Eicosanoid-regulated Myeloid ENaC and Isolevuglandin formation in Human Salt-sensitive Hypertension Immune ENaC Regulation by EETs in Salt Hypertension

Abstract

Background:

The mechanisms by which salt increases blood pressure in people with salt sensitivity remain unclear. Our previous studies found that high sodium enters antigen-presenting cells (APCs) via the epithelial sodium channel (ENaC) and leads to the production of isolevuglandins (IsoLGs) and hypertension. In the current mechanistic clinical study, we hypothesized that ENaC-dependent IsoLG-adduct formation in APCs is regulated by epoxyeicosatrienoic acids (EETs) and leads to salt sensitive hypertension in humans.

Methods:

Salt sensitivity was assessed in 19 hypertensive subjects using an inpatient salt loading and depletion protocol. IsoLG-adduct accumulation in APCs was analyzed using flow cytometry. Gene expression in APCs was analyzed using cellular indexing of transcriptomes and epitopes by sequencing (CITE-Seq) analysis of blood mononuclear cells. Plasma and urine EETs were measured using liquid chromatography-mass spectrometry.

Results:

Baseline IsoLG⁺ APCs correlated with higher salt sensitivity index. IsoLG⁺ APCs significantly decreased from salt loading to depletion with an increasing salt sensitivity index. We observed that human APCs express the ENaC-δ subunit, SGK1 and cytochrome P450 2S1. We found a direct correlation between baseline urinary EET 14–15 and salt sensitivity index, while changes in urinary EET 14–15 negatively correlated with IsoLG⁺ monocytes from salt loading to depletion. Co-incubation with EET 14–15 inhibited high-salt induced increase in IsoLG⁺ APC IsoLGs.

Conclusions:

IsoLG formation in APCs respond to acute changes in salt intake in salt-sensitive but not salt-resistant people with hypertension, and this may be regulated by renal EET 14–15. Baseline levels of IsoLG⁺ APCs or urinary EET 14–15 may provide diagnostic tools for salt sensitivity without a protocol of salt loading.

Introduction

Salt-sensitivity of blood pressure is a continuous trait characterized by changes in blood pressure that parallel salt intake and affects approximately 50% of people with and 25% of people without hypertension¹. Notably, salt sensitivity is a risk factor for cardiovascular morbidity and mortality, independent of, yet as powerful as hypertension².

Inflammation plays a critical role in the development of salt-sensitive hypertension and its associated end-organ damage^3–5. We have previously shown in mouse models that antigen-presenting cells (APCs), including dendritic cells and monocytes, are activated by high-salt microenvironments. Sodium enters APCs via epithelial sodium channel (ENaC) and triggers the production of isolevuglandins (IsoLGs). IsoLGs are highly reactive products of lipid peroxidation that rapidly adduct to proteins, forming neo-antigens. Accumulation of these immunogenic IsoLG-adducts activates APCs, promoting the secretion of pro-inflammatory cytokines and inflammation that results in the development of hypertension⁶. We also found that salt-induced expression of ENaC was promoted by salt-sensing kinase serum/glucocorticoid kinase 1 (SGK1)⁷.

Higher sodium accumulation in tissue is clinically associated with hypertension⁸ and end-organ damage, including left ventricular hypertrophy and endothelial dysfunction⁹. Whether the IsoLG response is associated with blood pressure response to salt in humans remains unknown, especially since the degree of salt-induced IsoLG production is highly variable among individuals.

Epoxyeicosatrienoic acids (EETs) are metabolites of arachidonic acid by cytochrome P450 (CYP) epoxygenases, including cytochrome P450 2S1 (encoded by CYP2S1) and have four regioisomers: 5,6-EET, 8,9-EET, 11,12-EET, and 14,15-EET. Among various biologic activities, EETs are potent inhibitors of ENaC and are suggested to regulate salt-sensitivity^10,11. Salt-sensitive rats have reduced urinary excretion of EETs and renal expression of epoxygenase¹², while inhibition of the renal epoxygenase induces salt-sensitivity in animal models¹³. Moreover, previous clinical studies found that the two pools of EETs, systemic and renal, are regulated differentially in salt-sensitive and salt-resistant normotensive individuals¹⁴. It is not known whether EETs inhibit ENaC activity and subsequent IsoLG production in APCs and whether such inhibition may play a role in salt-sensitive hypertension.

To address the questions above, we utilized a rigorous inpatient protocol to assess salt sensitivity in people with hypertension and investigated if IsoLG-adduct formation in APCs promotes salt sensitivity through an EET-regulated, ENaC-dependent pathway. We further studied salt-induced changes in plasma levels of APC and T-cell derived cytokines, explored gene expression profiles of ENaC subunits and ENaC-regulatory genes SGK1 and CYP2S1 in APCs and assessed salt-induced changes based on salt sensitivity.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request. The expanded Methods section is available in the Supplemental Material.

Study population and protocol

We recruited nineteen participants with hypertension for a rapid inpatient protocol of salt loading and depletion to assess salt sensitivity as described in detail in the Supplemental Material, and outlined in Figure 1A. Written informed consent was obtained from all participants before recruitment. The study protocol was approved by the Institutional Review Board of Vanderbilt University Medical Center and performed in accordance with the Declaration of Helsinki.

Figure 1. The study protocol and changes in urinary sodium and potassium excretion and blood pressures at baseline (B), salt loading (SL) and salt depletion (SD).

(A), Diagram showing the inpatient protocol of SL (day 1) and SD (day 2). Changes in 24-hour urinary sodium excretion (UNa⁺V) (B), 24-hour urinary potassium excretion (UK⁺V) (C), urinary sodium potassium ratio (Na⁺/K⁺) (D), and blood pressures (E) in each subject at baseline, SL and SD. ΔSBP, ΔDBP, ΔMAP and ΔPP represent the change from SL to SD (F). 1^st (blue line) and 3^rd (red line) salt sensitivity index tertiles represent the least and the most salt sensitive subjects, respectively.

Salt sensitivity index was defined as the mmHg change in daytime SBP from salt loading to depletion. A greater decrease in SBP from salt loading to depletion indicated a higher salt sensitivity index. The participants with the lowest salt sensitivity index were included in the lowest tertile (Tertile 1), while the participants with the highest salt sensitivity index were included in the highest tertile (Tertile 3). The tertiles were set prior to statistical analysis.

Results

Participant Characteristics

The demographics of all the participants are shown in Table 1. Nineteen (19) people with hypertension completed the protocol and were included in the analysis. The mean (±SEM) age was 51.9±1.7 years; 10 of the participants were female and 6 were Black. Mean (±SEM) SBP and DBP were 137.7±3.1 and 85±2.2 mmHg, respectively. All participants had normal baseline serum creatinine.

Table 1.

Baseline Characteristics of the study participants

Characteristic a	All n=19	Tertile 1 n=6	Tertile 2 n=7	Tertile 3 n=6
Age (years)	51.9±1.7	51.7±2.5	55.1±1.8	48.5±4.2
Female sex, n (%)	10 (52.6)	3 (50)	4 (57.1)	3 (50)
White race, n (%)	13 (68.4)	4 (66.7)	5 (71.4)	4 (66.7)
BMI (kg/m2)	32.7±2.3	31.5±4.0	32.2±3.5	34.7±4.8
SBP (mmHg)	137.7±3.1	146.3±7.8	132.9±1.8	134.8±5.0
DBP (mmHg)	85±2.2	89.9±5.6	85.7±3.3	84.3±2.2
Urinary Na+ Excretion (mmol/day)	150.4±13.7	155.8±18.9	155.7±30.6	138.8±20.4
Urinary K+ Excretion (mmol/day)	59.2±7.7	63.8±8.8	66.8±19.3	45.5±4.5
Serum creatinine (mg/dL)	0.87±0.03	0.88±0.05	0.83±0.06	0.92±0.06
PRC (ng/L)	9.3±3.1	9.0±4.3	11.6±7.6	7.1±3.1
Aldosterone (ng/dL)	12.1±2.2	10.8±4.5	13.2±4.7	12.2±2.2
ARR (ng/L/ng/L)	28.9±7.1	38.4±18.7	19.9±4.5	30.0±12.5

ARR, aldosterone renin ratio; BMI, body mass index; DBP, diastolic blood pressure; PRC, plasma renin concentration; SBP, systolic blood pressure.

^aData for continuous variables are presented as mean ± SEM.

Salt sensitivity index was defined as the change in SBP from salt loading to depletion, such that a positive salt sensitivity index shows a decrease in blood pressure from salt loading to depletion. Because salt-sensitivity of blood pressure is a normally distributed trait, we analyzed the data as a continuum instead of using arbitrary blood pressure cut-offs. However, we noticed that participants responded to salt loading/depletion in three major trends and thus represented them in three tertiles based on the salt sensitivity index. The lowest tertile (Tertile 1) included the least salt sensitive participants, while the highest tertile (Tertile 3) included the most salt sensitive participants. The mean (±SEM) salt sensitivity index was −4.9±1.3, 1.9±0.6 and 10.7±1.9 in Tertile 1,2 and 3, respectively.

Accuracy of the Salt Loading and Depletion Protocol

Blood pressures and urinary solute excretions at baseline, salt loading, and depletion along with changes across tertiles are shown in Figure 1. The accuracy and efficacy of salt loading and salt depletion protocols were confirmed with appropriate changes in 24-hour urinary Na⁺ excretion and urine Na⁺/K⁺ ratio (Table S1). The mean 24-hour urinary Na⁺ excretion (±SEM) increased from a baseline of 150.4±13.7 to 350.8±27.9 with salt loading and subsequently decreased to 73.5±13.9 after salt depletion. The mean urinary Na⁺/K⁺ ratios were 2.9±0.3, 6.8±1.6, and 1.9±0.3 at baseline, salt loading, and depletion, respectively, indicating that renal ENaC activity appropriately decreased in response to salt loading and increased in response to salt depletion.

Effects of Salt Loading and Salt Depletion on Blood Pressure and Baseline Characteristics

Table S1 shows the mean blood pressures and various characteristics of all participants at baseline, salt loading, and depletion. The mean salt sensitivity index (±SEM) was 2.5±1.6 mmHg. The salt sensitivity index was 17.3 mmHg and −7.3 mmHg in the most salt-sensitive and most salt-resistant participant, respectively.

Salt-induced changes in plasma renin concentration, aldosterone, and aldosterone renin ratio (ARR) are presented in Table S1. There was a trend towards decreased plasma aldosterone levels from baseline to salt loading and increased levels from salt loading to depletion in all tertiles. The changes in plasma renin, aldosterone, or ARR were not different across the tertiles and did not correlate with the salt sensitivity index.

IsoLGs Decrease with Salt Depletion in Salt-Sensitive People

To investigate the association of IsoLG-adduct accumulation in APCs with salt sensitivity, we determined the percent of dendritic cells and monocytes with IsoLG-adducts (IsoLG⁺) on each day of the protocol using flow cytometry. Figure 2 shows the experimental strategy and the salt-induced changes in IsoLG⁺ APCs. The baseline IsoLG⁺ APCs were highest among the lowest salt sensitivity tertile (Figure 2C) and decreased with increasing salt sensitivity index (Figure 2D), which was significant in classical and non-classical monocytes (Table S2). Hence, baseline levels of IsoLG⁺ decreased with increasing salt sensitivity.

Figure 2.

(A), Diagram showing the flow cytometry protocol. (B), Representative scatter plots showing flow cytometry gating strategy to identify IsoLG-adducts in dendritic cells (DCs) and Classical, Intermediate, and Non-Classical Monocytes (C), Percents of IsoLG⁺ APCs in each subject at baseline (B), salt loading (SL) and salt depletion (SD) are depicted on left. The mean percent of IsoLG⁺ APCs ±SEM in each tertile is depicted on right. 1^st (blue line) and 3^rd (red line) salt sensitivity index tertiles represent the least and the most salt sensitive subjects, respectively. (D), Correlations of salt sensitivity index with baseline percent of IsoLG⁺ APCs and difference in IsoLG⁺ APCs between salt loading and depletion (SL-SD).

The mean percent of IsoLG⁺ APCs among all participants decreased from salt loading to depletion (Table S1). To determine whether salt-induced changes in IsoLG⁺ APCs were dependent on salt sensitivity, subgroup analysis was performed across tertiles (Table S3). IsoLG⁺ in dendritic cells and classical and nonclassical monocytes significantly decreased from salt loading to depletion in the highest tertile of salt sensitivity. IsoLG⁺ APCs did not significantly change among the lower tertiles. Analysis of the linear relationship between IsoLG-adduct accumulation and salt sensitivity index revealed that IsoLG⁺ dendritic cells and non-classical monocytes demonstrated a greater decrease from salt loading to depletion with increasing salt-sensitivity (Figure 2 and Table S2). These results depict that changes in IsoLG-adduct accumulation mirror changes in blood pressure and salt balance among individuals with the highest salt sensitivity, implying that acute salt-induced IsoLG formation in APCs may contribute to salt-sensitive hypertension in humans.

Plasma Cytokine Levels Change with Salt Depletion in an IsoLG-Dependent Manner

To evaluate the effects of salt on the systemic inflammatory response, we measured plasma levels of APC and T-cell-derived inflammatory cytokines at baseline, salt loading, and depletion (Table S4). Plasma levels of cytokines at baseline and salt loading did not differ statistically. Plasma levels of fractalkine, IFN-γ, MIP-3α, IL-1β, IL-7, IL-12, IL17A, IL-21, and IL-23 were significantly lower, and plasma levels of IL-2 and IL-8 were significantly higher after salt loading compared to salt depletion. The changes in fractalkine, IFN-γ, IL-2, IL-10, IL-12, IL-13, IL-17A and IL-8 significantly correlated with changes in IsoLG-adduct accumulation in classical monocytes. Changes in IsoLG-adduct accumulation correlated with IL-8 in nonclassical monocytes, and with IL-10 and IL-13 in intermediate monocytes. The relationship between these plasma cytokines and IsoLG-adducts in all APCs is presented in Table S5.

Expression of ENaC subunits and Regulatory Genes CYP2S1 and SGK1 in Human Immune Cells

In order to investigate whether expression of ENaC subunits and ENaC regulator genes in APCs mirror changes in salt intake and relate with salt sensitivity, we profiled the transcriptomes of PBMCs at baseline, salt loading and salt depletion as outlined (Figure 3A). Different cell types were identified by combining surface protein expression and the transcriptomic signatures as previously published²⁹ (Figure 3B). For representation purposes, three subjects with the highest salt sensitivity index were presented as salt sensitive, and three subjects with the lowest salt sensitivity index were presented as salt resistant. We found that gene expression changed differently in the cells of salt-resistant versus salt-sensitive individuals during salt loading and depletion (Figure 3C). While the number of gene expression changes from baseline to salt-loading was greatest among naïve B cells and dendritic cells in salt-resistant participants, NK and NK/CD8⁺ T cells exhibited the greatest number of gene changes among salt-sensitive people. Gene expression in all cells was largely resistant to change from salt-loaded to salt-depleted state in the salt-resistant individuals. In striking contrast, there were marked changes in gene expression in the immune cells of salt-sensitive people, especially among monocytes, from salt-loaded to salt-depleted state.

Figure 3.

(A), Experimental design of the human peripheral blood mononuclear cells CITE-Sequence Analysis. (B), UMAP representation of different immune cell type clusters identified by antibody-derived tags andpercentage of differential gene expression between baseline and salt loading (B-SL) and between salt loading and depletion (SL-SD) are represented in select salt sensitive (SS) and salt resistant (SR) subjects. Three subjects with the highest salt sensitivity index were represented as salt sensitive, and three subjects with the lowest salt sensitivity index were represented as salt resistant. (C), Gene expression of ENaC subunits α, β, γ and δ (SCNN1A, SCNN1B, SCNN1G and SCNN1D), ASIC subunits 1 and 2, SGK1 and EET-related genes (soluble epoxide hydrolase (EPHX2) and epoxygenases CYP2S1, CYP2C9, CYP2C8 and CYP2J2) in different cell clusters are presented in UMAPs. (D), UMAP representation of gene expression of SCNN1D, CYP2S1, and SGK1 are shown at B, SL and SD in select salt sensitive and salt resistant subjects. Dot plot shows the gene expressions for the four ENaC and ASIC subunits, SGK1 and EET-related genes in different cell clusters. (E), Correlation between change in gene expression of SGK from salt loading to salt depletion and salt sensitivity index. Analyzed using linear regression and 2-tailed Pearson test. (F), Experimental design for bulk RNA sequencing. Human monocytes from 11 healthy subjects were isolated using magnetic separation and cultured in either normal salt (150mmol/L NaCl) or high salt (190 mmol/L NaCl) for 72 hours. Expression differences in EET-related genes in response to normal salt (NS) and high-salt (HS) treatment are presented in reads per kilobase per million mapped reads (RPKM). A heat map of RNA transcripts illustrating gene expression differences in response to normal salt (NS) and high-salt (HS) treatment is provided. Bright pink, bright blue, and white represent the highest, lowest, and median read values, respectively. Rows represent individual genes, and columns represent individual samples.

Expression of the four ENaC subunits and two ASIC subunits in cell clusters are presented in different UMAPs in Figure 3D. Immune cells were found to express the α and δ subunits, encoded by SCNN1A and SCNN1D, consistent with previous findings³⁰ (Figure 3F). There was also a trend towards decreased SCNN1D expression with salt depletion in monocytes in salt-sensitive but not salt-resistant individuals (Figure 3E), which did not reach statistical significance. The expression levels of SCNN1A were similar between salt loading and depletion in both groups.

To investigate whether autocrine EET signaling in monocytes plays a role in IsoLG formation and salt sensitivity through regulation of ENaC activity, we assessed changes in the expression of genes associated with the production and breakdown of EETs in response to salt in vivo and ex vivo. We employed RNA sequencing on the monocytes collected from 11 healthy subjects following ex vivo exposure to either normal or high sodium media for 72 hours (Figure 3H). A representative heatmap for the expression of genes associated with the production and breakdown of EETs is shown in Figure 3K. Exposure to elevated sodium decreased the expression of CYP2S1 (cytochrome P450 family 2 subfamily S member 1), the main epoxygenase expressed in APCs (p=0.0017). There was no significant difference in gene expressions of CYP2J2 (cytochrome P450 family 2 subfamily J member 2), CYP2C8 (cytochrome P450 family 2 subfamily C member 8), CYP2C9 (cytochrome P450 family 2 subfamily C member 9), CYP3A4 (cytochrome P450 family 3 subfamily A member 4), CYP4Z1 (cytochrome P450, family 4, subfamily Z, polypeptide 1) and EPHX2 (soluble epoxide hydrolase). These results indicate that EET production in monocytes through CYP2S1 expression is salt-responsive.

Cite-Seq analysis of the PBMCs collected from the participants on baseline, salt loading and salt depletion corroborated the findings of RNA sequencing, showing that EET regulator genes CYP2S1 and EPHX2 are expressed in PBMCs (Figure 3D and 3F). CYP2S1 was mainly expressed in dendritic cells, while EPHX2 was found in CD4 and CD8 T cells (Figure 3B). The changes in the expression of CYP2S1 on baseline, salt loading, and salt depletion in different cell clusters were similar between salt-sensitive and salt-resistant participants (Figure 3E). SGK1 was found to be strongly expressed in dendritic cells and all subtypes of monocytes. UMAP representation of SGK1 expression showed a decreasing trend in monocytes from salt loading to depletion in salt-sensitive, but not salt-resistant subjects. The decrease in SGK1 expression in monocytes from salt loading to depletion significantly correlated with salt sensitivity index (Figure 3G).

Urinary EETs Associate with Salt Sensitivity and IsoLG Formation

To determine whether systemic or renal EETs regulate ENaC-dependent IsoLG accumulation in APCs, plasma and urine EET 8–9, 11–2 and 14–15 were measured at baseline, salt loading and depletion (Table S1). Both plasma and urinary EETs showed a decreasing trend from salt loading to depletion in all tertiles (Table S6 and S7). Urine EET 11–12, 14–15 and total significantly decreased from salt loading to depletion in the lowest tertile of salt sensitivity. Plasma DHETs showed an increasing trend from salt loading to depletion in all tertiles while urine DHETs showed a decreasing trend from salt loading to depletion in all tertiles (Table S6 and S7). No correlation was found between changes in urine EETs and urinary Na+ excretion or urinary Na+/K+ ratio from salt loading to depletion (Table S8).

Significant positive correlations were found between baseline urinary EET14–15 and salt sensitivity index as well as between urinary total ETTs and salt sensitivity index, but not between plasma EETs and salt sensitivity index (Figure 4A and Table S9).

Figure 4. The relationship of baseline plasma and urine EETs with salt sensitivity index and IsoLG⁺ APCs.

(A) Plasma and urine EET 14–15 in each subject at baseline (B), salt loading (SL) and salt depletion (SD) are depicted on left. The mean ±SEM in each tertile is depicted on right. 1^st (blue line) and 3^rd (red line) salt sensitivity index tertiles represent the least and the most salt sensitive subjects, respectively. (B) Correlations of plasma and urine EET 14–15 and Total with salt sensitivity index (B), (C) baseline plasma and urine EET 14–15 with baseline IsoLG⁺ APCs, (D) changes in plasma and urine EET (ΔpEET and ΔuEET) 14–15 with changes in IsoLG⁺ APCs (ΔIsoLG%) from salt loading to salt depletion. The best fit line was analyzed using linear regression, r value and the significance was computed using Spearman’s rank correlation.(E) Effect of EET 14–15 on high salt-induced IsoLG accumulation in dendritic cells and classical monocytes of both salt-sensitive and salt-resistant people with hypertension.

Figure 4B shows the relationship between baseline IsoLG-adduct accumulation and plasma and urine EETs. Baseline levels of plasma EET 14–15 negatively correlated with IsoLGs in all subtypes of monocytes (Table S10). Similarly, baseline levels of urine EET 14–15 negatively correlated with baseline IsoLG⁺ in classical and intermediate monocytes.

The relationship between changes in IsoLGs and plasma and urine EETs are depicted in Figure 4C. No significant interaction was found between changes in plasma EETs and IsoLGs. Changes in urine EET 14–15 negatively related to changes in IsoLGs in classical and intermediate monocytes (Figure 4D and Table S11). These findings suggest that reductions in renal EET 14–15 contribute to ENaC activation and IsoLG accumulation in monocytes in salt-sensitive people.

To investigate whether EET 14,15 directly influences sodium-induced IsoLG-adduct formation in APCs during salt sensitivity, we treated baseline human PBMCs isolated from salt sensitive and resistant subjects with normal salt media (140 mM) and high salt media (190 mM) with or without EET 14,15. IsoLG-adduct formation in APCs was analyzed by flow cytometry using the previously described gating strategy. Compared to normal salt, exposure to high salt increased the percent of IsoLG+ dendritic cells and classical monocytes obtained from salt sensitive subjects, which was prevented by co-incubation with EET 14–15 (Figure 4E). High salt treatment did not change the percent of IsoLG+ dendritic cells or classical monocytes obtained from salt resistant subjects.

Discussion

In this study, we found that both baseline and acute salt-induced changes of IsoLG-adduct accumulation in APCs are linked to salt-sensitivity in hypertensive people. Salt-induced changes in IsoLGs in monocytes were markedly suppressed with increasing urine EET 14–15, suggesting a regulatory role of renal EET 14–15 in ENaC-dependent formation of IsoLGs in monocytes.

The association of salt-sensitivity with increased cardiovascular morbidity and mortality is well-established². However, the pathophysiological mechanisms remain elusive³¹, hindering the search for targeted therapies. Our previous animal studies established a pathway by which sodium entry into APCs through amiloride-sensitive ENaC leads to lipid peroxidation, IsoLG-adduct formation and immune activation that results in salt-sensitive hypertension^32,33. Here, using a rigorous inpatient salt loading and depletion protocol, we present the first clinical evidence that acute changes in dietary salt intake in humans result in acute changes in IsoLG formation in circulating APCs, such that salt loading increased and salt depletion decreased IsoLG formation. Importantly, this relationship between salt intake and IsoLG formation was found only among individuals in the most salt-sensitive tertile, further suggesting a causal link between salt sensitivity and salt-induced, ENaC- mediatedIsoLG response. In additional in vitro studies, we found that APCs isolated from salt sensitive subjects had a robust increase in IsoLG formation in response to high salt treatment, which was not seen in APCs obtained from salt resistant subjects. Furthermore, co-incubation with EET 14–15, a robust inhibitor of ENaC, prevented high salt-induced increase in IsoLG response. Thus, our current and previous findings ^32,33 suggest that the salt-sensitivity-dependent association between dietary changes in salt and IsoLG formation in APCs may be mediated by ENaC-induced sodium entry. Figure 5 illustrates the hypothesized mechanism of how dietary sodium may result in IsoLG production. Further studies are required to establish the underlying mechanisms of this association in humans and investigate whether inhibition of ENaC with amiloride could mitigate the effects of dietary salt intake on IsoLG formation in APCs.

Figure 5.

Proposed mechanism of IsoLG-mediated salt sensitive hypertension. Antigen presenting cells (APCs) predominantly express ENaC subunits α and δ, the expression of which is regulated by SGK1. Previous evidence shows that high extracellular sodium enters dendritic cells via amiloride-sensitive ENaC and reverses the direction of Na+/Ca2+ exchanger, resulting in Ca2+ influx into the cell.. Calcium induces the activation of protein kinase C that leads to the activation of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase and subsequent formation of IsoLGs. IsoLG adduction to proteins generates neo-antigens that drive the production of interleukins 1β and 6 in dendritic cells and activation of T cells, resulting in inflammation and hypertension^32,33. Renal EET 14–15 may regulate IsoLG response in APCs through ENaC regulation in the kidneys and play a role in the pathogenesis of salt-sensitive hypertension. The role of EET 14–15 production by CYP2S1 epoxygenase in ENaC inhibition in salt-sensitivity remains unclear.

Our single-cell RNA sequencing analyses indicate that monocyte gene expression is dynamically regulated by salt in salt-sensitive people, providing mechanistic insight into dietary salt-induced APC activation in vivo. Furthermore, APCs of salt resistant individuals show minimal salt-induced change in gene expression, supporting the hypothesis that salt-induced APC activation is a characteristic of salt sensitivity.

Our single-cell RNA sequencing analysis show that the ENaC α and δ subunits, encoded by SCNN1A and SCNN1D, are the main ENaC subunits expressed in APCs. We also observed a trend towards decreased SCNN1D expression with salt depletion in salt sensitive individuals. The δ subunit, the only of the four subunits not expressed in rodents, can replace the α subunit to couple with the other subunits or form a channel on its own, which alters the properties and functions of the channel. Our previous studies found a strong association between SCNN1D variants and blood pressure although ENaC in the human kidney does not express the δ subunit³⁰. We also reported increased expression of SCNN1D in human monocytes cultured in high salt compared to normal salt²⁹. The current findings support the hypothesis that extrarenal ENaC, particularly the δ subunit in APCs, play a role in blood pressure regulation.

A critical question is how the ENaC-dependent IsoLG response is regulated differently between the salt-sensitive and salt resistant populations We found that changes in urine EET 14–15, the most common EET regioisomer³⁴ are negatively associated with changes in IsoLGs in monocytes.EET 14–15 is a well-known autocrine and paracrine inhibitor of ENaC ^35,36. In the present study, EET 14–15 prevented high salt-induced IsoLG response in APCs, showing its inhibitory effect on APC ENaC. Indeed, EETs have been found to be regulated differentially between salt sensitive and resistant populations¹⁴. Previous studies ^37,38 have shown that APCs accumulate in the corticomedullary junction and medulla of kidneys where the interstitial sodium concentrations are exceedingly high and reach sodium concentrations used in previous in vitro experiments. Since APCs constitutively enter and re-emerge from nonlymphoid tissue without differentiation, APCs activated by the high sodium concentrations in the kidney can re-enter circulation and be detected in peripheral blood³⁹. Our previous observations revealed that high salt feeding results in the increased expression of chemokine ligand CCL2, and endothelial cellular adhesion molecules in the kidneys³⁷, which could facilitate increased transmigration of immune cells to the kidneys. Therefore, renal EETs may act locally to regulate the ENaC activity in the circulating monocytes entering the renal medulla.

While we found a direct link between urinary EETs and salt sensitivity, urinary sodium excretion and sodium/potassium ratio, a surrogate of renal tubular ENaC activity, did not associate with urinary EETs or salt sensitivity index. These observations support the hypothesis that the role of urinary 14,15-EET in salt sensitivity is independent of its natriuretic effect through the inhibition of renal tubular ENaC. We also investigated the expression of epoxygenases in APCs in response to in vitro and in vivo sodium. CYP2S1 was the most expressed epoxygenase in APCs. While the expression of CYP2S1 in monocytes was found to decrease in response to in vitro high salt exposure, CITE-Seq suggested a trend of decreasing expression after salt depletion in salt sensitivity. Since in vitro experiments were carried out on monocytes from healthy participants with unknown salt sensitivity status, it is possible and expected that monocytes of healthy people react differently to salt compared to APCs of people with salt sensitive hypertension. Thus, our results imply that CYP2S1 may potentially be a salt-sensitive gene in salt sensitivity that require further investigation. We did not find CYP2J2 and CYP2C8 expression in PBMCs. Previous studies have reported high expression of epoxygenases CYP2J2 and CYP2C8 in human acute monocytic leukemia cell line and PBMCs of patients with hematologic malignancies, but not in PBMCs of healthy people^40,41. Furthermore, monocyte CYP2J2 and CYP2C8 have been associated with more release of linoleic acid products compared to EETs⁴⁰.While further studies are needed to confirm the association between urinary EETs and IsoLG production in APCs and establish the underlying mechanisms, our current findings suggest that several pharmacological approaches including orally available EET analogs or soluble epoxide hydrolase inhibitors⁴² may be potential therapeutic options to modulate ENaC-mediated APC activation in salt-sensitive hypertension.

We also found that baseline levels of urine EETs are higher in more salt sensitive individuals, who are also characterized by lower baseline levels of IsoLG adducts in monocytes. The changes in urinary sodium excretion indicate that the salt intake of our participants at baseline was significantly lower than the salt loading day. Therefore, the high urine EETs and low IsoLG levels in salt sensitive participants at baseline may represent a salt depleted state rather than a salt loaded state. Our results imply that IsoLG formation in APCs is highly responsive to acute changes in salt. Therefore, another possibility is that IsoLG formation in more salt resistant individuals is driven by stimuli other than salt, rendering these APCs unresponsive to the effects of salt. Further mechanistic studies are required to unearth the reasons behind such a difference.

Nevertheless, our findings further suggest the baseline percentage of IsoLG-containing APCs may be the first marker of salt sensitivity that does not depend on a salt loading and depletion protocol. Salt sensitivity of blood pressure is a strong predictor of poor cardiovascular outcomes and mortality ². Currently, the diagnosis of salt sensitivity requires laborious and costly methods that cannot be undertaken in clinical practice⁴³. An easily accessible surrogate marker would enable mass phenotyping of populations, hasten research on genetic and environmental factors and help the development of targeted treatment options for the salt sensitive individuals.

Interestingly, the association between salt sensitivity and baseline levels or changes in IsoLG formation was not evident in intermediate monocytes. Intermediate monocytes represent a monocyte pool that either differentiates into nonclassical monocytes or leaves the circulation⁴⁴. In our recent studies, we found that in vitro and in vivo high salt exposure induces the conversion of monocytes into dendritic cells ²⁴. Differences in the respective roles of each monocyte subset or salt-induced dynamic changes in cell pools may account for this observed discrepancy.

We also observed that plasma levels of pro-inflammatory cytokines, including INF-γ and IL-17A, significantly increased from salt loading to depletion, which negatively correlated with the changes in IsoLG⁺ classical monocytes. In our previous findings, we showed that IsoLG formation in APCs promotes the production of INF-γ and IL-17A in T cells⁶. In the current study, we did not observe a significant change from baseline to salt loading, but a decrease from salt loading to depletion. It is possible that while IsoLG-production is an immediate response to salt, the systemic release of proinflammatory cytokines secondary to APC and subsequent T cell activation may be delayed. Thus, the increase in specific cytokines we observed on salt depletion may be representative of an expected increase in cytokines on salt loading. Data from follow-up samples would be needed to assess this possibility.

Our study has several limitations. First, this was an exploratory study with a small sample size. Second, this cross-sectional study was not designed to draw direct causal inferences on the association between IsoLG production in APCs and salt-sensitive hypertension, but rather to investigate the existence of such association. The limited sample size of the current study does not allow the investigation of whether IsoLG formation in APCs contributes to the gender differences observed in salt sensitive hypertension⁴⁵. Future studies with larger sample size are needed to reveal possible gender differences in the associations between IsoLG production, EETs and salt sensitive hypertension. We did not recruit normotensive individuals since the prevalence of salt sensitivity is higher in the hypertensive population; thus, whether such associations apply to the normotensive salt-sensitive population remains to be explored. Despite these limitations, the current study is the first to show that IsoLG formation in circulating APCs is responsive to dynamic changes in dietary salt intake in humans.

In conclusion, we have defined a dynamic relationship between dietary salt intake and IsoLG formation in APCs in salt-sensitive hypertensive humans, confirming the findings of previous in vitro and animal studies. Our observations support the role of ENaC-mediated, IsoLG-dependent APC activation in the pathogenesis of salt sensitivity and suggest that baseline IsoLGs may be the first biomarker for the diagnosis of salt sensitivity that may be used in the clinic instead of the currently used laborious salt loading and depletion protocols. The relationship between renal EET 14–15 and IsoLG formation in monocytes suggests that EET 14–15 may play a regulatory role in oxidative stress and inflammation in salt sensitivity.

Perspectives:

This is the first study to investigate the relationship between changes in acute dietary salt intake and IsoLG formation in salt sensitive hypertension in humans. We found that both baseline level and dietary salt-induced changes of IsoLG⁺ APCs significantly correlated with salt sensitivity. Moreover, ENaC-dependent, salt-responsive IsoLG formation may be regulated by renal EET 14–15. These findings provide a novel insight into the pathophysiological mechanism of salt sensitivity in people with hypertension. Furthermore, baseline levels of IsoLG⁺ APCs and urinary EET 14–15 may have profound clinical significance as diagnostic markers for salt sensitivity in an office setting.

Novelty and Relevance.

What is new?

Salt-sensitive isolevuglandin (IsoLG) formation in antigen presenting cells (APCs) associated with higher salt sensitivity in people with hypertension. Urinary EET 14–15 negatively correlated with salt-sensitive changes in IsoLG⁺ monocytes. Mechanistically, high-salt treatment induced IsoLG formation in APCs which was inhibited by EET 14–15 co-treatment.

What is relevant?

The mechanism underlying salt sensitive hypertension remains elusive. Our study is the first to demonstrate the relationship between IsoLG formation in APCs and salt sensitive hypertension in humans.

Clinical/Pathophysiological implications?

IsoLG formation in APCs may be a potential target for salt sensitive hypertension. Baseline levels of IsoLG⁺ APCs or urinary EET 14–15 may provide a diagnostic biomarker of salt sensitivity that does not require a protocol of salt loading.

Abbreviations:

ADTs

Antibody Derived Tags

APC

antigen presenting cells

BMI

body mass index

CI

confidence interval

CITE-seq

Cellular Indexing of Transcriptomes and Epitopes by Sequencing

CYP

cytochrome P450

DBP

diastolic blood pressure

DHETs

dihydroxyeicosatrienoic acids

ENaC

epithelial sodium channel

EETs

epoxyeicosatrienoic acids

IsoLGs

isolevuglandins

PBMCs

peripheral blood mononuclear cells

SBP

systolic blood pressure

sEH

soluble epoxide hydrolase

SEM

standard error of the mean

SGK1

salt-sensing kinase serum/glucocorticoid kinase 1