The importance of monocytes and macrophages in HIV pathogenesis, treatment, and cure

Abstract

Monocytes and macrophages play critical roles in HIV transmission, viral spread early in infection, and as a reservoir of virus throughout infection. There has been a recent resurgence of interest in the biology of monocyte subsets and macrophages and their role in HIV pathogenesis, partly fuelled by efforts to understand difficulties in achieving HIV eradication. This article examines the importance of monocyte subsets and tissue macrophages in HIV pathogenesis. Additionally, we will review the role of monocytes and macrophages in the development of serious non-AIDS events including cardiovascular disease and neurocognitive impairment, their significance in viral persistence, and how these cells represent an important obstacle to achieving HIV eradication.

Pathogenesis

Myeloid lineage cells are targets of HIV infection

By harboring HIV long term without cytolysis [1,2], monocytes and macrophages, compared with CD4⁺ T cells, provide a reservoir of HIV [3,4]. In the circulation, phenotypically unique subpopulations of monocytes exist that are distinguished based on size, granularity, and expression of surface markers [1,2,5]. In healthy people, cells expressing the lipopolysaccharide (LPS) receptor CD14 but not CD16 (classical CD14⁺⁺CD16⁻ monocytes) make up 80–90% of monocytes [5] (Table 1) [6–8]. Inflammation and peripheral immune activation during HIV infection are associated with increased numbers of circulating CD14⁺⁺CD16⁺ (intermediate) and CD14⁺CD16⁺⁺ (nonclassical) monocytes [9,10], and the degree of CD16⁺ monocyte expansion is closely linked to the rate of disease progression [11]. Then they can reach 10–50% of circulating monocytes depending upon disease state and antiretroviral therapy (ART) status [9–12] (Table 1).

Table 1.

Changes in the percentage of classical, intermediate, and nonclassical monocyte subpopulations under various conditions.

	Classical CD14++ CD16− monocytes	Intermediate CD14++ CD16+ monocytes	Nonclassical CD14+ CD16++ monocytes
Healthy individuals [6,8]	80–90%	11.2–28.2%	3.5–9.9%
Uncontrolled viremia [6] (viral load >400 copies/ml)		38.8% (29.6–59.8%)	17.1% (13.0–27.5%)
Controlled viremia [6,13] (viral load <400 copies/ml)	71.1%	20.7% (15.6%–28.2%)	9.3% (6.0–15.2%)
ART-treated individuals [7] (<3 years of therapy)		23.4% (18.6%–35.9%)	10.0% (7.5–14.3%)

Historically, monocytes have not been considered to be a significant target of HIV infection [14,15]. Early studies indicate that even though they are activated and express HLA-DR, infection is dependent on their differentiation into macrophages [16,17]. However, more recent literature demonstrates that monocytes, particularly the CD16⁺ intermediate and CD16⁺⁺ nonclassical cells, can be infected by HIV [18,19]. Of relevance, monocytes expressing CD16 are thought to have a more mature immune phenotype consistent with macrophages, which may partly account for their ability to become infected.

Relative to CD4⁺ T lymphocytes, monocytes and macrophages are more resistant to productive HIV infection as a result of maturation and activation-dependent host factors. Differential expression of host restriction factors including SAMHD1 (sterile motif and histidine aspartic domain and HD domain-containing protein 1), viperin [virus inhibitory protein, endoplasmic reticulum-associated, interferon (IFN)-inducible)], and members of the APOBEC (apolipoprotein B mRNA editing enzyme) 3G family can affect productive infection of monocytes and macrophages [20–23] (Fig. 1). Classical monocytes, in particular, express the low-molecular-weight forms of APOBEC3A and APOBEC3G that block HIV reverse transcription [18,24]. In contrast, intermediate and nonclassical monocytes express higher-molecular-weight forms of these enzymes that support HIV replication [18,21,24].

Fig. 1. Monocytes and macrophages are important targets of HIV.

Variable levels of expression of HIV co-receptors, as well as intracellular inhibitory factors, are responsible for differences in HIV infection and replication between macrophages and monocytes, and between monocyte subsets.

By hydrolyzing cellular deoxyribonucleotides and reducing the pool of nucleotides available for reverse transcription [25], SAMHD1 expression restricts viral replication in monocytes and macrophages [26]; cells lacking SAMHD1 are highly susceptible to infection [20]. Mechanisms underlying the antiviral activity of viperin remain unclear; however, viperin levels are increased upon HIV infection of monocytes and macrophages and inhibit production of infectious virus [22]. Recent studies implicate naturally occurring cellular microRNAs (miRNAs-28, -150, -223, -382) in restricting monocyte infection [27]. Chemokines and cytokines in the surrounding environment can also dictate expression of cellular restriction factors in monocytes and macrophages (reviewed in [28]), and influence their levels of productive HIV replication [29].

Monocytes and macrophages as viral reservoirs

Monocytes and macrophages are in anatomic reservoirs including tissues such as the brain and lung and can persist by avoiding immune system detection. In the central nervous system (CNS), parenchymal microglia, meningeal macrophages, choroid plexus macrophages, and perivascular macrophages all express viral co-receptors and are susceptible to HIV infection [30,31]. Perivascular macrophages are the most consistent targets of HIV in the brain [32], and viral DNA can be isolated from these cells throughout infection, indicating that they are viral reservoirs [33].

A recent study in simian imunodeficiency virus (SIV)-infected rhesus macaques underscores the importance of lung alveolar and interstitial macrophages in local viral infection [34]. Alveolar macrophages are a primary viral target in the lung [35]; interstitial macrophages that are repeatedly renewed from monocytes can also be infected [34]. Pneumocystis jirovecii coinfection augments viral infection and macrophage activation in the lung [36,37].

In the gastrointestinal tract (GIT) of ART-naive individuals, high levels of chemokines/cytokines (e.g. CCL2, IL1β, CCL5, CXCL9, CXCL10) are produced and activated mucosal macrophages with poor phagocytic activity accumulate [38]. Unlike the CNS and lungs, GIT macrophages downregulate CD4 and CCR5 in response to HIV [39]. Though this remains an open question, downregulation of CD4 and CCR5 may in part result in reduced viral infection in these GIT macrophages.

HIV can also infect marginal zone macrophages in the spleen but conflicting data exist on whether these cells maintain a reservoir of virus [40].

The low frequencies of HIV-infected monocytes and the difficulty in obtaining tissue samples present major barriers to elucidating the dynamics of HIV infection and productive viral replication in human monocytes and macrophages. Experiments in SIV-infected macaques have provided critical insights into the role of monocytes and macrophages in viral persistence and maintenance of tissue reservoirs (reviewed in [41,42]). There are higher levels of SIV-infected monocytes and macrophages in monkeys than HIV-infected cells of this lineage in humans, in part reflecting differences in the primate lentiviral auxiliary protein x (Vpx) between SIV and HIV [26]. By inducing proteasomal degradation of the host restriction factor SAMHD1, Vpx increases monocyte and macrophage susceptibility to SIV infection [26].

In the absence of inflammation, monocytes continuously leave the bone marrow and circulate in blood for 3–5 days before trafficking into tissues [43]. With SIV infection, there is increased egress of classical, nonclassical, and intermediate monocyte subsets from the bone marrow, and these cells remain in the circulation for a short amount of time before entering tissues [44]. We have shown that monocytes from HIV-infected individuals have shortened telomeres. Because monocytes do not divide in the blood, this observation may reflect increased cell division and subsequent shortening of telomeres in bone marrow monocyte precursor cells [45]. These findings suggest that monocyte turnover and release from the bone marrow may also increase during HIV infection [45].

Upon CD4⁺ T-cell depletion during chronic SIV infection in macaques, macrophages comprise more than 95% of productively infected cells in lymph nodes; high numbers of SIV RNA+ macrophages are also evident in the CNS and lungs [46,47]. Although sequencing studies have found little difference in HIV-1 subspecies in the blood and lungs [48], there appears to be independent evolution of virus in other tissues and phylogenetically distinct transcripts are present in the plasma, spleen, lymph nodes, and CNS [49–52]. Although all viral strains (X4, R5, or T cell vs. macrophage-tropic) can enter the CNS of HIV and SIV-infected humans and monkeys, the bulk of the data demonstrate that macrophage-tropic viruses preferentially replicate in the CNS [49,53–55]. It is quite possible that CD4⁺ T cells and monocytes/macrophages can bring virus into the CNS early in infection; analysis of sequence evolution within the CNS has revealed that these macrophage-tropic viruses seed the brain throughout infection and are the major strains that replicate with disease [56,57].

The proportion of intermediate and nonclassical monocytes expressing CD16⁺ varies between individuals but is higher in ART-naive vs. individuals on ART, with these cells contributing significantly to disease pathogenesis [19,45,58]. The proportion of CD16⁺ monocytes in virologically suppressed individuals on ART is comparable with that in HIV-infected elite controllers [59]. Despite this, CD16⁺ monocytes are preferentially infected and recovered from blood even during sustained virologic suppression when compared with classical monocytes [18]. CD16⁺ monocytes have also been shown to harbor viral variants that are genetically distinct from sequences in resting CD4⁺ T cells [60]. Taken together, the contribution of monocytes to the viral reservoir may vary throughout infection and with therapy.

Role of monocytes and macrophages in opportunistic infections

With HIV infection, the reduced phagocytic ability of monocytes and macrophages contributes to increased host susceptibility to opportunistic pathogens [61,62]. Both Mycobacterium tuberculosis (MTb) and HIV infect alveolar macrophages and coinfection can augment production of tumor necrosis factor (TNF)α, IFN-γ, and CCL2 [63]. In some coinfected patients, tuberculosis (TB)-associated immune reconstitution inflammatory syndrome (TB-IRIS) occurs following ART initiation. It reflects a heightened inflammatory response to MTb, with new or worsening symptoms [64]. TB-IRIS may present in one of two clinical forms: paradoxical TB-IRIS, when symptoms of TB develop or recur during antituberculosis treatment following ART initiation, or unmasked TB-IRIS with an initial clinical presentation of TB, often with a strong inflammatory component, during early ART. Although the pathogenesis of TB-IRIS is not well understood, recent gene expression studies suggest that inherent differences in gene expression by monocytes contribute to the development of TB-IRIS and that monocytes mediate important activities during TB-IRIS [65,66].

Role of monocytes in neuroAIDS

In the ART era, the incidence of HIV-associated dementia has declined, yet the prevalence of milder forms of HIV-associated neurocognitive disorders (HANDs) continues to increase [67,68]. Importantly, development of HAND is associated with higher rates of AIDS-defining illnesses [69] and increased mortality [70]. The events leading to neuronal injury with HAND are not well understood; however, the presence of activated and productively infected macrophages and microglial cells are the best correlates of CNS disease severity [11,71–79] (Table 2). Recent data suggest that the event initiating macrophage activation and accumulation in the CNS is activation, expansion, and viral infection of monocytes trafficking from bone marrow [73–75,80,81]. Supporting this finding, elevated levels of cytokines [interleukin (IL)-6, IL-8, IFNγ] and proteins linked with monocyte activation (sCD14) and chemotaxis (MCP-1, CCL3, CXCL10) in the cerebrospinal fluid (CSF) of HIV-infected individuals on ART are associated with impaired neurocognition and the severity of neuropathology [76,82].

Table 2.

Monocyte-specific correlates of neurocognitive decline or neuronal injury in HIV infection.

Compartment	Measure of monocyte dysfunction associated with neurocognitive decline or neuronal injury
Brain	High HIV viral load [72]; increased perivascular CD14+/CD16+ mononuclear cells [76]
Cerebrospinal fluid	Elevated sCD14 [76,77]
Blood	Elevated CD16+ monocytes [11,73,74]; high HIV DNA levels in CD16+ monocytes [75]; elevated sCD14 [78,79]

Persistently increased numbers and/or percentages of CD16⁺ monocytes are more tightly linked to the development of neurological disease than the number of HIV-infected cells in the CNS or CSF viral load [75,83]. Infection of CD16⁺ monocytes increases the capacity of these cells to migrate across the blood-brain barrier (BBB) [84] wherein they can stimulate productive viral replication in the brain [80]. Monocytes expressing the CD16 receptor are immunophenotypically similar to perivascular macrophages in the CNS, in that the majority of CD16⁺ monocytes express the haptoglobin-hemoglobin scavenger receptor CD163, and all CD163⁺ perivascular macrophages in the CNS express CD16 [76,85,86]. Although HIV does not directly infect neurons, increased production of proinflammatory chemokines/cytokines and other excitotoxic factors by perivascular macrophages and activated microglia contribute to neuronal injury and death [87,88] (Fig. 2). Release of inflammatory chemokines/cytokines (MCP-1, CX3CL1, CXCL10, CCL3, IL-1β, TNFα, IFNα) within the CNS also stimulates up regulation of adhesion molecules on the brain microvascular endothelia [89,90], thereby recruiting monocytes to the BBB and promoting their transmigration into the brain [91].

Fig. 2. Role of monocytes and macrophages in the development of neurocognitive impairment during FI IV infection.

(a) Infection of CD16⁺ monocytes by HIV promotes transmigration into the brain where (b) productive infection allows for the evolution of CCR5/macrophage tropic virus. (c) Chemokines and cytokines produced by infected cells promote further transmigration and ongoing infection and (d) induce the neuronal damage that leads to neurocognitive decline.

It is generally assumed that HIV enters the brain within infected monocytes, but this has not been directly demonstrated. Evidence that HIV enters the CNS soon after peripheral infection was demonstrated by accidental iatrogenic infection [92]. In SIV-infected rhesus macaques, gliosis, glial nodules, and scattered productively infected cells are evident in frontal lobes, temporal lobes, and white matter shortly after infection [93,94]. Interestingly, after initial seeding of virus, SIV RNA and p27 are not found until evidence of AIDS, probably the result of late seeding of the CNS by macrophages [95].

Evidence suggests that monocytes dictate the timing and severity of neurological sequelae during HIV infection. In virologically suppressed HIV-infected individuals on ART, levels of sCD163 in the plasma are associated with neurocognitive decline, suggesting that persistent activation of CD16⁺ monocytes plays a role in CNS disease [96]. Furthermore, experiments in rhesus monkeys using the thymidine analog 5′-bromo-2′-deoxyuridine (BrdU) to monitor monocyte dynamics during infection have underscored the importance of monocytes in CNS disease. In SIV-infected animals, the magnitude of BrdU+ monocytes released from bone marrow is a more accurate indicator of the rate of AIDS progression than plasma virus or CD4⁺ T-cell count [44]. Rapid development and histopathologic severity of SIV encephalitis are associated with the degree of BrdU⁺ monocyte expansion during the first 4 weeks of infection [97]. Moreover, sCD163 in plasma correlates with the percentage of BrdU-labeled monocytes and CD16⁺ monocytes [97]. These findings suggest that innate immune activation early in infection is critical for determining overall disease outcomes. Finally, in ART-naive individuals, HIV DNA levels in CD16+ monocytes are closely linked to the development and severity of neuropathology. This observation implicates the size of the monocyte HIV reservoir in neuropathogenesis [98,99].

Role of monocytes in serious non-AIDS events

HIV-infected individuals are at increased risk for a range of comorbidities including cardiovascular disease (CVD), non-AIDS malignancies, neurocognitive disorders, renal, bone and liver dysfunction, and frailty [100]. These conditions, collectively known as serious non-AIDS events (SNAEs), now represent the major cause of mortality in virologically suppressed HIV-infected individuals [101]. Persistent HIV-related inflammation and innate immune activation, both incompletely restored by ART, are thought to underpin the increased risk of these diseases, rather than T-cell activation [102]. This is supported by high plasma levels of inflammatory markers (e.g., hsCRP and IL-6 [103]) and monocyte activation (sCD14) [104]) that predict all-cause mortality in HIV-infected individuals. Chronic low-level endotoxemia in HIV-infected individuals can result from microbial translocation across the gut mucosa [105]. Subsequent passage of microbial products via the portal vein into the liver can lead to monocyte activation, altered coagulation, and systemic inflammation. This eventually can result in end-organ damage and development of SNAEs.

Low bone mineral density (BMD; reviewed in [106]) and osteoporosis [107] are common in HIV-infected people. Although T-cell activation has been linked to low BMD [108], monocyte/macrophage activation has not been directly implicated [109]. HIV infection is also associated with an increased risk of frailty and a lower age of onset [110]. Markers of monocyte activation including neopterin and CXCL10 [111] are associated with frailty in uninfected persons, but data are limited in HIV-infected individuals.

Monocyte activation and cardiovascular disease in HIV infection

HIV infection is associated with an approximately two-fold relative risk of CVD [112], persisting after adjustment for traditional risk factors. Moreover, CVD risk prediction algorithms including the Framingham Risk Score may underestimate the degree of atherosclerosis during HIV infection, with 56.4% of HIV-infected individuals with a low estimated CVD risk having evidence of subclinical atherosclerosis [113]. The prevalence of atherosclerosis in this low-risk group was independently associated with levels of oxidized LDL and MCP-1 [113], which promote monocyte recruitment to the subendothelial space where atherosclerotic lesions develop (reviewed in [114]).

Measures of HIV disease progression including viral load and CD4⁺ T-cell count and markers of T-cell activation are poor predictors of CVD risk and events in HIV-infected individuals [115]. These clinical outcomes are increasingly being shown to correlate with inflammatory markers such as hsCRP, IL-6, and TNF, and monocyte activation [116–118].

Monocyte activation during HIV infection persists despite ART [45,58]. We and others have shown that monocyte expression of CD11b and CX3CR1 are associated with carotid intima–media thickness (cIMT; a surrogate measure of atherosclerosis) in HIV-infected individuals [119,120], whereas others demonstrate that monocyte activation phenotype in HIV-infected individuals is similar to uninfected individuals with CVD [6]. The proportion of inflammatory CD16⁺ monocytes is increased in untreated HIV infection and predicts progression of coronary artery calcium, independent of traditional risk factors [115].

Monocytes play a critical role in atherosclerosis; they migrate across activated endothelial cells into fatty streaks within the subendothelium, where they mature into macrophages and endocytose lipids (Fig. 3). Monocytes can egress from the subendothelial space and remove lipid to prevent plaque progression [121]; however under inflammatory conditions, they are more likely to be retained and develop into lipid-laden foam cells. Lipid uptake by foam cells results in necrosis, inflammation, and expansion of the atherosclerotic core, whereas increased matrix metalloproteinase activity by resident macrophages increases plaque instability and promotes rupture [122].

Fig. 3. A model for the role of monocytes in the development of atherosclerosis in HIV-infected individuals.

Migration of monocytes across the blood vessel endothelium and the development of foam cells is the initiating step in the development of atherosclerosis. Alternatively, these monocytes can migrate out of the intima, carrying pro-atherosclerotic LDL away from the vessel wall. During HIV infection, several alterations in monocyte dynamics promote migration of monocytes and foam cell formation, rather than reverse migration. oxLDL, oxidized low density lipoprotein.

The relationship between monocyte/macrophage activation and CVD is supported by biomarker data. In HIV-infected individuals, plasma sCD163 levels are associated with noncalcified coronary plaques [123,124] and arterial inflammation [125]. Increased levels of sCD14 and LPS are independently associated with increased cIMT, in some [126,127] although not all cohorts [128]. Elevated sCD14 is also independently associated with coronary artery calcification [129]. Importantly, the association between monocyte activation and CVD risk (Table 3) [113,115,118,120,123–125,127,129,130] observed in cohorts of ART-treated individuals is independent of protease inhibitor use (linked to increased CVD risk in HIV infection) [131]. The increased prevalence of atherosclerosis and monocyte activation (evidenced by increased sCD163 levels) in HIV elite controllers who maintain viral suppression without ART [132] highlights the critical role of immune activation in HIV-related CVD risk independent of viral and ART parameters.

Table 3.

Markers of monocyte activation, recruitment into plaques and foam cell formation are associated with radiological measures of cardiovascular disease.

Monocyte	Characteristics	Associated measure of cardiovascular disease
Subset	CD16+ monocytes	CAC progression [115]
Phenotype	CD11b expression	cIMT [120]
	CX3CR1 expression	cIMT [120]
Activation markers	sCD163 level	Noncalcified coronary plaque size [123,124]; arterial wall inflammation [125]
	sCD14 level	cIMT progression [127]; CAC [129]
Recruitment	MCP-1 level	cIMT [113]; CAC [118]
	oxidized LDL level	cIMT [113]
	sVCAM1 level	cIMT [130]

CAC, coronary artery calcium; cIMT, carotid intima–media thickness; LDL, low-density lipoprotein.

Despite these associations, mechanistic explanations for how monocytes contribute to CVD in HIV infection are lacking. We recently demonstrated that elevated numbers of CD163⁺ macrophages in the hearts of SIV-infected monkeys correlate with the severity of fibrosis and overall cardiac damage [133]. Further, we found an increased percentage of CD16⁺ monocytes by 8 days postinfection in animals that developed cardiac disease and fibrosis (K. Williams unpublished results). We have previously shown that HIV impairs the ability of monocytes to egress from an in-vitro plaque model [134], which may promote their retention in atherosclerotic lesions in vivo. Our recent findings indicate that monocytes from HIV-infected individuals have a heightened potential to form foam cells [135]. Further work is essential to elucidate the association of HIV-associated inflammation and immune activation with atherogenesis.

Treatment and cure

In the era of effective ART, the new challenges in HIV treatment involve ameliorating the increased risk of SNAEs and moving toward a ‘functional HIV cure’, meaning undetectable viremia and no evidence of disease progression in the absence of ART.

Targeting inflammation to prevent serious non-AIDS events

Therapeutic strategies to prevent co-morbidities such as CVD in HIV-infected individuals need to target the component of this risk driven by inflammation and monocyte activation. The ability of the anti-inflammatory statin rosuvastatin to improve cardiovascular and skeletal health in HIV infection by simultaneously targeting inflammation and dyslipidemia is currently being evaluated in the SATURN-HIV trial. Preliminary data indicate rosuvastatin can reduce markers of monocyte activation, including sCD14 and tissue factor expression on monocytes, independent of its lipid-lowering effects [7]. This activity is also associated with increased bone mineral density [109]. Thus, anti-inflammatory drugs may be able to ameliorate markers of inflammation and monocyte activation in ART-treated individuals. Whether this effect translates to a reduction in SNAEs will be an area of intense interest in coming years.

HIV eradication

Experiments involving the transplantation of HIV-resistant CCR5Δ32 stem cells into an HIV-infected individual (the ‘Berlin patient’) have demonstrated that a sterilizing cure for HIV infection may be possible; however, this approach is definitely not a universal option for cure [136]. In this regard, after a period of sustained virologic suppression, cessation of ART was associated with HIV rebound in two HIV-infected individuals who received bone marrow transplantation with wild-type CCR5 stem cells [137]. This observation indicated that latent infection in long-lived cells and tissues can persist despite bone marrow transplantation and long-term ART [138].

HIV-infected individuals may also be ‘functionally cured’ of HIV, meaning that although HIV DNA and RNA persist in cells and tissues, they maintain undetectable plasma viral loads without ART [139]. Despite low levels of HIV in the plasma of elite controllers, replication-competent virus can be found in resting CD4⁺ T cells [140] and infrequently in circulating monocytes [141]. Elucidating the molecular mechanisms by which replication-competent virus is suppressed in these cellular reservoirs will be critical to design a therapeutic strategy that might affect a functional cure.

Persistent infection in latently infected cells

In the vast majority of virologically suppressed HIV-infected individuals on ART, reservoirs of latently infected cells (including resting CD4⁺ T cells and CD16⁺ monocytes in the bone marrow, thymus, blood, brain, and other tissues) persist, containing replication-competent, transcriptionally silent, latent provirus. This represents a major barrier to the eradication of HIV [142–144] (Fig. 4).

Fig. 4. The relevance of monocytes to HIV treatment and cure strategies.

Overview of the barriers which limit the ability of antiretroviral therapy (ART) to target monocytes and macrophages in tissues, and potential strategies to address these.

Although latently infected CD4⁺ T cells, particularly central memory T cells, comprise the majority of the HIV reservoir, analysis of viral sequences isolated during episodic increases in viral load in ART-treated individuals suggests this virus is coming from cells other than CD4⁺ T cells [145]. Importantly, even in the presence of ART, activated CD16⁺ monocytes are capable of perpetuating HIV replication through ongoing cell-to-cell transfer of virions and efficient infection of CD4⁺ T cells [146,147].

Efficacy of antiretroviral therapy in targeting monocytes and macrophages

Intensification of ART using maraviroc and raltegravir can reduce the size of the peripheral latently infected CD4⁺ T-cell reservoir [148]; however, specific data on the impact of ART intensification on the monocyte/macrophage reservoir are lacking. Initiating therapy early after seroconversion reduces the number of latently infected CD4⁺ T cells in the blood, GIT [149], and brain [150] and decreases HIV DNA to levels that are similar to that in elite controllers [151]. These findings suggest that both intensifying treatment and initiating ART during early infection can decrease the overall size of the latent HIV reservoir.

The biologic characteristics of monocytes and macrophages, however, may reduce the efficacy of such strategies in these cells. HIV-infected monocytes are more resistant than CD4⁺ T cells to the effects of antiretroviral compounds, although newer classes of drugs may hold more promise. Early in-vitro work indicated that chronically infected monocytes/macrophages are less susceptible to the effects of the nucleoside reverse transcriptase inhibitor (NRTI) zidovudine than acutely infected CD4⁺ T cells [152]. Although protease inhibitors are effective in stopping the release of infectious virions from productively infected macrophages, these drugs are less effective against latent infection [153,154] and high concentrations of protease inhibitors are required to suppress HIV replication in these cells [155].

The CCR5 inhibitor maraviroc can reach high concentrations in the GIT and reduce the size of the gut reservoir [156]. Maraviroc can also directly prevent infection of monocyte-derived macrophages ex vivo, thus potentially impacting the size of reservoirs in these cells in vivo. However, specific studies in HIV-infected individuals are currently lacking [156].

The integrase inhibitor raltegravir is equally potent in macrophages as in lymphocytes [157] and can reach therapeutic concentrations in the CSF [158]. This result is advantageous for targeting infected cells including macrophages in the brain. However, although multiple mutations are required to confer raltegravir resistance in T cells, this can be achieved via a single mutation in macrophages [159].

Factors contributing to persistent inflammation and HIV infection in the CNS despite effective therapy include limited neuropenetrance of many antiretroviral drugs [160,161] and neurotoxicity of some antiviral drugs capable of penetrating the CNS [162,163]. They may potentiate inflammation and immune activation. Even when therapy is initiated during primary infection, CNS immune activation is still evident more than 4 years after effective ART [164] and 80–90% of virologically suppressed HIV-infected individuals at autopsy have prominent activated and productively infected macrophages in the CNS [165]. This finding suggests that neuropenetrance of ART may not be sufficient to prevent HIV-related neuronal damage and target macrophage reservoirs of HIV in the brain. Three recent studies indicate that the ability of an antiviral drug to target CNS macrophages is the most accurate indicator of its utility in treating clinical symptoms associated with CNS HIV infection [75,83,151]. These data, along with the fact that viral sequences from the brains of HIV-infected individuals have macrophage-tropic motifs [56,166], demonstrate that to be effective in the CNS the therapy needs to target monocytes and macrophages.

Strategies other than activation of latent T-cell reservoirs are needed to achieve HIV eradication

Recent research toward a cure for HIV has focused on activation of latently infected T cells to induce HIV transcription, followed by elimination of these infected cells (reviewed in [167]). Although a number of drugs such as the phorbol ester prostratin [168] and histone deacetylase inhibitors including vorinostat [169,170] may also activate latent infection in monocytes and macrophages, there is no defined strategy to subsequently reduce the size of the viral reservoir, as these cells are resistant to ART-induced cell death. Thus, more plausible biologic approaches are required, particularly those that target immune cells in addition to lymphocytes, including monocytes and macrophages. These tactics merit serious consideration as a strategy to eradicate HIV or to effect significant and sustained change on the size of the HIV reservoir. Although directly eliminating HIV-infected monocytes/macrophages would seem an attractive approach to eradicate viral reservoirs, to date there are no current approaches for this. But, immune therapies that target monocyte/macrophage activation or differentiation might be considered to slow or stop HIV-related diseases or pathogenesis.

Conclusion

HIV-infected monocytes and macrophages contribute to viral persistence throughout infection providing an important reservoir of HIV and perpetuating HIV replication through ongoing cell-to-cell transfer of virions. These cells also play a critical role as inflammatory mediators of HIV-associated CNS disease, CVD, and other SNAEs. Eradication of HIV is not possible with current therapy. Strategies toward a cure thus far have overlooked the contributions of non-T-cell reservoirs to ongoing viral persistence. The future challenges for both treatment and toward a cure require the development of innovative and more effective therapeutics that target persistent HIV in these additional cellular reservoirs.