DIABETES, VOL. 47, AUGUST 1998 1365 Rapid Publication Human Preadipocytes Display a Depot-Specific Susceptibility to Apoptosis Carola U. Niesler, Kenneth Siddle, and Johannes B. Prins Adipose tissue mass is determined by both the number and volume of adipose cells. Adipose cell number reflects the balance of cell acquisition and cell loss, whereas adipose cell volume represents the balance of lipolysis and lipogenesis. It is well recognized that insulin resistance, NIDDM, and other metabolic disorders are associated more strongly with increased omental adiposity than with subcutaneous adiposity. Depot- related differences exist in adipocyte responses to lipolytic and lipogenic stimuli, in adipocyte apoptosis, and in the potential for preadipocyte replication and differentiation. In the present study, we address the question of whether there might also be a site-specific difference in the susceptibility of human preadipocytes to apoptosis. Paired samples of human omental and sub- cutaneous preadipocytes from 12 individuals were cul- tured, and apoptosis was induced by serum deprivation or treatment with tumor necrosis factor ( TNF)- for 4 h. Cells were then stained with acridine orange, and apop- totic indices were calculated as the fraction of cells showing nuclear condensation. Under both conditions, in 9 of 11 subjects, apoptotic indices were substantially greater in preadipocytes from the omental depot than in those from the subcutaneous depot, and mean apoptotic indices were more than twofold higher in omental cells ( serum-free medium: P < 0.05; TNF- : P < 0.02; paired t test). Omental preadipocytes are therefore more sus- ceptible to two different apoptotic stimuli than subcu- taneous preadipocytes, demonstrating another intrinsic site-specific difference between human adipose cells of the two depots. These results suggest that the regulation of adipose tissue distribution in humans could involve depot-specific differences in rates of preadipocyte apo- ptosis. Diabetes 47:1365–1368, 1998 O besity is characterized by an increased fat mass and occurs when the intake of food exceeds the energy requirement of the body for a sustained period (1). The degree of adiposity is determined by both the volume (lipid content) and the number of adipose cells (2). Adipocyte volume depends on the balance of lipol- ysis and lipogenesis, whereas adipose cell number is deter- mined by the rate of cell acquisition versus cell loss (2). Adi- pose cell acquisition occurs by proliferation and differentia- tion of preadipocytes and cell loss occurs via apoptosis of preadipocytes and adipocytes, and possibly by other processes such as adipocyte dedifferentiation (2). It is now well recognized that increased omental adiposity is more deleterious to health than subcutaneous adiposity. Numerous observations indicate that the metabolic syn- dromes closely associated with obesity, such as NIDDM, hyperlipidemia, cardiovascular disease, and syndrome X, are more tightly correlated with markers of central adiposity than with the degree of adiposity per se (3–5). Omental and subcutaneous adipose cells are known to display differences in various basal metabo lic pro perties, pointing to an intrinsic difference between cells from the two sites. In vivo, omental fat cells have a higher rate of lipid turnover than subcutaneous cells (4,6,7). In vitro, adipocytes isolated from the omental depot have higher basal levels of cAMP, a greater number of glucocorticoid receptors, greater sensitivity to the lipolytic effect of catecholamines, and decreased sensitivity to the antilipolytic effect of insulin (4,8–12). A further mechanism whereby the level of adiposity within each depot could be controlled is by depot-specific differences between cell acquisition and cell loss. There is evidence sug- gesting that preadipocytes from the omental depot display greater replication rates than their subcutaneous counterparts (13) and that subcutaneous preadipocytes differentiate more readily in response to thiazolidinediones than their omental counterparts (14). We have previously reported that human preadipocytes and adipocytes undergo apoptosis in vitro in response to tumor necrosis factor (TNF) and serum depriva- tion (15,16). We have also shown that human adipocyte apo- ptosis occurs in vivo and that rates of adipocyte apoptosis appear higher in omental cells than in subcutaneous cells in patients with malignancy (17). The in vitro effects of TNF- a n d the observation that adipocyte apoptosis rates are higher in malignancy (when TNF- levels are often elevated) support the likeliho od that TNF- –induced weight loss in vivo involves adipose cell apoptosis. This has not been investigated, but it has recently been reported that administration of leptin and thia- zolidinedione induces adipocyte apoptosis in rodents (18,19). In this study, we investigate whether human preadipocytes display a depot-spec ific sensitivity to the inductio n o f apo pto - sis by serum deprivation or the addition of TNF- . RESEARCH DESIGN AND METHODS Patients and sample acquisition. Patients undergo ing elective surgery at the Addenbro o ke’s Hospital (Cambridge, U.K.) were recruited for the study. The experimental protocols were approved by the institution’s ethics committee, and all patients gave their informed consent to the procedure. Patients’ characteris- tics are outlined in Table 1. No patients were on medications known to affect adi- pose tissue mass, distribution, or metabolism, and none had severe systemic ill- ness or known malignancy. All patients fasted for at least 8 h before surgery, and From the Departments o f Clinical Biochemistry (C.U.N., K.S., J.B.P.) and Medicine (J.B.P.), University of Cambridge, Addenbrooke’s Hospital, Cam- bridge, U.K. Address correspondence and reprint requests to Dr. John Prins, Depart- ment of Medicine, University of Cambridge, Addenbrooke’s Hospital, Hills Rd., Cambridge CB2 2QQ, U.K. E-mail: jprins@ hgmp.mrc.ac.uk. Received for publication 16 January 1998 and accepted in revised form 6 May 1998. PBS, phosphate-buffered saline; SFM, serum-free medium; TNF, tumor necrosis factor.