Nuclear domain 10 components promyelocytic leukemia protein and hDaxx independently contribute to an intrinsic antiviral defense against human cytomegalovirus infection

Abstract

Infection with DNA viruses commonly results in the association of viral genomes with a cellular subnuclear structure known as nuclear domain 10 (ND10). Recent studies demonstrated that individual ND10 components, like hDaxx or promyelocytic leukemia protein (PML), mediate an intrinsic immune response against human cytomegalovirus (HCMV) infection, strengthening the assumption that ND10 components are part of a cellular antiviral defense mechanism. In order to further define the role of hDaxx and PML for HCMV replication, we generated either primary human fibroblasts with a stable, individual knockdown of PML or hDaxx (PML-kd and hDaxx-kd, respectively) or cells exhibiting a double knockdown. Comparative analysis of HCMV replication in PML-kd or hDaxx-kd cells revealed that immediate-early (IE) gene expression increased to a similar extent, regardless of which ND10 constituent was depleted. Since a loss of PML, the defining component of ND10, results in a dispersal of the entire nuclear substructure, the increased replication efficacy of HCMV in PML-kd cells could be a consequence of the dissociation of the repressor protein hDaxx from its optimal subnuclear localization. However, experiments using three different recombinant HCMVs revealed a differential growth complementation in PML-kd versus hDaxx-kd cells, strongly arguing for an independent involvement in suppressing HCMV replication. Furthermore, infection experiments using double-knockdown cells devoid of both PML and hDaxx illustrated an additional enhancement in the replication efficacy of HCMV compared to the single-knockdown cells. Taken together, our data indicate that both proteins, PML and hDaxx, mediate an intrinsic immune response against HCMV infection by contributing independently to the silencing of HCMV IE gene expression.

FIG. 1.

Detection of endogenous ND10 proteins hDaxx and PML by Western blotting using cell lysates derived from primary human fibroblasts stably transduced with the respective shRNA expression vectors as indicated. The upper panel shows levels of hDaxx protein detected with the anti-hDaxx rabbit polyclonal antiserum M-117 in various cell types, as indicated above the lanes. The middle panel shows detection of the various SUMOylated as well as non-SUMOylated isoforms of PML using the monoclonal anti-PML antibody 5E10. The major PML isoform is indicated by an arrow. In the lower panel, β-actin detection served as a loading control.

FIG. 2.

Enhanced replication of HCMV in PML-kd as well as hDaxx-kd cells. Depletion of PML or hDaxx from primary human fibroblasts facilitates the initiation of HCMV gene expression, thus leading to a higher number of infected cells. (A) Quantification of the number of plaques by standard plaque assay 7 days postinfection of retrovirally transduced HFFs (vector, siC, siPML2, and siDaxx1) with 100 IEU of HCMV AD169. (B and C) Enhancement of HCMV IE gene expression in the absence of PML or hDaxx. In panel B, retrovirally transduced HFFs as indicated were infected with 50 IEU/well of HCMV AD169. Cells were fixed at 24 hpi followed by determination of the number of IE1-expressing fibroblasts via indirect immunofluorescence. Panel C shows Western blot analysis of IE1 gene expression 8 h after infection of PML- and hDaxx-negative cells and control fibroblasts with HCMV AD169 (MOI of 0.01).

FIG. 3.

Reconstitution of hDaxx expression in hDaxx-kd cells. (A) Detection of hDaxx protein levels by Western blotting after transduction of siDaxx1 fibroblasts or normal HFFs with the lentiviral Flag-Daxx expression vector pHM2499 resulting in siDaxx1-R and HFF+F-Daxx cells, respectively. hDaxx expression was assayed using an anti-hDaxx monoclonal rabbit antibody (lanes 1 to 4) as well as the anti-Flag monoclonal antibody M2 (lanes 5 to 8). Actin was included as an internal loading control. (B) Examination of HCMV IE gene expression after infection of cells with either a knockdown of hDaxx (siDaxx1) or after introduction of Flag-Daxx expression in siDaxx1-R and HFF+F-Daxx cells. Flag-Daxx-transduced cells (siDaxx1-R and HFF+F-Daxx), hDaxx-negative fibroblasts (siDaxx1), and control cells (vector and siC) were infected in parallel with 25 IEU per well of HCMV AD169. At 24 hpi the cells were stained with an anti-IE1 antibody (monoclonal antibody p63-27) followed by the quantification of IE1-positive fibroblasts. Error bars indicate standard deviations derived from three independent experiments.

FIG. 4.

Examination of hDaxx protein levels in wt AD169- and AD169/del-pp71-infected PML-positive and -negative HFFs. (A) HFFs were either mock infected or infected with wt HCMV AD169 (MOI of 2). Lysates were harvested at the indicated times postinfection (1 to 12 hpi) and analyzed by Western blotting for the expression of hDaxx, pp71, IE1, and actin (actin was included as an internal loading control). (B) Investigation of hDaxx levels after infection with the pp71 deletion virus AD169/del-pp71. HFFs were left uninfected (mock) or were infected with AD169/del-pp71 (MOI of 2). At the indicated hours postinfection cell lysates were sampled and examined for the presence of hDaxx, pp65, or actin via immunoblotting. (C) PML-kd and control HFFs (vector) were infected in parallel with wt AD169. Cell lysates were harvested directly after infection, as indicated, followed by the detection of hDaxx, pp65, or actin protein levels via Western blot analysis.

FIG. 5.

Analysis of IE gene expression after infection of knockdown (siPML2 and siDaxx) and control cells (vector and siC) with wt AD169 or the pp71 deletion virus AD169/del-pp71. (A) Retrovirally transduced HFFs as indicated were infected with 200 IEU/well of AD169 and 500 IEU/well of AD169/del-pp71, respectively. At 24 hpi the cells were immunostained for the IE1 protein, and the number of IE1-expressing cells was determined. Standard deviations derived from three independent experiments are illustrated by error bars. (B) Western blot experiments for comparison of IE1 gene expression in PML- and hDaxx-deficient cells and control fibroblasts. The indicated cells were infected in parallel with AD169 or AD169/del-pp71 at an MOI of 0.01. At 24 hpi lysates were harvested and assayed for IE1 expression. The detection of actin was used as a loading control.

FIG. 6.

Comparison of IE expression levels in PML-kd and hDaxx-kd and control fibroblasts infected with wt HCMV and the IE1-mutant virus CR208, respectively. siPML2, siDaxx1, and control cells (vector and siC) were infected with either 40 IEU/well of wt HCMV (A) or CR208 (B). At 24 hpi the number of IE-positive cells was determined by fluorescence microscopy using either monoclonal antibody p63-27 against IE1 (A) or antiserum anti-pHM178 against IE2 (B). Error bars represent standard deviations from three different experiments.

FIG. 7.

Investigation of AD169/del-MIEP infection in normal HFFs as well as in cells exhibiting a knockdown of PML or hDaxx. (A) Quantitative real-time PCR for evaluation of the viral load after infection of HFFs with wt HCMV AD169 and AD169/del-MIEP, respectively. HFF cells were infected with 25 IEU of AD169 or AD169/del-MIEP. Then, DNA was extracted at 14 hpi, and the number of HCMV genome copies was determined by real-time PCR. Amplification of the cellular albumin gene was utilized as a standard for calculation of cell numbers. (B, C, and D) Comparative analysis of HCMV IE gene expression following infection of knockdown and control cells with AD169/del-MIEP. PML-kd and hDaxx-kd cells together with control fibroblasts were infected in parallel with 25 IEU per well of AD169 and AD169/del-MIEP, respectively. At 24 hpi the amount of IE1-positive cells was evaluated by indirect immunofluorescence analysis (B). HFF cells transduced with the respective siRNA expression vectors as indicated were infected with AD169/del-MIEP at an MOI of 0.1. Cell lysates were prepared at 24 hpi and assayed for IE1 (C) as well as IE2 (D) expression by Western blotting. Detection of actin protein levels served as an internal loading control.

FIG. 8.

Verification of the generation of double-knockdown HFFs (siDaxx1+siPML2) devoid of both ND10 constituents, PML and hDaxx, followed by the analysis of HCMV replication in the constructed double-knockdown cells compared to single-knockdown fibroblasts. (A) Western blotting for detection of hDaxx and PML expression in the generated double-knockdown cells, siDaxx1+siPML2. Comparison of PML (upper panel), hDaxx (middle panel), and actin protein levels (lower panel) in whole-cell extracts from double-knockdown (siDaxx1+siPML2), single-knockdown (siPML2, siDaxx1), and control HFFs (vector and siC). (B) Analysis of HCMV IE gene expression after infection of double-knockdown cells in comparison to single-knockdown fibroblasts. The different cell populations as indicated were grown on coverslips in six-well dishes and infected with 100 IEU/well of wt HCMV AD169. At 24 hpi the amount of IE1-expressing cells was assessed via indirect immunofluorescence analysis. Error bars illustrate the standard deviation of three independent determinations.