Role of Meltrin (ADAM12) in Obesity Induced by High- Fat Diet
http://www.100md.com
内分泌学杂志 2005年第4期
Department of Growth Regulation, Institute for Frontier Medical Sciences, Kyoto University (M.M., T.K., A.S.-F.), Sakyo-ku, Kyoto 606-8507, Japan; Mochida Pharmaceutical Co. Ltd. (K.S.), Shizuoka 412-8524, Japan; and Japan Society for the Promotion of Science (M.M.), Tokyo 102-8471, Japan
Address all correspondence and requests for reprints to: Dr. Atsuko Sehara-Fujisawa, Department of Growth Regulation, Institute for Frontier Medical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: asehara@frontier.kyoto-u.ac.jp.
Abstract
Meltrin is a member of the metalloprotease-disintegrin (ADAM) family. In this paper we demonstrate that meltrin is involved in the development of white adipose tissue. Compared with wild-type mice, meltrin –/– mice displayed moderate resistance to weight gain induced by a high-fat diet, mainly because of an impaired increase in the number of adipocytes. There was no obvious difference in adipocyte size between wild-type and meltrin –/– mice, suggesting normal maturation of adipocytes of the latter under a high-fat diet. Embryonic fibroblasts and stromal-vascular cells lacking meltrin exhibited impaired cell proliferation upon adipogenic stimulation, which was accompanied by moderate defects in adipose differentiation. Addition of culture medium conditioned with wild-type cells in an early phase of adipose differentiation did not restore the defects in the meltrin –/– cells. These results uncover the involvement of meltrin in the development of obesity and in adipogenic cell proliferation.
Introduction
THERE ARE TWO types of adipose tissue in the mammalian body. White adipose tissue (WAT) is distributed in many locations, stores excess energy as triglycerides, and releases fatty acids in response to energy requirements. The second type of adipose tissue is brown adipose tissue (BAT), which is the main thermogenic tissue in rodents. In BAT, fatty acid oxidation stimulated by the sympathetic nervous system generates heat through the induction of uncoupling protein-1. The developmental patterns of these tissues are quite different. BAT develops during fetal stages and essentially acquires all the features of mature tissue at birth, when nonshivering thermogenesis is required (1). In contrast, the development of WAT continues after birth, and its mass increases during postnatal life (2).
In this study we report the roles of meltrin in obesity, a pathological development of WAT. Meltrin (ADAM12) is a member of the metalloprotease-disintegrin (ADAM) family of proteins that contain metalloprotease and disintegrin domains (3). ADAMs play important roles in fertilization (4, 5, 6) and various aspects of morphogenesis (3, 7, 8) and are implicated in certain pathogenetic processes (9, 10). Genetic and biochemical studies have revealed that some ADAMs participate in the ectodomain shedding of various membrane-anchored proteins (by proteolysis of these proteins at the extracellular juxtamembrane region), including growth factors, intercellular signaling molecules, and adhesion molecules (11, 12, 13). Meltrin modulates myotube formation in vitro (3) and in vivo (14, 15). Two candidate substrates for meltrin protease have been reported to date: heparin-binding epidermal growth factor (HB-EGF) (14) and IGF-binding protein-3 (IGFBP-3) (16, 17), which regulates the activation of IGF-I. Evidence suggests that different ADAM proteases participate in the phorbol 12-myristate 13-acetate (PMA)-induced ectodomain shedding of membrane-anchored ErbB ligands, including HB-EGF (9, 18), TGF- (8), and neuregulin (19). In addition to TNF--converting enzyme (TACE/ADAM17), meltrin and meltrin ? (ADAM19) are examples of such proteases, although their expression patterns, ligand specificity, and regulation of protease activity differ (9, 20, 21).
We reported previously that during embryogenesis, some meltrin –/– mice display impaired development of interscapular BAT and of the skeletal muscles that surround BAT (14). The increase in ectodomain shedding of HB-EGF in response to PMA was markedly reduced in meltrin –/– embryonic fibroblasts. In contrast, Kawaguchi et al. (22) reported that transgenic mice overexpressing a placental isoform of human meltrin exhibit increased adipogenesis. Although the lack of meltrin also affects WAT formation in some meltrin –/– mouse embryos, decreased formation of WAT is not as prominent as that of BAT and recovers after birth, suggesting that compensatory mechanisms restore the roles of meltrin in the surviving population of meltrin –/– mice during WAT formation in utero or before weaning (Kurisaki, T., and A. Sehara-Fujisawa, unpublished observations). Alternatively, the slight decrease in WAT formation in meltrin –/– mice and the enhanced formation of WAT in transgenic mice expressing a placental isoform of human meltrin could be secondary effects of other phenotypes, such as systemic growth retardation or aberrant placental development. To determine whether meltrin plays a direct role in WAT development and to reveal its role in adipogenesis, we examined the effects of meltrin deficiency on the induction of obesity by a high-fat diet and on adipogenesis of stromal-vascular (S-V) cells or embryonic fibroblasts cultured in vitro, both of which eliminate maternal factors and other compensatory factors in utero and before weaning. As a result, we demonstrated the involvement of meltrin in obesity induced by a high-fat diet. Meltrin participates mainly in increasing the number of adipogenic cells during the progression of obesity and in cell proliferation at an early stage of adipogenesis in S-V cells and embryonic fibroblasts, which are critical for the expansion of adipocytes in vivo and in vitro.
Materials and Methods
Animal experiments
The meltrin –/– mouse line was generated as described previously (14). The initial chimeras were backcrossed to C57BL/6J more than 12 times. The numbers of animals used in experiments are mentioned in Results and the figure legends. Animals were maintained in a temperature-controlled facility with a 12-h light, 12-h dark cycle. When male mice were 4 wk of age, C57BL/6J wild-type and meltrin –/– mice were divided into two groups. One group was given a high-fat diet containing 60% fat (Oriental Yeast Co. Ltd., Tokyo, Japan), and the second group was given a normal (10% fat) diet (Research Diets, Inc., New Brunswick, NJ). The diets contained 5.2 and 3.8 kcal/g. Body weight was recorded every week, and food intake for the high-fat diet was determined every second day. For the measurement of metabolic parameters, mice were fed the normal or high-fat diet for 12 wk. Blood was collected from the heart after an overnight fast. Plasma triglycerides, total cholesterol, and nonesterified fatty acids (NEFA) were measured using enzymatic assays: the triglyceride E test (Wako Pure Chemical Industries Ltd., Osaka, Japan), the cholesterol E test (Wako Pure Chemical Industries Ltd.), and the NEFA test (Wako Pure Chemical Industries Ltd.). For the determination of plasma leptin, adiponectin, and insulin (INS) levels, ELISA kits were purchased from Morinaga (Kanagawa, Japan), Otsuka (Tokyo, Japan), and Shibayagi (Gunma, Japan), respectively. Glucose tolerance testing and INS tolerance testing were performed with the C57BL/6J and meltrin –/– mice fed the normal diet at 25–35 wk of age or with the C57BL/6J and meltrin –/– mice fed the high-fat diet for 10 wk. Glucose at 1 g/kg body weight or INS at 0.75 U/kg was injected ip after an overnight fast. Blood was collected from the tail vein. Glucose quantification was performed with the One Touch Ultra blood glucose-monitoring system (Johnson & Johnson, Tokyo, Japan). For measurement of INS concentrations before and after glucose injection, glucose at 1 g/kg body weight was injected ip into the C57BL/6J and meltrin –/– mice fed the normal diet after overnight fasting. At each time point, blood was collected from an intraorbital vein. The serum INS concentration was determined by ELISA.
Histological analysis of adipose tissues and liver
Pieces of adipose tissues and livers were fixed in 4% paraformaldehyde (PFA), dehydrated in ethanol, embedded in paraffin, and sectioned at a thickness of 4 μm. Sections were deparaffinized, rehydrated, and stained with hematoxylin and eosin. The number of adipocytes was determined as described previously (23). Briefly, adipocytes in the nine slices for each animal (n = 3) were counted, and the average cell volume was determined. Relative cell numbers were calculated based on the average cell volume and tissue weights. For Oil Red-O staining, livers were immediately embedded in tissue-freezing medium (Tissue-Tek OCT compound, Miles, Inc., Elkhart, IN), and sections 7 μm thick were stained with Oil Red-O.
RNA preparation and real-time PCR
Total RNA was prepared from tissues of adult male mice with RNeasy Lipid Tissue Mini (Qiagen, Valencia, CA) in accordance with the manufacturer’s instructions. For the isolation of total RNA from mouse embryonic fibroblasts (MEFs), RNeasy mini (Qiagen) was used. Meltrin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA expression were determined by RT, followed by real-time TaqMan PCR analysis. Primers and probes of meltrin and GAPDH were purchased from Applied Biosystems (Foster City, CA; Mm00475719_m1and P/N 4308313, respectively). The primers used to detect mRNA of IGFBP-3 were 5'-GACACCCAGAACTTCTCCTCC-3' and 5'-CATACTTGTCCACACACCAGC-3'.
Preparation of S-V cells and immunofluorescent staining
Inguinal fat pads were harvested from 4- to 5-wk-old, wild-type or meltrin –/– male mice. After blood was washed out of the tissues, they were minced and digested with 1 mg/ml collagenase type I (Worthington Biochemical Corp., Freehold, NJ) for 30 min at 37 C. Cells were filtered through 200-μm pore size nylon meshes. The S-V cells were separated from adipocytes by centrifugation and washed with DMEM (Nacalai Tesque, Inc., Kyoto, Japan) supplemented with 10% heat-inactivated fetal bovine serum (FBS). S-V cells were plated and propagated to confluence in DMEM supplemented with 10% heat-inactivated FBS, 50 U/ml penicillin, and 50 μg/ml streptomycin. Two days later, the medium was replaced with differentiation induction medium (DIM) containing 1 μM dexamethasone (DEX; Sigma-Aldrich Corp., St. Louis, MO), 10 μg/ml INS (Sigma-Aldrich Corp.), and 0.5 mM 3-isobutyl-1-methylxanthine (MIX; Nacalai Tesque, Inc.) with 10% heat-inactivated FBS. After 24 h, cells were washed with PBS twice and fixed with 4% PFA for 30 min, followed by treatment with ice-cold 0.2% Triton X-20 for 5 min, then stained with anti-Ki-67 (BD PharMingen, San Diego, CA; 1:100 dilution) as a first antibody and antimouse IgG-Alexa 488-conjugated antibody (Molecular Probes, Eugene, OR; 1:400 dilution) as a secondary antibody. 4',6-Diamidino-2-phenylindole (DAPI; 0.2 μg/ml; Molecular Probes) was used for staining of all nuclei. Photographs were taken on a Zeiss fluorescent microscope (New York, NY) with MetaMorph software (Universal Imaging Corp., Brock and Michelsen, Briker?d, Denmark). The ratio of Ki-67-positive cells to DAPI-positive cells was determined.
Preparation of MEFs and induction of adipogenesis
Primary embryonic fibroblasts were prepared from 14.5 d postcoitus embryos. Cells (8 x 105) were plated on 12-well plastic dishes and cultured at 37 C in standard medium: -MEM (Nacalai Tesque, Inc.) supplemented with 10% heat-inactivated FBS, 50 U/ml penicillin, and 50 μg/ml streptomycin. Cells were propagated to confluence. Two days later, the medium was replaced with DIM with 10% heat-inactivated FBS. After 2 d, this medium was replaced with maturation medium, which contains 10 μg/ml INS. After another 3 d, the maturation medium was replaced with standard medium. For evaluation of the effect of HB-EGF, wild-type MEFs were cultured in DIM or -MEM containing 10 μg/ml INS with or without 100 ng/ml HB-EGF (R&D Systems, Inc., Minneapolis, MN). After 2 d, the medium was replaced with a maturation medium with or without HB-EGF. After another 3 d, the maturation medium was replaced with standard medium. In the medium exchange experiment, we inducted meltrin –/– and wild-type MEFs with DIM as described above (d 0). Then, culture medium of wild-type MEFs was replaced with meltrin –/–-conditioned medium and vice versa on d 1, 3, and 6. The medium was replaced with fresh maturation medium on d 2 and with standard medium on d 5. After 8 d, cytoplasmic lipid accumulation was observed by brightfield microscopy with Oil Red-O staining. The composition of each medium is shown in Tables 1 and 2 in detail. Oil Red-O staining was performed as follows. Cells were washed with PBS, fixed with 4% PFA for 10 min, then stained with 60% filtered Oil Red-O stock solution (0.15 g Oil Red-O in 50 ml isopropanol) for 30 min, washed with 60% isopropanol, then briefly washed with PBS twice before being visualized. To quantify the amount of lipid, stained oil was eluted with isopropanol, and the absorbance at a wavelength of 510 nm was read after dilution of the eluate to a linear range. To determine the number of cells, on d 0 (before induction with DIM) and d 3 (after induction), cells were trypsinized and well suspended with -MEM with 10% heat-inactivated FBS. Cells were counted using Neubauer’s hemocytometer (Erma, Tokyo, Japan) under light microscopy. Experiments were performed in triplicate with four different litters of mice, and values are shown as the mean ± SEM.
TABLE 1. Composition of each culture medium
TABLE 2. Cell culture conditions in the medium exchange experiment
Western blot
In preparation for Western blotting, 1.5 x 106 MEFs were placed in six-well dishes and cultured in a standard medium. Two days later, the medium was replaced with Opti-MEM (Invitrogen Life Technologies, Inc., Carlsbad, CA) reduced serum with either DIM or 10 μg/ml INS. After 18 or 36 h, the conditioned medium was concentrated 10-fold using a centrifuge tube filtration unit (Amicon Ultra-4 10,000 MWCO, Millipore Corp., Bedford, MA). Western blotting was performed as described previously (19, 21). Goat polyclonal antibody against IGFBP-3 (sc-6004) was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Statistical analysis
Data are expressed as the mean ± SEM. A two-tailed t test was used to calculate P values.
Results
Meltrin –/– mice are resistant to induction of obesity by a high-fat diet
To explore the role of meltrin in the development of WAT in adult mice, we examined the body weight gain of meltrin –/– mice on either a high-fat diet or a normal diet. After weaning, meltrin –/– mice fed a normal diet gained weight similarly to wild-type mice. However, meltrin –/– mice on a high-fat diet were protected against weight gain to some extent, despite having the same food intake as their wild-type counterparts (Fig. 1, A and B). Representative specimens of wild-type and meltrin –/– mice given the high-fat diet for 12 wk are shown in Fig. 1C. The decreased weight gain of meltrin –/– mice fed the high-fat diet was mainly reflected by the smaller mass and lower weight of their WAT compared with those of wild-type mice (Fig. 1, D and E). Resistance of meltrin –/– mice to high-fat diet-induced obesity was also shown by the normal appearance of their livers compared with those of the wild-type mice, which displayed hepatic steatosis. Furthermore, the interscapular BAT of meltrin –/– mice on the high-fat diet was darker and had less mass of WAT surrounding it than that of wild-type mice (Fig. 1D). These results indicate that meltrin –/– mice were resistant to the obesity induced by a high-fat diet.
FIG. 1. Decreased body weight gain of meltrin –/– mice on a high-fat diet. A, Body weight curves for male mice on normal or high-fat diets. Body weights were recorded weekly in wild-type (WT) and meltrin –/– (Mel.–/–) mice fed either a normal diet (ND) or a high-fat diet (HFD). Only on the high-fat diet did meltrin –/– mice have lower body weights than WT mice. Values are expressed as the mean ± SEM. WT-HFD, n = 8; WT-ND, n = 6; Mel.–/– ND, n = 6; Mel.–/– NFD, n = 6. Asterisks indicate a statistically significant difference (P < 0.05) between high fat-fed WT and meltrin –/– mice. B, Food intake was recorded every second day and was the same in WT and meltrin –/– mice (10.5 ± 0.1 and 10.7 ± 0.2 kcal/d, respectively). C, Morphology of 15-wk-old mice fed a high-fat diet. D, Morphology of interscapular BAT, liver, and inguinal WAT in 16-wk-old mice fed a high-fat diet. Scale bar, 1 cm. E, Tissue weights in mice after 12 wk on a high fat or a normal diet, expressed as the mean ± SEM (n = 6). Asterisks indicate statistically significant difference (P < 0.05) between two experimental groups.
Glucose homeostasis of meltrin –/– mice fed a high- fat diet
Defects in glucose homeostasis are important factors causing obesity. To evaluate the metabolic difference between wild-type and meltrin –/– mice, we performed INS tolerance and glucose tolerance tests on these mice fed a normal diet or a high-fat diet. In a normal diet condition, increased INS sensitivity was observed in meltrin –/– mice (Fig. 2A). Hypoglycemia was evident 30 and 60 min after INS injection in both groups of mice. Wild-type mice more or less recovered their plasma glucose concentrations by 120 min after injection, whereas meltrin –/– mice remained hypoglycemic (P < 0.05). Plasma triglyceride and NEFA levels were also significantly lower in meltrin –/– mice fed a normal diet than in wild-type (Table 3). Thus, such altered metabolic profiles of meltrin –/– mice fed a normal diet might be associated with the resistance of meltrin –/– mice to high-fat diet-induced weight gain. In contrast, no statistically significant difference was found between wild-type and meltrin –/– mice 10 wk after a high-fat diet. The glucose clearance rates after ip glucose injection in wild-type and meltrin –/– mice were almost the same whether animals were fed a normal diet or a high-fat diet (Fig. 2B). There was no difference in serum INS concentrations before and after glucose injection between wild-type and meltrin –/– mice (Fig. 2C).
FIG. 2. A, INS tolerance test in wild-type (WT) and meltrin –/– (Mel.–/–) mice fed a normal or a high-fat diet. INS (0.75 U/kg) was injected into wild-type and meltrin –/– 25- to 35-wk-old mice fed a normal diet or 15-wk-old mice fed a high-fat diet for 10 wk after overnight fasting. In each experiment, values are expressed as the mean ± SEM. ND, Normal diet (WT, n = 14; meltrin –/–, n = 9); HFD, high-fat diet (WT, n = 7; meltrin –/–, n = 6). B, Glucose tolerance test in wild-type (WT) and meltrin –/– (Mel.–/–) mice fed a normal or a high-fat diet. Glucose (1 g/kg) was injected ip into wild-type and meltrin –/– 25- to 35-wk-old mice fed a normal diet or 15-wk-old mice fed a high-fat diet for 10 wk after overnight fasting. In each experiment, values are expressed as the mean ± SEM. ND, Normal diet (WT, n = 9; meltrin –/–, n = 5); HFD, high-fat diet (WT, n = 7; meltrin –/–, n = 6). C, INS concentrations pre- and postinjection of glucose. Glucose (1 g/kg) was injected ip into wild-type and meltrin –/– 25- to 35-wk-old mice fed a normal diet after overnight fasting. Before and after 15-min glucose injection, blood was collected from an intraorbital vein. The serum INS concentration was determined. In each experiment, values are expressed as the mean ± SEM. WT, n = 5; meltrin –/–, n = 4.
TABLE 3. Metabolic parameters in wild-type (WT) and meltrin –/– (Mel.–/–) mice
Histological analysis of meltrin –/– mice fed a high fat diet
We performed a histological analysis of BAT, liver, and inguinal, epididymal, and retrorenal WAT of wild-type and meltrin –/– mice fed a normal or a high-fat diet. In the inguinal WAT, the adipocytes of wild-type and meltrin –/– mice on the normal diet were roughly the same size. On the high-fat diet, the adipocytes in inguinal WAT of both types of mice increased in size due to intracellular lipid accumulation (Fig. 3A). These lipid accumulation patterns were the same in retrorenal and epididymal WAT (data not shown). This indicates that fat storage in adipocytes was normal in meltrin –/– mice.
FIG. 3. A, Histological analysis of adipose tissues. Inguinal WAT and interscapular BAT sections were stained with hematoxylin and eosin. Magnifications to show morphology of inguinal WAT and BAT are x10 and x20, respectively. Scale bar, 100 μm. WT, Wild-type mice; Mel.–/–, meltrin –/– mice; HFD, high-fat diet; ND, normal diet. Average cell volumes for WT ND and Mel.–/– ND iguinal WAT were 1.51 x 104 ± 0.12 x 104 and 1.90 x 104 ± 0.44 x 104 μm3, respectively. Average cell volumes for WT HFD and Mel.–/– HFD inguinal WAT were 2.37 x 105 ± 0.30 x 105 and 2.07 x 105 ± 0.46 x 105 μm3, respectively. Data from nine slices for each animal (n = 3) were analyzed. B, Histological analysis of liver. Liver sections were stained with either hematoxylin and eosin (H & E) or Oil Red-O. Magnification for morphological examination, 10x and 20x, respectively. Scale bar, 100 μm. Abbreviations are explained in A. C, Relative cell numbers in wild-type and meltrin –/– WATs. The cell numbers in wild-type and meltrin –/– WATs were estimated as described in Materials and Methods. Data from nine slices for each animal (n = 3) were analyzed. Asterisks indicate a statistically significant difference (P < 0.05) between two experimental groups. Abbreviations are explained in A. D, Meltrin mRNA expression in WAT and liver of mice fed a normal diet. Total RNA was prepared from tissues of adult male mice, and the mRNA level of meltrin was determined by real-time PCR. Levels of mRNA were normalized to that of GAPDH. Values from epididymal WAT were set at 1. In each experiment, n = 3, and values are expressed as the mean ± SEM. E, High-fat diet-induced expression of meltrin mRNA in liver and epididymal WAT. Total RNA was prepared from tissues of adult male mice fed either a normal diet () or a high-fat diet (), and the mRNA level of meltrin was determined by real-time PCR. Levels of mRNA were normalized to that of GAPDH. Values from epididymal WAT of mice fed a normal diet were set at 1. In each experiment, n = 3, and values are expressed as the mean ± SEM. TA, Tibialis anterior muscle. Asterisks indicate a statistically significant difference (P < 0.05) between two experimental groups.
The adipocytes in the BAT of wild-type mice on a high-fat diet accumulated lipid in large droplets. However, the BAT adipocytes of meltrin –/– mice accumulated less lipid in smaller droplets after being fed a high-fat diet (Fig. 3A), accounting in part for the macroscopically darker color of the BAT of meltrin –/– mice. Hematoxylin and eosin staining and Oil Red-O staining of liver sections confirmed the presence of steatosis in the wild-type mice fed a high-fat diet, whereas there was no sign of steatosis in the livers of meltrin –/– mice on the same diet (Fig. 3B).
Because the fat pads of meltrin –/– mice on a high-fat diet were smaller than those of wild-type mice despite the similar size of individual adipocytes (Figs. 1E and 3A), we counted the number of cells in representative populations of white adipocytes from the inguinal, epididymal, and retrorenal WAT. There were fewer cells in meltrin –/– mice fed a high-fat diet than in wild-type mice on the same diet (Fig. 3C), but cell numbers were approximately the same in mice fed a normal diet. These data indicate that meltrin is involved in increasing the number of adipocytes in mice fed a high-fat diet.
Meltrin is expressed in various adipose tissues
We used quantitative real-time PCR to examine relative meltrin expression in several tissues of wild-type mice fed a normal diet. The level of meltrin mRNA was almost the same among epididymal and retrorenal WAT and interscapular BAT. The meltrin mRNA level in inguinal WAT was 12 times higher than that in other adipose tissues. Meltrin mRNA expression was much lower or was not recorded in the liver and tibialis anterior muscle (Fig. 3, D and E). This suggests that meltrin is expressed in adipose tissues and is required for the increase in cell number induced by a high-fat diet.
Although meltrin mRNA expression was low in the liver in the normal diet condition, meltrin –/– mice were resistant to hepatic steatosis, which was observed in wild-type mice fed a high-fat diet (Figs. 1D and 3B). We examined the influence of a high-fat diet on the expression of meltrin mRNA in the liver, tibialis anterior muscle, and epididymal WAT. As a result, meltrin mRNA expression was up-regulated in epididymal WAT (2-fold) and liver (12-fold), but not in muscle, when mice were fed a high-fat diet (Fig. 3E). Up-regulation of meltrin mRNA in the liver on a high-fat diet might be involved in hepatic steatosis.
S-V cells lacking meltrin have a defect(s) in cell proliferation during adipogenesis
We next asked whether decreased cell number in meltrin –/– adipose tissue after a high-fat diet is due to impaired proliferation of adipogenic cells in meltrin –/– mice. We found the most prominent difference in the development of inguinal WAT between wild-type and meltrin –/– mice after a high-fat diet. Previous studies indicate that inguinal WAT has a greater proliferation capacity than other WAT, such as epididymal fat pads (2). To analyze cell proliferation under adipogenic conditions in vitro, we isolated S-V cells, which contain preadipocytes and precursors of them, from inguinal WAT of wild-type and meltrin –/– mice. The S-V cells were propagated to confluence and induced to differentiate into adipocytes in DIM 2 d after confluence. After 24 h, cellular proliferation was assessed by immunofluorescent staining with an antibody against Ki-67 nuclear antigen, a marker of cellular proliferation, followed by quantification of the rate of Ki-67-positive cells. The proliferation rates of cells from wild-type and meltrin –/– mice were relatively low (30%) without DIM. Upon treatment with DIM, the proliferation rates of wild-type and meltrin –/– S-V cells were enhanced to 65% and 53%, respectively (Fig. 4). This modestly decreased proliferation rate of meltrin –/– S-V cells after DIM treatment suggests a role for meltrin in cell proliferation in vitro.
FIG. 4. A, Ki-67-positive cells in S-V cells after adipogenic induction. S-V cells from wild-type (WT) or meltrin –/– (Mel.–/–) mice were incubated with or without DIM for 24 h, and Ki-67-positive cells were visualized. DAPI stain was used for nuclei. The microscopic images show representative fields of individual treatments for three independent experiments. B, The rate of proliferating cells in S-V cells. Data from five fields for each well were analyzed. The ratios of Ki-67-positive cells to DAPI-positive cells were determined. Experiments were performed in duplicate (three independent experiments were carried out), and values are shown as the mean ± SEM. An asterisk indicates a statistically significant difference (P < 0.05) between two experimental groups.
MEFs lacking meltrin have a defect(s) in cell proliferation during adipogenesis
The conversion of MEFs to adipocytes after hormonal stimulation has been extensively studied as a means of identifying key regulatory factors in adipogenesis (24, 25). We prepared primary embryonic fibroblasts from wild-type and meltrin –/– mice to examine their adipogenic properties. The MEFs were induced to differentiate into adipocytes in DIM. Meltrin –/– MEFs showed a modestly reduced capacity to differentiate into adipocytes compared with wild-type MEFs. It is noteworthy that Oil Red-O-negative space was more pronounced in the culture of meltrin –/– MEFs than in that of wild-type cells (Fig. 6A). Quantification of triglyceride production confirmed a decreased differentiation in meltrin –/– MEFs (Fig. 6B). We performed quantitative real-time PCR to examine meltrin mRNA expression in each step of adipogenesis. Meltrin mRNA expression was transiently increased after induction of adipogenesis (highest 2 d after adipogenic differentiation) and decreased during the process of terminal differentiation (Fig. 6C). These results suggested that meltrin might have an essential role in the early steps of adipogenesis, including the production of adipogenic precursors and their mitotic clonal expansion, that precede terminal differentiation. To test this idea, we evaluated the increase in cell numbers during differentiation in wild-type and meltrin –/– MEFs. We found a significant increase in cell numbers in wild-type, but not meltrin –/–, MEFs 3 d after adipogenic differentiation (Fig. 6D). This impaired increase in the number of cells during adipogenic induction also suggests a role for meltrin in cell proliferation, which occurs in the early stages of differentiation and would affect the number of differentiated adipocytes produced.
FIG. 6. A, Adipogenic differentiation of primary MEFs prepared from wild-type and meltrin –/– mice. Primary MEFs from wild-type (WT) or meltrin –/– (Mel. –/–) mice were incubated in the absence (–) or presence of DIM. (–).ME and DIM.ME, Cultures in conditioned medium prepared from meltrin –/– and wild-type MEFs, respectively. After 8 d of differentiation, cells were fixed and stained with Oil Red-O. B, Triglyceride accumulation in wild-type and meltrin –/– MEFs after adipogenic induction. Triglyceride accumulation was expressed as absorbance at an OD of 510 nm. In each experiment, n = 3, and values were expressed as the mean ± SEM. Three independent experiments were carried out, and representative data are shown. Asterisks indicate a statistically significant difference (P < 0.05) between two experimental groups. C, Meltrin mRNA expression in MEFs during adipogenesis. Total RNA was prepared from wild-type MEFs, and the mRNA level of meltrin was determined by real-time PCR. Levels of mRNA were normalized to that of GAPDH. Values from growing cells were set at 1. In each experiment, n = 3, and values are expressed as the mean ± SEM. G, Growing cells; GA, growth-arrested cells; d1, d2, d3, d4, d6, and d8, 1, 2, 3, 4, 6, and 8 d after adipogenic induction. D, Cell proliferation of MEFs after adipogenic induction. On d 0 and 3 after induction with DIM, primary MEFs from WT or meltrin –/– mice were counted. Four independent experiments were carried out, and values are shown as the mean ± SEM. An asterisk indicates a statistically significant difference (P < 0.05) between two experimental groups.
Requirement for meltrin in autocrine or juxtacrine signaling
Meltrin encodes a membrane-anchored metalloprotease. Some ADAMs participate in the limited proteolysis of various membrane proteins and extracellular matrix proteins. Meltrin expressed in MEFs, therefore, may stimulate the secretion of some soluble factors into the culture medium that enhance adipogenesis. Alternatively, meltrin may degrade inhibitory factors for adipogenesis, thus activating adipogenesis. We previously showed reduced ectodomain shedding of HB-EGF in response to phorbol ester stimulation in meltrin –/– MEFs (14). Because HB-EGF is a growth factor, it might enhance adipogenesis by stimulating cell proliferation. Addition of soluble HB-EGF to the culture medium of wild-type cells, however, tended to inhibit adipogenesis (Fig. 5A). We also added soluble HB-EGF for the initial 6 h. In this case, HB-EGF also inhibited adipogenesis (data not-shown). IGFBP-3 is another candidate for a meltrin substrate. Proteolysis of IGFBP-3 can enhance adipogenesis through activation of IGF, which is kept in a latent form when it is associated with members of the IGFBP family. To test this model, we next examined IGFBP-3 behavior after adipogenic induction with DIM or with INS alone. Endogenous IGFBP-3 expression in wild-type MEFs was confirmed by RT-PCR and Western blot (Fig. 5, B and C). Although IGFBP-3 secreted into the culture medium was intact during the first 18 h after adipogenic induction with DIM, both wild-type and meltrin –/– MEFs completely cleaved IGFBP-3 within the next 18 h. IGFBP-3 was not cleaved when adipogenesis was induced by INS alone (Fig. 5C), suggesting that one or more components of DIM may play a role in the cleavage of IGFBP-3. We did not observe the cleavage of IGFBP-5 or IGFBP-6 under the same conditions (data not shown). Thus, meltrin –/– MEFs cleave IGFBPs as efficiently as do wild-type MEFs, suggesting that meltrin is not likely to be a key regulator of IGFBP-3 cleavage in response to DIM.
FIG. 5. A, The effect of HB-EGF on adipogenesis. Primary MEFs from wild-type (WT) mice were incubated in the absence (–) or presence of DIM or INS with or without HB-EGF. After 8 d of differentiation, cells were fixed and stained with Oil Red-O. B, IGFBP-3 expression in MEFs. Endogenous IGFBP-3 mRNA expression in wild-type and meltrin –/– was confirmed by RT-PCR. C, IGFBP-3 cleavage in response to culture in DIM. Primary MEFs from wild-type or meltrin –/– mice were cultured in Opti-MEM reduced serum medium containing DIM or INS. After 18 or 36 h, the conditioned media were collected, and full-length IGFBP-3 and degraded IGFBP-3 were detected by immunoblotting.
Finally, we asked whether meltrin plays a regulatory role in paracrine signaling in adipogenesis. If meltrin does modulate paracrine signaling, through either the secretion of activating factors or the degradation of inhibitory factors, then wild-type conditioned medium might compensate for the defect in meltrin –/– MEFs. To examine this possibility, we exchanged the conditioned medium of meltrin –/– and wild-type MEFs every day, except on d 2 and 5, when the medium was replaced with fresh medium. (Precise protocols are described in Materials and Methods and summarized in Table 2.) Surprisingly, medium conditioned by wild-type cells did not enhance differentiation in meltrin –/– MEFs, but medium conditioned by meltrin –/– MEFs enhanced adipocyte differentiation in wild-type MEFs (Fig. 6A). This suggests that although meltrin –/– cells secreted some stimulatory factors for adipogenesis into the conditioned medium, they could not use them because of the lack of protease activity. The implication is that meltrin mediates autocrine or juxtacrine signaling for adipogenesis, rather than paracrine signaling.
Discussion
One of the fundamental questions concerning adipogenesis is how adipogenic cell proliferation is regulated during adipogenesis in the processes of both development and the progression of obesity. In cell culture, growth-arrested preadipocytes undergo several rounds of mitotic clonal expansion upon adipogenic stimulation and differentiate to express various adipocyte-specific genes, such as those required for lipid accumulation (26, 27). Evidence suggests that both cell growth arrest and proliferation are required for adipogenesis in vitro. The importance of proliferation of preadipocytes and/or mesenchymal stem cells, however, has tended to be underestimated in studies of obesity. That is because proliferation proceeds together with differentiation and hypertrophy during increases in adipose tissue mass in obesity. Dissection of the individual steps in terms of genes and molecules will be required to evaluate the contribution of each step to obesity, and such a dissection will shed new light on the regulatory mechanisms of weight gain in obesity.
In the present study we demonstrated the role of meltrin in obesity. Meltrin –/– mice were moderately resistant to the weight gain induced by a high-fat diet. Although the lack of meltrin in these mice did not prevent an increase in the size of adipocytes in their WAT under a high-fat diet, the high-fat diet did prevent a significant increase in the number of adipocytes, unlike the situation in wild-type mice. This inhibition of increase in the number of adipocytes was therefore the gross mechanism of moderate weight gain resistance in the knockout mice. In contrast, there was no obvious difference in adipocyte size and number between wild-type and meltrin –/– mice on a normal diet, suggesting normal formation of adipose tissues. Thus, meltrin preferentially regulates adipose cell mass by adding new adipocytes to WAT, rather than by stimulating differentiation and/or maturation of preexisting adipocytes during obesity. The involvement of meltrin in the proliferation of preadipocytes or mesenchymal stem cells during a high-fat diet is supported by the results obtained with S-V cells and MEFs prepared from meltrin –/– mice. Meltrin -lacking cells do not proliferate as efficiently as wild-type cells in response to adipogenic stimuli, although they can differentiate into adipocytes. The moderate defects in the proliferation of S-V cells in meltrin –/– adipose tissues suggest that meltrin participates in the proliferation of subpopulations of S-V cells in response to DIM treatment. We believe that these cells are probably DIM-responsive cells in S-V cells. However, we have not determined what cell types are affected in meltrin –/– S-V cells and MEFs, because preadipocytes, which are determined or destined to give rise to adipocytes, cannot be distinguished from precursors of preadipocytes and mesenchymal stem cells by the surface markers suitable for FACS analyses currently available. We would like to continue our study to determine what kind of cell proliferation is affected during adipogenic induction in vitro and during high-fat diet-induced obesity. It is also possible that FBS in growth medium compensates to some extent for the role of meltrin in the growth of S-V cells.
To determine whether meltrin –/– mice suffered from any energy imbalance that might explain their resistance to the development of obesity, we measured their food intake and glucose metabolism. Meltrin –/– mice showed glucose tolerance similar to that in wild-type mice under either a normal diet or a high-fat diet. Interestingly, in contrast, increased INS sensitivity was observed in meltrin –/– mice in the normal diet condition, although a slight difference in INS sensitivity between these mice during a high-fat diet was not statistically significant. Plasma triglyceride and NEFA levels were also significantly lower in meltrin –/– mice than in wild-type mice only when they were fed a normal diet. More precise investigation will be necessary to determine whether the altered metabolic profile of meltrin –/– mice fed a normal diet is associated with the resistance of meltrin –/– mice to high-fat diet-induced weight gain and their decreased proliferation of preadipocytes.
Although we are currently in the process of investigating metabolic alterations in meltrin –/– mice, the marked reductions in intracellular lipid accumulation in their BAT and livers under a high-fat diet suggest that energy expenditure might be up-regulated in these tissues. It is especially noteworthy that the liver steatosis seen in wild-type mice after a high-fat diet was not observed in meltrin –/– mice. Meltrin mRNA expression was up-regulated when animals were fed a high-fat diet. Thus, meltrin induced in the liver may play a critical role in the steatotic liver. Alternatively, the decreased lipid accumulation in the livers of meltrin –/– mice may be caused by a systemic increase in energy expenditure or may reflect the reduced amount of visceral fat close to the liver in these mice.
Meltrin is a metalloprotease that belongs to the ADAM family. The evidence suggests that some of the ADAM proteases are involved in ectodomain shedding of membrane proteins and limited proteolysis of extracellular matrix proteins. We previously showed impaired ectodomain shedding of HB-EGF induced by PMA in meltrin –/– MEFs (14). The possible involvement of HB-EGF in adipogenesis has been reported: for example, the abundant expression of HB-EGF mRNA in human and mice adipose tissues (28, 29), the induction of HB-EGF with IGF-I during adipogenesis in vitro (28, 29), and the close link between HB-EGF and vascular disease in obesity (28). The inhibitory effects of HB-EGF on adipogenesis of wild-type MEFs excluded the possibility of involvement of meltrin in the release of mature soluble HB-EGF ligands into the culture medium during adipogenesis. These inhibitory effects of HB-EGF are similar to those of TGF- on adipogenesis (30). In contrast, IGFBP-3 has also been reported as a substrate of meltrin in vitro (16, 17). IGFBP-3 is the most abundant of the six IGFBPs in plasma (31, 32) that transport and modulate the biological actions of IGF-1, one of the key regulators of adipogenesis and myogenesis (33). Comparison of IGFBP-3 cleavage after adipogenic induction in meltrin –/– MEFs and wild-type MEFs showed no significant difference in cleavage. Therefore, meltrin is not the major protease that cleaves IGFBP-3 after the addition of DIM. Many metalloproteases, including matrix metalloproteases and ADAM28, degrade IGFBP-3 (34, 35, 36), and some matrix metalloproteases are up-regulated during adipocyte differentiation (37). These metalloproteases might contribute to the cleavage of IGFBP-3. Nevertheless, we cannot exclude the possibility that local cleavage of IGFBPs by meltrin is necessary for correct signal transduction.
There are different types of intercellular signaling mediated by growth and differentiation factors: paracrine, juxtacrine, and autocrine signaling. In an attempt to determine which type(s) of signaling is mediated by meltrin , we performed medium exchange experiments in which wild-type and meltrin –/– MEFs were incubated with DIM/maturation medium conditioned by each other’s MEFs. If meltrin modulates paracrine signaling, through either the secretion of activating factors or the degradation of inhibitory factors in the culture medium, we would expect that wild-type MEF-conditioned medium would enhance the adipogenesis of meltrin –/– MEFs and that meltrin –/–-conditioned medium would not enhance the adipogenesis of wild-type MEFs. Instead, the reverse occurred. These data suggest that meltrin is involved in adipogenesis in a cell-autonomous process or a process dependent on cell-cell contacts. It is plausible that soluble factors stimulating adipogenesis are not available in meltrin –/– MEFs. Meltrin might mediate the cell-autonomous activation of such soluble factors or the activation of their receptors. In this sense, we still cannot exclude the possibility that cell-autonomous and transient activation of HB-EGF or IGF by meltrin are necessary for obesity and adipogenesis.
Previous reports suggest that the expression of meltrin enhances myoblast aggregation (3) or causes morphological changes in 3T3L1 adipocytes (38). However, we did not observe cell morphological differences between wild-type and meltrin –/– MEFs. Additional investigations are required to determine whether disintegrin- and cystein-rich domains play regulatory roles in adipogenesis.
In conclusion, we demonstrated that meltrin is one of the key regulators of obesity and adipogenesis. Because meltrin participates mainly in increasing adipocyte numbers and not in adipocyte hypertrophy, meltrin –/– mice showed mild resistance to high-fat diet-induced obesity, in which adipocyte hypertrophy is prominent. Moderate defects in the proliferation of S-V cells and adipogenesis in MEFs in meltrin –/– mice also suggest that meltrin mainly plays a role in the proliferation of preadipocytes or mesenchymal stem cells that give rise to preadipocytes during high-fat diet-induced obesity and does not play a major role in the differentiation or cellular hypertrophy of preexisting adipocytes. Interestingly, meltrin –/– knockout mice develop WAT similarly to wild-type mice on a normal diet. It is plausible that the proliferation of preadipocytes in high-fat diet-induced obesity involves distinct pathways for cells in embryos and nurslings. Meltrin might participate mainly in a pathway(s) critical for high-fat diet-induced adipogenesis in adults, but not for adipogenesis during development. Alternatively, maternal factors provided in utero and during nursing may compensate for the defects in WAT development in meltrin –/– mice. The roles of meltrin in adult adipogenesis suggest novel directions in the development of therapies for obesity and lipoatrophy.
Acknowledgments
We thank Nobuyuki Itoh, Toshiyuki Asaki, Kazuwa Nakao, and Hiroaki Masuzaki (Kyoto University) for technical advice and helpful discussions.
References
Moulin K, Truel N, Andre M, Arnauld E, Nibbelink M, Cousin B, Dani C, Penicaud L, Casteilla L 2001 Emergence during development of the white-adipocyte cell phenotype is independent of the brown-adipocyte cell phenotype. Biochem J 356:659–664
Hausman DB, DiGirolamo M, Bartness TJ, Hausman GJ, Martin RJ 2001 The biology of white adipocyte proliferation. Obes Rev 2:239–254
Yagami-Hiromasa T, Sato T, Kurisaki T, Kamijo K, Nabeshima Y, Fujisawa-Sehara A 1995 A metalloprotease-disintegrin participating in myoblast fusion. Nature 377:652–656
Blobel CP, Wolfsberg TG, Turck CW, Myles DG, Primakoff P, White JM 1992 A potential fusion peptide and an integrin ligand domain in a protein active in sperm-egg fusion. Nature 356:248–252
Cho C, Bunch DO, Faure JE, Goulding EH, Eddy EM, Primakoff P, Myles DG 1998 Fertilization defects in sperm from mice lacking fertilin ?. Science 281:1857–1859
Nishimura H, Cho C, Branciforte DR, Myles DG, Primakoff P 2001 Analysis of loss of adhesive function in sperm lacking cyritestin or fertilin ?. Dev Biol 233:204–213
Rooke J, Pan D, Xu T, Rubin GM 1996 KUZ, a conserved metalloprotease-disintegrin protein with two roles in Drosophila neurogenesis. Science 273:1227–1231
Peschon JJ, Slack JL, Reddy P, Stocking KL, Sunnarborg SW, Lee DC, Russell WE, Castner BJ, Johnson RS, Fitzner JN, Boyce RW, Nelson N, Kozlosky CJ, Wolfson MF, Rauch CT, Cerretti DP, Paxton RJ, March CJ, Black RA 1998 An essential role for ectodomain shedding in mammalian development. Science 282:1281–1284
Asakura M, Kitakaze M, Takashima S, Liao Y, Ishikura F, Yoshinaka T, Ohmoto H, Node K, Yoshino K, Ishiguro H, Asanuma H, Sanada S, Matsumura Y, Takeda H, Beppu S, Tada M, Hori M, Higashiyama S 2002 Cardiac hypertrophy is inhibited by antagonism of ADAM12 processing of HB-EGF: metaloprotease inhibitors as a new therapy. Nat Med 8:35–40
Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, Walsh A, Liu Z, Hayward B, Folz C, Manning SP, Bawa A, Saracino L, Thackston M, Benchekroun Y, Capparell N, Wang M, Adair R, Feng Y, Dubois J, FitzGerald MG, Huang H, Gibson R, Allen KM, Pedan A, Danzig MR, Umland SP, Egan RW, Cuss FM, Rorke S, Clough JB, Holloway JW, Holgate ST, Keith TP 2002 Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 418:426–430
Black R, White JM 1998 ADAMs: focus on the protease domain. Curr Opin Cell Biol 10:654–659
Seals DF, Courtneidge SA 2003 The ADAMs family of metalloproteases: multidomain proteins with multiple functions. Genes Dev 17:7–30[Free Full Text]
Werb Z, Yan Y 1998 A cellular striptease act. Science 282:1279–1280
Kurisaki T, Masuda A, Sudo K, Sakagami J, Higashiyama S, Matsuda Y, Nagabukuro A, Tsuji A, Nabeshima Y, Asano M, Iwakura Y, Fujisawa-Sehara A 2003 Phenotypic analysis of meltrin (ADAM12)-deficient mice: involvement of meltrin in adipogenesis and myogenesis. Mol Cell Biol 23:55–61
Kronqvist P, Kawaguchi N, Albrechtsen R, Xu X, Schroder HD, Moghadaszadeh B, Nielsen FC, Frohlich C, Engvall E, Wewer UM 2002 ADAM12 alleviates the skeletal muscle pathology in mdx dystrophic mice. Am J Pathol 161:1535–1540
Loechel F, Fox JW, Murphy G, Albrechtsen R, Wewer UM 2000 ADAM 12-S cleaves IGFBP-3 and IGFBP-5 and is inhibited by TIMP-3. Biochem Biophys Res Commun 278:511–515
Shi Z, Xu W, Loechel F, Wewer UM, Murphy LJ 2000 ADAM12, a disintegrin metalloprotease, interacts with insulin-like growth factor-binding protein-3. J Biol Chem 275:18574–18580
Hinkle CL, Sunnarborg SW, Loiselle D, Parker CE, Stevenson M, Russell WE, Lee DC 2004 Selective roles for TACE/ADAM17 in the shedding of the epidermal growth factor receptor ligand family. The juxtamembrane stalk determines cleavage efficiency. J Biol Chem 279:24179–24188
Shirakabe K, Wakatsuki S, Kurisaki T, Sehara-Fujisawa A 2001 Roles of meltrin ? (ADAM19) in the processing of neuregulin. J Biol Chem 276:9352–9358
Mori S, Tanaka M, Nanba D, Nishiwaki E, Ishiguro H, Higashiyama S, Matsuura N 2003 PACSIN3 binds ADAM12/meltrin and up-regulates ectodomain shedding of heparin-binding epidermal growth factor-like growth factor. J Biol Chem 278:46029–46034
Kurohara K, Komatsu K, Kurisaki T, Masuda A, Irie N, Asano M, Sudo K, Nabeshima Y, Iwakura Y, Sehara-Fujisawa A 2004 Essential roles of meltrin ? (ADAM19) in heart development. Dev Biol 267:14–28
Kawaguchi N, Xu X, Tajima R, Kronqvist P, Sundberg C, Loechel F, Albrechtsen R, Wewer UM 2003 ADAM12 protease induces adipogenesis in transgenic mice. Am J Pathol 160:1895–1903
Picard F, Gehin M, Annicotte J, Rocchi S, Champy MF, O’Malley BW, Chambon P, Auwerx J 2002 SRC-1 and TIF2 control energy balance between white and brown adipose tissues. Cell 111:931–941
Tanaka T, Yoshida N, Kishimoto T, Akira S 1997 Defective adipocyte differentiation in mice lacking the C/EBP? and/or C/EBP gene. EMBO J 16:7432–7443
Sakaue H, Konishi M, Ogawa W, Asaki T, Mori T, Yamasaki M, Takata M, Ueno H, kato S, Kasuga M, Itoh N 2002 Requirement of fibroblast growth factor 10 in development of white adipose tissue. Genes Dev 16:908–912
Tang Q-Q, Otto TC, Lane MD 2002 Mitotic clonal expansion: A synchronouse process required for adipogenesis. Proc Natl Acad Sci USA 100:44–49
Tang Q-Q, Otto TC, Lane MD 2003 CCAAT/enhancer-binding protein ? is required for mitotic clonal expansion during adipogenesis. Proc Natl Acad Sci USA 100:850–855
Matsumoto S, Kishida K, Shimomura I, Maeda N, Nagaretani H, Matsuda M, Nishizawa H, Kihara S, Funahashi T, Matsuzawa Y, Yamada A, Yamashita S, Tamura S, Kawata S 2002 Increased plasma HB-EGF associated with obesity and coronary artery disease. Biochem Biophys Res Commun 292:781–786
Mulligan C, Rochford J, Denyer G, Stephens R, Yeo G, Freeman T, Siddle K, O’Rahilly S 2002 Microarray analysis of insulin and insulin-like growth factor-1 (IGF-1) receptor signaling reveals the selective up-regulation of the mitogen heparin-binding EGF-like growth factor by IGF-1. J Biol Chem 277:42480–42487
Luetteke NC, Lee DC, Palmiter RD, Brinster RL, Sandgren EP 1993 Regulation of fat and muscle development by transforming growth factor in transgenic mice and in cultured cells. Cell Growth Differ 4:203–213
Baxter RC 2001 Signalling pathways involved in antiproliferative effects of IGFBP-3: a review. Mol Pathol 54:145–148
Modric T, Silha JV, Shi Z, Gui Y, Suwanichkul A, Durham SK, Powell DR, Murphy LJ 2001 Phenotypic manifestations of insulin-like growth factor-binding protein-3 overexpression in transgenic mice. Endocrinology 142:1958–1967
Sordella R, Jiang W, Chen GC, Curto M, Settleman J 2003 Modulation of Rho GTPase signaling regulates a switch between adipogenesis and myogenesis. Cell 113:147–158
Sadowski T, Dietrich S, Koschinsky F, Sedlacek R 2003 Matrix metalloproteinase 19 regulates insulin-like growth factor-mediated proliferation, migration, and adhesion in human keratinocytes through proteolysis of insulin-like growth factor binding protein-3. Mol Biol Cell 14:4569–4580
Manes S, Llorente M, Lacalle RA, Gomez-Mouton C, Kremer L, Mira E, Martinez AC 1999 The matrix metalloproteinase-9 regulates the insulin-like growth factor-triggered autocrine response in DU-145 carcinoma cells. J Biol Chem 274:6935–6945
Mochizuki S, Shimoda M, Shiomi T, Fujii Y, Okada Y 2004 ADAM28 is activated by MMP-7 (matrilysin-1) and cleaves insulin-like growth factor binding protein-3. Biochem Biophys Res Commun 315:79–84
Chavey C, Mari B, Monthouel MN, Bonnafous S, Anglard P, Van Obberghen E, Tartare-Deckert S 2003 Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278:11888–11896
Kawaguchi N, Sundberg C, Kveiborg M, Moghadaszadeh B, Asmar M, Dietrich N, Thodeti CK, Nielsen FC, Moller P, Mercurio AM, Albrechtsen R, Wewer UM 2003 ADAM12 induces actin cytoskeleton and extracellular matrix reorganization during early adipocyte differentiation by regulating ?1 integrin function. J Cell Sci 116:3893–3904(Megumi Masaki, Tomohiro K)
Address all correspondence and requests for reprints to: Dr. Atsuko Sehara-Fujisawa, Department of Growth Regulation, Institute for Frontier Medical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: asehara@frontier.kyoto-u.ac.jp.
Abstract
Meltrin is a member of the metalloprotease-disintegrin (ADAM) family. In this paper we demonstrate that meltrin is involved in the development of white adipose tissue. Compared with wild-type mice, meltrin –/– mice displayed moderate resistance to weight gain induced by a high-fat diet, mainly because of an impaired increase in the number of adipocytes. There was no obvious difference in adipocyte size between wild-type and meltrin –/– mice, suggesting normal maturation of adipocytes of the latter under a high-fat diet. Embryonic fibroblasts and stromal-vascular cells lacking meltrin exhibited impaired cell proliferation upon adipogenic stimulation, which was accompanied by moderate defects in adipose differentiation. Addition of culture medium conditioned with wild-type cells in an early phase of adipose differentiation did not restore the defects in the meltrin –/– cells. These results uncover the involvement of meltrin in the development of obesity and in adipogenic cell proliferation.
Introduction
THERE ARE TWO types of adipose tissue in the mammalian body. White adipose tissue (WAT) is distributed in many locations, stores excess energy as triglycerides, and releases fatty acids in response to energy requirements. The second type of adipose tissue is brown adipose tissue (BAT), which is the main thermogenic tissue in rodents. In BAT, fatty acid oxidation stimulated by the sympathetic nervous system generates heat through the induction of uncoupling protein-1. The developmental patterns of these tissues are quite different. BAT develops during fetal stages and essentially acquires all the features of mature tissue at birth, when nonshivering thermogenesis is required (1). In contrast, the development of WAT continues after birth, and its mass increases during postnatal life (2).
In this study we report the roles of meltrin in obesity, a pathological development of WAT. Meltrin (ADAM12) is a member of the metalloprotease-disintegrin (ADAM) family of proteins that contain metalloprotease and disintegrin domains (3). ADAMs play important roles in fertilization (4, 5, 6) and various aspects of morphogenesis (3, 7, 8) and are implicated in certain pathogenetic processes (9, 10). Genetic and biochemical studies have revealed that some ADAMs participate in the ectodomain shedding of various membrane-anchored proteins (by proteolysis of these proteins at the extracellular juxtamembrane region), including growth factors, intercellular signaling molecules, and adhesion molecules (11, 12, 13). Meltrin modulates myotube formation in vitro (3) and in vivo (14, 15). Two candidate substrates for meltrin protease have been reported to date: heparin-binding epidermal growth factor (HB-EGF) (14) and IGF-binding protein-3 (IGFBP-3) (16, 17), which regulates the activation of IGF-I. Evidence suggests that different ADAM proteases participate in the phorbol 12-myristate 13-acetate (PMA)-induced ectodomain shedding of membrane-anchored ErbB ligands, including HB-EGF (9, 18), TGF- (8), and neuregulin (19). In addition to TNF--converting enzyme (TACE/ADAM17), meltrin and meltrin ? (ADAM19) are examples of such proteases, although their expression patterns, ligand specificity, and regulation of protease activity differ (9, 20, 21).
We reported previously that during embryogenesis, some meltrin –/– mice display impaired development of interscapular BAT and of the skeletal muscles that surround BAT (14). The increase in ectodomain shedding of HB-EGF in response to PMA was markedly reduced in meltrin –/– embryonic fibroblasts. In contrast, Kawaguchi et al. (22) reported that transgenic mice overexpressing a placental isoform of human meltrin exhibit increased adipogenesis. Although the lack of meltrin also affects WAT formation in some meltrin –/– mouse embryos, decreased formation of WAT is not as prominent as that of BAT and recovers after birth, suggesting that compensatory mechanisms restore the roles of meltrin in the surviving population of meltrin –/– mice during WAT formation in utero or before weaning (Kurisaki, T., and A. Sehara-Fujisawa, unpublished observations). Alternatively, the slight decrease in WAT formation in meltrin –/– mice and the enhanced formation of WAT in transgenic mice expressing a placental isoform of human meltrin could be secondary effects of other phenotypes, such as systemic growth retardation or aberrant placental development. To determine whether meltrin plays a direct role in WAT development and to reveal its role in adipogenesis, we examined the effects of meltrin deficiency on the induction of obesity by a high-fat diet and on adipogenesis of stromal-vascular (S-V) cells or embryonic fibroblasts cultured in vitro, both of which eliminate maternal factors and other compensatory factors in utero and before weaning. As a result, we demonstrated the involvement of meltrin in obesity induced by a high-fat diet. Meltrin participates mainly in increasing the number of adipogenic cells during the progression of obesity and in cell proliferation at an early stage of adipogenesis in S-V cells and embryonic fibroblasts, which are critical for the expansion of adipocytes in vivo and in vitro.
Materials and Methods
Animal experiments
The meltrin –/– mouse line was generated as described previously (14). The initial chimeras were backcrossed to C57BL/6J more than 12 times. The numbers of animals used in experiments are mentioned in Results and the figure legends. Animals were maintained in a temperature-controlled facility with a 12-h light, 12-h dark cycle. When male mice were 4 wk of age, C57BL/6J wild-type and meltrin –/– mice were divided into two groups. One group was given a high-fat diet containing 60% fat (Oriental Yeast Co. Ltd., Tokyo, Japan), and the second group was given a normal (10% fat) diet (Research Diets, Inc., New Brunswick, NJ). The diets contained 5.2 and 3.8 kcal/g. Body weight was recorded every week, and food intake for the high-fat diet was determined every second day. For the measurement of metabolic parameters, mice were fed the normal or high-fat diet for 12 wk. Blood was collected from the heart after an overnight fast. Plasma triglycerides, total cholesterol, and nonesterified fatty acids (NEFA) were measured using enzymatic assays: the triglyceride E test (Wako Pure Chemical Industries Ltd., Osaka, Japan), the cholesterol E test (Wako Pure Chemical Industries Ltd.), and the NEFA test (Wako Pure Chemical Industries Ltd.). For the determination of plasma leptin, adiponectin, and insulin (INS) levels, ELISA kits were purchased from Morinaga (Kanagawa, Japan), Otsuka (Tokyo, Japan), and Shibayagi (Gunma, Japan), respectively. Glucose tolerance testing and INS tolerance testing were performed with the C57BL/6J and meltrin –/– mice fed the normal diet at 25–35 wk of age or with the C57BL/6J and meltrin –/– mice fed the high-fat diet for 10 wk. Glucose at 1 g/kg body weight or INS at 0.75 U/kg was injected ip after an overnight fast. Blood was collected from the tail vein. Glucose quantification was performed with the One Touch Ultra blood glucose-monitoring system (Johnson & Johnson, Tokyo, Japan). For measurement of INS concentrations before and after glucose injection, glucose at 1 g/kg body weight was injected ip into the C57BL/6J and meltrin –/– mice fed the normal diet after overnight fasting. At each time point, blood was collected from an intraorbital vein. The serum INS concentration was determined by ELISA.
Histological analysis of adipose tissues and liver
Pieces of adipose tissues and livers were fixed in 4% paraformaldehyde (PFA), dehydrated in ethanol, embedded in paraffin, and sectioned at a thickness of 4 μm. Sections were deparaffinized, rehydrated, and stained with hematoxylin and eosin. The number of adipocytes was determined as described previously (23). Briefly, adipocytes in the nine slices for each animal (n = 3) were counted, and the average cell volume was determined. Relative cell numbers were calculated based on the average cell volume and tissue weights. For Oil Red-O staining, livers were immediately embedded in tissue-freezing medium (Tissue-Tek OCT compound, Miles, Inc., Elkhart, IN), and sections 7 μm thick were stained with Oil Red-O.
RNA preparation and real-time PCR
Total RNA was prepared from tissues of adult male mice with RNeasy Lipid Tissue Mini (Qiagen, Valencia, CA) in accordance with the manufacturer’s instructions. For the isolation of total RNA from mouse embryonic fibroblasts (MEFs), RNeasy mini (Qiagen) was used. Meltrin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA expression were determined by RT, followed by real-time TaqMan PCR analysis. Primers and probes of meltrin and GAPDH were purchased from Applied Biosystems (Foster City, CA; Mm00475719_m1and P/N 4308313, respectively). The primers used to detect mRNA of IGFBP-3 were 5'-GACACCCAGAACTTCTCCTCC-3' and 5'-CATACTTGTCCACACACCAGC-3'.
Preparation of S-V cells and immunofluorescent staining
Inguinal fat pads were harvested from 4- to 5-wk-old, wild-type or meltrin –/– male mice. After blood was washed out of the tissues, they were minced and digested with 1 mg/ml collagenase type I (Worthington Biochemical Corp., Freehold, NJ) for 30 min at 37 C. Cells were filtered through 200-μm pore size nylon meshes. The S-V cells were separated from adipocytes by centrifugation and washed with DMEM (Nacalai Tesque, Inc., Kyoto, Japan) supplemented with 10% heat-inactivated fetal bovine serum (FBS). S-V cells were plated and propagated to confluence in DMEM supplemented with 10% heat-inactivated FBS, 50 U/ml penicillin, and 50 μg/ml streptomycin. Two days later, the medium was replaced with differentiation induction medium (DIM) containing 1 μM dexamethasone (DEX; Sigma-Aldrich Corp., St. Louis, MO), 10 μg/ml INS (Sigma-Aldrich Corp.), and 0.5 mM 3-isobutyl-1-methylxanthine (MIX; Nacalai Tesque, Inc.) with 10% heat-inactivated FBS. After 24 h, cells were washed with PBS twice and fixed with 4% PFA for 30 min, followed by treatment with ice-cold 0.2% Triton X-20 for 5 min, then stained with anti-Ki-67 (BD PharMingen, San Diego, CA; 1:100 dilution) as a first antibody and antimouse IgG-Alexa 488-conjugated antibody (Molecular Probes, Eugene, OR; 1:400 dilution) as a secondary antibody. 4',6-Diamidino-2-phenylindole (DAPI; 0.2 μg/ml; Molecular Probes) was used for staining of all nuclei. Photographs were taken on a Zeiss fluorescent microscope (New York, NY) with MetaMorph software (Universal Imaging Corp., Brock and Michelsen, Briker?d, Denmark). The ratio of Ki-67-positive cells to DAPI-positive cells was determined.
Preparation of MEFs and induction of adipogenesis
Primary embryonic fibroblasts were prepared from 14.5 d postcoitus embryos. Cells (8 x 105) were plated on 12-well plastic dishes and cultured at 37 C in standard medium: -MEM (Nacalai Tesque, Inc.) supplemented with 10% heat-inactivated FBS, 50 U/ml penicillin, and 50 μg/ml streptomycin. Cells were propagated to confluence. Two days later, the medium was replaced with DIM with 10% heat-inactivated FBS. After 2 d, this medium was replaced with maturation medium, which contains 10 μg/ml INS. After another 3 d, the maturation medium was replaced with standard medium. For evaluation of the effect of HB-EGF, wild-type MEFs were cultured in DIM or -MEM containing 10 μg/ml INS with or without 100 ng/ml HB-EGF (R&D Systems, Inc., Minneapolis, MN). After 2 d, the medium was replaced with a maturation medium with or without HB-EGF. After another 3 d, the maturation medium was replaced with standard medium. In the medium exchange experiment, we inducted meltrin –/– and wild-type MEFs with DIM as described above (d 0). Then, culture medium of wild-type MEFs was replaced with meltrin –/–-conditioned medium and vice versa on d 1, 3, and 6. The medium was replaced with fresh maturation medium on d 2 and with standard medium on d 5. After 8 d, cytoplasmic lipid accumulation was observed by brightfield microscopy with Oil Red-O staining. The composition of each medium is shown in Tables 1 and 2 in detail. Oil Red-O staining was performed as follows. Cells were washed with PBS, fixed with 4% PFA for 10 min, then stained with 60% filtered Oil Red-O stock solution (0.15 g Oil Red-O in 50 ml isopropanol) for 30 min, washed with 60% isopropanol, then briefly washed with PBS twice before being visualized. To quantify the amount of lipid, stained oil was eluted with isopropanol, and the absorbance at a wavelength of 510 nm was read after dilution of the eluate to a linear range. To determine the number of cells, on d 0 (before induction with DIM) and d 3 (after induction), cells were trypsinized and well suspended with -MEM with 10% heat-inactivated FBS. Cells were counted using Neubauer’s hemocytometer (Erma, Tokyo, Japan) under light microscopy. Experiments were performed in triplicate with four different litters of mice, and values are shown as the mean ± SEM.
TABLE 1. Composition of each culture medium
TABLE 2. Cell culture conditions in the medium exchange experiment
Western blot
In preparation for Western blotting, 1.5 x 106 MEFs were placed in six-well dishes and cultured in a standard medium. Two days later, the medium was replaced with Opti-MEM (Invitrogen Life Technologies, Inc., Carlsbad, CA) reduced serum with either DIM or 10 μg/ml INS. After 18 or 36 h, the conditioned medium was concentrated 10-fold using a centrifuge tube filtration unit (Amicon Ultra-4 10,000 MWCO, Millipore Corp., Bedford, MA). Western blotting was performed as described previously (19, 21). Goat polyclonal antibody against IGFBP-3 (sc-6004) was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Statistical analysis
Data are expressed as the mean ± SEM. A two-tailed t test was used to calculate P values.
Results
Meltrin –/– mice are resistant to induction of obesity by a high-fat diet
To explore the role of meltrin in the development of WAT in adult mice, we examined the body weight gain of meltrin –/– mice on either a high-fat diet or a normal diet. After weaning, meltrin –/– mice fed a normal diet gained weight similarly to wild-type mice. However, meltrin –/– mice on a high-fat diet were protected against weight gain to some extent, despite having the same food intake as their wild-type counterparts (Fig. 1, A and B). Representative specimens of wild-type and meltrin –/– mice given the high-fat diet for 12 wk are shown in Fig. 1C. The decreased weight gain of meltrin –/– mice fed the high-fat diet was mainly reflected by the smaller mass and lower weight of their WAT compared with those of wild-type mice (Fig. 1, D and E). Resistance of meltrin –/– mice to high-fat diet-induced obesity was also shown by the normal appearance of their livers compared with those of the wild-type mice, which displayed hepatic steatosis. Furthermore, the interscapular BAT of meltrin –/– mice on the high-fat diet was darker and had less mass of WAT surrounding it than that of wild-type mice (Fig. 1D). These results indicate that meltrin –/– mice were resistant to the obesity induced by a high-fat diet.
FIG. 1. Decreased body weight gain of meltrin –/– mice on a high-fat diet. A, Body weight curves for male mice on normal or high-fat diets. Body weights were recorded weekly in wild-type (WT) and meltrin –/– (Mel.–/–) mice fed either a normal diet (ND) or a high-fat diet (HFD). Only on the high-fat diet did meltrin –/– mice have lower body weights than WT mice. Values are expressed as the mean ± SEM. WT-HFD, n = 8; WT-ND, n = 6; Mel.–/– ND, n = 6; Mel.–/– NFD, n = 6. Asterisks indicate a statistically significant difference (P < 0.05) between high fat-fed WT and meltrin –/– mice. B, Food intake was recorded every second day and was the same in WT and meltrin –/– mice (10.5 ± 0.1 and 10.7 ± 0.2 kcal/d, respectively). C, Morphology of 15-wk-old mice fed a high-fat diet. D, Morphology of interscapular BAT, liver, and inguinal WAT in 16-wk-old mice fed a high-fat diet. Scale bar, 1 cm. E, Tissue weights in mice after 12 wk on a high fat or a normal diet, expressed as the mean ± SEM (n = 6). Asterisks indicate statistically significant difference (P < 0.05) between two experimental groups.
Glucose homeostasis of meltrin –/– mice fed a high- fat diet
Defects in glucose homeostasis are important factors causing obesity. To evaluate the metabolic difference between wild-type and meltrin –/– mice, we performed INS tolerance and glucose tolerance tests on these mice fed a normal diet or a high-fat diet. In a normal diet condition, increased INS sensitivity was observed in meltrin –/– mice (Fig. 2A). Hypoglycemia was evident 30 and 60 min after INS injection in both groups of mice. Wild-type mice more or less recovered their plasma glucose concentrations by 120 min after injection, whereas meltrin –/– mice remained hypoglycemic (P < 0.05). Plasma triglyceride and NEFA levels were also significantly lower in meltrin –/– mice fed a normal diet than in wild-type (Table 3). Thus, such altered metabolic profiles of meltrin –/– mice fed a normal diet might be associated with the resistance of meltrin –/– mice to high-fat diet-induced weight gain. In contrast, no statistically significant difference was found between wild-type and meltrin –/– mice 10 wk after a high-fat diet. The glucose clearance rates after ip glucose injection in wild-type and meltrin –/– mice were almost the same whether animals were fed a normal diet or a high-fat diet (Fig. 2B). There was no difference in serum INS concentrations before and after glucose injection between wild-type and meltrin –/– mice (Fig. 2C).
FIG. 2. A, INS tolerance test in wild-type (WT) and meltrin –/– (Mel.–/–) mice fed a normal or a high-fat diet. INS (0.75 U/kg) was injected into wild-type and meltrin –/– 25- to 35-wk-old mice fed a normal diet or 15-wk-old mice fed a high-fat diet for 10 wk after overnight fasting. In each experiment, values are expressed as the mean ± SEM. ND, Normal diet (WT, n = 14; meltrin –/–, n = 9); HFD, high-fat diet (WT, n = 7; meltrin –/–, n = 6). B, Glucose tolerance test in wild-type (WT) and meltrin –/– (Mel.–/–) mice fed a normal or a high-fat diet. Glucose (1 g/kg) was injected ip into wild-type and meltrin –/– 25- to 35-wk-old mice fed a normal diet or 15-wk-old mice fed a high-fat diet for 10 wk after overnight fasting. In each experiment, values are expressed as the mean ± SEM. ND, Normal diet (WT, n = 9; meltrin –/–, n = 5); HFD, high-fat diet (WT, n = 7; meltrin –/–, n = 6). C, INS concentrations pre- and postinjection of glucose. Glucose (1 g/kg) was injected ip into wild-type and meltrin –/– 25- to 35-wk-old mice fed a normal diet after overnight fasting. Before and after 15-min glucose injection, blood was collected from an intraorbital vein. The serum INS concentration was determined. In each experiment, values are expressed as the mean ± SEM. WT, n = 5; meltrin –/–, n = 4.
TABLE 3. Metabolic parameters in wild-type (WT) and meltrin –/– (Mel.–/–) mice
Histological analysis of meltrin –/– mice fed a high fat diet
We performed a histological analysis of BAT, liver, and inguinal, epididymal, and retrorenal WAT of wild-type and meltrin –/– mice fed a normal or a high-fat diet. In the inguinal WAT, the adipocytes of wild-type and meltrin –/– mice on the normal diet were roughly the same size. On the high-fat diet, the adipocytes in inguinal WAT of both types of mice increased in size due to intracellular lipid accumulation (Fig. 3A). These lipid accumulation patterns were the same in retrorenal and epididymal WAT (data not shown). This indicates that fat storage in adipocytes was normal in meltrin –/– mice.
FIG. 3. A, Histological analysis of adipose tissues. Inguinal WAT and interscapular BAT sections were stained with hematoxylin and eosin. Magnifications to show morphology of inguinal WAT and BAT are x10 and x20, respectively. Scale bar, 100 μm. WT, Wild-type mice; Mel.–/–, meltrin –/– mice; HFD, high-fat diet; ND, normal diet. Average cell volumes for WT ND and Mel.–/– ND iguinal WAT were 1.51 x 104 ± 0.12 x 104 and 1.90 x 104 ± 0.44 x 104 μm3, respectively. Average cell volumes for WT HFD and Mel.–/– HFD inguinal WAT were 2.37 x 105 ± 0.30 x 105 and 2.07 x 105 ± 0.46 x 105 μm3, respectively. Data from nine slices for each animal (n = 3) were analyzed. B, Histological analysis of liver. Liver sections were stained with either hematoxylin and eosin (H & E) or Oil Red-O. Magnification for morphological examination, 10x and 20x, respectively. Scale bar, 100 μm. Abbreviations are explained in A. C, Relative cell numbers in wild-type and meltrin –/– WATs. The cell numbers in wild-type and meltrin –/– WATs were estimated as described in Materials and Methods. Data from nine slices for each animal (n = 3) were analyzed. Asterisks indicate a statistically significant difference (P < 0.05) between two experimental groups. Abbreviations are explained in A. D, Meltrin mRNA expression in WAT and liver of mice fed a normal diet. Total RNA was prepared from tissues of adult male mice, and the mRNA level of meltrin was determined by real-time PCR. Levels of mRNA were normalized to that of GAPDH. Values from epididymal WAT were set at 1. In each experiment, n = 3, and values are expressed as the mean ± SEM. E, High-fat diet-induced expression of meltrin mRNA in liver and epididymal WAT. Total RNA was prepared from tissues of adult male mice fed either a normal diet () or a high-fat diet (), and the mRNA level of meltrin was determined by real-time PCR. Levels of mRNA were normalized to that of GAPDH. Values from epididymal WAT of mice fed a normal diet were set at 1. In each experiment, n = 3, and values are expressed as the mean ± SEM. TA, Tibialis anterior muscle. Asterisks indicate a statistically significant difference (P < 0.05) between two experimental groups.
The adipocytes in the BAT of wild-type mice on a high-fat diet accumulated lipid in large droplets. However, the BAT adipocytes of meltrin –/– mice accumulated less lipid in smaller droplets after being fed a high-fat diet (Fig. 3A), accounting in part for the macroscopically darker color of the BAT of meltrin –/– mice. Hematoxylin and eosin staining and Oil Red-O staining of liver sections confirmed the presence of steatosis in the wild-type mice fed a high-fat diet, whereas there was no sign of steatosis in the livers of meltrin –/– mice on the same diet (Fig. 3B).
Because the fat pads of meltrin –/– mice on a high-fat diet were smaller than those of wild-type mice despite the similar size of individual adipocytes (Figs. 1E and 3A), we counted the number of cells in representative populations of white adipocytes from the inguinal, epididymal, and retrorenal WAT. There were fewer cells in meltrin –/– mice fed a high-fat diet than in wild-type mice on the same diet (Fig. 3C), but cell numbers were approximately the same in mice fed a normal diet. These data indicate that meltrin is involved in increasing the number of adipocytes in mice fed a high-fat diet.
Meltrin is expressed in various adipose tissues
We used quantitative real-time PCR to examine relative meltrin expression in several tissues of wild-type mice fed a normal diet. The level of meltrin mRNA was almost the same among epididymal and retrorenal WAT and interscapular BAT. The meltrin mRNA level in inguinal WAT was 12 times higher than that in other adipose tissues. Meltrin mRNA expression was much lower or was not recorded in the liver and tibialis anterior muscle (Fig. 3, D and E). This suggests that meltrin is expressed in adipose tissues and is required for the increase in cell number induced by a high-fat diet.
Although meltrin mRNA expression was low in the liver in the normal diet condition, meltrin –/– mice were resistant to hepatic steatosis, which was observed in wild-type mice fed a high-fat diet (Figs. 1D and 3B). We examined the influence of a high-fat diet on the expression of meltrin mRNA in the liver, tibialis anterior muscle, and epididymal WAT. As a result, meltrin mRNA expression was up-regulated in epididymal WAT (2-fold) and liver (12-fold), but not in muscle, when mice were fed a high-fat diet (Fig. 3E). Up-regulation of meltrin mRNA in the liver on a high-fat diet might be involved in hepatic steatosis.
S-V cells lacking meltrin have a defect(s) in cell proliferation during adipogenesis
We next asked whether decreased cell number in meltrin –/– adipose tissue after a high-fat diet is due to impaired proliferation of adipogenic cells in meltrin –/– mice. We found the most prominent difference in the development of inguinal WAT between wild-type and meltrin –/– mice after a high-fat diet. Previous studies indicate that inguinal WAT has a greater proliferation capacity than other WAT, such as epididymal fat pads (2). To analyze cell proliferation under adipogenic conditions in vitro, we isolated S-V cells, which contain preadipocytes and precursors of them, from inguinal WAT of wild-type and meltrin –/– mice. The S-V cells were propagated to confluence and induced to differentiate into adipocytes in DIM 2 d after confluence. After 24 h, cellular proliferation was assessed by immunofluorescent staining with an antibody against Ki-67 nuclear antigen, a marker of cellular proliferation, followed by quantification of the rate of Ki-67-positive cells. The proliferation rates of cells from wild-type and meltrin –/– mice were relatively low (30%) without DIM. Upon treatment with DIM, the proliferation rates of wild-type and meltrin –/– S-V cells were enhanced to 65% and 53%, respectively (Fig. 4). This modestly decreased proliferation rate of meltrin –/– S-V cells after DIM treatment suggests a role for meltrin in cell proliferation in vitro.
FIG. 4. A, Ki-67-positive cells in S-V cells after adipogenic induction. S-V cells from wild-type (WT) or meltrin –/– (Mel.–/–) mice were incubated with or without DIM for 24 h, and Ki-67-positive cells were visualized. DAPI stain was used for nuclei. The microscopic images show representative fields of individual treatments for three independent experiments. B, The rate of proliferating cells in S-V cells. Data from five fields for each well were analyzed. The ratios of Ki-67-positive cells to DAPI-positive cells were determined. Experiments were performed in duplicate (three independent experiments were carried out), and values are shown as the mean ± SEM. An asterisk indicates a statistically significant difference (P < 0.05) between two experimental groups.
MEFs lacking meltrin have a defect(s) in cell proliferation during adipogenesis
The conversion of MEFs to adipocytes after hormonal stimulation has been extensively studied as a means of identifying key regulatory factors in adipogenesis (24, 25). We prepared primary embryonic fibroblasts from wild-type and meltrin –/– mice to examine their adipogenic properties. The MEFs were induced to differentiate into adipocytes in DIM. Meltrin –/– MEFs showed a modestly reduced capacity to differentiate into adipocytes compared with wild-type MEFs. It is noteworthy that Oil Red-O-negative space was more pronounced in the culture of meltrin –/– MEFs than in that of wild-type cells (Fig. 6A). Quantification of triglyceride production confirmed a decreased differentiation in meltrin –/– MEFs (Fig. 6B). We performed quantitative real-time PCR to examine meltrin mRNA expression in each step of adipogenesis. Meltrin mRNA expression was transiently increased after induction of adipogenesis (highest 2 d after adipogenic differentiation) and decreased during the process of terminal differentiation (Fig. 6C). These results suggested that meltrin might have an essential role in the early steps of adipogenesis, including the production of adipogenic precursors and their mitotic clonal expansion, that precede terminal differentiation. To test this idea, we evaluated the increase in cell numbers during differentiation in wild-type and meltrin –/– MEFs. We found a significant increase in cell numbers in wild-type, but not meltrin –/–, MEFs 3 d after adipogenic differentiation (Fig. 6D). This impaired increase in the number of cells during adipogenic induction also suggests a role for meltrin in cell proliferation, which occurs in the early stages of differentiation and would affect the number of differentiated adipocytes produced.
FIG. 6. A, Adipogenic differentiation of primary MEFs prepared from wild-type and meltrin –/– mice. Primary MEFs from wild-type (WT) or meltrin –/– (Mel. –/–) mice were incubated in the absence (–) or presence of DIM. (–).ME and DIM.ME, Cultures in conditioned medium prepared from meltrin –/– and wild-type MEFs, respectively. After 8 d of differentiation, cells were fixed and stained with Oil Red-O. B, Triglyceride accumulation in wild-type and meltrin –/– MEFs after adipogenic induction. Triglyceride accumulation was expressed as absorbance at an OD of 510 nm. In each experiment, n = 3, and values were expressed as the mean ± SEM. Three independent experiments were carried out, and representative data are shown. Asterisks indicate a statistically significant difference (P < 0.05) between two experimental groups. C, Meltrin mRNA expression in MEFs during adipogenesis. Total RNA was prepared from wild-type MEFs, and the mRNA level of meltrin was determined by real-time PCR. Levels of mRNA were normalized to that of GAPDH. Values from growing cells were set at 1. In each experiment, n = 3, and values are expressed as the mean ± SEM. G, Growing cells; GA, growth-arrested cells; d1, d2, d3, d4, d6, and d8, 1, 2, 3, 4, 6, and 8 d after adipogenic induction. D, Cell proliferation of MEFs after adipogenic induction. On d 0 and 3 after induction with DIM, primary MEFs from WT or meltrin –/– mice were counted. Four independent experiments were carried out, and values are shown as the mean ± SEM. An asterisk indicates a statistically significant difference (P < 0.05) between two experimental groups.
Requirement for meltrin in autocrine or juxtacrine signaling
Meltrin encodes a membrane-anchored metalloprotease. Some ADAMs participate in the limited proteolysis of various membrane proteins and extracellular matrix proteins. Meltrin expressed in MEFs, therefore, may stimulate the secretion of some soluble factors into the culture medium that enhance adipogenesis. Alternatively, meltrin may degrade inhibitory factors for adipogenesis, thus activating adipogenesis. We previously showed reduced ectodomain shedding of HB-EGF in response to phorbol ester stimulation in meltrin –/– MEFs (14). Because HB-EGF is a growth factor, it might enhance adipogenesis by stimulating cell proliferation. Addition of soluble HB-EGF to the culture medium of wild-type cells, however, tended to inhibit adipogenesis (Fig. 5A). We also added soluble HB-EGF for the initial 6 h. In this case, HB-EGF also inhibited adipogenesis (data not-shown). IGFBP-3 is another candidate for a meltrin substrate. Proteolysis of IGFBP-3 can enhance adipogenesis through activation of IGF, which is kept in a latent form when it is associated with members of the IGFBP family. To test this model, we next examined IGFBP-3 behavior after adipogenic induction with DIM or with INS alone. Endogenous IGFBP-3 expression in wild-type MEFs was confirmed by RT-PCR and Western blot (Fig. 5, B and C). Although IGFBP-3 secreted into the culture medium was intact during the first 18 h after adipogenic induction with DIM, both wild-type and meltrin –/– MEFs completely cleaved IGFBP-3 within the next 18 h. IGFBP-3 was not cleaved when adipogenesis was induced by INS alone (Fig. 5C), suggesting that one or more components of DIM may play a role in the cleavage of IGFBP-3. We did not observe the cleavage of IGFBP-5 or IGFBP-6 under the same conditions (data not shown). Thus, meltrin –/– MEFs cleave IGFBPs as efficiently as do wild-type MEFs, suggesting that meltrin is not likely to be a key regulator of IGFBP-3 cleavage in response to DIM.
FIG. 5. A, The effect of HB-EGF on adipogenesis. Primary MEFs from wild-type (WT) mice were incubated in the absence (–) or presence of DIM or INS with or without HB-EGF. After 8 d of differentiation, cells were fixed and stained with Oil Red-O. B, IGFBP-3 expression in MEFs. Endogenous IGFBP-3 mRNA expression in wild-type and meltrin –/– was confirmed by RT-PCR. C, IGFBP-3 cleavage in response to culture in DIM. Primary MEFs from wild-type or meltrin –/– mice were cultured in Opti-MEM reduced serum medium containing DIM or INS. After 18 or 36 h, the conditioned media were collected, and full-length IGFBP-3 and degraded IGFBP-3 were detected by immunoblotting.
Finally, we asked whether meltrin plays a regulatory role in paracrine signaling in adipogenesis. If meltrin does modulate paracrine signaling, through either the secretion of activating factors or the degradation of inhibitory factors, then wild-type conditioned medium might compensate for the defect in meltrin –/– MEFs. To examine this possibility, we exchanged the conditioned medium of meltrin –/– and wild-type MEFs every day, except on d 2 and 5, when the medium was replaced with fresh medium. (Precise protocols are described in Materials and Methods and summarized in Table 2.) Surprisingly, medium conditioned by wild-type cells did not enhance differentiation in meltrin –/– MEFs, but medium conditioned by meltrin –/– MEFs enhanced adipocyte differentiation in wild-type MEFs (Fig. 6A). This suggests that although meltrin –/– cells secreted some stimulatory factors for adipogenesis into the conditioned medium, they could not use them because of the lack of protease activity. The implication is that meltrin mediates autocrine or juxtacrine signaling for adipogenesis, rather than paracrine signaling.
Discussion
One of the fundamental questions concerning adipogenesis is how adipogenic cell proliferation is regulated during adipogenesis in the processes of both development and the progression of obesity. In cell culture, growth-arrested preadipocytes undergo several rounds of mitotic clonal expansion upon adipogenic stimulation and differentiate to express various adipocyte-specific genes, such as those required for lipid accumulation (26, 27). Evidence suggests that both cell growth arrest and proliferation are required for adipogenesis in vitro. The importance of proliferation of preadipocytes and/or mesenchymal stem cells, however, has tended to be underestimated in studies of obesity. That is because proliferation proceeds together with differentiation and hypertrophy during increases in adipose tissue mass in obesity. Dissection of the individual steps in terms of genes and molecules will be required to evaluate the contribution of each step to obesity, and such a dissection will shed new light on the regulatory mechanisms of weight gain in obesity.
In the present study we demonstrated the role of meltrin in obesity. Meltrin –/– mice were moderately resistant to the weight gain induced by a high-fat diet. Although the lack of meltrin in these mice did not prevent an increase in the size of adipocytes in their WAT under a high-fat diet, the high-fat diet did prevent a significant increase in the number of adipocytes, unlike the situation in wild-type mice. This inhibition of increase in the number of adipocytes was therefore the gross mechanism of moderate weight gain resistance in the knockout mice. In contrast, there was no obvious difference in adipocyte size and number between wild-type and meltrin –/– mice on a normal diet, suggesting normal formation of adipose tissues. Thus, meltrin preferentially regulates adipose cell mass by adding new adipocytes to WAT, rather than by stimulating differentiation and/or maturation of preexisting adipocytes during obesity. The involvement of meltrin in the proliferation of preadipocytes or mesenchymal stem cells during a high-fat diet is supported by the results obtained with S-V cells and MEFs prepared from meltrin –/– mice. Meltrin -lacking cells do not proliferate as efficiently as wild-type cells in response to adipogenic stimuli, although they can differentiate into adipocytes. The moderate defects in the proliferation of S-V cells in meltrin –/– adipose tissues suggest that meltrin participates in the proliferation of subpopulations of S-V cells in response to DIM treatment. We believe that these cells are probably DIM-responsive cells in S-V cells. However, we have not determined what cell types are affected in meltrin –/– S-V cells and MEFs, because preadipocytes, which are determined or destined to give rise to adipocytes, cannot be distinguished from precursors of preadipocytes and mesenchymal stem cells by the surface markers suitable for FACS analyses currently available. We would like to continue our study to determine what kind of cell proliferation is affected during adipogenic induction in vitro and during high-fat diet-induced obesity. It is also possible that FBS in growth medium compensates to some extent for the role of meltrin in the growth of S-V cells.
To determine whether meltrin –/– mice suffered from any energy imbalance that might explain their resistance to the development of obesity, we measured their food intake and glucose metabolism. Meltrin –/– mice showed glucose tolerance similar to that in wild-type mice under either a normal diet or a high-fat diet. Interestingly, in contrast, increased INS sensitivity was observed in meltrin –/– mice in the normal diet condition, although a slight difference in INS sensitivity between these mice during a high-fat diet was not statistically significant. Plasma triglyceride and NEFA levels were also significantly lower in meltrin –/– mice than in wild-type mice only when they were fed a normal diet. More precise investigation will be necessary to determine whether the altered metabolic profile of meltrin –/– mice fed a normal diet is associated with the resistance of meltrin –/– mice to high-fat diet-induced weight gain and their decreased proliferation of preadipocytes.
Although we are currently in the process of investigating metabolic alterations in meltrin –/– mice, the marked reductions in intracellular lipid accumulation in their BAT and livers under a high-fat diet suggest that energy expenditure might be up-regulated in these tissues. It is especially noteworthy that the liver steatosis seen in wild-type mice after a high-fat diet was not observed in meltrin –/– mice. Meltrin mRNA expression was up-regulated when animals were fed a high-fat diet. Thus, meltrin induced in the liver may play a critical role in the steatotic liver. Alternatively, the decreased lipid accumulation in the livers of meltrin –/– mice may be caused by a systemic increase in energy expenditure or may reflect the reduced amount of visceral fat close to the liver in these mice.
Meltrin is a metalloprotease that belongs to the ADAM family. The evidence suggests that some of the ADAM proteases are involved in ectodomain shedding of membrane proteins and limited proteolysis of extracellular matrix proteins. We previously showed impaired ectodomain shedding of HB-EGF induced by PMA in meltrin –/– MEFs (14). The possible involvement of HB-EGF in adipogenesis has been reported: for example, the abundant expression of HB-EGF mRNA in human and mice adipose tissues (28, 29), the induction of HB-EGF with IGF-I during adipogenesis in vitro (28, 29), and the close link between HB-EGF and vascular disease in obesity (28). The inhibitory effects of HB-EGF on adipogenesis of wild-type MEFs excluded the possibility of involvement of meltrin in the release of mature soluble HB-EGF ligands into the culture medium during adipogenesis. These inhibitory effects of HB-EGF are similar to those of TGF- on adipogenesis (30). In contrast, IGFBP-3 has also been reported as a substrate of meltrin in vitro (16, 17). IGFBP-3 is the most abundant of the six IGFBPs in plasma (31, 32) that transport and modulate the biological actions of IGF-1, one of the key regulators of adipogenesis and myogenesis (33). Comparison of IGFBP-3 cleavage after adipogenic induction in meltrin –/– MEFs and wild-type MEFs showed no significant difference in cleavage. Therefore, meltrin is not the major protease that cleaves IGFBP-3 after the addition of DIM. Many metalloproteases, including matrix metalloproteases and ADAM28, degrade IGFBP-3 (34, 35, 36), and some matrix metalloproteases are up-regulated during adipocyte differentiation (37). These metalloproteases might contribute to the cleavage of IGFBP-3. Nevertheless, we cannot exclude the possibility that local cleavage of IGFBPs by meltrin is necessary for correct signal transduction.
There are different types of intercellular signaling mediated by growth and differentiation factors: paracrine, juxtacrine, and autocrine signaling. In an attempt to determine which type(s) of signaling is mediated by meltrin , we performed medium exchange experiments in which wild-type and meltrin –/– MEFs were incubated with DIM/maturation medium conditioned by each other’s MEFs. If meltrin modulates paracrine signaling, through either the secretion of activating factors or the degradation of inhibitory factors in the culture medium, we would expect that wild-type MEF-conditioned medium would enhance the adipogenesis of meltrin –/– MEFs and that meltrin –/–-conditioned medium would not enhance the adipogenesis of wild-type MEFs. Instead, the reverse occurred. These data suggest that meltrin is involved in adipogenesis in a cell-autonomous process or a process dependent on cell-cell contacts. It is plausible that soluble factors stimulating adipogenesis are not available in meltrin –/– MEFs. Meltrin might mediate the cell-autonomous activation of such soluble factors or the activation of their receptors. In this sense, we still cannot exclude the possibility that cell-autonomous and transient activation of HB-EGF or IGF by meltrin are necessary for obesity and adipogenesis.
Previous reports suggest that the expression of meltrin enhances myoblast aggregation (3) or causes morphological changes in 3T3L1 adipocytes (38). However, we did not observe cell morphological differences between wild-type and meltrin –/– MEFs. Additional investigations are required to determine whether disintegrin- and cystein-rich domains play regulatory roles in adipogenesis.
In conclusion, we demonstrated that meltrin is one of the key regulators of obesity and adipogenesis. Because meltrin participates mainly in increasing adipocyte numbers and not in adipocyte hypertrophy, meltrin –/– mice showed mild resistance to high-fat diet-induced obesity, in which adipocyte hypertrophy is prominent. Moderate defects in the proliferation of S-V cells and adipogenesis in MEFs in meltrin –/– mice also suggest that meltrin mainly plays a role in the proliferation of preadipocytes or mesenchymal stem cells that give rise to preadipocytes during high-fat diet-induced obesity and does not play a major role in the differentiation or cellular hypertrophy of preexisting adipocytes. Interestingly, meltrin –/– knockout mice develop WAT similarly to wild-type mice on a normal diet. It is plausible that the proliferation of preadipocytes in high-fat diet-induced obesity involves distinct pathways for cells in embryos and nurslings. Meltrin might participate mainly in a pathway(s) critical for high-fat diet-induced adipogenesis in adults, but not for adipogenesis during development. Alternatively, maternal factors provided in utero and during nursing may compensate for the defects in WAT development in meltrin –/– mice. The roles of meltrin in adult adipogenesis suggest novel directions in the development of therapies for obesity and lipoatrophy.
Acknowledgments
We thank Nobuyuki Itoh, Toshiyuki Asaki, Kazuwa Nakao, and Hiroaki Masuzaki (Kyoto University) for technical advice and helpful discussions.
References
Moulin K, Truel N, Andre M, Arnauld E, Nibbelink M, Cousin B, Dani C, Penicaud L, Casteilla L 2001 Emergence during development of the white-adipocyte cell phenotype is independent of the brown-adipocyte cell phenotype. Biochem J 356:659–664
Hausman DB, DiGirolamo M, Bartness TJ, Hausman GJ, Martin RJ 2001 The biology of white adipocyte proliferation. Obes Rev 2:239–254
Yagami-Hiromasa T, Sato T, Kurisaki T, Kamijo K, Nabeshima Y, Fujisawa-Sehara A 1995 A metalloprotease-disintegrin participating in myoblast fusion. Nature 377:652–656
Blobel CP, Wolfsberg TG, Turck CW, Myles DG, Primakoff P, White JM 1992 A potential fusion peptide and an integrin ligand domain in a protein active in sperm-egg fusion. Nature 356:248–252
Cho C, Bunch DO, Faure JE, Goulding EH, Eddy EM, Primakoff P, Myles DG 1998 Fertilization defects in sperm from mice lacking fertilin ?. Science 281:1857–1859
Nishimura H, Cho C, Branciforte DR, Myles DG, Primakoff P 2001 Analysis of loss of adhesive function in sperm lacking cyritestin or fertilin ?. Dev Biol 233:204–213
Rooke J, Pan D, Xu T, Rubin GM 1996 KUZ, a conserved metalloprotease-disintegrin protein with two roles in Drosophila neurogenesis. Science 273:1227–1231
Peschon JJ, Slack JL, Reddy P, Stocking KL, Sunnarborg SW, Lee DC, Russell WE, Castner BJ, Johnson RS, Fitzner JN, Boyce RW, Nelson N, Kozlosky CJ, Wolfson MF, Rauch CT, Cerretti DP, Paxton RJ, March CJ, Black RA 1998 An essential role for ectodomain shedding in mammalian development. Science 282:1281–1284
Asakura M, Kitakaze M, Takashima S, Liao Y, Ishikura F, Yoshinaka T, Ohmoto H, Node K, Yoshino K, Ishiguro H, Asanuma H, Sanada S, Matsumura Y, Takeda H, Beppu S, Tada M, Hori M, Higashiyama S 2002 Cardiac hypertrophy is inhibited by antagonism of ADAM12 processing of HB-EGF: metaloprotease inhibitors as a new therapy. Nat Med 8:35–40
Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, Walsh A, Liu Z, Hayward B, Folz C, Manning SP, Bawa A, Saracino L, Thackston M, Benchekroun Y, Capparell N, Wang M, Adair R, Feng Y, Dubois J, FitzGerald MG, Huang H, Gibson R, Allen KM, Pedan A, Danzig MR, Umland SP, Egan RW, Cuss FM, Rorke S, Clough JB, Holloway JW, Holgate ST, Keith TP 2002 Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 418:426–430
Black R, White JM 1998 ADAMs: focus on the protease domain. Curr Opin Cell Biol 10:654–659
Seals DF, Courtneidge SA 2003 The ADAMs family of metalloproteases: multidomain proteins with multiple functions. Genes Dev 17:7–30[Free Full Text]
Werb Z, Yan Y 1998 A cellular striptease act. Science 282:1279–1280
Kurisaki T, Masuda A, Sudo K, Sakagami J, Higashiyama S, Matsuda Y, Nagabukuro A, Tsuji A, Nabeshima Y, Asano M, Iwakura Y, Fujisawa-Sehara A 2003 Phenotypic analysis of meltrin (ADAM12)-deficient mice: involvement of meltrin in adipogenesis and myogenesis. Mol Cell Biol 23:55–61
Kronqvist P, Kawaguchi N, Albrechtsen R, Xu X, Schroder HD, Moghadaszadeh B, Nielsen FC, Frohlich C, Engvall E, Wewer UM 2002 ADAM12 alleviates the skeletal muscle pathology in mdx dystrophic mice. Am J Pathol 161:1535–1540
Loechel F, Fox JW, Murphy G, Albrechtsen R, Wewer UM 2000 ADAM 12-S cleaves IGFBP-3 and IGFBP-5 and is inhibited by TIMP-3. Biochem Biophys Res Commun 278:511–515
Shi Z, Xu W, Loechel F, Wewer UM, Murphy LJ 2000 ADAM12, a disintegrin metalloprotease, interacts with insulin-like growth factor-binding protein-3. J Biol Chem 275:18574–18580
Hinkle CL, Sunnarborg SW, Loiselle D, Parker CE, Stevenson M, Russell WE, Lee DC 2004 Selective roles for TACE/ADAM17 in the shedding of the epidermal growth factor receptor ligand family. The juxtamembrane stalk determines cleavage efficiency. J Biol Chem 279:24179–24188
Shirakabe K, Wakatsuki S, Kurisaki T, Sehara-Fujisawa A 2001 Roles of meltrin ? (ADAM19) in the processing of neuregulin. J Biol Chem 276:9352–9358
Mori S, Tanaka M, Nanba D, Nishiwaki E, Ishiguro H, Higashiyama S, Matsuura N 2003 PACSIN3 binds ADAM12/meltrin and up-regulates ectodomain shedding of heparin-binding epidermal growth factor-like growth factor. J Biol Chem 278:46029–46034
Kurohara K, Komatsu K, Kurisaki T, Masuda A, Irie N, Asano M, Sudo K, Nabeshima Y, Iwakura Y, Sehara-Fujisawa A 2004 Essential roles of meltrin ? (ADAM19) in heart development. Dev Biol 267:14–28
Kawaguchi N, Xu X, Tajima R, Kronqvist P, Sundberg C, Loechel F, Albrechtsen R, Wewer UM 2003 ADAM12 protease induces adipogenesis in transgenic mice. Am J Pathol 160:1895–1903
Picard F, Gehin M, Annicotte J, Rocchi S, Champy MF, O’Malley BW, Chambon P, Auwerx J 2002 SRC-1 and TIF2 control energy balance between white and brown adipose tissues. Cell 111:931–941
Tanaka T, Yoshida N, Kishimoto T, Akira S 1997 Defective adipocyte differentiation in mice lacking the C/EBP? and/or C/EBP gene. EMBO J 16:7432–7443
Sakaue H, Konishi M, Ogawa W, Asaki T, Mori T, Yamasaki M, Takata M, Ueno H, kato S, Kasuga M, Itoh N 2002 Requirement of fibroblast growth factor 10 in development of white adipose tissue. Genes Dev 16:908–912
Tang Q-Q, Otto TC, Lane MD 2002 Mitotic clonal expansion: A synchronouse process required for adipogenesis. Proc Natl Acad Sci USA 100:44–49
Tang Q-Q, Otto TC, Lane MD 2003 CCAAT/enhancer-binding protein ? is required for mitotic clonal expansion during adipogenesis. Proc Natl Acad Sci USA 100:850–855
Matsumoto S, Kishida K, Shimomura I, Maeda N, Nagaretani H, Matsuda M, Nishizawa H, Kihara S, Funahashi T, Matsuzawa Y, Yamada A, Yamashita S, Tamura S, Kawata S 2002 Increased plasma HB-EGF associated with obesity and coronary artery disease. Biochem Biophys Res Commun 292:781–786
Mulligan C, Rochford J, Denyer G, Stephens R, Yeo G, Freeman T, Siddle K, O’Rahilly S 2002 Microarray analysis of insulin and insulin-like growth factor-1 (IGF-1) receptor signaling reveals the selective up-regulation of the mitogen heparin-binding EGF-like growth factor by IGF-1. J Biol Chem 277:42480–42487
Luetteke NC, Lee DC, Palmiter RD, Brinster RL, Sandgren EP 1993 Regulation of fat and muscle development by transforming growth factor in transgenic mice and in cultured cells. Cell Growth Differ 4:203–213
Baxter RC 2001 Signalling pathways involved in antiproliferative effects of IGFBP-3: a review. Mol Pathol 54:145–148
Modric T, Silha JV, Shi Z, Gui Y, Suwanichkul A, Durham SK, Powell DR, Murphy LJ 2001 Phenotypic manifestations of insulin-like growth factor-binding protein-3 overexpression in transgenic mice. Endocrinology 142:1958–1967
Sordella R, Jiang W, Chen GC, Curto M, Settleman J 2003 Modulation of Rho GTPase signaling regulates a switch between adipogenesis and myogenesis. Cell 113:147–158
Sadowski T, Dietrich S, Koschinsky F, Sedlacek R 2003 Matrix metalloproteinase 19 regulates insulin-like growth factor-mediated proliferation, migration, and adhesion in human keratinocytes through proteolysis of insulin-like growth factor binding protein-3. Mol Biol Cell 14:4569–4580
Manes S, Llorente M, Lacalle RA, Gomez-Mouton C, Kremer L, Mira E, Martinez AC 1999 The matrix metalloproteinase-9 regulates the insulin-like growth factor-triggered autocrine response in DU-145 carcinoma cells. J Biol Chem 274:6935–6945
Mochizuki S, Shimoda M, Shiomi T, Fujii Y, Okada Y 2004 ADAM28 is activated by MMP-7 (matrilysin-1) and cleaves insulin-like growth factor binding protein-3. Biochem Biophys Res Commun 315:79–84
Chavey C, Mari B, Monthouel MN, Bonnafous S, Anglard P, Van Obberghen E, Tartare-Deckert S 2003 Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278:11888–11896
Kawaguchi N, Sundberg C, Kveiborg M, Moghadaszadeh B, Asmar M, Dietrich N, Thodeti CK, Nielsen FC, Moller P, Mercurio AM, Albrechtsen R, Wewer UM 2003 ADAM12 induces actin cytoskeleton and extracellular matrix reorganization during early adipocyte differentiation by regulating ?1 integrin function. J Cell Sci 116:3893–3904(Megumi Masaki, Tomohiro K)