CD8+ T Cell Dynamics during Primary Simian Immunodeficiency Virus Infection in Macaques: Relationship of Effector Cell Differentiation with
http://www.100md.com
免疫学杂志 2005年第11期
Abstract
Immunological and virological events that occur during the earliest stages of HIV-1 infection are now considered to have a major impact on subsequent disease progression. We observed changes in the frequencies of CD8bright T cells expressing different chemokine receptors in the peripheral blood and lymph nodes of rhesus macaques during the acute phase of the pathogenic SIVmac251 infection; the frequency of CD8bright T cells expressing CXCR4 decreased, while the frequency of those expressing CCR5 increased. These reciprocal changes in chemokine receptor expression were associated with changes in the proportion of cycling (Ki67+) CD8bright T cells, and with the pattern of CD8bright T cell differentiation as defined by expression of CCR7 and CD45RA. In contrast, during the primary phase of the attenuated SIVmac251nef infection, no major change was observed. Whereas during the acute phase of the infection with pathogenic SIV (2 wk postinfection) no correlate of disease protection was identified, once the viral load set points were established (2 mo postinfection), we found that the levels of cycling and of CCR5- and CXCR4-positive CD8bright T cells were correlated with the extent of viral replication and therefore with SIV-infection outcome. Our data reveal that, during primary SIV infection, despite intense CD8 T cell activation and an increase in CCR5 expression, which are considered as essential for optimal effector function of CD8+ T cells, these changes are associated with a poor prognosis for disease progression to AIDS.
Introduction
Chemokines and chemokine receptors play a critical role in T cell homeostasis and in the migration of various cell types including memory and effector T cells (1, 2, 3, 4). Two important subdivisions of chemokine receptors are those which control recirculation of resting naive and memory cells through T cell areas of secondary lymphoid tissue, CXCR4 and CCR7, and those which direct activated and effector T cells under inflammatory conditions, CXCR3 and CCR5 (5, 6, 7, 8, 9). In antiviral immunity, this localization of effector CD8+ T cells plays an extremely important role in promoting host recovery and virus clearance (10, 11, 12). The primary acute phase of HIV type-1 (HIV-1) and SIV infections is characterized by an early burst of viral replication, an exponential increase in plasma viral load, the dissemination and seeding of the virus in all peripheral lymphoid organs, and the induction of the host immune response against the virus (13, 14, 15, 16, 17, 18). The plasma viral load reaches a steady-state at the end of this primary phase, 2–6 mo after infection that is predictive of progression toward AIDS (19, 20, 21, 22, 23).
CD8+ T cells are thought to play an important role in controlling HIV and SIV viral replication and an HIV-specific CTL response is rapidly induced at the same time as the decline in viral titers (18, 24, 25, 26, 27). However, HIV-infected individuals usually fail to control virus replication efficiently. Several reports have shown an apparent nonresponsiveness of circulating CD8+ T cells that may reflect one mechanism by which immunodeficiency viruses escape the immune response (28, 29, 30, 31, 32).
However, an optimal antiviral defense requires efficient mechanisms for targeting activated T cells to sites of infection. Insight into the regulation of CD8 T cell trafficking is therefore essential to understanding how viral infections are controlled. CCR5 expression seems to be required for optimal CD8 effector T cell function in vivo and CCR5 may be involved in the recruitment and positioning of virus-specific T cells (33, 34, 35). In HIV-1 infection in particular, where CCR5+CD4+ T cells are the main targets in early infection, colocalization of CCR5+ effector CD8+ T cells would be predicted to be especially important. In contrast, CCR7 functions as a homing receptor in migration of CD8+ T cells to the lymph nodes (LNs), 4 but is down-regulated after stimulation with Ag (36, 37). The following differentiation lineage for CD8+ T cells has therefore been proposed: CD45RA+CCR7+ (naive) CD45RA-CCR7+ (central memory, TCM) CD45RA-CCR7– (effector memory, TEM) CD45RA+CCR7– (terminally differentiated T cells, TDT) (32). Moreover, the CCR7– subset of CD8+ T cells express perforin-containing granules (a feature of CTLs) (38, 39). It is unclear how the expression of chemokine receptors on CD8+ T cells is affected by the physiological response during primary HIV infection.
Circulating lymphocytes represent only 2% of all T lymphocytes, whereas LN lymphocytes account for 30% of the lymphocyte pool (40) and are considered as the main sites of both ongoing T cell activation and viral replication. The SIV-infection model is especially well-suited for analyzing T cell dynamics and changes in chemokine receptor expression on CD8+ T cells during primary infection, because it is possible to control precisely the inoculum used, the timing and route of inoculation, and to obtain LNs early in infection. In contrast, study of acute HIV-1 infection is extremely limited by availability of both peripheral blood and lymphoid tissue, and in nearly all cases involves highly variable cross-sectional data rather than longitudinal data from different individuals with different outcomes. In this study rhesus macaques were either infected with the pathogenic SIVmac251 strain or the attenuated SIVmac251nef molecular clone. We looked at the relationships between changes in chemokine receptor expression on CD8+ T cells, in both peripheral blood and LNs, and the rate of viral replication and progression to AIDS.
Materials and Methods
Statistical analyses
A Mann-Whitney U test was used to determine whether differences in means were significant. Differences were considered to be significant if p < 0.05. Spearman Rank correlations were calculated to evaluate correlations. Best-fit lines are shown.
Results
Dynamics of viral replication and cell counts in acute SIVmac251 infection
Ten rhesus macaques (seven females and three males) were infected with 10 animal-infectious doses of the pathogenic SIVmac251 isolate i.v. Animals were 8.50 ± 0.97 years old and had an initial CD4 T cell count in the blood of 875/mm3± 518. Infection of rhesus macaques with the SIVmac251 resulted in a peak of viremia (mean: 1.3 x 107 SIV RNA copies/ml, range: 2.6 x 105 to 3.6 x 107) by day 14 (Fig. 1). This peak was followed by a decrease in viremia that reached a steady state at 2 mo at different levels in different individuals. The steady states were 2 logs lower in five SIVmac251-infected macaques (nos. 0191, 0300, 0341, 0312, 94870; mean: 2.7 x 103 SIV RNA copies/ml, range: 102 to 104) than in the other five monkeys (nos. 94746, 94852, 94856, R97087, 94860; mean: 1.2 x 105 SIV RNA copies/ml, range: 2 x 104 to 3.4 x 105; p = 0.009). These set points are predictive of further progression to AIDS, identifying slow and moderate progressors, respectively (23, 42). Three animals (nos. 94746, 94852, and R97087), followed for clinical progression toward disease, developed AIDS (a wasting syndrome with cachexia and opportunistic infections) 1–2 years after infection (Table I). Two others (nos. 94856 and 94860) were asymptomatic during the year follow-up period of the study but they lost around 50% of their initial CD4+ T cell counts and displayed viral loads higher than 104 copies/ml. The five other primates remained asymptomatic during the study having lost only around 25% of the CD4+ T cells and with viral loads remaining <104 copies/ml (Table I). In all infected primates, anti-SIV Abs were detectable in blood 4 wk postinfection. The concentration of these Abs increased thereafter (data not shown). According to our recent results (23, 42), none of this new group of infected primates were rapid progressors, i.e., they did not develop AIDS (wasting syndrome with cachexia, opportunistic infections, and an absence of SIV Ab response) within 6 mo. To determine the extent of the early viral seeding in peripheral lymphoid organs, we also analyzed peripheral LNs. These LNs were sequentially retrieved from each of these animals 0, 7, 14 days and 2 mo after infection. The numbers of SIV productively infected cells peaked on day 14 postinfection (Fig. 1) and decreased by the end of this early phase. The set point levels at 2 mo postinfection were significantly lower in slow progressors (0.24 ± 0.24/2 mm2) than in moderate progressors (2.78 ± 2.84/2 mm2; p = 0.009). The set point level of SIV+ RNA-expressing cells at 2 mo postinfection was correlated with the plasma viral load at this time (r = 0.85, p = 0.006). Within the first week of infection, global lymphopenia (B and T cells), around the peak of replication, was observed in peripheral blood (Table II). This observation is in agreement with a recent report on SIV-infected macaques (44). Two months after infection, CD8+ T cells and B cells were restored to their preinfection levels or even higher, whereas CD4+ T cells remained at a lower level than before infection. The extent of further CD4 T cell decline, as mentioned before, is predictive of further disease evolution toward AIDS (Table I). In contrast to that observed in the blood, LN cell counts revealed that the size of the lymphocyte pools, 2 wk postinfection, is maintained or even higher whereas the enlargement of LNs at 2 mo postinfection was associated with a 2- to 3-fold increase in the pool of CD8+ and CD20+ cells (Table II). We do not know whether the early lymphopenia was due to cell death, cell redistribution from the blood into tissue sites of SIV replication, or both.
Discussion
Efficient initiation of the adaptive immune response upon viral infection is essential to combat rapidly replicating infectious pathogens. However, our understanding of how HIV and SIV avoid the immune system and become chronic infections remains incomplete, and is currently one of the greatest challenges to the rational design of effective HIV vaccines. Our findings indicate that the extent of viral replication at 2 mo correlates with a decrease in the proportion of CD8bright T cells expressing CXCR4 and an increase in CCR5 expression and cycling cells in the LNs. These changes were predictive of further disease evolution toward AIDS, defining slow and moderate progressors. In contrast, CD8 T cell changes occurring earlier, at day 14, did not differ between these two groups of SIV-infected macaques. Therefore, these results suggest, at least in part, the dramatic decrease in viremia occurring after the peak in slow progressors may not depend exclusively on an effective CD8 immune response to the virus.
CCR5 is considered as crucial in the recruitment and positioning of virus-specific T cells (33, 34, 35). Here, we found after in vitro stimulation that activated CD8bright T cells express high levels of CCR5 but lower CXCR4. In mice, it has been also reported that T cytotoxic type 1 effector CD8+ T cells, which produce type-1 cytokines, predominantly express CCR5 while T cytotoxic type 2 effector CD8+ T cells, which produce type-2 cytokines, express high levels of CXCR4 but not CCR5 (60). Our results on the dynamics of chemokine receptor expression upon pathogenic SIVmac251 infection may be at least in part the consequence of in vivo CD8 T cell activation as indicated by Ki67 staining and changes in the lineage commitment of T cell subpopulations. In nonpathogenic SIVmac251nef, despite intense viral replication at day 14, we did not observe major changes in the dynamics of CD8bright T cells, especially in the expression of CCR5. These observations suggest that during the acute phase, a certain threshold of SIV Ags is required to activate CD8+ T cells and trigger the expression of CCR5. However, the differences in CD8 phenotype (CXCR4 increase) in the SIVmac251nef could still suggest an indirect effect of nef in pathogenic SIV-infected macaques. Recently, we have observed that, during primary HIV infection, CCR5 expression and proliferation of CCR5+CD8+ T cells were also higher in HIV-infected individuals than in healthy donors (61). There is also evidence that viral infections induced, on virus epitope-specific tetramer-positive CD8+ T cells, cell surface expression of CCR5 (62, 63). Similar to humans (47, 50, 64), CCR5 is mainly expressed on the CD45RAlow T cell subset in macaques and CCR5 expression has been reported to be restricted to memory and effector CD8+ T cells but not naive CD8+ T cells (62). Although we have not directly shown the level of chemokine receptor on the different CD8+ T cell subsets, due to the limited panel of labeled Abs that cross-react in macaques, we found an increase in the proportion of effector memory (CD45RA–CCR7–) CD8bright T cells in moderate progressors concomitant with CCR5 increase. Moreover, our data showing that RANTES and MIP-1 induce a greater CD8 T cell migration in SIV-infected macaques than in healthy macaques confirm that CCR5 expressed on CD8+ T cells physiologically functions as a receptor for chemokines.
CCR5 is widely expressed on T lymphocytes in many different organs particularly at sites of inflammation (65, 66, 67, 68). The injection of RANTES into a chimeric SCID mouse model has been shown to recruit CD8+ T cells to the sites of injection (69), while another report demonstrated the localization of infused CCR5 HIV-specific T cells to inflamed tissue (70). Our results during primary SIV infection thus support the idea that these CCR5+CD8+ T cells, activated within the LNs, may re-enter the circulation, and be capable of directional migration to the inflammatory sites to control viral replication (33, 34, 35). Therefore, because CCR5 expression has been proposed to be essential for optimal CD8+ effector T cell function and exert its effector potential at the site of virus infection (33, 34, 35), our data raises a major issue. Indeed, despite higher activation states, higher numbers of cycling CD8+ T cells, and higher proportions of CCR5 expressing CD8+ T cells, the levels of plasma viremia set point and SIV+ replicating cells within LNs remained higher in moderate progressors than in slow progressors. These data support the idea that despite intense CD8+ T cell activation and proliferation within LNs, these cells are inefficient in controlling the virus replication during primary infection.
During the asymptomatic phase, it has been proposed that the CD8 T cell response fails to control HIV and SIV infections because CD8+ T cells are defective in perforin expression, but not in cytokine production, which could render them less efficient at killing virus-infected cells (59, 71). Here, we found that the proportion of TIA-1-expressing CD8bright T cells in the LNs is greater in moderate progressors than in slow progressors at 2 mo postinfection. TIA-1 expressing CD8bright T cells were, at 2 mo, in fact characterized mostly by effector memory phenotype (CD45RA–CCR7–).
A skewed maturation of HIV-specific memory CD8+ T cells has been previously reported in inducing the accumulation of a preterminally differentiated subset of memory T cells during the asymptomatic phase (32). Although only five animals were analyzed, we found in moderate progressors the accumulation of effector memory CD8+ T cells. The absence of a fully differentiated profile might be due to a deregulation of the cell cycle making CD8+ T cells more anergic and prone to undergo apoptosis as observed during the asymptomatic phase (30, 31, 53, 72, 73, 74). In this sense, it has been reported that HIV-specific CTL clones rapidly disappear (29), and activated CCR5+CD8+ T cells are more prone to undergo apoptosis in vitro during primary HIV infection (61). We and other laboratories have previously reported that CD45RA–CD8+ T cells are more prone to undergo apoptosis than CD45RA+ during the asymptomatic phase (31, 53, 75). Therefore, whether the abortive differentiation pattern in moderate progressors was the result of an increased death of either effector memory and/or terminal differentiated T cells, or both, during primary infection remains an open question.
Increasing evidence suggests an association between high levels of immune activation during the asymptomatic phase and poor outcome in HIV-infected individuals (76, 77) and in SIV-infected macaques (42). A recent report suggest that the CD8+ T cell activation "set point" is also an early predictive marker of further disease evolution toward AIDS in HIV-infected individuals (78) which is in agreement with our data. Moreover, introduction of antiretroviral treatment resulted in rapid decreases in the level of CD8 T cell activation (79, 80) supporting the idea that the threshold of viral replication determines the extent of CD8 T cell activation. In such chronic infection, viral immune evasion could be due to persistence of activated CD8+ T cells as proposed by Zajac et al. (81) following an examination of the regulation of virus-specific CD8+ T cells during chronic lymphocytic choriomeningitis infection of mice. This unresponsiveness was higher under conditions of CD4 T cell deficiency. Thus, the lack of CD4 T cell help may render these CD8+ T cells non-fully differentiated during primary SIV infection and warrants further exploration (82, 83, 84).
Because the duration of antigenic stimulation of CD8+ T cells can determine whether subsequent in vivo clonal expansion will be abortive or extensive (85) and that sustained stimulus may induce an extensive proliferation and production of cells potentially capable of inducing destruction of peripheral target tissues, it cannot be excluded that, although provocative, the CD8 T cell response may have a detrimental effect during SIV infection rather than a protective effect. Other reports have shown an active participation of activated CCR5+CD8+ T cells in the pathogenesis of graft-vs-host-disease (66).
Despite the intense T cell activation and changes in chemokine receptor expression that occur during primary infection, related to the extent of viral replication and further disease progression to AIDS, our results highlight the inability to control viremia that may be, at least in part, due to the lack of maturation of CD8bright T cells during the acute phase. Better understanding of the mechanisms involved should improve our knowledge of the early steps in the inability of the immune response to control HIV and SIV replication and prevent further progression to AIDS.
Acknowledgments
We acknowledge M. Muller-Trutwin and A. Brussel for helpful discussions and critical reading of our manuscript.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was funded by a grant from the Agence Nationale de Recherche sur le Sida (ANRS; to J.E.). F.P. was supported by a postdoctoral fellowship from Ensemble Contre le Sida and L.V. was supported by a doctoral fellowship from ANRS.
2 L.V. and F.P. were equal contributors.
3 Address correspondence and reprint requests to Dr. Jér?me Estaquier, Unité de Physiopathologie des Infections Lentivirales, Institut Pasteur, 28 rue du Docteur Roux, 75724 Paris cedex 15, France. E-mail address: jestaqui{at}pasteur.fr
4 Abbreviations used in this paper: LN, lymph node.
Received for publication May 14, 2004. Accepted for publication March 11, 2005.
References
Cyster, J. G.. 1999. Chemokines and cell migration in secondary lymphoid organs. Science 286: 2098-2103.[
Butcher, E. C., M. Williams, K. Youngman, L. Rott, M. Briskin. 1999. Lymphocyte trafficking and regional immunity. Adv. Immunol. 72: 209-253.
Kim, C. H., H. E. Broxmeyer. 1999. Chemokines: signal lamps for trafficking of T and B cells for development and effector function. J. Leukocyte Biol. 65: 6-15.
Sallusto, F., C. R. Mackay, A. Lanzavecchia. 2000. The role of chemokine receptors in primary, effector, and memory immune responses. Annu. Rev. Immunol. 18: 593-620.
Mackay, C. R.. 1996. Chemokine receptors and T cell chemotaxis. J. Exp. Med. 184: 799-802.
Sallusto, F., A. Lanzavecchia, C. R. Mackay. 1998. Chemokines and chemokine receptors in T-cell priming and Th1/Th2-mediated responses. Immunol. Today 19: 568-574.
Premack, B. A., T. J. Schall. 1996. Chemokine receptors: gateways to inflammation and infection. Nat. Med. 2: 1174-1178.
Wells, T. N., C. A. Power, A. E. Proudfoot. 1998. Definition, function and pathophysiological significance of chemokine receptors. Trends Pharmacol. Sci. 19: 376-380.
Baggiolini, M.. 1998. Chemokines and leukocyte traffic. Nature 392: 565-568.(Valérie Monceaux*, Lauren)
Immunological and virological events that occur during the earliest stages of HIV-1 infection are now considered to have a major impact on subsequent disease progression. We observed changes in the frequencies of CD8bright T cells expressing different chemokine receptors in the peripheral blood and lymph nodes of rhesus macaques during the acute phase of the pathogenic SIVmac251 infection; the frequency of CD8bright T cells expressing CXCR4 decreased, while the frequency of those expressing CCR5 increased. These reciprocal changes in chemokine receptor expression were associated with changes in the proportion of cycling (Ki67+) CD8bright T cells, and with the pattern of CD8bright T cell differentiation as defined by expression of CCR7 and CD45RA. In contrast, during the primary phase of the attenuated SIVmac251nef infection, no major change was observed. Whereas during the acute phase of the infection with pathogenic SIV (2 wk postinfection) no correlate of disease protection was identified, once the viral load set points were established (2 mo postinfection), we found that the levels of cycling and of CCR5- and CXCR4-positive CD8bright T cells were correlated with the extent of viral replication and therefore with SIV-infection outcome. Our data reveal that, during primary SIV infection, despite intense CD8 T cell activation and an increase in CCR5 expression, which are considered as essential for optimal effector function of CD8+ T cells, these changes are associated with a poor prognosis for disease progression to AIDS.
Introduction
Chemokines and chemokine receptors play a critical role in T cell homeostasis and in the migration of various cell types including memory and effector T cells (1, 2, 3, 4). Two important subdivisions of chemokine receptors are those which control recirculation of resting naive and memory cells through T cell areas of secondary lymphoid tissue, CXCR4 and CCR7, and those which direct activated and effector T cells under inflammatory conditions, CXCR3 and CCR5 (5, 6, 7, 8, 9). In antiviral immunity, this localization of effector CD8+ T cells plays an extremely important role in promoting host recovery and virus clearance (10, 11, 12). The primary acute phase of HIV type-1 (HIV-1) and SIV infections is characterized by an early burst of viral replication, an exponential increase in plasma viral load, the dissemination and seeding of the virus in all peripheral lymphoid organs, and the induction of the host immune response against the virus (13, 14, 15, 16, 17, 18). The plasma viral load reaches a steady-state at the end of this primary phase, 2–6 mo after infection that is predictive of progression toward AIDS (19, 20, 21, 22, 23).
CD8+ T cells are thought to play an important role in controlling HIV and SIV viral replication and an HIV-specific CTL response is rapidly induced at the same time as the decline in viral titers (18, 24, 25, 26, 27). However, HIV-infected individuals usually fail to control virus replication efficiently. Several reports have shown an apparent nonresponsiveness of circulating CD8+ T cells that may reflect one mechanism by which immunodeficiency viruses escape the immune response (28, 29, 30, 31, 32).
However, an optimal antiviral defense requires efficient mechanisms for targeting activated T cells to sites of infection. Insight into the regulation of CD8 T cell trafficking is therefore essential to understanding how viral infections are controlled. CCR5 expression seems to be required for optimal CD8 effector T cell function in vivo and CCR5 may be involved in the recruitment and positioning of virus-specific T cells (33, 34, 35). In HIV-1 infection in particular, where CCR5+CD4+ T cells are the main targets in early infection, colocalization of CCR5+ effector CD8+ T cells would be predicted to be especially important. In contrast, CCR7 functions as a homing receptor in migration of CD8+ T cells to the lymph nodes (LNs), 4 but is down-regulated after stimulation with Ag (36, 37). The following differentiation lineage for CD8+ T cells has therefore been proposed: CD45RA+CCR7+ (naive) CD45RA-CCR7+ (central memory, TCM) CD45RA-CCR7– (effector memory, TEM) CD45RA+CCR7– (terminally differentiated T cells, TDT) (32). Moreover, the CCR7– subset of CD8+ T cells express perforin-containing granules (a feature of CTLs) (38, 39). It is unclear how the expression of chemokine receptors on CD8+ T cells is affected by the physiological response during primary HIV infection.
Circulating lymphocytes represent only 2% of all T lymphocytes, whereas LN lymphocytes account for 30% of the lymphocyte pool (40) and are considered as the main sites of both ongoing T cell activation and viral replication. The SIV-infection model is especially well-suited for analyzing T cell dynamics and changes in chemokine receptor expression on CD8+ T cells during primary infection, because it is possible to control precisely the inoculum used, the timing and route of inoculation, and to obtain LNs early in infection. In contrast, study of acute HIV-1 infection is extremely limited by availability of both peripheral blood and lymphoid tissue, and in nearly all cases involves highly variable cross-sectional data rather than longitudinal data from different individuals with different outcomes. In this study rhesus macaques were either infected with the pathogenic SIVmac251 strain or the attenuated SIVmac251nef molecular clone. We looked at the relationships between changes in chemokine receptor expression on CD8+ T cells, in both peripheral blood and LNs, and the rate of viral replication and progression to AIDS.
Materials and Methods
Statistical analyses
A Mann-Whitney U test was used to determine whether differences in means were significant. Differences were considered to be significant if p < 0.05. Spearman Rank correlations were calculated to evaluate correlations. Best-fit lines are shown.
Results
Dynamics of viral replication and cell counts in acute SIVmac251 infection
Ten rhesus macaques (seven females and three males) were infected with 10 animal-infectious doses of the pathogenic SIVmac251 isolate i.v. Animals were 8.50 ± 0.97 years old and had an initial CD4 T cell count in the blood of 875/mm3± 518. Infection of rhesus macaques with the SIVmac251 resulted in a peak of viremia (mean: 1.3 x 107 SIV RNA copies/ml, range: 2.6 x 105 to 3.6 x 107) by day 14 (Fig. 1). This peak was followed by a decrease in viremia that reached a steady state at 2 mo at different levels in different individuals. The steady states were 2 logs lower in five SIVmac251-infected macaques (nos. 0191, 0300, 0341, 0312, 94870; mean: 2.7 x 103 SIV RNA copies/ml, range: 102 to 104) than in the other five monkeys (nos. 94746, 94852, 94856, R97087, 94860; mean: 1.2 x 105 SIV RNA copies/ml, range: 2 x 104 to 3.4 x 105; p = 0.009). These set points are predictive of further progression to AIDS, identifying slow and moderate progressors, respectively (23, 42). Three animals (nos. 94746, 94852, and R97087), followed for clinical progression toward disease, developed AIDS (a wasting syndrome with cachexia and opportunistic infections) 1–2 years after infection (Table I). Two others (nos. 94856 and 94860) were asymptomatic during the year follow-up period of the study but they lost around 50% of their initial CD4+ T cell counts and displayed viral loads higher than 104 copies/ml. The five other primates remained asymptomatic during the study having lost only around 25% of the CD4+ T cells and with viral loads remaining <104 copies/ml (Table I). In all infected primates, anti-SIV Abs were detectable in blood 4 wk postinfection. The concentration of these Abs increased thereafter (data not shown). According to our recent results (23, 42), none of this new group of infected primates were rapid progressors, i.e., they did not develop AIDS (wasting syndrome with cachexia, opportunistic infections, and an absence of SIV Ab response) within 6 mo. To determine the extent of the early viral seeding in peripheral lymphoid organs, we also analyzed peripheral LNs. These LNs were sequentially retrieved from each of these animals 0, 7, 14 days and 2 mo after infection. The numbers of SIV productively infected cells peaked on day 14 postinfection (Fig. 1) and decreased by the end of this early phase. The set point levels at 2 mo postinfection were significantly lower in slow progressors (0.24 ± 0.24/2 mm2) than in moderate progressors (2.78 ± 2.84/2 mm2; p = 0.009). The set point level of SIV+ RNA-expressing cells at 2 mo postinfection was correlated with the plasma viral load at this time (r = 0.85, p = 0.006). Within the first week of infection, global lymphopenia (B and T cells), around the peak of replication, was observed in peripheral blood (Table II). This observation is in agreement with a recent report on SIV-infected macaques (44). Two months after infection, CD8+ T cells and B cells were restored to their preinfection levels or even higher, whereas CD4+ T cells remained at a lower level than before infection. The extent of further CD4 T cell decline, as mentioned before, is predictive of further disease evolution toward AIDS (Table I). In contrast to that observed in the blood, LN cell counts revealed that the size of the lymphocyte pools, 2 wk postinfection, is maintained or even higher whereas the enlargement of LNs at 2 mo postinfection was associated with a 2- to 3-fold increase in the pool of CD8+ and CD20+ cells (Table II). We do not know whether the early lymphopenia was due to cell death, cell redistribution from the blood into tissue sites of SIV replication, or both.
Discussion
Efficient initiation of the adaptive immune response upon viral infection is essential to combat rapidly replicating infectious pathogens. However, our understanding of how HIV and SIV avoid the immune system and become chronic infections remains incomplete, and is currently one of the greatest challenges to the rational design of effective HIV vaccines. Our findings indicate that the extent of viral replication at 2 mo correlates with a decrease in the proportion of CD8bright T cells expressing CXCR4 and an increase in CCR5 expression and cycling cells in the LNs. These changes were predictive of further disease evolution toward AIDS, defining slow and moderate progressors. In contrast, CD8 T cell changes occurring earlier, at day 14, did not differ between these two groups of SIV-infected macaques. Therefore, these results suggest, at least in part, the dramatic decrease in viremia occurring after the peak in slow progressors may not depend exclusively on an effective CD8 immune response to the virus.
CCR5 is considered as crucial in the recruitment and positioning of virus-specific T cells (33, 34, 35). Here, we found after in vitro stimulation that activated CD8bright T cells express high levels of CCR5 but lower CXCR4. In mice, it has been also reported that T cytotoxic type 1 effector CD8+ T cells, which produce type-1 cytokines, predominantly express CCR5 while T cytotoxic type 2 effector CD8+ T cells, which produce type-2 cytokines, express high levels of CXCR4 but not CCR5 (60). Our results on the dynamics of chemokine receptor expression upon pathogenic SIVmac251 infection may be at least in part the consequence of in vivo CD8 T cell activation as indicated by Ki67 staining and changes in the lineage commitment of T cell subpopulations. In nonpathogenic SIVmac251nef, despite intense viral replication at day 14, we did not observe major changes in the dynamics of CD8bright T cells, especially in the expression of CCR5. These observations suggest that during the acute phase, a certain threshold of SIV Ags is required to activate CD8+ T cells and trigger the expression of CCR5. However, the differences in CD8 phenotype (CXCR4 increase) in the SIVmac251nef could still suggest an indirect effect of nef in pathogenic SIV-infected macaques. Recently, we have observed that, during primary HIV infection, CCR5 expression and proliferation of CCR5+CD8+ T cells were also higher in HIV-infected individuals than in healthy donors (61). There is also evidence that viral infections induced, on virus epitope-specific tetramer-positive CD8+ T cells, cell surface expression of CCR5 (62, 63). Similar to humans (47, 50, 64), CCR5 is mainly expressed on the CD45RAlow T cell subset in macaques and CCR5 expression has been reported to be restricted to memory and effector CD8+ T cells but not naive CD8+ T cells (62). Although we have not directly shown the level of chemokine receptor on the different CD8+ T cell subsets, due to the limited panel of labeled Abs that cross-react in macaques, we found an increase in the proportion of effector memory (CD45RA–CCR7–) CD8bright T cells in moderate progressors concomitant with CCR5 increase. Moreover, our data showing that RANTES and MIP-1 induce a greater CD8 T cell migration in SIV-infected macaques than in healthy macaques confirm that CCR5 expressed on CD8+ T cells physiologically functions as a receptor for chemokines.
CCR5 is widely expressed on T lymphocytes in many different organs particularly at sites of inflammation (65, 66, 67, 68). The injection of RANTES into a chimeric SCID mouse model has been shown to recruit CD8+ T cells to the sites of injection (69), while another report demonstrated the localization of infused CCR5 HIV-specific T cells to inflamed tissue (70). Our results during primary SIV infection thus support the idea that these CCR5+CD8+ T cells, activated within the LNs, may re-enter the circulation, and be capable of directional migration to the inflammatory sites to control viral replication (33, 34, 35). Therefore, because CCR5 expression has been proposed to be essential for optimal CD8+ effector T cell function and exert its effector potential at the site of virus infection (33, 34, 35), our data raises a major issue. Indeed, despite higher activation states, higher numbers of cycling CD8+ T cells, and higher proportions of CCR5 expressing CD8+ T cells, the levels of plasma viremia set point and SIV+ replicating cells within LNs remained higher in moderate progressors than in slow progressors. These data support the idea that despite intense CD8+ T cell activation and proliferation within LNs, these cells are inefficient in controlling the virus replication during primary infection.
During the asymptomatic phase, it has been proposed that the CD8 T cell response fails to control HIV and SIV infections because CD8+ T cells are defective in perforin expression, but not in cytokine production, which could render them less efficient at killing virus-infected cells (59, 71). Here, we found that the proportion of TIA-1-expressing CD8bright T cells in the LNs is greater in moderate progressors than in slow progressors at 2 mo postinfection. TIA-1 expressing CD8bright T cells were, at 2 mo, in fact characterized mostly by effector memory phenotype (CD45RA–CCR7–).
A skewed maturation of HIV-specific memory CD8+ T cells has been previously reported in inducing the accumulation of a preterminally differentiated subset of memory T cells during the asymptomatic phase (32). Although only five animals were analyzed, we found in moderate progressors the accumulation of effector memory CD8+ T cells. The absence of a fully differentiated profile might be due to a deregulation of the cell cycle making CD8+ T cells more anergic and prone to undergo apoptosis as observed during the asymptomatic phase (30, 31, 53, 72, 73, 74). In this sense, it has been reported that HIV-specific CTL clones rapidly disappear (29), and activated CCR5+CD8+ T cells are more prone to undergo apoptosis in vitro during primary HIV infection (61). We and other laboratories have previously reported that CD45RA–CD8+ T cells are more prone to undergo apoptosis than CD45RA+ during the asymptomatic phase (31, 53, 75). Therefore, whether the abortive differentiation pattern in moderate progressors was the result of an increased death of either effector memory and/or terminal differentiated T cells, or both, during primary infection remains an open question.
Increasing evidence suggests an association between high levels of immune activation during the asymptomatic phase and poor outcome in HIV-infected individuals (76, 77) and in SIV-infected macaques (42). A recent report suggest that the CD8+ T cell activation "set point" is also an early predictive marker of further disease evolution toward AIDS in HIV-infected individuals (78) which is in agreement with our data. Moreover, introduction of antiretroviral treatment resulted in rapid decreases in the level of CD8 T cell activation (79, 80) supporting the idea that the threshold of viral replication determines the extent of CD8 T cell activation. In such chronic infection, viral immune evasion could be due to persistence of activated CD8+ T cells as proposed by Zajac et al. (81) following an examination of the regulation of virus-specific CD8+ T cells during chronic lymphocytic choriomeningitis infection of mice. This unresponsiveness was higher under conditions of CD4 T cell deficiency. Thus, the lack of CD4 T cell help may render these CD8+ T cells non-fully differentiated during primary SIV infection and warrants further exploration (82, 83, 84).
Because the duration of antigenic stimulation of CD8+ T cells can determine whether subsequent in vivo clonal expansion will be abortive or extensive (85) and that sustained stimulus may induce an extensive proliferation and production of cells potentially capable of inducing destruction of peripheral target tissues, it cannot be excluded that, although provocative, the CD8 T cell response may have a detrimental effect during SIV infection rather than a protective effect. Other reports have shown an active participation of activated CCR5+CD8+ T cells in the pathogenesis of graft-vs-host-disease (66).
Despite the intense T cell activation and changes in chemokine receptor expression that occur during primary infection, related to the extent of viral replication and further disease progression to AIDS, our results highlight the inability to control viremia that may be, at least in part, due to the lack of maturation of CD8bright T cells during the acute phase. Better understanding of the mechanisms involved should improve our knowledge of the early steps in the inability of the immune response to control HIV and SIV replication and prevent further progression to AIDS.
Acknowledgments
We acknowledge M. Muller-Trutwin and A. Brussel for helpful discussions and critical reading of our manuscript.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was funded by a grant from the Agence Nationale de Recherche sur le Sida (ANRS; to J.E.). F.P. was supported by a postdoctoral fellowship from Ensemble Contre le Sida and L.V. was supported by a doctoral fellowship from ANRS.
2 L.V. and F.P. were equal contributors.
3 Address correspondence and reprint requests to Dr. Jér?me Estaquier, Unité de Physiopathologie des Infections Lentivirales, Institut Pasteur, 28 rue du Docteur Roux, 75724 Paris cedex 15, France. E-mail address: jestaqui{at}pasteur.fr
4 Abbreviations used in this paper: LN, lymph node.
Received for publication May 14, 2004. Accepted for publication March 11, 2005.
References
Cyster, J. G.. 1999. Chemokines and cell migration in secondary lymphoid organs. Science 286: 2098-2103.[
Butcher, E. C., M. Williams, K. Youngman, L. Rott, M. Briskin. 1999. Lymphocyte trafficking and regional immunity. Adv. Immunol. 72: 209-253.
Kim, C. H., H. E. Broxmeyer. 1999. Chemokines: signal lamps for trafficking of T and B cells for development and effector function. J. Leukocyte Biol. 65: 6-15.
Sallusto, F., C. R. Mackay, A. Lanzavecchia. 2000. The role of chemokine receptors in primary, effector, and memory immune responses. Annu. Rev. Immunol. 18: 593-620.
Mackay, C. R.. 1996. Chemokine receptors and T cell chemotaxis. J. Exp. Med. 184: 799-802.
Sallusto, F., A. Lanzavecchia, C. R. Mackay. 1998. Chemokines and chemokine receptors in T-cell priming and Th1/Th2-mediated responses. Immunol. Today 19: 568-574.
Premack, B. A., T. J. Schall. 1996. Chemokine receptors: gateways to inflammation and infection. Nat. Med. 2: 1174-1178.
Wells, T. N., C. A. Power, A. E. Proudfoot. 1998. Definition, function and pathophysiological significance of chemokine receptors. Trends Pharmacol. Sci. 19: 376-380.
Baggiolini, M.. 1998. Chemokines and leukocyte traffic. Nature 392: 565-568.(Valérie Monceaux*, Lauren)