> top > docs > CORD-19:ef8d5f283816c67a1af2a88c454c165b04946775

CORD-19:ef8d5f283816c67a1af2a88c454c165b04946775 JSONTXT

INTESTINAL ANTIBODY RESPONSE AFTER VACCINATION AND INFECTION WITH ROTAVIRUS OF CALVES FED COLOSTRUM WITH OR WITHOUT ROTAVIRUS ANTIBODY Abstract Van Zaane, D., IJzerman, 3. and De Leeuw, P.W., 1986. Intestinal antibody response after vaccination and infection with rotavirus of calves fed colostrum with or without rotavirus antibody. Vet. Immunol. Immunopathol., 11:45-63. The intestinal and systemic antibody response of cah'es vaccinated and/or challenged with rotavirus was studied employing isotype-specific ELISAs for the detection of IgG1, IgG2, IgM and IgA antibodies to rotavirus. Monoclonal antibodies to bovine immunoglobulin isotypes of proven specificity were used as conjugated or catching antibody. Five days after oral inoculation (dpi) of a 5-day-old gnotobiotic calf with rotavirus, IgM rotavirus antibodies were excreted in faeces, followed 5 days later by IgA rotavirus antibodies. The increase in IgM rotavirus antibody titre coincided with the inability to detect further rotavirus excretion. Faeces IgM and IgA rotavirus antibody titres fell to low levels within 3 weeks post infection. IgG 1 and IgG 2 rotavirus antibodies were not detected in faecal samples. In serum, antibodies to rotavirus of all four isotypes were detected, starting with IgM at 5 dpi. Two SPF-calves, which were fed colostrum free of rotavirus antibodies, were vaccinated with a modified live rotavirus vaccine and challenged with virulent rotavirus 6 days later. Upon vaccination, the cah'es showed an antibody response similar to the response of the infected gnotobiotic calf. Intestinal IgM rotavirus antibodies were excreted before or on the day of challenge and appeared to be associated with protection against challenge infection with virulent virus and rotavirus-induced diarrhoea. In 3 control calves, which were challenged only, the antibody patterns also resembled that of the gnotobiotic calf and again the appearance of IgM rotavirus antibodies coincided with the end of the rotavirus detection period. Two other groups of :3 SPF-calves were treated similarly, but the calves were fed co]ostrum with rotavirus antibodies during the first 48 h of life. These calves excreted passively acquired IgG 1 and IgG 2 rotavirus antibodies in their faeces from 2 to 6 days after birth. After vaccination, no IgM or IgA antibody activity in serum or faeces was detectable. Upon challenge, all calves developed diarrhoea and excreted rotavirus. Seven to 10 days after challenge low levels of IgM rotavirus antibody were detected for a short period. These data indicate that the intestinal antibody response of young calves to an enteric viral infection is associated with the excretion of IgM antibodies, immediately followed by IgA antibodies. This response is absent or diminished in ealves with passively acquired specific antibodies which may explain the failure to induce a protective intestinal immune response by oral vaccination with modified live rotavirus of calves fed co]ostrum containing rotavirus antibodies. The intestinal and systemic antibody response of cah'es vaccinated and/or challenged with rotavirus was studied employing isotype-specific ELISAs for the detection of IgG1, IgG2, IgM and IgA antibodies to rotavirus. Monoclonal antibodies to bovine immunoglobulin isotypes of proven specificity were used as conjugated or catching antibody. Five days after oral inoculation (dpi) of a 5-day-old gnotobiotic calf with rotavirus, IgM rotavirus antibodies were excreted in faeces, followed 5 days later by IgA rotavirus antibodies. The increase in IgM rotavirus antibody titre coincided with the inability to detect further rotavirus excretion. Faeces IgM and IgA rotavirus antibody titres fell to low levels within 3 weeks post infection. IgG 1 and IgG 2 rotavirus antibodies were not detected in faecal samples. In serum, antibodies to rotavirus of all four isotypes were detected, starting with IgM at 5 dpi. Two SPF-calves, which were fed colostrum free of rotavirus antibodies, were vaccinated with a modified live rotavirus vaccine and challenged with virulent rotavirus 6 days later. Upon vaccination, the cah'es showed an antibody response similar to the response of the infected gnotobiotic calf. Intestinal IgM rotavirus antibodies were excreted before or on the day of challenge and appeared to be associated with protection against challenge infection with virulent virus and rotavirus-induced diarrhoea. In 3 control calves, which were challenged only, the antibody patterns also resembled that of the gnotobiotic calf and again the appearance of IgM rotavirus antibodies coincided with the end of the rotavirus detection period. Two other groups of :3 SPF-calves were treated similarly, but the calves were fed co]ostrum with rotavirus antibodies during the first 48 h of life. These calves excreted passively acquired IgG 1 and IgG 2 rotavirus antibodies in their faeces from 2 to 6 days after birth. After vaccination, no IgM or IgA antibody activity in serum or faeces was detectable. Upon challenge, all calves developed diarrhoea and excreted rotavirus. Seven to 10 days after challenge low levels of IgM rotavirus antibody were detected for a short period. These data indicate that the intestinal antibody response of young calves to an enteric viral infection is associated with the excretion of IgM antibodies, immediately followed by IgA antibodies. This response is absent or diminished in ealves with passively acquired specific antibodies which may explain the failure to induce a protective intestinal immune response by oral vaccination with modified live rotavirus of calves fed co]ostrum containing rotavirus antibodies. Rotavirus infections occur worldwide and are an important cause of enteric disease in many species (Holmes, 1979) . In young farm animals the morbidity and mortality can be high, causing considerable damage. In cattle, vaccination against rotavirus-induced diarrhoea has been attempted but the results were usually disappointing (Mebus eta]., 1973; De Leeuw et aI., 1980b; Snodgrass et al., 19801 Snodgrass et al., 1982; BDrki et aI., 1983) or still need confirmation under field conditions (Saif et aI., 1983; Saif et al., 1984) . Part of the problem may be the empirical basis on which such vaccines and vaccination methods have to be developed, due to an incomplete understanding of the mucosal immune system. Particularly in cattle, only limited information is available and some of the results reported appear to be conflicting. For instance, whereas in most species [gA is the predominant immunoglobulin (Ig)-isotype* in secretions (Tomasi and Zigelbaum, 19631 Heremans, 1974) , IgG 1 has been found to be the major isotype in the milk of cows and in the intestinal contents of 2-week-old calves (Newby and Bourne, 1976; Butler, 1983) . [gA levels increased with age but IgG 1 was still the major isotype in the intestine of adult cattle. Newby and Bourne (1976) also demonstrated the presence of a high percentage of IgG 1 containing ceils in the small intestine. In contrast, Porter et aI. (1972) reported the predominance of IgM (and in some animals of IgG 2) in intestinal loop secretions of pre-ruminant calves. IgM-containing ceils indeed predominate in the intestine of young calves but they appeared to be outnumbered by IgAcontaining cells in older calves and adult cattle (Allen and Porter, 1975) . This latter observation is in agreement with the results of Cripps et aI. (1974) who found IgA to be the major isotype in intestinal loop secretions of adult sheep. As illustrated by these conflicting results, the mucosal immune system of cattle needs further study. This is particularly important as there is little doubt that resistance to enteric infection is largely independent of circulating antibody but relies on passively or actively acquired immunity in the gut (Mebus et aI., 1973; Snodgrass and Wells, 1978a) . Few papers relate specifically to local rotavirus immunity in the calf. Hess et al. (1981) reported that specific antibodies to rotavirus were present in bovine intestinal fluid obtained via jejunal fistulae and in faeces of calves orally infected with rotavirus. In general, [gG 1 was the predominant isotype in these samples, but this isotype was not shown to be directly associated with anti-rotavirus activity. Vonderfecht and Osburn (1982) presented evidence that rotavirus inoculation of neonatal calves resulted [n the appearance of large numbers of [gA rotavirus antibody producing ceils in the mucosa of the proximal small intestine. * If appropriate the word "isotype" includes also subisotypes IgO 1 and [gO2. Our approach to the study of the immune response in calves infected with rotavirus imolved first the development of isotype-specific ELISAs for the detection of rotavirus antibody in serum and secretions. The specificity of the anti-immunoglobulin reagents employed in such tests is of crucial importance and has to be proven in the same assay in which they will be used (Butler, t980; Townsend et al., 1982) . With conventional antisera against bovine immunoglobulin (Big) isotypes that appeared to be specific in agar gel precipitation techniques we observed marked cross-reactions in sensitive ELISAs. To circumvent these problems we produced monoclona] antibodies (MCA) to BIg-isotypes and characterized them in ELISAs (Van Zaane and I3zerman, 1984) . MCAs specific for [gOt, IgG 2 and lgA were obtained, while MCAs against IgM showed a weak cross-reaction with [gA. The MCAs were applied in isotype-specific ELISAs for the detection of lgG1, lgG2, IgM and [gA rota~irus antibody. E~idence was presented that these tests were isotype-specific, except for the [gM test, which showed at most 3% cross-reacti~ity with IgA rotavirus antibodies. During the development of the tests, an inherent problem in isotype-specific ELISAs, i.e. the competition effects that occur (Chantler and Oiment, 1981~ Townsend et al., 1982~ Van Zaane and 13zerman, 1084 , was also studied. When isotype-specific reagents are used as conjugates, competition between isotypes for the limited amount of antigen (inter-isotype competition) can pre~ent the detection of an isotype with a relatively low antibody concentration or affinity. Another disadvantage of the ELISA is the possible interference of rheumatoid factors. Both problems can be circumvented by applying isotype-specific reagents as catching antibodies. In this case, howe~er, competition within a particular isotype for the limited amount of catching antibody can occur (intra-isotype competition). Based on the evaluation of these competition effects optimum EL]SAs were selected (Van Zaane and I3zerman, 1984) . In this paper we report the results of preliminary studies on humoral aspects of the intestinal immune response of calves after oral vaccination and/or infection with rotavirus and the influence that passimely acquired rotavirus antibody may have. In all experiments Friesian cal~es were used. A colostrum-depri~'ed gnotobiotic calf was reared in a positi~e-pressure isolation unit. It was fed condensed milk (Nutrieia, Zoetermeer, the Netherlands) during the whole experiment. The calf was free of bovine ~irus diarrhoea (BVD) ~irus and it remained free of bacteria and serologically negative for bovine coronavirus and BVD-~'irus. The SPF-calves used formed part of another study described in detail elsewhere (De Leeuw and Tiessink, 1984) . The experimental design is outlined in Table [ . * Groups B and D in this paper correspond with groups B 2 and D1, respectively, in the paper of De Leeuw and Tiessink (1984) . For further explanation see the materials and methods section. Briefly, cahes in groups A and B received bovine colostrum free of antibodies to rotavirus, whereas calves in groups C and D were fed "positive" colostrum, i.e. with antibodies to rotavirus. Qroups B and D were vaccinated orally 1 hr after the first colostrum feeding. All groups were challenged on day 6 after birth with ~irulent rotavirus. At that time the peak of rotavirus infection normally occurs in the field (De Leeuw et al., 1980a) and colostral antibodies are probably no longer present in the gut. Calves used in experimental groups A and B were born to mothers in the institute's SPkherd, which is serologically negative among others for rotavirus and bovine coronavirus. Calves in experimental groups C and D were from cows obtained through a market. Calves were obtained by caesarian section and reared in isolation. The day of birth is taken as day 0. Calves used in the same experimental group were housed in individual boxes in one isolation room. They always arrived on the same day. Calves were fed 8 litres of colostrum, divided over 4 meals, during the first 2 days. Thereafter a commercial milk substitute was given, 2 times 2 ]itres daily. Co]ostrum without antibodies against rotavirus and bovine coronavirus was obtained from cows in the institute's SPF-herd; "positive" colostrum was obtained from cows in a dairy herd on a The gnotobiotic calf was infected with a Dutch virulent rotavirus field strain at 5 days of age. Three and /4 days later~ faecal samples containing rotavirus were obtained~ pooled and stored in small quantities at -7g°c. This was used as the challenge inoeulum for calves in experimental groups A-D~ and it had a titre of 105 tissue culture infective doses 50 per cent (TCIDs0) per ml. These calves were simultaneously challenged with virulent bovine coronavirus obtained from another experimentally infected gnotobiotic calf. Both viruses were given orally. The vaccine used was Scourvax~-2* containing both attenuated bovine rotavirus and bovine coronavirus. Each calf received 2 ml of the reeonstituted vaccine orally as specified by the manufacturer, The rota~irus titre of the vaccine in our hands was l0 6,2 TCIDso/ml. Vaccination was always performed one hour after the first colostrum feeding; the calves were then 5 to 8 hours old. Within one hour after collection, faecal samples were cooled to 4oc, homogenized the same day with 4 volumes of PBS containing 0.05% w/v Tween 80 and subsequently stored at -20°C. Before testing, samples were thawed and centrifuged for /0 rain at 250 x g. The supernatants were taken as "undiluted" test samples. No specific measures were taken to inhibit proteolytic activity~ but it was shown that incubation of purified lgG 1 rotavirus antibodies with a fresh faecal extract for 24 h at 37°C did not reduce the antibody titre. Therefore, proteolytic degradation of lg after collection of faecal samples was not considered a problem. The isotype-specific ELISAs for the detection of antibodies to rotavirus have been described in detail elsewhere (Van Zaane and [Jzerman, 1984) . MCAs to bovine Igisotypes G1, (32~ M and A were used as conjugate in the indirect double antibody sandwich assay (IDAS) or as catching antibody in the antibody capture assay (ACA) (Fig. 1 ). Based on an analysis of inter-and intra-isotype competition effects in IDAS and sample was also tested in a 1/20 dilution against a faecal sample free of rotavirus. For this~ a pool of faeces from conventional calves was used (see Fig. 1 , step 2 and 3 for the IDAS and ACA, respectively). All samples were negative in this test. A germ-free calf was infected orally with rotavirus on the fifth day of life. Rotavirus antibodies of all isotypes were detected in its serum starting with IgM 5 days post inoculation (dpi). In faeces only IgM (from 5 dpi on) and IgA rotavirus antibodies (from 10 dpi on) were found to be present (Fig. 2) . Rotavirus was detected in the faeces by ELISA from 2-6 dpi. The vaccinated calves of group B as well as the controls, which were challenged only (group A), showed a systemic and intestinal immune response similar to that observed in the gnotobiotic calf. In faeces of group A calves IgM rotavirus antibodies were detected starting on 4-6 dpi, followed by lgA rotavirus antibodies 2-6 days later. Antibody titres fell to low levels within 3 weeks post infection (Fig. 3) . In group B calves IgM rotavirus antibodies were detected for the first Lime 5-6 days post vaccination (dpv) (Fig. 4) . In In group A, rotavirus was excreted between i and 6 dpi. in group B vaccine virus was detected 2-5 dpv but after the challenge no rotavirus was found by ELISA or by electron microscopy (Figs. 3 and 4) . Calves in group A were reinoculated orally with rotavirus 2 to 4 weeks after primary infection. The rotavirus isolate originated from another herd than the first one. The inoculum also contained Cryptosporidium. All 3 calves developed diarrhoea and excreted Cryptosporidium oocysts in their faeces. Although no virus excretion could be detected, a marked rotavirus antibody response followed in two calves (nos. 28 and 33). In this case, IgA rotavirus antibody was the predominant antibody found in faeces; it was excreted in high titre 4-14 days after reinfection (Fig. 3) . IgM rotavirus antibodies were also excreted but at low titre. Again, no IgG 1 or IgG 2 rotavirus antibodies were detected in the faeces. The changes of the isotype-specific antibody titres in serum are difficult to interpret as they may still be related to the primary response. It appears, however, that after reJnoculation the IgG 1 and IgG 2 titres did increase and in nos. 28 and 58 also the [gA titre. This isotype-switch of the predominant rotavirus antibody might be characteristic for a secondary antibody response of primed cells. Recently~ evidence for the existence of a memory response of the mucosal immune system in rabbits (Keren et al.~ 1982) ~ rats (Andrew and Hall~ 1982) and man (Wright etal., 1983) has been presented. On the other hand, Allen and Porter (1975) were unable to show the existence of a memory response in the intestine of calves, using dead bacteria. Also, it cannot be excluded that the isotype-switch mentioned before simply reflects age-related changes of the primary On the other hand, results of other groups (Porter el: al., 1972~ Cripps et al., 1974~ Allen and Porter, 1975 demonstrated, in accordance with our results, a major role for IgM and IgA in the intestine of cattle (see introduction). The predominance of IgA rotavirus antibodies in faeces late after infection or after reinoculation of rotavirus described in this paper is furthermore in agreement with the results of Vonderfecht and Osburn (1982) . They demonstrated that 4 weeks after rotavirus inoculation in Thiry-Vella loops of calves the majority of the cells that produced rotavirus antibodies contained IgA and only a few per cent stained for IgG and lgM. Our results compare also rather well with those obtained in unweaned piglets. Evidence has been presented that their intestinal lamina propria contains equal numbers or more of IgM producing cells as compared to IgA producing cells (Allen and Porter, 1973; Brown and Bourne, 1976 ). In addition, rotavirus infections in pigs were associated with the excretion of IgM rotavirus antibodies which were found, however, in concert with an excess of IgA rotavirus antibodies (Corthier and F ranz, 1981) . It has been suggested earlier that IgM might play an important role in (the early phase of) the mucosal immune response (Porter et al., 1972; Allen and Porter, 1973; Brandtzaeg, 1975) . Differences in the results mentioned above may be due to a variety of factors, i.e. species, differences in methodology (techniques, isotype-specific reagents), parameters measured (immunoglobulin concentration, antibodies, plasma-cells) and samples examined (secretions of intestinal loops, intestinal content obtained post mortem, faeces). Final proof for the local production of IgM rotavirus antibodies in the bovine intestine awaits further studies using double immunofluorescence tests on gut tissue sections (Vonderfecht and Osburn, 1982) . Similarly, the relevance of IgG in the intestine should be studied further using monoclona] isotype-specific reagents and hopefully, this will shed some light on this controversial issue (Morgan et al., 1980) . Another important aspect of this study is the evaluation of the immune response after oral application of a modified live rotavirus vaccine in young calves. The results obtained with group A and B calves, fed colostrum without specific antibodies, suggest that IgM is the main antibody-isotype responsible for recovery from and early protection against viruses causing enteric disease. In the field, nearly all adult cattle possess rotavirus antibodies and can transfer these via colostrum and milk to their offspring. Evidence has been presented that effective oral vaccination of calves against rotavirus-induced diarrhoea is hindered by the presence of these lacteal antibodies , 1980b; BOrki et al., 1983; De Leeuw and Tiessink, 1984) . Our results with group C and D calves (Figs. 5 and 6) further substantiate this view. These calves excreted antibodies in their faeces up to 6 days after birth. The presence of these antibodies in the intestine most likely prevented the normal multiplication of the vaccine rotavirus. It appears from studies of transmissible gastroenteritis virus infections in swine that an intestinal immune response to a virus infection requires an extensive infection of the intestinal mucosa (Bohl and Saif, 1975) . Together, this probably explains why no intestinal antibody response was observed after vaccination of group D calves (Fig. 6 ), comparable to that in group B calves (Fig. 4) . However, a suppressing effect of serum antibodies on the mucosal antibody response cannot be excluded. This effect is suggested by the delayed and weak IgM-response after challenge of group C and D calves, which at that moment did not excrete detectable quantities of IgG rotavirus antibodies in their faeces. In some calves, the intestinal response was even absent. The observations with group C and D calves also confirm that feeding of colostral antibody for only a few days does not provide protection against challenge infection several days later. This requires continuous uptake of rotavirus antibodies (Bridget and Woode, 1975; Lecce et al., 1976; Snodgrass and Wells, 1978b) . Although our data suggest that lgM is the major antibody isotype involved in recovery from and early protection against enteric viral infection, th e precise role of lgM and [gA has yet to be defined. In addition, other immune mechanisms have not been studied and a possible protective role for lgA and [gG at a later age or longer after an infection or a vaccination still remains possible. In conclusion, our results indicate that the intestinal immune response of young calves to an enteric viral infection is associated, as in other species, with the excretion of IgA antibodies, which is preceded by the excretion of IgM antibodies. This response is absent or diminished in calves with passively acquired specific antibodies in serum and intestine. Oral vaccination with modified live rotavirus of calves fed colostrum with rotavirus antibodies did not induce a protective intestinal immune response. This is most likely explained by neutralization of the vaccine virus as suggested by De Leeuw and Tiessink (1984) . In addition or alternatively, serum antibodies might suppress the development of intestinal immunity. Although oral vaccination rapidly induced a protective intestinal immune response in calves which were not fed colostral rotavirus antibodies, it is not sensible to omit or delay colostrum feeding. Therefore, alternative vaccination methods are needed to induce an active intestinal immune response or a persistent secretion of rotavirus antibodies in milk. To facilitate this, further study of the mucosal immune system of young and adult cattle is necessary.

projects that include this document

Unselected / annnotation Selected / annnotation