Stress-induced disruption of parturition in the rat may bemediated by endogenous opioidsG. Leng, S. Mansfield, R. J. Bicknell, D. Brown, C. Chapman, S. Hollingsworth,C. D. Ingram, M. I. C. Marsh, J. O. Yates and R. G. DyerDepartment of Neuroendocrinology, AFRC Institute of Animal Physiology and Genetics Research,

Babraham, Cambridge cb2 4at

received 5 December 1986

ABSTRACT

Plasma samples were obtained before and duringparturition from conscious rats implanted chronicallywith a jugular cannula. Rats were either allowed toremain in their nesting cage throughout parturition,or were moved immediately following the birth of thesecond or third pup into an empty glass chamber. Thetime-course of parturition was prolonged for motherrats which were moved in mid-parturition to thisunfamiliar environment. However, in rats given an i.v.

injection of the opioid antagonist naloxone at the timeof transfer, parturition followed a normal time\x=req-\course, and plasma oxytocin levels were significantlyhigher than in animals injected with saline. Thus our

hypothesis is that stress activates opioid pathwayswhich delay parturition by inhibiting oxytocinrelease. Opioid-mediated mechanisms may similarly beresponsible for some problems in human parturition.J. Endocr. (1987) 114, 247\p=n-\252

INTRODUCTION

On the afternoon of day 21 of gestation, pregnant ratsusually give birth to between six and fourteen pupsover about an hour. During this time oxytocin isreleased from the neurohypophysis to stimulate con¬tractions of the uterus, which at this time has a veryhigh density of oxytocin receptors (Fuchs, 1969;Fuchs & Poblete, 1970; Summerlee, 1981). The nor¬mal course of parturition can, however, be severelydisrupted by environmental stress. In mice (Newton,Foshee & Newton, 1966; Newton, Peeler & Newton,1968) and in rats (Leng, Mansfield, Bicknell et al.1985), moving the mother from her nesting cage to an

empty glass chamber in mid-parturition can greatlydelay the subsequent delivery of young.

The mechanisms of this stress-induced disruption ofparturition are not known. One possibility is that thisdisruption is mediated, at least in part, by the actionsof endogenous opioids. Intravenous or intraventricu¬lar administration of morphine certainly results ina dramatic prolongation of parturition in rats, accom¬

panied by reduced levels of circulating oxytocin(Cutting, Fitzsimons, Gosden et al. 1985; Gosden,Humphreys, Johnston et al. 1985). This disruption ofparturition in rats injected with morphine can be

reversed by infusions ofoxytocin (Cutting et al. 1985).Several opioid species are present within the hypo-thalamo-neurohypophysial system (for review see

Bicknell, 1985). When released they suppress thesecretion of oxytocin (Clarke, Wood, Merrick &Lincoln, 1979; Bicknell & Leng, 1982), and there isevidence for an influence of endogenous opioids onnormal parturition in the rat (Leng et al. 1985; Jones& Summerlee, 1986). The present study sought evi¬dence for a physiological role of these endogenousopioids in parturition, and in particular during stress-induced disruption of parturition. Some of the resultshave been published previously in abstract form(Leng, Bicknell, Mansfield & Dyer, 1986).

MATERIALS AND METHODS

Virgin Wistar rats were housed overnight with sex¬

ually active males, and the following morning vaginalsmears were taken to detect the presence of sperm.The morning on which sperm was detected was desig¬nated day 0 of gestation. Early in the morning of day21 of gestation, rats were anaesthetized briefly withAvertin (no longer commercially available; Avertin isa mixture of tribromethanol and amylene hydrate: for

full details see Dyer, Weick, Mansfield & Corbet,1981) to allow a cannula to be placed in the jugularvein. The rats were watched continuously thereafterand the time of delivery of every pup was noted.

A pilot study was performed, in which no bloodsamples were taken, involving two groups of rats.Immediately after the birth of the second pup, the ratswere transferred from their cage into a glass obser¬vation chamber (disturbed rats), and were given ani.v. injection of either the opioid antagonist naloxone(Du Pont U.K. Ltd; 2 mg naloxone hydrochloride/kgbody weight) or an equivalent volume of saline.The interval between the birth of the first andsecond pups was found to be very variable in theseexperiments, and so for the main study involvingblood sampling the experimental protocol waschanged and rats were moved after the birth of thethird pup.

Two groups were moved after the birth of the thirdpup, and were given naloxone or saline at this time.Two further groups of rats remained in their cages(undisturbed rats) but again were given injections ofeither naloxone or saline following the birth of thethird pup. The glass chamber was devoid of any bed¬ding, food or water, and the rats remained in thisenvironment for 90 min, whereupon they werereturned to their cages and provided with nest-building material. Up to eight blood samples (0-3 ml)were collected from each rat. Each sample wasimmediately replaced with the same volume ofhepari¬nized, isotonic saline. Samples were taken before theonset of parturition, at the time of birth of pups,between the birth of pups (at 10-min intervals after thepreceding birth) and at the end of parturition, andwere immediately placed on ice before centrifugation.Plasma was stored at

—

20 °C until radioimmuno¬assay for oxytocin using an antiserum of which thespecificity and characteristics have been describedpreviously (Higuchi, Honda, Fukuoka et al. 1985). Inour studies the antiserum yielded assays of meansensitivity 0-45 pg (determined from twice the s.d.about the zero-binding points). Mean intra- and inter¬assay coefficients of variation were 10-2 and 7-3%respectively.

Four groups of female litter-mates of the rats usedin the above study were similarly implanted withjugular catheters. These rats had been mated at thesame time as the pregnant rats, but had failed tobecome pregnant. They were cannulated at the sametime as the pregnant rats, but their ovarian status wasnot determined. Eight blood samples were taken fromeach non-pregnant rat at 10-min intervals; followingthe third sample, rats were either moved to a glasschamber or allowed to remain in their cage, andwere given an i.v. injection of either naloxone orsaline.

RESULTS

Effects of disturbance on the time-course of parturitionIn the initial pilot study, rats were transferred fromtheir nesting cage to the glass chamber immediatelyfollowing the birth ofthe second pup. In the following40 min (approximately one half-life for naloxone inblood: Spector, Ngai, Hempstead & Berkowitz, 1978),no pups were delivered by three out of eight saline-treated rats, whereas at least three pups were deliveredby each of seven naloxone-treated rats. For sub¬sequent experiments involving blood sampling, ratswere transferred immediately after the birth of thethird pup. This change in protocol enabled a betterestimate of the birth rate in each rat before disturb¬ance, and allowed more blood samples to be takenbefore drug injection. Transferring the rats from theircage to the observation chamber after the birth of thethird pup again resulted in a marked disruption ofparturition. The mean intervals between pups threeand four, four and five, and five and six were all longerin the saline-treated disturbed rats than the corres¬ponding intervals in the undisturbed rats (Fig. 1). Inthe 40 min immediately following transfer, three ormore pups were delivered by nine out of 14 naloxone-treated rats, but by only three out of 15 saline-treatedrats. Two rats in the former group and seven in thelatter gave birth to one or no pups within 40 min. Theinterval between pups three and four was longer thanthat between pups two and three in the saline-treateddisturbed rats (mean difference 9-4 + 5-8 (s.e.m.) min, = 14, omitting one rat for which the interval betweenpups three and four exceeded 90 min), while the inter¬val between pups three and four was significantlyshorter than that between pups two and three in thenaloxone-treated disturbed rats (mean difference7-2 + 4-2; «=13): these differences were significantly(P<005) different between the two groups. In ratswhich were given naloxone at the time of transfer, thetime-course of parturition was indistinguishable fromthat in undisturbed rats, whereas giving naloxone torats which were not transferred did not affect the over¬all time-course of parturition (Fig. 1). Thus naloxoneproduced a significant effect upon parturition only inthe disturbed rats, and in these rats it completelyeliminated the disruption produced by transfer into aglass chamber.

Plasma concentrations of oxytocinThe plasma concentration of oxytocin rose from29-0+ 3-8 pmol/1 (s.e.m.; « = 26) before the onset ofparturition (at least 6 h before the birth of the firstpup) to 69-8+ 17-4 pmol/1 (w = 26) at or immediatelyafter the birth of the third pup. In the naloxone-treated disturbed rats, samples taken at the birth

figure 1. Intervals between the birth of pups three and four,pups four and five, and pups five and six in rats injected withsaline or naloxone after the birth of pup three, and whichwere either left undisturbed in their nesting cages (columnslabelled SC and NC for saline and naloxone treatmentsrespectively) or were transferred at this time to a glass obser¬vation chamber (columns labelled SG and NG for saline andnaloxone treatments respectively). For comparison, data areshown from a previous experiment in which rats were notcannulated (column labelled UC). Values are means ±s.e.m.; numbers of rats are shown in parentheses. The inter-birth intervals between births three and four, four and five,and five and six are all prolonged in the saline-treated dis¬turbed group compared with all other groups (P < 001 oncumulative interval three to six for each comparison; /-test,confirmed with Mann-Whitney U test). The amount of pro¬longation is underestimated by this analysis as three ofthe15 rats in this group had not given birth to pup six by the endofthe 90-min observation period: for this analysis the timeof birth of these pups was arbitrarily assumed to be at 90min from the birth of pup three.

of the fourth, fifth and sixth pups showed higheroxytocin levels than samples taken at the same birthsin the saline-treated disturbed rats (average of the hor¬mone levels for the three births: 105-4+16-9 pmol/1,« = 12 vs 60-8 + 10-3 pmol/1, « = 9; < 005, Mann-Whitney U test). However, there was no significant

difference between the naloxone-treated disturbedand naloxone-treated undisturbed rats at this time,nor between the saline-treated disturbed and saline-treated undisturbed rats.

Effects of disturbance and of naloxone in non-pregnantrats

The sampling procedure itself did not affect themeasured levels of oxytocin. Eight samples takenfrom non-pregnant female rats at 10-min intervalsshowed no progressive change in oxytocin: insaline-treated undisturbed rats, oxytocin levels in thefirst sample were 24-8+ 7-9 pmol/1 compared with29-5 + 5-8 pmol/1 in the eighth sample (n= 12). Nalox¬one produced an immediate and significant increase inthe plasma concentration of oxytocin in rats placed ina glass observation chamber (P<0-05; naloxone-treated disturbed rats vs saline-treated undisturbedrats; comparison of percentage differences, betweenthe medians for each animal, of oxytocin levelsmeasured in the two samples immediately before andthe four samples immediately after naloxone or salineinjection; a logarithmic transformation was applied tothe raw data to reduce the effects of outlying values;Mann-Whitney U test). In undisturbed rats naloxoneproduced a smaller but still significant increase inoxytocin (P<0-01; naloxone-treated undisturbedrats vs saline-treated undisturbed rats) (Fig. 2).The disturbance alone produced no significantchange in plasma oxytocin in the non-pregnant rats:levels in saline-treated rats were not significantlydifferent between the disturbed and the undisturbedgroups.

DISCUSSION

Recent experiments have shown that morphine,administered either peripherally or centrally, willinterrupt parturition (Gosden et al. 1985) and depressoxytocin release (Cutting et al. 1985) in rats. This pro¬longation of parturition produced by morphine can beovercome by oxytocin infusions (Cutting et al. 1985).Environmental disturbance has previously beenshown to prolong parturition in mice (Newton et al.1966, 1968) and in rats (Leng et al. 1985). The presentresults support the hypothesis that the prolongationof parturition following environmental disturbance ismediated, at least partly, by activation of endogenousopioid pathways.

Some behavioural stresses have been reported toreduce plasma oxytocin levels in primates (Kalin,Gibbs, Barksdale et al. 1985), although severe stressproduces increased oxytocin levels in non-pregnantrats (Lang, Heil, Ganten et al. 1983). In the present

140

120

S lOO-lE

.£ 80-

> oto w

CL

60-

40-

20-

Saline (12) Naloxone (13)-1-*Undisturbed

Saline (13) Naloxone (12)^Disturbed

figure 2. Plasma concentrations of oxytocin in non-pregnant control rats. Eightsamples (0-3 ml) ofblood were taken at 10-min intervals. After the third sample,two groups of rats (disturbed rats) were transferred from their cage into a glassobservation chamber, and were given an i.v. injection ofeither naloxone(2 mg/kg) or saline. Two further groups (undisturbed rats) remained in theircages throughout. The histograms show mean + s.e.m. plasma concentrations ofoxytocin in the third and fourth samples (immediately before and after injectionand transfer; open and cross-hatched bars respectively), and in the sixth sample(30 min after transfer; solid bars). Numbers of rats are shown in parentheses.Naloxone produced a small but significant (P < 001 ; Mann-Whitney U test,comparison ofthe median of samples two and three with the median of samplesfour to seven) increase in circulating oxytocin in the undisturbed rats. In thedisturbed rats, naloxone produced a larger but more variable increase (compari¬son with saline-treated undisturbed rats, < 0-05; Mann-Whitney U test as

before).

study we found normal oxytocin levels at times ofbirth in rats following environmental disturbance inmid-parturition. Interpretation of this observation iscomplicated: as a consequence of the prolongation inparturition, the samples in the two groups correspondto two different times, but for each group to timeswhen birth was occurring. A small number of samplescollected at 10 min after transfer, from rats in whichparturition was interrupted for at least 20 min,showed oxytocin levels lower than those seen atbirth in normal rats, but lower levels of oxytocin inrats in which parturition is interrupted is to beexpected as a consequence of reduced activation oftheoxytocin cells via the Ferguson reflex (Higuchi,Uchide, Honda & Negoro, 1986). We are thus unableto resolve whether the disruption of parturition canbe directly attributed to inhibition of oxytocinrelease.

The environmental disturbance did not influenceplasma oxytocin levels in non-pregnant rats. As in the

parturient rats, naloxone raised oxytocin levels in dis¬turbed rats. However, while naloxone also raisedplasma oxytocin levels in undisturbed parturient rats,naloxone had less effect on oxytocin in undisturbednon-pregnant rats. One interpretation of these obser¬vations is that both the stress of parturition and thestress of environmental disturbance activate bothstimulatory pathways and inhibitory opioid pathwaysin the control of oxytocin secretion. Little is known ofthe physiological stimuli for the release of brainopioids, but the stress of labour and delivery inwomen produces a marked release of maternal ß-endorphin (Kofinas, Kofinas & Tavakoli, 1985), andhypothalamic levels of ß-endorphin are increasedtowards the end of pregnancy in the rat (Wardlow &Frantz, 1983). Opioids have a substantial inhibitoryinfluence upon oxytocin secretion through actionsboth at the neurosecretory terminals in the neuro¬

hypophysis (Clarke et al. 1979; Rossier, Pittman,Bloom & Guilleman, 1980; Bicknell & Leng, 1982),

and at the neurosecretory cell bodies within thehypothalamus (Wakerley, Noble & Clarke, 1983;Bicknell, 1985). Recent evidence suggests that opioids,possibly from the adrenal medulla, may mediate atleast some of the inhibitory influences of relaxin uponoxytocin release (O'Byrne, Eltringham, Clarke &Summerlee, 1986). The importance of the presentwork is that it attributes a physiological role toendogenous opioid pathways. Our hypothesis is thatstress prolongs parturition by activating opioidpathways to restrict the release of oxytocin. Even inundisturbed rats chronic naloxone treatment willslightly but significantly shorten the intervals betweenbirths (Leng et al. 1985). The present study hasdemonstrated dramatic effects of acute naloxoneupon rats disturbed in mid-parturition by transfer toan unfamiliar environment, and these effects are

accompanied by a significant potentiation of oxytocinrelease.

In women, anxiety is known to prolong labour andthe level of anxiety is correlated with the requirementfor oxytocin augmentation during delivery (Haddad,Morris & Spielberger, 1985). There is also a correla¬tion between extended labour and the plasma concen¬tration of ß-endorphin (Thomas, Fletcher & Hill,1982)-an opioid thought to be secreted from theadenohypophysis during labour in response to painand/or stress. Furthermore, women who have a spon¬taneous labour have lower plasma ß-endorphin levelsthan those women requiring either artificial rupture ofmembranes or oxytocin supplements (Thomas et al.1982). Anxious women are also more likely to requireepidural anaesthesia (Haddad et al. 1985), a treat¬ment known to prolong the second stage of labour(Crawford, 1979). This prolongation probablyresults from a weakening of both the power of thelower abdominal and pelvic muscles and reflexabdominal straining. Systemically applied analgesicscan also prolong delivery, in this case by slowing thefrequency of uterine contractions during labour

-

aneffect observed both with morphine and the opioid-like drug pethidine (Pétrie, Wu, Miller et al.1976).

There is therefore abundant circumstantial evidenceto suggest that anxiety in women directly and indirectlysuppresses oxytocin secretion to prolong labour.Current evidence leads to the conclusion that opioidmechanisms may be responsible for the suppression ofoxytocin secretion during anxiety and/or stress inlabour. It remains possible, however, that stress andendogenous opioids prolong parturition by partlyindependent mechanisms. We have previously sug¬gested that opioids play a role in normal parturition inthe rat to control the spacing of successive births(Leng et al. 1985) and, in support of this suggestion,we found in the present experiments that naloxone

increased oxytocin levels in undisturbed rats as well asin disturbed rats, in confirmation of the results ofHartman, Rosella-Dampman, Emmert & Summy-Long, 1986). It may thus be anticipated that some ofthe facilitatory influence of naloxone upon parturitionmay not depend upon the presence of stress, but suchfacilitation may be more readily apparent during theprolongation of parturition by stress.

ACKNOWLEDGEMENTS

We thank Dr T. Higuchi for the gift of oxytocin anti¬serum and Du Pont U.K. Ltd for the gift of naloxone.We also thank Drs P. Cobbett, R. Grossman, K.Shibuki and D. J. S. Sirinathsinghi for their con¬structive criticism of an earlier draft of thismanuscript.

REFERENCES

Bicknell, R. J. (1985). Endogenous opioid peptides and hypo¬thalamic neuroendocrine neurones. Journal ofEndocrinology 107,437^*46.

Bicknell, R. J. & Leng, G. (1982). Endogenous opiates regulate oxy¬tocin but not vasopressin secretion from the neurohypophysis.Nature 298,161-163.

Clarke, G., Wood, P., Merrick, L. P. & Lincoln, D. W. (1979).Opiate inhibition of peptide release from the neuro-humoralterminals of hypothalamic neurones. Nature 182,746-748.

Crawford, J. S. (1979). Continuous lumbar epidural analgesia forlabour and delivery. British Medical Journal i (6156), 72-74.

Cutting, R., Fitzsimons, N., Gosden, R. G., Humphreys, E. M.,Russell, J. ., Scott, S. & Stirland, J. A. (1985). Evidence thatmorphine interrupts parturition in rats by inhibiting oxytocinsecretion. Journal ofPhysiology 311,182P.

Dyer, R. G., Weick, W. F., Mansfield, S. & Corbet, . (1981).Secretion of luteinizing hormone in ovariectomized adult ratstreated neonatally with monosodium glutamate. Journal ofEndocrinology 91,341-346.

Fuchs, A.-R. (1969). Uterine activity in late pregnancy and duringparturition in the rat. Biology ofReproduction 1,344-353.

Fuchs, A.-R. & Poblete, V. F. (1970). Oxytocin and uterine functionin pregnant and parturient rats. Biology ofReproduction 2,387^*00.

Gosden, R. G., Humphreys, E. M., Johnston, V., Liddle, S. &Russell, J. A. ( 1985). Morphine acts centrally to interruptestablished parturition in the rat. Journal ofPhysiology 364, 59P.

Haddad, P. F., Morris, N. F. & Spielberger, C. D. (1985). Anxietyin pregnancy and its relation to use of oxytocin and analgesia inlabour. Journal ofObstetrics and Gynaecology 6,77-81.

Hartman, R. D., Rosella-Dampman, L. M., Emmert, S. E. &Summy-Long, J. Y. (1986). Inhibition of neurohypophysial hor¬mones by endogenous opioid peptides in pregnant and parturientrats. Brain Research 382,352-359.

Higuchi, T., Honda, K.., Fukuoka, T., Negoro, H. & Wakabayashi,K. (1985). Release of oxytocin during suckling and parturition inthe rat. Journal ofEndocrinology 105, 339-346.

Higuchi, T., Uchide, K., Honda, K. & Negoro, H. (1986). Oxytocinrelease during parturition in the pelvic-neurectomized rat.Journal ofEndocrinology 109, 149-154.

Jones, S. A. & Summerlee, A. J. S. (1986). Effects of porcine relaxinon the length.of gestation and duration of parturition in the rat.Journal ofEndocrinology 109,85-88.

Kalin, N. H., Gibbs, D. M., Barksdale, C. M., Shelton, S. E. &Carnes, M. ( 1985). Behavioral stress decreases oxytocin concen¬

trations in primates. Life Sciences 36, 1275-1280.Kofinas, G. D„ Kofinas, A. D. & Tavakoli, F. M. (1985). Maternal

and fetal ß-endorphin release in response to the stress of laborand delivery. American Journal ofObstetrics and Gynecology 152,56-59.

Lang, R. E., Heil, J. W. E., Ganten, D., Hermann, K., Unger, T. &Rascher, W. ( 1983). Oxytocin unlike vasopressin is a stresshormone in the rat. Neuroendocrinology 37, 314-316.

Leng, G, Bicknell, R. J., Mansfield, S. & Dyer, R. G. (1986).Oxytocin, opioids and parturition. Journal ofEndocrinology 108(Suppl), Abstract No. 19.

Leng, G., Mansfield, S., Bicknell, R. J., Dean, A. D. P., Ingram,C. D., Marsh, M. I. C, Yates, J. O. & Dyer, R. G. (1985). Centralopioids: a possible role in parturition? Journal ofEndocrinology106,219-224.

Newton, N., Foshee, D. & Newton, M. (1966). Parturient mice:effect of environment on labor. Science 151,1560-1561.

Newton, N., Peeler, D. & Newton, M. (1968). Effect of disturbanceon labor. American Journal ofObstetrics and Gynecology 101,1096-1102.

O'Byrne, K. T., Eltringham, L., Clarke, G & Summerlee, A. J. S.(1986). Effects of porcine relaxin on oxytocin release from the

neurohypophysis in the anaesthetized lactating rat. Journal ofEndocrinology 109,393-397.

Pétrie, R. H, Wu, R., Miller, F. C, Sacks, D., Sugarman, R., Paul,R. H. & Hon, . H. (1976). The effect of drugs on uterine activity.Obstetrics and Gynecology 48,431-435.

Rossier, J„ Pittman, Q., Bloom, F. & Guilleman, R. (1980). Distri¬bution of opioid peptides in the pituitary: a new hypothalamic-pars nervosa enkephalinergic pathway. Federation Proceedings39,2555-2560.

Spector, S., Ngai, S. H., Hempstead, J. & Berkowitz, B. A. (1978).Pharmacokinetics of naloxone in rats. In Factors Affecting theAction ofNarcotics, pp. 249-343. Eds M. W. Adler, L. Manara &R. Samonin. New York: Raven Press.

Summerlee, A. J. S. (1981). Extracellular recordings from oxytocinneurones during the expulsive phase of birth in unanaesthetisedrats. Journal ofPhysiology 321,1-10.

Thomas, . ., Fletcher, J. E. & Hill, R. G. (1982). Influence ofmedication, pain and progress in labour on plasma ß-endorphin-like immunoreactivity. British Journal ofAnaesthesia 54,401^108.

Wakerley, J. B., Noble, R. & Clarke, G. (1983). Effects ofmorphineand D-Ala, D-Leu enkephalin on the electrical activity of supra-optic neurosecretory cells in vitro. Neuroscience 10,73-81.

Wardlow, S. L. & Frantz, A. G. (1983). Brain ß-endorphinduring pregnancy, parturition and the postpartum period.Endocrinology 113, 1664-1668.