Dapagliflozin Attenuates NKCC2 Protein Expression and Alleviates Diabetic Kidney Disease in Salt Loaded‐Hypertensive Diabetic Db/Db Mice

ABSTRACT

Dapagliflozin, a potent sodium‐glucose cotransporter 2 (SGLT2) inhibitor, slows diabetic kidney disease (DKD) progression and reduces blood pressure in diabetic mice, potentially via modulation of tubular sodium transporters and inflammatory pathways. RING‐type E3 ligases, including cullin‐based complexes, regulate the stability of with‐no‐lysine kinases (WNK1/4), key upstream activators of the sodium‐potassium‐chloride cotransporter 2 (NKCC2). Twelve‐week‐old male db/db mice were placed on a high‐salt diet to induce hypertension and accelerate DKD, then randomized to receive dapagliflozin or vehicle by oral gavage for 14 days. Metabolic cages were used for 24‐h urine collection, and glomerular filtration rate was measured using FITC‐sinistrin clearance. Blood pressure was assessed by tail‐cuff plethysmography. Kidneys were processed for histology, immunohistochemistry, Western blotting, and label‐free proteomics. Dapagliflozin markedly reduced systolic blood pressure and improved renal injury on Periodic acid–Schiff staining, accompanied by reduced lactic acid accumulation and a lower urinary albumin‐to‐creatinine ratio. Proteomic profiling demonstrated differential abundance of multiple signaling proteins between groups. Western blotting showed increased expression of cullin 1, 2, and 3 family members, with a significant reduction in NKCC2 and aquaporin‐2 protein levels in dapagliflozin‐treated mice. In cultured mouse thick ascending limb cells, dapagliflozin inhibited high glucose–induced activation of the NLRP3 inflammasome, indicating an additional anti‐inflammatory effect. Dapagliflozin confers renoprotection in salt‐loaded hypertensive db/db mice by lowering blood pressure, attenuating tubular and glomerular injury, reducing NKCC2 abundance possibly via enhanced cullin‐mediated degradation, and attenuating high‐glucose–induced NLRP3 protein expression.

In salt‐loaded, hypertensive diabetic db/db mice with diabetic kidney disease, dapagliflozin lowered blood pressure and reduced tubular and glomerular injury. Mechanistically, dapagliflozin increased cullin E3 ligase abundance, promoting degradation of NKCC2. Dapagliflozin treatment also suppressed aquaporin‐2 and NLRP3 inflammasome activation, collectively conferring renoprotection in diabetic kidney disease. The graphical abstract was created with BioRender.com.

1. Introduction

Sodium‐glucose cotransporter 2 (SGLT2) proteins are responsible for reabsorbing approximately 90% of glucose filtered by the kidneys and are predominantly located on the luminal surface of the proximal convoluted tubules. Additionally, they are also found in the human pancreatic cells, where they regulate the release of glucagon [1]. SGLT2 inhibitors lower blood sugar levels by increasing urinary glucose excretion [1]. FDA‐approved SGLT2 inhibitors that are frequently prescribed to diabetic patients include empagliflozin, canagliflozin, ertugliflozin, and dapagliflozin. SGLT2 inhibitors also contribute to glucose regulation by influencing multiple molecular pathways [2].

Recent studies show that dapagliflozin's actions are not exclusive to the proximal tubule; it also inhibits sodium‐coupled transport mechanisms in the distal tubule. In addition to normalizing blood glucose, dapagliflozin was shown to normalize high blood pressure in salt‐loaded hypertensive diabetic db/db mice in a mechanism involving inhibiting the association between the sodium chloride cotransporter (NCC) and the actin cytoskeleton linker protein ezrin in mouse DCT2 cells and inhibiting NCC phosphorylation [3]. A recent study suggested that dapagliflozin may promote urinary potassium excretion and correct plasma potassium levels in STZ mice on a high potassium diet by presumably increasing ENaC activity [4].

Cullins are scaffolding proteins that recruit multiple subunits to form the cullin‐RING E3 ubiquitin ligase (CRL) [5]. This complex has been shown to mediate a cGMP‐dependent increase in sodium‐potassium‐2 chloride cotransporter (NKCC2) ubiquitination in thick ascending limb (TAL). The ubiquitin ligase scaffold proteins cullin‐1, cullin‐2, cullin‐3, cullin‐4A, and cullin‐5 are expressed at the protein level in rat TALs [5]. CRLs transfer ubiquitin from an E2 enzyme to other proteins and the resulting ubiquitin chains are recognized by the 26S proteasome, resulting in proteolytic degradation of the ubiquitin‐tagged protein. Mutations in specific cullin genes have been shown to cause hereditary forms of hypertension. For instance, familial Hyperkalemic Hypertension (FHHt) is caused by mutations in the genes encoding cullin 3, Kelch‐Like 3 (KLHL3) and With‐No‐Lysine [K] (WNK) kinase family, WNK1 and WNK4. The activity of NCC is regulated by WNK1 and WNK4, and the abundances of these proteins are controlled by cullin 3 and its substrate adaptor KLHL3. Ferdaus et al. reported an increase in the abundance of phosphorylated NCC in CUL3‐Het/mRNA lacking exon 9 mice [6].

Diabetic kidney disease (DKD) remains a major complication of diabetes and a leading cause of chronic kidney disease and end‐stage kidney failure. SGLT2 inhibitors have been shown to have beneficial effects in the kidney [7], but the mechanisms underlying their kidney protective actions remain incompletely defined, particularly in settings where diabetes is combined with hypertension and high salt intake. In addition to improving glycemic control, SGLT2 inhibitors have been shown to reduce albuminuria, preserve renal structure, and modulate inflammatory [8] and hypoxic signaling pathways in an experimental DKD model [7]. However, most mechanistic studies have focused on proximal tubular responses, complement regulation, hypoxia‐inducible pathways, or general inflammatory signaling [7] leaving the distal nephron contribution less well understood.

The thick ascending limb is an important determinant of sodium handling and blood pressure regulation because it expresses the sodium‐potassium‐chloride cotransporter NKCC2, whose abundance and activity are tightly controlled by cullin‐RING E3 ubiquitin ligase pathways [5] and the WNK‐SPAK signaling cascade [9]. Prior studies have established that alterations in this pathway can influence NKCC2 abundance and salt‐sensitive blood pressure regulation [10, 11], but whether SGLT2 inhibition affects this distal nephron axis in diabetic kidney disease has not been directly addressed. This question is particularly relevant because SGLT2 inhibition triggers counter‐regulatory activation of sodium and water reabsorption in the loop of Henle and more distal nephron segments (Packer et al., PMID: 37486998) [12], raising the possibility that NKCC2 and its regulatory pathways contribute to this compensatory response in diabetic kidney disease.

We therefore investigated whether dapagliflozin could ameliorate renal injury and normalize blood pressure in salt‐loaded hypertensive db/db mice, and whether these beneficial effects are associated with altered NKCC2 abundance and cullin family protein expression in the kidney. In parallel, we examined the effects of dapagliflozin on aquaporin 2 (AQP2) protein expression in the kidney and on NLRP3 protein expression in mouse thick ascending limb cells under high‐glucose conditions to determine whether dapagliflozin also exerts direct effects on distal nephron water handling and inflammatory signaling. By integrating in vivo physiologic, histologic, proteomic, and in vitro cell‐based approaches, this study was designed to address the knowledge gap regarding whether dapagliflozin modulates sodium transport and inflammation in a diabetic kidney disease model complicated by hypertension and dietary salt loading.

2. Methods

2.1. Animals and Treatments

A total of 14 male db/db mice (BKS.Cg‐Dock7m +/+ Leprdb 116/J) 12 weeks of age were purchased from the Jackson Laboratory (Bar Harbor, Maine, USA). The mice were randomized into two groups with one group of mice receiving vehicle saline and the other group receiving dapagliflozin. All animal‐related procedures and experiments were carried out in accordance with a protocol that was reviewed and approved by the University of Florida Institutional Animal Care and Use Committee (IACUC). Furthermore, all aspects of the animal studies adhered to the guidelines and standards set forth by the National Institutes of Health in their publication titled “Guide for the Care and Use of Laboratory Animals,” ensuring ethical and humane treatment throughout the research process. The mice were fed a high‐salt diet containing 4% sodium chloride (NaCl), which was formulated using Type 1 water, agar, NaCl, and powdered chow (Teklad TD.94045.PWD; Envigo; Indianapolis, IN, USA). The prepared diet was portioned into small food cups and kept refrigerated at 4°C for a maximum of 3 days.

2.2. Blood Pressure Measurements and Randomization of Mice Into Groups

The mice were maintained on a 4% high‐salt diet for a period of 10 days to induce hypertension. The duration of salt‐loading for 10 days was chosen based on previous animal studies demonstrating that administration of 2%–4% NaCl for 10–14 days causes a significant rise in systolic blood pressure [13, 14]. On the tenth day, blood pressure measurements were taken via the tail‐cuff method to verify the onset of hypertension, defined as a systolic blood pressure exceeding 140 mmHg. The animals were then randomly assigned to either a control group, which received a vehicle (0.95 saline) or dapagliflozin at a dose of 1 mg/kg/day (MedChemExpress, Junction, NJ, USA) by oral gavage at 6 p.m. for 14 consecutive days. Blood pressure measurements were repeated on Day 24 of the study before the mice were euthanized (Figure 1).

FIGURE 1.

Experimental timeline for the db/db vehicle vs. dapagliflozin animal study. Diabetic db/db mice were individually housed and placed on a high‐salt diet at Day 0. Blood pressure was measured at Day 10 to confirm the establishment of hypertension. From Days 11–24, animals received daily oral gavage of dapagliflozin (1 mg/kg) or vehicle. At Day 24, blood pressure was reassessed and glomerular filtration rate (GFR) was measured. Animals were euthanized at Day 25, and organs were harvested for downstream analyses. N = 7 mice per group.

2.3. Metabolic Cage Studies

The mice were weighed and then individually housed in metabolic cages for 24‐h urine collections. 24‐h water intake and 24‐h urine collection were done at the beginning of the study (normal phase), after 10 days of salt loading (high salt phase), during the mid‐treatment phase and at the end of the study (completion of treatment) (Figure 1).

2.4. Preparation of Kidney Sections

Kidney sections cut at a thickness of 5 μm from db/db mice treated with either dapagliflozin (DAPA) or a vehicle (VEH) were mounted on the same microscopy slides. This co‐mounting technique was used to ensure that all comparative sections from both experimental groups were exposed to the same conditions during subsequent staining procedures.

2.5. Immunofluorescence Staining Protocol

Previously standardized immunofluorescence staining protocol was followed for all tissue sections. Slides were deparaffinized and antigen retrieval was performed by boiling in SG unmasking buffer. This buffer consisted of 10 mM sodium citrate, 1.85% v/v Simple Green, in nanopure water, with the pH adjusted to 6.0. Sections were washed in Phosphate Buffered Saline (PBS) before being incubated with FFPE blocking buffer (1% bovine serum albumin (Sigma‐Aldrich), 0.2% milk, 4% fish gelatin (Sigma‐Aldrich), and 0.3% Triton X‐100 (Sigma‐Aldrich) in PBS) for 1 h at room temperature. Primary antibodies (1:100 dilution ENaC γ (StressMarq), 1:15 dilution NKCC2 (DSHB), 1:200 dilution Aquaporin 2 (Novus)) were incubated for 1 h at room temperature. Slides were washed 3X in PBS (5 min each). Secondary antibodies (1:250 dilution CY3‐conjugated donkey anti‐rabbit, Jackson Immuno Research; 1:250 dilution Alexa‐Fluor 488‐conjugated donkey anti‐mouse, Jackson Immuno Research; 1:250 dilution CY5‐conjugated donkey anti‐mouse, Jackson Immuno Research) were incubated for 1 h at room temperature followed by 3X washes with PBS (5 min each). Coverslips were mounted using ProLong Gold mounting media (Thermofisher).

2.6. Imaging and Analysis

Immunostained kidney sections were imaged using a Nikon A1R confocal microscope, large, stitched images were obtained using a 20× (NA 1.2) and other images were obtained using a 40× oil objective (NA 1.3). All image processing and subsequent quantification of immunofluorescence were performed using the ImageJ/Fiji software. For quantification, regions of interest (ROIs) were manually selected to delineate tubular areas. Measurements for each ROI included area, mean intensity, and integrated density. To correct for background fluorescence, several non‐tubular ROIs were selected within each image, and their average mean intensity was calculated. The total corrected fluorescence for each tubule was then determined by the formula: Total Fluorescence = Integrated Density−(Area × Average Background Mean Intensity). Data on mean fluorescence intensity in DAPA treated mice (a.u.) was normalized to controls (vehicle treatment) and analyzed using unpaired t‐test/Mann–Whitney test on GraphPad Prism 10.

2.7. Preparation of Soluble and Membrane Fraction Kidney Lysates

The right kidney from each mouse in the two groups was subject to a lateral, longitudinal cut to expose the cortex and medulla. Based on the distinct texture and color difference between the two regions, a 50 mg section of tissue was excised from the outer region of the kidney while being careful not to obtain tissue from the medullary region. The tissue was homogenized in 500 μL of tissue protein extraction reagent (TPER; ThermoFisher Scientific, Waltham, Massachusetts) using an Omni TH homogenizer. The tissue lysates were solubilized on ice for 20 min, with vortexing every 5 min, and then centrifuged at 13,000 rpm for 10 min at room temperature using a Micromax benchtop centrifuge. Based on a previously published study that used a centrifugation speed of 100,000× g for 30 min to obtain membrane preparations [15], here we utilized a similar procedure but utilized a more optimal speed of 149,000× g for the same time period [16, 17, 18]. The supernatant was transferred to clean vials and subjected to ultracentrifugation at 34,000 rpm (149,000× g) for 30 min at 4°C using an Optima XE‐90 ultracentrifuge and SW55 rotor (Beckman Coulter). The soluble fraction (supernatant) was collected, while the pellets were resuspended in 200 μL of TPER and sonicated on ice for two 3‐s intervals, after which the membrane fraction was transferred to clean tubes.

2.8. BCA Protein Assay and Western Blotting

The total protein concentration of each tissue lysate was quantified using a Bicinchoninic acid (BCA) protein assay (ThermoFisher Scientific, Waltham, Massachusetts). Briefly, a 1:10 dilution of the soluble or membrane kidney cortex protein lysate was prepared in deionized water (ddH2O). Nine protein standards were prepared by serially diluting a stock solution of 2 mg/mL bovine serum albumin (BSA) (Sigma). The protein concentrations of the samples were then determined by using a standard regression equation generated after plotting the absorbance values of the known standards and concentrations.

Fifty micrograms of protein from soluble or membrane fraction kidney tissue lysates were loaded onto 4%–20% Tris·HCl polyacrylamide gels (ThermoFisher Scientific, Waltham, Massachusetts) and resolved using the Criterion electrophoresis system (Bio‐Rad, Hercules, California). The resolved proteins were then transferred to nitrocellulose membranes (ThermoFisher Scientific, Waltham, Massachusetts) using the Criterion transfer system (Bio‐Rad, Hercules, California) in Towbin buffer (25 mM Tris, 192 mM glycine, 20% methanol). The membrane was stained with 1× ponceau S for 10 min and imaged to verify uniform protein loading. After washing off the ponceau stain, the membranes were blocked with 5% nonfat dry milk in 1× Tris‐buffered saline (1X TBS) for 1 h at room temperature to prevent nonspecific binding. The membranes were subsequently washed three times with 1X TBS for 4 min each to remove excess blocking reagent. The membranes were then incubated overnight at 4°C with gentle rocking in primary antibody solutions prepared in 5% BSA in 1X TBS. After the incubation, the membranes were washed three times with 1X TBS for 4 min each to remove unbound primary antibodies. The membranes were then incubated with horseradish peroxidase (HRP)‐conjugated goat anti‐mouse or goat anti‐rabbit secondary (BioRad, Hercules, CA) antibody (1:3000 dilution) for 1 h at room temperature. Afterwards, the membranes were washed four times with 1× TBS for 4 min each to remove any excess secondary antibody. Protein bands were visualized by incubating the membranes with ECL reagent (Bio‐Rad) for 7 min, and the resulting chemiluminescent signals were captured using an iBright imager (Thermofisher Scientific, Waltham, Massachusetts). Densitometry was performed using ImageJ software [19].

2.9. Proteomics and Bioinformatics

Kidney samples from db/db treated with vehicle or dapagliflozin were homogenized in tissue protein extraction (TPER) buffer (ThermoFisher Scientific, Waltham, MA). The total protein concentration in each sample was determined using a BCA assay, and the sample volumes were then adjusted to ensure that 20 μg of total protein was used for the digestion process. Protein digestion was performed using a trypsin rapid digestion kit from Promega (Madison, Wisconsin) following the manufacturer's instructions. The samples were first incubated at 56°C for 30 min with 1 mL of dithiothreitol solution (0.1 M in 100 mM ammonium bicarbonate) to reduce disulfide bonds and denature the proteins. Next, 0.54 mL of 55 mM iodoacetamide was added to each sample, and the mixture was incubated in the dark at room temperature for 30 min to alkylate the cysteine residues and prevent reformation of disulfide bonds. Afterwards, 1 mL of freshly prepared trypsin enzyme was added to the samples, and digestion was carried out for 1 h at 70°C. The digestion reaction was then stopped by adding 0.5% trifluoroacetic acid (TFA), which inactivates the enzyme and stabilizes the peptides for downstream analysis. Immediately after digestion, mass spectrometry (MS) analysis was performed to minimize nonspecific cleavage and degradation. The MS analysis utilized a ThermoScientific Q Exactive HF Orbitrap mass spectrometer equipped with an EASY Spray nanospray source. For chromatographic separation prior to MS, a Nano‐LC/MS/MS system was used, specifically the UltiMate 3000 RSLCnano system. Mobile phase A consisted of water with 0.1% formic acid, while mobile phase B was acetonitrile with 0.1% formic acid. The loading pump used water with 0.1% trifluoroacetic acid as mobile phase A. Peptides were first desalted and concentrated using a Pharma Fluidics mPAC C18 trapping column, and then separated on a 50 cm mPAC analytical column maintained at 40°C. The flow rate was set at 750 nL/min for the initial 15 min and then reduced to 300 nL/min. The elution gradient started at 1% B and increased to 20% B over 100 min, then ramped up to 45% B in 20 min, for a total run time of 150 min. The EASY Spray source operated at 200°C and 1.5 kV, with full scan analysis from 375 to 1575 Da at 60,000 resolution, followed by MS/MS scans at 15,000 resolution.

Label‐free quantitation of precursor ion intensity was conducted using Proteome Discoverer (ThermoFisher Scientific, version 3.1.0638, Waltham, Massachusetts), searching against the Mus musculus database (including isoforms and sub taxonomies, version 2023‐06‐28) with trypsin specified as the digestion enzyme. The vehicle and dapagliflozin treatment groups were compared using a “non‐nested” study factor to account for variability. Normalization was performed using all detected peptides, and protein abundances were calculated by summing the abundances of the corresponding peptide groups. Statistical analysis was carried out using Fisher's exact test (pairwise ratio‐based) with low intensity resampling value imputation, and p‐values were adjusted using the Benjamini‐Hochberg method to control false discovery rate. The Reactome program was utilized for network analysis [20].

2.10. Urinary ACR

Twenty‐four‐hour urine samples were collected and centrifuged at 1500 rpm for 10 min at room temperature to remove cellular debris. The clarified supernatant was collected and diluted in the kit sample diluent to fall within the linear range of the assay, as recommended by the manufacturer. Urinary albumin was quantified using a mouse albumin ELISA kit (Immunology Consultants Laboratory, USA; Cat. No. E‐90AL) following the manufacturer's instructions. Briefly, standards (7.8–500 ng/mL) and urine samples were added to pre‐coated wells and incubated for 30 min at room temperature, followed by four wash cycles. Enzyme–antibody conjugate was then added and incubated for 30 min at room temperature; the plate was washed, and TMB substrate was applied for 10 min in the dark before addition of stop solution. Absorbance was read at 450 nm on a microplate reader, and albumin concentrations were calculated from a standard curve generated using known mouse albumin concentrations. Urinary creatinine was measured using a colorimetric creatinine assay kit (Abcam, ab204537). Twenty‐four‐hour urine samples were centrifuged at 1500 rpm for 10 min at room temperature, and the clarified supernatant was diluted in deionized water to fall within the linear range of the assay, as recommended by the manufacturer. Briefly, creatinine standards and diluted urine samples (50 μL/well) were added to a 96‐well microplate, followed by addition of the Jaffe reaction reagent, and incubated for 30 min at room temperature. Absorbance was then measured at 490 nm using a microplate reader, and creatinine concentrations were determined from a standard curve generated using known creatinine standards. Urinary albumin values were normalized to creatinine and expressed as the albumin‐to‐creatinine ratio (ACR).

2.11. Cell Culture and Immunofluorescence Microscopy for imTAL Cells

Immortalized Mouse Thick Ascending Limb (imTAL) cells were cultured on 35 mm glass bottom dishes (Mattek; Ashland, MA). Cells were cultured in Immorto media made of DMEM‐F12, 10% FBS, Glutamine (2 mM), INFγ (10 ng/mL), Insulin (10 mg/mL), Apo‐Transferrin (5.5 μg/mL), Sodium Selenite (5 ng/mL), and Antioxidant supplement (1×) and Penicillin/Streptomycin (1×). The cells were divided into 3 groups, one of which was maintained in normal glucose media (final concentration of 15 mM) and the other two were maintained in high glucose media (final concentration of 30 mM) for 3 days. The high glucose groups were then treated with vehicle (normal saline) or dapagliflozin (DAPA) (5 μM final concentration) for 24 h. The cells were fixed in methanol/acetone (1:1 ratio) for 10 min at −20°C, washed with 1× PBS, and then incubated in blocking solution (2.5% normal horse serum (Vector Labs; Newark, CA)) for 20 min. Afterwards, the cells were incubated with NLRP3 primary antibody (Abcam, Ab263899) (1:750 dilution) for 45 min and then subject to a series of 1× PBS washes before being incubated with Duet reagent (Vector Labs) for 30 min. To visualize nuclei, DAPI (Vector Labs) reagent was added to the mounting media (Vector Labs) and the solution was added to the dishes for cover slipping. Fluorescence imaging was performed using an Olympus microscope and the fluorescence intensity was measured using ImageJ.

2.12. Measurement of GFR

A 28 mg/mL FITC‐sinistrin stock was prepared by dissolving 56 mg of FITC‐sinistrin (Fisher, Medibeacon FTCFS001) in 2 mL of saline and incubating the solution overnight at 4°C, protected from light. Next, db/db mice receiving vehicle or dapagliflozin were shaved on the flank against the direction of hair growth, and the exposed skin was cleaned with water and thoroughly dried to provide a large measurement area. The body weight of each mouse was recorded and used to determine the injection volume using the calculation: body weight (g)/0.4 (μL). FITC‐sinistrin was drawn into 30‐gauge insulin syringes, which were then wrapped in foil, and the mice were anesthetized with isoflurane. The TGFR monitor was switched on, fixed to the depilated skin with adhesive while avoiding pigmented lesions, secured with tape, and baseline signals were recorded for 2–5 min. FITC‐sinistrin was administered via retro‐orbital injection. The device then continuously recorded fluorescence decay for up to 3 h, corresponding to sinistrin clearance. At the end of the recording period, the device and adhesive were gently removed, and the resulting data were processed using the Medibeacon analysis software.

2.13. Statistical Analysis

A Student's t‐test was utilized to determine whether there were statistically significant differences between two groups and a one way ANOVA was used to test significance between more than two groups. All statistical analyses in this study were carried out using the SigmaPlot 15.0 software package (Systat Software, San Jose, CA, USA). Statistical analysis for the immunofluorescence imaging of NKCC2/ENAC γ/Aquaporin 2 was done using GraphPad Prism 10. The results are reported as the mean value accompanied by the standard error (mean ± SE). For all analyses, a p‐value of less than 0.05 was considered to indicate statistical significance.

3. Results

3.1. Dapagliflozin Treatment Improves Kidney Function in Db/Db Mice With DKD After Salt Induced Hypertension

A published study from our group showed dapagliflozin alleviates hypertension induced by salt‐loading in diabetic db/db mice [3]. Spontaneous hypertension is known to have a deleterious effect on the kidney. Here, we induced hypertension by salt‐loading in db/db mice to exacerbate the development of DKD. Renal pathology was observed by periodic acid–Schiff (PAS) staining (Figure 2A,B) and by the accumulation of L‐lactate measured in kidney tissue lysates (Figure 2C). In addition, dapagliflozin improved urinary albumin‐to‐creatinine ratio (ACR) (Figure 2D) and improved GFR (Figure 2E) in db/db mice with DKD. These data suggest dapagliflozin ameliorates adverse effects of salt‐loading in the kidney of DKD mice.

FIGURE 2.

Analysis of pathological changes and kidney function in the kidneys of vehicle treated hypertensive diabetic db/db mice with DKD compared to dapagliflozin treated mice. (A) Representative PAS staining images from the two groups. Kidneys of vehicle treated mice (left) and dapagliflozin treated mice (right); (B) Quantification analysis of the PAS staining. N = 4 per group; (C) Quantification of L‐lactate concentration in kidney tissue lysates from the vehicle (VEH) treated group and dapagliflozin (DAPA) treated group. N = 4 per group; (D) Ratio of urinary albumin‐to‐creatine in the vehicle (VEH) treated group and dapagliflozin (DAPA) treated group. N = 4 per group; E. Glomerular filtration rate (GFR) in hypertensive db/db mice either treated with vehicle (VEH) or dapagliflozin (DAPA) for 14 days. N = 3 per group. A student's test is performed to determine the significance between the groups. * Represents p‐value < 0.05. Dashed lines within the boxed plots represent the means.

3.2. Dapagliflozin Lowers Blood Glucose and Blood Pressure in Hypertensive Diabetic Db/Db Mice With DKD

The beneficial effects of dapagliflozin are not limited to the inhibition of SGLT2. Dapagliflozin is thought to reduce sodium retention and hence blood pressure in the diabetic kidney [3]. Here we investigated whether dapagliflozin treatment could concurrently improve high blood glucose and high blood pressure in salt‐loaded hypertensive diabetic db/db mice with early stages of DKD. As shown in Figure 3, dapagliflozin treatment was effective in alleviating hyperglycemia (Figure 3A) and hypertension (Figure 3B) in these DKD mice. These data show dapagliflozin significantly normalizes both blood glucose levels and salt‐induced blood pressure in DKD mice.

FIGURE 3.

Dapagliflozin reduces blood glucose and blood pressure in hypertensive diabetic db/db mice with DKD. (A) Blood glucose measured in mouse plasma in vehicle (VEH) treated and dapagliflozin (DAPA) treated mice. N = 7 per group. A Student's t‐test was performed to compare the two groups. **** represents a p‐value < 0.0001. (B) Systolic blood pressure measurements taken during the PM of vehicle treated and dapagliflozin treated mice. N = 7 per group. A one‐way ANOVA was performed to compare the different groups. *** represents a p‐value < 0.001.

3.3. Dapagliflozin Differentially Regulates the Abundance of Various Types of Proteins in the Kidney of Db/Db Mice

Since our group previously showed dapagliflozin can inhibit the activity of NCC located in the renal cortex [3], we wanted to further identify other proteins in the cortex that may be affected by dapagliflozin treatment. We took a proteomic approach to identify peptides/proteins in the kidney cortex of salt‐loaded hypertensive diabetic db/db mice with DKD that are affected by dapagliflozin compared to vehicle treatment. Our proteomics data identified specific proteins in each group (Figure 4A–C, Table 1) and among these proteins were various signaling proteins including multiple kinases.

FIGURE 4.

Proteomic analysis of kidney cortex tissue lysates from vehicle treated hypertensive diabetic db/db mice with DKD compared to dapagliflozin treated mice. (A) Abundance ratio weights (DAPA/VEH), (B) Heatmap of the abundance ratio weights (DAPA/VEH), (C) Venn diagram showing the number of surrogate proteins and peptides identified between the two groups, Dapagliflozin treated (DAPA) and vehicle treated (VEH). N = 4 mice per group.

TABLE 1.

Table of various proteins found to be upregulated or downregulated between the two groups, Dapagliflozin treated (DAPA) and vehicle treated (VEH).

Protein	Enrichment in Dapagliflozin group	Abundance Ratio Adj. p‐Value: (DAPA)/(VEH)	Function	References
E3 ubiquitin‐protein ligase Itchy, RNF181	lower	< 0.0001	Transfers ubiquitin, tags proteins for degradation	[21]
E3 ubiquitin‐protein ligase RNF170	Higher	< 0.001	Transfers ubiquitin, tags proteins for degradation	[21]
Rho guanine nucleotide exchange factor 28	higher	< 0.0001	Regulates cytoskeleton, promotes cell motility	[22]
Phosphatidylserine synthase 1	lower	< 0.0001	Catalyzes phosphatidylserine formation from phosphatidylcholine or phosphatidylethanolamine	[23]
Non‐receptor tyrosine‐protein kinase TYK2	higher	< 0.0001	Mediates immune signaling for cytokine responses	[24]
Ras‐related protein Rab‐23	lower	< 0.0001	Regulates intracellular trafficking, inhibits Sonic hedgehog signaling	[25]
Sphingosine kinase 2	lower	< 0.0001	Catalyzes sphingosine phosphorylation producing signaling molecule	[26]
Sodium‐ and chloride‐dependent glycine transporter 1	higher	< 0.0001	Regulates glycine levels in neurotransmission	[27]
Serine/threonine‐protein kinase SMG1	lower	< 0.0001	Regulates mRNA decay, DNA damage response	[28]
Fatty acid desaturase 6	lower	< 0.0001	Introduces double bonds in fatty acids	[29]
Hephaestin	lower	< 0.0001	Facilitates iron metabolism in the kidney	[30]
Kelch‐like protein 9	higher	< 0.0001	Adapter in E3 ubiquitin ligase, regulates mitosis	[31]
Diacylglycerol kinase delta	lower	< 0.0001	Converts diacylglycerol to phosphatidic acid	[32]
Occludin	higher	< 0.0001	Regulates tight junction stability, controls barrier permeability	[33]
Renin receptor	higher	< 0.0001	Enhances angiotensinogen cleavage, activates intracellular signaling	[34]
Septin‐10	lower	< 0.0001	Cytoskeletal GTPase, regulates cell division.	[35]
Sulfiredoxin‐1	lower	< 0.0001	Restores antioxidant proteins, reduces oxidative stress	[36]
Phosphoserine phosphatase	lower	< 0.0001	Catalyzes final L‐serine biosynthesis step	[37]
Cadherin‐13	higher	< 0.0001	Modulates cell signaling, negative regulator of synaptic function	[38]
Death‐associated protein kinase 3	higher	< 0.0001	Regulates apoptosis, autophagy, cell migration	[39]
CaMKII subunit gamma	lower	< 0.0001	Calcium/calmodulin‐dependent kinase regulating memory	[40]
Spectrin alpha chain	higher	< 0.0001	Links plasma membrane to Actin cytoskeleton	[41]
Transportin‐3	lower	< 0.0001	Nuclear import receptor for splicing factors	[42]
MAP kinase‐activated protein kinase 2	lower	< 0.0001	Protects cells against stressors	[43]
Dysbindin	lower	< 0.0001	Regulates mitochondrial fission	[44]
Amiloride‐sensitive sodium channel subunit gamma	lower	< 0.0001	Regulate extracellular fluid volume and blood pressure	[45]
Prostasin	higher	< 0.0001	Regulating epithelial monolayer resistance and permeability in kidney	[46]
Alpha‐1‐antitrypsin 1–5	lower	< 0.0001	Serine protease inhibitor that inhibits activity of various proteases	[47]

Consistent with these results, bioinformatic analysis revealed that a large amount of the identified proteins that were found to be affected by dapagliflozin treatment are associated with the plasma membrane and involved in signal transduction (Figure 5). Further network analysis of the proteins that were found to be differentially expressed by the proteomic study indicated mRNA splicing, small molecule transport, PA synthesis, steroid hormone metabolism, and transport of vitamins, nucleotides, and related molecules were among the pathways affected by dapagliflozin treatment (Figure 5C). Collectively, these data show that dapagliflozin alters the function of a myriad of proteins in the kidney of DKD mice.

FIGURE 5.

Bioinformatic analysis of differentially enriched proteins in the kidney of db/db mice treated with vehicle or dapagliflozin. (A) cellular component, (B) molecular function. N = 4 per group, (C) Reactome was used for pathway analysis while inputting accession numbers corresponding to proteins with an abundance ratio adj. p‐Value: (DAPA)/(VEH) of < 0.005.

3.4. Dapagliflozin Treatment Reduces NKCC2 Protein Abundance in the Kidneys of Db/Db Mice

Motivated by data from our previous study that dapagliflozin treatment inhibits phospho‐NCC and total‐NCC protein expression [3] in the kidney of hypertensive diabetic db/db mice with DKD, in this study we investigated whether dapagliflozin could also have effects in the thick ascending limb to inhibit NKCC2 protein expression. First, immunofluorescence microscopy analysis showed NKCC2 protein expression is significantly decreased in the kidneys of db/db mice administered dapagliflozin compared to vehicle treated hypertensive diabetic db/db mice with DKD (Figure 6). Second, Western blotting and densitometric analysis corroborated these results and showed a significant decrease in an immunoreactive band corresponding to NKCC2 (Figure 7) after dapagliflozin treatment. Take together, these data show dapagliflozin at least indirectly attenuates NKCC2 protein expression in the kidney of DKD mice.

FIGURE 6.

Representative images of immunostained kidney sections from db/db mice that were treated with vehicle (A, B, F) or dapagliflozin (DAPA) (C, D, G) show NKCC2 (green), ENaCγ (red), Aqp2 (white) and nucleus (DAPI/blue). Images representing merged channels of NKCC2, ENaCy, and Aqp2 (B, D) are shown side by side with images showing the NKCC2 channel alone (A, C) to demonstrate decreased NKCC2 expression in the DAPA‐treated specimen. Scale bars = 50 μm. (E) Graph represents quantified mean fluorescence intensity results from the immunostained kidney sections. The box plot depicts the fold change in NKCC2 expression in the vehicle treated group and the DAPA treated group. The dotted horizontal line represents the baseline expression level in vehicle treated group at 1. The DAPA‐treated group shows a statistically significantly lower median fold change in NKCC2 expression compared to the vehicle group (Mann–Whitney t‐test, p‐value = 0.0199). N = 3 kidneys per group. Large, stitched mages represent immunostained whole kidney sections from vehicle‐treated (F) and DAPA‐treated mice (G) demonstrating expression of for NKCC2 (green), ENaCγ (red), Aqp2 (white) and nucleus (DAPI/blue). Scale bar = 1000 μm. VEH refers to vehicle treatment and DAPA refers to dapagliflozin treatment.

FIGURE 7.

Western blot and densitometric analysis of NKCC2 protein expression in membrane fractions from the kidney cortex of vehicle treated hypertensive diabetic db/db mice with DKD compared to dapagliflozin treated mice. (A) Western blot for NKCC2. (B) Ponceau stain for NKCC2 used to assess lane loading. (C) Densitometric analysis of the immunoreactive bands in the Western blot. A Student's t‐test was performed to compare differences between the two groups. *** represents a p‐value < 0.001. N = 7 mice per group. VEH refers to vehicle treatment and DAPA refers to dapagliflozin treatment.

3.5. Multiple Members of the Cullin Family Are Augmented in the Kidney of Db/Db Mice After Dapagliflozin Treatment

NKCC2 ubiquitination by a cullin‐RING ligase complex is known to regulate its surface expression in the thick ascending limb of the loop of Henle [5]. Since our proteomics data revealed that some E3 ubiquitin‐protein ligases are down‐regulated while others are upregulated in the kidneys of dapagliflozin treated mice compared to vehicle treated mice, we investigated specific cullin family members by Western blotting. Western blot and densitometric analysis showed protein expression of cullin‐1, cullin‐2, and cullin‐3 was significantly upregulated in the dapagliflozin treated group compared to the vehicle treated control group (Figure 8A–C). Cullin‐4A protein expression was comparable between the two groups (Figure 8D). These data suggest that dapagliflozin increases protein expression of multiple cullins in the kidney cortex of DKD mice.

FIGURE 8.

Western blot and densitometric analysis of Cullin‐1, −2, −3, and 4A protein expression in kidney cortex tissue lysates of vehicle treated hypertensive diabetic db/db mice with DKD treated compared to dapagliflozin treated mice. (A) Western blot for Cullin‐1 (top left) and Ponceau stain (bottom left) used to assess lane loading. Densitometric analysis of the immunoreactive bands in the Western blot (right). (B) Western blot for Cullin‐2 (top left) and Ponceau stain (bottom left) used to assess lane loading. Densitometric analysis of the immunoreactive bands in the Western blot (right). (C) Western blot for Cullin‐3 (top left) and Ponceau stain (bottom left) used to assess lane loading. Densitometric analysis of the immunoreactive bands in the Western blot (right). (D) Western blot for Cullin‐4A (top left) and Ponceau stain (bottom left) used to assess lane loading. Densitometric analysis of the immunoreactive bands in the Western blot (right). A Student's t‐test was performed to compare differences between the two groups. *, p‐value < 0.05; **, p‐value < 0.01. N = 4 mice per group. VEH refers to vehicle treatment and DAPA refers to dapagliflozin treatment.

3.6. Dapagliflozin Treatment Attenuates AQP2 Protein Expression in Db/Db Mice

Since dapagliflozin was found by our group and others to alter protein expression of multiple sodium coupled epithelial transport proteins in the kidney, we next investigated whether dapagliflozin treatment has an effect on the water channel AQP2. As shown in Figure 9, compared to vehicle treatment in hypertensive diabetic db/db mice with DKD, there was a significant reduction in renal AQP2 protein expression in db/db mice administered dapagliflozin.

FIGURE 9.

Western blot and densitometric analysis of aquaporin protein expression in membrane fractions from the kidney cortex of vehicle treated hypertensive diabetic db/db mice with DKD compared to dapagliflozin treated mice. (A) Western blot for APQ2. (B) Ponceau stain used to assess lane loading. (C) Densitometric analysis of the immunoreactive bands in the Western blot. A Student's t‐test was performed to compare differences between the two groups. * represents a p‐value < 0.05, *** represents a p‐value < 0.001. N = 7 mice per group. VEH refers to vehicle treatment and DAPA refers to dapagliflozin treatment.

3.7. Dapagliflozin Treatment Ameliorates High Glucose Induced Upregulation of the NLPR3 Inflammasome in Mouse Thick Ascending Limb Cells

Recent evidence suggests C‐reactive protein enhances inflammation in DKD via theNLRP3 inflammasome‐related pathway [48]. In addition, the NLRP3 inflammasome has emerged as an attractive target in the pathophysiology of DKD [49]. Due to the NLRP3 pathway being involved in the regulation of NKCC2 [50], here we investigated whether dapagliflozin treatment could affect NLRP3 inflammasome protein expression in thick ascending limb cells where NKCC2 is localized in the kidney. The immunofluorescence microscopy analysis of NLRP3 protein expression in cultured mouse thick ascending limb cells shows that high glucose stimulated upregulation of NLRP3 protein is blunted by dapagliflozin treatment (5 μM final concentration) (Figure 10). Because downstream inflammasome activation markers were not assessed in this study, these data indicate an effect on NLRP3 protein abundance rather than definitive suppression of inflammasome activation.

FIGURE 10.

Immunofluorescence microscopy of NLRP3 protein expression in mouse thick ascending limb cells treated with vehicle or dapagliflozin. Cells were maintained in normal glucose (NG) (final concentration of 15 mM) or high glucose (HG) (final concentration of 30 mM) and treated with vehicle (VEH) or dapagliflozin (DAPA). Representative images (top) of cells in each group. (A) Cells maintained in normal glucose (NG) media. (B) Cells maintained in high glucose (HG) media treated with vehicle (VEH). (C) Cells maintained in high glucose (HG) media treated with Dapagliflozin. Green fluorescence shows positive immunoreactivity for NLRP3. DAPI was used to stain cell nuclei. (D) Quantification plot of the immunofluorescence. A Student's t‐test was performed to compare differences between the two groups. *** represents a p‐value < 0.001. N = 3 independent experiments.

4. Discussion

In this study, we show for the first time that administration of dapagliflozin to salt‐loaded hypertensive diabetic db/db mice ameliorates high blood pressure, improves kidney function, and reduces renal NKCC2 protein expression likely due to increased protein expression of cullin protein family members, including cullin 3. Other data from this study show dapagliflozin treatment reduces AQP2 protein expression in the kidney. We also observed that dapagliflozin attenuates high glucose‐induced upregulation of NLRP3 protein expression in cultured mouse thick ascending limb cells. Although this finding is consistent with a reduction in inflammatory signaling, our analysis was limited to NLRP3 protein abundance and did not include downstream markers of inflammasome activation such as caspase‐1 cleavage or IL‐1β maturation. Therefore, these data do not allow us to conclude that dapagliflozin directly suppresses NLRP3 inflammasome activation, but they do suggest that dapagliflozin may modulate inflammatory pathways in thick ascending limb cells under diabetic conditions.

The 14‐day treatment period was selected to evaluate early renal and molecular responses to dapagliflozin in a salt‐loaded hypertensive db/db model, rather than to model chronic long‐term therapy. This timeframe is consistent with prior work showing that dapagliflozin can produce measurable biological effects in db/db mice over relatively short intervention windows. For example, Wu et al. reported that dapagliflozin altered renal disease progression in db/db mice during time‐course analysis, supporting the notion that short‐term treatment can reveal biologically meaningful responses [51]. Due to this study showing that dapagliflozin can induce measurable glycemic changes within as little as 2 weeks in preclinical studies, a 14‐day intervention was considered sufficient to evaluate early renal and molecular responses in our accelerated db/db model. Likewise, Chang et al. showed that dapagliflozin ameliorated diabetic kidney disease in db/db mice and modulated renal complement‐related signaling, indicating that renal protective mechanisms can be detected within a limited treatment period [7]. In addition, Lee et al. demonstrated that dapagliflozin improved generalized vascular dysfunction and reduced inflammatory markers in diabetic mice after 8 weeks of treatment [52], supporting the broader concept that dapagliflozin can induce meaningful physiologic responses within a finite experimental window. Taken together, these studies support our use of a 14‐day treatment period to assess early effects on blood pressure, renal injury, and transporter abundance in our model.

Ma et al. reported protein expression of the NaPi‐2a, NHE3, and ATP1b1, but not the SGLT1/2, is augmented in dapagliflozin‐treated diabetic mice when compared to vehicle control mice [53]. This group also showed dapagliflozin increased ENaC alpha, beta, and gamma mRNA expression in the collecting duct [53]. In one study, we showed that dapagliflozin treatment increased the concentration of multiple phosphatidylethanolamines in the kidneys of salt‐loaded hypertensive diabetic mice, and dapagliflozin treatment caused caveolin‐1 protein to shift from lipid raft fractions to non‐lipid raft fractions in cultured mouse proximal tubule cells [54]. Mechanistically, these studies suggest the effects of dapagliflozin may be due to an alteration of membrane lipids and lipid raft proteins. In another study by our group, we showed dapagliflozin's actions are not exclusive to the proximal tubule [3]. We showed salt‐loaded hypertensive diabetic db/db mice treated with dapagliflozin compared to vehicle have less active NCC [3]. In contrast to our study, a study by Chen et al. showed that in diabetic rats, dapagliflozin treatment does not affect AQP2 and NKCC2 protein expression after diabetes mellitus induced upregulation of these two proteins, although they did show AQP2 and NKCC2 proteins were upregulated in DM rats compared to control rats [55]. One possible reason for the differences in results of these two studies includes differences in species, as the study by Chen et al. used rats and we used mice in our study. Another reason for the discrepancy could be that diabetes was induced by streptozotocin in the study by Chen et al., and we used db/db mice predisposed to developing diabetes. A third reason is that we induced hypertension in our db/db mice by salt‐loading, while the rats in the study by Chen et al. were not hypertensive.

A number of experimental studies have likewise demonstrated renoprotective actions of dapagliflozin across diverse models of diabetic [7] or fibrotic kidney injury [56] but with distinct upstream targets compared with the cullin–NKCC2 axis identified here. In db/db mice, Chang et al. reported that dapagliflozin reduced albuminuria and ameliorated glomerular and tubular injury largely via upregulation of the complement regulatory protein Crry and suppression of tubular hypoxia and inflammation, without altering NKCC2 abundance, suggesting a primarily proximal tubular and immunometabolic mechanism that differs from the TAL‐focused effect observed in our salt‐loaded hypertensive db/db model [7]. Their study was conducted in standard‐chow db/db mice without a superimposed hypertensive stimulus, used a different treatment duration, and did not manipulate dietary sodium, factors that may explain the absence of NKCC2 changes and the predominance of glomerular and proximal tubular endpoints [7]. Other work in streptozotocin‐induced diabetic rats and adenine‐induced chronic kidney disease models showed that dapagliflozin attenuates renal fibrosis and collagen deposition, at least in part by inhibiting TGF‐β1/Smad or MAPK (ERK/p38) signaling, again without a defined role for TAL sodium transporters [56, 57]. These models differ from ours with respect to the etiology of kidney injury (toxin‐ or adenine‐driven rather than obesity‐ and salt‐driven), the absence of db/db‐associated metabolic syndrome, and less emphasis on blood pressure and NaCl transport phenotyping, which may shift the predominant dapagliflozin targets toward fibrogenic and inflammatory pathways rather than distal NaCl transport. Collectively, these studies support a broad renoprotective profile of dapagliflozin but contrast with the present findings, which highlight a previously unrecognized distal nephron mechanism in salt‐loaded hypertensive diabetic db/db mice whereby dapagliflozin reduces NKCC2 protein abundance along with changes in cullin expression, suggesting a possible cullin‐dependent mechanism and potentially linking SGLT2 inhibition to TAL NaCl handling and blood pressure regulation in DKD.

In salt‐loaded hypertensive db/db mice, dapagliflozin can improve or preserve GFR despite the high‐salt diet's tendency to worsen renal function because its beneficial hemodynamic, tubular, and anti‐fibrotic effects counteract the maladaptive responses to salt and diabetes, rather than simply adding to salt‐induced injury [58]. One of the suggested mechanisms behind it could be activation of tubuloglomerular feedback (TGF) via increased NaCl delivery to the macula densa, which causes afferent arteriolar vasoconstriction and lowers intraglomerular pressure, thereby reducing hyperfiltration and protecting the glomerulus [59]. Another mechanism could be lowering systemic and intrarenal blood pressure by promoting natriuresis and mild diuresis, which reduces the mechanical stress on the glomerular capillaries that would otherwise accelerate sclerosis and GFR decline in salt‐loaded hypertension [60]. By attenuating the salt‐induced upregulation of NCC and its interaction with ezrin in the distal convoluted tubule, Dapagliflozin reduces distal sodium reabsorption, enhances natriuresis, and further lowers blood pressure, [3] which could result in preservation of glomerular function.

Published studies on the role of SGLT2 inhibitors in DKD that specifically measured GFR have shown different effects depending on the disease stage. In early DKD, characterized by glomerular hyperfiltration [61, 62], SGLT2 inhibitors decrease or normalize high GFR [63, 64, 65], which leads to restoring tubuloglomerular feedback and lowering intraglomerular pressure. These results emphasize that SGLT2 inhibitors oppose the harmful hyperfiltration phase of DKD to help protect kidney function. However, published research indicates that in advanced or progressive kidney disease, where GFR is decreased, SGLT2 inhibitors like dapagliflozin stabilize GFR or slow its decline in patients with chronic kidney disease [66]. Overall, these findings suggest that the effects of SGLT2 inhibitors on GFR depend on the disease stage, with a protective reduction of hyperfiltration in early DKD and stabilization or improvement of kidney filtration capacity in later stages of DKD. Our findings align with this framework, as dapagliflozin‐treated mice with progressive DKD exhibited higher GFR compared to vehicle‐treated controls, suggesting preservation or improvement of filtration in the late disease stage.

A previous study reported that cullin‐1, cullin‐2, cullin‐3, and cullin‐4A protein are expressed in the kidneys of rats [5]. That study showed cGMP increases neddylation of cullins and cGMP dependent upregulation of NKCC2 ubiquitination is regulated by the cullin‐RING E3 ubiquitin ligase family [5]. Similarly, we show these cullin family members are also expressed in mouse kidney. Our Western blot studies show dapagliflozin treatment upregulates cullin‐1, cullin‐2, and cullin‐3A, but not cullin‐4 in hypertensive diabetic db/db mice. FHHt, also known as pseudohypoaldosteronism Type II (PHAII), has similar clinical characteristics as Gitelman syndrome and is caused by intronic deletions that increase WNK1 expression [67], missense mutations in WNK4 [67], recessive or dominant Kelch‐like 3 (KLHL3) mutations [68], and dominant cullin‐3 mutations [68]. KLHL3 is known to bind to substrate proteins and promote ubiquitination in a mechanism involving the interaction of the BTB domain with Cullin 3, a component of a Cullin/RING E3 ubiquitin ligase [69, 70]. The diverse role of ubiquitination includes targeting proteins for proteasomal degradation [71], alteration of protein–protein interactions [72], and modulation of protein activity [73]. Results from a study by Ohta et al. suggested that the CUL3‐KLHL3 E3 ligase complex regulates blood pressure by its ability to ubiquitylate WNK isoforms [74]. This group showed that WNK4 mutations that cause Gordon's syndrome inhibit interaction with KLHL3 that could raise blood pressure in a mechanism involving the increase in WNK4 expression leading to increased activation of NCC/NKCC2 and subsequent salt retention by the kidney. Although our protein biochemistry data suggest dapagliflozin treatment attenuates renal NKCC2 protein abundance and concurrently increases expression of multiple cullin family members, additional experiments are needed to further confirm whether cullins are directly involved in the downregulation of NKCC2.

The current findings of our study hold important translational significance, as they offer a novel molecular explanation for the known benefits of SGLT2 inhibitors in high‐risk patient populations. Dapagliflozin treatment in our model of salt‐loaded hypertensive diabetic db/db mice, which mirrors the clinical triad of diabetes, hypertension, and kidney injury, led to a marked reduction in systolic blood pressure and alleviation of renal injury, evidenced by improved urinary albumin‐to‐creatinine ratio. Mechanistically, we propose that dapagliflozin exerts these renoprotective effects by attenuating NKCC2 protein abundance, a key transporter regulating volume status, presumably by enhancing the cullin‐mediated degradation pathway through the observed upregulation of Cullin‐1, Cullin‐2, and Cullin‐3. These results link the drug's therapeutic action directly to the Cullin‐RING E3 ubiquitin ligase (CRL) system. Although we propose a Cullin‐dependent mechanism for the reduction in NKCC2 abundance after dapagliflozin treatment, we did not directly examine key intermediates of the canonical CUL3‐KLHL3‐WNK‐SPAK signaling axis. In particular, we did not assess WNK1, WNK4, SPAK, or their phosphorylated forms, which prevents definitive conclusion about whether dapagliflozin regulates NKCC2 through the established WNK‐SPAK pathway or via an alternative mechanism. Therefore, while our data are consistent with involvement of cullin‐mediated regulation, additional studies will be required to corroborate the cullin‐dependent mechanism and define the precise upstream signaling events linking dapagliflozin to NKCC2 downregulation. Likewise, our data suggest that dapagliflozin's efficacy in lowering blood pressure and protecting the kidneys in diabetic patients may involve boosting the kidney's intrinsic protein degradation system to selectively clear pro‐hypertensive sodium transporters, thereby providing a targetable pathway for managing salt‐sensitive hypertension and progression of DKD in the clinical settings.

Author Contributions

Z.S. and A.A.A. were involved in the design and conception of this study; Z.S., N.B., A.I., and A.A.A. performed experiments and analyzed data; Z.S., N.B., A.I., A.K., I.C., and A.A.A. drafted and edited this manuscript; A.A.A. and I.C. were involved in the supervision of this study; Z.S., N.B., A.I., A.K., I.C., and A.A.A. approved the final version of this manuscript.

Funding

This work was funded in part by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases Grant R01 DK123078 (to A.A.A.).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Original data generated and analyzed during this study are included in this published paper or in the data repositories.