- Volume 149, Issue 1, 2003

The taxonomic studies on the genus Pseudomonas performed in the Department of Bacteriology of the University of California at Berkeley played a significant rôle in the development of modern prokaryote taxonomy, which started in the 1960s. This impact was due to a revival of the method of den Dooren de Jong for the nutritional analysis of chemoorganotrophic organisms and, mostly, to the introduction of the determination of rRNA as a method of taxonomic analysis. While the introduction of the nutritional studies facilitated the characterization of Pseudomonas and other chemoorganotrophs, the applicability of the rRNA studies extended to all prokaryotes.

Bacteria have developed a wide arsenal of survival strategies to cope with the specific problems posed by their environment. These processes are carefully regulated and complex signal transduction cascades ensure proper activation of the adequate adaptive response. An intriguing observation is that generally the regulation pathways of the different adaptive processes are highly intertwined. In this review, this phenomenon is illustrated by the regulation of genetic competence development in Bacillus subtilis. The different regulation pathways which make up the gene regulation network that controls the development of competence are described, and their connections to other adaptive processes in B. subtilis are discussed.

Bacillus subtilis uses two-component signal transduction systems to sense intra- and extracellular stimuli to adapt to fluctuating environmental situations. Regulator aspartate phosphatases (Raps) have important roles in these processes, as they can dephosphorylate certain response-regulators, and are themselves subject to cell-density-controlled inhibition by secreted Phr (phosphate regulator) peptides. Eleven chromosomal genes encode this family of phosphatases, but in addition, certain strains contain endogenous plasmids with genes for homologous Rap–Phr systems. Plasmid pTA1060 encodes Rap60 and its antagonistic signalling molecule Phr60. Strikingly, expression of Rap60 in B. subtilis 168 strongly repressed the production of proteolytic enzymes. In fact, the transcription of the aprE gene, encoding a major extracellular protease, was shown to be decreased upon Rap60 expression, whereas this effect could be antagonized by the extracellular addition of synthetic Phr60 pentapeptide. Finally, transcription studies suggest that Rap60 dephosphorylates a component of the phosphorelay and is coupled to aprE transcription by the transition-state regulator AbrB. In conclusion, these data show that endogenous plasmids contain functional Rap–Phr systems and for the first time, that Rap–Phr systems can mediate cell-density controlled production of secreted proteases. This quorum-sensing mechanism might enable B. subtilis to suppress protease production under conditions of low cell densities when nutrients are still available in sufficient amounts.

The genes for nitrous oxide (N2O) reduction, nosRZDFYL, are clustered on the chromosome of Pseudomonas aeruginosa. Promoter assays using transcriptional fusions to lacZ revealed that the structural gene for nitrous oxide reductase, nosZ, is transcribed with the upstream nosR gene. The nosR gene product is not required for the activity of the nosR promoter. A sequence similar to the consensus FNR-binding motif was found 41·5 bp upstream from the major transcriptional start point of nosR. Mutation of the motif significantly reduced the promoter activity. DNR, an FNR-related transcriptional regulator required for the expression of denitrification genes in P. aeruginosa, is necessary for the transcription of nosR, indicating that the motif is recognized by DNR. Nitrite (NO−2), nitric oxide (NO) and NO-generating reagents induced nosR promoter activity, but N2O did not. The NO−2-induced nosR promoter activity was reduced by mutation of the NO−2 reductase gene. However, a low concentration of NO−2 induced the promoter activity in a NO reductase mutant. These results indicate that NO is the inducer molecule for transcription of the nos genes.

Fresh isolates of the oral bacterial pathogen Actinobacillus actinomycetemcomitans exhibit a fimbriated, rough colony phenotype. Evidence suggests that the fimbrial subunit gene flp is part of a cluster of 14 genes (flp to tadG) thought to encode proteins involved in the synthesis, assembly and export of these fimbriae. To determine the transcriptional organization of the 5′ terminus of this gene cluster, total RNA from rough and smooth phenotype variants of A. actinomycetemcomitans strain 283 were analysed by RT-PCR. Primers designed to amplify regions spanning gene junctions or multiple genes yielded amplicons at each individual gene junction from flp to tadD for both the rough and smooth variants. Semi-quantitative RT-PCR of the rcpA to tadZ amplicon revealed that significantly more mRNA was transcribed from the rough than the smooth variant. Longer amplicons encompassing flp to tadZ (3·9 kb) and tadA to tadD (2·1 kb) were also detected, but only from the rough variant. Rapid amplification of cDNA ends (RACE) was used to identify the 5′ end of the mRNA containing flp. Antisense primers located within rcpC, orfB and flp-2 enabled amplification of a RACE product that was subsequently isolated and subcloned into pGEM-T. DNA sequencing indicated that the 5′ end of the mRNA was located at a G or T nucleotide −102 to −101 nt upstream of flp. Corresponding σ70 consensus sequences were located at −10 (TATAAT) and −35 (TTGCAT) relative to the transcription start site. These data confirm that the flp gene cluster is an operon transcribed as a polycistronic message commencing from a G or T nucleotide located in the intergenic region upstream of flp. Promoter function of the flp upstream region was confirmed using a lacZ reporter gene construct transformed into Escherichia coli. RT-PCR analysis further suggests that although transcription does occur in both the rough and smooth variants, full-length transcripts are rapidly degraded or are significantly downregulated in the smooth variant.

PCR primers and hybridization probes were designed for the 16S rRNA genesof six bacterial species or groups typically present in human faeces or usedin the dairy industry. The primers and probes were applied for quantificationof the target bacterial genomes added in artificial DNA mixtures or faecalDNA preparations, using dot-blot hybridization and real-time PCR with SYBRGreen I and TaqMan chemistries. Dot-blot hybridization with 33P-labelled oligonucleotide probes was shown to detect a 10 %target DNA fraction present in mixed DNA samples. Applicability of the rDNA-targetedoligonucleotide probes without pre-enrichment of the 16S gene pool by PCRwas thus limited to the detection of the predominant microbial groups. Real-timePCR was performed using a 96-well format and was therefore feasible for straightforwardanalysis of large sample amounts. Both chemistries tested could detect andquantify a subpopulation of 0·01 % from the estimatednumber of total bacterial genomes present in a population sample. The linearrange of amplification varied between three and five orders of magnitude forthe specific target genome while the efficiency of amplification for the individualPCR assays was between 88·3 and 104 %. Use of a thermallyactivated polymerase was required with the SYBR Green I chemistry to obtaina similar sensitivity level to the TaqMan chemistry. In comparisonto dot-blot hybridization, real-time PCR was easier and faster to performand also proved to have a superior sensitivity. The results suggest that real-timePCR has a great potential for analysis of the faecal microflora.

Subgingival plaque samples were obtained from 26 subjects with advanced periodontal lesions. Bacterial diversity was analysed by amplification of the 16S rRNA genes with two different primer sets, and subsequent cloning and sequencing. A total of 578 sequences was analysed after the exclusion of chimeras. The authors found 148 phylotypes with the clone library 27f/519r (number of clones n=322; coverage, C=66 %) and 75 phylotypes with the clone library 515f/1525r (n=256; C=84 %). Comparative sequence analysis revealed that 70 % of all of the analysed sequences showed a similarity of at least 99 % to sequences deposited in public databases. The classes Actinobacteria, Bacilli, Bacteroidetes, Clostridia, Deferribacteres, Flavobacteria, Fusobacteria, Mollicutes, Spirochaetes and all classes of the Proteobacteria were represented. Sequences that were at least 99 % identical to Porphyromonas gingivalis, Filifactor alocis and Treponema socranskii were present in at least one-third of the patients. Libraries generated with the two PCR primer pairs differed significantly in their representation of the families Porphyromonadaceae, Prevotellaceae, Fusobacteriaceae, Eubacteriaceae, Streptococcaceae and Acidaminococcaceae. A total of 14 sequences exhibited less than 97 % identity to sequences published previously and were assigned to six different families within the phyla Bacteroidetes and Firmicutes. Twelve of 20 putative pathogens were recovered, which were recently proposed to be associated with periodontitis.

Infection with bacterial species belonging to the Burkholderia cepacia complex contribute significantly to morbidity and mortality in persons with cystic fibrosis (CF). The majority of isolates recovered from CF patients belong to B. cepacia genomovar III and several distinct ‘epidemic’ strains have been described. This study examined the population structure of B. cepacia genomovar III by using multilocus restriction typing, indexing allelic variation at five chromosomal genes by restriction analysis of PCR-amplified genes. A collection of 375 isolates, recovered from CF and non-CF patients and natural environments in North America, Europe and Australia, was examined. Among these isolates 144 different restriction types were found. Overall, the population is at linkage disequilibrium, indicating that it has a clonal structure. The majority (86·7 %) of restriction types grouped into three major clonal complexes, comprising the epidemic ET12, PHDC and Midwest clonal lineages. The analysis indicates that these complexes are geographically widespread and demonstrate varying degrees of genetic recombination. These differences in population structure among major clonal complexes within the same species are likely related to differences in evolutionary history and ecology. The observation that genetic recombination is frequent within some B. cepacia genomovar III populations has important implications for the biotechnological use of B. cepacia complex species.

This study was based on the hypothesis that groundwater-derived biofilms may provide a reservoir for coliform or pathogenic bacteria as has been observed in drinking water distribution systems. Escherichia coli, labelled with green fluorescent protein, was found to colonize all layers of mixed-population biofilms developed in association with indigenous groundwater micro-organisms in a laboratory-scale reactor. Biofilm-associated E. coli was removed at a slower rate from the reactor flasks than planktonic E. coli under a continuous flow regime. During flow-through of groundwater, planktonic E. coli removal was slower in flasks containing coverslips for enhanced biofilm development compared to a control flask without coverslips. Conversely, during flow-through of treated effluent, planktonic E. coli removal was faster in flasks with coverslips compared to without. Removal of attached E. coli was also fastest in the coverslip-containing flasks with effluent flow-through. This suggests that an increase in available nutrients may reduce E. coli survival potential due to either enhanced competition for nutrients or enhanced antagonism by the indigenous microbial population. Under identical conditions, GFP-labelled Pseudomonas aeruginosa was found to persist in the biofilms for longer than E. coli, most notably when exposed to flow-through of treated effluent. However, prolonged persistence of P. aeruginosa in the effluent could not be attributed to an association with the biofilms. This study has shown that under certain conditions the presence of mixed-population biofilms may limit the survival potential of enteric bacterial pathogens introduced into groundwater.

In this manuscript, the authors have sought to gain a better understanding of the interactions between Escherichia coli and lactic acid bacteria (LAB) isolated from Rogossa MRS agar along the digestive tract of grain- and forage-fed cattle. E. coli from cattle receiving a high-grain diet were more numerous (P<0·05) than from the high-forage diet and the highest numbers were in the faeces. Isolates on Rogossa MRS agar were always greater in the high-grain diet (P<0·05) and contained a significant number of LAB. A random set of Rogossa MRS agar colonies was selected and artificial neural networks were used to develop a relationship between colony description and species which was validated using sequence analysis (16S rDNA). The neural networks correctly predicted species in more than 80 % of cases and was composed, primarily, of Lactobacillus vitulinus, Lactobacillus ruminis, Selenomonas ruminantium, Streptococcus bovis, Acidaminococcus fermentans and Megasphaera elsdenii. In conjunction with statistical diversity indices, it was demonstrated that diversity in the high-fibre diet was always lower and was a consequence of the dominance of Str. bovis. In contrast, the diversity in the high-grain diet was greater (P<0·05) and was a consequence of the decline in Str. bovis. These data demonstrate that there is a positive relationship between coliform and LAB isolates throughout the digestive tract of cattle, and diet is the major factor regulating bacterial composition.

Small, coccoid and rod-shaped Synechococcus-type cyanobacteria with either phycoerythrin or phycocyanin as major accessory pigments were isolated from several large, temperate-zone lakes and the brackish Baltic Sea. The picocyanobacteria had two ribosomal operons with a long internal transcribed spacer (ITS-1) separating the 16S rDNA and 23S rDNA. A 16S rRNA-based phylogenetic analysis assigned all isolates to the picophytoplankton clade [sensu Urbach, E., Scanlan, D. J., Distel, D. L., Waterbury, J. B. & Chisholm, S. W. ( 1998 ). J Mol Evol 46, 188–201], which also comprises marine Synechococcus spp. and Prochlorococcus spp. The strains assorted to five paraphyletic clusters each containing two or more strains with 99·4–100 % 16S rRNA sequence identity. Five corresponding strain clusters were deduced from analysis of ITS-1 sequences. Sequence divergence in ITS-1 varied between 23 % in the most divergent and 1 % in the phylogenetically most conserved cluster. Clustered strains with low sequence divergence in ITS-1 were frequently isolated from a single ecosystem or hydrographically comparable lakes in the same region. They represent physiologically distinct ecotypes of species which, among other phenotypic variations, frequently differed in their major accessory pigments, the phycobiliproteins. The reproduction of the various pigment traits in different lineages was not correlated with the phylogenetic divergence deduced from 16S rRNA or ITS-1 sequences but rather seemed to be related to characteristics of the ecosystem and habitat from which the strains were isolated. The occurrence of a comparable spectrum of phenotypes in different lineages and ecosystems indicates that different strain clusters developed similar ecotypes during independent adaptive radiations.

Bacterial cultures capable of reducing nitrate to nitrite, or of complete denitrification, were established from 5, 10, 15 and 20 cm depths of a freshwater sediment. Taxonomic analysis of the 56 isolates using 16S rRNA gene sequences revealed an unexpected species richness, which included representatives of the γ-Proteobacteria, Bacillus spp., Staphylococcus spp. and members of the Actinobacteria. Gram-positive species tended to predominate in the lower depths of the sediment, where there was evidence of active sulphate respiration. Sequences (from the narG gene) potentially encoding the catalytic subunit of the membrane-associated nitrate reductase were successfully amplified from 46 of the isolates, using a nested PCR with four degenerate primers. NarG sequences clustered into three major groupings that were supported by alternative phylogenetic analyses. The NarG sequences from Gram-positive isolates (according to rRNA gene phylogeny) clustered together within sequences from the low-G+C Gram-positive bacteria. However, this cluster also included two sequences from members of the genus Pseudomonas. Another group contained mostly NarG sequences from the Proteobacteria (according to rRNA gene phylogeny), but also included five sequences from Gram-positive species. The third group of NarG sequences contained three sequences from Gram-positive species. Thus, the NarG-derived phylogeny is not entirely consistent with 16S rRNA-based taxonomy, precluding the use of the narG gene as a taxonomically useful tool for the characterization of nitrate-respiring bacteria. Total DNA was also extracted from the four depth intervals of the sediment sample and used in similar narG amplifications. Most sequences amplified directly from environmental DNA clustered in the Gram-negative group, and none was in the predominantly Gram-positive group. The study also revealed a degree of spatial organization of a nitrate-respiring community in terms of both microbiology and narG sequences.

The gene encoding cholera toxin, the principal virulence factor of Vibrio cholerae, is encoded by a filamentous, lysogenic bacteriophage known as CTXϕ. The genome of V. cholerae, the host for CTXϕ, consists of two chromosomes, one large and one small. Here, it is shown that localization and array of CTX prophage DNA in either the large or small chromosome of V. cholerae is likely to be one of the reasons for the emergence of O1 biotype El Tor variants isolated just before and after the V. cholerae O139 cholera outbreak in 1992. Analyses of the organization of the CTX region of the genome of pre-O139 El Tor strains revealed that these strains carry two distinct CTX prophages integrated in the small chromosome in tandem: CTXET, the prophage having a conserved NotI site in its repeat sequence segment which seems to be specific for the El Tor strains so far examined, followed by CTXcalc-like genome, the prophage found in recent O139 clinical isolates from Calcutta. In sharp contrast, in post-O139 El Tor strains only one copy of the CTXET prophage was found to be integrated in the large chromosome. To the authors' knowledge, the presence of CTX prophage in the small chromosome of O1 El Tor strains has not been reported previously. It is also shown that the difference in the CTX copy number and the position of the bacteriophage on the genomes of pre- and post-O139 El Tor strains have an effect on cholera toxin production. While a pre-O139 strain produced maximum cholera toxin in yeast extract/peptone medium at 30 °C, a post-O139 El Tor strain showed maximal yield at 37 °C, indicating differential regulation of cholera toxin between the strains. It appears from this study that the variation in the integration site of the CTX prophage, its copy number and the presence of diverse phage genomes in V. cholerae O1 biotype El Tor may be strategically important for generating variants with subtle phenotypic modulations of virulence factor production in this longest-ruling seventh pandemic strain.

A clfB : : tetK reporter was constructed in Staphylococcus aureus strains Newman and 8325-4, whereby the level of tetracycline resistance reflected the activity of the clfB promoter. Wild-type strains carrying a single copy of this construct exhibited a low level of tetracycline resistance, suggesting that the clfB promoter is weak. Spontaneous mutants that grew at higher tetracycline concentrations were isolated. Some were due to point mutations in the clfB promoter that led to increased expression of the tetK gene. The clfB promoter was identified by primer extension analysis and −35 and −10 elements were assigned. The promoter regions from the tetracycline-resistant mutants were sequenced and several had base changes within or adjacent to the −35 box. Three created the consensus −35 sequence TTGACA. The mutant clfB promoters were fused to lacZ. β-Galactosidase activity was six- to ninefold higher in the mutant strains compared to the wild-type. The wild-type clfB gene was placed under the control of the mutant promoters. ClfB expression was higher than the corresponding wild-type strains and the protein was present on bacteria from the stationary phase instead of being confined to the exponential phase. Therefore, mutations in the clfB promoter that cause changes in the −35 region produce a stronger promoter that is capable of increased transcription and, as a result, increased expression of ClfB.

Prenylation is a post-translational protein modification process that results in correct protein localization to membranes in the cell. It is mediated by at least three prenyltransferases (PTFs) in eukaryotic cells. The RAM2 gene encodes the common α-subunit of two PTFs, farnesyltransferase (FTase) and geranylgeranyltransferase I (GGTase I). In this study, the RAM2 gene of the prenylation pathway in the opportunistic pathogen Candida albicans was analysed. The heterozygote trisomy test was used to demonstrate that RAM2 is essential to the viability of C. albicans. RAM2 mRNA in the yeast and hyphae growth states was not detected by Northern blot analysis, but was detected by RT-PCR. Drugs that inhibit mammalian PTFs do not alter Candida cell growth, but they do inhibit FTase and GGTase I activities in cell-free enzyme assays. The data from genetic studies and cell-free enzyme assays suggest that the drugs may not have access to the Candida cytoplasm. The regulation of PTF subunits was also examined in a strain in which RAM2 was under the control of a regulable promoter. Overall, this study demonstrated that RAM2 is essential to C. albicans, indicating that protein prenylation is an indispensable cellular process in this yeast.

Listeria monocytogenes is a facultative intracellular bacterial pathogen that regulates the expression of virulence-associated gene products in response to specific host cell compartment environments. The PrfA protein of L. monocytogenes functions as a key regulatory factor required for the differential expression of bacterial virulence gene products within infected host cells. PrfA both positively and negatively regulates its own expression, and while PrfA positive regulation is required for cell-to-cell spread of L. monocytogenes and for full virulence in infected mice, a role for negative regulation has been of presumed importance but has yet to be established. To address the role of negative regulation of prfA expression in L. monocytogenes pathogenesis, prfA promoter mutations designed to reduce or eliminate negative regulation were introduced into L. monocytogenes and analysed for their effects on patterns of PrfA-dependent gene expression and virulence in murine models of infection. High level PrfA production resulting from the prfA promoter mutations produced significantly increased levels of PrfA-regulated gene expression in broth-grown cultures; however the apparent loss of negative prfA regulation had no deleterious effects on growth and spread of the bacteria within infected tissue culture cells or on virulence in mice. The results indicate that while negative regulation of prfA expression exists and provides a feedback system for the control of PrfA synthesis, this feedback system is dispensable for virulence.

Helicobacter pylori possesses two distinct thioredoxin proteins (Trx1 and Trx2) which may play important roles in the ability of this bacterium to survive oxidative stress. Trx1 has previously been shown to be an electron donor in vitro for alkyl-hydroperoxide reductase (AhpC), one of three members of the peroxiredoxin family of antioxidant peroxidases present in H. pylori. In this study, mutants in the trxA1 and trxA2 genes encoding Trx1 and Trx2, respectively, and in the tpx and bcp genes, which encode the remaining two members of the H. pylori peroxiredoxin family, were constructed in order to determine their roles in resistance to damage by reactive oxygen and nitrogen species. Mutation of trxA1 led to a pronounced increase in sensitivity to oxygen, hydrogen peroxide and the superoxide generator paraquat, as well as to the nitric oxide (NO) releasers sodium nitroprusside (SNP) and S-nitrosoglutathione (GSNO), consistent with an in vivo role for Trx1 as a reductant for AhpC. A trxA2 single mutant grew normally in an atmosphere of 2 % (v/v) O2 but grew very poorly in 10 % (v/v) O2. It showed slight increases in killing by hydrogen peroxide, paraquat, SNP and GSNO compared to the wild-type, but was significantly more sensitive to cumene hydroperoxide in disc-diffusion assays. A trxA1 trxA2 double mutant was very sensitive to all of the oxidative and nitrosative stresses applied. Comparison of the phenotypes of the tpx and bcp mutants showed that Tpx plays a significant role in peroxide and superoxide resistance in H. pylori, while the role of Bcp is minimal. No evidence was obtained for a role for either Tpx or Bcp in resistance to the toxic effects of NO. The results show that a functional thioredoxin system is necessary for both oxidative and nitrosative stress resistance in H. pylori but, surprisingly, is not essential for viability despite the absence of glutathione and a glutaredoxin system in this bacterium.

Cryptococcus neoformans comprises two varieties (neoformans and gattii) and four serotypes (A, B, C and D). Fertile isolates of both mating types have been identified in serotypes B, C and D; however, a fertile serotype A MAT a strain has not been confirmed, although serotype A MATα strains will mate with serotype D MAT a strains. Preliminary analysis of a recent Italian environmental isolate (IUM 96-2828) suggested that this strain was haploid, serotype A and MAT a. In this study, IUM 96-2828 has been characterized in detail. A mating reaction between IUM 96-2828 and H99 (serotype A MATα) produced abundant spores with an equal distribution of MAT a and MATα progeny, all of which were serotype A. Karyotypic analysis of F1 spores revealed evidence of recombination, confirming that IUM 96-2828 was fertile. The MAT a pheromone gene from IUM 96-2828 was sequenced and found to be most closely related to the serotype D MAT a pheromone gene. Phylogenetic comparisons of other genes not linked to mating type also suggested IUM 96-2828 was most closely related to serotype A strains. Biochemical analysis showed that the carbon assimilation profiles of H99 and IUM 96-2828 were identical for 97 % (30/31) of the substrates while isozyme analysis showed 89 % (17/19) identity. Assays of major virulence factors found no difference between H99 and IUM 96-2828. Virulence studies using the mouse model demonstrated that IUM 96-2828 was virulent for mice, although it was less virulent than H99. These data strongly suggest that IUM 96-2828 is a true haploid serotype A MAT a isolate that is fertile.

The naturally mummified remains of a mother and two daughters found in an 18th century Hungarian crypt were analysed, using multiple molecular genetic techniques to examine the epidemiology and evolution of tuberculosis. DNA was amplified from a number of targets on the Mycobacterium tuberculosis genome, including DNA from IS6110, gyrA, katG codon 463, oxyR, dnaA–dnaN, mtp40, plcD and the direct repeat (DR) region. The strains present in the mummified remains were identified as M. tuberculosis and not Mycobacterium bovis, from katG and gyrA genotyping, PCR from the oxyR and mtp40 loci, and spoligotyping. Spoligotyping divided the samples into two strain types, and screening for a deletion in the MT1801–plcD region initially divided the strains into three types. Further investigation showed, however, that an apparent deletion was due to poor DNA preservation. By comparing the effect of PCR target size on the yield of amplicon, a clear difference was shown between 18th century and modern M. tuberculosis DNA. A two-centre system was used to confirm the findings of this study, which clearly demonstrate the value of using molecular genetic techniques to study historical cases of tuberculosis and the care required in drawing conclusions. The genotyping and spoligotyping results are consistent with the most recent theory of the evolution and spread of the modern tuberculosis epidemic.

The binding of Campylobacter jejuni to fibronectin (Fn), a component of the extracellular matrix, is mediated by a 37 kDa outer-membrane protein termed CadF for Campylobacter adhesion to fibronectin. The specificity of C. jejuni binding to Fn, via CadF, was demonstrated using antibodies reactive against Fn and CadF. More specifically, the anti-CadF antibody reduced the binding of two C. jejuni clinical isolates to immobilized Fn by greater than 50 %. Furthermore, a C. jejuni wild-type isolate, in contrast to the isogenic CadF mutant, was found to compete with another C. jejuni wild-type isolate for host cell receptors. Given the relationship between the pericellular Fn matrix and the cytoskeleton, the involvement of host cell cytoskeletal components in C. jejuni internalization was also examined. Cytochalasin D and mycalolide B microfilament depolymerizing agents resulted in a significant reduction in C. jejuni invasion. Studies targeting paxillin, a focal adhesion signalling molecule, identified an increased level of tyrosine phosphorylation upon C. jejuni infection of INT 407 cells. Collectively, these data suggest CadF promotes the binding of C. jejuni to Fn, which in turn stimulates a signal transduction pathway involving paxillin.