Mitochondrial dysfunction-related lipid changes occur in nonalcoholic fatty liver disease progression
Nonalcoholic fatty liver disease (NAFLD) comprises fat-accumulating conditions within hepatocytes that can cause severe liver damage and metabolic comorbidities. Studies suggest that mitochondrial dysfunction contributes to its development and progression and that the hepatic lipidome changes extensively in obesity and in NAFLD. To gain insight into the relationship between lipid metabolism and disease progression through different stages of NAFLD, we performed lipidomic analysis of plasma and liver biopsy samples from obese patients with nonalcoholic fatty liver (NAFL) or nonalcoholic steatohepatitis (NASH) and from those without NAFLD. Congruent with earlier studies, hepatic lipid levels overall increased with NAFLD. Lipid species that differed with NAFLD severity were related to mitochondrial dysfunction; specifically, hepatic cardiolipin and ubiquinone accumulated in NAFL, and levels of acylcarnitine increased with NASH. We propose that increased levels of cardiolipin and ubiquinone may help to preserve mitochondrial function in early NAFLD, but that mitochondrial function eventually fails with progression to NASH, leading to increased acylcarnitine. We also found a negative association between hepatic odd-chain phosphatidylcholine and NAFLD, which may result from mitochondrial dysfunction-related impairment of branched-chain amino acid catabolism. Overall, these data suggest a close link between accumulation of specific hepatic lipid species, mitochondrial dysfunction, and the progression of NAFLD. Nonalcoholic fatty liver disease (NAFLD) comprises fat-accumulating conditions within hepatocytes that can cause severe liver damage and metabolic comorbidities. Studies suggest that mitochondrial dysfunction contributes to its development and progression and that the hepatic lipidome changes extensively in obesity and in NAFLD. To gain insight into the relationship between lipid metabolism and disease progression through different stages of NAFLD, we performed lipidomic analysis of plasma and liver biopsy samples from obese patients with nonalcoholic fatty liver (NAFL) or nonalcoholic steatohepatitis (NASH) and from those without NAFLD. Congruent with earlier studies, hepatic lipid levels overall increased with NAFLD. Lipid species that differed with NAFLD severity were related to mitochondrial dysfunction; specifically, hepatic cardiolipin and ubiquinone accumulated in NAFL, and levels of acylcarnitine increased with NASH. We propose that increased levels of cardiolipin and ubiquinone may help to preserve mitochondrial function in early NAFLD, but that mitochondrial function eventually fails with progression to NASH, leading to increased acylcarnitine. We also found a negative association between hepatic odd-chain phosphatidylcholine and NAFLD, which may result from mitochondrial dysfunction-related impairment of branched-chain amino acid catabolism. Overall, these data suggest a close link between accumulation of specific hepatic lipid species, mitochondrial dysfunction, and the progression of NAFLD. Nonalcoholic fatty liver disease (NAFLD) is an umbrella term that is used to describe a variety of conditions characterized by a build-up of fat within liver cells. NAFLD can range from mild hepatic steatosis [nonalcoholic fatty liver (NAFL)] to the more severe nonalcoholic steatohepatitis (NASH), and can progress to fibrosis, cirrhosis, and hepatocellular carcinoma. NAFLD is the predominant cause of chronic liver disease in many parts of the world, and is present in 75–90% of adults who are morbidly obese (1.Silverman J.F. O'Brien K.F. Long S. Leggett N. Khazanie P.G. Pories W.J. Norris H.T. Caro J.F. Liver pathology in morbidly obese patients with and without diabetes.Am. J. Gastroenterol. 1990; 85: 1349-1355PubMed Google Scholar, 2.Schwenger K.J.P. Fischer S.E. Jackson T.D. Okrainec A. Allard J.P. Non-alcoholic fatty liver disease in morbidly obese individuals undergoing bariatric surgery: prevalence and effect of the pre-bariatric very low calorie diet.Obes. Surg. 2018; 28: 1109-1116Crossref PubMed Scopus (26) Google Scholar). Given that the liver is a central regulator of whole-body energy homeostasis, it is not surprising that many individuals with NAFLD are also characterized by metabolic complications, including insulin resistance, muscle atrophy, and cardiovascular disease, resulting in increased morbidity and mortality (3.Fotbolcu H. Zorlu E. Nonalcoholic fatty liver disease as a multi-systemic disease.World J. Gastroenterol. 2016; 22: 4079-4090Crossref PubMed Scopus (80) Google Scholar). NAFL is diagnosed when liver fat exceeds 5% of the total liver weight. Lipid accumulation in the liver is regulated by the interplay between the delivery of lipids to the liver and their hepatic uptake, synthesis, oxidation, and secretion of lipids. Alterations in the equilibrium of one or more of these processes can promote hepatic steatosis (4.Meex R.C.R. Watt M.J. Hepatokines: linking nonalcoholic fatty liver disease and insulin resistance.Nat. Rev. Endocrinol. 2017; 13: 509-520Crossref PubMed Scopus (323) Google Scholar). NASH is characterized by hepatic steatosis, hepatocyte ballooning, and lobular inflammation. The mechanisms that trigger hepatic inflammation are poorly understood, and to date, a liver biopsy is the only reliable method to distinguish between the different stages of NAFLD. Several studies have tried to identify lipid biomarkers of NAFL and NASH, but this has proven to be very difficult because the correlations between blood and liver lipid species drastically decrease after the transition from NAFL to NASH (5.Sa R. Zhang W. Ge J. Wei X. Zhou Y. Landzberg D.R. Wang Z. Han X. Chen L. Yin H. Discovering a critical transition state from nonalcoholic hepatosteatosis to nonalcoholic steatohepatitis by lipidomics and dynamical network biomarkers.J. Mol. Cell Biol. 2016; 8: 195-206Crossref PubMed Scopus (25) Google Scholar). We previously used lipidomics to compare serum lipid profiles from excessive chronic drinkers without liver disease to those with advanced alcoholic cirrhosis, and showed that circulating lipids associate with alcoholic liver cirrhosis and represent potential biomarkers for risk assessment (6.Meikle P.J. Mundra P.A. Wong G. Rahman K. Huynh K. Barlow C.K. Duly A.M. Haber P.S. Whitfield J.B. Seth D. Circulating lipids are associated with alcoholic liver cirrhosis and represent potential biomarkers for risk assessment.PLoS One. 2015; 10: e0130346Crossref PubMed Scopus (24) Google Scholar). The aim of the current study was to gain a more comprehensive understanding of the hepatic and plasma lipidomic changes in morbidly obese subjects with NAFL or NASH compared with morbidly obese subjects without NAFLD, and to gain insight into lipid species that are linked to disease progression through the different stages of NAFLD. We performed lipidomic analysis in liver biopsies and plasma samples from 58 subjects classified on the basis of liver histology as either normal (n = 16), steatotic (n = 10), or NASH (n = 32). Our study cohort consisted of 58 subjects with severe obesity (BMI >35) who underwent bariatric surgery. The institutional medical ethical committee of Maastricht University approved the study, and all subjects gave written informed consent before participation in the study. Just before surgery, a fasting venous blood sample was obtained from each subject. During surgery, a fasted liver biopsy was obtained. Part of the liver biopsies were fixed in formalin and embedded in paraffin, and histopathology was performed by H&E staining and Masson's trichrome staining. Subjects were classified as control (CTRL), NAFL, or NASH according to the criteria of both Brunt et al. (7.Brunt E.M. Janney C.G. Di Bisceglie A.M. Neuschwander-Tetri B.A. Bacon B.R. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions.Am. J. Gastroenterol. 1999; 94: 2467-2474Crossref PubMed Scopus (3130) Google Scholar) and Kleiner et al. (8.Kleiner D.E. Brunt E.M. Van Natta M. Behling C. Contos M.J. Cummings O.W. Ferrell L.D. Liu Y.C. Torbenson M.S. Unalp-Arida A. Nonalcoholic Steatohepatitis Clinical Research Network Design and validation of a histological scoring system for nonalcoholic fatty liver disease.Hepatology. 2005; 41: 1313-1321Crossref PubMed Scopus (7113) Google Scholar). The CTRL group was comprised of subjects showing <5% steatotic hepatocytes. The steatotic group was comprised of individuals showing >5% steatotic hepatocytes without significant inflammation, as observed after H&E staining (no intra-acinar inflammatory foci per 20 fields with a 20 ocular and no portal tract inflammation). The third group was comprised of subjects with NASH, who were identified by the presence of steatosis, lobular inflammation, and hepatocellular ballooning. Lipid extraction of the liver and plasma was carried out using an earlier described single-phase chloroform:methanol (2:1) approach, with P.J. Wong G. D. Barlow C.K. M.J. L. A. lipidomic analysis of and Biol. PubMed Scopus Google Scholar). liver were with a in of and underwent on to the lipid a liver or plasma sample was with of and of The samples were and for to the The were to and using the system to lipidomic the lipid was and in of of were The samples were and the the lipids was to for lipidomic lipidomics for the liver and plasma was performed using with an as described earlier with Wong G. Barlow C.K. A. L. J.B. J. lipid in a Lipid PubMed Scopus Google Scholar). was using an system with a with the The system consisted of and with and and 5% with Our was as conditions to to by a to with a to to and The was to conditions and for for a of 20 method all lipid species, including and species, to be in a in the the of the species was on the of a and the were on comprehensive of each using the plasma control we were to the species from the on and the and and the and total of lipid species lipid and were in this study was used for of and data normal analysis we were to current of by their To the of was characterized by either in negative for or species, or by of in for species, and used to of on for species that were of the was performed on lipid of plasma the analysis of of phosphatidylcholine and the of in the extraction and were with of which the to for for J. studies on the PubMed Scopus Google Scholar, J. of and of by with PubMed Scopus Google Scholar). it was not to resulting from we the system of with an and a are as or out of was one data for and were with the The data were using the by the The data were with the the data were with the was The liver and plasma lipidomic data are as per of or to the CTRL group are used to the lipidomic The with the was used to the was with the were as the and a was To the association of the lipid species or to the disease NAFL and was performed between NAFL, and NASH. The was for and a was undergoing bariatric were into CTRL (n = 16), NAFL (n = 10), and NASH (n = as by histological of their liver subjects were morbidly obese and subjects with the of metabolic for and for and in and in assessment of insulin and blood levels were increased in NASH compared with a more levels were in NASH compared with NAFL to NASH, only levels were in individuals with NASH of NAFL, and NASH are as and were using the or in a of NAFL, and NASH are as and were using the or To the hepatic lipidome in individuals with NAFL and NASH, the of each lipid was to the CTRL with CTRL of and ubiquinone were increased in NAFL NASH compared with acylcarnitine and were in the NASH group compared with the NAFL group of excessive energy as is a and a of NAFLD. of the and species were in the NAFL and NASH compared with the CTRL the NASH all but not species were compared with the NAFL group To changes in the of and between was as a of total of each lipid with NASH more and in their compared with the CTRL group the not The of acid in was in with NAFLD the of acid which is in not between the total levels of increased to a as with NAFL and NASH, were between the lipid in of with NAFL and NASH levels of in their and of and and in the of the NAFL and NASH of the levels within the lipid between NAFL, and NASH in each lipid are as and were with the and for with the The with the was the analysis (n = for = for = for in a in each lipid are as and were with the and for with the The with the was the analysis (n = for = for = for and comprised lipid species that were more in the NAFL and NASH compared with the CTRL group the in and were increased in subjects with NASH compared with levels of and were The was also in NAFL individuals compared with CTRL to the and were increased in the NASH group as compared with the CTRL and were The was also in the NAFL group compared with CTRL and was in NASH compared with NAFL to and were only a the and species between species acid were when individuals with NAFL and NASH, including and CTRL and NASH and which both were also in individuals with NASH compared with CTRL significant were not observed in of the odd-chain species, analysis that total odd-chain was associated with NAFLD. The of total odd-chain NASH was and the only and were different between the CTRL and NAFLD both of were in individuals with NASH compared with levels of accumulated with the disease severity between NASH levels of were only increased in the NASH group NASH CTRL and NASH with its lipid the species, was increased in the NAFL and NASH group compared with the species, and were only increased in the NASH The only species that differed between the was which was in the NASH group compared with CTRL dysfunction has to a in the of NAFLD K. A. D. dysfunction in and to PubMed Scopus Google Scholar, C. J. K. C. M. M. of hepatic mitochondrial function in with fatty liver is in 2015; PubMed Scopus Google Scholar). and ubiquinone are lipid that have linked to mitochondrial function of cardiolipin in mitochondrial dysfunction and J. Cell PubMed Scopus Google Scholar, M. E. The of metabolism in nonalcoholic fatty liver disease development and 2015; PubMed Scopus Google Scholar). cardiolipin levels were in the NAFL and NASH as compared with the CTRL group the of cardiolipin was the of the in the cardiolipin was in NASH significant = The of was increased in NASH, the of was all acylcarnitine were increased in individuals with NASH were no in acylcarnitine between when as a of the total showed a acylcarnitine was increased in NASH compared with CTRL significant = NASH acylcarnitine was in NASH CTRL significant = NASH of ubiquinone were increased with NAFL and NASH, compared with CTRL Several earlier studies the potential of NAFLD by plasma lipids K. M. N. C. J. A. K. Circulating portal to NASH in 2015; PubMed Scopus Google Scholar, E. S. J. of NAFLD a lipidomics to an Lipid 2015; PubMed Scopus Google Scholar). of the lipid biomarkers have as or reliable to NAFLD in To the and of NAFLD by plasma we performed plasma lipid using plasma samples from the NAFL, and NASH The were using were the only plasma lipids that were different between more specifically, it was that were increased in NASH CTRL subjects lipid species as potential of of NAFL, and NASH were with lipid species with significant are The and the lipid of species that the lipid is one of the by but with an Lipid species are by the of their in a of NAFL, and NASH were with lipid species with significant are The and the lipid of species that the lipid is one of the by but with an Lipid species are by the of their To plasma and hepatic we performed correlations between the many plasma and hepatic lipid species of the lipid species to and and acylcarnitine showed a and between the plasma and hepatic but correlations were also found for and The hepatic and plasma lipidome changes extensively with the development of obesity and NAFLD, as by earlier lipidomic studies K. M. N. C. J. A. K. Circulating portal to NASH in 2015; PubMed Scopus Google C. Contos M.J. M. Zhou H. The plasma lipidomic of nonalcoholic PubMed Scopus Google Scholar). a lipidomic study not only the and but also low and lipid and of lipids that differed between the NAFL, and NASH are related to mitochondrial dysfunction, a that has to an in the development of NAFLD. study hepatic and plasma lipid profiles are with NAFLD and the of these changes on mitochondrial function and the progression of NAFLD. the current study, we found increased hepatic levels of and in individuals with NAFLD, which is in with earlier studies J. R. L. M. J. in fatty acid in to hepatic steatosis in patients with fatty liver PubMed Scopus Google Scholar, J. C. Contos M.J. lipidomic analysis of nonalcoholic fatty liver disease.Hepatology. PubMed Scopus Google Scholar). the fatty of these lipid was to earlier was a decrease in in and an in within and an accumulation of in the NAFL NASH species, those with acid in the NAFLD which is in with studies an increased in NAFLD J. R. L. M. J. in fatty acid in to hepatic steatosis in patients with fatty liver PubMed Scopus Google Scholar, J. C. Contos M.J. lipidomic analysis of nonalcoholic fatty liver disease.Hepatology. PubMed Scopus Google Scholar). Several studies suggest that mitochondrial dysfunction a in the development of advanced NAFLD. are in per D. J. N. R. M. A. C. A. and in chronic liver PubMed Scopus Google and mitochondrial associated with NAFLD and C. J. K. C. M. M. of hepatic mitochondrial function in with fatty liver is in 2015; PubMed Scopus Google Scholar, of in the of nonalcoholic Biol. PubMed Scopus Google Scholar). an study by et al. C. J. K. C. M. M. of hepatic mitochondrial function in with fatty liver is in 2015; PubMed Scopus Google it was that individuals with NAFL hepatic mitochondrial this was in individuals with NASH. The that increased lipid in the liver of individuals with NAFL hepatic mitochondrial and to NAFLD increased may not be to and may promote excessive hepatic hepatocellular these mechanisms mitochondrial resulting in the development of NASH C. J. K. C. M. M. of hepatic mitochondrial function in with fatty liver is in 2015; PubMed Scopus Google Scholar). and acylcarnitine an a in mitochondrial is a that is for mitochondrial function of cardiolipin in mitochondrial dysfunction and J. Cell PubMed Scopus Google Scholar). is in the mitochondrial it with and mitochondrial and a in and M. for the of and mitochondrial Lipid PubMed Scopus Google Scholar). is also for mitochondrial and the of K. C. H. Biol. PubMed Scopus Google Scholar). also as is in of mitochondrial 2017; PubMed Scopus Google Scholar). is in the of mitochondrial as transition and the of and it as an and a and from M. E. The of metabolism in nonalcoholic fatty liver disease development and 2015; PubMed Scopus Google Scholar). is an which is through the mitochondrial and its fatty group for or insulin PubMed Scopus Google Scholar). Our study that hepatic cardiolipin and ubiquinone accumulated in NAFL, but not with NASH. et al. R. P.J. M.S. metabolic in liver of and cardiolipin PubMed Scopus Google Scholar) also observed increased levels of hepatic cardiolipin and ubiquinone in and both of also a of liver to the of mitochondrial a that is by and that to mitochondrial and The that the in cardiolipin and ubiquinone were processes mitochondrial function R. P.J. M.S. metabolic in liver of and cardiolipin PubMed Scopus Google Scholar). we propose that the increased levels of cardiolipin and ubiquinone observed are of mitochondrial which may be to or mitochondrial function in to the excessive energy hepatic insulin of this studies in with NAFLD found increased liver cardiolipin levels M. G. J. J.P. C. C. fatty liver mitochondrial cardiolipin and its fatty acid in with nonalcoholic fatty liver PubMed Scopus Google Scholar, G. S. J. E. C. C. of hepatic steatosis and liver mitochondrial and and their in a J. 2018; PubMed Scopus Google and these levels to in of mitochondrial potential and G. S. J. E. C. C. of hepatic steatosis and liver mitochondrial and and their in a J. 2018; PubMed Scopus Google Scholar). increased mitochondrial has previously observed in patients with hepatic steatosis C. J. K. C. M. M. of hepatic mitochondrial function in with fatty liver is in 2015; PubMed Scopus Google Scholar). cardiolipin has also to have on is to to its close association with which may result in of the that to G. M. G. species from the mitochondrial from cardiolipin in the PubMed Scopus Google Scholar, G. G. cardiolipin and mitochondrial dysfunction in nonalcoholic fatty liver disease.World J. Gastroenterol. PubMed Scopus Google Scholar). cardiolipin is associated with mitochondrial dysfunction in including NAFLD G. S. J. E. C. C. of hepatic steatosis and liver mitochondrial and and their in a J. 2018; PubMed Scopus Google G. G. M. species mitochondrial through cardiolipin PubMed Scopus Google Scholar). We also observed an in cardiolipin and a decrease in in cardiolipin have previously to mitochondrial function by the are found in both and in both and have also linked to a of and associated with mitochondrial dysfunction of cardiolipin in mitochondrial dysfunction and J. Cell PubMed Scopus Google Scholar, M. D. C. A. S. W. The of in cardiolipin PubMed Scopus Google Scholar, M. M. Y. in mitochondrial 2005; PubMed Scopus Google Scholar). is also that the of hepatic ubiquinone was observed for cardiolipin and cardiolipin has to promote between and it is that the changes in cardiolipin and ubiquinone an for M. A. C. L. J. M. K. between ubiquinone and to 2017; Scholar). of acylcarnitine were increased in individuals with NASH, but not in individuals with NAFL, as compared with the CTRL The accumulation of acylcarnitine is to be a of mitochondrial mitochondrial dysfunction, and M. A. G. J. hepatic mitochondrial in patients with nonalcoholic PubMed Scopus Google Scholar, N. L.D. amino acid metabolism profiles in nonalcoholic fatty liver 2015; PubMed Scopus Google Scholar, S. M. J. in steatohepatitis an in acid J. Endocrinol. 2016; PubMed Scopus Google Scholar). Our that acylcarnitine accumulation mitochondrial dysfunction and is a of NASH. Our data also the of et al. C. J. K. C. M. M. of hepatic mitochondrial function in with fatty liver is in 2015; PubMed Scopus Google that increased lipid in the liver hepatic mitochondrial to NAFLD We suggest that levels of cardiolipin and ubiquinone are increased in individuals with NAFLD, in an to mitochondrial but that mitochondrial function eventually leading to increased levels of acylcarnitine and NASH We no of mitochondrial function in study it was not to link lipid to mitochondrial be the of We observed a negative association between hepatic odd-chain and NAFLD. that a of odd-chain can be from branched-chain amino and M. amino acid and Biol. 2016; PubMed Scopus Google Scholar, N. of amino contributes to of odd-chain and fatty in One. 2015; 10: PubMed Scopus Google Scholar). the of in in amino in metabolic and insulin resistance.Nat. Rev. Endocrinol. 10: PubMed Scopus Google the odd-chain of low is a of mitochondrial We propose that the impairment of to hepatic mitochondrial dysfunction may be for the negative association between odd-chain and NAFLD. of this et al. N. Z. C. C. K. Z. species and of and normal mitochondrial in PubMed Scopus Google Scholar) previously from the muscle of and specific in in and mitochondrial that are for and metabolism before into the acid as were in of obese The that this to the accumulation of that is associated with the of insulin N. Z. C. C. K. Z. species and of and normal mitochondrial in PubMed Scopus Google Scholar). Our is by studies showing increased hepatic levels in individuals with NAFLD S. M. J. in steatohepatitis an in acid J. Endocrinol. 2016; PubMed Scopus Google Scholar, E. M. A. M. of fatty liver in and and of the 2017; PubMed Scopus Google Scholar, M. N. D. C. E. H. G. X. branched-chain amino metabolic obese with fatty liver 2017; PubMed Scopus Google Scholar). study potential the low circulating odd-chain and observed in NAFLD and patients H. P.J. M. Barlow C.K. J.B. C. T.D. in of phosphatidylcholine species with metabolic in PubMed Scopus Google Scholar, A. J. M. in the association between plasma fatty and the Endocrinol. PubMed Scopus Google Scholar, X. Zhou D. E. S. branched-chain fatty and insulin of obesity and 2015; PubMed Scopus Google Scholar). we not an in in NAFLD, data suggest of the as by accumulation of is a of increased in NAFLD liver M. J.P. of in nonalcoholic fatty liver Endocrinol. PubMed Scopus Google Scholar). study also observed increased levels of species in obese individuals with NASH compared with control and a association between total hepatic levels and hepatic and inflammation M. R. C. S. E. C. D. hepatic to insulin resistance, and inflammation in nonalcoholic 2018; 41: PubMed Scopus Google Scholar). the increased levels of can species and S. N. A. the of 2015; PubMed Scopus Google Scholar, S. S. is a that and in hepatic steatosis 2017; PubMed Scopus Google which can cause S. S. is a that and in hepatic steatosis 2017; PubMed Scopus Google Scholar). is that increased may to the progression of NAFLD and related metabolic We also observed increased levels of in individuals with NASH, but not with We previously a association between circulating and alcoholic liver cirrhosis (6.Meikle P.J. Mundra P.A. Wong G. Rahman K. Huynh K. Barlow C.K. Duly A.M. Haber P.S. Whitfield J.B. Seth D. Circulating lipids are associated with alcoholic liver cirrhosis and represent potential biomarkers for risk assessment.PLoS One. 2015; 10: e0130346Crossref PubMed Scopus (24) Google and increased levels of hepatic have in hepatocellular compared with the hepatic Y. H. M. J. H. hepatocellular is associated with and changes in PubMed Scopus Google Scholar). studies showed that of insulin and steatosis H. M. Zhang J. C. S. control and insulin in of PubMed Scopus Google Scholar, H. M. Zhang J. J. D. C. hepatic steatosis in obese PubMed Scopus Google Scholar, N. M. C. M. R. N. J. of metabolism hepatic insulin and hepatic steatosis in PubMed Scopus Google Scholar). a for of in the transition from NAFLD to NASH, and hepatocellular carcinoma. lipidomic study an of the hepatic lipidome and the between mitochondrial and lipids and mitochondrial We the lipidomic changes the transition from NAFL to NASH. is for understanding the of this We propose that cardiolipin and ubiquinone levels in NAFL to mitochondrial as a to the increased levels of acylcarnitine in NASH are of mitochondrial dysfunction with the progression of The negative association between odd-chain and NAFLD is which mitochondrial catabolism. The of increased and with NAFLD an to understanding of metabolism in NAFLD. Our the to studies understanding the of these lipid in NAFLD The to and for with the lipidomic with branched-chain amino acid control assessment of insulin nonalcoholic fatty liver nonalcoholic fatty liver disease nonalcoholic steatohepatitis phosphatidylcholine