In the current study, we examined the role of SAMHD1 in restricting HIV-1 infection of DCs and compared SAMHD1 expression levels following treatment with IFNs. Our results show that Vpx-mediated degradation of SAMHD1 in DCs can relieve a post-entry restriction block against HIV-1 by increasing the intracellular dNTP pool and promoting the accumulation of HIV-1 late reverse transcription products. However, early reverse transcription products were not affected by SAMHD1 degradation, consistent with previous findings in macrophages
, suggesting that SAMHD1-mediated HIV-1 restriction mainly affects late reverse transcription. A previous study found that the HIV-1 genome is not completely reverse transcribed in quiescent lymphocytes, unlike in activated lymphocytes
. It is likely that HIV-1 early reverse transcription can be initiated in non-cycling cells with low dNTP concentrations, but the late reverse transcription cannot be completed. We also observed that there was no direct correlation between fold increase in HIV-1 late reverse transcription products and fold change in viral infectivity. A potential explanation for the difference is that HIV-1 late reverse transcription product levels may not fully reflect the efficiency of viral gene expression given the complexity of the viral life-cycle in DCs
SAMHD1 functions as a dGTP-dependent phosphohydrolase
[19, 20], and its degradation with Vpx treatment in DCs increased accumulation of HIV-1 late reverse transcription products, suggesting that SAMHD1 regulates intracellular dNTP levels in DCs. We show that DCs contain low levels of dNTPs (~11–415 nM), within the range of resting T-lymphocytes (300–5,000 nM)
, but below that in HIV-1 permissive cell types, such as activated peripheral blood mononuclear cells (PBMCs) (1.5–9.2 μM)
 and activated CD4+ T-lymphocytes (3–30 μM)
. It appears that DCs have dNTP levels ~1.9- to 2.3-fold higher than macrophages (~50 nM)
Although HIV-1 infection of DCs is enhanced in the presence of Vpx, the levels of p24 released into the supernatant from infected DCs are lower compared to those from macrophages and fully permissive target cells
[32, 33]. This suggests that SAMHD1 has a role in HIV-1 restriction in DCs, but it is likely that additional post-entry restriction steps exist to block HIV-1 infection in DCs. For example, APOBEC3A is highly expressed in myeloid-lineage cells and interacts with Vpx, leading to its degradation, which correlates with increased HIV-1 infection in primary monocytes
. Silencing of APOBEC3A relieves restriction of HIV-1 in macrophages, DCs and monocytes
, and abolished deaminase activity of APOBEC3A in monocytes
. There also remains an unidentified cryptic sensor for HIV-1 infection in DCs dependent on newly synthesized viral capsid
SAMHD1 restricts HIV1 infection in resting CD4+ T-lymphocytes by limiting reverse transcription through depleting intracellular dNTP concentrations
. Previous studies measuring dNTP levels in resting T-lymphocytes suggest that the intracellular dNTP pool is sufficiently low to restrict HIV-1 reverse transcription, which can be attributed to SAMHD1 activity
[37, 38]. Recent studies showed that T-cell activation does not significantly affect SAMHD1 expression in primary CD4+ T-cells treated with PHA or with anti-CD3/anti-CD28 for 2–3 days
[14, 15]. In agreement with these results, we observed that PHA-treatment of resting CD4+ T-cells for 2 days only slightly decreased SAMHD1 expression in activated CD4+ T-lymphocytes (by 10%- 30% in two donors, Figure
6). Activated CD4+ T-cells have a 3- to 8-fold higher dNTP concentration relative to resting CD4+ T-cells, while SAMHD1 expression remains the same in resting and activated CD4+ T-cells
. It is possible that intracellular dNTP levels can be significantly increased when CD4+ T-cells are activated and become dividing cells. How activated CD4+ T-cells upregulate the intracellular dNTP pool without decreasing SAMHD1 expression remains to be investigated.
We found that over-expression of SAMHD1 in dividing cell lines does not restrict HIV-1, similar to a study which found that SAMHD1 expression in dividing cell lines did not have an inhibitory effect on a range of viruses, including HIV-1
. Our dNTP analysis in HeLa cells suggests that SAMHD1 is able to moderately deplete the dNTP pool; but the concentration of dNTPs was within the range of activated T-lymphocytes
, suggesting dividing cell lines are capable of maintaining their dNTP pools in the presence of high levels of SAMHD1. It is possible that the catalytic activity of over-expressed SAMHD1 in HEK 293T and HeLa cells may be less stoichiometrically active than the endogenous protein in DCs, and/or that the transformed cell lines lack a potential cellular co-factor(s) for SAMHD1-mediated HIV-1 restriction function
Analysis of SAMHD1 after IFN treatment indicated that neither type I nor type II IFN treatment affected SAMHD1 protein levels in DCs or primary CD4+ T-lymphocytes. However, analysis of SAMHD1 mRNA levels at 6 hr post-treatment with IFN indicated a 2- to 4-fold increase in mRNA, suggesting that in DCs SAMHD1 is IFN sensitive, albeit transiently. Comprehensive analysis of the effect of IFNα treatment of DCs from 6 to 72 hrs indicated no change in SAMHD1 protein levels and a small transient increase in mRNA levels at 6 and 12 hr post-treatment. Furthermore, our data for IFN treatment of HEK 293T and HeLa cells, as well as previous studies
 show that SAMHD1 is type I IFN inducible. Although Berger et al. observed increased SAMHD1 protein levels in primary monocytes upon IFNα treatment, we observed that primary monocytes express lower levels of SAMHD1 relative to DCs (data not shown), which could partially explain the difference in response to IFNα treatment across the two cell types. As we show that DCs have low levels of dNTPs, it is plausible that SAMHD1 expression and/or its activity is tightly regulated in these cells to ensure a minimal dNTP pool is maintained without causing detrimental effects on the cell, for example, DNA repair within cells requires carefully modulated dNTP levels
[47, 48]. It is also possible that post-transcriptional regulation of SAMHD1 mRNA may affect SAMHD1 protein expression in the cell. A recent study identified naturally occurring splice variants of SAMHD1
, indicating that SAMHD1 expression and activity is regulated at a transcriptional level.
Interestingly, HIV-1 has no means of counteracting SAMHD1, and our recent study suggests that co-evolution of primate SAMHD1 and lentivirus Vpx led to the loss of the vpx gene in the HIV-1 precursor, SIVcpz, and consequently HIV-1
. Additional studies also suggested that SAMHD1 restriction toward HIV-1 was evolutionarily maintained under positive selection and that antagonism of SAMHD1 by Vpx is species-specific
[51, 52], but that Vpx degradation of SAMHD1 was an acquired ability that arose through positive selection in lentiviruses
. We recently reported that common polymorphisms of SAMHD1 are unlikely to contribute to the infection and natural control of HIV-1, at least in European and African-American individuals
. It is interesting to investigate whether polymorphisms of SAMHD1 are associated with HIV-2 and SIV infections in humans and non-human primates, respectively.
SAMHD1 has been suggested to play a role in the innate immune responses to viral infections
[54–58]. Our results indicate that SAMHD1 functions as an important restriction factor to counteract HIV-1 infection in DCs. Broader understanding of the mechanism of SAMHD1-mediated restriction in non-dividing cells and further investigation of the biological role of SAMHD1 is vital to enhancing our knowledge of HIV-1 infection and pathogenesis.