| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online June 13, 2005 as doi:10.1096/fj.04-3453fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of
* Internal Medicine and
Surgery, CNR Centro di Fisiopatologia dello Shock, Catholic University, School of Medicine, Rome, Italy; 
Computer and Systems Science Department, University of Rome "La Sapienza," Rome, Italy;
Institut National de la Santé et de la Recherche Médicale (INSERM) Unité 449, Lyon, France; and
|| Department of Internal Medicine and C.N.R. Institute of Clinical Physiology, University of Pisa, Italy
1Correspondence: Dipartimento di Medicina Interna, Catholic University, Largo A. Gemelli 8, Rome 00168, Italy. E-mail: gmingrone{at}rm.unicatt.it
SPECIFIC AIMS
Obese individuals have higher than normal plasma levels of leptin, which are, however, unable to achieve the appropriate responses. Among the possible mechanisms that can result in leptin resistance there is, at least in animals, the overfeeding of high-fat diets. Consumption of high-fat diets leads to increased energy intake and thus to gain weight up to frank obesity in humans as well. However, the mechanisms by which high-fat diets lead to weight gain are poorly understood. To our knowledge, no data are reported in the literature about the effect of long-term low-fat intake on 24 h leptin levels in obese subjects.
Bilio-pancreatic diversion (BPD) is a bariatric surgical technique that produces massive lipid malabsorption, thus mimicking the effects of a very low-fat diet with stable changes in fat-metabolizable energy from 
47% to 20%.
The long-term effects of a very low-fat diet and weight loss on insulin and glycemic responses to meals and 24 h circulating leptin concentration were examined. Since plasma cortisol and GH levels have been implicated in the regulation of 24 h circulating leptin profiles, plasma cortisol and GH profiles were also studied. To correlate the effects of leptin modification on 24 h substrate oxidation measured in the calorimetric chamber, levels of the isoform 2 of acetyl CoA carboxylase (ACC2) were measured by quantitative PCR in skeletal muscle tissue biopsies of 9 severely obese women, 35 ± 5-years-old, before and after BPD (BMI 28.9±0.9 vs.58.6±9.6 kg/m2).
1. Energy balance
Energy intake did not differ significantly before and after BPD; but after the bariatric operation, as a consequence of the fecal fat loss, the metabolizable energy became only 40% of the energy intake. The carbohydrate (CHO) fecal loss accounted for 10.78 ± 2.17% of the CHO intake, the loss of lipids with feces was 80.67 ± 5.10%, and that of proteins 17.33 ± 4.90%. However, the metabolizable energy normalized by kgFFM was not statistically different before and after BPD.
Twenty-four hour energy expenditure was 11300.58 ± 1457.57 kJ before BPD and was decreased to 7278.22 ± 773.61 kJ after BPD (P<0.0001). However, if 24 h EE was normalized by FFM, the mean value after BPD was significantly higher than that before BPD (87.04±7.58 vs. 68.51±3.39 kJ·kgFFM–1, P<0.001).
Twenty-four hour glucose oxidation was 323.94 ± 55.04 g (or 5077.83±862.80 kJ) before BPD and became 298.84 ± 33.31 g (or 4684.36±522.15 kJ) after BPD (P=n.s.), while 24 h lipid oxidation was significantly (P<0.0001) reduced (131.94±35.58 vs. 44.56±15.10 g; or 4963.83±1338.5 vs. 1636.39±568.27 kJ). However, the average lipid oxidation was 97.2 ± 3.1% (P<0.01) of the metabolizable lipid intake after BPD whereas it represented 69.2 ± 8.5% before operation.
2. Insulin sensitivity
Whole body glucose uptake significantly increased (P<0.0001) after BPD from 27.4 ± 2.20 to 57.3 ± 2.7 µmol·kgFFM–1·min–1. No significant difference was observed in the plasma level of insulin in the last 40 min of the clamp (550.50±52.87 before BPD vs. 552.75±51.95 pmol·L–1 after BPD).
3. ACC2 expression in the skeletal muscle
An average 59.26 ± 6.23% decrease of the ACC2 skeletal muscle level was observed, since it significantly (P<0.0001) fell from 452.82 ± 76.35 before BPD to 182.45 ± 40.69% of cyclophilin mRNA after BPD. No significant difference was found in the average concentration of cyclophilin mRNA before (20.7±2.4 amol·mg–1 total RNA) and after BPD (19.7±3.5 amol·mg–1 total RNA, P=0.84).
4. Skeletal muscle malonyl-CoA content
Weight loss resulted in a 41.40 ± 6.74% reduction in muscle malonyl-CoA levels (from 0.28±0.02 to 0.16±0.01 nmol·g-1; P<0.0001). ACC2 changes significantly (P<0.0001) correlated (changes in malonyl-CoA=0.3865xchanges in ACC2+19.048, R2=0.81) with the changes in malonyl-CoA levels in the skeletal muscle samples.
5. Ultradian variability (pulsatility) analysis
The area under the curve of the leptin levels before BPD, computed using a trapezoidal rule, was 1004.94 ± 132.95 ng·h–1·mL–1 and decreased significantly (P=0.002) to 286.10 ± 59.86 ng·h–1·mL–1 after BPD.
Plasma leptin showed diurnal variations with a peak (median) at 2:30 AM and a nadir at 10:30 AM (Fig. 1
). Levels of circulating leptin rose progressively after each meal and fell to trough levels within 1–2 h.
|
The maximum diurnal variation (acrophase value) of leptin levels was 22.60 ± 2.79 ng/mL–1, representing a 74.03 ± 4.44% of the mean 24 h leptin level, before BPD, and 10.27 ± 1.70 ng/mL, corresponding to 158.16 ± 18.86% of the mean 24 h leptin level after BPD (P=0.001).
The pulsatility index increased (P=0.02) from 1.050 ± 0.004 ng·mL–1·min–1 before BPD to 1.084 ± 0.005 ng·mL–1·min–1 after BPD. Leptin plasma clearance also increased (P=0.01) from 0.0010 ± 0.0001 before BPD to 0.0020 ± 0.0001·min–1 after BPD. The average number of peaks did not vary significantly (median 11.8 before and 12.7 after BPD).
The average plasma leptin concentration over 24 h normalized by FM did not change after weight loss (0.54 ± 0.24 vs. 0.57 ± 0.21 µg·L–1 · kgFM–1, P=n.s.).
The insulin AUC was significantly (P=0.005) reduced after BPD (from 4908 ± 786 to 1364 ± 118 pmol·h–1·L–1). Obviously, the insulin peaks were meal-related, but their amplitude markedly decreased after BPD.
The acrophase of insulin levels, calculated as the difference between the mean peak and trough values, was 475.86 ± 61.64 pM before BPD and 172.14 ± 12.39 pM after BPD (P=0.001).
The pulsatility index of insulin increased significantly after the surgical operation (1.53±0.13 vs. 1.10±0.12 pM·min–1, P=0.01) similarly to the insulin clearance rate (from 0.005 ± 6·10–4 min–1 before BPD to 0.009 ± 4·10–4 min–1 after BPD; P=0.01).
Plasma GH AUC significantly (P<0.0001) increased after BPD, from 11.17 ± 2.58 to 37.25 ± 1.41 µg· L–1·h–1.
Plasma GH acrophase was 0.91±0.20 µg·L–1·min–1 before BPD and 4.58 ± 0.80 µg·L–1·min–1 after BPD (P=0.0001). No significant change was observed after BPD in the cortisol AUC (from 4409.58 ± 520.18 to 4356.16 ± 429.42 nmol·L–1·h–1).
The influence of long-term insulin fluctuation on leptin was largely different before (lag period=2 min; coefficient=60%) than after (lag period=80 min; coefficient=70%) BPD.
6. Correlation
The correlation between fasting plasma leptin concentration and FM was stronger before BPD (fasting leptin=0.7415 FM–15.927, R2=0.71; P<0.001) than after BPD (fasting leptin=1.3703 FM–18.225, R2=0.57; P<0.05.)
Changes in leptin negatively correlated with changes in M with a R2 of 0.69 (
leptin=–0.4147 M+3.2098, P<0.001).
Using changes in leptin as an independent variable and changes in FM, GH, insulin, blood glucose, FFA, and glucose/insulin (G/I) ratio, all averaged over 24 h, as dependent variables in a stepwise regression (R2=0.87 with a significance of F equals to 0.0055), changes in FFA (B=0.105±0.018, P=0.002) and changes in G/I (B=0.247±0.081, P=0.029) resulted to be the best predictors of the leptin variations, whereas modification in insulin, GH and glucose levels did not affect the changes in leptin.
CONCLUSIONS AND SIGNIFICANCE
The principal findings of this study are that the pulsatility index as well as the clearance of plasma leptin, GH, and insulin significantly increased after weight loss in formerly obese women with massive lipid malabsorption consequent to bilio-pancreatic diversion. The major determinants of the changes in leptin concentration, averaged over 24 h, other than fat mass reduction, were the variation of mean daily circulating levels of free fatty acids together with that of glucose/insulin ratio; whereas insulin, cortisol, and GH did not seem to play a significant role in controlling the modification of leptin concentrations. Finally, muscle ACC2 mRNA levels were significantly reduced after BPD in these subjects.
Both in vitro and in vivo studies have suggested that the inhibitory role on leptin secretion is exerted by changes in insulin-mediated glucose disposal or, in other terms, in insulin sensitivity rather than by variations of insulin levels per se. Our data support this hypothesis since it is the glucose/insulin ratio that represents, together with FFA circulating levels, the major determinant of 24 h leptin changes. The changes in leptin plasma levels were well correlated with the variation of whole body glucose uptake.
Changes in glucose metabolism may also explain the observation that high-fat meals lower 24 h circulating leptin levels relative to high-carbohydrate meals in human subjects, suggesting a mechanism that may contribute to the effects that high-fat diets have in promoting increased energy intake, weight gain, and obesity.
High-fat diet seems to induce leptin resistance. In rats under high-fat dietary regimen for 4 wk, the stimulatory effect of leptin on lipid oxidation and hydrolysis in the soleus muscle was elicited. The human skeletal muscle of obese subjects is also leptin resistant.
Our findings seem to support the effect of a low-fat diet in reducing 24 h plasma leptin concentration, although it is very hard to discriminate between the contribution deriving from the fat mass loss and that depending on a very low-fat diet in our population. The lack of changes in leptin concentration normalized by kilogram of fat mass after weight loss, however, is supportive of the hypothesis that it is the reversal of insulin resistance, which has been reported to be associated in these subjects to the reduction of triglyceride depots in skeletal muscle fibers, rather than the degree of weight loss implicated in the restoration of leptin sensitivity. In fact, it has been proposed that the major components of the metabolic syndrome of obese humans, such as the insulin resistance, may result from a failure of leptin liporegulation to prevent lipid overload of lean body mass and lipoapoptosis in certain organ systems.
The reversibility of a leptin resistance state simultaneous or consequent to the reversion of insulin resistance might explain why lower plasma levels of leptin are able to maintain a constant body weight, close to normal standards in formerly obese subjects after BPD.
Leptin selectively stimulates phosphorylation and activation of the 
2 catalytic subunit of AMPK (2 AMPK) in skeletal muscle. 2 AMPK phosphorylates acetyl CoA (CoA) carboxylase (ACC) and malonyl CoA decarboxylase (MCD) (Fig. 2
). A reversal of leptin resistance might determine a reduced synthesis of malonyl-CoA resulting in increased fatty acid oxidation as observed in our series, which could account for the reduction in the triglyceride content of skeletal muscle tissue previously observed in these subjects.
|
In conclusion, the reversion of insulin resistance might allow to a reversal of leptin resistance, restoration of leptin pulsatility, and consequent inhibition of ACC2 mRNA expression translating in a reduced synthesis of malonyl-CoA that results in increased fatty acid oxidation. The latter could account for the reduction in the triglyceride content of skeletal muscle tissue previously observed in these subjects. Since leptin inhibits GH secretion, a reduction of circulating leptin levels can increase GH secretion as observed in our series.
We hypothesize that, under a high-fat diet, insulin resistance takes place and skeletal muscle becomes resistant to the effects of leptin, resulting in the accumulation of intramuscular TG. This may be an important initiating step in the development of leptin resistance so common in obesity.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-3453fje;
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
