Monash Centre for Health Research and Implementation (MCHRI), School of Public Health and Preventive Medicine, Monash University, 43-51 Kanooka Grove, Clayton, VIC 3168, Australia.
Metabolomics Laboratory, Baker Heart and Diabetes Institute, 75 Commercial Road, Melbourne, VIC 3004, Australia.
Biochim Biophys Acta Mol Cell Biol Lipids. 2019 Mar;1864(3):335-343. doi: 10.1016/j.bbalip.2018.12.014. Epub 2018 Dec 23.
Advances in mass spectrometry and lipidomics techniques are providing new insights into the role of lipid metabolism in obesity-related diseases. However, human lipidomic studies have been inconsistent, owing to the use of indirect proxy measures of metabolic outcomes and relatively limited coverage of the lipidome. Here, we employed comprehensive lipid profiling and gold-standard metabolic measures to test the hypothesis that distinct lipid signatures in obesity may signify early stages of pathogenesis toward type 2 diabetes.
Using high-performance liquid chromatography-electrospray tandem mass spectrometry, we profiled >450 lipid species across 26 classes in 65 overweight or obese non-diabetic individuals. Intensive metabolic testing was conducted using direct gold-standard measures of adiposity (% body fat by dual X-ray absorptiometry), insulin sensitivity (hyperinsulinaemic-euglycaemic clamps), and insulin secretion (intravenous glucose tolerance tests), as well as measurement of serum inflammatory cytokines and adipokines (multiplex assays; flow cytometry). Univariable and multivariable linear regression models were computed using Matlab R2011a, and all analyses were corrected for multiple testing using the Benjamini-Hochberg method.
We present new evidence showing a strong and independent positive correlation between the lysophosphatidylinositol (LPI) lipid class and insulin secretion in vivo in humans (β [95% CI] = 781.9 [353.3, 1210.4], p = 0.01), supporting the insulinotropic effects of LPI demonstrated in mouse islets. Dihydroceramide, a sphingolipid precursor, was independently and negatively correlated with insulin sensitivity (β [95% CI] = -1.9 [-2.9, -0.9], p = 0.01), indicating a possible upregulation in sphingolipid synthesis in obese individuals. These associations remained significant in multivariable models adjusted for age, sex, and % body fat. The dihexosylceramide class correlated positively with interleukin-10 before and after adjustment for age, sex, and % body fat (p = 0.02), while the phosphatidylethanolamine class and its vinyl ether-linked (plasmalogen) derivatives correlated negatively with % body fat in both univariable and age- and sex-adjusted models (all p < 0.04).
Our data suggest that these lipid classes may signify early pathogenesis toward type 2 diabetes and could serve as novel therapeutic targets or biomarkers for diabetes prevention.
质谱和脂质组学技术的进步为脂质代谢在肥胖相关疾病中的作用提供了新的见解。然而,由于代谢结果的间接替代指标的使用以及脂质组相对有限的覆盖范围,人类脂质组学研究一直不一致。在这里,我们采用全面的脂质分析和金标准代谢测量来检验这样一个假设,即在肥胖症中存在不同的脂质特征可能标志着 2 型糖尿病发病的早期阶段。
使用高效液相色谱-电喷雾串联质谱法,我们在 65 名超重或肥胖的非糖尿病个体中对 26 类超过 450 种脂质种类进行了分析。使用直接金标准方法对肥胖进行了强化代谢测试(双能 X 线吸收法测量的体脂百分比[%])、胰岛素敏感性(高胰岛素正葡萄糖钳夹试验)和胰岛素分泌(静脉葡萄糖耐量试验),以及血清炎症细胞因子和脂肪因子的测量(多重分析;流式细胞术)。使用 Matlab R2011a 计算单变量和多变量线性回归模型,使用 Benjamini-Hochberg 方法对所有分析进行了多重检验校正。
我们提供了新的证据,表明在人类体内,溶血磷脂酰肌醇(LPI)脂质类与胰岛素分泌之间存在强烈而独立的正相关(β[95%置信区间] = 781.9[353.3, 1210.4],p = 0.01),支持在小鼠胰岛中观察到的 LPI 的胰岛素促分泌作用。神经酰胺二氢,一种鞘脂前体,与胰岛素敏感性呈独立负相关(β[95%置信区间] = -1.9[-2.9, -0.9],p = 0.01),表明肥胖个体中鞘脂合成可能上调。这些关联在调整年龄、性别和体脂百分比后的多变量模型中仍然显著。二己糖神经酰胺类在调整年龄、性别和体脂百分比前后与白细胞介素-10 呈正相关(p = 0.02),而磷脂酰乙醇胺类及其乙烯基醚连接(血浆)衍生物在单变量和年龄及性别调整模型中均与体脂百分比呈负相关(均 p<0.04)。
我们的数据表明,这些脂质类可能标志着 2 型糖尿病发病的早期阶段,可作为糖尿病预防的新型治疗靶点或生物标志物。