The E3 ubiquitin ligase MARCHF8 restricts HSV-1 infection by inhibiting replication of the viral genome

Abstract

Host cell restriction factors are intracellular proteins that target and inhibit virus replication. Membrane-associated RING-CH finger (MARCHF) proteins are a family of intracellular E3-ubiquitin ligases and some, including MARCHF8, have been implicated in restricting the replication of diverse RNA viruses. However, little is currently known as to whether MARCHF proteins mediate antiviral activity against DNA viruses. Herein, we used a doxycycline-inducible overexpression system to demonstrate that human MARCHF1 and MARCHF8 potently restrict productive HSV-1 replication and assessed MARCHF8-mediated restriction of HSV-1 in detail. A functional RING-CH domain and a tyrosine-based motif, located in the N- and C-terminal cytoplasmic tail of MARCHF8, respectively, were required for HSV-1 restriction. For many RNA viruses, MARCHF8-mediated restriction has been associated with downregulation of viral envelope glycoproteins from the cell surface, thereby limiting their subsequent incorporation into nascent virions. However, while MARCHF8 expression did not affect virus entry, translocation of viral genome to the nucleus and immediate early (IE) gene expression, it did inhibit HSV-1 genomic replication, and therefore subsequent late gene expression and release of infectious virions. MARCHF8-mediated restriction of HSV-1 occurred independent of other cellular factors known to impact genomic replication of HSV-1, namely SAMHD1 and CD81, and could also proceed efficiently in cells expressing a functional cGAS-STING pathway. To our knowledge, these studies are the first to demonstrate MARCHF8-mediated restriction against a human DNA virus. Moreover, inhibition of HSV-1 genomic replication represents a novel mechanism of MARCHF8-mediated virus restriction that is distinct to its reported antiviral activity against different RNA viruses.

Keywords: E3 ubiquitin ligase; MARCHF8 protein; herpesvirus; innate immunity; viral immunology; virology; virus restriction.

Figure 1

Inducible expression of MARCHF8 in human epithelial cells inhibits productive replication of HSV-1. A, 293T cells with doxycycline (DOX)-inducible expression of FLAG-tagged MARCHF1 (M1) or MARCHF8 (M8), or of an irrelevant intracellular control protein lacking a FLAG tag (CTRL) were cultured in the presence (Dox) or absence (No Dox) of 1 μg/ml Dox for 24 h (solid line), or in the presence of Dox for 20 h before addition of MG132 and chloroquine (MG132+CQ) for the last 4 h. Cells were then fixed and stained for intracellular expression of FLAG-tagged proteins and analysed by flow cytometry. Representative histograms are shown. B and C, 293T cells with Dox-inducible protein expression were incubated in the presence (Dox, grey bars) or absence (No Dox, white bars) of Dox for 24 h and then (B) infected with the KOS strain of HSV-1 (MOI 0.1) and supernatants collected at 2, 24 and 48 h post-infection (hpi), or (C) infected with KOS strain of HSV-1 at different MOIs (0.01, 0.1 and 1) and supernatants collected at 48 hpi. Virus titres in cell-free supernatants were determined by plaque assay on Vero cells. Representative technical replicates from one of B) n = 2 or C) n = 3 independent experiments shown. D, 293T cells with Dox-inducible protein expression were incubated with 1ug/ml Dox for 24 h before infection (Before), both before and after infection (Before + After) or Dox was added only after infection (After). Cells were infected with HSV-1 (MOI 0.1) and virus titres in clarified supernatants collected at 48 hpi were determined by plaque assay. Representative technical replicates from one of n = 3 independent experiments are shown. E, BEAS2B or HeLa cells with Dox-inducible expression of FLAG-tagged M1 or M8 were cultured in the presence (Dox) or absence (No Dox) of 1 ug/ml Dox for 24 h (solid line), or in the presence of Dox for 20 h before addition of MG132 and chloroquine (MG132+CQ) for the last 4 h. Cells were then fixed and stained for intracellular expression of FLAG-tagged proteins and analysed by flow cytometry. Representative histograms are shown. F, BEAS2-B or HeLa cells with Dox-inducible protein expression were incubated in the presence (Dox, grey bars) or absence (No Dox, white bars) of Dox for 24 h, infected with HSV-1 (MOI = 0.1) and virus titres in clarified supernatants determined at 48 hpi. Technical replicates from one of two independent experiments are shown. Limit of detection for plaque assay results are shown as a horizontal line. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments, as described in Experimental procedures. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 2

A functional E3 ligase domain is required for antiviral activity against a-herpesviruses.A and B, 293T cells with Dox-inducible expression of FLAG-tagged parental M8 or E3 ligase mutants (M8 CS or M8 W114A) were cultured in the presence or absence of Dox for 24 h or in Dox for 20 h before addition of MG132+CQ for the last 4 h. Dox-inducible protein expression was then determined (A) by flow cytometry using an anti-FLAG mAb to stain for intracellular protein expression, or (B) by Western blot using a M8-specific polyclonal Ab, with a Calnexin loading control. C and D, 293T cells with Dox-inducible protein expression were cultured in the presence (Dox) or absence (No Dox) of Dox and then infected with (C) HSV-1 KOS (MOI 0.1), or (D) HSV-1 strains F or SC-16, HSV-2 strain 186 or VACV-OVA (all at MOI 0.1). Virus titres in clarified supernatants collected at 48 hpi were determined by plaque assay. Technical replicates from one of three independent experiments are shown. Limit of detection for plaque assay results are shown as a horizontal line. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments, as described in Experimental procedures. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 3

Structural characteristics of MARCHF8 and MARCHF1 that contribute to their ability to inhibit productive replication of HSV-1.A, schematic showing the domain structure of M8 and highlight M8 mutants utilized in this study. B and C, stable 293T cell lines with DOX-inducible expression of parental M8 or M1 or (B) M8 C-CT tyrosine motif mutants M8-²²²Axx^L225 or M8-²³²AxxV²³⁵, (C) M8 with N-CT deletion (M8_ΔN-CT) or substitution for the N-CT of M1 (M8_M1 N-CT), or M1 with N-CT deletion (M1_ΔN-CT) or substitution for the N-CT of M8 (M1_M8 N-CT) were generated and characterized previously (25). D/E, schematic of M1 isoform 2 (M1) showing specific lysine (K) residues in the N-CT, as well as M1 N-CT deletion mutants lacking the entire N-CT (M1_ΔN-CT) or lacking 15 residues from the N-CT of M1.2 (M1_Δ15). 293T cells with DOX-inducible expression of M1, M1 isoform 1 (M1.1) and M1_Δ15 were generated and characterized previously (10). B, C, and E, cells were cultured in the presence (Dox) or absence (No Dox) of 1 μg/ml Dox and then infected with HSV-1 KOS (MOI 0.1). Virus titres in clarified supernatants collected at 48 hpi were determined by plaque assay. Technical replicates from 1 of (A) n = 3, (B) n = 5, and (C) n = 5 independent experiments are shown. Limit of detection for plaque assay results are shown as a horizontal line. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments as described in Experimental Procedures. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 4

MARCHF8 does not inhibit HSV-1 entry and translocation to the nucleus but does inhibit HSV-1 genomic replication and late viral gene expression.A, 293T cells with Dox-inducible expression of FLAG-tagged M8 or M8-CS were cultured in the presence (Dox) or absence (No Dox) of 1ug/ml Dox for 24 h (solid line) and then infected with i) HSV-1 KOS-TK-GFP (MOI 2) or ii) HSV-1 KOS (MOI 4) and analysed at 8 hpi. (i) Cells were fixed and the percentage of GFP⁺ cells was determined by flow cytometry. (ii) Cells were fixed, permeabilized and stained with a mAb to the immediate early HSV-1 protein ICP4, followed by Alexa Fluor 488-conjugated chicken anti-mouse Ig (right panel). Representative technical replicates from (A) n = 3 and (B) n = 2 experiments are shown. B, following 24 h culture in the presence or absence of Dox, cells were infected with a double fluorescent HSV-1 virus (DF-HSV-1), expressing GFP and RFP under control of the HSV-1 gB or gC promoters, respectively. At either 8 (MOI = 2) or 18 (MOI = 5) hpi, cells were fixed and EGFP and RFP expression analysed by flow cytometry. Technical replicates from one of three independent experiments are shown. C, (i) A549 cells were transfected with 10 μM of M8-specific or NTC siRNA for 48 h, infected with DF-HSV-1 (MOI 1) and levels of GFP or RFP expression determined by flow cytometry at 18 hpi. Technical replicates from one of four independent experiments are shown. (ii) A549 cells were transfected with 10 μM of M8-specific or NTC siRNA for 48 h and then infected with HSV-1 KOS (MOI 0.001). Virus titres in clarified supernatants collected at 36 hpi were determined by plaque assay. Technical replicates from one of four independent experiments are shown. Limit of detection for plaque assay results are shown as a horizontal line. (iii) Pooled data of log₂ fold change in viral titres from average NTC titres across four experiments are shown, different symbol used for each experiment. D, 293T cells with Dox-inducible protein expression cultured for 24 h in the presence or absence of Dox were infected with HSV-1 (MOI 0.1). At 2 and 18 hpi, total cellular DNA was extracted for real-time qPCR. HSV-1 DNA copy number was determined relative to a standard curve generated using a plasmid expressing the HSV-1 TK gene. Technical replicates from one of five independent experiments are shown and the horizontal line indicates the limit of detection. E, infectivity per virion was determined as the ratio between the titre of infectious virus (Log₁₀ PFU/ml) and the copy number of the Tk gene (Log₁₀DNA copies/ml) from each sample. Technical replicates from one of three independent experiments are shown. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 5

Inducible expression of MARCHF8 in human macrophage-like THP-1 cells inhibits productive replication of HSV-1.A and B, PMA-differentiated THP-1 cells with Dox-inducible expression of FLAG-tagged M8 or M8-CS were cultured in No Dox, in Dox for 24 h or in Dox for 20 h before addition of MG132+CQ for the last 4 h. Dox-inducible protein expression was then determined (A) by flow cytometry using an anti-FLAG mAb to stain for intracellular protein expression, or (B) by Western blot using a M8-specific polyclonal Ab, with a Calnexin loading control. C–F, PMA-differentiated THP-1 cells were cultured 24 h in the presence (Dox) or absence (No Dox) of 1 μg/ml Dox. C, cells were collected and levels of cell surface CD86 were determined by flow cytometry (i) Representative histograms and (ii) geometric mean fluorescence intensity (gMFI) of CD86 from one of three independent experiments are shown. D, cells were then infected with HSV-1 KOS or HSV-2186 (both at MOI 5) and virus titres in clarified supernatants collected at 2 and 48 hpi were determined by plaque assay. Technical replicates from one of 5 (HSV-1) or one of 2 (HSV-2) independent experiments are shown. E, cells were then infected with HSV-1 KOS (MOI 10) and at 5 hpi cells were fixed, permeabilized and stained with a mAb to the immediate early HSV-1 protein ICP4, followed by Alexa Fluor 488-conjugated chicken anti-mouse Ig (right panel) and analysed by flow cytometry. Technical replicates from one of two independent experiments are shown. F, cells were then infected with HSV-1 KOS (MOI 1). At 2 and 18 hpi, total cellular DNA was extracted for real-time qPCR. (i) HSV-1 DNA copy number was determined relative to a standard curve generated using a plasmid expressing the HSV-1 TK gene. Technical replicates from one of three independent experiments are shown and the horizontal line indicates the limit of detection. (ii) Infectivity per virion was determined as the ratio between the titre of infectious virus (Log₁₀ PFU/ml) and the copy number of the TK gene (Log₁₀DNA copies/ml) from each sample. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 6

Cellular CD81 does not contribute to M8-mediated restriction of HSV-1 replication.A, 293T cells with Dox-inducible expression of M8 or M8-CS were treated with CD81-specific or NTC siRNA and, 24 h later, incubated with (Dox) or without (No Dox) one ug/ml DOX for 24 h. Cells were then washed and (i) cell surface CD81 expression was determined by flow cytometry, or (ii) infected with HSV-1 (MOI 0.1) and virus titres in cell-free supernatants collected at 48 hpi were determined by plaque assay. Technical replicates from one of two independent experiments are shown. B, parental 293T and THP-1 cells with Dox-inducible expression of M8 or M8-CS were treated with CD81-specific guide RNA to generate CD81 KO cells. Flow cytometry was used to confirm KO of cell surface CD81 in (i) 293T or (ii) THP-1, with minimal effects on expression of the tetraspanins CD9, CD62, and CD151. C, parental or CD81 KO (i) 293T or (ii) THP-1 cells cultured for 24 h in the presence (+Dox) or absence (−Dox) of 1 μg/ml Dox were lysed and analysed by Western blot using a polyclonal Ab to detect CD81 or M8, or with a Calnexin loading control. D and E, parental or CD81 KO (i) 293T or (ii) THP-1 cells with Dox-inducible expression of M8 or M8-CS were cultured for 24 h in the presence or absence of 1 μg/ml Dox. D, cells were infected with HSV-1 KOS (MOI 4 and 10 for 293T and THP-1, respectively) and, at 5 hpi, cells were fixed, permeabilized and stained and stained with a mAb to the immediate early HSV-1 protein ICP4, followed by Alexa Fluor 488-conjugated chicken anti-mouse Ig. Technical replicates from one of two independent experiments are shown. E, cells were infected with HSV-1 KOS (MOI 0.1 and five for 293T and THP-1, respectively) and virus titres were determined in cell-free supernatants collected at 48 hpi. Technical replicates from one of 3 (293T) or one of 2 (THP-1) independent experiments are shown. The limit of detection is indicated as a horizontal line. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments as described in Experimental procedures. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 7

Cellular SAMHD1 does not contribute to MARCHF8-mediated restriction of HSV-1 replication.A, THP-1 and Jurkat cells cultured 24 h in the presence (+) or absence (−) of 1000 mU/ml of recombinant human IFN-α were analyzed by Western blot with an anti-SAMHD1 mAb, and with a Calnexin loading control. B, Jurkat cells with Dox-inducible expression of FLAG-tagged M8 or M8-CS were cultured in No Dox, in Dox for 24 h or in Dox for 20 h before addition of MG132+CQ for the last 4 h. Dox-inducible protein expression was then determined (i) by flow cytometry using an anti-FLAG mAb to stain for intracellular protein expression, or (ii) by Western blot using a M8-specific polyclonal Ab, with a Calnexin loading control. C, Jurkat cells with Dox-inducible expression of M8 or M8-CS cultured for 24 h in the presence (Dox) or absence (No Dox) of 1 μg/ml Dox were then infected with HSV-1 KOS (MOI 0.1) and virus titres in cell-free supernatants collected at 48 hpi were determined by plaque assay. Technical replicates from one of three independent experiments are shown. D–F, PMA-differentiated parental (THP-1) and SAMHD1 knockout (SAMHD1 KO) THP-1 cells with inducible expression of M8 or M8-CS were cultured for 24 h in the presence (+Dox) or absence (-Dox) of 1 μg/ml Dox. D, cells were lysed and analysed by Western blot using a mAb to detect SAMHD1 or a polyclonal Ab to detect M8, or with a Calnexin loading control. E, cells were infected with HSV-1 KOS (MOI 10) and at 5 hpi, cells were fixed, permeabilized and stained and stained with a mAb to the immediate early HSV-1 protein ICP4, followed by Alexa Fluor 488-conjugated chicken anti-mouse Ig. Technical replicates from one of two independent experiments are shown. F, cells were infected with HSV-1 KOS (MOI 5), and virus titres in cell-free supernatant collected at 48 hpi were determined by plaque assay. Technical replicates from one of four independent experiments are shown. The horizontal line represents the limit of detection. Statistical analysis was done using a mixed-effects model utilizing data points from all experiments. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.

Figure 8

MARCHF8-mediated restriction of HSV-1 in THP-1 cells proceeds in the presence of an intact cGAS signaling pathway.A, PMA-differentiated parental or cGAS KO THP-1 cells were transfected with 5 ug/ml HT-DNA for 8 h, then lysed for isolation of RNA. qPCR was then used to determine induction of IFN-α or IFN-β relative to untransfected cells. Technical replicates from one of two independent experiments are shown. B, parental or cGAS KO with Dox-inducible expression of M8 or M8-CS were generated. Cells cultured for 24 h in the presence (+Dox) or absence (−Dox) of 1 μg/ml Dox, were then lysed and analysed by Western blot using a rabbit mAb to detect cGAS or a rabbit polyclonal Ab to M8, or with a Calnexin loading control. C and D, parental or cGAS KO PMA-differentiated THP-1 cells with Dox-inducible expression of M8 or M8-CS were cultured for 24 h in the presence or absence of 1 μg/ml Dox. C, cells were infected with HSV-1 KOS (MOI 10) and, at 5 hpi, cells were fixed, permeabilized and stained with a mAb to the immediate early HSV-1 protein ICP4, followed by Alexa Fluor 488-conjugated chicken anti-mouse Ig. Technical replicates from one of two independent experiments are shown. D, cells were infected with HSV-1 (MOI 5) and virus titres in cell-free supernatants collected 48 hpi were determined by plaque assay. Technical replicates from one of three independent experiments are shown. Limit of detection is indicated by a horizontal line. Statistical analysis was performed using a mixed effects model utilizing data points from all experiments as described in Experimental procedures. ∗p < 0.01, ∗∗p < 0.001, ∗∗∗p < 0.0001 and ns = not significant.