| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Vascular Medicine, University Medical Center, 3508 GA Utrecht, The Netherlands
Address all correspondence and requests for reprints to: M. Castro Cabezas, M.D., Ph.D., Department of Vascular Medicine F02.126, University Medical Center Utrecht, P.O Box 85500, 3508 GA Utrecht, The Netherlands. E-mail: m.castrocabezas{at}azu.nl
Abstract
The regulation of hormone-sensitive lipase activity in vivo has not been studied in detail before. We have performed noninvasive in vivo tests to measure hormone-sensitive lipase activity under high plasma levels of endogenous insulin and catecholamines. For this purpose, two mental stress tests were carried out at random in 13 healthy volunteers. The subjects ingested 200 ml of a placebo solution or 20% glucose, followed by 1 h of rest, 20 min of mental stress, and 40 min of rest. Twenty minutes after the ingestion of glucose, insulin levels increased from 6.8 ± 1.6 to a maximum of 30.5 ± 4.8 mU/liter (P < 0.01), whereas the increase in insulin was significantly less after placebo (from 5.7 ± 0.9 to 9.5 ± 1.5 mU/liter; P < 0.01). The increase in heart rate, as an estimate of the amount of stress, was similar in both tests (12% increase). During stress, plasma norepinephrine and epinephrine concentrations increased by 24% and 44%, respectively, after glucose and by 4% and 21%, respectively, after placebo (n = 6). Fasting plasma FFA were similar in both tests (placebo, 0.35 ± 0.07 mM; glucose, 0.46 ± 0.08 mM). Forty minutes after ingestion of placebo, plasma FFA concentrations decreased to 0.27 ± 0.07 mM, compared with a stronger suppression to 0.11 ± 0.02 mM after ingestion of glucose (P < 0.01). By 10 min after mental stress, plasma FFA concentrations increased by 53% after placebo (P < 0.01), in contrast to unchanged FFA concentrations after ingestion of glucose. Taken together, these results suggest that the suppression of hormone-sensitive lipase by endogenous insulin in healthy, insulin-sensitive subjects is stronger than the stimulation by endogenous catecholamines.
PLASMA FFA concentrations vary widely over the day and are highly regulated by hormonal, metabolic, and neuronal signals (1). Lipolysis in adipose tissue is mainly regulated by hormone-sensitive lipase (HSL) and lipoprotein lipase. In the fasting state, HSL is the rate-limiting enzyme determining adipose tissue triglyceride (TG) lipolysis (2). HSL is under tight hormonal and neuronal control and was first described by Hahn et al. (3). Variations in HSL expression probably modulate the extent of adipose tissue lipolysis. HSL-stimulating hormones, like catecholamines, activate this enzyme through cAMP-dependent phosphorylation (4). Insulin, in contrast, inhibits phosphorylation by hydrolysis of cAMP (5). It has been shown that HSL activity is related to the localization of adipose tissue in the human body, being lower in sc adipocytes compared with visceral adipocytes, which is probably related to the different fat cell sizes (6, 7). Also in various diseases, such as the insulin resistance syndrome (8), obesity (9), diabetes mellitus (10), and familial combined hyperlipidemia (FCHL) (11, 12), diminished activity of HSL has been described. However, the latter was not confirmed in a Finnish population of FCHL subjects (13).
As HSL is an intracellular enzyme, most data concerning HSL have been generated by in vitro studies. Frayn et al. (14, 15) calculated the action of HSL in vivo by microdialysis techniques, measuring glycerol, TG, and FFA concentrations after correction for bloodstream velocity. These elegant in vivo studies have enhanced our understanding of the regulation of HSL.
As there are scarce data on the acute regulation of HSL activity in vivo, and most data are derived from insulin clamp techniques, by which stimulation of HSL is inhibited, we investigated the effects of HSL metabolism in vivo in a more physiological, noninvasive situation. The potency of endogenous catecholamines, induced by mental stress, to activate HSL was investigated under low and high endogenous insulin levels. The aim of the present study was to evaluate in vivo whether catecholamine induction of HSL could override insulin-mediated HSL inhibition in healthy, insulin-sensitive subjects.
Subject and Methods
Subjects
The study protocol was approved by the human investigations review committee of the University Hospital (Utrecht, The Netherlands). Healthy normolipidemic volunteers were recruited by advertisement. All participants were healthy volunteers, with a negative family history for coronary artery disease and type 2 diabetes. All subjects had a body mass index less then 30 kg/m2. None of the volunteers used drugs known to affect lipid metabolism. On the morning of inclusion, blood pressure and waist to hip ratio were measured, and venous blood was drawn once for baseline clinical chemical evaluation.
Mental stress tests
Two mental stress tests were performed with at least a 1-wk interval. All participants visited our laboratory after a 12-h overnight fast where an iv cannula was placed in the brachial vein. The cannula was kept open by a continuous 0.9% saline infusion. All peripheral blood samples, obtained from the cannula in sodium EDTA (2 mg/ml), were placed on ice and centrifuged immediately for 15 min at 800 x g at 4 C. The participants ingested a solution of either 200 ml placebo containing 8% (wt/vol) saccharide or 200 ml 20% (wt/vol) glucose. The two solutions were identical in color, taste, and viscosity. Subjects were blinded to the type of solution, and tests were given randomly. The volunteers remained supine during the first 60 min of the test in a room without disturbing stimuli. Over the next 20 min the subjects were subjected to two types of mental stress tests consisting of letters and figures, as described by others with modifications (16). They were asked to subtract two letters of the alphabet from the given letters; for example, CD would give AB. When numbers where given, subjects had to reverse the order; for example, 1234567 would give 7654321. After the mental stress period, the subjects remained supine for 40 min. Peripheral blood samples were obtained before the mental stress test (-60 to 0 min) at 10-min intervals. During the 20 min of stress (0–20 min) blood was collected at 5-min intervals, and after the stress (30–60 min) blood was collected at 10-min intervals. During the test heart rate was also recorded. In a subgroup of six subjects, catecholamines were measured at -60, -40, 0, 10, and 20 min.
Analytical methods
Plasma samples were stored at -20 C immediately after centrifugation. TG and cholesterol were measured in duplicate by commercial colorimetric assay (GPO-PAP, and Monotest Cholesterol kit, respectively, Roche Molecular Biochemicals, Indianapolis, IN). FFA concentrations were measured in plasma samples by an enzymatic colorimetric method (Wako Chemicals GmbH, Neuss, Germany). The quantitative assays of apolipoprotein B have been described in detail previously (17). Insulin was measured by commercial ELISA (Mercodia, Uppsala, Sweden). Epinephrine and norepinephrine concentrations were measured by standard procedures by the Department of Clinical Chemistry of our hospital (University Medical Center, Utrecht, The Netherlands) using HPLC. The lowest detection limit is 0.1 nmol/liter. For estimation of insulin sensitivity the glucose/insulin ratio and homeostasis model assessment (glucose x insulin/22.5) were calculated (18).
Statistics
All values are expressed as the mean ± SEM. Correlations among variables were tested by linear regression analysis (Pearson’s correlation coefficient). For statistical analysis of changes in FFA concentrations and heart rate during the 20 min of stress, repeated measures ANOVA was used. The percent increase in FFA and heart rate in this period was calculated by correcting for basal values at rest (0 min).
The percent suppression of plasma FFA concentrations during glucose and placebo was calculated as the percent decrease from -60 to 0 min. Heart rate increase during stress was calculated as the percent increase from 0 to 20 min.
The differences between glucose and placebo tests were calculated by unpaired t test. Intraindividual differences were calculated by paired t test. Statistical significance was reached at P < 0.05 (two-tailed). Calculations were performed using SPSS/PC+ 9.0 (SPSS, Inc., Chicago, IL).
Results
General characteristics
Thirteen subjects were enrolled in this study, eight men and five
women. Anthropometric characteristics and fasting laboratory values are
shown in Table 1
.
|
)At the beginning of the test (-60 min) heart rate was similar in the placebo (63 ± 2 beats/min) and glucose (62 ± 2 beats/min; P = NS) groups. Heart rate tended to increase after both placebo (63 ± 2 to 66 ± 2 beats/min; P = 0.07) and glucose (62 ± 2 to 65 ± 2 beats/min; P = 0.12) from -60 to 0 min. There were no differences in heart rate at 0 min. The maximal heart rate was reached after 10 min of mental stress in both tests and was not different between placebo (78 ± 3 beats/min) and glucose (76 ± 3 beats/min) tests. The average heart rate during the stress period (5–20 min) increased after placebo (from 66 ± 2 at 0 min to 73 ± 3 beats/min; P < 0.01) as well as after glucose (from 65 ± 2 to 72 ± 3 beats/min; P < 0.01). However, the relative increase was not different between the tests (11 ± 2% vs. 11 ± 4%; P = NS).
|
Insulin concentrations (Fig. 2)
Fasting plasma insulin concentrations were not different between placebo and glucose (5.7 ± 0.9 and 6.8 ± 1.6 IU/liter, respectively) tests. Maximal insulin concentrations were reached after 20 min in both tests, although the increase was significantly lower after placebo (9.5 ± 1.5 IU/liter at -40 min) than after glucose (30.5 ± 4.8 IU/liter; -40 min). As the antilipolytic effect of insulin is decreased in the insulin resistance syndrome (19), and this may have an effect on the FFA concentrations during rest and stress, the subjects were subdivided into three groups according to their fasting plasma insulin concentrations (average, 3.7 ± 0.9 vs. 5.3 ± 1.3 vs. 7.8 ± 0.7 IU/liter) to evaluate the effects of insulin sensitivity on FFA metabolism. The maximal insulin concentrations after glucose during the first 60 min of rest were not significantly different between the tertiles (30.9 ± 6.7 vs. 45.4 ± 3.2 vs. 34.7 ± 6.7 IU/liter). The homeostasis model assessment index showed a significant gradual increase (0.39 ± 0.04 vs. 1.23 ± 0.08 vs. 1.76 ± 0.01; P < 0.01). Plasma insulin concentrations during stress after placebo were similar to basal values; after glucose, insulin concentrations were significantly higher at 0 min (14.3 ± 1.7 IU/liter) and 10 min (10.8 ± 1.5 IU/liter) compared with -60 min. Insulin returned to basal values after placebo (5.7 ± 0.9 IU/liter) and after glucose (6.3 ± 1.3 IU/liter) at the end of the test.
|
, A and
B)
Plasma FFA concentrations were similar at the beginning of the
placebo (0.35 ± 0.07 mmol/liter) and glucose (0.46 ± 0.08
mmol/liter) tests. After glucose, FFA concentrations tended to increase
from -60 to -50 min (to 0.47 ± 0.09 mmol/liter), in contrast to
those after placebo (to 0.30 ± 0.05 mmol/liter). Plasma FFA
concentrations tended to decrease (P = 0.07) during the
first 60 min of rest after placebo (to 0.27 ± 0.07 mmol/liter at
0 min compared with -60 min), but decreased significantly
(P < 0.01) after glucose (to 0.11 ± 0.02
mmol/liter at 0 min compared with -60 min) (Fig. 3A). The relative
decrease was lower after placebo (32 ± 8%) than after glucose
(70 ± 3%; P < 0.01). Glucose ingestion resulted
in a gradual suppression of FFA concentrations (61% vs.
73% vs. 88%; P < 0.01, lowest
vs. highest tertile), which was linked to fasting insulin
concentrations (Fig. 3B).
|
TG concentrations (Fig. 4)
Fasting plasma TG concentrations were similar in placebo (1.25 ± 0.17 mmol/liter) and glucose (1.28 ± 0.17 mmol/liter) tests. During mental stress, plasma TG concentrations did not change significantly.
|
Lipolysis in fat cells is stimulated via ß-adrenoceptors, initiating a chain of events finally activating HSL (20, 21). The ß-adrenergic stimulation of lipolysis involves both epinephrine (which is delivered through the systemic circulation) as well as norepinephrine (which is released locally in adipose tissue). The present study suggests that insulin-mediated inhibition of HSL is stronger than activation by endogenous catecholamines in vivo in healthy insulin-sensitive subjects. To stimulate endogenous catecholamine production, mental stress tests were used, as described previously by others (16). The first period of rest after ingestion of glucose reflected insulin-mediated HSL inhibition. The FFA changes seen during the stress period represent stimulation of HSL activity by endogenous catecholamines.
Regulation of HSL activity is of great clinical importance. In several diseases, such as type 2 diabetes mellitus, FCHL and obesity a cluster of metabolic disturbances has been described (e.g. impaired FFA metabolism, insulin resistance, and diminished HSL activity) (8, 9, 10, 11, 12). Impaired FFA metabolism itself may be one of the pathophysiological factors underlying the insulin resistance syndrome (22). Klannemark et al. (10) have linked insulin resistance in families with type 2 diabetes to a polymorphism of the HSL gene. They found a significant difference in allele distribution of the HSL gene between subjects with insulin resistance and controls. They concluded that this HSL polymorphism may increase the susceptibility to abdominal obesity and possibly also to type 2 diabetes (10). This could mean that abdominal obesity and impaired FFA metabolism are secondary to impaired HSL regulation. The test, as described in this paper, can be used to investigate whether this polymorphism is linked to impaired HSL activity in vivo.
We have shown that the decrease in FFA concentrations during the first 60 min of the test was higher after glucose than after placebo and that the subgroup with the highest fasting insulin tertile had the highest FFA suppression, suggesting that small changes in insulin concentrations may have large effects on the inhibition of HSL in insulin-sensitive subjects. The increased suppression of FFA in this subgroup could not be explained by higher insulin concentrations after glucose, as the maximal insulin concentration during the first 60 min of rest was not different between the subgroups. This is in line with the data reported by Horowitz et al. (23), who showed in lean and obese women a significant correlation between the decrease in fasting insulin concentrations and the increase in glycerol appearance, suggesting a causal relationship between changes in circulating insulin and lipolysis during fasting. Impaired FFA suppression has been described in the insulin resistance syndrome, suggesting that diminished FFA suppression may be one of the characteristics of this syndrome (24). Furthermore, in subjects with increased sc adipose tissue and increased insulin concentrations, ß-adrenoceptor-mediated lipolysis of abdominal adipocytes is increased in vitro (25). Our data suggest that when insulin sensitivity is normal in lean healthy subjects with relatively low abdominal fat tissue, it is the insulin-mediated inhibition of HSL that predominates above catecholamine-induced HSL activation.
It has been proposed that the antilipolytic effect of insulin is caused by direct inhibition of HSL activity. Frayn and co-workers (26) suggested that insulin does not affect HSL activity itself, but that insulin may be important in the reesterification of FFA in adipose tissue, affecting FFA partitioning. In this study, however, we could not differentiate between the two mechanisms, because FFA concentrations were only measured in plasma.
During mental stress, an elevation of the FFA concentration was seen after placebo, probably reflecting catecholamine-induced HSL activation and consequently adipose tissue lipolysis. As no differences were noticed in plasma TG, it is not likely that the FFA increase during stress was due to lipolysis of TG-rich particles by lipoprotein lipase. Although the volunteers were equally stressed on both occasions, as suggested by a similar increase in heart rate in both tests, FFA did not increase after glucose during the stress period. As spillover of norepinephrine to the systemic circulation reflects more accurately sympathetic activity than plasma catecholamines (27), we relied mainly on the increase in heart rate as a reflection of sympathetic activation. The lack of FFA increase, representing diminished HSL activity, is in agreement with in vitro data demonstrating efficient HSL inhibition (5). Although insulin concentrations decreased at the beginning of stress, the levels were significantly higher than those during the stress period in the placebo test. We cannot exclude that the ingestion of glucose caused a change in sympathetic activity, as shown by Patel et al. (28). These researchers observed a postprandial increase in norepinephrine and a postprandial decline in epinephrine, which influenced adipose tissue lipolysis during the postprandial period. This may explain the tendency of FFA to rise after glucose during the first 10 min, which was not seen after placebo in the present study. Moreover, Patel et al. (28) hypothesized that increased postprandial insulin concentrations would override the sympathetic HSL activation, as demonstrated in our experiments. Under these experimental conditions, resembling a normal life situation, insulin-mediated inhibition of HSL cannot be overruled by catecholamine-mediated activation. However, our experimental approach does not allow us to make quantitative comparisons between inhibitory (insulin-mediated) and activating (catecholamine-mediated) stimuli.
It has been shown that the antilipolytic effect of insulin is stronger in sc than in visceral fat (29). As the visceral fat depot is in direct contact with the liver through the portal circulation, the inability of insulin to decrease lipolysis in visceral fat cells may result in a higher FFA flux to the liver and consequently VLDL overproduction, leading to dyslipidemia. Supporting this concept, a recent study has shown that low and physiological increased insulin concentrations suppress lipolysis in skeletal muscle (30).
We cannot rule out that some of the results obtained in the present study reflect insulin and catecholamine effects on skeletal muscle. During the second rest period after placebo, when heart rate and insulin concentrations were low, FFA declined to basal levels (0 min). However, although after glucose, heart rate and insulin concentrations decreased, FFA concentrations remained below basal concentrations. This could be explained by a prolonged action of plasma insulin on the suppression of HSL. Further research will be necessary to investigate the duration of suppression of HSL by insulin after a decrease in plasma insulin concentrations.
In conclusion, in this study we have demonstrated that the catecholamine-induced HSL activation can be overruled by insulin-mediated inhibition in healthy, insulin-sensitive volunteers. Furthermore, the suppression of FFA by insulin under physiological conditions is positively associated with fasting plasma insulin concentrations. This noninvasive in vivo test may be used in patients to test the antilipolytic action of insulin, HSL stimulation, and suppression of HSL activation by insulin under different conditions. Future studies will investigate whether different patient groups with differences in insulin sensitivity will show similar results concerning HSL regulation by insulin and catecholamines in vivo.
Acknowledgments
Footnotes
Abbreviations: CHD, Coronary heart disease; HR, heart rate; HSL, hormone-sensitive lipase; FCHL, familial combined hyperlipidemia; TG, triglyceride.
Received September 27, 2000.
Accepted May 14, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  M. Badeau, V. Vihma, T. S. Mikkola, A. Tiitinen, and M. J. Tikkanen Estradiol Fatty Acid Esters in Adipose Tissue and Serum of Pregnant and Pre- and Postmenopausal Women J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4327 - 4331. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ribalta, C. J.M. Halkes, J. Salazar, L. Masana, and M. C. Cabezas Additive Effects of the PPAR{gamma}, APOE, and FABP-2 Genes in Increasing Daylong Triglycerides of Normolipidemic Women to Concentrations Comparable to Those in Men Clin. Chem., May 1, 2005; 51(5): 864 - 871. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J van Oostrom, H. van Dijk, C. Verseyden, A. D Sniderman, K. Cianflone, T. J Rabelink, and M. Castro Cabezas Addition of glucose to an oral fat load reduces postprandial free fatty acids and prevents the postprandial increase in complement component 3 Am. J. Clinical Nutrition, March 1, 2004; 79(3): 510 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Meijssen, R. J. Derksen, S. Bilecen, D. W. Erkelens, and M. C. Cabezas In Vivo Modulation of Plasma Free Fatty Acids in Patients with Familial Combined Hyperlipidemia Using Lipid-Lowering Medication J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1576 - 1580. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and glucose (•) tests in 13 healthy volunteers. Data are the
mean ± SEM.
