Protective Effects of Fermentable Dietary Fiber and Propionate in Dahl Salt-Sensitive Hypertension and Renal Damage

Abstract

The Dahl Salt-Sensitive (SS) rat is an established model of hypertensive kidney injury, where the gut microbiota has been shown to causally contribute to disease progression. Gut bacteria-derived metabolites serve as mechanistic links between the microbiota and disease, with dietary fiber providing a critical source of protective metabolites such as short-chain fatty acids (SCFAs). The current study hypothesized that the substitution of non-fermentable fiber cellulose with the fermentable fiber inulin would attenuate hypertension and renal damage in SS rats via increased circulating SCFAs. Male and female SS rats were placed on the 0.4% NaCl (LS) inulin diet for one week prior to the switch to a 4.0% NaCl (HS) inulin diet for 4 weeks. Controls were maintained on diets containing cellulose. Rats consuming inulin had a reduction in mean arterial pressure (MAP) compared to cellulose, though the antihypertensive effect was more robust in females. The inulin diet significantly protected both sexes from albuminuria, medullary protein cast formation, and renal immune cell infiltration, and was associated with specific changes to the fecal microbiota. Assessed by mass spectrometry, inulin consumption resulted in increased circulating propionate and butyrate, and the administration of these SCFAs revealed a protective effect of propionate against salt-sensitive hypertension and kidney damage in males, which coincided with an expansion of renal T regulatory cells. In conclusion, substitution of cellulose for the fermentable fiber inulin lowered blood pressure and significantly attenuated salt-induced renal damage in both sexes, which may be attributed to greater production of the protective, anti-inflammatory SCFA propionate.

New and Noteworthy

The dietary switch to inulin, a fermentable fiber, reduced salt-sensitive hypertension and kidney injury in male and female Dahl SS rats, and caused gut microbiota composition shifts and increases in SCFA production (propionate and butyrate). Direct administration of propionate ameliorated salt-sensitivity in males, which coincided with renal T regulatory cell expansion. These findings provide the mechanistic basis for leveraging the microbiota and its metabolites through dietary interventions as a therapeutic for hypertension and kidney disease.

Graphical Abstract

Introduction

Diet is widely recognized as a top modifiable lifestyle risk factor and the notorious Western Diet is strongly associated with increased risk for cardiovascular diseases and a wide range of other health concerns^1,2. Associated with the Western Diet is a suboptimal consumption of dietary fiber, and the CDC has specifically identified fiber as a critical dietary component that has important influences on health outcomes. Notably, the most recent National Health and Nutrition Examination Survey found that over 98% of both adults and children consume less than 50% of the recommended daily fiber intake levels³. As a modifiable dietary factor, increased fiber intake has been shown to reduce the risk of cardiovascular disease^4,5 and acts protectively against renal nephropathy⁶ and chronic kidney disease^7–9.

Our laboratory, as well as others, have demonstrated that diets rich in fiber, particularly plant-based diets, can reduce hypertension and associated hypertensive end-organ damage^10–14. These beneficial effects have been associated with the short-chain fatty acids (SCFAs) butyrate, acetate, and propionate, and their action on different G protein-coupled receptors^13,15–17. Furthermore, our research and that of others has shown that high-salt diets lead to significant alterations to the gut microbiota, contributing to increased gut dysbiosis and increased hypertension and renal damage^12,18,19.

Also associated with the Western Diet is an overconsumption of salt, and we utilize the Dahl Salt-Sensitive (SS) rat as a well-established translational model of salt-sensitive hypertension, since it mirrors key human pathophysiological features such as elevated blood pressure, increased renal damage, and enhanced renal immune cell infiltration when exposed to a high-salt diet^20–22. Our laboratory maintains the SS/JrHsdMcwi substrain of Dahl SS rats on a purified AIN-76A diet from Dyets Inc., which contains cellulose as its sole fiber source. Cellulose is an insoluble, non-fermentable fiber and is minimally digested by gut bacteria. In contrast, soluble, fermentable fibers can be metabolized by beneficial commensal bacteria to promote their growth and production of protective SCFAs, as noted above. However, to our knowledge, no studies have directly compared the effects of fermentable versus non-fermentable fiber, especially in the context of salt-sensitive hypertension. We therefore hypothesized that substitution of non-fermentable fiber cellulose with the fermentable fiber inulin would attenuate hypertension and renal damage in SS rats by increasing circulating SCFAs.

Materials and Methods

Experimental animals and diet.

Experiments were concurrently performed in both male and female inbred Dahl SS rats (SS/JrHsdMcwi) from our breeding colony that were maintained ad libitum on a low salt (LS) diet with cellulose as the fiber source in the diet (0.4% NaCl with cellulose, AIN-76A #113755GI, Dyets Inc). All animals were housed on aspen wood chip bedding (Sani Chips, PJ Murphy) in a specific pathogen-free facility in the vivarium at Augusta University. The experimental timeline for the dietary fiber study is provided in detail in Figure 1A. In brief, at 8 weeks of age, experimental animals were randomized to either continue on the LS diet containing cellulose or switch to the LS diet containing inulin as the fiber source (0.4% NaCl with inulin, AIN-76A #104773GI, Dyets Inc). At 9 weeks, rats were then switched to a high salt (HS, 4.0% NaCl) diet while continuing on their respective fiber source (4.0% NaCl with cellulose, AIN-76A #113756GI or 4.0% NaCl with inulin, AIN-76A #104988GI, Dyets Inc) for 4 weeks. At the end of the experimental protocol, all animals were euthanized between 7–9am. All protocols and procedures were approved by the Augusta University Institutional Animal Care and Use Committee.

Figure 1. Dietary substitution of cellulose for the fermentable fiber inulin attenuated blood pressure, with a greater effect in females.

(A) Schematic of experimental timeline. 24-hour mean arterial pressure (MAP) was measured continuously via telemetry in male (B) and female (C) SS rats through the LS and HS period. ΔMAP was also determined as the difference in pressure between HS day 28 and HS day 0. n=6–11. *p<0.05, **p<0.01 cellulose versus inulin via Two way RM ANOVA with Holm-Sidak posthoc test or unpaired Student’s t-test.

Blood pressure recording and urinalysis.

7 week old rats were deeply anesthetized via 2% isoflurane for aseptic implantation of radiotelemeters (HD-S10, Data Sciences International, St. Paul, MN) into the right carotid artery for continuous monitoring of blood pressure. Rats were postsurgically administered 0.3mg/kg Buprenorphine-SR, given one week to recover, and after baseline blood pressure recording, rats were switched diets according to the experimental design in Figure 1A. Blood pressure data presented in the figures only include days where full 24-hour recordings were continuously measured.

Rats were placed in metabolic cages during the LS period for baseline measurements and then again on HS days 7, 14, 21, and 28. Urinary total protein concentration was quantified using Weichselbaum’s biuret reagent and urinary albumin excretion was quantified utilizing a fluorescent assay assessing Albumin Blue 580 Dye (Molecular probes) on a fluorescent plate reader (FL600, BioTek). Creatinine clearance, blood urea nitrogen (BUN), and urine electrolytes were measured via autoanalyzer (ACE, Alfa Wasserman).

Assessment of renal histological damage.

Kidneys were flushed and the right kidney was halved longitudinally along the coronal plane, fixed in 10% neutral buffered formalin, paraffin embedded, cut into 5 μm sections, mounted, and stained with Masson’s Trichrome. Kidney slices were scanned at 20x magnification by a ZEISS Axioscan 7 slide scanner at the Augusta University Medical College of Georgia Cell Imaging Core Facility (RRID:SCR_026799). The percentage of renal damage and protein casting in the outer medulla was determined by color inclusion via ImageJ software (version 1.54d, National Institutes of Health). Glomerular damage was also assessed, where ~50 glomeruli per kidney section were blindly scored from a scale of 1 to 4 (least to worst damage), and data are presented as the average % of glomeruli with a score ≥3 (severely damaged glomeruli).

Immune cell profiling via flow cytometry.

At the end of the experimental timelines, rats were deeply anesthetized under isoflurane and the abdominal aorta was catheterized for blood collection and then used to flush the kidneys with 2% heparinized saline. Circulating immune cells were isolated by density gradient centrifugation of blood layered over Histopaque-1083 (Sigma-Aldrich). Renal immune cell isolation was performed as previously described²³; briefly, kidneys were minced and incubated for 30 mins in RPMI-1640 media containing collagenase type IV, L-glutamine, HEPES and DNase. After a series of gravity filters, cells were layered across a Percoll density gradient and centrifuged. Immune cells isolated from the resulting layer were counted using a hemocytometer, and 10⁶ cells were incubated with a cocktail of antibodies for extracellular markers: anti-CD45 for leukocytes (202214, BioLegend), anti-CD11b/c for monocytes and macrophages (50–0110-82, eBioscience), anti-CD45R for B cells (554881, BD Bioscience), anti-CD3 for T cells (46–0030-82, eBioscience), anti-CD4 for T helper cells (201518, BioLegend), and anti-CD8 for cytotoxic cells (201703, BioLegend), with DAPI to assess viability (422801, BioLegend). An additional 10⁶ cells were incubated with the extracellular anti-CD45, anti-CD3+, and anti-CD4+ from above, combined with anti-CD25 (activation marker, 5010344, Fishersci), then stained with Fixable Viability Dye eFluor450 (50–169-62, eBioscience). Cells were permeabilized and fixed using BD Cytofix/Cytoperm Fixation/Permeabilization Kit (554714, BD Biosciences), followed by nuclear staining with anti-FoxP3 (T regulatory cell transcription factor, 17–5773-82, Thermofisher). All cells were analyzed by flow cytometry (5-Laser AURORA Spectral Cytometer, Cytek) in the Flow and Mass Cytometry Core Facility at the Georgia Cancer Center (RRID: SCR_025747). Data were analyzed by FlowJo software (Tree Star), and example gating strategies are provided for the extracellular (Figure 4A) and T regulatory cell staining (Figure 9D).

Figure 4. Dietary substitution of cellulose for the fermentable fiber inulin significantly reduced renal immune cell infiltration in males, with no changes to overall circulating immune cell profile.

(A) Representative flow cytometry gating strategy used to identify CD45+ leukocytes, CD3+ T cells, CD4+ T helper cells, CD8+ cytotoxic T cells, CD45R+ B cells, and CD11b/c+ myeloid cells in the kidney. (B, D) Immune cell populations identified in the blood. (C, E) Immune cell populations identified in the renal tissue. n=7–11. *p<0.05 cellulose versus inulin via unpaired Student’s t-test.

Figure 9. Effect of propionic or butyric acid supplementation on blood and renal immune cell profile.

Flow cytometry was utilized to identify CD45+ leukocytes, CD3+ T cells, CD4+ T helper cells, CD8+ cytotoxic T cells, CD45R+ B cells, and CD11b/c+ myeloid cells in the circulation (A) and in the kidney (B). The proportion of renal T regulatory cells (CD25+FoxP3+) was determined from the CD3+CD4+ T helper cell population (C), with a representative gating strategy from kidney T cells shown in (D). n=4. *p<0.05 versus Vehicle via One way ANOVA with Holm-Sidak posthoc test.

Short-chain fatty acid quantification via mass spectrometry.

Quantification of eight short-chain fatty acids (acetic, propionic, isobutyric, butyric, isovaleric, valeric, isocaproic, hexanoic) was performed by the Mayo Clinic Metabolomics Core via gas chromatograph/mass spectrometer (GC-MS) as previously published^24,25 with a few modifications. In brief, an internal standard solution containing d5-propionic acid, d7-butyric acid, d9-valeric acid, and d11-caproic acid was added to serum samples and dichloromethane was used to extract SCFA from the mixture. The extract was derivatized with N-Methyl-N-tert-butyldimethylsilyltrifluoroacetamide prior to analysis on a GC-MS. Derivatized analytes were separated on a DB5MS column (30m x 0.25mm ID x 0.25um film thickness) prior to entering the MS detector (Agilent MSD5977A). Concentrations of acetic acid (m/z 117.0), propionic acid (m/z 131.1), isobutyric acid (m/z 145.1), butyric acid (m/z 145.1), isovaleric acid (m/z 159.1), valeric acid (m/z 159.1), isocaproic acid (m/z 173.2) and hexanoic acid (m/z 173.2) were measured against 12-point calibration curves that underwent the same derivatization. A representative chromatogram of the standard and a control serum sample is provided in Figure 5A.

Figure 5. Serum short-chain fatty acid (SCFA) profile between cellulose- and inulin-fed SS rats.

Representative GC-MS Total Ion Chromatogram-Selected Ion Monitoring trace of a standard and control serum sample (A). Eight SCFA analytes identified via GC-MS in the serum of male (B) and female (C) SS rats fed either the cellulose or inulin fiber HS diet. n=6–8. *p<0.05 cellulose versus inulin via unpaired Student’s t-test.

Fecal 16S rRNA Gene Sequencing and Analysis of Microbiota Composition.

Fecal samples were collected directly from the anus of the rats and stored at −80°C until shipment to the University of Toledo Microbiome Core for further processing. A total of 8–10 rats were used for each group analysis. As described previously²⁶, QIAamp Power Fecal Pro DNA kit (QIAGEN) was used to extract gDNA from the fecal pellets (~40–50 mg). The gDNA was eluted in low TE buffer (0.1 mM EDTA, Tris-HCl buffer, 10 mM, pH 8.5) instead of the AE buffer provided in the kit. DNA concentration was determined using a NanoDrop and samples were diluted to a final concentration of 5 ng/μl in low TE buffer. We utilized the Illumina User Guide, 16S Metagenomic Sequencing Library Preparation—Preparing 16S Ribosomal RNA Gene Amplicons for the Illumina MiSeq System (Part No. 15044223 Rev. B) for 16S Polymerase Chain Reaction library preparation, clean-up, normalization, and pooling. The 16S rRNA gene targeting the V3-V4 region was amplified by PCR using the Illumina sequencing primers: 5’ TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG and 5’ TCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGGACTACHVGGGTWTCTAAT. For index PCR, the Nextera XT index kit (FC-131–1002) from Illumina was used to attach dual indexes. Each 25μL reaction mixture contained 2.5μL of 10X reaction buffer (Invitrogen), 0.5μL of 10 mM dNTPs, 0.75 (for target PCR)/1μL (for index PCR) of 50 mM MgCl2, 0.1μL of 5U/μL of HotTaq polymerase (Invitrogen), 1μL of each primer (5 μM) and 2.5μL of 5 ng/μL DNA. All samples were reconstituted in water for a final volume of 25μL. Thermocycling was performed in a BioRad T100TM thermal cycler with the following cycling conditions: initial denaturation at 95°C for 5min, followed by 25 cycles of 95°C for 30s, 58°C for 30s, 72°C for 30s, and a final extension at 72°C for 5min for target PCR. Index PCR was carried out in 8 cycles, with an initial denaturation at 95°C for 3min, followed by 95°C for 30s, 55°C for 30s, 72°C for 30s, and a final extension at 72°C for 5min. Each PCR amplicon sample was purified in two rounds using AMPure XP beads (Beckman Coulter Inc.). The concentration of each purified index PCR product was measured using the Qubit dsDNA HS Assay kit with Qubit 3.0 fluorometer (Life Technologies). 4 nmol/L of each amplicon was pooled equally. The pooled library was checked for quality using a 2100 Bioanalyzer (Agilent) before sequencing. Library Denaturing and MiSeq Sample Loading was done according to the Illumina User Guide for the Illumina MiSeq System. The 10 pmol/L denatured and diluted library with 10% PhiX was loaded on an Illumina MiSeq V3 flow cell kit with 2×300 cycles.

For quality filtering, amplicon sequence variant (ASV) picking, and data analysis, chimeric sequences were identified and filtered using Quantitative Insights in Microbial Ecology (QIIME II) software package (version 2021.11)²⁷ with custom scripts relevant for each comparison. The ASVs were subsequently picked using QIIME II, and taxonomy assignment was performed using Silva as the reference database (version 138). Taxonomy was assigned to ASVs using a pre-trained Naïve Bayes classifier. Silva V138, clustered at 99% identity, was used as the taxonomic reference database. As previously reported²⁸, Principal Coordinate Analysis (PCoA) was performed using Microbiome analyst. For male cellulose versus inulin group comparisons, the reads were rarified at 28,000; for female cellulose versus inulin group comparisons, the reads were rarified at 41,000. For bacterial diversity, ANOSIM was used to determine significant differences between two groups of samples. An R value ≈1 indicates dissimilarity between groups, R value of 0.25 to 0.75 indicates moderate separation, and an R value ≈0 indicates no significant dissimilarity. The relative abundance of the microbial population at the genus level for each comparison was generated using QIIME2. Differential enrichment of gut microbiota was visualized using Linear discriminant analysis coupled with Effect Size (LEfSe) based on a threshold of LDA≥2 in Microbiome Analyst²⁸. The 16S sequencing data have been made publicly available under NCBI SRA BioProject PRJNA1381668.

Propionic and butyric acid treatment regiment.

The effects of propionic acid and butyric acid were tested in male Dahl SS rats maintained on the standard LS AIN-76A diet (0.4% NaCl with cellulose, AIN-76A #113755GI, Dyets Inc). At 9 weeks of age, rats were switched to the HS diet for 3 weeks (4.0% NaCl with cellulose, AIN-76A #113756GI) and concurrently assigned to one of the following groups: vehicle control, propionic acid, or butyric acid. Propionic acid (W292400, Sigma Millipore) or butyric acid (W222100, Sigma Millipore) was added to the drinking water at a concentration of 1g/L and neutralized by NaOH to pH 7.

Statistical analysis.

Data are presented as mean±SEM, with the number of animals indicated in the corresponding figure legends, and plotted utilizing GraphPad Prism 10. Statistical significance was defined as P<0.05, with specific levels of significance noted in each figure. Statistical analyses were performed using SigmaPlot version 14.0. The appropriate statistical tests were selected based on the experimental design and specific comparisons; this included the use of an unpaired Student’s t-test, one-way ANOVA, two-way ANOVA, or repeated measures two-way ANOVA, with a Holm-Sidak post hoc test as indicated in the figure legends.

Results

Attenuation of salt-induced blood pressure elevations upon dietary fiber substitution of cellulose for inulin

During the baseline 0.4% NaCl period (LS), there were no significant differences in mean arterial pressure (MAP) between rats fed the two types of dietary fiber, regardless of sex (Figure 1B–C, males: 117.2±2.0 vs 114.9±0.9 mmHg, females: 117.7±0.8 vs 117.1±1.8 mmHg, inulin vs cellulose at HS-1). After 4 weeks of 4.0% NaCl (HS) challenge, inulin-fed male SS rats exhibited a slight reduction in blood pressure at the end of the study (159.5±4.0 vs 170.7±4.8 mmHg, inulin vs cellulose, HS28). The effect of fiber on blood pressure in females was much more pronounced, where inulin-fed female SS rats had significantly lower MAP as compared to females on cellulose throughout the last week of HS (154.9±3.4 vs 170.8±5.9 mmHg, inulin vs cellulose, HS28).

Inulin preserved renal function by ameliorating salt-induced renal injury and inflammation

Despite a modest blood pressure effect in the males, inulin consumption significantly reduced endpoint albuminuria (Figure 2A, 174.4±30.0 vs 370.8±32.4 mg/day, inulin vs cellulose, HS28) and proteinuria (Figure 2B, 319.2±42.1 vs 575.7±45.5 mg/day, HS28). This protection from renal injury was also seen in inulin-fed females (Figure 2F–G, albuminuria: 68.0±8.3 vs 180.3±27.4 mg/day, proteinuria: 176.9±26.7 vs 333.5±39.5 mg/day, inulin vs cellulose, HS28). Interestingly, the inulin diet resulted in increased urinary K+ excretion in both males and females (Figure 2C and 2H, males: 1.7±0.1 vs 1.3±0.1 mEq/day, females: 1.6±0.1 vs 1.2±0.1 mEq/day, inulin vs cellulose, HS28), but there were no differences in urinary Na+ or creatinine excretion in either sex (Figure 2D–E, 2I–J). The protection from HS-induced renal injury was also reflected histologically, where males on the inulin diet displayed significantly less renal medullary protein casts (Figure 3A, 9.3±0.9 vs 14.2±0.9 %, inulin vs cellulose, HS28) and a reduction in the % of severely damaged glomeruli (19.2±2.0 vs 35.9±4.1 %). This protection was accompanied with increased creatinine clearance (Figure 3B, 0.55±0.02 vs 0.44±0.02 ml/min/g KW), reduced BUN (Figure 3C, 15.4±0.5 vs 21.6±2.0 mg/dL), and smaller kidney size (Figure 3D, 0.102±0.004 vs 0.124±0.005 kidney/BW ratio). In females, there was no significant difference in medullary protein casting (Figure 3E, 4.2±1.1 vs 4.6±0.7 %, inulin vs cellulose, HS28) or glomerular damage (12.0±1.1 vs 13.4±3.0 %) between the two fiber diets, though inulin still improved creatinine clearance (Figure 3F, 0.71±0.05 vs 0.53±0.04 ml/min/g KW), BUN (Figure 3G, 17.9±0.6 vs 21.7±1.2 mg/dL), and renal hypertrophy (Figure 3H, 0.097±0.002 vs 0.116±0.004 kidney/BW ratio), compared to females maintained on cellulose. At the end of the study, there were no differences in the circulating immune cell profile between inulin- or cellulose-fed rats of either sex (Figure 4B and 4D). However, there was a significant reduction in infiltrating immune cells in the kidney of male rats on the inulin diet (Figure 4C, 48.5% reduction in CD45+ leukocytes, 50.2% CD11b/c+ myeloid cells, 42.8% CD3+ T cells, 49.3% CD4+ T helper cells, 36.8% CD8+ cytotoxic T cells). No differences were observed in any of the immune cell populations in the kidney of females on the two diets (Figure 4E).

Figure 2. Dietary substitution of cellulose for the fermentable fiber inulin significantly protected against salt-induced kidney injury in both sexes.

Endpoint analysis of urinary excretion of albumin (A, F), protein (B, G), K+ (C, H), Na+ (D, I), and creatinine (E, J) after 4 weeks of HS challenge in males and females. n=7–11. **p<0.01, ***p<0.001 cellulose versus inulin via unpaired Student’s t-test.

Figure 3. Dietary substitution of cellulose for the fermentable fiber inulin significantly improved salt-induced renal histopathology, function, and hypertrophy in both sexes.

(A, E) Representative Masson’s Trichrome-stained kidneys and glomeruli with quantification of % protein casts in the outer medulla and % severely damaged glomeruli, (B, F) creatinine clearance, (C, G) serum blood urea nitrogen, and (D, H) kidney weights normalized to body weight; all measurements were made after 4 weeks of HS challenge in males and females. n=6–11. *p<0.05, **p<0.01 cellulose versus inulin via unpaired Student’s t-test.

Circulating SCFA profile was altered between cellulose- versus inulin-fed SS rats

Of the eight short-chain fatty acids (SCFAs) that were measured in the serum via mass spectrometry, only propionic and butyric acid were significantly elevated in the circulation of male rats consuming inulin versus cellulose (Figure 5B, propionic: 9.70±2.40 vs 3.17±0.32 μM, butyric: 7.43±0.92 vs 4.26±0.92 μM, inulin vs cellulose). No significant differences were observed in acetic, isobutyric, isovaleric, valeric, isocaproic, and hexanoic acid. In the females, only butyric acid was found to be significantly elevated with inulin consumption (Figure 5C, 3.13±0.22 vs 2.11±0.14 μM).

Sex differences in blood pressure, kidney damage and inflammation, and circulating SCFAs

Shown in Figure 6A, no statistically significant sex differences in MAP were observed, regardless of dietary fiber. This result was surprising, given our recent and historic data demonstrating significantly lower blood pressure in female versus male SS rats^29–31. Understanding the basis for this discrepancy is a focus of ongoing investigations in our laboratory. However, what remained consistent were the significant sex differences in albuminuria, medullary protein casting, creatinine clearance, and CD45+ leukocytes infiltrating the kidneys (Figure 6B and 6C), where females were significantly protected from developing the same extent of HS-induced kidney damage and inflammation compared to males, regardless of dietary fiber. There also appeared to be specific sex differences in circulating propionic and butyric acid, whether the rats consumed the cellulose or inulin fiber diet (Figure 6D).

Figure 6. Sex differences in blood pressure and kidney damage in cellulose- versus inulin-fed SS rats.

(A) 24-hour mean arterial pressure (MAP) was measured continuously via telemetry in male and female SS rats through the 0.4% and 4.0% NaCl period. ΔMAP was also determined as the difference in pressure between HS day 28 and HS day 0. **p<0.01 female-cellulose versus inulin, †p<0.05 male-cellulose versus inulin via Two way ANOVA or Two way RM ANOVA with Holm-Sidak posthoc test. Sex differences in extent of renal injury (B), total renal CD45+ leukocytes (C), and serum SCFAs (D) were also assessed. #p<0.05, ##p<0.01, ###p<0.001 between sex; ^p<0.05, ^^p<0.01, ^^^p<0.001 between diet via Two way ANOVA with Holm-Sidak posthoc test.

Differences in 16S fecal microbiota signature between cellulose- versus inulin-fed SS rats

To determine if the effect of inulin on SCFA production could be related to changes in gut bacteria, 16S sequencing of the stool of cellulose- versus inulin-fed rats was performed. As a measure of bacterial community similarity, weighted UniFrac beta diversity was significantly different between SS rats consuming the two different fiber diets, regardless of sex (Figure 7A and 7C, males R=0.30, p=0.008; females R=0.41, p=0.005). Figure 7B and 7D represents the taxa identified via LEfSe analysis, which determines the genomic features that characterize and most reliably differentiate between the two groups. Specifically, taxa like Bacteroides, Bifidobacterium, Allobaculum, Clostridia, Lachnospiraceae, and Adlercreutzia are more abundant in inulin-fed SS males, which are associated with SCFA production³². In cellulose-fed males, Tuzzerella, Peptococcus, and Colidextribacter were more abundant, which have been negatively correlated with SCFAs, with some linked to inflammation. In females, we similarly observed greater abundance in taxa like Allobaculum, Clostridia, and Lachnospiraceae in females consuming inulin, and Tuzzerella and Colidextribacter in those consuming cellulose (Figure 7D).

Figure 7. 16S fecal microbiota profile between cellulose- and inulin-fed SS rats.

In both sexes, weighted Unifrac beta diversity plots (A, C) demonstrated significant shifts to the fecal microbiota in response to inulin consumption. LEfSe analysis (B, D) identified the top 25 differentially abundant taxa between cellulose- and inulin-fed SS rats. n=7–10. ANOSIM non-parametric test for weighted Unifrac analysis; p-value cutoff=0.05, LDA score ≥2.0 for LEfSe analysis.

Differential effects of propionic and butyric acid supplementation on salt-sensitive hypertension and renal damage

Given the specific changes to circulating SCFAs that occurred in response to different dietary fibers, it became necessary to test the potential effect of propionic acid and butyric acid on the salt-sensitive phenotype. During the baseline LS period, there were no differences in MAP between vehicle-, propionic acid-, or butyric acid-treated male SS rats. After 3 weeks of HS, rats treated with propionic acid has significantly attenuated MAP compared to vehicle (Figure 8A, 133.5±1.2 vs 161.3±15.2 mmHg, propionic acid vs vehicle, HS21). However, treatment with butyric acid had no significant effect on blood pressure compared to vehicle (149.1±3.8 vs 161.3±15.2 mmHg, butyric acid vs vehicle, HS21). This pattern of protection via propionic acid, but not butyric acid, was also reflected in the urinary excretion of albumin (Figure 8B, vehicle: 195.2±74.0 mg/day, propionic: 91.9±8.4 mg/day, butyric: 142.8±18.5 mg/day, HS21) and protein (Figure 8C, vehicle: 359.7±82.9 mg/day, propionic: 228.3±5.8 mg/day, butyric: 288.0±23.2 mg/day, HS21). Interestingly, both propionic and butyric acid treatment significantly protected the kidney from histopathological damage and reduced the number of medullary protein casts (Figure 8D, vehicle: 19.5±1.0%, propionic: 12.9±0.1%, butyric: 12.5±1.1%).

Figure 8. Effect of propionic or butyric acid supplementation on salt-induced hypertension and renal damage in male SS rats.

(A) MAP was measured via telemetry throughout 3 weeks of HS challenge in male SS rats administered either vehicle, propionic acid, or butyric acid. ΔMAP was also determined as the difference in pressure between HS day 21 and HS day 0. Renal damage was assessed via urinanalysis of albumin (B) and protein (C), and histopathology of Masson’s Trichrome-stained kidneys and determination of medullary protein casts (D). n=4. *p<0.05, **p<0.01 versus Vehicle, or †p<0.05, ††p<0.01, †††p<0.001 versus LS, via Two way RM ANOVA with Holm-Sidak posthoc test.

Propionic and butyric acid specifically increased the proportion of T regulatory cells in the kidney

Though there appeared to be a trend for reduced immune cells in both the blood and kidney of propionic or butyric acid-treated rats compared to vehicle, there were no significant differences in the immune cell profile in either the circulation or kidney across groups (Figure 9A–B). However, closer examination of CD3+CD4+ T helper cells revealed that both propionic and butyric acid treatment increased the proportion of CD25+FoxP3+ activated T regulatory cells in the kidney (Figure 9C, vehicle: 23.4±0.9%, propionic: 30.6±1.9%, butyric: 28.5±1.3%). These differences were limited to the kidney and were not observed in the circulation.

Discussion

Comprehensive studies from our laboratory have elucidated the contribution of various dietary factors, specifically non-sodium components of the diet, in modulating the response to a high salt intake^{10–12,33,34}, and we have shown these effects to depend upon alterations to the gut microbiota and to T cell activation and metabolism. In the current study, we defined the contribution of different dietary fiber sources and their effect on the salt-sensitive phenotype. Overall, the switch to the fermentable fiber inulin reduced salt-induced hypertension and renal injury, which coincided with an increase in serum propionate and butyrate. Administration of propionate itself had anti-hypertensive and renoprotective effects in the SS rat and was accompanied by an expansion of renal T regulatory cells. By elucidating the mechanisms by which fiber and SCFAs can improve blood pressure and kidney damage, this study supports the leveraging of dietary fiber as a potential therapeutic in salt-sensitive hypertension.

Dietary fiber is considered to be an essential component of a balanced, heart healthy diet, and is associated with a wide range of CVD benefits like improved blood sugar and lipid profiles, relaxation of blood vessels and lowering of blood pressure, reduced inflammation, and prevention of premature mortality and noncommunicable disease incidence^35–37. However, there is a significant lack of fiber consumption at the global level (15–26g/day versus the recommended 20–35g/day³⁸), and insufficient fiber intake has been identified as one of the top leading dietary risk factors contributing to CVD mortality³⁹. Our study not only advocates for increasing fiber consumption for the benefit of health, but also for understanding how various fiber types may differentially impact hypertension and subsequent end-organ injury.

Fibers are non-digestible by human enzymes and are characterized by their solubility and microbial fermentation. Cellulose is a nonstarch polysaccharide and is the sole fiber source in the purified AIN-76A diet fed to the SS/JrHsdMcwi substrain of Dahl SS rats (Dyets Inc, #100000 AIN-76A Diet, 113755GI). Cellulose is an insoluble, non-fermentable fiber that is resistant to enzymatic digestion by the gut bacteria; it is therefore considered to be a bulking agent and not the most nutritious fiber source. Conversely, inulin is a soluble, fermentable fructo-oligosaccharide that is considered to be a prebiotic since it can be metabolized and utilized as a fuel source by the gut microbiota to produce bacterial metabolites like SCFAs. Though this study specifically looked at the effects of inulin, we would expect other non-digestible carbohydrates, such as resistant starches, to increase SCFA production⁴⁰ and have similar protective effects. While we observed sex differences in the degree of protection, the substitution of cellulose for the fermentable fiber inulin resulted in significant protection from salt-induced elevations in blood pressure, reductions in albuminuria and renal histopathological damage, with fewer immune cells infiltrating into the kidney, which was associated with significant shifts in gut microbiota composition and bacteria-derived metabolite production. Most interesting was the effect of inulin in males, where despite a modest effect on blood pressure, the drastic renoprotection from inulin consumption remained. This suggests the potential for direct effects of inulin and/or inulin-derived metabolites on the kidney, independent of blood pressure. These preclinical results align with multiple studies in humans that have explored the association between inulin-type fiber consumption with the gut microbiota and CVD, hypertension, and kidney disease outcomes. In a single group-design trial, healthy participants who consumed about 15g/day of inulin-type fructans saw positive changes in gut microbiota composition and function, improved preference for inulin-rich vegetables, and importantly tolerated the dietary fiber well, in terms of potential gastrointestinal symptoms⁴¹. These results demonstrate promise for the use of inulin as a therapeutic. In a prospective cohort study, higher levels of estimated dietary inulin intake were found to significantly associate with reduced risk of hypertension, though no relationship was found between inulin consumption and CVD or CKD incidence⁴². However, in patients with CKD, whether dialyzed⁴³ or nondialyzed⁴⁴, inulin consumption improved primary renal outcomes. Specifically, dialyzed ESRD patients receiving 10g/day of inulin-type prebiotics demonstrated a reduction in serum uric acid levels and enhanced intestinal uric acid degradation, though no improvements were observed in terms of ESRD symptoms⁴³. The extent of disease may contribute to the potential effectiveness of inulin as a therapeutic, since nondialyzed CKD patients administered a synbiotic (prebiotic inulin in conjunction with probiotics) resulted in significantly improved eGFR, lower C-reactive protein, and lower gut-derived uremic toxins indoxyl sulfate and TMAO, in the treatment group⁴⁴. Further supporting this balance between disease severity and efficacy of inulin treatment, administration of inulin in conjunction with other beneficial agents, like inorganic nitrates (critical for the nitrate–nitrite–NO pathway), to normotensive young adults had no effect on blood pressure, despite increases in plasma SCFAs⁴⁵.

In our study, the switch to dietary inulin was linked to increases in the circulating levels of the SCFAs propionate and butyrate, but only the direct administration of propionic acid demonstrated an anti-hypertensive and renoprotective effect, which was associated with an expansion of T regulatory cells. The current interventional study utilizing propionic and butyric acid only included male rats, since the inulin-induced increase in propionate and butyrate was observed specifically in males compared to females. This is a limitation, and future studies will be required to determine whether these findings are also applicable in females. These protective actions of propionate align with the current literature surrounding its use in hypertension, CVD, and CKD. Some of the earliest studies demonstrated that application of SCFAs, either in combination or alone, promoted relaxation of isolated human colonic resistance arteries⁴⁶. These vasodilatory properties likely contribute to the acute hypotensive response observed in mice administered propionate, which have been shown to be mediated by Gpr41^16,47. This antihypertensive effect has also been observed in humans, where propionate supplementation in a prospective observational study resulted in a mild blood pressure reduction and an increase in circulating T regulatory cells⁴⁸. This expansion of anti-inflammatory T regulatory cells by propionate has also been observed in ESRD patients, and likely contributes to the observed reduction of inflammatory marker CRP and the overall benefit to systemic inflammation⁴⁹. Beyond its impact on blood pressure, propionate has also been shown to have important cardioprotective effects, improving hypertensive cardiac hypertrophy, arrhythmias, fibrosis, vascular dysfunction, and systemic inflammation in a T regulatory cell-dependent manner⁵⁰.

Interestingly, butyric acid treatment in the current study had no obvious effect on salt-induced blood pressure, albuminuria, or renal immune cell infiltration, though this could in part be due to a slight reduction in the daily molarity dosing of butyric acid in the drinking water. However, butyric acid did cause a significant reduction in renal histopathological damage and medullary protein casting that was also associated with increased renal T regulatory cells (as similarly seen with propionic acid). These data indicate a specific influence of these particular SCFAs on T regulatory cells and a potential direct effect on renal morphological damage. These somewhat contradicting results mirror the conflicting evidence in the literature regarding butyrate. Butyrate administration has been shown to be protective in animal models of kidney disease⁵¹, and its levels significantly correlate with renal function (eGFR, creatinine, BUN) in patients with CKD⁵². While rodent models have also clearly demonstrated that butyrate reduces blood pressure in angiotensin II-induced hypertension^17,53, current human clinical evidence is less consistent⁵⁴. In the SPIRIT trial, a 10% increase in either serum or fecal butyrate concentration was associated with a significant decrease in SBP⁵⁵. Since then, multiple intervention studies have demonstrated lowering of SBP in hypertensive humans receiving butyrate supplementation^14,56. However, a recent double-blind randomized placebo-controlled trial found that oral butyrate instead caused a ~10mmHg increase in daytime SBP⁵⁷. These inconsistencies may possibly be attributed to differences in mode of butyrate delivery as well as the small sample size within each of these cohorts. Future investigations will need to elucidate these diverging effects of butyrate from these studies as well as from our own.

In conclusion, our study adds to the growing foundation of work connecting dietary changes to microbiota function, bacteria-derived metabolites, and supports the leveraging of the microbiota as a therapeutic for hypertension⁵⁸. The dietary substitution of a non-fermentable fiber for the prebiotic inulin had considerable protective effects against salt-induced blood pressure elevations, renal injury, and inflammation in both males and females. These findings notably highlight the importance of not only increasing fiber intake but recognizing the distinct physiological consequences of different types of fiber. Therefore, understanding the underlying mechanisms for how different dietary fibers exert these protective effects may lead to the discovery of novel, more specific therapeutics for blood pressure control and the prevention of end-organ damage.