IL-7 is a nonredundant cytokine for T cell homeostasis. Circulating IL-7 levels increase in lymphopenic clinical settings, including HIV-1 infection. HIV-2 infection is considered a “natural” model of attenuated HIV disease given its much slower rate of CD4 decline than HIV-1 and limited impact on the survival of the majority of infected adults. We compared untreated HIV-1- and HIV-2-infected patients and found that the HIV-2 cohort demonstrated a delayed increase in IL-7 levels during the progressive depletion of circulating CD4 T cells as well as a dissociation between the acquisition of markers of T cell effector differentiation and the loss of IL-7Rα expression. This comparison of two persistent infections associated with progressive CD4 depletion and immune activation demonstrates that a better prognosis is not necessarily associated with higher levels of IL-7. Moreover, the delayed increase in IL-7 coupled with sustained expression of IL-7Rα suggests a maximization of available resources in HIV-2. The observation that increased IL-7 levels early in HIV-1 infection were unable to reduce the rate of CD4 loss and the impaired expression of the IL-7Rα irrespective of the state of cell differentiation raises concerns regarding the use of IL-7 therapy in HIV-1 infection.
Interleukin-7 is considered a key cytokine in T cell homeostasis that acts both during thymopoiesis and in the periphery, promoting naive T cell proliferation and survival as well as the maintenance of memory cells (1, 2, 3, 4). Thus, much expectation has been placed on the therapeutic use of IL-7 to improve thymic output and broaden the T cell repertoire in lymphopenic clinical settings as well as on its use as adjuvant to improve the breadth of vaccine responses (5, 6, 7, 8, 9, 10). IL-7 is constitutively produced by stromal cells of the bone marrow, thymus, mucosal lymphoid tissues, and lymph nodes (11, 12, 13, 14). Circulating IL-7 levels increase in lymphopenic states, suggesting either increased production facilitated by a compensatory feedback loop or increased availability due to the reduction in cell targets (1, 14, 15, 16). Increasing data point to a critical role of the expression of the α-chain of the IL-7 receptor (IL-7Rα) in the regulation of IL-7 biology (17, 18). IL-7 signaling results in a transient down-regulation of the IL-7Rα that is thought to allow adequate sharing of available IL-7 by a large number of T cells (17). Moreover, IL-7Rα can also be down-regulated by other cytokines that share the common cytokine receptor chain γc, such as IL-2, and by TCR stimulation (19, 20). Of note, differentiated effector T cells have been shown to be IL-7Rαlow, a phenotype associated with decreased survival and proliferation in response to IL-7 (21, 22).
HIV-2, the second AIDS virus, is considered a “natural” model of attenuated HIV disease because it is associated with a much slower course of disease progression than HIV-1 with limited impact on the survival of the majority of infected adults (30, 31). More than 90% of HIV-2 infected individuals are thought to meet the standard criteria for “long-term nonprogressors” (30, 31, 32) but display a steady decline of CD4 counts (30, 32, 33, 34). Although the rate of CD4 depletion is markedly slower in HIV-2 than in HIV-1 infection, a direct correlation with the levels of immune activation was observed in both cases (34, 35). This is despite the low levels of viremia that characterize all stages of HIV-2 disease (33, 34, 36, 37, 38). The reduced viremia is considered the main reason why HIV-2 infection remains confined to West Africa (39, 40, 41). In Portugal, because of its connections with its past colonies there is a significant prevalence of HIV-2, currently accounting for 5% of the HIV infections (42).
We took advantage of this situation and compared untreated HIV-1- and HIV-2-infected patients living in Portugal to obtain insights into the role of IL-7 in the rate of CD4 decline and the imbalances of T cell subsets associated with HIV/AIDS pathogenesis as well as the regulation of IL-7 in other lymphopenic states.
We found that a delayed increase in IL-7 levels during the progressive depletion of circulating CD4 T cells as well as a dissociation between the acquisition of markers of T cell effector differentiation and the loss of IL-7Rα expression are distinctly observed in HIV-2 disease.
Materials and Methods
A cross-sectional study involving 50 HIV-2- and 45 HIV-1-infected patients currently living in Portugal and attending outpatient clinics in Lisbon, with no evidence of ongoing opportunistic infections or tumors, was performed. The clinical and epidemiological data of the two cohorts, as well as of the healthy controls, are summarized in Table I⇓. Additionally, a longitudinal study was performed in nine HIV-2 infected patients. All subjects gave informed consent to blood sampling and processing and the study was approved by the Ethical Board of the Faculty of Medicine of Lisbon, Lisbon, Portugal.
Phenotypic and intracellular characterization of lymphocyte subsets
PBMC were isolated from fresh heparinized blood using Ficoll-Hypaque density separation gradient. PBMC were surface stained as previously described (34
Plasma viral load assessment
HIV-1 viremia was quantified by RT-PCR (detection threshold of 50 RNA copies/ml; Roche Ultrasensitive test). HIV-2 viral load was quantified using a RT-PCR-based assay developed and performed by Gomes, Lourenço, and coworkers (42) that has a detection limit of 200 RNA copies/ml. The cutoff value of the tests was considered for the purpose of the analysis in the cases where detection was below this level.
Quantification of cellular proviral DNA load
Genomic DNA was extracted from 1062 gag gag, and HIV-2 gag plasmids ranging from 106 to 5 copies. Samples were run in duplicate and the input level of DNA was normalized to the albumin copy number. Data were expressed as the number of HIV DNA copies per 106 PBMC.
Statistical analysis was performed using GraphPad Prism version 4.00 (GraphPad Software). The data are presented as arithmetic mean ± SEM and were compared using ANOVA and an unpaired t test. Pearson’s correlation coefficient was used to assess the correlation between two variables. Value of p < 0.05 was considered to be significant.
Strong correlation between the increase in circulating IL-7 and CD4 depletion in HIV-2 infection
Serum IL-7 levels were significantly increased both in HIV-1 and HIV-2 infections as compared with healthy controls and there were no significant differences between the two infections (Fig. 1⇓A). However, we found a strong correlation between the increased levels of circulating IL-7 and the degree of CD4 depletion (p < 0.0001; r = −0.5929) in HIV-2+ patients that contrasts with the absence of a significant correlation in our HIV-1 cohort (Fig. 1⇓B). This cohort, like the HIV-2 cohort, did not include a significant proportion of patients with extreme CD4 lymphopenia, which might explain the discrepancy between our data and the previous reports of a significant negative correlation between CD4 counts and IL-7 levels in HIV-1 infection. In these cases, the significance was usually reached due to the very high IL-7 levels exhibited by patients with <100 CD4 cells/μl (1, 14, 23, 24, 25).
To further clarify the relationship between IL-7 levels and disease progression, HIV-1- and HIV-2-infected patients were divided into three groups according to CD4 T cell counts: >500 (early), 200–500 (intermediate), and <200 CD4 T cells/μl (advanced). Although both HIV-2 and HIV-1 patients at intermediate and advanced stages of the disease exhibited significantly higher IL-7 levels than healthy controls, increased IL-7 levels in the early disease stage was only observed in HIV-1-infected patients (Fig. 1⇑C). Thus, the absence of elevated IL-7 in early disease may contribute to the strong correlation documented between IL-7 and CD4 lymphopenia in HIV-2 infection.
The strong correlation between IL-7 and CD4 lymphopenia observed in HIV-2 infection is further supported by longitudinal studies of HIV-2 infected patients
A longitudinal analysis of circulating IL-7 levels and CD4 T cell counts was performed in nine HIV-2 infected patients. As shown in Fig. 2⇓, low CD4 T cell counts were associated with increased IL-7 levels in all patients. The period of follow-up ranged from 2 to 9 years. These untreated HIV-2 infected patients exhibited a reduced rate of CD4 decline as expected (32, 33, 36, 42). The graphs illustrate the consistency of the IL-7 measurements during the follow-up and document changes in serum IL-7 levels inversely related to the alterations found in CD4 T cell counts.
No correlation of serum IL-7 levels with age
It has been suggested that serum IL-7 levels decreases with age (43, 44). Because HIV-2-infected patients tend to be older than HIV-1-infected subjects, we looked for a possible impact of age on the circulating IL-7 levels and found no correlation between age and serum IL-7 in our three cohorts as illustrated in Fig. 3⇓A. Moreover, there were no statistically significant differences in IL-7 levels according to gender in the infected cohorts (p = 0.1194, derived from the ANOVA).
No significant correlation of serum IL-7 levels with viremia and proviral DNA
IL-7 is a powerful inducer of HIV-1 replication in vitro (45). However, there are conflicting data regarding the correlation between circulating IL-7 and HIV-1 viremia (1, 14). We found no correlation between serum IL-7 and HIV-1 viremia (r = 0.0324; p = 0.8094). Despite the lack of data on the ability of IL-7 to induce HIV-2 replication, it is unlikely that it will have a different effect to that reported for HIV-1 and SIV strains (45, 46, 47). Because the large majority of the HIV-2 infected patients had undetectable levels of viremia (<200 RNA copies/ml), it was impossible to correlate this parameter with IL-7 levels.
Despite the distinct viremia, HIV-1- and HIV-2-infected patients have been shown to have comparable levels of cell-associated viral load as assessed by proviral DNA (36, 48, 49). Accordingly, we found no significant difference between the levels of PBMC proviral DNA in the HIV-1 and HIV-2 cohorts. No correlation was found between circulating IL-7 and HIV-1 proviral DNA or with HIV-2 proviral DNA, as shown in Fig. 3⇑B.
Analysis of CD4 T cell subsets/IL-7Rα expression in relation to circulating IL-7
We then investigated whether the apparent close association of increased IL-7 levels with CD4 depletion in HIV-2 infection had an impact on naive/memory imbalances and on IL-7Rα expression as assessed by flow cytometry on freshly isolated PBMC.
As illustrated in Fig. 4⇓A, a significant negative correlation was found between circulating IL-7 levels and naive CD4 T cell counts in the HIV-2 cohort that was not observed in the HIV-1 cohort.
The CD31 molecule has been proposed to be a marker for recent thymic emigrants because the CD31+ subset of human naive T cells is enriched in TCR excision circles (50). When we looked for a correlation between IL-7 and the proportion of CD31+ cells within the naive CD4 subset, the linear regression showed a similar marked positive slope in both HIV-2 and healthy cohorts that contrasts with the negative slope found in the HIV-1 cohort (Fig. 4⇑B). Moreover, this correlation reaches significance when the number of subjects is increased by conjoint analysis of the HIV-2 and healthy cohorts (r = 0.3876; p = 0.0312). Although there were no significant differences between the frequency of CD31+ cells within the naive CD4 subset in the three cohorts (HIV-2, 48.41 ± 6.43; HIV-1, 43.87 ± 3.57; and healthy, 42.50 ± 3.11), the expected decline of this frequency with age was not observed in HIV-2-infected patients (Fig. 4⇑C). Further studies are required to elucidate the relative contribution of the thymus and the periphery in the maintenance of this population, including TCR excision circle quantification. Despite this, our data suggest that the better preservation of this subset in HIV-2 than in HIV-1 infected patients is closely related to IL-7 levels.
Fig. 4⇑, D and E, illustrate the IL-7Rα expression levels within the naive and memory CD4 subsets, respectively. The HIV-2 patients included in this analysis were older than the HIV-1 patients (49 ± 3 and 37 ± 3 years, respectively; p = 0.0058) but had similar CD4 counts and IL-7 levels. As expected, in all three cohorts there was a significant decrease in the levels of IL-7Rα (p < 0.006), with cell differentiation defined by the loss of CD45RA. The HIV-1-infected patients exhibited a major reduction in IL-7Rα expression as compared with healthy controls in both the naive and memory subsets; this in agreement with the low levels of IL-7Rα expression in HIV-1 infected patients reported by others (24, 26, 27). However, in HIV-2-infected patients there was only a significant decrease in IL-7Rα expression in the memory subset, suggesting maintenance of IL-7Rα in the naive CD4 pool. The levels of CD4 T cell activation as assessed by HLA-DR up-regulation were similar in the two infections (data not shown), suggesting that the discrepancies of IL-7Rα expression cannot be attributed to different states of T cell activation.
The ability of IL-7 to promote T cell survival has been shown to be related to the up-regulation of the antiapoptotic molecule Bcl-2 (3). We measured intracellular Bcl-2 by flow cytometry on freshly isolated PBMC and, despite an apparent trend in HIV-2 infected patients for a higher median intensity of fluorescence within the CD4 subset, no significant differences were found in the three cohorts (Fig. 4⇑F).
Analysis of CD8 T cell subsets/IL-7Rα expression in relation to circulating IL-7
CD8+ T cell differentiation was assessed by the expression of CD27, CCR7, and CD45RA in cohorts of HIV-2- and HIV-1-infected patients with comparable degrees of CD4 counts and IL-7 levels but differing viremia. We found similar levels of naive and central memory CD8 T cell pool depletion in both infections (Fig. 5⇓, A and B, respectively). However, HIV-1-infected patients exhibited an expansion of CD8+ T cells with the intermediate differentiated phenotype CD45RA−CCR7−CD27+, which was not found in the HIV-2 cohort (Fig. 5⇓C). In fact, the major CD8 expansion observed in HIV-2 infected patients was essentially due to terminally differentiated effector cells (CCR7−CD27−CD45RA+; Fig. 5⇓D). These results are in agreement with our previous data from other HIV-2 and HIV-1 cohorts where CD8 differentiation was assessed by CD62L, CD28, and CD27 expression as well as by IL-2 and/or IFN-γ production (34, 38). No significant correlations were found between serum IL-7 levels and the frequency of the different CD8 subsets in either infection.
All three cohorts exhibited a progressive loss of expression of IL-7Rα in the CD8 subpopulations as they acquired a more differentiated phenotype. However, HIV-2- and HIV-1-infected patients exhibited significantly lower expression levels of IL-7Rα in all of the CD8 subsets as compared with healthy controls (Fig. 5⇑E). When we compared both infected cohorts, the HIV-2 infected patients preserved significantly higher levels of IL-7Rα expression than the HIV-1+ patients in all of the CD8 effector memory subpopulations (Fig. 5⇑E). A similar level of CD8 activation assessed by CD38 and/or HLA-DR was observed in the two infected cohorts (data not shown) and, therefore, it is unlikely that the discrepant levels of IL-7Rα expression in the two infections were due to differences in the cell activation state.
In summary, the expanded CD8 T cell pool, observed in both HIV cohorts, showed skewing toward a terminally differentiated effector phenotype paralleled by a better preserved expression of IL-7Rα in the HIV-2+ individuals as compared with the HIV-1+ individuals.
This is the first study evaluating both the levels of circulating IL-7 and the expression of the IL-7Rα in patients infected with the second AIDS-associated virus, HIV-2. HIV-2 disease has several unique features that make it especially attractive for addressing the role of IL-7 in lymphopenic clinical settings. First, it is characterized by a progressive decrease in CD4 counts associated with pan-immune activation, though at a much slower rate than that observed in HIV-1 infection (30, 32, 33, 34). Second, in contrast with HIV-1, HIV-2 infection is associated with reduced viremia at all disease stages (33, 34, 36, 37, 38). Third, it has a relatively favorable clinical outcome with limited impact on the mortality of the majority of the infected adults (30, 31, 32), suggesting that HIV-2-infected patients somehow retain the capacity to replace and sustain numbers of immunologically competent CD4 T cells.
In this study we describe an increase in circulating IL-7 levels in strong correlation with the degree of CD4+ T cell depletion. This was documented in a cross-sectional study involving HIV-2 patients with different levels of CD4 counts without known ongoing opportunistic infections as well as in a longitudinal study. This cohort was compared with an HIV-1 infected cohort, similarly underrepresented by profoundly lymphopenic patients, which may explain the absence of this correlation in contrast to the majority of the HIV-1 studies reported in the literature (1, 14, 23, 24, 25).
Elevated IL-7 levels may result from increased production in response to CD4 lymphopenia (1, 14, 15). This would require a more marked CD4 loss in HIV-1- than in HIV-2-infected patients despite the same levels of peripheral blood CD4 counts to explain the disease-specific differences in the kinetics of the IL-7 increase in early infection. We found similar levels of circulating IL-7 in early HIV-2 disease and healthy subjects, in contrast to the significant increase found in early HIV-1 infection. However, there is cumulative evidence that during early HIV-1 disease the peripheral blood compartment overestimates the degree of CD4 depletion of the body due to the traffic alterations that promote lymphocyte retention in the lymphoid tissue as illustrated through the study of tonsils and lymph nodes (51, 52). Yet, recent data show a marked depletion of CD4+ T cell subsets in the gut that occurs during acute HIV-1 infection and persists throughout disease (53, 54, 55). There are no data on lymph node or mucosal pathology during HIV-2 infection, but it is reasonable to speculate that the establishment of HIV-2 infection may not be associated with a high viremia peak and major depletion of the memory compartment given the absence of clinical reports of HIV-2 acute infection. The possible contribution of gut-associated CD4 depletion in triggering the early increase of IL-7 production in HIV-1 infection deserves further exploration.
Alternatively, the levels of circulating IL-7 may increase as a result of its diminished adsorption by a reduced number of cells expressing the IL-7Rα (16). Our findings support this possibility given the reduced levels of IL-7Rα expression in HIV-1- as compared with HIV-2-infected patients with the same degree of CD4 depletion. Although IL-7 consumption has been shown to transiently induce IL-7Rα down-regulation, the high IL-7 levels observed in HIV-1 infection argue against this explanation. The different IL-7Rα expression in the two infections is particularly relevant in view of the similar state of immune activation as assessed by the expression of HLA-DR and CD38 within both CD4 and CD8 T cell subsets, in agreement with our previous studies in other cohorts using a larger panel of activation markers (34, 56). Moreover, several factors would favor a lower level of expression of the IL-7Rα in HIV-2 infection: 1) HIV-2-infected patients were shown to be older and aging is associated with impaired IL-7Rα expression (22); 2) the longer length of the infection in the case of HIV-2; and 3) the terminally differentiated profile exhibited by the CD8 T cells of these patients.
IL-7 administration to SIV-infected nonhuman primates has been shown to expand naive and memory T cell subsets (6, 9, 10). Although the frequency of naive and central memory T cells were not higher in HIV-2-infected individuals as compared with HIV-1 infected patients with the same degree of CD4 depletion, it is possible that a well adjusted and balanced increase in circulating IL-7 levels may contribute to the slower rate of loss of these populations in HIV-2 infection. Moreover, almost all subsets of CD4 and CD8 T cells, irrespective of their differentiation state, were shown to have higher levels of IL-7Rα expression in HIV-2- than in HIV-1-infected patients, possibly indicative of a higher proliferative capacity or survival in response to IL-7. The preserved expression of the IL-7Rα, in particular within the CD4 naive T cells, could also lead to an increased consumption of IL-7 in HIV-2 as compared with HIV-1 infection and partly explain the delayed increase in circulating IL-7 levels.
IL-7 is a strong inducer of HIV-1 replication (45, 57, 58), leading to its proposed use as a therapy to purge the viral reservoirs in HIV-1-infected patients virally suppressed by antiretroviral therapy (45). Although there are no data on the ability of IL-7 to promote HIV-2 replication, it would be expected to be comparable given the similarity of the two viral promoter regions (long terminal repeats) (59). No differences were found between the levels of HIV-2 and HIV-1 proviral DNA in PBMCs, in agreement with other studies (36, 48, 56). The low to undetectable HIV-2 viremia in this context further emphasizes the efficiency of the ill-defined mechanisms involved in the control of viral replication in HIV-2-infected patients in the absence of antiretroviral therapy.
In summary, this comparison of two persistent infections associated with progressive CD4 depletion and immune activation demonstrates that the better prognosis is not necessarily associated with higher levels of IL-7. Moreover, the delayed increase in IL-7 levels, coupled with sustained expression of IL-7Rα, suggests a maximization of available resources. The observation that increased IL-7 levels early in HIV-1 infection seem unable to reduce the rate of CD4 loss and the impaired expression of the IL-7Rα irrespective of the state of cell differentiation raises concerns regarding the use of IL-7 therapy in HIV-1 infection.
We gratefully acknowledge P. Gomes and M. H. Lourenço for the performance of HIV-2 viremia quantifications and the clinical collaboration of the following colleagues: A. Ribeiro, E. Valadas, F. Antunes, M. Doroana, M. Lucas, R. Dutchman, and R. Marçal.
The authors have no financial conflict of interest.
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 supported by a grant from Fundação para a Ciência e a Tecnologia and by Programa Operacional Ciência e Inovação 2010 Grant POCI2010 (to A.E.S.). A.S.A., C.S.C., R.B.F., and R.S.S. received scholarships from the Fundação para a Ciência e a Technologia.
↵2 A.S.A. and C.S.C. contributed equally to this paper.
↵3 Address correspondence and reprint requests to Dr. Ana Espada de Sousa, Unidade de Imunologia Clínica, Instituto de Medicina Molecular, Faculdade de Medicina de Lisboa, Avenida Professor Egas Moniz, Lisbon, Portugal. E-mail address:
- Received October 17, 2006.
- Accepted December 12, 2006.
- Copyright © 2007 by The American Association of Immunologists