Secondary Logo

Journal Logo

Increased hepatic peroxisome proliferator-activated receptor coactivator-lα expression precedes the development of insulin resistance in offspring of rats from severe hyperglycemic mothers

MA, Jing-mei; ZENG, Chan-juan; ZHANG, Li; SHOU, Chong; YANG, Hui-xia

doi: 10.3760/cma.j.issn.0366-6999.2012.07.007
Original article

Background Prenatal hyperglycaemia may increase metabolic syndrome susceptibility of the offspring. An underlying component of the development of these morbidities is hepatic gluconeogenic molecular dysfunction. We hypothesized that maternal hyperglycaemia will influence her offsprings hepatic peroxisome proliferator-activated receptor coactivator-1α (PGC-1α) expression, a key regulator of glucose production in hepatocytes.

Method We established maternal hyperglycaemia by streptozotocin injection to induce the maternal hyperglycaemic Wistar rat model. Offspring from the severe hyperglycemia group (SDO) and control group (CO) were monitored until 180 days after birth. Blood pressure, lipid metabolism indicators and insulin resistance (IR) were measured. Hepatic PGC-1α expression was analyzed by reverse transcription polymerase chain reaction and Western blotting. mRNA expression of two key enzymes in gluconeogenesis, glucose-6-phospha-tase (G-6-Pase) and phosphoenolpyruvate carboxykinase (PEPCK), were analyzed and compared.

Results In the SDO group, PGC-1α expression at protein and mRNA levels were increased, so were expression of G-6-Pase and PEPCK (P <0.05). The above effects were seen prior to the onset of IR.

Conclusion The hepatic gluconeogenic molecular dysfunction may contribute to the metabolic morbidities experienced by this population.

Department of Obstetrics and Gynecology, Peking University First Hospital, Beijing 100034, China (Ma JM, Zhang L, Shou C and Yang HX)

Department of Obstetrics and Gynecology, International Peace Maternity & Child Health Hospital, Shanghai 200030, China (Zeng CJ)

Correspondence to: Dr. YANG Hui-xia, Department of Obstetrics and Gynecology, Peking University First Hospital, Beijing 100034, China (Email:

MA Jing-mei and ZENG Chan-juan contributed equally to this paper.

This study was supported by a grant from National Natural Science Foundation of China (No. 30973212).

(Received January 12, 2012)

Edited by CHEN Li-min

Diabetes in pregnancy, including type 1 diabetes, type 2 diabetes and gestational diabetes mellitus (GDM) are all classified as hyperglycemia status in pregnancy and are associated with several adverse outcomes over the short- and long-term for both the mother and offspring.1,2 The frequency of GDM is consistently rising globally as well as China.3,4

There appears to be a “metabolic memory”. This memory may start with the environment in the uterus if the mother suffered from hyperglycemia. This “hyperglycemia memory” may involve the complex etiology of metabolic syndrome (MS); including obesity, hypertension, dyslipidemia and diabetes. Epidemiologic and animal studies5–7 have shown strong relationships between the early environment during development and the subsequent risk of features of the MS, especially if the developmental and adult environments were dissimilar. This forms the basis of what is known as the developmental origins of health and disease hypothesis (DOHaD). Among the mechanism of MS, insulin resistance (IR)8 is central to the development of MS, and will lead to the glucose and lipid metabolism disorders.

Peroxisome proliferator-activated receptor gamma coactivator 1-alpha9 (PGC-1α) is a transcriptional coactivator that regulates the genes involved in energy metabolism. This protein interacts with the nuclear receptor PPAR-γ, which permits the interaction with multiple transcription factors and regulates their activity; including cAMP response element-binding protein (CREB) and nuclear respiratory factors (NRFs). It provides a direct link between external physiological stimuli and the regulation of mitochondrial biogenesis, and is a major factor that regulates determination of muscle fiber type.

PGC-1α is also involved in controlling blood pressure, regulating cellular cholesterol homoeostasis, and the hepatic expression of key gluconeogenic enzymes as a key mechanism for controlling glucose production in hepatocytes. It is likely involved in pathogenic conditions such as obesity, diabetes. Some molecular evidence has shown that maternal placental ischemia10 or ethanol exposure 11 followed by catch-up growth of the offspring can persistently alter hepatic PGC-lα and gene expression and that physiological changes that persist into adulthood contribute to the onset of features of MS.

In our study a well-established animal model12,13 of severe maternal hyperglycemia was utilized to determine if prenatal hyperglycemia would influence the glucose and lipid metabolism of their offspring, and determine the effects on PGC-lα expression. From this study, we may observe whether expression change of PGC-lα may occur ahead of the insulin resistance.

Back to Top | Article Outline


The entire experiment was approved by the Ethics Committee of the Peking University First Hospital (Beijing, China) for animal procedures, which assures adherence to the standards established by the Guide for the Care and Use of Laboratory Animals.

Back to Top | Article Outline


Thirty female Wistar rats weighing 250 g-270 g and ten male rats weighing 360-400 g were obtained from the Vital River Laboratory Animal Technology Co., Ltd., China. They were maintained in an experimental room at Peking University First Hospital under controlled conditions of temperature (21±4)°C, a 12-hour light/dark cycle, and allowed to have free access to food, standard rat chow with 1.04% calcium and 0.9% phosphorus, and water.

Back to Top | Article Outline

Experimental procedures

From 5 p.m. to 7 p.m., three female were caged overnight with a male, and vaginal smears were taken the following morning around 7 a.m. to 8 a.m. Sperm found in the same vaginal smear at three different sites was designated as day 0 of gestation. On day 5 of gestation, 24 pregnant rats were randomly assigned to two experimental groups: severe hyperglycemia (SD group, n=16), rats that received a single intraperitoneal injection of 50 mg/kg streptozotocin (Sigma Chemical Company,) and control (C group, n=8), rats that received a single intraperitoneal injection of an equivalent amount of citrate buffer solution (0.1 mol/L, pH 6.5). During pregnancy, blood samples from both groups were obtained from cut tail tips for glycemic determination using a typical glucometer (One Touch Ultra/Blood glucose test). Inclusion criteria for the SD group required glycemic levels on day 8 of gestation of >20 mmol/L. There were eight SD and eight C rats. On day 20 of gestation, each rat was moved to their own cage. Fasting glucose levels were significantly increased in the SD group compared with the C group (6.25±1.41 vs. 28.34±5.14, P <0.01). After weaning, pups were fed standard laboratory chow ad libitum until the date of the experiments. These offsprings were chosen and divided randomly into two groups: (1) CO (offspring from control group), eight pups from control group, four female and four male, (2) SDO (offspring from the severe hyperglycemia group), eight pups from the severe hyperglycemia group, 4 female and 4 male. Two separate adult offspring cohorts were set up.

Back to Top | Article Outline

Offspring physical features

Offspring body weight was measured at birth and then weekly after weaning. Growth rate (GR= (Wx-W3)/W3) were calculated. Blood pressure monitoring was done on postnatal day 90, 120, 150, and 180. With an indirect tail-cuff method (ALC-NIBP, Shanghai Alcott Biotech, China), blood pressure was determined in conscious rats of the two groups. Rats were adequately trained before the effective measurements: they were briefly placed in restrainers several times and then covered with a dark tissue. After this step, they remained calm and the pressure records were made in 15 to 20 minutes in a state of quiet. The environment was silent, with constant temperature and the measurements were always performed by the same person. Three consecutive measurements were taken. All the values were discarded when a great variability was found, and the rat was measured the following day.

Back to Top | Article Outline

Plasma and tissue collection

After a 12-hour fast, 180-day-old offspring rats were anesthetized with sodium pentobarbital (2%, 40 mg/kg). The midneck was incised to isolate the common carotid artery, and a 4-ml blood sample drawn, centrifuged for 15 minutes and then stored at -20°C for subsequent determination of biochemical indexes. Following blood collection, the animals were decapitated to allow the rapid removal of the liver and pancreas. The tissues were fixed in 4% formaldehyde and embedded in paraffin for immunohistochemistry.

Back to Top | Article Outline

Biochemical index

Plasma insulin concentrations and serum glucose were assayed by ELASA kit (Sun Biomedical Technology Co., Ltd, Beijing, China). Total triglyceride (TG), total cholesterol (TC), low density lipoprotein (LDL) and high density lipoprotein (HDL) were measured in the Biochemical Laboratory of Peking University First Hospital.

Back to Top | Article Outline

Pancreas immunohistochemistry

Pancreas tissues were dehydrated and embedded in paraffin. Serial sections were cut at 5 μm and mounted on Superfrost Plus glass slides. The sections were rehydrated and immunostained with the primary antibody, rabbit anti-guinea pig insulin antibody (Santa Cruz), diluted 1:2000. After four washes in PBS, slides were incubated with the appropriate biotin-labeled secondary antibody. Following development with diaminobenzidine/H2O2, sections were counterstained with hematoxylin, cleared, and mounted in Permount. Morphometric analysis was performed using a Zeiss transmitted light microscope at magnifications of x100 or x400. Analysis was performed with morphometric analysis software (Image-Pro plus 5.0, Media Cybernetics). To calculate islet area, or the percentage of the islet area staining positive for insulin, individual islets were circled for image analysis and selected by RGB threshold: (1) The β cell area was quantified by acquiring adjacent x400 images of the entire pancreas from two anti-insulin stained sections per animal. (2) The ratio of β cells was calculated from the mean insulin-cell areas. (3) The proportion of β-cell surface area versus surface area of the whole pancreas was determined by acquiring adjacent x100 images of the entire pancreas.

Back to Top | Article Outline

Quantitative real-time reverse transcription-PCR (qRT-PCR)

The mRNA levels of PGC-1 alpha, G-6-Pase, PEPCK and GcK in the liver were measured at 180 days. cDNA was synthesized from 1 μg of deoxyribonuclease-treated mRNA, as described above. Target primers were designed using Primer Express Software (PE Applied Biosystems, Foster City, CA) (Table 1). qRT-PCR using SYBR Green chemistry was performed. Samples were subjected to a heat dissociation protocol after thefin al cycle of PCR to ensure that only one product was detected. Relative quantification of gene expression was performed by the comparative CT method, with β-actin as the endogenous control. Cycle parameters were 95°C for 30 seconds, then then 40 cycles of 95°C for 3 seconds, then 60°C for 25 seconds. Each sample was analyzed in triplicate in assays performed on three occasions to exclude the possibility that real-time PCR findings were a result of a nonspecific increase in mRNA levels.

Table 1

Table 1

Back to Top | Article Outline

Western blotting

Protein was isolated, by centrifugation, after liver homogenization in Laemmeli lysis buffer. A total of 30 μg of liver protein was separated on an SDS-PAGE gel along with molecular weight markers and subsequently transferred to nitrocellulose. The nitrocellulose was incubated in Blotto solution, and then with rabbit anti-PGC-1 alpha primary antibody (1:2000) (Santa Cruz, CA, USA). The filters were then washed before incubation with secondary anti-rabbit antibody (1:1000 dilutions). The filters were washed and immunoreactive proteins from the liver were visualized by ECL. The products were quantified by densitometry after standardization for loading. Each blot was repeated three times.

Back to Top | Article Outline

Statistical analysis

Data from different groups were combined and reported as the mean ± standard deviation (SD). Student's t test for independent means comparisons was used to provide a statistical analysis. Values were considered significant when P <0.05.

Back to Top | Article Outline


Weight and catch-up growth

At birth, body weights in the SDO group were significantly lower than those of the CO group (P <0.05). Although SDO rats displayed early catch-up growth, with a higher growth rate (P <0.05) and they attained similar body weights as the CO rats by 9 weeks of age (P >0.05). Ultimately, body weights of the SDO rats remained lower than those of the CO rats at 180 days of age (P >0.05) (Table 2).

Back to Top | Article Outline

The blood pressure and metabolic markers of the offspring

At 90 days and 120 days after birth, there was no significant difference between the CO and SDO groups in metabolic markers. At 180 days after birth, the systolic, diabolic and the mean arterial blood pressure (SBP, DBP, MAP) of the SDO group was higher than in the CO group. At 180 days of age in the SDO, fasting plasma TG and TC concentrations were notably increased compared with the CO rats (P <0.05) (Table 3).

To study the effect of prenatal hyperglycemia on the offsprings' glucose homeostasis, serum glucose and insulin levels in the fasting state were measured. FPG levels were similar among the groups (P >0.05). Plasma insulin concentrations were elevated in the SDO rats compared with the CO rats. The SDO group exhibited overt hyperinsulinemia, and higher homeostatic model assessment (HOMA-IR) relative to other group (P <0.05) (Table 3).

Table 2

Table 2

Table 3

Table 3

Back to Top | Article Outline

Pancreatic morphology

At 180 days of age, rats in the SDO group had no significant increase in β-cell mass, islet hypertrophy, or β-cell percentage, compared with the CO group (P >0.05) (Table 4 and Figure 1).

Figure 1.

Figure 1.

Table 4

Table 4

Back to Top | Article Outline

Expression of genes that regulate glyconeogenesis in liver

Compared with CO rats at 180 days of age, the SDO rats' hepatic PGC-1, PEPCK, and G-6-Pase mRNA levels were significantly higher (P <0.05). GCK mRNA levels were significantly lower (P <0.05) (Figure 2). At 180 days of age, the hepatic PGC-1 protein levels of in SDO rats were also significantly increased (P <0.05) (Figure 3).

Figure 2.

Figure 2.

Back to Top | Article Outline


Model of offspring of rats from severe diabetic mother

Experimental studies14–17 have demonstrated that an adverse embryonic or fetal environment, such as maternal hyperglycemia, can induce structural and/or functional abnormalities and lead to permanent changes in metabolic features during adult life. Links between prenatal growth and the later risk of non-communicable diseases (NCD) such as diabetes, cardiovascular disease and the metabolic syndrome are thought to reflect variations in the quality of the intra-uterine environment.18 As well as the limiting effects of small uterus size constrained fetus growth may reflect other aspects of the intra-uterine environment such as nutrition, oxygen supply and hormonal exposure. Both intra-uterine growth restriction and preterm birth appear to have long-term consequences for NCD risk.

Animal models play an important role in the study of this mechanism.14–17 Compared to human, rats have a brief accelerated childhood. Rats develop rapidly during infancy and become sexually mature at about six weeks of age. Humans, on the other hand, develop slowly and do not hit puberty until about 12 to 13 years. While rats become sexually mature at six weeks, they reach social maturity several months later at about five to six months of age, which is similar to eighteen years old for humans.

In this study, we established a rat model of maternal severe hyperglycemia, with offspring having a lower body weight at birth compared to the control group. This model mimics the hyperglycemia status during pregnancy in humans. The follow-up was focused on the offspring. From birth to 6 weeks, which is the childhood period of rats, the growth pattern showed a catch-up trend in the SDO group, with CO group as the baseline.

At the age of 180 days, which means the offspring rats had reached social maturity, we observed that the SDO group showed mild to moderate increases of MAP, SBP and DBP. We also found that the SDO rats developed hyperlipidemia. FPG in the SDO group was normal, but more insulin was needed to maintain the level. This showed that IR occurred, which is the key process for the development of metabolic syndrome. The process was similar to what had been found in epidemiological studies.18,19

Back to Top | Article Outline

Effects of maternal severe hyperglycemia on β cell function and mass

In rodents, the number of β cells in the endocrine pancreas seems to be determined during a critical window that is somewhere in the last quarter of fetal gestation and in the first few days after birth.20 Islet development may be altered by changes in nutrient availability. Maternal hyperglycemia constitutes the challenge in our study. The model shows that at 180 day after birth, the SDO group exhibit overt hyperinsulinemia, and a higher HOMA-IR relative to the CO group. The β cell mass and pancreatic islet size was increased in the SDO rats, while relative β cell volume was decreased compared with the CO group. No significant changes of islet morphological were found in SDO islets, which indicate that it was the function of β cells that was impaired. Experimental FGR studies (obtained through malnutrition or intrauterine artery ligation) had demonstrated that an adverse embryonic or fetal environment can induce structural and functional abnormalities in the pancreatic islet cells and that β cell mass persists into adulthood and can be the cause of permanent changes in insulin sensitivity.21

FGR leads to a decrease in the insulin content and β-cells degranulation in fetal sheep, as well as low circulating insulin levels and reduced β-cell response to exogenous glucose. Development of fetal islet cells has been shown to depend on the availability of glucose and amino acids.22 A poor nutrient supply fails to stimulate the development of the fetal endocrine pancreas. The elevated glucose concentrations in the mother lead to degranulation of the fetal β cells. When nutrients are restored after the postnatal period, β cell neogenesis and replication returns to normal, but fetal β cells function would be irreversible.

These findings indicate that the prenatal severe diabetic environment may program permanent alterations in pancreas development which may be linked to the earlier development of IR in the SDO group.

Back to Top | Article Outline

Molecular mechanisms underlying hepatic insulin resistance

Insulin resistance and the compensatory hyperinsulinaemia are thought to be the key elements of metabolic syndrome. Liver is a major buffering tissue that ensures nutrient homeostasis in fed and fasting conditions. Metabolic adaptation to fasting requires an important control at the transcriptional level. PGC-lα is induced in the fasted liver to activate gluconeogenic and fatty acid oxidation genes. Recent research has shifted the focus to a biochemical and molecular approach that concentrates on key metabolic genes and their transcriptional control, which lead to an increased risk of developing MS when disturbed. PGC-lα in hepatocytes strongly activates key gluconeogenic enzymes, including PEPCK and G-6-Pase, leading to increased glucose output. It has been seen as having a critical role in the maintenance of glucose, lipid, and energy homeostasis and is likely involved in pathogenic conditions such as obesity, diabetes and cardiomyopathy. It has been observed that PGC-1α mRNA levels are elevated in the liver and pancreas of rodent models of type 1 and type 2 diabetes.23,24 In these tissues, increased PGC-1α levels can account for elevated glucagon genesis and decreased insulin secretion in diabetes. In addition, acute knockdown of PGC-1α in liver improves hepatic insulin resistance and glucose tolerance in db/db mice.9

Alterations of glucagon-genic enzymes have been reported in adult rat offspring born with FGR induced by prenatal malnutrition,16 placental ischemia,25 ethanol exposure11 or glucocorticoid exposure.15 Furthermore, failure of insulin to suppress HGP has been reported in the offspring of mothers fed a low-protein diet.14 In the placental ischemia FGR model,25 where hyperinsulinemia was shown to precede the development of hyperglycemia, hepatic PEPCK and G-6-Pase mRNA were both increased and basal HGP was even higher. It has also been demonstrated that hepatic PGC-1α is increased in the livers of the offspring of malnutritional mother. Our findings were similar to the findings above.

Our study showed that PGC-1α, PEPCK and G-6-Pase mRNA levels were increased about two folds in the liver of SDO offspring compared with control rats. These processes occur early in the life, before the onset of hyperglycemia. Yet, the exact mechanism and its interaction with canonical-pathways await future investigations.

Overall, these findings indicate that maternal hyperglycemia leads to permanent changes in hepatic glucose metabolism in the offspring, and that hepatic gluconeogenic molecular dysfunction may contribute to the metabolic morbidities experienced by this population.

Back to Top | Article Outline


1. HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991-2002.
2. Fan ZT, Yang HX, Gao XL, Lintu H, Sun WJ. Pregnancy outcome in gestational diabetes. Int J Gyn Obstet 2006; 94: 12-16.
3. Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J, et al. Prevalence of diabetes among men and women in China. N Engl J Med 2010; 362: 1090-1101.
4. Yang H, Wei Y, Gao X, Xu X, Fan L, He J, et al. Risk factors for gestational diabetes mellitus in Chinese women: a prospective study of 16 286 pregnant women in China. Diabet Med 2009; 26: 1099-1104.
5. Gluckman PD, Hanson MA, Cooper C, Thornburg KL. Effect of in utero and early-life conditions on adult health and disease. N Engl J Med 2008; 359: 61-73.
6. Gillman MW. Developmental origins of health and disease. N Engl J Med 2005; 353: 1848-1850.
7. Gluckman PD, Hanson MA, Beedle AS. Early life events and their consequences for later disease: a life history and evolutionary perspective. Am J Human Biol 2007; 19: 1-19.
8. Poston L. Developmental programming and diabetes-The human experience and insight from animal models. Best Pract Res Clin Endocrinol Metab 2010; 24: 541-552.
9. Koo SH, Satoh H, Herzig S, Lee CH, Hedrick S, Kulkarni R, et al. PGC-1 promotes insulin resistance in liver through PPAR-α-dependent induction of TRB-3. Nat Med 2004; 7: 530-534.
10. Lane RH, MacLennan NK, Hsu JL, Janke SM, Pham TD. Increased hepatic peroxisome proliferator-activated receptor-coactiva-tor-1 gene expression in a rat model of intrauterine growth retardation and subsequent insulin resistance. Endocrinology 2002; 143: 2486-2490.
11. Vuguin P, Raab E, Liu B, Barzilai N, Simmons R. Hepatic insulin resistance precedes the development of diabetes in a model of intrauterine growth retardation. Diabetes 2004; 53: 2617-2622.
12. Fujisawa Y, Nakagawa Y, Ren-shan L, Ohzeki T. Streptozotocin-induced diabetes in the pregnant rat reduces 11 beta-hydroxysteroid dehydrogenase type 2 expression in placenta and fetal kidney. Life Sci 2004; 75: 2797-2805.
13. Liu HH, Zhang YL, Yang HX. Intrauterine hyperglycemia and leptin resistance of offsprings. Chin J Obstet Gynecol (Chin) 2009; 44: 841-845.
14. Ozanne SE, Smith GD, Tikerpae J, Hales CN. Altered regulation of hepatic glucose output in the male offspring of protein-malnourished rat dams. Am J Physiol Endocrinol Metab 1996; 270 (4 Pt 1): E559-E564.
15. Nyirenda MJ, Lindsay RS, Kenyon CJ, Burchell A, Seckl JR. Glucocorticoid exposure in late gestation permanently programs rat hepatic phosphoenolpyruvate carboxykinase and glucocorticoid receptor expression and causes glucose intolerance in adult offspring. J Clin Invest 1998; 101: 2174-2181.
16. Desai M, Byrne CD, Zhang J, Petry CJ, Lucas A, Hales CN. Programming of hepatic insulin-sensitive enzymes in offspring of rat dams fed an isocaloric protein-restricted diet. Am J Physiol Gastrointest Liver Physiol 1997; 272 (5 Pt 1): G1083-G1090.
17. Gómez-Ambrosi J, Rodriguez A, Catalán V, Becerril S, Gil MJ, Mugueta C, et al. Impaired adiponectin-AMPK signalling in insulin-sensitive tissues of hypertensive rats. Life Sci 2008; 83: 540-549.
18. Meigs JB, Cupples LA, Wilson PW. Parental transmission of type 2 diabetes: the Framingham Offspring Study. Diabetes 2000; 49: 2201-2207.
19. Bunt JC, Tataranni PA, Salbe AD. Intrauterine exposure to diabetes is a determinant of hemoglobin A(1)c and systolic blood pressure in pima Indian children. J Clin Endocrinol Metab 2005; 90: 3225-3229.
20. Hanson M, Godfrey KM, Lillycrop KA, Burdge GC, Gluckman PD. Developmental plasticity and developmental origins of non-communicable disease: theoretical considerations and epigenetic mechanisms. Prog Biophys Mol Biol 2011; 106: 272-280.
21. Zambrano E, Bautista CJ, Deás M, Martinez-Samayoa PM, González-Zamorano M, Ledesma H, et al. A low maternal protein diet during pregnancy and lactation has sex- and window of exposure-specific effects on offspring growth and food intake, glucose metabolism and serum leptin in the rat. J Physiol 2006; 571(Ptl): 221-230.
22. Yoon JC, Xu G, Deeney JT, Yang SN, Rhee J, Puigserver P, et al. Suppression of b cell Energy metabolism and insulin release by PGC-1α. Dev Cell 2003; 5: 73-83.
23. Yoon JC, Puigserver P, Chen G, Donovan J, Wu Z, Rhee J, et al. Control of hepatic gluconeogenesis through the transcriptional coactivator PGC-1. Nature 2001; 413: 131-138.
24. Kervran A, Guillaume M, Jost A. The endocrine pancreas of the fetus from diabetic pregnant rat. Diabetologia 1978; 15: 387-393.
25. Handschin C, Spiegelman BM. Peroxisome proliferator activated receptor gamma coactivator 1 coactivators, energy homeostasis, and metabolism. Endocr Rev 2006; 27: 728-735.

prenatal programming; peroxisome proliferator-activated receptor coactivator-la; phosphoenolpyruvate carboxykinase; insulin resistance

© 2012 Chinese Medical Association