**5. Mitochondria as a biosensor for drug development**

effect on K+ transport, diazoxide also exhibits a direct effect on mitochondrial energy metab‐ olism by inhibition of respiratory chain complex II in liver mitochondria. Recently, mitoKATP channel opener BMS-191095 with no peripheral vasodilator activity was described.Using isolated mitochondria or proteoliposomes reconstituted with partly purified mitoKATP channel and measuring potassium flux demonstrated that heart and liver liver mitochondrial KATP channels have some pharmacological similarities with the cell membrane KATP channel, i.e., both channels are activated by KCOs. Mitochondrial KATP channels are 1000 times more sensitive to diazoxide than that of cell membrane KATP channels. This document concluded that the interaction of mitochondrial KATP channels with KCOs plays a key role in cardiopro‐

Mitochondrial KATP channel: A Novel target for Cardioprotection. KCOs mimic hypoxic/ ischemic preconditioning in the absence of ischemia in the heart myocardial cells, the reason why antagonists of KATP channel, like 5-hydroxydecanoic acid and glibenclamide, ameliorate the positive effects of short time hypoxic/ischemic conditions on the heart myocardium. The primary postulation to justify these events includes cell membrane KATP channels. Newly, it was shown that in fact KCOs including diazoxide affect the mitoKATP channel in mitochondria. In a complementary approach, it was shown that diazoxide did not activate plasma membrane KATP channels, but induced oxidation of mitochondrial flavoproteins, due to the activation of mitoKATP channel. These findings established the fact that the target for the diazoxide protec‐ tive effects in heart myocytes is the mitochondrial KATP channel rather than the cell membrane KATP channel. It is also note worthy that evidence for mitochondrial KATP channels as effectors of cardiac myocardial preconditioning has also been proven in human subjects. The initial observations on the cardioprotective action of KCOs on mitochondria were further confirmed and developed in a series of reports. It has been shown that other KCOs such as nicorandil, cromakalim, and pinacidil modulate mitochondrial Ca2+ uptake, respiration, mitochondrial membrane potential, ATP generation, and mitochondrial Ca2+ uptake. The main question remains how the opening of the mitoKATP channel could protect cells against ischemic damage. 1) Opening of the mitoKATP channel followed by mitochondrial swelling could improve mitochondrial ATP handling and/or production. 2) The protective effect of mitoKATP activation could be mediated by lowering Ca2+ overloading of mitochondria. In fact, it was found that diazoxid preserves mitochondrial function in ischaemic rat cardiomyocyte. It is now proven that hypoxia approximately decreases mitochondrial oxygen consumption rate to 40% of the normal value, and administration of diazoxide maintains the prehypoxic mitochondrial oxygen consumption rate during hypoxia/ischemia. Cardiac ATP concentration was signifi‐ cantly raised following diazoxide treatment. Secondly, by lowering Ca2+ overloading of mitochondria the protective effect of mitochondrial KATP activation could be induced. It was shown that the opening of the mitochondrial KATP channel may increase mitochondrial reactive oxygen species (ROS) formation. This increase could lead to protein kinase C activa‐ tion, which is known to be necessary for the cardioprotection. Besides, it seems that mito‐ chondrial KATP channel is enrolled in delayed preconditioning because of an alteration in expression of "protective" proteins (3). It was that pretreatment of hippocampal neurons with

*4.4.2. Mitochondrial KATP channel: A novel target for cardioprotection*

tection [32].

76 Toxicology Studies - Cells, Drugs and Environment

Extensive study over the last 50 years indicates that many medications can induce mitochon‐ drial damage [33]. Medication- induced dysfunctions include the alteration of mitochondrial components and metabolic pathways. These dysfunctions are a major challenge and problem for drug development. There is mounting evidence of the mitotoxicity (table 2).

Interestingly knowledge of the mechanisms that trigger drug-induced mitochondrial damage will be helpful in the development of strategies to decrease the potentially toxic effects of medications. Additional, these issues affect the most aerobically poised organs such as heart and kidneys or organs exposed to higher concentrations of the drug for example liver. Recently using mitochondria as a biosensor for determination safety of drug development has in‐ creased. The reasons are as follows: A) in general, mitochondria control many of the pro-death and anti-death cell signals; B) a number of reports describe an association between patients receiving medication and effects on mitochondrial metabolism 3) drug safety has become a priority of many pharmaceutical companies [4].

It is quite obvious that mitochondria are key elements of cell life which several well known drugs induce toxic effects on them in several non-target and target organs. As soon as possible by improvement of preliminary drug safety assessment the possibility of drug toxic reactions during clinical practice will be avoided. Depending on the targeted organ, severe in vitro mitochondrial impairment may be sufficient to ban an efficient drug in the market or prevent‐ ing a promising drug candidate from further clinical trials. Drug companies now have a new dilemma, which is to realize how much of the evaluated mitochondrial toxicity is a key predictor of the drug pharmacological or adverse effects. Pharmaceutical suppliers have also now a difficult problem which is to know how much of the supposedly mitochondrial impairment is a component of the therapeutic effect. On the other hand, it may be a tough choice to remove dispensing drugs showing a certain degree of mitochondrial toxicity *in vitro* evaluations but with a very unique significant therapeutic effect. Despite showing mitochondrial toxicity, sometimes pharmaceutical companies may decide to push the lead candidate molecule forward for further in *vivo* assays in order to also clarify ways of reducing adverse mitochondrial toxicity [3]. Some pharmacological strategies could be used to decrease mitochondrial toxicity. For example in the cardiac toxicity of doxorubicin (DOX), One possibility is to improve drug targeting, decreasing the amount of drug that reach non-target organs. One successful example of this strategy is the use of pegylated liposomal DOX, which has a quite different pharmacokinetic profile including an increased circulation time and a decreased volume of distribution [34]. Another strategy is the co-administration of protective agents, one example of which being the preventive role of the beta-blocker carvedilol on DOX induced cardiac mitochondrial impairment.


**Table 2.** Examples of Drugs with Black Box Warnings for Mitochondrial Toxicity

Refining of the different methodologies results into higher achievement in the isolation of functional mitochondria from different organs, which can be used in further mechanistic studies to identify tissue-specific drug-induced mitochondrial toxicity. Nevertheless, studies with isolated mitochondria lack the complexity associated with experiments in intact cells, isolated organs or even in vivo studies. But the use of isolated mitochondrial fractions helps determining precise sites of action of the molecule on mitochondria. If everything works OK, the accurate drug safety assessment would correlate data in isolated mitochondria with data collected in intact cells and *in vivo* (figure 3).One important issue in this discussion includes what can be gained by using mitochondria as a biosensor to search drug safety. One particular example was nefazodone, an anti-depressant. This drug was withdrawn from the market in 2004, following several clinical reports of serious liver toxicity [35]. Would the use of mito‐ chondria as cost effective accelerated biosensors of drug safety helped in avoiding the entry of the drug in the market? The answer is, maybe so. Dykens*et al.* demonstrated that nefazodone is highly toxic to isolated liver mitochondria, human hepatocytes and HepG2 cells, showing an obvious cytotoxicity even when cell lines were used. [36] If an initial assessment of mitochondrial toxicity for nefazodone had been done, it is very unlikely that the drug would have been pushed forward for further clinical trials.

26 of a range of compounds can be performed in a fast and relatively 27 inexpensive way, avoiding some later human toxicity problems that may 28 arise during subsequent testing stages or even during clinical use. Some 29 companies will focus more on investigating direct drug-induced 30 mitochondrial dysfunction, others will rather measure drug-induced **Figure 3.** Basic scheme of the drug development process involving mitochondria as an important marker for drug-in‐ duced toxicity. The thickness of each arrow exemplifies the number of different molecules in each evaluation stage. Drug toxicity on mitochondria is proposed as the bottleneck step in decision-making [3].

25 interest to the industry, since a primary assessment of mitochondrial toxicity

31 alterations in mitochondrial-relevant genes. Whatever the chosen strategy is, 32 the final outcome is the prediction of drug safety based on a mitochondrial 33 end-point. 34 35 It is now apparent that mitochondrial toxicology has become an area of interest to the industry, since a primary assessment of mitochondrial toxicity of a range of compounds can be per‐ formed in a fast and relatively inexpensive way, avoiding some later human toxicity problems that may arise during subsequent testing stages or even during clinical use. Some companies will focus more on investigating direct drug-induced mitochondrial dysfunction, others will rather measure drug-induced alterations in mitochondrial-relevant genes. Whatever the chosen strategy is, the final outcome is the prediction of drug safety based on a mitochondrial end-point.
