Immune activation in the context of HIV infection (original) (raw)

In the article by Vingerhoets et al., published in this issue, the reasons for the defective response of T cells obtained from HIV-infected patients to stimulation by superantigens have been explored [1]. The authors describe a marked diminution in response of both T cell subsets during HIV infection that appears first in CD8 cells, is related to the progression of the disease, and is associated with increased levels of apoptosis. They ascribe these changes to extrinsic and intrinsic cellular defects, and rightly point out that several non-exclusive mechanisms could explain T cell dysfunction in the course of HIV infection. In concluding their article, the authors are still not clear as to the direct causes of the impaired responses they have described, and indicate that further studies are needed to identify and characterize these defects and the molecular regulation of apoptosis. However, beyond the questions of molecular mechanisms, there are conceptual issues regarding the nature of chronic infection and possible adaptive value of the functional modifications observed versus their pathogenic significance.

Persistent activation of the immune system is one of the hallmarks of HIV infection. Already in 1983, Shearer compared this activation to that occurring in graft versus host reaction [2]. Others have pointed to the role of immune activation in enhancement of viral replication and dissemination, thereby affecting progression of the disease [3, 4]. Ascher & Sheppard proposed, several years ago, that abnormal non-specific activation of CD4 T cells by the virus is responsible for a dysregulation of their turnover and for the development of immunodeficiency [5, 6]. We have proposed more recently that chronic immune activation caused by chronic infections is a major factor in the pathogenesis of AIDS in Africa, by making the host more susceptible to HIV infection and less able to cope with it once infected [7]. We have also proposed that persistent stimulation of the immune system by HIV infection, in itself, is sufficient to cause most if not all of the numerical, phenotypic and functional changes that are observed [8]. Progression of the disease was attributed to the persistent ‘stress’, resulting in an ageing-like process of the immune system in terms of the functional phenotype [8]. These ideas challenged the generally accepted view [10] that HIV ‘depletes’ the CD4 lymphocyte pool by effecting the killing of these cells, directly or indirectly.

Specifically, we have proposed that two major processes are responsible for the diminution of CD4 T cell counts in blood during the asymptomatic phase of the disease: (i) activation-associated redistribution of T cells between the blood compartment and the lymphoid tissues; and (ii) activation-associated attenuation of the proliferation of CD4 T cells in the infected tissues. Redistribution may be related to changes in expression of cell surface molecules on activated cells and consequently in cell migration patterns [9]. Indeed, there is now evidence supporting the view that the diminution of CD4 cell numbers in blood does not reflect comparable depletion in lymphoid tissue [11].

The effect of HIV infection on proliferation is the net result of different forms of ‘activation’. Broadly, there are two modes of T cell activation, depending on the pattern of stimulation (abrupt or continuous) to which particular cells are exposed over time [12–16]: ‘full activation’, leading to proliferation and often also to activation-associated cell death; and ‘partial activation’, resulting in a selective expression of some cellular functions but also in transient resistance to full activation (‘anergy’). In HIV-infected tissues, both forms of activation are induced, because the level and pattern of stimulation of individual cells is not uniform. Depending on the overall balance of the induction of cell proliferation, cell death, or cellular resistance to both, the steady-state number of cells in a cell population can either increase or decrease as a result of systemic activation. CD4 and CD8 lymphocytes are inherently different in terms of their response characteristics, and therefore, for the two subsets, the balance among these inducible cellular functions can be quite different. This could result in a substantial change in the cell-subset ratio compared with normal. Many of the anergic cells may be HIV-specific, but a large proportion may be non-HIV-specific lymphocytes. Such T cells are partially activated as ‘bystanders’ [17] in specific HIV–T cell encounters [8]. We and others [8, 12, 13, 18] postulated the concept of ‘infectious induction of anergy’, in which cells that respond to their own nominal antigens in the presence of partially activated cells tend themselves to become partially rather than fully activated and thus are recruited into the anergic pool, so that the state of ‘anergy’ is propagated by successive cycles of anergy induction and recruitment. There is now considerable evidence for such ‘infectious’ suppression of full activation and anergy induction [19]. A testable prediction is that a large proportion of phenotypically activated cells from lymphoid tissues of HIV-infected individuals is resistant to full activation and hence also to HIV replication, and will therefore display a degree of anergy. This interpretation is consistent with the functional ‘defects’ that occur in peripheral T cells from HIV-infected persons and that are also described in the article by Vingerhoets et al. in this issue [1].

Resistance of CD4 T cells to full activation limits HIV replication in this population. Therefore, we view the modified pattern of homeostasis, recirculation, and function of the cells of the immune system not as ‘dysregulation’ but as manifestations of effective host response to chronic HIV infection. AIDS occurs either because these modifications eventually become excessive, impairing the ability of the immune system to protect the host from opportunistic infections, or, because of a change in the mode of HIV dissemination, occurring when the viral load exceeds a certain threshold, which changes the mode of infection from local transmission of HIV during cell–cell interactions to efficient infection of activated CD4 T cells by ambient virus, resulting in accelerated destruction of immune cells by HIV ([20]; Z. Grossman, M. B. Feinberg and W. E. Paul, in preparation).

If manifestations of HIV infection reflect an effective host response and a general strategy against chronic infections [8], then similar manifestations should be observed in other chronic infections. Our recent observations in an Ethiopian immigrant population in Israel having a very high prevalence of several infectious diseases, particularly helminth infections [21], support this conjecture. All immigrants that were not infected by HIV had substantial markers of persistent immune activation [22]. This chronic immune activation was caused by the helminth infections and not any other infections or genetic factors (Kalinkovich et al., unpublished observations). In particular, activation was associated with remarkable changes in the distribution and phenotype of peripheral blood T cell populations: a marked decrease in CD4 cell numbers with significantly decreased mean CD4/CD8 ratios [23]; significantly elevated proportion of activated cells (HLA-DR/CD3, HLA-DR/CD4 and HLA-DR/CD8-positive cells); decrease in proportion of naive CD4 cells and concomitant increase in memory CD4 cells; and markedly increased rates of apoptosis [24]. The phenotypic and functional cellular changes observed by us in these HIV− individuals resemble those described in HIV-infected people [25–29]. Further comparative studies are required to assess the commonality of other functional changes, including decreased cellular responses to recall antigens, to allo-stimulation and to mitogens. Although we expect the chronic immune activation by infectious agents to render infected people's cells more susceptible to HIV infection compared with healthy people [7], because of the difference in the numbers of conventionally activated cells, we would also expect chronic activation to protect a proportion of the activated cells against de novo productive infection. Taken together, these findings lend further support to our view [8, 20] that many of the functional changes that occur during the asymptomatic phase of HIV infection reflect the chronic activation of the immune system, which reflects a normal mode of response to chronic infection.

References