To identify TBE virus-endemic areas, it is effective to conduct an epizootiological survey of wild rodents. The neutralizing test can be used for serological diagnosis of wild rodents, but it is time consuming and uses hazardous live viruses that require a high-level MAPK inhibitor biosafety facility. It is also known that non-infected wild rodents sometime indicated low neutralization antibody titers by the neutralization test. Therefore, a diagnosis which is more convenient for the epizootiological survey of wild rodents is required. In this study, we tried to develop ELISAs using two recombinant antigens

in the serological diagnosis of rodents for the first time. Domain III of the E protein was known to have the neutralizing epitopes (11) and was used for the serological diagnosis in several flaviviruses (13, 14). In this study, the recombinant domain III of the E protein was applied to the diagnosis ELISA for wild rodents. The EdIII-ELISA was shown to

have a relatively high sensitivity (27/35, 77.1%) and specificity (68/85, 80.0%) as compared with the neutralization test when the cut-off value for the ELISA was set at 0.64 (Fig. 2). Eight of 35 this website neutralization test-positive samples were negative in the EdIII-ELISA (Table 1). Several false-positive samples showed high reactivity to the negative control antigens, NusA (data not shown). In another study it was reported that a neutralizing response to West Nile virus in naturally infected horses was induced with epitopes within not only EdIII, but also other domains (25). It was suggested that these false-negatives were due to the lack of other domains and the Thiamine-diphosphate kinase conformational structure of the EdIII expressed in E. coli, and to the presence of antibodies that have high reactivity to NusA -Tag protein. In the flavivirus, co-expression of prM and E proteins in mammalian cells leads to the secretion of SPs to culture medium (19, 26, 27). The SPs have no viral

genome and do not produce progeny virus, and they have similar antigenicity and immunogenicity to the native virus. Therefore, SPs have been developed as a safe and useful alternative for live viruses as the antigen for serological diagnosis tests and vaccines (18, 20, 28, 29). In this study, the SPs were used as the antigens in ELISA to detect TBE virus-infected rodents. The SP-ELISA was shown to have a very high sensitivity (32/35, 91.4%) and specificity (85/85, 100%) as compared with the neutralization test when the cut-off value for the ELISA was set at a 0.089 (Fig. 4). In a recent study, it was reported that the antigenic structures of E proteins were disturbed when the ELISA plate was coated directly with the viral particles as solid-phase antigens (30). To avoid this, our SP-ELISA uses capture antibodies to coat the SP-antigen on the plate. And unlike infectious virions, the SPs do not require formalin inactivation, which affects the reactivity of several epitopes of the E proteins (31).