Carbamazepine induces mitotic arrest in mammalian Vero cells
J.M. Perez Martın, P. Fernandez Freire, V. Labrador, M.J. Hazen ∗Departamento de Biologıa, Facultad de Ciencias, Universidad Autonoma de Madrid, Cantoblanco, 28049 Madrid, Spain
Received 11 April 2007; received in revised form 24 July 2007; accepted 26 July 2007Available online 6 August 2007
bstract
We reported recently that the anticonvulsant drug carbamazepine, at supratherapeutic concentrations, exerts antiproliferativeffects in mammalian Vero cells, but the underlying mechanism has not been elucidated. This motivates us to examine rigorouslyhether growth arrest was associated with structural changes in cellular organization during mitosis. In the present work, we
ound that exposure of the cells to carbamazepine led to an increase in mitotic index, mainly due to the sustained block at theetaphase/anaphase boundary, with the consequent inhibition of cell proliferation. Indirect immunofluorescence, using antibodies
irected against spindle apparatus proteins, revealed that mitotic arrest was associated with formation of monopolar spindles, causedy impairment of centrosome separation. The final consequence of the spindle defects induced by carbamazepine, depended on theuration of cell cycle arrest. Following the time course of accumulation of metaphase and apoptotic cells during carbamazepinereatments, we observed a causative relationship between mitotic arrest and induction of cell death. Conversely, cells released fromhe block of metaphase by removal of the drug, continued to progress through mitosis and resume normal proliferation. Our results
how that carbamazepine shares a common antiproliferative mechanism with spindle-targeted drugs and contribute to a betternderstanding of the cytostatic activity previously described in Vero cells. Additional studies are in progress to extend these initialndings that define a novel mode of action of carbamazepine in cultured mammalian cells.2007 Elsevier B.V. All rights reserved.
ndle; Ap
[eo
cgai
eywords: Carbamazepine; Vero cells; Mitotic arrest; Monopolar spi
. Introduction
Carbamazepine (5H-dibenzo [b,f] azepine-5-car-oxamide, CBZ) is a tricyclic lipophilic compound,idely used for the treatment of epileptic seizures,
rigeminal neuralgia and psychiatric disorders [1–3].he therapeutic effects of this drug are not completelylucidated, but current experimental evidences suggest
hat carbamazepine acts primarily by inhibiting voltage-ependent sodium channels [4,5]. Although reportednfrequently, serious side-effects which involve the skin
6,7], liver [8,9], brain [10], kidney [11] and haematopoi-tic system [12,13], have been observed during the usef carbamazepine.
In vitro studies have suggested that CBZ displaysytotoxic activity against cell lines of neuronal andlial origin [14–16], human lymphocyte cultures [17]nd human erythrocyte membranes [12,18]. Moreover,n early studies we found that concentrations of CBZbove the therapeutic range (4–10 mg/l), elicited sig-ificant antiproliferative effects in monkey kidney Veroells [19]. Our results were in agreement with previ-
us observations in both murine lymphoid cells [20] anduman primary cell cultures [17,21,22], but the mech-nisms underlying the mode of action of this drug onammalian cell growth are poorly understood.
The experiments described in this report weredesigned to further study the cytostatic activity of CBZin mammalian Vero cells. The relationship between cellcycle progression and inhibition of cell proliferation wasexamined by determining the mitotic index at a range ofCBZ concentrations. Indirect immunofluorescent stain-ing with antibodies against mitotic apparatus proteins,was performed in order to evaluate the occurrence ofpossible abnormalities in spindle organization.
Carbamazepine caused reversible mitotic arrest, bypreventing progression from metaphase to anaphase,with the formation of monopolar spindles. These obser-vations strongly suggest that the antiproliferative activityof this anticonvulsive drug in cultured mammalian cells,arise by a mechanism involving aberrant mitotic spindleassembly and function. In addition, we found that Verocells treated with carbamazepine, similar to those treatedwith most anti-spindle agents, eventually trigger signal-ing pathways that lead to apoptosis following metaphaseblockage. This study defines a novel mode of action ofcarbamazepine and provides additional data that maycontribute to a better understanding of its activity incultured mammalian cells.
2. Material and methods
2.1. Cell culture and CBZ treatments
Vero cells were routinely maintained at 37 ◦C in a 5% CO2
humidified atmosphere, using Dulbecco’s modified Eagle’smedium (DMEM), supplemented with 5% fetal calf serum,100 U/ml penicillin, 100 mg/ml streptomycin and 2 mM l-glutamine (Bio-Whittaker, Belgium).
Exponentially growing Vero cells were seeded at105 cells/ml in different cell culture surfaces (Falcon, BectonDickinson, USA), depending on the experimental proce-dure. After an overnight incubation, the culture medium wasreplaced by new medium containing concentrations of carba-mazepine (Cas number 298-46-4), ranging from 10 to 500 �M(2.4–118.1 �g/ml). Stock solutions of CBZ (Sigma, USA) wereprepared in ethanol and maintained in darkness at room tem-perature. The exposure solutions were prepared before usein DMEM, and sterilized by filtration through a 0.22 �mMillipore® filter. Maximum ethanol concentration in mediumwas 1% including the control groups.
At given times during the experimental period (4, 9, 18, and24 h), both treated cells and untreated control cells were gentlywashed with phosphate-buffered saline (PBS) and processedaccording the different experimental analyses.
2.2. Cell proliferation assessment
To evaluate the effect of CBZ in Vero cells proliferationand/or detachment, two independent assays were carried out.
icmba
esearch 637 (2008) 124–133 125
otal protein content (TPC) was estimated in cells grownnto 24-well culture plates treated for 24 h with CBZ, by the
ethod of Bradford [23], using Coomassie® Brilliant Blue G-50 reagent (Bio-Rad, Germany) and bovine serum albuminSigma) as standard. Absorbance was measured at 595 nm in apectrafluor microplate reader (Tecan, Austria).
The determination of mitotic index was performed in cellsrowing on glass coverslips into 6-well culture plates. At var-ous time points during CBZ exposure, cells were fixed withold methanol and stained with 5 �g/ml Hoechst 33258 (Riedele Haen, Germany). Three thousand cells were counted at eachime point and mitotic index (MI) was calculated as the ratioetween the number of cells in mitosis and the total number ofells.
For analysis of cell detachment, floating cells after CBZreatments were collected by centrifugation at 1200 rpm for0 min. The pellets were resuspended in 100 �l cold methanolnd 10 �l (1 mg/ml) Hoechst 33258 (Riedel-de Haen, Ger-any) and kept in the dark at −20 ◦C until used. Both mitotic
nd apoptotic cells were scored, based on Hoechst morphology,n at least 1000 nuclei per sample.
.3. Reversibility of CBZ effects
To test whether the effect of CBZ on cell proliferationas reversible, metaphase-arrested cells after a treatment with00 �M CBZ for 9 h, were harvested and reseeded (5 × 104
ells/well) into 24-well culture plates to evaluate total pro-ein content, or on coverslips for immunofluorescence analysis.fter an additional incubation for 24–72 h in fresh cultureedium, cells were processed according the two experimental
rotocols.
.4. Immunofluorescence analysis
To evaluate mitotic spindle and chromosome organization,ero cells grown on glass coverslips were treated with CBZnd fixed with cold methanol for 5 min. After washing inhosphate-buffered saline (PBS), the cells were briefly per-eabilized with 0.5% (v/v) Triton X-100 in PBS at room
emperature, and processed for indirect immunofluorescence.he primary antibodies used were: mouse monoclonal anti-tubulin (1:100, Sigma) to label microtubules, rabbit ***poli-lonal anti CENP-E (1:600, gift from Dr. Suja, Universidadutonoma de Madrid) as centromere marker and mouse mon-clonal anti �-tubulin (1:50, Sigma), to detect the centrosomes.overslips were incubated with primary antibodies for 1 h
n a humidified chamber at 37 ◦C. In double-labeling experi-ents (�-tubulin/CENP-E), primary antibodies were incubated
imultaneously.Following extensive washing in PBS, samples were
◦
ncubated for 45 min at 37 C with either a secondary TRITC-onjugated anti-mouse IgG antibody (1:75, Sigma) or aixture of secondary FITC-conjugated anti-mouse IgG anti-
ody (1:100, Sigma) and Texas Red-conjugated anti-rabbit IgGntibody (1:200, Jackson ImmunoResearch, UK).
1 ation Research 637 (2008) 124–133
si5wt
2
Oe(tOpSwul
2
fwfmvntgBs
3
3c
T2coctv
m1iec
Fig. 1. Growth inhibitory effect of CBZ on Vero cells after a 24 h cul-tdd
ntoiSfwwTisa
a5pmrenr2tiwama
mwi
26 J.M. Perez Martın et al. / Mut
The coverslips were subsequently rinsed in PBS, counter-tained for 3 min with 5 �g/ml Hoechst 33258 and mountedn ProLong Gold antifade reagent (Invitrogen, UK). A total of00 randomly chosen metaphases were scored per slide, whichere divided into normal and abnormal figures depending on
he spindle organization.
.5. Microscopy
Microscopic observations were carried out using anlympus BX-61 epifluorescence microscope (Tokyo, Japan),
quipped with an HBO 100 W mercury lamp and ultravioletUV, 365 nm), blue (450–490 nm) and green (546 nm) excita-ion filters. The images were acquired with a CCD cameralympus DP-70 and processed using the software Olym-us DP controller 1.1.1.65 and Adobe Photoshop 8.0 (Adobeystems, Inc.). The immunofluorescence images of �-tubulinere deconvoluted with Huygens-2 software (Scientific Vol-me Imaging, The Netherlands) using the classic maximumikelihood estimation.
.6. Statistical analysis
The experiments were performed at least three timesor each dose group. Dose–response curves were generatedith individual data points, expressed as percentage of that
ound in control cultures, and presented as the arithmeticean ± standard deviation using Microsoft® Excel 2002. IC50
alues were obtained with the program Sigma Plot 9.0 usingonlinear regression. The data were analyzed statistically usinghe SPSS software 11.5 and significant differences betweenroups were determined by analysis of variance (ANOVA) withonferroni method as the post hoc test. The level of statistical
ignificance employed in all cases was p ≤ 0.05.
. Results
.1. Cell growth inhibition and M-phase arrestaused by CBZ
The effect on cell growth was determined by thePC assay after continuous exposure to the drug for4 h. CBZ significantly reduced cell proliferation atoncentrations ≥100 �M, with half-maximal inhibitionccurring at 406.2 �M (Fig. 1). The relationship betweenell cycle progression and inhibition of cell prolifera-ion was examined by determining the mitotic index atarious time points during drug exposure.
The data summarized in Table 1 revealed that treat-ent of Vero cells with CBZ at doses higher than
00 �M, led to an increase of MI reaching the max-mal values at 9 h. Longer incubation periods did notnhance mitotic activity but rather resulted in a signifi-ant decrease in the MI at all doses tested. The relative
tpoa
ure period. The values represent mean ± S.D. from three independenteterminations of total protein content. Asterisks indicate statisticallyifferent values in treated cells compared to control cells.
umbers of cells in the phases of mitosis indicated thathe rise in mitotic index was due to an accumulationf cells in prometaphase/metaphase and a correspond-ng decline in the number of cells undergoing anaphase.eventy percent of the mitotic cells were in metaphaseollowing a treatment for 9 h with 250 �M and the blockas maximum (95%) at 500 �M CBZ, as comparedith the percentage of control cells in metaphase stage.hese results indicate that CBZ at high concentrations
nterferes with cell cycle progression, arresting cell divi-ion specifically at the transition from metaphase tonaphase.
Since a significant number of detached cells waslready observed after 4 h in cultures treated with00 �M CBZ, this concentration was chosen for mor-hological characterization of cells by fluorescenticroscopy. Floating cells arrested in metaphase were
eadily distinguishable from apoptotic cells by differ-nces in chromatin compaction. As shown in Fig. 2, theumber of cells in metaphase increased progressively toeach maximum values at 9 h and then declined within4 h. Paralleling these results, we found a rise in the frac-ion of cells showing characteristic features of apoptosis,.e., chromatin condensation and nuclear fragmentation,hich was especially evident at the time of peak mitotic
rrest and increased to the highest level at 24 h after treat-ent. The proportion of apoptotic cells was typically
round 1–2%, in the control cell population.To assess the reversibility of the mitotic block at the
etaphase/anaphase transition, Vero cells treated for 9 hith 500 �M CBZ were recovered and allowed to grow
n fresh medium. Most cells (83.5 ± 2.1%) were adhered
o the glass surface 24 h after plating and resume normalroliferation (Fig. 3A), a small but significant proportionf cells (4.0 ± 0.7%) displayed multinucleation (Fig. 3B)nd the remaining cells were apoptotic.
J.M. Perez Martın et al. / Mutation Research 637 (2008) 124–133 127
Table 1Mitotic and phase indices of Vero cells untreated (control) and treated with increasing CBZ concentrations for indicated time periods
Time (h) Concentration (�M) Mitotic index (%) Phase index (%)
The mean values which differ significantly from controls are marked b
3.2. Effects of CBZ on mitotic spindle andchromosome congression
The spindle organization and chromosome congres-sion were concurrently examined by immunofluorescenttechniques and differential DNA staining, using attachedcells exposed to 250 and 500 �M CBZ for 9 h.These experimental conditions allowed us to discernaccurately changes in both number of chromosomes con-
gressed to the metaphase plate and number of spindlepoles.
Mitotic cells arrested in prometaphase/metaphase,displayed altered spindle organization and chromo-
cooo
Fig. 2. (A) Time course of metaphase arrest and apoptosis in floating Vero celmean ± S.D. from three independent experiments. (B) Fluorescence micrograpanel) and apoptotic cells (bottom panel), after 500 �M CBZ treatment for 24
3 ± 38.35 56.60 ± 36.32 11.48 ± 17.65*
isks.
ome congression defects as illustrated in Fig. 4. Mostontrol cells in prometaphase showed well-organizedipolar spindles and a small CENP-E signal at kineto-hores, indicating that chromosomes are being alignedn the equatorial plate (Fig. 4A–D). In CBZ-treated cellsFig. 4E–L), abnormal spindles were either bipolar, withore than two uncongressed chromosomes remaining at
he spindle poles, or monopolar, with a microtubule arrayurrounded by a ring of chromosomes. In both cases, a
rescent and brighter CENP-E signal, indicative of a lackf attachment or insufficient tension at kinetochores, wasbserved. Although CBZ exerted considerable effectsn spindle organization, these same concentrations did
ls, during continuous exposure to 500 �M CBZ. The values representphs showing the nuclear morphology of metaphase-arrested cells (toph. Bar, 5 �m.
128 J.M. Perez Martın et al. / Mutation Research 637 (2008) 124–133
Fig. 3. (A) Recovery of cell growth after removal of the drug. The values represent mean ± S.D. from three independent determinations of total proteincontent. (B) Morphology of culture of Vero cells released from CBZ arrest and examined 48 h after removal of the drug, by immunofluorescenceanalysis. Arrows indicate giant multinucleated cells. Bar, 20 �m.
Fig. 4. Representative images showing the effects of CBZ on spindle organization and chromosome congression after 500 �M CBZ treatment for9 h. (A–D) Control cell spindle with a well-defined prometaphase plate of chromosomes. (E–H) Bipolar spindle with weakly attached chromosomesnear the spindle poles. (I–L) Monoastral spindle surrounded by a ring of chromosomes. Scale bars indicate 5 �m.
J.M. Perez Martın et al. / Mutation R
tTwd
4
caTpdtwmw
opmvetmtcdmr
Fig. 5. Microtubule network in interphase Vero cells after a 24 h treat-ment with 500 �M carbamazepine. Bar, 20 �M.
not significantly affect the appearance of microtubulesin interphase cells (Fig. 5).
Quantitative analysis of mitotic spindles revealed thatthe number of congressed metaphases decreased andthe percentage of monopolar spindles increased in adose-dependent manner (Fig. 6). The frequency of non-congressed metaphases did not differ from control cells,which present occasionally some misaligned chromo-somes.
To further analyze spindle organization, weimmunolocalized �-tubulin, a marker for centrosomes
in control and CBZ-treated Vero cells (Fig. 7). Untreatedcells have two centrosomes, each with two centrioles,at opposite ends of a bipolar spindle. Monoastralcells showed two dots of �-tubulin, indicating that
a
tm
Fig. 6. Frequency of metaphase figures in CBZ-treated Vero cells for 9 h. ThAsterisks indicate statistically different values in treated cells compared to co
esearch 637 (2008) 124–133 129
he replicated centrosomes are only slightly separated.hese observations reveal that CBZ does not interfereith centriole replication, but inhibits the activitiesriving centrosome separation in Vero cells.
. Discussion
Previous studies from our laboratory revealed thatarbamazepine, a major antiepileptic drug, exertedntiproliferative effects on mammalian Vero cells [19].o extend the above in vitro findings, we investigated theossible alterations induced by CBZ in the mitotic spin-le apparatus, a vulnerable target for small moleculeshat interfere with cell cycle progression [24,25]. In thisork, we demonstrate that carbamazepine shares a com-on antiproliferative activity with drugs that interfereith the normal organization of the mitotic spindle.Spindle assembly and function require the regulation
f microtubule dynamics and the activity of a variety ofroteins [26–29]. Suppression of spindle dynamics is theechanism responsible for mitotic block induced by a
ariety of well-characterized antimicrotubule drugs, thatither inhibit the polymerization of tubulin or preventhe disassembly of microtubules [30–34]. Mitotic arrest
ay also be associated with inhibition of both micro-ubule motor proteins and mitotic kinases, which haverucial roles in spindle morphogenesis [35–37]. Despiteifferences in their mechanism of action for inhibitingitosis, the consequence appears to be the same: aber-
ant organization of mitotic spindles, metaphase arrest
nd induction of apoptosis [32,38–43].
An interesting observation in the present study washe finding that Vero cells treated with high carba-
azepine concentrations (≥250 �M), were arrested in a
e values are expressed as mean ± S.D. of three separate experiments.ntrol cells.
130 J.M. Perez Martın et al. / Mutation Research 637 (2008) 124–133
F ero cellst tion of
pmtcttidmaddatwbFmi[om
osvmotm
tsdpb
ctakdaocerHooFh[catp
ig. 7. Fluorescence images of mitotic spindles and centrosomes in Vreatment with 500 �M CBZ for 9 h. The inset shows a high magnifica
rometaphase-like state with a strikingly altered spindleorphology, without disruption of the interphase micro-
ubule network. Accordingly, we hypothesized that thisompound is primarily active in dividing cells and thathe spindle defects could arise from perturbation of pro-eins that drive mitotic machinery, rather than by directnteraction with tubulin and/or microtubules. The spin-le phenotype caused by carbamazepine consisted ofonoastral microtubule arrays that were nucleated fromsingle organizing centre. Although monopolar spin-
les might be generated by failure in either centrosomeuplication or separation [44–47], immunofluorescencenalysis revealed that monoasters resulting from CBZreatments had two foci of �-tubulin labeling, consistentith a mechanism in which the centrosomes duplicateut do not separate properly to form a bipolar spindle.urthermore, the centromeric region of condensed chro-osomes showed increased CENP-E staining intensity,
ndicating unattached or weakly attached kinetochores48], a condition likely responsible for the activationf the spindle checkpoint and consequent arrest at theetaphase–anaphase transition.It is noteworthy that incubation of different types
f mammalian cells with diazepam, a benzodiazepinetructurally and functionally related to CBZ, also pre-ents centrosome separation, leading to the formation of
onopolar spindles [49–52]. Interestingly, these effects
ccur in the same range of concentrations used for CBZreatments in the present study. This coincidence in the
ode of action could indicate a common molecular
ad
a
. (A–D) Metaphase in a control cell. (E–H) Monopolar spindles aftertwo unseparated centrosomes. Scale bars indicate 5 �m.
arget for these two clinically important anticonvul-ive drugs. Nevertheless, the subcellular location of theiazepam binding sites remain unknown, although theossible interference with spindle motor proteins haseen suggested [51].
The monopolar configuration induced by CBZ in Veroells, closely resembles that observed after inactiva-ion of the kinesin-related spindle protein (KSP) Eg5,
member of the BimC family of plus end-directedinesins, required for centrosome separation and spin-le bipolarity [53,54]. KSP inhibitors lead to mitoticrrest with monopolar spindles, without direct effectsn microtubule dynamics [41,55–59]. Based on theseonsiderations, it is tempting to speculate that under ourxperimental conditions, the mitotic kinesin Eg5 mightepresent a potential molecular target of carbamazepine.owever, we could not rule out the possibility that thebserved spindle defects may result from interferencef CBZ with some kinases associated to centrosomes.or instance, cAMP-dependent kinase (PKA) whichas been shown to regulate separation of centrosomes60] and spindle function [61], may be inhibited byarbamazepine treatments [62,63]. In support of thisssumption, it must be stated that inhibition of other cen-rosomal kinases, such as Aurora/Ip11p-related kinaseEg2 and Polo-like kinase1 (Plk1), results in metaphase
rrest with monoastral and other defective mitotic spin-les [64,65].
Following spindle damage cells frequently undergopoptosis, which can occur immediately or after an aber-
ation R
[
[
[
[
[
[
[
[
J.M. Perez Martın et al. / Mut
rant exit from mitosis [34,66,67]. Hence, we evaluatedmicroscopically the kinetics of CBZ-mediated mitoticarrest and apoptosis in Vero cells, in order to establisha possible causative relationship between these two pro-cesses. During carbamazepine treatments, the increase inapoptotic cells was detected in parallel with disappear-ance of the metaphase-blocked cells, thus indicating thatmitotic failure directly triggered the execution of apop-tosis. It is notable however, that upon removal of thedrug, mitotic arrest appears to be lifted and cells pro-gressed normally into the next cell cycle. Remarkably,catastrophic mitosis was found in only a few cells thatdisplayed enlarged shape, micronucleation and eventu-ally delayed apoptosis.
Taken together, these observations suggest that, if thecheckpoint signal disappears, duplicated centrosomesare able to separate and form a bipolar spindle, lead-ing to successful completion of mitosis. Alternatively,prolonged exposure to the drug or sustained mitoticblockage, i.e., longer than 9 h, inevitably resulted inapoptosis. It should be pointed out that carbamazepineconcentrations used in the present study, were sev-eral orders of magnitude above the therapeutic serumlevels, within an effective range of 13–50 �M [21].Therefore, it is very improbable that equivalent cellularresponses relevant to human health effects might occurin vivo.
In conclusion, this report provides unprecedentedevidence about a possible association between antiprolif-erative effects of carbamazepine on cultured mammaliancells and impairment of centrosome separation. Fur-ther studies are focusing on the precise mechanismunderlying the formation of aberrant spindles, the char-acterization of molecular target(s) and in particular, onthe signaling pathways triggered by carbamazepine, thatmay be complex and likely include the participation ofa number of distinct molecules.
Acknowledgements
This work was supported by a grant (CTM2005-02135) from the Spanish Ministry of Education andScience. J.M. Perez Martın is a recipient of a grant fromthe Universidad Autonoma de Madrid.
References
[1] E. Beghi, Carbamazepine, Clinical efficacy and use in other
neurologic disorders, in: R. Levy, R. Mattson, B. Meldrum,E. Perucca (Eds.), Antiepileptic Drugs, Lippincott Williams &Wilkins, Philadelphia, 2002, pp. 273–277.
[2] J. McNamara, Drugs effective in the therapy of the epilepsies,in: J. Hardman, L. Limbird (Eds.), Goodman and Gilman’s The
[
[
esearch 637 (2008) 124–133 131
Pharmacological Basis of Therapeutics, McGraw-Hill, New York,2001.
[3] M. Trimble, Carbamazepine, Clinical efficacy and use in psychi-atic disorders, in: R. Levy, R. Mattson, B. Meldrum, E. Perucca(Eds.), Antiepileptic Drugs, Lippincott Williams & Wilkins,Philadelphia, 2002, pp. 278–284.
[4] A. Ambrosio, P. Soares-da-Silva, C. Carvalho, A. Carvalho,Mechanisms of action of carbamazepine and its derivatives,oxcarbazepine, BIA 2-093, and BIA 2-024, Neurochem. Res. 27(2002) 121–130.
[5] R. Macdonald, Carbamazepine. Mechanism of action, in: R. Levy,R. Mattson, B. Meldrum, E. Perucca (Eds.), Antiepileptic Drugs,Lippincott Williams & Wilkins, Philadelphia, 2002, pp. 227–235.
[6] M. Fischer, H. Hamm, J. Wirbelauer, Severe drug-related skinreaction: toxic epidermal necrolysis caused by carbamazepine,Klin. Padiatr. 216 (2004) 288–293.
[7] M.S. Lin, Y.S. Dai, R.F. Pwu, Y.H. Chen, N.C. Chang, Riskestimates for drugs suspected of being associated with Stevens-Johnson syndrome and toxic epidermal necrolysis: a case-controlstudy, Intern. Med. J. 35 (2005) 188–190.
[8] M. Kalapos, Carbamazepine-provoked hepatotoxicity and pos-sible aetiopathological role of glutathione in the events.Retrospective review of old data and call for new investigation,Adverse Drug React. Toxicol. Rev. 21 (2002) 123–141.
[9] A. Grieco, B. Alfei, P. Di Rocco, L. Miele, G. Biolcati, D. Griso,F. Vecchio, A. Bianco, G. Gasbarrini, Non-alcoholic steatohep-atitis induced by carbamazepine and variegate porphyria, Eur. J.Gastroenterol. Hepatol. 13 (2001) 973–975.
10] J. Gorrel, Management of pediatric epilepsy syndromes, US Phar-macol. (2002) 75–84.
11] J. Hagarty, M. Picton, G. Agarwal, A. Pramanik, P. Kalra,Carbamazepine-induced acute granulomatous interstitial nephri-tis, Clin. Nephrol. 57 (2002) 310–313.
12] S. Konig, J. Knolle, S. Friedewald, W. Koelfen, E. Longin, T.Lenz, D. Hannak, Effects of valproic acid, carbamazepine andphenobarbitone on the fatty acid composition of erythrocyte mem-branes in children, Epilepsia 44 (2003) 708–711.
13] M. Cengiz, A. Yuksel, M. Seven, The effects of carbamazepineand valproic acid on the erythrocyte glutathione, glutathione per-oxidase, superoxide dismutase and serum lipid peroxidation inepileptic children, Pharmacol. Res. 41 (2000) 423–425.
14] X. Gao, R. Margolis, P. Leeds, C. Hough, R. Post, D. Chuang,Carbamazepine induction of apoptosis in cultured cerebellar neu-rons: effects of N-methyl-d-aspartate, aurintricarboxylic acid andcycloheximide, Brain Res. 703 (1995) 63–71.
15] A. Pavone, V. Cardile, An in vitro study of new antiepileptic drugsand astrocytes, Epilepsia 44 (2003) 34–39.
16] I.M. Araujo, A.F. Ambrosio, E.C. Leal, M.J. Verdasca, J.O.Malva, P. Soares-da-Silva, A.P. Carvalho, C.M. Carvalho, Neuro-toxicity induced by antiepileptic drugs in cultured hippocampalneurons: a comparative study between carbamazepine, oxcar-bazepine, and two new putative antiepileptic drugs, BIA 2-024and BIA 2-093, Epilepsia 45 (2004) 1498–1505.
17] W. Awara, M. El-Gohary, S. El-Nabi, W. Fadel, In vivo and in vitroevaluation of the mutagenic potential of carbamazepine: doesmelatonin have anti-mutagenic activity? Toxicology 125 (1998)45–52.
18] M. Suwalsky, S. Mennickent, B. Norris, F. Villena, C. Sotomayor,Effects of the antiepileptic drug carbamazepine on human erythro-cytes, Toxicol. in Vitro 20 (2006) 1363–1369.
19] A. Jos, G. Repetto, J. Rıos, M. Hazen, M. Molero, A. del Peso,M. Salguero, P. Fernandez Freire, J. Perez Martın, Ecotoxico-
1 ation R
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
32 J.M. Perez Martın et al. / Mut
logical evaluation of carbamazepine using six different modelsystems with eighteen endpoints, Toxicol. in Vitro 17 (2003) 525–532.
20] T. Tittle, B. Schaumann, Effects of antiepileptic drugs on growthof murine lymphoid tumor cells in single-cell culture, Epilepsia33 (1992) 729–735.
21] M. Stander, J. Dichgans, M. Weller, Anticonvulsant drugs fail tomodulate chemotherapy-induced cytotoxicity and growth inhibi-tion of human malignant glima cells, J. Neurooncol. 37 (1998)191–198.
22] J. Feldkamp, A. Becker, O. Witte, D. Scharff, W. Scherbaum,Long-term anticonvulsant therapy leads to low bone mineraldensity-evidence for direct drug effects of phenytoin and carba-mazepine on human osteoblast-like cells, Exp. Clin. Endocrinol.Diabetes 108 (2000) 37–43.
23] M. Bradford, A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254.
24] S. Haggarty, T. Mayer, D. Miyamoto, R. Fathi, R. King, T.Mitchinson, S. Schreiber, Dissecting cellular processes usingsmall molecules: identification of colchicine-like, taxol-like andother small molecules that perturb mitosis, Chem. Biol. 7 (2000)275–286.
25] M. Jordan, L. Wilson, Microtubules as a target for anticancerdrugs, Nat. Rev. Cancer 4 (2004) 253–265.
26] T. Mitchison, L. Evans, E. Schulze, M. Kirschner, Sites of micro-tubule assembly and disassembly in the mitotic spindle, Cell 23(1986) 515–527.
27] E. Karsenti, I. Vernos, The mitotic spindle, a self-made machine,Science 294 (2001) 543–547.
28] D. Sharp, G. Rogers, J. Scholey, Microtubule motors in mitosis,Nature 407 (2000) 41–47.
29] T. Wittmann, A. Hyman, A. Desai, The spindle: a dynamic assem-bly of microtubules and motors, Nat. Cell Biol. 3 (2001) E28–E34.
30] J. Correia, S. Lobert, Physicochemical aspects of tubulin-interacting antimitotic drugs, Curr. Pharm. Des. 7 (2001)1213–1228.
31] G. Bacher, T. Beckers, P. Emig, T. Klenner, B. Kutscher, B. Nicker,New small-molecule tubulin inhibitors, Pure Appl. Chem. 73(2001) 1459–1464.
32] M. Jordan, Mechanism of action of antitumor drugs that interactwith microtubules and tubulin, Curr. Med. Chem. Anti-CancerAgents 2 (2002) 1–7.
33] J. Zhou, J. Yao, H. Joshi, Attachment and tension in the spindleassembly checkpoint, J. Cell Sci. 115 (2002) 3547–3555.
34] J. Chen, S. Horwitz, Differential mitotic responses tomicrotubule-stabilizing and -destabilizing drugs, Cancer Res. 62(2002) 1935–1938.
35] V. Mountain, D. Compton, Dissecting the role of molecularmotors in the mitotic spindle, Anat. Rec. 261 (2000) 14–24.
36] P. Andrews, E. Knatko, W. Moore, J. Swedlow, Mitotic mechan-ics: the auroras come into view, Curr. Opin. Cell Biol. 15 (2003)672–683.
37] F. Barr, H. Sillje, E. Nigg, Polo-like kinases and the orchestrationof cell division, Nat. Rev. Mol. Cell Biol. 5 (2004) 429–440.
38] K. Bhalla, Microtubule-targeted anticancer agents and apoptosis,Oncogene 22 (2003) 9075–9086.
39] F. Mollinedo, C. Gajate, Microtubules, microtubule-interferingagents and apoptosis, Apoptosis 8 (2003) 413–450.
40] J. Zhou, P. Giannakakou, Targeting microtubules for cancerchemotherapy, Curr. Med. Chem. Anti-Cancer Agents 5 (2005)65–71.
[
esearch 637 (2008) 124–133
41] W. Tao, V. South, Y. Zhang, J. Davide, L. Farrel, N. Kohl, L. Sepp-Lorenzino, R. Lobell, Induction of apoptosis by an inhibitor ofthe mitotic kinesin KSP requires both activation of the spindleassembly checkpoint and mitotic slippage, Cancer Cell 8 (2005)49–59.
42] F. Gizatullin, Y. Yao, V. Kung, M. Harding, M. Loda, G.Shapiro, The aurora kinase inhibitor VX-680 induces endoredu-plication and apoptosis preferentially in cells with compromisedp53-dependent postmitotic checkpoint function, Cancer Res. 66(2006) 7668–7677.
43] T. Lansing, R. McConnell, D. Duckett, G. Spehar, V. Knick,D. Hassler, N. Noro, M. Furuta, K. Emmitte, T. Gilmer, T.J.Mook, M. Cheung, In vitro biological activity of novel small-molecule inhibitor of polo-like kinase 1, Mol. Cancer Ther. 6(2007) 450–459.
44] J. Waters, R. Cole, C. Rieder, The force-producing mechanism forcentrosome separation during spindle formation in vertebrates isintrinsic to each aster, J. Cell Biol. 122 (1993) 361–372.
45] M. Heck, A. Pereira, P. Pesavento, Y. Yannoni, A. Spradling,L. Goldstein, The kinesin-like protein KLP61F is essential formitosis in Drosophila, J. Cell Biol. 123 (1993) 665–679.
46] G. Sluder, F. Miller, C. Rieder, Reproductive capacity of seaurchin centrosomes without centrioles, Cell Motil. Cytoskeleton13 (1989) 264–273.
47] M. Winey, L. Goetsch, P. Baum, B. Byers, MSP1 and MSP2:novel yeast genes defining distinct steps of spindle pole bodyduplication, J. Cell Biol. 114 (1991) 745–754.
48] D. Thrower, M. Jordan, L. Wilson, Modulation of CENP-E orga-nization at kinetochores by spindle microtubule attachment, CellMotil. Cytoskeleton 35 (1996) 121–133.
49] L. Andersson, V.-P. Lehto, S. Stenman, R. Badley, I. Virtanen,Diazepam induces mitotic arrest at prometaphase by inhibitingcentriolar separation, Nature 291 (1981) 247–248.
50] A. Lafi, E. Parry, J. Parry, The effects of benzodiazepines uponthe fidelity of mitotic cell division in cultured Chinese hamstercells, Mutat. Res. 189 (1987) 319–332.
51] P. Gassner, I. Adler, Analysis of chemically induced spindle aber-rations in male mouse germ cells: comparison of differential andimmunofluorescent staining procedures, Mutagenesis 10 (1995)243–252.
52] I. Vitale, A. Antoccia, P. Crateri, S. Leone, G. Arancia,C. Tanzarella, Caspase-independent apoptosis is activated bydiazepam-induced mitotic failure in HeLa cells, but not in humanprimary fibroblasts, Apoptosis 10 (2005) 909–920.
53] A. Blangy, H. Lane, P. d’Herin, M. Harper, M. Kress, E. Nigg,Phosphorylation by p34 cdc2 regulates spindle association ofhuman Eg5, a kinesin-related motor essential for bipolar spindleformation in vivo, Cell 83 (1995) 1159–1169.
54] K. Sawin, T. Mitchinson, Mutations in the kinesin-like proteinEg5 disrupting localization to the mitotic spindle, Proc. Natl.Acad. Sci. U.S.A. 92 (1995) 4289–4293.
55] T. Mayer, T. Kapoor, S. Haggarty, R. King, S. Schreiber, T.Mitchison, Small molecule inhibitor of mitotic spindle bipolar-ity identified in a phenotype-based screen, Science 286 (1999)971–974.
56] T. Kapoor, T. Mayer, M. Coughlin, T. Mitchison, Probing spindleassembly mechanisms with monastrol, a small molecule inhibitor
of the mitotic kinesin, Eg5, J. Cell Biol. 150 (2000) 975–988.
57] R. Sakowicz, J. Finer, C. Beraud, A. Crompton, E. Lewis, A.Fritsch, Y. Lee, J. Mak, R. Moody, R. Turincio, J. Chabala, P.Gonzales, S. Roth, S. Weitman, K. Wood, Antitumor activity ofa kinesin inhibitor, Cancer Res. 64 (2004) 3276–3280.
ation R
[
[
[
[
[
[
[
[
[
J.M. Perez Martın et al. / Mut
58] A. Marcus, U. Peters, S. Thomas, S. Garrett, A. Zelnak, T. Kapoor,P. Giannakakou, Mitotic kinesin inhibitors induce mitotic arrestand cell death in taxol-resistant and -sensitive cancer cells, J. Biol.Chem. (2005).
59] N. Sunder-Plassman, V. Sarli, M. Gartner, M. Utz, J. Seiler, S.Huemmer, T. Mayer, T. Surrey, A. Gianni, Synthesis and biologi-cal evaluation of new tetrahydro-�-carbolines as inhibitors of themitotic kinesin Eg5, Bioorg. Med. Chem. 13 (2005) 6094–6111.
60] W. Lutz, W. Lingle, D. McCormick, T. Greenwood, J. Salisbury,Phosphorylation of centrin during the cell cycle and its role incentriole separation preceding centrosome duplication, J. Biol.Chem. 276 (2001) 20774–20780.
61] D. Diviani, L. Langeberg, S. Doxey, J. Scott, Pericentrin anchors
protein kinase A at the centrosome through a newly identifiedRII-binding domain, Curr. Biol. 10 (2000) 417–420.
62] J. Jensen, A. Mork, Altered protein phosphorylation in the ratbrain following chronic lithium and carbamazepine treatments,Eur. Neuropsychopharmacol. 7 (1997) 173–179.
[
esearch 637 (2008) 124–133 133
63] G. Chen, K. Hasanat, J. Bechuk, G. Moore, D. Glitz, H. Manji,Regulation of signal transduction pathways and gene expressionby mood stabilizers and antidepressants, Psychosom. Med. 61(1999) 599–617.
64] R. Giet, C. Prigent, The Xenopus laevis aurora/Ip11p-relatedkinase pEg2 participates in the stability of the bipolar mitoticspindle, Exp. Cell Res. 258 (2000) 145–151.
65] I. Sumara, J. Gimenez-Abian, D. Gerlich, T. Hirota, C. Kraft, C.de la Torre, J. Ellenberg, J. Peters, Roles of polo-like kinase 1 inthe assembly of functional mitotic splindles, Curr. Biol. 14 (2004)1712–1722.
66] M. Jordan, K. Wendell, S. Gardiner, W. Derry, H. Copp, L. Wil-son, Mitotic block induced in HeLa cells by low concentration of
paclitaxel (Taxol) results in abnormal mitotic exit and apoptoticcell death, Cancer Res. 56 (1996) 816–825.
67] T. Wang, H. Wang, Y. Soong, Paclitaxel-induced cell death. Wherethe cell cycle and apoptosis come together, Cancer 88 (2000)2619–2628.
