The Expression of SPARC in Adipose Tissue and Its Increased Plasma Concentration in Patients with Coronary Artery Disease

Transcription

1 The Expression of SPARC in Adipose Tissue and Its Increased Plasma Concentration in Patients with Coronary Artery Disease Masahiko Takahashi, Hiroyuki Nagaretani, Tohru Funahashi, Hitoshi Nishizawa, Norikazu Maeda, Ken Kishida, Hiroshi Kuriyama, Iichiro Shimomura, Kazuhisa Maeda, Kikuko Hotta, Noriyuki Ouchi, Shinji Kihara, Tadashi Nakamura, Shizuya Yamashita, and Yuji Matsuzawa Abstract TAKAHASHI, MASAHIKO, HIROYUKI NAGARETANI, TOHRU FUNAHASHI, HITOSHI NISHIZAWA, NORIKAZU MAEDA, KEN KISHIDA, HIROSHI KURIYAMA, IICHIRO SHIMOMURA, KAZUHISA MAEDA, KIKUKO HOTTA, NORIYUKI OUCHI, SHINJI KIHARA, TADASHI NAKAMURA, SHIZUYA YAMASHITA, AND YUJI MATSUZAWA. The expression of SPARC in adipose tissue and its increased plasma concentration in patients with coronary artery disease. Obes Res. 2001;9: Objective: Adipocytes secrete various cytokines and matrix proteins. Several of them precipitate in obesity-associated diseases, including atherosclerosis. In the current study, we have examined the expression of secreted protein, acidic and rich in cysteine (SPARC) in adipose tissue and its significance in obesity and coronary artery disease (CAD). Research Methods and Procedures: The SPARC mrna expressions both in vivo and in vitro were detected by Northern blot analysis. Plasma SPARC concentrations were measured by enzyme immunosorbent assay. First, we investigated the plasma SPARC levels of 88 unrelated adult Japanese subjects (62 men and 26 women; average age: [ SD] years; body mass index [BMI]: 16 to 46 kg/m 2 ). Additionally 31 subjects with CAD diagnosed by coronary angiography (20 men and 11 women) were also investigated. Results: Human adipose tissues expressed abundant SPARC mrna. SPARC expression in adipose tissues was Submitted for publication November 20, Accepted for publication in final form April 25, Department of Internal Medicine and Molecular Science, Graduate School of Medicine, Osaka University, Yamadaoka, Suita, Osaka, Japan. Address correspondence to Dr. Tohru Funahashi, Department of Internal Medicine and Molecular Science, Graduate School of Medicine, B5, Osaka University (Second Department of Internal Medicine, Osaka University Medical School), 2-2 Yamadaoka, Suita, Osaka , Japan. tohru@imed2.med.osaka-u.ac.jp Copyright 2001 NAASO upregulated in obese db/db mice. Markedly enhanced expression of SPARC mrna was observed in 3T3-L1 fibroblasts during adipocyte differentiation. Consistent with these results, plasma SPARC levels proved a positive correlation with BMI in humans (r 0.27; p 0.01). Interestingly, plasma SPARC concentrations were significantly elevated in age- and BMI-matched subjects with CAD (p 0.05). Discussion: SPARC was expressed in adipose tissues and its expression was enhanced in obese mice. In human, plasma SPARC levels were elevated in obesity and CAD patients. This elevated SPARC may be involved in the progression of CAD. Key words: secreted protein, acidic and rich in cysteine, adipocytokine, body mass index, coronary artery disease Introduction Osteonectin, also referred to as secreted protein, acidic and rich in cystein (SPARC), is an extracellular matrix (ECM)-associated glycoprotein first described by Termine et al. (1) as a major noncollagenous constituent of bovine bone. SPARC binds mineral to matrix and functions in the development of ossified and mineralized tissues. In addition to the classical function, it participates in tumor progression by promoting migration of neoplastic cells and neovascularization (2). The protein is also involved in wound healing by regulating the remodeling of matrix proteins and the facilitation of angiogenesis (3,4). In atherosclerotic lesion, SPARC is expressed in the vascular smooth muscle cells and is postulated to regulate the availability of dimeric forms of platelet-derived growth factor in vascular injury through the interaction with platelet-derived growth factor-b chain (5). Physiologically, SPARC expression is known in the heart, kidney, lung, gut, etc. A variety of cell types, such as osteoblasts, macrophages, fibroblasts, smooth muscle cells, and endothelial cells, expresses SPARC 388 OBESITY RESEARCH Vol. 9 No. 7 July 2001

2 mrna (3,6). Although the SPARC protein is also detected in plasma, its physiological role or clinical significance is not fully clear. To elucidate the molecular basis of obesity-related disorders, we performed the systemic analysis of adipose tissue expressed genes previously by a random sequencing (7). Surprisingly, adipocytes expressed many kinds of genes for secretory proteins, including matrix proteins. Plasminogen activator inhibitor (PAI)-1 regulates the activity of plasmin, which destroys the basement membrane during the growth of adipocytes. In obesity, the expression of PAI-1 is augmented in adipose tissue particularly intra-abdominal fat depot, resulting in increased circulating PAI-1 levels (8). The hypersecretion of PAI-1 from adipose tissue can be one of the causative factors for the development of thrombotic vascular diseases (9). We have also identified a novel collagen-like protein named adiponectin in human adipose tissue (10), which is abundantly present in systemic circulation and accumulates in injured vascular wall (11). In vitro studies revealed that adiponectin suppressed the expression of endothelial adhesion molecules, such as vascular cell adhesion molecule-1, endothelial-leukocyte adhesion molecule-1, and intracellular adhesion molecule-1 (12). In contrast, the plasma concentration of adiponectin is decreased in obesity (13). Thus, the hyposecretion of adiponectin in obesity may lead to the development of atherosclerosis. These findings imply that molecules secreted from adipose tissue directly affect vascular homeostasis, and the dysregulation of them in obesity may contribute to the development of atherosclerotic vascular disease (adipo-vascular axis) (9). We found SPARC cdna in a human adipose tissue library (7). Plasma SPARC can exert some effects on the migration or cell matrix interactions of smooth muscle cells and result in the remodeling of vascular walls. In the current study, we investigated the expression of SPARC in adipose tissues and the regulation of adipose and plasma SPARC in obesity. Research Methods and Procedures Northern Blot Analysis Human and mouse SPARC probes obtained by reverse transcription-polymerase chain reaction from human and mouse adipose tissue, respectively, were used for Northern blot analysis. Human total RNA of various tissues except adipose tissues was purchased from Clontech Laboratories (Palo Alto, CA). Total RNA from human adipose tissue of lean subjects was extracted with TRIZOL reagent kit (Life Technologies, Rockville, MD). Ten micrograms of total RNA was separated by 1% agarose gel containing 2.2 M formaldehyde. The RNA was transferred to nylon membranes (Hybond-N; Amersham Pharmacia Biotech, Piscataway, NJ). The membranes were hybridized in QuikHyb hybridization solution (Stratagene, La Jolla, CA), according to the manufacturer s protocol. The probes were labeled with [ -32P] d-ctp (3000 Ci/mmol; Amersham Pharmacia Biotech), using the randomprimer DNA-labeling system (Amersham Pharmacia Biotech). Mouse 3T3-L1 fibroblasts were grown in high-glucose Dulbecco s modified Eagle s medium (DMEM) containing 10% fetal calf serum (FCS). At 2 days postconfluent (day 0), the medium was returned to DMEM supplemented with 10% FCS, 5 g/ml insulin, 0.5 mm 1-methyl-3-isobutylxanthin, and 1 M dexamethazone as previously described (8). After 48 hours, the medium was replaced to DMEM supplemented with 10% FCS only. Total cellular RNA was extracted at preconfluent and on days 0, 3, 5, 7, and 10 after differentiation. Male C57BL/KsJ (db / m) and C57BL/KsJ (db /db ) mice (CLEA Japan, Osaka, Japan) were anesthetized by 5 mg/ml pentobarbital sodium salt and killed at 6, 12, and 18 weeks old after fasting overnight. Total RNA was extracted from mesenteric adipose tissue and abdominal subcutaneous adipose tissue (14). All procedures were conducted in accordance with the principles and guidelines established by the University Committee and National regulations for the care and use of laboratory animals. Measurement of Plasma Concentrations of SPARC We studied 88 adult Japanese subjects without coronary artery disease (CAD) who visited our hospital (62 men and 26 women; mean [ SD] age: years; body mass index [BMI]: 16 to 46 kg/m 2 ) with informed consent. Additionally, 31 subjects (20 men and 11 women) with CAD confirmed by coronary angiography were also investigated. None of them had renal failure. Subcutaneous fat area (SFA) and visceral fat area (VFA) were measured by crosssectional computed tomography at the umbilical level as described previously (15). Venous blood samples were collected after fasting overnight and human plasma SPARC was measured by duplicate assay using the SPARC EIA Kit (Takara Shuzo, Kyoto, Japan) (16). Briefly, plasma was applied onto ON-1 monoclonal antibodies coated 96-well plates. Then, peroxidase-labeled OSN4 2 mab was used as second antibody. The absorbance at 280 nm was measured and plasma SPARC concentration was calculated by using bovine SPARC as a standard. These two monoclonal antibodies can bind equally to both human and bovine SPARC (17). Plasma leptin (15) and adiponectin (13) were also measured as previously described. The significance of the difference between the mean values of the groups was evaluated by Student s t test or ANOVA with Fisher s protected least significant difference (PLSD) test. Results Through the systematic analysis of the gene expression profile of human adipose tissue, we have found that SPARC OBESITY RESEARCH Vol. 9 No. 7 July

3 cdna appeared frequently (7). We first examined tissue distribution of SPARC mrna in humans (Figure 1). SPARC mrna abundantly expressed in both mesenteric and subcutaneous adipose tissues as 2.2-kb major and 3.0-kb minor bands as reported previously (3). Expression of SPARC mrna was also confirmed in the lung. A trace of SPARC mrna expression was detected in the heart, brain, liver, kidney, and skeletal muscle, but not in the pancreas, consistent with previous reports (3). The expression in subcutaneous adipose tissue was strongest in all tissues examined. These results suggest that adipose tissues can be a primary source of SPARC in vivo. Because adipose tissue contains a variety of cells, such as fibroblasts, endotherial cells, smooth muscle cells, etc., in addition to adipocytes, it is important to evaluate the roles of adipocytes in the production of SPARC. We examined the SPARC mrna expression during adipocyte differentiation in 3T3-L1 cells (Figure 2). A low level of SPARC mrna was detected in undifferentiated 3T3-L1 fibroblasts as a single band consistent with previously report (3). The expression of SPARC mrna was markedly enhanced after exposure of 3T3-L1 fibroblasts to the differentiation medium. Thus, it was verified that the differentiated adipocytes produce SPARC. Next, we investigated the effect of fat accumulation on the SPARC expression in adipose tissues of obese model mice in vivo. The SPARC mrna signals were detected both in mesenteric adipose tissue (Figure 3A) and subcutaneous adipose tissue (Figure 3B). Obese db /db mice expressed more abundant SPARC mrna than did lean Figure 2. Expression of SPARC during the differentiation of 3T3-L1 fibroblasts into adipocytes. Five micrograms of total RNA of various tissues was resolved by electrophoresis and hybridized with mouse SPARC probe. The SPARC mrna levels were analyzed by Northern blot analysis. Ethidium bromide staining of 28S ribosomal RNA was used to normalized RNA loading. db / m mice at all stages in both adipose tissues. Although the lung and kidney expressed SPARC mrna in mice, the expression levels were not changed in obese mice. Although SPARC exists in plasma, the major source of this protein has not been investigated. The SPARC expression in adipose tissue increased in obese mice as mentioned above. We analyzed the plasma SPARC concentrations of 88 subjects in view of the adiposity to confirm whether the enhanced SPARC mrna of in adipose tissue may contrib- Figure 1. Tissue distribution of SPARC as tested by Northern blotting. Ten micrograms of total RNA of various tissues was resolved by electrophoresis and hybridized with human SPARC probe. Lanes: 1) heart, 2) whole brain, 3) lung, 4) liver, 5) kidney, 6) spleen, 7) pancreas, 8) skeletal muscle, 9) mesenteric adipose, and 10) subcutaneous adipose. Ethidium bromide staining was also shown. Figure 3. Expression of SPARC in adipose tissues of obese model mice. Male C57BL/KsJ (db / m) and C57BL/KsJ (db /db ) mice were killed at 6, 12, and 18 weeks. The SPARC mrna levels in mesenteric adipose (A) and subcutaneous adipose (B) were analyzed by Northern blot analysis. Ethidium bromide staining of 28S ribosomal RNA was used to normalized RNA loading. 390 OBESITY RESEARCH Vol. 9 No. 7 July 2001

4 ute the plasma concentrations. Plasma SPARC concentrations showed a significant positive correlation with BMI (r 0.27; p 0.01; Figure 4). SFA also had a significant correlation with plasma SPARC levels (r 0.38; p 0.01), but VFA did not (r 0.07; p not significant; data not shown). As to gender differences and aging, neither sex nor age influenced the plasma levels of SPARC. Next, among adipocytokines (8,9), we also measured the plasma leptin (15) and adiponectin (12,13). Plasma levels of leptin, known to be positively regulated by BMI (15), exhibited a significant correlation with plasma SPARC (r 0.24; p 0.05; data not shown). In contrast, plasma adiponectin, negatively controlled by BMI (12), showed a negative correlation with plasma SPARC (r 0.21; p 0.05; data not shown). In contrast, neither fasting plasma glucose (r 0.12; p 0.27) nor plasma insulin (r 0.11; p 0.33) presented a statistically significant correlation with plasma SPARC levels (data not shown). The above data suggest that adipose tissue could be recognized as the major source of SPARC. Evidence has revealed that SPARC is involved in the process of angiogenesis (4), and it is detected in atherosclerotic lesions (5,6). Next, we analyzed the plasma levels of SPARC in 31 patients with CAD, including 20 men and 11 women (mean [ SD] age: years; BMI [ SD]: kg/m 2 ) and compared them with 65 age- and BMI-matched subjects without CAD, including 53 men and 12 women (mean [ SD] age: years; BMI [ SD]: kg/m 2 ). Although no significant difference in BMI was observed between the two groups, VFA determined by computed tomography in the CAD ( ) group was significantly larger compared with the CAD ( ) group. The plasma level of leptin did not differ between the two groups (Table 1). As reported previously (12), plasma levels of adiponectin proved significantly lower in the CAD ( ) Table 1. Metabolic parameters (mean values SD) in Japanese subjects Parameter Non-CAD control subjects CAD subjects Patient number Men/women 53/12 20/11 Age BMI (kg/m 2 ) VFA (cm 2 ) * SFA (cm 2 ) sbp (mm Hg) * dbp (mm Hg) Cholesterol (mg/dl) * TG (mg/dl) High density lipoprotein (mg/dl) * Fasting plasma glucose (mg/dl) Fasting immunoreactive insulin ( U/mL) HbA1c (%) * Leptin (ng/ml) Adiponectin ( g/ml) * * p sbp, systolic blood pressure; dbp, diastolic blood pressure; TG, triglyceride; HbA1c, glycohemoglobin A1c. group than in the CAD ( ) group (mean [ SE]: vs g/ml; p 0.001; Table 1). Interestingly, plasma levels of SPARC were more elevated in the CAD ( ) group than in the CAD ( ) group (mean [ SE]: vs ng/ml; p 0.05; Figure 5). Figure 4. Scatter plots analyses of plasma SPARC concentrations in human subjects. Plasma SPRAC levels were plotted relative to BMI (kg/m 2 ; n 88). The line indicates the regression line. Discussion Although adipose tissue had been recognized as a relatively passive depot for lipid storage, recent reports, including our report, have established that adipocytes secrete various lines of molecules (adipocytokines) to elicit the obesity-associated diseases (7 9,18 20). Through systematic survey of active genes in the human adipose tissue using random cdna sequencing, we have identified SPARC in adipose cdna library. In this study, we clarified that the expression of SPARC was strongest in adipose tissue in all tissues examined. The 3T3-L1 preadipocytes expressed a low level of SPARC mrna and differentiated 3T3-L1 cells dramatically in- OBESITY RESEARCH Vol. 9 No. 7 July

5 Figure 5. Plasma SPARC concentrations in human subjects with CAD. The average plasma SPARC levels were shown in agematched and BMI-matched subjects without (n 65) and with CAD (n 31). Error bars represent the SE. A score 0.05 denotes statistical significance. creased its expression. In lean mice, there was no significant difference between mesenteric fat and subcutaneous fat tissue. Although the augmented expression of SPARC mrna was remarkable in subcutaneous fat at an early stage of obesity, SPARC mrna expression was enhanced in obese mice compared with their lean littermates throughout the period. Because these model mice are devoid of normal leptin signaling, we cannot rule out the influence of leptin. In humans, plasma level of SPARC significantly correlated to adiposity. Taken together, adipose tissues can be a main source of SPARC in vivo. In obesity, some of adipocyte-derived factors, such as leptin, PAI-1, and tumor necrosis factor- (8,17,20), are increased in plasma. In this report, we proved that plasma concentrations of SPARC showed a positive correlation with BMI. Furthermore, plasma SPARC levels also had a close correlation with SFA but not with VFA. Because SFA reflects more correctly the whole body fat mass, these results sustain the conception that adipose tissues mainly contribute the plasma levels of SPARC. The significance of the augmented SPARC in obesity remains obscure to date. The possibility is that SPARC affects the expression of matrix components and the activity of enzymes regulating cellular interactions with ECM to perform angiogenesis and neovascularization (2,4,5). This hypothesis is supported by the fact that adipose tissue can alter its size according to the differentiation and proliferation through the deposition or assembly of ECM proteins. In the current study, we have shown that the plasma level of SPARC is elevated in the patients with CAD. The detailed mechanism of increased SPARC in CAD subjects remains obscure. Although BMIs were matched, CAD patients have larger VFAs than do non-cad patients. Increased visceral fat may be a source of SPARC in CAD patients. We cannot exclude the possibility that SPARC may be ectopically expressed in CAD patients, such as in the activated platelets, macrophages, and smooth muscle cells in the atherosclerotic region (5,6). The roles of increased SPARC in CAD patients remain speculative. SPARC enhances detachment and subsequent migration of vascular smooth cells, leading to the invasion of the cells into intima (6). In vascular endothelial cells, exogenous SPARC facilitates endothelial permeability resulting in barrier dysfunction (21). SPARC inhibits endothelial cell proliferation induced by vascular endothelium growth factor (VEGF) through directly binding to VEGF and hindering the association of VEGF with its cell surface receptor Flt-1 (22). Overproduced SPARC prevents the proliferation of endothelial cells, failing to repair the injured vessels. The stimulating effect of SPARC on PAI-1 production in endothelial cells precipitates an acceleration of atherosclerosis (23). Considering these effects, SPARC in atherosclerotic regions may promote the process of atherosclerosis. In this point, it will be of interest whether plasma levels of SPARC are elevated in the patients with other atherosclerotic disease such as cerebrovascular disease or arteriosclerosis obliterans. Additionally, the regulation of SPARC was entirely opposite to that of adiponectin, showing a significant negative correlation between their plasma levels as described above. Both molecules share the following interesting properties: secreted from adipocytes, binding to various collagens in the ECMs, and observed in atherosclerotic regions (3,6,11). Increased SPARC and decreased adiponectin in obesity can exert cooperatively some roles in the development of atherosclerosis. Acknowledgments This work was supported in part by the Research for the Future Program from The Japan Society for the Promotion of Science: JSPS-RFTF97L00801 and Grants-in-Aid from the Ministry of Education, Science, Sports and Culture of Japan (Grants , , , and ). References 1. Termine JD, Kleinman HK, Whitson SW, Conn KM, Mc- Garvey ML, Martin GR. Osteonectin, a bone-specific protein linking mineral to collagen. Cell. 1981;26: Ledda MF, Adris S, Bravo AI, et al. Suppression of SPARC expression by antisense RNA abrogates the tumorigenicity of human melanoma cells. Nat Med. 1997;3: Lane TF, Saga FH. The biology of SPARC, a protein that modulates cell-matrix interaction. FASEB J. 1994;8: Lane TF, Iruela-Arispe ML, Sage EH. Regulation of gene expression by SPARC during angiogenesis in vitro. J Biol Chem. 1992;267: OBESITY RESEARCH Vol. 9 No. 7 July 2001

6 5. Raines EW, Lane TF, Iruela-Arispe ML, Ross R, Saga EH. The extracellular glycoprotein SPARC interacts with plateletderived growth factor (PDGF)-AB and -BB and inhibits the binding of PDGF to its receptors. Proc Natl Acad Sci USA. 1992;89: Hirota S, Imakita M, Kohri K, et al. Expression of osteopontin messenger RNA by macrophages in atherosclerotic plaques. Am J Pathol. 1993;143: Maeda K, Okubo K, Shimomura I, Mizuo K, Matsuzawa Y, Matsubara K. Analysis of an expression profile of genes in the human adipose tissue. Gene. 1997;190: Shimomura I, Funahashi T, Takahashi M, et al. Enhanced expression of PAI-1 in visceral fat: possible contributor to vascular disease in obesity. Nat Med. 1996;2: Funahashi T, Nakamura T, Shimomura I, et al. Role of adipocytokines on the pathogenesis of atherosclerosis in visceral obesity. Intern Med. 1999;38: Maeda K, Okubo K, Shimomura I, Funahashi T, Matsuzawa Y, Matsubara K. cdna cloning and expression of a novel adipose specific collagen-like factor, apm1 (adipose most abundant gene transcript 1). Biochem Biophys Res Commun. 1996;221: Okamoto Y, Arita Y, Nishida M, et al. An adipocyte-derived plasma protein, adiponectin, adheres to injured vascular walls. Horm Metab Res. 2000;32: Ouchi N, Kihara S, Arita Y, et al. Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein, adiponectin. Circulation. 1999;100: Arita Y, Kihara S, Ouchi N, et al. Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun. 1999;257: Shimomura I, Takahashi M, Tokunaga K, et al. Rapid enhancement of acyl-coa synthetase, LPL, and Glur-4 mrnas in adipose tissue of VMH rats. Am J Physiol. 1996; 270:E995 E Takahashi M, Funahashi T, Shimomura I, Miyaoka K, Matsuzawa Y. Plasma leptin levels and fat distribution. Horm Metab Res. 1996;28: Shiba H, Nakamura S, Shirakawa M, et al. Effects of basic fibroblast growth factor on proliferation, the expression of osteonectin (SPARC) and alkaline phosphatase, and calcification in cultures of human pulp cells. Dev Biol. 1995;170: Kamihagi K, Katayama M, Ouchi R, Kato I. Osteonectin/ SPARC regulates cellular secretion rates of fibronectin and laminin extracellular matrix proteins. Biochem Biophys Res Commun. 1994;200: White RT, Damm D, Hancock N, et al. Human adipsin is identical to complement factor D and is expressed at high levels in adipose tissue. J Biol Chem. 1992;267: Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature. 1994;372: Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest. 1995;95: Goldbulm SE, Ding X, Funk SE, Saga EH. SPARC (secreted protein acidic and rich in cystein) regulates endothelial cell shape and barrier function. Proc Natl Acad Sci USA. 1994;91: Kupprion C, Motamed K, Saga EH. SPARC inhibits the mitogenic effect of vascular endothelial growth factor on microvascular endothelial cells. J Biol Chem. 1998; 273: Hasselaar P, Loskutoff DJ, Sawdey M, Sage EH. SPARC induces the expression of type 1 plasminogen activator inhibitor in cultured bovine aortic endothelial cells. J Biol Chem. 1991;266: OBESITY RESEARCH Vol. 9 No. 7 July