NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Coffin JM, Hughes SH, Varmus HE, editors. Retroviruses. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 1997.

Cover of Retroviruses


Show details

Immunopathogenic Mechanisms of HIV Infection

Mechanisms of CD4+ T Lymphocyte Dysfunction

One of the early hallmarks of HIV infection, prior to the profound depletion in CD4+ T cells, is the impairment of a variety of CD4+ T-cell functions including T-cell colony formation, autologous mixed lymphocyte reactions, expression of interleukin-2 (IL-2) receptors, and IL-2 production (for review, see Rosenberg and Fauci 1989a). T-cell proliferative responses to a range of stimuli are diminished in all stages of HIV disease. The earliest T-cell proliferative response to be lost is the response to recall antigens such as tetanus toxoid, influenza, Candida albicans, and Cryptococcus neoformans. Defects in responses to soluble antigens are followed by the loss of T-cell proliferative responses to alloantigens and, subsequently, mitogens. Decreased CD4+ T-cell responses to antigens and mitogens have also been observed during acute HIV infection (Cooper et al. 1988). Some studies suggest that the defect in T-cell proliferative responses is independent of abnormalities in antigen-presenting cells, whereas others have suggested that abnormalities in antigen-presenting cell function caused a secondary defect in T-cell proliferation (Meyaard et al. 1993; see below).

Helper CD4+ T cells of both mice (Mossman et al. 1986) and humans (Romagnani 1991) can be separated into two groups according to their function and profiles of secreted cytokines (see Chapter 12. T-helper-1 (TH1) cells provide help to CTLs and secrete IL-2 and interferon-γ, whereas TH2 cells support B lymphocytes and secrete IL-4 and IL-10. It has been postulated that a switch from a TH1 to a TH2 cytokine pattern in HIV-infected individuals is a critical step in the pathogenesis of HIV disease (Clerici and Shearer 1993), but the sharp dichotomy between these two types of cytokine patterns related to disease progression has not been corroborated (Graziosi et al. 1994; Maggi et al. 1994).

The precise cause of the CD4+ T-cell abnormalities seen in HIV infection is not now known. Abnormalities occur in both the presence and the absence of direct infection of the CD4+ T cell (for review, see Rosenberg and Fauci 1989a, 1992). Direct killing may deplete certain functional subsets of CD4+ T cells eliminating that particular immune function. Even when an infected cell is not killed, its function may be compromised. The cell surface expression of the CD4 receptor in infected CD4+ T cells is down-modulated as the HIV gp120 molecules produced during viral replication bind to cytoplasmic CD4, forming intracellular gp120-CD4 complexes. Other viral products, Nef and Vpu, also contribute to the downregulation of CD4 (Willey et al. 1992a,b; Anderson et al. 1993; Benson et al. 1993; Chen et al. 1993; Garcia et al. 1993; Mariani and Skowronski 1993; see Chapter 7. These effects probably contribute to viral replication by reducing superinfection and permitting efficient release of virions. As a side effect, the decrease in surface CD4 expression reduces MHC class II–CD4 interactions, which could cause a decline in antigen-specific responses. Since HIV preferentially infects CD4+ T cells that belong to the memory subset (Schnittman et al. 1990a), an HIV-induced decrease in CD4 expression in these cells could account for the defect in response to soluble antigens. The importance of these effects relative to direct killing of activated CD4+ T cells remains to be established.

Failure to respond to antigenic stimuli has also been seen in uninfected CD4+ T cells in HIV-infected patients (for review, see Rosenberg and Fauci 1989a, 1992). HIV proteins such as purified Env glycoproteins (both HIV gp120 and gp41), Env peptides, and purified Tat protein have been reported to suppress, in vitro, proliferative responses to mitogens, antigens, and anti-CD3 antibody. Exposure of cells to HIV gp120 also inhibits both IL-2 production and IL-2 receptor expression. The mechanisms by which HIV envelope glycoproteins suppress immune function in the absence of infection of the cell involve the binding of gp120 to membrane-bound CD4. gp120-CD4 interactions at the cell surface could sterically interfere with the MHC class II–CD4-mediated cell adhesion necessary for the induction of antigen-specific responses. Inappropriate binding of CD4 by anti-CD4 antibodies results in a reduction in the expression of the T-cell receptor (TCR), suggesting that gp120-CD4 interactions may have a similar effect.

There is evidence for a negative effect of HIV on signal transduction. In vitro, the crosslinking of CD4 molecules that occurs when gp120 is bound in the presence of anti-gp120 antibodies causes suppression of T-cell activation (Mittler and Hoffmann 1989). Anti-gp120 antibody-coated CD4+ T cells have been detected in HIV-infected individuals, suggesting that soluble gp120 can bind CD4 molecules in vivo and HIV gp120-induced T-cell anergy could occur (Amadori et al. 1992).

In addition to suppression of T-cell activation, in vitro experiments have also shown that gp120-CD4 binding could cause inappropriate T-cell activation (Ascher and Sheppard 1990), and it is hypothesized that if these cells no longer respond to a subsequent antigenic challenge, a state of anergy thus ensues.

Although all of these mechanisms have received support from in vitro experimentation, it is important to remember that there is as yet no direct evidence that any of these mechanisms have a significant role in HIV-1 pathogenesis in vivo. A particularly important issue is whether there is a sufficient concentration of free (or virion-associated) gp120. Even at a high viral load of 106 virions per milliliter, the concentration of virion-associated gp120 is only about 2 picomolar. The concentration of anti-gp120 antibodies is far greater than this.

Dysfunction of Other Immune Competent Cells

Monocytes and Macrophages

As discussed above (see Cellular Targets of Infection, Host Range of Cellular Targets), HIV infects monocytes and macrophages in vitro. In vitro, the cytopathic effects are less than those observed when CD4+ T cells are infected. In vivo, HIV-infected macrophages are readily found in tissues throughout the body. Although there have been conflicting reports regarding the numbers of HIV-infected monocytes in the peripheral blood, most studies have demonstrated that the major reservoir of HIV in the peripheral blood is the CD4+ T cell; the numbers of infected monocytes in the blood are quite low compared to both CD4+ T cells in the blood and macrophages in the tissues (Psallidopoulos et al. 1989; Spear et al. 1990). This makes it unlikely that the range of abnormalities which have been reported in peripheral blood monocyte/macrophages, including defects in chemotaxis, monocyte-dependent T-cell proliferation, Fc receptor function, C3-receptor-mediated clearance, and oxidative burst responses, can be attributed to direct infection of these cells by HIV (for review, see Rosenberg and Fauci 1989a). However, it is possible that monocyte/macrophage dysfunction is caused by exposure to HIV proteins, since exposure of peripheral blood monocytes to noninfectious HIV, to gp120 alone, or to a synthetic peptide homologous to gp41 can cause the downregulation of chemotactic ligand receptors followed by suppression of chemotactic function as well as the respiratory burst (Harrell et al. 1986). Exposure of monocytes to HIV gp120 leads to cellular activation and the secretion of certain cytokines (Clouse et al. 1991) which might contribute to the observed abnormalities (Graziosi et al. 1994). Again, it is uncertain whether the concentrations of gp120 in blood are sufficiently high for these effects to be important.

CD8+ T Cells

During symptomatic primary infection with HIV, the level of CD8+ T cells generally declines; however, following this acute phase, CD8+ T-cell numbers rebound to higher than normal levels, sometimes remaining at these elevated levels throughout the clinically latent stage of disease (Cooper et al. 1988; Gaines et al. 1990). Relatively high levels of HIV-specific CD8+ CTLs can be demonstrated in many HIV-infected individuals during the early stages of HIV infection (Walker and Plata 1990; Koup et al. 1994; Safrit et al. 1994). This observation suggests that the observed CD8+ T lymphocytosis may be due, at least in part, to expansion of HIV-specific CTLs. The rise in CD8+ T cells may also reflect an attempt by the immune system to maintain a normal number of lymphocytes. It has been hypothesized that as CD4+ T cells are selectively depleted in HIV infection, the homeostatic mechanisms cause the lost CD4+ T cells to be replaced with CD8+ as well as CD4+ T cells (for review, see Stanley and Fauci 1993).

As HIV disease progresses, anti-HIV CTL activity is lost. This decline in HIV-specific CTLs may be explained by defects in the clonogenic potential (for review, see Rosenberg and Fauci 1989a). It has been demonstrated that a significant proportion of circulating CD8+ T cells in HIV-infected individuals have an unusual phenotype. They are HLA-DR-positive (activated), but they do not express the CD25 marker (the receptor for IL-2) which is required for optimal clonogenic potential (Pantaleo et al. 1990). Since HIV-specific CTL activity resides almost exclusively within the CD8+DR+CD25 subset of CD8+ T cells, as these cells lost clonogenic potential, HIV-specific CTLs would also be lost. The phenotype of the CD8+ T cell in HIV-infected individuals can have prognostic significance. Individuals who developed HLADR+ CD38CD8+ cells at seroconversion also stabilize their CD4+ T-cell counts (Giorgi et al. 1994). Elevated levels of CD38+CD8+ T cells are associated with a poor prognosis (Giorgi et al. 1993). In addition to the defects in HIV-specific CTLs, functional defects in other MHC-restricted CTLs have been demonstrated in HIV infection, including CTLs directed against influenza and CMV (for review, see Rosenberg and Fauci 1989a). The loss of CD4+ T cells as well as the defect in IL-2 production is characteristic of HIV infection and probably compounds the defect in CD8+ CTLs; the integrity of CD8+ T-cell function depends in part on adequate inductive signals from CD4+ helper/inducer T cells.

During the late stages of HIV disease when CD4+ T-cell levels are dramatically reduced and opportunistic infections are common, there is a significant reduction in the number of CD8+ T cells (Lane et al. 1985). One potential explanation for the decline in CD8+ T cells late in the course of disease is a defect in the ability of progenitor cells to regenerate mature T cells (Fauci 1988). Since cells in the thymus are infected during HIV infection in humans and SIV infection in macaques (J.G. Müller et al. 1993), the loss of immature thymocytes may result in a failure to regenerate adequate levels of all types of mature T cells. The rapid rise in CD4 counts following treatment with antiviral drugs even in patients with very low CD4+ T-cell levels implies that there is significant regenerative capacity even in advanced disease (Ho et al. 1995; Wei et al. 1995). However, these replacement CD4 cells are primarily memory cells, suggesting that they are created by division of existing cells, rather than by de novo generation (Lane 1995). Even with combination drug therapy that leads to long-term suppression of viral replication, CD4 cell numbers do not return to normal levels, despite the initial rapid rise, and there is some question as to whether the immune response can completely recover in such patients.

One of the problems in addressing these questions is the lack of a good in vivo experimental model in which HIV infection of various cell types can be studied. A useful model has been developed using genetically determined severe combined immunodeficiency (SCID) mice into which a human hematopoietic system (thymus and bone marrow) has been grafted. In such SCID-hu mice, HIV can infect CD4/CD8 double positive cells as well as CD4+ and CD8+ immature thymocytes (Stanley et al. 1993).

B Cells

Aberrant polyclonal activation of B cells is one of the first immunologic abnormalities to appear following HIV infection. B cells from HIV-infected individuals proliferate spontaneously in vitro, and hypergammaglobulinemia of IgG, IgA, and IgD is commonly observed in vivo and may result from the lymph node hyperplasia associated with HIV infection (Lane et al. 1983). Although spontaneous B-cell proliferation and antibody production do occur in HIV infection, B cells from HIV-infected individuals are defective in their response to additional activation signals. Defects in the production of immunoglobulins in response to specific antigens or pokeweed mitogen can be seen early in the course of infection (Terpstra et al. 1989; for review, see Rosenberg and Fauci 1989a). In addition, B-cell abnormalities also cause poor responses to primary and secondary immunizations to both protein and polysaccharide antigens. Studies of primary HIV infection indicate that the absolute numbers of circulating B cells may be depressed; however, this is usually a transient phenomenon (Reddy et al. 1991). In certain patients, the number of B cells may decrease late in the disease (Reddy et al. 1991).

Some of the loss of B-cell function is presumably due to the loss of CD4+ T-cell helper cells that are critical to proper B-cell function. However, HIV may directly cause functional B-cell aberrations via direct B-cell-stimulating effects. In this regard, exposure of B cells to intact or disrupted HIV in vitro can cause many functional abnormalities similar to the abnormalities seen in B cells in vivo (for review, see Rosenberg and Fauci 1989a). Portions of the HIV gp41 envelope glycoprotein have been shown to induce polyclonal B-cell activation (Chirmule et al. 1990). The receptor for HIV on the surface of human B cells has been demonstrated to be a product of the VH3 genes (Berberian et al. 1993). However, these correlations should be viewed with some caution. As was pointed out earlier in the discussion of T-cell defects in patients, the amount of circulating HIV-1 protein is relatively small. Other polyclonal B-cell activators, such as Epstein-Barr virus (EBV) and CMV, are commonly found in HIV-infected individuals and may also contribute to the functional defect (Pahwa et al. 1986).

Natural Killer Cells

Functional abnormalities in natural killer (NK) cells are seen in the course of HIV infection; the magnitude of the defect increases as HIV disease progresses (for review, see Rosenberg and Fauci 1989a). NK cell activity is reduced in both peripheral blood and bronchoalveolar lavage fluids from AIDS patients. The defect(s) in NK cells in HIV infection might involve the production or release of cytotoxic factors; neither NK cell binding nor antibody-dependent cellular cytotoxicity is impaired. NK function can be restored in vitro by the addition of IL-2, suggesting that IL-2 is required for the production of NK cell cytotoxins. This suggests that the reduction in CD4+ T cells, which results in lower amounts of IL-2, may contribute to NK cell dysfunction. IL-12 has also been shown to enhance the function of NK cells from HIV-infected individuals in vitro (Chehimi et al. 1992), and there are reports suggesting that NK cells may be depleted in AIDS patients (Mansour et al. 1990).

Mechanisms of CD4+ T Lymphocyte Destruction

An unresolved issue critical for understanding CD4+ T lymphocyte destruction and AIDS pathogenesis is accurate quantitation of the number of infected lymphoid cells; different results have been obtained with different methodologic approaches. Quantitative PCR measurements have found approximately 1 provirus per 1,000–20,000 PBMCs; different laboratories have obtained reasonably consistent results with these types of PCR measurements (Schnittman et al. 1989; Simmonds et al. 1990; Daar et al. 1991; Piatak et al. 1993). In contrast, in situ PCR techniques have yielded data suggesting that as many as 8–16% of PBMCs contain HIV DNA. In some cases, the numbers of virus-positive cells are higher than the numbers of CD4-positive cells. Here too, different laboratories have obtained reasonably consistent results (Bagasra et al. 1992; Bagasra and Pomerantz 1993; Embretson et al. 1993b; Patterson et al. 1993), and in situ PCR studies suggest that the relative number of infected cells is similar to those in lymphoid tissues and in the blood (Embretson et al. 1993a). However, other studies in which PCR is done in solution indicate that there are approximately three- to tenfold more infected cells in lymph nodes than in peripheral blood and that the number of infected cells is much lower than what was measured by in situ PCR (Pantaleo et al. 1993c). These discrepancies have not been resolved. It should be kept in mind, when trying to interpret any data based on measurements of viral DNA, that a substantial fraction of the proviruses present in cells may be defective copies that accumulate as infection continues (Coffin 1995; Nowak et al. 1997; Perelson et al. 1997). Only a small fraction of cells that contain viral DNA may be involved in active viral replication.

Even if the lower estimates of the numbers of infected cells are correct, the collapse of the immune system could be explained by the gradual loss of CD4+ T cells, particularly because the immune system seems to be unable to regenerate these cells adequately in HIV-infected individuals (see below). Although it has often been assumed that depletion of CD4+ T cells and cells of the monocyte/macrophage lineage is a direct result of HIV or SIV infection, the extent to which the observed depletion is the result of the virus directly killing the cells is unclear. As has already been discussed, it is relatively easy to demonstrate a number of virus-related mechanisms of cell killing in vitro (for review, see Fauci 1988; Rosenberg and Fauci 1989a); however, the extent to which these mechanisms are responsible for the depletion of immune competent cells in vivo has not been defined and likely represents a combination of processes.

Direct Killing of Single Cells

A number of mechanisms have been proposed for the killing in vitro of single HIV-infected cells. These include copious budding of virions from the cell surface with concomitant disruption of the integrity of the cell membrane, interference with cellular RNA processing, disruption of cellular protein synthesis due to high levels of viral RNA, and the accumulation of high levels of unintegrated viral DNA in the cell (for review, see Rosenberg and Fauci 1992). The intracellular interaction between HIV gp120 and CD4 has also been implicated in cytopathicity (Fisher et al. 1986; Hoxie et al. 1986; Koga et al. 1990).

There are clear strain differences in the killing of cells by HIV-1; these are determined largely by gp120 sequences, providing strong support for the importance of env in this process. This role need not involve toxicity directly; rather, as with some simple retroviruses (Keshet and Temin 1979; Weller et al. 1980), certain strains could be less effective in inducing resistance to superinfection than others, allowing reinfection of a cell by the virus it produces and a consequent buildup of viral replication to toxic levels.

Syncytium Formation

A potential explanation of the cause of the death of cells that are not directly infected with HIV is the formation of multinucleated giant cells or syncytia. As discussed above, some strains of HIV and SIV, known as SI strains, efficiently induce syncytia (for review, see Narayan et al. 1988; Rosenberg and Fauci 1989a).

Uninfected CD4+ T cells are killed in vitro by fusion with an HIV-infected cell; fusion occurs when CD4 molecules on the surface of the uninfected cell bind to gp120 on the infected cell. Multinucleated thymocytes that appear to be syncytia have been seen in the human thymus in HIV-infected SCID-hu mice (S. Stanley and A.S. Fauci, unpubl.). Multinucleated syncytial cells are also a common feature of both SIV- and HIV-induced encephalitis. However, the relative contribution of syncytium formation to the depletion of CD4+ T cells in vivo remains unknown. The appearance of SI virus, although it may accelerate the course of disease, is not essential for the process, since such viruses are found in only about half of people dying of AIDS (Tersmette et al. 1988; Jurriaans et al. 1994).

Immune-mediated Attack on Infected Cells and Cells Coated with Viral Proteins

Although it is possible to describe the early rise and decline of viremia following HIV infection on the basis of availability of target cells (Phillips 1996), immunologic responses directed against HIV or HIV-infected cells appear to have a significant role in limiting viral replication leading to the decline in viral load following the initial phase of primary infection. Both humoral and cell-mediated responses likely contribute to the partial suppression of virus and viral replication, acting both on the virus and on infected CD4+ T cells during the prolonged course of clinical latency (see below, Immune Responses to HIV and SIV). HIV-specific immune responses may also target uninfected immune competent cells that have viral proteins bound to their surfaces. Soluble gp120 can be shed from free virions and from HIV-infected cells and can bind to CD4 molecules on the surface of uninfected T cells and monocyte/macrophages. Anti-HIV gp120 antibody can recognize these bound gp120 molecules and cause the elimination of these cells by an antibody-dependent cell-mediated cytotoxicity (ADCC) pathway. In addition, gp120-specific CTLs may target uninfected cells that have bound gp120 (Lyerly et al. 1987; Weinhold et al. 1989). Similarly, virus and/or viral proteins may adhere to peripheral blood or follicular dendritic cells and mark them for destruction by ADCC, CTL, or other unidentified mechanisms. In this regard, effector cells from HIV-infected individuals have been shown to mediate both CTL and NK-associated ADCC against uninfected cells that were coated with HIV antigens in vitro (Katz et al. 1988). Again, there are substantial questions of whether there is sufficient gp120 in patients to mediate such effects in vivo.

Autoimmune Mechanisms

A number of clinical manifestations seen in HIV-infected individuals may be associated with autoimmunity. These include Reiter's syndrome, HIV-associated arthropathy, psoriatic arthritis, polymyositis/dermatomyositis, Sjogren's syndrome, necrotizing vasculitis, septic arthritis, and other connective tissue disorders (Espinoza et al. 1989; Kaye 1989). The occurrence of HIV-induced rheumatologic disease might involve opportunistic pathogens that induce a reactive arthritis; HIV antigens could also stimulate an immune reaction with subsequent tissue damage. A role for HIV antigens is supported by the CAEV model where CAEV-induced chronic arthritis is associated with tissue inflammation produced by cytokines induced by the immune response to viral proteins (Straub 1989). Similarly, human T-cell leukemia virus type 1 (HTLV-1)-associated arthropathy in humans is associated with activated lymphocytes in the synovium.

Another pathway by which uninfected cells might be eliminated involves antigenic cross-reactivity between HIV proteins (gp120 and gp41) and MHC class II determinants (Ziegler and Stites 1986). In this model, anti-HIV antibodies cause the elimination of uninfected MHC class II-bearing cells via ADCC. Anti-HLA-DR antibodies have been detected in HIV-infected individuals. Not only is there similarity between HIV proteins and MHC class II determinants, but similarity also exists between HIV envelope glycoproteins and MHC class I molecules (for review, see Rosenberg and Fauci 1989a). Whether these have any functional significance or are merely indicative of the power of computer searches to match random segments of unrelated sequences (see Appendix 1) remains to be determined.

A variety of antibodies that have the potential to react with uninfected normal cells have been identified in the sera of HIV-infected individuals. These include antibodies to lymphocytes, platelets and neutrophils, CD4 molecules, CD43 molecules, and IL-2 and autologous antibodies bound to uninfected T cells (Ardman et al. 1990a,b).

Defect in CD4+ T-cell Regeneration

The gradual decline in CD4+ T cells that occurs during HIV infection may result, at least in part, from the inability of the immune system to regenerate the CD4+ T-cell pool (Fauci 1988). At least two major mechanisms may contribute to the failure of CD4+ T-cell regeneration during the course of HIV or SIV infection: (1) the destruction of lymphoid precursor cells, perhaps by infection as they become activated to divide and (2) the disruption of the microenvironment required for efficient regeneration of immune competent cells. HIV infection of CD34+ bone marrow myeloid precursor cells occurs both in vivo and in vitro; in vitro experiments show a concomitant suppression of cell proliferation and colony formation (Folks et al. 1988; Stanley et al. 1992). Although it is not known whether infection of these cells leads to cell death in vivo, higher levels of HIV-infected CD34+ cells are associated with low CD4+ T-cell counts in North American patients (Stanley et al. 1992). However, HIV-mediated suppression of bone marrow precursor cells does not necessarily involve infection of these cells, since exposure of bone marrow progenitors to recombinant gp120 results in decreased cell viability (Zauli et al. 1991, 1992). The presence of a spectrum of bone marrow abnormalities in HIV-infected patients, including thrombocytopenia, granulocytopenia, anemia, and lymphopenia (Costello 1988), lends support to the hypothesis that HIV infection of bone marrow precursors has a role in HIV-mediated immunosuppression.

The lack of regeneration of CD4+ T cells may also be explained by the deleterious effects of HIV on the thymus. Immature CD4+ CD8+ thymic lymphocytes are highly susceptible to infection by HIV (Schnittman et al. 1990b; Tanaka et al. 1992). Even thymic precursor cells that were thought to have a triple-negative phenotype (CD3CD4CD8) express sufficient amounts of CD4 to be infected by HIV in vitro (Schnittman et al. 1990b). If HIV infection of thymic precursor cells results in cytopathicity, then the loss of these cells could interfere with regeneration of mature CD4+ T cells. This hypothesis is supported by experiments with the SCID-hu mouse model discussed earlier (Aldrovandi et al. 1993; Bonyhadi et al. 1993; Stanley et al. 1993).

Depletion of thymic precursor cells may also occur if the thymic microenvironment is disrupted during HIV infection. In the SCID-hu mouse, HIV infection results in destruction of the thymic stromal architecture and internalization of HIV within thymic stromal epithelial cells (Stanley et al. 1993). In HIV-infected patients, premature atrophy of the thymus is seen in conjunction with thymocyte depletion and fibrosis (Seemayer et al. 1984). As discussed in greater detail below (see Role of Lymphoid Organs in Immunopathogenesis), disruption of the lymph node architecture is seen in HIV-infected patients in vivo. It should, however, be kept in mind that, when HIV-1 replication is strongly suppressed by effective drug therapy, there is a concomitant rise in CD4+ cell counts (Ho et al. 1995; Wei et al. 1995). This observation implies that there is still a considerable ability to generate CD4+ cells—largely by division of existing memory cells (Lane 1995)—even at relatively late stages of AIDS.


It has been hypothesized that some of the quantitative and qualitative abnormalities in CD4+ T cells in HIV infection may be due to the presence of superantigens (for review, see Pantaleo et al. 1993a). Superantigens are proteins that bind directly both to MHC molecules and to a relatively constant region of the Vβ portion of the T-cell receptor. As such, they are recognized by and stimulate a higher proportion of T cells than most conventional antigens (see Chapter 10. Superantigens encoded by certain bacteria and by the sag gene of mouse mammary tumor virus (MMTV) induce the expansion of an entire Vβ subset which can represent from 1% to 10% of all T cells (Chapter 10. Depending on the state of activation of the T cells, and the developmental stage of the affected individual, superantigens can cause expansion, deletion, or anergy of the affected Vβ subset.

Perturbation of the Vβ repertoire in HIV-infected individuals has suggested a possible role for superantigens in HIV infection. In some individuals, specific Vβ subsets of CD4+ T cells can be either preferentially depleted or expanded (Imberti et al. 1991; Dalgleish et al. 1992). However, since opportunistic pathogens are common in HIV-infected individuals, the Vβ perturbations could be caused by a superantigen expressed by an opportunistic pathogen, rather than to HIV itself. Humans are outbred, and genetic differences in the expression of the Vβ repertoire may explain some of the data, although one recent study of monozygotic twins discordant for HIV infection and HIV-infected mother/ infant pairs showed differences (Soudeyns et al. 1993). The primary problem in proposing a role for an HIV superantigen is that, so far, despite extensive analysis of the virus, no HIV gene product with clear-cut superantigen activity has been found.


Apoptosis is a form of programmed cell death that normally occurs during organogenesis (for review, see Thompson 1995) and during thymic development, where it is a mechanism by which autoreactive T cells are clonally deleted. Apoptosis is also involved in the elimination of some activated cells during a normal immune response. The induction of apoptosis requires cellular activation, and it has been suggested that HIV infection causes apoptosis through an aberrant signaling mechanism. Apoptosis can be induced in murine T cells by crosslinking of the CD4 receptor and activation of the TCR (Newell et al. 1990). During HIV infection, crosslinking of the CD4 receptor could occur following binding of either gp120 or gp120/anti-gp120 complexes to CD4 (Ameisen and Capron 1991; Groux et al. 1992), thus providing the first of the two signals required for apoptosis (Fig. 9). The cell would then be primed for the second apoptotic signal, i.e., activation of the TCR which could be by either conventional antigens or superantigens.

Figure 9. Apoptosis in HIV infection.

Figure 9

Apoptosis in HIV infection. Complexes of gp120 and gp120 antibody crosslink the CD4 molecule on CD4+ T cells. This provides the first signal for apoptosis. Activation of the cell by antigen (more...)

Several experiments have suggested a role for HIV-induced apoptosis in CD4+ T-cell depletion. gp120/anti-gp120 complexes are present on the surface of CD4+ T cells in HIV-infected individuals (Amadori et al. 1992). Although it has not been formally demonstrated that gp120-induced CD4 crosslinking results in signal transduction, extremely low concentrations of gp120 can prime cells for apoptosis (Banda et al. 1992). Apoptosis can be induced in CD4+ T cells from asymptomatic HIV-infected individuals if the cells are exposed in vitro to a superantigen or mitogen (Groux et al. 1992). High levels of anti-histone and anti-double-stranded DNA antibodies are seen in HIV-infected individuals. This observation could be explained by the release of nuclear contents during apoptosis (Muller et al. 1992). Both CD4+ and CD8+ T cells from HIV-infected individuals can apparently undergo spontaneous apoptosis in vitro (Gougeon et al. 1991; Meyaard et al. 1992). However, most of the apoptosis in lymphoid tissue occurs in cells that are not infected with HIV (Finkel et al. 1995), and there is no correlation between apoptosis and the stage of HIV disease (Meyaard et al. 1994; Muro-Cacho et al. 1995). Apoptosis in lymphoid tissue of HIV-infected individuals may reflect the degree of activation of the lymphoid tissue (Muro-Cacho et al. 1995; Pantaleo and Fauci 1995).

Induction of Viral Expression

Role of Endogenous Cytokines

As discussed in detail above (see Cellular Targets of Infection, Host Range of Cellular Targets), activation of CD4+ T cells is necessary for efficient infection, replication, and production of HIV. Factors that influence the activation of CD4+ T cells and monocytes/ macrophages therefore have important roles in infection. One such class of factors is endogenous cytokines, the peptide hormones that regulate immune function. Enhancing the level of cell activation with phytohemagglutinin (PHA) and IL-2 has long been known to be necessary for efficient spread of HIV-1 in PBMC cultures. Furthermore, supernatant from PHA-stimulated PBMCs or lipopolysaccharide (LPS)-stimulated macrophages can induce HIV expression in chronically infected T or monocytoid cell lines.

Tumor necrosis factor-α (TNF-α) shows HIV induction activity in mitogen-activated cell supernatants; both TNF-α and TNF-β function in an autocrine/ paracrine manner to regulate HIV in acutely infected primary T cells and monocytoid cell lines. Induction of endogenous TNF-α is an important component of the phorbol myristate acetate (PMA)-induced expression of HIV in chronically infected T and monocytoid cell lines (for reviews, see Clouse et al. 1989b; Folks et al. 1989; Poli et al. 1990; Poli and Fauci 1992, 1993, 1995; Kinter et al. 1995).

The observation that TNF-α was an inducer of HIV-1 replication led to the discovery that several other cytokines also stimulated HIV-1 expression. The precise nature of its effects depend on whether the infected cell is of T-cell or monocytoid origin (for review, see Poli and Fauci 1992, 1993, 1995). Cytokines known to affect HIV infection are listed in Table 2. Not only do individual cytokines stimulate HIV-1, but certain combinations of cytokines also have synergistic effects. The molecular mechanisms of TNF-α are probably the best understood: It reduces NF-κB which acts as a transcriptional activator of HIV and also stimulates the expression of a variety of cellular genes (Chapter 6. IL-1 is thought to act on HIV replication at the level of viral transcription; however, NF-κB has not been shown to be involved in this process. Cytokines such as IL-6, granulocyte macrophage–colony-stimulating factor (GM-CSF), and IFN-γ act predominantly at the posttranscriptional level.

Table 2. Cytokine Regulation of HIV Expression.

Table 2

Cytokine Regulation of HIV Expression.

The timing of exposure to a cytokine relative to the timing of HIV infection may be important, and differences are seen if established cell lines or primary cultures are used. For example, IFN-γ functions as an autocrine/paracrine inducer of HIV expression in chronically infected monocytoid cell lines and in primary monocyte cultures if it is added prior to HIV infection, but downregulates HIV expression when primary cultures are infected with HIV before IFN-α is added. TGF-β also has dichotomous effects on HIV replication. HIV replication is enhanced in TGF-β-treated primary T cells; however, viral replication in monocytes or macrophages may be either boosted or suppressed depending on the timing of TGF-β treatment in relation to HIV infection. To date, the only cytokines that consistently result in a downregulation of HIV expression are IFN-α and IFN-β, probably due to a block in reverse transcription as well as suppression of virion release from the cell membrane (Clouse et al. 1989a; for reviews, see Poli and Fauci 1992, 1993, 1995).

In addition to the regulatory effects of cytokines on HIV replication, cytokine expression can be influenced by HIV (for review, see Poli and Fauci 1992, 1993, 1995). For example, IL-6 and TNF-α production is increased in cells infected with HIV. Similarly, HIV-infected monocytes have elevated levels of cytokine production following exposure to mitogens, antigens, or TNF-α. Precisely how HIV activates cytokine production is unclear. HIV replication may be necessary since expression of Tat protein can increase TNF-α gene expression. However, HIV-induced activation of cytokine expression has also been observed in the absence of viral infection following binding of CD4 with either gp120 or a monoclonal antibody that recognizes the gp120-binding site on CD4 (Rieckmann et al. 1991b; Biswas et al. 1994).

Although a role of endogenous cytokines in the regulation of HIV expression has not been formally proven, it is likely that these factors have some role in the modulation of viral replication and in inducing activation of target cells to an HIV-sensitive state. PBMCs in HIV-infected individuals secrete elevated levels of TNF-α, IL-6, and IL-1, and in some patients, there are elevated levels of TNF-α and IL-6 in body fluids. Activated B cells from HIV-infected individuals spontaneously secrete high levels of TNF-α and IL-6, which can induce HIV expression in chronically infected cells in vitro. B cells from HIV-infected individuals secrete increased levels of cytokines if these cells are exposed to gp120 in vitro. IL-4-treated B cells from uninfected individuals can be stimulated to secrete elevated levels of cytokines by exposure to HIV virions or HIV gp120, and CD4+ and CD8+ HIV-specific cytolytic T-cell clones produce increased amounts of TNF-α and IFN-γ following exposure to HIV in vitro. Lymph nodes from HIV-infected individuals have enhanced expression of a variety of HIV-inducing cytokines such as TNF-α, IFN-γ, and IL-6 when compared to lymph nodes obtained from uninfected individuals (for reviews, see Rieckmann et al. 1991a,c; Poli and Fauci 1992, 1993, 1995).

The increased expression of HIV-inductive cytokines may have special importance in lymphoid tissue where the proximity of uninfected and infected cells favors viral spread.

Role of Heterologous Pathogens

Since the immune system is activated by a variety of bacterial, viral, or fungal pathogens, the infection of HIV-infected individuals with additional pathogens may alter the progression of HIV disease (for review, see Rosenberg and Fauci 1989b). Activation of uninfected CD4+ T cells by their cognate antigens renders these cells susceptible to infection with HIV and facilitates the spread of virus from cell to cell. In addition, antigen-induced activation of cells already infected with HIV may increase viral replication. Although it seems reasonable to propose that the effect of heterologous pathogens on HIV expression is due in part to the induction of cytokines, the activation of other cellular and viral genes may also have important roles.

A number of heterologous viruses, such as herpes simplex virus type 1 (HSV-1), CMV, EBV, adenovirus, hepatitis B virus (HBV), human herpesvirus type 6 (HHV-6), pseudorabies virus, and HTLV-1, can enhance HIV expression in vitro (for review, see Rosenberg and Fauci 1989b; Levy 1993). This effect results from induction of expression of cellular transcription factors that activate the HIV LTR (see Chapter 6. Viral proteins that can enhance HIV expression include NF-κB-binding proteins, SP1, and TATA-binding factors. Other microbes, such as Mycoplasma, have also been reported to induce HIV expression. It must be kept in mind that these effects require coinfection of the target cells with HIV and the second pathogen which, in most cases, is unlikely to occur to a significant extent.

Heterologous coinfecting pathogens can also cause synergistic cytopathic effects. For example, HHV-6 coinfects HIV-1-infected CD4+ T cells, and these coinfected cells have increased levels of cytopathology (Lusso and Gallo 1995). Secondary pathogens can also contribute to the suppression of immune function and to the induction of cell surface molecules, such as the Fc receptor, that could, in principle, mediate HIV infection via virus-antibody complexes.

Copyright © 1997, Cold Spring Harbor Laboratory Press.
Bookshelf ID: NBK19451


  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...