TOXICOLOGY AND APPLIED PHARMACOLOGY 83, 525-530 (1986)

Stereospecific Effect of Naloxone Hydrochloride on Cyanide intoxication

PETER LEUNG,* DIANE M. SYLVESTER,? FAYE CHIOU,* LORI L. WAY,* E. LEONG WAY,$ AND JAMES L. WAY*.’

*Department of Medical Pharmacology and To.ricologj: Te.ras A&M University. College of Medicine. College Station, Te,vas 77843; 7 Washington State University, College ofPharmac.v, Pullman, U’ashington 99164; and

*Department qf Pharmacology, Sl210 University of California Medical Center, San Francisco, California 94143

Received September 21, 1985: accepted November 20. 198.5

Stereospecitic Effect of Naloxone Hydrochloride on Cyanide Intoxication. LEUNG, P., SYLVES-

TER. D. M.. CHIOU, F., WAY, L. L., WAY, E. L., AND WAY, J. L. (1986). Tosicol. Appl. Pharmacol. 83,525-530.Cyanideintoxicationinmicecanbeantagonizedbytheopiateantagonist.(-)naloxone HCl, alone or in combination with sodium thiosulfate and/or sodium nitrite. Potency ratios, derived from LD50 values were compared in groups of mice pretreated with sodium nitrite (SC, 100 mg/kg), sodium thiosulfate (ip, 1 g/kg), and (-)naloxone HCl (SC, 10 mg/kg) either alone or in various combinations. These results indicate that naloxone HCI provides a significant protection against the lethal effects of potassium cyanide. The protective effect of sodium thiosulfate, but not sodium nitrite, was enhanced with (-)naloxone HCl. The combined administration of sodium nitrite and sodium thiosulfate was further enhanced with (-)naloxone HCl. This protective effect of naloxone HCl against the lethal effect of cyanide appears to be restricted to the (-) stereoisomer, as the (+) stereoisomer, the inactive opiate antagonist, is also inactive in protecting against the lethal effects of cyanide. The mechanism of antagonism is discussed. o 1986 hdemic PES, I~C.

The site of action of the lethal effects of cyanide is believed to be predominantly the central nervous system, and the mechanism of the le- thal effects of cyanide is attributed to respi- ratory cessation. Since the opiate agonists and P-endorphin are reported to be associated with respiratory inhibition, which can be reversed by opiate antagonists, it is of interest to explore the use of opiate antagonists in reversing the lethal effects of cyanide. Most of the cyanide antagonists developed previousty have been systemic antagonists rather than site selective. This is the first attempt to target a cyanide antagonist predominantly to a specific organ, e.g., the brain. The present study was initiated to investigate the effect of naloxone. either alone or in various combinations with other cyanide antagonists, in protecting against the lethal effects of cyanide. Furthermore, the ste-

’ TO whom correspondence should be addressed.

reospecificity of this antagonist will be ex- plored in an attempt to relate its activity at opiate receptor sites with its antidotal prop- erties against cyanide.

METHODS

Materials. Potassium cyanide (KCN) and sodium thio- sulfate (Na&03 * 5H20) were obtained from MCB Man- ufacturing Chemists, Inc. (Cincinnati. Ohio). Sodium chloride(NaCI)andsodium nitrite(NaN02)wereanalyzed reagent grade chemicals purchased from J. T. Baker Chemicals (Phillipsburg. N.J.). (-)Naloxone HCl was do- nated by DuPont Pharmaceuticals (Garden City, N.Y.). (+)Naloxone HCI was generously provided by Chemistry and Life Science Division of the National Institute of Drug Abuse, Research Triangle Institute (Research Triangle Park, N.C.). All solutions were prepared in 0.9% NaCl immediately before use in glass-stoppered bottles and maintained at 0-2°C.

Methods. Swiss Webster [Crl:CFW(SW)BR] male mice (Charles River Breeding Laboratories, Inc., Wilmington, Mass.). weighing between 16 and 30 g and housed in tem-

525 0041-@08x/86 $3.00 Copyright 0 1986 by Academic Press. Inc. All rights of reproduction in any form resewed.

526 LEUNG ET AL.

perature-controlled (22-24°C) rooms, were maintained on Rat Chow 5012 (Ralston Purina Co., St. Louis, MO.) and water, ad libitum. These mice were fasted overnight and distributed randomly into experimental groups of 10 to assess the effects of different treatment on cyanide lethality. Median lethal dose (LDSO), based on 24-hour mortality, was determined for KCN either alone or in various com- binations with (-) or (+)naloxone HCl, NaNOz, and Na&Os, and the surviving mice were observed for one additional week. No apparent toxic effects were observed in animals receiving the indicated cyanide antagonists ei- ther alone or in various combinations. The dose and routes of administration of antagonists were as follows: sodium nitrite, 100 mg/kg, SC: sodium thiosulfate, 1 g/kg, ip; (+) and (-)naloxone HCl, 10 mg/kg, SC: potassium cyanide. 6 to 65 mg/kg, sc. Solutions were administered to the mice in a volume of less than 1% of the body weight per injection. The mice received sodium nitrite, sodium thiosulfate, (+) and (-)naloxone HCI 45, 15. and 10 min. respectively, prior to receiving potassium cyanide.

Cyanide concentrations in whole blood obtained from mice receiving a sublethal dose of potassium cyanide (5 mg/kg, SC) or pretreated with either (+) or (-)naloxone HCI (I 0 mg/kg, sc) 10 min prior to cyanide administration were determined by microdiffusion analysis with the pyr- azolone-pyridine calorimetric assay as described by Mor- gan et al. (1979). At the indicated time after cyanide ad- ministration, each mouse was anesthetized and exsangui- nated by cardiac puncture; 0.5-ml aliquots of whole blood were employed in each assay. Each experiment was re- peated three separate times and the results were expressed as the .U * 1 SE.

Statistical nnalws. The LD50 values were analyzed statistically by the method of Litchfield and Wilcoxon ( 1949). as adapted to computer analysis, with confidence limits of l9/20 probability, and all experiments were re- peated at least two times. Efficacy of the antagonist(s) was expressed as potency ratios, which were derived from LDSO values of KCN with and without antagonist.

RESULTS

The efficacy of (-)naloxone HCl, either alone or administered in combination with sodium nitrite and/or sodium thiosulfate, in antagonizing the lethal effects of cyanide is summarized in Table 1. The median lethal dose of potassium cyanide (experiment 1) is consistent with earlier studies (Schwartz et al.. 1979). Administration of (+)naloxone HCI provided no significant protection against cy- anide intoxication. In contrast, (-)naloxone HCl alone provided significant protection against the lethal effects of cyanide (experi- ment 3). Sodium nitrite (experiment 4) pro- duced a significant increase in the protection against cyanide toxicity, but when sodium ni- trite was administered in combination with (-)naloxone HCl or (+)naloxone HCl (exper-

TABLE I

EFFECT OF (-) AND (+)NALOXONE HCl, NaN02, AND Na&O, ON THE LD50 OF KCN IN MICE

Treatment before KCN (SC)

Expt. NaNOz (SC) Na&O, (ip) (+)Naloxone HCI (SC) (p)Naloxone HCl (SC) LD50 (limits. p = 0.05)” No. k/k) (g/k) k/W k/kg) (mg/kg)

1 0 0 0 0 7.0 (6.6-7.4) 2 0 0 0.01 0 7.0 (6.7-7.4) 3 0 0 0 0.01 9.5 (8.7-10.4) 4 0.1 0 0 0 24.5 (23.8-25.2) 5 0.1 0 0.0 1 0 22.0(21.1-23.0) 6 0.1 0 0 0.01 23.5 (23.0-24.7) 7 0 1.0 0 0 21.5 (19.9-23.2) 8 0 1.0 0.0 1 0 21.0 (19.8-22.3) 9 0 1.0 0 0.01 26.0(23.8-28.5)

10 0.1 1.0 0 0 45.0 (43.1-46.9) II 0.1 1.0 0.0 I 0 44.0 (42.0-46.0) 12 0.1 I.0 0 0.01 55.0 (49.1-61.6)

’ Each LD50 value was obtained from 4 or more graded doses of KCN administered to 4 or more groups of mice containing 10 mice/group.

NALOXONE HYDROCHLORIDE ON CYANIDE INTOXICATION 527

iments 5 and 6) no additional protection was observed. When sodium thiosulfate (experi- ment 7) was administered in combination with (+)naloxone (experiment 8), no enhancement was observed; however, with (-)naloxone HCl (experiment 9), a significant enhancement of the protective effect was observed. Further- more, the combined administration of sodium nitrite and sodium thiosulfate (experiment 10) provided additional protection, which was not enhanced with (+)naloxone (experiment 1 l), but its antidotal effect is further augmented when (-)naloxone HCl (experiment 12) is in- cluded with this regimen. Although the pro- tection afforded by (-)naloxone HCl alone is less than that of sodium nitrite or sodium thiosulfate, it should be noted that no con- vulsions occurred in groups receiving (-)nal- oxone HCl, whereas convulsions usually were noted in groups treated with (+)naloxone HCl, sodium nitrite, and/or sodium thiosulfate.

The slopes of the log dose-mortality regres- sion lines (Fig. 1) in all experiments were an- alyzed statistically and were determined to be not significantly different from each other. Therefore, the procedure employed in calcu- lating the potency ratios is valid. Other com- binations of drug regimens were compared and expressed as potency ratios (Table 2) to obtain a paired comparison of relative efficacy be- tween regimens. In this regard, (-)naloxone HCl alone produced moderate but significant

-- Naloxone HCI

90

E 70

3 50 2

.g

Lx BP

30 L- 10

3 20 30 50 70

DOSE CmWKg)

FIG. I. Dose-mortality regression lines for KCN with and without pretreatment by (-)naloxone HCI, sodium

nitrite, and sodium thiosulfate as described in Table 1.

protection against cyanide intoxication (po- tency ratio = 1.4). In addition, nitrite was ef- fective in antagonizing cyanide poisoning, but its efficacy was not augmented by (-)naloxone HCl. However, (-)naloxone HCl enhanced the antidotal potency of sodium thiosulfate (potency ratio = 1.21). Similarly, the com- bined effect of sodium nitrite and thiosulfate was also increased (potency ratio = 1.22).

Additional studies were performed to assess the stereospecificity of naloxone HCl on the antagonism of cyanide intoxication. An equivalent dose of (+)naloxone HCl was ad- ministered alone or in combination with other antidotes (Table 2). The data in Tables 1 and 2 indicate that the administration of (+)nal- oxone HCl, either alone or in various com- binations of sodium nitrite and sodium thio- sulfate, did not provide any or additional pro- tection against cyanide intoxication.

Concentrations of cyanide in whole blood obtained from mice treated with a sublethal dose of potassium cyanide (5 mg/kg) or pre- treated with either (+) or (-)naloxone HCl 10 min prior to cyanide administration are shown in Fig. 2. The concentration of cyanide in the blood increased rapidly after cyanide admin- istration and within 7 min reached a maxi- mum, after which it began to decrease expo- nentially. Pretreatment with (-)naIoxone HCl reduced the maximal concentration of cyanide in the blood by 45%, whereas pretreatment with (+)naloxone HCl did not alter the max- imal values of cyanide in the blood.

DISCUSSION

Physiologic and pharmacologic considera- tions have not played a major role in attempts to develop more effective cyanide antagonists. However, these facets should be emphasized since lethality in response to acute exposure to cyanide is due to respiratory cessation. Ir- respective of the in vitro biochemical mecha- nisms involved in cyanide toxicity, the pre- dominant sensitive site for the oxygen depriv- ing action of cyanide resides in the central

528 LEUNG ET AL

TABLE 3

COMPARISON OF (-)NALOXONE HCI AND (+)NALOXONE ON THE EFFECTS OF OTHER ANTILICWES IN THE PROTECTION AGAINST KCN INTOXICATION

Expt No.

I 2 3

4 5 6

7 8 9

Treatment

Control’ (+)Naloxone HCI (-)Naloxone HCI

N&03 (+)Naloxone HCI + Na&O, (-)Naloxone HCI + Na2S20,

NaNO, + Na2S20, (+)Naloxone HCI + NaN02 + NaZS203 (-)Naloxone HCI + NaNO? + Na2S203

Slope function a (limits: p = 0.05)

1.13 (1.06-1.21) 1.12(1.06-1.19) I.22 ( I .oo- I .48)

I.21 (1.05-1.39) 1.13 (0.97-1.31) 1.23 (0.94-1.62)

1.10 (0.98-1.23) 1.l0(0.91-1.34) 1.26 (0.98-1.61)

Potency ratiob (limits: y = 0.05)

l.OO(O.Yl-1.10) 1.40 (1.30-1.50)

0.99 (0.9 I-1.07) I.21 (1.08-1.36)

0.98 (0.90-I .08) 1.22 (1.16-1.X)

’ None of the slopes were significantly different from one another. b Potency ratio = LD50 of KCN (with antagonists)/LD50 of KCN without antagonists.

KCN in isotonic saline was employed as the control.

nervous system, especially those sites which are concerned with respiratory control. Therefore, in treating acute cyanide toxicity the major focus should be the reversal of the respiratory cessation and ameliorating factors which may contribute to this deleterious effect.

0- 10 20 30

TIME (minutes)

FIG. 2. Cyanide concentrations in whole blood obtained from mice receiving a sublethal dose ofpotassium cyanide (5 m&kg. sc; 0 - 0) or pretreated with either (+)naloxone HCI (0 --- 0; 10 mg/kg) or (-)naloxone HCI (0 - 0: 10 me/kg) 10 min prior to cyanide administration were de- termined by microdiffusion analysis in conjunction with the pyrazolone-pyridine calorimetric assay as described under Methods.

In cyanide poisoning, an extreme stress related condition develops and a release of endorphins could occur. Therefore, this mechanism may play a role in exacerbating cyanide-induced respiratory depression, particularly since morphine is a classic respiratory depressant, and the potency of the endorphins for eliciting many pharmacologic effects is much greater than morphine. Therefore, one of the mech- anisms of naloxone antagonism of the lethal effects of cyanide then may be attributed to the reversal of the inhibitory effects of the en- dorphins on respiration. Since the lethal effect of cyanide is ascribed to respiratory cessation, then it seems reasonable that naloxone may be exerting its effect by antagonizing the cy- anide-induced respiratory depression as is manifested by respiratory cessation. The pre- dominant site for the lethal actions of cyanide is probably the central nervous system. since the brain is a very sensitive organ to cyanide. In the event of cyanide poisoning, electrical silence and electrical abnormality have been observed in EEG recordings by several Iabo- ratories (Burrows et al.. 1973). The patholog- ical signs of cyanide poisoning also are related to the central nervous system. In this connec- tion, demyelination of the optic nerves and cerebral atrophy have been reported in patients

NALOXONE HYDROCHLORIDE ON CYANIDE INTOXICATION 529

with elevated amounts of cyanide in their blood resulting from excessive tobacco smok- ing and chronic alcoholism (Jestico et al., 1984).

To explore more in depth about the respi- ratory mechanism involved in cyanide lethal- ity, the role of the various opiate receptors and ligands are presently being pursued. Whether or not naloxone exerts its antidotal effect on cyanide lethality by inhibition of the phar- macologic effect of endogenous peptide has any contributory validity, the fact remains that it was a result of this hypothesis that a probable new class of cyanide antagonist has been dis- covered. An alternative mechanism in which naloxone may play a role in cyanide antago- nism is related to its cardiovascular effects that are mediated centrally by opiate receptors.

The results of the present study demonstrate that (-)naloxone HCl is similar in many re- spects to oxygen as a cyanide antagonist (Way et al., 1966). In this connection, (-)naloxone HCl alone provides moderate, but significant protection against the lethal effects of cyanide. In addition, (-)naloxone HCl enhances the protective effect of sodium thiosulfate, either alone or in combination with sodium nitrite, but does not potentiate that of sodium nitrite.

Other cyanide antagonists, including co- baltous chloride (Isom and Way, 1973) and sodium pyruvate (Schwartz et ul., 1979) ap- pear to exert their protective effects by inter- acting with cyanide to form a stable complex. In contrast, no intermediate complex was de- tected in vitro by thin-layer chromatography when (-)naloxone HCl was incubated with potassium cyanide under physiological con- ditions.

The inability of (+)naloxone HCl to provide protection against cyanide intoxication, or to enhance the antidotal potency of other cyanide antagonists, indicates that the protective effect of naloxone HCI may be stereospecific. This suggests that toxic manifestations of cyanide may be mediated in part by the release of en- dogenous opiate peptides. The mechanism is probably similar to the stereospecific reversal of the hypotension by the administration of naloxone to laboratory animals following

spinal cord transection (Faden et al.. 198 1). In addition, the association of an elevation in plasma /3-endorphin immunoreactivity with spinal cord injury (Faden et al.. 1982) suggests that the cardiovascular effects of naloxone are mediated through opiate receptors within the central nervous system. However, the fact that opiate receptors have been noted in red blood cells (Yamasaki and Way, 1983) and opiate ligands affect red cell deformability (Rhoades et al.. 1985) suggest that peripheral mecha- nisms cannot be excluded (Larsen et al., 1980). In any event, an improvement in the circu- lation or blood flow through organs which play a crucial role in the metabolism of cyanide would provide protection against cyanide in- toxication. This idea is consistent with the ob- servation that pretreatment with (-)naloxone HCl reduces the maximal concentration of cyanide in the blood. In contrast, pretreatment with the inactive opiate antagonist, (+)nal- oxone HCl. did not alter the maximal con- centration of cyanide in the blood. Therefore. naloxone may represent a new category of pharmacological agents which antagonize cy- anide intoxication by exerting its action(s) on the central nervous system. In conclusion, we have demonstrated that (-)naloxone HCI provides prophylactic protection against cya- nide intoxication and that its actions are ste- reospecific.

ACKNOWLEDGMENTS

This work was supported by research funds from the

National Institute of General Medical Sciences, National Institute of Environmental Health Sciences, and U.S.

Army Medical Research Development Command. The authors thank Dr. John W. Holaday for many valuable discussions and advice. Also the authors thank Mrs. Mar-

ion Wilson and Ms. Lynn Baxter for the preparation of

this manuscript.

REFERENCES

BURROWS, G. E.. LIU, D. H. W.. AND WAY, J. L. ( 1973).

Effect of oxygen on cyanide intoxication: V. Physiologic effect. J. Pharmacol. Exp. Ther. 184, 739-748.

FADEN, A. I., JACOBS, T. P., AND HOLAIIAY. J. W. (1982). Neuropeptides and spinal cord injury. Adv. Biochem. Psychopharm. 33. 13 I - 138.

530 LEUNG ET AL.

FADEN, A. I., JACOBS, T. P., MOUGEY, E., AND HOLADAY, J. W. ( I98 1). Endorphins in experimental spinal injury:

Therapeutic effect of Naloxone. Ann. Neural. 10, 326- 332.

ISOM. G.. AND WAY, J. L. (1973). Cyanide intoxication:

Protection with cobaltous chloride. Toxicol. Appl.

Pharmacol. 24,449~456. JESTICO. J. V.. O’BRIEN, M. D., TEOH, R., TOSELAND.

P. A., AND WONG, H. C. ( 1984). Whole blood cyanide

levels in patients with tobacco amblyopia. J Nemo/ Neurosurg. Psychiat. 47, 573-578.

LARSEN, F. L., KATZ, S., AND ROUFOGALIS, B. D. (1980).

Red blood cell viscosity: The effects of calcium and A23 187. Proc. West. Pharmacol. Sot. 23, 36 I-364.

LITCHFIELD, J. T., JR., AND WILCOXON, F. ( 1949). A sim-

plified method of evaluating dose-effect experiments. J. Pharmacol Exp. Ther. 96,99- 113.

MORGAN, R. L., ISOM, G. E., AND WAY, J. L. (1979).

Resolution of thiosulfate interference in cyanide deter- mination. Toxicol. Appl. Pharmacol. 50, 323-328.

RHOADES, D. L.. YAMASAKI, Y., WAY, E. L. (1985). Opi- ates reduce human red blood cell deformability. (Ab-

stract), International Narcotic Research Conference, p. 167.

SCHWARTZ, C., MORGAN, R. L., WAY, L. M., AND WAY, J. L. (1979). Antagonism of cyanide intoxication with

sodium pyruvate. Toxicol. Appl. Pharmacol. 50. 437- 441.

WAY, J. L.. GIBBON, S. L., AND SHEEHY, M. (1966). Effect of oxygen on cyanide intoxication. I. Prophylactic pro-

tection. J. Pharmacol. Exp. Ther. 153, 381-385. YAMASAKI, Y., AND WAY, E. L. (1983). Possible inhibition

of calcium pump of rat erythrocyte ghosts by opioid k

agonists. Life Sci. 33 (Suppl. I), 733-736.