Abstract

Aims: To assess the efficacy of rosiglitazone in blocking ethanol-induced hangover in rats. Methods: Rats injected with ethanol (4 g/kg body weight) were subjected to social interaction tests. Levels of aldehyde dehydrogenase 2 (ALD2), involved in an anti-hangover mechanism, were measured by semi-quantitative RT–PCR and western blot analysis. Results: Rosiglitazone caused an upregulation of mitochondrial ALD2, thus significantly detoxifying acetaldehyde. Conclusions: Rosiglitazone alleviated the symptoms of ethanol-induced hangover by inducing ALD2 expression; this result was reconfirmed by eliminating the effect of rosiglitazone by injecting cyanamide, an ALD2 inhibitor.

(Received 24 October 2005; first review notified 22 December 2005; in revised form 2 February 2006; accepted 3 February 2006)

INTRODUCTION

Although ethanol has been used to reduce stress, excessive ethanol consumption can have serious negative social consequences. Alcoholism is not a simple problem of heavy drinking. It substantially damages individuals' mental and physical health. Symptoms of ethanol abuse include vomiting, dizziness, thirst, headaches and muscle pain (Swift and Davidson, 1998). Chronic ethanol consumption leads to liver damage, pancreatitis, myocardial infarction and neuropathy and exacerbates the symptoms of tuberculosis and other diseases (Lieber and Leo, 1986, 1992).

Most ethanol consumed is metabolized by aldehyde/alcohol dehydrogenase (ALD), a microsomal ethanol-oxidizing system, and catalase, and by peroxidase in the liver; the remainder is emitted via the kidney and lung. Ethanol administration causes more damage to the liver than to any other organ (Lieber and DeCarli, 1968; Goodman and Gilman, 1975; Lieber, 1985; Lieber and Leo, 1986). Therefore, the efficient detoxification of ethanol is crucial for the preservation of proper liver function.

Acetaldehyde is known to be a primary metabolite of ethanol and is the main cause of hangovers (Kim, 1994). Lieber reported acetaldehyde to be more toxic to the body than ethanol itself (Lieber, 1973). It has been reported that acetaldehyde causes steatohepatitis, hepatic cirrhosis, downregulation of ALD2 expression and impaired vitamin activity via mitochondrial dysfunction (Helander, 1988). Therefore, problems of ethanol toxicity can potentially be solved by effectively decreasing the plasma acetaldehyde concentration.

Rosiglitazone, a class of insulin-sensitizing thiazolidinedione (TZD) drugs, is a peroxisome proliferator-activated receptor (PPAR)-γ agonist. It has been widely used for treating hyperglycaemia and improving insulin resistance (Patel et al., 1999). The metabolic effects of TZDs are mediated by receptor-dependent activation of the PPAR-γ–retinoid X receptor (RXR) complex and subsequent transcriptional activation of target genes (Yki-Jarvinen, 2004). Interestingly, TZDs may also be potent inhibitors of inflammation, in part through receptor-independent mechanisms (Chawla et al., 2001; Takata et al., 2002; Ruan et al., 2003; Mohanty et al., 2004). It is also thought that the anti-inflammatory effects of TZDs may reduce the risk of cardiovascular disease (van Wijk and Rabelink, 2004). There has been one report that the ALD2 promoter contains PPAR response elements (PPREs) (Pinaire et al., 1999). Thus, the expression of ALD2 could potentially be regulated by rosiglitazone.

The purpose of this study was to assess the effects of rosiglitazone on ALD2 expression and associated symptoms of ethanol-induced hangover.

MATERIALS AND METHODS

Animals and study design

Four-week-old male Sprague–Dawley (SD) rats were purchased from Samptako (O San, Korea) and were housed (two per cage) under 12 h light/dark cycles with ad libitum access to food and water for 24 weeks. Half of the animals (total n = 48) were given 5 mg/kg body weight of rosiglitazone orally for 2 weeks along with standard lab chow ad libitum. Because rosiglitazone was originally a metabolic disease drug, it needs time to induce ALD2 expression. This was confirmed in in vitro experiments using HepG2 cells (data not shown). After pre-treatment with rosiglitazone, the rats were given one 4 g/kg body weight dose of ethanol orally (AR, n = 14) or saline (A, n = 10) (Gauvin et al., 1992). Then 10 mg/kg body weight of cyanamide was administered to rosiglitazone plus ethanol-treated rats (ARI, n = 7) by intraperitoneal injection as an ALD2 inhibitor. Animals not treated with rosiglitazone (C, n = 24) were given standard lab chow ad libitum. The animals were given a 10 min social interaction test. Independent rat tissue samples under the same conditions were harvested according to time-dependent plasma alcohol concentration (Fig. 1).

Fig. 1.

Plasma ethanol concentration in experimental rats. Data are means ± SD (n = 4). *P < 0.05, ethanol-treated and ethanol plus rosiglitazone-treated group compared with the control group. The difference between the ethanol-treated and ethanol plus rosiglitazone-treated groups was not significant (N.S.). C, control (closed circle); A, ethanol 4 g/kg body weight treatment (open circle); AR, ethanol plus rosiglitazone 5 mg/kg body weight/day treatment (closed triangle).

Fig. 1.

Plasma ethanol concentration in experimental rats. Data are means ± SD (n = 4). *P < 0.05, ethanol-treated and ethanol plus rosiglitazone-treated group compared with the control group. The difference between the ethanol-treated and ethanol plus rosiglitazone-treated groups was not significant (N.S.). C, control (closed circle); A, ethanol 4 g/kg body weight treatment (open circle); AR, ethanol plus rosiglitazone 5 mg/kg body weight/day treatment (closed triangle).

Determination of plasma ethanol concentration

The plasma ethanol concentration was determined using a rapid analyzer (GL5, Analox Instruments, Lunenburg, MA). The plasma ethanol concentrations in three groups for 8 h are shown in Fig. 1. All experiments using rats were performed in compliance with the European Community Guide for the care and use of Laboratory Animals.

A social interaction test

Overall social activity was scored as the sum of social investigation, contact and play behaviour frequencies. Social investigation was defined as sniffing of any part of the partner's body. Frequency of contact behaviour was scored as the sum of crawling over and under the partner and social grooming. Play fighting was scored by the frequencies of the following behavioural acts and postures demonstrated by test subjects: pouncing or playful nape attack (the experimental subject lunges at the partner with its forepaws extended outward), following and chasing (the experimental animal rapidly pursues the partner) (Meaney and Stewart, 1981; Thor and Holloway, 1984; Vandershuren et al., 1997). The total number of crossovers (movements between compartments) demonstrated by each experimental subject was used as an index of general locomotor activity (Varlinskaya and Spear, 2002). The frequency of social investigation, contact behaviour and play fighting and the total number of crossovers were examined using separate between-group ANOVAs.

Immunoblot analysis

Immunoblot analysis was performed using crude extracts from liver tissue. Samples were resolved by 10% SDS–polyacrylamide gel electrophoresis, transferred to nitrocellulose membrane and blotted with appropriate primary antibodies. The membrane was incubated with peroxidase-conjugated secondary antibodies (1:5000; Santa Cruz Biotechnology, Santa Cruz, CA) and the bound antibody was visualized using a chemiluminescence reader (Intron, Sungnam, Korea) and X-ray film (Fuji Film, Uetake, Japan). The primary antibody, mouse anti-ALD2, was purchased as synthesized peptides from Peptron (1:1000; Daejeon, Korea).

Isolation of mitochondrial fraction

Tissue lysates were washed in buffer A (100 mM sucrose, 1 mM EGTA and 20 mM MOPS, pH 7.4) and resuspended in buffer B [buffer A plus 5% Percoll, 0.01% digitonin and a cocktail of protease inhibitors: 10 mM aprotinin, 10 mM pepstatin A, 10 mM leupeptin, 25 mM calpain inhibitor I and 1 mM phenylmethylsulfonyl fluoride (PMSF)]. After 15 min of incubation on ice, unbroken cells and nuclei were pelleted by centrifugation at 2500 g for 10 min. The supernatant was centrifuged at 15 000 g for 15 min to pellet mitochondria, which were resuspended in buffer C [300 mM sucrose, 1 mM EGTA, 20 mM MOPS (pH 7.4) and the aforementioned cocktail of protease inhibitors]. The supernatant was centrifuged further at 100 000 g for 1 h. The resulting supernatant and the pellet were designated as the cytosolic and microsomal fractions, respectively. The nuclear pellet was resuspended in 10 mM Tris–HCl (pH 7.5), 2.5 mM KCl and 2.5 mM MgCl2 and isolated after centrifugation at 90 000 g for 30 min through use of 2.1 M sucrose in 50 mM Tris–HCl (pH 7.5) and 5 mM MgCl2.

ALD2 activity assay

The mitochondrial fraction was used to assay the catalytic activity of ALD2. ALD2 catalytic activity was assayed by measuring the increase of absorbance at 340 nm due to the reduction of NAD+ to NADH following the oxidation of acetaldehyde to acetic acid. The assay method has been described previously (Bostian and Betts, 1978), but was adapted to a 96-well microplate reader using a final volume of 0.2 ml.

RESULTS

Ethanol-induced hangover causes suppressed social investigation in rats, as illustrated in Fig. 2A for our four groups of rats 18 h after ethanol challenge. In contrast to the hangover-associated suppression observed in the ethanol-treated group, rosiglitazone-treated rats showed recovery from hangover-associated social investigation suppression, but not completely. Rosiglitazone plus cyanamide-treated rats did not show a rosiglitazone effect. General locomotor activity (as indexed by the total number of crossovers between compartments) was significantly lower in ethanol-treated rats than in control and in ethanol plus rosiglitazone-treated rats. Decreased locomotor activity during acute ethanol withdrawal (hangover) has also been reported in adult animals tested on the elevated plus maze (Lal et al., 1991; File et al., 1993). Social investigation and contact behaviour tests showed the same patterns as the crossovers tests. But play fighting was not detected in any group (Fig. 2A). As expected, overall social activities were shown to vary among the groups (Fig. 2B).

Fig. 2.

Social investigation, contact behaviour, play fighting and locomotor activity (overall number of crossovers) during the test session, 18 h after an acute ethanol challenge (A). Overall social activity was scored as the sum of social investigation, contact, play fighting and crossover frequencies (B). Data for familiar and unfamiliar situations were similar (data not shown). Cyanamide-only treatment was also tested, but that group was not significantly different from the control group. Data are means ± SD. *P < 0.05, compared with the control group; +P < 0.05, compared with the ethanol treatment group; #P < 0.05, compared with the ethanol plus rosiglitazone treatment group. C, control; A, ethanol 4 g/kg body weight treatment; AR, ethanol plus rosiglitazone 5 mg/kg body weight/day treatment; ARI, ethanol plus rosiglitazone plus cyanamide 10 mg/kg body weight treatment.

Fig. 2.

Social investigation, contact behaviour, play fighting and locomotor activity (overall number of crossovers) during the test session, 18 h after an acute ethanol challenge (A). Overall social activity was scored as the sum of social investigation, contact, play fighting and crossover frequencies (B). Data for familiar and unfamiliar situations were similar (data not shown). Cyanamide-only treatment was also tested, but that group was not significantly different from the control group. Data are means ± SD. *P < 0.05, compared with the control group; +P < 0.05, compared with the ethanol treatment group; #P < 0.05, compared with the ethanol plus rosiglitazone treatment group. C, control; A, ethanol 4 g/kg body weight treatment; AR, ethanol plus rosiglitazone 5 mg/kg body weight/day treatment; ARI, ethanol plus rosiglitazone plus cyanamide 10 mg/kg body weight treatment.

Because we found behavioural changes in the social interaction test following rosiglitazone treatment, we decided to confirm the relationship of rosiglitazone treatment to ALD2 expression. To analyze aldehyde metabolism, expression of ALD2, which is known to catalyze aldehyde, was assessed by western blot analysis in the four groups. As shown in Fig. 3A, ALD2 expression was higher in rosiglitazone-treated rat livers than in those of control and ethanol-treated rats. But cyanamide influenced only ALD2 activity, not its expression (Fig. 3B).

Fig. 3.

ALD2 expression and activity in experimental rats. ALD2 expression was determined by western blot analysis. The expression of ALD2 was estimated by densitometric analysis of each protein band (A). ALD2 activity was measured (B). Cyanamide-only treatment was also tested, but that group was not significantly different from the control group. Data are means ± SD (n = 4). *P < 0.05, compared with the control group; +P < 0.05, compared with the ethanol treatment group; #P < 0.05, compared with the ethanol plus rosiglitazone treatment group. C, control; A, ethanol 4 g/kg body weight treatment; AR, ethanol plus rosiglitazone 5 mg/kg body weight/day treatment; ARI, ethanol plus rosiglitazone plus cyanamide 10 mg/kg body weight treatment.

Fig. 3.

ALD2 expression and activity in experimental rats. ALD2 expression was determined by western blot analysis. The expression of ALD2 was estimated by densitometric analysis of each protein band (A). ALD2 activity was measured (B). Cyanamide-only treatment was also tested, but that group was not significantly different from the control group. Data are means ± SD (n = 4). *P < 0.05, compared with the control group; +P < 0.05, compared with the ethanol treatment group; #P < 0.05, compared with the ethanol plus rosiglitazone treatment group. C, control; A, ethanol 4 g/kg body weight treatment; AR, ethanol plus rosiglitazone 5 mg/kg body weight/day treatment; ARI, ethanol plus rosiglitazone plus cyanamide 10 mg/kg body weight treatment.

DISCUSSION

Hangover-related anxiety can be indexed in laboratory animals by assessing the suppression of social interactions normally seen when two animals are placed together (social inhibition). Social interaction tests have been effectively applied to the investigation of anxiety-related behaviour in rats (File and Seth, 2003) including behaviours related to ethanol withdrawal (File et al., 1989, 1991, 1992). The social interaction test is a model of generalized anxiety (File, 1995) and has several advantages over other animal models of anxiety. It does not involve ecologically unusual conditions such as food and water deprivation or the use of electric shock. The social interaction test does not require extensive pre-training procedures and, hence, can be easily applied to the assessment of hangover-related anxiety in rats. In the conventional social interaction test, a pair of rats is placed in an experimental cage and the time that they spend in active social interaction is recorded. A summed measure of a variety of different elementary behavioural acts, including sniffing, grooming, chasing and wrestling, crawling under or over, nape attacks and biting, has traditionally been used for assessment of anxiogenic drug effects in this test situation. However, rats engage in distinct forms of interactive social behaviours that include social investigation, contact behaviour and play fighting. Each of these forms of social activity is influenced differently by drug treatment (Beatty et al., 1982; Soffie and Bronchart, 1988; Vandershuren et al., 1997), has a distinct ethological pattern (Meaney and Stewart, 1981; Panksepp et al., 1984; Thor and Holloway, 1984; Vandershuren et al., 1997) and hence may be mediated by different neural systems.

This study assessed the effects of acute ethanol hangover on different forms of social behaviour (social investigation, contact behaviour and play fighting) in control, ethanol-treated, ethanol plus rosiglitazone-treated and rosiglitazone-only-treated rats. The results of this study indicate that rosiglitazone reversed ethanol's depression of social activities and that it is able to eliminate acetaldehyde by regulating the expression of ALD2.

As expected, we confirmed more induced ALD2 expression in ethanol plus rosiglitazone-treated rats than in rats treated with ethanol only. Thus, we supposed that rosiglitazone enabled recovery from hangover in rats by effectively decreasing the plasma acetaldehyde concentration. The social activities of rosiglitazone-only-treated rats were similar to those of the controls (data not shown)

Unfortunately, we could not directly measure plasma acetaldehyde concentrations because of acetaldehyde's highly evaporative nature and the potential for measurement inaccuracy in our laboratory setting. But we used the second best option: we solved the problem by treating rats with cyanamide, an ALD2 inhibitor. Cyanamide eliminated the effects of rosiglitazone in the social interaction test (Fig. 2) as well as ALD2 activity (Fig. 3B) without changing the ALD2 expression level (Fig. 3).

Interestingly, the present study shows that rosiglitazone may be able to decrease plasma acetaldehyde concentration in a time-dependent manner without changing plasma ethanol concentrations (Fig. 1). This result shows that rosiglitazone can selectively eliminate the very toxic effects of acetaldehyde by inducing ALD2 expression.

In summary, our results show that rosiglitazone combats ethanol-induced hangover in rats. These results may open up a new clinical avenue for treating acute and severe ethanol-induced hangover. Further studies of the anti-hangover mechanism and ALD2 mechanism of rosiglitazone are necessary, and the cost of rosiglitazone needs to be reduced if it is to be used as a simple anti-hangover drug.

This study was supported by a grant from the Basic Research Program of the Korean Science and Engineering Foundation (R13-2002-054-01001-0, 2002).

REFERENCES

Beatty, W. W., Dodge, A. M., Dodge, L. J. et al. (
1982
) Psychomotor stimulants, social depriviation and play in juvenile rats.
Pharmacology, Biochemistry, and Behavior
 
16
,
417
–422.
Bostian, K. A. and Betts, G. F. (
1978
) Kinetics and reaction mechanism of potassium-activated aldehyde dehydrogenase from Saccharomyces cerevisiae.
Biochemical Journal
 
173
,
787
–798.
Chawla, Y. Barak, L., Nagy, L. et al. (
2001
) PPAR-γ dependent and independent effects on macrophage-gene expression in lipid metabolism and inflammation.
Nature Medicine
 
7
,
48
–52.
Ciardullo, A. V. (
2005
) Rosiglitazone improves insulin sensitivity, glucose tolerance and ambulatory blood pressure in subjects with impaired glucose tolerance: does it really work? If yes, is it a novelty?
Diabetic Medicine
 
22
,
666
–667.
File, S. E. (
1993
) The social interaction test of anxiety.
Neuroscience Protocol
 
10
,
1
–7.
File, S. E. (
1995
) Animal models of different anxiety states.
Advances in Biochemical Psychopharmacology
 
48
,
93
–113.
File, S. E. and Seth, P. (
2003
) A review of 25 years of the social interaction test.
European Journal of Pharmacology
 
463
,
35
–53.
File, S. E., Baldwin, H. A. and Hitchcott, P. K. (
1989
) Flumazenil but not nitrendipine reverses increased anxiety during ethanol withdrawal in the rat.
Psychopharmacology
 
98
,
262
–264.
File, S. E., Zharkovsky, A. and Hitchcott, P. K. (
1992
) Effects of nitrendipine, chlordiazepoxide, flumazenil and baclofen on the increased anxiety resulting from alcohol withdrawal.
Progress in Neuro-psychopharmacology and Biological Psychiatry
 
16
,
87
–93.
File, S. E., Zharkovsky, A. and Gulati, K. (
1991
) Effects of baclofen and nitrendipine on ethanol withdrawal responses in the rats.
Neuropharmacology
 
30
,
183
–190.
Freeman, B. A. and Crapo, J. D. (
1982
) Biology of disease: free radicals and tissue injury.
Laboratory Investigation
 
47
,
412
–426.
Gauvin, D. V., Youngblood, B. D. and Holloway, F. A. (
1992
) The discriminative stimulus properties of acute ethanol withdrawal (hangover) in rats.
Alcoholism, Clinical and Experimental Research
 
16
,
336
–341.
Goodman, L. S. and, Gilman, A. G. (
1975
) The Pharmacological Basis of Therapeutics, 5th edn. Macmillan, New York.
Helander, A. and Tottmar, O. (
1988
) Effect of acute ethanol administration on human blood aldehyde dehydrogenase activity.
Alcoholism, Clinical and Experimental Research
 
12
,
643
–643.
Hwang, J., Kleinhenz, D. J., Lassegue, B. et al. (
2005
) Peroxisome proliferator-activated receptor-ligands regulate endothelial membrane.
American Journal of Physiology. Cell Physiology
 
288
,
899
–905.
Jeong, C. S. and Jung, K. H. (
2002
) Protective effects of Angelica tenuissima Nakai on hepatoxicity by carbon tetrachloride in rats.
Journal of Applied Pharmacology
 
10
,
211
–217.
Jürgen, M. L., Linda, B. M., Tracey, A. S. et al. (
1995
) An antidiabetic thiazolidinedione is a high affinity ligand for peroxisome proliferator-activated receptor.
Journal of Biological Chemistry
 
270
,
12953
–12956.
Kim, K. W., Yang, J. S., Lee, J. S. et al. (
1994
) Activity of alcohol dehydrogenase and ethanol, acetaldehyde levels in normal adult blood.
The Korean Journal of Food and Nutrition
 
4
,
240
–247.
Lal, H., Prather, P. L. and Rezazadeh, S. M. (
1991
) Anxiogenic behavior in rats during acute and protracted ethanol withdrawal: reversal by buspirone.
Alcohol
 
8
,
467
–471.
Lieber, C. S. (
1973
) Liver adaptation and injury in alcoholism.
New England Journal of Medicine
 
288
,
356
–362.
Lieber, C. S. (
1985
) Alcohol and liver: metabolism and ethanol, metabolic effects and pathogenesis of injury.
Acta Medica Scandinavica Supplementum
 
703
,
11
–55.
Lieber, C. S. and DeCarli, L. M. (
1968
) Ethanol oxidation by hepatic microsomes: adaptive increase after ethanol feeding.
Science
 
162
,
917
–918.
Lieber, C. S. and Leo, M. A. (
1986
) Progress in liver diseases. In Interaction of alcohol and nutritional factors with hepatic fibrosis, Popper, H. and Schaffner, F. eds, p.
253
. Grune and Stration, New York.
Lieber, C. S. and Leo, M. A. (
1992
) Alcohol and liver. In Medical and Nutritional Complications of Alcoholism: Mechanism and Management, p.
185
. Plenum Medical Book Co., New York.
Lieber, C. S., Garro, A., Leo, M. A. et al. (
1986
) Alcohol and cancer.
Hepatology
 
6
,
1005
–1019.
Meaney, M. J. and Stewart, J. (
1981
) A descriptive study of social development in the rat (Rattus norvegicus).
Animal Behaviour
 ,
29
,
34
–35.
Mohanty, P., Aljada, A., Ghanim, H. et al. (
2004
) Evidence for a potent anti-inflammatory effect of rosiglitazone.
Journal of Clinical Endocrinology and Metabolism
 
89
,
2728
–2735.
Mukherjee, R., Davies, P. J., Crombie, D. L. et al. (
1997
) Sensitization of diabetic and obese rats to insulin by retinoid X receptor agonists.
Nature
 ,
386
,
407
–410.
Neuschwander-Tetri, B. A., Brunt, E. M., Wehmeier, K. R et al. (
2003
) Improved nonalcoholic steatohepatitis after 48 weeks of treatment with the PPAR-gamma ligand rosiglitazone.
Hepatology
 
38
,
1008
–1017.
Panksepp, J., Normansell, L., Siviy, S. et al. (
1984
) Casomorphins reduce separation distress in chicks.
Peptides
 
5
,
829
–831.
Patel, J., Weston, W. and Hemyari, P. (
1999
) Rosiglitazone (RSG) decreases insulin resistance (IR) and improves beta-cell function (BCF) in patients with type 2 diabetes mellitus (Abstract).
Program and Abstracts: Endocrine Society 81st Annual Meeting
 
420
,
3
–153.
Pinaire, J., Chou, W. Y., Stewart, M. et al. (
1999
) Activity of the human aldehyde dehydrogenase 2 promoter is influenced by the balance between activation by hepatocyte nuclear factor 4 and repression by perosixome proliferator activated receptor delta, chicken ovalbumin upstream promoter-transcription factor, and apolipoprotein regulatory protein-1.
Advances in Experimental Medicine and Biology
 
463
,
115
–121.
Ruan, H., Pownall, H. J. and Lodish, H. F. (
2003
) Troglitazone antagonizes tumor necrosis factor-alpha-induced reprogramming of adipocyte gene expression by inhibiting the transcriptional regulatory functions of NF-kappa B.
Journal of Biological Chemistry
 
278
,
28181
–28192.
Simon, R. H., Scoggin, C. M. and Patterson, D. (
1981
) Hydrogen peroxide causes the fatal injury to human fibroblasts exposed to oxygen radicals.
Journal of Biological Chemistry
 
266
,
7181
–7186.
Soffie, M. and Bronchart, M. (
1998
) Age-related scopolamine effects on social and individual behavior of rats.
Psychopharmacology
 
95
,
344
–350.
Swift, R. and Davidson, D. (
1998
) Alcohol hangover, mechanism and mediators.
Alcohol Health Research World
 
22
,
54
–60.
Takata, Y., Kitami, Y. and Yang, Z. H. (
2002
) Vascular inflammation is negatively autoregulated by interaction between CCAAT/enhancer-binding protein-delta and peroxisome proliferator-activated receptor-gamma.
Circulation Research
 
91
,
427
–433.
Thor, D. H. and Holloway, W. R. Jr. (
1983
) Play-solicitation behavior in juvenile male and female rats.
Animal Learning and Behavior
 
11
,
173
–178.
van Wijk, J. P. and Rabelink, T. J. (
2004
) PPAR-gamma agonists: shifting attention from the belly to the heart?
Arteriosclerosis, Thrombosis, and Vascular Biology
 
24
,
798
–800.
Vanderschuren, L. J. Niensink, R. J. and van Ree, J. M. (
1997
) The neurobiology of social play behavior in rats.
Neuroscience and Biobehavioral Reviews
 
21
,
309
–326.
Yki-Jarvinen, H. (
2004
) Thiazolidinediones.
New England Journal of Medicine
 
351
,
1106
–1118.

Author notes

1The Brain Korea 21 Project for Medical Science

2Division of Endocrinology and Metabolism, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea

3Department of Internal Medicine, Pochon CHA University College of Medicine, Sungnam, Korea