BacA-Mediated Bleomycin Sensitivity in Sinorhizobium meliloti Is Independent of the Unusual Lipid A Modification
http://www.100md.com
《细菌学杂志》
Institute of Structural and Molecular Biology and Centre for Science at Extreme Conditions, School of Biological Sciences, University of Edinburgh, Edinburgh, EH9 3JR, United Kingdom,Department of Biology, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139
ABSTRACT
Sinorhizobium meliloti bacA mutants are symbiotically defective, deoxycholate sensitive, and bleomycin resistant. We show that the bleomycin resistance phenotype is independent of the lipid A alteration and that the changes giving rise to both phenotypes are likely to be involved in the inability of bacA mutants to persist within their hosts.
TEXT
The BacA protein is essential for Sinorhizobium meliloti, a legume symbiont, and Brucella abortus, a phylogenetically related mammalian pathogen, to establish chronic intracellular infections in their respective hosts (9, 13). Although the precise function of BacA is unknown, S. meliloti and B. abortus bacA-null mutants display a range of phenotypes during growth in complex medium, including low-level resistance to bleomycin, a glycopeptide antibiotic, and increased sensitivity to detergents compared with their respective parent strains (7, 10). The detergent sensitivity phenotype led to the discovery that BacA affects the unusual very-long-chain fatty acid (VLCFA) modifications, 27-OHC28:0, 27-OH(OMeC4:0)C28, and 29-OHC30:0, of the lipid A in both S. meliloti and B. abortus (5, 7). Thus, in the absence of BacA, only 50% of the lipid A molecules of S. meliloti and B. abortus become modified with a VLCFA, in contrast to their respective parent strains, whose lipid A molecules all contain a VLCFA modification (5). However, recent evidence suggests that the unusual lipid A modification observed during growth of wild-type S. meliloti in complex medium is important, but not essential, for the legume symbiosis (6). Thus, since additional VLCFA modifications of the lipid A occur during the symbiosis of Rhizobium leguminosarum with peas (11), and we observed a similar increase in S. meliloti lipopolysaccharide (LPS) hydrophobicity during the alfalfa symbiosis (6), we proposed a model whereby BacA could be involved in host-induced lipid A changes. These BacA-dependent lipid A changes could be essential for the chronic infection of their eukaryotic hosts by S. meliloti and B. abortus (6).
Nevertheless, there also remained a formal possibility that additional lipid A-independent changes were occurring in the S. meliloti bacA-null mutant and that one or more of these changes could also be involved in the inability of bacA mutants to persist within their hosts. For example, it seemed unlikely that a reduction in the lipid A VLCFA content could account for the low-level bleomycin resistance phenotype of the S. meliloti bacA-null mutant, since deletion of the bacA homolog, sbmA, in Escherichia coli also gives rise to a similar phenotype (10), despite the fact that the lipid A of E. coli lacks VLCFA modifications (15). Thus, these data suggested that BacA might have additional effects on S. meliloti, resulting in increased sensitization of wild-type S. meliloti toward bleomycin relative to the S. meliloti bacA mutant.
S. meliloti mutants completely lacking the lipid A VLCFA modification have increased sensitivity to bleomycin. Our recent discovery that the lipid A species produced by the S. meliloti acpXL and lpxXL insertional mutants in LB medium supplemented with 2.5 mM MgSO4 and 2.5 mM CaCl2 (LB/MC medium) completely lacked the lipid A VLCFA modifications (6) provided us with the means to investigate how the absence of the lipid A VLCFA modification in S. meliloti influences bleomycin sensitivity. The acpXL and lpxXL gene products encode a VLCFA-acyl carrier protein (3) and a VLCFA-acyltransferase protein (1), which are directly involved in the biosynthesis of VFCFA-modified lipid A in S. meliloti. However, in contrast to S. meliloti bacA mutants, which display low-level resistance to bleomycin, the S. meliloti acpXL and lpxXL insertional mutants displayed increased sensitivity to bleomycin on LB/MC agar relative to the parent strain, Rm1021 (Fig. 1A). Interestingly, consistent with previous stress sensitivity assays (6), the lpxXL mutant was more sensitized to bleomycin than the acpXL mutant. Since the acpXL mutant, but not the lpxXL mutant, produces a significant percentage of its total lipid A molecules in the pentaacylated state (6), despite lacking the VLCFA modification, these pentaacylated lipid A molecules must confer some protection against bleomycin. However, these data show that the complete absence of the lipid A VLCFA modification increases the sensitivity of S. meliloti to bleomycin and does not lead to increased bleomycin resistance. Thus, these findings provide further support for the notion that the low-level bleomycin resistance phenotype of the S. meliloti bacA mutant is unlikely to be due to a reduction in the lipid A VLCFA modification content.
Deletion of bacA increases the resistance of S. meliloti to bleomycin even in mutants lacking the lipid A VLCFA modification. To further confirm that the low-level bleomycin resistance phenotype of the S. meliloti bacA-null mutant could occur independently of the lipid A VLCFA modification, we compared the bleomycin sensitivities of the S. meliloti acpXL bacA and lpxXL bacA double mutants to those of the acpXL and lpxXL single mutants, respectively (Fig. 1B and C, respectively). We demonstrated previously that the S. meliloti acpXL bacA and lpxXL bacA double mutants had lipid A profiles identical to those of the acpXL and lpxXL single mutants (6), respectively, and that their lipid A molecules completely lacked the VLCFA modification. However, despite the lack of the lipid A VLCFA modification, the absence of BacA still conferred low-level resistance to bleomycin in the acpXL bacA and lpxXL bacA double-mutant backgrounds relative to the respective acpXL and lpxXL single mutants (Fig. 1B and C, respectively). Thus, these data show that deletion of bacA increases the resistance of S. meliloti to bleomycin, even in mutant strains that completely lack the lipid A VLCFA modification. Additionally, these findings argue that BacA must exert a lipid A-independent effect on the parent strain and that the absence of BacA gives rise to the low-level bleomycin resistance phenotype displayed by the S. meliloti bacA mutant.
In the related legume symbiont, R. leguminosarum, deletion of acpXL results in growth defects in complex medium relative to the parent strain (17). However, we observed that S. meliloti mutants with insertions in the lpxXL gene, but not the acpXL gene, had a reduced growth rate on LB/MC agar without bleomycin (data not shown). Since deletion of bacA did not further affect the growth rate of either the S. meliloti parent strain or the lpxXL and acpXL mutants on LB/MC agar without bleomycin (data not shown), these findings rule out the possibility that deletion of bacA confers protection of S. meliloti against bleomycin due to growth rate alterations.
Transposon insertions in the bacA gene alone lead to bleomycin resistance in S. meliloti. To gain further insights into resistance to bleomycin in S. meliloti and to determine whether bleomycin resistance per se was linked to the inability of S. meliloti to form a successful legume symbiosis, a transposon mutant library was constructed in the S. meliloti Rm1021 parent strain background using Tn5-233 delivered on the suicide vector pRK607 (4). Tn5-233 was used instead of Tn5 because the latter transposon contains the ble gene, which encodes high-level resistance to bleomycin (4). The bleomycin sensitivity of the parental strain Rm1021 was initially determined by plating approximately 5 x 106 stationary-phase cells of Rm1021 onto LB agar supplemented with a range of bleomycin concentrations (0 to 2.5 μg ml–1). We discovered that the growth of the parent strain was severely affected by the inclusion of 0.5 μg ml–1 bleomycin in the agar. Thus, 8 x 105 S. meliloti Tn5-233 mutants were subsequently plated onto LB agar containing 0.5 to 2.5 μg ml–1 bleomycin, and 9 putative bleomycin-resistant mutants were purified. All nine Tn5-233 mutants were shown to display a range of bleomycin resistance phenotypes compared to the parent strain, some having a level of resistance similar to that of a bacA-null mutant and some having higher levels of resistance (Fig. 2A). However, after transduction of the Tn5-233 insertions from the bleomycin-resistant mutants into Rm1021, all of the resulting transposon mutants conferred low-level resistance to bleomycin, similar to that observed for the bacA-null mutant (Fig. 2A). Thus, these findings suggested that some of the original transposon-induced bleomycin-resistant mutants contained additional unlinked mutations that contributed to their higher-level bleomycin resistance phenotype. Consistent with this hypothesis, we discovered that, like the bacA-null mutant, all nine of the transduced Tn5-233 mutants had increased sensitivity to deoxycholate (DOC) (Fig. 2B) and were defective in alfalfa symbiosis (data not shown). Subsequent analysis by PCR, using a Tn5-233-specific forward primer and a bacA-specific reverse primer, confirmed that that all nine of the transposon mutants contained insertions disrupting their bacA genes (data not shown). Since the S. meliloti genome is predicted to encode 6,000 genes (8) and since we plated 8 x 105 transposon mutants, our selection was performed under saturating conditions. Additionally, since the S. meliloti bacA gene is not arranged in an operon and since a plasmid carrying the S. meliloti bacA gene (9) complemented all the reported phenotypes of the bacA transposon insertion mutants (data not shown), this provided evidence that the phenotypes of the transposon mutants were due to disruption of the bacA gene and not to polar effects on downstream genes. Thus, under our experimental conditions, only disruption of the bacA gene resulted in bleomycin resistance in S. meliloti Rm1021. However, we cannot rule out the possibility that essential genes could also affect the resistance of S. meliloti to killing by bleomycin.
Bleomycin resistance per se is not necessarily linked with the inability of S. meliloti to establish a successful symbiosis. Since only bacA mutants were isolated from our transposon mutagenesis study, we were unable to address whether bleomycin resistance per se was linked to the inability of S. meliloti to form a successful symbiosis. However, during our earlier assessment of the sensitivity of the parent strain, Rm1021, to bleomycin, we obtained 38 spontaneous bleomycin mutants with various levels of resistance to bleomycin (Fig. 3A). Although we termed these spontaneous mutants, since bleomycin has been shown to be a DNA-damaging agent in E. coli (18), it is also possible that some of these 38 bleomycin resistance mutants arose due to bleomycin-induced DNA damage. Interestingly, the level of bleomycin resistance in these 38 mutants did not necessarily correlate with the concentration of bleomycin used for the initial selection conditions (Fig. 3A). Since these mutants displayed a range of bleomycin resistances, despite the selection not being performed in the optimal manner (i.e., independent cultures were not used), they allowed us to investigate whether there was any link between the level of bleomycin resistance per se and the ability of S. meliloti to persist within legumes.
Inoculation of the 38 mutants onto individual alfalfa seedlings on Jensen's agar, which lacks a nitrogen and carbon source, enabled us to assess their symbiotic competency (12). Four weeks postinoculation, we discovered that 10/38 mutants were unable to establish a successful symbiosis with alfalfa (data not shown). Thus, in contrast to alfalfa seedlings inoculated with symbiotically competent S. meliloti, which were dark green and had elongated pink nodules, indicative of a successful nitrogen-fixing symbiosis, the alfalfa seedlings inoculated with the 10 mutants that were defective in establishing a symbiosis were stunted in height and yellowish green, with white nodules. Intriguing, all 10 of the bleomycin-resistant mutants that were unable to form a successful symbiosis with alfalfa (termed class I mutants) displayed low-level resistance to bleomycin (Fig. 3B), similar to the bacA-null and transposon mutants (Fig. 2A). However, since several of the mutants (termed class II mutants) also displayed low-level resistance to bleomycin (Fig. 3B) yet were fully competent in alfalfa symbiosis (data not shown), these data showed that the low-level bleomycin resistance phenotype per se was not necessarily linked to the inability of S. meliloti to form a successful legume symbiosis. Additionally, since mutants displaying a higher level of resistance to bleomycin (termed class III mutants) (Fig. 3B) were symbiotically proficient (data not shown), these findings provided further evidence that resistance to bleomycin per se is not necessarily linked to the inability of S. meliloti to persist within legumes.
Class I mutants all have mutations in their bacA genes. The fact that mutations in bacA are the only mutations known to date to give rise to low-level bleomycin resistance and symbiotic defects in S. meliloti led us to investigate whether our class I mutants contained mutations in their bacA genes. To investigate this further, the bacA genes and promoter regions from our class I mutants were sequenced, and we determined that all class I mutants had mutations in their bacA genes, which would result in the production of truncated BacA proteins (Table 1). Intriguingly, a number of the class I mutants had either an addition or a deletion at position 92 in their bacA gene (Table 1). Since the bacA gene sequence in this region contains a stretch of guanine residues, which are known to cause problems with DNA replication, leading to an increased frequency of frameshift mutations, this is likely to account for our observations. Additionally, we also observed that a number of the class I mutants had two mutations in their bacA gene (Table 1). Thus, it may be that frequently occurring mutations at hot spots in bacA lead to a growth disadvantage that results in selection for the acquisition of a second mutation in the bacA gene. However, these findings provide evidence that class I mutants produce mutant forms of the BacA protein. Consistent with this, all class I mutants displayed increased sensitivity to DOC on LB agar (Fig. 3C), and all of their phenotypes could be complemented by transformation with a plasmid harboring the intact S. meliloti bacA gene (pJG51A) but not with the control plasmid alone (pRK404) (9) (data not shown).
In contrast to the class I mutants, only one of the class II mutants was mutated in the bacA coding region and promoter (Table 1). The fact that this class II mutant differed from the class I mutants in being symbiotically proficient suggests that it must produce a functional BacA protein, despite having a mutation in its bacA promoter and a frame shift near the 3' end of the bacA coding region. The rest of the class II mutants did not have mutations in their bacA coding region or promoter (Table 1). Intriguingly, all the class II mutants were as resistant to DOC as the parent strain, including the class II mutant, which contained mutations in the bacA gene (data not shown). Thus, these findings suggest that the changes in the bacA mutants giving rise to increased sensitivity to DOC and increased resistance to bleomycin are both involved in the inability of bacA mutants to form a successful symbiosis.
Conclusions. In this study, we show that the bleomycin resistance phenotype of the S. meliloti bacA mutant is independent of the lipid A alteration in this mutant. We also show that bleomycin resistance per se is not necessarily linked to the inability of S. meliloti to establish a successful symbiosis. Instead, this study provides evidence that the specific changes in bacA mutants resulting in the DOC sensitivity and bleomycin resistance phenotypes appear to be involved in the inability of bacA mutants to form an effective legume symbiosis. However, the mechanism by which loss of BacA gives rise to bleomycin resistance is still unresolved.
Interestingly, an E. coli mutant lacking the BacA homolog, SbmA, is also resistant to bleomycin and microcin antibiotics (10, 16). Since the E. coli sbmA mutant is as sensitive to internally synthesized microcins as the parent strain, this finding led to a model whereby SbmA is proposed to directly transport unusual peptides, such as bleomycin and microcin antibiotics, into the E. coli cell (16). Thus, it is possible that S. meliloti BacA may also be involved in bleomycin transport. However, there is no direct evidence demonstrating transport of bleomycin by E. coli SbmA, and it is difficult to rationalize why loss of a transport system would result only in low-level resistance to bleomycin. Additionally, the fact that bleomycin can still inhibit the growth of the S. meliloti bacA mutants suggests that there would have to be another mechanism for bleomycin uptake and/or that bleomycin can cause additional damage to an extracytoplasmic component of S. meliloti. However, when we selected for bleomycin-resistant mutants after transposon mutagenesis, we obtained only bacA mutants, suggesting either that an additional uptake system does not exist or that BacA is essential for the growth of S. meliloti on LB agar. Bleomycin has been shown to induce DNA damage in E. coli, and RecA was found to be involved in repair of this damage (18). We also found that an S. meliloti recA::Tn5-233 mutant has increased sensitivity to bleomycin compared to the parent strain (V. L. Marlow, C. Rougier, G. C. Walker, and G. P. Ferguson, unpublished data), suggesting that bleomycin can enter into S. meliloti cells and cause DNA damage. However, bleomycin has been shown to damage the cell wall of Saccharomyces cerevisiae (2), leading to increased spheroplast production, and thus, bleomycin may also be causing cell wall damage, in addition to DNA damage, in S. meliloti. Thus, future studies will be required to determine the precise effects of bleomycin on S. meliloti, the molecular basis of BacA-dependent bleomycin sensitivity, and the exact role of BacA in persistent bacteriuml-host interactions. However, these studies provide further evidence that BacA is also capable of exerting lipid A-independent effects in S. meliloti.
ACKNOWLEDGMENTS
G.P.F. and A.J. were funded by National Institutes of Health grant GM31030, awarded to G.C.W. G.P.F. is currently supported by a University of Edinburgh Development Trust grant (D54305) and a BBSRC grant (BB/D000564/1). V.L.M. is funded by a BBSRC Ph.D. studentship. G.C.W. is also supported by an American Cancer Society Research Professorship and a Howard Hughes Medical Institute Professorship.
We thank Marty Roop II, Carole Rougier, and the Walker lab for many helpful discussions.
REFERENCES
Basu, S. S., M. J. Karbarz, and C. R. Raetz. 2002. Expression cloning and characterization of the C28 acyltransferase of lipid A biosynthesis in Rhizobium leguminosarum. J. Biol. Chem. 277:28959-28971.
Beaudouin, R., S. T. Lim, J. A. Steide, M. Powell, J. McKoy, A. J. Pramanik, E. Johnson, C. W. Moore, and P. N. Lipke. 1993. Bleomycin affects cell wall anchorage of mannoproteins in Saccharomyces cerevisiae. Antimicrob. Agents Chemother. 37:1264-1269.
Brozek, K. A., R. W. Carlson, and C. R. Raetz. 1996. A special acyl carrier protein for transferring long hydroxylated fatty acids to lipid A in Rhizobium. J. Biol. Chem. 271:32126-32136.
De Vos, G. F., G. C. Walker, and E. R. Signer. 1986. Genetic manipulations in Rhizobium meliloti using two new transposon Tn5 derivatives. Mol. Gen. Genet. 204:485-491.
Ferguson, G. P., A. Datta, J. Baumgartner, R. M. Roop II, R. W. Carlson, and G. C. Walker. 2004. Similarity to peroxisomal-membrane protein family reveals that Sinorhizobium and Brucella BacA affect lipid-A fatty acids. Proc. Natl. Acad. Sci. USA 101:5012-5017.
Ferguson, G. P., A. Datta, R. W. Carlson, and G. C. Walker. 2005. Importance of unusually modified lipid A in Sinorhizobium stress resistance and legume symbiosis. Mol. Microbiol. 56:68-80.
Ferguson, G. P., R. M. Roop II, and G. C. Walker. 2002. Deficiency of Sinorhizobium meliloti bacA mutant in alfalfa symbiosis correlates with an alteration of cell envelope. J. Bacteriol. 184:5625-5632.
Galibert, F., T. M. Finan, S. R. Long, A. Puhler, P. Abola, F. Ampe, F. Barloy-Hubler, M. J. Barnett, A. Becker, P. Boistard, G. Bothe, M. Boutry, L. Bowser, J. Buhrmester, E. Cadieu, D. Capela, P. Chain, A. Cowie, R. W. Davis, S. Dreano, N. A. Federspiel, R. F. Fisher, S. Gloux, T. Godrie, A. Goffeau, B. Golding, J. Gouzy, M. Gurjal, I. Hernandez-Lucas, A. Hong, L. Huizar, R. W. Hyman, T. Jones, D. Kahn, M. L. Kahn, S. Kalman, D. H. Keating, E. Kiss, C. Komp, V. Lelaure, D. Masuy, C. Palm, M. C. Peck, T. M. Pohl, D. Portetelle, B. Purnelle, U. Ramsperger, R. Surzycki, P. Thebault, M. Vandenbol, F. J. Vorhlter, S. Weidner, D. H. Wells, K. Wong, K. C. Yeh, and J. Batut. 2001. The composite genome of the legume symbiont Sinorhizobium meliloti. Science 293:668-672.
Glazebrook, J., A. Ichige, and G. C. Walker. 1993. A Rhizobium melioti homolog of the Escherichia coli peptide-antibiotic transport protein SbmA is essential for bacteroid development. Genes Dev. 7:1485-1497.
Ichige, A., and G. C. Walker. 1997. Genetic analysis of the Rhizobium meliloti bacA gene: functional interchangeability with the Escherichia coli sbmA gene and phenotype of mutants. J. Bacteriol. 179:209-216.
Kannenberg, E. L., and R. W. Carlson. 2001. Lipid A and O-chain modifications cause Rhizobium lipopolysaccharides to become hydrophobic during bacteroid development. Mol. Microbiol. 39:379-391.
Leigh, J. A., E. R. Signer, and G. C. Walker. 1985. Exopolysaccharide-deficient mutants of Rhizobium meliloti that form ineffective nodules. Proc. Natl. Acad. Sci. USA 82:6231-6235.
LeVier, K., R. W. Phillips, V. K. Grippe, R. M. Roop II, and G. C. Walker. 2000. Similar requirements of a plant symbiont and a mammalian pathogen for prolonged intracellular survival. Science 287:2492-2493.
LeVier, K., and G. C. Walker. 2001. Genetic analysis of the Sinorhizobium meliloti BacA protein: differential effects of mutations on phenotypes. J. Bacteriol. 183:6444-6453.
Raetz, C. R., and C. Whitfield. 2002. Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71:635-700.
Salomon, R. A., and R. N. Farías. 1995. The peptide antibiotic microcin 25 is imported through the TonB pathway and the SbmA protein. J. Bacteriol. 177:3323-3325.
Vedam, V., J. G. Haynes, E. L. Kannenberg, R. W. Carlson, and D. J. Sherrier. 2004. A Rhizobium leguminosarum lipopolysaccharide lipid-A mutant induces nitrogen-fixing nodules with delayed and defective bacteroid formation. Mol. Plant-Microbe Interact. 17:283-291.
Yamamoto, K., and F. Hutchinson. 1984. The effect of bleomycin on DNA in Escherichia coli K12 cells. Chem. Biol. Interact. 51:233-246.(Gail P. Ferguson, An Jans)
ABSTRACT
Sinorhizobium meliloti bacA mutants are symbiotically defective, deoxycholate sensitive, and bleomycin resistant. We show that the bleomycin resistance phenotype is independent of the lipid A alteration and that the changes giving rise to both phenotypes are likely to be involved in the inability of bacA mutants to persist within their hosts.
TEXT
The BacA protein is essential for Sinorhizobium meliloti, a legume symbiont, and Brucella abortus, a phylogenetically related mammalian pathogen, to establish chronic intracellular infections in their respective hosts (9, 13). Although the precise function of BacA is unknown, S. meliloti and B. abortus bacA-null mutants display a range of phenotypes during growth in complex medium, including low-level resistance to bleomycin, a glycopeptide antibiotic, and increased sensitivity to detergents compared with their respective parent strains (7, 10). The detergent sensitivity phenotype led to the discovery that BacA affects the unusual very-long-chain fatty acid (VLCFA) modifications, 27-OHC28:0, 27-OH(OMeC4:0)C28, and 29-OHC30:0, of the lipid A in both S. meliloti and B. abortus (5, 7). Thus, in the absence of BacA, only 50% of the lipid A molecules of S. meliloti and B. abortus become modified with a VLCFA, in contrast to their respective parent strains, whose lipid A molecules all contain a VLCFA modification (5). However, recent evidence suggests that the unusual lipid A modification observed during growth of wild-type S. meliloti in complex medium is important, but not essential, for the legume symbiosis (6). Thus, since additional VLCFA modifications of the lipid A occur during the symbiosis of Rhizobium leguminosarum with peas (11), and we observed a similar increase in S. meliloti lipopolysaccharide (LPS) hydrophobicity during the alfalfa symbiosis (6), we proposed a model whereby BacA could be involved in host-induced lipid A changes. These BacA-dependent lipid A changes could be essential for the chronic infection of their eukaryotic hosts by S. meliloti and B. abortus (6).
Nevertheless, there also remained a formal possibility that additional lipid A-independent changes were occurring in the S. meliloti bacA-null mutant and that one or more of these changes could also be involved in the inability of bacA mutants to persist within their hosts. For example, it seemed unlikely that a reduction in the lipid A VLCFA content could account for the low-level bleomycin resistance phenotype of the S. meliloti bacA-null mutant, since deletion of the bacA homolog, sbmA, in Escherichia coli also gives rise to a similar phenotype (10), despite the fact that the lipid A of E. coli lacks VLCFA modifications (15). Thus, these data suggested that BacA might have additional effects on S. meliloti, resulting in increased sensitization of wild-type S. meliloti toward bleomycin relative to the S. meliloti bacA mutant.
S. meliloti mutants completely lacking the lipid A VLCFA modification have increased sensitivity to bleomycin. Our recent discovery that the lipid A species produced by the S. meliloti acpXL and lpxXL insertional mutants in LB medium supplemented with 2.5 mM MgSO4 and 2.5 mM CaCl2 (LB/MC medium) completely lacked the lipid A VLCFA modifications (6) provided us with the means to investigate how the absence of the lipid A VLCFA modification in S. meliloti influences bleomycin sensitivity. The acpXL and lpxXL gene products encode a VLCFA-acyl carrier protein (3) and a VLCFA-acyltransferase protein (1), which are directly involved in the biosynthesis of VFCFA-modified lipid A in S. meliloti. However, in contrast to S. meliloti bacA mutants, which display low-level resistance to bleomycin, the S. meliloti acpXL and lpxXL insertional mutants displayed increased sensitivity to bleomycin on LB/MC agar relative to the parent strain, Rm1021 (Fig. 1A). Interestingly, consistent with previous stress sensitivity assays (6), the lpxXL mutant was more sensitized to bleomycin than the acpXL mutant. Since the acpXL mutant, but not the lpxXL mutant, produces a significant percentage of its total lipid A molecules in the pentaacylated state (6), despite lacking the VLCFA modification, these pentaacylated lipid A molecules must confer some protection against bleomycin. However, these data show that the complete absence of the lipid A VLCFA modification increases the sensitivity of S. meliloti to bleomycin and does not lead to increased bleomycin resistance. Thus, these findings provide further support for the notion that the low-level bleomycin resistance phenotype of the S. meliloti bacA mutant is unlikely to be due to a reduction in the lipid A VLCFA modification content.
Deletion of bacA increases the resistance of S. meliloti to bleomycin even in mutants lacking the lipid A VLCFA modification. To further confirm that the low-level bleomycin resistance phenotype of the S. meliloti bacA-null mutant could occur independently of the lipid A VLCFA modification, we compared the bleomycin sensitivities of the S. meliloti acpXL bacA and lpxXL bacA double mutants to those of the acpXL and lpxXL single mutants, respectively (Fig. 1B and C, respectively). We demonstrated previously that the S. meliloti acpXL bacA and lpxXL bacA double mutants had lipid A profiles identical to those of the acpXL and lpxXL single mutants (6), respectively, and that their lipid A molecules completely lacked the VLCFA modification. However, despite the lack of the lipid A VLCFA modification, the absence of BacA still conferred low-level resistance to bleomycin in the acpXL bacA and lpxXL bacA double-mutant backgrounds relative to the respective acpXL and lpxXL single mutants (Fig. 1B and C, respectively). Thus, these data show that deletion of bacA increases the resistance of S. meliloti to bleomycin, even in mutant strains that completely lack the lipid A VLCFA modification. Additionally, these findings argue that BacA must exert a lipid A-independent effect on the parent strain and that the absence of BacA gives rise to the low-level bleomycin resistance phenotype displayed by the S. meliloti bacA mutant.
In the related legume symbiont, R. leguminosarum, deletion of acpXL results in growth defects in complex medium relative to the parent strain (17). However, we observed that S. meliloti mutants with insertions in the lpxXL gene, but not the acpXL gene, had a reduced growth rate on LB/MC agar without bleomycin (data not shown). Since deletion of bacA did not further affect the growth rate of either the S. meliloti parent strain or the lpxXL and acpXL mutants on LB/MC agar without bleomycin (data not shown), these findings rule out the possibility that deletion of bacA confers protection of S. meliloti against bleomycin due to growth rate alterations.
Transposon insertions in the bacA gene alone lead to bleomycin resistance in S. meliloti. To gain further insights into resistance to bleomycin in S. meliloti and to determine whether bleomycin resistance per se was linked to the inability of S. meliloti to form a successful legume symbiosis, a transposon mutant library was constructed in the S. meliloti Rm1021 parent strain background using Tn5-233 delivered on the suicide vector pRK607 (4). Tn5-233 was used instead of Tn5 because the latter transposon contains the ble gene, which encodes high-level resistance to bleomycin (4). The bleomycin sensitivity of the parental strain Rm1021 was initially determined by plating approximately 5 x 106 stationary-phase cells of Rm1021 onto LB agar supplemented with a range of bleomycin concentrations (0 to 2.5 μg ml–1). We discovered that the growth of the parent strain was severely affected by the inclusion of 0.5 μg ml–1 bleomycin in the agar. Thus, 8 x 105 S. meliloti Tn5-233 mutants were subsequently plated onto LB agar containing 0.5 to 2.5 μg ml–1 bleomycin, and 9 putative bleomycin-resistant mutants were purified. All nine Tn5-233 mutants were shown to display a range of bleomycin resistance phenotypes compared to the parent strain, some having a level of resistance similar to that of a bacA-null mutant and some having higher levels of resistance (Fig. 2A). However, after transduction of the Tn5-233 insertions from the bleomycin-resistant mutants into Rm1021, all of the resulting transposon mutants conferred low-level resistance to bleomycin, similar to that observed for the bacA-null mutant (Fig. 2A). Thus, these findings suggested that some of the original transposon-induced bleomycin-resistant mutants contained additional unlinked mutations that contributed to their higher-level bleomycin resistance phenotype. Consistent with this hypothesis, we discovered that, like the bacA-null mutant, all nine of the transduced Tn5-233 mutants had increased sensitivity to deoxycholate (DOC) (Fig. 2B) and were defective in alfalfa symbiosis (data not shown). Subsequent analysis by PCR, using a Tn5-233-specific forward primer and a bacA-specific reverse primer, confirmed that that all nine of the transposon mutants contained insertions disrupting their bacA genes (data not shown). Since the S. meliloti genome is predicted to encode 6,000 genes (8) and since we plated 8 x 105 transposon mutants, our selection was performed under saturating conditions. Additionally, since the S. meliloti bacA gene is not arranged in an operon and since a plasmid carrying the S. meliloti bacA gene (9) complemented all the reported phenotypes of the bacA transposon insertion mutants (data not shown), this provided evidence that the phenotypes of the transposon mutants were due to disruption of the bacA gene and not to polar effects on downstream genes. Thus, under our experimental conditions, only disruption of the bacA gene resulted in bleomycin resistance in S. meliloti Rm1021. However, we cannot rule out the possibility that essential genes could also affect the resistance of S. meliloti to killing by bleomycin.
Bleomycin resistance per se is not necessarily linked with the inability of S. meliloti to establish a successful symbiosis. Since only bacA mutants were isolated from our transposon mutagenesis study, we were unable to address whether bleomycin resistance per se was linked to the inability of S. meliloti to form a successful symbiosis. However, during our earlier assessment of the sensitivity of the parent strain, Rm1021, to bleomycin, we obtained 38 spontaneous bleomycin mutants with various levels of resistance to bleomycin (Fig. 3A). Although we termed these spontaneous mutants, since bleomycin has been shown to be a DNA-damaging agent in E. coli (18), it is also possible that some of these 38 bleomycin resistance mutants arose due to bleomycin-induced DNA damage. Interestingly, the level of bleomycin resistance in these 38 mutants did not necessarily correlate with the concentration of bleomycin used for the initial selection conditions (Fig. 3A). Since these mutants displayed a range of bleomycin resistances, despite the selection not being performed in the optimal manner (i.e., independent cultures were not used), they allowed us to investigate whether there was any link between the level of bleomycin resistance per se and the ability of S. meliloti to persist within legumes.
Inoculation of the 38 mutants onto individual alfalfa seedlings on Jensen's agar, which lacks a nitrogen and carbon source, enabled us to assess their symbiotic competency (12). Four weeks postinoculation, we discovered that 10/38 mutants were unable to establish a successful symbiosis with alfalfa (data not shown). Thus, in contrast to alfalfa seedlings inoculated with symbiotically competent S. meliloti, which were dark green and had elongated pink nodules, indicative of a successful nitrogen-fixing symbiosis, the alfalfa seedlings inoculated with the 10 mutants that were defective in establishing a symbiosis were stunted in height and yellowish green, with white nodules. Intriguing, all 10 of the bleomycin-resistant mutants that were unable to form a successful symbiosis with alfalfa (termed class I mutants) displayed low-level resistance to bleomycin (Fig. 3B), similar to the bacA-null and transposon mutants (Fig. 2A). However, since several of the mutants (termed class II mutants) also displayed low-level resistance to bleomycin (Fig. 3B) yet were fully competent in alfalfa symbiosis (data not shown), these data showed that the low-level bleomycin resistance phenotype per se was not necessarily linked to the inability of S. meliloti to form a successful legume symbiosis. Additionally, since mutants displaying a higher level of resistance to bleomycin (termed class III mutants) (Fig. 3B) were symbiotically proficient (data not shown), these findings provided further evidence that resistance to bleomycin per se is not necessarily linked to the inability of S. meliloti to persist within legumes.
Class I mutants all have mutations in their bacA genes. The fact that mutations in bacA are the only mutations known to date to give rise to low-level bleomycin resistance and symbiotic defects in S. meliloti led us to investigate whether our class I mutants contained mutations in their bacA genes. To investigate this further, the bacA genes and promoter regions from our class I mutants were sequenced, and we determined that all class I mutants had mutations in their bacA genes, which would result in the production of truncated BacA proteins (Table 1). Intriguingly, a number of the class I mutants had either an addition or a deletion at position 92 in their bacA gene (Table 1). Since the bacA gene sequence in this region contains a stretch of guanine residues, which are known to cause problems with DNA replication, leading to an increased frequency of frameshift mutations, this is likely to account for our observations. Additionally, we also observed that a number of the class I mutants had two mutations in their bacA gene (Table 1). Thus, it may be that frequently occurring mutations at hot spots in bacA lead to a growth disadvantage that results in selection for the acquisition of a second mutation in the bacA gene. However, these findings provide evidence that class I mutants produce mutant forms of the BacA protein. Consistent with this, all class I mutants displayed increased sensitivity to DOC on LB agar (Fig. 3C), and all of their phenotypes could be complemented by transformation with a plasmid harboring the intact S. meliloti bacA gene (pJG51A) but not with the control plasmid alone (pRK404) (9) (data not shown).
In contrast to the class I mutants, only one of the class II mutants was mutated in the bacA coding region and promoter (Table 1). The fact that this class II mutant differed from the class I mutants in being symbiotically proficient suggests that it must produce a functional BacA protein, despite having a mutation in its bacA promoter and a frame shift near the 3' end of the bacA coding region. The rest of the class II mutants did not have mutations in their bacA coding region or promoter (Table 1). Intriguingly, all the class II mutants were as resistant to DOC as the parent strain, including the class II mutant, which contained mutations in the bacA gene (data not shown). Thus, these findings suggest that the changes in the bacA mutants giving rise to increased sensitivity to DOC and increased resistance to bleomycin are both involved in the inability of bacA mutants to form a successful symbiosis.
Conclusions. In this study, we show that the bleomycin resistance phenotype of the S. meliloti bacA mutant is independent of the lipid A alteration in this mutant. We also show that bleomycin resistance per se is not necessarily linked to the inability of S. meliloti to establish a successful symbiosis. Instead, this study provides evidence that the specific changes in bacA mutants resulting in the DOC sensitivity and bleomycin resistance phenotypes appear to be involved in the inability of bacA mutants to form an effective legume symbiosis. However, the mechanism by which loss of BacA gives rise to bleomycin resistance is still unresolved.
Interestingly, an E. coli mutant lacking the BacA homolog, SbmA, is also resistant to bleomycin and microcin antibiotics (10, 16). Since the E. coli sbmA mutant is as sensitive to internally synthesized microcins as the parent strain, this finding led to a model whereby SbmA is proposed to directly transport unusual peptides, such as bleomycin and microcin antibiotics, into the E. coli cell (16). Thus, it is possible that S. meliloti BacA may also be involved in bleomycin transport. However, there is no direct evidence demonstrating transport of bleomycin by E. coli SbmA, and it is difficult to rationalize why loss of a transport system would result only in low-level resistance to bleomycin. Additionally, the fact that bleomycin can still inhibit the growth of the S. meliloti bacA mutants suggests that there would have to be another mechanism for bleomycin uptake and/or that bleomycin can cause additional damage to an extracytoplasmic component of S. meliloti. However, when we selected for bleomycin-resistant mutants after transposon mutagenesis, we obtained only bacA mutants, suggesting either that an additional uptake system does not exist or that BacA is essential for the growth of S. meliloti on LB agar. Bleomycin has been shown to induce DNA damage in E. coli, and RecA was found to be involved in repair of this damage (18). We also found that an S. meliloti recA::Tn5-233 mutant has increased sensitivity to bleomycin compared to the parent strain (V. L. Marlow, C. Rougier, G. C. Walker, and G. P. Ferguson, unpublished data), suggesting that bleomycin can enter into S. meliloti cells and cause DNA damage. However, bleomycin has been shown to damage the cell wall of Saccharomyces cerevisiae (2), leading to increased spheroplast production, and thus, bleomycin may also be causing cell wall damage, in addition to DNA damage, in S. meliloti. Thus, future studies will be required to determine the precise effects of bleomycin on S. meliloti, the molecular basis of BacA-dependent bleomycin sensitivity, and the exact role of BacA in persistent bacteriuml-host interactions. However, these studies provide further evidence that BacA is also capable of exerting lipid A-independent effects in S. meliloti.
ACKNOWLEDGMENTS
G.P.F. and A.J. were funded by National Institutes of Health grant GM31030, awarded to G.C.W. G.P.F. is currently supported by a University of Edinburgh Development Trust grant (D54305) and a BBSRC grant (BB/D000564/1). V.L.M. is funded by a BBSRC Ph.D. studentship. G.C.W. is also supported by an American Cancer Society Research Professorship and a Howard Hughes Medical Institute Professorship.
We thank Marty Roop II, Carole Rougier, and the Walker lab for many helpful discussions.
REFERENCES
Basu, S. S., M. J. Karbarz, and C. R. Raetz. 2002. Expression cloning and characterization of the C28 acyltransferase of lipid A biosynthesis in Rhizobium leguminosarum. J. Biol. Chem. 277:28959-28971.
Beaudouin, R., S. T. Lim, J. A. Steide, M. Powell, J. McKoy, A. J. Pramanik, E. Johnson, C. W. Moore, and P. N. Lipke. 1993. Bleomycin affects cell wall anchorage of mannoproteins in Saccharomyces cerevisiae. Antimicrob. Agents Chemother. 37:1264-1269.
Brozek, K. A., R. W. Carlson, and C. R. Raetz. 1996. A special acyl carrier protein for transferring long hydroxylated fatty acids to lipid A in Rhizobium. J. Biol. Chem. 271:32126-32136.
De Vos, G. F., G. C. Walker, and E. R. Signer. 1986. Genetic manipulations in Rhizobium meliloti using two new transposon Tn5 derivatives. Mol. Gen. Genet. 204:485-491.
Ferguson, G. P., A. Datta, J. Baumgartner, R. M. Roop II, R. W. Carlson, and G. C. Walker. 2004. Similarity to peroxisomal-membrane protein family reveals that Sinorhizobium and Brucella BacA affect lipid-A fatty acids. Proc. Natl. Acad. Sci. USA 101:5012-5017.
Ferguson, G. P., A. Datta, R. W. Carlson, and G. C. Walker. 2005. Importance of unusually modified lipid A in Sinorhizobium stress resistance and legume symbiosis. Mol. Microbiol. 56:68-80.
Ferguson, G. P., R. M. Roop II, and G. C. Walker. 2002. Deficiency of Sinorhizobium meliloti bacA mutant in alfalfa symbiosis correlates with an alteration of cell envelope. J. Bacteriol. 184:5625-5632.
Galibert, F., T. M. Finan, S. R. Long, A. Puhler, P. Abola, F. Ampe, F. Barloy-Hubler, M. J. Barnett, A. Becker, P. Boistard, G. Bothe, M. Boutry, L. Bowser, J. Buhrmester, E. Cadieu, D. Capela, P. Chain, A. Cowie, R. W. Davis, S. Dreano, N. A. Federspiel, R. F. Fisher, S. Gloux, T. Godrie, A. Goffeau, B. Golding, J. Gouzy, M. Gurjal, I. Hernandez-Lucas, A. Hong, L. Huizar, R. W. Hyman, T. Jones, D. Kahn, M. L. Kahn, S. Kalman, D. H. Keating, E. Kiss, C. Komp, V. Lelaure, D. Masuy, C. Palm, M. C. Peck, T. M. Pohl, D. Portetelle, B. Purnelle, U. Ramsperger, R. Surzycki, P. Thebault, M. Vandenbol, F. J. Vorhlter, S. Weidner, D. H. Wells, K. Wong, K. C. Yeh, and J. Batut. 2001. The composite genome of the legume symbiont Sinorhizobium meliloti. Science 293:668-672.
Glazebrook, J., A. Ichige, and G. C. Walker. 1993. A Rhizobium melioti homolog of the Escherichia coli peptide-antibiotic transport protein SbmA is essential for bacteroid development. Genes Dev. 7:1485-1497.
Ichige, A., and G. C. Walker. 1997. Genetic analysis of the Rhizobium meliloti bacA gene: functional interchangeability with the Escherichia coli sbmA gene and phenotype of mutants. J. Bacteriol. 179:209-216.
Kannenberg, E. L., and R. W. Carlson. 2001. Lipid A and O-chain modifications cause Rhizobium lipopolysaccharides to become hydrophobic during bacteroid development. Mol. Microbiol. 39:379-391.
Leigh, J. A., E. R. Signer, and G. C. Walker. 1985. Exopolysaccharide-deficient mutants of Rhizobium meliloti that form ineffective nodules. Proc. Natl. Acad. Sci. USA 82:6231-6235.
LeVier, K., R. W. Phillips, V. K. Grippe, R. M. Roop II, and G. C. Walker. 2000. Similar requirements of a plant symbiont and a mammalian pathogen for prolonged intracellular survival. Science 287:2492-2493.
LeVier, K., and G. C. Walker. 2001. Genetic analysis of the Sinorhizobium meliloti BacA protein: differential effects of mutations on phenotypes. J. Bacteriol. 183:6444-6453.
Raetz, C. R., and C. Whitfield. 2002. Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71:635-700.
Salomon, R. A., and R. N. Farías. 1995. The peptide antibiotic microcin 25 is imported through the TonB pathway and the SbmA protein. J. Bacteriol. 177:3323-3325.
Vedam, V., J. G. Haynes, E. L. Kannenberg, R. W. Carlson, and D. J. Sherrier. 2004. A Rhizobium leguminosarum lipopolysaccharide lipid-A mutant induces nitrogen-fixing nodules with delayed and defective bacteroid formation. Mol. Plant-Microbe Interact. 17:283-291.
Yamamoto, K., and F. Hutchinson. 1984. The effect of bleomycin on DNA in Escherichia coli K12 cells. Chem. Biol. Interact. 51:233-246.(Gail P. Ferguson, An Jans)