Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation into the molecular basis of secondary vascular tissue formation in poplar and Arabidopsis… Johnson, Lee Alan 2006

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

Item Metadata

Download

Media
831-ubc_2006_fall_johnson_lee.pdf [ 2.56MB ]
Metadata
JSON: 831-1.0092930.json
JSON-LD: 831-1.0092930-ld.json
RDF/XML (Pretty): 831-1.0092930-rdf.xml
RDF/JSON: 831-1.0092930-rdf.json
Turtle: 831-1.0092930-turtle.txt
N-Triples: 831-1.0092930-rdf-ntriples.txt
Original Record: 831-1.0092930-source.json
Full Text
831-1.0092930-fulltext.txt
Citation
831-1.0092930.ris

Full Text

An Investigation Into the Molecular Basis of Secondary Vascular Tissue Formation in Poplar and Arabidopsis With an Emphasis on the Role of Auxin and the Auxin Response Factor MONOPTEROS by Lee Alan Johnson B.Sc. (Hon), University of Alberta, 1998 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 September 28, 2006 © 2006  Abstract The differentiation of plant vascular tissue is regulated by plant hormones and transcription factors. One of the key plant hormones involved in this process is auxin. Auxin signals are mediated by auxin response factor transcription factors (ARFs). These transcription factors are involved in the perception of auxin signals and the subsequent activation or deactivation of suites of downstream genes. Based on its mutant phenotype, one of the most interesting members of this family is the ARF MONOPTEROS (MP). This thesis investigates the role played by MP in secondary vascular differentiation, as well as taking a look at other molecular aspects of secondary vascular differentiation, with a focus on the model plants Arabidopsis thaliana and poplar (Populus trichocarpa and hybrid poplar). A dexamethasone inducible RNAi silencing strategy was developed, and transgenic Arabidopsis lines produced. When silencing was induced in these lines from germination, a phenotype closely resembling the mp mutant was observed. When MP silencing was induced in bolting stems, early senescence, as well as a dramatic reduction in interfascicular fibre production was observed, and these stems were thinner and less rigid than empty vector controls. RNA from these stems was isolated and used in a global transcript profiling microarray experiment. This experiment showed that several auxin-related genes, as well as several transcription factors, were differentially regulated in response to MP silencing. Because Arabidopsis is not a typical woody plant, further investigation into the role played by MP in wood formation was done using the model tree poplar. A BLAST search of a poplar xylem EST database identified a single promising partial sequence. Based on this sequence information, a poplar MP homolog was isolated and named PopMP1. The full-length sequence of this gene demonstrated remarkable structural conservation when compared with that of Arabidopsis. Subsequent complete sequencing of the poplar genome revealed a second copy of the MP gene in poplar and named PopMP2. Expression profiling across a range of tissues suggests that subfunctionalization has occurred between the two copies. Overexpression transgenic lines for PoptrMP1 were developed. AtHB8 is known to be regulated by MP in Arabidopsis, and a poplar HB8 homolog was upregulated in the transgenic lines. However, no obvious physical phenotype in these lines was apparent. To investigate the transcriptome-wide changes associated with initiation of cambium formation in poplar stems, a global transcript profiling experiment was performed. Out of 15400 genes tested, 2320 met an arbitrary cutoff of >1.3 fold and p-value <0.05 and were labeled differentially expressed (DE). These included several transcription factors and showed remarkable similarity to analogous data from Arabidopsis. The conclusions drawn from this thesis support the hypothesis that MP plays roles in later development, and do not rule out the possibility that MP is directly involved in wood development. The data reported also offer a large number of candidate for further investigation into the genetic control of wood development.  ii  Table of Contents Abstract………………………………………………………………………………...…..ii Table of Contents….………………………………….…..………………………......iii List of Tables……………………………………………………………………………..vii List of Figures…………………………………………………………………………...viii Acknowledgments………………………………………………………………………xi CHAPTER 1 INTRODUCTIO N...................................................1 1.1 The plant vascular system……………………………………………………1 1.2 Auxin: biosynthesis, transport, and other biological roles……………8 1.3 Auxin and hormonal control of vascular development………………14 1.4 The auxin response pathway, the MONOPTEROS gene, and vascular development……………………………………………………………18 1.5 Arabidopsis and Poplar as models for studying plant vascular development……………………………………………………………………….28 1.6 Transcription Factors as tools for the genetic study of complex physiological processes………………………………………………………….31 1.7 Microarray expression profiling of fiber and xylem development in Arabidopsis and poplar…………………………………………………………33 1.8 The objectives of this thesis………………………………………………..34  CHAPTER 2 MATERIALS AND METHO DS……………………………….36 2.1 General nucleic acid methods…………………………………………….36 2.1.1 Plasmid DNA preparation and sequencing…………………36 2.1.2 Genomic DNA and total RNA isolation……………………..36 2.1.3 DNA Extraction from Agarose Gels………………………….36 2.1.4 Sequence Alignment…………………………………………….37 2.2 Quantitative RT-PCR………………………………………………………..37 iii  2.2.1 Quantitative Real-Time PCR…………………………………….37 2.2.2 Semi-quantitative RT-PCR………………………………………..39 2.3 Materials and Methods specific to Chapter 3………………………...39 2.3.1 Poplar MP complementation in Arabidopsis (performed by Ulises Sanchez, Berleth Lab, U Toronto)…………………………….39 2.3.2 Amplification of the Activation Domain for the Study of Sequence Diversity……………………………………………………….40 2.3.3 Testing for presence of PoptrMP1 neighbouring genes…40 2.3.4 MP-silenced Arabidopsis Microarray Experiment…..……41 2.4 Materials and Methods specific to Chapter 4…………………………44 2.4.1 Isolation of the PoptrMP1 clone………………………………44 2.4.2 Isolation of the PoptrMP1 genomic clone from a BAC library………………………………………………………………………45 2.4.3 Poplar DNA Southern blot analysis………………………….46 2.4.4 Phylogenetic Tree Construction………………………………..47 2.4.5 Agrobacterium-mediated transformation of Populus………48 2.5 Poplar Stem Microarray Experiment…………………………………….48 2.6 Construct Design…………………………………………………………….53 CHAPTER 3  ANALYSIS OF THE ARABIDOPSIS MONOPTEROS  (AtMP) GENE FUNCTION IN MATU RE PLANTS……………………….55 3.1 Introduction……………………………………………………………….....55 3.2 Results………………………………………………………………………….59 3.2.1 Conditional silencing of AtMP: the strategy…………………59 3.2.2 Silencing AtMP in developing seedlings results in a mp-like phenotype………………………………………………………………….62 3.2.3 Silencing of AtMP during flowering stem formation and the resulting phenotype………………………………………………………67  iv  3.2.4 Global transcript profiling of AtMP silenced stems……….73 3.2.5 AtMP  sequence  is  invariable  across  a  range  of  geographically separated Arabidopsis ecotypes…………80 3.3 Discussion……………………………………………………………………..83 CHAPTER  4  CLONING  AND  CHARACTERIZATION  OF  A  MONOPTEROS GENE FROM HYBRID POPLAR………………………..88 4.1 Introduction…………………………………………………………………..88 4.2 Results………………………………………………………………………….93 4.2.1 Cloning of the poplar gene popMP1………………………...93 4.2.2 Determining MP copy number in poplar…………………..102 4.2.3 Comparison of expression patterns for the two poplar MP genes……………………………………………………………………...105 4.2.4 Investigation of synteny in the genomic regions surrounding the MP genes……………………………………………………………108 4.2.5 Attempted Rescue of the Arabidopsis mp mutant phenotype with PoptrMP1………………………………………………………….116 4.2.6 Testing PoptrMP1 function by overexpressing the gene in poplar…………………………………………………………………….116 4.3 Discussion……………………………………………………………………119 CHAPTER 5 GLOBAL TRANSCRIP T PROFILING OVER THE COURSE OF POPLAR GREEN STEM DEVELOPMENT……………………………125 5.1 Introduction…………………………………………………………………125 5.2 Results and Discussion…………………………………………………….128 5.2.1 2320 microarray elements that vary over the course of stem development fit into 7 distinct clusters……………………………...128 5.2.2 Elements associated with “The Liginification Toolbox” are generally upregulated during stem development………………..134  v  5.2.3 Differential expression of elements associated with auxin signaling………………………………………………………………….140 CHAPTER 6 CONCLU SION AND A LOOK TO THE FUTURE…….152 References……………………………………………………..……………………….155  vi  List of Tables Table 3.1 Auxin-related genes whose expression decreased as a result of MP silencing……………………………………………………………………………………..77 Table 3.2 Annotated transcription factors whose expression pattern in MPsilenced stems varied inversely with expression over the course of stem development……………………………………………………………………………….79 Table  4.1  The gene "neighbourhood" surrounding MONOPTEROS on  Arabidopsis chromosome I…………………………………………………………….110 Table 5.1 DEs associated with the genes in “The Lignification Toolbox”…138 Table 5.2 Differentially Expressed auxin related genes………………………141 Table 5.3 Putative poplar orthologues of Arabidopsis TF fibre regulatory genes and microarray expression data…………………………………………….149 Table 5.4 Differentially Expressed TFs of Particular Interest…………………150  vii  List of Figures Figure 1.1 A simplified cladogram of land plant evolution based on the information referenced in the text……………………………………………………….3 Figure 1.2 A simplified view of the origins and structure of primary and secondary vasculature (redrawn from (Fosket, 1994)………………………………6 Figure 1.3 The structure of indole-3-acetic acid, the most common biologically occurring auxin………………………………………………………………………………9 Figure 1.4 Auxin Transport…………………………………………………………..12 Figure 1.5 The feedforward mechanism envisioned by the canalization hypothesis results in auxin drainage through specific strands……………………23 Figure 1.6 The general structures of the ARF and AuxIAA gene families…..24 Figure 1.7 A current model for auxin signaling and gene regulation..........25 Figure 1.8 Neighbour-joining tree of Arabidopsis ARF loci…………………...26 Figure 2.1 Construct design…………………………………………………………..54 Figure 3.1 A schematic representation of the Dex-inducible RNAi…………..61 Figure 3.2 Activation of AtMP RNAi expression in transgenic lines results in an mp-like phenotype…………………………………………………………………….64 Figure 3.3 Semi-quantitative RT-PCR amplifying the MP transcript in Dex treated plants………………………………………………………………………………66 Figure 3.4 Premature senescence of silenced leaves is apparent after 7 days of Dex induction…………………………………………………………………………..70 Figure 3.5 Senescence of silenced leaves after 14 days of Dex induction…71 Figure 3.6 Cross-sections of bolting stems from Dex-sprayed transgenic silenced and empty vector plants………………………………………………..…….72 Figure 3.7 Experimental design for the microarray hybridization portion of the experiment………………………………………………………..……………………75  viii  Figure 3.8 Sites of original isolation for 8 phenotypically distinct Arabidopsis ecotypes…………………………………………………………………………………….82 Figure 4.1 A phylogeny of model plant organisms……………………………..92 Figure 4.2 Comparison of the Poplar PoptrMP1 and Arabidopis AtMP genes………………………………………………………………………………………...94 Figure 4.3 The strategy employed to clone PopMP1……………………………95 Figure 4.4 A phylogenetic tree constructed using 18 members of the Arabidopsis ARF family, the proposed rice MP gene, and PopMP1…………..99 Figure 4.5 A Southern Blot suggests that the full genomic sequence is contained in at least 3 of 5 BAC clones…………………………………………….101 Figure 4.6 Comparison of Poptr1 and Poptr2. a) The full-length genes are depicted……………………………………………….…………………………………..103 Figure 4.7 Alignments of PoptrMP1, PoptrMP2, and AtMP predicted aa sequences. ………………………………………..……………………………………..106 Figure 4.8 Quantitative RT-PCR expression data for PoptrMP1 and PoptrMP2 across a tissue palette……………………………………………………………….….107 Figure  4.9  PCR-based localization of poplar genes homologous to  Arabidopsis genes in the genomic region surrounding PoptrMP1……………112 Figure 4.10  Sequence-based comparison of the regions surrounding the  three MP genes……………………………………………………………………….....115 Figure 4.11  QRT-PCR quantification of PoptrMP1 and PopHB8 expression  levels in 35S:MP1 overexpressing lines leaf tissue……………………………….119 Figure 5.1 Green poplar stem dissected by internode………………………..130 Figure 5.2 cDNA representing the transcriptome…………………………….132 Figure 5.3 Elements with a fold-change > 1.5 and a p-value > 0.05 formed 10 clusters…………………………………………………………………..133  ix  Figure 5.4 Representation of the monolignol biosynthetic pathway, and the “Lignification Toolbox” in poplar and Arabidopsis………………………………139  x  Acknowledgements The work leading to this thesis was by far the most challenging and rewarding undertaking of my life: I have spent nearly a quarter of my life completing this document. It has been tremendously exciting to have the chance to contribute to human knowledge. It has also been awe-inspiring and, above all, “a blast” to have worked with and around the following people: My supervisor: C arl J . Do ugla s. Carl is an incredibly effective and patient teacher. From the moment I met Carl in February 1998, he has constantly imparted me with scientific knowledge and advice. More than that, he has taught me a great deal about how to manage a successful team and communicate in a cautious and sincere fashion. I will gratefully carry the lessons learned from Carl through all of my life’s endeavors. I should also thank Allen Go od at the U of A for introducing and recommending me to Carl and Pliny Haye s at Red Deer College for initially inspiring my pursuit of graduate studies in molecular biology. My committee members: Drs. J örg B ohlm an n, Brian Ellis, and Ge org e Ha ugh n. These gentlemen have been a continual source of knowledge and support. Without their generous leadership, the completion of this work would have been impossible. All my fellow Douglas Lab members over the years 1999-2006, especially those with whom I had the pleasure to work and socialize to a great extent: Jür gen Ehltin g, Björ n an d Britta Ha mb erg er, Cl arice Souz a, D ae Kyun R o, Bah ram Soltani, Qi ng Wa ng, Ery ang Li, Jun e Kim, St ev e Petr ar, Tyler Fra ser, T am ara All en, Sha na Gutm an, D avid Joh nst on Mo nje, B enja min Wil son, Ali Sam aei an and To m Liu. I would also like to heartily thank the Haughn and Kunst teams from over the years, especially Ljerka Kun st, Mar k Pidko wic h, Ow en R o wla nd, Ravi Kum ar and Ta nya Hook er. Other professors, students and post-docs in the Department that have helped me (you know who you are) are too numerous to name, but the efforts of Lac ey Sa muel s, Har dy Hall, Ste ve R alp h, Hesth er Yu eh, N atalie M atth eus, He athe r C olem an and Sh aw n Man sfield were of special importance to my time at UBC. The team at PAPRICAN, especially Se an R og ers and Simo n Potter , were also invaluable in helping me obtain funding and complete this work. Th om as B erlet h and his group at U of T Botany were a great scientific resource and inspiration. The staff at the Department of Botany, from the gentlemen in stores, to the handymen and the EM Lab staff, to the invaluable secretarial staff, continually amazed me with their competence, helpful attitude, and admirable work ethic. My dear friends and family, around Canada and the globe, have offered the love, support, and good times that made my life during these years the envy of anyone that knew me. My parents, G ord on and Pat J oh nso n have been supportive of every decision I have made. Many people claim this type of support, but in my case, it is absolutely true: I cannot remember a single life choice to which they have not offered their unconditional backing. This support included always providing an environment that balanced learning and fun. It is thanks largely to their love, kindness, and advice that I have succeeded in this daunting task. My most heartfelt thanks go to all of the above and many others for helping me to achieve this lifelong goal.  xi  I dedicate this thesis to my grandmother, Eileen Johnson, as well as the loving memory of my late grandparents Eve Smith, Harold Johnson, and Ray Smith. To them I owe my DNA and my parents.  xii  Chapter 1. Introduction  1.1 The plant vascular system  Approximately 500 million years ago, small, filamentous algae began colonizing the dry land surrounding the bodies of water in which they lived (Gensel et al., 2001). The best available information from the fossil record indicates that these were the origins of the first land plants (Wellman and Gray, 2000; Wellman et al., 2003; Wellman, 2004).  The subsequent  “invasion” of plants and fungi onto the land gave rise to soil, and is ultimately thought to have “paved the way” for the subsequent invasion of animals (Gensel et al., 2001). Hence, the colonization of dry land by plants should be seen as the key biological event that has led to the appearance of terrestrial life.  Growing on land posed major challenges for multicellular plants that had evolved in aquatic habitats.  As depicted in Figure 1.1, a series of key  evolutionary “branching” events took place, eventually resulting in the range of land plant morphologies and habitats we see presently. First, these plants were no longer growing in a bath of water, dissolved carbon dioxide, and minerals. They now had to transport water and minerals from the soil, and move photoassimilates from the photosynthetic aerial tissues to those located below ground. Simple diffusion is an inefficient means of accomplishing these tasks, and so it is unsurprising that the development of simple vascular systems took place rather quickly. It is estimated that just 14 million years after the first appearance of land plants, vascular pteridophytes appeared and began  1  colonizing further from the wet, swampy habitats to which plants had thus far been limited (Renzaglia et al., 2000).  The competition between plants for  sunlight must have presented another challenge to the rapidly invading new vascular plants: gravity. Herbaceous plants are only able to grow to a certain height without support against gravity. Thus, the subsequent development of secondary vascular tissue, commonly referred to as “wood”, allowed the progymnosperm Archaeopteris, the first known tree, to support large body masses, and quickly achieve heights of over 40 metres (Meyer-Berthaud, 2000). Subsequently, the development of the seed, about 380 million years ago, allowed gymnosperms to push into still drier environments (as seed plants do not require pools of water for gamete fertilization) (Frohlich, 2003). Much later, just 140 million years ago, flowering plants appeared, thought to mark the most recent major development in plant evolution (Kenrick, 2000).  2  3  Some non-vascular land plant species, such as the bryophytes, continue to exist. However, they are restricted to very moist habitats populated by few competitors. Thus, for the vast majority of land plant species, a vascular system for the transport of water and minerals (transported through the xylem), and nutrients and signals (transported through the phloem) is essential.  The  vascular system has additional importance in woody plants, which produce secondary xylem tissue, generally referred to as wood, as a means of physical support for large biomasses. Wood is also of major importance to mankind, having  uses  in  paper  production,  construction,  and  manufacturing.  Additionally, wood acts as a "carbon sink"; storing large masses of atmospheric carbon dioxide as fixed organic carbon in the form of molecules such as cellulose, hemicellulose and lignin. Thus, gaining an understanding of how plant vascular systems develop has importance well beyond that of the interest of the academic biologist.  Plant vascular systems are comprised of xylem and phloem. The procambium, originating immediately below the shoot apex (see Figure 1.2), is the tissue that produces the primary xylem and phloem (Ye, 2002; Ye et al., 2002). The vascular tissues of leaves, called leaf veins, along with those of the primary stem, both arise from the procambium (Scarpella et al., 2002; Scarpella et al., 2004).  As depicted in Figure 1.2, the cells of the procambium arise from  meristem cells in differentiation zones at specific locations, and it is these cells that give rise to differentiated xylem and phloem cells. The procambial cells form strands, which comprise an extension of the vascular tissues into the growing tip and thus form connections between newly developed organs with the conductive systems of the rest of the plant. The cells of the procambium differ from the surrounding parenchyma in that they are elongated, a key initial  4  step in the differentiation into xylem or phloem cells (Ye, 2002). I will here use the definition of Larson (Larson, 1994) of a cambium as a meristem of undifferentiated tissue that divides to form new cells (based on the Greek origin of the word, from meristos, meaning divided). Hence, while the procambium is occasionally referred to as a meristem, by this definition, the partially differentiated and non-dividing procambium must be considered as nonmeristematic.  5  6  Secondary vascular tissue is defined as vascular tissue that arises from the vascular cambium (Groover, 2005). In woody dicots and gymnosperms, as well as many non-woody dicots such as tobacco, secondary vascular tissue develops in older stems from the vascular cambium. The vascular cambium, as shown in Figure 1.2, develops from the procambium in the stem, forming a continuous ring of only partially differentiated cells that separates the xylem from the phloem. This cambium, unlike the procambium, is a lateral meristem, capable of dividing periclinally and anticlinally, giving rise to new xylem and phloem cells. The xylem that is derived from these divisions is found on the inside of the dividing cambium (Chaffey et al., 2002; Nieminen et al., 2004). This xylem is known as secondary xylem, or, informally, wood. In developing into tracheary elements within the xylem, cells undergo terminal differentiation. In angiosperm trees such as poplar, during this phase cells deposit secondary cell walls composed primarily of cellulose, become heavily lignified and finally undergo programmed cell death and autolyse, such that their fate is as a sturdy tubular shaped cell wall, with the top and bottom missing and all cellular contents having been degraded. The deposition of lignin marks the stage of differentiation prior to cell death. (Roberts and McCann, 2000; Samuels et al., 2002; Ehlting et al., 2005). Lignin, a large heavily cross-linked polymer of undefined size composed of cinnamyl alcohol subunits derived from phenylpropanoid metabolism (Hahlbrock and Scheel (1989), is thus a major component of wood, and is the second most abundant biopolymer on earth after cellulose (Raes et al., 2003).  7  1.2 Auxin: biosynthesis, transport, and biological roles  Auxin is the general name given to several compounds that impart common physiological effects on plants. These effects include stimulation of elongation of coleoptiles or stems. All of these compounds also have a general chemical resemblance to indole-3-acetic acid, depicted in Figure 1.3, which is the major natural occurring auxin (Taiz and Zeiger, 2002). Because it acts as a chemical signal that can act as a messenger, and, via a concentration gradient, as positional cue (Friml, 2003), auxin is referred to as both a plant hormone and a morphogen.  8  9  Auxin was the first plant hormone to be discovered, and its existence was hypothesized over 100 years ago. The first evidence for the existence of any mobile signal in plants comes from experiments by Charles Darwin and his son Francis.  These experiments, performed on the coleoptiles of canary grass  (Phalaris canariensis) that demonstrated the involvement of a tip-produced signal in phototropism (Darwin and Darwin, 1881).  The same signal was  shown to be of a chemical nature by Fritz Went, who, in what are now considered famous experiments, decapitated oat coleoptiles, stopping growth (Went and Thimann, 1937). He then allowed the signal to diffuse from the cut tops into gelatin blocks. These blocks were then placed atop the coleoptile stumps, restoring growth.  This signal was dubbed auxin, from the Greek  auxein, meaning “to grow” (Pennazio, 2002).  While it is known that some other chemicals have auxin function, the primary known plant-produced auxin is indole-3-acetic acid (IAA), the structure of which is depicted in Figure 1.3. IAA is derived from tryptophan, through a variety of intermediates and pathways (Taiz and Zeiger, 2002; Cohen et al., 2003). Auxin is transported polarly, in aerial tissues from the shoot apex downward. This basipetal movement results in a gradient of concentration, highest at the apex (Muday, 2000). In the root, the highest region of concentration is in the root tips, and transport proceeds acropetally, via the stele. This directional auxin transport is mediated by specific auxin transporters (Kepinski and Leyser, 2005; Leyser, 2005b). The current model for auxin transport is known as the chemiosmotic model of polar auxin transport (Jones, 1992, 1998). According to this model, illustrated in Figure 1.4, IAA in the acid form enters a cell through a carrier. Mutations to the gene AUX1 display an auxin-insensitive  10  phenotype, and the similarity of this gene to certain amino acid permeases has led to the hypothesis that it is an example of an auxin influx carrier, and is a member of a family of auxin influx carriers (Parry et al., 2001). Inside the cell, the high pH causes the acid to deprotonate, forming the conjugate base. Then, following the concentration gradient, auxin exits the cell basally through another specific carrier (see Figure 1.4). This process requires no net energy other than that required for maintenance of the electron potential across the cell membrane (Jones, 1998). Several of the specific auxin efflux transporters have been identified based on mutations in the genes, which form a “pinformed” apex (hence, the genes are named PIN1, PIN2, etc), and the genes have been cloned and characterized (reviewed in Blakeslee et al., 2005). The gene products for some of these have been localized to the plasma membrane, at either pole of the cell (Jones, 1998; Friml, 2003).  11  12  Auxin regulates in a number of other plant processes such as cell expansion (Mockaitis and Estelle, 2004), embryo patterning (Jenik and Barton, 2005), primordium formation at the shoot apex (Kessler and Sinha, 2004), phototropism and gravitropism (Muday, 2001), vascular differentiation (Reinhardt, 2003; Fukuda, 2004), the regulation of apical dominance (Leyser, 2005a) and leaf senescence and abscission (Ellis et al., 2005). The latter processes (vascular differentiation, apical dominance, and leaf senescence and abscission) are of particular interest to this thesis; the role of auxin in vascular development will be discussed in detail in section 1.3 (below).  The phenomenon of apical dominance is the inhibition, by the apex, of growth from lateral buds. If the plant is decapitated, the apical dominance is lost with the source of auxin, and one or more lateral buds are released from growth inhibition and start to grow.  That an auxin source could maintain apical  dominance in decapitated plants was first demonstrated over 60 years ago by Thimann and Skoog (Taiz and Zeiger, 2002) in Vicia faba.  However,  numerous experiments have  levels  since  demonstrated  that  auxin  in  decapitated plants do not change measurably, while concentrations of ABA vary greatly (reviewed in Taiz and Zeiger, 2002). As well, mutants defective in auxin reponse have been shown to have a reduced degree of apical dominance, as might be expected if auxin is directly responsible for apical dominance. These seemingly contradictory facts leave many questions about auxin’s involvement in maintaining apical dominance.  The role played by auxin in the process of leaf senescence is also unclear. However, there are reports that exogenous application of auxin delays senescence (Osborne, 1959; Kawa-Miszczak et al., 1992). This phenomenon  13  is exploited commercially, as small amounts of auxin are sprayed on decorative holly (Ilex aquifolium) to prevent leaf drop during shipment (Taiz and Zeiger, 2002). Until recently, no genetic evidence existed for a clear role of auxin response genes in the leaf senescence. This may be explained by the fact that most auxin biosynthesis, transport, and perception mutants have pleiotropic phenotypes, and directly linking an auxin-related phenotype to premature (or delayed) senescence would thus be difficult. Nonetheless, very recent evidence suggests that some auxin response transcription factors (ARFs) may play a role in the auxin-mediated transition to senescence and organ abscission (Ellis et al., 2005). This study demonstrated that Arabidopsis plants mutant for two ARFs, ARF1 and ARF2, demonstrated delayed leaf senescence.  1.3 Auxin and hormonal control of vascular development  In vascular plants, as described above, the vascular system is a network of continuous cell files, all necessarily interconnected in some fashion, just as the circulatory system of vertebrates is understood to be continuous (Steeves and Sussex, 1989). The establishment of patterning within this system is thought to involve molecular signals in a regulatory cascade. Plant hormones are known to be important regulators of development and patterning, and among the signals though to be involved in vascular patterning is the plant hormone auxin (Nelson and Dengler, 1997). Auxin has been shown to be the limiting and controlling factor in xylem regeneration around a wound (Jacobs, 1952), thus suggesting the ability of auxin to re-establish vascular connections. Strong evidence for the ability of auxin to induce vascular differentiation has come from much more recent work of a similar nature (Mattsson et al., 1999).  14  A key theoretical framework for the ability of auxin to induce vascular differentiation comes from the "canalization" hypothesis of Sachs (Sachs, 1981). This hypothesis suggests that auxin flow, initially driven by diffusion, results in locally higher auxin concentrations, which induce the polar auxin transport system, which then facilitates active auxin transport and higher local auxin concentrations. This feed-forward activation of the auxin transport system by auxin itself then leads to canalization of auxin transport along cell files corresponding to developing vascular strands.  Recent molecular studies of  auxin transporters have offered an explanation for this directional flow. The directional transport is explained, at least partially, by the specific localization of auxin transport proteins, both influx and efflux, to either the top or bottom of cells within these files, as discussed above (reviewed in Palme and Galweiler, 1999; Muday and DeLong, 2001; Kramer, 2004). This locally enhanced auxin flow is thought to be a key early signal in cuing these cells to form procambial initials, and ultimately lead to the differentiation of functional xylem strands (Sachs, 1986; Berleth et al., 2004).  Physiological studies support the hypothesis that auxin plays a general role in tissue and organ patterning, including vascular patterning. Exogenous application of the auxin indole-acetic acid (IAA) at both embryo and adult stages of monocots has been shown to affect the switch from radial to bilateral symmetry (Fischer et al., 1997). Further, recent work involving the exogenous application of the auxin-transport inhibitor naphthylphthalamic acid (NPA) has “revealed an enormous plasticity of the vascular pattern” (Mattsson et al., 1999). Another study, utilizing NPA as well as, 9-hydroxyfluorene-9-carboxylic acid, 2,3,5-triiodobenzoic acid 2-(p-chlorophynoxy)-2-methylpropionic acid (Sieburth, 1999) demonstrated a range of specific morphological effects of  15  auxin inhibition. In this study, organs formed when auxin transport is reduced showed increased numbers of vascular strands, and these strands showed improper cellular alignment. Leaves developing under conditions of auxin transport inhibition demonstrate a restriction of vein development to the leaf margin, suggesting a flow of auxin inward from the leaf margin controls vein development.  Additionally, auxin is capable of inducing tracheary element  differentiation from Zinnia mesophyll cells (Fukuda H, 1980).  However,  perhaps the most exciting and compelling recent evidence of auxin’s involvement in the process of patterning comes from genetic experiments, which have revealed the roles of a number of auxin-related genes (such as transporters and transcription factors) whose actions underlie auxin-induced embryo and vascular patterning. These experiments will be discussed in section 1.4.  There is also evidence of a role for auxin in secondary xylem differentiation from the vascular cambium. It has been well established that auxin depletion results in the premature differentiation of cambial cells to a parenchymatic state (Savidge, 1983), thus demonstrating an essential role for auxin in the maintenance of cambium in a meristematic state. Experiments such as those outlined in (Sundberg B, 2000) demonstrate that application of exogenous auxin to cambium results in stimulated xylem production. Earlier work (Uggla et al., 1996; Tuominen et al., 1997) had demonstrated a strong positive quantitative correlation in developing woody stems between the local concentration of auxin and the initiation of secondary vascular differentiation from the vascular cambium.  The model supported by these data has the  cambium defined by a region of relatively high auxin concentration, with the  16  concentration decreasing on either side of this region and defining, at least in part, the regions of secondary phloem and xylem differentiation.  One interesting line of experimentation taken to test the role of auxin in secondary vascular differentiation has been to assess the effect of increased auxin levels in aspen on xylem differentiation. This was accomplished by creating transgenics overexpressing the auxin biosynthesis genes from Agrobacterium tumefaciens (Tuominen et al., 1995).  These plants had an  altered auxin gradient across the vascular cambium. While these plants did have a somewhat higher overall level of auxin, it was concentrated in the root tips and leaves, and the cambial zone and secondary xylem appeared changed only in that the zone was slightly wider, though the cell number was constant. The only variation in the local auxin concentration in the cambial zone was that a slightly lower than wild-type concentration was observed in a slightly wider zone.  The fibres produced were of a larger radius.  These  observations correlate nicely with the model outlined above, where a wider zone of auxin would result in a wider region of “undifferentiated” cambial cells.  There is also evidence that supports the involvement of other hormones, notably gibberellins, cytokinins, brassinosteroids, and ethylene in the process of secondary vascular development.  Gibberellic acid (GA) application in  conjunction with auxin was first shown to stimulate xylem differentiation in 1966 (Digby and Wareing, 1966), and overexpressing GA synthesis in aspen (Eriksson et al., 2000) has been shown to alter secondary xylem development, notably resulting in longer fibres. One model, though probably overly simplistic, of the hormonal control of xylem and phloem development is  17  that auxin alone stimulates cambial initials to form xylem, while GA and auxin act in concert to stimulate phloem formation (Savidge, 1983).  While  cytokinins (along with auxin) are known to be involved in inducing the in vitro transdifferentiation of Zinnia mesophyll cells to tracheary elements (Demura et al., 2002), mass spectrometry experiments have revealed extremely low (nanomolar) concentrations of cytokinins in the Scots pine cambium (Nilsson et al., 1996). To date, little work has been done on studying the cytokinin-related genes within the differentiating xylem, and this might be a worthwhile area of future research.  Among the phytohormones, the roles of ethylene and  brassinosteroids in the process of secondary vascular development are the least clear.  Brassinosteroids are known to be involved in Zinnia mesophyll  transdifferentiation (Ohashi-Ito et al., 2002).  Additional experiments have  shown that brassinosteroid-responsive genes are involved in Arabidopsis xylem development (Ohashi-Ito and Fukuda, 2003). Evidence suggests that ethylene stimulates cambial division, and mass spectrometry has shown ethylene to be present at relevant concentrations in the stem (reviewed in Mellerowicz et al., 2001). It has also been shown that xylem development in wood-forming tissues is affected by exogenous ethylene application (Andersson-Gunneras et al., 2003).  1.4  The auxin response pathway, the MONOPTEROS gene, and vascular development  In addition to the many biochemical and physiological studies of auxin’s connection to vascular differentiation, current research has begun to elucidate the molecular basis of the process.  Initial mutational analyses of vascular  differentiation identified a number of genes affecting primary vascular  18  patterning (Nelson and Dengler, 1997). These genes were identified by loss of function mutations leading to visibly defective vascular patterning in largescale mutagenesis experiments. One such Arabidopsis gene, MONOPTEROS (MP), was originally identified as an embryo-patterning mutant (Berleth T, 1993; Hardtke and Berleth, 1998). Because this gene eventually turned out to have an auxin-related function, it is appropriate to focus here upon its mutant phenotype.  The MP gene contributes to apical-basal pattern formation in the Arabidopsis embryo.  While the mutant phenotype varies from "weak" to "strong"  depending on the allele, all mp mutant seedlings lack such basal structures as hypocotyl, radicle, and root meristem.  This basal tissue is replaced by a  truncated conical structure made of large, undifferentiated cells, referred to as the "basal peg".  Further, cotyledon position seems to be variable, and  cotyledon veins are reduced. The mutant’s name was derived from the Greek words for “one wing,” due to the appearance of a single cotyledon in some mutant individuals. The wild-type cotyledon vein system consists of two semicircles connected to a central strand, while in mp mutants the two semi-circles are not present. Based on this phenotype, the MP gene product was shown to be absolutely required for the basal pattern elements (hypocotyl, radicle, root meristem), and was thought to be required in apical patterning only for cotyledon vein development. (Berleth and Jurgens, 1993; Jurgens et al., 1994). Significantly, vascular differentiation in plants mutant for MP closely resembles that of plants in which auxin transport is restricted (Mattsson et al., 1999). In both mp mutants and auxin transport inhibited plants, all normal vascular cell types are present, but there is a defect in the improper alignment of cells within vascular strands, such that formation of a complete vascular  19  system is impaired. The combined data from biochemical studies and from the mp phenotype led to the initial hypothesis that MP may be involved in auxin signaling events known to control vascular differentiation (Hardtke and Berleth, 1998).  The isolation of MP by positional cloning has provided a predicted protein sequence for the gene product.  The predicted protein contains predicted  nuclear localization sequences, suggesting that it functions within the nucleus (Hardtke and Berleth, 1998). Also within the predicted amino acid sequence is a DNA binding domain homologous to a domain in the original Auxin Response Factor (ARF) gene, the maize ARF1, shown to bind auxin inducible promoters (Ulmasov et al., 1997). Subsequent work has done a great deal to elucidate this process, as detailed below.  The finding that MP is involved in mediating response to auxin signals and regulating expression of auxin-regulated genes provides a clear genetic link between auxin and vascular development to compliment the well-established physiological evidence.  None of the known mp alleles studied has a mutation  in the DNA binding domain, suggesting that such a potentially null mutation may result in a lethal phenotype. Expression studies on MP have shown that it is initially expressed in broad regions of the early embryo, implicating a role for MP and, by extension, auxin, in the earliest developmental processes (Hardtke and Berleth, 1998).  These studies also revealed that, during  development, MP expression is gradually confined to vascular tissues, and is eventually confined to vascular files. In summary, these data provide evidence that MP is a transcription factor, and that it is at least partially responsible for an auxin perception in vascular development.  20  All data collected about MP support the hypothesis that the MP gene product mediates the role of auxin in vascular patterning. The mutant phenotype is defective in vascular pattern from the early embryo stage.  This phenotype  bears important resemblance to that of plants grown under conditions of impaired auxin transport. These facts lend support to the idea that the MP transcription factor is critical to vascular differentiation, and that it is an intermediate in auxin-induced signaling during this process.  While auxin has long been established as being involved as a key signal and morphogen in many developmental processes, a detailed elucidation of the nature of auxin perception and signaling has only recently become possible, with the discovery of MP and many other auxin-related regulatory genes in Arabidopsis (reviewed in Dharmasiri and Estelle, 2004).  The present molecular model for auxin response involves two families of interacting  regulators,  as  well  as  a  ubiquitination  regulated  protein  degradation step. This process is depicted in Figure 1.7. The auxin response is regulated by two large families of regulators: the auxin/indole-3-acetic acid responsive proteins, encoded by a gene family known as the AUX/IAA genes, and the Auxin Response Factors (ARFs). The AUX/IAA proteins form dimers with the ARF transcription factors (ARFs), preventing ARFs from binding auxinresponse elements in the promoters of auxin-activated genes (Tiwari et al., 2004). Auxin stimulates the ubiquitination of the AUX/IAA proteins, resulting in their targeted degradation via the 26S proteosome, resulting in loss of negative ARF-AUX/IAA interactions allowing ARFs to assume a (usually) positive role in activating auxin target gene activation as described below  21  (Gray et al., 2001).  The AUX/IAA gene family in Arabidopsis contains at least 29 members, defined by 4 conserved domains, assigned Roman numerals I-IV ordered N to C terminus, and depicted in Figure 1.6. Domains III and IV are shared with ARF genes, and are thought to mediate protein-protein interactions between these gene products (Hagen and Guilfoyle, 2002). The family was initially named because many of its members are rapidly up-regulated by auxin. However, this is true only for part of the family, and it is hypothesized that the others may be involved in “late” auxin response.  AUX/IAA proteins have  been shown to dimerize with the ARF gene products, of which 23 exist in Arabidopsis (see Figure 1.8).  All but two (ARF3/ettin and ARF17) share  domains III and IV with the Aux/IAA proteins, and this is therefore considered a protein-protein interaction domain (Okushima et al., 2005).  ARFs, when  freed from interaction with a “partner” AUX/IAA, act to bind auxin-response elements (AuxREs), a conserved TGTCTC domain, in the promoters of early auxin-response genes, such as the GH3 and so-called “small auxin up RNA” (SAUR) genes. An instance where part of this pathway is well understood is that of MONOPTEROS/ARF5, which is specifically bound by AUX/IAA12 (also known by the mutant name bodenlos) (Hamann et al., 2002). Mutants for either of these genes have almost identical phenotypes. It has also been shown that another member of the ARF family, ARF7 (also known by the mutant name non-phototropic hypocotyl 4 or nph4), has a largely overlapping function with MP, suggesting some redundancy within ARF/AuxIAA regulation (Hardtke et al., 2004).  22  23  24  25  26  The involvement of ubiquitin-regulated protein degradation in the process of auxin perception was discovered with the isolation of the mutant tir1, which showed reduced auxin response, and turned out to be involved in ubuiquitinmediated protein degradation (Ruegger et al., 1998; Nemhauser and Chory, 2005). In this process, the protein destined for degradation is tagged with a polyubiquitin chain, which is subsequently recognized by the 26S proteosome. The TIR1 protein is an E3 ubiquitin ligase that catalyzes the formation of the polyubiquitin chain, and catalyzes the degradation of AUX/IAA regulatory proteins. Upon the induction of the auxin response, the AUX/IAA protein is degraded via ubiquitination, and its partner ARF is freed up to activate transcription of downstream genes (Dharmasiri and Estelle, 2004).  Very recently, it has been shown that TIR1 binds auxin directly, and that its interaction with AUX/IAA proteins is regulated by this interaction with auxin (Dharmasiri et al., 2005a; Leyser, 2005b).  This discovery marks a major  breakthrough in the understanding of auxin signaling. Soon after the publication of TIR1’s activity as an auxin receptor, several new auxin receptors were published (Dharmasiri et al., 2005b). The emerging model from these discoveries involves a family of at least four auxin receptors, named AFB (Auxin F-box Binding) proteins, each of which has similar function. In one key experiment, the genes encoding all four of these proteins were knocked out, and the resulting phenotype was very similar to that of mp mutants, suggesting that it may be these four receptors, combined with MP and NPH4, that mediate the auxin regulation of an entire suite of downstream genes (Dharmasiri et al., 2005b). This work has very broad implications, as it the first instance where a complete auxin signaling pathway, from the hormone to a downstream gene, appears to be worked out at the molecular level. 27  1.5 Arabidopsis and Poplar as models for studying plant vascular development  Arabidopis thaliana is the major model for most current research in plant biology (reviewed in Somerville and Koornneef, 2002). The main advantages of Arabidopsis as a genetic system include the small size, short generation time, large seed set, and ease of handling. The choice of Arabidopsis as a model has led to a variety of well-established techniques, such as those for Agrobacterium-mediated transformation and mutagenesis.  Currently, the  resources available to the Arabidopsis biologist are extensive, from the existence of T-DNA-tagged mutant collections, to a well-annotated genome sequence, completed in 2000 (Initiative, 2000). However, there are limitations to the number of biological questions that can be answered by the use of Arabidopsis as a model. This is due to the absence, or reduced prominence, of many important processes in the normal Arabidopsis life cycle. One such process is that of wood development, which, under laboratory conditions, Arabidopsis does not normally undergo.  However, Arabidopsis flowering stems do normally produce non-xylem interfascicular fibres with heavily lignified secondary cell walls, which act to provide mechanical support to the stem (Zhong et al., 2001). It has also been shown that woody growth does take place in Arabidopsis when grown under appropriate conditions. (Chaffey et al., 2002; Ko et al., 2004). Hence, some information about the process of auxin-mediated secondary cell wall and secondary vascular development can be ascertained from the study of Arabidopsis, and the ability of a herbaceous species such as Arabidopsis to  28  form secondary xylem suggests that primary and secondary vascular differentiation may be related processes and could involve the same or related developmental pathways, such as MP-mediated regulation of gene expression by auxin.  However, the process of secondary xylem formation and wood  formation may be more effectively studied in a woody perennial plant, where this process is central to the life history of the plant.  In selecting a woody angiosperm to study, the genus Populus (commonly: poplar, aspens and cottonwoods) is a choice model system.  Populus is of  economic importance in North American boreal forests and in temperate plantations for pulp and paper production (Tuskan, 2006).  Poplar is also  valued in Europe, where the wood is used in the manufacture of boxes and furniture, as well as South America and Australia.  Beyond its commercial importance, poplar is an amenable model system. It has a relatively small genome of about 500 Mb (Han et al., 1996) and rapidly accumulates biomass in the form of secondary xylem. Vegetative propagation from cuttings or root-borne sucker shoots allows for study of a single genome varied both temporally and spatially and simple transformation protocols are available for some genotypes (Bradshaw, 1996). Also important to the study of wood-formation processes is the ease with which poplar xylem can be isolated through the peeling of bark from the main stem in late spring. The differentiating xylem is left exposed on the debarked stem, easily isolated by scraping. Additionally, an Expressed Sequence Tag (EST) database for poplar xylem was developed and made partially publicly available in 1998 (Sterky et al., 1998), shortly before I began my thesis work, and a much more complete  29  resources having been more recently made available (Sterky et al., 2004) (Ralph, 2005).  When the research described in this thesis was commenced, the poplar EST database comprised of a total of 5692 different tags, with an average readlength of 401 nucleotides (Sterky et al., 1998).  The database was made  from two separate cDNA libraries. The first of these was a mixture of xylem, phloem, and cambium cells, and the second a sample of differentiating xylem cells.  Tags were prepared for over 3,700 separate transcripts.  This small  database allowed for rapid, efficient preliminary investigation of the poplar genome, such as are cross-genome searches (i.e. a known gene in Arabidopsis is searched for in poplar by sequence similarity using the common BLAST algorithm (Altschul et al., 1990)). Since this research was initiated, an explosion of poplar genomic research has resulted in the completed, currently under annotation, genome sequence for poplar (Tuskan et al., 2006). Additionally, huge leaps in the depth of EST libraries as well as the development of cDNA microarray resources have made research in poplar as a model woody plant even more attractive (Sterky et al., 2004); Ralph et al., 2006).  Despite the many resources now available to poplar molecular biologists, the development of mutant resources and reverse genetic analyses commonly used in Arabidopsis to determine gene function are difficult and time consuming, often to the point of impossibility in even model trees such as poplar. Thus, it has become pertinent that we attempt to maximize the usefulness of the existing Arabidopsis information in making predictions about the functions of genes controlling vascular development in other, more practically useful plant  30  species, such as trees. In this spirit, this thesis has focussed on the potential of using information from Arabidopsis to identify vascular regulatory genes in poplar. As part of the project, I identified a potential poplar ortholog to the MP gene, and established to what extent synteny exists between the Arabidopsis and poplar genomes in the region of orthologous MP genes in the two species. By cloning an auxin-related gene such as MP in poplar and comparing its relative positions to those in Arabidopsis, this project will investigate the feasibility of developing this as a general strategy for applying genetic and genomic information from Arabidopsis to the industrially important Populus.  1.6  Transcription  Factors  as  tools  for  the  genetic  study  of  complex physiological processes  Transcription factors (TFs) are proteins that regulate the expression of one or more “downstream” target genes. Through this action, a few specific TFs can be responsible for the regulation of, among other things, entire developmental processes (reviewed in Zhang, 2003).  This makes the study of TFs a very  direct and efficient means for asking broad questions about the regulation of such processes (plant secondary vascular development, for example). Complications, such as functional redundancy, pleiotropic phenotypes, and even lethality, present difficulties for use of the traditional “forward” genetic method of mutant isolation when studying TFs.  For the same reasons, the  established “reverse” genetic approach of constitutive transcript-level silencing can be of limited use.  For example, the apparently lethal nature of a  MONOPTEROS null mutant (implied by the partially-functional nature of all known mutant alleles) suggests that a “knockout” approach, such as  31  constitutive antisense or RNAi would yield a lethal phenotype and little information about its role in the later stages of development, such as secondary vascular differentiation.  RNAi (RNA interference) is a method that takes  advantage of many eukaryotes (including flowering plants) constitutive, apparently virus immunity-related, tendency to rapidly degrade any abnormal RNA molecules (most commonly, double-stranded RNAs).  In so doing, this  degradation applies to “related” RNA species, including single stranded transcripts having a high degree of identity to the abnormal RNAs. The result is genetically-engineerable transcript-level silencing of any gene in species that exhibits RNAi (reviewed in Szweykowska-Kulinska et al., 2003; Waterhouse and Helliwell, 2003; Baulcombe, 2005; Soosaar et al., 2005).  Thus, in addition to overexpression studies that can circumvent lethality problems, a method for conditionally removing TF function at later stages of development is desirable. Through the combination of a system for inducible transcription with a system for strong transcript-level gene silencing, such as RNAi, one should be able to investigate the role, if any, played by a transcription factor such as MP later in development. Since auxin acts in such a wide array of developmental processes, one might expect that an ARF such as MP plays some role in these processes, in addition to those already shown in early development. In the case of poplar, where such inducible techniques have not yet been developed, the work described has attempted to maximize the utility of the converse approach: that of overexpression. As reviewed in (Zhang, 2003), the overexpression of TFs has been used many times to avoid the complications mentioned above, and gain insight into the function of TF genes in many developmental processes.  32  1.7  Microarray  expression  profiling  of  fiber  and  xylem  development in Arabidopsi s and poplar  The sequencing of entire plant genomes, including that of poplar and Arabidopsis, has provided a huge resource of genes potentially useful in agricultural and forest species improvement.  It has also made possible  expression profiling experiments, where the status of the entire transcriptome can be assessed under given circumstances (in a mutant background, or under stress conditions, for example). The first reported poplar transcript profiling experiment (Hertzberg et al., 2001) assessed the expression profile of 2,995 ESTs in wood-forming tissues. This experiment confirmed that known lignin and cell wall related genes were upregulated, but also found some interesting upregulation for previously unstudied transcription factors.  A similar  experiment (Schrader et al., 2004) used a more complete poplar array to profile gene expression across the developing cambial region, and identified more potential transcriptional regulators of secondary vascular differentiation. (Ehlting et al., 2005) employed a microarray-based strategy to study the transcriptional events associated with interfascicular fibre development in bolting Arabidopsis stems, a process related to secondary vascular cell differentiation.  These experiments have demonstrated the potential for transcriptome profiling as a strategy for gene discovery, specifically in the field of plant vascular development. Brown et al., 2005, report an early example of an experiment that uses existing microarray data as a guide for reverse genetic experiments in investigating genes related to IRREGULAR XYLEM1 (IRX1) and IXR3, both genes whose phenotype is related to Arabidopsis vascular development.  33  Through this approach, these investigators were able to clone several “xylem” genes that previously had not been implicated in the process of vascular development. The results of that study confirm that transcriptome-level profiling is an extremely powerful tool for identifying previously unknown genes involved in developmental processes such as vascular development.  1.8 The objectives of this thesis  Differentiation of secondary vascular tissue in angiosperms may follow an MPmediated patterning process, similar to that involved in primary vascular differentiation. As discussed previously, auxin is involved in both processes, so conservation of the basic mechanisms of auxin-induced differentiation may be expected. Taking advantage of the excellent model plant, one method for the study of auxin’s role in secondary growth involves the direct use of the Arabidopsis system.  Because MP function is critical to very early  developmental processes (as discussed in 1.3), study of MP’s involvement in later, secondary processes is not possible using forward genetics. However, through the conditional, transcript-level silencing of MP in adult plants, it may be possible to determine what role MP plays in later auxin mediated processes.  Hence, as presented in Chapter 3, I have used RNAi gene  silencing, directed by an inducible promoter, to investigate the function of MP in interfascicular fibre development in the Arabidopsis bolting stem.  While the model herb Arabidopsis provides valuable information, the cloning of a MP ortholog in a woody plant species would offer further opportunity for the study of auxin and ARF involvement in secondary vascular development. No MP homologue has yet been isolated and characterized from a woody  34  angiosperm species that undergoes secondary differentiation, and no suggestion of a role of MP in secondary vascular differentiation has been published. One direct method for testing the role of MP in secondary xylem differentiation is to investigate the role of MP or MP-like genes in a woody plant.  Thus, Chapter 4 of this thesis discusses the cloning and partial  characterization of a MONOPTEROS homolog (and likely ortholog) in poplar.  At the commencement of my thesis work, the realm of research possibility for a molecular biologist interested in secondary vascular development was far more restricted than it is today.  The new possibilities are owing to the dramatic  expansion of available resources, most importantly the complete sequencing of the poplar genome, the first among woody species. With that development, researchers can today ask questions of a much broader nature than was previously possible. Thus, while the initial focus of this thesis was restricted to the involvement of auxin, represented through the important ARF gene MONOPTEROS, it has become possible at the end of my thesis work to ask questions of a large portion of the transcriptome of wood-forming tissue in a single experiment. As Demura et al. (2002) and Ehlting et al (2005) have demonstrated in identifying key genes involved in tracheary element and fibre differentiation, it is possible to use microarray technology to study molecular aspects of secondary vascular differentiation.  In this spirit, the research  described in Chapter 5 has taken advantage of the extremely powerful technique of global transcript profiling, using a cDNA microarray representing ~15,500 poplar genes (Ralph et al., 2006).  35  Chapter 2. Materials and methods  2.1 General nucleic acid methods  2.1.1 Plasmid DNA preparation and sequencing Plasmid DNA was prepared with the use of Qiagen spin Miniprep- and Midiprep- kits, following the manufacturer’s instructions (Qiagen).  DNA  sequencing was performed by the University of British Columbia Nucleic Acid and Protein Service unit, using BigDye 3.0 (Applied Biosystems) and a Prism Sequencer (Applied Biosystems).  2.1.2 Genomic DNA and total RNA isolation Total poplar genomic DNA extraction was carried out with the use of Nucleon PhytoPure beads (Amersham-Pharmacia). Young poplar (P. trichocarpa X P. deltoides, clone H11-11) leaf tissue tissue was frozen in liquid nitrogen, ground to a fine powder, and the beads used according to manufacturer’s instructions. Arabidopsis DNA was extracted using the same method, using Arabidopsis (ecotype Columbia, except when otherwise specified) young leaf tissue. Poplar RNA was isolated following the protocol of Kolosova et al., 2004 using approximately 1.5 g of frozen tissue per extraction. Arabidopsis RNA was isolated using Trizol reagent (Gibco-BRL), following manufacturer’s instructions.  2.1.3 DNA Extraction from Agarose Gels DNA bands were cut from 1% agarose gels and the DNA extracted therefrom using the QiaQuick Gel Extraction Kit (Qiagen) according to manufacturer’s instructions.  36  2.1.4 Sequence Alignment Multiple sequence alignment was done using the CLUSTALX package version 1.8 (Thompson, et al. 1997) and restriction mapping and sequence analysis was  done  using  the  SeqPup  software  (http://iubio.bio.indiana.edu/soft/molbio/seqpup/java/seqpup-doc.html).  2.2 Quantitative RT-PCR  2.2.1 Quantitative Real-Time PCR RNA samples were isolated, qualified, and quantified as described for RNA gel blot analysis. RNA samples (2 µg RNA/ 20 µl reaction), were used to generate first strand cDNA using Superscript II Reverse Transcriptase (Invitrogen Life Technology) following the manufacturer’s protocol.  For the real-time quantitative real-time RT-PCR described in Section 4.2.3, and the results of which are depicted in Figure 4.7, total RNA (15 µg) was first digested with 15U DNAse in 1x buffer (Invitrogen) for 15 min at room temperature.  The  reaction  was  stopped  with  EDTA  (2.5  mM  final  concentration) and heat-inactivation (65ºC, 10 min).  RNA was precipitated with 1 volume of isopropanol and a 1/10 volume of 3 M sodium acetate at - 80°C for at least 30 min, and subsequently pelleted at 14,000 rpm for 40 min at 4°C. The precipitate was washed with 70% ethanol, re-centrifuged, air dried and resuspended in RNAse free water to an approximate concentration of 0.5 µg/µl. Concentration was determined spectrophotometrically.  37  10 µg total RNA was used for reverse transcription with 0.27 µM T17VN primer, 0.15 mM dNTP's, 40 U RNAseOut, and 400 U SuperscriptII (Invitrogen) in 10 mM DTT and 1 x first strand buffer in a total volume of 40 µl. Prior to addition of enzymes the solution was heated to 65°C for 5 min and for primer annealing cooled to 42°C. Following an incubation at 42°C for 50 min, in activate the reaction by heating at 70°C for 15 min. Based on A260 concentrations determined for the DNAsel-treated total RNA samples, dilute cDNA to a concentration of 1.67ng/µl for use in real-time PCR.  For quantitative PCR reactions, 6 µl cDNA were incubated with 10 µl QuantiTect SYBR Green PCR mastermix (Qiagen) and 30 nmole of each a forward and a reverse primer in a total volume of 20 µl. After an initial denaturation at 95°C for 15 min, 40 cycles at 95°C for 15 sec, 55°C for 30 sec, and 68°C for 45 sec followed by a fluorescence reading were performed. After and final incubation at 68°C for 5 min, a melting curve was generated ranging from 90°C to 60°C. Threshold cycles were adjusted manually, and the resulting threshold cycles (CT) were subtracted from CT values obtained for a housekeeping control amplified in parallel on each plate thus generating normalized CT values.  The relative starting quantities of each gene were  determined by setting as a base value the gene with the highest CT value, and relative quantities were expressed as log2 values. The intron-spanning primers used were PopMP1RTF 5’ GCACAACCATTCTGAAATGTTC PopMP1RTR 5’ CACCAACGTGGCAGTATCTC CTCTGTCCTTGCACATCTGGGATG  PopMP2RTF PopMP2RT  5’ 5’  CCATCAGACGTTCCAGGACAT.  38  2.2.2 Semi-quantitative RT-PCR For the semi-quantitative RT-PCR described in Section 3.2.2. and depicted in Figure 3.4, gene-specific and intron-spanning primers were used in PCR reactions to amplify corresponding cDNA sequences as follows: general PCR conditions were 95°C for 2 min, followed by 22 cycles of (94°C for 30 sec, 60°C for 30 s), using the same PCR reagents as described in 2.1.4. The PCR products were separated on 1% ethidium bromide agarose gels, visualized, and photographed under UV transilluminator using AlphaImager 1220. The primers  used  for  the  RT-PCR  reactions  were:  983fwd  5’  –  CCTAGTAATCGATTACG and 1471rev 5’ -- TCTAGGTAGCATGATCGCAA.  2.3 Materials and Methods specific to Chapter 3  2.3.1  Poplar MP complementation in Arabidopsis (performed by  Ulises Sanchez, Berleth Lab, U Toronto). The plasmid was transformed in a mp G12 normal looking population of plants that came from heterozygous parents. Plants were transformed and collected individually to avoid seed contamination between mp true heterozygous and wild type homozygous plants. Twenty-three plants were harvested individually after transformation. Putative T1 seed was grown on plates with kanamycin to select resistant plats with a density of 200 plants/plate and 5 plates per line. The progeny was scored for the appearance of mp phenotypes on the selection plates. Plates without mp phenotype seedlings were not considered for the selection of transformants. The progeny from 16 plants segregating mp seedlings were tested for transformation efficiency.  39  2.3.2 Amplification of the Activation Domain for the Study of Sequence Diversity For the purposes of assessing potential sequence diversity in the activation domain among various Arabidopsis ecotypes, seeds of the ecotypes listed in 3.2.3 were obtained from Lehle Seeds (www.arabidopsis.com). These were grown as described as above for DNA and RNA isolation and 21-day old rosette leaves used for total genomic DNA extraction as described above. These DNA samples were diluted to 100mg/mL and 1µL was used as template for  PCR  reactions  using  the  following  primers:  fwd:  5’-  CAGCTCCATCCATCTTACTTTGC rev 3’-ATCAAGGACTGTGTTTGAGAGTGG. PCR conditions were as described above, and these products were sequenced as described above.  2.3.3 Testing for presence of PoptrMP1 neighbouring genes For the synteny experiment described in section 4.2.4, the following primers were  used:  (Poplar  homolog  GAAACAAGGGTATCATGACCAG At1g19835  Fwd  5’  Rev  to) F’  At1g19830:  Fwd  CTCAGTATCTGATCAGAATGG  GAGGAGTTCTCAATTACATTCAAG  GCACAACCATTCTGAAATGTTC  F’  At1g19840  Rev  Fwd  5’ 5’  CCATCAGACGTTCCAGCAGGACAT Rev 5’ CACCAACGTGGCAGTATCTC Atg19850  Fwd  5’  ATGGGTCCTGCTGAAGAG  CCTGCTCGCTGTGTCCTTG  At1g19860  Rev  Fwd  5’ 5’  CTGATATCTTCGAGTCTTACCAGTTG Rev 5’ GCCACCGGTGGTACAAAAC At1g19870  Fwd  5’  CTTTATCAGCAAGTAAGAGATCTCC  CTCTGTCCTTGCACATCTGGGCTGGATG GGCTCGTACCAAGCAAACTGC  Rev  At1g19890 5’  Rev Fwd  5’ 5’  GCACGTTCACCACGGATACG  40  At1g19900  Fwd  5’  CCGACCGCTTACCATTACAACC  Rev  5’  CGGAATGAACTCATAATTGAATTG.  2.3.4 MP-silenced Arabidopsi s Microarray Experiment  Growth conditions and Agrobacterium-mediated transformation of Arabidopsis Arabidopsis thaliana ecotype Columbia was germinated and grown on AT medium (Somerville and Ogren, 1982) for one week under constant light. These were then transferred to soil (Redi-Earth, WR Grace and Co, West Ajax, Ontario, CA or SunShine Mix, Sungrow Horticulture, Saba Beach, AB), and grown under 12h light/12h dark conditions for several weeks until abundant floral clusters had formed, at which point plants were switched to 16h light/8h dark. Plants were then transformed using the floral dip method (Clough and Bent, 1998) with 0.05% Silwet L-77 (Lehle Seeds) with the vector as described and depicted in Section 3.2.1 and Figure 3.1 of this thesis and Agrobacterium genotype GV3101 (Van Labereke et al, 1974) which had been transformed using heat-shock transformation. Selection for transformants was done on AT media containing 30µg/mL hygromycin.  The primary transformants were  selfed and brought to the T5 generation with antibiotic selection at each crossing to obtain homozygous lines.  Dexamethasone Induction of Silencing Plants from 35 lines were grown on AT media (to determine which lines responded most strongly to treatment and for Figures 3.2 and 3.3) containing 30µM dexamethasone. This resulted in the selection of two transgenic lines (1 and 19) for further analysis, as well as the confirmation of an empty vector control line.  41  Plants from each of the three lines were started on AT medium, transferred to soil after 14 days, and grown under 8h light conditions for 28 days before being switched to 18h days. After 7 days, floral buds began to emerge and spraying was commenced.  Plants were sprayed liberally with a 50µM  dexamethasone aqueous solution twice daily (at 8 and 16h intervals).  Isolation of Plant Material for Sectioning and Array After 14 days of spraying, 10cm synchronized stems from each line (two “silenced” and one empty vector, sprayed) were isolated with a razor, cauline leaves removed, and the material snap-frozen in liquid nitrogen.  Microscopy To generate Figure 3.7, stem cross-sections were prepared using a vibratome immersed in tapwater. The sections were 50µm. Fluorescent illumination was used to observe lignified portions of the stems on a Zeiss Axioplan light microscope.  RNA Isolation RNA was isolated following the Trizol (Gibco-BRL) manufacturer’s extraction protocol using approximately 1.5 g of frozen tissue per extraction. Total RNA was quantified and quality checked by spectrophotometer and agarose gel. Details of the RNA isolation and quality testing done can be found at http://treenomix0.forestry.ubc.ca/~arabidopsis/2_Array_Specs.html#mozTocI d823774.  42  Array hybridization and experimental design I used array-ready 70mer oligomer (or 'oligo') sets from Operon.  The  sequence of the 70mers was based upon Unigene models of 26,090 known or putative Arabidopsis genes (Douglas and Ehlting, 2005). Details of the 26 K array and its design is available at the Arabidopsis Microarray Project (AtMP) home page: (http://treenomix0.forestry.ubc.ca/ arabidopsis/1_AtMP_Home.html).  The pipeline used for quality control, normalization, and bioinformatic analysis of array results, using customized scripts for R and Bioconducter (The R Development Core Team, http://www. r-project.org) is described at the Arabidopsis Microarray Project (AtMP) Home Page: (http://treenomix0.forestry. ubc.ca/ arabidopsis/7_Pipeline_Tools.html). Optimized Cy3/Cy5 cDNA probe preparation, and microarray hybridization Protocols are also available at the AtMP home page.  Each of the two transgenic lines was compared directly with the control (sprayed, empty vector) plants. This was done with six replicates for each line, once in the Cy3 channel, once in the Cy5, for a total of 12 slides per silenced line. The tissue used was pooled at the earlier freezing step, but RNA isolation was repeated individually for each independent replicate. These data were then “pooled” and a cutoff of value less than 0.05 and fold change greater than 1.3 was used to screen “background” genes. Only elements that met this cutoff were considered as differentially expressed.  43  2.4 Materials and Methods specific to Chapter 4  2.4.1 Isolation of the PoptrMP1 clone A cDNA library from hybrid poplar H11-11 xylem mRNA was previously constructed in Lambda ZAPII (Subramaniam et al, 1993; Y. Tsutsumi, Shizouka University, Japan and C. Douglas, unpublished) according to manufacturer’s instructions (Stratagene).  100µL aliquots of this library were boiled for 1  minute and then 1µL used as template for a PCR reaction. The primers for this reaction were: forward (5’-ACCGCTAGTGATACAAGTACACATGGTGG-3’) and  reverse  (5’-GCAGCAGCAAGGACACCAATGTGCAT-3’).  The  PCR  conditions were: total volume 50µL, 20pmol each primer, 2.5 units Expand High Fidelity Taq Polymerase (Roche), 5µL template, 0.2mM dNTPs, and 2 mM Mg2+. The cycling was: 2 min at 94º, 29 cycles of 1 min at 60º, and 3 min at 72º. A 20µL aliquot of this PCR reaction was loaded on 1% agarose EtBR containing gel and a single band at the expected size of 800 bp was resolved. This band was extracted from the gel and eluted in 30µL water. 1 µL of this eluate was used as template for the Qiagen PCR Cloning Kit reaction, following manufacturer’s instructions (Qiagen). This clone was sequenced, and new primers were made to amplify in both 5’ and 3’ directions from the edges of this clone. A nested PCR procedure was used, using first the aforementioned ZAPII library as template, then a 2µL aliquot of the resulting PCR reaction. The same cycling and reaction conditions described above were used. The primers used were, for the nested reactions proceeding in the 5’ direction: forward  (5’-ACCGCTAGTGATACAAGTACACATGGTGG-3’)  GTTCTCAGTGCCACGCAGGGCAGC-3’) TTGTAATACGACTCACTATAGGGCG-3’)  and  and  reverse and  (5’(5’(5’-  GGTACCGGGCCCCCCCTGGA-3’). These reverse primers were designed to  44  anneal with the vector. In the 3’ direction, the primers used were: forward (in this  case,  the  forward  primers  annealing  with  CTCGAAATTAACCCTCACTAAAGGG-3’)  vector)  and  CTATGACCATGATTACGCCAAGCTC-3’)  and  GCAGCAGCAAGGACACCAATGTGCAT-3’) ATGTTGTTTGTTGACGGTTTGCATGCC-3’).  the  reverse and  (5’5’(5’(5’-  These reactions produced 1.4kb  and 2.4 kb bands, respectively. These bands were also cloned with the TOPO TA Cloning Kit (Invitrogen), following manufacturer’s instructions. Both of these clones were sequenced completely.  The resulting sequences were digitally  aligned, and new primers were designed to anneal to the extreme 5’ and 3’ of the resulting 2.8 kb open reading frame. ATGGGTCCTGCTGAAGAGAAAACC-3’) TCAAGCATGGATACCCTCTGTGATGG-3’).  These were: forward (5’and  reverse  (5’-  These primers were used to  amplify the complete 2.8 kb ORF, using 5µL of the aforementioned library as template, and the PCR conditions described above. The resulting 2.8 kb band was cloned using the TOPO TA Cloning Kit and sequenced.  2.4.2 Isolation of the PoptrMP1 genomic clone from a BAC library A genomic library for poplar clone H11 of Bacterial Artifical Chromosome (BAC) clones, containing an average insert size of 114 kb was created using pBELOBAC in E. coli strain DH10B by The Crop Biotechnology Center (Texas A & M University) and obtained from J. Giovannoni.  The library was then  spotted onto nitrocellulose. This library was probed separately with the first and last 150 bp of the poplar MP coding sequence, as well as the entire cDNA clone with the Amersham-Pharmacia Random-Prime labeling kit (Amersham) according to manufacturer’s instructions.  Eleven preliminary BACs were  isolated, and a PCR check was done to test for presence of the PopMP1 gene.  45  Subsequent Southern blots confirmed the presence of the full-length gene in three BACs.  These were numbered T0016N11, T0024O19, and T0067H1  (also referred to as BACs 3, 5 and 9, respectively, in this thesis).  2.4.3 Poplar DNA Southern blot analysis For the genomic Southern, 3 aliquots of 10µg (each) total genomic P. trichocarpa DNA was digested to completion with each of XbaI, XhoI, and BamHI. These digests were resolved on 0.8% agarose gel. This gel was poststained in ethidium bromide, and then denatured in 0.5M NaOH, 1.5M NaCl for 5 minutes, and finally renatured in 0.5M Tris pH 7.5, 1.5M NaCl for 5 minutes.  A blotting apparatus was assembled, and DNA transferred to  Hybond XL membranes (Amersham-Pharmacia) according to standard methods (Sambrook et al., 1989). The blot was rinsed briefly in 2X SSC and baked at 80º for 2h. Prehybridization was done for 3h at 65ºC in 1 M NaCl, 1% SDS, 10% dextran sulfate and 50 µg/mL denatured salmon sperm DNA. Radioactive probes were prepared using the Random Prime Labelling Kit (Gibco-BRL) according to manufacturer’s instructions.  These probes were  added to the prehybridization solution and hybridization proceeded overnight at 65ºC. Membranes were then washed twice for 30min at 65º in 2X SSC, 0.1% SDS. A final wash was performed at 65º for 30min in 0.2X SSC, 0.1% SDS.  Radioactive signals were detected with a Molecular Dynamics Storm  phosphoimager. Blots were stored, damp, at -20ºC between uses.  For the BAC DNA Southern, the exact same protocol was used, with the only difference being that 10µg aliquots of DNA from each of the 10 BACs were digested to completion with XbaI before running the gel.  46  The probes used were generated using the following primers: 1) “5’ 150 bp Probe”:  0fwd  5’-  ATGGGTCCTGCTGAAGAGAAAACC  ACCATGTCCAGTGGACAGGAACC  2)  “3’  150bp  150rev  Probe”:  TGATCCGATGCCAGTAGGCATT  3’  -  2.65fwd 2.8rev:  TCAAGCATGGATACCCTCTGTGATGG.  2.4.4 Phylogenetic Tree Construction An alignment of published YGI amino acid sequences was generated using the CLUSTALX package version 1.8 (Thompson, et al. 1997) with manual optimization using the SeqPup software. Based on this alignment, a maximum parsimony analysis was performed (Fitch, 1977) using the PAUP 3.1.1 program (Smithsonian Institution, Washington DC, USA). The most parsimonious tree was found using the heuristic search option with the ‘tree bisection reconnection’ (TBR) branch swapping algorithm (Swoffort and Olsen, 1990). For statistical analysis, 500 bootstrap replications (Felsenstein, 1985) were analysed. The following plant YGI sequences were used (GenBank accession numbers or locus identifier are given in brackets): AtARF9 (AT2G46530), AtARF18 (AT3G61830), ARF1 (AT1G59750), MP/AtARF5 (AT1G19850),  AtARF2  (AT5G62000),  AtARF6  (AT1G30330),  AtARF8  (AT5G37020), AtARF3 (AT2G33860), AtARF4 (AT5G60450), AtARF10 (AT2G28350), AtARF17 (At1g77850) AtARF30 (AT2G15770), AtARF22 (AT1G34390) AtARF75 (AT3G05480), AtARF32 (At2g46870) AtARF31 (At2g36080), AtARF14 (AT1G35540), and Oryza sativa putative MP (accession number AB071292). The tree is shown in Figure 4.4.  47  2.4.5 Agrobacterium-mediated transformation of Populus Agrobacterium  genotype  GV3101  was  generated  with  heat-shock  transformation containing pHannibal vector in which the intron had been replaced by the coding sequence of PoptrMP1 in the sense orientation (as described in Section 4.2.4 of this thesis). This was cultured overnight in woody plant  media  (WPM)  (Owen  and  Miller,  1992)  containing  100µM  acetosyringone. Leaf disks from 4-6 week old P. alba x P. grandidentata, 717 were cultured overnight in liquid WPM. (NAA),  6-benzylaminopurine  (BA),  0.1 M of 1-naphthaleneacetic acid and  thidiazuron  Agrobacterium culture was diluted to 0.1 OD600.  (TDZ).  The  The leaf disks were  cocultivated with shaking at room temperature in Agrobacterium (25 disks/10 mL culture, 30 min each). These disks were then placed in the dark on solid WPM  containing  0.1  M  of  1-naphthaleneacetic  acid  (NAA),  6-  benzylaminopurine (BA), and thidiazuron (TDZ) for 3 days, after which they were moved to fresh WPM media with the same hormones plus and 500 mg/L of carbenicillin, 250 mg/L of cefotaxime and 25 mg/L of kanamycin. Media was refreshed weekly.  After 9 weeks, shoots were transferred to WPM  containing only antibiotics and 0.1µM NAA for rooting. After 4 additional weeks, these plantlets were multiplied on the same rooting media. Once plants had established root systems, they wre transferred to soil (Redi Earth) under greenhouse conditions  2.5 Poplar Stem Microarray Experiment  Plant material Populus trichocarpa Torr. & Gray x P. deltoides Bartr, H11-11 genotype, was grown on the University of British Columbia South Campus farm. Cuttings of  48  30-100 cm were taken in February of 2003 from previous year shoots, placed in soil (35% peat, 15% perlite, 50% pasteurized mineral soil, 250 gm-3 OsmocoteTM 13-13-13 plus micronutrients) in two-gal. pots (Stuewe & sons Inc., Corvallis, USA), and watered daily. Trees were maintained in a greenhouse under constant summer conditions where a constant 16/8-hour photoperiod was provided by high-pressure sodium lamps, and were placed outside in pots in the summer of 2003. Trees of 60 to 70 cm in height were used for isolation of internode and leaf material in August 2003.  Sampling The top seven internodes were isolated from each of 12 ramets using a clean razor blade.  Leaves were removed from the internodes and these were  immediately snap-frozen in liquid nitrogen. Internodes of the same category from each tree were pooled together.  The internodes were numbered as  follows: the internode immediately up from the first fully expanded leaf: -1, the internode down from this leaf, +1. A survey of 25 trees revealved that in each case there were exactly 5 distinct internodes above the first fully expanded leaf. Subsequently, it was decided that only the top four internodes would be used for RNA isolation.  The topmost internode, -5, was then renamed  “Treatment 5” or T5, and the next T4, and so on down to T2.  Microscopy Stem cross-sections from the centre of 10 representative internodes from each stage (T5, T4, etc) were prepared using a vibratome immersed in tapwater. The sections were 100µm.  Fluorescent illumination was used to observe  lignified portions of the stems. A Zeiss Axioplan light microscope was used to magnify the images. All sections from the same internode stage were deemed  49  essentially  identical from  the  standpoint  of  lignification  and  cambial  development. A representative from each was photographed. In the case of the larger (T4, T3, T2) sections, it was necessary to take multiple digital photos and merge them using Photoshop (Adobe).  RNA isolation RNA was isolated following the protocol of Kolosova et al. (2004) using approximately 1.5 g of frozen tissue per extraction. Total RNA was quantified and quality checked by spectrophotometer and agarose gel. RNA was also evaluated for integrity and the presence of contaminants using reversetranscription with Superscript II reverse transcriptase (Invitrogen) with an oligo d(T18) primer and P  32  dGTP incorporation (Ralph et. al, 2006).  I used a cDNA microarray composed of 15,496 cDNA elements selected from 14 cDNA libraries representing leaves, buds, phloem, xylem, bark and root tissues (Ralph et al., 2006).  Functional annotation of array elements was  assigned according to the TAIR Arabidopsis protein set using BLASTX, as well as using BLASTX versus the set of ca. 45,000 protein models predicted from the draft version of the poplar genome sequence. Overall, 11,410 (73.6%) of 15,496 spotted cDNAs have similarity to the TAIR Arabidopsis protein set by BLASTX (E <10-05) (Ralph et al., 2006). Additional annotations of array elements of special interest were carried out by manual searches of corresponding poplar EST contigs and or gene models corresponding to genes in the TAIR database.  All microarray experiments were designed to comply with MIAME guidelines (Brazma et al., 2001). All scanned microarray TIF images, an ImaGene grid,  50  the gene identification file and Imagene quantified data files are available at http://douglas.bcgsc.bc.ca.  Total RNA from each internode segment was directly compared in the microarray to each of the other samples in the experiment. Consequently, all four internodes were compared to each other, in an each experiment containing 24 arrays. Each sample appeared on 12 arrays, six times in the Cy3 channel and six times in the Cy5 channel. Each sample was hybridized with each other sample on four arrays, twice in each channel. Therefore, for each sample A in Cy5 and sample B in Cy3 there were two arrays.  Briefly, hybridizations were performed using the Genisphere Array350 kit following manufacturer’s instructions and with the modifications described by Ralph et al. (2006), using 30 µg total RNA were used per sample.  After  hybridization and washes, the microarray slides were scanned using a PE Scanner, and fluorescent images of hybridized arrays were acquired by using ScanArray Express (Perkin Elmer, Foster City, USA).  The Cy3 and Cy5  cyanine fluors were excited at 543 nm and 633 nm, respectively. All scans were performed at the same laser power (90%), but with the photomultiplier tube settings for the two channels adjusted such that the ratio of the mean signal intensities was ~1, and the percentage of saturated array elements was < 0.5%, while minimizing background fluorescence. Fluorescent intensity data for median spot foreground and background signal were generated using ImaGene 5.5 software (Biodiscovery, El Segundo, USA).  51  Microarray experimental design, statistical analysis, clustering Before normalization, the lowest 10% of median foreground intensities was subtracted from the median foreground intensities to correct for background intensity for each channel of each array. After quantification of the signal intensities, data were normalized to compensate for nonlinearity of intensity distributions using the vsn method (Huber et al., 2002). The normalized score for the Cy3 channel was subtracted from the normalized score for the Cy5 channel resulting in a single score per gene per array. A statistical model was applied to each gene treating each array as a replicate. The model contained a dye effect (Cy5-Cy3) and a treatment effect (βA-βB) where βA is the effect of the sample in the Cy5 channel and βB is the effect of the sample in the Cy3 channel. Therefore, within each experiment, the effect of a specific internode segment, as compared to the other samples in the same experiment, was evaluated by extrapolating both the direct and indirect comparisons between samples. This increased the replication for any comparison from four direct comparisons to an estimated eight comparisons.  A  t statistic was computed for each of the 6 pairwise differences that resulted  from comparing between all the four conditions within an experiment, as well as an overall F statistic testing if at least one condition differed from the others. Those genes whose overall P value was less than 0.05 and whose maximum pairwise difference was greater than ln (1.3) were clustered using the DIANA algorithm (Kaufman, 1990), a divisive hierarchical clustering algorithm, to find those genes with a similar profile over the four samples within an experiment. A total of 5994 cDNAs were thus selected.  52  F value statistic The normalized data were fitted into an effects model, after the dye effect was substracted out. An F test was carried out to test for significant difference between at least two samples within an experiment.  For the internode  segments, a total of 8855 cDNAs had a F statistic with a p value < 0.05. Consequently, a large proportion of the elements on the array showed some degree of differential expression between the internodes. In order to identify those cDNAs whose expression would increase with the progression of secondary xylem development, a subset of these, showing at least a difference between any two samples of at least ln(1.3), were selected and placed into clusters. These clusters are shown in Figure 5.3. The progression along the X axis follows the progression of internode maturation, from T5 (youngest) to T2 (oldest) (Figure 5.1).  The ten clusters created for the experiment, allowed me  to identify sets of genes that showed increased expression with the progression of secondary xylem development. In addition, it showed other groups having unique expression patterns.  2.6 Construct Design For the purposes of this thesis, two constructs were designed. The first, AtMPSilencer was used in the inducible silencing of AtMP described in Chapter 3. The second, 35S:PopMP, was used both in the attempted rescue of Arabidopsis mp mutants described in Chapter 3 and the poplar overexpression described in Chapter 4. The construct design methodology is diagrammed in Figure 2.1. Both constructs are based on pHANNIBAL (Wesley et al., 2001 and viewable at http://www.pi.csiro.au/rnai/vectors.htm).  53  54  Chapter 3. Analysis of the Arabidopsis MONOPTEROS (AtMP) gene function in mature plants 3.1 Introduction  The roles of auxin and ARFs, specifically MONOPTEROS (MP), in the early stages of plant development, including embryo patterning and primary vascular tissue differentiation and patterning are reasonably well understood (Berleth and Mattsson, 2000; Berleth et al., 2000; Aida et al., 2002; Mattsson et al., 2003; Hardtke et al., 2004).  Many informative experiments have  yielded a working model for the interaction of auxin, auxin transport, and auxin perception in setting up the pattern of primary vasculature (reviewed in Mattsson et al., 2003). The cumulative evidence includes: the observation that auxin application leads to vascular strand formation (Sachs, 1991), the distortion of vascular pattern formation by auxin transport inhibitors (Mattsson et al., 1999), the impaired vasculature of plants mutant for auxin response transcription factor (ARF) genes (Liscum and Reed, 2002), and the use of auxin reporter genes to visualize auxin response in the early formation of the procambium (Mattsson et al., 2003). The working model that has emerged involves a feed-forward mechanism for auxin-mediated vascular patterning. Auxin is produced at meristems, and as it diffuses away, a suite of auxin responsive genes are turned on (mediated through the action of ARFs such as MP). Auxin responsive genes include auxin transport genes, which are upregulated in tissues of high auxin concentration, allowing for the auxin to be “drained” away (Mattsson et al., 1999). As this local “canalization” takes place, other auxin-responsive genes are turned on, and the differentiation into vascular tissue is initiated (Mattsson et al., 2003; Heisler et al., 2005). 55  Auxin also appears to play a role in later stages of vascular system and fibre cell development, such as secondary growth in woody plants (Mellerowicz et al., 2001) and interfascicular fibre differentiation in Arabidopsis (Mattsson et al., 1999; Zhong and Ye, 2001). However, to date, relatively little work has been done to determine the molecular basis by which auxin signals are interpreted during later stages of stem development. Thus, there is an impetus for attempting to unravel the role that auxin and ARFs play in these processes. A major impediment to these studies is the fact that mutants defective for the function of ARF genes, such as mp, development may be so dramatically impaired early in development that no appreciable secondary growth or interfascicular fibre differentiation ever takes place.  A role for AtMP function in later development in Arabidopsis has been suggested by constitutive overexpression experiments (Hardtke et al., 2004). These experiments demonstrate that MP overexpressers have abnormal floral inflorescences, occasionally terminated in pin-shaped tips. However, these studies did not investigate the nature of interfascicular fibres or secondary growth (Thomas Berleth, personal communication).  This chapter reports the results of a research strategy that was undertaken to attempt to investigate the role of AtMP in Arabidopsis flowering stem development. While such bolting stems fail to undergo secondary growth (as it has been defined for this thesis), they do produce abundant non-xylem interfascicular fibres that differentiate from non-lignified parenchyma cells during the course of flowering stem development (Zhong et al., 2001; Ehlting et al., 2005). Like fibres that differentiate from the vascular cambium during  56  secondary xylem development, these cells develop heavily lignified secondary cell walls and their formation is thought to be related to auxin as mentioned above (Zhong and Ye, 2001). Thus, I sought to develop an experimental system to test the role of AtMP-mediated auxin signaling in interfascicular fibre differentiation, to complement the abundant information on the role of this transcription factor during primary vascular system patterning.  To study the specific role(s) of AtMP later in development, a system was developed to remove gene function by silencing gene expression after primary patterning was already set up and flowering stems were beginning to bolt. This approach employs the dexamethasone inducible system (Zuo et al., 2000), which allows for conditional transcriptional activation of any transgene. This system was used to drive expression of an AtMP-specific RNAi-silencing transcript.  RNAi was chosen rather than antisense or co-supression as a means of AtMP transcript-level silencing. This choice was made based on the overwhelming evidence available in the literature showing that RNAi is a great deal more effective than these other methods (reviewed in Scherer and Rossi, 2003; Horiguchi, 2004). No advantage to antisense or co-supression was apparent, and recent evidence suggests that each of these methods of silencing involves activation of the same pathway in plants (Horiguchi, 2004). For these reasons, only RNAi was employed in conjunction with inducible expression.  In addition to visible and microscopic observation, stems in which AtMP had been silenced were used as a source of RNA for a global expression profiling experiment, employing an available Arabidopsis near-whole-genome longmer  57  microarray. This experiment allows for the identification of a large number of genes whose regulation is dependent upon MP function. As stated above, the severe, pleiotropic, phenotype of constitutive overexpressor and mutant phenotypes for AtMP suggests that analysis of the entire transcriptome would be uninformative. This combination of inducible silencing with global expression profiling offers the unique opportunity to reveal genes downstream of a key primary developmental transcription factor late in development.  The data from these experiments suggest that AtMP plays several roles in development after vascular patterning. Also, global transcript profiling of AtMP-silenced plants relative to wild-type plants revealed that removing this gene product from developing stems affects the expression of many downstream genes. The data from these experiments suggest that AtMP plays several roles in development in addition to vascular patterning. Also, global transcript profiling of AtMP-silenced plants relative to wild-type plants revealed that removing this gene product from developing stems effects the expression of many downstream genes.  By comparing the data from this profiling  experiment with those from the developing Arabidopsis stem array published in (Ehlting et al., 2005), I was able to strengthen the data from both experiments, and increase the likelihood that differentially expressed (DE) genes were actually MP and auxin related. These profiling data provide insights into the function of AtMP and auxin signaling in stem development, and provide many candidate genes that may play important developmental roles.  58  3.2 Results 3.2.1 Conditional silencing of AtMP: the strategy The dexamethasone (Dex) -inducible system was developed by Chua and his colleagues (2000a) as a means for the conditional expression of transgenes in plants. The advantages of this system include: a completely foreign inducer chemical (dexamethasone is a mammalian hormone), minimal side effects of the inducer substance, and ease of use (both the transcription factor and the expression construct are contained on a single binary vector) (Zuo and Chua, 2000a).  The system involves transforming the plant with a single DNA  construct, contained within the T-DNA region of the pTA7002 binary vector. This T-DNA contains three cassettes: one encoding a constitutively expressed transcriptional activator (a steroid-inducible transcription factor), a second containing an inducible promoter that is specifically activated by the transcription factor, located upstream of a multiple cloning site (MCS), and a constitutively expressed hygromycin-resistance selectable marker. The gene of interest (in this case, and AtMP RNAi construct) is introduced into the MCS, downstream of the Dex-inducible promoter.  Once one has selected transformants, one can induce the expression of the transgene by either spraying or feeding the plant with Dex. Dex, a steroid hormone analog, readily diffuses across cell membranes. Once in the cells of the transgenic plant, Dex interacts with the steroid-inducible transcription factor, causing a conformational change. This change results in the migration of the activator into the nucleus, where it activates the expression of the inducible promoter driving the gene of interest (Zuo and Chua, 2000a).  59  In this study, an AtMP-specific RNAi construct was cloned into the MCS of pTA7002, such that induction would result in expression of the RNAi gene, activating posttranscriptional gene silencing of AtMP.  The AtMP RNAi  construct was generated in the pHANNIBAL vector (Wesley et al., 2001), which contains two MCSs that flank an intron. A 150bp segment of the AtMP ORF was PCR amplified with two sets of primers. Both forward and reverse primer pairs were identical, except that added onto the ends of these primers were restriction enzyme recognition sites. These were different for each of the four primers, such that they enabled the two products to be cloned into the two MCSs in opposite orientations. The theoretical transcript that arises from the expression of this transgene forms a double-stranded RNA (dsRNA) hairpinloop structure. The intron loop is subsequently spliced out. This type of dsRNA structure has been shown to be very effective at inducing the degradation of both the dsRNA as well as the corresponding native mRNA transcript (Smith et al., 2000) via posttranscriptional gene silencing.  By combining the Dex-  inducible system with RNAi technology, I developed a system for the conditional post-transcriptional silencing of any gene (Figure 3.1). In this case, the AtMP gene was targeted for conditional down-regulation, allowing its function(s) late in development to be tested.  60  61  3.2.2 Silencing AtMP in developing seedlings results in a mp-like phenotype  To generate lines with conditional AtMP silencing, Arabidopsis plants (ecotype Columbia) were transformed with the PTA7002 binary vector containing the constructs described in 3.2.1.  As controls, transgenic lines harbouring the  vector without the AtMP RNAi construct (referred to as empty vector lines) were also generated. To test the effectiveness of MP gene silencing in these lines, seeds from T2 generation were sown on Dex-containing MS media, and seedling growth and development compared to seedlings grown on unsupplemented MS media. Of 21 AtMP RNAi transgenic lines, 2 (hereafter referred to by the line names: 1 and 19) demonstrated a dramatic phenotype when grown on Dex but were indistinguishable from wild-type and empty vector controls on non-Dex media.  This phenotype, which resembled the  seedling phenotype of some mp mutants, is demonstrated in Figure 3.2, and described below. Line 19 was somewhat more strongly affected by the Dex treatment than was line 1. These two lines were taken to the T5 generation, with selection on kanamycin at each generation, to obtain homozygous lines.  To more carefully analyze the phenotypes of lines 1 and 19, approximately 20 seeds from each line were then sown on Dex and non-Dex containing media, along with wild-type and empty vector controls.  The controls showed no  apparent difference between the two media.  However, as previously  observed, the phenotype of Dex-treated seeds from lines 1 and 19 was dramatic, with very slow growth and impaired seedling development (Figure 3.2), reminiscent of untransformed seeds germinated on hygromycin selection plates. However, no bleaching occurred on the Dex plates as it does on the  62  antibiotic.  At 10 days after germination, approximately half of the Dex-  induced seedlings from each line completely lacked roots, and while some of the induced seedlings in both lines demonstrated adventitious lateral root development, all showed a dramatic reduction in root structures (Figure 3.2). Aside from the small size, there was no apparent change of phenotype in the cotyledons, and the plants plated on Dex did eventually produce true leaves. In contrast, the empty vector and wildtype controls displayed no change in phenotype when grown on Dex and Dex-free MS media (Figure 3.2).  This seedling phenotype of lines 1 and 19 was severe and consistent with the phenotype of mp mutant seedlings as depicted in (Hardtke and Berleth, 1998). Surprisingly, however, the AtMP-silenced plants appeared even more impaired in growth than mp mutants (T. Berleth, personal communication). This may be explained by the fact that no null mutant alleles have been identified for AtMP, suggesting that a total loss of AtMP function would lead to a lethal phenotype, and that the extent of AtMP loss of function in Dex-induced lines 1 and 19 is greater than in the mp mutant alleles studied to date.  63  64  These results show that Dex treatment has no obvious phenotypic effect in wildtype plants, or in plants transgenic for the empty vector containing the Dexinducible transcription factor, but no RNAi construct. In contrast, the strong mplike phenotypes observed in seedlings of Dex-treated lines 1 and 19, but not in untreated seedlings, is consistent with effective Dex-induced silencing of the AtMP gene in both lines. To test directly whether AtMP expression is downregulated in Dex-treated lines 1 and 19, semi-quantitative RT-PCR was performed. This PCR (depicted in Figure 3.4) demonstrates the transcript-level downregulation of MP in the Dex-treated lines 1 and 19 as compared to the empty vector control.  65  66  3.2.3 Silencing of AtMP during flowering stem formation and the resulting phenotype  The results presented above show that Dex-inducible AtMP RNAi lines 1 and 19 are useful for conditional inactivation of AtMP expression, and thus could be effective tools to examine the role of AtMP late in development. To analyse the specific effect of AtMP silencing on flowering stem development, plants from lines 1 and 19, as well as the empty vector control, were first synchronized under short-day conditions (8h light) for six weeks. These plants were then switched to long day (18h light) to induce synchronized bolting stem development.  At  this  point,  plants  of  the  three  genotypes  were  indistinguishable to the naked eye. One week after the switch to long day conditions, the plants were sprayed once daily with Dex solution to induce AtMP RNAi expression and AtMP silencing.  One week after spraying began, a premature senescence phenotype became apparent in the rosette leaves of the “silenced” lines (Figure 3.4).  These  leaves yellowed and died, leaving only the youngest leaves looking green and actively photosynthetic. Even these youngest leaves were slightly yellowed, and this was more obvious in line 19. This phenotype was somewhat stronger in line 19 than in 1, consistent with a stronger mp mutant phenotype of line 19 seedlings on Dex-containing plates (Figure 3.2). For line 19, only those leaves under 1cm in length appeared healthy (absence of yellowing), while in line 1, the healthy leaves were as large as 2cm in length. This senescence phenotype intensified over the following week, and two weeks after spraying began, almost all rosette leaves in the putatively silenced lines were completely yellowed and desiccated. By contrast, the rosette leaves of the empty control  67  plants were still a dark green, showing very limited signs of senescence or stress (Fig 3.5). This unexpected phenotype suggests that AtMP (and auxininduced gene expression) plays a role in inhibiting the onset of leaf senescence in Arabidopsis.  Two weeks after the onset of spraying, stems of approximately 10 cm had emerged from the large majority of plants from all three lines. While they were of a similar length, the stems of putatively silenced lines 1 and 19 were much thinner and less sturdy than the empty vector control (Figure 3.5).  To examine interfascicular fibre development in these stems, cross-sections were made at distances of 2, 4, 8, and 10cm from the tips of these plants, and fresh section examined under phase contrast and with UV irradiation to reveal lignified fibre cells. The most striking difference observed between lines 1 and 19 and control plants was the great reduction in diameter of the stems of the putatively silenced plants (Fig 3.6). Additionally, UV autofluorescence showed that the stems from lines 1 and 19 displayed a general reduction in lignification of vascular bundles. While the vascular bundle number and arrangement was not noticeably different from that of the control stems, the general size and intensity of autofluorescence associated with the bundles was greatly reduced.  Interfascicular fibre (IF) formation was also reduced in the stems of lines 1 and 19 relative to the abundant formation in the empty vector plants (Figure 3.6). This was especially obvious at 8cm from the tip. Among the 10 stems from control empty vector lines that were sectioned, all displayed normal IF development. Among the 10 stems sectioned from line 1, no IF was seen, and very limited fibres were observed in line 9 (See Figure 3.6 for representative  68  stem micrographs). Thus, silencing of AtMP expression at the onset of bolting stem initiation appears to significantly inhibit both vascular lignification and IF differentiation in these stems, consistent with a role for AtMP-mediated gene expression in regulating these processes late in development.  69  70  71  72  3.2.4 Global transcript profiling of AtMP silenced stems The dramatic stem phenotype that results from the transcript level silencing of MP suggests a role for MP in transcriptional regulation of processes later in development. To gain insight into the transcriptional consequences of reducing MP function in adult stems, a global expression profiling experiment was conducted. Thirty 10-cm stems, that had bolted under conditions of continual DEX induction of transgene expression from two silenced lines as well as the empty vector control were harvested, after removal of rosettes, subaerial tissue, floral organs and cauline leaves, by snap freezing in liquid nitrogen. These samples were then used for RNA extraction. cDNA was then prepared from each RNA sample and labeled with Cy3 and Cy5 fluorescent dyes for the purposes of hybridization to a 26,000 gene 70mer Arabidopsis microarray. For more detailed information on the array design, see (Ehlting et al., 2005).  The experimental design employed is diagrammed in Figure 3.7. In order to generate biological replicates for the experiment, I used RNA from two independently transformed lines (1 and 19), both of which showed Dexinducible MP silencing. Stems from each Dex-induced line (21 days after induction began) were harvested, pooled and RNA extracted independently from the line 1 and line 19 pooled tissue. As a reference, RNA was extracted from the Dex-treated empty-vector control line grown up in the same chamber and sprayed in parallel to the two silenced lines. RNA from lines 1 and 19 was then labeled and directly compared to the control line, with 6 slides used per line: 3 replicates for each of two “dye flips” (experimental and reference probes labeled with the opposite dyes to compensate for dye bias). As shown in Figure 3.7, this resulted in a total of 12 hybridizations. The data from all 12  73  hybridizations were then pooled, such that the two completely independent biological replicates could be treated as one sample to compare against the control reference.  74  75  Of the approximately 26,000 transcripts represented on the array, 861, or 3.33%, met the cutoff for differential expression (DE), defined for my purposes as >1.3 fold change in expression in MP silenced lines relative to the vector control with a p-value <0.05.  Of these, 339 had a negative expression  change, and 522 a positive. Among the negatively effected genes was, as expected, MP, which had a fold change of -1.63. While apparently modest, this was the 48th most negatively effected gene of 339 negatively regulated DE genes. Because of this reduced expression for MP, for the remaining data analysis it can be assumed that partial loss of MP function is directly or indirectly responsible for changes in expression of other DE transcripts.  In analyzing DE genes, special attention was paid to genes annotated to respond to auxin, genes known to be involved in cell wall-related metabolism, and transcription factors, since each of these may be predicted to be direct or indirect  targets  development.  of A  MP-mediated list  of  all  signaling DE  during  genes  can  inflorescence be  found  stem at  www.botany.ubc.ca/leealanjohnson, and selected genes of interest in Table 3.1.  76  Table 3.1. Auxin-related genes whose expression decreased as a result of MP silencing. AGI code  Annotation  FC  p-value  At1g19850  ARF transcription factor (MP)  -1.63  .002  At1g44350  putative IAA amino acid conjugate hydrolase ILL6  -1.55  .001  At5g65670  auxin induced gene  -1.45  .013  At5g10990  auxin-responsive family protein  -1.41  .008  At5g57090  auxin efflux carrier EIR1 (PIN2)  -1.37  .008  At1g04240  SHY2/IAA3  -1.34  .006  Surprisingly few genes known to respond to auxin, such as the “Small AUxin Up” (SAUR) or “canonical Aux/IAA” genes were DE. Only six genes on the array (including MP) annotated as responding to auxin treatment or have known or hypothetical involvement in auxin signaling, transport, or metabolism were DE (see Table 3.1), all of which were down-regulated. These genes included one functionally characterized Aux/IAA gene, SHY2/IAA3, known to interact with ARFs in regulating specific auxin responses (Weijers and Jurgens, 2005), as well as two auxin response proteins of unknown specific function (At5g65670 and At5g10990). Finally, a gene implicated in release of IAA from amino acid conjugates  (At1g44350/ILL6), and an auxin efflux  transporter, (At5g57090/EIR1) expressed and functional in polar auxin transport in roots (Blilou et al., 2005) were downregulated in the MP silenced stems, suggesting that loss of MP activity may affect auxin homeostasis, which could affect expression of these genes.  The DE transcription factors belong to various different classes, including: basic helix-loop-helix/bHLH (7), myb and myb-related (6), MADS box (5), bZIP (2),  77  B3 (2, one of which is also classed as an ARF), ARF (1), ERF/AP2 (1), KOW (1), HD-ZIP (1), NAM (1), NAC (1), TCP (1), and WRKY (1).  To gain further insight into the potential functional significance of the DE TFs, their expression changes were compared with data from an experiment in which global expression patterns of 26,000 Arabidopsis genes were profiled over the course of inflorescence stem maturation and interfascicular fiber differentiation (Ehlting et al., 2005). In this experiment, relative expression levels were compared along the length of bolting stems, and TFs whose expression was shown to go up over the course of stem development (increasing from tip to base) were identified as candidate regulators of fibre development (Ehlting et al., 2005). To test the idea that MP-could mediate the expression of such TFs over the course of inflorescence stem development and fibre differentiation, I compared the expression data from that experiment with my data on DE TFs in the MP silenced lines. This type of meta-analysis of array data has been shown to be powerful in analyzing developmental networks (Levesque and Benfey, 2004).  78  Table 3.2. Annotated transcription factors whose expression pattern in MP-silenced stems varied inversely with expression over the course of stem development.  AGI code At2g46970  Annotation Myc-related bHLH TF AtBHLH124/PIL1  FC  p-value  Ehlting et al. category1  -1.92 .000 5 Myb-related transcription factor CCA1 -1.92 .000 5 MADS box protein ANR1 -1.71 .000 Up MADS box protein AGL1/SHP1 -1.48 .024 4 bZIP transcription factor AtbZIP42 -1.47 .000 Up MADS box protein AGL5/SHP2 -1.41 .001 4 bHLH family protein AtBHLH78 -1.39 .042 Up Myb transcription factor AtMYB77 -1.38 .030 4 bZIP transcription factor AtbZIP6 -1.33 .031 Up bHLH transcription factor AtBHLH64 -1.33 .042 4 HD-Zip transcription factor AtHB6 -1.32 .008 Up AP2-EREBP family DREB subfamily A-5 -1.32 .004 Up No apical meristem NAM family protein 1.30 .017 Down bHLH family protein AtBHLH69 1.32 .025 2 MADS-box protein AGL13 1.34 .002 Down Myb transcription factor AtMYB36 1.37 .007 Down MADS box protein AGL4/SEP2 1.48 .000 6 KOW domain plastid transcription factor 1.54 .002 Down 1 Data from Ehlting et al. (2005), Expression categories 4 and 5 are up-regulated; categories 2 and 6 down regulated. “Up”, significantly upwardly differentially expressed, but not in an expression category. “Down”, significantly downwardly differentially expressed, but not in an expression category. At2g46830 At2g14210 At3g58780 At3g30530 At2g42830 At5g48560 At3g50060 At2g22850 At2g18300 At2g22430 At1g74930 At1g25580 At4g30980 At3g61120 At5g57620 At3g02310 At3g09210  79  3.2.5  AtMP  sequence  is  invariable  across  a  range  of  geographically separated Arabidopsi s ecotypes  Adaptability among individuals within a species to flourish over a range of environmental conditions is dependent, at least partially, upon the plasticity of their genome.  Variability at single nucleotides, known as single nucleotide  polymorphisms (SNPs), can result in a change in a potentially beneficial phenotype. Studying these changes can reveal much about the adaptive and evolutionary strategies of a species.  Logically, the most variable regions of the genome will be the most “dispensable”, and thus the least well-preserved. Prima facie, it would seem that the least well-preserved nucleotides would be those in non-coding regions. The available evidence supports this (Gilchrist et al., 2006). Extending this logic, the best preserved, then, should be those nucleotides whose mutation would result in a change in the coding sequence of a single copy gene. Within coding sequences, we might expect that there would be a range of variability: nucleotides that encode key regions of a protein, whose exact amino acid sequence is critical to that protein’s function, should be the best preserved of all.  Since AtMP is an indispensable single-copy gene (Hardtke and Berleth, 1998), we might expect very little plasticity within its coding sequence. Comparing the sequence of the Arabidopsis gene and its poplar counterparts, we find that the genes are remarkably well-conserved between such divergent species (as demonstrated in Figure 4.2). However, within the central region of the gene, corresponding with the transcriptional activation domain (Ulmasov et al.,  80  1999) we find that the genes diverge considerably (Figure 4.2). This is not incommensurable with the body of evidence, which suggests that the activity of transcriptional activation domains is dependent primarily upon charge, rather that a particular shape conferred by particular amino acid residues.  Thus,  unlike the domains dependent heavily on specific aa sequence, such as the DNA binding region or protein-protein interaction domains (Gerber HP, 1994), it might be expected that this central region would be more plastic.  In order to determine how much intraspecific plasticity exists for AtMP, I isolated the activiation domain of MP loci from eight geographically isolated, phenotypicially distinct, Arabidopsis ecotypes.  These included: (Bensheim,  Landsberg erecta, No-0, (first isolated in Germany), Niederlenz (Switzerland) Columbia (Missouri, USA), Dijon (France), S96 (Holland), and Wassilewskija (Russia).  As depicted in Figure 3.8, seven of these ecotypes were isolated  from various points within Europe, and the ninth (Columbia) from the USA. (Source of the ecotype geography: The Arabidopsis Information Resource, www.arabidopsis.org).  Seeds from these ecotypes, obtained from Lehle  Seeds, were germinated and grown under short day conditions, and DNA was isolated from rosette leaves. These DNA samples were then used as templates for separate PCR reactions using primers that amplify only region encoding the putative activation domain.  Sequencing the corresponding 450 nt products  revealed that there were no SNPs whatsoever. Thus, even in phenotypically distinct ecotypes of Arabidopsis isolated from as far apart as Europe and North America, there appears to be no variability within the MP coding sequence, at least not within the region corresponding with the activation domain.  81  82  The result of this simple experiment, while it fails to correspond with the great deal of variability observed between poplar and Arabidopsis MP homologs, does suggest that an indispensable single-copy gene such as AtMP is resistant to natural variation.  3.3 Discussion This characterization of AtMP’s role in stem development offers new insights into the roles of auxin and ARFs in this important developmental process. The work  reported  in  this  chapter  employed  RNAi,  chemically-inducible  transcription, and microarray technology in a single experiment.  This  approach made it possible to gain information about the specific role, late in development, both at a phenotypic and molecular level, of AtMP.  This is  significant, as AtMP is a gene for which no null mutant exists and existing mutant lines demonstrate severe pleiotropic defects (Hardtke and Berleth, 1998). Since auxin is implicated in such a broad range of developmental and stress-related processes (as discussed in 1.1), it is likely that a single ARF, such as AtMP, is involved in a wide range of these processes as well. Thus, the ability to restrict the silencing of a regulatory gene such as MP to a specific time and place (here, the bolting stem) allows for the possibility of studying the specific role played by the gene in that time and place. Once generated, lines such as the inducibly-silenced 1 and 19 reported here could be used to determine the role, if any, of AtMP and auxin in a variety of other developmental processes, simply by manipulating the temporal and spatial application of the Dex treatment.  The mp phenotype is visible even at the early embryo stages (Hardtke and Berleth, 1998). Thus, the seedling phenotype may follow from a defect in a  83  developmental process set in motion during these embryonic stages. In the silenced seedlings tested here, MP function was not interrupted in the embryo, as Dex was first applied to the fully mature seed. Thus, the close resemblance of the silenced seedlings to the mutant mp phenotype was somewhat surprising. This may be explained by a partial leakiness to the inducible promoter, which would have allowed for a certain degree of MP silencing, even in the embryo. In this scenario, though, one would have to assume that this was later “rescued” once the plants germinated, as there was no appreciable phenotypic difference at the seedling stage between the untreated transgenic and wildtype lines (Figure 3.2). To test this possibility, one could use electron microscopy to assess the transgenic seedling phenotype as was done in (Hardtke and Berleth, 1998) with mutant seedlings.  In any case, the phenotype of the silenced  seedlings (see Figure 3.2) strongly suggests that MP continues to play a role in seedling development, especially that of roots (which were greatly reduced or absent in the silenced seedlings) well after the embryo is mature.  The premature yellowing observed in mature rosette leaves where MP was inducibly silenced suggests a role of auxin and MP in leaf senescence. As discussed in section 1.1, there are a number of lines of evidence that suggest that auxin action delays leaf senescence and abscission. Thus, while MP has not been directly implicated in this specific process, it is consistent with the overall role of auxin in leaf senescence that interrupting auxin perception by removing MP would result in early senescence.  As with the above  observations, this is not a role for MP that it would be possible to have pinpointed with the mutant due the general, pleiotropic phenotype.  84  The decreased thickness and general reduction in vigour of the induced bolting stems could be partially due to a reduced photosynthetic capacity of the sprayed, senescing leaves. However, the lack of rigidity of the stems is likely a partial result of the impaired development of interfascicular fibres (IF), something independent of the reduced overall vigour of the plants (depicted in Figure 3.7).  The connection between auxin and IF development has been  demonstrated in such experiments as (Zhong and Ye, 2001), where an interruption in auxin flow was seen in plants mutant for IF formation. In this instance, it appears that in the MP-silenced lines the impairment of auxin perception has similarly interfered with IF formation.  Confirmation of this  hypothesis is difficult. However, a comparison of the two phenotypes would be simple and likely to prove informative.  A key process regulated by auxin is cell expansion (reviewed in Mellerowicz et al., 2001). By stimulating the acidification and loosening of the cell wall, auxin stimulates a cascade of cell expansion and wall deposition related physiological events (Cosgrove, 1993).  This fact may help to explain the  observed small diameter of cells (Figure 3.7), as well as the limp or floppy nature, of the MP silenced stems (Figure 3.5).  Of the 339 genes with a  significant negative expression change, 33 (9.7%) are annotated as related to cell well expansion or carbohydrate metabolism. Of the 522 genes with a positive expression change, just 13 (2.5%) have the same involvement.  The observed downregulation of the transcription factor IAA2/SHY2 in the induced stems provides an interesting point of consideration. SHY2 encodes IAA3, a member of the auxin-induced Aux/IAA family (Tian et al., 2002). Gain-of-function mutants for this gene display short hypocotyls and altered  85  auxin-regulated root development. Gain-of-function mutants have altered auxin induction of many primary response genes, suggesting that it may repress primary auxin responses.  Aux/IAA genes and ARFs are known to work  antagonistically in tandem (as discussed in Section 1.4 of this thesis) in mediating auxin regulation, and recent data suggest that some ARFs act to upregulate Aux/IAA genes (Thomas Berleth, personal communication) in a possible form of feedback self-regulation.  Hence, the downregulation of  SHY2/IAA3 as a result of MP silencing in stems suggests that these transcription factors may interact. Thus, crossing shy2/iaa3 mutants into my inducibly silenced MP line may provide insight into one of the many possible ARFAux/IAA interactions.  The dramatic phenotypes of mp mutant and MP conditionally silenced plants, combined with the high number of DE genes observed in the MP silenced stems, suggest that MP may (either directly or indirectly) regulate many genes. A possible mechanism for indirect regulation would be via the direct regulation of the expression of other transcription factors (TFs) that, in turn, directly regulate suites of genes.  Among the 339 genes whose expression had a  significant drop in the silenced stems, 21 (6.2%) were TFs. Of the 522 genes whose expression increased, 15 (2.9%) were TFs. These numbers are in line with the generally accepted number of ~1550 TFs of the total 28000 Arabidopsis genes (5.5%), though it is interesting to note that much greater percentage of TFs whose expression dropped with that of MP compared with those that rose. This would fit with the hypothesis supported in (Ulmasov et al., 1999) that MP is a predominately positively-acting transcription factor. While the data here are merely suggestive, a possible model for the regulation of  86  many of the other 860 genes that were DE on in this array experiment would by via some of these DE transcription factors.  Among the 21 TFs whose expression dropped when MP was silenced, 12 showed increased expression in the Ehlting experiment (Ehlting et al., 2005), 7 did not vary significantly, and 2 showed decreased expression. Of the 15 TFs whose expression increased with the silencing of MP, there were only data for 12 from the Ehlting experiment.  Among these, 6 showed a decrease in  expression in the Ehlting experiment, 2 showed no significant change, and 4 showed an increase. The 18 genes whose expression patterns are consistent between the experiments (i.e. down regulated in the MP down regulated stems and up regulated over the course of stem development, and vice versa, are listed in Table 3.2. and reveal potential new candidate TFs involved in auxin/MP-mediated fibre development.  Among the 18 TFs listed in Table 3.2, an exercise in “data mining” turned up some additional experimental data and suggests some interesting possible regulatory roles for AtMP. Among the several bHLH TFs that were DE in this experiment, additional data was found for one: AtHB6 (At2g22430). AtHB6 was downregulated in the experiment reported here, and upregulated in the Ehlting experiment.  It was also reported as upregulated during “weight-  induced secondary vascular formation” (Ko et al., 2004). As AtMP is known to upregulate AtHB8 (Mattsson et al., 2003), it may be that regulation of many processes is mediated via more than one MP-HB interaction.  87  Chapter 4. Cloning and characterization of a MONOPTEROS gene from hybrid poplar  4.1 Introduction For important commercial and ecological reasons, there is a great interest in the rate at which wood is produced, and in mechanisms that control the quality of that wood. The study of the molecular biology of a complex physiological process such as that of wood development is aided by the use of model species. While great quantities of information have been gleaned from studies in a wide range of woody plants, Populus, (poplars and aspens) has emerged as a tree genus that is both economically important as and is relatively amenable as a model to study by molecular biologists. The reasons for this are described in section 1.5. Additionally, successful efforts are being made to apply the wealth of knowledge from such non-woody model plants Arabidopsis thaliana to the study of wood formation in trees such as poplar (Groover, 2005). Among the exciting potential applications of this knowledge are the modification of tree growth rate, improvement of the physical qualities of the wood produced, and the production of wood that is more chemically amenable to the pulping process.  The production of woody tissue is the result of xylogenesis: the differentiation of cambial initial cells into the elongated, highly lignified, and dead, fibers and tube-shaped vessel cells that make up the bulk of angiosperm secondary xylem. These cells serve the plant first as conducting elements for the movement of water and minerals (vessels), and secondly as elements of the woody “skeleton” which provides mechanical strength and stability to the tree (fibres).  88  The plant hormone auxin, in the form of indole acetic acid (IAA), has been shown to have a variety of roles in plant development, including a myriad of processes, and auxin has long been known as a key chemical regulator of xylogenesis and wood formation (Mellerowicz et al., 2001; Groover, 2005). Exogenous application of auxin can result in the formation of new vascular elements (Roberts and McCann, 2000). It is known that auxin concentration across the vascular cambium peaks toward the side of xylem formation (Uggla et al., 1996). Thus, study of the mechanisms by which plants perceive and respond to auxin at the vascular cambium and during secondary xylem differentiation provides one entry point into the regulation of wood formation.  In multicellular organisms, developmental processes are frequently mediated by chemical morphogens. These morphogens diffuse or are transported from a point of production outward, resulting in a concentration gradient.  Such  gradients then provide a set of positional cues to developing cells (Summerbell et al., 1991). In animals, it has been well established that these gradients are “interpreted” by key transcription factors (Tabata and Takei, 2004). In plants, recent experiments have suggested that auxin acts as a plant morphogen, and that many of the broad range of developmental roles played by auxin can be explained by changes in gene expression mediated by families of transcription factors and interacting gene products that respond to changes in auxin levels (reviewed in Guilfoyle et al., 1998; Bhalerao and Bennett, 2003). As a direct example of auxin’s capacity as a morphogen, recent work has demonstrated that auxin plays a key role in regulating phyllotaxis at plant shoot primordia. In this work, it was shown that certain local auxin concentrations at the shoot apex, located at a particular minimal distance from existing primordia, determine the site of new primordial formation (Reinhardt et al., 2003).  89  Gene activation in response to auxin involves a family of transcription factors known as the Auxin Response Factors (ARFs) (Leyser and Berleth, 1999; Liscum and Reed, 2002). ARF activity is negatively regulated by interacting AUX/IAA interacting proteins, whose levels are in turn modulated by ubiquitination and proteolysis by the 26S proteosome (Dharmasiri and Estelle, 2004; Dharmasiri et al., 2005b).  Auxin appears to control ARF activity  indirectly by binding to an F-box protein.  This F-box protein activates the  ubiquitination pathway of inhibitory IAA/AUX proteins, leading to their targeted degradation (Dharmasiri et al., 2005b). ARF5, originally identified as the MONOPTEROS (MP) gene in Arabidopsis, is among the best functionally characterized ARFs, and is known to be required for auxin-induced embryo patterning and primary vascular system development (Hardtke and Berleth, 1998).  Central to the utility of a model organism is the ability to clone orthologous genes first identified in the model from other, related organisms. For example, once a mutant in a model organism is isolated and the gene responsible for the mutant phenotype is cloned, one may be interested in moving from the easily studied model to other, species of practical importance. By using the gene sequence information from the model species, one can exploit conservation of gene structure and function between related species to attempt to identify and clone a homolog from those species. There are many examples of successful use of this strategy from the animal world, where the cloning of many medically important human genes has been made possible using sequence information from genes originally identified in model animals such as the fruit fly Drosophila melanogaster, the nematode Caenorhabditis elegans  90  (Shin and Fishman, 2002), the mouse Mus musculus (Hamet and Tremblay, 2004), and even (the non-animal) yeast Saccharomyces ceriviseae (Mager and Winderickx, 2005).  This movement from model organism to practical  application also has many success stories (reviewed in Zhang et al., 2004).  Using analogous thinking, this part of my project was focused on the cloning a MONOPTEROS ortholog from hybrid poplar, using the sequence information available from the Arabidopsis gene (Hardtke and Berleth, 1998), in order to study its potential role in the regulation of wood formation. As Arabidopsis and poplar are relatively closely related species, only having diverged 100120 million years ago (Tuskan et al., in preparation), I hypothesized that the sequence of such a key transcription factor as MP would be highly conserved. (For a detailed look at angiosperm evolution, including the relatively recent divergence of Arabidopsis and poplar, see Fig. 4.1. At the time when this work was initiated (1999), very little genomic information existed for poplar. The only publicly available genomic-level sequence information available was a fairly limited Expressed Sequence Tag (EST) library (Sterky et al., 2004). Using information from Arabidopsis and the poplar EST database, I set about cloning and functionally characterizing MP from hybrid poplar.  91  92  4.2 Results  4.2.1 Cloning of the poplar gene popMP1 A BLASTN (Altschul et al., 1990) search of the poplar EST database using the Arabidopsis MP open reading frame nucleotide sequence (Hardtke and Berleth, 1998) as a query yielded one very strong hit. This hit had an e value of 1.9e-100 and was over 80% identical over the 880 bp length of the EST. The poplar EST sequence retrieved aligned to the central region of the MP open reading frame, corresponding to the 3’ end of the region encoding the DNA binding site (Figure 4.3). Based on the high level of nucleotide identity between this poplar EST and the Arabidopsis MP gene (AtMP) (Figure 4.2), it was hypothesized that this EST represented an orthologous gene. Using the EST sequence, nested primers were designed to PCR amplify a full-length cDNA from a Lambda ZAP xylem cDNA library from the P. trichocarpa X P. deltoides hybrid H11 (Subramaniam et al., 1993). This cloning strategy is depicted in Figure 4.3.  93  94  95  Nested PCR amplification of the regions 5’ and 3’ to the EST yielded fragments of approximately 1.4 and 2.4 kb, respectively (Figure 4.3). Subtracting the 800 bp of overlapping sequence and the 100 bp of sequence derived from the vector resulted in an estimated cDNA length of 2.9 kb, very similar to the 2.7 kb known length of the AtMP mRNA.  These two fragments were end-  sequenced, and this sequence information was used to design primers to amplify the full-length cDNA from the lambda library in a single reaction. This yielded the expected 2.9-kb band. This band was cloned and sequenced, and found to contain a single open reading frame. It should be noted that 5 clones were end-sequenced, each from separate PCR reactions, and each clone had identical end sequences.  The cloned cDNA was named popMP1, and the  corresponding gene PopMP1.  96  97  Overall, the amino acid identity between the predicted AtMP and PopMP1 proteins was 57%, and Figure 4.2 shows that the popMP1 cDNA sequence revealed a very high level of conservation in three of the four ARF domains found in the predicted sequence of AtMP.  As depicted in Figure 4.2, the  domains conserved between the predicted product of AtMP and PopMP1 are: the ARF-specific amino terminal B3 DNA binding domain (DBD); and domains III and IV, both protein-protein interaction (thought to mediate the interaction with Aux/IAA proteins as discussed in section 1.4) domains. In contrast to the rest of the predicted protein, the highly variable ARF middle region (MR), which is thought to function as an activation domain (Ulmasov et al., 1999) is relatively divergent between AtMP and PopMP1, with a predicted amino acid identity of less than 30%. This corresponds with predicted aa residues ~460680 in AtMP and residues ~440-710 in PopMP1. This domain is ~60 aa longer in the poplar protein, accounting for almost all of the extra length in the poplar gene.  To confirm that PopMP1 was not more homologous to another member of the Arabidopsis ARF family, a phylogenetic tree was constructed of the then-known Arabidopsis ARF family members, as well as the rice MP homolog (Figure 4.4). Since PopMP1 branched off most closely to AtMP1, the support for these genes being orthologs was strengthened.  98  99  Once the PopMP1 cDNA sequence was known, it was used as a probe for cloning the full-length genomic copy. This was accomplished through the use of a 50,000 clone poplar BAC genomic library made from P.trichocarpa “Nisqually 1”, containing an average insert size of ~114kb, previously spotted on membranes.  Ten putative clones were isolated, and Southern analysis  suggested that 3 of them contained the full-length gene (Figure 4.5). Longtemplate PCR was performed, using these clones as template and the same fulllength 5’ and 3’ primers as were used to amplify the full-length cDNA. Two of the templates yielded single 5.2-kb bands, corresponding reasonably well with the transcribed portion of the Arabidopsis 4.3 kb MP gene. These were cloned using a PCR cloning vector and end-sequenced. The two clones were identical, so only one was sequenced.  This encoded the same predicted amino acid  sequence as PopMP1 within 15 exons. Because of its origin from P. trichocarpa, and in keeping with proposed poplar gene nomenclature, this gene was renamed PoptrMP1. The 13 introns of PoptrMP1comprised a total of 2.3kb. This corresponded with the Arabidopsis MP gene, known to contain 13 exons and 12 introns.  The intron positions were also remarkably well  preserved, in most cases bounded by the same codons in each gene.  100  101  4.2.2 Determining MP copy number in poplar  Because poplar is an ancient polyploid (Tuskan et al., 2006), it is predicted that many genes will have duplicate copies. To test the copy number of MP, Southern analysis of poplar genomic DNA was performed, using for a probe the full-length PoptrMP1, as well as 150bp probes corresponding with the 3’ and 5’ ends of the coding sequence.  Using two restriction enzymes, the  resulting banding patterns suggested a single copy of the poplar MP gene, but did not exclude the possibility of other, extremely well conserved copies with conserved restriction sites (Figure 4.6).  Upon completion of the poplar genome sequence (Tuskan et al., 2006), two copies of the poplar MP gene have been annotated. PoptrMP1is located on Linkage Group (LG) II.  The second copy is located on LG V, and will  henceforth be referred to as PoptrMP2. The two copies are highly similar to one another, and both share a very similar degree of identity to AtMP. Each predicted protein shares the property of an activation domain that diverges in amino acid sequence from AtMP, as described above. Figure 4.6 depicts the comparison between the sequences of AtMP and the two poplar genes. Figure 4.6 a) compares AtMP, PoptrMP1, and PoptrMP2, showing that the gene structures are remarkably similar, with the genes having 12, 14, and 13 introns, respectively. As can be seen in Figure 4.7 b), the genes are highly conserved at both the nucleotide and predicted amino acid level, with the most variable region corresponding with the middle region (MR), thought to contain the generally variable (among ARFs) activation domain.  102  103  4.2.3 Comparison of expression patterns for the two poplar MP genes  To gain insight into what roles the two genes play in poplar development, and whether they have undergone subfunctionalization subsequent to duplication, expression patterns for the two genes were assessed in a range of tissues. RNA was obtained from xylem, phloem, mature and juvenile roots, mature and juvenile leaves, and petioles. Primers were designed to specifically amplify each of the two PoptrMP copies, and transcript levels were assessed by quantitative real-time reverse transcription-PCR (QRT-PCR).  The organ and tissue-specific expression data are shown in Figure 4.8. The two genes displayed similar levels of expression in both root types, as well as in phloem and young leaves. In petioles, the expression data suggested that PoptrMP1is expressed approximately 8-fold more highly than PoptrMP2. Interestingly, in xylem, the QRT-PCR data showed that PoptrMP1transcripts outnumbered PoptrMP2transcripts by more than1000:1.  This dramatic  difference suggests that PoptrMP1plays a specific role in xylem development, a role apparently not shared by PoptrMP2. It also fits with the observation that only a popMP1 cDNA was identified among the 5 separate PCR products amplified from the xylem cDNA library, and that the original EST found the xylem EST library was derived from PoptrMP1(4.2.1). These data suggest that there may be sufunctionalization of the two genes.  Recently, “virtual northern” data based on poplar EST abundance (Sterky et al., 2004) has provided further evidence for the subfunctionalization of the two PoptrMPgenes, as these data indicates that PoptrMP2 is most highly expressed  104  in floral buds an organ type in which AtMP is known to be expressed (Hardtke and Berleth, 1998). In this tissue, twelve ESTs were counted for PoptrMP2, and 1 for PoptrMP1. This dramatic difference offers more evidence for the subfunctionalization of the two genes.  105  106  107  4.2.4 Investigation of synteny in the genomic regions surrounding the MP genes Synteny is the conservation of gene, or genetic marker, order between divergent species. This phenomenon has been most often demonstrated for genetic markers rather than actual gene loci, between the genomes of plants belonging to members of the same family in various instances. For example, comparison of the RFLP genetic maps of tomato, potato, and pepper (all members of the Solanaceae) has proven the conservation in the order of genetic markers (reviewed in Schmidt, 2000)). Additional studies have shown that the RFLP maps of Arabidopsis and other members of the Brassicaceae (mustard family) demonstrate a level of conserved organization (Lagercrantz and Lydiate, 1996). Recently, studies such as Mudge et al., 2005 have demonstrated that gene-level synteny between divergent angiosperm families including the Brassicaceae and the Fabaceae (legumes). The existence of a poplar BAC library, and the later completion of the complete poplar genome sequence,  have  greatly  facilitated  investigation  of  synteny  between  Arabidopsis and tree species such as poplar, especially at the single gene locus level, as demonstrated by work such as that presented in (Zuo and Chua, 2000b). This study aimed to determine the degree of conservation between the poplar and Arabidopsis genomes in the region containing AtMP and PoptrMP1 and PoptrMP2.  In work completed before assembly of the poplar genome sequence, I first investigated poplar-Arabidopsis synteny in the area of MP genes using the poplar BAC library and the Arabidopsis genome sequence. To determine whether the genes neighbouring MP in Arabidopsis are the same genes neighbouring MP in poplar, I probed a 50,000-member poplar BAC library arrayed on hybridization filters with a probe specific to PoptrMP1. Of the 108  eleven BACs isolated, Southern analysis demonstrated that only three contained the full-length open reading frame (Figure 4.5). DNA from these clones was then isolated and used as template for a PCR-based test for the presence of poplar genes similar to those found neighbouring MP in Arabidopsis.  To assay for the presence of poplar genes corresponding to genes immediately surrounding MONOPTEROS in Arabidopsis, a nucleotide-nucleotide BLASTN search of the Swedish poplar EST database (poppel.fysbot.umu.se) was performed. This search yielded strong poplar EST matches for the following genes neighbouring the Arabidopsis MONOPTEROS gene (At1g19850): At1g19830, At19835, At1g19840, At1g19860, At1g19870, At1g19890, and At1g19900, representing a 54kb "neighbourhood" from Arabidopsis chromosome I surrounding MP (see Table 1 and Figure 4.9). No strong match was found for At1g19880. The poplar EST matches to these genes had an average length of 297 and an average nucleic acid identity to the given Arabidopsis gene of 69.5%.  In order to determine whether the poplar genes corresponding to the AtMP neighbouring genes were present in the same region as PoptrMP1, PCR primers, based on the EST sequences, were designed to specifically amplify a portion of each poplar gene. These were then used in PCR reactions with template DNA from three BAC clones known to contain the putative MP orthologue. The PCR products were of the expected sizes, and were cloned and sequenced to confirm that they had the sequence of the corresponding EST. The results of these PCR tests, shown in Figure 4.9, demonstrated that poplar homologues for at least five of the Arabidopsis MP "neighbour" genes  109  are present in the genomic region immediately surrounding the putative poplar MP orthologue. Further, the PCR data suggest these genes may be arranged in a similar order on the chromosome in poplar as they are in Arabidopsis (Figure 4.9). These conclusions are based on the fact that many of the genes known to immediately surround MP in Arabidopsis are contained in a ~110 kb genomic clone that also contains a poplar MP ortholog.  Table 4.1. The gene "neighbourhood" surrounding MONOPTEROS on Arabidopsis chromosome I.  Arabidopsis  TAIR  annotated Poplar EST  Gene name  gene function  homologue  EST  EST nucleic  length  acid identity  (and ORF length) At1g19830  auxin- T106B10  Putative induced  241 nt  69%  protein  (354 nt) At1g19835  Unknown (2949 nt)  At1g19840  Putative  UL64PC03  auxin- T039D03  induced  241 nt  70%  327 nt  72%  protein  (462 nt) At1g19850  Auxin  response  (MONOPTERO  transcription  S)  (2889 nt)  At1g19860  Zinc finger protein G12OP73Y  factor  ---  ---  ---  153 nt  63%  294 nt  63%  (1242 nt) At1g19870  Calmodulin binding A033P28U  110  protein (2385 nt) At1g19880  Regulator  of NONE  NONE  NONE  410 nt  78%  269 nt  65%  chromomsome condensation  (538  nt) At1g19890  At1g19900  Putative  histone UA10BPE0  protein (414 nt)  6  Glyoxyl-oxidase  M101E08  related (1647 nt)  111  112  Subsequent to these experiments, the poplar genome was completely sequenced and assembled.  The sequence information has allowed for  confirmation of the PCR-based experiments. Additionally, it has allowed for an expansion of the area studied, and a comparison of the regions surrounding both PoptrMP1 and PoptrMP2.  Analysis of data from the completed poplar sequence, shown in Figure 4.10, shows that a striking level of synteny exists between the AtMP region on chromosome I and the chromosomal regions surrounding both poplar MP genes on linkage groups I and V (Fig 4.10). These data are based on BLAST searches of individual Arabidopsis genes against a 500 kb region (~500 kb in either direction from the PoptrMP genes) from each of the poplar linkage groups. Comparison of the poplar regions corresponding to the Arabidopsis BAC F6F9 (which contains AtMP) shows not only that microsynteny exists between  both  poplar  MP-containing  regions  and  the  corresponding  Arabidopsis MP-containing region, but also provides clues as to genomic rearrangements that have taken place since poplar and Arabidopsis diverged. Examination of the poplar sequence demonstrates that there has been an expansion of intergenic distance in poplar (or, conversely, a compression in Arabidopsis) since the divergence of Arabidopsis and poplar: the 117kb F6F9 region in Arabidopsis contains 28 genes. The corresponding 500 kb regions in poplar contain just 13 genes each. Additionally, within the regions studied, there appears to have been an inversion, either in Arabidopsis or poplar of a region containing 10 genes. This is based on an exact reversal in gene order for these ten genes.  Finally, since the duplication of the MP chromosomal  region in poplar (as depicted in Figure 4.10), presumably during the whole genome duplication event in the lineage that gave rise to Populus, there has  113  been an insertion of an intervening sequence, approximately 43 kb in size, into the PoptrMP1 region.  This region does not appear to contain any coding  sequence: a search for open reading frames found none over 250 bp in length.  114  115  4.2.5 Attempted Rescue of the Arabidopsis mp mutant phenotype with PoptrMP1  In order to prove conservation of function between the poplar and Arabidopsis MP genes, an attempt was made to rescue the Arabidopsis monopteros (allele G12) phenotype using the PoptrMP1 gene.  The PoptrMP1 ORF was cloned into the pHANNIBAL vector (Wesley et al., 2001), such that the ORF replaced the intron within the vector, resulting in a 35S: PoptrMP1:NOS- terminator cassette. This cassette was moved into the binary vector pART27 and used in Agrobacterium-mediated transformation of the mutant Arabidopsis population  (the transformation experiment was  performed in the lab of T. Berleth).  No complementation was observed,  suggesting an at best partial (and thus difficult to assess) rescue (T. Berleth, personal communication).  4.2.6 Testing PoptrMP1 function by overexpressing the gene in poplar  In order to learn about what role MP (and, by extension, auxin) plays in wood development, I initiated a functional study of PoptrMP1 in developing transgenic Populus. Functional characterization of transcription factors (TF), especially in organisms with very long generation times, can often be accomplished by overexpression of the TF in question.  Gene silencing  strategies often fail, due to the frequently redundant nature of TFs: related family members can often complement the lack of function of the gene whose function is disrupted. In the instance of PoptrMP1 this is especially true: two  116  copies of the gene exist, and other ARFs may compensate even if both of these were silenced. As well, the fact that no null mp mutant has been found in Arabidopsis suggests that such a mutant would have an embryo lethal phenotype, making the likelihood of generating such a “mutant” low. Additionally, the fact that this is an auxin response factor is especially problematic, as the methods for generating transgenic poplars involve tissue culture techniques that rely on the auxin response! (Auxin is used as a rooting hormone to generate plantlets from the transgenic calluses.) Hence, I chose the following overexpression strategy, using the strong CaMV 35S promoter to drive constitutive expression of the gene.  By misexpressing the PoptrMP1  transcript, it was hypothesized that wood development would be somehow altered, which would support a role for PoptrMP1 in this process. Since the role of ARFs in wood development has not yet been studied, this could provide novel new insights into ARF function in a woody species.  Poplar leaf disks were transformed with the 35S: PoptrMP1:NOS construct described in 4.2.4 using Agrobacterium-mediated tissue culture transformation.  This procedure yielded 16 apparently transformed plants, based on kanamycin resistance. To confirm that these were genuine transformants, genomic DNA was extracted from each putative transformant line, and primers designed to anneal to the 5’ end of the 35S promoter and the 3’ end of the kanamycin resistance gene were used in a PCR reaction.  12 of the 16 lines gave  appropriately-sized bands, confirming them as transformations, and these were named K1 – K12.  117  These plants were transferred to soil after 16 weeks and allowed to grow in greenhouse conditions for 40 additional weeks. RNA was then extracted from young leaf tissue, and real time RT-PCR was performed for each sample to assess expression of the MP transgene in each line. The results demonstrated that four lines were overexpressing the PoptrMP1 transcript by approximately 4-250 fold when compared with control untransformed plants also grown from leaf disks in the same experiment, as shown in Figure 4.11. However, none of these lines showed an obvious phenotypic difference from the control lines. In Arabidopsis, it is known that the xylem expressed AtHB8 transcription factor is dramatically up-regulated by auxin, and appears to be a direct target of AtMP, based on AtMP overexpression studies (Hardtke et al., 2004). Thus, I hypothesized that the PoptrMP1 overexpressers would have upregulated popHB8 expression, if functional PoptrMP1 overexpression had been achieved. A poplar homolog of AtHB8 was identified from the poplar genome sequence and specific RT-PCR primers were designed to allow specific amplification of its transcript. QRT-PCR analysis of popHB8 expression in young leaves showed significant upregulation of this putative PoptrHB8 ortholog in four of the PoptrMP1 overexpresser lines (Figure 4.11). Thus, these data suggest that upregulation of PoptrMP1 leads to upregulation of the PoptrHB8 gene, as expected from Arabidopsis results, although the level upregulation is much lower than that of PoptrMP1 itself.  118  119  4.3 Discussion The cloning and characterization of a key auxin response factor (ARF) such as MONOPTEROS from a woody plant offers has many practical and basic research implications.  Auxin response is involved in a myriad of plant  processes, especially in development (reviewed in Kepinski and Leyser, 2005). I have succeeded in cloning one of what appear to be two copies of the MONOPTEROS gene from poplar, and completed characterization of the genes.  Sequential analysis of the PopMP1 homolog demonstrated a remarkable degree of identity, and thus conservation, between poplar and Arabidopsis. Despite having diverged over 100 mya, these organisms have conserved the MP locus: amino acid sequences of the two poplar genes show a degree of identity of 57 and 58% respectively, overall (depicted in Figures 4.2 and 4.6). If the highly variable Middle Region (MR) is ignored, the remaining sequences are nearly 80%.  Interestingly, the intron/exon structure is remarkably well  conserved: in most cases the exact same codons flank the splice sites in each of the three genes. This high degree of conservation may be reflective of MP’s essential role in development (Hardtke and Berleth, 1998).  Also noteworthy is that poplar maintains two copies of the MP gene. Adding to the noteworthiness of this finding is the much different expression patterns of the two genes (shown in Figure 4.9), as it suggests the possibility of subfunctionalization of the two copies. The very high degree of expression (over 1000 fold greater) of PoptrMP1I in xylem, for example, may suggest that this copy is responsible for MP’s role in the auxin response in secondary vascular differentiation, a process that is to this point poorly understood (from  120  a genetic standpoint, at least).  The high relative degree of PoptrMP1  expression in xylem also probably explains the finding of only that copy of the gene, as it was amplified from a xylem cDNA library (the odds were thus 1000:1 of pulling out a PoptrMP1 clone). Finally, the “virtual northern” data suggesting that PoptrMP2 is the predominant MP functioning in floral tissue offers experimental evidence from another source of the subfunctionalization of these two genes.  There are three obvious reasons for determining the degree of synteny between two organisms around a given locus. Firstly, proving that synteny exists between two loci provides evidence that the loci are orthologous. Secondly, comparing the order of genes around a given loci may provide clues as to the types of chromosomal rearrangements, such as inversions, rearrangements, or duplications, which may have occurred since the two species diverged. Thirdly, the existence of syteny allows for the positions of potentially important genes in a commercially important species to be predicted relative to genetic landmarks, based on known gene order in Arabidopsis (from the completed Arabidopsis sequence) (Schmidt, 2000; McCouch, 2001; Job, 2002; Stirling et al., 2003).  At the time when this project commenced all three of these reasons provided the rationale for the investigation of synteny between Arabidopsis and poplar. With the unexpectedly quick completion of the poplar genome sequence, the methodology for proving synteny between the two organisms has become a facile exercise in looking through genome sequence and annotation data at genome databases. However, by utilizing a simple PCR-based approach, I was able to demonstrate synteny between Arabidopsis and poplar with only a  121  poplar BAC library and EST sequence available (Johnson et al., 2004). I have also have shown, in concept a very “quick and easy” manner with which one can quickly compare any genome to that of Arabidopsis, provided that there is a certain quantity of EST data and a large-insert genomic library available. This is an early example of an easily accomplished gene synteny for a dicot species relative to Arabidopsis. Based on these data, one can extrapolate that there should be a large degree of microsynteny between the poplar and Arabidopsis genomes, and thus between dicot genomes in general. As well, the high degree of synteny in the regions immediately surrounding AtMP, PoptrMP1 and PoptrMP2 offers additional proof that the poplar genes are actually MONOPTEROS orthologs.  Functional genetic analysis is difficult in a large, long-lived (in short: poorly amenable to experimentation) organism such as poplar. However, the options of “reverse genetics” as well as functional complementation of the Arabidopsis mutant were available to me. The attempt at rescuing mp Arabidopsis with the poplar homolog yielded only negative data.  There are several possible  explanations of these data. Firstly, it may simply be that an insufficient number of lines (10) were assessed.  It may also be that the promoter used, the  CMV35S constitutive promoter may not have been appropriate for these purposes. In fact, it may be that overexpression resulted from the use of this promoter, and overexpression has been shown to result in a phenotype that is difficult to distinguish from the mutant (Hardtke et al., 2004).  Finally, it is  possible that PoptrMP1 alone is not sufficient to rescue the loss of AtMP in early development, perhaps due to a functional difference resulting from the divergence in the MR. In this scenario, it may be that either PoptrMP2 alone, or the two poplar genes working in concert are necessary to compensate for  122  the loss of AtMP.  To test these hypotheses, an obvious future experiment  would be to repeat the rescue effort, using a variety of promoters, as well as both PoptrMP genes, individually and together.  Another approach used to assess the function of PoptrMP1 was the overexpression of the transcript in poplar. While transcript-level analysis of gene expression confirmed that several lines had dramatically increased expression of the transcript, no gross phenotype was observed. Microscopic observation of cross sections of stems from each line also showed no obvious morphological differences from the wildtype, even in the xylem, where it was anticipated that PoptrMP1 plays a role, especially considering that, according to my RT-PCR data, it is the only MP expressed there. This may be explained by some compensatory measure taken by the plant to compensate for the overexpression of PoptrMP1, resulting in, for example, a failure of the transcript to actually be translated. It may also be that popMP1 does not play any role in wood development. The possibility does exist, however, that the scope of our phenotypic observation simply failed to pick up subtle differences in wood development caused by the overexpression of popMP1. Thus, future assessment of these lines would benefit from a more intense microscopic analysis of the xylem tissue.  Also surprising was the modest increase in expression of PoptrHB8 associated with the overexpression of PoptrMP1, as compared with the analogous data from Arabidopsis (Hardtke et al., 2004). This may be explained by the fact that PoptrMP1 is ordinarily expressed at a high level in young leaves (Figure 4.7).  That is, perhaps PoptrHB8 expression is normally at a saturated or  maximum level in wildtype young leaves.  It may also suggest that the  123  PoptrHB8 gene analysed is not actually orthologous to AtHB8.  Another  possibility that must be acknowledged is that the level of the PoptrMP1 transcript does not actually reflect the level of the PoptrMP1 protein. Finally, the possibility exists that PoptrMP1 is not an activator of PoptrHB8, and that upregulation of PoptrMP2, again either individually or in concert with PoptrMP1 expressed in young leaves, would have yielded the expected upregulation of PoptrHB8.  124  Chapter 5.  Global transcript profiling over the course of poplar  green stem development  5.1 Introduction Since the commencement of this thesis, there have been enormous advances in the field of global transcript profiling (for a review of some historical highlights (see (Hoheisel, 2006); (Przemeck et al., 1996). These advances have made it possible for the relatively quick and non-labour intensive study of how the expression of much of the transcriptome of a given organism varies between tissues and treatments or environmental stimuli. The opportunity to investigate the global expression profile of a developing wood-forming poplar stem using a poplar cDNA microarray (Ralph et al., 2006) within the context of this thesis was an exciting and logical extension of the work discussed in the previous two chapters, enabling me to ask new questions about wood development. The work described in this chapter placed a special emphasis on the expression of auxin related genes and transcription factors in developing woody tissue.  Global transcript profiling utilizing microarray technology is an extremely powerful method for identification of genes involved in complex developmental processes such as vascular development, as has been demonstrated in several experiments, using various model systems. The general strategy successfully identified genes involved in morphogenic events during secondary cell wall biosynthesis in Zinnia cell cultures (Demura et al., 2002). (Hertzberg et al., 2001) profiled changes in gene expression at various stages of poplar secondary xylem differentiation, work that has been followed up on looking at global expression changes associated with activity of the vascular cambium (Schrader et al., 2004) and in formation of tension wood in poplar  125  (Andersson-Gunneras et al., 2003). (Oh, 2003) and (Ko et al., 2004) were able to employ array technology to identify Arabidopsis genes that display preferred expression in secondary xylem, and during the shift from primary to secondary growth in stems. An impressive experiment by (Birnbaum et al., 2005) combined cell sorting and array technology to profile the expression of genes in several different Arabidopsis root cell types and tissues.  The  expression data for genes preferentially expressed in the stele (where xylem and phloem differentiate) generated from this experiment offer indications of candidate genes that may be involved in vascular differentiation in the Arabidopsis root.  Recently, (Ehlting et al., 2005) profiled changes in gene expression along the Arabidopsis bolting stem, using a full-genome “longmer” array. By dividing the stem into 2cm sections, this experiment was able to follow the expression patterns of genes along the stem’s developmental course, with younger tissue represented by sections nearer the tip and older, more mature tissue represented by sections nearer the base. This strategy identified many genes that may be involved in the metabolic, developmental, and regulatory events that control stem (and thus secondary vascular tissue) development. Following a similar approach, the work described in this chapter employed microarray technology to profile global changes in gene expression during the development of poplar stems.  From a very general perspective, poplar stem development follows a similar course to that of a lab-grown Arabidopsis bolting stem. At the extreme tip, both Arabidopsis and woody dicot stems (such as poplar) consist of green and largely parenchymous tissue (Larsen, 1994).  The provascular tissue is  126  arranged in bundles of cells that have elongated shape and begin to take on the characteristics of vascular tissue as one moves down the stem (Larsen, 1994), resulting in the formation of vascular bundles (see also Figure 5.1). However, in poplar and not Arabidopsis (under normal growth conditions), a vascular cambium develops as the stem matures. This cylindrical stem cell population divides laterally, generating more vascular tissues (xylem towards the inside, phloem to the outside) resulting in the radial growth of the stem and formation of secondary xylem (wood) (Larsen, 1994). In addition to studying the changes in gene expression associated with the formation of primary vascular tissue and fibre differentiation, this study also sought to identify candidate genes involved in the process of initiation of the vascular cambium.  Global expression profiling experiments provide enormous quantities of data. For the purposes of my thesis, I chose to focus specifically on three classes of genes whose expression was differentially regulated over the course of poplar stem differentiation. Firstly, as discussed in section 1.3, auxin is known to be involved in the process of vascular differentiation, specifically the activity of the vascular cambium.  Thus, this work provided an opportunity to assess the  changes in auxin-related gene expression that may be involved in the auxinvascular cambium signaling process. Secondly, by spatially isolating the tissue where the vascular cambium first forms, I looked at the genes whose expression spiked specifically at this location as a source of potential cambiumregulating candidates. Finally, as transcription factors are known to be key players in the mediation of plant vascular development (Sieburth and Deyholos, 2006), I focused on transcription factors whose expression was dramatically altered over the course of stem development.  127  5.2 Results and Discussion  5.2.1 2,320 microarray elements that vary over the course of stem development fit into 7 distinct clusters  A conventional manner of dividing woody stems is based on the progression of internode and node differentiation (for example, see (Lybeer et al., 2006)). Each internode is separated from the adjacent by intervening buds and leaves at nodes. I commenced this work by examining, the state of vascular development in each of the first few internodes in a developing poplar stem. Young poplar stem tips were taken from ~2m hybrid poplar (Populus trichocarpa X P. deltoids clone H11-11) trees, grown outdoors from cuttings, in early June. Cross-sections were prepared from the middle of each of the first five internodes. By convention, these were named according to their proximity to the first fully-expanded leaf: The internode immediately above this leaf was named -1, the next node -2, and so on, as indicated in Figure 5.1. In the sample of approximately twenty plants, each had exactly five distinguishable internodes above the first fully expanded leaf. As a measure of primary and secondary xylem development in these sections, lignin deposition was monitored in sections from each internode by autofluorescence under UV light excitation. In all, ten stems were sectioned and observed.  The results are  shown in Figure 5.1. In internode -5, the youngest internode, discrete vascular bundles were observable, with no continuous ring of lignification present. By internode -3, a continuous ring of lignified vascular tissue was apparent, and this grew thicker by -2, suggesting the presence of an actively dividing cambium.  128  RNA was isolated from each of the first four internodes and used to generate Cy3 and Cy5 probes for hybridization to poplar cDNA microarrays (Ralph et al., 2006).  The experimental design employed for hybridization was a four element factorial design. In this approach, each sample was compared directly to each of the others, with a “dye-flip” performed for each.  This is represented  diagrammatically in Figure 5.2. The total slides needed for this design was 24 (6 comparisons x 2 technical replicates x 2 “dye-flips”).  129  130  Of the 15,496 elements present on the array, (Ralph et al., 2006) 5,994 were associated with genes whose expression varied with a fold change 1.5 or greater and had a p-value of <0.05.  Henceforth, these elements will be  referred to as differentially expressed (DE).  According to a clustering  algorithm, these DE were fitted into 10 clusters, each of which contained elements of comparable expression pattern. Figure 5.3.  These clusters are depicted in  Among these, a special focus was placed on cluster 5, as the  general pattern for this cluster showed a peak of expression at internode -3, corresponding with the observed initiation of the vascular cambium. As well, array elements associated with genes whose expression varied most dramatically (see Figure 5.3) were sorted into those that increased as vascular development increased down the stem (clusters 3, 4 and 9), and those that decreased (clusters 6, 8 and 10). Through this clustering of DE genes, a huge amount of information was filtered into a total of 7 clusters corresponding to a smaller dataset of 2,320 array elements.  131  132  133  5.2.2  Elements associated with “The Liginification Toolbox” are generally upregulated during stem development  In order to test the reliability of the data collected, I focused on the expression of those DE elements likely associated with phenylpropanoid metabolism and lignification.  Upregulation of genes required for lignin biosynthesis is  associated with formation of woody tissues (Mellerowicz et al., 2001). Thus, one would expect that these elements would be positively DE along the course of early poplar stem development, as wood deposition starts and continually increases among the sections sampled (see Figure 5.1). It was expected that genes associated with lignification that were DE would fit predominately into the “upregulated” clusters 3, 4, and 9. To conduct a brief survey, the data for elements whose best hit was to each of the genes in “The Arabidopsis Lignification Toolbox” (Raes et al., 2003) were compiled. In parallel, an annotation of the poplar lignin biosynthetic genes in poplar has been undertaken as part of the effort to annotate the poplar genome (B. Hamberger, M. Ellis, C. Souza, and C. Douglas, unpublished; Tuskan et al., 2006). A representation of the monolignol biosynthetic pathway, and the number of annotated genes in poplar and Arabidopsis encoding each enzyme is shown in Figure 5.4.  The data for DE elements associated with phenylpropanoid metabolism agree with the above prediction. As shown in Table 5.1 and Figure 5.4, DE elements were found corresponding with genes encoding: phenylalanine ammonia-lyase (PAL); 5 elements, hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyltransferase  (HCT);  1  element,  4-Coumarate-CoA-ligase  (4CL); 2 elements, cinnamic acid-4-Hydroxylases (C4H); 2 elements, 5  134  elements, caffeoyl-CoA 3-O-methyltransferase cinnamoyl-CoA  reductases  (CCR);  1  (CCoAOMT); 6  element,  cinnamyl  elements, alcohol  dehydrogenases (CADs); 3 elements, and ferulate-5-hydroxylase (F5H); 3 elements. The majority of these were in the upregulated clusters 3, 4, or 9 (Table 5.1, Figure 5.4).  There is only one annotated C4H gene in Arabidopsis, At2g30490 or AtC4H. The array element associated with this gene fell into upregulated Expression Cluster 3, fitting with the logic that with increased wood development, genes associated with lignification will be upregulated.  Similarly, three array  elements associated with AtF5H (At4g36220), specific to S-lignin biosynthesis, fit into clusters showing upregulation over the course of stem development: two elements in Cluster 4 and one in the most dramatically upregulated cluster, Cluster 9. All 6 of the array elements associated with CCoAOMT had AtCCoAOMT1 (At4g34050) as their best hit to the Arabidopsis genome. All 7 of these fell into cluster 4, fitting with data from (Raes et al., 2003) and (Ehlting et al., 2005) that showed AtCCoAOMT1 expressed more highly in the lignified bases of Arabidopsis stems than the apical portion. Two elements associated with CAD were DE.  These were associated with the Arabidopsis AtCAD1  (At4g34230) and AtCAD2 (At3g19450) genes, using the nomenclature of (Ehlting et al., 2005). Each fell into Cluster 4, again showing upregulation over the course of stem development. Again, this agrees with data from (Raes et al., 2003) and (Ehlting et al., 2005). Finally, a sole element associated with CCR was shown to be DE, and this had as its best Arabidopsis hit CCR2 (At1g80820). This element fit into Cluster 4.  These data show a strong trend of upregulation of genes annotated as being  135  involved in lignin biosynthesis over the course of stem maturation, consistent with the increased commitment to lignin biosynthesis during secondary growth and secondary xylem formation. However, elements corresponding to PAL and 4CL, phenylpropanoid genes in the general phenylpropanoid pathway, were not uniformly upregulated and in the case of 4CL showed differing trends between members of the same gene family.  Interestingly,  although  PAL  activity  is  required  for  entry  into  the  phenylpropanoid pathway and lignin biosynthesis, all 5 of the elements corresponding to PAL genes were found in the downregulated Cluster 6. Each of these elements had as its best hit to Arabidopsis PAL1 (At2g37040). This may be explained by the fact that PAL activity is at the entry point into the general phenylpropanoid pathway (Chapple et al., 1992; Subramaniam et al., 1993), and is required for the biosynthesis of many other phenylpropanoid derived natural products in addition to monolignols. Such products known to accumulate at high levels in young poplar tissues are condensed tannins (CT), flavonoid derivatives, and phenolic glycosides (Harding et al., 2005).  The data reported here suggest that the poplar PAL1 gene represented by elements in expression cluster 6 encodes an enzyme with maximum activity in younger tissues, and that PAL1 expression peaks before wood formation begins, and then drops off, as is seen in the trend for Cluster 6. This agrees with data from such work as that reported by (Subramaniam et al., 1993) and (Kao et al., 2002) that shows that the expression of distinct poplar PAL genes is strongest in young tissues where CT and PG accumulate. For example, the PAL array element, EST WS0154_H05, in cluster 6, is 96% identical to P.tremuloides PAL1, which whose expression is strongly associated with  136  phenolic metabolism in young foliar tissues, and is not expressed in concert with secondary growth (Kao et al., 2002), and this gene is nearly identical to P. trichocarpa X P. deltoides PAL1, which is highly expressed in the subepidermal cells of young leaves and green stems, but poorly expressed in secondary xylem (Subramaniam et al., 1993).  Two DE elements annotated to lignin biosynthetic genes corresponded to 4CLs. One annotated as being related to At4CL2 (At3g21240) fit in the upregulated Cluster 4, and that annotated as being related to At4CL3 (At1g65060) fell in the downregulated Cluster 6. This is in keeping with prior data on the Populus 4CL gene family, which showed that 4CL gene family members are differentially expressed during poplar development (Allina et al., 1998; Hu et al., 1998). The array element corresponding to At4CL3, WS0204_H13, is 98% identical to P. tremuloides 4CL2 (Hu et al., 1998) and P. trichocarpa X P. deltoides 4CL4 (Cukovic et al., 2001). The microarray results are consistent with the high expression of P. tremuloides 4CL2 in young tissues and its postulated role in CT biosynthesis, as well as the hypothesized role of At4CL3 in flavonoid biosynthesis in Arabidopsis (Ehlting et al., 1999). In contrast, the upregulated element in Cluster 4 is likely involved in lignin biosynthesis, although the element, WS0214_L08, is not identical to P.tremuloides 4CL1, highly expressed in secondary growth (Hu et al., 1998), and may represent another 4CL gene family member. Also appearing in the upregulated Cluster 4 was an element associated with the (as yet uncharacterized) 4CL-Like1 (CLL1) At1g20510.  137  Table 5.1.  Differentially Expressed Elements associated with the  genes in “The Lignification Toolbox”. Arabidopsis gene name 4CL2 (At3g21240) 4Cl3 (At1g65060) CLL1 (At1g20510) PAL1 (At2g37040)  C4H (At2g30490) CCoAOMT1 (At4g34050)  F5H1 (At4g36220)  CAD2 (At3g19450) CAD1 (At4g34230) CCR2 (At1g80820)  Array element  Cluster # (trend)  WS0214_L08 WS0204_H13 WS01127_F24 WS0154_H05 WS0182_B10 WS0186_F12 WS0188_J10 WS0212_I14 WS0154_H05 WS0182_B10 WS0113_E16 PX0015_K01 WS0114_A10 WS0121_N05 WS0151_G16 WS0153_I10 WS0155_L24 PX0019_K01 WS0133_H14 WS0195_K04 WS0171_F04 WS0153_H17 WS01222_P11  4 6 4 6 6 6 6 6 6 6 3 4 4 4 4 4  (upregulated) (downregulated) (upregulated) (downregulated) (downregulated) (downregulated) (downregulated) (downregulated) (downregulated) (downregulated) (upregulated) (upregulated) (upregulated) (upregulated) (upregulated) (upregulated)  4 4 9 4 4 4  (upregulated) (upregulated) (strongly upregulated) (upregulated) (upregulated) (upregulated)  138  139  5.2.3 Differential expression of elements associated with auxin signaling  As discussed in 1.3, the plant hormone auxin is a key hormonal regulator of secondary vascular development in woody plants. The signaling components that underlie this regulation, while well studied recently (see, for example, (Swarup et al., 2002; Dharmasiri et al., 2005b), are not yet well understood. Hence, in analyzing the DE elements in this experiment, a focus was placed on those elements whose closest hit in Arabidopsis are annotated to have an auxin-related function, since these should identify genes encoding potential auxin-related signaling proteins that may be important in secondary xylem development at the poplar shoot apex.  I identified 13 auxin-related DE  elements, each of which had as their best hit to Arabidopsis genes in one of three general classes of function: auxin transporters, auxin related transcription factors, and the general category of “auxin induced genes” (see Table 5.2). 10 of these were seen to be upregulated as vascular development proceeded, while 3 fit into clusters that were downregulated.  140  Table 5.2. Differentially Expressed auxin related genes. Best hit in Arabidopsis At2g17500  T5:T2 FC1 9.099  P value  Cluster  1.96E-13  WS01216_B0 7 WS0213_D07  4.49  3.12E-10  9 (dramatically upregulated) 4 (upregulated)  2.91  3.37E-14  4 (upregulated)  WS01111_H2 0 WS02010_C2 4 WS0207_D18  2.11  3.37E-14  3 (upregulated)  2.01  3.07E-09  3 (upregulated)  1.861  3.12E-08  3 (upregulated)  WS01214_O2 2  1.732  1.23E-05  5 (upregulated, peaks at T3)  Auxin Induced Auxin efflux carrier AUX/IAA protein  WS0193_A21  1.703  2.64E-11  3 (upregulated)  ARF transcription factor At3g02260 Auxin transport protein At1g04240 AUX/IAA (IAA3/SHY2) protein 1 fold change between internode  WS0158_M20  At5g43700 (IAA4) At2g46370 (ARF19) At4g32280 (IAA29) At4g32280 (IAA29) At5g20730 (ARF7/NPH4) AT4G14550 (SOLITARYROOT/IAA14) At5g27780 At1g20925 At4g14550 (SOLITARYROOT/IAA14) At5g62000 (ARF2)  Annotated Function Auxin efflux carrier AUX/IAA protein ARF transcription factor AUX/IAA protein AUX/IAA protein ARF transcription factor AUX/IAA protein  Array element WS0185_J05  WS01116_K1 4 WS01214_O2 2  1.038  5.79E-01  1.732  2.64E-11  (upregulated, at T3) (upregulated, at T3)  6 (downregulated) 0.466  5.64E-05  WS0161_E17  WS01118_E0 1 T5 and T3  5 peaks 5 peaks  6 (downregulated) 0.389  3.80E-10  0.184  1.18E-15  8 (dramatically downregulated)  The auxin-related transcription factors As described in 1.4, auxin regulation of gene expression is achieved via a signaling pathway that concludes with the direct activation or repression of transcription. The transcription factors that mediate that step are, generally, the Auxin Response Factors (ARFs) and the AUX/IAA transcription factors (see 141  1.4). AUX/IAA proteins are thought to bind to ARFs, repressing their functions, while the ARFs are thought to directly either promote or repress expression in the absence of AUX/IAA proteins.  In this experiment, five DE genes are  annotated as AUX/IAAs, and two are ARFs.  The DE array element annotated as IAA3/SHY2 fell into expression cluster 8, the most dramatically downregulated. This is in keeping with the role for this gene as a repressor of auxin-regulated gene expression in Arabidopsis (Tian et al., 2002).  This gene, when overexpressed in Arabidopsis, generates a  phenotype that involved enlarged cotyledons, short hypocotyls, and altered auxin-regulated root development.  As well, overexpression was shown to  result in altered expression of many genes known to respond to auxin (Tian et al., 2002). The data here, demonstrating a dramatic downregulation over the course of early secondary vascular development, suggest that this gene may be a key negative regulator of secondary vascular development makes it an attractive target for further study. It is especially noteworthy that this gene was also shown to be dramatically downregulated in response to AtMP silencing, as reported in 3.3 of this thesis.  The DE array element annotated as ARF2 fell into the downregulated cluster 6. As is the case for IAA3/SHY2, ARF2 is also thought to act as a repressor of auxin-regulated gene expression (Okushima et al., 2005).  Arabidopsis plants  with loss-of function mutations for this gene exhibited elongated hypocotyls with enlarged cotyledons (Okushima et al., 2005). Interestingly, this loss-of-function phenotype logically corresponds with that of the gain-of-function mutants for IAA3/SHY2, at least as regards the hypocotyls. While no data have been reported with regard to the secondary vascular development for either of these  142  mutants, the additional data presented here, showing both genes are downregulated as secondary vascular development begins, suggests that both would make excellent candidates for future reverse-genetics studies in poplar and other woody species.  The remaining 6 DE auxin-related transcription factors or AUX/IAA proteins were upregulated, falling in clusters 3, 4, 5, and 9. Two of the auxin TF DE genes fall in cluster 5 (upregulated specifically in internode T3) were both similar to SOLITARY-ROOT/IAA14. This transcription factor is thought to play a role in the promotion of lateral root formation: in (some) iaa14 mutants, as well as transgenic lines expressing IAA14 in the stele, lateral roots do not form (Okushima et al., 2005). The process of lateral root formation is influenced by auxin, and the data presented in (Okushima et al., 2005) suggest that IAA14 plays a role in this process by inhibiting the action of ARF7/NPH4 and ARF19. ARF7 and ARF19 double mutants also exhibit the solitary root phenotype. Interestingly, a DE gene similar to ARF7/NPH4 is also upregulated, and appears to have an expression pattern that overlaps with the IAA14-related genes.  It is also noteworthy that these poplar IAA14-related genes both  showed a peak of expression in T3, the first internode where the vascular cambium is apparent (see Figures 5.2 and 5.3).  Expression data from  Arabidopsis (Okushima et al., 2005) show that IAA14 expression is strongest in root tissue, specifically in the stele and root tip. Reduced IAA4 expression has also been observed previously in arf7/nph4 null mutants by RNA gel blot analysis (Stowe-Evans et al., 1998). This is in keeping with the data presented here, in which DE gene similar to IAA4 (about which little else is known) is also upregulated in cluster 4, in concert with NPH4.  The data presented here  suggest that this gene, as well as the NPH4/ARF7 and IAA14 related poplar  143  genes may have functions related to establishment of the vascular cambium, or could perhaps play roles in the suppression/promotion of buds rather than lateral roots.  NPH4/ARF7 is of additional interest to this thesis in that it has an overlapping function with MONOPTEROS (MP) in Arabidopsis (Hardtke et al., 2004). The data presented in that study show that co-overexpression of MP and NPH4 results in an additive phenotype: axis formation is more greatly altered than in the single overexpressors, and auxin responsive gene expression is also dramatically altered. While MP was not represented on the arrays used, and thus no data is available from the microarray experiment, the variation in expression of ARF7/NP4H suggests that further study of both this gene and MP in the regulation of secondary vascular formation in poplar is likely to provide useful information, as discussed in detail in Chapter 4.  None of the remaining 4 auxin-related DE genes has been well characterized in Arabidopsis. However, the data presented here, showing that these genes are differentially expressed in the early stages of secondary vascular differentiation makes them candidates for further functional characterization, both in Arabidopsis and poplar.  The putative auxin efflux carriers Of particular interest are the two DE elements whose best hit was to Arabidopsis genes thought to mediate auxin efflux.  One of these,  WS0185_J05, fit into cluster 9, and was upregulated by more than nine fold over the course of the experiment.  The best hit for this element was  At2g17500, an uncharacterized gene, but one annotated to function in auxin  144  efflux. The DE gene represented by element WS01116_K14, also having a best hit to a hypothetical auxin efflux carrier (At1g20925), fit into the interesting cluster 5, which showed a peak in expression corresponding with the initiation of vascular cambium at internode -3.  Since neither of these  putative efflux carriers in Arabidopsis has been functionally characterized, it is difficult to speculate as to their roles in poplar. However, it is known that there is a correlation between gradients of auxin concentration and the vascular cambium (see 1.3 and (Uggla et al., 1996; Tuominen et al., 1997), suggesting that auxin is involved in the transition to secondary growth. The work reported by (Schrader et al., 2004) confirms that there is a correlation between these auxin concentrations and the expression pattern of certain auxin transporters in the poplar vascular cambium. The interesting expression patterns of the two DE genes reported here suggest that they will make excellent candidates for further functional characterization in poplar or Arabidopsis.  5.2.4 52 Differentially expressed elements are associated with transcription factors  Among the 2320 DE elements, 52 had transcription factors (TFs, based on their GO annotated gene function) as their best hit in Arabidopsis. (These are in addition to the 9 DE auxin-related transcription factors discussed in 5.2.3.). These genes have potential as candidates for the regulation of the transition to secondary growth, secondary xylem formation, and/or secondary wall formation, as transcription factors often play vital roles in regulating complex developmental processes such as vascular differentiation (see 1.6). Thirty-two of these were upregulated over the course of stem development: 4 fit into the upregulated cluster 4, 23 into the slightly less strongly upregulated cluster 3,  145  and 3 into cluster 5, which peaks at internode -3, the site of cambial initiation. Twenty of the DE elements associated with TFs were downregulated: 2 (overall, very slightly downregulated, but peaking at internode -3) fit into cluster 5, 2 into the strongly downregulated cluster 8, 3 into the very strongly downregulated cluster 10, and the remaining 13 into the moderately downregulated cluster 6.  Cluster 5 Five non-auxin-related DE elements with best hits to TFs fit into Cluster 5, which contains genes with the very interesting property of peaking in expression at internode -3, expression coinciding with the initiation of the vascular cambium (see Figures 5.2 and 5.3). Their best hits were: At1g13450 (array element WS0165_N02), At3g58720  and  encoding  an  At3g02290  uncharacterized (array  elements  DNA  binding  protein,  WS01126_A05  and  WS0195_A09), encoding C3HC4-type RING finger proteins that are as yet not functionally  characterized,  At5g08790  (array  element  WS0234_I22),  encoding a NAC/NAM protein not yet functionally characterized, and At5g49520 (array element WS0234_G01), encoding a WRKY transcription factor that is also not yet functionally characterized.  All of these TFs are candidates for further characterization, as their unusual expression pattern strongly suggests an involvement in cambial initiation. Particularly promising are the putative RING finger and NAC/NAM genes. It has been shown in Arabidopsis that certain RING proteins promote the ubiquitination and degradation of NAC1 (Xie et al., 2002).  Further, other  interactions between members of the RING finger and NAC/NAM families have been demonstrated by yeast 2-hybrid experiments (Greve et al., 2003).  146  The picture is made even more interesting by the fact that certain members of the NAC/NAM family (such as NAC1) are induced by auxin (Xie et al., 2002). The combination of this information suggests that the poplar NAC/NAM gene represented by array element WS0234_I22 (corresponding with At5g08790) may be activated by high auxin concentrations in the newly-formed vascular cambium, and that one or both of the poplar RING finger genes represented by array elements WS01126_A05 and WS0195_A09 (those corresponding with At3g58720 and At3g02290) may act to negatively regulate this action. While this model remains highly speculative, these novel expression data warrant further investigation into what may be a key step in the transcriptional regulation of vascular cambium formation. Future experiments, such as yeast two-hybrid assays to investigate potential protein-protein interactions and reverse genetic functional analysis of these genes promises potential information regarding the initiation of the vascular cambium in woody dicots such as poplar.  TFs commonly upregulated in numerous array experiments In the work reported by (Ehlting et al., 2005), a “data mining” analysis of multiple microarray experiments identified ten Arabidopsis transcription factors associated with Arabidopsis bolting stem development.  These ten were  expressed in concert with three physiological phenomena: the onset of fibre cell development in stems, with secondary wall development, and auxin regulation, and have expression patterns in other microarray experiments which support potential roles in fibre and xylem differentiation (Ehlting et al, 2005). In order to transfer this knowledge to poplar, I compared this set of ten candidates with the array results reported here.  147  The Douglas lab has identified the putative poplar orthologues of the ten Arabidopsis TF genes by a combination BLAST searches and phylogenetic reconstructions (M. Ellis, E. Li, and C. Douglas, unpublished). Of these ten, seven had putative poplar orthologues represented by elements on the poplar microarray employed in this experiment (Table 5.3). Of these, four were DE over the course of poplar stem development. These were the HB gene KNAT7 (At1g62990), which corresponds with array element PX0015_H02, MYB20 (At1g66230), which corresponds with array element WS0208_D15, MYB63 (At1g79180), which corresponds with array element WS0206_D10, and a C3H zinc finger gene (At5g42200), which corresponds with array element WS0152_D07.  All four of these genes were upregulated in the Ehlting  experiment and other experiments to which that data was compared (Ehlting et al., 2005 and references therein).  In this experiment, these genes fit into  Clusters 4 (dramatically upregulated, KNAT7) and Cluster 3 (moderately upregulated, other three genes). No functional data has yet been published for MYB63, MYB20, or the C3H zinc finger TF. Although little functional information has been published with regard to KNAT7, similar analyses of multiple arrays (Persson et al., 2005) have concluded that KNAT7 is involved in the cellulose biosynthesis in the secondary cell wall, and Brown et al. (2005) reported that knat7 mutant Arabidopsis plants display an irregular xylem phenotype.  Clearly, further functional analysis of these four genes is  warranted, in both poplar and Arabidopsis.  148  Table 5.3. Putative poplar orthologues of Arabidopsi s TF fibre regulatory genes and microarray expression data.  1 2  3  Ar abi dops is can didat e 1 At1g29950  Gen e family bHLH  At4g29100  bHLH  At1g62990  HB  At1g66230  MYB  At5g16600  MYB  At1g79180  MYB  Na me  JGI gen e mo del 2  AtbHLH1 44 AtbHLH0 68 KNAT7  fgenesh4_pm.C_L G_XI000130 estExt_fgenesh4_p g.C_LG_XVIII0694 estExt_fgenesh1_p g_v1.C_LG_I0964 grail3.003801020 1 eugene3.0000226 1 fgenesh4_pg.C_LG _V000361 eugene3.0007079 93 eugene3.0001065 7 eugene3.0147006 9 gw1.VII.1997.13 gw1.V.3010.13 grail3.000301300 13  AtMYB2 03 AtMYB4 33 AtMYB6 33  At5g07580  AP2-EREBP  No name  At5g24800  bZIP  AtbZIP9  At5g65210  bZIP  AtbZIP47  At5g42200  C3H  No name  Array elem ent WS0186_A24  Cluste r  WS0165_C20  Not DE  PX0015_H02  4  WS0208_D15  3  No element  N/A  WS0206_D10  3  No element  N/A  No element  N/A  WS01116_M2 2 No element No element WS0152_D07  Not DE  Not DE  N/A N/A 3  Data fro m Ehltin g et al. (2005 ) Gen e mo dels of putativ e po plar orth olo gue s from J GI Po plar bro w ser ver sio n 1.0 (http:// ge no me .jgi-psf. org/P optr1/P optr 1.ho me .html Orthol ogy to th e po plar ge ne s un cle ar due to hig h le vel s of s equ enc e simil arity  My poplar microarray data, in combination with data from Arabidopsis, help to define the most promising TF genes as candidates for further gene-by-gene characterization, with respect to fibre and xylem differentiation and secondary cell wall biosynthesis and deposition (Table 5.3). Conversely, the data suggest that Arabidopsis candidates and corresponding poplar TF genes such as bHLH068, bHLH144, and AtbZIP9, that are not DE in the poplar experiment, are of less interest for further functional characterization.  149  Table  5.4.  Differentially  Expressed  Transcription  Factors  of  Particular Interest. Best hit in Arabidopsis At1g13450  Array element  At3g58720  WS01126_A05  At3g02290  WS0195_A09  At5g08790  WS0234_I22  At5g49520  WS0234_G01  At1g62990 (KNAT7)  PX0015_H02  MYB63  WS0206_D10  WS0165_N02  Interesting Property(ies) Cluster 5  Annotated TF Class “DNA binding”  Cluster 5, may interact with NAC/NAM Cluster 5, may interact with NAC/NAM Cluster 5, may interact with RING Cluster 5  RING Finger  Upregulated in many vascular arrays Upregulated in many vascular arrays  HB  Cluster # (trend) 5 (peaks at cambium initiation) 5 (peaks at cambium initiation)  RING finger  5 (peaks at cambium initiation)  NAC/NAM  5 (peaks at cambium initiation) 5 (peaks at cambium initiation) 4 (upregulated)  WRKY  MYB  3 (moderately upregulated)  The work reported here demonstrates the enormous potential of global transcript profiling as applied to a specific developmental process. Alone, the data provided from such an experiment as this are not necessarily conclusive, and it must be acknowledged that, due to the many potential sites for error inherent in the microarray process, some of the genes discussed here are probably “false positives”. However, differential expression of certain genes (KNAT7, MYB20, and MYB43 homologs) has already been verified by quantitative real time RT-PCR (E. Li and C. Douglas, unpublished), and other candidates could be similarly tested. I have, to the extent possible, combined  150  these data with those from other experiments (Birnbaum et al., 2005; Ehlting et al., 2005), and, in the case of IAA/SHY2, even the data from the Arabidopsis array experiment reported in Chapter 3 of this thesis.  This process greatly  magnifies the probability that the genes reported are “truly” involved in the process of stem development. Thus, for such genes as IAA/SHY2, or KNAT7, further characterization is very likely to lead to new information regarding the regulation of woody stem development. As well, the vast quantities of data generated from this array experiment have only been explored here at a very focused level.  For those interested in specific genes (for example other  signaling pathways than that of auxin) the data generated from this array will also provide a useful future point of reference. Finally, cross-comparison with comparable experiments, such as that of (Ehlting et al., 2005), which studied global expression changes during Arabidopsis stem development, or that of Friedmann et al. (submitted), which used an analogous expression profiling strategy with spruce apical shoots, might lead to broad insights about the complex process of wood development that would otherwise be extremely unlikely to have been made.  151  Chapter 6. Conclusion and a look to the future  This thesis reports a substantial investigation into the genetic regulation of wood and fibre development.  While the approaches taken in the three  chapters that report data (Chapters 3, 4, and 5) differ, there are several “threads” that can be considered to “tie” the data together and attempt to put the collection of conclusions into a few “big picture” considerations.  The data reported in Chapter 3 suggest that MONOPTEROS plays a role in the development of interfascicular fibres in the bolting stem of Arabidopsis. This conclusion, made possible only through the use of an inducible-silencing strategy, is especially important, as it represents a “piece in the puzzle” of how the hormone auxin can play such a variety of roles: a single ARF gene such as MP, critical to embryo development, is shown also to be necessary much later in development.  As stated in Chapter 3, such a conclusion would be  impossible with traditional techniques of forward-genetics and knock-outs, due to the pleiotropic effects of completely removing the MP gene product.  While  the wide array of roles played by auxin (as well as the other phytohormones) has long seemed mind-boggling, by considering the “multi-purpose” capacity of a single ARF such as MP, and then imaging roughly twenty of these interacting with roughly thirty Aux/IAA gene products, the mystery begins to unravel. The microarray experiment reported in Chapter 3 suggests candidates for transcription factor genes acting downstream of MP, and will be interesting to combine with unpublished expression profiling data on Arabidopsis lines misexpressing MP at the seedling stage (T. Berleth, personal communication).  152  The data reported in Chapter 4 show that MP has been duplicated and undergone subfunctionalization in poplar. This fact, when considered with the conclusions from Chapter 3, opens the possibility that other genes, active in a variety of tissues or developmental stages, may have undergone duplication and subfunctionalization in species such as poplar that have undergone genome duplications.  This may represent an evolutionary strategy that has  been used for a variety of other “multi-purpose” genes (such as the other auxin-related transcription factors) between Arabidopsis and poplar, and merits further investigation.  A second thread between the various Chapters involves the information derived with regard to the auxin related genes. Unfortunately, neither copy of PoptrMP was represented on the array used in Chapter 5. Nonetheless, the differential regulation of an array of auxin-related genes over the course of stem development adds to the body of evidence that auxin is an important regulator of woody stem development. Since IAA3/SHY2 was downregulated in both the MP-silenced Arabidopsis stem array and the poplar stem array, the suggestion that MP may be involved (whether directly or not) in wood formation, is supported.  A potential model would have IAA3/SHY2  downstream from MP in the auxin-wood signaling pathway. The other gene of particular interest that was downregulated in the poplar array was ARF7/NPH4. As this gene is known to share some functions with MP (Hardtke et al., 2004), its downregulation along the course of poplar stem development offers indirect support for the idea that MP is involved in that process.  The data presented in the three Results chapters, while providing new biological insights, may have its greatest impact in suggesting new genes for  153  which functional analysis, both in Arabidopsis and poplar. Several of these have been mentioned in the various chapters. These studies could range from reverse genetics (T-DNA knockouts in Arabidopsis, RNAi lines in poplar; targeted overexpression in both plants), to studies on protein-protein interaction, such as two-hybrid assays. Furthermore, conservation of genes and apparent gene function between Arabidopsis and Populus suggest that comparative studies will be fruitful, e.g. expressing poplar genes in Arabidopsis and vice versa, to gain insights into the degree to which gene function and regulatory pathways are conserved in the two plants, which represent extremes of angiosperm life history.  154  References Arabidopsis Genome Initiative Aida, M., Vernoux, T., Furutani, M., Traas, J., and Tasaka, M. (2002). Roles of PIN-FORMED1 and MONOPTEROS in pattern formation of the apical region of the Arabidopsis embryo. Development 129, 3965-3974. Allina, S.M., Pri-Hadash, A., Theilmann, D.A., Ellis, B.E., and Douglas, C.J. (1998). 4-Coumarate:coenzyme A ligase in hybrid poplar. Properties of native enzymes, cDNA cloning, and analysis of recombinant enzymes. Plant Physiol 116, 743-754. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990). Basic local alignment search tool. J Mol Biol 215, 403-410. Andersson-Gunneras, S., Hellgren, J.M., Bjorklund, S., Regan, S., Moritz, T., and Sundberg, B. (2003). Asymmetric expression of a poplar ACC oxidase controls ethylene production during gravitational induction of tension wood. Plant J 34, 339-349. Baulcombe, D. (2005). RNA silencing. Trends Biochem Sci 30, 290-293. Berleth, T., and Jurgens, G. (1993). The role of the monopteros gene in organising basal development in the Arabidopsis embryo. Development 118, 575-587. Berleth, T., and Mattsson, J. (2000). Vascular development: tracing signals along veins. Curr Opin Plant Biol 3, 406-411. Berleth, T., Mattsson, J., and Hardtke, C.S. (2000). Vascular continuity and auxin signals. Trends Plant Sci 5, 387-393. Berleth, T., Krogan, N.T., and Scarpella, E. (2004). Auxin signals-turning genes on and turning cells around. Curr Opin Plant Biol 7, 553563. Berleth T, J.G. (1993). The role of the monopteros gene in organising basal development in the Arabidopsis embryo. Development 118, 575-587.  155  Bhalerao, R.P., and Bennett, M.J. (2003). The case for morphogens in plants. Nat Cell Biol 5, 939-943.  Birnbaum, K., Jung, J.W., Wang, J.Y., Lambert, G.M., Hirst, J.A., Galbraith, D.W., and Benfey, P.N. (2005). Cell type-specific expression profiling in plants via cell sorting of protoplasts from fluorescent reporter lines. Nat Methods 2, 615-619. Blakeslee, J.J., Peer, W.A., and Murphy, A.S. (2005). Auxin transport. Curr Opin Plant Biol 8, 494-500. Blilou, I., Xu, J., Wildwater, M., Willemsen, V., Paponov, I., Friml, J., Heidstra, R., Aida, M., Palme, K., and Scheres, B. (2005). The PIN auxin efflux facilitator network controls growth and patterning in Arabidopsis roots. Nature 433, 39-44. Bradshaw, H. (1996). Molecular genetics of Populus. In Biology of Populus and its implication for management and conservation, H.B.J. RF Stettler, PE Heilman, and TM Hinckley, ed (Ottawa, ON: NRC Research Press. Brown, D.M., Zeef, L.A., Ellis, J., Goodacre, R., and Turner, S.R. (2005). Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell 17, 2281-2295. Chaffey, N., Cholewa, E., Regan, S., and Sundberg, B. (2002). Secondary xylem development in Arabidopsis: a model for wood formation. Physiol Plant 114, 594-600. Chapple, C.C., Vogt, T., Ellis, B.E., and Somerville, C.R. (1992). An Arabidopsis mutant defective in the general phenylpropanoid pathway. Plant Cell 4, 1413-1424. Clough, S.J., and Bent, A.F. (1998). Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16, 735-743.  156  Cohen, J.D., Slovin, J.P., and Hendrickson, A.M. (2003). Two genetically discrete pathways convert tryptophan to auxin: more redundancy in auxin biosynthesis. Trends Plant Sci 8, 197-199. Cosgrove, D.J. (1993). Wall extensibility: its nature, measurement and relationship to plant cell growth. New Phytol 124, 1-23. Cukovic, D., Ehlting, J., VanZiffle, J.A., and Douglas, C.J. (2001). Structure and evolution of 4-coumarate:coenzyme A ligase (4CL) gene families. Biol Chem 382, 645-654. Darwin, C., and Darwin, F. (1881). The power of movement in plants. (New York: D. Appleton and Company). Demura, T., Tashiro, G., Horiguchi, G., Kishimoto, N., Kubo, M., Matsuok a, N., Minami, A., Nagata-Hiwatashi, M., Nakamura, K., Okamura, Y., Sassa, N., Suzuki, S., Yazaki, J., Kikuchi, S., and Fukuda, H. (2002). Visualization by comprehensive microarray analysis of gene expression programs during transdifferentiation of mesophyll cells into xylem cells. Proc Natl Acad Sci U S A 99, 15794-15799. Dharmasiri, N., and Estelle, M. (2004). Auxin signaling and regulated protein degradation. Trends Plant Sci 9, 302-308. Dharmasiri, N., Dharmasiri, S., and Estelle, M. (2005a). The F-box protein TIR1 is an auxin receptor. Nature 435, 441-445. Dharmasiri, N., Dharmasiri, S., Weijers, D., Lechner, E., Yamada, M., Hobbie, L., Ehrismann, J.S., Jurgens, G., and Estelle, M. (2005b). Plant development is regulated by a family of auxin receptor F box proteins. Dev Cell 9, 109-119. Digby, J., and Wareing, P. (1966). The effect of applied growth hormones on cambial division and the differentiation of the cambial derivates. Annals of Botany 30, 539-548. Douglas, C.J., and Ehlting, J. (2005). Arabidopsis thaliana full genome longmer microarrays: a powerful gene discovery tool for agriculture and forestry. Transgenic Res 14, 551-561.  157  Ehlting, J., Buttner, D., Wang, Q., Douglas, C.J., Somssich, I.E., and Kombrink, E. (1999). Three 4-coumarate:coenzyme A ligases in Arabidopsis thaliana represent two evolutionarily divergent classes in angiosperms. Plant J 19, 9-20.  Ehlting, J., Mattheus, N., Aeschliman, D.S., Li, E., Hamberger, B., Cullis, I.F., Zhuang, J., Kaneda, M., Mansfield, S.D., Samuels, L., Ritland, K., Ellis, B.E., Bohlmann, J., and Douglas, C.J. (2005). Global transcript profiling of primary stems from Arabidopsis thaliana identifies candidate genes for missing links in lignin biosynthesis and transcriptional regulators of fiber differentiation. Plant J 42, 618-640. Ellis, C.M., Nagpal, P., Young, J.C., Hagen, G., Guilfoyle, T.J., and Reed, J.W. (2005). AUXIN RESPONSE FACTOR1 and AUXIN RESPONSE FACTOR2 regulate senescence and floral organ abscission in Arabidopsis thaliana. Development 132, 4563-4574. Eriksson, M.E., Israelsson, M., Olsson, O., and Moritz, T. (2000). Increased gibberellin biosynthesis in transgenic trees promotes growth, biomass production and xylem fiber length. Nat Biotechnol 18, 784788. Fischer, C., Speth, V., Fleig-Eberenz, S., and Neuhaus, G. (1997). Induction of Zygotic Polyembryos in Wheat: Influence of Auxin Polar Transport. Plant Cell 9, 1767-1780. Fosket, D. (1994). Plant Growth and Development. A Molecular Approach. (New York: Academic Press). Friml, J. (2003). Auxin transport - shaping the plant. Curr Opin Plant Biol 6, 7-12. Frohlich, M.W. (2003). An evolutionary scenario for the origin of flowers. Nat Rev Genet 4, 559-566. Fukuda, H. (2004). Signals that control plant vascular cell differentiation. Nat Rev Mol Cell Biol 5, 379-391.  158  Fukuda H, K.A. (1980). Establishment of an experimental system for the study of tracheary element differentiation from single cells isolated from the mesophyll cells of Zinnia elegans. Plant Physiol 65, 57-60.  Gensel, P.G., Edwards, D., and International Organization of Paleobotany. Conference. (2001). Plants invade the land: evolutionary and environmental perspectives. (New York: Columbia University Press). Gerber HP, S.K., Georgiev O, Hofferer M, Hug M, Rusconi S, Schaffner W. (1994). Transcriptional activation modulated by homopolymeric glutamine and proline stretches. Science 263, 808-811. Gilchrist, E.J., Haughn, G.W., Ying, C.C., Otto, S.P., Zhuang, J., Cheung, D., Hamberger, B., Aboutorabi, F., Kalynyak, T., Johnson, L., Bohlmann, J., Ellis, B.E., Douglas, C.J., and Cronk, Q.C. (2006). Use of Ecotilling as an efficient SNP discovery tool to survey genetic variation in wild populations of Populus trichocarpa. Mol Ecol 15, 1367-1378. Gray, W.M., Kepinski, S., Rouse, D., Leyser, O., and Estelle, M. (2001). Auxin regulates SCF(TIR1)-dependent degradation of AUX/IAA proteins. Nature 414, 271-276. Greve, K., La Cour, T., Jensen, M.K., Poulsen, F.M., and Skriver, K. (2003). Interactions between plant RING-H2 and plant-specific NAC (NAM/ATAF1/2/CUC2) proteins: RING-H2 molecular specificity and cellular localization. Biochem J 371, 97-108. Groover, A.T. (2005). What genes make a tree a tree? Trends Plant Sci 10, 210-214. Guilfoyle, T., Hagen, G., Ulmasov, T., and Murfett, J. (1998). How does auxin turn on genes? Plant Physiol 118 , 341-347.  159  Hagen, G., and Guilfoyle, T. (2002). Auxin-responsive gene expression: genes, promoters and regulatory factors. Plant Mol Biol 49, 373-385. Hahlbrock K., and Scheel D. (1989). Physiology and molecular biology of phenylpropanoid metabolism. Annual Review of Plant Physiology and Plant Molecular Biology 40, 347-369. Hamann, T., Benkova, E., Baurle, I., Kientz, M., and Jurgens, G. (2002). The Arabidopsis BODENLOS gene encodes an auxin response protein inhibiting MONOPTEROS-mediated embryo patterning. Genes Dev 16, 1610-1615. Hamet, P., and Tremblay, J. (2004). Novel genetic approaches to the resolution of complex diseases. Cas Lek Cesk 143, 651-656; discussion 656. Han, K.H., Gordon, M.P., and Strauss, S.H. (1996). Cellular and molecular biology of Agrobacterium-mediated transformation of plants and its application to genetic transformation of Populus. In Biology of Populus and its implication for management and conservation., H.B.J. RF Stettler, PE Heilman, and TM Hinckley, ed (Ottawa, ON: NRC Research Press. Harding, S.A., Jiang, H., Jeong, M.L., Casado, F.L., Lin, H.W., and Tsai, C.J. (2005). Functional genomics analysis of foliar condensed tannin and phenolic glycoside regulation in natural cottonwood hybrids. Tree Physiol 25, 1475-1486. Hardtke, C.S., and Berleth, T. (1998). The Arabidopsis gene MONOPTEROS encodes a transcription factor mediating embryo axis formation and vascular development. Embo J 17, 1405-1411. Hardtke, C.S., Ckurshumova, W., Vidaurre, D.P., Singh, S.A., Stamatiou, G., Tiwari, S.B., Hagen, G., Guilfoyle, T.J., and Berleth, T. (2004). Overlapping and non-redundant functions of the Arabidopsis auxin response factors MONOPTEROS and NONPHOTOTROPIC HYPOCOTYL 4. Development 131, 1089-1100. Heisler, M.G., Ohno, C., Das, P., Sieber, P., Reddy, G.V., Long, J.A., and Meyerowitz, E.M. (2005). Patterns of Auxin Transport  160  and Gene Expression during Primordium Development Revealed by Live Imaging of the Arabidopsis Inflorescence Meristem. Curr Biol 15, 18991911.  Hertzberg, M., Aspeborg, H., Schrader, J., Andersson, A., Erlandsson, R., Blomqvist, K., Bhalerao, R., Uhlen, M., Teeri, T.T., Lundeberg, J., Sundberg, B., Nilsson, P., and Sandberg, G. (2001). A transcriptional roadmap to wood formation. Proc Natl Acad Sci U S A 9 8, 14732-14737. Hoheisel, J.D. (2006). Microarray technology: beyond transcript profiling and genotype analysis. Nat Rev Genet 7, 200-210. Horiguchi, G. (2004). RNA silencing in plants: a shortcut to functional analysis. Differentiation 72, 65-73. Hu,  W.J., Kawaoka, A., Tsai, C.J., Lung, J., Osakabe, K., Ebinuma, H., and Chiang, V.L. (1998). Compartmentalized expression of two structurally and functionally distinct 4-coumarate:CoA ligase genes in aspen (Populus tremuloides). Proc Natl Acad Sci U S A 95, 5407-5412.  Initiative, A.G. (2000). Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796-815. Jacobs, W. (1952). The role of auxin in differentiation of xylem around a wound. The American Journal of Botany 39, 301-309. Jenik, P.D., and Barton, M.K. (2005). Surge and destroy: the role of auxin in plant embryogenesis. Development 132, 3577-3585. Job, D. (2002). Plant biotechnology in agriculture. Biochimie 84, 11051110. Johnson, L., Rogers, S., Douglas, C.J., and Potter, S. (2004). Using the Model Plant Arabidopsis for gene discovery in Poplars. Paprican University Report 842, 1-13.  161  Jones, A.M. (1992). What remains of the Cholodny-Went theory? Assymetric redistribution of auxin need only occur over the distance of one cell width. Plant Cell Environ 15, 775-776. Jones, A.M. (1998). Auxin transport: down and out and up again. Science 282, 2201-2203. Jurgens, G., Torres Ruiz, R.A., and Berleth, T. (1994). Embryonic pattern formation in flowering plants. Annu Rev Genet 28, 351-371. Kao, Y.Y., Harding, S.A., and Tsai, C.J. (2002). Differential expression of two distinct phenylalanine ammonia-lyase genes in condensed tanninaccumulating and lignifying cells of quaking aspen. Plant Physiol 1 30, 796-807. Kaufman, L.R., P.J. (1990). Finding Groups in Data: An Introduction to Cluster Analysis. (New York: Wiley). Kawa-Miszc zak, L., Wegrzynowicz, E., and Saniewski, M. (1992). THE EFFECT OF REMOVAL OF ROOTS AND APPLICATION OF PLANT GROWTH REGULATORS ON TULIP SHOOT GROWTH. In VI International Symposium on Flower Bulbs, J.C.M.B. M. Saniewski, W. Bogatko, ed (Skierniewice, Poland: ISHS Acta Horticulturae). Kenrick, P. (2000). The relationships of vascular plants. Philos Trans R Soc Lond B Biol Sci 355, 847-855. Kepinski, S., and Leyser, O. (2005). Plant development: auxin in loops. Curr Biol 15, R208-210. Kessler, S., and Sinha, N. (2004). Shaping up: the genetic control of leaf shape. Curr Opin Plant Biol 7, 65-72. Ko, J.H., Han, K.H., Park, S., and Yang, J. (2004). Plant body weightinduced secondary growth in Arabidopsis and its transcription phenotype revealed by whole-transcriptome profiling. Plant Physiol 135, 1069-1083.  162  Kolosova, N., Miller, B., Ralph, S., Ellis, B.E., Douglas, C., Ritland, K., and Bohlmann, J. (2004). Isolation of high-quality RNA from gymnosperm and angiosperm trees. Biotechniques 36, 821-824. Kramer, E.M. (2004). PIN and AUX/LAX proteins: their role in auxin accumulation. Trends Plant Sci 9, 578-582. Lagercrantz, U., and Lydiate, D.J. (1996). Comparative genome mapping in Brassica. Genetics 144, 1903-1910. Larsen, P. (1994). The vascular cambium. (Berlin: Springer-Verlag). Larson, P.R. (1994). The vascular cambium: development and structure. (Berlin; New York: Springer-Verlag). Levesque, M.P., and Benfey, P.N. (2004). Systems biology. Current Biology 14, R179. Leyser, O. (2005a). Auxin distribution and plant pattern formation: how many angels can dance on the point of PIN? Cell 12 1, 819-822. Leyser, O. (2005b). The fall and rise of apical dominance. Curr Opin Genet Dev 15, 468-471. Leyser, O., and Berleth, T. (1999). A molecular basis for auxin action. Semin Cell Dev Biol 10, 131-137. Liscum, E., and Reed, J.W. (2002). Genetics of Aux/IAA and ARF action in plant growth and development. Plant Mol Biol 49, 387-400. Lybeer, B., Koch, G., J, V.A.N.A., and Goetghebeur, P. (2006). Lignification and cell wall thickening in nodes of Phyllostachys viridiglaucescens and Phyllostachys nigra. Ann Bot (Lond) 97, 529-539. Mager, W.H., and Winderickx, J. (2005). Yeast as a model for medical and medicinal research. Trends Pharmacol Sci 26, 265-273. Mattsson, J., Sung, Z.R., and Berleth, T. (1999). Responses of plant vascular systems to auxin transport inhibition. Development 126, 29792991.  163  Mattsson, J., Ckurshumova, W., and Berleth, T. (2003). Auxin signaling in Arabidopsis leaf vascular development. Plant Physiol 131, 1327-1339. McCouch, S.R. (2001). Genomics and synteny. Plant Physiol 125, 152-155. Mellerowicz, E.J., Baucher, M., Sundberg, B., and Boerjan, W. (2001). Unravelling cell wall formation in the woody dicot stem. Plant Mol Biol 47, 239-274. Meyer-Berthaud, B. (2000). [The first trees. The Archaeopteris model]. J Soc Biol 194, 65-70. Mockaitis, K., and Estelle, M. (2004). Integrating transcriptional controls for plant cell expansion. Genome Biol 5, 245. Muday, G.K. (2000). Maintenance of asymmetric cellular localization of an auxin transport protein through interaction with the actin cytoskeleton. J Plant Growth Regul 19, 385-396. Muday, G.K. (2001). Auxins and tropisms. J Plant Growth Regul 20, 226243. Muday, G.K., and DeLong, A. (2001). Polar auxin transport: controlling where and how much. Trends Plant Sci 6, 535-542. Mudge, J., Cannon, S.B., Kalo, P., Oldroyd, G.E., Roe, B.A., Town, C.D., and Young, N.D. (2005). Highly syntenic regions in the genomes of soybean, Medicago truncatula, and Arabidopsis thaliana. BMC Plant Biol 5, 15. Nelson, T., and Dengler, N. (1997). Leaf Vascular Pattern Formation. Plant Cell 9, 1121-1135. Nemhauser, J.L., and Chory, J. (2005). A new FronTIR in targeted protein degradation and plant development. Cell 12 1, 970-972. Nieminen, K.M., Kauppinen, L., and Helariutta, Y. (2004). A weed for wood? Arabidopsis as a genetic model for xylem development. Plant Physiol 135, 653-659.  164  Nilsson, O., Moritz, T., Sundberg, B., Sandberg, G., and Olsson, O. (1996). Expression of the Agrobacterium rhizogenes rolC Gene in a Deciduous Forest Tree Alters Growth and Development and Leads to Stem Fasciation. Plant Physiol 112, 493-502. Oh, S., Park, S, Han, K-H. (2003). Transcriptional regulation of secondary growth in Arabidopsis thaliana. J Exp Bot 5 4, 2709–2722.  Ohashi-Ito, K., and Fukuda, H. (2003). HD-zip III homeobox genes that include a novel member, ZeHB-13 (Zinnia)/ATHB-15 (Arabidopsis), are involved in procambium and xylem cell differentiation. Plant Cell Physiol 44, 1350-1358. Ohashi-Ito, K., Demura, T., and Fukuda, H. (2002). Promotion of transcript accumulation of novel Zinnia immature xylem-specific HD-Zip III homeobox genes by brassinosteroids. Plant Cell Physiol 43, 11461153. Okushima, Y., Overvoorde, P.J., Arima, K., Alonso, J.M., Chan, A., Chang, C., Ecker, J.R., Hughes, B., Lui, A., Nguyen, D., Onodera, C., Quach, H., Smith, A., Yu, G., and Theologis, A. (2005). Functional genomic analysis of the AUXIN RESPONSE FACTOR gene family members in Arabidopsis thaliana: unique and overlapping functions of ARF7 and ARF19. Plant Cell 17, 444-463. Osborne, D.J. (1959). Control of leaf senescence by auxins. Nature 183, 1459-1460. Owen, H.R., and Miller, A.R. (1992). An examination and correction of plant tissue culture basal medium formulations. Plant Cell, Tissue and Organ Culture 28, 147-150. Palme, K., and Galweiler, L. (1999). PIN-pointing the molecular basis of auxin transport. Curr Opin Plant Biol 2, 375-381. Parry, G., Delbarre, A., Marchant, A., Swarup, R., Napier, R., Perrot-Rechenmann, C., and Bennett, M.J. (2001). Novel auxin transport inhibitors phenocopy the auxin influx carrier mutation aux1. Plant J 25, 399-406.  165  Pearson, L.C. (1995). The diversity and evolution of plants. (Boca Raton: CRC Press). Pennazio, S. (2002). The discovery of the chemical nature of the plant hormone auxin. Riv Biol 95, 289-308.  Persson, S., Wei, H., Milne, J., Page, G.P., and Somerville, C.R. (2005). Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. Proc Natl Acad Sci U S A 102, 8633-8638. Przemeck, G.K., Mattsson, J., Hardtke, C.S., Sung, Z.R., and Berleth, T. (1996). Studies on the role of the Arabidopsis gene MONOPTEROS in vascular development and plant cell axialization. Planta 200, 229-237. Raes, J., Rohde, A., Christensen, J.H., Van de Peer, Y., and Boerjan, W. (2003). Genome-wide characterization of the lignification toolbox in Arabidopsis. Plant Physiol 133, 1051-1071. Ralph, S., Oddy, C., Cooper, D., Yueh, H., Jancsik, S., Kolosova, N., Philippe, R.N., Aeschliman, D., White, R., Huber, D., Ritland, C.E., Benoit, F., Rigby, T., Nantel, A., Butterfield, Y.S., Kirkpatrick, R., Chun, E., Liu, J., Palmquist, D., Wynhoven, B., Stott, J., Yang, G., Barber, S., Holt, R.A., Siddiqui, A., Jones, S.J., Marra, M.A., Ellis, B.E., Douglas, C.J., Ritland, K., and Bohlmann, J. (2006). Genomics of hybrid poplar (Populus trichocarpax deltoides) interacting with forest tent caterpillars (Malacosoma disstria): normalized and full-length cDNA libraries, expressed sequence tags, and a cDNA microarray for the study of insect-induced defences in poplar. Mol Ecol 15, 1275-1297. Ralph, S., et al. (2005). Genomics of hybrid poplar (Populus trichocarpa x deltoides) interacting with forest tent caterpillars (Malacosoma disstria): Normalized and full-length cDNA libraries, expressed sequence tags (ESTs), and cDNA microarrays for the study of insect-induced defenses in poplar. Molecular Ecology, in press.  166  Reinhardt, D. (2003). Vascular patterning: more than just auxin? Curr Biol 13, R485-487. Reinhardt, D., Pesce, E.R., Stieger, P., Mandel, T., Baltensperger, K., Bennett, M., Traas, J., Friml, J., and Kuhlemeier, C. (2003). Regulation of phyllotaxis by polar auxin transport. Nature 426, 255-260. Renzaglia, K.S., Duff, R.J.T., Nickrent, D.L., and Garbary, D.J. (2000). Vegetative and reproductive innovations of early land plants: implications for a unified phylogeny. Philos Trans R Soc Lond B Biol Sci 355, 769-793. Roberts, K., and McCann, M.C. (2000). Xylogenesis: the birth of a corpse. Curr Opin Plant Biol 3, 517-522. Ruegger, M., Dewey, E., Gray, W.M., Hobbie, L., Turner, J., and Estelle, M. (1998). The TIR1 protein of Arabidopsis functions in auxin response and is related to human SKP2 and yeast grr1p. Genes Dev 12, 198-207. Sachs, T. (1981). The control of the patterned differentiation of vascular tissues. Advanced Botanical Research 9, 151-262. Sachs, T. (1986). Cellular interactions in tissue and organ development. Symp Soc Exp Biol 40, 181-210. Sachs, T. (1991). Cell polarity and tissue patterning in plants. Development Supplement 91, 83-93. Sambrook, J., Frisch, E., and Maniatis, T. (1989). Molecular Cloning. A Laboratorv Manual. (Cold. I Spring Hirbor, NY: Cold Spring Harbor Laboratorv Press). 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.  167  Savidge, R.A. (1983). The role of plant hormones in higher plant cellular differentiation. I. A critique. Histochem J 15, 437-445. Scarpella, E., Francis, P., and Berleth, T. (2004). Stage-specific markers define early steps of procambium development in Arabidopsis leaves and correlate termination of vein formation with mesophyll differentiation. Development 131, 3445-3455.  Scarpella, E., Boot, K.J., Rueb, S., and Meijer, A.H. (2002). The procambium specification gene Oshox1 promotes polar auxin transport capacity and reduces its sensitivity toward inhibition. Plant Physiol 130, 1349-1360. Scherer, L.J., and Rossi, J.J. (2003). Approaches for the sequencespecific knockdown of mRNA. Nat Biotechnol 21, 1457-1465. Schmidt, R. (2000). Synteny: recent advances and future prospects. Curr Opin Plant Biol 3, 97-102. Schrader, J., Nilsson, J., Mellerowicz, E., Berglund, A., Nilsson, P., Hertzberg, M., and Sandberg, G. (2004). A high-resolution transcript profile across the wood-forming meristem of poplar identifies potential regulators of cambial stem cell identity. Plant Cell 16, 22782292. Shin, J.T., and Fishman, M.C. (2002). From Zebrafish to human: modular medical models. Annu Rev Genomics Hum Genet 3, 311-340. Sieburth, L.E. (1999). Auxin is required for leaf vein pattern in Arabidopsis. Plant Physiol 121, 1179-1190. Sieburth, L.E., and Deyholos, M.K. (2006). Vascular development: the long and winding road. Curr Opin Plant Biol 9, 48-54. Smith, N.A., Singh, S.P., Wang, M.B., Stoutjesdijk, P.A., Green, A.G., and Waterhouse, P.M. (2000). Total silencing by intronspliced hairpin RNAs. Nature 407, 319-320.  168  Somerville, C., and Ogren, W. (1982). Isolation of photorespiration mutants in Arabidopsis thaliana. In Methods in Chloroplast Biology, M. Edelman, R. Hallick, and N. Chua, eds (Amsterdam: Elsevier Biomedical Press), pp. 129-138. Somerville, C., and Koornneef, M. (2002). A fortunate choice: the history of Arabidopsis as a model plant. Nat Rev Genet 3, 883-889.  Soosaar, J.L., Burch-Smith, T.M., and Dinesh-Kumar, S.P. (2005). Mechanisms of plant resistance to viruses. Nat Rev Microbiol 3, 789798. Steeves, T.A., and Sussex, I.M. (1989). Patterns in plant development. (Cambridge England; New York: Cambridge University Press). Sterky, F., Regan, S., Karlsson, J., Hertzberg, M., Rohde, A., Holmberg, A., Amini, B., Bhalerao, R., Larsson, M., Villarroel, R., Van Montagu, M., Sandberg, G., Olsson, O., Teeri, T.T., Boerjan, W., Gustafsson, P., Uhlen, M., Sundberg, B., and Lundeberg, J. (1998). Gene discovery in the wood-forming tissues of poplar: analysis of 5, 692 expressed sequence tags. Proc Natl Acad Sci U S A 95, 13330-13335. Sterky, F., Bhalerao, R.R., Unneberg, P., Segerman, B., Nilsson, P., Brunner, A.M., Charbonnel-Campaa, L., Lindvall, J.J., Tandre, K., Strauss, S.H., Sundberg, B., Gustafsson, P., Uhlen, M., Bhalerao, R.P., Nilsson, O., Sandberg, G., Karlsson, J., Lundeberg, J., and Jansson, S. (2004). A Populus EST resource for plant functional genomics. Proc Natl Acad Sci U S A 101, 13951-13956. Stirling, B., Koo Yang, Z., Gunter, L., Tuskan, G., and Bradshaw, H. (2003). Comparative sequence analysis between orthologous regions of the Arabidopsis and Populus genomes reveals substantial synteny and microcolinearlity. Can J For Res 33, 2245-2251.  169  Stowe-Evans, E.L., Harper, R.M., Motchoulski, A.V., and Liscum, E. (1998). NPH4, a conditional modulator of auxin-dependent differential growth responses in Arabidopsis. Plant Physiol 118, 1265-1275. Subramaniam, R., Reinold, S., Molitor, E.K., and Dougla s, C.J. (1993). Structure, inheritance, and expression of hybrid poplar (Populus trichocarpa x Populus deltoides) phenylalanine ammonia-lyase genes. Plant Physiol 102, 71-83. Summerbell, D., Smith, J.C., and Maden, M. (1991). The molecular basis of positional information. In Vivo 5, 457-471. Sundberg B, U.C., Tuominen H. (2000). Cambial growth and auxin gradient. In Cell and Molecular Biology of Wood Formation, J.B. R Savidge, R Napier, ed (Oxford: BIOS Scientific Publishers Ltd), pp. 169188. Swarup, R., Parry, G., Graham, N., Allen, T., and Bennett, M. (2002). Auxin cross-talk: integration of signalling pathways to control plant development. Plant Mol Biol 49, 411-426. Szweykowsk a-Kulinska, Z., Jarmolowski, A., and Figlerowicz, M. (2003). RNA interference and its role in the regulation of eucaryotic gene expression. Acta Biochim Pol 50, 217-229. Tabata, T., and Takei, Y. (2004). Morphogens, their identification and regulation. Development 131, 703-712. Taiz, L., and Zeiger, E. (2002). Plant physiology. (Sunderland, Mass.: Sinauer Associates). Tian, Q., Uhlir, N.J., and Reed, J.W. (2002). Arabidopsis SHY2/IAA3 inhibits auxin-regulated gene expression. Plant Cell 14, 301-319. Tiwari, S.B., Hagen, G., and Guilfoyle, T.J. (2004). Aux/IAA proteins contain a potent transcriptional repression domain. Plant Cell 16, 533543. Tuominen, H., Puech, L., Fink, S., and Sundberg, B. (1997). A Radial Concentration Gradient of Indole-3-Acetic Acid Is Related to Secondary Xylem Development in Hybrid Aspen. Plant Physiol 115, 577-585.  170  Tuominen, H., Sitbon, F., Jacobsson, C., Sandberg, G., Olsson, O., and Sundberg, B. (1995). Altered Growth and Wood Characteristics in Transgenic Hybrid Aspen Expressing Agrobacterium tumefaciens TDNA Indoleacetic Acid-Biosynthetic Genes. Plant Physiol 109, 11791189. Tuskan, G. (2006). The genome of black cottonwood, Populus trichocarpa. In Preparation.  Uggla, C., Moritz, T., Sandberg, G., and Sundberg, B. (1996). Auxin as a positional signal in pattern formation in plants. Proc Natl Acad Sci U S A 9 3, 9282-9286. Ulmasov, T., Hagen, G., and Guilfoyle, T.J. (1999). Activation and repression of transcription by auxin-response factors. Proc Natl Acad Sci U S A 96, 5844-5849. Ulmasov, T., Murfett, J., Hagen, G., and Guilfoyle, T.J. (1997). Aux/IAA proteins repress expression of reporter genes containing natural and highly active synthetic auxin response elements. Plant Cell 9, 1963-1971. Waterhouse, P.M., and Helliwell, C.A. (2003). Exploring plant genomes by RNA-induced gene silencing. Nat Rev Genet 4, 29-38. Weijers, D., and Jurgens, G. (2005). Auxin and embryo axis formation: the ends in sight? Curr Opin Plant Biol 8, 32-37. Wellman, C.H. (2004). Palaeoecology and palaeophytogeography of the rhynie chert plants: evidence from integrated analysis of in situ and dispersed spores. Proc Biol Sci 271, 985-992. Wellman, C.H., and Gray, J. (2000). The microfossil record of early land plants. Philos Trans R Soc Lond B Biol Sci 355, 717-731; discussion 731-712. Wellman, C.H., Osterloff, P.L., and Mohiuddin, U. (2003). Fragments of the earliest land plants. Nature 425, 282-285.  171  Went, F.W., and Thimann, K.V. (1937). Phytohormones. (New York: The Macmillan Company). Wesley, S.V., Helliwell, C.A., Smith, N.A., Wang, M.B., Rouse, D.T., Liu, Q., Gooding, P.S., Singh, S.P., Abbott, D., Stoutjesdijk, P.A., Robinson, S.P., Gleave, A.P., Green, A.G., and Waterhouse, P.M. (2001). Construct design for efficient, effective and high-throughput gene silencing in plants. Plant J 27, 581590. Xie, Q., Guo, H.S., Dallman, G., Fang, S., Weissman, A.M., and Chua, N.H. (2002). SINAT5 promotes ubiquitin-related degradation of NAC1 to attenuate auxin signals. Nature 41 9, 167-170. Ye, Z.H. (2002). Vascular tissue differentiation and pattern formation in plants. Annu Rev Plant Biol 53, 183-202. Ye, Z.H., Freshour, G., Hahn, M.G., Burk, D.H., and Zhong, R. (2002). Vascular development in Arabidopsis. Int Rev Cytol 220, 225256. Zhang, J.Z. (2003). Overexpression analysis of plant transcription factors. Curr Opin Plant Biol 6, 430-440. Zhang, J.Z., Creelman, R.A., and Zhu, J.K. (2004). From laboratory to field. Using information from Arabidopsis to engineer salt, cold, and drought tolerance in crops. Plant Physiol 13 5, 615-621. Zhong, R., and Ye, Z.H. (2001). Alteration of auxin polar transport in the Arabidopsis ifl1 mutants. Plant Physiol 12 6, 549-563. Zhong, R., Burk, D.H., and Ye, Z.H. (2001). Fibers. A model for studying cell differentiation, cell elongation, and cell wall biosynthesis. Plant Physiol 126, 477-479. Zuo, J., and Chua, N.H. (2000a). Curr Opin Biotechnol 11, 146-151. Zuo,  J., and Chua, N.H. (2000b). Chemical-inducible systems for regulated expression of plant genes. Curr Opin Biotechnol 11, 146151.  172  Zuo, J., Niu, Q.W., and Chua, N.H. (2000). Technical advance: An estrogen receptor-based transactivator XVE mediates highly inducible gene expression in transgenic plants. Plant J 24, 265-273.  173  

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.831.1-0092930/manifest

Comment

Related Items