Differential Expression of Matrix Metalloproteinase 3 (MMP3) in Preadi
http://www.100md.com
《糖尿病学杂志》
the Diabetes Molecular Genetics Section, Phoenix Epidemiology and Clinical Research Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Phoenix, Arizona
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MMP, matrix metalloproteinase; SNP, single nucleotide polymorphism; UTR, untranslated region
ABSTRACT
Prior microarray studies comparing global gene expression patterns in preadipocytes/stromal vascular cells isolated from nonobese nondiabetic versus obese nondiabetic Pima Indians showed that matrix metalloproteinase 9 (MMP9) is upregulated in obese subjects. The current study targeted analysis of nine additional MMP genes that cluster to a region on chromosome 11q22 that is linked to BMI and percent body fat. Differential-display PCR showed that MMP3 is downregulated in preadipocytes/stromal vascular cells from obese subjects, and real-time PCR showed that MMP3 expression levels are negatively correlated with percent body fat. To determine whether variants within MMP3 are responsible for its altered expression, MMP3 was sequenced, and seven representative variants were genotyped in 1,037 Pima subjects for association analyses. Two variants were associated with both BMI and type 2 diabetes, and two additional variants were associated with type 2 diabetes alone; however, none of these variants were associated with MMP3 expression levels. We propose that the MMP3 pathway is altered in human obesity, but this alteration may be the result of a combination of genetic variation within the MMP3 locus itself, as well as variation in additional factors, either primary or secondary to obesity, that regulate expression of the MMP3 gene.
The Pima Indians of Arizona have a high prevalence of obesity and type 2 diabetes, and both of these diseases have a strong genetic basis (1,2). To identify novel genes and/or genetic pathways that may be important for the development of obesity in this Native-American population, Nair et al. (3) compared Affymetrix oligonucleotide microarray expression profiles from adipocyte precursor cells (preadipocytes/stromal vascular cells) isolated from nonobese and obese subjects and identified 218 genes that were differentially expressed between these two groups. One of the genes most notably upregulated in preadipocytes/stromal vascular cells isolated from obese subjects was MMP9, and this upregulation was confirmed by real-time PCR (3). Matrix metalloproteinase 9 (MMP9) is a member of a large family of MMPs, which have been proposed to be involved in adipose tissue remodeling (4,5). Therefore, other MMP family members, in addition to MMP9, may potentially contribute to the etiology of obesity (4,5).
A cluster of nine additional MMP genes (MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP3, MMP12, and MMP13) maps within a 500-kb region located at chromosome 11q22. In the current study, this cluster of MMP genes at chromosome 11q22 was targeted for analysis because this chromosomal region has previously been linked to measures of obesity. Linkage to BMI at chromosome 11q22 has been reported in studies of Nigerian families (6) and Old Order Amish (7), and linkage to BMI at 11q22-25 has also been reported in a Dutch cohort of type 2 diabetic patients (8). Our group has also reported linkage to percent body fat at chromosome 11q22 in a study of 277 Pima-Indian siblings (9), and, in a larger study of 966 Pima-Indian siblings, we reported linkage to BMI and type 2 diabetes at a more telomeric region (11q23-24) on chromosome 11 (10). Among the nine MMP genes that map to chromosome 11q22, only MMP3 was detected in preadipocytes/stromal vascular cells isolated from Pima Indians. This gene was further analyzed using the techniques of differential-display PCR, real-time PCR, direct sequencing, and association analyses for obesity and type 2 diabetes.
RESEARCH DESIGN AND METHODS
Subjects were participants of ongoing longitudinal studies of the etiology of type 2 diabetes among the Gila River Indian Community in Arizona (11). Subjects for the adipose tissue biopsy were admitted as inpatients into our Clinical Research Center and were confirmed to be nondiabetic, as determined by an oral glucose tolerance test; the results of which were interpreted according to World Health Organization criteria (12). Body composition was estimated by underwater weighing until January 1996, and thereafter it was estimated by dual-energy X-ray absorptiometry (DPX-1; Lunar Radiation, Madison, WI). A conversion equation derived from comparative analyses was used to make the estimates of body composition equivalent between the two methods (9). For the differential-display PCR experiments, 32 nondiabetic full-blooded Pima-Indian subjects were studied. None of these subjects were first-degree relatives. Of these 32 subjects, 16 were nonobese and 16 obese, and they were pair matched for age and sex (Table 1). For real-time PCR, 47 nondiabetic full-blooded Pima subjects (BMI range 20–70 kg/m2) were studied (Table 1), of which 7 of these subjects had been previously studied for MMP9 expression. Sequencing of the MMP3 gene was performed on DNA from 18 (9 nonobese and 9 obese) subjects who had also been studied by differential-display PCR. Variants detected in the MMP3 gene were genotyped in a family-based sample of 1,037 (573 diabetic, 464 nondiabetic) Pima Indians for association analysis. All studies were approved by the Tribal Council and the institutional review board of the National Institutes of Diabetes and Digestive and Kidney Diseases.
Isolation of preadipocytes/stromal vascular cells.
Subjects for adipose tissue biopsies were admitted as inpatients to our Clinical Research Center, and, after an overnight fast, underwent a subcutaneous abdominal needle biopsy under local anesthesia with 1% lidocaine. Collagenase digestion of the subcutaneous abdominal adipose tissue biopsy samples was performed as previously described (13,14). Briefly, the preadipocyte-containing infranatant was collected into a separate tube and washed several times with Hanks’ balanced salt solution. The stromal vascular fraction pellet containing preadipocytes/stromal vascular cells was resuspended in standard medium consisting of Medium 199 (Life Technologies, Grand Island, NY) supplemented with 1 μg/ml amphotericin B, 100 units/ml penicillin G sodium, 100 μg/ml streptomycin sulfate, 2 mmol/l Glutamax-1, and 10% heat-inactivated fetal bovine serum (Life Technologies). The suspension was strained through a sterile 25-um stainless steel tissue sieve (Thermo EC, Holbrook, NY). The filtrate was transferred to a T75 culture flask and maintained in an incubator at 37°C in 5% CO2. Cells were allowed to attach, and the next day floating erythrocytes were removed by aspiration and the culture media replenished. At subconfluency, the cultured cells were trypsinized and plated at a concentration of 1.5 x 106 cells per 15-cm dish for RNA extraction, which was carried out 14 days after the biopsy date. Media was changed every 2–3 days throughout the culturing period. There were no differences (in terms of cell number or visible morphology) between the cultures from the nonobese and obese subjects. Cultures were routinely screened for expression levels of ECSM2 (endothelial cell–specific molecule 2), a corollary of endothelial contamination, to confirm that it is consistent and low among subjects.
Isolation of total RNA and synthesis of cDNA.
Total RNA was extracted from the preadipocytes/stromal vascular cells, using an RNeasy Mini Kit (Qiagen, Valencia, CA). To remove any residual DNA, the purified RNA was treated with DNase using an RNase-free DNase set (Qiagen). First-strand cDNA was synthesized from preadipocytes/stromal vascular cell total RNA, using a BD Advantage RT-for-PCR kit (BD Bioscience/Clontech, Palo Alto, CA), following the manufacturer’s instructions.
Differential-display PCR for the nine MMP genes located at 11q22.
PCR amplification of MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP3, MMP12, and MMP13 was performed using specific primers (sequences for the primers are available from the authors on request). The PCR mix consisted of the following components: 1 μl first-strand cDNA, 1 μl each of forward and reverse primers (20 μmol/l each), 2 μl of 10x cDNA PCR buffer, 0.2 μl of 50x dNTP mix, 0.4 μl Advantage cDNA polymerase mix (Advantage cDNA PCR kit; BD Bioscience/Clontech), and 14.4 μl dH2O for a total volume of 20 μl. Each of the cDNA samples was amplified in duplicate, and the products were run and visualized on 2% high-resolution agarose gels (Agarose SFR; AMRESCO, Solon, OH).
Real-time quantitative PCR for MMP3.
MMP3 expression was quantitated by real-time PCR, using a predesigned gene expression assay (Assays-on-Demand gene expression products; Applied Biosystems, Foster City, CA). Real-time PCR was performed using an ABI –7700 sequence detection system (Applied Biosystems), following the manufacturer’s instructions. Assays were performed in triplicate, and the mean values were used to calculate expression levels, using the relative standard curve method. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous control to obtain normalized values. Raw GAPDH levels were plotted against both BMI and percentage of body fat to establish that there was no relationship between the GAPDH housekeeping gene and these phenoytpes.
Sequencing of MMP3.
Overlapping primers were designed to sequence 2,900 nucleotides upstream of the transcription start site, all 10 exons, all 9 introns, and 2,400 nucleotides of the 3'-untranslated region (UTR) of the MMP3 gene. Sequencing was performed on DNA samples from 18 of the subjects (9 nonobese and 9 obese) studied by differential-display PCR, using a Big Dye Terminator v1.1 cycle sequencing kit (Applied Biosystems).
Genotyping of sequence variations in MMP3.
All variants were genotyped in 1,037 Pima Indians. The two insertion/deletions variants, –1613 5T/6T (rs3025058) and +1494 7T/8T, were genotyped by direct sequencing as described above. The variant –1475 (G/A) was genotyped by SNaPshot minisequencing (SNaPshot multiplex kit; Applied Biosystems), following the manufacturer’s protocol. PCR-amplified samples were analyzed on an ABI Prism 3700 DNA analyzer with GeneScan analysis software (Applied Biosystems).
All other polymorphisms were genotyped by allelic discrimination (either Assays-on-Demand single nucleotide polymorphism [SNP] genotyping products or Assays-by-Design service; Applied Biosystems) and were scanned using an ABI-7700 sequence detection system (Applied Biosystems).
Statistical analysis.
Statistical evaluations of the data were performed using a statistical analysis system (SAS Institute, Cary, NC). For continuous variables, the general estimating equation procedure was used to adjust for appropriate covariates, including age and sex. General linear regression models were used to analyze the relationships between mRNA expression levels and obesity (percent body fat).
RESULTS
Differential-display PCR of each of the nine MMP genes positioned at chromosome 11q22.
Gene expression levels for MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP3, MMP12, and MMP13 were analyzed by differential-display PCR, using cDNA synthesized from preadipocytes/stromal vascular cell total RNA isolated from 16 nonobese and 16 obese Pima subjects (Table 1). Among these nine MMP genes, only MMP3 expression was detected in preadipocytes/stromal vascular cells using this technique. Comparison of MMP3 expression levels between age- and sex-matched pairs of nonobese versus obese subjects showed that in 7 of the 16 pairs, MMP3 expression was high in the nonobese subjects but barely detectable or undetectable in the matched obese subjects (Fig. 1). In the other nine pairs, MMP3 expression was high in both the nonobese and matched obese subjects. All of the nonobese subjects (both male and female subjects) had comparable levels of MMP3 expression. Among the seven obese subjects that had downregulated MMP3 expression, four were men and three were women. The identity of the differentially expressed product was confirmed to be MMP3 by sequencing.
Real-time quantitative PCR for MMP3.
To confirm and quantitate the observed decrease in expression of MMP3 in some obese subjects, MMP3 expression levels were further examined by real-time quantitative PCR, using a larger sample (n = 47) of Pima Indians who had a range of BMI values (mean BMI = 40 kg/m2, range 20–70). Assays were performed in triplicate, and the averages were used to calculate expression levels, which were normalized to GAPDH as an endogenous control. Linear regression analyses were performed to determine whether mRNA expression levels were correlated with measures of obesity. MMP3 mRNA levels were negatively correlated with percent body fat (P = 0.004, adjusted for age and sex) (Fig. 2).
Identification of sequence variants in MMP3.
To determine whether the decrease in MMP3 expression in obese subjects is attributable to variation in the MMP3 gene itself, or is the result of variation in a factor/pathway that regulates MMP3, the MMP3 gene was sequenced in DNA from 18 of the subjects studied by differential-display PCR (9 nonobese and 9 obese). The entire gene was sequenced (2,900 bp upstream of the transcription start site, all 10 exons and 9 introns, and 2,400 bp of the 3'-UTR).
Sequencing of MMP3 identified 24 variants (Fig. 3). Among the eight variants found in the 5'-UTR, only the G/A transition at –1475 appears to be novel. Seven of the 5'-UTR variants were single–base pair substitutions, and one variant (rs3025058, often reported in the literature as 5A/6A) was a thymine insertion/deletion in a string of thymines (5T/6T) at position –1613. Among the 11 variants identified in the coding and intronic regions, 1 (rs679620) predicts a nonsynonymous substitution (Lys45Glu) in exon 2, and 2 (rs602128 and rs520540) predict synonymous substitutions in exons 2 and 8, respectively. Two of the intronic variants are thymine insertion/deletions at +496 and +1494. The T at +496 (rs11422799) is in a string of 11 thymines in intron 1, and the +T at +1494 (novel) is in a poly (T) tract located 37 bp downstream of the 3' splice site for exon 4. Among the five variants in the 3'-UTR, one is novel (T at +9505).
Genotyping and association analyses of the MMP3 variants.
Based on the genotypic information obtained while sequencing the 18 subjects, several of the variants identified in MMP3 were found to be in high linkage disequilibrium (D' >0.98, r2 = 1), and therefore all 24 variants could be divided into seven genotypic groups. To examine whether any of these variants were associated with obesity and/or type 2 diabetes, one representative SNP from each group was genotyped in 1,037 (573 diabetic/464 nondiabetic) Pima Indians for association analyses. None of the genotypic data deviated from the expected under Hardy-Weinberg equilibrium. The linkage disequilibrium coefficients (D' and r2) between each of the seven representative variants are given in Table 2.
Two representative variants, rs650108 and rs522616, were associated with BMI (P = 0.05 and 0.0009, respectively, adjusted for age, sex, birth date, family membership, and Pima heritage) when analyzed under a dominant model (Table 3). Both of these variants were also associated with type 2 diabetes (P = 0.05 and 0.005, respectively, adjusted for age, sex, birth date, family membership, and Pima heritage) when analyzed under a dominant model (Table 4). Two additional variants, rs3025058 and the Lys45Glu missense mutation (rs679620), were also associated with type 2 diabetes (P = 0.005 and 0.02, respectively, under a dominant model adjusted for age, sex, birth date, family membership, and Pima heritage) (Table 4), but they were not associated with BMI (Table 3). However, among the 47 subjects who had measurements of MMP3 expression levels by real-time PCR, there was no correlation between genotype and MMP3 expression for any of these variants (data not shown).
DISCUSSION
Previous Affymetrix oligonucleotide microarray expression profiling of cultured abdominal subcutaneous preadipocytes/stromal vascular cells isolated from the adipose tissue of 14 nonobese and 14 obese nondiabetic subjects identified MMP9 as being upregulated in obese subjects (3). A total of 7 subjects from this prior study of MMP9 and 40 additional subjects were studied in the current report, which shows that another member of the MMP family, MMP3, was differentially expressed in preadipocytes/stromal vascular cells, but the expression of several other MMP genes (MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP12, and MMP13) clustered on 11q22 was not detectable by differential-display PCR. Using differential-display PCR, we showed that MMP3 mRNA levels were decreased in preadipocytes/stromal vascular cells isolated from obese Pima-Indian subjects compared with nonobese subjects, and real-time quantitative PCR showed a significant negative correlation between MMP3 mRNA expression levels and percent body fat. It is possible that technical procedures during the initial adipose tissue biopsy may have resulted in different proportions of endothelial "contaminating" cells in the preadipocytes/stromal vascular cells isolated from obese versus nonobese subjects, which could have lead to altered levels of cell-specific genes. However, we have previously shown that the expression level of endothelial cell–specific molecule 2, a corollary of endothelial contamination, was consistent among these subjects (14).
MMPs may play an important role in adipose tissue remodeling (4,5). In two previous studies, Chavey et al. (4) and Maquoi et al. (5) examined the expression levels of various MMPs and TIMPs (tissue inhibitors of MMP) in adipose tissue from mouse models of obesity. Both studies found significant changes in MMP and TIMP expression levels in obese adipose tissue and suggested that these proteins may be involved in obesity-mediated adipose tissue remodeling. MMPs also appear to be important for preadipocyte differentiation (adipogenesis) (4,5,15–17). Croissandeau et al. (15) demonstrated that MMP2 and MMP9 play a role in the differentiation of murine 3T3-LI preadipocytes into mature adipocytes, and Bouloumie et al. (16) showed that MMP2 and MMP9 may be involved in modulating murine 3T3F442A adipocyte differentiation.
There are a number of recent studies both corroborating and conflicting our finding of decreased expression of MMP3 in preadipocytes/stromal vascular cells isolated from obese subjects. For example, Maquoi et al. (18) demonstrated that MMP3–/– mice had increased adipocyte hypertrophy when fed a high-fat diet, and Alexander et al. (17) showed that MMP3–/– (Str1–/–) mice had accelerated adipocyte differentiation during mammary gland involution. In the same study, Alexander et al. noted that MMP3–/– mice backcrossed onto another background (FVB/N) became obese with age (17). In contrast to our study, both Maquoi et al. (5), in an earlier study, and Chavey et al. (4) found that MMP3 expression levels were upregulated in adipose tissue isolated from obese mice. The reasons for these disparities with our results are unclear. They could possibly be caused by differences in model systems (human versus mouse models, or diet-induced versus non–diet-induced obesity) or alternatively by differences in adipose tissue depots (subcutaneous versus gonadal).
Because MMP3 mapped to a region of linkage to percent body fat and BMI, we sought to determine whether expression differences were attributable to variation at this locus, or whether the differential expression was caused by variation in a protein/factor that regulates MMP3 and the position of this gene in a region of linkage was simply serendipitous. Sequencing of 13 kb of the MMP3 gene identified 24 sequence variations, many of which were in very high or complete linkage disequilibrium. Among the representative SNPs that were genotyped for association analyses, rs650108 and rs522616 were associated with both BMI and type 2 diabetes. Because rs650108 is located in intron 8 500 bases from the splice-donor site and 650 bases from the splice-acceptor site, it is unlikely that this variant directly contributes to the two phenotypes, but instead it may be in linkage disequilibrium with a putative functional polymorphism(s). The rs522616 polymorphism is located in the 5'-UTR 709 bases upstream of the transcription start site, and it does not appear to be in a known conserved regulatory element; however, it is in a string of 4 bases that are conserved across several species.
Two additional SNPs, a promoter SNP (rs3025058) and a nonsynonymous coding SNP (rs679620), were not associated with BMI but were associated with type 2 diabetes. The promoter variant rs3025058 (alternatively termed 5A/6A) is a common polymorphism located in a SIRE (stromelysin interleukin-1–responsive element; –1614G(T)TTTTTCCCCCC-ATCAAAG–1595) important for interleukin-1–induced DNA binding (19). It has been shown that two proteins bind to this site, ZBP-89 and nuclear factor-B, and that binding of ZBP-89 is dependent on the string of Cs (20), whereas the binding of nuclear factor-B is dependent on the string of Ts (19). The coding variant rs679620 predicts an amino acid substitution (Lys45 [basic] to Glu [acidic]). Lys45 is part of the 82–amino acid propeptide pro-MMP3, which is involved in maintaining MMP3 latency (21); however, the functional importance of this particular lysine residue is unclear. These two variants are representative SNPs for several other polymorphisms that were detected by sequencing of this gene; however, none of the nongenotyped SNPs appear to be located within any regions that would support a functional role. It should also be mentioned that the homozygous minor genotype for these two variants is somewhat rare (4.87 and 5.22%, respectively) (data not shown), so it may be possible that this low frequency is exaggerating the association with diabetes.
The promoter polymorphism rs3025058 (5A/6A) has previously been shown to affect MMP3 promoter activity in a number of cell lines (22), including human fibroblasts (23), rat smooth muscle cells (20), and human macrophages (24). In each of these studies, the 5A allele had higher promoter activity compared with the 6A allele (20,22–24). In contrast, the 5A allele was associated with lower MMP3 serum levels, compared with the 6A allele, in human subjects (25). In our study, the promoter variant rs522616, but not rs3025058, was associated with BMI; however, no association was observed between the rs522616 genotype and MMP3 mRNA levels in preadipocytes/stromal vascular cells. Therefore, further functional studies are needed to assess the biological relevance of this SNP. It is possible that this variant has no role in controlling MMP3 activity, and the position of the MMP3 gene within a region of linkage to percent body fat/BMI occurred by chance. In support of this, we found that adjustment of our linkage peak at 11q22 for the effect of any of these single SNPs did not reduce the evidence for linkage to percent body fat previously observed at this region (data not shown). Therefore, it is possible the downregulation of this gene in some obese subjects is attributable to a change in the level of a protein/hormone that affects MMP3 expression, rather than variation in the MMP3 gene itself. Alternatively, MMP3 expression may be too complex (controlled by multiple variants within the 5'-UTR of MMP3, as well as being influenced by external factors secondary to obesity, such as insulin), such that we were unable to detect an association between a single promoter variant and MMP3 expression levels. Therefore, although MMP3–/– mice become obese, our data in Pima Indians cannot distinguish whether downregulation of MMP3 leads to obesity, and/or is a consequence of obesity, in humans. In light of our conflicting data that MMP3 mRNA levels are negatively correlated with percent body fat in a small number of research subjects, but that a MMP3 promoter variant that is associated with BMI is not correlated with MMP3 mRNA levels, further studies such as protein expression and/or activity assays are needed to substantiate our MMP3 expression data.
In conclusion, we propose that the MMP3 pathway is altered in human obesity. We believe this alteration may be the result of a combination of genetic variation within the MMP3 locus itself, as well as variation in additional factors, either primary or secondary to obesity, that regulate expression of the MMP3 gene.
ACKNOWLEDGMENTS
This research was supported by the Intramural Research Program of the National Institutes of Health, National Institutes of Diabetes and Digestive and Kidney Diseases. M.T.T. is supported by a Mentor Grant from the American Diabetes Association.
We gratefully acknowledge the volunteers and leaders of the Gila River Indian Community, whose cooperation made these studies possible. We also acknowledge Dr. Joy Bunt and the nurses of the Clinical Research Center, and Dr. Arline Salbe and the Metabolic Kitchen staff, for the care of the research volunteers.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Knowler WC, Pettitt DJ, Saad MF, Charles MA, Nelson RG, Howard BV, Bogardus C, Bennett PH: Obesity in the Pima Indians: its magnitude and relationship with diabetes. Am J Clin Nutr 53:S1543–S1551, 1991
Baier LJ, Hanson RL: Genetic studies of the etiology of type 2 diabetes mellitus in Pima Indians: hunting for pieces to a complicated puzzle (Review). Diabetes 53:1181–1186, 2004
Nair S, Lee YH, Rousseau E, Cam M, Tataranni PA, Baier LJ, Bogardus C, Permana PA: Increased expression of inflammation-related genes in cultured preadipocytes/ stromal vascular cells from obese compared with non-obese Pima Indians. Diabetologia 48:1784–1788, 2005
Chavey C, Mari B, Monthouel MN, Bonnafous S, Anglard P, Van Obberghen E, Tartare-Deckert S: Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278:11888–11896, 2003
Maquoi E, Munaut C, Colige A, Collen D, Lijnen HR: Modulation of adipose tissue expression of murine matrix metalloproteinases and their tissue inhibitors with obesity. Diabetes 51:1093–1101, 2002
Adeyemo A, Luke A, Cooper R, Wu X, Tayo B, Zhu X, Rotimi C, Bouzekri N, Ward R: A genome-wide scan for body mass index among Nigerian families. Obes Res 11:266–273, 2003
Platte P, Papanicolaou GJ, Johnston J, Klein CM, Doheny KF, Pugh EW, Roy-Gagnon MH, Stunkard AJ, Francomano CA, Wilson AF: A study of linkage and association of body mass index in the Old Order Amish. Am J Med Genet 121:71–80, 2003
van Tilburg JH, Sandkuijl LA, Strengman E, Pearson PL, van Haeften TW, Wijmenga C: Variance-component analysis of obesity in type 2 diabetes confirms loci on chromosomes 1q and 11q. Obes Res 11:1290–1294, 2003
Norman RA, Thompson DB, Foroud T, Garvey WT, Bennett PH, Bogardus C, Ravussin E: Genomewide search for genes influencing percent body fat in Pima Indians: suggestive linkage at chromosome 11q21–q22: Pima Diabetes Gene Group. Am J Hum Genet 60:166–173, 1997
Hanson RL, Ehm MG, Pettitt DJ, Prochazka M, Thompson DB, Timberlake D, Foroud T, Kobes S, Baier L, Burns DK, Almasy L, Blangero J, Garvey WT, Bennett PH, Knowler WC: An autosomal genomic scan for loci linked to type II diabetes mellitus and body mass index in Pima Indians. Am J Hum Genet 63:1130–1138, 1998
Knowler WC, Bennett PH, Hamman RF, Miller M: Diabetes incidence and prevalence in Pima Indians: a 19-fold greater incidence than in Rochester, Minnesota. Am J Epidemiol 108:497–505, 1978
World Health Organization. Diabetes Mellitus: Report of a WHO Study Group Geneva, World Health Org., 1985 (Tech. Rep. Ser., no. 727)
Lee YH, Nair S, Rousseau E, Allison DB, Page GP, Tataranni PA, Bogardus C, Permana PA: Microarray profiling of isolated abdominal subcutaneous adipocytes from obese vs non-obese Pima Indians: increased expression of inflammation-related genes. Diabetologia 48:1776–1783, 2005
Permana PA, Nair S, Lee YH, Luczy-Bachman G, Vozarova De Courten B, Tataranni PA: Subcutaneous abdominal preadipocyte differentiation in vitro inversely correlates with central obesity. Am J Physiol Endocrinol Metabol 286:E958–E962, 2004
Croissandeau G, Chretien M, Mbikay M: Involvement of matrix metalloproteinases in the adipose conversion of 3T3–L1 preadipocytes. Biochem J 364:739–746, 2002
Bouloumie A, Sengenes C, Portolan G, Galitzky J, Lafontan M: Adipocyte produces matrix metalloproteinases 2 and 9: involvement in adipose differentiation. Diabetes 50:2080–2086, 2001
Alexander CM, Selvarajan S, Mudgett J, Werb Z: Stromelysin-1 regulates adipogenesis during mammary gland involution. J Cell Biol 152:693–703, 2001
Maquoi E, Demeulemeester D, Voros G, Collen D, Lijnen HR: Enhanced nutritionally induced adipose tissue development in mice with stromelysin-1 gene inactivation. Thromb Haemost 89:696–704, 2003
Borghaei RC, Rawlings PL Jr, Javadi M, Woloshin J: NF-kappaB binds to a polymorphic repressor element in the MMP-3 promoter. Biochem Biophys Res Commun 316:182–188, 2004
Ye S, Whatling C, Watkins H, Henney AM: Human stromelysin gene promoter activity is modulated by transcription factor ZBP-89. FEBS Lett 450:268–272, 1999
Freimark BD, Feeser WS, Rosenfeld SA: Multiple sites of the propeptide region of human stromelysin-1 are required for maintaining a latent form of the enzyme. J Biol Chem 269:26982–26987, 1994
Ye S: Influence of matrix metalloproteinase genotype on cardiovascular disease susceptibility and outcome. Cardiovasc Res 69:636–645, 2006
Ye S, Eriksson P, Hamsten A, Kurkinen M, Humphries SE, Henney AM: Progression of coronary atherosclerosis is associated with a common genetic variant of the human stromelysin-1 promoter which results in reduced gene expression. J Biol Chem 271:13055–13060, 1996
Beyzade S, Zhang S, Wong YK, Day IN, Eriksson P, Ye S: Influences of matrix metalloproteinase-3 gene variation on extent of coronary atherosclerosis and risk of myocardial infarction. J Am Coll Cardiol 41:2130–2137, 2003
Samnegard A, Silveira A, Lundman P, Boquist S, Odeberg J, Hulthe J, McPheat W, Tornvall P, Bergstrand L, Ericsson CG, Hamsten A, Eriksson P: Serum matrix metalloproteinase-3 concentration is influenced by MMP-3–1612 5A/6A promoter genotype and associated with myocardial infarction. J Intern Med 258:411–419, 2005(Michael T. Traurig, Paska A. Permana, Sa)
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MMP, matrix metalloproteinase; SNP, single nucleotide polymorphism; UTR, untranslated region
ABSTRACT
Prior microarray studies comparing global gene expression patterns in preadipocytes/stromal vascular cells isolated from nonobese nondiabetic versus obese nondiabetic Pima Indians showed that matrix metalloproteinase 9 (MMP9) is upregulated in obese subjects. The current study targeted analysis of nine additional MMP genes that cluster to a region on chromosome 11q22 that is linked to BMI and percent body fat. Differential-display PCR showed that MMP3 is downregulated in preadipocytes/stromal vascular cells from obese subjects, and real-time PCR showed that MMP3 expression levels are negatively correlated with percent body fat. To determine whether variants within MMP3 are responsible for its altered expression, MMP3 was sequenced, and seven representative variants were genotyped in 1,037 Pima subjects for association analyses. Two variants were associated with both BMI and type 2 diabetes, and two additional variants were associated with type 2 diabetes alone; however, none of these variants were associated with MMP3 expression levels. We propose that the MMP3 pathway is altered in human obesity, but this alteration may be the result of a combination of genetic variation within the MMP3 locus itself, as well as variation in additional factors, either primary or secondary to obesity, that regulate expression of the MMP3 gene.
The Pima Indians of Arizona have a high prevalence of obesity and type 2 diabetes, and both of these diseases have a strong genetic basis (1,2). To identify novel genes and/or genetic pathways that may be important for the development of obesity in this Native-American population, Nair et al. (3) compared Affymetrix oligonucleotide microarray expression profiles from adipocyte precursor cells (preadipocytes/stromal vascular cells) isolated from nonobese and obese subjects and identified 218 genes that were differentially expressed between these two groups. One of the genes most notably upregulated in preadipocytes/stromal vascular cells isolated from obese subjects was MMP9, and this upregulation was confirmed by real-time PCR (3). Matrix metalloproteinase 9 (MMP9) is a member of a large family of MMPs, which have been proposed to be involved in adipose tissue remodeling (4,5). Therefore, other MMP family members, in addition to MMP9, may potentially contribute to the etiology of obesity (4,5).
A cluster of nine additional MMP genes (MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP3, MMP12, and MMP13) maps within a 500-kb region located at chromosome 11q22. In the current study, this cluster of MMP genes at chromosome 11q22 was targeted for analysis because this chromosomal region has previously been linked to measures of obesity. Linkage to BMI at chromosome 11q22 has been reported in studies of Nigerian families (6) and Old Order Amish (7), and linkage to BMI at 11q22-25 has also been reported in a Dutch cohort of type 2 diabetic patients (8). Our group has also reported linkage to percent body fat at chromosome 11q22 in a study of 277 Pima-Indian siblings (9), and, in a larger study of 966 Pima-Indian siblings, we reported linkage to BMI and type 2 diabetes at a more telomeric region (11q23-24) on chromosome 11 (10). Among the nine MMP genes that map to chromosome 11q22, only MMP3 was detected in preadipocytes/stromal vascular cells isolated from Pima Indians. This gene was further analyzed using the techniques of differential-display PCR, real-time PCR, direct sequencing, and association analyses for obesity and type 2 diabetes.
RESEARCH DESIGN AND METHODS
Subjects were participants of ongoing longitudinal studies of the etiology of type 2 diabetes among the Gila River Indian Community in Arizona (11). Subjects for the adipose tissue biopsy were admitted as inpatients into our Clinical Research Center and were confirmed to be nondiabetic, as determined by an oral glucose tolerance test; the results of which were interpreted according to World Health Organization criteria (12). Body composition was estimated by underwater weighing until January 1996, and thereafter it was estimated by dual-energy X-ray absorptiometry (DPX-1; Lunar Radiation, Madison, WI). A conversion equation derived from comparative analyses was used to make the estimates of body composition equivalent between the two methods (9). For the differential-display PCR experiments, 32 nondiabetic full-blooded Pima-Indian subjects were studied. None of these subjects were first-degree relatives. Of these 32 subjects, 16 were nonobese and 16 obese, and they were pair matched for age and sex (Table 1). For real-time PCR, 47 nondiabetic full-blooded Pima subjects (BMI range 20–70 kg/m2) were studied (Table 1), of which 7 of these subjects had been previously studied for MMP9 expression. Sequencing of the MMP3 gene was performed on DNA from 18 (9 nonobese and 9 obese) subjects who had also been studied by differential-display PCR. Variants detected in the MMP3 gene were genotyped in a family-based sample of 1,037 (573 diabetic, 464 nondiabetic) Pima Indians for association analysis. All studies were approved by the Tribal Council and the institutional review board of the National Institutes of Diabetes and Digestive and Kidney Diseases.
Isolation of preadipocytes/stromal vascular cells.
Subjects for adipose tissue biopsies were admitted as inpatients to our Clinical Research Center, and, after an overnight fast, underwent a subcutaneous abdominal needle biopsy under local anesthesia with 1% lidocaine. Collagenase digestion of the subcutaneous abdominal adipose tissue biopsy samples was performed as previously described (13,14). Briefly, the preadipocyte-containing infranatant was collected into a separate tube and washed several times with Hanks’ balanced salt solution. The stromal vascular fraction pellet containing preadipocytes/stromal vascular cells was resuspended in standard medium consisting of Medium 199 (Life Technologies, Grand Island, NY) supplemented with 1 μg/ml amphotericin B, 100 units/ml penicillin G sodium, 100 μg/ml streptomycin sulfate, 2 mmol/l Glutamax-1, and 10% heat-inactivated fetal bovine serum (Life Technologies). The suspension was strained through a sterile 25-um stainless steel tissue sieve (Thermo EC, Holbrook, NY). The filtrate was transferred to a T75 culture flask and maintained in an incubator at 37°C in 5% CO2. Cells were allowed to attach, and the next day floating erythrocytes were removed by aspiration and the culture media replenished. At subconfluency, the cultured cells were trypsinized and plated at a concentration of 1.5 x 106 cells per 15-cm dish for RNA extraction, which was carried out 14 days after the biopsy date. Media was changed every 2–3 days throughout the culturing period. There were no differences (in terms of cell number or visible morphology) between the cultures from the nonobese and obese subjects. Cultures were routinely screened for expression levels of ECSM2 (endothelial cell–specific molecule 2), a corollary of endothelial contamination, to confirm that it is consistent and low among subjects.
Isolation of total RNA and synthesis of cDNA.
Total RNA was extracted from the preadipocytes/stromal vascular cells, using an RNeasy Mini Kit (Qiagen, Valencia, CA). To remove any residual DNA, the purified RNA was treated with DNase using an RNase-free DNase set (Qiagen). First-strand cDNA was synthesized from preadipocytes/stromal vascular cell total RNA, using a BD Advantage RT-for-PCR kit (BD Bioscience/Clontech, Palo Alto, CA), following the manufacturer’s instructions.
Differential-display PCR for the nine MMP genes located at 11q22.
PCR amplification of MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP3, MMP12, and MMP13 was performed using specific primers (sequences for the primers are available from the authors on request). The PCR mix consisted of the following components: 1 μl first-strand cDNA, 1 μl each of forward and reverse primers (20 μmol/l each), 2 μl of 10x cDNA PCR buffer, 0.2 μl of 50x dNTP mix, 0.4 μl Advantage cDNA polymerase mix (Advantage cDNA PCR kit; BD Bioscience/Clontech), and 14.4 μl dH2O for a total volume of 20 μl. Each of the cDNA samples was amplified in duplicate, and the products were run and visualized on 2% high-resolution agarose gels (Agarose SFR; AMRESCO, Solon, OH).
Real-time quantitative PCR for MMP3.
MMP3 expression was quantitated by real-time PCR, using a predesigned gene expression assay (Assays-on-Demand gene expression products; Applied Biosystems, Foster City, CA). Real-time PCR was performed using an ABI –7700 sequence detection system (Applied Biosystems), following the manufacturer’s instructions. Assays were performed in triplicate, and the mean values were used to calculate expression levels, using the relative standard curve method. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous control to obtain normalized values. Raw GAPDH levels were plotted against both BMI and percentage of body fat to establish that there was no relationship between the GAPDH housekeeping gene and these phenoytpes.
Sequencing of MMP3.
Overlapping primers were designed to sequence 2,900 nucleotides upstream of the transcription start site, all 10 exons, all 9 introns, and 2,400 nucleotides of the 3'-untranslated region (UTR) of the MMP3 gene. Sequencing was performed on DNA samples from 18 of the subjects (9 nonobese and 9 obese) studied by differential-display PCR, using a Big Dye Terminator v1.1 cycle sequencing kit (Applied Biosystems).
Genotyping of sequence variations in MMP3.
All variants were genotyped in 1,037 Pima Indians. The two insertion/deletions variants, –1613 5T/6T (rs3025058) and +1494 7T/8T, were genotyped by direct sequencing as described above. The variant –1475 (G/A) was genotyped by SNaPshot minisequencing (SNaPshot multiplex kit; Applied Biosystems), following the manufacturer’s protocol. PCR-amplified samples were analyzed on an ABI Prism 3700 DNA analyzer with GeneScan analysis software (Applied Biosystems).
All other polymorphisms were genotyped by allelic discrimination (either Assays-on-Demand single nucleotide polymorphism [SNP] genotyping products or Assays-by-Design service; Applied Biosystems) and were scanned using an ABI-7700 sequence detection system (Applied Biosystems).
Statistical analysis.
Statistical evaluations of the data were performed using a statistical analysis system (SAS Institute, Cary, NC). For continuous variables, the general estimating equation procedure was used to adjust for appropriate covariates, including age and sex. General linear regression models were used to analyze the relationships between mRNA expression levels and obesity (percent body fat).
RESULTS
Differential-display PCR of each of the nine MMP genes positioned at chromosome 11q22.
Gene expression levels for MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP3, MMP12, and MMP13 were analyzed by differential-display PCR, using cDNA synthesized from preadipocytes/stromal vascular cell total RNA isolated from 16 nonobese and 16 obese Pima subjects (Table 1). Among these nine MMP genes, only MMP3 expression was detected in preadipocytes/stromal vascular cells using this technique. Comparison of MMP3 expression levels between age- and sex-matched pairs of nonobese versus obese subjects showed that in 7 of the 16 pairs, MMP3 expression was high in the nonobese subjects but barely detectable or undetectable in the matched obese subjects (Fig. 1). In the other nine pairs, MMP3 expression was high in both the nonobese and matched obese subjects. All of the nonobese subjects (both male and female subjects) had comparable levels of MMP3 expression. Among the seven obese subjects that had downregulated MMP3 expression, four were men and three were women. The identity of the differentially expressed product was confirmed to be MMP3 by sequencing.
Real-time quantitative PCR for MMP3.
To confirm and quantitate the observed decrease in expression of MMP3 in some obese subjects, MMP3 expression levels were further examined by real-time quantitative PCR, using a larger sample (n = 47) of Pima Indians who had a range of BMI values (mean BMI = 40 kg/m2, range 20–70). Assays were performed in triplicate, and the averages were used to calculate expression levels, which were normalized to GAPDH as an endogenous control. Linear regression analyses were performed to determine whether mRNA expression levels were correlated with measures of obesity. MMP3 mRNA levels were negatively correlated with percent body fat (P = 0.004, adjusted for age and sex) (Fig. 2).
Identification of sequence variants in MMP3.
To determine whether the decrease in MMP3 expression in obese subjects is attributable to variation in the MMP3 gene itself, or is the result of variation in a factor/pathway that regulates MMP3, the MMP3 gene was sequenced in DNA from 18 of the subjects studied by differential-display PCR (9 nonobese and 9 obese). The entire gene was sequenced (2,900 bp upstream of the transcription start site, all 10 exons and 9 introns, and 2,400 bp of the 3'-UTR).
Sequencing of MMP3 identified 24 variants (Fig. 3). Among the eight variants found in the 5'-UTR, only the G/A transition at –1475 appears to be novel. Seven of the 5'-UTR variants were single–base pair substitutions, and one variant (rs3025058, often reported in the literature as 5A/6A) was a thymine insertion/deletion in a string of thymines (5T/6T) at position –1613. Among the 11 variants identified in the coding and intronic regions, 1 (rs679620) predicts a nonsynonymous substitution (Lys45Glu) in exon 2, and 2 (rs602128 and rs520540) predict synonymous substitutions in exons 2 and 8, respectively. Two of the intronic variants are thymine insertion/deletions at +496 and +1494. The T at +496 (rs11422799) is in a string of 11 thymines in intron 1, and the +T at +1494 (novel) is in a poly (T) tract located 37 bp downstream of the 3' splice site for exon 4. Among the five variants in the 3'-UTR, one is novel (T at +9505).
Genotyping and association analyses of the MMP3 variants.
Based on the genotypic information obtained while sequencing the 18 subjects, several of the variants identified in MMP3 were found to be in high linkage disequilibrium (D' >0.98, r2 = 1), and therefore all 24 variants could be divided into seven genotypic groups. To examine whether any of these variants were associated with obesity and/or type 2 diabetes, one representative SNP from each group was genotyped in 1,037 (573 diabetic/464 nondiabetic) Pima Indians for association analyses. None of the genotypic data deviated from the expected under Hardy-Weinberg equilibrium. The linkage disequilibrium coefficients (D' and r2) between each of the seven representative variants are given in Table 2.
Two representative variants, rs650108 and rs522616, were associated with BMI (P = 0.05 and 0.0009, respectively, adjusted for age, sex, birth date, family membership, and Pima heritage) when analyzed under a dominant model (Table 3). Both of these variants were also associated with type 2 diabetes (P = 0.05 and 0.005, respectively, adjusted for age, sex, birth date, family membership, and Pima heritage) when analyzed under a dominant model (Table 4). Two additional variants, rs3025058 and the Lys45Glu missense mutation (rs679620), were also associated with type 2 diabetes (P = 0.005 and 0.02, respectively, under a dominant model adjusted for age, sex, birth date, family membership, and Pima heritage) (Table 4), but they were not associated with BMI (Table 3). However, among the 47 subjects who had measurements of MMP3 expression levels by real-time PCR, there was no correlation between genotype and MMP3 expression for any of these variants (data not shown).
DISCUSSION
Previous Affymetrix oligonucleotide microarray expression profiling of cultured abdominal subcutaneous preadipocytes/stromal vascular cells isolated from the adipose tissue of 14 nonobese and 14 obese nondiabetic subjects identified MMP9 as being upregulated in obese subjects (3). A total of 7 subjects from this prior study of MMP9 and 40 additional subjects were studied in the current report, which shows that another member of the MMP family, MMP3, was differentially expressed in preadipocytes/stromal vascular cells, but the expression of several other MMP genes (MMP7, MMP20, MMP27, MMP8, MMP10, MMP1, MMP12, and MMP13) clustered on 11q22 was not detectable by differential-display PCR. Using differential-display PCR, we showed that MMP3 mRNA levels were decreased in preadipocytes/stromal vascular cells isolated from obese Pima-Indian subjects compared with nonobese subjects, and real-time quantitative PCR showed a significant negative correlation between MMP3 mRNA expression levels and percent body fat. It is possible that technical procedures during the initial adipose tissue biopsy may have resulted in different proportions of endothelial "contaminating" cells in the preadipocytes/stromal vascular cells isolated from obese versus nonobese subjects, which could have lead to altered levels of cell-specific genes. However, we have previously shown that the expression level of endothelial cell–specific molecule 2, a corollary of endothelial contamination, was consistent among these subjects (14).
MMPs may play an important role in adipose tissue remodeling (4,5). In two previous studies, Chavey et al. (4) and Maquoi et al. (5) examined the expression levels of various MMPs and TIMPs (tissue inhibitors of MMP) in adipose tissue from mouse models of obesity. Both studies found significant changes in MMP and TIMP expression levels in obese adipose tissue and suggested that these proteins may be involved in obesity-mediated adipose tissue remodeling. MMPs also appear to be important for preadipocyte differentiation (adipogenesis) (4,5,15–17). Croissandeau et al. (15) demonstrated that MMP2 and MMP9 play a role in the differentiation of murine 3T3-LI preadipocytes into mature adipocytes, and Bouloumie et al. (16) showed that MMP2 and MMP9 may be involved in modulating murine 3T3F442A adipocyte differentiation.
There are a number of recent studies both corroborating and conflicting our finding of decreased expression of MMP3 in preadipocytes/stromal vascular cells isolated from obese subjects. For example, Maquoi et al. (18) demonstrated that MMP3–/– mice had increased adipocyte hypertrophy when fed a high-fat diet, and Alexander et al. (17) showed that MMP3–/– (Str1–/–) mice had accelerated adipocyte differentiation during mammary gland involution. In the same study, Alexander et al. noted that MMP3–/– mice backcrossed onto another background (FVB/N) became obese with age (17). In contrast to our study, both Maquoi et al. (5), in an earlier study, and Chavey et al. (4) found that MMP3 expression levels were upregulated in adipose tissue isolated from obese mice. The reasons for these disparities with our results are unclear. They could possibly be caused by differences in model systems (human versus mouse models, or diet-induced versus non–diet-induced obesity) or alternatively by differences in adipose tissue depots (subcutaneous versus gonadal).
Because MMP3 mapped to a region of linkage to percent body fat and BMI, we sought to determine whether expression differences were attributable to variation at this locus, or whether the differential expression was caused by variation in a protein/factor that regulates MMP3 and the position of this gene in a region of linkage was simply serendipitous. Sequencing of 13 kb of the MMP3 gene identified 24 sequence variations, many of which were in very high or complete linkage disequilibrium. Among the representative SNPs that were genotyped for association analyses, rs650108 and rs522616 were associated with both BMI and type 2 diabetes. Because rs650108 is located in intron 8 500 bases from the splice-donor site and 650 bases from the splice-acceptor site, it is unlikely that this variant directly contributes to the two phenotypes, but instead it may be in linkage disequilibrium with a putative functional polymorphism(s). The rs522616 polymorphism is located in the 5'-UTR 709 bases upstream of the transcription start site, and it does not appear to be in a known conserved regulatory element; however, it is in a string of 4 bases that are conserved across several species.
Two additional SNPs, a promoter SNP (rs3025058) and a nonsynonymous coding SNP (rs679620), were not associated with BMI but were associated with type 2 diabetes. The promoter variant rs3025058 (alternatively termed 5A/6A) is a common polymorphism located in a SIRE (stromelysin interleukin-1–responsive element; –1614G(T)TTTTTCCCCCC-ATCAAAG–1595) important for interleukin-1–induced DNA binding (19). It has been shown that two proteins bind to this site, ZBP-89 and nuclear factor-B, and that binding of ZBP-89 is dependent on the string of Cs (20), whereas the binding of nuclear factor-B is dependent on the string of Ts (19). The coding variant rs679620 predicts an amino acid substitution (Lys45 [basic] to Glu [acidic]). Lys45 is part of the 82–amino acid propeptide pro-MMP3, which is involved in maintaining MMP3 latency (21); however, the functional importance of this particular lysine residue is unclear. These two variants are representative SNPs for several other polymorphisms that were detected by sequencing of this gene; however, none of the nongenotyped SNPs appear to be located within any regions that would support a functional role. It should also be mentioned that the homozygous minor genotype for these two variants is somewhat rare (4.87 and 5.22%, respectively) (data not shown), so it may be possible that this low frequency is exaggerating the association with diabetes.
The promoter polymorphism rs3025058 (5A/6A) has previously been shown to affect MMP3 promoter activity in a number of cell lines (22), including human fibroblasts (23), rat smooth muscle cells (20), and human macrophages (24). In each of these studies, the 5A allele had higher promoter activity compared with the 6A allele (20,22–24). In contrast, the 5A allele was associated with lower MMP3 serum levels, compared with the 6A allele, in human subjects (25). In our study, the promoter variant rs522616, but not rs3025058, was associated with BMI; however, no association was observed between the rs522616 genotype and MMP3 mRNA levels in preadipocytes/stromal vascular cells. Therefore, further functional studies are needed to assess the biological relevance of this SNP. It is possible that this variant has no role in controlling MMP3 activity, and the position of the MMP3 gene within a region of linkage to percent body fat/BMI occurred by chance. In support of this, we found that adjustment of our linkage peak at 11q22 for the effect of any of these single SNPs did not reduce the evidence for linkage to percent body fat previously observed at this region (data not shown). Therefore, it is possible the downregulation of this gene in some obese subjects is attributable to a change in the level of a protein/hormone that affects MMP3 expression, rather than variation in the MMP3 gene itself. Alternatively, MMP3 expression may be too complex (controlled by multiple variants within the 5'-UTR of MMP3, as well as being influenced by external factors secondary to obesity, such as insulin), such that we were unable to detect an association between a single promoter variant and MMP3 expression levels. Therefore, although MMP3–/– mice become obese, our data in Pima Indians cannot distinguish whether downregulation of MMP3 leads to obesity, and/or is a consequence of obesity, in humans. In light of our conflicting data that MMP3 mRNA levels are negatively correlated with percent body fat in a small number of research subjects, but that a MMP3 promoter variant that is associated with BMI is not correlated with MMP3 mRNA levels, further studies such as protein expression and/or activity assays are needed to substantiate our MMP3 expression data.
In conclusion, we propose that the MMP3 pathway is altered in human obesity. We believe this alteration may be the result of a combination of genetic variation within the MMP3 locus itself, as well as variation in additional factors, either primary or secondary to obesity, that regulate expression of the MMP3 gene.
ACKNOWLEDGMENTS
This research was supported by the Intramural Research Program of the National Institutes of Health, National Institutes of Diabetes and Digestive and Kidney Diseases. M.T.T. is supported by a Mentor Grant from the American Diabetes Association.
We gratefully acknowledge the volunteers and leaders of the Gila River Indian Community, whose cooperation made these studies possible. We also acknowledge Dr. Joy Bunt and the nurses of the Clinical Research Center, and Dr. Arline Salbe and the Metabolic Kitchen staff, for the care of the research volunteers.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Knowler WC, Pettitt DJ, Saad MF, Charles MA, Nelson RG, Howard BV, Bogardus C, Bennett PH: Obesity in the Pima Indians: its magnitude and relationship with diabetes. Am J Clin Nutr 53:S1543–S1551, 1991
Baier LJ, Hanson RL: Genetic studies of the etiology of type 2 diabetes mellitus in Pima Indians: hunting for pieces to a complicated puzzle (Review). Diabetes 53:1181–1186, 2004
Nair S, Lee YH, Rousseau E, Cam M, Tataranni PA, Baier LJ, Bogardus C, Permana PA: Increased expression of inflammation-related genes in cultured preadipocytes/ stromal vascular cells from obese compared with non-obese Pima Indians. Diabetologia 48:1784–1788, 2005
Chavey C, Mari B, Monthouel MN, Bonnafous S, Anglard P, Van Obberghen E, Tartare-Deckert S: Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278:11888–11896, 2003
Maquoi E, Munaut C, Colige A, Collen D, Lijnen HR: Modulation of adipose tissue expression of murine matrix metalloproteinases and their tissue inhibitors with obesity. Diabetes 51:1093–1101, 2002
Adeyemo A, Luke A, Cooper R, Wu X, Tayo B, Zhu X, Rotimi C, Bouzekri N, Ward R: A genome-wide scan for body mass index among Nigerian families. Obes Res 11:266–273, 2003
Platte P, Papanicolaou GJ, Johnston J, Klein CM, Doheny KF, Pugh EW, Roy-Gagnon MH, Stunkard AJ, Francomano CA, Wilson AF: A study of linkage and association of body mass index in the Old Order Amish. Am J Med Genet 121:71–80, 2003
van Tilburg JH, Sandkuijl LA, Strengman E, Pearson PL, van Haeften TW, Wijmenga C: Variance-component analysis of obesity in type 2 diabetes confirms loci on chromosomes 1q and 11q. Obes Res 11:1290–1294, 2003
Norman RA, Thompson DB, Foroud T, Garvey WT, Bennett PH, Bogardus C, Ravussin E: Genomewide search for genes influencing percent body fat in Pima Indians: suggestive linkage at chromosome 11q21–q22: Pima Diabetes Gene Group. Am J Hum Genet 60:166–173, 1997
Hanson RL, Ehm MG, Pettitt DJ, Prochazka M, Thompson DB, Timberlake D, Foroud T, Kobes S, Baier L, Burns DK, Almasy L, Blangero J, Garvey WT, Bennett PH, Knowler WC: An autosomal genomic scan for loci linked to type II diabetes mellitus and body mass index in Pima Indians. Am J Hum Genet 63:1130–1138, 1998
Knowler WC, Bennett PH, Hamman RF, Miller M: Diabetes incidence and prevalence in Pima Indians: a 19-fold greater incidence than in Rochester, Minnesota. Am J Epidemiol 108:497–505, 1978
World Health Organization. Diabetes Mellitus: Report of a WHO Study Group Geneva, World Health Org., 1985 (Tech. Rep. Ser., no. 727)
Lee YH, Nair S, Rousseau E, Allison DB, Page GP, Tataranni PA, Bogardus C, Permana PA: Microarray profiling of isolated abdominal subcutaneous adipocytes from obese vs non-obese Pima Indians: increased expression of inflammation-related genes. Diabetologia 48:1776–1783, 2005
Permana PA, Nair S, Lee YH, Luczy-Bachman G, Vozarova De Courten B, Tataranni PA: Subcutaneous abdominal preadipocyte differentiation in vitro inversely correlates with central obesity. Am J Physiol Endocrinol Metabol 286:E958–E962, 2004
Croissandeau G, Chretien M, Mbikay M: Involvement of matrix metalloproteinases in the adipose conversion of 3T3–L1 preadipocytes. Biochem J 364:739–746, 2002
Bouloumie A, Sengenes C, Portolan G, Galitzky J, Lafontan M: Adipocyte produces matrix metalloproteinases 2 and 9: involvement in adipose differentiation. Diabetes 50:2080–2086, 2001
Alexander CM, Selvarajan S, Mudgett J, Werb Z: Stromelysin-1 regulates adipogenesis during mammary gland involution. J Cell Biol 152:693–703, 2001
Maquoi E, Demeulemeester D, Voros G, Collen D, Lijnen HR: Enhanced nutritionally induced adipose tissue development in mice with stromelysin-1 gene inactivation. Thromb Haemost 89:696–704, 2003
Borghaei RC, Rawlings PL Jr, Javadi M, Woloshin J: NF-kappaB binds to a polymorphic repressor element in the MMP-3 promoter. Biochem Biophys Res Commun 316:182–188, 2004
Ye S, Whatling C, Watkins H, Henney AM: Human stromelysin gene promoter activity is modulated by transcription factor ZBP-89. FEBS Lett 450:268–272, 1999
Freimark BD, Feeser WS, Rosenfeld SA: Multiple sites of the propeptide region of human stromelysin-1 are required for maintaining a latent form of the enzyme. J Biol Chem 269:26982–26987, 1994
Ye S: Influence of matrix metalloproteinase genotype on cardiovascular disease susceptibility and outcome. Cardiovasc Res 69:636–645, 2006
Ye S, Eriksson P, Hamsten A, Kurkinen M, Humphries SE, Henney AM: Progression of coronary atherosclerosis is associated with a common genetic variant of the human stromelysin-1 promoter which results in reduced gene expression. J Biol Chem 271:13055–13060, 1996
Beyzade S, Zhang S, Wong YK, Day IN, Eriksson P, Ye S: Influences of matrix metalloproteinase-3 gene variation on extent of coronary atherosclerosis and risk of myocardial infarction. J Am Coll Cardiol 41:2130–2137, 2003
Samnegard A, Silveira A, Lundman P, Boquist S, Odeberg J, Hulthe J, McPheat W, Tornvall P, Bergstrand L, Ericsson CG, Hamsten A, Eriksson P: Serum matrix metalloproteinase-3 concentration is influenced by MMP-3–1612 5A/6A promoter genotype and associated with myocardial infarction. J Intern Med 258:411–419, 2005(Michael T. Traurig, Paska A. Permana, Sa)