Neuroscience Letters, 137 (1992) 173-177

173

© 1992ElsevierScientificPublishers Ireland Ltd. All rights reserved0304-3940/92/$05.00 NSL 08488

Inhibition of cisplatin-induced emesis in ferrets by the non-NMDA receptor antagonists NBQX and CNQX A n d e r s F i n k - J e n s e n a, M a r t i n E. J u d g e a, J o h n B o n d o H a n s e n a, P o u l J a c o b s e n a, L e c h o s l a w T u r s k i b, J o h n O l n e y ~ a n d Tage H o n o r 6 a aCNS Division, Novo Nordisk A/S, MtJlOv (Denmark), bScheringAG, Department of Neuropsychopharmacology, Berlin ( FRG) and cWashington University School of Medicine, St. Louis, MO (USA)

(Received 10 June 1991;Revised version received 16 December 1991;Accepted20 December 1991) Key words: 2,3-Dihydroxy-6-nitro-7-sulphamoylbenzo(f)quinoxaline;NBQX; 6-Cyano-7-nitroquinoxaline-2,3-dione;CNQX; Excitatory amino

acid; Non-N-methyl-D-aspartatereceptor; Emesis; Cisplatin; Ferret The excitatory amino acid (EAA) receptor antagonists, 2,3-dihydroxy-6-nitro-7-sulpbamoylbenzo(f)quinoxaline(NBQX) and 6-cyano-7nitroquinoxaline-2,3-dione(CNQX), which preferentiallyblock non-N-methyl-D-aspartate(non-NMDA) subtypes of EAA receptors, effectively inhibit cisplatin-inducedemesis in ferrets. A high dose of cisplatin (10 mg/kg i.v.) was used which induced emesisin all saline-treatedcontrol ferrets. At 10 mg/kg i.v., NBQX totally prevented cisplatin-induced emesis in 5 of 6 ferrets and CNQX totally prevented emesis in 3 of 5 ferrets. By comparison,each of the 5-HT3inhibitors,zacoprideand ondansetron,at 1.0 mg/kgi.v.(a dose consideredin the high therapeuticrange for controlling emesis by these compounds), totally prevented emesis in 2 of 5 ferrets. It is concluded that non-NMDA antagonists effectivelyinhibit cisplatininduced emesis. They are potential antiemeticcompounds, alone or in combination with 5-HT3 antagonists or other more conventionaldrugs of choice.

Cancer chemotherapy is a prerequisite for the effective managenent of various kinds of malignancies, but the drugs used to control cancer have serious side effects. Nausea and vomiting are often observed, especially with drug combinations containing cisplatin. The discomfort may be so great that patients refuse further treatment. For many years, dopamine antagonists were considered the drugs of choice for counteracting nausea and vomiting, the most effective drug in this class being metoclopramide. However, because its antiemetic properties are weak, metoclopramide must be used in high doses which leads to blockade of dopamine receptors throughout the brain with resultant disagreeable side effects such as extra-pyramidal symptoms [10]. Recently, it was found that metoclopramide, in addition to being a dopamine antagonist, exerts antagonist activity against 5H T 3 receptors in the gut wall, a site of origin of emesis reflexes [9]. This prompted research identifying 5-HT 3 receptor antagonists as anti-emetic agents, and selective antagonists in this class have now been shown to be effective against cisplatin-induced emesis both in animal experiments and in the clinic (reviewed in ref. 1). HowCorrespondence." A. Fink-Jensen, CNS Division, Novo Nordisk A/S, Novo Nordisk Park, DK-2760 M~tl~v,Denmark. Fax: (45) 44663939.

ever, at low doses these agents are not able to protect completely against cancer chemotherapy-induced emesis in humans [1, 16], and at high doses they actually induce nausea and vomiting [5]. Thus, there is still a need for new effective anti-emetic compounds. In recent years ferrets have been used in anti-emetic research with cisplatin often being the agent used to elicit emesis. Data developed in the ferret model pertaining to metoclopramide [7] and 5-HT 3 receptor antagonists [4] proved reliable in predicting their efficacy as anti-emetics in humans. A recent abstract by Price and Olney [13] reported that emesis induced in ferrets by cisplatin could be suppressed by use of a combination of excitatory amino acid (EAA) antagonists which simultaneously block all 3 EAA receptor subtypes. However, the effects of N-methyl-D-aspartate (NMDA) and n o n - N M D A antagonism were not investigated separately. In the present study we used the ferret-cisplatin model to evaluate the anti-emetic potential of two EAA receptor antagonists, 6-cyano-7-nitroquinoxaline-2,3dione (CNQX) and 2,3-dihydroxy-6-nitro-7-sulphamoylbenzo(f)quinoxaline (NBQX), which are higly selective for n o n - N M D A receptors, especially the ~-amino-3hydroxy-4-isoxazole propionic acid (AMPA) receptor [14]. For comparison, using ferrets treated with same

174 dose o f cisplatin, we tested two 5-HT3 receptor antagonists, ondansetron and zacopride, alone or in combination with E A A antagonists, for ability to suppress the emesis response. Albino or Fitch male ferrets (obtained from a commercial breeder at Nexoe, D e n m a r k ) weighing 0.7-2.5 kg, housed singly and with free access to food and water were used. Each animal was used for one experiment only. Prior to surgery, each ferret was sedated with propionyl promazin (Combelen Vet; 1 mg/animal, i.p.) and anaesthetised with pentobarbital sodium solution (Mebumal; 35 mg/kg, i.p.). A chronic venous catheter (Intramedic polyethylene tubing (0.034 in i.d. and 0.050 in o.d.) cat no PE 90, Clay-Adams, New York) was implanted using a m o d ification o f the technique described by Florczyk and Schurig [7]. The catheter was fixed by use o f non-absorbable surgical suture silk (30, black, braided, Ethicon, Sollentuna, Sweden). Following surgery, each animal was treated with streptocillin (Streptocillin Vet; 0.2 ml 40,000 IE/animal i.m.). The ferrets were returned to their cages and allowed to recover for 3 days before testing. Animals were removed f r o m their h o m e cages and in-

jected with a test c o m p o u n d and cisplatin via i.v. catheter or a test c o m p o u n d i.p. and cisplatin via i.v. catheter. Cisplatin (10 mg/kg; vol. 5 ml/kg) was administered intravenously immediately following the injection o f one or more o f the following: N B Q X , C N Q X , zacopride, ondansetron. Immediately after injections, the animals were placed in individual observation cages and observed for 240 min. Emesis was defined by rhythmic abdominal contractions which were either associated with the expulsion o f solid or liquid material from the gastrointestinal tract (i.e. vomiting) or not associated with the passage o f material f r o m the gastrointestinal tract (i.e. retching), as described by B. Costall et al. [4]. The time o f onset o f vomiting or retching was measured for each animal as was the total n u m b e r o f vomiting and/or retching episodes. After the 240 min observation period, the ferrets were killed by an overdose o f pentobarbital solution (Mebumal). Cisplatin was obtained from Sigma and dissolved in 0.9% saline at 70-75°C followed by gradual cooling to 40°C, then administered i.v. N B Q X and C N Q X were prepared in distilled water, zacopride (AH Robins) in

TABLE I INHIBITION OF CISPLATIN-INDUCEDEMESIS IN FERRETS Ferrerts were injected with test compound(s) immediately followed by injection of cisplatin. Results are given as medians and minimum - maximum (in parentheses). Time for onset of emesis for all ferrets was taken as 240 min (the observation period) if a ferret was completely protected. Compound and dose

Number completely protected/ number treated

Onset of emesis (min) (all ferrets)

Total vomits/retches per ferret (all ferrets)

Saline control (i.v.) Saline control (i.p.)

0/18 0/8

66 (48 128) 70 (5~94)

118 (59-451 ) 110 (19-164)

NBQX (10 mg/kg i.p.) NBQX (30 mg/kg i.p.)

2/5 3/5

90 (53 240) 240 (57 240)*

14 ((~120)* 0 ((L47)**

NBQX (3 mg/kg i.v.) NBQX (10 mg/kg i.v.)

3/5 5/6

240 (53-240)* 240 (115 240)***

0 ((~62)*** 0 (0 8)***

Zacopride (0.1 mg/kg i.v.) Zacopride (1.0 mg/kg i.v.)

0/4 2/5

117 (100 145)** 116 (89 240)**

28 (4-33)*** 10 ((L74)***

NBQX (30 mg/kg i.p.) + Zacopride (0.1 mg/kg i.v.)

5/5

240 (240 240)***

0 ((~0)***

Significant reduction in total vomits and retches or delay in onset ofemesis (compared to controls) are indicated by P < 0.05, **P < 0.01, ***P < 0.001 (Mann-Whitney U-test). The responses and the onset of action in the two control groups (saline i.p. and saline i.v.) were not different (P > 0.5). There was a significant improvement of the anti-emetic action after combination of zacopride and NBQX compared with zacopride alone (P < 0.05), whereas the effect of the combination treatment did not differ from NBQX given alone (P > 0.3).

175

0.9% saline, and ondansetron (GR 38032F, Glaxo) in buffered saline solution (Zofran). All ferrets displayed exploratory behavior for approximately 30-60 min after being placed in the observation cages. After this exploratory phase, the non-emetic ferrets exhibited normal resting behavior throughout the observation period in contrast to the other ferrets showing increased activity prior to the emetic events. Since experimental drugs were administered in some cases intraperitoneally (i.p.) and in other cases intravenously (i.v.), two control groups were employed, one receiving saline i.p. and one i.v. In both control groups, intravenous injection of cisplatin (10 mg/kg) consistently caused an emetic response (retches and vomiting) in all of 24 ferrets (Table I). NBQX was effective in suppressing or preventing cisplatin-induced emesis (Table I). At 10 mg/kg i.v., NBQX completely abolished the emetic response in 5/6 ferrets, while delaying onset and reducing the intensity of the response in the remaining ferret. A lower i.v. dose and

i.p. administration of NBQX also antagonized cisplatininduced emesis. CNQX also conferred significant protection against cisplatin-induced emesis (Table II). At 10 mg/kg i.v., CNQX completely prevented emesis in 3/5 ferrets and reduced the retching and vomiting in the other two ferrets. Lower doses of CNQX (1 mg/kg i.v. and 7 mg/kg i.v.) were not able to inhibit emesis, and only a dose of 1 mg/kg i.v. increases the time for onset of emesis. The 5-HT 3 receptor antagonists, ondansetron and zacopride, resembled one another in anti-emetic potency. At 0.1 mg/kg i.v. emesis was not completely prevented by zacopride in any of 4 ferrets (Table I) and by ondansetron in only 1 of 5 ferrets (Table II). At this dose, neither compound significantly reduced the amount of retching or vomiting compared to i.v. saline controls. However, at a 10-fold higher dose, each compound completely protected 2 of 5 ferrets and significantly reduced retching and vomiting (Tables I and II). NBQX (30 mg/kg i.p.) in combination with zacopride

TABLE II INHIBITION OF CISPLATIN-INDUCED EMESIS IN FERRETS Experiments performed and analyzed as described in legend for Table I. Compound and dose

Number completely protected/ number treated

Saline control (i.v.)

0/18

CNQX (1 mg/kg i.v.) CNQX (7 mg/kg i.v.) CNQX (10 mg/kg i.v.)

0/5 0/5 3/5

83 (74~145)* 39 (36-106) 240 (80 240)***

126 (4~136) 126 (9 166) 0 (0-94)***

Ondansetron (0.1 mg/kg i.v.)

1/5

128 (86-240)**

64 (0-154)

Ondansetron (1.0 mg/kg i.v.)

2/5

154 (125 240)***

8 (0-52)***

3/7

194 (120240)***

11 (0-35)***

4/4

240 (24t~240)***

CNQX (7 mg/kg i.v.) +

Onset of emesis (min) (all ferrets) 66 (48-128)

Total vomits/retches per ferret (all ferrets) 118 (59~,51)

Ondansetron (0.1 mg/kg i.v.) CNQX (10 mg/kg i.v.) +

0 (04))***

Ondansetron (1.0 mg/kg i.v.) Significant reduction in total vomits and retches or delay in onset of emesis (compared to controls) are indicated by *P < 0.05, **P < 0.01, ***P < 0.001 (Mann-Whitney U-test). There was a significant improvement of the antiemetic action after combination of ondansetron (0.1 mg/kg i.v.) and CNQX (7 mg/kg i.v.) compared with ondansetron (0.1 mg/kg i.v.) alone (P < 0.05), and compared with CNQX (7 mg/kg i.v.) given alone (P < 0.05).

176 (0.1 mg/kg i.v.) completely protected 5 of 5 ferrets (Table I). CNQX (7 mg/kg i.v.) in combination with ondansetron (0.1 mg/kg i.v.) completely protected 2 of 5 ferrets and significantly reduced the emetic response (Table II), while CNQX (10 mg/kg, i.v.) in combination with ondansetron (1.0 mg/kg i.v.) completely protected 4 of 4 ferrets (Table II). The present study shows that the quinoxalinedione EAA receptor antagonists, NBQX and CNQX, are highly effective in blocking cisplatin-induced emesis in the ferret. Both of these agents are antagonists that inhibit binding to n o n - N M D A receptors, with preferential affinity being shown for the AMPA receptors [15]. The fact that NBQX is inactive at the N M D A receptor and at the glycine receptor [14], and that it has similar anti-emetic actions as CNQX, suggests that N M D A receptors are not of importance. It seems likely that both CNQX and NBQX block emesis by a central action, one reason being that excitatory amino acid agonists, even those that do not penetrate the blood-brain barrier (BBB), enter the area postrema, a brainstem region that lacks BBB. After reaching area postrema, such agents are known [12] to diffuse into immediately adjacent brain regions, referred to as the emesis motor centers [2]. It is believed that emesis signals are synaptically relayed from afferent to efferent limbs of emesis reflex arcs in this center, with E A A receptors being major mediators of this relay function [13]. The fact that binding sites for EAA receptor agonists have been found in the periphery (reviewed in ref. 6) - - e.g. on parasympathetic ganglia and vagal afferents - - makes a peripheral site of action possible too, however, the pharmacological importance of these binding sites needs further investigation. The specific 5-HT3 antagonists, zacopride and ondansetron, have been shown to antagonize cisplatin-induced emesis in ferrets [4, 11] and dogs [3]. We were, however, not able to confirm the previously reported finding [4] that zacopride (1.0 mg/kg i.v.) completely abolished the emetic response. Our failure to demonstrate such a high degree of efficacy for either zacopride or ondansetron may be explained in part by our longer observation time (4 h instead of 2 h) since emesis was delayed beyond 2 h in some of our ferrets and this would have been scored as complete protection in the study by Costall et al. [4]. In humans, 5-HT 3 antagonists have not been able to protect completely against cisplatin-induced emesis and it has been suggested that combining these agents with other anti-emetics may be necessary for optimal antiemetic protection [1, 15]. Moreover, it has been suggested that the dose-response curve for 5-HT 3 antagonists may have the shape of an inverted U, in that recent clinical trials of ondansetron for anti-psychotic proper-

ties in schizophrenic subjects revealed that progressive dose increases frequently induced nausea and vomiting as a side effect [5]. A very impressive anti-emetic effect was obtained with the combination of NBQX (30 mg/kg i.p.) and zacopride (0.1 mg/kg i.v.), which provided significantly better protection than the same dose of zacopride alone. We also found that combining CNQX (7 mg/kg i.v.) and ondansetron (0.1 mg/kg i.v.), each at a dose that did not significantly reduce the number of emetic or retching episodes by itself, did significantly reduce the number of such episodes, compared with saline, CNQX, or ondansetron. Combining the two compounds at higher doses (CNQX, 10 mg/kg i.v. and ondansetron, 1.0 mg/kg i.v.) gave a complete protection against emesis. Our findings indicate that combining EAA antagonists and 5-HT 3 inhibitors can substantially increase the anti-emetic effect above that obtainable with the 5-HT 3 inhibitor alone. It has to be clarified whether the combined treatment provides better protection than that achieved with the EAA antagonists alone. When complete dose-response data are available for EAA antagonists and other classes of anti-emetics, it may be possible to specify relatively low doses of each, which in combination will produce maximal antiemetic effects while minimizing the risk of serious side effects. This may be a realistic hope since CNQX and NBQX do not interact with either 5-HT 3 or dopaminergic receptor systems, which are the receptors through which most anti-emetic drugs produce their side effects (as well as therapeutic benefits). We would like to thank Dr. Hanne Gamst-Andersen for help with the surgical procedures and Karen Uth for excellent technical assistance.

1 Andrews, P.L.R., Rapeport, W.G. and Sanger, G.W., Neuropharmacology of emesis induced by anti-cancer therapy, Trends Pharmacol. Sci., 9 (1988) 334-341. 2 Carpenter, D.W., Neural mechanisms of emesis, Can. J. Physiol. Pharmacol., 68 (1990)230-236. 3 Cohen, M.L., Bloomquist,W., Gidda, J.S. and Lacefield,W., Comparison of the 5-HT~receptorantagonistpropertiesof ICS 205 930, GR38032F and zacopride, J. Pharmacol. Exp. Ther., 248 (1988) 197 201. 4 Costall, B., Domeney, A.M., Naylor, R.J. and Tattersall, F.D., Emesis inducedby cisplatin in the ferret as a modelfor the detection of anti-emeticdrugs, Neuropharmacology,26 (1987) 1321-1326. 5 Deveaugh-Geiss,J., 5-HT3 Antagonists, preclinical tests. In H.Y. Meltzer and R.E. Becker(Eds.), Proceedingsof NIHM-sponsored Conference on Atypical Antipsychotic Medications, Puerto Rico, Dec. 15, 1990, in press. 6 Erd6, S.L., Excitatoryamino acid receptorsin the mammalianperiphery, Trends Pharmacol. Sci., 12 (1991)426-429. 7 Florczyk,A.P. and Schuring, J.E., A techniquefor chronicjugular

177 catheterization in the ferret, Pharmacol. Biochem. Behav., 14 (1981) 255 257. 8 Florczyk, A.P., Schuring, J.E. and Bradner, W.T., Cisplatin-induced emesis in the ferret: a new animal model, Cancer Treat. Rep., 66 0982) 187-189. 9 Fozard, J.R., Neuronal 5-HT receptors in the periphery, Neuropharmacology, 23 (1984) 1473-1486. 10 Gralla, R.J., Metoclopramide, a review of clinical trials, Drugs, 25 (1983) 63-73. 11 Higgins, G.A., Kilpatrick, G.J., Bunce, K.R., Jones, B.J. and Tyers, M.B., 5-HT3 receptor antagonists injected into the area postrema inhibit cisplatin-induced emesis in the ferret, Br. J. Pharmacol., 97 (1989) 247-255. 12 Olney, J.W., Misra, C.H. and Rhee, V., Brain and retinal damage

from the lathyrus excitotoxin, fl-N-oxalyl-L-alpha-fl-diaminopropionic acid (OPAD), Nature, 264 (1976) 659-661. 13 Price, M.T. and Olney, J.E., Excitatory amino acid antagonists as anti-emetics, Soc. Neurosci. Abstr., 16 (1990) 377. 14 Sheardown, M.J., Nielsen, E.O., Hansen, A.J., Jacobsen, P. and Honor6, T., 2,3-Dihydroxy-6-nitro-7-sulfamoylbenzo(f)quinoxaline: a neuroprotectant for cerebral isehemia, Science, 247 (1990) 571-574. 15 Smith, D.B., Newlands, E.S., Rustin, G.J.S., Begent, R.H.J., Crawford, S.M., Bagshawe, K.D. and Carruthers, L., A phase I/II study of the 5-HT3 antagonist GR38032F in the anti-emetic prophylaxis of patients receiving high-dose cisplatin chemotherapy, Cancer Chemother. Pharmacol, 25 (1990) 291594.