- Email: [email protected]
Does leptin play a role in the pathogenesis of human nonalcoholic steatohepatitis?
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2003 by Am. Coll. of Gastroenterology Published by Elsevier Inc.
Vol. 98, No. 12, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/j.amjgastroenterol.2003.09.037
Does Leptin Play a Role in the Pathogenesis of Human Nonalcoholic Steatohepatitis? Naga Chalasani, M.D., David W. Crabb, M.D., Oscar W. Cummings, M.D., Paul Y. Kwo, M.D., Ali Asghar, Ph.D., Prashant K. Pandya, D.O., and Robert V. Considine, Ph.D. Departments of Medicine and Pathology, Indiana University School of Medicine, Indianapolis, Indiana
OBJECTIVES: Obesity is a risk factor for nonalcoholic steatohepatitis (NASH). Leptin plays an important role in the regulation of food intake, body composition, energy expenditure, and body weight. It has been suggested that leptin plays a role in the pathogenesis of NASH; however, adequate studies are lacking. We therefore conducted a study to explore the role of serum leptin in the pathogenesis of human NASH. METHODS: We measured the levels of serum leptin and its anthropometric, biochemical, metabolic, and histological correlates in a cohort of patients with NASH (n ⫽ 26) and well-matched controls (n ⫽ 20). Furthermore, we measured the levels of leptin in the serum and hepatic leptin and leptin receptor messenger RNA (mRNA) expression in liver biopsy specimens of patients with NASH (n ⫽ 5) and simple steatosis (n ⫽ 5). RESULTS: Serum leptin was not statistically different between patients with NASH and their controls (21 ⫾ 13 vs 18 ⫾ 11 ng/ml, respectively, p ⫽ 0.5). There was no correlation between serum leptin and hepatic histology, serum transaminases, fasting insulin levels, or a measure of insulin resistance. After adjusting for covariates in a multiple regression analysis, only percent body fat (p ⫽ 0.04) and subcutaneous abdominal fat area (p ⫽ 0.04) had significant correlation with serum leptin. There was no expression of leptin mRNA in the cell lysate of liver biopsy specimens of subjects with NASH or steatosis. Additionally, the serum leptin levels and the hepatic leptin receptor mRNA expression were not statistically different between patients with NASH and those with simple steatosis. CONCLUSION: These data do not support a direct role for leptin in the pathogenesis of human NASH. (Am J Gastroenterol 2003;98:2771⫺2776. © 2003 by Am. Coll. of Gastroenterology)
INTRODUCTION Obesity is one of the most common risk factors for nonalcoholic steatohepatitis (NASH), and its prevalence in patients with proven NASH has ranged from 39% to 95% in various studies (1). Because leptin is closely involved in the regulation of food intake, body composition, and energy
expenditure (and therefore in the pathogenesis of obesity), there has been a significant interest in exploring its role in the pathogenesis of hepatic steatosis and steatohepatitis (2, 3). Recent experiments showed that leptin promotes insulin resistance, elevates circulating insulin levels (4 – 6), and augments inflammatory and profibrogenic responses in the murine liver exposed to hepatotoxic chemicals (7). These data, in conjunction with the observation that leptin is expressed and synthesized by the activated hepatic stellate cells (8), support the notion that leptin might play a role in the pathogenesis of human hepatic steatosis and steatohepatitis. In a recent study of 47 subjects with NASH and 47 matched controls, Chitturi et al. (9) showed that serum leptin levels were significantly higher in subjects with NASH as compared with controls. Furthermore, they showed that serum leptin was independently associated with the degree of hepatic steatosis but not hepatic inflammation or hepatic fibrosis. However, the association between hepatic steatosis and serum leptin could not be confirmed in a subsequent preliminary study (10). In addition to these two reports, there are three other published reports in the literature that have yielded conflicting results regarding the role of leptin in the pathogenesis of NASH (11–13). Therefore, we conducted the following studies to further elucidate the role of leptin in the pathogenesis of NASH. First, we measured the serum levels of leptin and its metabolic, anthropometric, and histological correlates in patients with NASH and matched controls. Second, we measured the expression of hepatic leptin and leptin receptor genes in the liver biopsy specimens of patients with simple fatty liver and NASH.
MATERIALS AND METHODS This study was reviewed and approved by the institutional review board and the Advisory Committee for the General Clinical Research Center of Indiana University School of Medicine. All volunteers gave written informed consent before their participation. The majority of the participants who took part in this protocol have also participated in other NASH-related research studies conducted by our group; those results have been previously reported (14, 15).
2772
Chalasani et al.
AJG – Vol. 98, No. 12, 2003
Subjects Twenty-six patients with biopsy-proven NASH and 20 age-, gender-, and body mass index (BMI)-matched controls without liver disease participated in this prospective study. The details of criteria used to characterize the subjects with NASH and the controls of this cohort have been described previously (15). In brief, none of the subjects had histological evidence of established cirrhosis or significant alcohol consumption (defined as more than seven drinks per week in women or more than 14 drinks per week in men) or previous use of medications associated with fatty liver or evidence of other types of chronic liver disease. The absence of liver disease in the controls was established based on clinical, laboratory, and imaging criteria. All controls had normal liver biochemistries and lacked any evidence from physical examination of chronic liver disease. Additionally, an abdominal ultrasound was performed to exclude “bright liver” in a random sample of five control subjects. We recognize that occult liver disease in obese controls cannot be completely excluded by clinical, laboratory, and imaging criteria, but it is unethical to perform a liver biopsy on potential controls.
fibrosis was categorized as none (stage 0), mild (stage 1), or severe (stage 2).
Anthropometry The anthropometric measurements obtained for this study include height, weight, BMI, waist-to-hip ratio (WHR), body composition, and abdominal fat measurements. Body composition was measured with the air displacement method (BOD POD, Life Measurement Instruments, Concord, CA), as described previously (16). Total abdominal fat area (TFA), subcutaneous abdominal fat area (SFA), and visceral fat area (VFA) were measured at the level of the umbilicus (expressed as millimeters squared) with a single-slice CT image, according to the protocols described previously (14, 15).
Statistical Analysis Statistical analyses were performed with StatView software (SAS Institute, Cary, NC), and any p value less than 0.05 was considered statistically significant. Depending on the data distribution, comparisons were made between the two groups with the Student t test or Mann-Whitney test. Spearman rank correlations were used to detect the associations between serum leptin and various demographic, anthropometric, and metabolic variables. The relationship between hepatic histology and serum leptin was evaluated with analysis of variance. When necessary, multiple regression analysis was performed to take into account the linear effect of several independent variables predicting the dependent variable.
Serum Studies Liver biochemistries (AST, ALT, ALP, bilirubin) and fasting levels of serum lipids (cholesterol, low density lipoprotein, high density lipoprotein, and triglycerides), insulin, glucose, and leptin were measured. Serum for leptin was obtained between 7:30 and 8:30 AM from every participant after a 12-h fast and was measured by radioimmunoassay (Linco Research, St. Charles, MO). The homeostasis model assessment method (HOMA), a measure of insulin resistance, was calculated with this formula: (fasting insulin [U/ml] ⫻ fasting glucose [mmol/L])/22.5. Histology Subjects must have had either steatosis with balloon degeneration of hepatocytes or steatosis with sinusoidal fibrosis and polymorphonuclear infiltrates, with or without Mallory bodies, to meet the histological definition of NASH (14, 15). A dedicated hepatopathologist (O.W.C) who was blinded to the research data graded the liver biopsy specimens according to methods described by Brunt et al., with modifications (17). Steatosis was categorized as mild (stage 1), moderate (stage 2), or severe (stage 3); inflammation was categorized as none (stage 0), mild (stage 1), or severe (stage 2); and
Measurement of Leptin and Leptin Receptor Messenger RNA From the Liver Biopsy Material To examine whether leptin plays any role in the transition between fatty liver and NASH, we measured the hepatic expression of leptin and leptin receptor messenger RNA (mRNA) in the cell lysate of liver biopsy specimens from five subjects with NASH and five age-, gender-, and BMImatched subjects with fatty liver. Leptin and leptin receptor mRNA levels measured in the liver tissue of five age- and gender-matched healthy subjects (who had liver biopsies performed as part of donor evaluation for living related liver transplantation) served as controls for measuring the effect of obesity. The RNA isolation and complementary DNA synthesis, polymerase chain reaction, in vitro synthesis of competitive reference standard RNA, and competitive reverse transcriptase polymerase chain reaction for quantification of leptin and leptin receptor were performed with methods and primers described previously (14, 18 –20).
RESULTS Selected demographic, anthropometric, metabolic, and laboratory characteristics of patients with NASH and their matched controls are shown in Table 1. Whereas age, gender, BMI, total body fat, SFA, and WHR were similar between the groups, subjects with NASH had significantly higher VFA than the controls (Table 1). As expected, subjects with NASH had significantly higher levels of glucose, insulin, and HOMA than the controls (Table 1). Four subjects in the NASH group had known diabetes mellitus, but none were receiving any pharmacological therapy for the management of their diabetes; no one in the control group had diabetes mellitus. There was no statistical difference in the serum leptin levels in patients with NASH when compared with the control group (21 ⫾ 13 ng/ml vs 18 ⫾ 12 ng/ml, respectively, p ⫽ 0.5). On the univariate analysis, serum leptin was significantly associated with female gender (p ⬍ 0.001),
AJG – December, 2003
Serum Leptin in NASH
2773
Table 1. Demographic, Anthropometric, and Biochemical Measurements in NASH and Control Groups
Age (yr) Female (%) BMI (kg/m2) WHR Body fat (%) SFA (mm2) VFA (mm2) Visceral fat (%) Fasting lipids Cholesterol (mg/dl) HDL (mg/dl) LDL (mg/dl) Triglycerides (mg/dl) Serum glucose (mg/dl) Serum insulin (ng/ml) HOMA ASTm(IU/L) ALT (IU/L) Leptin (ng/ml)
NASH (n ⫽ 26)
Controls (n ⫽ 20)
45 ⫾ 14 50 33 ⫾ 5 0.96 ⫾ 0.06 40 ⫾ 9 31,294 ⫾ 11,734 14,616 ⫾ 6,361 29 ⫾ 12
41 ⫾ 14 50 32 ⫾ 4 0.94 ⫾ 0.05 39 ⫾ 10 32,809 ⫾ 12,834 9,938 ⫾ 7,390 21 ⫾ 14
205 ⫾ 36 36 ⫾ 7 131 ⫾ 30 228 ⫾ 83 114 ⫾ 42 1.0 ⫾ 0.7 7.5 ⫾ 5.5 71 ⫾ 49 90 ⫾ 73 21 ⫾ 13
201 ⫾ 46 46 ⫾ 14.5 125 ⫾ 38 160 ⫾ 74 90 ⫾ 11 0.55 ⫾ 0.4 3.2 ⫾ 3.0 26 ⫾ 12 27 ⫾ 8 18 ⫾ 12
p 0.4 0.9 0.1 0.4 0.2 0.7 0.002 0.007 0.7 0.003 0.3 0.006 0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.5
Values are expressed as mean ⫾ SD. HDL ⫽ high density lipoprotein; LDL ⫽ low density lipoprotein.
BMI (r ⫽ 0.41, p ⫽ 0.005), percent body fat (r ⫽ 0.75, p ⬍ 0.001), SFA (r ⫽ 0.71, p ⬍ 0.001), and high density lipoprotein (r ⫽ 0.40, p ⫽ 0.006). However, on the multivariate analysis, only percent body fat (p ⫽ 0.04) and SFA (p ⫽ 0.04) had statistically significant association with serum leptin (female gender had a marginally significant association, p ⫽ 0.08). Because it has been suggested that serum leptin might play a role in the pathogenesis of NASH through its effects on insulin resistance, we closely examined the relationship between leptin and carbohydrate metabolism. We found no association between serum leptin levels and the fasting levels of glucose (r ⫽ 0.10, p ⫽ 0.3), fasting insulin (r ⫽ 0.02, p ⫽ 0.7), or HOMA (r ⫽ 0.05, p ⫽ 0.8). Upon histological staging, steatosis was mild in seven, moderate in 12, and severe in seven patients; inflammation was none in six, mild in 12, and severe in eight; and fibrosis was none in two, mild in 17, and severe in seven patients. There was no statistically significant association between serum leptin levels and degree of hepatic steatosis (p ⫽ 0.06), hepatic inflammation (p ⫽ 0.7), or hepatic fibrosis (p ⫽ 0.4). Furthermore, there was no correlation between serum levels of leptin and AST (r ⫽ 0.07, p ⫽ 0.6) or ALT (r ⫽ 0.15, p ⫽ 0.4). We performed an analysis of liver biopsy material. Age, gender, and BMI of three groups of patients (normal liver, fatty liver, and NASH) are shown in Table 2. The analysis of leptin and leptin receptor expression in these liver biopsies revealed the following findings. First, there was no expression of leptin gene in the cell lysate of any of the liver biopsy specimens studied (leptin gene expression was seen in the human omental adipocytes, which were used as positive controls). Second, the levels of leptin receptor expression and serum leptin levels between subjects with steatosis and NASH were similar (Table 2, Fig. 1). Finally, when
compared with healthy subjects with normal liver biopsies, subjects with steatosis and NASH had higher serum leptin and hepatic leptin receptor expression (Table 2, Fig. 1). These differences between healthy subjects and subjects with NASH/steatosis are likely due to a significantly lower BMI of the healthy subjects.
DISCUSSION Our study makes several important observations that argue against a role for leptin in the pathogenesis of human NASH. In this study, we not only failed to detect a statistically significant difference in the serum leptin levels between patients with NASH and their controls, but we also failed to show any significant association between serum leptin and insulin levels, insulin resistance, hepatic histology, or serum markers of hepatic inflammation. Furthermore, the absence of a significant difference in the hepatic expression of leptin, leptin receptor, or serum leptin levels in patients with simple fatty liver and NASH argues against the role of leptin in the transition between fatty liver and NASH. To our knowledge, this is the first study related to NASH Table 2. Age, Gender, BMI, and Serum Leptin Levels in Subjects Whose Hepatic Expression of Leptin and Leptin Receptor Genes Were Measured
Age (yr) Gender (male/female) BMI (kg/m2)* Serum leptin (ng/ml)*
NASH (n ⫽ 5)
Fatty Liver (n ⫽ 5)
Healthy (n ⫽ 5)
38 ⫾ 8 3/2 32 ⫾ 2 19 ⫾ 4
35 ⫾ 5 3/2 31 ⫾ 2 14 ⫾ 7
34 ⫾ 6 3/2 24 ⫾ 2 7⫾3
* BMI and serum leptin levels of healthy volunteers were significantly lower than in subjects with NASH (p ⬍ 0.05) or fatty liver (p ⬍ 0.05).
2774
Chalasani et al.
Figure 1. Leptin receptor mRNA from cell lysate from liver biopsy specimens from healthy volunteers (bands 1–5), subjects with NASH (bands 6 –10), and subjects with steatosis (bands 11–15). The relative band density of leptin receptor mRNA was not statistically different between NASH and steatosis groups (7.7 ⫾ 2.0 vs 6.3 ⫾ 1.2, respectively, p ⫽ ns). However, the band density of leptin receptor mRNA in the normal liver group (3.6 ⫾ 2.7) was significantly lower than in the other two groups (p ⬍ 0.05). The stability of RNA from the normal liver biopsies was confirmed by measurement of the expression of other hepatic genes, such cytochrome P450 2E1 and 3A4 (data not shown). L ⫽ ladder; CT⫹ ⫽ positive control (human hypothalamus tissue); CT⫺ ⫽ negative control.
that measured the expression of leptin and leptin receptor genes in the human hepatic tissue. It has been suggested that leptin might play a role in hepatic fibrogenesis. Isolated hepatic stellate cells have been shown to produce leptin during the in vitro transactivation process (7), and more recently, leptin has been shown to augment the profibrogenic response in the murine liver upon exposure to hepatotoxic chemicals (8). However, we did not detect any leptin gene expression in the liver tissue of five subjects with NASH, and moreover, there was no association between serum leptin and degree of hepatic fibrosis. Our findings, in conjunction with those Chitturi et al. (see below) suggest that leptin is an unlikely “second hit” in the pathogenesis of human NASH. The results of previously published human studies that explored the association between leptin and human NASH have been conflicting. In the first study, Giannini et al. (12) observed a trend (but not statistically significant) toward higher leptin levels in six patients with NASH when compared with 12 patients with chronic hepatitis C (six with hepatic steatosis and six without steatosis) and six healthy volunteers. The results from this study are difficult to interpret not only because of its small sample size but also owing to the significant variation in the BMI among different study groups (i.e., subjects with NASH had a significantly higher
AJG – Vol. 98, No. 12, 2003
BMI relative to other groups [27 ⫾ 3 kg/m2 in the NASH group vs 23 ⫾ 1.2 kg/m2 in the healthy volunteers and 24 ⫾ 2.0 kg/m2 in the hepatitis C group, p ⬍ 0.05]). In another study of 49 patients with NASH, Uygun et al. (11) reported that mean serum leptin levels were significantly higher in patients with NASH in comparison with patients with chronic hepatitis C and healthy controls. However, because the BMI was significantly higher in patients with NASH relative to other control groups (27 ⫾ 0.3 kg/m2 in the NASH group vs 24 ⫾ 0.6 kg/m2 in the hepatitis C group and 25 ⫾ 0.3 kg/m2 in the healthy controls, p ⬍ 0.05), it is quite possible that the higher levels of leptin in the NASH group are a manifestation of a higher BMI rather than NASH itself. In a third study, Nakao et al. (13) explored and failed to find any significant association between serum leptin and nonalcoholic fatty liver in 105 Japanese obese young adults. In the multivariate analysis, this study found that BMI, percent body fat, or serum leptin levels lacked an independent association with the presence of fatty liver in either men or women (13). Finally, in a detailed study, Chitturi et al. (9) examined the relationship between leptin and NASH in 47 subjects with NASH and 47 BMI- and gender-matched healthy volunteers. This study showed subjects with NASH had significantly higher levels of serum leptin compared with healthy volunteers, and more importantly, they were able to demonstrate an independent association between serum leptin and the degree of hepatic steatosis (but not degree of hepatic inflammation or fibrosis). It previously has been shown that the amount of total body fat and subcutaneous fat are critical in determining the serum levels of leptin in obese people (21–23). Therefore, the failure to detect an association between serum leptin and NASH in our study is likely due to the similarity of percent body fat and SFA in patients with NASH and their controls. This might also explain the discrepancy between our findings and those of Chitturi et al. (9). In the study by Chitturi et al., subjects with NASH and controls were BMI matched; however, their total body fat or abdominal fat distribution were not determined (9). It is possible that subjects with NASH in their study might have had higher body fat or higher subcutaneous fat than the controls to account for the higher levels of serum leptin seen in patients with NASH. The relationship between serum leptin and insulin levels and insulin resistance is complex. For example, whereas some in vitro studies and animal experiments have shown that leptin might promote insulin resistance and hyperinsulinemia (4 – 6), recent studies have shown that leptin administration might actually improve the insulin resistance in patients with lipodystrophy (who have low serum leptin levels) (24) and in mice with congenital lipodystrophy (who also have low leptin levels) (25). In our study, there was no association between serum leptin and fasting serum insulin levels or insulin resistance. This further argues against any potential role played by leptin in the pathogenesis of human NASH because insulin resistance has now emerged as an important factor in the pathogenesis of NASH (26 –30).
AJG – December, 2003
Although we consider our study to be the most detailed on this subject, it suffers from the unavoidable limitation of inability to definitively exclude fatty liver–related pathology in the controls. Owing to ethical considerations, we could not obtain liver biopsies in the control group. To minimize the risk of misclassification, we obtained a liver ultrasound in a random sample of controls (n ⫽ 5) and did not find a bright liver in any of them. A comparable methodology has been used to characterize the control population by a number of previous studies (9, 14, 15). Numeric mismatch between the cases and the controls is another limitation of our study. Although we did not have one specific control for each NASH patient, every NASH patient had a fully matched control (i.e., some subjects in the control group served as a match for more than one subject in the NASH group). Because there is a circadian pattern to the serum levels of leptin, the best method for measuring leptin, whether single basal measurement or serial measurements, over a 24-h period is unknown. In our study, we measured basal serum leptin levels in a fasting state between 7:30 and 8:30 AM in all subjects to avoid circadian variation. Furthermore, one could argue that our sample size was not sufficient to detect a significant difference in the serum leptin levels between subjects with NASH and obese controls. We observed a mean difference in the serum leptin levels in patients with NASH and controls of 3 ng/ml with an SD of 15.7 ng/ml (95% CI ⫽ ⫺6.4 –12.4 ng/ml). On the basis of these data, we would need to study 274 subjects in each group to demonstrate a difference with 80% power at the 5% significance level. In summary, serum leptin levels were not different between patients with NASH and appropriately matched controls, and there existed no association between serum leptin and hepatic histology, serum markers of hepatic inflammation, fasting insulin levels, or insulin resistance. Additionally, there were no differences in the levels of hepatic leptin mRNA, hepatic leptin receptor mRNA, and serum leptin between patients with NASH and those with simple fatty liver. These data do not support a role for leptin in the pathogenesis of human NASH.
ACKNOWLEDGMENTS Supported by NIH grant DK 56012 (to N.C.), a grant from the American Diabetes Association (to R.V.C.), grant P60 AA07611 (to D.W.C.), and grant RR00750 (to the G.C.R.C.). Reprint requests and correspondence: Naga Chalasani, M.D., Indiana University School of Medicine, WD OPW 2005, 1001 West 10th Street, Indianapolis, IN 46202. Received Mar. 12, 2003; accepted Aug. 5, 2003.
REFERENCES 1. Falck-Ytter Y, Younossi ZM, Marchesini G, McCullough AJ. Clinical features and natural history of nonalcoholic steatosis syndromes. Semin Liver Dis 2001;21:17–26. 2. Kaplan LM. Leptin, obesity, and liver disease. Gastroenterology 1998;115:997–1001.
Serum Leptin in NASH
2775
3. Raguso R, McCullough AJ. Leptin and the gastrointestinal tract. Curr Opin Gastroenterol 2000;16:160 –5. 4. Cohen B, Novick D, Rubinstein M. Modulation of insulin activities by leptin. Science 1996;274:1185–8. 5. Muller G, Ertl J, Gerl M, et al. Leptin impairs metabolic actions of insulin in isolated rat adipocytes. J Biol Chem 1997;272:10585–93. 6. Sivitz WI, Walsh SA, Morgan DA, et al. Effects of leptin on insulin sensitivity in normal rats. Endocrinology 1997;138: 3395–401. 7. Ikejima K, Honda H, Yoshikawa M, et al. Leptin augments inflammatory and profibrogenic responses in the murine liver induced by hepatotoxic chemicals. Hepatology 2001;34:288 –97. 8. Potter JJ, Womack L, Mezey E, et al. Transdifferentiation of rat hepatic stellate cells results in leptin expression. Biochem Biophys Res Commun 1998;244:178 –82. 9. Chitturi S, Farrell G, Frost L, et al. Serum leptin in NASH correlates with hepatic steatosis but not fibrosis: A manifestation of lipotoxicity? Hepatology 2002;36:403–9. 10. Alba LM, Petrovic L, Lindor KD, et al. Leptin is an independent marker of liver fibrosis in humans with nonalcoholic fatty liver disease. Hepatology 2002;3:296A (abstract). 11. Uygun A, Kadayifci A, Yesilova Z, et al. Serum leptin levels in patients with nonalcoholic steatohepatitis. Am J Gastroenterol 2000;95:3584 –9. 12. Giannini E, Botta F, Cataldi A, et al. Leptin levels in nonalcoholic steatohepatitis and chronic hepatitis C. Hepatogastroenterology 1999;46:2422–5. 13. Nakao K, Nakata K, Ohtsubo N, et al. Association between nonalcoholic fatty liver, markers of obesity, and serum leptin level in young adults. Am J Gastroenterol 2002;97:1796 –801. 14. Chalasani N, Gorski JC, Asghar MS, et al. Hepatic cytochrome P450 2E1 activity in non-diabetic patients with nonalcoholic steatohepatitis. Hepatology 2003;37:544 –50. 15. Chalasani N, Deeg MA, Persohn S, et al. Metabolic and anthropometric evaluation of insulin resistance in non-diabetic patients with non-alcoholic steatohepatitis. Am J Gastroenterol 2003; 98:1849⫺55. 16. Dempster P, Aitkens S. A new air displacement method for the determination of human body composition. Med Sci Sports Exerc 1995;27:1692–7. 17. Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-Tetri BA, Bacon BR. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastro 1999;94:2467–2474. 18. Considine RV, Considine EL, Williams CJ, et al. Evidence against either a premature stop codon or the absence of Ob gene mRNA in human obesity. J Clin Invest 1995;95:2986 –8. 19. Considine RV, Considine EL, Williams CJ, et al. The hypothalamic leptin receptor in humans: Identification of incidental sequence polymorphisms and absence of the db/db mouse and fa/fa rat mutations. Diabetes 1996;19:992–4. 20. Considine RV, Cao GY, Lynn RB. The physiology of leptin in humans: An update. In: Bray GA, Ryan DH, eds. Nutrition, genetics and obesity. Baton Rouge, LA: Louisiana State University Press, 2003:156 –71. 21. van Harmelen V, Reynisdottir S, Eriksson P, et al. Leptin secretion from subcutaneous and visceral adipose tissue in women. Diabetes 1998;47:913–7. 22. Minocci A, Savia G, Lucantoni R, et al. Leptin plasma concentrations are dependent on body fat distribution in obese patients. Int J Obes Relat Metab Disord 2000;24:1139 –44. 23. Misra A, Arora N, Mondal S, et al. Relation between plasma leptin and anthropometric and metabolic covariates in lean and obese diabetic and hyperlipidaemic Asian Northern Indian subjects. Diabetes Nutr Metab 2001;14:18 –26.
2776
Chalasani et al.
24. Oral EA, Simha V, Ruiz E, et al. Leptin-replacement therapy for lipodystrophy. N Engl J Med 2002;346:570 –8. 25. Shimomura I, Hammer RE, Ikemoto S, et al. Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy. Nature 1999;401:73–6. 26. Sanyal AJ, Campbell-Sargent C, Mirshahi F, et al. Nonalcoholic steatohepatitis: Association of insulin resistance and mitochondrial abnormalities. Gastroenterology 2001; 120:1183–92. 27. Marchesini G, Brizi M, Morselli-Labate AM, et al. Association of nonalcoholic fatty liver disease with insulin resistance. Am J Med 1999;107:450 –5.
AJG – Vol. 98, No. 12, 2003
28. Chitturi S, Abeygunasekera S, Farrell GC, et al. NASH and insulin resistance. Insulin hypersecretion and specific association with the insulin resistance syndrome. Hepatology 2002; 35:373–9. 29. Willner IR, Waters B, Patil SR, et al. Ninety patients with nonalcoholic steatohepatitis: Insulin resistance, familial tendency, and severity of disease. Am J Gastroenterol 2001;96: 2957–61. 30. Pagano G, Pacini G, Musso G, et al. Nonalcoholic steatohepatitis, insulin resistance, and metabolic syndrome: Further evidence for an etiologic association. Hepatology 2002;35:367– 72.
Vol. 98, No. 12, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/j.amjgastroenterol.2003.09.037
Does Leptin Play a Role in the Pathogenesis of Human Nonalcoholic Steatohepatitis? Naga Chalasani, M.D., David W. Crabb, M.D., Oscar W. Cummings, M.D., Paul Y. Kwo, M.D., Ali Asghar, Ph.D., Prashant K. Pandya, D.O., and Robert V. Considine, Ph.D. Departments of Medicine and Pathology, Indiana University School of Medicine, Indianapolis, Indiana
OBJECTIVES: Obesity is a risk factor for nonalcoholic steatohepatitis (NASH). Leptin plays an important role in the regulation of food intake, body composition, energy expenditure, and body weight. It has been suggested that leptin plays a role in the pathogenesis of NASH; however, adequate studies are lacking. We therefore conducted a study to explore the role of serum leptin in the pathogenesis of human NASH. METHODS: We measured the levels of serum leptin and its anthropometric, biochemical, metabolic, and histological correlates in a cohort of patients with NASH (n ⫽ 26) and well-matched controls (n ⫽ 20). Furthermore, we measured the levels of leptin in the serum and hepatic leptin and leptin receptor messenger RNA (mRNA) expression in liver biopsy specimens of patients with NASH (n ⫽ 5) and simple steatosis (n ⫽ 5). RESULTS: Serum leptin was not statistically different between patients with NASH and their controls (21 ⫾ 13 vs 18 ⫾ 11 ng/ml, respectively, p ⫽ 0.5). There was no correlation between serum leptin and hepatic histology, serum transaminases, fasting insulin levels, or a measure of insulin resistance. After adjusting for covariates in a multiple regression analysis, only percent body fat (p ⫽ 0.04) and subcutaneous abdominal fat area (p ⫽ 0.04) had significant correlation with serum leptin. There was no expression of leptin mRNA in the cell lysate of liver biopsy specimens of subjects with NASH or steatosis. Additionally, the serum leptin levels and the hepatic leptin receptor mRNA expression were not statistically different between patients with NASH and those with simple steatosis. CONCLUSION: These data do not support a direct role for leptin in the pathogenesis of human NASH. (Am J Gastroenterol 2003;98:2771⫺2776. © 2003 by Am. Coll. of Gastroenterology)
INTRODUCTION Obesity is one of the most common risk factors for nonalcoholic steatohepatitis (NASH), and its prevalence in patients with proven NASH has ranged from 39% to 95% in various studies (1). Because leptin is closely involved in the regulation of food intake, body composition, and energy
expenditure (and therefore in the pathogenesis of obesity), there has been a significant interest in exploring its role in the pathogenesis of hepatic steatosis and steatohepatitis (2, 3). Recent experiments showed that leptin promotes insulin resistance, elevates circulating insulin levels (4 – 6), and augments inflammatory and profibrogenic responses in the murine liver exposed to hepatotoxic chemicals (7). These data, in conjunction with the observation that leptin is expressed and synthesized by the activated hepatic stellate cells (8), support the notion that leptin might play a role in the pathogenesis of human hepatic steatosis and steatohepatitis. In a recent study of 47 subjects with NASH and 47 matched controls, Chitturi et al. (9) showed that serum leptin levels were significantly higher in subjects with NASH as compared with controls. Furthermore, they showed that serum leptin was independently associated with the degree of hepatic steatosis but not hepatic inflammation or hepatic fibrosis. However, the association between hepatic steatosis and serum leptin could not be confirmed in a subsequent preliminary study (10). In addition to these two reports, there are three other published reports in the literature that have yielded conflicting results regarding the role of leptin in the pathogenesis of NASH (11–13). Therefore, we conducted the following studies to further elucidate the role of leptin in the pathogenesis of NASH. First, we measured the serum levels of leptin and its metabolic, anthropometric, and histological correlates in patients with NASH and matched controls. Second, we measured the expression of hepatic leptin and leptin receptor genes in the liver biopsy specimens of patients with simple fatty liver and NASH.
MATERIALS AND METHODS This study was reviewed and approved by the institutional review board and the Advisory Committee for the General Clinical Research Center of Indiana University School of Medicine. All volunteers gave written informed consent before their participation. The majority of the participants who took part in this protocol have also participated in other NASH-related research studies conducted by our group; those results have been previously reported (14, 15).
2772
Chalasani et al.
AJG – Vol. 98, No. 12, 2003
Subjects Twenty-six patients with biopsy-proven NASH and 20 age-, gender-, and body mass index (BMI)-matched controls without liver disease participated in this prospective study. The details of criteria used to characterize the subjects with NASH and the controls of this cohort have been described previously (15). In brief, none of the subjects had histological evidence of established cirrhosis or significant alcohol consumption (defined as more than seven drinks per week in women or more than 14 drinks per week in men) or previous use of medications associated with fatty liver or evidence of other types of chronic liver disease. The absence of liver disease in the controls was established based on clinical, laboratory, and imaging criteria. All controls had normal liver biochemistries and lacked any evidence from physical examination of chronic liver disease. Additionally, an abdominal ultrasound was performed to exclude “bright liver” in a random sample of five control subjects. We recognize that occult liver disease in obese controls cannot be completely excluded by clinical, laboratory, and imaging criteria, but it is unethical to perform a liver biopsy on potential controls.
fibrosis was categorized as none (stage 0), mild (stage 1), or severe (stage 2).
Anthropometry The anthropometric measurements obtained for this study include height, weight, BMI, waist-to-hip ratio (WHR), body composition, and abdominal fat measurements. Body composition was measured with the air displacement method (BOD POD, Life Measurement Instruments, Concord, CA), as described previously (16). Total abdominal fat area (TFA), subcutaneous abdominal fat area (SFA), and visceral fat area (VFA) were measured at the level of the umbilicus (expressed as millimeters squared) with a single-slice CT image, according to the protocols described previously (14, 15).
Statistical Analysis Statistical analyses were performed with StatView software (SAS Institute, Cary, NC), and any p value less than 0.05 was considered statistically significant. Depending on the data distribution, comparisons were made between the two groups with the Student t test or Mann-Whitney test. Spearman rank correlations were used to detect the associations between serum leptin and various demographic, anthropometric, and metabolic variables. The relationship between hepatic histology and serum leptin was evaluated with analysis of variance. When necessary, multiple regression analysis was performed to take into account the linear effect of several independent variables predicting the dependent variable.
Serum Studies Liver biochemistries (AST, ALT, ALP, bilirubin) and fasting levels of serum lipids (cholesterol, low density lipoprotein, high density lipoprotein, and triglycerides), insulin, glucose, and leptin were measured. Serum for leptin was obtained between 7:30 and 8:30 AM from every participant after a 12-h fast and was measured by radioimmunoassay (Linco Research, St. Charles, MO). The homeostasis model assessment method (HOMA), a measure of insulin resistance, was calculated with this formula: (fasting insulin [U/ml] ⫻ fasting glucose [mmol/L])/22.5. Histology Subjects must have had either steatosis with balloon degeneration of hepatocytes or steatosis with sinusoidal fibrosis and polymorphonuclear infiltrates, with or without Mallory bodies, to meet the histological definition of NASH (14, 15). A dedicated hepatopathologist (O.W.C) who was blinded to the research data graded the liver biopsy specimens according to methods described by Brunt et al., with modifications (17). Steatosis was categorized as mild (stage 1), moderate (stage 2), or severe (stage 3); inflammation was categorized as none (stage 0), mild (stage 1), or severe (stage 2); and
Measurement of Leptin and Leptin Receptor Messenger RNA From the Liver Biopsy Material To examine whether leptin plays any role in the transition between fatty liver and NASH, we measured the hepatic expression of leptin and leptin receptor messenger RNA (mRNA) in the cell lysate of liver biopsy specimens from five subjects with NASH and five age-, gender-, and BMImatched subjects with fatty liver. Leptin and leptin receptor mRNA levels measured in the liver tissue of five age- and gender-matched healthy subjects (who had liver biopsies performed as part of donor evaluation for living related liver transplantation) served as controls for measuring the effect of obesity. The RNA isolation and complementary DNA synthesis, polymerase chain reaction, in vitro synthesis of competitive reference standard RNA, and competitive reverse transcriptase polymerase chain reaction for quantification of leptin and leptin receptor were performed with methods and primers described previously (14, 18 –20).
RESULTS Selected demographic, anthropometric, metabolic, and laboratory characteristics of patients with NASH and their matched controls are shown in Table 1. Whereas age, gender, BMI, total body fat, SFA, and WHR were similar between the groups, subjects with NASH had significantly higher VFA than the controls (Table 1). As expected, subjects with NASH had significantly higher levels of glucose, insulin, and HOMA than the controls (Table 1). Four subjects in the NASH group had known diabetes mellitus, but none were receiving any pharmacological therapy for the management of their diabetes; no one in the control group had diabetes mellitus. There was no statistical difference in the serum leptin levels in patients with NASH when compared with the control group (21 ⫾ 13 ng/ml vs 18 ⫾ 12 ng/ml, respectively, p ⫽ 0.5). On the univariate analysis, serum leptin was significantly associated with female gender (p ⬍ 0.001),
AJG – December, 2003
Serum Leptin in NASH
2773
Table 1. Demographic, Anthropometric, and Biochemical Measurements in NASH and Control Groups
Age (yr) Female (%) BMI (kg/m2) WHR Body fat (%) SFA (mm2) VFA (mm2) Visceral fat (%) Fasting lipids Cholesterol (mg/dl) HDL (mg/dl) LDL (mg/dl) Triglycerides (mg/dl) Serum glucose (mg/dl) Serum insulin (ng/ml) HOMA ASTm(IU/L) ALT (IU/L) Leptin (ng/ml)
NASH (n ⫽ 26)
Controls (n ⫽ 20)
45 ⫾ 14 50 33 ⫾ 5 0.96 ⫾ 0.06 40 ⫾ 9 31,294 ⫾ 11,734 14,616 ⫾ 6,361 29 ⫾ 12
41 ⫾ 14 50 32 ⫾ 4 0.94 ⫾ 0.05 39 ⫾ 10 32,809 ⫾ 12,834 9,938 ⫾ 7,390 21 ⫾ 14
205 ⫾ 36 36 ⫾ 7 131 ⫾ 30 228 ⫾ 83 114 ⫾ 42 1.0 ⫾ 0.7 7.5 ⫾ 5.5 71 ⫾ 49 90 ⫾ 73 21 ⫾ 13
201 ⫾ 46 46 ⫾ 14.5 125 ⫾ 38 160 ⫾ 74 90 ⫾ 11 0.55 ⫾ 0.4 3.2 ⫾ 3.0 26 ⫾ 12 27 ⫾ 8 18 ⫾ 12
p 0.4 0.9 0.1 0.4 0.2 0.7 0.002 0.007 0.7 0.003 0.3 0.006 0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.5
Values are expressed as mean ⫾ SD. HDL ⫽ high density lipoprotein; LDL ⫽ low density lipoprotein.
BMI (r ⫽ 0.41, p ⫽ 0.005), percent body fat (r ⫽ 0.75, p ⬍ 0.001), SFA (r ⫽ 0.71, p ⬍ 0.001), and high density lipoprotein (r ⫽ 0.40, p ⫽ 0.006). However, on the multivariate analysis, only percent body fat (p ⫽ 0.04) and SFA (p ⫽ 0.04) had statistically significant association with serum leptin (female gender had a marginally significant association, p ⫽ 0.08). Because it has been suggested that serum leptin might play a role in the pathogenesis of NASH through its effects on insulin resistance, we closely examined the relationship between leptin and carbohydrate metabolism. We found no association between serum leptin levels and the fasting levels of glucose (r ⫽ 0.10, p ⫽ 0.3), fasting insulin (r ⫽ 0.02, p ⫽ 0.7), or HOMA (r ⫽ 0.05, p ⫽ 0.8). Upon histological staging, steatosis was mild in seven, moderate in 12, and severe in seven patients; inflammation was none in six, mild in 12, and severe in eight; and fibrosis was none in two, mild in 17, and severe in seven patients. There was no statistically significant association between serum leptin levels and degree of hepatic steatosis (p ⫽ 0.06), hepatic inflammation (p ⫽ 0.7), or hepatic fibrosis (p ⫽ 0.4). Furthermore, there was no correlation between serum levels of leptin and AST (r ⫽ 0.07, p ⫽ 0.6) or ALT (r ⫽ 0.15, p ⫽ 0.4). We performed an analysis of liver biopsy material. Age, gender, and BMI of three groups of patients (normal liver, fatty liver, and NASH) are shown in Table 2. The analysis of leptin and leptin receptor expression in these liver biopsies revealed the following findings. First, there was no expression of leptin gene in the cell lysate of any of the liver biopsy specimens studied (leptin gene expression was seen in the human omental adipocytes, which were used as positive controls). Second, the levels of leptin receptor expression and serum leptin levels between subjects with steatosis and NASH were similar (Table 2, Fig. 1). Finally, when
compared with healthy subjects with normal liver biopsies, subjects with steatosis and NASH had higher serum leptin and hepatic leptin receptor expression (Table 2, Fig. 1). These differences between healthy subjects and subjects with NASH/steatosis are likely due to a significantly lower BMI of the healthy subjects.
DISCUSSION Our study makes several important observations that argue against a role for leptin in the pathogenesis of human NASH. In this study, we not only failed to detect a statistically significant difference in the serum leptin levels between patients with NASH and their controls, but we also failed to show any significant association between serum leptin and insulin levels, insulin resistance, hepatic histology, or serum markers of hepatic inflammation. Furthermore, the absence of a significant difference in the hepatic expression of leptin, leptin receptor, or serum leptin levels in patients with simple fatty liver and NASH argues against the role of leptin in the transition between fatty liver and NASH. To our knowledge, this is the first study related to NASH Table 2. Age, Gender, BMI, and Serum Leptin Levels in Subjects Whose Hepatic Expression of Leptin and Leptin Receptor Genes Were Measured
Age (yr) Gender (male/female) BMI (kg/m2)* Serum leptin (ng/ml)*
NASH (n ⫽ 5)
Fatty Liver (n ⫽ 5)
Healthy (n ⫽ 5)
38 ⫾ 8 3/2 32 ⫾ 2 19 ⫾ 4
35 ⫾ 5 3/2 31 ⫾ 2 14 ⫾ 7
34 ⫾ 6 3/2 24 ⫾ 2 7⫾3
* BMI and serum leptin levels of healthy volunteers were significantly lower than in subjects with NASH (p ⬍ 0.05) or fatty liver (p ⬍ 0.05).
2774
Chalasani et al.
Figure 1. Leptin receptor mRNA from cell lysate from liver biopsy specimens from healthy volunteers (bands 1–5), subjects with NASH (bands 6 –10), and subjects with steatosis (bands 11–15). The relative band density of leptin receptor mRNA was not statistically different between NASH and steatosis groups (7.7 ⫾ 2.0 vs 6.3 ⫾ 1.2, respectively, p ⫽ ns). However, the band density of leptin receptor mRNA in the normal liver group (3.6 ⫾ 2.7) was significantly lower than in the other two groups (p ⬍ 0.05). The stability of RNA from the normal liver biopsies was confirmed by measurement of the expression of other hepatic genes, such cytochrome P450 2E1 and 3A4 (data not shown). L ⫽ ladder; CT⫹ ⫽ positive control (human hypothalamus tissue); CT⫺ ⫽ negative control.
that measured the expression of leptin and leptin receptor genes in the human hepatic tissue. It has been suggested that leptin might play a role in hepatic fibrogenesis. Isolated hepatic stellate cells have been shown to produce leptin during the in vitro transactivation process (7), and more recently, leptin has been shown to augment the profibrogenic response in the murine liver upon exposure to hepatotoxic chemicals (8). However, we did not detect any leptin gene expression in the liver tissue of five subjects with NASH, and moreover, there was no association between serum leptin and degree of hepatic fibrosis. Our findings, in conjunction with those Chitturi et al. (see below) suggest that leptin is an unlikely “second hit” in the pathogenesis of human NASH. The results of previously published human studies that explored the association between leptin and human NASH have been conflicting. In the first study, Giannini et al. (12) observed a trend (but not statistically significant) toward higher leptin levels in six patients with NASH when compared with 12 patients with chronic hepatitis C (six with hepatic steatosis and six without steatosis) and six healthy volunteers. The results from this study are difficult to interpret not only because of its small sample size but also owing to the significant variation in the BMI among different study groups (i.e., subjects with NASH had a significantly higher
AJG – Vol. 98, No. 12, 2003
BMI relative to other groups [27 ⫾ 3 kg/m2 in the NASH group vs 23 ⫾ 1.2 kg/m2 in the healthy volunteers and 24 ⫾ 2.0 kg/m2 in the hepatitis C group, p ⬍ 0.05]). In another study of 49 patients with NASH, Uygun et al. (11) reported that mean serum leptin levels were significantly higher in patients with NASH in comparison with patients with chronic hepatitis C and healthy controls. However, because the BMI was significantly higher in patients with NASH relative to other control groups (27 ⫾ 0.3 kg/m2 in the NASH group vs 24 ⫾ 0.6 kg/m2 in the hepatitis C group and 25 ⫾ 0.3 kg/m2 in the healthy controls, p ⬍ 0.05), it is quite possible that the higher levels of leptin in the NASH group are a manifestation of a higher BMI rather than NASH itself. In a third study, Nakao et al. (13) explored and failed to find any significant association between serum leptin and nonalcoholic fatty liver in 105 Japanese obese young adults. In the multivariate analysis, this study found that BMI, percent body fat, or serum leptin levels lacked an independent association with the presence of fatty liver in either men or women (13). Finally, in a detailed study, Chitturi et al. (9) examined the relationship between leptin and NASH in 47 subjects with NASH and 47 BMI- and gender-matched healthy volunteers. This study showed subjects with NASH had significantly higher levels of serum leptin compared with healthy volunteers, and more importantly, they were able to demonstrate an independent association between serum leptin and the degree of hepatic steatosis (but not degree of hepatic inflammation or fibrosis). It previously has been shown that the amount of total body fat and subcutaneous fat are critical in determining the serum levels of leptin in obese people (21–23). Therefore, the failure to detect an association between serum leptin and NASH in our study is likely due to the similarity of percent body fat and SFA in patients with NASH and their controls. This might also explain the discrepancy between our findings and those of Chitturi et al. (9). In the study by Chitturi et al., subjects with NASH and controls were BMI matched; however, their total body fat or abdominal fat distribution were not determined (9). It is possible that subjects with NASH in their study might have had higher body fat or higher subcutaneous fat than the controls to account for the higher levels of serum leptin seen in patients with NASH. The relationship between serum leptin and insulin levels and insulin resistance is complex. For example, whereas some in vitro studies and animal experiments have shown that leptin might promote insulin resistance and hyperinsulinemia (4 – 6), recent studies have shown that leptin administration might actually improve the insulin resistance in patients with lipodystrophy (who have low serum leptin levels) (24) and in mice with congenital lipodystrophy (who also have low leptin levels) (25). In our study, there was no association between serum leptin and fasting serum insulin levels or insulin resistance. This further argues against any potential role played by leptin in the pathogenesis of human NASH because insulin resistance has now emerged as an important factor in the pathogenesis of NASH (26 –30).
AJG – December, 2003
Although we consider our study to be the most detailed on this subject, it suffers from the unavoidable limitation of inability to definitively exclude fatty liver–related pathology in the controls. Owing to ethical considerations, we could not obtain liver biopsies in the control group. To minimize the risk of misclassification, we obtained a liver ultrasound in a random sample of controls (n ⫽ 5) and did not find a bright liver in any of them. A comparable methodology has been used to characterize the control population by a number of previous studies (9, 14, 15). Numeric mismatch between the cases and the controls is another limitation of our study. Although we did not have one specific control for each NASH patient, every NASH patient had a fully matched control (i.e., some subjects in the control group served as a match for more than one subject in the NASH group). Because there is a circadian pattern to the serum levels of leptin, the best method for measuring leptin, whether single basal measurement or serial measurements, over a 24-h period is unknown. In our study, we measured basal serum leptin levels in a fasting state between 7:30 and 8:30 AM in all subjects to avoid circadian variation. Furthermore, one could argue that our sample size was not sufficient to detect a significant difference in the serum leptin levels between subjects with NASH and obese controls. We observed a mean difference in the serum leptin levels in patients with NASH and controls of 3 ng/ml with an SD of 15.7 ng/ml (95% CI ⫽ ⫺6.4 –12.4 ng/ml). On the basis of these data, we would need to study 274 subjects in each group to demonstrate a difference with 80% power at the 5% significance level. In summary, serum leptin levels were not different between patients with NASH and appropriately matched controls, and there existed no association between serum leptin and hepatic histology, serum markers of hepatic inflammation, fasting insulin levels, or insulin resistance. Additionally, there were no differences in the levels of hepatic leptin mRNA, hepatic leptin receptor mRNA, and serum leptin between patients with NASH and those with simple fatty liver. These data do not support a role for leptin in the pathogenesis of human NASH.
ACKNOWLEDGMENTS Supported by NIH grant DK 56012 (to N.C.), a grant from the American Diabetes Association (to R.V.C.), grant P60 AA07611 (to D.W.C.), and grant RR00750 (to the G.C.R.C.). Reprint requests and correspondence: Naga Chalasani, M.D., Indiana University School of Medicine, WD OPW 2005, 1001 West 10th Street, Indianapolis, IN 46202. Received Mar. 12, 2003; accepted Aug. 5, 2003.
REFERENCES 1. Falck-Ytter Y, Younossi ZM, Marchesini G, McCullough AJ. Clinical features and natural history of nonalcoholic steatosis syndromes. Semin Liver Dis 2001;21:17–26. 2. Kaplan LM. Leptin, obesity, and liver disease. Gastroenterology 1998;115:997–1001.
Serum Leptin in NASH
2775
3. Raguso R, McCullough AJ. Leptin and the gastrointestinal tract. Curr Opin Gastroenterol 2000;16:160 –5. 4. Cohen B, Novick D, Rubinstein M. Modulation of insulin activities by leptin. Science 1996;274:1185–8. 5. Muller G, Ertl J, Gerl M, et al. Leptin impairs metabolic actions of insulin in isolated rat adipocytes. J Biol Chem 1997;272:10585–93. 6. Sivitz WI, Walsh SA, Morgan DA, et al. Effects of leptin on insulin sensitivity in normal rats. Endocrinology 1997;138: 3395–401. 7. Ikejima K, Honda H, Yoshikawa M, et al. Leptin augments inflammatory and profibrogenic responses in the murine liver induced by hepatotoxic chemicals. Hepatology 2001;34:288 –97. 8. Potter JJ, Womack L, Mezey E, et al. Transdifferentiation of rat hepatic stellate cells results in leptin expression. Biochem Biophys Res Commun 1998;244:178 –82. 9. Chitturi S, Farrell G, Frost L, et al. Serum leptin in NASH correlates with hepatic steatosis but not fibrosis: A manifestation of lipotoxicity? Hepatology 2002;36:403–9. 10. Alba LM, Petrovic L, Lindor KD, et al. Leptin is an independent marker of liver fibrosis in humans with nonalcoholic fatty liver disease. Hepatology 2002;3:296A (abstract). 11. Uygun A, Kadayifci A, Yesilova Z, et al. Serum leptin levels in patients with nonalcoholic steatohepatitis. Am J Gastroenterol 2000;95:3584 –9. 12. Giannini E, Botta F, Cataldi A, et al. Leptin levels in nonalcoholic steatohepatitis and chronic hepatitis C. Hepatogastroenterology 1999;46:2422–5. 13. Nakao K, Nakata K, Ohtsubo N, et al. Association between nonalcoholic fatty liver, markers of obesity, and serum leptin level in young adults. Am J Gastroenterol 2002;97:1796 –801. 14. Chalasani N, Gorski JC, Asghar MS, et al. Hepatic cytochrome P450 2E1 activity in non-diabetic patients with nonalcoholic steatohepatitis. Hepatology 2003;37:544 –50. 15. Chalasani N, Deeg MA, Persohn S, et al. Metabolic and anthropometric evaluation of insulin resistance in non-diabetic patients with non-alcoholic steatohepatitis. Am J Gastroenterol 2003; 98:1849⫺55. 16. Dempster P, Aitkens S. A new air displacement method for the determination of human body composition. Med Sci Sports Exerc 1995;27:1692–7. 17. Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-Tetri BA, Bacon BR. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastro 1999;94:2467–2474. 18. Considine RV, Considine EL, Williams CJ, et al. Evidence against either a premature stop codon or the absence of Ob gene mRNA in human obesity. J Clin Invest 1995;95:2986 –8. 19. Considine RV, Considine EL, Williams CJ, et al. The hypothalamic leptin receptor in humans: Identification of incidental sequence polymorphisms and absence of the db/db mouse and fa/fa rat mutations. Diabetes 1996;19:992–4. 20. Considine RV, Cao GY, Lynn RB. The physiology of leptin in humans: An update. In: Bray GA, Ryan DH, eds. Nutrition, genetics and obesity. Baton Rouge, LA: Louisiana State University Press, 2003:156 –71. 21. van Harmelen V, Reynisdottir S, Eriksson P, et al. Leptin secretion from subcutaneous and visceral adipose tissue in women. Diabetes 1998;47:913–7. 22. Minocci A, Savia G, Lucantoni R, et al. Leptin plasma concentrations are dependent on body fat distribution in obese patients. Int J Obes Relat Metab Disord 2000;24:1139 –44. 23. Misra A, Arora N, Mondal S, et al. Relation between plasma leptin and anthropometric and metabolic covariates in lean and obese diabetic and hyperlipidaemic Asian Northern Indian subjects. Diabetes Nutr Metab 2001;14:18 –26.
2776
Chalasani et al.
24. Oral EA, Simha V, Ruiz E, et al. Leptin-replacement therapy for lipodystrophy. N Engl J Med 2002;346:570 –8. 25. Shimomura I, Hammer RE, Ikemoto S, et al. Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy. Nature 1999;401:73–6. 26. Sanyal AJ, Campbell-Sargent C, Mirshahi F, et al. Nonalcoholic steatohepatitis: Association of insulin resistance and mitochondrial abnormalities. Gastroenterology 2001; 120:1183–92. 27. Marchesini G, Brizi M, Morselli-Labate AM, et al. Association of nonalcoholic fatty liver disease with insulin resistance. Am J Med 1999;107:450 –5.
AJG – Vol. 98, No. 12, 2003
28. Chitturi S, Abeygunasekera S, Farrell GC, et al. NASH and insulin resistance. Insulin hypersecretion and specific association with the insulin resistance syndrome. Hepatology 2002; 35:373–9. 29. Willner IR, Waters B, Patil SR, et al. Ninety patients with nonalcoholic steatohepatitis: Insulin resistance, familial tendency, and severity of disease. Am J Gastroenterol 2001;96: 2957–61. 30. Pagano G, Pacini G, Musso G, et al. Nonalcoholic steatohepatitis, insulin resistance, and metabolic syndrome: Further evidence for an etiologic association. Hepatology 2002;35:367– 72.