key: cord-0004915-jozdnhgy authors: Snodgrass, D. R.; Angus, K. W.; Gray, E. W.; Menzies, J. D.; Paul, G. title: Pathogenesis of diarrhoea caused by astrovirus infections in lambs date: 1979 journal: Arch Virol DOI: 10.1007/bf01317493 sha: 38b46ecdaeed8a1b2362d573a4bf91c33444cd30 doc_id: 4915 cord_uid: jozdnhgy Experimental infection of 2-day-old gnotobiotic lambs with lamb astrovirus produced mild diarrhoea after an incubation period of about 48 hours. No other clinical symptoms developed. Infection was studied by immunofluorescent and histological examination of tissues from the lambs. Astroviruses infected only mature villus epithelial cells and subepithelial macrophages in the small intestine, where they produced partial villus atrophy. Infected enterocytes were replaced with cuboidal cells from the crypts, and the lesion gradually healed by 5 days after infection. No serological relationship was detected by immunofluorescence between lamb astrovirus antigen in gut sections and antisera to either calf or human astrovirus. Viruses of 28--30 nm diameter with a circular outline and stellate surface structure have been observed in faeces from diarrhoeic children (7, 14) , lambs (15) and calves (19) . The name astrovirus has been suggested for these morphologically distinctive viruses (6) and will be used in this paper. Little is known of the pathogenic potential of these viruses. After oral inoculation they produced mild diarrhoea in lambs (15) , and partial villus atrophy in calves (19) and inconsistently caused diarrhoea in adult human volunteers (5) . This paper reports findings made on the pathogenesis of astrovirns infections in gnotobiotic lambs. Six gnotobiotic lambs were infected orally when approximately 24 hours old with intestinal contents from the third passage of lamb astrovirus in gnotobiotic lambs (15) . these intestinal contents, while the sixth (that killed 120 hours after infection) received 1.5 ml of ulfiltered bacteria-free contents. The inoeulum appeared by electron microscopic examination to contain many fewer astroviruses after filtration. One lamb was killed at each of the following hours after infection (p.i.): 14, 23, 38, 45, 70 and 120. Five gnotobiotie lambs killed between 72 and 144 hours of age served as controls for the histology. Six gnotobiotie lambs between 72 and 408 hours of age were controls for laetase estimations. Lambs were deeply anaesthetised with sodium pentobarbitone or halothane. Segments were obtained from jejunum (about 10 cm distal to the duodeno-jejunal flexure); from midgut; and from ileum (about 50 cm proximal to the ileo-caecM junction, from an area free of Peyer's patches). The lambs were then killed by exsanguination, and tissues collected from caecum, colon, kidney, liver and lung. Tissues for histological examination were fixed in 10 per cent formalsaline and processed as described previously (16) . Additional small (1 mm a) blocks of intestine were fixed in 3 per cent glutaratdehyde in phosphate buffer (pit 7.4) and processed to ArMdite. Sections 1 Bm thick were cut and stained by 10 per cent Giemsa, at 60 ° C. Villus heights and crypt depths in HE-stained sections were measured by ocular micrometer on ten vertically-cut, full length villi and crypts at each site of small intestine. Additional portions of all tissues were frozen in a CO2-ethanol freezing mixture. Lengths of small intestine and colon were filled with embedding medium (Tissue-Tek II, Lab-Tek Products) prior to freezing, to aid proper orientation of villi. Frozen tissues were mounted on microtome chucks and 6 ~m sections cut on a cryostat. An antiserum to lamb astrovirus was prepared as follows :--a gnotobiotic lamb was infected orally with astrovirus, and reinfected 10 days later. After a further 4 days the lamb was given by intramuscular inoculation an astrovirus preparation partially purified by differential centrifugation of intestinal contents. Blood was collected for serum preparation 5 days after final inoculation. Tissue sections were treated with this antiserum, followed by fluoreseein-conjugated rabbit anti-sheep globulin. Control sections were treated with gnotobiotic lamb antiserum to lamb rotavirus, followed by the conjugated anti sheep globulin. For purposes of serological comparison, astrovirus-containing gut sections were stained with calf antiserum to calf astrovirus (kindly supplied by Dr. J. C. Bridger, Compton, Berkshire) or human antiserum to human astrovirus (kindly supplied by Dr. J. B. Kurtz, Churchill Hospital, Oxford), followed by the appropriate fluoresceinconjugated globulins. Additional portions of tissue from the three sites of small intestine were collected and stored at --20 ° C. Laetase estimations used the methods of DAHLQVIST (1). Faeces samples were collected daily from all lambs, and suspensions examined by electron microscopy for the presence of astrovirus (15) . The l~mbs killed at 14, 23 and 38 hours p.i. did not develop diarrhoea. The other three infected lambs developed diarrhoea 44--48 hours after infection, faeces changing from firm and dark brown in character to very loose and yellow. Voluntary milk intake remained normal. Astrovirus was not observed in faeces from the lambs killed t4 and 23 hours p.i., but was first seen in the faeces of all other infected lambs between 38 and 48 hours p.i. At necropsy, astrovirus was detected in intestinal contents of all Iambs except that killed 14 hours p.i. The control lambs remained clinically normal, and passed firm brown faeces throughout the duration of the experiment. No virus particles were detected in their faeces. Specific immunofluorescent staining was detected only in scattered epithelial and subepithelial cells on small intestinal villi (Fig. 1) . The immunofluorescenee was usually fine and stippled in appearance (Fig. 1 b) . Infected cells were present between 14 and 70 hours p.i. (Table 1 ). Fewer infected celts were present in jejunum than in midgut or posterior ileum. The greatest number of infected cells was present early in infection, during the incubation period from 14 to 38 hours p.i. The infected enterocytes were generally scattered through the apical half of the villi. Occasional infected cells were observed in the villus lamina propria..No specific reaction was observed when the rotavirus antiserum was used. Tissues other than small intestine showed no specific reaction. No immunofluoreseenee was observed with lamb astrovirus-infeeted intestinal sections and the antisera to either calf or human astrovirus. The proximal intestine was unchanged throughout the experiment. No morphological alterations were seen in the midgut at 14 hours p.i. Many enterocytes in the ileum at 14 hours p.i. contained large ovoid vacuoles apical to the nucleus, although the villi were long and slender. Changes were first observed at 23 hours p.i. in the midgut. Here the villi were long and their lateral margins contained many clefts or mieroerypts (Fig. 2) , compared with control villi (Fig. 3) . The enteroeytes lining the lower one-third of the villi appeared normal, but those of the apical two-thirds had rounded margins and were cuboidal rather than columnar. Many enterocytes contained large single basal vacuoles and multiple small apicM vacuoles (Fig. 4) . The apical vacuoles often impinged upon and indented the nucleus, which consequently appeared collapsed or pyknotic. Both apical and basal vacuoles contained pleomorphie Schiff-negative bodies which stained deep mahagony-red by Pollak's trichrome method. Similar intra-eytoplasmie bodies were also seen in some enteroeytes and subepithelial maerophages, usually close to the nucleus. These bodies were most clearly demonstrated in araldite s~etions (Fig. 5) , and have been shown subsequently (GRAY, E. W., in preparation) to contain arrays of astrovirus partieles. The lamina propria of affected villi contained moderate numbers of maerophages with abundant cytoplasm. Goblet cell numbers were comparable to controls. The ileum was unaffected at this stage; single clear apical vacuoles were present in many of the enterocytes. By 38 hours p.i., villi in midgut and ileum were obviously shorter and more spatulate than those in equivalent control sites, or at earlier stages of the infection. At 45 hours p.i., villi in the midgut and ileum were short and blunt with crenated epithelium, and crypts which were elongated contained numerous mitotic figures. The lamina propria contained infiltrates of macrophages, lymphoeytes and neutrophils, as well as eosinophils in similar numbers to the control iambs. None of the enteroeytes in the midgut contained the multiple apical vacuoles and granular bodies seen at 23 hours p.i. IJ[owevcr, single apical vacuoles were seen in the ileum of the infected Iambs at 38 and 45 hours p.i. By 70 hours p.i. the villi in the midgut site were long and slender and indistinguishable from normal intestine, but those in the ileum were stunted and lined by a erenated, partly euboidal epithelium. At 5 days p.i., however, all three intestinal sites were morphologically normal. Large basal vacuoles were seen in m i d g u t enterocytes of control lambs at 72 and 96 hours of age, but apical vacuoles were confined to the ileum of control lambs. These findings are in accord with those m a d e in normal calves (8, 9) and piglets (2, 11, 12) . The vacuolated cells m a y be absorptive with a m a r k e d pinoeytotic c a p a c i t y (11) . A few neutrophils were present in b o t h caecal and colonic mucosa of the lambs killed 23 hours p.i., b u t not in a n y other lamb. No changes were found in a n y of the other tissues at a n y stage. The villus heights and crypt depths at the three sampling sites in the five control Iambs did not v a r y significantly with age. N o r m a l measurements for villi and crypts at each site were therefore obtained b y pooling observations for all five lambs. Measurements from individual infected lambs were compared with these n o r m a l values ( Table 2) . The length of the villi in j e j u n u m did not differ significantly from normal. Villus a t r o p h y was observed at 38 and 45 hours p.i. in midgut, and from 38 to 120 hours p.i. in ileum. The crypts in all three sites showed a progressive elongation t h r o u g h o u t the experiments. The most m a r k e d changes were observed in ileum, and are illustrated in Figure 6 . Lactase levels in the 6 control lambs were ¢.5±0.5, 5.14-0.6, and 2.44-1.2 units/g tissue for the proximal, mid, and distal small intestinal sites respectively. In midgut of infected lambs, observations were consistently below those of the controls, fMling to a minimum of 1.2 units/g tissue at 23 hours p.i. :No consistent change was observed in laetase concentrations in proximal and distal sites. These experiments confirmed the ability of lamb astrovirus to multiply in the intestinal tract of gnotobiotic lambs, and to cause diarrhoea. The site of multiplication was shown by immunofluorescence and electron microscopy (GRAY, E. W., in preparation) to be the small intestine, and these same techniques failed to find evidence of astrovirus infection in any other site. Immunofluorescence showed less evidence of virus multiplication in jejunum than in other levels of small intestine, and this was reflected in the absence of histological change. Crypt hypertrophy was, however, as marked in jejunmn as at more distal sites, but the stimulus for this is not known. Lamb rotaviruses have similarly been found to cause least damage in the jejunum (16) . Damage produced by astrovirus infection could be demonstrated by histopathology and estimations of lactase, and was consistently associated with a mild transient diarrhoea. A sequence of events with initial epithelial cell infection and destruction leading to partial villus atrophy, reclothing of the villi with relatively immature cells, and eventual healing of the lesion, can be postulated. This effect is similar to that of other viral infections of villus epithelial cells, particularly rotavirus (16, 18) and coronavirus (3, 10, 13) infections. In the lamb, astrovirus infection is at each stage less severe than rotavirus infection (16, 17) , with fewer enterocytes infected, a lesser degree of villus atrophy and a milder clinical disease. The astroviruses of lambs, calves, and man have not been shown to be serologically related by immunofluorescence. Further serological and biochemical studies are necessary to investigate the relationships between these viruses. This study has confirmed that lamb astrovirus is a pathogen of the small intestine of lambs. However, the only information available on the epidemiology of any of the astroviruses is that antibody to bovine astrovirus was detected in cattle in 11 or 22 herds (19) , so no attempt can be made as yet to define their role in causing diarrhoea under natural conditions. The excellent technical assistance of ~¢Iessrs J. Redmond and N. Inglis is gratefully acknowledged. Method for assay of intestinal disaccharidases Gnotobiotie piglets experimentally infected with neonatal calf diarrhoea reovirus-like agent (rotavirus) Intestinal infection of neonatal dogs with canine coronavirus 1--71: studies by virologic, histologic histochemieal and immunofluoreseent techniques Astrovirus associated gastroenteritis in a children's ward Astrovirus infection in volunteers 13.: 28 nm particles in faeces in infantile gastroenteritis Stool viruses in babies in Glasgow Pathology of neonatal calf diarrhoea induced by a reo]ike virus Scanning electron, light and transmission electron microscopy of intestine of gnotobiotie calf Scanning electron, light and immunefluorescent microscopy of intestine of gnotobiotic calf infected with calf diarrheal coronavirus Vacuolated villous epithelium of the small intestine of young pigs Pathological changes in the small intestine of neonatal pigs infected with a pig reovirus-like agent (rotavirus) Transmissible gastroenteritis of swine: Virus-intestinal cell interactions A survey of rotavirus associated with gastroenteritis in Aboriginal children in Western Australia Detection and transmission of 30 nm virus particles (astroviruses) in faeces of lambs with darrhoea Rotavirus infection in lambs: pathogenesis and pathology Experimental rotavirus infection in lambs Pathogenesis of porcine rotaviral infection in experimentally inoculated gnotobiotic pigs Isolation of small viruses, astrovirus-like and calicivirus-like, from acute enteritis of calves Authors' address: Dr. D. R. SNODGnASS, Animal Diseases 1%esearch Association, Moredun Institute, 408 Gilmerton/toad, Edinburgh, Scotland.1%eeeived December 14, 1978