The major obstacle to HIV-1 eradication is the establishment of a latent infection. In infected individuals, viral production is a dynamic process involving continuous rounds of infection of CD4+ T lymphocytes with rapid turnover of both free virus and virus-producing cells that have a half-life of 1–2 days 12. The decay curves of plasma viremia following antiretroviral treatment have shown that after an initial fast decay, that wipes out the majority of circulating viruses in 1–2 weeks, plasma virus declines at a lower rate 34. The half-life of this compartment was estimated to be 1–4 weeks, but the nature of the cellular reservoir responsible for the second phase in the decay curve is still unclear. These cells could be macrophages, which are less sensitive to the cytopathic effect of HIV-1 infection 5 and that once terminally differentiated have a turnover rate of approximately 2 weeks. In addition, cellular reservoirs for HIV-1 could also be CD4+ T lymphocytes not fully activated, which carry the integrated provirus in a non-replicative state until the activation process is complete. Finally, dendritic cells (DCs) may also delay the release of infectious virus, since they are not permissive for HIV infection but can carry the virus trapped on their surfaces 6.

After two months on HAART the plasma levels of genomic RNA falls below the limit of detection in most previously untreated patients. Therefore, it was initially assumed that prolonged treatment might lead to eradication of the virus in these patients 3. Unfortunately, it is now clear that long-lived reservoirs of HIV-1 can persist for years in the presence of HAART. Although certain tissues like the male urogenital tract or the central nervous system might preserve infectious virus 78 the reservoir that appears to be the major barrier to eradication is composed of latently infected resting memory CD4+ T cells that carry an integrated provirus that is transcriptionally silent 910. The extremely long half-life of these cells, combined with a tight control of HIV-1 expression, make this reservoir ideally suited to maintain hidden copies of the virus, which are in turn able to trigger a novel systemic infection upon discontinuation of therapy. Given the importance of this reservoir, a lot of effort has been invested to characterize these cells from infected patients. These studies will be discussed in the following chapters, which will also address the problem of choosing appropriate model systems to thoroughly characterize the molecular determinants that allow the provirus to remain silent. Such mechanisms are mostly related to transcriptional control of viral expression and they depend both on the host cells and the virus. Finally, some ideas on how to approach viral eradication in the light of these novel findings will be presented in the final chapter.

HIV-1 exploits different strategies to persist within infected individuals. In CD4+ T lymphocytes, the replicative state of the virus is dependent upon the cell cycle of the host cell. Whereas HIV-1 entry into activated CD4+ lymphocytes leads to a productive infection 11, the virus encounter several blocks prior to integration in resting CD4+ lymphocytes 12. Such post-entry blocks have been proposed to result from a delay in completing reverse transcription due to low nucleotide pools and to the inability to import the pre-integration complex into the nucleus 13141516. Most recently the anti-retroviral deoxycytidine deaminase APOBEC3G has been shown to strongly protect unstimulated peripheral blood CD4+ T cells against HIV-1 infection 17. Furthermore, in Old World primates, TRIM5α, a component of cytoplasmic bodies, confers a potent block to human immunodeficiency virus type 1 (HIV-1) infection that acts after virus entry into cells, probably at the level of capsid processing 18. While these blocks delay the production of progeny virus following the infection of CD4+ resting T cells, mitogenic stimuli are able to trigger viral replication and release of infectious virus 1314192021. Although one would expect that activation of the cell per se would allow more efficient reverse transcription and nuclear import, by increasing the pools of DNA precursors and the available ATP used for the active mobilization of the PIC, still it is possible that specific blocks must be removed to allow full recovery of HIV-1 infectivity. In the case of APOBEC3G for example, activation of resting T cells induces the shift of the active low-molecular-mass form of APOBEC3G to an inactive high-molecular-mass complex unable to restrict viral infection 17. Other types of replication blocks that act after provirus integration in resting T cells, like for example the inhibition of NF-κB activity by Murr1 22, will be discussed in the following chapters.

Regardless of the kind of restriction imposed by the resting T cells on viral replication, this reservoir of pre-integrated latent virus is relatively labile persisting for weeks and cannot be accounted for the long-term latency observed during HAART.

In addition to CD4+ T lymphocytes, dendritic cells and macrophages are considered reservoirs for HIV-1 infection, but information on the replicative state of the virus within these cells is limited. DCs capture and internalize extracellular virions via the DC-SIGN lectin. Captured virions can subsequently be transmitted to T cells in trans 6. However, DC-SIGN does not significantly protect captured virions against degradation, leading to loss of infectivity within several hours 23. In HIV-1-infected monocyte-derived macrophages, mature viral particles can be observed within late endosomes 24. Virions found within monocyte-derived macrophages persist and retain infectivity for weeks, thus providing an additional mechanism for viral persistence 25. HIV-1 hidden in DCs and macrophages certainly plays an important role for viral spread and cell-cell transmission, but its involvement in long-term viral persistence has yet to be demonstrated.

A more stable form of latency occurs in CD4+ T cells that carry an integrated provirus. In principle, since integration requires T cell activation to allow efficient reverse-transcription and nuclear import of the pre-integration complex, post-integration latency can result only from the return of an infected activated T cell to a quiescent state. Evidence to this model has come from studies from Siliciano and co-workers who demonstrated the existence of resting memory CD4+ T cells carrying an integrated provirus in vivo 9. The phenotype of these resting cells carrying a non-productive HIV-1 infection, derived from peripheral blood of patients undergoing antiretroviral therapy with low to undetectable viremia, indicates that they derive from infected CD4+ lymphoblasts that have reverted to a resting memory state. These cells show a specific set of surface markers (such as CD4+, CD25-, CD69-, HLA-DR-) and are positive for the integrated provirus. Most importantly, upon mitogen stimulation, infectious virus can be recovered from these cells, indeed demonstrating that they represent a true, inducible viral reservoir. Hence, in vivo, it appears that HIV-1 gets trapped in T cells that revert to a resting memory state. In some respect long term HIV-1 persistence reflects directly long-term T-cell memory. At any given time, most CD4+ T lymphocytes in the body are in a resting G0 state. In response to antigens, resting T cells undergo a burst of cellular proliferation and differentiation, giving rise to effector cells. Most effector cells die quickly, but a subset survives and reverts to a resting G0 state (contraction phase). These lymphocytes persist as memory cells, with an altered pattern of gene expression enabling long-term survival and rapid responses to the relevant antigen in the future. Activated CD4+ cells are highly susceptible to HIV-1 infection and typically die quickly as a result of the cytopathic effects either of the virus or of the host immune response. However, some activated CD4 cells may become infected and then survive long enough to revert back to a resting state. Unfortunately, our current understanding of the decisive factors that determine if a CD4+ T cells will die or become a memory cell, as well as those that allow the self-renewal of resting memory cells for a lifetime, is still largely incomplete.

Despite the great wealth of information on the regulation of HIV-1 transcription, the crucial molecular events that control maintenance of the quiescent state in resting T cells remain elusive. Part of the problem depends on the lack of an appropriate model system. Most cell lines carrying an integrated quiescent provirus have been derived from transformed lymphoblasts that have been infected ex vivo with HIV-1. After an initial burst of viral replication and cell death, a population of non-productive cells carrying an integrated provirus remains. Establishment of latency in these cell lines is driven by selection of cells that resist viral replication and has been linked to mutation in viral genes, to certain cellular proteins and to the site of integration 26272829. Alternatively, T cells can be transduced with an HIV-1 vector carrying Tat and a reporter gene. This method has allowed the characterization of the integration status irrespective of the replication of the virus and has provided useful insights on the status of the integrated provirus. However, the constantly activated and proliferating nature of these cells, either infected or transduced, does not accurately represent the quiescent cellular environment of latently infected cells in vivo. A convenient animal model that recapitulates HIV-1 latency does not exist. In fact, several blocks to HIV-1 infection in mice greatly impair the development of an animal model to study HIV-1 infection amenable to genetic manipulation. One possibility would be the use of the SCID-hu (Thy/Liv) mouse that carries a source of human hematopoietic progenitor cells and human fetal thymus to provide a microenvironment for human T cells lymphopoiesis. Latently HIV-1 infected CD4+ T cells can be obtained in this model, but the exact extent to which it can be applied to natural infection is not known 30. SIV macaque models of AIDS are well established and have been extremely useful in HIV-1 vaccine development and in advancing the understanding of the pathogenesis of AIDS. The first report exploiting the SIV-macaque model to study viral infection in the course of antiretroviral therapy showed persistence of the virus in resting CD4+ T cells with many similarities to the human situation 31. This study provides initial evidence for the utility of a closely related retroviral infection for the analysis of HIV-1 persistence during antiretroviral treatment. However, the complexity of the protocol, which also requires antiretroviral drugs especially designed to control SIV infection, makes this model impractical if only for the preclinical evaluation of novel strategies to target viral reservoirs.

Since the HIV-1 provirus is found integrated into the host genome, regulation of viral gene expression depends on the chromatin environment at the site of integration and on the interaction of the viral Tat trans-activator with host factors. Clearly, multiple mechanisms could concur in this process.

Although the implementation of HAART has improved the survival and quality of life of HIV-infected individuals, HIV cannot yet be eradicated from infected individuals. Several studies have demonstrated that in individuals receiving HAART, the frequency of HIV-infected cells is reduced to fewer than one cell per 10⁶ resting CD4+ T cells 108990. However, even after years with viremia below the limit of quantification, the frequency of these infected cells does not decrease further. The therapeutic approaches evaluated to date have failed to demonstrate a significant and persistent decline of this latent viral reservoir 91, which appears small but stable and contains both wild-type and drug-resistant viral species 92. Several studies have shown that intensive antiretroviral therapy in combination with interleukin-2 or global T cell activators fails to eradicate HIV-1 infection. Global T cell activation may instead induce viral replication and increase the number of susceptible uninfected target cells beyond the threshold that can be contained by antiretroviral therapy 93. Following this approach, a strategy that selectively activates quiescent proviral genomes with limited effects on the host cell exploited the properties of the phorbol ester Prostratin or the human cytokine interleukin-7 that have been reported to reactivate latent HIV-1 in the absence of cellular proliferation 30949596. Another promising agent that has been proposed is the histone-deacetylase inhibitor Valproic acid capable of inducing outgrowth of HIV-1 from resting CD4+ cells of aviremic patients without full activation of the cells from the quiescent state 97. Such treatments, in combination with antiretroviral therapy, should allow outgrowth of latent HIV-1 but avoid the pitfalls of global T cell activation.

Another approach would be to target and destroy CD4+ memory cells. A study has been conducted ex vivo with an anti-CD45RO ricin immunotoxin to decrease the number of latently infected CD4+ T cells obtained from HIV-infected individuals without detectable plasma viremia 98. Such treatment significantly reduced the frequency of CD4+ memory cells with only a modest effect on the memory responses of CD8+ T cells. Therefore, purging latent cells from infected individuals on highly active antiretroviral therapy might reduce the HIV latent reservoir without seriously compromising CD8+ T cell memory responses. However, this kind of approach targets both infected and non-infected resting T cells and would severely deplete the immunological memory of the patient.

Recent advances have identified a long-lived stable reservoir of HIV-1 in patients on effective antiretroviral therapy that could potentially persist for life. This reservoir consists of a small pool of resting CD4+ T cells carrying an integrated provirus that somehow control viral expression allowing the virus to remain undetectable in plasma (Figure 1). Discontinuation of therapy and activation of cells results in production of infectious virus leading to a novel systemic infection. Hence, elimination of this reservoir by novel therapeutic approaches will be required before eradication can be achieved. Our knowledge of the cellular and molecular mechanisms behind the establishment of this remarkably stable reservoir will depend on appropriate in vitro model systems and in vivo animal models. These should allow the dissection of the pathways leading to transcriptional control of HIV-1 replication. In particular it will be crucial to correlate the activation state of host cells with the transcription state of the provirus. Key elements to be studied are: the site of provirus integration, which is determined by the viral integrase, and the mechanisms that control gene expression at these chromatin loci, which depend mostly on the activity of the Tat transactivator.

It is time to seriously consider alternative therapies that take into account the aim of eradicating infectious virus from the patients. In order to do so an effort is needed toward the understanding of the intimate interplay that is established at the molecular level between the virus and the host cell. From such studies it will be possible to devise novel therapies that selectively target the hidden virus.

HIV-1, human immunodeficiency virus type 1; HAART, highly active antiretroviral therapy; CD4/CD3/CD28, cluster of differentiation 4, 3, 28; IL-2, interleukin 2; PHA, phytohemagglutinin; PIC, pre-integration complex; IN, integrase; HMG, high mobility group; BAF, barrier to autointegration factor; INI1, integrase interactor 1; LEDGF, lens epithelium-derived growth factor; P/CAF, p300/CBP associated factor; CBP, CREB binding protein; CDK9, cyclin-dependent kinase 9; TPA, 12-O-tetradecanoyl-phorbol-13-acetate; LTR, long terminal repeat TAR, trans-activating response region; SCID, Severe Combined Immune Deficiency; PML, promyelocytic leukemia; PTGS, post-transcriptional gene silencing; RISC, RNA-induced silencing complex; RITS, RNA-induced transcriptional silencing; siRNA, small-inhibitory RNA.

The author declares that he has no competing interests.