Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Throughout evolution, organisms have needed to adapt to their environments and develop defenses to pathogens in order to ensure survival and propagation. Although the skin and the epithelial lining of internal mucosal surfaces provide an effective physical barrier to exposure to most microorganisms, the immune system has developed in vertebrates to prevent or clear pathogenic infections through innate and adaptive immune responses. The conventional innate response combats invading pathogens through non-specific mechanisms such as inflammation, complement activation, and phagocytosis by leukocytes. This often results in activation of the adaptive response in which the immune system develops more potent, specific recognition of pathogens or pathogen-infected cells and immunological memory develops.

In addition to the classically defined immune effectors, intrinsic host factors have been acquired during evolution to inhibit viral replication. Also know as restriction factors, well-defined examples include TRIM5α, tetherin (BST2), and the APOBEC family of proteins. These are dominant genes that impart antiviral effects early after exposure to prevent or limit viral infections. The presence of intrinsic antiviral factors creates an unfavorable cellular environment for viruses, and they have developed mechanisms to overcome restrictions to replication and persistence. For example, Vpu downregulates CD4 from the plasma membrane to increase virion infectivity and enhances virus release by counteracting the effects of tetherin (Andrew and Strebel 2010; Dube et al. 2010). The Vif protein is essential in vivo to counteract the inhibition of reverse transcription and G to A hypermutation mediated by the APOBEC3G/3F proteins (Gillick et al. 2013). Vif recruits the cellular ubiquitin ligase machinery to target APOBEC3G/3F for ubiquitination and subsequent proteasomal degradation (Mehle et al. 2004; Yu et al. 2003). Vpr also utilizes a similar strategy to target an as yet undefined cellular target to create a more favorable cellular environment for virus replication (Belzile et al. 2007; Hrecka et al. 2007). Inducing degradation of a cellular protein, such as a restriction factor, is a common strategy employed by viruses to eliminate obstacles to efficient replication (Gustin et al. 2011). A recent addition to the growing list of viral restriction factors has been identified as the sterile alpha motif and HD domain-containing protein (SAMHD1). SAMHD1 expression in non-cycling myeloid cells potently inhibits infection by a divergent group of retroviruses including HIV-1 and MLV (Goujon et al. 2007; Kaushik et al. 2009). Vpx, an accessory gene encoded by some lentiviruses, is able to overcome this block to infection. This chapter will focus on the restriction to lentiviral replication in myeloid cells and the roles that Vpx and SAMHD1 have in promoting or inhibiting the process.

2 Restriction to Retroviral Infection in Myeloid Cells

The majority of HIV-1 replication occurs in activated CD4+ T cells, and it is the progressive loss of these immune cells that results in immunodeficiency and associated disease. In contrast, non-proliferating, quiescent CD4+ T cells exhibit a profound block to infection as evidenced by the labile nature of the products of reverse transcription (Pierson et al. 2002; Stevenson et al. 1990; Zack et al. 1990). Similarly, myeloid lineage cells of the immune system are susceptible to lentivirus infection, but the efficiency of infection is low. Peripheral blood monocytes are highly refractory to productive HIV-1 infection, but become more permissive upon differentiation into macrophages and dendritic cells (Neil et al. 2001; Rich et al. 1992; Sonza et al. 1996). In vitro, HIV-1 enters freshly isolated peripheral blood monocytes efficiently but fails to generate products of reverse transcription (Sonza et al. 1996; Triques and Stevenson 2004). Upon differentiation in culture, monocyte-derived dendritic cells and macrophages become more susceptible to infection (Bergamaschi and Pancino 2010), although virus replication is much slower and less efficient than in activated, cycling cells. Thus, cells such as macrophages are susceptible to HIV-1 infection, but restrict or limit the levels of replication through cellular processes. Observations that intracellular dNTP levels are low in monocytes suggest that viral cDNA synthesis in these cells is abortive (O’Brien et al. 1994; Triques and Stevenson 2004) and this might account for the low level of infection. A role for APOBEC3G in regulating the permissivity of myeloid cells to HIV-1 infection has also been proposed (Ellery et al. 2007; Peng et al. 2007; Pion et al. 2006). Although these previously described mechanisms may contribute to inhibition of virus replication in monocytes and macrophages, it has become evident through several studies that myeloid lineage cells harbor a dominant restriction to lentiviral infection (Bergamaschi et al. 2009; Goujon et al. 2007; Kaushik et al. 2009; Negri et al. 2012; Schule et al. 2009; Sharova et al. 2008; Srivastava et al. 2008; Triques and Stevenson 2004).

3 Vpr and Vpx

3.1 Vpr and Vpx are Virulence Factors

HIVs and SIVs are highly adapted primate lentiviruses that encode several accessory genes that have a general role in enhancing virus replication and persistence in the host cell (Fig. 1). While the Tat and Rev genes are essential regulators of viral expression, Vif, Vpu, Vpr, Vpx, and Nef are dispensable for virus growth in most cell culture systems in vitro. Originally thought to be marginally important for virus replication, recent advances are providing a greater understanding and appreciation for their importance in the setting of a natural infection.

Fig. 1
figure 1

Genomic organization of HIV-1 and HIV-2/SIVsm. In addition to the gag, pol, and env genes common to all retroviruses, primate lentiviruses encode regulatory and accessory genes that modify the host cell environment to benefit the virus. While the HIV-1 lineage encodes Vpr, the HIV-2/SIVsm lineage encodes both a Vpr gene and a Vpx gene. Reprinted from Ayinde et al. (2010)

Full size image

Vpx and the genetically related accessory protein Vpr are virulence factors originally reported to exert a positive effect on virus replication in myeloid lineage cells. Although dispensable in vitro, Vpr and Vpx are important in vivo as demonstrated in non-human primate models (Gibbs et al. 1995; Hirsch et al. 1998). Rhesus macaques infected with Vpx-deficient SIV have reduced viral loads and delayed progression to AIDS. SIV defective for both genes is greatly attenuated, and infected animals have low viral burdens and lack symptoms of disease. The two accessory genes are not encoded by all HIVs and SIVs. HIV-1, HIV-2, and SIV isolated from sooty mangabeys (SIVsm) encode Vpr, but Vpx is specific to the HIV-2/SIVsm lineage (Cohen et al. 1990; Henderson et al. 1988; Kappes et al. 1988). SIV isolated from red-capped mangabeys and mandrills also encode both Vpr and Vpx, but all other primate lentiviruses have a single related gene that has been designated Vpr (Beer et al. 2001; Hu et al. 2003). Vpx encoded by the HIV-2/SIVsm lineage is thought to have arisen by recombination between SIV from sooty mangabey and SIV from African green monkeys, essentially resulting in the addition of Vpx to the ancestral SIVsm genome (Sharp et al. 1996; Tristem et al. 1998). The most recent common ancestor of modern day SIVs and HIV-1 was an SIV that encoded a Vpx gene (Zhang et al. 2012). The Vpx gene was likely lost during long-term evolution of SIV within chimpanzees, and the zoonotic transfer of virus from chimpanzees to humans gave rise to an HIV-1 that encodes Vpr, but not Vpx (Zhang et al. 2012).

3.2 Vpr and Vpx are Packaged into Virions

Both proteins are packaged into the virion core via an interaction with the structural p6 component of the Gag gene (Accola et al. 1999; Bachand et al. 1999; Kewalramani and Emerman 1996; Wu et al. 1994). As integral virion-encapsidated proteins, they are delivered intracellularly upon virion fusion with the target cell membrane, suggesting that they function early in the viral life cycle. Several properties have been described for Vpr, but a clear consensus on its role in virus replication and pathogenesis is lacking. In addition to its properties in promoting cell cycle arrest and transcriptional activation (Planelles and Benichou 2009), early studies suggested that Vpr facilitates macrophage infection by driving the nuclear import of the preintegration complex (Fassati 2006; Heinzinger et al. 1994). Retroviruses must integrate into the host cell chromosome to transition to a mode of productive infection, and lentiviruses are unique in their ability to traverse an intact nuclear membrane to allow infection of non-cycling cells. Several lines of evidence support the conclusion that Vpr, along with gagMA protein, are nucleophilic proteins that drive nuclear localization of the preintegration complex (Bukrinsky et al. 1993; Popov et al. 1998; von Schwedler et al. 1994). Virus in which Vpr was inactivated replicated poorly in primary macrophages, although a similar effect for the same virus was not observed in activated lymphocytes (Balliet et al. 1994). Although this implies that Vpr is required for optimal viral replication in terminally differentiated macrophages, the effect can vary significantly from donor to donor (Balliet et al. 1994). Subsequent studies have also suggested that HIV infection of terminally differentiated cells is not dependent on any of the karyophilic properties of viral proteins (Fouchier et al. 1997; Yamashita and Emerman 2005). Therefore, a well-defined role for the biological function of Vpr remains to be elucidated. In contrast, the ability of Vpx in promoting infection of non-dividing cells is emerging partly due to the efforts in delineating the biological function of Vpr.

3.3 Vpr Induces Cell Cycle Arrest

In particular, investigations focused on understanding the underlying mechanism of Vpr-mediated cell cycle arrest were instrumental in gaining a better understanding of the functional importance of Vpx. Vpr-induced cell cycle arrest was found to be dependent on its interaction with a cellular host factor, identified as DDB1-cullin 4-associated factor 1 (DCAF1). DCAF1 is identical to a cellular protein, previously reported to bind Vpr with high affinity and aptly named Vpr-binding protein (Zhang et al. 2001). Vpr recruits an E3 ubiquitin ligase complex through its interaction with DCAF1 (Fig. 2) and induces the ubiquitinylation and proteasomal degradation of an unknown host factor that is required for entry into mitosis (Belzile et al. 2007; DeHart et al. 2007; Hrecka et al. 2007; Schrofelbauer et al. 2007). Although it has been proposed that arresting cells in G2 would provide a significant boost in virus production from short-lived infected cells (Goh et al. 1998), G2 arrest may be induced as a coincidental consequence of a viral strategy to eliminate an intrinsic antiviral factor whose expression is linked to the cell cycle. Vpr appears to enhance virus replication under certain conditions, but it is unlikely to be linked to cell cycle arrest in macrophages, since these cells do not divide. A better understanding of the biologically significant role of Vpr in the lentiviral life cycle awaits identification of the host cell factor that is targeted for degradation by Vpr.

Fig. 2
figure 2

Model for HIV-1 Vpr mechanism of action. HIV-1 Vpr recruits the CUL4A-DDB1 ubiquitin ligase through DCAF1 binding, which leads to the ubiquitination and inactivation of an unknown cellular target required for entry into mitosis. Reprinted from Ayinde et al. (2010)

Full size image

3.4 Early Functional Studies of Vpx

HIV-2 and some SIVs encode both of the genetically related, although functionally distinct, accessory genes designated Vpr and Vpx. Unlike HIV-1 Vpr, Vpx does not induce G2 cell cycle arrest in dividing cells. Early studies determined that Vpx functions to enhance nuclear import of the preintegration complex in non-dividing cells, while it was found to be dispensable in cycling cells (Belshan et al. 2006; Fletcher et al. 1996; Planelles et al. 1996). Considerable evidence supports a functional role for Vpx in promoting nuclear localization of the preintegration complex containing the viral genome. It is efficiently packaged into the virion and generally thought to facilitate an early step in the viral life cycle (Pancio and Ratner 1998). Vpx associates with the preintegration complex, and residues constituting nuclear import signals within the protein have been characterized (Belshan and Ratner 2003; Pancio et al. 2000) and found to be necessary for productive infection of non-dividing cells (Belshan et al. 2006; Pancio et al. 2000). Vpx nuclear localization is enhanced via its interaction with a member of the Hsp40/DnaJ family of cellular chaperones, and modulation of cellular Hsp40/DnaJB6 levels impacts transport of HIV-2 preintegration complexes (Cheng et al. 2008). Furthermore, Vpx is an important determinant for establishing SIV infection in pigtailed macaques (Belshan et al. 2012; Hirsch et al. 1998) and promoting virus dissemination and pathogenesis (Hirsch et al. 1998).

3.5 Vpx Enhances Reverse Transcription

In addition to promoting nuclear import of preintegration complexes, in vitro studies with myeloid lineage cells led to the unexpected conclusion that Vpx functions to enhance the efficiency of reverse transcription (Fujita et al. 2008; Goujon et al. 2007; Sharova et al. 2008; Srivastava et al. 2008). The Vpx-mediated increase in reverse transcripts is cell-type dependent, with dendritic cells being some of the most restrictive to retrovirus infection (Goujon et al. 2007). Treatment of dendritic cells with the proteasome inhibitor MG132 partially restores infectivity of a Vpx-defective SIV, suggesting that Vpx is needed to counteract a proteasome-dependent antiviral mechanism present in dendritic cells (Goujon et al. 2007). Perhaps not surprisingly, Vpx specifically binds to the same DCAF1 subunit as Vpr to assemble a CUL4A-DDB1 ubiquitin ligase complex to target a cellular protein for proteasomal destruction (Bergamaschi et al. 2009; Sharova et al. 2008; Srivastava et al. 2008). While the cellular binding partner of Vpr remains to be identified, two research groups have recently demonstrated that SAMHD1 is the antiviral protein expressed in monocytes, dendritic cells, and macrophages that Vpx targets for ubiquitinylation and rapid proteasomal degradation (Hrecka et al. 2011; Laguette et al. 2011).

4 Sterile Alpha Motif and Histidine–Aspartic Domain Protein 1

4.1 Identifying the Restriction Factor SAMHD1

Using affinity purification and mass spectrometry, two independent groups identified SAMHD1 as a cellular protein that strongly interacts with Vpx (Hrecka et al. 2011; Laguette et al. 2011). SAMHD1 also associates with the cullin 4-based E3 ubiquitin ligase complex, but only when Vpx is part of the complex (Hrecka et al. 2011). Vpx acts as an adaptor to load SAMHD1 onto the CUL4A/DDB1/DCAF1 E3 ubiquitin ligase complex, resulting in ubiquitin modification and ultimately proteasomal degradation (Hrecka et al. 2011; Laguette et al. 2011). RNAi-mediated silencing of SAMHD1 allows HIV-1 to infect macrophages and dendritic cells independently of Vpx (Hrecka et al. 2011; Laguette et al. 2011). A consistent inverse correlation exists between the expression of SAMHD1 and permissiveness to HIV-1 infection (Laguette et al. 2011). Permissive cells such as Jurkat and monocytic U937 cells do not express SAMHD1, but non-permissive cells, including THP-1, monocytes, macrophages, and dendritic cells, exhibit high expression levels (Baldauf et al. 2012; Laguette et al. 2011). Artificially elevating SAMHD1 levels in normally permissive cells render them resistant to infection, unless Vpx is also expressed (Laguette et al. 2011). This experimental evidence strongly supports the conclusion that SAMHD1, identified through its ability to interact with Vpx, is the novel restriction factor that limits viral replication in myeloid lineage cells.

4.2 SAMHD1 Properties

SAMHD1 is a 626 amino acid protein that localizes to the nucleus and is uniquely comprised of a sterile alpha motif (SAM) and a histidine–aspartic (HD) domain that occurs in tandem (Aravind and Koonin 1998; Qiao and Bowie 2005). SAM domains are abundant protein interaction modules involved in protein–protein and protein–nucleic acid interactions that mediate many cellular processes, such as signal transduction and transcriptional regulation (Qiao and Bowie 2005). HD domains contain highly conserved histidine and aspartic acid residues and are found in a superfamily of enzymes with a protein fold predicted to have phosphohydrolase activity (Aravind and Koonin 1998; Zimmerman et al. 2008). Previously characterized HD domain-containing enzymes utilize nucleotides as substrates (Zimmerman et al. 2008). Structural studies and analysis of enzymatic activity indicate that SAMHD1 cleaves dNTPs to produce deoxynucleosides and triphosphates (Goldstone et al. 2011; Powell et al. 2011). The predominant subcellular localization of SAMHD1 appears to be in the nucleus (Berger et al. 2012; Brandariz-Nunez et al. 2012; Hofmann et al. 2012), although conflicting data indicates it may also localize to the cytoplasm in some cell types (Baldauf et al. 2012). Analysis of the primary amino acid sequence of SAMHD1 identified residues 11–14 (KRPR) as a match to the classical monopartite consensus nuclear localization signal (Brandariz-Nunez et al. 2012; Hofmann et al. 2012). Mutagenesis of the signal residues demonstrates that colocalization of Vpx and SAMHD1 in the nucleus is functionally important. Inactivation of the NLS shifts accumulation of SAMHD1 to the cytoplasm, but the cytoplasmic protein retains catalytic activity and antiviral function (Brandariz-Nunez et al. 2012; Hofmann et al. 2012). Cytoplasmic SAMHD1 can interact with Vpx, but this interaction is not sufficient for inducing the degradation of SAMHD1. There is a strong dependence on the Vpx–SAMHD1 interaction to occur in the nuclear compartment to initiate the process that results in depletion of cellular SAMHD1 by the proteasome (Brandariz-Nunez et al. 2012; Hofmann et al. 2012).

The SAMHD1 gene was previously cloned from dendritic cells and reported to be the human homologue of an interferon-γ-inducible gene in the mouse (Li et al. 2000). Treatment of human dendritic cells with either interferon-α, interferon-β, or interferon-γ does not increase SAMHD1 protein levels despite a transient increase in mRNA abundance (St Gelais et al. 2012). The absence of a regulatory effect in response to interferon in this system would make SAMHD1 unique among most antiviral restriction factors which are typically responsive to type I interferon (Bieniasz 2004; Harris et al. 2012). In contrast, treatment of monocytes or monocyte-derived macrophages with interferon significantly upregulated SAMHD1 levels, although variability in baseline expression levels of SAMHD1 in the different systems may account for the observed discordance (Berger et al. 2011; Rice et al. 2009). Interferon induction of SAMHD1 expression is also suggested by studies in which Vpx greatly enhances HIV-1 infection in type I interferon–treated macrophages (Gramberg et al. 2010). Interestingly, the loss of SAMHD1 expression due to genetic abnormalities in individuals with Aicardi–Goutières syndrome causes an inflammatory autoimmune state associated with elevated expression of type I interferon in the brain (Crow and Rehwinkel 2009; Rice et al. 2009). Clearly, there is a linkage between SAMHD1 expression and the immune system that implicates SAMHD1 as a negative regulator of the innate immune response (Rice et al. 2009).

4.3 Defects in SAMHD1 Cause Aicardi–Goutières Syndrome

Aicardi–Goutières syndrome is an autosomal neurological disorder that results in elevated type I interferon production phenotypically similar to that observed in congenital viral infections (Rice et al. 2009). Mutations in the 3′ repair exonuclease 1 (TREX1) or either subunit of the RNase H2 complex is causatively linked to AGS. TREX1 is a nuclease that preferentially degrades ssDNA in the cytosol to prevent the accumulation of DNA from endogenous retroelements or reverse-transcribed viral genomes that could otherwise trigger an autoimmune response (Stetson et al. 2008; Yan et al. 2010). Accumulation of cytosolic nucleic acid is detected by cellular receptors which trigger immune activation through a TLR-independent antiviral pathway (Stetson et al. 2008). HIV is normally able to evade this response, but excess HIV DNA accumulates in TREX1-deficient cells, resulting in detection and signaling to induce type 1 interferon (Yan et al. 2010). The RNase H2 complex also functions as a nuclease to remove misincorporated ribonucleosides from products of DNA replication, providing a mechanism to maintain chromosome integrity by avoiding elevated responses to DNA damage (Hiller et al. 2012). Like TREX1, this enzyme complex appears to facilitate HIV-1 infection, although its mechanism of action is unclear (Genovesio et al. 2011).

Prior to being identified as a lentiviral restriction factor, SAMHD1 was also linked to AGS and known to account for approximately ten percent of cases (Crow and Rehwinkel 2009). The loss of SAMHD1 expression results in the accumulation of cytosolic DNA and the associated increase in production of type I interferon (Rice et al. 2009). Like TREX1 and the RNase H2 complexes, SAMHD1 is also a modulator of the HIV-1 life cycle, but it functions to restrict rather than promote infection. Monocytes from SAMHD1-deficient AGS patients are highly susceptible to HIV-1 infection, while HIV-1 infection of healthy donor monocytes is very inefficient (Berger et al. 2011). In vitro, SAMHD1 does not degrade single-stranded DNA, double-stranded DNA, or RNA (Goldstone et al. 2011), so the mechanism by which abnormalities in SAMHD1 expression lead to the development of AGS differs from those of TREX1 and the RNase H2 complexes. Rather than degradation, the loss of SAMHD1 expression eliminates a cellular mechanism to inhibit the synthesis of cytosolic nucleic acid (Goldstone et al. 2011). Thus, there is an intricate relationship between nucleic acid metabolism, innate immune response, and susceptibility to lentiviral infection.

4.4 SAMHD1 Modulates Nucleotide Pools

SAMHD1 is a deoxyribonucleoside triphosphate triphosphohydrolase that converts dNTPs to deoxynucleosides and inorganic phosphate (Goldstone et al. 2011; Powell et al. 2011). Recombinant SAMHD1 rapidly hydrolyzes dGTP in vitro, while dATP, dCTP, and dTTP are not substrates when assayed individually. However, dNTPs in combination with dGTP result in efficient hydrolysis of all four dNTPs (Goldstone et al. 2011; Powell et al. 2011). The crystal structure of the HD domain and molecular modeling indicate that dGTP binding mediates the allosteric activation of SAMHD1 to increase hydrolysis of all four dNTP substrates (Goldstone et al. 2011). Thus, dGTP is a preferred substrate and an enhancer of SAMHD1 enzymatic activity that modulates the intracellular levels of the four dNTPs.

Cellular dNTP levels are tightly regulated in response to expression of cell cycle–specific biosynthetic enzymes, such as ribonucleotide reductase (Bjursell and Skoog 1980). An optimized amount of the four dNTPs ensures fidelity during DNA replication and repair and minimizes negative consequences of inappropriate dNTP concentrations (Hofer et al. 2012; Mathews 2006; Niida et al. 2010). Accordingly, cells progressing through G1/S and S phase have relatively high concentrations of dNTPs (Bjursell and Skoog 1980). Conversely, in terminally differentiated non-cycling cells, such as macrophages, dNTP levels are relatively low (Diamond et al. 2004). Differential expression of SAMHD1 results in large cell-dependent differences in dNTP concentrations that highly correlate with antiviral activity (Table 1).

Table 1 dNTP levels are variable in cells targeted by HIV-1
Full size table

Levels of dNTPs in activated primary CD4+ T cells are approximately 100 times higher than in monocyte-derived macrophages (Diamond et al. 2004). Reducing cellular dNTP concentrations to levels below that which is required by reverse transcription for efficient cDNA synthesis results in potent inhibition of retroviral replication (Diamond et al. 2004; Kim et al. 2012; Lahouassa et al. 2012). This antiviral mechanism is supported by studies in which RNAi-mediated depletion of SAMHD1 increases dNTP pools and renders the cells sensitive to HIV-1 infection (Kim et al. 2012; Lahouassa et al. 2012). Furthermore, addition of exogenous deoxynucleosides to culture medium increases intracellular dNTP levels and sensitivity to HIV-1 infection (Lahouassa et al. 2012). Recent studies have addressed whether a similar antiviral mechanism accounts for the differences in permissivity to HIV-1 infection of activated versus resting CD4+ T cells.

5 SAMHD1 Restricts HIV-1 Infection of Quiescent CD4+ T Cells

Activated CD4+ T cells are highly susceptible to productive HIV-1 infection, but infection of quiescent CD4+ T cells is inefficient due to various post-entry blocks to replication (Pierson et al. 2002; Stevenson et al. 1990; Swiggard et al. 2004; Zack et al. 1990). Similar to infection of myeloid cells, reverse transcription in resting lymphocytes is inefficient (Plesa et al. 2007; Zack et al. 1990). Based on earlier work, it was unclear whether SAMHD1 contributed to the restriction to infection of resting lymphocytes, because dNTP levels are higher than in myeloid cells (Gao et al. 1993; Lahouassa et al. 2012) and SAMHD1 expression is not detected in T cell lines (Laguette et al. 2011). Analysis has revealed that SAMHD1 is highly expressed in resting CD4+ T cells isolated from peripheral blood and lymphoid tissue (Baldauf et al. 2012; Descours et al. 2012). Expression levels are similar to those detected in myeloid cells, and analysis of sorted cells indicates that the various subsets of peripheral blood CD4+ T lymphocytes express similarly high levels of SAMHD1 (Descours et al. 2012). Elevated expression of SAMHD1 and the profound inhibition of reverse transcription in quiescent lymphocytes raise the possibility that the mechanism of restriction detected in myeloid cells also functions in CD4+ T cells. In support of this, delivery of Vpx into resting target cells or supplying exogenous nucleosides to boost cellular dNTP levels promotes efficient DNA synthesis in the absence of cellular activation (Baldauf et al. 2012; Descours et al. 2012). Resting CD4+ T cells isolated from individuals with Aicardi–Goutières syndrome harboring a defective SAMHD1 gene are also susceptible to HIV-1 infection (Baldauf et al. 2012; Descours et al. 2012). Notably, SAMHD1 degradation was not sufficient for virus production, indicating that additional blocks in the viral life cycle are not bypassed by elevating dNTP levels in resting T cells (Baldauf et al. 2012). In addition, SAMHD1 expression levels are similar in both quiescent and activated CD4+ T cells, suggesting that additional regulatory mechanisms influence dNTP levels in activated cells that are needed for DNA synthesis during proliferation (Descours et al. 2012).

6 SAMHD1 Influences Innate Immunity

Expression levels of SAMHD1 in cells of the immune system can have a major impact on the ability to sense retroviral infection and mount an antiviral immune response. In the context of HIV-1 infection, immune cells differ in their capacity to respond to infection. Plasmacytoid dendritic cells detect viral nucleic acid through their expression of Toll-like receptors 7 and 9 to activate TLR signaling (Gilliet et al. 2008). They are a main component of the innate antiviral immune system and respond rapidly to produce a large amount of type I interferon, as well as other antiviral cytokines, and play an important role in linking innate and adaptive immune responses (Gilliet et al. 2008). In contrast, monocyte-derived dendritic cells are largely resistant to HIV-1 infection and produce very little type I interferon when exposed to HIV (Fonteneau et al. 2004; Manel et al. 2010). Monocyte-derived macrophages are naturally infected by HIV-1, but the inefficiency of infection is likely to allow these cells to evade detection and escape an effective antiviral immune response (Hrecka et al. 2011; Laguette et al. 2011). Vpx enhances infection of myeloid cells resulting in increased sensitivity of detecting HIV that triggers the innate antiviral response and may contribute to development of more effective adaptive immunity (Manel et al. 2010). Vpx may be required by HIV-2 to infect cells with low dNTP pools in order to compensate for a less efficient reverse transcriptase (Fujita et al. 2012). HIV-2 infection is much less pathogenic than HIV-1 infection as evidenced by lower viral loads, slower rates of CD4+ T cell decline, and delayed progression to clinical symptoms of immunodeficiency (Andersson et al. 2000; Marlink et al. 1994). Indeed, the vast majority of HIV-2 infected individuals are categorized as long-term non-progressors, who control viral loads in the absence of chemotherapy (de Silva et al. 2008; van der Loeff et al. 2010). It is likely that the tropism mediated by Vpx accounts substantially for the differences in the clinical outcomes of HIV-1 versus HIV-2 infection. Presumably, HIV-1 has not developed a strategy to counteract the antiviral activity of SAMHD1 because it is advantageous. Overall, the interplay between Vpx and SAMHD1 suggests that limiting HIV-1 infection of myeloid cells promotes viral persistence and escape from antiviral immune responses. Strategies designed to inhibit SAMHD1 expression and allow infection and activation of dendritic cells may increase HIV-1 detection and potentiate virus-specific adaptive immunity.

7 Conclusions

The accessory genes encoded by lentiviral genomes represent evolutionary adaptations that the virus relies on to enhance replication, and the study of accessory gene functions has revealed the complex interplay between viral and host factors that support or inhibit viral infection. Efforts to characterize the functional roles of lentiviral Vpr and Vpx have identified SAMHD1 as a host cell restriction factor that potently inhibits virus replication in non-cycling myeloid cells and quiescent CD4+ T cells. The enzymatic activity of SAMHD1 regulates cellular dNTP levels, such that key substrates for reverse transcription are below levels needed for efficient conversion of viral RNA to genomic DNA. Vpx counteracts the SAMHD1-mediated restriction by recruiting the protein to the CUL4-based E3 ubiquitin ligase complex, resulting in its degradation by the proteasome. Depletion of SAMHD1 by Vpx, or the absence of SAMHD1 expression in individuals with Aicardi–Goutières syndrome, allows accumulation of cytosolic DNA that triggers an innate immune response with elevated production of interferon. HIV-1 has lost the ability to antagonize SAMHD1, and the low level of myeloid cell infection may be a viral strategy to evade immune surveillance to promote virus persistence and dissemination. A better understanding of SAMHD1 and innate immune responses during HIV infection may drive the development of novel therapeutic interventions and more effective vaccine strategies.