Embryotoxic effects of sodium azide infusions in the Syrian hamster

754-759(1990)

FLJNDAMENTALANDAPPLIEDTOXICOLOGY1~,

Embryotoxic

Effects of Sodium Azide Infusions in the Syrian Hamster’

THEODORE R. SANA, VERGIL H. FERM,* ROGER P. SMITH, ROBERTKRUSZYNA, HARRIET KRUSZYNA, AND DEAN E. WILCOX? Department of Pharmacology and Toxicology New Hampshire 03756; and TDepartment

Received

and *Department ofAnatomy, Dartmouth of Chemistry, Dartmouth College, Hanover,

April

23, 1990; accepted

August

Medical School, Hanover, New Hampshire 03755

2, 1990

Embryotoxic Effects of Sodium Azide Infusions in the Syrian Hamster. SANA, T. R., FERM, V. H., SMITH, R. P., KRUSZYNA, R., KRUSZYNA, H., AND WILCOX, D. E. (1990). Fundam. Appl. Toxicol. 15, 754-759. Pairs of osmotic minipumps containing 400 mg/ml (6.15 M) sodium azide in distilled water were subcutaneously implanted in timed pregnancy Syrian golden hamsters. The total delivered dose was calculated as 6 X lo-* mmol kg-’ hr-’ at the maximal pumping rate. Most dams exhibited obvious signs of toxicity during the period of pump implantation which was Days 7 through 9 of gestation. After removal of the pumps the dams were euthanized on Day 13 of gestation, and the uteri were removed for counting of the number of living, malformed, and resorbed fetuses. This dose rate resulted in a significantly increased incidence of resorptions of embryos over that in a control group implanted with pumps delivering only distilled water. The incidence of gross malformations exclusively in the form of encephaloceles was not different between control and azide-infused groups. The extent of nitrosylation of circulating hemoglobin was followed with time and found to involve only about 0.1% of the total blood pigment. Thus, this commercially important and widely distributed chemical with high acute toxicity is not considered to be teratogenic in hamsters, and it produces embryotoxicity only at dose rates that result in toxic signs in the dams. o 1990 Society of Toxicology.

Sodium azide (NaN3) is used to provide nitrogen gas for the inflation of automobile “air” bags (Buckheit and Fan, 1978). Its controlled ignition in the air bag system in the presence of iron/copper catalysts very rapidly releases large volumes of nitrogen gas (Fig. 1). In contrast, heavy metal salts of azide such as copper, CuN, , and lead, Pb(N& , are violently explosive and can be detonated by physical impact. Sodium azide is also a widely used preservative and antifungal agent

in laboratory reagents, and it is occasionally encountered as an insecticide, herbicide, and fungicide. Azide’s high acute toxicity approaches that for sodium cyanide, but its mechanism of action is not clearly understood (Gosselin et al., 1984). It has other properties in common with cyanide as well such as its ability to produce a fulminating, convulsive death in central respiratory arrest, an inhibition of cytochrome aa3 (cytochrome oxidase) activity, stimulation of carotid body chemoreceptors, and an ionic complex with methemoglobin. Induced methemoglobin’ A preliminary report on these findings was made at emia provides a small but statistically signifithe 29th Annual Meeting of the Society of Toxicology, Miami Beach, FL, and published as Toxicologist 10,124, cant degree of protection against acute azide 1990. poisoning in mice (Abbanat and Smith, 0272-0590/90$3.00 Copyright 0 1990 by the Society of Toxicology. All rights of reproduction in any form reserved.

754

AZIDE EMBRYOTOXICITY Arbag

Reaction:

4NaN3 + 0, -

6N2+2Na,0

2H,O -4NaOH

Blood Vessel, Platelet Reaction: 4NaNj

+ 702 -

2H20 l2NO + 2Na20 4NaOH

FIG.1. Oxidation reactions of sodium azide that generate nitrogen in “air bags” and nitric oxide in blood vessels, platelets, and other biological systems. Alternative reactions are possible; for example, a balanced reaction for blood vessels which yields both NO and N2 can be written.

1964), but azide-induced inhibition of cytochrome aa activity does not appear to account for its high lethality (Smith et al., 1977). At the same time azide has biological activities that are not shared with cyanide in that it is an extremely potent, directly acting vasodilator and an inhibitor of platelet aggregation and adhesion (Kaplita et al., 1984; Schwerin et al., 1983). These effects may not be due to azide per se, but to its biotransformation product, nitric oxide (NO; Fig. 1). At least small amounts of azide are converted to NO in red blood cell suspensions and in intact mice (Kruszyna et al., 1987; 1988). Thus, it appears to be a member of a larger group of drugs and xenobiotics that are now known as the NO vasodilators. Because of the high biological activity of NO and newly discovered possible roles for it in the nervous and immune systems (Collier and Valiance, 1989), it is no longer clear whether azide or NO is primarily responsible for the high acute toxicity. Structural changes in the cerebellar cortex ranging from slight local alterations to almost complete decortication occurred in monkeys given azide iv (Mettler and Sax, 1972). Spratt ( 1950) found that azide caused degeneration of all tissues in explanted chick embryos; however, we have not been able to locate in vivo mammalian teratology data on this chemical. Sodium cyanide has been shown to be teratogenic in the golden hamster when in-

IN HAMSTERS

755

fused continuously via osmotic minipumps between Days 6 and 9 of gestation (Doherty et al., 1982). Because of the other similarities noted above between cyanide and azide, we have examined sodium azide for possible embryotoxic and/or teratogenic effects. METHODS Male and pregnant and nonpregnant female Syrian golden hamsters (LKV strain) of known gestational age were obtained from Charles River Laboratories (North Wilmington, MA) and maintained under a 12-hr dark/ 12-hr light cycle. They were provided with tap water and laboratory chow (Purina Lab Chow, Ralston Purina Co., St. Louis, MO) ad libitum and housed individually in cages with pine shaving bedding. The day after the evening of breeding was considered Day 1 of gestation. Experiments were conducted using both male and nonpregnant female hamsters in order to determine a range of dose rates that would permit most animals to survive, yet be of sufficient magnitude to detect adverse developmental events in pregnant hamsters if they occurred. There were no obvious differences in terms of their sensitivities to sodium azide among males, nonpregnant females, and pregnant females. Alzet osmotic minipumps (Alza, Palo Alto, CA) were filled with fresh solutions of sodium azide (reagent grade, Fisher Scientific Co., Fairlawn, NJ) in distilled water or distilled water alone. Animals were weighed on the day of implantation (the seventh day of gestation for pregnant animals) and solutions of appropriate concentration were prepared to deliver the chemicals at previously calculated dose rates in mg kg-’ hr-‘. After loading the pumps, they were soaked in 0.9% saline for up to 4 hr in accord with the manufacturer’s directions for initiating the pumping action. The hamsters were anesthetized with 65 mg/kg pentobarbital sodium (Abbott Laboratories, North Chicago, IL) and shaved with clippers along the midline of the back where a 1.5-cm incision was then made for subcutaneous minipump insertion. Because of the limits of solubility of the chemicals, it was necessary to implant two pumps in order to give the desired dose rate. Controls also were implanted with two pumps. The incision was then closed with wound clips. On Day 9 of gestation, 2 days after pump implantation, the surviving hamsters were anesthetized as before, and both pumps were removed through the original incision which was then reclosed. This interval of time in the pregnant hamster is the period of greatest susceptibility to teratogenic stimuli (Ferm, 1967; Ferm and Hanlon, 1985). Two azide-treated dams from a total of 15 died before pump removal, and these animals were not included in the study. In 6 of 13 dams pumps were re-

756

SANA ET AL.

moved before 48 hr (3 at 40 hr and 3 at 42 hr) because of the severity of the dyspnea, hypothermia, and ataxia. The condition of these animals subsequently improved, and they were included in the study. Pregnant hamsters were sacrificed on Day 13 of gestation by CO* asphyxiation. The gestational sacs were removed, and the number of living, malformed, and resorbed fetuses was counted. The living fetuses were weighed and fixed in Bouin’s solution. Statistical comparisons were made between sets of treated hamsters and control animals using the Wilcoxon ( 1945) rank sum test and a modified Student’s t test (Sokol and Rohlf, 198 I). In order to evaluate the extent of nitrosylation of circulating hemoglobin due to azide infusions, blood samples were taken from nonpregnant female hamsters for examination by electron paramagnetic resonance (EPR) spectroscopy (Kruszyna et al., 1988). The protocol for the blood studies was identical to that for the embryotoxicity studies except that samples at the indicated times were taken by closed chest cardiac puncture after euthanasia with COZ.

RESULTS In two male hamsters sodium azide produced death within 24 hr after implantation of minipumps delivering at dose rates of 6.8 and 9.1 X 10e2 mmol kg-’ hr-‘. Two other hamsters at 50 and 25% of the larger dose rate (4.6 and 2.3 X low2 mmol kg-’ hr-‘, respectively) survived for the 4%hr exposure period but exhibited signs of decreased food intake, hyperpnea, lethargy, and hypothermia. Thereafter, the condition of the hamster receiving 4.6 X low2 mmol kg-’ hr-’ further deteriorated so that after 3 days it exhibited labored breathing, ataxia, muscle rigidity, convulsions, and death. On autopsy all organs were intact. Fluid was found in the peritoneal cavity, and the lungs and peritoneal wall had a pinkish color. The pumps of the animal receiving 2.3 X 1OP2 mmol kg-’ hr-’ were removed after 72 hr, and it recovered completely. In timed pregnant hamsters no fetal resorptions or malformations were observed at dose rates of sodium azide from 1.1 up to 5.3 X lop2 mmol kg-’ hr-‘. The litter data obtained with 6 X 10m2mmol kg-’ hr-’ are sum-

TABLE 1 EFFECTOFSODIUMAZIDEINFUSIONS ONHAMSTERLITTERS' Parameter

Control

Azide

Number of litters Total implantationsb Total fetuses Average No. fetuses per litter Total adverse reactionsC Total malformations Percentage affected Average FTA’

14 185 183 13.1 3 1 1.6 9.8

13 176 146 11.2 35 5d 19.9 29.6s

’ Sodium azide dose rate at 6 X lop2 mmol kg-’ hr-’ for 40-48 hr (Days 7 through 9 of gestation). Controls received distilled water. Two azide-treated dams died before removal of the pumps and are not included above. b The sum of the total fetuses and resorption sites. ’ The sum of the resorptions and malformations (encephaloceles). d Distributed among four litters. e Freeman Tukey arcsin in degrees. fSignificantly different from control group (t, < 0.0065).

marized in Table 1. The number of resorption sites and malformations (exclusively in the form of encephaloceles) were counted and pooled under the designation of “adverse reactions.” The incidence of such adverse reactions was significantly (p < 0.0065) higher in treated hamsters than in controls as also was the number of resorptions. In contrast, the incidence of malformations alone was not significantly different between control and azide-treated animals (Table 1). Almost all dams showed signs of toxicity at this dose rate. For some animals the signs were so severe that the pumps were removed at 40 hr instead of the full 4%hr exposure period (see above). These hamsters usually showed a prompt amelioration of toxic signs and their fetuses were included in the study after euthanasia on Day 13. There were 6/6 litters with adverse reactions in the dams exposed for 40 hr and 7/7 litters with adverse reactions in the dams exposed for 48 hr.

AZIDE EMBRYOTOXICITY

IN HAMSTERS

757

FIG. 2. Representative EPR spectra of hamster blood drawn under various control and experimental conditions. All hamsters receiving only distilled water or last exposed to azide 16 hr previously had the type of EPR signal indicated by the light stippled trace. All hamsters that were being exposed to sodium azide at the time of blood sampling had the other type of EPR signal shown in bold with the three-line hyperline pattern around 3300 G characteristic of nitrosylated a-heme subunits under appropriate conditions (Kruszyna etal., 1987).

Figure 2 shows the EPR signals found in hamster blood from nonpregnant females under control and experimental conditions. All animals that were implanted with pumps delivering distilled water and hamsters that had been implanted with pumps delivering sodium azide which had been removed 16 hr earlier exhibited the EPR signal shown as control. The experimental trace shows the characteristic nitrosyl hemoglobin EPR signal found in all animals that were receiving sodium azide at the time that the blood sample was taken. The extent of nitrosylation of circulating hemoglobin from the latter group of hamsters at various times is summarized in Fig. 3. DISCUSSION The weak signal in the g = 2 region found in control animals and in animals in which

the pumps had been removed 16 hr earlier is believed to originate from a naturally occurring, paramagnetic, copper-containing plasma protein (Foster et al., 1973) and not endogenous nitrosylated heme for the following reasons: ( 1) it does not have the correct g value for nitrosylated heme; (2) even at high amplification it lacks the three-line hyperfine pattern characteristic of nitrosylated heme; (3) after centrifugation of human blood it was associated with plasma and not red cells; and (4) we have encountered a similar signal previously in control mouse and human blood while systematically searching for endogenous nitrosylated heme (unpublished observations). The levels of nitrosylated heme found agree in general with acute experiments conducted previously in mice (Kruszyna et al., 1988) in that they involve only about 0.1% of the total circulating hemoglobin. Clearly,

758

SANA ET AL.

Moreover, despite the high acute toxicity of sodium azide (Gosselin et al., 1984), embryotoxic effects were obtained only at very high dosing rates, and rates that produced signs of toxicity in the dams. ACKNOWLEDGMENTS

Hours

FIG. 3. Micromoles of total circulating nitrosylated heme(I1) in nonpregnant female hamsters as a function of time after implantation of osmotic mininpumps containing sodium azide. Time 0 indicates the time of implantation after equilibration of the pumps in saline in vitro. Vertical bars indicate the SEM for three or four animals; remaining values are single observations. Open squares indicate animals that received pumps containing distilled water only. Open triangles represent hamsters that received azide for 48 hr after which the pumps were removed; the closed triangle is a control animal that received distilled water for the entire 64 hr. Open circles indicate animals receiving sodium azide at a dose rate of 2.5 to 7.1 X lo-’ mmol kg-’ hr-‘. Only blood from the animals receiving azide showed the hyperfine EPR pattern characteristic of nitrosylated heme (see also Fig. 2 and text).

nitric oxide under these conditions does not produce an anemic hypoxia as does carbon monoxide; i.e., it is not a “blood” poison. Finally, we note the decrease in circulating nitrosylated hemoglobin beginning at about 48 hr in animals still being infused with azide. The model of osmotic minipump used in these studies is rated to continue delivery of material for a week. If the levels of HbNO truly fall off at 48 hr even though the pumps are still functional, there appears to be some kind of adaptation phenomenon involving nitric oxide formation, removal, or both for which we have no explanation at present. We conclude from these results that sodium azide infusions are embryotoxic in the Syrian golden hamster, but in contrast to sodium cyanide (Doherty et al., 1982), sodium azide was not considered to be teratogenic.

This work was supported in part by NIH Grant HL 14 127 from the National Heart, Lung and Blood Institute (R.P.S.). The Bruker ESP-300 spectrometer was purchased with funding from the NSF (Grant CHE870 1406). We are grateful to Dr. William M. Layton, Jr., of the Department of Anatomy for help with the statistical analyses and to Dr. Lucile Smith of the Department of Biochemistry for helpful discussions.

REFERENCES ABBANAT, R. A., AND SMITH, R. P. (1964). The influence of methemoglobinemia on the lethality of some toxic anions. I. Azide. Toxicol. Appl. Pharmacol. 6, 576-583.

BUCKHEIT, B., AND FAN, W. (1978). Sodium azide in automotive air bags. National Highway Traffic Safety Administration, U.S. Department of Transportation, Washington, DC, March 30. COLLIER, J., AND VALLANCE, P. ( 1989). Second messenger role for NO widens to nervous and immune systems. Trends Pharmacol. Sci. 10,427-43 1. DOHERTY, P.A., FERM, V.H., ANDSMITH, R.P.(1982). Congenital malformations induced by infusion of sodium cyanide in the golden hamster. Toxicol. Appl. Pharmacol.

64,456-464.

FERM, V. H. (1967). The use of the golden hamster in experimental teratology. Lab. Anim. Care 17, 452462.

FERM, V. H., AND HANLON, D. P. (1985). Constant rate exposure of pregnant hamsters to arsenate during early gestation. Environ. Res. 37,425-432. FOSTER, M. A., POCIUINGTON, T., MILLER, J. D. B., AND MALLARD, J. R. (1973). A study of electron spin resonance spectra of whole blood from normal and tumor bearing patients. Brit. J. Cancer 28,340-348. GOSSELIN, R. E., SMITH, R. P., AND HODGE, H. C. ( 1984).

Clinical

Toxicology

of Commercial

Products,

5th ed. Williams &Wilkins, Baltimore. KAPLITA, P.V., BORISON,H.L., MCCARTHY, L.E., AND SMITH, R. P. ( 1984). Peripheral and central actions of sodium azide on circulatory and respiratory hemostasis in anesthetized cats. J. Pharmacol. Exp. Ther. 231, 189-196.

AZIDE EMBRYOTOXICITY H., KRUSZYNA, R., SMITH, R. P., AND WILCOX, D. E. (1987). Red blood cells generate nitric oxide from directly acting nitrogenous vasodilators. Tox-

KRUSZYNA,

icol. Appl. Pharmacol. 91,429-438. KRUSZYNA, R., KRUSZYNA, H., SMITH,

R. P., AND WILCOX, D. E. (1988). Generation of valency hybrids and nitrosylated species of hemoglobin in mice by nitric oxide vasodilators. Toxicol. Appl. Pharmacol. 94,458465. METTLER, F. A., AND SAX, D. S. (1972). Cerebellar cortical degeneration due to acute azide poisoning. Brain 95,505-5 16. SCHWERIN, F. T., ROSENSTEIN, R., AND SMITH, R. P. (1983). Cyanide prevents the inhibition of platelet ag-

759

IN HAMSTERS

gregation by nitroprusside, hydroxylamine and azide. Thromb. Haemostasis SMITH, L., KRUSZYNA,

50,780-783.

H., AND SMITH, R. P. (1977). The effect of methemoglobin on the inhibition of cytochrome c oxidase by cyanide, sulfide or azide. Bio-

them. SOKOL,

Pharmacol.

26,2247-2250.

R. R., AND ROHLF, F. J. (198 1). Biometry, 2nd ed., p. 226. Freeman, San Francisco. SPRATT, N. T. (1950). Nutritional requirements of the early chick embryo. III. The metabolic basis of morphogenesis and differentiation as revealed by the use of inhibitors. Biol. Bull. 99, 120- 135. WILCOXON, F. ( 1945). Individual comparisons by ranking methods. Biometrics 1,80-83.