High-salt diet aggravates allergic rhinitis through the NFAT5 signaling pathway

Abstract

Allergic rhinitis (AR) affects 10–40% of the world’s population, and modern high-salt diets (HSD) may influence immune function. Currently, the influence and mechanism of a HSD on AR remain poorly understood. The study aims to investigate the role of a HSD in AR through the P38/MAPK–NFAT5–SGK1 signaling pathway. The serum IgE levels, 24-hour urinary sodium excretion, and AR symptom scores in patients were detected. To evaluate the immune responses and potential pathway, ovalbumin-induced AR mice were exposed to a HSD. Cellular gene silencing technology was employed to identify key regulators of the NFAT5 pathway. In AR patients, urinary sodium excretion was positively correlated with IgE levels and symptom scores. In the mouse model, a HSD altered the gut microbiota and upregulated NFAT5 expression, leading to nasal mucosal barrier disruption and inflammation exacerbation. Gene-silencing experiments confirmed the critical role of the P38/MAPK–NFAT5–SGK1 pathway in mediating both allergic responses and epithelial damage. Notably, switching from a HSD to a normal diet partially reversed clinical symptoms in mice, but the immune memory remained difficult to reset. In conclusion, this work provides the strong evidence of salt-immune axis in AR through osmotic sensing pathways, advancing mechanistic understanding and clinical management approaches.

Introduction

Allergic rhinitis (AR) is a non-infectious, chronic inflammatory disease of the nasal mucosa, primarily mediated by immunoglobulin E (IgE) following exposed to allergens¹. The incidence of the disease has increased substantially in recent years^2,3. In Europe, the prevalence of AR among adults in Denmark has gradually increased from 19% to 32% over the past three decades⁴. The standardized prevalence rate of AR among adults in Asian countries like China has increased by 6.5% in the last 6 years, affecting about 240 million people⁴. These high incidence of AR-induced respiratory infections has significantly burdened the overall health of patients and adversely affected the standards of their life-style and socioeconomic conditions⁵.

The increasing prevalence of AR has been strongly linked to modern dietary habits⁶. The rapid adaptation of Western-style diet corresponds to high salt consumption⁷, and WHO recommended just 5 gram intake of salt per day. However, the global average daily salt consumption is 10.8 grams, and only 10% is reported from direct consumption and utilization in cooking, while remaining intake is from processed foods⁸. The excessive intake of salt can alter gut microbiome and short-chain fatty acids⁹, causing chronic inflammation, congenital and adaptive immune response dysfunction^10–12. Consumption of a high-salt diet (HSD) can activate the NLRP3 inflammasome protein complex, resulting in an enhanced release of pro-inflammatory cytokines, including IL-1β, which in turn exacerbates inflammatory responses¹³. In vitro findings on human monocytes exposure to high concentration sodium (Na⁺), showed an increased acquisition of dendritic cell (DC)-like morphology and the production of pro-inflammatory cytokines such as TNF-α, IL-6, and IL-1β¹⁴. Additionally, the in vitro elevated levels of NaCl can stimulate the induction of IL-17A in naive CD4⁺ T cells and suppress the expression of Foxp3 in regulatory T cells (Tregs), thereby impairing immune tolerance^15,16.

Nuclear factor of activated T-cells 5 (NFAT5) is implicated in the regulation of water and salt osmotic balance within and outside cells, as well as the function of immune cells under hypertonic conditions¹⁷. The HSD can activate T cells via NFAT5, stimulating immune signal transduction pathways¹⁸. NFAT5 activation under high salt conditions, can induce pathogenic TH17 cells and upregulate cytokines such as IL-1β, TNF-α, and MCP-1, exacerbating the development of autoimmune diseases¹⁹. NaCl can also induce TH2 cell responses in atopic diseases by osmotically sensitive transcription factor NFAT5⁹. In AR patients, Th2 cell activation leads to the aggregation and activation of eosinophils and mast cells through the secretion of cytokines such as IL-4, IL-5, and IL-13, releasing pro-inflammatory mediators²⁰. Concurrently, this activation upregulates the pro-inflammatory factors like TNF-α and IL-6, exacerbating tissue inflammation and damage²¹. However, no studies have investigated the association between nasal mucosa and NFAT5, so in the current study we hypothesized the mediation of immune response to AR through NFAT5.

In the current research work, the human study was conducted to determine the effect of a HSD on AR. Animal experiments and cell gene silencing techniques were used to investigate the mechanism related to NFAT5 pathway. This article is the first so far to explore the mediation of AR by a HSD via NFAT5, aiming to provide foundational insights into the complex mechanisms underlying the aggravation of allergic diseases by a HSD consumption (Fig. S1).

Results

A HSD exacerbates the immune response in AR patients

The total number of AR patients selected for the proposed study were 59 and 8 patients were excluded from analysis based on pre-defined inclusion and exclusion criteria and remaining 51 patients’ data was analyzed. Among the eight excluded patients, two declined participations, and six did not meet the inclusion criteria (Fig. S2). The basic information of patients is shown in Table 1 and the flowchart of the patient status is shown in Fig. S2. Initially, patients were categorized into three groups according to symptom severity. The 24 h urinary Na⁺ levels of patients with severe symptoms were significantly higher than those in the moderate symptoms group, whereas no significant difference was observed between the moderate and mild symptom groups (Fig. 1A). However, no significant variation in plasma Na⁺ levels was detected across the three groups (Fig. 1B). Based on 24 h urinary Na⁺ levels, patients were subsequently divided into the HSD and the low-salt diet groups (Fig. 1C). The total IgE levels and total nasal symptom score (TNSS) in the HSD group were significantly higher than those in the low-salt diet group (Fig. 1D, E). Nasal obstruction, a common symptom of AR, showed significantly higher scores in the HSD group compared to the low-salt diet group (Fig. 1F). While no significant differences were observed in running nose, nasal itching, or sneezing between the two groups (Fig. 1G–I). Association analysis of 24 h urinary Na⁺ revealed a strong positive correlation with both the TNSS and total IgE level, suggesting that higher urinary Na⁺ levels are associated with more severe symptoms and elevated IgE levels (Fig. 1J, K).

Table 1.

Participant characteristics and symptoms

	AR patients	AR type by severity				AR type by urinary sodium
		Mild	Moderate	Severe	P value	Low-salt diet	High-salt diet	P
n	51	11	20	20		25	26
Age, mean (SD)	34.78 (10.25)	32.82 (11.04)	33.95 (8.33)	36.70 (11.67)	0.549	38.36 (10.16)	31.35 (9.27)	0.013
Sex					0.817			0.577
Male, no.(%)	23 (45.1%)	5 (45.5%)	10 (50.0%)	8 (40.0%)		10 (40.0%)	13 (50.0%)
Female, no.(%)	28 (54.9%)	6 (54.5%)	10 (50.0%)	12 (60.0%)		15 (60.0%)	13 (50.0%)
Nasal symptoms
Obstruction, mean (SD)	2.72 (1.06)	1.73 (1.10)	2.70 (1.03)	3.30 (0.57)	< 0.001	2.36 (1.28)	3.08 (0.63)	0.017
Running, mean (SD)	3.41 (1.00)	2.09 (1.30)	3.65 (0.59)	3.90 (0.31)	< 0.001	3.36 (1.08)	3.46 (0.95)	0.722
Itching, mean (SD)	2.39 (1.17)	1.45 (1.04)	2.20 (1.11)	3.10 (0.85)	< 0.001	2.24 (1.13)	2.54 (1.21)	0.367
Sneezing, mean (SD)	2.43 (1.24)	1.36 (0.50)	2.10 (1.21)	3.35 (0.88)	< 0.001	2.20 (1.12)	2.65 (1.32)	0.193
TNSS, mean (SD)	10.96 (2.89)	6.64 (1.57)	10.65 (1.14)	13.65 (0.93)	< 0.001	10.16 (2.95)	11.73 (2.66)	0.051

AR allergic rhinitis, TNSS Total Nasal Symptom Score, SD Standard Deviation.

Fig. 1. A HSD exacerbates the immune response in AR patients.

A The 24 h-urinary sodium levels in AR patients with different symptom severity (Mild, n = 11; Moderate, n = 20; Severe, n = 20). B The plasma sodium levels in AR patients with different symptom severity (Mild, n = 11; Moderate, n = 20; Severe, n = 20). C AR patients were divided into Low-salt diet (n = 25) and High-salt diet group (n = 26) based on 24 h-urinary sodium levels measured by ICP/MS. D Total serum IgE levels in the Low- vs. High-salt diet group. E The TNSS in the Low- vs. High-salt diet group. F–I The nasal symptom scores (obstruction, running, itching and sneezing) in the Low- vs. High-salt diet group. J Pearson correlation analysis between TNSS and 24 h-urinary sodium levels (n = 51). K Pearson correlation analysis between IgE levels and 24 h-urinary sodium levels (n = 51). T-test was used for two groups, one-way ANOVA and LSD post hoc test were used for 3 groups. AR allergic rhinitis. TNSS Total Nasal Symptom Score, ICP/MS Inductively coupled plasma mass spectrometry, ns no significance difference.

NaCl rich diet promotes Th2 cell immunity and exacerbates AR immune symptoms in ovalbumin (OVA)-induced mice

To explore the associations between a HSD and AR, we employed a mouse model of AR combined to a HSD feeding in Balb/c mice (Fig. 2A). Mice fed with a HSD developed an aggravated disease phenotype compared to those fed with regular chow. A slight decline in body weight was observed in the allergic groups at the time of sensitization; however, no significant differences in the overall trend were noted (Fig. 2B). The numbers of sneezing episodes, nasal scratching behaviors, and the total symptom score were significantly higher in the AR group compared to the control (CON) group, and these parameters were further elevated in the HSD group compared to the AR group (Fig. 2C–E). Serum levels of OVA-IgE and histamine increased in both groups following sensitization, with higher levels observed in the HSD-OVA compared to chow-OVA (Fig. 2F, G). Additionally, we assessed the impact of a HSD and OVA on nasal epithelial thickness. The epithelial thickness of the HSD group increased significantly compared to that of the AR group (Fig. 2H, Q).

Fig. 2. A HSD promotes Th2 cell immunity and exacerbates AR immune symptoms in OVA-induced mice.

A Experimental diagram. B Body weight changes in mice (n = 8 per group). C, D The nasal symptom scores of sneezing and scratching (n = 8 per group). E Total nasal symptom scores (n = 8 per group). F, G Serum levels of OVA-specific IgE and histamine (n = 6 per group). H Thickness of the nasal epithelium (n = 3 per group). I–N Levels of IFN gama, IL-4, IL-5, IL-13, TSLP, and IL-17 in nasal lavage fluid (n = 5 per group). O, P, R, S Percentages of TH1 and TH2 cell subsets in splenocytes analyzed by flow cytometry (n = 3 per group). Q Representative images of nasal histology (H&E staining). Scale bar: 20 μm. One-way ANOVA and LSD post hoc test were used for 3 groups. OVA ovalbumin, IgE immunoglobulin E, CON the control group, AR the allergic rhinitis group, HSD the high salt diet group, ns no significance difference.

Analysis of nasal lavage fluid in mice showed that the characteristic Th2 cytokine levels and pro-inflammatory cytokines were elevated in the HSD group compared to those of the AR group, while Th1 and Th17 levels remained unchanged. Compared to the CON group, IL-4 levels in the nasal lavage fluid of the HSD group were significantly higher than those in the AR group, and IL-5, IL-13, TSLP, IFN gama and IL-17 showed a trend toward increase (Fig. 2I–N). Flow cytometry analysis of mouse spleen cells demonstrated a significant decrease in the proportion of Th1 cells and a significant increase in the proportion of Th2 cells (Fig. 2O, P, R, S).

A HSD modulates gut microbiome and impairs mucosal barrier function

The dietary intake into the gastrointestinal tract for digestion can influence the composition and function of microbiome, modify the gut barrier, which may be related to local and systemic immune responses²². Therefore, we investigated the effects of a HSD on fecal microbes and the gut barrier in mice. The α-diversity index (ACE) in the HSD group was lower than that in the AR and CON groups (Fig. 3A). Significant differences in fecal microbial composition were observed among the three groups (Fig. 3B). The gut microbiome was dominated by members of Firmicutes and Bacteroidetes (Fig. 3C). After ingestion of a HSD, the relative abundance of Firmicutes increased, while the relative abundance of Bacteroidetes decreased, leading to an increased Firmicutes/Bacteroidetes (F/B) ratio (Fig. 3D–F). Analysis of the relative abundance of microbiome in the CON, AR, and HSD groups revealed differences in Colidextribacter, Lactobacillus, and Clostridiales (Fig. 3G, H).

Fig. 3. A HSD modulates gut microbiome and impairs intestinal barrier function.

A α-diversity (ACE index), and (B) β-diversity (Principal coordinates analysis,PCoA) of fecal microbiome samples. C Relative abundance of phyla (n = 5 per group). D–F Relative abundance of Firmicutes, Bacteroidota and Firmicutes/Bacteroidota ratio (n = 5 per group). G LEfSe analysis identifying differentially abundant taxa between the CON and AR groups (LDA score > 3.5, n = 5 per group). H LEfSe analysis identifying differentially abundant taxa between the AR and HSD group (LDA score > 3.5, n = 5 per group). I–M Relative mRNA expressions of E-cadherin, Claudin 1, Occludin, Zo-1 and Muc2 in the jejunum by qPCR (n = 5 per group). N–R Relative mRNA expressions of E-cadherin, Claudin 1, Occludin, Zo-1 and Muc2 in the ileum by qPCR (n = 5 per group). S Representative images of ileum histology (H&E staining). Scale bar: 100 μm. Muc2 Mucin 2, CON the control group, AR the allergic rhinitis group, HSD the high salt diet group. T-test was used for two groups, one-way ANOVA and LSD post hoc test were used for 3 groups, ns no significance difference.

We further examined the expression of barrier genes in the jejunum and ileum. Compared to the AR group, the expression of Occludin and E-cadherin in the jejunum of the HSD group were significantly decreased, while the expression of Zo-1 and Mucin 2 showed a decreasing trend (Fig. 3I, K–M). In the ileum, the expression of Zo-1 and Mucin 2 in the HSD group was significantly lower than that in the AR group, and the expression of Occludin and E-cadherin showed a decreasing trend (Fig. 3N, P–R). In both the jejunum and ileum, the expression of Claudin 1 in the HSD group was significantly higher than in the AR group (Fig. 3J, O). Ileal sections in the HSD group showed abnormal hyperplasia and inflammatory infiltration compared with those in the AR group (Fig. 3S).

A HSD consumption aggravates nasal mucosal barrier damage through the P38/MAPK-NFAT5-SGK1 pathway

The concentrations of Na⁺ and potassium (K⁺) in serum of mice were found to be significantly higher in the HSD group than in the CON and AR groups, while there was no significant difference between the CON and AR groups (Fig. 4A, B). The detection of P38/MAPK-NFAT5-SGK1 pathway in mouse nasal mucosa showed that the expression of Nfat5 gene in the AR group was significantly higher than that in the CON group, while the expression of Nfat5, and serum/glucocorticoid-regulated kinase 1 (Sgk1) genes in the HSD group was significantly higher than that in the AR group (Fig. 4C, D). We also detected the expression of phosphorylated-P38 (p-P38) and P38, and the results showed that the expression of p-P38 normalized to total p38 protein levels in the HSD group was significantly higher than that in the AR group (Fig. 4E, F). The protein expression of NFAT5 in the AR group was higher than that in the CON group, and the HSD group was significantly higher than that in the AR group (Fig. 4G, H). From CON to AR to HSD, the expression of Serum/glucocorticoid regulated kinase 1 (SGK1) protein increased successively (Fig. 4I, J). Immunofluorescence staining showed that the expression of OCCLUDIN and ZO-1 in the AR group was lower than that in the CON group, while that in the HSD group was lower than that in the AR group (Fig. 4K). The results indicated that the nasal mucosa was damaged in allergic mice, and a HSD would aggravate the damage of the nasal mucosa. The detection of Mucin 5ac (Muc5ac) gene expression in nasal mucosa tissues also showed that the expression in the AR group was significantly higher than that in the CON group, while that in the HSD group was significantly higher than that in the AR group (Fig. 4L).

Fig. 4. A HSD damages nasal mucosal barrier via the NFAT5 pathway.

A, B Serum Na⁺ and K⁺ levels (n = 5 per group). C, D, L Relative mRNA expressions of Nfat5, Sgk1 and Muc5ac in nasal mucosa analyzed by qPCR (n = 5 per group). E–J Protein levels of p-P38 (normalized to total p38), NFAT5 and SGK1 in nasal mucosa analyzed by Western blotting (n = 4 per group). K Representative immunofluorescence images showing localization of OCCLUDIN (red) and ZO-1 (red) in nasal mucosa; nuclei are counterstained with DAPI (blue). Scale bar: 20 μm. One-way ANOVA and LSD post hoc test were used for 3 groups. Na⁺ sodium, K⁺ potassium, CON the control group, AR the allergic rhinitis group, HSD the high salt diet group, Nfat5 or NFAT5, nuclear factor of activated T-cells 5. Sgk1 or SGK1, serum/glucocorticoid-regulated kinase 1, p-P38 phosphorylated-P38, Muc5ac, Mucin 5ac ICP/MS Inductively coupled plasma mass spectrometry, ns no significance difference.

P38/MAPK-NFAT5-SGK1 regulates β-Hex and epithelial barrier index expression in RBL-2H3 and RPMI-2650 cells in high-NaCl conditions

A HSD has been shown to upregulate the expression of p-P38, NFAT5, and SGK1 in the nasal mucosa, thereby exacerbating AR. Based on this observation, we hypothesized that elevated NaCl levels activate the P38/MAPK pathway, subsequently inducing NFAT5 and its downstream target SGK1, which in turn modulates immune responses and barrier function. To test this hypothesis, we employed RBL-2H3 cells and RPMI-2650 cells (Figs. 5A, 6A). The optimal NaCl concentration for intervention was determined using the CCK8 assay (Fig. S3). In RBL-2H3 cells, our results revealed that NaCl intervention significantly increased β-Hex levels (Fig. 5B). Under high-salt conditions, p-P38, but not total P38, was markedly elevated (Fig. 5C–E), accompanied by the induction of Nfat5 (Fig. 5F). The p38/MAPK inhibitor SB202190 (P38i) was utilized to investigate whether inhibition of the P38/MAPK pathway would alter these outcomes (Fig. S5A, B). The results demonstrated that P38i significantly suppressed Nfat5 mRNA induction and attenuated the high-salt-induced increase in β-Hex levels (Fig. 5G, H). Similarly, treatment with the NFAT5 inhibitor (KRN5) (Fig. S5C) significantly reduced NaCl-induced Nfat5 expression and decreased β-Hex levels (Fig. 5I, J). SGK1, a direct downstream target of NFAT5, is regulated by both the P38/MAPK pathway and NFAT5. To confirm this regulatory mechanism, we assessed Sgk1 expression following treatment with P38i and KRN5. The results indicated that both inhibitors reduced Sgk1 gene expression (Fig. 5K, L). Furthermore, the SGK1 inhibitor GSK650394 (SGKi) reduced Sgk1 gene expression (Fig. S5D) and mitigated hypersalt-induced increases in the expression of Sgk1 and β-Hex (Fig. 5M, N).

Fig. 5. The immunological tolerance of RBL-2H3 cells induced by NaCl is dependent on P38/MAPK, NFAT5 and SGK1.

A Experimental diagram. B β-Hex release in cellular supernatant (n = 4 per group). C–F RBL-2H3 cells were stimulated in the presence or absence of additional 50 mM NaCl and were analysed by qPCR for Mapk14 (encoding the p38 protein) and Nfat5 (n = 6 per group) and by western blotting for p-P38 protein expressions normalized to total p38 (n = 4 per group). G, I, K–M mRNA expressions of Nfat5 and sgk1 after treatment with inhibitors against p38 (P38i), NFAT5 (KRN5), or SGK1 (SGKi) (n = 6 per group) (H, J, N) β-Hex release under the indicated inhibitor treatments (KRN5, P38i and SGKi) with or without NaCl stimulation (n = 4 per group). One-way ANOVA and LSD post hoc test were used for 3 groups. β-Hex, β-Hexosaminidase. Con, the control group. Nfat5 nuclear factor of activated T-cells 5. Sgk1 serum/glucocorticoid-regulated kinase 1, p-P38 phosphorylated-P38, P38i p38/MAPK inhibitor SB202190, KRN5 an inhibitor of NFAT5. SGKi SGK1 inhibitor GSK650394, ns no significance difference.

Fig. 6. The barrier function of RPMI-2650 cells induced by NaCl is dependent on P38/MAPK, NFAT5 and SGK1.

A Experimental diagram. B–G RPMI cells were stimulated in the presence or absence of additional 50 mM NaCl and were analysed by western blotting for p-P38 protein expressions normalized to total p38 (n = 4 per group) and by qPCR for Nfat5, Sgk1, Zo-1, Muc5ac (n = 6 per group). H–J Cells transduced with NFAT5-specific (shNFAT5) or control (NC) shRNA were analyzed by qPCR for Nfat5, Zo-1, and Muc5ac. (n = 6 per group). K, L Cells were inhibited by P38i and SGKi inhibitors and the expressions of Muc5ac and Zo-1 were verified by qPCR (n = 6 per group). M Lentivirus-mediated transfection efficiency of shNFAT5 into RPMI-2650 cells at different times. N Transfection efficiency of lentivirus-mediated shNFAT5 transfection of RPMI-2650 cells with different MOI. One-way ANOVA and LSD post hoc test were used for 3 groups, Con the control group, Nfat5 nuclear factor of activated T-cells 5, Sgk1 serum/glucocorticoid-regulated kinase 1, p-P38 phosphorylated-P38. Muc5ac, Mucin 5ac. P38i p38/MAPK inhibitor SB202190, SGKi SGK1 inhibitor GSK650394, MOI multiplicity of infection. ns no significance difference.

RPMI-2650 cells were also utilized to investigate whether the P38/MAPK-NFAT5-SGK1 pathway is present in these cells and whether it affects their epithelial barrier function. The cytotoxicity CCK8 test of the intervention was explored in Fig. S4. The results indicated that NaCl intervention significantly increased the expressions of p-P38 normalized to total p38 protein levels, Nfat5, and Sgk1, while significantly decreasing the expression of Zo-1 and increasing the expression of Muc5ac (Fig. 6B–G). We conducted cell transfection at different times and with different multiplicity of infection (MOI), and finally determined that using MOI = 20 and conducting transfection for 120 h would yield the best result (Fig. 6M, N). We silenced NFAT5 using shRNA in RPMI-2650 cells (Fig. S5F), and the results showed that shNFAT5 significantly reduced the expression of Nfat5 after NaCl intervention, leading to a significant increase in Zo-1 and a decrease in Muc5ac expression (Fig. 6H–J). P38i inhibited the expression of p-P38 and also reduced the expression of Nfat5 and Sgk1 (Fig. S5E, S5G–I). SGKi could inhibit the expression of Sgk1 (Fig. S5J). Additionally, both P38i and SGKi increased the expression of Zo-1 and decreased the expression of Muc5ac (Fig. 6K–L).

Short-term improvement of a HSD reversibly alleviates AR

The impact of HSD on AR has been verified in the above studies, which utilized a two-month HSD. However, we aimed to investigate whether a short-term HSD affects AR and whether a long-term HSD followed by a short-term return to a normal diet can improve AR symptoms. Therefore, four groups were set up: the persistent normal diet (PND) group, the persistent salt diet (PSD) group, the short salt diet (SSD) group, and the short normal diet (SND) group (Fig. 7A). Serum Na⁺ and K⁺ levels were decreased in the SND group and the thickness of nasal epithelium in the SSD and SND groups was between the PND and PSD group (Fig. S6D–G).Compared with the PND group, the SSD group exhibited an increasing trend in sneezing frequency, nose-scratching frequency, and symptom score, although these levels were lower than those in the PSD group (Fig. 7B–D). Conversely, the SND group showed significantly lower levels than the PSD group (Fig. 7B–D). The levels of IL-4 and IFN gama in nasal lavage solution in the SSD group were significantly higher than those in the PND group, but IL-5 had no significant difference (Fig. S6A–C). IL-4 in the SND group was significantly lower than that in the PSD group (Fig. S6A). These findings suggest that a short-term HSD may exacerbate AR symptoms, but to a lesser extent than a PSD. Additionally, a short-term return to a normal diet can alleviate AR symptoms. However, OVA-IgE levels were significantly higher in the PSD, SSD, and SND groups compared to the PND group, with no significant differences among the three high-salt diet groups (Fig. 7E). Histamine levels in the SSD and SND groups were intermediate between those of the PND and PSD groups, with no significant differences (Fig. 7F). Compared with the PND group, the SSD group showed an increasing trend in Nfat5, Sgk1 and Muc5ac expression, while p-P38 protein expressions normalized to total p38 levels were higher in the SSD group (Fig. 7G–K). In contrast, Nfat5, Sgk1, Muc5ac and p-P38 levels in the SND group were significantly lower than in the PSD group (Fig. 7G–K). Immunofluorescence results also indicated that the expression of OCCLUDIN and ZO-1 in the SSD and SND groups was intermediate between that of the PND and PSD groups (Fig. 7L).

Fig. 7. Short-term improvement of a HSD reversibly alleviates AR.

A Experimental schema. B, C The nasal symptom scores of sneezing and scratching (n = 5 per group). D Total nasal symptom scores (n = 5 per group). E, F Serum levels of OVA-specific IgE and histamine (n = 5 per group). G, J, K Relative mRNA expressions of Nfat5, Sgk1 and Muc5ac in nasal mucosa analyzed by qPCR (n = 5 per group). H, I Protein levels of p-P38 (normalized to total p38) in nasal mucosa analyzed by Western blotting (n = 4 per group). L Representative immunofluorescence images showing localization of OCCLUDIN (red) and ZO-1 (red) in nasal mucosa; nuclei are counterstained with DAPI (blue). Scale bar: 20 μm. One-way ANOVA and LSD post hoc test were used for 3 groups. OVA ovalbumin. PND the persistent normal diet group, PSD the persistent salt diet group, SSD the short salt diet group, SND the short normal diet group. Nfat5 or NFAT5 nuclear factor of activated T-cells 5, Sgk1 or SGK1 serum/glucocorticoid-regulated kinase 1, p-P38 phosphorylated-P38, Muc5ac, Mucin 5ac, ns no significance difference.

Discussion

Numerous pathological conditions and diseases have been associated with Western dietary patterns in recent years. However, providing evidence of potential causation and determining the mechanisms involved remain challenging. This study provides the first evidence suggesting that a HSD is associated with exacerbated AR, potentially through the P38/MAPK-NFAT5-SGK1 pathway.

Our experimental results showed that a HSD is associated with a significant exacerbation of AR symptoms, enhanced IgE and Th2 responses in mice, aligning with human epidemiological data linking a HSD to elevated IgE levels. These findings suggest that dietary salt intake may exacerbate allergic inflammation by disrupting immune balance, particularly through the Th2/IgE pathway. Although direct evidence linking a HSD to AR remains limited, significant parallels have been identified in allergic airway diseases such as asthma, which also exhibit a Th2-dominant immune response. For instance, Demissie et al. demonstrated a significant association between childhood salt intake and increased bronchial hyperresponsiveness²³, while Musiol et al. directly observed a positive correlation between salt intake and asthma risk in women in a large population cohort study⁹. In experimental asthma models, Musiol et al. further showed that a HSD exacerbates eosinophilic inflammation and elevates Th2 cytokine levels (IL-4, IL-5, IL-13) in bronchoalveolar lavage fluid, consistent with our observations in the AR model⁹. However, studies on the direct effects of salt in human asthma patients have yielded mixed results, potentially influenced by factors such as gender. For example, Knox et al. found that increased salt intake enhances bronchial reactivity in males but had no significant effect on females²⁴, whereas Musiol et al. observed a positive correlation between salt intake and asthma incidence in females but not in males⁹. In contrast, our study found no significant gender differences in the relationship between salt intake and allergic responses, possibly due to similar dietary patterns among participants. This highlights the need for further research with larger sample sizes to validate these findings. Although no increase in Th17 responses is observed in our study, Matthias et al. demonstrated that NaCl under high-salt conditions can influence Th17 cell differentiation and function¹⁵. High-salt conditions can induce the formation of anti-inflammatory Th17 cells in certain microenvironments, such as those with high transforming growth factor-β cytokine levels, while promoting pathogenic Th17 cells in other pro-inflammatory microenvironments^15,25,26. Thus, a HSD may modulate the balance of different T cell subsets, influencing the switch between pro-inflammatory and anti-inflammatory states, thereby exacerbating allergic inflammation.

Our study showed that in AR mice fed a HSD, the composition of the gut microbiota is altered, and damage to the intestinal mucosal barrier is aggravated. This gut-centric disruption may provide a potential explanation for the associated immune impairment. Compared to the non-sensitized group, patients sensitized to inhalant allergens shows a significant reduction in α-diversity metrics²⁷. Similarly, in OVA-sensitized mice, the gut microbiota composition is notably altered, with a pronounced decrease in the proportion of the Bacteroidetes phylum²⁸. Bacteroides are highly adaptable and widely distributed in the gastrointestinal tract²⁸. They produce short-chain fatty acids (SCFAs) with anti-inflammatory and barrier-enhancing effects, regulating immune system²⁹. Notably, we observed an elevated Firmicutes/Bacteroidetes ratio in the AR group, which may be related to dysbiosis of the microbiota³⁰. This aligns with Fu et al.’s finding that restoring this ratio balance, increasing microbial diversity, and enriching butyrate-producing bacteria can ameliorate allergic responses³¹. In our study, Colidextribacter is reduced in the AR group compared to the CON group. Miranda et al. demonstrated that a HSD reduces butyrate levels, exacerbating colitis³². Colidextribacter, which is associated with butyrate production, can regulate gut barrier function and inflammatory responses³³.

In this work, key barrier genes (including Muc2, Occludin, E-cadherin, and Zo-1) are downregulated in the small intestine of the HSD group, which are consistent with the known barrier decline driven by microbial dysbiosis³⁴. MUC2 is an important secreted protein in the human intestine that plays a role in protecting the intestinal barrier and regulating microbiota homeostasis, and its expression is reduced in mice with intestinal injury^35,36. The increase in Claudin 1 expression observed in our study is contrary to some findings. Elevated expression of Claudin 1 appears to increase the integrity of the epithelial barrier in the physiological state, but increased expression of Claudin 1 may be detrimental to the integrity of the epithelium in the inflammatory state³⁷. A study by Sarah et al showed that Claudin 1 expression is elevated in active ulcerative colitis and negatively correlated with butyrate levels³⁸. Beyond the gut, this interplay between barrier dysfunction and inflammation is also relevant in the nasal mucosa.

Building on the link between barrier disruption and inflammation, we explored the molecular pathways involved. The mechanism by which a HSD exacerbates AR may involve the upregulation of NFAT5, a transcription factor known to be associated with various immune diseases. Recent studies have found that abnormal expression of NFAT5 is associated with various immune diseases. For instance, deletion of the NFAT5 gene in young patients with anti-goblet cell antibody positive autoimmune enterocolic disease leads to a severe deficiency of immune function³⁹. Kleinewietfeld et al. indicated that inhibiting NFAT5 suppresses high-salt-induced Th17 cell development⁴⁰. NFAT5, also known as tonicity-responsive enhancer-binding protein (TonEBP), is a member of the Rel family of transcription factors and is initially discovered as a key factor in cell homeostasis maintenance under hypertonic stress conditions⁴¹. Under hypertonic conditions, the expression of Nfat5 is significantly activated⁹. Consistent with its role as a tonicity-responsive factor, we confirmed that hypertonic stress (50 mM NaCl) induces Nfat5 expression in immune and epithelial cells, highlighting its potential role in allergic inflammation⁴². After stimulation of CD4⁺ T cells with hypertonic medium, NFAT5 is increased, along with the expression of IL-2 and TH17-related gene products⁴³. Gene silencing of NFAT5 in CD4⁺ T cells inhibits the expression of Th17-related genes IL-17 and Rorγt, reducing inflammation and has been reported in autoimmune diseases⁴⁰.

The direct downstream target of NFAT5 is SGK1, which is involved in multiple cell transduction pathways and is closely related to immune regulation⁴⁴. In mice and humans, elevated SGK1 under hypertonic conditions promotes the development of Th1 cells into Th2 cells, which contributes to allergic diseases⁴⁵. Reducing the activation of SGK1 reduces allergic asthma and increases anti-tumor and antiviral immune response⁴⁶. Activation of SGK1 also activates various epithelial ion channels and subunits that regulate fluid clearance in various body parts⁹. Patients with AR increase nasal secretions, and histamine in secretions can aggravate the progression of AR by disrupting the epithelial barrier function⁴⁷. The increase in Sgk1 caused by a HSD undoubtedly aggravates the allergy symptoms. Hypertonic stress-induced nasal barrier damage and SGK1 upregulation were reproduced in both in vivo and in vitro experiments; these effects were shown to be reversible with SGK1 inhibition in vitro. The activation of SGK1 and NFAT5 depends on the regulation of P38/MAPK⁴⁸. Previous studies have shown that P38/MAPK is increased under the condition of high osmotic pressure⁴⁸. Our in vitro and in vivo experiments showed that hypertonicity elevates P38/MAPK activity; conversely, P38 inhibitor treatment suppresses β-Hex release and decreases Zo-1 expression.

We further investigated the reversibility of these effects. Animal experiments revealed that while a short-term HSD is associated with exacerbated AR symptoms, these could be partially alleviated by a short-term return to a normal diet, although IgE levels remains elevated. In addition, short-term return to a normal diet may activate the relevant pathways by decreasing the osmotic pressure of the nasal mucosa, thus reducing inflammatory response. Although returning to a normal diet can reversibly improve symptoms and reduce NFAT5 pathway activation, OVA-IgE exposure highlights the persistent effects of immune memory. These findings indicated that individuals with AR who avoid consuming a HSD may experience a reduction in symptoms, especially during allergy season or immunosensitive periods.

In conclusion, our study provides novel evidence linking a HSD to exacerbated AR progression, potentially mediated via the P38/MAPK-NFAT5-SGK1 signaling pathways. Meanwhile, timely dietary intervention may alleviate AR burden. This finding not only enhances our understanding of the pathological mechanisms underlying AR but also opens new perspectives for its prevention and treatment. This study also has several limitations: gut dysbiosis and barrier impairment were observed concurrently with a HSD, their specific causal contribution to the nasal symptoms requires future investigation; The human study is correlational and may have confounding factors, making it hard to establish a causal relationship, and further research is needed to verify the mechanism in human population while addressing possible confounding variables.

Method details

Participant recruitment

The population of volunteers was recruited in a tertiary grade-A hospital in Nanjing, China. The whole process was performed according to the Declaration of Helsinki, approved by the IEC for Clinical Research of Zhongda Hospital, Affiliated to Southeast University on December 6, 2023 (no. 2023ZDSYLL410-P01). All the participants provided their written informed consent. The patients with AR were diagnosed according to the Chinese Guidelines for the Diagnosis and Treatment of AR (Revised Edition 2022). AR patients were diagnosed if three criteria were met: The patient has recurrent sneezing, itchy nose, heavy runny nose, nasal congestion, or a combination of two or more of these symptoms; Both nasal mucous membranes were pale and edema, with a large amount of watery nasal secretions; A positive skin prick test or serum specific IgE antibody level exceeding 0.35 KUA/L. The inclusion criteria were: age 18–65 years, absence of diagnosed mental disorders, willingness to complete the questionnaire and provide biological samples, diagnosis of AR according to established criteria, no use of allergy medications within the preceding two weeks or corticosteroids within the preceding week, and voluntary provision of signed informed consent.

Exclusion criteria comprised: a history of non-IgE-mediated allergy; significant cardiovascular, cerebrovascular, hepatic, renal, oncologic, hematologic, or cognitive disorders, or any other condition that could compromise study compliance; and use of probiotics, prebiotics, synbiotics, or similar supplements within the past three months, as well as antibiotics or gastric acid suppressants during the same period. Salt intake was assessed by measuring 24 h urinary Na⁺ levels using inductively coupled plasma mass spectrometry (ICP/MS). Participants were stratified into the HSD group and low-salt diet group. Nasal symptoms in patients with AR were evaluated by a specialist physician according to the scoring system presented in Table S1.

Animal and diets

Balb/c mice (6-week-old, 18–20 g) were purchased from the Beijing Vital River Experimental Animal Co. (Beijing, China). The mice were kept in specific pathogen-free (SPF) environment with a temperature of 22 °C, 50% humidity and a 12 h light/dark cycle. The mice were randomly separated into three groups after one-week adaptive feeding. The CON group and the AR group received the AIN-76 diets while the HSD group received AIN-76 with 4% NaCl.

For short-term dietary interventions: the PSD group received AIN-76 diets with 4% NaCl; the PND group received AIN-76 diets without NaCl; the SSD group consumed the AIN-76 diet and transitioned to AIN-76 diet with 4% NaCl for the last week; conversely, the SND group was initially supplemented with AIN-76 diet with 4% NaCl and then shifted to the standard AIN-76 diet for the final week.

Mice were euthanized by CO₂ inhalation. Detailed analyses were conducted including cytokine release in nasal lavage fluid (BAL), T-cell typing of splenocytes, serological assays, and measurement of Na⁺ and K⁺ concentrations. Stool and tissue samples were collected for microbial detection and pathway analysis. The animal study protocol was approved by the Animal Experiment Ethics Committee of Southeast University (approval no. 20240324005).

Induction of AR

After feeding for 5 weeks, AR was induced using an established OVA model. Starting from day 39, mice were sensitized by intraperitoneal injection with 50ug OVA (Sigma-Aldrich, USA) and 2 mg aluminum hydroxide adjuvant (Sigma-Aldrich, USA) in a total volume of 200 μL three times every 7 days. Beginning on day 60, 400ug of OVA was administered intranasal stimulation for 5 days. Mice in the CON group were pseudo-sensitized with Phosphate Buffered Saline (PBS). Each group of mice was assessed for symptoms within 15 min of the last provocation, and allergy symptoms were scored using the criteria listed in Table S2⁴⁹.

Analysis of BAL, nasal histology and serology

Serum levels of OVA-specific IgE (Cusabio, China) and histamine (Elabscience, China) in mice were measured using ELISA kits. The concentration of inflammatory cytokines IL-4, IL-5, IL-13, IFN gama, TSLP and IL-17 (Abclonal, China) were detected by ELISA kits from the supernatant of centrifuged mouse BAL. The nasal tissues were fixed with 4% paraformaldehyde and embedded in paraffin. Hematoxylin-eosin (H&E) stained sections were used to analyze the nasal epithelial injury.

Flow cytometry analysis of spleen cells

Mouse spleen was removed and placed in PBS, chopped with scissors and filtered with 70um cell filter. The spleen was rinsed with PBS three times after grinding. The filtrate was collected into a centrifugal tube, centrifuged at 1300 rpm for 7 min, and then the supernatant was discarded. Then, 2 mL of red blood cell lysis buffer was added, and the sediment was resuspended and lysed at room temperature for 4 min. The lysis reaction was stopped by adding 3 mL of PBS medium. After centrifugation at 1300 rpm for 7 min, the supernatant was discarded. Finally, the cells were resuspended in 10% RPMI 1640 medium.

Cells were stimulated with 50 ng/ml PMA, 750 ng/ml ionomycin, and 10 μg/ml brefeldin A for 5 hours. Spleen cells were taken and suspended with 100 μL PBS, followed by addition of 0.5u g FITC Anti-Mouse CD4 (BD Biosciences, USA) and incubation at 4 °C for 30 min. After washing the cells with PBS, 1 mL fixation/permeabilization buffer was added to each tube and the mixture was pulsed vortex. It was then incubated at 2–8 °C for 30–60 min in the dark. Next, 1 mL of permeabilization buffer was added to each tube, and then the samples were centrifuged at 600 g at room temperature for 5 min. The supernatant was discarded and 0.5ug APC Anti-Mouse IFN-γ (BD Biosciences, USA) antibody, 0.5 ug PE Anti-Mouse IL-4 (BD Biosciences, USA) were added, followed by incubation at room temperature in the dark for at least 30 min. After adding 1 mL of permeabilization buffer and centrifuging at 600 g for 5 min at room temperature, the supernatant was discarded. The sediment was resuspended in 200 μL of assay buffer and analyzed by flow cytometry.

Real-time quantitative PCR (RT-qPCR)

Total RNA was extracted from cells and tissues samples of the jejunum, ileum and nasal mucosa using TRIzol reagent (Vazyme, China). The extracted RNA was subsequently reverse-transcribed into complementary DNA (cDNA) using reverse transcriptase (Vazyme, China). RT-qPCR was conducted following the manufacturer’s protocol with the SYBR Green PCR Mix Kit from Vazyme (China). The PCR primers sequences are given in supplementary data (Table S3).

ICP/MS

For the ICP/MS analysis, 10 μL mouse serum, human plasma, or human urine were taken and treated with 10 mL of 5% nitric acid. The samples were then subjected to ultrasonication for 10 minutes to ensure complete dissolution and homogenization. The sonicated samples were filtered through a 0.45 μm aqueous phase filter membrane to remove particulates. The filtered samples were subsequently analyzed using an ICP/MS instrument (Agilent 8900, USA), with specific mass numbers selected for the detection of Na⁺ and K⁺ ions. To mitigate cross-contamination, the needle was cleaned after analyzing every three samples were tested. The signal strength of Na⁺ and K⁺ for each sample was recorded, and the concentration was calculated from the standard.

Western blot analysis

Mouse nasal mucosa tissues and cell samples were collected and proteins were extracted using RIPA lysis buffer. Protein concentrations were quantified using BCA assay kit. Equal amounts of protein were loaded onto a SDS-PAGE and subsequently transferred to a PVDF membrane. The membrane was blocked for 1 hours and incubated with the primary antibodies overnight against p-P38 (Abclonal, China), P38 (Abclonal, China), NFAT5 (Proteintech, USA), SGK1 (Proteintech, USA), and Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Abcam, USA). Following TBST washes, the membrane was incubated with the secondary antibody for 1 hour. Chemiluminescence was induced by adding the ECL substrate, and the protein intensity were visualized using a chemiluminescence imaging system. Quantitative analysis of the blots was conducted using ImageJ software. The phospho-P38 measurements were normalized by total P38 levels. The complete WB images are shown in Figs. S8–S12.

Immunofluorescence Analysis

The paraffin-embedded tissue sections were deparaffinized, followed by antigen retrieval. After the repair completion, the section was allowed to cool naturally. The slices were sealed with bovine serum albumin for 30 minutes, washed and dried. The slices were incubated with primary antibodies of anti-OCCLUDIN (Servicebio, China) and anti-ZO-1 (Servicebio, China) overnight at 4 °C. Subsequently, the slices were incubated with the secondary antibody at room temperature for 50 min in a dark environment. The slices were washed thrice times with PBS, stained with DAPI dye solution, and incubated at room temperature in the dark for 10 min. After additional PBS washing, an autofluorescence quencher was applied for 5 min and washed with water for 10 min. Finally, anti-fluorescence quenching sealing tablets were used and the signal was captured using a fluorescence microscope.

Sampling and sequencing for microbiome analysis

The fresh feces form the mouse were collected in an sterile tubes and stored at -80°C. The total DNA was isolated from fecal samples using the TGuide S96 magnetic bead fecal Genomic DNA Extraction kit (Tiangen, China). The V1-V9 region of the 16S rDNA was targeted for amplification with the primer pair 27 F and 1492 R. The amplicon was purified and quantified. The SMRTBELLS PREP KIT 3.0 was employed for damage repair, end repair, and adapter ligation of the pooled products. The library was then purified and concentrated using AMpure PB magnetic beads to prepare it for sequencing. The purified final product was sequenced on the REVIO sequencer to obtain high-throughput sequencing analysis. Sequencing data quality assessment encompassed several critical steps: Circular Consensus Sequencing (CCS) sequence acquisition, primer identification and removal, and chimera elimination. Initially, raw subreads were corrected to generate CCS sequences. Next, forward and reverse primers were identified using cutadapt v1.8.3. Then, CCS sequences lacking primers were discarded. Subsequently, CCS lengths were filtered, with sequences falling outside the expected range (16S: 1200 bp–1650 bp) were removed. Finally, chimeras were removed to yield high-quality CCS sequences. Our exclusion criteria for low-quality or low-coverage samples were based on the following parameters: 1) Sample integrity: Degraded samples (with failed PCR amplification) are excluded; 2) Effective CCS threshold: Samples with fewer than 10,000 effective CCS reads were discarded; 3) Rarefaction curve: Samples whose alpha dilution curves fail to reach the plateau phase were excluded; 4) Clean CCS / Raw CCS ratio: Samples with a ratio below 80% were removed; 5) Effective (%): This metric represents the percentage of effective CCS reads relative to clean CCS reads, and samples below 80% were excluded; 6) AvgLen: Sequences with average lengths outside the reasonable range of 1 kb to 3 kb were discarded; 7) Outlier detection: Abnormal samples identified as outliers in the PCoA plot were excluded. The Alpha Rarefaction Curve is illustrated in Fig. S7, and statistics of sample sequencing data processing results are summarized in Table S4. The α diversity (ACE index) and β diversity (binary_jaccard) were calculated using OTU to determine the relative abundance of taxa at multiple levels. Linear discriminant analysis effect size (LEfSe) was used to identify bacterial species that contributed significantly to differences in microbiota. The Linear Discriminant Analysis (LDA) threshold was set at 3.5.

Cell Culture

RBL-2H3 and RPMI-2650 cells (Zhong Qiao Xin Zhou, China) were cultured to optimal passage density. The appropriate density of cells was separated from the culture vial using a 0.25% trypsin-EDTA solution. The resulting cell suspension was then transferred to a fresh culture bottle having fresh medium and incubated. Furthermore, RBL-2H3 and RPMI-2650 cells were inoculated into 96-well plates at a density of approximately 1 × 10⁴ cells per well and allowed to grow overnight. According to the CCK8 kit instructions, the toxicity of the intervention to the cells was evaluated, and the appropriate concentration was selected for the experiment. For RBL-2H3 cells, after overnight cell culture, the patient serum with egg allergy was diluted at a ratio of 1:25 per well for 16–24 h. After washing with PBS once, 50 mM of NaCl was added for 1–2 h, followed by PBS washing along. 40 μg/mL OVA was added for 1 h The cell culture supernatant was collected to detect β-Hex activity. In the RBL-2H3 cell experiment, different inhibitors were used: the P38i, the KRN5, or the SGKi. The inhibitor was added to the culture at a concentration of 1 μM, 5 μM, or 10 μM, respectively, and co-cultured with NaCl for 24 h. Cells were collected for follow-up experiments. In the RPMI-2650 cell experiment, the P38i or SGKi was added to the culture at a concentration of 1 μM or 10 μM, respectively, followed by co-incubation with NaCl for 24 hours.

Lentiviral-Mediated Gene Silencing

Lentiviral particles expressing shRNAs were procured from GenePharma (NFAT5: 5’-AGCACTCGTGCCAGATTGGTT-3′). RPMI-2650 cells were cultured overnight after seeding. RPMI-2650 cells were cultured overnight and then transduced either by viral particles containing vectors expressing specific shRNA or by control vectors expressing non-specific shRNA. After incubation for 120 hours, the cells sustained purinomycin selection to isolate transduced cells successfully. qRT-PCR was used to quantitatively detect the expression level of target genes.

Statistical analysis

SPSS software was used for statistical analysis. GraphPad Prism (version 10) was employed to generate graphs, and the BioRender website was utilized to create schematic mechanism diagrams. For comparison between the two groups, we used the unpaired T-test and for analysis of differences among more than two groups, we used one-way ANOVA and LSD post hoc test. These analysis were considered statistically significant at p < 0.05.

Author contributions

**L.J.** : Study design, Methodology, Validation, Writing-Original Draft, Writing-Review & Editing. **Y.W.**: Methodology. **J.H.**: Study design, Writing-Review & Editing. **S.Y., Y.Y., A.Z.,** and **V.R.** : Writing-Review & Editing. **L.C.**: Data Curation; Writing-Review & Editing. **J.W.**: Study design, Investigation, Validation, Supervision, Project Administration, Funding Acquisition, Writing-Review & Editing.

Data Availability

The datasets generated and analyzed during the current study are not publicly available due to privacy and ethical restrictions but are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41538-026-00760-4.