- Email: [email protected]
Cerebrospinal fluid and plasma concentrations of oxytocin and vasopressin during parturition and vaginocervical stimulation in the sheep
0361-9230191 $3.@l + .OO
Brain Research Bulletin,Vol. 26, pp. 803-807. e PergamonPress plc, 1991.Printedin the U.S.A
Cerebrospinal Fluid and Plasma Concentrations of Oxytocin and Vasopressin During Parturition and Vaginocervical Stimulation in the Sheep K. M. KENDRICK, E. B. KEVERNE,” M. R. HINTON AND J. A. GOODE
A.F.R.C. Institute of Animal Physiology and Genetics Research Cambridge Research Station, Babraham, Cambridge CB2 4AT ~~ub-De~ar~ent of Ani~l ge~viour, Universi~ of Cambridge, ~adingle~~ Cambridge Cl33 8AA Received 29 October 1990
KENDRICK, K. M., E. B. KEVERNE, M. R. HINTON AND J. A. GOODE. Cerebrospinalfluid andplasma concentrations of oxyocin and vasopressin during parturition and vaginocervical stimulation in the sheep. BRAIN RES BULL 26(5) 803-807, 1991.-Simultaneous blood and cerebrospinal fluid (CSF) samples were taken from conscious sheep before, during and after parturition. Concentrations of plasma and CSF oxytocin were significantly elevated during con~ctions and particularly at birth. Mean prepartum CSF concentrations of oxytocin were around 55% of those found in plasma but postpartum they were up to 2-fold higher than those in plasma. Plasma concentrations of oxytocin were only significantly elevated, compared to prepartum levels, for 15 min postpartum whereas those in CSF were increased for the whole of the 120 min postpartum sampling period. Plasma, but not CSF, con~n~tions of ~g~nine-v~opressin (AVP) were sig~cantly raised during con~actio~ and birth, and for 15 mm postpartum. During the prepartum period CSF AVP concentrations were 67% of those found in plasma whereas at birth plasma levels were lofold higher than in CSF. In a separate experiment it was shown that 5 mm of mechanical vaginocervical stimulation also stimulated significant increases in CSF and plasma oxytocin concentrations and in plasma vasopressin. Results support previous work suggesting an important role for central oxytocin release in the postpartum induction of maternal behavior and demonstrate that elevated concentrations of oxytocin in the CSF are present for a greater period than in blood. Elevated plasma AVP concentrations during contractions, birth or vaginocetvical stimulation may be stimulated by stress associated with these stimuli. Cerebrospinal fluid
Oxytocin
Parturition
Plasma
Sheep
IN sheep we have previously shown that cerebrospinal fluid (CSF) concen~tions of oxytocin increase during parturition and following artificial vaginocervical stimulation (11). A similar finding has also been reported in humans during labour (32). Other studies on sheep have found that there is increased release of oxytotin in the olfactory bulb, substantia nigra and medial preoptic area during pafturition (9, 12, 13). This central oxytocin release may be important for stimulating postpartum maternal behavior since central infusions of this peptide can induce maternal behavior in nonpregnant rats (4, 23, 25) and sheep (10). Central infusions of oxytocin antagonists or antisera also delay the onset of estrogen-stimulated maternal behavior in the rat (4,24). Further, vaginocervical stimulation, which increases CSF oxytocin concentrations (1 l), stimulates maternal behavior in nonpregnant ewes (15), whereas epiduml anaesthesia, which prevents the CSF rise in oxytocin during parturition in the sheep (19), blocks the induction of postpartum maternal behavior (16,lQ). The original study showing changes in CSF oxytocin during parturition (11) did not attempt to plot the precise time-course of this release and did not measure corresponding changes in plasma oxytocin concentrations. The present study therefore used more
Vaginocervical stimulation
Vasopressin
frequent CSF sampling together with blood samples to address this question. As a further control, CSF and plasma concentrations of another posterior pituitary peptide, arginine vasopressin (AVP), were measured. In the rat, for example, it has been shown that central infusions of AVP can stimulate maternal behavior, although with a long latency (23). METHOD
Nine adult female multiparous Clun Forest ewes were used. The animals were surgically implanted with bilateral, lateral ventricular cannulae, under general anaesthesia (10 mg sodium methohexitone IV-Brietal, Elanco Ltd. -followed by closed-circuit halothane) and with full sterile precautions, 6-7 weeks prior to parturition as previously described (1). The animals were housed inside and kept in individual pens. CSF and Plasma Sampiing Protocol
Blood samples (10 ml) were taken, by venipuncture, from the
803
804
KENDRICK.
jugular vein and CSF samples (I.5 ml) collected using a technique described previously (11). For more frequent sampling periods this technique was adapted slightly with the intracerebroventricular sampling needle being left in situ in the lateral ventricle. During periods between sampling the polythene cannula attached to the needle was attached to a syringe full of sterile 0.9% NaCl, which was tied to the fleece on the animal’s back. After each CSF sample the cannula was refilled with 0.9% NaCl. In general, CSF samples took l-5 min to collect and blood samples were taken approximately halfway through the CSF collection period. However, 2-3 blood samples were taken during CSF samples taken during the expulsion of the lamb and during vaginocervical stimulation to ensure that maximum concentrations were recorded. At 142 days of pregnancy the animals were given 20 mg IM injection of dexamethasone sodium phosphate (1 ml Decadron, Merck Sharpe & Dohme) to induce parturition (approximately 48 h after injection). For each animal 34 samples were taken during the period following these injections prior to the start of visible contractions being shown by the animals. A further 2-5 samples were taken during contractions and one immediately following the birth of the lamb. After the birth of the lamb samples were taken at 15. 30. 60, 90 and 120 min postpartum. Three months after parturition, and at least I month after the lambs had been removed from their mothers, 6 out of the 9 ewes which could still be sampled were given daily IM injections of 200 pg estradiol benzoate for 48 h and sampled immediately before, during and after a 5 min period of mechanical vaginocervical stimulation (VCS) using a plastic probe (4 cm diameter X 40 cm long). This stimulation comprises rhythmically inserting and withdrawing the probe (approximately 5-10 cm) with maximum pressure exerted on the cervix. During this stimulation the animals typically lie down and show head raising and lip curling (flehmen) similar to labour. Samples were taken at 10, 20 and 30 min post-VCS. The animals were estrogen primed since VCS will only induce maternal behavior in the presence of such priming. RIA All blood samples were centrifuged and frozen until assayed. Cerebrospinal fluid samples were frozen until assayed. The radioimmunoassays for plasma oxytocin (14,29) and AVP (22,33) were as previously described and lower detection limits for both were 0.1-0.3 pmol/l (1 ml samples run in duplicate). For CSF the assay procedures were the same except that the extraction step was omitted and lower detection limits were 0.4-0.8 pmol/I (300 p,l samples dried down and reconstituted in 50 ~1 of assay buffer and run in duplicate). Inter- and intra-assay coefficients of variation for AVP were 12.2 and 10.1% respectively and 12.6 and 9.5% for oxytocin. Some CSF samples were subjected to HPLC separation prior to radioimmunoassay and results confumed that oxytocin and vasopressin were being measured. Statistics Mean concentrations of oxytocin and AVP in plasma and CSF were calculated and subjected to a two-way analysis of variance (Friedman test). The 2-3 plasma samples taken following the birth of the lamb and during vaginocervical stimulation were reduced to a single mean value. Comparisons between individual sampling times were done using the Wilcoxon matchedpairs test. RESULTS
KEVERNE.
CSF
5,
HINTON AND GOODF
AVP
;
;2.;Illiiii* PRE-
LAB
B
15 Time
30
60
so
post-partum
PLASMA
120 (min)
AVP
20
PRE-LAB
B
15 Time
30
60
post-partum
90
120 (mtn)
FIG. 1. Mean + S.E.M. concentrations of arginine vasopressin in the CSF and plasma of nine sheep before during and after labour contractions (LAB) and birth (B). **p
Parturition Significant differences between concentrations of plasma AVP and oxytocin, and CSF oxytocin were found across the various
Vaginocervical Stimulation Significant differences
between concentrations
of plasma AVP
CSF AND PLASMA AVP AND OXYTOCIN
CSF *
805
DURING PARTURITION
OXYTOCIN
AVP
OXYTOCIN
CSF
CSF
l
50
l
I .
30 Time
PLASMA
60
90
post-partum
120 (min)
0 PRE-
VCS
10
Time
post-VCS
OXYTOCIN
20
30
i
PRE-
(min)
l
iii
i
VCS Time
PLASMA
10
20
post-VCS
30 (min)
PLASMA 200
:
*
PREPRE-
LAB
B
15 Time
30
60
post-partum
90
120
Time
post-VCS
(min)
I
VCS
10
Tame
post-VCS
20
30 (min)
(mln)
FIG. 2. Mean c S.E.M. concentrations of oxytocin in the CSF and plasma of nine sheep before during and after labour contractions (LAB) and birth (B). **p
and oxytocin, were found across the various sampling periods. Figure 3 shows that plasma AVP concentrations were significantly increased during VCS (from mean 2.8 pmol/l to 7.4 pmol/l, pcO.05) but not in the poststimulation samples taken at 10, 20 and 30 min. The mean of the maximum plasma AVP concentrations measured during VCS was 10.5 pmol/l. No significant changes in CSF concentrations of AVP were found. Figure 3 shows that plasma oxytocin concentrations were significantly increased following VCS (from 16.1 pmol/l to 100.5 pmol/l, p
These results confii our previous findings that CSF concentrations of oxytocin are increased significantly during parturition and following artificial vaginocervical stimulation. Oxytocin concentrations were still significantly elevated in CSF two hours postpartum, compared to prepartum levels, although plasma concentrations were only increased up to 15 min postpartum. Following VCS, CSF concentrations of oxytocin were raised significantly for 20 min, whereas plasma concentrations had fallen to prestimulation levels within 10 min. Plasma concentrations of AVP were also significantly increased during contractions and birth, whereas those in CSF remained unchanged. This increase in plasma AVP concentrations, like that of oxytocin, lasted for 15 min postpartum. Plasma AVP concentrations were also significantly increased
FIG. 3. Mean? S.E.M. concentrations of arginine vasopressin and oxytocin in the CSF and plasma of six sheep before (PEE-), during and after a 5 min period of mechanical vaginocervical stimulation (VCS). *p
following VCS but this increase disappeared by 10 min poststimulation. The longer time courses for the postpartum and post-VCS increases in CSF oxytocin, compared to those in plasma, probably mainly reflect the fact that this peptide has a longer half-life in the CSF compared to blood due to the lower concentrations of peptidases. In rats (21) and guinea pigs (8), for example, increased concentrations of AVP and oxytocin take 20-30 min to clear. However, the expulsion of fetal membranes and the placenta, and suckling during the 2 h postpartum may also have stimulated small increases in CSF oxytocin concentrations. Suckling, for example, does stimulate small increases in CSF oxytotin concentrations in sheep (11). The half-life of AVP and oxytocin in the blood is somewhat variable both within and across species with estimates ranging between 2 and 15 min (18). In sheep, an estimate of 8-10 min for clearance of AVP from the blood has been demonstrated (13). These figures would seem to be consistent with the rate of decline shown with both AVP and oxytocin concentrations postpartum and post-VCS . The durations of elevated concentrations of oxytocin and vasopressin in plasma and of oxytocin in CSF were shorter after VCS than postpartum, but this may simply reflect the fact that initial increases following VCS were much smaller than those seen at birth. In both cases there was in fact some indication that plasma AVP and oxytocin were still slightly elevated at 10 min post-VCS but this did not quite achieve significance. The increase in plasma concentrations of AVP during contractions and birth, are in agreement with previous work (30) although another study reported maximum levels of AVP in sheep plasma 48 h prepartum (1). Our finding that plasma AVP is also raised during VCS has not, however, been reported before in this species. The increases in plasma AVP concentrations could have been caused by a number of factors. In sheep, plasma cortisol
806
KENDRICK,
concentrations increase at parturition (31) and it is therefore possible that changes in Mood levels of AVP are stress related. Plasma AVP concentrations have also been shown to rise in cattle (17) and horses (2) during parturition and in cattle it has been claimed that changes in plasma osmolality may be responsible (17). It is also possible, although unlikely, that changes in blood pressure might have been responsible for the AVP increases seen in the present study, but since blood pressure measurements were not taken it remains to be seen whether decreases in blood pressure occur during parturition and following VCS. At present the most likely explanation is that plasma AVP changes during p~urition and following VCS are stress related. In some other species, such as the rat (7), rabbit (6) and goat (ZO), plasma vasopressin concentrations do not rise during parturition and there may therefore be species differences. Similarly, in the goat, vaginal distension was reported not to increase plasma AVP concentrations unless the animals were pretreated with naloxone (27). However, the authors also failed to find increased CSF concentrations of oxytocin following this stimulus and, since the animals were not estrogen primed and given a weaker form of VCS than we use on sheep, it is possible that either or both of these factors might have contributed to their negative findings. In the present study basal CSF concentrations of AVP and oxytocin were 67 and 56% of those in plasma, respectively. At birth plasma oxytocin concentrations were 2-3 times higher than those in CSF, however, postpartum concentrations in CSF were up to 2-fold higher than those found in the plasma. With AVP, on the other hand, at birth plasma concentrations were 9-12-fold greater than in CSF and postpartum concentrations were still up to IO-fold greater than those in CSF. Taken together, these findings confirm our previous results (I l), and those of others in the rat (21). that CSF concentrations of these two peptides are unlikely to be significantly influenced by plasma levels since in sheep only 0.01% of oxytocin passes across the blood-brain barrier (I 1) and in rats 0.002% (2 1). Further, our findings of increased oxytocin release within the brain of the sheep during parturition (9, 12, 13) also strongly suggest that the source of CSF oxytocin, and probably vasopressin. is the terminals containing these peptides within the brain.
KEVERNE,
HINTON AND GOODE
The relatively high concentrations of CSF oxytocin do raise the possibility that the peptide might be influencing receptor sites via transport in the brain ventricular system. With both oxytocin and vasopressin there is often an apparent mismatch between terminal and receptor localisations (26.34). Recent evidence has also shown that both in rats (5) and the goat (28) central administration of oxytocin can potentiate peripheral release of the peptide in response to suckling. It is possible that the long-fasting postpartum elevation of oxytocin in the CSF of the sheep might therefore facilitate peripheral oxytocin release during the important first few suckling bouts exhibited by the lambs. This would obviously ensure that the lambs gained the maximum amount of milk. for the minimum effort, during their first few hours of life. Our previous work has shown that central oxytocin may be important for the postpartum stimulation of maternal behavior in sheep. Central infusions of oxytocin in estrogen-primed, nonpregnant ewes, stimulate most aspects of maternal behavior within 30 s (10,14) and peridural anaesthesia, which severely disrupts the normal display of postpartum maternal care ( 16.19). prevents the increased central release of oxytocin (19). The long time course for raised CSF concentrations of oxytocin wouid also maximise the chances of a successful postpartum induction of postpartum maternal care. This would be of particular importance in animals that have a difficult birth, or are inexperienced, and take a longer time postpartum before taking an interest in the presence of their lambs. It is also interesting to note that raised CSF ~oncent~tions of oxytocin are present for the whole of the 2 h period postpartum during which the selective olfactory bond is made by the ewe for its lambs. Oxytocin concentrations are increased in the olfactory bulb during parturition ( 12) and in response to vaginocervical stimulation (13), and it is possible that this peptide might also be involved in the formation of the olfactory memory that a ewe uses to selectively recognise its own iambs. ACKNOWLEDGEMENTS
This work was partly supported by an A.F.R.C. Project grant to E.B.K. and by a M.A.F.F. commission (K.M.K.). We thank Mr. Duller for his help with the care and maintenance of the animals.
REFERENCES
t . Alexander, D. P.; Bashore, R. A.; Britton, H. G.; Forsling, M. L. 2.
3.
4.
5.
6. 7.
8.
9.
Maternal and fetal arginine vasopressin in the chronically catheterised sheep. Biol. Neonate 25:242-248; 1974. Allen, W. E.; Chard, T.; Forsling, M. L. Peripheral plasma levels of oxytocin and vasopressin in the mare during parturition. J. Endocrinol. 57:17.5-176; 1973. Blair-West. J. R.: Gibson, A.; Sheather, S. J.; Woods, R. L.; Brook, A. H. Vasopressin release in sheep following various degrees of rehydration. Am. J. Physiol. 253:R640-R645; 1987. Fahrbach, S. E.; Morrell, J. I.; Pfaff, D. W. Possible role for endogenous oxytocin in es~gen-f~ilitated maternal behaviour in rats. Neuroendocrinology 40:526-532; 1985. Freund-Mercier. M. J.: Moos, F.; Poulain, D. A.; Richard, P.; Rodriguez, F.; Theodosis, D. T.; Vincent, J. D. Role of central oxytotin in the control of the milk ejection reflex. Brain Res. Bull. 20: 737-741: 1988. Haldar, J. Independent release of oxytocin and vasopressin during oazturition in the rabbit. J. Phvsiol. (Land). 206723-730; 1970. kgman, R. D.; Rosella-Dampman, L. M.; Emmett, S. E.; SummyLong, .I. Y. Inhibition of release of neurohypophyseal hormones by endogenous opioid peptides in pregnant and parturient rats. Brain Res. 382352-359; 1986. Jones, P. M.; Robinson, I. C. A. F. Differential clearance of neurophysin and neurohypophyseal pepti&s from the cerebrospinal fluid in conscious guinea-pigs. Neuroendocrinology 34297-302; 1982. Kendrick, K. M.: Keveme, E. B. Release of mono~ines, amino
10.
Il.
12.
13.
14.
15.
16.
acids and oxytocin from the sheep brain during portion. Curr. Sep. 9:114; 1989. Kendrick. K. M.: Keveme, E. 8.; Baldwin, B. A. Intracerebroventricular oxytocin stimulates maternal behaviour in the sheep. Neuroendocrinology 46:56-61; 1987. Kendrick, K. M.; Keveme, E. B.; Baldwin, B. A.: Sharman, D. F. Cerebrospinal fluid levels of acetylcholinesterase, monamines and oxytocin during iabour, pa~urition, vagino-cervical stimulation, lamb separation and suckling in sheep. Neuroendocrinology 44: 149-l 56; 1986. Kendrick, K. M.; Keveme, E. B.; Chapman. C.; Baldwin, B. A. Intracraniai dialysis measurement of oxytocin, monoamines and uric acid release from the olfactory bulb and substantia nigra of sheep during parturition, suckling, separation from lambs and eating. Brain Res. 439:1-10; 1988. Kendrick, K. M.; Keveme, E. B.; Chapman, C.; Baldwin, B. A. Microdialysis measurement of oxytocin, aspartate, y-aminobutyric acid and glutamate release from the olfactory bulb of the sheep during vaginocervical stimulation. Brain Res. 442:17 l-174; 1988. Keverne, E. B.; Kendrick. K. M. Morphine and corticotrophin releasing factor potentiate maternal acceptance in multiparous ewes after vaginocervical stimulation. Brain Res.; in press. Keveme, E. B.; Levy, F.; Poindron, P.; Lindsay, D. R. Vaginal stimulation: an important determinant of maternal bonding in sheep. Science 219:81-83; 1983. Krehbiel, D.; Poindron. P.; Levy, F.: ~d’Homme~ M. J. Feridurai
CSF AND PLASMA AVP AND OXYTOCIN DURING PARTURITION
17.
18.
19.
20.
21.
22.
23.
24.
25.
anaesthesia disturbs maternal behaviour in primiparous and multiparous parturient ewes. Physiol. Behav. 40:463-472; 1987. Landgraf, R.; Schulz, J.; Eulenberger, K.; Wilhelm, J. Plasma levels of oxytocin and vasopressin before, during and after parturition in cows. Exp. Clin. Endocrinol. 81:321-328; 1983. Lauson, H. D. Metabolism of the neurohypophyseal hormones. In: Handbook of physiology. Endocrinology. sect. 7, vol. IV, part 1. Washington, DC: Am. Physiol. Sot.; 1984:287-393. Levy, F.; Kendrick, K. M.; Keveme, E. B.; Grgeur, P.; Piketty, V.; Poindron, P. (1989) Neuroendocrine mechanisms of maternal behaviour, facilitation by genital stimulation in sheep. In: Symposium of the European society for comparative physiology and biochemistry. Liege: Karger; 1984:107-108. McNeilly, A. S.; Martin, M. J.; Chard, T.; Hart, I. C. Simultaneous release of oxytocin and neurophysin during parturition in the goat. J. Endocrinol. 52:213-214; 1972. Mens, W. B. J.; Witter, A.; Van Wimersma Greidanus, T. B. Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): half-times of disappearance of these neuropeptides from CSF. Brain Res. 262:143-149; 1983. Moore, G.; Lutterodt, A.; Burford, G.; Lederis, K. A highly specific antiserum for arginine vasopressin. Endocrinology 101:1421-1435; 1977. Pedersen, C. A.; Ascher, J. A.; Monroe, Y. L.; Prange, A. J. Oxytocin induces maternal behaviour in virgin female rats. Science 216: 648-649; 1982. Pedersen, C. A.; Caldwell, J. D.; Johnson, M. F.; Fort, S. A.; Prange, A. J., Jr. Oxytocin antiserum delays onset of ovarian steroidinduced maternal behaviour. Neuropeptides 6:175-182; 1985. Pedersen, C. A.; Prange, A. J. Induction of maternal behaviour in virgin rats after intracerebroventricular administration of oxytocin.
807
Proc. Natl. Acad. Sci. USA 76:6661-6665; 1979. M.; Charpak, S.; Dreifuss, J. J. 26. Raggenbass, M.; Dubois-Dauphin, Neurons in the dorsal motor nucleus of the vagus nerve are excited by oxytocin in the rat but not in the guinea pig. Proc. Natl. Acad. Sci. USA 84:3926-3930; 1987. 27. Seckl, J. R.; Lightman, S. L. Effect of naloxone on oxytocin and vasopressin release during vaginocervical stimulation in the goat. J. Endocrinol. 115:317-322; 1987. 28. Se&l, J. R.; Lightman, S. L. Potentiation of lactation-induced oxytocin secretion by intracerebroventricular oxytocin in the conscious goat. J. Endocrinol. 116:273-277; 1988. 29. Sheldrick, E. L.; Flint, A. P. S. Circulating concentrations of oxytotin during the oestrous cycle and early pregnancy in sheep. Prostaglandins 22:631-636; 1981. 30. Stark, R. I.; Daniel, S. S.; Husain, K. M.; James, L. S.; Vande Wiele, R. L. Arginine vasopressin during gestation and parturition in sheen fetus. Biol. Neonate 35:235-241: 1979. 31. Strott, C.; Sundel, H.; Stahlman, M. T. Maternal and fetal plasma progesterone, cortisol, testosterone and 17B-estradiol in preparturient sheep: response to fetal ACTH infusion. Endocrinology 5:1327-1339; 1974. 32. Takeda, S.; Kiwabara, Y.; Mizuno, M. Effects of pregnancy and labor on oxytocin levels in human plasma and cerebrospinal fluid. Endocrinol. Jpn. 32:875-880; 1985. 33. Thornton, S.; Leng, G.; Bicknell, R. 3.; Chapman, C.; Purdew, T. Vasopressin, but not oxytocin, is released in response to water deprivation in conscious goats. J. Endocrinol. 110:335-340; 1986. 34. Van Leeuwen, F. W. Vasopressin receptors in the brain and pituitary. In: Gash, D. M.; Boer, G. J., eds. Vasopressin principles and properties. New York: Plenum Press; 1987:477-496.
Brain Research Bulletin,Vol. 26, pp. 803-807. e PergamonPress plc, 1991.Printedin the U.S.A
Cerebrospinal Fluid and Plasma Concentrations of Oxytocin and Vasopressin During Parturition and Vaginocervical Stimulation in the Sheep K. M. KENDRICK, E. B. KEVERNE,” M. R. HINTON AND J. A. GOODE
A.F.R.C. Institute of Animal Physiology and Genetics Research Cambridge Research Station, Babraham, Cambridge CB2 4AT ~~ub-De~ar~ent of Ani~l ge~viour, Universi~ of Cambridge, ~adingle~~ Cambridge Cl33 8AA Received 29 October 1990
KENDRICK, K. M., E. B. KEVERNE, M. R. HINTON AND J. A. GOODE. Cerebrospinalfluid andplasma concentrations of oxyocin and vasopressin during parturition and vaginocervical stimulation in the sheep. BRAIN RES BULL 26(5) 803-807, 1991.-Simultaneous blood and cerebrospinal fluid (CSF) samples were taken from conscious sheep before, during and after parturition. Concentrations of plasma and CSF oxytocin were significantly elevated during con~ctions and particularly at birth. Mean prepartum CSF concentrations of oxytocin were around 55% of those found in plasma but postpartum they were up to 2-fold higher than those in plasma. Plasma concentrations of oxytocin were only significantly elevated, compared to prepartum levels, for 15 min postpartum whereas those in CSF were increased for the whole of the 120 min postpartum sampling period. Plasma, but not CSF, con~n~tions of ~g~nine-v~opressin (AVP) were sig~cantly raised during con~actio~ and birth, and for 15 mm postpartum. During the prepartum period CSF AVP concentrations were 67% of those found in plasma whereas at birth plasma levels were lofold higher than in CSF. In a separate experiment it was shown that 5 mm of mechanical vaginocervical stimulation also stimulated significant increases in CSF and plasma oxytocin concentrations and in plasma vasopressin. Results support previous work suggesting an important role for central oxytocin release in the postpartum induction of maternal behavior and demonstrate that elevated concentrations of oxytocin in the CSF are present for a greater period than in blood. Elevated plasma AVP concentrations during contractions, birth or vaginocetvical stimulation may be stimulated by stress associated with these stimuli. Cerebrospinal fluid
Oxytocin
Parturition
Plasma
Sheep
IN sheep we have previously shown that cerebrospinal fluid (CSF) concen~tions of oxytocin increase during parturition and following artificial vaginocervical stimulation (11). A similar finding has also been reported in humans during labour (32). Other studies on sheep have found that there is increased release of oxytotin in the olfactory bulb, substantia nigra and medial preoptic area during pafturition (9, 12, 13). This central oxytocin release may be important for stimulating postpartum maternal behavior since central infusions of this peptide can induce maternal behavior in nonpregnant rats (4, 23, 25) and sheep (10). Central infusions of oxytocin antagonists or antisera also delay the onset of estrogen-stimulated maternal behavior in the rat (4,24). Further, vaginocervical stimulation, which increases CSF oxytocin concentrations (1 l), stimulates maternal behavior in nonpregnant ewes (15), whereas epiduml anaesthesia, which prevents the CSF rise in oxytocin during parturition in the sheep (19), blocks the induction of postpartum maternal behavior (16,lQ). The original study showing changes in CSF oxytocin during parturition (11) did not attempt to plot the precise time-course of this release and did not measure corresponding changes in plasma oxytocin concentrations. The present study therefore used more
Vaginocervical stimulation
Vasopressin
frequent CSF sampling together with blood samples to address this question. As a further control, CSF and plasma concentrations of another posterior pituitary peptide, arginine vasopressin (AVP), were measured. In the rat, for example, it has been shown that central infusions of AVP can stimulate maternal behavior, although with a long latency (23). METHOD
Nine adult female multiparous Clun Forest ewes were used. The animals were surgically implanted with bilateral, lateral ventricular cannulae, under general anaesthesia (10 mg sodium methohexitone IV-Brietal, Elanco Ltd. -followed by closed-circuit halothane) and with full sterile precautions, 6-7 weeks prior to parturition as previously described (1). The animals were housed inside and kept in individual pens. CSF and Plasma Sampiing Protocol
Blood samples (10 ml) were taken, by venipuncture, from the
803
804
KENDRICK.
jugular vein and CSF samples (I.5 ml) collected using a technique described previously (11). For more frequent sampling periods this technique was adapted slightly with the intracerebroventricular sampling needle being left in situ in the lateral ventricle. During periods between sampling the polythene cannula attached to the needle was attached to a syringe full of sterile 0.9% NaCl, which was tied to the fleece on the animal’s back. After each CSF sample the cannula was refilled with 0.9% NaCl. In general, CSF samples took l-5 min to collect and blood samples were taken approximately halfway through the CSF collection period. However, 2-3 blood samples were taken during CSF samples taken during the expulsion of the lamb and during vaginocervical stimulation to ensure that maximum concentrations were recorded. At 142 days of pregnancy the animals were given 20 mg IM injection of dexamethasone sodium phosphate (1 ml Decadron, Merck Sharpe & Dohme) to induce parturition (approximately 48 h after injection). For each animal 34 samples were taken during the period following these injections prior to the start of visible contractions being shown by the animals. A further 2-5 samples were taken during contractions and one immediately following the birth of the lamb. After the birth of the lamb samples were taken at 15. 30. 60, 90 and 120 min postpartum. Three months after parturition, and at least I month after the lambs had been removed from their mothers, 6 out of the 9 ewes which could still be sampled were given daily IM injections of 200 pg estradiol benzoate for 48 h and sampled immediately before, during and after a 5 min period of mechanical vaginocervical stimulation (VCS) using a plastic probe (4 cm diameter X 40 cm long). This stimulation comprises rhythmically inserting and withdrawing the probe (approximately 5-10 cm) with maximum pressure exerted on the cervix. During this stimulation the animals typically lie down and show head raising and lip curling (flehmen) similar to labour. Samples were taken at 10, 20 and 30 min post-VCS. The animals were estrogen primed since VCS will only induce maternal behavior in the presence of such priming. RIA All blood samples were centrifuged and frozen until assayed. Cerebrospinal fluid samples were frozen until assayed. The radioimmunoassays for plasma oxytocin (14,29) and AVP (22,33) were as previously described and lower detection limits for both were 0.1-0.3 pmol/l (1 ml samples run in duplicate). For CSF the assay procedures were the same except that the extraction step was omitted and lower detection limits were 0.4-0.8 pmol/I (300 p,l samples dried down and reconstituted in 50 ~1 of assay buffer and run in duplicate). Inter- and intra-assay coefficients of variation for AVP were 12.2 and 10.1% respectively and 12.6 and 9.5% for oxytocin. Some CSF samples were subjected to HPLC separation prior to radioimmunoassay and results confumed that oxytocin and vasopressin were being measured. Statistics Mean concentrations of oxytocin and AVP in plasma and CSF were calculated and subjected to a two-way analysis of variance (Friedman test). The 2-3 plasma samples taken following the birth of the lamb and during vaginocervical stimulation were reduced to a single mean value. Comparisons between individual sampling times were done using the Wilcoxon matchedpairs test. RESULTS
KEVERNE.
CSF
5,
HINTON AND GOODF
AVP
;
;2.;Illiiii* PRE-
LAB
B
15 Time
30
60
so
post-partum
PLASMA
120 (min)
AVP
20
PRE-LAB
B
15 Time
30
60
post-partum
90
120 (mtn)
FIG. 1. Mean + S.E.M. concentrations of arginine vasopressin in the CSF and plasma of nine sheep before during and after labour contractions (LAB) and birth (B). **p
Parturition Significant differences between concentrations of plasma AVP and oxytocin, and CSF oxytocin were found across the various
Vaginocervical Stimulation Significant differences
between concentrations
of plasma AVP
CSF AND PLASMA AVP AND OXYTOCIN
CSF *
805
DURING PARTURITION
OXYTOCIN
AVP
OXYTOCIN
CSF
CSF
l
50
l
I .
30 Time
PLASMA
60
90
post-partum
120 (min)
0 PRE-
VCS
10
Time
post-VCS
OXYTOCIN
20
30
i
PRE-
(min)
l
iii
i
VCS Time
PLASMA
10
20
post-VCS
30 (min)
PLASMA 200
:
*
PREPRE-
LAB
B
15 Time
30
60
post-partum
90
120
Time
post-VCS
(min)
I
VCS
10
Tame
post-VCS
20
30 (min)
(mln)
FIG. 2. Mean c S.E.M. concentrations of oxytocin in the CSF and plasma of nine sheep before during and after labour contractions (LAB) and birth (B). **p
and oxytocin, were found across the various sampling periods. Figure 3 shows that plasma AVP concentrations were significantly increased during VCS (from mean 2.8 pmol/l to 7.4 pmol/l, pcO.05) but not in the poststimulation samples taken at 10, 20 and 30 min. The mean of the maximum plasma AVP concentrations measured during VCS was 10.5 pmol/l. No significant changes in CSF concentrations of AVP were found. Figure 3 shows that plasma oxytocin concentrations were significantly increased following VCS (from 16.1 pmol/l to 100.5 pmol/l, p
These results confii our previous findings that CSF concentrations of oxytocin are increased significantly during parturition and following artificial vaginocervical stimulation. Oxytocin concentrations were still significantly elevated in CSF two hours postpartum, compared to prepartum levels, although plasma concentrations were only increased up to 15 min postpartum. Following VCS, CSF concentrations of oxytocin were raised significantly for 20 min, whereas plasma concentrations had fallen to prestimulation levels within 10 min. Plasma concentrations of AVP were also significantly increased during contractions and birth, whereas those in CSF remained unchanged. This increase in plasma AVP concentrations, like that of oxytocin, lasted for 15 min postpartum. Plasma AVP concentrations were also significantly increased
FIG. 3. Mean? S.E.M. concentrations of arginine vasopressin and oxytocin in the CSF and plasma of six sheep before (PEE-), during and after a 5 min period of mechanical vaginocervical stimulation (VCS). *p
following VCS but this increase disappeared by 10 min poststimulation. The longer time courses for the postpartum and post-VCS increases in CSF oxytocin, compared to those in plasma, probably mainly reflect the fact that this peptide has a longer half-life in the CSF compared to blood due to the lower concentrations of peptidases. In rats (21) and guinea pigs (8), for example, increased concentrations of AVP and oxytocin take 20-30 min to clear. However, the expulsion of fetal membranes and the placenta, and suckling during the 2 h postpartum may also have stimulated small increases in CSF oxytocin concentrations. Suckling, for example, does stimulate small increases in CSF oxytotin concentrations in sheep (11). The half-life of AVP and oxytocin in the blood is somewhat variable both within and across species with estimates ranging between 2 and 15 min (18). In sheep, an estimate of 8-10 min for clearance of AVP from the blood has been demonstrated (13). These figures would seem to be consistent with the rate of decline shown with both AVP and oxytocin concentrations postpartum and post-VCS . The durations of elevated concentrations of oxytocin and vasopressin in plasma and of oxytocin in CSF were shorter after VCS than postpartum, but this may simply reflect the fact that initial increases following VCS were much smaller than those seen at birth. In both cases there was in fact some indication that plasma AVP and oxytocin were still slightly elevated at 10 min post-VCS but this did not quite achieve significance. The increase in plasma concentrations of AVP during contractions and birth, are in agreement with previous work (30) although another study reported maximum levels of AVP in sheep plasma 48 h prepartum (1). Our finding that plasma AVP is also raised during VCS has not, however, been reported before in this species. The increases in plasma AVP concentrations could have been caused by a number of factors. In sheep, plasma cortisol
806
KENDRICK,
concentrations increase at parturition (31) and it is therefore possible that changes in Mood levels of AVP are stress related. Plasma AVP concentrations have also been shown to rise in cattle (17) and horses (2) during parturition and in cattle it has been claimed that changes in plasma osmolality may be responsible (17). It is also possible, although unlikely, that changes in blood pressure might have been responsible for the AVP increases seen in the present study, but since blood pressure measurements were not taken it remains to be seen whether decreases in blood pressure occur during parturition and following VCS. At present the most likely explanation is that plasma AVP changes during p~urition and following VCS are stress related. In some other species, such as the rat (7), rabbit (6) and goat (ZO), plasma vasopressin concentrations do not rise during parturition and there may therefore be species differences. Similarly, in the goat, vaginal distension was reported not to increase plasma AVP concentrations unless the animals were pretreated with naloxone (27). However, the authors also failed to find increased CSF concentrations of oxytocin following this stimulus and, since the animals were not estrogen primed and given a weaker form of VCS than we use on sheep, it is possible that either or both of these factors might have contributed to their negative findings. In the present study basal CSF concentrations of AVP and oxytocin were 67 and 56% of those in plasma, respectively. At birth plasma oxytocin concentrations were 2-3 times higher than those in CSF, however, postpartum concentrations in CSF were up to 2-fold higher than those found in the plasma. With AVP, on the other hand, at birth plasma concentrations were 9-12-fold greater than in CSF and postpartum concentrations were still up to IO-fold greater than those in CSF. Taken together, these findings confirm our previous results (I l), and those of others in the rat (21). that CSF concentrations of these two peptides are unlikely to be significantly influenced by plasma levels since in sheep only 0.01% of oxytocin passes across the blood-brain barrier (I 1) and in rats 0.002% (2 1). Further, our findings of increased oxytocin release within the brain of the sheep during parturition (9, 12, 13) also strongly suggest that the source of CSF oxytocin, and probably vasopressin. is the terminals containing these peptides within the brain.
KEVERNE,
HINTON AND GOODE
The relatively high concentrations of CSF oxytocin do raise the possibility that the peptide might be influencing receptor sites via transport in the brain ventricular system. With both oxytocin and vasopressin there is often an apparent mismatch between terminal and receptor localisations (26.34). Recent evidence has also shown that both in rats (5) and the goat (28) central administration of oxytocin can potentiate peripheral release of the peptide in response to suckling. It is possible that the long-fasting postpartum elevation of oxytocin in the CSF of the sheep might therefore facilitate peripheral oxytocin release during the important first few suckling bouts exhibited by the lambs. This would obviously ensure that the lambs gained the maximum amount of milk. for the minimum effort, during their first few hours of life. Our previous work has shown that central oxytocin may be important for the postpartum stimulation of maternal behavior in sheep. Central infusions of oxytocin in estrogen-primed, nonpregnant ewes, stimulate most aspects of maternal behavior within 30 s (10,14) and peridural anaesthesia, which severely disrupts the normal display of postpartum maternal care ( 16.19). prevents the increased central release of oxytocin (19). The long time course for raised CSF concentrations of oxytocin wouid also maximise the chances of a successful postpartum induction of postpartum maternal care. This would be of particular importance in animals that have a difficult birth, or are inexperienced, and take a longer time postpartum before taking an interest in the presence of their lambs. It is also interesting to note that raised CSF ~oncent~tions of oxytocin are present for the whole of the 2 h period postpartum during which the selective olfactory bond is made by the ewe for its lambs. Oxytocin concentrations are increased in the olfactory bulb during parturition ( 12) and in response to vaginocervical stimulation (13), and it is possible that this peptide might also be involved in the formation of the olfactory memory that a ewe uses to selectively recognise its own iambs. ACKNOWLEDGEMENTS
This work was partly supported by an A.F.R.C. Project grant to E.B.K. and by a M.A.F.F. commission (K.M.K.). We thank Mr. Duller for his help with the care and maintenance of the animals.
REFERENCES
t . Alexander, D. P.; Bashore, R. A.; Britton, H. G.; Forsling, M. L. 2.
3.
4.
5.
6. 7.
8.
9.
Maternal and fetal arginine vasopressin in the chronically catheterised sheep. Biol. Neonate 25:242-248; 1974. Allen, W. E.; Chard, T.; Forsling, M. L. Peripheral plasma levels of oxytocin and vasopressin in the mare during parturition. J. Endocrinol. 57:17.5-176; 1973. Blair-West. J. R.: Gibson, A.; Sheather, S. J.; Woods, R. L.; Brook, A. H. Vasopressin release in sheep following various degrees of rehydration. Am. J. Physiol. 253:R640-R645; 1987. Fahrbach, S. E.; Morrell, J. I.; Pfaff, D. W. Possible role for endogenous oxytocin in es~gen-f~ilitated maternal behaviour in rats. Neuroendocrinology 40:526-532; 1985. Freund-Mercier. M. J.: Moos, F.; Poulain, D. A.; Richard, P.; Rodriguez, F.; Theodosis, D. T.; Vincent, J. D. Role of central oxytotin in the control of the milk ejection reflex. Brain Res. Bull. 20: 737-741: 1988. Haldar, J. Independent release of oxytocin and vasopressin during oazturition in the rabbit. J. Phvsiol. (Land). 206723-730; 1970. kgman, R. D.; Rosella-Dampman, L. M.; Emmett, S. E.; SummyLong, .I. Y. Inhibition of release of neurohypophyseal hormones by endogenous opioid peptides in pregnant and parturient rats. Brain Res. 382352-359; 1986. Jones, P. M.; Robinson, I. C. A. F. Differential clearance of neurophysin and neurohypophyseal pepti&s from the cerebrospinal fluid in conscious guinea-pigs. Neuroendocrinology 34297-302; 1982. Kendrick, K. M.: Keveme, E. B. Release of mono~ines, amino
10.
Il.
12.
13.
14.
15.
16.
acids and oxytocin from the sheep brain during portion. Curr. Sep. 9:114; 1989. Kendrick. K. M.: Keveme, E. 8.; Baldwin, B. A. Intracerebroventricular oxytocin stimulates maternal behaviour in the sheep. Neuroendocrinology 46:56-61; 1987. Kendrick, K. M.; Keveme, E. B.; Baldwin, B. A.: Sharman, D. F. Cerebrospinal fluid levels of acetylcholinesterase, monamines and oxytocin during iabour, pa~urition, vagino-cervical stimulation, lamb separation and suckling in sheep. Neuroendocrinology 44: 149-l 56; 1986. Kendrick, K. M.; Keveme, E. B.; Chapman. C.; Baldwin, B. A. Intracraniai dialysis measurement of oxytocin, monoamines and uric acid release from the olfactory bulb and substantia nigra of sheep during parturition, suckling, separation from lambs and eating. Brain Res. 439:1-10; 1988. Kendrick, K. M.; Keveme, E. B.; Chapman, C.; Baldwin, B. A. Microdialysis measurement of oxytocin, aspartate, y-aminobutyric acid and glutamate release from the olfactory bulb of the sheep during vaginocervical stimulation. Brain Res. 442:17 l-174; 1988. Keverne, E. B.; Kendrick. K. M. Morphine and corticotrophin releasing factor potentiate maternal acceptance in multiparous ewes after vaginocervical stimulation. Brain Res.; in press. Keveme, E. B.; Levy, F.; Poindron, P.; Lindsay, D. R. Vaginal stimulation: an important determinant of maternal bonding in sheep. Science 219:81-83; 1983. Krehbiel, D.; Poindron. P.; Levy, F.: ~d’Homme~ M. J. Feridurai
CSF AND PLASMA AVP AND OXYTOCIN DURING PARTURITION
17.
18.
19.
20.
21.
22.
23.
24.
25.
anaesthesia disturbs maternal behaviour in primiparous and multiparous parturient ewes. Physiol. Behav. 40:463-472; 1987. Landgraf, R.; Schulz, J.; Eulenberger, K.; Wilhelm, J. Plasma levels of oxytocin and vasopressin before, during and after parturition in cows. Exp. Clin. Endocrinol. 81:321-328; 1983. Lauson, H. D. Metabolism of the neurohypophyseal hormones. In: Handbook of physiology. Endocrinology. sect. 7, vol. IV, part 1. Washington, DC: Am. Physiol. Sot.; 1984:287-393. Levy, F.; Kendrick, K. M.; Keveme, E. B.; Grgeur, P.; Piketty, V.; Poindron, P. (1989) Neuroendocrine mechanisms of maternal behaviour, facilitation by genital stimulation in sheep. In: Symposium of the European society for comparative physiology and biochemistry. Liege: Karger; 1984:107-108. McNeilly, A. S.; Martin, M. J.; Chard, T.; Hart, I. C. Simultaneous release of oxytocin and neurophysin during parturition in the goat. J. Endocrinol. 52:213-214; 1972. Mens, W. B. J.; Witter, A.; Van Wimersma Greidanus, T. B. Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): half-times of disappearance of these neuropeptides from CSF. Brain Res. 262:143-149; 1983. Moore, G.; Lutterodt, A.; Burford, G.; Lederis, K. A highly specific antiserum for arginine vasopressin. Endocrinology 101:1421-1435; 1977. Pedersen, C. A.; Ascher, J. A.; Monroe, Y. L.; Prange, A. J. Oxytocin induces maternal behaviour in virgin female rats. Science 216: 648-649; 1982. Pedersen, C. A.; Caldwell, J. D.; Johnson, M. F.; Fort, S. A.; Prange, A. J., Jr. Oxytocin antiserum delays onset of ovarian steroidinduced maternal behaviour. Neuropeptides 6:175-182; 1985. Pedersen, C. A.; Prange, A. J. Induction of maternal behaviour in virgin rats after intracerebroventricular administration of oxytocin.
807
Proc. Natl. Acad. Sci. USA 76:6661-6665; 1979. M.; Charpak, S.; Dreifuss, J. J. 26. Raggenbass, M.; Dubois-Dauphin, Neurons in the dorsal motor nucleus of the vagus nerve are excited by oxytocin in the rat but not in the guinea pig. Proc. Natl. Acad. Sci. USA 84:3926-3930; 1987. 27. Seckl, J. R.; Lightman, S. L. Effect of naloxone on oxytocin and vasopressin release during vaginocervical stimulation in the goat. J. Endocrinol. 115:317-322; 1987. 28. Se&l, J. R.; Lightman, S. L. Potentiation of lactation-induced oxytocin secretion by intracerebroventricular oxytocin in the conscious goat. J. Endocrinol. 116:273-277; 1988. 29. Sheldrick, E. L.; Flint, A. P. S. Circulating concentrations of oxytotin during the oestrous cycle and early pregnancy in sheep. Prostaglandins 22:631-636; 1981. 30. Stark, R. I.; Daniel, S. S.; Husain, K. M.; James, L. S.; Vande Wiele, R. L. Arginine vasopressin during gestation and parturition in sheen fetus. Biol. Neonate 35:235-241: 1979. 31. Strott, C.; Sundel, H.; Stahlman, M. T. Maternal and fetal plasma progesterone, cortisol, testosterone and 17B-estradiol in preparturient sheep: response to fetal ACTH infusion. Endocrinology 5:1327-1339; 1974. 32. Takeda, S.; Kiwabara, Y.; Mizuno, M. Effects of pregnancy and labor on oxytocin levels in human plasma and cerebrospinal fluid. Endocrinol. Jpn. 32:875-880; 1985. 33. Thornton, S.; Leng, G.; Bicknell, R. 3.; Chapman, C.; Purdew, T. Vasopressin, but not oxytocin, is released in response to water deprivation in conscious goats. J. Endocrinol. 110:335-340; 1986. 34. Van Leeuwen, F. W. Vasopressin receptors in the brain and pituitary. In: Gash, D. M.; Boer, G. J., eds. Vasopressin principles and properties. New York: Plenum Press; 1987:477-496.