Incomplete Suppression of HIV-1 by SAMHD1 Permits Efficient Macrophage Infection

Abstract

Background: Sterile alpha motif and histidine/aspartic acid domain-containing protein (SAMHD1) is a dNTP triphosphorylase that reduces cellular dNTP levels in non-dividing cells, such as macrophages. Since dNTPs are required for reverse transcription, HIV-2 and most SIVs encode a Vpx protein that promotes proteasomal degradation of SAMHD1. It is unclear how HIV-1, which does not appear to harbor a SAMHD1 escape mechanism, is able to infect macrophages in the face of SAMHD1 restriction.

Methods: To assess whether HIV-1 had a mechanism to negate SAMHD1 activity, we compared SAMHD1 and dNTP levels in macrophages infected by HIV-1 and SIV. We examined whether macrophages infected by HIV-1 still harbored antiviral levels of SAMHD1 by assessing their susceptibility to superinfection by vpx-deleted SIV. Finally, to assess whether HIV-1 reverse transcriptase (RT) has adapted to a low dNTP environment, we evaluated SAMHD1 sensitivity of chimeric HIV-1 and SIV variants in which the RT regions were functionally exchanged.

Results: Here, we demonstrate that HIV-1 efficiently infects macrophages without modulating SAMHD1 activity or cellular dNTP levels, and that macrophages permissive to HIV-1 infection remained refractory to superinfection by vpx-deleted SIV. Furthermore, through the use of chimeric HIV/SIV, we demonstrate that the differential sensitivity of HIV-1 and SIV to SAMHD1 restriction is not dictated by RT.

Conclusions: Our study reveals fundamental differences between HIV-1 and SIV in the strategy used to evade restriction by SAMHD1 and suggests a degree of resistance of HIV-1 to the antiviral environment created by SAMHD1. Understanding how these cellular restrictions antagonize viral replication will be important for the design of novel antiviral strategies.

Keywords: HIV-1; SAMHD1; macrophages.

Figure 1.

SIV but not HIV-1 eliminates SAMHD1 and augments cellular dNTP levels in macrophages. (A-B) FACS analysis of viral Gag and SAMHD1 levels in uninfected macrophages and in macrophages infected with HIV-1 or SIV. Individual FACS profiles together with combined profiles are indicated in panel A and B respectively. (C-D) Quantitative (geometric means) and statistical comparisons of viral Gag (C) and SAMHD1 (D). (E) Western blot analysis (cropped image) of SAMHD1 levels in uninfected, HIV-1, SIVwt, and SIVΔvpx-infected macrophages together with quantitative and statistical analysis of SAMHD1 levels (F) are indicated. Lysates for Western blotting and intracellular dNTP analysis (H) were generated 1 day post-infection. Cells from the same infection were sampled 4 days post-infection for intracellular Gag staining. (G) The percentage of viral Gag-positive cells from the same experiments as samples collected for Western blotting and dNTP measurements and above the Western blot for that particular experiment. Results represent pooled data from 3 or more replicate experiments +/− SD. Welch's t-test: *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.

Figure 2.

SAMHD1 knock-down renders macrophages permissive to infection by vpx-deleted SIV. (A) Complete elimination of SAMHD1 expression in macrophages through successive (3X) rounds of siRNA transfection. Levels of SAMHD1 in macrophages 14 days after transfection with SAMHD1-specific or scrambled siRNAs or mock-transfected macrophages are shown by Western blot (cropped image). (B) SAMHD1 levels were quantitated by densitometry for statistical comparisons of SAMHD1 as normalized to GAPDH levels. (C-D) SAMHD1 knock-down permits macrophage infection by SIVΔvpx. Fourteen days after siRNA transfection, macrophages were infected with wild-type (wt) or vpx-deleted (SIVΔvpx) variants, and infection of siRNA transfected macrophages was gauged from the levels of full-length viral cDNA by qPCR analysis (C) and by viral reverse transcriptase activity in culture supernatants (D) at the indicated intervals post-infection. Bar graphs represent pooled data from 3 or more replicate experiments +/− SD. Welch's t-test: *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.

Figure 3.

HIV-1 preinfection does not render macrophages permissive to superinfection by vpx-deleted SIV. (A) Experimental design. Macrophages were first infected with HIV-1 or SIV and after 2 hours, superinfected with wild-type (SIVwt) or vpx-deleted (SIVΔvpx) SIV variants expressing GFP. (B) Macrophage infection by wild-type and vpx-deleted SIV variants was gauged from the number of GFP positive cells at 3-4 days post-infection. To gauge possible contribution of macrophage autofluorescence to the GFP signal and to ascertain de novo infection, tenofovir, which inhibits HIV-1 and SIV reverse transcriptases, was added to duplicate cultures 2 hours prior to infection. (C) SIVwt pre-infection also increased macrophage permissivity to HIV-1-GFP infection. Bar graphs represent pooled data from 3 or more replicate experiments +/− SD. Welch's t-test: *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.

Figure 4.

Construction and validation of RT-SHIV in which HIV-1 RT is inserted into an SIV backbone. (A) schematic of RT-SHIV in which the HIV-1 reverse transcriptase was inserted into an SIV backbone. (B) differential sensitivity of RT-SHIV to the RT inhibitors nevirapine (NVP) and tenofovir (TFV) was assessed in sMAGI cells (susceptible to infection by HIV-1 and SIV). Tenofovir is active against HIV-1 and SIV RT while nevirapine is only active against HIV-1. sMAGI cells were pretreated with the RT inhibitors for 2 hours and then infected with wild type or vpx-deleted SIV variants or with RT-SHIV variants harboring a functional (RT-SHIVwt) or inactive vpx (RT-SHIVΔvpx) open reading frame. Infection of sMAGI cells was gauged from levels of viral cDNA by qPCR at the indicated intervals post-infection. Results represent pooled data from 3 or more replicate experiments +/− SD. *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.

Figure 5.

HIV-1 RT, within the context of an SIV backbone, requires Vpx for reverse transcription in macrophages. (A-B) Permissivity of macrophages to SIV infection is absolutely dependent on Vpx. Macrophages were infected with wild type or vpx-deleted SIV variants and infection was gauged by levels of viral cDNA by qPCR (A) or Gag-positive cells by FACS (B). Introduction of Vpx in trans (SIVΔvpx + Vpx) permitted macrophage infection by vpx-deleted SIV. (C-D) Permissivity of macrophages to RT-SHIV infection is absolutely dependent on Vpx. Macrophages were infected with RT-SHIV variants containing intact (RT-SHIVwt) or defective (RT-SHIVΔvpx) vpx open reading frames and infection gauged by qPCR for viral cDNA (C) and FACS for Gag-positive cells (D). As with parental SIVΔvpx, introduction of Vpx in trans rescued vpx-defective RT-SHIV (RT-SHIVΔvpx+Vpx). Bar graphs represent pooled data from 3 or more replicate experiments +/− SD. Statistical comparisons for panels A and C are based on Area Under the Curve analysis, and on Welch's t-test for panels B and D) *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.

Figure 6.

SIV RT, within the context of an HIV-1 backbone, does not require Vpx for reverse transcription in macrophages. (A) Schematic of RT-HSIV depicting HIV-1 harboring an SIV RT. (B) RTHSIV infection of sMAGI cells is sensitive to inhibition by tenofovir but not nevirapine. sMAGI cells were infected with wild type HIV-1 (HIV-1) or HIV-1 harboring SIV RT (RT-HSIV) and viral cDNA synthesis assessed by qPCR at the indicated intervals post-infection. (C) Macrophages are equally permissive to infection by wild type HIV-1 and HIV-1 containing SIV RT (RT-HSIV) despite the absence of Vpx. Macrophage infection was assessed by qPCR for viral cDNA at the indicated intervals post-infection. (D) Levels of Gag-positive cells were assessed by flow cytometry after HIV-1 or RT-HSIV infection. (E) Macrophages were infected for 2 hours with SIVwt, then superinfected with RT-HSIV. RT-HSIV cDNA levels were measured by qPCR. Bar graphs represent pooled data from 3 or more replicate experiments +/− SD. Statistical comparisons for panels B, C, and E are based on Area Under the Curve analysis. Panel D statistical comparisons were based on Welch's t-test. *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.

Supplemental Figure.

HIV-1 RT, within the context of an SIVmac316e backbone, requires Vpx for reverse transcription in macrophages. (A) Differential sensitivity of RT-SHIV316e to the RT inhibitors nevirapine (NVP) and tenofovir (TFV) was assessed in sMAGI cells (susceptible to infection by HIV-1 and SIV). Tenofovir is active against HIV-1 and SIV RT while nevirapine is only active against HIV-1. sMAGI cells were pretreated with the RT inhibitors and then infected with wild type or vpx-deleted SIVmac316e variants or with RT-SHIVmac316e variants harboring a functional (RT-SHIVwt) or inactive vpx (RT-SHIVΔvpx) open reading frame. Infection of sMAGI cells was gauged from levels of viral cDNA by qPCR at the indicated intervals post-infection. Pooled data from 3 or more replicate experiments +/− SD. (B) Infection of macrophages by SIV harboring an SIVmac316e RT is absolutely dependent on Vpx. Macrophages were infected with RT-SHIV316e variants containing intact (RT-SHIV316ewt) or defective (RTSHIV316eΔvpx) vpx open reading frames and infection gauged by qPCR for viral cDNA at the indicated intervals post-infection. Bar graphs in panel A represent pooled data from 3 or more replicate experiments +/− SD. Statistical comparisons for panel B were made from Area Under the Curve analysis. *P < 0.05 **P < 0.01 ***P < 0.001 ****P < 0.0001.