Previous Article | Next Article ![]()
Journal of Virology, March 2005, p. 3429-3437, Vol. 79, No. 6
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.6.3429-3437.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Center for Molecular Medicine and Infectious Diseases, Department of Biomedical Sciences and Pathobiology, College of Veterinary Medicine, Virginia Polytechnic Institute and State University, Blacksburg, Virginia
Received 16 July 2004/ Accepted 18 October 2004
|
|
|---|
|
|
|---|
Nonhuman primates have been used as animal models for HEV (3, 5, 10, 45, 48). However, due to the limited resources, ethical concerns, and restricted experimental procedures available for the use of nonhuman primates, little has been learned about the pathogenesis of HEV by the use of primate models. In addition, extrapolating from or interpreting the significance of human HEV pathogenesis in nonhuman primates may be difficult, as nonhuman primates are not the natural hosts of human HEV. The first animal strain of HEV, swine HEV, was discovered in 1997 in a pig in the United States (30). Since then, swine HEV has been identified in pigs in many other countries and has been shown to be genetically closely related to human HEV, especially genotype 3 and 4 strains of human HEV (11, 14, 17, 18, 21, 33, 43, 44, 46, 47). Interspecies transmissions of swine HEV to nonhuman primates (29) and of a U.S. strain of human HEV to pigs (14) have been documented. Increasing evidence indicates that hepatitis E is a zoonotic disease (25, 27, 31). Swine HEV infections in pigs have been evaluated as an experimental model of HEV (14, 21, 46). However, the potential to use swine as a model system is limited by the fact that swine HEV causes only subclinical infections (14, 28, 29). Therefore, only certain aspects of HEV replication and pathogenesis can be studied with the pig model.
More recently, another animal strain of HEV, avian HEV, was identified and characterized from chickens with hepatitis-splenomegaly (HS) syndrome in the United States (16). Like swine HEV, avian HEV is also genetically and antigenically related to human HEV. Unlike swine HEV, however, avian HEV is associated with a hepatic disease (HS syndrome) (16, 19). The complete genomic sequence of avian HEV was determined (20). The genomic organization of avian HEV is very similar to that of mammalian HEVs (20). Although avian HEV has only about 50% nucleotide sequence identity with mammalian HEVs, they share many significant structural and functional features (20), supporting the conclusion that avian HEV and mammalian HEVs belong to the same genus, Hepevirus (9). The discovery of avian HEV and its association with a hepatic disease provided a homologous animal model system for the study of HEV pathogenesis and replication. For this study, we attempted to experimentally infect specific-pathogen-free (SPF) adult chickens by the natural fecal-oral route, to systematically study HEV pathogenesis and replication in a homologous animal model via a natural route of infection, and to characterize the clinical course and pathological lesions associated with avian HEV infection.
|
|
|---|
Chickens. Eighty-five 60-week-old SPF chickens were purchased from Charles River SPAFAS Inc. (Wilmington, Mass.). The chickens were in the late stage of egg production. Prior to inoculation, all birds were confirmed to be negative for avian HEV antibodies by an enzyme-linked immunosorbent assay (ELISA) (19).
Experimental design. The chickens were randomly divided into three groups, of 28, 29, and 28 chickens. The twenty-eight chickens in group 1 were each inoculated by the oronasal route with 5 x 104.5 CID50 of the avian HEV infectious stock. One-fourth of the 1-ml inoculum was given nasally and the remaining inoculum was given orally by the use of gavage needles. Chickens in group 2 (n = 29) were each inoculated intravenously (i.v.) with 5 x 104.5 CID50 of the avian HEV infectious stock. The twenty-eight chickens in group 3 served as uninoculated controls. Each group was housed in a separate isolation room, and the chickens were allowed access to feed and drinking water ad libitum.
Sample collection and processing. Blood and fecal swab materials were collected prior to inoculation and weekly thereafter. Weekly blood plasmas were tested for liver enzyme profiles. Weekly serum samples were tested by ELISA for anti-avian HEV antibodies. Weekly serum samples and fecal swab materials were tested for avian HEV RNA by reverse transcription-PCR (RT-PCR). Two chickens from each group were necropsied at 1, 3, 5, 7, 10, 13, 16, 20, 24, 28, 35, and 42 days postinoculation (dpi), and the remaining chickens were necropsied at 56 dpi. Samples of serum, feces, bile, and 13 different tissues were collected during each necropsy and stored at 80°C. A portion of the liver tissue samples collected at each necropsy was homogenized in 10% (wt/vol) sterile phosphate-buffered saline. The liver homogenates were clarified by centrifugation at 3,000 rpm for 15 min at 4°C (Eppendorf centrifuge 5810, rotor A-4-44) and then used for the detection of avian HEV RNA by RT-PCR.
Pathology and histopathology evaluations. Gross pathological lesions from livers and spleens were evaluated during necropsies and were also recorded as digital pictures. Tissue samples collected at each necropsy, including thymus, heart, liver, lung, spleen, kidney, colon, cecal tonsil, cecum, ileum, jejunum, pancreas, and duodenum samples, were fixed in 10% neutral buffered formalin and processed for routine histological examinations. Histopathological lesions in various tissues were evaluated in a blinded fashion by a veterinary pathologist and were scored according to lesion severity based on standard scoring systems. Liver lesion scores ranged from 0 to 4 (0, no lesions; 1, <5 foci; 2, 5 to 8 foci; 3, 9 to 15 foci; 4, >15 foci). Thymus lesions were given scores from 0 to 4 (0, none; 1, minimal; 2, mild; 3, moderate; 4, severe) based on the severity of the lesions. Lung lesion scores were expressed as numbers of foci. Kidney lesion scores ranged from 0 to 4 (0, no lesions or nonspecific foci; 1, minimal interstitial nephritis; 2, mild interstitial nephritis; 3, moderate interstitial nephritis; 4, severe interstitial nephritis).
Serum biochemical profiles. A total of 13 chickens (4 from the oronasal group, 5 from the i.v. group, and 4 from the control group) were monitored weekly throughout the entire study of 56 days. The levels of liver enzymes in sera from the 13 chickens, including aspartate transferase (AST), lactate dehydrogenase (LDH), creatine phosphokinase (CPK), the albumin/globulin (A/G) ratio, bile acids, and total proteins, were determined by standard methods (Avian and Exotic Animal Clinical Pathology Labs, Wilmington, Ohio).
ELISA for avian HEV antibodies. A purified truncated recombinant ORF2 capsid protein of avian HEV expressed in Escherichia coli was used as the antigen for an ELISA to detect avian HEV antibodies in chickens as previously described (15, 19, 40). Briefly, 96-well plates (Thermo Labsystems, Franklin, Mass.) were coated with the purified avian HEV antigen. Horseradish peroxidase-conjugated rabbit anti-chicken immunoglobulin G (IgG; Sigma Chemical Co., St. Louis, Mo.) was used as the secondary antibody. Optical density (OD) values were measured at 405 nm. Samples with OD values of >0.30 were considered positive, as determined previously (19, 40). Convalescent-phase sera from experimentally infected chickens (41) and sera from SPF chickens were included as positive and negative controls, respectively.
RT-PCR to detect avian HEV RNA. To detect avian HEV RNAs in fecal, serum, bile, and liver tissue homogenates, we performed RT-PCR as previously described (19). Briefly, RNAs were extracted by the use of Trizol reagent (GIBCO-BRL) from 100 µl of serum, a 10% fecal suspension, a 10% liver homogenate, or a bile sample. The total RNA was resuspended in 12.25 µl of DNase-, RNase-, and proteinase-free water (Invitrogen). Reverse transcription was performed at 42°C for 60 min with 1 µl of N2 reverse primer (5'-CCGGGCTGATGGTCTCGATTAG-3'), 0.25 µl of Superscript II reverse transcriptase (Invitrogen), 1 µl of 0.1 M dithiothreitol, 4 µl of 5x RT buffer, 0.5 µl of RNase inhibitor, and 1 µl of 10 mM deoxynucleoside triphosphates. Five microliters of the resulting cDNA was amplified in a 50-µl reaction with AmpliTaq Gold DNA polymerase (Applied Biosystems).
For confirmation purposes, two nested RT-PCR assays targeted to different regions were used to test the samples. For the first nested RT-PCR assay, the first-round PCR, performed with a primer set located in the ORF1 region (forward primer N1 [5'-TTACCATTGACTTTGAACGGCG-3'] and reverse primer N2), produced an expected fragment of 643 bp. For the second-round PCR, the forward primer N3 (5'-GCTTGTGCATTGACGATTTCCC-3') and the reverse primer N4 (5'-CAATAGGTTACCCACGATGACG-3') produced an expected fragment of 500 bp. For the second nested RT-PCR assay, the first-round PCR produced an expected fragment of 595 bp with the forward primer P1 (5'-ACAACATCCACCCCTACAAG-3') and the reverse primer P2 (5'-ACAGTTTCACCTCAGGCTCG-3'). For the second-round PCR, the forward primer P3 (5'-AGAACAATGGTTGGCGGTCC-3') and the reverse primer P4 (5'-GAGGGCAAGCCACCTAAAAC-3') amplified an expected fragment of 394 bp. The PCR parameters included an initial incubation at 94°C for 9 min to activate the AmpliTaq Gold DNA polymerase, followed by 39 cycles of denaturation at 94°C for 0.5 min, annealing at 58°C for 0.5 min, and extension at 72°C for 1 min and a final extension at 72°C for 7 min.
The PCR products amplified from serum, fecal, bile, and liver samples from two selected chickens from the oronasally inoculated group and one selected chicken from the i.v. inoculated group were sequenced to confirm the identities of the viruses recovered from the experimentally infected chickens.
Statistical analyses. Gross and histopathologic lesions were recorded either as the presence or absence of lesions, as lesion scores, or as counts of lesion foci. Categorical (dichotomous) variables were analyzed by logistic regression by use of either the LOGISTIC or the GENMOD procedure in SAS version 8.02 (SAS Institute, Inc., Cary, N.C.). Lesion scores were ranked, and medians were compared by analysis of variance according to the GLM procedure in SAS. Counts of lesion foci were modeled as either Poisson or negative binomially distributed variables by use of the GENMOD procedure in SAS. Models included treatment (TRT) and dpi models, and in addition, for lesion scores, included a TRT x dpi interaction model.
|
|
|---|
Seroconversion to avian HEV antibodies in both oronasally and intravenously inoculated chickens. Prior to inoculation, all of the chickens were seronegative for HEV. All control chickens were seronegative throughout the study. Anti-avian HEV IgG was detected in 9 of 22 oronasally inoculated and 10 of 23 i.v. inoculated chickens at 1 week postinoculation (wpi) (Table 1). By 3 wpi, all remaining oronasally and i.v. inoculated chickens had seroconverted. OD values differed between the treatment groups (P < 0.0001) and behaved differently for each treatment group over the duration of the study (P < 0.0001) (Fig. 1). Mean OD values for i.v. inoculated chickens peaked at 2 wpi (mean OD ± standard error of the mean [SEM], 0.876 ± 0.034), decreasing by week 5 to 0.417 ± 0.046. Mean OD values for oronasally inoculated chickens increased gradually up to 6 wpi (mean OD ± SEM, 0.778 ± 0.056) and then remained relatively stable.
|
View this table: [in a new window] |
TABLE 1. Seroconversion to avian HEV antibodies in oronasally and i.v. inoculated chickens
|
![]() View larger version (14K): [in a new window] |
FIG. 1. Time courses of seroconversion to avian HEV antibodies for inoculated SPF chickens. The mean ELISA OD values of all chickens from the oral, i.v., and control groups at each week postinoculation are plotted.
|
|
View this table: [in a new window] |
TABLE 2. Detection of avian HEV RNA in fecal, serum, bile, and liver samples from chickens necropsied at different timesa
|
Weekly fecal swab materials and serum samples from inoculated chickens that were not necropsied were also tested for the presence of avian HEV RNA by RT-PCR (Table 3). Avian HEV RNA was detected variably in fecal swab materials from both oronasally and i.v. inoculated chickens (Table 3). Viremia was also detected variably in weekly serum samples (Table 3). For the 13 chickens (4 from the oronasal group, 5 from the i.v. group, and 4 from the control group) that were not necropsied until the end of the study, fecal shedding of viruses and viremia were detected mostly during the first 2 weeks for i.v. inoculated chickens (Table 4). For chickens inoculated by the oronasal route, avian HEV was shed in the feces from 1 to 8 wpi. Viremia in this group lasted from 2 to 5 wpi.
|
View this table: [in a new window] |
TABLE 3. Detection of avian HEV RNA in weekly sera and fecal swabs from chickens during the course of the study
|
|
View this table: [in a new window] |
TABLE 4. Courses of viremia and fecal virus shedding in the 13 chickens that were not necropsied until the end of the study
|
Gross lesions. Gross lesions were observed primarily in the liver. Subcapsular hemorrhages were noticed for 3 of 28 oronasally inoculated chickens, necropsied at 5, 16, and 35 dpi, and for 5 of 29 i.v.-inoculated chickens, necropsied at 3, 5, 7, 16, and 24 dpi (Fig. 2). A slightly enlarged right intermediate lobe of the liver was evident for 4 of 28 oronasally inoculated chickens (necropsied at 5, 7, 20, and 42 dpi) and for 2 of 29 i.v. inoculated chickens (necropsied at 5 and 10 dpi). Control chickens showed no gross hepatic lesions.
![]() View larger version (142K): [in a new window] |
FIG. 2. Gross lesion on a liver from an i.v. inoculated chicken showing subcapsular hemorrhages (arrows).
|
|
View this table: [in a new window] |
TABLE 5. Microscopic liver lesions in control, oronasal, and i.v. group chickensa
|
![]() View larger version (126K): [in a new window] |
FIG. 3. Microscopic lesions of the liver. (A) Liver section from an oronasally inoculated chicken, showing lymphocytic and scattered heterophilic portal vein periphlebitis. (B) Liver section from an i.v. inoculated chicken, showing focally intense lymphocytic venous phlebitis and periphlebitis. (C) Liver section from an i.v. inoculated chicken, showing locally extensive hepatocellular necrosis with lymphocytic inflammatory cell infiltration. (D) Liver section from an i.v. inoculated chicken. Note the architectural disruption and coalescing deposition of hypocellular homogenous eosinophilic matrix with displacement of the hepatocellular cords. (E) Liver section from an oronasally inoculated chicken. Note the large focus of acute hemorrhage with local architectural disruption of the hepatocellular cords and hepatic sinusoids. The tissues were stained with hematoxylin and eosin.
|
|
View this table: [in a new window] |
TABLE 6. Presence of histopathological lesions in various tissues collected at necropsy
|
![]() View larger version (193K): [in a new window] |
FIG. 4. Microscopic lesions in the spleen from an i.v. inoculated chicken. Note the coalescing focus of lymphoid hyperplasia surrounding several ellipsoid artery profiles. The tissue was stained with hematoxylin and eosin.
|
![]() View larger version (20K): [in a new window] |
FIG. 5. Levels of the liver enzyme LDH in sera from inoculated and control chickens. The mean LDH values at each week postinoculation for 13 chickens (4 oronasal, 5 i.v., and 4 control chickens) that were monitored for the entire duration of the study were plotted.
|
|
|
|---|
The discovery of avian HEV in chickens with HS syndrome and the demonstrated antigenic and genetic relatedness between avian HEV and human HEV allowed us to use chickens as a homologous small animal model system to study HEV replication and pathogenesis. In the present study, we successfully infected 60-week-old SPF chickens with a strain of HEV from a chicken by the fecal-oral route as well as the i.v. route. The course of pathogenesis and virus replication in chickens was characterized.
Under field conditions, HS syndrome is characterized by ovarian regression, red fluid in the abdomen, and an enlarged liver and spleen (36, 37). The avian HEV-infected chickens in this study exhibited mild gross pathological lesions characteristic of HS syndrome, such as subcapsular hemorrhages and slight swelling of the liver lobes (16, 19, 40), but the gross lesions were mild and limited to only one-fourth of the infected chickens. Therefore, the gross lesions characteristic of HS syndrome, such as enlargement of the liver and spleen, could not be consistently reproduced in experimentally infected SPF chickens. This was not surprising since our recent study showed that chickens from clinically healthy flocks are also infected by avian HEV (40). It is likely that avian HEV infection is an important factor, but not the sole factor, for the development of clinical HS syndrome. The microscopic liver lesions were mainly lymphocytic, heterophilic periphlebitis and phlebitis with occasional biliary vacuolation, amorphous hypocellular eosinophilic matrixes, hemorrhages, and necrotic foci. Such types of lesions in the liver are characteristic of HS syndrome, which is also called necrotic, hemorrhagic, hepatomegalic hepatitis (42). The foci containing amorphous hypocellular eosinophilic matrixes were possibly made up of serum within ectatic vascular spaces and were similar to the changes described as amyloid-like materials by Tablante et al. (42). However, Congo red staining revealed that the foci were not amyloid. Mild lymphoplasmacytic heterophilic periphlebitic lesions were also observed in some seronegative control chickens. These mild liver lesions are considered normal background for older chickens. Lymphoplasmacytic inflammation and rare focal necrotic foci were also observed in uninfected control pigs and were considered to be normal background changes for pig livers (14). The mean scores of the histopathological liver lesions were statistically significant between either of the two inoculated groups and the negative control group, indicating that the liver lesions in the inoculated chickens could be attributed to avian HEV infection.
Chickens inoculated by either the oronasal or i.v. route seroconverted to avian HEV antibodies, became viremic, and shed the virus in feces. Avian HEV RNA was detected in bile and liver samples, indicating that the virus must have replicated in the liver. Anti-avian HEV IgG appeared and peaked much earlier in i.v. inoculated chickens (2 to 3 weeks) than in oronasally inoculated chickens (4 to 6 weeks). This was anticipated, since in i.v. inoculated chickens, avian HEV directly reached its target organ, the liver, whereas in oronasally inoculated chickens, the virus had to first replicate at primary sites before entering the bloodstream and reaching the liver. Similar to these results, anti-HEV IgG was detected at 2.5 wpi in rhesus monkeys that were intravenously inoculated with a genotype 1 HEV (48). Also, SPF pigs that were intravenously inoculated with a US-2 strain of human HEV seroconverted at 2 to 3 wpi (14, 28). Clearly, the results from this study indicate that the timing of the development of anti-HEV IgG is related to the route of inoculation. In a study on acute sporadic hepatitis E in Egyptian children, anti-HEV IgG was reported to disappear within 6 to 12 months after infection (12). Similarly, the infected chickens in this study, especially the i.v. infected ones, displayed a waning trend in the level of anti-HEV IgG antibodies. This diminishing titer of IgG antibody was also reported for cases in which acute-phase and serial convalescent-phase human or monkey sera were tested by immunoelectron microscopy (5, 32). The decrease in anti-avian HEV IgG titers was less evident for oronasally inoculated chickens than for i.v. inoculated chickens. The pattern of antibody decay observed for oronasally infected chickens likely represents the true pattern of natural HEV infection in humans.
The levels of liver enzymes in sera, including the levels of LDH, AST, CPK, bile acid, total protein, and the A/G ratio, were analyzed. No significant elevations of the liver enzymes AST and CPK or of bile acids were observed. The LDH levels, which were indicative of recent damages to the liver and suggestive of an acute infection, peaked at 1 wpi (Fig. 5) in the oronasally inoculated chickens, which corresponded to seroconversion to avian HEV antibodies (Fig. 1). In the i.v. inoculated group, the LDH levels peaked at 1 wpi (Fig. 5), which preceded the highest titer of anti-avian HEV at 2 wpi (Fig. 1) and severe histopathological lesions in the liver (Table 5). LDH was reported to be the most sensitive indicator of liver cell damage based on tissue enzyme profile studies in racing pigeons (24). Increased LDH activities were observed in 33% of pigeons with aflatoxin B1-induced liver damage (6). It appears that LDH is also a good indicator of hepatic damage in chickens.
The disappearance of viremia corresponded to the rising titer of anti-avian HEV IgG. An oronasally inoculated chicken (chicken no. 4428) (Table 4) shed the virus in its feces for up to 8 wpi in the presence of anti-avian HEV IgG, suggesting the possibility of a persistent infection. This observation prompted us to sequence the virus recovered from this chicken at 8 wpi to determine whether the virus had undergone any mutations during replication that would render it able to escape the neutralizing antibody. However, the sequence recovered from this infected chicken at 8 wpi was identical to that of the inoculum.
Overall, the time to seroconversion and detection of avian HEV in feces, serum, bile, and the liver occurred earlier for i.v. inoculated chickens than for naturally oronasally infected chickens, suggesting that extrahepatic sites of replication exist under natural conditions. It has been suggested that hepatic damage in hepatitis E patients is caused by the immune response to the invading virus and not by the direct replication of the virus in the liver (34). It is unclear how the virus reaches the liver, as HEV is transmitted by the fecal-oral route. By using a pig model, Williams et al. (46) reported that HEV replicates extrahepatically. Replicative, negative-strand HEV RNA was detected in the small intestine, lymph nodes, and the colon (46). Similarly, extrahepatic sites of replication were also reported for pigs that were naturally infected by swine HEV (8). Further studies are warranted to investigate the extrahepatic sites of avian HEV replication.
In summary, we successfully infected chickens with a strain of HEV via the natural route. To our knowledge, this is the first report of a successful experimental oral transmission of HEV in a homologous animal model. Although clinical signs and gross lesions characteristic of HS syndrome were not consistently reproduced, characteristic microscopic liver lesions consistent with HS syndrome were reproduced in both oronasally and i.v. infected birds. Both oronasally and i.v. infected chickens developed an infection similar to that of human HEV infection in monkeys. The availability of a chicken model should help us to further study the mechanisms of HEV replication and pathogenesis in the future.
This study was supported by grants from the National Institutes of Health (AI 01653, AI 46505, and AI 50611) and from the U.S. Department of Agriculture National Research Initiative Competitive Grant Program (NRI 35204-12531).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»