GRAPHICAL ABSTRACT

Highlights

• Liver fat and insulin resistance decreased markedly after sleeve gastrectomy (SG)

• Lipidomic shifts in lipid species correlate with hormonal and clinical improvements.

• Glycerophospholipid remodeling correlates with changes in MASH-related parameters.

• Sphingolipid remodeling includes lipid ratio changes and increased hexosylceramide.

• Post-SG lipidomic signatures have metabolic and therapeutic relevance.

INTRODUCTION

Bariatric surgery, particularly sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB), is among the most effective treatments for severe obesity, producing substantial weight loss and improvements in conditions such as type 2 diabetes mellitus (T2DM), metabolic dysfunction-associated steatotic liver disease (MASLD), and cardiovascular disease [1-4]. While RYGB has superior weight loss effects, SG has become the most common bariatric surgery, with comparable long-term outcomes and notable metabolic benefits [2,3,5-8]. These benefits are linked to both weight loss-dependent and weight loss-independent mechanisms, including improved insulin sensitivity, a reduction in visceral fat area, and changes in hepatic and adipose tissue lipid metabolism [9].

Despite extensive results on RYGB, the detailed mechanisms underlying the metabolic effects of SG remain underexplored [10-15]. Thus, comprehensive analyses of various metabolic pathways related to SG-specific metabolic effects and accompanying changes in clinical parameters are still needed to understand the long-term metabolic effects following SG.

Lipids, which are essential in processes such as insulin regulation, hormone activity, and inflammation, are key metabolites for studying obesity and metabolic disorders [16]. High-resolution lipidomics, enabled by advanced liquid chromatography-mass spectrometry (LC-MS), provides a detailed profile of over 750 lipid species. This technology allows the analysis of lipid patterns, phenotypic correlations, and network models, revealing novel insights into the mechanisms and efficacy of SG [17].

In this prospective study, serum lipidomic analyses were performed on 50 obese subjects before and 6 months after SG to identify lipidomic signatures and their correlations with clinical parameters, providing a deeper understanding of the metabolic effects of bariatric surgery and potentially uncovering the novel clinical implications of SG.

METHODS

Study subjects and design

A pooled cohort study was developed from two primary prospective observational studies involving obese subjects undergoing bariatric surgery (Clinical trial registration No. KCT0003527 and KCT0009704). The first (principal investigator Dae Ho Lee) and the second (principal investigator Ji-Hyeon Park) studies shared nearly identical protocols and inclusion/exclusion criteria and primarily focused on evaluating MASLD before and after metabolic surgery, along with comprehensive metabolic assessment to identify blood and imaging biomarkers associated with disease progression. The second study was planned and conducted after the completion of the first study to focus more intensively on non-invasive biomarker research. Thus, they included research on the clinical, metabolic, and omics aspects related to MASLD. Between March 14, 2018 and January 10, 2024, a total of 81 obese subjects scheduled for bariatric surgery at Gachon University Gil Medical Center were recruited from the two primary cohorts of bariatric surgery. Of these 81 participants, 11 were enrolled in the second study. Laparoscopic vertical SG or RYGB was systematically performed by experienced surgeons (Ji-Hyeon Park and Seong Min Kim) in accordance with established guidelines and based on patients’ preferences following a thorough explanation of both procedures. The study protocols included baseline and follow-up studies 6 months after bariatric surgery for the evaluation of effects of bariatric surgery on various metabolic parameters including body weight, body fat, MASLD, and other pathophysiological factors as detailed below. The ages of the study subjects were required to be at least 19 years old, while the upper age limit in the first cohort study was 60 years. In the second cohort study, this limit was increased to 66 years. Key exclusion criteria were drug abuse, excessive alcohol consumption (alcohol intake >20 g/day for women and >30 g/day for men), evidence of another coexistent liver or biliary disease except for MASLD, use of medications known to cause secondary hepatic steatosis within the past 1 year, and any conditions that might affect patient competence or participation as determined by the opinion of the principal investigator. In the present study, we performed lipidomics analysis on serum samples from 50 subjects who received SG and completed post-operative follow-up visits. Each of the two primary studies was approved by the Institutional Review Board of Gachon University Gil Medical Center (GBIRB2017-200 and GCIRB2023-197) and conducted in accordance with the Declaration of Helsinki, and written informed consent was obtained from all participants. The two study protocols were registered at https://cris.nih.go.kr in accordance with the World Health Organization International Clinical Trials Registry Platform.

Clinical and laboratory evaluation

Various clinical and laboratory data were collected in the present study, as detailed previously in our previous studies on MASLD biomarkers [18-20]. After an overnight fast, blood samples were collected in a BDTM P800 tube (BD Bioscience, San Jose, CA, USA; Cat. No. 366420), which are pre-coated with a variety of (non-disclosed) pro-tease inhibitors, and an SST Tube with gel and clot activator (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) on the same day or within days of the imaging studies or several days before surgery to examine various biomarkers and metabolites and perform routine blood biochemical tests, which included liver function, glucose, insulin, a complete blood count with a platelet count, albumin, glycosylated hemoglobin (HbA1c), lipid panels, complement factors C3, and the enhanced liver fibrosis test parameters (matrix metabolism: hyaluronic acid, procollagen 3 N-terminal peptide [PIIINP], and tissue inhibitor of metalloproteinase 1 [TIMP-1]). All plasma and serum samples for additional laboratory measurements were frozen at –80°C until analysis. Currently available commercial kits were used for the measurement of blood levels of the apoptosis-associated neoepitope in the C-terminal domain of cytokeratin-18 (the M30-Apoptosense enzyme-linked immunosorbent assay [ELISA] kit, PEVIVA, Alexis, Grünwald, Germany), aldo-keto reductase family 1 member B10 (AKR1B10; Abcam, Cambridge, UK), complement factors C3 (turbidimetric immunoassay Tina-quant C3c, Roche Diagnostics Ltd., Rotkreuz, Switzerland), fasting plasma glucagon (Mercodia, Uppsala, Sweden), total glucose-dependent insulinotropic polypeptide (GIP; Merck, Darmstadt, Germany), active glucagon-like peptide-1 (GLP-1; Merck), oxyntomodulin (MyBioSource, San Diego, CA, USA), and fibroblast growth factor 21 (FGF21; Invitrogen, Carlsbad, CA, USA) were also measured using commercial ELISA kits. Body composition was measured using the dual-energy X-ray absorptiometry (DXA) technique (GE Healthcare, Wauwatosa, WI, USA) on the same day as the imaging studies. The hepatic and pancreatic fat content by magnetic resonance imaging-estimated proton density fat fraction (MRI-PDFF), the visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) areas at the umbilicus level, skeletal muscle area at the L3 lumbar spine level, and MR elastography-liver stiffness measurement (MRE-LSM) for hepatic fibrosis were all measured with a 3-T scanner (MAGNETOM Skyra, Siemens Healthineers, Erlangen, Germany) using an 18-channel body matrix coil and table-mounted 32-channel spine matrix coil. All participants underwent MR-based imaging studies and body factor assessment by DXA: hepatic and pancreatic PDFF, MRE-LSM, VAT, SAT, total body fat (%), and appendicular skeletal muscle mass. Other clinical indices and scores were calculated as previously described [18-20].

Lipid extraction

Serum lipids were extracted using a modified Folch method as described previously [21]. Briefly, 2 μL of an internal standard mixture (SPLASH Lipidomix standard, Avanti Polar Lipids, Alabaster, AL, USA) was spiked into 20 μL of each serum sample. Lipid extraction was performed by adding 600 μL of ice-cold methanol containing 1 mM butylated hydroxytoluene: chloroform (1:2, v/v). The mixtures were incubated on ice for 1 hour. To induce phase separation, 150 μL of distilled water was added to each sample, followed by vortexing for 30 seconds. After centrifugation, the lower organic phase was transferred to new tubes and the upper aqueous phase was re-extracted by adding an additional 600 μL of ice-cold methanol: chloroform (1:2, v/v). After re-extraction, the lower organic phase was combined with the initial extraction and evaporated. The residues were reconstituted in 100 μL of the solvent used as mobile phase B in LC-MS analysis.

Lipidomics data acquisition using ultra-performance liquid chromatography with orbitrap mass spectrometry

Lipidomic analysis was performed using an Orbitrap Exploris^TM 120 mass spectrometer coupled with a Vanquish Flex UHPLC system (Thermo Fisher Scientific, San Jose, CA, USA) at the Biopolymer Research Center for Advanced Materials (Sejong University, Seoul, Korea). An ACQUITY ultra-performance liquid chromatography (UPLC) BEH C18 column (2.1×100 mm, 1.7 μm; Waters, Milford, MA, USA) was used for chromatographic separation with 40% acetonitrile (v/v, mobile phase A) and isopropanol/acetonitrile (90:10, v/v, mobile phase B), both containing 2 mM ammonium formate and 0.1% formic acid. The gradient conditions were as follows: 0–1 minute, 10% B; 1–6 minutes, 10%−70% B; 6–12 minutes, 70%−90% B; 12–13 minutes, 90%–100% B; 13–13.5 minutes, 100%–10% B; 13.5–16 minutes, 10% B for re-equilibration with a flow rate of 0.35 mL/min. The MS scan was operated in both positive and negative ion modes with the following parameters: sheath gas flow at 50 arb, auxiliary gas flow at 10 arb, sweep gas flow at 1 arb, spray voltage at 3,500 V for positive mode and 2,500 V for negative mode, vaporizer temperature at 350°C.

MS data processing

Both full scan and data-dependent MS/MS spectra were acquired using Xcalibur 4.6 and pre-processed with Compound Discoverer 3.3 (Thermo Fisher Scientific) as previous described [21]. The fatty acyl composition of lipid species was determined based on MS/MS fragments and the relative quantification of each individual lipid species was performed by comparing peak areas normalized to a class-specific internal standard. Prior to statistical analysis, the reproducibility and reliability of the lipidomic data were ensured using quality control samples, which were repeatedly analyzed across the sample set (Supplementary Fig. 1).

Statistical analysis

To identify differences in the serum lipidome between the pre-SG and post-SG groups, unsupervised principal component analysis was performed using Umetrics SIMCA-P software version 17.0 (Umetrics AB, Umea, Sweden). The analytical workflow using the measured clinical parameters and lipidomic data is presented in Supplementary Fig. 2. First, a paired t-test was performed to compare clinically measured parameters before and after surgery in 50 patients, excluding those with missing data (Supplementary Table 1). We computed Spearman’s correlation coefficients between the average log2 (fold change) and the total carbon chain length or number of double bonds.

Next, Spearman’s correlation analysis was performed between the changes in the total abundance of each lipid subclass, lipid ratio, or lipid species and clinical parameters, with P values adjusted for multiple comparisons using the Benjamini-Hochberg procedure to control for type I errors. A Circos plot was constructed to visualize the associations between lipids and clinical markers [22].

RESULTS

Clinical characteristics of the patients before and after SG

The characteristics of the 50 obese subjects at baseline and 6 months after laparoscopic vertical SG are summarized in Table 1 (Supplementary Fig. 3). All obese subjects were Korean, with a mean±standard deviation age of 33.0±7.7 years and a mean body mass index of 37.9±5.2 kg/m².

Six months after SG, body weight decreased by 25%, whereas the homoeostatic model assessment of insulin resistance showed a reduction of more than 70%. Among the 12 patients with T2DM at baseline, 10 achieved diabetes remission 6 months after SG, resulting in a diabetes remission rate of 83.3%. The blood lipid profiles significantly improved, as expected (Table 1). Notably, the hepatic PDFF decreased by 70.2%. Blood and imaging biomarkers related to MASLD [18,20], including blood levels of PIIINP, cytokeratin-18, AKR1B10, and liver enzymes, and MRE-LSM, were significantly reduced after SG. Among the hormones and secretory factors, fasting blood levels of insulin and total GIP decreased significantly, whereas GLP-1 (active), FGF21, and oxyntomodulin levels increased, with a marked 2.5-fold increase in oxyntomodulin levels observed after SG. Additionally, significant alterations in the serum complement C3 level and leukocyte count were detected following SG.

Dynamic serum lipidome alterations after SG

We identified 751 lipids classified into five major classes (fatty acyls, glycerolipids, glycerophospholipids, sphingolipids, and sterol lipids) and 30 subclasses with dynamic changes and distinct clustering after SG (Fig. 1A and B). A total of 524 lipid species showed significant changes, with glycerolipids, glycerophospholipids, and sphingolipids contributing the most to these alterations (Supplementary Table 2, Supplementary Fig. 4).

For fatty acyls and glycerolipids, the total abundance, which was calculated by summing all lipid species within the subclasses, including free fatty acids (FFA), aylcarnitines (Car), diacylglycerols (DG), triacylglycerols (TG), monoacylglycerols (MG), and alkyltriacylglycerol (TG-O), decreased significantly following SG (Supplementary Table 3, Supplementary Fig. 5). However, despite the significant decrease in total abundance, the log2 (fold change) values of lipid species within the subclasses varied widely, ranging from –2.88 to 0.27, indicating differential effects of SG on individual lipid species (Fig. 1C).

Among the glycerophospholipid, sphingolipid, and sterol lipid classes, the levels of subclasses such as sphingoid base (SPB), ceramide (Cer), phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI), and alkylphosphatidylglycerol (PG-O) decreased significantly after surgery. In contrast, the levels of subclasses such as dihexosylceramide (Hex2Cer), lysophosphatidylcholine (LPC), lysoalkylphosphatidylcholine (LPC-O), alkylphosphatidylcholine or alkenylphosphatidylcholine (PC-O[P]), lysophosphatidylethanolamine (LPE), alkyllysophosphatidylethanolamine (LPE-O), and alkylphosphatidylethanolamine or alkenylphosphatidylethanolamine (PE-O[P]) increased significantly.

Molecular shifts in lipid species following SG

Considering the varied patterns of changes in specific lipid species within subclasses or major classes, we examined the relationship between the carbon chain length or the degree of unsaturation of fatty acyls within each lipid subclass and post-SG changes in the levels of specific lipids (Supplementary Table 4).

As shown in Fig. 2A, the levels of TG species with shorter and less unsaturated (i.e., more saturated) fatty acyl chains presented more pronounced reductions (P=2.619×10^–26 and P=3.598×10^–19, respectively) after SG, indicating a distinctive molecular shift in the serum TG pool toward longer and more unsaturated fatty acyl chains.

Among glycerophospholipids, the levels of species with shorter fatty acyl chains presented significant reductions, particularly PC (P=5.452×10^–6), the major subclass of mammalian glycerophospholipids [23]. Additionally, a significant increase in the level of lipid species with longer fatty acyl chains was observed for LPC (P=4.796×10^–6) and LPE (P=1.488×10^–2), suggesting a molecular shift toward longer fatty acyl chains (Fig. 2B). A similar shift pattern was also observed for cholesteryl esters (CE) (Supplementary Fig. 6).

Similarly, among sphingolipids, the levels of lipid species with fewer unsaturated fatty acyls decreased, whereas the levels of those with more unsaturated fatty acyl chains increased within sphingomyelin (SM, P=1.809×10^–3), Cer (P=9.096×10^–3), and hexosylceramide (HexCer, P=3.899×10^–3) after surgery (Fig. 2C).

Collectively, characteristic molecular shifts in fatty acyl composition were observed after SG across a wide range of lipid metabolic pathways (Fig. 2D).

Metabolic regulation of lipid pathways following SG

Since certain lipid ratios have the potential to serve as biomarkers for various diseases and health conditions [24,25], we assessed changes in several key lipid ratios after SG (Supplementary Table 5).

In particular, in glycerophospholipid metabolism, changes in various lipid ratios, including those of PC and PE, the most abundant phospholipids in mammalian cells, were identified (Fig. 3A). The PC/CE ratio, which is associated with the conversion of free cholesterol to CE [26,27], decreased significantly by 0.7-fold. In contrast, the LPC/PC and LPE/PE ratios, indicating the conversion of phospholipids to lysophospholipids (LysoPLs), were both significantly increased by approximately 1.6-fold. These findings suggest the upregulation of the pathway for the conversion of phospholipids to LysoPLs, along with increased cholesterol esterification (Fig. 3B).

In terms of the ether lipid-related ratios, the LPC-O/LPC, LPE-O/LPE, PE-O(P)/PE, and PC-O(P)/PC ratios increased significantly, with fold changes ranging from approximately 1.2 to 1.7, confirming the upregulation of ether lipid synthesis.

In sphingolipid metabolism, the HexCer/Cer, Hex2Cer/Cer, and sulfated hexosylceramide (SHexCer)/Cer ratios increased significantly, with fold changes ranging from approximately 1.2 to 1.4 (Fig. 3C), whereas the sphingosine level decreased significantly without a change in the sphingosine-1-phosphate level after SG (Fig. 3D, Supplementary Fig. 7).

Correlations between changes in lipidomic parameters and clinical and laboratory parameters

We performed a correlation analysis to investigate the associations between changes in various clinical and laboratory parameters and changes in the abovementioned lipidomic parameters following SG (Supplementary Tables 6-8).

First, changes in several MASLD-related parameters were exclusively associated with changes in sphingolipid metabolism. Specifically, changes in cytokeratin-18 levels were positively correlated with changes in SM (P=0.020) and SPB (P=0.006) levels (Fig. 4).

In the blood lipid panel, changes in total TG levels were negatively correlated with changes in LPE and LPE-O levels but positively correlated with changes in PC, PE, and PI abundances, as well as were correlated with changes in the ratios of several glycerophospholipid-related lipids. In contrast, changes in total cholesterol and low-density lipoprotein cholesterol (LDL-C) levels were positively correlated with changes in sphingolipid metabolism, suggesting the complex regulation of lipid metabolism.

Alterations in plasma oxyntomodulin levels were correlated with changes in all TG- and PE-related lipid markers. Additionally, the shorter the carbon chain length and the greater the degree of saturation of fatty acyl chains in TG and PE, the greater the correlation with changes in oxyntomodulin concentrations (Supplementary Fig. 8).

Correlations between changes in the levels of specific lipid species and clinical and laboratory parameters

Notably, characteristic correlations between changes in clinical variables and changes in the levels of individual fatty acyl-based lipid species were observed (Supplementary Table 8). Accordingly, the correlations between changes in the levels of all identified lipid species and clinical measures are shown in Fig. 5 (Supplementary Table 9).

First, changes in LDL-C and total cholesterol levels correlated with changes in the levels of numerous sphingolipid species (Fig. 4). Moreover, changes in the levels of PC species correlated significantly with changes in total cholesterol and total TG levels based on the carbon chain length, with R values of –0.556 and –0.274, respectively. Therefore, these correlations were observed only for PC species with fatty acyl chain lengths below C41.

With respect to MASLD-related parameters, changes in AKR1B10, cytokeratin-18, alanine aminotransferase (ALT), gamma-glutamyl transpeptidase (GGT), and liver R2* (an MR-based iron measurement) values [18] were positively correlated with changes in the levels of several sphingolipid species. Specifically, changes in D-sphingosine (SPB 18:1;O2) levels were correlated with changes in ALT and cytokeratin-18 levels, whereas changes in Cer 42:1;O4 levels were correlated with changes in both ALT and GGT levels (Supplementary Fig. 9). Additionally, changes in the levels of TG 40:1, a TG with relatively shorter and less unsaturated fatty acyls, were correlated (P=0.025) with changes in liver MRI-PDFF.

Changes in Cer 24:0;O3 levels were correlated with changes in both glucose and HbA1c levels (Supplementary Fig. 10). Additionally, changes in the pancreatic fat content were negatively correlated with changes in TG 54:10 levels.

Changes in oxyntomodulin levels were more strongly positively correlated with changes in the levels of shorter, less unsaturated TG species, as well as with PC 32:4 and PE 34:3 levels, but negatively correlated with LPC 40:7 levels (Supplementary Fig. 11). These results align with the observed negative correlation between changes in the levels of fatty acyl-based lipid species and correlation coefficients for changes in oxyntomodulin levels with TG, PC, PE, and LPC levels (Supplementary Fig. 8).

Changes in white blood cell (WBC) counts were positively correlated with changes in LPC 34:2, PC 32:3, PC 34:5, PC 36:6, and PC 39:8 levels but negatively correlated with changes in LPE 16:0 levels (Supplementary Fig. 12).

Additionally, the subcutaneous fat area was correlated with PG O-37:3 levels (Supplementary Fig. 13).

DISCUSSION

In this study, we observed significant metabolic improvements 6 months after SG, which included a 25% reduction in body weight and a marked decrease (by >70%) in both the intrahepatic TG content and insulin resistance. Despite being measured in a fasting state, plasma levels of oxyntomodulin doubled after surgery, accompanied by moderate increases in GLP-1 and FGF21 concentrations. A previous study reported that postprandial increases in oxyntomodulin and glicentin levels during the first 3 months after surgery are strongly linked to weight loss after 12 months, surpassing the role of GLP-1 [28]. Oxyntomodulin, a gut hormone, functions as a dual agonist for both GLP-1 and glucagon receptors. A study revealed that this kind of dual agonist (G49) has characteristic metabolic actions that increase energy expenditure by increasing adipose lipolysis, in which immune cells, HexCer (see below), and FGF21 are concurrently involved [29].

The lipidomic analysis of paired pre- and post-SG serum samples from 50 study subjects identified 751 lipids, among which prominent post-SG changes in glycerolipids, glycerophospholipids, and sphingolipids were observed.

We observed that SG not only reduced total serum TG levels but also induced a specific molecular shift toward TG species with longer fatty acyl chains and greater degrees of unsaturation, suggesting targeted metabolic remodeling. This shift implies enhanced β-oxidation or the preferential reduction of TG species with greater lipotoxic potential, particularly those containing saturated and relatively short-chain fatty acyls, which have been associated with hepatic fat accumulation and insulin resistance [30]. These lipidomic changes likely reflect metabolic adaptations in peripheral tissues such as adipose tissue and the liver, which are central regulators of lipid homeostasis, and are consistent with the observed improvements in clinical parameters. Notably, post-SG changes in the levels of TG 40:1, which is characterized by a shorter chain and lower degree of unsaturation, were strongly positively correlated with changes in MRI-PDFF, whereas changes in the levels of TG 54:10, which is characterized by a longer chain and greater degree of unsaturation, were negatively correlated with changes in the pancreatic fat content, a risk factor for T2DM [31]. These findings suggest that the SG-induced TG remodeling not only reflects improved lipid utilization but may also contribute to the broader metabolic benefits, including reductions in fat accumulation and improvements in insulin sensitivity [13,32-34].

In addition to the observed shifts in TGs, similar molecular alterations in glycerophospholipid and sphingolipid classes may also reflect underlying changes in endogenous fatty acid metabolism. Previous studies have reported that bariatric surgery enhances the activity of key lipid metabolic enzymes, including enzymes that catalyze the elongation of very long-chain fatty acids (ELOVL3, ELOVL5, and ELOVL6) and desaturases such as delta-5 desaturase (D5D), which are involved in fatty acid chain elongation and unsaturation [35]. These enzymatic changes have been associated with increased polyunsaturated fatty acid (PUFA) elongation and enhanced β-oxidation capacity in peripheral tissues, contributing to improved lipid handling and insulin sensitivity. Moreover, bariatric surgery has been shown to restore the reduced n-3 PUFA profile commonly observed in morbid obesity by significantly increasing serum n-3 PUFA levels during the post-surgical period [36]. Collectively, these molecular alterations across multiple lipid classes may represent a lipidomic signature of enzymatic and metabolic adaptation following SG, potentially contributing to the observed clinical improvements and offering valuable insights into future therapeutic targets.

Changes in plasma oxyntomodulin levels were significantly correlated not only with changes in total TG but also with four specific TG species and the molecular shift patterns of TG metabolism (Supplementary Figs. 8 and 11). These correlations were particularly pronounced in TG species with shorter carbon chains and lower degrees of unsaturation, suggesting that oxyntomodulin may play a role in the selective mobilization or catabolism of specific lipid species. Previous studies have reported that postprandial increases in oxyntomodulin after bariatric surgery are more strongly associated with weight loss than other gut hormones such as GLP-1 [28,37]. Moreover, oxyntomodulin has a longer half-life than GLP-1, which may contribute to its sustained metabolic effect [38]. Although measured in the fasting state, oxyntomodulin levels increased by approximately 2.5-fold after SG in our study, and the observed correlations with lipidomic changes support its potential involvement in post-surgical metabolic remodeling. These findings extend previous preclinical evidence that oxyntomodulin analogs can stimulate lipid oxidation and thermogenesis through FGF21-mediated pathways [39], by identifying distinct molecular lipid targets that may be preferentially influenced by oxyntomodulin.

Glycerophospholipids play critical roles in lipid metabolism processes, such as lipoprotein formation, lipid droplet regulation, mitochondrial homeostasis, and immune regulation [40]. A reduced PC/CE ratio following SG was associated with the reduction in total cholesterol levels and increased high-density lipoprotein cholesterol levels.

After SG, a reduction in major phospholipid levels was observed alongside an increase in LysoPL levels and elevated LPC/PC and LPE/PE ratios. The conversion of PC to LPC is associated with the alleviation of the metabolic stress and chronic inflammation, particularly in the context of weight loss [41]. Consistent with these findings, our results showed a significant correlation between the reduction in specific PC levels and WBC counts.

Ether lipids, including alkyl- and alkenyl-linked phospholipids, constitute approximately 20% of mammalian phospholipids and play essential roles in membrane structure, antioxidant defense, and signal transduction. Dysregulation of ether lipid metabolism has been implicated in obesity, insulin resistance, and cardiovascular disease [42-45]. In the present study, ether lipid levels in major phospholipids subclasses increased significantly after SG, suggesting remodeling may enhance membrane fluidity and promote antioxidant and anti-inflammatory effects.

Additionally, the observed negative correlation between changes in PC O-41:4 and liver stiffness (MRE-LSM) supports a possible antifibrotic role, consistent with previous reports suggesting that ether lipids may mitigate oxidative stress and fibrosis in MASLD.

The upregulation of ether lipids may also reflect a compensatory mechanism to restore redox balance in response to increased mitochondrial function and oxidative metabolism following weight loss. Given the growing interest in ether lipids as biomarkers of cardiometabolic health [44], our results underscore their potential as targets for monitoring or modulating post-SG metabolic adaptations.

Consistent with previous studies on sphingolipid and insulin resistance [46-48] the total Cer and sphingosine levels decreased significantly following SG. Specifically, changes in Cer 24:0;O3 levels were strongly positively correlated with changes in glucose and HbA1c levels.

Regulation of the Cer/SM balance plays a critical role in cellular functions such as apoptosis, autophagy, inflammation, and fibrosis [49,50]. The reduction in the Cer/SM ratio after SG indicates a metabolic improvement in the Cer/SM imbalance associated with obesity. Cer can also be glucosylated by glucosylceramide synthase to generate HexCer, which is known to regulate biological processes such as cell proliferation, apoptosis, and inflammation [51-54]. The levels of circulating Hex-Cer are lower in patients with coronary artery disease accompanied by T2DM than in those without T2DM [52-54]. In the present study, the upregulation of the HexCer pathway following SG suggests a metabolic effect of SG. Interestingly, one of the HexCer, α-galactosylceramide (αGalCer), has been shown to activate invariant natural killer T (iNKT) cells, promoting weight loss by inducing browning of white adipose tissue, thermogenesis, and β-oxidation, which is mediated by FGF21 [55]. Further studies on the tissue origins of these HexCer and the biological actions of each HexCer diastereomer are warranted [56].

Our study has several limitations, including potential biases related to sex, age, and lifestyle differences, despite the use of paired pre- and postsurgery samples. The study was conducted at a single center with a 6-month follow-up, limiting insights into long-term effects. Some subgroup-level comparisons may have been underpowered. Future studies with larger cohorts may improve statistical power and allow the identification of additional lipidomic associations with clinical relevance.

This study highlights significant lipidomic remodeling following SG, characterized by molecular shifts toward longer, more unsaturated lipid species that are associated with improvements in hepatic steatosis, insulin sensitivity, and elevated plasma oxyntomodulin levels. Enhanced activity of glycerophospholipid-to-LysoPL pathways may reduce inflammation, whereas increased ether lipid levels suggest antifibrotic effects on MASLD. An improved Cer/SM balance is correlated with reduced insulin resistance and inflammation, whereas HexCer upregulation indicates cardiometabolic benefits and increased energy metabolism. These lipidomic signatures may serve not only as biomarkers of metabolic improvement but also as dynamic indicators of treatment response and predictive markers of outcomes, such as T2DM remission or hepatic fibrosis. From a clinical perspective, several lipid species or ratios (e.g., TG 40:1, LPC/PC, HexCer/Cer) demonstrated strong correlations with key metabolic parameters and could be developed into non-invasive markers for monitoring post-surgical metabolic adaptation. Some of these lipid changes may serve as novel targets for therapies aiming to mimic the metabolic improvements achieved through SG. Altogether, our findings underscore the translational value of lipidomics in understanding, monitoring, and potentially enhancing the metabolic efficacy of bariatric surgery.

SUPPLEMENTARY MATERIALS

NOTES

CONFLICTS OF INTEREST

Dae Ho Lee has been an international editorial board member of the Diabetes & Metabolism Journal since 2023. He was not involved in the review process of this article. Otherwise, there was no conflict of interest.

AUTHOR CONTRIBUTIONS

Conception or design: G.L., S.M.K., J.H.P., D.H.L.

Acquisition, analysis, or interpretation of data: G.L., Y.C.L., M.P., S.M.K., J.H.P.

Drafting the work or revising: G.L., Y.C.L., D.H.L.

Final approval of the manuscript: all authors.

FUNDING

This study was supported by grants from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), which is funded by the Ministry of Health & Welfare, Korea (No. HI14C1135, to Dae Ho Lee); the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. NFR-2019 R1I1A2A02062305, to Dae Ho Lee and RS-2023-00245412, to Yeong Chan Lee); the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2021R1A5A2030333, to Dae Ho Lee); and the Gachon University Gil Medical Center (FRD2021-03, to Dae Ho Lee); and by Korea Basic Science Institute (National Research Facilities and Equipment Center) grant funded by the Ministry of Education (No. 2023R1A6C101A045).

ACKNOWLEDGMENTS

None

Fig. 1.

Serum lipidomic profiles of the study subjects before and 6 months after sleeve gastrectomy (SG) (n=50). (A) Total number of lipid species detected in the lipidomic platform, grouped by lipid subclasses. (B) Principal component analysis (PCA) plots generated from the lipidomic profiling data. (C) Bubble plots illustrating the changes in lipid species within each subclass after SG surgery. Each circle represents an individual lipid species within a lipid subclass, with its size reflecting the P value. The gray circles indicate nonsignificant changes, whereas the colored circles denote statistically significant changes after SG. FFA, free fatty acid; Car, aylcarnitines; TG, triacylglycerol; DG, diacylglycerol; TG-O, alkyltriacylglycerol; MG, monoacylglycerol; PC, phosphatidylcholine; PC-O, alkylphosphatidylcholine; PC-P, alkenylphosphatidylcholine; LPC, lysophosphatidylcholine; PE-O, alkylphosphatidylethanolamine; PE-P, alkenylphosphatidylethanolamine; PE, phosphatidylethanolamine; LPC-O, lysoalkylphosphatidylcholine; LPE, lysophosphatidylethanolamine; PI, phosphatidylinositol; PG, phosphatidylglycerol; PG-O, alkylphosphatidylglycerol; PA, phosphatidic acid; PS, phosphatidylserine; PS-O, alkylphosphatidylserine; PA-O, alkylphosphatidic acid; LPE-O, lysoalkylphatidylethanolamine; PI-O, alkylphosphatidylinositol; SM, sphingomyelin; Cer, ceramide; HexCer, hexosylceramide; Hex2Cer, dihexosylceramide; SHexCer, sulfated hexosylceramide; SPB, sphingoid base; CE, cholesteryl ester; ST, sterols; NS, not significant.

Fig. 2.

Molecular shift in the serum lipid pool of obese patients following sleeve gastrectomy (SG) surgery. (A) Changes in triacylglycerol (TG) species following surgery based on the fatty acyl carbon chain length and degree of unsaturation and their correlations. (B) Changes in glycerophospholipid species (phosphatidylcholine [PC], lysophosphatidylcholine [LPC], and lysophosphatidylethanolamine [LPE]) after SG surgery based on the fatty acyl carbon chain length and their correlations. (C) Changes in sphingolipid (sphingomyelin [SM], ceramide [Cer], and hexosylceramide [HexCer]) species after surgery based on the degree of unsaturation of fatty acyl chains and their correlations. (D) Molecular shift of lipid species within the serum lipid pool following SG surgery. The bar graphs display the mean log2 (fold change [FC]) values with standard deviations of TG species, grouped by the fatty acyl chain length and degree of unsaturation arranged along the x-axis. Each symbol represents data from an individual patient. In the correlation plot, each data point represents a distinct lipid species, organized along the x-axis by the total number of double bonds or carbon numbers in the acyl chains.

Fig. 3.

Regulation of lipid metabolic pathways following triacylglycerol (TG) surgery. (A) Ratios of lipid markers related to glycerophospholipid metabolism before and after sleeve gastrectomy (SG) surgery. (B) Regulation of metabolic pathways involved in glycerophospholipid metabolism following SG surgery. (C) Ratios of lipid markers related to sphingolipid metabolism after surgery. (D) Regulation of metabolic pathways involved in sphingolipid metabolism following SG surgery. The box-whisker plots display the 90th/10th percentiles at the whiskers, with the median represented by the center line and the mean represented by the ‘+.’ PC, phosphatidylcholine; CE, cholesteryl ester; LPC, lysophosphatidylcholine; LPE, lysophosphatidylethanolamine; PE, phosphatidylethanolamine; LPC-O, lysoalkylphosphatidylcholine; LPE-O, lysoalkylphatidylethanolamine; PC-O, alkylphosphatidylcholine; PC-P, alkenylphosphatidylcholine; PE-O, alkylphosphatidylethanolamine; PE-P, alkenylphosphatidylethanolamine; HDL, high-density lipoprotein; LCAT, lecithin-cholesterol acyltransferase; PLA, phospholipase A; PEMT, phosphatidylethanolamine N-methyltransferase; DHAP, dihydroxyacetone phosphate; Cer, ceramide; SM, sphingomyelin; HexCer, hexosylceramide; Cer, ceramide; Hex2Cer, dihexosylceramide; SHexCer, sulfated hexosylceramide; CERS, ceramide synthase; S1P, sphingosine-1-phosphate; SMS, sphingomyelin synthase; SMase, sphingomyelinase; GCS, glucosyl/galactosyl-ceramide synthase; GCDase, glucosyl/galactosyl-ceramidase. Statistically significance based on paired t-tests, with P values adjusted using the Benjamini-Hochberg method: ^aP<0.01, ^bP<0.001.

Fig. 4.

Correlations between changes in lipid subclasses or ratios and changes in clinical parameters. The dot size represents the Benjamini-Hochberg-adjusted −log10 (false discovery rate [FDR]). The color scale indicates the strength and direction of the correlation. R2*, R2* relaxation rate; ALT, alanine aminotransferase; AKR1B10, aldo-keto reductase family 1 member B10; HOMA-IR, homoeostatic model assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; FGF21, fibroblast growth factor 21; WBC, white blood cell; FFA, free fatty acid; DG, diacylglycerol; MG, monoacylglycerol; TG, triacylglycerol; TG-O, alkyltriacylglycerol; LPC, lysophosphatidylcholine; LPE, lysophosphatidylethanolamine; LPE-O, lysoalkylphatidylethanolamine; PA, phosphatidic acid; PA-O, alkylphosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG-O, alkylphosphatidylglycerol; PI, phosphatidylinositol; PS, phosphatidylserine; Cer, ceramide; Hex2Cer, dihexosylceramide; HexCer, hexosylceramide; SHexCer, sulfated hexosylceramide; SM, sphingomyelin; SPB, sphingoid base; ST, sterols; PC-O, alkylphosphatidylcholine; PC-P, alkenylphosphatidylcholine; CE, cholesteryl ester; PE-O, alkylphosphatidylethanolamine; PE-P, alkenylphosphatidylethanolamine.

Fig. 5.

Correlations between changes in lipid species levels and clinical parameters. The Circos plot illustrates the correlations between lipid species and clinical measures. Triacylglycerols (TGs) analyzed through lipidomics and total TG levels from clinical laboratory data were naturally correlated and, therefore, excluded from the plot. CE, cholesteryl ester; ST, sterols; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; DG, diacylglycerol; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PC-O, alkylphosphatidylcholine; Cer, ceramide; Hex2Cer, dihexosylceramide; HexCer, hexosylceramide; SM, sphingomyelin; SHexCer, sulfated hexosylceramide; HbA1c, glycosylated hemoglobin; MRI-PDFF, magnetic resonance imaging-estimated proton density fat fraction; AKR1B10, aldo-keto reductase family 1 member B10; ALT, alanine aminotransferase; GGT, gamma-glutamyl transpeptidase; MRE-LSM, MR elastography-liver stiffness measurement; R2*, R2* relaxation rate; PIIINP, procollagen 3 N-terminal peptide; TIMP1, tissue inhibitor of metalloproteinase 1; GLP-1, glucagon-like peptide-1; WBC, white blood cell; hsCRP, high-sensitivity C-reactive protein; BUN, blood urea nitrogen; FFA, free fatty acid; Car, aylcarnitines; TG, triacylglycerol; PG-O, alkylphosphatidylglycerol; LPC, lysophosphatidylcholine; LPC-O, lysoalkylphosphatidylcholine; PE-P, alkenylphosphatidylethanolamine; LPE, lysophosphatidylethanolamine; SPB, sphingoid base.

REFERENCES

• 1. Krishnan A, Hadi Y, Alqahtani SA, Woreta TA, Fang W, Abunnaja S, et al. Cardiovascular outcomes and mortality after bariatric surgery in patients with nonalcoholic fatty liver disease and obesity. JAMA Netw Open 2023;6:e237188.ArticlePubMedPMC
• 2. Lee G, Park YS, Cho C, Lee H, Park J, Park DJ, et al. Short-term changes in the serum metabolome after laparoscopic sleeve gastrectomy and Roux-en-Y gastric bypass. Metabolomics 2021;17:71.ArticlePubMedPDF
• 3. Aminian A, Al-Kurd A, Wilson R, Bena J, Fayazzadeh H, Singh T, et al. Association of bariatric surgery with major adverse liver and cardiovascular outcomes in patients with biopsy-proven nonalcoholic steatohepatitis. JAMA 2021;326:2031-42.ArticlePubMedPMC
• 4. Petrov MS, Taylor R. Intra-pancreatic fat deposition: bringing hidden fat to the fore. Nat Rev Gastroenterol Hepatol 2022;19:153-68.ArticlePubMedPDF
• 5. Oh TJ, Lee HJ, Cho YM. East Asian perspectives in metabolic and bariatric surgery. J Diabetes Investig 2022;13:756-61.ArticlePubMedPMCPDF
• 6. Peterli R, Wolnerhanssen BK, Peters T, Vetter D, Kroll D, Borbely Y, et al. Effect of laparoscopic sleeve gastrectomy vs laparoscopic Roux-en-Y gastric bypass on weight loss in patients with morbid obesity: the SM-BOSS randomized clinical trial. JAMA 2018;319:255-65.ArticlePubMedPMC
• 7. Coleman KJ, Wellman R, Fitzpatrick SL, Conroy MB, Hlavin C, Lewis KH, et al. Comparative safety and effectiveness of Rouxen-Y gastric bypass and sleeve gastrectomy for weight loss and type 2 diabetes across race and ethnicity in the PCORnet bariatric study cohort. JAMA Surg 2022;157:897-906.ArticlePubMedPMC
• 8. Salminen P, Gronroos S, Helmio M, Hurme S, Juuti A, Juusela R, et al. Effect of laparoscopic sleeve gastrectomy vs Roux-en-Y gastric bypass on weight loss, comorbidities, and reflux at 10 years in adult patients with obesity: the SLEEVEPASS randomized clinical trial. JAMA Surg 2022;157:656-66.ArticlePubMedPMC
• 9. Sandoval DA, Patti ME. Glucose metabolism after bariatric surgery: implications for T2DM remission and hypoglycaemia. Nat Rev Endocrinol 2023;19:164-76.ArticlePubMedPMCPDF
• 10. Grayson BE, Gutierrez-Aguilar R, Sorrell JE, Matter EK, Adams MR, Howles P, et al. Bariatric surgery emphasizes biological sex differences in rodent hepatic lipid handling. Biol Sex Differ 2017;8:4.ArticlePubMedPMCPDF
• 11. Yoshino M, Kayser BD, Yoshino J, Stein RI, Reeds D, Eagon JC, et al. Effects of diet versus gastric bypass on metabolic function in diabetes. N Engl J Med 2020;383:721-32.ArticlePubMedPMC
• 12. Chambers AP, Jessen L, Ryan KK, Sisley S, Wilson-Perez HE, Stefater MA, et al. Weight-independent changes in blood glucose homeostasis after gastric bypass or vertical sleeve gastrectomy in rats. Gastroenterology 2011;141:950-8.ArticlePubMedPMC
• 13. Bagheri M, Tanriverdi K, Iafrati MD, Mosley JD, Freedman JE, Ferguson JF. Characterization of the plasma metabolome and lipidome in response to sleeve gastrectomy and gastric bypass surgeries reveals molecular patterns of surgical weight loss. Metabolism 2024;158:155955.ArticlePubMedPMC
• 14. Vaz M, Pereira SS, Monteiro MP. Metabolomic signatures after bariatric surgery: a systematic review. Rev Endocr Metab Disord 2022;23:503-19.ArticlePubMedPMCPDF
• 15. Halasz M, Luczaj W, Jarocka-Karpowicz I, Stasiewicz A, Soldo AM, Soldo I, et al. Relationship between systemic biomarker of lipid peroxidation 4-hydroxynonenal and lipidomic profile of morbidly obese patients undergoing bariatric surgery. Free Radic Biol Med 2024;224:564-73.ArticlePubMed
• 16. Kimura T, Jennings W, Epand RM. Roles of specific lipid species in the cell and their molecular mechanism. Prog Lipid Res 2016;62:75-92.ArticlePubMed
• 17. Ni Z, Wolk M, Jukes G, Mendivelso Espinosa K, Ahrends R, Aimo L, et al. Guiding the choice of informatics software and tools for lipidomics research applications. Nat Methods 2023;20:193-204.ArticlePubMedPMCPDF
• 18. Choi SJ, Kim SM, Kim YS, Kwon OS, Shin SK, Kim KK, et al. Magnetic resonance-based assessments better capture pathophysiologic profiles and progression in nonalcoholic fatty liver disease. Diabetes Metab J 2021;45:739-52.ArticlePubMedPMCPDF
• 19. Ji M, Jo Y, Choi SJ, Kim SM, Kim KK, Oh BC, et al. Plasma metabolomics and machine learning-driven novel diagnostic signature for non-alcoholic steatohepatitis. Biomedicines 2022;10:1669.ArticlePubMedPMC
• 20. Choi SJ, Yoon S, Kim KK, Kim D, Lee HE, Kim KG, et al. A composite blood biomarker including AKR1B10 and cytokeratin 18 for progressive types of nonalcoholic fatty liver disease. Diabetes Metab J 2024;48:740-51.ArticlePubMedPMCPDF
• 21. Yu JS, Kim HJ, Kim YE, Yang HO, Shin YK, Kim H, et al. Lipidomic assessment of the inhibitory effect of standardized water extract of Hydrangea serrata (Thunb.) Ser. leaves during adipogenesis. Nutrients 2024;16:1508.ArticlePubMedPMC
• 22. Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. Circos: an information aesthetic for comparative genomics. Genome Res 2009;19:1639-45.ArticlePubMedPMC
• 23. van der Veen JN, Kennelly JP, Wan S, Vance JE, Vance DE, Jacobs RL. The critical role of phosphatidylcholine and phosphatidylethanolamine metabolism in health and disease. Biochim Biophys Acta Biomembr 2017;1859(9 Pt B):1558-72.ArticlePubMedPMC
• 24. Liu D, Meister M, Zhang S, Vong CI, Wang S, Fang R, et al. Identification of lipid biomarker from serum in patients with chronic obstructive pulmonary disease. Respir Res 2020;21:242.ArticlePubMedPMCPDF
• 25. Apostolopoulou M, Gordillo R, Gancheva S, Strassburger K, Herder C, Esposito I, et al. Role of ceramide-to-dihydroceramide ratios for insulin resistance and non-alcoholic fatty liver disease in humans. BMJ Open Diabetes Res Care 2020;8:e001860.ArticlePubMedPMC
• 26. Thacker SG, Rousset X, Esmail S, Zarzour A, Jin X, Collins HL, et al. Increased plasma cholesterol esterification by LCAT reduces diet-induced atherosclerosis in SR-BI knockout mice. J Lipid Res 2015;56:1282-95.ArticlePubMedPMC
• 27. Ma X, Chen L, He Y, Zhao L, Yu W, Xie Y, et al. Targeted lipidomics reveals phospholipids and lysophospholipids as biomarkers for evaluating community-acquired pneumonia. Ann Transl Med 2022;10:395.ArticlePubMedPMC
• 28. Perakakis N, Kokkinos A, Peradze N, Tentolouris N, Ghaly W, Pilitsi E, et al. Circulating levels of gastrointestinal hormones in response to the most common types of bariatric surgery and predictive value for weight loss over one year: evidence from two independent trials. Metabolism 2019;101:153997.ArticlePubMed
• 29. Valdecantos MP, Ruiz L, Folgueira C, Rada P, Gomez-Santos B, Solas M, et al. The dual GLP-1/glucagon receptor agonist G49 mimics bariatric surgery effects by inducing metabolic rewiring and inter-organ crosstalk. Nat Commun 2024;15:10342.ArticlePubMedPMCPDF
• 30. Luukkonen PK, Sadevirta S, Zhou Y, Kayser B, Ali A, Ahonen L, et al. Saturated fat is more metabolically harmful for the human liver than unsaturated fat or simple sugars. Diabetes Care 2018;41:1732-9.ArticlePubMedPMCPDF
• 31. Nadarajah C, Fananapazir G, Cui E, Gichoya J, Thayalan N, Asare-Sawiri M, et al. Association of pancreatic fat content with type II diabetes mellitus. Clin Radiol 2020;75:51-6.ArticlePubMed
• 32. Vessby B. Dietary fat, fatty acid composition in plasma and the metabolic syndrome. Curr Opin Lipidol 2003;14:15-9.ArticlePubMed
• 33. Warensjo E, Riserus U, Vessby B. Fatty acid composition of serum lipids predicts the development of the metabolic syndrome in men. Diabetologia 2005;48:1999-2005.ArticlePubMedPMCPDF
• 34. Rhee EP, Cheng S, Larson MG, Walford GA, Lewis GD, Mc-Cabe E, et al. Lipid profiling identifies a triacylglycerol signature of insulin resistance and improves diabetes prediction in humans. J Clin Invest 2011;121:1402-11.PubMedPMC
• 35. Walle P, Takkunen M, Mannisto V, Vaittinen M, Kakela P, Agren J, et al. Alterations in fatty acid metabolism in response to obesity surgery combined with dietary counseling. Nutr Diabetes 2017;7:e285.ArticlePubMedPMCPDF
• 36. Hovland A, Nestvold T, Bohov P, Troseid M, Aukrust P, Berge RK, et al. Bariatric surgery reduces fasting total fatty acids and increases n-3 polyunsaturated fatty acids in morbidly obese individuals. Scand J Clin Lab Invest 2017;77:628-33.ArticlePubMed
• 37. Nielsen MS, Ritz C, Wewer Albrechtsen NJ, Holst JJ, le Roux CW, Sjodin A. Oxyntomodulin and glicentin may predict the effect of bariatric surgery on food preferences and weight loss. J Clin Endocrinol Metab 2020;105:dgaa061.ArticlePubMedPDF
• 38. Schjoldager BT, Baldissera FG, Mortensen PE, Holst JJ, Christiansen J. Oxyntomodulin: a potential hormone from the distal gut. Pharmacokinetics and effects on gastric acid and insulin secretion in man. Eur J Clin Invest 1988;18:499-503.ArticlePubMed
• 39. Kueh MT, Chong MC, Miras AD, le Roux CW. Oxyntomodulin physiology and its therapeutic development in obesity and associated complications. J Physiol 2025;603:7683-93.ArticlePubMedPMC
• 40. Lagace TA, Ridgway ND. The role of phospholipids in the biological activity and structure of the endoplasmic reticulum. Biochim Biophys Acta 2013;1833:2499-510.ArticlePubMed
• 41. Heimerl S, Fischer M, Baessler A, Liebisch G, Sigruener A, Wallner S, et al. Alterations of plasma lysophosphatidylcholine species in obesity and weight loss. PLoS One 2014;9:e111348.ArticlePubMedPMC
• 42. Nagan N, Zoeller RA. Plasmalogens: biosynthesis and functions. Prog Lipid Res 2001;40:199-229.ArticlePubMed
• 43. Garces F, Lopez F, Nino C, Fernandez A, Chacin L, Hurt-Camejo E, et al. High plasma phospholipase A2 activity, inflammation markers, and LDL alterations in obesity with or without type 2 diabetes. Obesity (Silver Spring) 2010;18:2023-9.ArticlePubMedPDF
• 44. Graessler J, Schwudke D, Schwarz PE, Herzog R, Shevchenko A, Bornstein SR. Top-down lipidomics reveals ether lipid deficiency in blood plasma of hypertensive patients. PLoS One 2009;4:e6261.ArticlePubMedPMC
• 45. Pietilainen KH, Rog T, Seppanen-Laakso T, Virtue S, Gopalacharyulu P, Tang J, et al. Association of lipidome remodeling in the adipocyte membrane with acquired obesity in humans. PLoS Biol 2011;9:e1000623.ArticlePubMedPMC
• 46. Hannun YA, Obeid LM. Sphingolipids and their metabolism in physiology and disease. Nat Rev Mol Cell Biol 2018;19:175-91.ArticlePubMedPMCPDF
• 47. Sokolowska E, Blachnio-Zabielska A. The role of ceramides in insulin resistance. Front Endocrinol (Lausanne) 2019;10:577.ArticlePubMedPMC
• 48. Kang SC, Kim BR, Lee SY, Park TS. Sphingolipid metabolism and obesity-induced inflammation. Front Endocrinol (Lausanne) 2013;4:67.ArticlePubMedPMC
• 49. Kolesnick R. The therapeutic potential of modulating the ceramide/sphingomyelin pathway. J Clin Invest 2002;110:3-8.ArticlePubMedPMC
• 50. Grammatikos G, Ferreiros N, Waidmann O, Bon D, Schroeter S, Koch A, et al. Serum sphingolipid variations associate with hepatic decompensation and survival in patients with cirrhosis. PLoS One 2015;10:e0138130.ArticlePubMedPMC
• 51. DeVeaux SA, Ogle ME, Vyshnya S, Chiappa NF, Leitmann B, Rudy R, et al. Characterizing human mesenchymal stromal cells’ immune-modulatory potency using targeted lipidomic profiling of sphingolipids. Cytotherapy 2022;24:608-18.ArticlePubMedPMC
• 52. Dusing P, Heinrich NN, Al-Kassou B, Gutbrod K, Dormann P, Nickenig G, et al. Analysis of circulating ceramides and hexosylceramides in patients with coronary artery disease and type II diabetes mellitus. BMC Cardiovasc Disord 2023;23:454.PubMedPMC
• 53. Zigmond E, Zangen SW, Pappo O, Sklair-Levy M, Lalazar G, Zolotaryova L, et al. Beta-glycosphingolipids improve glucose intolerance and hepatic steatosis of the Cohen diabetic rat. Am J Physiol Endocrinol Metab 2009;296:E72-8.PubMed
• 54. Margalit M, Shalev Z, Pappo O, Sklair-Levy M, Alper R, Gomori M, et al. Glucocerebroside ameliorates the metabolic syndrome in OB/OB mice. J Pharmacol Exp Ther 2006;319:105-10.ArticlePubMed
• 55. Lynch L, Hogan AE, Duquette D, Lester C, Banks A, LeClair K, et al. iNKT cells induce FGF21 for thermogenesis and are required for maximal weight loss in GLP1 therapy. Cell Metab 2016;24:510-9.ArticlePubMedPMC
• 56. Hosono Y, Tomiyasu N, Kasai H, Ishikawa E, Takahashi M, Imamura A, et al. Identification of α-galactosylceramide as an endogenous mammalian antigen for iNKT cells. J Exp Med 2025;222:e20240728.ArticlePubMedPMCPDF

Figure & Data

References

Citations

Citations to this article as recorded by  

• Hepatic Insulin Resistance and Steatosis in Metabolic Dysfunction-Associated Steatotic Liver Disease: New Insights into Mechanisms and Clinical Implications
  Xuan Trong Truong, Dae Ho Lee
  Diabetes & Metabolism Journal.2025; 49(5): 964.     CrossRef

About this article

Lee G, Lee YC, Park M, Kim SM, Park JH, Lee DH. Dynamic Lipidomic Remodeling and Clinical Correlations after Sleeve Gastrectomy in Obese Subjects. Diabetes Metab J. 2026;50(2):396-411.

Received: Feb 12, 2025; Accepted: Apr 17, 2025

DOI: https://doi.org/10.4093/dmj.2025.0120.

Download citation ↓