key: cord-0039225-k9jagbob
authors: Dubourguier, H.C.; Mandard, O.; Contrepois, M.; Gouet, P.
title: Light and electronic microscopic studies of the changes in the intestinal mucosa of gnotobiotic calves infected with a wild rotavirus
date: 2006-11-03
journal: Ann Inst Pasteur Virol
DOI: 10.1016/s0769-2617(81)80023-8
sha: 30eae3d28d8186a69cd90f6a7bf796ea2184fe6f
doc_id: 39225
cord_uid: k9jagbob

Experimental infection of four gnotobiotic calves by a rotavirus isolated from the faeces of diarrheic calves caused non-fatal diarrhoea without dehydration. Lesions of the alimentary mucosa appear well before the onset of diarrhoea and are characterized by a release of mucus in the anterior part of the intestine. By the time diarrhoea begin, different types of rotavirus particules are present in the enterocytes of the mucosa, and mucus contents of distal intestine are affected. At this stage of the disease, the virus causes epithelial cells in the small intestine to desquamate and changes the secretion of mucus in the small as well as the large intestine. The very intense virus multiplication is completed within 18 h. Intracellular rotavirus disappear very quickly after the onset of diarrhoea, and cannot be detected even during the acute phase of the disease. Regeneration of the mucosa takes place slowly, and the cellular and structural lesions are still visible more than 4 days after the diarrhoea stopped.

The presence of rotavirus has been shown in many cases of neonatal diarrhoea, in man [1] , calves [8, 14] , piglets [18] or lambs [16] . A strain of bovine rotavirus, isolated in the USA by Mebus el al. [8] was shown to be pathogenic in gnotobiotic calves deprived of colostrum [9] . The wave of infection ~ affects the entire small intestine and may proceed from the duodenum to the ileum [9, 15] . In similar experiments, using a strain of rotavirus isolated in France, Gouet el al., have shown that diarrhoea is transient, does not result in dehydration and that severity varies according to the age of the animals [6] . This report describes the development of the lesions of the intestinal mucosa at the time of experimental infection of the gnotobiotic calves with rotavirus, and particularly alterations in quantities of mucus.

All the calves used in these experiments were delivered and reared according to gnotobiological techniques. Seven gnotobiotic calves were delivered either by aseptic c~esarian section [13] , or by septic caesarian section and transfer to an isolator through an antiseptic liquid chamber [4] , or per vaginam with decontamination by antibiotics [5] (table I) . Three of the calves were not inoculated and served as control. The wild rotavirus was supplied by Dr Scherrer and inoculated as previously described [6] . Using general anaesthesia, samples for histological examination were obtained at different times after inoculation from the following location of the gastro-intestinat tract: the antral area o[ the abomasum, duodenum (distal to the pancreatic duct), middle jejunum, ileum (50 cm before the ileo-caecal valve), caecum (opposite the ileal-c~ecal valve) and spiral colon. [16] . For scanning electron microscopy (SEN), samples were prepared as previously reported [3] . For light microscopy (LM), samples were fixed in Baker's formol mixture, cut into 7 ~m sections and stained either with Giemsa or Aleian blue-PAS with a general eosin staining. Histological analysis of the quantity of mucus present in the tissues was carried out on the sections stained with Alcian blue-PAS. This analysis was performed with a (( Wild )) microscope connected to a surface analyser (prototype apparatus) which converts the optical image to an electronic signal by a television camera (I.T.C.). A synthetiser linked to a television screen allows the comparison of the reconstructed image with the real one, after regulation of the selection thresholds of analyser. When both images are superposed, the analyser converts the surface of each selected shade into arbitrary proportional values. It is then possible to determine the ratio (R) between the surface area of the secretory granules of the goblet cells (stained purple by the Aleian blue-PAS) and the total surface area of tile intestinal tissue (stained pink by the eosin). In this manner, measurements are not made by image intensity, but only by surface area. Surfaces areas of the goblet cells and therefore their volume, are related to their mucus content.

Bacteriological tests were performed on the faeces of the animals. Samples were examined microscopically, stained by the Gram stain and inoculated on the following media: tryptiease soja broth (B. D. Merieux) and agar slant, thioglycollate broth (Difco) inoculated before and after heating the faeces at 75 ~ C for 10 rain. The incubation was made at 37 ~ C and results checked after 24, 48, 72 h and after 1 week.

Of the 7 calves used, 4 of them remained germ-free, and non-pathogenic bacteria were found in the other 3 (table I) .

The control calves remained normal and were sacrified when 12, 21 and 120 h old (table I) . With microscopic technics, the 3 control calves were similar as early reported [3] .

In the duodenum, goblet cells were seen between the enterocytes. Mitochondria and Golgi bodies were common in these cells and much of the cytoplasm was occupied by secretory granules. In the jejunum and ileum, goblet cells had a similar appearance ( fig. 1 ). The surface area of goblet cells was smaller in the jejunum ( fig. 2 ). In the caecum and the colon, goblet cells were plentiful and their area was two to three times observed in the jejunum ( fig. 3 ). Results of histo-quantitative analysis of goblet cells were similar in the 3 control calves except in the c~eeum of the 12 h old calf due to meconium (table II) .

In calf n ~ 4, 10 h after inoculation and before the appearance of diarrhoea, the intestine did not contain any liquid. By light or SEM microscopy, the structure of the entire small intestine was normal except for vacuoles in enterocytes of the ileum.

No rotavirus was observed by TEM in the intestinal contents or in epithelial cells during this early stage of infection. 

Before diarrhoea Acute phase of diarrhoea Recovery phase Calf n ~ 5 became diarrheic 18.5 h post-inoculation and was sampled 1 h later at which time the small intestine was slightly flushed. Botavirus particles were found in the intestinal contents at each level except that of the abomasnm. The abomasum did not show any lesion for an increased number of mast cells in the lamina propria and turgidity of the epithelial cells. In the duodenum, the amount of cellular vacuolisation decreased from the tip to the base of the villi. The nucleus was surrounded by free spherical granules with an homogenous diameter of about 54 nm (fig. 4) . The basolateral cytoplasmic membrane was partially destroyed. In the jejunum, the tips of some villi were desquamated ( fig. 5 ) and the enterocytes near the apex of the villi were cuboidal. Cytoplasmic changes of the enterocytes were similar to those observed in the duodenum. No free granules of 54 nm were observed. However a few granules were seen inside vacuoles and were associated with viral particles which had a dark nucleus surrounded by a clear ring, all of which had a total diameter of about 69 nm. Some viral particles of this type were also free in the cytoplasm. In the ileum, the shape of the villi was altered by the release of mucus which reduced the number of goblets cells. In the upper half of the villi, the enterocytes had an irregular brush border and had lost thei~ cytoplasmic contents as well as their nuclei. The cytoplasm, which was very clear, contained mainly vacuoles and polysomes. Vacuoles in the basal region of the cell contained rotavirus ( fig. 6 ) whose diameter varied between 60 and 73 nm ( fig. 7 ). There were no free-virus particules in the cells. The lamina propria contained a few mast cells. Changes were not marked in the c~ecum and colon. Intracellular viral particles were not observed. In addition, we never found any virus in goblet cells in small intestine and large intestine.

Calf no 6 was sampled 26 h after inoculation and 11 h after the onset of diarrhoea at which time the abomasal mucosa was covered by a layer of mucus. The folds of the mucosa were very long, resembling small villi and desquamated at the apex. In the duodenum, the villi were short and very rounded. Numerous eosinophilic polymorphonuclear cells and a few mast cells were present in the lamina propria. The jejunum and the ileum showed similar lesions. The jejunal villi were distored and swollen by a pronounced infiltration of endothelial cells into the lamina propria. Some villus tips were desquamated and the distal two thirds of most villi were covered with cuboidal cells. The c~eeal and colonic mucosa appeared to be normal. No rotavirus particles were seen in epithelial cells at any of the sampling locations; however virus was observed in the c~ecal contents. Calf no 7 was diarrheic from 19 to 53 h post-inoculation. It was sampled when 8 days old, 111 h after diarrhoea had stopped at which time it was clinically normal. Some abomasal cells at the tips of the folds had expelled their contents. The duodenal villi were distored, flattened, short and appeared to be regenerating. The outer third of some jejunal villi was desquamated or covered with cuboidal cells and seemed to be soft. The ideal villi were leaf-shaped very indented and their tips covered with cuboidal cells. Eosinophilic polymorphonnclear cells were visible in the lamina propria. The mucosa of the caecum and colon was similar to that in the control calf.

Histo-quantitative analysis of the goblet cells (table II) showed that mucus was affected at all levels of the small and large intestines. Before the onset of diarrhoea (calf no 4), the duodenum and jejunum were affected. The changes reached the ileum and the caecum by the time diarrhoea begin (calf no 5) and persisted throughout the acute phase of diarrhoea (calf n o 5 and n ~ 6) when the large intestine was also affected. Alterations were also observed during the recovery phase as long as 111 h after diarrhoea had ceased (calf n ~ 7).

All samples were taken under general anaesthesia without exsanguination in order to avoid the epithelial desquamation observed during preliminary test (unpublished results) [12] . The double fixation method used far SEM allowed us to observe the surface of samples with as much precision as with the chemical preparation used by Mebus [7] and the physical preparation (critical point) performed by Newman and Trapp [11] . In control calves, there is no change in epithelial cells depending upon age (12, 21 and 120 h old). Only in the caecum, more mucus was observed in the youngest calf. This might be due to residual meeonium.

The rotavirus infection of gnotoxenie calves is not fatal and causes transient diarrhoea without dehydration [6] . The time of onset (15-19 h) and the duration (24 h) of rotavirus diarrhoea agree with our earlier observation [6] and with those of Mebus [7] . During the incubation and initial phases of diarrhoea, a prominent vacuolization of the enteroeytes specially in the mueosa of the small intestine was seen. Later in the course of diarrhoea, cells were desquamated and replaced by cuboidal cells. Similar changes have been shown in calves by Mebus el al. [91 and in the piglet suffering from transmissible gastroenteritis [17] . This important vacuolization might correspond to a early cell degeneration, followed by desquamation. Lesions were seen also in the abomasum before the occurrence of diarrhoea and persisted for several days after the disappearance of the clinical sylnptoms. However, these lesions were probably not a direct result of viral infection since no virus was observed either in the epithelial cells or in the abomasal contents.

Ten hours post-inoculation, no rotavirus was seen in the intestinal cells. Later in the course of diarrhoea, numerous circular granules of 54 nm in diameter were observed. According to Chasey [2] they may correspond to non-infectious incomplete virus particles called type IV particles by Chasey. To our knowledge, type IV particles have never been reported in vivo in the calf. For Chasey this viral type is found in damaged cells and more often in lesser number in small vacuoles [9. ]. In our experiments, this type of particles were found in great number inside the badly damaged cells and around the vacuoles which seemed to have burst. At this same stage of infection, we saw complete viral particles with a diameter of 60 to 73 nm both in the epithelial cells of the small intestine and in the intestinal contents. The dimensions are similar to those found by Stair el al. [15] . According to Chasey, this type of particles could correspond to the final stage of maturation of the rotavirus. These different types of particles are before the appearance of diarrhoea. In the acute phase of diarrhoea, no viral particles were observed in the cells, although they were still present in the c~ecal contents. These results indicate that after infection, the rotavirus multiplies very quickly inside the enteroeyte, and its complete maturation must take place in vivo within about 18 h post-inoculation. From the micrographs obtained, the number of viral particles released by infected cells can be estimated at about 10,000.

The gastro-intestinal lesions had not completely disappeared 111 h after diarrhoea had ceased since the tips of some abomasal folds and some intestinal villi were still covered by anormal cells. This observation suggest that following rotavirus infection epithelial cell replacement time is longer than the 48-72 h observed in normal calves [10] . The renewal time of the lamina propria may increase the apparent replacement tinie of epithelium. So, in the rotaviral diarrhoea, the digestive and absorptive functions of the cells are altered, which gives rise to the diarrhoea. From the first emission of diarrhoea, the virus is present in each part of small intestine, but not in the large intestine which confirms the observations made by Mebus [7] . The changes in mucus contents are accentuated and are developped concurrently with the wave of infection. But rotavirus particles were never found in goblet cells. So, this alteration in mucus content may probably be related to inflammatory response of the calf. The important expulsion of infected cells, Undoubtedly quicker than the migration of euboidal immature replacement cells, may explain the onset of desquamation in the apexes of the villi. The normal functions of the intestine are definitely not restored before the regeneration of the intestinal epithelium and lamina propria is complete which requires several days. During this period, the change in the digestive functions might increase the sensitivity of the animals to enteropathogenic Escherichia coli, as the results of Gouet el al. [6] Les 16sions gastro-intestinales apparaissent m@~e avant le d6but de la diarrh@ et sont earact6ris6es par une diminution de la quantit6 de mucus prdsente dans l'intestin grille. Apr6s l'infeetion, diff6rents types de particules sont pr6sentes dans les ent6rocytes de la muqueuse et le contenu en mucus de l'intestin gr41e distal est modifi6. Les partieules provoquent une desquamation eellulaire et des modifications du mucus intestinal. La multiplication virale est achev6e en 18 h. Les rotavirus ne peuvent ~tre mis en 6videnee darts les ent6roeytes apr6s la phase aigufi de la maladie. La r6g6n6ration muqueuse n'est pas achev6e dans les 4 jours suivant la diarrh6e.

MOTS-CLES : Flotavirus, Veau, Diarrh6e, Intestin; Muqueuse.

Virus particules in epithelial cells of duodenal mucosa from children with acute non-bacterial gastroenteritis

Different particles types in tissue culture and intestinal epithelium infected with rotavirus

Scanning electron microscopy of abomasum and intestine of gnotoxenic calves infected either with rotavirus, eoronavirus or enteropathogenic E. coli

l~levage de gros animaux axdniques et oligoxdniques en isolateurs

Immediate postal decontamination as a means of obtaining axenie animals and human infants

The experimental production of diarrhea in eolostrum deprived axenic and gnotoxenie calves with enteropathogenic E. coli, rotavirus, coronavirus and in a combined infection of rotavirus and E. coli

Viral calf enteritis

Calf diarrhea reproduced with a virus from a field outbreak

Pathology of neonatal calf diarrhea induced by a reo-like virus

Epithelial cell migration in the small intestine of sheep and calves

Lesions of experimentally induced coli= bacillosis in neonatal gnotobiotic pigs: a scanning electron microscopic study

Pathological changes in the small intestine of neonatal calves naturally infected with reo-like virus (rotavirus)

Technique for obtaining, fistulisation and reari~g of axenic or gnotoxenique lambs, kids and calves

Reovirus-like agent (rotavirus) associated with neonatal calf gastroenteritis in France

Neonatal calf diarrhea. Electron microscopy of intestine infected with a reovirus-like agent

A rotavirus in lambs with diarrhea

Epithelial cell dynamics in transmissible gastroenteritis of neonatal pigs

The isolation of reovirus like agents (rotavirus) from acute gastroenteritis of piglets