Abstract
Familial partial lipodystrophy (FPLD) 3 is a rare genetic disorder caused by peroxisome proliferator-activated receptor γ gene (PPARG) mutations. Most cases have been reported in Western patients. Here, we describe a first pedigree of FPLD 3 in Japanese. The proband was a 51-year-old woman. She was diagnosed with fatty liver at age 32 years, dyslipidemia at age 37 years, and diabetes mellitus at age 41 years. Her body mass index was 18.5 kg/m2, and body fat percentage was 19.2%. On physical examination, she had less subcutaneous fat in the upper limbs than in other sites. On magnetic resonance imaging, atrophy of subcutaneous adipose tissue was seen in the upper limbs and lower legs. Fasting serum C-peptide immunoreactivity was high (3.4 ng/mL), and the plasma glucose disappearance rate was low (2.07%/min) on an insulin tolerance test, both suggesting apparent insulin resistance. The serum total adiponectin level was low (2.3 μg/mL). Mild fatty liver was seen on abdominal computed tomography. On genetic analysis, a P495L mutation in PPARG was identified. The same mutation was also seen in her father, who had non-obese diabetes mellitus, and FPLD 3 was diagnosed. Modest increases in body fat and serum total adiponectin were seen with pioglitazone treatment. Attention should be paid to avoid overlooking lipodystrophy syndromes even in non-obese diabetic patients if they show features of insulin resistance.
LIPODYSTROPHY SYNDROMES are heterogeneous diseases in which adipose tissue atrophies irrespective of the energy balance. Together with the progressive decrease and disappearance of adipose tissue, there is a high frequency of comorbidity with various metabolic disorders, such as insulin resistance, diabetes mellitus, hypertriglyceridemia, and fatty liver. Lipodystrophy syndromes are divided into 2 types that occur from congenital gene mutations or from acquired etiologies, and they are also classified as generalized, partial, or localized depending on the distribution of fat atrophy [1, 2].
Familial partial lipodystrophy (FPLD) is known to consist of FPLD types 1–7, and FPLD 3 is a rare autosomal dominant genetic disorder caused by mutations of peroxisome proliferator-activated receptor γ gene (PPARG) [3]. In cases of FPLD3, the amount and distribution of adipose tissue are almost normal in childhood. After puberty, atrophy of subcutaneous adipose tissue, mainly in the limbs, is seen, but subcutaneous and visceral adipose tissue in the abdomen are usually maintained. Ectopic fat accumulates in the liver and skeletal muscle, leading to insulin resistance [4]. The lipodystrophy is considered due to decreased activity of peroxisome proliferator-activated receptor γ (PPARγ), encoded by PPARG. PPARγ is a member of the nuclear receptor superfamily of ligand-dependent transcription factors which is required for adipocyte differentiation, regulation of insulin sensitivity, lipogenesis, and adipocyte survival and function [5]. So far, approximately 60 cases of FPLD 3 have been reported, most of those in Western patients [6]. Most Western patients displayed obesity, early-onset diabetes mellitus and severe insulin resistance. Very recently, a Chinese patient with FPLD 3 due to a de novo mutation in PPARG was reported [7]. The patient was lean (BMI 17.5 kg/m2), but had severe insulin resistance. Here, we report a first pedigree of FPLD 3 in Japanese.
The proband was a middle-aged woman who had been treated for type 2 diabetes mellitus for 10 years. She was lean, but had fatty liver and insulin resistance. There was lipodystrophy in the upper limbs and lower legs. Genetic analysis showed a P495L mutation in PPARG in the proband and her father, and a diagnosis of FPLD 3 was made. Since degree of obesity is often mild in East Asian patients with diabetes mellitus, the FPLD phenotype could be mild and difficult to recognize.
Case Presentation
The patient was a 51-year-old woman. She was diagnosed with fatty liver at age 32 years and with type IV dyslipidemia at age 37 years (total cholesterol 195 mg/dL, HDL cholesterol 37 mg/dL and triglycerides 192 mg/dL). At age 41 years, her fasting glucose was 128 mg/dL, and HbA1c was 6.9%, and she was diagnosed with diabetes mellitus. She started to take sitagliptin and voglibose at age 46 years. At age 49 years, her HbA1c worsened to around 9%, and at age 50 years, she was hospitalized for the first time in our department. Her endogenous insulin secretion was maintained, and with the administration of vildagliptin, metformin, and voglibose, her glycemic control improved. Five months after discharge, her HbA1c had again worsened to 10.2%, and she was hospitalized for the second time at age 51 years. The reason for worsening of glycemic control seemed mostly due to binge eating.
Her medical history included depression, panic disorder, binge eating disorder, and bronchial asthma. She had no history of pregnancy or childbirth. She had irregular menstruation in her 20s and 30s, but her menstruation was mostly regular thereafter. Her birth weight was 2,100 g (length 43.5 cm) at full term, which was small for dates. After reaching adulthood, her weight was around 40 kg. At the time she was diagnosed with diabetes (age 41 years) she weighed 42.3 kg.
Her family pedigree is shown in Fig. 1. Her father and elder brother had diabetes mellitus. Her father developed diabetes mellitus at age 77 years and was non-obese. He had a cerebral infarction at age 80 years. Her elder brother developed diabetes mellitus at age 47 years and was obese, and he was treated with oral hypoglycemic drugs.
On admission, the patient’s height was 150.5 cm, her weight was 41.8 kg (body mass index [BMI] 18.5 kg/m2), and her waist circumference was 74 cm. Her blood pressure was 124/86 mmHg, and pulse rate was 96 beats/min and regular. She had a somewhat distended abdomen and less subcutaneous fat in her upper limbs than in other areas. There were no abnormalities of deep tendon reflexes or vibratory sensation in the upper or lower limbs. Diabetic retinopathy was not seen. Her body fat percentage measured by bioelectrical impedance analysis (BIA) (InBody 770, In Body Japan Inc., Tokyo, Japan) was 19.2%.
Her laboratory findings on admission are shown in Table 1. Her fasting glucose was 185 mg/dL, and HbA1c was 10.4%. Triglycerides were slightly high (165 mg/dL). She was negative for islet autoantibodies. Her endogenous insulin secretion was maintained, with urinary C-peptide immunoreactivity (CPR) of 153.7 μg/day and fasting serum CPR of 3.4 ng/mL. Fasting immunoreactive insulin was 10.1 μU/mL, and homeostasis model assessment of insulin resistance (HOMA-IR) was as high as 4.61. The serum leptin level was 8.2 ng/mL, within the normal reference range (2.5–21.8 ng/mL), but the serum total adiponectin level was low (2.3 μg/mL), both measured with radioimmunoassay and latex agglutination turbidimetry at BML, Inc., Tokyo, Japan. Insulin sensitivity was evaluated with a short insulin tolerance test [8], and a low plasma glucose disappearance rate (Kitt) of 2.07%/min was seen, indicating apparent insulin resistance (normal reference range: 3.50 ± 0.75%/min) [9].
Table 1
Laboratory findings on admission
| Complete blood cell counts |
Blood chemistry |
| WBC |
7,900/μL |
TP |
6.9 g/dL |
Fasting glucose |
185 mg/dL |
[<110] |
| RBC |
485 × 104/μL |
Alb |
4.2 g/dL |
HbA1c |
10.4% |
[4.6–6.2] |
| Hb |
15.2 g/dL |
BUN |
12.3 mg/dL |
T-chol |
157 mg/dL |
|
| Ht |
44.9% |
Cre |
0.55 mg/dL |
LDL-chol |
81 mg/dL |
|
| Plt |
10.4 × 104/μL |
UA |
3.6 mg/dL |
HDL-chol |
43 mg/dL |
|
|
|
Na |
139 mEq/L |
TG |
165 mg/dL |
|
| Urinalysis |
Cl |
100 mEq/L |
|
|
|
| Protein |
(–) |
K |
3.8 mEq/L |
Endocrinological examinations |
| Glucose |
(3+) |
T-Bil |
0.9 mg/dL |
IRI |
10.1 μU/mL |
[2.2–12.4] |
| Occult blood |
(–) |
AST |
25 U/L |
Fasting CPR |
3.4 ng/mL |
[0.8–2.5] |
| Ketone |
(–) |
ALT |
37 U/L |
Postprandial (2-h) CPR |
4.3 ng/mL |
|
|
|
LDH |
100 U/L |
GAD antibody |
<5.0 U/mL |
[<5.0] |
| Urinary storage |
ALP |
173 U/L |
IA-2 antibody |
<0.6 U/mL |
[<0.6] |
| CPR |
153.7 μg/day |
γ-GTP |
23 U/L |
Leptin |
8.2 ng/mL |
[2.5–21.8] |
| Albumin |
1.3 mg/day |
CK |
96 U/L |
Total adiponectin |
2.3 μg/mL |
[>4] |
CPR: C-peptide immunoreactivity, IRI: Immunoreactive insulin, GAD: Glutamic acid decarboxylase, IA-2: Insulinoma-associated protein-2, [ ]: normal reference ranges
On abdominal computed tomography (CT), mild fatty liver was seen (Fig. 2a). No increase in intraperitoneal fat was seen (Fig. 2b). On whole-body magnetic resonance imaging (MRI), subcutaneous adipose tissue was atrophied in the upper limbs and lower legs and was maintained in the head, orbits, abdominal cavity, lower abdomen subcutaneous tissue, buttocks, and thigh subcutaneous tissue (Fig. 3).
After admission, her glycemic control improved, but pioglitazone 7.5 mg/day was added, considering the presence of insulin resistance and hypoadiponectinemia.
The patient was lean, but she had fatty liver and diabetes mellitus with apparent insulin resistance. Decreases in subcutaneous adipose tissue were seen particularly in the upper limbs and lower legs, and partial lipodystrophy was suspected. Her father was non-obese with elderly onset diabetes, and the possibility of FPLD was considered.
Genetic Analysis
A genetic analysis was conducted in the patient and her parents. In brief, genomic DNA was extracted from peripheral venous blood using the QIAmp DNA blood Midi Kit (QIAGEN, Valencia, CA, USA). Targeted sequencing was performed using the TruSight One panel (targeting 4813 OMIM genes, Illumina, San Diego, CA, USA) and the MiSeq Sequencer (Illumina). The mutations detected through panel sequencing were confirmed by Sanger sequencing. A P495L mutation in PPARG was identified in the patient and her father (Fig. 4), and a diagnosis of FPLD 3 was made. Her elder brother did not consent to a genetic analysis. The genetic analysis was approved by the Human Genome, Gene Analysis Research Ethics Committee of Showa University (approval number 293), and written, informed consent was obtained from the patient and her parents.
Course after Hospital Discharge
Three months after discharge, her fasting glucose had improved to 168 mg/dL and HbA1c to 7.8%, but at 5 months, her HbA1c had worsened again to 9.6%. The dose of pioglitazone was increased to 15 mg/day from 8 months after discharge, and at 12 months, a moderate improvement of glucose control was seen, with a fasting glucose of 158 mg/dL and HbA1c of 8.9% (weight 42.4 kg).
After pioglitazone was started, the serum leptin level fluctuated within the normal reference range, but the serum total adiponectin level tended to increase. Pioglitazone had no effect at a dose of 7.5 mg/day, but it apparently decreased HOMA-IR at a dose of 15 mg/day (Table 2a). On body composition measurements with BIA, there were no marked changes in muscle mass or the skeletal muscle mass index, but her body fat mass and body fat percentage tended to increase. Specifically, the increase in the body fat mass in the trunk was conspicuous compared with the limbs (Table 2b).
Table 2
Course from before to after the start of pioglitazone (Pio) of (a) blood leptin, total adiponectin, insulin level, and HOMA-IR and (b) body composition (BIA)
(a)
|
On admission
at age 50 |
On admission
at age 51 |
6 mo. after
start of Pio |
12 mo. after
start of Pio |
| Leptin (ng/mL) [2.5–21.8] |
8.4 |
8.2 |
14.0 |
9.2 |
| Total adiponectin (μg/mL) [>4] |
2.0 |
2.3 |
3.2 |
3.9 |
| Immunoreactive insulin (μU/mL) |
6.5 |
10.1 |
9.6 |
4.1 |
| HOMA-IR |
2.74 |
4.61 |
4.72 |
1.60 |
(b)
|
On admission
at age 50 |
On admission
at age 51 |
6 mo. after
start of Pio |
12 mo. after
start of Pio |
| Height (cm) |
150 |
|
|
|
| Weight (kg) |
40.6 |
40.7 |
42.3 |
41.9 |
| Body mass index (kg/m2) |
18.0 |
18.1 |
18.8 |
18.6 |
| Muscle mass (kg) |
31.0 |
30.9 |
31.7 |
30.9 |
| Skeletal muscle mass index (kg/m2) |
5.5 |
5.2 |
5.5 |
5.4 |
| Body fat mass (kg) |
7.7 |
7.8 |
8.6 |
9.0 |
| Body fat percentage (%) |
19.0 |
19.2 |
20.4 |
21.4 |
| Fat mass by site (kg) |
Right arm |
0.5 |
0.5 |
0.5 |
0.6 |
| Left arm |
0.5 |
0.5 |
0.6 |
0.6 |
| Trunk |
3.0 |
3.2 |
3.6 |
3.8 |
| Right leg |
1.4 |
1.4 |
1.5 |
1.6 |
| Left leg |
1.4 |
1.4 |
1.5 |
1.6 |
HOMA-IR: homeostasis model assessment of insulin resistance, BIA: bioelectrical impedance analysis, skeletal muscle mass index is calculated as the appendicular skeletal muscle mass (kg)/(height in m)2, [ ]: normal reference ranges
Sixteen months after discharge, visceral fat area remained unchanged (from 37.4 cm2 to 39.5 cm2), whereas subcutaneous fat area increased (from 56.2 cm2 to 77.8 cm2) on the CT scan at the umbilical level.
Discussion
This patient was lean, but she had partial lipodystrophy and apparent insulin resistance. A P495L mutation in PPARG was identified in the patient and her father, and a diagnosis of FPLD 3 was made. To the best of our knowledge, this is the first report of FPLD 3 in a Japanese patient.
The phenotype of FPLD 3 is more pronounced in women than in men, and in women, symptoms such as hirsutism and dysmenorrhea, which are the characteristics of polycystic ovarian syndrome, are often seen [6, 10]. Regarding the reasons why the features of FPLD 3 are more common in premenopausal women than in men, the possibility has been suggested that estrogen receptor β impairs PPARγ transcriptional activity and inhibits adipocyte differentiation [11]. In addition, fat accumulation in subcutaneous adipose tissue is more predominant in women than in men, and therefore, atrophy of subcutaneous adipose tissue may result in severe metabolic phenotypes in women [6]. In the present family, the proband developed diabetes mellitus at age 41 years, whereas her father, who had the same mutation, developed diabetes mellitus at a much older age.
Since the PPARG P495L mutation is located in a ligand binding domain, binding of ligands could be impaired. The mutation also decreases PPARγ activity with a dominant-negative effect, and these disturbances are involved in fat atrophy [3]. Since the DNA binding domain is intact, the binding affinity for the target gene promoter region is thought to be the same as in the wild type. In cases of P495L mutation reported by Barroso et al., diabetes mellitus and hypertension occurred at a young age, but there was no fat atrophy or abnormal fat distribution [3]. In other cases of P495L mutation reported by Savage et al., the body fat percentage was lower than predicted from the BMI, and there was remarkably little subcutaneous fat in the limbs and buttocks on MRI [12]. In other cases of P495L mutation reported by Tan et al., plasma non-esterified fatty acid levels were high due to excessive lipolysis of triglyceride-rich lipoproteins [13]. In all of these reports, moderate to severe insulin resistance was seen; fasting insulin levels were as high as 85–325 pM (approximately 12–47 μU/mL) [3, 12, 13]. In the present case, the insulin resistance, diabetes mellitus, and dyslipidemia were mild compared with previous reports, and hypertension was not seen. In cases of P495L mutation in Western patients [3, 12, 13], the BMI was 24.2–25.9 kg/m2, higher than in the present case. This higher degree of obesity may result in severe insulin resistance and the earlier onset of diabetes mellitus, dyslipidemia, and hypertension in the Western patients.
Nazare et al. evaluated the abdominal fat distribution on CT scans of White, African and Caribbean Black, Hispanic, Southeast Asian (Malaysian, Vietnamese, and others), and East Asian (Chinese, Korean, Japanese, and others) individuals. They reported that, whereas visceral adipose tissue area and subcutaneous adipose tissue area were the smallest in East Asians for both men and women, the visceral adipose tissue/subcutaneous adipose tissue area ratio was the highest in East Asians in men, and in East Asians and Southeast Asians in women [14]. An interpretation of this finding is that subcutaneous fat in East Asians has low surplus energy storage capacity and that the portion that cannot be stored in subcutaneous fat is stored in visceral fat ectopically. The reasons for this are thought to involve environmental and lifestyle factors, along with differences in genetic backgrounds [15]. In the present patient, BMI was very low, and although mild fatty liver was seen, there was no increase in visceral fat. Because of these characteristics, the FPLD phenotype was considered to be inconspicuous.
Decreased blood leptin levels are involved in the pathogenesis of lipodystrophy syndromes, and leptin replacement therapy is effective for lipodystrophy where blood leptin levels have decreased below the reference range [2]. In the present patient, a decreased blood leptin level was not seen, and body fat mass was thought to have been maintained to some degree, since blood leptin levels are strongly associated with body fat mass per se [16]. At the same time, even though the present patient was a non-obese woman, the decrease in the blood adiponectin level was marked. Blood adiponectin levels and its gene expression are known to decrease when adipocytes enlarge, which is usually seen in obesity [17]. It might be possible that the size of adipocytes was enlarged in non-atrophied adipose tissue of the present lean patient to accommodate surplus energy. Since the insulin-sensitizing adiponectin level was low, she was newly treated with a thiazolidinedione.
In the cases with reported by Savage et al., in which the patient had the same P495L mutation as the present patient, rosiglitazone 8 mg/day was administered for 6 months, after which total body fat mass increased by 3.5 kg, and the fat mass increased slightly in the limbs and buttocks, where body fat had decreased [12]. In the present patient as well, there was a modest increase in body fat mass after the start of pioglitazone. By body site, the increase in the body fat mass, especially in subcutaneous fat mass in the trunk, was marked compared with the limbs, where the decrease in adipose tissue was conspicuous. Thiazolidinediones were reported not to affect adipose tissue in atrophied sites in a patient with FPLD 2 from an LMNA mutation, while inducing increases in adipose tissue in non-atrophied sites [18]. Thiazolidinediones may exert an effect on sites where fat atrophy is mild. In the above cases of Savage et al., it was reported that insulin sensitivity and HbA1c were both normalized, and blood leptin and adiponectin levels increased (from 0.7 to 1.9 μg/L and 0.25 to 0.5 units/mL) [12]. In the present patient, the blood adiponectin level, while low, increased after the start of pioglitazone as expected [19], and HOMA-IR tended to decrease, suggesting an improvement in insulin resistance. Since PPARγ transcriptional activity and the response to a thiazolidinedione were blunted in in vitro adipocytes with P495L mutation [3], an increased dose of pioglitazone might have increased blood adiponectin levels in the present case. Despite increased subcutaneous fat mass, blood leptin levels remained essentially unchanged. This may be explained by the finding that leptin gene expression is negatively regulated by thiazolidinediones [20]. A larger number of cases will be needed to examine the effectiveness of thiazolidinediones in FPLD 3.
In conclusion, though this patient was lean, she had diabetes mellitus with apparent insulin resistance. A P495L mutation in PPARG was identified on genetic analysis, and she was diagnosed with FPLD 3. The phenotype of PPARG mutations is thought to be diverse, and this patient had a markedly decreased blood adiponectin level without an obvious decrease in the blood leptin level. Her diabetes mellitus, hypertension, and dyslipidemia were mild compared with previous reports. Particularly in Japan and other East Asian countries, the degree of obesity is often mild in patients with diabetes mellitus. In non-obese patients with diabetes mellitus, the FPLD phenotype could be mild. Attention should be paid to avoid overlooking lipodystrophy syndromes even in non-obese diabetic patients if they display features of insulin resistance.
Disclosure
None of the authors has any potential conflicts of interest associated with this research.
References
- 1 Brown RJ, Araujo-Vilar D, Cheung PT, Dunger D, Garg A, et al. (2016) The diagnosis and management of lipodystrophy syndromes: a multi-society practice guideline. J Clin Endocrinol Metab 101: 4500–4511.
- 2 Tanaka T, Kusakabe T, Ebihara K, Aizawa-Abe M, Aotani D, et al. (2021) Practice guideline for lipodystrophy syndromes-clinically important diseases of the Japan Endocrine Society (JES). Endocr J 68: 1027–1042.
- 3 Barroso I, Gurnell M, Crowley VE, Agostini M, Schwabe JW, et al. (1999) Dominant negative mutations in human PPAR gamma associated with severe insulin resistance, diabetes mellitus and hypertension. Nature 402: 880–883.
- 4 Vigouroux C, Caron-Debarle M, Le Dour C, Magré J, Capeau J (2011) Molecular mechanisms of human lipodystrophies: from adipocyte lipid droplet to oxidative stress and lipotoxicity. Int J Biochem Cell Biol 43: 862–876.
- 5 Lefterova MI, Haakonsson AK, Lazar MA, Mandrup S (2014) PPARγ and the global map of adipogenesis and beyond. Trends Endocrinol Metab 25: 293–302.
- 6 Broekema MF, Savage DB, Monajemi H, Kalkhoven E (2019) Gene-gene and gene-environment interactions in lipodystrophy: lessons learned from natural PPARγ mutants. Biochim Biophys Acta Mol Cell Biol Lipids 1864: 715–732.
- 7 Chen X, Ma Z, Chen P, Song X, Li W, et al. (2022) Case report: a new peroxisome proliferator-activated receptor gamma mutation causes familial partial lipodystrophy type 3 in a Chinese patient. Front Endocrinol (Lausanne) 13: 830708.
- 8 Bonora E, Moghetti P, Zancanaro C, Cigolini M, Querena M, et al. (1989) Estimates of in vivo insulin action in man: comparison of insulin tolerance tests with euglycemic and hyperglycemic glucose clamp studies. J Clin Endocrinol Metab 68: 374–378.
- 9 Park SW, Yun YS, Ahn CW, Nam JH, Kwon SH, et al. (1998) Short insulin tolerance test (SITT) for the determination of in vivo insulin sensitivity-a comparison with euglycemic clamp test. J Korean Diabetes Assoc 22: 199–208.
- 10 Lau E, Carvalho D, Oliveira J, Fernandes S, Freitas P (2015) Familial partial lipodystrophy type 3: a new mutation on the PPARG gene. Hormones (Athens) 14: 317–320.
- 11 Foryst-Ludwig A, Clemenz M, Hohmann S, Hartge M, Sprang C, et al. (2008) Metabolic actions of estrogen receptor beta (ER beta) are mediated by a negative cross-talk with PPAR gamma. PLoS Genet 4: e1000108.
- 12 Savage DB, Tan GD, Acerini CL, Jebb SA, Agostini M, et al. (2003) Human metabolic syndrome resulting from dominant-negative mutations in the nuclear receptor peroxisome proliferator-activated receptor-gamma. Diabetes 52: 910–917.
- 13 Tan GD, Savage DB, Fielding BA, Collins J, Hodson L, et al. (2008) Fatty acid metabolism in patients with PPAR gamma mutations. J Clin Endocrinol Metab 93: 4462–4470.
- 14 Nazare JA, Smith JD, Borel AL, Haffner SM, Balkau B, et al. (2012) Ethnic influences on the relations between abdominal subcutaneous and visceral adiposity, liver fat, and cardiometabolic risk profile: the international study of prediction of intra-abdominal adiposity and its relationship with cardiometabolic risk/intra-abdominal adiposity. Am J Clin Nutr 96: 714–726.
- 15 Williams R, Periasamy M (2020) Genetic and environmental factors contributing to visceral adiposity in Asian populations. Endocrinol Metab (Seoul) 35: 681–695.
- 16 Nagasaka S, Ishikawa S, Nakamura T, Kawakami A, Rokkaku K, et al. (1998) Association of endogenous insulin secretion and mode of therapy with body fat and serum leptin levels in diabetic subjects. Metabolism 47: 1391–1396.
- 17 Liu YM, Lacorte JM, Viguerie N, Poitou C, Pelloux V, et al. (2003) Adiponectin gene expression in subcutaneous adipose tissue of obese women in response to short-term very low calorie diet and refeeding. J Clin Endocrinol Metab 88: 5881–5886.
- 18 Simha V, Rao S, Garg A (2008) Prolonged thiazolidinedione therapy does not reverse fat loss in patients with familial partial lipodystrophy, Dunnigan variety. Diabetes Obes Metab 10: 1275–1276.
- 19 Kadowaki T, Yamauchi T, Kubota N, Hara K, Ueki K, et al. (2006) Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest 116: 1784–1792.
- 20 Okuno A, Tamemoto H, Tobe K, Ueki K, Mori Y, et al. (1998) Troglitazone increases the number of small adipocytes without the change of white adipose tissue mass in obese Zucker rats. J Clin Invest 101: 1354–1361.
