Effect of the DASH diet on the sodium-chloride cotransporter and aquaporin-2 in urinary extracellular vesicles

Abstract

The dietary approach to stop hypertension (DASH) diet combines the antihypertensive effect of a low sodium and high potassium diet. In particular, the potassium component of the diet acts as a switch in the distal convoluted tubule to reduce sodium reabsorption, similar to a diuretic but without the side effects. Previous trials to understand the mechanism of the DASH diet were based on animal models and did not characterize changes in human ion channel protein abundance. More recently, protein cargo of urinary extracellular vesicles (uEVs) has been shown to mirror tissue content and physiological changes within the kidney. We designed an inpatient open label nutritional study transitioning hypertensive volunteers from an American style diet to DASH diet to examine physiological changes in adults with stage 1 hypertension otherwise untreated (Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, Obarzanek E, Conlin PR, Miller ER 3rd, Simons-Morton DG, Karanja N, Lin PH; DASH-Sodium Collaborative Research Group. N Engl J Med 344: 3-10, 2001). Urine samples from this study were used for proteomic characterization of a large range of pure uEVs (small to large) to reveal kidney epithelium changes in response to the DASH diet. These samples were collected from nine volunteers at three time points, and mass spectrometry identified 1,800 proteins from all 27 samples. We demonstrated an increase in total SLC12A3 [sodium-chloride cotransporter (NCC)] abundance and a decrease in aquaporin-2 (AQP2) in uEVs with this mass spectrometry analysis, immunoblotting revealed a significant increase in the proportion of activated (phosphorylated) NCC to total NCC and a decrease in AQP2 from day 5 to day 11. This data demonstrates that the human kidney's response to nutritional interventions may be captured noninvasively by uEV protein abundance changes. Future studies need to confirm these findings in a larger cohort and focus on which factor drove the changes in NCC and AQP2, to which degree NCC and AQP2 contributed to the antihypertensive effect and address if some uEVs function also as a waste pathway for functionally inactive proteins rather than mirroring protein changes.NEW & NOTEWORTHY Numerous studies link DASH diet to lower blood pressure, but its mechanism is unclear. Urinary extracellular vesicles (uEVs) offer noninvasive insights, potentially replacing tissue sampling. Transitioning to DASH diet alters kidney transporters in our stage 1 hypertension cohort: AQP2 decreases, NCC increases in uEVs. This aligns with increased urine volume, reduced sodium reabsorption, and blood pressure decline. Our data highlight uEV protein changes as diet markers, suggesting some uEVs may function as waste pathways. We analyzed larger EVs alongside small EVs, and NCC in immunoblots across its molecular weight range.

Keywords: AQP2; DASH diet; NCC; extracellular vesicles; proteomic analysis.

Graphical abstract

Figure 1.

Validation of clean and highly enriched urinary extracellular vesicle (uEV) preparations after low and high centrifugation to separate uEV from raw urine and uromodulin. A and B: cryo-transmission electron microscopy of EVs without uromodulin cosedimentation from low centrifugation pellet treated with low ionic strength buffer and the first two fractions of a size exclusion chromatography (SEC) of a high centrifugation pellet. C and E: Coomassie blue stained gels showing a bulk of protein at a height of 100 (y axis) corresponding to uromodulin. C: relates to low centrifugation pellet, untreated (1), the supernatant treated with low ionic strength buffer (2), and the pellet after treatment showing removal of uromodulin (3). E: corresponds to the high centrifugation pellet fractionation by SEC (x axis) showing the first two fractions of SEC containing very clean uEV prep which were used for the mass spectrometry analysis. D and F: Western blot analysis of the membrane corresponding to the gel, reacted with antibody directed against TSG101, an EV protein marker.

Figure 2.

Heatmap of proteins with significant expression change over time. A and B: proteins with reduced expression (A) and proteins with increased expression (B). The x-axis shows all patients in the same order across the different days. The y-axis shows the different clusters and names of the proteins. The color scale relates to z-score where orange shades show increased expression and blue shades show decreased expression. C and D: mean expression of NCC (C) and mean expression of aquaporin-2 (AQP2, D), according to iBAQ, across trial days.

Figure 3.

Correlation between extracellular vesicle (EV) protein expression and systolic blood pressure. The x axis shows trial days (days 1, 5, and 11) when measurements and samples were collected. The y-axis shows on the left systolic blood pressure (red, mmHg) and on the right protein z-scores [blue:NCC (A) and aquaporin-2 (AQP2, B)]. Each measurement is a mean of all participants on that day.

Figure 4.

Quantification of Western immunoblots shown in Supplemental Fig. S1 of urinary extracellular vesicle (uEV) samples isolated from samples collected at three time points for each subject (A–H). Blots were sequentially probed for phosphorylated T55-NCC (pNCC), total NCC, aquaporin-2 (AQP2), and syntenin-1 proteins, respectively. The two different sized bands detected for pNCC and total NCC were quantified together. Fluorescent signals were quantified using an Odyssey CLx imager and Image Studio software. The mean value for each target was graphed with standard error of the mean depicted with error bars. Statistically significant changes are indicated with top bars with an asterisk (*). A–H: phosphorylated T55-NCC (pNCC, A), total NCC (B), pNCC normalized to total NCC (C), the pNCC/NCC ratio normalized to creatinine (D), pNCC/NCC ratio normalized to syntenin-1 (E), aquaporin-2 (AQP2) normalized to syntenin-1 (F), AQP2 normalized to creatinine (G), and AQP2 (H). The table shows the statistical analysis of immunoblot quantities using a Wilcoxon signed-rank test for comparing 2 days and nonparametric Friedman ANOVA for comparing all three time points. Results of tests that are below or approaching the significance threshold (P < 0.05) are in bold.