Immunosuppressive regulatory T cells (Tregs) are responsible for the induction of immune tolerance to prevent autoimmunity and for the maintenance of immune homeostasis (Sakaguchi et al., 1995; Seddon, 2000; Kim et al., 2007; Sakaguchi et al., 2013). The expression of several genes governing the differentiation of Tregs, the maintenance of the phenotype of these cells, as well as their suppressive functions (Fontenot et al., 2005; Sakaguchi et al., 2013), is dictated by the protein FoxP3. This transcription factor is therefore referred to as the “master regulator” of Tregs. Mature human Tregs are described phenotypically as CD3⁺CD4⁺CD25^highCD127^lowFoxP3^high (Jenabian et al., 2012; Chevalier and Weiss, 2013; Sakaguchi et al., 2013). Tregs may have two developmental origins: (1) natural or thymic Tregs (tTregs), which originate within the thymus as a T-cell lineage and (2) induced Tregs (iTregs), which are derived from peripheral naive CD4+ T-cells that start expressing FoxP3 under inflammatory conditions and/or in vitro, upon stimulation with Transforming Growth Factor Beta (TGF-β) (Chen et al., 2003; Curotto de Lafaille et al., 2004; Curotto de Lafaille and Lafaille, 2009; Sakaguchi et al., 2013). Within the thymus, tTreg development occurs in two main steps: survival of thymocytes bearing TCR with relatively high affinity to self-antigens and cytokine-dependent induction and maintenance of FoxP3 expression (Lio and Hsieh, 2008). TGF-β is a pleiotropic cytokine, involved in many biological proceeses such as differentiation of cells, their proliferation, embryonic development, carcinogenesis, fibrosis, tissue repair and healing, and angiogenesis (Blobe et al., 2000). In mammals, there are three different isoforms of TGF-β (TGF-β1, TGF-β2, and TGF-β3), each encoded by a different gene (Govinden and Bhoola, 2003). The TGF-β1 isoform is implicated in the regulatory aspects of the immune response (Li et al., 2006). TGF-β is also known to regulate FoxP3 expression during tTreg development and inflammation in other tissues, notably in gut lymphoid tissues (Chen et al., 2003; Li and Flavell, 2008; Konkel and Chen, 2011; Zeng et al., 2011; Chen and Konkel, 2015; Jurberg et al., 2015).

It is well documented that the proportion of Tregs in blood and secondary lymphoid organs is increased during HIV/SIV infections (Andersson et al., 2005; Bi et al., 2009; Zeng et al., 2011; Jenabian et al., 2012; Moreno-Fernandez et al., 2012; Chevalier and Weiss, 2013; Kleinman et al., 2018; Yero et al., 2019). Importantly, in HIV/SIV infections, Treg accumulation in lymphoid tissues has been directly correlated with the severity of the disease and immunodeficiency (Nikolova et al., 2011; Zeng et al., 2011; Jenabian et al., 2012; Moreno-Fernandez et al., 2012; Chevalier and Weiss, 2013; Kleinman et al., 2018; Yero et al., 2023). This phenomenon is most obvious in the gut associated lymphoid tissue (GALT), where the increase in proportion of Tregs results in an imbalance between Tregs and T helper 17 (Th17) cells, which results in impaired mucosal immune responses, which consequently leads to microbial translocation and exacerbated systemic activation of the immune system (Favre et al., 2009; Kim et al., 2013; Maartens et al., 2014; Yero et al., 2023). Indeed, even under effective antiretroviral therapy (ART), immunosuppressive Tregs can inhibit both the HIV-specific immune responses and the proliferation of effector CD4 and CD8 T-cells (Aandahl et al., 2004; Nilsson et al., 2006; Favre et al., 2009; Zeng et al., 2011; Jenabian et al., 2012; Chevalier and Weiss, 2013; Jenabian et al., 2013; Kim et al., 2013; Maartens et al., 2014; Kleinman et al., 2018; Yero et al., 2019). Our team reported a decrease in the proportions of Tregs in the peripheral blood following early initiation of ART in acute HIV infection compared to healthy controls (Jenabian et al., 2015; Yero et al., 2021). Using SIV-infected rhesus macaques, we demonstrated that even early ART initiation cannot prevent the increase in Treg frequencies or migration of Tregs into mesenteric lymph nodes, thereby contributing to immune dysfuction in the gut, and probably also leading to fibrosis and reduced functioning of these lymph nodes (Yero et al., 2019). This negative role of Tregs during HIV infection is mostly via the production of TGF-β. Indeed, Tregs producing TGF-β are implicated in collagen deposition and gut lymphoid tissue fibrosis and immune dysfunction, which in turn, causes generalized inflammation and HIV/SIV disease progression (Estes et al., 2007; Zeng et al., 2011; Jenabian et al., 2012; Moreno-Fernandez et al., 2012; Chevalier and Weiss, 2013; Yero et al., 2019, 2023). Therefore, a rational approach to predictably inhibit Treg development and function could be an important therapeutic tool in the context of HIV infection. However, this requires a better understanding of the mechanisms of differentiation and activation of tTregs versus iTregs during HIV infection.

Thymic involution has been reported in chronically HIV-infected individuals and in children with AIDS (Grody et al., 1985; Joshi et al., 1986; Haynes et al., 1999). Indeed, thymocytes and thymic dendritic cells are susceptible to HIV infection and, consequently, their immunological and biological functions are altered by HIV infection (Evans et al., 2011; Young and Angel, 2011; Nunes-Cabaço et al., 2015). Importantly, SIV infection of rhesus macaques showed that lymphocytes and macrophages within the thymus are infected early (within 2 weeks) by SIV (Baskin et al., 1991). However, the role of the thymus and the mechanisms of increased Treg frequencies during HIV infection have not yet been well described. It is plausible that besides inducing peripheral Treg generation, TGF-β might play a part in the induction and maintenance of Tregs in the thymus during HIV infection. Hence, we hypothesized that TGF-β induces FoxP3 expression in HIV-infected thymocytes and thus contributes to higher thymic Treg frequencies during HIV infection. Herein, we assessed the impact of in vitro HIV infection of human thymocytes on the differentiation of tTregs and the influence of TGF-β1 in a co-culture model with OP9-DL1 cells expressing Notch ligand delta-like 1.

Tregs can differentiate either as a T-cell lineage within the thymus (tTregs) or be derived from naïve CD4+ T-cells in blood and secondary lymphoid tissues during inflammation (Chen et al., 2003; Curotto de Lafaille et al., 2004; Curotto de Lafaille and Lafaille, 2009; Sakaguchi et al., 2013). Although Treg frequencies and their immunosuppressive activities increase in individuals with chronic HIV infection, and TGF-β is known to cause fibrosis of lymphoid tissues and disease exacerbation in such individuals, the direct effects of HIV alone or combined with TGF-β in the induction of thymic tTregs remains unknown. In this study, in vitro HIV infection of human thymocytes was performed to assess its impact on FoxP3 expression and differentiation of tTregs along with the influence of TGF-β.

We first assessed if tTregs are susceptible to HIV infection and whether infection of thymocytes could result in increased Treg frequencies. Our results showed that although the majority of thymocytes are CXCR4+, tTregs express high levels of CCR5 and CXCR4, suggesting a potentially higher susceptibility of tTregs to HIV infection. Notably, in contrast to our observation on tTregs from thymic tissue, expression levels of CCR5 and CXCR4 in circulating Tregs were previously shown to be comparable to those of effector CD4 T-cells (Moreno-Fernandez et al., 2009). Previous reports also indicate that thymocytes are more susceptible to in vitro X4 infection compared to infection with R5-tropic HIV strains (Nunes-Cabaço et al., 2015), which correlates with higher expression of HIV co-receptor CXCR4 compared to CCR5 in human thymocytes, while this was not significant in our co-culture model of HIV infection. We thus assessed the permissiveness of Tregs versus non-Treg thymocytes to HIV infection using a co-culture model of human thymocytes and the OP9-DL1 cell line (Ramsdell et al., 2006). One of the main reasons for using OP9-DL1 co-culture, instead of the in vitro stimulation of thymocytes via anti-CD3 or cytokines such as IL-2, is that such stimulations either increase FoxP3 and/or CD25 expression on CD4 T cells (Sereti et al., 2000; Wang et al., 2007). Furthermore, our team previously reported this system as an optimal model to study the differentiation of human thymocytes in vitro, as well as their exogenous stimulation or HIV infection (Young and Angel, 2011; Young, 2012). Our in vitro data indicate that tTregs are refractory to HIV infection, which could be related to the fact that the transcription factor FoxP3 may block HIV viral transcription by restraining the expression of NFAT (Selliah et al., 2008). In agreement with this, it was demonstrated that Tregs and effector T-cells isolated from peripheral blood of HIV-infected individuals without treatment, have comparable levels of HIV p24 ex vivo; although in vitro HIV infection of Tregs is lower compared to effector T cells (Moreno-Fernandez et al., 2009).

TGF-β1 regulates FoxP3 expression in the thymus and lymphoid tissues during inflammation (Chen et al., 2003; Li and Flavell, 2008; Chevalier and Weiss, 2013; Jenabian et al., 2013; Chen and Konkel, 2015; Jurberg et al., 2015). Overexpression of TGF-β protein and mRNA by peripheral blood mononuclear cells (PBMCs) from HIV-infected donors has been previously demonstrated (Hu et al., 1996; Wiercinska-Drapalo et al., 2004; Theron et al., 2017). Indeed, HIV-gp160 and HIV-Tat induced TGF-β expression and increased TGF-β levels in the plasma have been correlated with disease progression (Kekow et al., 1990; Hu et al., 1996; Reinhold et al., 1999; Wiercinska-Drapalo et al., 2004). A major negative effect of TGF-β (one of the main cytokines produced by Tregs) in HIV-infected individuals is its role in lymphoid tissue fibrosis as it enhances the function of fibroblast cells, which results in excessive collagen deposition and lymphoid tissue dysfunction (Nilsson et al., 2006; Zeng et al., 2011; Theron et al., 2017). Based on these facts, we postulated that TGF-β1 might contribute to the induction and maintenance of thymic Tregs during HIV infection. Co-culture of human thymocytes with OP9-DL1 cells in presence of recombinant human TGF-β1 led to increased frequencies of CD127 (IL-7 receptor)-expressing CD4 T-cells, which confirmed that TGF-β1 promotes CD127 expression (Ouyang et al., 2013). Importantly, compared to untreated controls, FoxP3 expression and Treg frequencies were higher in all TGF-β1-treated specimens, whereas the presence of TGF-β1 inhibitor PFD blocked this effect. Our results indicate that HIV infection, by itself, does not increase FoxP3 expression, nor does the combination of HIV infection and TGF-β treatment. Indeed, Treg differentiation during HIV infection depends on a combination of multiple factors including, among others, TCR stimulation, immune activation/inflammation, IL-2 supply, and TGF-β1 activity. In our in vitro experiments only TGF-β1 was added without TCR activation (anti-CD3/CD28) which might have affected our observations. Moreover, the mRNA levels of TGF-β in in vitro HIV-infected thymocytes remained unchanged. Of note, we observed relatively low infection rates in thymocytes, which could be due to the fact that laboratory-constructed viral strains rather than clinical isolates were used, which may explain the differences when compared to previously reported data (Young and Angel, 2011; Young, 2012). Our results indicate that TGF-β does not alter HIV infection rates in vitro. It is to be noted that in vitro infection only mimics an in vivo infection setting, and does not serve as an exact replica of an in vivo infection. We do not observe TGF-β production during in vitro infection possibly for the reason that Tregs remain unaffected by infection or probably because tTregs present in the thymus are not yet functionally active. Finally, we acknowledge that the relatively small sample size in our study could impact the outcome of the results. However, due to the rareness of these specimens, it was not possible to increase the sample size for some experiments.

Altogether, our results indicate that direct in vitro HIV infection of human thymocytes in absence of TCR activation does not increase FoxP3 expression and tTreg differentiation, nor does the combination of HIV infection and TGF-β treatment. Although our initial hypothesis about the impact of TGF-β on thymic tTreg differentiation during HIV infection was not confirmed, our results suggest that differentiation of thymic tTregs within the thymus is distinct from peripheral blood and secondary lymphoid tissues and additional inflammatory mechanisms may be involved in the differentiation and thymic output of tTregs during HIV infection.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by Université du Québec à Montréal (UQAM, #216_e_453). Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

M-AJ designed the study. SS, AY, and OF performed the experiments and generated data. SB and JA provided access to specimens. SS, TS, AY, and OF analysed, discussed, and interpreted results throughout the study. SS, TS, and M-AJ wrote the paper. SS, TS, AY, OF, SB, JA, and M-AJ contributed to the refinement of the study and reviewed study outcomes. All authors contributed to the article and approved the submitted version.

This study was funded by the Canadian Institutes of Health Research (CIHR, grant MOP 142294) and in part by the AIDS and Infectious Diseases Network of the Réseau SIDA et maladies infectieuses du Fonds de recherche du Québec-Santé (FRQ-S) to M-AJ. SS was supported by an MSc excellence scholarship from UQAM’s foundation. AY is the recipient of an FRQ-S doctorate fellowship. M-AJ holds the tier 2 CIHR Canada Research Chair in Immuno-Virology.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.