Supervisors info:
Ευάγγελος Γκίκας, Καθηγητής Τμήμα Χημείας ΕΚΠΑ,
Ειρήνη Παντερή Καθηγήτρια Τμήμα Φαρμακευτικής ΕΚΠΑ,
Εμμανουήλ Μικρός Καθηγητής Τμήμα Φαρμακευτικής ΕΚΠΑ
Summary:
Colistin or polymyxin E is a well-known cyclic polypeptide antibiotic drug. Its use was limited to last-line treatment, due to the extensive toxicity and mainly the nephrotoxicity it causes. However, in the last decade, the use of inhaled colistin has been reconsidered, for the treatment of respiratory infections due to multidrug-resistant Gram (-) pathogenic bacteria.
To investigate the toxicity of colistin in relation to the dosage, but also, the mechanisms that induce this toxicity, it is appropriate to study the effect of the drug at the level of metabolites. Metabolomic studies have been widely used in toxicity, and dose response (positive/negative) studies. In general, metabolomics offer knowledge about the normal or disturbed functioning of an organism, under the influence of exogenous and endogenous factors. The analysis of the total metabolic profile is a sensitive tool for predicting toxicity and can help the field of drug development in the pre-clinical and clinical stage. Comparing the metabolic profiles of two groups, for example a group receiving drug and a control group, can lead both to the finding of metabolic intermediates and metabolites, whose expression is dysregulated, and which are possible indicators of toxicity (biomarkers).
In this thesis, a metabolomic study was developed for the toxic effect of colistin. Specifically, the variables that differ during the effect of an increased dose of colistin, in kidney samples, were determined, after in vivo induction in mice. 15 mice were used (6 female and 9 male), which were divided into 3 groups. These were the reference group C (control group), maximum non-toxic dose group A, which was administered 1 mg/kg, and a toxic dose group B, which received 1.5 mg/kg. Then, the kidneys were obtained, according to the specific protocol of kidney tissue extraction (MeOH/H2O system) and their metabolic profiles were determined using UPLC-QTOF-MS in positive and negative ESI ionization. Processing of the acquired spectra was performed using MSDial 4.90 software. Spectral data was subjected to multiparametric statistical analysis with supervised and unsupervised Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) and Orthogonal-PLS (OPLS-DA) methods. In the case of the identified metabolites, ROC analysis was applied, to highlight the statistically significant biomarkers and the factors that impact them.
The results of the study showed differentiation of the metabolic profiles of the mice that had not received the drug but also those that received different dose. (1030) metabolites, with statistical significance were identified, which participate in the metabolic pathways of purine metabolism, phenylalanine, tyrosine, tryptophan and acetyl-tRNA biosynthesis, as well as tryptophan metabolism.
Keywords:
colistin, nephrotoxicity, metabolomics, liquid chromatography, mass spectrometry, chemometrics, statistical analysis
