6828 Exploring Metabolomic Clues in Diabetic Retinopathy

• Published in: Journal of the Endocrine Society Vol. 8; no. Supplement_1
• Main Authors: Simonson, M W, Li, Y, McAnany, J J, Prasad, B, Hanlon, E C, Pannain, S, Layden, B T, Van Cauter, E, Park, J C, Crowley, S J, Chau, F Y, Danielson, K K, Chen, H, Chlipala, G E, Martinez, C, Reutrakul, S
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 05-10-2024
• Subjects: Abstract
• ISSN: 2472-1972; 2472-1972
• DOI: 10.1210/jendso/bvae163.769

Disclosure: M.W. Simonson: None. Y. Li: None. J.J. McAnany: None. B. Prasad: None. E.C. Hanlon: None. S. Pannain: None. B.T. Layden: None. E. Van Cauter: None. J.C. Park: None. S.J. Crowley: None. F.Y. Chau: None. K.K. Danielson: None. H. Chen: None. G.E. Chlipala: None. C. Martinez: None. S. Reutrakul: Speaker; Self; Eli Lilly & Company.
Background: Diabetic retinopathy (DR) is a common microvascular complication of diabetes. Metabolomics offers the possibility for novel biomarker discovery. The purpose of this study was to determine metabolomic differences that characterized patients with versus without DR. Methods: A total of 62 serum samples were collected from 36 type 2 diabetes (T2D) patients with DR and 26 T2D patients without DR, and analyzed via UPLC-MS. The metabolite data were normalized using median normalization. The differential expression of metabolites was compared among the groups using the limma package in R. The covariates (age, sex, BMI and sleep apnea severity) were accounted for by using a generalized linear model fitted to the data. Significantly different metabolites were filtered by an absolute log2 Fold Change (FC) of 1.5 or greater and Q values < 0.05. Results: Mean (SD) age for DR and no-DR group was 55 (±5.9) versus 54 (±6.7) years, while BMI was 33 (±5.8) versus 33 (±6.3) kg/m2 respectively. Both DR and no-DR groups were 58% female. A1C levels in DR group was marginally higher than in no-DR group (p=0.09). There were significant differences in metabolomic changes between DR versus no-DR, with 166 total differential metabolites after adjusting for covariates and filtering by FC. Notable metabolites included those involved in fatty acid metabolism, acyl carnitines, prostaglandins, and phospholipids. There was an overall upregulation of serum acyl carnitines in DR as compared to no-DR, including 2,6 dimethylheptanoyl carnitine (FC = 3.0) and cis-5-Tetradecenoylcarnitine (FC = 2.0). Additionally, there was a decrease in long chain fatty acids in DR as compared to no-DR, which included 6-hydroxypentadecanedioic acid (FC = -6.6), 10,16-dihydroxy-palmitic acid (FC = -6.0), and tetracosatetraenoyl CoA (FC = -6.1). The 11-beta prostaglandin F2 (11-PGF2) was increased in DR (FC = 1.9). Circulating phospholipids including lysophosphatidylcholine (22:2) (lysoPC) (FC = 1.8) and phosphatidylcholine (24:1/24:1) (PC) (FC = 5.9) were also significantly dysregulated. Conclusions: We identified a subset of metabolites that discriminate T2D patients with or without DR. The changes in acyl carnitines and long chain fatty acids may reflect mitochondrial dysfunction in DR. The presence of certain prostaglandins highlights the inflammatory environment which may facilitate DR progression. Increases in phospholipids such as lysoPC suggest an atherogenic state and higher LDL oxidation, while increases in PC may indicate altered lipogenesis in DR. These findings may serve as new potential serum biomarkers to promote efficient preventive care and provide mechanistic insight into understanding DR pathophysiology.
Presentation: 6/2/2024