Ioanna Barla ¹ Ioanna V. Dagla ² Aikaterini M. Daskalopoulou ³ Maria Panagiotopoulou ³ Maria Kritikaki ³ Panagiotis Dalezis ⁴ Nikolaos S. Thomaidis ¹ Antony Tsarbopoulos ⁴ Dimitrios Trafalis ⁴ Evagelos Gikas ^1*

• ¹ Laboratory of Analytical Chemistry, School of Science, Department of Chemistry, National and Kapodistrian University of Athens, Athens, Greece
• ² GAIA Research Center, The Goulandris Natural History Museum, Kifissia, Greece
• ³ Laboratory of Pharmaceutical Analysis, School of Health Science, Department of Pharmacy, National and Kapodistrian University of Athens, Athens, Greece
• ⁴ Laboratory of Pharmacology, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece

Colistin (CMS) is used for the curation of infections caused by multidrug-resistant bacteria. CMS is constrained by toxicity, particularly in kidney and neuronal cells. The recommended human doses are 2.5-5 mg/kg/day, and the toxicity is linked to higher doses. So far, the in vivo toxicity studies have used doses even 10-fold higher than human doses. It is essential to investigate the impact of metabolic response of doses, that are comparable to human doses, to identify biomarkers of latent toxicity. The innovation of the current study is the in vivo CMS impact using a range of CMS doses that have never been investigated before, i.e., 1 mg/kg and 1.5 mg/kg. The 1 mg/kg and 1.5 mg/kg, administered in mice, correspond to the therapeutic and toxic human doses, based on previous expertise of our team, regarding the human exposure. The study mainly focused on the biochemical impact of CMS on the metabolome, and on the alterations provoked by 50%-fold of dose increase. The main objectives were (i) the comprehension of the biochemical changes resulting after CMS administration and (ii) from its dose increase; and (iii) the determination of dose-related metabolites that could be considered as toxicity monitoring biomarkers. The in vivo experiment employed two doses of CMS versus a control group treated with normal saline, and samples of plasma, kidney, and liver were analyzed with a UPLC-MS-based metabolomics protocol. Both univariate and multivariate statistical approaches (PCA, OPLS-DA, PLS regression, ROC) and pathway analysis were combined for the data interpretation. The results pointed out six dose-responding metabolites (PAA, DA4S, 2,8-DHA, etc.), dysregulation of renal dopamine , and extended perturbations in renal purine metabolism. Also, the study determined altered levels of liver suberylglycine, a metabolite linked to hepatic steatosis. One of the most intriguing findings was detection of elevated levels of renal xanthine and uric acid, that act as AChE activators, leading to the rapid degradation of acetylcholine. This evidence provides a naïve hypothesis, for the potential association between the CMS induced nephrotoxicity and CMS induced neurotoxicity, that should be further investigated.