Identification of inducible HIV reservoirs in tonsillar, intestinal and cervical tissue models of HIV latency

Gallego-Cortés, A.; Sánchez-Gaona, N.; Mancebo-Pérez, C.; Ruiz-Isant, O.; Benítez-Martínez, A.; Landolfi, S.; Castellví, J.; Pumarola, F.; Ortiz, N.; Llano, I.; Lorente, J.; Mañalich-Barrachina, L.; Prado, JG.; Martín-Gayo, E.; Falcó, V.; Genescà, M.; Buzon, MJ.

doi:10.1038/s41467-025-65288-9

Download PDF

Article
Open access
Published: 24 November 2025

Identification of inducible HIV reservoirs in tonsillar, intestinal and cervical tissue models of HIV latency

Nature Communications volume 16, Article number: 10353 (2025) Cite this article

3064 Accesses
1 Citations
46 Altmetric
Metrics details

Subjects

Abstract

HIV persists in diverse tissues, with distinct cellular reservoirs presenting a major barrier to a cure and requiring targeted therapeutic strategies to address this heterogeneity. Here, we develop tissue models of HIV latency using human tonsillar, intestinal and cervicovaginal tissues. These models reveal differential HIV infection across CD4⁺ T-cell subpopulations, with ART partially restoring CD4⁺ T cells and reducing intact HIV DNA. T follicular helper cells (T_{FH CD69+ CCR7-}) are the primary inducible reservoir in tonsils, while tissue-resident memory cells (T_{RM CD69+ CD49a+}) dominate in the intestine. Identification of markers for inducible reservoirs shows that CD69, CD45RO, and PD-1 are shared across tissues, while CXCR5 in tonsils and CD49a in the intestine act as tissue-specific markers. Furthermore, using different latency reversal agents (LRAs) demonstrates that Histone Deacetylase Inhibitors (HDACis) fail to induce HIV in any tissue, the SMAC mimetic AZD5582 is effective only in a resident-memory CD4⁺ T-cell subpopulation in the intestine, and IL-15 exhibits the broadest reactivation potential across tissues and CD4⁺ T-cell subsets. These models provide insights into the inducible reservoir’s composition in different tissues and inform strategies for its elimination.

Profound phenotypic and epigenetic heterogeneity of the HIV-1-infected CD4⁺ T cell reservoir

Article Open access 19 December 2022

HIV infection reprogrammes CD4+ T cells for quiescence and entry into proviral latency

Article Open access 26 September 2025

Phenotypic signatures of immune selection in HIV-1 reservoir cells

Article Open access 04 January 2023

Introduction

Despite significant progress in antiretroviral therapy (ART), HIV infection is still an incurable disease^1,2. While ART effectively suppresses HIV viremia in the bloodstream of people with HIV (PWH), it cannot completely eradicate the virus from the body due to the establishment of viral reservoirs. These reservoirs are mainly comprised of a pool of long-lived HIV-infected cells generated early after infection^3,4 and upon ART initiation⁵, which harbour integrated, intact HIV DNA capable of producing replicative viruses and largely evading the immune system^6,7,8. HIV reservoirs are distributed across diverse cell subtypes and tissue compartments^{9,10,11,12,13,14,15,16,17}, representing the primary barrier to curing HIV. Thus, identifying and understanding the distinctive characteristics of HIV reservoirs’ cells is essential for developing strategies aimed at eliminating HIV from the body.

HIV infects multiple immune cell populations^18,19, but memory CD4⁺ T cells are recognized as the primary contributors to the long-lived viral reservoir^20,21,22. Specifically, central memory (T_CM), transitional memory (T_TM), and effector memory (T_EM) CD4⁺ T cells are identified as major cellular reservoirs of latent HIV in the peripheral blood of PWH on ART^21,22,23,24. The pivotal role of less differentiated cell populations in HIV persistence has also been described^25,26. While cellular reservoirs have been extensively studied in the bloodstream of PWH, the majority of CD4⁺ T cells, and consequently viral reservoirs, reside in tissues. Therefore, it is crucial to investigate HIV persistence within tissue compartments.

HIV has been detected in most tissues throughout the human body, including critical sanctuary sites such as lymphoid tissues (LT), the gastrointestinal tract, the central nervous system, and the reproductive tract^17,27,28. These anatomical compartments may sustain low levels of viral production due to their immune-privileged environment^29,30,31,32 or reduced penetration of antiretroviral drugs^33,34,35. Furthermore, clonal expansion of latently infected CD4⁺ T cells represents another mechanism contributing to HIV persistence in these tissues^36,37,38. Compelling evidence indicates that lymph nodes (LN), gut mucosa, and gut-associated lymphoid tissue (GALT) serve as major reservoirs for HIV in PWH on ART^39,40,41. Moreover, a higher viral burden has been observed in the gastrointestinal and cervical tissues compared to peripheral blood in PWH on suppressive treatment^9,41,42. In these tissue reservoirs, a significant proportion of the HIV resides within CD4⁺ T cells, with varying phenotypes and proportions across tissues, adding to the complexity of HIV reservoirs. Therefore, it is imperative to identify the specific CD4⁺ T-cell reservoirs within these tissues and develop strategies to purge them.

Certain CD4⁺ T-cell populations, such as follicular helper T cells (T_FH) in LN and type 17 helper T cells (T_H17) in the gut, are pivotal in sustaining persistent HIV infection within these tissues^43,44,45. Moreover, tissue-resident memory T cells (T_RM) have been identified as a critical cellular reservoir in the cervix and LN^9,46. T_RM cells are a non-circulating memory subpopulation found in multiple tissues that mediate rapid protective responses to site-specific infections⁴⁷. The presence of T_RM within distinct locations drives the development of T_RM cells functionally and phenotypically diverse, yet all T_RM cells share a core gene and protein profile^47,48,49. They are characterized by the expression of CD69, which confers tissue retention properties^47,50. T_RM can also express the adhesion molecules integrin αE (CD103) and integrin α1 (CD49a), the immune checkpoint receptor PD-1, and the IL7 receptor-alpha chain (CD127)^47,51,52. Notably, T_RM are highly susceptible to HIV infection and possess enhanced proliferative capacity and longevity, making them potential targets for HIV persistence^{9,47,53,54,55,56}. Moreover, the expression of PD-1 and CD127 on CD4⁺ T cells has been associated with a phenotype that predominantly supports latent HIV infection^13,14. Our current understanding of how T_RM cells contribute to tissue reservoirs is limited. Considering their widespread distribution and abundance throughout the body, further research into the role of T_RM cells in HIV persistence is essential.

The diverse array of cellular reservoirs within tissues underscores the need to develop curative strategies capable of effectively targeting a wide range of cell populations to cure HIV. Recent efforts to eliminate the latent reservoir have focused on the “shock and kill” strategy. This approach uses latency reversal agents (LRAs) to reactivate HIV from latently infected cells, which are then cleared by virus-induced cytopathic effects or the immune system^57,58. The effect of LRAs on circulating cellular reservoirs has been extensively studied^{22,59,60,61,62,63}; however, information regarding their impact on HIV reservoirs within tissues is scarce. Differential responses to current LRAs have been observed among the various CD4⁺ T-cell subsets from the blood^22,59,61, suggesting that cell populations within tissue compartments may also exhibit varying responses. Characterizing the susceptibility of distinct tissue CD4⁺ T-cell subpopulations to pharmacological HIV reactivation is crucial for designing more effective therapies aimed at reducing HIV reservoirs. Nevertheless, the limited and challenging accessibility to tissue samples from PWH hinders these studies.

In our study, we established models of latent HIV infection using human tonsillar, intestinal and cervicovaginal explants. We aimed to characterize productive and persistent viral infection, identify inducible HIV reservoirs, and evaluate the efficacy of various LRAs in these tissues. Our results showed a differential distribution of HIV infection among CD4⁺ T-cell subpopulations across tissues. We also demonstrated the presence of inducible HIV reservoirs in our tissue models, primarily localized within T_FH cells in the tonsils and T_RM subsets in the intestinal tissue. Importantly, our study revealed distinct reactivation patterns of several LRAs, not only between tissues but also across different CD4⁺ T-cell populations in these tissues, highlighting the tissue-specific and cell-specific nature of both HIV reservoir establishment and viral reactivation. These findings underscore the importance of developing therapeutic strategies with a broader spectrum of activity to target the different cellular reservoirs within anatomical compartments.

Results

Development of explant models for the study of HIV persistence in tissues

We initially worked on the establishment of three different explant models to characterize persistent HIV infection in human tissues. Tonsillar, intestinal and cervicovaginal explants were obtained from routine surgeries of uninfected donors and dissected into small blocks. Tissue blocks were cultured, HIV-infected and ART-treated as described in the schematic representation in Fig. 1a. First, we assessed the viability of total tissue-derived cells and CD4⁺ T cells throughout the culture period (Supplementary Fig. 1). While a general decline in overall tissue viability was observed over time, CD4⁺ T-cell viability was comparatively less impacted and remained largely stable, supporting the use of these tissue models for reliable analysis of HIV infection dynamics and reservoir establishment. Next, we identified the optimal conditions for establishing HIV infection in the explant models. To achieve this, we monitored the progression of infection by assessing intracellular p24 expression in CD4⁺ T cells throughout both productive and treated viral infection. Longitudinal analyses were performed for the tonsillar and intestinal models, but were not feasible for the cervicovaginal model due to the limited size of the explants. A robust and consistent HIV infection was evident by day 5 in the tonsillar tissue (median_TO= 1.0% p24⁺ in CD4⁺ T cells, p_TO < 0.0001) and by day 6 in the intestinal and cervical tissues (median_GUT = 3.6% p24⁺ in CD4⁺ T cells, p_GUT < 0.0001; median_CVX = 0.8% p24 in CD4⁺ T cells, p_CVX = 0.06) (Fig. 1b, c). Intestinal explants consistently exhibited higher levels of HIV infection compared to tonsillar and cervicovaginal explants, and ART administration effectively and significantly reduced HIV infection in all tissues within two to three days (median decrease in the percentage of p24⁺ cells: TO = 77%, GUT = 68% and CVX = 84%) (Fig. 1c, d). Notably, the production of p24 was not entirely abrogated by ART (median_TO= 0.5%, median_GUT = 2.3% and median_CVX = 0.05%) (Fig. 1d). Additionally, we analysed longitudinal changes in key immune populations in the tonsillar and intestinal models, including CD4⁺ T and CD8⁺ T cells; NK cells (total CD56⁺ NK cells, as well as CD56⁺ CD16⁺ and CD56⁺ CD16^- subsets) and B cells. An example of the gating strategy used for these analyses is shown in Supplementary Fig. 2a–c. Overall, HIV infection did not cause significant alterations in the proportion of CD8⁺ T cells, NK cells and B cells (Supplementary Fig. 3). However, viral infection reduced the percentage of CD4⁺ T cells in both tissues, which was partially reverted after ART introduction (Fig. 1e, Supplementary Fig. 3). We also examined the impact of HIV infection and ART administration on CD4⁺ T cells within cervical tissue (Supplementary Fig. 4); however, no significant changes were observed, potentially due to fluctuations in the proportions of other cellular populations that may have overshadowed these effects (Fig. 1e).

**Fig. 1: Tonsillar, intestinal and cervical models of latent HIV infection.**

Next, we evaluated the presence of inducible reservoirs by triggering viral reactivation in the tissue models under ART. Pharmacological stimulation using phorbol 12-myristate 13-acetate (PMA) and ionomycin on isolated ART-treated CD4⁺ T cells from all tissues modestly reduced cell viability (Supplementary Fig. 5), but led to a significant increase in the percentage of intracellular p24 (median increase in the percentage of p24⁺ cells: TO = 66%, GUT = 41% and CVX = 44%), confirming the presence and reactivation of HIV reservoirs within tissues (Fig. 1f). Interestingly, the percentage of p24⁺ cells under PMA/ionomycin stimulation was higher in the cervix (median = 0.8%) compared to the tonsils (median = 0.1%) (Fig. 1f), despite the cervix having lower levels of productive infection (Fig. 1b, d). This discrepancy may be due to the smaller number of cells isolated from cervicovaginal tissue, which could overestimate its reactivation potential.

Finally, we analysed the size of the established HIV reservoirs within the tissues after 7-8 days of productive infection. The Intact Proviral DNA Assay (IPDA) was performed on positively isolated CD4⁺ T cells from tonsillar and intestinal tissue blocks to quantify the different forms of HIV DNA (intact, 3’ deleted/hypermutated and 5’ deleted variants). Due to the limited number of CD4⁺ T cells isolated from cervicovaginal tissue, we measured only the amount of total HIV DNA using quantitative PCR (qPCR). Intestinal CD4⁺ T cells harboured the most HIV DNA, followed by tonsillar and cervicovaginal cells (Fig. 2a), consistent with the infection frequencies observed (Fig. 1d). When quantifying the distinct HIV DNA forms, we found that 14% of the proviruses in tonsillar CD4⁺ T cells exhibited intact DNA, a proportion comparable to the 15% observed in CD4⁺ T cells from the intestine (Fig. 2b). In both tissues, the 5’-deleted variant was the most prevalent form (mean_TO = 49%, mean_GUT = 50%), followed closely by the hypermutated variant (mean_TO= 37%, mean_GUT = 35%). Overall, the proportions of HIV DNA forms were similar between the two tissues, with no statistically significant differences observed (Fig. 2c). After ART administration, we observed a significant decline in total HIV DNA levels across all tissues (Fig. 2d). All forms of HIV DNA in tonsillar and intestinal tissues were reduced, with the decrease of the intact HIV DNA form being slightly more pronounced (Fig. 2d). It is important to note that, during active infection, a fraction of productively infected cells may still express CD4 and therefore be included in the positively selected CD4⁺ population. This could partly account for the pronounced decline in HIV DNA observed following ART initiation.

**Fig. 2: HIV reservoir size in the tissue models.**

Overall, we developed ex vivo tissue-based models to study inducible HIV reservoirs during ART in tonsillar, intestinal and cervicovaginal tissues. While we acknowledge the limitations inherent to in vitro systems, our models enable the assessment of HIV-associated changes, including a reduction in CD4⁺ T cells following infection, the partial protective effect of ART, and a measurable impact of ART on HIV DNA levels. In addition, we observed that these models respond to pharmacological agents known to induce viral reactivation. We believe that our tissue-based approach provides a more physiologically relevant context to study HIV latency and its potential reversal in tissues.

Distinct representation of CD4⁺ T-cell subpopulations across tissues and differential contribution to HIV infection in tonsillar and intestinal compartments

We then characterized the subset composition and phenotypic heterogeneity of the CD4⁺ T-cell repertoires in tonsillar, intestinal and cervicovaginal tissues. CD4⁺ T cells isolated from the tissue blocks were analysed via flow cytometry using a comprehensive panel of markers for T-cell differentiation, function and migration. Through unsupervised clustering based on the expression profiles of these markers (Supplementary Fig. 6a, b), we identified 12 clusters of CD4⁺ T cells (Fig. 3a, b). These CD4⁺ T-cell subsets included naïve T cells (C01-02 T_NA: CD45RO^- CCR7⁺), central memory T cells (C03, T_CM: CD45RO⁺ CCR7⁺), effector memory T cells (C04, T_EM: CD45RO⁺ CCR7^-) and terminal effector memory T cells (C05-06, T_TEM: CD45RO^- CCR7^-). T_NA and T_TEM populations were further differentiated based on the expression of the activation marker CD69: clusters C01 and C05 represented cells in a resting state (CD69^-), while clusters C02 and C06 showed stimulated phenotypes (CD69⁺). Moreover, we identified follicular helper T cells (C07-09), characterized by the expression of the chemokine receptor CXCR5, which facilitates their proximity to B cells within lymph nodes, and PD-1⁶⁴ (T_FH: CXCR5⁺ PD-1⁺). These T_FH cells were further classified based on the differential expression of CCR7, which, together with CXCR5, is crucial for their positioning within lymph nodes and maturation^65,66. Additionally, T_FH cells were categorized based on the presence or absence of CD69, a receptor also associated with tissue retention⁴⁷, distinguishing between resident (CD69⁺) and non-resident T_FH cells (CD69^-)⁶⁷. Considering the marker expression profiles and in line with previous studies^66,68,69, we defined C07:T_{FH CCR7+} as an early precursor population of pre-T_FH cells, C08:T_{FH CD69+ CCR7+} as pre-T_FH cells, and C09:T_{FH CD69+ CCR7-} as mature germinal center (GC) T_FH cells. Lastly, we identified three clusters of resident memory T cells, characterized by the essential expression of CD69 and the downregulation of CCR7, to ensure their retention within non-lymphoid tissues, as well as a memory phenotype (C10-12, T_RM: CD45RO⁺ CCR7^- CD69⁺). These T_RM clusters differed in the expression of the adhesion molecules CD103 and CD49a, which determine their localization within tissues; as well as PD-1 expression⁴⁷. The expression of the homoeostatic cytokine IL7 receptor CD127, commonly present in long-living memory T cells⁷⁰, was analysed in the identified subsets and found to be higher in T_RM clusters (C11 and C12), T_FH cells (C07 and C08) and in the T_CM subset (C03) (Fig. 3b, Supplementary Fig. 6a). Importantly, tissue memory CD4⁺ T cells expressing CD127 have previously been shown to contain latent HIV and be prone to viral reactivation¹⁴. Additionally, we assessed the expression of KLRG1, a marker associated with a more differentiated CD4⁺ T-cell profile^71,72 which has also been identified as an immune inhibitory checkpoint receptor⁷³. KLRG1 is expressed on CD4⁺ T cells harbouring inducible HIV genomes⁷³ and clonally expanded proviruses⁴⁶. We found that the T_RM cluster C11 exhibited slightly higher KLRG1 expression (Fig. 3b, Supplementary Fig. 6a).

**Fig. 3: Distinct CD4⁺ T-cell compartments in the tissue models and differential distribution of HIV infection in tonsillar and intestinal tissues.**

Next, we determined the proportions of the identified clusters across tissues (Fig. 3c). The tonsillar tissue exhibited a high abundance of less differentiated subsets and T_FH cells, consistent with existing literature^{74,75,76,77,78}. Specifically, we observed a greater representation of clusters C01:T_NA and C09:T_{FH CD69+ CCR7-}, comprising 37% and 23% of the pool of CD4⁺ T cells, respectively (Fig. 3c). Conversely, T_RM clusters predominated in the intestinal tissue. In particular, clusters C10:T_{RM CD69+ PD-1+} and C11:T_{RM CD69+ CD49a+} were the most abundant, accounting for 20% and 36% of the total CD4⁺ T cells, respectively (Fig. 3c). This prevalence of T_RM cells within the human intestinal CD4⁺ T-cell compartment has been previously documented^54,79. Similarly, in the cervicovaginal tissue, cluster C11:T_{RM CD69+ CD49a+} was the most abundant subset, comprising 27% of total CD4⁺ T cells, followed closely by cluster C04:T_EM, which represented 24% of the pool (Fig. 3c). These findings align with prior reports describing a predominance of T_EM and T_RM CD4⁺ T cells in the cervical immune environment^9,80,81. When comparing cluster distributions across tissues, only a few significant differences were observed between intestinal and cervical compartments, likely reflecting the greater divergence of tonsils from these tissues (Fig. 3c). Nevertheless, pairwise comparisons revealed that the majority of CD4⁺ T-cell populations differed significantly between tissues (Fig. 3d, Supplementary Fig. 7a).

We also evaluated HIV infection among all the identified CD4⁺ T-cell subsets from the tonsils and the intestine. This analysis could not be performed for the cervicovaginal tissue due to the limited amount of CD4⁺ T cells isolated. Firstly, we assessed the impact of productive infection in the abundance of the different clusters. Despite the ability of HIV to modify the phenotype of CD4⁺ T cells^82,83,84, no novel clusters emerged in the HIV-infected conditions and we only observed a slight yet significant difference in the proportion of cluster C05:T_TEM in the intestine (p = 0.049) (Fig. 3e). The proportions of the other clusters remained consistent in both tissues, indicating that the analysed phenotypes were not significantly altered. We then quantified HIV infection across the distinct CD4⁺ T-cell clusters by measuring the expression of the p24 viral protein. As expected, the percentages of p24⁺ cells were higher in the intestinal compared to the tonsillar CD4⁺ T cells (Fig. 3f). A significant HIV infection was observed across most subpopulations in tonsillar tissues, with notably higher infection levels detected in the T_RM clusters, particularly in C10:T_{RM CD69+ PD-1+} (Fig. 3f). In the intestine, viral infection predominantly occurred within the T_RM clusters C10:T_{RM CD69+ PD-1+}, C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+}. Comparable percentages of p24⁺ cells were detected in the subset C04:T_EM, while lower yet significant levels of viral infection were observed in the clusters C03:T_CM, C05:T_TEM, C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} (Fig. 3f). Interestingly, the subpopulation with the highest CD127 levels and slight KLRG1 expression (C11:T_{RM CD69+ CD49a+}) was among the most significantly infected clusters in both tissues (Fig. 3f). Ultimately, we determined the contribution of each cluster to the overall infection within the total CD4⁺ T-cell population (Fig. 3g). The tonsils exhibited a dispersed distribution of infection but it primarily localized in T_FH and T_RM subsets. Clusters C08:T_{FH CD69+ CCR7+}, C09:T_{FH CD69+ CCR7-} and C10:T_{RM CD69+ PD-1+} made the greatest contributions to total levels of infection accounting for 13%, 40%, and 22% of the total p24⁺ CD4⁺ T cells, respectively (Fig. 3g, Supplementary Fig. 7b). Moreover, the contributions of clusters C01:T_NA, C07:T_{FH CCR7+}, C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} to the overall infection were significantly higher in the tonsils than in the intestine (Fig. 3g). In contrast, infection concentrated in the T_RM clusters C10:T_{RM CD69+ PD-1+}, C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+} in the intestine, accounting for 26%, 49%, and 9% of the total p24⁺ CD4⁺ T cells respectively, with the contribution of C11 and C12 significantly surpassing those in the tonsils (Fig. 3g, Supplementary Fig. 7b).

Overall, our findings unveiled notable differences in the composition of the CD4⁺ T-cell compartments across the studied tissues, with T_NA and T_FH cells predominating in the tonsils, while intestinal and cervical CD4⁺ T cells primarily exhibited a T_RM phenotype. The T_FH fraction in the tonsils harboured the greatest number of infected cells, while HIV infection was primarily localized within the T_RM subpopulations in the intestinal tissue. Moreover, the contribution of the CD4⁺ T-cell subsets to the total pool of HIV-infected cells differed markedly between tonsillar and intestinal tissues.

Tonsillar and intestinal inducible HIV reservoirs are constituted of distinct CD4⁺ T-cell subpopulations

As HIV reservoirs are established promptly following primary infection and shortly after the initiation of ART⁵, we aimed to identify the specific CD4⁺ T-cell clusters that constituted inducible HIV reservoirs within the pool of tonsillar and intestinal infected CD4⁺ T cells. To achieve this, tissue blocks were HIV-infected and treated with ART. Inducible viral reservoirs were identified after maximal cell activation. For that, we stimulated isolated ART-treated HIV-infected CD4⁺ T cells with PMA and ionomycin (Fig. 1a). First, we compared the proportions of CD4⁺ T-cell clusters under unstimulated and PMA/ionomycin-stimulated conditions, as treatment with these agents might induce phenotypic changes associated with T-cell activation and differentiation^{15,22,85,86,87}. Thus, as previously reported^22,86,88,89, we anticipated an upregulation of CD69 and PD-1 expression, and a downregulation of CXCR5 and CD127. Besides significantly altering some of the CD69^- and non-T_RM CD69⁺ clusters (C01:T_NA, C02:T_{NA CD69+}, C03:T_CM, C04:T_EM, C05:T_TEM, C06:T_{TEM CD69+} and C07:T_{FH CCR7+}), the proportion of the T_RM subsets remained largely unchanged in both tissues (Supplementary Fig. 8a, b). Additionally, T_FH cells expressing CD69 in the tonsils, which contributed the most to HIV infection, were similarly unaffected (Supplementary Fig. 8a). This indicated that these cells retained their phenotype following stimulation with PMA and ionomycin. One possible explanation is that CD69 expression was already elevated in these subsets and did not increase further following cell activation, as previously shown for T_RM in cervix⁹. Next, we measured p24 expression in the clusters following viral reactivation (Fig. 4a). Five tonsillar and three intestinal clusters exhibited statistically significant viral reactivation, identifying them as inducible HIV reservoirs during ART. Specifically, cluster C10:T_{RM CD69+ PD-1+} showed significant reactivation in both tissues; clusters C03:T_CM, C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} were predominantly reactivated in the tonsils; and clusters C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+} showed higher reactivation in the intestine (Fig. 4a). Additionally, the percentage of p24⁺ cells was elevated in cluster C02:T_{NA CD69+} in the tonsil, despite not detecting significant infection in this cluster (Fig. 3f). We believe that this might be due to the upregulation of CD69 in cells from cluster C01:T_NA, which could transition to C02:T_{NA CD69+} upon activation.

**Fig. 4: Tonsillar and intestinal inducible HIV reservoirs are constituted of distinct subpopulations of CD4⁺ T cells.**

Subsequently, we analysed the contribution of each cluster to the overall inducible HIV reservoir in CD4⁺ T cells (Fig. 4b, Supplementary Fig. 8c). In the tonsils, cluster C09:T_{FH CD69+ CCR7-} was the primary contributor to total levels of inducible HIV, accounting for 64% of the p24⁺ CD4⁺ T cells. Reactivated cells from cluster C08:T_{FH CD69+ CCR7+} also comprised a significant portion, representing 21% of the total reactivated cells (Fig. 4b). Among intestinal cells, cluster C11:T_{RM CD69+ CD49a+} (the one expressing the highest levels of CD127 and KLRG1) predominated, representing 42% of virally reactivated cells. Cluster C10:T_{RM CD69+ PD-1+} also significantly contributed to the total inducible cell reservoir, accounting for 24% of the p24⁺ CD4⁺ T cells. Although cluster C09:T_{FH CD69+ CCR7-} did not consistently reactivate in all intestinal samples, it substantially contributed to the reactivated cell pool in those intestines where it did (Fig. 4b). Significant differences were observed in the contribution of certain clusters to reactivation between tissues. In particular, p24⁺ cells from clusters C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} were significantly more prevalent in the pool of reactivated tonsillar CD4⁺ T cells compared to their intestinal counterparts. In contrast, clusters C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+} exhibited the opposite pattern (Fig. 4b).

To further characterize the profile of reactivated cells and identify antigens that could define inducible HIV reservoirs in these tissues, we analysed the expression of the phenotypic markers within the pool of p24⁺ CD4⁺ T cells following PMA/ionomycin stimulation (Fig. 4c). In both tonsillar and intestinal tissues, reactivated cells consistently expressed CD69, and nearly all of them displayed the memory marker CD45RO. This may appear inconsistent with our previous observations, where T_CM cells, which lack CD69, reactivated significantly (Fig. 4a); however, it is plausible that CD69 expression is induced in the p24⁺ CD4⁺ T cells from this cluster upon reactivation. Notably, the majority of the reactivated cells from the tonsils expressed CXCR5 and PD-1, whereas those from the intestine predominantly expressed CD49a and PD-1. CD49a and CD103 were absent in tonsillar reactivated cells, potentially due to the under-representation of T_RM cells within the tonsillar compartment (Fig. 3c, d). These findings identify CD69, CD45RO, and PD-1 as shared markers of inducible reservoirs in both tonsillar and intestinal tissues. Additionally, CXCR5 emerges as predominantly associated with tonsillar reservoirs, whereas CD49a and, to a lesser extent, CD103 are specific to intestinal reservoirs.

Overall, while some clusters comprise potential HIV reservoirs in both tissues, there are unique tonsillar and intestinal CD4⁺ T-cell reservoirs. In the tonsils, the primary reservoir is formed by T_FH cells, whereas T_RM populations constitute the largest inducible HIV reservoir in the intestinal tissue. Inducible HIV reservoirs within these tissue compartments shared CD69, CD45RO and PD-1 expression, while distinct markers, such as CXCR5 in the tonsils and CD49a in the intestine, delineated specific reservoir populations. Together, these results suggest both shared and tissue-restricted features of inducible HIV reservoirs across these compartments and identify potential marker candidates for characterizing inducible reservoirs in tonsils and intestinal tissues.

Current LRAs demonstrated differing potency and breath of viral reactivation between tissue reservoirs

We aimed to evaluate the efficacy of different LRAs in inducing viral reactivation in the HIV-latency tissue models. The LRAs tested included the protein kinase C (PKC) agonist Ingenol (ING), histone deacetylase inhibitors (HDACis) Romidepsin (RMD) and Panobinostat (PNB), the SMAC mimetic AZD5582 (AZD) and the interleukin IL-15. Additionally, Ingenol and Romidepsin were evaluated in combination (I + R). First, we assessed whether LRA stimulation affected the viability of isolated ART-treated CD4⁺ T cells (Supplementary Fig. 9). Overall, CD4⁺ T cells remained viable following LRA treatment, with some subtle reductions observed that did not compromise data interpretation. Next, the efficacy of the LRAs was determined by quantifying the percentage of p24⁺ cells after LRA stimulation in isolated and ART-treated CD4⁺ T cells. IL-15 induced a potent HIV reactivation across all tissue reservoirs (Fig. 5a). The combination of Ingenol and Romidepsin significantly reactivated HIV in tonsillar and intestinal CD4⁺ T cells, while Ingenol alone was sufficient to reactivate the virus in tonsillar and cervicovaginal reservoir cells (Fig. 5a). AZD5582 demonstrated effectiveness exclusively in reactivating the intestinal reservoirs (Fig. 5a). Subsequently, we compared the efficacy of the LRAs between tonsillar and intestinal tissues only, as the limited CD4⁺ T-cell yield from cervicovaginal tissues precluded testing all LRAs in a single sample. A marked difference was observed in the efficacy of Ingenol between the intestinal and tonsillar reservoirs (Fig. 5b). Additionally, we sought to validate the observed HIV-reactivation patterns by quantifying p24 protein levels in the supernatants of reactivated CD4⁺ T-cell cultures using the ultrasensitive SIMOA assay (Fig. 5c). This analysis was performed exclusively on tonsillar and intestinal samples, as the limited number of cells recovered from cervical tissue did not allow for reliable detection of p24 in the supernatant following reactivation. For this validation, we selected the LRAs that had elicited the most robust responses in our prior assays: the combination of Ingenol and Romidepsin, and IL-15. In tonsillar cultures, Ingenol was also tested as a single agent. All evaluated LRAs induced a substantial increase in extracellular p24 levels, further confirming their ability to effectively reactivate latent HIV from tissue-derived CD4⁺ T cells (Fig. 5c).

**Fig. 5: Differences in the efficacy of current LRAs in inducing viral reactivation of tissue reservoirs.**

Next, we evaluated the impact of these LRAs on the distinct subpopulations of CD4⁺ T cells, focusing on the clusters identified as potential HIV reservoirs in the tonsillar and intestinal tissues. We first examined whether the abundance of these CD4⁺ T-cell subpopulations differed between the non-reactivated (ART) and LRA-treated conditions (Supplemental Fig. 10a, b). In the tonsils, the proportion of the primary contributor to the inducible HIV reservoir, C09:T_{FH CD69+ CCR7-}, remained consistently unaffected by all LRAs. The second major inducible cluster, C08:T_{FH CD69+ CCR7+}, was reduced in the IL-15 treatment condition and marginally increased with Ingenol, both alone and in combination with Romidepsin. This combination also minimally impacted the proportions of clusters C03:T_CM and C10:T_{RM CD69+ PD-1+} (Supplementary Fig. 10a, c). In contrast, we observed consistent proportions of all clusters identified as inducible HIV reservoirs in the intestine (C10:T_{RM CD69+ PD-1+}, C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+}) before and after treatment with LRAs (Supplementary Fig. 10b, c). Then, we measured the percentage of p24⁺ cells within each cluster after treatment with the different LRAs (Fig. 5d). Ingenol alone and in combination with Romidepsin demonstrated a potent capacity to reactivate the tonsillar subsets C03:T_CM, C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} and the intestinal clusters C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+} (Fig. 5d). Moreover, Ingenol with Romidepsin successfully reactivated the intestinal reservoir C10:T_{RM CD69+ PD-1+}. AZD5582 showed no effect in any of the tonsillar subsets but was highly effective in reactivating the intestinal T_RM reservoir C11:T_{RM CD69+ CD49a+}. IL-15 exhibited substantial reactivation capacity in the tonsillar reservoirs C03:T_CM, C08:T_{FH CD69+ CCR7+} and C10:T_{RM CD69+ PD-1+} and the intestinal clusters C11:T_{RM CD69+ CD49a+} and C12:T_{RM CD69+ CD49a+ CD103+}. Interestingly, while any other LRA failed to induce a significant reactivation in the tonsillar subsets C06:T_{TEM CD69+} and C11:T_{RM CD69+ CD49a+} and the intestinal cluster C08:T_{FH CD69+ CCR7+}, IL-15 could reactivate these clusters. Romidepsin and Panobinostat showed no reactivation potential in any tonsillar nor intestinal CD4⁺ T-cell subpopulations (Supplementary Fig. 10d). Importantly, despite being susceptible to HIV infection (Fig. 3f), tonsillar clusters C04:T_EM, C07:T_{FH CCR7+} and C12:T_{RM CD69+ CD49a+ CD103+} and intestinal clusters C03:T_CM, C04:T_EM, C05:T_TEM were not reactivated with any of the LRAs tested.

Finally, we assessed the contribution of each cluster to the total p24⁺ CD4⁺ T cells following stimulation with the LRAs (Fig. 5e). In the tonsils, reactivated cells from clusters C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} were the most abundant in all LRA-stimulated conditions, with a minor yet significant contribution from cluster C03:T_CM. Conversely, across most LRAs, the intestinal cluster C11:T_{RM CD69+ CD49a+} made the largest contribution to the total number of reactivated cells, followed by cluster C12:T_{RM CD69+ CD49a+ CD103+}. Notably, Ingenol and, to a lesser extent, IL-15, were the most effective LRAs in the tonsils, while IL-15 induced the broadest reactivation in the intestine (Fig. 5e).

In summary, we observed that HIV reservoirs from the tonsils, intestine and cervix are more efficiently reactivated by Ingenol alone or in combination with Romidepsin, and IL-15. However, at the level of CD4⁺ T-cell subpopulations, LRAs induce reactivation in distinct subsets both across different tissues and within the same tissue. Specifically, while T_CM and T_FH cells in the tonsils and the T_RM fraction in the intestine were reactivated by most tested LRAs, the specific clusters affected differed between LRAs. Notably, IL-15 was capable of reactivating populations in both tissues that remained unresponsive to other LRAs.

Heterogeneous responses of tonsillar and intestinal CD4⁺ T-cell subpopulations to LRAs

Subsequently, we evaluated the differential response of reservoir cells from the same clusters across tonsillar and intestinal tissues to the selected LRAs. Our previous data indicated that HIV predominantly persists in resident CD4⁺ T cells in both tissues, and that these subpopulations exhibited enhanced responsiveness to the tested LRAs. Nonetheless, inducible HIV reservoirs were also identified in the tonsillar clusters C03:T_CM and C06:T_{TEM CD69+} (Fig. 6a, b). The reservoir in cluster C03:T_CM could be reactivated by Ingenol alone and in combination with Romidepsin, as well as with IL-15. Remarkably, the combination treatment was most potent for cluster C03:T_CM, achieving reactivation in nearly all tonsillar samples (Fig. 6a). In contrast, reactivation of cluster C06:T_{TEM CD69+} was solely observed with IL-15 (Fig. 6b). In the tonsils, HIV reservoirs primarily localized within the resident T_FH fraction (C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-}), which notably exhibited the most robust reactivation response to the LRAs. However, differences were observed in the efficacy and extent of viral reactivation (Fig. 6c, d). Both tonsillar clusters C08:T_{FH CD69+ CCR7+} and C09:T_{FH CD69+ CCR7-} responded to stimulation with Ingenol alone and in combination with Romidepsin, with subset C08:T_{FH CD69+ CCR7+} exhibiting a greater response to the combined treatment (Fig. 6c); while cluster C09:T_{FH CD69+ CCR7-}, to the single Ingenol (Fig. 6d). Additionally, IL-15 could significantly reactivate reservoir cells from cluster C08:T_{FH CD69+ CCR7+}. Conversely, in the intestine, despite being significantly infected, these clusters were rarely virally reactivated by the LRAs. Only HIV present in cluster C08:T_{FH CD69+ CCR7+} was reactivated with IL-15 (Fig. 6c). As IL-15 could reactivate HIV from cluster C08:T_{FH CD69+ CCR7+} in both tonsillar and intestinal tissues, we compared its potency between the tissues and observed that they exhibited similar reactivation efficiencies (Supplementary Fig. 11). Ultimately, we evaluated the reactivation in the T_RM subsets, which comprise the majority of the HIV reservoir in the intestine but only a small proportion in the tonsils. The T_RM cluster C10:T_{RM CD69+ PD-1+} harboured HIV reservoirs in both tonsillar and intestinal tissues (Fig. 6e), yet they were reactivated by distinct LRAs. Specifically, this reservoir was selectively reactivated by IL-15 in the tonsils, and by the co-administration of Ingenol with Romidepsin in the intestine. The largest HIV reservoir in the intestinal tissue consisted of infected cells from cluster C11:T_{RM CD69+ CD49a+}, which exhibit the highest levels of CD127 and slight KLRG1 expression. These cells could be reactivated by nearly all tested LRAs except for Romidepsin alone and Panobinostat (Fig. 6f). Notably, this was the only cluster responsive to reactivation with AZD5582. The most potent reactivation was achieved with IL-15, which also successfully reactivated the same cluster in the tonsils (Fig. 6f). Moreover, the efficacy of IL-15 was comparable in both the tonsillar and intestinal clusters C11:T_{RM CD69+ CD49a+} (Supplementary Fig. 11). Finally, we analysed the response of cluster C12:T_{RM CD69+ CD49a+ CD103+} to the tested LRAs (Fig. 6g). We observed that an inducible HIV reservoir was present exclusively in the intestinal cluster. Cells from this subset were successfully reactivated using Ingenol, both as a single treatment and in combination with Romidepsin, as well as with IL-15. The combination of Ingenol and Romidepsin proved to be the most effective reactivation strategy for this subset.

In summary, our findings reveal distinct reactivation patterns to current LRAs within the same CD4⁺ T-cell subpopulations across the tissues. HIV reservoirs within resident T_FH subsets and T_CM cells exhibited superior responsiveness to current LRAs in the tonsils, whereas HIV within T_RM clusters showed greater reactivation in the intestine. This underscores the tissue-specific nature of both HIV reservoir establishment and reactivation dynamics.

Validation of LRA efficacy in isolated CD4⁺ T-cell subpopulations

Lastly, we sought to assess the capacity of the most potent LRAs to reactivate latent HIV in isolated CD4⁺ T-cell subsets without substantially altering their phenotypic identity. To this end, specific CD4⁺ T-cell subpopulations from both tonsillar and intestinal tissues were isolated by fluorescence-activated cell sorting (FACS) (Supplementary Fig. 12), stimulated with the selected LRAs, and evaluated for HIV reactivation by quantifying intracellular p24 expression via flow cytometry (Fig. 7a, b). The LRAs tested included the combination of Ingenol and Romidepsin, and IL-15 for both tissue types, as well as Ingenol alone for tonsillar cells. The CD4⁺ T-cell subpopulations sorted from tonsils included naïve CD4⁺ T cells (T_NA: CD45RO⁻ CCR7⁺ CD69⁻) and follicular helper cells (T_FH: CXCR5⁺ PD-1⁺), including the T_{FH CD69⁻ CCR7⁺}, T_{FH CD69⁺ CCR7⁺} and T_{FH CD69⁺ CCR7⁻} subsets. From intestinal tissues, resident memory CD4⁺ T cells (T_RM: CD45RO⁺ CCR7⁻ CD69⁺ CXCR5⁻), including the T_{RM CD69⁺ CD103⁺ CD49a⁺}, T_{RM CD69⁺ CD49a⁺} and T_{RM CD69⁺ PD-1⁺} subsets, and non-resident (T_non-RM: CD69⁻) CD4⁺ T cells were isolated. Among tonsillar subsets, T_FH cells exhibited higher levels of reactivation than T_NA cells (Fig. 7a), confirming our earlier results (Fig. 5d). In both subpopulations, the combination of Ingenol and Romidepsin induced higher reactivation compared to Ingenol alone or IL-15 (Fig. 7a). Regarding intestinal CD4⁺ T-cell subpopulations, T_RM cells demonstrated strong HIV reactivation in response to both the Ingenol–Romidepsin combination and IL-15, while T_non-RM population displayed lower but still detectable reactivation levels (Fig. 7b), in line with our previous findings (Fig. 5d). Phenotypic analysis post-stimulation revealed that subsets largely maintained their original profiles (Fig. 7c). The most prominent changes observed were a consistent downregulation of CD127 across all stimulated conditions and an upregulation of CD69 in all LRA-treated samples, with the exception of those treated with IL-15. The reduction in CD127 expression aligns with its established role as a marker of quiescent T cells¹⁴. In contrast, the upregulation of CD69 has variable implications depending on the baseline expression profile of the subsets. In populations that constitutively express CD69, such as most tonsillar T_FH and intestinal T_RM cells, this change did not interfere with their phenotypic classification. However, in the tonsillar T_NA and intestinal T_non-RM subsets, the induction of CD69 expression may confound accurate phenotypic identification. In the case of tonsillar T_NA cells, this upregulation likely reflects early activation and possibly correspond to a transition from cluster C01 to C02 (Fig. 3b), since other phenotypic markers remained unaltered. Similarly, intestinal T_NON-RM cells, did not substantially upregulate T_RM-associated markers (CD103, CD49a, PD-1), suggesting that they acquired an activated state without adopting a resident memory phenotype.

**Fig. 7: HIV-reactivation of sorted CD4⁺ T-cell subpopulations from tonsillar and intestinal tissues.**

Overall, our assays with FACS-isolated CD4⁺ T subpopulations demonstrate that, while LRA stimulation can induce some phenotypic alterations, the major HIV reservoirs, T_FH cells in tonsils and T_RM cells in the intestine, largely retain their phenotypic identity. These findings align with our prior observations derived from analyses of the total CD4⁺ T cell compartment.

Discussion

Despite significant advancements in antiretroviral therapy, which have transformed HIV into a manageable chronic condition, a cure remains elusive. The persistence of HIV, mainly in long-lived CD4⁺ T cells where it can evade immune surveillance, poses a major barrier to eradication efforts⁸. These HIV reservoirs span diverse CD4⁺ T-cell populations and anatomical compartments¹⁹, making it essential to understand their distinct characteristics and distribution to develop effective therapeutic strategies. In this study, we developed explant models of human tonsils, intestinal mucosa and cervicovaginal tissue to investigate the susceptibility of CD4⁺ T-cell subpopulations within these tissues to HIV infection and reservoir establishment. We revealed tissue and cell-specific dynamics influencing HIV infection, viral persistence under ART, and HIV reactivation, demonstrating differential responses of the CD4⁺ T-cell viral reservoirs to various LRAs. Our findings highlight the need for broad-spectrum therapies capable of effectively targeting the diverse cellular reservoirs of HIV across different anatomical compartments.

We developed explant models of HIV latency, based on an established experimental setup^90,91, using three crucial tissues: tonsils, intestinal mucosa and cervix. First, we observed a decline in CD4⁺ T cells during untreated HIV infection^92,93,94. Second, ART administration partially reversed CD4⁺ T-cell loss and facilitated viral reservoir establishment^5,95,96,97. Third, ART caused a more pronounced decrease in intact HIV DNA compared to defective forms^98,99. Fourth, we detected residual viral production despite ART, a phenomenon commonly observed in blood, lymph nodes and gastrointestinal tissue from PWH^{100,101,102,103,104}. Although the first three observations could not be conclusively established in the cervix, residual viral transcription was detected in cervical mucosa despite ART suppression, consistent with prior reports from our group⁹. Overall, our explant models, provide valuable tools for investigating the mechanisms of HIV persistence and strategies directed to target tissue reservoirs.

In our study, we reported a predominance of T_NA and T_FH cells in tonsils and T_RM subsets in the intestinal mucosa and cervix, aligning with previous reports^{9,47,54,75,76,77,80,81}. It is important to consider that age differences among tissue donors may influence these proportions and, as a result, impact the susceptibility to HIV infection of the tissues. Tonsils derived from paediatric individuals exhibit a higher proportion of less differentiated cells compared to those from adults¹⁰⁵. Furthermore, aging is known to increase the frequency of T_RM cells in cervical tissue⁸¹. Similarly, intestinal samples from adult donors may display a higher prevalence of more differentiated and senescent cells due to the advanced age of the individuals. While we acknowledge that age-related differences among donors represent a limitation of our study, we believe that our tissue models serve as suitable and valuable platforms for studying HIV dynamics in these anatomical sites.

Productive HIV infection was detected in the majority of the clusters in the tonsils and intestine but, remarkably, T_RM cells consistently exhibited the highest levels of HIV infection and significantly contributed to the pool of infected CD4⁺ T cells in both tissues. This was particularly noteworthy in the tonsils, where, despite being scarce (approximately 5% of total CD4⁺ T cells), T_RM-phenotype infected cells were notably abundant comprising approximately 27% of total infected cells. Moreover, we have previously demonstrated a preferential infection of T_RM cells in cervical tissue⁹. Consequently, our findings, supported by other studies^106,107, strongly suggest that T_RM cells are preferential targets for HIV infection within tissues. In addition, HIV infection has been reported to induce a T_RM-like phenotype in resting memory CD4⁺ T cells¹⁰⁸ which enhances their susceptibility to productive infection. This further supports the notion that the T_RM-phenotype might be associated with increased permissiveness to HIV infection.

After ART administration, inducible viral reservoirs were identified in all tissue models. In tonsils, we detected HIV reservoirs across a broad range of cell subpopulations. Notably, resident T_FH cell clusters and the T_RM subset C10:T_{RM CD69+ PD-1+} were prominent contributors to the inducible HIV reservoir. Although not in tonsils, previous studies have specifically linked CXCR5, PD-1 and CD69 expression to latent HIV infection in CD4⁺ T cells from other lymph-node tissues^15,109. Furthermore, CD4⁺ T_CM cells, recognized as major HIV reservoirs in peripheral blood^21,59,61, also play a significant role as inducible viral reservoirs within the tonsillar lymphoid tissue. In the intestine, T_RM cells, particularly those with high CD127 and slight KLRG1 expression, emerged as the predominant HIV reservoir. CD4⁺ T cells expressing CD69 and PD-1 have been previously shown to support latent HIV infection in the intestinal tissue¹⁵. Moreover, T_RM markers CD49a and CD127, as well as KLRG1 have been associated with CD4⁺ T cells harbouring HIV reservoirs^9,14,46,73. Notably, our study identifies CD69, CD45RO and PD-1 as common markers of inducible reservoirs in both tissues, while CXCR5 is specific to tonsils and CD49a and CD103 are exclusive markers of HIV reservoirs in the intestine. These findings reveal a spectrum of phenotypes contributing to HIV persistence in tissues, with populations expressing markers previously associated with HIV latency, underscoring their potential as therapeutic targets.

Our study evaluated several LRAs for their capacity to reactivate HIV within tissue reservoirs. Such investigations with human tissues are infrequent, primarily due to the challenges involved in obtaining samples. Among the LRAs tested, Ingenol demonstrated great potency particularly in tonsillar and cervical reservoirs and broad reactivation activity across the distinct tonsillar CD4⁺ T-cell subpopulations. In tonsils, Ingenol preferentially reactivated the T_CM subset over the other non-resident populations, mirroring similar patterns observed in peripheral blood⁶¹. Moreover, Ingenol showed robust reactivation of T_FH subsets in the tonsils and T_RM cells in the intestine. The differential responsiveness among these subsets may be attributed to varying levels of PKC isoforms and NF-κB pathway components. In contrast, both HDACis Romidepsin and Panobinostat, which have been reported to reactivate latent HIV in blood cells^{60,61,63,110,111,112}, showed no effect on any tissue. Notably, Panobinostat’s ineffectiveness in reactivating tissue reservoirs has also been corroborated in animal models¹¹⁰. Previous research has demonstrated that these drugs differ in their ability to inhibit cell-associated HDAC activity⁶⁰. Therefore, it is plausible to suggest that the differential expression of HDAC isoforms between the CD4⁺ T-cell subsets in blood and tissues might contribute to their intrinsic limitations in reactivating HIV in tissues.

Recently, the potential use of IL-15 as LRA has been extensively evaluated. IL-15 has demonstrated efficacy in reactivating HIV^{62,113,114,115,116} and, unlike most LRAs, is capable of priming latently infected cells for recognition by CD8⁺ T cells and NK cells while also enhancing the cytotoxic activity of these cells^62,114,117. Our study revealed a potent effect of IL-15 in inducing HIV reactivation across all studied tissues, demonstrating efficacy even in CD4⁺ T-cell subsets unresponsive to PMA/ionomycin stimulation. The dual role of IL-15 in both reactivating HIV and promoting cell survival and proliferation^118,119 may account for the observed increase in p24 production. The heightened responsiveness of certain CD4⁺ T-cell subsets to IL-15 is likely attributable to increased expression of the IL-15 receptor (IL-15R). IL-15 dependence for CD8⁺ T_RM cell maintenance varies across tissues¹²⁰, suggesting tissue-specific differences in IL-15R expression within these populations, a pattern that may extend to the CD4⁺ T_RM compartment. Although T_CM cells are generally not associated with elevated IL-15R expression¹²¹, our findings revealed IL-15-mediated reactivation in the tonsillar T_CM subset. Additionally, while T_FH cells from LNs have been reported to express higher levels of IL-15R compared to other memory subsets⁴⁶, only select T_FH subpopulations responded to IL-15 stimulation. These observations suggest heterogeneity in IL-15R expression across tissues and within distinct CD4⁺ T-cell subpopulations. Beyond receptor expression, IL-15-induced intracellular signalling pathways may further modulate HIV reactivation. IL-15 has been reported to inhibit the action of SAM domain and HD domain-containing protein 1 (SAMHD1)¹¹⁹, an enzyme that impairs HIV gene expression and negatively modulates viral latency reversal in CD4⁺ T cells¹²². Notably, T_FH cells exhibit low level of SAMHD1¹²³, which may facilitate their susceptibility to IL-15-mediated reactivation. Collectively, these findings suggest that the differential responsiveness of CD4⁺ T-cell subsets to IL-15 may arise from the combined effects of IL-15 receptor expression and SAMHD1 activity, a hypothesis that warrants further investigation.

AZD5582 has demonstrated significant efficacy in reactivating HIV in humanized mice and SIV-infected rhesus macaques under ART^124,125. In these animal models, HIV reactivation was observed in lymph nodes¹²⁴, but its impact in the gastrointestinal tract remained unexplored. Conversely, we revealed that AZD5582 exclusively induced HIV reactivation in the intestine, with minimal effect on CD4⁺ T cells from the tonsillar and cervical tissues. The differences in findings may arise from the detection techniques employed; while that study measured viral RNA, our detection relied on p24 protein levels which may selectively capture only the most robust reactivations^126,127. Indeed, AZD5582’s effect was only observed in the T_RM subset with the highest levels of infection and susceptibility to viral reactivation by most LRAs studied. These observations suggest that AZD5582 may exhibit only a modest potency in CD4⁺ T cells from intestine, triggering viral reactivation primarily in the subsets harbouring more inducible HIV.

We observed that certain CD4⁺ T-cell subsets in tonsillar and intestinal tissues, despite exhibiting high infection rates, showed resistance to viral reactivation. One possible explanation is that LRAs might enhance HIV RNA transcription in specific cells but fail to induce robust production of viral particles, as previously demonstrated in blood^126,127. Another possibility is that these cell types tend to sustain productive rather than latent HIV infection and may undergo rapid cell death following infection^128,129. Additionally, LRAs might induce phenotypic changes that hinder the identification of the original cell subset. Ultimately, specific phenotypes of tonsillar and intestinal CD4⁺ T cells could be in a deeper latent state than those in peripheral blood, rendering them less responsive to LRA stimulation. Indeed, this could be the case for intestinal CD4⁺ T cells, as elevated HIV transcription has been observed in circulating CD4⁺ T cells compared to rectal tissue CD4⁺ T cells following LRA stimulation. A greater block to HIV transcription has been described as responsible for this phenomenon^130,131. Moreover, silencing due to repressive chromatin modifications or transcriptional interference^132,133,134, as well as integration into heterochromatin¹³⁵, could profoundly impact viral reactivation in specific cell subsets. Importantly, these mechanisms are largely unexplored in tissue cells. Overall, the absence of detectable viral reactivation in certain CD4⁺ T-cell subsets susceptible to productive HIV infection may be attributed to a combination of factors, including defective production of viral particles, LRA-induced phenotypic changes, and profound transcriptional silencing within the cells.

Our study has several limitations. Firstly, the lifespan of the tissues is limited, preventing long-term studies of HIV persistence. However, since the majority of HIV reservoirs are established shortly after initial infection and at the onset of ART⁵, we consider our models well-suited for identifying potential cell reservoirs. Secondly, we used a CCR5-tropic HIV strain and did not test CXCR4-tropic viruses. While differences in infection rates and affected CD4⁺ T-cell subpopulations may occur, we believe the inducibility of the established reservoirs would not be significantly altered. Furthermore, using a CCR5-tropic strain more accurately models an early HIV infection, making it more suitable for studying viral reservoir establishment. Thirdly, due to the difficulty in isolating individual tissue cell subsets for reservoir identification, we identified inducible reservoirs after maximal in vitro activation of the total tissue CD4⁺ T-cell compartments. While most non-inducible proviruses in blood of PWH on ART are likely defective, one study suggests that 11.7% may comprise intact HIV genomes⁹⁹. Whether this fraction differs in tissue cells remains unclear. Additionally, it is uncertain if the inducible reservoirs identified in our study contain replication-competent proviruses. Moreover, it remains unknown whether the observed increase in p24 expression detected by flow cytometry reflects truly viral reactivation from latently infected cells or enhanced viral protein production in previously undetectable virus-producing cells. Finally, a principal limitation of our study is the uncertainty regarding potential phenotypic alterations in certain CD4⁺ T-cell populations following pharmacological reactivation^{15,22,85,86,87}. Nonetheless, the frequency of the main clusters in both unstimulated and stimulated conditions remained consistent, as also observed in a previous study⁵⁹, and we did not observe the creation of new cell clusters after viral reactivation, as reported in an early study¹⁵. Furthermore, post-LRA stimulation analysis of major HIV reservoirs sorted from tonsillar and intestinal tissues demonstrated that these populations largely retained their phenotypic profiles.

It should also be acknowledged that, while our ex vivo tissue models support infection of physiologically relevant CD4⁺ T-cell subsets and allow the detection of intact and inducible proviruses under ART, they differ in certain respects from the in vivo reservoirs observed in PWH on long-term ART. Notably, studies analyzing lymph nodes from ART-treated individuals have reported significantly lower frequencies of both inducible and intact proviruses compared to our ex vivo tissue models. For instance, two studies^136,137 found that only approximately 8-9 proviruses per million CD4⁺ T cells were inducible, based on the detection of a small number of reactivated cells. In contrast, we observed inducible viruses in approximately 0.05% of CD4⁺ T cells in tonsils and about 0.5% in both intestinal and cervical tissues. Likewise, intact proviral frequencies in vivo are considerably lower, ranging from ~100 intact proviruses per million CD4⁺ T cells¹³⁶ to fewer than 2% of total HIV genomes¹³⁷, whereas our ex vivo models exhibited frequencies on the order of 10,000–100,000 intact proviruses per million CD4⁺ T cells. These discrepancies likely reflect the favourable conditions of ex vivo models, including recent infection, limited immune pressure, and short culture duration. Nonetheless, our model provides a valuable platform for the controlled study of HIV dynamics within tissue-resident compartments under ART and for the preclinical evaluation of latency-reversing and curative strategies. Future refinements, such as extending culture duration under ART while maintaining tissue viability through the addition of supportive factors, may better replicate long-term latency and clonal expansion observed in vivo.

In summary, using highly physiological tissue models, we identified inducible cellular HIV reservoirs in critical tissues with varying susceptibility to current LRAs. The differential responses to pharmacological viral reactivation among different CD4⁺ T-cell subpopulations within tonsillar and intestinal tissues, as well as variations within the same subpopulations across the two tissues, suggest that viral reactivation is influenced by both cellular and tissue-specific factors. Therefore, our findings have significant implications for designing therapies aimed at promoting HIV reactivation in tissues.

Methods

Patient samples and ethics statement

Tonsils were collected from routine tonsillectomies and tonsil reduction surgeries performed at the Otorhinolaryngology Department of Hospital Universitari Vall d’Hebron (HUVH). Gastrointestinal tissues, specifically colon resections, were procured from colorectal cancer patients undergoing colectomies at the General and Digestive Surgery department of HUVH. Cervicovaginal tissues were obtained from non-neoplastic hysterectomies and cervical amputations performed in the Gynaecology Department of HUVH. The study protocols received ethical approval from the Institutional Review Board at HUVH [PR(AG)582/2020PR]. All tissues were surgically removed for non-inflammatory reasons, and only healthy regions of the collected samples were used in the study. Participants included paediatric patients for tonsillar samples and adult patients for intestinal and cervical tissues, all confirmed to be HIV-negative at the time of collection. Samples from each donor were randomly allocated to experiments. Donor age in tonsillar tissues from paediatric donors could represent a potential covariate across experiments; however, this was not controlled for, as all experimental conditions were performed in parallel using tissue from the same tonsil. In contrast, intestinal and cervical tissues were obtained from adult donors, from whom age was not considered a relevant covariate. Sample sizes for each experiment were determined by tissue availability, and results were reproduced across independent donors. Participation in the study required written informed consent provided by adult participants themselves or by legal guardians for paediatric patients. Participants did not receive monetary compensation for participating in this study. The anonymity and untraceability of the collected samples were ensured. Information on the number of participants, their gender, age, surgical procedure and indication for surgery is provided in Supplementary Table 1.

Tissue explant models of productive and latent HIV infection

Tonsillar, intestinal and cervicovaginal tissues were processed as described below. Tissue resections were delivered in RPMI 1640 medium (Life Technologies, cat.no. 12004997) containing antibiotics and antifungals within 4 h post-surgery. The tissues were dissected into uniform blocks of approximately 2 mm × 2 mm x 2 mm in size. For tonsillar and intestinal tissues, 10-12 blocks were placed on a piece of absorbable gelatine sponge (Surgispon, Aegis Lifesciences, cat. no. SSP-805010) suspended in RPMI 1640 medium supplemented with 20% foetal bovine serum (FBS, Life Technologies, cat. no. A5256701) (R20 medium), 100 U/ml penicillin, 100 µg/ml streptomycin (Capricorn Scientific, cat. no. PS-B), and 66 µg/ml Amikacin (Normon, cat. no. 791301.6) in a 6-well plate. Timentin (Caisson Labs, cat. no. T034) at a concentration of 310 µg/ml was added to the media only on the first day of culture. Viral infection was initiated by depositing 5 µl of the CCR5-tropic HIV viral strain HIV_BAL (156,250 TCID₅₀) onto each tissue block, with some blocks left uninfected to serve as controls. Infected and control blocks were cultured for an additional 7-9 days at 37 °C and 5% CO₂, with the media and sponges replaced every 3 days. On day 5 (tonsils) or day 6 (intestine), ART comprising 1 µM Raltegravir (NIH AIDS Reagent Programme, cat. no. 0980), 1 µM Darunavir (NIH AIDS Reagent Programme, cat. no. 0989), and 1 µM Nevirapine (Sigma-Aldrich, cat.no. SML0097), was added on top of the infected tissue blocks and left for a period of 2-3 days. For cervicovaginal explants, 8-12 blocks of tissue per condition were immersed in 310 µl R20 media with 100 U/ml penicillin and 100 µg/ml streptomycin in a 24-well plate and infected with 40 µl of HIV_BAL virus (156,250 TCID50). Some blocks were placed in 350 µl of medium to serve as controls. After a 2-hour incubation at 37 °C and 5% CO₂, tissue blocks were rinsed three times with 3 mL of 1X PBS in 6-well plates and transferred back into a 12-well plate, placing 8-12 blocks per well in 1 mL of R20 media with antibiotics. Infected and control blocks were cultured for an additional 8 days, with the medium replaced every 3 days. On day 6, ART comprising 1 µM Raltegravir, 1 µM Darunavir and 1 µM Nevirapine was added to the culture medium containing some of the infected tissue blocks for 2 days^90,91.

Digestion of tissue blocks

Digestion procedures were optimized for each tissue. Tonsillar blocks were mechanically digested using disposable pellet pestles in RPMI 1640 supplemented with 5% FBS (R5). For intestinal and cervicovaginal tissues, enzymatic digestion was required prior to the mechanical disruption of the blocks. Intestinal blocks were incubated in R5 containing 2.5 mg/ml collagenase IV (Fisher Scientific, cat. no. 10780004) and 100 μg/ml DNase I (Roche, cat. no.10104159001) while cervicovaginal blocks were submerged in R5 with 5 mg/ml collagenase IV, both for 30 min at 37 °C and 400 rpm. We verified that the enzymatic digestion did not significantly alter the expression of key phenotypic markers used in our study (Supplementary Fig. 13). Following digestion, all cellular suspensions were filtered through 70 μm cell strainers (Labclinics, cat.no. PLC93070) to remove large aggregates. Finally, cells were washed twice with 1X PBS^90,91.

Longitudinal phenotyping of key immune populations and HIV infection assessment by flow cytometry

To characterize changes in pivotal immune populations and measure HIV infection in CD4⁺ T cells during the culture of the infected explants, tissue blocks were digested at various time points and subsequently stained for both surface and intracellular markers. Cellular suspensions were stained with LIVE/DEAD™ Fixable Aqua Dead Cell Stain (1/250, Invitrogen, cat. no. L34966) for 20 min at room temperature (RT). After washing with 1X PBS, cells were incubated with anti-HLA-DR-PerCP-Cy5.5 (1/100, G46-6, Beckton Dickinson, cat. no. 560652), anti-CD56-FITC (2/100, B159, Beckton Dickinson, cat. no. 562794), anti-CD3-PE-Cy5 (0.6/100, UCHT-1, Biolegend, cat. no. 300410), anti-CD45-AF700 (1/100, HI30, Biolegend, cat. no. 304024), anti-CD8-APC (2.5/100, RPA-T8, Beckton Dickinson, cat. no. 555369), anti-CD16-BV605 (2.5/100, 3G8, Beckton Dickinson, cat. no. 563172) and anti-CD19-V500 (2.5/100, HIB19, Beckton Dickinson, cat. no. 561121) antibodies in staining buffer (1X PBS 3% FBS) for 20 min at RT. After washing with staining buffer, cells were fixed and permeabilized using Fixation/Permeabilization Solution (Beckton Dickinson; cat. no. 554714;) for 20 min at 4 °C, followed by two washes with BD Perm/Wash™ buffer. Then, intracellular staining with anti-p24 antibody (0.5/100, KC57, Beckman Coulter, cat. no. 6604667) was performed for 30 min on ice, followed by an additional 30 min at RT. After washing, cells were fixed with 2% PFA. Samples were acquired in a LSRFortessa (Becton Dickinson) flow cytometer and analysed with FlowJo v10 software (TreeStar).

The degree of HIV infection in CD4⁺ T cells was also assessed by staining digested tissue blocks using a panel comprising: LIVE/DEAD™ Fixable Far Red Dead Cell Stain (1/250, Invitrogen, cat. no. L34974) or LIVE/DEAD™ Fixable Aqua Dead Cell Stain (1/250, Invitrogen, cat. no. L34966), anti-CD45-FITC (1/100, HI30, Biolegend, cat. no. 304006) or anti-CD45-BV605 (2.5/100, HI30, Becton Dickinson, cat. no. 564047), anti-CD3-Per-Cp (10/100, SK7, Becton Dickinson, cat. no. 345766) or anti-CD3-AF700 (1/100, SK7, Biolegend, cat. no. 344822), anti-CD8-APC (2.5/100, RPA-T8, Beckton Dickinson, cat. no. 555369) and anti-p24-PE (0.5/100, KC57, Beckman Coulter, cat. no. 6604667) antibodies. Viability, surface, and intracellular staining, as well as fixation, were performed as previously described. Sample acquisition was conducted using a BD FACSCalibur or a BD LSRFortessa flow cytometer, and data analysis was carried out with FlowJo v10 software. All antibodies are commercially available and validated by their respective manufacturers.

Quantification of HIV DNA in isolated CD4⁺ T cells by Intact Proviral DNA Assay (IPDA) and Quantitative PCR (qPCR)

To quantify all forms of HIV DNA within ART-treated and untreated infected explants, tissue blocks from tonsils collected on day 7, and from intestinal and cervicovaginal tissues collected on day 8, were digested as previously described. Following digestion, cellular suspensions were processed with the EasySep™ Dead Cell Removal (Annexin V) Kit (StemCell, cat. no. 17899) to remove apoptotic cells. Two rounds of immunomagnetic negative selection were performed to ensure isolation of only live cells. Subsequently, CD4⁺ T cells were isolated from the live cell fraction using the Dynabeads™ CD4 Positive Isolation Kit (Invitrogen, cat. no. 11331D). Isolated cells were lysed in a Proteinase K-containing lysis buffer at 55 °C over-night and at 95 °C for 5 min. IPDA was then conducted on lysed CD4⁺ T cells from tonsillar and intestinal tissues¹³⁸. Quantification of intact, 5’ deleted and 3’ deleted/hypermutated proviruses was carried out using specific primers and probes targeting the Ψ (psi) HIV-1 gene (HIV-1 Ψ forward: 5’-CAGGACTCGGCTTGCTGAAG-3’, HIV-1 Ψ reverse: 5’-GCACCCATCTCTCTCCTTCTAGC-3’ and, HIV-1 Ψ probe: 5’−6-FAM-TTTTGGCGTACTCACCAGT-MGBNFQ-3’) and the env HIV gene (HIV-1 env forward: 5’-AGTGGTGCAGAGAGAAAAAAGAGC-3’, HIV-1 env reverse: 5’-GTCTGGCCTGTACCGTCAGC-3’, HIV-1 env intact probe: 5’-VIC-CCTTGGGTTCTTGGGA-MGBNFQ-3’, and HIV-1 anti-Hypermutant env probe: 5’-CCTTAGGTTCTTAGGAGC-MGBNFQ-3’) (generated by Sigma-Aldrich and IDT). The human RPP30 gene served as a reference for cell input normalization (RPP30_1 forward 5’-GATTTGGACCTGCGAGCG-3’, RPP30_1 reverse 5’-GCGGCTGTCTCCACAAGT-3’, RPP30_1 probe 5’-FAM-CTGACCTGA-ZEN-AGGCTCT-IABkFQ-3’, RPP30_2 forward 5’-CCATTTGCTGCTCCTTGGG-3’, RPP30_2 reverse 5’-CATGCAAAGGAGGAAGCCG-3’ and RPP30_2 probe 5’-HEX-AAGGAGCAA-ZEN-GGTTCTATTGTAG-IABkFQ-3’) (generated by Sigma-Aldrich and IDT). The cycling parameters were 95 °C for 2 min, and then 95 °C for 15 s and 60 °C for 30 s for 40 cycles of amplification. Samples were analysed using a QIAcuity One 2 plex System (Qiagen). Intact provirus measurements were corrected for DNA shearing, and samples with excessive shearing (>0.85) were excluded from the final analysis. The mean shearing value for the intestinal samples was 0.69, while the mean for tonsillar samples was 0.49. After correction for DNA shearing, results were expressed as proviral copies per 10⁶ CD4⁺ T cells. To evaluate the effect of ART, the percentage reduction in the frequency of each HIV DNA form was calculated for each tissue sample by comparing the ART-treated and untreated conditions using the following formula: [1 − (frequency in treated condition/frequency in the untreated condition)] × 100.

IPDA could not be performed on isolated CD4⁺ T cells from cervicovaginal tissue due to the limited cell yield. Consequently, qPCR was employed as an alternative method. Total HIV DNA in the CD4⁺ T-cell lysates was quantified by qPCR using primers and probes specific for the 1-LTR HIV region (LTR forward: 5’-TTAAGCCTCAATAAAGCTTGCC-3’, LTR reverse: 5’-GTTCGGGCGCCACTGCTAG-3’, and probe: 5’-CCAGAGTCACACAACAGACGGGCA-3’) (generated by IDT). The CCR5 gene was used for cell input normalization (CCR5 forward: 5′-GCTGTGTTTGCGTCTCTCCCAGGA-3′, CCR5 reverse: 5′-CTCACAGCCCTGTGCCTCTTCTTC-3′, and probe: 5′-AGCAGCGGCAGGACCAGCCCCAAG-3′) (generated by IDT). The cycling parameters were 50 °C for 2 min, 95 °C for 10 min and then 95 °C for 15 s and 60 °C for 1 min for 55 cycles of amplification. Samples were analysed in a QuantStudio^TM 5 Real-Time PCR System, and quantification of DNA copies was performed using a standard curve. Results were expressed as HIV-DNA copies per 10⁶ CD4⁺ T cells.

Ex vivo reactivation with LRAs of HIV-infected and ART-treated CD4⁺ T cells

Tissue blocks from uninfected, HIV-infected, and HIV-infected and ART-treated conditions collected on day 7 for tonsils, and on day 8 for the intestinal and cervicovaginal tissue, were digested according to the previously described protocol. CD4⁺ T cells were then isolated using the MagniSort Human CD4⁺ T Cell Enrichment commercial kit (Invitrogen, cat. no. 8804-6811-74), with two rounds of cell separation performed to maximize cell purity. CD4⁺ T-cell purity following magnetic enrichment was evaluated (Supplementary Fig. 14). In all conditions, isolated CD4⁺ T cells were cultured at a density of 10⁶ cells/ml in R10 medium containing 10 μM Q-VD-OPh (Selleckchem, cat. no. S7311) for a minimum of 2 h at 37 °C and 5% CO₂. For HIV-infected and ART-treated CD4⁺ T cells, 1 µM Raltegravir, 1 µM Darunavir and 1 µM Nevirapine were added to the R10 media. After the incubation, uninfected and HIV-infected CD4⁺ T cells remained untreated, while HIV-infected and ART-treated CD4⁺ T cells were stimulated for 20-22 h with LRAs. LRAs were added to the media at the following concentrations: 100 nM Ingenol-3-angelate (Sigma-Aldrich, cat. no. SML1318), 40 nM Romidepsin (Selleckchem, cat. no. S3020), 30 nM Panobinostat (Selleckchem, cat. no. S1030), 100 nM AZD5582 (MedChem, cat.no. HY-12600), 50 ng/ml IL-15 (Myltenyi Biotech, cat. no. 130-095-760). The positive control consisted of 81 nM PMA and 1 μM Ionomycin (Abcam, cat. no. ab120297 and ab120370) in the medium, while the negative control comprised solely of R10 medium with ART. All compounds were reconstituted in either deionized sterile-filtered water, 0.1% BSA in 1X PBS, or DMSO (at a maximum concentration of 0.006%).

CD4⁺ T-cell subsets phenotyping and detection of HIV-1 p24 protein by flow cytometry

Identification of CD4⁺ T-cell subpopulations present in the tissues and measurement of HIV infection through p24 protein detection within these subsets was performed using flow cytometry. Previously cultured CD4⁺ T cells were stained using LIVE/DEAD™ Fixable Aqua Dead Cell Stain (1/250, Invitrogen, cat. no. L34966) for 20 min at RT. Following washing with 1X PBS, cells were stained with anti-CXCR5-BB700 (2/100, RF8B2, Beckton Dickinson, cat. no. 566470) and anti-CCR7-AF647 (10/100, 150503, Beckton Dickinson, cat. no. 560816) in staining buffer for 30 min at 37 °C. After washing with staining buffer, cells were incubated with the following antibodies for 20 min at RT: anti-CD127-FITC (1/100, A019D5, Biolegend, cat. no. 351312), anti-CD49a-PE-Cy7 (0.6/100, TS2/7, Biolegend, cat. no. 328311), anti-CD3-PE-Cy5 (0.6/100, UCHT-1, Biolegend, cat. no. 300410), anti-CD69-PE-CF594 (1/100, FN50, Beckton Dickinson, cat. no. 562617), anti-PD-1-APC-Fire750 (1/100, EH12.2H7, Biolegend, cat. no. 329954), anti-CD45-AF700 (1/100, HI30, Biolegend, cat. no. 304024), anti-CD103-BV650 (0.6/100, Ber-ACT8, Beckton Dickinson, cat. no. 743653), anti-CD45RO-BV605 (1.25/100, UCHL1, Biolegend, cat. no. 304238) and anti-KLRG1-BV421 (5/100, 14C2A07, Biolegend, cat. no. 368604). Next, cells were washed with staining buffer and fixed and permeabilized using Fixation/Permeabilization Solution (Beckton Dickinson; cat. no. 554714) for 20 min at 4 ^oC. Two washes with BD Perm/Wash™ buffer were conducted and intracellular staining with anti-p24-PE antibody (0.5/100, KC57, Beckman Coulter, cat. no. 6604667) was then performed for 30 min on ice, followed by an additional 30 min at RT. Cells were then washed and fixed with 2% PFA. Samples were acquired using a BD LSR Fortessa flow cytometer. All antibodies are commercially available and validated by their respective manufacturers.

The acquired data were analysed using the online platform OMIQ (www.omiq.ai, www.dotmatics.com). The clustering algorithm Flow-Self Organizing Maps (FlowSOM) was employed to identify clusters based on the expression of CD45RO, CCR7, CD69, CD103, CD49a, CD127, CXCR5, PD-1 and KLRG1. All events within pre-gated live CD3⁺ cells were concatenated across tissues and experimental conditions for subsequent analysis. To visualize the identified clusters, the dimensionality reduction algorithm opt-SNE was applied. Specifically, opt-SNE was performed on a randomly selected subset of 6 × 10⁵ cells from uninfected tonsillar and intestinal tissues, and on all available uninfected cells (4 × 10³) from cervical tissue, based on the expression of the aforementioned markers. Manual gating was used to assess p24 expression, and viral reactivation was represented as the difference in the frequency of p24⁺ cells between each LRA-stimulated condition and the ART-treated control. Additionally, FlowJo v10 software was employed to apply Boolean filters to identify specific subsets of live CD3⁺ p24⁺ cells, based on the expression of the selected markers, in the PMA/ionomycin-stimulated conditions from tonsillar and intestinal tissues. PESTLE v2.0 and SPICE v6.0 software were used to generate pie charts illustrating the proportions of the different subsets and their marker expression profiles.

Detection of HIV-1 p24 in CD4⁺ T-cell culture supernatants via Single Molecule Array (SIMOA) assay

Supernatants from overnight cultures of isolated CD4⁺ T cells were collected and stored at –80 °C. HIV-1 Gag p24 concentrations in the supernatants were quantified using the Simoa® HIV p24 Advantage Kit (Quanterix, USA; cat. no. 102215) on a Simoa HD-1 analyzer, following the manufacturer’s instructions. Samples were thawed at room temperature and centrifuged at 10,000 G for 5 min prior to p24 quantification. Four-parameter logistic (4PL) regression fitting was used to calculate p24 concentrations. The lower limit of quantification (LLOQ) was 0.010 pg/mL.

Sorting and viral reactivation of tonsillar and intestinal CD4⁺ T-cell subpopulations

HIV-infected and ART-treated tissue blocks collected on day 7 (tonsils) and day 8 (intestine) were digested and CD4⁺ T cells were isolated using the MagniSort™ Human CD4⁺ T Cell Enrichment Kit, as previously described. Isolated CD4⁺ T cells were stained with LIVE/DEAD™ Fixable Aqua viability dye (1/250, Invitrogen, cat. no. L34966) and the following panel of antibodies: anti-CD4-BUV496 (2.5/100, RPA-T4, eBioscience, cat. no. 364-0049-42), anti-CD8-APC-Fire810 (1.25/100, SK1, Biolegend, cat. no. 344764), anti-CXCR5-BB700 (2/100, RF8B2, Beckton Dickinson, cat. no. 566470), anti-CCR7-AF647 (10/100, 150503, Beckton Dickinson, cat. no. 560816), anti-CD49a-PE-Cy7 (0.6/100, TS2/7, Biolegend, cat. no. 328311), anti-CD3-PE-Cy5 (0.6/100, UCHT-1, Biolegend, cat. no. 300410), anti-CD69-PE-CF594 (1/100, FN50, Beckton Dickinson, cat. no. 562617), anti-PD-1-APC-Fire750 (1/100, EH12.2H7, Biolegend, cat. no. 329954), anti-CD45-AF700 (1/100, HI30, Biolegend, cat. no. 304024), anti-CD103-BV650 (0.6/100, Ber-ACT8, Beckton Dickinson, cat. no. 743653), and anti-CD45RO-BV605 (1.25/100, UCHL1, Biolegend, cat. no. 304238). CD4⁺ T-cell populations (CD45⁺ CD3⁺ CD4⁺ CD8⁻) were sorted using Fluorescence-Activated Cell Sorting (FACS) on an Aurora CS flow cytometer (Cytek Biosciences). For tonsillar tissue, the three subpopulations of follicular helper CD4⁺ T cells (T_{FH: CXCR5⁺ PD-1⁺}) were isolated: T_{FH CD69⁻ CCR7⁺}, T_{FH CD69⁺ CCR7⁺} and T_{FH CD69⁺ CCR7⁻} subsets. Naïve CD4⁺ T cells (_{TNA: CXCR5⁻ CD69⁻ CD45RO⁻ CCR7⁺}) were also sorted. For intestinal samples, resident memory CD4⁺ T cells (T_{RM: CD69⁺ CXCR5⁻ CD45RO⁺ CCR7⁻}), including T_{RM CD69⁺ CD103⁺ CD49a⁺}, T_{RM CD69⁺ CD49a⁺} and T_{RM CD69⁺ PD-1⁺} subsets, were isolated. Non-resident CD4⁺ T cells (T_{NON-RM: CD69⁻}) were also sorted. T_FH subsets from tonsillar samples and T_RM subsets from intestinal samples were pooled for reactivation assays. All subpopulations were cultured and stimulated with LRAs as previously described. HIV reactivation was assessed by measuring intracellular p24 expression via flow cytometry, using the same staining panel as for total CD4⁺ T cells. All antibodies are commercially available and validated by their respective manufacturers.

Statistical analyses

All data were analysed using GraphPad Prism 8.3.0 software (GraphPad Software, La Jolla, CA, USA), with statistical significance defined as p < 0.05. Volcano plots, generated using OMIQ software (www.omiq.ai, www.dotmatics.com), represent statistical comparisons conducted via the edgeR method, with significance determined at p < 0.05. Detailed statistical information for each experiment is provided in the respective figure legends.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data associated with this study are provided within the main text or the Supplementary Information. Source data are provided with this paper. Any other requested data related with the human tissue samples from paediatric and adult donors is restricted by ethical and privacy considerations. Access can be obtained after IRB approval of the formal data-sharing agreement in a process that can last up to 24 weeks. For questions related to the data request and usage, contact Maria J. Buzon at mariajose.buzon@vhir.org. Source data are provided with this paper.

Code availability

This study does not report original code.

References

Palmer, S., Josefsson, L. & Coffin, J. M. HIV reservoirs and the possibility of a cure for HIV infection. J. Intern. Med. 270, 550–560 (2011).
Article PubMed CAS Google Scholar
Coiras, M., López-Huertas, M. R., Pérez-Olmeda, M. & Alcamí, J. Understanding HIV-1 latency provides clues for the eradication of long-term reservoirs. Nat. Rev. Microbiol. 7, 798–812 (2009).
Article PubMed CAS Google Scholar
Gantner, P. et al. HIV rapidly targets a diverse pool of CD4 + T cells to establish productive and latent infections. Immunity 56, 653–668.e5 (2023).
Article PubMed PubMed Central CAS Google Scholar
Tokarev, A. et al. Single-Cell Profiling of Latently SIV-Infected CD4 ⁺ T Cells Directly Ex Vivo to Reveal Host Factors Supporting Reservoir Persistence. Microbiol. Spectr. 10, e0060422 (2022).
Article PubMed Google Scholar
Abrahams, M.-R. et al. The replication-competent HIV-1 latent reservoir is primarily established near the time of therapy initiation. Sci. Transl. Med. 11, (2019).
Finzi, D. et al. Identification of a Reservoir for HIV-1 in Patients on Highly Active Antiretroviral Therapy. Science (1979) 278, 1295–1300 (1997).
CAS Google Scholar
Lichterfeld, M., Gao, C. & Yu, X. G. An ordeal that does not heal: understanding barriers to a cure for HIV-1 infection. Trends Immunol. 43, 608–616 (2022).
Article PubMed PubMed Central CAS Google Scholar
Castro-Gonzalez, S., Colomer-Lluch, M. & Serra-Moreno, R. Barriers for HIV Cure: The Latent Reservoir. AIDS Res. Hum. Retroviruses 34, 739–759 (2018).
Article PubMed PubMed Central Google Scholar
Cantero-Pérez, J. et al. Resident memory T cells are a cellular reservoir for HIV in the cervical mucosa. Nat. Commun. 10, 4739 (2019).
Article ADS PubMed PubMed Central Google Scholar
Serra-Peinado, C. et al. Expression of CD20 after viral reactivation renders HIV-reservoir cells susceptible to Rituximab. Nat. Commun. 10, 3705 (2019).
Article ADS PubMed PubMed Central Google Scholar
Gálvez, C. et al. Atlas of the HIV-1 Reservoir in Peripheral CD4 T Cells of Individuals on Successful Antiretroviral Therapy. mBio 12, e0307821 (2021).
Article PubMed Google Scholar
Grau-Expósito, J. et al. A Novel Single-Cell FISH-Flow Assay Identifies Effector Memory CD4 ⁺ T cells as a Major Niche for HIV-1 Transcription in HIV-Infected Patients. mBio 8, (2017).
Fromentin, R. et al. CD4 + T Cells Expressing PD-1, TIGIT and LAG-3 Contribute to HIV Persistence during ART. PLoS Pathog. 12, e1005761 (2016).
Article PubMed PubMed Central Google Scholar
Hsiao, F. et al. Tissue memory CD4 + T cells expressing IL-7 receptor-alpha (CD127) preferentially support latent HIV-1 infection. PLoS Pathog. 16, e1008450 (2020).
Article PubMed PubMed Central Google Scholar
Neidleman, J. et al. Phenotypic analysis of the unstimulated in vivo HIV CD4 T cell reservoir. Elife 9, (2020).
Astorga-Gamaza, A. et al. Identification of HIV-reservoir cells with reduced susceptibility to antibody-dependent immune response. Elife 11, (2022).
Pieren, D. K. J., Benítez-Martínez, A. & Genescà, M. Targeting HIV persistence in the tissue. Curr. Opin. HIV AIDS 19, 69–78 (2024).
Article PubMed CAS Google Scholar
Kandathil, A. J., Sugawara, S. & Balagopal, A. Are T cells the only HIV-1 reservoir?. Retrovirology 13, 86 (2016).
Article PubMed PubMed Central Google Scholar
Chen, J. et al. The reservoir of latent HIV. Front. Cell Infect. Microbiol. 12, 945956 (2022).
Article PubMed PubMed Central CAS Google Scholar
Bosque, A. & Planelles, V. Studies of HIV-1 latency in an ex vivo model that uses primary central memory T cells. Methods 53, 54–61 (2011).
Article PubMed CAS Google Scholar
Chomont, N. et al. HIV reservoir size and persistence are driven by T cell survival and homeostatic proliferation. Nat. Med. 15, 893–900 (2009).
Article PubMed PubMed Central CAS Google Scholar
Kulpa, D. A. et al. Differentiation into an Effector Memory Phenotype Potentiates HIV-1 Latency Reversal in CD4 ⁺ T Cells. J. Virol. 93, e00969-19 (2019).
Article PubMed PubMed Central Google Scholar
Hiener, B. et al. Identification of Genetically Intact HIV-1 Proviruses in Specific CD4 + T Cells from Effectively Treated Participants. Cell Rep. 21, 813–822 (2017).
Article PubMed PubMed Central CAS Google Scholar
De Scheerder, M.-A. et al. HIV Rebound Is Predominantly Fueled by Genetically Identical Viral Expansions from Diverse Reservoirs. Cell Host Microbe 26, 347–358.e7 (2019).
Article PubMed PubMed Central Google Scholar
Buzon, M. J. et al. HIV-1 persistence in CD4 + T cells with stem cell–like properties. Nat. Med 20, 139–142 (2014).
Article PubMed PubMed Central CAS Google Scholar
Venanzi Rullo, E. et al. Persistence of an intact HIV reservoir in phenotypically naive T cells. JCI Insight 5, e133157 (2020).
Article PubMed PubMed Central Google Scholar
Chaillon, A. et al. HIV persists throughout deep tissues with repopulation from multiple anatomical sources. J. Clin. Investig. 130, 1699–1712 (2020).
Article PubMed PubMed Central CAS Google Scholar
Wong, J. K. & Yukl, S. A. Tissue reservoirs of HIV. Curr. Opin. HIV AIDS 11, 362–370 (2016).
Article PubMed PubMed Central CAS Google Scholar
Yukl, S. A. et al. Differences in HIV Burden and Immune Activation within the Gut of HIV-Positive Patients Receiving Suppressive Antiretroviral Therapy. J. Infect. Dis. 202, 1553–1561 (2010).
Article PubMed Google Scholar
Jenabian, M.-A. et al. Immune tolerance properties of the testicular tissue as a viral sanctuary site in ART-treated HIV-infected adults. AIDS 30, 2777–2786 (2016).
Article PubMed CAS Google Scholar
Fukazawa, Y. et al. B cell follicle sanctuary permits persistent productive simian immunodeficiency virus infection in elite controllers. Nat. Med. 21, 132–139 (2015).
Article PubMed PubMed Central CAS Google Scholar
Persidsky, Y. & Poluektova, L. Immune privilege and HIV-1 persistence in the CNS. Immunol. Rev. 213, 180–194 (2006).
Article PubMed CAS Google Scholar
Fletcher, C. V. et al. Persistent HIV-1 replication is associated with lower antiretroviral drug concentrations in lymphatic tissues. Proc. Natl. Acad. Sci. 111, 2307–2312 (2014).
Article ADS PubMed PubMed Central CAS Google Scholar
Rosen, E. P. et al. Antiretroviral drug exposure in lymph nodes is heterogeneous and drug dependent. J. Int AIDS Soc. 25, e25895 (2022).
Article PubMed PubMed Central CAS Google Scholar
Lamers, S. L. et al. HIV DNA Is Frequently Present within Pathologic Tissues Evaluated at Autopsy from Combined Antiretroviral Therapy-Treated Patients with Undetectable Viral Loads. J. Virol. 90, 8968–8983 (2016).
Article PubMed PubMed Central CAS Google Scholar
Dufour, C. et al. Near full-length HIV sequencing in multiple tissues collected postmortem reveals shared clonal expansions across distinct reservoirs during ART. Cell Rep. 42, 113053 (2023).
Article PubMed CAS Google Scholar
McManus, W. R. et al. HIV-1 in lymph nodes is maintained by cellular proliferation during antiretroviral therapy. J. Clin. Investig. 129, 4629–4642 (2019).
Article PubMed PubMed Central CAS Google Scholar
Wu, V. H. et al. Assessment of HIV-1 integration in tissues and subsets across infection stages. JCI Insight 5, e139783 (2020).
Article PubMed PubMed Central Google Scholar
Pantaleo, G. et al. Lymphoid organs function as major reservoirs for human immunodeficiency virus. Proc. Natl. Acad. Sci. 88, 9838–9842 (1991).
Article ADS PubMed PubMed Central CAS Google Scholar
Pantaleo, G. et al. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature 362, 355–358 (1993).
Article ADS PubMed CAS Google Scholar
Chun, T. W. et al. Persistence of HIV in Gut-Associated Lymphoid Tissue despite Long-Term Antiretroviral Therapy. J. Infect. Dis. 197, 714–720 (2008).
Article PubMed CAS Google Scholar
Poles, M. A. et al. Lack of Decay of HIV-1 in Gut-Associated Lymphoid Tissue Reservoirs in Maximally Suppressed Individuals. JAIDS J. Acquir. Immune Defic. Synd. 43, 65–68 (2006).
Article Google Scholar
Perreau, M. et al. Follicular helper T cells serve as the major CD4 T cell compartment for HIV-1 infection, replication, and production. J. Exp. Med. 210, 143–156 (2013).
Article PubMed PubMed Central CAS Google Scholar
Renault, C. et al. Th17 CD4 + T-Cell as a preferential target for HIV reservoirs. Front Immunol. 13, 822576 (2022).
Article PubMed PubMed Central CAS Google Scholar
Miles, B. & Connick, E. T FH in HIV latency and as sources of replication-competent virus. Trends Microbiol 24, 338–344 (2016).
Article PubMed PubMed Central CAS Google Scholar
Sun, W. et al. Phenotypic signatures of immune selection in HIV-1 reservoir cells. Nature 614, 309–317 (2023).
Article ADS PubMed PubMed Central CAS Google Scholar
Kumar, B. V. et al. Human tissue-resident memory T cells are defined by core transcriptional and functional signatures in lymphoid and mucosal sites. Cell Rep. 20, 2921–2934 (2017).
Article PubMed PubMed Central CAS Google Scholar
Szabo, P. A. et al. Single-cell transcriptomics of human T cells reveals tissue and activation signatures in health and disease. Nat. Commun. 10, 4706 (2019).
Article ADS PubMed PubMed Central CAS Google Scholar
Poon, M. M. L. et al. Tissue adaptation and clonal segregation of human memory T cells in barrier sites. Nat. Immunol. 24, 309–319 (2023).
Article PubMed PubMed Central CAS Google Scholar
Shiow, L. R. et al. CD69 acts downstream of interferon-α/β to inhibit S1P1 and lymphocyte egress from lymphoid organs. Nature 440, 540–544 (2006).
Article ADS PubMed CAS Google Scholar
Thome, J. J. C. et al. Spatial map of human T cell compartmentalization and maintenance over decades of life. Cell 159, 814–828 (2014).
Article PubMed PubMed Central CAS Google Scholar
Schenkel, J. M. & Masopust, D. Tissue-resident memory T Cells. Immunity 41, 886–897 (2014).
Article PubMed PubMed Central CAS Google Scholar
Park, S. L. et al. Local proliferation maintains a stable pool of tissue-resident memory T cells after antiviral recall responses. Nat. Immunol. 19, 183–191 (2018).
Article PubMed CAS Google Scholar
Wong, M. T. et al. A High-Dimensional Atlas of Human T Cell Diversity Reveals Tissue-Specific Trafficking and Cytokine Signatures. Immunity 45, 442–456 (2016).
Article PubMed CAS Google Scholar
Yokoi, T. et al. Identification of a unique subset of tissue-resident memory CD4 ⁺ T cells in Crohn’s disease. Proc. Natl. Acad. Sci. 120, e2204269120 (2023).
Article PubMed CAS Google Scholar
Reilly, E. C. et al. T _RM integrins CD103 and CD49a differentially support adherence and motility after resolution of influenza virus infection. Proc. Natl. Acad. Sci. 117, 12306–12314 (2020).
Article ADS PubMed PubMed Central CAS Google Scholar
Kim, Y., Anderson, J. L. & Lewin, S. R. Getting the “Kill” into “Shock and Kill”: Strategies to Eliminate Latent HIV. Cell Host Microbe 23, 14–26 (2018).
Article PubMed PubMed Central CAS Google Scholar
Board, N. L., Moskovljevic, M., Wu, F., Siliciano, R. F. & Siliciano, J. D. Engaging innate immunity in HIV-1 cure strategies. Nat. Rev. Immunol. 22, 499–512 (2022).
Article PubMed CAS Google Scholar
Pardons, M., Fromentin, R., Pagliuzza, A., Routy, J.-P. & Chomont, N. Latency-reversing agents induce differential responses in distinct memory CD4 T cell subsets in individuals on antiretroviral therapy. Cell Rep. 29, 2783–2795.e5 (2019).
Article PubMed PubMed Central CAS Google Scholar
Wei, D. G. et al. Histone Deacetylase Inhibitor Romidepsin Induces HIV Expression in CD4 T Cells from Patients on Suppressive Antiretroviral Therapy at Concentrations Achieved by Clinical Dosing. PLoS Pathog. 10, e1004071 (2014).
Article PubMed PubMed Central Google Scholar
Grau-Expósito, J. et al. Latency reversal agents affect differently the latent reservoir present in distinct CD4 + T subpopulations. PLoS Pathog. 15, e1007991 (2019).
Article PubMed PubMed Central Google Scholar
Covino, D. A., Desimio, M. G. & Doria, M. Impact of IL-15 and latency reversing agent combinations in the reactivation and NK cell-mediated suppression of the HIV reservoir. Sci. Rep. 12, 18567 (2022).
Article ADS PubMed PubMed Central CAS Google Scholar
Banga, R., Procopio, F. A., Cavassini, M. & Perreau, M. In Vitro Reactivation of Replication-Competent and Infectious HIV-1 by Histone Deacetylase Inhibitors. J. Virol. 90, 1858–1871 (2016).
Article PubMed PubMed Central CAS Google Scholar
Shi, J. et al. PD-1 controls follicular T Helper Cell Positioning and Function. Immunity 49, 264–274.e4 (2018).
Article PubMed PubMed Central CAS Google Scholar
Hardtke, S., Ohl, L. & Förster, R. Balanced expression of CXCR5 and CCR7 on follicular T helper cells determines their transient positioning to lymph node follicles and is essential for efficient B-cell help. Blood 106, 1924–1931 (2005).
Article PubMed CAS Google Scholar
Haynes, N. M. et al. Role of CXCR5 and CCR7 in Follicular Th Cell Positioning and Appearance of a Programmed Cell Death Gene-1High Germinal Center-Associated Subpopulation. J. Immunol. 179, 5099–5108 (2007).
Article PubMed CAS Google Scholar
Asrir, A., Aloulou, M., Gador, M., Pérals, C. & Fazilleau, N. Interconnected subsets of memory follicular helper T cells have different effector functions. Nat. Commun. 8, 847 (2017).
Article ADS PubMed PubMed Central Google Scholar
Song, W. & Craft, J. T follicular helper cell heterogeneity: Time, space, and function. Immunol. Rev. 288, 85–96 (2019).
Article PubMed PubMed Central CAS Google Scholar
Crotty, S. T follicular helper cell differentiation, function, and roles in disease. Immunity 41, 529–542 (2014).
Article PubMed PubMed Central CAS Google Scholar
Huster, K. M. et al. Selective expression of IL-7 receptor on memory T cells identifies early CD40L-dependent generation of distinct CD8 ⁺ memory T cell subsets. Proc. Natl. Acad. Sci. 101, 5610–5615 (2004).
Article ADS PubMed PubMed Central CAS Google Scholar
Robbins, S. H., Terrizzi, S. C., Sydora, B. C., Mikayama, T. & Brossay, L. Differential regulation of killer cell lectin-like receptor G1 expression on T cells. J. Immunol. 170, 5876–5885 (2003).
Article PubMed CAS Google Scholar
Voehringer, D., Koschella, M. & Pircher, H. Lack of proliferative capacity of human effector and memory T cells expressing killer cell lectinlike receptor G1 (KLRG1). Blood 100, 3698–3702 (2002).
Article PubMed CAS Google Scholar
Astorga-Gamaza, A. et al. KLRG1 expression on natural killer cells is associated with HIV persistence, and its targeting promotes the reduction of the viral reservoir. Cell Rep. Med. 4, 101202 (2023).
Article PubMed PubMed Central CAS Google Scholar
Nave, H., Gebert, A. & Pabst, R. Morphology and immunology of the human palatine tonsil. Anat. Embryol. 204, 367–373 (2001).
Article CAS Google Scholar
Mueller, S. N., Gebhardt, T., Carbone, F. R. & Heath, W. R. Memory T cell subsets, migration patterns, and tissue residence. Annu. Rev. Immunol. 31, 137–161 (2013).
Article PubMed CAS Google Scholar
Hagel, J. P. et al. Defining T cell subsets in human tonsils using ChipCytometry. J. Immunol. 206, 3073–3082 (2021).
Article PubMed PubMed Central CAS Google Scholar
Crotty, S. Follicular helper CD4 T Cells (T _FH). Annu. Rev. Immunol. 29, 621–663 (2011).
Article PubMed CAS Google Scholar
Thornhill, J. P., Fidler, S., Klenerman, P., Frater, J. & Phetsouphanh, C. The Role of CD4 + T Follicular Helper Cells in HIV infection: from the germinal center to the periphery. Front Immunol. 8, 46 (2017).
Article PubMed PubMed Central Google Scholar
Sathaliyawala, T. et al. Distribution and compartmentalization of human circulating and tissue-resident memory T Cell Subsets. Immunity 38, 187–197 (2013).
Article PubMed CAS Google Scholar
Woodward Davis, A. S. et al. The human memory T cell compartment changes across tissues of the female reproductive tract. Mucosal Immunol. 14, 862–872 (2021).
Article PubMed CAS Google Scholar
Shen, Z., vom Steeg, L. G., Patel, M. V., Rodriguez-Garcia, M. & Wira, C. R. Impact of aging on the frequency, phenotype, and function of CD4 + T cells in the human female reproductive tract. Front. Immunol. 15, 1465124 (2024).
Article PubMed PubMed Central CAS Google Scholar
Xie, G. et al. Characterization of HIV-induced remodeling reveals differences in infection susceptibility of memory CD4 + T cell subsets in vivo. Cell Rep. 35, 109038 (2021).
Article PubMed PubMed Central CAS Google Scholar
Matheson, N. J. et al. Cell Surface Proteomic Map of HIV Infection Reveals Antagonism of Amino Acid Metabolism by Vpu and Nef. Cell Host Microbe 18, 409–423 (2015).
Article PubMed PubMed Central CAS Google Scholar
Cavrois, M. et al. Mass Cytometric Analysis of HIV Entry, Replication, and Remodeling in Tissue CD4 + T Cells. Cell Rep. 20, 984–998 (2017).
Article PubMed PubMed Central CAS Google Scholar
Cohn, L. B. et al. Clonal CD4 + T cells in the HIV-1 latent reservoir display a distinct gene profile upon reactivation. Nat. Med. 24, 604–609 (2018).
Article PubMed PubMed Central CAS Google Scholar
Pardons, M. et al. Single-cell characterization and quantification of translation-competent viral reservoirs in treated and untreated HIV infection. PLoS Pathog. 15, e1007619 (2019).
Article PubMed PubMed Central CAS Google Scholar
Baxter, A. E. et al. Single-Cell Characterization of Viral Translation-Competent Reservoirs in HIV-Infected Individuals. Cell Host Microbe 20, 368–380 (2016).
Article PubMed PubMed Central CAS Google Scholar
Bui, J. K. et al. Ex vivo activation of CD4 + T-cells from donors on suppressive ART can lead to sustained production of infectious HIV-1 from a subset of infected cells. PLoS Pathog. 13, e1006230 (2017).
Article PubMed PubMed Central Google Scholar
Jubel, J. M., Barbati, Z. R., Burger, C., Wirtz, D. C. & Schildberg, F. A. The Role of PD-1 in Acute and Chronic Infection. Front. Immunol. 11, 487 (2020).
Article PubMed PubMed Central CAS Google Scholar
Grivel, J.-C. & Margolis, L. Use of human tissue explants to study human infectious agents. Nat. Protoc. 4, 256–269 (2009).
Article PubMed PubMed Central CAS Google Scholar
Cantero, J. & Genescà, M. Maximizing the immunological output of the cervicovaginal explant model. J. Immunol. Methods 460, 26–35 (2018).
Article PubMed CAS Google Scholar
Grossman, Z., Meier-Schellersheim, M., Paul, W. E. & Picker, L. J. Pathogenesis of HIV infection: what the virus spares is as important as what it destroys. Nat. Med. 12, 289–295 (2006).
Article PubMed CAS Google Scholar
Veazey, R. S. et al. Gastrointestinal Tract as a Major Site of CD4 ⁺ T Cell Depletion and Viral Replication in SIV Infection. Science (1979) 280, 427–431 (1998).
CAS Google Scholar
Brenchley, J. M. et al. CD4 + T Cell Depletion during all Stages of HIV Disease Occurs Predominantly in the Gastrointestinal Tract. J. Exp. Med. 200, 749–759 (2004).
Article PubMed PubMed Central CAS Google Scholar
Suanzes, P. et al. Impact of very early antiretroviral therapy during acute HIV infection on long-term immunovirological outcomes. Int. J. Infect. Dis. 136, 100–106 (2023).
Article PubMed CAS Google Scholar
Le, T. et al. Enhanced CD4 + T-Cell Recovery with Earlier HIV-1 Antiretroviral Therapy. N. Engl. J. Med. 368, 218–230 (2013).
Article ADS PubMed PubMed Central CAS Google Scholar
Schuetz, A. et al. Initiation of ART during Early Acute HIV Infection Preserves Mucosal Th17 Function and Reverses HIV-Related Immune Activation. PLoS Pathog. 10, e1004543 (2014).
Article PubMed PubMed Central Google Scholar
Peluso, M. J. et al. Differential decay of intact and defective proviral DNA in HIV-1–infected individuals on suppressive antiretroviral therapy. JCI Insight 5, e132997 (2020).
Article PubMed PubMed Central Google Scholar
Gandhi, R. T. et al. Selective Decay of Intact HIV-1 Proviral DNA on Antiretroviral Therapy. J. Infect. Dis. 223, 225–233 (2021).
Article PubMed CAS Google Scholar
Palmer, S. et al. Low-level viremia persists for at least 7 years in patients on suppressive antiretroviral therapy. Proc. Natl. Acad. Sci. USA 105, 3879–3884 (2008).
Article ADS PubMed PubMed Central CAS Google Scholar
Wong, J. K. et al. Reduction of HIV-1 in blood and lymph nodes following potent antiretroviral therapy and the virologic correlates of treatment failure. Proc. Natl. Acad. Sci. 94, 12574–12579 (1997).
Article ADS PubMed PubMed Central CAS Google Scholar
Yukl, S. A. et al. The Distribution of HIV DNA and RNA in Cell Subsets Differs in Gut and Blood of HIV-Positive Patients on ART: Implications for Viral Persistence. J. Infect. Dis. 208, 1212–1220 (2013).
Article PubMed PubMed Central CAS Google Scholar
Belmonte, L. et al. The intestinal mucosa as a reservoir of HIV-1 infection after successful HAART. AIDS 21, 2106–2108 (2007).
Article PubMed Google Scholar
Asowata, O. E. et al. Irreversible depletion of intestinal CD4 + T cells is associated with T cell activation during chronic HIV infection. JCI Insight 6, e146162 (2021).
Article PubMed PubMed Central Google Scholar
Bergler, W., Adam, S., Gross, H. J., Hörmann, K. & Schwartz-Albiez, R. Age-dependent altered proportions in subpopulations of tonsillar lymphocytes. Clin. Exp. Immunol. 116, 9–18 (1999).
Article PubMed PubMed Central CAS Google Scholar
Saluzzo, S. et al. Delayed antiretroviral therapy in HIV-infected individuals leads to irreversible depletion of skin- and mucosa-resident memory T cells. Immunity 54, 2842–2858.e5 (2021).
Article PubMed CAS Google Scholar
Wang, X. et al. Distinct Expression Patterns of CD69 in Mucosal and Systemic Lymphoid Tissues in Primary SIV Infection of Rhesus Macaques. PLoS One 6, e27207 (2011).
Article ADS PubMed PubMed Central CAS Google Scholar
Reuschl, A.-K. et al. HIV-1 Vpr drives a tissue residency-like phenotype during selective infection of resting memory T cells. Cell Rep. 39, 110650 (2022).
Article PubMed PubMed Central CAS Google Scholar
Banga, R. et al. PD-1+ and follicular helper T cells are responsible for persistent HIV-1 transcription in treated aviremic individuals. Nat. Med. 22, 754–761 (2016).
Article PubMed CAS Google Scholar
Tsai, P. et al. In vivo analysis of the effect of panobinostat on cell-associated HIV RNA and DNA levels and latent HIV infection. Retrovirology 13, 36 (2016).
Article PubMed PubMed Central Google Scholar
Søgaard, O. S. et al. The Depsipeptide Romidepsin Reverses HIV-1 Latency In Vivo. PLoS Pathog. 11, e1005142 (2015).
Article PubMed PubMed Central Google Scholar
Rasmussen, T. A. et al. Panobinostat, a histone deacetylase inhibitor, for latent-virus reactivation in HIV-infected patients on suppressive antiretroviral therapy: a phase 1/2, single group, clinical trial. Lancet HIV 1, e13–e21 (2014).
Article PubMed Google Scholar
McBrien, J. B. et al. Robust and persistent reactivation of SIV and HIV by N-803 and depletion of CD8+ cells. Nature 578, 154–159 (2020).
Article ADS PubMed PubMed Central CAS Google Scholar
Jones, R. B. et al. A subset of latency-reversing agents expose HIV-Infected Resting CD4 + T-Cells to Recognition by Cytotoxic T-Lymphocytes. PLoS Pathog. 12, e1005545 (2016).
Article PubMed PubMed Central Google Scholar
Miller, J. S. et al. Safety and virologic impact of the IL-15 superagonist N-803 in people living with HIV: a phase 1 trial. Nat. Med. 28, 392–400 (2022).
Article PubMed CAS Google Scholar
Chehimi, J. et al. IL-15 enhances immune functions during HIV infection. J. Immunol. 158, 5978–5987 (1997).
Article PubMed CAS Google Scholar
Garrido, C. et al. Interleukin-15-Stimulated natural killer cells clear HIV-1-infected cells following latency reversal ex vivo.J. Virol 92, (2018).
Li, Y., Gao, H., Clark, K. M. & Shan, L. IL-15 enhances HIV-1 infection by promoting survival and proliferation of CCR5 + CD4 + T cells. JCI Insight 8, e166292 (2023).
Article PubMed PubMed Central Google Scholar
Manganaro, L. et al. IL-15 regulates susceptibility of CD4 ⁺ T cells to HIV infection. Proc. Natl. Acad. Sci. 115, E9659–E9667 (2018).
Article PubMed PubMed Central CAS Google Scholar
Schenkel, J. M. et al. IL-15 independent maintenance of tissue resident and boosted effector memory CD8 T cells. J. Immunol. 196, 3920–3926 (2016).
Geginat, J., Sallusto, F. & Lanzavecchia, A. Cytokine-driven Proliferation and Differentiation of Human Naive, Central Memory, and Effector Memory CD4 + T Cells. J. Exp. Med. 194, 1711–1720 (2001).
Article PubMed PubMed Central CAS Google Scholar
Antonucci, J. M. et al. SAMHD1 Impairs HIV-1 gene expression and negatively modulates reactivation of viral latency in CD4 + T Cells. J. Virol. 92, e00292-18 (2018).
Article PubMed PubMed Central Google Scholar
Ruffin, N. et al. Low SAMHD1 expression following T-cell activation and proliferation renders CD4 + T cells susceptible to HIV-1. AIDS 29, 519–530 (2015).
Article PubMed CAS Google Scholar
Nixon, C. C. et al. Systemic HIV and SIV latency reversal via non-canonical NF-κB signalling in vivo. Nature 578, 160–165 (2020).
Article ADS PubMed PubMed Central CAS Google Scholar
Mavigner, M. et al. CD8 lymphocyte depletion enhances the latency reversal activity of the SMAC Mimetic AZD5582 in ART-suppressed simian immunodeficiency virus-infected Rhesus Macaques. J. Virol. 95, e01429-20 (2021).
Article PubMed PubMed Central Google Scholar
Bullen, C. K., Laird, G. M., Durand, C. M., Siliciano, J. D. & Siliciano, R. F. New ex vivo approaches distinguish effective and ineffective single agents for reversing HIV-1 latency in vivo. Nat. Med. 20, 425–429 (2014).
Article PubMed PubMed Central CAS Google Scholar
Laird, G. M. et al. Ex vivo analysis identifies effective HIV-1 latency–reversing drug combinations. J. Clin. Investig. 125, 1901–1912 (2015).
Article PubMed PubMed Central Google Scholar
Perelson, A. S., Neumann, A. U., Markowitz, M., Leonard, J. M. & Ho, D. D. HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science (1979) 271, 1582–1586 (1996).
CAS Google Scholar
Wei, X. et al. Viral dynamics in human immunodeficiency virus type 1 infection. Nature 373, 117–122 (1995).
Article ADS PubMed CAS Google Scholar
Telwatte, S. et al. Gut and blood differ in constitutive blocks to HIV transcription, suggesting tissue-specific differences in the mechanisms that govern HIV latency. PLoS Pathog. 14, e1007357 (2018).
Article PubMed PubMed Central Google Scholar
Elliott, J. H. et al. Activation of HIV Transcription with Short-Course Vorinostat in HIV-Infected Patients on Suppressive Antiretroviral Therapy. PLoS Pathog. 10, e1004473 (2014).
Article PubMed PubMed Central Google Scholar
Lenasi, T., Contreras, X. & Peterlin, B. M. Transcriptional interference antagonizes proviral gene expression to promote HIV latency. Cell Host Microbe 4, 123–133 (2008).
Article PubMed PubMed Central CAS Google Scholar
Pearson, R. et al. Epigenetic Silencing of Human Immunodeficiency Virus (HIV) Transcription by Formation of Restrictive Chromatin Structures at the Viral Long Terminal Repeat Drives the Progressive Entry of HIV into Latency. J. Virol. 82, 12291–12303 (2008).
Article PubMed PubMed Central CAS Google Scholar
Han, Y. et al. Orientation-Dependent Regulation of Integrated HIV-1 Expression by Host Gene Transcriptional Readthrough. Cell Host Microbe 4, 134–146 (2008).
Article PubMed PubMed Central CAS Google Scholar
Schröder, A. R. W. et al. HIV-1 Integration in the Human Genome Favors Active Genes and Local Hotspots. Cell 110, 521–529 (2002).
Article PubMed Google Scholar
Martin, A. R. et al. Similar Frequency and Inducibility of Intact Human Immunodeficiency Virus-1 Proviruses in Blood and Lymph Nodes. J. Infect. Dis. 224, 258–268 (2021).
Article PubMed CAS Google Scholar
Pardons, M. et al. Blood and tissue HIV-1 reservoirs display plasticity and lack of compartmentalization in virally suppressed people. Nat. Commun. 16, 2173 (2025).
Article ADS PubMed PubMed Central CAS Google Scholar
Bruner, K. M. et al. A quantitative approach for measuring the reservoir of latent HIV-1 proviruses. Nature 566, 120–125 (2019).
Article ADS PubMed PubMed Central CAS Google Scholar

Download references

Acknowledgements

This study was funded by “la Caixa” Foundation (ID 100010434) under the project code LCF/PR/HR20/00218. This study also received funding from the Agencia Estatal de Investigación project PID2021-123321OB-I00 and PID2024-155881OB-100 supported by MCIN/AEI/10.13039/501100011033/FEDER, UE; The Spanish “Ministerio de Economía y Competitividad, Instituto de Salud Carlos III” (ISCIII, PI20/00160); and the Gilead fellowships GLD21-00049, GLD22/00152 and GLD24/00014. MJB is supported by the Miguel Servet programme of the Spanish Health Institute Carlos III (CP17/00179 and CPII22/00005), and AGC is a recipient of a PhD fellowship from “la Caixa” Foundation (LCF/BQ/DI20/11780014). The funders had no role in the design of the study, data collection or analysis, decision to publish, or manuscript preparation. We acknowledge María Lázaro and Ruth Peña (JGP research group) for their technical assistance in the production of the HIV_BAL viral stock.

Author information

Authors and Affiliations

Infectious Diseases Department, Hospital Universitari Vall d’Hebron, Vall d’Hebron Institut de Recerca (VHIR), Universitat Autònoma de Barcelona, Barcelona, Spain
A. Gallego-Cortés, N. Sánchez-Gaona, C. Mancebo-Pérez, O. Ruiz-Isant, A. Benítez-Martínez, V. Falcó, M. Genescà & MJ. Buzon
Department of Pathology, Hospital Universitari Vall d’Hebron, Barcelona, Spain
S. Landolfi & J. Castellví
Department of Otorhinolaryngology, Hospital Universitari Vall d’Hebron, Barcelona, Spain
F. Pumarola, N. Ortiz, I. Llano & J. Lorente
Department of Obstetrics and Gynecology, Hospital Universitari Vall d’Hebron, Barcelona, Spain
L. Mañalich-Barrachina
IrsiCaixa AIDS Research Institute, Germans Trias i Pujol Research Institute (IGTP), Badalona, Spain
JG. Prado
CIBER Enfermedades Infecciosas (CIBERINFEC), Instituto de Salud Carlos III, Madrid, Spain
JG. Prado
Centro de Biología Molecular Severo Ochoa, CSIC-UAM, Madrid, Spain
E. Martín-Gayo
Immunology Unit, Hospital Universitario de La Princesa and Instituto de Investigación Sanitaria Princesa, Madrid, Spain
E. Martín-Gayo

Authors

A. Gallego-Cortés
View author publications
Search author on:PubMed Google Scholar
N. Sánchez-Gaona
View author publications
Search author on:PubMed Google Scholar
C. Mancebo-Pérez
View author publications
Search author on:PubMed Google Scholar
O. Ruiz-Isant
View author publications
Search author on:PubMed Google Scholar
A. Benítez-Martínez
View author publications
Search author on:PubMed Google Scholar
S. Landolfi
View author publications
Search author on:PubMed Google Scholar
J. Castellví
View author publications
Search author on:PubMed Google Scholar
F. Pumarola
View author publications
Search author on:PubMed Google Scholar
N. Ortiz
View author publications
Search author on:PubMed Google Scholar
I. Llano
View author publications
Search author on:PubMed Google Scholar
J. Lorente
View author publications
Search author on:PubMed Google Scholar
L. Mañalich-Barrachina
View author publications
Search author on:PubMed Google Scholar
JG. Prado
View author publications
Search author on:PubMed Google Scholar
E. Martín-Gayo
View author publications
Search author on:PubMed Google Scholar
V. Falcó
View author publications
Search author on:PubMed Google Scholar
M. Genescà
View author publications
Search author on:PubMed Google Scholar
MJ. Buzon
View author publications
Search author on:PubMed Google Scholar

Contributions

A.G.C. and M.J.B. designed, directed, and interpreted experiments. A.G.C., N.S.G., C.M.P., O.R.I., and A.B.M. performed the experiments and analysed the data. M.G. and E.M.G. contributed to the design and interpretation of the experiments. S.L., J.C., F.P., N.O., I.L., J.L., V.F. and L.M.B. were responsible for recruitment, specimen handling and storage, and the collection of related clinical data. J.G.P. produced and provided the HIV_BAL viral stock. A.G.C. and M.J.B. wrote the initial manuscript, and all authors contributed to its editing.

Corresponding author

Correspondence to MJ. Buzon.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Talia Swartz and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review file

Reporting Summary

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Gallego-Cortés, A., Sánchez-Gaona, N., Mancebo-Pérez, C. et al. Identification of inducible HIV reservoirs in tonsillar, intestinal and cervical tissue models of HIV latency. Nat Commun 16, 10353 (2025). https://doi.org/10.1038/s41467-025-65288-9

Download citation

Received: 12 February 2025
Accepted: 08 October 2025
Published: 24 November 2025
Version of record: 24 November 2025
DOI: https://doi.org/10.1038/s41467-025-65288-9

Subjects

Abstract

Similar content being viewed by others

Profound phenotypic and epigenetic heterogeneity of the HIV-1-infected CD4+ T cell reservoir

HIV infection reprogrammes CD4+ T cells for quiescence and entry into proviral latency

Phenotypic signatures of immune selection in HIV-1 reservoir cells

Introduction

Results

Development of explant models for the study of HIV persistence in tissues

Distinct representation of CD4+ T-cell subpopulations across tissues and differential contribution to HIV infection in tonsillar and intestinal compartments

Tonsillar and intestinal inducible HIV reservoirs are constituted of distinct CD4+ T-cell subpopulations

Current LRAs demonstrated differing potency and breath of viral reactivation between tissue reservoirs

Heterogeneous responses of tonsillar and intestinal CD4+ T-cell subpopulations to LRAs

Validation of LRA efficacy in isolated CD4⁺ T-cell subpopulations

Discussion

Methods

Patient samples and ethics statement

Tissue explant models of productive and latent HIV infection

Digestion of tissue blocks

Longitudinal phenotyping of key immune populations and HIV infection assessment by flow cytometry

Quantification of HIV DNA in isolated CD4+ T cells by Intact Proviral DNA Assay (IPDA) and Quantitative PCR (qPCR)

Ex vivo reactivation with LRAs of HIV-infected and ART-treated CD4+ T cells

CD4+ T-cell subsets phenotyping and detection of HIV-1 p24 protein by flow cytometry

Detection of HIV-1 p24 in CD4+ T-cell culture supernatants via Single Molecule Array (SIMOA) assay

Sorting and viral reactivation of tonsillar and intestinal CD4+ T-cell subpopulations

Statistical analyses

Reporting summary

Data availability

Code availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Supplementary information

Supplementary Information

Peer Review file

Reporting Summary

Source data

Source data

Rights and permissions

About this article

Cite this article

Share this article

Search

Quick links

Profound phenotypic and epigenetic heterogeneity of the HIV-1-infected CD4⁺ T cell reservoir

Distinct representation of CD4⁺ T-cell subpopulations across tissues and differential contribution to HIV infection in tonsillar and intestinal compartments

Tonsillar and intestinal inducible HIV reservoirs are constituted of distinct CD4⁺ T-cell subpopulations

Heterogeneous responses of tonsillar and intestinal CD4⁺ T-cell subpopulations to LRAs

Quantification of HIV DNA in isolated CD4⁺ T cells by Intact Proviral DNA Assay (IPDA) and Quantitative PCR (qPCR)

Ex vivo reactivation with LRAs of HIV-infected and ART-treated CD4⁺ T cells

CD4⁺ T-cell subsets phenotyping and detection of HIV-1 p24 protein by flow cytometry

Detection of HIV-1 p24 in CD4⁺ T-cell culture supernatants via Single Molecule Array (SIMOA) assay

Sorting and viral reactivation of tonsillar and intestinal CD4⁺ T-cell subpopulations