Recently, we reported the use of a low-density cDNA chicken immune-specific microarray to investigate the host immune response to Cp infection by examining gene expression profiles in spleens from experimentally infected chickens . Our results indicated that a more than one immune response pathway was targeted, including the differential expression of genes within the MHC class I and II and apoptosis pathways. Very recently, a 44 K chicken whole genome custom oligo array was developed, which was manufactured with the Agilent Technology . This allowed us to expand our research by compensating for the size limitation of the previous array and thus to more robustly investigate the host response to Cp infection. Importantly, the genes that were previously identified by our group as functionally relevant in the spleen of Cp-challenged chickens (e.g. MHC class I and II family members) were also shown to be differentially expressed using the large-scale array, confirming the technical integrity of both experiments and the biological importance of the gene families.
Upon obtaining the large set of statistically significant gene expression data derived from the analysis of the 44 K microarray, further and extensive data mining and annotation were required to interpret and correlate the results. Numerous reports of expression data annotation have been previously described, in order to assess the large data sets that are considered to be the fruition of high-throughput methodologies. Annotation is necessary because typical methods used to filter the data to a manageable load, such as clustering, can further complicate the dataset, as opposed to converge the output. The Gene Ontology Consortium is a collection of databases for a variety of organisms, which is commonly used to annotate gene expression based on existing knowledge of biological function and a limited functionally-derived vocabulary. This method is organised in a hierarchy of key words including molecular function, biological process, and cellular component . Biological function can then be applied to relevant pathways, through databases such as KEGG (Kyoto Encyclopedia of Genes and Genomes). However, these databases are not nearly as well-developed as the previously mentioned GO databases, suggesting that gaps in the annotation process exist . Furthermore, when studying species that have less known gene functionality (e.g. the chicken) compared with the more widely-studied human and mouse, the annotation process can be more challenging. Nonetheless, as shown by the present study (Figure 3 and Additional file 1), GO annotation did provide a multitude of biological processes potentially involved in the chicken response to Cp infection and NE development, and identified fewer processes that are solely affected by feeding the birds a Medicated diet (bacitracin) compared to the Non-medicated. One might note that this process did not narrow the results to a manageable size. On the contrary, the GO annotation established a data categorization that leads to a new realm of questions regarding the roles of these processes in Cp infection. Moreover, the results of GO annotation required a user-imposed, arbitrary clustering of the processes into more general terms, as demonstrated in Figure 3.
Specific target genes were selected from the list of genes that were differentially expressed to a significant degree for further studies to infer biological function. Selection was based on GO annotation and previous reports of disease pathology associated with NE in chickens or other species, and other diseases that manifest similarly. Specifically, NE lesions have been observed in the small intestine, caeca, liver and kidney of both chickens and turkeys , and the small intestine is the main site and characterized by thin friable walls indicative of exacerbated inflammation [43–45]. Histopathological studies in NE diseased birds have revealed sloughed intestinal epithelium, heterophil infiltration into the lamina propria and enterocyte necrosis as indicated by the presence of matrix metalloproteinase (MMP)-2 enzyme [46, 47]. In humans, perivascular infiltration with polymorphonuclear, mononuclear, and eosinophil cells, in the presence of intestinal necrosis, is observed in cases of necrotizing enteritis . The occurrence of eosinophilia, manifesting as a local type I hypersensitivity response, suggests that this mechanism may also be present in the chicken during the course of NE lesion development. Interestingly, our results from the present study concur with both controlled and excessive inflammatory processes, as well as an antibody receptor gene up-regulation potentially related to an eosinophil response. Regardless of the comparison made between Medicated and Non-medicated chickens, with the aim of deciphering protective and compromised immune responses during the course of NE development, the largest difference in expression profiles were observed before and after clostridial infection irrespective of antibiotic treatments. Thus during the process of correlating biological significance with gene expression data, the focus was shifted primarily to CP-infected birds in comparison to non-infected birds.
Of the genes identified as biologically-relevant from functional annotation, galectin 3, IFNAR1, IgY R and TCR-γ appear to be significantly relevant to the mechanism of development of NE. Expression profiles of IgY R and TCR-γ can be speculated to be correlated with innate immunity responses and eosinophil-regulated inflammation, respectively, whereas galectin 3 and IFNAR1 biological functions throughout CP infection are primarily based on previous characterization of inflammatory disease processes. Type I interferons (IFNs) are typically described as important mediators of the anti-viral cytokine response and more generally speaking, mediators for a pro-inflammatory response . For example, type I IFNs are produced upon bacterial activation of macrophages, which initiates an up-regulatory autocrine-type production of IFNs. Specifically, IFN-α and IFN-β, whose expression patterns are mediated by availability of receptors IFNAR1 and IFNAR2, induce IFN-γ and IL-18 production . In concert, our results indicate up-regulated IFNAR1 expression in Medicated birds on (D1 and D2 PI) compared to D0 PI baseline and the same time point ratios in the Non-medicated groups. At the latest time point, the expression ratio is reversed, in which Non-medicated chickens had the highest level of expression on D4 PI compared to pre-challenged chickens and compared to the same ratio in the Medicated group of chickens. This increase in inflammation at early time points in Medicated birds may suggest a correlation between the protective antibacterial effect of IFNAR1 on D1 PI and D2 PI. However, the rise in IFNAR1 expression on D4 PI in Non-medicated birds may suggest a shift from a protective immune response to an exacerbated, lesion-forming inflammation as observed in Cp-infected chickens developing NE. Importantly, IL-18 expression was also up-regulated in clostridial challenged chickens on D1-D4 PI compared with pre-challenged controls for both the Medicated and Non-medicated groups, with the Medicated group showing the highest level of expression of all.
Galectin 3 is a β-galactoside-specific lectin that is expressed on intestinal epithelial cells (IELs) . Aside from facilitating adhesion, galectin 3 is strongly involved in inflammatory processes and mRNA expression has been observed in IELs from patients with Crohn's disease and bowel information, colon carcinoma and colitis . Suggesting a connection to the antibody-mediated response, serum from Crohn's disease patients was shown to contain anti-galectin-3 IgG antibodies. Moreover, galectin 3 acts as receptor for IgE, facilitating the up-regulation of IgE production in atopic patients . Specifically, B cells expressing surface IgE expressed higher levels of galectin 3 than B cells of other phenotypes . The relationship between IgE, galectin 3 and eosinophil activity in allergy-induced inflammation is such that eosinophils, which are activated upon IgE production, are recruited to sites of inflammation by galectin 3 . Similarly, other members of the galectin family have been shown to stimulate production of pro-inflammatory cytokines under intestinal inflammatory conditions [55, 56] suggesting an integral role of galectins in inflammatory host responses, with emphasis on the intestinal induction site.
In general, the antibody response in chickens is similar to mammals [57, 58]. Avian IgY shares properties with both mammalian IgG and IgE. Although in most cases, IgY molecule is still referred to as chicken IgG, as it appears to be functional equivalent to mammalian IgG. Despite lacking evidence that birds produce the mammalian equivalent of IgE, there is similarity in NE pathogenesis in chickens compared to other IgE-mediated host responses to gut-associated pathogens inducing intestinal inflammation in mammals [59–61], suggesting that our observation on IgY R gene induction and previously reported IgY antibody production following exposure to Cp proteins may be representative of IgE activity. To validate the hypothesis, further functional studies with chicken eosinophils, antibody titres and hypersensitivity are required.
Lastly, TCR-γ expression was shown to be consistently up-regulated in infected chickens, regardless of antibiotic treatment, indicative of the involvement of γδ-T cells in the host response to Cp infection. Specifically, γδ-T cells, which are dominant at mucosal surfaces, have been described to interact with innate immunity cells including mammalian dendritic cells  and chicken NK-like cells  during cellular activation, and presumably at the outset of pathogenic infection. In chickens, NK-like cell activity has been shown in intestinal epithelial lymphocyte populations potentially containing chicken γδ-T cells , and γδ-T cells have been reported within the caecum . Although further studies are required to determine the role of this T cell sub-population in chickens during Cp infection, our results suggest γδ-T cells may represent a previously unreported innate immunity mechanism to combat NE disease.