Biotin Limitation in Sinorhizobium meliloti Strain 1021 Alters ...

2 downloads 0 Views 518KB Size Report
Jul 8, 2002 - Hofmann, K., E. B. Heinz, T. C. Charles, M. Hoppert, W. Liebl, and W. R.. Streit. ... Meade, H. M., S. R. Long, G. B. Ruvkun, S. E. Brown, and F. M. ...
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 2003, p. 1206–1213 0099-2240/03/$08.00⫹0 DOI: 10.1128/AEM.69.2.1206–1213.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 69, No. 2

Biotin Limitation in Sinorhizobium meliloti Strain 1021 Alters Transcription and Translation Elke B. Heinz and Wolfgang R. Streit* Institut fu ¨r Mikrobiologie und Genetik der Universita ¨t Go ¨ttingen, D-37077 Go ¨ttingen, Germany Received 8 July 2002/Accepted 18 October 2002

Most Sinorhizobium meliloti strains lack several key genes involved in microbial biotin biosynthesis, and it is assumed that this may be a special adaptation which allows the microbe to down-regulate metabolic activities in the absence of a host plant. To further explore this hypothesis, we employed two different strategies. (i) Searches of the S. meliloti genome database in combination with the construction of nine different gusA reporter fusions identified three genes involved in a biotin starvation response in this microbe. A gene coding for a protein-methyl carboxyl transferase (pcm) exhibited 13.6-fold-higher transcription under biotin-limiting conditions than cells grown in the presence of 40 nM biotin. Consistent with this observation, biotin-limiting conditions resulted in a significantly decreased survival of pcm mutant cells compared to parental cells or cells grown in the presence of 40 nM biotin. Further studies indicated that the autoinducer synthase gene, sinI, was transcribed at a 4.5-fold-higher level in early stationary phase in biotin-starved cells than in biotin-supplemented cells. Lastly, we observed that open reading frame smc02283, which codes for a putative copper resistance protein (CopC), was 21-fold down-regulated in response to biotin starvation. (ii) In a second approach, proteome analysis identified 10 proteins which were significantly down-regulated under the biotinlimiting conditions. Among the proteins identified by using matrix-assisted laser desorption ionization–time of flight mass spectrometry were the ␼ subunit of the RNA polymerase and the 50S ribosomal protein L7/L12 (L8) subunit, indicating that biotin-limiting conditions generally affect transcription and translation in S. meliloti.

Microorganisms from the gram-negative genera Rhizobium, Sinorhizobium, Bradyrhizobium, Mesorhizobium, and Azorhizobium, collectively termed rhizobia, are well known for their capacity to establish N2-fixing symbioses with legume plants (2). Although the molecular basis of rhizobial N2 reduction is defined (12) and a foundation has been constructed for understanding other bacterial genes expressed in the plant (22), our knowledge of how rhizobia grow on the surface of plant roots or survive in the absence of a plant is less well characterized. Soil is in general considered to be a harsh and oligotrophic environment with a limited availability of carbon, energy sources, and other nutrients. As a result, bacterial growth in soil is in general very slow, with frequent periods of no growth, and only those bacteria which are well adapted will persist (8). Interestingly, in Sinorhizobium meliloti metabolic activities depend strongly on the availability of external biotin, and this may be a special adaptation for the organism to better survive periods of starvation outside the root nodule. Whether biotin is essential or simply stimulatory for rhizobial growth has long been debated, but clearly, cell densities of strain Rm1021 and many other rhizobia under biotin-limiting conditions are increased greatly by small amounts of biotin (6, 27, 31), and biotin biosynthesis appears to be limited because several key genes are nonfunctional or absent (7). Growing S. meliloti serially under biotin-limiting conditions produces several physiological and metabolic changes, including the accumulation of

poly-3-hydroxybutyrate and a significant reduction in cell size (10). In addition, biotin-dependent enzymes such as pyruvate carboxylase are affected under biotin-limiting conditions, and several tricarboxylic acid cycle auxiliary enzymes show decreased activities (3–5). Also, the regulatory gene bioS helps S. meliloti to compete under such biotin-limiting conditions (9, 28). The present study, therefore, was initiated to identify additional genes and proteins, which might be involved in a biotin starvation response in S. meliloti. Our results indicate that pcm gene transcription in S. meliloti depends on biotin availability in defined media. Consistent with this finding, we also show that a pcm mutation in S. meliloti results in significantly decreased survival in stationary-phase cells but, surprisingly, only under biotin-limiting conditions. Further tests indicated that biotin starvation in S. meliloti probably induces a general stress response which results in an increased transcription of the acyl-homoserine lactone (acyl-HSL) autoinducer synthase homolog (sinI). Additional data from two-dimensional (2D) gel analyses identified 10 proteins that are down-regulated during biotin starvation, two of which are essential for transcription and translation and one of which is possibly involved in copper tolerance. MATERIALS AND METHODS Bacterial strains and plasmids. The microbiological materials used in the present work are listed in Table 1. Escherichia coli was grown at 37°C on Luria-Bertani medium (23) supplemented with appropriate antibiotics. Rhizobia were cultured at 28°C on GTS (14) or TY medium (0.5% tryptone, 0.25% yeast extract, and 10 mM CaCl2), and growth was monitored by absorbance at 600 nm. For growth under the biotin-limiting conditions, cells were grown in GTS medium as previously described (7). If required, GTS medium was complemented

* Corresponding author. Mailing address: Institut fu ¨r Mikrobiologie und Genetik, Universita¨t Go ¨ttingen, Grisebachstr. 8, 37077 Go ¨ttingen, Germany. Phone: (49) 551-393775. Fax: (49) 551-393793. E-mail: [email protected]. 1206

BIOTIN LIMITATION IN S. MELILOTI

VOL. 69, 2003

1207

TABLE 1. Bacterial strains and plasmids used in this work Strain or plasmid

Relevant trait(s)

Source or reference

E. coli strains DH5␣ S17-1 S17-1-pCAM140

recA1 ⌬lacZ Modified RP4 plasmid integrated into genome Promoterless gusA in mini-Tn5

GibcoBRL 26 32

Plasmids pBK3 PBBR1MCS2 pK18mobsac pK18mobsac-E pBSK⫹

Multipurpose promoter fusion vector Broad-host-range cloning vector; Kmr pACYC184ori mob sacB; Kmr pACYC184ori mob sacB; Kmr, lacking the EcoRI site pBluescriptSK(⫹) multicopy cloning vector; Ampr

16 15 24 This work Stratagene

S. meliloti strains Rm1021 Rm1021-EH45 Rm1021-EH47 Rm1021-B5 Rm1021-sinI-gusA Rm1021-sinR-gusA Rm1021-nodD1-gusA Rm1021-nodD3-gusA Rm1021-plsX-gusA Rm1021-02283-gusA Rm1021-04882-gusA Rm1021-birS-gusA

S. meliloti 1021 parent strain; Smr Rm1021 carrying a gusA gene in pcm in direction of transcription Rm1021 carrying a gfp gene in surE in direction of transcription Rm1021 carrying Tn5-B30 in bioS; Tetr Kmr Rm1021 carrying a sinI-gusA fusion in pBK3 Rm1021 carrying a sinR-gusA fusion in pBK3 Rm1021 carrying a nodD1-gusA fusion in pBK3 Rm1021 carrying a nodD3-gusA fusion in pBK3 Rm1021 carrying a plsXI-gusA fusion in pBK3 Rm1021 carrying a 02283-gusA (copC-gusA) fusion in pBK3 Rm1021 carrying a 04882-gusA fusion in pBK3 Rm1021 carrying a birS-gusA fusion in pBK3

21 This This 27 This This This This This This This This

A. tumefaciens strains NTL4 KYC6

A. tumefaciens traI-lacZ reporter strain A. tumefaciens control strain carrying pCF218 traM mutant

C. Fuqua C. Fuqua

with 40 nM biotin. For the 2D gel analysis, 200-ml cultures were grown under biotin-limiting conditions to stationary phase. One hundred milliliters of the culture was removed, transferred into a sterile flask, and supplemented with biotin (40 nM), and cultures were employed for the preparation of protein extracts as soon as the biotin-supplemented culture resumed growth. ␤-Glucuronidase assays. ␤-Glucuronidase activities in cells harvested from cultures in TY or GTS medium were measured. All measurements were repeated at least three times, using standard ␤-glucuronidase assays (32). Construction of mutant strains and gusA reporter fusions. To obtain a pcmgusA mutant, pcm was amplified by PCR with primers PCM-1 (5⬘-GGGGATA TGCGCATCCTGCTGACGAATGACGA-3⬘) and PCM-2 (5⬘-GGGGCATAT GTTAAAGGAACGACGCTACCTGGTG-3⬘). The resulting fragment was ligated into the SmaI restriction site of pk18mobsac, and a 2-kb DNA fragment containing a promoterless gusA gene was inserted into the 3⬘ XhoI restriction site in the pcm gene to yield pEH45. Prior to this, the gusA gene was amplified from pCAM140 by using gusA-specific primers (GUSA-1 [5⬘-GGGGCTCGAGTCA TTGTTTGCCTCCCTG-3⬘] and GUSA-2 [5⬘-GGGGCTCGAGGAGTCCCTT ATGTTACGTCCTGTA-3⬘]). The gusA-specific primers were designed to introduce an XhoI site at each end of the reporter gene. The correctness of the insert in pEH45 was verified by sequencing from both ends with standard primers. This construct was transformed to E. coli S17-1 and mobilized into S. meliloti. Double recombinants were obtained by plating the transconjugants onto TY medium containing 15% sucrose. Approximately 100 clones were analyzed by PCR and Southern blotting, and several clones carrying a gusA gene in the pcm gene but lacking the parental gene were identified. The S. meliloti clone finally employed in this study was designated Rm1021-EH45. The surE mutation was constructed by employing the same 2-kb DNA fragment but ligating into pk18mobsacE lacking the EcoRI site. To mutate the surE gene, a gfp gene was inserted into the unique EcoRI restriction site of the construct. The resulting clone was verified by sequencing. Double recombinant clones were selected as described above, and the mutant finally used was designated Rm1021-EH47. For the construction of the rhizobial sinI reporter fusion, the DNA region containing the putative promoter region was amplified by using primers SINI-1 (5⬘-GCAAGCTGCAGCGCACGCTG-3⬘) and SINI-2 (5⬘-GAACATCTAGAC GATGGCCTGG-3⬘). Primers were designed to introduce PstI and XbaI restriction sites at the ends of the DNA fragments to allow ligation into the corresponding restriction sites in the pBK3 vector. The resulting sinI-gusA reporter

work work work work work work work work work work

fusion was analyzed by DNA sequencing and mobilized into S. meliloti strain 1021. For construction of all other reporter fusions, published or putative promoter regions were amplified by PCR with specific primers. The primers employed were NODD3-1 (5⬘-GCATCTGCAGGGACAGCATCTTC ⫺3⬘), NODD3-2 (5⬘-CTCGTTCTA GAAGGAGCGTAGGCG-3⬘), NODD1-1 (5⬘-TT CCACTGCAGTTTTAAGGACATGTAAC-3⬘), NODD1-2 (5⬘-GACGAG GTCTAGATCTAGGCCCCTA-3⬘), SINR-1 (5⬘-GGCGCTGCAGGCGCATA TTCTG-3⬘), SINR-2 (5⬘-GATGTTCTAGACGCATCAGGGCG-3⬘), ORF04882-1 (5⬘-CGTCTGCAGATACGAAACTATTCTC-3⬘), ORF04882-2 (5⬘-CACGTC TAGAACCAGTCGGCGGC-3⬘), COPC-1 (5⬘-GGTGCTGCAGGGCATCTC GGCGG-3⬘), COPC-2 (5⬘-TGAGTCTAGAGCCGCGGCAAGAAG-3⬘), PLSX-1 (5⬘-GGTGCTGCAGGAC TGGAAAAAGG-3⬘), PLSX-2 (5⬘-GACTTCTAGA CCATAGTCGCCTCC-3⬘), BIRS-1 (5⬘-GACCTGCAGCCGGAAAGGCGCC GATGATGCCGGCCTTGC-3⬘), and BIRS-2 (5⬘-GACTCTAGATCCTTCTG CGCGAGTCTTCCGGTTGCCGC-3⬘); cloning was performed as described above. The Agrobacterium tumefaciens traI-lacZ reporter strain was used for the specific detection of acyl-HSLs by the method described by Zhu et al. (33) and Shaw et al. (25). 2D gel analysis. Proteins on analytical 2D gels were visualized by silver staining and digitized with a Mikrotek Scanmaker 4. Spot detection, gel alignment, and gel-to-gel protein spot matching were performed with ImageMaster 2D software (Amersham-Pharmacia, Freiburg, Germany). For the analysis of regulated protein spots by matrix-assisted laser desorption ionization–time of flight (MALDITOF) mass spectrometry, an analyzer from Bruke Daltonik (Bremen, Germany) was employed and masses were evaluated by using the Mascot program (Matrix Science). For the comparison of the theoretical masses of the corresponding S. meliloti proteins after tryptic digests, we used the protein data available at http://sequence.toulouse.inra.fr/meliloti.html, data published for the S. meliloti proteome reference maps (20), and data for theoretical tryptic digests available at http://www.expasy.ch/tools/peptide-mass.html.

RESULTS Transcriptional analysis of the pcm-gusA reporter strain. To identify possible biotin-regulated genes in S. meliloti, searches

1208

HEINZ AND STREIT

FIG. 1. (A) Physical organization of the pcm gene on the S. meliloti chromosome. surE codes for an acid phosphatase (unpublished data), and bioS codes for a regulator. (B) Transcription of a pcm-gusA fusion in Rm1021-EH45. Data represent mean values from three different cultures, and error bars indicate the standard deviations. For measurements under biotin-limiting conditions, cells were transferred three times into medium without biotin prior to the experiment. Open bars, pcm-gusA activities in cells grown under biotin-limiting conditions; shaded bars, pcm-gusA transcription in the presence of 40 nM biotin in defined medium. (C) Survival of S. meliloti strain Rm1021 in GTS medium lacking additional biotin. Data represent mean values from three different cultures, and error bars indicate the standard deviations. ■, wild-type cultures; F, pcm mutant cultures.

in the genome database were performed. Since the rhizobial pcm gene is located upstream of the unique biotin-regulated gene bioS (Fig. 1A), it was of interest to investigate the regulation of pcm. For this purpose the reporter strain Rm1021EH45 was constructed and used to measure transcription. When pcm-gusA transcription was measured in GTS medium containing small amounts of biotin, transcription increased in the first days of incubation but declined sharply over time and was hardly detectable in cultures which had been grown for 49 days (Fig. 1B). Interestingly, pcm-gusA transcription was higher in cells, which had been growing under biotin-limiting conditions. Under those conditions, pcm-gusA transcription

APPL. ENVIRON. MICROBIOL.

increased during the first 25 days of incubation and declined from then on (Fig. 1B). Surprisingly, after prolonged incubation for 70 days under biotin-limiting conditions, pcm-gusA was still transcribed at 14-fold-higher levels than in cells which had been growing in the presence of small amounts of biotin (40 nM). Also, when we measured transcription of the pcm-gusA reporter fusion in an S. meliloti strain which carried an intact copy of the wild-type gene, a similar result was observed (data not shown). Together, these findings suggested that pcm transcription was influenced by biotin availability. Survival of Rm1021 and Rm1021-EH45 in defined media. Since the S. meliloti pcm gene was transcribed at higher levels under the biotin-limiting conditions (Fig. 1B) and pcm mutations affect survival in E. coli, we were interested in the importance of the rhizobial pcm gene for survival under these conditions. Therefore, we analyzed survival of Rm1021 wild-type and Rm1021-EH45 pcm mutant cells under biotin-limiting conditions. In these tests the pcm mutant was impaired in its ability to survive (Fig. 1C). After 6 days of stationary-phase survival, the CFU measured for the pcm mutant were already 3.2-fold lower than the CFU measured for the parent strain. This difference was even more pronounced after 35 days of incubation under the biotin-limiting conditions, when CFU observed for the mutant strain were 29.1-fold lower than those measured for the wild-type strain (Fig. 1C). These findings supported the hypothesis that pcm is important for S. meliloti survival. Because the longevity of Rm1021-EH45 was adversely affected in the absence of exogenous biotin, we speculated that the lack of biotin was the cause of the observed phenotype, and therefore we analyzed its survival in the presence of 40 nM biotin. Although the initial decline in CFU was much faster than under the biotin-limiting conditions, the mutant and wildtype strains survived equally well over a time period of 33 days. In one such experiment, we measured for the parent strain 1.9 ⫻ 109 CFU and for the mutant strain 1.8 ⫻ 109 CFU after 2 days of growth in the presence of 40 nM biotin in defined medium. However, the titers of parent and mutant strains declined within 10 days to 6.7 ⫻ 107 and 7.6 ⫻ 107 CFU, respectively. After 33 days of growth, we measured for the parent strain 9.0 ⫻ 107 CFU and for the mutant 8.6 ⫻ 107 CFU. Thus, these data indicate that the lack of biotin indeed resulted in a decreased longevity of Rm1021-EH45. Additional tests were performed to ensure that the observed phenotype was linked to pcm only and that none of the flanking genes was involved. Therefore, growth and survival of a surE and a bioS mutant strain were analyzed. However, mutations in surE or bioS had no obvious influence on survival under biotinlimiting conditions (data not shown). Further tests were performed to analyze the influence of other environmental factors on S. meliloti survival. However, no differences in CFU between parent and mutant cells were observed when cells were subjected to various types of stress in growth and starvation experiments. Those tests included exposure of Rm1021 and Rm1021-EH45 cells to higher temperatures (41°C) for 2 and 20 min. Also, no influence of the pcm mutation on growth and survival in the presence of 0.2 M NaCl, H2O2 (27 and 40 mM), or methanol (0.5 and 1%) was detected (data not shown). Therefore, our findings suggested that the mutation in pcm was the cause of the decreased survival under biotin-limiting con-

VOL. 69, 2003

ditions and that the mutation probably had no polar effect on the genes flanking pcm. Biotin limitation induces genes involved in quorum sensing. To test the influence of biotin limitation on the expression of genes involved in rhizobial quorum sensing, an S. meliloti autoinducer synthase reporter fusion (sinI-gusA) was constructed as described above. The resulting sinI-gusA fusion was mobilized into Rm1021, and transcription in response to biotin availability in defined medium was analyzed. As expected, cells grown in exponential-phase cultures in GTS medium not containing biotin showed higher sinI-gusA transcription than cells which had been growing in the presence of 40 nM biotin (Fig. 2B). This difference was most pronounced in cultures growing for less than 48 h. Under those conditions, the sinI-gusA fusion was transcribed at a 4.5-fold-higher level in biotin-limited cultures than in cultures which had been supplemented with biotin. However, in biotin-limited cultures which had been growing for 5 days, the sinI-gusA transcription was only 30% higher than in cells which had been growing in the presence of 40 nM biotin (Fig. 2B). Further tests indicated that the addition of small amounts of the biotin-complexing protein avidin to cultures which had been grown in biotin-supplemented medium resulted in significantly increased sinI-gusA transcription (Fig. 2C). As a control, we also measured the amount of HSLs produced by S. meliloti strain 1021 by using the A. tumefaciens traI-lacZ reporter strain (NTL4). The amounts of HSLs detected in those tests confirmed our finding with the rhizobial sinI-gusA fusion (Fig. 3). The amounts of HSLs detected in the culture supernatants from cells which had been grown under biotin-limiting conditions were severalfold higher than those detected in supernatants from biotin-supplemented cultures. Together, our data indicated that in S. meliloti, biotin-limiting conditions either directly or indirectly induce increased sinI transcription. Transcriptional studies with additional gusA fusions. Besides the sinI and pcm promoters, several other S. meliloti promoters were chosen for the construction of gusA reporter strains. The promoters tested for biotin responsiveness in this work together with the sinI and pcm promoter are summarized in Table 2, and the overall induction rates observed are indicated. The additional promoters or genes were selected because of their known or speculated roles in S. meliloti fatty acid biosynthesis (plsX), possible involvement in biotin sensing (birS and orf04882) (unpublished data), role in nodulation regulation (nodD1 and nodD3), or possible role in culturability under nutrient starvation (orf02283 [copC]). In addition to the pcm and sinI reporter genes, only the possible copC gene transcription locus appeared to be significantly affected by biotin availability. The transcription of the putative copC reporter fusion was decreased 21-fold under the biotin-limiting conditions compared to the biotin-sufficient growth conditions. This effect was most pronounced in cultures which had been growing for 48 h and which were still in the exponential growth phase (data not shown). Transcription of all other tested reporter fusions was not significantly altered under the conditions tested (Table 2). Comparative analysis of 2D gels of biotin-limited and biotin-sufficient cultures. To further identify possible biotin-regulated genes and proteins, 2D protein profiles of biotin-limited

BIOTIN LIMITATION IN S. MELILOTI

1209

FIG. 2. (A) Physical organization of the sinI gene on the S. meliloti chromosome. sinR codes for the autoinducer regulator, and sinI codes for an autoinducer synthase. (B) Transcription of a sinI-gusA fusion in Rm1021-sinI-gusA. For measurements under biotin-limiting conditions, cells were transferred three times into medium without biotin prior to the experiment. Open bars, sinI-gusA activities in cells grown under biotin limiting-conditions; shaded bars, sinI-gusA transcription in the presence of 40 nM biotin in defined medium. (C) Transcription of a sinI-gusA fusion in cells which were grown for 24 h in GTS medium containing biotin and then washed and transferred to fresh medium containing no additional biotin (open bar), 40 nM biotin (light shaded bar), or 40 nm biotin plus 0.65 U of avidin ml⫺1 (dark shaded bar). Data represent mean values from three different cultures. Errors bars indicate standard deviations.

cells and biotin-supplied cells were compared. Figure 4 shows the overall expression patterns of proteins from biotin-limited and biotin-supplied cells obtained with pI strips from pI 4 to 7. Overall, 50 to 100 differentially expressed proteins are visible on the two gels. A more detailed analysis of the boxed regions in Fig. 4 by using pI strips from pI 4.5 to 5.5 identified at least 15 to 20 differentially expressed protein spots. From those, 10 abundant spots were selected for a MALDI-TOF mass spectrometry analysis, and functions were assigned based on their homologies to known proteins in the S. meliloti genome database. The proteins identified are listed in Table 3, and the

1210

HEINZ AND STREIT

APPL. ENVIRON. MICROBIOL.

FIG. 3. Thin-layer chromatography analysis of HSLs produced in S. meliloti and detected by using the A. tumefaciens NTL4 (traI-lacZ) reporter strain. Cells were harvested at early exponential growth phase, and the HSLs were extracted as described in Materials and Methods. Two-microliter portions of the extracts were loaded onto the thin-layer chromatography plates. Lane A, HSL extracts derived from biotin-starved cells; lane B, HSL from cells which were growing in the presence of 40 nM biotin.

corresponding spots are depicted in Fig. 5. Protein spot A matched the 50S ribosomal protein subunit L7/L12 (L8), thus indicating that in biotin-starved cells, protein biosynthesis might be affected. Spot D matched the ␼-subunit of the RNA polymerase. This finding further suggests that biotin starvation affects transcription in S. meliloti. Most other proteins identified are involved in transport of potential carbon sources (spot F), amino acids (spots E, K, and L), or intermediates from central metabolism (spot I), and one enzyme is involved in sugar metabolism (spot G). Surprisingly, spots B and C were identified as potential copper resistance proteins matching open reading frame smc02283. Therefore, these protein data confirm the data obtained by using the smc02283-gusA (copCgusA) reporter fusion and suggest a possible link between copper availability and biotin starvation in S. meliloti 1021. DISCUSSION The results from this study clearly show that S. meliloti 1021 has evolved several different strategies to survive periods of

biotin starvation: (i) synthesis of an L-isoaspartyl methyltransferase involved in protein repair; (ii) elevated transcription of a traI (sinI) homologue commonly associated with quorum sensing in other microbes, (iii) decreased transcription of a possible copC homologue, and (iv) significantly reduced expression of proteins involved in transcription, translation, and metabolic activities. Susceptibility of pcm mutants to biotin limitation and other environmental stresses in extended stationary-phase cultures. The S. meliloti L-isoaspartyl protein repair methyltransferase (Pcm) (EC 2.1.1.77) belongs to a group of highly conserved enzymes (11). The L-isoaspartyl residues occur in aging cells as a result of spontaneous deamidations of asparaginyl residues and isomerizations of aspartyl residues. Pcm is known to be involved in the repair of these accumulating L-isoaspartyl residues, and it has been shown in several cases that a lack of pcm is detrimental to cell survival (13, 17, 30). In E. coli, pcm mutations affect survival, but mutant phenotypes are observed only when cells are subjected to various types of stress in stationary phase (30). Among the reported stresses influencing the survival of an E. coli mutant was the presence of high salt concentrations in the medium and heat treatments. Interestingly, in S. meliloti these treatments did not lead to significantly decreased survival of the pcm mutant employed in this study. Consistent with increased transcription of pcm (Fig. 1B) under biotin-limiting conditions, mutations in this gene compromised rhizobial survival (Fig. 1C). However, it is highly intriguing that this phenotype was observed only in cells which were not supplied with exogenous biotin. We can only speculate about the ecological significance of this observation, but a possible solution to this puzzling phenomenon may be found in the unique organization of the S. meliloti pcm gene together with the bioS gene. This unique organization may be a special adaptation of S. meliloti to survival under biotin-limiting conditions. This hypothesis is supported by searches in genomes of 183 bacteria in the ERGO database (Integrated Genomics) and by analyzing the flanking genes of pcm in these microbes. Interestingly, none of the microbes whose genome sequences are available in this database carry a pcm gene linked to a biotin-regulated gene. Biotin starvation results in increased sinI transcription. The results from this study offer positive evidence that S. meliloti 1021 sinI transcription is increased by the absence of sufficient biotin in defined media. Growing S. meliloti serially under biotin-limiting conditions produces several physiological

TABLE 2. Promoters tested for biotin-dependent expression in S. melioti and the overall induction observed Promoter

Gene and/or function

Induction (fold)a

nodD1 nodD3 plsX pcm sinI sinR orf02283 orf04882 birS

Nodulation-regulatory locus, LysR-type regulator Nodulation-regulatory locus, LysR-type regulator Involved in fatty acid and phospholipid biosynthesis L-Isoaspartyl protein repair methyltransferase HSL autoinducer synthase, traI homologue HSL autoinducer transcriptional regulator, traR homologue Possible copper resistance gene copC Possible LysR-type regulator involved in biotin response Regulatory locus involved in starvation response (GenBank accession no. AF274306)

None None None ⫹13.6 ⫹4.3 ⫹1.3 ⫺21 None None

a Induction observed under biotin-limiting conditions compared to induction under biotin-sufficient conditions. None, no significant induction; ⫹, positive induction; ⫺, negative induction.

BIOTIN LIMITATION IN S. MELILOTI

VOL. 69, 2003

1211

FIG. 4. 2D gel image representing a protein expression window showing cellular proteins from S. meliloti 1021 between pI 4 and 7. Immobilized pH gradient strips 4 to 7 were used for the first dimension and loaded with 100 ␮g of protein. (A) Cellular proteins of S. meliloti supplied with 40 nM biotin; (B) proteins of cells grown in the absence of biotin. The most obvious differences were observed in the boxed area.

and metabolic changes, including the accumulation of poly-3hydroxybutyrate and a significant reduction in cell size (6, 10). Biotin-dependent enzymes such as pyruvate carboxylase also are affected under biotin-limiting conditions, and several tricarboxylic acid cycle auxiliary enzymes show decreased activities (3, 4). However, it is an absolutely novel finding that biotin-limiting growth conditions directly or indirectly induce elevated transcription of genes related to quorum sensing. The S. meliloti sinI/sinR locus has been identified only recently (19) together with a second locus involved in HSL synthesis. Within this framework, it should be noted that stationary-phase production of HSLs helps Rhizobium leguminosarum to better survive other types of stress, such as carbon and nitrogenlimiting conditions or osmotic stress (29). Therefore, future work needs to explore whether the increased HSL production under biotin-limiting conditions will also result in increased stress survival in S. meliloti. Biotin-limiting growth conditions induce decreased copCgusA transcription. It is highly intriguing that transcription and translation of the possible copC homologue are affected by biotin availability (Fig. 5). Interestingly, under nutrient starva-

tion, small amounts of copper induce a viable-but-nonculturable condition (VBNC) in A. tumefaciens and R. leguminosarum. (1), and a similar effect has been described for S. meliloti (18). Unfortunately, the molecular mechanisms of the rhizobial response are not understood. CopA and CopC are assumed to be involved in copper detoxification and are probably part of a transport system. However, no direct link has been made between these genes and the VBNC of S. meliloti or other rhizobial species. Although we have not estimated in our experiments to what extent biotin-limiting growth conditions affect the VBNC of S. meliloti, our findings might offer initial clues to better understand the VBNC of S. meliloti under conditions of nutrient starvation. Reduced expression of proteins involved in transcription and translation. The 2D gel profile comparison of biotin-supplied and biotin-limited cells confirmed that the availability of biotin has a strong influence on cellular processes in S. meliloti. Overall, between 50 and 100 differentially expressed proteins were observed (Fig. 4). However, it should be noted that it is to be expected that under these two extreme conditions of biotin starvation versus biotin sufficiency many genes will be ex-

TABLE 3. Identified S. meliloti 1021 proteins which are down-regulated in response to biotin starvation Spot

a

A B/C D E F G H I J K/L

S. meliloti protein; function

pI/molecular mass (kDa)

SMc01318/RpIL; 50S ribosomal protein L7/L12 (L8) SMc02283; probable copper resistance protein SMc02408/RpoZ; ␻ subunit of RNA polymerase SMc02118/AapJ; ABC transporter, L-amino acid periplasmic binding protein SMb20484; ABC transporter periplasmic sugar binding protein SMc03153/Eda2; 2-keto-3-deoxy-6-phosphogluconate aldolase SMc02774; ABC transporter periplasmic sugar binding protein SMb20416/UgpB; ABC transporter periplasmic glycerol-3-phosphate binding protein SMb20025; hypothetical protein SMc01946/LivK; leucine-specific binding protein

4.8/12.8 5.3/17.2 4.2/14.9 5.0/36.8 5.4/34.9 4.7/21.6 5.1/34.0 5.2/46.6 5.3/34.1 5.0/39.0

a The corresponding protein spots are depicted in Fig. 5 in the 2D gel arrays. The protein spots B/C and K/L were identified as identical proteins and are probably modified forms of the same protein.

1212

HEINZ AND STREIT

APPL. ENVIRON. MICROBIOL. ACKNOWLEDGMENTS This work was supported by the Fonds der Chemischen Industrie and by DFG grant STR415-2-3 to W.R.S. W.R.S. thanks C. Fuqua for providing the A. tumefaciens reporter strain and W. Liebl for helpful discussions.

FIG. 5. 2D gel images showing cellular proteins from S. meliloti 1021 between pI 4 and 7 (upper four panels) and between pI 4.5 and 5.5 (lower eight panels). Protein patterns from biotin-supplied and biotin-starved cells were analyzed. Letters and arrows indicate the spots identified by MALDI-TOF mass spectrometry. The proteins identified are listed in Table 3 together with their assumed functions, pI values, and molecular masses.

pressed differentially and that many of these changes are probably minor metabolic adjustments. Also, within this framework it should be mentioned that growth rates are affected by biotin availability (6, 27), and this may also influence protein patterns. The expression of the ribosomal protein (L7/L12) and a core subunit of the RNA polymerase clearly suggests that transcription and translation are affected in biotin-limited cells, and this will certainly lead to a shut down of metabolic activities (Table 3). Together, the identification of two genes, sinI and pcm, which are up-regulated in response to biotin-limitation and the identification of several proteins that are down-regulated under these conditions suggest that biotin is a regulatory compound for S. meliloti. Therefore, defining the molecular mechanisms that S. meliloti employs to sense biotin-limiting conditions remains a worthwhile future research objective.

REFERENCES 1. Alexander, E., D. Pham, and T. R. Steck. 1999. The viable-but-nonculturable condition is induced by copper in Agrobacterium tumefaciens and Rhizobium leguminosarum. Appl. Environ. Microbiol. 65:3754–3756. 2. Broughton, W. J., and X. Perret. 1999. Genealogy of legume-Rhizobium symbioses. Curr. Opin. Plant Biol. 2:305–311. 3. Dunn, M. F. 1998. Tricarboxylic acid cycle and anaplerotic enzymes in rhizobia. FEMS Microbiol. Rev. 22:105–123. 4. Dunn, M. F., G. Araiza, M. A. Cevallos, and J. Mora. 1997. Regulation of pyruvate carboxylase in Rhizobium etli. FEMS Microbiol. Lett. 157:301–306. 5. Dunn, M. F., G. Araiza, and T. M. Finan. 2001. Cloning and characterization of the pyruvate carboxylase from Sinorhizobium meliloti Rm1021. Arch. Microbiol. 176:355–363. 6. Encarnacion, S., M. Dunn, K. Willms, and J. Mora. 1995. Fermentative and aerobic metabolism in Rhizobium etli. J. Bacteriol. 177:3058–3066. 7. Entcheva, P., D. A. Phillips, and W. R. Streit. 2002. Functional analysis of Sinorhizobium meliloti genes involved in biotin synthesis and transport. Appl. Environ. Microbiol. 68:2843–2848. 8. Ferenci, T. 2001. Hungry bacteria—definition and properties of a nutritional state. Environ. Microbiol. 3:605–611. 9. Heinz, E. B., D. A. Phillips, and W. R. Streit. 1999. BioS, a biotin-induced, stationary-phase, and possible LysR-type regulator in Sinorhizobium meliloti. Mol. Plant Microbe Interact. 12:803–812. 10. Hofmann, K., E. B. Heinz, T. C. Charles, M. Hoppert, W. Liebl, and W. R. Streit. 2000. Sinorhizobium meliloti strain 1021 bioS and bdhA gene transcriptions are both affected by biotin available in defined medium. FEMS Microbiol. Lett. 182:41–44. 11. Kagan, R. M., H. J. McFadden, P. N. McFadden, C. O’Connor, and S. Clarke. 1997. Molecular phylogenetics of a protein repair methyltransferase. Comp. Biochem. Physiol. B 117:379–385. 12. Kaminski, P. A., J. Batut, and P. Boistard. 1998. A survey of symbiotic nitrogen fixation by rhizobia, p. 431–460. In H. P. Spaink, A. Kondorosi, and P. J. J. Hooykaas (ed.), The Rhizobiaceae. Kluwer Academic Publishers, Dordrecht, The Netherlands. 13. Kim, E., J. D. Lowenson, D. C. MacLaren, S. Clarke, and S. G. Young. 1997. Deficiency of a protein-repair enzyme results in the accumulation of altered proteins, retardation of growth, and fatal seizures in mice. Proc. Natl. Acad. Sci. USA 94:6132–6137. 14. Kiss, G. B., E. Vincze, Z. Kalman, T. Forrai, and A. Kondorosi. 1979. Genetic and biochemical analysis of mutants affected in nitrate reduction in Rhizobium meliloti. J. Gen. Microbiol. 113:105–118. 15. Kovach, M. E., P. H. Elzer, D. S. Hill, G. T. Robertson, M. A. Farris, R. M. Roop II, and K. M. Peterson. 1995. Four new derivatives of the broad-hostrange cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166:175–176. 16. Kusian, B., R. Bednarski, M. Husemann, and B. Bowien. 1995. Characterization of the duplicate ribulose-1,5-bisphosphate carboxylase genes and cbb promoters of Alcaligenes eutrophus. J. Bacteriol. 177:4442–4450. 17. Lowenson, J. D., E. Kim, S. G. Young, and S. Clarke. 2001. Limited accumulation of damaged proteins in L-isoaspartyl (D-aspartyl) O-methyltransferase-deficient mice. J. Biol. Chem. 276:20695–20702. 18. Manahan, S. R., and T. R. Steck. 1997. The viable but nonculturable state in Agrobacterium tumefaciens and Rhizobium meliloti. FEMS Microbiol. Ecol. 22:29–38. 19. Marketon, M. M., and J. E. Gonzalez. 2002. Identification of two quorumsensing systems in Sinorhizobium meliloti. J. Bacteriol. 184:3466–3475. 20. Mathesius, U., N. Imin, H. Chen, M. A. Djordjevic, J. J. Weinman, S. H. Natera, A. C. Morris, T. Kerim, S. Paul, C. Menzel, G. F. Weiller, and B. G. Rolfe. 2002. Evaluation of proteome reference maps for cross-species identification of proteins by peptide mass fingerprinting. Proteomics 2:1288–1303. 21. Meade, H. M., S. R. Long, G. B. Ruvkun, S. E. Brown, and F. M. Ausubel. 1982. Physical and genetic characterization of symbiotic and auxotrophic mutants of Rhizobium meliloti induced by transposon Tn5 mutagenesis. J. Bacteriol. 149:114–122. 22. Oke, V., and S. R. Long. 1999. Bacterial genes induced within the nodule during the Rhizobium-legume symbiosis. Mol. Microbiol. 32:837–849. 23. Sambrook, J., E. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 24. Schafer, A., A. Tauch, W. Jager, J. Kalinowski, G. Thierbach, and A. Puhler. 1994. Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145:69–73. 25. Shaw, P. D., G. Ping, S. L. Daly, C. Cha, J. E. Cronan, Jr., K. L. Rinehart, and S. K. Farrand. 1997. Detecting and characterizing N-acyl-homoserine

VOL. 69, 2003

26. 27. 28. 29.

lactone signal molecules by thin-layer chromatography. Proc. Natl. Acad. Sci. USA 94:6036–6041. Simon, R., U. Priefer, and A. Pu ¨hler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gramnegative bacteria. Bio/Technology 1:784–791. Streit, W. R., C. M. Joseph, and D. A. Phillips. 1996. Biotin and other water-soluble vitamins are key growth factors for alfalfa root colonization by Rhizobium meliloti 1021. Mol. Plant Microbe Interact. 9:330–338. Streit, W. R., and D. A. Phillips. 1997. A biotin-regulated locus, bioS, in a possible survival operon of Rhizobium meliloti. Mol. Plant Microbe Interact. 10:933–937. Thorne, S. H., and H. D. Williams. 1999. Cell density-dependent starvation survival of Rhizobium leguminosarum bv. phaseoli: identification of the role of an N-acyl homoserine lactone in adaptation to stationary-phase survival. J. Bacteriol. 181:981–990.

BIOTIN LIMITATION IN S. MELILOTI

1213

30. Visick, J. E., H. Cai, and S. Clarke. 1998. The L-isoaspartyl protein repair methyltransferase enhances survival of aging Escherichia coli subjected to secondary environmental stresses. J. Bacteriol. 180:2623–2629. 31. Watson, R. J., R. Heys, T. Martin, and M. Savard. 2001. Sinorhizobium meliloti cells require biotin and either cobalt or methionine for growth. Appl. Environ. Microbiol. 67:3767–3770. 32. Wilson, K. J., A. Sessitsch, J. C. Corbo, K. E. Giller, A. D. Akkermans, and R. A. Jefferson. 1995. ␤-Glucuronidase (GUS) transposons for ecological and genetic studies of rhizobia and other gram-negative bacteria. Microbiology 141:1691–1705. 33. Zhu, J., J. W. Beaber, M. I. More, C. Fuqua, A. Eberhard, and S. C. Winans. 1998. Analogs of the autoinducer 3-oxooctanoyl-homoserine lactone strongly inhibit activity of the TraR protein of Agrobacterium tumefaciens. J. Bacteriol. 180:5398–5405.