Gianluigi La Piana

Mitochondrial diseases (MDs) may result from mutations affecting nuclear or mitochondrial genes, encoding mitochondrial proteins, or non-protein-coding mitochondrial RNA. Despite the great variability of affected genes, in the most severe cases, a neuromuscular and neurodegenerative phenotype is observed, and no specific therapy exists for a complete recovery from the disease. The most used treatments are symptomatic and based on the administration of antioxidant cocktails combined with antiepileptic/antipsychotic drugs and supportive therapy for multiorgan involvement. Nevertheless, the real utility of antioxidant cocktail treatments for patients affected by MDs still needs to be scientifically demonstrated. Unfortunately, clinical trials for antioxidant therapies using α-tocopherol, ascorbate, glutathione, riboflavin, niacin, acetyl-carnitine and coenzyme Q have met a limited success. Indeed, it would be expected that the employed antioxidants can only be effective if they are able to target the specific mechanism, i.e., involving the central and peripheral nervous system, responsible for the clinical manifestations of the disease. Noteworthily, very often the phenotypes characterizing MD patients are associated with mutations in proteins whose function does not depend on specific cofactors. Conversely, the administration of the antioxidant cocktails might determine the suppression of endogenous oxidants resulting in deleterious effects on cell viability and/or toxicity for patients. In order to avoid toxicity effects and before administering the antioxidant therapy, it might be useful to ascertain the blood serum levels of antioxidants and cofactors to be administered in MD patients. It would be also worthwhile to check the localization of mutations affecting proteins whose function should depend (less or more directly) on the cofactors to be administered, for estimating the real need and predicting the success of the proposed cofactor/antioxidant-based therapy.

Valinomycin (VLM, 1) is a K⁺ ionophore cyclodepsipeptide capable of depolarizing mitochondria and inducing apoptosis to several mammalian cell types, including a number of tumor cell lines. With the aim of creating VLM‐based ligand‐targeted anticancer drugs that may selectively convey VLM to pathological cells, we have previously introduced derivatizable hydroxyl handles into the VLM structure, allowing to access a three‐entity library of monohydroxyl VLMs (HyVLMs) bearing the OH group at the isopropyl side chain of a d‐Hyi, d‐Val, or l‐Val residue (analogs 2–4, respectively). Herein, the levels of bioactivity retained by the conjugable HyVLMs have been assessed on the basis of their ability to alter the functionality of isolated rat‐liver mitochondria. Experiments run with HyVLMs in the range 1–10 nM and in 20 or 125 mM KCl medium show that the hydroxyl group reduces the potency of HyVLMs relative to VLM to an extent that depends upon the molecular site involved in the hydroxylation. On the other hand, estimation of the stability constants of complexes (in methanol at 25 °C) of each analog with Na⁺, K⁺, and Cs⁺ reveals that HyVLMs nicely retain the VLM binding features, except for a moderate increase in the stability of Na⁺ complexes. These findings, along with pertinent structural considerations, suggest that the incorporation of OH into the VLM structure might actually have altered its K⁺ transporting ability across mitochondrial membranes. Besides facing new aspects of VLM structure–activity relationship, these studies set the basis for the rational design of ligand‐HyVLMs conjugates through derivatization of hanging OH group. Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.