g ' .l 1!: I. 00:71 for V0 1:: J .. . us... .7. .. z 3.3....» . . F.) ‘70-..."0'O4-I", ’0 3"! A .1". Pliil‘V‘: EC. '31? n ‘ I. .1. {71 r Ice-5! f. 0 III!" § ’3. 3. «9:11.. .‘ (interiuafl is .. 1-0.1.7 .v I it}. ‘- Cvrcf tilii'V'V .31! . Villrtqoz; l'lettv (it'll—9i luffuvuulr'zo!’ vlvluovocVTaY-a. I: I: II. I‘.‘n!fl.a It! 74' g .L‘OYDIQVI) I!!! «6.1),!!! :0. “.i.> 3v: I'lll'lr.ul $1454 (Cvrll )llcl .1; A...J..I..£:‘ Ii r SAC-I. 11.52. , 0!... v... . :..Illo .u Iffu 4:.ot .VI .3 1.1.1.! 4 ‘6 ill.- Ilitl.‘ rt!!! .1: I Ix .1. In“! I \A Illa , .nfn‘v I'll; lsivil 49.! III: a)!!! v.4, Ifttnttid ‘f'clpllvUOvt' it! a. I”: IIVVvII?! will; It! .0. [(0 fill. ‘1 ‘- I". r. 1.61."! I!!! l.l.llll1lll.x)vl#ll\:'t. if I.r1\.llir'v(‘ Ltlllvlvb’ tutti 113.0}..1‘ . A 2.; a! .11!!! I. 1.00215. {Sixltbtl . . 1.3 51.-.? i .315; ..v(7..._l.1¢ ii» roif‘.)1uuca. . . . . 15.3.2... (21.1-32! .Ii . . i---.:.:.:... . O a ‘ . ., fiat»: . ‘1 2 a} F... _ .8... 4.3231"! 3fir.323fimgrkmrs.s ‘z . 13111.. ‘ n .huo.l.. 1’1). .43. a, A 1 1 ‘ . . . a . .3 .u. as. u, ::uwu§un% ism.“ Vsusnwfifiegx .8 ‘ ‘ udodts|lig ITTITTTTTTTTTTITTTTTTTTTTTTTTTTTTIT TTTITTTTTT TITTTTTI 31293 00794 9773 This is to certify that the dissertation entitled Interactions of Arabidopsis thaliana with two bacterial plant pathogens: zanthomonas campestris pv. campestris and Pseudomonas syringae pv. sxringae presented by Jun Tsuji has been accepted towards fulfillment of the requirements for / f T 4/» (mLa—vmxw /¢~ Major professor Date 9-23-92 MSU 15 an Affirmative Action/Equal Opportunity Institution 0-12771 ZBRARY Michigan State T University u T *— a PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID F INES return on or before date due. DATE DUE DATE DUE DATE DUE 1 T_:TL__j fl _____JT_ fiT—T—T MSU Is An Affirmative Action/Equal Opportunity Institution cMWMt INTERACTIONS OF ARABIDOPSIS THALIANA WITH TWO BACTERIAL PLANT PATHOGENS: XANTHOMONAS CAMPESTRIS PV. CAMPESTFIIS AND PSEUDOMONAS SYRINGAE PV. SYRINGAE By Jun Tsuji A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY GENETICS PROGRAM 1992 ABSTRACT INTERACTIONS OF ARABIDOPSIS THALIANA WITH TWO BACTERIAL PLANT PATHOGENS: XANTHOMONAS CAMPESTRIS PV. CAMPESTRIS AND PSEUDOMONAS SYRINGAE PV. SYRINGAE By Jun Tsuji A new model system was developed to study the molecular and biochemical basis of the disease resistance response of plants to phytopathogenic bacteria. This model system consisted of the interactions at Arabidopsis thaliana with the bacteria Xanthomonas campestris pv. campestris and Pseudomonas syringae pv. syringae. X. c. campestris multiplied approximately three orders of magnitude and caused extensive chlorosis and necrosis when inoculated into leaves of Arabidopsis landrace Pro. In contrast, leaves of Arabidopsis landrace Columbia showed slight chlorosis or were symptomless when inoculated in the same manner. By genetic analysis, resistance in landrace Columbia was found to be determined by a dominant allele of a single, nuclear gene. This locus, RXC1, was mapped to approximately 2.9 cM from restriction fragment length polymorphism marker 251 on chromosome 2. This marker may serve as a starting point in a chromosome walk directed at cloning RXC1. When inoculated into leaves of Arabidopsis, Pseudomonas syringae pv. syringae induced the hypersensitive reaction and the accumulation of a phytoalexin. This phytoalexin, camalexin, was purified and the structure was identified as 3-thiazoI-2'-yl-indole. Camalexin was found to inhibit the growth of both the bacterium P. s. syringes and the fungus Cladosporium cucumerinum, and the accumulation of this phytoalexin was correlated with the restricted in planta growth of P. s. syringae. The hypothesis that tryptophan is a precursor to camalexin was tested. Feeding studies using 1‘I'TC-anthranilate and 3H-tryptophan suggested that tryptophan is not a precursor to camalexin. The use of tryptophan-requiring mutants of Arabidopsis supported this conclusion since two different tryptophan synthase mutants accumulated wild-type levels of camalexin. Using the molecular and genetic tools available with Arabidopsis, the role of this phytoalexin in disease resistance can now be addressed. ACKNOWLEDGMENTS I thank Dr. Shauna Somerville and Dr. Ray Hammerschmidt for their guidance and support during the course of this thesis. I also thank Dr. Jonathon Walton and Dr. Dennis Fulbright for serving on my committee. I am especially indebted to my wife, Wendy, and daughter, Haley, for their love and encouragement. iv TABLE OF CONTENTS Page List of Tables ................................................................................................... v i i i List of Figures ................................................................................................. x A Review of the Literature ......................................................................... 1 General Introduction ...................................................................................... 2 Arabidopsis as a Model Host for Studies of Disease Resistance...3 Pathogens of Arabidopsis ................................................................ 6 A Viral ............................................................................................... 6 B Bacterial ....................................................................................... 9 C Fungal ............................................................................................. 1 3 D Nematode ...................................................................................... 1 5 E. Fastidious Prokaryotic ............................................................ 1 6 Defense-related Responses .............................................................. 1 7 References ................................................................................................ 2 2 Chapter I. First Report of the Natural Infection of Arabidopsis Chapter II. Xanthomonas campestris pv. campestris—induced Chlorosis in Arabidopsis thaliana ............................................................ 3 5 Introduction ............................................................................................ 3 6 Materials and Methods ........................................................................ 3 7 Results and Discussion ...................................................................... 3 9 References ............................................................................................... 4 3 Chapter III. Identification of a Gene in Arabidopsis thaliana that Controls Resistance to Xanthomonas campestris pv. canpestn‘s .......................................................................................................... 44 Introduction ............................................................................................ 4 5 Materials and Methods ........................................................................ 4 6 Results ......................................................................................................... 47 Discussion ................................................................................................ 5 1 References ............................................................................................... 5 2 Chapter IV. Genetic Mapping of RXCl, A Xanthomonas Resistance Gene in Arabidopsis thaliana .................................. 5 4 Introduction ............................................................................................ 5 5 Materials and Methods ............................................ . ............................ 5 6 Results ......................................................................................................... 5 7 Discussion ................................................................................................ 7 1 References ................................................................................................ 7 5 Chapter V. Phytoalexin Accumulation in Arabidopsis thaliana vi During the Hypersensitive Reaction to Pseudomonas syringae pv. syringes ....................................................................................................... 8 1 Introduction ............................................................................................ 8 2 Materials and Methods ........................................................................ 8 2 Results ......................................................................................................... 83 Discussion ................................................................................................ 8 6 Literature cited ..................................................................................... 8 7 Chapter VI. Camalexin Biosynthesis in Arabidopsis lhalr‘ana ............................................................................................................... 8 8 Introduction ............................................................................................ 8 9 Materials and Methods ........................................................................ 9 0 Results ......................................................................................................... 9 2 Discussion ................................................................................................ 9 8 References ............................................................................................... 1 04 Chapter VII. Future Directions ............................................................... 1 07 References ............................................................................................... 1 1 4 Appendix .............................................................................................................. 1 1 6 vii Table Page 1. Arabidopsis thaliana races .................................................................. 3 8 2. Mean chlorophyll contents for Arabidopsis races Columbia, Bur-O, and Pr—O ................................................................... 42 3. Mean and standard error of the mean of chlorophyll contents for Arabidopsis leaves inoculated with X. 0. Wins ................................................................................................. 50 4. Formulae for Castle-Wright indicies (nE) and "E values for the minimum number of genes (effective factors) controlling the response of Arabidopsis to inoculation withX.c.campestn’s .............................................................................. 50 5. Number of F2 individuals with the specified morphological phenotype and disease response from crosses between Pro and six Landsberg mapping lines ...................................................... 61 6. Markers, X2, and P values used for the detection of LIST OF TABLES viii 10. 11. linkage with RXC1 .................................................................................... s 3 Restriction enzymes yielding restriction fragment length polymorphisms between Arabidopsis landraces Columbia and Pro for markers ASA2, 251, 605, and ASA2... 65 Disease phenotypes and RFLP genotypes for F3 families derived from a cross between Arabidopsis landraces Columbia and Pro ...................................................................................... 6 7 Recombination fractions and map distances between loci....72 Incorporation of radiolabel from 14C-anthranilate or 3H-tryptophan into camalexin ............................................................ 9 7 Camalexin levels in three tryptophan-requiring mutants of Arabidopsis eighteen hours after elicitation with 10 mM silver nitrate, 0.1% Tween 20 .............................................. 99 ix LIST OF FIGURES Figure , Page 1. Symptom development of Arabidopsis race Pr-O 5 days after inoculation with X. c. campestris; initial leaf yellowing, a, and spreading chlorosis, b ........................................ 40 2. Chlorophyll profiles of three races of Arabidopsis inoculated with X. c. campestris. Chlorophyll values (ug/cm2 1: SE) of the controls at day 0 were 63.77 :I: 0.53 (Bur-0), 57.92 :I: 0.88 (Columbia), and 50.22 :I: 1.66 (Pr-0) .................................................................................................. 42 3. Symptoms on leaves of (A) Pro and (8) Columbia 6 days after infiltration with a suspension of X. c. campestris at a concentration of 1-5X106 cfu ml'1. Arrows indicate healthy (H), chlorotic (C), and necrotic (N) tissues ................. 47 4. Chlorophyll contents of leaves of Arabidopsis inoculated with a 1-5X106 cfu ml'1 suspension of X. c. campestris. Solid bar: Columbia inoculated with X. c. campesfris 20520; Stippled bar: Pro inoculated with the black rot mutant X. c. campestris JS111; Open bar: Pr0 inoculated with X. c. campestris 2D520. Values are the mean percentages of ten inoculated plants per time point. Standard error bars are shown. Chlorophyll contents (pg/cm2 1 SE) of the uninoculated control leaves at day 0 were 21.47 i 0.53 for Columbia; 17.72 i: 0.42 for Pro, X. c. campestris 20520; and 19.07 :t 0.71 for Pro, X. c. campestrisJS111 ........................................................................... 48 Symptoms on leaves of (A) Pro and (8) Columbia 2 weeks after inoculation with X. c. campestris via the hydathodes. Arrows indicate healthy (H), chlorotic (C), and necrotic (N) tissues ........................................................................ 4 9 Time course of growth of X. c. campestris infiltrated into leaves of ([3) Pro and (I) Columbia. Each data point represents the mean of six replicates. Standard error bars are shown. Similar results were obtained in two additional experiments .......................................................................... 4 9 Frequency distribution histogram of chlorophyll contents of X. c. campestris—inoculated leaves for 102 F2 individuals derived from a cross between Arabidopsis landraces Columbia and Pro ................................................................. 5 8 Restriction fragment length polymorphisms between landraces Columbia (left lane) and Pro (right lane) for the markers 605 (Bgl ll), ASA2 (Bgl ll), GPA1 pCIT1838 (fijnd Ill), and 251 (Hind Ill). Lefthand margin indicates x1 size in kilobases. The RFLP shown for marker 251 was between two F3 families. This RFLP is representative of the polymorphism observed between Columbia and PrO .................................................................................................................. 66 Reverse phase HPLC of fungitoxic fractions obtained from silica gel TLC plates. Elution was with a linear gradient of 1 to 100% acetonitrile .................................................. 83 10. 1H-NMFi spectrum (500 MHz, CDCI3) of the Arabidopsis 11. 12. 13. phytoalexin .................................................................................................. 8 4 Structure of the Arabidopsis phytoalexin, 3-thiazol-2'-yl-indole ........................................................................... 8 5 Time course of growth of P. s. syringae PSSD20 infiltrated into leaves of Arabidopsis. Each point is the mean of six replicates. Standard error bars are shown. Similar results were obtained in two additional experiments in which leaf discs were surface sterilized in 0.5% sodium hypochlorite for 1 to 2 min before maceration .................................................................................................. 8 5 Time course of the accumulation of phytoalexin activity. Leaves of Arabidopsis race Columbia were infiltrated with (o) P. s. syringae PSSD220, (o) P. s. syringae PSSDZZ, (x) X. c. campestris 2D520, or (I) 10 mM potassium x11 14. 15. 16. phosphate buffer (pH 6.9). Chloroform-soluble fractions of methanol leaf extracts (50 mg fresh weight) were assayed for fungitoxic activity, and the area of fungal growth inhibition in mm2 was recorded. Each point is the mean of three separate experiments. Standard error barsareshown .......................................................................................... 85 Time course of the accumulation of 3-thiazol-2'-yl-indole following inoculation with P. s. syringae PSSD20. The phytoalexin was purified from inoculated leaves of Arabidopsis and quantified using a molar extinction coefficient of 14,800 M'1 cm". Each point is the mean of two separate experiments. Standard error bars are shown ............................................................................................................ 86 Time course of camalexin accumulation following elicitation with 10 mM silver nitrate, 0.1% Tween 20. Each point is the mean of three separate experiments. Standard error bars are shown ........................................................... 9 4 TLC profiles of 14C-anthranilate-labelled and 3H-trypt0phan-labelled metabolites. Extracts prepared from control (A and C) or elicited (B and D) leaves fed with 10 mCi/mmole 14C-anthranilate (A and B) or 10 mCi/mmole 3H-tryptophan (C and D) were spotted on a TLC plate. The plate was developed in chloroform: methanol (9:1, v/v), and scanned for radioactivity using x111 17. 18. a Bioscan beta-detector. The arrow indicates the camalexin peak .......................................................................................... 95 The tryptophan biosynthetic pathway. The enzymatic steps with decreased activity in three tryptophan -requiring mutants of Arabidopsis are shown ............................ 101 Phytoalexins and stress metabolites of crucifers ..................... 117 xiv A REVIEW OF THE LITERATURE 2 GENERAL INTRODUCTION Although plants are exposed to a wide array of potentially pathogenic microorganisms throughout their life cycle, successful infections are rare. The mechanisms by which plants resist infection are not fully understood; however, a number of laboratories have shown that a wide array of host responses are induced during an incompatible reaction, which are either absent or less induced during a compatible interaction. These resistance ~associated responses include, but are not limited to, the hypersensitive reaction, the accumulation of antimicrobial compounds called phytoalexins, the accumulation of enzymes with fungal cell wall degrading activities such as chitinase and beta-1,3-glucanase, and the accumulation of proteins involved in plant cell wall modifications like hydroxyproline-rich glycoproteins and peroxidase (10). However, because a number of host responses have been correlated with disease resistance, the role that any one host response plays in resisting infection remains unclear. Genetic analysis of a number of plant-pathogen interactions has shown that resistance in the plant and avirulence in the pathogen are conditioned by dominant alleles of single genes. A common interpretation of these observations is that the product of the pathogen's avirulence gene interacts directly or indirectly with the product of the plant's resistance gene. The biochemical function of avirulence genes are unknown; however evidence from Keen's laboratory suggests that at least one avirulence gene is involved in the production of a low molecular weight, extracellular elicitor 3 (26). The biochemical function of resistance genes is also not fully understood, with the exception of the Hm1 locus in maize. Walton, Briggs, and co-workers have demonstrated that the Hm1 locus most likely encodes an enzyme that inactivates HC-toxin by pyridine nucleotide-dependent reduction of an essential carbonyl group (25,40). With the exception of this one case involving a host-specific toxin, resistance genes have not been cloned and their mRNA and protein products are unknown. The cloning of disease resistance genes and the analysis of mutants altered in their disease phenotype are required to fully understand the complexities of disease resistance. However, the majority of plants used for biochemical and molecular studies of disease resistance are not amenable to molecular and genetic techniques. Recently, Arabidopsis thaliana has been shown to be a host for a number of bacterial, fungal, viral, nematode, and fastidious prokaryotic pathogens. These studies have demonstrated that Arabidopsis offers tremendous potential as a model host for molecular and biochemical studies of disease resistance. ARABIDOPSIS AS A MODEL HOST FOR STUDIES OF DISEASE RESISTANCE Arabidopsis thaliana posseses a number of attributes which make it an attractive model organism for studies of plant-pathogen interactions (16,39,43,44,59). Arabidopsis is a small, self-fertile, diploid weed in the family Cruciferae with a rapid life cycle and abundant seed set. Thus, genetic analysis of disease resistance 4 genes can be conducted within a relatively short period of time and in a relatively small space such as a growth chamber. These features also make Arabidopsis well suited for mutational analysis of disease resistance-related processes (16,39). Also available for studies of disease resistance is a large collection of naturally occurring landraces of Arabidopsis which can be screened for variability in resistance to a particular plant pathogen. A number of mutants of Arabidopsis affecting different aspects of metabolism and development are available (16.39). For example, tryptophan-requiring mutants of Arabidopsis isolated by Last and Fink enabled us to test the hypothesis that tryptophan is a precursor to camalexin (33,34). Arabidopsis has one of the smallest genomes of any higher plant, with a haploid nuclear genome size of approximately 70,000 kb, as compared to 5,900,000 kb for wheat (43). The small genome size makes it easier to screen a genomic library to isolate a disease resistance gene. Only 16,000 random lambda clones of 20-kb average insert size must be screened to have a 99% chance of obtaining any fragment of Arabidopsis DNA (43). In contrast, 370,000 clones in tobacco, 1,000,000 clones in pea, and 1,400,000 clones in wheat would have to be screened from a similar library to have the same expectation of success (43). Arabidopsis has very little, interspersed repetitive DNA. The average length of unique sequences in the Arabidopsis genome is 125 kb, as compared to 1.4 kb in tobacco (43). This feature is important because it makes chromosome walking feasible. Overlapping genomic clones for large contiguous regions of the 5 Arabidopsis genome can be obtained. The presence of repetitive DNA in other plant genomes makes it difficult to select for single-copy probes for each step in the chromosome walk (43). Before walking to a particular genetic locus, it is necessary to accurately map its location in the genome. Mapping is accomplished by linkage analysis of the disease resistant phenotype with various genetic markers. The Arabidopsis genetic map consists of approximately 80 morphological markers (30). In addition, two restriction fragment length polymorphism (RFLP) maps have been constructed for Arabidopsis (8,45). The two RFLP maps collectively consist of about 200 markers with an average distance of 350 kb between each marker. Mapping can also be conducted using random amplified polymorphic DNA (RAPD) markers. A RAPD map has recently been constructed for Arabidopsis (49). This map has been integrated with the existing maps of RFLP and morphological markers, and contains 320 markers with an average distance of 2 cM between each marker. The chromosome walk is facilitated by the construction of several YAC libraries (21,70). YACs are yeast artificial chromosomes that contain a yeast replication origin, centromere, telomeres, and selectable markers. YAC vectors have the advantage over cosmid vectors in that YAC vectors can hold inserts of several hundred kb as opposed to cosmid vectors which can hold inserts of less than 50 kb (7). The YAC library constructed by Grill and Somerville consists of 2300 YAC clones, which is equivalent to about 3 genomes of Arabidopsis, and the YAC clones have an average insert size of 150 kb (21). As a first step towards the construction 6 of an overlapping YAC library of the Arabidopsis genome, YAC clones corresponding to 125 RFLP markers have been identified (22). These YAC clones of known genetic map location encompass about 30% of the Arabidopsis genome. The last step in cloning a disease resistance gene by chromosome walking is to identify which YAC clone contains the gene of interest. This step is accomplished by transforming the susceptible plant line with YAC subclones and testing the transformants for disease resistance. This approach is possible in Arabidopsis because this plant can be efficiently transformed using standared Agrobacterium-based vectors (68). In addition, the development of T-DNA tagging (17), transposon tagging (20) and genomic subtraction (60) in Arabidopsis further expands the approaches available in cloning genes involved in disease resistance. PATHOGENS OF ARABIDOPSIS A. VIRAL PATHOGENS One of the first reported viral pathogens of Arabidopsis was cauliflower mosaic virus (CaMV). Balazs and Lebeurier inoculated two different strains of CaMV onto five landraces of Arabidopsis (3). Seven to nine days post-inoculation, inoculated plants appeared slightly stunted and leaves showed typical mosaic symptoms. Flowers were distorted and seed yield markedly reduced. All of the five landraces of Arabidopsis tested were susceptible to infection, but with slight differences in symptom severity. 7 The ability of CaMV to infect Arabidopsis was later confirmed by Melcher (42). Melcher reported differences in symptoms depending on the CaMV isolate used. Isolate NY8153 and isolates derived from CM4-184 caused stunting of rosette leaves, while CM4-184 did not. Furthermore, the group of isolates that caused stunting on Arabidopsis were different from those that caused stunting on turnip. Differential reactions of Arabidopsis landraces to inoculation with CaMV were also reported by Leisner and Howell (37). These researchers inoculated four different strains of CaMV onto several different landraces of Arabidopsis. They observed that the landraces varied in their resistance to specific viral isolates, but no landrace was resistant to all four isolates. The differences in disease phenotype were also accompanied by differences in the distribution of CaMV DNA in infected plants. Ramachandra et al. tested five landraces of Arabidopsis for differential responses to turnip yellow mosaic virus (TYMV) (48). Using three different inoculum concentrations and two different temperature regimes, they reported that landraces Landsberg and Nd-0 were susceptible to infection, but systemic infections occured only at high inoculum concentrations and at high temperatures. In contrast, systemic infection occured on landraces Col-0 and Bur-0 at all inoculum concentrations and temperatures tested. Landrace Fi-3 was found to be resistant to TYMV, although no genetic analysis of resistance was reported. Skotnicki et al. also reported the infection of Arabidopsis by TYMV (56). The virus caused necrotic lesions, strong yellow mosaic 8 of systemically infected leaves, flower sterililty, and eventually plant death. Mutants of Arabidopsis that survived TYMV inoculation were selected. Twenty mutants were isolated that exhibited a range of disease phenotypes from symptomless to strong yellow mosaic. Simon and co-workers have found that Arabidopsis is also susceptible to turnip crinkle virus (11). Symptoms appear one week after inoculation as stunting and browning. By screening 24 landraces of Arabidopsis, Columbia was identified as susceptible and Dijon as resistant to infection. Resistance was reported as a delay in symptom development and in viral replication. In a preliminary study of the inheritance of resistance to turnip crinkle virus, Simon reported that resistance was conferred by a partially dominant allele of a single gene. Davis et al. have recently described susceptibility of Arabidopsis to Beet Curly Top Virus (BCTV) (11). Infected plants are stunted and develop abnormal floral structures. In some cases, leaves become chlorotic and wrinkled. Preliminary studies by Urban et al. have shown that Arabidopsis is susceptible to tobacco mosaic virus (TMV) (67). Two different strains of TMV were used in cross-protection experiments. Leaves were inoculated with one strain, and at various times after inoculation, the leaves were subsequently challenged with the second strain. Regardless of the viral strain inoculated first, no movement of the second virus strain was detected. In contrast, the strain that was inoculated first was always found to spread systemically. 9 The ability of TMV to infect Arabidopsis was later confirmed by lshikawa et al. (24). lshikawa and co-workers used a crucifer strain of TMV (TMV-Cg) to screen for mutants of Arabidopsis in which the multiplication of TMV-Cg was impaired. Of 6,000 M2 plants that were screened, two were identified that had lower levels of TMV-Cg than in the wild-type. Genetic analysis revealed that the two mutants belong to the same complementation group and that the phenotype was caused by a 'single, nuclear, recessive mutation. The ability of turnip crinkle virus or turnip yellow mosaic virus to replicate in the mutants was not altered, suggesting that the mutation is specific for TMV. B. BACTERIAL PATHOGENS Several groups have found that Arabidopsis thaliana is susceptible to infection by Xanthomonas campestris pv. campestris (9,55,62,63). X. c. campestris, the causal agent of black rot disease, is the most economically important bacterial pathogen of crucifers (72). X. c. campestris has a wide host range and has been reported to infect cruciferous weeds (31,52,53). Not surprisingly, Tsuji and Somerville reported the natural infection of a wild Michigan landrace of Arabidopsis by X. c. campestris (63). Symptoms appeared as chlorotic and necrotic lesions along the margins of the leaves. Similar symptoms were observed when plants were artificially inoculated in the laboratory by leaf infiltrations or by hydathode inoculations. Symptoms were associated with about a 103-fold increase in bacterial growth over a five day period (64). 10 _ Simpson and Johnson also reported the ability of X. c. campestris to cause vein blackening on leaves of Arabidopsis by wound inoculations (55). These symptoms were accompanied by approximately a 103-fold increase in bacterial growth over a six day period. Dark veins is one of the symptoms of black rot disease observed on Brassica, but was not observed on naturally infected Arabidopsis plants (63). Differential reactions of Arabidopsis landraces to inoculation with X. c. campestris have been reported (9,35,36,55,62,64). Tsuji et al. reported resistance in Columbia that appeared as little or no chlorosis and susceptibility in Pro that appeared as extensive chlorophyll loss by either leaf infiltrations or hydathode inoculations (64). However, no differences in bacterial growth were observed in either landrace Columbia or Pro. Simpson and Johnson reported resistance to X. c. campestris in landrace Pn-0 and susceptibility in Ob-1 (55). Resistance appeared as less vein blackening and a 10-fold reduction in bacterial growth. Simpson and Johnson also examined the disease phenotypes of three landraces of Arabidopsis inoculated with two different strains of X. c. campestris (55). Landraces Uk-2 and Bu-20 responded similarly to X. c. campestris strains KXCC and XCCS. However, landrace Edi-0 was resistant to KXCC but susceptible to XCCS. The bacterium X. c. armoraceae was also reported to be a pathogen of Arabidopsis (35,36). When infiltrated into leaves, X. c. armoraceae was found to cause leaf spotting on Arabidopsis landrace Landsberg but not on landrace Wassileskija. By genetic analysis, resistance was found to be determined by a semidominant 11 allele of a single locus. This locus, Rxaf, was mapped to chromosome 5 based on cosegregation with a single morphological marker. In addition, Daniels et al. have reported the cloning of an avirulence gene from X. c. armoraceae that differentially elicits a hypersensitive response on different landraces of Arabidopsis (9). Several groups have reported that Arabidopsis is also susceptible to infection by P. s. maculicola (13,14,15,71). Infiltration of P. s. maculicola into leaves of Arabidopsis causes greyish, water-soaked lesions, chlorosis of the surrounding tissue, and eventually necrosis. Symptoms are associated with a 103 to 104-fold increase in bacterial growth. Debener et al. have reported resistance to P. s. maculicola which appeared as the expression of the hypersensitive response and limited in planta bacterial growth (14). Resistance to P. s. maculicola m2 was observed in landraces Oy-0 and Col-0, and susceptibility was observed in landrace Nd-0. An avirulence gene (aerpmf) was cloned from P. s. maculicola m2 by the ability to confer an HR phenotype on Oy-0 when mobilized into the virulent isolate P. s. maculicola m4. Resistance to P. s. maculicola m2 was conferred by a dominant allele of a single gene (Rpmf), which was mapped to chromosome 3. Furthermore, resistance to aerpmf co-segregated with resistance to P. s. maculicola m2, suggesting that Rpmf conditions resistance to aerpmf. In addition to P. s. maculicola, the closely related pathovar P. s. tomato was also found to infect Arabidopsis (13,15,71). Symptoms caused by P. s. tomato were the same as those reported for P. s. maculicola. Disease symptoms appeared as a greyish-brown 12 (sometimes water-soaked) lesions with spreading chlorosis that was accompanied by 104 to 105-fold increase in bacterial growth. Resistance to P. s. tomato 1065 was identified in Arabidopsis landraces Col-0, Po-1, and H50 An avirulence gene (aerpt2) was cloned from P. s. tomato 1065 by the ability to confer an HR phenotype on Col-0 but not Po-1 when mobilized into P. s. tomato DC3000 (71). The same avirulence gene was independently cloned by the ability to confer an HR phenotype on Col-0 when mobilized into P. s. maculicola E84326 (15). In a preliminary study of the inheritance of resistance to P. s. tomato DCSOOO(aerpt2), resistance in landrace Po-1 was found to be conditioned by more than one gene (4). In addition, mutants of landrace Col-0 that no longer respond with an HR to inoculation with P. s. tomato DCSOOO(aerpt2) and P. s. maculicola ES4326(aerpt2) have recently been isolated (32,73). These mutants are altered specifically in the ability to recognize bacteria expressing aerpt2 since they retain resistance to bacteria carrying other avirulence genes. By genetic analysis, the mutant phenotype was found to be controlled by a recessive allele of a single gene. This locus, Rpt2, was located 4.3 cM from marker At600 between markers At600 and g3088 on chromosome 4. When the avirulence gene avrB was introduced into P. s. tomato 003000, the expression of this gene was found to elicit the HR on Arabidopsis landrace Columbia (23). However, P. s. tomato DC3000(avrB) failed to elicit an HR on landrace Bla-2. By crossing Columbia with Bla-2 and scoring the disease phenotypes in the F2 generation, lnnes et al. found that resistance was determined by a 13 dominant allele of the locus HRB1 (23). When P. s. tomato DCSOOO(avrB) was inoculated into leaves of soybean, this bacterium was able to elicit an HR on cultivar Merit but not Flambeau. These results suggest that HRBf may be functionally homologous to the soybean resistance gene RPGf. Tapio Palva has also reported that Erwinia carotovora can cause tissue maceration of Arabidopsis (46). Treatment of plants with cell wall degrading, extracellular enzymes induced the accumulation of several classes of PR proteins including PR1 and beta-1,3-glucanase (46,58). Inoculation of Arabidopsis with mutants of Erwinia carotovora that are reduced in virulence but still retain the ability to secrete pectic enzymes, induced the accumulation of these PR proteins. However, these PR proteins accumulated to much lower levels when leaves were inoculated with the wild-type bacterium. Palva hypothesized that the pectic enzymes secreted by Erwinia carotovora release pectic fragments from the plant cell wall that act as elicitors of PR protein accumulation. Palva speculated that the wild-type bacterium is able to suppress the plant's defense responses such that PR protein accumulation occurs more slowly. C. FUNGAL PATHOGENS Peronospora parasitica, Rhizoctonia solani, and Botrytis cinerea were isolated from naturally infected Arabidopsis plants (28,29). The Arabidopsis strain Weiningen was found to be highly susceptible to P. parasitica and displayed extensive intercellular 14 mycelial growth, haustoria formation, conidiophore production, and copious sporulation. In contrast, the RLD landrace of Arabidopsis was resistant to infection and reacted with a typical hypersensitive response. In a preliminary report of the inheritance of resistance, Slusarenko and Mauch-Mani reported that two or more genes determine the differential response of Arabidopsis landrace Weiningen and RLD to inoculation with P. parasitica (57). In a more thorough examination, Tor et al. have identified at least six loci that condition resistance to various isolates of P. parasitica (61). The chromosomal locations of four of the resistance loci have been identified (47,61). The resistance gene Rpr was mapped 4.7 cM from the marker M249 between the markers 91-1 and M249 on chromosome 3. The other resistance genes RPp2, RPp4, and RPp5 were all located to chromosome 4. The gene RPp2 was mapped 6 cM from the marker M557 between the markers M557 and M600, RPp4 was mapped 17.5 cM from M557 between the markers M557 and M326, and RPp5 was mapped 0.3 cM from both the markers 3845 and RAPD 018a. R. solani and B. cinerea were isolated from naturally infected Arabidopsis plants grown in the greenhouse (29). Plants infected with R. solani showed pale to dark brown lesions which later developed into a soft rot. Similar symptoms were observed when three different landraces of Arabidopsis were artificially inoculated with R. solani. No differences in susceptibility were observed among the three landraces of Arabidopsis inoculated with R. solani. The ability of R. solani to infect Arabidopsis was later confirmed by Berger et al. (5). A greyish mold was observed on 15 plants infected with B. cinerea. The incidence of this disease increased greatly following shoot formation and flowering. Sporulation was observed on the rosette leaves and on the lower portion of the shoots. Artificial inoculations of B. cinerea onto healthy plants were not performed. By artificial inoculations, Erysiphe cruciferarum was observed to infect leaves of Arabidopsis landrace Weiningen (29). Although macroscopic disease symptoms were absent, vegetative development was identical to the colonization of other host plants by E. cruciferarum. By 18 hours after inoculation, conidia germinated and formed large, irregularly lobed appresoria. Appresoria developed directly from conidia or less frequently from well-defined germ tubes. Haustoria then formed from appresoria which were accompanied by the formation of secondary hyphae. By five days after inoculation, conidiophores were observed. The lack of macroscopic symptoms and sparse hyphal growth and conidophore formation suggests a low degree of compatibility or unfavorable growth conditions. Albugo candida is an obligate, biotrophic, oomycete fungus that is able to infect Arabidopsis landrace Weiningen (6). Resistance to infection was identified in the landrace Keswick 37. By genetic analysis, at least two genes were found to condition resistance against this fungus. One of the resistance genes, RAcf, was mapped 12.2 cM from RFLP marker 271 on chromosome 1. D. NEMATODE PATHOGENS 16 Sijmons et al. reported susceptibility of Arabidopsis to a number of cyst-forming and root-knot nematodes (54). Sijmons and co-workers observed complete life cycles of the nematodes Heterodera schachtii, H. trifolii, H. cajani, Meloidogyne incognita, M arenaria, and Pratylenchus penetrans on Arabidopsis. In contrast, H goettingiana and Globodera rostochiensis were unable to infect Arabidopsis roots. Seventy-four different landraces of Arabidopsis were screened for susceptibility to H. schachtii. The landraces differed in their degree of susceptibility to H. schachtii, but no landrace exhibited complete resistance. E. FASTIDIOUS PROKARYOTIC PATHOGENS Golino et al. have reported infections of Arabidopsis by the beet leafhopper transmitted virescence agent (19). When Arabidopsis landrace La-0 was fed upon by the beet leafhopper (Circulifer tenellus) carrying the mycoplasma-like organism, symptoms that included virescence, phyllody, and the proliferation of flowers appeared five weeks later. Transmission electron microscopy revealed the presence of mycloplasma-like bodies in the phloem of symptomatic plants that were absent in asymptomatic plants. Fletcher used the beet leafhopper to infect Arabidopsis landraces Columbia and Landsberg with Spiroplasma citri (18). Symptoms appeared two weeks after inoculation as stunting of the basal rosette, curled and deformed cauline leaves, floral stunting and necrosis, reduced silique size and seed set, and reduced internode length of the floral stalk with terminal bunching of 17 flowers and siliques. Spiroplasmas were recovered from 23 of the 26 exposed plants, but none were recovered from uninoculated control plants. By polyacrylamide gel electrophoresis, the protein patterns of re-isolated spiroplasmas were indistinguishable from those of cultured S. citri, but were different from other spiroplasmas. DEFENSE-RELATED RESPONSES IN ARABIDOPSIS Initial studies of the induction of defense-related responses in Arabidopsis were conducted on suspension-cultured cells elicited with alpha-1,4-endopolygalacturonic acid (PGA) lyase (12). Treatment with PGA lyase induced the activation of a number of enzymes involved in phenylpropanoid metabolism including phenylalanine ammonia-lyase (PAL), 4-coumarate:CoA ligase (4CL), caffeic acid o-methyl transferase (CMT), and peroxidase. PAL and 4CL activities were transiently induced with similar induction kinetics and reached maximum levels at eight to ten hours after elicitation. CMT and peroxidase activities were induced more slowly and reached maximum levels at 24 hours post elicitor treatment. The transient increase in PAL and 4CL activities were preceded by transient increases in their mRNA levels. Cells treated with PGA lyase also had increased levels of beta-1,3-glucanase mRNA. Later studies demonstrated that similar patterns of gene expression were observed in leaves of Arabidopsis inoculated with pathogenic bacteria (15). Avirulent bacteria, Pseudomonas cichorii 18 83-1 and P. s. tomato MM1065, induced a 15 to 30-fold accumulation of PAL mRNA while virulent bacteria, P. s. maculicola ES4326, P. s. tomato 003000, and P. s. maculicola 795, induced a 5 to 10-fold transient accumulation of PAL mRNA. In contrast to PAL, accumulation of beta-1,3-glucanase mRNA was induced after inoculation with the virulent strain P. s. maculicola ES4326 but to lower levels with the avirulent strain P. s. tomato MM1065. Neither inoculation with virulent nor avirulent bacteria induced chalcone synthase mRNA accumulation. With the introduction of an avirulence gene (aeriptZ) from P. s. tomato MM1065 into P. s. maculicola ES4326, this bacterium was able to elicit PAL mRNA accumulation but not beta-1,3-glucanase mRNA accumulation. To further distinguish the virulent from avirulent Pseudomonas interactions with Arabidopsis, Davis et al. examined the alkalinization of cultured Arabidopsis cells treated with various Pseudomonas strains (13). P. cichorri 83-1, which induces a strong HR phenotype, induced an increase of approximately 1-1.5 pH units in the culture medium in five hours. The avirulent strain P. s. tomato 1065 induced an increase of about 0.3-0.6 pH units in five hours. In contrast, the virulent strains P. s. tomato DC3000, P. s. maculicola ES4326, and P. s. tomato 5034 did not significantly effect the pH of the culture medium. These results are similar to those of Atkinson et al. and Baker et al., who observed changes in the pH of cultured cells associated with a K"‘/H+ exchange during the hypersensitive response of tobacco to bacterial pathogens (1,2). In addition to PAL, a number of other defense-related genes have been cloned from Arabidopsis. A gene encoding lipoxygenase 19 has recently been isolated from Arabidopsis (41). The cDNA sequence exhits 64% and 80% nucleotide and amino acid similarity, respectively, when compared to soybean lipoxygenase 3. ln Arabidopsis, lipoxygenase appears to be encoded by a single gene. Lipoxygenase mRNA was abundant in the roots with lower levels found in healthy leaves. Levels of lipoxygenase mRNA increased rapidly in leaves following inoculation with the avirulent strain P. s. tomato MM1065 reaching maximum levels by 12 hours after inoculation. In contrast, lipoxygenase mRNA levels increased more slowly after inoculation with the virulent strain P. s. maculicola E84326 and reached a maximum by 48 hours. Infiltration of P. s. maculicola E84326 harboring the avirulence gene aerpt2 induced lipoxygenase mRNA accumulation similar to that observed with P. s. tomato MM1065. Leaf lipoxygenase mRNA levels also increased after exposure to methyl jasmonate, ABA, and wounding. Three glutathione-S-transferase genes (GST1, GST2, and GST3) have been cloned from Arabidopsis (73). GST1 mRNA accumulated rapidly to high levels in leaves three hours after inoculation with either the virulent bacterium P. s. maculicola ES4326or the avirulent bacterium P. s. maculicola ES4326(aerpt2). At 6-12 hours post inoculation, mRNA levels of GST1 remained high in leaves inoculated with P. s. maculicola ES4326, but GST1 mRNA levels decreased to low levels in leaves inoculated with P. s. maculicola ES4326(aerpt2). In contrast to GST1, GST2 and GST3 were not induced by inoculation. An Arabidopsis gene (A.t. [EL/3) homologous to the parsley ELI3 gene has been cloned (27). A.t. ELI3 is rapidly induced after 20 inoculation with the avirulent isolate P. s. maculicola m2, but is induced later and more weakly after inoculation with virulent isolates. In addition, A.t. ELI3 gene activation co-segregates with RPMf function, supporting the hypothesis that A.t. ELI3 is involved in the resistance response. Genes encoding basic and acidic chitinases, and a chitinase enzyme have been isolated from Arabidopsis (50,69). Both chitinases are encoded by single copy genes. The basic chitinase has 73% amino acid sequence similarity to the basic chitinase from tobacco, and the acidic chitinase has 60% amino acid similarity to the acidic chitinase from cucumber. Expression of the basic chitinase is organ specific and age dependent. A high level of constitutive expression was observed in the roots with lower levels in the leaves and flowering shoots. Exposure of plants to ethylene induced high levels of systemic expression of basic chitinase with expression increasing with plant age. Expression of the acidic chitinase was not observed in healthy plants or plants treated with ethylene. However, a transient expression assay indicated that the acidic chitinase promoter is active in leaf tissue. The expression of the acidic chitinase promoter was studied in transgenic Arabidopsis by fusing the promoter to the beta-glucuronidase (GUS) coding region (51). In healthy transgenic plants, acidic chitinase promoter activity was observed in the roots, leaf vascular tissue, hydathodes, guard cells, and anthers. After inoculation with Rhizoctonia solani, GUS expression was induced in the mesophyll cells surrounding the lesions. The basic chitinase was purified from Arabidopsis and has a 21 molecular mass of approximately 32 kilodaltons and an apparent isoelectric point of 8.7 (69). The purified chitinase inhibited the in vitro growth of the fungus Trichoderma reesei, but not the growth of Alternaria solani, Fusarium oxysporum, Sclerotinia sclerotiorum, Gaeumannomyces graminis, or Phytophthora megasperma. Amino acid sequence analysis revealed that the protein is most likely the product of the basic chitinase gene. A thaumatin-like protein (TLP) and its corresponding gene have been isolated from Arabidopsis (38). Similar to the Arabidopsis basic chitinase, TLP mRNA and protein levels are constitutively high in roots, but low in leaves. The expression of the TLP promoter was studied in transgenic Arabidopsis by fusing the promoter to the beta-glucuronidase (GUS) coding region. Similar to the Arabidopsis acidic chitinase promoter, the TLP promoter-GUS fusion in transgenic Arabidopsis was constitutively expressed in the leaves, hydathodes, and floral organs. Expression of the TLP promoter was also induced by ethephon treatment and R. solani infection. The antifungal activity of TLP was not tested. Uknes et al. have recently demonstrated that leaves of Arabidopsis develop resistance to P. s. tomato and P. parasitica following treatment with dichloroisonicotinic acid (INA) (66). This chemically-induced resistance is associated with the accumulation of three pathogenesis-related proteins (PR-1, PR-2, and PR-S). Genes encoding the three proteins were cloned and found to be induced by INA, pathogen infection, and salicylic acid. Recently, Tsuji et al. reported the purification of a phytoalexin, camalexin, from Arabidopsis (65). The structure was elucidated as 22 3-thiazol-2'-yI-indole on the basis of UV, IR, MS, 1H-NMR, and 13C-NMR data. The phytoalexin accumulated rapidly after inoculation with P. s. syringae and reached maximum levels 24 to 48 hours after inoculation. The purified phytoalexin was found to inhibit the growth of the fungus Cladosporium cucumerinum and the bacterium P. s. syringae. The studies described here demonstrate that Arabidopsis is susceptible to a broad range of pathogens, and that the defense responses of Arabidopsis are similar to those described with other plants. Thus, studies using Arabidopsis may provide insight into the nature of disease resistance genes and the mechanisms by which plants resist infection. REFERENCES 1. Atkinson, M., Huang, J., and Knopp, J. 1985. The hypersensitive reaction of tobacco to Pseudomonas syringae pv. pisi: Activation of a plasmalemma K"/H+ exchange mechanism. Plant Physiol. 79: 843-847. 2. Baker, C., Atkinson, M., and Collmer, A. 1987. Concurrent loss in Tn5 mutants of Pseudomonas syringae pv. syringae of the ability to induce the HR and host plasma membrane KT/HT exchange in tobacco. Phytopathology 77: 1268-1272. 3. Balazs, E., and Lebeurier, G. 1981. Arabidopsis as a host of cauliflower mosaic virus. Arabidopsis Information Service 18: 130-134. 4. Bent, A., lnnes, R., Kunkel, B., and Staskawicz, B. 1991. 23 Molecular genetic analysis of Arabidopsis resistant to Pseudomonas syringae pathogens. Third international congress of the international society of plant molecular biology. Abstract 1229. 5. Berger, M., Handelsman, J., and Goodman, R. M. 1992. Infection of Arabidopsis thaliana by Rhizoctonia solani and suppression of disease by Bacillus cereus. Sixth international symposium on molecular plant-microbe interactions. Abstract 403. 6. Brose, E., Holub, E., Crute, I., and Beynon, J. 1992. Mapping of genes that code for resistance to the fungal pathogen Albugo candida in Arabidopsis thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 228. 7. Burke, 0., Carla, G., and Olson, M. 1987. Cloning of large segments of exogenous DNA into yeast by means of artificial chromosome vectors. Science 236: 806-812. 8. Chang, C., Bowman, J., DeJohn, A., Landers, E., and Meyerowitz, E. 1988. Restriction fragment length polymorphism linkage map for Arabidopsis thaliana. Proc. Natl. Acad. Sci., USA 85: 6856-6860. 9. Daniels, M., Fan, M., Barber, C., Clarke, 8., and Parker, J. 1991. Interaction between Arabidopsis thaliana and Xanthomonas campestris. H. Hennecke and D. P. S. Verma (eds). Advances in Molecular Genetics of Plant-Microbe lnterations. Vol. 1. Kluwer Academic Publishers, Netherlands, pp. 84-89. 10. Darvill, A. G., and Albersheim, P. 1984. Phytoalexins and their elicitors-a defense against microbial infection in plants. Annu. Rev. Plant Physiol. 35: 243-275 11. Davis, K. R. 1991. Arabidopsis thaliana as a model host for 24 studying plant-pathogen interactions. Desh Pal Verma (ed.) Molecular Signals in Plant-Microbe Communications. CRC Press, Boca Raton, pp. 393-406. 12. Davis, K. R., and Ausubel, F. M. 1989. Characterization of elicitor-induced defense responses in suspension-cultured cells of Arabidopsis. Mol. Plant-Microbe Interact. 2: 363-368. 13. Davis, K., Schott, E., and Ausubel, F. M. 1991. Virulence of selected phytopathogenic Pseudomonads in Arabidopsis thaliana. Mol. Plant-Microbe Interact. 4: 477-488. 14. Debener, T., Lehnackers, H., Arnold, M., and Dangl, J. 1991. Identification and molecular mapping of a single Arabidopsis thaliana locus determining resistance to a phytopathogenic Pseudomonas syringae isolate. Plant Journal 1: 289-302. 15. Dong, X., Mindrinos, M., Davis, K., and Ausubel, F. M. 1991. Induction of Arabidopsis defense genes by virulent and avirulent Pseudomonas syringae strains and by a cloned avirulence gene. Plant Cell 3: 61-72. 16. Estelle, M., and Somerville, C. 1986. The mutants of Arabidopsis. Trends in Genetics 2: 89-93. 17. Feldmann, K., Marks, 0., Christianson, M, and Quatrano, R. 1989. A dwarf mutant of Arabidopsis generated by T-DNA insertion mutagenesis. Science 243: 1351-1354. 18. Fletcher, J., and Eastman, C. 1991. Arabidopsis thaliana as an experimental host plant of Spiroplasma citri. Phytopathology. 81: 1209. 19. Golino, 0., Shaw, M., and Rappaport, L. 1988. Infection of Arabidopsis thaliana (L.) Heynh. with a mycoplasma-Iike organism, 25 the beet leafhopper transmitted virescence agent. Arabidopsis Information Service 26: 9-14. 20. Grevelding, C., Becker, 0., Kunze, R., Von Menges, A., Fantes, V., Schell, J., and Masterson, R. 1992. High rates of Ac/Ds germinal transposition in Arabidopsis suitable for gene isolation by insertional mutagenesis. Proc. Natl. Acad. Sci., USA 89: 6085-6089. 21. Grill, E., and Somerville, C. 1991. Construction and characterization of a yeast artificial chromosome library of Arabidopsis. Mol. Gen. Genet. 226: 484-490. 22. Hwang, l., Kohchi, R., Hauge, B., Goodman, H., Schmidt, R., Cnops, 6., Dean, C., Gibson, 8., lba, K., Lemieux, B., Arondel, V., Danhoff, L., and Somerville, C. 1992. Identification and map position of YAC clones comprising one-third of the Arabidopsis genome. Plant Journal 1: 367-374. 23. lnnes, R., Bisgrove, S., Smith, N., Bent, A., and Staskawicz, B. 1992. Genetic mapping of a disease resistance gene in Arabidopsis that is functionally homologous to the RPG1 gene of soybean. Sixth international symposium on molecular plant-microbe interactions. Abstract 343. 24. lshikawa, M., Obata, F., Kumagai, T., and Ohno, T. 1991. Isolation of mutants of Arabidopsis in which the accumulation of tobacco mosaic virus coat protein is reduced to low levels. Third international congress of the international society of plant molecular biology. Abstract 1189. 25. Johal, G., and Briggs, 8. P. 1991. Molecular cloning of the Hm1 disease resistance gene in maize. Third international congress of the international society of plant molecular biology. Abstract 26 1384. 26. Keen, N., Kobayashi, D., Tamaki, 8., Shen, H., Stayton, M., Lawrence, D., Sharma, A., Midland, S., Smith, M., and Sims, J. 1991. Avirulence gene D from Pseudomonas syringae pv. tomato and its interaction with resistance gene Rpg4 in soybean. H. Hennecke and D. P. S. Verma (eds). Advances in Molecular Genetics of Plant-Microbe lnterations. Vol. 1. Kluwer Academic Publishers, Netherlands, pp. 37-44. 27. Kiedrowski, S., Somssich, I. E., and Dangl, J. 1992. A novel Arabidopsis defense gene, A.t.ELI3: Structure and pathogen induced expression. Sixth international symposium on molecular plant-microbe interactions. Abstract 181. 28. Koch, E., and Slusarenko, A. 1990. Arabidopsis is susceptible to infection by a downy mildew fungus. Plant Cell 2: 437-445. 29. Koch, E., and Slusarenko, A. 1990. Fungal pathogens of Arabidopsis thaliana (I.) Heyhn. Botanica Helvetica 100: 257-268. 30. Koornneef, M., Van Eden, J., Hanhart, C., Stam, P., Braaksma, F., and Feenstra, W 1983. Linkage map of Arabidopsis thaliana. J. Hered. 74: 265-272. 31. Kuan, T., Minsavage, G., and Schaad, N. 1986. Aerial dispersal of Xanthomonas campestris pv. campestris from naturally infected Brassica campestris. Plant Disease 70: 409-413. 32. Kunkel, B., Bent, A., and Staskawicz, B. 1992. Arabidopsis mutants with altered resistance to the bacterial pathogen Pseudomonas syringae pv. tomato. Sixth international symposium on molecular plant-microbe interactions. Abstract 349. 33. Last, R., and Fink, G. 1988. Tryptophan-requiring mutants of 27 the plant Arabidopsis thaliana. Science 240: 305-310. 34. Last, R., Bissinger, P., Mahoney, D., Radwanski, E., and Fink, G. 1991. Tryptophan mutants in Arabidopsis: the consequence of duplicated tryptophan synthase beta genes. Plant Cell 3: 345-358. 35. Lazo, G. R., and Ludwig, R. A. 1988. Search for race-specific interactions between bacterial pathogens of crucifers and the plant host Arabidopsis thaliana. Phytopathology 78: 1529 (abstract). 36. Lazo, G. R., and Dixon, R. A. 1992. Screening transformation-competent DNA libraries for plant defense genes in Arabidopsis thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 351. 37. Leisner, S., and Howell, S. 1991. Long distance movement of cauliflower mosaic virus in Arabidopsis thaliana. Third international congress of the international society of plant molecular biology. Abstract 1205. 38. Malehorn, D. E., Verburg, J. G., and Shah, D. 1992. Characterization of a gene encoding a thaumatin-Iike protein of A. thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 355. 39. McCourt, P., and Somerville, C. 1987. The use of mutants for the study of plant metabolism. Davies, 0., (ed.) The Biochemistry of Plants. Vol. 13. Academic Press, New York, pp. 33-64. 40. Meeley, R., Johal, G., Briggs, S., and Walton, J. 1992. A biochemical phenotype for a disease resistance gene of maize. Plant Cell 4: 71-77. 41. Melan, M., Dong, X., Endata, M., Davis, K., Ausubel, F., and Peterman, T. 1992. Lipoxygenase is induced during the Arabidopsis 28 thaliana pathogen defense response. Plant Physiol. 99: 37 (supplement). 42. Melcher, U. 1989. Symptoms of cauliflower mosaic virus infection in Arabidopsis thaliana and turnip. Botanical Gazette 150: 139-147. 43. Meyerowitz, E. 1987. Arabidopsis thaliana. Ann. Rev. Genet. 21: 93-111. 44. Meyerowitz, E., and Pruitt, R. 1985. Arabidopsis thaliana and plant molecular genetics. Science 229: 1214-1218. 45. Nam, H-G., Giraudat, J., den Boer, B., Moonan, F., Loos, W., Hauge, B., and Goodman, H. 1989. Restriction fragment length polymorphism linkage map of Arabidopsis thaliana. Plant Cell 1: 699-705. 46. Palva, T. 1990. Expression of stress response genes in A. thaliana. The fourth international conference on Arabidopsis research. P. 168 (abstract). 47. Parker, J. E., Szabo, V., Staskawicz, B. J., Lister, C., Dean, C., Daniels, M. J., and Jones, J. D. 1992. Analysis of an Arabidopsis gene for resistance to Peronospora parasitica. Sixth international symposium on molecular plant-microbe interactions. Abstract 364. 48. Ramachandra, 8., Lommel S. A., and White, F. F. 1988. Interaction of turnip yellow mosaic virus (TYMV) with Arabidopsis thaliana. Phytopathology 78: 1559 (abstract). 49. Reiter, R., Williams, J., Feldmann, K., Rafalski, J., Tingey, 8., and Scolnik, P. 1992. Global and local genome mapping in Arabidopsis thaliana using recombinant inbreds and random amplified polymorphic DNAs (RAPDs). Proc. Natl. Acad. Sci. 89: 29 1477-1481. 50. Samac, D., Hironaka, C., Yallaly, P., and Shah, D. 1990. Isolation and characterization of the genes encoding basic and acidic chitinase in Arabidopsis thaliana. Plant Physiol. 93: 907-914. 51. Samac, D., and Shah, D. Developmental and pathogen-induced activiation of the Arabidopsis acidic chitinase promoter. Plant Cell 3: 1063-1072. 52. Schaad, N., and Dianese, J. 1981. Cruciferous weeds as sources of inoculum of Xanthomonas campestris in black rot of crucifers. Phytopathology 71: 1215-1220. 53. Schultz, T., and Gabrielson, R. 1986. Xanthomonas campestris pv. campestris in western Washington crucifer seed fields: occurence and survival. Phytopathology 76: 1306-1309. 54. Sijmons, P., Grundler, F., von Mende, N., Burrows, P. and Wyss, U. 1991. Arabidopsis thaliana as a new model host for plant-parasitic nematodes. Plant Journal 1: 245-254. 55. Simpson, R., and Johnson, L. 1990. Arabidopsis as a host for Xanthomonas campestris pv. campestris. Mol. Plant-Microbe Interact. 3: 233-237. 56. Skotnicki, M., Mackenzie, A., and Gibbs, A. 1991. Mutations of Arabidopsis thaliana for tolerance to tymovirus infection. Third international congress of the international society of plant molecular biology. Abstract 1219. - 57. Slusarenko, A., and Mauch-Mani, B. 1991. Downy mildew of Arabidopsis thaliana caused by Peronospora parasitica: a model system for the investigation of the molecular biology of host-pathogen interactions. H. Hennecke and D. P. S. Verma (eds). 3o Advances in Molecular Genetics of Plant-Microbe lnterations. Vol. 1. Kluwer Academic Publishers, Netherlands, pp. 280-283. 58. Somerville, CR. 1989. Arabidopsis Blooms. Plant Cell 1: 1131-1135. 59. Somerville, C., McCourt, P., Caspar, T., Estelle, M., and Keith, K. 1985. Arabidopsis thaliana as a model system for plant genetics and molecular biology. Freeling, M (ed.). Plant Genetics. Alan R. Liss, Inc, New York, pp. 651-660. 60. Sun, T., Goodman, H. M., and Ausubel, F. M. 1992. Cloning the 6A1 locus by genomic subtraction. Plant Cell 4: 119-128. 61. Tor, M., Holub, E., Crute, I., and Beynon, J. 1992. Mapping of genes that code for resistance to the fungal pathogen Peronospora parasitica in Arabidopsis thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 229. 62. Tsuji, J., and Somerville, S. 1988. Xanthomonas campestris pv. campestris-induced chlorosis in Arabidopsis thaliana. Arabidopsis Information Service. 26: 1-8. 63. Tsuji, J., and Somerville, S. 1992. First report of natural infection of Arabidopsis thaliana by Xanthomonas campestris pv. campestris. Plant Disease 76: 539. 64. Tsuji, J., Somerville, 8., and Hammerschmidt, R. 1991. Identification of a gene in Arabidopsis thaliana that controls resistance to Xanthomonas campestris pv. campestris. Physiol. Mol. Plant Pathol. 38: 57-65. 65. Tsuji, J., Jackson, E., Gage, D., Hammerschmidt, R., and Somerville, S. 1992. Phytoalexin accumulation in Arabidopsis thaliana during the hypersensitive reaction to Pseudomonas syringae 31 pv. syringae. Plant Physiol. 98: 1304-1309. 66. Uknes, 8., Mauch-Mani, B., Moyer, M., Potter, 8., Williams, 8., Dincher, 8., Chandler, D., Slusarenko, A., Ward, E., and Ryals, J. 1992. Acquired resistance in Arabidopsis. Plant Cell 4: 645-656. 67. Urban, L. A., Sherwood, J. L., and Melcher, U. 1988. The use of Arabidopsis thaliana cv. Columbia in plant virus interaction studies. Phytopathology 78: 1586 (abstract). 68. Valvekens, D., Van Montagu, M., and Lijsebettens, M. 1988. Agrobacterium tumefaciens-mediated transformation of Arabidopsis thaliana root explants by using kanamycin selection. Proc. Natl. Acad. Sci., USA 85: 5536-5540. 69. Verburg, J., and Khai Huynh, Q. 1991. Purification and characterization of an antifungal chitinase from Arabidopsis thaliana. Plant Physiol. 95: 450-455. 70. Ward, E., and Jen, G. 1990. Isolation of single-copy sequence clones from a yeast artificial chromosome library of randomly-sheared Arabidopsis thaliana DNA. Plant Mol. Biol. 14: 561-568. 71. Whalen, M. C., lnnes, R. W., Bent, A. F., and Staskawicz, B. J. 1991. Identification of Pseudomonas syringae pathogens of Arabidopsis and a bacterial locus determining avirulence on both Arabidopsis and soybean. Plant Cell 3: 49-59. 72. Williams, P. 1980. Black rot: a continuing threat to world crucifers. Plant Disease 64: 736-742. 73. Yu, G.-L., and Ausubel, F. M. 1992. Defense genes expression in wild-type and mutant Arabidopsis during bacterial infection. Sixth international symposium on molecular plant-microbe interactions. Abstract 388. 32 CHAPTER I FIRST REPORT OF THE NATURAL INFECTION OF ARABIDOPSIS THALIANA BY XANTHOMONAS CAMPES TRIS PV. CAMPESTRIS This chapter was published in Plant Disease (1992) 76: 539. 33 First Report of Natural Infection of Arabrdapsis Union by Xantho- monas coupes-vi: pv. man-is. J. Tsuji and S. C. Somerville. MSU- DOE Plant Research Laboratory, Michigan State University, East Lansing 488244312. Plant Dis. 76:539, I992. Accepted for publication [3 November I992. Semioval necrotic and chlorotic lesions were observed on the margins of leaves of the cruciferous weed Arabidopsis thaliana (L) Heynh. growing along a roadside in Kalamazoo. Michigan. A gram-negative. rod-shaped bacterium was isolated from diseased tissue and identified as thrhomonas campestris pv. campestris (Pammel) Dowson by the following criteria; yellow mucoid colonies on YDC. growth on SX. xanthomonadin pigment production. growth at 35 C. protease activity, and ability to cause black rot disease on turnip (Brassica campestris L. subsp. campestris) from Bentley Seed Co. (Cambridge, NY). To fulfill Koch’s postulates, we sprayed guttating leaves of 3-wIt-old A. thaliana race Pro with the bacterium. Symptoms identical to those described above were observed l-Z wk after inoculation, and the same bacterium was reisolated from the diseased leaves. No blackened veins were observed on either naturally infected or artificially inoculated leaves. This is the first report of natural infection of A. thaliana by X. c. campestris and further extends the range of cruciferous weeds that are alternative hosts for this bacterium. CHAPTER II XANTHOMONAS CAMPESTRIS PV. CAMPESTRIS -|NDUCED CHLOROSIS IN ARABIDOPSIS THALIANA This chapter was published in Arabidopsis Information Service (1988) 26: 1-8. 35 Ar- ORIGINAL CONTRIBUTIONS WWW. WIS-mm GAMMA J. TSUJI and 3.6. scrum 3180-“)! Plant Research laboratory. flichigan State University .. East Lansing. MI 48824. USA received Oct. 26. 1988. accepted Nov. Us. 1988 Am Developaant or‘ a aodel Arabidopsis disease is of interest so that the pou- ertul solecular genetic techniques adapted for use with this plant can be used to characterise disease resistance. Fifteen races of Arabidopsis m- liana were tested for susceptibility to the crucifer pathogen Xanthomonas cupastris pv. campestris. Leaves of race Pr-O exhibited a spreading chlo- rosis 72 h after infiltration. Based on assays of chlorophyll loss. race Fr-O appeared sore susceptible to this bacterial pathogen than the other Arabidopsis races tested. This newly discovered host-pathogen interaction say prove to be a useful aodel syste- Eor studying disease developaant in plants. momma Arabidopsis thaliana has may attractive features as a endel organic for classical and anlecular genetics. Arabidopsis has a rapid reproductive cycle and a sell genoae with little repetitive Illa (W et al.. 1986). This crucifer can be transformed using Agrobacteriu—derivad vec- tors (LlDYD at al.. 1986). can be regenerated Erna protoplasts (DA!!! and mm. 1988). and can maintain an active. nice transposable eleaenc (VAN sun's et al.. 1987). These characteristics suggest that Arabidopsis sight be a powerful tool for studying the aolecular basis of disease re- sistance. Xanthoannas c-pestris pv. cups-atria. the causal agent of black rot of any crucifers (WILLING. 19$). has been studied as a nodal organise for investigating the genetics of bacterial pathogenicity (this at al.. 1987). 3y coapleaenting non-pathogenic eutants with frag-ants of the wild type 37 pathogen DNA. a locus for pathogenicity has been identified that restores pathogenicity to the eutsnts (DOE et al., 1987 and SHAH et al.. 1988). Recently. BOUCBER et al. (1987) found hoeology in Xanthnennes with the hrp locus in Pseudoeones snlaneceerue that controls the initiation of disease syeptoes on host plants and the developeent of the hypersensitive. resist- ance response on non-host plants (LINDGREN er al.. 1986). The hrp locus is believed to encode a regulatory protein that controls the expression of genes that elicit plant defenses against bacterial infection (GRIN! and PANOPOULOS. in press). While. a hrp locus has not been identified in Xan— thoennas. an analogous regulatory protein could be encoded by the pathoge- nicity locus. Previously. BALASZ and LEBEURIER (1981) have shown that Arabidopsis is host of the Cauliflower eosaic Virus. In this paper we present evidence that Arabidopsis is also a host of Xanthoennas caepeetris pv. caepestris. Inoculated Arabidopsis leaves turned chlorotic within 72 hours after in- oculation. and the syeptoes snow specificity along different races of Are- bidopsis. This new host-pathogen interaCtion eey prove to be a useful endel systee for studying the enleeular basis of disease resistance in plants. MATERIALS AND METHODS Plggt ggtgrigl The rates of Arabidopsis thaliana utilized in this study are listed in Table 1. Plants were grown in a l:l:l eixture of perlite. vereiculita. and spnagnue (SOHERVILLE and OGREN. 1982) in a growth chaeber eaintained at 25'C for 16 hours during the day cycle (250 "Ee'zs-l) and at 20'C for the night cycle. Arabidopsis plants were grown for 3 weeks before use. while brassita caepestris plants were grown for & weeks before use. Bagggrgs Spontaneous rifaepicin-resistant autents of Xanthnennas caepestris pv. ceepestris (reportedly strain 86-1 II. a derivative of 20250 described in SHAH and um. I986) were selected on YDC agar WILSON et al.. 1967) con- taining 100 ug rifeepicin/el. Colonies were streaked onto a second rife- picin plate. transferred to liquid eediue 523 (DflINCRA and SINCLAIR. 1985) containing 50 ug rifaepicin/el, and grown overnight at 30‘C (shaken at 200 38 revolutions per dance) in the darts. W syringae p‘r- syringae (strain P883220 described in Wand NIGHT. 1983) was transferred Ernetings Bandit. agar-(W. 19E!) totingsbbrothandgrownover- night at 30‘C (m at 200 revolutions per einute). Table l- Arabidopsis thaliana races. Arabidopsis race source flcb-l 3.2. Irena 31.-A. A3. {rant Der-O 521" “I168 (Iii-l LR. Irena Colt-bis G. Redei Colubia-glabroue 6.3. We 01-0 LR. trans Estland R. Schull La-O A3. (ran: Landsberg (erecta) r‘l. loomeaf Montcal- F. Lehle Nudes-sens E. Meyerowitz Pr-O LR. Iran: lechew 6.1. Peldnn Vesailewauja 6.1. Feld-n Z | o' . I Broth culture was centrifuged for Z ainutes at high speed on a Fisher si- cro-centrifuge (aodel 2353). The supernatant was decanted. and the pellet was resuspended in IO all potassiue phosphate buffer pi! 6.9. The bacterial suspennonwas weshedwith bufferasecnnd ties. and theinocultnconcen- tration. detersined by eaaauring the absorbsnce of the bacterial stupen- sion at 640 ne using a Perkin-Elsa Laebda 33 uv/vis spectrophotoeeter. was adjusted to l x 108 cell/al. " of Asinglesiteoneachoftwoleavesperplantwesinoculatedusinga lsl plastic syringe without needle by the cached of 601m at. al. (1987). About to ill. of the inoculu- was infiltrated through the stoeates by gently applying force on the plunger. Excess inoculum was blot-dried free- the surface of the leaf. and the petioles of infiltrated leaves were mrlted using a Sharpie pen. Control leaves received'no treatment since- buffer was found to have no effect on the leaf chlorophyll content. 3%? £bln:222zll.ssassssssats At daily intervals. five inoculated Arabidopsis leaves were harvested. Leaf discs were rcoved free the inoculation site and were extracted for chlorophyll in It”: eethanol (1 el per disc) at 50'C for 20 einutes. The absorbanceof tbesolutionwaseaasuredet666neand653ne usingaPer— hin-Eleer spectrophotoeeter. and the assent of chlorophyll was deter-med using the foreule of um and MN (1983). Each experieent was repeated three tines. and eeen chlorophyll values were detersined in eicrograes per square centieeter of leaf tissue. Mean chlorophyll values for inoculated plants were standardized by dividing by the ease chloro- phyll values of control plants seepled the sees day, and expressing the result as a percentage. called the chlorophyll content. The chlorophyll contents were subjected to an analysis of variance. and eean chlorophyll contents. averaged over 8 days of assume-ant. were coepsred using Fisher's least significant difference test. km AND DISCUSSION ézsnsse_2szslsissat Leaves of Arabidopsis race Pr-O infiltrated with X.c. c-pestris (c. 10 cells/e1) showed an initial water seeking that cleared within 26 hours. Leaf yellowing was visible 72 hours after inoculation which subsequently spread beyond the inoculated area (Figure la). This was soon followed by browning and cell collapse within the inoculated region (Figure lb). In contrast. leaves infiltrated with Panda-nuns syringae pv. syringae (c. 108 cells/e1) showed no yellowing. Leaves infiltrated with P. s. syringae 'turned light brown—grey within 68 hours. and the necrosis did not sprad beyond the inoculation site. leaves infiltrated with 10 it potassiue phos- phate buffer pH 6.9 also showed an initial water snaking but otherwise had the seas appearance as uninoculated leaves. 8 Thediseeeesyepto-aeen onArabidopeis weresieilar tothoeeobserved whenlreesicaeqeecris.altnovmhost.wesinoculatedwith1.c.m- riabytheesee eethod.hrassics leavesshowed brightyellow'lng atthe infiltration site 72 hours after inoculation which later turned brown and necrotic (m at al.. 1987). Incontrsst tothese syeptoea.when Brassica leaves were inoculated with P. s. syringae leaf yallowingwes absent. Instead. the infiltration site turned dark grey-within.2£.hours. Figure l. Syaptoa developsent of Arabidopsis race Pr-O leaves 5 days after inoculation with X.c. c-pestris: initial leaf yellowing, a. and spreading chlorosis. b. 1+1 and the necrosis was contained within the inoculation site. Therefore. the X. c. e-pastria-induced disease reaction could be distinguished froa the hypersensitive. resistance response by the presence of spreading chloro- sis. Froa an infected Arabidopsis leaf we recovered a bacteria that was identified as the inoculated Ishthuonad by the following criteriairod- shaped: gran negative: rifaapicin resistance: yellow. mtoid colony on m agar: non-fluorescence on Kings 3 agar: and growth on S! agar (SGMD. 19”). This bacterius was able to induce the saae syaptoas in a second inoculation. Further-ore. aetabolically active bacteria were necessary to induce syaptoas; leaves inoculated with bacteria that were hast-killed in a boiling water bath showed no yellowing. m... results suggest that X.c. c-pastris was the infectious agent that caused leaf yellowing and necro- sis. W In addition to M. In other Arabidopsis races were screened for suscept- ibility to X.c. c-pestris. All the races responded siailarly to infection with X.c. caapeatris: however. the chlorosis appeared to have spread sore rapidly on Arabidopsis race Pr-O. 0f the 15 races asaained. Colubia. lur- 0 and Pr-O were selected for further study. In order to quantitata the responses of the three races of Arabidopsis to infection with X.c. c.- paatris. a aethod based on chlorophyll loss was developed. his analysis is based on the assuaption that sore susceptible races of Arabidopsis lose chlorophyll faster than less susceptible races. Since the total leaf chlo— rophyll value varies asong the different races. one race -y yellow sore quickly than another yet both aay lose chlorophyll at the ease rate. Thus a subjective. disease rating scale for deteraining relative susceptibility is insufficient. Polynoaical regression analysis of the chlorophyll con- tent of inoculated leaves relative to the controls over a seven day period gave a quantitative criterion for coaparing the responses of different races to inoculation with X.c. W. has resulting chlorophyll pro- files suggest that the races Col-bis and hero responded very siailarly to inoculation with X.c. c-pastris. but the race Pr-O was sore suscept- ible than both Coll-his and Bur-0 (Figure 2). Using Fisher's least signi- ficant difference test we found that the aasn chlorophyll content of Colubia over the seven day period was not significantly different froa Bur-0: however. the aaan chlorophyll content of Pr-O was significantly different froa both Colt-bis and Bur-O (Table 2). Thus. the race M was statistically sore susceptible than both Coll-bis and Bur-0. l+2 lii§| an . - . w o 2 4 a a Day after "'00."!th Chlorophyll contem (percent 0! control) Figure 2. Chlorophyll profiles of three races of Arabidopsis inoculated with X.c. caspastris. Chlorophyll values (pg/ca‘:$E) of the controls at day 0 were 63.77 3 0.53 (Bar-0). 57.923038 (Calm-us). and 50.22 31.56 (Pr-0). Table 2. Mean chlorophyll contents for Arabidopsis races Coll-bis. Bar-0. and Pr-O. Arabidopsis race seen chlorophyll content (percent of control) Coll-bis 80.63a‘ Bur-0 79.3“ H 65.“!!! Chlorophyll contents with the seas letter are not significantly different at the 0.01 level using Fisher‘s least significant difference test. Experieents are now in progress to deter-ins the inheritance of the sus- ceptibility observed in Pr-O. If this response proves to be a siaply in- herited trait. then by mutagenesis. we my be able to identify the plant cosponents that initiate coapatihle reactions. 43 AW The authors than): A. [LEAN for her assistance with the growth of the cul- tures. Photographs were taken by I. ST'EPNITZ. Instructional Media Center. Michigan State University. East Lansing. Michigan 6882‘. USA. Financial support was provided by 0.5. Depart-ant of Energy (0W-M—EROT338). REFERENCES M1482. E. and 6. mm: Arabid.Inf.Serv.(I-'ramtfurt a Main) E. 130- 131- (1981) sum. C.A.. R. VAN GUST-Sal. A. WIS. H. AILAT’ and C. ZISCHEI: J. of Bacteriol L62. 5626—5632 (1987) COLLINS. 0.6.. 0.5. HILLIGAN and J.H. M: Plant 'Hol. Biol. _8_. 605-616 (1987) DAIOI. B. and L. W: Hol. Gen. Genet. _2_T_3. lS—ZO (1988) mum. 0.0. and J.B. SINCLAIR: Basic plant pathology aathods. CIC Press. Inc. Boca Raton. Florida p.315 (1985) W. J.H.. G. SQFIELD. K. “AFFORD. P.C. mm and H.J. DANIELS: Phy- siol. Hol. Plant Pathol. ll. 261-27] (1987) mm. C. and N.J. PANOPOUwS: The predicted protein of a pathogenicity locus froa Pseudoaonas syringae pv. pinaaolicola is hoaologous to a highly conserved doaain of several prokaryotic regulatory pro- teins. In press (1988) W. L.S.. 3.x. Nil-EVANS and E. m: Hol. Gen. Genet. fl. 15-23 (198‘) mm. 0.1. and A.R. m: Biochn. Soc. Transactions fl. 591-592 (1983) LINWJ’JH R.C.PEET and mam: J.Becteriol.jfl.512—522 (1986) LlDYD. A.H.. A.R. MASON. 5.6. acorns. I'LC. m. R.T'. FRAYLET. and 3.3. MSG: Science A. M (1986) SCHAAD. N’.U.: Laboratory guide for identification of plant pathogenic bacteria. Aaer. Phytopathol. Soc. St. Paul. Minnesota ; (19K)) SHAH. J.J. and C.I. KAN: Biotechnology A. 560-565 (1986) SHAH. J.J.. L.G. 5m and (2.1. KAN: Hol. Plant-Microbe Interactions l. 339-55 (1988) mm. (2.1!. and 0.1... com: Isolation of photorespiration sutants in ~Arabidopsis thaliana. H. 30mm. 8.8. HALLICX. NJL CHUA. eds. Methods in Chloroplast Molecular Biology. Elsevier. Aaaterdas. pp. 129-138 (1982) mac. J.L.. (2.1.. m. C.£. BARBER. J.H. W and H.J. mums: Hol. Gen. Genet. 3.12: “3468 (1987) VAN SUITS.H.A.. J. ms and N. m: nno Journal 9. 3881-3889 (1987) VINE“. JJI. and 0.8. WIGHT: J. Bacteriol. 12,. 139—1351 (1983) WILLIAMS. PJL: Plant Disease 2". 736—71-2 (19$) VILSQI. 2.2.. RH. 22mm and 0.1.. mxcxson: Phytopathology S_7. 618- 621 (1967) CHAPTER III IDENTIFICATION OF A GENE IN ARABIDOPSIS THALIANA THAT CONTROLS RESISTANCE TO XANTHOMONAS CAMPES TRIS PV. CAMPESTRIS This chapter was published in Physiological and Molecular Plant Pathology (1991) 38: 57-65. all 45 W and Molecular Plant Faraday (I99l) 38. 57—65 57 Identification of a gene in Arabidopsis thaliana that controls resistance to Xanthomonas campestris pv. campestris jun 131111.71 SHAUNA C. Sowenvrtnz.1't§ and RAYMOND Hmumcmanfi t MSU-DOE Plat Ram/r Laboratory. :Gacn'c: Program. and §Doperouu of Betas} and Plant Pathology, Michigan Sm: Um. Earl lasing, Mic/tiger W4. U.S.A. (Accepted for publication December 1990) Xanthomonas memoir pv. campestris caused black ror disease when inoculated via the hydathodes and caused a spreading chlorosis when infiltrated into the leaves of Arabidopsis thaliana race PrO. In contrast. leaves of Arabidopsis race Columbia were resistant to .l’. c. mm: and were slightly chlorotic or symptomless when inoculated by either method of inoculation. By genetic analysis. we found that resistance to X. c. rampant: was determined by a dominant allele of a single. nuclear gene. INTRODUCTION Disease resistance genes have been identified in many agronomically important crop species; however, no disease resistance gene has ever been cloned. Most strategies for cloning genes involve prior knowledge of the gene’s mRNA or prOtein product. Such approaches for cloning disease resistance genes are precluded because the products of disease resistance genes are unknown. An alternate approach is to closely map disease resistance genes relative to RF LP markers and then to use these RF LP markers as starting points to chromosome walk into the gene of interest. However, the application of this strategy is limited because mOSt plants have relatively large genomes, contain repetitive DNA. lack a detailed genetic map of morphological and RFLP markers, and lack an elficient transformation system. Arabidopsis thaliana (L.) Heynh.. a small weed in the family Cruciferae, is one of a few plants for which chromosome walking is feasible. Arabidopsis possesses a small genome (approx. 100 Mb) with little interspersed repetitive DNA [10, 20]. In addition to a map of morphological markers [14], two independent RFLP maps consisting collectively of about 200 markers have been constructed for Arabidopsis [4, 21]. Also, chromosome walking in Arabidopsis is facilitated by the recent construction of several yeast artificial chromosome libraries [12, 38]. Since Arabidopsis can be transformed and re-generated [18. 37], subclones from yeast artificial chromosomes, which span the gene of interest, can be identified by the ability to complement the recessive phenorype. No Other plant Address correspondence to: S. C. Somerville. MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, Michigan +8824. U.S.A. Abbreviations used in text: cfu. colony fornnng units: RF LP. restricnon fragment length polymorphism. 0885—5765/9l/0l0057+09 $03.00/0 Q [99! Academic Press Limited 58 Jun Tsuji at al. species offers these advantages for cloning disease resistance genes by chromosome walking. . Several investigators [15, 25, 26] have reported that cruciferous weeds are hosts for Xanthomonas campestris pv. campestris (Pammel) Dowson, the causal agent of black rot disease [39]. Previously [31, 36], evidence was presented that A. thaliana is also a host for this pathogen. In this paper, we describe resistance in Arabidopsis to X. c. campestris and report on the inheritance of this trait. MATERIALS AND METHODS Plant material Arabidopsis thaliana races Columbia and PrQ were grown as previously described [.13] in a growth chamber maintained at 25 °C for 16 h during the light cycle ( 250 ME. m" s") and at 20 °C for the night cycle. Crosses were performed as described [33]. Bactcria The following bacterial strains were used in this study: Xanthomonas campestris pv. campestris 2D520 [28], rifampicin resistant, cauliflower pathogen; Xanthomonas campestris pv. campestris jSl 11 [30], nonpathogenic mutant of strain 2D520, rifampicin and tetracycline resistant. BaCteria were grown overnight in liquid medium 523 [9] in the presence of the appropriate antibiotic at 30 °C on a rotary shaker (200 revolutions min“) in the dark. Preparation of inoculum The liquid culture of baCteria was centrifuged at 13000 g for 30 s. The supernatant was decanted, and the pellet resuspended in 10 mu p0tassium phosphate buffer, pH 69. The baCterial suspension was washed with buffer a second time, and the inoculum was adjuSIed to the desired concentration by measuring the absorbance of the bacrerial suspension at 640 nm [24]. Inoculation: Leaves of Arabidopsis were infiltrated with inoculum as described [5]. Excess inoculum was blatted from the surface of the leaf, and the petioles of infiltrated leaves were marked using a Sharpie” pen. Hydathode inoculations were performed as described previously [29] or by placing plants in a 23 °C dew chamber overnight, allowing gutation droplets to form. and then lightly misting leaves with a suspension of X. c. campestris 2D52O at a concentration of IX 10‘ cfu ml“. Pats were covered with a transparent plastic bag to maintain high humidity, and the plants were returned to the growth chamber. Bacterial growl/t stud) Leaves of Arabidopsis were infiltrated with a suspension of X. c. campestris 2D520 at a concentration of l x 10" cfu ml". At 0, l, 3, and 5 days post-inoculation, an oval leaf disc (044- cm’) was removed from the inoculation site, rinsed three times in sterile disdlled water, and macerated in 1 ml of 10 mar potassium phosphate, pH 6-9. Viable counts were determined by plating serial dilutions of the leaf extract on YDC [42] containing 100 pg rifampicin ml“. Values in Fig. 4 are the mean and standard error 1+7 Single gene resistance to X. c. campestris 59 of the mean of six replicates. Similar results were obtained in two additional experiments when leaf discs were surface sterilized in 0'5 °/o sodium-hypochlorite for l-‘Z min before maceration. Chlorophyll mearurernenb' Leaves of Arabidopsis were infiltrated with a suspension of X. c. campestris 2D520 or X. c. campestris JSl ll at a concentration of 1-5 x 10' cfu ml“. At 0, 2, 4, and 6 days post- inoculation, a leaf disc (044 cm’) was removed from each of two inoculated leaves and two uninoculated leaves of comparable age from the same plant. Chlorophyll was extracted from the two samples of leaf dises in 100% methanol (1 ml per disc) at 50 °C for l h. Afterwards, . u and A“1 of the methanol solution were recorded. The amount of chlorophyll was determined using the formula of Lichtenthaler & Wellburn [17], and the amount of chlorophyll of the inoculated leaf discs was expressed as a percentage of the amount of chlorophyll of the uninoculated, control discs. Values in Fig. 2 are the mean percentages of 10 inoculated plants per time point. Quantitative inheritance stud] Six days after inoculation with a suspension of X. c. campestris 2D520 at a concentration of 1—5 x lO'I cfu ml“, leaves of Pr0, Columbia, the F,, and the F._. (a toral of 768 leaves) were extracted for chlorophyll as described above. Genetic effecrs and the degree of dominance were estimated [[9], and the minimum number ol‘efiective factors was estimated using the formulae described by Lande [16]. RESULTS Previously, 15 races of Arabidopsis were screened for resistance to X. c. campestris [.36]. Of the races examined, Columbia and PM were selecred for further study. When leaves of Arabidopsis race PrG were infiltrated with X. c. campestris (1-5 x [0‘ cfu ml“), chlorosis was visible by 6 days after inoculation and subsequently spread beyond the infiltration site. This was later followed by browning and cell collapse within the inoculated region (Fig. IA). During the same period, leaves of Columbia infiltrated .‘tbfieké‘~ 3,..— -::)‘-‘A_k "\—.‘ # Fin. 1. Symptoms on leaves of (A) M and (B) Columbia 6 days after infiltration with a suspension of X. c. campestris at a concentration of 1-5 x 10‘ cfu ml“. Arrows indicate healthy : H), chlorotic (C), and necrotic (Ni tissues. 60 Jun Tsuji er al. ‘ “i?"‘ii‘T-‘f‘if‘il Chlorophyll conlanl lpercsnl ol control) Time post inoculation (days) FIG. ‘2. Chlorophyll contents of leaves of Arabidopsis inoculated with a l-3x IO‘cfis ml“ suspension of X. c. campestris. Solid bar: Columbia inoculated with X. c. caspmm 2D520: Stippled bar: PrO inoculated with the black rot mutant X. c. WmJSl l l ; Open bar: PrO inoculated with X. c. campestris 2D520. Values are the mean percentages of ten inoculated plants per time point. Standard error bars are shown. Chlorophyll contents tug cm“ iSEM) of the uninoculated control leaves at day 0 were “II-47:05:! for Columbia; l7-72to-4-2 for Pr0, X. c. campestris 2D520; and 190710-71 for PM, X. c canpmrisJSlll. with the same concentration of bacteria showed only slight chlorosis or were symptomless (Fig. [3). Leaves infiltrated with phosphate buffer or heat-killed bacteria were symptomless. Analysis of the chlorophyll content of inoculated leaves relative to the uninoculated control leaves over a 6 day period gave a quantitative criterion for comparing the responses of the two rades to inoculation with X. c. campestris. Based on the resulting chlorophyll contents, Columbia was clearly more resistant to X. c. campestris-induced chlorosis than PrQ (Fig. 2), confirming our visual observations of the differential responses of PrG and Columbia. By this method of inoculation, chlorosis was reproducibly observed in more than 95 % of all inoculated leaves of PrO while extensive chlorosis was never observed in inoculated leaves of Columbia. Hydathode inoculations of Arabidopsis race PM with X. c. campestris produced symptoms similar to those observed on infected cabbage plants in the field. One to two weeks after inoculation, symptoms appeared along the leaf margin as angular, necrotic lesions bordered by a yellow halo (Fig. 3A). Leaves later wilted. In contrast, leaves of Columbia showed only a slight marginal chlorosis or were entirely symptomless when inoculated by the same method (Fig. 33). By this method of inoculation, about 10% of all inoculated leaves of 80—120 Pr0 plants developed black rot disease while black rot was never observed on inoculated leaves of 80—120 Columbia plants. To determine whether a pathogenicity locus of X. c. campestris that is required to cause black rot on cauliflower is also necessary to induce chlorosis in Arabidopsis, we inoculated Arabidopsis race PrQ with X. c. campestris strain JSlll. Strain JSlll is a transposon-induced mutant of X. c. campestris 2D520 that is deficient in the ability to cause black rot on cauliflower [30]. When a suspension of strain JSl ll at a Single gene resistance to X. c. campestris 61 concentration of l x 10‘ cfu ml“ was infiltrated into leaves of Pr@, the plants were found to be nearly symptomless 6 days after inoculation (Fig. 2). The small loss of chlorophyll that was detected at day 6 may be due to the instability of Tn443l in planta [30]. By 6 days after inoculation, we found that 32: l 9.1, of the bacreria had lost the transposon as determined by antibiotic resistance. By losing the transposon, the bacteria likely regained the ability to induce the chlorotic reaction. In addition, black r0t disease was never observed on Arabidopsis when strain JSI 11 was inoculated via the hydathodes [13]. To further characterize the differential response of PM and Columbia to inoculation with X. c. campestris, we conducted an in viva growth study. Leaves of Arabidopsis races Fri?) and Columbia were infiltrated with a suspension of X. c. campestris 2D520. In three separate experiments, we observed that the baCterium multiplied 2-3 orders of magnitude over a 5 day period, and that the growth of the bacterium was nor significantly different in PrQ than in Columbia (Fig. 4). Thus, symptom development was not correlated with bacterial growth. Fin. 3. Symptoms on leave of (A) PM and (B) Columbia ‘2 weeks after inoculation with X. c. campestris via the hydathodes. Arrows indimte healthy (H), chlorotic (Cl, and necrotic (N) tissue. L on lclu cm'al rirne nest-inoculation (days) Fin. +. Time course of growth of X. c. campestris infiltrated into leaves of (D) M and .I) Columbia. Each data point represents the mean of six replicates. Standard error bars are shown. Similar mults were obtained in two additional experiments. 50 62 Jun Tsuji at al. To determine the genetic basis for the differential response to X. c. campestris, Columbia and PrQ) were reciprocally crossed and the F1 was evaluated visually for symptom development following inoculation by the leaf infiltration method. We observed that the response of all the F1 resembled that of the Columbia parent, and we detected no maternal efi'ect. F , plants were selfed, and the F2 was scored for either the Columbia-type of response (resistance) or the PrO-type of response (susceptibility). We observed that resistance to X. c. campestris-induced chlorosis segregated in a manner consistent with a 3:1 ratio (331 resistant, 104 susceptible; Z" =- 0°277, P > 0-5). The inheritance pattern was also confirmed in testcrosses between the F1 and Pr@ (data not presented). These results support the hypOthesis that Columbia and PrO differ in their response to X. c. campestris at a single, nuclear gene, and that resistance is encoded by a dominant allele. We have named this gene RXCI (reacrion to X. c. campestris), and according to convention [2, 3], we have designated the dominant resistance allele in Columbia as rch-l and the recessive allele in PrQ as rxc1-2. In addition, the quantitative inheritance of resistance in Columbia was determined using measurements of chlorophyll content 6 days after inoculation of the two parental populations, the F,, and the F 3 (Table 1). From these data, we estimated the additive TAIL! l .Mmdeo/WmdchlaefiyflmmfwAnbidopsnmmu-de X. c. campestris Arabidopsis No. of plants Chlorophyll content‘ PrO 30 30-75 1- 547° Columbia 30 8630 j; 244 F, 30 93-89 1 3°25 F, [02 69-92 1 2°69 ‘Percent of control. ”Mean and standard error of the mean. genetic effect to be 27°77, the dominance genetic efi'ecr to be 35-36, and the degree of dominance to be 127. By using two related Castle-Wright index (as) formulae, the minimum number of segregating genes (effective factors) was estimated to be 092 or 124 (Table 2). By this independent, quantitative genetic test, we confirmed that a Taste 2 Females/or Castle-Wright indicies 'jo and n, oalaesjor the resistant number ofgcnes reflective/actors) controlling the response of Arabidopsis to inoculation with X. c. campestris Formula‘ n8 gP, -P,)‘/8lV,,-V,,, 0-92 (P,-P,;‘/8[V,,—l l/2\'n+ l/QVm + l/4V,,',] l-‘24 ‘P, and I"2 indicate the mean chlorophyll content ol'PrO and Columbia. respectively. V", V”, V", and V" indicate the variance associated with the mean chlorophyll content of PrO. Columbia. the F1, and the F2, respecrively. 51 Single gene resistance to X. c. campestris 63 single, major gene determines the differential response Of Arabidopsis races PrO and Columbia to inoculation with X. c. campestris, and that resistance to X. c. campestris is determined by a completely dominant allele. DISCUSSION Shaw 8t Kado [3] have recommended wound inoculations of petioles or hydathode inoculations for experimental studies of the black rOt disease. In their experiments, only these two methods consistently gave rise to vein blackening and dry wt symptoms charaCteristic Of the black rOt disease. Simpson & johnson [31] reported that about 60 % Of all wound inoculated leaves OfArabidopsis never developed disease. In addition, we found that about 90% of all leaves inoculated via the hydathodes escaped disease. Thus, hydathode and wound inoculations are unsuitable for screening individuals in genetic experiments. In contrast, leaf infiltration gave highly reproducible results and has been successfully used as a screen for X. c. campestris mutants with altered pathogenicity [8. 22]. Although the method Of infiltration into attached leaves is nOt considered a typical route of natural infecrion, we believe that the phenotype we observed is a reliable indicator of disease because chlorotic mesophyll tissue is one of the symptoms of black rOt disease. the non-pathogenic mutant JSI l I did not cause chlorOtic lesions on Arabidopsis race PrO. and resistance to X. c. campestris in Arabidopsis race Columbia was observed by bOth methods of inoculation. When the growth Of X. c. campestris 2D520 was examined, we observed that the growth curves in Arabidopsis were similar to those reported in cabbage [7] and turnip [11, 35]. We also observed no significant difference in bacterial growth between PrO and Columbia. Thus, symptom development was nOt correlated with baCterial growth. The lack Of correlation between bacterial growth and symptom development has been observed with Other bacterial plant pathogens. Smidt 8t Kosuge [32] observed that indole acetic acid-deficient strains Of P. s. saoastanoi failed to produce galls on Oleander but exhibited similar growth patterns as the gall-eliciting wild-type strain. In addition, Erwinia chrysantlmni [23] and E. carotovora [6] were able to multiply as well in aerobic, symptomless pOtatO tubers as in tubers that were incubated anaerobically and became macerated. Willis et al. [41] observed that a Tn5-induced mutant of P. s. gin-ingot. deficient in the ability to cause disease on bean pods or leaves, was still able to attain wild-type levels of growth within or on bean leaves. Osbourn ct al. [22] found that pathogenicity mutants of X. c. campestris multiplied in leaves of turnip at the same rate as- the wild—type, pathogenic baCterium. Furthermore, Stall 8: Hall [34] found that X. c. vesicatoria multiplied equally well in chlorosis-prone pepper plants as in chlorosis- resistant plants. These results suggest that the ability to multiply in plant tissues is not closely tied to symptom induction. A limited number Of crucifer genotypes resistant to the black rOt disease have been identified. Bains [1] and Williams et al. [40] have described resistance in the cabbage cultivar, Early Fuji, which has been introduced into commercial lines. This resistance appears to be encoded by a single gene, designated f; the dominant allele encodes resistance. The resistance in Early Fuji is associated with the development of small necrOtic lesions at the leaf margins, limited bacterial growth, and the expression of the hypersensitive response. In cauliflower, the genetics Of resistance to black rOt disease is 52 64 Jun Tsuji at al. less clear. Resistance was reported to be a dominant character; however, Sharma et al. [27] reported that resistance in cauliflower was a polygenic trait rather than a simply inherited charaCter. The relationship between f and RXCI is unclear at this time. In conclusion, we have demonstrated that resistance in Arabidopsis race Columbia to X. c. campestris is controlled by a dominant allele of a single, nuclear gene. We are currently mapping RXCI relative to morphological. and RF LP markers in prelude to cloning this gene by chromosome walking. The authors thank C. I. Kado (University of California, Davis) for providing the strains of X. c. campestris, K. Stepnitz (Instructional Media Center) for phOtography, and E. Light for dew chamber set-up and growth chamber maintenance. Columbia and PrO seeds were kindly provided by G. Redei (University of Missouri, Columbia) and A. Kranz (BOtanisches Institut,j. W. Goethe-Universitat, Frankfurt), respectively. The financial support Of the US. Department of Energy (DE-AC02-76-ER01338) and the Michigan Agricultural Experiment Station are gratefully acknowledged. REFERENCES l. Bum. D. C. (1955). Resistance of cabbage to black rOt. Phytopathology 45. 35-37. 2. Btuxaa. A. 8.. Beretta. M. .A.. Sousavrtu. C. R. St Kenna. H. {1988). Insensitivity to ethylene conferred by a dominant mutation in Arabidopsis thaliana. Science 241. 1086-1089. 3. Bowsuur. J. L. Sum. D. R. St Msmowrrz. E. M. 11989). Genes directing flower development in Arabidopsis. Plant Cell 1, 37-52. 4. Crown. C., Bowman.) 1... DeJOittt. A. W., Larson. E. S. & Msvsaowrrz, E. M. ( 1988). Restriction fragment length polymorphism linkage map for Arabidopsis thaliana. Proceedings oft/re National Academy of Sciences. (1.3.14. 85. 6856-6860. 5. COLLINOI. D. 3.. MILUGAN. D. 15.. Dow.J. M.. Soomto, G. & Dawns. M.J. (1987). Gene expression in Brassica campestris showing a hypersensitive response to the incompatible pathogen Xanthomonas campestris pv. oiliaas. Plant Molecular Biology 8. 405—114. 6. Cousin. .A. 8: Kurt. N. T. (1986). The role of pecuc enzymes in plant pathogenesis. Annual Review of Plgrtopetltolog 24. 383-409. 7. Coon. D. R. 8: Roasson. D. 1. (1986). Active resistance of cabbage (Brassica oleracea) to Xanthomonas campestris pv. carotae and protection against the causal agent of black cat Xanthomonas campestris pv. cantpestrit. by co-inoculatron. Physiological and Molecular Plant Pathology 28. «ll—52. 8. Dunsts. M. J., Banana. C. E.. Toma. P. C.. Cunt. W. G. St waczvc. M. K. (1984). Isolation of mutants of Xanthomonas campestris pv. campestris showing altered pathogenicity. Journal of General Microbiology 130, 2447—2455. 9. Damon. O. D. & Sarcasm]. B. (1985). Basic Plant Pathology Methods. p. 315. CRC Press. Inc. Boca Raton, Florida. 10. Errata. M. A. St Sounvrtts, C. R. (1986). The mutants of Arabidopsir. Trends in Genetics 2, 89-93. 11. Conan. C. I... Dow.J. M.. Bmu. C. E. & Dunes. M. J. (1988). Cloning of two endoglucanase genes of Xanthomonas campestris pv. campestris: Analysis of the role of the major endoglucanase in pathogenesis. Molecular Plain-Microbe Interactions 1. 275-281. 12. Grunt. E. (St Sounvrtu. C. R. (_ 1991). Construction and characterization of a yeast artificial chromosome library of Arabidopsis. Molecular and General Genetics (in press). 13. KAIOUN. S. &. Kano. C. I. if 1990). A plant-inducible gene ofXant/tontonas cantpestris pv. campestris encodes an exocellular component required for growth in the host and hypersensitivity on nonhosts. Journal of Bacteriology 172. 5165-5172. 14. Kooarmur. M.. van EosN.J.. Hartman C.J.. Strut. P.. am F.J. & Funerals. W. F. (1983). Linkage map of Arabidopm thaliana. Journal of Heredity 74. 265-272. 15. KHAN. T. I... MINMVAGE, G. V. St Scum, N. W. (1986). Aerial dispersal OfXantItontonar comm} pv. campestris from naturally infected Bram'ca campestris. Plant Dictate 70. +09—ll3. 16. Lanna. R. .j 1981). The minimum number of genes contributing to quantitative variation between and within populations. Genetics 99. 541-553. 17. Lacuna-rm. H. K. St WELLBCRN. A. R. {1983). Determinations OftOtal carotenoids and chlorophyll: a and b of leaf extracrs in dilferent solvents. Biochemical Society Transactions 603. 591-592. 53 Single gene resisranco to X. c. campestris 65 18. I9. 20. 21. 24. 26. 27. 28. Ill. 32. 33. 34. 35. 36. 37. 39. .Wtusanss. P. H.. Sraua. T. & SU'r'ronn.J. C. (1972). Inheritance ofresistance in cabbage to black rot. 41. 42. Lwrn. A. M.. Bananas. .A. R.. Roons. S. C., Ems. M. C.. Peanut. R. T. & Hansen. R. B. ( 1986). Transformation of Arabidopsis thaliana with Agrobacterium nuns/adorns. Science 236. 464—466. Marissa. K. &Jttnss.J. L. (1971). [sand-cools lo Bionnetrical Genetics. Cornell University Press. Ithaca. NY. Mavsaowrrz. E. M. (1987). Arabian: thaliana. Annual Review clam 21. 93—111. Nani. H.-G., Gmnnxr. J.. our Bonn. 8.. Moonan, F., Loos. W. D. 8.. Hanna. 8. M. a Goonnuur. H. M. (1989). Resnicnon fragment length polymorphism linkage map ofdraon'dopn': thaliana. Plant Cell 1. 699-705. . Osaouast. .A. E.. Guiana. B. R. & Danna. M.J. (1990). Identification and DNA sequence of a pathogenicity gene of Xanthomonas was: pv. car-wan. Molecular Plains—Microbe Interaction: 3. 280-285. . PamJ. H., Scisosnst..C., Kurt,N. T. &Cou.nssa.A. (.1987) Multiplicanonandvirulcnceinplant tissuesofErc/tnioh’acoliclonesprnducing pccratelyaseisozymes PLband PLeathighlevelsandofan Erwinia durum mutant deficient in PLc. Applied and Environ-nasal Microbiology 53. 2315-2320. Senna. N. W. (1980). laboratory Gsidefor Ute Identifications ofPlal Pathogenic Banana. p. 91. American Phytopathological Society. St. Paul. Minn. U.S.A. . Scsuun. N. W. & DWISI. J. C. (1981). Cruciferous weeds as sources ofinoculum of Katha-nan: cannon: in black rot ol'crucifers. Phytopathology 71.1215-1220. Set-wan, T. & OWN. R. L. (1986). .l’achonaaascarntpesan: pv. manpower in western Washington crucifer seed fields: Occurence and survival. Phytopathology 76.1306-1309. 5m 8. R.. Swaaur. V. & Cuarraayss. S. S. (1972). Inheritance ofresisnanee no black rot in cauliflower. Canadian Journal of Genetic: and Cytology 14, 363-370. SHAW. J.J. & Kano. C. I. (1986). Development of a mm biolumineseence genes“ we monitor phytopathogenic bacteria during the ongoing disease process in a non-disruptive manner. Biotechnology 4. 560-564. . SHAW. J.J. & Kano. C. I. (1988). Whole plant wound inoculation for consistent reproduction of black not of crucifers. Pigstopanholog 78. 981-986. . Snuw.J.J.. Ssi'rus. L. G. & Kano. C. 1. (1988). Transposon mutagenes'nofXashsanaaascanpestrn': pv. : Characterization of a nonpathogenic mutant and cloning of a locus for pathogenicity. Molecular Plat-Microbe Interactions: 1. 39—45. Snsrsonn. R. B. & Jounsou. LJ. (1990). Arabidopsis thaliana as a host for Katha-noises caanpeun': pv. W. Molecular Plane-Microbe 1mm 3. 233—237. Snnn‘r. M. St Kosuos. T. (1978). The role of indole-3-acetic acid accumulation by alpha methyl tryptophan-resistant mutants of Pseudomonas mastaan' in gall formation on oleanders. Physiological Plat Pathology 13. 203-214. Sontaavrau. C. R. & Conan. W. L. (1982). Isolation of photorespiration mutants in Arabidopsis thaliana. In Method: in Clnloroflau Molecular Biology. Ed. by Edelman et al.. pp. 129-138. Elsevier Biomedical Press. New York. STALL, R. E. & Han. C. B. (1984). Chlorosis and ethylene production in pepper leaves infected with Xanthomonas catfish: pv. anaemia. W9 74. 373-375. Tune.J. 1... Conan. C. 1... Banana, C. E.. Dow. J. M. & Dumas. M. J. (1987). Molecular cloning of protease gene(s) from Xanthomonas mourn: pv. mesons: Expression in Escherichia coli and role in pathogenicity. Molecular and General Genetic: 210, 443-448. Tsuji.J. &. Sousavnua. S. C. (1988). Xarlnoanoaa: campannspv. consonant-induced chlorosisindrabidopsir Malia-a. Arabidopsis Inforrnnuioa Service 26.1-8. Vacuum. D.. VAN MONTAGU. M. & Van Luann-rm. M. (1988). Aguilar-tannin “foam-mediated transformation of Arabidopsis thaliana rent explants by using kanamycin selection. Proceeding: of the National Acadanny ofScn'enccr. U.S.A. 85. 5536-5540. . Wm. E. R. & Jets. G. C. (1990). Isolation ot single-eopy-sequence clones from a yeast artificial chromosome library of randomly-sheared Arabidopsis llnaln'aa DNA. Plat Molecular Biology 14. 561 -568. Wtuaanss. P. H. (1980). Black rot: .A continuing threat to world crucifers. Pleas Disease 64. 736-742. Wgnwwan Wn.us.D...K Hamsfll-ZHM chn.J..,J BurmTwM annow,S...(E.&Pauoeom.os.NJ (1990). Isolation and characrerizanon of a P:eadoanarna: syringae pv. syringae mutant deficient in lesion formation on bean. Molecular Plane—Microbe laceration 3. 149-156. Wtuou. E. E.. Zarrouu. F. M. & Fannucssorr. D. L. ( 1967). Bacterial phloem canker. a new disease of Persian walnut trees. Phytopathology 57. 618—621. 5 MP?” CHAPTER IV GENETIC MAPPING OF FIXCl, A XANTHOMONAS RESISTANCE GENE IN ARABIDOPSIS THALIANA 55 INTRODUCTION Most approaches for cloning genes involve a prior knowledge of the gene's mRNA or protein product. However, for most genes this information is not available. An alternative approach to cloning genes that are known only by their phenotype is map-based cloning. By this method, the target gene is mapped relative to genetic markers (for example RFLP or RAPD markers), a physical map of the region defined by the markers is created using pulsed-field gel electrophoresis, and a chromosome walk aimed at cloning the target gene is initiated from tightly-linked, flanking markers. This approach has been successfully employed in cloning genes responsible for a number of human genetic diseases (26.33). Arabidopsis thaliana (L.) Heynh. is a small weed in the family Cruciferae that is well suited for map-based cloning. Arabidopsis has a small genome (about 70,000 kb) with little interspersed repetitive DNA (18); has a comprehensive genetic map of morphological, RFLP, and RAPD markers (5,12,19,23); and has several YAC libraries (8,34). In addition, Arabidopsis is the only plant for which genes (fed? and fadD) have been cloned by chromosome walking (Arondel et al., in press). Previously, we identified the gene RXCi in Arabidopsis thaliana which confers resistance to Xanthomonas campestris pv. campestris (Pammel) Dowson, an important global pathogen of crucifers (32). In this chapter, we report on the genetic mapping of RXC1 to chromosome 2. Markers that flank FlXCl may serve as starting points in a chromosome walk directed at cloning this gene. 56 MATERIALS AND METHODS Plant Materials A. thaliana landraces Columbia (rxc1-1/rxcl-1) and Landsberg, which are resistant to X. c. campestris, and Pro (rxc1-2/rxc1-2), which is susceptible to infection, were used as the parental lines. Mapping RXC1 relative to morphological markers was conducted using 617 F2 individuals derived from crosses between Pro and six Landsberg mapping lines: MSU12 (op-2, er, as), MSU14 (ttg, yr), MSU15 (ap-Z, oer-2, bp), MSU16 (an, dis-1 ), MSU25 (hy-2, gI-1, "-6), and MSU30 (ch-1, ap-1, gl-Z) (11). Mapping RXCi relative to RFLP markers was conducted using approximately 100 F2 individuals derived from a cross between Columbia and Pro. Seeds were collected from individual plants of the this F2 population, and the resulting F3 families were used for DNA isolations. X. c. campestris inoculations Two leaves per plant were infiltrated with a suspension of X.c. campestris at a concentration of approximately 5x106 cfu/mL. After one week, leaves were visually scored for resistance. Plants were classified into two categories: resistant, with very little or no chlorophyll loss, and susceptible, with extensive chlorosis. Quantitative assesments of disease were performed by measurements of the chlorophyll contents of inoculated leaves (32). RFLP analysis 57 In order to identify RFLPs suitable for mapping analysis, DNA was extracted (25) from Columbia and Pro plants and digested with the following six restriction enzymes: EQQRI, Knal,xn_QI,B_g1ll,D_d_e_l, and HJ_n_dlll. The digested DNA was size fractionated by electrophoresis in TBE on 0.6% agarose at 20 volts for 15-16 hours. The DNA was transferred onto Hybond N (Amersham) membrane by the alkaline transfer method according to the manufacturer's instructions. Four RFLP markers that map to chromosome 2, 605 (5), 251 (5), GPA1 pCIT1838 (9), and ASA2 (20), were radiolabelled by the random hexamer priming method (7) and hybridized to the Arabidopsis DNA blots. Restriction enzymes revealing polymorphisms for a given RFLP marker were used to digest the DNA of the approximately 100 F3 families, and the marker was subsequently used as the hybridization probe. RESULTS Inheritance of resistance Previously, in a screen of 15 landraces of Arabidopsis for resistance to X. c. campestris, we identified resistance in the landrace Columbia. A cross was made between Columbia and the susceptible landrace Pro, and F1 seeds were collected. The F1 plants were then selfed and F2 seeds were harvested. The F2 plants were inoculated with X. c. campestris and individually scored for resistance. Figure 7 shows the distribution of disease severity, as determined by measurements of chlorophyll content of inoculated leaves, for an F2 population of 102 individuals from such a cross. 58 12 10‘ g 8- E O. * 0 6-1 L— g . 3 41 Z 1 2- .l 'l l ' 1 l l 0' ' cmomcmomomomomomomomomomomomomo wwmmmmvvmmmmnnoommOOwwmmnmvvm v-v-v-v-Pv-v-v-v-v-F chlorophyll content (percent of control) Figure 7. Frequency distribution histogram of chlorophyll contents of X. c. campestris-inoculated leaves for 102 F2 individuals derived from a cross between Arabidopsis landraces Columbia and Pro. 59 We classified these individuals into two genotypic classes based on their disease phenotype. The first class (rxcl-2/rxcl-2) comprised about 25% (23/102) of the F2 population and phenotypically showed extensive chlorophyll loss (mean % chlorophyll content, 30.7%). The second class (rxc1-1/rxc1-2 and rxc1-1/rxc1-1) comprised about 75% (79/102) of the F2 population and showed little chlorophyll loss (mean % chlorophyll content, 81.3%). Because the quantitative assessment of disease using measurements of chlorophyll content is a destructive process, RFLP linkage analysis was conducted on an additional 105 F2 individuals derived from the same cross. These plants were inoculated with X. c. campestris and visually scored for resistance. Eighty of these plants were found to be resistant and 25 of the plants were susceptible. Taken-together, these data are consistent with a 3:1 ratio (X2- 0.079 , p> 0.50) as predicted for the segregation of a single gene with resistance being dominant to susceptibility. Individual F2 plants were then allowed to self fertilize and F3 seeds were collected. DNA was then extracted from 27 susceptible and 76 resistant F3 families and used for RFLP analysis. During the initial screen of 15 landraces for resistance to X. c. campestris, resistance was also identified in the landrace Landsberg; however, resistance was observed at a lower inoculum concentration (5x105-1x106cfulmL) than in Columbia (1-5x106 cfu/mL). Therefore, the lower inoculum concentration was used to follow the inheritance of Xanthomonas resistance in Landsberg. A cross was made between the resistant landrace Landsberg and the susceptible landrace Pr0. F1 plants were allowed to self fertilize, 60 and the resulting F2 generation was inoculated with X. c. campestris. Of the 90 F2 individuals scored, 66 were resistant and 24 were susceptible. These results are also consistent with a 3:1 ratio (x2-o.133, p> 0.50) Linkage analysis Mapping rxcl-l on the classical genetic map (12) was conducted by linkage analysis of rxc1—1 to 16 morphological markers using F2 individuals derived from crosses between Pro and Landsberg mapping lines. Six mapping lines each possessing two to four morphological markers on a single chromosome in the Landsberg genetic background were used. The F2 individuals derived from these crosses were scored for each morphological marker and their disease response. These results are presented in table 5. Chi-square tests were then performed to test the goodness-of-fit of the observed segregation ratios with those expected for the independent assortment of unlinked loci. The chi-square values and the associated P values from these tests are shown in table 6. Four markers (dis-1, cp-2, as, and tt-6) were observed to have P values less than 0.05, suggesting that the deviations from the ratios expected for independent assortment were not due to chance alone. Significant deviations of the segregation ratios from those expected for independent assortment may be due to four factors: improper segregation of RXCl, improper segregation of the morphological marker, linkage of the morphological marker with RXCl, or incorrect scoring of a marker. To distinguish these sources of discrepancy, additional chi-square tests were performed to test the deviation of 61 Table 5. Number of F2 individuals with the specified morphological phenotype and disease response from crosses between Pro and six Landsberg mapping lines. Chromosome Cross Phenotype Resistant Susceptible 1 Pro X MSU30 Wild-type 38 1 1 ch-1 5 3 ap-l 2 0 gl-2 7 3 ch-l & ap-1 1 1 ch-1 & gl-2 2 0 ap-1 8 gl-2 2 1 ch-1, ap-1, & gl-2 3 1 Pro X MSU16 Wild-type 37 12 an 10 2 dis-1 2 1 an & dis-1 2 1 2 Pro X MSU12 Wild-type 49 30 as 0 0 cp-2 2 0 er 0 0 as & cp-2 14 2 as & er 0 0 cp-2 & er 3 as, cp-2, & er 14 1 3 Pro X MSU 25 Wild-type 81 27 hy-2 0 0 gl-l 6 0 hy-2 & gl-1 23 10 Wild-type 102 32 tt-6 15 3 4 Pro X MSU 15 Wild-type 42 14 ap-2 2 0 oer-2 5 1 bp 5 2 oer-2 & bp 3 2 ap-2 & bp 1 0 Table 5. Continued. 62 Chromosome Cross Phenotype Resistant Susceptible ap-2 & oer-2 12 2 ap-2, cer-2, & bp 2 5 5 Pro X MSU14 Wild-type 52 25 ttg 0 0 yi 5 0 ttg & yi 12 4 Table 6. Markers, X2, and P values used for the detection of linkage with RXCf. Chromosome Marker X§(9:3:3:1) P X§(no linkage) P 1 gl-2 0.49 we -‘ - ap-1 1.64 >0.5 - - ch-1 0.52 >0.9 - - an 0.28 >O.9 - - dis-1 10.29 <0.05 0.03 >08 2 cp-2 14.87 <0.005 8.02 <0.005 er 5.26 >0.05 3.53 >0.05 as 10.52 <0.01 9.43 <0.005 3 hy-2 0.66 >0.9 - - gl-1 0.19 >0.9 - - tt-6 14.45 <0.005 0.150 >05 4 ap-2 1.34 >0.5 - - oer-2 7.03 >0.05 0.73 >0.2 bp 4.55 >0.2 - - 5 ttg 5.56 >0.1 - - yi 3.37 >0.2 - - 1 -. not determined. 64 each of the two, single factor ratios from their expectations, and the deviation of the joint segregation from its expectation of no linkage (17). The chi-square values and the associated P values from these tests are also shown in table 6. The morphological markers dis-1 and tt-6 were found not to segregate in the expected segregation ratio of 3:1 in the F2 population, and were also unlinked to rxc1-1 (Table 6). However, the joint segregation of the marker cp-2 and HXCl deviated significantly from the results expected for the independent assortment of unlinked loci. This result was also observed with the marker as (Table 6). Hence, RXCl was linked with two morphological markers on chromosome 2. To create a more accurate map of the region surrounding RXC1, RXC1 was mapped relative to RFLP markers. Since RXCl was located to chromosome 2, four DNA clones (605, 251, GPA1 pCIT1838, and ASA2) from this chromosome were used as hybridization probes to Southern blots containing DNA of the susceptible landrace Pro and the resistant landrace Columbia digested with six restriction enzymes. Using the restriction enzymes EQQR I, m l, £09. l, 831 ll, D_d_e_ I, and Hind Ill, RFLPs were observed (Table 7). The RFLPs used in this study to map RXCl are shown in figure 8. These RFLPs were then used to analyze about 100 F3 families segregating for Xanthomonas resistance. The disease phenotype and RFLP genotype of each F3 family used in this study are shown in Table 8. From these data, pair-wise comparisons were made between FiXCl and each of the four RFLP markers using the LINKAGE-1 computer program (30). For each pair-wise comparison, the expected ratio for each locus was entered into the program. For example, during the 65 Table 7. Restriction enzymes yielding RFLPs between Arabidopsis landraces Columbia and Pro for markers 605, 251, GPA1, and ASA2. Restriction Enzyme1 EmbeEmBJ XhaJXIJQJBQJJ QdeJHindJll 605 + + + + .. ... 251 - - - - - + GPA1 + + . - + + + ASA2 - - - + - - 1 +, RFLP observed -, RFLP not observed 605 ASA2 GPA1 251 23.1 - .. . 9" . 6 U 6.6- ',. ‘0‘ - N N a o o W Figure 8. Restriction fragment length polymorphisms between landraces Columbia (left lane) and Pro (right lane) for the markers 605 (flgl II), ASA2 (531 ll), GPA1 pCIT1838 (Hind Ill), and 251 (Hind Ill). Lefthand margin indicates size in kilobases. The RFLP shown for marker 251 was between two F3 families. This RFLP is representative of the polymorphism observed between Columbia and Pro. 67 Table 8. Disease phenotypes and RFLP genotypes for F3 families derived from a cross between Arabidopsis landrace Columbia and Pro. RFLP Marker}: F3 Family Phenotype1 ASA2 251 GPA1 605 1 R 0 2 R 0 - +/- 3 R 0 0 0 4 R 0 O 0 5 R O O - 6 R O 0 0 7 R + + +/- 8 R + +/- - 9 R + +/- + 10 R O - +/- 11 R + 12 R +/- + + +/- 13 R 14 R 15 R 16 R 17 R 18 R +/- + 19 R 20 R 21 R 22 R 23 R +/- +/- 24 R 25 R 26 R +/- +/- 27 R 0 28 R 0 29 R 0 30 R +/- +/- 31 R + + 32 R +/- +/- 33 R + + 68 Table 8. Continued. RFLP Marker2 F3 Family Phenotype‘ ASA2 251 GPA1 605 34 R +l- +/— 35 R +/- +/- 36 R +/- +/- 37 R +/- +/- 38 R +/- +/- 39 R +/- +/- 40 R +/- +/- 41 R +l- + +/- 42 R +/- + +/- 43 R +/- + +/- 44 R + + +/- 45 R + + +/- 46 R +/- + - 47 R + + - 48 R +/- + + 49 R +/- + +/- 50 R + + + 51 R +/- + - 52 R +/- + +/- 53 R +/- + +/- - 54 R +/- +/- 55 R + + + + 56 R + + + + 57 R + + + + 58 R + + + + 59 R +/- + +/- +/- 60 R + + + +/- 61 R +/- + +/- +/- 62 R + + +/- + 63 R + +/- 0 64 R + +/- - 65 R +/- + +' +/- 66 R + 0 O + 67 R +/- - 68 R +/- +/- Table 8. Continued. 69 RFLP Marker2 F3 Family Phenotype1 ASA2 251 GPA1 605 69 R +/- +/- 70 R + +/- 71 R - +/- 72 R + + 73 R - + +/- +/- 74 R +/- O 0 +/- 75 R +/- O O +/- 76 R NT NT NT NT 77 S - - 78 S 0 0 79 S +/- +/- 80 S 0 0 81 S +/- - 82 S +/- +/- 83 S - - 84 S - - 85 S - - 86 S + - 87 S - 88 S 89 S +/- - 90 S +/- - 91 S - - 92 S - 0 0 +/- 93 S - +l- +/- 94 S - +/- +/- 95 S - +/- +/- 96 S - - - 97 S - - - 98 S - +l- - 99 S - - - - 100 S - + - - 101 S - O O +/- 102 S - - S 103 70 1Disease phenotypes: R, resistant to X. c. campestris; and S, susceptible to X. c. campestris. 2RFLP genotypes: +, Columbia-type of RFLP pattern; -, Pro-type of RFLP pattern; +/—, heterozygote; 0, unscorable; and NT, not tested. Genotypes were labelled unscorable when RFLP bands were not clearly identified. 71 analysis of linkage between RXCl and RFLP marker 251, an expected segregation ratio of 3:1 in the F2 generation was entered for both loci. The number of F3 families in each RFLP genotypic class and disease phenotypic class was also entered into the program. For example, 29 F3 families were observed to have the Columbia-type of RFLP pattern for marker 251 and were also resistant to X. c. campestris, one F3 family was observed to have the Columbia-type of RFLP pattern for marker 251 and was susceptible to X. c. campestris, seven F3 families were observed to have the Pro-type of RFLP pattern for marker 251 and were susceptible to X. c. campestris, and no F3 families were observed that had the Pro-type- of RFLP pattern for marker 251 and were resistant to X. c. campestris. From these data, the recombination fraction and its standard error were calculated by the LINKAGE-1 program using maximum likelihood formulae. Using the recombination fraction, the map distance in centimorgans was then calculated using the Kosambi mapping function (21). The recombination fraction and map distance between RXCl and each of the four RFLP markers is shown in Table 9. The marker 251 is the closest RFLP marker to FlXCl, located approximately 2.9 cM from FiXC1. DISCUSSION The task of mapping new genes to the Arabidopsis genetic map was greatly facilitated by the construction of two, high density RFLP maps (5,19). These two maps make it possible to follow the segregation of hundreds of genetic markers in a single population. 72 Table 9. Recombination fractions and map distances between loci. Loci Recombination fraction1 Map distance?— 605 and RXCl 0.181 :l: 0.047 18.96 251 and RXCl 0.029 :1: 0.164 2.90 GPA1 and RXCl 0.227 :f: 0.090 24.49 ASA2 and HXCl 0.140 :t: 0.045 14.38 605 and 251 0.309 t 0.090 36.09 251 and GPA1 0.216 :t 0.091 23.12 1 The recombination fraction and standard error were calculated using the LINKAGE-1 computer program. 2The map distance was calculated using the Kosambi mapping function, X . 1/4 In (1+2r/1-2r), where X is the map distance in centimorgans, and r is the recombination fraction. 73 Since these markers are distributed over all five chromosomes of Arabidopsis, the map position of any new gene can be readily identified. Using these markers, we mapped the Xanthomonas resistance gene, HXCl, to chromosome 2. A number of other disease resistance genes have recently been mapped in Arabidopsis. The gene RPM 1, which confers resistance to Pseudomonas syringae pv. maculicola was located to chromosome 3 (6). RAcl, an Albugo candida resistance gene, was mapped to chromosome 1 (4). The locus Flpt2, which determines resistance to P. s. tomato, was mapped to chromosome 4 (13,37). lea1, a X. c. armoraceae resistance gene, was mapped to chromomome 5 (14). Fle1, RPp2, Fle4, and RPp5, genes that confer resistance to Peronospora parasitica, were mapped to chromosomes 3 and 4 (22,31). A number of disease resistance genes in other plants have also been mapped relative to RFLP or RAPD markers. For example, RFLP or RAPD markers have been identified that are closely linked to Tm-2, a tobacco mosaic virus resistance gene from tomato (35); Rxl and Fix2, genes conferring resistance to potato virus X in potato (24); Pto, a Pseudomonas syringae pv. tomato resistance gene from tomato (16); It and I2, genes conferring resistance to Fusarium oxysporum f. sp. chospersici race 1 and 2, respectively, in tomato (27,28,29); Ht-1, a Helminthosporium turcicum race 1 resistance gene from maize (3); Grol, a gene conferring resistance to the cyst nematode Globodera rostochiensis in potato (1); R1, a Phytophthora infestans resistance gene from potato (15); Pi-2(t) and Pi-4(t), genes conferring resistance to Pyricularia oryzae in rice (36); and Sm, a Stemphylium resistance gene from tomato (2). The 71+ identification of genetic markers tightly linked to disease resistance genes is of interest to plant breeders, because these markers can be used to screen the available germplasm for resistance without having to perform inoculations, and to molecular biologists, because these markers can be used as starting points in a chromosome walk directed at cloning these genes. By mapping disease resistance genes relative to RFLP and RAPD markers, researchers have found that some of these genes are located on the same chromosomal regions. For example, the RFLP marker GP21, linked (4.5 cM) to the gene le2 that confers resistance to potato virus X and is located on chromosome V in potato (24), was also found to be tightly linked (2.5 cM) to the locus R1 that confers resistance to P. infestans (15). In tomato, which has a genome homologous to potato, Pto, which confers resistance to P. s. tomato, was also mapped to chromosome V (16). In addition, the RFLP marker TG20, which was linked (8.2 cM) to the gene Grol that confers resistance to the cyst nematode Globodera and is located on potato chromosome VII (1), was also found to be linked (10.6 cM) to the locus l1, that confers resistance to F. oxysporum and is located on tomato chromosome VII (28). 'Linkage of an additional RFLP marker to Gro1 and [1, confirmed that the two resistance genes are located on the same chromosomal segment that is homologous in potato and tomato. The location of disease resistance genes on the same chromosomal segments may indicate that they shared a common ancestral origin. The ancestral resistance gene may have undergone gene duplication giving rise to two, tightly-linked genes. Over time, 75 the two genes may then have evolved the ability to function against different plant pathogens. On the otherhand, the clustering of different resistance genes on the same chromosome may have occurred by chance and may not reflect an evolutionary relationship. The cloning and characterization of these genes may help to resolve this question. The genetic mapping of RXC1 represents the first step towards the cloning of this gene by chromosome walking. The next step is to isolate cloned DNA fragments that bridge the gap between the closest RFLP marker (251) and FlXCl. Towards this end, Hwang et al. have already used the marker 251 as a probe to identify the YAC clone EG7E2 (10). By CHEF gel electrophoresis, YAC clone EG7E2 was found to have an insert size of approximately 130 kb. By generating end-specific probes from YAC clone EG7E2 and using them as RFLP markers to orient the direction of the chromosome walk and as probes to identify other YAC clones, it should be possible to walk into RXC1. The actual physical distance to be covered by the chromosome walk, however, cannot be accurately determined based on the genetic distance because the correlation between the genetic and physical distance is not uniform throughout the Arabidopsis genome (8). REFERENCES 1. Barone, A., Ritter, E.. Schachtschabel, U., Debener, T., Salamini, F., and Gebhardt, C. 1990. Localization by restriction fragment length polymorphism mapping in potato of a major 76 dominant gene conferring resistance to the potato cyst nematode Globodera rostochiensis. Mol. Gen. Genet. 224: 177-182. 2. Behare, J., Laterrot, H., Sarfatti, M.. and Zamir, D. 1991. Restriction fragment length polymorphism mapping of the Stemphylium resistance gene in tomato. Mol. Plant-Microbe Interact. 4: 489-492. 3. Bentolila, S., Guitton, C., Bouvet, N., Sailland, A., Nykaza, S., and Freyssinet, G. 1991. Identification of an RFLP marker tightly linked to the Hit gene in maize. Theor. Appl. Genet. 82: 393-398. 4. Brose, E.. Holub, E.. Crute, I., and Beynon, J. 1992. Mapping of genes that code for resistance to the fungal pathogen Albugo candida in Arabidopsis thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 228. 5. Chang, C., Bowman, J., DeJohn, A., Landers, E.. and Meyerowitz, E. 1988. Restriction fragment length polymorphism linkage map for Arabidopsis thaliana. Proc. Natl. Acad. Sci., USA 85: 6856-6860. 6. Debener, T., Lehnackers, H., Arnold, M.. and Dangl, J. 1991. Identification and molecular mapping of a single Arabidopsis thaliana locus determining resistance to a phytopathogenic Pseudomonas syringae isolate. Plant Journal 1: 289-302. 7. Feinberg, A., and Vogelstein, B. 1983. A technique for radiolabelling DNA restriction endocuclease fragments to high specific activity. Anal Biochem 132: 6-13. 8. Grill, E., and Somerville, C. 1991. Construction and characterization of a yeast artificial chromosome library of Arabidopsis. Mol. Gen. Genet. 226: 484-490. 77 9. Hong, M.. Yanofsky, M. F., and Meyerowitz, E. M. 1990. Molecular cloning and characterization of GPA1, a G protein subunit gene from Arabidopsis thaliana. Proc. Natl. Acad. Sci., USA 87: 3821-3825. 10. Hwang, I., Kohchi, R.. Hauge, B., Goodman, H., Schmidt, R., Cnops, G., Dean, C., Gibson, 6., lba, K., Lemieux, B., Arondel, V., Danhoff, L., and Somerville, C. 1992. Identification and map position of YAC clones comprising one-third of the Arabidopsis genome. Plant Journal 1: 367-374. 11. Koornneef, M.. and Hanhart, C. 1983. Linkage marker stocks of Arabidopsis thaliana. Arabidopsis Information Service 20: 89-92. 12. Koornneef, M., van Eden, J., Hanhart, C., Stam, P.. Braaksma, F., and Feenstra, W. 1983. Linkage map of Arabidopsis thaliana. J. Hered. 74: 265-272. 13. Kunkel, B., Bent, A., and Staskawicz, B. 1992. Arabidopsis mutants with altered resistance to the bacterial pathogen Pseudomonas syringae pv. tomato. Sixth international symposium on molecular plant-microbe interactions. Abstract 349. 14. Lazo, G. R.. and Dixon, R. A. 1992. Screening transformation-competent DNA libraries for plant defense genes in Arabidopsis thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 351. 15. Leonards-Schippers, C., Gieffers, W., Salamini, F., and Gebhardt, C. 1992. The R1 gene conferring race-specific resistance to Phytophthora infestans in potato is located on potato chromosome V. Mol. Gen. Genet. 233: 278-283. 16. Martin, G., Williams, J., and Tanksley, S. 1991. Rapid 78 identification of markers linked to a Pseudomonas resistance gene in tomato using random primers and near-isogenic lines. Proc. Natl. Acad. Sci., USA 88: 2336-2340. 17. Mather, K. 1951. The Measurement of Linkage in Heredity. John Wiley & Sons, New York, pp. 32-43. 18. Meyerowitz, E. 1987. Arabidopsis thaliana. Ann. Rev. Genet. 21: 93-111. 19. Nam, H-G., Giraudat, J., den Boer, B., Moonan, F., Loos, W., Hauge, B., and Goodman, H. 1989. Restriction fragment length polymorphism linkage map of Arabidopsis thaliana. Plant Cell 1: 699-705. 20. Niyogi, K. K., and Fink, G. R. 1992. Two anthranilate synthase genes in Arabidopsis: defense-related regulation of the tryptophan pathway. Plant Cell 4: 721-733. 21. Ott, J. 1988. Analysis of Human Genetic Linkage. John Hopkins University Press. Baltimore, Maryland, p. 9. 22. Parker, J. E.. Szabo, V., Staskawicz, B. J., Lister, C., Dean, C., Daniels, M. J., and Jones, J. D. 1992. Analysis of an Arabidopsis gene for resistance to Peronospora parasitica. Sixth international symposium on molecular plant-microbe interactions. Abstract 364. 23. Reiter, R.. Williams, J., Feldmann, K., Rafalski, J., Tingey, S., and Scolnik, P. 1992. Global and local genome mapping in Arabidopsis thaliana using recombinant inbreds and random amplified polymorphic DNAs (RAPDs). Proc. Natl. Acad. Sci., USA 89: 1477-1481. 24. Ritter, E.. Debener, T., Barone, A., Salamini, F., and Gebhardt, C. 1991. RFLP mapping on potato chromosomes of two genes 79 controlling extreme resistance to potato virus X (PVX). Mol. Gen.. Genet. 227: 81-85. 25. Rogers, 8., and Bendich, A. 1988. Extraction of DNA from plants. Mol. Biol. Manual A6: 1-10. 26. Rommens, J., lannuzzi, M.. Kerem, B., Drumm, M.. Melmer, G., Dean, M.. Rozmahel, R.. Cole, J., Kennedy, 0., Hidaka, N., Zsiga, M., Buchwald, M.. Riordan, J., Tsui, L., and Collins, F. 1989. Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245: 1059-1065. 27. Sarfatti, M.. Katan, J., Fluhr, R.. and Zamir, D. 1989. An RFLP marker in tomato linked to the Fusarium oxysporum resistance gene I2. Theor. Appl. Genet. 78: 755-759. 28. Sarfatti, M., Abu-Abied, M.. Katan, J., and Zamir, D. 1991. RFLP mapping of l1, a new locus in tomato conferring resistance against Fusarium oxysporum f. sp. lycopersici race 1. Theor. Appl. Genet. 82: 22-26. 29. Segal, G., Sarfatti, M.. Schaffer, M.. Ori, N., Zamir, D., and Fluhr, R. 1992. Correlation of genetic and physical structure in the region surrounding the [2 Fusarium oxysporum resistance locus in tomato. Mol. Gen. Genet. 231: 179-185. 30. Suiter, K. A., Wendel, J. F., and Case, J. S. 1983. LINKAGE-1: a PASCAL computer program for the detection and analysis of genetic linkage. J. Herod. 74: 203-204. 31. Tor, M., Holub, E.. Crute, I., and Beynon, J. 1992. Mapping of genes that code for resistance to the fungal pathogen Peronospora parasitica in Arabidopsis thaliana. Sixth international symposium on molecular plant-microbe interactions. Abstract 229. 80 32. Tsuji, J., Somerville, S., and Hammerschmidt, R. 1991. Identification of a gene in Arabidopsis thaliana that controls resistance to Xanthomonas campestris pv. campestris. Physiol. Mol. Plant Pathol. 38: 57-65. ' 33. Wallace, M., Marchuk, D., Anderson, L., Letcher, R.. Odeh, H., Saulino, A., Fountain, J., Brereton, A., Nicholson, J., Mitchell, A., Brownstein, B., and Collins, F. 1990. Type 1 neurofibromatis gene: identification of a large transcript disrupted in three NF1 patients. Science 249: 181-186. 34. Ward, E. and Jen, G. 1990. Isolation of single-copy sequence clones from a yeast artificial chromosome library of randomly-sheared Arabidopsis thaliana DNA. Plant Mol. Biol. 14: 561-568. 35. Young, N., Zamir, D., Ganal, M.. and Tanksley, S. 1988. Use of isogenic lines and simultaneous probing to identify DNA markers tightly linked to the Tm-2a gene in tomato. Genetics 120: 579-585. 36. Yu, 2., Mackill, D., Bonman, J., and Tanksley, S. 1991. Tagging genes for blast resistance in rice via linkage to RFLP markers. Theor. Appl. Genet. 81: 471-476. 37. Yu, G.-L. and Ausubel, F. M. 1992. Defense genes expression in wild-type and mutant Arabidopsis during bacterial infection. Sixth international symposium on molecular plant-microbe interactions. Abstract 388. CHAPTER V PHYTOALEXIN ACCUMULATION IN ARABIDOPSIS THALIANA DURING THE HYPERSENSITIVE REACTION TO PSEUDOMONAS SYRINGAE PV. SYRIA/GAE This chapter was published in Plant Physiology (1992) 98: 1304-1309. 81 M W. (1992) 98.1304-1309 WMIQZISBNWIOBISOLMIO Recalvedlorptaricau'onSeptambara. 1991 AcceptadNovambarts. 1991 Phytoalexin Accumulation in Arabidopsis thaliana during the Hypersensitive Reaction to Pseudomonas syringae pv syringae1 Jun Tsuji, Evelyn P. Jackson, Douglas A. Gage, Raymond-Hammerschmidt. and Shauna C. Somerville' Michigan State University—Department at Energy Plant Research Laboratory (J. T., S. 6.8. ). Max 1'. Rogers Nuclear Magnetic Resonance Facility (E.P.J.), Michigan State University—National Institute of Health Mass Spectrometry Facility, Department of Biochemistry (DA. G. ), and Department of Botany and Plant Pathology (R.I-I.. 8.0.8.), Michigan State University, East Lansing, Michigan 48824 ABSTRACT WGWMWW(L)WM mmmmsmpvmm mmoexpreasionofmehypanerlaltivemaetionand in phyto- mmmmmmmm attarirtfiltrltionotlaavaawithamntofP.s.syringaedefieiertt mmahilitytoalicrtahyparaanmmetiorewlmmm pamogamxmscarnpestflspvcampastrtxorwith 1o Warm ptmaphata buffer(pH 6.9). Phytoalexin ae- Wmmmmmmmmua asyrhrgaeAphytoatexmwaaptuifiadbyacemb'matlonot mphaaeflaahchromatography,thinlayarchrematography. folewadbympltaaehighpartmllquidchmateg- raphy.ThaArahIdopeta phytoalexmwaaidendtledaaa-thiazol- 2'-yl-indoloonthehaaiaotlrltraviolet.lrrtmrad.massspaetnl. ‘mmmmeW hate-data. Phytoalexins are low mol wt. antimicrobial compounds of plant origin that accumulate after inoculation with a plant pathogen ( )4). A number of observations support the hypoth- esis that phytoalexin: play a role in the defense response of plants to pathogens. Phytoalexins are absent in healthy tissues and accumulate after infecrion by fungal (I8. 24. 30) or bacterial (9. ll) pathogens in monocotyledonous plants (24) as well as in dicotyledonous plants (9, I8. 30). Phytoalexins have been demonstrated to accumulate rapidly at the site of attempted infection in sufficient quantities to inhibit the in vitro growth of fungi 1 IS. 30) and bacteria (9). Virulence of the fungus Necm’a haematococca on pea is correlated with the ability to detoxify the phytoalexin pisatin (28). Further- more. transformation of Cochll‘obolus heterostroplm with the gene encoding pisatin demethylase allowed this maize patho- 'Supported by Michigan Agricultural Experiment Station (No. l648). National lnstrtutes for Health-National Centers for Research Resources lPHI-RR00480 to J. Watson). US. Department of Energy 1 DE-FGOZ-90ER2002 I ). US. Department ongriculture I No. 8080). and the Michigan State University Foundation. .I.T. was supported in part by a fellowship from the College of Natural Science. 1304 gen to cause limited infections on pea (20). These observations suggest that phytoalexins contribute to disease resistance. Although a large body of evidence supports the defensive role for phytoalexins, phytoalexin accumulation is not the only disease resistance-associated response that has been ob- served in plants. Plants often respond to infection with a HR.2 a rapid localized necrosis that is a common response of plants to bacteria. fungi. and viruses (10). Plants have also been observed to deposit lignin and hydroxyproline-rich glycopro- teins in their cell walls and to synthesize the fungal cell wall- degrading enzymes chitinase and 3-I.3-glucanase after infec- tion (5). Because a number of host responses have been correlated with disease resistance. the relative contribution of phytoalexin: to disease resistance remains 00an We have initiated an investigation of phytoalexin accu- mulation in Arabidopsis thaliana (L) Heynh. to address the contribution of phytoalexins to disease resistance. Arabidopsis offers many advantages for physiological and molecular stud- ies of disease resistance in plans ( 12). Pseudomonas syringae pv syringae van Hall. a bacterial wheat pathogen. is nonpath- ogenic on Arabidopsis and elicits a HR. In this paper. we report on the purification and structural determination of a phytoalexin from Arabidopsis and examine the accumulation of this phytoalexin during the HR to P. s. syn‘ngae. MATERIALS AND METHODS Biologic-librarian Arabidopsis thaliana race Columbia was grown as described (29). The following bacterial strains were used in this study: Pseudomonas syringae pv syringae Nal', PSSD20, a wheat pathogen (23); P. s. syringae Nal', Kan‘. Hrp‘. PSSD20::Tn5. PSSDZZ. a nonpathogenic mutant (23); and Xanthomonas campestris pv campestris (Pammel) Dowson Rif'. 2D520. a cauliflower pathogen (2 I ). Bacteria were grown. suspended in I0 mu potassium phosphate buffer (pH 6.9) to IO‘ cfu/mL. and infiltrated into leaves of 3-weelt-old plants as described previously (29). ’ Abbreviauons: HR hypersensitive reaction: cfu. colony forming unit: El. electron impact. 83 AN AW PHYTOALEXIN 1305 PurificadonandOuantiflcationoftheArahidopais Phytoalexin Phytoalexin accumulation was elicited by spraying leaves of 3-weelt-old Arabidopsis with IO mM silver nitrate in 0.1% Tween 20. After I d. leaves were harvested. quickly weighed, and placed in boiling 80% methanol for IS min. The extract was cooled to room temperature and then filtered through Whatman filter paper No. I. The methanol was then removed by rotary evaporation at 35'C. and the resulting aqueous solution was extracted three times each with 2 volumes of chloroform. The chloroform-soluble extracts were pooled. evaporated to dryness. and Stored at -20‘C. The chloroforrn-soluble residue was dissolved in 50% ethanol and fracrionated by flash chromatography on Cu (25). The Cu column (2 x 7.5 cm) was developed with a srep gradient of 0. 25. 50. 75. and 95% ethanol. and IO-ml. fractions were collected and assayed for antifungal activity as described below. Fungitoxic fractions were combined. and the solvent was removed by evaporation under nitrogen. The sample was dissolved in chloroform and fractionated by TLC on silica gel. TLC plates were developed in chloro- form:methanol (9:1. V/V), air-dried. and assayed for inhibition of fungal growth by the Cladosportum bioasaay as described below. The Arabidopsis phytoalexin was eluted from the silica gel with methanol. concentrated under nitrogen. and purified by reverse phase HPLC. HPLC was performed on a Varian 500 instrument with a Waters nBondapac C.. column (0.78 x 30cmlaspreviouslydescribed(l6)exceptelutionwaswith a linear gradient of l to I00% acetonitrile at a flow rate of 2.0 mL/min. Purified Arabidopsrr phytoalexin was quantified in metha- nol at 318 nm using a molar extincuon coefficient of l4.800 M“cm". This molar extinction coefficient was mlculated using the Beer-Lambert law by measuring the UV absorbances ofweighed samples. A linear relationship between phytoalexin concentration and abaorbance at 318 nm was observed with phytoalexin concentrations between 300 ng/ml. and 30 ug/ ml. methanol. 390mm UV absorption spectra were obtained on a Beckman DU- 70 spectrophorometer'. IR spectra were obtained on a Nicolet FUR/42: El and high resolution (R a 5000) mass spectra were obtained on a JEOL AXSOS double focusing mass spectrometer: and the fast atom bombardment mass spectrum was acquired on a JEOL HXI 10 double focusing mass spec- trometer with nitrobenzyl alcohol as the liquid matrix. 'H- and "C-NMR spectra were obtained on a Varian VXR 500 spectrometer at 500 and 125 MHz. respectively. Chemical shifts were referenced to the solvent. CDCIt. FungalandBaetsriaIGrowthlnhibltIonAssays Antifungal activity was detected by the Cladospon'um-TLC bioassay (l). Extracts were sported onto silica gel 60A KGF TLC plates (Whatman) and the plates were developed in chloroform:methanol (9:1, v/v). The plates were then air- dried and sprayed with a dense conidia) suspension of Cla- dosporium cucumerinum Ellis 6t Arth. suspended in double- strengthpotatodextrpsebrothwarmedto40'C.Theplates were then incubated under high humidity in the dark at room temperature for 3 d. Antibacterial activitywasdetectedasdescribedunexcept that an agar overlay was used. Purified phytoalexin prepara- tions were spotted onto silica gel TLC plates. and the plates were overlayed with a mixture of P. s. syringae suspended in 50°C King's B agar ( 19) amended with 0.05% 2.3.5-triphenyl tetrazoliumchloride.Theplateswerethenincubatedina moistchamberin thedarltatroom temperature for3d. BacterialGrowthShrdy Leaves of Arabidopsis were infiltrated with a suspension of P. s. syringae PSSD20 at a concentration of l x l0' cfu/mL. At 0. l. 2. 3, and 5 d post-inoculation. an oval Ieafdisc (0.44 cm‘) was removed from the inoculation site. rinsed three times in sterile distilled water. and then macerated in l ml. of IO mu potassium phosphate buffer (pH 6.9). Viable counts were determined by plating serial dilutions of the leaf extract on King's B agar ( 19) containing 200 ug nalidixic acid/ml. RESULTS WMWWWM The Arabidopsis phytoalexin was eluted from the Cl. flash chromatography column with 95% ethanol. This fraction was further purified by TLC on srlica gel. The phytoalexin mi- grated on the TLC plates with an R: value of about 0.56. Final purification of the Arabidopsis phytoalexin was achieved by HPLC on a Cu column. Purified phytoalexin eluted from the column as a single peak with 6 l % acetonitrile (retention time. 37 min) (Fig. I). By this purification method. 2.31 mg of the Arabidopsis phytoalexin was recovered fi'om 1.28 kg fresh weight of elicited leaf tissue. The phytoalexin was also purified from P. s. syringae-inoculated leaves and was determined to be identical to that purified from silver nitrate-treated leaves based on the UV and mass spectra. I p00 '1 A . I I ’- j I LJJ 9 l .' : — i ” 90- N 'D l , "3 r: N ' X I Z v ; l . '3 L: l ,' I f- v i .r so -'-l 2 : , "I L) < l ,I < a ' I ii i 1' i: v i .’ on: .‘,' :n 'E' ’, -‘ Lu e—J :2 l .’ A z. < J ,’ “' 5" i l l I l 3 'o 20 30 40 so so RETENTiCN TIME (mm) c; «--— \ Figurst. ReversephaseHPLCotImgrtoxr’elractionsobta‘nedlmrn slicagelTLCplataaEIutionwaswlmalineargrars'entott t0100% acetonnnle. 84 race rsuseru. - . PlattPhyaid.Vd.98.1992 Antifungal activity was also detected in a more polar frac- system involved four protons: a one-proton multiplet at 8.25 tion that eluted from the Cu HPLC column with 42% ace- ppm. a one-proton multiplet at 7.43 ppm, and a unto-proton. tonitrile (retention time. 25 min). This fraction was present ABXX' resonance at 7.28 ppm. Taken together, these two in minor quantities and was unstable when rechromato- spin systems were suggestive of a monosubstituted indole graphed on the reverse phme column. This fraction was nor substituted at the 3 position. with the broad singlet at 8.45 purified to homogeneity and was not investigated further. ppm representing the N-H proton and the narrow doublet at 7.87 ppm representing H-Z. The two one-proton signals at Structural Detaruinatlon or the Arabidopsis Phytoalexin 8-25 and 7-43 pom Wewe asisned to H-4 and 114. respectively. _ _ and the two-proton signal at 7.28 ppm was assigned to [+5 The 4'00?me 131171011600 W obtarned 3‘ a stable. and l-l-6. The assignments were confirmed by comparison of colorless sohd that fluoreeoed boaht blue-Mic on o TLC the 500 MHz 'H-NMR spectrum of the Arabidopsis phyto- plate under l0'18 waveleosth UV (302 nm) illumination. The alexin with those of 3-cyanorndole. 3-methyl indole. and fluorescence was visible when as little as 3.5 ng of theArabi- brassinin at the same field m The '3C-NMR m of 40PM phytoalexin was spotted on a TLC plate. the Arabidopsis phytoalexin were also consistent with these In the El mass spectrum. the purified Arabidopsis phyto- interpretations. alum dismayed a prominent molecular ion as the we oelk The two additional coupled doublets at 7.82 and 723 ppm at I'll/2 200. This 11101 W! determination was confirmed by fast (I a 3.3 Hz) remained to be account“! for in the spectrum, atom bombardment MS. WhiCh displayed an intense peak for Excluding these [mm m morons, (he indole mm no. the protonated molecule (MH’) at m/z 201. High resolution counted for eight cal-bong, six hydrogens. and one of the El MS indicated that the fonnula for the Arabidopsis ohvto- nitrogen atoms in the molecular formula. In addition, six of alexin was CrrHths (measured. 300-0401: calculated. the nine degrees of unsaturation were fulfilled. Only a few 200.0408), which is consistent with the isotope disrriburion possible structures can be prom to fit the last three car- pattern observed in the low resolution mass spectrum. This bong. one sulfur. one aim two hydrogengb and the re. formula 808853“ that the compound contained a combined maining three degrees of unsaturation into the molecule. One total of nine rings and double bonds (rings + double bonds - possibility, the cad-unsaturated isothiocyanate. can be ex- ! " Y/2 + 2/2 + I. for CHM). eluded on the basis of the stability of the Arabidopsis phyto- The UV mm 01' a methanolic $01060!) 01' the .lrabi- alexin relative to the expected instability of the isothiocyanate doom phytoalexin contained absorbance maxima (IO! 6) at and the amence of a prominent IR absorption between 2000 318 nm (4.17), 275 am (3.92), and 215 nm (4.36). in accord and 2300 cm". with thehighdegreeof conjugation indimtedbythe molecular Three alternate structures seemed more likely. The peaks formula. at l63.l and 142.6 ppm in the "GM spectrum. the two The 'H-NMR spectrum (500 MHz. CDCl;) displayed six coupled prorons at 7.82 and 7.23 ppm in the ‘H-‘IMR one-proton and one two-proton downfield signals between spectrum that could be assigned to the two carbon peaks at 7.2 and 8.5 ppm. which could be grouped into three spin 142.6 and l15.9 ppm by heteronuclear multiquantum coher- coupling sysrems based on selective decoupling experiments ence data. and the fragment ion m/z 58 (Cd-11S) in the mass (Fig. 2). The first spin system consisted of the broad singlet at spectrum indicated that the substituent attached to the indole 8.45 ppm. which was exchangeable with 0:0. and coupled to was a thiazole or an isothiazolc (15). These possibilities were a narrow doublet at 7.87 ppm (J =- 2.2 Hz). The second spin discriminated by ‘H-NMR. The coupling constant (J - 3.3 0005' H5 H6 H5’ H2H4’ H7 H4 .../\¥ IIITIIIIIITWTITTT—Tl—VTTIITFIE...IIIFTIIIITI]ITTTIIII‘IITTIIIIITITTITT] 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.8 7.5 7.4 ppm Figural ‘H-NMRspacmanlSOOMHLCDCIa)ottheAraoidopsrsphytoalexln. ANARABIDOPSISPD-WTOALEXIN 1307 4’ (3150 L I E E 5' 3.5. I >3:- 22100‘ i l 85 . :5 1 1 " 3 E .3. 3'3 50‘ "' 2‘ figures. ShilcmreolttioArahidopsrsphytodaxn.3-tbiazo1-2’-yt- 2’3 3 a. 5 08/1 rat 1 P Hz) between the doublets at 7.82 and 7.23 ppm is character- istic ot‘a monosubstrtuted thiazole substituted at the 2' posi- tion (3). Hence. the structure of the Arabidopsis phytoalexin is 3-thiazol-2'—yl-indole as shown in Figure 3. PhytoaloxlnAeeurnulatloninArabidopslsand Rolatlonahipwlththol-lfltoP.s.synngao When leaves of Arabidopsis were inoculated with the non- host pathogen. P. s. syringae PSSDZO (approximately 1 x 10‘ cfu/mL), a confluent. grayish. sunken necrosis characteristic of the HR (10) was observed within 24 h. During the same period. leaves infiltrated with P. s. syringae PSSD22. the I-Irp‘ mutant of P. s. syringae PSSDZO: X. c. campestris: or 10 mar pomssium phosphate buffer (pH 6.9) were symptomless. The viability of P. s. syringae inoculated into leaves of Arabidopsis was investigated by reisolating P. s. syringae from inoculated leaves and determining the number of living bac- A E U Q L O 3 U’ U} \ 3 \- U V O) O ..J 4fi.v.e.v.v . 0123456 Days post-inoculation M04. TirnecollrseotgrowthotP.s.symgaePSSO20hfltramd htolaaveaolAraDidopsrs.Eachpcmisthameanotsrxrepicataa. SWerrcrbarsaresnown.Sirri-lsresutswareobtanadhn~o adritiorialexpamiantainvvtaclibaldseswaresuiacastarizedm 0.5%scdlsnhypoaianeta1t02mbeloremacsranon. o 10 20 so 40 so Hours post-inoculation Figures. Thnacottaeotmsaeclmaatlonotpnymuxnactivrty. LaavasotAracrdcpsrsracaCclumbiamhfltratadwlmlolP.s. syringae PSSDZO. (O) P. s. syringae PSSDZZ. (x) X. c. cm 20520.0rm10mupctassunphospnatebuttarlpl16.9).0hloro- form-schnalracnorisolmathariollaaiaxtractslsomglreanwayn) massayadtorhmgitcxicactivlty.anctnaareaollungagrowtn Wmmn‘wasmcomsd.Eacnpositisthamaanotttm soparamaxpenmantsfitandudarrcrnaraareatrwn. teria by dilution plating. in Arabidopsis. populations of P. s. syringae increased within 24 h alter inoculation but then exhibited a decline in the number ofViable bacteria after 24 h (Fis. 4). To detect the presence of a phytoalexin. inoculated leaves were harvested and chloroform-soluble fractions of methanol extracts were prepared. From the Cladospon’um-TLC bioas- say, antifungal activity was detected as a single zone of fungal growth inhibition with an RF value of approximately 0.56. Phytoalexin aetivity increased rapidly between 12 and 24 h after inoculation with P. s. syringae and reached maximum activity between 24 and 48 h postinoculation (Fig. 5). During the same time period. little or no antifungal aetivity was detected by the same method in extracts prepared from leaves infiltrated with 10 mM potassium phosphate buffer (pH 6.9); the crucifer pathogen. X. c. campestris: or a mutant of P. s. syringae deficient in the ability to elicit a hypersensitive reaction in tobacco or Arabidopsis (Fig. 5). Once a method for purifying the Arabidopsis phytoalexin had been develOped. the accumulation of this compound in leaves of Arabidopsis was assessed. After inoculation with P. s. syringae. the amount of the phytoalexin increased in par- allel with the level of phytoalexin aetivity to a maximum level of about 8 lug/8 fresh weight (Fig. 6). No fluorescence under long UV illumination was observed with extracts prepared from uninoculated leaves. The antimicrobial activity of the purified phytoalexin was quantified by determining the min- imum amount of the phytoalexin able to reproducibly inhibit the growth of either the fungus C. cucumerinum or the bac- terium P. s. syringae. An aliquot of 250 ng completely inhib- ited the growth of C. cucumerinum. A higher amount. 1 ttg. 1313 TSLUIETAL. "" 4L 24 .. pg 3-thlazol-2'-gl-lndola per gram (rash welght o a . . . . o 10 20 so 40 so Hours post-inoculation Flame. Tlrriacotlrsaottneaccurailatioriota-ttiiazol-Z'-yl-iriools MWWP.s.synngaaPSSD20.mpnymdaxn waspixlfiedlrominociaatedloavesotAraOidopsisandquantifiad tisaigarriouaxthetloncoalficiaritot14.800M"crn“.Eacnpont islttenramoltwosepuateaxparmmswidamarmrbusm shown. was required to inhibit the growth of P. s. syringae (data not shown). DISCUSSION Leaves of Arabidopsis exhibited a classic HR when inocu- lated with the wheat pathogen P. s. syringae. The bacteria were restricted within the necrotic tissue and exhibited a decline in the number of viable cells. To determine whether the decline in P. s. syringae population levels and the expres- sion of nonhost resistance may be due in part to the accu- mulation of antimicrobial compounds by Arabidopsis. we used the Cladospori'um-TLC bioassay to detect the presence of phytoalexins. Using this relatively sensitive bioassay, we were able to detect the accumulation ofa low mol wt molecule that met the criteria of a phytoalexin as defined by Paxton (14). This phytoalexin accumulated rapidly in leaves within 12 h postinoculation with P. s. syringae and was associated. with the cell collapse that occurred during the expression of the HR. The Arabidopsis phytoalexin was found to inhibit the in vitro growth of P. s. syringae. and the accumulation of the Arabidopsis phytoalexin was negatively correlated with the in vivo growth of P. s. syringae. Although further studies are needed. these initial observations suggest that the phyto- alexin plays a role in the general defense mechanism of Arabidopsis against pathogens. The Arabid0psis phytoalexin was found to accumulate to a maximum level of about 8 rig/s fresh weight by 36 h. This accumulation was greater than the maximum level of accu- mulation of brassilexin (240 ng/g fresh weight) in leaves of Brassica juncea 6 d after inoculation with Leptasphaeri'a mam/ans (17), but less than the maximum level of accumu- lation of cyclobrassinin and methoxybrassinin (about 50 and 86 MWVG.” 1992 100 ttg/g fresh weight. respectively) in leaves of B. napus 10 d alter inoculation with L maculans (4). Thus. although the levels of accumulation of the Arabidopsis phytoalexin were not comparable to those of the isotlavanoid phytoalexins (9. 30), the levelswerewithintheranpexhibitedbythemher cruciferous phytoalexins. Whilethispaperwasin preparanon.BrowneetaL(2) reported the isolation of two phytoalexin. camalexin and methoxycamalexin. from leaves of Carnelian saliva. The chemical structure reported for camalexin is identical to the structure we elucidated for the Arabidopsis phytoalexin. Com- parison of the spectral data of the Arabidopsis phytoalexin with those reported for camalexin suggests that the two phy- toalexin: are the same compound. thus providing independ- ent confirmation for the structure of 3-thiazol-2'-yl-indole. In addition. Slusarenko and Mauch-Mani (22) reported the pres- ence of an antibacrerial compound in extracts prepared from inoculated Arabidopsis. Further spectral data are required to determine whether this compound is identical or related to 3. thiazol-2’-yl-indole. Eight other phytoalexins have been isolated from five other species of the Cruciferae (6. 7. 13. 26. 27). Camalexin is structurally similar to thme phytoalexins in that it is a allfur- containing indole derivative. The structural similarity be- tween the cruciferous phytoalexins suggests that they may share common biosynthetic intermediates. All of the phyto- alexin: are structurally similar to indolyl 3-methyl isOthiocy- anate. which is the product of the hydrolysis of glucobrassicin by myrosinase. However. a direct link between the indole glucosinolates and the cruciferous phytoalexin: has nor been established. Devys er al. (6) speculated that indole-3arbox- aldehyde may be a common precursor of the indole phyto- alexins and have recently isolated a significant amount of this compound from leaves of Brassica oleracea (8). indole} mrboxaldehyde may also be an intermediate to camalexin biosynthesis. The Arabidopsis phytoalexin may be biosynthe- sized through the cyclization of indole-3-carboxaldehyde with cysteine with the subsequent loss of C02. We are currently investigating this possible biosynthetic pathway. With the description of an Arabidopsis phytoalexin. the molecular and genetic tools available in Arabidopsis can be employed to determine the relative contribution of this phy- toalexin in host defenses against both bacrerial (29) and fungal pathogens (2). In addition, the collections of auxou'ophic mutants of Arabidopsis ( 12) will aid in the elucidation of the biosynthetic pathway of 3-thiazol-2'-yl-indole and its meta— bolic regulation in Arabidopsis. WNW The authors thank Dr. Soledad Pedras (Plant Biorechnology [nati- tute. National Research Council. Saskatoon. Saskatchewan. Canada) for the samples of brassinin and 3-cyanoindole. Philip Jensen ( Mich- igan State University) for the sample of 3-methyl indole. and Dr. William Reuscb (Michigan State University) for the Fourier trans- form lR spectra. APPENDIX 3-Thiazol-2'-yl.indole. isolated as a colorless solid. UV max (MeOH. log t): 318 nm (4.17). 275 nm (3.92). and 215 nm (4.36): Fourier transform IR (CHClr. cm"): 3692 (med). 3610 (small). 3480 AN ARABIDOPSE PHYTOALEXIN (large). 16“) (med). 1585 (mod). and 1580 (med): HRMS (El) m/z measured 200.0401: calculated 200.0408 for CtthNtS; EIMS m/z (relative intensity): 200 (C..H.N§. 100). 142 (CeHeNi. 21). 115 (CoflsN. 7). 1m (6). 86 (4). 58 (cuts. )7): 'H-NMR (500 MHz. C00,): 8.45 (brs. 1H. 020 eachanuable. N-H). 8.251m. 1H. H4), 7.87 (d. .l - 2.2 Hz. 1H. H-Z). 7.82 (d. J - 3.4 Hz. 1H. H-4’), 7.43 (m. 1H. H-7). 7.28 (m. 2H. H-5. H-6). 7.23 (d. J - 3.2 Hz. 1H. H- 5'). "C-NMR (125 MHz. C003): 163.1 (s. C-Z'). 142.6 (d. C-4'). 136.4 (s. CJa). 124.7 (d. C-2). 124.4 (s. C-3a). 123.2 (d. C-5). 121.5 (d. C-6). 120.7 (d. C-4). 115.9 (d. C-5'). 112.7 (s. C-3). 1 11.9 (d. C-7). UTERAME CITED 1. AllsaEl-LKneJ(1968)a~Solanineanda-chaconineasfungitoaic compounds in extracu of Irish pomto tubers. Phytopathology 58: 776-781 Browne LNLCoonKI.AyerWA.TewariJP(1991)Thecamal- exins: new phytodexins produced in the leaves of Camelr'na saliva (Cruciferae). Tetrahedron 47: 3909-3914 . Brigal W (1979) Handbook of NMR Spectral Parameters. Vol 3. Heyden and Son Limited. London. pp 646-651 4. Dahiya JS. Rimmer SR (1988) Phytoalexin accumulation in tissues of Brassica napus inoculated with Laxasphaen‘a nia- crdans. Phytochemistry 27: 3105-3107 5. Darvill AG. Albersheim P (1984) Phytoalexins and their elici- tors—a defense against microbial infection in plans. Annu Rev Plant Physiol 35: 243-275 6. DevysMBarbierM.Loiseletl.RoaxelT.SanigaatA.Koll- maria A. Bousqoet J-F (1988) Brasrlexin. a novel sulphur- containing phytoalexin from Brassica junta: L. (Cniciferae). Tetrahedron Lett 29: 6447-6448 7. Devys M. Barbier M. Kollrrlarln A. Round T. Bouquet J-F (1990) Cyclobrassinin sulphoxide. a sulphur-containing phy- toalexin from Brassica juncea. Phytochemistry 29: 1087-1088 Devys M. Barbies M (1991) lndole-J-carboxaldehydc in the cabhge Brassica oleracea: a systematic determination. Phy- tochemistry 30: 389-391 9. Keen NT. Kennedy BW ( 1974) Hydroxyphaseollin and related isollavanoids in the hypersensitive resistance reaction of soy- beans to Pseudomonas glycine. Physiol Plant Pathol 4: 173-185 Kiss-sat Z (1982) Hypersensitivity. In MS Mount. GH Lacy. eds. Phytopathogenic Prokaryotes. V012. Academic Press. New York. pp 149-177 Laid BM. Lyon GD (1975) Detection of inhibitors of Erwinia carotovora and E. herbicala on thin-layer chromatograms. J Chromatogr I10: 193-196 12. Meyerowitz EM ( 1987) Arabidopsis thaliana. Annu Rev Genet 21:93-111 Moods KSasalrl K.ShlrataA.Tallasagi M(l990)4-Methoxy- brassinin. a sulphur-containing phytoalexin from Brassica aler- aaa. Phytochemistry 29: 1499-1500 PaxtoaJD(1981) Phytoalexins—aworkingredefinition. Phyto- pathol Z 101: 106-109 a 6. U 10. 13. I4. 87 22. . Yoshikawa M. Yamallchl K. Maaago H (1978) 1309 . Porter-QN(1985) MassSpectrometryofHeterocyclicCorn- pounds John WileytltSons. New York. p 899 . Rasmussen JB. Schefler RP (1988) Isolation and biological activities of four selective toxins from Hdmintlmporium car- bommr. Plant Physiol 86: 187-191 . RoarteIT. SarnigoetA.KolImaanA.BoaaaaetJ-F(l989)Ac~ cumulation of a phytoalexin in Brassica spp. in relation to a hypersensitive reaction to Leprosphaert'a inoculum. Physiol Mol Plant Path0134: 507-517 . SatoN.KitaaawaK.TomiyamaK(l971)'l'heroleofrishitinin loalizing the invading hyphae of Pllyioplliliora iii/attain in infecnonsitesatulecutsiin‘acesofpotatomberstysiolPlant Pathol 1: 289-295 . Schaad NW (1988) Laboratory Guide for the Identification of Plant Pathogenic Bacteria. American Phytopathologiml Soci- ety. St. Paul. MN. p 3 Sehil'er W. Smiley D. Ciul'fetti L Van Etta! HD. Yodar OC (1989) One enzyme makes a fungal pathogen. but not a sap- rophyte. virulent on a new host plant. Science 246: 247-249 . Shaw JJ. Kado CL (1986) Development of a Vibno biolumi- nescence geneset to monitor phytopathogenic bacteria during the ongoing disease process in a non-disruptive manner. Biol technology 4: 560-564 Slusarenko AJ. Much-Mani B (1991) Downy mildew ofAra- bidopsis thaliana caused by Permspora parasitica: a model system for the investigation of the molecular biology of host- pathogen interactions. In H Hennecke. DPS Verma. eds. Ad- vances in Molecular Genetics of Plant-Microbe Interactions. Vol 1. Kluwer Academic Publishers. Dordrecht. pp 280-283 . Smith .M. Hammerschmidt R. Fulbright DW (1991) Rapid induction of systemic resistance in cucumber by Pseudomonas syringae pv. syringae. Physiol Mol Plant Pathol 38: 223-235 . Snyder BA. Nicholson RL (1990) Synthesis of phytoalexins in sorghum as a site-specific response to fungal ingrem. Science 248: 1637-1638 . Still WC. Kahn M. Mitra A (1978) Rapid chromatographic technique for preparative separatiom with moderate resolu- tion. J Org Chem 43: 2923-2925 . ‘I‘akasagi M. Meade K. KamiN.ShintaA(l987)Spiiobrassi- nin. a novel sulfur-containing phytoalexin from the daikon Raphanus sarivus 1... var. horrensis (Cniciferae). Chem Lett 1631-1632 . TaknogiMModoKKatsniN.ShhataA(l988)Novelsulfilr- containing phytoaleains from the Chinese cabbage Brassica campestris L. spp. pekineris13(Crudfer'ae). Bull Chem Soc Jpn 61: 285-289 . Tegtrneier 10. Van Ester: HD(1982)Theroleofpisatin tolerance and deyadation in the virulence of Necrn'a hmocomr on peas:a genetic analysis. Phytopathology 72: 608-612 .TsoiiJ. SomervilleSC. Hammerschmidt R(l991) Identification of a gene in Arabidopsis thaliana that controls resistance to Xanthomonas campestris pv. campestris. Physiol Mol Plant Pathol 38: 57—65 Glyceollin: its rolein restrictingl'ungalgrowthinresistantsoybmnbypocotyls infected with Phyom'uhm W var. sojoe. Physiol Plant Pathol 12: 73-82 CHAPTER VI CAMALEXIN BIOSYNTHESIS IN AHABIDOPSIS THALIANA 39 INTRODUCTION The family Brassicaceae is susceptible to a number of fungal and bacterial pathogens. Recent work has suggested that resistance to these pathogens may be mediated by the accumulation of low molecular weight antimicrobial compounds known as phytoalexins (1 ,2,7,14,15). The crucifers accumulate an array of phytoalexins all belonging to a novel class of sulfur-containing indoles (3,4,5,12,17,18,19). In contrast to our knowledge of the chemistry of these compounds, nothing is known concerning the biosynthesis of these phytoalexins. The biosynthetic intermediates, enzymes, and genes that encode the biosynthetic enzymes have yet to be identified. We have initiated an investigation of phytoalexin metabolism in Arabidopsis thaliana (L.) Heynh. in order to define the biosynthetic pathway of the sulfur-containing indole camalexin, to identify the genes encoding the biosynthetic enzymes, and to evaluate the physiological effects of blocks in specific biosynthetic steps. Such work is prerequisite for the genetic engineering of resistance in the Brassicaceae through modifications of the phytoalexin pathway. We have recently described the purification and structural characterization of the phytoalexin camalexin (3-thiazol-2'-yl-indole) from leaves of Arabidopsis (19). Similar to other cruciferous phytoalexins, camalexin is a sulphur-containing indole derivative. Most indole-based metabolites like indole-3-acetic acid (1AA) and the indole glucosinolates are thought to be derived from tryptOphan because of the structural similarities 90 between the molecules. However, recent work characterizing a mutant of maize mutated in both genes encoding the tryptophan synthase beta subunit has shown that tryptophan is not a precursor to 1AA in maize (20). In this paper, we investigated whether camalexin is biosynthesized from tryptophan. MATERIALS AND METHODS Plant material Three tryptophan-requiring mutants of Arabidopsis were used in this study. The Arabidopsis mutant trp1-100 (10) is significantly reduced in anthranilate transferase activity, the mutant trp2-1 (11) is mutated in the major tryptophan synthase beta subunit gene, and the mutant trp3-1 (11) is believed to be mutated in the major tryptophan synthase alpha subunit gene. Seeds of the three tryptophan-requiring mutants were germinated on medium (16) containing 50uM L-tryptophan for one week. Seedlings were then transplanted to soil without further tryptophan supplementation. Plants were grown as described (16) in a growth chamber under the following conditions: 25°C for 16 hours of light (mercury vapor and high pressure sodium lamps, 40 to 50 pEinsteins PAR/mzlsec) and 20°C in the dark. Phytoalexin isolation and quantification Leaves of Arabidopsis were sprayed with 10mM silver nitrate, 0.1% Tween 20 to induce the accumulation of camalexin. Leaves from 5 to 20 plants were then harvested, pooled, quickly weighed, 91 and placed in boiling 80% methanol. After the methanol had evaporated, sufficient water was added to the samples to restore the original volume. The extracts were then partitioned twice each with one volume of chloroform. The chloroform phases were pooled and evaporated using a rotary evaporator. The samples were then resuspended in chloroform, and the extracts were fractionated by silica gel thin layer chromatography using chloroform: methanol (9:1, v/v) as the solvent. The location of the phytoalexin on the TLC plate was visualized under long UV illumination, and the phytoalexin was eluted from the silica gel with methanol. The amount of camalexin was then quantified by fluorometry as described previously (5). Precursor feeding study Labelled anthranilate (ring 14C, 10 mCi/mmol) was obtained from Sigma, and labelled D/L-tryptophan (5-[3HJ-trp, 20 Ci/mmol) was obtained from from New England Nuclear. The 3H-tryptophan was diluted with unlabelled L-tryptophan to the same specific activity as 1"”C-anthranilate, and the two labels were fed to 3- to 4-week old leaves as follows: Ten leaves per treatment were excised under water with a razor blade and the petiole of each leaf was immersed in a SOuL solution (5 nmole, 50 nCi) of one of the labelled compounds in the tip of a 0.5 mL microcentrifuge tube. Following uptake of the label, water was added to the tubes and fed to the leaves. Six hours after the label was administered to the leaves, the leaves were sprayed with a solution of 10mM silver nitrate, 0.1% Tween 20 to elicit the accumulation of camalexin. 92 Control leaves that were fed with labelled compound were left untreated since spraying leaves with a solution of 0.1% Tween 20 does not elicit the accumulation of camalexin. Uptake of the compounds was calculated by determining the amount of radioactivity remaining in the tubes. Eighteen hours after elicitation, the ten leaves per treatment were pooled and camalexin was extracted as described above. Extracts were fractionated by thin layer chromatography, and the amount of radioactivity on the TLC plate was measured with a Bioscan beta-detector. Camalexin-containing zones on the TLC plate (approximately 1 cm) were scraped-off, and the amount of radioactivity measured using a scintillation counter. RESULTS Phytoalexin accumulation in lea ves of Arabidopsis sprayed with silver nitrate. Spraying leaves of Arabidopsis with a solution of 10mM silver nitrate, 0.1% Tween 20 resulted in the accumulation of camalexin. The phytoalexin that was purified from silver nitrate-treated leaves possessed identical ultraviolet absorption spectra and mass spectra as the phytoalexin that was purified from P. s. syringae -inoculated leaves (19). Thus, the structure of camalexin is not altered by spraying leaves with silver nitrate. The abiotic elicitor silver nitrate was used in this study to insure that labelled compounds that were fed to the plant were metabolized by the plant and not by the bacterium. 93 A time course of camalexin accumulation following elicitation with silver nitrate was performed to determine when the maximum rate of phytoalexin accumulation occurred. Camalexin accumulated rapidly within 12 hours after elicitation and reached maximum levels between 24 and 48 hours. Levels of camalexin then declined over the next three days (Fig. 14). The maximum rate of camalexin accumulation occurred between 12 and 24 hours post elicitation. Incorporation of radioactivity from 14C-anthranilate and 3H—tryptophan into camalexin after elicitation with silver nitrate. We hypothesized that camalexin is biosynthesized from tryptophan because of the structural similarities between the two molecules. To test this hypothesis, we fed equal molar amounts of labelled tryptophan and anthranilate of the same specific activity to leaves of Arabidopsis, and measured the incorporation of radioactivity into camalexin. Figure 15 shows the TLC profiles of 14C-anthranilate-labelled and 3H-trypt0phan-Iabelled metabolites. The profiles of 14C-anthranilate-Iabelled metabolites differed between the control leaves and the elicited leaves. The profile of 14C-anthranilate-labelled metabolites in the control leaves revealed a single major peak of radioactivity. In contrast, the profile of 14C-anthranilate-labelled metabolites in elicited leaves revealed a number of peaks of radioactivity. One of the major peaks was identified as camalexin based on co-chromatography with camalexin. In contrast to 14C-anthranilate, the profiles of 3H-tryptophan-labelled metabolites did not differ between the control leaves and the elicited leaves. In the TLC profiles of both 3000 camalexin (rig/gram fresh weight) Days post-elicitation Figure 14. Time course of camalexin accumulation following elicitation with 10 mM silver nitrate, 0.1% Tween 20. Each point is the mean of three separate experiments. Standard error bars are shown. 95 A 1001- . sor- B 100? 1 L3? 50)- ;; 1M“ - f - A.“ -u‘ E? 1 =3 so) C i so! ‘0 DAM“) 20 100i D 30+ 60)- 40+ 20+ co 2 4 e a 1'01'21'4131320 Distance from the origin (cm) Figure 15. TLC profiles of 1‘j’C-anthranilate and 3H-tryptophan-labelled metabolites. Extracts prepared from control (A and C) or elicited (B and D) leaves fed with 10 mCi/mmole 14C-anthranilate (A and B) or 10 mCi/mmole 3H-tryptophan (C and D) were spotted on a silica gel TLC plate. The plate was developed in CHCIS: MeOH (9:1, v/v), and scanned for radioactivity using a Bioscan beta-detector. The arrow indicates the camalexin peak. 96 the control and elicited leaves, three major peaks of labelled compounds were observed. None of the three peaks co-chromatographed with camalexin. Quantitative data was obtained by determining the amount of radioactivity taken-up by the leaves and the amount present in the camalexin fractions. Uptake of 14C-anthranilate by the detached leaves in three separate experiments was found to range between 94% to 99%. Similarly, uptake of 3H-tryptophan in three different experiments ranged between 97% to 99%. The percent incorporation of radioactivity from 14C-anthranilate into the camalexin fraction of elicited leaves was measured to be about 11 times greater than that of the controls. In contrast, the percent incorporation of radioactivity from 3H-tryptophan into the camalexin fraction of elicited leaves was not significantly greater than that of the controls (Table 10). Furthermore, the specific activity of camalexin labelled with 1"’C-anthranilate was 5 to 6 times greater than the specific activity of camalexin labelled with 3H-tryptophan (Table 10). These results do not support the hypothesis that tryptophan is a precursor to camalexin. Camalexin levels in three tryptophan-requiring mutants of Arabidopsis after elicitation with silver nitrate. As an additional test of the hypothesis that camalexin is biosynthesized from tryptophan, camalexin levels were measured in three tryptophan-requiring mutants of Arabidopsis (10,11). For these experiments, the wild-type Columbia and the three tryptophan-requiring mutants were sprayed with 10mM silver 97 Table 10. Incorporation of radiolabel from 14C-anthranilate or 3H-trypt0phan into camalexin.1 Incorporation Percent Specific Activity3 Label Treatment (dpm) Incorporation2 (mCi/mmole) Izc-Anthranilate Control 616 a 40 0.046 a 0.004 -4 C'Anih'ani'ale AUN°3 7.050 :1: 920 0.515 :1: 0.074 2.32 a 1.00 3H-Tryptophan Control 1,190 :l: 130 0.033 3; 0.014 - 3H-Tryptophan AleOa 1.633 3: 560 0.123 2 0.044 0.53 2 0.04 1Data are expressed as the mean and standard deviations of three separate experiments. 2The percent incorporation was determined by dividing the amount of radioactivity on the TLC plate corresponding to the Rf 0f camalexin, by the amount of radioactivity taken up by the leaves. 3The specific activity was determined by dividing the amount of radioactivity in the camalexin fractions by the amount of camalexin in the fractions as determined by fluorometry. The specific activity was calculated from two independent experiments. 4-, Not determined. 98 nitrate, 0.1% Tween 20 to elicit the accumulation of camalexin. Eighteen hours later, the plants were harvested, and leaves of 5 to 20 wild-type, trp1-100, trp2-1, or trp3-1 plants were pooled and extracted for camalexin. In three separate experiments, camalexin levels in the mutants trp2-1 and trp3-1 were measured and found to be comparable to those of the wild-type Columbia; however, camalexin levels in the mutant trp1-100 were significantly lower than those of the wild-type (Table 11). These data also do not support the hypothesis that tryptophan is a precursor to camalexin. DISCUSSION To test the hypothesis that camalexin is biosynthesized from tryptophan, we fed equal molar amounts of labelled tryptophan and anthranilate of the same specific activity to leaves of Arabidopsis, and measured the incorporation of radioactivity into camalexin. If significant incorportation of radioactivity into camalexin was observed with anthranilate and tryptophan, then these data would support the hypothesis that camalexin is biosynthesized from tryptophan. However, significant incorportation of radioactivity into camalexin was observed only with anthranilate and not with tryptophan. In addition, if tryptophan is a precursor to camalexin, then the specific activity of camalexin labelled with 3H-tryptophan should be greater than the specific activity of camalexin labelled with 1‘j'C-anthranilate. However, we found that the specific activity of camalexin labelled with 14C-anthranilate was 5 to 6 times greater than the specific activity of camalexin labelled with 99 Table 11. Camalexin levels in three tryptophan-requiring mutants of Arabidopsis eighteen hours after elicitation with 10 mM silver nitrate, 0.1% Tween 20. Arabidopsis Camalexin (ng/gram fresh weight)1 Columbia 1418.8 1 177.2 a trp1-100 . 763.4 1 93.09 b trp2-1 1282.5 2 211.4 a trp3-1 1485.0 :1: 411.5 a 1Values are the mean and standard deviation of three separate experiments. Values with the same letter are not significantly different (P-0.05) by Duncan's new multiple range test. 100 3H--tryptophan. These results do not support the hypothesis that camalexin is biosynthesized from tryptophan. Instead, we have adopted the alternate hypothesis that the biosynthetic pathway of camalexin branches off the tryptophan pathway before tryptophan (Fig. 16). As an additional test of the hypothesis that camalexin is biosynthesized from tryptoohan, we measured camalexin levels in three tryptophan-requiring mutants of Arabidopsis. Most if not all of the genes encoding enzymes in the tryptophan biosynthetic pathway in Arabidopsis are duplicated (10,11). For example, the mutant trp2-1 is mutated in the major tryptophan synthase beta subunit gene. This mutation is lethal under high light conditions, but is not lethal under low light conditions. The ability to grow under low light conditions is presumably due to the presence of a functional minor tryptophan synthase beta subunit gene. Likewise, the mutants trp1-100 and trp3-1 are also light-conditional mutants and are thought to have very low levels of enzyme activity due to the presence of a functional minor gene. Therefore, if camalexin is biosynthesized from tryptophan, then camalexin levels in these mutants are expected to be significantly reduced, but not zero. If camalexin levels were significantly reduced in trp1-100, trp2-1, and trp3-1, then these results would support the hypothesis that camalexin is biosynthesized from tryptophan. However, camalexin levels were reduced only in trp1-100, supporting the hypothesis that the biosynthetic pathway of camalexin branches off the tryptophan pathway before tryptophan. Since camalexin levels in the mutant trp1-100 were lower than those in the wild-type 101 TRYPTOPHAN BIOSYNTHET 1C PATHWAY Chorismate l Anthranilate trp 1- 700 l Anthranilate Transferase S-Phosphoribosylanthranilate l 1 -(O-Caroxvphenvlaminol-1-deoxyribulose—5-P lndole-3-glycerol phosphate trp3-l Tryptophan synthase 0 lndole trp2-1 l Tryptophan synthase ,8 Tryptophan Figure 16. The tryptophan biosynthetic pathway. The enzymatic steps with decreased activity are shown for the three tryptophan-requiring mutants of Arabidopsis. 102 Columbia, the branch point in the tryptophan pathway must occur after anthranilate. Furthermore, since camalexin levels in the mutant trp3-1 were comparable to those of the wild-type plant, the branch point must be before indole (Fig. 16). This paper is the first report to address the biosynthesis of cruciferous phytoalexins. All the cruciferous phytoalexins that have been characterized to date share a common indole skeleton. Therefore, we believe camalexin biosynthesis in Arabidopsis to be an appropriate model system to study the early steps in the biosynthesis of these phytoalexins. Previously, Takasugi speculated that the indole glucosinolates may be precursors to the cruciferous phytoalexins (18). Based on labelling experiments and the analysis of glucosinolate mutants, all glucosinolates are believed to be derived from amino acids (6,9). For example, the indole glucosinolates are thought to be derived from tryptophan. Since tryptophan is at best a poor precursor to camalexin, the results from this study do not support the hypothesis that the indole glucosinolates are precursors to camalexin. The results presented in this paper are consistent with the hypothesis that the biosynthesis of camalexin occurs from a branch off the tryptophan pathway. Similar results have recently been reported for the biosynthesis of lAA in maize (20). By feeding maize seedlings labelled tryptophan, Wright et al. reported no incorporation of label from 15N-tryptophan or 2H5-tryptophan into 1AA. Wright et al. also measured 1AA levels in the maize mutant orange pericarp, a tryptophan auxotroph mutated in both tryptophan synthase beta subunit genes. Despite lower levels of free 103 tryptophan, this mutant had a dramatically increased 1AA content. Whether the branch point off the tryptophan pathway for camalexin biosynthesis is the same as that for 1AA biosynthesis remains to be determined. Recent studies of aromatic amino acid biosynthesis in Arabidopsis have revealed that several enzymes are induced by inoculation with pathogenic bacteria (8,13). Keith et al. have recently cloned two genes encoding 3-deoxy-D-arabino -heptulosonate 7-phosphate synthase (DAHP), which catalyzes the first step in the shikimic acid pathway (8). RNA of the more highly expressed DAHP gene (DHSl) accumulated rapidly and reached maximum levels by approximately six hours after inoculation with the avirulent bacterium P. s. tomato MM1065. In contrast, RNA of DHS1 accumulated more slowly and reached maximum levels by 24 hours after inoculation with the virulent bacterium P. s. maculicola E34326. In addition, Niyogi and Fink have cloned two genes encoding the alpha subunit of anthranilate synthase, the enzyme catalyzing the first step in the tryptophan biosynthetic pathway (13). RNA of the more highly expressed anthranilate synthase gene (ASA2) increased transiently six hours after inoculation with the avirulent strain P. s. tomato MM1065. In contrast, RNA of ASA2 did not accumulate until 12.5 hours after inoculation with the virulent strain P. s. maculicola E84326. The induction of these genes may reflect the coordinate regulation of the genes involved in the biosynthesis of secondary defense compounds like camalexin. ACKNOWLEDGEMENTS 104 I thank Dr. R. Last for the seeds of the tryptophan-requiring mutants of Arabidopsis, and Dr. M. 200k for the assistance with the 14C-anthranilate and 3H-tryptophan labelling experiments. R E F E R E N C E S 1. Conn, K. L., Tewari, J. P., and Dahiya, J. S. 1988. Resistance to Alternaria brassicae and phytoalexin-elicitation in rapeseed and other crucifers. Plant Science 56: 21-25. 2. Dahiya, J. S., and Rimmer, S. R. 1988. Phytoalexin accumulation in tissues of Brassica napus inoculated with Leptosphaeria maculans. Phytochemistry 27: 3105-3107. 3. Devys, M.. Barbier, M.. Loiselet, l., Rouxel, T., Sarniguet, A., Kollmann, A., and Bousquet, J.-F. 1988. Brassilexin, a novel sulphur-containing phytoalexin from Brassica juncea L., (Cruciferae). Tet. Letters 29: 6447-6448. 4. Devys, M.. Barbier, M., Kollmann, A., Rouxel, T., and Bousquet, J.-F. 1990. Cyclobrassinin sulphoxide, a sulphur-containing phytoalexin from Brassica juncea. Phytochemistry 29: 1087-1088. 5. Hammerschmidt, R.. Tsuji, J., Zook, M.. and Somerville, S. C. 1992. A phytoalexin from Arabidopsis thaliana and its relation to other phytoalexins of crucifers. K. R. Davis and R. Hammerschmidt (eds) Arabidopsis as a Model System for Studying Plant-Pathogen Interactions. APS Press, St. Paul. in press. 6. Haughn, G. W., Davin, L., Giblin, M., and Underhill, E. W. 1991. Biochemical genetics of plant secondary metabolites in Arabidopsis thaliana. The glucosinolates. Plant Physiol. 97: 217-226. 105 7. Jejelowo, O. A., Conn, K. L., and Tewari, J. P. 1991. Relationship between conidial concentration, germling growth, and phytoalexin production by Camelina sativa leaves inoculated with Alternaria brassicae. Mycol. Res. 95: 928-934. 8. Keith, B., Dong, X., Ausubel, F. M.. and Fink, G. R. 1991. Differential induction of 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase genes in Arabidopsis thaliana by wounding and pathogenic bacteria. Proc. Natl. Acad. Sci., USA 88: 8821-8825. ‘ 9. Larsen, P. O. 1981. Glucosinolates. E. E. Conn (ed.) The Biochemistry of Plants. Vol 7. Academic Press, New York, pp. 501-525. 10. Last, R. L., and Fink, G. R. 1988. Tryptophan-requiring mutants of the plant Arabidopsis thaliana. Science 240: 305-310. 11. Last, R. L., Bissinger, P. H., Mahoney, D. J., Radwanski, E. R., and Fink, G. R. 1991. Tryptophan mutants in Arabidopsis: the consequences of duplicated tryptophan synthase beta genes. Plant Cell 3: 345-358. 12. Monde, K., Sasaki, K., Shirata, A., and Takasugi, M. 1990. 4-Methoxybrassinin, a sulphur-containing phytoalexin from Brassica oleracea. Phytochemistry 29: 1499-1500 . 13. Niyogi, K. K., and Fink, G. R. 1992. Two anthranilate synthase genes in Arabidopsis: Defense-related regulation of the tryptophan pathway. Plant Cell 4: 721-733. 14. Rouxel, T., Sarniguet, A., Kollmann, A., and Bousquet, J.-F. 1989. Accumulation of a phytoalexin in Brassica spp in relation to a hypersensitive reaction to Leptosphaeria maculans. Physiol. Mol. Plant Pathol. 34: 507-517. 106 15. Rouxel, T., Kollmann, A., Boulidard, L., and Mithen, R. 1991. Abiotic elicitation of indole phytoalexins and resistance to Leptosphaeria maculans within Brassiceae. Planta 184: 271-278. 16. Somerville, C. R.. and Ogren, W. L. 1982. Isolation of photorespiration mutants in Arabidopsis thaliana. M. Edelman, R. B. Hallick, and N. H. Chua (eds). Methods in Chloroplast Molecular Biology. Elsevier, Amsterdam, pp. 129-138. 17. Takasugi, M., Monde, K., Katsui, N., and Shirata, A. 1987. Spirobrassinin, a novel sulfur-containing phytoalexin from the Daikon Rhaphanus sativus L. var. hortensis (Cruciferae). Chem. Letters: 1631-1632. 18. Takasugi, M., Monde, K., Katsui, N., and Shirata, A. 1988. Novel sulfur-containing phytoalexins from the Chinese cabbage Brassica campestris 1.. spp. pekinensis (Cruciferae). Bull. Chem. Soc. Japan 61: 285-289. 19. Tsuji, J., Jackson, E.. Gage, D., Hammerschmidt, R.. and Somerville, S. C. 1992. Phytoalexin accumulation in Arabidopsis thaliana during the hypersensitive reaction to Pseudomonas syringae pv. syringae. Plant Physiology 98: 1304—1309. 20. Wright, A. D., Sampson, M. B., Neuffer, M.. G., Michalczuk, L., Pernise Slovin, J., and Cohen, J. D. 1991. lndole-3-Acetic acid biosynthesis in the mutant maize orange pericarp, a tryptophan auxotroph. Science 254: 998-1000. CHAPTER VII FUTURE DIRECTIONS 107 108 In this thesis, a locus, RXC1. was identified in Arabidopsis thaliana that confers resistance to X. c. campestris. As a prelude to cloning this gene, an RFLP marker about 2.9 cM from RXCl was found. This marker can now be used as a starting point in a chromosome walk directed at cloning this gene. YAC clones containing inserts of much of chromosome 2 near the region of RXCl have already been oriented. End-specific probes from these YACs can then be used as additional markers to orient these YAC clones relative to RXCl. lfa marker is found that is less than one cM from RXCl, then subclones from that YAC can be used to transform the susceptible landrace Pro. Transformants can then be tested for resistance to X. c. campestris. If a transformant is found that is resistant to X. c. campestris, then the subclone used to transform that plant most likely contains RXC1. The clone can then be further subcloned and tested for the ability to confer resistance to X. c. campestris. If Pro is not readily transformable, then the landrace RLD that has been backcrossed to Pro for two generations can be used. Once the genomic and cDNA clones of RXCl have been obtained, then a number of questions can be addressed. 1. What is the function of RXC1? To address this question, the RXCl cDNA should be sequenced to determine the amino sequence of the RXCl gene product. The nucleotide and amino acid sequence should then be compared with those of other known genes and proteins. If significant homology exists between RXCl and another known gene, then this information may give a clue as to how RXCl functions in disease resistance. The 109 nucleotide and amino acid sequence may also indicate whether the RXCl gene product is compartmentalized in an organelle, is membrane-bound, or is located in the cytoplasm. One common interpretation of the gene-for-gene hypothesis is that the product of the pathogen's avirulence gene interacts directly with the product of the plant's resistance gene at the host plasma membrane. Knowledge of the hydrophobicity of the RXCl gene product, knowledge of the presence or absence of the appropriate signal sequences, and the use of antibodies to localize the RXCl gene product will help to address the suitability of such a model for the X. c. campestris-Arabidopsis interaction. If no significant sequence homology is found with other known genes or proteins, then further studies are required to elucidate the function of RXCl. One possible biochemical phenotype of the RXCl gene product is toxin catabolism. X. c. campestris may synthesize a chlorosis-inducing toxin that landrace Columbia is able to catabolize but landrace Pro is not. To test this hypothesis, 1 infiltrated concentrates, prepared from culture filtrates of X. c. campestris grown in liquid medium 523, into leaves of landrace Columbia and Pro. No chlorosis-inducing activity was observed. However, hrp genes have recently been isolated from X. c. campestris and shown to be repressed in complex medium but induced in minimal medium (1). Thus, X. c. campestris may produce a toxin in planta that is not synthesized in liquid medium 523. To test this hypothesis, 1 prepared concentrates from intercellular wash fluids of chlorotic Pro leaves previously inoculated with X. c. campestris. Again, no chlorosis-inducing activity was observed. 110 Even though a toxin was not detected in these preliminary experiments, it does not rule out the possibility that X. c. campestris produces a toxin in planta that is unstable during isolation. Although a few other pathovars of X. campestris have been reported to produce a toxin (3,4), there is no precedence in the literature of a toxin produced by X. c. campestris. Another possible function of RXCl is in conferring ethylene insensitivity. Leaves of Arabidopsis inoculated with X. c. campestris could produce high levels of ethylene at the site of inoculation. Leaves of landrace Pro may be more sensitive to ethylene than leaves of Columbia, and thus become symptomatic. Differential sensitivity to ethylene may also explain why X. o. campestris is able to multiply equally well in landrace Columbia as in Pro. To test this hypothesis, I infiltrated suspensions of ethephon into leaves of landrace Columbia and PrO. Although ethephon induced chlorosis, no differential response was observed between the two landraces of Arabidopsis. 2. How is RXC1 regulated? To determine whether RXCl is constitutively expressed or is induced by inoculation, Northern blot analysis should be performed. Also, the expression of RXC1 in landrace Columbia should be compared to that in landrace Pro. The expression of the RXCl promoter can also be studied in transgenic plants. For example, by fusing the RXCl promoter to the beta-glucuronidase coding region, the expression of the RXCl promoter can be studied during 111 development and under different stresses. Furthermore, the sis-elements in the RXCl promoter recognized by Arabidopsis transcription factors can be studied. For example, the expression of deletion and site-specific mutation constructs of the RXCl promoter can be analyzed in transgenic plants. Also, the binding sites in the RXCl promoter for Arabidopsis proteins can be mapped using gel retardation assays and by DNase I footprinting. Once the sis-elements have been identified, then the transcription factors can be isolated using sequence-specific DNA affinity columns. lf RXC1 is inducible rather than constitutively expressed, then these studies may help to elucidate the signal transduction pathway leading to the activation of the RXCl promoter. 3. What is the relationship between RXCl and other disease resistance genes? By the time RXCl is cloned, nucleotide sequence information will most likely be available for a number of other disease resistance genes like Hm1, RPM 1, and Pto. Such information may reveal whether there are functionally important regions that are conserved among different disease resistance genes. Also, the RXCl cDNA and antibody should be used to test whether similar genes or proteins exist in other crucifers, dicots, and other plants. Such information may indicate whether the tolerance encoded by RXCl is specific for X. c. campestris or functions against a broad range of pathogens. Also, these data would indicate the suitability of using the RXCl cDNA as a hybridization probe to clone other disease resistance genes. 112 4. Can RXCl enhance disease resistance in Brassica? One of the purposes of cloning disease resistance genes in Arabidopsis is to use those genes to enhance the resistance of agronomically important crops to plant pathogens. If the resistance encoded by RXCl is specific to X. c. campestris, then Brassica species can be transformed with RXCl and tested for enhanced disease resistance. If the transformants are more resistant to the disease symptoms caused by X. c. campestris, then this result would indicate that a similar resistance mechanism operates in Arabidopsis and Brassica. In this thesis, the phytoalexin camalexin was purified from Arabidopsis and the structure was elucidated as 3-thiazol-2'-yl-indole. Two main questions arise from this work. How is camalexin biosynthesized? What role does camalexin play in disease resistance? 1. How is camalexin biosynthesized? Similar to IAA biosynthesis in maize, camalexin was found to be biosynthesized from a branch off the tryptophan pathway. The work described in this thesis suggests that the branch point is at N-5'-phosphoribosylanthranilate, 1-(o-carboxyphenylamino)-1 -deoxyribulose 5-phosphate, or indole-3-glycerol phosphate. Although labelled forms of these compounds are not commercially available, labelled forms of these putative intermediates may be biologically or chemically synthesized. A number of E. coli mutants 113 have been isolated that are blocked in various steps in the tryptophan pathway. By feeding these bacteria labelled anthranilate, these mutants can accumulate intermediates and thus be a source of labelled metabolites. Alternatively, labelled putative intermediates may also be chemically synthesized. Once labelled intermediates are obtained, the compounds can then be tested for the ability to label camalexin. For example, if label from N-5'-phosphoribosylanthranilate, 1-(o-carboxyphenylamino)-1 -deoxyribulose 5-phosphate, and indole-3-glycerol phosphate are incorporated into camalexin, then this result would suggest that the branch point is at indole-3-glycerol phosphate. lf labelled intermediates cannot be obtained, then the individual biosynthetic enzymes in the tryptophan pathway can be measured to see if they are induced after inoculation. For example, if anthranilate synthase, transferase, isomerase, and indole-3-glycerol phosphate synthase are induced by inoculation, but tryptophan synthase is not, then these results would suggest that the branch point is at indole-3-glycerol phosphate. On the otherhand, if the entire tryptophan pathway is induced, then the branch point cannot be determined by this approach. After the branch point has been determined, then the intermediates between the branch point and camalexin can be isolated and identified. Putative intermediates can be isolated from elicited Arabidopsis leaves during the period of phytoalexin accumulation and purified by HPLC. The putative intermediates can then be identified by GC/MS analysis. This approach has been successfully used to identify intermediates in the biosynthesis of 114 cotton sequiterpene phytoalexins (Essenberg, unpublished results). Once the intermediates are known, they can be used as substrates in the purification of the enzymes. Once the enzymes have been purified, then the corresponding genes can be cloned. After the genes encoding the biosynthetic enzymes have been cloned, then the genes can be introduced and expressed in Brassica. The transformants can then be tested for camalexin synthesis and enhanced disease resistance. 2. What role does camalexin play in disease resistance? The best approach to address this question is to isolate mutants of Arabidopsis that are altered in the ability to accumulate camalexin. A screen for such mutants has just recently been performed (4). Glazebrook and Ausubel screened approximately 7,000 M2 plants following inoculation with P. s. maculicola E84326 and isolated three phytoalexin deficient mutants. Two mutants accumulated lower levels of camalexin than the wild-type. The third mutant did not accumulate any detectable levels of camalexin. After these mutants have been backcrossed to remove any unlinked mutations, these mutants will be an invaluable tool in addressing the role camalexin plays in resistance against bacterial, fungal, viral, nematode, and fastidious prokaryotic pathogens. REFERENCES 1. Arlat, M. Gough, C. L., Barber, C. E.. Boucher, C., and Daniels, M. J. 1991. Xanthmonas campestris contains a cluster of hrp genes 115 related to the larger hrp cluster of Pseudomonas solanacearum. Mol. Plant-Microbe Interact. 4: 593-601. 2. Mitchell, R. E. Coronatine analogues produced by Xanthomonas campestris pv. phormiicola. 1991. Phytochemistry 30: 3917-3920. 3. Hammerschlag, F. A., Bottino, P. J., and Stewart, R. N. 1982. Effect of culture filtrates of Xanthomonas campestris pv. pelargonii and extracts of geranium stems inoculated with X. c.pv. pelargonii on geranium callus and seedlings. Plant Cell Tissue Organ Culture 1: 247-254. 4. Glazebrook, J., and Ausubel, F. M. 1992. The role of Arabidopsis phytoalexin in defense against bacterial pathogens. Sixth international symposium on molecular plant-microbe interactions. Abstract 171 APPENDIX I116 0 ,1 I... 0 BRASSININ (Brassica campestris) 0:01... 1 H CYCLOBRASSININ (Brassica campestris) 0M0 NH / N S 8M0 4—METHOXYBRASSININ (Brassica oleracoa) .55.»... SPIROBRASSININ (Raphanus sativus) Figure 17. 117 | NH 1:1 8’ Shin 0M0 METHOXYBRASSININ (Brassica campestrls) (:61... CYCLOBRASSININ SULPHOXIDE (Brassica )uncea) (01.141 BRASSILEXIN (Brassica juncoa) 0M0 METHOXYBRASSITIN (Brassica campestris) Phytoalexins and stress metabolites of crucifers. 118 l a N fiOMo 1'1 0 BRASSICANAL C (Brassica oleracoe) 0M0 METHOXYBRASSENIN A (Brassica oloracoa) 73 9 I 8 1'! M00 H CAMALEXIN Camelina sativa a) rabldopsls thallan HO 0 .. ..1 l H BRASSICANAL A (Brassica oleracoa) 0 II 0'31 l 0M0 METHOXYBRASSENIN B (Brassica oleracoa) 0:0 METHOXYCAMALEXI N (Camelina sativa) DIOXIBRASSININ (Brassica oleracoa) Figure 17. Continued. lllllllll