Identification of Putative Novel Biomarkers of Organic Anion Transporter 1 and 3 for the Prediction of Transporter-Mediated Drug-Drug Interactions

Abstract

<b>Abstract ID 24507</b> <b>Poster Board 187</b> Endogenous metabolites can be used as phenotypic biomarkers of solute carrier (SLC) transporters to predict transporter-mediated drug-drug interactions (DDIs). The SLC22 subfamily members organic anion transporter (OAT) 1 and 3 are expressed on the basolateral membrane of kidney proximal tubule cells. Both facilitate the uptake of several drugs and endogenous metabolites for renal secretion. Probenecid-mediated inhibition of OAT1/3 increases plasma concentrations of endogenous metabolites, including pyridoxic acid and kynurenic acid.^1,2 The objective of this study was to assess the effects of OAT1/3 inhibition on the metabolic profile in human plasma and urine using targeted and untargeted metabolomics approaches. Plasma and urine samples from 7 healthy adults who participated in an ongoing clinical pharmacokinetic DDI study (n=16) involving oral furosemide (OAT1/3 substrate, 5 mg) and probenecid (OAT1/3 inhibitor, 1000 mg) were selected for untargeted analysis. Plasma (pool of 1, 1.5, and 2 h collections) and urine (0-4 h) samples representing timepoints that captured maximum probenecid plasma concentrations were analyzed <i>via</i> liquid chromatography-mass spectrometry. Using XCMS Online software (xcmsonline.scripps.edu), &gt;10,000 individual features (<i>m/z</i> values) were detected in the pooled plasma samples that were either elevated or repressed in the presence of probenecid (Fig 1). Co-administration of probenecid increased furosemide plasma concentrations by 1.5- to 3-fold (p &lt; 0.05), indicating a DDI. However, pyridoxic acid and kynurenic acid, known biomarkers of OATs, showed no statistically significant differences due to high interindividual variability. This variability is likely because the precursors of these biomarkers are derived from diet.¹ More sensitive, selective, and less variable putative OAT1/3 biomarkers were identified by shortlisting the detected features that 1) were elevated in plasma by 2- to 10-fold in presence of probenecid (p &lt; 0.05) and 2) showed acceptable chromatographic peak signal to noise ratios. Correlations between the fold increase in furosemide plasma concentration and fold increase in plasma concentrations of the putative hits were next evaluated. Of these hits, 16 putative biomarkers (<i>m/z</i> 100-505) that correlated with furosemide (r &gt; 0.50) were selected for a METLIN database search. The identified putative biomarkers were mostly anionic compounds with variable hydrophilicity (LogP: -1.0 to 6.5). Five of the putative biomarkers were carboxylic acids, two were glucuronide conjugates, three belonged to a steroidal pathway, and one each belonged to the bile acid and tryptophan metabolism pathways. A detailed pharmacokinetic analysis (0-24 h) of furosemide, pyridoxic acid, and kynurenic acid concentrations measured in plasma and urine samples from 2 participants revealed a 2.2-, 2.3-, and 1.7-fold increase in the AUC, respectively, in the presence of probenecid. Targeted metabolomics analysis of the identified metabolites in all study participants will validate their utility as biomarkers of OAT1/3 for predicting transporter-mediated DDIs. Ref: Shen <i>et&nbsp;al</i>., 2019. <i>JPET</i> 368:136-45. Tang <i>et&nbsp;al</i>., 2021. <i>DMD</i> 49:1063-69. This work was funded by NIH/NICHD R01HD081299.