New clues to understanding HIV nonprogressors: low cholesterol blocks HIV trans infection.

A small percentage of HIV-infected subjects (2 to 15%) are able to control disease progression for many years without antiretroviral therapy. Years of intense studies of virologic and immunologic mechanisms of disease control in such individuals yielded a number of possible host genes that could be responsible for the preservation of immune functions, from immune surveillance genes, chemokines, or their receptors to anti-HIV restriction factors. A recent mBio paper by Rappocciolo et al. (G. Rappocciolo, M. Jais, P. Piazza, T. A. Reinhart, S. J. Berendam, L. Garcia-Exposito, P. Gupta, and C. R. Rinaldo, mBio 5:e01031-13, 2014) describes another potential factor controlling disease progression: cholesterol levels in antigen-presenting cells. In this commentary, we provide a brief background of the role of cholesterol in HIV infection, discuss the results of the study by Rappocciolo et al., and present the implications of their findings.

COMMENTARY

Understanding the genetic basis of natural resistance to HIV-1 is a major goal in the effort to control HIV. Previous studies to identify and delineate host genetic variants responsible for complete or partial resistance to HIV infection or disease progression have pointed to host genes involved in the HIV-1 replication cycle (CCR5, CCR2b, chemokines), immune surveillance (major histocompatibility complex class I) or restriction factors (members of the apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like, family; tripartite motif-containing protein 5α; and tetherin) (recently reviewed in reference 1). Such studies are valuable, as they provide important clues to identifying the viral and host targets for developing more efficacious therapies. The recent report in mBio by Rappocciolo et al. (2) throws new light on a molecule whose role in aiding and abetting HIV replication is already known—cholesterol. That study shows that the level of cholesterol in immune cells may be a determinant of HIV spread within the body. While the underlying genetic basis of this is still unknown, the study provides some important hints.

Cholesterol has been shown to be required for the entry or fusion of many different viruses (3–5), and the list continues to expand (6). The role of cholesterol in HIV replication has been studied for over 10 years. Early reports demonstrated that cholesterol depletion from virus-producing cells suppresses virus production (7), whereas depletion of cholesterol from mature virions or target cells inhibits virus-cell fusion and infection (8–11). This dependence of HIV on host cell cholesterol was related to viral assembly on cholesterol-rich lipid rafts of the plasma membrane and the role of cholesterol and lipid rafts in membrane fluidity (7, 12, 13). Thus, reports that HIV actively controls the cholesterol metabolism of the host cell came as little surprise. HIV, via its protein Nef, stimulates cholesterol uptake and biosynthesis by activating the transcription of sterol-responsive element binding factor 2 (SREBF-2) and SREBF-2-regulated genes (14), and Nef also inhibits the activity of the cellular cholesterol transporter ATP-binding cassette A1 (ABCA1), thus reducing cholesterol efflux from cells (15). In addition, Nef binds cholesterol and delivers it to lipid rafts (16). Together, these effects lead to an increase in intracellular cholesterol, an increase in lipid raft abundance, and an increase in viral production and infectivity (17). Conversely, depletion of cellular cholesterol by ABCA1 activation potently inhibits HIV replication (18–20).

The paper by Rappocciolo and colleagues describes another role that cholesterol plays in HIV infection (2). Investigating the mechanistic basis underlying the ability of a small group of HIV-infected persons to control HIV infection for many years without antiretroviral treatment, the authors compared antigen-presenting cells (APCs) from nonprogressors (NPs) and progressors (PRs) for the ability to trans infect susceptible CD4+ T cells. Trans infection is the process by which APCs such as dendritic cells (DCs) or certain B cells can take up HIV-1 and participate in enabling the infection of CD4+ T cells (21). It is different from the standard cell-to-cell infection (cis infection) in that the APCs are not productively infected by HIV, so the virus they transduce to target cells is the virus they have captured and preserved. However, trans infection is similar to cis infection in that the target cells must express CD4 and coreceptor and the formation of a virological synapse is involved. Previous study by Viglianti’s group demonstrated that trans infection by DCs depends on cholesterol levels and can be suppressed or stimulated by manipulating cellular cholesterol content via stimulation of the nuclear receptors LXR and PPARγ or the targeted knockdown of ABCA1, respectively (22). Rappocciolo et al. found that APCs (DCs and B cells) from NPs, when pulsed with HIV in vitro, did not trans infect autologous or heterologous CD4+ T cells, whereas cells from PRs efficiently transferred the virus to susceptible cells. Importantly, the defect was localized to APCs, as T cells from both NPs and PRs were equally efficiently infected by cell-free virus.

In the authors’ search for a mechanistic basis for this finding, they made a startling observation—the ability of APCs to trans infect CD4+ T cells was directly correlated with APC cholesterol content. In other words, cells from NPs had lower cholesterol than cells from PRs. Replenishment of cholesterol in APCs from NPs reversed the block to trans infection, and depletion of cholesterol or inhibition of cholesterol biosynthesis with statins blocked the ability of APCs from PRs to transfer HIV to susceptible cells. Of note, the difference in APC cholesterol content between PRs and NPs is likely due, at least in part, to genetic or epigenetic regulation, as ABCA1 expression was higher in cells from NPs. These findings support a previously proposed role for ABCA1 as an innate anti-HIV-1 (and probably a more general antiviral) factor (17). In this case, the decreased cholesterol levels controlled by ABCA1 affect the ability of DCs and B cells to capture HIV and transfer it to susceptible cells in a process that requires tight cell-to-cell contact to form a virological synapse (Fig. 1). Jolly and Sattentau previously showed that the formation of virological synapses between HIV-infected and uninfected T cells requires the integrity of lipid rafts (23). It is likely that these are the same membrane structures as those involved in the assembly of infectious HIV-1 particles. Thus, the mechanism of anti-HIV-1 activity of ABCA1 may be the same in all cells and comes down to reduction of lipid raft abundance. The authors examined the ability of B cells, isolated from PRs and NPs both before and after seroconversion, to participate in trans infection and found that the inability to trans infect existed prior to seroconversion, thus suggesting that the inability of APC to trans infect is an innate property of NPs. The specific genetic or epigenetic mechanisms underlying the high ABCA1 expression in APCs from nonprogressors remain to be uncovered. It is not unlikely that this is due to single nucleotide polymorphism in ABCA1 or genes encoding transcription factors that specifically regulate ABCA1 expression in APCs.

FIG 1 

Schematic depicting the role of ABCA1 expression levels in APCs during HIV-1 trans infection. Individuals with genetically determined high levels of ABCA1 gene expression have lower cholesterol levels than normal subjects in cells expressing ABCA1, such as DCs, whereas the cholesterol levels in cells where ABCA1 is not expressed, such as T cells, are similar in these two groups. The paper by Rappocciolo et al. suggests that HIV disease in individuals with high ABCA1 expression progresses more slowly than in normal individuals. They demonstrate an absence of trans infection of CD4+ T cells by DCs and B cells derived from NPs (individuals with high ABCA1 expression levels). Cholesterol depletion has been shown to inhibit HIV uptake by DCs. Therefore, high ABCA1 expression is expected to reduce HIV uptake and consequently trans infection. Trans infection proceeds through a virological synapse, which is formed with the participation of lipid rafts. Lipid rafts are reduced when ABCA1 expression increases. See text for details.

Results of this study, while provocative and exciting, are only the first step in understanding the mechanisms that operate in the ability of some individuals to control HIV replication. It is most likely that several mechanisms that engage different molecular players are involved, and careful studies are needed to pinpoint the contribution of cholesterol to each mechanism. It will be important to determine whether cholesterol levels of APCs can explain the differences between elite controllers (undetectable viral loads), viremic controllers (plasma HIV-1 levels of 50 to 2,000 copies/ml), and viremic nonprogressors (variable viral loads but stable CD4+ T cell counts). Although the sample size in the study by Rappocciolo et al. was small (n = 8), it is striking that the APCs from all 8 NP individuals were unable to effect trans infection.

Some questions remain unanswered. The authors demonstrated the dependence of trans infection on cholesterol by repleting APCs with cholesterol prior to infection. However, since repletion was performed prior to loading of the APCs with HIV, it is unclear if cholesterol is required for the process of HIV-1 capture by APCs, for the actual trans infection process, or for both. Trans infection of CD4+ T cells by APCs, in particular DCs, is known to play a significant role in the mucosal transmission of HIV (24), but its contribution to the posttransmission stage of disease progression remains unknown. It is possible that B cells play a role at this step (25), but evidence for this remains scarce. It is also intriguing that the cholesterol levels in the recipient CD4+ T cells did not matter; rather, the cholesterol levels in the APCs controlled trans infection. This conclusion was based on experiments where the authors used 20% serum to culture T cells; given low ABCA1 expression in T cells, such culture conditions could overload the cells with cholesterol, masking differences between cells from PRs and those from NPs. High ABCA1 expression in nonprogressors may also be responsible for the restricted infection of macrophages and other myeloid cells that are productively infected with HIV. This may limit the viral reservoir (26), contributing to nonprogression. In any case, the possibility that levels of ABCA1 expression may determine the control of HIV infection deserves further study. The implications and translational opportunities are colossal, from preexposure prophylaxis to microbicides to clinical maintenance of HIV-infected subjects.