Differentially expressed genes in visceral or subcutaneous adipose tissue of obese men and women

There is growing evidence that the distribution of adipose tissue in the body is of importance in the development of metabolic complications of obesity, such as diabetes, hypertension, and hyperlipidemia. The aim of this study was to identify differentially expressed genes in subcutaneous and omental human adipose tissue in obese men, using a subtractive hybridization strategy. From the obtained set of differentially expressed transcripts, we also aimed to identify genes that have a sex-specific pattern of expression in omental or subcutaneous adipose tissue. Representational difference analysis (RDA) was performed on cDNA from subcutaneous and omental fat tissue from a man with extreme abdominal obesity. Forty-four putatively differentially expressed genes were identified. The obtained RDA products were spotted onto glass slides to screen for differential expression in other obese patients by using a microarray hybridization procedure. Five genes were confirmed to be differentially expressed in subcutaneous or omental adipose tissue from male or female obese patients. One gene was detected only in males and was found to be upregulated in subcutaneous tissue.The findings extend previous knowledge that different fat depots have differential gene expression and indicate that sex differences exist in adipose gene expression patterns. Linder, K., P. Arner, A. Flores-Morales, P. Tollet-Egnell, and G. Norstedt. Differentially expressed genes in visceral or subcutaneous adipose tissue of obese men and women. J. Lipid Res. 2004. 45: 148–154. There is growing evidence that the distribution of adipose tissue in the body is of importance in the development of metabolic complications of obesity, such as diabetes, hypertension, and hyperlipidemia. The aim of this study was to identify differentially expressed genes in subcutaneous and omental human adipose tissue in obese men, using a subtractive hybridization strategy. From the obtained set of differentially expressed transcripts, we also aimed to identify genes that have a sex-specific pattern of expression in omental or subcutaneous adipose tissue. Representational difference analysis (RDA) was performed on cDNA from subcutaneous and omental fat tissue from a man with extreme abdominal obesity. Forty-four putatively differentially expressed genes were identified. The obtained RDA products were spotted onto glass slides to screen for differential expression in other obese patients by using a microarray hybridization procedure. Five genes were confirmed to be differentially expressed in subcutaneous or omental adipose tissue from male or female obese patients. One gene was detected only in males and was found to be upregulated in subcutaneous tissue. The findings extend previous knowledge that different fat depots have differential gene expression and indicate that sex differences exist in adipose gene expression patterns. Linder, K., P. Arner, A. Flores-Morales, P. Tollet-Egnell, and G. Norstedt. Differentially expressed genes in visceral or subcutaneous adipose tissue of obese men and women. J. Lipid Res. 2004. 45: 148–154. It is well established that accumulation of visceral fat is associated with a higher risk for development of obesity-related diseases such as type 2 diabetes, cardiovascular disease, hypertension, and hyperlipidemia (1Larsson B. Bengtsson C. Bjorntorp P. Lapidus L. Sjostrom L. Svardsudd K. Tibblin G. Wedel H. Welin L. Wilhelmsen L. Is abdominal body fat distribution a major explanation for the sex difference in the incidence of myocardial infarction? The study of men born in 1913 and the study of women, Goteborg, Sweden.Am. J. Epidemiol. 1992; 135: 266-273Crossref PubMed Scopus (184) Google Scholar). Adipose tissue distribution differs between men and women, and visceral obesity is much more common among men than women (2Pi-Sunyer F.X. Dowling H.J. The effects of race and body fat distribution on insulin sensitivity.Trans. Am. Clin. Climatol. Assoc. 1995; 107: 175-185Google Scholar, 3Kissebah A.H. Krakower G.R. Regional adiposity and morbidity.Physiol. Rev. 1994; 74: 761-811Crossref PubMed Scopus (1009) Google Scholar). The metabolic and endocrine functions of adipose tissue from various depots differ in a way that may explain the association of visceral but not subcutaneous fat with obesity-related cardiovascular and metabolic problems (4Arner P. Not all fat is alike.Lancet. 1998; 351: 1301-1302Abstract Full Text Full Text PDF PubMed Scopus (139) Google Scholar). Regarding the metabolic function of fat, visceral adipose tissue is more sensitive to the stimulation of lipolysis by cathecolamines, whereas subcutaneous fat is more sensitive to the antilipolytic effects of insulin. Concerning endocrine function, visceral and subcutaneous adipocytes have different capacities to produce hormones and enzymes. Depot-related variation in mRNA expression has been shown for several genes, including leptin, TNF-α, angiotensinogen, PAI-1 (4Arner P. Not all fat is alike.Lancet. 1998; 351: 1301-1302Abstract Full Text Full Text PDF PubMed Scopus (139) Google Scholar), and recently, carboxypeptidase E and thrombospondin-1 (5Ramis J.M. Franssen-van Hal N.L. Kramer E. Llado I. Bouillaud F. Palou A. Keijer J. Carboxypeptidase E and thrombospondin-1 are differently expressed in subcutaneous and visceral fat of obese subjects.Cell. Mol. Life Sci. 2002; 59: 1960-1971Crossref PubMed Scopus (52) Google Scholar). The mechanisms responsible for depot differences in adipose function are unknown. It is possible that fat cells in various regions have different origins and, because of this, express different genes. Recent indirect evidence supports this idea, because newly formed adipocytes in human subcutaneous and visceral fat were shown to maintain the phenotypic site differences of mature adipocytes (6van Harmelen V. Dicker A. Ryden M. Hauner H. Lonnqvist F. Naslund E. Arner P. Increased lipolysis and decreased leptin production by human omental as compared with subcutaneous preadipocytes.Diabetes. 2002; 51: 2029-2036Crossref PubMed Scopus (158) Google Scholar). The major aim of the present study was to determine differences in gene expression patterns between subcutaneous and omental adipose tissue. We have used representational difference analysis (RDA) and microarrays to identify depot-specific genes that might explain the different metabolic and endocrine actions of these tissues, and possibly their different roles in the development of obesity-related diseases. Using RDA in two directions, i.e., with cDNA derived from subcutaneous adipose tissue subtracted from omental adipose tissue and the other way around, we cloned 44 cDNA fragments from a man with extreme abdominal obesity. We further used these 44 cDNA fragments to test differential expression in 6 other obese males and in 11 female patients using microarray hybridization. Combining RDA with microarray analysis enabled high throughput screening of the differentially cloned products. This procedure led to the identification of five genes with depot-specific expression in either males alone or in both sexes. The obese subjects (7 males and 11 females) included were members of a subgroup of patients participating in a clinical trial of adjustable gastric banding (7Thorne A. Lonnqvist F. Apelman J. Hellers G. Arner P. A pilot study of long-term effects of a novel obesity treatment: omentectomy in connection with adjustable gastric banding.Int. J. Obes. Relat. Metab. Disord. 2002; 26: 193-199Crossref PubMed Scopus (266) Google Scholar). From these subjects, it was possible to obtain fat specimens from omental as well as subcutaneous adipose tissue (the latter was taken from the surgical incision). All subjects were Caucasian and born in Sweden. Except for obesity, they were healthy and did not use any regular medication. Clinical data are recorded in Table 1. The study was approved by the Ethics Committee of Karolinska Institute, Stockholm. All individuals gave informed consent to participate in the study. The patients fasted from 10 PM the day before surgery, and only saline was given intravenously before adipose tissue was removed, which took place at the beginning of surgery. A tissue specimen (∼10 g) was taken from the abdominal surgical incision (subcutaneous fat) and from the major omentum. Premedication and general anesthesia were given as described (7Thorne A. Lonnqvist F. Apelman J. Hellers G. Arner P. A pilot study of long-term effects of a novel obesity treatment: omentectomy in connection with adjustable gastric banding.Int. J. Obes. Relat. Metab. Disord. 2002; 26: 193-199Crossref PubMed Scopus (266) Google Scholar). The specimen was immediately frozen in liquid nitrogen and stored at −70°C. Subcutaneous and omental adipose tissue from one male patient with extreme abdominal obesity was chosen for the procedure of subtractive cDNA hybridization using RDA (8Odeberg J. Wood T. Blucher A. Rafter J. Norstedt G. Lundeberg J. A cDNA RDA protocol using solid-phase technology suited for analysis in small tissue samples.Biomol. Eng. 2000; 17: 1-9Crossref PubMed Scopus (19) Google Scholar). This patient was selected from 16 of the above-mentioned patients who had undergone an abdominal computerized tomography (CT) scan prior to surgery (the remaining two subjects could not be investigated by CT for technical reasons). Intra-abdominal and subcutaneous fat areas were determined at vertebrae L4-L5. We selected the subject that had the greatest intra-abdominal:subcutaneous adipose area ratio. He was 57 years of age and had a body mass index of 39 kg/m2. His proportion of visceral versus total adipose area was 53% (426 cm2 visceral fat area, 352 cm2 subcutaneous fat area) and waist-to-hip ratio was 1.10. In attempts to verify differential gene expression, we used a pool of patient samples to minimize biological noise. Adipose tissue from 6 males and 11 females was used to prepare female or male, or visceral or subcutaneous RNA pools, respectively.TABLE 1Clinical characteristicsMeasureMalesFemalesPn711Age, years 40 ± 1 40 ± 110.87Body mass index, kg/m2 41 ± 2 43 ± 10.46Waist-to-hip ratio1.05 ± 0.040.94 ± 0.080.005PL-glucose, mmol/l6.1 ± 2.36.2 ± 2.50.96PL-insulin, mU/l22.6 ± 13.618.6 ± 9.50.47PL-cholesterol, mmol/l6.4 ± 0.95.4 ± 1.00.05PL-HDL-cholesterol, mmol/l1.1 ± 0.11.2 ± 0.10.85PL-triglycerides2.5 ± 1.11.8 ± 0.90.15PL, plasma. Values are mean ± SD; comparisons were by Student's unpaired t-test. Open table in a new tab PL, plasma. Values are mean ± SD; comparisons were by Student's unpaired t-test. Subcutaneous and omental adipose tissue from individual subjects was homogenized, and total RNA was isolated using Trizol reagent (Life Technologies, Inc., Grand Island, NY) according to the protocol supplied by the manufacturer. The quality of the RNA samples was ascertained on denaturing agarose gels, and the concentration was determined spectrophotometrically. For verification of RDA results, four RNA pools were prepared. Equal amounts of total RNA were mixed to generate omental and subcutaneous RNA pools, using starting material from 6 males or 11 females. cDNA RDA was performed as previously described (8Odeberg J. Wood T. Blucher A. Rafter J. Norstedt G. Lundeberg J. A cDNA RDA protocol using solid-phase technology suited for analysis in small tissue samples.Biomol. Eng. 2000; 17: 1-9Crossref PubMed Scopus (19) Google Scholar). One hundred thirty micrograms of total RNA from subcutaneous or omental fat tissue, respectively, was used to enrich poly(A)+ RNA (mRNA) using oligo-(deoxythymidine) paramagnetic beads (Dynal AS, Oslo, Norway). cDNA was synthesized using a kit purchased from Promega Corp. (Madison, WI). cDNA obtained from subcutaneous and omental adipose tissue from the chosen male subject was used as driver (subcutaneous) and tester (omental), or vice versa, to generate gene products (representations) that were induced or repressed in omental adipose tissue. After two rounds of subtraction and amplification, using tester:driver ratios of 1:100 and 1:800, difference products (DP2) were visualized on a 2% agarose gel. After being excised and from the were cloned the site of the and isolated were from excised and in were using the analysis of differentially expressed cDNA in total was performed using with on a analysis and were performed using data The Mol. 2000; Google Scholar). The were for with in the and expressed of the of the for using the Mol. PubMed Scopus Google Scholar). with more than one gene were than with more than to human genes or were for with were also for in the was performed by using the at in and of the cloned RDA products were further using and The by two were by and in 40 and 1 was to was on an agarose gel. that were from further The RDA products were on slides Inc., using a with four of RDA was spotted in at different on the genes, and were as to comparisons between cDNA was also as an After the slides were as described previously for analysis of gene PubMed Scopus Google and stored in a hybridization. the slides were in hybridization at for The protocol for and was as described previously for analysis of gene PubMed Scopus Google Scholar). micrograms of total RNA from the subcutaneous and omental female or male pool was used to generate cDNA The procedure was and was performed to differences in cDNA were synthesized by using (Life Technologies, in the of or for analysis of gene PubMed Scopus Google Scholar). and cDNA were and using The was to with hybridization of 10 and 10 After at for 2 the were to the and with a The was in a hybridization and the hybridization took place at for The was for analysis of gene PubMed Scopus Google and immediately using a The hybridization was for with The and were and using The at from and was and were used to or that were from The from was as the of the the a for a from we used a that was than the C. A. H. T. L. T. J. L. G. of by gene expression 2000; Scopus Google Scholar). For between we used the gene and as a gene in K. J. of for gene expression in human adipocytes and Metab. Res. PubMed Scopus Google Scholar). that had a ratio higher than the in at two of the were used for further The for of upregulated and genes was set to A. P. Lundeberg J. J. Norstedt G. analysis of the in effects of and on gene expression in the PubMed Scopus Google Scholar, P. A. Norstedt G. expression of the in effects of PubMed Scopus Google Scholar). in gene expression that were in both are Values are mean ± men and women were Student's unpaired was Concerning previous In the present we have used RDA and cDNA microarray screening to identify genes that might be in the development of obesity. The aim was to novel genes differentially expressed in subcutaneous and visceral adipose tissue and also to these genes in a male and a female obese RDA is a sensitive and subtraction of differentially expressed gene subtraction and rounds an of gene fragments from both cDNA the genes that are differentially for differences in gene expression between subcutaneous and omental human adipose tissue that are for visceral obesity, cDNA RDA was performed on adipose tissue obtained from a man with extreme visceral obesity. This who was selected from among subjects who had undergone abdominal had the greatest intra-abdominal:subcutaneous adipose area ratio. and from subcutaneous and omental adipose tissue, respectively, were used as driver and or vice versa, to generate differentially expressed The RDA in a in of the two cDNA from the to the products the a of products from to has from to in as different gene products as from and four from were excised from the and from were and After using data The Mol. 2000; Google Scholar), the were for with in the and of the using the Mol. PubMed Scopus Google Scholar). of 44 for and were and according to or The of the and the are shown in Table and of cDNA obtained in by cDNA gene or genes to function upregulated by in found in on from from expressed representational difference and in were compared with of in the and of the using and the with the was to the The were further according to or in adipose expression according to is The the was detected had a ratio higher than the in at of four in the microarray a pool of female or male obese Open table in a new tab expressed representational difference and in were compared with of in the and of the using and the with the was to the The were further according to or in adipose expression according to is The the was detected had a ratio higher than the in at of four in the microarray a pool of female or male obese the RDA obtained could also gene expression differences in other obese we the 44 different cDNA for differential expression using The obese subjects males and 11 females) included were members of a subgroup of patients participating in a clinical trial of adjustable gastric banding (7Thorne A. Lonnqvist F. Apelman J. Hellers G. Arner P. A pilot study of long-term effects of a novel obesity treatment: omentectomy in connection with adjustable gastric banding.Int. J. Obes. Relat. Metab. Disord. 2002; 26: 193-199Crossref PubMed Scopus (266) Google Scholar). the subjects were obese mass index The microarrays were with derived from subcutaneous and omental fat from male or female shown in Table 16 of the 44 RDA were were differentially expressed in the female pool and five in the male gene products are in Table A of the RDA and microarray is shown in gene expression differences in subcutaneous and omental fat in obese as upregulated and in subcutaneous versus omental adipose tissue using cDNA microarray in omental gene in subcutaneous are Values are the of in two different hybridization of of The variation of in one for these between and as upregulated and in subcutaneous versus omental adipose tissue using cDNA microarray Open table in a new tab The are Values are the of in two different hybridization of of The variation of in one for these between and not among the RDA was on the as a cDNA gave a in subcutaneous fat which several previous findings in P. of adipose J. 2000; PubMed Google an mRNA of leptin were also determined by analysis Harmelen V. P. A. J. Lonnqvist F. Arner P. from subcutaneous and visceral adipose tissue in 1998; PubMed Scopus Google Scholar), which in the subcutaneous adipose not The of this study was to for novel genes that may be to a accumulation of adipose tissue in the abdominal P. for human obesity in the adipose J. Obes. Relat. Metab. Disord. 2000; PubMed Scopus Google have used a gene to for depot-specific adipose gene We used RDA for differential because it is a that genes that are expressed at different in one the other in an We to for differential gene expression between omental and subcutaneous fat in a man with extreme visceral obesity, because we that we be to genes of in such a hundred cDNA fragments could be to 44 gene fragments were identified. of these had previously been found to be expressed in adipose tissue, and in and several K. K. I. K. K. of an expression of genes in the human adipose PubMed Scopus Google Scholar, M. H. T. H. K. H. I. K. K. T. The expression of in adipose tissue and concentration in patients with Res. PubMed Scopus Google Scholar). had not previously been to be expressed in adipose tissue. In differential one may i.e., genes that are not differentially In this we to this by a new set of male patients. In we to for sex differences in expression by a of obese females. a we used cDNA microarray The 44 RDA were spotted in onto glass slides and to cDNA from the above-mentioned male and female RNA Using the microarray we could verify that RDA for differentially expressed genes. of the RDA were detected by cDNA and of these were differentially expressed according to the chosen The of using differences in one patient as the for screening in set of patients was chosen as a to for common The of detected could indicate biological variation between the RDA patient and the but are also The cDNA microarray technology has in of and in using RDA in with cDNA one might be the cDNA The RDA are between and and the for cDNA microarray is by microarray has also been used in two on adipose tissue K. J. H. K. K. H. of genes expressed in the visceral adipose tissue of Lipid Res. 2000; Full Text Full Text PDF PubMed Google Scholar, M. Hal N.L. T. Keijer J. gene expression in and PubMed Google Scholar). the as described by i.e., of the RDA were not detected with the microarray in a of detected We could of the RDA and of these were differentially expressed according to chosen Using the M. Hal N.L. T. Keijer J. gene expression in and PubMed Google confirmed differential expression for of obtained RDA with microarray It is that to in this did not all Concerning the not all of the differentially expressed genes are the procedure. of in the mRNA may generate than of expressed genes in the fragments may not be by which has a differential expression with in subcutaneous fat, was not among the the site difference in leptin expression was found in the on the cDNA The that of the were found in only a one that and more more genes. The in the RDA small expression differences and of The of microarray might to verify these including such as might be more and the RNA amounts could also analysis of individual patient the above-mentioned it that the differences between subcutaneous and omental adipose tissue are in this study. a is because is to mRNA expression and the of the biological is it is of to the possible clinical of Intra-abdominal obesity is associated with in the We an of mRNA in the omental fat in obese males and females. is a of the It is a that between to and has an in of mRNA in human adipose tissue was in by and E. P. H. in mRNA expression of the by the adipocytes in human subcutaneous and visceral adipose 2000; PubMed Scopus Google Scholar). In with the present they found a difference in mRNA with of in omental adipose tissue in subjects with obesity. The present study and previous E. P. H. in mRNA expression of the by the adipocytes in human subcutaneous and visceral adipose 2000; PubMed Scopus Google that from adipose tissue in might to of has been to insulin and to in in obese and obese patients A. is to insulin by insulin in obese 1998; PubMed Scopus Google Scholar). A novel and is the of in subcutaneous adipose tissue of men and women. or is a expression is upregulated in and cells on a and 1998; PubMed Scopus Google Scholar). to a of that have been in several human such as disease, and G. functions and 2002; PubMed Google Scholar). is induced in cells to adipocytes and the cells are with hormones K. F. of in cells to adipocytes and by J. PubMed Scopus Google Scholar). The from fat cells is also by K. F. of in cells to adipocytes and by J. PubMed Scopus Google Scholar), and this stimulation be by insulin F. K. of adipose by PubMed Scopus Google Scholar). We also found of in the subcutaneous adipose tissue in both males and females. This is with by P. is to and is expressed at high in adipose 1992; Full Text PDF PubMed Google Scholar), including high expression of in subcutaneous adipose tissue. is to The of an in and induced It is that adipocytes are to the of the and and M. Hal N.L. T. Keijer J. gene expression in and PubMed Google Scholar, P. H. A. A. P. differences in expression of in adipose and Clin. Metab. Google Scholar, B. analysis of gene expression in human adipose tissue using Res. 2000; PubMed Scopus Google Scholar). of these are for which is a of in human fat cells Harmelen V. K. E. J. K. A. Arner P. in the of from isolated human fat cells by and Full Text Full Text PDF PubMed Scopus Google Scholar, M. K. and men and the 2000; PubMed Scopus Google Scholar). is a of of the which of the P. P. expression of a new of the gene 1992; PubMed Scopus Google Scholar). The that is differentially with in omental adipose of either is in of the functions of shown in Table we cloned several of one of for differential it was detected and differentially expressed only in the male subcutaneous The was not found in the performed in the expressed in the which and of genes, a gene on was This gene for a with the present we are as to this a new a or an In we have used an RDA in with cDNA microarray to identify genes with different expression in different fat We for the differential expression of and between subcutaneous adipose tissue and omental adipose tissue, which might be of importance in development of visceral obesity. This also detected a gene not previously described included in which might be a that is in subcutaneous adipose tissue of men but in women. is to the biological of the latter for development of visceral obesity in The for technical This study was by from the the the of the and the and