The E3 ligase TRIM56 is a host restriction factor of Zika virus and depends on its RNA-binding activity but not miRNA regulation, for antiviral function

Yang, Darong; Li, Nan L.; Wei, Dahai; Liu, Baoming; Guo, Fang; Elbahesh, Husni; Zhang, Yunzhi; Zhou, Zhi; Chen, Guo-Yun; Li, Kui

To study host factors that impact ZIKV replication, we first surveyed a number of human and animal cell lines of various tissue origins for ZIKV susceptibility. Cells were infected by MR766, a prototype African strain of ZIKV, followed by immunofluorescence staining of viral E protein expression at 72 h.p.i. Three human hepatocyte cell lines, including Huh7, PH5CH8, and HLCZ01, were all susceptible to ZIKV infection, albeit to different degrees (S1 Fig). The hepatoma Huh7 cells were previously reported to support ZIKV infection [44] . We found Huh7 cells supported the highest efficiency of ZIKV infection, with 100% of cells infected and exhibiting significant cytopathic effect (CPE) at 72 h.p.i. HLCZ01, a recently established human hepatoma line, was also highly susceptible, with >90% cells infected by ZIKV. In comparison, PH5CH8, a non-neoplastic hepatocyte line that harbors intact innate immune responses resembling primary human hepatocytes [38, 40, 45] , was permissive for ZIKV but less susceptible than Huh7 and HLCZ01. HeLa-FitA2 cells, which were derived from the human cervical epithelial carcinoma cell line HeLa and stably express the Tet repressor and harbors a single integrated Flp recombination target site allowing for rapid generation of Tetinducible stable cell lines, also supported robust ZIKV infection (S2 Fig). The human embryonic kidney HEK293 cells had been reported to support low efficiency of ZIKV infection [11] . We examined three different HEK293-derivatives, HEK293-FIT, HEK293-T3Y and 293T cells, and found they supported varying degrees of ZIKV replication (see below). A notable feature of ZIKV infection in humans is its neurotropism [46] [47] [48] . We thus examined two human cell lines of neural origin, i.e., the neuroblastoma line SK-N-SH and SVGA, a fetal glial cell line, and found both to be permissive for ZIKV infection (S3 Fig). Apart from the human cell lines, we found the African green monkey kidney cell line BSC1 and the Madin-Darby canine kidney (MDCK) cell line both to be susceptible to ZIKV infection (S4 Fig). We also examined the susceptibility of three commonly used mouse cell lines, Hepa1-6, L929 and MEF, to ZIKV infection and compare these lines with Huh7 and HEK293, which represent human cell lines with relatively high and low permissiveness for ZIKV infection, respectively. Cells were infected with ZIKV-MR766 for different time periods, followed by quantification of intracellular viral RNA abundance by qPCR that sensitively detects viral RNA replication. As shown in S5 Fig, Huh7 cells were highly permissive, harboring 10 8 copies of ZIKV RNA per microgram of total RNA already at 24 h.p.i. There was a further,~4-fold increase in viral RNA levels at 48 h.p.i., which reached sub-10 9 copies/μg total RNA range. In contrast, HEK293 cells were much less permissive, harboring mid-10 6 copies/μg total RNA range of viral RNAs at 24 h.p.i. ZIKV RNA replication increased by~18-fold in HEK293 cells at 48 h.p.i, approaching that of Huh7 cells at 24 h.p.i. However, all three murine cell lines examined (hepa1-6, L929 and MEF) exhibited limited permissiveness, if any, for ZIKV, harboring mid-10 5 to~10 6 copies/μg total RNA range viral RNA at 24 h.p.i. Moreover, viral RNA levels did not increase in hepa1-6 and MEFs, while had a mere~2-fold uptick in L929 cells, at 48 h.p.i. Collectively, these data demonstrate that ZIKV infects and replicates in a broad range of human and animal cell lines, although the susceptib