Aryl Hydrocarbon Receptor Activation Promotes Effector CD4⁺ T Cell Homeostasis and Restrains Salt-Sensitive Hypertension

Abstract

Excess dietary salt and salt-sensitivity contribute to cardiovascular disease. Distinct T cell phenotypic responses to high salt and hypertension, as well as influences from environmental cues, are not well understood. The aryl hydrocarbon receptor (AhR) is activated by dietary ligands, promoting T cell and systemic homeostasis. We hypothesized that activating AhR supports CD4⁺ homeostatic functions, such as cytokine production and mobilization, in response to high salt intake while mitigating salt-sensitive hypertension. In the intestinal mucosa, we demonstrate that a high-salt diet (HSD) is a key driving factor, independent of hypertension, in diminishing interleukin 17A (IL-17A) production by CD4⁺ T (Th17) cells without disrupting circulating cytokines associated with Th17 function. Previous studies suggest that hypertensive patients and individuals on a HSD are deficient in AhR ligands or agonistic metabolites. We found that activating AhR augments Th17 cells during experimental salt-sensitive hypertension. Further, we demonstrate that activating AhR in vitro contributes to sustaining Th17 cells in the setting of excess salt. Using photoconvertible Kikume Green-Red mice, we also revealed that HSD drives CD4⁺ T cell mobilization. Next, we found that excess salt augments T cell mobilization markers, validating HSD-driven T cell migration. Also, we found that activating AhR mitigates HSD-induced T cell migration markers. Using telemetry in a model of experimental salt-sensitivity, we found that activating AhR prevents the development of salt-sensitive hypertension. Collectively, stimulating AhR through dietary ligands facilitates immunologic and systemic functions amid excess salt intake and restrains the development of salt-sensitive hypertension.

Graphical Abstract

AhR ligand in the presence of high salt diet chow increases colonic homeostatic Th17 cells as well as restrains salt-sensitive hypertension.

Open in new tabDownload slide

Introduction

High salt intake and hypertension are both significant factors leading to cardiovascular disease contributing to grave morbidity and a marked reduction in quality of life.^1–4 Economically, excess salt intake leads to the loss of tens of billions of dollars due to hospitalization and loss in productivity.^{5,  6} Thus, enumeration of mechanisms involved in high salt intake, with and without hypertension, is critical to US and global health. Environmental cues play a key role in regulating immune cell function and fate.^{7,  8} Studies have demonstrated distinct alterations in adaptive immunity, specifically effector CD4⁺ T cells in response to high dietary salt^9–14 as well as in hypertension.^15–18

At homeostasis, Th17 cells are critical to maintaining health and function.^19–23 However, when homeostasis is disrupted, Th17 and IL-17A producing cells are known to promote various disease states, including hypertension.^{13,  24–29} Studies show that exogenous administration of interleukin 17A (IL-17A) further augments hypertension during angiotensin II-dependent hypertension,^{24,  26} while other studies suggest that the loss of IL-17A abrogates end-organ damage during induced experimental hypertension.^{25,  28} Key studies show that Th17 cells are salt responsive with several mechanisms revealed.^{12–14,  30} Recent studies indicate that specific Th17 programs, aligned as homeostatic or pathogenic, may also influence cytokine production.^{29,  31–33} Previous work from Wu et al. and Matthias et al. identified a pathogenic phenotype of mouse and human Th17 cells in the context of high salt.^{14,  30} Homeostatic and pathogenic Th17 cells have been described in the gut, and very little is known about how high salt intake affects the responses in homeostatic and pathogenic programming of Th17 cells.

The gut is highly responsive to changes in the diet and has the highest number of immune cells for all non-lymphoid tissues^34–36 making it a tissue of interest. Omenetti et al. identified distinct homeostatic and pathogenic Th17 subsets in the intestines dependent on segmented filamentous bacteria (SFB) status with SFB having a specific effect on homeostatic Th17 cell differentiation.³³ Previous studies have shown that high salt intake promotes colitis progression due to changes in the microbial environment and inflammation; however, the colonic pathogenic and homeostatic Th17 cell response to high salt was not reported.^{37,  38} Furthermore, mobilization and infiltration of CD4⁺ effector T cells is central to end-organ pathogenesis across multiple diseases,^{31,  39–44} but it is unclear if excess dietary salt directly impacts CD4⁺ T cell mobilization. As the prevalence of excess dietary salt intake in today’s society is high, further study is required to elucidate the mechanism(s) regulating homeostatic or pathogenic T cell functions by high salt.^{16,  30,  45–47}

The aryl hydrocarbon receptor (AhR) has been described as a key mediator of homeostatic Th17 responses.^48–51 AhR, which is ubiquitously expressed across diverse cell types,⁵² is critical for T cell function,^53–55 and is highly dependent on ligand availability.⁴⁸ Numerous ligands enhance AhR activity with the most abundant dietary source derived from cruciferous vegetables.⁵⁶ Recent studies suggest that patients with hypertension and/or individuals who consume excess salt are deficient in tryptophan-derived indoles and AhR-binding metabolites.^{12,  57,  58} AhR activation is also critical for protection against colitis.⁵⁹ However, it is unclear if reduced AhR activation is key to excess salt-responsiveness in Th17 cells or plays a role in salt-sensitive hypertension.

We, therefore, hypothesized that AhR activation in mice on a high-salt diet (HSD) and/or salt-sensitive hypertension supports CD4⁺ homeostatic functions, specifically cytokine production and mobilization. We further hypothesized that loss of AhR activation contributes to salt-sensitive hypertension and that dietary supplementation of an AhR ligand mitigates salt-sensitive hypertension. To test these hypotheses, we employed an experimental model of salt-sensitive hypertension to evaluate the Th17 cell responses and blood pressure to HSD and/or salt-sensitive hypertension with AhR activation. We utilized a model of salt-sensitive hypertension, L-NAME + HSD to determine whether HSD alone is responsible for Th17 cell responses independent of augmented blood pressure or whether a systemic blood pressure response to HSD is required to disrupt Th17 cell functions. The pre-clinical L-NAME/high salt experimental mouse model of salt-sensitive hypertension is widely used to recapitulate human salt-sensitivity.

Collectively, our study indicates that AhR activation in the presence of excess salt restores key facets of Th17 cell functionality and the homeostatic Th17 phenotype. Also, we show that the development of salt-sensitive hypertension is prevented with dietary supplementation to promote AhR activation. These pre-clinical studies would indicate that in the clinical setting activating AhR through dietary intervention may be a target in stabilizing CD4⁺ T cell disruption in the setting of excess dietary salt consumption as well as restraining the development of salt-sensitive hypertension.

Methods

Animal Ethics

All animal and cellular experiments, derived from animals, were conducted following institutional, local, state, and federal guidelines. All permissions were provided by the Institutional Animal Care and Use Committee at the University of Alabama at Birmingham. All animals, unless indicated, were maintained in a humidity- and temperature-controlled room on a 12:12 h lights on/off cycle (on 7 am–7 pm; off 7 pm–7 am) with ad libitum access to food and water at the University of Alabama at Birmingham. Details about the diet and water are given below.

Mice

Specific-pathogen-free C57BL/6J male mice were purchased from Jackson Laboratory (Pathogen & Opportunist Free Maximum Barrier Facility RB10 with regular interval testing, Sacramento, CA) at weaning (18–21 days old). Jackson laboratory mice are devoid of SFB (Candidatus Arthromitus) while housed within the Jackson lab facilities.^60–63 Upon arrival at UAB, newly weaned 18–21-day-old mice were randomly assigned for co-housing to acquire the UAB Kaul vivarium microbiome for at least 5 weeks and to normalize microbial composition across groups.^{64,  65} Following a 5–9-week acclimatization period, in a random cohort of experimental C57BL6/J mice, 92% (11 of 12) tested PCR-positive for SFB and all tested PCR-negative for Helicobacter hepaticus, indicating acquisition of SFB during the weaning reaction window. The Jackson laboratory confirms the use of a genomic stability program using monthly introduced C57BL6/J founder mice into breeding colonies to minimize genetic drift. All mice entered experimental protocols between 8 and 12 weeks of age. IL17A-IRES-GFP-KI (IL-17A-eGFP) and Il17a^CreR26R^eYFP (IL-17Afm-eYFP) fate-mapping mice were kindly provided by Dr. Laurie Harrington. Breeding pairs of CAG:: KikGR (KikGR) photoconvertible mice were kindly provided by Dr. Troy Randall.⁶⁶ Littermate-matched male mice were utilized for all reporter colonies between 8 and 12 weeks of age. Where appropriate, tissues were collected during the inactive period between ZT1 and ZT3.

Diet

All diets were dye-free and matched nutritionally and provided ultra-filtered sterile drinking water. Where indicated, mice were placed on a dye-free normal salt diet (NSD) (0.4% NaCl, TD.96208, Teklad) or a nutritionally matched HSD (4.0% NaCl, TD.92034, Teklad) plus 1% NaCl (wt/vol) sterile drinking water (HSD) for either 2 weeks or 5 weeks. Diets are identical in composition except for NaCl content. To investigate HSD-mediated blood pressure disruption, beginning between 8 and 12 weeks of age, mice were randomly assigned to the following groups: (a) NSD (TD.96208, Teklad) and sterile drinking water for the entire protocol, (b) NSD (TD.96208, Teklad) for 4 weeks followed by a 2-week HSD period, or (c) provided NSD alongside L-NAME-supplemented sterile drinking water for 3 weeks followed by withdrawal of L-NAME and continued NSD and normal sterile drinking water for 3 weeks, or (d) mice were given NSD with 0.5 mg/mL L-NAME (Alfa Aesar; cat# AAH6366614) supplemented sterile drinking water for 3 weeks. Afterward, L-NAME was removed, and mice received NSD with normal sterile drinking water for 1 week. Following the 1-week hiatus, mice were then freely fed HSD for 2 weeks. L-NAME-supplemented sterile drinking water was changed every third day, as previously described.^{12,  67,  68} In specified experiments, Teklad directly incorporated indole-3-carbinol (I3C) into chow at 200 parts per million (mg I3C/kg chow).^69–71

Photoconversion of Kikume Green-Red Mice

Photoconversion procedures were adapted from protocols previously published and validated.^{66,  72–76} Kikume Green-Red male mice were anesthetized with 2% isoflurane (vol/vol) and maintained at 37°C on a warming platform. Abdominal hair was shaved, followed by the application of a depilatory cream for 2 min. Once the abdomen was cleared of all hair, a sterile environment was prepared with betadine surgical solution. A 1-cm incision was made through the abdominal skin and abdominal wall. The small intestine with Peyer’s patches (PP) was gently withdrawn with minimal disruption and laid onto a sterile aluminum foil blanket that covered the mouse to block and avoid light exposure to other tissues and lymph nodes. A Dymax QX4 405 nm wand system with an LED head and collimator was placed 3.4 cm above the withdrawn small intestine. The front and back of small intestinal tissue were exposed to 405 nm light at an intensity of 95 mW/cm² for 13 min. During illumination, tissue was kept moist with sterile, warmed PBS. The intestine with PP was returned to the abdominal cavity, followed by closure of the abdominal wall and skin with sterile suture. Mice were allowed to recover and provided analgesic therapy as directed and needed based on daily observation. Tissues were collected 3 days post-photoconversion to evaluate the presence of photoconverted KikumeRed⁺ cells in tissues of interest, by flow cytometry, indicating immune cell egress from the small intestine. Ex vivo staining protocol below provides additional details for analysis by flow cytometry.

Preparation of Tissues

All mice were anesthetized using an isoflurane aerosolizer with 1.5%–2% isoflurane, with the depth of anesthesia confirmed by gentle paw pressure. Blood was directly collected by cardiac puncture into an EDTA-coated syringe and centrifuged at 2000 g for 10 min at 4°C to separate plasma from red cell fraction. Plasma was immediately snap-frozen through liquid nitrogen immersion followed by –80°C storage until further analysis. Mice were then gently perfused with 10 mL ice-cold PBS to remove blood-borne leukocytes in tissues. Tissue preparations for single-cell suspensions are described below.

Spleens and Lymph Nodes

Spleens and lymph nodes were gently removed and crushed between 2 sterile frosted slides. Suspensions were filtered through 70 and 40 |$\mu $|m filters and centrifuged at 400 g for 5 min at 4°C. Splenocytes were ACK lysed, followed by quenching and re-centrifugation. Splenic and lymph node single-cell suspensions were resuspended in R10 and placed on ice until further processing.^{77,  78}

Intestines and Kidneys

The entire large intestine (without cecum) was harvested and cleaned of luminal contents and feces. Peyer’s patches, from the distal two-thirds of the small intestine, were removed using curved scissors and placed into ice-cold R10 until processing. Intestines were cut into tissue segments followed by a quick wash with ice-cold buffer. Epithelial cells and intraepithelial lymphocytes were washed off by incubating in 37°C stripping buffer (for 20 min at 400 x g on a flat-bottom shaker). Tissue pieces were transferred and resuspended in copious warm H2/wash buffer, followed by mincing in 2 mL R10. Minced tissue was digested at 37°C at 400 x g in 10 mL total digest buffer (1 mg/mL collagenase type IV, 0.1 U/mL dispase II, and 0.2 mg/mL DNase type I) for 45 min, followed by gentle disruption using a transfer pipette and gentleMACS dissociation (m_intestine_01). Enzymes were quenched with a large volume of ice-cold R5. Single-cell suspensions were filtered through 70 and 40 |${\rm{\mu }}$|m filters and centrifuged at 400 g for 10 min at 4°C. Kidneys were immediately collected, decapsulated, and placed into 2 mL of complete media and minced. Subsequently, kidney pieces were digested (1 mg/mL collagenase type IV, 0.1 U/mL dispase II, and 0.2 mg/mL DNase type I) for 45 min followed by gentle disruption using a transfer pipette and gentle MACS dissociation (non-lymphoid tissue program), and enzymes were quenched with a large volume of kidney wash buffer and filtered through 70 and 40 |${\rm{\mu }}$|m filters and centrifuged at 400 g for 10 min at 4°C. All single-cell suspensions were gently resuspended in 5 mL of 37% isosmotic Percoll and centrifuged at 700 g for 20 min at 22°C (acceleration 3, brake 0) to enrich colonic lamina propria lymphocytes (cLPL) and kidney immune cells. Debris was removed, and the pellet was washed and centrifuged twice with ice-cold R10.^79–81

Peyer’s Patches

Isolated PPs were processed using the GentleMacs dissociator (m_spleen_02) followed by digestion in digest buffer (0.5 mg/mL collagenase type IV, 0.1 U/mL dispase II, 200 |${\rm{\mu }}$|g/mL DNase type I) at 37°C for 20 min at 400 x g on a flat-bottom shaker. PP underwent GentleMacs dissociation (m_spleen_03) and mashed through a 70 |${\rm{\mu }}$|m filter, followed by washing and filtering through a 40 |${\rm{\mu }}$|m filter. Samples were centrifuged at 400 g for 7 min at 4°C, and pellets were resuspended in R10.

Buffers

Complete RPMI media: RPMI-1640 supplemented with 2 mm L-glutamine, 1% Pen-Strep, 50 |$\mu $|m 2-mercaptoethanol (2-ME), 1x NEAA, 10 mm HEPES, 1 mm Sodium Pyruvate, and 10% fetal bovine serum (FBS). Complete IMDM media: IMDM with 25 mm HEPES supplemented with 2 mm L-glutamine, 1% Pen-Strep, 50 |$\mu $|m 2-ME, 1x NEAA, 1 mm Sodium Pyruvate, and 10% FBS. H2/wash buffer: HBSS (w/o Ca/Mg), 2% FBS, and 1% Pen-Strep. Stripping Buffer (H2/EDTA/DTT): HBSS (w/o Ca/Mg), 5 mm EDTA, 1 mm dithiothreitol (DTT), 1% Pen-Strep, and 2% FBS. Kidney Wash buffer: HBSS (without Ca/Mg), 2% FBS, 2 mm EDTA. Flow cytometry (FACS) buffer: PBS (w/o Ca/Mg) with 2% FBS, 1% Pen-Strep, and 2 mm EDTA. Digestion buffer in R10: 1 mg/mL collagenase type IV (∼285 U/mL), 0.1 U/mL dispase II, 200 |$\mu $|g/mL Deoxyribonuclease I from bovine pancreas (DNase I). R5: RPMI-1640 supplemented with 5% FBS, and 1% pen-strep. Room temperature 37% Percoll in R5. Cell stimulation media: T cells: 50 ng/mL phorbol 12-myristate 13-acetate (PMA), 1 μg/mL ionomycin, 10 |$\mu $|g/mL Brefeldin A (BFA); ILC3: PMA/Ionomycin/BFA with 20 ng/mL recombinant mouse IL-1|$\beta $| and 20 ng/mL recombinant mouse IL-23. Fixation buffer: 2% PFA fixation buffer in FACS.

Catalog Numbers

RPMI 1640 without L glutamine (SH30096FS, Fisher), HbSS without Ca, Mg, phenol Red (SH3058802, Fisher), Hyclone IMDM without glutamine (SH3025901, Fisher), PBS without Ca2+/Mg2+, Corning (MT21040CV, Fisher), Sodium Pyruvate (BW13-115E, Fisher), 100x non-essential amino acids (SH3023801, Fisher), HEPES (BP299100, Fisher), and 55 mm 2-Mercaptoethanol in DPBS (21985023, Thermo Fisher), Collagenase type IV, Gibco (17-104-019, Fisher), DNase Type I (DN25-100 mg, Sigma), Dispase II (5 U/mL), StemCell Tech (NC9886504, Fisher), EDTA, 0.5 M, pH 8.0 (15-575-020, Fisher), Percoll (45-001-747, Fisher), Dithiothreitol, DTT (43819-5 G, sigma), PMA (P1585-1MG, Sigma), Ionomycin, calcium salt, >98% (I0634-5MG, Sigma), eBioscience FOXP3 Kit (staining buffer kit Invitrogen™ 00 552 300) (50-112-8857, Fisher), Brilliant Violet buffer plus 1000T (BDB566385, Fisher), Ultra-comp beads (50-112-9040, Fisher) or Ultra-comp beads Plus (01-3333-42, Thermo Fisher), Brefeldin A (B7651-25MG), 32% Methanol-free PFA.

Ex Vivo Stimulation

For intracellular cytokine determination, single-cell suspensions were stimulated with 50 ng/mL phorbol myristate acetate (PMA) 1 mg/mL ionomycin) at 37°C in complete media for 1 h. After 1 h of incubation, brefeldin A (10 μg/mL final concentration) was supplemented for the final 3 h at 37°C to sequester cytokines. For ILC3 stimulations, stimulation media was supplemented with 20 ng/mL recombinant mouse IL-1|$\beta $| (130-101-682, Miltenyi Biotec) and 20 ng/mL recombinant mouse IL-23 (34-8231-82, eBioscience), in addition to PMA, Ionomycin, and brefeldin A protocol. Once complete, samples were spun down, stimulation media decanted, and washed twice with ice-cold FACS to quench stimulation⁷⁹.

Ex Vivo Staining

Single-cell suspensions were stained with amine-reactive live-dead dye (Thermo Fisher amine-reactive live-dead dye (live/dead aqua or live/dead blue) or Biolegend zombie near-infrared (NIR)) for dead cell exclusion in ice-cold PBS for 15 min protected from light at 4°C. After washing with cold FACS buffer, cells were incubated with Fc receptor blockade (FcX plus and anti-CD16.2 FcγRIV) for 25 min, followed by the addition of fluorophore-conjugated surface markers in ice-cold FACS (PBS, 2% FBS, 2 mm EDTA) with brilliant violet plus buffer for an additional 25 min on ice protected from light. Cells were washed and fixed overnight at 4°C in FOXP3 eBioscience fixation buffer protected from light. Subsequently, samples were washed and washed with room-temperature eBioscience FOXP3/transcription factor perm buffer, along with a brief 10 min incubation in perm buffer. Samples were then stained in permeabilization buffer for 45 min at room temperature protected from light. Afterward, samples were washed and fixed in 2% PFA for 10 min on ice protected from light. Samples were then washed twice, resuspended in cold FACS buffer, and protected from light at 4°C until acquisition. In experiments with fluorescent reporter proteins, following surface staining, samples were gently fixed with 3.7% FA (Image-IT fixative, methanol-free) at room temperature for 20 min. Cells were washed and permeabilized with 0.1% IGEPAL CA-630 for 4–5 min at room temperature protected from light. Cells were washed twice with cold FACS and stained for intracellular cytokines overnight at 4°C protected from light in FACS buffer. Samples were then washed twice, fixed, and resuspended in FACS. Samples were acquired on a BD FACSymphony A3/A5 flow cytometer or a Cytek Northern Lights spectral cytometer. In experiments where KikGR mice are used, KikGreen signal is detected using the 488 nm laser (BP 515/20, LP 505), and KikRed signal is detected using the 561 nm laser (BP 586/15, LP 570) on the BD FACSymphony A3/A5.⁸² Flow cytometry data were analyzed with Flowjo software (V10, Tree Star). Single-color controls were utilized to establish compensation and unmixing parameters (beads and cells), and fluorescence-minus-one (FMO), no-stimulation controls, and no-stain controls were utilized to assign gating. Dead cells and doublets were excluded from analysis and absolute cell counts.

Fluorophore-Conjugated Antibodies Used for Flow Cytometry

CD45 (30-F11), TCR|$\beta $| (H57-597), CD3 (145-2C11 or 17A2), TCR|$\gamma \delta $| (GL3), CD4 (GK1.5), CD8|$\alpha $| (53-6.7), CD25 (PC61), CD11b (M1/70), CD11c (N418), CD44 (IM7), CCR6 (29-2L17), CXCR3 (CXCR3-173), IL-17A (TC11-18H10), IL-22 (IL22JOP or 1H8PWSR), Tbet (ebio4B10), FOXP3 (FJK-165), ROR|$\gamma $|t (AFKJ5-9), and IFN|$\gamma $| (XMG1.2). For ILC3 staining, biotinylated antibodies were used against CD11b (M1/70), CD11c (N418), TCR|$\beta $| (H57-597), TCR|$\gamma \delta $| (UC7-13D5), CD3|$\varepsilon $| (145-2C11), CD8a (53-6.7), CD8b (YTS156.7.7), F4/80 (BM8), Ter119 (TER-119), NK1.1 (PK136), KLRG1 (2F1), CD19 (6D5), and Gr-1 (GR-1), then conjugated to streptavidin conjugated-PerCP-Cy5.5.

Isolation of Naïve CD4⁺ T Cells

Mice were sacrificed, and spleens were dissected and placed into sterile, cold R10. Spleens were crushed between 2 sterile fully frosted glass slides and filtered through 70 |$\mu $|m mesh. Red cells were ACK lysed, and the reaction was quenched with ice-cold MACS buffer, with single-cell suspension run through 40 |$\mu $|m filters. Samples were centrifuged, followed by gentle resuspension. CD4⁺ T cells were purified using CD4 (L3T4) microbeads according to manufacturer instructions (Miltenyi). Subsequently, the positively selected fraction was stained with fluorophore-conjugated antibodies to TCR|$\beta $|⁠, CD4, CD25, CD62L, and CD44. We sorted and collected naïve CD4⁺ T cells (TCR|$\beta $|⁺CD4⁺CD25⁻CD44^loCD62L^hi) at the UAB Comprehensive Flow Cytometry Core using the BD FACS ARIA II cell sorter. In 96-well flat-bottom plates, 1-2 × 10⁵ naïve CD4⁺ T cells were cultured in complete R10 (L-glutamine-free RPMI-1640 supplemented with 2 mm L-glutamine, 1% Pen-Strep, 50|$\mu $|m 2-ME, 1x NEAA, 10 mm HEPES, 1 mm Sodium Pyruvate, and 10% FBS). Using APC-free differentiation, we differentiated CD4⁺ T cells under Th17 conditions using plate-bound |$\alpha $|CD3|$\varepsilon $| and soluble |$\alpha $|CD28. All cultures were split on day 3 and harvested on day 5. For homeostatic Th17 conditions, as defined by O'Shea, Kuchroo, Littman, and Weaver, among others,^{42,  44,  83,  84} we stimulated cells with |$\alpha $|CD3|${\rm{\varepsilon }}$| (2.5–10 |$\mu $|g/mL, as indicated), |$\alpha $|CD28 (2 |$\mu $|g/mL), IL-6 (25 ng/mL), rhTGF|$\beta $|1 (2 ng/mL), |$\alpha $|IL-4 (10 |$\mu $|g/mL), and |$\alpha $|IFN|${\rm{\gamma }}$| (10 |$\mu $|g/mL). For pathogenic Th17 conditions, as defined by O'Shea, Kuchroo, Littman, and Weaver, among others, ^{42,  44,  83,  84} cells were stimulated with |$\alpha $|CD3|${\rm{\varepsilon }}$| (10 |$\mu $|g/mL), |$\alpha $|CD28 (2 |$\mu $|g/mL), IL-6 (25 ng/mL), IL-23 (20 ng/mL), IL-1|$\beta $| (20 ng/mL), |$\alpha $|IL-4 (10 |$\mu $|g/mL), and |$\alpha $|IFN|${\rm{\gamma }}$| (10 |$\mu $|g/mL). Where indicated, cultures were treated with 40 mm NaCl (Quality Biological), 100 nm FICZ (6-Formylindolo(3,2-b) carbazole, Enzo), 80 mOsm D-mannitol (Fisher Scientific), and untreated cultures were provided with the appropriate vehicle. |$\alpha $|CD3|${\rm{\varepsilon }}$| (145-2C11), |$\alpha $|CD28 (37.51), |$\alpha $|IL-4 (11B11), and |$\alpha $|IFN|${\rm{\gamma }}$| (XMG1.2) were purchased from Bio X cell. IL-6 (130-096-684), IL-1|$\beta $| (130-101-682), and rhTGF|$\beta $|1 (130-095-067) were purchased from Miltenyi Biotec, and IL-23 (34-8231-82) was purchased from eBioscience (Thermo Fisher Scientific). For cytokine determination, cultures were re-stimulated in 96-well U-bottom plates in complete media with 50 ng/mL PMA and 1 mg/mL ionomycin. After 1 h of incubation, brefeldin A (B7651, Sigma) was supplemented for the final 3-h to sequester cytokines. Cells were stained with amine-reactive live-dead dye, in cold PBS, to exclude dead cells, and stained for surface and intracellular markers as described above (using either eBioscience Foxp3 kit or BD Fix/Perm kit). Single-color controls, FMO, no-stimulation controls, and no-stain controls were utilized to establish compensation and unmixing parameters and gating, respectively.

Luminex Multiplexed Analyses

Plasma collected from mice was snap-frozen by liquid nitrogen immersion followed by storage at –80°C. Once needed, plasma was thawed on ice. Following manufacturer instructions, plasma was analyzed for cytokines using the MT17MAG47K-PX25 Milliplex Luminex panel.

FITC-Dextran Gut-Permeability Assay

Mice are transferred to a new cage and fasted for 4 h to limit coprophagy and access to chow. Mice are given 400 mg/kg body weight FITC-dextran (3–5 kDa FD4, Sigma Aldrich) dissolved in water by oral gavage. Next, mice continue fasting for an additional 4 h. Mice were anesthetized, and 50–75 μL of blood was collected from the retro-orbital sinus. Blood collected is spun at 10 000 g for 10 min at room temperature, and the plasma is diluted 1:4 with water. Sample FITC-dextran concentration was analyzed with a spectrophotometer (excitation 485 nm, emission wavelength 528 nm). Plasma from a mouse given water by oral gavage was used to subtract background autofluorescence.

Telemetry

Telemetry devices (Data Sciences, PA-C10, St. Paul, MN) were implanted into the left carotid artery of isoflurane anesthetized mice. Mice were allowed to recover for 10 days before collection of baseline telemetry readings and subsequent dietary interventions. Recordings were obtained in 2-min intervals every 10 min.

Segmented Filamentous Bacteria and H. hepat icus Determination

Fecal pellets from a cohort of UAB-housed male 8–12-week-old C57BL6/J mice were collected into separate sterile microfuge tubes and immediately snap-frozen. SFB and H. hepaticus colonization was qualitatively confirmed by analyzing DNA extracted from murine fecal pellets (fecal DNA extraction kit, Cat. No. D6010, Zymo Research) through PCR (SFB Primers: forward: 5′-ACGCTACATCGTCTTATCTTCCCGC-3′, reverse: 5′-TCCCCCAAGACCAAGTTCACGA-3′; H. hepaticus Primers: forward 5′ATGGGTAAGAAAATAGCAAAAAGATTGCAA3′, reverse: 5′CTATTTCATATCCATAAGCTCTTGAGAATC3′).

Colonic Tissue Sodium Content Analysis

Colons were placed in nitric acid-cleaned ceramic crucibles, weighed, and desiccated for 48 h in a 75°C drying oven. Dried tissues were then weighed, and the difference in weight prior to and after drying was considered water content as previously described.⁸⁵ Crucibles were then placed in a kiln for 40 h at 450°C. After cooling, ash was reconstituted in 5% nitric acid, and Na⁺ content was measured by atomic absorption (iCE 330 AAS, Thermo Fisher Scientific, Waltham, MA) and normalized to dry colon weight.

Statistical Analysis

Statistical analyses were performed with GraphPad Prism 9. Where appropriate and indicated, data were analyzed by paired or unpaired, two-tailed Student’s t-test, ANOVA (one- and two-way), or two-way repeated-measures ANOVA. Data are plotted as mean ± SEM. In cases in which non-normal distribution is found, appropriate non-parametric statistical analysis is completed (eg, Mann–Whitney, etc.). Tukey’s and Holms–Sidak multiple comparison’s tests were carried out for post hoc analysis. P < .05 was considered statistically significant. All data are pooled from, at least, 2–4 independent experiments, with each n indicating a biologically independent sample.

Results

HSD Depresses CD4⁺ T Cell Cytokine Expression

Three-week-old C57BL6/J males were purchased from Jackson Laboratory and accustomed to NSD and housed for 5–9 weeks, allowing them to acquire SFB (Figure S1A).^60–62 Th17 cells and their cytokines, which include IL-17A and IL-22, are considered essential to homeostasis, found abundantly in the intestinal tract, and mediate host protection.^86–88 Therefore, as the intestinal luminal content contains high amounts of salt in mice and humans that consume a HSD, we sought to determine if excess dietary salt alone or in tandem with salt-sensitive hypertension predisposes CD4⁺ T cells to produce IL-17A and IL-22. Thus, to evaluate the Th17 cell response to HSD, we fed mice either a NSD or a HSD. Further, we induced experimental salt-sensitive hypertension by conditioning mice with N(gamma)-nitro-L-arginine methyl ester (L-NAME), pan-nitric oxide synthase inhibitor, followed by feeding with HSD alone, recapitulating salt-sensitive hypertension.^{67,  68} Some experiments include conditioning with L-NAME followed with NSD fed mice to inform whether systemic nitric oxide synthase inhibition and short-term rise in blood pressure may be involved in Th17 cell disruptions. As HSD feeding alone does not promote a rise in blood pressure in C57BL6/J mice,^{89,  90} the experimental salt-sensitive hypertension model informs whether HSD alone is responsible for Th17 cell responses independent of augmented blood pressure or whether a systemic blood pressure response to HSD is required to disrupt Th17 cell functions (Figure 1A).

We interrogated the expression and population abundance of IL-17A- and IL-22-expressing CD4⁺ T cells in the colons of NSD-fed mice, HSD-fed mice, L-NAME-conditioned mice fed NSD, or L-NAME-conditioned mice fed a HSD (salt-sensitive mice). Using the gating scheme shown in Figure S1B, we determined the abundance of IL-17A- and IL-22-expressing CD4⁺ T cells in the colons of the 4 groups. We found that CD4⁺ IL-17A and IL-22 were markedly reduced in HSD-fed mice, with parallel responses by L-NAME-conditioned mice fed NSD, and salt-sensitive mice fed a HSD compared to NSD fed control mice (Figure 1B). These findings show that Th17 cells are sensitive to excess salt diet as well as a short-term rise in blood pressure. Total CD45⁺ and CD4⁺ T cell cLPL compartments were unaffected in HSD-fed, L-NAME-conditioned mice fed NSD, or salt-sensitive mice, suggesting a specificity for Th17 cells by excess salt diet and L-NAME conditioning (Figure S2A). The pro-inflammatory CD4⁺ T cell population, IL-17A⁺TNF|$\alpha $|⁺CD4⁺ T cells, was blunted in response to excess salt diet and in salt-sensitive mice; however, no significant difference was found in L-NAME-conditioned mice fed NSD (Figure S2B). Thus, suggesting that pro-inflammatory Th17 cells are sensitive to excess salt diet but not a rise in blood pressure.

Since Th17 cells can contribute to intestinal homeostasis,^{33,  91,  92} we reasoned that the experimental salt-sensitive mice may have altered colonic permeability. Indeed, we observed that salt-sensitive mice had significantly increased colonic permeability to FITC-dextran compared to NSD mice (Figure S2C). Many investigations have previously shown that HSD increases gut absorption of salt,⁹³ acute elevations in plasma sodium,^{94,  95} and increased excretion of urinary sodium.^{96,  97} We confirmed that sodium content in colonic tissue from HSD and L-NAME/HSD mice was elevated in the colon compared to NSD mice (Figure S2D). Collectively, these data indicate that salt-sensitivity is associated with reduced colonic Th17 cells, increased colonic permeability, and higher colonic sodium content.

Further, we assessed responses of retinoic acid-related orphan receptor gamma t⁺ (ROR|$\gamma $|t⁺) Foxp3⁻CD4⁺ T cells as well as forkhead box protein P3 (Foxp3)⁺ CD4⁺ T regulatory cells, and T-box transcription factor TBX21 (T-bet)⁺CD4⁺ T cells to excess salt diet and in salt-sensitive mice to determine the specificity of the Th17 response. ROR|$\gamma $|t⁺ is the master transcription factor for Th17 cells.⁹⁸ As expected, we found that ROR|$\gamma $|t⁺Foxp3⁻CD4⁺ T cells were decreased by excess dietary salt, L-NAME-conditioned mice fed NSD, and in salt-sensitive mice (Figure S3A), in agreement with the Th17 cell response. Whereas Foxp3⁺CD4⁺ T regulatory cells and T-bet⁺CD4⁺ T cells remained stable and were unaffected in all groups of mice (Figure S3A and S3B). Finally, Type 3 innate lymphoid cells (ILC3) were also unaffected in all groups of mice (Figure S3C). Thus, these observations further validate the specificity for the Th17 cell response to high dietary salt and salt-sensitivity.

We examined whether longer HSD feeding for 5 weeks would elicit a similar Th17 response as 2 weeks. Colonic IL-17A and IL-22 expressing CD4⁺ T cells were also decreased in response to 5 weeks of HSD, suggesting that acute and longer-term consumption of HSD promotes contraction of the Th17 cell pool without an apparent pro-inflammatory switch (Figure 1C). We also assessed whether the Th17-associated cytokines IL-17A, IL-22, GM-CSF, and TNF|${\rm{\alpha }}$| were disrupted by excess salt diet and experimental salt-sensitive hypertension. We found that circulating IL-17A, IL-22, GM-CSF, and TNF|${\rm{\alpha }}$| were unchanged by HSD, L-NAME-conditioned mice fed NSD, or salt-sensitive mice (Figure 1D) suggesting that these HSD responses are stable and most likely are not systemically mediated.

We determined whether the naïve CD4⁺ T cell compartment is sensitized to salt-sensitivity in vitro. We utilized naïve CD4⁺ T cells isolated from NSD-fed and salt-sensitive mice, polarized them under Th17-polarizing conditions in the presence or absence of additional NaCl in the cell media. Polarized Th17 cells from NSD-fed and salt-sensitive mice both had decreased IL-17A production in response to excess salt in vitro, indicating that the induced IL-17A T cell program remained stable among the naïve CD4 compartment (Figure 1E). Peyer’s patches serve as a key site for priming CD4⁺ T cells and for sensing luminal contents along the intestinal tract.^99–103 Therefore, we utilized the IL-17 ^fm/eYFP fate-mapping mouse¹⁰² and sought to confirm that HSD directly disrupts Th17 cells in PP, a key immunologic priming site. In concert with colonic and in vitro data, HSD reduced the Th17^efm/eYFP pool found in PP (Figure 1F). Taken together, we established that HSD depresses IL-17A and IL-22 expression by CD4⁺ T cells among the intestinal mucosa in the absence of circulating cytokine perturbations.^{74,  104–106}

At homeostasis, Th17 cells have an enriched non-pathogenic signature, which includes expression of the AhR, an established trans-activator of IL-17A among CD4⁺ T cells. Using next-generation RNA sequencing published by Matthias et al.,¹⁴ we evaluated unexplored candidate genes among polarized human Th17 cells exposed to high and normal salt. We specifically focused on candidate genes involved in T cell stability, AhR metabolite response, and migration. CYP1A1 (Cytochrome P450, family 1, subfamily A, polypeptide 1) catabolizes AhR ligands¹⁰⁷ suggesting that an imbalanced CYP1A1: AhR presence would lead to decreased AhR ligands. We noted that exposure to high NaCl in vitro led to an increased CYP1A1: AhR ratio in Th17 cells (Figure 1G). Further, we observed a significant increase in the expression of CCR6 and CXCL10 (C-X-C Motif Chemokine Ligand 10), a key ligand for CXCR3,^{108,  109} with marked depression in CXCR6, a key pathogenic marker, in Th17 cells exposed to high salt conditions (Figure 1G). Analysis of key regulatory pathways shows increased negative regulation of cell–cell adhesion, augmented cell migration programs, and reduced cytokine signaling pathways (Figure 1H). These data indicate that excess salt links human T cell programs leading to increased mobilization and disrupted cytokine expression.

AhR Activation Promotes Th17 Expansion During Excess Salt Intake and Exposure

Recent work has demonstrated that AhR activating ligands are depleted in patients that consume a HSD or display hypertension, and salt-sensitive animal models have further validated these observations.^{12,  57} As AhR is involved in promoting IL-17A expression,^{110,  111} we reasoned that reduced AhR ligand availability or decreased AhR activation may be responsible for blunted Th17 cell abundance. We assessed whether activating AhR in the setting of salt-sensitive hypertension promotes IL-17A-expressing CD4⁺ T cells in vivo and in vitro. L-NAME conditioned mice were either given HSD chow or HSD chow supplemented with the AhR activator I3C,⁷¹ followed by evaluation of the Peyer’s patch Th17 compartment to examine whether the Th17 compartment was augmented in AhR-activated HSD-fed mice (Figure 2A). We observed in the mice fed the HSD + I3C chow an expansion of Th17 cells among the Peyer’s patch compartment with augmented expression of CD69, a marker of early activation and mobilization arrest (Figure 2B). We next assessed if this expansion was a function of cellular proliferation by examining the expression of Ki-67, a marker of active proliferation. Interestingly, Ki-67 was not augmented in mice with the AhR-activating chow indicating that the IL-17A program is upregulated independent of cellular proliferation (Figure 2B). Importantly, these data indicate that activating AhR through dietary ligands represents a physiologic stimulus for expanding non-pathogenic Th17 cells among priming sites.

Th17 cells can generally be classified as homeostatic or pathogenic based on their differentiation pathway and functional phenotypes.⁷⁸ We designed in vitro studies to determine if directly activating AhR with 6-Formylindolo(3,2-b) carbazole (FICZ)^{112,  113} upregulates IL-17A among excess salt-treated homeostatic or pathogenic Th17 cells (Figure 2C). Consistent with our earlier observations, increased NaCl reduced IL-17A among CD4⁺ T cells under homeostatic polarizing conditions (Figure 2D). In contrast, naïve CD4⁺ T cells polarized under pathogenic conditions showed an expanded IL-17A response to increased NaCl (Figure 2D). To determine a direct effect of NaCl or a change in osmolality on polarizing Th17 cells under homeostatic or pathogenic conditions, we treated polarizing cells with mannitol, an osmotically active agent. Th17 cells polarized under homeostatic conditions were non-responsive to mannitol treatment, yet under pathogenic conditions Th17 cells were expanded. Thus, indicating a clear dichotomous response that homeostatic Th17 cells are sodium-dependent whereas pathogenic Th17 cells are osmotic-dependent (Figure 2D). Further studies are needed to determine the molecular mechanisms dictating these distinct responses.

Activating AhR in these homeostatic polarizing conditions even in the presence of excess NaCl promoted IL-17A expression (Figure 2D). Thus, suggesting that AhR responsiveness remains intact in homeostatic cultures despite excess salt conditions. However, naïve CD4⁺ T cells polarized under pathogenic conditions showed an expanded IL-17A in response to increased NaCl but lacked an AhR response (Figure 2D) confirming that the AhR program is inactive in Th17 cells polarized under pathogenic conditions. Taken together, these data suggest that activating AhR selectively expands homeostatic Th17 cells while limiting pathogenic Th17 cell activity. To further confirm regulation of IL-17A expression in the presence of AhR activation, we employed the IL-17A^fm/eYFP fate mouse and polarized naïve CD4⁺ T cells under homeostatic and pathogenic conditions and determined if activating AhR regulated IL-17A^fm/eYFP expression (Figure 2E). In agreement with our previous findings, Th17 cell fate was augmented in response to AhR activation with high salt exposure under homeostatic conditions (Figure 2F). Th17 cells polarized under pathogenic conditions were expanded in response to salt without additional IL-17A expression following AhR activation (Figure 2F). These in vitro data further validate key programmatic differences between homeostatic and pathogenic Th17 cells, with AhR activity being enriched among homeostatic Th17 cells.

Activating AhR Regulates Mobilization Induced By HSD in CD4⁺ T Cells

Key studies demonstrate that AhR activation also regulates immune cell mobilization.^{114,  115} Since we found that AhR activation regulates IL-17A expression in vivo and in vitro during excess salt conditions, we reasoned that activating AhR may regulate CD4⁺ T cell migration markers under high salt exposure. First, we examined whether the naïve CD4⁺ T cell compartment in NSD or salt-sensitive mice was primed with respect to expression of CXCR3,¹⁰⁹ a key migration marker among CD4⁺ T cells, during Th17 cell differentiation by in vitro experiments. We utilized naïve CD4⁺ T cells isolated from NSD-fed and salt sensitive mice, polarized under homeostatic conditions in the presence or absence of excess NaCl (40 mm) in the cell media. Homeostatic Th17 cells from NSD-fed and salt sensitive mice both responded to excess salt in which we observed upregulation of CXCR3 among Th17 cells in vitro (Figure 3A). Interestingly, Th17 cells from salt-sensitive mice had a significantly enhanced response to excess salt than the Th17 cells from NSD fed mice (Figure 3A). We also examined whether CCR6,^{116,  117} another key migration marker, would be upregulated in response to high salt under homeostatic Th17-polarizing in vitro conditions (Figure 3B). We observed that CCR6 expression was upregulated in Th17 cells isolated from either NSD-fed or HSD-fed mice when exposed to excess salt, and the response from HSD-fed mice was significantly higher than from NSD-fed mice in vitro (Figure 3B). We next examined whether the migration marker, CXCR3, was upregulated in Th17 cells from HSD fed mice in vivo (Figure 3C). Similarly, we found upregulation of CXCR3 among fated Th17 cells in the PP of HSD-fed mice (Figure 3C). These findings suggest that high salt directly reprograms mobilization potential for egress from the PP, thus, further studies determined the CD4⁺ T cell migration with HSD from the small intestine/PP.

We employed the CAG:: KikGR³³ transgenic mouse model, which expresses the photoconvertible fluorescent reporter protein, Kikume Green-Red (KikGR)55, to substantiate the HSD-induced CD4 + T cell migration from the small intestine to extra-intestinal sites. Upon illumination with 405 nm light, the KikGR protein undergoes irreversible photoconversion from its native green state to a red isoform (KikGreen to KikRed), which allows for cell tracking (Figure 3D). Furthermore, photoconversion is limited to tissues illuminated allowing for specific control in tissue-specific mobilization, and does not photoconvert circulating immune cells nor tissue protected by the sterile aluminum shield (Figure S4A). We examined if mice fed 5 weeks of HSD promote small intestinal (including PP) CD4⁺ T cell egress to mucosal and non-mucosal associated tissues. Intriguingly, HSD increased the presence of KikRed⁺ CD4⁺ T cells in the mesenteric lymph node (mLN), colonic lamina propria (cLP), and pooled caudal/iliac LN, all key mucosal-associated tissues (Figure 3E). The mLN and caudal/iliac LN specifically drain sites within the colon. Whereas the frequency of KikRed⁺ CD4⁺ T cells remained stable with 5 weeks of HSD in the inguinal LNs (ingLNs), spleen, kidneys, and whole blood (all non-mucosal-associated tissues) (Figure S4B). Whether Th17 cells are the primary CD4⁺ T cell subset that is migrating remains to be determined. Regardless, these data provide a proof of concept that HSD stimulates CD4⁺ T cell migration between mucosal tissues. Altogether, excess salt exposure in vitro and HSD in vivo sensitizes and promotes CD4⁺ T cell migration as evidenced by upregulation of key migration markers and migration to remote mucosal-associated tissues.

We sought to determine if activating AhR reprograms CD4⁺ T cell migration markers altered by HSD. We reasoned that activating AhR may blunt the expression of CXCR3 and CCR6 under high salt conditions among CD4⁺ T cells. We utilized polarized naïve CD4⁺ T cells under homeostatic and pathogenic Th17 in vitro conditions in the presence of high salt or high salt coupled with the AhR activator, FICZ (Figure 3F). Indeed, activating AhR with FICZ under high salt conditions normalized the frequency of CXCR3 expressing Th17 cells exposed to high salt under homeostatic conditions (Figure 3F). In contrast, neither high salt nor AhR activation with FICZ affected CXCR3 expressing Th17 cells under pathogenic conditions (Figure 3F). These findings further validate a role for AhR in homeostatic Th17 function in the context of high salt.

To further confirm AhR activation as a regulatory focus for effector CD4⁺ T cells under high salt conditions, naïve CD4⁺ T cells were isolated from mice fed 2 weeks of NSD or HSD and polarized under homeostatic and pathogenic conditions with excess salt in the absence or presence of FICZ in vitro (Figure 3G) and examined whether CCR6 expression was regulated under these conditions. We found that 2 weeks of HSD led to increased CCR6⁺ homeostatic Th17 cells compared to mice fed a NSD (Figure 3G). Interestingly, both NSD and HSD fed mice saw an increase in CCR6⁺ homeostatic Th17 cells with excess salt in the media that was attenuated with FICZ treatment (Figure 3G). In contrast, Th17 cells polarized under pathogenic conditions did not demonstrate regulation of CCR6 expression with AhR activation. Taken together, these findings indicate that activating AhR may offer a physiological avenue to limit inappropriate CD4⁺ T cell egress from the gut by blunting migration in response to excess salt.

AhR Activation Blunts the Development of Salt-Sensitive Hypertension

Foundational studies demonstrate control of blood pressure by immune cells, and notably T cells.^{45,  118} However, whether AhR activation plays a key role in salt-sensitive hypertension is not established. We examined if activating AhR supplants HSD-dependent increased blood pressure in a mouse model of salt-sensitivity. We provided HSD without or with the diet-derived, high-affinity AhR pro-ligand I3C to L-NAME conditioned salt-sensitive mice and evaluated the blood pressure response by telemetry (Figure 4A). All mice displayed a diurnal rhythm in blood pressure and heart rate (Figure 4B–I) with averages of light and dark periods denoted by the alternating white and gray areas. As expected, mice fed chow with HSD after L-NAME conditioning displayed increased salt-sensitive hypertension, as observed with significantly elevated mean arterial, systolic, and diastolic blood pressure (MAP, SBP, and DBP, respectively) during the light and dark periods (Figure 4B–G). Interestingly, we found that mice fed chow with HSD supplemented with the AhR activator I3C after L-NAME conditioning did not develop salt-sensitive hypertension and showed a salt-resistant blood pressure phenotype (Figure 4B–G) but without an alteration in heart rate (Figure 4H–I). It is known that metabolism of AhR ligands, specifically I3C, occurs in the intestine with limited I3C-derived metabolites found systemically.⁷⁰ While more research is needed, these findings suggest that the control of salt-sensitive hypertension by AhR-activating ligands may be limited to the intestinal tract.

Discussion

Here, we demonstrate that activating AhR directs homeostatic Th17 cells amid excess salt as well as restrains the development of experimental salt-sensitive hypertension. Our study unraveled novel non-canonical salt-sensitive CD4⁺ T cell responses highlighting a central role for AhR in blunting the salt-sensitive Th17 cell fate. Th17 cells have extreme plasticity, and the environment determines their homeostatic or pathogenic fate.³³ Distinct homeostatic Th17 cells are present in the intestine as well and are critical for barrier integrity.^{33,  119,  120} While several groups have defined a pathogenic phenotype of Th17 cells that is present in the intestine with an increase in IL-23 signaling.^{120,  121} Our study found that direct AhR activation in polarized homeostatic and pathogenic Th17 cells shows divergent responses to high salt analogous to in vivo conditions. We also discovered salt-responsive CD4⁺ T cell mobilization as well as increased expression of migration markers with high salt in both in vivo and in vitro conditions. AhR activation in the presence of excess salt leads to blunted migration marker expression specifically in homeostatic Th17 cells. Furthermore, our study revealed that AhR activation via dietary supplementation with I3C blocks the development of experimental salt-sensitive hypertension. In summary, activating AhR fosters homeostatic control of effector CD4⁺ T cells in the presence of excess salt and prevents the development of salt-sensitive hypertension.

Numerous mechanisms have been proposed as cooperative factors in the response to excess dietary salt and salt-sensitive hypertension.^122–124 Several contrasting studies have demonstrated that excess dietary salt does not elicit an inflammatory response or may lead to immunosuppression.^{24,  125} Kleinewietfeld et al. and Wu et al. published seminal studies outlining a role for excess salt in promoting Th17 cells in mice with induced myelin oligodendrocyte glycoprotein (MOG) peptide 35-55 (MOG_35-55)-dependent experimental autoimmune encephalomyelitis, a mouse model of multiple sclerosis.^{13,  30} Wilck et al. elegantly demonstrated an association between the Th17 compartment expansion amidst salt-sensitive hypertension¹² and Madhur and colleagues show a role for IL-17A to worsen hypertension.^{24–26,  28} While our study found mucosal Th17 cell contraction in response to HSD and salt-sensitive mice, the murine strain differences or microbiota differences may underlie the potential discrepant study results. Notably, Wilck et al. used specific pathogen-free/virus antibody-free FVB/N mice acquired from Charles River at 12 weeks of age,¹² while our study used 8–12-week-old C57BL6/J mice from Jackson Laboratory acquired at 3 weeks old, which were subsequently colonized with SFB during the post-weaning reaction window. Indeed, Robertson et al. show the importance of acclimatization protocols and breeding strategies to standardize microbiota for reproducible results.¹²⁶ Thus, differences in the murine genetic background and microbial imprinting between our study and the previous reports may explain the key discrepancies.^127–130

We observed that the in vivo response to a HSD is reflected in the in vitro response with Th17 cells cultured under homeostatic conditions (TGF|$\beta \ $| and IL-6). Ghoreschi et al. and others have elegantly demonstrated key programmatic differences between homeostatic (or protective) and pathogenic Th17 cells.^{42, 131, 132} Among homeostatic Th17 cells, AhR activation was enriched, whereas pathogenic Th17 cells show prominent T-bet induction.^{43,  44} Therefore, we hypothesized that activating AhR may expand Th17 cells despite high salt conditions. Indeed, in vitro and in vivo AhR activation in the presence of high salt promoted Th17 cells. Recent clinical trials evaluating the efficacy of anti-IL-17A treatment in patients with colitis noted local and systemic clinical deterioration, presumably through disruption of local indispensable Th17-dependent homeostatic mechanisms.^{21,  33} Homeostatic Th17 cells were identified in the intestines and are studied in their role related to the intestinal barrier.^{33,  119} In line with these studies, we found that the L-NAME-induced model of salt-sensitivity showed increased intestinal permeability as well as higher sodium content in the colon. However, identifying a role for homeostatic Th17 cells in other organs and tissues is yet to be described. Thus, further research is needed to determine whether homeostatic Th17 cells play a role in blood pressure responses to dietary interventions.

Key facets of T cell function, such as mobilization and cytokine expression, and how these are influenced by excess salt are not well established.¹⁸ We found that small intestinal-derived CD4⁺ T cells migrate in response to high salt suggesting disruption of homeostasis. In vitro models corroborated that excess salt heightens CD4⁺ T cell function and migration directly. Interestingly, multiple groups have demonstrated that autoimmune experimental models promote migration of small intestine-derived CD4⁺ T cells.¹³³ We also demonstrate that effector CD4⁺ T cells upregulate the canonical migration marker CXCR3, in vitro and in vivo, with high salt. Intriguingly, the population of CD4⁺ T cells fated to express IL-17A is reduced in PP. Using varied in vitro conditions, we found that specific T cell priming conditions dictate responsiveness to excess salt and AhR activation. Specifically, a reduction in CXCR3 following AhR activation suggests sustained homeostatic Th17 programs. In brief, when excess salt is present in the diet or in cell culture conditions, CD4⁺ T cells are primed to mobilize with tandem inactivation of the IL-17A program. Intriguingly, the loss of IL-17A and increased CXCR3 among CD4⁺ T cells may reflect the reprogramming of pathogenic ex-Th17 cells.⁷⁸ Collectively, this study provides primary evidence for HSD-induced migration of intestinal-derived CD4⁺ T cells.

Matthias et al. revealed key differences in the response to high salt between human induced homeostatic and pathogenic Th17 cell fates.¹⁴ Using the Matthias database of human transcriptomics with single-cell RNA sequencing, we identified Th17 candidate genes that approximate the murine responses to excess salt found in our study. For example, increased CCR6 and CXCL10 with reduced CXCR6 may reflect preponderance to mobilize with high salt. Further, an augmented CYP1A1: AhR ratio in human Th17 cells exposed to high salt may indicate decreased AhR ligand availability for Th17 stability due to CYP1A1-dependent ligand metabolism.^{134,  135}

Critical studies yielded that exogenous administration of IL-17A augments blood pressure,^{24,  26} whereas, in some cases, infusion with anti-IL-17A did not reduce blood pressure.¹³⁶ Subsequently, Krebs and Lange et al. demonstrated that loss of IL-17A and IL-23 accelerates end-organ injury in deoxycorticosterone acetate + angiotensin II-induced hypertension.^{17,  136} L-NAME-induced salt-sensitive hypertension, which presents with lower circulating angiotensin II levels, is distinct from angiotensin II-dependent hypertension with divergent mechanistic underpinnings.^{137,  138,  139} Harrison’s laboratory first demonstrated a linkage between L-NAME-induced salt-sensitive hypertension and increased non-lymphoid Th17 cells.⁶⁸ Our study, however, finds blunted Th17 cell responses in the colon with L-NAME induced salt sensitive hypertension as well as with high salt only and L-NAME conditioning. Our study focused on the mechanisms of how high salt conditions influence Th17 cell differentiation and activation. It is interesting that L-NAME conditioning itself has a significant effect on Th17 cells; however, we focused on the high salt and salt-sensitive dependent mechanisms. Thus, a limitation of our study is the lack of understanding of whether L-NAME conditioning and HSD work through the same mechanism affecting the Th17 cell fate.

Overall, our current work highlights the diversity and complexity of Th17 responses, especially with regard to AhR activation, in the presence of high salt and in the development of salt-sensitive hypertension. These data show that AhR activation via dietary supplementation with I3C reduces blood pressure in an L-NAME-induced salt sensitive model. Several investigations^140–142 show distinct differences in the systemic activation of AhR by I3C or when AhR activation is limited to the intestine with dietary I3C supplementation similar to our study. During digestion, I3C is metabolized into AhR ligands. I3C-supplemented diet in the non-obese type 1 diabetic mouse model, NOD mice, leads to strong AhR activation in the small intestine but minimal systemic AhR activity.¹⁴⁰ Dietary I3C did not alter T helper cell differentiation in the spleen or pancreatic draining lymph nodes. However, dietary I3C increased the percentage of CD4⁺RORγt⁺Foxp3⁻ (Th17 cells) in the lamina propria, intraepithelial layer, and PP of the small intestine of NOD mice. The immune modulation in the gut was accompanied by alterations to the intestinal microbiome, with changes in bacterial communities observed within 1 week of I3C supplementation.¹⁴⁰ Our study indicates a potential role of colon-derived Th17 cells associated with driving the blood pressure response; however, these data are not conclusive. Although, it is tempting to speculate that a basal reservoir of Th17 cells within the intestinal compartment is required and indispensable for local homeostasis and potentially systemic blood pressure regulation. Further research will test the hypothesis that the activation of AhR in the colon is driving the blood pressure response.

In conclusion, these findings show that in the setting of experimental salt-sensitivity, activating AhR promotes T cell stability and controls the sensitivity of blood pressure response. Experimentally, the deletion of AhR in mice seems an obvious model to study the relationship between AhR and salt-sensitive hypertension. However, the absence of AhR, systemically or cell-specific, leads to deleterious physiological consequences. For example, loss of CD4⁺ T cell-specific AhR leads to disruption in the Th17 differentiation, inability for Th17-transdifferentiation, and inability to sustain epithelial integrity.¹⁴³ Deletion of AhR among epithelial cells results in intestinal epithelial stem cell disruption leading to unrestricted proliferation and impaired differentiation.^144–146 Myeloid-specific AhR deletion leads to inappropriate intestinal morphogenesis, epithelial differentiation, and decreased epithelial integrity.¹¹⁵ Studying salt-sensitivity through constitutive AhR knockout mouse models would most likely be confounded by these physiological consequences. The pathogenesis of salt-sensitivity is not restricted to adaptive immune mechanisms but also relies on renal, endothelial, and/or neural signaling. Thus, cell-specific inducible AhR knockout mouse models would be a valuable tool for future research efforts.

Perspectives

Blunting salt-sensitive hypertension through specific dietary modifications is sought after clinically. Guidance by the American Heart Association Life’s Essential 8 recommends sustained intake of vegetables, such as broccoli, which are enriched in AhR activating ligands.⁶ Clinically, the recent generation of designer indoles with high affinity and high potency for AhR in therapeutic ranges provides a promising tool in targeting intestinal disruptions, hypertension, and immune cell dysfunction.¹⁴⁷ Overall, this work provides evidence that increasing high-affinity AhR ligand availability provides a diet-based therapeutic platform to minimize mucosal disruption as well as reduce the hypertensive response to excess dietary salt intake.

Acknowledgements

We appreciate the efforts of Vidya Sagar Hanumanthu, Associate Director of the UAB Flow Cytometry and Single Cell Core. We also gratefully acknowledge the scientific discussions and technical assistance by Ryan McMonigle. Further, the UAB/UCSD O'Brien Core Center for Acute Kidney Injury Research provided excellent surgical support with telemetry implantation.

Author Contributions

Patrick A. Molina (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing), Claudia J. Edell (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing), Luke S. Dunaway (Investigation, Methodology, Writing – review & editing), Cailin E. Kellum (Data curation, Investigation, Methodology, Writing – review & editing), Rachel Q. Muir (Data curation, Investigation, Writing – review & editing), Melissa S. Jennings (Investigation, Methodology, Writing – review & editing), Jackson C. Colson (Investigation, Methodology, Validation, Writing – review & editing), Carmen De Miguel (Validation, Visualization, Writing – review & editing), Megan K. Rhoads (Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing), Ashlyn A. Buzzelli (Methodology, Writing – review & editing), Laurie E. Harrington (Resources, Validation, Writing – review & editing), Selene Meza-Perez (Investigation, Resources, Writing – review & editing), Troy D. Randall (Methodology, Resources, Writing – review & editing), Davide Botta (Investigation, Methodology, Resources, Writing – review & editing), Dominik N. Müller (Methodology, Resources, Writing – original draft, Writing – review & editing), David M. Pollock (Methodology, Project administration, Resources, Funding acquisition, Visualization, Writing – review & editing), Craig L. Maynard (Conceptualization, Project administration, Supervision, Validation, Writing – review & editing), and Jennifer S. Pollock (Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing)

Funding

This research was supported by grants from the National Institues of Health: F31HL151264, T32DK116672, and T32GM8361 to P.A.M., F31HL167626, T32GM811135, and TL1TR003106 to C.J.E., F31HL149235 to L.S.D., K01HL145324 to C.D.M.; F31HL165863 to C.E.K., R01DK125870 to L.E.H., TL1DK139566 to M.K.R., R01DK118386 to C.L.M. P01HL136267, P01HL158500, and R01DK134562 to D.M.P. and J.S.P.; Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Projektnummer 394046635-SFB 1365 and DFG-SFB1470/A06, by the Deutsches Zentrum für Herz-Kreislauf-Forschung (DZHK, 81Z0100106) and by the BMBF (The Federal Ministry of Education and Research), Foerderkennzeichen 01EJ2202D (TAhRget consortium) to D.N.M.; Crohn's and Colitis Foundation Research Initiatives Award 558 420 to C.L.M. and J.S.P.

Conflict of Interest

DMP holds the position of editor-in-chief of FUNCTION and is blinded from reviewing or making decisions for the manuscript.

References

Author notes