Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Signal compound specificity in agrobacterium tumefaciens Spencer, Paul Anthony 1988

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

Item Metadata


831-UBC_1988_A6_7 S69.pdf [ 3.21MB ]
JSON: 831-1.0097848.json
JSON-LD: 831-1.0097848-ld.json
RDF/XML (Pretty): 831-1.0097848-rdf.xml
RDF/JSON: 831-1.0097848-rdf.json
Turtle: 831-1.0097848-turtle.txt
N-Triples: 831-1.0097848-rdf-ntriples.txt
Original Record: 831-1.0097848-source.json
Full Text

Full Text

SIGNAL COMPOUND SPECIFICITY IN AGROBACTERIUM TUMEFACIENS By PAUL ANTHONY SPENCER B.Sc, The University of Victoria, 1985  A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in THE FACULTY OF GRADUATE STUDIES (Department of Botany)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA September, 1988 © Paul Anthony Spencer, 1988.  In  presenting  degree freely  at  this  the  available  copying  of  department publication  of  in  partial  fulfilment  University  of  British  Columbia,  for  this or  thesis  reference  thesis by  this  for  his  and  scholarly  or  thesis  for  her  Department  Date  DE-6(3/81)  Columbia  \  f  I further  purposes  gain  the  requirements  I agree  shall  that  agree  may  representatives.  financial  permission.  The University of British 1956 Main Mall Vancouver, Canada V6T 1Y3  study.  of  be  It not  is  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall for  the that  without  make  it  extensive  head  of  my  copying  or  my  written  ABSTRACT Agrobacterium  tumefaciens  ,  a soil-borne gram negative bacterium, is the  causative agent of crown gall disease and one of the most promising vectors for genetic engineering in plants. It is known to respond to the presence of certain plant-derived phenolic compounds by expressing an essential set of genes for virulence (vir) (Bolton  et al., 1986; Stachel et al., 1985-a).  However, only one report has described the  isolation and identification of virulence inducing phytochemicals producecd by a host plant (Stachel  etal.,  1985-a).  These compounds are acetosyringone (AS) and a-  hydroxyacetosyringone (OH-AS), or 3,5-dimethoxy-4-hydroxy-  and a-hydroxy-3,5-  dimethoxy-4-hydroxy-acetophenone, respectively. Since these compounds have never previously been reported from plant tissues and are not likely to be of widespread occurrence, it seemed unlikely that these were the only signal compounds for this wide host range pathogen. In addition, the results of Bolton et al. (1986), who found that a mixture of lower molecular weight phenolics could also induce vir gene expression, raised the question as to exactly which chemical structures could act as vir -inducers. This thesis reports a quantitative re-examination of the results of Bolton et al. (1986), describes more fully the structure-activity specificity of vir -induction in a wide host range (WHR) strain of  A.  tumefaciens  presents results which indicate that WHR  than did Stachel et al. (1985-a), and  Agrobacterium  is capable of detecting  phytochemicals which are ubiquitous or at least widespread amongst susceptible hosts. The relative vir - inducing activities of the lignin precursors coniferyl and sinapyl alcohols, a variety of cinnamic acid derivatives and two chalcones are presented and discussed in terms of the early events in crown gall tumorogenesis and the sophisticated use of Agrobacterium  in Ti plasmid-mediated transformation of plants.  i i  TABLE OF CONTENTS  Abstract Table of Contents  i»  List of Figures  iv  Acknowledgements  v  Introduction  1  Methods  13  Result and Discussion  16  References  42  iii  List of Figures Figure 1: The WHR plasmid pTiA6 and its vir region  6  Figure 2: Strategy for analysis of signal compound inducible pTi genes  8  Figure 3: Structures of the seven phenolic compounds which were examined for vir -induction by Bolton et al. (1986) Figure 4: The structures of acetosyringone and a-hydroxy-acetosyringone  10 11  Figure 5: Transmission electron micrographs of potato tuber tissue 1 hour after inoculation with Agrobacterium  tumefaciens  17  Figure 6: pH curve of activity of 100 uM acetosyringone  19  Figure 7: Activity curves for some of the seven phenolics reported by Bolton etal. (1986) Figure 8: The 16 chemicals used in the present study of structure-activity relationships, arranged into 4 classes Figure 9: The effect of citrate/phosphate buffer on the dose-response curves of a few phenolics  21 22 23  Figure 10: vir -inducing activity of the lignin precursors coniferyl alcohol and sinapyl alcohol  25  Figure 11: vir -induction by sinapic (sinapinic) acid and related structures ....  26  Figure 12: vir -induction by ferulic acid and related structures  28  Figure 13: vir -induction by syringic acid and related structures  29  Figure 14: vir -induction by vanillin and other structures  30  Figure 15: vir -induction by chalcones of guaiacyl and syringyl substitution Figure 16: Electron flow through the vir -inducer structures and the peroxidase reaction which is known to occur in the production of lignin  32 36  Figure 17: The structures nod -inducingflavonoidsrecently identified from host plants  38  Figure 18: Commercial flavonoids used in the analysis of R. leguminosarum nod - induction  39  iv  ACKNOWLEDGEMENTS  I would like to give special thanks to my Graduate Supervisor, tennis partner and friend Dr. G. H. N. Towers who continually supported me and guided me in the fields of phytochemistry and biotechnology, and also to Dr. Eugene W. Nester (University of Washington) who provided the  Agrobacterium  strains used in the course of this research  and discussed freely the developments in his lab. Thanks are also due to Dr. Tony Warren for his support, Dr. Bruce Bohm for sharing his knowledge, and freely giving his advice and samples of flavonoids, chalcones and aurones, Dr. Norman Lewis (Virginia Polytechnical Institute) who supplied the monolignols, Dr. Albert Stoessl who synthesized methyl ferulate specifically for this research, and Dr. Joan MacPherson for her excellent introduction to plant molecular genetics and for her moral and intellectual support. I gratefully acknowledge the numerous and useful discussions concerning this work with my fellow lab workers and graduate students.  The University of British  Columbia provided financial support in the form of a University Graduate Fellowship and Teaching Assistantships, and the research was funded by the Natural Sciences and Engineering Research Council of Canada.  v  INTRODUCTION  Plant-microbe interactions must be characterized initially by a stage in which the microbe detects a susceptible host. The mechanism of detection is on a chemical level, in which the host normally exudes or under certain conditions produces, or already possesses as part of its cell wall or membrane, compounds which act as signals for the microbial pathogen or symbiont. The microbe responds to these signals through spore germination (Fries etal., 1987), directed growth or chemotaxis (Ashby etal., 1987), and expression of genes necessary in subsequent stages of the interaction, for example, expression of genes directing morphogenesis from saprophytic to parasitic stages (Castle and Day, 1984). At present there are few examples of cell-cell signalling between plants and microbes in which the chemicals that induce such responses have been identified. However, in some cases of bacterial-plant interactions involving the Rhizobiaceae, certain signal compounds involved have recently been identified (Stachel et al., 1985-a; Peters et al., 1986;  Redmond et al., 1986;  Firmin et al., 1986;  Sadowsky etal., 1988). The study of signal compounds in relation to the biology of tumefaciens  Agrobacterium  is potentially of great importance because successful gene manipulation in  plants using this microbial vector is absolutely dependent on these chemicals. Agrobacterium  is a gram negative soil bacterium which causes crown gall disease of a  wide variety of dicotyledonous plants (DeCleen and Deley, 1976). Smith and Townsend (1907) originally described the disease and the incitant. Since that time, and particularly recently, a phenomenal amount of research has been conducted on this subject. It is now known that the bacterium causes a neoplastic growth of the plant tissue by passing copies of its T-DNA, a part of its tumor inducing plasmid (pTi), into the host  plant genome (Chilton et al.,  1977; Thomashow et al., 1980; Yadav et al.,  Chilton et al., 1980; Willmitzer et al., 1980; Zambryski et al., 1980).  1980,  No other  procaryotic organism is known to be capable of such a feat of natural genetic engineering. The T-DNA includes genes which encode enzymes of auxin (Schroder et al., 1984; Thomashow etal.,  1984) and cytokinin biosynthesis (Akiyoshi etal.,  1983;  Akiyoshi et al., 1984), and these genes are expressed in the transformed plant cell (Hille et al.,  1984; Willmitzer et al., 1981). Also present in the T-DNA is a locus  (ocs ) conferring on the transformed plant cells the ability to synthesize one or another characteristic amino acids called opines (Nester et al., 1984), which are then catabolized by the bacteria possessing the occ (octopine catabolism) locus. This confers an advantage on the infecting Agrobacterium  strain in that it has the ability to use the  opines as a sole carbon and nitrogen source. In addition, it has been found that the opines stimulate conjugative pTi transfer in Agrobacterium al., 1978).  (Klapwijk et al., 1978; Petit et  How could such a complex system have evolved?  By what route and  mechanism does the transforming DNA reach the plant cell nucleus and become incorporated in the plant genome? These and other questions concerning the biology of Agrobacterium  presently  remain unanswered. Nevertheless, the profound implications of such a system were immediately recognized and they have rapidly been exploited by molecular biologists. A useful vector for genetic engineering in plants may be constructed by replacing certain of the normal T-DNA genes with new genes of interest. Plant cells are then infected with the bacterium containing the modified Ti-plasmid, transformants are selected, and finally plants are regenerated from the transformed cells. Since the modified T-DNA is stably incorporated into the genome of the transformed cell, every cell of the regenerated plant should possesses the new DNA. This system has been successful in a number of cases (Goodman etal., 1987); however, it should be emphasized that it is ultimately dependent on the ability of the bacterium to detect susceptible cells and  express a number of genes essential for successful transformation of the plant genome, it has therefore become of considerable importance that the molecular mechanism by which the bacterium accomplishes this feat be understood in greater detail. Early studies indicated that attachment of the bacteria to the host plant cells occurs in a site-specific manner (Lippincott and Lippincott, 1969; Smith and Hindiey, 1978), and has been considered an essential stage in crown gall tumorogenesis (Lippincott and Lippincott, 1969).  Attachment of agrobacteria to the surfaces of a  diversity of plant cell types has been studied by a variety of methods (Gee et al., 1967; Bogers, 1972; Smith and Hindiey, 1978; Douglas et al., 1982; Matthysse and Gurlitz,1982; Sigee et al., 1982, Draper etal., 1983; Douglas et al., 1985; Graves et al., 1988).  Hasezawa et al. (1983) reported the introduction by endocytosis of  Agrobacterium  spheroplasts into Vinca rosea protoplasts. Agrobacterium  can attach to  suspension culture cells of both monocotyledons and dicotyledons (Matthysse and Gurlitz, 1982; Douglas et al., 1985) and can attach either to both intact cells or plant protoplasts (Matthysse  et al., 1982), but, interestingly, it cannot bind to carrot  suspension culture cells which have been induced to form embryos (Matthysse and Gurlitz,1982). In general, the evidence from several labs suggests that young, developing cell walls (or damaged cell walls and those under repair at wound sites) are prime targets for binding by the bacteria. Both bacterial and plant components for this binding have been examined (Lippincott etal., 1977; Lippincott and Lippincott, 1978; Gurlitz etal., 1987). The plant component appears to be a plant cell wall pectin or protein to which a bacterial lipopolysaccharide binds. A number of these studies show the bacterial cells attached in a nonrandom manner; the cells attach to numerous sites on the cell surface first singly (by one end) and later in clusters.  The elaboration of cellulose microfibrils is thought to firmly bind  the bacteria to one another and onto the host cell surface (Matthysse et al., 1981;  Matthysse, 1983).  In addition, numerous pili are produced (Lacey Sameuls,  unpublished observations). Electron microscopy by a number of groups have revealed that a membranous structure subsequently envelopes the attached bacteria. The number of attachment sites per host cell has been investigated by Gurlitz et al. (1987). It was estimated that there existed approximately 200 sites of attachment per carrot suspension culture ceil. Electron microscopy has revealed that, at least in binding to certain plants tissues, cells within vascular bundles are preferred (Graves et al., 1988). Early electron microscopy suggested that the bacterium could enter the host cell and persist intracellular^ (Gee et al., 1967; Sigee et al.,  1982). One group  conducted autoradiographic studies which seemed to support their model of entry of the bacterium into the vacuole of the host followed by passage of the T-DNA from the bacterium to the host ceil nucleus (Sigee et al., 1982). Despite the existing evidence for the passage of agrobacteria into the vacuoles of the host cells most groups now believe that this is not, in general, part of the route by which T-DNA reaches the plant cell nucleus. The genetics of the early events of the infection process by Agrobacterium  is  rapidly becoming better understood. In addition to the much studied virulence (vir) genes described below, chromosomal genes affecting cellulose synthesis  (eel)  (Mattysse, 1983), bacterial attachment (att) (Robertson etal., 1988) and cellobiase (abg) from Agrobacterium  (Wakarchuk et al., 1988) have been examined.  The  nonattaching mutants (chv AB and an* mutants) reported to date have all been found to be chromosomal mutants. Prior to attachment of Agrobacterium  to the plant cells, the bacterium must  detect susceptible (wounded) host tissue. Detection of susceptible host cells and early stages of tumorogenesis are mainly controlled by a set of pTi genes known as the virulence (vir ) genes (Horsch et al., 1986; Klee et al., 1983).  Stachel and Nester  (1986) investigated the genetic and transcriptional organization of the vir region of  Agrobacterium  tumefaciens  by saturation transposon mutagensis of cloned portions of  the vir region with Tn3::HoHo1. This modified transposable element contains the Bgalactosidase structural gene (lacZ) and is engineered such that both transcriptional and translational gene fusions may be obtained. Recently it was found that the vir genes are expressed upon cocultivation of the bacteria with host plant cells (Stachel et al., 1985b; Stachel et al., 1986), and that some diffusible plant cell factor (or factors) induced this response. Because of their role in the early stages of tumorogenesis, and therefore their central importance in transformation of plant genomes, intense research has been directed to understanding the mechanism involved in vir gene expression and towards identifying the vir gene products. The genetic and transcriptional organization of the vir region contained in the wide host range  Agrobacterium  Ti-plasmid pTiA6 (Figure 1) has been examined by a  number of independent research groups. Two of the vir genes (A and G) are regulatory in nature (Winans era/., 1986; Stachel and Zambryski, 1986-a). Apparently these act in a coordinated manner to transmit to the bacterium information about external conditions; they trigger expression of other vir genes if the external conditions are appropriate,  vir A is also a host range determinant and is thought to be the  environmental sensor of the plant derived inducer molecules (Leroux et al., 1987). At least one more vir locus (vir C) is connected with host range (Hille et al., 1984-b; Hooykaas et al., 1984; Yanofsky et al., 1985), and another (the vir D operon) is now known to encode a site specific endonuclease which recognizes and cleaves the left and right border sequences of T-DNA (Yanofsky et al., 1986). vir B is a region encoding polypeptides, thought to be homologous in function to pilin proteins, which are required for a presumed transfer.  Agrobacterium  -plant cell conjugation event resulting in T-DNA  Figure 1. Typical organization of the vir region of a Ti plasmid (arrows indicate transcriptional orientation of the vir loci) .  Activation of vir gene expression is known to result in the production of multiple, linear, single-stranded T-DNA molecules within the bacterium (Stachel et al., 1987).  However, this is just one of a few potential intermediates in T-DNA  transmission known to occur in A g r o b a c t e r i u m .  Circular forms of T-DNA have been  reported by several groups (Koukolikova-Nicola et al., 1985; Machida et al., 1986; Alt-Moerbe et al., 1986; Yamamoto et al., 1987) and double-stranded cleavages, mediated by the vir D gene product, also occur within the T-DNA borders (Veluthambi et al., 1987). Presumably, one or more of these T-DNA molecules are the elements which are transferred to the plant genome, however they may be transported as a package along with a vir gene or otherwise encoded protein. The vir E2 gene product was recently determined to be a single-stranded-DNA-binding protein that associates with T-DNA (Christie et al.,  1988).  It was proposed that the VirE2 protein is involved in the  processing of T-DNA and in T-strand protection during transfer to the plant cell.  In  summary, the vir loci are the set of genes that, when expressed, enable A g r o b a c t e r i u m to detect susceptible plant tissue, that prepare the bacterium for its interaction with the plant cell, and result in the production of the transforming DNA in a form apparently ready for transmission to the host. Originally, as is still true of some of the vir genes, the actual functions of the vir  loci were unknown. Their existence was known only as a result of transposon  mutagenesis which at these loci result in avi rule nee or altered virulence (Garfinkel and Nester, 1980). Fortunately, in order to monitor the expression of genes of unknown products it is possible to replace the coding region of the unknown gene with the coding region of a "reporter" gene with some easily measurable activity. As was stated above, Stachel et al. (1985-b) prepared a Tn3::/ac Z transposon for the random generation of 6-galactosidase gene fusions and used it to study gene expression in A g r o b a c t e r i u m . This system is shown diagrammatically in Figure 2. In a strain  Figure 2. Strategy for analysis of signal compound inducible pTi genes. The modified transposon Tn3::HoHo1 carrying lac Z is inserted at numerous sites in the vir region on a cloning vehicle and pTi-carrying strains also carrying vir v.lac Z gene fusioncontaining (pSM) plasmids are used in vir -induction bioassays.  carrying a vir ::lac Z gene fusion plasmid, the degree of vir -induction can be determined simply by assaying 13-galactosidase activity. Using this system it was established that, with the exception of the regulatory locus vir A, expression of each of the vir genes can be induced by the presence of certain phenolic compounds (Bolton etal.,  1986). The  regulation of the vir genes of pTiC58 has recently been examined with vir wlux fusions (Rogowsky etal.,  gene  1987).  After it was established that cocultivation of the bacteria with the host plant cells resulted in vir -induction, it was determined that the inducing agent must have a molecular weight below 1000 (Stachel etal.,  1986). Bolton etal.  (1986) found that a  mixture of low molecular weight phenolic compounds (Figure 3) could be used to induce expression of most of the vir genes, however, as will be discussed, quantitative analysis of vir gene induction by each component of the mixture was not reported. Stachel et al. (1985-a) identified  two  active  signal compounds, acetosyringone  and  a-  hyroxyacetosyringone (Figure 4), from a transformed tobacco root culture and from leaf discs. In that report a few other related compounds were assayed at one or more concentrations for their vir -inducing activity.  This comprised a very brief structure-  activity study which yielded useful but limited information about the structural features required to confer activity.  At the concentrations tested, none of these compounds  examined by Stachel et al. (1985-a) displayed the level of activity observed with acetosyringone, but, as will be discussed later, within their results was sufficient information to predict which of the seven phenolics assayed in mixture by Bolton et al. (1986) was significantly active. The reported activity of these lower molecular weight phenolic compounds is at odds with an earlier report of a higher molecular weight and apparently proteinaceous inducer of XhevirC  locus (Okker et al., 1984).  Stachel and Zambryski (1986-b) have referred to the A g r o b a c t e r i u m  -plant  cell interaction as "a novel adaptation of extracellular recognition and DNA conjugation".  10  COOH  COOH  OH OH  OH OH  OH Protocatechuate  COOH  OH  Catechol  8-resorcylate  COOH  COOH OH  HO'  OH OH Gallate  OH OH Pyrogallate  OH p-hydroxy benzoate  Figure 3. Structures of the seven phenolic compounds which, as a mixture, were examined for vir -induction by Bolton etal. (1986).  11  Figure 4. The structures of acetosyringone (AS) and a-hydroxyacetosyringone (HO-AS). These acetophenone vir -inducers were isolated from transformed tobacco root culture conditioned medium by Stachel etal. (1985-a).  In view of the major cytological differences involved, i.e the presence of the plant cell wall and nuclear envelope, the suggestion that T-DNA transfer is a modified version of conjugation is rather difficult to accept.  However, in the Rhizobiaceae chemotaxis  towards aromatic and hydroaromatic compounds has been demonstrated (Parke et al., 1985).  It has been implicit in the reports concerning signal compounds for  Agrobacterium,  that the bacteria are attracted to susceptible plant tissues by following a  concentration gradient of the virulence inducing substances and some results which support this idea were obtained by Ashby etal. (1987). Parke era/.  (1985) have  shown that AS is not a potent chemoattractant, but that certain other phenolics (gallate, B-resorcylate, protocatechuate, p -hydroxybenzoate, and vanillin, i.e. a number of the phenolics examined by Bolton etal., 1986) induced chemotactic behaviour. The exact mechanism by which T-DNA reaches the plant genome from the Ti plasmid remains unknown. Because they remain the only phytochemicals to be identified from host plants, the compounds acetosyringone and a-hydroxyacetosyringone have come to be regarded as the unique chemicals which Agrobacterium  detects in nature and which trigger the  initial events within the bacterium, resulting in tumor formation.  Reports have  appeared in the literature concerning the use of wound exudates from host plants or of acetosyringone to induce virulence of Agrobacterium  and thereby extend the normal host  range (Schafer et al., 1987) or to boost transformation efficiency (Sheikolleslam and Weeks, 1987). However, it has yet to be shown that these acetophenones are the signal compounds produced by all susceptible hosts. In fact, the results reported here indicate that other phytochemicals are likely involved in the induction of virulence in Agrobacterium.  This thesis reports the vir -inducing activity over a range of concentrations of a variety of plant-derived and synthetic phenolic compounds with structures related to that of acetosyringone and presents some new information regarding the structural  features involved in the activation of vir genes. Included are the activities of some cinnamic acid derivatives, chalcones, and of the lignin precursors sinapyl alcohol and coniferyl alcohol. A number of these compounds are known to be of widespread occurrence, and the monolignols ubiquitous, amongst plants. The results are discussed in relation to the biology of A g r o b a c t e r i u m  tumefaciens  and their potential practical  significance in Ti-mediated gene transfer technology.  METHODS  Electron microscopy. 1cm. potato discs (ca. 5mm. in thickness) were inoculated with one drop of an overnight culture (LB broth, 150 rpm., 28° C) of A . tumefaciens  strain B6. Following  1 hr. incubation at room temperature suitably sized pyramidal sections were cut from the discs and immersed in a glutaraldehyde fixative (2% paraformaldehyde/2% glutaraldehyde in 0.05 M cacodylate buffer at pH 7.2) for 3 hrs.  After 3 washes in  cacodylate buffer, the specimens were post-fixed for 12 hrs. at 4° C with potassium dichromate and 2% OSO4 in 0.05 M cacodylate buffer. The ethanolic dehydration series used consisted of 5 min. washes in 30, 50, 70, 90, 95, and twice in 100% ethanol, then in 1:1 ethanohpropylene oxide (P.O.), and finally 100% (P.O.). Infiltration with Polybed 812 was achieved by placing the specimens in 3:1 P.O.:Polybed 812 for 3hrs., 1:1 P.O.:Polybed 812 overnight (O/N), 1:3 P.O.:Polybed 812 (catalyzed) O/N, and finally 100% catalyzed Polybed 812 O/N. The specimens were embedded by polymerizing the catalyzed resin at 60° C O/N. After sectioning on an ultramicrotome and mounting on transmission electron microscopy (TEM) grids, the sections were stained with uranyl acetate (15 min., 4 g in 50 ml MeOH), washed then counter-stained for 7 min. with Reynold's lead citrate, washed in basic water, then deionized water, then allowed to dry.  The specimens were examined with a Zeis model 10A electron microscope at 60kV.  Chemicals. The chemicals used in this study were commercially available (Sigma, Aid rich, Fluka Co.'s), synthesized in our lab, or supplied by others. Dr. Bruce Bohm (University of British Columbia) donated samples of flavonoid glycosides and aglycones, cinnamic acid derivatives, chalcones and aurones. Some of these compounds were plant derived, others were of commercial origin, and a number of them (notably the chalcones) were synthesized by Dr. Bohm. Dr. Norman Lewis (Virginia Polytechnical Institute and State University) supplied the monolignols coniferyl and sinapyl alcohol. Dr. Albert Stoessl (London Research Center, Agriculture Canada) synthesized methyl ferulate specifically for this research. The starting materials were purified ferulic acid and methyl iodide (iodomethane). The purity of each of the compounds used in this study was assessed by TLC on silica gel or polyamide, using organic solvent mixtures to develop them, and short or long wave UV, or H2SO4 spray and charring for detection. Where necessary, chemicals were purified by recrystallization or preparative TLC on silica gel plates.  vir -induction assay. 8- galactosidase activity was assayed (as described below) as a measure of virgene induction in a wide host range strain (A348, a C58 derivative with pTiA6) carrying a vir E::/ac Z gene fusion plasmid (pSM358). pSM358 carries both kanamycin and carbenicillin resistance markers, so A348/pSM358 was grown in the presence of these antibiotics (50 and 100 pg/mL, respectively) to ensure the presence of the plasmid.  Preliminary assays were conducted as described below, with the exceptions that DMSO was not used to dissolve the compounds used, and no citrate-phosphate buffer was added to the MS medium. In this way a rough pH curve of activity was obtained, and the seven phenolics reported by Bolton et al., (1986) were individually screened for activity. For the structure-activity study, the compounds tested were dissolved in DMSO and diluted in citrate-phosphate (Mcllvane) buffered pH 5.70 MS medium (Murashige and Skoog, 1962) to a final concentration of 0.1% DMSO. 100 pL of bacterial cells from an overnight culture of A348/pSM358 (Stachel etal., 1985-b) were inoculated into each 50 mL culture tube and subjected to continuous shaking at 200 RPM and at 28°C for 8 hours to allow for induction of vir E::/ac Z expression. Cell density was determined by measuring absorbance at 600 nm and 1 mL aliquots were removed for 6galactosidase assay essentially as described by Miller (1972). The relatively acidic induction medium was removed and the induced bacteria were resuspended in a neutral assay medium (Z buffer). The bacteria were then treated with 20pL each of 0.05%SDS and CHCI3, and vortexed for 10 seconds to puncture the cells and permit passage of the chromogenic substrate orthonitrophenyl-R-D-galactoside (4mg/mL ONPG, 200u.L). The assays were terminated by addition of 250u.L of 1M Na2C03. The bacteria were then pelleted and the absorbance at 420nm of the supernatant was recorded. The following equation was used to calculate the 6-galactosidase activity in Miller units: 6-galactosidase activity « O.D.420nm x 10  3  O.D.600nm x T where T = assay time in minutes. Each point on the activity curve of a test compound represents the average of the results of each concentration tested in triplicate. Acetosyringone was used as a positive control in each experiment to ensure that the system was functioning correctly.  RESULTS AND DISCUSSION  Electron  microscopy.  The electron microscopy conducted as a part of this thesis (Figure 5 ) yielded no novel results to add to those in the literature. It was hoped that some new data on the early interaction of the bacteria with the plant cell surface might be obtained,  in  particular the route by which T-DNA is passed through the plant cell wall, and that this interaction could be examined in the presence of varying concentrations of vir -inducing compounds. The methods used resulted in good fixation and staining, and may be used in subsequent studies of the early interaction of Agrobacterium  with host cells, but certain  problems became apparent. The host tissue used (potato tuber tissue, with large cells mostly occupied by vacuoles) was not the most appropriate for T E M .  Bacterial cells  were observed in the intercellular spaces, and in close association with the damaged plant cell membrane. However, after numerous sections were examined, it became clear that in order to efficiently examine binding phenomena it would be necessary to establish a means by which the numerous binding sites could be readily observed. There must exist, at least long enough for T-DNA transfer, some very close contact between the bacteria and the host cell, and perhaps this interaction involves pili or other structures which should be observable by transmission electron microscopy. However, because there are a limited number of host cell binding sites available to the bacteria, the number of successful transformation events on a 1cm potato tuber disc is limited, and it was impossible to select in advance the correct areas from the inoculated disc, the probability of observing by T E M the structural features of the T-DNA transfer event was anticipated to be extremely low. Indeed, no structures  F i g u r e 5. T r a n s m i s s i o n e l e c t r o n m i c r o g r a p h s of potato t u b e r t i s s u e 1 h o u r after inoculation with Agrobacterium tumefaciens. A. b a c t e r i a in a n intercellular s p a c e (23,100 X). B. a b a c t e r i u m c o n t a i n e d within a n invagination of a d a m a g e d plant c e l l m e m b r a n e ( 5 0 , 0 0 0 X). c w = plant cell wall; b - b a c t e r i u m ; c m - plant c e l l membrane.  were observed through which T-DNA transfer could occur and this aspect of crown gall tumorogenesis remains unknown. Host species highly susceptible to transformation, and inoculated cells from suspension cultures, in which there may exist significantly greater surface area for binding, may be superior subjects for future cytological research by TEM. Transformation of carrot root disc cambial cells has been reported (Bercetche et al., 1987). The cambial layers of inoculated carrot root discs could be isolated from surrounding tissues for fixation and TEM.  Determination of pH optimum for vir -induction. In order to confirm the pH optimum for vir -induction reported by Stachel et al. (1986) a curve of activity of 100 u.M acetosyringone over a range of pH values was obtained (Figure 6) for the strain A348 containing the v/r E::/ac Z gene fusion plasmid pSM358. The curve obtained matches that of Stachel et al. (1986), and confirms that, under the conditions described, the pH optimum forv/'r-induction is pH 5.1. tempting to conduct all futureWr -induction assays at pH 5.1.  It was  However, plant tissue  cultures are commonly grown on media of pH 5.7, and previously reported research concerning vir gene expression was conducted at pH 5.7, therefore subsequentv/r induction assays were conducted at this sub-optimal pH.  Assay of phenolics reported by Bolton et al. (1986). Quantitative assays of  \hevir  -inducing activity of each of the seven compounds  which were examined as a mixture by Bolton et al. (1986) yielded results which indicated the need for the structure-activity analyses described in this thesis. These preliminary results suggested that only one compound (vanillin) out of the seven phenolics reported by Bolton et al. (1986) significantly induced the vir genes of Agrobacterium.  This result was predicted on the basis of information regarding the  structural features required for activity as reported by Stachel  etal.  (1985-a). The  Figure 6. vir -inducing activity of 100 u.M acetosyringone over a range of pH values. The strain A348/pSM358 was incubated for 10 hrs. in MS medium containing 100pM acetosyringone and at various pH's.  activity curves obtained are shown, along with that of acetosyringone, in Figure 7. The compounds catechol, 6-resorcylic acid, and gallic acid exhibited low, but detectable levels of induction.  Protocatechuic, pyrogallic, and p-hydroxybenzoic acids did not  induce vir expression above background levels.  Structure-activity  analysis of  vir -induction.  The structures of sixteen vir -inducing phenolic compounds were examined for the purpose of obtaining an understanding of the structure-activity relationships of vir -induction.  These compounds may be categorized into four groups (Figure 8): 1.  acetophenones and related structures, 2. monolignols, 3. structures related to cinnamic acid, and 4. chalcone derivatives.  Each structure contains a guaiacyl or syringyl  nucleus, most possess a carbonyl group, and many are common plant-derived compounds of the phenylpropanoid pathway. The results of an initial set of assays indicated the need to include a buffer system in the induction medium. Uncontrolled pH shifts influenced the degree of induction, and a citrate/phosphate (Mcllvane) buffer was found which demonstrably improved the accuracy of the results, reducing variation between replicate assays to less than 10%. The dose-response curves for a few of the compounds in Figure 8, in the presence and absence of citrate/phosphate buffer, are shown in Figure 9. Visual comparison of the curves in Figure 9-A and 9-B reveals the need for the use of a buffer when attempting to analyze structure-activity relationships wherein pH effects are significant.  Without  the buffer replicate assays varied by as much as 20%. It should be noted that the levels of induction for each of the compounds was decreased as a result of the presence of the buffer, but that the dose-response curve shapes remained the same. The buffer system was used in all assays concerning structure-activity relationships. The dose-response curves obtained for these compounds are shown in Figures 10-15 and are discussed below. Each point on the activity curve of a test compound  21  CONCENTRATION (uM)  F i g u r e 7. Activity c u r v e s for s o m e of the s e v e n p h e n o l i c s reported by Bolton et al. (1986). E x c e p t for vanillin, most of the c o m p o u n d s d i d not display significant activity at the concentrations e x a m i n e d .  Ri  1 a  b  c  d  e f  2a b  H H CH CH CH OMe  a b  c d e f  H OMe H CMe CMe CMe  R= OMe  H CH CH OMe CMe CMe  vanillin syringaldehyde acetovanillone acetosyringone syringic acid Me-syringate  coniferyl alcohol  R= H  R1  3  R2  sinapyl alcohol  R2  H H CMe H CH CMe  vanillalacetone ferulic acid sinapic acid Me-ferulate 5-OH-MF Me-sinapate  Chalcones 4 a R= H  2\4\4-(pH) -3-OMe 3  b R= OMe  OH  O  2\4\4-(OH) -3,5-(OMe)2  Figure 8. The 16 chemicals used in the present study of relationships, arranged into 4 classes.  3  structure-activity  2000  CONCENTRATION (uJWI)  Figure 9. The effect of citrate/phosphate (Mcllvane) buffer on the dose-response curves of a few phenolics. 9-A: dose-response curves of acetosyringone (AS), sinapyl alcohol (SINAPYL ALC), coniferyl alcohol (CONIFERYL ALC), and 2',4',4-trihydroxy-3methoxy-chalcone (CHALCONE) in the absence of buffer; 9-B: activity curves in the presence of buffer, and using the same compounds as in 9-A.  represents the average of the results of each concentration tested in triplicate. The buffer system was used to minimize variation due to uncontrollable pH shifts otherwise observed in the results. In this way, standard deviations rarely reached 10%, the average being 4.7% (n=92) for results of 100 Miller units and above. The curve shown for acetosyringone in each of Figures 10-15 represents the average of the results of numerous assays conducted under identical conditions. The activities of the lignin precursors sinapyl alcohol (2b) and coniferyl (2a) alcohol are shown in Figure 10. The activity of coniferyl alcohol approaches that of acetosyringone (1d) while that of sinapyl alcohol is somewhat less, for the greater part of its range approximating that of sinapic acid (3c), which is shown in Figure 11. To our knowledge this is the first report of virulence induction by compounds generally regarded as immediate precursors of lignin (Freudenberg and Neish, 1968). This important result establishes that  Agrobacterium  may be capable of detecting cells which  are undergoing lignin synthesis or cell wall regeneration and thereby target those cells for transformation. This correlates with the requirement of wounded plant tissue for transformation by binding of  A.  tumefaciens.  Agrobacterium  In addition, this result could explain the preferential  to vascular bundles in some plant tissues (Graves  et  al.,  1988), where lignified cells are typically found . The tzs locus, located within the nopaline-type Ti plasmidv/V region, encodes a dimethylallyl transferase which is induced in a manner similar to that of the other vir loci (John and Amasino, 1988). Induction of this locus by acetosyringone was found to be pH-dependent in octopine strains and pH-independent in nopaline strains. Three possible roles (which need not be exclusive) were suggested for this plant-inducible cytokinin production. These are "(i) to condition plant cells to a state in which the transfer of DNA from the bacteria to the plant cell is optimal, (ii) to ensure that plant cells pass through stages of the cell cycle in which T-DNA integration can occur, or  Figure 10. vir -inducing activity of the lignin precursors coniferyl alcohol (CONIFERYL ALC.) and sinapyl alcohol (SINAPYL ALC). The virulence inducing activity of a variety of phenolic compounds are shown in this and Figures 11-15. Following incubation with a compound in aqueous solution, B-galactosidase activity in a strain of A g r o b a c t e r i u m carrying a vir E::/ac Z fusion plasmid (A348/pSM358) was assayed as an indicator of vir gene induction. Each point on the activity curve of a test compound represents the average of the results of each concentration tested in triplicate. To assist in comparing the curves, the activity curve of acetosyringone (AS) is shown with each group of compounds. The curve shown for acetosyringone represents the average of the results of numerous assays conducted under identical conditions.  CONCENTRATION (uM)  Figure 1 1 . vir -induction by sinapic (sinapinic) acid and related structures SINAPATE= sinapic acid methyl ester). See Figure 1 0 for details.  (M  (iii) to stimulate high levels of postintegration T-DNA expression, thus leading to rapid tumor development". It is speculated here that following detection of lignin precursors and other phenylpropanoid metabolites the increased levels of cytokinins, resulting from the plant-induced expression of the tzs locus, induces cell divisions and such changes in the host cell wall as to expose binding sites to the bacteria.  Perhaps after binding, a  bacterial cellulase assists in creating a passageway for T-DNA transfer in the presumed conjugation-like process. Subsequent to passage of T-DNA into the host cytoplasm, the DNA of a plant cell in the process of division would be more easily accessible to T-DNA integration.  This series of events takes into consideration preferential binding to  certain, likely lignified, tissues in vascular bundles, the activity of the monolignols probably both in vir - and tzs - induction, the concept of specific cell wall binding sites and the route by which T-DNA can gain access to host plant cell DNA. vir -induction by ferulic acid (3b)(Figure 12), syringic acid (1e)(Figure 13) and acetovanillone (1c)(Figure 14) was significantly less than that of any of the other compounds tested. The low activity of acetovanillone (of guaiacyi substitution) in comparison with that of acetosyringone (of syringyl substitution) indicates that for acetophenones a syringyl nucleus is more effective at vir -induction than is a guaiacyi nucleus. This supports the results of Stachel et al. (1985-a) who, by using a vir B::/ac Z strain of Agrobacterium, assayed the relative activities of these acetophenones at four concentrations.  Vanillalacetone (3a), which is not a natural product, possesses a  structure similar to that of acetovanillone except that its carbonyl group is separated from the guaiacyi nucleus by a C-C double bond, as is present in the cinnamic acid derivatives.  Its dose-response curve (Figure 14) lies between that of the two  acetophenones, indicating either that the double bond or simply the increased distance between the carbonyl group and the guaiacyi nucleus enhances the activity of the  Figure 12. vir -induction by ferulic acid and related structures (Me-FERULATE= ferulic acid methyl ester; 5-OH,MF= 5-hydroxy ferulic acid methyl ester). See Figure 10 for details.  Figure 13. vir -induction by syringic acid and related structures syringic acid methyl ester). See Figure 10 for details.  (Me-SYRINGATE  Figure 14. vir -induction by vanillin and other structures (VAA= vanillalacetone). See Figure 10 for details.  structure.  However, the former appears to be the case because dihyroferulic acid,  which is saturated at this bond , was found to be inactive. Interestingly, the methyl esters of ferulic (3d), syringic (1f), and sinapic acids (3f) (Figures 12, 13, and 11, respectively) exhibited significantly greater activity than the corresponding free acids. The possible effects of this esterification are discussed below. The ethyl esters tested were less active again (data not shown), perhaps due to some steric hindrance at the bacterial receptor site not evident with the methyl esters. The curves of activity induced by the chalcones (Figure 15) are somewhat different than those of any of the other compounds tested.  2',4',4-Trihydroxy-3-  methoxy-chalcone (4a) displayed its greatest vir -inducing activity at 10 u.M. The maximum levels of induction by all of the other compounds (except syringaldehyde, 1 b) were obtained at the highest concentration tested, 200 u,M. Unlike any of the other compounds, this chalcone was capable of low level vir -induction at 0.1 uM. The curve for 2',4',4-trihydroxy-3,5-dimethoxy-chalcone  (4b) is shifted more to the right than  the other chalcone, closer to that of acetosyringone, and its maximum activity is observed at 50 u.M. Perhaps this is as a result of the syringyl substitution of its Bring, thereby affording a structure more similar to that of acetosyringone. Neither of these chalcones is very soluble in the aqueous medium used, but they do exhibit significantly greater activity at lower concentrations than do any of the simpler phenolics tested. A number of chalcones are known to exhibit biological activity (Dhar, 1981) and it will be interesting to investigate whether chalcones play any part in virinduction in nature. Other compounds tested, most of which possessed guaiacyl or syringyl substitution patterns but which exhibited little or nov/r -inducing activity, include the following:  phloridzin, chrysosplenol-6-C-glucoside, homoeridiodictyol, tricin,  3,5,7,4'-tetrahydroxy-3'-methoxy  flavonol, plicatic acid, conidendron, substituted  Figure 15. vir -induction by chalcones of guaiacyl and syringyl substitution (2\4\4/3,5 CH= 2\4 ,4-trihydroxy-3,5 dimethoxy chalcone; 2\4 ,4/3 CH= 2\4',4trihydroxy-3-methoxy chalcone). See Figure 10 for details. ,  ,  aurones, vanillic acid, vanilloyl methyl ketone, 5-hydroxyvanillin, dihydrodiferulic acid, 5-hydroxyferulic acid, isoferulic acid, gluco-ferulaldehyde, and syringic, ferulic and vanillic acid ethyl esters. Apparently, hydroxylation at the number 5 position of any active compound possessing a guaiacyi nucleus decreases the compound's activity. This appears to be true even of 5-hydroxy methyl ferulate (3e), which retains activity similar to that of methyl ferulate, but reaches its maximum at a concentration of 50 u.M (Figure 12). The lack of activity of 5-hydroxyferulic acid is of interest because it has recently been found as one of a few cell wall bound cinnamic acid derivatives in monocots (Ohashi et al.,  1987).  Although inhibition of vir -induction by a phenolic compound was not  demonstrated during the course of this research, these observations lead one to speculate about the possibility of phenolic vir -inhibitors common to all monocots. This would explain why the monocots are naturally resistant to infection by A g r o b a c t e r i u m and also why preincubation of the bacteria with a known inducer (Schafer et al., 1987) then allows transformation.  In fact, during the preparation of this thesis, inhibitory  compounds were obtained from Zea mays  (E.W. Nester, personal communication) and  the structures of the inhibitory compounds will be elucidated as part of a collaborative project. The structure of the aglycone of gluco-ferulaldehyde meets the putative requirements of an active signal compound (see below) and yet the glycoside itself was found to be inactive. Although only two phenolic glycosides were tested, the results indicate that, during the exposure time of these assays, A g r o b a c t e r i u m vir genes are not induced by such compounds. If the vacuolar phenolics, which must be exuded upon wounding, are a source of vir - inducing phytochemical precursors, then it appears that plant glucosidases must act to yield the effective compound. The activity of other glycosides  such  as  coniferin  glucopyranoside] and isoconiferin  [4-(3-hydroxy-1-propenyl)-2-methoxyphenyl-D[1-(4-hydroxy-3-methoxyphenyl)-propenyl-3-  D-glucopyranoside] remain to be investigated. Asparagus  Coincidental^/, coniferin is found in  , 10th edn. (1983) Merck Rahway), which appears to be one  (Merckjndex  of the few monocots from which A g r o b a c t e r i u m -transformed tissue has been obtained by the usual methods (Hernalsteens etal.,  1984). Recently, results which will appear in a  subsequent thesis were obtained that indicate that glucose esters of vir -inducing phenolics are also effective vir -inducers (data not shown). It is interesting to note that agrobacteria are known to degrade the lignin model compounds a-conidendron and veratrylglycerol-8-coniferyl ether (Subba Rao et al., 1971). In this study, conidendron was found to be inactive. However, this brings up an important point, namely, that A g r o b a c t e r i u m may chemically alter the compounds which induce virulence. Tracer studies should be conducted in order to determine whether this is indeed the case, and if so, then the metabolites should be identified. Dr. Nestef s group has recently initiated such studies using C -labelled sinapyl alcohol. 14  In addition to the inactive compounds listed above, each of the compounds used by Bolton etal.  (1986) was assayed individually, and only vanillin (1a) resulted in any  significant vir -induction (Figure 7 and 14). Vanillin is not a lignin precursor, however it is regarded as a breakdown product from lignin (Freudenberg and Neish, 1968), as is syringaldehyde (1b).  This may indicate that both lignin precursors and  lignin degradation products function in nature as vir -inducers, which in turn suggests that A g r o b a c t e r i u m  is generally attuned to lignin metabolites.  Interestingly, in  comparison with the response induced by vanillin, the remaining compounds (gallic, 8resorcylic, pyrogallic, p-hydroxybenzoic, and protocatechuic acids, and catechol) were essentially inactive. None of these inactive compounds possesses a guaiacyl or syringyl nucleus. It remains to be determined in what manner each of these compounds effects the activity curve of vanillin. The results clearly indicate that, in general, two basic structural features together are required to confer activity upon a compound. With the exception of the  monolignols and the chalcones, these features are: 1) guaiacyi or (conferring enhanced activity) syringyl substitution on a benzene ring, and 2) a carbonyl  group on a  substituent para to the hydroxy substituent on the ring. There are restrictions on the nature of the carbonyl carbon. It may be one or three carbon atoms removed from the ring. However, to confer maximal activity, in the latter case there must be a double bond between the carbonyl carbon and the ring, as is present in the chalcones and cinnamic acid derivatives. Furthermore, the carbonyl group of a free acid is less effective than that of the corresponding ester. Perhaps the activity of esters of phenolic acids could be used to explain the early result (Okker et al., 1984) which indicated the involvement of a proteinaceous inducer. The inducer obtained may have been a protein ester of some phenolic acid.  As was stated above, a recent result, which will be included in a  subsequent thesis, was the finding that a glucose ester of a phenolic acid was an effective vir -inducer (data not shown). Esterification alters the solubility of the compound. In addition, esterification prevents one oxygen of the carboxyl group from forming a partial double bond, thereby rendering the carbonyl group more reactive.  In these  cases, and the case of the aldehydes and chalcones, this carbonyl group forms the terminus of a long conjugated double bond system running from the hydroxyl group and through the ring.  Electron flow through this system, or peroxidase activity (which  results in monolignol dehydrogenation and aroxyl radical formation in the synthesis of lignin), are possible mechanisms by which the structure interacts with the vir A gene product (Figure 16).  The observation that there is a drop in the pH of the induction  medium after incubation with the bacteria (data not shown) is evidence in support of such models, perhaps indicating that the terminal hydroxy group of active compounds lose the hydrogen ion to the medium. Lastly, the presence of a C ring in the more typical fiavonoids tested virtually abolished activity.  H  A.  Receptor  (VirA)  Figure 16. A. Hypothetical electron flow through the vir -inducer structures (example using a chalcone), and B. the dehydrogenation of coniferyl alcohol and the production of aroxyl radicals which is known to occur in the production of lignin.  37  It may be of considerable importance to compare the chemistry of  Agrobacterium  virulence-induction with that of nodulation (nod ) gene induction in the closely related genus Rhizobium.  The nod genes control early events in the formation of nitrogen fixing  nodules on the roots of leguminous plants. Expression of the nod genes by Rhizobium species is essential for the successful nodulation of host plant species, and is induced by host plant-derived flavonoids. The activity of the more typical flavonoids represents a significant difference in the range of structures inducing gene expression in the two genera.  The structure-activity relationships reported in this thesis are clearly  different from those reported for the activation of nod genes in Rhizobium  species  (Peters and Long, 1987). Hydroxylated flavones, isoflavones, or flavanones in nM to pM concentrations induce expression of nod genes (Peters etal., 1986; Redmond etal., 1986; Firmin etal., 1986; Sadowsky etal., 1988). The compounds involved in nod gene induction include the well known flavonoids luteolin, apigenin, inteolin, naringenin, eriodictyol, geraldone, and daidzein.  The  structures of the active flavonoids which have been identified from host plants are shown in Figure 17. Firmin et al. (1986) did not isolate nod -inducing flavonoids from host tissues, but several commercially available flavonoids were examined (Figure 18) in order to obtain results indicating the range of compounds which could induce nod gene expression in R. leguminosarum.  Most Rhizobium  species are highly specific for its  host plant species and each appears to exhibit a high degree of specificity towards its signal compound. The compound which most strongly induces one species of Rhizobium can act as a potent inhibitor of nod-induction in another species of Rhizobium. original strain of Agrobacterium  tumefaciens  The  from which the strain used in this study  was derived exhibits a wide host range (WHR) and, as we have shown in this report, a comparatively much lower degree of signal compound specificity.  Rhizobium trifolii Flavones a. R=H, 7,4'-&hydroxyflavone (DHF) b. R=OMe, 7,4'-dihydroxy3'-methoxvflavone (Geraldone)  HO.  .0,  O  Rhizobium fredii Isoflavone 4',7-dihydroxyisoflavone (Daidzein)  Rhizobium meliloti Flavone 3',4',5,7-tetrahydroxyflavone (luteolin)  Figure 17. The structures nod -inducing flavonoids recently identified from host plants.  39  Rhizobium leguminosarum Flavones a. R!=H, R =OH, R = H  (Apigenin)  b. R ^ O H , R =OH, R = H  (Inteolin)  2  3  2  3  c. R ^ O H , R =H, R = H 2  3  d. R ^ H , R =OH, R =glc 2  3  (Apigenin7-O-glucoside)  Flavanones a. R ^ H ,  OH  o  R =OH 2  (Naringenin)  b. R!=OH, R =OMe (Hesperitin) 2  c. R j O H , R = O H 2  (Eriodictyol)  Figure 18. Commercial flavonoids used in the analysis of R. leguminosarum induction.  nod -  Another potentially important piece to this biochemical puzzle is the finding that some of the very compounds which induce vir genes in Agrobacterium,  including  acetosyringone, strongly inhibit nod gene activation by these flavonoids (Firmin er al., 1986). At higher concentrations most of the vir - inducing phenolics are bacteriostatic even against Agrobacterium against Rhizobium nod -induction.  (data not shown), and presumably they act in this way  species, or they may act more directly by competitive inhibition of  As was proposed at the outset of the present research, cell-cell  signalling must naturally occur in media of considerable chemical complexity, with both inducers and inhibitors present.  Therefore, successful induction nod or vir genes  would require the correct balance between inducer and inhibitor molecules. These factors can be used to explain the artificial extension of the host range of  Agrobacterium  (Schafer etal., 1987), as well as the increased transformation efficiency obtained (Sheikolleslam and Weeks, 1987) by pre-induction of the agrobacteria with inducing exudates or acetosyringone. A number of the vir -inducing compounds reported here are of widespread occurrence amongst dicotyledonous plants.  Indeed, the lignin precursors must be  ubiquitous amongst susceptible hosts. It would therefore be possible to conclude that, if present in the correct concentration, the presence of one or another of these compounds alone would determine whether a given plant is susceptible to infection by Agrobacterium.  However, it is well known that monocots also produce phenolic  compounds, even exuding phenolic acids such as syringic, sinapic and ferulic acid into the rhizosphere from intact roots (Tang and Young, 1982), and yet, with few exceptions (Hernalsteens era/., 1984), monocots lie outside of the natural host range of any strain of Agrobacterium.  Concentration effects, as well as the action of inhibitors of vir -  induction, may be factors in the resistance of monocots to crown gall tumorogenesis.  This limitation of host range remains a significant problem in the use of this organism as a vector for genetic engineering in plants.  Factors other than  inducer/inhibitor phytochemistry may well be involved in the immunity of monocots to transformation by A g r o b a c t e r i u m .  The present results may facilitate research into  such other factors that may exist. It is hoped that clues to a better understanding of the underlying mechanism of host range determination may be obtained through the results presented here.  Using the results presented here as the foundation for subsequent  research, work is already underway on the isolation and identification of phenolics affecting virulence in both monocotyledons and dicotyledons. Perhaps, at least in part, through a sophisticated application of phytochemicals the barriers to Ti-plasmid mediated plant transformation may be overcome.  REFERENCES Akiyoshi, P.E., Klee, H., Amasino, R.M., Nester, E.W., and Gordon, M.P. (1984) T-DNA of Agrobacterium tumefaciens encodes an enzyme of cytokinin biosynthesis. Proc. Natl. Acad. Sci. 81:5994-5998. Akiyoshi, P.E., Monis, R.O., Hing, R., Mischke, B.S., Kosuge, T., Garfinkel, D.J., Gordon, M.P., and Nester, E.W. (1983) Cytokinin/auxin balance in crown gall tumors is regulated by specific loci in the T-DNA. Proc. Natl. Acad. Sci. 80:407-411. Alt-Moerbe, J., Bak, B., and Schroder, J. (1986) A 3.6-kbp segment from the vir region of Ti plasmids contains genes responsible for border sequencedirected production of T-region circles in E. coli. EMBO J. 5:1129-1135. Ashby, A.M., Watson, M.D., and Shaw, C H . (1987) A Ti-plasmid function is responsible for chemotaxis of Agrobacterium tumefaciens towards the plant wound product acetosyringone. Fed. Eur. Micro. Soc. 41:189-192. Bercetche, J., Chriqui, D., Adam, S., and David, C. (1987) Morphogenetic and cellular reorienations induced by Agrobacterium rhizogenes (strains 1855 and 8196) on carrot, pea and tobacco. Plant Science 52: 195-210. Bogers, R.J. 1972. On the interaction of Agrobacterium  tumefaciens  with cells of  Kalanchoe diagramontiana, In, Proceedings of toe third international conference on plant pathogenic bacteria, Wageningen, 14-21 April 1971.  Bolton, G., Nester, E.W., and Gordon, M. (1986) Plant phenolic compounds induce expression of the Agrobacterium tumefaciens loci needed for virulence. Science 232:983-985. Castle, A.J., and Day, A.W. (1984) Isolation and identification of a-tocopherol as an inducer of the parasitic phase of Ustilago violacea. Phytopathology 74:1 184-1200. Chilton, M.-D., Saiki, R.K., Yadav, N., Gordon, M.P., and Quetier, F. (1980) T-DNA from Agrobacterium Ti-plasmid is in the nuclear DNA fraction of the crown gall tumor cells. Proc. Natl. Acad. Sci. 77:4060-4064. Chilton, M.D., Montoya, A.L, Merlo, D.J., Drummond, M.H., Nutter, R., Gordon, M.P., and Nester, E.W. (1977) Stable incorporation of plasmid DNA into higher plantcells: the molecular basis of crown gall tumorogenesis. Cell 11:263271. Christie, P.J., Ward, J.E., Winans, S.C, and Nester, E.W. (1988) The Agrobacterium tumefaciens vir E2 gene product is a single-stranded-DNA-binding protein that associates with T-DNA. J. Bacteriol. 170:2659-2667. DeCleen, M., and Deley, J. (1976) The host range of crown gall. Bot. Rev. 42:389466.  Dhar, D.N. (1981) The Chemistry of Chalcones and Related Compounds, pp. 284. John Wiley and Sons, New York. pp. 285. Douglas, C , Halperin, W., Gordon, M., and Nester, E.W. (1985) Specific attachment of Agrobacterium tumefaciens to bamboo cells in suspension cultures. J. Bacteriol. 161:764-766. Douglas, C.J., Halperin, W., and Nester, E.W. (1982) Agrobacterium tumefaciens mutants affected in attachment to plant cells. J. Bacteriol. 152:12651275. Draper, J., MacKenzie, I.A., Davy, M.R., and Freeman, J.P. (1983) Attachment of Agrobacterium tumefaciens to mechanically isolated Asparagus cells. Plant Sci. Lett. 29:227-236. Firmin, J.I., Wilson, K.E., Rossen, I., and Johnston, A.W.B. (1986) Flavonoid activation of nodulation genes in Rhizobium reversed by other compounds present in plants. Nature 324:90-92. Freudenberg, K., and Neish, A.C. (1968) Constitution and Biosynthesis of Lignin. Molecular Biology Biochemistry and Biophysics 2. pp. 129. SpringerVerlag, New York. Fries, N., Serck-Hassen, K., and Dimberg, L.H. (1987) Abietic acid, an activator of spore germination in ectomycorrhizal species of the genus Suillus (Boletaceae). Exp. My col. 11:360-363. Garfinkel, D.J., and Nester, E.W. (1980) Agrobacterium tumefaciens mutants affected in crown gall tumorogenesis and octopine catabolism. J. Bacteriol. 144:732-743. Gee, M.M., Sun, C.N., and Dwyer, J.D. 1967. An electron microscope study of sunflower crown gall tumor. Protoplasma 6 4 : 195-200. Goodman, R.M., Hauptli, H., Crossway, A., and Knauf, V.C. (1987) Gene transfer in crop improvement. Science 2 3 6 : 48-54. Graves, A.E., Goldman, S.L, Banks, S.W., and Graves, A.C.F. (1988) Scanning electron microscope studies of Agrobacterium tumefaciens attachment to Zea mays, Gladiolus sp., and Triticum aestivum. J. Bacteriol. 170:2395-2400. Gurlitz, R.H.G., Lamb, P.W., and Matthysse, A.G. (1987) Involvement of carrot cell surface proteins in attachment of Agrobacterium tumefaciens . Plant Physiol. 83:564-568. Hasezawa, S., Matsui, C , Nagata, T., and Syono, K. (1983) Cytological study of the introduction of Agrobacterium tumefaciens spheroplasts into Vinca rosea protoplasts. Can. J. Bot. 61:1052-1057. Hernalsteens, J.P., Thia-Toong, L., Schell J, and van Montagu, M. (1984) An Agrobacterium -transformed cell culture from the monocot Asparagus officinalis. EMBO J. 3:3039-3041.  Hille, J., Hoekema, P., Hoojkaas, P., and Shilperoort, R. (1984) Gene organization of the Ti plasmid, p. 287-309. In D.P.S. Verma and T Hohn (ed.), Plant gene research.  New York.  Genes involved in plant-microbe interacations.  Spinger-Verlag,  Hille, J., van Kan, J., and Shilperoort, R. (1984) trans -acting virulence functions of the octopine Ti-plasmid from Agrobacterium tumefaciens. J. Bacteriol. 158:754-756. Hooykaas, P.J.J., Hofker, M., Den Dulk-Ras, H., and Shilperoort, R.A. (1984) A comparison of virulence determinants in an octopine Ti-plasmid, a nopaline Ti-plasmid and an Ri-plasmid by complimentation analysis of Agrobacterium tumefaciens mutants. Plasmid 11:195-205. Horsch, R.B., Klee, H.J., Stachel, S., Winans, S.C, Nester, E.W., Rogers, S.G., and Fraley.R.T. (1986) Analysis of Agrobacterium tumefaciens virulence mutants in leaf discs. Proc. Nat. Acad. Sci. 83:2571-2575. John, M., and Amasino, R.M. (1988) Expression of an Agrobacterium Ti plasmid gene involved in cytokinin biosynthesis is regulated by virulence loci and induced by plant phenolic compounds. J. Bacteriol. 170:790-795. Klapwijk, P.M., Scheuldermon, T., and Schilperoort, R.A. (1978) Coordinated regulation of octopine degradation and conjugative transfer of Ti-plasmids in Agrobacterium tumefaciens : Evidence for a common regulatory gene and separate operons. J. Bacteriol. 136: 775-785. Klee, H.J., White, F.F., Iyer, V.N., Gordon, M.P., Nester, E.W. (1983) Mutational analysis of the virulence region of an Agrobacterium tumefaciens Ti plasmid. J. Bacteriol. 153:878-883. Koukolikova-Nicola, Z., Shillito, R.D., Hohn, B., Wang, K., Van Montagu, M. and Zambryski, P. (1985) Involvement of circular intermediates in the transfer of T-DNA from Agrobacterium tumefaciens to plant cells. Nature 313:191-196. Leroux, B., Yanofsky, M.F., Winans, S.C, Ward, J.E., Ziegler, S.F., and Nester, E.W. (1987) Characterization of the vir A locus of Agrobacterium tumefaciens : a transcriptional regulator and host range determinant. EMBO J. 6:849-856. Lippincott, B.B., and Lippincott, J.A. (1969) Bacterial attachment to a specific wound site as an essential stage in tumor formation by Agrobacterium tumefaciens. J. Bacteriol. 97:620-628. Lippincott, J.A., and Lippincott, B.B. (1978) Ceil walls of crown gall tumors and embryonic plant tissues lack Agrobacterium adherence sites. Science 199:1075-1078. Lippincott, B.B., Whatley, M.H., and Lippincott, J.A. (1977) Tumor induction by Agrobacterium involvesd attachment of the bacterium to a site on the host plant cell wall. Plant Physiology 59:388-390.  Machida, Y., Usami, S., Yamamoto, A., Niwa, Y., and Takebe.l. (1986) Plant-inducible recombination between the 25bp border sequences of T-DNA in Agrobacterium tumefaciens . Mol. Gen. Genet. 204:374-382. Matthysse, A.G. (1983) Role of bacterial cellulose microfibrils in Agrobacterium tumefaciens infection. J. Bacteriol. 154:906-915. Matthysse, A.G., Holmes, K.V., and Guriitz, R.H.G. (1981) Elaboration of cellulose microfibrils by Agrobacterium tumefaciens during attachment to carrot cells. J. Bacteriol. 145:583-595. Matthysse, A.G., Holmes, K.V., and Guriitz, R.H.G. (1982) Binding of Agrobacterium tumefaciens to carrot protoplasts. Physiol. Plant Pathol. 20:27-33. Mattysse, A.G., and Guriitz, R.H. (1982) Plant cell range for attachment of Agrobacterium tumefaciens to tissue culture cells. Physiol. Plant Pathol. 21:381-387. Miller, J.H. (1972) Experiments in Molecular Genetics, Cold Spring Harbour Laboratory Press, New York. Murashige, T., and Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiologia PI. 15:473-497. Nester, E.W., Gordon, M.P., Amasino, R.M., and Yanofsky, M.F. (1984) Crown gall: a molecular and physiological analysis. Annu. Rev. Plant Physiol. 35:387413. Ohashi, H., Yamamamoto, E., Lewis, N., and Towers, G.H.N. (1987) 5-Hydroxyferulic acid in Zea mays and Hordeum vulgare cell walls. Phytochemistry 26:915-916. Okker, R.J.H., Spaink, H., Hille, J., van Brussel, T.A.N., Lugtenberg, B., and Schilperoort, R.A. (1984) Plant-inducible virulence promoter of the Agrobacterium tumefaciens Ti plasmid. Nature 312:564-566. Parke, D., Rivelli, M., and Ornston, LN. (1985) Chemotaxis to aromatic and hydroaromatic acids: comparison of Bradyrhizobium japonicum and Rhizobium trifolii. J. Bacteriol. 163:417-422. Parke, D., Ornston, L.N., and Nester, E.W. (1987) Chemotaxis to plant phenolic inducers of virulence genes is constitutively expressed in the absence of the Ti plasmid in Agrobacterium tumefaciens . J. Bacteriol. 169:5336-5338. Peters, K. and S. Long (1987) The role of plant compounds in the regulation of Rhizobium nodulation genes. Symposium #7. 27th Annual Meeting of the Phytochemical Society of North America. Tampa, Florida. June 21-26, 1987. Peters, N.K., Frost, J.W., and Long, S.R. (1986) A plant flavone, luteolin, induces expression of Rhizobium meliloti nodulation genes. Science 233:977-980.  Petit, A., Temp6, J., Kerr, A., Holsters, M., Van Montagu, M., and Schell, J. (1978) Substrate induction of conjugative activity of Agrobacterium tumefaciens. Nature 2 7 1 : 570-572. Redmond, J.W., Batley, M., Djordjevic, M.A., Innes, R.W., Kuempel, P.L, and Rolfe, B.G. (1986) Flavones induce expression of nodulation genes in Rhizobium. Nature 323:632-635. Robertson, J.L., Holliday, T., and Matthysse, A.G. (1988) Mapping of Agrobacterium tumefaciens chromosomal genes affecting cellulose sunthesis and bacterial attachment to host cells. J. Bacteriol. 170:1408-1411. Rogowsky, P.M., Close, T.J., Chimera, J.A., Shaw, J.J., and Kado, C.I. (1987) Regulation of thev/r genes of Agrobacterium tumefaciens plasmid pTiC58. J. Bacteriol. 169:5101-5112. Sadowsky, M.J., Olson, E.R., Foster, V.E., Kosslak, R.M., and Verma D.P.S. (1988) Two host-inducible genes of Rhizobium fredii and characterization of the inducing compound. J . Bacteriol. 170:171-178. Schafer, W., Gorz, A., and Kahl, G. (1987) T-DNA integration and expression in a monocot crop plant after induction of Agrobacterium. Nature 327:529532. Schroder, G., Waffenschmidt, S., Weiler, F.W., and Schroder, J. (1984) The T-region of Ti plasmids codes for an enzyme synthesizing indole-3-acetic acid. Eur. J. Biochem. 138:387-391. Sheikholleslam, S.N., and Weeks, D.P. (1987) Acetosyringone promotes high efficiency transformation of Arabidopsis thaliana explants by Agrobacterium tumefaciens. Plant Mol. Biol. 8:291-298. Sigee, D.C, Smith, V.A., and Hindiey, J. (1982) Passage of bacterial DNA into host cells during in vitro transformation of Nicotiana tabacum by Agrobacterium tumefaciens. Microbios 34:113-132. Smith, E.F., and Townsend, C.0.(1907) A plant tumor of bacterial origin. Science 25:671-673. Smith, V.A. and Hindiey, J. (1978) Effect of agrocin 84 on attachment of Agrobacterium tumefaciens to cultured tobacco cells. Nature 276:498-500. Stachel, S.E. and Zambryski, P.C. (1986) Agrobacterium tumefaciens and the susceptible plant cell: anovel adaptation of extracellular recognition and DNA conjugation. Cell 47:155-157. Stachel, S.E., An, G., Flores, C , and Nester, E.W. (1985) A Tn3 lacZ transposon for the random generation of 6-galactosidase gene fusions: application to the analysis of gene expression in Agrobacterium. EM BO J. 4:891-898. Stachel, S.E., and Zambryski, P.(1986) vir A and virG control the plant-induced activation of the T-DNA transfer process of Agrobacterium tumefaciens. Cell 46:325-333.  Stachel, S.E., Messens, E., Van Montagu, M., and Zambryski, P. (1985) Identification of the signal molecules produced by wounded plant cells that activate T-DNA transfer in A g r o b a c t e r i u m tumefaciens. Nature 318:624-629. Stachel, S.E., Nester, E.W., and Zambryski, P.C. (1986) A plant cell factor induces A g r o b a c t e r i u m tumefaciens vir gene expression. Proc. Natl. Acad. Sci. 83:379-383. Stachel, S.E., Timmerman, B., and Zambryski, P. (1987) Activation of A g r o b a c t e r i u m tumefaciens vir gene expression generates multiple single-stranded Tstrand molecules from the pTiA6 T-region: requirement for 5' vir D gene products. EMBO J. 6:857-863. Subba Rao, P.V., Nambudiri, A.M.D., and Bhat, J.V. (1971) Microbial degradation of phenylpropanoid compounds. J. Sci. and Ind. Res. 30:663-679. Tang, C , and Young, C. (1982) Collection and identification of allelopathic compounds from the undisturbed root system of bigalta limpograss ( H e m a r t h r i a a l t i s s i m a ) . Plant Physiol. 69:155-160. Thomashow, L.S., Reeves, S., and Thomashow, M.F. (1984) Crown gall oncogenesis: evidence that a T-DNA gene from the A g r o b a c t e r i u m , Ti plasmid pTiA6 encodes an enzyme that catalyzes synthesis of indole acetic acid. Proc. Natl. Acad. Sci. 81:5071-5075. Thomashow, M.F., Nuter.R., Montoya, A.L, Gordon, M.P., and Nester, E.W. (1980) Integration and organization of Ti-plasmid sequences in crown gall tumors. Cell 19:729-739. Veluthambi, K., Jayaswal, R.K., and Gelvin, S.B. (1987) Virulence genes A, G, and D mediates the double-stranded border cleavage of T-DNA from the A g r o b a c t e r i u m Ti plasmid. Proc. Nat. Acad. Sci. 84:1881-1885. Wakarchuk, W.W., Greenberg, N.M., Kilburn, D.G., Miller, R.C.Jr., and Warren, R.A.J. (1988) Structure and transcriptuional analysis of the gene encoding a cellobiase from A g r o b a c t e r i u m sp. strain ATCC 21400. J. Bacteriol. 170:301-307. Willmitzer, L, De Beuckeleer, M., Lemmers, M., Van Montagu, M., and Schell, J. (1980) The Ti-plasmid derived T-DNA is present in the nucleus and absent from the plastids of plant crown gall cells. Nature 287:359-361. Willmitzer, L, Schmalenbach, W., and Schell, J. (1981) Transcription of T-DNA in octopine and nopaline crown gall tumors is inhibited by low concentrations of a-aminitin. Nucleic Acid Res. 9:4801-4812. Winans, S.C, Ebert.P.R., Stachel, S.E., Gordon, M.P., and Nester, E.W. (1986) A gene essential for A g r o b a c t e r i u m virulence is homologous to a family of positive regulatory loci. Proc. Natl. Acad. Sci. 83:8278-8282. Yadav, N.S., Postle, K., Saiki, R.K., Thomashow, M.F., and Chilton, M. -D. (1980) TDNA of a crown gall teratoma is covalently joined to host plant DNA. Nature 287: 458-461.  Yamamoto, A., Iwahashi, M., Yanafsky, M.F., Nester, E.W., Takebe, I., and Machida, Y. (1987) The promoter proximal region in the virD locus of Agrobacterium tumefaciens is necessary for the plant-inducible circularization of T-DNA. Mol. Gen. Genet. 206:174-177. Yanofsky, M., Lowe, B„ Montoya, A., Rubin, R., Krul, W., Gordon, M., and Nester, E.W. (1985) Molecular and genetic analysis of factors controlling host range in Agrobacterium tumefaciens. Molec. Gen. Genet. 201:237-246. Yanofsky, M.F., Porter, S.G., Young, C , Albright, L.M., Gordon, M.P., and Nester, E.W. (1986) The vir D operon of Agrobacterium tumefaciens encodes a sitespecific endonuclease. Cell 47:471 -7. Zambryski, P., Holsters, M., Kruger, K., Depicker, A., Schell, J., Van Montagu, M., and Goodman, H.M. (1980) Tumor DNA structure in plant cells transformed by A. tumefaciens. Science 209:1385-1391.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items