Dissecting How CD4 T Cells Are Lost During HIV Infection

Abstract

Although the replicative life cycle of HIV within CD4 T cells is understood in molecular detail, less is known about how this human retrovirus promotes the loss of CD4 T lymphocytes. It is this cell death process that drives clinical progression to acquired immune deficiency syndrome (AIDS). Recent studies have highlighted how abortive infection of resting and thus nonpermissive CD4 T cells in lymphoid tissues triggers a lethal innate immune response against the incomplete DNA products generated by inefficient viral reverse transcription in these cells. Sensing of these DNA fragments results in pyroptosis, a highly inflammatory form of programmed cell death, that potentially further perpetuates chronic inflammation and immune activation. As discussed here, these studies cast CD4 T cell death during HIV infection in a different light. Further, they identify drug targets that may be exploited to both block CD4 T cell demise and the chronic inflammatory response generated during pyroptosis.

Figure 1. Roles of Caspase-3-Dependent Apoptosis and Caspase-1-Dependent Pyroptosis in CD4 T Death during HIV Infection

HIV-1 infection in a biologically relevant human lymphoid aggregate culture (HLAC) system reveals that the permissivity status of target CD4 T cells dictates how they die. If the virus infects an activated and thus permissive cell (i.e., 5% of the total CD4 T cells), productive infection ensues, and the cell dies from a silent caspase-3-mediated apoptosis. Conversely, if the virus infects a resting, nonpermissive cell (i.e., >95% of target CD4 T cells), abortive infection occurs, leading to the accumulation incomplete cytosolic viral DNA transcripts that are detected by IFI16. This sensor assembles into an inflammasome where caspase-1 becomes activated, which in turn triggers pyroptosis, a highly inflammatory form of programmed cell death. Elicitation of pyroptosis absolutely requires cell-to-cell spread of the virus; cell-free virions are not able to activate this response (see Figure 3). During chronic infection, most HIV-1 replication and loss of CD4 T cells occurs in such secondary lymphoid tissues (Zeng et al., 2012a).

Figure 2. Biological Differences between Tissue-Derived and Blood CD4 T Cells Have Important Implications for Viral Pathogenesis

(A) Target CD4 T cells derived from lymphoid tissue undergo abortive infection and pyroptotic cell death. Resting peripheral blood CD4 T cells are susceptible to HIV-1 fusion (with X4-tropic virus) and become abortively infected but are highly resistant to pyroptosis. This intrinsic resistance likely involves factors affecting both viral replication (i.e., early block of reverse transcription, arrested or delayed uncoating as a result of unsuccessful reverse transcription, reverse-transcribed DNA intermediates below the optimal length required for recognition), and the ability of peripheral blood CD4 T cells to mount an efficient antiviral response (i.e., low expression of IFI16 and other PRRs, a general defect in innate antiviral pathways in response to cytosolic DNA [Berg et al., 2014]). (B) Viral spread from productively infected cells extensively depletes target CD4 T cells in lymphoid cultures but not in cultures of peripheral blood cells. (C) Resting blood-derived CD4 T cells are rendered sensitive to cell death and massively depleted when cocultured with lymphoid cells (CD4 T, CD8 T or B cells), suggesting that the lymphoid microenvironment sensitizes CD4 T cells to depletion by abortive HIV infection and caspase-1-mediated pyroptosis. Thus, the resistance of target blood cells to pyroptosis may not be due to inefficient viral production or transfer from blood-derived CD4 T cells. Supernatants from tonsil cultures do not render blood CD4 T cells susceptible to pyroptosis, indicating that close interactions between the lymphoid-derived and blood cells are required for sensitization, as occurs when CD4 T cells in bloodstream traffic back to lymphoid tissue.

Figure 3. The Mode of HIV-1 Spread Determines the Outcome Form of Programmed Cell Death and Has a Key Role in HIV Pathogenesis

Free HIV particles kill only CD4 T cells that are permissive, undergo productive infection, and die from caspase-3-mediated apoptosis. However, in human lymphoid tissues such as tonsil and spleen, activated and permissive cells constitute <5% of all CD4 T cells. Free HIV-1 particles, even in large quantities, cannot directly trigger innate immune recognition and pyroptosis of nonpermissive target CD4 T cells, which constitute >95% of CD4 T cells in lymphoid tissues. Similarly, infection with lentiviral vectors does not kill nonpermissive target CD4 T cells (Doitsh et al., 2014). Thus, HIV particles themselves do not directly cause pathogenesis and AIDS. Conversely, it is the small fraction of permissive cells that become productively infected and mediate cell-to-cell spread across viral synapses culminating in the pyroptotic death on nonpermissive CD4 T cells. Thus, productively infected cells, not free HIV particles, are the fundamental “killing units” of CD4 T cells in lymphoid tissues. Productive (“direct”) and abortive (“bystander”) infections are therefore not independent pathways of CD4 T cell depletion; they are linked in a single pathogenic cascade. Along with playing a critical role in the virological synapse the interaction of LFA-1 on T cells with ICAM-1 also mediates the arrest and migration of leukocytes on surfaces of postcapillary venules at sites of infection or injury, as well as the ability of these cells to crawl out of the blood stream between high endothelial venules and into lymph nodes (Girard et al., 2012). Importantly, IL-1β and other inflammatory signals increase the expression of adhesion molecules such as ICAM-1 on endothelial cells (Barreiro et al., 2002; Carman and Springer, 2004; Dinarello, 2009; Dustin et al., 2011; Hubbard and Rothlein, 2000). The release of IL-1β by dying pyroptotic CD4 T cells in HIV-infected lymphoid tissues likely attracts more cells from the blood into the infected lymph nodes to die and produce more inflammation. Thus, the interaction of LFA-1 with ICAM-1 contributes to HIV pathogenesis by both promoting the depletion of CD4 T cells and facilitating a state of chronic inflammation, two key processes that propel disease progression to AIDS.

Figure 4. The Set of DNA Sensors Expressed in Cells Does Not Necessarily Define the Innate Response Pathway Against Intracellular DNA

(A) IFI16 displays two HIN domains (designated HIN-A and HIN-B) separated by a spacer region containing several serine, threonine, and prolines. The length of this region is regulated by alternative mRNA splicing, giving rise to three IFI16 isoforms (designated A–C). The predominant B isoform of IFI16 is detectable in various cell types, including human fibroblasts, epithelial cells, macrophages, and T cells (Dell'Oste et al., 2015). Interestingly, while all A–C IFI16 isoforms are equally expressed in tonsillar CD4 T cells and bind dsDNA, the B form specifically possess high affinity to ssDNA. (B) Biochemical analysis of cytosolic DNA-binding proteins in tonsillar CD4 T cells to identify potential viral DNA sensors (Monroe et al., 2014). Despite their significant endogenous expression, the known DNA binding proteins DAI, STING, and AIM2 were not recovered by immunoprecipitation of biotinylated HIV DNA. APOBEC3G, another endogenously expressed DNA-binding proteins, which has high affinity for ssDNA, was not identified in these analyses. The DNA sensor cGAS was detected neither at the protein level in tonsillar CD4 T cells nor in the affinity chromatography-mass spectrometry experiments. Conversely, the proteins DNA-PK and IFIX were expressed and associated with cytoplasmic HIV-1 DNA but were not involved in IFN-β induction and pyroptosis of CD4 T cells abortively infected with HIV-1. Thus, the surveillance activity of host innate sensors comprises a diverse set of PRRs that act nonredundantly against cytoplasmic DNA ligands. (C) Top-ranked hits of cytoplasmic DNA-binding protein in tonsillar CD4 T cells based on DNA affinity chromatography and mass spectrometry protein discriminant scores. Among the broad array of DNA-binding proteins identified, only IFI16 was required and sufficient to induce IFN-β and pyroptosis of lymphoid-derived CD4 T cells abortively infected by HIV-1. Some of the identified proteins might not be true sensors but instead have other regulatory roles involving innate sensing pathways, DNA damage repair, or the cell cycle. (D) A comparison of evidence supporting the function of the various receptors in lymphoid CD4 T cells and myeloid cells in detecting intracellular DNA. Despite some redundancy at the level of expression, most sensors act in a cell-type-specific manner. More detailed studies are needed to identify the dynamic role of individual sensors in the context of disparate viral infections and to assess crosstalk between the different sensing pathways. (A)–(C) were adapted from (Monroe et al., 2014). (D) is based on experimental data sets from Civril et al. [2013]; DeYoung et al. [1997]; Ding et al. [2004]; Ferguson et al. [2012]; Fernandes-Alnemri et al. [2009]; Gao et al. [2013]; Hornung et al. [2009]; Jakobsen et al. [2013]; Jin et al. [2012]; Lee [2013]; Li et al. [2007]; Lu et al. [2015]; Shindo et al. [2012]; Sun et al. [2013]; Takaoka et al. [2007]; Unterholzner et al. [2010]; Wang et al. [2008]; Wu et al. [2013]; Yan et al. [2008]; Zhang et al. [2011a, .

Figure 5. Caspase-1 Activity in Lymphoid Tissue May Persist Independently of Viral Replication and Promote a Chronic State of Inflammation and Immune Activation

Pyroptosis provides a nexus between CD4 T cell death and inflammation—the two key drivers of HIV pathogenesis. Abortive HIV infection of CD4 T cells in lymphoid tissues results in sensing of the viral cytosolic DNA products by IFI16 leading to caspase-1 activation in inflammasomes and pyroptosis (steps 1 and 2). Dying cells release large amounts of proinflammatory cytokines including IL-1β and cellular contents such as 5′-ATP into the extracellular milieu (step 2). These events promote local inflammation (step 3), which mediates the migration of new circulating CD4 T cells (predominantly central memory CD4 T cells, containing large amounts of pro-IL-1β) into the lymph node (step 4) and establishes a vicious cycle of HIV spread, CD4 T cell death, and inflammation. Inflammatory signals induce transendothelial migration of other types of circulating leukocytes into the inflamed lymphoid organs, particularly neutrophils and monocytes; the latter cells differentiate into tissue-resident macrophages or dendritic cells (step 5). These tissue-resident cells are primed to mount inflammatory responses and constitutively express high levels of cytoplasmic pro-IL-18, as well as the caspase-1 adaptor ASC and NLRP3 inflammasome. The release of proinflammatory cellular contents and ATP by nearby pyroptotic cells may activate the NLRP3 inflammasome in nearby primed cells (including primed naive CD4 T cells), leading to new rounds of pyroptosis (step 6). Such an “autoinflammation” scenario could generate persistent rounds of pyroptosis, chronic inflammation, and loss of CD4 T cells even when viral replication is suppressed by antiretroviral therapy.