S Goodrick

We measured arterio-venous differences in concentrations of tumor necrosis factor-alpha (TNF alpha) and interleukin-6 (IL-6) across a sc adipose tissue bed in the postabsorptive state in 39 subjects [22 women and 17 men; median age, 36 yr (interquartile range, 26-48 yr); body mass index, 31.8 kg/m2 (range, 22.3- 38.7 kg/m2); percent body fat, 28.7% (range, 17.6-50.7%)]. A subgroup of 8 subjects had arteriovenous differences measured across forearm muscle. Thirty subjects were studied from late morning to early evening; 19 ate a high carbohydrate meal around 1300 h, and 11 continued to fast. We found a greater than 2-fold increase in IL-6 concentrations across the adipose tissue bed [arterial, 2.27 pg/mL (range, 1.42-3.53 pg/mL); venous, 6.71 pg/mL (range, 3.36-9.62 pg/mL); P < 0.001], but not across forearm muscle. Arterial plasma concentrations of IL-6 correlated significantly with body mass index (Spearman's r = 0.48; P < 0.01) and percent body fat (Spearman's r = 0.49; P < 0.01). Subcutaneous adipose tissue IL-6 production increased by the early evening (1800-1900 h) in both subjects who had extended their fasting and those who had eaten. Neither deep forearm nor sc adipose tissue consistently released TNF alpha [across adipose tissue: arterial, 1.83 pg/mL (range, 1.36-2.34 pg/mL); venous, 1.85 pg/mL (range, 1.44-2.53 pg/mL); P = NS: across forearm muscle: arterial, 1.22 pg/mL (range, 0.74-2.76 pg/mL); venous, 0.99 pg/mL (range, 0.69-1.70 pg/mL); P = NS]. Although both IL-6 and TNF alpha are expressed by adipose tissue, our results show that there are important differences in their systemic release. TNF alpha is not released by this sc depot. In contrast, IL-6 is released from the depot and is thereby able to signal systemically.

In non-diabetic subjects, insulin concentrations and insulin resistance are clearly connected, and both correlate with leptin levels, making interpretations about mechanisms difficult. In non-insulin-dependent (Type 2) diabetes mellitus (NIDDM), however, insulin concentrations and insulin resistance are less closely associated. Therefore, we examined the relationship of plasma leptin concentrations within insulin resistance and insulin levels in 32 subjects with NIDDM, who underwent measurement of insulin resistance with an insulin sensitivity test. Plasma leptin was measured with an in-house monoclonal immunoradiometric assay. Fasting leptin level correlated with BMI (r = 0.78; p < 0.001), metabolic clearance rate of glucose (= -0.44; p = 0.015), and fasting specific insulin (r = 0.58; p = 0.001), but not with age, cholesterol, triglycerides or blood pressure (r = -0.26 to 0.21; p = NS). In linear regression analysis, after adjustment for BMI and gender, leptin concentrations correlated with those of insulin (partial r = 0.42; p = 0.025), but not insulin resistance (partial r = -0.10; p = NS). We conclude that in NIDDM, concentrations of plasma leptin are closely related to those of insulin per se and to obesity, but not to insulin resistance. Insulin may be an important regulator of leptin concentration in NIDDM.

We evaluated abdominal adipose tissue leptin production during short-term fasting in nine lean [body mass index (BMI) 21 +/- 1 kg/m(2)] and nine upper body obese (BMI 36 +/- 1 kg/m(2)) women. Leptin kinetics were determined by arteriovenous balance across abdominal subcutaneous adipose tissue at 14 and 22 h of fasting. At 14 h of fasting, net leptin release from abdominal adipose tissue in obese subjects (10.9 +/- 1.9 ng x 100 g tissue x (-1) x min(-1)) was not significantly greater than the values observed in the lean group (7.6 +/- 2.1 ng x 100 g(-1) x min(-1)). Estimated whole body leptin production was approximately fivefold greater in obese (6.97 +/- 1.18 microg/min) than lean subjects (1.25 +/- 0.28 microg/min) (P < 0.005). At 22 h of fasting, leptin production rates decreased in both lean and obese groups (to 3.10 +/- 1.31 and 10.5 +/- 2.3 ng x 100 g adipose tissue(-1) x min(-1), respectively). However, the relative declines in both arterial leptin concentration and local leptin production in obese women (arterial concentration 13.8 +/- 4.4%, local production 10.0 +/- 12.3%) were less (P < 0.05 for both) than the relative decline in lean women (arterial concentration 39.0 +/- 5.5%, local production 56.9 +/- 13.0%). This study demonstrates that decreased leptin production accounts for the decline in plasma leptin concentration observed after fasting. However, compared with lean women, the fasting-induced decline in leptin production is blunted in women with upper body obesity. Differences in leptin production during fasting may be responsible for differences in the neuroendocrine response to fasting previously observed in lean and obese women.