• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Mar 2011; 193(5): 1065–1075.
Published online Dec 30, 2010. doi:  10.1128/JB.01252-10
PMCID: PMC3067607

Phenotypic and Genotypic Evidence for l-Fucose Utilization by Campylobacter jejuni[down-pointing small open triangle]


Campylobacter jejuni remains among the leading causes of bacterial food-borne illness. The current understanding of Campylobacter physiology suggests that it is asaccharolytic and is unable to catabolize exogenous carbohydrates. Contrary to this paradigm, we provide evidence for l-fucose utilization by C. jejuni. The fucose phenotype, shown in chemically defined medium, is strain specific and linked to an 11-open reading frame (ORF) plasticity region of the bacterial chromosome. By constructing a mutation in fucP (encoding a putative fucose permease), one of the genes in the plasticity region, we found that this locus is required for fucose utilization. Consistent with their function in fucose utilization, transcription of the genes in the locus is highly inducible by fucose. PCR screening revealed a broad distribution of this genetic locus in strains derived from various host species, and the presence of this locus was consistently associated with fucose utilization. Birds inoculated with the fucP mutant strain alone were colonized at a level comparable to that by the wild-type strain; however, in cocolonization experiments, the mutant was significantly outcompeted by the wild-type strain when birds were inoculated with a low dose (105 CFU per bird). This advantage was not observed when birds were inoculated at a higher inoculum dose (108 CFU per bird). These results demonstrated a previously undescribed substrate that supports growth of C. jejuni and identified the genetic locus associated with the utilization of this substrate. These findings substantially enhance our understanding of the metabolic repertoire of C. jejuni and the role of metabolic diversity in Campylobacter pathobiology.

Campylobacteriosis remains among the leading causes of bacterial food-borne illness worldwide and is a significant burden to public health due to lost worker productivity, health care cost, and premature deaths (21). Additionally, Campylobacter infection is a primary antecedent to severe sequelae, such as Guillain-Barré syndrome and Miller-Fisher syndrome. Campylobacter jejuni, the etiologic agent implicated in the preponderance of campylobacteriosis cases, is a microaerophilic thermophile of the class Epsilonproteobacteria that grows best between 37°C and 42°C. Sources of C. jejuni infection include contaminated drinking water or milk and contact with infected pets; however, consumption of undercooked poultry meat is the most frequent source (34).

Despite recent advances in understanding Campylobacter physiology, relatively little is known about this organism's metabolism, particularly within the host. Initial studies reported that Campylobacter fetus neither ferments nor oxidizes common carbohydrates (2, 25, 46), and the genus is often deemed asaccharolytic. Consistent with these observations, the NCTC 11168 genome encodes an incomplete Embden Meyerhof Parnas (EMP) pathway due to the absence of phosphofructokinase (36), rendering the organism unable to catabolize exogenous sugars through fructose-1,6-bisphosphate. Interestingly, pyruvate kinase (pyk) activity can be detected in cell extracts at high levels, potentially creating a futile cycle among pyruvate, phosphoenolpyruvate (PEP), and oxaloacetate (49). Velayudhan and Kelly hypothesize that pyruvate kinase functions in a catabolic role where its activity is exerted on an unidentified substrate(s). This enigmatic finding highlights our dearth of knowledge of Campylobacter physiology, particularly in terms of growth substrates that confer a metabolic advantage in different environmental niches.

Several amino acids (Asp, Glu, Ser, Asn, and Pro) are utilized by C. jejuni (16, 26, 32), and mutations to key genes necessary for their metabolism attenuate C. jejuni's in vivo colonization potential (16, 48). In addition to amino acids, respiratory activity is detected in the presence of 2-ketoacids, such as pyruvate, 2-ketoglutarate, succinate, and fumarate, most of which are noted to be citric acid cycle intermediates (13, 17). The prevailing understanding of C. jejuni's in vivo metabolism is centered on amino acids to support growth and establish colonization of the host intestines. While general conclusions can be made for isogenic mutants, the metabolic diversity of C. jejuni is evidenced by amino acid auxotrophs and differential carbon utilization (12, 13). For example, strain-specific catabolism of glutathione and asparagine provided evidence of metabolism contributing to tissue tropism (18). Thus, ancillary metabolic features may present strains with a competitive advantage allowing for the colonization of a broad range of niches. Although metabolism has not traditionally been considered a virulence factor, the abilities to survive and persist in the environment and to cause disease in the host are predicated on nutrient availability. Therefore, the metabolic diversity within pathogens may be an overlooked feature of an organism's virulence potential.

Indirect evidence of an exogenous carbohydrate playing a role in C. jejuni's physiology was produced by chemotaxis and cell adherence assays. As a commensal residing primarily in the mucosal layer of the intestines, C. jejuni demonstrates motility toward and binds to mucin (20, 31, 45). l-Fucose is a prominent component of eukaryotic glycoproteins and comprises 4 to 14% of the total oligosaccharide component of mucin (3). Specifically, fucose has been described as a chemoattractant for C. jejuni (20, 45), increasing flaA promoter activity (3). In vitro binding of C. jejuni to α1,2-fucosylated glycans has been demonstrated previously (11, 42), and free fucose reduces adherence to intestinal cells (10), suggesting recognition of this oligosaccharide by the bacterium. In addition, respiratory activity can be detected with fucose as a carbon source (8, 15), although the assay identified other sugars believed to be false positives (29).

In this report, we provided evidence of strain-specific utilization of fucose for Campylobacter growth. Furthermore, the genetic locus for the fucose phenotype (Fuc) was mapped to a previously described hypervariable plasticity region (PR2) (38). It was also shown that expression of this genetic locus was induced by fucose. Chick colonization experiments revealed that fucose utilization is not essential for Campylobacter colonization of this species but may be advantageous when birds are exposed to low levels of C. jejuni. Together, these results identified a new metabolic substrate for C. jejuni, which may facilitate adaptation of this organism to nutrient-poor conditions.


Bacterial strains and conditions.

Bacterial strains and plasmids used in this study are listed in Table Table1.1. Escherichia coli DH5α was routinely cultured in lysogeny broth or on lysogeny agar at 37°C. C. jejuni strains were cultured on Mueller-Hinton (MH) agar at 37°C microaerobically (85% N2, 10% CO2, 5% O2) in a gas incubator (model 3130; Forma Scientific, Pittsburgh, PA).

Bacterial strains and plasmids used in this study

Phenotypic MicroArray.

The Phenotypic MicroArray (Biolog, Hayward, CA) was used according to the manufacturer's recommendation to identify differential carbon utilization by C. jejuni strains NCTC 11168, 81116, and 81-176, with dye mix D as the redox indicator of respiration. Plates were incubated for 48 h at 42°C microaerobically as described above.

Growth in chemically defined medium.

Unless otherwise noted, all chemicals were acquired from Sigma (St. Louis, MO). For experiments using defined medium, minimum essential medium (no. 11430-030; Invitrogen, Carlsbad, CA) was supplemented with FeSO4 (12.5 mg liter−1) and ammonium formate (0.3625 mM) to produce basal defined medium (BDM) (pH 7.0 with NaOH). Batch cultures of the BDM (10 ml) were aliquoted into a sterile 18- by 150-mm glass tube containing a 10- by 3-mm stir bar and capped with a Morton closure. Growth substrates were individually added to each tube from 100× stock solutions stored at −20°C. Stock growth substrates included l-fucose (5 M), d-glucose (5 M), sodium pyruvate (2 M), and a mixture of l-aspartate, l-glutamate, l-proline, and l-serine (DEPS) at 39 mM, 36 mM, 56 mM, and 130 mM, respectively. An inoculum (optical density at 600 nm [OD600] of 0.1; 1-cm path length) was prepared for each strain in the BDM and diluted 1:10, and 100 μl of the diluted inoculum was added to each tube to give an initial cell density of ~105 CFU ml1. The tubes were placed upright in a custom acrylic tube holder mounted on a stir plate set at a low setting (~230 rpm) and incubated. After 24 h and 48 h of incubation, 10-fold serial dilutions were prepared in phosphate-buffered saline (PBS), plated on MH agar, and incubated for 48 h. After incubation, plates were enumerated, and results were recorded and are presented as log10 CFU ml1. Growth experiments for each strain were repeated on three separate occasions.

For serial passage experiments, the BDM was supplemented with l-fucose (50 mM final concentration), inoculated at a density of ~106 CFU ml1, and grown with gentle agitation as described above. After incubation for 24 h, 100 μl was inoculated into 10 ml of fresh BDM containing fucose. The freshly inoculated culture was incubated for 24 h, after which 100 μl was inoculated into 10 ml of fresh BDM containing fucose. This process was repeated two more times for a total of four serial passages. Serial dilutions of the cultures grown for 24 h were prepared in PBS and plated on MH agar. The density of the freshly inoculated culture was estimated from these plate counts. The experiment was repeated three times.

Construction of fucP mutant.

An isogenic fucP mutant was constructed using an inverse PCR strategy described previously (47). The fucP gene was PCR amplified with primers fucP_L and fucP_R (Table (Table2),2), ligated with pGEM-T (Promega, Madison, WI), and transformed into chemically competent E. coli DH5α. Recombinant plasmid DNA was isolated and used as a template for inverse PCR using primers ifucP_L and ifucP_R, deleting an internal fragment of 512 nucleotides. The cat cassette was PCR amplified from pUOA18 (50) with the cat_L and cat_R primers. The inverse PCR and cat amplicons were purified by ethanol precipitation and digested with BglII (New England BioLabs, Ipswich, MA) according to the manufacturer's recommendations. The digested antibiotic cassette and inverse PCR product were ligated and used for transforming E. coli DH5α. For transformation of naturally competent C. jejuni NCTC 11168, purified recombinant plasmid DNA (pWM10) was diluted to 100 ng μl1, spotted (5 μl) onto a overnight lawn of C. jejuni, and incubated for 6 h at 37°C. After incubation, the lawn was harvested and isolates harboring the ΔfucP::cat mutation were selected on MH agar containing chloramphenicol (10 μg ml1). Individual colonies were streaked for isolation on MH agar containing chloramphenicol and incubated at 37°C for 48 h. After incubation, isolated colonies were plated on MH agar and incubated at 37°C for 24 h. Mutation of fucP and orientation of the cat cassette were confirmed by PCR. Cells were harvested in brain heart infusion broth containing 20% glycerol and stored at −80°C.

Primers used in this study

Complementation of C. jejuni ΔfucP::cat mutation.

The ΔfucP::cat mutation was complemented by inserting fucP and its downstream genes between the structural genes of the ribosomal gene cluster as described by Karlyshev and Wren (24), with the following modifications. The ribosomal cluster was amplified using rrs and rrl primers (24) and cloned into pGEM-T (Promega) to produce plasmid pRR. pRR was digested with MfeI and ligated with the aphA3 cassette, which was amplified from pMW10 (51) using primers aphA3_L and aphA3_R. The resulting plasmid, pRRK, transcribes the aphA3 cassette in the opposite orientation to the ribosomal genes. pRRK was digested with XbaI and used to clone the complementing open reading frame(s) (ORF). DNA fragments of various lengths were amplified from NCTC 11168. In all, four complementation constructs were created, i.e., fucP, fucP-cj0487, fucP-cj0488, and fucP-aldA constructs (see Fig. Fig.1A1A for details). The left primer, KIfucP_L, served as an anchor for which different right primers were used (Table (Table2;2; Fig. Fig.1A).1A). The fucP, fucP-cj0487, fucP-cj0488, and fucP-aldA ORF fragments were amplified using KIfucP_R, KIfucP_R4, KIfucP_R5, and KIfucP_R3, respectively. After amplification of the ORF fragments, the amplicons were digested with the appropriate restriction endonuclease (either XbaI or AvrII) and cloned into pRRK. Constructs in which the ORF(s) is oriented in the same transcriptional direction as the ribosomal genes were purified and used as suicide vectors as described above for mutant construction. Complemented fucP strains were selected on MH agar containing kanamycin (50 μg ml1) and confirmed by PCR.

FIG. 1.
Identification of the genetic locus for fucose utilization. (A) Schematic representation of the chromosomal region associated with the Fuc+ phenotype in NCTC 11168. Solid arrows depict open reading frames. Vertical bars indicate locations of frameshift ...

To assess the effects of the ΔfucP::cat mutation on the transcription of downstream genes, an inoculum (OD600 of 1.0) was prepared for strains NCTC 11168 and CjWM226a (see above). One milliliter of this inoculum was added to two 18- by 150-mm tubes containing 10 ml MH broth and a 10- by 3-mm stir bar. To one tube for each strain, fucose was added to a final concentration of 25 mM. After 4 h of incubation at 37°C with stirring, RNA was extracted from 1.5 ml of each culture by use of an RNeasy minikit (Qiagen, Valencia, CA). RNA was DNA-free (Ambion, Austin, TX) treated and reverse transcribed using 3 pmol of primers qfucP_RT, cjr01_RT, pcj0487_R, pcj0488_R, and ald_R using SuperScript RT (Invitrogen) per the manufacturer's recommendations. The diluted (1:2) cDNA was used with primers qfucP_L and qfucP_RT, cjr01_L and cjr01_RT, pcj0487_L and pcj0487_R, tcj0488_L and pcj0488_R, or ald_L and ald_R (0.4 μM each primer) in separate PCR mixtures and cycled for 94°C for 3 min followed by 30 cycles of 95°C for 30 s, 51°C for 30 s, and 72°C for 30 s. PCR products were detected following electrophoresis through a 1% agarose gel and staining with ethidium bromide.

Reverse transcription-PCR (RT-PCR) analysis of gene induction by fucose.

To examine the expression of the fucose utilization gene cluster in the presence or absence of fucose, RNA was extracted from cells grown in defined medium containing fucose, pyruvate, or DEPS for 48 h using the RNeasy minikit. Nucleic acid (150 ng) was treated with DNA-free to remove residual DNA. Fifty nanograms of DNase I-treated RNA was reverse transcribed with 3 pmol of primers qfucP_RT, pdapA_R, ald_R, and cjr01_RT (Table (Table2)2) using SuperScript RT (Invitrogen) per the manufacturer's recommendations. SuperScript RT enzyme was omitted from control reaction mixtures. The resulting cDNA was diluted 1:2 and used as the template in separate PCR mixtures containing primers qfucP_RT and qfucP_L, ald_L and ald_R, pdapA_R and tdapA_L, or cjr01_RT and cjr01_L (0.4 μM each primer) and cycled as described above. PCR products were detected following electrophoresis through a 1% agarose gel and staining with ethidium bromide.

The relative fucP transcript abundance in wild-type NCTC 11168 was measured by SYBR green RT-PCR in batch cultures with or without fucose. Overnight growth on MH agar was harvested using MH broth to an OD600 of 1.0. One milliliter of this inoculum was transferred to 18- by 150-mm glass test tubes containing 10 ml of MH broth and a 10- by 3-mm stir bar and capped with a Morton closure. l-Fucose was serially diluted from a 5 M stock solution and added to each tube containing MH broth to a final concentration of 0.5 μM to 5 mM. After incubation at 37°C for 4 h, RNA was extracted from a 1.5-ml aliquot using the RNeasy minikit (Qiagen) according to the manufacturer's recommendation. RNA (150 ng) was treated with DNA-free (Ambion) to remove contaminating genomic DNA. cDNA was synthesized from 50 ng of DNA-free-treated RNA that was reverse transcribed with 3 pmol of gene-specific primers (qfucP_RT and cjr01_RT) using SuperScript RT (Invitrogen). Reaction mixtures in which the SuperScript RT enzyme was omitted served as controls. Amplification reaction mixtures (25 μl) consisted of 1× iQ SYBR green Supermix (Bio-Rad, Hercules, CA), 10 pmol each of forward and reverse primers, and 2 μl of 1:2-diluted cDNA in duplicate or triplicate technical replicates. Thermocycling parameters were 95°C for 3 min, followed by 35 cycles of 95°C for 30 s, 52°C for 30 s, and 72°C for 30 s, in a Bio-Rad iQ5 thermocycler. Fluorescence data were acquired just prior to extension at 72°C. Amplicon melt curve analysis was added to the end of the run that consisted of continuous fluorescence acquisition from 65°C to 95°C. Upon completion of the run, the second derivative of the melt curve was analyzed for spurious peaks and shoulders, indicative of nonspecific amplification. Additionally, peak symmetry was assessed with melting temperatures (Tm) of 82.3°C and 79.2°C for the 16S and fucP amplicons, respectively. Background-subtracted fluorescence values were exported as an Excel file (Microsoft, Redmond, WA) for analysis by LinRegPCR v11.1 software (39) using the program's baseline determination algorithm (41). fucP transcript abundance was normalized to that of the 16S transcript, and the PavrgECt method of efficiency averaging was applied to determine differences in transcript abundance (23). For statistical purposes, the ratios of induced/uninduced (Rinduced) values were log2 transformed to yield log2-fold differences. Assumptions of normality and variance were assessed by the Kolmogorov-Smirnov and modified Levene tests, respectively. A one-sample t test was applied to test the null hypothesis that Rinduced equaled 0.

PCR screen for fucP and ggt among field isolates of C. jejuni.

To measure the metabolic diversity among C. jejuni strains, we analyzed the distribution of fucP and ggt in C. jejuni isolates derived from various host species. The ggt gene encodes the γ-glutamyltranspeptidase protein, which has previously been described as being strain specific and conferring the ability to utilize glutamine and glutathione (18). C. jejuni strains isolated from humans, sheep, turkeys, and chickens were screened for the fucP and ggt genes by PCR. Selected strains were defined as having at least one band difference by SmaI or KpnI pulsed-field gel electrophoresis (PFGE) (40). The ggt gene was amplified by primers ggt_L and ggt_R (0.4 μM each) to produce a 285-bp product. The fucP gene was detected in a separate multiplex reaction using cj0486FWD and cj0486REV (14) and rpoC and rpsL primers (0.4 μM each). The cj0486FWD and cj0486REV primers target the fucP gene; thus, fucP-positive strains produce a 1.2-kb amplicon. The rpoC and rpsL primers anneal to the rpoC and rpsL genes, respectively; thus, fucP-negative strains produce a 446-bp amplicon with the rpoC and rpsL primers due to the absence of the 9.5-kb gene cluster. This allows for an unambiguous assessment of the presence/absence of fucP. All reaction mixtures were thermocycled using the parameters for fucP as described earlier (14). To associate the fucP-positive genotype with the Fuc+ phenotype, one fucP-positive and one fucP-negative strain representing each host species were grown in basal defined medium or defined medium supplemented with fucose (50 mM) or pyruvate (20 mM). The growth experiment for field isolates was performed once for each isolate. Viable cell numbers were determined after incubation for 24 h as described above. The ratio (R) of growth was calculated as the number of viable cells when grown with substrate divided by the number of viable cells in basal medium (e.g., Rfucose = viable cells when grown with fucose/viable cells in basal defined medium).

Chick colonization by wild-type and fucP mutant strains.

Both mono- and coinoculation strategies were employed to assess the role of fucose utilization in intestinal colonization. Day-old chicks (Ross × Cobb) were obtained from a commercial hatchery and determined to be free of C. jejuni by culture of cloacal swabs prior to inoculation. Chicks were randomly assigned to three groups. In experiment I, each bird in groups 1 and 2 (n = 10 for each group) was orally challenged with 3 × 108 CFU of either NCTC 11168 wild type or the fucP mutant. Birds in group 3 (n = 10) were challenged with a mixture (1:1) of the wild type and the mutant strain (~1.5 × 108 CFU of each strain). At 10 and 20 days postinoculation, five birds from each group were humanely sacrificed and the ceca removed. Contents of the ceca (~1 g) were squeezed into a 16- by 100-mm round-bottomed tube, and an appropriate volume of buffered peptone water (BPW) was added to achieve a 10% (wt/vol) suspension. The sample was homogenized by vortexing and serially diluted in BPW. Aliquots (100 μl) of the dilution were plated onto duplicate MH agar plates containing growth and selective supplements (SR0232E and SR0183E; Oxoid, Lenexa, KS) and incubated at 42°C for 48 h microaerobically. For the cocolonization group, aliquots (100 μl) were plated onto both supplemented MH agar and supplemented MH agar containing chloramphenicol (10 μg ml1) in duplicate. Chloramphenicol-resistant colonies represent the mutant fraction, whereas colonies on MH agar without chloramphenicol represent the total Campylobacter bacterial number (i.e., sensitive fraction = total fraction − resistant fraction). In experiment II, chicks were assigned to three groups (n = 16 per group). Each bird in groups 1 and 2 was inoculated with either the wild type or the mutant strain (1.5 × 105 CFU), while birds in group 3 served as the cocolonization group (~7.5 × 104 CFU of each strain). At 5 and 10 days postinoculation, eight birds from each group were sacrificed, the ceca were removed, and C. jejuni bacteria were enumerated as described above. For graphical and statistical purposes, the viable plate counts (CFU/g) were log10 transformed. Differences between strains were evaluated by a two-sample t test after assessing assumptions of normality and variance by D'Agostino Omnibus and modified Levene tests, respectively.


l-Fucose supports growth of C. jejuni NCTC 11168 in chemically defined medium.

The Phenotypic MicroArray was initially used to identify carbon substrates metabolized by strains NCTC 11168, 81116, and 81-176. The ability to reduce the tetrazolium dye in the presence of l-fucose, glycyl-l-glutamic acid, and glycyl-l-proline was a metabolic feature distinguishing the three strains (Fig. (Fig.2).2). Since tetrazolium reduction in the presence of fucose was evident only for strain NCTC 11168, abiotic reduction of the dye was unlikely, as had been suggested for other sugars (29). Independent confirmation of the Fuc+ phenotype was obtained using a chemically defined medium with different supplemented carbon sources (Fig. (Fig.3A).3A). Because formate had been described as a potent electron donor (27), ammonium formate was added to the defined medium as a respiration substrate. No growth was observed in the basal defined medium (BDM) (no additional carbon source added) for any of the strains tested, allowing for the detection of carbon utilization deficiencies. High cell densities (~108 CFU ml1) were obtained for all strains when pyruvate was added to the BDM. Furthermore, like pyruvate, the mixture of catabolizable amino acids (aspartate, glutamate, proline, and serine; DEPS) (16) supported high cell densities that were sustained throughout the experiment, suggestive of common metabolic pathways for all strains tested. A 3-log10 increase (P = 0.00) in viable cell numbers with fucose was evident for strain NCTC 11168 but not for 81116 or 81-176 at 24 h. Because NCTC 11168 grew in media containing at least 20 mM solutes, it is plausible that growth is due to an osmotic imbalance and is not substrate specific. To rule out this possibility, growth with an equimolar concentration of d-glucose was included in three subsequent experiments. Consistent with previous observations, growth on glucose produced no measurable increase in viable cell numbers and was similar to that in BDM (Fig. (Fig.3A)3A) (P = 0.08). After 48 h of incubation, viability decreased in both the BDM and the BDM supplemented with glucose compared to that at the start of the experiment. In contrast, growth with pyruvate and DEPS was sustained at high levels after 48 h. A slight reduction (~0.7 log10) in cell viability was observed for 48 h of growth on fucose; however, viability was still >1 log10 greater than that at the start of the experiment (P = 0.01). This reduction in cell viability may be attributed to exhaustion of the utilizable carbon source or accumulation of nonmetabolizable end products in the batch culture, prohibiting further growth.

FIG. 2.
Results of Phenotypic MicroArray carbon utilization plates (PM1) after incubation for 48 h with strain NCTC 11168, 81116, or 81-176. Wells indicating differential carbon utilization are circled. B4, l-fucose; G1, glycyl-l-glutamic acid; H1, glycyl-l-proline. ...
FIG. 3.
Growth of wild-type C. jejuni strains in defined medium supplemented with different carbon sources. (A) Basal medium was supplemented with d-glucose (50 mM), l-fucose (50 mM), sodium pyruvate (20 mM), or a combination of l-aspartate, l-glutamate, l-proline, ...

To determine whether fucose would sustain C. jejuni growth, cultures were serially passaged (1:100) at 24-h intervals in BDM containing fucose. In all three experiments, C. jejuni reached cell densities of ~7.5 log10 CFU ml1 after a 24-h grow-out period from the previous passage (Fig. (Fig.3B).3B). Additionally, the culture could be passaged in fucose-containing medium at least four times without losing viability, indicating that fucose is sufficient as a carbon and/or energy source to maintain viability.

Identification of the genetic locus associated with the Fuc+ phenotype.

No obvious homologs to the E. coli fucose catabolism genes fucR, fucI, fucA, and fucK were evident in any of the sequenced C. jejuni genomes; however, ORF cj0486 in the NCTC 11168 genome encodes a protein with significant homology (E value of 4e−64, 38% identity over 91% of the length of the protein) to the E. coli l-fucose:H+ symporter (fucP). This fucP homolog is a member of an 11-ORF cluster located between the rpoC and rpsL genes (Fig. (Fig.1A).1A). Significantly, this cluster is present in strain RM1221, which also displayed the ability to reduce tetrazolium in the presence of fucose (15), while absent in the genomes of the Fuc strains 81116 and 81-176 (Fig. (Fig.3A).3A). The C. jejuni chromosome contains regions that are highly variable or are differentially present among strains. These regions are considered plasticity regions (PR), and the 11-ORF cluster containing cj0486 has previously been characterized as PR2 (38). With the exception of the 11-ORF cluster, the flanking regions of PR2 in all four genomes (NCTC 11168, 81-176, 81116, and RM1221) display a high degree of similarity. Within the PR2 cluster are homologs to genes encoding an IclR-type transcriptional regulator (cj0480c), dipicolinate synthase (dapA), altronate dehydratase (uxaA), a probable membrane transport protein (cj0484), a possible oxidoreductase (cj0485), a putative fucose permease (fucP), aldehyde dehydrogenase (aldA), and two proteins of unknown function (cj0487 and cj0488). uxaA and aldA are predicted to be pseudogenes due to a frameshift mutation resulting in a premature stop codon in NCTC 11168 (36). Similarly, uxaA is predicted to be a pseudogene in RM1221 (35); however, aldA encodes a full-length protein. Partial sequencing of aldA and uxaA confirmed the presence of the predicted premature stop codon in the NCTC 11168 strain used in this study (data not shown).

The PR2 locus is associated with growth of C. jejuni on fucose.

To assess whether the fucP-containing PR2 locus is required for growth on fucose, an isogenic mutant was constructed by deleting an internal fragment (512 nucleotides) of fucP and replacing it with a chloramphenicol resistance cassette (cat) (Fig. (Fig.1A).1A). The cassette was confirmed by PCR to be in the same transcriptional orientation to fucP (data not shown). No growth of the mutant strain was observed when fucose was added to the defined medium (Fig. (Fig.4),4), indicating a defect of the mutant in fucose utilization. When pyruvate or DEPS was added to the medium, the mutant and wild-type strains demonstrated similar growth levels at 24 and 48 h, indicating that no other metabolic defects were evident for the mutant strain.

FIG. 4.
Growth comparison of various strains in defined medium supplemented with different carbon sources. White columns, NCTC 11168 wild type; gray columns, fucP derivative of NCTC 11168 (CjWM114a); black columns, ΔfucP::cat strain complemented with ...

To complement the growth defect of the fucP mutant, we first examined whether the fucP mutation affected the expression of its downstream genes in the PR2 locus. RT-PCR analysis revealed that disruption of fucP introduced a polar effect on the transcription of cj0487, cj0488, and aldA, because transcription of the three downstream genes was detected only in the wild-type strain but not in CjWM226a (partially shown in Fig. Fig.1B).1B). Thus, a sequential complementation strategy was employed, where fucP and one, two, or all three downstream genes were cloned into the fucP mutant background (strains CjWM230a, CjWM231a, and CjWM227a, respectively) (Fig. (Fig.1A).1A). fucP alone (strain CjWM226a) or fucP plus cj0487 (CjWM230a) did not restore the Fuc+ phenotype (Fig. (Fig.4);4); however, the Fuc+ phenotype was partially rescued in CjWM231a (complemented with fucP, cj0487, and cj0488) and CjWM227a (complemented with fucP, cj0487, cj0488, and aldA) (Fig. (Fig.4).4). This complementing effect was particularly evident after 48 h of incubation, where the mean number of viable cells was 2.5 log10 greater than that of the mutant strain when grown with fucose as a carbon source (P = 0.01). These results are consistent with the RT-PCR results that polar effects had been introduced by the mutation and suggest that multiple genes in the locus contribute to the Fuc+ phenotype.

fucP and other genes in the cluster are transcriptionally upregulated in the presence of fucose.

To determine if genes in PR2 are inducible by fucose, RT-PCR was used to qualitatively determine the transcription levels of dapA, aldA, and fucP in defined medium with different metabolizable substrates. RNA was extracted from NCTC 11168 cultures grown for 48 h in defined medium with fucose, pyruvate, or amino acids as the carbon source. dapA, aldA, and fucP transcripts were amplified from cells cultured with fucose but not from cells grown with pyruvate or amino acids (Fig. (Fig.5A).5A). Amplification of the 16S RNA transcript was evident for cells grown under all conditions (Fig. (Fig.5A).5A). No amplification of any transcript was detected when the SuperScript RT enzyme was withheld from the reaction mixture (control), indicating that genomic DNA contamination of the RNA is negligible. Furthermore, semiquantitative RT-PCR demonstrated a dose-dependent transcriptional response of fucP to fucose. The abundance of the fucP transcript was significantly greater than that for the uninduced culture at 50 μM fucose (Rinduced = 5.79; P = 0.025) (Fig. (Fig.5B)5B) and increased with increasing concentrations of fucose up to 5 mM (Rinduced = 13.99; P = 0.000).

FIG. 5.
RT-PCR analysis of fucose induction of gene expression in C. jejuni NCTC 11168. (A) Expression of dapA, fucP, and aldA in wild-type NCTC 11168 in defined medium supplemented with fucose (50 mM), pyruvate (20 mM), or a mixture of aspartate, glutamate, ...

Distribution of fucP among C. jejuni strains and its association with the Fuc+ phenotype.

To determine the prevalence of the fucP gene among C. jejuni strains isolated from different host species, a PCR screen was applied to strains isolated from humans, sheep, turkeys, and chickens. Additionally, the ggt gene was included in the screen, due to its association with the strain-specific utilization of glutamine and glutathione. Thus, this PCR assay detects two of the genetic elements that contribute to the known metabolic diversity among C. jejuni strains. The overall prevalence of fucP was 47%, and the gene was distributed with parity among host species (Table (Table3),3), suggesting that fucose utilization by Campylobacter is not associated exclusively with certain host species. For all strains tested, the lack of a fucP band was correlated with the presence of the rpoC rpsL amplicon, indicating that the fucP gene is gained or lost with the entire PR2. Overall, ggt was observed at a much lower frequency (24.4%) than fucP. When strains from each host species were analyzed separately, the ggt gene was represented in 47% of human strains, but in only 30%, 8%, and 17% of strains from turkey, chicken, and sheep sources, respectively. Interestingly, with two exceptions (C. jejuni subsp. doylei 269.97 and C. jejuni 1336), the fucP gene is mutually exclusive with the ggt gene (i.e., strains are not positive for both fucP ggt) (Table (Table3).3). Two strains (one fucP-negative and one fucP-positive strain) from each host species were grown in the defined medium to confirm the association between the Fuc+ phenotype and the presence of fucP. In all strains examined, the phenotype and genotype association was confirmed (Table (Table4),4), indicating that the presence of fucP (and other genes in PR2) predicts fucose utilization and that fucose metabolism is not limited to NCTC 11168.

Distribution of fucP and ggt among C. jejuni isolates from turkey, chicken, sheep, and human
Growth of selected field isolates from different host species in defined medium supplemented with pyruvate or fucosea

fucP is not required for colonization of the chick ceca.

We further examined whether fucose utilization is required for Campylobacter colonization in chickens. Prior to inoculation, all chicks were free of Campylobacter as determined by cloacal swab enrichments. In the high-dose experiment (experiment I), monoinoculated groups were challenged with either the wild-type NCTC 11168 strain or the fucP isogenic mutant at 108 CFU per bird. For the competitive colonization group, a 1:1 mixture of each strain was used to inoculate chicks. Birds were colonized at high densities throughout the 20-day experiment (range of 7.50 × 107 to 5.95 × 109 CFU/g). For the monoinoculated groups, the wild-type strain colonized birds at significantly higher levels than the fucP mutant at 10 days but not at 20 days postinoculation (P = 0.007 and P = 0.141, respectively) (Fig. (Fig.6A).6A). In contrast, the competition group yielded equivocal results at 10 days (P = 0.437), suggesting that no competitive advantage exists for either strain (Fig. (Fig.6A).6A). A slightly (P = 0.021) greater number of fucP mutant colonies than wild-type colonies was recovered at the 20-day sampling when strains were in competition.

FIG. 6.
Colonization of chicken ceca by wild-type C. jejuni NCTC 11168 (circles) and its fucP mutant (squares) in mono- and coinoculated groups. In two separate experiments, young chicks were inoculated with a high dose (108 CFU per chicken) (A) or a low dose ...

To further explore the fitness of the wild-type and fucP strains, a second colonization experiment was performed at a lower inoculum dose. For the low-dose experiment (experiment II), birds were challenged with 105 CFU per bird of the wild-type or mutant strain or a 1:1 mixture of each strain in competition. For the monoinoculated groups, higher numbers of the fucP strain than the wild-type strain were recovered from the ceca 5 and 10 days after inoculation (P = 0.052 and P = 0.018, respectively) (Fig. (Fig.6B).6B). In contrast, when in competition, the wild-type strain outcompeted the fucP strain at both 5 and 10 days postinoculation (P = 0.014 and P = 0.005, respectively) (Fig. (Fig.6B).6B). Taken together, the data suggest that fucose utilization may be advantageous when birds are exposed to low levels of Campylobacter but is not essential for chick colonization.


Contrary to the dogma that C. jejuni is an asaccharolytic bacterium, we provide evidence of strain-specific utilization of fucose, a prominent oligosaccharide component of mucin and other glycoproteins, to support growth. This conclusion is based on multiple lines of evidence, including the strain-specific reduction of a tetrazolium dye in the presence of fucose (Fig. (Fig.2),2), identification of the genetic locus responsible for the fucose phenotype (Fig. (Fig.1),1), growth of wild-type, mutant, and complemented strains in chemically defined medium (Fig. (Fig.4),4), association of the fucose phenotype with the fucP-positive genotype in various strains (Fig. (Fig.33 and Table Table4),4), and induction of gene expression in the presence of fucose (Fig. (Fig.5).5). These results have conclusively identified a previously unrecognized metabolic function in C. jejuni.

The unusual physiology of C. jejuni has contributed to the paucity of information regarding its metabolic capability and diversity, and relatively little is known about its growth substrates under various conditions. Amino acids and alternative electron acceptors play an important role during colonization (16, 44, 48). These metabolic features are common to the majority of C. jejuni strains and likely represent the organism's primary metabolism. The genetic diversity among strains of C. jejuni manifests in various secondary metabolic capabilities. For example, strains expressing γ-glutamyltranspeptidase are able to utilize glutathione and glutamine as carbon sources (18). Furthermore, asparagine can support the growth of strains that encode an exported form of asparaginase (AnsB) (18). These types of metabolic diversities may afford a competitive advantage to strains under certain conditions or environments. Indeed, the periplasmic form of AnsB allowed C. jejuni to colonize the liver at significantly higher levels. In this report, we describe fucose as another substrate that supports growth in a strain-specific manner.

The Phenotypic MicroArray provides a global perspective of cellular metabolism in response to a variety of growth conditions. Application of this technology identified d-ribose, l-lyxose, d-xylose, and l-arabinose as carbohydrates with which an abiotic reduction of tetrazolium was evident (Fig. (Fig.2,2, wells C4, H6, B8, and A2, respectively) (29). The response to these pentose sugars is likely an artifact of microaerobic incubation at an elevated temperature producing an abiotic reduction of the tetrazolium dye (7, 29). Others have reported tetrazolium reduction with arabinose but have also described fucose as a possible utilizable monosaccharide (8, 15). In this study, we identified this phenotype in strain NCTC 11168; however, the fucose phenotype (Fuc+) was not observed with strains 81116 and 81-176 (Fig. (Fig.2),2), suggesting that biological metabolism of this deoxyaldohexose is responsible for the color change. Independent confirmation of this distinguishing phenotype was obtained using a nutrient defined medium (Fig. (Fig.3A).3A). Furthermore, fucose was sufficient as a carbon and/or energy source to maintain growth and viability after four serial passages in the defined medium (Fig. (Fig.3B).3B). This report is the first description of a carbohydrate as a carbon and/or energy source sustaining C. jejuni growth.

Whole-genome microarrays estimate the genetic variability among C. jejuni strains to be 16.3% (38), and the strain-specific metabolism of fucose would likely be associated with genomic plasticity regions. Genomic plasticity regions (PR) are characterized as gene clusters that are hypervariable or vary in presence among strains. The genetic locus associated with the Fuc+ phenotype was localized to a 9.5-kb region in the chromosome of NCTC 11168 between two highly conserved genes (rpoC and rpsL). In Fuc strains (e.g., 81116 and 81-176), the 9.5-kb ORF cluster is absent, consistent with the characterization of this region of the chromosome as PR2 (38). Mutation of the predicted fucose permease gene (fucP) rendered the mutant strain unable to utilize fucose in defined medium. As demonstrated by RT-PCR (Fig. (Fig.1B),1B), the ΔfucP::cat mutation caused a polar effect on the transcription of genes downstream of fucP in the PR2 cluster. Thus, it is not surprising that the Fuc+ phenotype could not be restored by reconstituting the fucP gene alone. The phenotype was partially rescued by complementation of the fucP, cj0487, and cj0488 genes, suggesting that multiple genes in the PR2 gene cluster are necessary for the phenotype. The reason that full restoration of the phenotype was not achieved may be due to expression levels of the complemented genes. In the wild-type strain, gene expression is significantly induced in the presence of fucose; however, this increase in transcript abundance would not occur in the complemented strain, since the native promoter was not included in the construct. Alternatively, disruption of the fucP gene may destabilize the entire transcript, causing reduced expression of genes upstream of the mutation. Cj0487 shows homology to an amidohydrolase from Gardnerella vaginalis (E value of 7e−22), while Cj0488 displays similarity to the Bacillus subtilis l-rhamnose mutarotase (E value of 1e−05). The function of these genes and their contribution to the Fuc+ phenotype warrant further investigation, especially considering the lack of homologs to previously described fucose catabolism genes, as discussed below.

Because C. jejuni responds dynamically to environmental cues, it is possible that the genes necessary for fucose metabolism are differentially regulated depending on substrate availability. Qualitative RT-PCR revealed that fucP, dapA, and aldA transcripts were observed in the presence of fucose but not with pyruvate or amino acids (Fig. (Fig.5A).5A). Semiquantitative RT-PCR data estimated the fucP transcript abundance to be significantly increased in cultures with added fucose compared to that in uninduced cultures (Fig. (Fig.5B).5B). This response was dose dependent, suggestive of a transcriptional regulatory mechanism that was responsive to fucose. Interestingly, the first gene (cj0480c) in the 11-ORF cluster is predicted to encode a 253-amino-acid protein with similarities to a DNA binding protein of the IclR-type regulator family (36). IclR-type transcriptional regulators are known to be involved in the regulation of several metabolic processes and are capable of binding effector molecules which affect DNA binding (33). Whether Cj0480c is involved in regulation of fucose utilization in Campylobacter is being investigated in our laboratory.

Fucose utilization has been described for both pathogenic and commensal bacteria, including Salmonella enterica serovar Typhimurium LT2, Escherichia coli, Klebsiella pneumoniae, Roseburia inulinivorans, and Bacteroides thetaiotaomicron (4, 5, 9, 19, 43). Under both anaerobic and aerobic conditions, fucose is metabolized to dihydroxyacetone phosphate (DHAP) and l-lactaldehyde by the key enzymes fucose isomerase, fuculose kinase, and fuculose aldolase (FucI, FucK, and FucA, respectively) (9). Particularly enigmatic is the notable absence of FucI, FucK, and FucA homologs in the sequenced C. jejuni genomes. DHAP is metabolized to pyruvate through phosphoenolpyruvate in the lower portion of the Embden Meyerhof Parnas (EMP) pathway. Interestingly, the genome sequence of C. jejuni predicts that the genes necessary for the conversion of DHAP to pyruvate are intact (36), supporting Velayudhan and Kelly's hypothesis that the lower portion of the EMP pathway functions in a catabolic role for an unidentified substrate (49). Pyruvate kinase (pyk) plays an important function in the catabolic role of glycolysis, catalyzing the transfer of a phosphate group from PEP to ADP, yielding pyruvate and ATP. Whether carbons from fucose contribute to DHAP formation in Campylobacter is currently under investigation in our laboratory.

Under aerobic conditions, E. coli oxidizes lactaldehyde to lactate through the activity of aldehyde dehydrogenase (9). Within the 11-ORF cluster of the Fuc+ locus, an aldehyde dehydrogenase homolog is present; however, a frameshift mutation introduces a predicted premature stop codon that truncates the protein at the amino terminus, producing a 73-amino-acid N-terminal fragment and a 393-amino-acid C-terminal fragment (36). By sequencing the region containing the premature stop codon, we confirm that aldA and uxaA contain the predicted premature stop codons in the NCTC 11168 strain used in this study, although no functional enzymatic assays were performed to assess activity of the truncated proteins. In the absence of oxygen, lactaldehyde is reduced to 1,2-propanediol by propanediol oxidoreductase (5), which is either excreted (4) or fermented to propionate and propanol (5, 43). A BLAST query using the amino acid sequence of propanediol oxidoreductase from E. coli produced one significant hit to a hypothetical protein in strain 81116 (27% identity, E value of 5e−23) but no hits in the NCTC 11168 or other C. jejuni genomes. In addition, no homologs of fucI, fucK, or fucA genes were identified in C. jejuni NCTC 11168 when the E. coli orthologs were used in a BLAST query. Based on these homology searches, it appears that C. jejuni may catabolize fucose through an uncharacterized and potentially novel pathway. More work to elucidate the pathway of fucose utilization by C. jejuni is warranted, considering the apparent lack of known or previously identified enzymes necessary for fucose dissimilation.

Strain-specific secondary metabolism could confer an advantage in the colonization of different host species. To explore this possibility, strains from chicken, turkey, human, and sheep sources were screened for the fucP and ggt genes by PCR. Since all currently sequenced genomes position the PR2 gene cluster between rpoC and rpsL, the inclusion of rpoC and rpsL primers in the fucP multiplex reaction mixture ensures that allelic polymorphism at the primer binding site(s) does not produce a false-negative result. This strategy, however, does not preclude the possibility of a false negative due to the joint chance of allelic divergence at the primer binding site(s) and translocation of the gene due to intramolecular recombination. A previous study assessing the distribution of fucP among isolates of poultry and human origin estimated the prevalence of the gene at 64.5% (14). Here, we report the overall prevalence of fucP to be 47.7%, a slightly lower estimate than reported earlier (14). This discrepancy may be due to the sampling strategy employed to identify the isolate population. In this study, the use of PFGE-defined strains ensured that isolates were genetically distinct from one another and that a single clonal strain is not overrepresented in the sample population. No association between the fucP-positive genotype and host species examined in this study was observed. Additionally, the fucP gene is distributed with parity among human-derived strains originating from Asia, South Africa, the United States, and the United Kingdom. A more comprehensive sampling, however, is required to assess any spatial effects on the distribution of the fucP gene. With two exceptions (C. jejuni subsp. doylei 269.97 and C. jejuni 1336), the fucP and ggt genes appear to be mutually exclusive, which suggests vertical dissemination and divergence from a common ancestor. Alternatively, the two genes may be functionally incompatible in the same cell.

The selective pressures involved in the maintenance of the Fuc+ phenotype remain undefined but likely include nutrient availability. How C. jejuni obtains free fucose from its microenvironment remains unknown. The human commensal Bacteroides thetaiotaomicron induces intestinal cells of the host to produce more fucosylated glycans, directly affecting nutrient availability (19). Additionally, patients infected with C. jejuni expressed Muc1 at higher levels than uninfected cohorts (28); thus, C. jejuni may also be capable of inducing host intestinal cells to produce fucosylated glycans. Several bacteria secrete fucosidase, which cleaves the terminal fucose from the glycosylated protein, producing free fucose which can subsequently be taken up. Although no fucosidase homolog was identified in the C. jejuni genome, it is plausible that intestinal commensals produce the enzyme to cleave the fucose from the host glycoconjugate. Alternatively, Helicobacter pylori, a close relative of Campylobacter, has been shown to direct host cells to produce α-l-fucosidase, freeing the fucose from the glycoconjugate directly (30). Secreted mucin has been suggested to increase the viability of C. jejuni (1), indicating that C. jejuni is capable of directly degrading or inducing the host to degrade mucin into components that support growth. The structure of the fucosylated glycan may vary among species, due to differences in the fucosyltransferases that modify the protein. Differences in the glycan structure may confer a species adaptation of Fuc+ strains, although our survey of chicken, turkey, sheep, and human strains did not identify the fucP gene as overly represented in any particular host examined in this study.

The chick colonization model was used to assess the colonization potential and competitive fitness of the wild-type strain and an isogenic Fuc mutant. Birds inoculated with the mutant strain displayed high C. jejuni densities in the ceca, indicating that fucose utilization is not essential for chicken colonization. This is consistent with the proportion and isolation of Fuc strains from chickens and turkeys. The competition model of colonization yielded equivocal results, depending on the inoculation dose and duration of colonization. These results suggest that fucose utilization may be advantageous but is not essential for the colonization of the chick intestine. This is likely due to the presence of amino acids in the chicken intestinal tract, which are preferred substrates for Campylobacter metabolism, overshadowing the role of fucose utilization in colonization. However, it may be possible that Fuc+ strains of C. jejuni have a competitive advantage in another host species due to variation in nutrient availability (e.g., amino acids), microbiome, and structural differences of host mucin. Interestingly, transposon mutagenesis identified fucP as an invasion determinant, since transposon insertion in this gene led to reduced invasion of cultured cells (22). How fucP contributes to the invasion process is unknown, but the finding illustrates the functional diversity of this gene.

The findings reported here identified a carbohydrate substrate that supports C. jejuni growth and formally linked this phenotype to the Fuc+ locus localized to a plasticity region of the Campylobacter chromosome. The Fuc+ phenotype likely involves adaptation to nutrient-poor niches where utilization of this carbon and/or energy source affords the organism a competitive advantage. How C. jejuni metabolizes fucose and how this metabolic pathway interacts with other processes in facilitating Campylobacter adaptation remain to be elucidated in future studies.


We are grateful to Orhan Sahin for his help with the animal work. We thank Thad Stanton and Tom Casey for insightful discussions and critical readings of the manuscript and Irene Wesley for providing research materials for this work.

Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.

This work was supported by National Institutes of Health grant RO1DK063008 and National Research Initiative Competitive Grants Program grant no. 2007-35201-18278 from the USDA National Institute of Food and Agriculture.


[down-pointing small open triangle]Published ahead of print on 30 December 2010.


1. Alemka, A., et al. 2010. Probiotic colonization of mucus-adherent HT29MTXE12 cells attenuates Campylobacter jejuni virulence properties. Infect. Immun. 78:2812-2822. [PMC free article] [PubMed]
2. Alexander, J. K. 1957. Energy sources utilized by Vibrio fetus. J. Bacteriol. 74:168-170. [PMC free article] [PubMed]
3. Allen, K. J., and M. W. Griffiths. 2001. Effect of environmental and chemotactic stimuli on the activity of the Campylobacter jejuni flaA sigma(28) promoter. FEMS Microbiol. Lett. 205:43-48. [PubMed]
4. Badia, J., J. Ros, and J. Aguilar. 1985. Fermentation mechanism of fucose and rhamnose in Salmonella typhimurium and Klebsiella pneumoniae. J. Bacteriol. 161:435-437. [PMC free article] [PubMed]
5. Baldoma, L., and J. Aguilar. 1988. Metabolism of L-fucose and L-rhamnose in Escherichia coli: aerobic-anaerobic regulation of L-lactaldehyde dissimilation. J. Bacteriol. 170:416-421. [PMC free article] [PubMed]
6. Black, R. E., M. M. Levine, M. L. Clements, T. P. Hughes, and M. J. Blaser. 1988. Experimental Campylobacter jejuni infection in humans. J. Infect. Dis. 157:472-479. [PubMed]
7. Bochner, B. R. 2009. Global phenotypic characterization of bacteria. FEMS Microbiol. Rev. 33:191-205. [PMC free article] [PubMed]
8. Brandl, M. T., A. F. Haxo, A. H. Bates, and R. E. Mandrell. 2004. Comparison of survival of Campylobacter jejuni in the phyllosphere with that in the rhizosphere of spinach and radish plants. Appl. Environ. Microbiol. 70:1182-1189. [PMC free article] [PubMed]
9. Chen, Y. M., Y. Zhu, and E. C. Lin. 1987. NAD-linked aldehyde dehydrogenase for aerobic utilization of L-fucose and L-rhamnose by Escherichia coli. J. Bacteriol. 169:3289-3294. [PMC free article] [PubMed]
10. Cinco, M., E. Banfi, E. Ruaro, D. Crevatin, and D. Crotti. 1984. Evidence for L-fucose (6-deoxy-l-galactopyranose)-mediated adherence of Campylobacter spp. to epithelial cells. FEMS Microbiol. Lett. 21:347-351.
11. Day, C. J., et al. 2009. Differential carbohydrate recognition by Campylobacter jejuni strain 11168: influences of temperature and growth conditions. PLoS One 4:e4927. [PMC free article] [PubMed]
12. Dickgiesser, N., and D. Czylwik. 1985. Chemically defined media for auxotyping of Campylobacter jejuni. Zentralbl. Bakteriol. Mikrobiol. Hyg. A 260:57-64. [PubMed]
13. Elharrif, Z., and F. Megraud. 1986. Characterization of thermophilic Campylobacter. I. Carbon-substrate utilization tests. Curr. Microbiol. 13:117-122.
14. Fearnley, C., et al. 2008. Identification of hyperinvasive Campylobacter jejuni strains isolated from poultry and human clinical sources. J. Med. Microbiol. 57:570-580. [PubMed]
15. Fouts, D. E., et al. 2005. Major structural differences and novel potential virulence mechanisms from the genomes of multiple campylobacter species. PLoS Biol. 3:e15. [PMC free article] [PubMed]
16. Guccione, E., et al. 2008. Amino acid-dependent growth of Campylobacter jejuni: key roles for aspartase (AspA) under microaerobic and oxygen-limited conditions and identification of AspB (Cj0762), essential for growth on glutamate. Mol. Microbiol. 69:77-93. [PubMed]
17. Hoffman, P. S., and T. G. Goodman. 1982. Respiratory physiology and energy conservation efficiency of Campylobacter jejuni. J. Bacteriol. 150:319-326. [PMC free article] [PubMed]
18. Hofreuter, D., V. Novik, and J. E. Galan. 2008. Metabolic diversity in Campylobacter jejuni enhances specific tissue colonization. Cell Host Microbe 4:425-433. [PubMed]
19. Hooper, L. V., J. Xu, P. G. Falk, T. Midtvedt, and J. I. Gordon. 1999. A molecular sensor that allows a gut commensal to control its nutrient foundation in a competitive ecosystem. Proc. Natl. Acad. Sci. U. S. A. 96:9833-9838. [PMC free article] [PubMed]
20. Hugdahl, M. B., J. T. Beery, and M. P. Doyle. 1988. Chemotactic behavior of Campylobacter jejuni. Infect. Immun. 56:1560-1566. [PMC free article] [PubMed]
21. Humphrey, T., S. O'Brien, and M. Madsen. 2007. Campylobacters as zoonotic pathogens: a food production perspective. Int. J. Food Microbiol. 117:237-257. [PubMed]
22. Javed, M. A., et al. 2010. Transposon mutagenesis in a hyper-invasive clinical isolate of Campylobacter jejuni reveals a number of genes with potential roles in invasion. Microbiology 156:1134-1143. [PubMed]
23. Karlen, Y., A. McNair, S. Perseguers, C. Mazza, and N. Mermod. 2007. Statistical significance of quantitative PCR. BMC Bioinformatics 8:131. [PMC free article] [PubMed]
24. Karlyshev, A. V., and B. W. Wren. 2005. Development and application of an insertional system for gene delivery and expression in Campylobacter jejuni. Appl. Environ. Microbiol. 71:4004-4013. [PMC free article] [PubMed]
25. Kiggins, E. M., and W. N. Plastridge. 1958. Some metabolic activities of Vibrio fetus of bovine origin. J. Bacteriol. 75:205-208. [PMC free article] [PubMed]
26. Leach, S., P. Harvey, and R. Wali. 1997. Changes with growth rate in the membrane lipid composition of and amino acid utilization by continuous cultures of Campylobacter jejuni. J. Appl. Microbiol. 82:631-640. [PubMed]
27. Lecce, J. G. 1958. Some biochemical characteristics of Vibrio fetus and other related vibrios isolated from animals. J. Bacteriol. 76:312-316. [PMC free article] [PubMed]
28. Linden, S. K., T. H. Florin, and M. A. McGuckin. 2008. Mucin dynamics in intestinal bacterial infection. PLoS One 3:e3952. [PMC free article] [PubMed]
29. Line, J. E., K. L. Hiett, J. Guard-Bouldin, and B. S. Seal. 2010. Differential carbon source utilization by Campylobacter jejuni 11168 in response to growth temperature variation. J. Microbiol. Methods 80:198-202. [PubMed]
30. Liu, T. W., et al. 2009. Role for alpha-L-fucosidase in the control of Helicobacter pylori-infected gastric cancer cells. Proc. Natl. Acad. Sci. U. S. A. 106:14581-14586. [PMC free article] [PubMed]
31. McAuley, J. L., et al. 2007. MUC1 cell surface mucin is a critical element of the mucosal barrier to infection. J. Clin. Invest. 117:2313-2324. [PMC free article] [PubMed]
32. Mohammed, K. A., R. J. Miles, and M. A. Halablab. 2004. The pattern and kinetics of substrate metabolism of Campylobacter jejuni and Campylobacter coli. Lett. Appl. Microbiol. 39:261-266. [PubMed]
33. Molina-Henares, A. J., T. Krell, G. M. Eugenia, A. Segura, and J. L. Ramos. 2006. Members of the IclR family of bacterial transcriptional regulators function as activators and/or repressors. FEMS Microbiol. Rev. 30:157-186. [PubMed]
34. Olson, C. K., S. Ethelber, W. van Pelt, and R. V. Tauxe. 2008. Epidemiology of Campylobacter jejuni infections in industrialized nations, p. 163-189. In I. Nachamkin, C. Szymanski, and M. J. Blaser (ed.), Campylobacter. ASM Press, Washington, DC.
35. Parker, C. T., B. Quinones, W. G. Miller, S. T. Horn, and R. E. Mandrell. 2006. Comparative genomic analysis of Campylobacter jejuni strains reveals diversity due to genomic elements similar to those present in C. jejuni strain RM1221. J. Clin. Microbiol. 44:4125-4135. [PMC free article] [PubMed]
36. Parkhill, J., et al. 2000. The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403:665-668. [PubMed]
37. Pearson, B. M., et al. 2007. The complete genome sequence of Campylobacter jejuni strain 81116 (NCTC11828). J. Bacteriol. 189:8402-8403. [PMC free article] [PubMed]
38. Pearson, B. M., et al. 2003. Comparative genome analysis of Campylobacter jejuni using whole genome DNA microarrays. FEBS Lett. 554:224-230. [PubMed]
39. Ramakers, C., J. M. Ruijter, R. H. Deprez, and A. F. Moorman. 2003. Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci. Lett. 339:62-66. [PubMed]
40. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. J. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J. Clin. Microbiol. 39:1889-1894. [PMC free article] [PubMed]
41. Ruijter, J. M., et al. 2009. Amplification efficiency: linking baseline and bias in the analysis of quantitative PCR data. Nucleic Acids Res. 37:e45. [PMC free article] [PubMed]
42. Ruiz-Palacios, G. M., L. E. Cervantes, P. Ramos, B. Chavez-Munguia, and D. S. Newburg. 2003. Campylobacter jejuni binds intestinal H(O) antigen (Fuc alpha 1, 2Gal beta 1, 4GlcNAc), and fucosyloligosaccharides of human milk inhibit its binding and infection. J. Biol. Chem. 278:14112-14120. [PubMed]
43. Scott, K. P., J. C. Martin, G. Campbell, C. D. Mayer, and H. J. Flint. 2006. Whole-genome transcription profiling reveals genes up-regulated by growth on fucose in the human gut bacterium “Roseburia inulinivorans.” J. Bacteriol. 188:4340-4349. [PMC free article] [PubMed]
44. Sellars, M. J., S. J. Hall, and D. J. Kelly. 2002. Growth of Campylobacter jejuni supported by respiration of fumarate, nitrate, nitrite, trimethylamine-N-oxide, or dimethyl sulfoxide requires oxygen. J. Bacteriol. 184:4187-4196. [PMC free article] [PubMed]
45. Tu, Q. V., M. A. McGuckin, and G. L. Mendz. 2008. Campylobacter jejuni response to human mucin MUC2: modulation of colonization and pathogenicity determinants. J. Med. Microbiol. 57:795-802. [PubMed]
46. Vandamme, P., F. E. Dewirst, B. J. Paster, and S. L. On. 2005. Genus I. Campylobacter, p. 1147-1161. In D. J. Brenner et al. (ed.), Bergey's manual of systematic bacteriology, 2nd ed. Springer, New York, NY.
47. van Vliet, A. H., A. C. Wood, J. Henderson, K. Wooldridge, and J. Ketley. 1998. Genetic manipulation of enteric Campylobacter species, p. 407-419. In P. Williams et al. (ed.), Methods in microbiology, vol. 27. Academic Press Ltd., London, United Kingdom.
48. Velayudhan, J., M. A. Jones, P. A. Barrow, and D. J. Kelly. 2004. L-serine catabolism via an oxygen-labile L-serine dehydratase is essential for colonization of the avian gut by Campylobacter jejuni. Infect. Immun. 72:260-268. [PMC free article] [PubMed]
49. Velayudhan, J., and D. J. Kelly. 2002. Analysis of gluconeogenic and anaplerotic enzymes in Campylobacter jejuni: an essential role for phosphoenolpyruvate carboxykinase. Microbiology 148:685-694. [PubMed]
50. Wang, Y., and D. E. Taylor. 1990. Chloramphenicol resistance in Campylobacter coli: nucleotide sequence, expression, and cloning vector construction. Gene 94:23-28. [PubMed]
51. Wosten, M. M., M. Boeve, M. G. Koot, A. C. van Nuenen, and B. A. van der Zeijst. 1998. Identification of Campylobacter jejuni promoter sequences. J. Bacteriol. 180:594-599. [PMC free article] [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...