Human Immunodeficiency Virus (HIV) disease progression to acquired immunodeficiency syndrome (AIDS) is impeded by antiretroviral drugs (ARVs) that inhibit critical steps in the HIV life cycle [1]. However, ARVs do not target or eliminate the viral genome integrated into the cellular genome. Once integration occurs, curing people becomes extremely difficult because a small percentage of HIV-infected cells remain dormant (latent) or suppressed at specific anatomical sites throughout the body. Within these reservoirs, HIV-infected cells can be suppressed by ARVs, persistent or latent. HIV-persistent cells produce few virions, and cells latently infected do not produce new virions but can produce viral transcripts and proteins or remain transcriptionally silent [2]. However, once reactivated or upon ARV treatment interruption, these cells can rebound HIV viral load to detectable levels within two weeks [3].

2. Molecular Mechanisms of HIV Latency

At the beginning of HIV cellular infection, HIV glycoprotein (gp120) binds to the host CD4 receptor. HIV can then enter the cell in two common pathways. In the primary pathway, gp120 undergoes a conformational change and exposes the gp120’s V3 domain. The V3 domain binds to the host chemokine coreceptors, such as CCR5 or CXCR4, depending on the viral tropism caused by a point mutation ^[4][11]. A post-conformational change dissociates gp120 from gp41, which facilitates fusion with the host membrane ^[5][12]. The second pathway involves endocytosis. Upon gp120 binding to the CD4+ receptor, the virus enters the cell via an endocytic vesicle ^[6][13]. Chauhal et al. demonstrated that HIV-1 enters astrocytes via the endocytosis pathway in which gp41 facilitates fusion with the luminal membrane of the endosome. The endocytosis pathway has been reported to produce more defective viruses ^[7][8][9][14,15,16]. Once fusion occurs in both pathways, the capsid is released into the cytosol. Cells latently infected with HIV establish reservoirs at specific anatomical sites such as the brain ^[10][25]. At these reservoirs, two types of latency exist, pre-integration and post-integration (Figure 1). Pre-integration latency primarily occurs when the host cells transition to a quiescent state before HIV viral DNA is integrated. The mechanisms of pre-integration latency include poor nuclear transportation of the PIC, defective viral proteins, and the state of the cell life cycle ^[11][22]. In addition, the viral PIC can remain stable for several weeks on centrosomes and possibly integrate into the host genome when the host cell becomes reactivated ^[12][26]. Pre-integration latency is less stable than post-integration latency. Post-integration latency occurs after the viral genome has integrated into the host genome and remains transcriptionally silenced ^[13][27]. Memory CD4+ T-cells transitioning from an active to a quiescent state also exhibit post-integration latency. Several factors can contribute to this state, such as low transcriptional factors, DNA modification, transcriptional suppression, and low Tat and P-TEFb (Figure 1). Post-integration latency can depend on the expression threshold of the HIV Tat protein, which forms a positive feedback loop that upregulates HIV transcription. When Tat expression is above the threshold, transcription can occur. However, when Tat expression is below the threshold, the provirus is considered dormant/latent. As a result, HIV LTR transcription is significantly reduced/inhibited. Several factors can contribute to the low expression of Tat, such as low levels of host transcriptional factors, the presence of host transcriptional repressors, and the integration of the HIV genome into a heterochromatin region ^[14][28]. For example, in microglia, Sp3 (a host transcriptional repressor) is regulated and induces latency by blocking HIV transcription. However, latency in astrocytes is induced by low expression of Sam68 or by high expression of cellular proteins with Rev-interacting domain (Risp), class I histone deacetylases (HDACs), and a lysine-specific histone methyltransferase, Su(var)3–9 ^[15][16][17][29,30,31]. Additionally, the microenvironment of the cell can also determine latency. Zhuang et al. observed that under hypoxic conditions, hypoxia-inducible factors could restrict HIV transcription by binding to the viral promoter regions ^[18][32]. There are multiple types of HIV infections including latent, active, and persistent (Figure 2). Persistent infection is believed to result from low ARV concentrations in anatomical HIV reservoirs. Within these reservoirs, HIV-infected cells can have transcriptionally active proviruses that are persistent (producing little virions) or latent (producing no virions) ^[19][33]. Moreover, infected cells can contain replication-competent or defective proviruses ^[20][34]. Replication-competent proviruses produce virions with the ability to replicate. On the other hand, defective proviruses produce imperfect proteins that prevent the virions from reproducing. Approximately 5–11% of CD4 T-cells within the reservoirs are replication-competent ^[21][35]. These reservoirs are established throughout the body where ARVs are limited ^[22][36] (Figure 3). In some unique microenvironments such as the CNS, compartmentalization can occur, resulting in HIV variants. In addition, infected cells can also undergo normal cellular division (clonal expansion), which replication-competent proviruses can increase in number ^[23][37]. It is estimated that more than half of the latent cells are maintained by clonal expansion ^[24][25][38,39]. Antigen-driven proliferation is considered one of the forces that drive clonal expansion. The location of these replication-competent cells needs to be identified to develop an effective strategy to prevent viral rebound. Specific tissues within the body serve as reservoirs and possess persistent and latently infected cells, described below.