Machine learning-assisted SERS approach enables the biochemical discrimination in Bcl-2 and Mcl-1 expressing yeast cells treated with ketoconazole and fluconazole antifungals

Antifungal medications are important due to their potential application in cancer treatment either on their own or with traditional treatments. The mechanisms that prevent the effects of these medications and restrict their usage in cancer treatment are not completely understood. The evaluation and discrimination of the possible protective effects of the anti-apoptotic members of the Bcl-2 family of proteins, critical regulators of mitochondrial apoptosis, against antifungal drug-induced cell death has still scientific uncertainties that must be considered. Novel, simple, and reliable strategies are highly demanded to identify the biochemical signature of this phenomenon. However, the complex nature of cells poses challenges for the analysis of cellular biochemical changes or classification. In this study, for the first time, we investigated the probable protective activities of Bcl-2 and Mcl-1 proteins against cell damage induced by ketoconazole (KET) and fluconazole (FLU) antifungal drugs in a yeast model through surface-enhanced Raman spectroscopy (SERS) approach. The proposed SERS platform created robust Raman spectra with a high signal-to-noise ratio. The analysis of SERS spectral data via advanced unsupervised and supervised machine learning methods enabled unquestionable differentiation (100 %) in samples and biomolecular identification. Various SERS bands related to lipids and proteins observed in the analyses suggest that the expression of these anti-apoptotic proteins reduces oxidative biomolecule damage induced by the antifungals. Also, cell viability assay, Annexin V-FITC/PI double staining, and total oxidant and antioxidant status analyses were performed to support Raman measurements. We strongly believe that the proposed approach paves the way for the evaluation of various biochemical structures/changes in various cells.