Loss of biliverdin reductase-A promotes lipid accumulation and lipotoxicity in mouse proximal tubule cells
Abstract
Obesity and increased lipid availability have been implicated in the development and progression of chronic kidney disease. One of the major sites of renal lipid accumulation is in the proximal tubule cells of the kidney, suggesting that these cells may be susceptible to lipotoxicity. We previously demonstrated that loss of hepatic biliverdin reductase A (BVRA) causes fat accumulation in livers of mice on a high-fat diet. To determine the role of BVRA in mouse proximal tubule cells, we generated a CRISPR targeting BVRA for a knockout in mouse proximal tubule cells (BVRA KO). The BVRA KO cells had significantly less metabolic potential and mitochondrial respiration, which was exacerbated by treatment with palmitic acid, a saturated fatty acid. The BVRA KO cells also showed increased intracellular triglycerides which were associated with higher fatty acid uptake gene cluster of differentiation 36 as well as increased de novo lipogenesis as measured by higher neutral lipids. Additionally, neutrophil gelatinase-associated lipocalin 1 expression, annexin-V FITC staining, and lactate dehydrogenase assays all demonstrated that BVRA KO cells are more sensitive to palmitic acid-induced lipotoxicity than wild-type cells. Phosphorylation of BAD which plays a role in cell survival pathways, was significantly reduced in palmitic acid-treated BVRA KO cells. These data demonstrate the protective role of BVRA in proximal tubule cells against saturated fatty acid-induced lipotoxicity and suggest that activating BVRA could provide a benefit in protecting from obesity-induced kidney injury.
INTRODUCTION
Obesity is a significant public health issue that contributes to several cardiovascular and kidney diseases (11, 12, 30). Pathways through which obesity causes kidney disease are not well understood. Recent studies in animal models and humans have associated ectopic lipid accumulation in the kidney with obesity-related renal disease (15, 24). There is also growing evidence that ectopic lipid accumulation is associated with structural and functional changes of mesangial cells, podocytes, and proximal tubular cells; however, the mechanisms by which lipid accumulation causes cellular dysfunction in these various cell types are not known (8).
Bilirubin (BR) is a bile pigment generated by catabolism of heme released during the recycling of red blood cells in the spleen. Heme is initially broken down to biliverdin (BV) via heme oxygenase (HO), and then BV is rapidly reduced to BR via the enzyme biliverdin reductase A (BVRA) (37). Several large population studies have demonstrated a negative correlation between serum BR levels and obesity (4, 20, 23). Additional investigations have shown an inverse relationship between serum BR levels and hepatic steatosis and development of nonalcoholic fatty liver disease (25, 26, 28, 38). Apart from BR being found in the plasma, it can also be generated inside the cell via recycling of intracellular heme from cytochrome P450 proteins through the actions of both HO and BVRA, which implies that regulation of these enzymes may control responses to cellular lipotoxicity.
In the kidney, BVRA is ubiquitously expressed in both tubular segments as well as the renal vasculature (39). Despite its ubiquitous expression and important function, the regulation of BVRA in normal and diseased states of the kidney, such as diabetes, hypertension, and obesity, is not known. Moreover, little is known about the role of BVRA in proximal tubule cells, which are the primary renal cell type susceptible to obesity-induced lipotoxicity. We recently unveiled an antisteatotic role for BVRA in the liver using hepatocyte-specific BVRA knockout (KO) mice (17). These mice that lack BVRA, specifically in hepatocytes of the liver, developed marked hepatic steatosis in response to high-fat diet feeding. This increase in hepatic steatosis in hepatocyte-specific BVRA KO mice was associated with alterations in genes involved in fatty acid synthesis as well as genes involved in the β-oxidation of fatty acids (17). The goal of the present study was to test the hypothesis that BVRA protects proximal tubule cells against lipotoxicity. We used a CRISPR mediated approach to explicitly delete BVRA from cultured mouse proximal tubule cells and to determine the effect of loss of BVRA on development of fatty-acid induced lipotoxicity.
MATERIALS AND METHODS
Materials.
Mouse proximal tubule (MCT) kidney cells were as originally described (14). Bacteria protein expression [pET-32a (+)] plasmid DNA (Cat# 69015-3) was from EMD Millipore (Billerica, MA), NiCO21 (DE3) bacteria (Cat# C-2529-H) were from New England Biolabs (Ipswich, MA), and Ni-NTA Fast start kit 6xHis protein purification kit (Cat# 30600) was from Qiagen Inc. (Valencia, CA). Lactate dehydrogenase (LDH) Cytotoxicity detection kit (Cat# MK-401) and GuideIt-CRISPR/Cas9 system (Green) (Cat# 632601) plasmid were obtained from Clonetech (Takara Bio USA Holdings, Inc., Mountain View, CA). PrecisionX Multiplex guide RNA (gRNA) Cloning Kit (CAS9-GRNA-KIT) was obtained from System Biosciences (Palo Alto, CA). Free fatty acid (FFA)-free Bovine Serum Albumin (Cat# A-8806), Nile red (Cat# N-3013), Oleic acid (Cat# O-1383), Palmitic acid (Cat# P-0500), Protease inhibitor cocktail, phosphatase inhibitor cocktail, and NADPH/NADH, etc. were from SigmaAldrich (St. Louis, MO). BR and BV were from Frontier Scientific (Logan, UT; Cat# B-584-9 and 655-9, respectively). UnaG-pCDNA3 Flag plasmid was a kind gift from Dr. Atsushi Miyawaki (RIKEN Brain Science Institute, Saitama, Japan). iScript Reverse transcriptase supermix (Cat# 1708841), End point PCR master mix, and iTaq Universal SYBR Green Supermix (Cat# 172-5121) were from Bio-Rad Laboratories (Hercules, CA). Oligonucleotides (Table 1) for multiplex single guide RNA (sgRNA) targeting mouse biliverdin reductase A gene (Blvra), primers for real-time quantitative PCR (qPCR) and for end point PCR (genotyping) were all synthesized by Eurofins Genomics LLC (Louisville, KY). eBiosciences Annexin V-FITC Apoptosis Kit (Cat# BMS-500-FI-100) was from Thermo Fisher Scientific (Waltham, MA).
| ID | Oligonucleotide | Primer (5′-3′) | |
|---|---|---|---|
| 1 | sgRNA1 | aaaggacgaaacaccgAGCGGAAGCTCGTGAGCCGCgttttagagctagaaatagcaag | |
| 2 | sgRNA2 | ttctagctctaaaacGCGCTGCCGCAAATGACAATggatccaaggtgtctcatac | |
| Forward | Reverse | ||
| 3 | BVRA | CTCGTGAGCCGCTGGTAA | CTTCAGTCAGTGTCCGTGAAG |
| 4 | CD36 | CCTGCAAATGTCAGAGGAAA | GCGACATGATTAATGGCACA |
| 5 | BVRA (Exn2→2) genotyping | GTAAGGGACACCTTTGCTGC | CTGGTTGGTGGCTAGAGTGG |
| 6 | BVRA (Exn1→5) genotyping | AGCGCTCTGTCTGTAGCTG | ACTGTGACATGTGGCAACCA |
Cell culture.
MCT mouse proximal tubule cells were grown in 100 mm plates in maintenance medium made up of DMEM/F12 medium supplemented with 1% antibiotic (penicillin/streptomycin) and 10% fetal bovine serum in a humidified cell culture incubator at 37°C, 5% CO2.
Generation of BVRA Knockout KO MCT cells.
Two sgRNAs with high efficacy and low off-target scores were identified on Exon 1 and 5 of the mouse Blvra gene using Benchling online software. The two Cas9 targets were separated by 16,513 bases. All of the off-targets to our BVRA sgRNA had 4 mismatches, of which at least 1–2 were within the seed region (up to 12 bases proximal to the protospacer adjacent motif (PAM) site) which reduces the likelihood of Cas9 off-target effects. The multiplex sgRNAs were generated using the PrecisionX Multiplex gRNA Cloning Kit according to manufacturer instructions. Oligonucleotides used are listed in Table 1. The multiplex sgRNA fragments were then cloned into the GuideIt-zsGreen plasmid according to the manufacturer’s instructions. After sequence verification, 1 μg of the plasmid was transfected into cells in 12-well plates. After 36 h of transfection, cells with the top 10% level of fluorescence were single-sorted into 96-well plates by fluorescent activated cell sorting. After cells grew to confluence, individual wells were harvested with trypsin, and crude genomic DNA was obtained from two-thirds of the cells while the remaining one-third was left to continue growing. PCR was carried out on the genomic DNA samples using primers flanking the two cut sites (Exon 1→5; Table 1). Positive clones were identified by the presence of a 316-bp product (+/− depending on whether there is further insertion or deletion) indicative of Cas9-mediated targeting. Clones with the ≈ 316-bp product were sequentially expanded in 24-well and 6-well plates and then in 10-cm culture dishes. Definitive BVRA KO clones were finally determined by qPCR and Western blot.
Fatty acid treatment and Nile red staining.
One day after seeding cells (1.5 × 105 per well) in a 12-well plate, cells were treated with 400 μM Palmitic acid conjugated to FFA-free BSA (final BSA concentration of 0.5%) for 24 h. The FFA treatment medium was removed and cells detached with trypsin and centrifuged at 1,500 revolutions·min−1 for 5 min. Pellets were washed with 1 ml PBS and then resuspended in 400 µl solution of 1 µg·min−1 Nile red (in PBS) in the dark to measure triglyceride levels. Nile red is a lipophilic dye which only fluoresces when bound to lipids. This method has been shown to be comparable to both the enzymatic method as well as to BODIPY staining (16). Cell suspensions were then analyzed by flow cytometry in a Beckman Coulter Gallios Machine at Ex/Em: 488, Ex: 580 (16).
Real-time qPCR.
Total RNA was extracted from cells with Tryzol, and 1 µg was reverse-transcribed and diluted 1:15. Diluted cDNA (10 µl) was used in duplicates in a 25-µl real-time qPCR reaction. The relative expression of each gene was determined by the 2ΔC method with 18S as the housekeeping gene.
Western blot.
Cells were lysed with radioimmunoprecipitation assay (RIPA) buffer supplemented with 1:100 each of protease and phosphatase inhibitor cocktails. After clearing by centrifugation, 3 volumes of lysates were mixed with 1 volume of 4× loading buffer and heated at 95°C for 5 min before resolving by SDS-PAGE. Protein was transferred onto nitrocellulose membrane and detected by incubation with the respective primary and secondary antibodies sequentially. Membranes were scanned in a Li-Cor IR scanner, and optical density was measured with Image Studio. Level of expression was reported as relative optical densities of the respective protein normalized to the relative control. Antibodies were as follows: BVRA antibody (Cat# ADI-OSA-450-E) was from Enzo Life Sciences, Inc. (Farmingdale, NY), and heat shock protein α/β (F-8) (Cat# SC-13119), BVRA (SC-393385), and neutrophil gelatinase-associated lipocalin 1 (NGAL1) (SC-515876) antibodies were from Santa Cruz Biotechnology (Dallas, TX). PhosphoBAD s136 (4366), PhosphoBAD s112 (5284), and Caspase-3 (9662) were from Cell Signaling Technology (Danvers, MA).
IncuCyte fluorescent apoptosis assay.
IncuCyte ZOOM Live-Cell Imaging system (Essen Bioscience Inc., Ann Arbor, MI) was used for kinetic monitoring of apoptotic activity of wild-type (WT) and BVRA KO MCT proximal tubule cells. Cells were seeded at 5,000 cells per well in 96-well plates. Cells were treated with 400 μM of palmitic acid or vehicle in the presence of 5 μM of Caspase 3/7 Apoptosis Assay Reagent (Essen Bioscience). The Caspase 3/7 reagent labels dead cells yielding green fluorescence. Cells were cultured at 37°C and 5% CO2 and the plate scanned, and fluorescent and phase-contrast images were taken in real time every 2 h from 0 to 24 h posttreatment. Normalized green object count per well at each time point and quantified time-lapse curves were generated by IncuCyte ZOOM software. Data at the 24-h end point was graphed using GraphPad Prism software.
UnaG-based BVRA assay.
The method is described in more detail in (2). In brief, the DNA sequence of UnaG protein which specifically fluoresce with BR was cloned into the EcoRI-BamHI site of pET-32a(+) plasmid produced in NiCo21 bacteria and purified with a Ni-NTA Fast Start 6xHis Column kit. For the intracellular BVRA assay, cells were seeded overnight in 12-well plates. On the next day, the growth medium was replaced with serum-free medium containing 0 µmol/l or 40 µmol/l BV. After a 21-h exposure, cells were lysed with RIPA buffer, protein concentrations of the cleared lysate were determined, and 100 µl of lysate were incubated with UnaG 17.5 µg in a total volume of 500 µl made up with PBS. BR standards of 0 nmol/l, 10 nmol/l, 100 nmol/l, and 1,000 nmol/l containing the same concentration of UnaG were also used. After incubating at 37°C with shaking for 1 hr, BR concentration in the reactions were read using 200 µl of the reaction loaded in duplicates in a 96-well plate and read in a fluorometer (Biotek Synergy 2 multi-mode reader) with Ex 485 nm and Em 528 nm.
Cytotoxicity assay.
Following FFA-BSA treatments, medium was collected from each well and centrifuged at maximum speed for 5 min. The cleared medium (100 µl) was transferred in duplicates into a 96-well plate. High controls for each cell line were made by adding medium containing 1% Triton X-100 to the cells. One hundred microliters of 12.5%, 25%, and 50% of the respective high standards were also transferred into the wells. LDH reaction mix (100 µl) was added to the wells and incubated at room temperature for 20 min, and the optical densities were read at 491 nm and 611 nm. The % cytotoxicity was calculated according to manufacturer instructions.
Determination of apoptotsis (annexin-V FITC).
Treated cells were harvested with trypsin after collecting all the floating cells by centrifuging the medium. The combined pellets were washed with 1× binding buffer and then stained at room temperature with 10 µl annexin-V added to 195 µl 1× binding buffer for 10 min. Annexin-V solution was removed from the cells and washed with binding buffer. Cells were then stained with 10 µl propidium iodide solution added to 190 µl binding buffer. Binding buffer (200 µl) was later added, and the cell suspensions were analyzed by flow cytometry.
Seahorse cellular respiration analysis.
Cellular respiration was quantified using the Seahorse Extracellular Flux Analyzer XF-96 (Agilent Technologies, Cedar Creek, TX). We used the Seahorse XF Cell Energy Phenotype Test Kit (Agilent Technologies, Cat # 103325-100) for analysis of cellular respiration. WT and BVRA KO MCT cells were seeded on a 96-well plate at 20,000 cells per well. Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were used to quantify the cellular energy phenotype of WT and BVRA KO MCT cells with and without palmitic acid treatment. The cells were treated with 400 µmol/l palmitic acid or DMSO for 24 h in DMEM/F12 (1:1 of DMEM/High Glucose and F12/GlutaMAX) media with 0.5% BSA. Cells were washed twice with Seahorse Bioscience Assay Media (XF Base media with 25 mmol/l glucose, 2 mmol/l L-glutamate, and 1 mmol/l sodium pyruvate) then incubated with the buffer for 1 h in a nonCO2 incubator. The Seahorse Cartridge ports were loaded with 20 ml of assay media with 10 μM FCCP and 10 μM oligomycin.
Statistical analysis.
Data were analyzed with Prism 7.02 (GraphPad Software, San Diego, CA) using analysis of variance combined with Tukey’s post hoc test to compare pairs of group means or unpaired t-tests. Results are expressed as mean ± SE. Additionally, one-way ANOVA with a least significant difference post hoc test was used to compare mean values between multiple groups, and a two-tailed and a two-way ANOVA were used to analyze the genotype (WT vs. BVRA KO), treatment (DMSO vs. palmitic acid), and genotype treatment interaction effects followed by the recommended Tukey post hoc analysis for multiple comparison. P values of 0.05 or smaller were considered statistically significant.
RESULTS
Characterization of BVRA KO cells.
We confirmed disruption of the Blvra gene by PCR using genomic DNA isolated from the individual clones selected by flow cytometry for green fluorescent protein fluorescence. One clone was selected among several of the clones that showed the 316-bp signal which confirms the deletion of the targeted fragment within the Blvra gene (Fig. 1A). We avoided clones with any apparent structural abnormality as compared with WT cells. To further confirm the phenotype of the selected clone, we measured BVRA expression in the cells at both mRNA and protein level using specific primers and antibody, respectively. We confirmed the complete KO of BVRA from the cells by both real-time qPCR and Western blot (Fig. 1, B and C). We also confirmed the loss of BVRA by assessing its reductase activity (conversion of BV to BR) in the cells. As expected, WT cells converted BV to BR as indicated by the significant increase (14-fold) in intracellular BR after BV treatment (P < 0.0001) (Fig. 1D). However, there was no significant increase in intracellular BR levels after BV treatment in the KO cells. Thus, we confirmed loss of BVRA mRNA, protein, and activity in our CRISPR-mediated KO cells.
Fig. 1.Characterization of biliverdin reductase A (BVRA)-deficient mouse proximal tubule cells. A. genotyping scheme of CRISPR-Cas9 double targeted BVRA knockout (KO) cells. PCR primers were designed outside of exon 1 and in exon 5 to detect a 316 bp fragment (normally 16,513 bp) in double mutant cells. Primers were also designed in exon 2 and exon 3 to generate a control, 706-bp PCR product. B. real time PCR expression of biliverdin reductase A gene (Blvra) mRNA in wild-type (WT) (open bar) and KO (closed bar) cells, n = 3 per group. C. Western blot of BVRA protein in WT and KO cells, n = 3 per group. D. bilirubin concentration in WT and KO cells under basal conditions and 21 h after treatment with biliverdin (40 µmol/l). *P < 0.05 as compared with WT. #P < 0.05 as compared with WT vehicle. ^P < 0.05 as compared with biliverdin treated WT. HSP90, heat shock protein 90.
The loss of BVRA alters the energy phenotype following palmitic acid treatment.
We assessed changes in cellular energy and respiration using a Seahorse Extracellular Flux Analyzer XF-96 in WT and BVRA KO MCT cells with and without palmitic acid treatment. Palmitic acid-treated WT cells were less energetic and aerobic (Fig. 2A). Interestingly, the BVRA KO MCT cells were less energetic and aerobic than WT with no treatment and much more quiescent with palmitic acid. The loss of BVRA reduced the metabolic potential which indicates the energy demand via respiration and glycolysis. The loss of BVRA in the MCT kidney cells caused less metabolic potential basally and significantly (P < 0.05) less with palmitic acid treatment (Fig. 2B). Palmitic acid treatment in the WT MCT cells reduced OCR (Fig. 2C) and ECAR (Fig. 2D). The OCR and ECAR were significantly (P < 0.05) less in the BVRA KO MCT cells compared with WT and greatly reduced with palmitic acid. Cells treated with palmitic acid overall exhibited a more constant quiescent phenotype, whereas nontreatment cells displayed more metabolic flexibility in energetic capacity. The loss of BVRA and treatment with palmitic acid caused an exacerbated reduction in OCR, which is a measure of the rate of mitochondrial respiration of the cell and ECAR, which measures the rate of glycolysis.
Fig. 2.The loss of biliverdin reductase A (BVRA) changes the mitochondrial respiration and metabolic potential of mouse proximal tubule cells. A. oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were used to quantify the cellular energy phenotype of wild-type (WT) and BVRA knockout (KO) mouse proximal tubule (MCT) cells with and without palmitic acid (400 μM) treatment. B. metabolic potential measured as the percentage increase of stressed OCR over baseline OCR and stressed ECAR over baseline ECAR were analyzed in the WT and BVRA KO MCT cells with and without palmitic acid (400 μM) treatment. C. mitochondrial respiration rates of the WT and BVRA KO cells were measured by OCR. D: the rate of glycolysis of the WT and BVRA KO cells was measured by the ECAR. *P < 0.05 as compared with WT vehicle. #P < 0.05 as compared with palmitic acid-treated WT. ^P < 0.05 as compared with KO vehicle. N = 10–20 per group.
Loss of BVRA results in increased lipid accumulation.
Exposure of cells to high levels of saturated fatty acids is known to result in increased formation of lipid droplets which are composed mainly of triglycerides. We have recently shown that mice with a hepatocyte-specific BVRA KO accumulate more fat (17). To determine the effect of BVRA in palmitic acid-induced lipid accumulation, we treated WT and BVRA KO MCT cells with palmitic acid and BR for 24 h. Palmitic acid treatment increased fatty acid accumulation in both groups (Fig. 3A, n = 3 per group). However, the fat accumulation was significantly (P < 0.05) higher in the BVRA KO compared with WT with palmitic acid treatment. BR treatment resulted in a modest (~5%) but statistically significant (P < 0.05) effect on lipid accumulation in both WT and BVRA KO cells as compared with palmitic acid treatment alone.
Fatty acid translocase (CD36) is one of the major cell surface proteins responsible for uptake and transport of extracellular fatty acid across the membrane. To determine whether the higher levels of triglycerides in BVRA KO cells may depend on the increased uptake of the extracellular fatty acids mediated by CD36, we measured the mRNA expression of fatty acid translocase gene (Cd36) in the cells by real-time qPCR. Cd36 mRNA levels were significantly higher (P < 0.05) in BVAR KO as compared with WT vehicle (Fig. 3B, n = 6 per group). Palmitic acid treatment had no significant effect on Cd36 levels in either WT or KO cells. These results demonstrate that loss of BVRA increases lipid accumulation in response to increased delivery of palmitic acid in mouse proximal tubule cells possibly by increased Cd36 expression.
Fig. 3.Effect of biliverdin reductase A (BVRA) on palmitic acid-induced intracellular lipid accumulation. A. intracellular lipid accumulation as measured by Nile red fluorescence. B. (Cd36) mRNA levels were determined by real-time PCR in vehicle and palmitic acid (400 μM) treated cells. #P < 0.05 as compared with WT vehicle. *P < 0.05 as compared with WT palmitic acid. ^P < 0.05 as compared with KO vehicle. †P < 0.05 as compared with KO palmitic acid. n = 6 per group. BR, bilirubin.
BVRA KO cells are more sensitive to palmitic acid-induced lipotoxicity.
To determine if the effects of palmitic acid treatment on the cell pathology in WT and BVRA KO MCT cells, we first quantitated the percent population of healthy cells. In Fig. 4A, we show that palmitic acid treatment significantly (P < 0.05) decreased the number of healthy cells in both groups, but the effect was more pronounced in the BVRA KO cells (Fig. 4A, n = 3 per group). In comparison, palmitic acid increased the number of apoptotic cells in both groups (Fig. 4B, n = 3 per group). However, the effect was significantly (P < 0.05) greater in BVRA KO compared with WT cells treated with palmitic acid. NGAL1 (LCN2) is a known marker of kidney damage and has been shown to be important for the progression of kidney disease (5, 35, 36). Palmitic acid treatment significantly (P < 0.05) increased NGAL levels in both WT and BVAR KO cells (Fig. 4C, n = 3). The loss of BVRA resulted in a significant (P < 0.05) increase in NGAL levels under basal conditions. Importantly, palmitic acid treatment significantly (P < 0.05) increased NGAL levels in the BVRA KO cells higher than WT cells treated with palmitic acid. To evaluate the effect of BR on palmitic acid-induced cellular toxicity, we measured the LDH activity released from the cytosol of damaged cells into the supernatant. The LDH released was higher in palmitic acid-treated WT and BVRA KO cells (Fig. 4D, n = 3 per group). However, LDH release was markedly enhanced in the BVRA KO cells. BR treatment did not affect palmitic acid-induced LDH release in WT cells treated with palmitic acid. However, it did have a modest (5%) but significant (P < 0.05) effect to lower LDH release in BVRA KO cells. However, the level of LDH released in the BR and palmitic acid-treated BVRA KO cells was lower. These data demonstrate that loss of BVRA renders mouse proximal tubule cells more susceptible to fatty acid-induced cell damage, apoptosis, and cell death.
Fig. 4.Cell toxicity is higher in biliverdin reductase A knockout (BVRA KO) cells after treatment with palmitic acid. A. percentage of healthy cells. B. percentage of apoptotic cells following treatment with vehicle or palmitic acid (400 μM). C. Western blot of Neutrophil gelatinase-associated lipocalin (NGAL) in vehicle or palmitic acid (400 μM) treated cells. D: palmitic acid-induced cytotoxicity. #P < 0.05 as compared with WT vehicle. *P < 0.05 as compared with WT palmitic acid. ^P < 0.05 as compared with KO vehicle. †P < 0.05 as compared with KO palmitic acid. n = 3–4 per group. HSP90, heat shock protein 90.
A loss of BVRA in mouse proximal cells increases sensitivity to apoptosis.
The BVRA KO cells have increased sensitivity to palmitic acid-induced cytotoxicity, which might be due to enhanced apoptotic signaling. Apoptosis is mediated by the phosphorylation of the death signal and member of the BCL-2 family, BCL-2-associated death promoter (BAD), at Ser 136 and Ser 112 (7, 9). Thus, we measured BAD phosphorylation at both phosphorylation sites (pBADser136 and pBADser112) in cells treated with palmitic acid or vehicle. Palmitic acid treatment did not affect WT cells but significantly reduced pBADser136 levels in BVRA KO cells (P = 0.0061) (Fig. 5A). On the other hand, pBADser112 was significantly higher in the BVRA KO cells with vehicle treatment (P = 0.0001) (Fig. 5B). While palmitic acid did affect WT cells, it significantly reduced pBADser112 phosphorylation in the BVRA KO cells (P < 0.0001). To further determine if palmitic acid or the loss of BVRA in mouse proximal tubule cells causes the activation of apoptotic signaling pathways, we measured caspase-3 expression by immunoblotting which is a surrogate marker of apoptosis signaling and has been shown to be cleaved during lipotoxicity-induced apoptosis (29). As expected, the palmitic acid treatment caused caspase-3 cleavage, thus leading to reduced uncleaved and higher cleaved caspase-3 in the BVRA KO compared with WT (Fig. 6A, n = 4 per group). To further define the role of BVRA in apoptotic signaling, we used the Caspase-3/7 Green Apoptosis Assay for the IncuCyte live-cell analysis system that enables quantification of apoptosis over time. We treated the WT and BVRA KO MCT cells with palmitic acid or vehicle over 24 h and quantitated at every 2 h. At the 24-h end point, the palmitic acid-induced caspase 3/7 activation, as measured by Total Green Object, was significantly higher in WT and BVRA KO cells (Fig. 6B, n = 10 per group). The palmitic acid treatment in the BVRA KO cells had significantly (P < 0.05) higher caspase 3/7 apoptotic signaling compared with WT, which was also shown in Fig. 6C for the mean caspase 3/7 ratio normalized to the vehicle over the 24-h time span (Fig. 6C, n = 10 per group). Taken together, these data demonstrate that BVRA prevents palmitic acid-induced lipotoxicity in proximal tubule cells.
Fig. 5.BCL-2-associated death promoter (BAD) phosphorylation levels in wild-type (WT) and knockout (KO) cells after 24-h treatment with vehicle or palmitic acid (400 μM). A. levels of serine 136 phosphorylated BAD (pBADs136) normalized to heat shock protein 90 (HSP90). B. levels of serine 112 pBADs112 normalized to HSP90. All blots were quantitated by densitometry and graphed below the blots, respectively. #P < 0.05 as compared with WT vehicle. ^P < 0.05 as compared with KO vehicle. N = 3 per group.
Fig. 6.Caspase-3 levels in biliverdin reductase A knockout (BVRA KO MCT) cells with palmitic acid treatment. A. Western blot and densitometry of cleaved and uncleaved caspase-3 (closed arrows) and heat shock protein 90 (HSP90) in wild-type (WT) and BVRA KO MCT cells after 24-h treatment with vehicle or palmitic acid (400 μM). n = 4 per group. The uncleaved and cleaved caspase-3 were quantitated by densitometry and normalized to HSP90 in the graph below the blots. B. quantification of apoptosis using the Caspase-3/7 Green Apoptosis Assay for the IncuCyte live-cell analysis system at the 24-h end point (n = 10). C. mean caspase 3/7 ratio normalized to the vehicle over each 2-h time point for the 24-h time span (n = 10). *P < 0.05 as compared with WT palmitic acid. #P < 0.05 as compared with WT vehicle. ^P < 0.05 as compared with KO vehicle.
DISCUSSION
The role of lipids in inducing kidney injury in obese individuals has been widely reported (15, 24). Proximal tubule cells are more susceptible to lipotoxicity partly because of the high energy demand resulting in their preference for lipids as the main energy source (33). In the present study, we have shown that the loss of BVRA in an MCT resulted in increased intracellular lipid accumulation and lower metabolic potential. We recently demonstrated the important role of hepatic BVRA in protecting the liver against steatosis using a mouse model of hepatocyte-specific deletion of BVRA (17). Thus, the increased level of lipid accumulation in BVRA KO mouse proximal tubule cells in the current study is consistent with an antisteatotic, protective role of BVRA. High levels of intracellular lipid accumulation cause lipotoxicity in which fatty acids enter the cell either through passive diffusion via membrane-bound fatty acid transport proteins including CD36 and fatty acid transport proteins (1, 6, 21). Long chain fatty acids enter the cell by CD36 (22, 43). CD36 allows for the buildup of intracellular fatty acids which can cause lipotoxicity, and targeting CD36 may be useful in the treatment of renal disease (45). The loss of BVRA in cultured mouse proximal tubule cells caused a 6-fold increase in Cd36 mRNA levels, which may be related to the greater accumulation of lipids in the BVRA KO cells when treated with palmitic acid. Our results also reveal a shift in the metabolic profile highlighting BVRA’s role in sustaining metabolic capacity and energetic potential in proximal tubule cells.
Like several other long-chain saturated fatty acids, palmitic acid is known to induce apoptosis in several kidney cell types (24, 27, 34). Our results support increased susceptibility of BVRA KO MCT cells to cytotoxicity, apoptosis, and cell death from palmitic acid. BVRA may function to protect the cell by the production of BR. These functions include the reductase activity by which BV from heme breakdown is reduced to BR in the cells which can function as an antioxidant and promote cell survival (10, 31, 32). We have shown that mice with hyperbilirubinemia had reduced hepatic fat accumulation (18, 19). In this study, we showed that BR significantly reduced lipid accumulation in both cell types. Our previous studies in adipocytes showed that BV-derived BR reduces lipid accumulation through its activation of peroxisome proliferator-activated receptor α (PPARα) (41). We have also shown that the hepatocyte-specific BVRA KO mice on a high-fat diet develop severe hepatic steatosis and have lower PPARα expression and target genes (17). Conversely, mice with hyperbilirubinemia because of a polymorphism in the hepatic UDP-glucuronosyltransferase 1-1 (UGT1A1) gene, which conjugates BR for deposal into bile and the intestine (13, 37, 42, 44), have reduced hepatic steatosis (18). BVRA generation of BR plays an integral part in the regulation of hepatic fat accumulation in the liver as well as in other tissues (13, 17–19, 37, 41). Our study in MCT shows that the BVRA is a mediator of lipid toxicity, which may also be due to lower PPARα expression. This low level of PPARα may account for the lack of a robust effect of BR on lipid accumulation in these studies. However, this is yet to be tested. Our results suggest that the increased sensitivity to palmitic acid-induced toxicity in BVRA KO MCT cells may depend on other functions of BVRA than on its reductase activity, which may be through its regulation of intracellular signaling. However, this issue needs to be resolved using appropriate in vivo models in which BR or BVRA levels in the kidney can be specifically altered.
Lipotoxicity enhances NGAL1 levels and apoptosis of the cells, which was significantly higher in the BVRA KO MCT cells. One of the principal mechanisms by which palmitic acid induces lipoapoptosis in different cell types is by the induction of endoplasmic reticulum (ER) stress. ER stress feeds into apoptosis and cell death by inhibiting several survival-signaling pathways while activating cell death and apoptotic pathways (24). One of the survival pathways inhibited by palmitic acid-induced ER stress is phosphorylation of BAD (40). We showed that the BVRA KO cells had increased palmitic acid-induced lipotoxicity and phosphorylation of BAD. When BAD is phosphorylated, it is sequestered into the cytosol by the protein 14-3-3. This sequestration prevents it from inducing apoptosis where it inhibits another BCL2 family member, BCL-XL, a membrane-bound antagonist of apoptosis (46). Apoptotic signaling was shown to be enhanced by palmitic acid in the BVRA KO cells by increased caspase-3 cleavage by immunoblotting and by the caspase 3/7 IncuCyte data. These data, therefore, suggest that the increased sensitivity of BVRA KO cells to lipid-induced apoptosis is most likely mediated by the activation of BAD to cause cleavage of caspase-3 and apoptosis (Fig. 7).
Fig. 7.Mechanism by which biliverdin reductase A (BVRA) protects proximal tubule cells from lipoapoptosis. Saturated fatty acids (e.g., palmitic acid) reduce phosphorylation of BCL-2-associated death promoter (BAD) at Ser112 and Ser136 (pBAD) to initiate lipotoxicity-induced apoptosis by cleavage of caspase-3. BVRA prevents lipotoxicity-induced BAD activation and cellular toxicity. The loss of BVRA in proximal tubule cells increases BAD, cleavage of caspase-3, and apoptosis.
In conclusion, this study developed the first gene disruption/deletion of the mouse Blvra gene using two different gRNAs with the CRISPR-Cas9 system (3). The results show that BVRA protects mouse proximal tubule cells from palmitic acid-induced lipotoxicity and increases the energy state of the cell. BVRA is part of a critical survival pathway which protects from injury and lipotoxicity in kidney proximal tubule cells. Further studies to determine the importance of BVRA in protecting proximal tubule cells in vivo from lipotoxic injury are needed. Activating BVRA in the proximal tubule may be a novel therapeutic target to protect the kidney against obesity-induced renal damage.
GRANTS
This work was supported by grants from the National Heart, Lung and Blood Institute [K01HL-125445 (to T. D. Hinds), RO1HL088421 (to D. E. Stec), PO1HL-051971 (to J. E. Hall), and 1T32HL105324 (to S. O. Adeosun)] and the National Institute of General Medical Sciences [P20GM-104357 (to J. E. Hall)].
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
AUTHOR CONTRIBUTIONS
S.O.A., J.E.H., T.D.H., and D.E.S. conceived and designed research; D.M.G., M.F.W., K.H.M., T.D.H., and D.E.S. performed experiments; S.O.A., D.M.G., M.F.W., K.H.M., T.D.H., and D.E.S. analyzed data; S.O.A., D.M.G., M.F.W., K.H.M., J.E.H., T.D.H., and D.E.S. interpreted results of experiments; S.O.A., T.D.H., and D.E.S. prepared figures; S.O.A., T.D.H., and D.E.S. drafted manuscript; S.O.A., D.M.G., M.F.W., K.H.M., J.E.H., T.D.H., and D.E.S. edited and revised manuscript; S.O.A., D.M.G., M.F.W., K.H.M., J.E.H., T.D.H., and D.E.S. approved final version of manuscript.
ACKNOWLEDGMENTS
The authors thank Dr. Atsushi Miyawaki (RIKEN Brain Science Institute, Saitama, Japan) for the gift of the UNaG plasmid. We also thank the Flow Cytometry Core of the Univ. of Mississippi Medical Center Cancer Institute and Dr. Sibali Bandyopadhyay for helping with the flow cytometry analysis and FACS sorting.
REFERENCES
- 1. . Membrane proteins implicated in long-chain fatty acid uptake by mammalian cells: CD36, FATP and FABPm. Biochim Biophys Acta 1441: 4–13, 1999. doi:10.1016/S1388-1981(99)00137-7.
Crossref | PubMed | Web of Science | Google Scholar - 2. . A novel fluorescence-based assay for the measurement of biliverdin reductase activity. React Oxyg Species (Apex) 5: 35–45, 2018. doi:10.20455/ros.2018.809.
Crossref | PubMed | Google Scholar - 3. . Generation of genomic deletions in mammalian cell lines via CRISPR/Cas9. J Vis Exp: e52118, 2015. doi:10.3791/52118.
Crossref | PubMed | Web of Science | Google Scholar - 4. . Body fat percentage is a major determinant of total bilirubin independently of UGT1A1*28 polymorphism in young obese. PLoS One 9: e98467, 2014. doi:10.1371/journal.pone.0098467.
Crossref | PubMed | Web of Science | Google Scholar - 5. . Neutrophil gelatinase-associated lipocalin in patients with autosomal-dominant polycystic kidney disease. Am J Nephrol 27: 373–378, 2007. doi:10.1159/000103912.
Crossref | PubMed | Web of Science | Google Scholar - 6. . The molecular pathogenic role of inflammatory stress in dysregulation of lipid homeostasis and hepatic steatosis. Genes Dis 1: 106–112, 2014. doi:10.1016/j.gendis.2014.07.007.
Crossref | Google Scholar - 7. . Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91: 231–241, 1997. doi:10.1016/S0092-8674(00)80405-5.
Crossref | PubMed | Web of Science | Google Scholar - 8. . Fatty kidney: emerging role of ectopic lipid in obesity-related renal disease. Lancet Diabetes Endocrinol 2: 417–426, 2014. doi:10.1016/S2213-8587(14)70065-8.
Crossref | PubMed | Web of Science | Google Scholar - 9. . Interleukin-3-induced phosphorylation of BAD through the protein kinase Akt. Science 278: 687–689, 1997. doi:10.1126/science.278.5338.687.
Crossref | PubMed | Web of Science | Google Scholar - 10. . Overexpression of biliverdin reductase enhances resistance to chemotherapeutics. Cancer Lett 300: 40–47, 2011. doi:10.1016/j.canlet.2010.09.003.
Crossref | PubMed | Web of Science | Google Scholar - 11. . Cardiovascular disease, chronic kidney disease, and diabetes mortality burden of cardiometabolic risk factors from 1980 to 2010: a comparative risk assessment. Lancet Diabetes Endocrinol 2: 634–647, 2014. doi:10.1016/S2213-8587(14)70102-0.
Crossref | PubMed | Web of Science | Google Scholar - 12. . Obesity, hypertension, and chronic kidney disease. Int J Nephrol Renovasc Dis 7: 75–88, 2014. doi:10.2147/IJNRD.S39739.
Crossref | PubMed | Google Scholar - 13. . Bilirubin in the liver-gut signaling axis. Trends Endocrinol Metab 29: 140–150, 2018. doi:10.1016/j.tem.2018.01.002.
Crossref | PubMed | Web of Science | Google Scholar - 14. . Characterization of a renal tubular epithelial cell line which secretes the autologous target antigen of autoimmune experimental interstitial nephritis. J Cell Biol 107: 1359–1368, 1988. doi:10.1083/jcb.107.4.1359.
Crossref | PubMed | Web of Science | Google Scholar - 15. . Altered renal lipid metabolism and renal lipid accumulation in human diabetic nephropathy. J Lipid Res 55: 561–572, 2014. doi:10.1194/jlr.P040501.
Crossref | PubMed | Web of Science | Google Scholar - 16. . Cell-to-cell heterogeneity in lipid droplets suggests a mechanism to reduce lipotoxicity. Curr Biol 23: 1489–1496, 2013. doi:10.1016/j.cub.2013.06.032.
Crossref | PubMed | Web of Science | Google Scholar - 17. . Biliverdin reductase A attenuates hepatic steatosis by inhibition of glycogen synthase kinase (GSK) 3beta phosphorylation of serine 73 of peroxisome proliferator-activated receptor (PPAR) alpha. J Biol Chem 291: 25179–25191, 2016. doi:10.1074/jbc.M116.731703.
Crossref | PubMed | Web of Science | Google Scholar - 18. . Mice with hyperbilirubinemia due to Gilbert’s syndrome polymorphism are resistant to hepatic steatosis by decreased serine 73 phosphorylation of PPARα. Am J Physiol Endocrinol Metab 312: E244–E252, 2017. doi:10.1152/ajpendo.00396.2016.
Link | Web of Science | Google Scholar - 19. . Increased HO-1 levels ameliorate fatty liver development through a reduction of heme and recruitment of FGF21. Obesity (Silver Spring) 22: 705–712, 2014. doi:10.1002/oby.20559.
Crossref | PubMed | Web of Science | Google Scholar - 20. . Serum bilirubin levels are lower in overweight asymptomatic middle-aged adults: an early indicator of metabolic syndrome? Metabolism 62: 976–985, 2013. doi:10.1016/j.metabol.2013.01.011.
Crossref | PubMed | Web of Science | Google Scholar - 21. . Fatty acid regulation of hepatic gene transcription. J Nutr 135: 2503–2506, 2005. doi:10.1093/jn/135.11.2503.
Crossref | PubMed | Web of Science | Google Scholar - 22. . Defective fatty acid oxidation in renal tubular epithelial cells has a key role in kidney fibrosis development. Nat Med 21: 37–46, 2015. doi:10.1038/nm.3762.
Crossref | PubMed | Web of Science | Google Scholar - 23. . Relationship between serum bilirubin levels and metabolic syndrome in patients with schizophrenia spectrum disorders. Clin Psychopharmacol Neurosci 15: 153–162, 2017. doi:10.9758/cpn.2017.15.2.153.
Crossref | PubMed | Web of Science | Google Scholar - 24. . Lipotoxicity in renal proximal tubular cells: relationship between endoplasmic reticulum stress and oxidative stress pathways. Free Radic Biol Med 48: 1654–1662, 2010. doi:10.1016/j.freeradbiomed.2010.03.021.
Crossref | PubMed | Web of Science | Google Scholar - 25. . Unconjugated hyperbilirubinemia in patients with non-alcoholic fatty liver disease: a favorable endogenous response. Clin Biochem 45: 272–274, 2012. doi:10.1016/j.clinbiochem.2011.11.017.
Crossref | PubMed | Web of Science | Google Scholar - 26. . Serum bilirubin levels are inversely associated with nonalcoholic fatty liver disease. Clin Mol Hepatol 18: 383–390, 2012. doi:10.3350/cmh.2012.18.4.383.
Crossref | PubMed | Google Scholar - 27. . Intrarenal renin-angiotensin system mediates fatty acid-induced ER stress in the kidney. Am J Physiol Renal Physiol 310: F351–F363, 2016. doi:10.1152/ajprenal.00223.2015.
Link | Web of Science | Google Scholar - 28. . Variants in the UGT1A1 gene and the risk of pediatric nonalcoholic fatty liver disease. Pediatrics 124: e1221–e1227, 2009. doi:10.1542/peds.2008-3087.
Crossref | PubMed | Web of Science | Google Scholar - 29. . Wogonin ameliorates lipotoxicity-induced apoptosis of cultured vascular smooth muscle cells via interfering with DAG-PKC pathway. Acta Pharmacol Sin 32: 1475–1482, 2011. doi:10.1038/aps.2011.120.
Crossref | PubMed | Web of Science | Google Scholar - 30. . Association of age and BMI with kidney function and mortality: a cohort study. Lancet Diabetes Endocrinol 3: 704–714, 2015. doi:10.1016/S2213-8587(15)00128-X.
Crossref | PubMed | Web of Science | Google Scholar - 31. . Biliverdin reductase: PKC interaction at the cross-talk of MAPK and PI3K signaling pathways. Antioxid Redox Signal 9: 2187–2195, 2007. doi:10.1089/ars.2007.1805.
Crossref | PubMed | Web of Science | Google Scholar - 32. . New insights into biliverdin reductase functions: linking heme metabolism to cell signaling. Physiology (Bethesda) 20: 382–389, 2005. doi:10.1152/physiol.00029.2005.
Link | Web of Science | Google Scholar - 33. . Metabolic substrates, cellular energy production, and the regulation of proximal tubular transport. Annu Rev Physiol 47: 85–101, 1985. doi:10.1146/annurev.ph.47.030185.000505.
Crossref | PubMed | Web of Science | Google Scholar - 34. . Renal lipotoxicity-associated inflammation and insulin resistance affects actin cytoskeleton organization in podocytes. PLoS One 10: e0142291, 2015. doi:10.1371/journal.pone.0142291.
Crossref | PubMed | Web of Science | Google Scholar - 35. . Identification of neutrophil gelatinase-associated lipocalin as a novel early urinary biomarker for ischemic renal injury. J Am Soc Nephrol 14: 2534–2543, 2003. doi:10.1097/01.ASN.0000088027.54400.C6.
Crossref | PubMed | Web of Science | Google Scholar - 36. . Neutrophil gelatinase-associated lipocalin: a novel early urinary biomarker for cisplatin nephrotoxicity. Am J Nephrol 24: 307–315, 2004. doi:10.1159/000078452.
Crossref | PubMed | Web of Science | Google Scholar - 37. . Biliverdin reductase isozymes in metabolism. Trends Endocrinol Metab 26: 212–220, 2015. doi:10.1016/j.tem.2015.02.001.
Crossref | PubMed | Web of Science | Google Scholar - 38. . Serum bilirubin level is inversely associated with nonalcoholic steatohepatitis in children. J Pediatr Gastroenterol Nutr 57: 114–118, 2013. doi:10.1097/MPG.0b013e318291fefe.
Crossref | PubMed | Web of Science | Google Scholar - 39. . Some physical and immunological properties of ox kidney biliverdin reductase. Biochem J 255: 431–435, 1988. doi:10.1042/bj2550431.
Crossref | PubMed | Web of Science | Google Scholar - 40. . Oleate and eicosapentaenoic acid attenuate palmitate-induced inflammation and apoptosis in renal proximal tubular cell. Biochem Biophys Res Commun 402: 265–271, 2010. doi:10.1016/j.bbrc.2010.10.012.
Crossref | PubMed | Web of Science | Google Scholar - 41. . Bilirubin binding to PPARα inhibits lipid accumulation. PLoS One 11: e0153427, 2016. doi:10.1371/journal.pone.0153427.
Crossref | PubMed | Web of Science | Google Scholar - 42. . Glucuronidation and UGT isozymes in bladder: new targets for the treatment of uroepithelial carcinomas? Oncotarget 8: 3640–3648, 2017. doi:10.18632/oncotarget.12277.
Crossref | PubMed | Google Scholar - 43. . Multiple metabolic hits converge on CD36 as novel mediator of tubular epithelial apoptosis in diabetic nephropathy. PLoS Med 2: e45, 2005. doi:10.1371/journal.pmed.0020045.
Crossref | PubMed | Web of Science | Google Scholar - 44. . Biliverdin reductase and bilirubin in hepatic disease. Am J Physiol Gastrointest Liver Physiol 314: G668–G676, 2018. doi:10.1152/ajpgi.00026.2018.
Link | Web of Science | Google Scholar - 45. . Targeting the fatty acid transport protein CD36, a class B scavenger receptor, in the treatment of renal disease. Kidney Int 89: 740–742, 2016. doi:10.1016/j.kint.2016.01.009.
Crossref | PubMed | Web of Science | Google Scholar - 46. . Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14-3-3 not BCL-X(L). Cell 87: 619–628, 1996. doi:10.1016/S0092-8674(00)81382-3.
Crossref | PubMed | Web of Science | Google Scholar
