Analyses in mice suggest that dietary salt increases blood pressure partly by affecting some of the microbes that inhabit the gut. The implications of this work for hypertension warrant further study in humans. See Article p.585
High blood pressure is a leading cause of cardiovascular disease and hence preventable death in the United States1, and is an increasingly prevalent and costly global health burden. Blood-pressure control in humans is inadequately understood — debate swirls around the contribution of dietary salt in particular. On page 585, Wilck et al.2 draw connections between dietary salt, microorganisms in the gut, immune responses and blood pressure.
Processed foods and Western diets are packed with salt. Average daily sodium intake in the United States is more than 3.4 grams (equivalent to 8.5 g of table salt)3, despite the fact that guidelines4 recommend an intake of less than 2.3 g (5.8 g of salt). Most studies show that excess sodium consumption raises blood pressure in a dose-dependent manner4. But blood-pressure responses to salt are variable and are generally detected in fewer than half of all subjects5. Known sources of such variability include genetics, dietary intake of other nutrients such as potassium, and kidney disease. In addition, responses to sodium intake are more pronounced if individuals have high blood pressure (hypertension)5.
Previous work6 has suggested that high salt intake increases the number and activity of immune cells called T lymphocytes, especially a pro-inflammatory subset called TH17 cells. Activated TH17 cells produce the immune-cell signalling factor interleukin-17, which not only promotes hypertension and inflammation in artery walls6,7, but also induces autoimmune diseases8.
Wilck et al. sought to determine whether gut microbes (collectively known as the gut microbiota) have a role in mediating the effects of a high-salt diet (HSD). The authors fed mice an HSD for three weeks and analysed the composition of the animals' gut microbiota by sequencing DNA from faecal samples. There were no major changes in composition; however, a machine-learning algorithm identified a DNA sequence that became less abundant in mice during HSD and more abundant when mice were returned to a normal diet. The sequence matched that of the bacterium Lactobacillus murinus.
The researchers recovered L. murinus from mouse faeces, and demonstrated in vitro that its growth was inhibited by salt concentrations equivalent to those found in the colons of mice fed an HSD. This bacterial species is not seen in humans, but the researchers found similar salt-sensitive growth in some human-associated Lactobacillus species. As expected, HSD caused hypertension in mice, but oral administration of L. murinus blunted this effect.
Wilck and colleagues found that L. murinus administration also prevented HSD-induced exacerbation of an autoimmune disease that can be experimentally generated in mice — actively induced experimental autoimmune encephalomyelitis (EAE). Administration of L. murinus was associated with a reduction in the numbers of TH17 cells, which mediate EAE, in the intestinal wall, spleen and spinal cord (Fig. 1).
What links decreases in L. murinus to increased numbers of TH17 cells? Lactobacillus species produce compounds called indoles from the dietary amino acid tryptophan, and treatment with indoles reduces the severity of actively induced EAE in mice9. The authors demonstrated that levels of indoles decreased when mice were fed an HSD, but were restored by administration of L. murinus. They then showed that one indole metabolite downregulated differentiation of T lymphocytes into TH17 cells.
Do these observations translate to humans? Wilck et al. gave 12 healthy people 6 g of supplemental sodium chloride per day for 14 days. The subjects' mean blood pressure increased over time, as did the numbers of TH17 cells in their blood. In four of the five individuals who had Lactobacillus species in their guts before treatment, five species became undetectable after two weeks. Ten of the people gained extra Lactobacillus species during treatment, but overall Lactobacillus abundance declined. The authors drew control data from previous studies.
Wilck and colleagues' animal analysis, like many such studies, has provided key insights and raised compelling questions about the role of the microbiota in regulating host traits. However, there are several caveats when considering whether these findings will translate to humans. First, the authors gave mice 0.3 g of salt daily, which is a 15-fold increase from their normal intake. This high-salt environment would be difficult to create in humans because the equivalent daily dose for a 70-kg human, scaled by surface area, would be roughly 170 g — much more than humans could tolerate. Furthermore, most dietary sodium in humans is typically absorbed in the small intestine10. Thus, if elevated salt intake has direct effects on the gut microbiota, these effects probably occur before nutrients reach the colon, which was studied here.
Second, Wilck and colleagues' human data concern a specific scenario — people with normal blood pressure on an HSD for a relatively short time. This scenario and these findings are not necessarily informative about people who have chronic hypertension, particularly those with other underlying health problems, because of additional confounding factors.
Finally, the authors did not analyse whether changes in the abundance of Lactobacillus species correlated with the magnitude of changes in blood pressure, which could have helped to show a causal relationship between these factors. A larger study that includes people with normal and high blood pressure, compares various salt doses and a placebo, and takes functional readouts of the microbiota over time is eagerly awaited.
The list of complex diseases in which the gut microbiota might play a part is growing, but careful attention to interacting co-factors is needed when testing hypotheses. For instance, when considering a proposed role11 for the microbiota in cardiovascular disease, several variables must be taken into account. Gut microbes produce the molecule trimethylamine from dietary compounds found in red meat, and the liver converts it to trimethylamine oxide (TMAO). In animals, increases in TMAO promote a condition called atherosclerosis11, which increases the risk of heart attack and stroke. However, TMAO production is not necessary for the development of atherosclerosis, and other factors are required in addition to gut microbes for TMAO production. Thus, the gut microbiota contributes to atherosclerosis, but only in individuals who have certain combinations of microbes, diet and other risk factors.
Should hypertension be added to the list of conditions promoted by the gut microbiota? Future studies will no doubt tell. If this connection is real, one might predict — because of other variable factors — that the effects will be modest and restricted to a subset of individuals. Nonetheless, even modest effects are more than worthy of further study, because of the profound potential for global health benefits. Footnote 1
Institute of Medicine. A Population-Based Policy and Systems Change Approach to Prevent and Control Hypertension (Natl Acad. Press, 2010).
Wilck, N. et al. Nature 551, 585–589 (2017).
US Department of Health and Human Services, US Department of Agriculture. What We Eat in America. NHANES 2011-2012. Available at http://www.ars.usda.gov/ARSUserFiles/80400530/pdf/1112/tables_1-40_2011-2012.pdf
Institute of Medicine. Dietary Reference Intakes for Water, Potassium, Sodium Chloride, and Sulfate (Natl Acad. Press, 2005).
Kotchen, T. A., Cowley A. W. Jr & Frohlich, E. D. N. Engl. J. Med. 368, 1229–1237 (2013).
Kleinewietfeld, M. et al. Nature 496, 518–522 (2013).
Madhur, M. S. et al. Hypertension 55, 500–507 (2010).
Burkett, P. R., Meyer zu Horste, G. & Kuchroo, V. K. J. Clin. Invest. 125, 2211–2219 (2015).
Rothhammer, V. et al. Nature Med. 22, 586–597 (2016).
Sandle, G. I. Gut 43, 294–299 (1998).
Tang, W. H. W. & Hazen, S. L. J. Clin. Invest. 124, 4204–4211 (2014).