We examined two variants of the genome-sequenced strain, Campylobacter jejuni NCTC11168, which show marked differences in their virulence properties including colonization of poultry, invasion of Caco-2 cells, and motility. Transcript profiles obtained from whole genome DNA microarrays and proteome analyses demonstrated that these differences are reflected in late flagellar structural components and in virulence factors including those involved in flagellar glycosylation and cytolethal distending toxin production. We identified putative 28 and 54 promoters for many of the affected genes and found that greater differences in expression were observed for 28 -controlled genes. Inactivation of the gene encoding 28 , fliA, resulted in an unexpected increase in transcripts with 54 promoters, as well as decreased transcription of 28 -regulated genes. This was unlike the transcription profile observed for the attenuated C. jejuni variant, suggesting that the reduced virulence of this organism was not entirely due to impaired function of 28 . However, inactivation of flhA, an important component of the flagellar export apparatus, resulted in expression patterns similar to that of the attenuated variant. These findings indicate that the flagellar regulatory system plays an important role in campylobacter pathogenesis and that flhA is a key element involved in the coordinate regulation of late flagellar genes and of virulence factors in C. jejuni.

We have used comparative genomic hybridization (CGH) on a full-genome Campylobacter jejuni microarray to examine genome-wide gene conservation patterns among 51 strains isolated from food and clinical sources. These data have been integrated with data from three previous C. jejuni CGH studies to perform a metaanalysis that included 97 strains from the four separate data sets. Although many genes were found to be divergent across multiple strains (n ‫؍‬ 350), many genes (n ‫؍‬ 249) were uniquely variable in single strains. Thus, the strains in each data set comprise strains with a unique genetic diversity not found in the strains in the other data sets. Despite the large increase in the collective number of variable C. jejuni genes (n ‫؍‬ 599) found in the meta-analysis data set, nearly half of these (n ‫؍‬ 276) mapped to previously defined variable loci, and it therefore appears that large regions of the C. jejuni genome are genetically stable. A detailed analysis of the microarray data revealed that divergent genes could be differentiated on the basis of the amplitudes of their differential microarray signals. Of 599 variable genes, 122 could be classified as highly divergent on the basis of CGH data. Nearly all highly divergent genes (117 of 122) had divergent neighbors and showed high levels of intraspecies variability. The approach outlined here has enabled us to distinguish global trends of gene conservation in C. jejuni and has enabled us to define this group of genes as a robust set of variable markers that can become the cornerstone of a new generation of genotyping methods that use genome-wide C. jejuni gene variability data.Campylobacter jejuni is a human pathogen, a commensal inhabitant of many domestic animals, and globally, the most common cause of acute bacterial enteritis (for a review, see reference 31). Two well-established serotyping methods, namely, Penner typing based on heat-stable antigens and Lior typing based on heat-labile antigens, have been in use for more than two decades to study species diversity, to track epidemiological trends, and to determine important epidemiological correlations (15,23). Technical limitations on the production of high-quality typing sera have limited the availability of these reagents. Culturing conditions can affect the expression of serotyping determinants, which affects serotyping results, and several strains are nontypeable (32). Additionally, serotype relatedness is not always indicative of genetic relatedness since members of different serotypes of C. jejuni are genetically related, despite differences in heat-stable antigen expression (16).The need for alternative subtyping schemes has been recognized, leading to the development of a number of different methods based on differences at the DNA level (i.e., genotyping). The techniques used at present range from analysis of polymorphisms in groups of housekeeping genes (multilocus sequence typing [5,26]), amplified fragment length polymorphism analysis (28), restriction fragment length polymorphism analysi...

Activity screening and insertional inactivation of lipopolysaccharide (LPS) biosynthetic genes in Helicobacter pylori have led to the successful characterization of two key enzymes encoded by HP0159 (JHP0147) and HP1105 (JHP1032) open reading frames (ORFs) which are members of the large and diverse carbohydrate active enzymes (CAZY) GT-8 (rfaJ) family of glycosyltransferases. Activity screening of a genomic library led to the identification of the enzyme involved in the biosynthesis of the type 2 N-acetyl-lactosamine O-chain backbone, the beta-1,3-N-acetyl-glucosaminyl transferase. In addition, the activity screening approach led to the identification and characterization of a key core biosynthetic enzyme responsible for the biosynthesis of the alpha-1,6-glucan polymer. This alpha-1,6-glucosyltransferase protein is encoded by the HP0159 ORF. Both enzymes play an integral part in the biosynthesis of LPS, and insertional inactivation leads to the production of a truncated LPS molecule on the bacterial cell surface. The LPS structures were determined by mass spectrometry and chemical analyses. The linkage specificity of each glycosyltransferase was determined by nuclear magnetic resonance (NMR) analysis of model compounds synthesized in vitro. A cryogenic probe was used to structurally characterize nanomole amounts of the product of the HP1105 (JHP1032) enzyme. In contrast to the HP0159 enzyme, which displays the GT-8-predicted retaining stereochemistry for the reaction product, HP1105 (JHP1032) is the first member of this GT-8 family to have been shown to have an inverting stereochemistry in its reaction products.

Regardless of the dose of Ag, Ag presentation occurs rapidly within the first few days which results in rapid expansion of the CD8+ T cell response that peaks at day 7. However, we have previously shown that this rapid priming of CD8+ T cells is absent during infection of mice with Mycobacterium bovis (bacillus Calmette-Guérin (BCG)). In this study, we have evaluated the mechanisms responsible for the delayed CD8+ T cell priming. Because BCG replicates poorly and survives within phagosomes we considered whether 1) generation of reduced amounts of Ag or 2) weaker activation by pathogen-associated molecular patterns (PAMPs) during BCG infection is responsible for the delay in CD8+ T cell priming. Using rOVA-expressing bacteria, our results indicate that infection of mice with BCG-OVA generates greatly reduced levels of OVA, which are 70-fold lower in comparison to the levels generated during infection of mice with Listeria monocytogenes-expressing OVA. Furthermore, increasing the dose of OVA, but not PAMP signaling during BCG-OVA infection resulted in rapid Ag presentation and consequent expansion of the CD8+ T cell response, indicating that the generation of reduced Ag levels, not lack of PAMP-associated inflammation, was responsible for delayed priming of CD8+ T cells. There was a strong correlation between the relative timing of Ag presentation and the increase in the level of OVA in vivo. Taken together, these results reveal that some slowly replicating pathogens, such as mycobacteria, may facilitate their chronicity by generating reduced Ag levels which causes a substantial delay in the development of acquired immune responses.