Regulation of Renal Fatty Acid and Cholesterol Metabolism, Inflammation, and Fibrosis in Akita and OVE26 Mice With Type 1 Diabetes

Male C57BL/6-Ins2^Akita (Akita) mice in the C57BL/6 background and C57BL/6 control mice were obtained from The Jackson Laboratories (Bar Harbor, ME). Ins2^Akita is a model of type 1 diabetes. The Akita spontaneous mutation is an autosomal dominant mutation in the insulin II gene (Ins2) (15,16). This missense mutation results in an amino acid substitution (cysteine 96 to tyrosine), which corresponds to the seventh amino acid position of the insulin II A chain. The mice were maintained on a 12-h light/dark cycle and fed standard mouse chow (Mouse Chow 5015; Ralston Purina, St. Louis, MO) ad libitum for 3 months until they were killed. The animal studies and the protocols were approved by the institutional review boards at the Denver Veterans Medical Center and the University of Colorado Health Sciences Center.

At the end of the 3 months, spot urine samples were obtained so that albumin and creatinine levels could be measured. In one group of animals (n = 8 in each experimental group), after anesthesia (intraperitoneal injection of pentobarbital; Abbott Laboratories, Chicago, IL), blood was drawn from the inferior vena cava, and then the kidneys were removed and processed for biochemical studies that included 1) RNA extraction and real-time quantitative PCR, 2) isolation of nuclei and microsomes and Western blotting, 3) measurement of stearoyl CoA desaturase-1 (SCD-1) enzyme activity, and 4) lipid extraction and lipid composition analysis. In another group of animals (n = 4 in each experimental group), after anesthesia, the kidneys were fixed via in vivo perfusion fixation and the kidneys were processed for 1) periodic acid Schiff (PAS) stain for renal histopathology, 2) oil red O stain for neutral lipids, and 3) immunofluorescence microscopy for extracellular matrix proteins (5,6,12,18,19).

Parallel studies were also performed in the OVE26 model of type 1 diabetes. These transgenic mice express a chicken calmodulin minigene controlled by the rat insulin II promoter and develop hyperglycemia within 24 h of age because of decreased pancreatic insulin (17). Male OVE26 mice on the FVB background and control FVB mice were obtained from Dr. Paul Epstein at the University of Louisville, Kentucky, and from The Jackson Laboratories.

Urine albumin concentration was determined by competitive enzyme-linked immunosorbent assay using an Albuwell M kit (Exocell, Philadelphia, PA). Urine creatinine concentration was determined by Jaffe’s reaction of alkaline picrate with creatinine using a Creatinine Companion kit (Exocell). Results were expressed as the urine albumin-to-creatinine ratio (micrograms per milligram) (5,6,12,18,19).

Total RNA was extracted according to the Trizol protocol (Invitrogen Life Technologies, Carlsbad, CA). Total RNA (2 μg) was subjected to DNase digestion followed by cDNA synthesis using the iScript cDNA Synthesis kit (Bio-Rad, Richmond, CA). We used iQSYBR Green Supermix (Bio-Rad) for real-time PCR according to the manufacturer’s instructions. Cyclophilin was used as internal control, and the amount of RNA was calculated by the comparative C_T method. All data were calculated from duplicate reactions, as we have previously described (5,6,12,18,19). Primers used are listed in Table 1.

Kidneys were homogenized at 4°C in a buffer (20 mmol/l Tris-Cl, pH 7.4, 75 mmol/l NaCl, 2 mmol/l EGTA, 2 mmol/l EDTA, 1 mmol/l Na₃VO₄, and 1 mmol/l dithiothreitol), supplemented with a protease inhibitor cocktail that consisted of 104 mmol/l AEBSF, 0.08 mmol/l aprotinin, 2 mmol/l leupeptin, 4 mmol/l bestatin, 1.5 mmol/l pepstatin A, and 1.4 mmol/l E-64 (Sigma-Aldrich, St. Louis, MO). Nuclear extracts and microsomes were prepared as we have previously described (5,6,12,18,19).

Equal amounts of protein samples were subjected to SDS-PAGE (10% wt/vol), and they were then transferred to nitrocellulose membranes. After blockage with 5% fat-free milk powder with 1% Triton X-100 in Tris-buffered saline (20 mmol/l Tris-Cl, 150 mmol/l NaCl, pH 7.4), the blots were incubated with antibodies against SREBP-1 (Santa Cruz Biotechnology, Santa Cruz, CA; 1:1,000), SREBP-2 (Santa Cruz Biotechnology; 1:1,000), and SCD-1 (Santa Cruz Biotechnology; 1:1,000). The SREBP-1 antibody recognizes both SREBP-1a and -1c. Corresponding secondary antibodies were visualized using enhanced chemiluminescence (Pierce, Bradford, IL). The signals were quantified with a Phosphor Imager with chemiluminescence detector and the accompanying densitometry software (Bio-Rad).

SCD enzyme activity was measured in microsomes prepared from individual kidney cortex extracts by determining the conversion of [1-¹⁴C] stearoyl-CoA to [1-¹⁴C] oleate (20).

Lipids from the renal cortex were extracted by the method of Bligh and Dyer, as we have previously described (5,6,12,18,19). Triglyceride and cholesterol content was measured using kits from Sigma according to instructions of the manufacturer.

Mice were anesthetized and perfused at a pressure of 180 mmHg through the abdominal aorta, as previously described (5,6,12,18,19).

Paraffin sections were stained for PAS. Frozen sections were used for oil red O staining to determine the renal accumulation of neutral fats (5,6,12,18,19). The stained kidney sections were imaged with an Olympus microscope. Immunofluorescence microscopy for type IV collagen and fibronectin were performed as previously described (5,6,12,18,19). The kidney sections were then imaged with a laser scanning confocal microscope (Zeiss LSM 510).

SPSS 11.0 for Windows was used for statistical analysis. The results were expressed as the means ± SE. The statistical significance of differences was assessed by unpaired t test.

In Ins2^Akita mice, PAS staining clearly reveals mesangial expansion, increased matrix protein accumulation, and tubulointerstitial fibrosis when compared with age- and sex-matched control mice (Fig. 1A). Immunofluorescence microscopy with anti-fibronectin (results not shown) and anti-type IV collagen (Fig. 1B) antibodies indicates increased intensity of immunofluorescence in the glomeruli and tubulointerstitial cells in Ins2^Akita mice, indicating accumulation of extracellular matrix proteins. Ins2^Akita mice have significant decreases in the mRNA abundance of the podocyte marker proteins synaptopodin and podocin (Table 2). These changes are associated with a significant increase in the urine albumin-to-creatinine ratio in Ins2^Akita when compared with control mice (Table 2).

The glomerulosclerosis, tubulointerstitial fibrosis, and proteinuria in Ins2^Akita mice are associated with significant increases in renal cortical mRNA abundance of 1) fibrosis-inducing growth factors TGF-β and PAI-1, 2) glomerular permeability–inducing growth factor VEGF, and 3) proinflammatory cytokine tumor necrosis factor (TNF)-α (Table 2).

In contrast to minimal or absent oil red O staining in the kidneys of the control mice, there is very strong staining of oil red O in the glomeruli of Ins2^Akita mice, which indicates accumulation of neutral lipid deposits (Fig. 2A). Biochemical analysis of lipid composition of the renal cortex indicates that the accumulation of neutral lipid deposits corresponds to significant increases in triglyceride and cholesterol content (Fig. 2B).

Ins2^Akita mice have a significant increase in SREBP-1 nuclear protein abundance, a nuclear receptor/transcription factor that is a master regulator of fatty acid and triglyceride synthesis (Table 3). To determine the functional significance of the increase in SREBP-1 protein, we determined the mRNA, protein, and enzyme activity of SCD-1, a key enzyme in regulation of fatty acid and triglyceride synthesis. We found that Ins2^Akita mice have significant increases in SCD-1 mRNA, SCD-1 protein, and SCD-1 enzyme activity (Table 3).

Interestingly, the expression of FXR, a nuclear receptor that is an important negative regulator of SREBP-1c and TGF-β, and its major target small heterodimer partner (SHP) were significantly decreased in the kidneys of Ins2^Akita mice (Table 3).

Ins2^Akita mice also have a significant increase in SREBP-2 nuclear protein abundance, a nuclear receptor/transcription factor that is a master regulator of cholesterol synthesis (Table 3). In addition, Ins2^Akita mice have a significant decrease in ABCA-1 mRNA abundance, a key enzyme that mediates cholesterol efflux (Table 3).

Interestingly, the expression of LXR-α and LXR-β, nuclear receptors that are important positive regulators of ABCA-1 and negative regulators of proinflammatory cytokines, were significantly reduced in the kidneys of Ins2^Akita mice (Table 3).

Previous studies determined that OVE26 mice on the FVB genetic background develop severe glomerulosclerosis and proteinuria (17). In our study, we confirmed these finding (results not shown), and we have also determined that there is a marked increase in the mRNA abundance for type IV collagen and fibronectin (Table 4), which suggests that the increased expression and accumulation of the extracellular matrix proteins are transcriptionally regulated. In addition, these mice also have significant decreases in the mRNA abundance for the podocyte marker proteins podocin and synaptopodin (Table 4).

OVE26 mice have significant increases in renal cortical mRNA abundance of 1) the proinflammatory cytokines TNF-α and interleukin (IL)-6 (Table 4), 2) markers of oxidative stress, including components of the NADPH oxidase system (Nox4, p22phox, and p47phox) (Table 4), and 3) receptor for advanced glycation end products (Table 4).

OVE26 mice have significant increases in mRNA abundance of SREBP-1c target enzymes acetyl-CoA carboxylase (ACC) and fatty acid synthase (FAS) (Table 5), enzymes that initiate the cascade of fatty acid synthesis. In addition, there is also marked upregulation of ChREBP and its major target enzyme, liver pyruvate kinase (L-PK) (Table 5), which, in concert with the SREBP-1 pathway, mediates fatty acid and triglyceride synthesis.

The increases in SREBP-1–and ChREBP-dependent fatty acid synthesis pathways are paralleled by marked decreases in the expression of PPAR-α, PPAR-δ, and their target enzyme acyl-CoA oxidase (ACO) (Table 5), which mediate fatty acid oxidation. These marked increases in fatty acid synthesis and decreased fatty acid oxidation are paralleled by a marked decrease in FXR expression (Table 5), a nuclear receptor that has been shown in the liver to be a negative regulator of SREBP-1 and ChREBP and a positive regulator of PPAR-α.

OVE26 mice also have a significant increase in SREBP-2 mRNA abundance, a nuclear receptor and transcription factor that is a master regulator of cholesterol synthesis (Table 5) and its major target enzyme HMG-CoA reductase (Table 5), which is a key enzyme in cholesterol synthesis.

Interestingly, the expression of LXR-α and LXR-β, nuclear receptors that are important positive regulators of ABCA-1 and negative regulators of proinflammatory cytokines, were significantly reduced in the kidneys of OVE26 mice (Table 5).

Well-known changes in diabetes-related renal function and structure include mesangial expansion and podocyte injury and loss, which result in glomerulosclerosis, proteinuria, and a decline in glomerular filtration rate (21–29). Several hormonal and metabolic factors, including profibrotic growth factors (angiotensin II, TGF-β, and VEGF), proinflammatory cytokines (TNF-α and IL-6), oxidative stress, and advanced glycation end products have been shown to modulate diabetes-related renal disease (30,31). Renal accumulation of lipids has also been thought to play a role in the pathogenesis of diabetic nephropathy (1–6). Our previous study of streptozotocin-induced type 1 diabetes in the rat showed that increased lipid synthesis mediated by increased activity of SREBP-1 plays an important role in the development of diabetic nephropathy (5). Furthermore, we were able to demonstrate that increased renal expression of SREBP-1a per se is the key factor linking increased fatty acid synthesis and accumulation of lipids to development of nephropathy (5).

In this study of two different genetic mouse models of type 1 diabetes, we also demonstrate that the glomerulosclerosis and proteinuria are associated with increased triglyceride and cholesterol accumulation.

A novel finding of our study is that we demonstrate that in addition to increased SREBP-1 activity, increased ChREBP activity most likely plays a role in increased fatty acid synthesis and accumulation of triglycerides. ChREBP has previously been shown to play an important role in triglyceride synthesis in the liver (7–9) and adipose tissue (32). ChREBP mediates insulin-independent, glucose-stimulated gene expression of multiple liver enzymes that are responsible for converting excess carbohydrate to fatty acids for long-term storage (7–9). This is the first demonstration of increased expression of ChREBP and L-PK in diabetic kidneys, which, along with SREBP-1c, may play an important role in mediating increased fatty acid synthesis in the kidney.

In addition to increased fatty acid synthesis induced by coordinate upregulation of SREBP-1 and ChREBP, our study also demonstrates that decreased fatty acid oxidation mediated by decreased expression and activity of PPAR-α, PPAR-δ, and ACO also may play a role in the eventual triglyceride accumulation (33,34). Interestingly, a recent study found accelerated diabetic nephropathy in mice lacking PPAR-α (35).

Another novel finding of our study is that we have found decreased expression of FXR and its immediate target (SHP) in the kidneys of the Akita and OVE26 mice. FXR is highly expressed in liver, kidney, intestine, colon, and adrenal glands. Similar to PPARs, FXR binds DNA as an obligate heterodimer with the retinoid X receptor (RXR) (10,11). The regulation of FXR in the kidney, however, has not been determined.

In the liver, FXR has been shown to play an important role in regulation of triglyceride metabolism by decreasing the expression of SREBP-1 and ChREBP and increasing the expression of PPAR-α (10,11). Indeed, FXR knockout mice have increased liver triglyceride content (36). Our results, therefore, indicate that the decrease in FXR and its major target, SHP, could mediate the increases in SREBP-1 and ChREBP and decreases in PPAR-α and -δ, which result in increased fatty acid and triglyceride synthesis and accumulation of triglyceride in the kidney.

Recent studies indicate that FXR agonists also modulate fibrosis. In rat models of extrahepatic and intrahepatic cholestasis, the FXR agonist GW4064 protects against cholestatic liver damage in part by decreasing the expression of TGF-β (37). In additional models of liver cirrhosis and fibrosis, treatment with the FXR agonist 6-ethyl chenodeoxycholic acid causes decreased expression of 1) TGF-β1, 2) tissue metalloproteinase inhibitor-1 and -2, 3) α1 (I) collagen, 4) α smooth muscle actin, and 5) increased expression and activity of matrix metalloproteinase-2. These effects result in resolution of hepatic fibrosis (38). Our results therefore also indicate that the decrease in FXR and its major target (SHP) could also mediate the increased expression of TGF-β in the kidney, resulting in podocyte loss, accumulation of extracellular matrix proteins, renal fibrosis, and proteinuria.

In addition to increased triglyceride content, we have also found a significant increase in renal cholesterol content. The cholesterol accumulation is mediated by SREBP-2–and HMG-CoA reductase–induced increases in cholesterol synthesis. At the same time, we also found evidence for a potential decrease in cholesterol efflux mediated by decreased expression of ABCA-1.

An additional novel finding of this study is that in addition to the regulation of SREBP-1, ChREBP, PPAR-α, PPAR-δ, FXR, and SREBP-2, we have also found significant modulation of LXR-α and LXR-β in the kidneys of the Akita and OVE26 mice.

LXRs are members of the ligand-activated nuclear receptor family of transcription factors, and they function as intracellular sensors of sterols (11,39). LXRs regulate intestinal uptake of dietary cholesterol, cellular efflux of cholesterol in peripheral tissues to HDL, and its eventual excretion from the body via the bile. In response to binding their sterol ligands, LXRs stimulate ABCA-1–mediated transport of cholesterol and phospholipids out of the cell interior to apolipoprotein A1–containing HDL. Hepatic expression of LXR promotes excretion of cholesterol from the body via two processes: 1) direct transport of cholesterol into biliary canaliculi by way of the ABC transporter proteins ABCG5 and ABCG8 and 2) upregulation of cytochrome P450 7α-hydroxylase (CYP7A1), which converts cholesterol to bile acids.

In addition to their roles in cholesterol metabolism, LXRs exert important regulatory effects on triglyceride metabolism. These effects are largely caused by their coordinate regulation of SREBP-1c, the master regulator of triglyceride and fatty acid synthesis. LXR transcriptional response elements are present in the promoter of SREBP-1c; however, the net effect on hepatic and serum triglycerides has been variable after treatment with different synthetic LXR agonists (40–42).

The decrease in LXR expression in the diabetic kidney may mediate the decreased expression of ABCA-1 and result in decreased cholesterol efflux. A significant increase in cholesterol accumulation in the kidney is the net result of increased cholesterol synthesis mediated by increased expression of SREBP-2 and HMG-CoA reductase and decreased cholesterol efflux mediated by decreased expression of LXR-α, LXR-β, and ABCA-1.

As modulators of inflammatory cytokines, LXRs may be important in diabetic kidney disease (39). LXRs and their ligands are negative regulators of macrophage inflammatory gene expression. In activated macrophages, LXRs repress expression of the proinflammatory genes IL-6, IL-1β, cyclooxygenase-2, granulocyte colony–stimulating factor, inducible nitric oxide synthase, and matrix metalloproteinase-9, as well as the chemokines monocyte chemoattractant protein-1 and macrophage inflammatory protein-1β. In our studies, we found increased expression of TNF-α and IL-6 in diabetic kidneys of OVE26 and Akita mice, which could be mediated by decreased expression of LXR.

Proinflammatory cytokines in the diabetic kidney can also induce downregulation of LXR and FXR (43,44), which would then modulate renal lipid metabolism and add to the complexity of the interactions between nuclear receptors and proinflammatory cytokines in the pathogenesis of diabetic kidney disease.

We propose that the increases in renal triglyceride and cholesterol accumulation play an important role in the pathogenesis of diabetic nephropathy. Three lines of evidence suggest that altered lipid metabolism is associated with diabetic and nondiabetic renal disease. First, there are a number of genetic abnormalities of lipid metabolism in humans and experimental animals, including Fabry’s disease (45); lecithin cholesterol acyltransferase deficiency (46); type IA glycogen storage disease (von Gierke’s disease) (47); genetic and acquired lipodystrophy (48); ABCA-1 knockout mice (a murine model of Tangiers disease) and familial HDL deficiency, with defects in ABCA-1–and HDL-mediated reverse cholesterol transport (49); and ApoE knockout mice (50), in which abnormalities in serum and tissue lipids, including renal lipid composition, are associated with glomerular disease and proteinuria. Second, increases in serum lipids have been associated with a faster decline of renal function (51). Third, there is increasing evidence that inhibition of cholesterol synthesis by HMG-CoA reductase inhibitors (statins), inhibition of triglyceride synthesis by PPAR-α agonists (fibrates) (52), or a decrease in LDL achieved by LDL apheresis (29) protect against diabetic and nondiabetic renal disease.

In summary, our findings in Akita and OVE26 mice with type 1 diabetes show renal lipid accumulation, glomerular matrix expansion and sclerosis, podocyte loss, and development of proteinuria (Fig. 3). Our lipid metabolism pathway analysis suggests a global shift in diabetic renal metabolism toward 1) increased fatty acid synthesis mediated by increases in SREBP-1, ACC, FAS, SCD-1, ChREBP, and L-PK; 2) decreased fatty acid oxidation mediated by decreases in PPAR-α, PPAR-δ, and ACO; 3) increased cholesterol synthesis mediated by increases in SREBP-2 and HMG-CoA reductase; and 4) decreased cholesterol efflux mediated by decreases in LXR-α, LXR-β, and ABCA-1. We also found increased levels of the profibrotic growth factors TGF-β, PAI-1, and VEGF; the proinflammatory cytokines IL-6 and TNF-α; oxidative stress; and receptor for advanced glycation end products. We have observed a pattern of reciprocal changes in renal lipid biosynthesis and fibrosis/inflammation/oxidative stress in diabetic kidneys of Akita and OVE26 mice, which could be explained mechanistically by the corresponding changes we found in LXR and FXR expression. This parallels similar findings in liver and macrophages by other investigators. Our findings suggest but do not prove a significant role for LXR and FXR in the development of the proinflammatory, profibrotic, and prolipogenic environment characteristic of diabetic kidney disease and warrant future investigations of synthetic agonists of FXR and LXR for their potential roles as therapeutic agents for the treatment of diabetic kidney disease.

These studies were supported by grants from the Juvenile Diabetes Research Foundation (1-2003-108), the National Institutes of Health (NIH) (R01 DK062209, R01 066029, and RO3 AG20361 to M.L.), NIH Training Grants (T32 DK-07038-29 and T32 DK-007135-32 to G.P. and Z.W.), and a National Kidney Foundation Fellowship (to G.P.).

The authors thank Dr. Paul N. Epstein for providing the OVE26 mice, Dr. Makoto Miyazaki and Dr. James M. Ntambi for helping to perform the SCD activity assays, and Carissa Gould and Julie McCollam for the illustrations.