Too much of a good thing? IFITM3 only restricts Zika virus when overexpressed

From the Overbaugh lab, Human Biology

The first known cases of Zika virus (ZIKV) in the Americas were documented in March of 2015 when a massive outbreak of febrile illness was detected in a northeastern state in Brazil. A year later, the virus had spread to at least 33 countries and territories in the Americas. ZIKV was introduced into Brazil from the Pacific islands, where large outbreaks of Asian-lineage ZIKV strains had occurred since 2007. As the virus spread throughout the Pacific and later in the Americas, cases of fetal microcephaly and other congenital malformations surged, a new disease syndrome of ZIKV infection. Before 2007, ZIKV infection caused sporadic outbreaks of mild febrile illness in Africa, which suggested that the ZIKV lineages circulating in the Pacific regions and the Americas might have distinct pathogenic properties from the African ZIKV lineages. Ted Gobillot, a former M.D./Ph.D. student in the Overbaugh lab, investigated if the differences in the pathogenic features between African-lineage and Asian-lineage ZIKV strains were associated with different susceptibilities to the cell-intrinsic antiviral state induced by interferon (IFN). The group found that African-lineage and Asian-lineage ZIKV strains have different sensitivities to IFN and ruled out an antiviral gene as the basis for the difference. The findings were recently published in the journal Viruses.

Interferon controls an arm of the innate immune system that responds to common features of pathogens as the first line of defense.  IFN signaling turns on the expression of hundreds of Interferon Stimulated Genes (ISGs), some of which encode antiviral proteins that can affect virus control and overall pathogenesis. Before this study, it had been well-documented –in vitro and in vivo– that overall, ZIKV strains are highly sensitive to the inhibitory effects of IFN. Still, it was unknown if African-lineage and Asian-lineage strains differed in IFN-sensitivity. To address this question, the investigators selected nine ZIKV strains: five isolates from the African-lineage and four isolates from the American outbreak within the Asian-lineage. Each strain was used to infect the human epithelial cell line A549 with or without pretreatment of IFN. After a few days, the investigators determined if IFN pretreatment affected viral replication by collecting the supernatant from cells grown under both conditions and measuring the output virus. In agreement with previous reports, for all strains, IFN treatment reduced the amount of output virus compared with the cells without IFN treatment. However, Asian-lineage strains were more sensitive to IFN treatment than African-lineage strains.

Zika-infected cell produces interferon. IFN restricts American-outbreak Zika strains more potently than African Zika strains. The ISG IFITM3 only restricts Zika virus when over-expressed in vitro.
IFN restricts American-outbreak Zika strains more potently than African Zika strains. The ISG IFITM3 only restricts Zika virus when over-expressed in vitro. Image provided by Ted Gobillot

The transmembrane protein IFITM3 was one of the first ISGs described to have antiviral activity against ZIKV. To determine if IFITM3 expression could explain the IFN sensitivity of between ZIKV strains, the investigators engineered an A549 cell line that expressed a tagged version of IFITM3 and selected cells with physiologically relevant levels of IFITM3 protein –relative to endogenous levels induced by IFN– and called it “IFIM3-rel”. They then infected the A549-IFTIM3-rel cells with the most IFN-sensitive African-lineage and Asian-lineage ZIKV strains in the absence and presence of IFN. The results surprised the researchers: IFITM3 alone did not block replication of either ZIKV strain; the levels of infection were similar to the control condition without IFN treatment or exogenous IFITM3. Importantly, IFN treatment in IFITM3-rel cells potently reduced ZIKV replication. Principal investigator, Dr. Julie Overbaugh, reflected on this unexpected finding: “I will say he [Ted Gobillot] worked very hard on this paper because he was showing that previous studies missed the mark, which is always an uphill battle, but very important in getting the science right for the long term.”

Previous studies have reported potent inhibition of ZIKV infection mediated by IFITM3; therefore, investigators sought to determine if overexpression was the culprit. To this end, they developed an A549 cell line with high IFITM3 expression “IFTIM3-high” and infected it with an African-lineage isolate. Indeed, ZIKV replication was potently inhibited in the IFITM3-high cell line. Next, the investigators questioned if their method of detecting infection at the final stage of infection (i.e., detection of output virus) could miss restriction at earlier time points, which could explain the lack of antiviral activity in the A549-rel cell line. To test this hypothesis, they stained ZIKV-infected IFITM3-rel and IFITM3-high cell lines a day after infection with an antibody to detect the ZIKV E-protein, and calculated the percentage of cells that were positive. Consistent with their previous findings in the study, the IFITM3-rel had similar levels of E-protein positive cells as the control cell line; however, the IFITM3-high levels had a significant reduction in E-protein positive cells. This set of experiments demonstrated that IFITM3 restricts ZIKV replication only when expressed at high levels.

Finally, to address if IFITM3 loss had any effects in the IFN-mediated restriction of African-lineage or Asian-lineage ZIKV strains, the investigators knocked out IFITM3 in the A549 cell line using the CRISPR/Cas9 system. They infected IFIM3-knock out cells with both strains in the presence or absence of IFN. They found that loss of IFITM3 did not impair the IFN-induced restriction of ZIKV, showing that an ISG, other than IFITM3, is responsible for the antiviral activity induced by IFN treatment. Ted Gobillot, shared new questions that stemmed from the study and future directions:

“(1) What ISGs are responsible for the potent restriction of Zika? We are currently using some new CRISPR screening approaches to identify interferon-induced host proteins that restrict Zika. (2) Are specific ISGs driving differences in interferon susceptibility between African and American Zika strains?  Are there viral sequence determinants that could be driving differences in interferon-susceptibility?”

This research was supported by grants from National Institutes of Health.

UW/Fred Hutch Cancer Consortium member Dr. Julie Overbaugh contributed to this work.

Gobillot, T. A., Humes, D., Sharma, A., Kikawa, C., & Overbaugh, J. 2020. The robust restriction of Zika virus by type-I interferon in A549 cells varies by viral lineage and is not determined by IFITM3. Viruses, 12(5).