Endotoxin and Plasminogen Activator Inhibitor-1 Serum Levels Associated With Nonalcoholic Steatohepatitis in Children

Alisi, Anna*; Manco, Melania; Devito, Rita; Piemonte, Fiorella§; Nobili, Valerio*

Journal of Pediatric Gastroenterology & Nutrition: June 2010 - Volume 50 - Issue 6 - p 645–649
doi: 10.1097/MPG.0b013e3181c7bdf1
Original Articles: Hepatology and Nutrition

Objectives: Recent evidence supports a role for endotoxemia in the progression from nonalcoholic fatty liver disease (NAFLD) to nonalcoholic steatohepatitis (NASH). We investigated the association between serum levels of endotoxin, proinflammatory molecules, and histology in children with NAFLD.

Patients and Methods: A total of 40 children, mean age of 11.9 ± 2.8 years (27 male and 13 female), with biopsy-proven NAFLD were consecutively enrolled. Anthropometrics, blood pressure, and parameters of the metabolic syndrome were collected. Serum levels of endotoxin, plasminogen activator inhibitor-1 (PAI-1), tumor necrosis factor (TNF)-α, and interleukin (IL)-6 were measured by an enzyme-linked immunosorbent assay and compared with circulating levels of the same soluble factors in 9 age- and sex-matched normal weight controls (age 11.4 ± 1.7 years; 6 boys and 3 girls).

Results: Children with NAFLD had markedly higher serum concentrations of endotoxin (P < 0.01), PAI-1 (P < 0.001), TNF-α, and IL-6 (P < 0.05) than control subjects. Endotoxin (P = 0.002), PAI-1, (P < 0.001), IL-6 (P = 0.002), TNF-α (P = 0.02), and body mass index (P = 0.03) were significantly associated with a NAFLD activity score ≥5 at the univariate analysis. At the stepwise regression analysis, endotoxin (P < 0.0001) and PAI-1 (P = 0.009) were the most significant predictors for NAFLD activity score.

Conclusions: Our findings demonstrate that, apart from TNF-α and IL-6, endotoxin and PAI-1 may represent good markers of NASH. They also reinforce the hypothesis that elevated levels of endotoxin may contribute to the progression from NAFLD to NASH.

*Liver Unit, Italy

Scientific Directorate, Italy

Pathology Department, Italy

§Molecular Medicine of Bambino Gesù Children's Hospital and Research Institute, Rome, Italy.

Received 15 April, 2009

Accepted 11 October, 2009

Address correspondence and reprint requests to Anna Alisi, PhD, Liver Unit of “Bambino Gesù” Children's Hospital and Research Institute, Piazzale S. Onofrio 4, 00165 Rome, Italy (e-mail: anna.alisi@opbg.net).

Drs Alisi and Manco contributed equally to this article.

This study was funded by “Bambino Gesù” Children's Hospital and Research Institute.

The authors have no conflicts of interest.

Article Outline

Nonalcoholic fatty liver disease (NAFLD) has become the most frequent chronic liver disease in children and adolescents in westernized countries (1–4).

NAFLD encompasses a spectrum of diseases ranging from simple steatosis to nonalcoholic steatohepatitis (NASH), which is defined by the association of lipid accumulation with evidence of cellular damage, inflammation, and different degree of scarring or fibrosis (5). Although the pathogenesis of NAFLD/NASH, both in children and in adults, remains unclear, a “2-hit hypothesis” is considered most likely (6–8). With this theory, increased fatty liver (steatosis) and hyperinsulinemia may play crucial roles in the pathogenesis of insulin resistance as the first step; then a second event or a sum of multiple hits (multiple-hit hypothesis) lead to the development of NASH. They include abnormalities in lipid metabolism, insulin resistance, oxidative stress, and impaired secretion of cytokines and adipocytokines. However, no conclusive data have been obtained so far (9,10).

Inflammation associated with insulin resistance is characterized by an overproduction of a number of proinflammatory cytokines, including tumor necrosis factor (TNF)-α, leptin, resistin, and conversely, reduced secretion of adiponectin, an anti-inflammatory molecule (9,11,12). All of these factors can promote hepatic inflammation (9,11). Recently, it has been observed that morbidly obese subjects have increased levels of circulating lipopolysaccharide (LPS) and LPS-binding protein, which correlate with both a major liver expression of TNF-α and the presence of NASH (13). Interestingly, LPS is known as an inductor of the expression of fibrinolysis inhibitors (eg, PAI-1). Elevated PAI-1 has been correlated with enhanced LPS-induced liver damage and induction of liver inflammatory response (14,15).

The endotoxin (LPS), derived from Gram-negative gut bacteria, LPS-related sensors, such as Toll-like receptor-4 (TLR4), and plasminogen activator inhibitor-1 (PAI-1) can act in the progression from NAFLD to NASH (16–18). A pivotal role of systemic endotoxemia in the NAFLD pathogenesis has been suggested in both human and animal studies. However, some mechanisms remain obscure. In addition, there are no data about a possible association of endotoxin serum levels and correlated factors (ie, PAI-1, TNF-α, and IL-6) with pediatric NAFLD. A better understanding of the roles of these serum molecules and their pathological changes associated with the development of NAFLD in humans is desirable to unravel the mechanisms leading to NASH but also to ameliorate intervention strategies.

Starting from this background, the aim of our study was to determine whether children with NAFLD/NASH present systemic endotoxemia and whether LPS and other proinflammatory molecules, including TNF-α, IL-6, and PAI-1, may represent serum biomarkers of the severity of liver damage.

Back to Top | Article Outline


Patients, Anthropometrics, and Clinical Chemistry

The study protocol was approved by the ethics committee of the “Bambino Gesù” Children's Hospital (Rome, Italy), and written informed consent was obtained from all of the subjects included in this study. Forty consecutive patients diagnosed with NAFLD (27 boys and 13 girls), mean age 11.9 ± 2.8 (SD), seen in our center from April 2008 to March 2009 were included in the study.

NAFLD subjects had no medical records of alcohol consumption (15 g/day ethanol); drug-induced hepatotoxicity; evidence of other forms of liver disease, which were excluded based on viral serology, serum iron-binding capacity, ceruloplasmin, anti-nuclear antibodies, anti-mitochondrial and smooth muscle antibodies, and liver histology; history of taking lipid-lowering drugs or drugs affecting lipid metabolism; and clinical indication of impaired nutritional status. Controls (6 boys and 3 girls) were nonobese healthy children. They were available from a previous study (19). Samples were drawn from fasting patients and controls.

Weight and height were measured using standard procedures (20). Body mass index (BMI) was calculated from the formula (weight in kilograms/height in square meters) and data converted into BMI percentile using the UK 1990 reference data for BMI in childhood (21).

Laboratory tests were performed at the first visit. Levels of transaminases (ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, γ-glutamyltransferase) were measured by a routine clinical chemistry laboratory using a MODULAR analyzer (Hitachi/Roche, Milan, Italy). Plasma glucose was measured in triplicate (Beckman, Fullerton, CA); insulin by a specific radioimmunoassay (MYRIA Technogenetics, Milan, Italy). Triglycerides and total cholesterol were measured spectrophotometrically.

Insulin resistance and sensitivity were determined, respectively, by the homeostatic model assessment insulin resistance (HOMA-IR) using the formula IR = (insulin × glucose)/22.5; and by the insulin sensitivity index (ISI) derived from the oral glucose tolerance test (OGTT) using the formula ISI = (10,000/square root of [fasting glucose × fasting insulin] × [mean glucose × mean insulin during OGTT]) (22,23).

Back to Top | Article Outline

Enzyme-linked Immunosorbent Assay (ELISA)

Blood samples were stored at −70°C until analysis. Endotoxin serum concentration was measured using a commercially available limulus amebocyte lysate chromogenic endpoint assay with a concentration range of 0.01 to 10 EU/mL (Hycult Biotechnology, Uden, The Netherlands). Levels of PAI-1, TNF-α, and IL-6 were determined with a commercial ELISA kit (R&D Systems, Wiensbaden, Germany). Levels of endotoxin, PAI-1, TNF-α, and IL-6 were measured in samples of patients (40) with NAFLD and in control healthy subjects (9) matched for age and sex.

Back to Top | Article Outline

Liver Histology

Biopsies were performed and processed as described elsewhere (24). Liver tissues were immediately fixed in 10% buffered formalin until later analysis. Steatosis, inflammation, hepatocyte ballooning, and fibrosis were scored using the NAFLD Clinical Research Network criteria. Features of steatosis, lobular inflammation, and hepatocyte ballooning were combined to obtained the NAFLD activity score (NAS) and cases with NAS ≥5 were considered diagnostic of NASH (25).

Back to Top | Article Outline

Statistical Analyses

The Shapiro-Wilk W test was applied for determining whether sample data likely derive from a normal-distributed population. Continuous data were reported as mean ± SD, whereas categorical data were reported as counts and percentages. To compare means between groups, Mann-Whitney U test and analysis of variance were performed. Stepwise linear regression analysis was performed using NAS as the dependent variable and endotoxin, PAI-1, TNF-α, IL-6, and BMI as independent variables. Statistical significance was set to a 2-tailed P < 0.05 value. Analysis was performed using SPSS statistical software version 12.0 (SPSS Inc, Chicago, IL).

Back to Top | Article Outline


Patients' Characteristics and Serum Endotoxin Concentration

Anthropometrics and main laboratory and histological parameters for the 40 patients and the 9 controls are reported in Table 1. Steatosis was present in all of the patients and it was of moderate to severe grade in most patients. Inflammation was of mild severity in most (n = 23, 57.5%) children. 62.5% (n = 12) of cases revealed ballooning of the hepatocytes. Stage 1 fibrosis was present in 29 patients (72.5%). 67.5% of children had NASH, and the average NAS was 4.9 ± 1.6 (SD). No patient had established cirrhosis.

As shown in Figure 1A, levels of endotoxin in serum controls (N = 9) ranged from 0.05 to 0.16 EU/mL (average value 0.10 ± 0.04 EU/mL).

In NAFLD (N = 40), serum concentrations of endotoxin ranged from 0.05 to 3.55 EU/mL (average value 0.85 ± 0.9 EU/mL). Nine (22.5%) patients of 40 had levels of endotoxin in the range of normality and 31 (77.5%) had levels above it.

Back to Top | Article Outline

Serum Levels of TNF-α and IL-6

Next, levels of TNF-α and IL-6 were measured in serum samples of control subjects and patients with NAFLD. As shown in Figure 1B, in patients with NAFLD, TNF-α ranged from 3.08 to 28.98 pg/mL, with an average value of 14.20 ± 5.92 pg/mL versus controls (2.3 ± 0.51 pg/mL). The serum IL-6 concentration in patients with NAFLD ranged from 4.90 to 34.09 pg/mL, and as reported in Figure 1C, the mean (12.92 ± 6.55 pg/mL) was greater than in controls (14.4 ± 0.71 pg/mL).

Back to Top | Article Outline

Analysis of PAI-1 Serum Concentration

We also evaluated the serum concentration of PAI-1 in controls and patients with NAFLD. As shown in Figure 1D, the average serum levels of PAI were 345.95 ± 94.04 ng/mL in NAFLD, ranging from 19.51 to 608.24 ng/mL, which was higher than controls (30.10 ± 3.91 ng/mL).

Back to Top | Article Outline

Associations Between Serum Parameters, BMI, and NASH

At the univariate analysis, serum levels of endotoxin (1.18; 0.75–1.60, 95% confidence interval [CI]; P = 0.002), PAI-1 (381; 352.40–410.18, 95% CI; P < 0.001), TNF-α (15.69; 13.89–17.48, 95% CI; P = 0.02), and IL-6 (15.02; 12.30–17.74, 95% CI; P = 0.002), and BMI (93.30; 90.80–95.79, 95% CI; P = 0.028) were the variables that differed significantly in patients with NASH (NAS ≥5) as compared with those without.

We next wondered which parameter would best predict NAS. Stepwise regression analysis, performed with NAS as dependent continuous variable, resulted in 2 models (Tables 2 and 3). In the first model, endotoxin alone explained 47% of the variance in the NAS (P < 0.0001). In the second model, endotoxin and PAI-1 explained together 56% of variance (P < 0.0001). No additional information was gained by modeling for sex, age, waist circumference, AST and ALT levels, insulin resistance, triglycerides, and cholesterol.

Back to Top | Article Outline


Endotoxemia in NAFLD may be a cofactor leading to NAFLD/NASH. This concept is strongly suggested by several studies conducted in both animals and humans (13,26–28). In particular, animal studies demonstrate that LPS in genetically obese rodents with simple fatty liver can induce steatohepatitis, and endotoxin in a nutritional model of NAFLD increases hepatic lipid accumulation and promote development of steatohepatitis (26–28). In human studies, it was reported that obese patients with NAFLD have elevated plasma levels of LPS, which further increase with the severity of liver disease (13).

A possible association between endotoxemia and NASH includes bacterial overgrowth and disruption of intestinal mucosa barrier integrity, which may lead to an increased intestinal permeability, a consequent excessive absorption of endotoxin, an alteration of innate immune response, and liver tissue damage (29). This hypothesis is strongly supported by the high incidence of NASH in obese patients undergoing jejunoileal bypass surgery, a procedure known to be associated with small intestinal bacterial overgrowth (SIBO) and portal endotoxemia (30). In addition, genetically obese mice display enhanced intestinal permeability, which leads to increased portal endotoxemia, which makes hepatic stellate cells more sensitive to bacterial endotoxin (31).

Interestingly, some authors postulate an increased translocation of endotoxin due to disordered intestinal motility as mechanism for NAFLD pathogenesis (16,32). A rich-in-carbohydrate diet may produce ethanol when intestinal stasis favors bacterial overgrowth in the upper parts of the gastrointestinal tract. The increased portal endotoxemia could initiate TLR signaling. Accordingly, the authors found significant correlations between hepatic expression of TLR-4, plasma PAI-1, and endotoxin, even though they are still unable to explain the molecular signaling pathways (32). As occurs in many models of chemical liver inflammation (15), the activation of PAI-1 may be critical for the inflammatory response, which leads to NASH.

Here, we found that the mean value of serum concentration of endotoxin and PAI-1 in pediatric patients with NAFLD was greater, respectively, 8 (P < 0.01) and 11 times (P < 0.0001) than in controls. Furthermore, for the first time our results demonstrate that an increased of gut-derived bacterial products such as endotoxin strongly associates with an increased NAFLD activity score in children, suggesting a potential role of endotoxin as trigger of hepatic necroinflammatory cascades resulting in liver cell injury. Moreover, although the association of PAI-1 and severity of NAFLD histology has been reported in adults by Targher et al (17), our results reinforce the significance of PAI-1 serum levels as a predictor of NASH in children.

Of course, we are aware that clinical significance of the present study is biased by the small number of patients. However, it represents just a preliminary study aimed at verifying our working hypothesis, and novelty in findings must be balanced against difficulties in recruiting a homogeneous sample of such well-characterized children with biopsy-proven NAFLD.

Because the mechanisms involved in the development of NAFLD are not yet fully clarified, therapeutic options are still limited. However, the mechanisms that link serum levels of endotoxin, PAI-1, and pathogenesis of steatohepatitis in children require further investigation and may have significant implications for designing new therapeutic strategies.

Back to Top | Article Outline


1. Ong JP, Younossi ZM. Epidemiology and natural history of NAFLD and NASH. Clin Liver Dis 2007; 11:1–16.
2. Barshop NJ, Sirlin CB, Schwimmer JB, et al. Epidemiology, pathogenesis and potential treatments of paediatric non-alcoholic fatty liver disease [review article]. Aliment Pharmacol Ther 2008; 28:13–24.
3. Patton HM, Sirlin C, Behling C, et al. Pediatric nonalcoholic fatty liver disease: a critical appraisal of current data and implications for future research. J Pediatr Gastroenterol Nutr 2006; 43:413–427.
4. Roberts EA. Pediatric nonalcoholic fatty liver disease (NAFLD): a “growing” problem? J Hepatol 2007; 46:1133–1142.
5. Yeh MM, Brunt EM. Pathology of nonalcoholic fatty liver disease. Am J Clin Pathol 2007; 128:837–847.
6. Schwimmer JB, Behling C, Newbury R, et al. Histopathology of pediatric nonalcoholic fatty liver disease. Hepatology 2005; 42:641–649.
7. Mager DR, Roberts EA. Nonalcoholic fatty liver disease in children. Clin Liver Dis 2006; 10:109–131.
8. Day CP. Genes or environment to determine alcoholic liver disease and non-alcoholic fatty liver disease. Liver Int 2006; 26:1021–1028.
9. Kim CH, Younossi ZM. Nonalcoholic fatty liver disease: a manifestation of the metabolic syndrome. Cleve Clin J Med 2008; 75:721–728.
10. de Alwis NM, Day CP. Non-alcoholic fatty liver disease: the mist gradually clears. J Hepatol 2008; 48(Suppl 1):S104–S112.
11. Manco M, Marcellini M, Giannone G, et al. Correlation of serum TNF-alpha levels and histologic liver injury scores in pediatric nonalcoholic fatty liver disease. Am J Clin Pathol 2007; 127:954–960.
12. Jarrar MH, Baranova A, Collantes R, et al. Adipokines and cytokines in non-alcoholic fatty liver disease. Aliment Pharmacol Ther 2008; 27:412–421.
13. Ruiz AG, Casafont F, Crespo J, et al. Lipopolysaccharide-binding protein plasma levels and liver TNF-alpha gene expression in obese patients: evidence for the potential role of endotoxin in the pathogenesis of non-alcoholic steatohepatitis. Obes Surg 2007; 17:1374–1380.
14. Luyendyk JP, Maddox JF, Green CD, et al. Role of hepatic fibrin in idiosyncrasy-like liver injury from lipopolysaccharide-ranitidine coexposure in rats. Hepatology 2004; 40:1342–1351.
15. Beier JI, Luyendyk JP, Guo L, et al. Fibrin accumulation plays a critical role in the sensitization to lipopolysaccharide-induced liver injury caused by ethanol in mice. Hepatology 2009;49:1545–53.
16. Maher JJ, Leon P, Ryan JC. Beyond insulin resistance: innate immunity in nonalcoholic steatohepatitis. Hepatology 2008; 48:670–678.
17. Targher G, Bertolini L, Scala L, et al. Plasma PAI-1 levels are increased in patients with nonalcoholic steatohepatitis. Diabetes Care 2007; 30:e31–e32.
18. Siebler J, Galle PR, Weber MM. The gut-liver-axis: endotoxemia, inflammation, insulin resistance and NASH. J Hepatol 2008; 48:1032–1034.
19. Nobili V, Marcellini M, Marchesini G, et al. Intrauterine growth retardation, insulin resistance, and nonalcoholic fatty liver disease in children. Diabetes Care 2007; 30:2638–2640.
20. Lohman TG, Roche AF, Martorell R. Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics Books; 1988.
21. Cole TJ, Freeman JV, Preece MA. Body mass index reference curves for the UK, 1990. Arch Dis Child 1995; 73:25–29.
22. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999; 22:1462–1470.
23. Abdul-Ghani MA, Matsuda M, Balas B, et al. Muscle and liver insulin resistance indexes derived from the oral glucose tolerance test. Diabetes Care 2007; 30:89–94.
24. Manco M, Bedogni G, Marcellini M, et al. Waist circumference correlates with liver fibrosis in children with nonalcoholic steatohepatitis. Gut 2008; 57:1283–1287.
25. Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005; 41:1313–1321.
26. Yang SQ, Lin HZ, Lane MD, et al. Obesity increases sensitivity to endotoxin liver injury: implications for the pathogenesis of steatohepatitis. Proc Natl Acad Sci U S A 1997; 94:2557–2562.
27. Dickerson RN, Karwoski CB. Endotoxin-mediated hepatic lipid accumulation during parenteral nutrition in rats. J Am Coll Nutr 2002; 21:351–356.
28. Kirsch R, Clarkson V, Verdonk RC, et al. Rodent nutritional model of steatohepatitis: effects of endotoxin (lipopolysaccharide) and tumor necrosis factor alpha deficiency. J Gastroenterol Hepatol 2006; 21(1 Pt 1):174–182.
29. Wigg AJ, Roberts-Thomson IC, Dymock RB, et al. The role of small intestinal bacterial overgrowth, intestinal permeability, endotoxaemia, and tumour necrosis factor alpha in the pathogenesis of nonalcoholic steatohepatitis. Gut 2001; 48:206–211.
30. Hocking MP, Davis GL, Franzini DA, et al. Long-term consequences after jejunoileal bypass for morbid obesity. Dig Dis Sci 1998; 43:2493–2499.
31. Brun P, Castagliuolo I, Di Leo V, et al. Increased intestinal permeability in obese mice: new evidence in the pathogenesis of nonalcoholic steatohepatitis. Am J Physiol Gastrointest Liver Physiol 2007; 292:G518–G525.
32. Thuy S, Ladurner R, Volynets V, et al. Nonalcoholic fatty liver disease in humans is associated with increased plasma endotoxin and plasminogen activator inhibitor 1 concentrations and with fructose intake. J Nutr 2008; 138:1452–1455.

endotoxin; inflammation; NAFLD; NASH; PAI-1

© 2010 Lippincott Williams & Wilkins, Inc.