, 44:1125

Effects of Weight Loss on Lipid Transfer Proteins in Morbidly Obese Women


  • Markus W. Laimer
    • Department of Internal Medicine IMedical University Innsbruck
  • Julia Engl
    • Department of Internal Medicine IMedical University Innsbruck
  • Alexander Tschoner
    • Department of Internal Medicine IMedical University Innsbruck
  • Susanne Kaser
    • Department of Internal Medicine IMedical University Innsbruck
  • Andreas Ritsch
    • Department of Internal Medicine IMedical University Innsbruck
  • Tobias Tatarczyk
    • Department of Internal Medicine IMedical University Innsbruck
  • Markus Rauchenzauner
    • Department of PediatricsMedical University Innsbruck
  • Helmut Weiss
    • Department of General SurgerySJOG Hospital Salzburg
  • Franz Aigner
    • Department of General, Thoracic and Transplant SurgeryMedical University Innsbruck
  • Josef R. Patsch
    • Department of Internal Medicine IMedical University Innsbruck
    • Department of Internal Medicine IMedical University Innsbruck
Original Article

DOI: 10.1007/s11745-009-3349-8

Cite this article as:
Laimer, M.W., Engl, J., Tschoner, A. et al. Lipids (2009) 44: 1125. doi:10.1007/s11745-009-3349-8


Obesity is associated with lipid abnormalities leading to an increased morbidity and mortality from atherosclerotic disease. Lipid transfer proteins such as Cholesteryl Ester Transfer Protein (CETP) and Phospholipid Transfer Protein (PLTP), and lipases such as lipoprotein lipase (LPL) and hepatic lipase (HL) are involved in the pathogenesis of the obesity associated proatherogenic dyslipidemia. Nineteen severely obese female subjects undergoing laparosopic gastric banding participated in this prospective study. Subjects were examined with respect to body composition, lipid profile, CETP, PLTP, LPL and HL before and 1 year after surgical treatment. Mean weight loss was 22.2 kg, mainly due to losses in the fat depots. Triglycerides decreased and HDL2-C increased significantly. In respect to transfer proteins mean CETP mass decreased from 1.82 to 1.71 μg mL−1 (P = 0.043) and mean PLTP activity was reduced from 7.15 to 6.12 μmol mL−1 h−1 (P = 0.002), in parallel. In addition, both mean LPL activity and mean HL activity tended to decrease from 297 to 248 nmol mL−1 h−1 for LPL (P = 0.139) and from 371 to 319 nmol mL−1 h−1 for HL (P = 0.170), respectively. We conclude that weight loss induced by bariatric surgery is associated with the amelioration of the obesity-associated dyslipidemic state. This improvement may be attributable to decreased mass and action of the adipocyte tissue derived lipid transfer proteins CETP and PLTP.


Weight lossCholesteryl ester transfer proteinHepatic lipaseLipidsLipoprotein lipasePhospholipid transfer protein



Cholesteryl esters


Cholesteryl ester transfer protein


High density lipoprotein


Hepatic lipase


Laparoscopic adjustable gastric banding


Low density lipoprotein


Lipoprotein lipase


Phospholipid transfer protein


Very low-density lipoprotein


Obesity is linked to a variety of metabolic and hormonal dysfunctions such as the development of insulin resistance and dyslipidemia leading to increased morbidity and mortality in these subjects [1]. Obesity is associated with dyslipidemia, elevated triglyceride concentrations, low high density lipoprotein (HDL) cholesterol levels and small, dense low-density lipoprotein (LDL) particles. These alterations in plasma lipoprotein-lipid metabolism have been shown to increase the risk for cardiovascular disease in both men and women [2].

The state of dyslipidemia is influenced by the body fat distribution with central obesity leading to a higher risk for cardiovascular disease [3]. Both, transfer proteins such as cholesteryl ester transfer protein (CETP) and phospholipid transfer protein (PLTP), and lipases such as lipoprotein lipase (LPL) and hepatic lipase (HL) are candidates to mediate the higher risk for cardiovascular disease [4].

Adipose tissue is a prominent source of CETP and obesity is associated with elevated plasma CETP concentrations [5, 6]. CETP is an important determinant of lipoprotein composition due to its capacity to mediate the transfer of cholesteryl esters (CE) from CE-rich lipoproteins to TG-rich lipoproteins in exchange for triglycerides. It has been suggested that CETP plays a key role in the reversibility of the atherogenic lipoprotein profile seen in obese subjects [7]. During marked reduction in fat mass the serum levels of both CETP-mass and CETP-activity are reduced in parallel [8, 9].

PLTP circulates bound to HDL and mediates the transfer of phospholipids from apo B-containing lipoproteins into HDL, thus modulating HDL size and lipid composition. PLTP activity generates pre-beta HDL, the major acceptor of cholesterol in the reverse-cholesterol transport route [10]. Liver, adipose tissue and lung are presumably the major sources of circulating PLTP [11]. PLTP-knockout mice have been shown to have markedly reduced HDL levels due to defective transfer of phospholipids from triglyceride-rich lipoproteins into HDL [12]. In a cross-sectional study elevated PLTP levels were associated with an increased risk for coronary artery disease [13]. Furthermore, it has been hypothesized that reduction of PLTP activity and increase of HDL particle size are important component factors in converting the atherogenic lipoprotein profile of obese subjects into a less atherogenic profile with weight loss [14].

The HL plays a role in the metabolism of chylomicrons and very low-density lipoprotein (VLDL) remnants, LDL, and high-density lipoproteins (HDL), which are all implicated in atherosclerosis [1517]. A decrease in the activity of postheparin HL was found in 21 obese older men after weight loss, suggesting that the reduced HL concentrations might mediate the improvement in the lipoprotein profile [18].

LPL is a rate-limiting enzyme that hydrolyzes circulating triglyceride (TG)-rich lipoproteins, such as VLDL and chylomicrons [19]. A decrease in LPL activity is associated with an increase in plasma TG levels and a decrease in HDL-cholesterol [19].

As compared to previous weight loss studies, which evaluated either lipid transfer proteins or lipolytic enzymes, we determined—to our knowledge for the first time—the influence of weight reduction on both the lipid transfer proteins CETP and PLTP and the lipolytic enzymes LPL and HL in parallel and in a prospective study design.

Subjects and Methods


Nineteen morbidly obese women defined by a BMI of more than 40 kg/m² participated in this prospective study. Subjects desiring surgical intervention for the treatment of obesity were referred from the surgical department to the outpatient clinic for metabolism, where they were consecutively screened for eligibility. Exclusion criteria were secondary causes of adiposity, diabetes mellitus, pregnancy, intake of lipid lowering drugs or other medically significant illness. Examinations of the study subjects were undertaken within 2 months prior to Laparoscopic Adjustable Gastric Banding (LAGB) and 1 year post LAGB. Informed consent was obtained from each participant before entering the study, and all procedures were performed in accordance with institutional guidelines at the Internal Department of the medical faculty of the University of Innsbruck. The study protocol was approved by the Ethical Committee of the Medical University Innsbruck.

Surgical Procedure

The surgical procedure was performed as described by Forsell [20] at the Department of Surgery, University of Innsbruck [21]. The Swedish Adjustable Gastric Band was used in all of the study patients (SAGB Obtech Medical AG, Zug, Switzerland).

Analysis of Body Composition

BMI was calculated as body weight in kilograms divided by height in meters squared. Body composition was determined by impedance analysis using InBody 3.0 Body Composition Analyzer from Biospace Europe (Dietzenbach, Germany). Measurements were taken in the morning in the fasted state.

Glucose Metabolism, Leptin and Lipoprotein Analysis

Blood was collected after an overnight fast. Plasma was separated from erythrocytes by centrifugation at 3,000 rpm for 10 min at 4 °C immediately after collection. Plasma samples were stored frozen at −80 °C until assayed. Plasma TG, cholesterol, apo-AI and apo-B concentrations were quantified using a commercially available enzymatic kit (Roche Diagnostic Systems, Basel, Switzerland) on a Cobas Mira analyzer. HDL-cholesterol and HDL3-C concentrations were determined using precipitation procedures with polyethylene glycol (Immuno, Vienna, Austria) [22]. HDL2-C concentrations were calculated by subtracting HDL3-C from HDL-C. LDL-C was calculated according to the formula of Friedewald et al. [23]. Plasma glucose was measured by the hexokinase method on a Cobas MIRA analyzer. Plasma insulin was determined by a micro particle enzyme immunoassay from Abbott (Wiesbaden, Germany). The homeostasis model of assessment of insulin resistance (HOMA-IR) was calculated by the following formula: fasting serum insulin concentration (μIU mL−1) × blood glucose concentration (mmol L−1)/22.5. Leptin concentrations were measured using an ELISA (R&D Systems, Wiesbaden, Germany).


CETP concentrations were determined by capture enzyme-linked immunosorbent assay. Wells were coated with recombinant single-chain antibody fragments 1CL8 [24]. CETP was detected using a polyclonal anti-CETP antibody conjugated directly to alkaline phosphatase [25, 26].

PLTP Activity

The ability of plasma PLTP to transfer [3H]-dipalmitoyl-phosphatidylcholine (NEN Life science Products, Boston, MA, USA) from phosphatidylcholine vesicles to HDL3 was measured as described previously [27].

LPL- and HL-Activity

LPL activity and HL activity were measured as described previously [28, 29]. Briefly, postheparin plasma was collected 15 min after an intravenous heparin dose of 50 Units per kg body weight into cold tubes to maintain the enzyme activity. For HL activity, we used a gum arabic emulsion of [3H]-glycerol-trioleate, under these conditions LPL is inactivated. To measure LPL activity we used an emulsion of Intralipid® into which [3H]-glycerol-trioleate was incorporated by sonication. HL activity was inhibited by goat immunoglobulins to human HL, that was a kindly provided by Thomas Olivecrona, University of Umea, Sweden.

Statistical Analysis

Descriptive data are expressed as mean values ± SD. Normal distribution was estimated using a Shapiro–Wilk test. Normally distributed data from the pre- and post-gastric banding group were compared using a paired-samples t test. Not normally distributed data from the pre- and post-gastric banding group were compared using a Wilcoxon test for paired samples. Associations between PLTP activity and CETP mass were assessed using the Spearman r correlation coefficient. Statistical significance was inferred at a two-tailed P value of less than 0.05. Statistical analyses were calculated using SPSS release 11.5 for Windows (SPSS, Chicago, USA).


Clinical Characteristics

Anthropomorphic measures of the study subjects are shown in Table 1. The mean age was 37.2 ± 11.6 years in pre-gastric banding subjects. In this group the mean body weight was 116.8 ± 12.2 kg and the BMI was 41.7 ± 2.9 kg/m2. Mean body weight loss after the surgical procedure was 22.2 kg (P < 0.001), respectively. Mean BMI decreased in parallel by 8 kg/m2 after 1 year (P < 0.001). The body fat mass was 57.5 ± 8.1 kg in the subjects undergoing surgical intervention. After bariatric surgery body fat mass decreased to 37.0 ± 8.4 kg, corresponding to a loss of 20.5 kg fat mass (P < 0.001). Thus, in these study subjects weight loss mainly due to loss of fat mass.
Table 1

Anthropomorphic measures of the study participants at baseline and after weight loss induced by LAGB




Test probability

Age (years)

37.2 ± 11.6

38.5 ± 11.7


Height (cm)

167 ± 6


Weight (kg)

116.8 ± 12.2

94.6 ± 14.4


BMI (kg/m²)

41.7 ± 2.9

33.7 ± 4.3


Fat mass (kg)

57.5 ± 8.1

37.0 ± 8.4


Data are expressed as means ± SD. Statistical significance was estimated by paired-samples t test

Lipids and Apolipoproteins

Lipid parameters at baseline and after LAGB are shown in Table 2. Mean triglyceride levels decreased by 0.45 mmol L−1 after weight loss (P = 0.027). HDL-cholesterol tended to increase from 1.37 ± 0.31 mmol L−1 in the pre-gastric banding group to 1.43 ± 0.28 mmol L−1 in the post-gastric banding group. Mean HDL2-C concentrations increased significantly by 0.02 mmol L−1 (P = 0.009). Further mean Apo-B levels decreased 4 mg dL−1 from baseline concentrations 1 year after LAGB, respectively.
Table 2

Lipid parameters and apolipoproteins at baseline and after weight loss induced by LAGB




Test probability

Glucose (mmol L−1)a

5.28 ± 1.07

5.23 ± 0.62


Insulin (pmol L−1)a

122.7 ± 113.4

61.3 ± 38.1



4.40 ± 4.79

2.06 ± 1.26



44.4 ± 17.0

18.5 ± 11.9


Total cholesterol (mmol L−1)

5.22 ± 0.72

4.86 ± 0.91


Triglyceride (mmol L−1)

1.52 ± 0.97

1.07 ± 0.47


LDL-C (mmol L−1)

3.12 ± 0.57

2.88 ± 0.75


HDL-C (mmol L−1)

1.37 ± 0.31

1.43 ± 0.28


HDL2-C (mmol L−1)

0.26 ± 0.13

0.28 ± 0.10


HDL3-C (mmol L−1)

1.14 ± 0.20

1.14 ± 0.20


apo-AI (mg dL−1)

159 ± 22

157 ± 24


apo-B (mg dL−1)

89 ± 18

85 ± 17


Data are expressed as means ± SD

ns not significant

Statistical significance was estimated by paired-samples t test or aWilcoxon-signed-ranks test

Lipid Transfer Proteins and Lipolytic Enzymes

The CETP mass was 1.81 ± 0.5 μg mL−1 at baseline. After a weight loss of 20.5 kg in fat mass, CETP mass decreased by 8.3% (P = 0.043; Table 3). Baseline PLTP activity of 7.15 ± 1.0 (μmol mL−1 h−1) decreased by 14.4% (P = 0.002; Table 3). In univariate analysis CETP mass and PLTP activity were significantly associated at baseline (r = .539, P = 0.026), but not after weight loss (r = .248, P = 0.338). The correlation coefficient for Δ CETP and Δ PLTP was not significant (r = .074, P = 0.787). LPL activity tended to decrease from baseline levels of 297 ± 96 to 248 ± 74 nmol mL−1 h−1 (P = 0.139) and, also, the HL activity tended to decrease from 371 ± 148 to 319 ± 151 nmol mL−1 h−1 (P = 0.170).
Table 3

Lipoprotein transfer proteins and lipases at baseline and after weight loss induced by LAGB




Test probability

CETP (μg mL−1)

1.81 ± 0.50

1.66 ± 0.44


PLTP activity (μmol mL−1 h−1)

7.15 ± 1.00

6.12 ± 1.07


LPL activity (nmol mL−1 h−1)

297 ± 96

248 ± 74


HL activity (nmol mL−1 h−1)

371 ± 148

319 ± 151


Data are expressed as means ± SD. Statistical significance was estimated by paired-samples t test


Obesity is associated with a higher risk for morbidity and mortality from atherosclerotic disease [1]. Bariatric surgery is an efficient method for reducing body weight and thereby reducing mortality and morbidity [30]. Weight loss leads to beneficial metabolic effects such as lowering serum triglycerides, improving the insulin sensitivity and increasing HDL-cholesterol. The Swedish Obese Subjects Study represents the first prospective controlled intervention study on overall mortality after bariatric surgery [31]. After an observation period of 10 years several lipid parameters ameliorated in the bariatric surgery group: triglycerides decreased by 16%, HDL-cholesterol increased by 24% and total cholesterol decreased by 5.4% [31]. Changes in lipid transfer proteins and lipolytic enzymes could account for these improvements.

Consequently, we evaluated the influence of weight reduction by bariatric surgery on the lipid transfer proteins CETP and PLTP and the lipolytic enzymes LPL and HL in parallel in a prospective study. The study cohort comprised healthy obese women to minimize potential confounding factors.

One year post gastric banding fat mass decreased by 35.7% and, in parallel, CETP mass and PLTP activity decreased significantly by 8.3% and by 14.4%, respectively. We previously reported that CETP mass and activity decreased in subjects undergoing weight loss after bariatric surgery leading to a more favorable lipoprotein profile. After the bariatric procedure a consistent increase in LDL particle diameters was found in parallel to the plasma CETP diminution [8]. PLTP is another lipid transfer protein that plays a crucial role in improving the atherogenic lipoprotein profile of obese subjects during weight loss. In a previous study we found a significant reduction of PLTP activity and a concomitant increase in HDL2 particle size during weight loss, whereas no alteration of HDL3 particle size was observed [14]. We hypothesized that a reduction in PLTP activity and an increase in HDL particle size are important component factors in converting the atherogenic lipoprotein profile of obese subjects into a less atherogenic profile with weight loss [14]. Furthermore, decreased PLTP activity was positively correlated with the change in subcutaneous fat after weight loss, but there was no significant relationship between the change in PLTP activity and the change in intraabdominal fat and insulin sensitivity [32]. Parallel CETP and PLTP plasma level reduction has already been found after slight weight loss induced by caloric restriction in obese women [33]. CETP and PLTP showed to correlate positively in obese females before weight reduction [33]. After weight loss the positive relationship between CETP and PLTP disappeared [33]. Both lipid transfer proteins seem to play a favorable role in the variances of lipids after weight reduction.

In contrast, postheparin HL activity and LPL activity did not change significantly after LAGB induced weight loss in the present study. However, we noted a tendency to decrease in both lipolytic enzymes. The role of HL and LPL in weight loss associated improvements of the lipid profile has been explored in previous studies. Pardina et al. [34] recently reported a significant reduction of plasma and liver HL activity after gastric bypass surgery. Purnell et al. [18] observed a reduction in postheparin HL activity after weight loss which contributed to the observed increase in LDL size and HDL2-C, especially in those subjects with pattern B LDL particles before weight loss. A recently published study postulated that the lowering of plasma lipids following a weight reduction program was due to increased expression of both LPL and LDL receptor mRNA [35]. However, in another study by Berman et al. [36] regional adipose tissue LPL did not change after weight loss. The different weight loss inducing interventions investigated in the above-mentioned trials may be responsible for the conflicting results. In our study, HL and LPL activity seem to be unaffected by LAGB, a caloric intake restricting procedure.

A major limitation of this study is the restriction to women and the limited number of subjects. The latter could explain that, although we found a decrease in both LPL and HL activity, these differences were not significant in the respective statistical analysis.

In conclusion weight loss induced by bariatric surgery results in amelioration of TG and HDL-2 concentrations in normolipidemic obese women. These improvements may be attributable to decreased mass and action of the adipocyte tissue derived lipid transfer proteins CETP and PLTP, while the modest increase in HDL-2 could be the consequence of the change in TG independent of, or in combination with, the changes in the transfer proteins.


The expert technical assistance of Ursula Stanzl and Karin Salzmann is gratefully acknowledged.

Copyright information

© AOCS 2009