This study revealed the association between salt intake assessed by 24-h urine and intestinal microbiota and metabolites in children and adolescents. Our findings suggested that salt intake was associated with microbial diversity (alpha-diversity and beta-diversity) and composition, especially sodium-induced conditional pathogenic bacteria increased that have been linked to metabolic syndrome risk factors, such as Prevotella [22] and Lachnospira [23]. In metabolomic profiles analysis, we found that although the two groups had no significant differences in the metabolomic composition, the high salt diet had overall adverse effects on several significant biomarkers. For instance, high salt diet led to up-regulation of fecal SCFAs (Isobutyric acid) and down-regulation of the urinary amino acids (Proline), and up-regulation of the urinary fatty acids (Pentadecanoic acid). The above findings suggested that alterations in the gut microbiota and host metabolites may play an important role in diseases associated with high-salt diets in adolescents.
Many rodent and patient studies have shown that a high-salt diet could alter intestinal microbiota composition and is strongly associated with many diseases. For example, the overall microbial diversity (Shannon index) was negatively correlated with salinity, and a high-sodium diet decreased Lactobacillus [7], Bacteroidetes, and increased Firmicutes/Bacteroidetes ratio Lachnospiraceae, Ruminococcus [6], and SCFA concentrations [24]. Recently, high salt intake was also reported to increase the abundance of genus Prevotella bacteria [25]. Our study showed similar results, with a higher abundance of Lachnospira, Prevotella, Prevotella_9, and Ruminoccus_bicirculans in the high-salt diet group. A growing number of studies are now focusing on the relationship between metabolic diseases associated with high-salt diets and intestinal microbiota and metabolites: studies from Chinese adults had shown that sodium depletion was associated with microbiota and metabolites associated with inflammation and cardiovascular disease (CVD) risk factors [10]. Recently, animal studies from mice found a high salt diet induced a decrease in Bacteroides_fragilis and arachidonic acid levels, which led to increased levels of gut-derived corticosterone and increased levels of blood pressure. In addition, they found that the abundance of Lactobacillus spp. was significantly lower in the HSD group of mice [26]. Similarly, we found that Bacteroides_fragilis decreased abundance in the HSD group and negatively correlated with urinary sodium. Moreover, accumulating evidence suggested that the concentration of salt in the feces was associated with the consumption of Akkermansia_muciniphila [27], which had a consistent anti-obesity effect and a beneficial effect on the regulation of body metabolism [28–30]. Our study also found that 24-h urinary sodium excretion correlated with decreased abundance of Akkermansia_muciniphila; since high salt intake leads to obesity [31–33], we hypothesized that a high salt diet reduced the abundance of Akkermansia_muciniphila, leading to a reduction in the ability of the intestinal microbiota to resist the risk of obesity, and is also responsible for the development of obesity or other metabolic diseases syndrome and other diseases [34, 35]. However, although several studies have shown that a high-salt diet reduces the abundance of Lactobacillus spp. [7], we did not find significant changes in Lactobacillus. There could be many reasons for the inconsistencies. First, the above studies have been carried out in adults or mice other than in children or adolescents (individual differences lead to differences in the intestinal microbiota). Second, some population studies used a Food Frequency Questionnaires (FFQ), or 24-h diet recalls, to assess habitual dietary intake, which we believe is potentially subject to measurement error compared with the 24-h urine results. Third, the software and supporting databases used were different, such as Usearch, QIIME2, SILVA, and GreenGenes reference database.
Previous mice and population studies had reported that sodium intake or high salt intake could affect multiple metabolic pathways [34]. Mice study demonstrated that a high-salt diet reprioritized liver energy metabolism [31] and the untargeted metabolomics analysis showed lower circulating levels of the ketone body, beta-hydroxybutyrate (βOHB) in high salt-fed hypertensive rats [36]. The population-based DASH (Dietary Approaches to Stop Hypertension Trial)-sodium trial showed the sodium-restricted diet decreased the level of isovalerate in plasma or the level of isovalerate in the urine [37, 38]. Similarly, plasma metabolomic analysis from Chinese adults found that dietary sodium was associated with the overall metabolome and several metabolites involved in inflammation and etiology of CVD, including three gut-derived phenolics (1,2,3-benzenetriol sulfate, 3-methoxycatechol sulfate, and 4-methylcatechol sulfate) and 2 SCFAs (butyrate/isobutyrate and isovalerate) [10]. The results of the above studies are not identical due to the source of the samples, the assay method (targeted or untargeted), the sequencing instrumentation, or the statistical methods; however, all of these studies suggest that the high-salt diet could lead to changes in the concentration of plasma metabolites, especially the SCFAs, which in turn affected various physiological mechanisms through different metabolic pathways or even caused the development of diseases such as hypertension. Given that the fecal metabolome can thus be used as an intermediate phenotype that promotes microbial impact on the host and vice-versa [39]. We performed metabolomic assays on 15 fecal samples and found similar results. For example, we found that high-salt diet group adolescents have a higher level of Isobutyric acid in SCFAs. This was consistent with the results of rat model experiments [24].
Recent findings have suggested Proline biosynthesis and catabolism were essential processes in disease [40]. Proline supplementation has been found to confer resistance against oxidative damage and improve visual function [41], as well as reduce oxidative stress and apoptosis [42]. In the urine metabolomics analysis, we found a down-regulation of Proline levels in the high-salt diet group, suggesting that dietary salt content may regulate the metabolic process of Proline in humans, thus bringing about diseases associated with reduced Proline metabolism due to a high-salt diet. Besides that, a biomarker of CVD risk, pentadecanoic acid [43], was also found to be a differential metabolite across salt intake levels in this study. And interestingly, these metabolites showed strong associations with the intestinal microbiota; for example, the genus Prevotella_9 which was enriched in the HSD group in the LEfSe analysis was positively associated with the concentration of Isobutyric acid. In addition, a negative correlation with sodium excretion Romboustia and Akkermansia were also associated with many urinary metabolites. We, therefore, infer that this may be the result of the interaction between the intestinal microbiota and host metabolism. However, given the complexity of the mechanisms of action of intestinal microbiota and host metabolic pathway, we also need to validate the results of this study with targeted metabolomics in the future.
The major advantage of our research is that we first explored the association of a high salt diet assessed by three-day’s 24-h urine on intestinal microbiota and metabolites in children and adolescents. Previous studies on the effects of high salt intake on intestinal microbiota and metabolites have been conducted in adults or rodents, probably because 24-h urine and feces are not easily collected in children. In addition, our study was conducted in boarding schools where participants' meals were sourced from the school cafeteria, thus better demonstrating the relationship between a high-salt diet on intestinal microbiota and metabolites and filling a gap in the field of children. Our study also had several limitations. Firstly, this study was conducted in three boarding schools in Hunan Province and the proportions of overweight/obesity or hypertension were higher in our population compared to the general (overweight/obesity: 26.49% [44]; hypertension: ~4.0% [13]). Therefore, we should be cautious to generalize our results to other populations. Secondly, only 15 normal-weight participants’ fecal and urinary metabolomics were analyzed and the results needed to validate in larges samples. Last but not the least, since our study was a cross-sectional survey, we cannot infer whether salt intake levels affect the intestinal microbiota, and fecal and urine metabolites, but can only explain a certain correlation between them.
In conclusion, we explored the relationship between salt intake and intestinal microbiota and host metabolites in children and adolescents based on 24-h urine over three consecutive days. We found that a high salt diet is strongly asscociated with the intestinal microbiota and host metabolites. These results suggest that intestinal microbiota and related metabolites might play an important role in salt-related diseases. In the future, we will conduct more salt restriction intervention studies to validate and explore the biological pathways involved in this study.