It is common to classify a stimulus as stressful if its exposure to an organism ... (the same rock station, volume and tone settings were used). After 30 rain, the ...
Physiology&Behavior,Vol.51, pp. 1157-1163, 1992
0031-9384/92$5.00 + .00 Copyright© 1992PergamonPressLtd.
Printed in the USA.
Effects of Various Stressors on Milk Release in the Rat CHANTAL LAU 1
Department of Pediatrics, Baylor College of Medicine, Houston, TX 77030, and Department of Biology, University of Houston, Houston, TX 77204-5513 Received 30 September 1991 LAU, C. Effectsof variousstressorson milk releasein the rat. PHYSIOL BEHAV 51(6) 1157-1163, 1992.--This study investigated the effect of four stimuli on milk release (MR), namely, sound, nociception, novelty, and restraint. The role of the ensuing adrenocortical response in the suppression of MR was also evaluated. Plasma corticosterone (CORT) levelswere measured at 0 (basal), 15, 30, and 60 min during the suckling sessions to determine whether elevated CORT normally associated with stress could be inhibitory to MR. Compared to nonstressed lactators, dams exposed to sound demonstrated no suppression in MR, but a significant increase in plasma CORT. Pain did not alter milk yield and elevated CORT only at the end of the first hour of exposure. During novelty, MR was suppressed and again CORT was only elevated at the end of the sampling period. Restraint decreased milk yield and increased CORT. During novelty, MR appeared to be regulated by an adrenal factor, which remains to be identified.The peripheral opiates seem to be partially involved during restraint. In conclusion, not all types ofaversive stimuli interfere with MR. Of those which do, different mechanisms seem to be implicated depending upon the nature of the stressor. Furthermore, reduced MR during stress is not a direct consequence of increased CORT. Adrenocortical activation
Lactation
Opiates
THE notion that stress suppresses lactation is based primarily on studies which have shown that stressors such as immobilization, ether stress, auditory, and olfactory stimuli can interfere with milk release (7,9,14,15). It is, however, unclear whether all types of stressors necessarily interfere with milk release (MR). Furthermore, as stress is normally accompanied by elevated levels of adrenocortical steroids, it remains to be determined whether increased circulating glucocorticoids play a role in such suppression. Findlay and Grosvenor (12) have proposed that activation of the peripheral sympatho-adrenal system may be implicated in the suppression of milk yield; this likely occurring via vasoconstriction at the level of the mammary glands and/or increased mammary ductal tone. It has also been shown that threshold sensitivity to nociceptive stimuli may be altered by opiates (31). For instance, pain reactions measured by various nociceptive tests, can be reduced during and immediately following acute immobilization. Such suppression is believed to result from the analgesic properties of increased endogenous opiates during exposure to such stressor. This notion is supported by the observation that treatment with opiate antagonists such as naloxone or naltrexone can reverse the above observation. Thus, the goals of this study focused on the effect of four different types of stressors on MR and on the role of the ensuing adrenocortical responses in its suppression. The protocols were designed so as not to affect milk supply, as the latter will affect the process of MR. The effects of the stressors on the adrenocortical
response of suckling dams was assessed and MR was measured indirectly by the body weight gain of pups. Whenever MR was altered, the role of the peripheral sympatho-adrenal system and that of the opiates were evaluated. Due to the importance and necessity of maintaining mother/young contact in assessing MR, the stressors selected had to meet the following conditions: a) exposure to the stressful stimulus must not drive the dam away from the nest as she will not suckle her young; b) it must not alter the sucking ability of the pups as this would inaccurately reflect changes in MR; and c) humoral stressors should not be used, as the humoral agent may directly affect MR. Thus, the following stimuli were chosen: 1) a neurogenic acoustic stimulus; 2) a nociceptive pain stimulus applied to the dam; 3) exposure of the dam (prior to suckling) to an affective/emotional stimulus, a 30-rain novelty stress; and 4) restraint. The protocols used were approved by the Animal Care Protocol Committee of the University of Houston. METHOD
Animals Primiparous Sprague-Dawley rats (CrI:CD[SD]BR) were received from Charles River Breeding Laboratories (Wilmington, MA) on day 15 of gestation. They were housed individually in opaque polystyrene cages (45 cm × 25 cm × 22 cm) containing wood chips (2-3 cm thick) with food (Rodent Laboratory Chow
zRequests for reprints should be addressed to C. Lau at her present address: Department of Pediatrics, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030.
1157
1158 5001, Ralston Purina Co., St. Louis, MO) and water available ad lib. Animal quarters were air-conditioned (21 _+ 1°C) and set to a 12-h light/12-h dark cycle with lights on at 0600 h. On the expected date of parturition, cages were checked every, 2 h for the presence of pups. The birth date was designated day 0. Litters were culled to eight pups within 48 h of birth. It is common to classify a stimulus as stressful if its exposure to an organism leads to an increase in circulating levels of adrenocortical steroids. However, inasmuch as studies have suggested that dams show a dampened response to stress when compared to nonlactators (33,36), in the present situation, the criterion used to characterize the stressful nature of the various stimuli was based on their ability to induce elevated levels of circulating corticosterone in at least nonlactating rats, if not in both lactating and nonlactating animals. As virgin rats may exhibit different physiological responses than do rats which have had litters, only nonlactators which had had at least one pregnancy were used. Furthermore, these animals were studied only during the diestrous stage as hormonal patterns are affected by the estrous cycle (22,37). At the end of the experiments, the animals were sacrificed by decapitation or a lethal dose of sodium pentobarbital 1V (Nembutal, Abbott Labs, Chicago, IL).
Eivperimental Procedures A protocol similar to that of earlier studies was used (23,24). Briefly, MR was determined on day 13/14 postpartum as follows. Dams were separated from their young at 0800 h for 5 h in order to allow accumulation of the same "milk supply" and a concurrent 5-h fast of the young. During the separation period, the pups were kept at nest temperature in styrofoam buckets (3335°C). A 2-h refeeding period began at 1300 h. Milk intake was calculated as the difference between the hourly weights of the pups and those at the start of the refeeding session. Inasmuch as rat pups do not urinate on their own and the darns do not consistently empty all the pups' bladders, urination was induced by stroking the ano-genital area with a tissue prior to each weighing. Blood samples were collected through an atrial cannula implanted 24 h prior to the experimental day. These procedures have been described in an earlier study (23). Plasma CORT was measured at time 0 (just before the return of the pups and/or induction of the stressors) and 15, 30, and 60 minutes thereafter. CORT determination was measured by competitive proteinbinding assay (19). Intraassay variability was 2.3% and recovery of a spiked sample between assays averaged 92%. The following stressors were used. The neurogenie sound stimulus was a 5000 Hz tone (90 dB, background 70 dB) OCCuTTing 10 S on/10 s off for a 10-rain period followed by a 10rain silence. The speakers were placed 30 cm above the cage. This was presented to the clams at the return of the young for the duration of the refeeding period. The nociceptive stimulus consisted of clipping 3 mm off the tip of the tail of the dams immediately following the return of pups. The affective/emotional stress, a 30-rain novelty, included the following: 30 rain prior to testing, the rats were transferred to a clean cage and moved into the experimental room where a radio placed 30 cm above the cage played for the duration of the refeeding period (the same rock station, volume and tone settings were used). After 30 rain, the dams were returned to their respective home cage and reunited with pups for a 2-h refeeding period. Restraint consisted of the following procedure: to induce pups to attach readily after restraining the lactator, the young were returned to the unrestrained dams for l 0 rain prior to immobilization. This step was necessary as the nipples were recessed following the
I_A[! lack of stimulation during the 5-h isolation and not all the young succeeded in attaching without maternal help. The 10-rain reunion allowed the pups to settle down and attach to the nipples, but was not sufficient to allow for the occurrence of a milk release as no characteristic stretch response of pups was observed (10). The dams were then immobilized with their [bur limbs and tail securely taped to a board (25 cm ;x 10 cm) in a supine position. An additional tape was placed around the ventrum without obstructing the nipples to help support their back. The board had an opening at the level of the neck of the animal such that the head could be further restrained from downward movements by means of a metal plate taped firmly on the board under the chin. This was necessary in order to prevent the dam from reaching down and licking her young. The animals were then turned over on their ventrum. They rested on woodchip bedding with the area underneath the ventrum slightl~ cleared such that the pups could freely attach to the nipples, Immobilization of the dam from the time the animal was removed r-ore her cage was performed within 3 min. It is after this procedure that the initial body weight of the pups was taken (i,e., prefeed weight). Control nonstressed lactators for sound, pain, and novelty were subjected to the 5-h isolation, but were left undisturbed during the refeeding session. Sham-restrained lactators were immobilized for 3 rain and then released into their cage for the retkeding sessions. Nonlactators were subjected to the same stressors for an equivalent period at the same time of the day. They were treated in the absence of pups inasmuch as tinder normal conditions, they were housed singly in individual cages. It is recognized that the time spent by dams in the nest are not necessarily correlated to milk release. In addition. Stern and Johnson (34.35) have recently described the complex interaction ofperioral somatosensory determinants implicated in the suckling performance of dams. However, as milk release was indirectly measured by the weight gain of the pups and as it was evident that milk intake by the pups would be reduced if the dams spent more time out of the nest, the percent time spent by the lactators away from the young was recorded to insure that any decrease in MR during exposure to the various stimuli was not the result of a change in such maternal behavior. This parameter was not assessed during the restraint experiment as it was not applicable. Such monitoring confirmed that none of the experimental dams spent more time away from their young than nonstressed control counterparts. In order to maintain proper lactation (30), adrenalectomy with CORT replacement was performed as described in an earlier study (23). Briefly, bilateral adrenalectomy and CORT replacement consisting of a 100 mg pellet (50% cholesterol, 50% CORT) implanted subcutaneously in the nape of the neck were performed on days 2-3 postpartum under ether anesthesia. Drinking water was replaced with normal saline. Proper glucocorticoid replacement was confirmed as the litters from adrenalectomized CORT dams demonstrated normal weight gain for the duration of the study. At the end of each experiment, the animals were decapitated within 1 min of opening the top of the cage. Trunk blood was collected for determination of plasma CORT. Diarrhea and chromodacryorrhea (CDR) were monitored in all the animals subjected to restraint. CDR is commonly observed in rats undergoing severe stress or withdrawal symptoms (3). It is characterized by the secretion of a red exudate around the eyes. Morphine sulfate (NIDA Drug Supply Program; 0.5 mg/ kg IV), dissolved in 0.15 ml normal saline, was administered to restrained animals immediately following blood sampling at time 0 at the time when flushing of the cannula with saline was done. Naloxone HCI, generously provided by Dupont de Nemours & Co. (Wilmington, DE), was administered in 0.15 ml saline at a
STRESS AND MILK RELEASE
1159
dose of 2 mg/kg, IV, at the beginning of the feed, similarly to the morphine administration and at 5 mg/kg at the beginning of the second hour, at the time when the 60-min blood sample was collected. The time and procedure at which these drugs were administered allowed the use of the respective nontreated counterparts to be used as injection (vehicle) controls.
A. 8O
=e 60
Statistical Analyses
o
In all studies, the dam was considered the significant variable (1). Therefore, individual weight gains of all pups sucking on a particular dam were averaged to give a single value (g/pup) for that dam. The means + SEM were then calculated for all lactators in each experimental group. Body weight change of the pups are presented as increments from time 0. Two-way repeated measures ANOVA was used to assess the effect of treatment (stress/no stress) by time ( 1 and 2 h) on the body weight gain of the pups. However, for novelty, treatment comparisons included nonstressed lactators and both intact and adrenalectomized dams exposed to that stimulus. In the restraint study, treatment defined not only the animals exposed to the presence/absence of the stressor, hut also the adrenalectomized and drug-treated restrained lactators. An independent t-test was used to assess significance of time during the first hour of refeed. CORT levels were analyzed also by a two-way repeated measures ANOVA between treatment and time. Treatment encompassed the nonstressed and stressed lactators and stressed nonlactators. However, in the restraint study, naloxone-treated sham and restrained animals were also included in the analysis. When no significant interaction effect was noted, and treatment effect was significant, comparisons between specific treatment groups were obtained using the Bonferroni multiple comparison adjusted t-test. When the interaction effect was significant, comparisons between specific treatment groups used the approximated t-test (28).
4o 0
== 20
E
0
0.00
,
0.25
=
Time
~
,
0.50
0.75
i
1.00
(h)
so
.Q ~ 4o U
RESULTS
As shown in Fig. l, MR as reflected by the body weight gain of the pups was similar in both sound-stressed and nonstressed groups during the 2-h refeeding session, with a significant increase occurring during the first hour of refeed in both groups (p < 0.01). Figure 2A shows the plasma CORT responses to sound.
~ IO
E
20 i s ~"
0 o.O0
i 0.25
i
=
,
0,50
0.75
1.00
Time 2
='. _=o
........ z 1
z= g. mO
0,~ 0.00
I 1.00
Time
I 2.00
(h)
FIG. I. Hourly body weight gain per pup (means + SEM) of 5-h fasted pups nursing 5-h isolated nonstressed lactators (i, n = 5) and dams exposed to the sound (I, n = 5) and the pain stimuli (0, n = 5),
(h)
FIG. 2. (A) Changes in plasma corticosterone concentrations (means _+ SEM) in nonlactators (El, n = 5) and lactators (i, n = 5) subjected to the sound stress, and nonstressed lactators ( t n = 5). (B) Changes in plasma corticosterone concentrations(means _+SEM) from nonlactators (q, n = 5) and lactators (0, n = 5) subjected to the pain stress, and nonstressed lactators (m, n = 5). There was a significant interaction effect, F(2, 36) --- 3.88, p < 0.01. CORT levels were significantly greater in the dams subjected to the sound stimulus than nonstressed counterparts at all times following the beginning of the feed (p < 0.05). However, this response was significantly less than that observed in their nonlactating counterparts (p < 0.001). When compared to the nonstressed group, exposure to the pain stimulus did not affect the body weight gain of the pups (Fig. l). There was again a significant weight increase after the first hour (p < 0.01). A significant interaction effect, F(2, 36) = 3.7 l, p < 0.0 l, was noted in plasma CORT between pain-exposed and nonstressed animals (Fig. 2B). The levels of stressed dams were similar to those of controls at 15 and 30 min, but were significantly higher at 60 min (p < 0.001). In turn, nonlactators
1160
I,AU
A ",,,i
60
v Q C 0
E
"--_ T
I,.
.j.j.J
40
O mC
0 0
...........o,.-'""''"'t
0
O
20
m
E ID O 0
0.00
I
I
1.00
2.00
Time
r,,
0 0.00
I
I
I
I
0.25
0.50
0.75
1.00
Time
(h)
FIG. 3. Hourly body weight change per pup (means _+SEM) of 5-h fasted pups nursing 5-h isolated nonstressed lactators (U, n = 5), intact (A, n = 6) and adrenalectomized/CORT-replaced([], n = 6) dams exposed to the novelty stress.
subjected to the pain stimulus demonstrated a marked adrenocortical activation (Fig. 2B) when compared to their lactating counterparts (p < 0.001) at all times. Figure 3 shows the effect of novelty, the affective/emotional stimulus, on the body weight of pups. The interaction effect was not significant, but treatment effect was, F(2, 39) = 17.50, p < 0.001. Further analyses showed a significantly lower body weight gain of pups from novelty-exposed dams (p < 0.05) at both time points, indicating that MR was suppressed. To determine whether an adrenal factor could be implicated in the reduction of MR during novelty, dams subjected to bilateral adrenalectomy/CORT replacement were exposed to that stimulus and MR was assessed. Following removal of the adrenal glands, milk yield was restored to that of their nonstressed counterparts; no difference was noted between nonstressed dams and noveltyexposed adrenalectomized counterparts (Fig. 3). Pups' body weight increment was significant after the first hour of refeed in both groups of novelty-exposed lactators (p < 0.01 ). Circulating CORT were compared between nonstressed dams and noveltyexposed lactators and nonlactators (Fig. 4). No significant interaction effect was observed. The novel environment did not elevate the CORT levels of dams when compared with that of nonstressed counterparts at 15 and 30 min, but did at 60 min (p < 0.05). Circulating CORT levels in nonlactators exposed to this stressor were significantly higher (p < 0.01) than those of their lactating counterparts at all times (Fig. 4). CORT replacement in the adrenalectomized dams maintained plasma levels at 9.46 _+ 0.56 ug/dl. The body weight gain of pups from restrained lactators are presented in Fig. 5. The interaction effect (treatment vs. time) was not significant, but treatment effect was, F(5, 31) = 20.78, p < 0.001. Pups from restrained dams ingested significantly less milk than those from sham counterparts (p < 0.01) during the 2 h of refeed. In order to determine whether such inhibition in MR could also be a consequence of an adrenal component as observed during exposure to novelty, dams subjected to bilateral adrenalectomy plus CORT replacement were exposed to restraint. Under these conditions, the body weight gains of the young were further suppressed (p < 0.01) when compared to that of intact restrained counterparts. Immobilization stimulates
(h)
FIG. 4. Changes in plasma corticosteroneconcentrations(means _+SEM) from nonlactators (A, n = 5) and lactators (A, n = 6) subjected to the novelty stress, and nonstressed lactators (U, n = 5).
the release of opiates inasmuch as restraint-induced analgesia has been shown to be opiate mediated (31). In addition, the adrenal medulla is a primary source of peripheral opiates (26,39). Therefore, to investigate whether opiates are implicated in the observed suppression of MR during immobilization, morphine sulfate was administered to adrenalectomized/CORT-replaced dams subjected to restraint. Morphine treatment, at the dose given, brought back the pups' body weight gain to that of intact restrained intact counterparts. To substantiate this opiate-mediated effect on MR, intact restrained lactators, in turn, were treated with the opiate antagonist, naloxone. Such treatment significantly suppressed the weight gain of the pups compared to that of intact restrained counterparts (p < 0.01). The degree of inhibition was actually similar to that observed in the adrenalectomized/CORT restrained group. To verify that such
A
Q.
1.5
1.0 o
I / ..... m
......................... ........
0.00
,00
Time
I. ,00
(h)
FIG. 5. Hourly body weightchange per pup (means _+SEM) of 5-h fasted pups nursing 5-h isolated dams subjected to the followingtreatments: sham-restraint (n, n = 5), restraint (O, n = 6), restraint plus adrenalectomy/CORT replacement (O, n = 6), sham-restraint plus naloxone (1~, n = 8), restraint plus adrenalectomy/CORT-replacementplus morphine sulfate (~, n = 6), restraint plus naloxone (O, n = 6).
STRESS AND MILK RELEASE suppression was not the direct result of naloxone on MR, shamrestrained intact lactators were similarly treated with naloxone. Under these conditions, weight gain of the young from naloxonetreated animals were similar to those from controls at 1 h, but were decreased at 2 h (p < 0.01 ). The decrease at the latter time, nevertheless, was significantly greater than any of the restrained groups (p < 0.01). With the exception of the pups from the restrained adrenalectomized dams, those from all the other groups demonstrated a significant increase in weight gain by the first hour of refeed (p < 0.05). Figure 6 shows the effect of immobilization on plasma CORT during restraint. There was a significant interaction effect, F(4, 57) = 2.56, p < 0.01. At all times, restrained lactators showed a significant increase over sham counterparts and restrained nonlactators (p < 0.001). The latter two groups showed similar levels. When compared to naloxone-treated counterparts, restrained dams showed higher hormonal levels at all times (p < 0.01) except at 30 min. The levels of sham controls were significantly lower than those of their naloxone-treated counterparts at 15, 30, and 60 min (p < 0.001). Following adrenalectomy/ CORT replacement, the hormonal levels of the restrained dams averaged 23.02 _+ 2.51 t~g/dl. Diarrhea and CDR were observed in all restrained dams (six/ six). After naloxone treatment, five of six and six of six immobilized lactators still demonstrated diarrhea and CDR, respectively. In restrained adrenalectomized/CORT lactators, five of six demonstrated diarrhea and CDR. However, following morphine treatment of restrained adrenalectomized/CORT replaced dams, diarrhea was observed in only one of six animals and CDR in four out of six dams. In nonlactators, restraint only induced two of eight animals to have diarrhea, and none exhibited CDR. DISCUSSION The first goal of the present study was to investigate the effect on MR of four types of stressors which are commonly used in studies on the effect of stress on rats (i 1,29,31,38). Because lactators exhibit a suppressed glucocorticoid response to various stressors (20,21,33,36), a stimulus was classified as stressful if it elevated plasma CORT at least in nonlactators if not in both categories of animals. The observations that CORT levels were greater in nonlactators than lactating counterparts following sound, pain and novelty and in lactators following restraint assured that this criterion was met. Sound and pain did not affect MR as measured by body weight gain of the pups. On the other hand, circulating levels of CORT was elevated following the former treatment at all sampiing times, but only at the end of the exposure following the latter. It is possible that the stimuli needed to be more intense for MR suppression or that the nature of these stressors does not interfere with MR. The observation made during exposure to sound is contradictory to that of Grosvenor and Mena (14) who noted a decrease in milk ingestion by the pups during exposure to a different type of acoustic stimulus. It is difficult to compare the results from the two studies inasmuch as the auditory stimulus used by these authors consisted of sounds made by personnel during routine tasks associated with animal care and their ongoing study. As such, the nature of their stressor may have included more than just an acoustic stimulus. Novelty, which affected CORT levels only at the end of the exposure, significantly suppressed MR. The observation that the latter was reinstated following adrenalectomy/CORT replacement supports the involvement of an adrenal component(s) which is inhibitory to MR during exposure to novelty. It is con-
1161
60
/
o
40 ~"
. . . . . . . . . . . . . .
~.
,"//
/ g~
,".-~",,e,,''~-~-'-"
20
g ~-
o
0.00
0.25
0.50 Time
0.75
1.00
(h)
FIG. 6. Changes in plasma corticostemne concentrations (means _+SEM) from restrained nonlactators (r"3,n = 5), restrained lactators (O, n = 5), sham-restrained lactators (I, n = 5) naloxone-treated restrained lactators (0, n = 6), and sham-restrained dams treated with naloxone ([], n = 5).
ceivable that sympatho-adrenomedullary participation was implicated. Indeed, it has been proposed that such activation may lead to vasoconstriction, reducing delivery of oxytocin to the mammary gland and/or increased ductal tone at the mammary level, decreasing milk flow out of the mammary ducts (12). Further studies are necessary to identify the nature of the adrenal component(s) implicated in the suppression of MR during exposure to novelty. During restraint, the partial maintenance of MR, which was abolished in adrenalectomized/CORT animals, also suggests the involvement of an adrenal component(s). However, contrary to the situation observed during novelty, the presence of such factor(s) would play a permissive rather than inhibitory role in MR. With the knowledge that peripheral opiates are involved in stressinduced analgesia (25,26,31,38), it was hypothesized that adrenal opiates were implicated in the partial maintenance of MR observed during restraint. This is supported by the observations that adrenalectomy/CORT replacement plus morphine sulfate reinstated MR to the level of intact restrained animals and that restrained dams treated with naloxone demonstrated a suppressed MR similar to that observed in adrenalectomized/ CORT-replaced dams under restraint. The observation that naloxone did not affect MR in sham-restrained counterparts verified that the suppressed MR in these restrained dams did not result from a direct effect of the drug on MR. Thus, the present data support the involvement of peripheral opiates in the maintenance of MR during immobilization. Further studies, however, are necessary to understand the exact mechanism(s) by which such action takes place. Comparison of the milk intake obtained by the pups of sham-restrained dams with that of restrained counterparts may suggest that opiates are only partially implicated in the maintenance of MR. It is conceivable that full restoration of MR could have been obtained if a higher dose of morphine had been used. However, this should be examined with care because the normal expression of maternal behavior (e.g., stimulation of the pups) during suckling was prevented in the immobilized animals. Maternal stimulation and the integrity of mother-pop interaction are important for optimal milk intake by the young (32). The difference in milk yield between restrained and sham-restrained dams may be a consequence of the absence ofperioral somatosensory stimulation experienced by these lac-
1162
I.A [.
tators, inasmuch as the immobilization prevented the animals from licking their young. Indeed, Stern and Johnson (34) have discussed the importance of adequate perioral stimulation in establishing appropriate nursing behavior and milk release in rats. During restraint, the high plasma C O R T observed in the restrained lactators showed that dams can demonstrate greater response to stress than nonlactators. Interesting questions are raised by the differential effects of naloxone on the hormonal responses induced by this stimulus. However, this will not be discussed in the present report as it is being investigated further in our laboratory. Intact dams under restraint exhibited diarrhea and CDR, whereas morphine treatment of adrenalectomized/CORT-replaced dams under the same condition prevented these symptoms in the majority of the animals. This is contrary to the observations made in nonlactators which showed reduced occurrence of diarrhea and no CDR. These latter observation supports studies by other investigators who have demonstrated that immobilization stimulates opiate release in nonlactators (31). In the present situation, it appears that if opiates were released during suckling/lactation, they were suppressed when the dam was exposed to a stressor such as restraint. Stressors may thus have different effects depending upon the state of the organism, i.e., pregnant, lactating/suckling or not. Precedence supporting such notion has already been presented. For instance, during pregnancy, an opiate-induced elevation in pain threshold has been observed (2,13). Lactators demonstrate decreased fearfulness, as measured by the freezing response in rats and increased aggression towards intruders, when compared to virgins ( 17,18). Numerous studies have shown that opiates are inhibitory to oxytocin release with a direct effect at the level of the neurohypophysis (4-6,16,27). The exact mechanism of such action. however, remains to be elucidated. The suppression of MR during restraint, in the present study, corroborates the observations made by Clarke and Wright (8). These authors investigated the potency of opiates in the suppression of oxytocin relative to their effects as analgesics in lactating rats and suggested that oxytocin secretion is reduced during opiate-induced analgesia.
~Fhe second aim of this study was to investigate whether increased glucocorticoids, normally associated with stress, play a role in regulating MR. From the results obtained, it becomes apparent that elevated glucocorticoids per se resulting from the experimental procedures need not be present for suppression of MR to occur. Indeed, a normal response was seen with moderate (pain) and marked (sound) adrenocortical activation. Similarly, a decreased milk ingestion by the pups was observed also with moderate (novelty) and marked elevation in plasma C O R T (restraint). Finally, the difference noted in the hormonal levels in adrenalectomized/CORT rats subjected to restraint versus novelty needs clarification. It is conceivable that the volume of distribution of C O R T in the organism was differentially altered in response to the different types of stressors. Further studies will be necessary to answer this question. The likelihood that this disparity resulted from the manuthcture of the C O R T pellets is doubtful inasmuch as the C O R T levels obtained within each group cannot explain the 2.4-fold difference seen between groups. In summary, the following observations can be made from the present work. First, not all types of aversive stimuli interfere with MR: sound and pain, as used here, had no effect, whereas novelty and restraint were suppressive. It appears that during exposure to novelty an adrenal component(s), which needs to be identified, is involved. During restraint, there is evidence that the peripheral opiates are implicated. Second, suppression of MR during stress cannot be attributed to a direct effect of elevated circulating glucocorticoids. ACKNOWLEDGEMENTS The author wishes to express her gratitude to S. Henning for the generous use of her laboratory,, S. Henning and L. Leeper for their critical review of this work, L. Ferlic for statistical consultation and M. Stevens for secretarial assistance. This study was supported in part by grant HD 19639 from the National Institutes of Health, by #S07 RR 07147-17 from the University of Houston. and by the Department of Pediatrics Fund, Baylor College of Medicine.
REFERENCES 1. Abbey, H.; Howard, E. Statistical procedure in developmental studies on species with multiple offspring. Dev. Psychobiol. 6:329-335: 1973. 2. Baron, S. A.; Gintzler, A. R. Effects of hypophysectomy and dexamethasone treatment on plasma fl-endorphin and pain threshold during pregnancy. Brain Res. 418:138-145; 1987. 3. Bell, J.; Adler, M. W. Comparison of peripheral and central administration of naloxone in precipitating abstinence in morphine-dependent rats. Drug Alcohol Depend. 21:189-194; 1988. 4. Bicknell, R. J.; Chapman, C.; Leng, G. Neurohypophysial opioids and oxytocin secretion: Source of inhibitory opioids. Exp. Brain Res. 60:192-196: 1985. 5. Bicknell, R. J.; Leng, G.; Lincoln, D. W.; Russell, J. A. Naloxone excites oxytocin neurones in the supraoptic nucleus of lactating rats after chronic morphine treatment. J. Physiol. (Lond.) 396:297-318: 1988. 6. Bondy, C. A.; Gainer, H.; Russell, J. T. Dynorphin A inhibits and naloxone increases the electrically stimulated release ofoxytocin but not vasopressin from the terminals of the neural lobe. Endocrinology 122:1321-1327; 1988. 7. Chaudhury, R. R.; Chaudhury, M. R.; Lu, F. C. Stress-induced block of milk release. Br. J. Pharmacol. 17:305-309:1961. 8. Clarke, G.; Wright, D. M. A comparison of analgesia and suppression of oxytocin release by opiates. Br. J. Pharmacol. 83:799-806; 1984. 9. Cross, B. A. Neurohormonal mechanisms in emotional inhibition of milk release. J. Endocrinol. 12:29-37; 1955.
10. Deis, R. P. The effect of an exteroceptive stimulus on milk release in lactating rats. J. Physiol. (Lond.) 197:37-46; 1968. I 1. Feldman, S. Neural pathways mediating adrenocortical responses. Fed. Proc. 44:169-175: 1985. 12. Findlay, A. L. R.: Grosvenor, C. E. The role of mammary gland innervation in the control of the motor apparatus of the mammary gland: A review. Dairy Sci. Abstr. 31:109-116; 1969. 13. Gintzler, A. R. Endorphin mediated increases in pain threshold during pregnancy. Science 210:193-195; 1980. 14. Grosvenor, C. E.; Mena, F. Effect of auditory, olfactory and optic stimuli upon milk release and suckling-induced release of prolactin in lactating rats. Endocrinology 80:840-848; 1967. 15. Grosvenor, C. E.; Mena, F. Regulating mechanisms for oxytocin and prolactin secretion during lactation. In: Muller, E. E.; MacLeod, R. M., eds. Neuroendocrine perspectives. Amsterdam: Elsevier/North Holland Biomedical Press; 1982:69-110. 16, Haldar, J.; Hoffman, D. L.; Zimmerman, E. A. Morphine, fl-endorphin and [D-AIa2] met-enkephalin inhibit oxytocin release by acetylcholine and suckling. Peptides 3:663-668; 1982, 17. Hansen, S.; Ferreira, A. Food intake, aggression, and fear behavior in the mother rat: Control by neural systems concerned with milk release and maternal behavior. Behav. Neurosci. 100:64-70; 1986. 18. Hard, E.; Hansen, S. Reduced fearfulness in the lactating rat. Physiol. Behav. 35:641-644; 1985. 19. Henning, S. J. A sensitive and convenient method for measurement of corticosterone in rat serum. Steroids 35:673-683: 1980.
STRESS A N D M I L K RELEASE
20. Higuchi, T.; Honda, K.; Takano, S.; Negoro, H. Reduced oxytocin response to osmotic stimulus and immobilization stress in lactating rats. J. Endocrinol. 116:225-230; 1988. 21. Higuchi, T.; Negoro, H.; Arita, J. Reduced responses of prolactin and catecholamine to stress in the lactating rat. J. Endocrinol. 122: 495-498; 1989. 22. Kamoun, A. Activit6 cortico-surrrnale au cours de la gestation, de la lactation et du drveloppement pr6 et post-natal chez le rat. I. Concentration et cinrtique de disparition de la corticosterone. J. Physiol. (Paris) 62:5-32; 1970. 23. Lau, C. Activation of the adrenal cortex or the peripheral sympathoadrenomedullary system does not necessarily influence milk release in the rat. J. Endocrinol. 118:399-405; 1988. 24. Lau, C.; Henning, S. J. Regulation of milk ingestion in the infant rat. Physiol. Behav. 33:809-815; 1984. 25. Lewis, J. W.; Cannon, J. T.; Liebeskind, J. C. Opioid and nonopioid mechanism of stress analgesia. Science 208:623-625; 1980. 26. Lewis, J. W.; Tordoff, M. G.; Sherman, J. E.; Liebeskind, J. C. Adrenal medullary enkephalin-like peptides may mediate opioid stress analgesia. Science 217:557-559; 1982. 27. Lincoln, D. W.; Clarke, G.; Merrick, L. An opiate-mediated inhibition ofpeptide release from neurosecretory terminals. In: Farmer, D. S.; Lederis, K., eds. Neurosecretion, molecules, cells, systems. New York: Plenum Press; 1980:269-279. 28. Milliken, G. A.; Johnson, D. E. Analysis of messy data. New York: Van Nostrand Reinhold Co.; 1984. 29. Muir, J. L.; Pfister, H. P. Corticosterone and prolactin responses to predictable and unpredictable novelty stress in rats. Physiol. Behav. 37:285-288; 1986.
1163 30. Neville, M. C.; Berga, S. E. Cellular and molecular aspects of the hormonal control of mammary function. In: Neville, M. C.; Neifest, M. R., eds. Lactation, physiology, nutrition and breast-feeding. New York: Plenum Press; 1983:141-177. 31. Porro, C. A.; Cadi, G. Immobilization and restraint effects on pain reactions in animals. Pain 32:289-308; 1988. 32. Rosenblatt, J. S. The basis of synchrony in the behavioral interaction between the mother and her offspring in the laboratory rat. In: Foss, B. M., ed. Determinants of infant behavior, vol. 3. New York: Wiley & Sons; 1965:3--45. 33. Stem, J. M.; Goldman, L.; Levine, S. Pituitary-adrenal responsiveness during lactation in rat. Neuroendocrinoiogy 12:179-191; 1973. 34. Stem, J. M.; Johnson, S. K. Perioral somatosensory determinants of nursing behavior in Norway rats (Rattus norvegicus). J. Comp. Psychol. 103:269-280; 1989. 35. Stern, J. M.; Johnson, S. K. Ventral somatosensory determinants of nursing behavior in Norway rats. 1. Effects of variations in the quality and quantity of pups stimuli. Physiol. Behav. 47:993-1011; 1990. 36. Thoman, E. G.; Conner, R. L.; Levine, S. Lactation suppresses adrenal corticosteroid activity and aggressiveness in rats. J. Comp. Physiol. Psychol. 70:364-369; 1970. 37. Viau, V.; Meaney, M. J. Variation in the hypothalamic-pituitaryadrenal response to stress during the estrous cycle in the rat. Endocrinology 129:2503-2511; 1991. 38. Watkins, L. R.; Mayer, D. J. Organization of endogenous opiate and nonopiate pain control systems. Science 216:1185-1192; 1982. 39. Yang, H. Y. T.; Hexum, T.; Costa, E. Opioid peptides in adrenal gland. Life Sci. 27:1119-1125; 1980.
