AUTHOR=Tofte Nete , Suvitaival Tommi , Trost Kajetan , Mattila Ismo Matias , Theilade Simone , Winther Signe Abitz , Ahluwalia Tarunveer Singh , Frimodt-Møller Marie , Legido-Quigley Cristina , Rossing Peter 

TITLE=Metabolomic Assessment Reveals Alteration in Polyols and Branched Chain Amino Acids Associated With Present and Future Renal Impairment in a Discovery Cohort of 637 Persons With Type 1 Diabetes

JOURNAL=Frontiers in Endocrinology

VOLUME=10

YEAR=2019

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2019.00818

DOI=10.3389/fendo.2019.00818

ISSN=1664-2392

ABSTRACT=<p><bold>Background:</bold> Improved understanding of the pathophysiology causing diabetic kidney disease (DKD) is imperative. The aim of this study was to uncover associations between serum metabolites and renal outcomes.</p><p><bold>Methods:</bold> Non-targeted serum metabolomics analyses were performed in samples from 637 persons with type 1 diabetes using two-dimensional gas chromatography coupled to time-of-flight mass-spectrometry. Longitudinal data at follow-up (median 5.5 years) on renal events were obtained from national Danish health registries. A composite renal endpoint (<italic>n</italic> = 123) consisted of estimated glomerular filtration rate (eGFR) decline from baseline (≥30%), progression to end-stage renal disease and all-cause mortality. Metabolites with significant associations (<italic>p</italic> &lt; 0.05) in any of the cross-sectional analyses with eGFR and albuminuria were analyzed for specific and composite endpoints. Adjustments included traditional cardiovascular risk factors and correction for multiple testing.</p><p><bold>Results:</bold> A data-driven partial correlation analysis revealed a dense fabric of co-regulated metabolites and clinical variables dominated by eGFR. Ribonic acid and myo-inositol were inversely associated with eGFR, positively associated with macroalbuminuria (<italic>p</italic> &lt; 0.02) and longitudinally associated with higher risk of eGFR decline ≥30% (HR 2.2–2.7, CI [1.3–4.3], <italic>p</italic> &lt; 0.001). Ribonic acid was associated with a combined renal endpoint (HR 1.8, CI [1.3–2.3], <italic>p</italic> = 0.001). The hydroxy butyrate 3,4-dihydroxybutanoic acid was cross-sectionally associated with micro- and macroalbuminuria, urinary albumin excretion rate and inversely associated with eGFR (<italic>p</italic> &lt; 0.04) while branched chain amino acids were associated with eGFR and lower risk of the combined renal endpoint (<italic>p</italic> &lt; 0.02).</p><p><bold>Conclusions:</bold> Alterations in serum metabolites, particularly polyols and amino acids, were associated with renal endpoints in type 1 diabetes highlighting molecular pathways associated with progression of kidney disease. External validation is needed to further assess their roles and potentials as future therapeutic targets.</p>