The modulatory effect of high salt on immune cells and related diseases

Abstract Background The adverse effect of excessive salt intake has been recognized in decades. Researchers have mainly focused on the association between salt intake and hypertension. However, studies in recent years have proposed the existence of extra‐renal sodium storage and provided insight into the immunomodulatory function of sodium. Objectives In this review, we discuss the modulatory effects of high salt on various innate and adaptive immune cells and immune‐regulated diseases. Methods We identified papers through electronic searches of PubMed database from inception to March 2022. Results An increasing body of evidence has demonstrated that high salt can modulate the differentiation, activation and function of multiple immune cells. Furthermore, a high‐salt diet can increase tissue sodium concentrations and influence the immune responses in microenvironments, thereby affecting the development of immune‐regulated diseases, including hypertension, multiple sclerosis, cancer and infections. These findings provide a novel mechanism for the pathology of certain diseases and indicate that salt might serve as a target or potential therapeutic agent in different disease contexts. Conclusion High salt has a profound impact on the differentiation, activation and function of multiple immune cells. Additionally, an HSD can modulate the development of various immune‐regulated diseases.

immunomodulatory effect of high salt. We will first introduce sodium homeostasis and its physiological functions. Then, we focused on the effect of high salt on various immune cells in microenvironments. Furthermore, the influence of HSD on immune-regulated diseases and relevant immune responses will be reviewed. Finally, we will briefly discuss the present and potential clinical applications of high salt.

| SODIUM HOMEOSTASIS AND PHYSIOLOGICAL FUNCTIONS
Dietary salt is an important source of sodium, of which homeostasis is tightly regulated. 5 Sodium homeostasis in the traditional twocompartment model is regulated by the kidney. 6 However, in recent years, studies have suggested that large amounts of Na + are stored in extra-renal tissues, particularly the skin and muscles. [7][8][9][10] Mechanistically, excessive Na + in bone, cartilage and skin is stored by non-osmotic binding with polyanionic proteoglycans, generating a hypertonic environment in local tissue and acting as an osmotically inactive Na + reservoir. [11][12][13][14] Moreover, tissue sodium storage is linked to certain diseases. Increased skin or muscle Na + storage can be observed in patients with multiple sclerosis (MS), 15 refractory hypertension, 10 lipedema, 16 systemic sclerosis 17 or end-stage renal disease. 18,19 Tissue Na + storage may reflect the activity and progression of some diseases, such as systemic lupus erythematosus (SLE), MS and psoriasis. [20][21][22][23][24] Of note, Na + tends to accumulate in inflammatory and infection sites. In subtotal-nephrectomized mice (5/6Nx) fed an HSD, sodium storage in the abdominal wall tissues was increased, 25 and the total sodium concentration of patients with triple-negative breast cancer was two to threefold higher in the tumour than in normal tissue. 26 This dynamic and tightly regulated sodium homeostasis contributes to critical physiological functions in the body. 27 Specifically, sodium is the major osmotically effective cation that keeps the plasma volume within normal limits and maintains suitable cell volume necessary for cellular survival and function. [28][29][30] The resting membrane potential of cells largely depends on the transmembrane sodium concentration gradient, whereas the generation of action potential is mediated by a brief sodium influx. 28,31 Sodium also participates in various metabolic reactions. Data from humans showed that severe sodium deprivation can negatively affect glucose metabolism. 32 Moreover, an adequate sodium intake is required for mammalian survival and development. 33,34 The growth of protoplasm, fat and bone was retarded in sodium-deprived rats, 35 while gestational sodium restriction can impair foetal growth and nervous system development. 36,37 3 | THE EFFECT OF HIGH SALT ON IMMUNE CELLS Immune cells comprise innate and adaptive immune cells. Innate immune cells include granulocytes, natural killer (NK) cells, macrophages, monocytes and dendritic cells (DCs). Adaptive immune cells include T lymphocytes and B lymphocytes, which express T cell receptor (TCR) and B cell receptor (BCR), respectively. 38 Extensive studies have suggested the immunomodulatory function of salt, and we particularly focus on the effect of high salt on the differentiation, activation and function of immune cells in microenvironments. Furthermore, considering the difference between mouse and human immune cells, 39 the high salt effects on immune cells showed discrepancies among species (Figure 1).

| The effect of high salt on immune cell proliferation and death
In humans and rodents, plasma Na + concentration is approximately 140 millimolar (mM), while Na + concentrations in interstitium and lymphoid tissue range between 160 and 250 mM. 40,41 Based on these findings, most studies enrich the media with additional 20-100 mM NaCl to investigate the effect of high salt on immune cells. [42][43][44][45] Importantly, adding 40 mM NaCl can simulate the Na + concentrations found in the skin of high-salt-diet rodents. 46 These clinically relevant high salt can trigger osmoprotective responses in immune cells and affect cellular homeostasis.
Studies have demonstrated that immune cells rely on the nuclear factor of activated T cells-5 (NFAT5 or TonEBP) to adapt to hypertonic stress. 47 The guanine nucleotide exchange factor (GEF) Brx and p38 mitogen-activated protein kinase (MAPK) can activate NFAT5 and mediate the osmoprotective response in immune cells. 48,49 In terms of immune cell proliferation, NFAT5 can induce osmoprotective responses that enable T lymphocytes to resume cyclin expression. 50 Furthermore, CD24 was identified as an NFAT5-regulated gene involved in the adaptation of murine proliferating T cells to high salt. 51 Additionally, the exposure conditions and cell lines are critical. Human and murine T cell proliferation was suppressed at additional NaCl above 40 mM, but increasing NaCl below 40 mM promoted T cell proliferation. [51][52][53][54][55] The augmented T cell proliferation is related to p38/MAPK activation, 52 enhanced monocyte function 53 and altered macrophage function. 45,56 Moreover, adding 40 mM NaCl can inhibit B cell proliferation. 57 In addition, high NaCl can damage cytoskeleton, increase DNA breaks, inhibit translation and cause cell cycle arrest. [58][59][60] Osmoprotective mechanisms enable cells to survive and function. Failure to accommodate leads to cell death. 58 Partial loss of NFAT5 function reportedly led to impaired lymphocyte growth and function. 40,61 Additionally, salt exposure conditions and cell lines can influence the outcome of adaptation. In PBMCs, additional NaCl below 80 mM had no significant effect on cell apoptosis, whereas severe cell death was observed at 800 mM. 62 Adding 100 mM NaCl for 24 h potentiated murine macrophage apoptosis, along with inhibited protein kinase B (PKB or Akt) activation. 63,64 Moreover, adding 40 mM NaCl had little impact on the growth or apoptosis of CD4 + T cells, but increasing concentrations further led to cell death. 43,65 3.2 | The effect of high salt on innate immune cells

| Neutrophils
Neutrophils, a type of polymorphonuclear leukocyte, are the major effectors of acute inflammation. 66 High-salt treatment after human neutrophil activation can increase superoxide production and elastase release. 67,68 In contrast, high salt pretreatment can suppress diverse human neutrophil functions such as degranulation, 69 superoxide production, 70,71 phagocytosis, 72 adhesion 73,74 and migration. 42,75 These effects are related to adenosine triphosphate (ATP) release, p38/MAPK activation and cytoskeletal remodelling. [76][77][78][79] Moreover, NaCl above 209 mM can significantly suppress neutrophil extracellular trap (NET) formation while promoting apoptosis in human neutrophils. 80 Consistently, high NaCl can impaire the antimicrobial capacities of both human and murine neutrophils. In human neutrophils, the reduced bactericidal activity was due to the decreased reactive oxygen species (ROS) production. 81 Furthermore, an HSD in mice can increase the urea accumulation in the renal medulla and elevate the glucocorticoid level in the blood, resulting in the suppressed neutrophil antibacterial functionality. Healthy humans who accepted an HSD also showed hyperglucocorticoidism and inhibited neutrophil function. 82 In summary, the sequence of high-salt exposure and stimulation critically affects the outcome of human neutrophil activation, whereas high salt can suppress the antibacterial functionality of murine neutrophils.

| Natural killer cells
NK cell is a prototypical member of Group 1 innate lymphoid cells (ILC1) that can produce interferon-γ (IFN-γ). 83,84 A study demonstrated that HSD suppressed the proliferation, activation and function of NK cells in mice. 85 Mice fed an HSD displayed decreased NK cells in the spleen and lungs, while NK cell maturation in the spleen and bone marrow was blocked. Mechanistically, HSD downregulated CD122 expression in NK cells via ROS signalling and thus reduced the responsiveness to interleukin (IL)-15 in NK cells and inhibited their function. 85 However, Rizvi et al. 86 found that an HSD in mice enhanced NK cell function in tumour immunity by decreasing PD-1 expression and increasing IFN-γ and serum hippurate. These two distinct conclusions indicated the importance of the disease context in the high salt effect on murine NK cells.

| Mononuclear phagocyte family
Macrophages, DCs and monocytes form a family of mononuclear myeloid cells that are specialized in antigen presentation. 87 These three myeloid cells also belong to the mononuclear phagocyte system (MPS), a concept encompassing progenitors, macrophages, DCs and monocytes. 88 Tissue-resident macrophages can be derived from embryonic progenitors or they can be of monocytic origin under F I G U R E 1 The effect of high salt on innate and adaptive immune cells. High salt enhances the formation of IsoLG-protein adducts in murine dendritic cells, along with increased IL-1β secretion and subsequent activation of T cells. An HSD increases circulating monocytes in mice and humans, and high salt can induce pro-inflammatory human monocytes. Moreover, high salt induces M1 macrophages, Th17 cells and Th2 cells, while suppressing M2 macrophages, Treg cells and Th1 cells. The differentiation of human Tfh cells is increased by high salt. The sequence of high salt exposure and stimulation affects the outcome of human neutrophil activation, whereas the antibacterial functionality of murine neutrophils is suppressed by high salt. The high salt effects on murine NK cells and CD8 + T cells are context-dependent, while murine B cells respond to high salt in a biphasic manner.
inflammatory conditions, the relative proportion of which depends on the tissue. 89,90 DCs and monocytes originate from haematopoietic stem cells, but they have distinct precursors. 88 Moreover, DCs comprise conventional DCs (cDCs) and plasmacytoid DCs (pDCs). 91 In this review, we considered cDCs and not pDCs. Monocytes demonstrate plasticity and can be differentiated into macrophages and dendriticlike cells. 92

Macrophages
Macrophage activation phenotypes have been extended to a spectrum model with two extremes, M1 and M2 macrophages. 93 M1 macrophages are polarized by inflammatory stimuli, mediating the immune response to bacteria and intracellular pathogens. M2 macrophages are anti-inflammatory macrophages, existing in settings such as helminth immunity, asthma and allergy. 94,95 Murine macrophages. Generally, high salt potentiates the activation and function of M1 macrophages, while suppressing the activation and function of M2 macrophages ( Figure 2). 45,56,96,97 First, high salt can induce p38/MAPK-dependent NFAT5 activation in M1 macrophages, resulting in inducible nitric oxide synthase (iNOS or NOS2)dependent nitric oxide (NO) production and tumour necrosis factor (TNF) secretion. 41 These responses boosted the function of macrophages, facilitating anti-leishmanial control and regulating salt-induced hypertension. 41,46 Additionally, the Na + /Ca 2+ exchanger 1 (NCX1) expressed on murine macrophages can sense Na + and contribute to NFAT5 accumulation. 98,99 Second, high salt can activate caspase-1 and trigger IL-1β release in macrophages. These responses required the activation of the nucleotide-binding domain and leucine-rich repeat pyrin domain 3 (NLRP3) and the nucleotide-binding domain and leucine-rich repeat caspase recruitment domain 4 (NLRC4) inflammasomes via mitochondrial ROS. These salt-activated inflammasomes can act as innate sensing components and promote T helper (Th)17 response. 44 Third, the p38/cFos/ activator protein 1 (AP1) and extracellular signal-regulated kinase (Erk) 1/2/cFos/AP1 pathways mediated the salt-driven pro-inflammatory profile in M1 macrophages, while the Erk1/2/signal transducer and activator of transcription (STAT)6 pathway mediated the salt-induced suppression of M2 macrophages. 97 Furthermore, high NaCl can activate macrophages without extra activators, leading to a novel activation state named M(Na). 97 Last, high salt can inhibit the mitochondrial respiration in macrophages, resulting in the improved bactericidal function of M1 macrophages and suppressed function of M2 macrophages. 45,56,99 The inhibited M2 macrophage activation by high salt is associated with blunted Akt/mechanistic target of rapamycin signalling. 56 Apart from the direct activation of pro-inflammatory macrophages, high salt can induce macrophage migration. In vitro, macrophages can migrate towards NaCl in a dose-dependent manner. 100 In subtotal-nephrectomized mice (5/6Nx), an HSD increased macrophage infiltration in the peritoneal wall, heart and para-aortic tissues. 25 The mechanisms involved the enhanced expression of F I G U R E 2 High salt induces a pro-inflammatory profile in murine macrophages. NCX1 expressed on M1 macrophages can sense extracellular Na + , and high salt increases NO and TNF production by p38/MAPK-dependent NFAT5 activation and downstream iNOS upregulation. Moreover, high salt activates NLRP3 and NLRC4 inflammasomes via mitochondrial ROS, thereby increasing IL-1β in a caspase-1-dependent manner. Additionally, high salt induces a pro-inflammatory profile in M1 and M(Na) macrophages through the p38/cFos/AP1 and Erk1/2/cFos/ AP1 pathways, whereas the Erk1/2/STAT6 pathway mediates the salt-driven suppression of M2 and M(Na) macrophages. In M2 macrophages, high salt downregulates the expression of Arg1, mannose receptor, C type 1 (Mrc1), inflammatory zone 1 (Fizz1), chitinase-like 3 (Chil3 or Ym1), interferon regulatory factor 4 (IRF4) and macrophage mannose receptor (MMR). Furthermore, high salt suppresses mitochondrial metabolism and AKT/mTOR signalling in M2 macrophages. The mitochondrial function of M1 macrophages is also suppressed by high salt.
NFAT5-dependent monocyte chemotactic protein-1 (MCP-1) in mesothelial cells and cardiomyocytes. 25 Human macrophages. High salt can induce M(Na) in human macrophages, characterized by enhanced pro-inflammatory and suppressed anti-inflammatory gene expression. 97 Salt-activated M1 macrophages with increased pro-inflammatory cytokine production were also observed. 45,97 Moreover, high salt inhibited the mitochondrial respiration in M1 and M2 macrophages, resulting in the suppressed function of M2 macrophages. 99 However, studies have identified the discrepancies between mouse and human macrophages. For instance, iNOS and arginase-1 (Arg1), two enzymes important for murine macrophage arginine metabolism, might not be functional in human macrophages. 39,101 Thus, the investigations in murine macrophages cannot be simply extrapolated to human macrophages, and the high salt effect on human macrophages needs to be further explored.

Monocytes
Two studies in healthy subjects have revealed a positive association between salt intake levels and monocyte numbers. 102,103 Data showed that minimal changes in plasma sodium concentration caused by HSD can transiently inhibit the mitochondrial function of human circulating monocytes. 99 A recent study found that high salt can drive human monocytes to a DC-like phenotype, characterized by the formation of isolevuglandin (IsoLG) adducts, expression of CD83 and increased production of IL-1β. These responses induced subsequent T-cell activation with increased IL-17A production. 104 Consistently, monocytes from humans with high skin sodium exhibited enhanced IsoLG adduct accumulation and CD83 expression. 104 Other studies also observed pro-inflammatory human monocytes induced by high salt in vitro or by an HSD. 45,105 In addition, a long-term HSD remarkably increased the circulating monocytes in mice by driving monocyte mobilization from bone marrow. 106 Taken together, an HSD can increase circulating monocytes in mice and humans, and high salt can induce pro-inflammatory human monocytes.

Dendritic cells
Current evidence has revealed the impact of high salt on murine DCs.
High salt impeded the cross-priming capacity of murine DCs in a toll/ IL1 receptor domain-containing adapter-inducing interferon-beta (TRIF)-dependent manner. 107 Moreover, murine DCs treated with high salt acquired an M2-like signature. 108 Nevertheless, the maturation, antigen presentation and inflammatory cytokine expression of murine DCs were reportedly enhanced by high salt. 107,109,110 Specifically, elevated sodium enters DCs through amiloride-sensitive channels and consequently increases the formation of IsoLG adducts and the production of IL-1β. These salt-activated DCs can promote the secretion of IL-17A and IFN-γ in T cells. 110 Data in mice fed an HSD corroborated these findings. 111 In addition, an HSD in SLE mice can activate DCs through the p38/MAPK-STAT1 pathway. 112 Tubbs et al. 113 found that the increased cytokine production in murine DCs upon high NaCl treatment was mediated by p38/MAPK, serum and glucocorticoid-inducible kinase 1 (SGK1) and downstream toll-like receptor 4 (TLR4). Collectively, these findings suggested that high salt can activate murine DCs and increase their production of inflammatory cytokines. deficiency. 43,126 In addition, Na + -K + -2Cl À cotransporter 1 may participate in sensing extracellular NaCl in murine Th17 cells and mediate the salt-induced increase in SGK1 and IL-23R. 127 In contrast, Matthias et al. 128  In addition, the high salt effect on Th17 cells can be mediated by gut microbiota and their metabolites. 129 Indeed, an HSD in mice significantly reduced several intestinal bacteria, particularly the Lactobacillus murinus. Studies have indicated that increased salt consumption results in the depletion of L. murinus and thereby induces proinflammatory Th17 cells in mice. 130,131 Mechanistically, Lactobacilli can metabolize tryptophan to indole metabolites. Faecal indoles, confirmed to be an inhibitory factor of Th17 differentiation, were reportedly decreased by HSD. 130 Further investigations suggested that oral gavage of L. murinus or Lactobacillus reuteri in mice restored the decrease of faecal indoles caused by HSD and blunted high-saltinduced Th17 activation. Consistently, a high salt challenge on healthy volunteers increased the peripheral blood Th17 cells, accompanied by reduced intestinal Lactobacillus spp. survival. 130 In summary, these studies linked HSD to the gut-immune axis and suggested that HSD can induce pro-inflammatory Th17 cells by decreasing gut Lactobacillus spp. in mice and humans.

Treg cells
High salt enables murine Treg cells to acquire an SGK1-dependent, pro-inflammatory phenotype, which has proved to be Th1-like 132 or Th17-like 133 in different reports, leading to inhibited suppressive function of Treg cells. By contrast, a salt-induced Th1-like phenotype has been observed in human Treg cells (Figure 4).
On the one hand, high NaCl impaired the suppressive capacity of murine and human Treg cells in vitro and in vivo, leading to a Th1-type phenotype with increased SGK1-dependent IFN-γ secretion. 132 Specifically, FOXO1, a downstream target of SGK1, can stabilize the Foxp3 locus alongside FOXO3. Treg cells exposed to high salt displayed enhanced phosphorylation of FOXO1/FOXO3, and the decreased Foxp3 stability may be responsible for Treg suppression. 132 On the other hand, Yang et al. 133 found a Th17-like phenotype of murine Treg cells with preserved suppressive function upon high salt treatment. These salt-induced RORγt + Foxp3 + Treg cells were associated with SGK1 activation and were not significant producers of IL-17A. 133 Consistently, an HSD decreased the Treg proliferation and proportion in a murine transplantation model. 55 Although it has been corroborated that high NaCl can promote RORγt expression in murine tTreg cells, whether high salt affects murine iTreg cells is controversial. 133,134 Data also suggested that high salt had little impact on the development and function of human iTreg cells. 134

Other T helper cells
High salt can induce human Th2 but suppress Th1 cell responses on multiple regulatory levels. 65,135 IL-4 and IL-13, the Th2 signature cytokines, were enhanced in human naïve and memory T cells under high NaCl conditions, whereas the production of IFN-γ was decreased.
These effects were dependent on NFAT5 and SGK1. 65 Consistently, an HSD augmented Th2 responses in food allergy mice, 136 while the high-salt formulation of Al(OH) 3 enhanced the ovalbumin (OVA)induced Th2 response in mice. 109 In addition, high NaCl can significantly increase the polarization of human Tfh cells. 62 Mechanistically, high NaCl enhanced the gene expression of ten-eleven translocation (TET)2 and TET3 in human CD4 + T cells, along with upregulated F I G U R E 3 The context-dependent effect of high salt on Th17 cells. (A) In the presence of Th17-polarizing cytokines, high salt boosts the induction of pathogenic murine and human Th17 cells. These responses are related to p38/MAPK, NFAT5 and SGK1 activation. (B) In the absence of Th17-polarizing cytokines, high salt induces antiinflammatory human Th17 cells by a significant upregulation of FoxP3 and the autocrine of TGF-β. However, additional pro-inflammatory cytokines can block TGF-β secretion, and therefore, human Th17 cells exhibit a pro-inflammatory phenotype. Similarly, high salt shows context-dependent effect on the pathogenicity of murine Th17 cells, but the autocrine TGF-β production was below the detection level. expression of sialophorin (spn), and then induced the DNA hypomethylation, ultimately contributing to enhanced Tfh cell differentiation. These findings provide a potential epigenetic mechanism for high-salt-induced autoimmunity. 62 3.3.2 | CD8 + T cells CD8 + T cells play a vital role in immunity to intracellular pathogens and tumours. 137 HSD increased the renal infiltration of CD8 + T lymphocytes in mice. 138,139 In an E.G7-OVA tumour mouse model, OVA/Al/high salt formulation displayed an enhanced antitumor effect via CD8 + cytotoxic T lymphocytes-mediated immunity. 109 By contrast, Popovic et al. 107 found that high NaCl inhibited the DCdependent activation of murine CD8 + T cells in a TRIF-dependent manner.

| B cells
B lymphocytes are immune cells that express clonally diverse cell surface immunoglobulin receptors. 140 High salt might activate murine splenic B cells and increase immunoglobulin production via the Brx/p38/MAPK/NFAT5 pathway. 49 Additionally, a short-term increase in NaCl concentration fostered murine B cell activation and differentiation, whereas chronic exposure to high NaCl dampened the differentiation of plasmablasts. Mechanistically, murine B cells treated with high NaCl showed inhibited p38/MAPK pathway activity and delayed NFAT5 response. 57

| THE EFFECT OF HIGH-SALT DIET ON IMMUNE-REGULATED DISEASES
Given the high salt effects on immune cells and the excessive sodium storage in local tissues caused by an HSD, 46 extensive studies have demonstrated the impact of an HSD on various immune-regulated disorders in mouse models (Table 1) and in humans ( Table 2).

| Hypertension and associated cardiovascular disease
It is well recognized that dietary salt intake has a direct causal relationship with blood pressure and can increase the risk of cardiovascular events and death. 166,167,193 Studies have demonstrated that an HSD can contribute to hypertension via the immune system. First, an HSD in mice can prime hypertension through DC-dependent T cell activation. 110,111 A study on human monocytes implied a similar mechanism. 104 Second, an HSD can induce the generation of Th17 cells by reducing Lactobacillus species, contributing to hypertension in mice and humans. 130 Last, salt-driven alterations in short-chain fatty acids (SCFAs), a subset of fatty acids generated by gut microbiota, were observed in hypertensives 194 and mice with salt-sensitive hypertension. 195 Given the immunomodulatory functions of SCFAs, 196 saltinduced hypertension might be mediated by SCFAs. In contrast, in response to skin Na + accumulation caused by HSD, interstitial MPS cells secrete NFAT5-dependent vascular endothelial growth factor-C (VEGF-C), which increases the hyperplasia of lymph capillaries and provides a buffering mechanism for salt-driven hypertension. 46,197,198

| Multiple sclerosis
MS is a chronic autoimmune demyelinating disease of the CNS. 199 EAE, the model of MS, can be induced by active or passive immunization. It can also be developed from opticospinal EAE (OSE) spontaneously. 200 Data showed that an HSD can aggravate actively induced EAE. First, exacerbated EAE by an HSD showed enhanced CNS infiltration and peripherally induced pathogenic Th17 cells. 43,126 Another study highlighted the role of gut microbiota in this process, as supplementation with L. murinus or L. reuteri blunted salt-induced pathogenic Th17 cells and ameliorated EAE exacerbation. 130 Second, EAE mice fed an HSD showed augmented macrophage infiltration in the CNS, and enhanced pro-inflammatory cytokine production in myeloid cells. 45 Last, high salt intake can exacerbate EAE by increasing blood-brain barrier (BBB) permeability. 141 This effect might be related to the decreased tight junction (TJ) proteins in endothelial cells. 158,201 However, another study reported that an HSD suppressed spontaneous EAE by Data from humans suggested that the high salt effect on MS is controversial. A cohort study reported that high salt intake was associated with increased disease activity in patients with MS. 168 Conversely, four other human studies found no association between HSD and MS progression. [169][170][171][172] However, these clinical studies measured dietary salt intake by spot urine collections or food questionnaires, which may lead to invalid results, 202,203 and more accurate measurement methods for salt intake are needed in follow-up studies.

| Intestinal bowel disease
IBD, encompassing ulcerative colitis and Crohn's disease, is a chronic relapsing inflammatory disorder of the gastrointestinal tract. 204,205 Data showed that an HSD can exacerbate DSS-and TNBS-induced colitis, leading to increased mortality in mice. 113,146 T A B L E 1 The effect of high-salt diet on immune-regulated diseases in mouse models This exacerbation of colitis is associated with the p38/MAPKdependent production of pro-inflammatory cytokines by intestinal mononuclear cells. 143,144 Further investigations corroborated the participation of CD4 + T cells and macrophages. 145 Depleting macrophages reduced the severity of DSS-induced colitis promoted by high salt intake. 144 Furthermore, an HSD exacerbated TNBSinduced colitis by enhancing the intestinal Th17 response. 146 The salt-driven reduction in Lactobacillus and butyrate levels might mediate this salt-induced Th17 response. 131 In addition, Type 3 innate lymphoid cells ILC3, a kind of IL-17-producing cell increased in the colon of high-salt-diet mice, may participate in the saltdriven aggravation of experimental colitis. 147 Moreover, an HSD can aggravate experimental colitis by blocking the suppressive function of Treg cells. 132 Last, data suggested that high dietary salt can promote intestinal fibrosis in TNBS-induced colitis by activating intestinal fibroblasts. 148 Nevertheless, a nested casecontrol study in women did not find an association between dietary sodium and the risk of ulcerative colitis or Crohn's disease. 173 More investigations in humans are warranted to clarify the role of HSD in IBD.

| Systemic lupus erythematosus and rheumatoid arthritis
SLE is an autoimmune, connective-tissue disorder that involves multiple systems. 206 Lupus nephritis is one of the most severe organ manifestations in SLE. 207 Data showed that an HSD accelerated lupus progression and increased the mortality in MRL/lpr mice, a mouse model of SLE. 62,149 In MRL/lpr mice fed an HSD, the ratio of Th17/ Treg was significantly increased, 149 and a higher proportion of Tfh cells was observed in the spleen. Given the pathogenic role of Tfh cells in lupus, an HSD might accelerate the SLE progression by inducing Tfh cell differentiation. 62 Additionally, an HSD accelerated the progression of murine lupus by activating DCs through the p38/MAPK-STAT1 pathway. 112 RA is a chronic inflammatory joint disease that can cause cartilage, bone damage and disability. 208 Collagen-induced arthritis (CIA) and K/BxN serum transfer-induced arthritis (STIA) are mouse models of RA. CIA depends on adaptive and innate immunity, while STIA predominantly mimics the innate effector phase. 151 CIA mice fed an HSD showed more severe arthritis, higher Th17 cell proportion in splenocytes and increased IL-17 expression in synovium and intestine. 150 Sehnert et al. 151 did not observe aggravated CIA or STIA in high-salt-diet mice, but they found that a low-salt diet ameliorated the severity of CIA and STIA. Consistently, a study enrolled RA and SLE patients demonstrated that a restricted dietary salt intake can dampen the pro-inflammatory response in patients with autoimmune diseases. 174 Moreover, high salt consumption among smokers was reportedly associated with increased RA risk. 175,176 By contrast, another study confirmed the association between high sodium intake and RA, particularly in nonsmokers. 177 Increased sodium excretion was also observed in patients with RA, 179 and further investigation suggested a correlation between RA disease activity and urinary Na/K ratio. 178

| Cancer
It is known that the Na + concentration is raised in solid tumours and can affect cell metabolism and immune function. Thus, it seems promising that dietary salt can influence the development of tumours and has the potential to be administered in cancer therapy. 209 Compelling evidence suggests that an HSD can suppress the progression of tumours. High salt intake suppressed the tumour growth and lung metastasis in a breast cancer murine model, 210 154 Additionally, an HSD induced NK cell-mediated tumour immunity by augmenting the intratumor localization of Bifidobacterium. 86 However, given that high salt intake is a potent inducer of proinflammatory states, the adverse effect of HSD on cancer has also been reported. 212 In vitro, high salt can synergize with IL-17 to enhance the proliferation, treatment resistance and Warburg-like metabolism of breast cancer cells. [213][214][215] In mice, an HSD accelerated the development and lung metastasis of breast cancer. 152,216 Data suggested that the IL-17F produced by salt-induced Th17 cells activated the MAPK signalling in breast cancer cells. 152 In humans, high salt intake is a risk factor for lung, testicular, bladder, 182 renal cell, 183 pancreatic, 184 oesophageal 181 and gastric cancer. 180 Particularly, an HSD can promote Helicobacter pylori infection, gastric mucosa damage, hypergastrinemia and cell proliferation, therefore contributing to gastric carcinogenesis. 153 These double-sided effects indicated that the high salt effect on tumours might change with different tissues and phases of tumours.

| Infections
An HSD might promote the elimination of Escherichia coli (E. coli) and Leishmania major (L. major) infection by increasing skin Na + concentrations. 41

| Ischemic stroke
Ischemic stroke is a cerebrovascular disease that causes high mortality worldwide. 217 It has been reported that HSD in mice can exacerbate ischemic stroke. 218 An animal study linked HSD to ischemic brain damage. 219 Mechanistically, an HSD in mice enhanced BBB disruption during ischemia via the p38/MAPK/SGK1 pathway, accompanied by downregulated TJ protein expression in endothelial cells. 158 Additionally, an HSD induced pro-inflammatory microglia by increasing aldose reductase (AR) protein expression via p38/MAPK and thus exacerbated ischemic stroke. 159 HSD also decreased the expression of the phagocytic molecule triggering receptor expressed on myeloid cells 2 TREM2 and induced a pro-inflammatory phenotype in macrophages, leading to the postponed recovery of stroke lesions. 160 Moreover, high urinary sodium levels in humans were associated with large ischemic lesions. 158 Data from humans also showed that high salt intake was associated with the risk of ischemic stroke death, but was not associated with the risk of ischemic stroke onset. 185

| Cognitive disorders
Cognitive disorders, common in psychiatric and neurological diseases, are a major societal burden. 220 High salt intake has been reported to impair cognitive functions via the gut-brain axis. HSD can induce gut Th17 responses and increase circulating IL-17, which suppresses cerebral endothelial NO production, leading to cerebral hypoperfusion, neurovascular dysregulation and cognitive impairment in mice. 161 Furthermore, an HSD caused cognitive dysfunction in mice by inducing an inflammatory environment and triggering apoptosis in the brain. 162 Several human studies linked HSD to impaired cognitive function, while a clinical trial showed that a low-salt intake might improve cognition. 186 Nonetheless, most studies employed food questionnaires or diet records, and higher-quality studies are needed.

| Transplantation rejection
In a humanized xenogeneic graft-versus-host disease x-GvHD murine model, an HSD blocked the immunosuppressive function of Treg cells and consequently worsened the severity and onset of transplantation rejection. 132 Another study demonstrated that mice fed an HSD displayed accelerated cardiac allograft rejection, accompanied by decreased Treg cells. 55

| Others
Wound healing was delayed in high-salt-diet mice owing to reduced M2 activation. 56 Additionally, HSD in mice during recovery from acute renal failure accelerated the progress towards chronic kidney disease and interstitial fibrosis. These responses were associated with Th17 cell activation in the kidney. 163,164 Moreover, an HSD might exacerbate food allergy in mice, 136 while sodium may participate in the progression of atopic dermatitis by regulating Th2 responses. 65

| CLINICAL APPLICATION OF HIGH SALT
The clinical applications of salt can date back to the middle ages, when salt was used as a treatment for toothaches, upset stomachs and so forth. 222 Nowadays, high salt in clinical practice is primarily applied in the form of intravenous injection or nebulization. Hypertonic saline resuscitation can promote volume expansion, improve microcirculation and modulate immune responses in certain critically ill patients. 223,224 In addition, hypertonic saline has been adopted as an alternative to mannitol in patients with raised intracranial pressure. 225 Nebulized hypertonic saline is already used in the treatment of cystic fibrosis, 226 non-cystic fibrosis bronchiectasis 227 and viral bronchiolitis. 228 Hypertonic saline inhalation in patients with cystic fibrosis can rehydrate the airway surface liquid, increase mucociliary clearance and improve lung function. 229,230 Most guidelines recommend the use of nebulized hypertonic saline in bronchiectasis therapy to facilitate airway clearance. 227,231 Nonetheless, studies about the application of nebulized hypertonic saline in patients with viral bronchiolitis demonstrated conflicting results. 228 Additionally, hypertonic saline nasal irrigation might be helpful in patients with chronic rhinosinusitis. 232 Furthermore, an increase in salt concentration in cancer vaccines can significantly change the physicochemical properties of the vaccine formulation and enhance its efficacy, showing invaluable potential in cancer therapy. 109,233 Salt-activated CD4 + T cells that were derived from tumour-bearing mice and injected into mice with breast cancer also elicited a strong anticancer response. 234 As the effect of high salt on immune cells and immune-regulated diseases has been studied in recent decades, further investigation on its feasibility and applicability in clinical therapy should be conducted.

| CONCLUSION AND FUTURE DIRECTION
In recent decades, we have witnessed the immunomodulatory effect

ACKNOWLEDGEMENTS
This work was supported by the National Science Foundation for Excellent Young Scholars (32122052) and the National Natural Science Foundation Regional Innovation and Development (no. U19A2003).

CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.