stacks in the apical versus basolateral regions of the cells (Figure 5-1C). This indicates that mucilage-secreting Golgi stacks are found throughout the cell, not just in the area near where mucilage is being deposited.  5.2.2 Vesicles with electron-dense cargo are observed in developing seed coat epidermal cells In order to examine how cargo is delivered in the developing seed coat of Arabidopsis, a survey of post-Golgi structures was undertaken. In addition to the interconnected vesicular clusters of the TGN (See Chapter 3) a unique population of vesicles with electron-dense cargo was observed (hereafter named electron-dense vesicles, Figure 5-2). Electron-dense vesicles were smaller than the vesicular clusters of the TGN, but larger than clathrin-coated vesicles (Figure 5-3 inset). As well, no associated coat was observed with electron-dense vesicles. Electron-dense vesicles were observed throughout the cytoplasm at 7 and 9 DPA, including in small clusters near Golgi stacks (Figure 5-2). At 7 DPA they were also seen in close proximity to the cortical microtubule array below the mucilage pocket (Figure 5-3 and 5-4, A-B) whereas at 9 DPA they were observed near the apex of the cytoplasmic column among a population of vesicles believed to be fusing with the plasma membrane (Figure 5-4C and D). Because the morphology of the electron-dense vesicles did not display any similarities to vesicles such as clathrin-coated vesicles that are involved in endocytosis, and because it is unlikely that seed coat epidermal cells undertake the storage of proteins via a storage vacuole, the electron-dense vesicles are hypothesized to belong to the post-Golgi secretory pathway of the 89  cells. In this model, once the cargo exits the Golgi apparatus via the TGN, the contents of the vesicle are somehow condensed, making them appear smaller and more electron-dense in TEM, as they travel from the Golgi apparatus to the plasma membrane. If the electron-dense vesicles are involved in secretion, then at 7 DPA one of their most likely cargos is mucilage (See Chapter 4). If this is the case, it is possible that their contents can be immunolabeled with antibodies known to label mucilage, such as CCRC-M36 (labeling mucilage RGI, see Table 4-1) or antixyloglucan (-XG). Non-osmicated samples were double immunogold labeled with CCRC-M36 and -XG (Figure 5-5). Electron-dense vesicles were more difficult to differentiate from other uncoated vesicles in non-osmicated samples, but could still be identified (Figure 5-5, B-C). Also, efficiency of labeling was highest in the TGN and then decreased in vesicles, especially in electron-dense vesicles. Regardless, labeling was observed in both vesicles (Figure 5-5, A-C) and electron-dense vesicles (Figure 5-5C). Also, vesicular clusters were observed to label with XG and CCRC-M36 antibodies (Figure 5-5D), though it is possible that this cluster was a TGN with its associated Golgi stack outside of the plane of section. Interestingly, no vesicles were observed which labeled with both -XG and CCRC-M36 at the same time. Whether this is because of steric hindrance of multiple antibodies attached to epitopes in a single vesicle, or due to specific sorting and packaging is unclear at this time. In order to provide further evidence that electron-dense vesicles were not endocytic in nature, 3 different antibodies were acquired that have been used previously as endocytic markers in plants. The first two were antibodies to the rice homolog of SCAMP1 (OsSCAMP1, a kind gift from Dr. Liwen Jiang, Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China), which is believed to identify an early endosomal compartment that resembles the partially coated reticulum and TGN (Lam et al., 2007). Each antibody was raised against a different amino acid sequence found in SCAMP1 of Oryza sativa (Lam et al., 2007). BLAST 90  searching using these peptide sequences revealed that neither had an exact match in Arabidopsis, but that several AtSCAMP proteins had partial sequence similarity, including AtSCAMP1. However, attempts to immunogold label 7 DPA seed coat epidermal cells with either OsSCAMP1 antibody yielded no obvious labeling of any specific compartment or vesicle in the developing seed coat (data not shown). The second endosomal compartment marker that was used to label electron-dense vesicles was an antibody to AtRabA4b (Preuss et al., 2004). AtRabA4b bears distinct homology to the mammalian endosomal pathway marker, Rab11, and is ubiquitously expressed in Arabidopsis (Preuss et al., 2004). When 7 DPA seed coat epidermal cells were immunogold labeled using the AtRabA4b antibody, gold particles were observed associated with numerous membrane-bound compartments (Figure 5-6). Sometimes these compartments bore resemblance to a vesicular cluster, such as the partially coated reticulum. Although no Golgi stack was observed directly apposed to the labeled compartment, the associated Golgi stack could have been outside of the plane of section. Thus it cannot be ruled out that the compartments labeled by AtRabA4b were TGN. However, no electron-dense vesicles were observed to have gold particles associated with them (Figure 5-6B). Since electron-dense vesicles lacked the proper morphology for endocytic vesicles, they did not label with antibodies to known endocytic markers such as RabA4b and OsSCAMP1 and finally, because these vesicles do label with antibodies known to label mucilage, which is secreted in high quantities at 7 DPA, my conclusion is that these vesicles are most likely to be secretory vesicles destined for the plasma membrane.  5.2.3 Electron-dense vesicles also do not cluster near the site of secretion at 7 DPA Since mucilage is being secreted in high quantities to a particular area of the apoplast, namely the mucilage pocket, and Golgi stacks are not observed to cluster in any one area of the 91  cell, the implication is that vesicles must carry the appropriate molecular machinery for targeting of mucilage to the correct area of the plasma membrane. In sensitive antibody immunofluorescence and immunogold localization experiments with multiple mucilage epitopes, mucilage was not observed to accumulate in any other area of the apoplast surrounding the seed coat epidermal cell, indicating that mucilage-bearing vesicles must fuse with the plasma membrane directly below the growing mucilage pocket. Therefore, one might expect to see an increased number of vesicles near the site of deposition of the cargo. At 9 DPA, secondary cell wall deposition is the major activity of the secretory apparatus, and an obvious clustering of vesicles can be observed fusing with the apex of the cytoplasmic column (Figure 5-4, C-D). Therefore it seems reasonable that some kind of clustering should be happening at 7 DPA as well. This idea was tested by quantifying the distribution of electron-dense vesicles in different regions of the cell. Based on this hypothesis and the observations in TEM of clustered vesicles at 9 DPA, one might expect that there would be more vesicles in the apical region of the cell if vesicles were traveling towards this area for release to the mucilage pocket. The arrangement of electrondense vesicles was quantified in a similar manner to the Golgi stacks: electron-dense vesicles were counted in TEM sections and reported as a measurement of electron-dense vesicles per square micron (d.v./m2) of cytosol in either the apical or basolateral region of the cell (Figure 57). The rationale behind this is similar to that described for the Golgi stack distribution (See Chapter 3): Since there is a significant difference in the amount of visible cytosol in the apical and basolateral regions of the cell, statistical comparison of the absolute numbers of vesicles would not be a true indication of whether more vesicles were released in one area over the other. When vesicle density in the cytosol was measured, the apical region of the cell showed a decrease in the number of electron-dense vesicles per square micron in the apical region (0.36 92  d.v./m2), as compared to the basolateral region of the cell (0.64 d.v./m2). This was considered to be statistically significant (Paired samples t-test, t(11) = -3.174, p<0.05, Figure 5-7D). This result was initially surprising, considering that vesicle clustering and fusion was so obvious at 9 DPA. However, the vesicles at 7 DPA may move more rapidly through the peripheral region than those observed at 9 DPA and fuse immediately upon arrival at the plasma membrane. If this alternate hypothesis were true, it is possible that fewer vesicles would be observed in the area immediately adjacent to the plasma membrane, compared to other areas of the cell. Further examination confirmed that there appeared to be fewer electron-dense vesicles within 500 nm of the plasma membrane than there were in the central region of the cell.  5.2.4 Cortical microtubules line the mucilage pocket at 7 DPA, but not 9 DPA Results from Section 1.2.1 show that the location of the Golgi stack has little bearing on the final destination of the cargo. It stands to reason, then, that there are other factors influencing the proper targeting of mucilage-bearing vesicles in the seed coat. It has been noted in Chapter 3 that microtubules show a specific organization that changes during seed coat development. Since the microtubules are observed to be As the cortical microtubules were observed in the area where vesicle fusion was most likely to be taking place in the 7 DPA seed coat epidermal cells, they were examined in more detail, in order to discern whether microtubules could be involved in vesicle targeting to the mucilage pocket. The cortical microtubule array was examined in 7 and 9 DPA seed coat epidermal cells by TEM on cryofixed sections of wild-type seed coat epidermal cells (Figure 5-8). At 7 DPA, microtubules lie in close proximity to the plasma membrane directly below the mucilage pocket and not other regions of the cell (See also McFarlane et al., 2008). They appear to encircle the mucilage pocket, forming a series of bands that are perpendicular to the apical-basal axis of the 93  cell (Figure 5-8, A, C-D and F). Microtubules were often observed in cross section, showing the characteristic organization of 13 protofilaments (Figure 5-8D). Interestingly, despite the fact that the microtubules lie directly adjacent to the area where mucilage-bearing vesicles should be fusing with the plasma membrane in high numbers, vesicle fusion was never observed. In contrast, at 9 DPA, no microtubules could be found below the mucilage pocket (5-8, B, E and G). However, several vesicles were observed closely associated with the mucilage pocket, including vesicles that were fusing, as described previously (Figure 5-4D). The changes in the microtubule array between 7 and 9 DPA cells were confirmed by confocal imaging of chemically fixed whole seeds labeled with a monoclonal antibody to tubulin (Figure 5-9). As the cytoplasmic column of 7 and 9 DPA seed coat epidermal cells contains several starch granules that have been found to obscure confocal imaging results (McFarlane et al., 2008) the starchless mutant phosphoglucomutase 1 (pgm1), which is deficient in chloroplast-localized phosphoglucomutase activity (Caspar et al., 1985) was used for all confocal imaging. In order to confirm that the lack of starch did not affect mucilage production, several control experiments were done, including mucilage release assays, and immunolabeling of extruded mucilage with CCRC-M36 (data not shown). Under no circumstances were any differences observed between wild type and pgm1, other than the expected lack of starch granules. At 7 DPA, confocal imaging showed an obvious cortical microtubule array in seed coat epidermal cells (Figure 5-9, A and D). Microtubules appeared to line the apical side of the cell most prominently, confirming the arrangement that had been suggested by TEM. Labeling of 9 DPA cells was more difficult, presumably because of the increased secondary wall that the antibodies have to penetrate due to the deposition of the nascent columella. Additionally, 9 DPA seeds are more sensitive to fixatives and, as such, are more likely to collapse inwards, destroying the remnants of the cytoplasmic column (Figure 5-8C). However, under no circumstances was a 94  microtubule array observed in 9 DPA cells (Figure 5-8, B-C and E), also confirming results observed by TEM. It was our initial hypothesis that these microtubules were involved in the targeting of the mucilage-bearing vesicles. However, work done by Heather McFarlane on mor1-1, a temperature sensitive mutant which results in short, disorganized microtubules at restrictive temperature (Whittington et al., 2001), showed that mucilage was properly deposited even in the absence of well-organized microtubules, though mucilage extrusion was impaired (McFarlane et al., 2008). In addition, at 9 DPA many vesicles are observed fusing with the plasma membrane at the apex of the cytoplasmic column (Figure 5-4, C-D), even though no microtubules are observed. Thus this hypothesis is not supported by our results. An alternate hypothesis was developed, based on comparisons between the arrangement of developing seed coat cells and other cell types with well-defined microtubule arrays. Microtubules have been considered to be important for the maintenance of cell shape, among other functions, in several systems that display differential growth, such as the cells of the expanding regions of the hypocotyl or root, leaf pavement cells and in trichomes (Wasteneys, 2000; Kost and Chua, 2002; Smith, 2003). Measurement of the dimensions of seed coat epidermal cells using Openlab software (Improvision, Inc. Waltham, MA, USA) at 4, 7 and 9 DPA show that seed coat epidermal cells grow in both width and height between 4 and 7 DPA, and then remain unchanged between 7 and 9 DPA (Figure 5-10). Thus, the main axis of growth at 7 DPA can be considered to be the apical-basal axis. It is possible that the cortical microtubule arrangement in 7 DPA cells plays a role in the maintenance of cell shape during the deposition of mucilage. Between 4 and 7 DPA, seed coat epidermal cells are still in the growth phase and, as such, 7 DPA cells should be both wider and taller than 4 DPA cells. This was found to be true, 95  although they changed more in height than in width (Figure 5-10). If a given mutation, such as those found in mor1-1 or mum4, affected the cell’s overall growth, the cell would be smaller, and therfore both the height and width of the cell would likely be reduced equally. If, on the other hand, the cell’s ability to form its proper shape was inhibited, then height and width would not be affected equally, and one may change more, or in a different direction, than the other. The widths and heights of seed coat epidermal cells in resin-embedded sections of cryofixed wild-type cells at 4, 7 and 9 DPA, as well as on 7 DPA cells of the mutants mor1-1 (grown at restrictive temperature, 29°C; mor1-1 samples only were prepared and sectioned, but not measured, by Heather McFarlane) and mum4 were measured and compared (Figure 5-10 and Table 5-1). As both sets of measurements passed the test for normality, One-Way Analysis Of Variance (ANOVA) showed that there were significant differences in both the widths (F(4, 236) = 12.6, p<0.05) and heights (F(4, 236) = 27.5, p<0.05) of the groups. Post-hoc analysis using Bonferroni’s test showed that in the case of cell widths, wild-type 4 DPA cells were significantly smaller than all of the others (Table 5-1 and Figure 5-10B), indicating that the cell is still growing in width until 7 DPA, but does not change size between 7 and 9 DPA. Additionally, neither mutation affected the width of the seed coat epidermal cells. This provides evidence that the overall size of the cells was not affected by the mutations in mor1-1 or mum4. TABLE 5-1: Results of statistical analysis of mean heights and sectioned seed coat epidermal cells, with standard deviations Cell Width Number Mean Standard Group of cells (m) Deviation Wild-type 4 DPA 110 22.6a 7.4 Wild-type 7 DPA 37 28.4b 5.1 b mor1-1 7 DPA grown at 29°C 20 29.1 7.2 b mum4 7 DPA 49 28.9 5.9 Wild-type 9 DPA 25 28.5b 5.7  widths of cryofixed, Cell Height Mean Standard (m) Deviation 14.7c 3.9 20.7d 3.5 d 18.6 3.4 c 15.1 2.7 18.9d 2.9  Letters following width and height measurements indicate results of one-way ANOVA post-hoc analysis using Bonferroni’s test. Values followed by the same letter are not significantly different at the corrected p-value, p<0.0125.  96  When cell heights were analyzed post-hoc (Using Bonferroni’s test), a statistically significant difference was again observed in wild-type cells between 4 and 7 DPA, but not between 7 and 9 DPA (Table 5-1 and Figure 5-10C). This indicates that growth continues in height as well as width only until 7 DPA. Despite predictions that the disrupted microtubules in mor1-1 would drastically affect cell shape, there were no significant differences between seed coat epidermal cells in wild type and mor1-1 at 7 DPA, despite a slight decrease in cell height in the mor1-1 cells (Table 5-1 and Figure 5-10D). On the other hand, mum4 cells at 7 DPA were found to be more similar to wild-type cells at 4 DPA than those at 7 DPA. It should be noted that the vacuole size in wild type and mum4 were compared, and no differences were observed. This implied that the large amounts of mucilage secreted at 7 DPA have an important role in determining cell height, but the presence of organized microtubules is not required.  5.3 Discussion 5.3.1 Targeting mucilage to the correct domain of the mucilage pocket During the development of the Arabidopsis seed coat, the targeted secretion of matrix polysaccharides is not accompanied by a concentration of Golgi stacks near the site of cargo deposition. This indicates that the clustering of Golgi stacks is not a mechanism for the proper targeting of cell wall carbohydrates to their destination in the developing mucilage pocket. The dispersed Golgi stacks observed in this study are contrary to the situation observed in other plant cells where polarized growth has been observed. In root hairs (Sherrier and Vandenbosch, 1994), pollen tubes (Mascarenhas, 1993) and in cell-plate formation during cytokinesis (Nebenführ et al., 2000; Segui-Simarro and Staehelin, 2006), targeted secretion of cell wall products is accompanied by a clustering of Golgi stacks in the region of the cell where deposition is high. In the case of cell plate formation, the overall density of Golgi stacks in meristematic root cells 97  remains relatively constant despite local increases in the number of Golgi stacks near the cell plate (Segui-Simarro and Staehelin, 2006). The non-polarized distribution of Golgi stacks seen here is more similar to the arrangement in diffusely growing interphase plant cells (Boevink et al., 1998; Segui-Simarro and Staehelin, 2006), especially in mucilage-secreting root cap cells, where Golgi stacks are dispersed throughout the cytoplasm (Staehelin et al., 1990). However, even in such cases of diffuse growth, there is some evidence of targeted secretion of macromolecules to specific wall domains, as seen in the case of COBRA (COB) in the periclinal walls of the root (Roudier et al., 2005). If the Golgi stacks are dispersed, yet specifically deposit pectins and other matrix components at the mucilage pocket, then post-Golgi mechanisms for vesicle targeting must be acting to ensure that cell wall products are targeted to the appropriate domain of the plasma membrane. The major plasma membrane targeting complex in yeast and animals is the exocyst (Guo et al., 2000), an eight-protein complex that most likely forms a rod-like structure to facilitate docking of the incoming vesicles at the target membrane (Munson and Novick, 2006). All eight subunits have been identified in Arabidopsis and shown to be important for targeted secretion in developing pollen tubes (Hala et al., 2008). However, whether the exocyst also has a role in diffuse growth has yet to be determined.  5.3.2 Electron-dense vesicles In this study we observed a population of vesicles that were electron dense, labeled with antibodies known to label mucilage, did not bear any morphological resemblance to clathrincoated vesicles, and did not label with antibodies to endocytic markers such as RabA4b and SCAMP1. For these reasons, we have classified these vesicles as secretory. Most of the work that has been done on electron-dense vesicles in plants has been on those observed in pea cotyledons (Hohl et al., 1996; Robinson et al., 1997; Hillmer et al., 2001). 98  In this organ, electron-dense vesicles have been found to begin to form as early as the cis-Golgi, through the observation of electron-dense bulges at the periphery of the cisternae (Robinson et al., 1997; Hillmer et al., 2001). These electron-dense vesicles were shown to label with antibodies to specific storage proteins that are destined for the storage vacuole of the cotyledons (Hohl et al., 1996). In seed coat epidermal cells, it is unlikely that the observed electron-dense vesicles are performing the same role as those of the pea-cotyledon. Seed coat epidermal cells are destined for programmed cell death (Western et al., 2000) and, as such, are not expected to be involved in the storage of proteins and nutrients for germination. Electron-dense vesicles have been observed in developing xylem cells (Cronshaw and Bouck, 1965), a cell type that is also involved in intense polysaccharide secretion prior to programmed cell death. For the same reasons, it is unlikely that storage proteins are the contents of the electron-dense vesicles of developing xylem. However, immunolabeling of these particular vesicles has not been described. In this study we have shown that the immunolabeling of electron-dense vesicles in the seed coat is reduced, compared to labeling in the Golgi apparatus and TGN, but that electrondense vesicles do label with antibodies to specific mucilage components, such as RGI and XG. In the developing root hair, electron-dense vesicles were observed at the root hair tip, but researchers were unable to label them using antibodies (Sherrier and Vandenbosch, 1994). Another study of several plant cell types has shown that both RGI and XG can be labeled in different vesicles, but no distinction was made between electron-dense vesicles and other types of uncoated vesicles (Moore et al., 1991). It is possible that the electron-dense vesicles of the seed coat represent a condensation step in the delivery of polysaccharides to the plasma membrane, which could have the side effect of decreasing labeling of the cargo of these vesicles. It is interesting to note that we were unable to detect XG and RGI epitopes in the same vesicles.  99  This is consistent with previous reports using the same anti-XG and a different polyclonal antibody to RGI (Moore et al., 1991). A related but different hypothesis is that the more electron dense appearance of certain vesicles could be the result of an increased protein component to the cargo of certain vesicles. This could speak to how polysaccharides and proteins destined for the cell wall are sorted and packaged by the Golgi apparatus. Some proteins are believed to be secreted to the mucilage pocket, such as MUM2 (Dean et al., 2007). It would be interesting to compare the distribution of a protein such as MUM2 with the pattern of polysaccharide epitope signals described earlier in this thesis (See Chapter 4). Electron-dense vesicles were observed in higher concentrations in areas of the cell that were distal to the mucilage pocket but were not found in areas adjacent to the mucilage pocket. This result is likely due to the rapid movement of vesicles through the peripheral region of the apical seed coat epidermal cell, and their immediate fusion with the plasma membrane, whereas vesicles in other regions of the cell are still in transit towards their final destination. This scenario is different from that which has been documented in highly polarized cells such as pollen tubes and root hairs. Examination of root hairs by TEM shows a high concentration of vesicles near the root tip, where exocytosis is taking place (Sherrier and Vandenbosch, 1994). Additionally, vesicles in pollen tubes have also been observed to collect at the apex (Bove et al., 2008). This has been attributed in part to the fact that there is a low efficiency of fusion at the pollen tube apex, and that vesicles often cycle through the apex more than once prior to fusion with the plasma membrane.  5.3.3 The role of microtubules in seed coat development In the Arabidopsis seed coat, an array of cortical microtubules is specifically observed to form at 7 DPA, while mucilage is being deposited. This array apparently disappears at 9 DPA, 100  when the secondary cell wall that forms the columella is secreted. Since development at both 7 and 9 DPA requires intense secretion of cargo in a polarized fashion, the role of the cortical microtubule array at 7 DPA it is not entirely clear. My initial was that the array had a role in the proper targeting of vesicles to the growing mucilage pocket. However, studies of the mor1-1 mutant, whose microtubules are short and disorganized, revealed that mucilage was deposited to its proper location even in the absence of well-organized microtubules, although mucilage release was affected (McFarlane et al., 2008). In addition, morphological characterization of 9 DPA seed coat epidermal cells showed a high number of vesicles clustering and fusing with the plasma membrane at the apex of the cytoplasmic column, even though the microtubule array was not observed at this stage (Figure 5-4, C-D). Together, these results indicate that an organized cortical microtubule array is not required for proper targeting of vesicles in the seed coat. During the formation of xylem vessels, microtubules arrays are observed at future sites of secondary cell wall deposition (Oda et al., 2005). As the spiral secondary cell wall is secreted, the microtubule array splits and remains associated with the edges of the growing wall (Oda et al., 2005; Wightman and Turner, 2008). The presence of microtubules under the mucilage pocket at 7 DPA may be analogous with the microtubules observed prior to cell wall formation in xylem, instead of being directly related to the growth of the cell at 7 DPA. However, there are two problems with this hypothesis. First, secondary cell wall is also deposited at the apex of the cytoplasmic column, and yet no microtubules are observed in this area of the cell at 7 DPA (McFarlane et al., 2008). Second, bundles of microtubules have not been observed at the edges of the developing secondary cell wall at 9 DPA (Figures 5-8 and 5-9), therefore it is not likely that the function of the 7 DPA microtubule array is analogous to that of microtubules in xylogenesis.  101  The most likely explanation for the presence of microtubules at 7 DPA is that they are somehow helping to maintain the shape of the cytoplasmic column, until mucilage production finishes and the secondary cell wall of the columella can be laid down. Studies of the size of seed coat epidermal cells at different stages have shown that these cells are still growing isotropically until 7 DPA (Figure 5-10 and Table 5-1). This requires the deposition of new primary cell wall products, as well as the remodeling of the primary wall to allow expansion. Between 7 and 9 DPA, the cell continues to grow anisotropically, but accompanied now by intense mucilage production. The fact that mum4 seed coat epidermal cells are of the appropriate width, but not height, at 7 DPA is an indication that the deposition of mucilage may be integral to the ability of the cell to reach its final size. Despite the fact that this study, as well as the study by Heather McFarlane, found no difference in the shape of cells between wild type and mor1-1, mor1-1 seeds grown partially at restrictive temperature did not release mucilage properly (McFarlane et al., 2008). Thus disruption of microtubules did affect seed development in some way. Given the cell size phenotype in mum4, it is possible that studying the effect of crossing mum4 and mor1-1 on seed coat development would provide further insight into the relationship between mucilage production, microtubules and the shape of Arabidopsis seed coat epidermal cells. Inhibitor studies using microtubule-depolymerizing drugs would also be informative as to the role of microtubules in seed coat development. A series of inhibitor studies were attempted by Heather McFarlane (Samuels lab) but delivery of the drugs to the developing seeds was a major impediment. Removal of seeds from the silique arrested development, whereas ‘feeding’ developing seeds inside the silique via the cut stem resulted in cell death when reagents were applied at concentrations that were high enough to disrupt microtubules. If one could find a way to overcome the difficulties inherent in microtubule inhibitor delivery to the developing seed, 102  inhibitor studies of this kind would be important to aid in understanding the relationship between microtubules and the development of the Arabidopsis seed coat.  103  FIGURE 5-1: Distribution of Golgi stacks in wild-type, 7 DPA seed coat epidermal cells. (A) Schematic representation of how the apical and basolateral regions of seed coat epidermal cells are defined. (B) Comparison of the Golgi stack density in apical and basolateral regions of 7 DPA seed coat epidermal cells. This analysis was based on three separate experiments, which resulted in a total of 24 cells, and 474 Golgi stacks being examined. Error bars represent 95% confidence intervals. Differences were not considered significant (p>0.05). (C) Comparison of regions of 7 DPA cells that label with immunogold (anti-mucilage (CCRC-M36) or antixyloglucan (-XG)), indicating an even distribution of epitopes in all regions of the cell. For this analysis, a total of 31 cells and 287 Golgi stacks using CCRC-M36, and 28 cells and 252 Golgi stacks using -XG were examined, in three separate experiments. Error bars represent the range of percentages obtained in different replicates. (Figure originally published by Young et al., 2008. Copyright American Society of Plant Biologists).  104  FIGURE 5-2: Electron-dense vesicles in 7 DPA seed coat epidermal cells. (A-I) TEM Images of electron-dense vesicles (arrows) in sections of 7 DPA seed coat epidermal cells. (J) Schematic of 7 DPA seed coat epidermal cell showing the location within the 7 DPA cell of each of the images in this figure. a – amyloplast; G – Golgi stack; m – mucilage. Scale bars are 500 nm.  105  FIGURE 5-3: Tomographic slice (underlying image) overlain with a 3-dimensional reconstruction modeled from the whole electron tomogram. Cortical microtubules (yellow) line the plasma membrane (brown). In addition, a trans Golgi network is visible (purple) with several vesicles (pink), including a population of electron-dense vesicles (blue) near the plasma membrane. (inset) Tomographic slice from boxed area, showing morphology of electron-dense vesicles and a clathrin-coated vesicle (arrowhead). Scale bar 250 nm; Inset, 50 nm (With kind permission from Springer Science+Business Media: McFarlane et al., 2008).  106  FIGURE 5-4: Electron-dense vesicles near plasma membrane in columella of 7 and 9 DPA seed coat epidermal cells. (A-B) Oblique sections through columella showing mucilage pocket and underlying cytosol. In both cases electron-dense vesicles can be seen closely associated with cortical microtubule array that underlies the mucilage pocket, as well as near Golgi stacks (arrows). However, no vesicle fusion is observed. (C) Intense vesicle activity is observed at the apex of the 9 DPA columella. (D) Higher magnification of area in box in C. Electron-dense vesicles are obious (arrows) as well as some vesicles that have fused with the plasma membrane (arrowheads). m – mucilage; G – Golgi stack; TGN – trans Golgi network; 2cw – secondary cell wall. Scale bars A, B, D 250 nm; C 1 m.  107  FIGURE 5-5: Double labeling of 7 DPA seed coat epidermal cells using the anti-mucilage RGI CCRC-M36 (15 nm gold, squares) and anti-xyloglucan (-XG, 10 nm gold, circles). (A) Both xyloglucan and CCRC-M36 epitopes can be labeled in the mucilage (m). A single vesicle is observed near the mucilage pocket, labeled with CCRC-M36. (B) A number of vesicles are observed near the primary cell wall at the base of the seed coat epidermal cell, including some which can be positively identified as electron-dense vesicles (arrows). (C) Two Golgi stacks (G), with several vesicles, including electron-dense vesicles (arrows). (D) A vesicular cluster at the base of the cytoplasmic column, overtop of the vacuole (v). Arrows indicate electron-dense vesicles. G – Golgi stack; m – mucilage; v – vacuole; 1cw – primary cell wall. Scale bars are 250 nm. 108  FIGURE 5-6: Immunogold labeling with antibodies to RabA4b in wild-type, 7 DPA seed coat epidermal cells. Anti-RabA4b (-RabA4b) labeling using 15 nm gold (triangles). (A) A Golgi stack (G) with a cluster of vesicles nearby. (B) The region at the base of the cytoplasmic column, between the mucilage (m) and the vacuole (v) with several vesicles. Arrows indicate electrondense vesicles. G – Golgi stack, m – mucilage, v – vacuole. Scale bars are 250 nm.  109  FIGURE 5-7: Distribution of electron-dense vesicles in different regions of wild-type, 7 DPA seed coat epidermal cells. (A-C) TEM images of 7 DPA seed coat epidermal cells; each star represents one electron-dense vesicle. Scale bars represent 5 m. (D) Quantification of distribution of electron-dense vesicles in 7 DPA seed coat epidermal cells. These data have been normalized to account for the fact that systemic variance is 0 in a repeated-measures test such as this. This analysis was based on three separate experiments, which resulted in a total of 10 cells, and 761 electron-dense vesicles. Error bars represent 95% confidence intervals.  110  111  FIGURE 5-8 (overleaf): TEM images of cortical microtubule arrangement under the mucilage pocket at 7 and 9 DPA. (A-D) Images of 7 DPA cells. (A, B) Oblique sections through the 7 DPA mucilage pocket and underlying cytosol of the columella, showing numerous cortical microtubules adjacent to the plasma membrane under the mucilage pocket (arrows). (C) Cross section through cortical microtubules under the mucilage pocket of 7 DPA seed coat epidermal cells showing characteristic morphology and protofilament arrangement. (D) Oblique section through a 7 DPA mucilage pocket and underlying cytosol of columella showing longitudinal view of cortical microtubules array. (E-G) Images of 9 DPA cells. (E) Apex of 9 DPA columella showing intense vesicle activity, but no obvious cortical microtubule array. (F) Higher magnification of apex of developing columella. (G) Oblique section through secondary cell wall deposited below mucilage at 9 DPA, as well as underlying cytosol of columella. Intense vesicle activity is visible, but no cortical microtubule array. Arrows point to microtubules. m – mucilage; 2cw – secondary cell wall. Scale bar A, C-G 250 nm; B, 50 nm.  112  FIGURE 5-9: Confocal images of microtubule bundle distribution in 7 and 9 DPA seed coat epidermal cells. Cells were chemically fixed, permeabilized and immunolabeled with anti- tubulin. (A, D) Confocal image of 7 DPA seed coat epidermal cell labeled with anti-tubulin. Bands of cortical microtubules are obvious in the cytoplasm and the cytoplasmic column. (B, C, E) Confocal image of 9 DPA seed coat epidermal cell labeled with anti-tubulin. Imaging of 9 DPA cells was more variable, as a result of the developing secondary cell wall of the columella. (B) Although cells could sometimes be observed that had fluorescent label, there was no obvious cortical array as observed at 7 DPA. (C) Many 9 DPA cells that appeared to have strong label had the characteristic appearance of a collapsed cell, with a cytoplasmic column that appeared tall and narrow (asterisk, compare to cytoplasmic column of 7 DPA cell outlined in D). (E) Most often, no fluorescence could be observed. One cell in each image has been outlined in white. Scale bars are 15 m.  113  FIGURE 5-10: Height and width of seed coat epidermal cells of different ages and mutant lines. (A) Light microscope images of sections of seed coat epidermal cells stained with Toluidine Blue. Height and width of cells was measured at approximately the centre of each cell. Scale bars represent 15m. (B-C) Mean width (B) and height (C) of seed coat epidermal cells (For values, see Table 5-1). For this analysis, cells were measured from three separate experiments, resulting in a total of 110 wild-type cells at 4 DPA, 37 wild-type cells at 7 DPA, 20 7 DPA mor1-1 cells that were grown at the restrictive temperature (29°C), 49 mum4-1 cells at 7 DPA and 25 wildtype cells at 9 DPA. Error bars represent 95% confidence intervals.  114  6 SECRETION OF MUCILAGE AND MUCILAGE-MODIFYING ENZYMES 6.1 Introduction In Chapters 4 and 5, I investigated how different polysaccharides were produced in the scattered Golgi stacks and packaged into vesicles for delivery to the plasma membrane. This chapter examines how the Golgi apparatus manages the delivery of different cargo molecules to the plasma membrane. The specific goal of this chapter is to examine if mucilage catabolic enzymes are packaged by the Golgi apparatus for delivery to the plasma membrane in the same vesicles that contain their polysaccharide substrates. Although there have been several immunolabeling experiments, including the ones described in this thesis (Chapter 4 and 5), that examine whether a single Golgi stack can produce several different polysaccharide components, very few studies have examined proteins and polysaccharides together. Moore et al. (1991) examined the secretion of glycoproteins and polysaccharides, using an antibody to extensin (Stafstrom and Staehelin, 1988). They found that extensins and polysaccharides could be produced by the same Golgi stacks. However extensins are structural glycoproteins and, as such, are not being secreted with potential substrates if packaged together into vesicles with other compounds. It is possible that this area has not yet been explored because the range of enzymes that are secreted and that are known to modify pectin is small. The largest group of secreted pectin-modifying enzymes is the pectin methylesterases, which remove methyl groups from linear portions of the pectin, homoglalacturonan (HGA), thereby allowing the formation of rigid, calcium cross-linked sections of pectin (Micheli, 2001; Pelloux et al., 2007). Most pectin methylesterases have a leader sequence that is believed to be involved in targeting them to the correct domain of the 115  plasma membrane, in addition to the regulation of enzyme activity during transit (Bosch et al., 2005; Dorokhov et al., 2006). This sequence is cleaved prior to secretion of the pectin methylesterase to the apoplast (Wolf et al., 2009). However, despite the existence of a hypothesis that describes co-secretion of pectin methylesterases and polysaccharides in the same vesicles (Micheli, 2001), no research has been done that can support or refute it. This could be due to the fact that no antibodies exist that can identify pectin methylesterases. In fact, the only available antibody that recognizes an enzyme involved in polysaccharide synthesis or modification is one which recognizes GAUT1, a homogalacturonan galacturonosyltransferase involved in pectin biosynthesis and thus predicted to reside in the Golgi apparatus (Sterling et al., 2006). However, even this antibody has not yet been used for electron microscopic localization of GAUT1. The recent characterization of MUM2, which codes for a -galactosidase that is postulated to modify pectin in mucilage, based on linkage analysis of extracted mucilage in wild type and mum2 (Dean et al., 2007) has provided an ideal enzyme with which to address the cosecretion question. MUM2 has been hypothesized to play a role in the removal of side-chains from RGI, which results in changes in hydration of mucilage. Since it is a secreted enzyme, it must travel through the Golgi apparatus, alongside rhamnogalacturonan I (RGI), its predicted substrate (Dean et al., 2007), prior to its deposition in the mucilage pocket. As Golgi stacks and vesicles bearing mucilage can be identified by the mucilage antibody, CCRC-M36, the Arabidopsis seed coat provides a unique opportunity to examine the mechanism by which the plant secretes enzymes and their substrates using the same organelle.  6.1.1 Previous and related work The mum2 mutant was originally identified in an EMS-mutagenized population (Western et al., 2001). Arabidopsis seeds homozygous for a mutation in MUM2 do not release mucilage because mucilage in mum2-1 is unable to expand when exposed to water (Dean et al., 2007). 116  MUM2 was cloned and shown to encode a -galactosidase (At5g63800). A genomic fragment including 2 kb of DNA sequence upstream of the start codon (ATG) of the At5g63800 gene and 1 Kb downstream of the stop codon was cloned into pCAMBIA1200, creating the plasmid pMUM2g. This vector was shown to be able to complement the mum2-1 mutation, suggesting that this fragment includes the entire MUM2 gene (Dean et al., 2007). The MUM2 gene has 16 exons and 15 introns and encodes a protein 718 amino acids long that includes an N-terminal secretion signal sequence, a glycosyl hydrolase domain near the N-terminus, as well as two putative galactose-binding-like domains near the C-terminus. It has a predicted size of 80 kD. Several alleles of this gene have been identified, including one that is considered to be a naturally occurring mutation in the Shadhara accession (Macquet et al., 2007b). For the project described in this chapter, only mum2-1 was used, which has a point mutation at 1537 bp, resulting in a G-to-A transition. RT-PCR analysis (Dean et al., 2007) indicates that MUM2 is expressed throughout the plant (roots, cauline and rosette leaves, stems and open flowers) as well as in siliques throughout development.  6.1.2 Objectives The secretion of matrix polysaccharide-modifying enzymes poses a major dilemma for plants, as both the enzyme and its potential substrate must pass through the Golgi apparatus prior to secretion to the apoplast. This chapter focuses on how the seed coat epidermal cell secretes mucilage modifying-enzymes, such as the -galactosidase, MUM2, which travels through the Golgi apparatus during the same time-frame as its predicted substrate, RGI-rich mucilage. In order to determine whether enzymes and substrates get targeted to different vesicles, or whether they travel together through the endomembrane system, I constructed a histidine-tagged version of MUM2. Using this tagged version of MUM2 I plan to localize the protein in the Golgi 117  apparatus, vesicles and mucilage of the developing seed coat, and to compare its localization with the distribution of mucilage-specific epitopes, such as those identified by the RGI antibody, CCRC-M36.  6.2 Results In order to examine how the Golgi apparatus manages the secretion of mucilage, and mucilage-modifying enzymes such as MUM2, it was important to be able to localize both MUM2 and its putative substrate, RGI. CCRC-M36 was an ideal probe to localize mucilage RGI. Unfortunately, no antibody that recognizes MUM2 protein is available. Since producing antibodies is costly, time-consuming and uncertain, I decided to use an epitope-tagged version of MUM2. Epitope tags can be added to a protein by genetic engineering, and then localized using commercially available epitope-specific antibodies (Jarvik and Telmer, 1998). Although there are several epitopes that are used in epitope-tagging, I considered only two, 6xHis and cMyc, since we have had experience with each. In addition, a GFP-tagged version of MUM2 (35SMUM2-GFP) had been engineered previously (Dean et al., 2007) and this was also considered for use as an epitope-tagged version of MUM2 for this project. However, the latter was not an ideal choice for two reasons. The 35S promoter has been shown to sometimes result in patchy expression in the seed coat (Young et al., 2008), which could make localization using antibodies problematic. Second, the form of GFP that was used was not acid-resistant and, as such, would not fluoresce in the lower pH environment of the mucilage pocket. For the reasons stated above I chose to engineer a new epitope-tagged MUM2 protein. The 6xHis tag was chosen for two main reasons. First, prior to this decision, I tested several antibodies on cryofixed, embedded, sectioned material from wild-type, 7 DPA seed coats (Figure 6-1) in order to assess the amount of background label associated with use of each antibody. Samples were immunogold-labeled with anti-cMyc (Molecular Probes), or one of two different 118  antibodies to 6xHis (Penta-His, from Qiagen Inc., or an anti-6xHis by Sigma-Aldrich Canada, Ltd.) followed by silver enhancement, to amplify signal from gold particles so that the relative amounts of labeling could be assessed quickly. Interestingly, there were very different levels of background between the two antibodies to 6xHis (Figure 6-1, compare D and E). However, background Penta-His antibody labeling was virtually indistinguishable from the negative control (Figure 6-1, compare A and D) and thus was considered to be the best antibody of all those tested, including the antibody to cMyc. Second, 6xHis is a very small epitope (only 6 amino acids, or 18 nucleotides) compared to cMyc (10 amino acids, or 30 nucleotides), and it can therefore be added to the gene by incorporation into the PCR primer used for amplification of the MUM2 gene fragment, rather than by amplifying the epitope separately and introducing it into the construct as a separate piece, thereby saving time in the cloning process.  6.2.1 Production of MUM2-His Previous work on MUM2 has shown that it is a very large gene (>5kb), with 16 introns and 15 exons (Figure 6-2). This includes a particularly large first intron, 1879 bp in length. Despite the fact that the cDNA was a much more manageable size (2596 bp), the genomic DNA construct, pMUM2g, was used for this project. The reason for this was two-fold. First, there is some evidence to suggest that important regulatory elements may exist within the large first intron (Gill Dean and George Haughn, personal communication). Promoter-GUS constructs engineered using 2 kb of the sequence upstream of the ATG of MUM2 cDNA (i.e. including the 5’UTR) did not result in seed coat expression, despite the fact that mum2 is known to be expressed there (Dean et al., 2007). Second, the pMUM2g construct completely rescued the mum2 phenotype in 26 of 30 independently transformed lines (Dean et al., 2007). Since MUM2 is predicted to have an N-terminal signal sequence for secretion to the apoplast, which is likely to be cleaved during co-translational insertion of MUM2 into the ER 119  (Dean et al., 2007), the 6xHis tag was fused in frame with the C-terminus of the full-length MUM2 genomic construct. My strategy was to find a unique restriction enzyme site upstream of the stop codon of MUM2 that could be used to insert a fragment of DNA that included the 6xHis tag (Figure 6-2A and B). This resulted in some duplication of the 3’ region of the gene in the 3’ untranslated region (UTR) of the new gene (Figure 6-2C). To remove this duplicated region, a second unique restriction enzyme site was found in the 3’ untranslated region (UTR) that could be used to excise a portion of the 3’ UTR (including the duplicated region). The original 3’ UTR would be amplified by PCR, and this could then be inserted by restriction enzyme digestion and ligation. The result would be that the DNA found in pMUM2g would be identical, except for the addition of the codons for the 6xHis tag (Figure 6-2D). PCR was used to amplify a 370 bp fragment of the 3’ end of the gene from the pMUM2g vector. This fragment contained the last 2 exons (15 and 16), an intron and the stop codon (Figure 6-2B). Since the 6xHis tag codons were incorporated into the PCR reverse primer, no extra step was required to add the 6xHis tag to the amplified fragment. The 6xHis was followed by the stop codon, an ApaI site, and an Xba1 site for excision of the duplicated region of the gene. The pMUM2g vector and the fragment were digested using ApaI, and the fragment was inserted at the ApaI site using DNA ligase, and sequenced (Figure 6-2C). This vector, named pM2H, was used to transform mum2-1 plants. Seeds that had been transformed with Agrobacterium tumefaciens were grown on media that selected for hygromycin B resistance, which  is  conferred  by  a  gene  within  the  T-DNA  of  the  pCAMBIA  vector  (http://www.cambia.org/daisy/cambia/home.html). Six independently transformed lines were obtained after selection. As seed coat tissue is maternal (Figure 6-3), it was not until the T2 generation of seeds that rescue of the mum2 phenotype could be determined. Of those six lines, two lines showed complete rescue of the mum2 phenotype (M2H-1a and M2H-1b) two lines 120  were partially rescued (M2H-1c and M2H-1d) and two did not rescue at all (M2H-2a, M2H-2b, Figure 6-4). Based on the growth of the seeds on selective media, as well as the rescue of the phenotype, only the two best lines (M2H-1a, M2H-1b) were examined in the T3 generation for homozygous lines (Figure 6-5). Two sister T3 plants from M2H-1a and two sister T3 plants from M2H-1b were chosen to use for immunolocalization studies as they were homozygous, and completely rescued the mum2 phenotype. An attempt was made to use the engineered XbaI site to remove the duplicated region of the pM2H construct. Despite the fact that sequencing of the plasmid indicated that both the original and the engineered XbaI sites were present in pM2H, several attempts to use XbaI to digest pM2H only cut the vector in a single location. As pM2H successfully rescued the mum2 phenotype in transformed plants despite the small duplicated end of the gene in the 3’UTR, this part of the cloning strategy was not pursued further.  6.2.2 Immunolabeling of MUM2-His and mum2-1 6.2.2.1 MUM2-His immunolabeling Prior to attempting to immunolabel 6xHis tagged MUM2 in the seed coat, several tests were carried out on leaves of transformed plants. The rationale was that since MUM2 is ubiquitously expressed (Dean et al., 2007), leaf tissue could be used to test anti-His antibodies, and determine working concentrations prior to the development of seeds. This also gave me more opportunities to sample tissue per generation, since immunofluorescence experiments were lengthy (2 days to 2 weeks, depending on the protocol), and only a few siliques from each generation could be imaged as a result. As a positive control, a strain of recombinant Pichia pastoris that had been genetically engineered to express a 6xHis tagged MUM2 by Dr. Gill Dean was used. Cell pellet extracts from this yeast line that had been immunolabeled with Penta-His 121  antibodies in Western blots identified protein of the proper size, that was not detected in yeast strains transformed with the empty vector (Dean et al., 2007). Leaves of T2 lines were pressed onto nitrocellulose and immunoblotted, using silver enhancement of immunogold to detect signal (Figure 6-6A). This experiment was not very successful for 2 reasons: Silver enhancement turned out to be a poor choice for detecting the results of an immunoblot; even positive controls did not label well by this method. Also since pressing leaves released cell contents, including chlorophyll, onto nitrocellulose the weak signal of the silver enhancement could not be detected over the intense green colour of the leaf extract (Figure 6-6A). In order to troubleshoot the previous results an antibody screen was done, using only positive and negative controls of yeast extracts (Figure 6-6B). Pellets and supernatants of extracts were dot blotted onto nitrocellulose so that proteins were not denatured, to ensure that native protein could be positively identified by the anti-His antibodies, and immunolabeled using 3 antibodies available from Qiagen, Inc.: RGS-His, Penta-His and Tetra-His. Epitopes that can be recognized by these three antibodies are in Table 6-1. This time, detection of primary antibody was done by chromogenic detection of secondary antibodies conjugated to alkaline phosphatase. Results were consistent with previous experiments with these yeast strains. A strong signal was identified in the cell pellet extract, but not the supernatant, of P. pastoris strains that carried MUM2-His by both Penta-His and Tetra-His (Figure 6-6B). P. pastoris that had been transformed with the empty vector had faint labeling in the pellet, and none in the supernatant. RGS-His labeling was consistent with the other two antibodies, but was not as strong. This was expected, considering the epitopes of RGS-His (Table 6-1) and, as such, RGSHis was not considered in future studies. Based on the results of yeast dot blots, leaf blots were again attempted (Figure 6-6C). This time, dots of leaf extracts were used instead of leaf presses, as the amount of chloropohyll 122  in extracts was lower in relative terms, and the green colour of extracts was not as strong. Detection of label was done by chromogenic methods, using secondary antibodies conjugated to alkaline phosphatase. Unfortunately, despite strong label in the control yeast, especially in the cell pellet extract of the MUM2-His, labeling of leaf extracts did not give a positive signal when labeled with antibodies to 6xHis (Figure 6-6C). Virtually no differences could be seen between transgenic plant lines that carried MUM2-His and plants that did not (including transgenic lines carrying the original pMUM2g vector, mum2-1 and wild-type plants). TABLE 6-1: Epitopes recognized by RGS-His, Penta-His and Tetra-His1 Peptide Penta-His Tetra-His RGS-His xHHHHHHx (6xHis) + + xxHHHHHx (5xHis) + + xHxHHHHx (4xHis) + xHHHHxHx (4xHis) + xRGSHHHHx (RGS+4xHis) + + xRGSHHHHHHx (RGS+6xHis) + + + xHxHHxHx 1  Based on information in the QIAexpress Detections and Assay Handbook (Qiagen, Inc, Oct 2002).  Despite limited success with immunoblot experiments, the ultimate goal was to use MUM2-His to localize the MUM2 protein in planta, thus immunofluorescence and confocal microscopy were still attempted. Variability in detection capabilities of antibodies in different types of experiments is a relatively well-known occurrence. A study that compared different antiHis antibodies (including Sigma-His, RGS-His, Tetra-His and Penta-His) found that the different antibodies were highly variable in their capacity to bind recombinant proteins tagged with 6xHis under different conditions (Debeljak et al., 2006). They also found that of the four different antibodies tested, the Tetra-His antibody was found to be the most reliable under different conditions. For this reason, and because the Tetra-His antibody had the best signal to noise ratio in dot blots of yeast extracts (Figure 6-6B), the Tetra-His antibody was chosen for immunofluorescence experiments.  123  Developing seed coats of T2 lines were cryofixed at 7 DPA by high pressure freezing, embedded in resin and sectioned for light microscopy. Samples were immunolabeled using the Tetra-His antibody and secondary antibodies conjugated to fluorophores, and imaged by epifluorescence (Figure 6-7). As a negative control, the same experiment was performed using wild-type 7 DPA seeds instead of one of the transgenic lines. All transgenic T2 lines had a labeling pattern that was consistent with the negative controls. Since yeast extracts bearing undenatured MUM2-His could be labeled in dot blots using the Tetra-His antibody, it was possible that tissue preparation (i.e. resin embedding) was affecting ability of the Tetra-His antibody to bind to the antigen and the experiment was repeated using whole seeds instead of resin embedded sections. Developing seeds from homozygous T3 lines (instead of T2) were used for whole seed immunofluorescence since the two copies of MUM2-His in homozygous lines would increase the chances of successful MUM2-His labeling (Figure 6-3). 7 DPA seeds were chemically fixed, permeabilized and labeled with the Tetra-His antibody as the primary antibody; secondary antibodies were conjugated to fluorophores for fluorescent detection via confocal laser microscopy (Figure 6-8). Negative controls included 7 DPA mum2-1 seeds, labeled with the Tetra-His antibody, and the positive control was mum2-1 labeled with anti-tubulin. Consistent with previous results, no differences could be observed between negative controls and transgenic plants expressing 6xHis-tagged MUM2. Since positive controls successfully and consistently labeled microtubules in mum2-1 seed coat epidermal cells, the only conclusion that can be drawn is that MUM2-His labeling was not successful under our conditions  6.2.2.2 Characterization of mum2-1 The loss of the putative MUM2 -galactosidase could lead to changes in the polysaccharides of the mucilage that could be detected using cell wall probes. It is possible that the mutation in mum2-1 affected the binding profile of anti-pectin antibodies and since the 124  MUM2-His lines were in the mum2-1 background, it was important to determine if this genotype had altered binding to anti-pectin probes such as CCRC-M36, which was the antibody of choice for tracking mucilage. Therfore, wild-type and mum2-1 seeds at 7 and 9 days post anthesis (DPA) were cryofixed, resin embedded, sectioned and immunolabeled for fluorescence microscopy. Several antibodies were tried that had been used previously to characterize wildtype mucilage (CCRC-M36, CCRC-M1, CCRC-M2, CCRC-M7, JIM5, JIM7, LM5 and LM7. See Table 4-1 for details of antigens). In addition, another antibody to RGI side chains, LM6 was used. LM6 specifically binds arabinan side chains ((15)--L-arabinan) of RGI (Willats et al., 1998). No antibodies showed any differences at 7 DPA between wild type and mum2-1, including CCRC-M36 and LM6 (data not shown). At 9 DPA, only LM6 showed a difference in labeling (Figure 6-9). LM6 labeled very little in wild type seeds at 9 DPA, and nothing in the seed coat (Figure 6-9C). In mum2-1, a thin lining of fluorescent labeling was observed on either side of the developing secondary cell wall (Figure 6-9D). Despite the fact that MUM2 is a galactosidase, and LM6 labels arabinan side chains, this result is not entirely surprising, as linkage analysis of mum2 mucilage showed an increase in (15)--arabinan, among others, when compared to wild type (Dean et al., 2007).  6.3 Discussion In this chapter my objective was to examine how polysaccharide-modifying enzymes and their substrate mucilage transit through the Golgi apparatus on their way to the mucilage pocket. To do this I constructed an epitope-tagged version of MUM2, with the goal of using it for immunolabelling studies alongside antibodies to mucilage polysaccharides, such as CCRC-M36. Transgenic mum2-1 Arabidopsis lines bearing the pM2H vector completely rescued the mum2 phenotype, indicating that the genetically engineered MUM2 protein bearing the 6xHis tag is fully active and capable of performing the functions of the original MUM2 protein. Additionally, 125  P. pastoris cells bearing enzymatically active MUM2-His have been shown to consistently label with three different antibodies that recognize the 6xHis tag, not only in the experiments described here, but in previous experiments as well (Dean et al., 2007). Despite the functionality of the MUM2-His protein in Arabidopsis, immunolabeling of the 6xHis-tagged protein was not successful.  6.3.1 Troubleshooting MUM2-His labeling in planta There are several possibilities that could explain why the antibody labeling was not successful. First, it is possible that the 6xHis tagged is hidden or modified in planta during the normal production of the MUM2 protein. There is a precedent for secreted polysaccharidemodifying enzymes being modified as a part of normal function. Pectin methylesterases are formed originally as a pro-protein, with a leader sequence that must be removed for the enzyme to become functional (Micheli, 2001; Pelloux et al., 2007) it is conceivable that MUM2 would follow a similar pathway and that a portion of the protein, including the 6xHis tag, is somehow removed or altered during protein processing. However, there is no evidence of such a leader region in the predicted protein structure, and the His-tagged MUM2 that was transfected into yeast was shown to maintain enzymatic function, as well as being reactive to anti-His antibodies (Dean et al., 2007 and Figure 6-6). Therefore, this hypothesis seems unsupported by the data. Another possibility is that despite the fact that the protein is expressed in all tissues and throughout seed development, it was not present in a form that could be recognized by the antibody in these experiments. In the case of leaf blots, both leaf presses and leaf extracts, MUM2-His could have been retained in the tissue, instead of being released onto the nitrocellulose. Considering that MUM2-His is maintained in cell pellets in yeast extracts (Figure 6-6) this is not unreasonable. In the case of immunofluorescent labeling of MUM2-His in planta, it is possible that the chemicals involved in sample preparation, such as glutaraldehyde, in the 126  fixation step in whole seeds, or the LR White resin used for sectioned material, binds to the 6xHis tag in such a way that the antigen can non longer be recognized by the antibody. Additionally, cryofixed, resin embedded samples of Arabidopsis stems bearing a cMyc-tagged version of a lipid transfer protein also did not yield labeling above background levels, despite the fact that the tagged version of the protein was able to rescue the mutant phenotype (Allan DeBono and Lacey Samuels, personal communications). Considering that two different proteins, tagged with two different epitope tags and constructed by different members of the Samuels Lab, could not be localized by antibodies to their epitopes, one can only deduce that the problem is more likely to be in the sample preparation and not necessarily in the epitope-tagged proteins themselves. During the course of the project described in this thesis, several attempts were made to overcome the problem of antigenicity in resin-embedded material. The freeze-substitution medium was altered to decrease or omit glutaraldehyde. Different resins were tried at different temperatures, including low-temperature embedding resins such as HM20 and K4M. Resins were even omitted altogether, by attempting whole-mount immunoflourescence (Figure 6-8). The problem of antigenicity was compounded by the fact that seed tissue is very difficult to preserve. The cuticle can cause tissue to separate from surrounding resin (as was observed with low-temperature embedding techniques) making it impossible to preserve specimens well enough for imaging. Despite several attempts, none were successful. However, a new method of crypreservation, geared specifically toward difficult to preserve specimens, such as plant tissue, has been developed and shows promise (Ripper et al., 2008). This method uses a combination of cryofixation and the Tokuyasu method for cryo-sectioning (Tokuyasu, 1997), in order to preserve antigenicity and structural integrity of samples while minimizing exposure to strong fixatives or resins. Samples are kept a low temperatures and only weakly fixed by freeze127  substitution, rehydrated, sucrose infiltrated, refrozen, cryo-sectioned and transferred to a TEM grid, immunolabeled, fixed and post-stained and then imaged (Ripper et al., 2008). Using this method, several sensitive antigens, such as the syntaxin KNOLLE (Lauber et al., 1997), were shown to label in a superior fashion, even in hard to preserve tissue such as the Arabidopsis embryo (Ripper et al., 2008). I feel that despite the complicated nature of this methodology, it could provide superior antibody labeling of epitope tagged proteins. I suggest that in the case of the MUM2-His project, that P. pastoris strains could be used in order to trouble shoot this experiment, as labeling capacity has been established using several anti-His antibodies, and yeast are much easier to preserve cryogenically.  6.3.2 Exploration of the mum2-1 phenotype The experiments described in this chapter, which further characterize the phenotype of mum2-1 seeds coats, provide valuable insight into the function of the MUM2 protein. The difference in LM6 labeling in the cell wall of mum2-1 mutants and wild type at 9 DPA suggests MUM2 has finished acting on its substrate by this stage of development. Since the epitope for LM6 ((15)--L-arabinan) is a side chain of RGI, and CCRC-M36 is predicted to label the backbone of RGI in mucilage, LM6 and CCRC-M36 double labeling may provide further insight into when MUM2 becomes functional. If LM6 epitopes can be observed in the Golgi apparatus, but not in vesicles or the seed coat, then it might be an indication that MUM2 is acting in the Golgi apparatus prior to secretion to the mucilage pocket. An additional benefit to exploration of the phenotype of mum2-1 would be that it could provide insight into the electron-dense vesicles described in Chapter 5 (See section 5.2.2 for details). One hypothesis for the appearance of electron-dense vesicles is that the darker staining is due to an increased protein concentration in these vesicles, relative to other less electron-dense  128  vesicles observed in the cytoplasm. Since MUM2 is a secreted protein, exploration of mum2-1 seed coat cells by high resolution TEM might reveal a decrease in electron-dense vesicles.  129  FIGURE 6-1: Testing of anti-epitope antibodies for background labeling. Cryofixed samples of wild-type 7 DPA seeds were resin-embedded, sectioned and immunogold labeled for TEM. Immunogold particles were amplified by silver enhancement so that the relative amounts of signal could be observed more easily. (A) Negative control. No primary antibody. (B) Positive control. CCRC-M36 antibody was used as a positive control as its labelling pattern has been well characterized. (C) Anti-cMyc antibody (Molecular Probes). (D) Penta-His antibody (Qiagen, Inc.) (E) Anti-His (Sigma-Aldrich Canada, Ltd). Scale bars are 5 m.  130  FIGURE 6-2: Cloning strategy for the production of 6xHis tagged MUM2. (A) pMUM2g vector was used to produce MUM2-His. MUM2 has 16 exons and 15 introns (Gene structure drawn here is based on Dean et al., 2007; Macquet et al., 2007b). The first intron is very large, >1Kb. The site of the mum2-1 G-to-A transition is marked on original fragment, as well as the sites of the unique restriction enzyme sites used, ApaI and XbaI. This DNA fragment was inserted into the pCAMBIA1200 vector to form the pMUM2g vector (Dean et al., 2007). (B) PCR was used to amplify a 370 bp section of DNA, which included the 3’ end of the MUM2 coding region, 6xHis, an ApaI restriction enzyme site, and an XbaI restriction enzyme site. (C) The fragment was inserted into pMUM2g at ApaI site. (D) XbaI restriction enzyme sites could be used to remove duplicated region of 3’MUM2 gene, if necessary.  131  FIGURE 6-3: Schematic of origin and genotype of seed tissues in each generation. Seed coat tissue is maternal, and thus is always from the same generation as the mother plant, despite the embryo being of the next generation.  132  133  FIGURE 6-4 (overleaf): T2 seed phenotypes and seedling growth on selective medium. T2 seeds were stained with Ruthenium Red, in order to determine phenotype of seeds, prior to growing them on selective AT media. (A, C, E, G) Controls. (A) Wild-type seeds all extrude mucilage, which stains clearly with Ruthenium Red. However, none of these plants is capable of growing on selective media. (C) mum2-1 seeds do not extrude mucilage, and also do not grow on selective media (not shown). (E, G) mum2-1 seeds transformed with pMUM2g vector, as control. 2 independent T1 lines grew on selective media, but only one rescued the phenotype. (B, D, F, H, I) mum2-1 seeds transformed with pM2H. 6 independent T1 lines grew on selective media. 2 lines rescued the phenotype (B, D), 2 lines partially rescued (F, H) and 2 did not rescue the phenotype (I, J). N/S – Not shown.  134  FIGURE 6-5: T3 seed phenotypes and seedling growth on selective media. T3 seeds were stained with Ruthenium Red, in order to determine the phenotype of seeds, prior to growing them on selective AT media. Several plants of each line were grown, in order to find a homozygous line. T3 seeds and seedlings shown are all homozygous, and completely rescue the mum2 phenotype. Thus these lines were used for immunolocalization of His-tagged proteins.  135  136  FIGURE 6-6 (overleaf): Immunoblotting of leaf and transformed Pichia pastoris extracts (that have not been denatured) to detect 6xHis epitope. (A) P. pastoris extracts (top row) and leaf presses of T2 lines (bottom two rows). Leaves were pressed onto nitrocellulose, and immunolabeled using Penta-His as the primary, and gold-conjugated secondary antibodies. Secondary antibodies conjugated to colloidal gold were detected by silver enhancement. (B) Dot blots of P. pastoris extracts, using 3 different anti-His primary antibodies (listed along bottom row), alkaline phosphatase-conjugated secondary antibodies, and chromogenic detection of secondary antibodies. (C) Dot blots of T2 leaf extracts, using 2 different anti-His primary antibodies, alkaline phosphatase-conjugated secondary antibodies, and chromogenic detection of secondary antibodies. In all panels, GS115-s indicates supernatant from P. pastoris transformed with empty vector; GS115-p indicates pellet from P. pastoris transformed with empty vector; M2H-s indicates supernatant from P. pastoris transformed with MUM2-His; M2H-p indicates pellet from P. pastoris transformed with MUM2-His. Similarly, in all panels, M2H-1a, M2H-1b (or -1b), M2H-1c (or -1c), M2H-1d (or -1d), M2H-2a (or -2a), M2H-2b (or -2b) indicate independent T2 transgenic lines carrying MUM2-His; M2-1a, M2-2a indicate T2 transgenic lines carrying pMUM2g vector; wt is Wild type (Col-2); mum2-1 is untransformed mum2-1 mutant background.  137  FIGURE 6-7: Immunoflourescent labeling of cryofixed, resin embedded sections of 7 DPA seeds of T2 lines of plants transformed with pM2H vector. (A-B) Positive controls. The antimucilage, CCRC-M36, was used to label wild-type (A) and mum2-1 (B) mucilage at 7 DPA. (C) Negative control, Wild type seeds at 7 DPA. (D-E) Transgenic lines at 7 DPA. Scale bars are 100m.  138  FIGURE 6-8: Confocal images of developing T3 seeds at 7 DPA immunolabeled with the TetraHis antibody. Samples were chemically fixed, permabilized and labeled with the Tetra-His antibody. (A-C) Background labeling of non-seed coat cells. Some immunofluorescent label was observed in non-seed coat tissues, but no differences could be observed between control (A) and T3 lines (B, C). (D-H) Immunofluorescent labeling of 7 DPA seed coat cells. (D) Positive control. Anti--tubulin was used to immunolabel microtubules in mum2-1. (E-G) Transgenic lines bearing MUM2-His. Footprint of a seed coat epidermal cell is outlined in white. Scale bars are 15 m.  139  FIGURE 6-9: Immunolabeling of wild-type and mum2-1 seeds at 9 DPA. Seeds were cryofixed, resin-embedded and sectioned. (A) Wild-type seeds labeled with CCRC-M36, which labels RGIlike polysaccharides in mucilage. (B) mum2-1 seeds labeled with CCRC-M36. (C-D) Wild-type (C) and mum2-1 seeds (D) labeled with LM6, which labels arabinan side chains of RGI. Seed coat is inside white lines. Scale bars in A, B are 100 m; in C and D, bars are 15m.  140  7 CONCLUSIONS AND FUTURE DIRECTIONS Pectin is a commercially important polysaccharide as well as a critical component of plant cell walls. It is synthesized in the Golgi apparatus, along with other matrix polysaccharides and then secreted to the apoplast by vesicular transport. This thesis describes research that addresses an important, but largely overlooked aspect of cell wall biology, namely how the fragmented Golgi stacks of the Golgi apparatus function individually and as a collective to produce matrix polysaccharides such pectin, as well as how these polysaccharides are packaged into vesicles for transport to a specific domain of the cell wall. Several new discoveries have arisen from this work and have been published (McFarlane et al., 2008; Young et al., 2008; See page 43 for full citation). The major conclusions of this thesis are listed below, along with suggested areas of future study. Research described in Chapter 3 confirms that the morphology of Golgi stacks is dependant on the amount of product. Though a correlation has been observed in the past between the amount of secretion and Golgi stack morphology, this is the first time that a direct connection between amount of product and Golgi stack ultrastructure has been made. The seed coat epidermal cells of Arabidopsis thaliana were ideal for addressing this question, as levels of secretion change dramatically during differentiation. The obvious cellular rearrangements that accompany changes in secretion make it easy to accurately determine the stage of differentiation of a given cell and, as such, scoring Golgi morphology is also more accurate. Also, mucilage deficient mutants such as mum4 have allowed for the uncoupling of product and Golgi morphology for the first time, allowing for this important discovery. The mum4 mutation was also integral in a second conclusion, that the number of Golgi stacks within the cell is developmentally programmed. Though cell cycle changes in number had been observed in the past (Garcia-Herdugo et al., 1988; Gonzalez-Reyes et al., 1988; Segui-Simarro and Staehelin, 141  2006) there was no way to separate the amount of polysaccharide being produced from programmed regulation of Golgi stack numbers that might occur. The fact that in mum4 the number of Golgi stacks did not change provided conclusive evidence for the first time that Golgi stack duplication is not merely an effect of increased secretion, but also specific component of the developmental program of the cell. A major question in plant cell biology has been whether the fragmented stacks of the Golgi apparatus can perform different functions within the cell, or whether they work collectively to produce the polysaccharides that are required at any given stage. As mucilage is composed of more than one type of matrix polysaccharide, antibody labeling of Golgi stacks during mucilage production has shown that the majority of Golgi stacks are involved in mucilage production. This work is described in Chapter 4 and provides important evidence that the multiple Golgi stacks of the plant cell are in fact a single unit that act in unison to meet the secretory demands of the cell. According to the growth continuum concept, tip growth and diffuse growth can be considered opposite extremes of a single continuum on which all plant growth falls (Wasteneys and Galway, 2003; Wasteneys and Ambrose, 2009). The work described in Chapter 5 supports that idea, as polarized secretion in seed coat epidermal cells occurred in the absence of clustered Golgi stacks. This is important, as cases such as tip growing root hairs and pollen tubes, as well as during the formation of the cell plate, Golgi stacks are observed to cluster near the site of secretion. In case of the Arabidopsis seed coat, secretion is also obviously polarized as mucilage is only deposited in a specific region of the apical domain of the cell, but there is no evidence of Golgi clustering near the mucilage pocket, much like what is observed in diffusely growing cells. This supports the idea that diffuse growth can result in polarized deposition of  142  molecules, such as that observed in COBRA (Roudier et al., 2005) even though there is no obvious polarization of the Golgi apparatus. With regards to post-Golgi traffic to the plasma membrane (as visualized in Figure 1-1), Chapter 5 also details work that identified a population of electron-dense vesicles as secretory vesicles, based on their contents and their morphology. The fact that there is a morphological characteristic that can be used to identify secretory vesicles after they have left the Golgi apparatus could be important in the study of post-Golgi traffic. However, further research on electron-dense vesicles in the Arabidopsis seed coat could provide important insight into how cargo destined for the cell wall is sorted and packaged for transport. This could be especially important, considering that enzymes known to act upon cell wall components, such as the mucilage modifying enzyme MUM2, must travel through the Golgi apparatus along with their substrates. One possible explanation is that electron-dense vesicles have a higher protein content than vesicles that are less densely stained (See section 5.3.2). Changes in the vesicle profile of mum2-1 seed coat epidermal cells could address this question, as there is a known decrease in the amount of protein being secreted in mum2-1, which might be reflected by a decrease in electrondense vesicles. Alternately, if MUM2 function is important to allow for condensation of polysaccharides within the vesicle, the ratio of electron-dense vesicles to other types of vesicles might reflect this change. Comparisons of vesicle labeling using the anti-mucilage RGI antibody, CCRC-M36, with the RGI sidechain antibody, LM6, might also give an indication of when MUM2 is expected to begin performing its putative function, namely the removal of side chains from RGI. At 7 DPA, seed coat cells have an interesting array of cortical microtubules adjacent to the plasma membrane below the mucilage pocket. Work done by Heather McFarlane (McFarlane et al., 2008), in conjunction with work described in this thesis (Section 5.2.4) refutes the 143  hypothesis that these microtubules are involved in the targeting of mucilage bearing vesicles. A second hypothesis was formed, stating that the 7 DPA cortical microtubule array was important for the maintenance of cell shape. Though examination of the temperature sensitive microtubule organizing mutant, mor1-1 showed very little difference in cell size, examination of mum4 revealed that cell size is dependent on mucilage production. Despite the fact that studies with mor1-1 have been negative thus far, it is currently unclear whether this should be construed as evidence that microtubules play no role in the determination of chape in the seed coat epidermal cells. If the problem of delivery of microtubule depolymerizing drugs could be resolved (See section 5.3.3), inhibitor studies would provide powerful evidence for or against microtubules being integral to the maintenance of cell shape. As mucilage also appears to be important to proper formation of the seed coat epidermal cell, crossing mor1-1 into mucilage mutants such as mum4 or mum2 could provide further evidence for the inter-relationship between mucilage, microtubules and the shape of seed coat cells. Although the work described in this thesis led to several important discoveries, there are also several questions that remain, the most important of which is the question of how mucilage and mucilage modifying enzymes are secreted by the Golgi apparatus? Although an epitopetagged version of MUM2 was constructed that completely rescued the mum2 phenotype, immunolocalization of MUM2-His in the seed coat was unsuccessful. Based on results from other epitope-tagged protein localization experiments in the Samuels lab, it is likely that much of the difficulty stems from sample preparation for immunolabeling, and not necessarily the epitope tags themselves. A new method has recently been suggested for cryopreservation of sensitive materials that combines cryofixation with the Tokuyasu cryo-sectioning method (Ripper et al., 2008). It is very likely that this method could be effectively used for immunolabeling of 6xHis, and that the Pichia pastoris strains bearing the MUM2-His contructs that were used as a control 144  are ideal for trouble shooting this protocol. Yeast is considered to be relatively easy to cryo-fix, and immunolabeling has already been successful in dot blots (Figure 6-6) and in Western blots (Dean et al., 2007). Once the protocol was successful using P.pastoris seed coat epidermal cells could again be attempted. Another intriguing aspect of seed coat ultrastructure is that mucilage is being deposited in vast quantities to the mucilage pocket, and yet vesicle fusion has not been observed during this stage of development. On the other hand, vesicle fusion is obvious during the deposition of the columella at 9 DPA, after mucilage production has finished. It would be interesting to understand why this is so. Freeze fracture could be used to examine the plasma membrane to look for vesicle fusion profiles, as identified previously using suspension-culture cells (Staehelin and Chapman, 1987). Also, proteins involved in targeting are beginning to be identified by other laboratories, including the components of the exocyst (Hala et al., 2008). If one of these mutants have a seed coat phenotype, the seed coat would be a powerful system in which to examine the role of any targeting mutants, for the reasons stated previously. The work in Chapter 4 shows that the contents of the vesicles differ between 7 and 9 DPA, since more cellulose is required for the formation of the rigid columella at 9 DPA, it is possible that one of the major differences between vesicle traffic during mucilage and the columella production is the fact that cellulose synthase complexes must be deposited and assembled in the plasma membrane to form the columella. Specific compartments have recently been observed to transport cellulose synthases to the plasma membrane (Crowell et al., 2009). Microarray data comparing different stages of seed development (Dean, Ahad and Haughn, unpublished results) could be used to determine which cellulose synthases are upregulated at 9 DPA, and then examine the traffic of these complexes by using GFP-tagged versions of those cellulose synthases. 145  8 BIBLIOGRAPHY Andreeva, A.V., Kutuzov, M.A., Evans, D.E., and Hawes, C. (1998). The structure and function of the Golgi aparatus: A hundred years of questions. J Exp Bot 49, 1281-1291. Battey, N.H., James, N.C., Greenland, A.J., and Brownlee, C. (1999). Exocytosis and Endocytosis. Plant Cell 11, 643-660. Beeckman, T., De Rycke, R., Viane, R., and Inze, D. (2000). Histological study of seed coat development in Arabidopsis thaliana. J Plant Res 113, 139-148. Bibikova, T.N., Blancaflor, E.B., and Gilroy, S. (1999). Microtubules regulate tip growth orientation in root hairs of Arabidopsis thaliana. Plant J 17, 657-665. Blatt, M.R., Leyman, B., and Geelen, D. (1999). Tansley Review No. 108: Molecular events of vesicle trafficking and control by SNARE proteins in plants. New Phytologist 144, 389418. Bloom, G.S., and Goldstein, L.S. (1998). Cruising along microtubule highways: how membranes move through the secretory pathway. J Cell Biol 140, 1277-1280. Bock, J.B., Matern, H.T., Peden, A.A., and Scheller, R.H. (2001). A genomic perspective on membrane compartment organization. Nature 409, 839-841. Boevink, P., Oparka, K., Santa Cruz, S., Martin, B., Betteridge, A., and Hawes, C. (1998). Stacks on tracks: the plant Golgi apparatus traffics on an actin/ER network. Plant J 15, 441-447. Bolwell, G.P., and Northcote, D.H. (1983). Arabinan synthase and xylan synthase of Phaseolus vulgaris. Biochem J 210, 497-507. Bosabalidis, A.M. (1985). A possible new mode of dictyosome duplication in plant cells. Tissue & Cell 17, 279-286. Bosch, M., Cheung, A.Y., and Hepler, P.K. (2005). Pectin Methylesterase, a regulator of pollen tube growth. Plant Physiol 138, 1334-1346. Bourquin, V., Nishikubo, N., Abe, H., Brumer, H., Denman, S., Eklund, M., Christiernin, M., Teeri, T.T., Sundberg, B., and Mellerowicz, E.J. (2002). Xyloglucan endotransglycosylases have a function during the formation of secondary cell walls of vascular tissues. Plant Cell 14, 3073-3088. Bouton, S., Leboeuf, E., Mouille, G., Leydecker, M.T., Talbotec, J., Granier, F., Lahaye, M., Hofte, H., and Truong, H.N. (2002). QUASIMODO1 encodes a putative membrane-bound glycosyltransferase required for normal pectin synthesis and cell adhesion in Arabidopsis. Plant Cell 14, 2577-2590. 146  Bove, J., Vaillancourt, B., Kroeger, J., Hepler, P.K., Wiseman, P.W., and Geitmann, A. (2008). Magnitude and direction of vesicle dynamics in growing pollen tubes using spatiotemporal image correlation spectroscopy and fluorescence recovery after photobleaching. Plant Physiol 147, 1646-1658. Brandizzi, F., Snapp, E.L., Roberts, A.G., Lippincott-Schwartz, J., and Hawes, C. (2002). Membrane Protein Transport between the Endoplasmic Reticulum and the Golgi in Tobacco Leaves Is Energy Dependent but Cytoskeleton Independent: Evidence from Selective Photobleaching. Plant Cell 14, 1293-1309. Burgert, I. (2006). Exploring the micromechanical design of plant cell walls. Am. J. Bot. 93, 1391-1401. Campanoni, P., and Blatt, M.R. (2007). Membrane trafficking and polar growth in root hairs and pollen tubes. J Exp Bot 58, 65-74. Cannon, M.C., Terneus, K., Hall, Q., Tan, L., Wang, Y., Wegenhart, B.L., Chen, L., Lamport, D.T.A., Chen, Y., and Kieliszewski, M.J. (2008). Self-assembly of the plant cell wall requires an extensin scaffold. Proceedings of the National Academy of Sciences 105, 2226-2231. Carpita, N.C., and Gibeaut, D.M. (1993). Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J 3, 1-30. Caspar, T., Huber, S.C., and Somerville, C.R. (1985). Alterations in growth, photosynthesis, and respiration in a starchless mutant of Arabidopsis thaliana (l.) deficient in chloroplast phosphoglucomutase activity. Plant Physiol 79, 11-17. Clough, S.J., and Bent, A.F. (1998). Floral dip: a simplified method for Agrobacteriummediated transformation of Arabidopsis thaliana. Plant J 16, 735-743. Cole, N.B., and Lippincott-Schwartz, J. (1995). Organization of organelles and membrane traffic by microtubules. Curr Opin Cell Biol 7, 55-64. Collings, D.W., and Wasteneys, G.O. (2005). Actin microfilament and microtubules distribution patterns in the expanding root of Arabidopsis thaliana. Can J Bot 83, 579590. Cosgrove, D.J. (2000). Loosening of plant cell walls by expansins. Nature 407, 321-326. Cosgrove, D.J. (2003). Expansion of the plant cell wall. In The Plant Cell Wall, J.K.C. Rose, ed (Oxford: Blackwell Publishing. Cosgrove, D.J. (2005). Growth of the plant cell wall. Nat Rev Mol Cell Biol 6, 850-861. Craig, S., and Staehelin, L.A. (1988). High pressure freezing of intact plant tissues evaluation and characterization of novel features of the endoplasmic reticulum and associated membrane systems. European Journal of Cell Biology 46, 80-93. 147  Cronshaw, J., and Bouck, G.B. (1965). The fine structure of differentiating xylem cells. J. Cell Biol. 24, 415-431. Crowell, E.F., Bischoff, V., Desprez, T., Rolland, A., Stierhof, Y.-D., Schumacher, K., Gonneau, M., Hofte, H., and Vernhettes, S. (2009). Pausing of Golgi Bodies on Microtubules Regulates Secretion of Cellulose Synthase Complexes in Arabidopsis. Plant Cell, tpc.108.065334. Dauwalder, M., Whaley, W.G., and Kephart, J.E. (1969). Phosphatases and differentiation of the Golgi apparatus. J Cell Sci 4, 455-497. De Matteis, M.A., and Luini, A. (2008). Exiting the Golgi complex. Nat Rev Mol Cell Biol 9, 273-284. Dean, G.H., Zheng, H., Tewari, J., Huang, J., Young, D.S., Hwang, Y.T., Western, T.L., Carpita, N.C., McCann, M.C., Mansfield, S.D., and Haughn, G.W. (2007). The Arabidopsis MUM2 gene encodes a beta-galactosidase required for the production of seed coat mucilage with correct hydration properties. Plant Cell 19, 4007-4021. Debeljak, N., Feldman, L., Davis, K.L., Komel, R., and Sytkowski, A.J. (2006). Variability in the immunodetection of His-tagged recombinant proteins. Anal Biochem 359, 216-223. Denecke, J., Botterman, J., and Deblaere, R. (1990). Protein Secretion in Plant Cells Can Occur via a Default Pathway. Plant Cell 2, 51-59. Doblin, M.S., Vergara, C.E., Read, S., Newbigin, E., and Bacic, A. (2003). The Plant Cell Wall: making the bricks. In The Plant Cell Wall, J.K.C. Rose, ed (Oxford: Blackwell Publishing), pp. 183-222. Donohoe, B.S., Mogelsvang, S., and Staehelin, L.A. (2006). Electron tomography of ER, Golgi and related membrane systems. Methods 39, 154-162. Dorokhov, Y.L., Skurat, E.V., Frolova, O.Y., Gasanova, T.V., Ivanov, P.A., Ravin, N.V., Skryabin, K.G., Makinen, K.M., Klimyuk, V.I., Gleba, Y.Y., and Atabekov, J.G. (2006). Role of the leader sequence in tobacco pectin methylesterase secretion. FEBS Letters 580, 3329-3334. Driouich, A., and Staehelin, L.A. (1997). The plant Golgi apparatus: Structural organization and functional properties. In The Golgi Apparatus, E.G. Berger and J. Roth, eds (Basel (Switzerland): Birkhauser Verlag. Egelund, J., Petersen, B.L., Motawia, M.S., Damager, I., Faik, A., Olsen, C.E., Ishii, T., Clausen, H., Ulvskov, P., and Geshi, N. (2006). Arabidopsis thaliana RGXT1 and RGXT2 encode Golgi-localized (1,3)-alpha-D-xylosyltransferases involved in the synthesis of pectic rhamnogalacturonan-II. Plant Cell 18, 2593-2607. Esseling-Ozdoba, A., Vos, J.W., van Lammeren, A.A., and Emons, A.M. (2008a). Synthetic lipid (DOPG) vesicles accumulate in the cell plate region but do not fuse. Plant Physiol 147, 1699-1709. 148  Esseling-Ozdoba, A., Houtman, D., Van Lammeren, A.A.M., Eiser, E., and Emons, A.M.C. (2008b). Hydrodynamic flow in the cytoplasm of plant cells. J Microsc 231, 274-283. Feraru, E., and Friml, J. (2008). PIN Polar Targeting. Plant Physiol. 147, 1553-1559. Freshour, G., Clay, R.P., Fuller, M.S., Albersheim, P., Darvill, A.G., and Hahn, M.G. (1996). Developmental and Tissue-Specific Structural Alterations of the Cell-Wall Polysaccharides of Arabidopsis thaliana Roots. Plant Physiol 110, 1413-1429. Gälweiler, L., Guan, C., Müller, A., Wisman, E., Mendgen, K., Yephremov, A., and Palme, K. (1998). Regulation of Polar Auxin Transport by AtPIN1 in Arabidopsis Vascular Tissue. Science 282, 2226-2230. Garcia-Herdugo, G., Gonzalez-Reyes, J.A., Gracia-Navarro, F., and Navas, P. (1988). Growth kinetics of the golgi apparatus during the cell cycle in onion root meristems. Planta (Heidelberg) 175, 305-312. Geitmann, A., and Emons, A.M. (2000). The cytoskeleton in plant and fungal cell tip growth. J Microsc 198, 218-245. Geitmann, A., Snowman, B.N., Emons, A.M., and Franklin-Tong, V.E. (2000). Alterations in the actin cytoskeleton of pollen tubes are induced by the self-incompatibility reaction in Papaver rhoeas. Plant Cell 12, 1239-1251. Geldner, N., Anders, N., Wolters, H., Keicher, J., Kornberger, W., Muller, P., Delbarre, A., Ueda, T., Nakano, A., and Jurgens, G. (2003). The Arabidopsis GNOM ARF-GEF mediates endosomal recycling, auxin transport, and auxin-dependent plant growth. Cell 112, 219-230. Gonzalez-Reyes, J.A., Gracia-Navarro, F., Garcia-Herdugo, G., and Navas, P. (1988). Changes of dictyosome ultrastructure during the cell cycle in onion root meristematic cells a morphometric and stereological study. Protoplasma 146, 35-40. Griffing, L.R. (1991). Comparisons of Golgi structure and dynamics in plant and animal cells. Journal of Electron Microscopy Technique 17, 179-199. Guo, W., Sacher, M., Barrowman, J., Ferro-Novick, S., and Novick, P. (2000). Protein complexes in transport vesicle targeting. Trends Cell Biol 10, 251-255. Hala, M., Cole, R., Synek, L., Drdova, E., Pecenkova, T., Nordheim, A., Lamkemeyer, T., Madlung, J., Hochholdinger, F., Fowler, J.E., and Zarsky, V. (2008). An Exocyst Complex Functions in Plant Cell Growth in Arabidopsis and Tobacco. Plant Cell 20, 1330-1345. Harris, P.J., and Northcote, D.H. (1971). Polysaccharide formation in plant Golgi bodies. Biochim Biophys Acta 237, 56-64. Haughn, G.W., and Somerville, C. (1986). Sulfonylurea-resistant mutants of arabidopsisthaliana. Molecular & General Genetics 204, 430-434. 149  Haughn, G.W., and Chaudhury, A. (2005). Genetic analysis of seed coat development in Arabidopsis. Trends Plant Sci 10, 472-477. Hawes, C.R., Brandizzi, F., and Andreeva, A.V. (1999). Endomembranes and vesicle trafficking. Curr Opin Plant Biol 2, 454-461. Hepler, P.K., Vidali, L., and Cheung, A.Y. (2001). Polarized cell growth in higher plants. Annu Rev Cell Dev Biol 17, 159-187. Hillmer, S., Movafeghi, A., Robinson, D.G., and Hinz, G. (2001). Vacuolar storage proteins are sorted in the cis-cisternae of the pea cotyledon Golgi apparatus. J Cell Biol 152, 4150. Hirose, S., and Komamine, A. (1989). Changes in ultrastructure of Golgi apparatus during the cell cycle in a synchronous culture of Catharanthus roseus. New Phytol. 111, 599-605. Hoekema, A., Hirsch, P.R., Hooykaas, P.J.J., and Schilperoort, R.A. (1983). A binary plant vector strategy based on separation of vir- and T-region of the Agrobacterium tumefaciens Ti-plasmid. Nature 303, 179-180. Hohl, I., Robinson, D.G., Chrispeels, M.J., and Hinz, G. (1996). Transport of storage proteins to the vacuole is mediated by vesicles without a clathrin coat. J Cell Sci 109 ( Pt 10), 2539-2550. Houtman, D., Pagonabarraga, I., Lowe, C.P., Esseling-Ozdoba, A., Emons, A.M.C., and Eiser, E. (2007). Hydrodynamic flow caused by active transport along cytoskeletal elements. EPL (Europhysics Letters) 78, 18001. Hwang, I. (2008). Sorting and Anterograde Trafficking at the Golgi Apparatus. Plant Physiol. 148, 673-683. Jahn, R., Lang, T., and Sudhof, T.C. (2003). Membrane fusion. Cell 112, 519-533. Jamet, E., Albenne, C., Boudart, G., Irshad, M., Canut, H., and Pont-Lezica, R. (2008). Recent advances in plant cell wall proteomics. Proteomics 8, 893-908. Jarvik, J.W., and Telmer, C.A. (1998). Epitope Tagging. Annual Review of Genetics 32, 601618. Jarvis, M.C., and McCann, M.C. (2000). Macromolecular biophysics of the plant cell wall: Concepts and methodology. Plant Physiol. Biochem. 38, 1-13. Jervis, E.J., Haynes, C.A., and Kilburn, D.G. (1997). Surface diffusion of cellulases and their isolated binding domains on cellulose. J Biol Chem 272, 24016-24023. Johnson, K.L., Jones, B.J., Schultz, C.J., and Bacic, A. (2003). Non-enzymatic (glyco)proteins. In The Plant Cell Wall, J.K.C. Rose, ed (Oxford: Blackwell Publishing.  150  Jones, L., Seymour, G.B., and Knox, J.P. (1997). Localization of Pectic Galactan in Tomato Cell Walls Using a Monoclonal Antibody Specific to (1[->]4)-[beta]-D-Galactan. Plant Physiol. 113, 1405-1412. Jürgens, G. (2005). Cytokinesis in higher plants. Annu Rev Plant Biol 56, 281-299. Jürgens, G., and Geldner, N. (2002). Protein secretion in plants: from the trans-Golgi network to the outer space. Traffic 3, 605-613. Kachar, B., and Reese, T.S. (1988). The mechanism of cytoplasmic streaming in characean algal cells: sliding of endoplasmic reticulum along actin filaments. J Cell Biol 106, 15451552. Ketelaar, T., de Ruijter, N.C., and Emons, A.M. (2003). Unstable F-actin specifies the area and microtubule direction of cell expansion in Arabidopsis root hairs. Plant Cell 15, 285292. Kimura, S., Laosinchai, W., Itoh, T., Cui, X., Lainder, R., and Brown, R.M.J. (1999). Immunogold labelling of rosette terminal cellulose-synthesizing complexes in the vascular plant Vigna angularis. Plant Cell 11, 2075-2085. Kirsch, T., Paris, N., Butler, J.M., Beevers, L., and Rogers, J.C. (1994). Purification and initial characterization of a potential plant vacuolar targeting receptor. Proceedings of the National Academy of Sciences of the United States of America 91, 3403-3407. Kiss, J.Z., Giddings, T.H., Staehelin, L.A., and Sack, F.D. (1990). Comparison of the ultrastructure of conventionally fixed and high pressure frozen/ freeze substituted root tips of Nicotiana and Arabidopsis. Protoplasma 157, 64-74. Knox, J.P. (1997). The use of antibodies to study the architecture and developmental regulation of plant cell walls. Int Rev Cytol 171, 79-120. Knox, J.P. (2008). Revealing the structural and functional diversity of plant cell walls. Curr Opin Plant Biol 11, 308-313. Knox, J.P., Linstead, P.J., King, J., Cooper, C., and Roberts, K. (1990). Pectin esterification is spatially regulated both within cell walls and between developing tissues of root apices. Planta 181, 512-521. Knox, J.P., Linstead, P.J., Peart, J.M., Cooper, C., and Roberts, K. (1991). Developmentally-regulated epitopes of cell surface arabinogalactan-proteins and their relation to root tissue pattern formation. Plant J 1, 317-326. Kobayashi, M., Matoh, T., and Azuma, J. (1996). Two Chains of Rhamnogalacturonan II Are Cross-Linked by Borate-Diol Ester Bonds in Higher Plant Cell Walls. Plant Physiol 110, 1017-1020. Kost, B., and Chua, N.H. (2002). The plant cytoskeleton: vacuoles and cell walls make the difference. Cell 108, 9-12. 151  Kost, B., Mathur, J., and Chua, N.H. (1999). Cytoskeleton in plant development. Curr Opin Plant Biol 2, 462-470. Kremer, J.R., Mastronarde, D.N., and McIntosh, J.R. (1996). Computer visualization of three-dimensional image data using IMOD. J Struct Biol 116, 71-76. Kuzmin, P.I., Zimmerberg, J., Chizmadzhev, Y.A., and Cohen, F.S. (2001). A quantitative model for membrane fusion based on low-energy intermediates. Proc Natl Acad Sci U S A 98, 7235-7240. Lam, S.K., Siu, C.L., Hillmer, S., Jang, S., An, G., Robinson, D.G., and Jiang, L. (2007). Rice SCAMP1 defines clathrin-coated, trans-golgi-located tubular-vesicular structures as an early endosome in tobacco BY-2 cells. Plant Cell 19, 296-319. Langhans, M., Hawes, C., Hillmer, S., Hummel, E., and Robinson, D.G. (2007). Golgi Regeneration after Brefeldin A Treatment in BY-2 Cells Entails Stack Enlargement and Cisternal Growth followed by Division. Plant Physiol. 145, 527-538. Lau, J.M., McNeil, M., Darvill, A.G., and Albersheim, P. (1985). Structure of the backbone of rhamnogalacturonan I, a pectic polysaccharide in the primary cell walls of plants. Carbohydr Res 137, 111-125. Lauber, M.H., Waizenegger, I., Steinmann, T., Schwarz, H., Mayer, U., Hwang, I., Lukowitz, W., and Jurgens, G. (1997). The Arabidopsis KNOLLE Protein Is a Cytokinesis-specific Syntaxin. J. Cell Biol. 139, 1485-1493. Lefebvre, B., Batoko, H., Duby, G., and Boutry, M. (2004). Targeting of a Nicotiana plumbaginifolia H+-ATPase to the Plasma Membrane Is Not by Default and Requires Cytosolic Structural Determinants. Plant Cell 16, 1772-1789. Lipka, V., Kwon, C., and Panstruga, R. (2007). SNARE-Ware: The role of SNARE-domain proteins in plant biology. Annu Rev Cell Dev Biol 23, 147-174. Lynch, M.A., and Staehelin, L.A. (1992). Domain-specific and cell type-specific localization of two types of cell wall matrix polysaccharides in the clover root tip. J Cell Biol 118, 467479. Macquet, A., Ralet, M.-C., Kronenberger, J., Marion-Poll, A., and North, H.M. (2007a). In situ, Chemical and Macromolecular Study of the Composition of Arabidopsis thaliana Seed Coat Mucilage. Plant Cell Physiol. 48, 984-999. Macquet, A., Ralet, M.C., Loudet, O., Kronenberger, J., Mouille, G., Marion-Poll, A., and North, H.M. (2007b). A naturally occurring mutation in an Arabidopsis accession affects a beta-D-galactosidase that increases the hydrophilic potential of rhamnogalacturonan I in seed mucilage. Plant Cell 19, 3990-4006. Martin, C., Bhatt, K., and Baumann, K. (2001). Shaping in plant cells. Curr Opin Plant Biol 4, 540-549.  152  Mascarenhas, J.P. (1993). Molecular Mechanisms of Pollen Tube Growth and Differentiation. Plant Cell 5, 1303-1314. Mastronarde, D.N. (1997). Dual-axis tomography: an approach with alignment methods that preserve resolution. J Struct Biol 120, 343-352. McCartney, L., Marcus, S.E., and Knox, J.P. (2005). Monoclonal antibodies to plant cell wall xylans and arabinoxylans. J Histochem Cytochem 53, 543-546. McFarlane, H.E., Young, R.E., Wasteneys, G.O., and Samuels, A.L. (2008). Cortical microtubules mark the mucilage secretion domain of the plasma membrane in Arabidopsis seed coat cells. Planta 227, 1363–1375. Micheli, F. (2001). Pectin methylesterases: cell wall enzymes with important roles in plant physiology. Trends in Plant Science 6, 414-419. Miller, D.D., de Ruijter, N.C.A., Bisseling, T., and Emons, A.M.C. (1999). The role of actin in root hair morphogenesis: studies with lipochito-oligosaccharide as a growth stimulator and cytochalasin as an actin perturbing drug. Plant J 17, 141-154. Mohnen, D. (2008). Pectin structure and biosynthesis. Curr Opin Plant Biol 11, 266-277. Mollenhauer, H.H., and Morre, D.J. (1976). Cytochalasin B, but not colchicine, inhibits migration of secretory vesicles in root tip of maize. Protoplasma 87, 39-48. Mollenhauer, H.H., and Morre, D.J. (1991). Perspectives on Golgi apparatus form and function. J Electron Microsc Tech 17, 2-14. Mollenhauer, H.H., and Morre, D.J. (1994). Structure of Golgi apparatus. Protoplasma 180, 14-28. Mollenhauer, H.H., Morre, D.J., and Griffing, L.R. (1991). Post-Golgi apparatus structures and membrane removal in plants. Protoplasma 162, 55-60. Moore, P.J., and Staehelin, L.A. (1988). Immunogold localization of the cell-wall-matrix polysaccharides rhamnogalacturonan I and xyloglucan during cell expansion and cytokinesis in Trifolium pratense L.; implication for secretory pathways. Planta 174, 433445. Moore, P.J., Darvill, A.G., Albersheim, P., and Staehelin, L.A. (1986). Immunogold localization of xyloglucan and rhamnogalacturonan I in the cell walls of suspensioncultured sycamore cells. Plant Physiol 82, 787-794. Moore, P.J., Swords, K.M.M., Lynch, M.A., and Staehelin, L.A. (1991). Spatial organization of the assembly pathways of glycoproteins and complex polysaccharides in the Golgi apparatus of plants. J Cell Biol 112, 589-602. Mouille, G., Ralet, M.C., Cavelier, C., Eland, C., Effroy, D., Hematy, K., McCartney, L., Truong, H.N., Gaudon, V., Thibault, J.-F., Marchant, A., and Hofte, H. (2007). 153  Homogalacturonan synthesis in Arabidopsis thaliana requires a Golgi-localized protein with a putative methyltransferase domain. The Plant Journal 50, 605-614. Munson, M., and Novick, P. (2006). The exocyst defrocked, a framework of rods revealed. Nat Struct Mol Biol 13, 577-581. Nebenführ, A., Frohlick, J.A., and Staehelin, L.A. (2000). Redistribution of Golgi stacks and other organelles during mitosis and cytokinesis in plant cells. Plant Physiol 124, 135-151. Nebenführ, A., Gallagher, L.A., Dunahay, T.G., Frohlick, J.A., Mazurkiewicz, A.M., Meehl, J.B., and Staehelin, L.A. (1999). Stop-and-go movements of plant Golgi stacks are mediated by the acto-myosin system. Plant Physiol 121, 1127-1142. Neuhaus, J.M., and Rogers, J.C. (1998). Sorting of proteins to vacuoles in plant cells. Plant Mol Biol 38, 127-144. Nielsen, E., Cheung, A.Y., and Ueda, T. (2008). The Regulatory RAB and ARF GTPases for Vesicular Trafficking. Plant Physiol. 147, 1516-1526. Northcote, D.H., and Pickett-Heaps, J.D. (1966). A function of the Golgi apparatus in polysaccharide synthesis and transport in the root-cap cells of wheat. Biochem J 98, 159167. O'Neill, M.A., and York, W.S. (2003). The composition and structure of plant primary cell walls. In The Plant Cell Wall, J.K.C. Rose, ed (Oxford: Blackwell Publishing), pp. 1-54. Oda, Y., Mimura, T., and Hasezawa, S. (2005). Regulation of secondary cell wall development by cortical microtubules during tracheary element differentiation in Arabidopsis cell suspensions. Plant Physiol 137, 1027-1036. Orfila, C., Oxenboll Sorenson, S., Harholt, J., Geshi, N., Crombie, H., Truong, H.N., Reid, J.S.G., Knox, J.P., and Scheller, H.V. (2005). QUASIMODO1 is expressed in vascular tissue of Arabidopsis thaliana inflorescence stems, and affects homogalacturonan and xylan biosynthesis. Planta 222, 613-622. Paredez, A.R., Somerville, C.R., and Ehrhardt, D.W. (2006). Visualization of cellulose synthase demonstrates functional association with microtubules. Science 312, 1491-1495. Park, P., Ohno, T., Kato-Kikuchi, H., and Miki, H. (1987). Alkaline bismuth stain as a tracer for Golgi vesicles of plant cells. Stain Technol 62, 253-256. Pelloux, J., Rusterucci, C., and Mellerowicz, E.J. (2007). New insights into pectin methylesterase structure and function. Trends in Plant Science 12, 267-277. Pennell, R.I., Knox, J.P., Scofield, G.N., Selvendran, R.R., and Roberts, K. (1989). A family of abundant plasma membrane-associated glycoproteins related to the arabinogalactan proteins is unique to flowering plants. J Cell Biol 108, 1967-1977.  154  Pennell, R.I., Janniche, L., Kjellbom, P., Scofield, G.N., Peart, J.M., and Roberts, K. (1991). Developmental Regulation of a Plasma Membrane Arabinogalactan Protein Epitope in Oilseed Rape Flowers. Plant Cell 3, 1317-1326. Perez, S., Mazeau, K., and Herve du Penhoat, C. (2000). The three-dimensional structures of the pectic polysaccharides. Plant Physiol. Biochem. 38, 37-55. Perez, S., Rodriguez-Carvajal, M.A., and Doco, T. (2003). A complex plant cell wall polysaccharide: rhamnogalacturonan II. A structure in quest of a function. Biochimie 85, 109-121. Persson, S., Paredez, A., Carroll, A., Palsdottir, H., Doblin, M., Poindexter, P., Khitrov, N., Auer, M., and Somerville, C.R. (2007a). Genetic evidence for three unique components in primary cell-wall cellulose synthase complexes in Arabidopsis. Proc Natl Acad Sci U S A 104, 15566-15571. Persson, S., Caffall, K.H., Freshour, G., Hilley, M.T., Bauer, S., Poindexter, P., Hahn, M.G., Mohnen, D., and Somerville, C. (2007b). The Arabidopsis irregular xylem8 mutant is deficient in glucuronoxylan and homogalacturonan, which are essential for secondary cell wall integrity. Plant Cell 19, 237-255. Pesacreta, T.C., and Lucas, W.J. (1985). Presence of a Partially-Coated Reticulum in Angiosperms. Protoplasma 125, 173-184. Popper, Z.A. (2008). Evolution and diversity of green plant cell walls. Curr Opin Plant Biol 11, 286-292. Preuss, M.L., Serna, J., Falbel, T.G., Bednarek, S.Y., and Nielsen, E. (2004). The Arabidopsis Rab GTPase RabA4b localizes to the tips of growing root hair cells. Plant Cell 16, 1589-1603. Puhlmann, J., Bucheli, E., Swain, M.J., Dunning, N., Albersheim, P., Darvill, A.G., and Hahn, M.G. (1994). Generation of monoclonal antibodies against plant cell-wall polysaccharides. Plant Physiol 104, 699-710. Reiter, W.D. (2002). Biosynthesis and properties of the plant cell wall. Curr Opin Plant Biol 5, 536-542. Rensing, K.H., Samuels, A.L., and Savidge, R.A. (2002). Ultrastructure of vascular cambial cell cytokinesis in pine seedlings preserved by cryofixation and substitution. Protoplasma 220, 39-49. Ridley, B.L., O'Neill, M.A., and Mohnen, D. (2001). Pectins: structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry 57, 929-967. Ripper, D., Schwarz, H., and Stierhof, Y.D. (2008). Cryo-section immunolabeling of difficult to preserve specimens: advantages of cryofixation, freeze-substitution and rehydration. Biol Cell 100, 109-123. Roberts, K. (2001). How the cell wall acquired a cellular context. Plant Physiol 125, 127-130. 155  Robinson, D.G., and Kristen, U. (1982). Membrane flow via the Golgi apparatus of higher plant cells. International Review of Cytology 77, 89-127. Robinson, D.G., and Hinz, G. (2005). Protein Sorting to the Storage Vacuoles of Plants: A Critical Appraisal. Traffic 6, 615-625. Robinson, D.G., Baumer, M., Hinz, G., and Hohl, I. (1997). Ultrastructure of the pea cotyledon Golgi apparatus: origin of dense vesicles and the action of Brefeldin A. Protoplasma 200, 198-209. Roelofsen, P.A. (1965). Ultrastructure of the wall and its relation to the direction of the growth. Adv Bot Res 2, 69-149. Rojo, E., and Denecke, J. (2008). What Is Moving in the Secretory Pathway of Plants? Plant Physiol. 147, 1493-1503. Rose, J.K.C., Catala, C., Gonzalez-Caranza, Z.H., and Roberts, J.A. (2003). Cell wall dissasembly. In The Plant Cell Wall, J.K.C. Rose, ed (Oxford: Blackwell Publishing. Roudier, F., Fernandez, A.G., Fujita, M., Himmelspach, R., Borner, G.H.H., Schindelman, G., Song, S., Baskin, T.I., Dupree, P., Wasteneys, G.O., and Benfey, P.N. (2005). COBRA, an Arabidopsis Extracellular Glycosylphosphatidylinositol-Anchored Protein, Specifically Controls Highly Anisotropic Expansion through Its Involvement in Cellulose Microfibril Orientation. Plant Cell, 1749-1763. Saint-Jore, C.M., Evins, J., Batoko, H., Brandizzi, F., Moore, I., and Hawes, C. (2002). Redistribution of membrane proteins between the Golgi apparatus and endoplasmic reticulum in plants is reversible and not dependent on cytoskeletal networks. Plant J 29, 661-678. Sakai, T., Honing, H., Nishioka, M., Uehara, Y., Takahashi, M., Fujisawa, N., Saji, K., Seki, M., Shinozaki, K., Jones, M.A., Smirnoff, N., Okada, K., and Wasteneys, G.O. (2008). Armadillo repeat-containing kinesins and a NIMA-related kinase are required for epidermal-cell morphogenesis in Arabidopsis. Plant J 53, 157-171. Samuels, A.L., Giddings, T.H., Jr., and Staehelin, L.A. (1995). Cytokinesis in tobacco BY-2 and root tip cells: a new model of cell plate formation in higher plants. J. Cell Biol. 130, 1345-1357. Samuels, A.L., Rensing, K.H., Douglas, C.J., Mansfield, S.D., Dharmawardhana, D.P., and Ellis, B.E. (2002). Cellular machinery of wood production: differentiation of secondary xylem in Pinus contorta var. latifolia. Planta 216, 72-82. Schnepf, E. (1986). Cellular polarity. Annu Rev Plant Physiol Plant Mol Biol 37, 23-47. Segui-Simarro, J.M., and Staehelin, L.A. (2006). Cell cycle-dependent changes in Golgi stacks, vacuoles, clathrin-coated vesicles and multivesicular bodies in meristematic cells of Arabidopsis thaliana: A quantitative and spatial analysis. Planta 223, 223-236.  156  Segui-Simarro, J.M., Austin, J.R., 2nd, White, E.A., and Staehelin, L.A. (2004). Electron tomographic analysis of somatic cell plate formation in meristematic cells of Arabidopsis preserved by high-pressure freezing. Plant Cell 16, 836-856. Sherrier, D.J., and Vandenbosch, K.A. (1994). Secretion of cell wall polysaccharides in Vicia root hairs. Plant J 5, 185-195. Shimmen, T., and Yokota, E. (2004). Cytoplasmic streaming in plants. Curr Opin Cell Biol 16, 68-72. Smith, L.G. (2003). Cytoskeletal control of plant cell shape: getting the fine points. Curr Opin Plant Biol 6, 63-73. Smith, L.G., and Oppenheimer, D.G. (2005). Spatial control of cell expansion by the plant cytoskeleton. Annu Rev Cell Dev Biol 21, 271-295. Somerville, C., Bauer, S., Brininstool, G., Facette, M., Hamann, T., Milne, J., Osborne, E., Paredez, A., Persson, S., Raab, T., Vorwerk, S., and Youngs, H. (2004). Toward a systems approach to understanding plant cell walls. Science 306, 2206-2211. Spurr, A.R. (1969). A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26, 31-43. Staehelin, L.A., and Chapman, R.L. (1987). Secretion and membrane recycling in plant cells: novel intermediary strctures visualized in ultrarapidly frozen sycamore and carrot suspension-culture cells. Planta 171, 43-57. Staehelin, L.A., and Moore, I. (1995). The plant Golgi apparatus: Structure, functional organization and trafficking mechanisms. Annu Rev Plant Physiol Plant Mol Biol 46, 261-288. Staehelin, L.A., and Kang, B.H. (2008). Nanoscale architecture of endoplasmic reticulum export sites and of Golgi membranes as determined by electron tomography. Plant Physiol 147, 1454-1468. Staehelin, L.A., Giddings, T.H., Kiss, J.Z., and Sack, F.D. (1990). Macromolecular differentiation of Golgi stacks in root tips of Arabidopsis and Nicotiana seedlings as visualized in high pressure frozen and freeze-substituted samples. Protoplasma 157, 7591. Staehelin, L.A., Giddings, T.H., S., L., Lynch, M.A., Moore, P.J., and Swords, K.M.M. (1991). Organization of the secretory pathway of cell wall glycoproteins and complex polysaccharides in plant cells. In Endocytosis, Exocytosis and Vesicle Traffic, C.R. Hawes, J.O.D. Coleman, and D.E. Evans, eds (Cambridge: Cambridge University Press), pp. 183-198. Stafstrom, J.P., and Staehelin, L.A. (1988). Antibody localization of extensin in cell walls of carrot storage root. Planta 174, 321-332.  157  Steinmann, T., Geldner, N., Grebe, M., Mangold, S., Jackson, C.L., Paris, S., Galweiler, L., Palme, K., and Jurgens, G. (1999). Coordinated polar localization of auxin efflux carrier PIN1 by GNOM ARF GEF. Science 286, 316-318. Sterling, J.D., Atmodjo, M.A., Inwood, S.E., Kumar Kolli, V.S., Quigley, H.F., Hahn, M.G., and Mohnen, D. (2006). Functional identification of an Arabidopsis pectin biosynthetic homogalacturonan galacturonosyltransferase. Proceedings of the National Academy of Sciences of the United States of America 103, 5236-5241. Surpin, M., and Raikhel, N. (2004). Traffic jams affect plant development and signal transduction. Nat Rev Mol Cell Biol 5, 100-109. Tokuyasu, K.T. (1997). immunocytochemistry on ultrathin cryosections. In Cells: A laboratory manual, D.L. Spector, R.D. Goldman, and L.A. Leonwand, eds (Cold Spring Harbour, NY: Cold Spring Harbour Laboratory Press), pp. 131.131-131.127. Tominaga, M., Morita, K., Sonobe, S., Yokuta, E., and Shimmen, T. (1997). Microtubules regulate the organization of actin filaments at the cortical region in root hair cells of Hydrocharis. Protoplasma 199, 83-92. Tse, Y.C., Mo, B., Hillmer, S., Zhao, M., Lo, S.W., Robinson, D.G., and Jiang, L. (2004). Identification of multivesicular bodies as prevacuolar compartments in Nicotiana tabacum BY-2 cells. Plant Cell 16, 672-693. Usadel, B., Kuschinsky, A.M., Rosso, M.G., Eckermann, N., and Pauly, M. (2004). RHM2 is involved in mucilage pectin synthesis and is required for the development of the seed coat in Arabidopsis. Plant Physiol 134, 286-295. Varner, J.E., and Taylor, R. (1989). New ways to look at the architecture of plant cell walls: Localization of polygalacturonate blocks in plant tissue. Plant Physiol 91, 31-33. Verhertbruggen, Y., Marcus, S.E., Haeger, A., Ordaz-Ortiz, J.J., and Knox, J.P. (2009). An extended set of monoclonal antibodies to pectic homogalaturonan. Carbohydr Res doi:10.1016/j.carres.2008.11.010. Vincken, J.P., Schols, H.A., Oomen, R.J., McCann, M.C., Ulvskov, P., Voragen, A.G., and Visser, R.G. (2003). If homogalacturonan were a side chain of rhamnogalacturonan I. Implications for cell wall architecture. Plant Physiol 132, 1781-1789. Vitale, A., and Hinz, G. (2005). Sorting of proteins to storage vacuoles: how many mechanisms? Trends in Plant Science 10, 316-323. Wasteneys, G.O. (2000). The cytoskeleton and growth polarity. Curr Opin Plant Biol 3, 503511. Wasteneys, G.O., and Galway, M.E. (2003). Remodeling the cytoskeleton for growth and form: an overview with some new views. Annu Rev Plant Biol 54, 691-722. Wasteneys, G.O., and Ambrose, J.C. (2009). Spatial organization of plant cortical microtubules: close encounters of the 2D kind. Trends Cell Biol 19, 62-71. 158  Western, T.L., Skinner, D.J., and Haughn, G.W. (2000). Differentiation of mucilage secretory cells of the Arabidopsis seed coat. Plant Physiol 122, 345-356. Western, T.L., Young, D.S., Dean, G.H., Tan, W.L., Samuels, A.L., and Haughn, G.W. (2004). MUCILAGE-MODIFIED4 encodes a putative pectin biosynthetic enzyme developmentally regulated by APETALA2, TRANSPARENT TESTA GLABRA1, and GLABRA2 in the Arabidopsis seed coat. Plant Physiol 134, 296-306. Western, T.L., Burn, J., Tan, W.L., Skinner, D.J., Martin-McCaffrey, L., Moffatt, B.A., and Haughn, G.W. (2001). Isolation and characterization of mutants defective in seed coat mucilage secretory cell development in Arabidopsis. Plant Physiol 127, 998-1011. Whittington, A.T., Vugrek, O., Wei, K.J., Hasenbein, N.G., Sugimoto, K., Rashbrooke, M.C., and Wasteneys, G.O. (2001). MOR1 is essential for organizing cortical microtubules in plants. Nature 411, 610-613. Wightman, R., and Turner, S.R. (2008). The roles of the cytoskeleton during cellulose deposition at the secondary cell wall. Plant J 54, 794-805. Willats, W., Knox, J.P., and Mikkelsen, J.D. (2006). Pectin: new insights into an old polymer are starting to gel. Trends in Food Science and Technology 17, 97-104. Willats, W.G., McCartney, L., and Knox, J.P. (2001a). In-situ analysis of pectic polysaccharides in seed mucilage and at the root surface of Arabidopsis thaliana. Planta 213, 37-44. Willats, W.G., Gilmartin, P.M., Mikkelsen, J.D., and Knox, J.P. (1999). Cell wall antibodies without immunization: generation and use of de-esterified homogalacturonan blockspecific antibodies from a naive phage display library. Plant J 18, 57-65. Willats, W.G., McCartney, L., Mackie, W., and Knox, J.P. (2001b). Pectin: cell biology and prospects for functional analysis. Plant Mol Biol 47, 9-27. Willats, W.G., Orfila, C., Limberg, G., Buchholt, H.C., van Alebeek, G.J., Voragen, A.G., Marcus, S.E., Christensen, T.M., Mikkelsen, J.D., Murray, B.S., and Knox, J.P. (2001c). Modulation of the degree and pattern of methyl-esterification of pectic homogalacturonan in plant cell walls. Implications for pectin methyl esterase action, matrix properties, and cell adhesion. J Biol Chem 276, 19404-19413. Willats, W.G.T., Marcus, S.E., and Knox, J.P. (1998). Generation of a monoclonal antibody specific to (1-->5)-[alpha]--arabinan. Carbohydrate Research 308, 149-152. Windsor, J.B., Symonds, V.V., Mendenhall, J., and Lloyd, A.M. (2000). Arabidopsis seed coat development: morphological differentiation of the outer integument. Plant J 22, 483493. Winicur, Z.M., Zhang, G.F., and Staehelin, L.A. (1998). Auxin Deprivation Induces Sychronous Golgi Differentiation in Suspension-Cultures Tobacco BY-2 Cells. Plant Physiol 117, 501-513. 159  Wolf, S., Rausch, T., and Greiner, S. (2009). The N-terminal pro region mediates retention of unprocessed type-I PME in the Golgi apparatus. Plant J. Young, R.E., McFarlane, H.E., Hahn, M.G., Western, T.L., Haughn, G.W., and Samuels, A.L. (2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized secretion of pectin-rich mucilage. Plant Cell 20, 1623-1638. Zhang, G.F., and Staehelin, L.A. (1992). Functional compartementation of the Golgi apparatus of plant cells. Plant Physiology 99, 1070-1083.  160  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0067318/manifest

Comment

Related Items