key: cord-0865511-m7rqr886 authors: de Arriba, M.L; Carvajal, A; Pozo, J; Rubio, P title: Isotype-specific antibody-secreting cells in systemic and mucosal associated lymphoid tissues and antibody responses in serum of conventional pigs inoculated with PEDV date: 2002-01-01 journal: Vet Immunol Immunopathol DOI: 10.1016/s0165-2427(01)00386-5 sha: d817da9fc7dd94d21db3b72baa274a6cb924ee7a doc_id: 865511 cord_uid: m7rqr886 An enzyme-linked immunospot (ELISPOT) has been developed to detect porcine epidemic diarrhea virus (PEDV)-specific antibody secreting cells (ASC) in gut associated lymphoid tissues (duodenum and ileum lamina propria and mesenteric lymph nodes) and systemic locations (spleen and blood) of conventional pigs so as to characterise the mucosal and systemic antibody response generated by the infection with PEDV. A total number of 28 eleven-day-old conventional pigs were orally inoculated with the field isolate of the PEDV strain CV-777. Diarrhea was observed in 32% of the pigs and virus shedding was demonstrated in 100% between postinoculation day (PID) 1 and 8. Serum IgG and IgA antibodies to PEDV were detected by isotype ELISA from PID 12 and 15, respectively, reaching maximum values at PID 32 (IgG) and 21 (IgA). PEDV specific IgM ASC occurred in all the tissues between PID 4 and 7, with the strongest response in the intestinal lamina propria. IgA and IgG ASC responses were evident in the intestinal lymphoid tissues from PID 21, the highest number of specific ASC corresponded to the duodenum lamina propria. In the systemic lymphoid tissues the number of IgG and IgA ASC detected were lower than in the mucosal tissues, however, in the blood, presence of IgA ASC was constantly detected from PID 14 until the end of the experiment. Memory antibody response to the PEDV was also studied by secondary in vitro stimulation of the mononuclear cells (MNC) isolated from mesenteric lymph nodes, spleen and blood. The memory B cell response was prominent at PID 21 and 25 and consisted in IgG and IgA ASC. To our knowledge, this is the first report to research into the presence and distribution of specific ASC in different locations of the systemic and the gut associated lymphoid tissues after a PEDV infection as well as the presence of memory B cells. Porcine epidemic diarrhoea virus (PEDV) is an important pathogen causing severe gastroenteritis with a clinical picture similar to that of transmissible gastroenteritis (TGE). PEDV has been classi®ed in group I of the Coronaviridae family, which also includes transmissible gastroenteritis virus (TGEV), feline infectious peritonitis virus (FIPV) and human respiratory coronavirus 229E (HCV229E) (Cavanagh et al., 1994; Murphy et al., 1999) . Porcine epidemic diarrhoea (PED) was ®rst described in Great Britain in 1971 (Oldham, 1972) , and nowadays the virus is widely distributed throughout Europe and Asia, although has not been detected in America. In Spain, a serological survey carried out in 1993±1994 detected antibodies against the virus in 54% of the breeding herds (Carvajal et al., 1995b) . Moreover, over the last 4 years the virus has been found in a high percentage of the clinical cases of diarrhoea analysed at our laboratory. These data indicate that PEDV is currently one of the most important causes of gastroenterical disorders in pigs in Spain, which seems to be the same case in other countries (Van Reeth and Pensaert, 1994) . Clinically, the disease can appear in two forms, PED Type I, which affects only pigs older than 4±5 weeks, and PED Type II, which affects pigs of all ages. Both are characterised by profuse, watery diarrhoea, depression and anorexia. Morbidity is high, close to 100%, but mortality rate is relatively low in adult pigs (3%), whereas not in suckling piglets among which the severity of the disease increases and mortality can reach 90% (Pensaert, 1999) . The epithelium of the small intestine and the colon is the site for virus replication and although the presence of viral antigen in the mesenteric lymph nodes has been described, there is no evidence of virus replication in tissues other than in those in the gastroenteric tract. Infection of the enterocytes causes vacuolisation and ®nally destruction, which leads to villous atrophy and watery diarrhoea due to malabsorcion. So far, there is no effective vaccine or speci®c treatment available, and the only measures to control the disease are those directed to preventing the entrance of the virus on the farm (Pensaert, 1999) . The development of immunological strategies in order to induce protection would be desirable, mainly those involving the protection of suckling piglets less than 2±3 weeks old. Little has been reported relating to the immunological aspects of the disease other than detection of serum antibodies against the virus in convalescent animals (Carvajal et al., 1995a; Van Niewstadt and Zetstra, 1991; Debouck and Pensaert, 1984) . However, due to the special features of the mucosal immune system of the pigs, the presence of serum antibodies against gastroenteric pathogens is not always correlated with protection (Saif and Wesley, 1999; Saif, 1996; Saif et al., 1994 , To à et al., 1998 Ward et al., 1996) and only proves the contact with the microorganism. In the present work, we have tried to contribute to the information on the immune mechanisms occurring after PEDV infection by making a ®rst approach to the characterisation of the antibody response generated by the virus. Considering the fact that in the enteric infections in porcine is the local immunity which plays the main role in protection instead of the systemic immunity, as previously mentioned, the ®rst step was to develop techniques to investigate the immune response to PEDV in different tissues. Finally, we tried to emulate a natural infection by inoculating conventional pigs with a virulent strain of PEDV and monitorised the humoral response in different locations of the lymphoid system, involving gut associated lymphoid tissues as well as systemic tissues. The wild type isolated of the CV-777 strain of PEDV, kindly provided by Dr. Peansert (Gent, Belgium) was ampli®ed by passages in conventional 1-week-old piglets without antibodies against PEDV. Animals were orally inoculated and sacri®ced in the acute phase of diarrhoea, collecting the intestinal contains and the small intestine. The small intestine from each animal was macerated in PBS (1:2 (w/v)) and, like the intestinal contains, clari®ed by centrifugation at 5000  g for 20 min at 4 8C. Finally, all the fractions were pooled and stored at À70 8C. The cell-culture adapted PEDV, strain CV-777, was propagated in Vero cells as previously described (Hofmann and Wyler, 1988) . Brie¯y, Vero cells were grown with Eagle's minimum essential medium (Gibco, Life Technologies) buffered with bicarbonate and supplemented with 5% (v/v) foetal calf serum (Gibco), 0.04% (w/v) yeast extract (Difco, MI, USA), streptomycin (10 mg/l) and penicillin (10,000 UI/l) (Penicillin-streptomycin, Gibco). Con¯uent monolayers were infected by removing the growth medium and adding the viral inoculum diluted in medium without foetal calf serum but containing 10 ml/ml trypsin (Difco). In order to standardise the ELISPOT, three conventional 4-week-old pigs were hyperimmunisated and used as a source of lymphocytes primed against PEDV. Pigs were orally inoculated with a suspension of the virulent, wild type, PEDV and boostered 4 and 12 weeks later by peritoneal injection of the same inoculum concentrated by ultracentrifugation at 10 000  g and diluted (1:1) in Freund's adjuvant (complete in the ®rst injection and incomplete in the second) with antibiotics (gentamycin: 500 mg/ml (Gibco), streptomycin: 20 mg/ml, penicillin: 20,000 UI/ml (Penicillin-streptomycin, Gibco)). Serum samples were taken weekly after the ®rst inoculation to monitorise the antibody production. Once a high serum antibody titter was reached, blood samples were collected periodically in 25% (v/v) acid citrate glucose to obtain the mononuclear cells (MNC). Finally, the pigs were sacri®ced and the spleen and mesenteric lymph nodes were aseptically collected. As a negative control, blood samples from PEDV seronegative conventional pigs were collected in the same manner. A total number of 28 conventional pigs, seronegative to PEDV and from a herd with no previous history of the disease, were weaned at 11±12 days of age and maintained in isolation facilities. Pigs were inoculated orally with 3 ml of the virulent CV-777, a dose previously established which assesses a high rate of infection and the effective development of the immune response but without causing mortality among the pigs. The animals were observed daily for clinical signs and rectal swabs from all of them were taken for 11 days after the inoculation. Blood samples were also collected twice a week until the end of the experiment. At postinoculation days (PID) 4, 7, 14, 21, 25 and 31, 4±5 selected pigs were euthanised by a sodic penthotal injection (Eutalender, Normon, Madrid, Spain). The small intestine (duodenum and ileum), spleen, mesenteric lymph nodes and blood were aseptically collected for isolation of MNC. Three conventional, seronegative, pigs served as negative controls and were sacri®ced without previous inoculation. An ELISA that combines the use of two monoclonal antibodies (Mab) against the S protein of PEDV (CVI-PEDV 66.31 and 66.49) and a blocking step with rabbit-anti PEDV hyperimmune serum or gut-origin PEDV, was carried out as previously described (Carvajal et al., 1995a) to detect speci®c antibodies in serum or viral antigen in rectal swab samples, respectively. Antibody isotypes IgG and IgA to PEDV in serum samples were detected and titrated using an indirect ELISA as previously described (de Arriba et al., 1994) . Antigen was obtained from PEDV infected cell culture supernatants that were lysated, concentrated 50 times by ultracentrifugation (100 000  g, 4 8C, 2 h) and semi-puri®ed by ultracentrifugation under the same conditions through 20% sucrose. Mock-infected cultures were given the same treatment in order to obtain a control antigen. Viral or control antigens were immunocapturated in polystyrene microtiter plates (Costar, MA, USA) previously coated with the Mab CVI-PEDV-66.31. In the next step, serial 2-fold dilutions, starting at 1:20, of each serum sample were incubated in paired wells, containing viral or control antigen. Two biotinylated Mabs 3H7 (80 ng/ml) and 6D11 (65 ng/ml) against porcine IgG and IgA, respectively (Paul et al., 1989) , followed by horseradish peroxidase-conjugated streptavidin (KPL, MD, USA) and ABTS substrate were used for the detection of both isotypes of antibodies. Titres were expressed as the reciprocal of the lowest positive sample dilution and titres <20 were assigned a value of 10 for calculation of geometric mean titre (GMT). MNC were isolated from MLN, blood, spleen and the lamina propria of the small intestine (duodenum and ileum) by using modi®cations of previously described methods (Chen et al., 1995; Van Cott et al., 1993; Yuan et al., 1996) . Blood was collected aseptically in 25% (v/v) acid citrate glucose and the peripheral blood lymphocytes were isolated by Ficoll-Paque (Ficoll-Paque Research Grade, Pharmacia Biotech., Upsala, Sweden) density gradient centrifugation. Lymphocytes collected from the interface were washed twice in Hanks' balanced salt solution and resuspended in enriched medium (RPMI 1640 containing 8% fetal calf serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 0.1 mM nonessential aminoacids, 20 mM HEPES, and 20 mg of ampicillin and 100 mg of gentamicin per millilitre). Spleen, mesenteric lymph nodes and two fragments of the small intestine (one from duodenum and the other from ileum) were collected aseptically and placed in ice-cold wash medium (RPMI 1640 containing 10 mM HEPES and 200 mg of gentamicin and 20 mg ampicillin per ml). Spleen and mesenteric lymph nodes were homogenised pressing through stainless steel screens (80 mesh) of a cell collector (Cell-Selector, E-C Apparatus, FLA, USA). Cell suspensions were placed in a 30% Percoll (Pharmacia Biotech.) solution and centrifugated at 1200  g for 30 min at 4 8C. The resulting pellets were subjected to a discontinuous gradient centrifugation in Percoll: the cells were resuspended in 43% Percoll, underlined with 70% Percoll and centrifuged at 1800  g for 20 min at 4 8C. Finally, the MNC were aspirated from the Percoll interface, washed twice with wash medium and suspended in enriched medium. Fragments of approximately 20 g of duodenum and ileum were cut in small pieces, washed twice with wash solution and twice with Hanks' balanced salt solution. In order to remove epithelial cells, the tissues were placed in Hank's balanced salt solution containing 1 mM dithiothreitol and 5 mM EDTA and vigorously shaken for 30 min. Tissues were then digested for two 31-min periods at 37 8C in gentle shaking with enriched medium containing 400 UI of Type II collagenase (Sigma, MO, USA) per millilitre and 5 mM EDTA. Digested supernatants were collected and the remaining tissues were pressed through the stainless steel 80-mesh screens. The single cellular suspensions obtained were pooled with the digested supernatants and subjected to the gradient centrifugation in Percoll as described for mesenteric lymph nodes and splenic MNC. Viability of all MNC preparations was proved by the trypan blue exclusion test, being in any case higher than 95%. Two different kind of plates were assayed to set up the ELISPOT, on one hand PEDV infected and ®xed cell monolayers and on the other, semi-puri®ed antigen that was immunocaptured in plates previously coated with a Mab against the S protein of the virus. Vero cells, grown in 96-well tissue culture plates, were inoculated with the cell culture adapted PEDV as described (Hofmann and Wyler, 1988) . After incubation at 37 8C in a humid incubator with 5% CO 2 , plates were ®xed in 80% acetone in PBS for 20 min at room temperature and stored at À20 8C. Different dilutions were tested in order to ®nd the optimal multiplicity of infection to yield over 90% of infected cells. Control plates were mock-inoculated and treated in the same way. Optimal antigenic expression was veri®ed in every batch of plates by indirect immunouorescence using the Mab 113B, directed to the S protein of the PEDV and kindly provided by Dr. M. Ackermann (Zurich, Switzerland). Only the plates showing more than 80% of¯uorescent surface were used to perform the ELISPOT. Ninety-six-well microtiter plates were coated and dilutions of viral or control antigen were immunocaptured as described for the isotype ELISA. The ELISPOT technique was based on previously published methods (Chen et al., 1995; Czerkinsky et al., 1983, Sedgwick and Holt, 1983; Van Cott et al., 1993; Yuan et al., 1996) modi®ed and adapted by us for the detection of PEDV-speci®c antibody secreting cells (ASC). Fixed cell plates were thawed and rehydrated by incubation with PBS for 5 min at room temperature while antigen captured plates were washed three times before use in the ELISPOT. Different amounts of MNC (5  10 3 , 5  10 4 and 5  10 5 ) from each tissue were added to duplicate wells of the ®xed-PEDV infected or mock-infected cell plates or to the plates with the immunocaptured antigen (viral and control antigen). Plates were centrifuged at low speed 50  g for 5 min and incubated at 37 8C with 5% CO 2 for different periods. In order to remove the cells and between steps, the plates were washed 5 times with 0.05% Tween 20-PBS (PBST). The antibody production was detected by using biotinylated mouse Mabs 3H7 (80 ng/ml) and 6D11 (65 ng/ml) against porcine IgG and IgA, respectively (Paul et al., 1989) and goat anti-porcine IgM serum (1:20,000, KPL) diluted in PBST. After 2 h incubation at room temperature, horseradish peroxidase-conjugated streptavidin (KPL) was added (1:20,000) and incubated for 1 h at room temperature. Finally, the spots were developed by tetramethylbencidine (TMB) with H 2 O 2 membrane peroxidase substrate system (KPL) and counted under an optic microscope. Counts were averaged from the duplicated wells at the dilutions showing less than 40 spots per well and were expressed relative to 5  10 5 MNC. Working conditions were optimised to detect the highest number of speci®c spots against PEDV. The in vitro viral stimulation technique was modi®ed from published methods (Van Cott et al., 1993 . MNC puri®ed from spleen, blood and mesenteric lymph nodes were diluted in enriched medium containing 50 mM 2-mercaptoethanol (2ME-enriched medium) (Sigma) to 5  10 6 MNC per millilitre. 750 ml of each cell preparation were added to two consecutive wells of a 12-well tissue culture plate and stimulated with semi-puri®ed PEDV viral antigen, prepared as described for the isotype ELISA and diluted in 750 ml of the 2ME-enriched medium. Plates were maintained in a humid incubator with 5% CO 2 for 5 days and, from the second day on, 500 ml per well of fresh 2ME-enriched medium were added. On the ®fth day, MNC were harvested, rinsed twice with wash medium, suspended in 2ME-enriched medium and tested by ELISPOT (testing 5  10 2 , 5  10 3 and 5  10 4 MNC per well). Kruskal-Wallis non-parametric analysis of variance was used to prove statistically signi®cant differences in the number of ASC between different days. Antibody GMT were compared by means of Student t-test at each point in time. Signi®cance was assessed at p < 0:05. For the analysis the SYSTAT for Windows v.5.03 (SYSTAT) and the spreadsheet Microsoft EXCEL v.7.0 (Microsoft) were used. Pigs inoculated with the virulent isolate of the PEDV strain CV-777 exhibited moderate signs of the disease, mainly semi-liquid diarrhoea, shown in 32% of the pigs. The onset of the diarrhoea was between PID 2 and 4 and its average duration was 1.7 days. Rectal virus shedding was detected by blocking-ELISA in 100% of the PEDV exposed animals. Viral antigen was present in faecal samples from PID 1 to 8, but most of the pigs shed the virus in faeces between PID 3 and 6. The average duration of the shedding period was 5.4 days and the peak of the GMT of viral antigen in rectal swabs was at PID 5 ( Fig. 1) . After inoculation with virulent PEDV, seroconversion was demonstrated in 100% of the pigs by blocking ELISA. Speci®c antibodies were detected in 52.6% of the pigs at PID 4, in 96% at PID 7 and in 100% at PID 12. The isotype-speci®c ELISA antibody titters in the serum of inoculated piglets are shown in Fig. 2 . Titters of IgG increased signi®cantly from PID 12 over the duration of the experiment, reaching the maximum at that moment (PID 32, GMT 2650). IgA serum Optimal antigenic expression with the infected and ®xed cell monolayers was obtained by inoculating 2:8  10 3¯u orescent focus-forming units of viral inoculum per well and ®xing the plates after 12±15 h of incubation. These conditions guaranteed a citopatic effect between 80 and 90%, which gave a very high rate of¯uorescence, but keeping the integrity of the monolayer. Plates with the immunocaptured antigen yielded the best results when the semi-puri®ed viral antigen diluted 1:25 was used (2.05 mg of protein per well) and the plates were incubated for 4 h at 37 8C or overnight at 4 8C. Using infected and ®xed cell monolayers unspeci®c spots were non-detected either in the plates with mock-infected cell monolayers or with MNC from naõ Ève piglets and, therefore, every spot detected was considered a positive result. On the other hand, plates with the immunocaptured antigen showed some unspeci®c spots in the control wells and also when MNC from negative pigs were tested. Moreover, a higher number of spots were usually detected when the assay was performed over infected and ®xed cell monolayers. Consequently, due to the high speci®city shown for the ®xed PEDV infected cell plates and its higher sensitivity, we decided to use these plates for the rest of the experiments. In every assay mock-infected plates were included as negative control to assess the speci®city. In order to evaluate the distribution of PEDV-speci®c ASC, MNC from mesenteric lymph nodes, lamina propria of duodenum and ileum, spleen and blood were recovered at various PID from PEDV exposed pigs and tested for antibody production by ELISPOT. Kinetics of those responses are summarised in Figs. 3±5 and Table 1 . No virus-speci®c ASC were detected by ELISPOT either with the control pigs or when the assay was performed with mock-infected cell plates. PEDV-speci®c IgM ASC were the predominant response between PID 4 and PID 7. At PID 4 speci®c IgM ASC were found in all the tissues with the only exception in the blood, the strongest response corresponded to the intestinal lamina propria, where the number of IgM ASC was 6±7-fold greater than in mesenteric lymph nodes and spleen. At PID 7, the number of IgM ASC declined in the duodenum and the ileum, but not in mesenteric lymph nodes and systemic tissues (spleen and blood) and at PID 14 blood was the only tissue in which these cells were detected. Although some IgA ASC were detected as early as PID 7 in duodenum lamina propria, the IgG and IgA ASC responses in the intestinal lymphoid tissues were evident from PID 21 on. IgG ASC were demonstrated in the three tissues at PID 21, being maximum in the duodenum (Fig. 4, 23 .5 IgG ASC per 5  10 5 MNC). From this day and until the end of the experiment, the number of IgG ASC declined in the duodenum, but not in the mesenteric lymph nodes, where maximum was reached on PID 32. Although no signi®cant statistical difference was demonstrated, the number of IgA ASC detected in the intestinal lymphoid tissues by ELISPOT was, in general, lower than the number of IgG ASC. The highest IgA ASC response was also observed in the duodenum, where these cells were shown on PID 7, 21 and 25. The lowest response was obtained in ileum, in which IgA ASC were demonstrated only in a limited number at PID 25 (<1 IgA ASC per 5  10 5 MNC) (Fig. 5) . The number of IgG ASC in spleen and blood was lower than in the intestinal lymphoid tissues and these cells were evidenced later, between PID 25 and 32. A slow number of IgA ASC were detected in the spleen, but only at PID 14. However, in the blood, the presence of IgA ASC was proved from PID 14 until the end of the experiment on PID 32. Due to the individual variability in the responses observed and mainly to the limited number of pigs used at each point in time, statistical comparison among tissues or days showed no signi®cant differences. MNC were stimulated in vitro with different amounts of virus and over different periods. The best results were obtained when culturing the cells for 5 days with 4.6 ml (238.6 ng of protein) of the semi-puri®ed viral antigen per 5  10 5 MNC. In vitro stimulation of the cells was only possible with the MNC from mesenteric lymph nodes, spleen and blood, but not with those from the intestinal lamina propria since they had an irregular survival due to their frequent contamination with the enteric microbial ora. Results from the secondary immune response obtained after the in vitro stimulation are summarised in Table 2 . Even though low numbers of IgM ASC were detected on PID 4, secondary antibody response was more prominent on PID 21 and 25, consisting of IgG and IgA ASC. The number of speci®c ASC increased from 38 to 1137 times after the secondary stimulation and in a similar way to the primary responses, IgG ASC were predominant in the systemic tissues (spleen and blood), whereas not in the MLN, in which the secondary IgA ASC response was higher at PID 21. In vitro stimulation of naõ Ève MNC from the control group did not yield virus-speci®c ASC after 5 days of culture. Table 1 Numbers of isotype-specific ASC to PEDV (per 5  10 5 MNC) in duodenum and ileum lamina propria, mesenteric lymph nodes (MLN), spleen and blood from pigs experimentally exposed to the virulent isolate of the PEDV strain CV-777 and sacrificed on PID 4, 7, 14, 21, 25 In the present work, we have made a characterisation of the humoral immune response occurring after the infection with PEDV. The study was carried out with conventional pigs, infected with a virulent, wild type isolate of the virus, with the aim of emulating the natural conditions in which the infection took place. Experimental models were 12-day-old, PEDV seronegative piglets. No older animals were used due to the high growth rhythm of porcines, which makes it very dif®cult to handle a large number of pigs for an extended period. However, these young animals, in spite of having a non-totally-mature-immune system, are immunocompetent from the birth and are capable of developing a complete immune response (Roth, 1999) . Virulent PEDV caused moderate to severe diarrhea in 30% of the inoculated piglets with a average duration of 1.7 days. The incidence and severity of the disease in our experiment was lower than previously described by other authors (Carvajal et al., 1995a; Debouck and Pensaert, 1984; Debouck et al., 1981) who found that almost 100% of the inoculated piglets, gnotobiotic and conventional, developed diarrhea over an average period of 7±10 days. This difference may be explained by the low dose of virus used in our experiment. This dose was experimentally determined in previous assays in order to guarantee infection but without causing severe disease, in order that immune response could be developed and detected in the piglets. However, viral antigen was detected in faeces of all the pigs for the average period of 5.4 days, starting between PID 1 and 3, as it has been described in similar experiments (Carvajal et al., 1995a) . Speci®c antibodies against PEDV were demonstrated by blocking ELISA in 100% of the inoculated pigs between PID 4 and 12. This result is closely related with previous studies which report seroconversion 1 week after PEDV infection (Carvajal et al., 1995a; Van Niewstadt and Zetstra, 1991) . Although PEDV-IgG antibodies were always the predominant response in blood, in consistence with the patron for distribution of the different immunoglobulin isotypes, an important IgA response was detected in blood from PID 15±25, simultaneous to the detection of IgA ASC in blood. These results indicate that even blood is not considered an important source of the typically mucosal-associated Table 2 Numbers of isotype-specific ASC to PEDV (per 5  10 5 MNC) after in vitro secondary stimulation of the MNC isolated from mesenteric lymph nodes (MLN), spleen and blood of conventional pigs experimentally exposed to the virulent isolate of the PEDV strain 7, 14, 21, 25 IgM IgA IgG IgM IgA IgG IgM IgA IgG 4 11 (44) 0.7 (±) 0 (±) 22 (36) 0 (±) 0 (±) 8 (±) 0 (±) 0 (±) 7 0 (±) 0 (±) 0 (±) 0 (±) 0 (±) 0 (±) 0 (±) 0 (±) 0 (±) 14 0 (±) 0 (±) 0 (±) 0 (±) 0 (±) 0.8 (±) 0 (±) 0 (±) 0 (±) 21 0 (±) 44 (35) 38 (14) 0 (±) 220 (±) 586 (±) 0 (±) 29 (46) 267 (2136) 25 0 (±) 156 (±) 1137 (247) 0 (±) 7 (±) 97 (34) 0 (±) 49 (163) 1038 (478) a ASC after secondary in vitro stimulation/ASC after primary in vivo stimulation is given within the parenthesis. immunoglobulin. Its detection is possible in the early stages after infection, probably in relation to the lymphocyte homing process (Corthesy and Kraehenbuhl, 1999; Kagnoff, 1996; Kantele et al., 1997; Salmi and Jalkanen, 1997) . In order to standardise ELISPOT, two different kinds of antigenated plates were compared, infected and ®xed cell culture plates and semi-puri®ed antigen immunocaptured plates. In spite of the fact that no statistical studies were carried out, ELISPOT performed with infected cell culture plates gave better results in terms of sensitivity and speci®city. In the immunocaptured plates we used a Mab directed against S-protein of PEDV and, therefore, most of the viral antigen captured in the plate could be fragments of the spike more than the complete viral particle. On the other hand, viral inoculum for peritoneal immunisation of the animals was concentrated using ultracentrifugation and may then induce a high immune response against N and M protein which could be better detected with the infected and ®xed cell plates. Using the ELISPOT, we demonstrated the early development of a primary speci®c immune response against PEDV, formerly IgM ASC followed by IgG and IgA ASC responses. PEDV-speci®c IgM ASC were detected in all tissues between PID 4 and 7, with this response being higher and faster in duodenum and ileum lamina propria than in the other tissues, including mesenteric lymph nodes. The earlier presence of speci®c ASC in these tissues could be related to its anatomic proximity to the virus replication site in the enterocytes of the small intestine. The antigen could be delivered immediately to the lamina propria where it would stimulate the local immune response whereas its diffusion to the mesenteric lymph nodes would be more delayed (Corthesy and Kraehenbuhl, 1999; Kagnoff, 1996; Kraehenbuhl and Neutra, 1992; . After this initial response, there was no important detection of PEDV-ASC in the intestine or mesenteric lymph nodes until PID 21. Maximum values of ASC were reached between PID 21 and 32, at the end of the experiment. Similar experiences with other porcine enteric viruses such as TGEV or rotavirus (Chen et al., 1995; Van Cott et al., 1993; Yuan et al., 1996) showed consistent IgA and IgG ASC responses as early as PID 12, with the maximum amounts of cells detected also being higher than those described in the report. This difference could be due to a lower sensitivity of the ELISPOT performed for PEDV which would not detect low numbers of speci®c-IgA and IgG ASC present in the ®rst weeks after the infection. We also have to consider that experiences with rotavirus were carried out with gnotobiotic pigs in which the antigen-speci®c ASC/total ASC ratio is higher than in conventional animals. However, at least for TGEV, we cannot rule out the possibility of a higher antigenic stimulation as compared with PEDV. Although duodenum registered the highest number of speci®c-IgG and IgA ASC of the three mucosal associated lymphoid tissues, we consider that this result is more related to the distribution and organisation of the intestinal lymphoid tissue than to real differences in the immune response between tissues. MNC puri®ed from duodenal lamina propria belong to the diffuse lymphoid tissue, in which most of the cells, mainly the lymphocytes, are mature and active (Kraehenbuhl and Neutra, 1992; Pescovitz, 1999) . On the other hand, MNC from ileum and mesenteric lymph nodes also include large numbers of naive cells from the germinal centres of the lymph nodes or the Peyer patches that were not removed from the ileum (Heel et al., 1997; Pescovitz, 1999) . As a constant in our results, the number of IgG ASC in gut associated tissues was higher than the number of IgA ASC. This fact has also been described in previous reports (Van Cott et al., 1993; Yuan et al., 1996 Yuan et al., , 1998 studying immunity by ELISPOT in TGEV and rotavirus infections. Even though there is no clear explanation for this fact, the possibility of extraintestinal stimulation of the immune system has been proposed. However, PEDV has only been located in mesenteric lymph nodes apart from the intestinal mucosa (Debouck et al., 1981; Pensaert, 1999) and the possibility of systemic stimulation is very low. Further investigation is necessary to elucidate this question. With regard to the response detected in the systemic lymph tissues and mainly in the blood, our results showed the presence of speci®c IgA ASC at PID 14, earlier than in mucosal associate lymphoid tissues. As mentioned previously, we explain this ®nding as a consequence of the lymphocyte homing, a process in which the lymphocytes migrate through the blood to the efector sites, such as the intestinal lamina propria, after completing its maturating in distal sites of the mucosal immune system (Corthesy and Kraehenbuhl, 1999; Kagnoff, 1996; Kantele et al., 1997; Salmi and Jalkanen, 1997) . Thus, at least shortly after the infection, the PEDV speci®c ASC present in the blood could be considered more related to the response in mucosal associated lymph tissues than to the systemic response. Secondary in vitro stimulation of the MNC from inoculated piglets was performed in order to determine the potential B-cell memory response after PEDV infection. No speci®c-ASC was evidenced after the secondary in vitro stimulation of MNC obtained from control animals with no previous contact with PEDV. This result con®rms, as it has been described for other virus (Berthon et al., 1990; Van Cott et al., 1993; Yuan et al., 1996) , that only previously in vivo stimulated cells are capable of developing a speci®c response when stimulated in vitro with PEDV antigen. Moreover, a limited number of IgM ASC were detected at PID 4, memory B cells appeared consistently at PID 21 in mesenteric lymph nodes, spleen and blood from PEDV infected pigs. Similar results have been found after oral inoculation of pigs with porcine rotavirus while in TGEV infected animals speci®c B memory cells were detected from PID 12 (Van Cott et al., 1993) . Again, the number of PEDV±ASC per 5  10 5 MNC after secondary in vitro stimulation were lower than reported by Van Cott et al. (1993) in similar studies for TGEV. As we have suggested for the in vivo response, this difference could be attributed either to a lower sensitivity of the ELISPOT performed with PEDV or to differences in the immune response against each virus. In agreement with other authors (Berthon et al., 1990; Van Cott et al., 1993 and as described for the in vivo stimulated ASC, clear predominance of PEDV-speci®c IgG ASC was observed in systemic lymphoid tissues, blood and spleen, but not in the mesenteric lymph nodes where, at PID 21, IgA ASC outnumbered IgG ASC although reverse circumstances occurred at PID 25. To sum up, a ®rst attempt has been made to study the immune response developed in conventional piglets after infection with PEDV. This report shows the presence of speci®c-ASC from the different isotypes and in several locations of gut associated and systemic lymphoid tissues and also describes the IgA and IgG kinetics in serum. In addition, our study shows the presence of speci®c memory B cells from the third week after infection. Present results have important implications for future studies of immunity and protection against PEDV infection and consequently in the development and evaluation of future immunopro®lactic strategies to prevent and control the disease. Kinetics of the in vitro antibody response to transmissible gastroenteritis (TGE) virus from pig mesenteric lymph node cells, using the ELISASPOT and ELISA tests Evaluation of a bloking ELISA using monoclonal antibodies for the detection of porcine epidemic diarrhea virus and its antibodies Seroprevalence of porcine epidemic diarrhea virus infection among different types of breeding swine farms in Spain Revision of the taxonomy of the Coronavirus, Torovirus and Arterivirus genera Enumeration of isotype-specific antibody-secreting cells derived from gnotobiotic piglets inoculated with porcine rotaviruses Antibody-mediated protection of mucosal surfaces A solid-phase enzyme-linked immunospot (ELISPOT) assay for enumeration of specific antibody-secreting cells Development of an ELISA for the detection of antibody isotypes to porcine epidemic diarrhea virus (PEDV) in sow's milk Porcine epidemic diarrhea: kinetics of actively and passively acquired serum antibodies and the effect of reinfection The pathogenesis of an enteric infection in pigs, experimentally induced by the coronavirus like agent, CV-777 Review: Peyer's patches Propagation of the virus of porcine epidemic diarrhea in cell culture Mucosal immunology: new frontiers Homing potentials of circulating lymphocytes in humans depend on the site of activation Molecular and cellular basis of immune protection of mucosal surfaces Viral taxonomy and nomenclature Epithelial M cells: gateways for mucosal infection and inmunization Antigen sampling across epithelial barriers and induction of mucosal immune responses Pig farming Production and characterization of monoclonal antibodies to porcine immunoglobulin gamma, alpha and light chains Porcine epidemic diarrhea Immunology of the pig The immune system Mucosal immunity: an overview and studies of enteric and respiratory coronavirus infections in a swine model of enteric disease Transmissible gastroenteritis and porcine respiratory coronavirus Immunity to transmissible gastroenteritis virus and porcine respiratory coronavirus infections in swine How do lymphocytes know where to go: current concepts and enigmas of lymphocyte homing A solid phase immunoenzymatic technique for enumeration of specific antibody-secreting cells Serum and intestinal isotype antibody responses and correlates of protective immunity to human rotavirus in a gnotobiotic pig model of disease Isotype-specific antibody-secreting cells to transmissible gastroenteritis virus and porcine respiratory coronavirus in gut-and bronchus-associated lymphoid tissues of suckling pigs Contribution of antibody-secreting cells induced in mucosal lymphoid tissues of pigs inoculated with respiratory or enteric strains of coronavirus to immunity against enteric coronavirus challenge Use of two enzime-linked inmunosorbent assays to monitor antibody responses in swine with experimentally induced infection with porcine epidemic diarrhea virus Prevalence of infections with enzootic respiratory and enteric viruses in feeder pigs entering fattening herds Development of mucosal and systemic lymphoproliferative responses and protective immunity to human Group A rotavirus in a gnotobiotic pig model Systemic and intestinal antibody-secreting cell responses and correlates of protective immunity to human rotavirus in a gnotobiotic pig model of disease Antibody-secreting cell responses and protective immunity assessed in gnotobiotic pigs inoculated orally or intramuscularly with inactivated human rotavirus We wish to thank Dr. L.J. Saif and Dr. M. Ackermann for providing Mabs and Dr. M.B. Pensaert for providing the wild type isolated of the PEDV strain CV-777. We also wish to thank G.F. Bayo Ân for her excellent technical assistance. This work was funded by the Comisio Ân Interministerial de Ciencia y Tecnologõ Âa (CICYT) project No. AGF-960486. Salaries were provided by the Excelentisima Diputacio Ân Provincial de Leo Ân.