Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Microbiology (2003), 149, 3473–3484 DOI 10.1099/mic.0.26541-0 Attachment to and biofilm formation on abiotic surfaces by Acinetobacter baumannii: involvement of a novel chaperone-usher pili assembly system Andrew P. Tomaras,1 Caleb W. Dorsey,13 Richard E. Edelmann2 and Luis A. Actis1 Departments of Microbiology1 and Botany2, Miami University, 40 Pearson Hall, Oxford, OH 45056, USA Correspondence Luis A. Actis [email protected] Received 4 June 2003 Revised 27 August 2003 Accepted 5 September 2003 Acinetobacter baumannii causes severe infections in compromised patients, survives on abiotic surfaces in hospital environments and colonizes different medical devices. In this study the analysis of the processes involved in surface attachment and biofilm formation by the prototype strain 19606 was initiated. This strain attaches to and forms biofilm structures on plastic and glass surfaces, particularly at the liquid–air interface of cultures incubated stagnantly. The cell aggregates, which contain cell stacks separated by water channels, formed under different culture conditions and were significantly enhanced under iron limitation. Electron and fluorescence microscopy showed that pili and exopolysaccharides are part of the cell aggregates formed by this strain. Electron microscopy of two insertion derivatives deficient in attachment and biofilm formation revealed the disappearance of pili-like structures and DNA sequencing analysis showed that the transposon insertions interrupted genes with the highest similarity to hypothetical genes found in Pseudomonas aeruginosa, Pseudomonas putida and Vibrio parahaemolyticus. Although the products of these genes, which have been named csuC and csuE, have no known functions, they are located within a polycistronic operon that includes four other genes, two of which encode proteins related to chaperones and ushers involved in pili assembly in other bacteria. Introduction of a copy of the csuE parental gene restored the adherence phenotype and the presence of pili on the cell surface of the csuE mutant, but not that of the csuC derivative. These results demonstrate that the expression of a chaperone-usher secretion system, some of whose components appear to be acquired from unrelated sources, is required for pili formation and the concomitant attachment to plastic surfaces and the ensuing formation of biofilms by A. baumannii cells. INTRODUCTION Members of the genus Acinetobacter are non-motile, ubiquitous Gram-negative bacteria that can be recovered from a wide range of sources such as soil, water, food products and medical environments (Bergogne-Berenzin & Towner, 1996). The latter source is of particular significance because A. baumannii is the Acinetobacter genomic species of greatest importance in human medicine. A large number of reports describe outbreaks of nosocomial infections that 3Present address: Department of Medical Microbiology and Immunology, Texas A&M University Health Science Center, College Station, TX, USA. Abbreviations: DIP, 2,29-dipyridyl; EDDHA, ethylenediamine-di-(ohydroxyphenyl) acetic acid; GFP, green fluorescent protein; SEM, scanning electron microscopy; TEM, transmission electron microscopy. The GenBank accession number for the sequence reported in this paper is AY241696. 0002-6541 G 2003 SGM include urinary tract infections, secondary meningitis, wound and burn infections, and particularly nosocomial pneumonia (Bergogne-Berenzin et al., 1993, 1996). Some of the challenges in the prevention and treatment of the infections caused by this opportunistic pathogen are its remarkable widespread resistance to different antibiotics and its ability to persist in nosocomial environments and medical devices (Bergogne-Berenzin & Towner, 1996). A. baumannii survives for several days on inanimate objects and surfaces found normally in medical environments, even in dry conditions on dust particles. These survival properties most likely play a significant role in the outbreaks caused by this pathogen. The potential ability of A. baumannii to form biofilms could explain its outstanding antibiotic resistance and survival properties. This possibility is supported by a very limited number of publications which showed that a clinical isolate of this bacterium is able to attach to and form biofilm Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Printed in Great Britain 3473 A. P. Tomaras and others structures on glass surfaces (Vidal et al., 1996, 1997). Bacterial biofilms, arrangements in which the cells are morphologically, metabolically and physiologically different from their planktonic counterparts (Stoodley et al., 2002), have been found on the surface of medical devices such as intubation tubes, catheters, artificial heart valves, water lines and cleaning instruments (Donlan & Costerton, 2002). The surfaces of all of these medical and dental devices are normal targets for colonization by complex microbial communities. Previous research efforts have shown that biofilm formation proceeds via a series of steps which, upon completion, produce a mature, three-dimensional structure on both biotic and abiotic surfaces. Some of the current working models for biofilm formation implicate the participation of either bacterial surface motility mediated by pili and flagella (O’Toole & Kolter, 1998a), the flagellummediated recruitment of planktonic cells by the developing biofilm from the liquid medium (Tolker-Nielsen et al., 2000) or the formation and growth of microcolonies formed as a consequence of the multiplication of cells attached to solid surfaces (Heydorn et al., 2000). While forming and establishing these multicellular structures, the cells composing them secrete exopolysaccharides which serve to fortify and maintain the structure of the biofilm. It is not well understood whether these steps and cell components are involved in the apparent ability of A. baumannii to form biofilms on abiotic surfaces. Furthermore, the mechanism by which this bacterium forms biofilms may pose a challenge because of its wellestablished non-motile phenotype (Bergogne-Berenzin & Towner, 1996). In this work, we report the initial characterization of the biofilm structures formed by the A. baumannii 19606 prototype strain under different culture conditions. In addition, we have initiated the genetic and molecular analysis of some of the factors that are involved in this important cellular process that could explain the resistance and survival of this opportunistic pathogen under harsh conditions such as those found in patients and nosocomial environments. METHODS Bacterial strains, plasmids and culture conditions. The bacter- ial strains and plasmids used in this work are listed in Table 1. Luria–Bertani (LB) broth and agar (Sambrook et al., 1989) were used to maintain all bacterial strains. M9 minimal medium (Miller, 1972) was used for sliding agar motility assays and growth under chemically defined conditions. The effect of iron on biofilm formation was tested by using M9 minimal medium supplemented with 200 mM FeCl3, 200 mM ethylenediamine-di-(o-hydroxyphenyl) acetic acid (EDDHA) or 200 mM 2,29-dipyridyl (DIP) to simulate ironreplete and two iron-limited conditions, respectively. The effect of different carbon sources on biofilm production was examined through the addition of either 10 mM sodium citrate, 25 mM sodium succinate, 50 mM ethanol, 0?5 % pyruvate, 0?5 % acetate or 0?5 % lactate to Simmons salts. The latter consisted of 0?2 g MgSO4 l21, 1?0 g NH4H2PO4 l21, 1?0 g K2HPO4 l21 and 5?0 g NaCl g l21, adjusted to pH 7?0. Transmission electron microscopy (TEM) experiments were done using cells lifted from agar plates incubated at 37 uC. Scanning electron microscopy (SEM) and light microscopy experiments were done using cells cultured in LB broth. Swimming, swarming, sliding and twitching agar plates (Rashid & Kornberg, 2000) were used to test different types of cell motility. Plates were stab-inoculated through to the bottom with a sterile toothpick and incubated at 37 uC for 16 h. Table 1. Bacterial strains and plasmids used in this work Strain/plasmid Strains A. baumannii 19606 19606 #37 Derivative of 19606 19606 #144 Derivative of 19606 E. coli DH5a Top10 EC100D Plasmids pCR-Blunt II-TOPO pUC4K pWH1266 pMU125 pMU243 pMU348 pMU349 Relevant characteristics* Source/reference Clinical isolate, prototype strain csuC : : EZ : : TN <R6Kcori/KAN-2> Bouvet & Grimont (1986) This work csuE : : EZ : : TN <R6Kcori/KAN-2> This work Used for recombinant DNA methods Used for recombinant DNA methods pir+; used for plasmid rescue Gibco-BRL Invitrogen Epicentre Used to clone amplicons Source of the Kmr gene Acinetobacter lwoffii plasmid cloned into pBR322 PvuII site; Apr Tcr pWH1266 harbouring gfp, Apr pWH1266 harbouring csuE, Apr Plasmid rescued by self-ligation of NdeI-digested DNA from mutant #144 Plasmid rescued by self ligation of EcoRI-digested DNA from mutant #144 Invitrogen PharmaciaBiotech Hunger et al. (1990) Dorsey et al. (2002) This work This work This work *Apr, ampicillin resistance; Kmr, kanamycin resistance; Tcr, tetracycline resistance. 3474 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Microbiology 149 Acinetobacter baumannii biofilms General DNA procedures. Total DNA was isolated either by ultra- centrifugation in CsCl density gradients (Meade et al., 1982) or using a mini-scale method adapted from previously published research (Barcak et al., 1991). Plasmid DNA was isolated using commercial kits (Qiagen). DNA digestions were performed with restriction enzymes as indicated by the supplier (New England Biolabs) and size-fractionated by agarose gel electrophoresis (Sambrook et al., 1989). Southern blot analyses were conducted using standard protocols (Sambrook et al., 1989) using high-stringency conditions (Graber et al., 1998). The pUC4K aph gene, which encodes kanamycin resistance and was used as a probe to detect the insertion of the EZ : : TN <R6Kcori/KAN-2> transposon, was PCR-amplified with Pfu DNA polymerase (Stratagene) and the primers 133 (59-GGCGCTGAGGTCTGCCTCCTG-39) and 134 (59-GAGCCATATTCAACGGG-39). The amplicon was purified using the GeneClean II kit (QBiogene) and labelled with [a-32P]dCTP (Feinberg & Vogelstein, 1983). The radioactive bands were detected with a Storm 860 scanner (Molecular Dynamics). Transcriptional analysis of gene expression. Expression of polycistronic genes was tested by RT-PCR analysis using total RNA isolated from bacteria grown in LB broth as described by Wu & Janssen (1996). The RNA samples were treated with RNase-free DNase I (Roche) and used with an RT-PCR commercial kit (Qiagen), under the conditions suggested by the manufacturer. The amplicons were analysed by agarose gel electrophoresis (Sambrook et al., 1989). PCR of total RNA without reverse transcription was used to test for DNA contamination of RNA samples. The nature of the amplicons was confirmed by automated DNA sequencing. Biofilm assays. One millilitre of fresh medium in borosilicate (156125 mm), polystyrene (12675 mm) or polypropylene (126 75 mm) sterile tubes was inoculated with 0?01 ml of an overnight culture. Duplicate cultures for each sample were incubated for 8 h either shaking (at 200 r.p.m. in an orbital shaker) or stagnant at 37 uC. One tube was sonicated immediately for 5 s with a thin probe and the OD600 of the culture was determined to estimate total cell biomass. The supernatant of the other tube was aspirated and rinsed thoroughly with distilled water. The cells attached to the tube walls were visualized and quantified by staining with crystal violet and solubilization with ethanol–acetone as described by O’Toole et al. (1999). The OD580/OD600 ratio was used to normalize the amount of biofilm formed to the total cell content of each sample tested to avoid variations due to differences in bacterial growth under different experimental conditions. All assays were done at least twice using fresh samples each time. In addition, every test was done in duplicate each time. Square (1006100 mm) polystyrene Petri dishes were also used to detect and characterize the structures formed by A. baumannii cells on plastic surfaces. Microscopy experiments. Cells attached to the bottom and the side of square plates were visualized with regular (brightfield) light microscopy after staining with crystal violet as described by O’Toole et al. (1999). A. baumannii 19606 cells harbouring pMU125 and expressing the green fluorescent protein (GFP) were visualized by epifluorescence microscopy using a Nikon Eclipse E400 microscope equipped with a SPOT digital camera (Diagnostic Instruments). The same equipment was used to visualize samples stained with calcofluor white as described by Neu et al. (2002). Colonies grown overnight on Tris-M9 agar were sampled with Formvar-coated gold grids, negative-stained with 1?5 % ammonium molybdate and visualized using a Zeiss EM 10C transmission electron microscope. For SEM, the cells were cultured in square Petri dishes without shaking at 37 uC either overnight or for up to 10 days replacing the culture medium daily. Upon removal of the culture medium, the plates were immediately flooded with 2?5 % gluteraldehyde in 0?05 M sodium cacodylate and incubated at room temperature for 2 h. The fixative was removed and replaced immediately with distilled water to prevent sample dehydration. Small samples (161 cm) were cut http://mic.sgmjournals.org from the sides of each plate, rinsed with distilled water, dehydrated with increasing concentrations of ethanol, ranging from 25 to 100 %, and CO2 critical point dried. Samples were gold-coated and visualized with a JEOL T200 scanning electron microscope. Transposition mutagenesis and rescue of interrupted genes. A. baumannii 19606 was mutagenized using the EZ : : TN <R6Kcori/ KAN-2> Tnp Transposome transposition mutagenesis system and electroporation as described by Dorsey et al. (2002). Transformants that grew after plating on LB agar containing 40 mg kanamycin ml21 were toothpicked into separate wells of 96-well polystyrene microtitre plates, each containing 200 ml LB broth. Plates were incubated stagnant at 37 uC for 24 h, the culture supernatant was aspirated and the wells were rinsed with water and stained with crystal violet. Potential biofilm mutants were validated using tube and Petri dish assays as described above. The genomic regions harbouring the insertion of the EZ : : TN <R6Kcori/KAN-2> transposon were rescued by self-ligation of EcoRI- or NdeI-digested DNA and electroporation into Escherichia coli EC100D pir+ cells. Plasmid DNA isolated from colonies that grew on LB agar containing 40 mg kanamycin ml21 was used as a template to determine the nucleotide sequence of the genomic DNA flanking the transposon element with the primers complementary to this insertion element that were supplied with the mutagenesis kit. Further extension of nucleotide sequences was done using custom-designed primers. DNA sequencing was conducted using the DYEnamic ET Terminator Cycle Sequencing Kit (Amersham Pharmacia Biotech) and an ABI 3100 automated DNA sequencer (Applied Biosystems). Sequences were examined and assembled using Sequencher 4.1.2 (Gene Codes). Nucleotide and amino acid sequences were analysed with DNASTAR, BLAST and the analysis tools available through the ExPASy Molecular Biology Server (http://www.expasy.ch), such as SIGNALP, PSORT, HMMTOP and TMHMM to predict the presence of signal peptides and the cellular locations of proteins. G+C content was determined with Artemis (http://www.sanger.ac.uk), using a 120 nt window. Genetic complementation of a biofilm mutant. The A. bau- mannii 19606 csuE-like gene, whose disruption by the insertion of EZ : : TN <R6Kcori/KAN-2> caused a biofilm-deficient phenotype in the derivative #144, was PCR-amplified from the parental strain genome with Pfu DNA polymerase and the primers 1671 (59-CGGGATCCCGTATTGCTGCTAAACGTGGCTCGGGTGTTGTG-39) and 1672 (39-CGGGATCCCGGCTATAGAGCTTATGCAAAACTCATGCAGTGCC-59) that included BamHI restriction sites. The blunt-ended amplicon was ligated into pCR-Blunt II-TOPO and transformed into E. coli Top10. Plasmid DNA, which was isolated from a kanamycin-resistant colony and proved to have an insert with the same nucleotide sequence as the parental DNA, was digested with BamHI, ligated into the cognate site of the shuttle vector pWH1266 and transformed into E. coli DH5a. Plasmid DNA was isolated from a colony that was resistant to 100 mg ampicillin ml21 and sensitive to 20 mg tetracycline ml21 and electroporated into A. baumannii 19606 #37 and #144 cells. Transformants that grew after overnight incubation at 37 uC on LB agar containing 100 mg ampicillin ml21 were tested for their ability to form biofilms on plastic surfaces as described above. The presence of the complementing plasmid was verified by Southern blot analysis using csuE amplified from the parental strain as a probe. RESULTS Attachment to and formation of biofilms on abiotic surfaces The initial assays showed that A. baumannii 19606 cells which were taken from an agar plate with a bacteriological Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 3475 A. P. Tomaras and others (b) 2 1 2 OD580/OD600 (a) 1 (a) 3 (b) 6 2 4 1 2 (c) 20 16 12 8 4 0 Fig. 1. Detection of biofilms formed on plastic surfaces. E. coli DH5a (sample 1) and A. baumannii 19606 (sample 2) were incubated either in polystyrene tubes (a) or square Petri dishes (b) overnight at 37 6C without shaking in LB broth. Attached cells were detected by staining with crystal violet after washing with tap water. loop, deposited on the surface of a Petri dish or a Teflon strip and left for 10–15 min at room temperature remained tightly adhered after washing with tap water and staining with crystal violet. Incubation of A. baumannii 19606 cells overnight in LB broth without shaking at 37 uC either in polystyrene tubes or square Petri dishes also resulted in their adherence to these surfaces. In contrast, no adherence was detected with E. coli DH5a cells cultured under similar conditions (compare tubes and plates shown in Fig. 1a and b). These observations were confirmed further when cell attachment to plastic was quantified by relating the total cell mass to the stain retained by the attached cells as described in Methods. It is interesting to note that the A. baumannii 19606 cells form denser aggregates at the liquid–air interface than on the side and bottom surfaces of the tubes and plates. Furthermore, the biofilm grows upwards from the liquid–air interface onto the walls of the plate (Fig. 1b), a phenomenon that is not caused by the movement of the broth, since the inoculated plate was incubated without any movement until stained with crystal violet. Quantitative analysis showed that while the amount of A. baumannii 19606 cells attached to polystyrene and polypropylene is similar, their attachment to borosilicate is significantly reduced when tested under the same experimental conditions (Fig. 2a). Culture shaking also reduced the amount of cells attached to these three abiotic surfaces when compared with cultures incubated stagnantly (Fig. 2a). The values shown in Fig. 2(a), particularly those obtained with glass tubes, do not represent background values since crystal violet does not attach by itself to the surface of borosilicate. A. baumannii 19606 attaches to polystyrene tubes when cultured stagnantly at different temperatures, although this process is more efficient at 30 uC than at 37 uC (Fig. 2b). The addition of inorganic iron to M9 minimal medium reduced the amount of attached cells, while the supplementation of this chemically defined medium with the iron chelators EDDHA or DIP 3476 1 2 0 0 3 1 2 Experimental conditions 1 2 3 4 Fig. 2. Quantification of A. baumannii 19606 biofilms formed under different culture conditions on plastic and glass surfaces. (a) Cells cultured in LB broth in polystyrene (1), polypropylene (2) or borosilicate (3) tubes with (open bars) or without (closed bars) shaking overnight at 37 6C. (b) Cells cultured in LB broth in polystyrene tubes without shaking either at 30 (1) or 37 6C (2). (c) Cells cultured in Tris-M9 minimal medium in polystyrene with no supplementation (1) or supplemented with 200 mM FeCl3 (2), 200 mM EDDHA (3) or 200 mM DIP (4) at 37 6C overnight. OD600 was determined after the cultures were sonicated briefly to resuspend most of the cells to determine total cell mass. OD580 was determined after the stained tubes were incubated with ethanol-acetone. increased cell attachment significantly (Fig. 2c). In contrast, no significant effects were observed when the cells were cultured in a chemically defined medium supplemented with different carbon sources such as citrate, succinate, ethanol, pyruvate, acetate or lactate (data not shown). Microscopy analysis of cells attached to plastic surfaces Light microscopy examination of A. baumannii 19606 cells attached to the bottom surface of a Petri dish and stained with crystal violet showed the presence of discrete cell clumps (Fig. 3a), which did not coalesce to form at least a monolayer even after several days of incubation with or without renewing the culture medium. In contrast, compact cell aggregates were seen on the sides of the plates at the liquid–air interface. Similar results were obtained when A. baumannii 19606 cells producing GFP encoded by the recombinant plasmid pMU125 were examined with epifluorescence microscopy (Fig. 3b). This figure also shows that the fluorescent cellular aggregates are interspersed with areas devoid of cells, an arrangement that is very similar to those described in other bacteria capable of forming biofilms on abiotic surfaces, such as Pseudomonas aeruginosa (Lawrence et al., 1991). Fluorescence microscopy of samples stained with calcofluor white showed that the cells attached either to the walls or the bottom of the plates were embedded within a blue fluorescent material (Fig. 3c). This non-specific UV-excitable dye that binds to exopolysaccharides such as those produced by Rhizobium meliloti Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Microbiology 149 Acinetobacter baumannii biofilms (a) (b) (c) Fig. 3. Light microscopy of A. baumannii 19606 cells attached to square Petri dishes. (a) Cells attached to the bottom (upper panel) or the side (lower panel) were stained with crystal violet. (b) Cells harbouring the plasmid pMU125 attached either to the bottom (upper panel) or the side (lower panel) were examined with epifluorescence microscopy. (c) Cells attached to the bottom (upper panel) or the side (lower panel) were stained with calcofluor white and examined with epifluorescence microscopy. All cells were cultured in LB broth without shaking overnight at 37 6C. (Leigh et al., 1985) was recently used to study biofilms formed by Salmonella enteritidis (Solano et al., 2002). These reports together with our data support the hypothesis that A. baumannii 19606 produces exopolysaccharides that are part of the cellular structures formed by this bacterium when it grows attached to a plastic surface. SEM analysis of samples similar to those used for light microscopy showed that while few cells were clustered together on the sides of the plate below the meniscus of the liquid medium, compact cell conglomerates separated by channel-like spaces were formed at the meniscus of the broth (compare Fig. 4b and c). Much denser and tighter conglomerates were formed by cells located just above the meniscus, which appear to form multilayer structures, with the top cell layer covered with a film that most likely represents the exopolysaccharides produced by this bacterium (Fig. 4a). Alternatively, this film could also have been the result of the dehydration of the cell aggregates and the exopolysaccharides that surround them, which are located immediately above the liquid–air interface, during incubation. http://mic.sgmjournals.org SEM also showed that the cells were linked to each other through extracellular appendages that resemble pili structures (see insets of Fig. 4a and b). The presence of such a type of appendage was investigated by TEM of cells cultured on agar plates and stained with ammonium molybdate. This approach showed the presence of structures that resemble pili projecting from the surface of the cells (Fig. 5a), which were found to be fairly equally distributed on the cell surfaces without any particular organization. Cell motility and the presence of cellular appendages have been implicated in the formation of biofilms by other bacteria (O’Toole & Kolter, 1998a; Tolker-Nielsen et al., 2000). However, members of the Acinetobacter genus, including A. baumannii, are taxonomically defined to lack flagella and thus are non-motile (Bergogne-Berenzin & Towner, 1996). To confirm this inability to move, we tested 19606 cells by inoculating defined solid media used to detect swimming, swarming, sliding and twitching motility as described by Rashid & Kornberg (2000). These assays proved that A. baumannii 19606 lacked the ability to move Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 3477 A. P. Tomaras and others Fig. 4. SEM of A. baumannii 19606 cells attached to square Petri dishes. Attachment of the parental strain (left) and the insertion derivative #144 harbouring plasmid pMU243 (right) above (a and d), at (b and e) and below (c and f) the liquid–air interface. Bars: (a, c, d–f) 5 mm; (b) 10 mm; all insets, 1 mm. via these mechanisms under the experimental conditions used in this study. Genetic analysis of cell attachment and biofilm formation analysis of the side of the plates in which the mutant #144 derivative was incubated showed the presence of very few cell clusters, with no more than two or three cells grouped together that did not display pili on their surfaces (data not shown). The EZ : : TN <R6Kcori/KAN-2> Tnp Transposome system from Epicentre was used to initiate the identification and characterization of some of the genetic determinants required by A. baumannii 19606 to attach to and form biofilms on plastic surfaces. As we reported recently (Dorsey et al., 2002), screening of insertion derivatives of this strain resulted in the identification of mutants affected in cell attachment and biofilm formation without affecting their growth in LB broth. The rescue cloning and initial nucleotide sequence analysis of the mutant #144 resulted in the identification of a gene encoding a protein highly similar to that encoded by the Vibrio parahaemolyticus csuE gene. TEM analysis of mutant #144 proved that the disruption of the A. baumannii 19606 csuE-like gene resulted in the disappearance of the pili detected in the parental strain (compare Fig. 5a and b). Although not shown, SEM The role of the A. baumannii 19606 csuE-like gene was confirmed further by genetic complementation of the insertion derivative with the parental gene PCR-amplified and cloned in the shuttle vector pWH1266 (Hunger et al., 1990). Electroporation of the recombinant plasmid pMU243 into the A. baumannii 19606 #144 insertion derivative restored its ability to attach and form biofilms in a fashion similar to that displayed by the 19606 parental strain (sample 2 in Fig. 1a). In contrast, the electroporation of pWH1266 did not change the phenotype of the biofilmdeficient derivative, which produced an image identical to that displayed by sample 1 in Fig. 1(a). SEM and TEM analyses also showed that the introduction of the parental csuE-like gene restored the ability of A. baumannii 19606 #144 to form structures similar to those formed by the parental strain (Fig. 4d–f) and the concomitant presence of 3478 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Microbiology 149 Acinetobacter baumannii biofilms total DNA isolated from this biofilm mutant, showed that the A. baumannii 19606 csuE-like gene is the last component of a gene cluster that encompasses six ORFs (Fig. 6a). The mean G+C content of this region is 37?5 mol%; however, ORF 2 (28?5 mol%) and 3 (29?7 mol%) displayed a significantly lower value than the mean (Fig. 6b and Table 2). This observation suggests that these two ORFs were acquired from a source different from that of the remaining ORFs that compose this apparently polycistronic locus, whose G+C content is closer to the 40–43 mol% values assigned to different A. baumannii isolates (Bouvet & Grimont, 1986). The polycistronic nature of this locus was confirmed by RT-PCR analysis, which produced amplicons with the predicted sizes (Fig. 7), and nucleotide sequences when the appropriate pairs of primers were used. None of these amplicons could be detected when total RNA was used as a template for PCR. Fig. 5. TEM of A. baumannii 19606 cells cultured on agar and stained with ammonium molybdate. (a) 19606 parental strain. (b) Insertion mutant #144. (c) Insertion mutant #144 harbouring the plasmid pMU243. Bars, 100 nm. pili on the surface of the mutant cells (Fig. 5c). Southern blot analysis of total DNA isolated from A. baumannii 19606 #144 and the complemented mutant probed with the csuElike gene validated the attachment and biofilm results by confirming the presence of the complementing plasmid pMU243 as an independent replicon without detectable rearrangements (data not shown). Sequence analysis of plasmid DNA rescued from mutant #144 Sequence analysis of plasmids pMU348 and pMU349, which were rescued by self-ligation of NdeI- and EcoRI-digested http://mic.sgmjournals.org The first ORF encodes a potential protein with a predicted 25 aa signal peptide, which showed the highest similarity to the P. aeruginosa hypothetical protein PA4648 (Stover et al., 2000) and the Pseudomonas putida PP2358–PP2360 hypothetical proteins (Nelson et al., 2002) (Table 2). All these proteins contain the conserved domain COG5430, which has been identified in other hypothetical proteins produced by bacteria such as P. aeruginosa, Ralstonia solanacearum and Yersinia pestis. These proteins were annotated as potential secreted proteins related to the type 1 pili subunit CsuA and CsuB proteins. Significant homology was also detected with the Myxococcus xanthus spore protein U, which is produced as a secretory precursor and then secreted across the membrane to be assembled on the spore surface (Gollop et al., 1990). Interestingly, the protein U and other proteins annotated as protein U-like proteins also showed significant similarity to the conserved domain COG5430. This similarity may reflect the fact that most of the proteins that belong to this family have been annotated as secreted proteins and therefore may use similar secretion mechanisms. The product of ORF 2, which is predicted to have a 31 aa signal peptide, displayed significant similarity only to the hypothetical V. parahaemolyticus CsuA protein (Table 2). The predicted product of ORF 3 is a protein with a potential 20 aa signal sequence that is related to the proteins PP2360, PA4648 and CsuB from P. putida, P. aeruginosa and V. parahaemolyticus, respectively (Table 2). Interestingly, all these proteins contain a sequence highly related to the conserved domain COG5430 that has been found in proteins annotated as type 1 pili subunit CsuA/B or CsuB in different Gramnegative bacteria, as well as sequences related to the spore protein U. Further analysis using GAP showed that the identity of the A. baumannii 19606 CsuA/B, CsuA and CsuB proteins ranged from 23?7 to 27?7 %, while their similarity was between 34?1 and 36?6 %. The predicted product of the fourth ORF is highly related to the hypothetical PP2361, CsuC and PA4651 proteins of P. putida, V. parahaemolyticus and P. aeruginosa, respectively Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 3479 A. P. Tomaras and others Fig. 6. Analysis of the A. baumannii 19606 DNA region harbouring the csu locus. (a) Genetic and physical map of the csu locus. The horizontal arrows and numbers indicate the location and direction of transcription of the ORFs identified by the computer analysis of nucleotide sequence of the region rescued from mutant #144 as plasmids pMU348 and pMU349. Vertical lines indicate the location of the NdeI and EcoRI restriction sites. The vertical arrows indicate the transposon insertion site in the isogenic mutants #37 and #144. The name of each ORF was assigned based on its similarity to genes described in other bacteria. (b) Percentage G+C of the csu locus determined with Artemis using a 120 nt window. (c) Alignment of csu-like ORFs present in the genomes of different bacteria. The numbers within the arrows represent the gene numbers assigned in the genome of the cognate bacterial strains. (Table 2). Analysis with BLASTP showed that all four proteins are highly related to fimbrial chaperones and contain the COG 3121.1 and Pfam 00345.6 conserved domains, which have been either described or annotated as pili assembly factors, with the FimC and PapD chaperones being some of the best characterized members of this protein family that Table 2. Sequence analysis of the ORFs located within the csu locus of A. baumannii 19606 ORF G+C (mol%) Protein (aa/kDa) Similarity* E valueD Identity/similarityd Accession no. 1 39?0 180/18?7 2 3 28?5 29?7 192/20?8 172/19?3 4 37?6 277/30?6 5 39?7 832/92?7 6 42?5 339/36?6 P. aeruginosa PA4648 P. putida PP2358 P. putida PP2360 P. putida PP2359 V. parahaemolyticus CsuA P. putida PP2360 P. aeruginosa PA4648 V. parahaemolyticus CsuB P. putida PP2361 V. parahaemolyticus CsuC P. aeruginosa PA4651 P. putida PP2362 P. aeruginosa PA4652 P. putida PP2363 V. parahaemolyticus CsuE P. aeruginosa PA4653 5e-19 2e-12 1e-10 7e-10 3e-06 4e-07 1e-06 3e-06 1e-35 2e-35 2e-32 e-144 e-143 4e-26 2e-23 6e-18 42?4/49?4 28?4/39?2 32?9/39?0 33?7/40?7 26?0/40?5 31?9/41?1 30?7/36?1 25?3/36?7 35?4/45?4 35?3/54?1 31?0/41?1 38?0/46?0 41?7/50?4 34?2/38?9 33?2/41?6 29?8/38?1 Q9HVE4 AAN67971 AAN67973 AAN67972 AF339087 AAN67973 Q9HVE4 AF339087 AAN67974 AF339087 Q9HVE1 AAN67975 Q9HVE0 AAN67976 AF339087 Q9HVD9 *Identification of bacterial products is based on BLASTP analysis. DThe E values of the highest and more significant matches obtained during the dValues were obtained with the GCG GAP program. 3480 BLASTP analyses are shown. Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Microbiology 149 Acinetobacter baumannii biofilms 1 2 3 4 5 6 kbp 2.0 0.6 Fig. 7. Agarose gel electrophoresis of the amplicons obtained by RT-PCR of total RNA extracted from A. baumannii 19606 cells. The primers used were designed to detect the expression of ORF 1–ORF 2 (lane 1), ORF 2–ORF 3 (lane 2), ORF 3– ORF 4 (lane 3), ORF 4–ORF 5 (lane 4) and ORF 5–ORF 6 (lane 5). Lane 6, HindIII-digested l DNA. are involved in the biogenesis of pili in E. coli (Hermanns et al., 2000; Sauer et al., 2000). The expression and role of this gene is supported by the phenotype of the derivative #37, which harbours a transposon insertion in csuC (Fig. 6a) and displays the same biofilm-deficient phenotype as mutant #144 (data not shown). TEM analysis showed that CsuC-deficient cells do not have pili on the cell surface, an observation that confirms the role of this protein in the assembly of these cell appendages (data not shown). The product of ORF 5 is a large protein that is highly similar to the hypothetical proteins PP2363 and PA4652 of P. putida and P. aeruginosa, respectively (Table 2). These three proteins contain sequences related to COG3188 and Pfam 00577, conserved domains that have been described in fimbrial usher proteins involved in the biogenesis of Gram-negative bacterial pili. This protein family includes the FimD and PapC usher proteins that are required for the biogenesis of pili (Klemm et al., 1985; Thanassi et al., 1998). The last ORF of this A. baumannii 19606 gene cluster, whose disruption by the insertion of the EZ : : TN <R6Kcori/KAN2> transposon (Fig. 6a) resulted in the isolation of the attachment and biofilm deficient derivative #144, encodes a predicted protein that has a 27 aa signal peptide. BLASTP analysis showed that the most related proteins are all hypothetical proteins, with the highest similarity to the PP2363, CsuE and PA4653 proteins of P. putida, V. parahaemolyticus and P. aeruginosa, respectively (Table 2). This predicted A. baumannii protein has sequences highly related to the COG5430 conserved domain that has been found in uncharacterized secreted proteins in P. aeruginosa, Y. pestis, Ralstonia solanacearum and Rickettsia conorii. Furthermore, a search of the Pfam database with MOTIFSCAN showed that the product of this ORF has a motif with some similarity to fimbrial proteins found in different bacteria. As described for the products of ORFs 1 and 3, the product of ORF 6 also has similarity with proteins harbouring the protein U domain found in the family of spore coat proteins. Fig. 6(c) shows that the A. baumannii 19606 gene cluster has the same number of genes and genetic organization as in P. aeruginosa (Stover et al., 2000) and Y. pestis (Parkhill et al., 2001), while the cognate loci in P. putida (Nelson et al., 2002) and V. parahaemolyticus (GenBank http://mic.sgmjournals.org accession no. AF339087; Makino et al., 2003) are one ORF longer and shorter, respectively. It is worthy to note that all the genes and gene products described in Table 2 are hypothetical bacterial elements described during the annotation of the cognate genomes. Therefore, neither their expression nor biological role(s) were tested experimentally as we have done with the A. baumannii 19606 csu locus. Furthermore, the P. aeruginosa PA4648–PA4653 gene cluster, which showed significant similarity to the A. baumannii 19606 csu locus, is different from the cupA1–A5 fimbrial gene cluster that specifies a chaperone-usher pathway that was found to be involved in the formation of biofilm formation by this pathogen (Vallet et al., 2001). Nevertheless, the A. baumannii 19606 csu locus showed significant similarity with the cupA cluster when compared with BLASTX. However, the similarity was only with CupA3 (E value 1e-20), the usher component of this P. aeruginosa secretory system that was annotated as the PA2130 protein. Therefore, the expression and biological role of the P. aeruginosa PA4648–PA4653 gene cluster remains to be tested. DISCUSSION Several reports have shown that Acinetobacter environmental isolates form biofilm communities composed of different bacterial species. In contrast, only two brief reports describe the ability of A. baumannii clinical isolates to attach to and form biofilms on glass surfaces (Vidal et al., 1996, 1997), a property that is most likely to be associated with the capacity of this pathogen to survive in hospital environments and medical devices, and cause severe infections in compromised patients (BergogneBerenzin & Towner, 1996). In this report we have extended these initial observations by showing that the prototype strain 19606 also attaches to and forms biofilms on the surface of different plastics (polystyrene, polypropylene and Teflon), some of which are used in the fabrication of medical devices. It is also evident that this bacterium forms biofilms under static as well as dynamic conditions, although the latter results in less biofilm when compared with samples incubated stagnantly. This property could explain the ability of this pathogen to persist successfully in medical environments where cells could be subjected to the shearing forces of a liquid stream, such as those found with catheters and respiratory tubes, or allowed to persist on less disturbed surfaces such as those of hospital furniture and bed linen (Bergogne-Berenzin & Towner, 1996). The formation of biofilms at different temperatures and the presence of extracellular material surrounding the attached cells that is stained with calcofluor white are also in accordance with the ability of this pathogen to grow within a wide range of temperatures and produce capsule and exopolysaccharides (Bergogne-Berenzin & Towner, 1996; Towner et al., 1991). It is also clear that A. baumannii 19606 attaches to and forms biofilm structures on hydrophobic (plastic) as well as hydrophilic (glass) surfaces, with Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 3481 A. P. Tomaras and others much denser cell conglomerates located at the liquid–air interface. This observation is significant because A. baumannii is a strict aerobic bacterium, a phenotype that could explain this behaviour as well as the observation that the bottom of the plates or tubes did not have a contiguous cell layer even after extended incubation. No significant effects on biofilm formation were observed when the cells were cultured in a chemically defined medium supplemented with different carbon sources. In contrast, the incubation of A. baumannii 19606 cells under iron limitation resulted in a significant increase of biofilm when compared to that obtained with cells cultured under iron-rich conditions. This expressional behaviour could be explained by the presence of the nucleotide sequence 59-TACAAATCTAATCATTATA-39 10 nt upstream of the first ORF, which matches 13 of the 19 nt of the binding site for the Fur iron repressor protein that controls the expression of ironregulated protein in bacteria (Calderwood & Mekalanos, 1988). These observations are quite significant since these bacterial cells would most likely be exposed to iron-limiting conditions either in the host or the nosocomial environment (Neilands, 1981). The light microscopy of crystal-violet-stained cells and fluorescence microscopy of GFP-expressing bacteria showed that the conglomerates located at the liquid–air interface indeed form biofilm structures similar to those described for other bacteria, with channels that would provide nutrients and remove waste products. The presence of these channels is supported further by the data obtained by SEM, which showed stacks of cells surrounded by open areas on the sides of the plates at the liquid–air interface (Fig. 4b). The cell stacks appear to be 4–5 cells tall with the cells attached to each other by pili-like structures and amorphous material, with the latter being more evident in the structures formed just above the meniscus of the broth. The pili-like structures, whose presence was confirmed by TEM analysis, were also seen on the surface of A. calcoaceticus RAG-1, a hydrocarbon-degrading environmental isolate that adheres to plastic surfaces and hexadecane droplets (Rosenberg et al., 1982). Interestingly, a mutation that abolished the formation of these cell-surface appendages impaired the ability of this bacterium to attach to hydrophobic surfaces and grow in the presence of hexadecane as sole carbon source. This attachment phenotype that depends on the formation of thin filaments is very similar to what we have observed with A. baumannii 19606, which is discussed in more detail below. The amorphous material seen between cells, cell stacks and particularly covering the top cell layer formed above the broth meniscus may represent exopolysaccharides, compounds known to be produced by environmental and clinical isolates of Acinetobacter (Towner et al., 1991). Taken together, the data indicate that environmental and clinical members of the Acinetobacter genus use similar strategies to attach to solid surfaces and to initiate the formation of biofilm structures that allow them to persist and proliferate under harsh conditions. 3482 Cell motility mediated either by appendages such as flagella and pili or the action of glycopeptidolipids is required for biofilm formation by bacteria such as P. aeruginosa (O’Toole & Kolter, 1998a), P. fluorescens (O’Toole & Kolter, 1998b) and Mycobacterium smegmatis (Recht et al., 2000), respectively. The lack of this cell behaviour was one of the criteria used by Baumann et al. (1968) to name this genus Acinetobacter, although twitching and gliding motility have been observed when some isolates were tested on semi-solid media (Towner et al., 1991). All our attempts failed to detect any type of motility with the A. baumannii 19606 prototype strain, indicating that its ability to form biofilms does not depend on this cell property, at least under the experimental conditions used in this work. Based on this behaviour, it is possible to speculate that the multiplication and growth of cells and microcolonies already attached to solid surfaces is the mechanism by which this strain forms biofilms on plastics and glass, as described for some Pseudomonas strains (Heydorn et al., 2000). The genetic approach used in this work proved that the presence of pili-like structures on the surface of A. baumannii 19606 cells is essential in the early steps of the process that leads to the formation of biofilm structures on plastic surfaces. The disruption of the csuC and csuE ORFs resulted in non-piliated cells and abolished cell attachment and biofilm formation. These defects were restored to levels similar to those of the parental strain when the insertion derivative #144 was electroporated with a shuttle vector harbouring the wild-type copy of the csuE gene. In contrast, this recombinant construct could not reverse the phenotype of the mutant #37, in which the insertion interrupted the csuC chaperone-encoding gene, to that of the parental strain. Taken together, the data show that this is indeed a polycistronic operon that encodes functions required for pili assembly and biofilm formation. Based on the nucleotide composition, it appears that part of this gene cluster has been acquired by A. baumannii 19606 from an unrelated source by lateral gene transfer, a possibility that is in accordance with the well known natural competence and ability of members of this genus to acquire DNA easily from different sources (Towner et al., 1991). While the translational products of the fourth and fifth ORFs are the most recognizable since they are highly related to chaperone and usher bacterial proteins, respectively, the four remaining ORFs encode hypothetical proteins potentially involved in pili assembly. Searches with PSI-BLAST, SCANPROSITE and MOTIFSCAN failed to identify the ORF encoding the major pilus protein among the four ORFs whose predicted products are not related to chaperone and usher proteins. On one hand, this observation may indicate that the A. baumannii 19606 gene encoding this subunit is one of the four ORFs located in this cluster, although its translation product is not significantly related to known pili subunit sequences. On the other hand, it is possible that the gene encoding this subunit is located outside of this Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Microbiology 149 Acinetobacter baumannii biofilms gene cluster. This analysis also showed that the systems most related to that of A. baumannii 19606 are found in the human pathogens P. aeruginosa (Stover et al., 2000) and Y. pestis, and the environmental bacterium P. putida (Nelson et al., 2002), all of which have been annotated as hypothetical genes and proteins. In addition, a similar locus was found in the strains BB22 (GenBank accession no. AF339087) and RIMD 2210633 (KXV237) (Makino et al., 2003) of the food-borne pathogen V. parahaemolyticus. This observation suggests that the csu operon is widespread among unrelated bacteria, which could play a central role in the ability of these micro-organisms to attach to and form biofilms on abiotic surfaces, a property that would allow them to persist in their natural environments. In summary, the results presented in this study demonstrate the ability of the prototype strain 19606 of the opportunistic pathogen A. baumannii to attach to and form structures typical of bacterial biofilms on abiotic surfaces under different experimental conditions. In addition, the genetic approach used in this work proved that the expression of csuC and csuE, which belong to a gene cluster related to bacterial loci encoding secretion and pili assembly functions and the production of pili are required in the early steps of the process that leads to biofilm formation. Prior to this study, the loci most closely related to the A. baumannii 19606 csu locus were annotated as hypothetical genes and proteins. This study demonstrates the expression of and identifies a biological function for one member of this uncharacterized group of operons of the chaperone-usher superfamily. Epidemiology, Infections, Management, pp. 2–12. Edited by E. Bergogne-Berenzin, M. L. Joly-Guillou & K. Towner, J. Boca Raton, FL: CRC Press. Bouvet, P. J. M. & Grimont, P. A. D. (1986). Taxonomy of the genus Acinetobacter with the recognition of Acinetobacter baumannii sp. nov., Acinetobacter haemolyticus sp. nov., Acinetobacter johnsonii sp. nov., and Acinetobacter junii sp. nov. and emended descriptions of Acinetobacter calcoaceticus and Acinetobacter lwoffii. Int J Syst Bacteriol 36, 228–240. Calderwood, S. B. & Mekalanos, J. (1988). Confirmation of the fur operator site by insertion of a synthetic oligonucleotide into an operon fusion plasmid. J Bacteriol 170, 1015–1017. Donlan, R. M. & Costerton, J. W. (2002). Biofilms: survival mechanisms of clinically relevant microorganisms. J Clin Microbiol Rev 15, 167–193. Dorsey, C. W., Tomaras, A. P. & Actis, L. A. (2002). Genetic and phenotypic analysis of Acinetobacter baumannii insertion derivatives generated with a Transposome system. Appl Environ Microbiol 68, 6353–6360. Feinberg, A. P. & Vogelstein, B. (1983). A technique for radio- labeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132, 6–13. Gollop, R., Inouye, M. & Inouye, S. (1990). Protein U, a late- developmental spore coat protein of Myxococcus xanthus, is a secretory protein. J Bacteriol 173, 3597–3600. Graber, K., Smoot, L. M. & Actis, L. A. (1998). Expression of iron binding proteins and hemin binding activity in the dental pathogen Actinobacillus actinomycetemcomitans. FEMS Microbiol Lett 163, 135–142. Hermanns, U., Sebbel, P., Eggli, V. & Glockshuber, R. (2000). Characterization of FimC, a periplasmic assembly factor for biogenesis of type 1 pili in Escherichia coli. Biochemistry 39, 1564–1570. Heydorn, A., Nielsen, A. T., Hentzer, M., Sternberg, C., Givskov, M., Ersboll, B. K. & Molin, S. (2000). Quantification of biofilm struc- tures by the novel computer program 2395–2407. ACKNOWLEDGEMENTS COMSTAT. Microbiology 146, Hunger, M., Schumucker, R., Kishan, V. & Hillen, W. (1990). Miami University research funds and Public Health Grants AI4477601A1 and DE13657-02 supported part of this work. We are grateful to C. Wood, coordinator of the Miami University Center of Bioinformatics and Functional Genomics, for his support and assistance with automated DNA sequencing and nucleotide sequence analysis. Analysis and nucleotide sequence of an origin of DNA replication in Acinetobacter calcoaceticus and its use for Escherichia coli shuttle plasmids. Gene 87, 45–51. Klemm, P., Jorgensen, B. J., van Die, I., de Ree, H. & Bergmans, H. (1985). The fim genes responsible for synthesis of type 1 fimbriae in Escherichia coli, cloning and genetic organization. Mol Gen Genet 199, 410–414. Lawrence, J. R., Kroger, D. R., Hoyle, B. D., Costerton, J. W. & Caldwell, D. E. (1991). Optical sectioning of microbial biofilms. REFERENCES Barcak, G. J., Chandler, M. S., Redfield, R. J. & Tomb, J. F. (1991). Genetic systems in Haemophilus influenzae. Methods Enzymol 204, 321–342. Baumann, P., Doudoroff, M. & Stanier, R. Y. (1968). A study of the J Bacteriol 173, 6558–6567. Leigh, J. A., Signer, E. R. & Walker, G. C. (1985). Exopolysaccharide- deficient mutants of Rhizobium meliloti that form ineffective nodules. Proc Natl Acad Sci U S A 82, 6231–6235. Moraxella group II. Oxidase negative species (genus Acinetobacter). J Bacteriol 95, 1520–1541. Makino, K., Oshima, K., Kurokawa, K. & 14 other authors (2003). Genome sequence of Vibrio parahaemolyticus: a pathogenic Bergogne-Berenzin, E. & Towner, K. J. (1996). Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin Microbiol Rev 9, 148–165. Meade, H. M., Long, S. R., Ruvkum, S. E., Brown, S. E. & Ausubel, F. M. (1982). Physical and genetic characterization of symbiotic Bergogne-Berenzin, E., Decre, D. & Joly-Guillou, M. L. (1993). and auxotrophic mutants Rhizobium meliloti induced by transposon Tn5 mutagenesis. J Bacteriol 149, 114–122. Opportunistic nosocomial multiply resistant bacterial infections – their treatment and prevention. J Antimicrob Chemother 32, 39–47. Bergogne-Berenzin, E., Joly-Guillou, M. L. & Towner, K. J. (1996). History and importance of Acinetobacter spp., role in infection, treatment and cost implications. In Acinetobacter: Microbiology, http://mic.sgmjournals.org mechanism distinct from that of V. cholerae. Lancet 361, 743–749. Miller, J. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Neilands, J. (1981). Microbial iron compounds. Annu Rev Biochem 50, 715–731. Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 3483 A. P. Tomaras and others Nelson, K., Weinel, C., Paulsen, I. & 40 other authors (2002). Com- plete genome sequence and comparative analysis of the metabolically versatile Pseudomonas putida KT2440. Environ Microbiol 4, 799–808. Neu, T. R., Kuhlicke, U. & Lawrence, J. R. (2002). Assessment of fluorochromes for two-photon laser scanning microscopy of biofilms. Appl Environ Microbiol 68, 901–909. O’Toole, G. A. & Kolter, R. (1998a). Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol Microbiol 30, 295–304. O’Toole, G. A. & Kolter, R. (1998b). Initiation of biofilm forma- tion in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signaling pathways: a genetic analysis. Mol Microbiol 28, 449–461. O’Toole, G. A., Pratt, L. A., Watnick, P. I., Newman, D. K., Weaver, V. B. & Kolter, R. (1999). Genetic approaches to the study of biofilms. Methods Enzymol 310, 91–109. Solano, C., Garcia, B., Valle, J., Berasain, C., Ghigo, J. M., Gamazo, C. & Lasa, I. (2002). Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. Mol Microbiol 43, 793–808. Stoodley, P., Sauer, K., Davies, D. G. & Costerton, J. W. (2002). Biofilms as complex differentiated communities. Annu Rev Microbiol 56, 187–209. Stover, C. K., Pham, X. Q., Erwin, A. L. & 28 other authors (2000). Complete genome sequence of Pseudomonas aeruginosa PA01, an opportunistic pathogen. Nature 406, 959–964. Thanassi, D. G., Saulino, E. T., Lombardo, M. J., Roth, R., Heuser, J. & Hultgren, S. J. (1998). The PapC usher forms an oligomeric channel: implications for pilus biogenesis across the outer membrane. Proc Natl Acad Sci U S A 95, 3146–3151. Tolker-Nielsen, T., Brinch, U. C., Ragas, P. C., Andersen, J. B., Jacobsen, C. S. & Molin, S. (2000). Development and dynamics of Parkhill, J., Wren, B. W., Thomson, N. R. & 32 other authors (2001). Pseudomonas sp. biofilms. J Bacteriol 182, 6482–6489. Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413, 523–527. Towner, K. J., Bergogne-Berenzin, E. & Fewson, C. A. (1991). Rashid, M. H. & Kornberg, A. (2000). Inorganic polyphosphate is needed for swimming, swarming, and twitching motilities of Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 97, 4885–4890. Recht, J., Martinez, A., Torello, S. & Kolter, R. (2000). Genetic analysis of sliding motility in Mycobacterium smegmatis. J Bacteriol 182, 4348–4351. Rosenberg, M., Bayer, E. A., Delarea, J. & Rosenberg, E. (1982). Role of thin fimbriae in adherence and growth of Acinetobacter calcoaceticus RAG-1 on hexadecane. Appl Environ Microbiol 44, 929–937. Acinetobacter portrait of a genus. In The Biology of Acinetobacter, pp. 1–24. Edited by K. J. Towner, E. Bergogne-Berenzin & C. A. Fewson. New York: Plenum. Vallet, I., Olson, J. W., Lory, S., Lazdunski, A. & Filloux, A. (2001). The chaperone/usher pathways of Pseudomonas aeruginosa: identification of fimbrial gene clusters (cup) and their involvement in biofilm formation. Proc Natl Acad Sci U S A 98, 6911–6916. Vidal, R., Dominguez, M., Urrutia, H., Bello, H., Gonzalez, G., Garcia, A. & Zemelman, R. (1996). Biofilm formation by Acinetobacter baumannii. Microbios 86, 49–58. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: Vidal, R., Dominguez, M., Urrutia, H., Bello, H., Garcia, A., Gonzalez, G. & Zemelman, R. (1997). Effect of imipenem and a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. sulbactam on sessile cells of Acinetobacter baumannii growing in biofilm. Microbios 91, 79–87. Sauer, F. G., Barnhart, M., Choudhury, D., Knight, S. D., Waksman, G. & Hultgren, S. J. (2000). Chaperone-assisted pilus assembly and Wu, C.-J. & Janssen, G. R. (1996). Translation of vph mRNA in bacterial attachment. Curr Opin Struct Biol 10, 548–556. 3484 Streptomyces lividans and Escherichia coli after removal of the 59 untranslated leader. Mol Microbiol 22, 339–355. Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Sun, 18 Jun 2017 17:27:57 Microbiology 149