Regulatory t cells as potential therapeutic targets to improve the response to the hepatitis b virus vaccine in hiv-infected subjects

Resumen

The main function of our immune system is to protect us from invading pathogens and microorganisms by destroying infected cells, while minimizing collateral damage to tissues. In order to maintain this balance between immunity and tolerance, current understanding of the immune system attributes a major role to regulatory T cells (Tregs). These cells, which suppress the intensity and duration of immune responses, play a key role in infectious diseases, including HIV-associated immunopathology, and are involved in the suppression of T cell activation, proliferation, inflammation and cytokine production. Tregs are a small subpopulation of CD4 T lymphocytes that constitutively express the extracellular Interleuquine-2 Receptor alpha-chain (CD25) and the intracellular Forkhead Box P3 Transcription Factor (FoxP3). A good correlation between a novel phenotype, CD4+CD25+/hiCD127lo, and CD4+CD25hiFoxP3+ was reported in healthy subjects. The staining of CD127, unlike FoxP3, does not require fixation and permeabilization of cells, and gave a great advantage to the researchers, because it was cheaper, easier to perform, staining was less-time consuming and allowed Tregs isolation. Nevertheless, most of Tregs studies in HIV-infected populations started to use this novel phenotype not only to isolate, but also to directly characterize Tregs. However, if this novel phenotype accurately identifies Tregs in HIV-infected subjects remained unexplored. Therefore, we compared the most commonly used Tregs phenotypes, CD4+CD25hiFoxP3+ and CD4+CD25+/hiCD127lo, to characterize and quantify Tregs in two different groups of HIV-infected subjects, one viremic and the other one aviremic. The most important finding was, like as previously reported in healthy people, a strong correlation between both Tregs phenotypes in the aviremic group, but surprisingly, this correlation was completely absent in the viremic group. Additional data on T cell activation levels of both HIV-infected groups suggested that the CD4+CD25+/hiCD127lo phenotype could be just mirroring the elevated numbers of activated non-regulatory T cells in the viremic HIV group. Thus, the identification and quantification of Tregs on the basis of the CD4+CD25+/hiCD127lo phenotypes could be severely compromised in HIV-infected subjects with high viral load levels and/or elevated T cell immune activation. These results are included in the article entitled "CD4+CD25+/hiCD127lo phenotype does not accurately identify regulatory T cells in all populations of HIV-infected persons" (Pozo-Balado et al. J Infect Dis 2010; 201:331¿335). Most of the studies in HIV-infected people have reported a relative increase of Tregs among CD4 T cells, which could have important clinical implications. Thus, an increased suppressive function could reduce the immune response not only to HIV, but also to other pathogens and/or new and recall antigens, which could explain at least in part, the impaired response of HIV-infected subjects to several vaccines, including Hepatitis B virus (HBV) vaccine, which has important clinical implications. In fact, HIV and HBV coinfection is a main cause of liver-related mortality among HIV-infected subjects. Vaccination against HBV is the most important way to prevent HBV coinfection; thus, an effective immunization against HBV is advisable in all susceptible HIV-infected subjects. Unfortunately, HIV-infected subjects show a low rate of response to standard HBV vaccination (23%¿56%), in contrast to rates higher than 90% found in the general population. The molecular and cellular bases underlaying this phenomenon were poorly understood, especially in HIV-infected people, were low levels of CD4 T cells and high viral load (VL) were the best predictors of unresponsiveness to the vaccine. Nevertheless, the HIV-associated Tregs expansion together with the potent suppressive function that these cells display on several components of the immunity suggest that they could be implicated, at least in part, in the low rate of seroconversion to the HBV vaccine described in HIV-infected subjects. In fact, we found an increased baseline Tregs frequency in HIV-infected subjects non responders to the HBV vaccine. Moreover, Tregs frequency was, between several clinical and immunological parameters, the only factor independently associated with the final non response to the vaccine. Furthermore, we observed a negative correlation between baseline Tregs frequency and levels of anti-HBs antibody titers in the final response. This finding could be useful for design innovative therapies complementary to the vaccination protocols to improve the seroconversion rate in HIV-infected subjects. These results have been included in the article "Increased Regulatory T Cell Counts in HIV-Infected Nonresponders to Hepatitis B Virus Vaccine", (Pozo-Balado et al. J Infect Dis 2010; 201:331¿3352010). On the other hand, it is also known that the anti-HBs antibody titers decrease over time and can fall below protective serologic levels, increasing the risk of HBV infection. Although in healthy people, the duration of protection is enough to prevent infection for at least 20 years, only 58-63% of HIV-infected subjects maintain protective anti-HBs titers 12 months after vaccination. To date, the extent of the maximum antibody response after vaccination is the only factor associated to the persistence of protective antibody titers. Due to the important role that Tregs display in the response of HBV vaccine and their relationship with the anti-HBs antibody production, we addressed the role of Tregs on the loss of protective antibody titers after one year the initialization of the protocol of immunization. We found that the frequency of baseline Tregs was independently associated to such loss of protective anti-HBs antibody titers. This finding underlines the significance of baseline Tregs, not only in the response, but also in the subsequent maintenance of specific-memory responses against HBV vaccine along the time, and emphasizes the potential benefit of immunomodulatory strategies directed to reduce Tregs before or during the HBV immunization protocol. These results have been included in the article entitled "Regulatory T-cells are associated with the loss of anti-HBs antibody titers one year after vaccination against Hepatitis B virus in HIV-infected individuals responding to the vaccine", (Pozo-Balado et al., under review). Recent new lines of investigation suggest that maraviroc (MVC), a new generation antiretroviral drug, has a favorable immunological impact on T cell responses, increasing certain anti-HIV immune function, such as the humoral and cellular responses both to recall and new immunization antigens. MVC is the first antiretroviral drug to target a human protein, the CCR5 coreceptor; however, the mechanisms of MVC-associated immunomodulation in HIV-infected subjects remained to be elucidated. Since Tregs express the CCR5 coreceptor, they are susceptible to MVC-mediated CCR5 blockade. This observation prompt us to hypothesize that MVC could improve the responses to vaccines through a reduction in the Tregs frequency. Thus, we analyzed the effect of MVC on Tregs. To asses this issue, we compared the effect of an MVC-containing or sparing combined-ART (cART) on ART-naïve HIV subjects. In vivo, MVC significantly reduced Tregs both at short-term and one year after treatment initiation. Furthermore, we observed an in vitro dose-dependent reduction of Tregs after treatment of PBMCs with MVC. However, their suppressive function was not altered. These findings partially explain MVC-associated immunomodulatory effects and could be relevant for the development of new Tregs-depleting immunotherapies useful, for instance, in chronic infections, cancer, and vaccination. These results have been included in the article "Maraviroc reduces the regulatory T-cell frequency in antiretroviral-naïve HIV-infected subjects", (Pozo-Balado et al., J Infect Dis. doi: 10.1093/infdis/jiu180). Finally, Tregs, as other T-cell compartments, show a high degree of heterogeneity and can be further subdivided into different subpopulations. Thus, the expression of CD45RA on FoxP3 expressing cells may differentiate Tregs into different phenotypic and functional subpopulations: naïve Tregs (nTregs) (FoxP3lowCD45RA+) and effector Tregs (eTregs) (FoxP3highCD45RA-) both with suppressor activity and non-Tregs (FoxP3lowCD45RA-) lacking suppressive function. FoxP3 expressing cells show a differential expression of CCR5 (eTregs>> non-Tregs> nTregs). The different pattern of CCR5-expression on Foxp3 subsets could make them differentially susceptible to MVC. Thus, we analyzed the effect of MVC on nTregs, eTregs and non-Tregs subsets, and on several homeostatic-related parameters. To address this issue, we extended the previous design to further analyze FoxP3 subsets. In vivo, only MVC significantly reduced all FoxP3 subsets at the short-term and even normalized eTregs and non-Tregs at long-term. In addition, the correlation between eTregs and CD4 T cells, lost at the acute phase of HIV-infection, was only restored in the MVC group. In vitro experiments showed a dose-dependent reduction only for eTregs and non-Tregs after treatment of PBMCs with MVC. The particular effect of MVC on different Tregs subsets, including the normalization of homeostatic parameters related to effector Tregs, could have consequences in the clinical progression of HIV subjects that merits further investigation. These results have been included in the article "Maraviroc restores the homeostasis of Regulatory T-cell subsets in antiretroviral-naïve HIV-infected subjects", (Pozo-Balado et al., submitted article). CONCLUSIONS The conclusions obtained from the studies presented in this doctoral thesis are presented below: 1. The identification of Tregs based on the CD4+CD25+/hiCD127lo phenotype, not including the staining of FoxP3, could be severely compromised in HIV-infected patients with elevated immune activation and/or high viral load levels. Consequently, caution would be needed when interpreting data from Tregs frequency by using this phenotype. 2. Baseline Tregs frequency is independently associated with the non response to the HBV vaccine, suggesting an active role played by Tregs on the immunization antigen-specific T and/or B cell responses, with the final consequence of a lower anti-HBs production. This finding also suggests that a reduction of Tregs before vaccination should be desirable to improve the seroconversion rate in HIV-infected subjects. 3. The loss of protective anti-HBs antibodies in those HIV subjects responding to the vaccine may be due to the ability of Tregs to actively suppress T and/or B immune responses driving the maintenance of function and/or survival of memory cells over time. This finding emphasizes the potential benefit of immunomodulatory strategies directed to reduce Tregs before or during the HBV immunization protocol. 4. The finding that Maraviroc reduces the frequency of Tregs in ART-naïve HIV-infected subjects could explain, at least in part, the immunomodulatory properties associated to this drug. This finding could be relevant for the development of new Tregs-depleting immunotherapies useful, for instance, in chronic infections, cancer, and vaccination. 5. The particular effect of Maraviroc on different Tregs subsets, including the normalization of homeostatic parameters related to effector Tregs, could have consequences in the clinical progression of HIV subjects that merits further investigation.