Saliva and Plasma Reflect Metabolism Altered by Diabetes and Periodontitis

• Published in: Frontiers in molecular biosciences Vol. 8; p. 742002
• Main Authors: Sakanaka, Akito, Kuboniwa, Masae, Katakami, Naoto, Furuno, Masahiro, Nishizawa, Hitoshi, Omori, Kazuo, Taya, Naohiro, Ishikawa, Asuka, Mayumi, Shota, Tanaka Isomura, Emiko, Shimomura, Iichiro, Fukusaki, Eiichiro, Amano, Atsuo
• Format: Journal Article
• Language: English
• Published: Frontiers Media S.A 13-09-2021
• Subjects: biomarkers; inflammation; metabolome; Molecular Biosciences; non-alcoholic fatty liver disease; saliva
• ISSN: 2296-889X; 2296-889X
• DOI: 10.3389/fmolb.2021.742002

Periodontitis is an inflammatory disorder caused by disintegration of the balance between the periodontal microbiome and host response. While growing evidence suggests links between periodontitis and various metabolic disorders including type 2 diabetes (T2D), non-alcoholic liver disease, and cardiovascular disease (CVD), which often coexist in individuals with abdominal obesity, factors linking periodontal inflammation to common metabolic alterations remain to be fully elucidated. More detailed characterization of metabolomic profiles associated with multiple oral and cardiometabolic traits may provide better understanding of the complexity of oral-systemic crosstalk and its underlying mechanism. We performed comprehensive profiling of plasma and salivary metabolomes using untargeted gas chromatography/mass spectrometry to investigate multivariate covariation with clinical markers of oral and systemic health in 31 T2D patients with metabolic comorbidities and 30 control subjects. Orthogonal partial least squares (OPLS) results enabled more accurate characterization of associations among 11 oral and 25 systemic clinical outcomes, and 143 salivary and 78 plasma metabolites. In particular, metabolites that reflect cardiometabolic changes were identified in both plasma and saliva, with plasma and salivary ratios of (mannose + allose):1,5-anhydroglucitol achieving areas under the curve of 0.99 and 0.92, respectively, for T2D diagnosis. Additionally, OPLS analysis of periodontal inflamed surface area (PISA) as the numerical response variable revealed shared and unique responses of metabolomic and clinical markers to PISA between healthy and T2D groups. When combined with linear regression models, we found a significant correlation between PISA and multiple metabolites in both groups, including threonate, cadaverine and hydrocinnamate in saliva, as well as lactate and pentadecanoic acid in plasma, of which plasma lactate showed a predominant trend in the healthy group. Unique metabolites associated with PISA in the T2D group included plasma phosphate and salivary malate, while those in the healthy group included plasma gluconate and salivary adenosine. Remarkably, higher PISA was correlated with altered hepatic lipid metabolism in both groups, including higher levels of triglycerides, aspartate aminotransferase and alanine aminotransferase, leading to increased risk of cardiometabolic disease based on a score summarizing levels of CVD-related biomarkers. These findings revealed the potential utility of saliva for evaluating the risk of metabolic disorders without need for a blood test, and provide evidence that disrupted liver lipid metabolism may underlie the link between periodontitis and cardiometabolic disease.