Reassessing Retinal Pigment Epithelial Ketogenesis: Enzymatic Assays for Ketone Body Levels Provide Inaccurate Results

The retinal pigment epithelium (RPE) is omnivorous and can utilize a wide range of substrates for oxidative phosphorylation. Certain tissues with high mitochondrial metabolic load are capable of ketogenesis, a biochemical pathway that consolidates acetyl-CoA into ketone bodies. Earlier work demonstrated that the RPE expresses the rate-limiting enzyme for ketogenesis, 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2), and that the RPE indeed produces ketone bodies, including beta-hydroxybutyrate ({beta}-HB). Prior work, based on detecting {beta}-HB via enzymatic assays, suggested that differentiated cultures of primary RPE preferentially export {beta}-HB across the apical membrane. Here, we compare the accuracy of measuring {beta}-HB by enzymatic assay kits to mass spectrometry analysis. We found that commercial kits lack the sensitivity to accurately measure the levels of {beta}-HB in RPE cultures and are prone to artifact. Using mass spectrometry, we found that while RPE cultures secrete {beta}-HB, they do so equally to both apical and basal sides. We also find RPE is capable of consuming {beta}-HB as levels rise. Using isotopically labeled glucose, amino acid, and fatty acid tracers, we found that carbons from both fatty acids and ketogenic amino acids, but not from glucose, produce {beta}-HB. Altogether, we substantiate {beta}-HB secretion in RPE but find that the secretion is equal apically and basally, RPE {beta}-HB can derive from ketogenic amino acids or fatty acids, and accurate {beta}-HB assessment requires mass spectrometric analysis.