Dysregulation of IL-17/IL-22 Effector Functions in Blood and Gut Mucosal Gamma Delta T Cells Correlates With Increase in Circulating Leaky Gut and Inflammatory Markers During cART-Treated Chronic SIV Infection in Macaques

HIV-associated inflammation has been implicated in the premature aging and increased risk of age-associated comorbidities in cART-treated individuals. However, the immune mechanisms underlying the chronic inflammatory state of cART-suppressed HIV infection remain unclear. Here, we investigated the role of γδT cells, a group of innate IL-17 producing T lymphocytes, in the development of systemic inflammation and leaky gut phenotype during cART-suppressed SIV infection of macaques. Plasma levels of inflammatory mediators, intestinal epithelial barrier disruption (IEBD) and microbial translocation (MT) biomarkers, and Th1/Th17-type cytokine functions were longitudinally assessed in blood and gut mucosa of SIV-infected, cART-suppressed macaques. Among the various gut mucosal IL-17/IL-22-producing T lymphocyte subsets including Th17, γδT, CD161+ CD8+ T, and MAIT cells, a specific decline in the Vδ2 subset of γδT cells and impaired IL-17/IL-22 production in γδT cells significantly correlated with the subsequent increase in plasma IEBD/MT markers (IFABP, LPS-binding protein, and sCD14) and pro-inflammatory cytokines (IL-6, IL-1β, IP10, etc.) despite continued viral suppression during long-term cART. Further, the plasma inflammatory cytokine signature during long-term cART was distinct from acute SIV infection and resembled the inflammatory cytokine profile of uninfected aging (inflammaging) macaques. Overall, our data suggest that during cART-suppressed chronic SIV infection, dysregulation of IL-17/IL-22 cytokine effector functions and decline of Vδ2 γδT cell subsets may contribute to gut epithelial barrier disruption and development of a distinct plasma inflammatory signature characteristic of inflammaging. Our results advance the current understanding of the impact of chronic HIV/SIV infection on γδT cell functions and demonstrate that in the setting of long-term cART, the loss of epithelial barrier-protective functions of Vδ2 T cells and ensuing IEBD/MT occurs before the hallmark expansion of Vδ1 subsets and skewed Vδ2/Vδ1 ratio. Thus, our work suggests that novel therapeutic approaches toward restoring IL-17/IL-22 cytokine functions of intestinal Vδ2 T cells may be beneficial in preserving gut epithelial barrier function and reducing chronic inflammation in HIV-infected individuals.


INTRODUCTION
The discovery of efficient, well-tolerated combinational antiretroviral therapy (cART) regimens has remarkably reduced the rates of HIV-associated mortality and transformed it into a chronic, manageable disease that requires life-long treatment. However, people living with HIV (PLWH) still remain at unusually high risk of age-associated diseases including neurocognitive, metabolic, cardiovascular disorders, and other non-AIDS-defining comorbidities (1). The pathogenesis of ageassociated diseases is complex but there is growing consensus on the substantial role of HIV-associated chronic inflammation and immune activation in the premature onset of immunosenescence and early aging in PLWH despite effective viral suppression (2)(3)(4). A similar process characterized by increased pro-inflammatory mediators leading to a chronic inflammatory state, termed "inflammaging", is a significant risk factor for morbidity and mortality in the elderly people, even those without HIV infection (5). As more PLWH are transitioning to the middleage and older age-groups, it is important to develop a detailed understanding of the immune mechanisms underlying HIVassociated inflammaging and determine the immune mechanisms that may synergize with aging and may accelerate this process.
Aberrant immune activation and chronic inflammation during HIV/SIV infection is strongly associated with loss of Th17-type mucosal immune functions and intestinal epithelial barrier damage (IEBD) resulting in intestinal permeability (leaky gut) and microbial translocation (MT) (6)(7)(8)(9). This leaky gut-associated chronic inflammation persists even with long-term effective cART and predicts mortality and incidence of age-related co-morbidities (e.g. neurocognitive, metabolic, cardiovascular disorders) (10)(11)(12). Accordingly, an inflammaging phenotype comprising of IEBD, MT, and inflammatory biomarkers has been described in both HIV-negative older subjects and PLWH (8). Indeed, we have recently demonstrated a similar inflammaging phenotype in SIVnegative aging rhesus macaques with elevated plasma levels of IEBD, MT, and inflammatory markers, which is associated with significant impairment in production of the cytokines IL-17 and IL-22 by classical and innate Th17-type immune cells (13). Gamma delta (gd) T cells are an important innate source of IL-17 and IL-22 cytokines and are key players in gut barrier functions, MT, and immune activation during HIV infection (14), and may demonstrate both inflammatory (15,16) and/or immunoregulatory potential (17,18). HIV infection results in changes in circulating gd T cell subsets with an increase in Vd1 and depletion of Vd2 T cells that may persist with viral suppression (19,20). Similar expansion in Vd1 T cells during chronic untreated SIV infection has been associated with MT (21). However, the dynamic role of gut mucosal gd T cells in IEBD, MT and persistent inflammation through the course of treated HIV infection remains unclear.
Here we assessed the development of systemic inflammation during chronic SIV infection suppressed with cART in the rhesus macaque model of treated HIV infection and evaluated longitudinal changes in systemic and mucosal gd T cell functions. We demonstrate that following early resolution of circulating inflammatory markers by effective viral suppression with cART, a plasma inflammatory and leaky gut phenotype similar to inflammaging re-emerged during later stages of chronic SIV infection despite continued treatment and viral suppression. The development of this phenotype was preceded by a significant loss of systemic and gut mucosal IL-17/IL-22 cytokine producing functions of gdT cells along with a significant decline in Vd2 cells and subsequent increase in Vd1 cells resulting in skewing of the Vd2/Vd1 subset ratio during long-term suppressive ART. These data are the first to demonstrate that dysregulation of gdT cells, particularly Vd2 T cells in the gut mucosa, is likely a cause rather than an effect of leaky gut and systemic inflammation during chronic treated SIV infection.

Ethics Statement
Animals in this study were housed at the Tulane National Primate Research Center (TNPRC), accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International. The study was approved by the TNPRC Institutional Animal Care and Use Committee (IACUC) and was conducted under the standards of the US National Institutes of Health Guide for the Care and Use of Laboratory Animals. Following SIV infection, animals were housed in Animal Biosafety Level 2 indoor housing. All animal procedures including virus administration, sample collection, and euthanasia were carried out under the direction of TNPRC veterinarians.

Animals, Viral Inoculation, and cART
Six healthy female Indian origin rhesus macaques ranging in age from 5-10 years old and seronegative for SIV, HIV-2, STLV-1 (Simian T Leukemia Virus type-1), SRV-1 (type D retrovirus), and herpes-B viruses were used in this study. MHC-1 genotyping for exclusion of the common Mamu alleles Mamu-A*01/-A*02 and Mamu-B*08/-B*17 was performed by sequence-specific priming PCR. Animals were infected with 2500 TCID50 SIV mac 251 via intrarectal (IR) route using the pathogenic SIV challenge stocks obtained from the Preclinical Research and Development Branch of Vaccine and Prevention Research Program, Division of AIDS, NIAID. cART consisted of daily subcutaneous injection of 5.1 mg/kg Tenofovir Disoproxil Fumarate (TDF), 30 mg/kg Emtricitabine (FTC) and 2.5 mg/kg Dolutegravir (DTG) in a solution containing 15% (v/v) kleptose at pH 4.2, as previously described (22).

Sample Collection and Processing
Blood samples in EDTA vacutainer tubes (Sarstedt Inc Newton, NC) were taken for a complete blood count and routine chemical analysis, and centrifuged within 1 h of phlebotomy for plasma separation. Processing of blood and rectal (RB) biopsies were performed as previously described for isolation of PBMC and lamina propria lymphocytes (23). Plasma viral load quantification was performed using Roche High Pure Viral RNA Kit (Catalog #11858882001) as earlier described (24).

Plasma Markers of Inflammation, Microbial Translocation, and Intestinal Damage
Frozen plasma samples were thawed and cleared using Ultrafree Centrifugal Filters (Millipore, Billerica, MA). The filtered plasma samples were used for simultaneous quantification of cytokines, chemokines, and growth factors using the multiplexed-bead assay Non-Human Primate Cytokine & Chemokine & Growth Factor 37-plex ProcartaPlex (Invitrogen, Life Technologies), following manufacturers' instructions. Data were acquired with a Bio-Plex 200 analyzer (Bio-Rad, Hercules, CA) and analyzed using Bio-Plex Manager software v6.1 (BioRad). For the analysis of markers of leaky gut and microbial translocation, plasma IFABP and LBP were quantified using the commercially available Monkey IFABP/FABP2 and LBP ELISA kits (MyBioSource, San Diego, CA). Commercially available ELISA kit for Human sCD14 (R&D Systems, Minneapolis, MN) was used according to the manufacturer's protocols with 1:200 diluted plasma samples. All tests were performed according to the manufacturer's guidelines. The assays were performed in duplicate, and data were analyzed using Gen 5 software (BioTek).

Intracellular Cytokine Staining
Freshly isolated PBMCs and rectal biopsy lamina propria lymphocytes were resuspended at a concentration of 1 × 10 6 cells per ml in RPMI 1640 medium supplemented with 10% FBS, 100 IU ml − 1 penicillin, and 100 mg ml − 1 streptomycin. Stimulations were conducted for 16 h at 37°C in the presence of phorbol myristate acetate (PMA; 10 ng ml −1 ), ionomycin (Sigma-Aldrich 1 mg ml −1 ), in the presence of brefeldin A (5 mg/ml, Sigma-Aldrich). After 16 h, cells were washed once with PBS to remove stimuli and stained with surface markers for CD45, CD3, CD4, CD8, TCR gd, TCR Vd1, or TCR Vd2, and CD161 for 30 min at room temperature. Cells were then fixed with cytofix/cytoperm (BD Pharmingen, San Diego, CA), washed, and stained intracellularly with antibodies specific for CD69, IL-17, IL-22, TNF-a, and IFN-g for 30 min at room temperature. Following staining, cells were washed and fixed with PBS containing 1% paraformaldehyde prior to acquiring on BD Fortessa cytometer.

Statistical Analyses
All statistical analysis was performed using GraphPad Prism Software (Version 8.4.3). Data were analyzed by one-way analysis of variance (ANOVA) with multiple comparisons, one-way ANOVA with a test for linear trend, or two-way ANOVA with repeated measures. Tukey's and Dunnett's post hoc tests were used for multiple comparisons. All correlations were computed using non-parametric Spearman rank correlation test. P values of 0.05 or lower were considered significant, * p<0.05, * * p<0.01, * * * p<0.001, * * * * p<0.0001. Polyfunctional responses were compared using SPICE 6 software (25).

Early Resolution of SIV-Induced Increase in Inflammatory Cytokines and IEBD Biomarkers Following Viral Suppression With cART
Six adult rhesus macaques were infected via intra-rectal route with SIVmac251 to represent mucosal HIV infection. Following establishment of set-point viremia, all animals were treated daily with a three-drug antiretroviral (cART) regimen, Tenofovir Disoproxil Fumarate (TDF), Emtricitabine (FTC) and Dolutegravir (DTG) (as indicated by the blue shading in Figure 1A). As expected, plasma viremia peaked at 2 weeks of infection to an average 7.2 log 10 copies/mL and by 4 weeks had reduced 2 logs and settled at~5.8 log 10 copies/mL ( Figure 1B). Stable suppression of viremia below the limit of detection (LOD) of the assay (83 viral copy Eq/ml) was achieved in all 6 animals 8 weeks from cART initiation (18 weeks post-SIV infection). To assess the impact of SIV infection on systemic inflammation, plasma concentrations of inflammatory and IEBD biomarkers were measured at pre-infection (2-4 weeks before SIV infection), around peak viremia (1-2 weeks post-SIV), during early cART (1-6 months of cART), and late chronic SIV/cART phase (7-12 months of cART).
In agreement with previous work (26)(27)(28)(29), acute SIV infection resulted in a surge in plasma cytokines, compared to baseline levels ( Figure 2). Of the 40 analytes measured, all animals displayed significantly elevated levels of pro-inflammatory cytokines IL-1b, IL-18, IL-1Ra, MIG, ITAC, MIP-1b, G-CSF, and CXCL13 ( Figure  2A). Further, in accordance with several studies demonstrating compromised gastrointestinal barrier integrity early during HIV/ SIV infections (30)(31)(32), the study animals displayed a significant increase in the surrogate markers of IEBD and MT IFABP, and LBP (but not sCD14) at 1 month post-SIV infection ( Figure 2B). This coincided with a significant decline in peripheral CD4 T lymphocytes frequencies ( Figure 2C). After an initial increase in overall T lymphocyte numbers in the 1st week of infection followed by decreased CD4 T lymphocytes at 1-month post-SIV infection, the T lymphocytes resolved to baseline by the 1-month time-point, which was maintained it throughout set-point viremia and treatment phase (Supplementary Figure 1). Effective viral suppression with cART resolved the plasma inflammatory cytokines/chemokines and leaky gut biomarkers by 3 months of cART (Figures 2A, B).

Recurrence of Systemic Inflammation and IEBD Biomarkers During the Chronic Phase of cART-Suppressed SIV Infection
Despite continued treatment with cART ( Figure 1B) multiple plasma inflammatory and IEBD markers re-emerged around 8-10 months of cART in all six macaques (Figure 3), suggesting the development of an inflammatory phenotype during chronic SIV infection that is independent of suppression of viremia and recovery of circulating CD4 T cell frequencies to baseline levels. Interestingly, at 10-12 months of cART, plasma levels of IL-1b (p = 0.007 for acute vs 10 month; p=0.001 for 12 month), GCSF (p=0.003 for acute vs 10 month; p = 0.008 for 12 month), IL-1Ra (p=0.04 for acute vs 10 month; p=0.013 for 12 month), and IL-18 (p = 0.03 for acute vs 12 month) were observed at higher levels than in the acute, pre-treatment phase ( Figure 3). Notably, additional inflammatory cytokines including GM-CSF, MCP-1, IP-10, IL-6, IFN-g, IL-12, and TNF-a were elevated only during the chronic phase following 8 months of cART ( Figure 3), indicating that the plasma inflammatory cytokine signature during long-term controlled SIV infection is distinct from the cytokine storm of acute SIV infection. The Th2 cytokine, IL-4, was also significantly elevated at the 10-and 12-month time-point. Since the inflammatory mediators tracked with markers of IEBD and MT during acute SIV infection before treatment, plasma levels of IFABP, LBP, and sCD14 were also assessed longitudinally in parallel with cytokines/chemokines. Increased levels of IFABP and LBP were observed at 8-11 months post-cART ( Figure 4). Further, consistent with several previous studies (33)(34)(35), we observed a surge in plasma sCD14 levels from 7 month onwards that were significantly higher than baseline (p=0.0018 at 8 month; p=0.03 at 11 month cART) as well as acute SIV/short-term antiretroviral treatment levels (p=0.003 at 8 month; p=0.047 at 11 month cART). The absence of any opportunistic infections or significant fluctuation in body temperature and weight of the animals (Supplementary Figure 5) suggested that the increase in plasma inflammatory markers was likely due to intrinsic changes in the host immune cell functions.    We recently reported that similar to humans (8), inflammaging phenotype in older SIV-naïve macaques is associated with a signature of increased circulating inflammatory cytokines and elevations in biomarkers of leaky-gut including IFABP, LBP, and sCD14 (13). Further analysis in this cross-sectional data to visualize the pattern of IFABP, LBP, and sCD14 expression (represented as a heat map for each individual across a row) revealed a gradient towards higher levels with increased age ( Figure 5A; vertical columns) that was more pronounced from age 17 years and up. Likewise, in this longitudinal study of SIVinfected, cART suppressed young macaques, a strikingly similar pattern of increase in IFABP, LBP, and sCD14 developed during the chronic phase of infection ( Figure 5B; vertical columns of the heat map). It should be noted that since the older SIV-naïve macaques are from a cross-sectional study, any contribution of differences in diet or social environment cannot be fully accounted for in this comparison. Nonetheless, it is noteworthy that IFABP and LBP that were maintained at baseline levels from 3-7 months of cART, increased significantly around 8 months and stayed elevated till 1 year of cART ( Figures 4 and 5B). The increase in sCD14 was earlier and did not always track with IFABP and LBP within individuals in both the cross-sectional data and the longitudinal cohort ( Figures 5A, B). Moreover, there was no significant correlation between sCD14 and the IEBD/MT markers IFABP and LBP (Supplementary Figure 2A), suggesting that factors besides MT may contribute to the increased secretion of sCD14, and that it likely is a more sensitive biomarker of inflammation. However, as anticipated, there was a highly significant correlation between IFABP and LBP (Supplementary Figure 2B) associated with the increase in inflammaging phenotype during the chronic phase of treated SIV infection.   (23,(37)(38)(39). During early acute SIV infection in our study, peripheral blood gd T cells displayed a very significant increase in frequency (p<0.0001) and absolute numbers (p=0.003) that returned to baseline by 1 month post-infection ( Figure 6). However, later during chronic phase of infection, gd T cell numbers declined to significantly low levels at the 7-9-month cART time-points (p=0.046 and 0.01 respectively; Figure 6). In agreement with previous studies, a persistent significant decline in Th17 cell frequencies was observed in the rectal mucosa through the course of SIV infection and suppressive cART in this study ( Figure 7). However, no significant differences were observed in the frequencies of mucosal Tc17 cells and MAIT cells through the course of chronic treated SIV infection ( Figure 7). Interestingly, gut mucosal gd T cell frequencies increased significantly by 7 days post-SIV infection (p=0.005) and returned to baseline during early cART (Figure 7), suggesting an early response to ongoing viral replication in the gut. However, during the long-term treatment phase, a decline in gut mucosal gd T cells was observed reaching significance at the 7-9-month cART time-points (p=0.04 and 0.03 respectively; Figure 7) demonstrating similar kinetics of responses between blood and  gut mucosal compartments. Notably, this decline in gut mucosal gd T cells preceded the elevation in plasma IFABP, LBP and sCD14 (Figure 4), suggesting the loss of gd T cells may play a role in the subsequent increase of leaky gut biomarkers during chronic treated SIV infection. Thus, among the various subsets of Th17type T cells in the gut mucosa, significant changes in gd T cell frequencies aligned with subsequent increase in plasma inflammatory and leaky-gut markers during long-term cART suppressed SIV infection (Figures 3 and 4).

Impaired Gut Mucosal gd T Cell Effector Function Is Associated With Development of IEBD and Systemic Inflammation During Long-Term cART Suppressed SIV Infection
We next investigated the gd T cell effector functions and activation status in the context of acute SIV-infection, shortterm cART, and during the chronic phase of treated SIV infection (Figure 8 and Supplementary Figure 6). The rectal mucosal gd T cells in healthy SIV-naive macaques displayed a dominant IL-17 and IL-22 cytokine response to PMA/Ca Ionomycin stimulation that is comparable to the classical Th17 subsets (Figures 8 and 9B). However, production of the Th1 cytokine, TNF-a, was lower in gd T cells in comparison to both Th17 and Tc17 cells. We have previously shown that CD161+ Th17 type mucosal T cells, besides being capable of both Th17 and Th1 cytokine production, were more polyfunctional than CD161-negative T cells (23). However, it remains unclear whether there are any differences in the polyfunctionality or the balance between Th1 and Th17 cytokine responses between the subsets of gut mucosal Th17 type T cells. Thus, we compared the cytokine response of rectal mucosal gd T cells  Figure 8A), the rectal mucosal Th17 cells displayed sustained loss in IL-17/IL-22 production throughout the course of cART ( Figure 8B). Indeed, a persistent impact of SIV infection was observed on both frequencies ( Figure 6) and epithelial barrier-protective effector functions of mucosal Th17 cells. Likewise, IL-17 and IL-22 production was significantly reduced in gd T cells (Figure 8), which was evident earlier in PBMC (7 days post-SIV; Figure 8A) than in the rectal mucosa (30 days post-SIV; Figure 8B). There was an early increase in IFN-g and TNF-a production that returned to baseline by 30 days post-SIV in blood and rectal mucosal gd T cells (Figures 8A, B). The kinetics of the Th1/Th17 cytokine response was similar between gd T and Th17 cells in blood and rectal mucosa during acute SIV infection prior to cART ( Figure 8). However, unlike the continued decline in IL-17/IL-22 cytokine responses of rectal mucosal Th17 cells, gd T and Tc17 cells demonstrated a recovery to baseline following cART initiation ( Figure 8B). Intriguingly, even though IL-17/IL-22 effector functions were maintained during short-term (3-5 months) cART, a significant loss of IL-17 and IL-22 producing ability was observed in blood and rectal mucosal gd T cells around 7 month of cART ( Figure 8). Notably, during this stage of treated chronic SIV infection, gd T cells displayed significantly higher production of IFN-g in both peripheral blood (p=0.003 at 7 months, p=0.014 at 9 months cART) and rectal mucosa (p<0.01). Additionally, a trend towards enhanced TNF-a production was also observed in rectal mucosal gd T cells that reached significance (p=0.016) at 9 months post-cART ( Figure 8). Thus, despite return to baseline during early cART and effective suppression of viremia, gd T cells exhibited significant changes in effector functions reflecting a specific loss of Th17 type cytokine producing ability and skewing towards Th1 type cytokine responses. This dysfunction aligned with the reduced frequencies of rectal mucosal gd T cells, thus, strongly suggesting a role in the subsequent increase in plasma IEBD/MT biomarkers and inflammation observed in this study (heat maps in , which returned to baseline by 3-month cART in both Tc17 and gdT cells but not in Th17 cells. Significant reduction in IL-17-and IL-22-production of gdT cells was observed again at 7-month cART time point along with increase in IFN-g production. One-way ANOVA with Dunnett's multiple comparisons test was used to determine significant differences between baseline and different time points post-SIV infection and cART. Asterisks indicate significant differences in gdT cell cytokine production between time points (*p < 0.05; **p < 0.01, ***p < 0.001).  Figures 5B and 9A). Indeed, comparison of gut mucosal gd T cell polyfunctional responses for IL-17, IL-22, TNF-a and IFN-g through the course of SIV infection and cART demonstrated an initial loss of Th17 cytokine functions during acute SIV infection followed by return to baseline by 5 months of cART and subsequently a significant decline in IL-17/IL-22 polyfunctional, and increased TNF-a and IFN-g mono-functional responses (pie charts in Figure 9B). In contrast there was a sustained loss of polyfunctional cytokine responses in CD4 Th17 cells throughout acute and treated SIV infection (Supplementary Figure 4A) and no significant change in Tc17 cells until 9 months of cART (Supplementary Figure 4B).

Perturbation in Vd1 and Vd2 Subset Ratios During Chronic Treatment Correlates With Loss of Mucosal Th17-Type Cytokine Functions and Increased Plasma IEBD/MT Biomarkers
Primate gd T cells have a relatively restricted repertoire of V gene segments, with Vd1 and Vd2 being the most commonly used Vd gene segments in blood and mucosal tissues (44)(45)(46) and are termed Vd1 and Vd2 subsets. An inversion of the typical Vd2/ Vd1 subset ratio in circulating blood has been associated with HIV/SIV infection (16,45,47), however the impact of chronic treated infection on gut mucosal Vd1 and Vd2 subsets remains unclear. Since the overall gd T cell compartment displayed significant perturbations in frequencies and function aligning with the recurrence of IEBD and inflammation during the chronic phase of treated SIV infection in our study, Vd1 and Vd2 subset composition of blood and rectal mucosal gd T cells were examined over the course of long-term viral suppression with cART. The average Vd2/Vd1 ratio in blood and rectal mucosa was 1.7-1.9 at baseline prior to SIV challenge ( Figure  10B). Notably, by 7-day post-SIV infection there was a remarkable increase in circulating Vd2 frequencies, resulting in significantly higher numbers of total gd T cells ( Figure 7) and higher Vd2/Vd1 ratio (average of 13.76; Figure 10B) that returned to just below baseline during early cART. The Vd1 and Vd2 subset distribution in rectal mucosal lymphocytes could only be evaluated at baseline and chronic SIV infected time-points owing to the low cell yields during acute SIV infection and early cART time-points. For the time-points analyzed, the dynamics of Vd2/Vd1 ratio in rectal mucosa resembled the dynamics observed in blood (Figures 10A, B). Significantly reduced Vd2 frequencies from 7-months cART onwards with a concomitant increase in Vd1 frequencies was observed in PBMC and rectal biopsies ( Figure 10A) resulting in significant decline of the Vd2/Vd1 ratio from baseline ( Figure  10B). Furthermore, during the chronic phase of treatment, the specific loss of rectal mucosal Vd2 subsets significantly correlated with plasma IFABP levels ( Figure 11A). On the other hand, elevated levels of mucosal Vd1 cells displayed significantly correlated with plasma sCD14 concentrations ( Figure 11B). Taken together, the overall decline in IL-17/IL-22 functions along with significant loss of Vd2 cells, particularly in the rectal mucosa, coupled with significant correlations with plasma IFABP levels, points towards a likely role for Vd2 cell impairment in the loss of gut barrier functions, despite long-term effective cART in SIV-infected macaques. Several studies have examined inflammaging phenotype in PLWH by evaluating large datasets comprised of wide ageranges, often grouped as young and old categories (8,(49)(50)(51). Since biological aging is a continuum and is impacted by several factors including HIV infection itself, elucidating the immune mechanisms during virus infection and its treatment before the onset of biological aging is an important first step toward understanding the process of inflammaging in PLWH. Aging rhesus macaques typically begin exhibiting relatively higher levels of circulating cytokines/chemokines associated with inflammaging around 18 years of age (13,52). In this study, we evaluated the development of inflammaging phenotype through the course of treated SIV infection in young adult macaques (5-10 years old) well before the onset of non-infectious inflammaging. As reported in early HIV infection (11,53,54), we observed that systemic inflammation following acute mucosal (IR) SIV infection was associated with elevation in plasma LBP and IFABP, markers of MT and enterocyte damage respectively. It is important to note that during acute SIV infection prior to treatment with cART, while we observed a correlation between IFABP and LBP, we did not establish any relationship between either IFABP, or LBP and sCD14. Indeed, there was no significant increase in plasma levels of sCD14 during the inflammatory response of acute SIV infection despite a set-point viremia of 5.8 log and increase in several proinflammatory mediators including IL-1b, IL-18, IL-1Ra, MIG, ITAC, MIP-1b, G-CSF, and CXCL13 as well as elevated IFABP and LBP levels. Despite its wide use as a plasma MT marker, sCD14 is not an exclusive marker of ongoing translocation of bacterial by- products from gut lumen into blood circulation but also serves as a biomarker of monocyte activation (55) and there are conflicting reports on the levels of plasma sCD14 reflecting MT in PLWH (55,56) and SIV-infected macaques (31,57). It is likely that a more significant monocyte/macrophage activation and prolonged intestinal mucosal damage may be required for significantly elevated plasma sCD14 levels. Thus, the increase of sCD14 in the chronic phase of cART (7 months onwards in this study) possibly reflects a combination of persistent monocyte/macrophage activation and MT beyond the levels observed in the acute phase of SIV infection.
To understand the potential immune mechanisms of IEBD and MT associated with persistent inflammation during chronic cART-suppressed SIV infection, we longitudinally compared diverse subpopulations of Th17-type immune cells in blood and rectal mucosa. The main finding was the detection of a significant correlation of IEBD and systemic inflammation with a specific loss of Th17-type gdT effector functions during longterm suppressive ART. gdT cells represent a unique subset of effector T cells that can traffic to tissues, are enriched at gut mucosa where HIV prevails, and selectively target cancer or virally infected cells (47). Chronic HIV/SIV infection is associated with impaired proliferative responses of gdT cells (58,59) and inversion of Vd2:Vd1 subset ratios (16,45,60). However, the role of gut mucosal gdT cell functions in IEBD, MT and persistent inflammation during the course of treated HIV infection remains unclear. In this prospective study, we demonstrated that gut mucosal gdT cells are a dominant source of polyfunctional IL-17 and IL-22 responses among the diverse subsets of Th17-type immune cells including classical Th17, Tc17, and MAIT cells in healthy SIV-naïve macaques. Furthermore, gut mucosal gdT cells displayed important differences in the kinetics of IL-17/IL-22 cytokine responses from other Th17-type cells through the course of treated SIV infection. Notably, an early expansion of gdT cells in all of the study animals within a week of SIV infection, particularly in rectal mucosa, suggests a likely role in compensatory immune mechanisms for the preferential and sustained loss of gut CD4+ Th17 cells. However, this early expansion was not sufficient to prevent subsequent IEBD and MT, as plasma IFABP and LBP increased later in acute SIV infection, prior to treatment with cART. Indeed, despite the transient expansion in numbers, both peripheral and rectal gdT cells displayed a significant decline in IL-17 and IL-22 cytokine responses indicating that a specific decline in gdT IL-17/IL-22 effector functions, besides the significant gut mucosal Th17 depletion, likely contributed to the overall loss of epithelial barrier integrity during acute phase of infection. It is likely that the innate immune activation and the cytokine storm induced by acute infection modulated gdT cell functions toward loss of Th17-type cytokine expression and increases in Th1 cytokine production. This is supported by the increased production of TNF-a and IFN-g by both blood and rectal gdT cells observed in this study.
Intriguingly, in contrast to the widely reported expansion of circulating Vd1 cells during chronic HIV/SIV infection (14,21), we found that a significant increase in Vd2 rather than Vd1 subset contributed to the early expansion of gdT cell numbers within a week of SIV infection. However, our results are consistent with an earlier report in rhesus macaques (61) showing peripheral expansion of Vd2 cells that returned to baseline by 1-month post-SIV infection. Thus, it is likely that a very early expansion of Vd2 subset occurs in HIV infection as well, which needs further investigation. In the lack of data on Vd1 and Vd2 subset distribution in rectal mucosa during the acute pre-treatment phase of SIV infection in the current study, it remains to be confirmed whether a similar increase of Vd2 cells occurs in the gut mucosal compartment. However, based on the comparable kinetics of the changes in subset ratios between blood and rectal mucosa at baseline and throughout the treatment phase, we speculate that a similar expansion of gut mucosal Vd2 cells contributed to the increase in total gdT cell frequencies observed 1-week post-SIV infection in rectal mucosa. Normalization of this early switch in gdT subset ratios during suppression of viremia with cART and recovery of polyfunctional IL-17/IL-22 cytokine functions indicate that the perturbations in gdT cell functions and subset balance were driven by active viral replication prior to initiation of cART. Moreover, the significantly enhanced IL-22 responses in gut mucosal gdT cells during early viral suppression with cART suggests an active role in epithelial repair process of the IEBD caused during acute SIV infection. IL-22 is required for the preservation of the intestinal epithelial barrier and wound healing processes via the maintenance and proliferation of epithelial stem cells (62,63). However, in agreement with other reports of treated HIV infection (47), the recovery of gdT effector functions and subset ratios was not maintained through longterm cART and it was reversed to significantly lower Vd2 and higher Vd1 subset frequencies around 7 months of continued treatment and effective viral suppression. Importantly, this inversion of Vd1 and Vd2 subset ratio along with the recurrent loss of IL-17-and IL-22-producing effector functions preceded the subsequent increase in plasma IFABP, LBP, and sCD14 levels. This suggests that IL-17-and IL-22-producing gdT cells may have an essential role in the maintenance of the gut epithelial barrier integrity during effective suppression of viremia with cART, and that loss of these cells during chronic treated SIV infection may contribute to increase in IEBD and MT biomarkers. The mechanisms of dysregulation in gdT cell numbers and function during long-term suppressed SIV infection are unclear at this point. Several factors including ongoing viral replication in the gut tissue and/or long-term exposure to cART drugs, changes in the gut microbiome (64), or depletion of helper CD4+ T cells (65), etc. are likely to drive gdT cell dysfunction, thereby contributing to IEBD and MT. Although our data are not conclusive to demonstrate that this link is causative, it is noteworthy that plasma IFABP levels showed a significant negative correlation with frequency of gut mucosal Vd2 T cells during long-term cART phase. This argues for a causative role of decline in epithelial barrier-protective effector functions of Vd2 cells and the emergence of a leaky gut phenotype. On the other hand, the attendant increase in Vd1 subset and enhanced Th1 cytokine production may reflect a response of Vd1 cells to leaky gut mediated systemic inflammation, as Vd1 cells proliferate in response to inflammatory cytokines (66). Thus, besides the association of impaired Vd2 functions with leaky gut biomarkers, increases in7nbsp;Vd1 cells and their corresponding inflammatory cytokine production in the gut mucosa may further contribute to systemic inflammation during chronic cART-treated SIV infection.
Notably, the resurgence of plasma IEBD/MT biomarkers and inflammatory cytokine/chemokine markers resembled the inflammaging phenotype of older SIV-naïve macaques comprising of elevated levels of IFABP, LBP, sCD14, GM-CSF, IL-1b, IL-12, IL-6, and TNF-a (13), and was distinct from the cytokine storm of acute SIV infection phase. A similar inflammaging phenotype has been described in chronic HIV infected persons (8). Based on our data, we propose that the loss of Vd2 T cells initiated a vicious cycle of inflammation with the inversion of Vd1/Vd2 subset ratio and dysregulated cytokine effector functions in the residual gdT cells thereby contributing to development of inflammaging phenotype. Of note, a recent study has demonstrated that among diverse lymphocyte subsets including NK cells, T cells, Tregs, and iNKT cells, only gdT cell phenotype (activated/exhausted TIGIT+PD-1+ phenotype associated with plasma pro-inflammatory profile) could distinguish inflammaging in aviremic HIV infected individuals from HIV-negative younger and older individuals (50), implicating gdT cells as the key inflammatory driver in cARTsuppressed HIV infection. Moreover, HIV elite/viral controllers maintain significantly higher Vd2 frequencies than untreated or cART-treated PLWH and display preserved IL-17 function and lower immune activation (67,68). Further research is warranted in order to deeply understand the precise mechanism of HIV/ SIV-mediated gdT cell dysfunction in gut mucosa and the specific role of Vd1 and Vd2 subsets in driving leaky gutmediated inflammaging process. Our observation opens the possibility that augmenting IL-17/IL-22 cytokine functions of gdT cells may ameliorate persistent systemic inflammation of chronic treated SIV infection by promoting the maintenance of epithelial barrier functions and preventing MT. Clinical augmentation of Vd2 cells has been demonstrated in cancer patients through administration of bisphosphonates, which are a class of compounds recognized only by the Vd2 subset of gdT cells (69,70). A strength of this study is the longitudinal assessment of the development of inflammation during the course of effectively suppressed chronic SIV infection, which enabled the assessment of changes in immune cell functions in blood and gut mucosa prior to the emergence of leaky gut phenotype and systemic inflammation associated with chronic treated lentiviral infection. However, a limitation of our study design is the use of only female rhesus macaques to enable direct comparison with the inflammaging phenotype established in aging female rhesus macaques in our previous study. It will be of interest to evaluate whether there are gender-based differences in the functions of gdT cells and their role in the development of inflammaging phenotype in male macaques with chronic SIV infection and long-term cART.
In summary, we demonstrated that the inflammatory cytokine/chemokine signature of long-term cART is distinct from acute SIV infection. Furthermore, we have shown that significant changes in gut mucosal gdT cell functions, particularly loss of Vd2 cells and impaired IL-17/IL-22 producing ability are associated with development of systemic inflammation during chronic treated SIV infection in rhesus macaques. This study contributes to the current understanding of gdT cell dysregulation as a mechanism of HIV/SIV immune evasion and provides deeper insights into the potential role of gdT cell dysfunction in the re-emergence of leaky gut and inflammatory phenotype following initial control of immune activation during effective suppression of viremia by ART. Further studies aimed at identifying the mechanisms driving gdT cell dysfunction and increased gut permeability during aviremic SIV infection are required to test its potential as an immune-based intervention to improve mucosal homeostasis and reduce HIV-associated chronic inflammation.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The animal study was reviewed and approved by Tulane University Institutional Animal Care and Use Committee (IACUC).

AUTHOR CONTRIBUTIONS
NR conceived the project, designed experiments, and supervised the work. EW, NS, and GG performed experiments and analyzed data. BG is the veterinarian in this study. JM provided additional samples and PK, DW, RV, and SJ helped with overall data interpretation. All authors contributed to the article and approved the submitted version.

FUNDING
This study is supported by the NIGMS, NIH award P20 GM103629 awarded to SJ and NR, in part by the NIH grant P51OD011104 to Tulane National Primate Research Center, and in part by the Intramural Research Program, National Institute on Aging, NIH.

ACKNOWLEDGMENTS
We gratefully acknowledge Romas Geleziunas (Gilead), and Katie Kitrinos and James Demarest (ViiV) for supplying the antiretroviral drugs. The authors acknowledge Mary Barnes and Melissa Pattison of the Pathogen Detection and Quantification Core of Tulane National Primate Research Center (TNPRC) for assistance with the multiplex cytokine detection assays and use of the Bioplex-200 instrumentation, and the clinical veterinary staff in the Division of Veterinary Medicine at TNPRC and the NIA Nonhuman Primate Core technicians for coordinating the biospecimen collections. Technical assistance of the flow cytometry core facility staff at the TNPRC is greatly appreciated.

SUPPLEMENTARY MATERIAL
The One-way ANOVA with Dunnett's multiple comparisons test was used to determine significant differences from baseline. Asterisks indicate significant differences between time points (*p < 0.05; **p < 0.01; ***p < 0.001).