INTRODUCTION

Top Abstract Introduction Materials & Methods Results & Discussion References

ADIPOSE TISSUE IS THE MAJOR SITE for energy storage and plays an important role in maintaining glucose homeostasis (12). Abnormalities in hormonal and nutritional regulation of this tissue have been implicated in the pathophysiology of obesity, diabetes, and atherosclerosis. It is therefore crucial to investigate regulation of adipose tissue metabolism, in particular that of human adipose tissue.

Although the causes of human obesity are not yet elucidated, this disease is commonly associated with excessive fat storage, leading to adipocyte hypertrophy. In addition, obesity is highly prevalent in type II diabetic patients, and diabetic patients on insulin therapy tend to gain weight. Therefore, the objective of this study was to investigate the role of insulin in regulating fatty acid synthesis in human adipose tissue.

Several studies have demonstrated the presence of the key enzymes for fatty acid synthesis (4, 23) and the importance of human adipose tissue in de novo fatty acid synthesis (2, 5). In addition, recent studies demonstrated that liver plays a minor role in de novo human lipogenesis and suggested that adipose tissue may be the principal lipogenic tissue in humans (1). As an approach toward understanding the role that nutrients and hormones play in regulation of human adipose tissue metabolism, we recently developed a cell culture system in which human adipocytes can be maintained viable and metabolically active for several days (21). We have demonstrated that glucose utilization as well as activities of lipogenic enzymes including fatty acid synthase (FAS) were increased by insulin in a dose-dependent manner in cultured human adipocytes (21).

FAS is a key lipogenic enzyme that catalyzes all of the reactions involved in the synthesis of long-chain saturated fatty acid (palmitate) from acetyl CoA, malonyl CoA, and NADPH. In addition, this enzyme is highly regulated by nutrients and hormones in all species tested. In rodent and murine cell lines, FAS is suppressed by fasting (25), polyunsaturated fatty acids (8), and diabetes (25), whereas it is induced by feeding high-carbohydrate diets (25), glucose (10), obesity (9), and insulin (25). We have recently identified an insulin response sequence in the proximal promoter of the rat FAS gene (20) that mediates regulation of FAS gene transcription by insulin. This element overlaps the CAAT box region and binds upstream stimulatory factor (29). A second insulin response element has been identified within a DNase hypersensitive region (30). Although regulation of human FAS gene serves as a better tool toward our understanding of lipid metabolism disorders in human obesity and diabetes, very limited information is available concerning its regulation. Human FAS has been shown to be expressed in several human tissues (14, 26). Semenkovich and his collaborators (26, 27) have reported that FAS expression was induced by glucose in Hep G2 cells at the posttranscriptional level. However, neither nutritional nor hormonal regulation of FAS gene expression in human adipocytes has been reported before.

Because our previous study showed induction of human lipogenic enzymes, including FAS, by insulin (21), the objective of the present work was to determine the mechanism(s) whereby insulin regulates expression of the FAS gene in cultured human adipocytes from lean and obese patients as well as in Hep G2 human hepatoma cells. We report for the first time that FAS expression was induced by insulin at the transcriptional level in human adipocytes and at the posttranscriptional level in Hep G2 cells.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results & Discussion References

Human subjects. Nonobese, nondiabetic [body mass index (BMI) <27] as well as morbidly obese women (BMI >34) with an average age of 37 ± 15 yr were used in this study. To our knowledge, these patients did not exhibit any other disorders or diseases. These patients required or elected abdominal surgery or liposuction. No information was available on the medications that the patients were taking or on their dietary habits or lipid profile before surgery. We maintained adipocytes isolated from these patients in culture for several days to eliminate any differences contributed by in vivo circulating factors. This proposal was approved by the institutional review board for human subjects and the Committee for Research Protocols of the University of Tennessee in Knoxville.

Isolation of human adipocytes and culture conditions. Adipose tissue was obtained from subcutaneous abdominal fat of the above fasted patients and cultured as we recently described (21). Fat was removed at the time of the surgery in a sterile environment and was immersed in Hanks' balanced salt solution (GIBCO BRL) supplemented with penicillin (100 U/ml), streptomycin (100 µg/ml), and gentamicin (50 µg/ml). Adipose tissue was then washed several times with Hanks' balanced salt solution to remove the majority of connective tissue and blood clots. The tissue was minced into small fragments, which were digested with type I collagenase (1 mg/ml, GIBCO BRL) in a shaking water bath at 37°C for 30-60 min in a polypropylene flask. Cells were then filtered through a sterile nylon filter (350 µm mesh). The suspension was centrifuged at 500 g for 1 min to separate the pelleted stromal vascular fraction from the floating adipocyte fraction and was washed three times with Hanks' balanced salt solution. Adipocytes were then resuspended in Dulbecco's modified Eagle's medium (DMEM) supplemented with HEPES (15 mM), glucose (25 mM), bovine serum albumin (1%), 50 nM adenosine, antibiotics, and 1% fetal bovine serum (standard medium). Cells were subsequently cultured in suspension in sterile polypropylene tubes in a humidified incubator at 37°C under 5% CO₂ and 95% air. The culture medium containing the adipocytes (still in suspension) was removed 24 h later, and the adipocytes were cultured in fresh standard medium. The cells were maintained for an additional 3-6 days in this medium and then incubated overnight in serum-free medium before insulin treatment as indicated in figure legends. Media were changed every day during the culture. Trypan blue exclusion test was conducted in all cultures to confirm cell viability.

Culture of Hep G2 cells. Hep G2 cells were purchased from American Type Culture Collection (Rockville, MD). Cells were grown and maintained in DMEM supplemented with 10% fetal bovine serum. Subconfluent cells were incubated overnight in serum-free medium before insulin treatment as indicated in figure legends.

FAS enzyme activity. Adipocytes were washed with phosphate-buffered saline (PBS) and homogenized in sucrose buffer, and then FAS activity was assayed spectrophotometrically as we previously described by measuring the rate of oxidation of NADPH (15). One unit of enzyme activity equals one nanomole of NADPH oxidized per minute per microgram DNA that was assayed fluorometrically (3).

RNA isolation and hybridizations. RNA was isolated by the cesium chloride density gradient method and analyzed by Northern and/or dot blotting (15, 22). A 2-kb human FAS cDNA probe, HFAS (27), was kindly provided by Dr. C. F. Semenkovich (St. Louis, MO). The 18S rRNA probe was obtained from Clonetech laboratories (Palo Alto, CA). Sequential hybridizations with FAS, -actin, and 18S probes were conducted as we previously described (15, 22). Changes in FAS mRNA were normalized to 18S ribosomal RNA or to -actin.

Preparation of nuclei and nuclear transcription run-on assay. Preparation of nuclei from human adipocytes has not been previously reported. We performed this preparation in various buffer conditions and found that addition of very low concentrations of Nonidet P-40 (NP-40, 0.005%) to the cell lysis buffer containing 5 mM MgCl₂, 10 mM Tris, pH 7.5, 25 mM KCl, 0.1 mM EDTA, and 1 mM dithiothreitol (DTT) yielded satisfactory nuclei recovery. Adipocytes were first rinsed in PBS and resuspended in the above NP-40-supplemented lysis buffer. The homogenate was then centrifuged at 500 g at 4°C for 5 min, and the nuclei pellet was recovered by pipetting through the solid fat layer and transferred to a fresh tube. The suspension was rinsed with the same buffer without NP-40 and then centrifuged as above and resuspended in nuclei storage buffer containing 50 mM Tris, pH 7.8, 5 mM MgCl₂, 0.1 mM EDTA, 0.1 mM DTT, and 40% glycerol. Nuclei were prepared from Hep G2 cells as we previously described (22). Nuclear run on assay and hybridizations were conducted on both human adipocytes and Hep G2 cells as we previously described for 3T3-L1 adipocytes (22). The following cDNA plasmids were used: human FAS plasmid (pHFAS), kindly provided by Dr. C. F. Semenkovich, St. Louis, MO (27); pFos plasmid, kindly provided by Dr. M. S. Miller, University of Tennessee, Knoxville (18); and angiotensin II type 2 receptor (pAR) cDNA, kindly provided by Dr. T. S. Elton, University of Alabama at Birmingham (18).

mRNA stability. Hep G2 cells were cultured as described above. Subconfluent cells were maintained overnight in serum-free medium and then treated with insulin for 24 h. Actinomycin D (5 µg/ml) was then added to cells as we previously described (22). Cells were harvested at different time points after actinomycin D treatment. RNA was then isolated and analyzed by Northern blot. Sequential hybridizations with FAS and 18S probes were conducted as we previously described (15, 22). The relative abundance of FAS mRNA as a function of time was used to determine FAS mRNA half-life.

Data analysis. Autoradiograms from Northern and dot blot analyses and run on assays were quantified by densitometric scanning. Alternatively, membranes were counted using Ambis 4000 direct -imaging system (Billerica, MA). Post hoc comparisons between groups were made using Student's t-test. Data are expressed ± SE.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials & Methods Results & Discussion References

Regulation of expression of human FAS mRNA by insulin in cultured adipocytes from lean and obese subjects. We have recently described a cell culture system of human adipocytes in which glucose consumption and activities of lipogenic enzymes were increased by insulin (21). In these studies, we have also shown that insulin increases FAS activity in a dose-dependent manner, with a maximal induction at <10 nM. Furthermore, FAS activity was increased by insulin within 3 days of treatment (21). In the present work, we investigated the molecular mechanisms whereby this key lipogenic gene was regulated by insulin in cultured adipocytes from lean and obese patients.

View larger version (93K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Fig. 1.   Insulin increases fatty acid synthase (FAS) expression in human adipocytes. Total RNA was isolated from human adipocytes treated with or without insulin (20 nM) for 3 days. A: 20 µg RNA was analyzed by Northern blot; hybridization with human FAS and ribosomal (18S) probes are shown in this representative Northern blot. Lane 1, freshly isolated adipocytes; lane 2, adipocytes cultured without insulin; lane 3, adipocytes cultured with insulin. B: autoradiography obtained from Northern and/or dot blot analysis of RNA isolated from adipocytes of lean and obese patients (n = 17) was analyzed by densitometric scanning. Data are expressed ± SE after normalization to 18S ribosomal RNA.

View larger version (16K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   Fig. 2.   Continuous presence of insulin is required to maintain high levels of FAS activity. A: human adipocytes were cultured without insulin for 3 days (days 0-3) and then maintained without () or with (+/ and +) 20 nM insulin for an additional 3 days (days 3-6). At day 6, cells were either deprived (+/) or maintained in presence of 20 nM insulin (+). Cells were harvested and FAS activity was measured at the indicated times. This experiment was repeated twice. B: human adipocytes were cultured as described then maintained without () or with (+) 20 nM insulin for 2 days. Cells deprived of insulin () were then treated with the hormone (/+) for 2 additional days while cells previously treated with insulin (+) were deprived from the hormone for 2 additional days (+/). Cells were harvested and RNA was extracted from all 4 groups at same time and then analyzed by Northern blotting. Membranes were counted using Ambis 4000 -imaging system. Values from cells treated without insulin were set as 100%. This experiment represents average of 2 independent cultures that yielded comparable results.

Regulation of FAS mRNA by insulin in Hep G2 cells. To determine whether regulation of expression of FAS by insulin was tissue specific, we compared regulation of this gene in adipocytes and Hep G2 cells. Recent studies have reported that FAS gene was posttranscriptionally regulated by glucose in Hep G2 cells (26, 27). However, regulation of FAS gene in these cells by insulin has not been reported. Accordingly, we analyzed RNA isolated from Hep G2 cells treated for 48 h with or without 20 nM insulin. A representative Northern blot is shown in Fig. 3A. Results shown in this figure indicate that human FAS is expressed in Hep G2 cells and that expression of this message is induced upon insulin treatment. Data from densitometric scanning of three independent measurements are shown in Fig. 3B. These results indicate that insulin increases FAS mRNA content approximately threefold in human hepatoma cells. Taken together, the above data indicate that both hepatic and adipocyte FAS mRNA are upregulated by insulin. It is worth noting that angiotensinogen mRNA was not significantly changed by insulin in Hep G2 cells (Fig. 3A). Our findings in Hep G2 cells are supported by other studies in H4 hepatoma cells that demonstrated that glucocorticoid is the only hormone that regulates angiotensinogen expression in hepatoma cells. This is in contrast to previous reports suggesting that insulin is a key regulator of angiotensinogen expression in rodent adipose tissue (7). We have also recently confirmed these findings by demonstrating that insulin increases angiotensinogen gene expression in cultured adipocytes (17). Taken together, these data suggest that insulin regulation of angiotensinogen gene may be tissue specific.

View larger version (92K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Fig. 3.   Insulin increases FAS mRNA content in Hep G2 cells. Total RNA was isolated from hepatoma cells treated with or without insulin (20 nM) for 48 h. A: RNA were analyzed by Northern blotting after hybridization with human FAS and angiotensinogen (AGT) cDNAs as shown in this representative Northern blot. This experiment was repeated 4 times. B: scanning of autoradiograms obtained from 4 independent experiments is shown after normalization to -actin mRNA.

Mechanism of insulin regulation of the FAS gene in human adipocytes. To gain insight into mechanisms involved in regulation of FAS by insulin, we next investigated whether human FAS gene was regulated at the transcriptional level. Accordingly, we performed run-on transcription assay in nuclei isolated from control and insulin-treated human adipocytes. These assays are difficult to perform in human adipocytes and require large amounts of cells. To our knowledge, this is the first report on nuclei isolation and measurement of transcription rate in human adipocytes. A representative assay is illustrated in Fig. 4A, and scanning of autoradiograms from independent run-on assays performed in five different patients are presented in Fig. 4B.

View larger version (59K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Fig. 4.   Insulin induces the transcription rate of the FAS gene in human adipocytes. Nuclei were isolated from human adipocytes treated for 48 h with or without 20 nM insulin. RNA were labeled with [32P]UTP, purified, and hybridized with cDNA plasmids immobilized on "gene screen" nylon membranes. A: representative autoradiogram is shown. pBS, BlueScript control plasmid; AR, angiotensin II type 2 receptor plasmid; Fos, Fos oncogene plasmid; HFas, human FAS plasmid. B: densitometric scanning of run-on assays performed using human FAS plasmid (FAS, n = 5), pAR (AR, n = 2), and pFos (Fos, n = 3).

Mechanism of insulin regulation of the FAS gene in Hep G2 cells. Previous reports in rodents have demonstrated that the rodent FAS gene was primarily regulated at the transcriptional level in liver and adipose tissue as well as in hepatoma cells and preadipocyte cell lines (9, 20, 25). Interestingly, our study demonstrated that transcriptional regulation of the human FAS gene by insulin is tissue specific because insulin did not affect FAS gene transcription in Hep G2 cells. However, FAS gene transcription was increased by insulin in rat hepatoma H4-II-E cells (20), suggesting species-specific differential regulation of the FAS gene by insulin.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Insulin stabilizes FAS mRNA in Hep G2 cells. Subconfluent Hep G2 cells were maintained overnight in serum-free medium then treated with or without insulin (20 nM) for 24 h before addition of actinomycin D (5 µg/ml) for up to 24 h. Cells were harvested at indicated times, and RNA was isolated and analyzed by Northern blot using FAS and 18S probes as indicated in MATERIALS AND METHODS. This figure is representative of 3 separate experiments.

Perspectives