Binding of Escherichia coli does not protect tulane virus from heat-inactivation regardless the expression of HBGA-like molecules

Authors: LI, QIANQIAN - Shanghai Institute Of Technology; WANG, DAPENG - Xan Jiao Tong University; Yang, David; SHAN, LEI - University Of Minnesota; Tian, Peng

Submitted to: Frontiers in Microbiology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/28/2017
Publication Date: 9/21/2017
Citation: Li, Q., Wang, D., Yang, D., Shan, L., Tian, P. 2017. Binding of Escherichia coli does not protect tulane virus from heat-inactivation regardless the expression of HBGA-like molecules. Frontiers in Microbiology. 8:1746. https://doi.org/10.3389/fmicb.2017.01746.
DOI: https://doi.org/10.3389/fmicb.2017.01746

Interpretive Summary: Human noroviruses (HuNoVs) are the major cause of gastroenteritis outbreaks worldwide. Human noroviruses can interact with histo-blood group antigens (HBGAs) on the surface of mammalian cells as well as bacterial cells. HBGAs have been considered as putative receptors or co-receptors for HuNoVs in mammalian cells. HBGAs could also be expressed on the surface of bacteria. It has been suggested that binding of HuNoV-derived virus-like particles to HBGA-expressing bacteria increases the stability of the viral capsid to better resist heat denaturation. It remains unknown if viral binding to HGBA-expressing bacteria protects the virus against heat denaturation. In this study, we tested for HBGA-binding-conveyed protection against heat-denaturation of the viral capsid using Tulane virus (TV) and Escherichia coli O86:H2, with E. coli K-12 used as a control. Expression of HBGA type B was confirmed by ELISA to be positive with E. coli O86:H2 and negative with E. coli K-12. Binding of TV was confirmed by ELISA to be strong with E. coli O86:H2 and weak with E. coli K-12. We determined via ELISA that pre-incubation of TV with free HBGA could completely inhibit its ability to bind to E. coli O86:H2 (p=0.006), while producing no significant change in its ability to bind E. coli K-12 (p=0.28). We utilized a bacterial-capture (BC) RT-qPCR procedure to confirm that both bacterial strains were capable of binding TV, and that E. coli O86:H2 exhibited fivefold greater binding capacity than E. coli K-12. Pre-incubation of TV with free HBGA would partially inhibit the binding of TV to E. coli O86:H2 (p=0.047). In contrast, not only did pre-incubation of TV with free HBGA not inhibit the binding of TV to E. coli K-12, binding was slightly (but not significantly) enhanced (p=0.13). The viral infectivity assay allowed us to conduct a direct evaluation of the ability of HBGA-bound bacteria to confer heat-denaturation protection to TV. TV was pre-incubated with each bacterial strain at ratios of 1:1 and 100:1, then both partially and fully heat-denatured. The heat-denatured complexes were infected to host cell culture (LLC-MK2), and viral amplification was quantitated by RT-qPCR 48h later. The binding of bacteria to TV reduced viral replication in host cell culture (LLC-MK2) in a dose-dependent matter. We found that neither bound E. coli O86:H2 nor K-12 conferred protection of TV against partial or full heat denaturation conditions. Partial heat-denaturation reduction of viral replication was not significantly impacted by the binding of either bacterial strain, with infectivity losses of 99.23%, 99.67%, 97.92%, 97.58%, and 91.37% for TV w/o bacteria, TV w/O86:H2 (1:1), TV w/O86:H2 (1:100), TV w/K-12 (1:1), and TV w/K-12 (1:100), respectively. Full heat-denaturation reduction of viral replication was not impacted by the binding of either bacterial strain, as full loss of infectivity was observed in all cases.

Technical Abstract: Human noroviruses (HuNoVs) are the major cause of gastroenteritis outbreaks worldwide. Human noroviruses can interact with histo-blood group antigens (HBGAs) on the surface of mammalian cells as well as bacterial cells. HBGAs have been considered as putative receptors or co-receptors for HuNoVs in mammalian cells. HBGAs could also be expressed on the surface of bacteria. It has been suggested that binding of HuNoV-derived virus-like particles to HBGA-expressing bacteria increases the stability of the viral capsid to better resist heat denaturation. It remains unknown if viral binding to HGBA-expressing bacteria protects the virus against heat denaturation. In this study, we tested for HBGA-binding-conveyed protection against heat-denaturation of the viral capsid using Tulane virus (TV) and Escherichia coli O86:H2, with E. coli K-12 used as a control. Expression of HBGA type B was confirmed by ELISA to be positive with E. coli O86:H2 and negative with E. coli K-12. Binding of TV was confirmed by ELISA to be strong with E. coli O86:H2 and weak with E. coli K-12. We determined via ELISA that pre-incubation of TV with free HBGA could completely inhibit its ability to bind to E. coli O86:H2 (p=0.006), while producing no significant change in its ability to bind E. coli K-12 (p=0.28). We utilized a bacterial-capture (BC) RT-qPCR procedure to confirm that both bacterial strains were capable of binding TV, and that E. coli O86:H2 exhibited fivefold greater binding capacity than E. coli K-12. Pre-incubation of TV with free HBGA would partially inhibit the binding of TV to E. coli O86:H2 (p=0.047). In contrast, not only did pre-incubation of TV with free HBGA not inhibit the binding of TV to E. coli K-12, binding was slightly (but not significantly) enhanced (p=0.13). The viral infectivity assay allowed us to conduct a direct evaluation of the ability of HBGA-bound bacteria to confer heat-denaturation protection to TV. TV was pre-incubated with each bacterial strain at ratios of 1:1 and 100:1, then both partially and fully heat-denatured. The heat-denatured complexes were infected to host cell culture (LLC-MK2), and viral amplification was quantitated by RT-qPCR 48h later. The binding of bacteria to TV reduced viral replication in host cell culture (LLC-MK2) in a dose-dependent matter. We found that neither bound E. coli O86:H2 nor K-12 conferred protection of TV against partial or full heat denaturation conditions. Partial heat-denaturation reduction of viral replication was not significantly impacted by the binding of either bacterial strain, with infectivity losses of 99.23%, 99.67%, 97.92%, 97.58%, and 91.37% for TV w/o bacteria, TV w/O86:H2 (1:1), TV w/O86:H2 (1:100), TV w/K-12 (1:1), and TV w/K-12 (1:100), respectively. Full heat-denaturation reduction of viral replication was not impacted by the binding of either bacterial strain, as full loss of infectivity was observed in all cases.