Secondary Logo

Journal Logo

Original Articles: Hepatology

Lipidomics in Nonalcoholic Fatty Liver Disease: Exploring Serum Lipids as Biomarkers for Pediatric Nonalcoholic Fatty Liver Disease

Draijer, Laura G.∗,†,‡; Froon-Torenstra, Denise; van Weeghel, Michel§,||; Vaz, Frédéric M.§,||; Bohte, Anneloes E.; Holleboom, Adriaan G.#; Benninga, Marc A.; Koot, Bart G.P.

Author Information
Journal of Pediatric Gastroenterology and Nutrition: October 2020 - Volume 71 - Issue 4 - p 433-439
doi: 10.1097/MPG.0000000000002875

Abstract

What Is Known/What Is New

What Is Known

  • Current screening tools have mediocre accuracy in detecting steatosis in children with obesity.
  • Lipidomic studies in adults showed that lipid profiles significantly differed in patients with nonalcoholic fatty liver disease compared with healthy controls.

What Is New

  • In children with obesity, nonalcoholic fatty liver disease is associated with alterations in alkyl-diacylglycerols (TG[O]), phosphatidylethanolamine (PE), alkyl/alkenyl-phosphatidylethanolamine (PE[O]), alkyl/alkenyl-lysophosphatidylethanolamine (LPE[O]) and alkyl/alkenyl-phosphatidylcholine (PC[O]) classes and with alterations in 12 individual lipid species, suggesting that lipid classes and species are potential biomarkers for steatosis.
  • Increased lipid species in children with nonalcoholic fatty liver disease contain mostly monounsaturated and saturated fatty acids.

Concurrent with the rise in obesity, nonalcoholic fatty liver disease (NAFLD) has become the most prevalent chronic liver disease in children in industrialized countries (1). The pooled prevalence of NAFLD in studies in children with obesity is 34.2% (95% confidence interval [CI]: 27.8%--41.2%) (2). Although simple steatosis (‘NAFL’) is the most common and reversible form in children, presence of liver fibrosis at diagnosis has been reported in 16% to 17% (3,4) and progression to cirrhosis and liver failure has been reported in children (5,6). In addition, NAFLD increases the risk of type 2 diabetes mellitus and atherosclerotic cardiovascular disease in adulthood (7). Although it remains to be established, children may be at higher risk to develop these long-term liver and nonliver-related complications in their lifetime, simply as they have more life years ahead of them than adults. Due to these health-related risks of NAFLD in children, it is of great importance to diagnose NAFLD timely in order to start treatment and thereby prevent unfavorable health outcomes (8).

Current guidelines propagate screening for NAFLD in children with obesity using liver transaminases or ultrasonography as primary screening tools (9,10). Both are considered markers of steatosis but lack accuracy in children (11,12). Therefore, there is an urgent need for an accurate, noninvasive, and easily available diagnostic tool for detecting this endemic disorder. Magnetic Resonance Imaging techniques have a high performance in detecting and grading steatosis in children but are poorly suited as screening tests for such a prevalent disorder because of limited availability and high costs (13).

Disturbances in lipid metabolism have a close bidirectional relation with NAFLD: accumulation of toxic lipid classes, such as diacylglycerol (DG) and lysophosphatidylcholine (LPC) may drive intrahepatic lipotoxicity and steatohepatitis (14); steatotic insulin-resistant livers have increased secretion of triglyceride-rich very-low density lipoprotein (VLDL)-cholesterol, which drives the mixed hyperlipidemia often observed in patients with NAFLD (15). Lipidomics is an analytical technique that is increasingly used to survey the lipidome of tissues and body fluids for discovery of noninvasive biomarkers to improve diagnosis of complex diseases (16). Although lipidomic analyses are not yet widely available, the field of lipidomics is expected to follow a trajectory comparable with the developments seen in genomics over the past decade (17). Previous studies in adults showed that lipid profiles significantly differed in patients with NAFLD compared with healthy controls (18–21). Therefore, this technique seems promising for the development of new noninvasive diagnostic biomarkers for NAFLD, yet data from pediatric patients with NAFLD are lacking. We therefore used a lipidomics platform that allows the identification and semiquantitative determination of >800 lipids to identify plasma lipids that are correlated with the presence of steatosis in children with obesity.

METHODS

Study Design and Population

For this pilot study, 21 children with obesity and steatosis were matched with 21 children with obesity without steatosis. Matching was based on age, gender, and body mass index (BMI). Children were selected from a cohort of 119 children with obesity who participated in a study that evaluated the efficacy of lifestyle interventions in an obesity clinic in the Netherlands. The design and inclusion criteria have been previously reported (22). In all children, presence of hepatic steatosis was determined using proton magnetic resonance spectroscopy (1H-MRS). Steatosis was defined as >1.8% fat fraction on 1H-MRS, which corresponds with >5% fat containing hepatocytes on liver histology in a validation study using a similar 1H-MRS setting (23). The study was approved by the Medical Ethics Committee of the Amsterdam UMC, location AMC. Each participant and/or its legal guardians provided written informed consent.

Sample Collection and Lipidomics

Venous blood was collected after an overnight fast. Serum was stored at −80 °C until lipidomics was performed. Blood collection, anthropometric measurements, and MR scanning were carried out on the same day. Lipids were extracted from serum in the presence of internal standards (IS) by Folch method as described previously (24). In short, 2 ultra-high performance liquid chromatography (UHPLC) separations, using a reversed phase and a normal phase column, were analyzed in positive and negative ionization mode resulting in 4 data files per sample. A defined amount of internal standards dissolved in 120 μL of chloroform/methanol (1 : 1, v/v), and 1.5 mL of chloroform/methanol (1 : 1, v/v) was added to 20 μL serum. The internal standard mixture consisted of: 0.5 nmol of DG(14:0)2, 0.5 nmol of TG(14:0)2, 0.5 nmol of CE(14:0), 0.1 nmol of CL(14:0)4, 0.2 nmol of BMP(14:0)2, 2.0 nmol of PC(14:0)2, 0.1 nmol of PG(14:0)2, 5.0 nmol of PS(14:0)2, 0.5 nmol of PE(14:0)2, 0.5 nmol of PA(14:0)2, 0.5 nmol of PI(8:0)2, 2.0 nmol of SM(d18:1/12:0), 0.02 nmol of LPG(14:0), 0.1 nmol of LPE(14:0), 0.5 nmol of LPC(14:0), 0.1 nmol of LPA(14:0) (purchased from Avanti Polar Lipids, Alabaster, AL). This technique enables to identify >1200 lipid species depending on the type of matrix used (24).

Bioinformatics for Lipid Identification

The raw LC/MS data were converted to mzXML format using MSConvert (25). The dataset was processed using an in-house developed lipidomics pipeline written in the R programming language 3 (R Foundation for Statistical Computing, Vienna, Austria, www.r-project.org). In brief, it consisted of the following steps: pre-processing using the R package XCMS (26) with minor changes to some functions in order to better suit the Q Exactive data; notably, the definition of noise level in centWave was adjusted and the stepsize in fillPeaks; identification of metabolites using an in-house database of (phospho)lipids, with known internal standards indicating the position of most of the lipid clusters, matching m/z values within 3 ppm deviation; isotope correction to obtain deconvoluted intensities for overlapping peak groups; normalization on the intensity of the internal standard for lipid classes for which an internal standard was available and scaling on measured protein content per sample.

Data Analyses and Visualizations

Analyses were done with R version 3.5.1 and Bioconductor version 3.7. Principal component analyses were performed using the R package MixOmics. Differential abundance of lipids or lipid classes was calculated using Student t test, with significance cutoff set at P < 0.05. Unless implemented through an aforementioned R package or base R graphics, visualization of data was performed using ggplot2 (27).

Statistical Analysis

Patients’ demographic data, laboratory, and imaging data were described using standard descriptive statistics and were compared using a Student t test or a Mann-Whitney U test (continuous variables) or Chi-square test (categorical variables). In the lipidomics analysis, summary data for the lipid classes were calculated. A Student t-test was used to compare lipid classes and lipid species between the 2 groups. A P-value of 0.05 or less was considered significant. P-values were corrected for multiple testing per lipid class by applying a Benjamini & Hochberg correction with FDR <5%. A partial least squares (PLS) regression analysis was performed to identify common factors between the 2 groups. This was combined with a Discriminant Analysis (PLS-DA), to examine whether significant differences exist among the groups, in terms of the underlying metabolite intensities. From the PLS-DA, the variable influence on projection (VIP) score was calculated. The VIP score estimates the importance of each variable in the in PLS model. A variable with a VIP score close to or greater than 1 is considered important (28). A heatmap was created by calculating the z score of each individual sample in a row (metabolite). z = (X [value of the individual sample] − μ [average of the row])/σ (standard deviation of the row). This is a graphical representation of data where the individual values contained in the sample set are represented as colors. A volcano plot was created to quickly identify the most meaningful changes in this large data set composed of replicate data points between subjects with steatosis and controls. It plots significance versus change in concentration on the y-axis and x-axis, respectively.

RESULTS

Patients

Clinical characteristics are summarized in Table 1. In both groups, median age was 15 years (range 10–18) and 57% were boys. Children with steatosis were more insulin-resistant (higher homeostatic assessment for insulin resistance; HOMA-IR) and had significantly higher alanine aminotransferase (ALT) levels, total cholesterol, and triglycerides compared with controls. The median fat fraction measured by 1H-MRS was 9% (interquartile range [IQR 5.7–13.6%]) in children with steatosis versus 1% (IQR 0.6–1.3%) in children without steatosis.

TABLE 1
TABLE 1:
Clinical characteristics

Lipidomics

A total of 18 lipid classes constituting 839 different lipid species were identified and analyzed. When looking at the summations of the individual species within the 18 lipid classes (total level of a particular lipid class), there was an overall significant increase in NAFLD of alkyl-diacylglycerols (TG[O]) and phosphatidylethanolamine (PE) and a significant decrease of different etherphospholipid classes namely alkyl/alkenyl-phosphatidylethanolamine (PE[O]), alkyl/alkenyl-lysophosphatidylethanolamine (LPE[O]) and alkyl/alkenyl-phosphatidylcholine (PC[O]) classes (Fig. 1).

FIGURE 1
FIGURE 1:
Total levels of the different major lipid classes. Boxplots of the summed subspecies of the different major lipid classes (y-axis is a log10 scale). Boxes show median and 25th and 75th percentiles. TG[O] and PE were significantly increased and LPC[O], LPE[O] and PC[O] were significantly decreased in children with steatosis compared with children without steatosis. DG = diacylglycerol; DG[O] = alkylacylglycerol; LPA = lysophosphatic acid; LPC = lysophosphatidylcholine; LPC[O] = alkyl/alkenyl-lysophosphatidylcholine; LPE = lysophosphatidylethanolamine; LPE[O] = alkyl/alkenyl-lysophosphatidylethanolamine; PA = phosphatidic acid; PC = phosphatidylcholine; PC[O] = alkyl/alkenyl-phosphatidylcholine; PE = phosphatidylethanolamine; PE[O] = alkyl/alkenyl-phosphatidylethanolamine; PI = phosphatidylinositol; PS = phosphatidylserine; SM(d) = sphingomyelin; SM(t) = hydroxysphingomyelin; TG = triacylglycerol; TG[O] = alkyldiacylglycerol.

When looking at individual lipid species, the heatmap in Supplemental Digital Content 1 (http://links.lww.com/MPG/B900) shows the top 30 most changed individual lipid species based on their uncorrected P-value. After FDR <5% limitation, 12 lipid species remained significant, which are highlighted in this figure. These 12 lipids constituted 2 sphingomyelin (SM) species [SM(d39:0) and SM(d41:0), 2 lysophosphatidylethanolamine (LPE) species (LPE(20:3) and LPE(22:5) and 8 TG[O] species (TG(O-52:0), (TG(O-52:1) TG(O-52:2), TG(O-54:1), TG(O-54:2), TG(O-52:3), TG(O-56:1) and TG(O-56:2)], all with increased levels in children with hepatic steatosis compared with controls (Table 2, Supplemental Digital Content 2, http://links.lww.com/MPG/B900). The PLS-DA that was used to calculate the VIP score of each lipid in this table is provided in Supplemental Digital Content 3 (http://links.lww.com/MPG/B900).

TABLE 2
TABLE 2:
Significant lipid species

The volcano plot analysis presented in Figure 2 shows that increased lipid species in subjects with steatosis mostly contain monounsaturated and saturated fatty acids.

FIGURE 2
FIGURE 2:
Volcano plot. Volcano plot depicting the lipidomics data. Significance cut off is shown in the legend of the volcano plot. The changed lipid species with P value <0.01 are labeled. The y-axis plots significance level. The 3 horizontal dotted lines indicate P values 0.05, 0.01, and 0.001, respectively. The x-axis plots change of lipid species. The 2 vertical dotted lines indicate log2(fold change) −2 and 2, respectively. PC = phosphatidylcholine; SM(d) = sphingomyelin; TG = triacylglycerol; TG[O] = alkyldiacylglycerol.

The complete dataset can be found in Supplemental Digital Content 4 (http://links.lww.com/MPG/B901).

DISCUSSION

In this case-control study in children with obesity, we investigated the relation between plasma lipids, assessed by lipidomics, and NAFLD. To the best of our knowledge, no study on this has previously been published. In children with obesity and hepatic steatosis compared with children with obesity without steatosis, we found an overall significant increase in TG[O] and PE and a significant decrease of species from the PC[O], LPC[O], and LPE[O] classes. In addition, 12 individual lipid species of 3 lipid classes were significantly elevated in children with hepatic steatosis.

The pathogenesis of NAFLD is complex, multifactorial, and only partially understood (29). The onset of the disease is characterized by an increased influx of free fatty acids derived from adipose tissue lipolysis under insulin-resistant conditions, diet, and de novo hepatic lipogenesis. Subsequently in some patients, lipotoxic insults related to the lipid overload can cause oxidative stress, mitochondrial dysfunction, and activation of inflammatory pathways, thus causing steatohepatitis (NASH) and fibrosis. In adults, alterations in several lipid species have been linked to the development of both NAFL and NASH (30).

An interesting finding in our study is that most altered lipid species are from ether lipid classes that contain an alkyl or alkenyl-substituent at the sn-1 position of the glycerol backbone. Previous studies confirmed this finding in plasma of adults with NAFLD (31) with a more pronounced effect in NASH (32). In addition, our study showed higher levels of certain sphingolipids and PE species, and lower levels of selected PC species. Similar changes in this cluster of lipids were found by Gorden et al (32) when comparing adults with histologically normal livers to those with all stages of NAFLD. Many of these lipid species showed overlap with analytes from liver tissue, which suggests that plasma reflects lipid changes in the liver affected by NAFLD. These changes have also been consistently reported by other studies that evaluated liver and serum samples of adults with NAFLD compared with healthy living liver donors (33) and to adults with severe obesity (31,34,35). Ether-phospholipids (PC[O] and PE[O]) are mainly produced by the liver (36). The described alterations in these lipids could be caused by a decreased dietary choline and ethanolamine intake in those with steatosis, as these are essential components for the ether-phospholipid synthesis (37). An alternative explanation could be that the increased saturated/mono-unsaturated fatty acids are derived from de novo synthesis and that this, in combination with a diet high in saturated fats and low in polyunsaturated fatty acids, leads to a depletion of essential fatty acids, such as docosahexaenoic acid (DHA) and arachidonic acid (AA). As ether-phospholipids are enriched with DHA and AA, this could result in lower levels of ether-phospholipids (38) and accumulation of these more saturated fatty acids in TG[O] species resulting in increased TG[O] levels.

Similar to our study, increased sphingomyelin species in both liver and serum in patients with NAFLD were also found in the study of Apostolopoulou et al (21) who analyzed 21 adults with obesity. Sphingomyelin belongs to the main class of sphingolipids, which are mainly produced in the liver and are components of the cell membrane. They function as signalling molecules that regulate cell growth, survival, and immune responses (39). Sphingolipids have been shown to play a role in lipid-induced insulin resistance and were found to be increased in patients with NAFLD (40,41). However, increased sphingolipids have also been associated with other liver diseases, such as intrahepatic cholestasis (42), and might therefore not be specific for NAFLD.

Increased LPE was also found by Gorden et al (32) when steatotic liver tissue was compared with normal liver tissue, but not in NASH. Only few other studies reported on LPE in NAFLD but all showed a decrease of LPE in NAFLD (41,43). LPE results from partial hydrolysis of PE and is a minor constituent of the cell membrane. The exact function of LPE and its role in NAFLD, however, remains unknown (43).

Several lipidomic studies in adults showed increased TG and DG in plasma of steatotic subjects (20,31,44). This is likely because of increased secretion of VLDL, packing fatty acids that were released from autophagy of lipid droplets in the steatotic liver, or derived from up-regulated de novo lipogenesis, well known to be increased in NAFLD (45). Although we observed a trend towards increased TG species in children with NAFLD in our study and found 51 TG species that were significantly increased, none of them remained significant after correction for multiple testing. This is likely because of the large number of TG species that were tested.

Another interesting observation in our study is a shift towards lipid species containing saturated or monounsaturated fatty acids, in particular in TG and TG[O] in children with NAFLD, as shown in the Volcano plot analysis (Fig. 2). This finding is consistent with previous studies that evaluated steatotic liver tissue (18,46) and plasma samples of patients with NAFLD (19,31,34). This could reflect the high insulin levels driving de novo lipogenesis as this produces saturated fatty acids (47). The reached limit of the cell's capacity to store saturated fatty acids as TG, might explain the observed shift in plasma. Furthermore, the production of monounsaturated fatty acids is driven by the activity of steroyl Coenzyme A (CoA) desaturase, an enzyme that is dependent on sterol response element binding protein-1 (SREBP-1) (48). Previous studies showed increased steroyl CoA desaturase and SREBP-1 activation in NAFLD because of hyperinsulinemia (49).

The strength of this study is the inclusion of patients who had only obesity and were not referred to a liver clinic because of liver abnormalities (ie, elevated liver enzymes), resulting in a cohort that reflects the true population in need for NAFLD screening. Furthermore, the HPLC-MS used for lipidomic analysis provides an outright overview of the human lipidome and the use of a validated 1H-MRS setting allowed to detect steatosis with a high accuracy. A limitation is the small sample size as this was a pilot study. However, even with a small number of patients we found significant changes in NAFLD that need to be further evaluated in studies with larger cohorts including those with different ethnicities and age. Finally, this study does not relate lipid profiles to disease severity, that is NASH and fibrosis as we do not have histopathological data. The latter would allow to study lipids as markers of disease severity and could provide insights in the pathophysiology of NAFLD.

In conclusion, in this pilot study, we observed an overall increase of TG[O] and PE species and a decrease of PC[O], LPC[O], and LPE[O] species in children with obesity with NAFLD compared with children with obesity without NAFLD. We identified significant alterations in 12 individual lipid species of 3 lipid classes in children with NAFLD. Furthermore, we observed a shift towards lipid species containing saturated or monounsaturated fatty acids in children with NAFLD. These findings suggest that lipid species and classes are potential biomarkers for steatosis. Larger lipidomic studies are needed to determine the diagnostic value of these single lipids or a lipid panel as markers for steatosis.

REFERENCES

1. Strauss RS, Barlow SE, Dietz WH. Prevalence of abnormal serum aminotransferase values in overweight and obese adolescents. J Pediatr 2000; 136:727733.
2. Anderson EL, Howe LD, Jones HE, et al. The prevalence of non-alcoholic fatty liver disease in children and adolescents: a systematic review and meta-analysis. PLoS One 2015; 10:e0140908.
3. Schwimmer JB, Newton KP, Awai HI, et al. Paediatric gastroenterology evaluation of overweight and obese children referred from primary care for suspected non-alcoholic fatty liver disease. Aliment Pharmacol Ther 2013; 38:12671277.
4. Nobili V, Reale A, Alisi A, et al. Elevated serum ALT in children presenting to the emergency unit: relationship with NAFLD. Dig Liver Dis 2009; 41:749752.
5. Molleston JP, White F, Teckman J, et al. Obese children with steatohepatitis can develop cirrhosis in childhood. Am J Gastroenterol 2002; 97:24602462.
6. Feldstein AE, Charatcharoenwitthaya P, Treeprasertsuk S, et al. The natural history of non-alcoholic fatty liver disease in children: a follow-up study for up to 20 years. Gut 2009; 58:15381544.
7. Armstrong MJ, Adams LA, Canbay A, et al. Extrahepatic complications of nonalcoholic fatty liver disease. Hepatology 2014; 59:11741197.
8. Vilar-Gomez E, Martinez-Perez Y, Calzadilla-Bertot L, et al. Weight loss through lifestyle modification significantly reduces features of nonalcoholic steatohepatitis. Gastroenterology 2015; 149:367.e5378.e5.
9. Vajro P, Lenta S, Socha P, et al. Diagnosis of nonalcoholic fatty liver disease in children and adolescents: position paper of the ESPGHAN Hepatology Committee. J Pediatr Gastroenterol Nutr 2012; 54:700713.
10. Vos MB, Abrams SH, Barlow SE, et al. NASPGHAN Clinical Practice Guideline for the diagnosis and treatment of nonalcoholic fatty liver disease in children. J Pediatr Gastroenterol Nutr 2017; 64:319334.
11. Draijer LG, Feddouli S, Bohte AE, et al. Comparison of diagnostic accuracy of screening tests ALT and ultrasound for pediatric non-alcoholic fatty liver disease. Eur J Pediatr 2019; 178:863870.
12. Bohte AE, Koot BG, van der Baan-Slootweg OH, et al. US cannot be used to predict the presence or severity of hepatic steatosis in severely obese adolescents. Radiology 2012; 262:327334.
13. Awai HI, Newton KP, Sirlin CB, et al. Evidence and recommendations for imaging liver fat in children, based on systematic review. Clin Gastroenterol Hepatol 2014; 12:765773.
14. Friedman SL, Neuschwander-Tetri BA, Rinella M, et al. Mechanisms of NAFLD development and therapeutic strategies. Nat Med 2018; 24:908922.
15. Choi SH, Ginsberg HN. Increased very low density lipoprotein (VLDL) secretion, hepatic steatosis, and insulin resistance. Trends Endocrinol Metab 2011; 22:353363.
16. Wu Z, Shon JC, Liu KH. Mass spectrometry-based lipidomics and its application to biomedical research. J Lifestyle Med 2014; 4:1733.
17. Wenk MR. Lipidomics: new tools and applications. Cell 2010; 143:888895.
18. Gorden DL, Ivanova PT, Myers DS, et al. Increased diacylglycerols characterize hepatic lipid changes in progression of human nonalcoholic fatty liver disease; comparison to a murine model. PLoS One 2011; 6:e22775.
19. Puri P, Wiest MM, Cheung O, et al. The plasma lipidomic signature of nonalcoholic steatohepatitis. Hepatology 2009; 50:18271838.
20. Mayo R, Crespo J, Martinez-Arranz I, et al. Metabolomic-based noninvasive serum test to diagnose nonalcoholic steatohepatitis: results from discovery and validation cohorts. Hepatol Commun 2018; 2:807820.
21. Apostolopoulou M, Gordillo R, Koliaki C, et al. Specific hepatic sphingolipids relate to insulin resistance, oxidative stress, and inflammation in nonalcoholic steatohepatitis. Diabetes Care 2018; 41:12351243.
22. Koot BG, van der Baan-Slootweg OH, Vinke S, et al. Intensive lifestyle treatment for non-alcoholic fatty liver disease in children with severe obesity: inpatient versus ambulatory treatment. Int J Obes (Lond) 2016; 40:5157.
23. van Werven JR, Marsman HA, Nederveen AJ, et al. Assessment of hepatic steatosis in patients undergoing liver resection: comparison of US, CT, T1-weighted dual-echo MR imaging, and point-resolved 1H MR spectroscopy. Radiology 2010; 256:159168.
24. Herzog K, Pras-Raves ML, Vervaart MA, et al. Lipidomic analysis of fibroblasts from Zellweger spectrum disorder patients identifies disease-specific phospholipid ratios. J Lipid Res 2016; 57:14471454.
25. Chambers MC, Maclean B, Burke R, et al. A cross-platform toolkit for mass spectrometry and proteomics. Nat Biotechnol 2012; 30:918920.
26. Smith CA, Want EJ, O’Maille G, et al. XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal Chem 2006; 78:779787.
27. Wickham H. ggplot2. WIREs Comput Stat 2011; 3:180185.
28. Chong IG, Jun CH. Performance of some variable selection methods when multicollinearity is present. Chemometrics Intelligent Lab Syst 2005; 78:103112.
29. Fang YL, Chen H, Wang CL, et al. Pathogenesis of non-alcoholic fatty liver disease in children and adolescence: from “two hit theory” to “multiple hit model”. World J Gastroenterol 2018; 24:29742983.
30. Musso G, Cassader M, Paschetta E, et al. Bioactive lipid species and metabolic pathways in progression and resolution of nonalcoholic steatohepatitis. Gastroenterology 2018; 155:282.e8302.e8.
31. Oresic M, Hyotylainen T, Kotronen A, et al. Prediction of non-alcoholic fatty-liver disease and liver fat content by serum molecular lipids. Diabetologia 2013; 56:22662274.
32. Gorden DL, Myers DS, Ivanova PT, et al. Biomarkers of NAFLD progression: a lipidomics approach to an epidemic. J Lipid Res 2015; 56:722736.
33. Arendt BM, Ma DW, Simons B, et al. Nonalcoholic fatty liver disease is associated with lower hepatic and erythrocyte ratios of phosphatidylcholine to phosphatidylethanolamine. Appl Physiol Nutr Metab 2013; 38:334340.
34. Peng KY, Watt MJ, Rensen S, et al. Mitochondrial dysfunction-related lipid changes occur in nonalcoholic fatty liver disease progression. J Lipid Res 2018; 59:19771986.
35. Puri P, Baillie RA, Wiest MM, et al. A lipidomic analysis of nonalcoholic fatty liver disease. Hepatology 2007; 46:10811090.
36. Vance JE. Lipoproteins secreted by cultured rat hepatocytes contain the antioxidant 1-alk-1-enyl-2-acylglycerophosphoethanolamine. Biochim Biophys Acta 1990; 1045:128134.
37. Wiedeman AM, Barr SI, Green TJ, et al. Dietary choline intake: current state of knowledge across the life cycle. Nutrients 2018; 10.
38. Gaposchkin DP, Zoeller RA. Plasmalogen status influences docosahexaenoic acid levels in a macrophage cell line. Insights using ether lipid-deficient variants. J Lipid Res 1999; 40:495503.
39. Bikman BT, Summers SA. Sphingolipids and hepatic steatosis. Adv Exp Med Biol 2011; 721:8797.
40. Luukkonen PK, Zhou Y, Sadevirta S, et al. Hepatic ceramides dissociate steatosis and insulin resistance in patients with non-alcoholic fatty liver disease. J Hepatol 2016; 64:11671175.
41. Tiwari-Heckler S, Gan-Schreier H, Stremmel W, et al. Circulating phospholipid patterns in nafld patients associated with a combination of metabolic risk factors. Nutrients 2018; 10:649.
42. Mikucka-Niczyporuk A, Pierzynski P, Lemancewicz A, et al. Role of sphingolipids in the pathogenesis of intrahepatic cholestasis. Prostaglandins Other Lipid Mediat 2019; 147:106399.
43. Furukawa T, Fuda H, Miyanaga S, et al. Rapid tin-mediated access to a lysophosphatidylethanolamine (LPE) library: application to positional LC/MS analysis for hepatic LPEs in non-alcoholic steatohepatitis model mice. Chem Phys Lipids 2016; 200:133138.
44. Barr J, Caballeria J, Martinez-Arranz I, et al. Obesity-dependent metabolic signatures associated with nonalcoholic fatty liver disease progression. J Proteome Res 2012; 11:25212532.
45. Donnelly KL, Smith CI, Schwarzenberg SJ, et al. Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest 2005; 115:13431351.
46. Kotronen A, Seppanen-Laakso T, Westerbacka J, et al. Hepatic stearoyl-CoA desaturase (SCD)-1 activity and diacylglycerol but not ceramide concentrations are increased in the nonalcoholic human fatty liver. Diabetes 2009; 58:203208.
47. Aarsland A, Wolfe RR. Hepatic secretion of VLDL fatty acids during stimulated lipogenesis in men. J Lipid Res 1998; 39:12801286.
48. Sanyal AJ, Pacana T. A lipidomic readout of disease progression in a diet-induced mouse model of nonalcoholic fatty liver disease. Trans Am Clin Climatol Assoc 2015; 126:271288.
49. Shimomura I, Bashmakov Y, Horton JD. Increased levels of nuclear SREBP-1c associated with fatty livers in two mouse models of diabetes mellitus. J Biol Chem 1999; 274:3002830032.
Keywords:

biomarkers; liver; obesity; pediatric; steatosis

Supplemental Digital Content

Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition