INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Caliciviruses are small, nonenveloped viruses 27 to 38 nm in diameter and possess a single-stranded, plus-sense RNA genome of 7.3 to 8.3 kb and a single capsid protein of 56 to 71 kDa. Caliciviruses are divided into four distinct genera: Vesivirus, Lagovirus, Norwalk-like viruses (NLVs), and Sapporo-like viruses (SLVs) (14). Human caliciviruses (HuCVs) have emerged as the leading cause of food- and waterborne, acute, nonbacterial gastroenteritis in humans worldwide (10, 23, 38). These uncultivable enteric caliciviruses belong to either the NLV or SLV genus. The NLVs are commonly identified as causative pathogens in outbreaks of gastroenteritis in humans of all ages (10, 21, 38, 41). The SLVs are mainly associated with sporadic, acute gastroenteritis in infants and young children (9, 20), but also cause viral gastroenteritis in the elderly or other age groups (26, 39).

Animal enteric caliciviruses also cause gastroenteritis in swine, calves, chickens, cats, dogs, and mink (4, 16, 32, 42). A number of these animal enteric caliciviruses are closely related genetically to HuCVs (7, 15, 16, 22), raising public health concerns for potential cross-species transmission and animal reservoirs for enteric caliciviruses related to HuCVs (36, 37).

The HuCVs remain refractory to cell culture propagation, and no susceptible animal models are available, which impedes the understanding of their replication strategies, pathogenesis, and host immunity. Although the genomes of a number of HuCVs have been sequenced, knowledge about viral replication and pathogenesis is still limited and derived mainly from volunteer studies (20). The HuCVs such as Norwalk virus (NV), Snow Mountain virus, and Hawaii virus (HV) caused diarrhea in volunteers, and histologic lesions were evident in jejunal biopsies (20). However, the extent of small intestinal involvement in HuCV infections is unknown, and the major site of viral replication is undetermined (9). The bovine enteric caliciviruses, Jena virus and Newbury agent, caused diarrhea and villous atrophy in the small intestine of gnotobiotic (Gn) calves (18, 22). A porcine enteric calicivirus (PEC) was first identified in the feces of a piglet with diarrhea in 1980 (32), which is morphologically and genetically similar to SLV HuCVs (15). The wild-type (WT) PEC Cowden strain induced diarrhea and villous atrophy in orally inoculated Gn pigs, with PEC-specific immunofluorescence detected in villous epithelial cells and villous atrophy observed in the proximal small intestine (12). The PEC/Cowden is the only enteric calicivirus that has been adapted to cell culture using a porcine kidney cell line (LLC-PK) with incorporation of intestinal content (IC) preparations from uninfected Gn pigs into the medium (11, 29). Genomic sequence analyses indicated that the TC PEC has one distant and three clustered amino acid substitutions in the hypervariable region 2 of the predicted capsid protein, and two amino acid changes in the RNA-dependent RNA polymerase region, compared with the WT PEC (15). Thus, it was of interest to examine if the virulence of the TC PEC in pigs was changed after serial passage in the LLC-PK cells. Future studies could then address if changes detected in the virulence of the TC PEC are associated with the predicted amino acid substitutions in the capsid region of the TC PEC.

In this report, the serially passaged TC PEC/Cowden was used to orally inoculate Gn pigs, and its pathogenesis was compared with that of WT PEC. We also examined the influence of the route of inoculation (oral versus intravenous [i.v.]) on the pathogenesis of the WT PEC, including analyzing the sites of intestinal and extraintestinal replication (lung, liver, kidney, and spleen) and lesions, determined by the immunofluorescence (IF) test and histopathology, respectively. We assessed the occurrence of diarrhea and fecal virus shedding by reverse transcription (RT)-PCR, enzyme-linked immunosorbent assay (ELISA), and immunoelectron microscopy (IEM). The presence of viremia following PEC infection was also examined by RT-PCR and ELISA in serum from the pigs inoculated orally or i.v. with WT PEC and confirmed by inoculation of additional pigs with acute-phase serum from the viremic pigs.

(This report represents a portion of a dissertation submitted by M. Guo to the Graduate School of The Ohio State University as partial fulfillment of the requirements for the Ph.D. degree.)

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Calicivirus inocula. The WT PEC/Cowden was originally obtained from the intestinal contents of a 27-day-old diarrheic suckling pig (32). The WT PEC/Cowden was passaged 15 times in Gn pigs, consistently causing diarrhea at each passage (12). The WT PEC/Cowden inoculum was obtained from the 15th Gn pig passage and was prepared by making a 20% suspension of the pooled intestinal contents in serum-free Eagle's minimal essential medium (EMEM). For i.v. inoculation, the WT PEC inoculum was centrifuged at 1,200 × g for 30 min and filtered through 0.45-µm-pore-size filters. As a control, WT PEC was also inactivated by using formalin. Briefly, formalin was added to the WT PEC filtrate at a final concentration of 0.4%, and the mixture was incubated at 37°C for 48 h with occasional shaking every 2 to 3 h.

Animals, experimental design, and samples. Gn pigs were procured and maintained as previously described (24). A total of 24 3- to 6-day-old Gn pigs were used in this study and assigned to eight groups as indicated in Table 1. Groups I, II, and III were inoculated orally with TC PEC, WT PEC, or the serum-free EMEM (mock inoculation), respectively. For i.v. inoculation, the inocula were injected slowly via venipuncture into the anterior vena cava, jugular vein, or femoral vein. The Gn pigs in groups IV, V, and VI were inoculated i.v. with WT PEC, formalin-inactivated WT PEC, or serum-free EMEM, respectively. The Gn pigs in groups VII and VIII were inoculated orally or i.v. with PEC-positive acute-phase serum from WT-PEC-infected pigs inoculated orally or i.v. All inoculated pigs were observed daily for diarrhea, and their feces were scored as follows: 0, normal; 1, pasty; 2, semiliquid; 3, liquid. Diarrhea was indicated as fecal scores of 2. Rectal swab fluids were collected daily and processed by suspension into 8 ml of serum-free EMEM and stored at 20°C for detection of viral shedding. For i.v.-inoculated pigs, blood was drawn at PID 0, 2, 4, 6, and 8 or at euthanasia, and the serum was separated and stored at 20°C. The inoculated pigs were euthanatized at the onset of diarrhea, 1 to 5 days after the onset of diarrhea, or at PID 7 if they did not show diarrhea. Three orally or i.v.-inoculated control pigs were euthanatized at either PID 4 or 8. Euthanasia was performed by electrocution. For the Gn pigs in groups VII and VIII orally inoculated with PEC-positive acute-phase serum, blood was drawn at PID 0, 2, 4, 6, 8, 10, 14, and 21 or at euthanasia, and both serum and white blood cells (WBC) (see "RT-PCR" below) were collected in glass tubes. Convalescent-phase serum samples were examined for seroconversion to PEC by using a virus-like particle ELISA (VLP-ELISA) for detection of antibodies to PEC (18).

Histologic examination. Formalin-fixed sections of the duodenum, jejunum, ileum, colon, lungs, spleen, liver, and kidneys were routinely processed, embedded in paraffin, sectioned at 5 µm, stained with Mayer's hematoxylin and eosin, and examined microscopically (12). The histologic evaluation was done in a blind fashion on coded samples, and a comparison was made with tissues from age-matched controls. Villous length and crypt depth were measured for histologic sections of the duodenum, jejunum, and ileum by using an ocular micrometer. The mean villous length and crypt depth were determined by measurement of 10 randomly selected villi and crypts on intestinal histologic sections, respectively, similar to methods described previously (12).

Detection of PEC in tissues by IF staining. To evaluate the distribution of PEC antigens in the tissues, impression smears prepared from fresh specimens of the duodenum, jejunum, ileum, colon, liver, spleen, lungs and kidneys collected at necropsy were stained directly by using hyperimmune antisera to PEC conjugated to FITC. The smears were prepared, stained, and examined by IF microscopy as described (3, 12).

Scanning electron microscopy. Segments of the duodenum, jejunum, ileum, and colon were fixed in a mixture containing 3% glutaraldehyde, 2% paraformaldehyde, and 1.5% acrolein in 0.1 M collidine buffer, pH 7.3, as described previously (12). The specimens were dehydrated in an ethanol-dry ice series and gently vacuum dried. The dried tissues were sputter coated with approximately 150 Å of platinum and observed using a scanning electron microscope (ISI-40; International Scientific Instruments Inc., Tokyo, Japan), and the tissues were photographed.

Detection of PEC in rectal swab fluids, intestinal contents, and serum samples. (i) IEM. IEM was performed as described (32). Small and large intestinal contents and fecal samples from inoculated pigs were diluted 1:5 in 0.01 M PBS, pH 7.2, and filtered (0.45-µm pore size) after centrifugation at 1,200 × g for 30 min. The filtrates were incubated with 1:500 diluted hyperimmune antiserum to PEC at 4°C overnight, followed by ultracentrifugation at 69,020 × g twice for 35 min each time. The final pellet was resuspended in 0.05 ml of distilled water; negatively stained with an equal volume of 2% phosphotungstic acid, pH 7.0; and examined using an electron microscope (model 201; Philips-Norelco, Eindhoven, The Netherlands).

(ii) RT-PCR. The PEC RNA was extracted and purified from rectal swab fluids, intestinal contents, serum and WBC (only for pigs in group VIII) of inoculated Gn pigs by using TRIZOL LS or TRIZOL reagent (only for WBC) according to the instructions provided by the supplier (Life Technologies, Grand Island, N.Y.). Rectal swab fluids and 20% suspensions of intestinal contents in 0.01 M PBS, pH 7.2, were centrifuged at 1,200 × g for 30 min, and the supernatants were used for RNA extraction. For groups VII and VIII, WBC were prepared by adding 40 volumes of sterile distilled water to blood samples from pigs to lyse red blood cells, followed by centrifugation at 800 × g for 10 min. The cell pellets were washed twice with distilled water and resuspended in 0.01 M PBS, pH 7.2, to one-fifth of the volume of the original blood sample. For PEC RNA extraction, rectal swab fluids, 20% suspensions of intestinal contents, serum samples, or WBC suspensions were mixed with 3 volumes of TRIZOL LS or TRIZOL reagent by vortexing and incubated at 15 to 30°C for 5 min. The mixture was mixed with four-fifths volume of chloroform by vigorous vortexing for 1 min followed by centrifugation at 14,000 × g for 15 min at 4°C. The viral RNA in the upper aqueous phase was precipitated with 1 µl of glycogen (20 µg/ml) and an equal volume of isopropanol. The RNA pellet was resuspended in 50 µl of diethyl pyrocarbonate-treated water and stored at 20°C until use.

(iii) Antigen-ELISA. An antigen-ELISA was performed as described elsewhere (19; M. Guo, G. J. Bowman, Q. Wang, and L. J. Saif, unpublished data). Hyperimmune guinea pig antiserum to PEC/Cowden was prepared by immunizing guinea pigs with PEC virus-like particles (VLPs) as described previously (17) and was used to coat Nunc-Immuno plates (MaxiSorp; Nalge Nunc International, Roskilde, Demark) at a dilution of 1:2,000 in 0.05 M carbonate buffer, pH 9.6. The plates were incubated at 4°C overnight, followed by blocking with 4% nonfat dry milk in 0.01 M PBS, pH 7.2. After washing the plates three times, PEC-positive, PEC-negative control fecal samples or IC, and PEC-negative serum samples from Gn pigs and test samples (rectal swab suspensions or 1:25 diluted intestinal contents supernatants or serum samples from i.v.-inoculated pigs) were added to the wells, followed by incubation at 37°C for 120 min. After washing three times, 1:2,000-diluted hyperimmune pig antisera to PEC/Cowden were added to the wells, and the plates were then incubated at 37°C for 90 min. Antigen binding was detected by adding 1:2,000-diluted horseradish peroxidase-labeled goat anti-pig immunoglobulin G-Fc conjugate (Bethyl Laboratories, Inc., Montgomery, Tex.) to the wells (100 µl/well) followed by incubation at 37°C for 90 min. After washing the plates three times, the substrate, 2,2'-azino-bis-3-ethylbenz-thiazoline sulfonic acid (ABTS) (Sigma, St. Louis, Mo.) was added to the wells for color development (at 37°C for 30 min). The samples were tested at a single sample dilution (1:25), and positive samples were defined as ones with an absorbance greater than or equal to the mean absorbance (A) of the antigen-negative control wells + 3 standard deviations (SD). The comparative absorbances of positive samples were designated as follows: +++, A₄₉₂ >1; ++, A₄₉₂ = 0.5 to 1.0; +, A₄₉₂ < 0.5; , A₄₉₂ < A + 3 SD. For titration by the antigen-ELISA, both the TC-PEC and WT-PEC inocula were twofold serially diluted beginning at 1:40, and their PEC antigen titers were expressed as the reciprocal of the highest dilution with an absorbance of A + 3 SD.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Diarrhea and virus shedding in Gn pigs. (i) Oral inoculation with TC PEC or WT PEC. The TC-PEC inoculum had a virus titer of 10⁶ fluorescent focus-forming units/ml by CCIF (11). Both the TC-PEC and WT-PEC inocula had a comparable PEC antigen titer of 1:2,560 by the antigen-ELISA. Marked differences were observed between the WT-PEC and TC-PEC orally inoculated Gn pigs in terms of diarrhea and fecal virus shedding (Table 2). Mild to moderate diarrhea developed in all WT-PEC orally inoculated pigs by PID 2 to 4 and persisted for 2 to 5 days (Table 2). Diarrhea was not observed in either the TC-PEC orally inoculated or the mock-inoculated pig. The WT-PEC orally inoculated pigs also had higher cumulative fecal scores compared with the TC-PEC orally inoculated and the mock-orally inoculated pigs (data not shown). The PEC RNA was detected by RT-PCR in rectal swab fluids from PID 1 to PID 7 or 9 (or until euthanasia) and in the intestinal contents of both WT-PEC and TC-PEC orally inoculated pigs at euthanasia, but not in the intestinal contents of the mock-inoculated pig. The PEC particles were detected by IEM in the small and large intestinal contents from WT-PEC orally inoculated pigs that were euthanatized between PID 3 and PID 9, but not from the TC-PEC orally inoculated pigs (Table 2). Virus particles were also detected by IEM from fecal samples (when available) collected from the WT-PEC-inoculated diarrheic pigs.

(ii) i.v. inoculation with WT PEC. Mild to moderate diarrhea was observed at PID 4 to 5 in all WT-PEC i.v.-inoculated Gn pigs (Table 3). The PEC RNA was detected in both rectal swab fluids and intestinal contents from WT-PEC i.v.-inoculated pigs from PID 1 to PID 8 (or until euthanasia), and PEC particles were detected in intestinal contents of all 3 pigs. The PEC antigens were detected by ELISA in rectal swab fluids from PID 3 to 8. High ELISA absorbance values for virus antigens were detected in rectal swab fluids or intestinal contents from PID 4 to 8. No diarrhea developed in the mock-i.v.-inoculated Gn pigs or the pig i.v.-inoculated with formalin-inactivated WT PEC. No PEC particles, antigens, or RNA was detected by IEM, ELISA, and RT-PCR, respectively, in the fecal swab fluids and intestinal contents from the mock- and formalin-inactivated WT-PEC i.v.-inoculated Gn pigs.

(iii) Oral or i.v. inoculation with WT-PEC-positive acute-phase sera. Moderate diarrhea developed in Gn pigs inoculated both orally and i.v. with PEC-positive acute-phase sera from WT-PEC orally or i.v.-inoculated pigs at PID 2 to 5 (Table 4). The diarrhea persisted for 5 to 8 days. The PEC RNA and antigens were detected in rectal swab fluids or intestinal contents of i.v.-inoculated Gn pigs from PID 2 to 8; PEC particles were detected by IEM in feces or intestinal contents of 2 i.v.-inoculated pigs euthanatized at PID 3 or 7 and in feces collected from PID 2 to PID 4 of 2 orally inoculated pigs. Of the 2 orally inoculated pigs, the PEC antigens were detected by using an antigen-ELISA in rectal swab fluids from PID 2 and up to PID 17, whereas PEC RNA was detected from PID 1 to 27 (data not shown). High ELISA absorbance values for PEC antigens were detected in rectal swab fluids, feces or intestinal contents from PID 2 to 8 (Table 4).

Examination of serum samples from WT-PEC i.v.- or orally inoculated pigs for PEC RNA or antigens by using RT-PCR and ELISA, respectively. Serum samples collected from the WT-PEC i.v.-inoculated pigs (group IV) at PID 2, 4, 6, and 8 (or until euthanasia) were examined for PEC RNA by RT-PCR. The PEC RNA was detected in serum samples from all three of the WT-PEC i.v.-inoculated pigs from PID 2 to PID 8, but not from two mock-i.v.-inoculated pigs or one formalin-inactivated WT-PEC i.v.-inoculated pig (Table 3). The PEC antigens were detected by ELISA at low absorbance values in serum samples (1:25 dilution) from all three WT-PEC i.v.-inoculated pigs at PID 2 to 6, but not from the mock- or formalin-inactivated WT-PEC-inoculated pigs (Table 3). The PEC RNA or antigens were also detected in acute-phase serum samples from seven of nine WT-PEC orally inoculated pigs from PID 2 to 10 (data not shown). Furthermore, the PEC RNA or antigens were detected in sera from all four pigs in groups VII and VIII inoculated with acute-phase serum from pigs i.v. or orally inoculated with WT PEC (Table 4). The duration for PEC RNA or antigen detection was 5 to 6 days. The PEC RNA was also detected in WBC from both pigs in group VII at PID 6 to 14 and from one of two pigs in group VIII at PID 8 to 14 (partial data shown in Table 4).

Examination of PEC antigen distribution in tissues. (i) Oral inoculation of Gn pigs with TC PEC or WT PEC. By using IF staining, the PEC antigens were detected in impression smears from the small intestinal tissues from all the WT-PEC-inoculated pigs except for two recovered pigs euthanatized at PID 9 (Table 5). There were consistently higher numbers of PEC antigen-positive villous epithelial cells in the proximal small intestine (duodenum and jejunum, 0.05 to 15%) than in the distal small intestine (ileum, 0.01 to 0.5%). The PEC antigens were also detected by IF staining in small intestinal impression smears from only one of four TC-PEC-inoculated pigs, but the numbers of PEC-positive cells were much lower (0.01%) and limited to the jejunum only. Mucosal impression smears from the small intestine of a mock-inoculated pig were negative for PEC antigen by IF staining, and so were impression smears from the colon and extraintestinal tissues (lungs, liver, spleen and kidneys) of all orally inoculated pigs.

(ii) i.v. inoculation of Gn pigs with WT PEC. The villous epithelial cells in the small intestine were positive for PEC antigens by IF staining of the small intestinal impression smears from the Gn pigs inoculated i.v. with WT PEC (Table 5). Viral antigens were not detected by IF in the small intestinal impression smears from two mock-i.v.-inoculated pigs or one pig inoculated i.v. with formalin-inactivated WT PEC. Impression smears prepared from the colon, liver, lungs, spleen and kidneys of all pigs were negative for PEC by IF staining.

(iii) Oral or i.v. inoculation of Gn pigs with PEC-positive acute-phase sera. The PEC antigens were detected in the small intestinal impression smears from the inoculated pigs euthanatized at PID 3 and 7 (Table 5), but not from the ones euthanatized at PID 21 or 28 (data not shown). However, serum antibodies to PEC/Cowden were detected by using VLP-ELISA in convalescent-phase serum samples collected at PID 21 (pig 14-4, 1:800; pig 15-9, 1:1600) and 28 (pig 14-4, 1:1,600). Impression smears prepared from the colon, liver, lungs, spleen, and kidneys of all pigs were negative for PEC by IF staining.

Histologic findings. (i) Oral inoculation of Gn pigs with TC PEC or WT PEC. Mild to severe villous atrophy, mild to moderate and multifocal villous fusion, and crypt hyperplasia were observed in the small intestine (mainly in the duodenum and jejunum) of all WT-PEC orally inoculated pigs that were euthanatized from PID 3 to PID 9 (Table 5; Fig. 1C). Villous atrophy in the duodenum was moderate to marked in five of six pigs examined on PID 3 and 4, and was mild in one pig each at PID 4 and PID 7, and two pigs at PID 9. In the jejunum, villous atrophy was absent at PID 3, was marked to severe in four of six pigs at PID 4 and 7, and was resolving by PID 9 (mild to moderate). Only mild villous atrophy was observed in the ileum of two of five pigs at PID 4 and in one pig at PID 7, and no villous atrophy in the ileum was observed in the other pigs. Enterocytes on the tips of the villi were occasionally cuboidal or flattened, lacked cytoplasmic droplets, and acquired a foamy, vacuolated cytoplasm. Epithelial cell vacuolization was seen primarily in the duodenum and jejunum, and seldom in the ileum. Crypt hyperplasia was demonstrated by elongation of crypt length, increased numbers of mitoses, and a disordered arrangement of crypt epithelial cells. Villous fusion was observed mainly in the duodenum and occasionally in the jejunum, but not in the ileum. Expansion of the villous lamina propria and dilatation of lacteal vessels in the submucosa were indicative of edema in some sections of the duodenum and jejunum.

View larger version (136K): [in this window] [in a new window]   FIG. 1.   Histologic lesions in the duodenum or jejunum of Gn pigs following oral or intravenous inoculation with PEC/Cowden. Hematoxylin and eosin stain. (A) Normal-appearing, long villi of the jejunum from a mock-inoculated Gn pig (pig 13-5). (B) Villi of the duodenum showing mild villous atrophy change from the TC-PEC orally inoculated pig (pig 4-6). (C) The severe blunting, shortening, and fusion of jejunal villi and crypt hyperplasia from the WT-PEC orally inoculated Gn pig (pig 8-9). (D) The severe, widespread villous atrophy and crypt hyperplasia and elongation in the jejunum from the WT-PEC i.v.-inoculated Gn pig (pig 8-1). Bar, 100 µm.

(ii) i.v. inoculation of Gn pigs with WT PEC. Marked, widespread villous atrophy, crypt hyperplasia, and elongation were observed in the proximal small intestine (duodenum and jejunum) of WT-PEC i.v.-inoculated pigs (Table 5; Fig. 1D); this was similar to lesions observed in the WT-PEC orally inoculated pigs. Superficial epithelial cells were mildly attenuated (tall cuboidal), often retaining cytoplasmic vacuoles. There was mild and multifocal exfoliation and loss of enterocytes from the villous tips at PID 6. The crypt length increased significantly and the villus/crypt ratios were decreased dramatically. In contrast, no villous atrophy was observed in the ileum of the three pigs in this group. No lesions were observed in the small intestine of the mock- and formalin-inactivated WT-PEC i.v.-inoculated pigs, nor in the colon, liver, lungs, spleen and kidneys of all i.v.-inoculated Gn pigs.

iii) Oral or i.v. inoculation of Gn pigs with PEC-positive acute-phase sera. Moderate villous atrophy (duodenum) and severe villous atrophy and fusion (jejunum) were observed in the proximal small intestine of a Gn pig (pig 14-5) i.v. inoculated with PEC-positive acute-phase sera from WT-PEC i.v.-inoculated pigs and euthanatized at PID 7 (3 days after diarrhea onset) (Table 4). The pig (pig 15-10) inoculated i.v. but with PEC-positive acute-phase sera from WT-PEC orally inoculated pigs showed mild villous atrophy only in the duodenum when euthanatized at PID 3 (1 day after diarrhea onset). No villous atrophy was observed in the ileum of either pig. No villous atrophy was observed in the small intestines of the two Gn pigs orally inoculated with PEC-positive acute-phase sera and euthanatized at either PID 21 or 28. Both of these pigs seroconverted to PEC at PID 21 to 28.

Scanning electron microscopy. Small intestinal villi were long and finger-like, with circular transverse grooves on their surface, and the microvillous coat was uniform and densely packed in the mock-inoculated pigs (Fig. 2A, D, and G). In the WT-PEC-inoculated (orally or i.v.) pigs, denuding and shortening of the duodenal and jejunal villi was apparent, and the microvillous coat appeared irregular and patchy (Fig. 2C, F, and I). Some severely stunted villi appeared fused, cornical, or leaf-shaped. Enterocytes on the villous tips appeared swollen and degenerate and were exfoliated (Fig. 2F and I). Villous atrophy and fusion in the duodenum and jejunum were pronounced in contrast to minor changes in the ileum. In comparison, only mild or no villous shortening and irregularity along with occasional patchy microvillous coat were observed in the duodenum and jejunum of the TC-PEC-inoculated pigs (Fig. 2B, E, and H). No changes were observed in the ileum.

View larger version (135K): [in this window] [in a new window]   FIG. 2.   Scanning electron micrographs of duodenum segments from Gn pigs following oral inoculation with TC PEC/Cowden or WT PEC/Cowden. (A) Typical, long, finger-like villi showing circular transverse grooves on their surface from a mock-inoculated pig (pig 13-5). Magnification, ×130. (B) Finger-like villi with mildly irregular outer surface appearance from the TC-PEC-inoculated pig (pig 4-6). Magnification, ×120. (C) Shortened and fused villi with irregular appearance from the WT-PEC-inoculated pig (pig 4-4). Magnification, ×140. (D) Typical villous apex showing smooth surface from a mock-inoculated pig (pig 13-5). Magnification, ×490. (E) Villous apex with mildly irregular surface appearance from the TC-PEC-inoculated pig (pig 4-6). Magnification, ×510. (F) Shortened and fused villous apex, showing swollen enterocytes, exfoliation, and loss of enterocytes from the WT-PEC-inoculated pig (pig 4-4). Magnification, ×490. (G) Dense, uniform microvillous coat of the enterocytes from a mock-inoculated pig (pig 13-5). Magnification, ×3,870. (H) Smooth microvillous coat of enterocytes, showing mild irregularity, from the TC-PEC-inoculated pig (pig 4-6). Magnification, ×3,870. (I) Irregular microvillous coat of enterocytes from WT-PEC-inoculated pig (pig 4-4). Microvilli are fused, shortened, and more sparse. Magnification, ×3,870.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The HuCVs are the leading cause of food- and waterborne viral gastroenteritis worldwide. In a recent report it was estimated that the HuCVs cause 23 million cases of food-borne illnesses annually, accounting for 67% of the cases caused by food-borne pathogens in the United States and 33% of annual hospitalizations due to food-borne illnesses (23), although most cases used for estimation lacked identifiable pathogens. Animal enteric caliciviruses are emerging pathogens that cause diarrhea in the respective animal host (4, 32; K. W. Theil and C. M. McCloskey, Abstr. 76th Annu. Meet. Conference of Research Workers in Animal Diseases, abstr. 110, 1995). These mostly uncultivable enteric caliciviruses are genetically related (7, 15, 22) and induce similar histologic lesions in the proximal small intestines of their respective hosts (12, 19, 20).

To date the PEC/Cowden is the only cultivable enteric calicivirus (11, 29), but for growth it requires an IC preparation from uninfected Gn pigs as a medium supplement. Previously the TC PEC was serially passaged 19 times in primary porcine kidney cells (11). In this study, the TC PEC was passaged another 19 times in a continuous porcine kidney cell line (LLC-PK) and then used to orally inoculate Gn pigs to examine its virulence. No diarrhea developed in the TC-PEC orally inoculated Gn pigs, and only mild or no villous atrophy was observed in the small intestine. In contrast, all WT-PEC orally inoculated Gn pigs developed diarrhea, and all of them euthanatized at PID 3 to 4 demonstrated moderate to severe villous atrophy and fusion in the proximal small intestine (duodenum, nine of nine; jejunum, six of nine from PID 3 to 9), which was similar to results previously reported (12). The histologic lesions in the WT-PEC-inoculated pigs correlated with the PEC antigen distribution, and coincided with the detection of fecal virus shedding and clinical signs. The ELISA absorbance values of fecal PEC antigen shed were consistently higher in the WT-PEC-inoculated pigs than in the TC-PEC-inoculated pigs, in spite of the consistent detection of fecal PEC RNA in pigs of both groups. These data indicate that the TC PEC apparently infected Gn pigs and induced limited proximal small intestinal lesions and fecal virus shedding (with much lower virus titers), but did not cause clinical illness, suggesting the less efficient replication and growth of the TC PEC in Gn pigs following infection. Thus, the virulence of the TC PEC is at least partially attenuated after serial passage in cell culture. No dose-response was done in this experiment, but the pigs were given the highest available dose of the TC PEC, which had a comparable antigen titer (1:2,560) to that of the WT PEC inoculum. Previous studies indicated that the TC PEC has two amino acid changes in the polymerase region and one distant and three clustered amino acid substitutions in the capsid region (15). This hypervariable capsid region with three clustered amino acid changes forms the externally located P2 subdomain on the virus surface and corresponds to the binding region of the Norwalk virus (NV) capsid (rNV virus-like particles) to human and animal cells in vitro (30, 40). It is likely that this P2 subdomain determines host specificity (30, 40). The limited propagation of TC PEC in the small intestine of inoculated Gn pigs and its reduced virulence may be the result of a potential change in tissue tropism or binding that may be related to the amino acid substitutions in the hypervariable capsid region. Thus, the amino acid changes in the capsid hypervariable region may be associated with both the cell culture adaptation and the attenuation of the virulence of the TC PEC in Gn pigs. The TC PEC may be a useful candidate vaccine for future evaluation for prevention of PEC infections of swine.

In early volunteer studies, the HuCVs (NV, Snow Mountain virus, and HV) caused diarrhea or vomiting and induced histologic lesions evident in jejunal biopsies, which included broadening and blunting of the intestinal villi, crypt cell hyperplasia, cytoplasmic vacuolization, and infiltration of polymorphonuclear and mononuclear cells into the lamina propria (9, 20). In a previous study, it was reported that WT PEC/Cowden induced diarrhea and villous atrophy in orally inoculated neonatal Gn pigs (12). The present study confirmed these findings for Gn pigs inoculated orally with WT PEC. The proximal small intestinal villous atrophy and fusion, crypt cell hyperplasia and reduction of villus/crypt ratios, cytoplasmic vacuolization, and infiltration of polymorphonuclear and mononuclear cells into the lamina propria coincided with the appearance of clinical illness in the infected human volunteers and in the Gn pigs (1, 2, 8, 33, 34).

The diarrhea, fecal virus shedding, and intestinal lesions in Gn pigs in our study resembled those observed in human volunteer studies for the HuCVs (NV, HV, and Montgomery County agent) (1, 9,13, 20, 34) and in Gn calves inoculated orally with bovine enteric calicivirus Newbury agents (5, 19). The HuCVs and enteric caliciviruses in animals reportedly infect the proximal small intestine (1, 12, 19, 20, 33). However, in infected volunteers only jejunal biopsies were examined and the extent of involvement of the small intestine in disease progression, lesions and virus replication remains unclear. Our present and previous (12) data indicated that WT PEC infected villous enterocytes and induced histologic lesions mainly in the duodenum and jejunum, confirming that they are the major sites for PEC replication in the small intestine. Colon and extraintestinal tissues or organs may not support PEC replication, because no PEC-positive cells were detected by IF staining in impression smears of these tissues, nor were lesions evident, even from the pigs shown to have viremia following PEC infection. It is possible that the restricted growth of PEC to the small intestine relates to its requirement for specific receptors and factors present in intestinal contents as demonstrated for its in vitro cultivation (11, 29), and such factors or receptors may not be present at extraintestinal sites.

In NV-infected volunteers, the peak of virus shedding in stool as detected by a sensitive antigen-ELISA was between 25 and 72 h after oral inoculation, and its duration was at least 7 days (13). In WT-PEC orally inoculated pigs, fecal virus shedding was detected by both the RT-PCR and an antigen-ELISA from PID 1 to PID 9, with a peak from PID 2 to 7. Interestingly, in two Gn pigs that were orally inoculated with PEC-positive acute-phase sera and euthanatized at PID 21 or 28, the PEC RNA and antigens were detected in rectal swab fluids up to 27 days by RT-PCR and 17 days by ELISA, respectively. A recent study indicated that stool virus shedding was detected by RT-PCR and southern hybridization up to 28 days in patients naturally infected with a GII NLV during a long-term- care hospital outbreak (P. R. Hazelton, K. M. Combs, T. B. Ball, L. Klass, and P. Plourde, Abstr. 19th Annu. Meet. Am. Soc. Virol., abstr. W14-4, 2000). The long duration of fecal virus shedding may facilitate virus transmission, causing secondary infections or outbreaks. Thus, proper planning for intervention strategies and handling of outbreaks associated with HuCVs may help to control virus transmission and disease spread.

Enteropathogenic viruses like HuCVs and animal enteric caliciviruses are usually transmitted through the fecal-oral route. Some nonenteric caliciviruses, such as the vesiviruses, feline calicivirus, vesicular exanthema of swine virus, and San Miguel sea lion virus, are transmitted through direct contact, infected fomites, or respiratory routes (feline calicivirus) (6, 35). However, inoculation of pigs with vesicular exanthema of swine virus via intradermal, subcutaneous, intramuscular or i.v. routes, also produced vesicular disease in swine (35). The rabbit hemorrhagic disease virus (RHDV) causes a fatal systemic hemorrhage in rabbits that can be infected via many inoculation routes (27). Viremia appears within 24 h after inoculation and likely plays an important role in virus spread to the target organs or tissues including the liver. In this study, Gn pigs were inoculated i.v. with WT PEC. Interestingly, all WT-PEC i.v.-inoculated pigs developed diarrhea, characteristic histologic lesions and PEC antigens detectable in the proximal small intestine (mainly duodenum and jejunum), which were similar to those observed in WT-PEC orally inoculated pigs, although more severe villous atrophy was consistently seen in the jejunum. Other major differences were that the incubation period for clinical diarrhea was 1 to 2 days longer than that in the orally inoculated pigs and correspondingly the PEC antigens were first detected by ELISA in feces at PID 3 instead of PID 1 in most of the orally inoculated pigs, although fecal virus RNA was detected by RT-PCR in both groups at PID 1. The delay in clinical illness coincided with the initial presence of PEC antigen shedding detected in feces. The peak of fecal virus shedding detected by antigen-ELISA was from PID 4 to PID 7, and high numbers of PEC-infected enterocytes were evident by IF in the small intestine, suggesting that WT PEC replicated efficiently in the infected enterocytes. Subsequently, marked villous atrophy and fusion were observed in the proximal small intestine as a result of the damage, exfoliation and loss of enterocytes. This is further supported by the presence of high virus numbers (by IEM) in the intestinal contents from the WT-PEC i.v.-inoculated Gn pigs. In contrast, no diarrhea and small intestinal histologic lesions were evident in mock- or formalin-inactivated WT-PEC i.v.-inoculated pigs. Thus, WT PEC administered i.v. reached the small intestine presumably via the bloodstream, where it localized, propagated efficiently and induced small intestinal lesions characteristic of those observed in the WT-PEC orally inoculated pigs. To our knowledge, this is the first report of an enteric calicivirus causing symptomatic illness, fecal virus shedding and histologic lesions in the susceptible host following i.v. inoculation. How the WT PEC reaches the small intestine from the bloodstream and infects the villous enterocytes is unknown. In type 1 reovirus infection of mice, reoviruses in the bloodstream during viremia or after i.v. inoculation are transported to the ileum, where they infect crypt cells possibly via attachment to the basolateral membrane but not via the luminal surface (28). Other enteropathogenic animal viruses, such as adenovirus, parvovirus, and bovine virus diarrhea virus may infect crypt enterocytes via hematogenous dissemination after viremia in a similar manner (31). The PEC may be unique in this regard since i.v. inoculation leads to infection of mainly villous and not crypt enterocytes as detected by IF staining of small intestinal impression smears.

Previously it was unknown if enteric caliciviruses induced viremia. In this study, the PEC RNA and low titers of virus antigen were detected in serum samples from the WT-PEC i.v.-inoculated pigs and from seven of nine pigs inoculated with WT PEC orally. It is unlikely that the PEC RNA or antigen in serum was from the initial inoculum because the PEC RNA or antigens were also detected in sera from WT-PEC orally inoculated pigs and not from a control pig inoculated with killed WT PEC. In addition, the longer incubation periods for the onset of diarrhea and fecal viral antigen shedding coincided with a delay in transport of PEC from the bloodstream to the small intestine. Many viruses induce viremia during which the viruses circulate in the blood serum or WBCs and are spread to the target organs to initiate infection (25). To determine if the PEC-positive acute-phase sera contained infectious virus, the serum samples from the WT-PEC i.v.- and orally inoculated pigs were used to inoculate additional Gn pigs via the i.v. or oral routes. Diarrhea and fecal virus shedding developed in all inoculated pigs, with a pattern similar to that of intestinally derived WT-PEC i.v.- or orally inoculated pigs. In addition, seroconversion to PEC was detected at PID 21 in 2 orally inoculated pigs. Characteristic small intestinal lesions (described earlier) and PEC-positive enterocytes were observed in both of the i.v.-inoculated pigs examined, and the PEC RNA or antigens were detected in serum samples from all four inoculated pigs (Table 4). Additionally PEC RNA was also detected in the washed WBC from three of these pigs, but only later after PID 6 to 8. Because phagocytic cells in the blood phagocytize viruses in the process of viral destruction, we did not conduct pig infectivity experiments with the WBC to determine if they contained infectious PEC. However, it is interesting that after PEC and HuCV infections, polymorphonuclear and mononuclear cells infiltrate the small intestine; whether subsets of these are derived from WBC that might contain infectious PEC should be examined in the future studies. Collectively, these results suggest that viremia occurs following PEC infection and that the PEC-positive sera contain infectious virus. Considering the low virus titers by ELISA in PEC-positive acute-phase sera which induced illness in inoculated Gn pigs, the WT PEC/Cowden must be highly infectious for pigs. This is an important common property for PEC and the NV and related HuCVs associated with food- and waterborne viral gastroenteritis (9, 20).

In conclusion, the TC PEC induced a limited small intestinal infection and low levels of fecal virus shedding, but no diarrhea, in the Gn pigs. Infection with TC PEC caused only mild or no lesions in the small intestine. Thus, the TC PEC is at least partially attenuated after serial passage in cell culture in vitro, and it may be a potential candidate vaccine for further evaluation. The WT PEC induced diarrhea and characteristic histologic lesions (villous atrophy and fusion) in the proximal small intestine of Gn pigs following oral or intravenous inoculation. The WT PEC detected in serum of pigs i.v. or orally inoculated with WT PEC was infectious when inoculated into additional pigs. To our knowledge, this is the first report of an attenuated cell culture-adapted enteric calicivirus and of a WT enteric calicivirus that causes symptomatic illness and intestinal lesions via intravenous inoculation as well as the occurrence of viremia following PEC infection.