J. ELECTROCARDIOLOGY 14 (3), 1981, 219-224

Mechanism of Chlorpromazine-induced Arrhythmia Arrhythmia and Mitochondrial Dysfunction

BY MASAYASU KITAZAWA, PH.D.,* SATORU SUGIYAMA, M.D.,* TAKAYUKI OZAWA, M.D.,* YUTAKA MIYAZAKI, M.D.t AND KAZUNOBU KOTAKA, M.D.t

SUMMARY In this study, we investigated the mechanism of the arrhythmogenic action of chlor-

promazine (CPZ). Thirty-two anesthetized mongrel dogs were used. In each, the chest was opened and a st imulating electrode was at tached to the apex of the left ventricle and the ventricular multiple response threshold (VMRT) was measured. The carotid artery was cannulated to measure aortic pressure. The dogs were divided into four groups, and the t ime course of YMRT, blood pressure, and heart rate were determined. All groups were placed under observation for 30 min after CPZ infusion. In the control group, only saline (2ml/kg) was infused; CPZ group: CPZ (lmg/kg) was infused 10 rain after saline (2ml/kg) infusion; CoQlo group: Coenzyme Qlo (CoQlo) (5mg/kg) was in- fused 10 rain before CPZ (lmg/kg) infusion; FAD group: Flavin-adenine-dinucleotide (FAD) (2mg/kg) was infused 10 rain before CPZ (lmg/kg) infusion. In each group, myocardial mitochondria were prepared 3{} min after CPZ infusion. The mitochondrial functions, respiratory control index, ADP/{}, State III rate of oxygen consumption, and activities of two segments of the electron-transport chain (NADH --* CoQ --* cyt.v and cyt.c --* eyt.a, a3 ~ 02) were measured separately. Ca ++- binding activity of the mito- chondria was also determined.

CPZ administrat ion decreased VMRT and blood pressure, and caused mitochondrial dysfunction which derived from a disturbance in the first segment of the electron- t ransport chain. Decreased Ca++-binding activity was observed when mitochondrial function was disturbed. CoQlo prevented significantly the decrease in VMRT and the dis turbance of mitochondrial funct ion induced by CPZ, but did not prevent the hypotensive effect of CPZ. FAD prevented not only the decrease in VMRT and the disturbance of mitochondrial function, but also the hypotensive effect of CPZ.

These results suggest that the decrease in VMRT is closely related to mitochondrial dysfunction induced by CPZ. Moreover, it is suggested that the arrhythmogenic effect of CPZ is derived from the decreased mitochondrial Ca ++- binding activity.

Although much research has been done on the electrophysiological mechanism underlying ven- tricular arrhythmias, a complete understanding of this condition has not been achieved. In par- t icular, the re la t ionship be tween electrophy- siological factors and biochemical factors in

From the Faculty of Medicine, University of Nagoya, Tsuruma, Showa-ku, Nagoya 466, Japan. *Department of Biomedical Chemistry tDepartment of Internal Medicine

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked ~'advertisement" in accordance with 18 U.S.C. w 1734 solely to indicate this fact. Reprint requests to: Takayuki Ozawa, Professor, Department of Biomedical Chemistry, Faculty of Medicine, University of Nagoya, Tsuruma, Showa-ku, Nagoya 466, Japan.

developing arrhythmias has not been evaluated. It has been reported 1-5 that psychotropic drugs, such as chlorpromazine (CPZ), have an adverse effect on the cardiovascular system, such as the induction of abnormal changes in ECG and the deve lopment of a r rhy thmias . Alexander and Nino 6 reported that abnormal changes in ECG induced by CPZ administration are associated w i t h a b n o r m a l m o r p h o l o g i c a l c h a n g e s in mitochondria, and we have reported that CPZ i n d u c e s m i t o c h o n d r i a l d y s f u n c t i o n . 7 We studied 7-12 the mechanism of vent r icu la r ar- rhythmia through both electrophysiological and biochemical approaches, and repor ted 12 tha t some type of ventricular ar rhythmia and ven- tricular fibrillation might be derived from hear t mitochondrial dysfunction. In the present study, we inves t iga ted the e lect rophysiological and

219

220 KITAZAWA ET AL

b i o c h e m i c a l m e c h a n i s m of C P Z - i n d u c e d ar - r h y t h m i a s and the protec t ive effects of Coen- zyme Qlo (COQlo) and f lavin-adenine-dinucleo- tide (FAD).

MATERIALS AND METHODS Thirty-two mongrel dogs of both sexes, weighing 8-12 kg, were anesthetized with sodium pentobarbital (50mg/kg) intraperitoneally. Under artificial respira- tion, the chest was opened in the 4th or 5th intercostal space and the heart was suspended in a pericardial cradle. For the measurement of VMRT, a platinum stimulating electrode with 3 mm interpolar distance was attached to the apex of the left ventricle. As we have described previously, v VMRT was measured as a minimum current necessary to induce more than three serial ventricular extrasystoles. Aortic blood pressure was measured through a catheter which was cannu- lated into the carotid artery and connected to an elec- tromanometer (Nihonkoden Ltd. MP-4T). A second catheter was inserted into the femoral vein to infuse the drugs. Lead II ECG was monitored. The dogs were divided into four groups; each group consisted of eight dogs. VMRT, blood pressure, and heart rate were mea- sured every 10 min.

Control group: For the control, only saline (2ml/kg) was infused.

CPZ group: CPZ (lmg/kg) was infused 10 min after saline (2ml/kg) infusion.

CoQlo group: CoQlo (5mg/kg) was infused 10 min be- fore CPZ (lmg/kg) infusion.

FAD group: FAD (2mg/kg) was infused 10 min before CPZ (lmg/kg) infusion.

Each group was placed under observation for 30 min after saline or CPZ infusion. After the last measure- ments , the h e a r t was i so la ted and m y o c a r d i a l mitochondria were prepared as reported previously. 7 Immediately after the preparation of the mitochondria, the respiratory control index (RCI), ADP/O, and the rate of oxygen consumption in State III (St. III 02) were measured using an oxygen electrode mounted in a closed cell. A part of the mitochondria were frozen to determine the activity of the electron-transport chain. A mannitol reaction mixture of 2.8 ml (0.3 M mannitol, 10 mM potassium chloride, 10 mM potassium phos- phate, pH 7.4, 2.5 mM magnesium chloride and 0.25 mM EDTA) was used for the respiration studies. Respi- ration was initiated by sequential addition of 0.1 ml of mitochondria (10 - 15 mg protein/ml), 0.1 ml of succi- nate (0.2 M) and 0.05 ml of ADP (0.01M). The frozen myocardial mitochondria used for the estimation of the electron-transport chain was divided into two seg- ments: the NADH ~ CoQ ~ cyt.c and the cyt.c -~ cyt.a,a3 ~ 02. The electron-transport activity of the first segment was determined 13 as the specific activity of NADH-cyt.c reductase; the reaction mixture con- sisted of 0.06 ml I~HPO4-HC1 buffer (1.0 M, pH 8.0),

0.1 ml of NaN3 (0.1M), 0.06 ml of EDTA (1 mM), 5 tL1 of 1% DOC (pH 8.0), 0.18 ml of 1% ferricytochrome c (Sigma, Type III) and 2.6 ml of distilled water. The reac- tion was initiated by adding 5 t~l of mitochondria sus- pension (10 - 15 mg protein/ml) and 75 ~1 of 10 mM NADH. After 15 sec of incubation, the reaction rate was followed for 1 min by recording the increase in absor- bance of cyt.c at 550 nm. The specific activity of NADH - cyt.c reductase was calculated from the increasing rate of the absorbance.

Electron-transport activity of the second segment was determined TM as the specific activity of cyt.c oxidase. To prepare ferrocytochrome c, 1% ferricyto- chrome c (Sigma, Type III) was reduced completely by dithionite, and excess dithionite was removed by pass- ing the solution through a column of Sephadex G-25 (fine). I~HPO4-HC1 buffer (2.67 ml, 50 mM, pH 7.0) and 30 ~l of 10% Triton-X 100 were added to 100 tL1 of ferrocytochrome c solution. Immediately after the addi- t ion of 0.1 ml m i t o c h o n d r i a suspens ion (50 t~g protein/ml), the reaction rate was followed for 1 min by recording the decrease in absorbance at 550 nm. Specific activity of cty.c oxidase was calculated from the decrease of the absorbance.

In order to assay the Ca++-binding activity of the mitochondria, the reaction medium contained 3.0 ml of Tris-HC1 buffer, pH 7.8, and 0.125 ml of succinate (0.2 M), and 0.1 ml of calcium chloride (10 mM). One ml of mitochondria suspension (10-15 mg protein/ml) was rapidly added to the reaction medium. After 30 sec of incubation, the reaction mixture was centrifuged at 25,000 g for 5 min. Ca§ activity of the mitochondria was calculated from the decrease in Ca + + concentration of the supernatant, which was estimated by an atomic absorption spectrophotometer (Hitachi, type 207).

RESULTS Fig. 1 shows the t ime course of VMRT in each group. In the control group, VMRT was 1.66 _+ 0.05 m A (mean _ S.E.) before saline infusion, and no s ignif icant change was observed du r ing the exper iment . In the CPZ group, VMRT was de- creased s ignif icant ly to 1.27 +_ 0.08 m A and 1.22 _+ 0.05 m A at 10 and 20 min after CPZ infusion, respectively. In the CoQlo group, pr 'emedicat ion wi th CoQlo prevented the decrease in VMRT in- duced by CPZ admin i s t r a t ion wi th values of 1.70 _+ 0.10 m A and 1.67 _+ 0.07 m A being obta ined at 10 and 20 min af ter the infusion, respectively. In the FAD group, p remed ica t ion wi th F A D also p reven ted the decrease in VMRT despite subse- quen t CPZ admin i s t r a t i on . Fig. 2A shows the t ime course of systolic blood pressure in each group. In the control group, no s ignif icant change was observed. In the CPZ group, systolic blood pressure was decreased s ignif icant ly f rom 108.3

J. ELECTROCARDIOLOGY, VOL. 14, NO. 3, 1981

VMRT

Fig. 1 Time course of VMRT in each group. In the CPZ group, VMRT was decreased significantly. In the CoQ~o group, premedication with CoQ1 o pre- vented the decrease in VMRT. In the FAD group, premedication with FAD also prevented significantly the decrease in VMRT.

2.0

1.5

AoP: (rnmHg

1.0

O'

(~3~)" Control ~ 1 ~ �9 CPZ

~: CoQ,0+CPZ ~ : FAD+CPZ (Mean+SE)

�9 : P<0.05 �9 , : P<0.01

i i

--10 6 1'0 20 30 (min)

_+ 2.3 m m H g ( m e a n _+ S.E.) to 87.3 _+ 2.9 m m H g a t 10 m i n a f t e r the infusion. In the CoQlo group, CoQ~o did not p r e v e n t s ign i f ican t ly the decrease in systol ic blood p ressure . In the FAD group, F A D

110

100

90

8O

0

$ $

$ $

* * �9 P<O.01 I , A

- '10 6 10 20 30 (min)

CHLORPROMAZINE- INDUCED ARRHYTHMIA 221

Fig. 2A Time course of systolic blood pressure in each group. In the CPZ group, systolic blood pressure de- creased significantly. In the CoQlo group, CoQlo did not prevent the decrease in systolic blood pressure. In the FAD group, FAD prevented significantly the decrease in systolic blood pressure.

p r e v e n t e d s ign i f ican t ly the decrease in systol ic blood p re s su re w i th a va lue of 100.7 _+ 3.8 m m H g a t 10 m i n a f t e r the infusion. Fig. 2B shows the t ime course of diastol ic blood pressure . In the CPZ group, diastol ic blood p re s su re was decreased sig- n i f i can t ly f rom 86.3 + 2.2 m m H g to 65.7 _+ 3.6 m m H g a t 10 ra in a f t e r the infusion. In the CoQlo

AoPd (mmH~) 9O

80

70

60

5C

' Control ' CPZ

)~ &~ " CoQ,o+CPZ Z~:~-~ �9 FAD+CPZ (Mean+SE)

* * : P<0.01

-'10 6 17 20 30 (min)

Fig. 2B Time course of diastolic blood pressure in each group. The time course of diastolic blood pressure was similar to that of systolic blood pressure in each group.

J. ELECTROCARDIOLOGY, VOL. 14, NO. 3, 1981

222 K ITAZAWA ET AL

TABLE I RCI, ADP/O, AND ST.III 02 OF MITOCHONDRIA

Group RCI ADP/O St.lll 02*

1. Control 4.15 +_ 0.20 1.95 +_ 0.05 308.8 _+ 18.2 2. CPZ 2.80 _+ 0.27* 1.52 +_ 0.16' 167.3 + 37.2* 3. CoQlo + CPZ 3.98 -+ 0.37* 1.83 _+ 0.07' 270.5 + 34.5* 4. FAD + CPZ 3.67 +- 0.31' 1.82 --- 0.11' 299.0 _+ 29.4*

(mean -+ S.D.) *n atoms/rag protein/rain tp < 0.01 (group 2 vs. group 1) *p < 0.01 (group 3 or 4 vs. group 2)

CPZ administration disturbed mitochondrial function. Premedication with CoQ~o or FAD prevented the disturbance of mitochondrial function.

group, CoQ~o did not prevent the decrease in dia- stolic blood pressure. In the FAD group, FAD pre- vented the decrease in diastolic blood pressure with a value of 77.2 _+ 2.4 mmHg. As for heart rate, no significant change was observed among the groups.

Table I shows RCI, ADP/O, and St. III 02. In the CPZ group, all of these values were decreased significantly. CoQlo or FAD cancelled signifi- cantly the decrease in these values. Fig. 3 shows typical traces of mitochondrial respiration in each group. Table II shows activities of the two seg- ments of the mitochondrial e lectron-transport chain. Electron-transport activity of the second segment did not show any significant change. In the CPZ group, the activity of the first segment was decreased significantly to 132.5 +_ 14.4. Ad- ministration of CoQlo or FAD prevented signifi- cantly the decrease in the activity of the first segment of the mitochondrial electron-transport chain. Table III shows Ca + + - binding activity of mitochondria. In the CPZ group, Ca ++- binding activity was decreased significantly. Premedica- tion with CoQ~o or FAD prevented significantly the decrease in Ca ++- binding activity.

TABLE II ACTIVITIES OF TWO SEGMENTS OF MITOCHONDRIAL

ELECTRON-TRANSPORT CHAIN (NADH ~ CoQ,o ~ cyt.c, and cyt.c ~ cyt.a, a3 ~ 02)

Specific activity of

G r o u p NADH-cyt.c reductase* cyt.c oxidase**

1. Control 214.5 +- 27.9 1188.5 ___ 154.6 2. CPZ 132.5-+ 14.4 t 1138.4 _+ 147.9 3. CoQlo + CPZ 197.0 +- 23.5* 1210.0 + 164.8 4. FAD + CPZ 178.8 +- 6.6* 1134.3 _+ 232.1

(mean _+ S.D.) *pH 8.0, 18~ n moles/rag protein/rain **pH 7.0, 18~ n moles/rag protein/min tp < 0.01 (group 2 vs. group 1) :~p < 0.01 (group 3 or 4 vs. group 2)

CPZ reduced the electron-transport activity of NADH --~ CoQlo --> cyt.c, but did not affect that of cyt.c --> cyt.a, a3 --* 02. Premedication with CoQ~o or FAD prevented the decrease of electron-transport activity induced by CPZ.

TABLE III Ca++-BINDING ACTIVITY OF MITOCHONDRIA

DISCUSSION It has been reported 15 that the electrophysiolog- ical mechanism of CPZ-induced arrhythmias re- sults from prolongation of the vulnerable period of ventricles. However, the precise mechanism of CPZ-induced arrhythmias remains unknown. In this study, we demonstrated that administration of CPZ (lmg/kg) decreases significantly VMRT and causes hear t mitochondrial dysfunction. We also revealed that premedication with CoQlo or FAD preven ted s igni f icant ly the decrease of VMRT and mitochondrial dysfunction. These re-

Group Ca++-Binding Activity*

1. Control 39.0 - 4.1 2. CPZ 17.7 ___ 4.5 t 3. CoO.to + CPZ 28.1 +- 6.8* 4. FAD + CPZ 29.1 +- 6.6*

(mean +- S.D.) *n motes/mg protein/30 sec tp < 0.01 (group 2 vs. group I) *p < 0,01 (group 3 or 4 vs. group 2)

CPZ administration suppressed mitochondrial Ca++-binding activity, but premedication with CoQlo or FAD prevented the disturbance of the Ca++-binding activity.

J. ELECTROCARDIOLOGY, VOL. 14, NO. 3, 1981

CHLORPROMAZINE-INDUCED ARRHYTHMIA 223

RCI : 4.12 ADP/o: 1.97 S t a t e ~ O= -Consumption : 315.8 (natoms/mg p r o t e i n / m i r

- - ~ r

' ' i - " ' / 1 " ~ ' '

C o n t r o l

RCI : 3.19 I ADP/o: 1.59 : i

: 2 "

[ / " t sue. !

! J z t i .... _/_2.1 .... t"

cpz

Fig. 3 Typical traces of mitochondrial respiration in each group. CPZ ad- ministration disturbed mitochondrial function. Premedication with CoQlo or FAD prevented the disturbance of mito- chondrial function.

RCI : 3.92 ADP/O:l.82 .... I ~ ; State II O=-Consumption : 293. ~ i (natoms/mg protein/min) ! fT~ lv l ' i t .

" . . . . ? i i -

CoQ, o-t-CPZ

sults suggest that there exists a close relationship between the development of CPZ-induced ar- rhythmias and heart mitochondrial dysfunction.

It has also been reported 1647 tha t mitochon- dria act as an intracellular Ca ++ pool to bind in- tracellular Ca ++. Ca§ activity of mito- ch o nd r i a ~s is k n o w n to be decreased when mitochondrial function is disturbed. In this study, we demonstrated tha t mitochondrial Ca + +-bind- ing activity was decreased when mitochondrial function was disturbed by CPZ administration. We also observed that when mitochondrial func- tion was protected by premedication of CoQ~o or FAD, the decrease of mitochondrial Ca++-bind- ing activity was prevented. These results suggest how arrhythmias are developed by mitochondrial dysfunction caused by CPZ. Inward Ca ++ current is known to contribute to action potentials and the influxed Ca ++ is taken up partially by the mitochondria. Therefore, the decreased Ca ++- binding activity of mitochondria caused by their dysfunction might result in relatively high levels

i ) State ~ Oz-Consumption : 289.0 _ _ J . ~ _ (natoms/mg protein/rain) ~,a-]~ I~.

I__- I . . . . . .

FAD- -CPZ

of free Ca ++ in the myocardium during the relax- ation phase. This might result in the cardiac cells responding to relatively high action potentials which in turn result in arrhythmias.

It has been reported that CPZ-induced mito- chondrial dysfunction is based on the deformation of mitochondrial membrane. ~9 Our results sug- gest tha t disturbance of the first segment (NADH --* CoQ --~ cyt.c) of the mitochondrial electron- transport chain, especially the CoQ site, is re- sponsible for mitochondrial dysfunction induced by CPZ. As described in Results, CoQlo did not show any protective action against the hypoten- sive effect of CPZ. Therefore, the hypotensive ef- fect of CPZ is suggested to have no relationship to mitochondrial dysfunction but to have a direct ac- tion on vascular smooth muscle as Elias and Boyer 2~ reported.

In conclusion, we would like to stress in this paper that CoQ1o and FAD can prevent effectively CPZ-induced arrhythmias and mitochondrial dys- function, and tha t the arrhythmogenic effect of

d. ELECTROCARDIOLOGY, VOL. 14, NO. 3, 1981

224 KITAZAWA ET AL

CPZ c a n be e x p l a i n e d by t h e d i s t u r b a n c e of m i t o c h o n d r i a l Ca + + - b i n d i n g ac t iv i ty . D i s t u r - b a n c e of t h e f i r s t s e g m e n t o f t h e e l e c t r o n - t r a n s p o r t cha in by CPZ is cons idered to cause ra i tochondr ia l dysfunct ion. Therefore , i t is con- s idered t h a t CoQlo shows specific a n t a g o n i s m to t h e e f fec t of CPZ on m i t o c h o n d r i a l f u n c t i o n , w h e r e a s FAD competes wi th CPZ both on mito- chondr ia l func t ion and on h y p o t e n s i v e effect be- cause of t he i r m u t u a l s t r u c t u r a l s imi la r i ty .

R E F E R E N C E S

1. HOLLISTER, L E AND KOSEK, J C: Sudden death dur- ing t r ea tmen t with phenothiazine derivatives. JAMA 192:1035, 1965

2. GILES, T D AND MODLIN, R K: Death associated with ventricular arrhythmia and thioridazine hy- drochloride. JAMA 205:10g, 1968

3. SACKS, M H, BONFORTE, R J, LASSER, R P AND DI- MICH, I: Cardiovascular complications of imipra- mine intoxication. JAMA 205:588, 1968

4. BURDA, C D: Electrocardiographic abnormalities induced by thioridazine. Am Heart J 76:153, 1968

5. FOWLER, N O, MCCALL, D, CHOU, T, HOLMES, J C AND HANENSON, R B: Electrocardiographic changes and cardiac a r rhy thmias in pat ients receiving psychotropic drugs. Am J Cardiol 37:223, 1976

6. ALEXANDER, C S AND NINO, A: Cardiovascular complications ia young patients taking psycho- tropic drugs. Am Heart J 78:757, 1969

7. SUGIYAMA, S AND OZAWA, T: Protection of chlor- promazine- induced a r r h y t h m i a by f lavin-ade- nine-dinucleotide in canine heart . Jpn Hear t J 20:657, 1979

8. KATO, T, SUZUKI, S, KAMBE, T, SAKAMOTO, N, SUGIYAMI, S AND OZAWA, T: Arrhythmogenic effect of free fatty acids in relation to heart mitochondrial function. J Appl Biochem 1:139, 1979

9. SUZUKI, S, KATO, T, KAMBE, T, SAKAMOTO, N, SUGIYAMA, S AND OZAWA, T: An exper imenta l s tudy of re lease a r rhy thmia : Occlusion t ime- dependent changes in ven t r icu la r f ibr i l la t ion threshold. Am Heart J 98:727, 1979

10. SUGIYAMA, S, NORIMATSU, I AND OZAWA, T: Possible roles of prostaglandins and calcium in ventricular vulnerability during coronary reperfusion. J Appl Biochem 1:402, 1979

11. SUGIYAMA, S, OZAWA, T, SUZUKI, S AND KATO, T: Effects of verapamil and propranolol on venticular vulnerability after coronary reperfusion. J Elec- trocardio! 13:49, 1980

12. SUGIYAMA, S, OZAWA, T, KATO, T AND SUZUKI, S: Recovery time course of ventricular vulnerability after coronary reperfusion in relation to mitochon- drial function in ischemic myocardium. Am Heart J 100:829, 1980

13. HATEFI, Y AND RIESKE, J S: The preparation and properties of DPNH-cytochrome c reductase (Com- plex I-III of the respiratory chain). In Methods in Enzymology, Vol. 10, R W ESTABROOK AND M E PULLMAN, eds. Academic Press, New York, 1967, pp 225-231

14. WHARTON, D C AND TZAGOLOFF, A: Cytochrome oxidase from beef heart mitochondria. In Methods in Enzymology, Vol. 10, R W ESTABROOK AND M E PULLMAN, eds. Academic Press, New York, 1967, pp 245-250

15. ARITA, M, NAGAMOTO, Y AND SAIKAWA, T: Automa- ticity and time-dependent conduction disturbance produced in canine ventricular myocardium. New aspects for initiation of ventricular arrhythmias. Jpn Circ J 40:1409, 1976

16. RASMUSSEN, H: Ion as ~Second Messenger.' In Cell Membrane, G WEISSMANN AND R CLAIBORNE, eds. HP Publishing Co, New York, 1975, pp 203-212

17. VASINGTON, F D AND MURPHY, J V: Ca ++ uptake by rat kidney mitochondria and its dependence on re- sp i ra t ion and phosphory la t ion . J Biol Chem 237:2670, 1962

18. BRIERLEY, G P, MURER, E AND GREEN, D E: Partici- pation of an intermediate of oxidative phosphoryla- tion in ion accumulation by mitochondria. Science 140:60, 1963

19. MATSUBARA, T AND HAGIHARA, B: Action mecha- nism of phenothiazine derivatives on mitochon- drial respiration. J Biochem 63:156, 1968

20. ELIAS, E AND BOYER, J L: Chlorpromazine and its metabolites alter polymerization and gelation of actin. Science 206:1404, 1979

J. ELECTROCARDIOLOGY, VOL. 14, NO. 3, 1981