Ceramide d18:1/24:1 as a potential biomarker to differentiate obesity subtypes with unfavorable health outcomes

Background The criteria for metabolically healthy obesity (MHO) and metabolically unhealthy obesity (MUO) remain controversial. This research aimed to identify a potential biomarker to differentiate the subtypes of obesity. Methods The study conducted a lipidomic evaluation of ceramide in the serum of 77 Chinese adults who had undergone hyperinsulinemic-euglycemic clamps. These adults were divided into three groups according to the clinical data: normal weight control group (N = 21), MHO (N = 20), and MUO (N = 36). Results The serum Cer d18:1/24:1 level in the MHO group was lower than that in the MUO group. As the Cer d18:1/24:1 level increased, insulin sensitivity decreased, and the unfavorable parameters increased in parallel. Multivariate logistic regression analysis revealed that serum Cer d18:1/24:1 levels were independently correlated with MUO in obesity. Individuals with higher levels of Cer d18:1/24:1 also had an elevated risk of cardiovascular disease. Most ceramide subtype levels increased in obesity compared to normal-weight individuals, but the levels of serum Cer d18:0/18:0 and Cer d18:1/16:0 decreased in obesity. Conclusions The relationships between ceramide subtypes and metabolic profiles might be heterogeneous in populations with different body weights. Cer d18:1/24:1 could be a biomarker that can be used to differentiate MUO from MHO, and to better predict who will develop unfavorable health outcomes among obese individuals. Trial registration The First Affiliated Hospital of Nanjing Medical University’s Institutional Review Board authorized this study protocol, and all participants provided written informed consent (2014-SR-003) prior to study entry. Supplementary Information The online version contains supplementary material available at 10.1186/s12944-023-01921-0.

and they may represent a unique group or a group that is in the process of transitioning to metabolically unhealthy obesity (MUO) [3].In contrast, MUO was reported as a less favorable type of obesity with impaired lipid and glucose https://app.ithenticate.com/en_us/report/101295808/similarity3/15 31 metabolism and a high risk of cardiovascular and in ammatory abnormalities and other metabolic disorders [4].There are many de nitions for MUO, most of which are based on phenotypes with unfavorable laboratory ndings, such as in ammatory markers, metabolic parameters, insulin sensitivity, brinolytic activity, and liver function [2].Some researchers also believed that the different risks of cardiometabolic diseases are the main identi ed factors for MHO and MUO [5].However, the various de nitions of MHO and MUO are still controversial [3].The current classi cations lack standardization and may not be su cient or accurate to identify the speci c obese subgroup.There is thus a need to have a better biological marker that can make the classi cation of obesity more accurate, with better prediction of health outcomes so that interventions can be initiated earlier to improve the disease outcomes.Sphingolipids (SLs), the minor components of membranes, have crucial biological functions such as altering the physiochemical characteristics of lipid bilayers and in uencing the activation of intracellular proteins and receptors.Although SLs comprise only 2-15% of the total cellular lipidome, they play important roles in the development of metabolic diseases and CVD [6].Ceramide, one of the best characterized SLs, represents a heterogeneous group of lipids that are identi ed by the speci c fatty acyl moiety bonded to sphingosine with an amide bond.The different fatty acyl moieties encompass short to long fatty acids (C2-C34) [7].
Ceramides, acting as secondary messengers for cellular signaling, are related to both lipid and glucose metabolism [8].
As reported before, ceramides were proven to play crucial roles in cell proliferation, autophagy, apoptosis, senescence, migration, regulating mitochondrial dynamics, lipid utilization, glucose sensitivity and in ammation [9-11]and were associated with depression [12], cancer [13], and neurodegenerative disorders [14].However, in recent years, ceramides are getting increased attention for their important roles in metabolic dysfunctions such as obesity, hepatic steatosis, diabetes, and CVD [15].Although increased levels of ceramides were shown to be associated with obesity [16], no study has assessed ceramides levels with regard to MHO or MUO.This study measured the ceramide levels in both MHO and MUO to evaluate the e ciency of ceramides in differentiating the subtypes of obesity with unfavorable health outcomes.Methods Data source.Data were derived from a cohort of participants who had undergone both hyperinsulinemic-euglycemic clamp and lipidomic analysis in the First A liated Hospital of Nanjing Medical University, Nanjing, China [17].The participants who underwent the hyperinsulinemic-euglycemic clamp had not taken any medication or supplement, had no history of smoking or high alcohol intake (four or more standard drinks per week for men and and two or more standard drinks per week for women), severe disease, acute in ammation or pregnancy.
Risk assessment models.The risk of arteriosclerotic cardiovascular disease (ASCVD) was assessed by the prediction model of China-PAR.The total study sample size of the China-PAR project was more than 127 thousand and the longest follow-up time was more than 23 years [21,22].This effective tool has already been proven to have good performance in predicting ASCVD risk in the Chinese population [21,22].The risks of individuals were assessed using the following website: http://www.cvdrisk.com.cn(10-year ASCVD risk: high risk≥10.0%, medium risk: 5.0%~9.9%,low risk<5.0%).
Lifetime ASCVD risk was de ned as follows: the risk of developing ASCVD from now to 85 years old, with the cutoff for low and high risk set to 32.8%.)[21,22].The differences in ceramide levels among the three groups are shown in Table 2. .033)contributed to the presence of the MUO phenotype in obesity (Fig. 3).In this study, the obesity subtypes were classi ed based on metabolic syndrome (MS).Therefore, the parameters in MS were not taken to adjust the models.were adjusted in a multivariate logistic regression analysis.The results are displayed in Table 3. Cer d18:1/24:1 was independently correlated with MUO in obesity.Ceramides and obesity related diseases.In addition, the ASCVD risks between different serum levels of ceramides were assessed.Six participants aged 18-20 years old were excluded according to the prediction model.One participant was excluded due to missing data.The 10year ASCVD risk was generally low as our study population tended to be young.The obese subjects were divided into two groups based on serum levels of Cer d18:1/24:1 which were below or above the median.The lifetime ASCVD risks were signi cantly higher in the above median group than in the below median group (Table 4 ).To further explore the relationship between Cer d18:1/24:1 and liver brosis in NAFLD, we calculated FIB4 and NFS in 47 obese subjects (nine subjects were excluded for missing platelet data).However, we didn't come to the conclusion since no individual in the study population meet the standard of brosis (due to the young age).Discussion Ceramides and outcomes of obesity subtypes Most previous studies have explored the relationships between ceramides and obesity.This study explored the roles of ceramide in two obesity subtypes and its impact on obesity related diseases, especially CVD.According to previous studies, there are many classi cation criteria for MHO and MUO, most of which are based on MS, insulin sensitivity or in ammatory indices [2].A important stage in the development of MS is insulin resistance.Ceramides bound to LDL stimulate the expressions of in ammatory genes in macrophages, such as IL-6 and TNF, and increase insulin resistance in skeletal myocytes [26].The review of 2022 suggested that sphingolipid pro ling, especially ceramide pro ling, was considered as a potential tool for MetS-associated risk strati cation [27].Many studies have proven that activation of β-cell apoptosis is a diabetogenic effect of elevatedceramide levels [28].Furthermore, speci c ceramide subtypes are related to insulin resistance .Early in 2017, dihydroceramide levels, reported as predictors of diabetes, are elevated up to 9 years prior to the onset of diabetes according to a research conducted in two cohort studies by Wigger et al [29].We have also previously reported the relationships between sphingolipids and insulin sensitivity [17].In addition to the aforementioned classi cation criteria for MHO and MUO, some researchers also believed that the different risks of cardiometabolic diseases are the main identi ed factors for MHO and MUO [5].As there are many types of ceramide-based scores that can reliably predict the risk or severity of CVDs [30], the relationship between the levels of ceramides and the risks of CVD in Chinese people was examined to prove that ceramide might be a potential biomarker for MUO in this study.Circulating ceramides have been proved to be used as biomarkers for the development and progression of CVD according to the recent studies.They have already been suggested to predict the cardiovascular events more accurately .Although the increased levels of ceramides are proved to be associated with obesity, the ceramide subtype levels in MHO and MUO have not been assessed.In this study, the levels of total ceramides and Cer d18:1/24:1 were higher in MUO than those in MHO.Meanwhile, the correlation between Cer d18:1/24:1 levels and the presence of MUO further indicated that the metabolism of Cer d18:1/24 :1 might play an important role in the MUO.To test the relationships between Cer d18:1/24:1 and comorbidities related to obesity, we used the Chinese ASCVD risk prediction 14 equations (China-PAR) to assess cardiovascular outcomes in groups with high or low circulating Cer d18:1/24:1 .We found that the circulating Cer d18:1/24 :1 level was higher in obese individuals with higher ASCVD risks.Many researches indicated that people in Asian Paci c region, including those in China, have a higher risk developing obesity-related diseases, even when their BMI is lower than that of Caucasians.We conducted our study based on the Chinese standard for obesity [18,38].We presumed that certain ceramide might provide similar information as China-PAR to predict cardiovascular events.Similarly, higher circulating Cer d18:1/24:1 levels were found in obese subjects with higher ASCVD risks and had already been proven to be a biomarker of high risk of CVDs according to a large number of could be a potential biomarker to differentiate MUO with unfavorable health outcomes from obesity.However, the limitations of this study cannot be ignored.The predominantly Chinese population in this study limits the conclusion generalized to the whole populations.The sample size was small.Some demographic data were collected based on participants recall, which is subject to bias.Conclusions Even though there might be the same risk of developing cardiovascular disease eventually for both MHO and MUO, MUO is worse than MHO due to its rapid progression to poor outcomes.The precise detection of obesity outcomes and early intervention are in need to improve the prognosis.This study showed that the levels of Cer d18:1/24 :1 were higher in the MUO group than in the MHO group and were closely associated with higher risks of CVD.

Age and GIR30
The results indicated that the level of Cer d18:1/24:1 might be a potential biomarker to differentiate MUO from MHO, and to better predict the unfavorable health outcomes of MUO.Individuals with obesity with increased levels of Cer d18:1/24:1 might pay attention to losing weight as early as possible.Although our ndings are interesting, a large scale, longitudinal, prospective study is still required to further explore and validate the predictive and prognostic value of Cer d18:1/24:1 as a biomarker to differentiate MUO from MHO, and to identify the best cutoff value.Meanwhile, whether ceremide could be considered a therapeutic target for metabolic disorders of obesity remains unknown.190 191 192 193 194 195 196 197 198 199 200 201

107. 7
million children and 603.7 million adults were identi ed as obese in 2015 according to the Global Burden of Disease Obesity Collaborators .Approximatedly 4.0 million deaths, mostly due to cardiovascular disease (CVD), are related tohigh body mass index (BMI ) [1].These ndings indicated the importance of effective treatments for decreasing obesity prevalence and disease burdens.Obesity is associated with many kinds of metabolic abnormalities and diseases, including type 2 diabetes mellitus (T2DM), nonalcoholic fatty liver disease (NAFLD), and CVD [2].Obese individuals who exhibit fewer metabolic dysfunctions are considered metabolically healthy obesity (MHO), heatmap of ceramides (Fig.1A) visualizes the concentrations.The level of Cer d18:1/24:1 was lower in both the NC and MHO groups than in the MUO group.Total serum ceramide showed asimilar trend to Cer d18:1/24 :1.Most notably, there were signi cant increases in serum Cer d18 :0/ 18:0 and Cer d18:1 /16: 0 levels in the NC group compared with the MHO group.We also compared the levels of Cer d18 :0/ 18:0 and Cer d18:1 /16: 0 between normal weight and obese individuals.Our results showed that the levels of Cer d18 :0/18: 0 and Cer d18:1 /16: 0 were signi cantly lower in people with obesity.The obese individuals were then divided into 5 groups on the basis of the numbers of metabolic risk abnormalities (Method: ve criteria of metabolic syndrome) (Fig.1B).The percentage of individuals with two to four metabolic abnormalities was 76.8%.There is no signi cance different distribution of https://app.ithenticate.com/en_us/report/10129580810(P=0.435).The level of Cer d18:1/24:1 showed an increasing trend with the accumulation of abnormality numbers (Fig.1C).Correlations between Ceramides and multiple clinical characteristics.As the serum level of Cer d18:1/24:1 increased, the TC, LDL-c,FPG, TG, SBP, and DBPalso increased in parallel.With regard to insulin sensitivity, the levels of Cer d18:1/16:0, Cer d18 :0/ 18:0 and Cer d18 :0/ 24:1 had positive relations with the changes in GIR30.Meanwhile, Cer d18:1 /16: 0 had a positive correlation with HDL-c).The less favorable metabolic phenotypes including high serum TC, TG, LDL-c, AST, ALT, SBP, DBP, WC, and UA, were positively associated with BMI and inversely correlated with the changes in GIR30.All the correlations among clinical characteristics are presented in the correlation heatmap (Fig.2).Binary logistic regression in obesity.The relationship between serum ceramide levels or demographic parameters and the presence of MUO versus MHO in the obese population (n=56) was further examined.In the univariate logistic regression models, . A total of 77 individuals were nally enrolled.
(BMI), systolic/diastolic blood pressure (SBP/DBP) and waist circumference (WC).These individuals were divided into three groups based on their BMI and metabolic parameters: normal weight control (NC) group, MHO and MUO.None of Continuous variables are represented as the mean ± standard deviation (SD) or median (quartile 1, quartile 3), and categorical variables are represented as the frequency.Individuals with values of variables above or below the mean ± 5 SD were regarded as outliers, and these individuals were excluded.The Mann-Whitney U test was used for data that were non-normally distributed, while Student's t test was used for data that were normally distributed when the differences between two groups were analyzed.The Kruskal-Wallis test was used for data that were non-normally distributed, while one-way ANOVA was used for data normally distributed when multiple group differences were analyzed.Categorical variables were compared by the chi-square test.The relationships between ceramides and metabolic characteristics were analyzed by Pearson correlation (normally distributed data) or Spearman correlation (non-normally distributed data).Models of binary logistic regression were tted to estimate the associations between ceramides and the presence of MUO.The software IBM SPSS 22.0 and statistical software R 4.1.0were used to conduct the statistical analysis.Statistical signi cance was de ned as P<0.05.Study approval.This study approval was granted by the Institutional Review Board of the First A liated Hospital of Nanjing Medical University, and all participants provided written informed consent (2014-SR-003) prior to study entry.Results Clinical characteristics of the subjects.
and nonalcoholic fatty liver disease score (NFS) were used to assess hepatic brosis.FIB-4= [age (years) × AST(U/L) ]/[PLT (×109/L) ×√ALT (U/L)].The cutoff values for stage 0-2 hepatic brosis and signi cant brosis are <1.3 and >2.67 respectively[23,24].NFS = − 1.675 + 0.037 × age (y) + 0.094 × BMI (kg/m2) + 1.13 × impaired fasting glucose or diabetes (yes=1, no=0) + 0.99 × AST/ALT ratio − 0.013 × platelet count (×109/L) −0.66 × albumin (g/dL).The presence of advanced brosis was detected with good accuracy using the high cut-off point of NFS (0.676), while it was excluded using the low cut-off point(-1.455)ofNFS.[25].Statistics.https://app.ithenticate.com/en_us/report/101295808/similarityclassied as the NC group, 20 (26%) were classi ed as MHO and another 36 (47%) were classi ed as MUO.Basic clinical characteristics based on metabolic phenotypes are displayed in Table 1(the proposed normal ranges of values given by the laboratory for blood parameters are displayed in Supplementary Table 1).There was no statistical difference in BMI between the MHO and MUOgroups.The distributions of sexes were not signi cantly different among the three groups .GIR30 ( which represents whole body insulin sensitivity.see methods.)tended to decrease in MHO/MUO groups compared to the NC group.However , GIR30 showed no signi cant difference between MHO and MUO.LDL-c, the biomarker to assess CVD risk, increased signi cantly in both MHO and MUO.In general, all the metabolic parameters in the obesity group were worse than those in the NC group .Ceramides among the three groups.The https://app.ithenticate.com/en_us/report/101295808/ To t the clinical characteristics of obesity subtypes, we divided 56 obese individuals into two groups based on the classical de nition of MUO (MS standard)[19].Compared to the NC group , the individuals in either the MHO or MUO group had higher blood glucose, and blood pressure, and unfavorable liver function.Although MHO is regarded as a more favorable subtype than MUO, the health outcomes of MHO remain unclear and controversial.According to previous studies, the risks of metabolic diseases such as T2DM, CVD or NAFLD in individuals with MHO appear to be higher than those in normal-weight individuals[34][35][36][37].Nevertheless, active and effective interventions should be given in a timely manner for individuals with MUO.Ceramide subtypes and metabolism Ceramides are essential elements of biological membranes and signaling molecules involved in a variety of cellular processes[16] https://app.ithenticate.com/en_us/report/101295808/

concentration of Cer d18:1/24:1 and have
studies[8].Based on previous ndings, there are many types of ceramide-based scores that can reliably predict the risk or severity of CVDs, including the ceramide test score (CERT1)[30,39], the new ceramide test score (CERT2)[33], and the Sphingolipid Inclusive Score (SIC)[40].The rst two calculated scores involve the already been used in Mayo Clinic[30].NAFLD is another obesity-related disease especially in MUO.In this study, the hepatic brosis scores including FIB4 and NFS were also calculated, but there was no rm conclusion since no individual in this study met the standard of brosis due to their young age.However, based on previous studies, a high levels were observed in the normal weight group compared to the obese subjects in this study.As the BMI and age of study population for these two studies differ from ours, we assumed that these subtypes of ceramide 16 might play different roles in populations with different ages and body weights.Further prospective studies in various populations are still needed.Recent studies have demonstrated the heritability of plasma concentrations of C18:1/22:0 and C18:1/24:0 which are also associated These data suggested that the subtypes of ceramides might have different impacts on the body metabolic mechanism and the progression of diseases.Ceramide-targeting treatment The study results indicated that more attention should be focused on ceramides because many functions of ceramides are still unknown and should be further explored.Studies in mice demonstrated that the inhibition of ceramide synthesis improved hepatic steatosis and slowed the progression of cardiometabolic diseases[15].Liraglutide can prevent hepatic in ammation and brosis by inhibiting the accumulation of Cer 18:1/16:0 and Cer 18:1/24:0 in the liver of methionine-choline de cient dietary mice[49].Liraglutide, through inhibiting ceramide levels, may alleviate the adverse effects of cardiac dysfunctions[45].As most recent studies have targeted the ceramides to treat insulin resistance, fatty liver disease or some comorbidities of obesity[50], further study of ceramide subtypes and mechanisms can be useful to explore speci c cures.