Zinc-α2-Glycoprotein Expression in Adipose Tissue of Obese Postmenopausal Women before and after Weight Loss with and without Exercise

Abstract

Objective

Zinc-Alpha 2-Glycoprotein (ZAG) has recently been implicated in the regulation of adipose tissue metabolism due to its negative association with obesity and insulin resistance. The purpose of this study is to investigate the relationships between adipose tissue ZAG expression and central obesity, and the effects of six-months of weight loss (WL) or aerobic exercise + weight loss (AEX+WL) on ZAG expression.

Design and Methods

A six-month, longitudinal study of 33 healthy, overweight or obese postmenopausal women (BMI: 25–46 kg/m²) was conducted. Abdominal and gluteal adipose tissue samples were obtained before and after AEX+WL (n=17) and WL (n=16). ZAG expression was determined by RT-PCR.

Results

Prior to interventions, abdominal ZAG expression was negatively correlated with visceral fat (r=−0.50, P<0.005), sagittal diameter (r=−0.42, P<0.05), and positively related to VO₂max (r=0.37, P<0.05). Gluteal ZAG expression was negatively correlated with weight, fat-free mass, visceral fat, resting metabolic rate, and fasting insulin (r=−0.39 to −0.50, all P<0.05). Abdominal ZAG mRNA levels increased, though not significantly, 5% after AEX+WL and 11% after WL. Gluteal ZAG mRNA levels also did not change significantly with AEX+WL and WL.

Conclusions

Abdominal ZAG expression may be important in central fat accumulation and fitness that modestly but not significantly increases with weight reduction alone or with aerobic training in obese postmenopausal women.

INTRODUCTION

Twenty-five percent of adults in the U.S. suffer from obesity and obesity-related health risks¹. ZAG has recently been implicated in the regulation of adipose tissue metabolism due to its negative correlation with obesity². Originally discovered in plasma of cancer cachexia patients, ZAG is overexpressed by various malignant tumors and later shown secreted by healthy tissue^3,4,5.

Much of the known molecular mechanisms of ZAG are derived from animal studies. Research shows that the introduction of ZAG producing tumors leads to up-regulation of ZAG in adipose tissue accompanied by profound WL due to fat depletion, without changing FFM or nutrient intake^{1, 6}. ZAG knockout animals are more prone to weight gain than their wild-type siblings when given a high-fat diet^{7, 8}. In addition, treatment with β-AR agonists stimulates ZAG production while β-AR antagonists suppress it, implying overexpression of ZAG is most likely regulated through β-AR and the sympathetic system in cachectic mice^{3, 4}. Since the administration of isolated human ZAG to mice via drinking water causes an increase in endogenous murine serum and adipose ZAG protein and mRNA, ZAG may induce its own regulation⁴.

There are limited human studies of ZAG expression and little is known of ZAG’s role in fitness, energy expenditure and body composition. Plasma ZAG levels are not different between lean and obese young and middle-aged adults⁷. In contrast, ZAG expression is significantly lower in omental and subcutaneous adipose tissue as well as liver of morbidly obese individuals compared to non-obese individuals². Considering the interplay between physical activity and obesity, it is conceivable that ZAG expression could be influenced by WL and exercise. We propose a novel hypothesis that adipose tissue ZAG expression is inversely related to visceral and subcutaneous abdominal fat area, aerobic capacity, and glucose tolerance, and that exercise and WL could increase ZAG expression. Our purpose was to investigate these relationships and to test the effects of six months of AEX+WL or WL alone on ZAG expression.

RESEARCH METHODS

Healthy, overweight/obese (BMI>25 kg/m²), postmenopausal women aged 50–78 years were screened by medical history, physical examination, fasting blood profile, and treadmill test, and showed no evidence of any medical disorders. Women were nonsmokers, weight stable, and sedentary. Data are reported here on 33 women who met study criteria, completed AEX+WL (n=17) and WL (n=16), and had adipose tissue aspirations. The study sample was Caucasian (n=22) and African American (n=11). The Institutional Review Board of University of Maryland approved all methods and procedures. Each participant provided written informed consent.

Study Protocol

Physical Characteristics

Height, weight, waist circumference, VO₂max and body composition by dual-energy X-ray absorptiometry and computed tomography were determined as described⁹. CT Scans were missing due to scheduling conflicts (n=3 WL and n=3 AEX+WL) or small field of view to determine abdominal subcutaneous fat (WL n=9 and AEX+WL n=7).

Fat aspirations with RT-PCR, OGTT, and blood samples

Women were instructed in TLC diet by a registered dietitian 1x/week for 6–8 weeks prior to the fat aspiration and OGTT¹⁰. Subjects were weight stable (±2%) at least two weeks prior to baseline testing, and maintained the TLC diet throughout the study. Women were provided with an eucaloric diet for two days before abdominal and gluteal fat aspirations were performed after a 12-hr overnight fast and 36–48 hours after the last exercise bout for AEX+WL¹¹. Total RNA was first extracted from adipose tissue and RT-PCR conducted as described¹¹. A two-hour OGTT with measurement of plasma glucose and insulin were performed⁹.

WL and AEX+WL

For six months, all women attended weekly WL classes and restricted caloric intake by 500 kcal/d. The AEX+WL group used treadmills and elliptical trainers three times a week for 6 months at the exercise facility⁹. There was a ~86% average compliance to exercise and WL classes.

Statistical Analyses

Univariate ANOVA tested for differences in changes between AEX+WL and WL. Paired t-tests tested differences between pre- and post-intervention measures within each intervention. Regression analyses determined relationships between adipose tissue ZAG, body composition, and glucose metabolism. Data were analyzed by SPSS (PASW Statistics 18) (SPSS Inc., Chicago), expressed as mean±SEM, and significance at P<0.05.

RESULTS

Prior to the interventions, abdominal ZAG mRNA were negatively correlated with VAT (r=−0.50, P=0.003, Figure 1a) and sagittal diameter (r=−0.42, P<0.05). In addition, abdominal ZAG was related to VO₂max (r=0.34, P=0.05, n=31). In a multiple regression analyses with VAT and VO₂max, VAT remained an independent predictor of abdominal ZAG (P=0.03). Gluteal ZAG mRNA were negatively correlated with weight (r=−0.39, P<0.05, n=32), FFM (r=−0.39, P<0.05, n=32), VAT (r=−0.41, P<0.05, n=33), RMR (r=−0.50, P<0.01, n=29), and fasting insulin (r=−0.42, P<0.05, n=32, Figure 1b). In a multiple regression analyses with weight, FFM, VAT, RMR and insulin, RMR remained an independent predictor of gluteal ZAG (P=0.02). Abdominal and gluteal ZAG mRNA were both related to muscle attenuation (r=0.39, P<0.05 and r=0.31, P=0.08, n=32, respectively). Neither abdominal nor gluteal mRNA were related to subcutaneous abdominal fat, fasting or 120 min glucose.

Figure 1.

(a) The relationship between abdominal ZAG mRNA levels and VAT mass (r=−0.50, P<0.005). (b) The relationship between gluteal ZAG mRNA levels and fasting insulin (r=−0.42, P< 0.05). (c) The relationship between absolute changes in abdominal ZAG mRNA expression and absolute changes in VAT mass (r=−0.40, P=0.05). (d) The relationship between percent change in gluteal ZAG mRNA expression and percent change in fasting glucose (r=−0.54, P<0.01).

There were no significant differences in baseline age, weight, BMI, VO₂max, body composition, fasting glucose, 2-hour glucose, and fasting insulin between WL and AEX+WL (Table 1). Fifteen women had impaired glucose tolerance in the combined group (WL, n=8 and AEX+WL, n=7).

Table 1.

Effects of AEX+WL and WL interventions

	AEX+WL (n=17)		WL (n=16)
	Pre	Post	Pre	Post
Weight (kg)	87 ± 5	80 ± 5‡	96 ± 3	87 ± 3‡
Percent body fat	47 ± 2	44± 2‡	49 ± 1	47 ± 1‡
Fat mass (kg)	43 ± 4	36 ± 3‡	48 ± 2	41 ± 2‡
FFM (kg)	46 ± 2	45 ± 2	49 ± 1	46 ± 1‡
VAT (cm2)	145 ± 25	130 ± 22	164 ± 17	131 ± 10*
Sagittal diameter (mm)	25 ± 2	23 ± 1‡	27 ± 1	24.8 ± 1†
Subcutaneous abdominal fat (cm2)	437 ± 38	347 ± 46†	430 ± 45	420 ± 35
Muscle attenuation (HU)	37 ± 3	41 ± 3	36 ± 2	34 ± 2
RMR (kcal/day)	1468 ± 59	1368 ± 75	1704 ± 89	1536 ± 109
VO2max (1/min)	1.74 ± 0.20	1.95 ± 0.23†	1.59 ± 0.12	1.60 ± 0.10
Fasting glucose (mmol/l)	93 ± 2	89 ± 2	100 ± 3	91 ± 2†
Glucose@ 120 min (mmol/1)	130 ± 8	125 ± 7	139 ± 14	124 ± 8
Fasting insulin (pmol/1)	83 ± 12	63 ± 7*	102 ± 10	74 ± 7*

^*P<0.05;

^†P<0.01;

^‡P<0.001.

There was a ~8% decrease in body weight after AEX+WL and WL (both P<0.001, Table 1) with reductions in percent body fat (both P<0.001), fat mass (both P<0.001), and sagittal diameter (P<0.01). VAT decreased after WL (P<0.05) and AEX+WL (P=0.05). Subcutaneous abdominal fat decreased after AEX+WL (P<0.01). FFM decreased only with WL (P<0.001). VO₂max increased 10% after AEX+WL (P<0.01). RMR did not change. Fasting insulin decreased 24% after AEX+WL and 27% after WL (both P<0.05).

Abdominal ZAG mRNA increased 5% after AEX+WL (39.5±5.4 vs. 41.4±6.8) and 11% after WL (29.2±4.8 vs. 32.3±3.5) but were not statistically significant. After removal of four people whose VO₂max did not improve more than 5% after AEX+WL, abdominal ZAG mRNA increased 20% (36.8±6.9 vs. 44.0±8.7, n=12, P=0.15). Gluteal ZAG mRNA also did not change significantly with AEX+WL (46.6±7.2 vs. 44.0±6.7) and WL (28.8±2.9 vs. 24.3±2.7). In the total group, the absolute change in abdominal ZAG mRNA was negatively correlated with changes in VAT (r=−0.40, P=0.05, n=24, Figure 1c). In the total group, the absolute and percent changes in gluteal ZAG mRNA are negatively correlated with absolute (r=−0.45, P<0.05, n=25) as well as percent changes (r=−0.54, P=0.005, n=25, Figure 1d) in fasting plasma glucose, respectively. In AEX+WL group, the percent change in gluteal ZAG mRNA was negatively correlated with the percent change in fasting glucose (r=−0.79, P<0.0001).

DISCUSSION

We are the first to investigate the effects of AEX+WL and WL on adipose tissue ZAG expression. Our study demonstrates that low ZAG expression is associated with obesity, fitness, and hyperinsulinemia. Although increases in ZAG expression after WL or AEX+WL were not statistically significant, the inverse relationships between increased ZAG expression and loss of VAT and reduction in fasting glucose lend support to the idea that modifications to ZAG are potentially important to changes in abdominal obesity and glycemia.

ZAG promotes lipid metabolism through β-AR, stimulates uncoupler proteins in the mitochondria inner membrane to increase energy expenditure, increases number of skeletal muscle glucose transporters, stimulates AMP-activated protein kinase and acetyl-CoA carboxylase activity that promote fatty acid oxidation and glycerol release, and inhibits the activity of several key enzymes in the lipogenesis pathway^{1, 4, 8, 12}. Due to its anti-obesity effects, ZAG is now considered a target for future therapeutic approaches.

In our subjects, weight, fat mass, VAT, and sagittal diameter are negatively correlated with ZAG, which could support that ZAG promotes lipolysis and the utilization of body fat via the stimulation of important enzymes that promote fatty acid oxidation and inhibit lipogenesis. The strong negative correlation of ZAG with VAT suggests that ZAG protein may induce loss of fat around organs. The positive correlation of ZAG with muscle attenuation suggests that ZAG may promote fat oxidation in skeletal muscle and reduce intramuscular fat. The changes in ZAG with interventions suggest that those with a greater increase in ZAG lost more VAT and reduced hyperglycemia.

Previous murine studies have shown that ZAG may stimulate uncoupler proteins in the mitochondria inner membrane to increase energy expenditure through proton gradient dissipation¹³. This may suggest that an increase in ZAG could increase RMR. We observed a negative correlation between ZAG expression and RMR prior to the interventions. It is possible that an exercise intervention that increases RMR and muscle mass such as resistive training is necessary to increase ZAG^14–18. Given that WL can drive a reduction in RMR due to a loss of FFM, the reduction of body weight and loss of FFM with WL may be counterproductive to an increase in ZAG expression^{19, 20}.

There are several limitations to the present study including the small sample size, and representation of only postmenopausal women. Nevertheless, this study gave us new insight that ZAG expression can be upregulated modestly by WL and exercise training.

CONCLUSIONS

Our major novel finding is that ZAG expression is related to markers of abdominal obesity. We have shown for the first time that WL and AEX+WL have only a modest but not significant effect on altering ZAG levels. Future studies could be conducted to understand the role of ZAG expression on fat oxidation and it’s implication in both total and central obesity.

Abbreviations

ZAG

zinc-α2-glycoprotein

WL

weight loss

AEX+WL

aerobic exercise+weight loss

β-AR

β-adrenergic receptors

RMR

resting metabolic rate

FFM

fat-freemass

TLC

therapeutic lifestyle changes

OGTT

oral glucose tolerance test

VAT

visceral fat