Analysis of Mitochondrial Function in Cell Membranes as Indicator of Tissue Vulnerability to Drugs in Humans

Citation: Elexpe, A.; SánchezSánchez, L.; Tolentino-Cortez, T.; Astigarraga, E.; Torrecilla, M.; Barreda-Gómez, G. Analysis of Mitochondrial Function in Cell Membranes as Indicator of Tissue Vulnerability to Drugs in Humans. Biomedicines 2022,10, 980. https://doi.org/10.3390/ biomedicines10050980 Academic Editors: Silvia OrtegaGutierrez, Amirata Saei Dibavar and María L. López-Rodríguez Received: 31 March 2022 Accepted: 21 April 2022 Published: 23 April 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). biomedicines Article Analysis of Mitochondrial Function in Cell Membranes as Indicator of Tissue Vulnerability to Drugs in Humans Ane Elexpe 1,2, Laura Sánchez-Sánchez 1,3 , Tarson Tolentino-Cortez 1, Egoitz Astigarraga 1, María Torrecilla 2 and Gabriel Barreda-Gómez 1,* 1Research and Development Department, IMG Pharma Biotech S.L, 48160 Derio, Spain; [email protected] (A.E.); [email protected] (L.S.-S.); [email protected] (T.T.-C.); [email protected] (E.A.) 2Department of Pharmacology, Faculty of Medicine and Nursing, University of the Basque Country UPV/EHU, 48940 Leioa, Spain; [email protected] 3Institute of Molecular Genetics and Biology (IBGM), University of Valladolid-CSIC, 47003 Valladolid, Spain *Correspondence: gabriel.barr[email protected]; Tel.: +34-94-4316-577; Fax: +34-94-6013-455 Abstract: Drug side effects are one of the main reasons for treatment withdrawal during clinical trials. Reactive oxygen species formation is involved in many of the drug side effects, mainly by interacting with the components of the cellular respiration. Thus, the early detection of these effects in the drug discovery process is a key aspect for the optimization of pharmacological research. To this end, the superoxide formation of a series of drugs and compounds with antidepressant, antipsychotic, anticholinergic, narcotic, and analgesic properties was evaluated in isolated bovine heart membranes and on cell membrane microarrays from a collection of human tissues, together with specific inhibitors of the mitochondrial electron transport chain. Fluphenazine and PB28 promoted similar effects to those of rotenone, but with lower potency, indicating a direct action on mitochondrial complex I. Moreover, nefazodone, a drug withdrawn from the market due to its mitochondrial hepatotoxic effects, evoked the highest superoxide formation in human liver cell membranes, suggesting the potential of this technology to anticipate adverse effects in preclinical phases. Keywords: microarray; superoxide; antipsychotic; mitochondria 1. Introduction Reactive oxygen species (ROS) formation is a physiological process produced by different pathways and controlled by diverse antioxidant mechanisms; however, it can turn into a pathological state due to an imbalance between oxidant and antioxidant compounds [ 1 ]. These compounds are a series of radical and nonradical oxygen species formed upon incomplete oxygen reduction [ 2 ] whose augmentation can induce lipid peroxidation and multiple alterations in proteins and nuclei acids. They can be produced by several sources, such as NADPH oxidase (NOX) enzyme family [ 3 ], dual oxidase (DUOX) [ 4 ], monoamine oxidase (MAO) [ 5 ], peroxisomes, and mitochondrial electron transport chain (mETC) [ 6 ]. The different types of NADPH oxidases are implicated in reactive oxygen species formation in a variety of tissues, such as brain, heart, or liver [ 7 ]; specifically, NOX2, NOX3, and NOX4 are expressed throughout the nervous system [ 7 ], whereas other isoforms such as NOX5 are found only in the lymph nodes and spleen [ 8 ]. Additionally, dual oxidase enzymes (DUOX) produce hydrogen peroxide directly or indirectly [ 7 ]. Monoamine oxidase is an enzyme located in the mitochondrial outer membrane, with two different isoforms (MAO-A and MAO-B). One of its catalytic products is hydrogen peroxide [ 9 , 10 ] whose accumulation provokes the damage of many cell types including neural cells [ 11 ] and is implicated in many brain pathologies [ 12 ], specifically Alzheimer’s [ 13 ] and Parkinson’s disease [ 14 ], Friedreich Ataxia [ 15 ], multiple sclerosis [ 16 ], and some psychotic disorders such as bipolar disorder [ 17 ] and schizophrenia [ 18 , 19 ]. Among these proteins, the main Biomedicines 2022,10, 980. https://doi.org/10.3390/biomedicines10050980 https://www.mdpi.com/journal/biomedicines Biomedicines 2022,10, 980 2 of 16 ones responsible for cytosolic hydrogen peroxide and other ROS are those involved in the mitochondrial electron transport chain and cytochrome P450 enzymatic system [ 20 ]. On the one hand, the oxidative phosphorylation (OXPHOS) process, performed in the mitochondrial inner membrane and composed of five enzymatic complexes, is the main source of energy as well as reactive oxygen species. Mitochondria are organelles present in most eukaryotic cells that can perform a variety of metabolic functions [ 21 ], such as oxidative phosphorylation [ 22 , 23 ] and metabolite regulation. They are also implicated in homeostatic signaling [ 22 ] and lipid biosynthetic pathways [ 23 ]. Principally, superoxide is produced by mitochondrial complexes I and III [ 24 ]. Their dysregulation leads to an ATP production decrease, an increase in oxidative stress, and may even initiate apoptotic processes leading to drug side effects [ 25 ]. On the other hand, cytochrome P450 enzymes are involved in the metabolization of compounds [ 20 ]. These enzymes oxygenate organic xenobiotic substrates and catalyze the reduction of molecular oxygen simultaneously; if it is not performed correctly, oxygen uncouples from the substrate and leads to ROS formation [26]. In physiological states, enzymatic and nonenzymatic antioxidant mechanisms regulate the cellular redox status by controlling the production of second messengers [ 27 , 28 ] and transcription factors [ 29 ] in diverse signaling pathways. For instance, superoxide dismutase (SOD) scavenges superoxide radicals [ 29 ], while catalase [ 30 ], glutathione peroxidase, and peroxyredoxin [ 24 ] protect the cell from the adverse effects of hydrogen peroxide through its breakdown into water and oxygen. Nevertheless, the exposure to some stress conditions, such as different pathologies or even medication intake, can make these mechanisms insufficient. In a certain way, many drugs that are being tested in clinical trials are withdrawn in preclinical phases because of their high toxicity due to the formation of reactive oxygen species. Considering that the major source of these small molecules is the mitochondrial electron transport chain, their dysfunction can generate an energetic reduction that can have a particularly pronounced impact on the efficiency of neuronal functions compared to other tissues [ 31 ]. Schizophrenia is a cognitive disorder that affects 1% of the population in which dissociation or thought disruption is experienced [ 32 ]. This illness denotes alterations in the neurotransmission of dopamine inside the mesolimbic system and mesocortical pathway [ 33 ]. Moreover, first-episode schizophrenic patients have decreased SOD activity making them prone to suffer from oxidative stress conditions [ 34 ]. It has been observed that some antipsychotics have been associated with mETC inhibition causing an increase in reactive oxygen species formation [ 35 ] by complex I inhibition [ 36 ]. Complex I inhibition contributes to a further reduction of mitochondrial energy production, which is already limited in some cognitive disorders, such as bipolar disorder [ 31 ] and schizophrenia [ 31 , 37 ]. For instance, clozapine can induce oxidative stress and apoptosis in neutrophil cells [ 38 ], whereas another antipsychotic drugs, such as pentazocine [ 39 ], seem to reduce it. In addition, these oxidative dysbalanced conditions have been related to extrapyramidal adverse effects [40,41]. Oxidative stress is also a contributing factor in other mental disorders, such as depression [ 42 ], the most common of this type of disease. It is characterized by apathy, anhedonia, sleep disturbance, and psychomotor retardation [ 43 ]. Among the antidepressant drugs, nefazodone, an antagonist of the 5-hydroxitriptamine (5-HT) receptor, was used for several years to treat depression until it was withdrawal from the market. It was reported that nefazodone caused liver toxicity and hepatic failure [ 44 ] due to the inhibition of cytochrome P450 [ 45 ], interference with OXPHOS enzymes, and generation of reactive oxygen species [ 44 ]. Other drugs, such as clozapine or fluphenazine, have also been associated with some hepatic alterations [ 46 ] and cardiac alterations such as cardiomyopathy or myocarditis [47]. Alzheimer’s disease, the most common neurodegenerative cause of dementia [ 48 ], is another mental illness in which oxidative stress seems to play a particular role [48,49] . This neurogenerative disorder has alterations such as the deposit of amyloid-beta plaques, dysregulation of central nervous system immune response [ 50 ], dysfunction of oxidative Biomedicines 2022,10, 980 3 of 16 phosphorylation, and reactive oxygen species generation due to mitochondrial disfunction [ 51 ]. Regarding the pathologic effects of Alzheimer’s disease, ligands for sigma 1 and 2 receptors can regulate the CNS immune response and modulate amyloid-beta production [ 35 , 50 ]. Furthermore, it has been reported that sigma receptor ligands strongly increased mitochondrial superoxide radicals [52]. As these medicines are usually tested in samples from animal models in vitro and in vivo , their side effects could be different to those present in human samples. In this regard, different fluorescence spectroscopy and colorimetric techniques [ 53 ] have been carried out to determine the production of ROS in a variety of organisms. In this sense, the aim of this study was to analyze the superoxide formation evoked by drugs and compounds with antipsychotic, anticholinergic, narcotic, and analgesic properties in isolated bovine heart membranes and on human cell membrane microarrays (CMMAs). These CMMAs (Figure 1) consisted of a collection of membranes isolated from 10 human tissues, which maintain the membrane environment and protein functionality, enabling their use in superoxide assays [54]. Biomedicines 2022, 10, x FOR PEER REVIEW 3 of 18 This neurogenerative disorder has alterations such as the deposit of amyloid-beta plaques, dysregulation of central nervous system immune response [50], dysfunction of oxidative phosphorylation, and reactive oxygen species generation due to mitochondrial disfunction [51]. Regarding the pathologic effects of Alzheimer’s disease, ligands for sigma 1 and 2 receptors can regulate the CNS immune response and modulate amyloid-beta production [35,50]. Furthermore, it has been reported that sigma receptor ligands strongly increased mitochondrial superoxide radicals [52]. As these medicines are usually tested in samples from animal models in vitro and in vivo, their side effects could be different to those present in human samples. In this regard, different fluorescence spectroscopy and colorimetric techniques [53] have been carried out to determine the production of ROS in a variety of organisms. In this sense, the aim of this study was to analyze the superoxide formation evoked by drugs and compounds with antipsychotic, anticholinergic, narcotic, and analgesic properties in isolated bovine heart membranes and on human cell membrane microarrays (CMMAs). These CMMAs (Figure 1) consisted of a collection of membranes isolated from 10 human tissues, which maintain the membrane environment and protein functionality, enabling their use in superoxide assays [54]. Figure 1. Scheme of the methodology used to evaluate the tissue-specific drug-mediated superoxide formation in human tissues. CMMAs were composed of human cell membranes from 10 different organs and tissues. Drugs and compounds with antipsychotic, antidepressant, anticholinergic, narcotic, and analgesic properties were incubated with CMMAs, and superoxide formation was detected by a colorimetric assay. CMMAs were digitalized, and the data after normalization were analyzed to determine ROS formation in each human tissue. 2. Materials and Methods 2.1. Drugs and Reagents Nitrotetrazolium blue chloride (NBT), 3,3′-Diaminobenzidine (DAB), beta-nicotinamide adenine dinucleotide (NADH), sodium succinate dibasic (SDH), decylubiquinone (dUQ), rotenone, antimycin A, sodium azide, cytochrome c from equine heart, olanzapine, and fluphenazine dichloride were purchased from Sigma-Aldrich (Saint Louis, MO, USA). (−)-Pentazocine, PPCC oxalate, PB 28 dihydrochloride, N’N dimethyltryptamine (DMT), BD 1047 dihydrobromide, nefazodone hydrochloride, dextromethorphan hydrochloride, and NE 100 hydrochloride were purchased from Tocris Bioscience (Bristol, UK). Figure 1. Scheme of the methodology used to evaluate the tissue-specific drug-mediated superoxide formation in human tissues. CMMAs were composed of human cell membranes from 10 different organs and tissues. Drugs and compounds with antipsychotic, antidepressant, anticholinergic, narcotic, and analgesic properties were incubated with CMMAs, and superoxide formation was detected by a colorimetric assay. CMMAs were digitalized, and the data after normalization were analyzed to determine ROS formation in each human tissue. 2. Materials and Methods 2.1. Drugs and Reagents Nitrotetrazolium blue chloride (NBT), 3,3 0 -Diaminobenzidine (DAB), beta-nicotinamide adenine dinucleotide (NADH), sodium succinate dibasic (SDH), decylubiquinone (dUQ), rotenone, antimycin A, sodium azide, cytochrome c from equine heart, olanzapine, and fluphenazine dichloride were purchased from Sigma-Aldrich (Saint Louis, MO, USA). ( − )-Pentazocine, PPCC oxalate, PB 28 dihydrochloride, N’N dimethyltryptamine (DMT), BD 1047 dihydrobromide, nefazodone hydrochloride, dextromethorphan hydrochloride, and NE 100 hydrochloride were purchased from Tocris Bioscience (Bristol, UK). 2.2. Tissue Samples Heart samples from Bos taurus were supplied by Llodio municipal slaughterhouse (Alava, Spain). Human biopsy tissues were supplied by the AMSBIO (Abingdon, Oxfordshire, UK ) tissue bank according to its ethical protocols (Table S2). Biomedicines 2022,10, 980 4 of 16 2.3. Cell Membrane Microarray Fabrication Cell membrane microarrays were composed of a collection of cell membrane homogenates isolated from different human tissues (liver, jejunum, lung, renal medulla, renal cortex, adrenal gland, myocardium, adipose tissue, duodenum, and spleen). Briefly, samples were homogenized using a disperser (Ultra-Turrax ® T10 basic, IKA, Staufen, Germany) or a Teflon-glass grinder (Heidolph RZR 2020, Schwabach, Germany) in 20 volumes of homogenization buffer (1 mM EGTA, 3 mM MgCl 2 , and 50 mM Tris-HCl, pH 7.4) supplemented with 250 mM sucrose. The crude homogenate was subjected to a 1500-rpm centrifugation (AllegraTM X 22R centrifuge, Beckman Coulter, Brea, CA, USA) for 5 min at 4 ◦ C, and the resultant supernatant was collected and centrifuged at 18,000 g (Microfuge ® 22R centrifuge, Beckman Coulter, Brea, CA, USA) for 15 min (4 ◦ C). The pellet was washed in 20 volumes of homogenized buffer and recentrifuged under the same conditions. The tubes were finally decanted, and the pellets were frozen at −80 ◦C, except for one aliquot, which was used to determine the protein concentration. Protein concentration was determined by the Bradford method [55,56] and adjusted to the final concentration. Membrane homogenates were resuspended in buffer and printed onto glass slides using a noncontact microarrayer (Nanoplotter NP 2.1), placing 3 replicates of each sample (7 nL/spot) onto preactivated glass microscope slides. Membrane homogenates of each tissue were obtained from three different individuals. The printing was carried out under controlled humidity (relative humidity 60%) at a controlled temperature of 4 ◦ C. CMMAs were stored at − 20 ◦ C until usage [ 54 ]. CMMAs were validated before usage by different methods including Bradford staining for protein determination, enzyme activity assays (NADH oxidoreductase, succinate dehydrogenase, and cytochrome c oxidase), and radioligand binding assays. [54,57–61] 2.4. Determination of Drug Effects on Superoxide Formation Promoted by NADH in Isolated Bovine Heart Cell Membranes The NADH-ubiquinone oxidoreductase activity assay was performed on cell membranes isolated from bovine heart tissue. For this purpose, membrane homogenates (0.1 mg/mL) were incubated in the presence of 0.35 mM NADH and 0.5 mg/mL NBT in phosphate buffer (5 mM; pH 7.4) for 3 h at 25 ◦ C with increasing concentrations (from 0.1 nM to 1 mM) of drugs and compounds with antipsychotic, anticholinergic, narcotic, and analgesic properties (olanzapine, clozapine, desclozapine, fluphenazine, pentazocine, PB 28, DXT, DMT, donepezil, BD 1047, PPCC, and NE 100) in the presence and absence of 50 µ M dUQ. The reaction started by the addition of membrane homogenates, and NBT oxidation was measured every 5 min spectrophotometrically at 595 nM in a Multiskan FC microtiter plate reader (Thermo Scientific ® , Waltham, MA, USA). Biochemical data on NADH oxidoreductase-evoked superoxide production were presented as a percentage of basal activity in the absence of the drug under study, with or without decylubiquinone. Every CMMA had three replicates of each tissue, and all the experiments were performed in duplicate. 2.5. Determination of Tissue-Specific Effects of Drugs on Superoxide Formation Triggered by NADH Using Human CMMAs The NADH-ubiquinone oxidoreductase activity was performed using cell membrane microarrays from a human tissue collection (heart, liver, jejunum, duodenum, renal medulla, renal cortex, adrenal gland, adipose tissue, spleen, and lung). CMMAs were incubated in the presence of 0.35 mM NADH and 0.05 mg/mL NBT in phosphate buffer (5 mM; pH 7.4) for 1 h at 25 ◦ C, with the compounds under study (olanzapine, fluphenazine, pentazocine, PB28, and nefazodone) at 30 µ M in the presence or absence of 50 µ M dUQ. The reaction was started by the addition of the reagents to the CMMAs. After the incubation time, the reaction was stopped by a dipping in dH 2 O. Once dried, the CMMA color signal was acquired with an Epson V750 pro scanner, and digital images were analyzed with the Biomedicines 2022,10, 980 5 of 16 software Adobe Photoshop CS5 (Adobe Systems Incorporated, Mountain View, CA, USA) and quantified using software ImageScanner (IMG Pharma S.L, Derio, Spain). 2.6. Determination of Tissue-Specific Effect on Superoxide Formation Induced by Succinate Using Human CMMAs Succinate dehydrogenase activity was performed on cell membrane microarrays from a human tissue collection (heart, liver, jejunum, duodenum, renal medulla, renal cortex, adrenal gland, adipose tissue, spleen, and lung). CMMAs were incubated in the presence of 1 mM succinate, 0.05 mg/mL NBT, and 50 µ M dUQ in phosphate buffer (5 mM; pH 7.4) for 16 h at 25 ◦ C. The reaction was started by the addition of the reagents to the CMMAs. After the incubation time, the reaction was stopped by a dipping in dH 2 O. Once dried, the CMMA color signal was acquired with an Epson V750 pro scanner, and digital images were analyzed with the software Adobe Photoshop CS5 (Adobe Systems Incorporated, Mountain View, CA, USA) and quantified using software ImageScanner (IMG Pharma S.L, Derio, Spain). 2.7. Determination of Drug Tissue-Specific Effect on Cytochrome C Oxidase Activity Cytochrome c oxidase activity was assayed on cell membrane microarrays from a human tissue collection. CMMAs were incubated in the presence of 1.3 mM of DAB and 0.01% of cytochrome c in phosphate buffer (0.1 M; pH 7.4) for 16 h at 37 ◦ C in darkness. After the incubation time, the reaction was stopped by a dipping in dH 2 O. Once dried, the CMMA color signal was acquired with an Epson V750 pro scanner, and digital images were analyzed with the software Adobe Photoshop CS5 (Adobe Systems Incorporated, Mountain View, CA, USA) and quantified using software ImageScanner (IMG Pharma S.L, Derio, Spain). 2.8. Data Analysis and Normalization Data handling and analysis were carried out using Excel and GraphPad software (version 9.2). The identification of outliers was carried out applying the following formulas: CV =SD X Y1=X−DF ×SD Y2=X+DF ×SD SD = standard deviation; DF = deviation factor; and CV = variation coefficient. Points were identified as outliers and excluded if the variation coefficient (CV) was higher than 0.15, Y 1 was higher than the point analyzed, or Y 2 was lower than the point examined. We used a deviation factor of 1 in our analysis. In experiments performed with bovine heart membranes homogenates, a nonlinear analysis (log (agonist) vs. response and log (inhibitor) vs. response) was performed. For microarrays, the analysis data obtained were normalized to the amount of total protein and were expressed as means of independent data points ± S.E.M. The normality of the data was tested using Shapiro–Wilk statistical test with α : 0.05. For Gaussian distributed data, a statistical analysis was performed by one-way, two-tailed ANOVA with Tukey’s multiple comparison test. To analyze nonparametrical data, the Kruskal–Wallis test with Dunn’s multiple comparison test was performed. Statistical differences were indicated by p-values ≤0.05. 3. Results 3.1. Protocol Optimization of Superoxide Formation in Bovine Heart Membranes Homogenates The effect of the different compounds studied relating to superoxide formation was determined in isolated bovine heart membranes following the protocol described in the materials and methods section. Biomedicines 2022,10, 980 6 of 16 Firstly, succinate dehydrogenase activity assays were performed determining DCIP reduction (Figure S1), and NADH consumption assays (Figure S2) were performed to ensure that membranes were preserved and functional. After validation, the formation of superoxide in these membranes was evaluated by determining the reduction of NBT in the presence of NADH as substrate, with or without decylubiquinone (Figure S3) to study NADH dehydrogenase activity (complex I). As expected, the presence of the decylubiquinone transporter increased the superoxide formation rate. Moreover, mETC superoxide generation was tested in the presence of rotenone (5 µ M) and antimycin A (5 µ M) as inhibitors of complexes I and III, respectively (Figure S3). Afterwards, NADH dehydrogenase activity assays were performed with different concentrations of these inhibitors. To determine the maximum effect, dose–response curves were plotted, and nonlinear regression was used for each one (Figure 2A, Table 1). Moreover, superoxide formation velocities were determined for each concentration; a dose–response curve was then plotted, and nonlinear regression was used in order to calculate the maximum superoxide formation velocity (Figure 2B, Table 1). Biomedicines 2022, 10, x FOR PEER REVIEW 6 of 18 3. Results 3.1. Protocol Optimization of Superoxide Formation in Bovine Heart Membranes Homogenates The effect of the different compounds studied relating to superoxide formation was determined in isolated bovine heart membranes following the protocol described in the materials and methods section. Firstly, succinate dehydrogenase activity assays were performed determining DCIP reduction (Figure S1), and NADH consumption assays (Figure S2) were performed to ensure that membranes were preserved and functional. After validation, the formation of superoxide in these membranes was evaluated by determining the reduction of NBT in the presence of NADH as substrate, with or without decylubiquinone (Figure S3) to study NADH dehydrogenase activity (complex I). As expected, the presence of the decylubiquinone transporter increased the superoxide formation rate. Moreover, mETC superoxide generation was tested in the presence of rotenone (5 μM) and antimycin A (5 μM) as inhibitors of complexes I and III, respectively (Figure S3). Afterwards, NADH dehydrogenase activity assays were performed with different concentrations of these inhibitors. To determine the maximum effect, dose–response curves were plotted, and nonlinear regression was used for each one (Figure 2A, Table 1). Moreover, superoxide formation velocities were determined for each concentration; a dose–response curve was then plotted, and nonlinear regression was used in order to calculate the maximum superoxide formation velocity (Figure 2B, Table 1). –7 –5 –3 100 120 140 160 Log [Inhibitor] (M) Superoxide formation mediated by NADH dehydrogenase Rotenone Antimycin A –9 –7 –5 –3 60 80 100 120 140 Log [Inhibitor] (M) Superoxide formation speed mediated by NADH dehydrogenase Rotenone Antimycin A A) B) Figure 2. (A) Concentration–response curves obtained from the maximum superoxide formation evoked by NADH in the presence of mitochondrial electron transport chain inhibitors of the complex I (rotenone) and III (Antimycin A). (B) Superoxide formation rate evoked by NADH in the presence of rotenone and antimycin A determined from the lineal range of the kinetic assays. (A) Nonlinear regression was performed with log(agonist) vs. response model and least squares regression as fitting method. (B) Nonlinear regression was performed with log(inhibitor) vs. response model for rotenone and log(agonist) vs. response model for antimycin A. In both cases, least squares regression was used as fitting method. Table 1. Logarithms of half-maximum effective or inhibitory concentrations (pEC50 or pIC50), maximum superoxide formation (Emax), and maximum superoxide formation velocity (Vmax) determined from concentration–response curves for each specific inhibitor. Superoxide Formation Maximum Amount Production Rate Rotenone pEC50 −7.7 ± 0.23 42.79 ± 4.88 pIC50 −7.47 ± 0.15 41.90 ± 2.94 Emax Vmax Antimycin A pEC50 −5.94 ± 0.14 34.20 ± 2.29 pEC50 −6.30 ± 0.35 21.04 ± 3.35 Emax Vmax Figure 2. ( A ) Concentration–response curves obtained from the maximum superoxide formation evoked by NADH in the presence of mitochondrial electron transport chain inhibitors of the complex I (rotenone) and III (Antimycin A). ( B ) Superoxide formation rate evoked by NADH in the presence of rotenone and antimycin A determined from the lineal range of the kinetic assays. ( A ) Nonlinear regression was performed with log(agonist) vs. response model and least squares regression as fitting method. ( B ) Nonlinear regression was performed with log(inhibitor) vs. response model for rotenone and log(agonist) vs. response model for antimycin A. In both cases, least squares regression was used as fitting method. Table 1. Logarithms of half-maximum effective or inhibitory concentrations (pEC50 or pIC50), maximum superoxide formation (Emax), and maximum superoxide formation velocity (Vmax) determined from concentration–response curves for each specific inhibitor. Superoxide Formation Maximum Amount Production Rate Rotenone pEC50 −7.7 ±0.23 pIC50 −7.47 ±0.15 Emax 42.79 ±4.88 Vmax 41.90 ±2.94 Antimycin A pEC50 −5.94 ±0.14 pEC50 −6.30 ±0.35 Emax 34.20 ±2.29 Vmax 21.04 ±3.35 Both inhibitors increased the total amount of superoxide produced (42.8% rotenone vs. 34.2% antimycin A) but with different potency (Figure 2A, Table 1). By contrast, their effects on reaction velocity were substantially different as rotenone reduced it, whereas antimycin A increased it (Figure 2B, Table 1). Maximum velocities were calculated by a nonlinear model using a velocity vs. log [inhibitor] curve (Figure 2B). The rates at each concentration were achieved from a lineal Biomedicines 2022,10, 980 7 of 16 range of superoxide formation assays (Figure S3) determining the slopes for every different inhibitor concentration. To calculate pharmacological parameters, we used these velocities and the maximum effect (Emax) obtained from the kinetic dose–response curves (Figure 2B). 3.2. Effect of Drugs and Compounds on Superoxide Formation in Isolated Bovine Heart Membranes The capacity of superoxide formation of 12 different medicines with antipsychotic, anticholinergic, narcotic, and analgesic properties was assessed: BD1047, PB28, NE100, and PPCC as sigma receptor antagonists; fluphenazine, pentazocine, olanzapine, clozapine, and desclozapine as first and second-generation antipsychotics; and N, N-dimethyltryptamine (DMT), donepezil, and dextromethorphan (DXT) as other drugs with neurological actions. An increase in reactive oxygen species formation was observed with certain drugs, namely, NE100 (sigma 1 receptor antagonist), pentazocine (analgesic), and PB28 (sigma 2 agonist) with the higher maximum effect of 60.1%, 45.5%, and 43.4%, respectively (Figure 3; Table 2). On the other hand, PPCC (sigma 1 receptor agonist) and fluphenazine (antipsychotic) reached an Emax of 35.4% and 36.8%, respectively. Finally, the ones with a lower Emax were BD1047 (sigma 1 receptor antagonist), olanzapine (antipsychotic), and DXT (antitussive). DMT (sigma 1 receptor agonist), clozapine (antipsychotic), and desclozapine (antipsychotic) did not promote any effect. Biomedicines 2022, 10, x FOR PEER REVIEW 7 of 18 Both inhibitors increased the total amount of superoxide produced (42.8% rotenone vs. 34.2% antimycin A) but with different potency (Figure 2A, Table 1). By contrast, their effects on reaction velocity were substantially different as rotenone reduced it, whereas antimycin A increased it (Figure 2B, Table 1). Maximum velocities were calculated by a nonlinear model using a velocity vs. log [inhibitor] curve (Figure 2B). The rates at each concentration were achieved from a lineal range of superoxide formation assays (Figure S3) determining the slopes for every different inhibitor concentration. To calculate pharmacological parameters, we used these velocities and the maximum effect (Emax) obtained from the kinetic dose–response curves (Figure 2B). 3.2. Effect of Drugs and Compounds on Superoxide Formation in Isolated Bovine Heart Membranes The capacity of superoxide formation of 12 different medicines with antipsychotic, anticholinergic, narcotic, and analgesic properties was assessed: BD1047, PB28, NE100, and PPCC as sigma receptor antagonists; fluphenazine, pentazocine, olanzapine, clozapine, and desclozapine as first and second-generation antipsychotics; and N, N-dimethyltryptamine (DMT), donepezil, and dextromethorphan (DXT) as other drugs with neurological actions. An increase in reactive oxygen species formation was observed with certain drugs, namely, NE100 (sigma 1 receptor antagonist), pentazocine (analgesic), and PB28 (sigma 2 agonist) with the higher maximum effect of 60.1%, 45.5%, and 43.4%, respectively (Figure 3; Table 2). On the other hand, PPCC (sigma 1 receptor agonist) and fluphenazine (antipsychotic) reached an Emax of 35.4% and 36.8%, respectively. Finally, the ones with a lower Emax were BD1047 (sigma 1 receptor antagonist), olanzapine (antipsychotic), and DXT (antitussive). DMT (sigma 1 receptor agonist), clozapine (antipsychotic), and desclozapine (antipsychotic) did not promote any effect. –7 –6 –5 –4 –3 100 120 140 160 Log [Drug] (M) Superoxide formation mediated by NADH dehydrogenase PENTAZOCINE BD1047 DXT –7 –6 –5 –4 –3 100 120 140 160 Log [Drug] (M) Superoxide formation mediated by NADH dehydrogenase OLANZAPINE DESCLOZAPINE –7 –6 –5 –4 –3 100 120 140 160 Log [Drug] (M) Superoxide formation mediated by NADH dehydrogenase PB28 NE100 PPCC DMT –7 –6 –5 –4 –3 100 120 140 160 Log [Drug] (M) Superoxide formation mediated by NADH dehydrogenase CLOZAPINE FLUPHENAZINE DONEPEZIL Figure 3. Concentration–response curves of superoxide formation in isolated bovine heart membranes induced by NADH dehydrogenase modulation with pentazocine, BD1047, DXT, PB28, NE100, PPCC, DMT, olanzapine, desclozapine, clozapine, fluphenazine, and donepezil. The Figure 3. Concentration–response curves of superoxide formation in isolated bovine heart membranes induced by NADH dehydrogenase modulation with pentazocine, BD1047, DXT, PB28, NE100, PPCC, DMT, olanzapine, desclozapine, clozapine, fluphenazine, and donepezil. The superoxide formation promoted by NADH dehydrogenase is represented in percentages versus the activity measured in absence of the tested drug. Nonlinear regression was performed with the log (agonist) vs. response (three parameters) model and least squares regression as fitting method. Regarding the rate of reactive oxygen species formation, most of the drugs studied reduced the rate at high concentrations, except for DXT, olanzapine, clozapine, and desclozapine, which did not affect it (Figure S4). The kinetic parameters were determined, and a reduction in the rate of superoxide formation promoted by fluphenazine and PB28 was observed, reaching 51.9% and 34.0%, respectively. Pentazocine, PPCC, and donepezil also promoted a rate reduction of 23.9%, 21.6%, and 17.3%, respectively (Table S1). By Biomedicines 2022,10, 980 8 of 16 contrast, clozapine, desclozapine, olanzapine, DMT, DXT, and BD1047 did not significantly change the rate of superoxide formation evoked by NADH dehydrogenase. Table 2. Potency (pEC50) and maximum effect (Emax) calculated for each drug on superoxide formation evoked by NADH dehydrogenase activity in isolated bovine heart membranes. pEC50 Emax (%) pEC50 Emax (%) PB 28 −5.4 ±0.2 43.4 ±3.0 Pentzazocine −3.9 ±0.3 45.5 ±11.1 NE 100 −3.6 ±0.3 60.1 ±23.1 Donepezil UD UD PPCC −4.5 ±0.3 35.4 ±5.3 Fluphenazine −4.6 ±0.3 36.8 ±4.9 BD 1047 −6.3 ±0.4 20.6 ±4.1 Clozapine UD UD DMT UD UD Olanzapine −5.5 ±0.4 15.7 ±3.1 DXT −3.9 ±0.3 19.0 ±5.2 Desclozapine −3.2 ±1.0 UD UD = undetermined. 3.3. Tissue-Specific Effects of Drugs on Superoxide Formation Using CMMAs of Human Tissues To ensure that the immobilized membranes that constitute the CMMAs were functional and the mitochondrial membranes were preserved, the superoxide formation triggered by complex I and complex II substrates and the cytochrome c Oxidase activity were assayed (Figure 4). Biomedicines 2022, 10, x FOR PEER REVIEW 8 of 18 superoxide formation promoted by NADH dehydrogenase is represented in percentages versus the activity measured in absence of the tested drug. Nonlinear regression was performed with the log (agonist) vs. response (three parameters) model and least squares regression as fitting method. Table 2. Potency (pEC50) and maximum effect (Emax) calculated for each drug on superoxide formation evoked by NADH dehydrogenase activity in isolated bovine heart membranes. pEC50 Emax (%) pEC50 Emax (%) PB 28 −5.4 ± 0.2 43.4 ± 3.0 Pentzazocine −3.9 ±0.3 45.5 ± 11.1 NE 100 −3.6 ± 0.3 60.1 ± 23.1 Donepezil UD UD PPCC −4.5 ± 0.3 35.4 ± 5.3 Fluphenazine −4.6 ± 0.3 36.8 ± 4.9 BD 1047 −6.3 ± 0.4 20.6 ± 4.1 Clozapine UD UD DMT UD UD Olanzapine −5.5 ± 0.4 15.7 ± 3.1 DXT −3.9 ± 0.3 19.0 ± 5.2 Desclozapine −3.2 ± 1.0 UD * UD = undetermined. Regarding the rate of reactive oxygen species formation, most of the drugs studied reduced the rate at high concentrations, except for DXT, olanzapine, clozapine, and desclozapine, which did not affect it (Figure S4). The kinetic parameters were determined, and a reduction in the rate of superoxide formation promoted by fluphenazine and PB28 was observed, reaching 51.9% and 34.0%, respectively. Pentazocine, PPCC, and donepezil also promoted a rate reduction of 23.9%, 21.6%, and 17.3%, respectively (Table S1). By contrast, clozapine, desclozapine, olanzapine, DMT, DXT, and BD1047 did not significantly change the rate of superoxide formation evoked by NADH dehydrogenase. 3.3. Tissue-Specific Effects of Drugs on Superoxide Formation Using CMMAs of Human Tissues To ensure that the immobilized membranes that constitute the CMMAs were functional and the mitochondrial membranes were preserved, the superoxide formation triggered by complex I and complex II substrates and the cytochrome c Oxidase activity were assayed (Figure 4). Figure 4. Representative image of human CMMAs showing the immobilized protein (Bradford), the activity of the NADH dehydrogenase (NADH), and succinate dehydrogenase (SDH) together with the activity of citochrome c oxidase (CcO) of the mitochondrial electron transport chain. The selection of drugs tested on human CMMAs was based on the results of the experiments with bovine heart membranes, the information available in the literature, a link between the adverse effects of these drugs, and mitochondrial oxidative stress [62,63]. The drugs selected were fluphenazine, pentazocine, olanzapine, PB28, and nefazodone. In the human heart, all drugs except nefazodone promoted an increase in superoxide formation Figure 4. Representative image of human CMMAs showing the immobilized protein (Bradford), the activity of the NADH dehydrogenase (NADH), and succinate dehydrogenase (SDH) together with the activity of citochrome c oxidase (CcO) of the mitochondrial electron transport chain. The selection of drugs tested on human CMMAs was based on the results of the experiments with bovine heart membranes, the information available in the literature, a link between the adverse effects of these drugs, and mitochondrial oxidative stress [ 62 , 63 ]. The drugs selected were fluphenazine, pentazocine, olanzapine, PB28, and nefazodone. In the human heart, all drugs except nefazodone promoted an increase in superoxide formation in the absence of dUQ (13.8% olanzapine <49.1% fluphenazine <72.5% PB28 < 85.0% pentazocine). In the presence of the dUQ electron transporter, olanzapine increased the superoxide formation from 13.8% to 52.0%, while fluphenazine, PB28, and pentazocine reduced it (from 49.1% to 8.3% for fluphenazine; 72.5% to − 4.1% for PB28; and 85.0% to 46.5% for pentazocine) (Figures 5and 6). Regarding other tissues, olanzapine significantly enhanced the superoxide production in liver, duodenum, adrenal gland, and renal medulla. The presence of dUQ did not alter the effect promoted by olanzapine alone in any of the tissues, except in spleen and lung, where a significant increase in superoxide formation was observed (Figure 5A). Fluphenazine only induced a significant increase in superoxide production in liver and Biomedicines 2022,10, 980 9 of 16 heart, while in all other tissues, it did not enhance and even seemed to decrease it. These actions were blocked by dUQ or even reverted in some tissues, such as spleen (Figure 5B). Pentazocine evoked a substantial increase in superoxide production in heart, liver, and jejunum tissues. A reduction in other tissues, such as renal medulla, adrenal gland, or spleen, was achieved with this compound in the absence of dUQ, while the presence of the transporter reverted these actions (Figure 6A). PB28 increased the superoxide formation in heart, liver, jejunum, renal medulla, adipose tissue, and spleen. However, the coincubation of PB28 with dUQ significantly blocked it in heart, jejunum, renal medulla, and adipose tissue (Figure 6B). Finally, nefazodone caused an increase in superoxide formation in liver, jejunum, duodenum, and lung tissues in the absence of dUQ, while in the presence of this electron transporter, it was not affected, except in the case of the duodenum, where it was also enhanced (Figure 6C). In renal medulla, renal cortex, adrenal gland, and adipose tissue, the coincubation of nefazodone with dUQ evoked an increase in superoxide formation. In contrast, in heart tissue, a reduction of superoxide formation was observed (Figure 6C). Biomedicines 2022, 10, x FOR PEER REVIEW 9 of 18 in the absence of dUQ (13.8% olanzapine <49.1% fluphenazine <72.5% PB28 < 85.0% pentazocine). In the presence of the dUQ electron transporter, olanzapine increased the superoxide formation from 13.8% to 52.0%, while fluphenazine, PB28, and pentazocine reduced it (from 49.1% to 8.3% for fluphenazine; 72.5% to −4.1% for PB28; and 85.0% to 46.5% for pentazocine) (Figures 5 and 6). Heart Liver Jejunum Duodenum Renal medulla Renal cortex Adrenal gland Adipode tissue Spleen Lung -100 -50 0 50 100 150 400 Superoxide formation mediated by NADH dehydrogenase (% vs control) Olanzapine * * #### # * A) Drug + dUQ Drug Heart Liver Jejunum Duodenum Renal medulla Renal cortex Adrenal gland Adipode tissue Spleen Lung -100 -50 0 50 100 150 400 Superoxide formation mediated by NADH dehydrogenase (% vs control) Fluphenazine *** # ## ** # ** ### B) Figure 5. (A) NADH-mediated superoxide formation promoted by olanzapine in different human tissues in the presence and absence of decylubiquinone. (B) NADH-mediated superoxide formation promoted by fluphenazine in different human tissues in the presence and absence of decylubiquinone. Results expressed as percentages of increase versus the basal activity without the drug (mean ± SEM). Shapiro–Wilk test was performed to test normality. For the tissues that present a Gaussian distribution, one-way ANOVA statistical test with α: 0.05 was performed. For tissues without a normal distribution, Kruskal–Wallis statistical test with α: 0.05 was performed: * p < 0.05; ** p < 0.01; *** p < 0.001 drug vs. control; # p < 0.05; ## p < 0.01; ### p < 0.001; and #### p < 0.0001 drug alone vs. drug + dUQ. Figure 5. ( A ) NADH-mediated superoxide formation promoted by olanzapine in different human tissues in the presence and absence of decylubiquinone. ( B ) NADH-mediated superoxide formation promoted by fluphenazine in different human tissues in the presence and absence of decylubiquinone. Results expressed as percentages of increase versus the basal activity without the drug ( mean ±SEM ). Shapiro–Wilk test was performed to test normality. For the tissues that present a Gaussian distribution, one-way ANOVA statistical test with α : 0.05 was performed. For tissues without a normal distribution, Kruskal–Wallis statistical test with α : 0.05 was performed: * p< 0.05; ** p< 0.01; *** p< 0.001 drug vs. control; # p< 0.05; ## p< 0.01; ### p< 0.001; and #### p< 0.0001 drug alone vs. drug + dUQ. Biomedicines 2022,10, 980 16 of 16 62. Dietrich-Muszalska, A.; Malinowska, J.; Olas, B.; Głowacki, R.; Bald, E.; Wachowicz, B.; Rabe-Jabło´nska, J. The oxidative stress may be induced by the elevated homocysteine in schizophrenic patients. Neurochem. Res. 2012,37, 1057–1062. [CrossRef] 63. Lepping, P.; Delieu, J.; Mellor, R.; Williams, J.H.H.; Hudson, P.R.; Hunter-Lavin, C. Antipsychotic medication and oxidative cell stress: A systematic review. J. Clin. Psychiatry 2011,72, 273–285. [CrossRef] [PubMed] 64. Suárez, H.; Andreu, Z.; Mazzeo, C.; Toribio, V.; Pérez-Rivera, A.E.; López-Martín, S.; García-Silva, S.; Hurtado, B.; Morato, E.; Peláez, L.; et al. CD9 inhibition reveals a functional connection of extracellular vesicle secretion with mitophagy in melanoma cells. J. Extracell. Vesicles 2021,10, e12082. [CrossRef] [PubMed] 65. Nolfi-Donegan, D.; Braganza, A.; Shiva, S. Mitochondrial electron transport chain: Oxidative phosphorylation, oxidant production, and methods of measurement. Redox Biol. 2020,37, 101674. [CrossRef] [PubMed] 66. Paradies, G.; Paradies, V.; Ruggiero, F.M.; Petrosillo, G. Oxidative stress, cardiolipin and mitochondrial dysfunction in nonalcoholic fatty liver disease. World J. Gastroenterol. 2014,20, 14205–14218. [CrossRef] 67. Simões, I.C.M.; Fontes, A.; Pinton, P.; Zischka, H.; Wieckowski, M.R. Mitochondria in non-alcoholic fatty liver disease. Int. J. Biochem. Cell Biol. 2018,95, 93–99. [CrossRef] 68. Salimi, A.; Razian, M.; Pourahmad, J. Analysis of toxicity effects of buspirone, cetirizine and olanzapine on human blood lymphocytes: In vitro model. Curr. Clin. Pharmacol. 2018,13, 120–127. [CrossRef] 69. Khorassani, F.; Saad, M. Intravenous olanzapine for the management of agitation: Review of the literature. Ann. Pharmacother. 2019,53, 853–859. [CrossRef] 70. Eftekhari, A.; Azarmi, Y.; Parvizpur, A.; Eghbal, M.A. Involvement of oxidative stress and mitochondrial/lysosomal cross-talk in olanzapine cytotoxicity in freshly isolated rat hepatocytes. Xenobiotica 2016,46, 363–378. [CrossRef] 71. Polydoro, M.; Schröder, N.; Lima, M.N.M.; Caldana, F.; Laranja, D.C.; Bromberg, E.; Roesler, R.; Quevedo, J.; Moreira, J.C.F.; Dal-Pizzol, F. Haloperidoland clozapine-induced oxidative stress in the rat brain. Pharmacol. Biochem. Behav. 2004,78, 751–756. [CrossRef] 72. Reinke, A.; Martins, M.R.; Lima, M.S.; Moreira, J.C.; Dal-Pizzol, F.; Quevedo, J. Haloperidol and clozapine, but not olanzapine, induces oxidative stress in rat brain. Neurosci. Lett. 2004,372, 157–160. [CrossRef] 73. Martins, M.R.; Petronilho, F.C.; Gomes, K.M.; Dal-Pizzol, F.; Streck, E.L.; Quevedo, J. Antipsychotic-induced oxidative stress in rat brain. Neurotox. Res. 2008,13, 63–69. [CrossRef] [PubMed] 74. Rodríguez Carranza, R. Pentazocina. Available online: https://www.vademecum.es/principios-activos-pentazocina-n02ad01 (accessed on 26 December 2021). 75. Heiss, K.; Vanella, L.; Murabito, P.; Prezzavento, O.; Marrazzo, A.; Castruccio Castracani, C.; Barbagallo, I.; Zappalà, A.; Arena, E.; Astuto, M.; et al. (+)-Pentazocine reduces oxidative stress and apoptosis in microglia following hypoxia/reoxygenation injury. Neurosci. Lett. 2016,626, 142–148. [CrossRef] [PubMed] 76. Natsvlishvili, N.; Goguadze, N.; Zhuravliova, E.; Mikeladze, D. Sigma-1 receptor directly interacts with Rac1-GTPase in the brain mitochondria. BMC Biochem. 2015,16, 1–7. [CrossRef] [PubMed] 77. Kurogi, K.; Chen, M.; Lee, Y.; Shi, B.; Yan, T.; Liu, M.-Y.; Sakakibara, Y.; Suiko, M.; Liu, M.-C. Sulfation of buprenorphine, pentazocine, and naloxone by human cytosolic sulfotransferases. Drug Metab. Lett. 2012,6, 109–115. [CrossRef] 78. Appleyard, T.N. Effects of pentazocine on cardiac function. Proc. Roy. Soc. Med. 1975,68, 770. [CrossRef] 79. Lal, S.; Savidge, R.S.; Chhabra, G.P. Cardiovascular and respiratory effects of morphine and pentazocine in patients with myocardial infarction. Lancet 1969,293, 379–381. [CrossRef] 80. Hargreaves, I.P.; Al Shahrani, M.; Wainwright, L.; Heales, S.J.R. Drug-induced mitochondrial toxicity. Drug Saf. 2016 ,39, 661–674. [CrossRef] 81. Balijepalli, S.; Kenchappa, R.S.; Boyd, M.R.; Ravindranath, V. Protein thiol oxidation by haloperidol results in inhibition of mitochondrial complex I in brain regions: Comparison with atypical antipsychotics. Neurochem. Int. 2001 ,38, 425–435. [CrossRef] 82. Jackson-Lewis, V.; Przedborski, S. Neuroleptic medications inhibit complex I of the electron transport chain. Ann. Neurol. 1994 , 35, 244–245. [CrossRef] 83. Xu, F.; Xia, Y.; Feng, Z.; Lin, W.; Xue, Q.; Jiang, J.; Yu, X.; Peng, C.; Luo, M.; Yang, Y.; et al. Repositioning antipsychotic fluphenazine hydrochloride for treating triple negative breast cancer with brain metastases and lung metastases. Am. J. Cancer Res. 2019 ,9, 459. 84. Marsh, W. Nefazodone. Available online: https://pubmed.ncbi.nlm.nih.gov/31643507 (accessed on 17 March 2022). 85. Sedky, K.; Nazir, R.; Joshi, A.; Kaur, G.; Lippmann, S. Which psychotropic medications induce hepatotoxicity? Gen. Hosp. Psychiatry 2012,34, 53–61. [CrossRef] [PubMed] 86. DeSanty, K.P.; Amabile, C.M. Antidepressant-induced liver injury. Ann. Pharmacother. 2007 ,41, 1201–1211. [CrossRef] [PubMed] 87. Robinson, D.S.; Roberts, D.L.; Smith, J.M.; Stringfellow, J.C.; Kaplita, S.B.; Seminara, J.A.; Marcus, R.N. The safety profile of nefazodone. J. Clin. Psychiatry 1996,57 (Suppl. 2), 31–38. [PubMed]