线粒体毒性研究

2A high-throughput dual parameter assay for assessing drug-induced 345Q2

6

[***********][1**********]

[***********][***********]58

drug-induced mitochondrial dysfunction has now been shown to

In addition to post-market drug withdrawals, mitochondrial liabil-ities have also been associated with many drugs carrying a black box label for hepatic or cardiac toxicity (Dykens et al., 2007), high-Corresponding author. Tel.:+[1**********].

lighting again the importance of mitochondrial toxicity when con-[***********][1**********]839

[***********][***********]77787980

2

[***********]91929394J. Hynes et al. /Toxicology in Vitro xxx (2012)

xxx–xxx

in the approaches used for assessing drug safety. The animals used in regulatory toxicology studies are generally young and healthy and are thus unsuitable for identifying compounds that may cause sub-lethal reductions of mitochondrial capacity. The development of suitable in vitro assays is therefore of particular importance in the investigation and prediction of human mitochondrial toxicity. Traditionally, in vitro analysis of oxygen consumption of both mitochondria and whole cells was conducted using a Clark elec-trode (Clark, 1959) but this lacks the throughput required in the application for large compound libraries and IC 50determination. This limitation was addressed through the use of an oxygen-sensi-tive fluorescentprobe which formed the basis for a microtitre plate-based oxygen galactose media and the THLE cytotoxicity assay, cell culture media and supplements were purchased from Invitrogen (Carlsbad,CA) and Lonza (Walkersville,MD), and tissue culture flasks(BDBio-coat), 96-well plates and 384-well plates (BDBiocoat) were pur-chased from VWR (Westchester,PA). CellTiter-Glo Luminescent Cell Viability Assay kits were purchased from Promega (Madison,WI). MitoXpress Ò-Xtra and pH-Xtra were purchased from Luxcel Biosciences (Cork,Ireland). 2.2. Methods

2.2.1. Cell culture

11mM glucose, [***********][**************]95analysis of isolated glutamine and supple-96Using this platform, units/mlpenicillin and 97for members of a in 175cm 2tissue 98donediones, grown in a cell culture 99et al., 2008a,b; 100uses an isolated and galactose-con-101on respiration ). High-glucose med-102the development of medium (DMEM)103sensitivity of glucose and 1mM so-104that impair 5mM N -2-hydroxyethyl-105resultant cellular 10%fetal bovine 106conditions are l g/mlstreptomycin. Gal-107of >3is taken as (Invitrogen11966–025)108chondrial liability mM glutamine (6mM fi-109certain 1mM sodium pyruvate, 110manner (Dykens et in collagen-coated 111the drawback in cell culture incubator at 112toxicity, it is not 113ity occurs or not.

(THLE-2)cells were cul-114In light of this, (Lonza,Walkers-115capable of components of a 116lar acidificationkit (withthe exception 117refineour current (Walkersville,MD), 118made possible by growth factor, 70ng/119based extracellular penicillin and 50l g/ml120allel oxygen cell culture incubator at 121drug displays 122be affected and 123itantly as cells try to 124increased glycolytic 125would therefore cells/wellon collagen-126drial oxygen cells were plated at 127present the bottom 384-well plates. 128of a 200compound were prepared in 129this assay with data indicated finaldrug con-130drial oxygen 0.5%.Cellular ATP con-131assay with the aim Cell Viability 132combination of 24h after drug addi-133

chondrial toxicity.

and 72h after drug 1342. Materials and 1352.1. Materials

Sprague–Dawleyrats (Hynes et al., 2012):136All chemicals of carbon dioxide. 137MO), or Toronto Research Chemicals Inc. (NorthYork, Ontario, The liver was rapidly excised and placed in ice-cold Buffer I 138Canada), and were of highest purity available. HL60cells, HepG2(210mM mannitol, 70mM sucrose, 5mM Hepes, 1mM EGTA, 139cells and THLE cells were purchased from the American Tissue Cul-0.5%BSA, pH 7.). Using a pair of scissors, approximately 6g of liver [***********][***********][***********][***********][***********][***********][***********]197198199

J. Hynes et al. /Toxicology in Vitro xxx (2012)xxx–xxx

[***********][***********]3

[***********][***********]278on low speed. The homogenate was then adjusted to 8vol. with Buffer I and centrifuged at 700g at 4°C for 10min, then filteredthrough two layers of cheesecloth and re-centrifuged for 10min at 14,000g to precipitate the mitochondrial fraction. The superna-tant was discarded and using a glass stirring rod, the mitochondrial pellet was re-suspended in 20ml of Isolation Buffer I and re-centri-fuged at 10,000g for 10min at 4°C. This wash step was repeated using buffer II (210mM mannitol, 70mM sucrose, 10mM MgCl 2, 5mM K 2HPO 4, 10mM MOPS, 1mM EGTA, pH 7.4) and the mito-chondrial pellet was re-suspended in 0.7ml of Buffer II and store on ice until required.

2.5. Measurement of ments were performed in the buffers outlined above. 70l l was added per test well and 7drugs and an untreated control were as-sessed at n =2per plate. Appropriate controls were included as per Fig. 2:HPE 1=Antimycin A 1l M:100%Inhibition effect, HPE Q12=FCCP 2.5l M:100%uncoupling effect, HPE 3=Oxamate 50mM, 100%ECA inhibition. Plates were prepared ahead of time and prior to cell addition were placed on a plate heat block at 30°C to ensure temperature equilibration prior to measurement. A concentrated stock of HL60cells was prepared and 5l l was added to test wells to yield a finalconcentration of 3Â106cells/ml and 1.5Â106cells/mlfor oxygen and pH analysis respectively. Fifty microlitres of mineral oil was then added to wells intended read in a pre-heated 216mitochondria

labels:an oxygen l s/70l s, gate 30l s and a 217Oxygen 100l s/300l s, gate 30l s. 218using a described (Hynes 219scribed (Hynes et transposed into hydrogen 220gen probe was relationships thereby 221concentration of determining IC 50values 222were pipetted into of the fluorescencemea-223per well. For drug into specificallydesigned 224pared in DMSO and conversion of raw data 225concentrations and ion scale as well as sub-2260.5%(v/v).All drug control in row A. This 227mitochondrial impact of tempera-228strate (12.5/12.5229or without 230tion) in respiration of oxygen 231EGTA, 5mM MgCl 2, 232mitochondria stock 233sired finalcells/wellon collagen-234oil was added to cells were plated at 235gen, and the plate plates. HL60cells 236Tecan, Innsbruck, 96-well plates at 237over a period of 20stock solutions 238each well every 1.5wells to give the indi-239tion/emission,a and cells were pre-240100l s. To ensure 241pensing steps were l l volume per well at Compound addi-242tions and a all samples were overlaid 243Park, IL). After 244profilesof the plates were then using the Victor X4fluo-245Magellan (Tecan,as described above. 246CA) to determine 247known relationship 248centration (Hynes et 249were subsequently 250tion profilesover were utilized in this in 100%DMSO as 30mM 2512.6. Dual-parameter assay and THLE 252acidificationassay

consumption assay in for the dual parameter 253Measurement were tested at indi-254tham, MA) DMSO concentration 255measurement 256with low buffering 2574g/LKCl, 0.08g/L258say). The 384-well 259The emission of the pH probe pH-Xtra is modulated by sample 3.1. Development of a high throughputdual-parameter cellular oxygen 260pH whereby decreasing pH causes an increase in probe emission consumption and extracellular acidificationassay for identifying 261(Hynes et al., 2009b ). A 75mM compound stock (150Â) was pre-mitochondrial toxicity

[***********][***********]291292

[***********][***********]305306307

[***********]314315316

[1**********]0

4

[***********][***********]336J. Hynes et al. /Toxicology in Vitro xxx (2012)

xxx–xxx

fluorescentprobes. To achieve this, a high throughput assay capable of parallel oxygen (O2) consumption and extracellular acidification(ECA)analysis was developed.

Table 1outlines the sensitivities of 3cells types to a panel of compounds including uncouplers, inhibitors of oxidative phos-phorylation and modulators of glycolytic fluxand extracellular acidification.The compounds used were FCCP, an uncoupler, oligo-mycin, an inhibitor of the F 1/Fo-ATPase,antimycin, an inhibitor of Complex III, 2-deoxyglucose, an inhibitor of hexokinase, sodium oxamate, an inhibitor of lactate dehydrogenase, and two drugs pre-viously shown to inhibit mitochondrial respiration, flutamideand nefazadone. The cells used were HL60, a non-adherent human pro-myelocytic 3.3. Application of the dual parameter assay and data correlations across two other mitochondrial toxicity assays

In order to establish the value of this new assay, 200drugs were screened and the results were compared to those obtained for the same set of drugs in an oxygen consumption assay using isolated mitochondria (Hynes et al., 2006) and a HepG2glucose/galactoseassay (Marroquin et al., 2007), two assays that have been routinely used by us in the past several years. In addition, the drugs were also tested for cytotoxicity over 72h in THLE cells.

The inter-assay variation of the oxygen consumption assay (Ta-ble 4) and the HepG2glucose/galactoseassay (Table 5) show that [***********][***********]337cell line, and THLE, assays are known to 338The data in Table 1cardiotoxicity or neph-339gen consumption the literature to cause 340the compounds. were taken 341pension cells and the dual parameter assay 342lines, HL60cells assay, respec-343

pounds for effects the initial 8min after mitochondrial assay. 3443.2. Data output and based on the follow-345assay

l M in the THLE assay had an UC 503in 355crease in oxygen impairment. 356extracellular sorted into 9categories 357reached where the 358and the cell dies. over 72h in THLE 359for mitochondrial were positive in all three 360a dose dependant 50Glucose:IC50Galactose 361cellular acidificationisolated mitochondrial 362tory increase in the dual parameter assay 363mitochondrial one drug, droperidol, 364to the compound’sratio >3,was positive in 365To assess the in the isolated mito-366proach, both Z 0at concentrations up to 367tency were be noted, however, 368were typically above mitochondrial protein, 369cellular acidificationvalue of 202nmol/mito-370To further examine of drugs that had an IC 50371ing perturbed cell in the isolated mito-372were performance the dual parameter assay. 373served inter-assay mechanisms of toxicity 374eters show with mitochondrial 375values typically indicated by the isolated 376oxygen assay. Category 4con-377ments. These data 50Galactose ratio

the high-throughput assay but were not

Table 1

Measured IC 50values in l M comparing apparent sensitivity of HepG2, THLE and HL60cells to oxygen consumption.

FCCP a

Oligomycin Antimycin Nefazadone Flutamide Oxamate 2DG [***********][***********][***********][***********][***********][***********][***********]433434

J. Hynes et al. /Toxicology in Vitro xxx (2012)xxx–xxx

5

435

[***********]442HL60cells used in the dual parameter assay. Category 5consisted 2007), the simplest of which uses differential sensitivity in HepG2443of drugs that had an IC 50Glucose:IC50Galactose ratio

[***********]

6

J. Hynes et al. /Toxicology in Vitro xxx (2012)xxx–xxx

Table 3

Inter-assay variation of validation compounds in both oxygen consumption [IC50(l M)]and extracellular acidification[EC50(l M)]measurements. Average values and associated standard deviations are presented calculated from 6independent experiments with each independent value being an average of 2EC 50values generated as outlined in Fig. 1.

Average

StDev %CVOxygen

FCCP

5.0711.11221.9Antimycin 0.0390.01230.7Oligomycin 0.0380.00718.1Rotenone 0.0200.00733.8Flutamide 19.5064.22521.7Nefazadone 12.6542.43019.2Extracellular acidification

FCCP

0.0530.01529.1Antimycin 0.0740.02634.7Oligomycin 0.0770.02734.6473

[***********][***********][***********][***********]

Inter-assay variation of the HepG2glucose/galactoseassay. n =5separate experiments. Compound

Glucose medium Galactose medium Average IC 50(l M)

StDev IC 50(l M) CV (%)Average IC 50(l M) StDev IC 50(l M) CV (%)[***********][***********][***********][***********][***********][***********]534535

J. Hynes et al. /Toxicology in Vitro xxx (2012)xxx–xxx

7

Table 6

The effect of 200drugs in three mitochondrial toxicity assays. NE:no effect at the highest concentration tested. Drug

IC 50Glucose:IC50Galactose ratio in HepG2cells

O 2consumption in isolated mitochondria

O 2consumption and extracellular acidification(ECA)in dual parameter

assay in HL60cells UC 50(nmol/mg

IC 50(nmol/mg

IC 50(l M) for O 2Max%increase mitochondrial protein)

mitochondrial protein) consumption in ECA Category 1:positive in all three mitochondrial toxicity assays. Cytotoxic in THLE cells (IC5033342.5357.632.18600Ketoconazole >3NE

78.74

2.9

222

536

[***********]543544Li et al., 2012; Moreno-Sanchez et al., 1999; Ong et al., 2006; Ru-dysfunction with hepatotoxicity (Pessayre D, 2008). Since the 545ben and Rasmussen, 1981). We are not aware of any published IC 50Glucose:IC50Galactose ratio in HepG2cells was

8

[***********][***********]578J. Hynes et al. /Toxicology in Vitro xxx (2012)

xxx–xxx

Category 4contained seven drugs:diflunisal,bepridil, tamoxi-fen, clomiphene, promethazine, chlorpromazine and imipramine (Table 6). All of them had an IC 50Glucose:IC50Galactose ratio

There were 75drugs which belonged to category 6. These com-pounds had no effect in any of the three mitochondrial toxicity as-says but caused cytotoxicity with an IC 50value

Only one drug, benziodarone, a vasodilator, fell into category 7, having no effect in the HepG2glucose/galactoseassay at concen-trations

582eter assay, these fell into category 8, 583egory 3. These data assay at concen-584weak mitochondrial consumption in isolated 585difficultythat can mitochondrial pro-586on such assays. The causing an increase in 587oxygen dual parameter assay in 588but, under the in THLE cells 589nificanttested (300l M). The 590rate, suggesting that are no published re-591cells was function. Molindo-592lation in lysosomes of schizophrenia, has 593cytotoxicity since and Cole, 1977). Our re-594chlorpromazine, with that of an 595been reported to respiration in intact 596in H9c2cells (597results on the having no effect in any of 598based mitochondrial cytotoxicity in THLE 599compounds can some of these drugs may 600Seven drugs fell longer duration or that 601cose:IC50Galactose in other types 602consumption in the dysfunction. More-603inhibition of such as mtDNA repli-604in the extracellular or opening of the 605parameter assay. would not be identified606ole, risperidone, 607study appears to tested in the three mito-608and pamidronate to HepG2glucose/galact-609say. Ciclopirox is a (categories1and 2) 610infections (Sehgal, The oxygen consumption 611likely that able to identify 4of 612ciclopirox due to the as mitochondrial toxi-613anti-glutamatergic as well as l5other 614trophic lateral (categories3, 4and 615with rare cases of caused mitochondrial 616a bisphosphonate performed with the HL60617has been associated assay was able to 618Mitochondrial mitochondrial toxicants 619et al., 2010) which as 16other drugs that 620on pamidronate. 3, 5, and 8) but it 621remaining four consumption in iso-622azothioprine and 7). Hence, although the 623indicate that these to identify compounds 624(Balijepalli et al., to its ease of use, it fails 6252006; Maurer and Moller, 1997; Menor et al., 2004). Although to identify many compounds (24out of 200in our study) that can 626the published information on azothioprine, risperidone, amioda-potentially cause mitochondrial impairment when the latter mech-627rone and fluvastatinwas from studies performed on isolated mito-anism of toxicity is masked by multi-factorial toxicity. The results [***********][***********][***********][***********][***********][***********][***********][***********]692

J. Hynes et al. /Toxicology in Vitro xxx (2012)xxx–xxx

[***********][***********]7089

[***********][***********][**************]respiration in comparison with any other pair of the three assays or each of the three assays on its own. Although the dual parameter assay and isolated mitochondrial assay were each able to identify 21and 19drugs, respectively, as causing mitochondrial impair-ment, each assay failed to identify some drugs that the other assay was able to identify. Moreover, performing both assays would also reduce the likelihood of identifying false positives that may arise if only one of these assays is performed. The isolated mitochondrial assay could easily be upgraded into a 384well format, which would allow for the same throughput as the dual parameter assay. One consideration that needs to be taken into account when trying to predict whether mitochondrial toxicity may occur in hu-mans based on Dykens, J.A., Jamieson, J.D., Marroquin, L.D., Nadanaciva, S., Xu, J.J., Dunn, M.C.,

Smith, A.R., Will, Y., 2008b. In vitro assessment of mitochondrial dysfunction and cytotoxicity of nefazodone, trazodone, and buspirone. Toxicol. Sci. 103, 335–345.

Dykens, J.A., Marroquin, L.D., Will, Y., 2007. Strategies to reduce late-stage drug

attrition due to mitochondrial toxicity. Exp. Rev. Mol. Diagn. 2, 161–175.

Dykens, J.A., Will, Y., 2007. The significanceof mitochondrial toxicity testing in drug

development. Drug Discov. Today 12, 777–785.

Eto, K., Fukuda, T., Araki, Y., Inoue, B., Ogata, M., 1985. Effect of tricyclic drugs on

mitochondrial membrane. Acta Med. Okayama 39, 289–295.

Evans, M., Rees, A., 2002. Effects of HMG-CoA reductase inhibitors on skeletal

muscle:are all statins the same? Drug Saf. 25, 649–663.

Fromenty, B., Fisch, C., Labbe, G., Degott, C., Deschamps, D., Berson, A., Letteron, P.,

Pessayre, D., 1990. Amiodarone inhibits the mitochondrial beta-oxidation of fatty acids and produces microvesicular steatosis of the liver in mice. J. by molindone. Am. J.

709pharmacological 710the possibility of N., 2007. K201, a multi-711to know the Cmax de pointes and attenuates 555, 54–60.

712average plasma treatment. Curr.

713dose in humans, in 714mitochondrial K.N., Stevens, G.J., Papkovsky,

mitochondrial toxicity using 715toxicity of a drug in Sci. 92, 186–200.

716cannot be K.N., Stevens, G.J., Papkovsky,

717in an in vitro mitochondrial toxicity using Sci. 92, 186–200.

718that the drug will pH and oxygen probes of the

719In summary, we mitochondria and whole cells. 720based assay The use of oxygen sensitive

721and extracellular function. In:Will, Y., 722available drugs in Dysfunction. John Wiley &723drugs already Y., Papkovsky, D.B., 2009b. In

724over, we obtained ph-sensitive probe and 725viously not 390, 21–28.

726Comparison of the analysis of mitochondrial

727assay with 59–72.

developments in the

728consumption assay Opin. Investig. Drugs 11, 729that the 730mitochondrial C.L., Branham, W.S., New,

Underlying mitochondrial 731pounds that caused liver injury in a mouse model 732of the three assays 238, 150–159.

R., Krahenbuhl, S., 2005.

hepatic toxicity. 733References

Brecht, K., Krahenbuhl, S.,

734Amacher, D.E., 2005. mitochondria. Cell. Mol. Life Sci. 735Med. Chem. 12, 736Arena, J.P., Kass, R.S., Mitochondrial dysfunction

737tertiary analogs:Toxicol. Sci. 127, 582–591.

738blocked. Mol. Will, Y., 2007. Circumventing

739Balijepalli, S., Boyd, with galactose increases 740complex I by toxicants. Toxicol. Sci. 97, 539–741567–577.

742Beeson, C.C., Beeson, I by neuroleptics in normal

743respirometric assay toxicity of neuroleptics. Mol. 744404, 75–81.

745Berson, A., Cazanave, S., A.J., 1983. The mechanism of

746Pessayre, D., 2006. by the nonsteroidal anti-74732, 2595–2598.

748mitochondrial O., Olleros, T., Arriaza, E.,

749by albumin. J. L.G., 2004. Azathioprine 750Boelsterli, U.A., Ho, H.K., protein kinases 751of nitroaromatic L -cysteine. J. Pharmacol. Exp. 752Brown, D.A., O’Rourke,B., 753Res. 88, 241–249.

Silveira, L.H., Martinez-Lavin, M.,

754Cammer, W., Moore, C.L., by nonsteroidal 755brain and liver submitochondrial particles, 756Clark, L.C., 1959. 743–752.

7572913386.

oxidative phosphorylation in rat

758Durand, F., Samuel, D., Bernuau, J., Saliba, F., Pariente, E.A., Marion, S., Benhamou,

15–20.

759J.P., Bismuth, H., 1992. Fipexide-induced fulminant hepatitis. Report of three Nadanaciva, S., Dykens, J.A., Bernal, A., Capaldi, R.A., Will, Y., 2007. Mitochondrial

760cases with emergency liver transplantation. J. Hepatol. 15, 144–146.

impairment by PPAR agonists and statins identifiedvia immunocaptured 761Dykens, J.A., 2007. RedOx targets:enzyme systems and drug development

OXPHOS complex activities and respiration. Toxicol. Appl. Pharmacol. 223, 277–762strategies for mitochondrial dysfunction. Comprehensive Medicinal 287.

Nadanaciva, S., Lu, S., Gebhard, D.F., Jessen, B.A., Pennie, W.D., Will, Y., 2011. A high

[***********][***********][***********][***********][***********][***********][***********][***********][***********][***********][**************]

10

[***********][***********][**************]

J. Hynes et al. /Toxicology in Vitro xxx (2012)

xxx–xxx

O’Bryant,C.L., Flaig, T.W., Utz, K.J., 2008. Bicalutamide-associated fulminant

hepatotoxicity. Pharmacotherapy 28, 1071–1075.

Ong, M.M., Wang, A.S., Leow, K.Y., Khoo, Y.M., Boelsterli, U.A., 2006. Nimesulide-induced hepatic mitochondrial injury in heterozygous Sod2(+/À) mice. Free Radic. Biol. Med. 40, 420–429.

Pessayre, D., Berson, A., Fromenty, B., 2008. Drug-induced mitochondrial

dysfunction. In:Dykens, J.A., Will, Y. (Eds.),Drug-Induced Mitochondrial Dysfunction. John Wiley &Sons, Hoboken, NJ, pp. 143–202.

Remy, A.J., Camu, W., Ramos, J., Blanc, P., Larrey, D., 1999. Acute hepatitis after

riluzole administration. J. Hepatol. 30, 527–530.

Rodriguez, R.J., Acosta Jr., D., 1996. Inhibition of mitochondrial function in isolated

rate liver mitochondria by azole antifungals. J. Biochem. Toxicol. 11, 127–131.Ruben, L., Rasmussen, H., 1981. Phenothiazines and related compounds disrupt

mitochondrial energy production by a calmodulin-independent reaction. Biochim.Biophys. Acta 637, 415–422.

Sehgal, V.N., 1976. double-blind study in Sleno, L., Staack, R.F., Varesio, E., Hopfgartner, G., 2007. Investigating the in vitro

metabolism of fipexide:characterization of reactive metabolites using liquid chromatography/massspectrometry. Rapid Commun. Mass Spectrom. 21, 2301–2311.

Sneyd, J.R., 2009. Droperidol:past, present and future. Anaesthesia 64, 1161–1164.Tuquet, C., Dupont, J., Mesneau, A., Roussaux, J., 2000. Effects of tamoxifen on the

electron transport chain of isolated rat liver mitochondria. Cell Biol. Toxicol. 16, 207–219.

Wallace, K.B., 2008. Mitochondrial off targets of drug therapy. Trends Pharmacol.

Sci. 29, 361–366.

Wallace, K.B., Starkov, A., 2000. Mitochondrial targets of drug toxicity. Annu. Rev.

Pharmacol. Toxicol. 40, 353–388.

Will, Y., 2007. Analysis of mitochondrial function using phosphorescent oxygen-sensitive probes. Nat. Protoc. 1, 2563–2572.

Zhou, S., Wallace, K.B., 1999. The effect of peroxisome proliferators on

[***********][***********][**************]