Endogenous MOV10 inhibits the retrotransposition of endogenous retroelements but not the replication of exogenous retroviruses

Abstract

Background: The identification of cellular factors that regulate the replication of exogenous viruses and endogenous mobile elements provides fundamental understanding of host-pathogen relationships. MOV10 is a superfamily 1 putative RNA helicase that controls the replication of several RNA viruses and whose homologs are necessary for the repression of endogenous mobile elements. Here, we employ both ectopic expression and gene knockdown approaches to analyse the role of human MOV10 in the replication of a panel of exogenous retroviruses and endogenous retroelements.

Results: MOV10 overexpression substantially decreased the production of infectious retrovirus particles, as well the propagation of LTR and non-LTR endogenous retroelements. Most significantly, RNAi-mediated silencing of endogenous MOV10 enhanced the replication of both LTR and non-LTR endogenous retroelements, but not the production of infectious retrovirus particles demonstrating that natural levels of MOV10 suppress retrotransposition, but have no impact on infection by exogenous retroviruses. Furthermore, functional studies showed that MOV10 is not necessary for miRNA or siRNA-mediated mRNA silencing.

Conclusions: We have identified novel specificity for human MOV10 in the control of retroelement replication and hypothesise that MOV10 may be a component of a cellular pathway or process that selectively regulates the replication of endogenous retroelements in somatic cells.

Figure 1

MOV10 overexpression restricts the production of infectious retrovirus particles for a broad range of exogenous retroviruses. (A) MOV10 overexpression decreases HIV-1 virus production. HeLa or 293T cells were co- transfected with pHIV-1_NL4-3 and increasing amounts of pT7-MOV10 as indicated or the pT7-Luc control. Virus concentration in the medium was determined by a p24^Gag ELISA. Cell lysates were analysed by quantitative immunoblotting with anti-T7, anti- p24^Gag and anti-Hsp90 antibodies. (B) Overexpression of MOV10 inhibits the infectivity of HIV-1 virions. The TZM-bl reporter cell line expressing a HIV-1 Tat inducible β-gal reporter gene was infected with equal amounts of virus normalised by the p24^Gag concentration from each of the indicated samples. Cells were lysed and β-gal activity was measured to determine virus infectivity. (C) MOV10 overexpression inhibits the production of infectious SIVmac, MLV and M-PMV particles. For HIV-1 and SIVmac lentiviral vector production, 293T cells were co-transfected with p8.91, pCSGW and pVSV-G, or pSIV3+, pSIV-RMES4 and pVSV-G, respectively, together with pT7-MOV10 or pT7-Luc. 293T cells were infected with lentiviral particles and infectivity was determined by measuring the percentage of GFP-positive 293T cells by FACS. For MLV and M-PMV virion production, 293T cells were co-transfected with pNCS/FLAG, pMSCV/Tat and pVSV-G, or pMTΔE and pVSV-G, respectively, together with pT7-MOV10 or pT7-Luc. Infectivity was determined using TZM-bl cells. (D) Overexpression of MOV10 decreases the production of M-PMV virions. Cell lysates and sucrose cushion purified M-PMV virions were analysed by immunoblotting with anti p27^Gag and anti-Hsp90 antibodies (* refers to non-specific bands). For (A), (B) and (C) results are normalised to the pLuc control, which is set at 100%. For (C) a single control bar set at 100% is graphed for simplicity. Values are the mean ± SD of 3 independent experiments.

Figure 2

Overexpression of MOV10 suppresses the retrotransposition of both LTR and non-LTR endogenous retroelements. (A) MOV10 overexpression restricts the replication of LINE-1, Alu and IAP. HeLa cells were co-transfected with pLINE-1 (pJM101/L1.3), pAlu (pAlu-neo^Tet plus pCEP-ORF2), or pIAP (pGL3- IAP92L23neo^TNF) together with pT7-MOV10 or pT7-Luc. Cells were selected with G418 for 12-14 days to measure retrotransposition frequency and then fixed and stained with Giemsa. (B) MOV10 overexpression has no affect on neo expression or selection. HeLa cells were co-transfected with a pcDNA3.1 control vector containing a neomycin resistance cassette (pcDNA3.1-neo) together with pT7-MOV10 or pT7-Luc. The cells were G418 selected and the colonies were quantified as described in panel (A). Results are normalised to the pLuc control, which is set at 100%. For (A) a single control bar set at 100% is graphed for simplicity. Values are the mean ± SD of 3 independent experiments.

Figure 3

Silencing endogenous MOV10 has no significant effect on the production of infectious retrovirus particles for a panel of exogenous retroviruses. (A) Depletion of endogenous MOV10 has no effect on HIV-1 virus production. Stable MOV10 KD cells were produced by transducing HeLa or 293T cells with lentiviral vectors expressing either a non-silencing control shRNA or a MOV10-specific shRNA. HeLa or 293T non-silencing control and MOV10 KD cells were infected with VSV-G pseudotyped HIV-1_NL4-3. Virus concentration in the medium was determined as described in Figure 1A. Cell lysates were analysed by immunoblotting with anti-p24^Gag, anti-Hsp90 or anti-MOV10 antibodies, the latter of which was used to verify the MOV10 KD. (HeLa virus production p = 0.0611, 293T virus production p = 0.2007). (B) Silencing of endogenous MOV10 has no effect on HIV-1 virion infectivity. Virion infectivity was determined as described in Figure 1B. (HeLa infectivity p = 0.3080, 293T infectivity p = 0.4812). (C) Depleting endogenous MOV10 has no effect on spreading HIV-1 replication. Hut78 non-silencing control or MOV10 KD cells were infected with equal amounts of HIV-1_NL4-3 and passaged every 2 days. Medium was harvested on days 2, 4, 6 and 8 and virus production was determined as described in Figure 1A. Cell lysates were analysed by immunoblotting with anti-MOV10 and anti-Hsp90 antibodies. (D) MOV10 silencing has no effect on the production of infectious SIVmac, MLV and M-PMV particles. 293T non-silencing control or MOV10 KD stable cells were transfected as described in Figure 1C for the production of HIV-1, SIVmac, MLV and M-PMV particles. Infectivity was determined as described in Figure 1C. (HIV-1 p = 0.1358, SIVmac p = 0.1040, MLV p = 0.4907, M- PMV p = 0.4919). For (A), (B) and (D) results are normalised to the non-silencing control, which is set at 100%. For (D) a single control bar set at 100% is graphed for simplicity. Values are the mean ± SD of 7 independent experiments for (A) and (B) or 3 independent experiments for (D). The data were analysed with an unpaired one- tailed t test.

Figure 4

Depletion of endogenous MOV10 significantly enhances the retrotransposition of endogenous retroelements. (A) Silencing of endogenous MOV10 significantly enhances the retrotransposition of LINE-1, Alu and IAP. As described in Figure 2A, HeLa non-silencing control or MOV10 KD stable cell lines were transfected with retroelement expression plasmids and retrotransposition was quantified. (LINE-1 **p ≤ 0.0056, Alu ***p ≤ 0.0005, IAP **p ≤ 0.0096). (B) MOV10 silencing does not effect neo expression or selection directly. As described in Figure 2B, HeLa non-silencing control or MOV10 KD cells were transfected with the pcDNA3.1-neo control vector and colonies were quantified. Results are normalised to the non-silencing control, which is set at 100%. For (A) a single control bar set at 100% is graphed for simplicity. Values are the mean ± SD of 3 independent experiments. The data were analysed with an unpaired one-tailed t test.

Figure 5

MOV10 is not necessary for miRNA or siRNA-mediated mRNA silencing. (A) MOV10 is not necessary for miRNA-mediated mRNA repression. HeLa non-silencing control or MOV10 KD cells were co-transfected with either FF4LCS (let-7 WT) or FFr4mLCS (let-7 mutant) together with pRenilla. Relative luciferase activities were measured using a Dual-Luciferase® Reporter Assay System and FF luciferase activity was normalised to the renilla luciferase activity. (B) MOV10 is not required for siRNA-mediated mRNA cleavage. HeLa non-silencing control or MOV10 KD cells were transfected with either psi-CHECK2-let-7X3 (let-7 WT) or psi-CHECK2-let-7X3m (let-7 mutant). Luciferase activities were measured as described in panel (A) and renilla luciferase activity was normalised to FF luciferase activity. Results are normalised to the non-silencing control, which is set at 100%. Values are the mean ± SD of 3 independent experiments.