The persistence of a leaky gut in HIV-treated patients leads to chronic inflammation with increased rates of cardiovascular, liver, kidney, and neurological diseases. Tissue regulatory T (tTreg) cells are involved in the maintenance of intestinal homeostasis and wound repair through the IL-33 pathway. In this study, we investigated whether the persistence of gut mucosal injury during HIV infection might be explained in part by a flaw in the mechanisms involved in tissue repair. We observed an increased level of IL-33 in the gut of HIV-infected patients, which is associated with an increased level of fibrosis and a low peripheral reconstitution of CD4+ T cells. Our results showed that intestinal Treg cells from HIV-infected patients were enriched in tTreg cells prone to support tissue repair. However, we observed a functional defect in tTreg cells caused by the lack of amphiregulin secretion, which could contribute to the maintenance of intestinal damage. Our data suggest a mechanism by which the lack of amphiregulin secretion by tTreg may contribute to the lack of repair of the epithelial barrier.

Human immunodeficiency virus type 1 (HIV-1) infection is lifelong and to date, none of the available treatments eliminate the virus from infected patients. Indeed, effective antiretroviral therapy (ART) suppresses HIV-1 replication to undetectable levels but is unable to cure infection in HIV-infected people because the virus integrates its genetic information into the host cell’s own DNA. Moreover, HIV-infected patients receiving treatment have chronic inflammation with increased rates of cardiovascular, liver, kidney, and neurological diseases (1). Microbial translocation was described as a mechanism for immune activation and an underlying cause of T cell activation in HIV infection (2), which persists in ART-treated patients at significantly higher levels than in HIV-uninfected control subjects. Impairment of the gut barrier is a prerequisite for microbial translocation, and cross-sectional studies have shown the presence of gut mucosal injury and microbial translocation in long-term ART-treated individuals (3).

A novel concept in immunology is that tissue-resident immune cells not only mediate immune homeostasis and host defense but also critically contribute to the maintenance of organismal physiology (4). In this regard, T regulatory (Treg) cells have been shown to perform essential functions in maintenance of tissue tolerance by promoting tissue homeostasis and regeneration in nonlymphoid tissues. These nonlymphoid tissue Treg (tTreg) cells are thought to promote tissue repair by sensing tissue damage in situ via soluble factors, including the proinflammatory cytokine IL-18 and the alarmin IL-33, at least in part independent of TCR signaling (5, 6). A major component of this tissue regenerative pathway in Treg cells is the production of amphiregulin (57), a growth factor that plays an important role in healing. Amphiregulin activates locally latent TGF-β (8) and contributes to both the local suppression of inflammation and the local differentiation of tissue stem cells, and in this way to the process of wound healing and restoration of tissue homeostasis (9).

IL-33 is a tissue‐derived nuclear cytokine principally produced by endothelial cells, epithelial cells, fibroblast‐like cells, and myofibroblasts, rather than CD45+ hematopoietic cells, during both steady state and inflammation. IL-33, along with its receptor suppression of tumorigenicity 2 (ST2), has been shown to modulate both the innate and adaptive immune system. ST2 exists in two forms as splice variants: a soluble form (sST2), which acts as a decoy receptor, sequesters free IL-33, and does not signal; and a membrane-bound form (ST2), which activates the MyD88/NF-κB signaling pathway to enhance mast cell, Th2, Treg, and innate lymphoid cell type 2 functions (10). IL-33 regulates the homeostasis of tTreg cells in damaged tissues. Indeed, IL-33 signaling in tTreg cells provides a critical signal for tTreg accumulation and maintenance in inflamed tissues (11, 12). Thus, ST2+ tTreg cells’ function is to go to sites of damage and in the intestine ST2+ tTreg cells mediate repair (13).

We have previously reported that, in a large cohort of patients followed longitudinally, sST2 is a significant predictor of all-cause mortality (14). During acute HIV infection, sST2 levels correlated with the levels of intestinal fatty acid binding protein and of sCD14, markers of gut epithelial damage and microbial translocation, respectively (15). These observations establish a link between IL-33/sST2 and gut tissue damage during HIV infection.

In this study, we investigated whether the persistence of gut mucosal injury during HIV infection might be explained in part by a lack of tissue repair, namely, the IL-33/ST2 and the tTreg cells/amphiregulin pathways.

Our study population included patients infected with HIV on combined ART (cART). We enrolled study participants from the Agence Nationale de Recherches sur le Sida et les Hépatites Virales EP65 AMVIH cohort. Rectal tissue and blood samples were collected during clinics for anal dysplasia screening from 31 patients recruited at the Clinical Immunology and Infectious Diseases department (Henri Mondor Hospital, Créteil, France). Tissue samples were also collected from 14 healthy donors (HDs) during colonoscopy for colorectal cancer screening (Clinique Paris-Bercy, Charenton le Pont, France). For HIV-infected patients, inclusion criteria included: (1) HIV-1 infection, (2) age >18 y, (3) duration of HIV disease >2 y, (4) CD4 T cells >200/mm3, and (5) undetectable plasma HIV RNA (<20 copies/ml for at least 1 y). Median age of the 31 HIV patients was 44.3 y; 83.9% were men. None of the study participants were suffering from hepatitis B virus (HBV) or hepatitis C virus (HCV) coinfection. Significantly, the HDs were slightly older than the HIV patients (median: 52 y [range 41–75 y] versus 44 y [range, 23–66 y], respectively; p = 0.032). The sex ratio (female/male) was statistically similar in the HD group (0.4) and in the HIV patients group (0.2) (p = 0.43). Characteristics of HIV-1 infected and HD patients are summarized in Table I.

For HIV patients, PBMCs and rectal biopsies were collected on the same day. All rectal biopsies (around eight punches for each patient) were performed at the same site (i.e., at 10–15 cm from the anal margin) to avoid potential regional variation among patients and were collected in cold RPMI medium (supplemented with antibiotics). About 30–50 ml of blood was drawn by sterile venipuncture on the same day from HIV patients. Samples were transported at room temperature for immediate processing.

All study participants provided written informed consent to participation in the study. The study protocol was reviewed and approved by the Sud-Méditerranée II ethics committee (Comité de Protection des Personnes), was conducted in accordance with good clinical practice and the Declaration of Helsinki, and was registered at ClinicalTrials.gov (NCT03622177).

PBMCs were isolated by Ficoll (Lymphoprep ProteoGenix) density gradient centrifugation and used immediately, as well as the following day after overnight resting in culture medium at 37°C, simultaneously with the recovered lamina propria lymphocytes (LPLs) from biopsies. To isolate LPLs from biopsies, the tissue was first subjected to vigorous shaking incubation at room temperature for 25 min in HBSS medium (without Ca2+ and Mg2+) (Life Technologies) supplemented with 1 mM DTT (Sigma) and 1 mM EDTA, to release the intraepithelial lymphocytes, which were discarded. This mechanical disruption is followed by transfer of the tissue specimen to R-10 medium supplemented with penicillin (100 U/ml), streptomycin (100 µg/ml), amphotericin B (1.25 µg/ml), and gentamicin (50 µM) (Life Technologies), and rested overnight at 37°C, 5% CO2 in a six-well plate. The following day, the culture supernatants are collected to recover released cells.

The phenotype of regulatory T (Treg) cells was assessed using FACS staining with the following Abs: CD3 Alexa 700 (UCHT1; BD Pharmingen), CD4 BV605 (RPA-T4; BD Horizon), CD8 efluor780 (SK1; eBioscience), CD25 PeCy7 (M-A251; BD), IL-18Ra allophycocyanin (H44; eBioscience), ST2L FITC (B4E6; MD Biosciences), and CD161 BV421 (BioLegend). For intracellular staining, cells were fixed and permeabilized using the FOXP3 staining buffer set (eBioscience, Thermo Fisher Scientific), washed, and incubated with FOXP3 Alexa 488/PE-CF594 (259D/C7; BD) and amphiregulin-PE (AREG559). For all cell stainings, dead cells were excluded from the gating by using the LIVE/DEAD fixable dead cell stain kit (Molecular Probes, Invitrogen). Cells were acquired on an LSRII flow cytometer (BD Biosciences) and analyzed with FlowJo software, version 10 (FlowJo LLC).

Isolated LPLs and PBMCs suspended in R10 medium were supplemented with antibiotics in a 24-well plate and stimulated for 4 h at 37°C, 5% CO2 with anti-CD3-CD28 Dynabeads (Thermo Fisher) (cell/bead ratio of 1:1) in the presence of GolgiStop (BD Biosciences) and brefeldin A, to study amphiregulin expression. To study Tregs, we stimulated the cells with medium alone in the presence of GolgiStop and brefeldin A. For ST2L staining, the Ab was added to all the wells for the 4 h of incubation.

Deparaffinized rectal tissue sections were stained with polyclonal goat anti-human IL-33 (cat. AF3625; R&D Systems) at a 1:100 dilution at room temperature for 1 h. Enzymatic activity was blocked using BLOXALL Blocking Solution (SP-6000; Vector Labs) (phosphatase and peroxidase block). ImmPRESS HRP REAGENT KIT Anti-GOAT IgG (MP-7405; Vector Labs) was used containing horse blocking serum and Peroxidase Polymer anti-goat IgG reagent. For nonspecific blockage prior to primary Ab staining, horse blocking serum was used for 30 min (ImmPRESS kit). Following the primary Ab wash, the secondary anti-goat Ab was used for 30–40 min (ImmPRESS kit). Immunoreactive cells were visualized by addition of a diaminobenzidine substrate (DAB Substrate Kit for peroxidase) and were counterstained with hematoxylin.

Following the immunohistochemistry (IHC) IL-33 staining described, the images obtained were subsequently analyzed using the ImageJ macro. The ImageJ macro was written to quantify the percentage of area expressing IL-33 within an image, using the example in the following link (C. Shang, LSU Health Sciences Center-Shreveport, Research Core Facility, Shreveport, LA): https://resources.finalsite.net/images/v1567623985/lsuhscshreveportedu/awsil3hxeld0mkym4bzl/immunohistochemistrystaining.pdf.

One or more fragments of rectal biopsies were collected in a small volume of RNAlater (QIAGEN) and kept at 4°C for 24 h followed by transfer to −20°C.

The rectal biopsy specimens were cleared from the RNAlater, transferred to an aliquot containing a metallic bead and lysing solution constituting a mix of 2-ME and RLT plus buffer (QIAGEN) (volume ratio of 1:100), and disrupted in a cold mechanical disruptor (TissueLyser QIAGEN). RNA extraction was performed according to the manufacturer’s protocol (RNeasy Plus Mini Kit; QIAGEN). RNA was converted to cDNA with High-Capacity RNA-to-cDNA Kit (Applied Biosystems).

Quantitative PCRs were performed using the Brilliant II SYBR GREEN QPCR Master Mix kit (Agilent Technologies) for the expression of IL-33, epidermal growth factor receptor (EGFR), and TGF-β. OAZ-1 mRNA was used as a control for sample normalization. Expression of the gene OAZ-1 was used as a reference, and the relative levels of each gene were calculated using the 22ΔΔCT method.

The following primers were used in this study: IL-33 (forward, 5′-CAAATGAATCAGGTGACGGTGT-3′; reverse, 5′-TTGTTGGCATGCAACCAGAAG-3′), OAZ1 (forward, 5′-ACTTATTCTACTCCGATGATC-3′; reverse, 5′-GAGAATCCTCGTCTTGTC-3′), EGFR (forward, 5′-GGTGGCATTTAGGGGTGACT-3′; reverse, 5′-TGTGATAATTCAGCTCAAACCTGTG-3′), and TGF-β (forward, 5′-GCAAGTGGACATCAACGGGT-3′; reverse, 5′-TCCGTGGAGCTGAAGCAATA-3′).

Cytokine concentrations in the plasma of HIV patients were measured using Simoa Human Cytokine 6-plex assay, which simultaneously measures six cytokines with high sensitivity (∼10 fg/ml). The assay measured human IFN-γ, IL-6, IL-10, IL-12p70, IL-17A, and TNF-α using the Simoa SR-X benchtop biomarker detection system. The plasma samples that were stored at −80°C were thawed at room temperature prior to the assay. The assay was performed in triplicates for each sample, in a 96-well plate, following the manufacturer's instructions.

Human IL-33 protein concentrations in patients’ plasma were measured using the Bio-Plex 200 system (Bio-Rad). Plasma samples were stored at −80°C and thawed at room temperature for the assay. Luminex assays (R&D Systems) were performed according to the manufacturers’ instructions.

Paraffin-embedded rectal tissue sections were stained with Sirius red staining using Gemini AS Slide stainer. Images were taken at 4× magnification and subsequently analyzed using the batch mode of the ImageJ macro. The ImageJ macro was written to quantify the percentage of fibrosis within an image, using the example in the following link (Dr. A. N. DebRoy, Pediatric Gastroenterology Fellow, University of Iowa, Iowa City, IA): https://imagej.nih.gov/ij/docs/examples/stained-sections/index.html.

Briefly, for all images to be analyzed, the macro first splits the image into red, green, and blue channels with color transformation to grayscale. The green channel is selected for thresholding, followed by analyzing to obtain % area of fibrosis.

Groups were compared using the two-sided nonparametric Mann–Whitney U test. A p value <0.05 was considered significant. All statistical analyses were performed using Prism (version 6.0; GraphPad Software).

We previously reported an increased level of sST2 in the serum of a large cohort of HIV-1–infected patients (14). Because sST2 is an indirect measure of IL-33 activity, we sought to measure IL-33 levels in the plasma and in gut biopsies from a cohort of 31 HIV-1–infected individuals under long-term cART (Table I) and HDs. Previous studies have shown that plasma levels of IL-33 are usually very low and/or undetectable (16). Accordingly, we found that IL-33 protein concentration in the plasma of patients with HIV-1 and HDs were below the limit of detection (0.68 pg/ml) (data not shown). In contrast, IL-33 mRNA relative expression was significantly increased in gut biopsies from HIV-1–infected patients compared with HD (1.68 ± 0.3% versus 0.58 ± 0.1%; p = 0.0003; (Fig. 1A). An IHC study showed that IL-33 expression was weak, limited to cell nuclei, or absent in biopsies from HD controls (Fig. 1B). We observed two patterns of IL-33 expression in gut biopsies from HIV-1–infected individuals: (1) one pattern characterized by a high level of IL-33 protein expressed predominantly in epithelial cells, and (2) a second pattern characterized by a weak expression of IL-33 protein as shown in HDs (see (Fig. 1B for a representative staining in two different patients). For the first panel defined as high IL-33 expression in epithelial cells, IL-33 protein was also expressed by endothelial cells, fibroblasts, fibroblast-like cells, as well as in the extracellular matrix. As expected, quantitative analysis of IL-33 expression revealed a higher expression of IL-33 in patients with high IL-33 expression in epithelial cells compared with patients with low expression (Fig. 1C). Because IL-33 is primarily produced by intestinal epithelial cells on local damage (10), we analyzed samples based on the level of IL-33 protein expression specifically in epithelial cells for further study.

FIGURE 1.

Increased IL-33 levels in the intestine of chronic HIV infection on cART. (A) Comparison of gut mucosal tissue IL-33 mRNA relative expression in HIV-infected patients on cART (n = 21) and HDs (n = 11). (B) IL-33 protein expression by IHC on sections from healthy mucosal tissue (left) from one HIV-infected patient with a weak nuclear staining of mononuclear cells and no epithelial cell staining (middle) and from one HIV-infected patient with a strong IL-33 expression on sections (right). (C) Quantification of IL-33 protein by ImageJ software on sections of mucosal tissue from HIV-infected patients either with low-expressing tissue sections with weak staining and no epithelial expression or with high IL-33 epithelial staining. Data represent mean ± SEM. Statistical analyses were performed using nonparametric Mann–Whitney U test. ***p < 0.001, ****p < 0.0001.

FIGURE 1.

Increased IL-33 levels in the intestine of chronic HIV infection on cART. (A) Comparison of gut mucosal tissue IL-33 mRNA relative expression in HIV-infected patients on cART (n = 21) and HDs (n = 11). (B) IL-33 protein expression by IHC on sections from healthy mucosal tissue (left) from one HIV-infected patient with a weak nuclear staining of mononuclear cells and no epithelial cell staining (middle) and from one HIV-infected patient with a strong IL-33 expression on sections (right). (C) Quantification of IL-33 protein by ImageJ software on sections of mucosal tissue from HIV-infected patients either with low-expressing tissue sections with weak staining and no epithelial expression or with high IL-33 epithelial staining. Data represent mean ± SEM. Statistical analyses were performed using nonparametric Mann–Whitney U test. ***p < 0.001, ****p < 0.0001.

Close modal
Table I.

Main characteristics of HIV-1–infected patients

Median (Interquartile Range) or %HIV+ cART (n = 31)HIV (n = 14)p Value
Demography    
 Men 83.9% 57.1% 0.4%a 
Ethnicity    
 White 80.6% ND  
 Sub-Saharan African 19.4% ND  
Year of HIV diagnosis 2013 (2001–2014)   
Pre-ART    
 CD4+ T cell nadir, cells/ml (n = 28) 266 (62.75–346.25)   
 CD4+ T cell nadir, 0.500 cell/ml (n = 28) 3.60%   
 PVL log10 copies/ml (n = 28) 77,970 (29,586.25–362,686)   
At inclusion    
 Age, y 44.3 (37.9–46.9) 52 (41–75) 0.032b 
 Duration with HIV, y 4.8 (3.2–17.6)   
 Overall cART exposure, y (n = 29) 4.4 (2.9–11.7)   
 CD4+ T cells/ml (n = 30) 594 (453.5–751)   
 CD4+:CD8+ ratio (n = 29) 1 (0.55–1.28)   
Coinfectionc (%)    
 None 100%   
Median (Interquartile Range) or %HIV+ cART (n = 31)HIV (n = 14)p Value
Demography    
 Men 83.9% 57.1% 0.4%a 
Ethnicity    
 White 80.6% ND  
 Sub-Saharan African 19.4% ND  
Year of HIV diagnosis 2013 (2001–2014)   
Pre-ART    
 CD4+ T cell nadir, cells/ml (n = 28) 266 (62.75–346.25)   
 CD4+ T cell nadir, 0.500 cell/ml (n = 28) 3.60%   
 PVL log10 copies/ml (n = 28) 77,970 (29,586.25–362,686)   
At inclusion    
 Age, y 44.3 (37.9–46.9) 52 (41–75) 0.032b 
 Duration with HIV, y 4.8 (3.2–17.6)   
 Overall cART exposure, y (n = 29) 4.4 (2.9–11.7)   
 CD4+ T cells/ml (n = 30) 594 (453.5–751)   
 CD4+:CD8+ ratio (n = 29) 1 (0.55–1.28)   
Coinfectionc (%)    
 None 100%   
a

Fisher exact test.

b

Mann–Whitney U test.

c

Coinfection (HBV/HCV).

PVL, plasma viral load.

IL-33 is an alarmin released after mucosal injury leading to control inflammation and fibrosis as a mechanism of repair. Therefore, we performed Sirius red staining to analyze collagen deposition in gut biopsies as a marker of fibrosis (17). As shown in (Fig. 2A, we found an increased deposition of collagen in gut biopsies from HIV-1–infected patients compared with those from HDs. Moreover, the frequency of areas exhibiting fibrosis was higher in biopsies from patients with high expression of IL-33 protein in epithelial cells than in patients with low or no IL-33 expression (p = 0.02; (Fig. 2B).

FIGURE 2.

Fibrosis and inflammation linked to the pattern of IL-33 expression in the gut mucosa of HIV-infected patients. (A) Sirius red staining on gut mucosal tissue sections (top), from HIV seronegative donor (left) and HIV-infected patients on cART (two on right). Lower three images showing the Sirius red staining quantification by ImageJ software for the respective images on top. (B) Graph showing the association between % tissue fibrosis quantified by ImageJ macros and IL-33 protein expression levels in the gut by IHC. HIV-seropositive patients on cART (HIV, n = 27). (C) Graphs representing the correlation of CD4 reconstitution in the peripheral blood with gut IL-33 protein expression levels. Data represent mean ± SEM. Statistical analyses were performed using nonparametric Mann–Whitney U test. *p < 0.05.

FIGURE 2.

Fibrosis and inflammation linked to the pattern of IL-33 expression in the gut mucosa of HIV-infected patients. (A) Sirius red staining on gut mucosal tissue sections (top), from HIV seronegative donor (left) and HIV-infected patients on cART (two on right). Lower three images showing the Sirius red staining quantification by ImageJ software for the respective images on top. (B) Graph showing the association between % tissue fibrosis quantified by ImageJ macros and IL-33 protein expression levels in the gut by IHC. HIV-seropositive patients on cART (HIV, n = 27). (C) Graphs representing the correlation of CD4 reconstitution in the peripheral blood with gut IL-33 protein expression levels. Data represent mean ± SEM. Statistical analyses were performed using nonparametric Mann–Whitney U test. *p < 0.05.

Close modal

Fibrosis within the parenchyma could have a profound impact on immunological function and lymphocyte population dynamics (18). Accordingly, we found significantly lower blood CD4+ T cell counts in patients with high IL-33 expression than with low IL-33 expression (470.5 ± 75.8 cells/ml versus 694.0 ± 81.9 cells/ml, respectively; p = 0.03; (Fig. 2C). This distinct profile of patients was underscored by the measurement of several inflammatory markers using a Simoa technology. Among several factors measured (Supplemental Fig. 1), we observed a trend toward an increase of IFN-γ plasma levels in patients with high expression of IL-33 as compared with those with low levels (0.28 ± 0.1 versus 0.11 ± 0.01 pg/ml, respectively; p = 0.08), whereas IL-12p70 was significantly increased (0.27 ± 0.05 versus 0.17 ± 0.01 pg/ml, respectively; p = 0.04; Supplemental Fig. 1). IL-6, TNF-α, IL-17, and IL-10 concentrations were similar in both groups (Supplemental Fig. 1).

We showed earlier an association of increased expression of IL-33, a marker of tissue damage, and gut fibrosis in a group of HIV-1–infected patients. The IL-33–ST2 axis mediates repair of intestinal integrity through mucosal Treg cells (19). We investigated the frequency of tTreg cells in gut biopsies from eight HIV-infected patients and eight HDs. In a first attempt, we did not find differences between groups in the frequency of global Treg cells, defined as FOXP3+CD25+ T cells (Fig. 3A). However, the absolute number of global Treg cells was decreased in the gut of HIV-infected patients compared with HDs (243.8 versus 415.0; p = 0.03) (Fig. 3A). Among mucosal Treg cells, CD161+ tTreg cells, a distinct highly suppressive population of Treg cells, were described to be enriched in intestinal lamina propria (20). We then analyzed expression of CD161 and IL-18Rα on Treg cells in the lamina propria (Fig. 3B, Supplemental Fig. 2A). The frequency of tTreg CD161+IL-18Rα+ was increased in the gut of HIV-1+ patients as compared with HDs (11.52 ± 2.9% versus 3.20 ± 0.9%; p = 0.007; (Fig. 3B). Because CD161+CD4+ T cells serve as an important compartment of the HIV-1 latent reservoir and contain a significant amount of clonally expanded proviruses (21, 22), we wondered whether the total compartment of CD161+CD4+ T cells also increases in the gut of HIV-infected patients. We observed no significant difference in either frequency or absolute number of LPL CD4+CD161+ T cells between HIV-infected patients and HDs (data not shown). The CD161+IL-18Rα+ Treg cell population was present only in the gut, but not in the blood, of HIV-1–infected patients, suggesting that these cells are tissue-resident cells (Fig. 3B). In a second step, we were able to characterize ST2+ tTreg cells in the gut of four HIV-infected patients and five HDs (Fig. 3C, Supplemental Fig. 2B). The frequency of the ST2+ tTreg population was also increased in the gut of HIV patients compared with HDs without reaching significance (12.60 ± 4.787% versus 4.77 ± 0.19%, respectively; p = 0.29; (Fig. 3C).

FIGURE 3.

Repair Tregs in the intestinal mucosa in chronic HIV infection. (A) Frequency of total Treg population within CD4 T cells and number of total Treg population in HIV-infected patients on cART (n = 8) and HDs (n = 8), using the nonparametric Mann–Whitney U test. Data represent mean. (B) Representative FACS plots of CD161 IL-18Rα coexpression on FOXP3+CD25+CD4+ T cells. Individual data showing CD161+IL-18Rα+ Treg cells frequencies in PBMCs and LPLs of HD (n = 8) and HIV-infected patients (n = 8). (C) Representative FACS plots of ST2 expression on FOXP3+CD25+CD4+ T (in blue) or conventional CD4+ T cells (in red). Individual data showing ST2 expressing Treg frequencies in PBMCs and LPLs of HDs (n = 5) and HIV-infected patients (n = 4). Statistical analyses of paired PBMCs and LPLs of HIV patients are performed by Wilcoxon matched-pairs signed rank test. Comparison of HD and HIV LPLs is performed using the nonparametric Mann–Whitney U test. Data represent mean ± SEM. *p < 0.05, **p < 0.01.

FIGURE 3.

Repair Tregs in the intestinal mucosa in chronic HIV infection. (A) Frequency of total Treg population within CD4 T cells and number of total Treg population in HIV-infected patients on cART (n = 8) and HDs (n = 8), using the nonparametric Mann–Whitney U test. Data represent mean. (B) Representative FACS plots of CD161 IL-18Rα coexpression on FOXP3+CD25+CD4+ T cells. Individual data showing CD161+IL-18Rα+ Treg cells frequencies in PBMCs and LPLs of HD (n = 8) and HIV-infected patients (n = 8). (C) Representative FACS plots of ST2 expression on FOXP3+CD25+CD4+ T (in blue) or conventional CD4+ T cells (in red). Individual data showing ST2 expressing Treg frequencies in PBMCs and LPLs of HDs (n = 5) and HIV-infected patients (n = 4). Statistical analyses of paired PBMCs and LPLs of HIV patients are performed by Wilcoxon matched-pairs signed rank test. Comparison of HD and HIV LPLs is performed using the nonparametric Mann–Whitney U test. Data represent mean ± SEM. *p < 0.05, **p < 0.01.

Close modal

We looked at three well-defined molecules involved in tissue repair and the wound healing process, i.e., TGF-β, amphiregulin, and its receptor, the EGFR. We showed an increase of TGF-β mRNA in gut biopsies from HIV-1–infected individuals as compared with HDs (8.14 ± 1.9% versus 1.4 ± 0.13%, respectively; p = 0.06; (Fig. 4A). Interestingly, IL-33 and TGF-β mRNA had a significant negative correlation (r = −0.88; p < 0.0001; (Fig. 4B). Through the EGFR binding, amphiregulin contributes to wound healing, mainly by inducing proliferation of epithelial cells. We found an upregulation of EGFR mRNA in HIV-1–infected individuals as compared with HDs (0.07 ± 0.01 versus 0.03 ± 0.0; p = 0.009; (Fig. 4C). Noteworthy, amphiregulin specifically produced by tTreg cells has a key role in protecting against tissue damage and maintaining barrier integrity during virus-induced inflammatory responses in mucosa (5). We looked first at the capacity of total CD4+ T cells isolated from gut biopsies to produce amphiregulin after TCR stimulation. We found a decreased frequency of amphiregulin producing CD4+ T cells in gut biopsies from HIV-1–infected individuals as compared with HD (1.90 ± 0.3% versus 4.15 ± 0.8%; p = 0.06; (Fig. 4E). Interestingly, we found a profound defect of FOXP3+CD25+ Treg cells producing amphiregulin in gut biopsies from HIV-1–infected individuals with virtually no cells as compared with HDs in whom 7.3% of FOXP3+CD25+ Treg cells produce amphiregulin (p = 0.02; (Fig. 5A). Among the amphiregulin-secreting Treg cells in the lamina propria from HDs, ST2+ Treg cells secrete more amphiregulin than their ST2 counterparts do (Fig. 5B). Altogether, these data demonstrate a profound defect of amphiregulin production by tTreg cells in the gut of HIV-1–infected patients despite the presence of ST2+ tTreg cells.

FIGURE 4.

Factors involved in repair of the intestinal mucosa in chronic HIV infection. (A) Comparison of mucosal tissue TGF-β mRNA relative expression in HIV-infected patients on cART (n = 16) and HDs (n = 10), using the nonparametric Mann–Whitney U test. Data represent mean ± SEM. (B) Graph representing IL-33 correlation with TGF-β, in HIV-infected patients on cART (n = 16), by Spearman correlation. (C) Comparison of mucosal tissue EGFR mRNA relative expression in HIV-infected patients on cART (n = 16) and HDs (n = 10), using the nonparametric Mann–Whitney U test. Data represent mean ± SEM. (D) Representative FACS plots of AREG expression gated on CD4+ T cells in healthy LPLs before and after TCR stimulation. (E) Frequency and total number of AREG+ CD4+ T cells in the lamina propria of HD (n = 5) and HIV-infected samples (HIV, n = 4). Data represent mean ± SEM. *p < 0.05, **p < 0.01.

FIGURE 4.

Factors involved in repair of the intestinal mucosa in chronic HIV infection. (A) Comparison of mucosal tissue TGF-β mRNA relative expression in HIV-infected patients on cART (n = 16) and HDs (n = 10), using the nonparametric Mann–Whitney U test. Data represent mean ± SEM. (B) Graph representing IL-33 correlation with TGF-β, in HIV-infected patients on cART (n = 16), by Spearman correlation. (C) Comparison of mucosal tissue EGFR mRNA relative expression in HIV-infected patients on cART (n = 16) and HDs (n = 10), using the nonparametric Mann–Whitney U test. Data represent mean ± SEM. (D) Representative FACS plots of AREG expression gated on CD4+ T cells in healthy LPLs before and after TCR stimulation. (E) Frequency and total number of AREG+ CD4+ T cells in the lamina propria of HD (n = 5) and HIV-infected samples (HIV, n = 4). Data represent mean ± SEM. *p < 0.05, **p < 0.01.

Close modal
FIGURE 5.

CD25+FOXP3+ Tregs expressing AREG after TCR stimulation in LPLs. (A) Summary plot, comparing frequency and cell count of AREG expression on Tregs of LPLs between HDs (n = 5) and HIV-infected (HIV) patients (n = 4) after TCR stimulation. (B) CD25+FOXP3+ Tregs expressing AREG after TCR stimulation in LPLs isolated from HDs. Representative FACS plots of ST2+ and ST2 cells gated on CD25+FOXP3+ CD4 T cells, and the expression of AREG by ST2+ (blue) and ST2 (red) Treg subsets was assessed by flow cytometry. Values of AREG in ST2+ Treg cells and ST2 Treg cells represent mean fluorescence intensity (MFI). *p < 0.05, **p < 0.01.

FIGURE 5.

CD25+FOXP3+ Tregs expressing AREG after TCR stimulation in LPLs. (A) Summary plot, comparing frequency and cell count of AREG expression on Tregs of LPLs between HDs (n = 5) and HIV-infected (HIV) patients (n = 4) after TCR stimulation. (B) CD25+FOXP3+ Tregs expressing AREG after TCR stimulation in LPLs isolated from HDs. Representative FACS plots of ST2+ and ST2 cells gated on CD25+FOXP3+ CD4 T cells, and the expression of AREG by ST2+ (blue) and ST2 (red) Treg subsets was assessed by flow cytometry. Values of AREG in ST2+ Treg cells and ST2 Treg cells represent mean fluorescence intensity (MFI). *p < 0.05, **p < 0.01.

Close modal

In this report, we showed that IL-33 expression is increased in the gut of HIV-infected patients compared with non–HIV-infected healthy individuals. When IL-33 secretion participates in fibrosis, mucosal Treg population from HIV-infected patients is enriched in tTreg cells prone to support tissue repair. However, we observed a functional defect in tTreg cells because of the lack of amphiregulin production, which could contribute to the persistence of local damage and release of IL-33. Our data point out a vicious circle where the absence of amphiregulin secretion by tTreg cells might contribute to IL-33 secretion leading to fibrosis and to a defect in epithelial barrier function involved in microbial translocation and chronic inflammation.

Our results showed an increased area of fibrosis in patients who displayed a high level of gut IL-33 expression in epithelial cells. A dysregulated IL-33 pathway can play a critical role in the progression of chronic inflammation and fibrosis. Increased expression/production of collagen in mucosal, GALT, and lymph node has been reported previously in the setting of HIV-1 infection (18). GALT fibrosis is a contributor to incomplete CD4+ T lymphocyte cell restoration in chronic HIV infection despite suppressive ART and is associated with a decreased frequency of lymphoid aggregate CD4+ T lymphocytes. During the chronic phase of HIV infection, collagen deposition in lymphoid tissues, including Peyer's patches in the gut, disrupts the T cell zone and limits the number of resident CD4 T cells (23). The development of fibrosis in lymphoid tissue has a profound impact on immunological function and lymphocyte population dynamics. According to this, we showed that patients with high expression of IL-33 in epithelial cells displayed a lower blood CD4+ T count than patients with low expression of IL-33. Longer duration of HIV infection or cART treatment (Table II), which may also impact the development of fibrosis, did not reach significance probably because of the small number of patients.

Table II.

Comparing main characteristics between two groups of HIV-1–infected patients

Median (Interquartile Range) or %Low IL-33 (n = 15)High IL-33 (n = 14)p Value
Demography    
 Men 73.3% 92.9% 0.3 
Ethnicity   0.65 
 White 73.3% 85.71%  
 Sub-Saharan African 26.7% 14.29%  
Year of HIV diagnosis 2014 (2002–2015) 2007 (1999–2013) 0.09 
Pre-ART    
 CD4+ T cell nadir, cells/ml (n = 13) 291 (72–371) 199 (35–338) 0.9 
 CD4+ T cell nadir, 0.500 cell/ml (n = 13) 0.00% 7.70%  
 PVL log10 copies/ml (n = 13) 65,895 (25,414–452,362) 128,397 (48,385–332,794) 0.6 
At inclusion    
 Age, y 44.3 (41.5–46) 40.7 (35.4–50) 0.8 
 Duration with HIV, y 4.4 (3.1–17.06) 9.1 (3.5–19.3) 0.4 
 Overall cART exposure, y 3.3 (2.8–10.24) (n = 13) 7.0 (3.1–19.3) (n = 14) 0.2 
 CD4+ T cells/ml 683 (560–862) (n = 15) 467 (387–620) (n = 13) 0.03 
 CD4+:CD8+ ratio 1.1 (0.6–1.35) (n = 14) 0.8 (0.4–1.23) (n = 13) 0.2 
Coinfectiona (%)    
 None 100% 100%  
Median (Interquartile Range) or %Low IL-33 (n = 15)High IL-33 (n = 14)p Value
Demography    
 Men 73.3% 92.9% 0.3 
Ethnicity   0.65 
 White 73.3% 85.71%  
 Sub-Saharan African 26.7% 14.29%  
Year of HIV diagnosis 2014 (2002–2015) 2007 (1999–2013) 0.09 
Pre-ART    
 CD4+ T cell nadir, cells/ml (n = 13) 291 (72–371) 199 (35–338) 0.9 
 CD4+ T cell nadir, 0.500 cell/ml (n = 13) 0.00% 7.70%  
 PVL log10 copies/ml (n = 13) 65,895 (25,414–452,362) 128,397 (48,385–332,794) 0.6 
At inclusion    
 Age, y 44.3 (41.5–46) 40.7 (35.4–50) 0.8 
 Duration with HIV, y 4.4 (3.1–17.06) 9.1 (3.5–19.3) 0.4 
 Overall cART exposure, y 3.3 (2.8–10.24) (n = 13) 7.0 (3.1–19.3) (n = 14) 0.2 
 CD4+ T cells/ml 683 (560–862) (n = 15) 467 (387–620) (n = 13) 0.03 
 CD4+:CD8+ ratio 1.1 (0.6–1.35) (n = 14) 0.8 (0.4–1.23) (n = 13) 0.2 
Coinfectiona (%)    
 None 100% 100%  
a

Coinfection (HBV/HCV).

PVL, plasma viral load.

IL-33 stimulation of human subepithelial myofibroblasts induced the expression of extracellular matrix components and several growth factors known to promote epithelial cell proliferation (24). Previous studies reported that intestinal myofibroblasts displayed an activated phenotype and contributed to the profibrotic changes in HIV patients under cART (25). Another study has also reported that flagellin produced by adherent-invasive Escherichia coli induces the induction of ST2 expression in IECs, which in turn increases IL-33 signaling and promotes the development of intestinal fibrosis (26). Therefore, it is tempting to speculate that dysbiosis and microbial translocation observed in HIV infection contribute to intestinal fibrosis through the IL-33 pathway activation.

Chronic immune activation elicits a Treg counterresponse with the adverse consequence of increased collagen production and deposition. Therefore, IL-33 may contribute to the development of fibrosis by promoting Treg accumulation during the chronic phase of HIV infection. We did not detect any difference in the frequency of global Treg cells between the gut of HIV-infected subjects and healthy individuals, but a depletion of total Treg in absolute number was observed. Our data are consistent with two previous studies showing that ART normalizes Treg cells frequency in the gut (27, 28). However, we extend significantly these results by deciphering the phenotype of these tTreg cells showing a shift toward tTreg cells prone to tissue repair. The tTreg cell populations in nonlymphoid tissue exhibit phenotypic heterogeneity. Both populations’ CD161+ Treg cells and ST2+ Treg cells were detected in tissues and not in blood from healthy individuals, suggesting their function of tissue-resident Tregs. IL-33 triggers retinoic acid signaling on CD4+ T cells (29), which induces CD161 expression on Treg cells (20). We could speculate IL-33 supports tTreg cell maintenance in the gut of HIV-infected patients.

These cells are critical for nonimmunological tissue repair function. We highlighted a functional defect of amphiregulin secretion by tTreg cells. The Rudensky group (5), using a mouse strain with a Treg-specific deficiency of amphiregulin, showed a substantially more severe form of symptoms after influenza infection than wild-type mice with concomitant reduction in blood oxygen saturation. ST2 deficiency limited to Treg cells led to disease exacerbation in a model of experimental autoimmune encephalomyelitis (30). Our report extends these data showing a profound defect of amphiregulin-secreting Treg cells in the gut of HIV-infected patients. The mechanism underlying this defect remains to be explored. A recent article from Bhaskaran et al. (31) showed that FOXP3+ cells require IL-1β– and PD1-dependant asparaginyl peptidase activation for expansion and sustained expression of FOXP3 and amphiregulin. In the context of HIV infection, they showed that only the endogenous IL-1β released because of caspase-1 activity upregulated amphiregulin in vitro. Whether this pathway is involved in the defect of amphiregulin expression needs to be investigated in the future.

The exact mechanism by which amphiregulin contributes to wound healing has remained unresolved. Amphiregulin is a poor mitogen for epithelial cells, and recent data show that one major function of amphiregulin is the local induction of TGF-β (8). During wound healing, amphiregulin induced the TGF-β–mediated differentiation of tissue stem cells and in this way critically contributed to the restoration of tissue homeostasis. Our data also showed an increased level of TGF-β mRNA in the gut of HIV-infected patients compared with HDs and an inverse correlation between mRNA–IL-33 and TGF-β mRNA. The inverse correlation between IL-33 and TGF-β levels is surprising and might be explained, in this context, by the deficit of amphiregulin secretion. Amphiregulin is critical for efficient Treg cell function (32) and enhances the suppressive capacity of Treg cells through the EGFR pathway (33). As reported, Treg cells exhibited reduced immunosuppressive capacity during HIV infection (34). It is tempting to speculate that the defect of amphiregulin secretion may participate in altered Treg cells function, which is not able to control local inflammation, an obstacle to the repair of the epithelial barrier.

Our study has several limitations. Because of the low number of cells, we cannot determine whether ST2+ Treg cells express CD161 or IL-18Rα and whether ST2+ Treg cells lack repair function because of the absence of amphiregulin secretion. The small number of patients included in this study makes it impossible to determine whether the IL-33 expression pattern is associated with a different systemic inflammatory profile. Furthermore, it will be of great interest to determine whether the defect in amphiregulin secretion is observed in all patients regardless of this IL-33 expression pattern. The injection of recombinant amphiregulin has demonstrated the alleviation of symptoms in several different experimental settings, such as during influenza infection or after viral–bacterial coinfections (6, 35). Regardless of whether all patients displayed a defect of amphiregulin-secreting Treg cells, this strategy could be considered for treating HIV-infected patients.

In conclusion, we observed an increased level of IL-33 in the gut of HIV-infected patients, which is associated with an increased level of fibrosis and a low peripheral reconstitution of CD4+ T cells. Our results suggested that IL-33 promotes maintenance of tTreg cells, but IL-33 did not induce secretion of amphiregulin by these cells. It remains to determine the reasons for this deficit. Because recombinant amphiregulin enhances wound healing (6, 35), it will be interesting to test whether amphiregulin may enhance epithelial barrier function and then decrease the symptoms of the leaky gut observed during HIV infection.

We thank the study volunteers for participation. We thank Philippe Gaulard and Nadine Martin-Garcia of the pathology department for providing support for our IHC experiments. We also thank Christelle Gandolphe and Wilfred Verbecq-Morlot from the Imaging platform for providing help with the Sirius red stainings.

This work was supported by Agence Nationale de la Recherche (ANR), Emerging Infectious Diseases and the Labex Vaccine Research Institute (Investissements d’Avenir program managed by the ANR under reference ANR-10-LABX-77-01) and ANRS|Maladies Infectieuses Émergentes, Agence Nationale de Recherches sur le Sida et les Hépatites Virales (ANRS) (Grant AMVIH). M.T. was supported by CARMA (Investissements d’Avenir program managed by the ANR under reference ANR-15-RHUS-0003).

S.H., S.G., J.-D.L., and Y.L. conceived and supervised the study. S.H., S.G., J.-D.L., Y.L., and M.T. designed the experiments and analyzed the data. M.T., M.S., A.W., C.L., and F.J.-L. performed the experiments. S.G., J.L.L.Z., J.-D.Z., and S.Y.-D. recruited the participants and collected the samples. S.H., Y.L., J.-D.L., and M.T. wrote the manuscript, with contributions from all authors.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ART

antiretroviral therapy

cART

combined antiretroviral therapy

EGFR

epidermal growth factor receptor

HBV

hepatitis B virus

HCV

hepatitis C virus

HD

healthy donor

IHC

immunohistochemistry

LPL

lamina propria lymphocyte

sST2

soluble form of suppression of tumorigenicity 2

ST2

suppression of tumorigenicity 2

Treg

regulatory T

tTreg

tissue regulatory T

1.
Deeks
S. G.
2011
.
HIV infection, inflammation, immunosenescence, and aging.
Annu. Rev. Med.
62
:
141
155
.
2.
Brenchley
J. M.
,
D. A.
Price
,
T. W.
Schacker
,
T. E.
Asher
,
G.
Silvestri
,
S.
Rao
,
Z.
Kazzaz
,
E.
Bornstein
,
O.
Lambotte
,
D.
Altmann
, et al
2006
.
Microbial translocation is a cause of systemic immune activation in chronic HIV infection.
Nat. Med.
12
:
1365
1371
.
3.
Mudd
J. C.
,
J. M.
Brenchley
.
2016
.
Gut mucosal barrier dysfunction, microbial dysbiosis, and their role in HIV-1 disease progression.
J. Infect. Dis.
214
(
Suppl. 2
):
S58
S66
.
4.
Panduro
M.
,
C.
Benoist
,
D.
Mathis
.
2016
.
Tissue Tregs.
Annu. Rev. Immunol.
34
:
609
633
.
5.
Arpaia
N.
,
J. A.
Green
,
B.
Moltedo
,
A.
Arvey
,
S.
Hemmers
,
S.
Yuan
,
P. M.
Treuting
,
A. Y.
Rudensky
.
2015
.
A distinct function of regulatory T cells in tissue protection.
Cell
162
:
1078
1089
.
6.
Burzyn
D.
,
W.
Kuswanto
,
D.
Kolodin
,
J. L.
Shadrach
,
M.
Cerletti
,
Y.
Jang
,
E.
Sefik
,
T. G.
Tan
,
A. J.
Wagers
,
C.
Benoist
,
D.
Mathis
.
2013
.
A special population of regulatory T cells potentiates muscle repair.
Cell
155
:
1282
1295
.
7.
Ito
M.
,
K.
Komai
,
S.
Mise-Omata
,
M.
Iizuka-Koga
,
Y.
Noguchi
,
T.
Kondo
,
R.
Sakai
,
K.
Matsuo
,
T.
Nakayama
,
O.
Yoshie
, et al
2019
.
Brain regulatory T cells suppress astrogliosis and potentiate neurological recovery.
Nature
565
:
246
250
.
8.
Minutti
C. M.
,
R. V.
Modak
,
F.
Macdonald
,
F.
Li
,
D. J.
Smyth
,
D. A.
Dorward
,
N.
Blair
,
C.
Husovsky
,
A.
Muir
,
E.
Giampazolias
, et al
2019
.
A macrophage-pericyte axis directs tissue restoration via amphiregulin-induced transforming growth factor beta activation.
Immunity
50
:
645
654.e6
.
9.
Zaiss
D. M.
,
C. M.
Minutti
,
J. A.
Knipper
.
2019
.
Immune- and non-immune-mediated roles of regulatory T-cells during wound healing.
Immunology
157
:
190
197
.
10.
Griesenauer
B.
,
S.
Paczesny
.
2017
.
The ST2/IL-33 axis in immune cells during inflammatory diseases.
Front. Immunol.
8
:
475
.
11.
Schiering
C.
,
T.
Krausgruber
,
A.
Chomka
,
A.
Fröhlich
,
K.
Adelmann
,
E. A.
Wohlfert
,
J.
Pott
,
T.
Griseri
,
J.
Bollrath
,
A. N.
Hegazy
, et al
2014
.
The alarmin IL-33 promotes regulatory T-cell function in the intestine.
Nature
513
:
564
568
.
12.
Sefik
E.
,
N.
Geva-Zatorsky
,
S.
Oh
,
L.
Konnikova
,
D.
Zemmour
,
A. M.
McGuire
,
D.
Burzyn
,
A.
Ortiz-Lopez
,
M.
Lobera
,
J.
Yang
, et al
2015
.
MUCOSAL IMMUNOLOGY. Individual intestinal symbionts induce a distinct population of RORγ+ regulatory T cells.
Science
349
:
993
997
.
13.
Siede
J.
,
A.
Fröhlich
,
A.
Datsi
,
A. N.
Hegazy
,
D. V.
Varga
,
V.
Holecska
,
H.
Saito
,
S.
Nakae
,
M.
Löhning
.
2016
.
IL-33 receptor-expressing regulatory T cells are highly activated, Th2 biased and suppress CD4 T cell proliferation through IL-10 and TGFβ release.
PLoS One
11
:
e0161507
.
14.
Thiébaut
R.
,
S.
Hue
,
F.
Le Marec
,
J.-D.
Lelièvre
,
M.
Dupon
,
E.
Foucat
,
E.
Lazaro
,
F.
Dabis
,
P.
Duffau
,
L.
Wittkop
, et al
ANRS CO3 Aquitaine Cohort
.
2017
.
Serum suppression of tumorigenicity 2 level is an independent predictor of all-cause mortality in HIV-infected patients.
AIDS
31
:
2355
2365
.
15.
Jenabian
M.-A.
,
M.
El-Far
,
K.
Vyboh
,
I.
Kema
,
C. T.
Costiniuk
,
R.
Thomas
,
J.-G.
Baril
,
R.
LeBlanc
,
C.
Kanagaratham
,
D.
Radzioch
, et al
Montreal Primary Infection and Slow Progressor Study Groups
.
2015
.
Immunosuppressive tryptophan catabolism and gut mucosal dysfunction following early HIV infection.
J. Infect. Dis.
212
:
355
366
.
16.
Sun
Z.
,
B.
Chang
,
A.
Huang
,
S.
Hao
,
M.
Gao
,
Y.
Sun
,
M.
Shi
,
L.
Jin
,
W.
Zhang
,
J.
Zhao
, et al
2019
.
Plasma levels of soluble ST2, but not IL-33, correlate with the severity of alcoholic liver disease.
J. Cell. Mol. Med.
23
:
887
897
.
17.
Dao
M.
,
C.
Pouliquen
,
A.
Duquesne
,
K.
Posseme
,
C.
Mussini
,
A.
Durrbach
,
C.
Guettier
,
H.
François
,
S.
Ferlicot
.
2020
.
Usefulness of morphometric image analysis with Sirius Red to assess interstitial fibrosis after renal transplantation from uncontrolled circulatory death donors.
Sci. Rep.
10
:
6894
.
18.
Estes
J. D.
,
A. T.
Haase
,
T. W.
Schacker
.
2008
.
The role of collagen deposition in depleting CD4+ T cells and limiting reconstitution in HIV-1 and SIV infections through damage to the secondary lymphoid organ niche.
Semin. Immunol.
20
:
181
186
.
19.
Whibley
N.
,
A.
Tucci
,
F.
Powrie
.
2019
.
Regulatory T cell adaptation in the intestine and skin.
Nat. Immunol.
20
:
386
396
.
20.
Povoleri
G. A. M.
,
E.
Nova-Lamperti
,
C.
Scottà
,
G.
Fanelli
,
Y.-C.
Chen
,
P. D.
Becker
,
D.
Boardman
,
B.
Costantini
,
M.
Romano
,
P.
Pavlidis
, et al
2018
.
Human retinoic acid-regulated CD161+ regulatory T cells support wound repair in intestinal mucosa.
Nat. Immunol.
19
:
1403
1414
.
21.
Li
X.
,
Z.
Liu
,
Q.
Li
,
R.
Hu
,
L.
Zhao
,
Y.
Yang
,
J.
Zhao
,
Z.
Huang
,
H.
Gao
,
L.
Li
, et al
2019
.
CD161+ CD4+ T cells harbor clonally expanded replication-competent HIV-1 in antiretroviral therapy-suppressed individuals.
MBio
10
:
e02121-19
.
22.
Lal
K. G.
,
Y.
Phuang-Ngern
,
S.
Suhkumvittaya
,
E.
Leeansyah
,
A.
Alrubayyi
,
J.
Dias
,
A.
Waickman
,
D.
Kim
,
E.
Kroon
,
S.
Pinyakorn
, et al
2020
.
Longitudinal analysis of peripheral and colonic CD161+ CD4+ T cell dysfunction in acute HIV-1 infection and effects of early treatment initiation.
Viruses
12
:
1426
.
23.
Schacker
T. W.
,
C.
Reilly
,
G. J.
Beilman
,
J.
Taylor
,
D.
Skarda
,
D.
Krason
,
M.
Larson
,
A. T.
Haase
.
2005
.
Amount of lymphatic tissue fibrosis in HIV infection predicts magnitude of HAART-associated change in peripheral CD4 cell count.
AIDS
19
:
2169
2171
.
24.
Maywald
R. L.
,
S. K.
Doerner
,
L.
Pastorelli
,
C.
De Salvo
,
S. M.
Benton
,
E. P.
Dawson
,
D. G.
Lanza
,
N. A.
Berger
,
S. D.
Markowitz
,
H.-J.
Lenz
, et al
2015
.
IL-33 activates tumor stroma to promote intestinal polyposis.
Proc. Natl. Acad. Sci. USA
112
:
E2487
E2496
.
25.
Asmuth
D. M.
,
I. V.
Pinchuk
,
J.
Wu
,
G.
Vargas
,
X.
Chen
,
S.
Mann
,
A.
Albanese
,
Z.-M.
Ma
,
R.
Saroufeem
,
G. P.
Melcher
, et al
2015
.
Role of intestinal myofibroblasts in HIV-associated intestinal collagen deposition and immune reconstitution following combination antiretroviral therapy.
AIDS
29
:
877
888
.
26.
Imai
J.
,
S.
Kitamoto
,
K.
Sugihara
,
H.
Nagao-Kitamoto
,
A.
Hayashi
,
T. L.
Morhardt
,
P.
Kuffa
,
P. D. R.
Higgins
,
N.
Barnich
,
N.
Kamada
.
2019
.
Flagellin-mediated activation of IL-33-ST2 signaling by a pathobiont promotes intestinal fibrosis.
Mucosal Immunol.
12
:
632
643
.
27.
Epple
H.-J.
,
C.
Loddenkemper
,
D.
Kunkel
,
H.
Tröger
,
J.
Maul
,
V.
Moos
,
E.
Berg
,
R.
Ullrich
,
J.-D.
Schulzke
,
H.
Stein
, et al
2006
.
Mucosal but not peripheral FOXP3+ regulatory T cells are highly increased in untreated HIV infection and normalize after suppressive HAART.
Blood
108
:
3072
3078
.
28.
Favre
D.
,
J.
Mold
,
P. W.
Hunt
,
B.
Kanwar
,
P.
Loke
,
L.
Seu
,
J. D.
Barbour
,
M. M.
Lowe
,
A.
Jayawardene
,
F.
Aweeka
, et al
2010
.
Tryptophan catabolism by indoleamine 2,3-dioxygenase 1 alters the balance of TH17 to regulatory T cells in HIV disease.
Sci. Transl. Med.
2
:
32ra36
.
29.
Gajardo
T.
,
F.
Pérez
,
C.
Terraza
,
M.
Campos-Mora
,
R. J.
Noelle
,
K.
Pino-Lagos
.
2016
.
IL-33 enhances retinoic acid signaling on CD4+ T cells.
Cytokine
85
:
120
122
.
30.
Hemmers
S.
,
M.
Schizas
,
A. Y.
Rudensky
.
2021
.
T reg cell-intrinsic requirements for ST2 signaling in health and neuroinflammation.
J. Exp. Med.
218
:
e20201234
.
31.
Bhaskaran
N.
,
E.
Schneider
,
F.
Faddoul
,
A.
Paes da Silva
,
R.
Asaad
,
A.
Talla
,
N.
Greenspan
,
A. D.
Levine
,
D.
McDonald
,
J.
Karn
, et al
2021
.
Oral immune dysfunction is associated with the expansion of FOXP3+PD-1+Amphiregulin+ T cells during HIV infection.
Nat. Commun.
12
:
5143
.
32.
Zaiss
D. M. W.
,
J.
van Loosdregt
,
A.
Gorlani
,
C. P. J.
Bekker
,
A.
Gröne
,
M.
Sibilia
,
P. M.
van Bergen en Henegouwen
,
R. C.
Roovers
,
P. J.
Coffer
,
A. J.
Sijts
.
2013
.
Amphiregulin enhances regulatory T cell-suppressive function via the epidermal growth factor receptor.
Immunity
38
:
275
284
.
33.
Wang
S.
,
Y.
Zhang
,
Y.
Wang
,
P.
Ye
,
J.
Li
,
H.
Li
,
Q.
Ding
,
J.
Xia
.
2016
.
Amphiregulin confers regulatory T cell suppressive function and tumor invasion via the EGFR/GSK-3β/Foxp3 axis.
J. Biol. Chem.
291
:
21085
21095
.
34.
López-Abente
J.
,
R.
Correa-Rocha
,
M.
Pion
.
2016
.
Functional mechanisms of Treg in the context of HIV infection and the Janus face of immune suppression. [Published erratum appears in 2018 Front. Immunol. 9: 792.]
Front. Immunol.
7
:
192
.
35.
Monticelli
L. A.
,
L. C.
Osborne
,
M.
Noti
,
S. V.
Tran
,
D. M. W.
Zaiss
,
D.
Artis
.
2015
.
IL-33 promotes an innate immune pathway of intestinal tissue protection dependent on amphiregulin-EGFR interactions.
Proc. Natl. Acad. Sci. USA
112
:
10762
10767
.

The authors have no financial conflicts of interest.

Supplementary data