JB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Prigent-Combaret, C.
Right arrow Articles by Lejeune, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Prigent-Combaret, C.
Right arrow Articles by Lejeune, P.

Journal of Bacteriology, October 1999, p. 5993-6002, Vol. 181, No. 19
0021-9193/99/$04.00+0
Copyright © 1999, American Society for Microbiology. All rights reserved.

Abiotic Surface Sensing and Biofilm-Dependent Regulation of Gene Expression in Escherichia coli

Claire Prigent-Combaret, Olivier Vidal,dagger Corinne Dorel, and Philippe Lejeune*

Laboratoire de Génétique Moléculaire des Microorganismes et des Interactions Cellulaires, CNRS UMR 5577, Institut National des Sciences Appliquées de Lyon, 69621 Villeurbanne, France

Received 10 March 1999/Accepted 23 July 1999


    ABSTRACT
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

To get further information on bacterial surface sensing and biofilm-dependent regulation of gene expression in Escherichia coli K-12, random insertion mutagenesis with Mu dX, a mini-Mu carrying the promoterless lacZ gene, was performed with an ompR234 adherent strain, and a simple screen was developed to assess changes in gene expression in biofilm cells versus planktonic cells. This screen revealed that major changes in the pattern of gene expression occur during biofilm development: the transcription of 38% of the genes was affected within biofilms. Different cell functions were more expressed in sessile bacteria: the OmpC porin, the high-affinity transport system of glycine betaine (encoded by the proU operon), the colanic acid exopolysaccharide (wca locus, formerly called cps), tripeptidase T (pepT), and the nickel high-affinity transport system (nikA). On the other hand, the syntheses of flagellin (fliC) and of a putative protein of 92 amino acids (f92) were both reduced in biofilms. Such a genetic reprogramming of gene expression in biofilms seems to result from changes in multiple environmental physicochemical conditions. In this work, we show that bacteria within biofilms encounter higher-osmolarity conditions, greater oxygen limitation, and higher cell density than in the liquid phase.


    INTRODUCTION
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Many bacteria can attach to solid surfaces and form biofilms, which are defined as matrix-enclosed microbial populations adherent to each other and to surfaces or interfaces (11). Biofilm formation is such a common phenomenon that virtually every material that comes into contact with naturally occurring fluids, such as blood or seawater, becomes rapidly covered with bacteria (8, 10). Given the important medical and economic consequences of this detrimental situation, understanding the colonization process would help in the design of surface coating methods able to prevent biofilm formation.

Abundant evidence indicates that cells grown on solid surfaces within biofilms are in a physiological state different from that of planktonic cells (reviewed in reference 22). The most difficult properties of biofilm bacteria are their extreme resistance to treatment with biocides and detergents and their high tolerance to prolonged antibiotic therapy in human infections (19, 26). Bacterial structures involved in biofilm formation have been well characterized for some bacteria; AtlE (an autolysin) in Staphylococcus epidermidis (25), type I pili (42) and curli (56) in Escherichia coli, and type IV pili in Pseudomonas aeruginosa (41) have been described as major structures for interaction with the surface. The importance of flagellar motility in initiation of biofilm development in Pseudomonas fluorescens, P. aeruginosa, and E. coli was also reported (40-42).

However, there is still limited information available on bacterial surface sensing at the gene expression level. Surface-induced expression of the laf genes, which are responsible for the synthesis of peritrichous lateral flagella, was correlated with a decreased rotation ability of the unique polar flagellum in Vibrio parahaemolyticus (6, 35) and Proteus mirabilis (5). Specific induced expressions of the P. aeruginosa genes algC (13, 14) and algD (27) and of the sfaA gene (encoding S fimbrial adhesins) of a pathogenic strain of E. coli (47) were also observed after contact of the bacteria with a solid surface, but the surface-sensing mechanisms were not investigated. Furthermore, in many bacterial species, regulation of metabolic functions, particularly those relating to virulence, involves cell-to-cell signalling molecules (such as N-acylhomoserine lactones in gram-negative bacteria). These signal molecules accumulate in the bacterial environment as a function of cell number. Thus, the high density of bacteria within biofilms led to the hypothesis that cell-to-cell signal mechanisms may play an important role in the establishment of the biofilm-specific physiological state. Evidence in support of this hypothesis has been recently provided. Acylhomoserine lactones were detected in biofilms formed on urethral catheters removed from patients (52) and on immersed stones from the San Marcos river in Texas (36). Moreover, Davies and coworkers (15) demonstrated that the LasI-LasR system (one of the two cell-to-cell signalling systems of P. aeruginosa) is required for the normal development of biofilms.

The present work was undertaken in the very well characterized E. coli K-12 context in order to identify biofilm-regulated functions and to gather information on the regulation processes triggered by bacterial contact with the surface. Using a new, reliable system to assess changes in gene expression in biofilm cells versus planktonic cells, we found that 38% of the E. coli genes are differentially expressed in the biofilm. In addition to cell-to-cell signalling mechanisms, microenvironmental conditions of osmolarity and oxygen concentration can be correlated with this major change in gene expression.


    MATERIALS AND METHODS
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Bacterial strains, plasmids, and media. All of the E. coli K-12 strains and plasmids used in this work are listed in Table 1. The bacteria were grown in complete Luria-Bertani (LB) medium (37) or in minimal M63 medium (37) supplemented with mannitol (0.2%) or glucose (0.2%) as carbon sources. MOPS (morpholinepropanesulfonic acid) medium supplemented with glycerol (0.4%) was prepared as described previously (39). Antibiotics were used at the following concentrations: ampicillin, 50 µg/ml; kanamycin, 50 µg/ml; tetracycline, 10 µg/ml; and chloramphenicol, 20 µg/ml. For all experiments with derivatives of strain YK3421, histidine (50 µg/ml), thymine (25 µg/ml), uracil (100 µg/ml), and cytosine (25 µg/ml) were added to the M63 medium.

                              
View this table:
[in this window]
[in a new window]
  		TABLE 1.   E. coli K-12 strains and plasmids used

Genetic methods. Random insertion mutagenesis with phage Mu dX (4) carrying the promoterless lacZ gene was performed by the procedure described by Miller (38). Phage P1 vir was used for transductions, which were carried out as described by Miller (37). The ompR234 mutation was transferred by using its genetic linkage (cotransduction at 50%) with malA or malT followed by screening of adherent transductants in 24-well microtitration plates.

Screening of lacZ fusions with altered expression within biofilms. Mutant clones obtained by transposition of Mu dX were grown in 200 µl of M63-mannitol (0.2%) medium in the wells of a 96-well microtitration plate. After 24 h of incubation at 30°C, a visible biofilm was present on the wall of each well. The medium containing the free-living bacteria of each well was removed carefully and was introduced into the corresponding well of a precooled microtitration plate, whereas attached bacteria were suspended in 200 µl of cold M63-mannitol medium by vigorous shaking of the plate. The turbidity (optical density at 630 nm [OD630]) and the initial absorbance of suspensions at 405 nm were estimated by using a microplate photometer. Bacteria were then permeabilized by the addition of 20 µl of a permeabilization reagent (48), and o-nitrophenyl-beta -D-galactoside was added to a final concentration of 2.7 mM. Quantification of beta -galactosidase activity was performed at 405 nm with a microplate photometer after various periods of incubation at room temperature. The ratio between the OD405 and OD630 was calculated for the free-living and biofilm bacteria (44).

Monitoring of lacZ fusion expression within biofilms developed on inert surfaces. The wells of 24-well microtitration plates, petri dishes, or glass tubes containing Plexiglas strips were filled with minimal M63 medium and inoculated with about 5 × 106 cells of an overnight culture of the clone of interest. After different incubation times at 30°C, the liquid medium containing free-living bacteria was carefully removed; biofilms developed on inert surfaces were washed, and sessile bacteria were suspended in cold medium by pipetting up and down. Cell biomass was estimated by OD600 measurements, and beta -galactosidase assays were performed as described below.

To be sure that no alteration of absorbance values occurred due to cell clumps remaining after resuspension and dispersal, both sessile and planktonic cell suspensions were stained with a 2-µg/ml solution of 4',6-diamidino-2-phenylindole (DAPI) and visualized by epifluorescence microscopy. No more cell clumps were observed in either situation. Furthermore, the total amounts of protein per OD600 unit were estimated, at different times, for both sessile and planktonic populations of strain PHL644 grown in M63-mannitol medium in petri dishes. Protein amounts were determined after cell lysis in the presence of 1% sodium dodecyl sulfate by using the bicinchoninic acid protein assay reagent, as recommended by the manufacturer (Pierce), with gamma globulin as a standard. Equal ratios were obtained for free-living and attached bacteria (10 independent measurements). One OD600 unit (1-cm path length, 1 ml) contains 0.16 mg of total proteins for both populations.

DNA manipulations. Standard techniques were used for chromosomal DNA preparation, plasmid extraction, gel electrophoresis, and DNA sequencing (45). Restriction endonucleases and T4 DNA ligase were used as recommended by the manufacturers.

Construction of plasmid pCP994. A 697-bp fragment containing the csgD open reading frame (ORF) was amplified by PCR with chromosomal DNA of strain MC4100 as the template and two primers (5'-CCGCCACACTGCAGCGTAAATAACG-3', and 5'-CGGGGTTTCATCATGACTAATGAAG-3') overlapping the csgD ORF and containing PstI and BspHI cutting sites, respectively. The PCR fragment was digested with PstI and BspHI and then purified and cloned in the PstI and NcoI sites of the plasmid cloning vector pKK233-2 (Apr) (2).

Identification of genes affected by Mu dX insertion. The chromosomal DNAs of mutants were extracted and digested with PstI, and fragments of more than 10 kb were cloned in pBR322 or pPH126 (51). Transformants displaying beta -galactosidase activity were analyzed. The chromosomal region adjacent to the lacZ end of the Mu genome was sequenced with a 20-bp lacZ-targeted oligonucleotide (5'-CAGGCATCAGGATTTGTGGC-3').

beta -Galactosidase assay. beta -Galactosidase specific activity in toluenized samples was measured by spectrophotometrically monitoring the hydrolysis of o-nitrophenyl-beta -D-galactoside into o-nitrophenol at 420 nm and 37°C (37) and was expressed as units per milligram of protein, where 1 U corresponds to 1 nmol of product liberated per min.

Potassium assay. Cells from 1-ml portions of suspensions of free-living and sessile bacteria grown in M63-mannitol medium in petri dishes, and containing the same number of cells (108), were harvested by centrifugation for 30 s, the supernatants were removed, and the pellets were suspended in 1 ml of deionized water. The suspensions were boiled for 1 min to release potassium from the cells, and the potassium concentrations were determined by flame photometry (55).

Preparation of conditioned medium. An actively growing culture of E. coli DH5alpha was diluted 1,000-fold in LB medium and grown for 24 h at 37°C to a final OD600 of 3.5. Cells were removed by centrifugation for 15 min at 4°C. The supernatant was sterilized by filtration through a 0.2-µm-pore-size filter and stored at -80°C. When conditioned medium was used to support growth, tryptone and yeast extract were added to concentrations of 10 and 5 g/liter, respectively (20). Fifty milliliters of conditioned medium or fresh LB medium was inoculated at a final dilution of 1/100 with a stationary-phase culture of the appropriate strain. The cultures were incubated at 30°C with vigorous shaking. The conditioned medium supported growth to a cell division rate equal to that of cells grown in fresh LB medium.

Electron microscopy. Cell cultures were grown in M63-mannitol medium at 30°C in gently shaken glass tubes, each containing one plastic strip (attached bacteria) or not containing a strip (free-living bacteria). At different incubation times between 8 and 48 h, suspensions of free-living bacteria and attached bacteria were carefully recovered to prevent breakdown of flagella and were allowed to adhere to carbon-coated 200-mesh grids. After being stained with 1% phosphotungstic acid (Sigma), the grids were examined with a Philips CM120 electron microscope.


    RESULTS
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

A major change in the pattern of gene expression occurs in E. coli biofilms. Random insertion mutagenesis with Mu dX, a bacteriophage Mu derivative carrying the promoterless lacZ reporter gene (4), was performed on the biofilm-forming E. coli K-12 strain PHL644, and, for 885 clones, the beta -galactosidase specific activities of the biofilm population and the nonattached cells were compared in microtitration plates (see Materials and Methods). A total of 446 clones were Lac+, and three classes of lacZ fusions could be distinguished. Ninety-eight fusions belong to the first class, corresponding to fusions significantly (1.5- to 10-fold) more expressed in the biofilm cells (up-regulated fusions). In the second class, 73 fusions were significantly (2- to 10-fold) less expressed in the biofilm (down-regulated fusions). Most of the fusions (275) belong to the third class, corresponding to invariant fusions. A total of 38% of the fusions, therefore, were differently expressed in biofilm. A second beta -galactosidase assay was conducted on a representative sample of 36 clones to compare the levels of expression in biofilm and free-living cells grown in 24-well microtitration plates (see Materials and Methods). The results supported those of the first analysis.

The higher osmolarity encountered in biofilm triggers major changes in gene expression. As detailed by Goodman and Marshall (22), a bacterium approaching a solid-water interface could encounter a gradient of inorganic ions and organic ionized molecules (attracted to counterbalance the electric charges existing at the solid surface), alterations in surface free energy, a lower pH level (resulting from proton accumulation), modified viscosity and osmolarity, and altered rates of gas exchange. Among these conditions, we chose to focus first on osmolarity.

It is well established that in E. coli, the intracellular content of potassium ions varies proportionally to the external osmotic pressure (12, 18). Intracellular K+ concentrations were measured for attached and planktonic PHL644 bacteria in 20 ml of liquid M63-mannitol medium introduced in petri dishes and incubated at 30°C without agitation (see Materials and Methods). After 10 h of incubation at 30°C, the intracellular potassium content in free-living bacteria was 1.81 ± 0.08 mmol/g of protein, and that in attached bacteria was 2.84 ± 0.28 mmol/g of protein (means ± standard errors for 10 repetitions; P < 0.001 by Student's t test). This corresponds to 223 ± 10 and 349 ± 34 mmol/liter for free-living and attached bacteria, respectively, where the cell volume was estimated to 1.2 µm3 for both by electron microscopy observations (more than 10 independent measurements; this result is in good agreement with data reported previously [17]). These results suggest that attached bacteria indeed encounter higher-osmolarity conditions than planktonic cells from the beginning of the colonization process.

To study the effects of osmolarity further, the expression of four genes known to be osmoregulated was compared for attached and free-living bacteria grown in petri dishes, as described in Materials and Methods. The porin ompC gene (46); the proU operon, which encodes a high-affinity glycine betaine transport system (23); and the wcaB gene (formerly called cpsB), which is involved in the synthesis of the capsular exopolysaccharide colanic acid (50) are up-regulated by high osmolarity. On the other hand, the flagellin gene fliC is down-regulated by high salt concentrations (49). For each fusion up-regulated by osmolarity, a clear induction (two- to threefold) was observed in sessile populations during the first 40 h (Fig. 1A, B, and D). The down-regulated fusion fliC is consistently less expressed in biofilm bacteria (Fig. 1C). Similar weak differences in fliC-lacZ expression in the presence of regulatory mutations conferring a flagellum-deficient phenotype have been reported (30). This suggests that flagellum synthesis is reduced in biofilms, even if the level of fliC expression is far from being totally abolished. An electron microscopic study (see Materials and Methods) revealed that biofilm cells of the motile strain PHL628 did not synthesize flagella (and produced a copious amount of curli); no flagella were visualized on sessile bacteria and no broken flagella were released in biofilm cell suspensions (Fig. 2B), whereas flagellated free-living bacteria were easy to visualize (Fig. 2A). Synthesis of curli was lower in these last cells (43).


View larger version (13K):
[in this window]
[in a new window]
  		FIG. 1.   Kinetics of osmoregulated lacZ fusion expression in attached and free-living bacteria. Specific beta -galactosidase activities of ompC-lacZ (A), proU-lacZ (B), fliC-lacZ (C), and wcaB-lacZ (D) fusions were measured in free-living (open squares) and sessile (filled squares) bacteria during biofilm development on petri dishes in M63-mannitol medium (A, B, and C) or on Plexiglas strips in M63-glucose medium (D) as described in Materials and Methods. Results are means ± standard deviations of 3 or 10 (*) independent measurements.


View larger version (63K):
[in this window]
[in a new window]
  		FIG. 2.   Electron micrographs of negatively stained bacteria of planktonic (A) or biofilm (B) cell suspensions of the motile PHL628 strain. The cells were grown in M63-mannitol medium, and biofilms were formed on plastic strips. The micrographs, corresponding to observations made after 24 h of culture, are representative of several microscopic observations performed at between 8 and 48 h of biofilm development.

To identify new genes regulated both by biofilm state and osmolarity, 40 clones isolated by the insertion mutagenesis reported above and showing biofilm regulation of their lacZ fusions were tested for osmoregulation. This screening was performed on M63-glucose agar plates, supplemented with X-Gal (5-bromo-4-chloro-3-indolyl-beta -D-galactopyranoside), in the presence or absence of NaCl (0.3 M). Differences in the intensity of the blue coloration permitted the isolation of PHL825, a clone containing a biofilm-down-regulated fusion which is repressed by a high salt concentration. A kinetic analysis of the expression of this fusion confirmed its biofilm regulation (Fig. 3A). To determine the precise extent of osmoregulation, the expression of the fusion in shaken liquid cultures containing different salt concentrations was monitored (Fig. 3B); a fourfold-lower expression was observed at high osmolarity (in the presence of 0.3 M NaCl). To identify the target gene of the Mu dX insertion, the DNA fragment surrounding the lacZ gene was cloned on a multicopy plasmid to give pCP832 and sequenced (see Materials and Methods). A BLAST search identified this gene as the uncharacterized f92 ORF (ECAE000245; min 33.5), encoding a putative short protein (92 amino acids).


View larger version (16K):
[in this window]
[in a new window]
  		FIG. 3.   Comparison of down-regulated lacZ fusion expression (strain PHL825) in attached and free-living bacteria during biofilm development on petri dishes (A) and in shaken liquid cultures under different osmolarity conditions (B). (A) Biofilm cultures were performed on petri dishes in M63-mannitol medium, and specific beta -galactosidase activities in free-living (open squares) and sessile (filled squares) bacteria were measured as described in Materials and Methods. Results are means ± standard deviations of three independent measurements. (B) Cells were grown at 30°C in low-osmolarity MOPS-glycerol medium supplemented with increasing NaCl concentrations of 0 to 0.3 M. Two independent measurements of specific beta -galactosidase activities were performed on mid-exponential-growth-phase cells. Results are means ± standard deviations of two independent measurements.

Effects of the ompR234 allele. Since wild-type E. coli K-12 strains are not able to attach to surfaces, the experiments reported above were done in the ompR234 genetic background. EnvZ-OmpR is a two-component signal transduction system involved in bacterial osmoregulation. The ompR234 allele increases the expression of csgA, the curlin-encoding gene (the resulting overproduction of curli confers the adherence properties), and of porin genes ompC and ompF (56). Although the osmoregulation mediated by EnvZ-OmpR is still efficient in ompR234 strains (56), the effect of this allele on the biofilm-regulated fusions had to be examined. The expressions of f92-lacZ, fliC-lacZ, wcaB-lacZ, and proU-lacZ in wild-type, ompR234, and ompR::Tn10 strains in planktonically grown cultures were compared (Table 2). For all of these strains, similar growth rates were observed (data not shown). The expression of proU, a gene which is osmoregulated independently of the EnvZ-OmpR system (32), was similar in the three backgrounds. The same results were observed for fliC, whereas f92 and the colanic acid gene wcaB were strongly dependent on the ompR allele. This last result indicates that these two genes belong to the OmpR regulon.

                              
View this table:
[in this window]
[in a new window]
  		TABLE 2.   Comparison of biofilm-regulated lacZ fusions in wild-type ompR, ompR234, and ompR null contextsa

To control for the possibility that the differences in gene expression observed for the attached and the planktonic bacteria do not result from an unexpected effect of the ompR234 allele, the differential expression of the two osmoregulated fusions proU-lacZ (EnvZ-OmpR independent) and wcaB-lacZ (EnvZ-OmpR dependent) was assessed in an ompR+ background. For these experiments, bacterial adherence properties were obtained by overexpression of the csgD gene (a specific activator of the csgBA operon [24]) cloned on a multicopy plasmid to give pCP994 (see Materials and Methods). ompR+ strains harboring this plasmid adhered to surfaces with the same effectiveness as ompR234 strains and formed biofilms as thick as those formed by ompR234 strains (20 µm after 24 h). As shown in Fig. 4, the same differences in expression as in the ompR234 background were observed. These results validate the approach used for identifying loci whose expression changes in planktonic versus biofilm-grown cells.


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 4.   Kinetics of proU-lacZ (A) and wcaB-lacZ (B) fusion expression in attached and free-living bacteria of adherent ompR wild-type strains (PHL1064 and PHL1036, respectively) overproducing the CsgD specific activator of the curlin-encoding csgBA operon. Specific beta -galactosidase activities were measured in free-living (open squares) and sessile (filled squares) bacteria during biofilm development on petri dishes in M63 medium supplemented with ampicillin and mannitol (A) or glucose (B), as described in Materials and Methods. Results are means ± standard deviations of three independent measurements.

Cells in biofilm also encounter limited diffusion of oxygen. Because of the particular physicochemical conditions on uncolonized surfaces and the high cellular density in established biofilms, oxygen diffusion could be an important parameter to explain differential gene expression in attached and free-living bacteria. Furthermore, in all of the experiments reported above, the microtitration and petri dishes were incubated in the absence of agitation. Due to the small total amount present in the liquid medium, the local oxygen availability could rapidly become a discriminating factor between the bacteria colonizing the bottom of the vessel and the cells swimming in the liquid phase. A nitrate assay (28) performed on strain PHL644 cultivated in liquid medium in unshaken petri dishes indicated that microaerophilic conditions were reached after 16 h (5 mM nitrite was detected in the medium). nikA encodes a high-affinity nickel transport system, and the expression of this gene is known to be precisely tuned by the level of oxygen availability (57). The nikA-lacZ fusion of strain HYD723 was introduced in an ompR234 background (to give strain PHL965), and its expressions in biofilm and planktonic bacteria under standard incubation conditions were compared. A fivefold-higher expression was observed in biofilm after 23 and 45 h (Fig. 5A). These differences were not observable in anaerobiosis (Fig. 5B), indicating that the local oxygen concentration is the main parameter responsible for the differential nikA expression under the standard conditions.


View larger version (22K):
[in this window]
[in a new window]
  		FIG. 5.   Comparison of nikA-lacZ expression in attached and free-living bacteria grown under standard conditions (A) or in anaerobiosis (B). Biofilm cultures were performed in M63-mannitol medium on petri dishes incubated in Generbox anaerobic jars (B) or not (A) for 23 and 45 h and specific beta -galactosidase activities were measured in free-living (open bars) and sessile (hatched bars) bacteria as described in Materials and Methods. Results are means ± standard deviations of three independent measurements.

To identify other genes regulated by both biofilm and anaerobiosis, the expressions of a sample of 48 lacZ fusions on solid LB medium supplemented with X-Gal during aerobiosis and anaerobiosis were compared. A clone (PHL665) carrying a biofilm-up-regulated fusion that was highly expressed in anaerobiosis was isolated. After cloning and sequencing of the chromosomal region in contact with the end of Mu dX (plasmid pCP752), a BLAST search identified the target gene as pepT (min 25.5). This gene encodes aminotripeptidase T, which is able to remove the N-terminal amino acid from various tripeptides (53). The expression of the pepT-lacZ fusion was monitored during biofilm development in petri dishes during anaerobiosis and under standard conditions (Fig. 6). A consistently higher level of expression was observed in attached bacteria from 12 to 33 h of incubation under standard conditions (Fig. 6A). This result confirms the up-regulation of pepT in sessile bacteria. A comparison of pepT expression under anaerobic and standard conditions revealed a slightly but significantly higher expression in the two cell populations during anaerobiosis (Fig. 6B). This induction of the E. coli pepT gene in anaerobiosis is in good agreement with the positive regulation by oxygen limitation of the pepT locus reported for Salmonella typhimurium (53). However, in contrast to the nikA-lacZ fusion, pepT-lacZ remained differently expressed in attached and free-living bacteria in the absence of oxygen; similar results were obtained for the f92, ompC, and proU osmoregulated fusions (Table 3). Although microaerophilic conditions seem to be one of the factors implicated in regulation of gene expression in biofilms, differences in oxygen availability are therefore not responsible for the differential expression of these four genes.


View larger version (20K):
[in this window]
[in a new window]
  		FIG. 6.   Kinetics of up-regulated lacZ fusion expression (strain PHL665) in attached and free-living bacteria (A) and comparison of fusion expression under standard conditions and during anaerobiosis after 28 h of biofilm cultures (B). Biofilm cultures were performed in M63-mannitol medium on petri dishes, and specific beta -galactosidase activities were measured in free-living bacteria (open squares and bars) and sessile bacteria (filled squares and hatched bars) as described in Materials and Methods. Results are means ± standard deviations of three independent measurements.

                              
View this table:
[in this window]
[in a new window]
  		TABLE 3.   Comparison of osmoregulated fusion expression in attached and free-living bacteria grown under standard conditions or under anaerobiosis

Cell-to-cell signalling. Since we have shown that osmolarity is a key factor in biofilm-dependent gene regulation, a possible osmoregulation of pepT was examined. Similar levels of expression of the pepT-lacZ fusion were obtained in liquid culture in MOPS-glycerol medium (126 U/mg of protein) and in MOPS-glycerol containing 0.3 M NaCl (118 U/mg of protein), indicating that pepT, in contrast to f92, ompC, fliC, wcaB, and proU, is not osmoregulated. In addition, a similar expression was observed in the ompR+, ompR234, and ompR::Tn10 backgrounds (Table 2). Kinetic analysis of the expression of the pepT-lacZ fusion during growth in liquid M63-mannitol medium gave some clues concerning its regulation. A sudden increase in beta -galactosidase activity was measured at the end of the exponential phase (Fig. 7A), probably resulting from a cell density or starvation signalling mechanism. As a cell-to-cell signalling mechanism has been shown to be involved in P. aeruginosa biofilm development (15), a quorum-sensing mechanism was investigated for pepT regulation. Conditioned medium (i.e., medium containing the suspected signalling factors) was removed from a stationary-phase culture of strain DH5alpha , as described previously (3, 20) (see Materials and Methods) and used to support growth of the pepT-lacZ strain. Overexpression of the fusion was observed at a low cell density (Fig. 7B), confirming the emission of some cell-to-cell signalling factors in the conditioned medium in response to high cell density. As the cellular density is far higher in our biofilm (1011 cells/ml are reached in a 20-µm-thick biofilm after 24 h) than in the liquid phase (109 cells/ml), we can suggest that the biofilm-dependent activation of pepT resulted from a cell-to-cell signalling mechanism.


View larger version (17K):
[in this window]
[in a new window]
  		FIG. 7.   Effects of cell density on expression of the pepT-lacZ fusion. (A) The PHL665 strain was grown in M63-mannitol medium in a shaker at 30°C. Samples were taken at intervals for determination of OD600 (open circles) and specific beta -galactosidase activity (filled circles). (B) Bacteria were grown in LB medium (open squares) or conditioned LB medium prepared as described in Materials and Methods from a stationary-phase culture of strain DH5alpha (filled squares). Growth curves in the two media were similar. This experiment is representative of three.


    DISCUSSION
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

In natural environments, bacteria are often found as sessile communities known as biofilms (8, 10). To date, the bacterial structures of adherence (25, 40-42, 56) and the physiological processes involved in bacterial surface colonization (15, 40, 41) are better understood than the genetic responses of bacteria adhering to a surface. By using a library of lacZ fusions and a reliable screen for identifying genes whose expression changes in biofilm versus planktonic cells, the transcription of 38% of the E. coli genes was shown to be modified during the colonization process. Several genes with altered expression in biofilms were identified. Different cellular functions were induced in attached bacteria: the OmpC porin, the high-affinity transport system of glycine betaine, colanic acid production (the E. coli class I exopolysaccharide), tripeptidase T, and synthesis of a nickel high-affinity transport system. On the other hand, the syntheses of flagella and of a putative protein of 92 amino acids were both reduced in biofilms. The induction of colanic acid synthesis at the gene expression level is consistent with the results of other authors, who have described a similar surface activation of exopolysaccharidic alginate genes in P. aeruginosa (13, 14, 27). These results emphasize the major role of exopolysaccharides in biofilm development, as previously reported (1, 9). Another important aspect of the biofilm formation process is triggered by flagella in different organisms, such as P. fluorescens, P. aeruginosa, and some strains of E. coli (40-42). However, in biofilms of E. coli K-12 strains overproducing curli, no flagella are produced (Fig. 2B). Furthermore, nonmotile strains (such as adherent derivatives of MC4100 overproducing curli) form thick biofilms as well as motile strains (such as adherent derivatives of MG1655 overproducing curli) do. It seems that, depending on environmental conditions and biofilm communities, different approaches and cell surface structures such as flagella, pili, S-fimbrial adhesins, autolysins, and curli could be used to initiate biofilm formation (25, 41, 42, 47, 56).

The surprising major change in gene expression observed within E. coli biofilms is consistent with the results of several teams that have shown that new protein synthesis is required for biofilm formation (40) and that patterns of proteins synthesized by attached cells versus planktonic cells are significantly different (11). Our study suggests that such a genetic reprogramming of gene expression in E. coli biofilms results from changes in multiple environmental physicochemical conditions: bacteria within biofilms encounter higher-osmolarity conditions, a lower oxygen supply, and higher cell density. Indeed, the intracellular concentration of potassium ions, which is essentially proportional to the osmolarity of the external medium in the absence of exogenous solutes such as proline or betaine (18), is 1.6-fold higher in attached bacteria than in free-living bacteria (see Results). This magnitude of change is weak but corresponds to that previously published (18; see Fig. 3 of reference 29). According to these studies, our data correspond to external osmolarities of about 220 mosM (corresponding to an intracellular potassium concentration of 223 mM) for free-living bacteria and about 380 mosM (corresponding to an intracellular potassium concentration of 349 mM) for attached cells. The osmolarity of the freshly prepared sterile M63-mannitol medium was 250 mosM (as measured with a Fiske OS/220 osmometer) (data not shown). A bacterium approaching a surface may encounter a gradient of organic and inorganic ions attracted to counterbalance the negative electric charges existing at the solid surface (22). Moreover, the copious amount of appendages present at the cell surface (such as curli [Fig. 2B]) and the various exopolymers excreted by bacteria could concentrate ionic molecules from the bulk phase as the biofilm develops.

Osmoregulated genes, whether they belong to the OmpR regulon (such as ompC, wcaB, and f92) (Table 2) or not (such as fliC, and proU) (Table 2), respond within biofilms to changes in osmolarity. Fusions up-regulated by high osmolarity (such as ompC, wcaB, and proU-lacZ) were more expressed in sessile cells, whereas fusions down-regulated by high osmolarity (such as fliC and f92-lacZ) were more expressed in planktonic cells (Fig. 1 and 3). The effect of ompR234, the particular allele used in this work, on these responses was investigated for one OmpR-independent osmoregulated gene, proU, and for one OmpR-dependent osmoregulated gene, wcaB. For these two genes, the same differential expression of free-living bacteria and attached bacteria was observed in biofilms of wild-type ompR strains overproducing curli through the overexpression of the activator CsgD (Fig. 4). Thus, an ompR234 strain is still able to detect osmotic changes and react efficiently (56); the OmpR234 protein is thought to interact more strongly with the regulatory sites of the target genes or with the RNA polymerase. This hypothesis is in good agreement with the higher repression of the wca promoter observed with OmpR234 than with OmpR (Table 2).

As the biofilm develops, oxygen consumption and emission of density signals could cause new patterns of gene expression. Using microelectrodes, Costerton and collaborators have shown that microaerophilic conditions were encountered by attached bacteria (11). Here, by using a fusion in nikA, a gene which is highly expressed in anaerobiosis and belongs to the FNR regulon (57), limitation of oxygen availability to cells within biofilms and its impact on gene expression were demonstrated (Fig. 5). A 100-fold-higher density was encountered in biofilms of ompR234 E. coli K-12 mutants than in the liquid phase. Cell-to-cell signals may therefore be accumulated and be involved in regulation of gene expression. This seems to be the case for pepT. This gene was shown to be more highly expressed in biofilm cells than in free-living bacteria and to be highly induced in the late exponential growth phase, suggesting its probable regulation by quorum sensing. In E. coli K-12 prototroph strains, such as MG1655, two types of autoinducers are produced: the first, encoded by the luxS gene (similar to the AI-2 factor produced by Vibrio harveyi), is expressed in the mid-exponential phase and is degraded when bacteria enter the stationary phase, whereas the second operates in the stationary phase (3). The first one is not produced by the domesticated laboratory strain DH5alpha , due to the presence of a frameshift mutation in the luxS gene (54). Since the pepT gene was induced in the late exponential phase, only the role of the second type of signal molecules was investigated: a conditioned medium, prepared from a stationary-phase culture of DH5alpha cells, was able to induce the expression of the pepT-lacZ fusion at a low cell density. pepT could therefore be considered a new cma (for conditioned medium activated) gene (3). Previously described cma genes were shown to be involved in amino acid metabolism; this fact is in good agreement with the function of tripeptidase T.

Biofilm gene expression patterns appear to be modulated by multiple changing external physicochemical conditions and to involve very complex regulation pathways; several bacterial sensors, many two-component signal transducing systems (such as OmpR-EnvZ), and perhaps some alternative sigma factors (such as AlgU, which is required for alginate gene expression in P. aeruginosa [16, 33]), may be suggested to coordinately regulate genes within biofilms. The analysis of additional biofilm-regulated genes isolated in the screen described in this study is in progress. It will allow us to assess the role of physicochemical factors other than osmolarity, oxygen limitation, and cell density in biofilm gene expression in E. coli K-12. Further studies of bacterial structures involved in surface sensing and biofilm formation may provide ways to move forward in the search for efficient surface coating methods able to prevent biofilm formation or, at least, to interfere with their inconvenient increased resistance to biocides.


    ACKNOWLEDGMENTS

We acknowledge Rémy Gourdon and Karim Lounis for helping us with the potassium assay, Valérie de Crécy and Philippe Bertin for gifts of strains, and the members of the Laboratoire de Génétique des Microorganismes for their interest in this work and helpful discussions.

This work was supported by grants from the French Defense Ministry (96-048/DRET) and the Centre National de la Recherche Scientifique (Réseau "Infections Nosocomiales").


    FOOTNOTES

* Corresponding author. Mailing address: Laboratoire de Génétique des Microorganismes, INSA de Lyon, 20 avenue A. Einstein, 69621 Villeurbanne, France. Phone: (33) 4 72 43 87 06. Fax: (33) 4 72 43 87 14. E-mail: lejeune{at}insa.insa-lyon.fr.

dagger Present address: Department of Animal Sciences, University of Illinois, Urbana, IL 61801.


    REFERENCES
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

1. Allison, D. G., and I. W. Sutherland. 1987. The role of exopolysaccharides in adhesion of freshwater bacteria. J. Gen. Microbiol. 133:1319-1327.
2. Amann, E., and J. Brosius. 1985. 'ATG' for regulated high-level expression of cloned genes in Escherichia coli. Gene 40:183-190[Medline].
3. Baca-DeLancey, R. R., M. M. South, X. Ding, and P. N. Rather. 1999. Escherichia coli genes regulated by cell-to-cell signalling. Proc. Natl. Acad. Sci. USA 96:4610-4614[Abstract/Free Full Text].
4. Baker, T. A., M. M. Howe, and C. A. Gross. 1983. Mu dX, a derivative of Mu d1 (lac Apr) which makes stable lacZ fusions at high temperature. J. Bacteriol. 156:970-974[Abstract/Free Full Text].
5. Belas, R. 1994. Expression of multiple flagellin-encoding genes of Proteus mirabilis. J. Bacteriol. 176:7169-7181[Abstract/Free Full Text].
6. Belas, R., M. Simon, and M. Silverman. 1986. Regulation of lateral flagella gene transcription in Vibrio parahaemolyticus. J. Bacteriol. 167:210-218[Abstract/Free Full Text].
7. Brill, J. A., C. Quinlan-Walshe, and S. Gottesman. 1988. Fine-structure mapping and identification of two regulators of capsule synthesis in Escherichia coli K-12. J. Bacteriol. 170:2599-2611[Abstract/Free Full Text].
8. Costerton, J. W., K.-J. Cheng, G. G. Geesey, T. I. Ladd, J. C. Nickel, M. Dasgupta, and T. J. Marrie. 1987. Bacterial biofilms in nature and disease. Annu. Rev. Microbiol. 41:435-464[Medline].
9. Costerton, J. W., and R. T. Irvin. 1981. The bacterial glycocalyx in nature and disease. Annu. Rev. Microbiol. 35:299-324[Medline].
10. Costerton, J. W., Z. Lewandowski, D. DeBeer, D. Caldwell, D. Korber, and G. James. 1994. Biofilms, the customized microniche. J. Bacteriol. 176:2137-2142[Free Full Text].
11. Costerton, J. W., Z. Lewandowski, D. E. Caldwell, D. R. Korber, and H. M. Lappin-Scott. 1995. Microbial biofilms. Annu. Rev. Microbiol. 49:711-745[Medline].
12. Csonka, L. N., and W. Epstein. 1996. Osmoregulation, p. 1210-1223. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. American Society for Microbiology, Washington, D.C..
13. Davies, D. G., A. M. Chakrabarty, and G. G. Geesey. 1993. Exopolysaccharide production in biofilms: substratum activation of alginate gene expression by Pseudomonas aeruginosa. Appl. Environ. Microbiol. 59:1181-1186[Abstract/Free Full Text].
14. Davies, D. G., and G. G. Geesey. 1995. Regulation of the alginate biosynthesis gene algC in Pseudomonas aeruginosa during biofilm development in continuous culture. Appl. Environ. Microbiol. 61:860-867[Abstract].
15. Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280:295-298[Abstract/Free Full Text].
16. DeVries, C. A., and D. E. Ohman. 1994. Mucoid-to-nonmucoid conversion in alginate-producing Pseudomonas aeruginosa often results from spontaneous mutations in algT, encoding a putative alternate sigma factor, and shows evidence for autoregulation. J. Bacteriol. 176:6677-6687[Abstract/Free Full Text].
17. Donachie, W. D., and A. C. Robinson. 1987. Cell division: parameter values and the process, p. 1578-1593. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C..
18. Epstein, W., and S. G. Schultz. 1965. Cation transport in Escherichia coli. V. Regulation of cation content. J. Gen. Physiol. 49:221-234[Abstract/Free Full Text].
19. Finlay, B. B., and S. Falkow. 1997. Common themes in microbial pathogenicity revisited. Microbiol. Mol. Biol. Rev. 61:136-169[Abstract].
20. Garcia-Lara, J., L. H. Shang, and L. I. Rothfield. 1996. An extracellular factor regulates expression of SdiA, a transcriptional activator of cell division genes in Escherichia coli. J. Bacteriol. 178:2742-2748[Abstract/Free Full Text].
21. Garrett, S., R. K. Taylor, and T. J. Silhavy. 1983. Isolation and characterization of chain-terminating nonsense mutations in a porin regulator gene, envZ. J. Bacteriol. 156:62-69[Abstract/Free Full Text].
22. Goodman, A. E., and K. C. Marshall. 1995. Genetic responses of bacteria at surfaces, p. 80-98. In J. W. Costerton, and H. M. Lappin-Scott (ed.), Microbial biofilms. Cambridge University Press, Cambridge, United Kingdom.
23. Gowrishankar, J. 1985. Identification of osmoresponsive genes in Escherichia coli: evidence for participation of potassium and proline transport systems in osmoregulation. J. Bacteriol. 164:434-445[Abstract/Free Full Text].
24. Hammar, M., A. Arnqvist, Z. Bian, A. Olsén, and S. Normark. 1995. Expression of two csg operons is required for production of fibronectin- and Congo red-binding curli polymers in Escherichia coli K-12. Mol. Microbiol. 18:661-670[Medline].
25. Heilmann, C., M. Hussain, G. Peters, and F. Götz. 1997. Evidence for autolysin-mediated primary attachment of Staphylococcus epidermidis to a polystyrene surface. Mol. Microbiol. 24:1013-1024[Medline].
26. Hoyle, B. D., and J. W. Costerton. 1991. Bacterial resistance to antibiotics: the role of biofilms. Prog. Drug Res. 37:91-105[Medline].
27. Hoyle, B. D., L. J. Williams, and J. W. Costerton. 1993. Production of mucoid exopolysaccharide during development of Pseudomonas aeruginosa biofilms. Infect. Immun. 61:777-780[Abstract/Free Full Text].
28. Ingledew, W. J., and R. K. Poole. 1984. The respiratory chains of Escherichia coli. Microbiol. Rev. 48:222-271[Free Full Text].
29. Ingraham, J. 1987. Effect of temperature, pH, water activity, and pressure on growth, p. 1543-1554. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C..
30. Komeda, Y. 1982. Fusions of flagellar operons to lactose genes on a Mu lac bacteriophage. J. Bacteriol. 150:16-26[Abstract/Free Full Text].
31. Komeda, Y., and T. Iino. 1979. Regulation of expression of the flagellin gene (hag) in Escherichia coli K-12: analysis of hag-lac gene fusions. J. Bacteriol. 139:721-729[Abstract/Free Full Text].
32. Lucht, J. M., and E. Bremer. 1994. Adaptation of Escherichia coli to high osmolarity environments: osmoregulation of the high-affinity glycine betaine transport system ProU. FEMS Microbiol. Rev. 14:3-20[Medline].
33. Martin, D. W., B. W. Holloway, and V. Deretic. 1993. Characterization of a locus determining the mucoid status of Pseudomonas aeruginosa: AlgU shows sequence similarities with a Bacillus sigma factor. J. Bacteriol. 175:1153-1164[Abstract/Free Full Text].
34. May, G., E. Faatz, M. Villarejo, and E. Bremer. 1986. Binding protein dependent transport of glycine betaine and its osmotic regulation in Escherichia coli K-12. Mol. Gen. Genet. 205:225-233[Medline].
35. McCarter, L., M. Hilmen, and M. Silverman. 1988. Flagellar dynamometer controls swarmer cell differentiation of Vibrio parahaemolyticus. Cell 54:345-351[Medline].
36. McLean, R. J. C., M. Whitely, D. J. Stickler, and W. C. Fuqua. 1997. Evidence of autoinducer activity in naturally occuring biofilms. FEMS Microbiol. Lett. 154:259-263[Medline].
37. Miller, J. H. 1972. Experiments in molecular genetics, p. 352-355. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
38. Miller, J. H. 1992. A short course in bacterial genetics, p. 412-416. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
39. Neidhardt, F. C., P. L. Bloch, and D. F. Smith. 1974. Culture medium for enterobacteria. J. Bacteriol. 119:736-747[Abstract/Free Full Text].
40. O'Toole, G. A., and R. Kolter. 1998. Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol. Microbiol. 28:449-461[Medline].
41. O'Toole, G. A., and R. Kolter. 1998. Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol. Microbiol. 30:295-304[Medline].
42. Pratt, L. A., and R. Kolter. 1998. Genetic analysis of Escherichia coli biofilm formation: roles of flagella, motility, chemotaxis and type I pili. Mol. Microbiol. 30:285-293[Medline].
43. Prigent-Combaret, C., G. Prensier, T. T. Le Thi, C. Dorel, O. Vidal, and P. Lejeune. Unpublished data.
44. Prigent-Combaret, C., and P. Lejeune. Monitoring gene expression in biofilms. Methods Enzymol., in press.
45. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
46. Sarma, V., and P. Reeves. 1977. Genetic locus (ompB) affecting a major outer-membrane protein in Escherichia coli K-12. J. Bacteriol. 132:23-27[Abstract/Free Full Text].
47. Schmoll, T., M. Ott, B. Oudega, and J. Hacker. 1990. Use of a wild-type gene fusion to determine the influence of environmental conditions on expression of the S fimbrial adhesin in an Escherichia coli pathogen. J. Bacteriol. 172:5103-5111[Abstract/Free Full Text].
48. Schupp, J. M., S. E. Travis, L. B. Price, R. F. Shand, and P. Keim. 1995. Rapid bacterial permeabilization reagent useful for enzyme assays. BioTechniques 19:18-20[Medline].
49. Shi, W., C. Li, C. J. Louise, and J. Adler. 1993. Mechanism of adverse conditions causing lack of flagella in Escherichia coli. J. Bacteriol. 175:2236-2240[Abstract/Free Full Text].
50. Sledjeski, D. D., and S. Gottesman. 1996. Osmotic shock induction of capsule synthesis in Escherichia coli K-12. J. Bacteriol. 178:1204-1206[Abstract/Free Full Text].
51. Spratt, B. G., P. J. Hedge, S. te Heesen, A. Edelman, and J. K. Broome-Smith. 1986. Kanamycin-resistant vectors that are analogues of plasmids pUC8, pUC9, pEMBL8 and pEMBL9. Gene 41:337-342[Medline].
52. Stickler, D. J., N. A. Morris, R. J. C. McLean, and C. Fuqua. 1998. Biofilms on indwelling urethral catheters produce quorum-sensing molecules in situ and in vitro. Appl. Environ. Microbiol. 64:3486-3490[Abstract/Free Full Text].
53. Strauch, K. L., J. B. Lenk, B. L. Gamble, and C. G. Miller. 1985. Oxygen regulation in Salmonella typhimurium. J. Bacteriol. 161:673-680[Abstract/Free Full Text].
54. Surette, M. G., M. B. Miller, and B. L. Bassler. 1999. Quorum sensing in Escherichia coli, Salmonella typhimurium, and Vibrio harveyi: a new family of genes responsible for autoinducer production. Proc. Natl. Acad. Sci. USA 96:1639-1644[Abstract/Free Full Text].
55. Sutherland, L., J. Cairney, M. J. Elmore, I. R. Booth, and C. F. Higgins. 1986. Osmotic regulation of transcription: induction of the proU betaine transport gene is dependent on accumulation of intracellular potassium. J. Bacteriol. 168:805-814[Abstract/Free Full Text].
56. Vidal, O., R. Longin, C. Prigent-Combaret, C. Dorel, M. Hooreman, and P. Lejeune. 1998. Isolation of an Escherichia coli K-12 mutant strain able to form biofilms on inert surfaces: involvement of a new ompR allele that increases curli expression. J. Bacteriol. 180:2442-244[Abstract/Free Full Text].
57. Wu, L. F., and M. A. Mandrand-Berthelot. 1986. Genetic and physiological characterization of new Escherichia coli mutants impaired in hydrogenase activity. Biochimie 68:167-179[Medline].

Journal of Bacteriology, October 1999, p. 5993-6002, Vol. 181, No. 19
0021-9193/99/$04.00+0
Copyright © 1999, American Society for Microbiology. All rights reserved.


This article has been cited by other articles: