An E2F1-mediated DNA damage response contributes to the replication of human cytomegalovirus

Abstract

DNA damage resulting from intrinsic or extrinsic sources activates DNA damage responses (DDRs) centered on protein kinase signaling cascades. The usual consequences of inducing DDRs include the activation of cell cycle checkpoints together with repair of the damaged DNA or induction of apoptosis. Many DNA viruses elicit host DDRs during infection and some viruses require the DDR for efficient replication. However, the mechanism by which DDRs are activated by viral infection is poorly understood. Human cytomegalovirus (HCMV) infection induces a DDR centered on the activation of ataxia telangiectasia mutated (ATM) protein kinase. Here we show that HCMV replication is compromised in cells with inactivated or depleted ATM and that ATM is essential for the host DDR early during infection. Likewise, a downstream target of ATM phosphorylation, H2AX, also contributes to viral replication. The ATM-dependent DDR is detected as discrete, nuclear γH2AX foci early in infection and can be activated by IE proteins. By 24 hpi, γH2AX is observed primarily in HCMV DNA replication compartments. We identified a role for the E2F1 transcription factor in mediating this DDR and viral replication. E2F1, but not E2F2 or E2F3, promotes the accumulation of γH2AX during HCMV infection or IE protein expression. Moreover, E2F1 expression, but not the expression of E2F2 or E2F3, is required for efficient HCMV replication. These results reveal a novel role for E2F1 in mediating an ATM-dependent DDR that contributes to viral replication. Given that E2F activity is often deregulated by infection with DNA viruses, these observations raise the possibility that an E2F1-mediated mechanism of DDR activation may be conserved among DNA viruses.

Figure 1. ATM is required for HCMV replication.

(A) Caffeine blocks HCMV replication. HEL fibroblasts were infected at MOIs of 0.3 or 1.0 as noted in the figure. Cells were treated with 10 mM caffeine following virus absorption and the drug was replenished every 24 h. Cell supernatants were assayed for infectious virus production by plaque assay. The mean values are shown with bars denoting standard error for three independent experiments. (B) ATM is required for HCMV infection. Normal (Con) and AT dermal fibroblasts were infected at an MOI of 0.3, 1.0 or 3.0. Cell supernatants were assayed for infectious virus production by plaque assay. The mean values are shown with bars denoting standard error for three independent experiments. (C) Viral protein expression is altered in fibroblasts lacking ATM. Immunoblot analyses for IE (IE1/IE2), E (pp65) and L (gB55) HCMV protein expression in normal (Con) and AT dermal fibroblasts. (D) ATM depletion compromises HCMV replication. HEL fibroblasts were transfected with siRNAs specific for ATM (siATMa or siATMc) or with a control siRNA (NS) 24 h prior to infection with HCMV at an MOI of 0.1. Cell supernatants were assayed for infectious virus production by plaque assay. Note that siATMa did not deplete ATM. The mean values are shown with bars denoting standard error for three independent experiments. (E) Transient depletion of ATM alters viral expression. HEL fibroblasts were transfected with the indicated siRNA, and infected with HCMV at an MOI of 0.1. The levels of ATM and viral IE, E and L protein expression were assessed by immunoblot analysis for ATM, IE1/IE2, pp65 and gB55, respectively.

Figure 2. Reduced formation of “mature” viral replication compartments (RCs) in AT fibroblasts and siATM-transfected HEL fibroblasts.

Normal (Con) and AT dermal fibroblasts were infected at an MOI of 0.3. HEL fibroblasts were transfected with control (NS) or siATMc, and subsequently infected with HCMV at an MOI of 0.1. Cells were fixed at 72 hpi and pUL44 detected by immunostaining. (A) Localization pUL44. Immunofluorescent images of normal (Con) and AT dermal fibroblasts infected with HCMV or HEL fibroblasts transfected with control (NS) or siATMc, subsequently infected with HCMV. Cells with “immature” RCs were defined as those with multiple, small pUL44 compartments (yellow arrows) and cells with “mature” RCs were identified as those composed of single, larger pUL44 compartments (white arrows). DAPI staining is used to define nuclei. (B) The percentage of fibroblasts with mature RCs was plotted relative to those lacking or having immature RCs. Over 200 cells were scored per sample.

Figure 3. Accumulation and localization of host damage response proteins varies during infection.

(A) γH2AX localization following HCMV infection. Immunofluorescent images of mock and virus-infected HEL fibroblasts (MOI = 1.0) are shown for IE and γH2AX localization. (B) Localization of HCMV IE and pUL44. Immunofluorescent images of mock and virus-infected HEL fibroblasts (MOI = 1.0) are shown for IE and pUL44 localization. (C) γH2AX localizes to HCMV replication compartments. Immunofluorescent detection of pUL44 and γH2AX in HCMV (MOI = 1.0) and mock-infected HEL fibroblasts is shown. (D) γH2AX accumulates during infection. HEL fibroblasts were infected with HCMV at the indicated MOI. Immunoblot analyses detected γH2AX protein in virus-infected fibroblasts. (E) ATM is required for γH2AX accumulation prior to the formation of viral DNA replication compartments. Immunofluorescent detection of γH2AX and phosphoserine 1981 ATM (p-ATM) or HCMV IE and p-ATM are shown in HCMV-infected cells fixed at 5 hpi and 48 hpi. Normal (Con) or AT dermal fibroblasts were infected with HCMV at an MOI of 5 and fixed at the indicated times pi. (A–C, E) DAPI staining is shown to identify nuclei.

Figure 4. H2AX contributes to HCMV replication.

(A) H2AX depletion compromises HCMV replication. HEL fibroblasts were transfected with siRNAs specific for H2AX (siH2AXa or siH2AXb) or with a control siRNA (NS) 24h prior to infection with HCMV at an MOI of 0.1. Cell supernatants were assayed for infectious virus production by plaque assay. (B) Depletion of H2AX alters viral protein accumulation. HEL fibroblasts were transfected with the indicated siRNA, and infected with HCMV at an MOI of 0.1. The levels of γH2AX and viral IE, E and L protein expression were assessed by immunoblot analysis for γH2AX, IE1/IE2, pp65, and gB55, respectively.

Figure 5. IE1 and IE2 expression induces a host DDR.

(A) γH2AX foci form within 5 h of HCMV infection in HEL fibroblasts (MOI = 5). Immunofluorescent detection of HCMV IE and γH2AX proteins is shown. DAPI staining identifies the cell nuclei. (B) Accumulation of γH2AX protein at 5 h following HCMV infection. Immunoblot detection of γH2AX protein at 5 hpi (MOI = 5) is shown. (C) IE1-72 protein accumulation following transduction with Ad-IE1 into HEL fibroblasts. Immunoblot analysis of major HCMV IE proteins from whole cell lysates of HEL fibroblasts infected with HCMV (MOI = 0.1, 120 hpi) or transduced with either Ad-IE1 or Ad β-gal (MOI = 250, 48 hpi). HCMV IE proteins were identified with a monoclonal antibody that detects both IE1-72 and IE2-86 proteins. (D) Accumulation of p-ATM in HEL fibroblasts transduced with Ad-IE1 or Ad-IE2. Immunofluorescent detection of HCMV IE and p-ATM proteins is shown. DAPI staining identifies the cell nuclei. (E) Quantitation of the cells positive for IE or IE plus p-ATM observed in (D). Histograms show the average of three independent experiments and the error bars denote the standard deviation. (F) IE1 expression leads to nuclear γH2AX accumulation. HEL fibroblasts were transduced with Ad-IE1 or Ad β-gal and fixed for immunofluorescent detection of γH2AX at the indicated times post transduction. DAPI staining identifies nuclei. (G) Accumulation of γH2AX protein following IE1-72 expression. Immunoblot analysis for γH2AX protein in lysates of HEL fibroblasts transduced with Ad-IE1 or Ad β-gal and harvested at the indicated times post transduction. (H) Accumulation of γH2AX protein following IE2-86 expression. Immunoblot analysis for γH2AX protein in lysates of HEL fibroblasts transduced with Ad-IE2 and harvested at the indicated times post transduction.

Figure 6. E2F1-specific contributes to the DDR associated with HCMV infection, IE1 or IE2 expression.

(A) E2F1 contributes to the DDR following HCMV infection. Quantitation of γH2AX-positive cells in mock and HCMV-infected cells. HEL fibroblasts were transfected with siRNAs specific for E2F1 (1A, 1C), E2F2 (2A, 2B), or E2F3 (3A, 3B) or with a control siRNA (Con) 24 h prior to infection with HCMV at an MOI of 1.0. At 24 hpi, cells were fixed and γH2AX was detected by immunofluorescent staining. *P<0.003. (B) E2F1 contributes to the DDR induced by IE1. Quantitation of γH2AX-positive cells following transduction with recombinant adenoviruses. HEL fibroblasts were transfected with siRNAs as described in (A) 24 h prior to infection with a recombinant adenovirus that encodes Ad-IE1. At 24 hpi, cells were fixed and γH2AX detected by immunofluorescent staining. Transduction of cells with a recombinant adenovirus that encodes HPV-16 E7 (Ad-E7, MOI = 250) was used as a positive control for activation of the host DDR. *P<0.007; **P<0.02; ***P<0.001. (C) E2F1 contributes to the DDR induced by IE2. Quantitation of γH2AX-positive cells following siRNA transfection and transduction with Ad-IE2. At 24 hpi or 48 hpi, HEL fibroblasts were fixed and γH2AX detected by immunofluorescent staining. ***P<0.001. (A–C) Cells containing >2 γH2AX foci were scored as positive and plotted. Histograms indicate the average of three independent experiments and the error bars denote the standard deviation. P values were determined by Student's t-test.

Figure 7. E2F1 is required for efficient virus replication.

(A) Production of progeny virus in HEL fibroblasts following transfection with siRNAs that deplete E2F1 levels (1A or 1C) or with a control siRNA (NS) 24 h prior to HCMV infection (MOI = 0.1). Culture supernatants from infected cells were assayed for infectious virus production by plaque assay. (B) Expression of E2F1 and HCMV proteins during infection. E2F1 and markers of viral IE, E, and L protein expression were detected by immunoblotting of cells lysates from (A). (C) Production of progeny virus in HEL fibroblasts following transfection with siRNAs that deplete E2F2 levels (E2A or E2B) or with a control siRNA (NS) 24 h prior to HCMV infection (MOI = 0.1). Samples processed as described in (A). (D) Production of progeny virus in HEL fibroblasts following transfection with siRNAs that deplete the levels of E2F3a (E3a), E2F3b (E3b), or both E2F3a and E2F3b (E3a+b), or with a control siRNA (NS) 24 h prior to HCMV infection (MOI = 0.1). Samples processed as described in (A).

Figure 8. Model showing the relationship between infection, E2F1, ATM activation and HCMV replication.