ABSTRACT
The role of the maternal adrenal medulla on frequency of stillbirths from chronic stress administered throughout gestation was investigated in CF/1 mice.
Evidence showing a decreased frequency of stillbirths in adrenal demedullated mice compared to normals is presented.
Mechanisms are discussed whereby the medullary hormones epinephrine and norepinephrine may affect prenatal development during maternal stress.
INTRODUCTION
Apaucity of experimental data exists relating the effect of chronic stress experienced throughout gestation to the incidence of stillbirths. Dornhorst & Young (1952) demonstrated that low, pressor dosages of epinephrine and norepinephrine administered intravenously to gravid rabbits and guinea-pigs produced vigorous uterine contractions with concomitant cyanosis of the placenta and decreases in both foetal heart-rate and blood-pressure. They interpreted their findings as indicating the action of these hormones to be on the placental vasculature and not the blockage of uterine vessels supplying the placenta.
Since the secretion of the adrenal medullary hormones is part of the organism’s stress response, the findings of the Dornhorst & Young investigation provide evidence for a mechanism which might be used to account, in part or whole, for the deleterious effects to prenatal development arising from maternal stress. The present investigation undertook to test this hypothesis by determining the incidence of stillbirths from adrenal demedullated and non-adrenal demedullated mice subjected to stress throughout gestation or allowed a stress-free gestation. It was predicted that adrenal demedullated mice receiving stress throughout gestation would manifest a lower incidence of stillbirths than similarly stressed, gravid, non-adrenal demedullated subjects.
METHODS
Thirty-nine female CF/1 mice served as the subjects. At 82 days of age, each animal began instrumental conditioning training. The rationale of this procedure was to create a situation which would predictably arouse strong ‘anxiety’ in the animal and to train it to reduce this anxiety by making an appropriate response. Stress to the gravid animal would be accomplished by exposing it once again to the anxiety-arousing situation and preventing the animal from making its learned response to reduce the anxiety.
The apparatus consisted of a box 10 in. by 6 in. and 19 in. high divided by a barrier extending in. above the floor. Two red lights extended from the wall above the barrier and one white light was positioned below the two red lights. A buzzer was mounted outside the apparatus beneath the barrier. Both red lights and buzzer were presented together and served as the conditioned or warning stimulus (CS); the white light was on between trials. An electric current of 0·2 mA. was delivered through the grid floor of the apparatus to the animal and served as the unconditioned stimulus (US).
Fifteen CS– US presentations were administered to each animal daily. The CS preceded the US by 5 seconds. Both the CS and US stayed on until the animal crossed the barrier. This response terminated the CS and US. If the barrier-crossing response was made during the 5 seconds CS presentation it was scored as a conditioned avoidance response. The performance criterion was 12 avoidance responses during any single day’s training. At this point training trials were terminated and the subject was assigned to one of three surgical treatment groups.
The surgical treatment was performed at 85 days of age. The animals were divided into adrenal demedullated, sham adrenal demedullated, and nonoperated groups. Surgery was accomplished under sterile conditions. Anaesthesia was chloral hydrate (9·0 mg./kg.) given intraperitoneally. For the adrenal demedullated group, a bilateral adrenalectomy was performed followed by the demedullation of the gland utilizing the technique described by Farris & Griffith (1949). The cortex from each gland was unilaterally transplanted beneath the kidney capsule. Immediately after surgery each animal was given an intramuscular injection of cortisone (4·17 mg./kg.). The sham adrenal demedullated group received the same bilateral cutaneous muscle and kidney capsule incisions as the demedullated subjects. Both surgically treated groups were maintained on a postoperative ad libitum feeding and drinking schedule of Purina laboratory chow and solutions of 1 per cent. NaCl and 5 per cent, dextrose. Tap-water was substituted 2 weeks postoperatively. The non-operated group remained in their individual cages and were on the usual ad libitum feeding and drinking schedule for the mouse colony.
After the surgical treatments and prior to breeding, all animals received additional training on the conditioned avoidance task to strengthen the response. Breeding was then started. The onset of conception was confirmed by the presence of a vaginal plug. After the animal became gravid, it was assigned to either a stress or no-stress treatment.
Stress sessions commenced for animals comprising the stress condition when the vaginal plug was found. Subjects were placed in the conditioned avoidance apparatus three times a day at intervals of 8 hours for a single 5-second presentation of only the CS with the escape route blocked. Thus, stress was defined as the animals emotional response of fear and anxiety when it was prevented from making its previously learned response to the CS. During these same periods the no-stress group animals were given similar handling as the stressed animals. Stress administration and handling were terminated at each animal’s parturition.
RESULTS AND DISCUSSION
Table 1 gives the number of litters, total born, and total stillbirths for the six experimental treatment groups. The stillbirth data is expressed as a percentage of the total number born in each group. Statistical analyses were computed for the difference between percentages (Walker & Lev, 1953) between all groups.
Inspection of Table 1 reveals that the sham demedullated and non-operated stress groups had a significantly higher incidence of stillbirths as compared to all other groups. Furthermore, the adrenal demedullated stress group did not differ significantly for percentage of stillbirths from any of the no-stress groups.
Table 2 presents the litter size and stillbirth frequency for each of the maternal organisms constituting the six experimental treatments. The likelihood that stillbirth frequency was simply a function of litter size may be ruled out. A correlation between number of stillbirths and total litter size for all subjects was not significant. Similarly, an analysis of the differences between the six groups for mean litter size showed no differences.
Since none of the six groups differed for total litter size the experimental manipulations in this study did not appear to produce any marked degree of foetal resorptions. This point was checked further by an inspection of the uteri of all animals for foetal resorption sites. This inspection revealed no differences between the groups. Since this inspection could not be accomplished until after 21 days post-partum, it may not have been reliable. At this time also serial sections of the adrenal cortical transplant sites were prepared. All animals in the demedullated treatment group demonstrated proliferation of the cortex. In several mice of both the stress and no-stress groups the transplant was successful only unilaterally. Histological examinations of the extirpated medullae revealed that the entire gland had been successfully removed.
Another consideration is the role of the medullary hormones in producing hyperglycemia by the conversion of glycogen to glucose in the liver. It is reasonable to assume that a chronic stress might result in prolonged hyperglycemia brought about by the action of the medullary hormones, and this state, in turn, would cause an increase in the amount of circulating insulin. It has been shown that insulin has a marked teratogenic effect on the developing rat foetus when administered into the blood-stream of the maternal organism (Lichtenstein, Guest, & Warkany, 1951; Ross & Spector, 1952). In the present investigation, only two of the stillbirth progeny from the entire study showed gross morphological anomalies. Both manifested anencephaly with one of the animals from the sham-demedullated stress condition and the other from the normal stress treatment.
Although these results do not provide a conclusive test of the proposed mechanisms, the fact that in the absence of the primary source of epinephrine and norepinephrine a significantly smaller proportion of stillbirths occurred following chronic, prenatal stress when compared with non-demedullated subjects is certainly suggestive. If the results of this investigation were in fact the result of a uterine vasoconstrictive effect caused by maternal medullary hormones, the question of both quantity of hormone and duration of action necessary for prenatal damage remains to be answered.
RÉSUMÉ
Fréquence des mort-nés chez des Sauris médullo-surrénalectomisées soumises à un ‘stress’ chronique pendant la gestation
On étudie le role joué par la médullosurrénale maternelle dans la fréquence des mort-nés observés chez les Souris CF/1 soumises à un stress chronique pendant la gestation.
Il est prouvé que la proportion des mort-nés est moindre chez les souris médullo-surrénalectomisées que chez les normales.
On discute les mécanismes par lesquels les hormones surrénales — épi-néphrine et norépinéphrine — peuvent affecter le développement prénatal chez des mères soumises à un stress.
ACKNOWLEDGEMENTS
This research was conducted at Vanderbilt University, Nashville, Tennessee, and supported by a Predoctoral Research Fellowship Grant (MF-10, 246) from the National Institute of Mental Health, United States Public Health Service. Grateful acknowledgement is made to G. W. Meier for his invaluable aid.