Mol Gen Genet (1990) 222:157-160 © Springer-Verlag 1990
Cloning of genes involved in negative regulation of production of extracellular enzymes and polysaccharide of Xanthomonas campestris pathovar campestris Ji-Liang Tang*, Clare L. Gough**, and Michael J. Daniels The Sainsbury Laboratory, John Innes Institute, Colney Lane, Norwich NR4 7UH, UK Summary. A recombinant plasmid pIJ3079 contains D N A sequences from Xanthomonas campestris pv campestris involved in coordinate negative regulation of production of the extracellular enzymes protease, endoglucanase, amylase and polygalacturonate lyase, and extracellular polysaccharide (EPS). Wild-type bacteria harbouring pIJ3079 and therefore carrying extra copies of the gene(s) therein showed reduced enzyme and EPS production and reduced aggresiveness to plants. Localised Tn5 mutagenesis of the corresponding region of the genome gave mutants producing higher levels of enzymes and EPS than the wild type, suggesting that the gene(s) may negatively regulate production in the normal cell. Enzyme and EPS production in the mutants was still dependent on previously characterised positive regulatory genes. Key words: Amylase - Endoglucanase - Protease - Polygalacturonate lyase - Pathogenicity
existence of a parallel system which negatively regulates the synthesis of the same factors in a coordinate manner. Techniques. Bacterial strains and plasrnids, culture conditions, genetic and molecular genetic techniques were as described by Daniels et al. (1984a, b) and Turner et al. (1984; 1985). Detection and assay of enzymes and EPS were described by Dow et al. (1987) for PGL, Tang et al. (1987) for PRT, Gough et al. (1988) for EGL and Barr+re et al. (1986) for EPS. A M L activity was detected by growing bacteria on plates supplemented with 0.1% soluble starch developed after incubation with I2/KI solution. Enzyme activities were deduced from diameters of haloes or zones of clearing around 5 m m diameter wells containing enzyme solutions, by comparison with standard curves. Pathogenicity to turnip seedlings (Brassica campestris vat Just Right) and bacterial growth in seedlings were tested as described previously (Daniels et al. 1984b; Tang et al. 1987). Mature turnip plants were also inoculated at the margins of leaves (Gough et al. 1988).
Little is known about the mechanisms by which plant pathogenic bacteria regulate the production of pathogenicity factors, including plant cell wall degrading enzymes and extracellular polysaccharides (EPS) (Daniels et al. 1988). The crucifer pathogen J(anthomonas campestris pathovar campestris produces extracellular enzymes including amylase (AML), endoglucanase (EGL), polygalacturonate lyase (PGL) and protease (PRT) together with EPS (xanthan gum). Genes encoding PRT (Tang et al. 1987), EGL (Gough et al. 1988) and P G L (Dow et al. 1989), and genes involved in EPS production (Barr+re et al. 1986) have been cloned and studied in some detail. A non-pathogenic mutant, 8237, deficient in synthesis of all the extracellular enzymes and EPS, could be complemented by genes cloned in pIJ3020 (Daniels et al. 1984a). pIJ3020 contains at least seven genes involved in a "global" system which positively regulates enzyme and EPS synthesis (Daniels et al. 1989; Tang 1989). Here we demonstrate the
Isolation of plJ3079. A genomic library of X.c. campestris D N A (Gough et al. 1988) in pLAFR3 (Staskawicz et al. 1987) was transferred from Escherichia coli HB101 (Boyer and Roulland-Dussoix 1969) into X.c. campestris 8004 (wild type, Turner et al. 1984). Of 2000 transconjugant colonies tested, one, harbouring the plasmid designated pIJ3079, produced much lower levels of enzymes and EPS while all the others gave wild-type levels (Table 1). Since 8004/pIJ3079 probably contains at least five copies of the genes in the pIJ3079 region of the genome, the data suggest that increasing the copy number of these sequences represses the production of extracellular enzymes and EPS. Lysis of bacteria by exposure of plates to CHC13 vapour (Tang et al. 1987) gave no evidence for increased intracellular or periplasmic accumulation of enzymes in 8004/pIJ3079, such as is observed in X.c. campestris mutants defective in enzyme export rather than synthesis (Dow et al. 1987).
* Present address: Laboratory of Molecular Genetics, Guangxi Agricultural College, Nanning, China ** Present address: Laboratoire de Biologic Molrculaire des Relations Plantes-Microorganismes, INRA-CNRS, 31326 Castanet-Tolosan, France
Effect of pIJ3079 on bacterial growth and pathogenicity. 8004/pIJ3079 was prototrophic and achieved the same growth rate and cell density as 8004 in both complete and minimal media (plus tetracycline in the case of 8004/pIJ3079), suggesting that the plasmid does not affect general protein synthesis in X.c. campestris.
Offprint requests to: M.J. Daniels
158 Table 1. Production of extracellular enzymes and extracellular polysaccharide of Xanthomonas campestris pathovar campestris strainsa Strains
8004 8004/pIJ3079 8004/pIJ30792 8004/pIJ30793 8004/pIJ30794 8004/pIJ30795 8004/pIJ30796 8004/pIJ30797 8004/pIJ30798 8004/pIJ307910 8004::Tn5-2 8004::Tn5-4 8004 : : Tn5-2 /pIJ3079 8237 8237/pIJ3079 8237::Tn5-2 8237::Tn5-4
Relative level (plate assays)b
1 0.13 0.98 0.13 1 0.13 0.14 0.t0 0.09 0.09 3.70 3.50 ND
i 0.59 0.97 0.55 0.97 0.50 0.55 0.57 0.55 0.59 1.13 :[.13 ND
1 0.29 ND ND ND ND ND ND ND ND 3.48 ND 0.32
0.09 0.06 0.13 0.:[3
0.68 0.50 0.61 0.69
0.11 0 0.:[5 ND
0.08 t 0.16 0.17 0.17 0.16 0.2:[ 2.97 2.97 ND
0.04 1 0.12 0.08 0.13 0.06 0.14 2.52 2.58 ND
1 0.12 0.97 0.12 0.96 0.12 0.12 0.:[2 0.12 0.12 3.10 3.:[0 ND
0.:[3 0.04 0.14 0.14
0.07 0 0.:[0 0.10
0.08 0.04 0.:[:[ 0.11
ND, not done a The relative extracellular polysaccharide production was obtained by measuring the mean diameter of a sample of 20 colonies 4 days after inoculation of 106 cfu (5 gl drops) on plates of NYGA plus 2% glucose, and taking the diameter of the wild-type 8004 colonies as 1.0 b PRT, protease; PGL, polygalacturonate lyase; EGL, endoglucanase; AML, amylase; EPS, extracellular polysaccharide
In turnip seedlings, 8004/pIJ3079 had a similar growth pattern to 8004 after inoculation with 105 colony forming units (cfu) per seedling (cf. G o u g h et al. 1988), b u t after inoculation with 102 cfu per seedling the growth rate o f 8004/pIJ3079 was less than half that o f 8004 and the p o p u lation after 5 days was 1-2 orders o f magnitude less (data not shown). Reisolation o f bacteria from seedlings infected with 8004/plJ3079 showed that the plasmid was lost at a rate o f ca. 5% per generation. However, although much of the bacterial growth was accounted for by 8004 cured o f pIJ3079, 8004/pIJ3079 was capable o f multiplying in seedlings, shown by an increase in the n u m b e r o f tetracycline resistant cells. 8004/pIJ3079 gave much less extensive disease symptoms than 8004 after inoculation at the margins o f turnip leaves. Seedlings inoculated with 105 cfu 8004/pIJ3079 showed typical symptoms (Daniels et al. 1984b), but the development was retarded by i d a y compared with 8004. Inoculation with 10 z cfu gave n o r m a l symptoms with 8004, but little damage was caused by 8004/pIJ3079. The instability o f pIJ3079 suggests that the symptoms observed were caused by 8004 cured o f the plasmid. Deletion analysis o f pIJ3079. A n E c o R I restriction m a p o f pIJ3079 (Fig. 1) was obtained by analysing a set o f deletion derivative constructed by partial E c o R I digestion and religation o f fragments larger than 22 kb. The derivatives were transferred into Jf.c. campestris 8004 and the transconjugants tested for p r o d u c t i o n of enzymes and EPS. The results are shown in Table 1 and Fig. 1. All the derivative plasmids which h a d lost the phenotype o f pIJ3079 (i.e. pIJ30792 and pIJ30794) lacked the 2.7 kb E c o R I fragment, but 8004 carrying this fragment cloned in b o t h orientations in p L A F R 3 produced wild-type levels o f enzymes and EPS. These data
Phenotype of 8004/ptasmid RI1.3R1.4 IR 2.7 R
I I - - I
- - [ - - I - - I - - I - - [
I ................................................. ~......................................... } .................................................................. t ..................................
Fig. 1. Restriction map, deletion analysis and Tn5 mutagenesis of pIJ3079. The numbers above the map indicate the distances in kb between EcoRI sites, marked R. The open bars represent the insert DNA and the dotted lines indicate deleted portions in the derivatives. The Tn5 insertions were only mapped to the EcoR1 fragments. The phenotype + or -- refers to the production or non-production of the set of factors AML, EGL, PGL, PRT and EPS by X.c. campestris 8004 harbouring the plasmids or, at the bottom of the figure, by 8004 and the two Tn5 mutants derived by marker-exchange from pIJ3079 :: Tn5. ND, not done
159 suggest that the 2.7 kb fragment carries sequences necessary but not sufficient for the function of pIJ3079, and that other sequences, probably within the 3.8 or 13.1 kb EcoRI fragments, are also necessary.
Mutation of plJ3079. A collection of Tn5 insertions in pIJ3079 was tested for its ability to depress enzyme and EPS production in 8004. Two out of 140 colonies tested showed wild-type enzyme and EPS levels (i.e. the function of pIJ3079 had been lost), while all the remainder retained the phenotype of 8004/pIJ3079. No mutants were found which were positive for some enzymes but negative for others. The plasmids which had lost the function, pIJ3079 : : Tn5-2 and pIJ3079: : Tn5-4, had insertions within the 13.1 kb and the 2.7 kb EcoRI fragments, respectively. Examination of a random sample of the 140 mutant plasraids showed that insertion had occurred into all the EcoRI fragments. The properties of the mutant plasmids, together with the deletion data, indicate that the presumed regulatory activity of pIJ3079 is located in the 2.7 and 13.1 kb fragments. Attempts to map the Tn5 insertions more precisely were frustrated by the lack of convenient restriction enzyme sites (e.g. HindIII and BamHl) in pIJ3079. Tn5 insertions -2 and -4 were transferred from pIJ3079 into the corresponding sites of the X.c. campestris genome by markerexchange (Turner et al. 1985). The resulting mutants, 8004: : Tn5, both produced about 3 times more AML, EGL, PGL and PRT, and more EPS than 8004 (Table 1), suggesting that the disrupted gene(s) may be involved in normal negative regulation of the production of extracellular enzymes and EPS in X.c. campestris. The overproduction of P G L in the 8004 : : Tn5 mutants was eliminated by introduction of pIJ3079. Interaction of the genes in the pIJ3079 and plJ3020 regions.A mutant X.c. campestris strain, 8237, produces 7-10 times less extracellular enzymes and EPS than the wild type, and the mutation can be complemented by the recombinant plasmid pIJ3020 (Daniels et al. 1984a; Tang 1989). pIJ3079 was introduced into 8237 and the resulting strain produced significantly lower levels of enzymes and EPS than either 8237 or 8004/pIJ3079. Double mutants constructed by transferring Tn5 insertions -2 and -4 from pIJ3079 into the genome of 8237 produced only low levels of enzymes and EPS, similar to those given by 8237 and 8004/pIJ3079. Thus the enhanced production characteristic of 8004: : Tn52 and -4 was not obtained when the mutation of 8237 was also present. The insert D N A of pIJ3079 did not hybridise with pIJ3020 (or with structural genes for EGL, PGL and P R T or genes involved in enzyme export). Discussion
We have previously cloned structural genes encoding certain X.c. campestris extracellular enzymes, (AML, EGL, P G L (one of three isozymes) and PRT) and enzymes of EPS biosynthesis and demonstrated that at least some of these factors are important for pathogenicity. Investigation of the regulation of enzyme and EPS biosynthesis may contribute to an understanding of the ecology of X.e. campestris and to host-pathogen interaction. A cluster of genes cloned in the plasmid pIJ3020 is involved in positive global regulation of synthesis o f A M L , AGL, PGL, PRT and EPS (Daniels et al. 1984a; Tang 1989), and the present paper
describes a "balancing" negative regulatory system. The increase in copy number of the negative regulatory genes from one in 8004 to ca. five in 8004/pIJ3079 may give a concomitant increase in the intracellular concentration of regulatory gene products, thereby titrating the molecular target or otherwise altering the balance between positive and negative regulation. The sequences thus found in pIJ3079 did not affect growth (and presumably did not therefore reduce general protein synthesis), but depressed aggressiveness to plants, probably as a consequence of the effect on enzyme and EPS synthesis. The residual symptoms produced were similar to those incited by mutants such as 8237 which also produce lower enzyme levels but for other reasons (Daniels et al. 1984a). Mutants in which the chromosomal copy of the pIJ3079 gene(s) had been inactivated overproduced enzymes and EPS in culture and retained pathogenicity, but we have not determined whether increased enzyme and EPS synthesis also takes place when the bacteria are inside the plants and, if so, whether they are hyperaggressive compared with the wild type. Sequencing has shown that some of the positive regulatory genes in the pIJ3020 cluster have the characteristics of "two component" regulatory systems, suggesting that regulation takes place at the level of transcription (Tang 1989; Daniels et al. 1989; Y.N. Liu, J.L. Tang and M.J. Daniels, unpublished). A number of general models for action of the negative regulator system can be envisaged: (1) the pIJ3079 sequences act by blocking the activation of enzyme and EPS biosynthesis by the positive regulators (pIJ3020), or (2) vice versa: the positive elements block the pIJ3079 system, or (3) both positive and negative systems operate independently, in parallel but in opposition, on the target genes. Results of preliminary experiments to study the interaction of the pIJ3020 and pIJ3079 systems (Table 1) showed that transfer of pIJ3079 into 8237 (with a mutation in the pIJ3020 cluster) further depresses enzyme levels below the values given by either 8237 or 8004/pIJ3079 and double mutants 8237: : Tn5 with defects in both systems produced low enzyme and EPS levels similar to 8237 or 8004/pIJ3079, indicating that over-production seen in 8004 :: Tn5 depends on the function of the pIJ3020 system. These results probably exclude model (2) above, but much more detailed genetic and biochemical experimentation will be necessary to understand the systems and their interactions fully. Two of the target genes have been sequenced, those for EGL (Gough et al. 1990) and PRT (Liu et al. 1990), and the sequence data will facilitate the search for D N A binding proteins which might be products of regulatory genes. Nothing is yet known about environmental or physiological factors which modulate the global regulatory systems, either positive or negative. The products of the genes which are coordinately regulated have little in common except that they (or the products of their enzymatic action in the case of the enzymes of EPS biosynthesis) contribute to symptom development in infected plants or degrade plant polymers. It is likely that any regulatory effectors will be substances of plant origin. Coordinate regulation by plantderived substances of expression of bacterial genes involved in interaction with plants is well established for Rhizobium and Agrobacterium (Firmin et al. 1986; Peters et al. 1986; Redmond et al. 1986; Stachel et al. 1986), but it has often been assumed that the apparently less specialised mode of interaction shown by necrotrophic pathogens would make
160 such refinements unecessary. X.c. eampestris is believed to be associated in nature exclusively with plants, growing as an epiphyte on external plant surfaces or as a parasite within tissues, principally the vascular system (Williams 1980). Perhaps these two types of environment are so dissimilar that extensive changes in patterns of gene expression are required u p o n transition from the surface to internal tissues. Alternatively, the changes in the physico-chemical properties of the infected plant tissues which accompany the progression of symptom development may require co-ordinate changes in gene expression by the bacteria.
Acknowledgements. JLT was supported by a studentship from the Government of the Chinese People's Republic and Guangxi Agricultural College and CLG by a studentship from the Agricultural and Food Research Council. This work was carried out under the provision of Licence PHF1185/8(48) issued by the Ministry of Agriculture, Fisheries and Food under the Plant Health (Great Britain) Order 1987. The Sainsbury Laboratory is supported by a Grant from the Gatsby Charitable Foundation. References Barr6re GC, Barber CE, Daniels MJ (1986) Molecular cloning of genes involved in the biosynthesis of the extracellular polysaccharide xanthan by Xanthomonas campestris pv. eampestris. Int J Biol Macromol 8 : 372-374 Boyer HW, Roulland-Dussoix D (1969) A complementation analysis of the restriction and modification of DNA in Escherichia eoli. J Mol Biol 41:459-47 Daniels M J, Barber CE, Turner PC, Sawczyc MK, Byrde RJW, Fielding AH (1984 a) Cloning of genes involved in pathogenicity of Xanthomonas campestris pv. campestris using the broad host range cosmid pLAFR1. EMBO J 3:3323-3328 Daniels M J, Barber CE, Turner PC, Cleary WG, Sawczyc MK (1984b) Isolation of mutants of Xanthomonas campestris pv. campestris showing altered pathogenicity. J Gen Microbiol 130:2447-2455 Daniels MJ, Dow JM, Osbourn AE (1988) Molecular genetics of pathogenicity in phytopathogenie bacteria. Annu Rev Phytopathol 26:285 312 Daniels M J, Osbourn AE, Tang JL (1989) Regulation in Xanthomonas-plant interactions. In: Lugtenberg BJJ (ed) Signal Molecules in Plants and Plant-Microbe Interactions. Springer Verlag, Berlin, Heidelberg, pp 189-196 Dow JM, Scofield G, Trafford K, Turner PC, Daniels MJ (1987) A gene cluster in Xanthomonas campestris pv. eampestris required for pathogenicity controls the excretion of polygalacturonate lyase and other enzymes. Physiol Mol Plant Pathol 31:261-27l Dow JM, Milligan DE, Jamieson L, Barber CE, Daniels MJ (1989) Molecular cloning of a polygalacturonate lyase gene from Xan-
thomonas campestris pv campestris and role of the gene product in pathogenicity. Physiol Mol Plant Pathol 35:113-120 Firmin JL, Wilson KE, Rossen L, Johnston AWB (1986) Flavonoid activation of nodulation genes in Rhizobium reversed by other compounds present in plants. Nature 324:90-92 Gough CL, Dow JM, Barber CE, Daniels MJ (1988) Cloning of two endoglucanase genes of Xanthomonas campestris pv. campestris: Analysis of the role of the major endoglucanase in pathogenicity. Mol Plant-Microbe Interactions 1:275-281 Gough CL, Dow JM, Keen J, Henrissat B, Daniels MJ (1990) Nucleotide sequence of the gene encoding the major endoglucanase of Xanthomonas eampestris pv campestris. Gene (in press) Liu YN, Tang JL, Clarke BR, Dow JM, Daniels MJ (1990) A multipurpose broad host range cloning vector and its use to characterise an extracellular protease gene of Xanthomonas eampestris pv campestris. Mol Gen Genet 220:433-440 Peters NK, Frost JW, Long SR (1986) A plant flavone, luteolin, induces expression of Rhizobium meliloti nodulation genes. Science 233:977-979 Redmond JW, Barley M, Djordjevic MA, Innes RW, Kuempel PL, Rolfe BG (a 986) Flavones induce expression of nodulation genes in Rhizobium. Nature 323:632-635 Stachel SE, Nester EW, Zambryski PC (1986) A plant cell factor induces Agrobacterium tumefaciens vir gene expression. Proc Natl Acad Sci USA 83 : 379--383 Staskawicz B, Dahlbeck D, Keen N, Napoli C (1987) Molecular characterization of cloned avirulence genes from race 0 and rase 1 of Pseudomonas syringae pv glycinea. J Bacteriol 169: 5789-5794 Tang J-L (1989) Aspects of extracellular enzyme production by Xanthomonas campestris. PhD Thesis, University of East Anglia, Norwich, UK Tang J-L, Gough C, Barber CE, Dow JM, Daniels MJ (1987) Molecular cloning of protease gene(s) from Xanthomonas campestris pv. campestris: expression in Escherichia coli and role in pathogenicity. Mol Gen Genet 210: 443-448 Turner P, Barber C, Daniels M (1984) Behavior of the transposons Tn5 and Tn7 in Xanthomonas campestris pv. campestris. Mol Gen Genet 195:101-107 Turner PC, Barber CE, Daniels MJ (1985) Evidence for clustered pathogenicity genes in Xanthomonas eampestris pv. eampestris. Mol Gen Genet 199:338-343 Williams PH (1980) Black rot: a continuing threat to world crucifers. Plant Dis 64:736-742
C o m m u n i c a t e d by H. Hennecke
Received November 13, 1989