INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Microorganisms are able to survive under a wide range of environmental conditions (e.g., osmolarity, temperature, and nutrient availability) by rapidly adapting their structure and physiology. These mechanisms are based on the existence of multiple regulatory systems in which gene expression is controlled in a coordinate manner in response to environmental stimuli. One example of such a complex process is the regulation of motility and chemotaxis in bacteria (16).

More than 80% of the known bacterial species are motile by means of flagella (18). The structure and arrangement of flagella differ from species to species and seem to be related to the specific environments in which the cells live (29). Flagella can be arranged on the cell body in a variety of configurations, including single polar, multiple polar, and many peritrichous (or lateral) configurations. Motility by means of flagella is thought to provide a specific advantage for a bacterium (18), because it helps the bacterium to reach the most favorable environment and to successfully compete with other microorganisms. However, the cost of maintenance of a flagellar motility system is high for bacteria (about 2% of biosynthetic energy expenditure in Escherichia coli) due to the number of genes and the energy required for flagellum synthesis and functioning. As a result, the flagellar system is highly regulated (16). Given its importance for bacterial survival under specific conditions, the efficiency of control of this complex system seems to be under strong selective pressure in the environment.

In the present study, we analyzed motility regulation by environmental factors in bacterial strains isolated from a specific natural habitat. Lake Baikal, located in eastern Siberia, is one of the oldest (25 million years) and the deepest (maximum depth, 1,637 m) lakes in the world (11). It represents a particular ecosystem with unique characteristics. Significant seasonal changes in temperature take place only in the top layers of water up to depths of 200 to 250 m. In summer, the surface layers of open deepwater regions of Lake Baikal reach a maximum of 12 to 16°C. The waters of Lake Baikal are poorly mineralized soft waters of the hydrocarbonate class, calcium group. In this oligotrophic lake the sum of the concentrations of major ions is about 100 mg/liter, and the content of biogenic elements and organic matter is insignificant (11). Thus, the aquatic bacteria grow under conditions of low temperature and low contents of mineral and other nutrient compounds. Our results demonstrated that the control of motility in response to changes in these environmental parameters is largely conserved in bacteria with different optimal growth temperatures and with different type of flagellation, despite different organizations of their flagellum regulatory cascades.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and growth conditions. The bacterial strains and plasmids used in this study are listed in Table 1. Lake Baikal strains were isolated on 10-fold-diluted RPA solid medium, containing 1.79 g of pancreatic fish hydrolysate per liter, 0.59 g of NaCl per liter, and 11.2 g of Bacto-agar per liter (6). Strains were then grown at 25°C in Luria-Bertani medium (17), tryptone medium (17), or M9 medium (17) supplemented with 0.1% (wt/vol) Casamino Acids. Tryptone swarm plates containing 1% Bacto-tryptone, 0.5% NaCl, and 0.3% Bacto-agar or 10-fold diluted RPA medium (6) with 0.3% agar were used to test bacterial motility as previously described (1). When required, tetracycline was added at a concentration of 15 µg/ml. All experiments were performed in accordance with the European regulation requirements concerning the contained use of genetically modified organisms of group I and group II (agreement no. 2735 and 2736 CAII).

Phylogenetic analysis. DNA fragments of about 1,380 nucleotides encompassing the 16S rRNA gene were PCR amplified and sequenced on both strands by Genome Express (Montreuil, France). The 16S rRNA gene sequences were screened against sequences deposited in databases using the Blast program (2). DNA sequences were aligned, and a phylogenetic tree was constructed as previously described (21).

PCR amplification with degenerate primers. The degenerate primers fleQ5' and fleQ3' (Table 2) used for the PCR amplification of Pseudomonas fleQ-homologous genes were designed on the basis of specifically conserved regions in the nucleotide alignment of the Pseudomonas aeruginosa fleQ (GenBank accession no. L49378) and Vibrio cholerae flrA (GenBank accession no. AF014113) genes.

Direct sequencing of chromosomal DNA. The sequencing of the flhDC operon of Enterobacter strain 22 was performed as previously described (13) with two degenerate oligonucleotides, i.e., flhD3' and flhC3' (Table 2), hybridizing with the more conserved region in homologous sequences available in databases. This was followed by several direct sequencing steps allowing the determination of the entire flhDC operon sequence with oligonucleotides Ent1 to Ent8 (Table 2). To complete the fleQ gene sequence of Pseudomonas strain Y1000, several direct sequencing steps were performed on chromosomal DNA with oligonucleotides Y3 to Y6 (Table 2).

Plasmid construction. Plasmid pDIA575 was constructed by PCR amplification of the flhDC fragment with primers EntXba5 and EntXho3 (Table 2) from chromosomal DNA of Enterobacter strain 22. The 1,481-nucleotide fragment was cloned into the XbaI and XhoI sites of the broad-host-range vector pBBR1MCS-3 (10).

Primer extension. Total RNA was extracted from 20 ml of culture grown in M9 minimal medium (17) supplemented with 0.1% Casamino Acids or in tryptone medium (17) to an optical density at 600 nm (OD₆₀₀) of 0.3 using FastPrep system (Bio 101 Savant) and Trizol solution (Gibco BRL) (I. Guillouard, unpublished data). RNA concentration and purity were determined by OD₂₆₀ and OD₂₈₀ measurements. Transcriptional start sites were determined as previously described (26). The reactions were performed with 10 and 20 µg of total RNA of Enterobacter and Pseudomonas, respectively, with -³²P-end-labeled oligonucleotides E1 and Y1 or Y2 (Table 2). As a reference, sequencing reactions were performed on plasmid pDIA575 or pDIA576, using a Thermosequenase radiolabeled terminator cycle sequencing kit from Amersham with the same primer as used in primer extension experiments. Bands from Hyperfilm-MP X-ray film (Amersham) were scanned with a JX-330 SHARP scanner and quantified with the PDI software PDQuest on a SUN computer system.

Nucleotide sequence accession numbers. The 1,379-nucleotide sequence of the 16S rRNA gene of Pseudomonas strain Y1000 was in accordance with the partial sequence in databases under accession number X99676. The 1,330-nucleotide 16S rRNA sequence of Enterobacter strain 22 has been assigned EMBL nucleotide sequence database accession no. AJ308467. The 2,026- and 1,481-nucleotide sequences containing the complete sequences of the fleQ gene and the flhDC operon have been assigned EMBL nucleotide sequence database accession no. AJ308470 and AJ308469, respectively.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Characterization of natural isolates from Lake Baikal. Two gram-negative bacteria were isolated from Lake Baikal water samples collected from the South Baikal (strain Y1000) in summer 1995 and from the Central Basin (strain 22) in September 1996 during water sampling expeditions of the Limnological Institute of the Siberian Division of the Russian Academy of Sciences. Water samples were taken at 1,000 and 1,200 m below the surface of the lake for strains Y1000 and 22, respectively. The cultures were enriched by plating on diluted medium (6) and incubation at 5°C or at room temperature for 3 days. All strains were able to grow at a wide range of temperatures, extending from 4 to 37°C, with an optimum growth temperature of around 25°C. Morphological, biochemical, and phenotypic characterization (Table 3) demonstrated that strains 22 and Y1000 belong to the Enterobacter and Pseudomonas genera, respectively, of the Proteobacteria gamma subdivision (12). The taxonomic positions of these strains were further investigated by determination of 16S rRNA gene sequences and their comparative analysis with different DNA sequences present in databases. The construction of a phylogenetic tree (data not shown) further supports the phylogenetic positions of these strains, largely in accordance with morphological characterizations, and suggests that strain 22 was closely related to Enterobacter amnigenus JCM1237 and Enterobacter intermedius JCM1238 (99 and 98% sequence identity, respectively), while the nearest relatives of strain Y1000 were Pseudomonas putida ATCC 17522 and Pseudomonas graminis DSM 11363 (98 and 97% sequence identity, respectively).

Effect of environmental factors on bacterial motility. We examined the effects of environmental conditions of Lake Baikal on the motility of isolated strains on semisolid plates at several temperatures ranging from 4 to 37°C and with different concentrations of NaCl. The Enterobacter and Pseudomonas strains were motile at 25°C and remained motile at low temperature (4°C), especially Pseudomonas strain Y1000 (Table 3 and Fig. 1). In contrast, a strong inhibition of motility was observed for both strains at 30 and 37°C or in the presence of NaCl (Fig. 1). Similar effects of these growth conditions on motility were observed by examination of the strains under a light microscope (data not shown).

View larger version (77K): [in this window] [in a new window]   FIG. 1.   Motility assay on semisolid medium plates. 1, Pseudomonas strain Y1000; 2, Enterobacter strain 22. Assays were performed at 25°C (A), at 25°C in the presence of 500 mM NaCl (B), at 37°C (C), or at 4°C (D). Plates were incubated for 13 to 15 h at 25 and 37°C and for 132 h at 4°C. The results are representative of those from three independent experiments.

Identification of flagellar regulatory proteins. The taxonomic positions of Lake Baikal motile strains, i.e., within the laterally flagellated (peritrichous) Enterobacteriaceae family or the polarly flagellated Pseudomonadaceae family (12), suggest the existence of FlhDC regulators (16) in Enterobacter strain 22 and of the FleQ regulator (3) in Pseudomonas strain Y1000. This prompted us to identify the genes encoding these putative master regulatory proteins of flagellar biosynthesis.

Effect of environmental conditions on transcription initiation. To further characterize the master regulator operon of Enterobacter strain 22, we determined the flhDC transcription start site by primer extension experiments with total RNA and primer E1, located upstream from the translational start site of flhDC (Fig. 2A). A single major band for transcription initiation was detected (data not shown), which indicates that transcription of flhDC arises from the A residue located 211 nucleotides upstream from the putative ATG start codon (Fig. 2A). 35 and 10 hexamers showing 50 and 67% similarity, respectively, with the canonic ⁷⁰ consensus sequence were identified upstream from the transcriptional start site. These two boxes are separated by a 17-bp spacer. Moreover, a catabolite gene activator protein (CAP)-like protein-binding site was identified in the regulatory region of the Enterobacter flhDC operon centered at position 71.5 with respect to the transcription start site (Fig. 2A). The presence of a long untranslated region was observed between the transcriptional start site and the ATG initiation codon. Taken together, these observations strongly suggest a mechanism of flhDC regulation similar to the one that we recently characterized for E. coli (26).

View larger version (27K): [in this window] [in a new window]   FIG. 2.   (A) Regulatory region of the flhDC master operon in Enterobacter strain 22. Nucleotides are numbered relative to the transcriptional start site (+1), indicated by a broken arrow. The unique CAP-binding site consensus sequence is indicated by a box. The positions of the 10 and 35 sequences are underlined. A putative ribosome-binding site (RBS) is indicated in boldface. Only the residues corresponding to the N- and C-terminal parts of FlhD and FlhC are indicated. The dashed arrow labeled E1 represents the oligonucleotide used in +1 mapping. (B) Regulatory region of the fleQ gene in Pseudomonas strain Y1000. Transcriptional start sites are indicated by broken arrows; the two major transcriptional start sites P1 and P2 are indicated in boldface. The position of a putative 54-binding site is indicated by a box. A putative ribosome-binding site with similarities with corresponding regions in the P. aeruginosa fleQ gene (accession no. L49378) and V. cholerae flrA gene (accession no. AF014113) and close to the consensus sequence of P. aeruginosa 16S rRNA (23) is indicated in boldface. Only the residues corresponding to the N- and C-terminal parts of FleQ are indicated. Dashed arrows labeled Y1 and Y2 represent oligonucleotides used in +1 mapping.

View larger version (70K): [in this window] [in a new window]   FIG. 3.   Identification of the fleQ transcriptional start site and effects of growth conditions on fleQ expression with primer Y1 (A) and with primer Y2 (B) (see Materials and Methods). Primer extension analysis was performed with RNA extracted from Pseudomonas strain Y1000 grown in exponential phase at room temperature (lanes 1); to late logarithmic phase (lanes 2); at 4°C (lanes 3); in the presence of 500 mM NaCl (lanes 5); at 30°C (lanes 6); or in the presence of 100 µM (lanes 7), 200 µM (lanes 8), and 400 µM (lanes 9) novobiocin and from Pseudomonas strain Y1000 carrying plasmid pDIA576 grown in exponential phase at 25°C (lanes 4). As a reference, a DNA sequencing ladder is shown (lanes G, A, T, and C). The sequence is complementary to the strand shown on the right and was obtained with the same primer as that used for primer extension. Major transcription start sites P1 and P2 are indicated by arrows. The mRNA level was quantified with the PDI software PDQuest on a SUN computer system. Quantitative data are indicated at the bottom of each lane and were expressed relative to the standard condition level (lanes 1), which was assigned a value of 100%. -, background level undetectable by quantification procedure; N.D., not determined.

Role of DNA supercoiling level in control of motility. A link between DNA topology and regulation of gene expression in response to environmental cues has been proposed (28). In particular, a reduction in motility has been observed in E. coli in the presence of DNA gyrase inhibitors, suggesting the involvement of DNA supercoiling in the regulation of bacterial motility (22). To investigate the role of DNA topology in the control of flagellar gene expression in Pseudomonas strain Y1000, we performed a motility assay in the presence of novobiocin, a DNA gyrase inhibitor. A strong reduction in motility was observed in the presence of 200 µM novobiocin, as in Enterobacter strain 22 (data not shown). To further characterize the effect of these conditions on the master regulator gene expression, we carried out primer extension experiments with RNA isolated from Pseudomonas strain Y1000 grown in the presence of 100, 200, and 400 µM novobiocin. No significant effect on growth was observed in the presence of 100 µM novobiocin, and only a small decrease in growth rate was observed in the presence of 200 and 400 µM novobiocin (data not shown). In contrast, a progressive and important decrease in the fleQ major transcript level was measured with the increase in novobiocin concentration. In particular, low expression of fleQ was measured at a 400 µM concentration of DNA gyrase inhibitor (30% and undetectable for P2 and P1 transcripts, respectively), which is quite similar to that observed under various environmental conditions (Fig. 3). Thus, these results suggest a role for DNA supercoiling in the regulation of polar flagellum master gene fleQ expression in Pseudomonas strain Y1000.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In the past years, the hierarchical organization of regulatory systems in gram-negative bacteria has been extensively studied. Bacterial flagellum genes form an ordered cascade in which the expression of one gene located at a given level requires the transcription of another gene at a higher level (16). At the top of the hierarchy are located the flhDC master operon in enterobacteria (16) and the fleQ and flrA master genes in P. aeruginosa (3) and V. cholerae (9), respectively. Flagellar regulatory cascades have also been recently identified in other bacteria, such as Caulobacter crescentus (30), Sinorhizobium (Rhizobium) meliloti (25), and Helicobacter pylori (27). In enterobacteria, the ⁷⁰-dependent flhDC operon is controlled by numerous environmental signals (15, 22) and global regulatory proteins such as H-NS and the cyclic AMP-CAP complex (4, 26). In contrast, the regulation of master genes governing the synthesis of polar flagella remains largely unknown. In the present study we characterized two gram-negative motile bacteria isolated from Lake Baikal. Morphological and phylogenetic analyses suggested that they belong to the laterally flagellated (peritrichous) Enterobacter and polarly flagellated Pseudomonas genera (Table 3). Both isolated strains are psychrotrophic bacteria, especially strain Y1000, suggesting that this organism is more adapted to deep-layer water conditions of Lake Baikal (7). These psychrotrophic properties were further supported by conservation of the motility of these strains at low temperatures (Fig. 1). The physical and chemical characteristics of Lake Baikal waters (11) prompted us to test whether isolated bacteria are able to respond to modifications in these parameters. We showed here that an increase in temperature and a high salt concentration resulted in inhibition of motility in both laterally and polarly flagellated psychrotrophic bacteria isolated from natural environment (Fig. 1). The stationary phase of growth also suppressed motility (Fig. 3 and data not shown), in accordance with growth phase dependence of flagellar gene expression in E. coli (19). Moreover, our results suggest that the master regulatory protein-encoding genes located at the first level of the cascade, i.e., flhDC in Enterobacter and fleQ in Pseudomonas (Fig. 2 and 3), are the major targets for environmental factors and growth phase. Indeed, increasing the fleQ gene dosage from plasmid pDIA576 in Pseudomonas strain Y1000 reversed the reduced motility caused by environmental conditions (data not shown).

A unique transcriptional start site was identified in the flhDC promoter region of Enterobacter strain 22 (Fig. 2A), consistent with the flhDC single major start site observed in E. coli (26) and in Proteus mirabilis (5). By primer extension experiments we identified multiple transcriptional start sites in Pseudomonas strain Y1000, including two major fleQ transcription initiation sites (Fig. 2B and 3). Such a complex promoter structure suggests possible multiple regulations of this gene. In this respect, it should be emphasized that transcription from the first major transcriptional start site (P1 in Fig. 3A) was more sensitive to all adverse conditions tested, including the presence of a gyrase inhibitor, than the second major transcriptional start site (P2 in Fig. 3B). This could result from different rates of transcription from the promoters and/or different mRNA stabilities of the corresponding transcripts. Finally, we have previously shown that a long untranslated region plays an important role in the transcriptional control of the master flagellar operon in E. coli (26). Whether the presence of such a regulatory region (Fig. 2) is involved in the regulation of motility and flagellum biosynthesis in Pseudomonas strain Y1000 and Enterobacter strain 22 remains to be determined.

Despite several studies devoted to motility control in response to environmental conditions (15, 22), the mechanism by which bacteria sense environmental changes remains unknown. Nevertheless, environmental factors such as high temperature and high osmolarity, which are known to induce changes in DNA topology and regulation of gene expression (8, 28), also affect bacterial motility in various microorganisms, such as H. pylori (27) and Y. enterocolitica (20). The loss of motility in E. coli in the presence of novobiocin, a gyrase inhibitor (22), further supports a link between the DNA supercoiling level and motility regulation. Consistent with these data, we demonstrated here that novobiocin and adverse conditions strongly reduce motility and/or flagellar gene expression in Enterobacter strain 22 (data not shown) and Pseudomonas strain Y1000 (Fig. 3), while they have no significant effect on growth. This suggests that the regulation of flagellar systems in both strains could correlate with specific variations in DNA supercoiling.

In this study we demonstrated the existence of remarkable similarities in responses to environmental factors, not only between the flagellar system of natural isolates and that of mesophilic bacteria but also between bacteria with different types of flagellation and regulatory cascades. These observations suggest that the control of bacterial motility has evolved toward a similar mechanism in different bacterial systems. Even though stable physicochemical conditions, i.e., low temperature and low salt concentration, characterize Lake Baikal, bacteria may encounter different temperature and salt conditions due to multiple currents and seasonal or local changes on surface layers or shallow parts (11). Thus, the ability of Lake Baikal bacteria to react to adverse conditions can be of prime importance in the adaptive process of these strains in response to environmental challenges.