UFMylation promotes orthoflavivirus infectious particle production

Abstract

Post-translational modifications play crucial roles in regulating viral infections, yet roles for many modifications remain unexplored in orthoflavivirus biology. Here, we demonstrate that the UFMylation system, a post-translational modification pathway that catalyzes the transfer of UFM1 onto proteins and promotes infection by multiple orthoflaviviruses, including dengue virus (DENV), Zika virus (ZIKV), West Nile virus, and yellow fever virus. We found that depletion of the UFMylation E3 ligase complex proteins UFL1 and UFBP1, as well as other UFMylation machinery components (UBA5, UFC1, and UFM1), significantly reduces orthoflavivirus infectious virion production. This regulation was specific to orthoflaviviruses as the hepacivirus and member of the broader Flaviviridae family, hepatitis C virus, was not regulated by UFL1. Mechanistically, UFMylation did not regulate viral RNA translation, RNA replication, or virion egress but instead affected the assembly of infectious virions. Furthermore, we identified novel interactions between UFL1 and several viral proteins involved in orthoflavivirus virion assembly, including NS2A, NS2B-NS3, and capsid. These findings establish UFMylation as a previously unrecognized post-translational modification pathway that promotes orthoflavivirus infection through modulation of viral assembly. This work expands our understanding of the post-translational modifications that control orthoflavivirus infection and identifies new potential therapeutic targets.IMPORTANCEOrthoflaviviruses depend on host-mediated post-translational modifications to successfully complete their life cycle, yet many of these critical interactions remain undefined. Here, we describe a role for a post-translational modification pathway, UFMylation, in promoting infectious particle production of ZIKV and DENV. We show that UFMylation is dispensable for initial RNA translation and RNA replication but promotes the assembly of infectious virions. Additionally, we find that regulation of infection by UFMylation extends to other orthoflaviviruses, including West Nile virus and yellow fever virus, but not to the broader Flaviviridae family. Finally, we demonstrate that UFMylation machinery directly interacts with specific DENV and ZIKV proteins during infection. These studies reveal a previously unrecognized role for UFMylation in regulating orthoflavivirus infection.

Keywords: PTM; UFM1; UFMylation; flavivirus; orthoflavivirus; post-translational modification; ubiquitination.

Fig 1

The UFMylation E3 ligase complex proteins promote mosquito-borne orthoflavivirus infection. (A and B) Focus-forming assay of supernatants from Huh7 cells infected with DENV^NGC or ZIKV^PRVABC59 (48 h, MOI 0.1) after siRNA depletion of the indicated transcripts or non-targeting control (CTRL), shown as percentage of siCTRL. (C) Cell viability measured after siRNA depletion of the indicated transcripts at 72 h post-transfection, relative to that of siCTRL, as measured by Cell-Titer GLO assay. (D) Immunoblot analysis of protein expression from Huh7 cells treated with the indicated siRNAs for 72 h. (E) Focus-forming assay of supernatants harvested from A549 cells infected with ZIKV^PRVABC59 (48 h, MOI 0.1) after siRNA depletion of the indicated transcripts. (F) Immunofluorescence micrographs of Huh7 cells treated with the indicated siRNA and then infected with the following viruses for 48 h: ZIKV^PRVABC59, MOI 0.1; YFV^17D, MOI 0.01; WNV^NY2000, MOI 0.01, or HCV^JFH1, MOI 1, as measured by immunostaining of viral antigens (E for ZIKV, YFV^17D, and WNV, and NS5A for HCV; green). Nuclei were stained with Hoechst (blue). (Right) Quantification of the percentage of virus-infected Huh7 cells, shown relative to siCTRL with an average of 5,000 cells counted for each condition. For all panels, n = 3 biologically independent experiments, with bars indicating mean and error bars showing the standard error of the mean. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, or not significant (ns) as determined by paired t-test (A and B), one-way analysis of variance (ANOVA) with Dunnett’s multiple comparison test (C and E), or two-way ANOVA followed by Šidák’s multiple comparison test (F).

Fig 2

The UFMylation E3 ligase complex does not regulate interferon induction during ZIKV infection. RT-qPCR analysis (relative to RPL30) of RNA extracted from A549 cells transfected with either siCTRL or siUFL1 followed by mock or ZIKV^PRVABC59 infection (24 h, MOI 0.1). n = 3 biologically independent experiments, with bars indicating the mean values and error bars showing the standard errors of the mean. ***P < 0.001, ****P < 0.0001, or not significant (ns), as determined by two-way ANOVA of log-transformed data followed by Šidák’s multiple comparisons test.

Fig 3

The UFMylation E3 ligase complex does not regulate orthoflavivirus translation or RNA replication. (A) Luciferase activity of Gaussia luciferase-encoding ZIKV^MR766 (ZIKV-GLuc, MOI 0.1) from infected Huh7 cells treated with dimethyl sulfoxide (DMSO), MK0608, or cycloheximide during infection and harvested at the indicated time points. (B) Normalized expression of ZIKV-GLuc from infected Huh7 cells treated with non-targeting control (CTRL) or UFL1 siRNA harvested at the indicated time points. (C) Luciferase activity of Gaussia luciferase-encoding ZIKV^MR766 (ZIKV-GLuc, MOI 0.1) from the supernatant of infected Huh7 cells treated with DMSO or MK0608 harvested at 24, 48, or 72 hpi. (D) Normalized luciferase activity of ZIKV-GLuc from the supernatant of infected Huh7 cells treated with CTRL or UFL1 siRNA and harvested at 24, 48, or 72 hpi. (E) Immunofluorescence micrographs of Huh7 cells treated with the indicated siRNA and then infected with ZIKV^PRVABC59 (36 h, MOI 1) that were immunostained with anti-calnexin (green) and anti-J2 (red) for dsRNA, with the nuclei stained with Hoechst (blue). Scale bar, 10 µm. (F) Normalized luciferase expression of lysates from the expression of Huh7 cells transfected with the indicated siRNA and electroporated with a DENV¹⁶⁶⁸¹ subgenomic RNA replicon expressing Renilla luciferase harvested at the indicated time points. Treatment with MK0608 was as in panel A. For all panels, n = 3 biologically independent experiments, with bars indicating mean and error bars showing standard error of the mean. **P < 0.01, or not significant (ns), determined by two-way ANOVA with Dunnett’s multiple comparison test (B, D, and F).

Fig 4

The UFMylation E3 ligase complex regulates assembly of intracellular infectious virions. (A–C) Huh7 cells were treated with the indicated siRNA and infected with ZIKV^PRVABC59 for 48 h (MOI 0.1), and the following assays were performed: (A) focus-forming assay from supernatants to measure extracellular titer, (B) RNA copy number from supernatants, and (C) focus-forming assay from cellular lysates for intracellular titer. (D–F) Huh7 cells were treated with the indicated siRNA and infected with DENV^NGC for 48 h (MOI 0.1), and the following assays were performed: (D) focus-forming assay from supernatants for extracellular titer, (E) RNA copy number from supernatants, and (F) focus-forming assay from cellular lysates for intracellular titer. n = 3 (A–C) or n = 4 (D–F) biologically independent experiments, with bars indicating mean values and error bars showing standard errors of the mean. *P < 0.05, **P < 0.01, or not significant (ns), determined by paired t-test.

Fig 5

The UFMylation machinery promotes orthoflavivirus infection. (A) Immunoblot analysis of Huh7 cells after siRNA depletion of the indicated transcripts or non-targeting control (CTRL). (B) Cell viability measured after siRNA depletion of the indicated transcripts at 72 h post-transfection, as measured by Cell-Titer GLO assay, relative to the viability of siCTRL. (C and D) Focus-forming assay of supernatants harvested from Huh7 cells infected with DENV^NGC or ZIKV^PRVABC59 (48 h, MOI 0.1) after siRNA depletion of the indicated transcripts, shown as percentage of siCTRL. (E and F) Focus-forming assay of supernatants harvested from Huh7-UFM1 KO cells transduced with Flag-UFM1^WT or Flag-UFM1^ΔC3 and infected with either DENV^NGC (72 h, MOI 0.1) or ZIKV ^PRVABC59 (48 h, MOI 0.1), shown as percentage of Flag-UFM1^WT. Immunoblots indicate UFM1-conjugated proteins as those that are of higher molecular weight from unconjugated UFM1 but are detected with the anti-UFM1 antibody. For all panels, n = 3 biologically independent experiments, with bars indicating mean values and error bars showing standard errors of the mean. *P < 0.05, **P < 0.001, ***P < 0.001, ****P < 0.0001, or not significant (ns), determined by one-way ANOVA with Dunnett’s multiple comparison test (B, C, and D) or paired t-test (E and F).

Fig 6

UFL1 interacts with several DENV and ZIKV proteins. (A) Schematic of DENV polyprotein, showing membrane topology of viral proteins. (B and C) Immunoblot analysis of anti-V5 immunoprecipitated extracts and inputs, lysed in NP40 buffer (B) or TX-100-RIPA buffer (C), from Huh7 cells stably expressing Flag-UFL1 transfected with plasmids expressing V5-tagged DENV¹⁶⁶⁸¹ proteins. (D and E) Immunoblot analysis of anti-Flag immunoprecipitated extracts and inputs from DENV^NGC-infected or ZIKV^PRVABC59-infected (48 h, MOI 1) Huh7 cells stably expressing Flag-UFL1 or vector. (F) Immunoblot analysis of anti-Flag immunoprecipitated extracts and inputs from Huh7 cells transfected with DNA plasmids encoding the plasmid-launched ZIKV^MR766-WT or pZIKV ^{MR766-Flag-NS2A} and harvested at 72 hpi. Representative immunoblots from n = 3 biologically independent experiments are shown. (G) Model depicting UFL1 interactions wtih viral proteins.