INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

ADIPOSE TISSUE triglycerides are the body's major source of fuel during periods of food deprivation. Therefore, increased mobilization of adipose tissue triglycerides is an important adaptive response to fasting. In lean adults, whole body lipolytic rates double during a 3-day fast (19, 40), and most of this increase occurs between 12 and 24 h of fasting (19). However, the increase in whole body lipolysis during 3 days of fasting is blunted in severely obese subjects (40). The mechanism(s) responsible for the differences in the lipolytic response to fasting in lean and obese subjects is not known but must be related to alterations in the factors that regulate lipolysis.

Insulin and catecholamines are the two major hormones that regulate lipolytic activity in humans. During fasting, a decline in plasma insulin concentration (19, 31) and an increase in adipose tissue resistance to the antilipolytic effect of insulin (14) enhance lipolysis. In addition, increased adrenal medullary catecholamine secretion (27) and increased adipose tissue lipolytic sensitivity to catecholamines (40) contribute to an increase in -adrenergic-mediated lipolysis (17). Although fasting causes a decrease in whole body sympathetic nervous system (SNS) activity in lean adults (43), the effect of fasting on adipose tissue SNS activity is unknown. Heterogeneity in the SNS response of different tissues to fasting could be advantageous by decreasing whole body SNS activity to conserve energy (43) and increasing adipose tissue SNS activity to enhance endogenous fuel mobilization.

In the present study, we evaluated whole body and abdominal subcutaneous adipose tissue lipolytic rates, and the major factors that regulate lipolyisis, in vivo during short-term fasting in lean and obese women. Both stable and radioactive isotopes were infused to evaluate lipid and norepinephrine (NE) kinetics. Subjects were studied at 14 and 22 h of fasting because maximal changes in fasting-induced lipolysis occur during this period (19). The obese group contained only women with upper-body obesity (UBO) because of the increase in basal lipolytic rates observed in this phenotype (24). We hypothesized that the increase in lipolysis that occurs during fasting is blunted in obese compared with lean women because of differences in circulating insulin and adipose tissue SNS activity.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects. Six women with UBO (>40% body wt as fat; waist-to-hip circumference ratio >0.85, and waist circumference >100 cm; 38 ± 3 yr old) and six lean women (<30% body wt as fat; 28 ± 2 yr old) participated in this study (Table 1). Although all women were premenopausal, the differences in age between lean and obese women were statistically significant (P < 0.05). The obese and lean subjects were matched for fat-free mass. Fat mass and fat-free mass were determined by dual energy X-ray absorptiometry (QDR 1000/W; Hologic, Waltham, MA). All subjects were considered to be in good health after a comprehensive medical examination, which included a history and physical examination, blood tests, and an electrocardiogram; obese subjects had normal glucose tolerance based on a 2-h oral glucose tolerance test. No subjects were taking any medications, and all were weight stable for at least 2 mo before the study, which was performed within the first 2 wk of the follicular phase of their menstrual cycle. Written informed consent was obtained before participating in this study, which was approved by the Institutional Review Board and the General Clinical Research Center of Washington University School of Medicine.

Experimental procedure. Subjects were admitted to the General Clinical Research Center at Washington University School of Medicine in the evening before the infusion study. At 1800, subjects ingested a standard meal containing 12 kcal/kg body wt for lean subjects and 12 kcal/kg adjusted body wt for obese subjects {adjusted body wt = ideal body wt + [(actual body wt  ideal body wt) × (0.25)]}. Therefore, this meal contained a total of ~700 and 800 kcal for the lean and obese women, respectively. Carbohydrate, fat, and protein represented 55, 30, and 15%, respectively, of total energy intake. At 2000, subjects ingested a defined liquid formula snack containing 250 kcal, 40 g carbohydrate, 6.1 g fat, and 8.8 g protein (Ensure, Ross Laboratories, Columbus, OH). After this snack, all subjects fasted until completion of the study the following day.

1

1

1

1

1

1

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Schematic diagram of study protocol. Isotope infusion studies were timed to ensure completion of each study at precisely 14 and 22 h of fasting. NE, norepinephrine; ATBF, adipose tissue blood flow.

Analytical procedures. Plasma insulin (8) and glucagon (9) concentrations were measured by radioimmunoassay. Plasma catecholamine concentrations were determined by a radioenzymatic method (32). Plasma fatty acid concentrations were determined by gas chromatography (25).

Calculations. Physiological and isotopic steady states were present during the last 15 min of isotope infusion at 14 and 22 h of fasting, so Steele's equation for steady-state conditions (36) was used to calculate whole body glycerol, FFA, and NE kinetics.

Statistical analysis. Student's t-test for paired samples was used to test the significance of differences between 14 and 22 h fasting data within lean and obese groups. Student's t-test for independent samples was used to test for significant differences between lean and obese subjects. One-tailed Student's t-test was used for a priori comparisons. Pearson product-moment correlation coefficient was computed to determine the relationship between specific variables. A value of P  0.05 was considered to be statistically significant. All data are expressed as means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Plasma hormone concentrations. At 14 h of fasting, plasma insulin concentration was greater (P < 0.05) in obese than in lean subjects (Table 2). By 22 h of fasting, plasma insulin concentration declined significantly (P < 0.05) in both groups, but the percent decline was less in obese than in lean women (20 ± 3 and 32 ± 5%, respectively; P < 0.05), and mean plasma insulin concentration remained more than twofold greater in the obese than in the lean group. Plasma glucagon concentrations were similar in both groups and increased slightly with continued fasting. At 14 h of fasting, mean plasma epinephrine concentration in obese women was about one-half as great as that observed in lean women (P < 0.05), whereas plasma NE concentrations were similar in both groups (Table 2). Continued fasting did not cause a significant change in plasma epinephrine or NE concentrations in either the lean or obese subjects.

Plasma substrate concentrations. Abdominal vein plasma glycerol and FFA concentrations were always greater than arterial concentrations in lean and obese subjects, indicating net glycerol and FFA release from subcutaneous abdominal adipose tissue at both 14 and 22 h of fasting (Table 3). Arterial plasma glycerol and FFA concentrations increased significantly (P < 0.05) with continued fasting in both lean and obese subjects.

ATBF. ATBF was >50% lower in obese compared with lean subjects (1.84 ± 0.18 and 4.29 ± 0.48 ml · 100 g adipose tissue¹ · min¹, respectively; P < 0.05) and did not change between 14 and 22 h of fasting in either lean or obese subjects.

Substrate kinetics. At 14 h of fasting, whole body glycerol rate of appearance (R_a), an index of whole body lipolysis, was greater in obese than lean women (201.4 ± 17.1 and 123.5 ± 12.3 µmol/min, respectively; P < 0.05; Fig. 2). Whole body FFA R_a, an index of fatty acid availability in plasma, was also greater in obese than lean women (526.8 ± 60.6 and 362.2 ± 41.2 µmol/min, respectively; P < 0.05; Fig. 2). However, glycerol and FFA kinetics normalized for body fat mass were ~60% lower in obese than in lean subjects (3.97 ± 0.33 vs. 9.29 ± 1.19 µmol glycerol · kg fat mass¹ · min¹ and 10.4 ± 1.2 vs. 26.9 ± 4.5 µmol FFA · kg fat mass¹ · min¹; P < 0.05). Continued fasting increased glycerol R_a and FFA R_a in both lean and obese groups (Fig. 2). However, the absolute increase in glycerol R_a and FFA R_a in obese subjects (19.8 ± 7.0 and 87.1 ± 30.3 µmol/min, respectively) was less than the increase observed in lean subjects (44.2 ± 6.6 and 137.4 ± 30.4 µmol/min, respectively; P < 0.05 for both glycerol and nonesterified fatty acid R_a). The percent increase in glycerol R_a correlated closely with the percent decline in plasma insulin concentration in both lean and obese subjects (r² = 0.85; P < 0.05; Fig. 3).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Whole body glycerol rate of appearance (Ra; A) and free fatty acid (FFA) Ra (B) at 14 h (open bars) and 22 h (hatched bars) of fasting in lean and obese women. *Significantly different from lean subjects, P < 0.05. Significantly different from 14-h values, P < 0.05. Values are means ±SE.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Relationship between the relative decrease in plasma insulin concentration and the relative increase in lipolysis (whole body glycerol Ra) during brief fasting (between 14 and 22 h of fasting) in lean () and obese () women.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Net regional subcutaneous abdominal adipose tissue glycerol (A) and FFA (B) release at 14 h (open bars) and 22 h (hatched bars) of fasting in lean and obese women. *Significantly different from lean subjects, P < 0.05. Significantly different from 14-h value, P < 0.05. Values are means ±SE.

NE kinetics. Whole body NE spillover in the systemic circulation, a marker of SNS activity, was similar in both lean and obese groups at 14 h of fasting (Fig. 5). Fasting caused a significant decrease in NE spillover in lean but not obese subjects. Regional abdominal subcutaneous adipose tissue NE spillover, an index of regional adipose tissue SNS activity, was more than threefold greater in lean than obese subjects (0.91 ± 0.08 and 0.26 ± 0.06 nmol · 100 g adipose tissue¹ · min¹, respectively; P < 0.05). Regional adipose tissue NE spillover did not change with continued fasting in either lean or obese subjects (Fig. 5).

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Whole body (A) and regional subcutaneous abdominal adipose tissue (B) NE spillover at 14 h (open bars) and 22 h (hatched bars) of fasting in lean and obese women. *Significantly different from lean subjects, P < 0.05. Significantly different from 14-h value, P < 0.05. Values are means ±SE.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The mobilization of endogenous triglycerides stored in adipose tissue is important for survival during starvation. In lean men, most of the increase in whole body lipolytic rates that occur during starvation takes place within the first 24 h of fasting (19). The results of the present study demonstrate, for the first time, that the increase in lipolysis during early starvation (between 14 and 22 h of fasting) is blunted in women with UBO. However, basal (14-h fast) whole body lipolytic rates were greater in UBO than lean women and were similar to values observed in lean women after 22 h of fasting. Therefore, the attenuated increase in lipolysis during fasting in UBO women did not compromise fatty acid availability as a fuel in these subjects. In fact, the blunted lipolytic response in UBO women may be beneficial by preventing excessive and potentially harmful increases in plasma FFA concentrations (35, 38).

Our data suggest that the mechanism responsible for differences in the lipolytic response to fasting in lean and UBO women may be related to differences in their insulin response. Insulin is a potent inhibitor of lipolysis in both lean and obese persons (3). Therefore, an alteration in plasma insulin concentration represents an important adaptive response to fasting. The decline in plasma insulin observed during fasting is not simply a response to decreased blood glucose but occurs even when euglycemia is maintained by exogenous low-dose glucose infusion (16). In the present study, the relative decrease in plasma insulin concentration between 14 and 22 h of fasting was smaller in obese than in lean subjects. In both groups, the relative decrease in insulin correlated closely with the relative increase in glycerol R_a, suggesting a causal relationship between changes in circulating insulin and lipolysis during fasting.

Increased stimulation of adipose tissue -adrenergic receptors is another important mechanism for increasing lipolysis during fasting (18). The increase in -adrenergic-mediated lipolysis during fasting is caused, in part, by an increase in adrenal medullary secretion of epinephrine (27). In the present study, plasma epinephrine concentration tended to increase between 14 and 22 h but was still below the threshold level known to stimulate lipolysis (7). However, fasting increases adipose tissue lipolytic sensitivity to epinephrine (14, 40), which likely decreases the threshold for epinephrine-stimulated lipolysis. We have previously found the increase in lipolysis during epinephrine infusion after short-term fasting is less in obese than in lean subjects (40). Therefore, altered adipose tissue sensitivity to epinephrine in our obese subjects may have also contributed to their blunted lipolytic response to fasting.

This is the first study to evaluate the effects of fasting on adipose tissue and whole body SNS activity in lean and obese subjects. We used tracer methodology to measure NE spillover in the systemic circulation to provide an index of whole body SNS activity (23). Signal transmission within the SNS involves NE release from sympathetic postganglionic neurons. Most of the released NE is cleared locally by neuronal reuptake and effector cell metabolism while a portion spills over in the bloodstream. Therefore, NE spillover in the systemic circulation represents NE released from sympathetic neurons distributed throughout the body and NE secreted from the adrenal medullas. We combined arteriovenous balance and tracer methodology to determine regional adipose tissue NE spillover to provide an index of adipose tissue SNS activity. This approach eliminates any contribution of NE from the adrenal medullas.

Our results demonstrate that the SNS response to early fasting differs between lean and obese women. In lean women, whole body NE spillover declined, but adipose tissue NE spillover did not change between 14 and 22 h of fasting. This heterogeneity may be advantageous by decreasing SNS activity in lean tissue and thereby decreasing energy expenditure, while maintaining SNS activity in adipose tissue and thereby stimulating the mobilization of endogenous triglycerides. These results are consistent with previous studies that found whole body NE spillover decreased in lean subjects after 10 days of hypocaloric feeding (26) and cardiac muscle NE turnover decreased in rats fasted for 48 h (42). In contrast to the lean women, whole body NE spillover rates did not decrease in obese women during short-term fasting. These results extend those of Bazelmans et al. (1), who found that 10 days of hypocaloric feeding did not affect whole body NE spillover in obese subjects. Therefore, the normal decline in whole body SNS activity that occurs in response to energy deprivation in lean subjects is blunted in obese persons. However, adipose tissue SNS activity is maintained during early starvation in both lean and obese subjects.

Postabsorptive (14-h fast) regional adipose tissue lipolytic rates, expressed per 100 g of adipose tissue, was 50% lower in our obese than in our lean subjects. However, the rate of whole body lipolysis was >60% greater in the obese group because of their large amount of fat mass; total body fat was more than threefold greater in our obese than in our lean subjects. Excessive release of FFA in plasma in persons with UBO may be responsible for several metabolic diseases associated with obesity by impairing the ability of insulin to stimulate muscle glucose uptake (6) and suppress hepatic glucose production (6, 15), by increasing pancreatic insulin secretion (2) and by inhibiting hepatic insulin clearance (28). Furthermore, increased delivery of FFA to the liver can increase hepatic very low density lipoprotein production and plasma triglyceride and cholesterol concentrations (21). The downregulation of lipolysis per unit of fat mass in obese persons helps prevent the generation of even greater whole body lipolytic rates.

The marked differences that we observed in regional lipolytic rates between lean and obese women is inconsistent with the results of a previous study by Jansson et al. (11), who found the rate of glycerol release from abdominal subcutaneous adipose tissue was the same in both lean and obese subjects. It is unlikely that subcutaneous adipose tissue lipolytic rates were the same in our lean and obese subjects for several reasons. First, the greater amount of body fat in our obese subjects would have caused more than a threefold difference in whole body lipolytic rates between the two groups. Instead, whole body lipolytic rates were only 60% greater in our obese than in our lean group. Second, we also found that regional glycerol production measured by tracer balance methodology (39) gave the same results as the arteriovenous concentration balance approach (data not shown). Third, the differences we found in regional glycerol kinetics between lean and obese subjects were concordant with the differences we found in FFA kinetics. The reason for the discrepancy between the Jansson et al. (11) study and ours may be related to differences in gender or methodology. Jansson et al. studied lean and UBO men and measured regional lipolytic activity by using microdialysis probes to sample interstitial adipose tissue glycerol. This latter approach requires an extrapolation of interstitial glycerol concentration to venous values, which may generate errors in the estimation of regional glycerol kinetics. In fact, direct comparisons of the two techniques in the same subjects have found that regional glycerol output measured by arteriovenous balance using abdominal vein samples was double the values obtained by arteriovenous balance using microdialysis samples (34, 37).

Although we found regional adipose tissue lipolysis in our obese subjects was one-half the value observed in our lean subjects, in vivo lipolytic rates per fat cell were probably similar in both groups. Fat cells from obese persons are larger and contain more lipid than fat cells from lean persons (11, 29). Subcutaneous abdominal adipocytes obtained from UBO women with similar body mass index and waist-to-hip ratio as our subjects have two times the cell volume as adipocytes obtained from lean women (29). Therefore, our regional measurement of glycerol and FFA kinetics, which was expressed per 100 g of adipose tissue, represented glycerol and FFA release from approximately one-half as many fat cells in the obese than in the lean group. Thus the rates of glycerol and FFA release per fat cell were similar in both groups. In contrast, studies performed in vitro have demonstrated that basal lipolytic rates in isolated adipocytes obtained from UBO women were threefold greater than lipolytic rates in cells obtained from lean women (29). However, this discrepancy is probably related to differences between in vivo and in vitro studies. Other investigators have also found that lipolysis measured in isolated adipocytes do not agree with in vivo data from the same subjects (22). The differences in lipolytic activity observed between in vitro and in vivo studies may be related to alterations in local adipocyte environment caused by removal of tissue and plasma factors, such as insulin, epinephrine, and adipose tissue SNS activity.

ATBF, expressed per 100 g of adipose tissue, in our obese subjects was less than one-half the rate measured in the lean group. The calculations we used in estimating blood flow assumed that the partition coefficient for ¹³³Xe is similar in both lean and obese women (13). Although the lower rate of blood flow in subcutaneous adipose tissue in obese compared with lean subjects has been observed previously (11), the mechanism for this phenomenon is not clear but may be a simple function of the anatomical relationship between capillaries and adipocytes. Each fat cell is located in proximity to a capillary (5). Blood flow per fat cell remains constant, independent of fat cell size, so that blood flow per 100 g of adipose tissue decreases with increasing cell volume (12). It is unlikely that decreased local SNS was responsible for the lower rate of adipose tissue blood flow even though subcutaneous fat houses a rich network of blood vessels that contain vascular -adrenergic receptors (10). Simonsen et al. (33) found that -adrenergic blockade prevented the normal meal-induced increase in ATBF but did not affect basal blood flow values. Therefore, adipose tissue SNS innervation may be responsible for the increase in ATBF after carbohydrate ingestion but does not appear to be an important regulator of basal, postabsorptive ATBF.

In summary, basal lipolytic rates per unit of fat mass are lower in UBO women than in lean women, but whole body lipolytic rates are greater in obese women because of their increased adiposity. During short-term fasting, whole body lipolytic rates increase in both lean and obese women, but the increase is blunted in the obese group. This downregulation of adipose tissue lipolysis in UBO women during fasting may be advantageous by preventing excessive and potentially harmful increases in plasma FFA. Our data suggest that the attenuated lipolytic response to fasting in obese women is related to a blunted decline in circulating insulin. Altered adipose tissue SNS activity during fasting does not contribute to the increase in lipolysis in either lean or obese women. Whole body SNS activity decreased in lean but did not change in obese women, whereas adipose tissue SNS activity remained the same during fasting in both groups.