Soon after the discovery of AIDS viruses it became apparent that natural hosts of SIV do not generally develop immunodeficiency in association with SIV infection, whereas these viruses readily induce disease in Asian nonhuman primates and humans 
. However, the mechanisms employed by these primates to avoid the pathogenic consequences of SIV-infection remain unclear.
In the present study, we made a direct comparison of the role of SIV-specific adaptive immune responses in a nonnatural host and natural host of SIV infection, PTM and vervet AGM, respectively. To do this, we evaluated the effect of antibody-mediated temporal inhibition of cellular and humoral immune responses during primary infection with the uncloned SIV virus, SIVagmVer90
, in PTM and AGM. Recent investigations have shown that this virus does not induce an AIDS-like disease in AGM but is not inherently nonpathogenic as infection studies in PTM have shown 
. Here in this study, temporal inhibition of adaptive immune responses in primary SIV infection of PTM resulted in an increased peak and set point viremia and accelerated disease progression similar to observations recently made in CD8+ lymphocyte-depleted rhesus macaques infected with pathogenic SIV 
. Interestingly, in the present study and in unpublished observations evaluating sabaeus AGM (R.C. Zahn et al.), peak viremia was not increased in lymphocyte-depleted AGM but only a relatively brief prolongation of peak viremia was observed during primary infection. The delay in resolution of primary viremia was very likely due to the inhibition of cellular immune responses since primary viremia had resolved before the appearance of humoral immune responses. Also, the eventual generation of humoral immune responses in B cell-depleted animals following reappearance of these cells did not appear to have a significant influence on the magnitude of viremia. This observation was recently confirmed by others when B cell depletion during primary and chronic SIV infection of AGM did not result in an increased viremia or clinical signs of illness (personal communications with Ivona Pandrea). Thus, the data presented here and unpublished observations using sabaeus AGM (R.C. Zahn et al.) suggest that cellular immune responses contribute to viral containment in AGM but humoral immune responses appear to be less critical. However, in contrast to rhesus monkeys the absence of CD8+ lymphocytes in AGM resulted in a much more subdued impact on viremia, similar to observations recently made in sooty mangabeys 
Although no overt signs of SIV disease were seen in the lymphocyte-depleted AGM, the depletion of these cells was associated with reactivation of CMV. A similar transient reactivation of CMV without clinical signs was observed in one of the two PTM that were lymphocyte-depleted but not challenged with SIV. The most significant signs of pathogenicity were seen in PTM that were depleted of CD8+ and CD20+ lymphocytes and inoculated with SIV: in these animals, we observed a massive increase in plasma CMV DNA copies, precipitous loss of CD4+ T cells and an increase in naive/memory CD4+ T cell ratio, indicating a rapid loss of memory CD4+ T cells.
These observations raise a number of questions. Is it possible that the impairment of adaptive immune responses in AGM was not of sufficient duration to negatively impact the health of these animals? Or, is the inherent ability to suppress immune activation in SIV-infected AGM the critical factor that helps AGM to cope with chronic SIV infection?
CD8+ lymphocyte depletion in rhesus macaques results in a significantly enhanced disease progression. However, as recently shown 
and observed here in the PTM depleted of CD8+ lymphocytes, the fastest disease acceleration is seen in nonnatural hosts when CD8+ lymphocytes are depleted for at least the first 4 weeks during primary SIV infection. However, the CD8+ lymphocyte depletion in most AGM in this study was of fairly short duration. But even the one relatively long-term CD8+ lymphocyte-depleted AGM (A13; about 6 weeks) did not develop an AIDS like-disease with rapid disease progression as we have seen in all rhesus macaques studied so far with a similar length of CD8+ lymphocyte-depletion 
. In addition, PTM P24 in this study also was only depleted for 2 weeks, but succumbed to AIDS within 18 weeks p.i.
A number of recent investigations have shown that natural hosts exhibit a much lower level of immune activation during chronic viremia compared to nonnatural hosts 
. The low level immune activation may protect the natural host species from more aggressive virus replication and the development of an AIDS-like disease. Recent investigations have shown that short-term immune activation using LPS or an IL-2/diphtheria toxin fusion protein in AGM can result in an increased viremia, supporting the notion that hyperactivation of the immune system plays a role in virus replication and disease progression 
. However, these brief in vivo
manipulations had no apparent effect on the health of the animals. Thus, in vivo
manipulations in AGM that are capable of suppressing adaptive immune responses for a longer duration than in the present study and/or induce a prolonged immune activation may result in a different outcome.
In addition, there may be a number of possible caveats in our study. (1). Since CD4+ T cells in vervet AGM dimly coexpress the CD8α molecule, administration of the anti-CD8α antibody may have also affected CD4+ T cell targets and thus limited virus replication. However, a similar result was recently observed in sabaeus AGM that did not coexpress the CD8α molecule on CD4+ T cells (R.C. Zahn et al., unpublished observations). We also cannot clearly rule out that NK cells that express the CD8α molecule may contribute to viral containment as these cells are also depleted by the anti-CD8α antibody. It is also possible that some of the differences observed in the control group and the antibody-treated group may be due to utilizing historical controls for this study. However, both the historical controls and antibody-treated group were treated identically except for receiving lymphocyte-depleting antibodies. Finally, it is conceivable that the combined depletion of CD8+ and CD20+ lymphocytes may affect AGM and PTM differently. However, to formally rule out that the antibody treatment may have a more significant pathogenic effect on PTM we have performed the lymphocyte depletion experiments as well in SIV-negative AGM and PTM which both did not exhibit any signs of disease following the antibody administrations.
Recent investigations have shown that evolutionary adaptations in natural hosts (paucity of CCR5+ cells, decreased immune activation, and ability to down modulate the CD4 molecule on CD4+ T cells) may assist adaptive immune responses or may even render SIV-specific adaptive immune responses unnecessary 
. In contrast, if the AIDS virus can bypass restriction factors in nonnatural hosts, adaptive immune responses appear to be the major defense against uncontrolled virus replication. However, the eventual failure of viral control is due to inevitable viral immune escape 
. A sign of the incredible plasticity of the immune system of natural hosts to cope with SIV infection was seen in one of the lymphocyte-depleted AGM (A346). This animal eventually lost all of its peripheral blood CD4+ T cells (both memory and naïve cells), suggesting that the virus in this animal might have changed coreceptor usage (characterization of coreceptor usage is still ongoing). Even with an almost complete loss of peripheral blood CD4+ T cells the animal showed no signs of disease. Recently, a similarly abrupt decline in CD4+ T cells was observed in another natural host of SIV, sooty mangabeys, upon emergence of a CXCR4-tropic SIV variant without inducing disease 
. This abrupt decline of CD4+ T cells in sooty mangabeys does not necessarily always depend on switching the tropism of the virus to CXCR4 
. AGM may be capable of utilizing a large fraction of CD4- T cells, which can be found in peripheral blood and tissues, as surrogate T helper cells 
Investigations into natural hosts of SIV, including AGM, will allow us to understand how these animals can coexist with SIV without developing disease. The observations made here and in sabaeus AGM (R.C. Zahn et al., unpublished observations) suggest that CD8+ T cell responses participate to some degree in controlling viral replication in natural hosts. However, the effects were considerably more limited than observations made in macaques and it is not clear whether a more long-term increase in viremia would precipitate disease progression in AGM. Further investigations are required to assess the relative contribution of adaptive immune responses versus non-adaptive mechanisms in the maintenance of an AIDS-free course of infection in natural host species. Our aim is that these investigations will provide clues how pathogenic AIDS virus infections could be limited, identify new therapeutic approaches, and contribute to the development of a successful HIV vaccine.