Immunohistochemical study on the delayed progression of epithelial apoptosis in follicle-associated epithelium of rat Peyer's patch

It is well known that some caspases in apoptosis is involved in determinant of terminal differentiation and maturation of various cells. Our previous study ultrastructurally clarified the differentiation into M cells from immature microvillous epithelial cells and the redifferentiation from M cells to microvillous epithelial cells in the follicle-associated epithelium (FAE) of rat Peyer's patch. In this study, the difference of epithelial apoptosis between the FAE of Peyer's patch and intestinal villi was immunohistochemically investigated in rat jejunoileum. As a result, cleaved caspase-3 was limited to several epithelial cells at the tip of FAE, whereas almost all of the epithelial cells were cleaved caspase-3 positive in intestinal villi. Cleaved caspase-9 was detected only in a few exfoliating or exfoliated epithelial cells of both FAE and intestinal villi. Nuclear DNA-fragmentation was detected only in several epithelial cells of the tip of FAE, while it was expressed from the middle regions in the intestinal villi. The DNase I expression of the epithelial cytoplasm was much weaker in FAE than in intestinal villi. Bcl-x expression was restricted in the apical cytoplasms of epithelial cells in the FAE, whereas it was restricted in whole cytoplasms in villous epithelial cells. These findings suggest that the progression of the apoptotic process in the epithelial cells of FAE is later than in the intestinal villi, so that the possibility of epithelial differentiation might be remained in the FAE, unlike in the intestinal villi.     

Differentiation of epithelial cells to M cells in response to bacterial colonization on the follicle-associated epithelium of Peyer's patch in rat small intestine

To clarify the relationship between M cells and intestinal microflora, histoplanimetrical investigation into the bacterial colonization and the differentiation to M cells was carried out in rat Peyer's patch under physiological conditions. The follicle-associated epithelium (FAE), except for the narrow area of apical region, was closely covered with both neighboring intestinal villi and a thick mucous layer, the latter of which also filled the intervillous spaces as well as the space between the FAE and the neighboring intestinal villi. Indigenous bacteria adhered almost constantly to the narrow areas of apical regions of both intestinal villi and the FAE. Bacterial colonies were occasionally located on the basal to middle region of FAE, where M cells also appeared, forming large pockets. When bacterial colonies were located on the basal to middle region of FAE, bacteria with the same morphological characteristics also proliferated in the intervillous spaces neighboring the Peyer's patch. In cases with no bacterial colonies on the basal to middle region of FAE, however, M cells were rare in the FAE. Histoplanimetrical analysis showed the similar distribution pattern of bacterial colonies on the FAE and M cells in the FAE. M cells ultrastructurally engulfed indigenous bacteria, which were then transported to the pockets. These results suggest that indigenous bacterial colonization on the FAE stimulates the differentiation of M cells in the FAE under physiological conditions. The uptake of bacteria by M cells might contribute the regulation of the development of indigenous bacterial colonies in the small intestine.     

Distribution of the pores of epithelial basement membrane in the rat small intestine

The distribution and diameter of the pores of epithelial basement membrane in the intestinal villi and the lymph nodules of ileal Peyer's patches were investigated in the rat small intestine by scanning electron microscopy after the removal of the overlying epithelial cells with OsO(4) maceration. In the duodenum, jejunum and ileum, the pores were mainly distributed at the upper three fourths of the villi, but were scarce around the top of the villi. The diameter of some of the pores in the upper three fourths of the villi was larger than that of those in the lower portion. The protrusion of lymphocytes and the cytoplasmic processes of macrophages were also seen at the orifices of the pores. In ileal Peyer's patches, in contrast, pores were densely distributed in the lower one third of the follicle-associated epithelium (FAE) where M cells were mainly seen. Furthermore, these pores were larger than those found in the upper two thirds. Lymphocytes or cytoplasmic processes of macrophages were frequently seen in the lower one third of FAE. These results suggest that the pores at the basement membrane correspond to the passage of the immunocompetent cells which are in contact with M cells or villous columnar epithelial cells and that the abundance of pores is a sign of aggressive interaction between the particular epithelial cells and the immunocompetent cells at the upper three fourths of intestinal villi and the lower one third of FAE in the rat small intestine.     

Apoptosis of villous epithelial cells and follicle-associated epithelial cells in chicken cecum

The process of the disappearance of epithelial cells was examined in chicken cecal villi and follicle-associated epithelium (FAE). The apoptotic epithelial cells with intense DNA-fragmentation and their exfoliation were found in the villous tips. The epithelial cells with weak DNA-fragmentation were seen in the upper portion of the villi and their sparse exfoliations were also found there. Numerous epithelial cells in the intestinal lumen expressed the apoptotic features. A row of apoptotic epithelial cells with DNA-fragmentation was also found in the apical FAE, whereas no M cells exhibited any apoptotic signs. In all cecal regions, CD3+, CD8+, and TCR2+ lymphocytes were predominant in the epithelium at the upper portion of the villi and the FAE. CD4+ lymphocytes were mainly seen in the lamina propria. TCR1+ lymphocytes were not abundant in comparison with TCR2+ lymphocytes in the epithelium. TCR3+ T lymphocytes were rarely detected. These results suggest that the chicken cecal epithelial cells exfoliated into the lumen after the induction of the apoptosis, and that the induction may be involved with CD3+, CD8+, and TCR2+ lymphocytes. No death in M cells suggests that M cells may transform into microvillous epithelial cells.     

In vitro adhesion of K88ac+ Escherichia coli to Peyer''s patch and peripheral blood lymphocytes, buccal and rectal epithelial cells or intestinal epithelial brush borders of weaned pigs

Escherichia coli adhesion assays were conducted using isolated porcine peripheral blood lymphocytes, Peyer's patch lymphocytes, rectal epithelial cells or brush borders, buccal epithelial cells and brush borders from small intestinal epithelial cells. The cells and brush borders were tested for their ability to bind K88-piliated exterotoxigenic E. coliStrain M1823B (K88ac) and E. coli Strain 1476 (K-12, K88ac). Comparison of adhesive phenotypes of 37 weaned pigs as determined by the adhesion assay with small intestinal brush borders and the adherence of K88ac⁺ enterotoxigenic E. coli to peripheral blood lymphocytes, Peyer's patch lymphocytes and rectal epithelial cells or brush borders, revealed no correlation. In vitro adhesion of K88ac-bearing E. coli was always negative with buccal epithelial cells. K88ac strains varied in their ability to adhere to lymphocytes and rectale pithelial cells or brush borders, indicating that the mechanism of adherence is unrelated to K88-mediated adhesion observed in animals that had the receptors on small-intestinal epithelial-cell brush borders. The non-piliated control E. coli Strain 123 adhered to fresh peripheral blood lymphocytes, and less intensively to frozen-thawed peripheral blood lymphocytes or Peyer's patch lymphocytes. It was concluded that none of the cell types or brush borders, except small-intestinal epithelial-cell brush borders, could be used as targets for phenotyping pigs for the presence of the K88 receptors that have been associated with adhesion and colonization of K88⁺ enterotoxigenic E. coli in the porcine small intestine.     

In vitro adhesion of K88acEscherichia coli to Peyer's patch and peripheral blood lymphocytes, buccal and rectal epithelial cells or intestinal epithelial brush borders of weaned pigs

Escherichia coli adhesion assays were conducted using isolated porcine peripheral blood lymphocytes, Peyer's patch lymphocytes, rectal epithelial cells or brush borders, buccal epithelial cells and brush borders from small intestinal epithelial cells. The cells and brush borders were tested for their ability to bind K88-piliated exterotoxigenic E. coliStrain M1823B (K88ac) and E. coli Strain 1476 (K-12, K88ac). Comparison of adhesive phenotypes of 37 weaned pigs as determined by the adhesion assay with small intestinal brush borders and the adherence of K88ac⁺ enterotoxigenic E. coli to peripheral blood lymphocytes, Peyer's patch lymphocytes and rectal epithelial cells or brush borders, revealed no correlation. In vitro adhesion of K88ac-bearing E. coli was always negative with buccal epithelial cells. K88ac strains varied in their ability to adhere to lymphocytes and rectale pithelial cells or brush borders, indicating that the mechanism of adherence is unrelated to K88-mediated adhesion observed in animals that had the receptors on small-intestinal epithelial-cell brush borders. The non-piliated control E. coli Strain 123 adhered to fresh peripheral blood lymphocytes, and less intensively to frozen-thawed peripheral blood lymphocytes or Peyer's patch lymphocytes. It was concluded that none of the cell types or brush borders, except small-intestinal epithelial-cell brush borders, could be used as targets for phenotyping pigs for the presence of the K88 receptors that have been associated with adhesion and colonization of K88⁺ enterotoxigenic E. coli in the porcine small intestine.     

Ultrastructural study on the epithelial responses against attachment of indigenous bacteria to epithelial membranes in peyer's patches of rat small intestine

The ultrastructure of epithelial responses against the membrane adhesion of indigenous bacteria was investigated in the follicle-associated epithelium (FAE) of rat small intestine. The most frequent adherence of the various morphological types of bacteria to the epithelial membranes was found at the apex of the FAE. The attachment sites were deeply invaginated, and their bottoms were deformed into a sharp cone shape. Four layers with different electron densities were formed just beneath the apical membranes by microfilaments which surrounded the invaginations. The electron density of each layer was gradually decreased as being apart from the invaginations. The extremities of some bacteria in the invaginations were deformed into sharpened shapes. The cell walls of the extremities of the bacteria were occasionally dissolved in the invaginations, and their cytoplasms were slightly swollen with low electron densities. In some invaginations, the attached bacteria were eliminated to leave their fragments such as filamentous debris and a part of cell walls. Finally these remnants disappeared completely. When the bacterial colonies existed in the middle region of the FAE, the attachment of bacteria resulted in the engulfment of bacteria by M cells. The degenerated bacteria whose cytoplasmic matrices were separated into high electron dense materials and cleared materials were occasionally engulfed by ordinary microvillous columnar epithelial cells or goblet cells throughout the FAE. These findings suggest that the epithelial cells reject the attachment of live indigenous bacteria and that the M cells absorb indigenous bacteria in rat Peyer's patches.     

Cellular kinetics of villous epithelial cells and m cells in rabbit small intestine

The cellular kinetics of villous columnar epithelial cells and M cells in the rabbit small intestine were determined by the use of 5-bromo-2'-deoxyuridine (BrdU) as a tracer. To identify M cells, vimentin antibody was used. The BrdU-labeled nuclei of columnar epithelial cells reached the base of intestinal villi in all portions at 1 day after BrdU administration. Thereafter, BrdU-labeled cells migrated toward the villous tip, but they did not move at a uniform speed. The epithelial cells which existed in intestinal villi on circular folds moved faster than those on mucosa other than circular folds. At 7 days after BrdU administration, the leading edge of BrdU-labeled epithelial cells already disappeared from the villous tip in all portions of the small intestine. In the ileal Peyer's patch, the BrdU-labeled nuclei of microvillous epithelial cells and vimentin-positive M cells appeared near the intestinal crypt orifice at 1 day after BrdU administration, and then migrated toward the luminal surface of the follicle-associated epithelium (FAE). As they moved toward the upper portion of FAE, the number of BrdU-labeled M cells on the side of the dome decreased simultaneously. The leading edge of BrdU-labeled epithelial cells disappeared from the top of the FAE within 7 days. These results suggest that M cells may differentiate from the undifferentiated cells in intestinal crypts within 1 day and disappear from the top of the FAE after the change of their form from M cells into microvillous epithelial cells.     

The cellular differentiation of M cells from crypt undifferentiated epithelial cells into microvillous epithelial cells in follicle-associated epithelia of chicken cecal tonsils

To clarify the cellular origin and the fate of M cells, detailed distributions of the epithelial cells were investigated scanning electron microscopically on the follicle-associated epithelia (FAE) of chicken cecal tonsils. The distribution of M cells was closely related with the situation of the crypt orifices in chicken cecal tonsils. In undeveloped cecal tonsils, the intestinal crypts were localized at the periphery of the FAE. In these tonsils, M cells without microvilli (M(0)) were predominantly populated in the basal region of the FAE, whereas goblet cells and microvillous epithelial cells (MV) were more distributed in the middle to the apical region of the FAE. A few M cells with short microvilli were dispersed throughout the FAE. Significantly shrunk MV (MVs) clustered together in transitional portions from the lateral face to the roof of the FAE. In well-developed cecal tonsils, the crypts also opened at the lateral surface in addition to the periphery of the FAE. In these tonsils, the M(0) accumulated densely in the small areas around the crypt orifices exclusively. No sign of exfoliation of apoptotic epithelial cells was found in the M(0)-accumulated areas and at their peripheral boundaries. The MVs were often clustered in the central regions among the crypt orifices in addition to the roof of the FAE. These findings suggest that M cells are directly derived from the undifferentiated crypt epithelial cells, not fall into apoptotic cell death and further differentiate into MV in the FAE of chicken cecal tonsils.     

Cryptosporidiosis and the follicle-associated epithelium over the ileal Peyer's patch in calves

Three calves were studied in stages of spontaneous cryptosporidial infection with particular reference to the relation of the cryptosporidia to the follicle-associated epithelium (FAE) over the ileal Peyer's patch (IPP). In early infection scanning electron microscopy and streptavidin immunoperoxidase staining showed marked predilection of cryptosporidia for the FAE. Cryptosporidial antigen was also found in subepithelial tissue, both in the domes over the IPP and in villi, apparently in macrophages, where the parasites seemed to be progressively degraded. The FAE showed long tightly spaced microvilli, replacing normal low folds and protrusions, particularly in late infection. Endocytosis of indian ink was restricted to the cell periphery in late infection, contrasting the normal, more even distribution of endocytosis in the FAE apical cytoplasm. Few parasites were seen in the intestinal mucosa at this stage. At convalescence the FAE was normal, but all stages of infection were characterised by elongation of microvilli in absorptive cells.     

Differences in lectin-binding properties between the common mucosal epithelium and follicle-associated epithelium in the rabbit small intestine

Differences in sugar distribution between the villous epithelium and follicle-associated epithelium (FAE) were compared using lectins in the rabbit small intestine. In every portion, villous columnar epithelial cells primarily exhibited a positive reaction to the GalNAc, GlcNAc, galactose, and oligosaccharide. In the ileal Peyer's patch (PP), whereas microvillous epithelial cells exhibited positive reactions, M cells tended to be negative. The villous epithelial reaction to the fucose group was negative, but M cells and microvillous epithelial cells showed a positive to the fucose. No epithelium had a positive reaction to the mannose and glucose. The variety of lectin-binding properties of villous epithelial cells and M cells may reflect specificity for the recognizing luminal substances such as antigenic molecules and bacterial elements.     

Ultrastructural study on the differentiation and the fate of M cells in follicle-associated epithelium of rat Peyer's patch

The differentiation process of immature microvillous epithelial cells to M cells and the fate of M cells in the follicle-associated epithelium (FAE) of the mucosa-associated lymphoid tissues are still unclear. In this study, the differentiation process and the fate of M cells were clarified in rat Peyer's patches under a transmission electron microscope. Almost all immature epithelial cells were found to possess long, slender microvilli, which gradually shortened, thickened and dispersed as the immature epithelial cells migrated away from the crypt orifices. These morphological changes started in the centers and moved to the peripheries of the apical surfaces of epithelial cells, accompanied by the protrusion of apical cytoplasm out of the terminal web. During these changes, the bundles of microfilaments of microvilli never shortened, and both small vesicles in the apical cytoplasm and tiny invaginations of the apical membranes were found. The intraepithelial migrating cells gradually accumulated to form typical intraepithelial pockets. In all FAE, there was no morphological sign of cell death in M cells. The rearrangement of microfilament bundles, the reconstruction of microvilli and the disappearance of pockets resulted in the transformation of M cells into microvillous epithelial cells. These serial ultrastructural changes suggest that M cells are a temporal and transitional cell type caused by the active engulfment of luminal substances and that when the engulfment ceases, the M cells transform into mature microvillous epithelial cells.     

Lectin histochemical investigations of the distal gut of chicks with special emphasis on the follicle-associated epithelium

Carbohydrates on epithelial cell surfaces play an important role as attachment sites for different microorganisms like bacteria, viruses and protozoa. To obtain more information about the distribution of carbohydrates on the luminal surface along the intestine, lectin histochemical studies on different gut segments of chicks of different age groups were carried out using a panel of 13 lectins with specificities for Man, Glc, Gal, GalNAc, GlcNAc or GlcNAc oligosaccharides and Sia. Furthermore, we tried to find out whether previously reported specificities of certain lectins for M cells (membranous or multifold cells) in the bursa of Fabricius (BF) can be observed also on M cells of the intestine. As a result we were able to demonstrate binding of all lectins employed in these studies in all investigated gut segments. In some cases, the application of the same lectin led to varying staining intensities of the same histological structures in different age-groups (e.g. staining of the brush border with WGA, LEA, MAA or Conarva) or different gut segments (e.g. staining of goblet cells with CMA II, LEA and MPA). Hence, terminal carbohydrate residues of glycoconjugates on the intestinal epithelium vary depending on age and organ site. As glycoconjugates can act as attachment sites for microorganisms, these differences in the distribution of sugar residues may be one explanation for the site-specificity of certain pathogens. Furthermore, the binding of lectins to the follicle-associated epithelium (FAE) of the BF differs from that to the FAE of the intestine again stressing the site specificity of lectin binding. Thus, up to now no universal M-cell marker along the chicken intestine exists.     

Development of ileal Peyer's patches and follicle associated epithelium in bovine foetuses

The development of the ileal Peyer's patches (ilPP) and follicle associated epithelium (FAE) was examined in 30 bovine foetuses ranging from 73 to 271 days of gestation by light and transmission electron microscopic methods. The first primordial ilPP was encountered in the foetus at 164 days of gestation The ilPP were found to have been formed from the aggregation of lymph follicles in the foetus at 227 days of gestation whereas in the foetus at 271 days of gestation the follicular development was observed to have been completed. While the cells in the FAE in the foetus at 164 days of gestation and those older were cuboidal, those of the foetus at 271 days of gestation were columnar. As from the foetus at 227 days of gestation, however, the FAE was found to be composed of uniform lymphoepithelial cells with an increase in the number of intraepithelial leukocytes. In the early stages, whereas the apical surfaces of the FAE cells appeared shorter with microfolds, with advancing age the apical surfaces of the FAE cells were observed to be heterogeneous. Our results suggest that bovine ilPP and FAE cells are histologically and functionally mature before birth.     

Effect of rotavirus and/or Escherichia coli infection on the aggregated lymphoid follicles in the small intestine of neonatal gnotobiotic calves

The effect of rotavirus and/or Escherichia coli infections on the follicle-associated epithelium (FAE or M cells) of the domes of the aggregated lymphoid follicles (ALF, or Peyer's patches) of gnotobiotic calves was evaluated by light, scanning electron, transmission electron, and immunofluorescence microscopies. Calf rotavirus (CRV) infection produced loss of FAE cell microvilli, and virions were observed in cytoplasmic vacuoles of FAE cells, as well as in intercellular spaces between FAE cells and lymphoid cells migrating through the dome epithelium. The CRV particles appeared to have entered the FAE cells by phagocytosis, with no subsequent cytoplasmic replication. Enterotoxigenic E coli (ETEC) induced more severe alterations including marked microvilli loss and ballooning in the FAE cells. There was no adhesion to, or colonization of FAE cells by ETEC, but bacteria were observed free or phagocytized within the dome and the germinal centers of the ALF. There were no ETEC observed in the cytoplasm of FAE cells. The presence of nonenterotoxigenic E coli (NETEC) in the intestine of calves had no effect on the intestinal FAE cells. The addition of NETEC to CRV infections did not enhance or modify in any way the response of FAE cells to the viral infection; however, the combination of CRV + ETEC produced severe necrosis of the FAE cells, and loss of dome epithelium of ALF.     

The Epithelium Covering Peyer’s Patches in Young Milk-Fed Calves: An Ultrastructural and Enzyme Histochemical Investigation

Between the ordinary villi over Peyer’s patches there are small domes or “pseudovilli” caused by bulges in the lymphoid tissue. These “pseudovilli” were studied in 5 healthy milk-fed, about 3-week-old, pre-ruminant calves. Scanning electron microscopy revealed that the “pseudovilli” were covered by a specialized follicle associated epithelium (FAE). The FAE had poorly developed microvilli and often extensive folding of the cell surface close to the cell borders. By transmission electron microscopy the tips of the marginal folds of the FAE seemed to fuse, probably in the process of enfolding bulk material from the lumen. The FAE apical cytoplasm contained numerous thick-walled and bristle-coated invaginations, tubules and vesicles indicative of micropinocytosis. Multivesicular bodies and large vacuoles were frequent. Indications of extracellular unloading of residual bodies were found. Intraepithelial lymphocytes tended to group together, and some were rich in rough endoplasmic reticulum. Enzyme histochemistry showed weak reactions of adenosine triphosphate splitting enzyme and aminopeptidase in the FAE luminal cell border. Cytoplasmic acid phosrphatase showed a marked basal-apical decrease along the “pseudovillus” probably caused by the onset of endocytosis. The results of this study appear compatible with the concept that the FAE takes up macromolecules from the lumen by pinocytosis and sensitizes lymphocytes.     

Temporal lectin histochemical characterization of porcine small intestine.

A variety of biotinylated lectins was applied to formalin-fixed intestinal sections from isolator-reared pigs ranging in age from newborn through 12 weeks. Lectin binding to brush borders of villus enterocytes, crypt enterocytes, and dome epithelium, and lectin reactivity within goblet cells and Brunner's glands was semiquantified by microscopy and was used to estimate temporal changes in complex carbohydrates of enteric epithelium. Although variability in binding scores often was observed among pigs of the same age, several general patterns of lectin binding were detected. Dolichos biflorus and Ulex europaeus lectins had increasing binding to brush border membranes as pigs aged. The Dolichos biflorus, however, had decreased binding at the 12-week time point. Neuraminidase-treated Arachis hypogaea and Triticum vulgaris were associated with high mean binding scores at all time points. Canavalia ensiformis bound, with high mean score at all time points, to villus but not to crypt enterocytes. Arachis hypogaea was associated with variable but often high binding scores, regardless of pig age. Succinylated wheat germ agglutinin bound more to crypt than to villus enterocytes. Goblet cells were generally less reactive than were corresponding villi and crypts. Dome epithelium reactivity varied with the lectin used, whereas Brunner's glands reacted with all lectins tested. We conclude that age and regional variations in lectin binding may reflect differences in intestinal function and differentiation. Because complex carbohydrates may act as cell surface receptors for a variety of enteric pathogens, our results indicate that these differences may be partially responsible for age and anatomic differences in susceptibility or resistance to enteric disease.     

Follicle associated epithelium of the gut associated lymphoid tissue of cattle

The morphology of the gut-associated lymphoid tissue of the small and large intestine in three gnotobiotic calves was examined by scanning and transmission electron microscopy, and the distribution of specialized membranous cells present in the follicle associated epithelium was defined. Isolated follicles remaining in the ileum of a cow after involution of the continuous Peyer's patch were examined by scanning electron microscopy. The presence of membrane-bound particles, reported to be exclusively associated with the continuous Peyer's patch, was investigated in other gut-associated tissue of the small and large intestine of the calf. The presence of two types of follicle associated epithelium in the small intestine of the calf was confirmed, and the follicle associated epithelium of the large intestine proved to be a homogeneous population of specialized membranous cells, similar to that of the continuous Peyer's patch of the small intestine. In the discrete Peyer's patches, some specialized membranous cells were completely hidden by adjacent enterocytes and could only be identified by cytoplasmic extensions into the intestinal lumen. In the proximal part of the continuous Peyer's patch, a transitional zone was detected where the follicle associated epithelium of some doomed villi was composed of a homogeneous population of specialized membranous cells, while the epithelium covering other doomed villi consisted of a mixture of absorptive and specialized membranous cells, usually only found in the discrete Peyer's patches. Membrane-bound particles were observed associated with gut-associated lymphoid tissue in the small and large intestine.     

Experimentally induced porcine proliferative enteritis in specific-pathogen-free pigs

Thirty-three, 10-week-old, specific-pathogen-free pigs were randomly allotted to 3 treatment groups: group 1--intragastrically given homogenized intestinal mucosa (crude inoculum) from pigs with naturally occurring proliferative enteritis; group 2--given cultures of Campylobacter sputorum subsp mucosalis; and group 3--controls. One pig from each group was killed 4, 7, 10, 14, 18, 21, 24, 28, 31, 36, and 38 days after inoculation. The earliest intestinal lesion observed in groups 1 and 2 was leukocytic exudate within crypt lumina and focal inflammation of the surrounding lamina propria. The lesions occurred primarily over ileal aggregated lymphoid nodules (Peyer's patches). These changes were followed by focal proliferation of immature crypt epithelial cells and infiltration of increasing numbers of macrophages into the lamina propria. Campylobacter sp-like organisms were observed within the cytoplasm of affected epithelial cells by light and electron microscopies. Lesions progressed to diffuse crypt cell proliferation, elongation of crypts, and loss of villi. Mucosal necrosis was not a prominent feature.     

A scanning and transmission electron microscopic study of rotavirus-induced intestinal lesions in neonatal gnotobiotic dogs

Neonatal gnotobiotic dogs orally inoculated with canine rotavirus had ultrastructural changes limited to the jejunal and ileal regions of the small intestine. Early scanning electron microscopic findings consisted of swollen villus epithelial cells, denuded foci on intestinal villi, and slight to moderate villus atrophy. Later changes were slight villus atrophy with no denuded intestinal villi. Transmission electron microscopic changes in villus epithelial cells from 12 to 48 hours post-inoculation included: rotavirus particles associated with intracytoplasmic vacuoles near the terminal web and apical tubules; viral particles in dilated cisternae of rough endoplasmic reticulum; and moderate numbers of necrotic cells having no microvilli, swollen mitochondria, membrane-bound lipid-like material in the cytoplasm, clumped chromatin around the periphery of the nucleus, and disruption of the cytoplasmic membrane. In jejunum and ileum at 72 to 154 hours post-inoculation, there were fewer necrotic villus epithelial cells and fewer virus particles. In addition, the ultrastructural morphology of the majority of the villus epithelial cells was similar to crypt epithelium. These studies showed that rotavirus infected the villus epithelial cells with subsequent propagation of the rotavirus and destruction of villus epithelial cells.