Epigenetic Histone β-Hydroxybutyrylation Contributes to Renoprotection by β-Hydroxybutyrate in the Dahl Rat

Abstract

BACKGROUND:

Previously, we demonstrated that the ketone body, β-hydroxybutyrate, is a potent antihypertensive and reno-protective metabolite in Dahl Salt-Sensitive rats. However, the mechanism by which β-hydroxybutyrate confers these beneficial effects is understudied. Here we focused on determining whether the reno-protective effect of β-hydroxybutyrate is due to its known ability to epigenetically remodel chromatin via histone β-hydroxybutyrylation.

METHODS:

We used the same animal protocol previously used for the discovery of the reno-protective effect of β-hydroxybutyrate. Briefly, postweaning, male and female Dahl Salt-Sensitive rats were split into 2 groups and supplemented with or without 1,3-butanediol for 6 weeks. At euthanasia, circulating β-hydroxybutyrate was quantitated. Renal homogenates were examined for histone 3 lysine 9 β-hydroxybutyrylation, chromatin occupancy, transcriptomic and proteomic profiles with validations

RESULTS:

Rats supplemented with 1,3-butanediol had higher circulating β-hydroxybutyrate, renal histone β-hydroxybutyrylation, and significant remodeling of chromatin. Notably, regions of the genome associated with lipid catabolism were predominantly in an open chromatin configuration, leading to active transcription and translation. The most highly upregulated gene actively transcribed and translated was Hmgcs2 (3-hydroxy-3-methylglutaryl CoA synthase 2), a gene responsible for the biosynthesis of β-hydroxybutyrate in mitochondria. In contrast, regions with more compact chromatin structures contained immune function genes, Ptprc (protein tyrosine phosphatase receptor type C) and Lcp1 (lymphocyte cytosolic protein 1), which were suppressed.

CONCLUSIONS:

These results reveal that renal epigenetic histone β-hydroxybutyrylation is a novel mechanism by which transcriptional regulation of both energy metabolism and immune function occur concomitantly and contribute to renoprotection in the hypertensive Dahl rat.

NOVELTY AND RELEVANCE.

What Is New?

Discovery of the epigenetic modification, H3K9 β-hydroxybutyrylation, as a novel mechanism underlying the beneficial effect of the ketone body β-hydroxybutyrate in curtailing kidney damage and lowering hypertension.

What Is Relevant?

Our study expands current limited mechanistic knowledge on how a relatively new histone modification by the ketone body, β-hydroxybutyrate, can specifically upregulate lipid energy metabolism and downregulate harmful overactive immune cell function to improve kidney health.

Clinical/Pathophysiological Implications?

Starvation, intermittent fasting, keto diet, and exercise, all are lifestyle modifications recommended for patients with hypertension. Precisely how these operate to lower blood pressure is largely unknown. In this context, our finding that a ketone body, β-hydroxybutyrate, which is generated in all these lifestyle modifications can protect kidneys of a hypertensive rat model via histone β-hydroxybutyrylation to protect kidneys, is a notable mechanistic advancement.

β-hydroxybutyrate is a ketone body produced in the liver during the breakdown of fatty acids during low-carbohydrate intake, fasting, or exercise.¹ It is a crucial alternate energy substrate which makes β-hydroxybutyrate essential during states of ketosis, where the body shifts from using carbohydrates to using fats for energy.^1,2 β-hydroxybutyrate is not only a fuel; it also functions as a signaling molecule that can influence gene expression.² It modulates pathways related to energy metabolism, inflammation, and oxidative stress.^3–5 β-hydroxybutyrate is shown to inhibit histone deacetylases, which affect gene transcription, potentially promoting longevity and reducing inflammation.^4,6,7 Particularly related to inflammation, β-hydroxybutyrate is reported to downregulate the expression of the NLRP3 inflammasome.^7–9 Due to all of these known beneficial effects, there is a growing interest in leveraging β-hydroxybutyrate to help manage chronic metabolic diseases, such as type 2 diabetes, obesity, and cardiovascular diseases.^3,9–13

Metabolism plays a key role in hypertension.¹⁴ Previously, we and others reported metabolic perturbations in salt-sensitive hypertension.^14–16 Previously, we examined the effect of reconstituting β-hydroxybutyrate in the Dahl S rat and discovered that it has a profound antihypertensive effect, which was associated with a betterment of kidney damage and renal function.⁹ However, the molecular mechanism by which β-hydroxybutyrate protects kidneys remains largely unknown.

Here we focused on the kidney and specifically examined whether the mechanism underlying the beneficial effect of β-hydroxybutyrate was due to β-hydroxybutyrylation, which is a newer epigenetic histone modification caused by β-hydroxybutyrate.¹⁷ Data in support of histone β-hydroxybutyrylation-mediated epi-transcriptional regulation of energy metabolism and immune function as a novel reno-protective mechanism are presented.

Methods

Please see the Major Resources Table in the Supplemental Material (Table S1).

Data Availability

All data and materials have been made publicly available in the GEO database (GSE298451 and GSE298452).

Experimental Model and Study Details

Male and female Dahl Salt-Sensitive rats from the original colony maintained at the University of Toledo College of Medicine and Life Sciences were used for this study. Rats weaned at 28 days were placed on a high-salt diet (2% NaCl) with or without 1,3-butanediol (20% v/v in drinking water). After 6 weeks of supplementation, rats were sacrificed and tissues were harvested.

BP Measurements

Rats were surgically implanted with radiotelemetry transmitters as described previously.¹⁸ Postsurgery, rats were housed individually and allowed to recover before recording their blood pressure (BP) using the DSI software and equipment (https://www.datasci.com/). Systolic BP was collected at 5-minute intervals and analyzed using the Dataquest A.R.T 4.2 software.

Monitoring Energy Metabolism and Activity

Metabolic parameters were measured in 12- to 13-week-old male rats after 6 weeks of 1,3-butanediol supplementation. Rats were housed individually for 24 hours in the Comprehensive Laboratory Animal Monitoring System (Columbus Instruments). Volumes of oxygen consumption and carbon dioxide production were sampled sequentially for 5 seconds every 10 minutes and motor activity was recorded every second in the X and Z dimensions. Respiratory exchange ratio was calculated as the volume of carbon dioxide production/volume of oxygen consumption.

Food Intake

Food given to rats housed in the Comprehensive Laboratory Animal Monitoring System was measured before and after 24 hours. The difference was recorded as food intake.

Measurement of Serum β-hydroxybutyrate

Serum β-hydroxybutyrate levels were measured using a colorimetric assay kit from Cayman Chemicals as previously described.^9,10

Histone Extraction

Snap-frozen kidney samples were homogenized for histone extraction using the EpiQuik histone extraction kit and quantitated using the Bicinchoninic Acid method.¹⁹ Proteins were aliquoted and stored at –80 °C until further use.

Western Blotting

Histones were resolved by gel electrophoresis on a 4% to 20% gradient SDS-PAGE gel and probed using primary antibodies against histone H3 lysine 9 (H3K9) β-hydroxybutyrylation, H3, ribosomal protein S6, and phospho-S6 ribosomal protein. from PTM Biolabs. Please refer to the major resources table in Table S1 for details.

Molecular Analyses

Detailed methods for molecular analyses are provided under Supplemental Methods. Briefly, kidneys were examined for chromatin accessibility via assay for transposase-accessible chromatin with sequencing (ATAC-seq)^20–31 and chromatin immunoprecipitation (ChIP) assays,^32–36 gene expression via RNA-seq,^37,38 and real-time polymerase chain reaction,³⁹ and protein abundance by mass spectrometry-based quantitative proteomic analysis.⁴⁰ Primer sequences are listed in Tables S2 and S3.

Pathway Analyses

KEGG-based pathway enrichment analyses were conducted using ShinyGO, v0.77 (false discovery rate cutoff <0.05).⁴¹ The reactome Pathway Knowledgebase was used to generate pathways.⁴²

Transmission Electron Microscopy and Analysis of Mitochondrial Morphology

Transmission electron microscopy was used to examine the impact of β-hydroxybutyrylation on mitochondrial morphology in the proximal tubule epithelial cells, as detailed under Supplemental Methods.

Quantification of Peripheral T Cells Proliferation

Peripheral lymphocytes were isolated from blood using the Histopaque method.⁴³ Briefly, lymphocytes (1×10⁶ cells/well) were stained with carboxyfluorescein diacetate succinimidyl ester (carboxyfluorescein diacetate succinimidyl ester [CFSE], Sigma, 5 µmol/L), washed 3× in cold IX PBS, and cultured in 1% heat-inactivated fetal bovine serum in Dulbecco's Modified Eagle Medium (DMEM) for 5 days. This was followed by staining with anti-rat CD3-APC antibody for getting CD3⁺ T cell populations. The proliferation of CFSE-stained cells was analyzed by quantifying the degree of CFSE dilution in proliferated cells via flow cytometry (BD Accuri C6 Plus, BD Biosciences). BD Accuri software was used to analyze the proliferation. Data presented in the histogram indicate CFSE-labeled CD3⁺ T cells.

Histological Analyses

Kidneys were fixed in 10% neutral buffered formalin for 24 hours and transferred to 70% ethanol. Then, kidney tissues were embedded in paraffin and 5-µm-thick sections were prepared. Rabbit anti-PMP70 antibody (Thermo Fisher Scientific, Cat# PA1-650, 1:100 dilution) was used and detected using the Vectastain Elite ABC kit and SigmaFast 3,3-Diaminobenzidine tablets. The images were captured by the VS120 Virtual Slide Microscope (Olympus) and analyzed using Olyvia Ver.2.9.

Quantification and Statistical Analyses

GraphPad Prism, version 9.3.1 was used for data analysis. Statistical analyses were performed using the students’ t test. For comparisons between 2 groups across multiple time intervals, data were analyzed by 2-way repeated-measures ANOVA. A P<0.05 was considered significant. All data are presented as mean±SEM. Statistical analyses for multiomics data are presented in the Supplemental Methods.

Results

1,3-Butanediol is a Precursor of β-Hydroxybutyrate and Lowers BP in Both Sexes

Oral administration of 1,3-butanediol to male rats increased their circulating β-hydroxybutyrate and lowered BP.⁹ Here, we first examined whether the ability of 1,3-butanediol to reconstitute serum β-hydroxybutyrate and lower BP was sex-independent. As seen in Figure 1A through 1D, both male and female rats responded to 1,3-butanediol treatment with a significant elevation of circulating β-hydroxybutyrate and reduced systolic BP. However, the magnitude of the 1,3-butanediol to β-hydroxybutyrate conversion, as well as the BP-lowering effect, was higher in males than in females. This indicates that the conversion of 1,3-butanediol to β-hydroxybutyrate and the BP-lowering effect occurred independently of sex, but to a much greater extent in males. Unfortunately, while we did not collect renal parameters from the current study, our previous study has demonstrated that the observed reduction in BP was strongly associated with decreased urinary protein excretion, lower renal fibrosis, and lower protein casts.⁹

Figure 1.

Histone β-hydroxybutyrylation is elevated with 1,3-butanediol treatment. Groups of Dahl salt-sensitive rats on a high-salt diet (2% Nacl) were supplemented with or without 1,3-butanediol (20% v/v) in their drinking water as described in the methods section. A and B, Systolic blood pressure of males (n=5–6/group) and females (n=5/group). C and D, Serum levels of β-hydroxybutyrate in males (n=4–6/group) and females (n=6–8/group). E and F, Immunoblotting for β-hydroxybutyrylation. G and H, Quantification of the blots (n=4–7/group). All data are mean±SEM,*P<0.05, **P<0.01, ****P<0.0001. BHB indicates β-hydroxybutyrate; H3K9BHB, histone 3 lysine 9 β-hydroxybutyrylation; and H3, histone 3

Enhanced Histone β-Hydroxybutyrylation in 1,3-Butanediol-Treated Rats

Next, we examined histone 3 lysine-9 β-hydroxybutyrylation, which is the modification reported to be affected during starvation.¹⁷ As seen in Figure 1C and 1D, both males and females treated with 1,3-butanediol exhibited a significant increase in renal histone 3 lysine-9 β-hydroxybutyrylation. These data are presented with equal protein loading of isolated histones. Despite this, for reasons unknown, we noticed that the levels of H3 were also elevated in samples from the rats treated with 1,3-butanediol. However, to maintain rigor, β-hydroxybutyrylation was quantified using H3 as the normalizing factor. As seen in Figure 1E and 1F, β-hydroxybutyrylation was more prominent in males than in females. Subsequent multiomics studies were conducted in male rats and validated in both sexes. Henceforth, the group of rats with increased renal histone 3 lysine-9 β-hydroxybutyrylation will be referred to as the β-hydroxybutyrate (BHB) group.

Histone β-Hydroxybutyrylation Promoted Large-Scale Chromatin Remodeling

To assess the extent of differential chromatin accessibility attributed to histone β-hydroxybutyrylation, we performed ATAC-seq. In the BHB group, 3494 genomic regions or loci (regardless of whether they were protein-coding or not) were detected in the open configuration, implying enhanced accessibility for transcriptional regulation (Figure 2A). Similarly, 7404 loci were open in the control group indicating that in the BHB group these regions were closed or compacted regions of the genome with decreased accessibility for transcription (Figure 2A). In Figure 2B, the x axis denotes the genomic annotations, and the y axis denotes proportions of loci (genomic regions) in each category falling into the annotation type. The proportion is calculated using the total number of loci in each category with a particular annotation (number at the top of each bar divided by the total number of annotations in the respective category). Most of the remodeling occurred in the interCGI regions (regions of the genome that lie between CpG islands), followed by intergenic regions (parts of the genome that lie between genes) and introns. Interestingly, 249 and 196 promoter regions were open in the BHB and control groups respectively (Figure 2B).

Figure 2.

Chromatin accessibility, gene expression, and protein levels are altered with 1,3-butanediol treatment. A, Volcano plot for β-hydroxybutyrate (BHB) vs control groups. Each data point is a tested locus. The vertical lines correspond to the logFC cutoff. B, Annotation summary of loci tested in BHB vs control groups. x axis: genomic annotations; y axis proportion of loci in each category within each type of annotation. All Tested loci (red bars), serve as a background by which to compare the significant loci (blue and green bars). Loci or genomic regions were calculated based on differentially open if false discovery rate (FDR) <0.05 and |logFC| >0.585. C, Pathways associated with the upregulated genes of BHB group. D, Pathways associated with upregulated proteins in the BHB group. E, Venn diagram showing common upregulated genes. Different shades of blue circles depict ATAC-seq, RNA-seq, and Proteomics data and numbers within the circles are numbers of genes/proteins significantly upregulated. F, Chromatin accessibility for genes of interest. The y axis represents chromatin cut sites and thus open chromatin, and x axis represents the chromatin location of genes of interest. Black peaks represent the control group, and green peaks represent BHB group (n=3/group).

Histone β-Hydroxybutyrylation-Mediated Remodeling of the Renal Transcriptome and Proteome

To uncover the extent to which histone β-hydroxybutyrylation regulated the transcriptome, we performed renal RNA-sequencing. In the BHB group compared with the control, a total of 186 genes were upregulated and 490 genes were downregulated (Table S4). Pathway analysis using these upregulated genes revealed that the top 2 upregulated pathways were pantothenate and CoA biosynthesis and fatty acid degradation (Figure 2C).

Next, to examine the impact of the chromatin-remodeling responsive transcriptome on the proteome, we conducted a mass spectrometry-based quantitative proteomic analysis. A total of 119 and 79 proteins were significantly upregulated and downregulated, respectively, in the BHB group compared with control (Table S5). Pathway analysis of the proteomics data using these upregulated proteins indicated that butanoate metabolism, fatty acid biosynthesis, and fatty acid metabolism were the upregulated pathways based on the differentially expressed proteins (Figure 2D). Interestingly, pathways identified in the transcriptome that were also upregulated in the proteome included fatty acid metabolism, tryptophan metabolism, PPAR signaling and metabolic pathways (Figure 2C and 2D).

In contrast, the downregulated transcriptome (Table S4) and proteome signatures (Table S5) were highly related to the overall theme of immunity and infection. The common pathways identified in both downregulated transcriptome and proteome included extracellular matrix-receptor interaction, focal adhesion, phagosome, toxoplasmosis, viral myocarditis, amebiasis, and leishmaniasis (Figure S1A and S1B).

Chromatin Remodeling by Histone β-Hydroxybutyrylation Upregulated Lipid Catabolism

Next, we explored whether the observed alterations in the transcriptome and proteome of the BHB group was specifically due to chromatin remodeling via histone β-hydroxybutyrylation. We prioritized common differentially regulated outputs between chromatin states identified by ATAC-seq, RNA-seq, and proteomics data sets. Such a combinatorial analysis revealed that there were 10 upregulated proteins aligned with enhanced transcription caused by histone β-hydroxybutyrylation (Figure 2E). Intriguingly, 4 of the top genes shared a common function, lipid metabolism. These were Hmgcs2 (3-hydroxy-3-methylglutaryl-coenzyme A synthase 2), Acaa1b (acetyl-coenzyme A acetyltransferase 1B), Cyp2d4 (cytochrome P450, family 2, subfamily d, polypeptide 4), and Cyp2e1 (cytochrome P450, family 2, subfamily e, polypeptide 1; Figure 2E). Chromatin accessibility of the promoter regions of Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 was higher in the BHB group compared with control (Figure 2F). These data indicated that histone β-hydroxybutyrylation caused open chromatin within the promoter regions of Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1. Reactome pathway analysis using the 10 genes identified metabolism as the top upregulated pathway along with metabolism of lipids, synthesis and metabolism of ketone bodies, and β-oxidation of long-chain fatty acids (Figure S2E).

Collectively, these results led us to further prioritize β-hydroxybutyrylation as a key mechanism promoting lipid catabolism as a primary source of energy.

Metabolic Reprogramming in Response to Elevated Histone β-Hydroxybutyrylation

To determine whether rats treated with 1,3-butanediol were preferring lipids for energy, we housed both groups of control and BHB rats in the Comprehensive Laboratory Animal Monitoring System and monitored their metabolic parameters. Overall, both the volume of oxygen consumption and volume of carbon dioxide production were lower in the BHB group compared with control (Figure 3A and 3B). Importantly, the respiratory exchange ratio was dramatically decreased in the BHB group compared with the control group (Figure 3C). A lower respiratory exchange ratio points to lipids, not carbohydrates, as the primary source of energy.^44,45 These data provided further evidence that histone β-hydroxybutyrylation enhanced transcription of key metabolic genes in the lipid-mobilizing pathways. Those key genes, Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 likely contributed to a preferential switch in bioenergetic fuel source to lipids.

Figure 3.

1,3-butanediol treatment reduces respiratory exchange ratio (RER) and inhibits mTORC1 (mammalian target of rapamycin complex 1). A through C, Metabolic comprehensive laboratory animal monitoring system data showed reduced volume of oxygen consumed (VO₂), volume of carbon dioxide production (VCO₂), and respiratory exchange ratio (RER) level in the β-hydroxybutyrate (BHB) group compared with controls. Black line: control group and green line: BHB group (n=5/group). D and E, Reduced phospho-S6 ribosomal protein levels were found in male rats treated with BHB compared with control. See also Figure S5 for data from female rats (n=5–6/group) All data are mean±SEM; *P≤0.05, **P≤0.01, and ***P≤0.001.

Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 Are Bona Fide Targets Linking Epigenetic Renal Histone-3 β-Hydroxybutyrylation to Energy Metabolism

In addition to β-hydroxybutyrylation, β-hydroxybutyrate is a histone deacetylase inhibitor and is also known to epigenetically modify histones by acetylation.^7,46 Therefore, we tested and confirmed that histone-3 lysine-9 acetylation and histone-3 lysine 23 acetylation were significantly upregulated with 1,3-butanediol treatment (Figure S3A and S3B). Hence, it was important to determine whether the 4 upregulated energy metabolism genes, Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1, which were prioritized in our study were specific targets of β-hydroxybutyrylation. We therefore performed ChIP assays using an antibody specific to β-hydroxybutyryl-histone H3 (Lys 9) β-hydroxybutyrate. ChIP assay using this antibody revealed that promoter regions of Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 were significantly enriched in the BHB group (Figure 4A), thus confirming that these genes were bonafide targets of β-hydroxybutyrylation.

Figure 4.

Validation of common upregulated genes through chromatin-immunoprecipitation quantitative polymerase chain reaction (ChIP-qPCR), real-time PCR, and proteomics. A, ChIP-qPCR data demonstrating enrichment of Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 in the β-hydroxybutyrate (BHB) group compared with control (n=3 replicates/group). Black bar: Control, Green bar: BHB. B, Real-time PCR showed higher expression of Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 in BHB group compared with control (housekeeping gene is L36a; n=5–6/group). C, Normalized relative abundance of Hmgcs2, Acaa1b, Cyp2d4, Cyp2e1 proteins (n=3–6/group). See also Figure S2 for female data. Black open circle-control, green closed circle-BHB. All data are mean±SEM; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 and #P=0.0567.

Next, transcripts of these genes were quantitated by real-time quantitative polymerase chain reaction analysis. The abundance of Hmgcs2 was prominently upregulated in the BHB group compared with control (Figure 4B; Figure S2A). Similarly, the abundances of Acaa1b, Cyp2d4, and Cyp2e1 were also upregulated in the BHB group compared with controls (Figure 4B; Figure S2B through S2D). Aligned with these data, quantitative mass spectrometry using high-quality MS3 spectra indicated that all 4 protein products of Hmgcs2, Acaa1b, Cyp2d4 and Cyp2e1 were significantly upregulated in the BHB group compared with control (Figure 4C). Collectively, these data demonstrate that β-hydroxybutyrate, via conferring epigenetic histone β-hydroxybutyrylation, upregulated enzymes that mobilize lipids for energy metabolism.

Upregulation of Hmgcs2-Mediated Fatty Acid Oxidation Increased Renal Proximal Tubule Epithelial Cell Mitochondrial Circularity Index

One of the prominent targets upregulated due to histone β-hydroxybutyrylation was the mitochondrial protein Hmgcs2. Hmgcs2 is a cytosolic and rate-limiting enzyme regulating mitochondrial fatty acid oxidation to promote ketogenesis.^47,48 Ketogenesis is largely attributed to the liver, but recent work demonstrates that Hmgcs2 in the kidneys also participates in this pathway.^49,50 Because ketogenesis in the liver increases mitochondrial circularity, we asked whether such alterations in the ultrastructure of mitochondria also occur in the kidney. Transmission electron microscopy revealed more circular mitochondria in the proximal tubule epithelial cells from the BHB group compared with the elongated mitochondria in the controls (Figure 5A). This observation was confirmed by morphometric analyses of the measurement of average mitochondrial area, area/perimeter, and circularity index of these mitochondria. As seen in Figure 5B through 5D, each of these morphometric measures was elevated in the BHB-supplemented group compared with the controls. These data support our conclusion that histone β-hydroxybutyrylation mediated upregulation of Hmgcs2 to promote ketosis contributed to enhanced renal mitochondrial circularity index.

Figure 5.

Kidney proximal tubule epithelial cell mitochondrial circulatory index was found to be higher in β-hydroxybutyrate (BHB) group compared with control. A, Representative transmission electron microscopy images of renal mitochondria. Orange arrows point to mitochondria. B through D, Bar graph shows quantification (B) Average mitochondrial area (C) area/perimeter, and (D) Circularity index (n=30 images per group). A total of 3 rats per group were used. All data are mean±SEM, ***P<0.001, and ****P<0.0001.

Upregulation of Acaa1b-Mediated Fatty Acid Oxidation Increased Peroxisomal Biogenesis

Fatty acid oxidation occurs both in mitochondria and peroxisomes.^51–56 Whereas mitochondria are the main site of oxidation of medium- and long-chain fatty acids, peroxisomes catalyze the β-oxidation of a distinct set of fatty acids, including long-chain fatty acids.^54–56 While Hmgcs2 is located within mitochondria, interestingly, another target of histone β-hydroxybutyrylation, Acaa1b, which is responsible for β-oxidation of fatty acids, is located within peroxisomes.^57,58 In response to metabolic stress, peroxisomes proliferate by upregulating their biogenesis.⁵⁹ One of the factors required for peroxisomal biogenesis is peroxisomal biogenesis factor 11γ.⁶⁰ Interestingly, proteomic data revealed that Pex11γ (peroxisomal biogenesis factor 11 gamma) was increased in the BHB group, which indicated that peroxisomal biogenesis was promoted in the BHB group (Figure S4A). To further confirm this observation, we examined the renal sections for the abundance of peroxisomes using the classical marker Pmp70 (peroxisomal membrane protein 70). Peroxisome staining was more prominent in the BHB group compared with control (Figure S4B and S4C), suggesting that the upregulation of the peroxisomal target of histone-β-hydroxybutyrylation, Acaa1b, placed an increased demand for fatty acid oxidation, which in turn promoted peroxisome biogenesis.

β-Hydroxybutyrylation-Mediated Enhanced Fatty Acid Oxidation Inhibits Mammalian Target of Rapamycin Complex 1

Stimulation of peroxisome biogenesis by drugs such as rapamycin are known to inhibit mTORC1 (mammalian target of rapamycin complex 1) activity.^61,62 Hyperactivation of mTORC1 signaling is associated with several human diseases, whereas suppression of mTORC1 is known to curb senescence, extend lifespan in yeast, C. elegans, Drosophila, and mice, and promote autophagy.⁶³ Importantly, mTORC1 inhibits fatty acid β-oxidation, whereas inhibition of mTORC1 by rapamycin promotes fatty acid oxidation.⁶⁴ Rapamycin suppresses mTORC1 signaling to ameliorate kidney injury and hypertension in Dahl salt-sensitive rats.¹⁶ Based on the similarity in function between rapamycin and the targets of β-hydroxybutyrylation to promote fatty acid oxidation, we hypothesized that mTORC1 is inhibited in the BHB group to protect kidneys. In support, mTORC1 was significantly inhibited in the BHB group compared with control (Figure 3D and 3E; Figure S5A and S5B).

Chromatin Remodeling by Histone β-Hydroxybutyrylation Resulted in Downregulation of Immune Function Genes

Next, we examined the downregulated loci, transcripts, and proteins in the BHB group (Figure 2A). These are open chromatin regions in the control group and closed chromatin region in the BHB group. Similar to the shared upregulated genes and proteins, the combinatorial analysis using ATAC-seq, transcriptomic, and proteomic data sets showed that 9 proteins were downregulated in all the 3 analyses (Figure 6A). Among these, the top 2 were Ptprc (protein tyrosine phosphatase receptor type C) and Lcp1 (lymphocyte cytosolic protein 1). Chromatin accessibility of the promoter regions of Ptprc and Lcp1 was lower in the BHB group compared with control (Figure 6B). These results indicate that histone β-hydroxybutyrylation mediated chromatin compaction of the promoter regions of Ptprc and Lcp1 contributed to the observed lower transcription (Figure 6C) and translation (Figure 6D) of these loci.

Figure 6.

Chromatin accessibility, gene expression, protein levels, and CD3⁺ T cells proliferation were altered with 1,3-butanediol treatment. A, Venn diagrams showing common downregulated genes and proteins leading to downregulation of immune function pathways. Different shades of blue circles are used to indicate ATAC-seq, RNA-seq, and Proteomics data. The gray box lists the names of common downregulated genes and proteins, beneath which are shown the associated downregulated pathways. B, Chromatin accessibility at the promoter region of Ptprc and Lcp1 promoter. The y axis represents chromatin cut sites and thus open chromatin, and x axis represents the chromatin location of genes of interest. Black peaks represent the control group, and green peaks represent β-hydroxybutyrate (BHB) group. C, Real-time polymerase chain reaction data showing reduced expression with Ptprc and Lcp1 in the 1,3-butanediol supplementation (n=5–6/group). D, Normalized relative abundances of Ptprc (Cd45) and Lcp1 detected in the quantitative proteomics study. Black open circle-control, green closed circle-BHB (n=3/group). E, Peripheral white blood cells and lymphocytes in the control and BHB treated male Dahl salt-sensitive rats. F, Representative histogram for percent carboxyfluorescein diacetate succinimidyl ester (CFSE)-positive CD3+ T cells andquantification for percent CFSE-positive CD3⁺ T cells in control and BHB groups after 5 days. Data are presented as mean±SEM and N was plotted for control and BHB group (n=7–8/group). All data are mean±SEM, *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001. CD3 indicates cluster of differentiation 3; CFSE, carboxyfluorescein diacetate succinimidyl ester; D1, daughter 1; D2, daughter 2 population; Lym, lymphocytes; P, parent population; and WBC, white blood cell.

Unlike the upregulated pathways of energy metabolism, Reactome pathway analysis of the downregulated genes showed that the overall downregulated pathways due to histone β-hydroxybutyrylation were related to immune function. These were phosphorylation of CD3 and TCR zeta chains, downstream TCR signaling, translocation of ZAP-70 to the immunologic synapse, PD-1 signaling, immune system, interferon gamma signaling, interleukin-12 family signaling, and TCR signaling (Figure 6A).

Histone β-Hydroxybutyrylation Promoted Downregulation of CD45 and LCP-1

Next, we focused on the 2 top genes which were prominently downregulated in our ATAC-seq study, Ptprc and Lcp-1. Lower expression of both Ptprc and Lcp1 in the BHB group was confirmed by quantitative polymerase chain reaction (Figure 6C). These data correlated with lower abundance of the protein products of these genes (CD45 and LCP-1, respectively) in the BHB group compared with the controls (Figure 6D). CD45 and LCP-1 are both important for the function of T cells, and dysfunctional regulation of T cells are implicated in renal disease and hypertension. Therefore, we focused on examining T cells using CD3⁺ as a pan-marker reporting on all T cells in our experimental groups. However, as expected, the ChIP assay showed comparable enrichment of promoter regions of Ptprc and Lcp1 in both the groups (Figure S6A and S6B), indicating that these are not direct targets of β-hydroxybutyrylation but consequential due to chromatin compaction.

Lower Expression of CD45 and LCP-1 Enhanced CD3⁺ T Cell Proliferation

Both CD45 and LCP-1 are present in all hematopoietic cells.^65–68 Therefore, we examined whether the downregulation of these 2 genes in the BHB group was facilitated by the epigenetic action of β-hydroxybutyrate reflected in immunomodulatory effects. Analysis of the complete blood count showed a trend toward reduced circulating monocyte and neutrophil counts with BHB treatment; but these differences did not reach statistical significance. However, a significant decrease in the level of circulating white blood cells, specifically lymphocytes, was noted in the BHB group (Figure 6E). Similarly, mean corpuscular volume, mean corpuscular hemoglobin, mean platelet volume, plateletcrit and platelet distribution width were also lower in the BHB group (Figure S7). Taken together, these data affirm an immunomodulatory function of β-hydroxybutyrate (Figure 6A; Figure S7A through S7F). In alignment, the parent population of CD3+ T cells was lower in the BHB group compared with controls (Figure 6F). Intriguingly, CD3+ T cells from the BHB group exhibited increased proliferation as indicated by a higher population of dividing cells (D2), while maintaining a similar population of D1 cells (Figure 6F). These findings suggest that the lower expression of CD45 and LCP-1 in the BHB group likely contributed to a rapid turnover of T lymphocytes by depleting existing parental cells and promoting the generation of a sufficient number of new effector T cells to maintain immune homeostasis. However, a key limitation is that we were unable to quantify T cells in the kidney tissues. Notably, in our previously published data, we demonstrated that BHB inhibits the infiltration of CD68+ macrophages into the kidneys.⁹

Discussion

This study was designed to delineate the molecular mechanism underlying the previously documented protective effect of BHB.⁹ Specifically, we examined the known function of β-hydroxybutyrate to epigenetically remodel chromatin by histone β-hydroxybutyrylation.¹⁷ Chromatin remodeling by histone β-hydroxybutyrylation exerted dynamic effects of distinctly modifying the renal transcriptome and proteome to promote both mitochondrial and peroxisomal regulation of fatty acid catabolism while parallelly dampening immune cell function. We have methodically dissected and characterized the major loci, transcripts, and proteins contributing to the specific consequence of renal histone β-hydroxybutyrylation to conclude that epigenetic chromatin remodeling by histone β-hydroxybutyrylation simultaneously regulates the dual processes of upregulation of fatty acid catabolism and dampening of immune cells. These results constitute a new mechanism underlying the dynamic beneficial reno-protective effect of β-hydroxybutyrate.

Histone β-hydroxybutyrylation was first identified as a new epigenetic modification occurring in the mouse liver in response to starvation,¹⁷ which is a condition simultaneously promoting ketosis and production of β-hydroxybutyrate.^11,69,70 Its epigenetic action to posttranslationally modify histones by β-hydroxybutyrylation has been subsequently studied in the context of a variety of pathologies including cardiomyopathy,⁷¹ depression,⁷² glomerulosclerosis,⁷³ and lung adenocarcinoma,⁷⁴ but none in any renal or hemodynamic studies. While our previous study on the benefit of β-hydroxybutyrate⁹ has been reproduced by others in different contexts such as preeclampsia and diabetic kidney disease, the underlying molecular mechanism has remained elusive.^75–77 In this context, our results presented in this study are the first to reveal that posttranslational modification of histones by β-hydroxybutyrylation is a novel mechanism protecting kidneys.

BHB is a known epigenetic modifier.⁴⁶ Along with its direct modification of histones, it can also indirectly modify histones by histone deacetylase inhibition. Our results here showed an increase of β-hydroxybutyrylation as well as acetylation (Figure 1; Figure S3A and S3B). This raised the question of which modification was responsible for the upregulation of lipid catabolism. To address this, we performed ChIP-PCR using a specific anti-H3K9 β-hydroxybutyrylation antibody and identified Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1 as bona fide targets of histone β-hydroxybutyrylation.

The food intake was significantly lowered in the BHB group, whereas the 24-hour activity was not different between the BHB and control groups (Figure S6C and S6D). Pathway analysis revealed an upregulation of fatty acid metabolism, which could be the reason for lowering of food intake in the BHB group. Mitochondria, a major organelle involved in fat metabolism, undergo morphological changes based on nutrient availability and stress conditions. mTORC1 is a known regulator of numerous cellular processes, such as cell proliferation, metabolism, and cell growth.⁷⁸ mTORC1 regulates mitochondrial metabolism and controls mitochondrial biogenesis.^79,80 Mitochondrial fission and apoptosis are controlled by mechanistic/mTORC1-mediated stimulated translation of MTFP1 (mitochondrial fission process 1).⁸⁰ Our study showed a distinct mitochondrial morphological alteration occurring in response to β-hydroxybutyrate, whereby mitochondria were more circular. Reduced circularity of mitochondria is reported to increase fatty acid oxidation, which concurs with our findings.⁸¹ Further, our data demonstrate that increased fatty acid oxidation was linked to a significant inhibition of mTORC1. In support, rapamycin, which is a uniquely specific mTOR inhibitor, increases fatty acid oxidation.⁶⁴ Similarly, fenofibrate, which is reported to have beneficial cardiovascular and renal effects, increased fatty acid oxidation and inhibited mTOR.⁸² Importantly, in the context of salt-sensitive hypertension and kidney injury, Kumar et al,⁸³ have shown that the inhibition of mTORC1 ameliorates both hypertension and renal injury. These data taken together with our findings reported in the current study, lead us to propose that epigenetic chromatin remodeling by histone β-hydroxybutyrylation promotes the renal fatty acid oxidation-mTORC1 axis to protect kidneys and lower hypertension.

An interesting finding of our study is that Hmgcs2, which is the main enzyme catalyzing the production of β-hydroxybutyrate from acetyl CoA, is also the target locus impacted by β-hydroxybutyrylation. This indicates that β-hydroxybutyrate autoregulates its own production by epigenetically remodeling chromatin to promote the transcription of Hmgcs2. If this process is perpetual, systemic β-hydroxybutyrate levels would remain consistently elevated after a single intervention to raise systemic β-hydroxybutyrate. Such a dysregulated continuous production of β-hydroxybutyrate would be detrimental as it would lead to ketoacidosis. Besides, β-hydroxybutyrate is elevated only during fasting and is lowered during the fed state.⁹ Similarly, β-hydroxybutyrylation was discovered to be elevated only during fasting.¹⁷ Collectively, these data support our conclusion that histone β-hydroxybutyrylation by β-hydroxybutyrate is not perpetual but turned off when β-hydroxybutyrate is limiting by yet undiscovered mechanisms.

A low-carbohydrate diet is reported to enhance human T cell immunity through immunometabolic reprogramming.⁸⁴ Such a low-carbohydrate diet could promote ketosis and inhibit mTOR. mTOR inhibition by rapamycin, which also lowers hypertension, is shown to promote T cell hypo-responsiveness or anergy.⁸³ In alignment with these observations, the downregulated pathways generated from downregulated genes/proteins led to lowering of T cell signaling, which implies that the mechanism by which T cell signaling is affected by β-hydroxybutyrate is a result of epigenetic remodeling of chromatin to a condensed state in regions harboring Ptprc and Lcp1.

The ChIP assays performed in this study conclusively demonstrate that H3K9 β-hydroxybutyrylation contributes to the transcriptional upregulation of the genes Hmgcs2, Acaa1b, Cyp2d4, and Cyp2e1. These are the only genes that are validated and thereby directly related to increased histone 3 lysine-9 β-hydroxybutyrylation. All other genes enlisted as identified via ATAC-seq and RNA-seq require additional confirmations.

The combinatorial ATAC-seq, RNA-seq, and proteomic data sets allow for the interpretation that the mechanistic effect of this remodeling is traced to impact renal energy metabolism and dampen the transcription of key genes in immune cell pathways. While our data are convincing to demonstrate that H3K9 β-hydroxybutyrylation cannot be ruled out as a mechanistic contributor, it may not be the sole mechanism. Further studies will be needed to clarify whether other factors could also contribute to the observed differential gene and protein expressions reported in our study.

Although our study reports a new epigenetic mechanism for lowering BP and kidney damage, it has some additional limitations. First, the study was conducted in Dahl salt-sensitive rats, which needs further validation from other experimental models of hypertension and kidney damage. Whether this operates in humans remains unknown. Second, the observations presented are largely in males. Future studies are required to establish whether BHB-mediated chromatin modification is a primary driver of the observed beneficial effect on BP and kidney damage. In addition, further in-depth studies are required to rule out the involvement of other known histone modifications affecting the genes identified as targets in our study. Lastly, the mechanism converging on mTOR requires additional studies to understand how mTOR inhibition lowers BP and protects kidneys.

In summary, our findings reveal that renal histone β-hydroxybutyrylation causes a dynamic increase in chromatin accessibility to the promoter regions harboring lipid catabolizing genes while simultaneously decreasing accessibility to the promoter regions of genes contributing to immune-related functions. Together, these newly discovered mechanisms contribute to the beneficial effects of β-hydroxybutyrate on renal-hemodynamic health.

Perspectives

This is the first study which demonstrates that in a ketogenic state, histone β-hydroxybutyrylation indeed occurs in the kidney to remodel renal chromatin. Key target genes impacted for their transcription via this remodeling of chromatin have been identified. The mechanistic effect of this remodeling is traced to impact renal energy metabolism and dampen the transcription of key genes in immune cell pathways. Our study serves as the foundation for further research into intricate details on which cell types are affected by this epigenetic effect of β-hydroxybutyrate. Nevertheless, it establishes that histone β-hydroxybutyrylation is at least in part contributing to the dynamic effects of a ketone body on the kidney and hypertension.

Article Information

Author Contributions

B. Joe designed the study; J. Mandal, S. Chakraborty, S. Aryal, I. Manandhar, X. Mei, B.S. Yeoh, B. Mell, A. Kleinhenz, R. Tummala, P. Saha, W.T. Gunning III, V. Basrur, I. de la Serna collected and analyzed data; J. Mandal, S. Chakraborty, S. Aryal, I. Manandhar, X. Mei, B.S. Yeoh, B. Mell, A. Kleinhenz, R. Tummala, P. Saha, W.T. Gunning III, V. Basrur, I. de la Serna, T. Yang, M. Vijay-Kumar interpreted the data; J. Mandal, S. Aryal, I. Manandhar, B. Joe, S. Chakraborty, V. Basrur, W.T. Gunning III, I. de la Serna, P. Saha, T. Yang contributed to writing the article, All authors read and edited the article.

Sources of Funding

National Institutes of Health: B. Joe (R01-HL171401, R01-HL143082); M. Vijay-Kumar (R01DK134053); American Heart Association: S. Aryal (25PRE1375711), I. Manandhar (24PRE1186688), P. Saha (855256), T. Yang (852969); Crohns and Colitis Foundation: P. Saha (854385); American Liver Foundation Liver Scholar Award: B.S. Yeoh; University of Toledo Startup funds: T. Yang; Melanoma Research Foundation: MRF; University of Toledo Stimulus and Bridge Awards: I. de la Serna.

Disclosures

None.

Supplemental Material

Tables S1–S5

Figures S1–S7

Supplementary Material

hyp-82-1729-s001.pdf ^{(337.8KB, pdf)}

hyp-82-1729-s002.pdf ^{(45.8MB, pdf)}

hyp-82-1729-s003.pdf ^{(21.5KB, pdf)}