Plant lectin-like antibacterial proteins from phytopathogens Pseudomonas syringae and Xanthomonas citri

Maarten G. K. Ghequire1, Wen Li1, Paul Proost2, Remy Loris3,4 and René De Mot1*

1 Centre of Microbial and Plant Genetics, KU Leuven, Kasteelpark Arenberg 20, 3001 Heverlee, Belgium

2 Laboratory of Molecular Immunology, Rega Institute for Medical Research, KU Leuven, 3000 Leuven, Belgium

3Molecular Recognition Unit, Department of Structural Biology, Vlaams Instituut voor Biotechnologie, 1050 Brussel, Belgium

4Structural Biology Brussels, Department of Biotechnology (DBIT), Vrije Universiteit Brussel, 1050 Brussel, Belgium

*For correspondence. E-mail ; Tel. +32 (0)16329681, Fax: +32(0)16 321963

Running title: P. syringae and X. citri lectin-likebacteriocins

Keywords: bacteriocin, MMBL lectin, LlpA, antibacterial activity

Abstract

The genomes of Pseudomonas syringae pv. syringae 642 and Xanthomonas citri pv. malvacearum LMG 761 each carry a putative homologue of the plant lectin-like bacteriocin (llpA) genes previously identified in the rhizosphere isolate Pseudomonas putida BW11M1 and the biocontrol strain Pseudomonas fluorescens Pf-5. The respective purified recombinant proteins, LlpAPss642 and LlpAXcm761, display genus-specific antibacterial activity across species boundaries.The inhibitory spectrum of the P. syringae bacteriocin overlaps partially with those of the P. putida and P. fluorescens LlpAs. Notably, Xanthomonas axonopodis pv. citri str. 306 secretes a protein identical to LlpAXcm761. The functional characterization of LlpA proteins from two different phytopathogenic γ-proteobacterial species expandsthe lectin-like bacteriocin family beyond the Pseudomonas genus and suggests its involvement in competition among closely related plant-associated bacteria with different lifestyles.

Introduction

The ability for bacteria to produce antagonistic compounds targeting competing micro-organisms is an important asset for colonization of nutrient-proficient environments. In such highly competitive habitats the secretion of antagonistic molecules can provide the producing bacterium with a significant eco-evolutionary advantage. Characterization of antimicrobial compounds in bacteria has revealed an enormous diversity of the molecular armamentarium. Traditionally, these molecules are divided in two major classes: secondary metabolites (antibiotics) with broad-spectrum activity and ribosomally-synthesized peptides or proteins (bacteriocins) with narrow killing spectrum (Riley, 1998; Hibbing et al., 2010). Bacteriocins vary in structural complexity from small peptides to large multisubunit complexesacting as molecular machines (Ennahar et al., 2000; Michel-Briand and Baysse, 2002; Papagianni, 2003). Their compound-specific mode of action ranges from, for example, membrane pore formation to inhibition of cellwall biosynthesis or degradation of nucleic acids.

Among Gram-negative bacteria, next to colicins produced by E. coli strains,pyocins produced by Pseudomonas aeruginosa have been extensively studied (reviewed by Michel-Briand and Baysse, 2002). S-type pyocins are multi-domain toxins with a colicin-like modular structure (Denayer et al., 2007; Ling et al., 2010). Also colicin M-like bacteriocins have been identified in P. aeruginosa (Barreteau et al., 2009). The F- and R-type pyocins consist of multi-subunit particles resembling bacteriophage tails (Nakayama et al., 2000; Williams et al., 2008). Several reports have documented the ecological significance of pyocins for competitive interactions among P. aeruginosa strains (Inglis et al., 2009; Waite and Curtis, 2009; Bakkal et al., 2010).

An unusual type of Pseudomonas bacteriocin was found in the rhizosphere isolate Pseudomonas putida BW11M1 (LlpABW11M1; Parret et al., 2003) and biocontrol strain Pseudomonas fluorescens Pf-5 (LlpA1Pf-5 and LlpA2Pf-5; Parret et al., 2005). These proteins consist of a tandem of monocot mannose-binding lectin (MMBL) domains, sharing similarity with certain plant lectins. Both domains of the antibacterial lectins seem to have evolved independently but no specific function has yet been assigned to the individual domains. The phylogenetically unrelated Gram-positive species Ruminococcus albusproduces albusin B, a bacteriocin with a single MMBL domain in addition to a second domain of unknown function (Chen et al., 2004). In recent years, several genes encoding hypothetical lectins with one or two MMBL domain(s), in some cases fused to additional unknown domains, have emerged from bacterial genomic sequencing projects.

In this study we report on the functional characterization of two novel tandem-MMBL bacteriocins in plant-associated -proteobacteria with a different lifestyle, namely pathovars of Pseudomonas syringae and Xanthomonas citri. This reveals a wider distribution of the LlpA family proteins, now including bacteriocins from strains belonging to two major groups of plant pathogens (Ryan et al., 2011; Silby et al., 2011).

Results and discussion

New candidate lectin-like bacteriocins

Several hypothetical bacterial proteins with a tandem MMBL structure were identified by BlastP homology searches ( with the known PseudomonasLlpA sequences as amino acid queries. Hits were found forpredicted gene products of P. syringae (two pathovars) as well as for several strains of non-pseudomonads: Xanthomonas axonopodis(-Proteobacteria), Burkholderia cenocepacia and Burkholderia ambifaria(-Proteobacteria), and Arthrobacter sp. (Actinobacteria). A phylogenetic tree was constructed to compare these hypothetical proteins with the three previously characterized LlpAs (Fig. 1).The protein from P. syringae pv. aptatastr. DSM50252(Baltrus et al., 2011) is most similar to LlpABW11M1 (66% amino acid sequence identity), while the one from P. syringae pv. syringae 642 (Clarke et al., 2010) exhibits stronger divergence (< 50% identity to the known LlpAs). The lowest similarity is apparent for the Burkholderia proteins. Intermediate positions between the Pseudomonas and Burkholderia clusters are taken by the Arthrobacter and Xanthomonas sequences. The latter gene product (XAC0868) was fortuitously identified among extracellular proteins by Yamazaki et al. (2008) when studying HrpG-regulated protein secretion by Xanthomonas axonopodis pv. citri str. 306, but it was not further characterized. Using PCR primers based on the XAC0868-encoding gene (Supplementary Table S1) we identified an identical copy in X. axonopodis pv. malvacearumLMG 761. According to Schaad et al. (2006), these strains represent two distinct pathovars of Xanthomonascitri, of which only pv. citri causes bacterial cancer of citrus. The presence of a highly conserved xac0868 gene in both pathovars reflects their close phylogenetic relatedness, as apparent from multilocus sequence analysis (Young et al., 2008). No amplicon was detected for X. alfalfaesubsp. alfalfae LMG 497 (previously X. axonopodis subsp. alfalfae) and X. axonopodis pv. manihotis LMG 784 (data not shown). It was decided to undertake functional characterization of the presumptive LlpA-like bacteriocin from Xanthomonascitri pv. malvacearum LMG 761 (LlpAXcm761), along with the putative one from P. syringae pv. syringae 642 (LlpAPss642).

Expression of recombinant tandem MMBL lectins

Sequence-verified PCR amplicons of llpAPss642 and llpAXcm761 were ligated in pET28a as described by Parret et al. (2004). Fusions were created to encode N-terminal His6-tags, since difficulties were encountered in obtaining active recombinant LlpA proteins with a C-terminal His-tag (unpublished data). Unlike the Pseudomonas LlpAs, the predicted LlpAXcm761precursor is preceded by a cleavable type-II secretion (T2SS)recognition sequence(SignalP analysis; Consequently, two different constructs were made: oneretaining the Xanthomonas T2SS signal sequence and one devoid of it. In the latter case the polyhistidine tag fusion was immediately adjacent to the expected start of the mature LlpAXcm761 protein. Induction of expression, cell harvest and disruption, and extraction of the recombinant proteins was performed as described previously (Parret et al., 2004).

Analysis of induced cell lysates by SDS-PAGE revealed the expected size (33 kDaobserved versus 33,795 Dapredicted) for recombinant His6-taggedLlpAPss642 (Fig. 2). The presence of the histidine-tag, with removal of the N-terminal methionine, was confirmed by N-terminal amino acid sequencing (according to Parret et al., 2003) and immunodetection by Western blot using polyhistidine-specific antibodies (Roche Diagnostics) (data not shown).

For LlpAXcm761, no recombinant protein expression was observed with the construct lacking the cognate T2SS signal sequence, while the one retaining this sequence yielded a protein band (23 kDa) significantly smaller than expected for the His6-tagged product ( 30,244 Da)(Fig. 2).This protein was however not detected with the anti-polyhistidine antibodies. In line with this, N-terminal sequencing yielded QMLRANFPGQ, corresponding to the N-terminal sequence of the mature form of LlpAXcm761following removal of the T2SS signal sequence and the preceding His6-tag. The same proteolytic processing was observed by Chen et al. (2004) for the LlpAXcm761counterpart, protein XAC0868, upon secretion by X. axonopodis pv. citri, indicating the unexpected heterologous expression of the native form of LlpAXcm761.

Purification of recombinant proteins

Purification of recombinant His6-tagged LlpAPss642 was performed on Ni-NTA agarose by HPLC, using an Äkta Purifier (GE Healthcare Life Sciences, Amersham Bioscience) with a 5ml Histrap column (GE Healthcare Life Sciences) using buffers, washing steps and elution conditions described previously (Parret et al., 2004). Due to the lack of the His6-tag in recombinant LlpAXcm761, it was purified by cation ion exchange chromatography (given the high predicted pI of mature LlpAXcm761: 8.66) instead of affinity chromatography, using a SP Sepharose HP column (XK 16/20, GE Healthcare Life Sciences) in combination with an Äkta Purifier. At pH 6.0 with malonic acid (50 mM) only contaminating proteins from the raw lysate were retained on the column. A second purification step with adsorption at pH 4.8 (50 mM acetic acid) and salt gradient elution (flow rate 2 ml/min; 0 to 1 M NaCl in 100 min) resulted in elution of LlpAXcm761without contaminating proteins at a NaCl concentration of approx. 300 mM.

Both recombinant LlpAPss642 and LlpAXcm761 were dialyzed against bis-tris propane buffer (20 mM, NaCl 200 mM, pH 7.0). As the N-terminal polyhistidine tag does not interfere with LlpA activity (Parret et al., 2004), it was not removed from LlpAPss642. The concentration of the purified proteins was determined by UV absorbance measurement (A280), using molar extinction coefficients of 60,850 M-1 cm-1 for His6-LlpAPss642and 38,055 M-1 cm-1 for LlpAXcm761, calculated according to Pace et al. (1995).

Bacteriocin activity of novel LlpAs

Potential bacteriocin activity of His6-LlpAPss642and LlpAXcm761was assayed by applying 10 µl spots of pure protein (conc. 1-2 mg/ml) on TSB agar plates, overlaid with 5 ml soft agar (0.5%) seeded with 200 µl of an overnight culture (108 CFU/ml) of the respective indicator strains. Plates were incubated overnight at 30°C, except for P. aeruginosa (37°C). The following day, inhibitory activity was observed as a halo in a confluent lawn of bacterial cells (Fig. 3). Dialysis buffer was taken as negative control. The test panel consisted of representative Pseudomonas and Xanthomonas culture collections strains and fluorescent Pseudomonas rhizosphere isolates from our lab collection. In parallel, the sensitivity of these strains to purified LlpABW11M1(Parret et al., 2003) and LlpA1Pf-5(Parret et al., 2005) was determined. The results are summarized in Table 1 and Table S2.

For LlpAPss642, a sensitive P. syringae, P. fluorescens and P. resinovorans strain, along with two partially inhibited strains, were identified. The sensitive P. syringae strain (rifampicin-resistant derivative of GR12-2) was until recently described asa P. putida strain but now reclassified by Blakney and Patten (2011). The LlpAPss642activity indicates that LlpA-mediated killing is not restricted to strains of the same species, as observed before for LlpABW11M1(Parret et al., 2003) and LlpA1Pf-5 and LlpA2Pf-5 (Parret et al., 2005). The LlpAPss642target strains are also sensitive to at least one other antibacterial lectin described before (Parret et al., 2003; Parret et al., 2005). Only a limited number of LlpAPss642-sensitive strains (about 4% of tested Pseudomonasstrains) were identified, compared to the much more frequently found LlpABW11M1-inhibited strains (about 42%) and LlpA1Pf-5-inhibited strains (about 12%).This is probably due to the biased species distribution in the test panel that is dominated by environmental isolates related to P. putida and P. fluorescens while only few strains of the producer’s taxonomic group (P. syringae) were included.

Growth of none of the Xanthomonas strains tested was affected by any of the Pseudomonas LlpAs. However, several of these strains (6 out of 16) were sensitive to recombinant LlpAXcm761. These target strains represent six different Xanthomonas species, including a pathovar of X. axonopodis. Much like the antagonism among Pseudomonas, LlpA-mediated inhibition is not confined to Xanthomonas strains belonging to the same species as the bacteriocin producer.In line with the intragenus-specific killing of the Pseudomonas LlpAs, none of the tested Pseudomonas strains was inhibited by LlpAXcm761.

Concluding remarks

Although bacteriocin production is also common among plant-associated bacteria, relatively few bacteriocins have been characterized at the molecular level (reviewed by De Mot, 2007). Bacteriocins LlpAPss642 and LlpAXcm761 identified in this work are unrelated to other bacteriocins produced by P. syringae strains (such as phage-like syringacins and colicin M-like bacteriocins: Smidt and Vidaver, 1986; Lavermicocca et al., 2002; Barreteau et al., 2009; Sisto et al., 2010) or by Xanthomonas species (glycinecin A: Heu et al., 2001; Pham et al., 2004).

Present results assigning a role as antibacterial molecules to thetandem-MMBL lectin-like proteins in the γ-Proteobacteria P. syringae and X. citri,add to the diversity of the LlpA family with respect to producers and targets. In line with previous results for LlpA proteins from P. putida and P. fluorescens strains, the antagonism is strain-specific and difficult to predict but appears to be confined to the cognate genus, albeit not limited by species boundaries. The Pseudomonas LlpA proteins share some target strains but display distinct inhibitory spectra. It will be of interest to characterize additional, more divergent LlpA family members, like the proteins from Burkholderia (β-Proteobacteria) and Arthrobacter (Actinobacteria) to verify whether the intra-genus but species border-crossing activity applies more widely. In addition, this information may provide more insight in the evolutionary origin of these antibacterial proteins and their relatedness to the MMBL-type lectins broadly distributed in monocot plants. Conceivably, the ancestral antibacterial activity may have evolved following acquisition of such eukaryotic gene by a plant-colonizing, possibly phytopathogenic bacterium.

Our current research aims at elucidation of the role played by the respective MMBL domains in overall structure and toxic activity and at identification oftarget specificity determinants. Of additional interest is the apparent difference in secretion pathways that is inferred from the presence of a T2SS signal sequence in some LlpAs (Xanthomonas, Burkholderia) but not in others (Pseudomonas, Arthrobacter). For the latter, alternative export routes may be involved. Recently, delivery of antibacterial effector molecules by the type-VI secretion machinery has been reported for several bacteria including P. aeruginosa (Russell et al., 2011). In X. citri pv. citri str. 308 (previously X. axonopodis pv. citri), secretion of the counterpart of LlpAXcm761 (XAC0868) is apparently coordinated with HrpG-dependent type III secretion, which plays a crucial role in infection and pathogenicity (Yamazaki et al., 2008). This suggests that the antibacterial activity of this protein may be important for competition with other Xanthomonas in the host plant environment prior to or during establishment of the infection.

Acknowledgements

This work was supported by grant G.0393.09N from FWO-Vlaanderen (to R.D.M. and R.L.). The authors wish to thank B. Vinatzer (Department of Plant Pathology, Virginia Tech, USA) for supplying genomic DNA ofPseudomonas syringae pv. syringae 642, M. Höfte and J. Xu (Laboratory of Phytopathology, Ghent University, Belgium) for assisting in antagonistic testing with different Xanthomonas oryzae pv. oryzae strains, J. Desair (CMPG, KU Leuven) for assisting in purification of the recombinantXanthomonas lectin, and the Interfaculty Centre for Proteomics and Metabolomics (ProMeta, KU Leuven) for MALDI-TOF analysis.

References

  1. Bakkal, S., Robinson, S.M., Ordonez, C.L., Waltz, D.A., and Riley, M.A. (2010) Role of bacteriocins in mediating interactions of bacterial isolates taken from cystic fibrosis patients. Microbiology156: 2058-2067
  2. Baltrus, D.A., Nishimura, M.T., Romanchuk, A., Chang, J.H., Mukhtar, M.S., Cherkis, K., et al. (2011) Dynamic evolution of pathogenicity revealed by sequencing and comparative genomics of 19 Pseudomonas syringae isolates. PLoS Pathog7: e1002132
  3. Barreteau, H., Bouhss, A., Fourgeaud, M., Mainardi, J.L., Touzé, T., Gérard, F., et al.(2009) Human- and plant-pathogenic Pseudomonas species produce bacteriocins exhibiting colicin M-like hydrolase activity towards peptidoglycan precursors. J Bacteriol191: 3657-3664
  4. Blakney, A.J., and Patten, C.L. (2011) A plant growth-promoting pseudomonad is closely related to the Pseudomonas syringae complex of plant pathogens. FEMS Microbiol Ecol77: 546-557
  5. Chen, J., Stevenson, D.M., and Weimer, P.J. (2004) Albusin B, a bacteriocin from the ruminal bacterium Ruminococcus albus 7 that inhibits growth of Ruminococcus flavefaciens. Appl Environ Microbiol70: 3167-3170
  6. Clarke, C.R., Cai, R., Studholme, D.J., Guttman, D.S., and Vinatzer, B.A. (2010) Pseudomonas syringae strains naturally lacking the classical P. syringae hrp/hrc locus are common leaf colonizers equipped with an atypical type III secretion system. Mol Plant Microbe Interact23: 198-210
  7. De Mot, R. (2007) Bacteriocins of plant-associated bacteria and their potential as biocontrol agents against phytopathogens. In Research and applications in bacteriocins. Riley, M., and Gillor, O. (eds). Norfolk: Horizon Bioscience, pp. 131-151
  8. Denayer, S., Matthijs, S., and Cornelis, P. (2007) Pyocin S2 (Sa) kills Pseudomonas aeruginosa strains via the FpvA type I ferripyoverdine receptor. J Bacteriol189: 7663-7668
  9. Drummond, A.J., Ashton, B., Buxton, S., Cheung, M., Cooper, A., Duran, C., et al. (2011) Geneious v5.4. Available from:
  10. Ennahar, S., Sashihara, T., Sonomoto, K., and Ishizaki, A. (2000) Class IIa bacteriocins: biosynthesis, structure and activity. FEMS Microbiol Rev24: 85-106
  11. Guindon, S., and Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol52: 696-704
  12. Heu, S., Oh, J., Kang, Y., Ryu, S., Cho, S.K., Cho, Y., and Cho, M. (2001) gly gene cloning and expression and purification of glycinecin A, a bacteriocin produced by Xanthomonas campestris pv. glycines 8ra. Appl Environ Microbiol67: 4105-4110
  13. Hibbing, M.E., Fuqua, C., Parsek, M.R., and Peterson, S.B. (2010) Bacterial competition: surviving and thriving in the microbial jungle. Nat Rev Microbiol8: 15-25
  14. Inglis, R.F., Gardner, A., Cornelis, P., and Buckling, A. (2009) Spite and virulence in the bacterium Pseudomonas aeruginosa. Proc Natl Acad Sci USA 106: 5703-5707
  15. Lavermicocca, P., Lonigro, S.L., Valerio, F., Evidente, A., and Visconti, A. (2002) Reduction of olive knot disease by a bacteriocin from Pseudomonassyringae pv. ciccaronei. Appl Environ Microbiol68: 1403-1407
  16. Ling, H., Saeidi, N., Rasouliha, B.H., and Chang, M.W. (2010) A predicted S-type pyocin shows a bactericidal activity against clinical Pseudomonas aeruginosa isolates through membrane damage. FEBS Lett584: 3354-3358
  17. Michel-Briand, Y., and Baysse, C. (2002) The pyocins of Pseudomonas aeruginosa. Biochimie 84: 499-510
  18. Nakayama, K., Takashima, K., Ishihara, H., Shinomiya, T., Kageyama, M. Kanaya, S., et al. (2000) The R-type pyocin of Pseudomonas aeruginosa is related to P2 phage, and the F-type is related to lambda phage. Mol Microbiol38: 213-231
  19. Pace, C.N., Vajdos, F., Fee, L., Grimsley, G., Gray, T. (1995) How to measure and predict the molar absorption coefficient of a protein. Protein Sci4: 2411-2423
  20. Papagianni, M. (2003) Ribosomally synthesized peptides with antimicrobial properties: biosynthesis, structure, function, and applications. Biotechnol Adv21: 465-499
  21. Parret, A.H.A., Schoofs, G., Proost, P., and De Mot, R. (2003) Plant lectin-like bacteriocin from a rhizosphere-colonizing Pseudomonas isolate. J Bacteriol185: 897-908
  22. Parret, A.H.A., Temmerman, K., and De Mot, R. (2005) Novel lectin-like bacteriocins of biocontrol strain Pseudomonas fluorescens Pf-5. Appl Environ Microbiol71: 5197-5207
  23. Parret, A.H.A., Wyns, L., De Mot, R., and Loris, R. (2004) Overexpression, purification and crystallization of bacteriocin LlpA from Pseudomonas sp. BW11M1. Acta Crystallogr D Biol Crystallogr60: 1922-1924
  24. Pham, H.T., Riu, K.Z., Jang, K.M., Cho, S.K., and Cho, M. (2004) Bactericidal activity of glycinecin A, a bacteriocin derived from Xanthomonas campestris pv. glycines, on phytopathogenic Xanthomonas campestris pv. vesicatoria cells. Appl Environ Microbiol70: 4486-4490
  25. Riley, M.A. (1998) Molecular mechanisms of bacteriocin evolution. Annu Rev Genet32: 255-278
  26. Russell, A.B., Hood, R.D., Bui, N.K., Leroux, M., Vollmer, W., and Mougous, J.D. (2011) Type VI secretion delivers bacteriolytic effectors to target cells. Nature475: 343-347
  27. Ryan, R.P., Vorhölter, F.J., Potnis, N., Jones, J.B., Van Sluys, M.A., Bogdanove, A.J., and Dow, J.M. (2011) Pathogenomics of Xanthomonas: understanding bacterium-plant interactions. Nat Rev Microbiol9: 344-355
  28. Schaad, N.W., Postnikova, E., Lacy, G., Sechler, A., Agarkova, I., Stromberg, P.E., et al. (2006) Emended classification of xanthomonad pathogens on citrus. Syst Appl Microbiol29: 690-695
  29. Silby, M.W., Winstanley, C., Godfrey, S.A., Levy, S.B., and Jackson R.W. (2011) Pseudomonas genomes: diverse and adaptable. FEMS Microbiol Rev35: 652-680
  30. Sisto, A., Cipriani, M.G., Morea, M., Lonigro, S.L., Valerio, F., and Lavermicocca, P. (2010) An Rhs-like genetic element is involved in bacteriocin production by Pseudomonas savastanoi pv. savastanoi. Antonie Van Leeuwenhoek98: 505-517
  31. Smidt, M.L., and Vidaver, A.K. (1986) Isolation and characterization of syringacin W-1, a bacteriocin produced by Pseudomonas syringae pv. syringae. Can J Microbiol32: 231-236
  32. Waite, R.D., and Curtis, M.A. (2009) Pseudomonas aeruginosa PAO1 pyocin production affects population dynamics within mixed-culture biofilms. J Bacteriol191: 1349-1354
  33. Williams, S.R., Gebhart, D., Martin, D.W., and Scholl, D. (2008) Retargeting R-type pyocins to generate novel bactericidal protein complexes. Appl Environ Microbiol74: 3868-3876
  34. Yamazaki, A., Hirata, H., and Tsuyumu, S. (2008) HrpG regulates type II secretory proteins in Xanthomonas axonopodis pv. citri. J Gen Plant Pathol74: 138-150
  35. Young, J.M., Park, D.C., Shearman, H.M., and Fargier, E. (2008) A multilocus sequence analysis of the genus Xanthomonas. Syst Appl Microbiol31: 366-377

Table legends