Lipidomics Combined with Network Pharmacology to Explore Differences in the Mechanisms of Grey Hair Development Between Type 2 Diabetes Mellitus and Normal Populations (Female).

Type 2 diabetes is usually accompanied by premature grey hair. In this study, we analysed differences in the lipid composition of black and white hair follicles between women with type 2 diabetes and healthy populations, using lipidomic methods. We examined the correlation between the lipid composition of female grey hair follicles and type 2 diabetes mellitus, and we screened for potential grey-hair-delaying ingredients using network pharmacology. Forty-one female volunteers with type 2 diabetes (diabetes, D) and thirty-five healthy volunteers (healthy, H) aged 55-65 years were recruited. Hair roots, including the follicular portion, were collected from grey hair (D-W for diabetic volunteers and H-W for healthy volunteers) and black hair (D-B for diabetic volunteers and H-B for healthy volunteer). Lipids were extracted separately and analysed using UPLC-QTOF-MS (Ultra-Performance Liquid Chromatography-Tandem Time-of-Flight Mass Spectrometry), combined with an OPLS-DA (Orthogonal Partial Least Squares Discriminant Analysis) model to identify different lipids among different groups under VIP conditions (VIP > 1, < 0.05, and fold change ≥ 2). Further screening was performed using the ROC (receiver operating characteristic) curve method, selecting lipids with an AUC (area under the curve) value greater than 0.8 and specificity plus sensitivity greater than 1.6. Finally, bioinformatics and reverse network pharmacology were used to screen relevant targets, ingredients, and herbs to find suitable raw materials with anti-grey-hair effects. We found the following: (1) Ten significant differential lipids were identified under VIP conditions in the D-W and D-B groups, and five potential differential lipids (1-O-alpha-D-glucopyranosyl-1,2-eicosandiol, emmotin A, odyssic acid, PI-Cer(t18:0/26:0(2OH)), and NAPE(18:1(9Z)/16:1(9Z)/18:0)) were further screened using ROC analysis. The levels of all five lipids were significantly higher in D-W than in D-B, and these elevated levels may have been related to the production of grey hair in diabetic patients. (2) Thirteen significantly different lipids were screened under VIP conditions in the H-W and H-B groups, and five potential differential lipids were screened via ROC analysis (PS(O-16:0/13:0), PA(12:0/16:1(9Z)), PS(13:0/20:3(8Z,11Z,14Z)), GlcCer(d18:1/24:1(15Z)), and PS(O-20:0/17:2(9Z,12Z))). The levels of all five lipids were significantly higher in H-B than in H-W, and we hypothesised that their reduced levels were associated with the production of grey hair in the healthy population. (3) Twelve significantly different lipids were screened under VIP conditions in the D-W and H-W groups, and two potential differential lipids were screened via ROC analysis (fucoxanthinol 3-heptadecanoate 3'-myristate and 2-(3-hydroxyphytanyl)-3-phytanyl-sn-glycerol). The contents of both lipids were significantly higher in H-W than in D-W, and there were differences in the lipid composition of grey hair in the D and H populations. (4) Important ingredients with possible therapeutic effects were obtained through lipid-matched target screening: resveratrol, calycosin, epigallocatechin 3-gallate, and herbs such as the fruit of the glossy privet, etc. In summary, the production of grey hair in the D and H populations may be affected by different lipids. The lipid components emmotin A and fucoxanthinol 3-heptadecanoate 3'-myristate were significantly higher in the D and H populations than in the same groups (D-B, H-B), and these are pregnenolone lipids (PRs). We hypothesised that PRs can influence the production of grey hair in both populations. The screening of important differential lipids may serve to provide diagnostic loci or therapeutic targets, while matching ingredients and herbs may provide a basis and direction for the subsequent development of anti-grey-hair ingredients.