Skip to main content
ARS Home » Pacific West Area » Albany, California » Western Regional Research Center » Produce Safety and Microbiology Research » Research » Publications at this Location » Publication #321269
Title: Comparison of in vivo and in vitro assays as methods for measuring infectivity of tulane virus

Author
item Tian, Peng
item LEI, SHAN - Volunteer
item WANG, DEPENG - University Of Shanghai
item Yang, David

Submitted to: Journal of Virological Methods
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/3/2016
Publication Date: 1/3/2016
Citation: Tian, P., Lei, S., Wang, D., Yang, D.H. 2016. Comparison of in vivo and in vitro assays as methods for measuring infectivity of tulane virus. Journal of Virological Methods. doi: 10.1016.

Interpretive Summary: Human noroviruses (HuNoV) are pathogens of significant concern, but as of current they cannot be cultured, which limits the ways in which they can be studied. Culturable viruses like feline calicivirus (FCV), murine norovirus (MNV) and Tulane virus (TV) have been used as surrogates for HuNoV. Plaque assay and the tissue culture 50% infectious dose (TCID50) assay are the most-commonly used in vivo assays for quantifying the infectivity of surrogate viruses. However, the correlation between TCID50 and plaque forming unit (pfu) has not been directly compared. Not knowing the link between the PFU and TCID50 makes it is harder to verify results between laboratories. The molecular approaches are the commonly used in vitro assays to identify the presence of HuNoV and/or surrogates. However, the correlation between the in vitro assays and in vivo assays remains unknown. Scientists at Produce Safety and Microbiology Research Unit at WRRC evaluated the correlation between two in vivo assays as well as correlations between the in vivo assays and in vitro assays by using TV as a surrogate for HuNoV. Ten batches of viral stocks ranging from 3.4x105 to 6.7x106 pfu were used for vivo and vitro assays side by side. Plaque assay and TCID50 assay were used as in vivo assays to measure infectivity of the virus. RNase exposure assay, viral-receptor-binding-based In Situ Capture-qRT-PCR (PGM-ISC-RTq-PCR) and antibody-binding-based In Situ Capture qRT-PCR (Ab-ISC-RTq-PCR) assay were used as in vitro assays to measure encapsidated viral RNA. We demonstrated that the in vivo assays were mostly consistent each other, and the TCID50 assay was more sensitive than the plaque assay. There was a 6.7-fold difference in the numerical results of these methods of enumeration. However, the results of in vivo assays were not consistent with the results from in vitro assays. The results suggested that the encapsidated viral RNA level was not an indicator for viral infectivity.

Technical Abstract: Human noroviruses (HuNoV) are pathogens of significant concern, but as of current they cannot be cultured, which limits the ways in which they can be studied. Therefore, culturable viruses like feline calicivirus (FCV), murine norovirus (MNV) and Tulane virus (TV) have been used as surrogates for HuNoV. The in vivo assays most-commonly used to quantify the infectivity of these surrogate viruses are the plaque assay and tissue culture 50% infectious dose (TCID50) assay. In vitro assays utilizing molecular approaches are also commonly used to identify the presence of HuNoV and/or its surrogates. However, the exact correlation (if any) between the in vivo and in vitro assays remains unknown. In this study, we used TV as a surrogate for HuNoV to evaluate for correlation between two in vivo assays and three in vitro assays. Specifically, the in vivo plaque and TCID50 assays measure for infectious virus particles, while the in vitro RNase exposure, Porcine Gastric Mucin In-Situ-Capture qRT-PCR (PGM-ISC-qRT-PCR), and Antibody In-Situ-Capture qRT-PCR (Ab-ISC-qRT-PCR) assays measure for an amplicon within viral genomic RNA that were correctly packaged into integritous viral capsids. Ten batches of viral stocks ranging from 3.4 x105 to 6.7x106 plaque forming units (PFUs) were used for side by side comparison with pfu as a reference. The results indicate that one PFU was equivalent to 6.7 ± 2.3 TCID50 units, 9.8 ± 10.9 RNase-untreated genomic copies (GCs), 2.8 ± 3.0 RNase-treated GCs, 0.05±0.07 PGM-ISC-qRT-PCR GCs, and 0.54±0.7 Ab-ISC-qRT-PCR GCs. We observed that while the in vivo assays were consistent with each other, the TCID50 assay was 6.7 times more sensitive than the plaque assay. In contrast, the in vitro assays were not consistent with the in vivo assays. An inverse correlation was found between GCs as measured by RNase exposure assay, and PFUs as measured by plaque assay. The very high variations in GCs as measured by both ISC-RT-qPCR assays made them impossible to correlate against the relatively small variations (<20-fold) in the PFUs or TCID50 units as measured by the in vivo assays. Overall, our results suggest that our targeted amplicons within integritous viral capsids was not an indicator for viral infectivity.