Immune response to and pathogenesis of porcine epidemic diarrhea virus
Porcine epidemic diarrhea (PED) virus causes effusive diarrhea in pigs of all ages and high mortality among pre-weaning pigs. The PED virus (PEDV) is one of several important swine pathogens, which causes significant economic losses to the swine industry. The first outbreak of PED in the United States occurred in 2013. Since then, various studies have been conducted to characterize the virus and its pathogenesis/pathogenicity and to develop diagnostic testing and reagents for the virus. The goals of the following three experiments in this dissertation were to characterize protective host antibody response against the PEDV, elucidate mechanisms responsible for the long duration of viral fecal shedding, and identify the putative receptor-binding domain of the virus.
The first study characterized clinical and antibody response of pigs following experimental PEDV infection/re-infection. After experimental inoculation, naive 3-week-old pigs developed diarrhea starting 2 days post infection (dpi). While most of the pigs recovered by 10 dpi, the virus continued to shed in feces from all pigs until 14 dpi and in some pigs up to 35 dpi. The pigs developed virus-specific antibodies of all major isotypes as early as 7-10 dpi, and serum virus neutralizing (VN) antibody was detectable in the majority of pigs (92%) by 14 dpi, suggesting that antibody appearance, particularly VN antibody, may coincide with the progress toward recovery. While all structural proteins of PEDV except envelope protein were found to be immunogenic, the appearance of antibody against spike (S) protein seemed to be associated with the development of neutralizing activity in blood circulation. When the pigs were re-exposed to the same strain of PEDV (i.e., homologous challenge) almost 2 months after the initial infection, neither clinical signs of diarrhea and other abnormalities nor viremia were observed in the pigs. Fecal shedding of the virus, however, occurred for 14 dpi even though those pigs had detectable antibodies in serum and oral fluid at the time of re-exposure, suggesting that prior infection can provide clinical protection against a homologous virus challenge but not sterile immunity. As antibody profiling of oral fluids demonstrated that a relatively high level of IgA was maintained until re-inoculated whereas IgG in both sera and oral fluids continued to decline after 21-28 dpi, mucosal IgA may be responsible for clinical protection from the subsequent inoculation.
The second study was conducted to evaluate if the presence of PEDV in non-intestinal tissue(s) contributes to prolonged fecal shedding of the virus following infection. Blood and various organs collected from pigs after experimental inoculation of PEDV were examined by a real-time RT-PCR and/or tissue assays such as immunohistochemistry and fluorescent in situ hybridization for viral RNA or antigen. Viremia observed in the challenged pigs starting on dpi 1 and then disappeared after 7 dpi. PEDV RNA was also detected in various tissues besides the small intestine as early as 1 dpi, which coincided with viremia, but was no longer detectable after 14 dpi except in the small intestine and mesenteric lymph node (MLN) in which viral RNA continued to be detected at 28 dpi. The virus could be isolated from MLN until 14 dpi. The findings from the study did not support the contribution of non-intestinal tissues to fecal shedding for the extended period (35 dpi). Longer duration of fecal shedding of PEDV may be attributed to viral replication in the small intestine while viremia may contribute to wide tissue distribution of the virus.
In the third study, a putative receptor-binding domain (RBD) of PEDV was identified in the S1 region of the S protein via bioinformatics, comparative structural biology, and homology modeling for protein structure in comparison to alphacoronaviruses with known RBD sequence. A recombinant protein was produced from that gene fragment, which was recognized by anti-PEDV swine sera from pigs experimentally inoculated with the virus, indicating the recombinant RBD protein maintained its native form as it was in the virus. Rabbit polyclonal and rat monoclonal antibodies that were generated against the putative RBD protein were able to neutralize PEDV infection in Vero cells, which are permissive to PEDV. These findings suggest the presence of neutralizing epitope(s) in the putative RBD protein and its potential utility as a vaccine antigen or antigenic basis for serodiagnosis. Furthermore, the protein bound to the surface of Vero cell, suggesting that the putative RBD identified in the study should be a receptor-binding domain of PEDV. Since Vero cells do not possess aminopeptidase N (APN) which is a known cellular receptor for alphacoronaviruses, PEDV may bind to a cellular receptor other than APN.
Overall, the present research contributes to the understanding the ontogeny of antibody response to PEDV and protective immunity to a subsequent infection, the pathogenesis of PEDV in weaned pigs, and virus attachment to permissive cells at the cellular level, leading to better prevention and control of PED.