TY - JOUR
T1 - Serum metabolomics reveals γ-glutamyl dipeptides as biomarkers for discrimination among different forms of liver disease
AU - Soga, Tomoyoshi
AU - Sugimoto, Masahiro
AU - Honma, Masashi
AU - Mori, Masayo
AU - Igarashi, Kaori
AU - Kashikura, Kasumi
AU - Ikeda, Satsuki
AU - Hirayama, Akiyoshi
AU - Yamamoto, Takehito
AU - Yoshida, Haruhiko
AU - Otsuka, Motoyuki
AU - Tsuji, Shoji
AU - Yatomi, Yutaka
AU - Sakuragawa, Tadayuki
AU - Watanabe, Hisayoshi
AU - Nihei, Kouei
AU - Saito, Takafumi
AU - Kawata, Sumio
AU - Suzuki, Hiroshi
AU - Tomita, Masaru
AU - Suematsu, Makoto
N1 - Funding Information:
This work was supported by Health and Labour Sciences Research Grants “Research on Biological Markers for New Drug Development” (T.S.) and “Research on Risk of Chemical Substances” (T.S.). Additional support was obtained through grants from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) for a Global COE Program entitled “Human Metabolomic Systems Biology” in Life Sciences (T.S., M.T. and M.S.) and the ERATO Gas Biology Project (M.S.), as well as research funds from the Yamagata Prefectural Government and City of Tsuruoka.
PY - 2011/10
Y1 - 2011/10
N2 - Background & Aims: We applied a metabolome profiling approach to serum samples obtained from patients with different liver diseases, to discover noninvasive and reliable biomarkers for rapid-screening diagnosis of liver diseases. Methods: Using capillary electrophoresis and liquid chromatography mass spectrometry, we analyzed low molecular weight metabolites in a total of 248 serum samples obtained from patients with nine types of liver disease and healthy controls. Results: We found that γ-glutamyl dipeptides, which were biosynthesized through a reaction with γ-glutamylcysteine synthetase, were indicative of the production of reduced glutathione, and that measurement of their levels could distinguish among different liver diseases. Multiple logistic regression models facilitated the discrimination between specific and other liver diseases and yielded high areas under receiver-operating characteristic curves. The area under the curve values in training and independent validation data were 0.952 and 0.967 in healthy controls, 0.817 and 0.849 in drug-induced liver injury, 0.754 and 0.763 in asymptomatic hepatitis B virus infection, 0.820 and 0.762 in chronic hepatitis B, 0.972 and 0.895 in hepatitis C with persistently normal alanine transaminase, 0.917 and 0.707 in chronic hepatitis C, 0.803 and 0.993 in cirrhosis type C, and 0.762 and 0.803 in hepatocellular carcinoma, respectively. Several γ-glutamyl dipeptides also manifested potential for differentiating between nonalcoholic steatohepatitis and simple steatosis. Conclusions: γ-Glutamyl dipeptides are novel biomarkers for liver diseases, and varying levels of individual or groups of these peptides have the power to discriminate among different forms of hepatic disease.
AB - Background & Aims: We applied a metabolome profiling approach to serum samples obtained from patients with different liver diseases, to discover noninvasive and reliable biomarkers for rapid-screening diagnosis of liver diseases. Methods: Using capillary electrophoresis and liquid chromatography mass spectrometry, we analyzed low molecular weight metabolites in a total of 248 serum samples obtained from patients with nine types of liver disease and healthy controls. Results: We found that γ-glutamyl dipeptides, which were biosynthesized through a reaction with γ-glutamylcysteine synthetase, were indicative of the production of reduced glutathione, and that measurement of their levels could distinguish among different liver diseases. Multiple logistic regression models facilitated the discrimination between specific and other liver diseases and yielded high areas under receiver-operating characteristic curves. The area under the curve values in training and independent validation data were 0.952 and 0.967 in healthy controls, 0.817 and 0.849 in drug-induced liver injury, 0.754 and 0.763 in asymptomatic hepatitis B virus infection, 0.820 and 0.762 in chronic hepatitis B, 0.972 and 0.895 in hepatitis C with persistently normal alanine transaminase, 0.917 and 0.707 in chronic hepatitis C, 0.803 and 0.993 in cirrhosis type C, and 0.762 and 0.803 in hepatocellular carcinoma, respectively. Several γ-glutamyl dipeptides also manifested potential for differentiating between nonalcoholic steatohepatitis and simple steatosis. Conclusions: γ-Glutamyl dipeptides are novel biomarkers for liver diseases, and varying levels of individual or groups of these peptides have the power to discriminate among different forms of hepatic disease.
KW - Biomarker
KW - Capillary electrophoresis mass spectrometry
KW - Glutathione
KW - Hepatitis C virus
KW - Hepatocellular carcinoma
KW - Metabolomics
KW - Nonalcoholic steatohepatitis
KW - Oxidative stress
KW - γ-Glutamyl dipeptides
UR - http://www.scopus.com/inward/record.url?scp=80051623540&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80051623540&partnerID=8YFLogxK
U2 - 10.1016/j.jhep.2011.01.031
DO - 10.1016/j.jhep.2011.01.031
M3 - Article
C2 - 21334394
AN - SCOPUS:80051623540
SN - 0168-8278
VL - 55
SP - 896
EP - 905
JO - Journal of Hepatology
JF - Journal of Hepatology
IS - 4
ER -