RNA helicase A modulates translation of HIV-1 and infectivity of progeny virions

Abstract

Retroviruses rely on host RNA-binding proteins to modulate various steps in their replication. Previously several animal retroviruses were determined to mediate Dhx9/RNA helicase A (RHA) interaction with a 5' terminal post-transcriptional control element (PCE) for efficient translation. Herein PCE reporter assays determined HTLV-1 and HIV-1 RU5 confer orientation-dependent PCE activity. The effect of Dhx9/RHA down-regulation and rescue with siRNA-resistant RHA on expression of HIV-1(NL4-3) provirus determined that RHA is necessary for efficient HIV-1 RNA translation and requires ATPase-dependent helicase function. Quantitative analysis determined HIV-1 RNA steady-state and cytoplasmic accumulation were not reduced; rather the translational activity of viral RNA was reduced. Western blotting determined that RHA-deficient virions assemble with Lys-tRNA synthetase, exhibit processed reverse transcriptase and contain similar level of viral RNA, but they are poorly infectious on primary lymphocytes and HeLa cells. The results demonstrate RHA is an important host factor within the virus-producer cell and within the viral particle. The identification of RHA-dependent PCE activity in cellular junD RNA and in six of seven genera of Retroviridae suggests conservation of this translational control mechanism among vertebrates, and convergent evolution of Retroviridae to utilize this host mechanism.

Figure 1.

HTLV-1 RU5 is necessary for PCE activity and interaction with RHA. (A) Drawing of HTLV-1 PCE-gag reporter plasmid with HTLV-1 R depicted in blue and U5 in purple with corresponding sequence shown below. Arrow indicates transcription start site; 5′ ss, splice site; pA_(n), polyadenylation signal. Dashed line denotes deletion of RU5 and reverse arrow indicates antisense orientation (AS) of RU5. HEK 293 cells were transfected with indicated reporter plasmid and HTLV-1 tax-1 expression plasmid to trans-activate the HTLV-1 U3 promoter for 48 h. Gag protein production was measured by Gag p24 ELISA and gag mRNA was quantified by RT-real-time PCR in four independent experiments. ANOVA model was used to study the differences in Gag protein level and gag RNA level and data are presented as mean fold change with corresponding P-values relative to pHTLV-1Δ with P-value indicated in parentheses. Asterisks indicate statistically significant difference (P ≤ 0.05). (B) RNA affinity chromatography was performed with HeLa nuclear extract and equimolar HTLV-1 RU5, RU5 AS, R, U5 or c-myc RNA bait. After sequential washes with KCl, high affinity proteins were eluted in 2 M KCl and immunoblotted with RHA antiserum lanes: input lysate; M, marker; Beads, reaction with no RNA control and indicated RNA bait.

Figure 2.

HIV-1 RU5 is necessary for PCE activity and interaction with RHA. (A) Drawing of HIV-1 PCE-gag reporter plasmid with HIV-1 R depicted in green and U5 in orange with corresponding sequence shown below. Arrow indicates transcription start site; 5′ ss, splice site; pA_(n), polyadenylation signal. Dashed line denotes deletion of R and reverse arrow indicates antisense orientation (AS) of RU5. HEK 293 cells were transfected with the indicated plasmid, Gag production was measured by Gag p24 ELISA and gag mRNA was quantified by RT-real-time PCR. ANOVA model was used to study the differences in Gag protein level and gag RNA level. Data are presented as mean fold change with corresponding P-values. Results are presented as mean fold change relative to pHIV-1U5 with P-value indicated in parentheses. Asterisks indicate statistically significant difference (P ≤ 0.05). (B) RNA affinity chromatography was performed with HeLa nuclear extract and equal moles of indicated HIV-1 RNA or c-myc RNA bait. High affinity proteins were eluted in 2 M KCl and immunoblotted with RHA antiserum. Lanes: input lysate; M, marker; Beads, reaction with no RNA control and indicated RNA bait.

Figure 3.

RHA is necessary for efficient production of HIV-1 Gag. HEK 293 cells were transfected consecutively with non-silencing (Sc) or RHA siRNAs (RHA) and HIV-1^NL4–3. (A) Immunoblot of total cell protein with antiserum to RHA or α-Tubulin verified RHA down-regulation. (B) Cell-associated (n = 5) and cell-free (n = 16) Gag levels measured by Gag immunoblot or Gag ELISA, respectively. Asterisks indicate statistically significant difference from Sc siRNA (P = 0.0006 for cell-associated and P < 0.0001 for cell-free Gag, respectively). (C) Cell-free Gag production measured by Gag ELISA (n = 3). Asterisks indicate statistically significant difference from Sc siRNA at indicated time point (P ≤ 0.0001).

Figure 4.

RHA is necessary for efficient translation of HIV-1 gag RNA. HEK 293 cells were transfected consecutively with non-silencing (Sc) or RHA siRNAs (RHA) and HIV-1^NL4–3 for 48 h, labeled for 1 h with [³⁵S]-cysteine/methionine and immunoprecipitation was performed in triplicate. Graph summarizes densitometry of [³⁵S]-labeled Gag and β-Actin protein. Asterisks indicate statistically significant difference from Sc siRNA control (P = 0.002).

Figure 5.

RHA down-regulation does not affect cytoplasmic accumulation of HIV-1 gag RNA. HEK 293 cells were transfected consecutively with non-silencing (Sc) or RHA siRNAs (RHA) and HIV-1^NL4–3 for 48 h and aliquots harvested for fractionation or 1 h labeling with [³⁵S]-cysteine/methionine. (A) Immunoblot of total cell protein with antiserum to RHA or β-Actin. (B) Immunoblot of nuclear or cytoplasmic extracts with Histone H1 or Tubulin antibody. (C) RT real-time PCR measured nuclear or cytoplasmic gag RNA and β-actin RNA. Cytoplasmic accumulation of gag RNA relative to β-actin is expressed as percentage. Translational efficiency is expressed as ratio of [³⁵S]-labeled Gag protein to cytoplasmic gag RNA.

Figure 6.

ATPase domain of RHA is necessary for efficient translation of HIV-1 gag RNA. HEK 293 cells were transfected consecutively with non-silencing (Sc) or RHA siRNAs (RHA) and HIV-1^NL4–3 for 48 h. Cells were labeled with [³⁵S]-cysteine/methionine for 1 h for IP with Gag and β-Actin antibody. (A) Evaluation of rescue by siRNA-resistant FLAG-RHA by IP (top panel). Immunoblot of total cell protein with indicated antiserum verified RHA down-regulation, expression of siRNA-resistant FLAG-RHA and equal protein loading, respectively (bottom panel). (B) Evaluation of rescue by siRNA-resistant FLAG-RHA K417R by IP (top panel). Immunoblot of total cell protein with indicated antiserum verified RHA down-regulation, expression of siRNA-resistant FLAG-RHA K417R and equal protein loading, respectively (bottom panel).

Figure 7.

RHA down-regulation reduces virus protein but not HIV-1 RNA packaging. HEK 293 cells were transfected consecutively with scrambled (Sc) or RHA (RHA) siRNAs and HIV-1^NL4–3 provirus and cellular protein and RNA were isolated. Gag p24 ELISA on cell-free medium was performed. (A) Immunoblot of total cell protein (10 µg) with antiserum to Gag, RHA, LysRS, Vif, Nef, Rev or β-Actin. (B) Immunoblot of virion preparation (equivalent to 5 ng of p24) with antiserum to RHA, LysRS, RT, Gag. (C) RT-real-time PCR determined copy number. ^aGag copy number in virion preparation equivalent to 25 ng of p24. ^bCellular gag RNA copy number standardized to β-actin RNA.

Figure 8.

RHA down-regulation reduces infectivity of progeny HIV-1. (A) HEK 293 cells were transfected with scrambled (Sc) or RHA (RHA) siRNAs, and then second dose of siRNA and HIV-1^NL4–3 for 48 h. Cell-free virus equivalent to 2 ng Gag was used to infect activated PBMCs. Virus growth on PBMC was measured by Gag ELISA (n = 3). Asterisks indicate statistically significant difference from Sc siRNA control was observed at day 6, 9 and 12 (P ≤ 0.0005, 0.0008, 0.0174). (B) HEK 293 cells were transfected with Sc or RHA siRNAs, and then second dose of siRNA with either empty vector or siRNA-resistant FLAG-RHA and HIV-1^NL4–3. Cell-free virus equivalent to 2 ng Gag was used to infect TZM-bl cells and luciferase activity determined at 48 h (n = 3).

Figure 9.

Model summarizing that DExD/H proteins contribute to many steps in retroviral posttranscriptional gene expression and coordination of virus assembly. Nuclear export of HIV-1 Rev/Rev responsive element (RRE)-dependent HIV-1 RNA requires DDX3 with CRM1 nuclear export receptor (27). Alternatively, the nuclear export ribonucleoprotein (RNP) of completely spliced viral RNA involves superfamily members RBM15 and DBP5 and NXF1 nuclear export receptor (28,29). Efficient viral translation requires DHX9/RNA helicase A [herein and (5)]. RNA helicase A recognizes structural features of virus 5′ RNA terminus and possibly distal structures, and facilitates efficient protein synthesis in an ATP-dependent manner (2). RNA helicase A also assembles into virions and supports virus infectivity, perhaps by promoting rearrangement of the viral RNP for reverse transcription or by carrying in a host cofactor (12). DDX24 promotes packaging of HIV-1 RNA (30).