Although several novel viruses have been identified in human stool 
, few studies have examined humoral responses to them. We have established a robust, sensitive serology platform that is ideally suited for pathogen discovery applications. LIPS holds several advantages over ELISA for studying humoral responses against potentially new infectious agents stemming from the low background binding, high sensitivity innate to this assay and the ability to rapidly test different antigens/antigen fragments using a standard format with little or no assay optimization 
. Here we used LIPS assays to demonstrate humoral responses to the HMOAstV-C Astrovirus capsid protein and provide experimental data supporting human infection.
Several different lines of evidence indicate that the HMOAstV-C capsid fragment is a target of humoral responses. First, only the C-terminal capsid region of HMOAstV-C was immunoreactive with human serum samples. Although RNA from HMOAstV-A was previously identified in human stool samples, we were unable to detect human immunoreactivity to this analogous C-terminal capsid region of HMOAstV-A. The exact reason for the lack of immunoreactivity is not clear, but it is possible that this capsid region of HMOAstV-A is less antigenic or its conformational folding is not recapitulated in the current Renilla
luciferase fusion protein using the LIPS system. Second, the HMOAstV-C astrovirus capsid region used in LIPS was relatively unique and had no significant homology with the eight known serotypes of HAstV strains. Based on its homology, it is possible that some of the observed HMOAstV-C capsid immunoreactivity is against the related HMOAstV-B capsid. However, our previous LIPS studies demonstrate markedly different serologic responses to similarly homologous antigens in subjects infected with related pathogens 
. In these studies, variations in both linear and conformational epitopes are more clearly differentiated using the LIPS liquid assay than using ELISAs 
. Further studies examining the humoral responses against the HMOAstV-B capsid, as well as additional proteins from the HMOAstV viruses should resolve this issue. Third, the anti-HMOAstV-C capsid antibody titers detected by LIPS assay in human samples was substantially higher than the background binding associated with mock protein A/G beads alone. This magnitude is comparable to those seen by LIPS against other infection agents such as Borrelia burgdorferi 
and Kaposi Sarcoma associated virus 
. Our strategy of using buffer blanks instead of seronegative uninfected samples has been used in other seroepidemiologic investigations and provide comparable diagnostic thresholds as seronegative samples 
. Fourth, the relatively low immunoreactivity against HMOAstV-C in young children (age 6-12 months) and a corresponding increase in seroprevalence with childhood age is consistent with human infection. Lastly, additional studies evaluating rabbit, pig and horse samples showed that anti-HMOAstV-C antibodies were rare or absent in animals. The lack of immunoreactivity against HMOAstV-C in animals, despite the greater homology of HMOAstV-C to animal Astroviruses 
, supports the notion that humans are a host for these viruses. The approach of analyzing humoral responses to a panel of different animal serums side-by-side human samples by LIPS is useful for understanding the host range for this and other potential pathogens.
Overall the results suggest that HMOAstV-C is a common infectious agent circulating in human populations. It is also possible that the studies described here underestimate the true seroprevalence of HMOAstV-C that may result from transient antibody titers from older infections and the genetic diversity of this newly recognized agent. The higher prevalence of antibodies in adults (65%) suggests that most adults are seropositive; however, it is unknown whether the presence of antibody is protective against infection. Nevertheless, compared to children, adults show a markedly higher incidence of HMOAstV-C infection. Furthermore, the detection of high prevalence of antibodies in the 0-6 month old children is consistent with the presence of maternal antibodies. In infants, waning maternal antibodies for many infectious agents is often observed 
Additional investigation including prospective analysis of antibody responses by LIPS from acute infection would be worthwhile. It is interesting to note that HMOAstV-C/AstV-VA1 was identified as an Astrovirus associated with an outbreak of gastroenteritis in a child daycare center in Virginia 
. RT-PCR and sequencing of affected individuals showed identical HMOAstV-C sequences. It is highly likely that there are diverse HMOAstV-C strains causing a spectrum of clinical symptoms. Along these lines, an Astrovirus, phylogenetically related to HMOAstV-C (95% identical capsids) was recently detected in the brain of a child with encephalitis who had a genetic form of agammaglobulinemia 
. Further studies are clearly needed to establish the clinical spectrum of disease caused by HMOAstV-C. Although a single antigenic capsid target was diagnostically useful here with HMOAstV-C, the ability of detecting antibody responses to multiple independent protein fragments can increase assay sensitivity and has the potential to stratify different conditions caused by the same infectious agent 
. The development of Renilla
luciferase antigen fusions for additional HMOAstV-C Astrovirus proteins and other Astrovirus in the LIPS system will likely be useful for identifying and understanding their role in diarrheal and other illnesses.