Quantification of the relative roles of niche and neutral processes in structuring gastrointestinal microbiomes.

The theoretical description of the forces that shape ecological communities focuses around two classes of models. In niche theory, deterministic interactions between species, individuals, and the environment are considered the dominant factor, whereas in neutral theory, stochastic forces, such as demographic noise, speciation, and immigration, are dominant. Species abundance distributions predicted by the two classes of theory are difficult to distinguish empirically, making it problematic to deduce ecological dynamics from typical measures of diversity and community structure. Here, we show that the fusion of species abundance data with genome-derived measures of evolutionary distance can provide a clear indication of ecological dynamics, capable of quantifying the relative roles played by niche and neutral forces. We apply this technique to six gastrointestinal microbiomes drawn from three different domesticated vertebrates, using high-resolution surveys of microbial species abundance obtained from carefully curated deep 16S rRNA hypervariable tag sequencing data. Although the species abundance patterns are seemingly well fit by the neutral theory of metacommunity assembly, we show that this theory cannot account for the evolutionary patterns in the genomic data; moreover, our analyses strongly suggest that these microbiomes have, in fact, been assembled through processes that involve a significant nonneutral (niche) contribution. Our results demonstrate that high-resolution genomics can remove the ambiguities of process inference inherent in classic ecological measures and permits quantification of the forces shaping complex microbial communities.

Robust computational analysis of rRNA hypervariable tag datasets.

Next-generation DNA sequencing is increasingly being utilized to probe microbial communities, such as gastrointestinal microbiomes, where it is important to be able to quantify measures of abundance and diversity. The fragmented nature of the 16S rRNA datasets obtained, coupled with their unprecedented size, has led to the recognition that the results of such analyses are potentially contaminated by a variety of artifacts, both experimental and computational. Here we quantify how multiple alignment and clustering errors contribute to overestimates of abundance and diversity, reflected by incorrect OTU assignment, corrupted phylogenies, inaccurate species diversity estimators, and rank abundance distribution functions. We show that straightforward procedural optimizations, combining preexisting tools, are effective in handling large (10(5)-10(6)) 16S rRNA datasets, and we describe metrics to measure the effectiveness and quality of the estimators obtained. We introduce two metrics to ascertain the quality of clustering of pyrosequenced rRNA data, and show that complete linkage clustering greatly outperforms other widely used methods.

Examination of Campylobacter jejuni putative adhesins leads to the identification of a new protein, designated FlpA, required for chicken colonization.

Campylobacter jejuni colonization of chickens is presumably dependent upon multiple surface-exposed proteins termed adhesins. Putative C. jejuni adhesins include CadF, CapA, JlpA, major outer membrane protein, PEB1, Cj1279c, and Cj1349c. We examined the genetic relatedness of 97 C. jejuni isolates recovered from human, poultry, bovine, porcine, ovine, and canine sources by multilocus sequence typing (MLST) and examined their profile of putative adhesin-encoding genes by dot blot hybridization. To assess the individual contribution of each protein in bacterium-host cell adherence, the C. jejuni genes encoding the putative adhesins were disrupted by insertional mutagenesis. The phenotype of each mutant was judged by performing in vitro cell adherence assays with chicken LMH hepatocellular carcinoma epithelial cells and in vivo colonization assays with broiler chicks. MLST analysis indicated that the C. jejuni isolates utilized in this study were genetically diverse. Dot blot hybridization revealed that the C. jejuni genes encoding the putative adhesins, with the exception of capA, were conserved among the isolates. The C. jejuni CadF, CapA, Cj1279c, and Cj1349c proteins were found to play a significant role in the bacterium's in vitro adherence to chicken epithelial cells, while CadF, PEB1, and Cj1279c were determined to play a significant role in the bacterium's in vivo colonization of broiler chicks. Collectively, the data indicate that Cj1279c is a novel adhesin. Because Cj1279c harbors fibronectin type III domains, we designated the protein FlpA, for fibronectin-like protein A.

Campylobacter jejuni invade chicken LMH cells inefficiently and stimulate differential expression of the chicken CXCLi1 and CXCLi2 cytokines.

Campylobacter jejuni is a major food-borne bacterial pathogen, which is capable of causing diarrhoea containing blood and leukocytes. C. jejuni invasion of the intestinal epithelial cells and the release of proinflammatory molecules contribute to the pathophysiology of campylobacteriosis. Given the commensal relationship of C. jejuni with chickens, we hypothesized that C. jejuni invasion of chicken cells and the release of host cell cytokines would be significantly less than with human cells. To test our hypothesis, we examined the interactions of C. jejuni with chicken LMH cells, and performed in vivo experiments with chickens. The binding and internalization assays revealed that C. jejuni was significantly less invasive of LMH cells relative to human INT 407 cells, even though the bacteria bound to each host cell species equally. We also assessed interleukin-8 (IL-8) transcript, IL-8 secretion, and the release of chemoattractant molecules from the inoculated cells. Inoculation of LMH cells with C. jejuni stimulated expression of both chicken IL-8 orthologues, chCXCLi2 and chCXCLi1, but at levels significantly less than human IL-8 (huCXCL8) expressed from human INT 407 cells inoculated with C. jejuni. Moreover, the supernatant fluids of the C. jejuni-inoculated LMH cells resulted in little heterophil migration. In vivo, C. jejuni were observed bound to the cells lining the glandular crypts, but overt signs of cell invasion or pathology were not observed. These results indicate that cytokine expression in chicken LMH cells in response to C. jejuni is distinct from that of Salmonella typhimurium.

Campylobacter jejuni strains compete for colonization in broiler chicks.

Campylobacter jejuni isolates possess multiple adhesive proteins termed adhesins, which promote the organism's attachment to epithelial cells. Based on the proposal that one or more adhesins are shared among C. jejuni isolates, we hypothesized that C. jejuni strains would compete for intestinal and cecal colonization in broiler chicks. To test this hypothesis, we selected two C. jejuni strains with unique SmaI pulsed-field gel electrophoresis macrorestriction profiles and generated one nalidixic acid-resistant strain (the F38011 Nal(r) strain) and one streptomycin-resistant strain (the 02-833L Str(r) strain). In vitro binding assays revealed that the C. jejuni F38011 Nal(r) and 02-833L Str(r) strains adhered to LMH chicken hepatocellular carcinoma epithelial cells and that neither strain influenced the binding potential of the other strain at low inoculation doses. However, an increase in the dose of the C. jejuni 02-833L Str(r) strain relative to that of the C. jejuni F38011 Nal(r) strain competitively inhibited the binding of the C. jejuni F38011 Nal(r) strain to LMH cells in a dose-dependent fashion. Similarly, the C. jejuni 02-833L Str(r) strain was found to significantly reduce the efficiency of intestinal and cecal colonization by the C. jejuni F38011 Nal(r) strain in broiler chickens. Based on the number of bacteria recovered from the ceca, the maximum number of bacteria that can colonize the digestive tracts of chickens may be limited by host constraints. Collectively, these data support the hypothesis that C. jejuni strains compete for colonization in chicks and suggest that it may be possible to design novel intervention strategies for reducing the level at which C. jejuni colonizes the cecum.

Campylobacter jejuni infection in broiler chickens.

Day-old, straight-run broiler chickens were procured from a hatchery located in the Pacific Northwest. The chickens were subdivided individually into nine groups of 20 chickens. The chickens were tagged, housed in isolation chambers on wire, fed commercial broiler feed, and given water ad libitum. Three isolates of Campylobacter jejuni of poultry origin and one of human origin were tested in this study. Various C. jejuni cultures were inoculated into 9-day-old chickens by crop gavage. Four groups of 20 chickens were inoculated at a dose level of 0.5 ml of 1 x 10(2) colony-forming units (CFU)/ml. The other four groups were inoculated with 0.5 ml of 1 X 10(4) CFU/ml. One group of 20 chickens was kept as an uninoculated control group. Four randomly selected chickens from each of the inoculated and uninoculated groups were necropsied at 5, 12, and 19 days postinoculation (DPI). The C. jejuni was cultured and enumerated from a composite of the upper and midintestine and the cecum. Body weights of all chicken groups at 7 days of age and at 5, 12, and 19 DPI were measured and statistically analyzed. No significant differences were present in the mean body weights (MBWs) of 7-day-old, 5 DPI, and 12 DPI male and female broiler chickens inoculated with C. jejuni at both dose levels compared with uninoculated controls. Differences in MBWs of the male and female broilers at 19 DPI were observed in some of the groups. Results of the C. jejuni culture enumeration mean (CEM) of composite intestine samples at 5 DPI from all inoculated chicken groups, irrespective of the dose level, ranged from (2.5 +/- 5.0) x 10(2) to (2.8 +/- 4.8) x 10(5) CFU/g (mean +/- SD). Results of cecum C. jejuni CEM at 5 DPI inoculated at both dose levels ranged from (2.5 +/- 5.0) x 10(6) to (1 +/- 0.0) x 10(7) CFU/g in all treatment groups irrespective of the dose level. CEM results from the composite intestine samples at 12 and 19 DPI increased by 1 log unit, or sometimes more. Results of cecum C. jejuni CEM at 5 DPI inoculated at both dose levels ranged from (2.5 +/- 5.0) x 10(6) to (1 +/- 0.0) x 10(7) CFU/g in all treatment groups irrespective of the dose level. Increases of 2-5 log units in C. jejuni CEM was present in chicken groups inoculated with 1 X 10(2) CFU of C. jejuni, and a 2- to 3-log increase was present in groups inoculated with a higher dose level of C. jejuni at 12 DPI. The results of C. jejuni CEM from cecal samples at 19 DPI were similar to chicken groups at 12 DPI. Campylobacterjejuni was not isolated from the uninoculated control chickens at 5, 12, and 19 DPI. Clinical signs of illness or gross pathologic lesions were not present in any of the chicken groups during this study. No lesions were present on histopathologic evaluations in C. jejuni-inoculated chickens or uninoculated control chickens.