Commonly prescribed medications often cause off-target mitochondrial dysfunction, but the underlying molecular mechanisms are largely unknown. Mitochondrial transport proteins form a significant class of potential off-targets that remain largely unexplored. The main aim of this thesis is to evaluate the role of human mitochondrial transporters in adverse drug effects, to overcome the technical challenges in their functional characterisation and to identify compounds that can inhibit substrate transport.

So far, most drug interactions have been reported for the mitochondrial ADP/ATP carrier (AAC), therefore it was validated first as a model for studying drug-carrier interactions. For the first time, pure human ADP/ATP carrier 1 (hAAC1∆1-10), heterologously expressed in yeast, was obtained in a functional form, and two types of studies were carried out. Firstly, thermostability shift assays were used to investigate the binding of drugs, previously reported to inhibit AAC1. Secondly, their effect on transport was assessed in proteoliposomes with reconstituted hAAC1∆1-10, enabling characterization of their inhibition kinetics. This approach confirmed that chebulinic acid, CD-437 and suramin are potent hAAC1∆1-10 inhibitors with IC₅₀-values in the low micromolar range.

Next, GSK compiled a library of 33 medications that failed clinical trials or were discontinued due primarily to mitochondrial toxicity. To investigate whether mitochondrial carrier inhibition could cause mitochondrial dysfunction, hAAC1∆1-10 and the human citrate carrier (hCIC), another abundant carrier, were purified and assayed using the strategies described above. Remarkably, from this limited subset of mitotoxic compounds, benzbromarone was shown to inhibit hAAC1∆1-10 and hCIC with an IC50 of 8.4 μM and 12.5 μM respectively. Moreover, tolcapone inhibits hCIC with an IC50 of 5 μM, three times more potent than the canonical inhibitor benzene tricarboxylic acid.

Moreover, detailed understanding of the transport processes underpinning the translocation of substrates of other carriers may also provide opportunities to study potential drug off-targets. Therefore, an investigation was carried out into the broad substrate specificity and proton coupling properties of hCIC, which are poorly understood. As inhibitors above have features in common with endogenous substrates, it is also important to identify orphan transporters as new potential off-targets. Hence, human SLC25A44, a transporter that was recently proposed to transport branched-chain amino acids, was also purified, and identified as a potential candidate for structural work, as it is remarkably stable in some short-chain detergents, therefore, crystallisation trials were set up.

Consequently, this study has demonstrated that mitochondrial transport proteins can be drug off-targets, since prescription drugs inhibit multiple mitochondrial transport proteins with low micromolar affinity. Thus, these results highlight the importance of exploring drug-transporter interactions further, identifying orphan transporters, and understanding the transport processes. Better evaluation methods of drug-induced inhibition of mitochondrial transport proteins will ultimately contribute to the development of drugs with an improved safety profile.