| |
IL-2 therapy and thymic production of naive CD4 T cells in
HIV-infected patients with severe CD4 lymphopenia
|
| |
| |
In this current issue of the journal AIDS (2003; 17(6):841-850), Frebnch researchers Guislaine Carcelain et al including Christine Katlama and Brittete Autran reported this paper on IL-2.
Summary: IL-2 therapy increases memory and naive CD4 T cells in HIV-infected patients, but its effect on thymopoiesis is unknown. To investigate this effect, we quantified T-cell receptor rearrangement excision circles (TREC) in CD4 T cells from lymphopenic AIDS patients treated with highly active antiretroviral therapy and IL-2. CD4 cell subsets were evaluated by flow cytometry using anti-CD45RO/RA, CD62L, Ki67 and CD95 monoclonal antibodies. The proportion of recent thymic emigrant had been quantified by a real-time polymerase chain reaction assay for signal joint TREC in peripheral blood mononuclear and purified CD4 T cells. At initiation of IL-2, TREC copies/[mu]l of blood were correlated with naive T cell numbers and age. Both naive and TREC numbers/[mu]l significantly increased over time in all patients, with a wide range of TREC increases. Higher percentages of CD4+CD45RO-negative cells positive for the Ki67 cell-cycle marker were found in patients with a low TREC increase, but remained stable under IL-2. TREC and naive cell recovery were correlated; they also correlated with the numbers of TREC and naive cells at the start of IL-2, and with age, suggesting a thymic origin for naive T-cell recovery. A mathematical model showing the linear recovery of naive cells and TREC under IL-2 also strongly suggested that a naive T-cell increase reflects thymic export and involves little net death and proliferation. Conclusion: Although we cannot rule out a mechanism of altered proliferation or death rate, the thymus plays an important role in the long-term recovery of naive T cells under IL-2 therapy.
IL-2 regulates the turnover of mature T lymphocytes and is currently being tested as an immunotherapeutic agent to improve CD4 cell counts in HIV-infected patients. Indeed, IL-2 in conjunction with antiretroviral therapy, allows for a sustained increase in CD4 cell counts in patients with moderate CD4 cell depletion. Of interest is the fact that this CD4 cell increase involves both memory CD4+CD45RO+ and naive CD4+CD45RA+ T cells. It is known,
however, that IL-2 almost exclusively enhances the peripheral turnover of mature memory T cells but has no known effect on naive T-cell production. These observations raise the question of the role of IL-2 in naive T-cell turnover. The CD45RA isoform is not an accurate marker of the naive cell status. Memory T cells can revert their CD45 isoform from RO to RA. Whether IL-2 therapy acts on this small subset of revertant memory CD4 T cells or really enhances the production of naive T cells is not known. Although extensively debated, IL-2 is not believed to amplify thymus output, which is mostly regulated by IL-7. Finally, none or very little proliferation of naive T cells is observed in the periphery, where naive T cells proliferate and convert to effector or memory T cells when encountering their specific antigen. In severe lymphopenia, however, some naive T cells might undergo some homeostatic proliferation, during which IL-7 is also believed to play a major role.
Rearrangement of the T-cell-receptor (TCR) delta genes leads to the excision of circular DNA fragments from genomic DNA during the early stages of double positive CD4+CD8+ thymocyte development in the thymus. T-cell receptor rearrangement excision circles (TREC) within peripheral blood T cells have been used as markers of naive T cell production. The proportion of naive T cells carrying TREC declines with age and during HIV infection, and
increases under potent antiretroviral therapy.
We have recently shown that IL-2 significantly increases the CD4 and naive CD4+CD45RA+CD62L+ cell numbers in patients with severe CD4 cell depletion failing to reconstitute their CD4 cells after antiretroviral therapy alone.
To evaluate whether this increase reflects a positive effect of IL-2 on thymus production, we quantified the TREC in purified CD4 T cells. To avoid problems of interpretation as a result of the different capacity of naive and memory T cells to proliferate and to allow an unbiased estimate of thymic output, results were expressed as TREC copies/[mu]l and not as TREC content per cell. To assist in the interpretation of our data, a mathematical model was formulated. This model, describing the dynamics of the naive CD4 T-cell population and the TREC, allowed us to estimate thymic production and the parameters relating to the proliferation and death rates of naive T cells under IL-2 therapy.
Patients were selected from a double-arm randomized clinical trial studying the efficacy of IL-2 infusions. Patients were selected as follows: CD4 cell counts remained between 25 and 200 cells/mm3, despite an antiretroviral
regimen containing two nucleoside analogues and a protease inhibitor for at least 6 months, which was efficient at reducing the plasma HIV-RNA load below 200 copies/ml. Patients (n = 72) were randomly assigned to receive adjuvant sc-IL2 4.5 MIU twice a day for 5 days every 6 weeks from day 0 of the study or after a delayed period of 24 weeks, in addition to their previous highly active antiretroviral therapy (HAART). After week 24, IL-2 was given as 9 MIU for eight cycles of 5 days every 6 weeks or as 9 MIU for seven cycles of 5 days every 8 weeks. Blood samples were collected at day 0, weeks 6 or 12, weeks 24 or 28, weeks 50 or 52 and week 80. Patients (n = 13) with the highest CD4 cell counts at baseline were selected for TREC quantification on frozen cells. Eight age-matched healthy individuals were used as controls for TREC analysis.
Results
T-cell receptor rearrangement excision circles at baseline of IL-2 therapy in patients failing CD4 cell reconstitution under highly active antiretroviral therapy alone
We first set up our assay by measuring the TREC content in naive and memory CD4 T cell sub-populations purified from PBMC of healthy controls (26-50 years). To quantify simultaneously and precisely the number of cells and TREC copies an internal control (GAPDH) was amplified in every sample. The TREC content was 10-fold higher in naive CD4+CD45RA+ CD62L+ than in memory CD4+CD45RO+ T cells, as shown in Fig. 1a, yielding similar results to
those previously reported. A non-significant correlation between the age and the TREC content of PBMC was observed in healthy donors (r = -0.53; P = 0.18).
TREC counts in CD4 T cells from the 13 HIV-positive patients (median age 48 years, range 32-66) who failed to reconstitute CD4 cell counts with HAART despite undetectable plasma viremia were determined at the baseline of IL-2
therapy. Patients had a median CD4 cell count of 131 cells/mm3 (range 79-191) and a severe depletion of naive CD4+ CD45RA+CD62L+ T cells (median 30 cells/mm3, range 6-97). The median plasma IL-7 level was 2.6 pg/ml (range
0.3-7.5), which is less than the levels observed in healthy controls (17 pg/ml, range 9.4-24.1). The median number of TREC per 106 PBMC in the 13 HIV-positive patients was 3799 copies (range 567-30 429). In patients, the
TREC numbers per PBMC were positively correlated (r = 0.72; P < 0.001) with the percentage of CD4+CD45 RA+CD62L+ cells (Fig. 1c), and negatively correlated (r = -0.55; P = 0.049) with age.
Effects of IL-2 therapy on naive T cells and T-cell receptor rearrangement excision circles
Nine million IU IL-2 were injected subcutaneously for 5 days every 6-8 weeks. As described in the Methods section, six patients received IL-2 at the initiation of the trial, and seven after a period of 24 weeks. No increase in naive CD4 T cells was observed in these seven patients during the first 24 weeks under HAART alone, with a mean change in CD4+CD45RA+ CD62L+ cells of 8 ± 13 cells/mm3 over that period (P = 0.13). Overall, the 13 patients received a median of eight cycles of IL-2 (range eight to 11) over a period of 52 weeks. IL-2 therapy induced a significant median increase of 84 cells/mm3 (range 14-179) naive CD45RA+CD62L+CD4+ T cells at week 52 (P <
0.001) (Fig. 2 and Fig. 3). The memory CD4+ CD45RO+ T cell counts showed a similar increase of 76 cells/mm3 (range 26-205) (P < 0.001). The increase in naive CD4 T cells was accompanied by a significant decrease in plasma levels of IL-7 from a median concentration of 2.6 pg/ml (range 0.2-7.5) at baseline to 1.4 pg/ml (range 0.1-7.8) after 52 weeks of IL-2 therapy (P = 0.03).
Discussion
To investigate whether IL-2 acts on thymopoiesis, we quantified TREC and naive CD4 T cells during the course of intermittent IL-2 therapy in HIV-1 patients with severe CD4 cell depletion. TREC measurements allowed us to evaluate whether IL-2 acts before or after the TCR rearrangement events. A mathematical model helped to discriminate between the role of thymic production and peripheral proliferation in naive T-cell reconstitution.
Our results and the mathematical model suggest that in these HIV-1-infected patients with severe CD4 T-cell depletion, the thymus plays an important role in naive T-cell recovery. Indeed, the fit of the model for the naive T-cell dynamics suggests that in most cases the naive cells are contributed by the thymus. However, as the patients were in steady state under HAART alone according to the entry criteria, thymus output cannot be estimated before the initiation of IL-2 therapy. Consequently, an enhanced survival of naive cells in the periphery cannot be ruled out with these models. Peripheral proliferation could still be substantial, but its effect, together with the conversion of naive to memory T cells, would be nullified by the death rate of cells.
To test the hypothesis of thymic involvement in naive T-cell recovery further, we analysed biological parameters of proliferation and death on CD4 naive cells. The more than twofold increase in TREC seen in half the patients could result either from an enhanced thymic production, or from an overall decrease of naive cell death in the periphery as a result of IL-2 treatment. According to the thymic hypothesis, IL-2 would act on thymopoiesis by inducing the proliferation of thymocytes before TCR rearrangement. Such an effect of IL-2 on the intrathymic development of mature T cells has been strongly debated. In vitro, IL-2 has been shown to promote proliferation and differentiation of human pro-T cells into CD3+CD4+CD8+ mature thymocytes. Although IL-7 is the main known immune modulator of thymic production, a smaller but direct effect of IL-2 on thymus production has recently been proposed in mice. We observed at baseline a low seric level of IL-7 in patients compared with healthy individuals. These low values were even lower than those reported by Napolitano et al., and may be linked to the incapacity of our patients to reconstitute
their CD4 cell numbers under HAART. These values of IL-7 decrease even further in our patients under IL-2 therapy. Altogether, the low seric level of IL-7 and its decrease in our patients under IL-2 does not support the hypothesis of an indirect effect of IL-2 through increased production of IL-7. However, we cannot rule out a pharmacodynamic effect of increased IL-7 binding to IL-7R, which is highly expressed on naive T cells.
Alternatively, IL-2 could increase absolute TREC numbers by decreasing naive T-cell death in the periphery. However, assuming that Fas is a marker for susceptibility to cell death, this hypothesis is not supported by the changes observed in Fas expression, which rather increases on naive CD45RO-negative T cells. Therefore, pre-TCR rearrangement expansion of thymocytes seems the likely mechanism through which both naive cells and TREC increase under IL-2.
In half the patients, however, TREC do not increase in contrast to naive T cells, raising the hypothesis of naive T-cell proliferation. It is generally considered in normal individuals that naive cell proliferation is antigen dependent and results in rapid conversion to memory T cells. However, some homeostatic proliferation might be possible in profoundly lymphopenic mice. A higher percentage of Ki67+CD45RO-negative cells in group B suggests that at least in some of these profoundly lymphopenic individuals, homeostatic proliferation of naive CD45RO-negative T cells might indeed occur. However, this phenomenon is observed at entry and was not enhanced by IL-2. In addition, the linear increase of naive T cells suggests that if naive cell proliferation exists, it would take place in the thymus. Indeed, if this proliferation had occurred in the periphery, the naive increase would have been exponential. The hypothesis of proliferation after TCR rearrangement is supported by recent findings in vivo in a severe combined immunodeficiency human model. In that model, IL-2 may prevent HIV-induced depletion of thymocytes by the expansion of a relatively mature subset of CD4+CD8+ thymocytes. Finally, the priming rate of naive cells remains an unknown factor. Indeed, one could argue that because IL-2 stimulates memory T-cell proliferation, there might be less need for the conversion of naive cells
into memory cells. However, we could not investigate the effect of IL-2 on the conversion rate of naive to memory cells. Furthermore, the absence of correlation of naive cell recovery with memory cell recovery under IL-2 does not support the hypothesis of a peripheral homeostatic mechanism.
We have assumed that the effect of IL-2 was continuous, i.e. not limited to the 5-day period of the IL-2 therapy itself. The reality probably lies somewhere in between: the effect of therapy extends well beyond the 5-day cycle. Indeed, in many patients the total CD4 cell count decreases during the course of an IL-2 cycle, but then increases to levels higher than before the cycle, suggesting that there is a long-term effect of IL-2 (data not shown). Sereti et al. Also observed an increase of activation and apoptosis markers during the 5-day IL-2 cycle, and a decrease to previous levels after IL-2 therapy. In conclusion, altered proliferation and increased death rates of CD4 cells during an IL-2 cycle may well play a role in naive cell reconstitution under IL-2, but they cannot explain the observed increase in TREC seen in at least half the patients.
The recovery rate of naive cells under IL-2, on average, is similar to that observed in other studies under HAART (0.24 ± 0.04/mm3 a day, mean ± SEM, versus 0.34 ± 0.04/mm3 a day, respectively). Cohen Stuart et al. (in preparation) noted that recovery rates in these patients were similar to those in patients after anti-CD4 monoclonal antibody treatment.
Similarly, we found that the recovery rates were not significantly different between the larger cohort on IL-2 therapy and those in patients after bone marrow transplantation (0.21 ± 0.25 versus 0.16 ± 0.24).
The variability in TREC and naive cell recovery rates during IL-2 therapy in this study can be explained by parameters of T-cell homeostasis and thymic production before the initiation of IL-2 therapy, such as the differences in numbers of TREC, naive cells and age at initiation of IL-2. IL-2 therapy thus appears to enhance the pre-existing mechanisms of T-cell homeostasis that were ineffective under HAART alone in these profoundly lymphopenic patients. Dependency on these three parameters adds support to the hypothesis of thymus involvement in the process of regeneration. We cannot
rule out the hypothesis that several non-exclusive mechanisms might be involved in the naive T-cell reconstitution under discontinuous IL-2 therapy, such as altered proliferation or death rate, and future larger studies will be needed to analyse these points. Our results strongly suggest, however, that the thymus plays an important role in the long-term-recovery of naive T cells under IL-2 therapy.
|
|
| |
| |
|
|
|
