BASIC RESEARCH
www.jasn.org
CD36 Regulates Oxidative Stress and Inflammation in Hypercholesterolemic CKD Daryl M. Okamura,* Subramaniam Pennathur,† Katie Pasichnyk,* Jesu´s M. Lo´pez-Guisa,* Sarah Collins,* Maria Febbraio,‡ Jay Heinecke,§ and Allison A. Eddy* *Department of Pediatrics, Seattle Children’s Research Institute, and §Department of Endocrinology and Metabolism, University of Washington, Seattle, Washington; †Department of Internal Medicine, University of Michigan, Ann Arbor, Michigan; and ‡Department of Cell Biology, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio
BASIC RESEARCH
ABSTRACT Scavenger receptors play a central role in atherosclerosis by processing oxidized lipoproteins and mediating their cellular effects. Recent studies suggested that the atherogenic state correlates with progression of chronic kidney disease (CKD); therefore, scavenger receptors are candidate mediators of renal fibrogenesis. Here, we investigated the role of CD36, a class B scavenger receptor, in a hypercholesterolemic model of CKD. We placed CD36-deficient mice and wild-type male mice on a high-fat Western diet for 7 to 8 wk and then performed either sham or unilateral ureteral obstruction surgery. CD36-deficient mice developed significantly less fibrosis compared with wild-type mice at days 3, 7, and 14 after obstruction. Compared with wild-type mice, CD36-deficient mice had significantly more interstitial macrophages at 7 d but not at 14 d. CD36-deficient mice exhibited reduced levels of activated NF-B and oxidative stress (assessed by measuring fatty acid– derived hydroxyoctadecadienoic acid and protein carbonyl content) and decreased accumulation of interstitial myofibroblasts compared with wild-type mice. These data suggest that CD36 is a key modulator of proinflammatory and oxidative pathways that promote fibrogenesis in CKD. J Am Soc Nephrol 20: 495–505, 2009. doi: 10.1681/ASN.2008010009
The prevalence of chronic kidney disease (CKD) is increasing at an alarming rate, with the largest increases seen in the moderate to advanced stages.1,2 Furthermore, CKD is a major risk factor for cardiovascular morbidity and mortality.3 Patients with CKD not only have a more atherogenic lipid profile but also have evidence of chronic inflammation and oxidative stress.4,5 The combination of these factors likely accounts for the accelerated atherosclerotic disease that develops in patients with advanced CKD. Several epidemiologic studies suggested that this highly atherogenic state is correlated with more rapid progression of kidney fibrosis.6 – 8 Other studies further suggested that specific lipoproteins are linked to an increased risk for renal dysfunction and progression of CKD.9,10 Studies of hypercholesterolemic animal models support the view that oxidized lipoproteins are directly involved in the J Am Soc Nephrol 20: 495–505, 2009
pathogenetic pathways that trigger inflammation and oxidative stress leading to cardiovascular disease. Much less is known about receptor-dependent pathways that perpetuate renal parenchymal inflammation and fibrosis within the dyslipidemic environment of CKD. In particular, there is a paucity of information on the renal pattern of expression and function of specific scavenger receptors that figure prominently in atherogenesis. Received January 4, 2008. Accepted October 15, 2008. Published online ahead of print. Publication date available at www.jasn.org. Correspondence: Dr. Daryl M. Okamura, Seattle Children’s Research Institute, Division of Nephrology, 4800 Sand Point Way NE, A7931, Seattle, WA 98105. Phone: 206-987-2524; Fax: 206987-2636; E-mail:
[email protected] Copyright 䊚 2009 by the American Society of Nephrology
ISSN : 1046-6673/2003-495
495
BASIC RESEARCH
www.jasn.org
Scavenger receptors comprise a family of nine classes of structurally similar receptors that share oxidized lipoproteins as their primary ligand.11 CD36 is a class B transmembrane scavenger receptor that is known to be expressed by multiple cell types, such as macrophages, microvascular endothelial cells, platelets, adipocytes, and epithelial cells.12 CD36 has been proposed to be a major endocytic receptor for oxidized lipoproteins and accounts for 50 to 60% of the uptake of oxidized LDLs (oxLDLs) by macrophages.13 Binding of oxLDLs to CD36 may also activate intracellular signaling pathways that lead to proinflammatory chemokine and cytokine production.14 –16 Peripheral blood monocyte CD36 levels are significantly higher in patients on dialysis than in normal individuals.17 In a recent study, we demonstrated the kidneys of mice with diet-induced hypercholesterolemia developed worse fibrosis after unilateral ureteral obstruction (UUO) than did normocholesterolemic mice. This difference was associated with higher renal oxidized lipoprotein levels, activation of inflammatory pathways, and a striking change in the intensity and cell-specific pattern of CD36 expression. CD36 and other scavenger receptors were predominantly expressed in tubular epithelial cells and interstitial macrophages with specific modulation of their expression pattern during chronic renal injury by obstruction.18 To gain further insight into the mechanisms that modulate fibrogenic pathways during hypercholesterolemia, we designed this study to investigate the hypothesis that CD36 promotes renal fibrosis by activating proinflammatory and oxidative pathways. We compared the severity of renal fibrosis induced by UUO in CD36 wild-type (CD36⫹/⫹) and deficient (CD36⫺/⫺) hypercholesterolemic mice.
Figure 1. Total kidney collagen content determined by hydroxyproline assay is significantly decreased in obstructed kidneys from hypercholesterolemic CD36⫺/⫺ mice compared with CD36⫹/⫹ mice. f, CD36⫹/⫹ mice (n ⫽ 6 to 7); 䡺, CD36⫺/⫺ mice (n ⫽ 8 to 10). Data are means ⫾ SEM. ‡P ⬍ 0.05, †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺.
were matched for mean body weight. The average body weights were similar between CD36⫹/⫹ and CD36⫺/⫺ mice at all time points, with the exception of day 7 (23.8 ⫾ 0.5 versus 25.7 ⫾ 0.6 g; n ⫽ 6 to 10 per group; P ⫽ 0.04). Fibrosis Severity. To investigate whether CD36 modulated the severity of renal fibrosis induced by UUO in mice with diet-induced hypercholesterolemia, we measured total collagen levels as hydroxyproline content per wet weight of kidney tissue. Fibrosis severity was significantly reduced by 19 to 29% in obstructed kidneys from hypercholesterolemic CD36⫺/⫺ mice compared with CD36⫹/⫹ mice at each time point (n ⫽ 6 to 10 per group; P ⬍ 0.01; Figure 1).
RESULTS Attenuated Renal Injury in Hypercholesterolemic CD36ⴚ/ⴚ Mice after UUO
Metabolic Effects of Diet-Induced Hypercholesterolemia. After 7 to 8 wk on a high-fat Western diet, both CD36⫹/⫹ and CD36⫺/⫺ mice developed significant hypercholesterolemia (Table 1). The mean serum cholesterol levels were similar between the CD36⫹/⫹ and CD36⫺/⫺ mice at all time points, with the exception of day 3, although this did not affect intrarenal oxidized lipoprotein levels at this time point. On the day of UUO surgery, the two groups of male mice (10 to 11 wk) Table 1. Mean cholesterol levels on high-fat diet at killinga Time Point Day Day Day Day
14 3 7 14
Surgery
CD36ⴙ/ⴙ (mg/dl) (n ⴝ 6 to 8)
CD36ⴚ/ⴚ (mg/dl) (n ⴝ 6 to 8)
Sham UUO UUO UUO
188 ⫾ 17 140 ⫾ 6b 193 ⫾ 29 191 ⫾ 15
180 ⫾ 4 215 ⫾ 6 196 ⫾ 10 172 ⫾ 11
a
Mean cholesterol levels on standard chow diet from previous experiments were 80 to 102 mg/dl.18 P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺.
b
496
Journal of the American Society of Nephrology
Tubular Epithelial Cell Damage. A limitation of the UUO model is the inability to measure functional consequences of the structural damage as a result of contralateral kidney compensation. Loss of expression of Ecadherin has been used as a surrogate marker of tubular cell damage and atrophy.19,20 Tubular integrity was significantly preserved with a 2.9-fold higher level of E-cadherin protein in obstructed kidneys at day 14 in the CD36⫺/⫺ group compared with CD36⫹/⫹ group (n ⫽ 5 to 6 per group; P ⫽ 0.002; Figure 2A). Although E-cadherin can be expressed by macrophages,21 by immunostaining, the predominant expression in the kidney during UUO is by tubular epithelial cells (Figure 2B). Biphasic Interstitial Macrophage Response in CD36ⴚ/ⴚ Mice
Macrophage Infiltration. We observed a 68% increase in F4/80⫹ interstitial area at day 7 in CD36⫺/⫺ mice compared with CD36⫹/⫹ mice despite significantly less fibrosis (n ⫽ 5 per group; F ⫽ 38, P ⫽ 0.0003; Figure 3). By day 14 after UUO, the pattern of macrophage infiltration had changed, now showing a NS decrease in the F4/80⫹ interstitial area in CD36⫺/⫺ mice compared with J Am Soc Nephrol 20: 495–505, 2009
www.jasn.org
BASIC RESEARCH
mRNA levels, a key chemokine in macrophage migration, at any of the time points investigated (Figure 4B). Among the chemokines examined, CCL5 and CXCL10 had the highest steady-state mRNA levels at the time points investigated. There was a 140% increase in steady-state CCL5 mRNA levels at day 7 after UUO in hypercholesterolemic CD36⫺/⫺ mice compared with CD36⫹/⫹ mice (n ⫽ 5 to 6 per group; P ⫽ 0.02; Figure 4C). Steady-state CXCL10 mRNA levels were significantly increased at days 7 and 14 with the largest increase (153%) at day 7 in CD36⫺/⫺ mice compared with CD36⫹/⫹ mice (n ⫽ 5 to 6 per group; P ⬍ 0.01; Figure 4D). CXCL2, a marker of activated macrophages, was significantly decreased by 33% in CD36⫺/⫺ mice compared with CD36⫹/⫹ mice at day 14 after UUO (n ⫽ 5 to 6 per group; P ⫽ 0.008; Figure 4, A and E). Attenuation of Proinflammatory Pathways in CD36ⴚ/ⴚ Mice
Figure 2. (A) Renal E-cadherin Western blot (top) illustrates the changes in E-cadherin protein levels (120 kD) on days 7 and 14 after UUO. E-cadherin expression was normalized to -actin. The graph summarizes the results of single-band density measurements, expressed as -actin–normalized E-cadherin band levels. Data are means ⫾ SEM. †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. (B) Representative E-cadherin immunohistochemistry photomicrographs demonstrate increased tubular localization of E-cadherin in CD36⫺/⫺ mice 14 d after UUO. Magnification, ⫻400.
To investigate whether the proinflammatory pathways that are activated in obstructed kidneys are modulated by CD36, we examined levels of activated NF-B at days 3 and 7, representing time points of early and established inflammation, respectively. Nuclear translocation of NF-B was initially evaluated indirectly by Western blot analysis of phosphorylated IB-␣ and total IB-␣. Phosphorylated IB-␣–to–total IB-␣ ratios were significantly decreased by 56% at day 7 in obstructed kidneys of hypercholesterolemic CD36⫺/⫺ mice compared with CD36⫹/⫹ mice; a NS 30% decrease was observed at day 3 (CD36⫹/⫹ versus CD36⫺/⫺: day 3, 0.70 ⫾ 0.20 versus 0.50 ⫾ 0.06 [P ⫽ 0.36]; day 7, 2.10 ⫾ 0.40 versus 0.90 ⫾ 0.20 [P ⫽ 0.02]; n ⫽ 6 per group; Figure 5, A and B). To confirm the
CD36⫹/⫹ mice (9.7 ⫾ 0.8 versus 11.9 ⫾ 0.7%; n ⫽ 5 per group; F ⫽ 0.88, P ⫽ 0.4). To investigate whether the increase in F4/80⫹ cells at day 7 after UUO represented an increase in renal dendritic cells, we compared kidney CD68 and CD11b/c levels in CD36⫺/⫺ mice with those of CD36⫹/⫹ mice. There was no difference in either the CD11b/c protein levels by Western blotting (CD36⫹/⫹ versus CD36⫺/⫺ 0.15 ⫾ 0.02 versus 0.19 ⫾ 0.03; n ⫽ 6 per group; P ⫽ 0.3) or the steady-state CD68 mRNA levels measured by semiquantitative real-time qPCR (normalized to 18S, relative expression CD36⫺/⫺ to CD36⫹/⫹ ⫽ 1.33; 95% confidence interval 0.50 to 5.02; P ⫽ 0.8; n ⫽ 5 per group). Differences in F4/80⫹ interstitial area between CD36⫹/⫹ and CD36⫺/⫺ mice were NS at day 3 (4.0 ⫾ 0.3 versus 3.0 ⫾ 0.6%; n ⫽ 6 per group; P ⫽ 0.5). Chemokine Expression. To investigate the effect of CD36 expression on macrophage profile in obstructed kidneys, we compared renal chemokine expression levels in CD36⫹/⫹ and CD36⫺/⫺ mice by RNAse protection assay. There was no difference in steady-state CCL2 J Am Soc Nephrol 20: 495–505, 2009
Figure 3. F4/80⫹ interstitial macrophage area is higher in CD36⫺/⫺ mice 7 d after UUO despite less fibrosis. (Top) Representative F4/80 immunohistochemistry photomicrographs (400x) of CD36⫹/⫹ and CD36⫺/⫺ mice at day 7 after UUO. (Bottom) Summary of the percentage F4/80⫹ interstitial area ⫾ SEM. †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. CD36 Regulates Oxidative Inflammation
497
BASIC RESEARCH
www.jasn.org
ceptible to peroxidation. Linoleates are the major polyunsaturated lipids in vivo and are oxidized by free radicals into hydroperoxy octadecadienoates. Total hydroxyoctadecadienoic acid (HODE) levels account for the majority of linoleate oxidation products.23 We measured kidney HODE levels and protein carbonyl content in CD36⫹/⫹ and CD36⫺/⫺ mice at day 7, representing a key UUO time point when both inflammation and oxLDL generation were significantly attenuated. HODE levels were significantly increased after UUO in mice of both genotypes compared with genotype-matched sham kidneys; on UUO Figure 4. Renal chemokine expression in CD36⫺/⫺ mice after UUO. (A) Represenday 7, HODE levels were 41% lower in the tative RNAse protection assay blot of CD36⫹/⫹ and CD36⫺/⫺ mice at day 7 after UUO. (B through E) Graphs summarize the results of band density measurements CD36⫺/⫺ kidneys (CD36⫹/⫹ versus normalized against the ribosomal protein L32 band. Data are means ⫾ SEM. ‡P ⬍ 0.05, CD36⫺/⫺, HODE pmol/mg protein: sham, 134 ⫾ 10 versus 132 ⫾ 13 [P ⫽ 0.9]; †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. UUO, 409 ⫾ 31 versus 241 ⫾ 15 [P ⫽ Western blot findings, we performed electromobility shift as- 0.0002]; n ⫽ 7 to 8 per group). Protein oxidation as measured says (EMSAs) on nuclear extracts obtained from obstructed by protein carbonyl content was also reduced by 41% in obkidneys. These studies confirmed a 52% decrease in nuclear structed kidneys from CD36⫺/⫺ mice compared with translocation of NF-B heterodimers in CD36⫺/⫺ mice at CD36⫹/⫹ mice at day 7 (CD36⫹/⫹ versus CD36⫺/⫺, proday 7 compared with CD36⫹/⫹ mice (n ⫽ 7 to 8 per group; tein carbonyl pmol/mg protein: sham, 103 ⫾ 4 versus 101 ⫾ 4 P ⫽ 0.02; Figure 5C). We performed supershift assays to deter- [P ⫽ 0.8]; UUO, 275 ⫾ 9 versus 163 ⫾ 5 [P ⫽ 0.00001]; n ⫽ 7 mine whether nuclear translocation of specific NF-B het- to 8/group). Renal protein and lipid oxidation measures erodimers was altered. These studies demonstrated a signifi- showed a strong positive correlation, evidence of an overall cant decrease in nuclear translocation of the proinflammatory reduction in the degree of oxidative stress in obstructed kidp50RelA (p50p65) heterodimer in CD36⫺/⫺ mice compared neys of hypercholesterolemic CD36⫺/⫺ mice (n ⫽ 29; r2 ⫽ with CD36⫹/⫹ mice (Figure 5D). Phosphorylated RelA im- 0.8374; F 139.1, df ⫽ 27; P ⬍ 0.01; Figure 7). munostaining confirmed decreased p50RelA nuclear staining in both tubular and interstitial cells in CD36⫺/⫺ obstructed Inflammation Attenuation in CD36ⴚ/ⴚ Mice kidneys (Figure 5, E and F). Modulates Fibrogenic Pathways TGF- Activation. We performed studies to determine whether the CD36-associAttenuation of Oxidative Pathways in CD36ⴚ/ⴚ Mice Accumulation of Oxidized Lipoproteins. ated changes in inflammatory and oxidative pathways resulted We compared renal oxLDL levels in obstructed kidneys of hy- in changes in the activity of the profibrotic growth factor percholesterolemic CD36⫹/⫹ and CD36⫺/⫺ mice to deter- TGF-. We examined phosphorylated Smad2 and total Smad2 mine whether they are modulated by CD36. Semiquantitative levels in obstructed kidneys of CD36⫹/⫹ and CD36⫺/⫺ mice analysis of kidney sections stained for hypochlorous (HOCl)- by Western blotting for each time point. There was a NS trend modified LDL showed significantly decreased levels in the toward lower ratio of phosphorylated Smad2 to total Smad2 CD36⫺/⫺ group at each time point investigated (n ⫽ 6 per levels at days 3 and 7 (n ⫽ 6 per group; P ⫽ 0.07; Figure 8). By group; P ⬍ 0.01; Figure 6). By immunostaining, less HOCl- immunostaining, there was a reduction in nuclear expression modified LDL was detected within both renal tubules and the of phosphorylated Smad2 in both tubular and interstitial cells interstitium of CD36⫺/⫺ obstructed kidneys (Figure 6). in CD36⫺/⫺ mice (data not shown). There was no difference in the ratio of phosphorylated Smad2 to total Smad2 between genoOxidative Stress. types at day 14 after UUO (CD36⫹/⫹ versus CD36⫺/⫺, day 14, Because oxidized lipoprotein accumulation was diminished, 0.07 ⫾ 0.01 versus 0.08 ⫾ 0.01; P ⫽ 0.6; n ⫽ 5 per group). the degree of oxidative stress was compared in CD36⫹/⫹ and CD36⫺/⫺ mice. Lipid, protein, and DNA may be targets of Interstitial Myofibroblast Numbers. oxidation by free radical species such as reactive oxygen during Because fibrosis severity was significantly attenuated, we perchronic renal injury. The most commonly used indicator of formed studies to examine the effect of CD36 expression on the generalized protein oxidation is protein carbonyl content.22 number of interstitial myofibroblasts. ␣-Smooth muscle actin Lipids contain polyunsaturated fatty acid moieties that are sus- (␣-SMA)-positive myofibroblasts are considered the primary 498
Journal of the American Society of Nephrology
J Am Soc Nephrol 20: 495–505, 2009
www.jasn.org
BASIC RESEARCH
7 after UUO (n ⫽ 6 per group; P ⬍ 0.01; Figure 10). By immunostaining, CXCL10 protein was identified in both tubular and interstitial cells. Attenuated Fibrosis in CD36ⴚ/ⴚ Mice Is Independent of Diet
To investigate whether the decrease in fibrosis severity in CD36⫺/⫺ mice was dependent on hypercholesterolemia, we measured total collagen levels in CD36⫺/⫺ and CD36⫹/⫹ mice on a control diet. As seen in hypercholesterolemic mice, fibrosis severity was significantly reduced in obstructed kidneys from CD36⫺/⫺ mice by 16 and 13% compared with CD36⫹/⫹ mice on a control diet at days 7 and 14 after UUO, respectively (n ⫽ 6 to 8 per group; P ⬍ 0.01; Figure 11).
DISCUSSION
Dyslipidemia not only is a risk factor for CKD progression but also independently can lead to renal dysfunction.6,9 We reFigure 5. Proinflammatory pathway attenuation in CD36⫺/⫺ mice after UUO. (A) Representative Western blot of phosphorylated IB-␣ and total IB-␣ in CD36⫹/⫹ and cently demonstrated that hypercholesterolCD36⫺/⫺ mice at day 3 and day 7 after UUO. (B) Summary of results of phosphory- emia exacerbates renal fibrosis in associalated IB-␣ to total IB-␣ ratios. (C) Representative EMSA illustrates translocated tion with increased oxLDL deposition and NF-B heterodimer levels on pooled nuclear extracts from day 7 UUO total kidney more severe inflammation.18 We now dochomogenate. Arrows indicate specific NF-B bands that were quantified for individual ument in this study in vivo profibrotic efsamples. Phorbol myristate acetate (PMA)-stimulated and unstimulated Jurkat T cells fects of the CD36 scavenger receptor in a were used as positive and negative controls, respectively. The graph summarizes the model of progressive renal disease associresults of NF-B EMSA results performed using individual mouse samples (n ⫽ 7 to 8 ated with hypercholesterolemia. The geper group; P ⫽ 0.02). (D) Representative supershift assay with p65 and p50 antibodies netic deficiency of CD36 resulted in a performed to identify NF-B heterodimers in nuclear extract isolated 7 d after UUO. (E ⬎20% reduction in renal fibrosis severity and F) Representative phosphorylated p65 immunohistochemistry photomicrographs compared with CD36⫹/⫹ mice at each ⫹ in CD36⫹/⫹ and CD36⫺/⫺ mice on day 7 after UUO. Arrows indicate p65 nuclei in tubular and interstitial cells. Data are means ⫾ SEM. ‡P ⬍ 0.05, CD36⫹/⫹ versus time point investigated after UUO. Although measurement of renal functional CD36⫺/⫺. Magnification, ⫻400. differences is not possible in the UUO source of interstitial matrix proteins that accumulate during model, the major structural consequence of interstitial fibrosis renal fibrosis. There was a ⬎30% reduction in the mean num- that determines functional decline is tubular damage and deber of interstitial ␣-SMA–positive myofibroblasts in ob- struction.25 Tubular injury was significantly less in hyperchostructed kidneys of hypercholesterolemic CD36⫺/⫺ mice at lesterolemic CD36⫺/⫺ mice 14 d after UUO. The results of days 7 and 14 compared with CD36⫹/⫹ mice (n ⫽ 6 per this study further suggest three important features of the group; P ⬍ 0.01; Figure 9). pathologic mechanisms that contribute to CKD. First, CD36 expression by interstitial macrophages determines an activated CXCL10 Expression. macrophage phenotype that is profibrotic. Second, CD36 conCXCL10 is a chemokine that has been reported to reduce fi- tributes to the profibrotic response to chronic renal injury via broblast migration.24 Steady-state CXCL10 mRNA levels were activation of proinflammatory pathways and generation of oxsignificantly higher in UUO kidneys from CD36⫺/⫺ mice at idative stress. Third, downregulation of proinflammatory and days 7 and 14 compared with CD36⫹/⫹ mice (Figure 4D). oxidative pathways as a consequence of genetic CD36 defiComputer-assisted image analysis of kidney sections stained ciency limits myofibroblast accumulation and fibrosis within for CXCL10 confirmed a significant 152% increase in the obstructed kidney. CD36⫺/⫺ kidneys compared with CD36⫹/⫹ kidneys on day It has long been known that macrophages are phenotypiJ Am Soc Nephrol 20: 495–505, 2009
CD36 Regulates Oxidative Inflammation
499
BASIC RESEARCH
www.jasn.org
Figure 8. Reduction in fibrosis severity after UUO in CD36⫺/⫺ mice is independent of TGF- activation. (Top) Representative Western blot of phosphorylated Smad2 (pSmad2) and total Smad2 for days 3 and 7 after UUO in CD36⫹/⫹ and CD36⫺/⫺ mice. (Bottom) Summary of the results of band density measurements for pSmad2 normalized to total Smad2. Data are means ⫾ SEM. P ⬎ 0.05, CD36⫹/⫹ versus CD36⫺/⫺. Figure 6. Kidney oxLDL levels are reduced in CD36⫺/⫺ mice after UUO. (Top) Representative HOCl-LDL/oxLDL immunohistochemical photomicrographs in CD36⫹/⫹ and CD36⫺/⫺ mice at day 14 after UUO. (Bottom) Summary of the quantification results obtained by computer-assisted image analysis expressed as the percentage of tubulointerstitial area staining positive for HOClLDL. Data are means ⫾ SEM. †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. Magnification, ⫻400.
Figure 7. Reduction in total oxidative stress in obstructed kidneys of CD36⫺/⫺ mice after UUO. Protein oxidation was measured as total protein carbonyl content by ELISA. Lipid peroxidation was measured as HODE by RP-HPLC. The levels of HODE and protein carbonyl content were reduced at day 7 after UUO in CD36⫹/⫹ and CD36⫺/⫺ mice. Linear regression demonstrated a strong positive correlation between HODE levels and protein carbonyl content and indicated much lower levels in CD36⫺/⫺ mice 7 d after UUO (r2 ⫽ 0.8374; F ⫽ 139.1, P ⬍ 0.01). ⽧, CD36⫹/⫹ sham-operated (n ⫽ 7); Œ, CD36⫺/⫺ sham-operated (n ⫽ 7); f, CD36⫹/⫹ day 7 UUO (n ⫽ 7); F, CD36⫺/⫺ day 7 UUO (n ⫽ 8). 500
Journal of the American Society of Nephrology
Figure 9. Myofibroblast accumulation is attenuated in CD36⫺/⫺ mice after UUO. (Top) Representative ␣-SMA immunohistochemical photomicrographs in CD36⫹/⫹ and CD36⫺/⫺ mice 14 d after UUO. (Bottom) Summary of the quantification by computerassisted image analysis, expressed as the percentage of tubulointerstitial area staining positive for ␣-SMA. Data are as means ⫾ SEM. †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. Magnification, ⫻400.
cally and functionally heterogeneous, potentially able to promote inflammation and fibrosis or to prevent it via scavenging activities. In most studies, the number of renal interstitial macrophages has correlated closely with fibrosis severity. One noJ Am Soc Nephrol 20: 495–505, 2009
www.jasn.org
Figure 10. CXCL10 levels are increased in CD36⫺/⫺ mice after UUO. (Top) Representative CXCL10/IP-10 immunohistochemical photomicrographs in CD36⫹/⫹ and CD36⫺/⫺ mice 7 d after UUO. (Bottom) Summary of the quantification by computer-assisted image analysis, expressed as the percentage of tubulointerstitial area staining positive for CXCL10. Data are means ⫾ SEM. †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. Magnification, ⫻400.
table exception has been in models of renal injury induced in genetically engineered animals with impaired macrophage phagocytic activities.1,26 The findings in this study suggest that the CD36 scavenger receptor can be added to this list. We previously demonstrated that CD36⫹ cells infiltrated the interstitium 7 and 14 d after UUO; dual staining indicated that most of the CD36⫹ interstitial cells were F4/80⫹ macrophages.18 In this study, interstitial macrophages were significantly more numerous in CD36⫺/⫺ mice at day 7 after UUO despite reduced fibrosis. The increased numbers of F4/80⫹ interstitial cells in CD36⫺/⫺ mice could not be attributed to an increase in renal CD11b/c⫹ dendritic cells27 and likely represents a separate population of macrophages that will be characterized in future studies. Despite an increase in macrophage number, functionally these interstitial cells did not augment proinflammatory and oxidative pathways. CD36 expression has been reported to increase as monocytes differentiate into activated macrophages.28,29 In this study, the expression pattern of CXCL2 corresponded to the severity of fibrosis in CD36⫺/⫺ mice. Recent studies suggested that CXCL2 may be an important marker of macrophage activation and is important in the generation of proinflammatory mediators.30 –32 Taken together, these data suggest that CD36 expression by renal interstitial macrophages may serve as a marker for a profibrotic phenotype in chronic renal injury. The initial phase of cellular activation and tissue injury that J Am Soc Nephrol 20: 495–505, 2009
BASIC RESEARCH
leads to renal fibrosis is characterized by the generation of proinflammatory mediators. NF-B transcription factors are key regulators of inflammation and immunity that control expression of important immunoregulatory genes.33 In this study, there was a ⬎50% reduction in nuclear translocation of NF-B heterodimers in hypercholesterolemic CD36⫺/⫺ mice compared with CD36⫹/⫹ mice. Studies suggested that RelA (p65) is a key regulator in the requisite inflammatory phase of progressive renal fibrosis.34,35 By supershift assay and immunostaining, the decrease in NF-B nuclear translocation in both tubular and interstitial cells after UUO in CD36⫺/⫺ mice was largely attributable to a reduction in the proinflammatory p50RelA heterodimer. Whether this decrease is related to a reduction in a RelA-dependent signaling pathway, increased ubiquitination, or proteasomal degradation of RelA is not clear and deserves further investigation. This study also detected a significant reduction in both HODE and carbonyl protein formation in hypercholesterolemic CD36⫺/⫺ mice compared with CD36⫹/⫹ mice 7 d after UUO. Generation of lipid peroxides such as HODE and carbonyl proteins are established markers of oxidative stress and reflect the balance between pro-oxidant and antioxidant mechanisms within the tissue. There was a strong positive correlation between the presence of CD36 and the level of oxidative stress within obstructed kidneys. In an experimental stroke model, Cho et al.36 demonstrated that the degree of ischemia was reduced in CD36⫺/⫺ mice and was associated with less reactive oxygen species (ROS) production in the ischemic area; however, the exact mechanism by which CD36 regulates ROS production is not known. There is increasing evidence that ROS and other reactive species may also be directly involved in redox signaling pathways that promote inflammation and apoptosis.37–39 Susztak et al.40 reported that hyperglycemia in-
Figure 11. Attenuated renal fibrosis in CD36⫺/⫺ mice is independent of cholesterol levels. Total kidney collagen content determined by hydroxyproline assay is significantly decreased in obstructed kidneys from CD36⫺/⫺ mice compared with CD36⫹/⫹ mice on a control diet. There was no significant difference between CD36⫺/⫺ mice on a control diet and a high-fat diet after UUO. u, CD36⫹/⫹ mice on control diet (n ⫽ 6 to 7); o, CD36⫺/⫺ mice on control diet (n ⫽ 7 to 8); f, CD36⫹/⫹ mice on high-fat diet (n ⫽ 6 to 7); 䡺, CD36⫺/⫺ mice on high-fat diet (n ⫽ 8 to 10). Data are means ⫾ SEM. †P ⬍ 0.01, CD36⫹/⫹ versus CD36⫺/⫺. CD36 Regulates Oxidative Inflammation
501
BASIC RESEARCH
www.jasn.org
duced CD36 expression and apoptosis in cultured proximal tubular cells. In this study, the decrease in fibrosis severity was also seen in CD36⫺/⫺ mice on a normal diet. These data are consistent with the view that CD36 not only is a receptor for oxidized ligands such as oxLDL but also CD36 propagates a state of oxidative stress and worsening fibrosis in chronically damaged kidneys. Within the renal interstitium, the transition from inflammation to fibrosis is characterized by the appearance of interstitial myofibroblasts. Several studies have demonstrated a close association between inflammation intensity and severity of renal fibrosis.41– 43 In this study, a reduction in proinflammatory and oxidative pathways in the obstructed kidneys of hypercholesterolemic CD36⫺/⫺ mice resulted in a significant reduction in interstitial myofibroblast density and subsequent collagen deposition. These findings extend those of Duffield et al.41 and suggest that CD36-positive macrophages regulate interstitial myofibroblast accumulation, but the mechanisms by which this occurs remains unclear. CXCL10 is an important chemokine expressed in the later phases of wound remodeling in the skin and serves to communicate key signals between dermal layers to end the regenerative phase and initiate the remodeling phase.44 In this study, renal tubular CXCL10 levels remained elevated at day 14 after UUO in CD36⫺/⫺ mice compared with decreasing levels in CD36⫹/⫹ mice. Studies of in vivo models of pulmonary and renal fibrosis have suggested that the increased fibrosis reported with genetic deficiency or pharmacologic blockade of CXCL10 was due to greater myofibroblast accumulation.24,45 Evidence suggested that CXCL10 serves as an important inhibitor of fibroblast motility via suppression of m-calpain activation46,47 and may promote fibroblast/myofibroblast differentiation.44,48 Whether CD36 directly or indirectly modulates CXCL10 expression is unclear but suggests another potential mechanism that bridges CD36 activity and the degree of renal fibrosis. The results of this study begin to elucidate the cellular pathways involved in hypercholesterolemic chronic renal injury. Similar to its role in atherogenesis, CD36 is a major receptor for oxLDL within the kidney, where it functions as a key modulator of inflammatory, oxidative, and fibrogenic pathways. Further studies are under investigation to determine the relative contribution of CD36⫹ tubules and CD36⫹ macrophages and their potential as therapeutic targets to prevent CKD progression.
CONCISE METHODS Experimental Design Breeding pairs of CD36⫺/⫺ and CD36⫹/⫹ mice on C57BL/6 background were obtained from Dr. Maria Febbraio’s colony13 and were bred in our animal facility in Seattle. Male mice were fed a high-fat Western diet (15.8% total fat, 1.25% cholesterol with 0.5% sodium cholate [TD 88051; Harlan Teklad, Madison, WI])49 beginning after weaning at 3 to 4 wk of age. After a run-in period of 7 to 8 wk on the 502
Journal of the American Society of Nephrology
high-fat Western diet, animals were randomly assigned to one of four experimental groups: CD36⫹/⫹, CD36⫺/⫺, sham surgery, or UUO. Groups of male mice (n ⫽ 6 to 10 each) were killed at 3, 7, and 14 d after surgery. Groups of CD36⫹/⫹ and CD36⫺/⫺ mice male mice (n ⫽ 6 to 8) were also placed on a standard chow diet with 0.5% sodium cholate for 7 to 8 wk after weaning, and UUO surgery was performed at 10 to 11 wk of age. For mice in the UUO group, the left ureter was exposed through a mid-abdominal incision and ligated using 4-0 silk. All surgeries were performed under general anesthesia with isoflurane. All procedures were performed in accordance with the guidelines established by National Research Council Guide for the Care and Use of Laboratory Animals and approval of our Institute Animal Care and Use Committee. Sham and UUO kidneys were harvested and processed for RNA and protein extraction and histologic studies as described previously.19 Frozen tissue samples were stored at ⫺80°C. Serum cholesterol levels were measured in blood samples that were obtained at killing using the Total Cholesterol Kit (WAKO Chemicals USA, Richmond, VA).
Genotyping Genotyping was performed by PCR using genomic DNA isolated from tails. PCR primer sequences were obtained from Dr. Febbraio, and genotyping was performed as described previously.13 Primers for the CD36⫹/⫹ allele are 5⬘-CAGCTCATACATTGCTGTTTATGCATG and 3⬘-GGTACAATCACAGTGTTTTCTACGTGG (band size approximately 600 bp). The primers for the CD36⫺/⫺ allele include the CD36⫹/⫹ 5⬘ primer and 3⬘-CGCTTCCTCGTGCTTTACGGTATC (band size approximately 750 bp).
Collagen Content
Hydroxyproline content of kidney tissue (g of hydroxyproline per mg of wet weight kidney section) was performed by acid hydrolysis of the tissue section using procedures established in our laboratory.19
Histologic Examination Immunohistochemical staining was performed on sections of paraffin-embedded tissue or cryosections of snap-frozen tissue using procedures established in our laboratory with VECTASTAIN Elite ABC Kits (Vector Laboratories, Burlingame, CA) and AEC Substrate Chromogen K3464 (Dako Corp., Carpinteria, CA) or DAB (Sigma, St. Louis, MO) as the peroxidase substrate. Sections were blocked with Avidin/Biotin blocking kit (Vector Laboratories). Primary antibodies used were reactive with F4/80 (rat anti-mouse F4/80 monoclonal; AbD Serotec, Raleigh, NC), CD11b/c (rabbit anti-mouse CD11b/c; Abcam, Cambridge, MA), CXCL10 (rabbit anti-mouse CXCL10 polyclonal; PeproTech, Rocky Hill, NJ), IB-␣ and phosphorylated IB-␣ (rabbit anti-human IB␣ polyclonal and rabbit anti-human phospho IB-␣ polyclonal; Cell Signaling Technology, Danvers, MA), phosphorylated p65 (rabbit anti-mouse p65 polyclonal; Cell Signaling Technology), Smad2 (rabbit anti-human Smad2 monoclonal; Cell Signaling Technology), phosphorylated Smad2 (rabbit antihuman phospho Smad2 polyclonal; Upstate Biotechnology, Billerica, MA), and oxLDL (rabbit anti-human hypochlorous acid modified LDL polyclonal; Chemicon Int., Temecula, CA). Interstitial myofibroblasts were quantified by staining using peroxidase-conjugated J Am Soc Nephrol 20: 495–505, 2009
www.jasn.org
murine anti-human ␣-SMA 1A4 mAb (Dako). The 1A4 antibody was detected using the Enhanced Polymer One-Step Staining reagent (Dako) as described previously.19,20,50 Secondary antibodies were shown to be nonreactive with tissue sections stained without primary antibody. Semiquantitative computer-assisted image analysis of tubulointerstitial proteins was performed on six randomly selected ⫻400 magnified images of slides from individual animals with ImagePro Plus software (Media Cybernetics, Bethesda, MD). Glomerular areas and space not occupied by tissue were subtracted in the analysis. Interstitial macrophage density was expressed as percentage of F4/ 80⫹ interstitial area on DAB-stained cryosections. The percentage of macrophage density was determined from results of five randomly selected ⫻400 magnified images of slides from individual animals with assistance from Image-Pro Plus software (Mediatech). The investigator was blinded to the experimental groups at the time of analysis.
RNAse Protection Assay Total kidney RNA was extracted using the TRIZOL single-step reagent (Life Technologies BRL Life Technologies, Grand Island, NY). A mouse chemokine RNase protection assay (mCK-5c) was performed according to protocol (BD Biosciences Pharmingen, San Diego, CA). Mouse chemokine multiprobe templates were radiolabeled with [␣-32P]UTP (Perkin Elmer Life Sciences, Boston, MA). Probe hybridization with 15 g of total RNA was performed overnight at 56°C, and unbound probe or single-stranded RNA was digested with RNase. Hybridized RNA samples were separated by electrophoresis, and gels were dried and visualized with the Typhoon 9410 PhosphorImager (GE Healthcare, Pittsburgh, PA). Chemokine band intensities were normalized relative L32 band intensities because it was closer to chemokine band levels. There was difference in significance between normalizing between glyceraldehyde-3-phosphate dehydrogenase and L32.
EMSA and Supershift Assay Kidney tissue nuclear extracts were prepared according to the “smallscale” method of Andrews and Faller.51 In brief, frozen tissue was homogenized on ice in Buffer A (0.5 M sucrose, 10 mM HEPES, 10 mM KCl, 0.1 mM EDTA ⫹ 0.5 mM dithiothreitol, and 0.5 mM PMSF) including Proteinase Inhibitor cocktail (Sigma-Aldrich). Homogenates were centrifuged and resuspended in Buffer C (10% glycerol, 20 mM HEPES, 420 mM NaCl, 1.5 mM MgCl2, 0.1 mM EDTA, 0.5 mM dithiothreitol, 0.5 mM PMSF, and proteinase inhibitor cocktail). Nuclear extract protein concentrations were determined using the BCA Protein Assay Kit (Pierce Biotechnology, Rockford, IL). Complementary oligonucleotides were annealed by heating and cooling overnight: NF-B1 (5⬘-AGCTTAGAGGGACTTTCCGAGAGGA-3⬘), NF-B2 (5⬘-AGCTTCCTCTCGGAAAGTCCCCTCTA-3⬘), AP1–1 (5⬘-GATCCATGACTCAGAGGAAAACA-3⬘), and AP1–2 (5⬘-GATCTGTTTTCCTCTGAGTCATG-3⬘). NF-B or AP1 radiolabeled probes were prepared using Klenow enzyme and [␣-32P]dATP (Perkin Elmer Life Sciences). Specific (unlabeled or cold probe) and nonspecific (AP-1) competitors were included to determine probe specificity. Phorbol myristate acetate–stimulated and unstimulated Jurkat T cells were used as positive and negative controls, respectively. SuJ Am Soc Nephrol 20: 495–505, 2009
BASIC RESEARCH
pershift was achieved using polyclonal rabbit anti-human p65 and p50 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA). Gels were dried and visualized with the Typhoon 9410 PhosphorImager.
Western Blotting Protein was isolated from homogenized frozen kidney, and Western blotting was performed as described previously.18 Bands were normalized against -actin (anti-mouse -actin; Sigma). The secondary antibodies were anti-rabbit and anti-mouse IR700Dye and IR800Dye (Rockland Immunochemicals, Gilbertsville, PA). Protein bands were visualized and quantified using the Odyssey (Li-Cor Biosciences, Lincoln, NE).
Oxidative Stress Anesthetized mice were perfused through the left ventricle with cold antioxidant solution [100 M diethylenetriaminepentaacetic acid dianhydride, 1 mM 2,6-bis(1,1-dimethylethyl)-4-methylphenol, 10 mM 3-amino-1,2,4-triazole stock, 50 mM sodium phosphate (pH 7); Sigma-Aldrich)] at the time of killing. Kidney tissue was immediately harvested, immersed in antioxidant buffer, and stored at ⫺80°C until analysis. Lipids were extracted from kidney tissue using modified Dole procedure.52 HODEs were quantified by reverse-phase hydrolysis C-18 HPLC analysis of triphenylphosphine-reduced lipid extracts after base hydrolysis. The protein content of tissue pellets was determined by Bradford assay using BSA as standard. The quantity of protein carbonyls in a protein sample were determined by derivatizing with dinitrophenylhydrazine (DNP) and measuring bound DNP immunologicaly using the Zentech PC Test EIA Kit (Biocell Corp., Auckland, New Zealand). Samples containing protein were reacted with DNP, and then the protein was nonspecifically adsorbed to an ELISA plate. Unconjugated DNP and nonprotein constituents are washed away. The adsorbed protein was probed with biotinylated anti-DNP antibody followed by streptavidin-linked horseradish peroxidase.53,54 Absorbances were related to standard curve prepared for BSA containing increasing proportions of hypochlorous acid– oxidized protein.52
Semiquantitative Real-Time qPCR
First-strand cDNA was prepared from 1 g of total RNA using the Bio-Rad iScript cDNA Synthesis kit (Bio-Rad Laboratories, Hercules, CA). Semiquantitative real-time qPCR was performed according to the IQ SYBR Green Supermix kit (Bio-Rad Laboratories) using CD68 primers (forward 5⬘-TGGCAGGGTTATGAGTGACA; reverse 5⬘TAGCCCAAGGAACAGAGGAA) and primers for the housekeeping gene 18S (forward 5⬘-GGTGAAATTCTTGGACCGGC; reverse 5⬘GACTTTGGTTTCCCTTAAGC; Invitrogen, Carlsbad, CA). Realtime qPCR reactions containing 1.5 l of cDNA, 0.2 M primers, and the 2X SYBR Green Supermix were run in the Bio-Rad iQ thermal cycler with programs specific for the respective gene. Reactions were run in triplicate and CD68 was normalized to 18s mRNA. Absence of primer-dimers was confirmed for primer specificity. Single amplicons were confirmed by gel electrophoresis. Data analysis was perCD36 Regulates Oxidative Inflammation
503
BASIC RESEARCH
www.jasn.org
formed using the Pfaffl algorithm with the REST analysis software version 1.9.9 (Corbett Research Pty. Ltd., Mortlake, NSW, Australia).
Statistical Analysis
All data are presented as means ⫾ SE. A nested ANOVA was used for all semiquantitative computer-assisted image analysis. For image analysis data, the arithmetic mean of six randomly selected images of slides for each animal was used to calculate the reported mean of the group and the SE. A Pearson linear pairwise correlation was performed on lipid peroxidation and protein oxidation data. All other results were analyzed by unpaired t test. P ⬍ 0.05 was considered statistically significant.
8.
9.
10.
11. 12.
ACKNOWLEDGMENTS We gratefully acknowledge research grant support from the National Institutes of Health (DK54500 to A.A.E., K08 DK073497 to D.M.O., 5 P50 DK039255 to S.P., HL70083 to M.F., and DK44757 to A.A.E.), Child Health Research Center Scholar Award (K12 HD043376 to D.M.O.), National Kidney Foundation Post-Doctoral Research Fellowship, and National Kidney Foundation Young Investigator award (D.M.O.). Part of this work was presented as an abstract at the annual meeting of the American Society of Nephrology; Philadelphia, PA; November 9, 2005 and as an oral presentation at the annual meeting of the American Society of Pediatric Nephrology; Washington, DC; May 15, 2005.
13.
14.
15.
16.
17.
DISCLOSURES None.
18.
REFERENCES 19. 1. Jones CA, McQuillan GM, Kusek JW, Eberhardt MS, Herman WH, Coresh J, Salive M, Jones CP, Agodoa LY: Serum creatinine levels in the US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis 32: 992–999, 1998 2. US Renal Data System: USRDS 2007 Annual Data Report: Atlas of end-stage renal disease in the United States, Bethesda, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2007 3. Levey AS, Beto JA, Coronado BE, Eknoyan G, Foley RN, Kasiske BL, Klag MJ, Mailloux LU, Manske CL, Meyer KB, Parfrey PS, Pfeffer MA, Wenger NK, Wilson PW, Wright JT Jr: Controlling the epidemic of cardiovascular disease in chronic renal disease: What do we know? What do we need to learn? Where do we go from here? National Kidney Foundation Task Force on Cardiovascular Disease. Am J Kidney Dis 32: 853–906, 1998 4. Himmelfarb J: Linking oxidative stress and inflammation in kidney disease: Which is the chicken and which is the egg? Semin Dial 17: 449 – 454, 2004 5. Himmelfarb J: Relevance of oxidative pathways in the pathophysiology of chronic kidney disease. Cardiol Clin 23: 319 –330, 2005 6. Muntner P, Coresh J, Smith JC, Eckfeldt J, Klag MJ: Plasma lipids and risk of developing renal dysfunction: The atherosclerosis risk in communities study. Kidney Int 58: 293–301, 2000 7. Hunsicker LG, Adler S, Caggiula A, England BK, Greene T, Kusek JW,
504
Journal of the American Society of Nephrology
20.
21.
22. 23. 24.
25. 26.
27.
Rogers NL, Teschan PE: Predictors of the progression of renal disease in the Modification of Diet in Renal Disease Study. Kidney Int 51: 1908 –1919, 1997 Washio M, Okuda S, Ikeda M, Hirakata H, Nanishi F, Onoyama K, Yoshimura T, Fujishima M: Hypercholesterolemia and the progression of the renal dysfunction in chronic renal failure patients. J Epidemiol 6: 172–177, 1996 Boes E, Fliser D, Ritz E, Ko¨nig P, Lhotta K, Mann JF, Mu¨ller GA, Neyer U, Riegel W, Riegler P, Kronenberg F: Apolipoprotein A-IV predicts progression of chronic kidney disease: The mild to moderate kidney disease study. J Am Soc Nephrol 17: 528 –536, 2006 Hsu CC, Kao WH, Coresh J, Pankow JS, Marsh-Manzi J, Boerwinkle E, Bray MS: Apolipoprotein E and progression of chronic kidney disease. JAMA 293: 2892–2899, 2005 Moore KJ, Freeman MW: Scavenger receptors in atherosclerosis: Beyond lipid uptake. Arterioscler Thromb Vasc Biol 26: 1702–1711, 2006 Nicholson AC, Febbraio M, Han J, Silverstein RL, Hajjar DP: CD36 in atherosclerosis: The role of a class B macrophage scavenger receptor. Ann N Y Acad Sci 902: 128 –131, discussion 131–123, 2000 Febbraio M, Abumrad NA, Hajjar DP, Sharma K, Cheng W, Pearce SF, Silverstein RL: A null mutation in murine CD36 reveals an important role in fatty acid and lipoprotein metabolism. J Biol Chem 274: 19055– 19062, 1999 Rahaman SO, Lennon DJ, Febbraio M, Podrez EA, Hazen SL, Silverstein RL: A CD36-dependent signaling cascade is necessary for macrophage foam cell formation. Cell Metab 4: 211–221, 2006 Yamashita S, Hirano K, Kuwasako T, Janabi M, Toyama Y, Ishigami M, Sakai N: Physiological and pathological roles of a multi-ligand receptor CD36 in atherogenesis: Insights from CD36-deficient patients. Mol Cell Biochem 299: 19 –22, 2007 Lipsky RH, Eckert DM, Tang Y, Ockenhouse CF: The carboxyl-terminal cytoplasmic domain of CD36 is required for oxidized low-density lipoprotein modulation of NF-kappaB activity by tumor necrosis factor-alpha. Recept Signal Transduct 7: 1–11, 1997 Chmielewski M, Bryl E, Marzec L, Aleksandrowicz E, Witkowski JM, Rutkowski B: Expression of scavenger receptor CD36 in chronic renal failure patients. Artif Organs 29: 608 – 614, 2005 Okamura DM, Lopez-Guisa JM, Koelsch K, Collins S, Eddy AA: Atherogenic scavenger receptor modulation in the tubulointerstitium in response to chronic renal injury. Am J Physiol Renal Physiol 293: F575–F585, 2007 Matsuo S, Lopez-Guisa JM, Cai X, Okamura DM, Alpers CE, Bumgarner RE, Peters MA, Zhang G, Eddy AA: Multifunctionality of PAI-1 in fibrogenesis: Evidence from obstructive nephropathy in PAI-1-overexpressing mice. Kidney Int 67: 2221–2238, 2005 Yamaguchi I, Lopez-Guisa JM, Cai X, Collins SJ, Okamura DM, Eddy AA: Endogenous urokinase lacks antifibrotic activity during progressive renal injury. Am J Physiol Renal Physiol 293: F12–F19, 2007 Khan-Dawood FS, Yang J, Dawood MY: Immunocytochemical localization and expression of E-cadherin and beta-catenin in the human corpus luteum. Mol Hum Reprod 2: 753–757, 1996 Stadtman ER, Oliver CN: Metal-catalyzed oxidation of proteins: Physiological consequences. J Biol Chem 266: 2005–2008, 1991 Yoshida Y, Hayakawa M, Niki E: Total hydroxyoctadecadienoic acid as a marker for lipid peroxidation in vivo. Biofactors 24: 7–15, 2005 Tager AM, Kradin RL, LaCamera P, Bercury SD, Campanella GS, Leary CP, Polosukhin V, Zhao LH, Sakamoto H, Blackwell TS, Luster AD: Inhibition of pulmonary fibrosis by the chemokine IP-10/CXCL10. Am J Respir Cell Mol Biol 31: 395– 404, 2004 Eddy AA: Molecular basis of renal fibrosis. Pediatr Nephrol 15: 290 – 301, 2000 Zhang X, Fitzsimmons RL, Cleland LG, Ey PL, Zannettino AC, Farmer EA, Sincock P, Mayrhofer G: CD36/fatty acid translocase in rats: Distribution, isolation from hepatocytes, and comparison with the scavenger receptor SR-B1. Lab Invest 83: 317–332, 2003 Ma X, Bacci S, Mlynarski W, Gottardo L, Soccio T, Menzaghi C, Iori E,
J Am Soc Nephrol 20: 495–505, 2009
www.jasn.org
28.
29.
30.
31.
32.
33. 34.
35.
36.
37.
38.
39.
40.
41.
Lager RA, Shroff AR, Gervino EV, Nesto RW, Johnstone MT, Abumrad NA, Avogaro A, Trischitta V, Doria A: A common haplotype at the CD36 locus is associated with high free fatty acid levels and increased cardiovascular risk in Caucasians. Hum Mol Genet 13: 2197–2205, 2004 Yesner LM, Huh HY, Pearce SF, Silverstein RL: Regulation of monocyte CD36 and thrombospondin-1 expression by soluble mediators. Arterioscler Thromb Vasc Biol 16: 1019 –1025, 1996 Huh HY, Pearce SF, Yesner LM, Schindler JL, Silverstein RL: Regulated expression of CD36 during monocyte-to-macrophage differentiation: Potential role of CD36 in foam cell formation. Blood 87: 2020 –2028, 1996 De Plaen IG, Han XB, Liu X, Hsueh W, Ghosh S, May MJ: Lipopolysaccharide induces CXCL2/macrophage inflammatory protein-2 gene expression in enterocytes via NF-kappaB activation: Independence from endogenous TNF-alpha and platelet-activating factor. Immunology 118: 153–163, 2006 Jalonen U, Nieminen R, Vuolteenaho K, Kankaanranta H, Moilanen E: Down-regulation of tristetraprolin expression results in enhanced IL-12 and MIP-2 production and reduced MIP-3alpha synthesis in activated macrophages. Mediators Inflamm 2006: 40691, 2006 Shibata T, Nagata K, Kobayashi Y: Cutting edge: A critical role of nitrogen oxide in preventing inflammation upon apoptotic cell clearance. J Immunol 179: 3407–3411, 2007 Bonizzi G, Karin M: The two NF-kappaB activation pathways and their role in innate and adaptive immunity. Trends Immunol 25: 280–288, 2004 Fujihara CK, Antunes GR, Mattar AL, Malheiros DM, Vieira JM Jr, Zatz R: Chronic inhibition of nuclear factor-kappaB attenuates renal injury in the 5/6 renal ablation model. Am J Physiol Renal Physiol 292: F92–F99, 2007 Lloberas N, Torras J, Herrero-Fresneda I, Cruzado JM, Riera M, Hurtado I, Grinyo JM: Postischemic renal oxidative stress induces inflammatory response through PAF and oxidized phospholipids: Prevention by antioxidant treatment. FASEB J 16: 908 –910, 2002 Cho S, Park EM, Febbraio M, Anrather J, Park L, Racchumi G, Silverstein RL, Iadecola C: The class B scavenger receptor CD36 mediates free radical production and tissue injury in cerebral ischemia. J Neurosci 25: 2504 –2512, 2005 Yang YL, Lin SH, Chuang LY, Guh JY, Liao TN, Lee TC, Chang WT, Chang FR, Hung MY, Chiang TA, Hung CY: CD36 is a novel and potential anti-fibrogenic target in albumin-induced renal proximal tubule fibrosis. J Cell Biochem 101: 735–744, 2007 Leonarduzzi G, Arkan MC, Basaga H, Chiarpotto E, Sevanian A, Poli G: Lipid oxidation products in cell signaling. Free Radic Biol Med 28: 1370 –1378, 2000 Zhao M, Fernandez LG, Doctor A, Sharma AK, Zarbock A, Tribble CG, Kron IL, Laubach VE: Alveolar macrophage activation is a key initiation signal for acute lung ischemia-reperfusion injury. Am J Physiol Lung Cell Mol Physiol 291: L1018 –L1026, 2006 Susztak K, Ciccone E, McCue P, Sharma K, Bottinger EP: Multiple metabolic hits converge on CD36 as novel mediator of tubular epithelial apoptosis in diabetic nephropathy. PLoS Med 2: e45, 2005 Duffield JS, Tipping PG, Kipari T, Cailhier JF, Clay S, Lang R, Bonventre JV, Hughes J: Conditional ablation of macrophages halts progres-
J Am Soc Nephrol 20: 495–505, 2009
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
BASIC RESEARCH
sion of crescentic glomerulonephritis. Am J Pathol 167: 1207–1219, 2005 Ninichuk V, Khandoga AG, Segerer S, Loetscher P, Schlapbach A, Revesz L, Feifel R, Khandoga A, Krombach F, Nelson PJ, Schlo¨ndorff D, Anders HJ: The role of interstitial macrophages in nephropathy of type 2 diabetic db/db mice. Am J Pathol 170: 1267–1276, 2007 Tesch GH, Schwarting A, Kinoshita K, Lan HY, Rollins BJ, Kelley VR: Monocyte chemoattractant protein-1 promotes macrophage-mediated tubular injury, but not glomerular injury, in nephrotoxic serum nephritis. J Clin Invest 103: 73– 80, 1999 Yates CC, Whaley D, Kulasekeran P, Hancock WW, Lu B, Bodnar R, Newsome J, Hebda PA, Wells A: Delayed and deficient dermal maturation in mice lacking the CXCR3 ELR-negative CXC chemokine receptor. Am J Pathol 171: 484 – 495, 2007 Nakaya I, Wada T, Furuichi K, Sakai N, Kitagawa K, Yokoyama H, Ishida Y, Kondo T, Sugaya T, Kawachi H, Shimizu F, Narumi S, Haino M, Gerard C, Matsushima K, Kaneko S: Blockade of IP-10/CXCR3 promotes progressive renal fibrosis. Nephron Exp Nephrol 107: e12– e21, 2007 Shiraha H, Glading A, Chou J, Jia Z, Wells A: Activation of m-calpain (calpain II) by epidermal growth factor is limited by protein kinase A phosphorylation of m-calpain. Mol Cell Biol 22: 2716 –2727, 2002 Shiraha H, Glading A, Gupta K, Wells A: IP-10 inhibits epidermal growth factor-induced motility by decreasing epidermal growth factor receptor-mediated calpain activity. J Cell Biol 146: 243–254, 1999 Kouroumalis A, Nibbs RJ, Aptel H, Wright KL, Kolios G, Ward SG: The chemokines CXCL9, CXCL10, and CXCL11 differentially stimulate G alpha i-independent signaling and actin responses in human intestinal myofibroblasts. J Immunol 175: 5403–5411, 2005 Paigen B, Holmes PA, Mitchell D, Albee D: Comparison of atherosclerotic lesions and HDL-lipid levels in male, female, and testosteronetreated female mice from strains C57BL/6, BALB/c, and C3H. Atherosclerosis 64: 215–221, 1987 Kim H, Oda T, Lopez-Guisa J, Wing D, Edwards DR, Soloway PD, Eddy AA: TIMP-1 deficiency does not attenuate interstitial fibrosis in obstructive nephropathy. J Am Soc Nephrol 12: 736 –748, 2001 Andrews NC, Faller DV: A rapid micropreparation technique for extraction of DNA-binding proteins from limiting numbers of mammalian cells. Nucleic Acids Res 19: 2499, 1991 Pennathur S, Ido Y, Heller JI, Byun J, Danda R, Pergola P, Williamson JR, Heinecke JW: Reactive carbonyls and polyunsaturated fatty acids produce a hydroxyl radical-like species: A potential pathway for oxidative damage of retinal proteins in diabetes. J Biol Chem 280: 22706 –22714, 2005 Buss H, Chan TP, Sluis KB, Domigan NM, Winterbourn CC: Protein carbonyl measurement by a sensitive ELISA method. Free Radic Biol Med 23: 361–366, 1997 Winterbourn CC, Buss IH: Protein carbonyl measurement by enzymelinked immunosorbent assay. Methods Enzymol 300: 106 –111, 1999
Supplemental information for this article is available online at http://www.jasn. org/.
CD36 Regulates Oxidative Inflammation
505
