|
|
||||||||
Call For Papers: 2nd International Symposium on Animal Functional Genomics
Department of Veterinary & Biomedical Sciences, University of Minnesota, St. Paul, Minnesota
ABSTRACT
The enteric immune system of swine protects against infectious and noninfectious environmental insults and discriminates ingested nutrients, food, and commensal microflora from pathogenic agents. The molecular and cellular elements of the immune system have been selected over evolutionary time in response to the specific environment of pigs. Thus, models of immune function based on mouse and human need to be applied cautiously in the pig. To better understand how the mucosal immune system of the small intestine accomplishes the conflicting functions of food tolerance and immunity to enteric infection, we used a genomic approach to profile gene expression in the Peyer's patch. More than 40% of mRNA enriched by differential subtraction for Peyer's patch-specific expressed sequences represented genes of unknown function or had no match in GenBank. Microarray analysis and radiation hybrid mapping validated their porcine origin and provided additional insights into putative functions. The abundance of expressed genes of unknown function indicates that a substantial fraction of the immunological and physiological processes of the Peyer's patch remains to be discovered. It further suggests that swine have evolved specialized biochemical and immunological processes in the small intestine. Further elucidation of these processes are expected to provide novel insights into swine enteric mucosal immune function.
swine; gene expression; gut-associated lymphoid tissue; mucosal immunity; enteric system; Salmonella
INTRODUCTION
The Porcine Enteric System
The intestinal tract encompasses a vast surface area, averaging 300 m2 in humans, and is the largest interface between an animal host and its environment. It is colonized by up to 400 species of commensal bacteria at densities reaching 1011 organisms/ml of luminal fluid (36, 80). The symbiotic microflora provide assistance to the host in nutrient processing and resistance to colonization by exogenous pathogens or opportunistic bacteria that present themselves in the gut lumen (87). The intestine has evolved a diverse array of nonspecific and specific protective mechanisms that allow it to coexist benignly with resident flora yet effectively clear pathogenic microorganisms (33).
The external surface of the intestine is a single layer of epithelial cells that, along with the underlying lamina propria and muscularis mucosae, comprises the mucosa or mucous membrane (59). The nature of the mucosal epithelium varies with location and reflects the functional requirements for absorption of nutrients and microbial defense. These two functions, i.e., as a nutrient absorber and microbial barrier, create a conflict in function that necessitates a complex system of physical, biochemical, and cellular mechanisms for protection of the intestinal mucosa from invading pathogens. The epithelium of the small intestine is specialized along its length for absorptive and secretory functions. It consists of a single layer of tall columnar cells arranged as villi projecting into the lumen and intestinal glands or crypts. The luminal border of the absorptive cells consists of microvilli, which greatly increase the absorptive area of the epithelium. Goblet cells secrete mucus, which protects the epithelium by binding to pathogens and slowing their mobility (9). Lysozyme secreted by Paneth cells in the crypts and proline-rich peptide defensins such as PR-39 are antibacterial factors also present in the mucous layer (9, 45, 59, 68, 84).1
A highly organized immune compartment, the gut-associated lymphoid tissue (GALT), is intimately associated with the gut epithelium and constitutes the largest mass of immune cells in the body (77). The GALT contains cells that function in antigen-specific immune responses. The organized GALT includes the Peyer's patches of the small intestine, which are discrete areas of organized lymphoid tissues in the lamina propria and submucosa with defined B and T lymphocyte areas characteristic of secondary lymphoid organs (59, 64, 77). In pigs, there are two separate categories of Peyer's patch (61). Peyer's patches distributed along the jejunum and proximal ileum (jejunal Peyer's patch) number
2535, are comparatively small, and persist throughout life. The very large single ileocecal patch (ileal Peyer's patch) may extend for up to 2.5 m along the terminal ileum but involutes at
1 yr of age. These two types of patches differ in structure, lymphocyte trafficking characteristics, and lymphocyte production (reviewed in Refs. 40, 74).
The ileal Peyer's patch serves as a primary source of B cells in young pigs (62, 63). It consists of lymphoid follicles embedded in lamina propria, which lie underneath an absorptive columnar epithelium, and so cannot sample antigens directly from the gut lumen (40). By contrast, the jejunal Peyer's patch follicles are found in a region of specialized dome epithelium, the follicle-associated epithelium (FAE). In addition to columnar epithelial cells, the FAE contains cuboidal epithelial cells referred to as microfold or "M" cells that lack an organized brush-border, a thick glycocalyx, and an abundance of hydrolytic enzymes (24). M cells are the only epithelial cells that endocytose and process luminal antigens for subsequent exposure to underlying B and T lymphocytes and macrophages present in the "intracellular pocket" of the M cell. The region underlying the Peyer's patch dome harbors a variety of antigen-processing and -presenting cells and is a principal inductive site for immunity (59). The uptake of antigens and other macromolecules involves the adherence of luminal material to the apical membrane of the M cell and its subsequent capture and transcellular transport by endosomes (73). Because of its specialized structure, its adaptation for antigen sampling, its capacity for induction of immune responses, and its ability to maintain tolerance to food antigens, the jejunal Peyer's patch-containing mucosa is an ideal intestinal tissue in which to examine the full range of GALT immune function.
The Enteric Immune System
A wide assortment of physical, biochemical, and cellular functions in the mucosa participate in the acute host defense to infection. The physical barrier of intestinal epithelium is maintained by a continuous turnover of epithelial cells and rapid repair and reconstitution of injured mucosa. These processes are modulated by a variety of regulatory peptides and growth factors operating at both the apical and basolateral domains of columnar epithelial cells (73). Intestinal epithelial cells adhere to each other through a series of junctional complexes. The apical tight junctions (or zonulae occludens), which form a circumferential belt, prevent paracellular invasion of pathogens and leakage of luminal contents. However, tight junctions are disrupted by stress, cytokines, various neurotransmitters, and the actions of several pathogens (55, 71). Epithelial cells in rodents, human, and swine also have accessory immune functions, including the ability to secrete cytokines, express cytokine receptors, process and present antigens, and transport IgA (35, 46, 67, 91).
Professional phagocytes, macrophages, and neutrophils play a key role in innate immune resistance. However, in contrast to the abundant information concerning phagocyte function during systemic bacterial infections, relatively little is known about their roles in innate mucosal immunity. The antimicrobial actions of macrophages serve a central role in determining the outcome of disease, so understanding the alterations in gene expression associated with infection is essential to elucidating the molecular mechanisms of enteric immunity (42). Peyer's patch dendritic cells are one of the cell types implicated in the immune response to intestinal antigens and pathogens. Mucosal dendritic cells also are key inducers of oral tolerance through the regulation of T cell activity and migration and regulation of IgA production (31, 79). Moreover, a multitude of genes whose protein products are involved in inflammation, immune cell recruitment, and induction of antigen-specific immunity appear to be rapidly induced upon infection in both epithelial cells and macrophages (18, 76), arguing persuasively that phagocyte recruitment and activation are essential components of the enteric host immune response.
Immune nonresponsiveness to food antigens and normal flora, known as oral tolerance, distinguishes the Peyer's patch from all other lymphoid tissues. Oral tolerance is an antigen-specific state of systemic hypo- or nonresponsiveness induced by oral administration of the antigen (89, 90). Oral tolerance prevents unnecessary and detrimental inflammatory responses to resident intestinal flora and to dietary nutrients. Loss of oral tolerance results in inflammatory bowel disease and food allergies (15, 52, 66). A variety of specific mechanisms are involved in oral tolerance (reviewed in Refs. 25, 89). For example, oral tolerance is mediated by the actions of T cells and can be divided into active modulation (suppression) or direct cellular inactivation. Active suppression of immune responses to antigens occurs via the secretion of inhibitory cytokines. IL-10 plays a central role in the maintenance of normal intestinal immune homeostasis, as demonstrated by the intestinal inflammation observed in IL-10 knockout mice (75). Tolerance is also induced under conditions of high-dose oral antigen exposure and involves apoptosis of both Th1 and Th2 cells or the induction of antigen-specific T cell anergy (11, 23, 29). Cytokine expression contributes to direct inactivation of T cell responses since neutralization of IL-12 results in specific clonal T cell deletion mediated by apoptosis (53).
Unique Aspects of the Porcine Immune System
The swine immune system appears to function in a similar fashion to other mammalian species, although few studies of porcine mucosal immunity have been performed at the molecular and cellular levels (10, 70). There is no evidence that the induction of an immune response differs substantially from that observed in the extensively studied murine model. Thus, immune responses are assumed to initiate and proceed similarly with essentially the same result in swine as in other livestock species. That said, there are notable differences between the porcine and murine or human immune systems that indicate a compelling likelihood that the details of immune induction and response in the pig can only be fully discerned by studying the pig itself.
Swine, like other species, have evolved numerous immunological specializations in response to unique environmental and infectious selection pressures. Pigs are omnivores, and their enteric system logically must be adapted to respond to a range of intestinal pathogens. Swine have unusual immune cell populations, such as CD4/CD8 double-positive T cells (78, 100), that play an unknown role in gut mucosal immunity (99). Lymph node function is similar in pigs as in other mammals, but the unusual inverted structure of the lymph nodes relative to mouse and man means that the details of lymphocyte trafficking must be different (reviewed in Ref. 8). Furthermore, the presence of peripheral, circulating 
T cells and CD4/CD8 double-positive
ßT cells in pigs, but not mice or humans, indicates that pigs regulate T cell differentiation differently and have cells whose function cannot be predicted from mouse models (7, 72).
Porcine IgG subclasses, unlike those of mouse and human, are diverse, with up to eight forms (10, 41). Since subclass diversification occurred after mammalian radiation and speciation (43, 65), the functional properties of porcine IgG subclasses cannot be inferred from biochemical similarities to IgG subclasses of other species. Thus, the functional roles and properties of IgG1 and IgG2 in mice or humans are unreliable guides to the functions of IgG subclasses in the pig.
Notable differences between porcine and human or murine innate immunity also exist. IL-8, a key neutrophil chemoattractant, is highly expressed in porcine macrophages (47) but is absent in mice. Inducible nitric oxide synthase is an important component of innate antibacterial defense in murine leukocytes. It is not induced in pig immune cells (69) but is readily expressed in porcine small intestine villous epithelium (28). IL-4, a key differentiation and immunoregulatory cytokine, has a 17-amino acid deletion in artiodactyls (swine and cattle) compared with humans. The structure and diversity of IgA alleles differ from that of human and mouse and consist of dimers and larger polymers in the serum of pigs, whereas humans have mostly monomeric IgA in the serum (82). Also, porcine antibodies may be diversified in the lower gut (Peyer's patches) early in life rather than in bone marrow throughout life, as in humans and rodents (82). The response to CpG differs between human and mouse, and the porcine response is similar to that of human (30).
Functional Characterization of Porcine Enteric Immunity
Immunological function in the gut requires a balance between oral tolerance of food antigens and immunological responsiveness to microbial pathogens. The current understanding of oral tolerance in pigs is based primarily on descriptive studies of hypersensitivity to food antigens at the time of weaning (25, 57, 85, 86). These studies described the T cell populations and the cytokine microenvironment of intestinal lamina propria and epithelium (6). The porcine gastrointestinal tract is heavily biased toward immunological nonresponsiveness and oral tolerance (83), which further emphasizes the challenges inherent in eliciting effective immunity against enteric pathogens.
Saif and colleagues (88, 9296) conducted a series of studies using a neonatal gnotobiotic pig model of rotavirus infection and diarrhea to identify correlates of protective immunity and to evaluate vaccine approaches for the induction of protective mucosal immune responses. They determined that production of intestinal IgA antibodies, induction of neutralizing antibody responses, and use of effective mucosal adjuvants or antigen delivery systems were necessary for effective enteric protection. These elegant studies examined effector responses after immunity had been acquired to live virus or virus-like particle immunization but did not elucidate the initiating events necessary for induction of a protective immune response.
An oral immunization model was used to characterize the adjuvanticity and immunogenicity of cholera toxin (CT) in the pig and showed that CT was a potent mucosal adjuvant, whereas its CT-B subunit was a powerful immunogen (22). CT adjuvanticity in the porcine gut is dependent on physical association with an antigen (22), but in mice CT need only be mixed together with antigen before oral administration (20, 50). The adjuvant activity of CT in the pig was associated with induction of inflammatory cytokines and the expression of costimulatory molecules on antigen-presenting cells (21, 22, 60). In contrast to mice, in which interferon-
(IFN-
) does not appear to be required for responses to CT, oral immunization with CT-B subunit plus CT significantly increases IFN-
-secreting cells in the mesenteric lymph node (MLN) and increases the amount of IFN-
secreted by lamina propria cells after antigen stimulation (Ref. 21 and unpublished data). The information obtained from these studies established that CT can break oral tolerance and elicit potent enteric immunity in the pig. Moreover, these data indicate that an early alteration of the inflammatory cytokine environment is important to CT adjuvant action in porcine intestinal mucosal immune tissues.
GENOMIC DISSECTION OF PORCINE INTESTINAL IMMUNITY
Identifying Genes of the Porcine Enteric Immune System
Oral tolerance is a unique and vital feature of the GALT that affects individual immunity and potential applications of oral vaccines. Thus, a catalog of sequences expressed in Peyer's patches is expected to reveal novel insights into the biochemical and molecular regulation of oral tolerance and both innate immunity and pathogen-specific immune responses. In the interaction with intestinal mucosa, invasive bacteria such as Salmonella unleash a broad range of host cell-modifying activities that facilitate bacterial entry and survival, ranging from enhancing the development of M cells in Peyer's patches for additional gateways of infection (44) to apoptosis of phagocytic macrophages (34, 49, 58). Whereas identification of bacterial genes that contribute to invasion and survival within the intestinal environment has rapidly accumulated (1, 13), analyses of concomitant alterations in host cell gene expression are just beginning. Elucidation of the genes expressed in intestinal tissues would provide a useful framework for more rapid progress.
Toward this end, we have identified a set of porcine Peyer's patch-expressed genes (17). Peyer's patch cDNA libraries were created from freshly isolated and in vitro-stimulated tissues, combined, and subtracted with fibroblast mRNA. The resulting subtracted library of
6,000 clones was presumably enriched in sequences selectively expressed in the porcine Peyer's patch. Single-pass 5'-end sequencing and bioinformatics analysis indicated the presence of 3,687 expressed sequence tags (ESTs), which further reduced to 2,414 unique sequences with a Poisson distribution of abundance (Fig. 1) (17). Some 1,638 sequences were represented by single clones, and the most abundant cDNA, the Na+/K+ ATPase inhibitor SPAI-2, was present in 47 clones (Fig. 1). Sequences from the normal, stimulated, and subtracted libraries were deposited in the GenBank EST database under the library names UMNMPM1, UMNMPM2, and UMNMPM3, respectively.
|
Of the identified porcine Peyer's patch sequences, 50% corresponded to genes with known functions in the public databases. Only 65% of these genes contained gene ontology (GO) information from the ENSEMBL database (May 2006), and were categorized according to biological process as shown in Fig. 2. Nine percent of genes with GO information were in the category of "response to stimulus," which includes "stress response" and "immune-related processes." In the initial analysis, 30% of the sequences hit to entries in the public databases with no known functions, and 20% had no significant hits in the public databases, perhaps representing novel porcine genes (17). As of May 2006, 31% of the unigene set corresponded to sequences of unknown function, and 4% had no matches in the public databases (novel ESTs).
|
|
|
|
|
A primary goal of our research program is to understand the molecular processes leading to immune protection or tolerance in the porcine gut. Therefore, determining the function of novel porcine genes expressed in mucosal immune tissues is required. Open reading frames (ORF) for all 10 sequences were examined using ORF finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html) to identify conserved protein motifs or identify putative functions for the novel porcine ESTs shown in Table 2. No large ORFs were found for any of the ESTs, suggesting that the mRNA sequence still may be incomplete or it may be noncoding RNA. Current studies are focused on characterizing spatial distribution of these transcripts within porcine Peyer's patches, determining whether they encode proteins, and bioinformatics analysis of putative functional domains.
Differential Gene Expression in the Porcine GALT
To gain insight into the temporal regulation of identified porcine Peyer's patch ESTs, gene expression changes were examined in the GALT in response to Salmonella choleraesuis infection. S. choleraesuis is an enteric pathogen of swine whose pathogenesis and food safety risks have been extensively studied. Initially, ileal Peyer's patch tissue harvested 6 h after Salmonella infection was competitively hybridized on the cDNA microarray with mRNA from control tissue. Seven genes showed expression ratios >1.5-fold in response to infection. They were matrix metalloproteinase-1, cell death activator (CIDE-B), X-box binding protein 1, protein differentially expressed in hematopoietic lineages (GW112), 2',5' oligoadenylate synthetase, lithostathine precursor [pancreatic stone protein (PSP)], and inhibitor of apoptosis protein-1 (unpublished data). To further evaluate these preliminary findings, quantitative real-time PCR was performed on these mRNA and on various inflammatory mediators and signaling molecules, including IL-1ß, IL-6, IL-8, TNF-
, CCL19, CXCL2, lysozyme, and calprotectin (S100A9) from Peyer's patch tissues obtained from six Salmonella-infected and five control animals. After Salmonella infection, levels of S100A9, IL-1ß, and IL-8 mRNAs were consistently increased and correlated in both proximal and distal ileal Peyer's patch, whereas other mRNAs showed no apparent pattern by cluster analysis. IL-1ß and IL-8 mRNA levels were positively correlated with numbers of mucosal neutrophils in distal ileal Peyer's patch, and the presence of Salmonella DNA (39).
Expression of IL-1ß and IL-8 mRNA in Peyer's patch in response to Salmonella challenge was further examined in an ex vivo Ussing chamber model. An increase of approximately twofold in transcripts for IL-8 and IL-1ß were observed after a 2-h infection compared with mock-infected tissues, further indicating that these cytokines are involved in the early response to S. choleraesuis (38).
The findings from these studies and related published literature in other species suggests a model in which flagellin of invading Salmonella is translocated to the epithelial basolateral membrane and binds Toll-like receptor (TLR)5 (26, 27). Ligation of TLR5 activates NF-
B and MAPK signaling pathways and induces an epithelial inflammatory gene expression program that includes IL-1ß and IL-8 (19, 26, 56). In combination with SopE2 injection via a type III secretion system, flagellin binding to TLR5 induces maximal IL-8 secretion by epithelial cells (37). Secretion of IL-8, along with other signals, such as complement 5, macrophage inflammatory protein (MIP)-2, and MIP-3
, induce neutrophil and dendritic cell chemotaxis to the site of invasion (12, 54, 81, 97). Neutrophil-secreted proteases then degrade flagellin, halting expression of epithelial inflammatory gene expression and preventing further neutrophil-induced tissue damage (48). The details of the molecular interactions of Salmonella and the porcine GALT are likely to vary compared with other species, depending in part on differences in their expressed sequences. Mice, for example, lack homologs of six human chemokines, including IL-8 (CXCL8), which is a key neutrophil chemoattractant in swine and humans (98).
The variation observed in calprotectin subunit S100A9 expression in Salmonella-infected and uninfected pigs demonstrates the extent of natural variation that may occur among outbred animals. Age is one of the variables that contributes to variation. For example, adult and juvenile pigs are different physiologically and developmentally, suggesting that patterns of Peyer's patch gene expression that affect immune function may vary. Thus we examined the expression profile in juvenile and adult pigs of 19 sequences that were more highly expressed in jejunal Peyer's patch than in MLN (51). Hierarchical clustering of the 19 ESTs by the distance between the expression profiles demonstrated that an adult jejunal Peyer's patch has a distinct pattern of expression separate from juvenile jejunal Peyer's patch and that there was significant animal-to-animal variation. ESTs with a high degree of variation [(coefficient of variation CV ) > 80%] in both juvenile and adult jejunal Peyer's patch were MHCII (major histocompatibility complex class II)-DQ, epithelial chloride channel protein, and PSP/Reg I, whereas CCL19, myristoylated alanine-rich C kinase substrate (MARCKS), basic proline-rich protein (PROL)1, CIDE-B, and GW112 had CVs >80% only in juvenile jejunal Peyer's patch (51). Eight of these genes (CB286471, CB287260, CIDE-B, GW112, MHCII-DR, PSP, MARCKS, and PROL1) showed differential expression between the two ages of pigs (51). Genes expressed in higher levels in the juvenile Peyer's patches were involved in growth and apoptosis-related processes consistent with rapid physiological growth and maturation of young animals (51). Given the dynamic local environment of Peyer's patches in the small intestine due to episodic feeding, peristalsis, and changes in the microbial community, it is not surprising and presents a challenge to the design of well-controlled animal experiments.
CONCLUSIONS
The initial unbiased genomic exploration of porcine Peyer's patch-expressed sequences reviewed here supports the notion that specialized biochemical and molecular processes unique to the small intestine are required to carry out its physiological activities and that these processes are highly adapted to individual species. The pig, for example, must respond immunologically to commensal and pathogenic microflora that have coevolved over evolutionary time and again were selected through the pressures imposed by domestication giving rise to present day commercial breeds and husbandry practices. It is reasonable to assume that the details of these interactions are determined at a fundamental level by protein-protein interactions, which will be different, especially with respect to immunological interactions, to the degree that the host and microbial protein sequences are specific to themselves. Indeed, antigen-specific immunity relies on this assumption and predicts that the induction of tolerance or immune responsiveness in the small intestine will be highly specific at the molecular level.
Genomic approaches provide an unparalleled opportunity to directly explore and dissect food animal mucosal immunology. Although our early genomic approaches used homemade cDNA microarrays, high-quality, commercial porcine long oligonucleotide array sets (Qiagen, Valencia, CA; and Illumina, San Diego, CA) and printed Affymetrix arrays (Affymetrix, Santa Clara, CA) that are now available allow the examination of large numbers of genes with increased reproducibility and an expanded range of analytical and visualization software. Here we demonstrated the feasibility of identifying differentially expressed genes of the Peyer's patch, which can be further studied experimentally to reveal their roles in intestinal biology. Peyer's patches play a central role in balancing absorption of nutrients, tolerance of commensals, and resistance to pathogens that is essential to health. Elucidation of the molecular and cellular processes that maintain this balance is expected to facilitate the development of better tools for disease prevention and improved nutrition.
ACKNOWLEDGMENTS
The authors thank Dr. Zheng Jin Tu at the University of Minnesota Supercomputing Institute for technical support.
FOOTNOTES
Address for reprint requests and other correspondence: M. P. Murtaugh, Dept. of Veterinary & Biomedical Sciences, Univ. of Minnesota, 1971 Commonwealth Ave., St. Paul, MN 55108 (e-mail: murta001{at}umn.edu).
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
1 The 2nd International Symposium on Animal Functional Genomics was held May 1619, 2006, at Michigan State University in East Lansing, MI, and was organized by Jeanne Burton of Michigan State University and Guilherme J. M. Rosa of University of Wisconsin-Madison (see meeting report by Drs. Burton and Rosa, Physiol Genomics 28: 1-4, 2006). ![]()
REFERENCES
This article has been cited by other articles:
![]() |
T. E. Burkey, K. A. Skjolaas, and J. E. Minton BOARD-INVITED REVIEW: Porcine mucosal immunity of the gastrointestinal tract J Anim Sci, April 1, 2009; 87(4): 1493 - 1501. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. S. Rogers, W. M. Abraham, K. A. Brogden, J. F. Engelhardt, J. T. Fisher, P. B. McCray Jr., G. McLennan, D. K. Meyerholz, E. Namati, L. S. Ostedgaard, et al. The porcine lung as a potential model for cystic fibrosis Am J Physiol Lung Cell Mol Physiol, August 1, 2008; 295(2): L240 - L263. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. L. Burton and G. J. M. Rosa Physiological genomics special issue on animal functional genomics Physiol Genomics, December 13, 2006; 28(1): 1 - 4. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |