Evidence for the secretion of immunoreactive neurophysin I in addition to from the ovary in cattle D. Schams, E. Schallenberger and J. J. Legros Institut für Physiologie, Süddeutsche Versuchs- und Forschungsanstalt für Milchwirtschaft, Technische Universität München, 8050 Freising- Weihenstephan, Federal Republic of Germany and * Unité de Psychoneuroendocrinologie, Institut de Médecine et Laboratoire de Radioimmunologie, Université de Liège, Sart-Tilman, 4000 Liège, Belgium

Summary. In Exp. I, blood samples were collected simultaneously from the posterior vena cava and jugular vein or aorta from 7 heifers every 5\p=n-\20min for 2\p=n-\5h. Concomi- tant pulsatile secretion of oxytocin and immunoreactive neurophysin I was detected in the vena cava, but not in the jugular vein or aorta. Concentrations of oxytocin and immunoreactive neurophysin increased earlier and were higher in the vena cava than in the jugular vein or aorta after the injection of a luteolytic dose of prostaglandin F-2\g=a\ analogue during the mid-luteal phase of the oestrous cycle, demonstrating its ovarian but not pituitary origin. In Exp. II, blood samples were collected from the jugular vein every 12 h during 1 week after oestrus. Follicular growth had been stimulated during the preceding oestrous cycle with PMSG (10 heifers and cows) or with FSH (5 animals); 6 heifers served as controls. There was a high correlation between the number of follicles or CL and the increase in oxytocin and immunoreactive neurophysin I. Although PMSG had a greater luteotrophic effect than did FSH on progesterone secre- tion, a similar stimulation of oxytocin and immunoreactive neurophysin I was not observed. It is concluded that immunoreactive neurophysin I and oxytocin are secreted from the ovary in concentrations dependent upon the number of corpora lutea (and of follicles) present. During the mid-luteal period the secretion occurs in a concomitant pulsatile fashion.

Introduction

A cycle-dependent release pattern of oxytocin can be measured in the peripheral blood plasma in ewes (Schams, Lahlou-Kassi & Glatzel, 1981 ; Sheldrick & Flint, 1981 ; Webb, Mitchell, Falconer & Robinson, 1981) and in cows (Schams, 1983). Only basal concentrations of oxytocin were measurable after ovariectomy (Schams et ai, 1982). It has been demonstrated that the corpora lutea of sheep and cows (Wathes & Swann, 1982; Wathes, Swann, Birkett, Porter & Pickering, 1983) contain high concentrations of oxytocin which is secreted into the ovarian vein (Flint & Sheldrick, 1982), exhibiting pulsatile fluctuations parallel with progesterone (Walters, Schams & Schallen- berger, 1984; Schallenberger, Schams, Bullermann & Walters, 1984). In the hypothalamo- neurohypophysial system oxytocin and oxytocin-correlated neurophysin I are synthesized together as part of a precursor molecule, which is cleaved before exocytosis of the neurosecretory granules (Richter, 1983). Ovarian neurophysin has been demonstrated in extracts of bovine corpora lutea (Wathes et al, 1983). High peripheral plasma concentrations of immunoreactive neurophysin I/II were measured in ewes before luteolysis (Fairclough et al., 1980). Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access The aim of the present study was to provide further evidence for ovarian neurophysin and oxytocin secretion which might indicate biosynthesis within a common precursor molecule in the hypothalamo-neurohypophysial system and in the ovary.

Materials and Methods Animals and experiments Twenty-three regularly cyclic heifers (18-24 months of age, body weight 450-550 kg) and 5 non- lactating cows (2-4 years old, body weight 550-650 kg) of the local Braunvieh or Fleckvieh breeds were used. All animals were tethered indoors during the experiment. They were fed hay, corn silage and energy supplement and had access to water ad libitum. Blood sampling. All blood samples were placed into heparinized tubes and centrifuged at 4°C ; the plasma was stored at 20°C until analysis. Experiment I: short-term— secretion of oxytocin and immunoreactive neurophysin I. Vinyl medical grade cannulae were inserted during the mid-luteal phase into the posterior vena cava and into the abdominal aorta or jugular vein in 7 heifers 1 day before the start of the experiment. The detailed surgical procedure is given by Walters et al. (1984). Blood samples (10 ml) were collected simulta¬ neously from each vessel every 5-20 min. The release pattern of oxytocin and immunoreative neurophysin I was studied during series of frequent samplings of 2-5 h in 4 animals and after induction ofluteolysis with 500 pg cloprostenol (a PGF-2a analogue ; Estrumate : ICI, England) in 3 heifers. Experiment II: oxytocin and immunoreactive neurophysin I during and after superovulation. An i.m. injection of 1500-3100 i.u. PMSG (Anteron, 1000 i.u./mg: Schering, West Berlin) was given during the mid-luteal phase in 8 heifers and 2 cows. Five heifers were stimulated twice daily for 5 days with an i.m. injection of 5 mg of a crude porcine pituitary FSH preparation (Armour, U.S.A.). Six heifers served as controls. At 2 days after start of the treatments, luteolysis was induced by i.m. injection of 500 pg cloprostenol. Details of similar approaches to induce superovulation in cattle are given elsewhere (Schams et al., 1979b). Symptoms of oestrus were observed about 2 days after the prostaglandin in all experimental animals. Blood was collected every 12 h from jugular catheters from the day of oestrus (= Day 0) until Day 6 of the following cycle. The number of corpora lutea was then counted after exteriorizing the ovaries at laparoscopy.

Hormone determinations The concentrations of oxytocin, neurophysin I and progesterone were determined radio- immunologically as described by Schams et al. (1979a), Schams (1983), Legros (1976) and Hoffmann, Kyrein & Ender (1973), respectively. Oxytocin. The antiserum used (final dilution 1/80 000) was raised in a rabbit against synthetic oxytocin coupled to bovine thyroglobulin and showed no cross-reaction with related peptides such as lysine- or arginine- or pituitary (e.g. prolactin, growth hormone, luteinizing hormone, follicle-stimulating hormone or adrenocorticotrophic hormone). Oxytocin was extracted from blood plasma with SEP-PAK C18 cartridges (recovery was 71-3 ± 3-7%), which also allows concentration of the samples. Total incubation volume of the assay was 0-4 ml. Separation of free and bound hormone was performed by the charcoal method. The reference preparation (synthetic oxytocin, 200 i.u./ml, kindly donated by Sandoz, Basle, Switzerland) had the same potency as the biologically and immunologically well defined 4th International Standard from the National Institute for Biological Standards and Control, London. The lower limit of assay sensitivity was 0-25 pg (0125 pU)/tube. The inter-assay CV of pooled control plasma with low and high oxytocin concentrations varied between 10-8 and 17-3%. The intra-assay CV was 7-3 ± 3T%. Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access Immunoreactive neurophysin I. The antiserum against bovine oxytocin-neurophysin was raised in a rabbit and used at a final dilution of 1/8000. The cross-reaction with bovine neurophysin II (vasopressin-neurophysin) was <0-5% and with other bovine anterior pituitary hormones (prolactin, GH, LH, FSH and thyreotrophic hormone) or oxytocin and vasopressin was <0T%. Bovine neurophysin I was a gift from Dr P. Cohen, Institut Pierre et Marie Curie, Paris, France, and was used as reference preparation and for labelling. Neurophysin was labelled by the lactoper- oxidase method and purified on a Sephadex G-25 column. The assay was performed using 200 µ unextracted plasma. Separation of free and bound hormone was performed by the double-antibody method. Sensitivity of the assay was 0-5 ng/ml. The intra- and inter-assay CVs were 9% and 15-5%, respectively. Progesterone. The antiserum was raised in a rabbit against lloc-hydroxyprogesterone- hemisuccinate-bovine serum albumin. Endogenous oestrogens, androgens and corticosteroids exhibited no cross-reactivity with the antiserum. A weak cross-reaction with desoxycorticosterone (7-5%) was observed. The specificity was further ensured by an extraction step with light petroleum. The reagent blank was < 10 pg/ml. Average recovery with labelled progesterone added to bovine plasma was 78%. Sensitivity was 20 pg/ml. The mean intra- and inter-assay CVs were 8-5% and 14-8%, respectively.

Statistics

Differences between groups were tested with Student's / test and Wilcoxon range test. Significances are given only when proved by both methods.

Results Experiment I Immunoreactive neurophysin and oxytocin exhibited a pulsatile secretion pattern (Text-fig. 1) in the plasma of the vena cava : the concentrations of neurophysin were higher (P < 0-05) than those in the aorta or jugular vein (Text-fig. lb). After the injection of cloprostenol immunoreactive neurophysin I and oxytocin concentrations increased more (P < 0-005) and appeared first in the vena cava then in the jugular vein (Text-fig. 2). Initial concentrations remained lower in the peri¬ pheral circulation.

Experiment II The results of this experiment are summarized in Table 1. The animals receiving PMSG stimu¬ lation were classified according to the number of corpora lutea (Group III = 1-6 and Group IV = > 6) which was not dependent on the PMSG dose used. In control animals (Group I) bearing one corpus luteum, progesterone, oxytocin and immunoreactive neurophysin I concentrations increased throughout Days 4-6 of the cycle. After FSH stimulation (Group II) oxytocin, immunoreactive neurophysin and progesterone concentrations were significantly higher, even during oestrus, than in Group I animals. The oxytocin values were the highest recorded in the study. Progesterone concentrations of Group III animals were significantly different from those of Group I after Day 0 but not from those of Group II, although the number of corpora lutea was much lower than in Group II (14 compared with 3-6), thus indicating an additional luteotrophic potency of PMSG. This stimulating effect was not seen for oxytocin and immunoreactive neurophysin I concentrations. Oxytocin values in Group III were significantly higher than in Group I but lower than in Group II animals. After Day 0 immuno¬ reactive neurophysin I concentrations of Group III were significantly lower than those of Group II animals. In Group IV, progesterone concentrations were significantly higher for all days examined Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access 130b

110

Text-fig. 1. Concentrations of oxytocin (·-#) and immunoreactive neurophysin I ( - ) in the posterior vena cava and of immunoreactive neurophysin I in the abdominal aorta (O-CO- Blood samples were collected every 15 min on (a) Day 12 of the oestrous cycle in Heifer R and (b) Day 8 of the oestrous cycle in Heifer E.

1000

·

-60-20 0 20 40 60 -60-20 0 20 40 60 Minutes Text-fig. 2. Concentrations of oxytocin and immunoreactive neurophysin I from vena cava and jugular vein blood samples collected at 5-20-min intervals before and after an i .m. injection of a prostaglandin F-2a analogue (0 min) on Day 12 of the oestrous cycle in two heifers (a, b). Note the logarithmic scale of the ordinate. Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access Table 1. Mean ( ± s.e.m.) neurophysin I, oxytocin and progesterone concentrations in jugular vein blood at Days 0 (oestrus), 1-3 and 4-6 of the oestrous cycle in control cows and those induced to superovulate with FSH or PMSG (samples were taken every 12 h) Group II Group III Group IV Days Group I (superovulation with (superovulation with (superovulation with of (controls, = 6; FSH, = 5; PMSG, = 5; PMSG, = 5; Hormone cycle 1 CL/cow) 14 ± 2-9 CL/cow) 3-6 ±1-4 CL/cow) 16 ± 3-6 CL/cow) Neurophysin I 0 0-8±0-la l-6±0-3b l-4±0-3ab 1-1 ±0-2ab (ng/ml) 1-3 l-4±0-la 6-l±0-9b 2-2 ± 0-2" 3-8 ± 0-4c 4-6 21 ±0-2a 10-6 ±1-8" 3-2 ± 0-4a 13-2±l-4b Oxytocin 0 11 ±01a 8-5 ±1-1" 4-1 ±0-9c 4-8 ± 0-9° (pg/ml) 1-3 1-2 ±0-1" 21-5 ± 2-5b 7-4 ± 09e 11-7 ±1-3" 4-6 1-9 ±0-2» 40-1 ±4-3" 12-2 ±l-lc 27-1 ± 1-9" Progesterone 0 0-3 ±002a 0-5 ±005b 0-6±0-lab 1-1 ±0-2c (ng/ml) 1-3 0-5 ± 01a 1-5 ±0-3" l-0±0-2a-b 4-0 ± 0-7c 4-6 2-1 +0-2a 7-4 + l-lb 7-9 + l-2b 20-5 + 2-6c

For corresponding days of the cycle, values with different superscript letters are significantly different (P < 005). in comparison with those of the other groups. If compared with Group II (about the same number of corpora lutea) an additional luteotrophic effect after PMSG stimulation was again observed. Oxytocin and neurophysin I concentrations were significantly higher than in Groups I and III after Day 0 (lower number of corpora lutea).

Discussion

The results of Exp. I lead to the conclusion that oxytocin and immunoreactive neurophysin I are simultaneously secreted from the corpus luteum into the venous drainage. Due to dilution within the circulation, short-term changes cannot be detected in the peripheral venous or arterial circu¬ lation. In sheep, the large luteal cells are the source of luteal oxytocin (Rodgers, O'Shea, Findlay, Flint & Sheldrick, 1983). In the brain oxytocin-associated neurophysin is synthesized as part of the same precursor molecule as is oxytocin (Richter, 1983). Therefore, one might assume that similar interrelationships occur when oxytocin is secreted elsewhere. The similarity of release of ovarian oxytocin and immunoreactive neurophysin indicates a potentially similar synthesis mechanism in the corpus luteum as in the brain. Oxytocin and bovine neurophysin concentrations in plasma should also be balanced in about molar equivalence if hormone release were by exocytosis and oxy¬ tocin and neurophysin I were stored within the same granules in equimolar concentrations (as in the brain). As this did not happen in our experiments the following points might be considered. Variable relationships between release of oxytocin and immunoreactive neurophysin I were observed by Peeters et al. (1983) in bulls. One explanation could be that the actual blood level does not necessarily reflect the secretion pattern of both hormones, but rather represents a relative balance depending on secretion, dilution, degradation and sensitivities of the assays. Watkins, Moore, Flint & Sheldrick (1984) have demonstrated, in a similar study carried out in sheep, that the corpus luteum secretes both oxytocin and neurophysin I/II at a mean molar ratio of 1-2. As shown also by Walters et al. (1984) the measurement of ovarian hormones in peripheral circulation is not sufficiently sensitive. Blood collection from the vena cava is superior even when hormone concentrations are dependent on the position of the catheter tip relative to the outflow of the ovarian vessels, and we have demonstrated a close temporal relationship between the discontinuous secretion of immunoreactive neurophysin I and oxytocin in addition to that established between oxytocin and progesterone (Walters et al., 1984). The aim of Exp. II was to determine whether (1) the concentrations of oxytocin and immuno- Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access reactive neurophysin I were dependent on the number of corpora lutea or on the relative progester¬ one concentration and (2) the luteotrophic abilities of PMSG (Schams et ai, 1979b) were exerted on oxytocin and immunoreactive neurophysin secretion. LH, hCG and PMSG are well known as potent stimulators for progesterone synthesis. PMSG exerts luteotrophic activities even after 10 days due to its slow biological disappearance time in cattle (Menzer & Schams, 1979). Pituitary FSH and LH have a shorter disappearance time and so a luteotrophic action of FSH stimulation can be excluded. The present results showed that the circulating PMSG stimulated progesterone production of the corpora lutea to the extent that an average of 3-6 corpora lutea (Group III) secreted about the same amount of progesterone as 14 corpora lutea in Group II after previous FSH stimulation. This effect was even more pronounced when Groups IV and II were compared. A stimulatory effect of PMSG on oxytocin or immunoreactive neurophysin I was not observed. Ovine LH and FSH as well as hCG and PMSG showed a stimulatory effect on progesterone secretion but a lack of stimulation of oxytocin secretion from corpora lutea slices in vitro (unpublished data). The follicle seems to be able to synthesize oxytocin in cattle because concentrations in follicular fluid are much higher than those in the peripheral blood (Schams, Schallenberger, Walters, Meyer & Bullermann, 1983). After ovarian stimulation for superovulation (with an increased number of follicles), oxytocin and neurophysin concentrations were significantly higher when an increased number of follicles was present compared to the control group. Luteinization of follicles which may contribute to the higher oxytocin concentrations can not be excluded since the crude FSH prepara¬ tion was contaminated with LH, and PMSG exerts also LH-like activity. The highest concen¬ trations ofoxytocin during oestrus occurred after stimulation with pituitary FSH. Follicular growth and quality of follicles and potential quality of resulting corpora lutea might be different after FSH stimulation compared to PMSG stimulation. According to Elsden, Nelson & Seidel (1978) super¬ ovulation with FSH results in more corpora lutea, recovered ova and pregnancies than does super- ovulation with PMSG. It is concluded from these experiments that immunoreactive neurophysin I is always secreted concomitantly with oxytocin in major quantities from the corpus luteum but also in minor quanti¬ ties from the follicle. The amount of neurophysin and oxytocin depends on the number of corpora lutea or follicles present but not on the relative progesterone secretion.

This study was supported by the Deutsche Forschungsgemeinschaft, Bonn-Bad Godesberg.

References

Elsden, R.P., Nelson, L.D. & Seidel, G.E., Jr (1978) Menzer, Ch. & Schams, D. (1979) Radioimmunoassay for Superovulating cows with follicle stimulating hor¬ PMSG and its application to in-vivo studies. J. mone and pregnant mare's serum gonadotrophin. Reprod. Fert. 55, 339-345. Theriogenology 9, 17-26. Peeters, G., Legros, J.J., Piron-Bossuyt, C, Reynaert, R., Fairclough, R.J., Moore, L.G., McGowan, L.T., Peter¬ Vanden Driessche, R. & Vannieuwenhuyse, E. (1983) son, A.J., Smith, J.F., Tervit, H.R. & Watkins, W.B. Release of neurophysin I and oxytocin by stimulation (1980) Temporal relationship between plasma con¬ of the genital organs in bulls. J. Endocr. 99, 161-171. centrations of neurophysin I/II and 13,14-dihydro- Richter, D. (1983) Vasopressin and oxytocin are ex¬ 15-keto-prostaglandin F around luteolysis in sheep. pressed as polyproteins. Trends in Biochem. Sciences Prostaglandins 20, 199-208. 8, 278-281. Flint, A.P.F. & Sheldrick, EX. (1982) Ovarian secretion Rodgers, R.J., O'Shea, J.D., Findlay, J.K., Flint, A.P.F. of oxytocin is stimulated by prostaglandin. Nature, & Sheldrick, E.L. (1983) Large luteal cells the source Lond. 297, 587-588. of luteal oxytocin in the sheep. Endocrinology 113, Hoffmann, B., Kyrein, H.J. & Ender, M.L. (1973) An 2302-2304. efficient procedure for the determination of proges¬ Schallenberger, E., Schams, 1).. Bullennann, . & terone by radioimmunoassay applied to bovine peri¬ Walters, D.L. (1984) Pulsatile secretion of gonado¬ pheral plasma. Hormone Res. 4, 302-310. trophins, ovarian steroids and ovarian oxytocin Legros, J.J. (1976) Les neurophysines. In Recherches during prostaglandin-induced regression of the cor¬ Méthodologiques Expérimentales et Cliniques, pp. 21- pus luteum in the cow. J. Reprod. Fert. 71, 493- 77. Masson Pubi. Paris. 501. Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access Schams, D. (1983) Oxytocin determination by radio¬ Sheldrick, E.L. & Flint, A.P.F. (1981) Circulating immunoassay. III. Improvement to subpicogramme concentrations of oxytocin during the estrous cycle sensitivity and application to blood levels in cyclic and early pregnancy in sheep. Prostaglandins 22,631- cattle. Acta endocr., Copenh. 103, 180-183. 636. Schams, D., Schmidt-Polex. B. & Kruse, V. (1979a) Walters, D.L., Schams, D. & Schallenberger, E. (1984) Oxytocin determination by radioimmunoassay in Pulsatile secretion of gonadotrophins, ovarian cattle. Acta endocr., Copenh. 92, 258-270. steroids and ovarian oxytocin during the luteal phase Schams, D., Menzer, Ch., Schallenberger, E., Hoffmann, of the oestrous cycle in the cow. J. Reprod. Fert. 71, B., Prokopp, ., Hahn, . & Hahn, R. (1979b) Super¬ 479-491. ovulation beim Rind : Hormonprofile bei Stimulation Wathes, D.C. & Swann, R.W. (1982) Is oxytocin an mit Serumgonadotropin (PMSG) bzw. hypophy- ovarian hormone? Nature, Lond. 297, 225-227. särem FSH. Zuchthygiene 14, 11-25. Wathes, D.C, Swann, R.W., Birkett, S.D., Porter, D.G. Schams, D., Lahlou-Kassi, A. & Glatzel, P. (1981) & Pickering, B.T. (1983) Characterization of oxyto¬ Oxytocin concentrations in peripheral blood during cin, vasopressin, and neurophysin from the bovine the estrous cycle and after ovariectomy in two breeds corpus luteum. Endocrinology 113, 693-698. of sheep with low and high fecundity. Acta endocr., Watkins, W.B., Moore, L.G., Flint, A.P.F. & Sheldrick, Copenh., Suppl. 240, 49-50. E.L. (1984) Secretion of by the ovary in Schams, D., Lahlou-Kassi, A. & Glatzel, P. (1982) sheep. Peptides 5, 61-64. Oxytocin concentrations in peripheral blood during Webb, R., Mitchell, M.D., Falconer, J. & Robinson, J.S. the oestrous cycle and after ovariectomy in two (1981) Temporal relationship between peripheral breeds of sheep with low and high fecundity. J. plasma concentrations of oxytocin, progesterone and Endocr. 92, 9-13. 13,14-dihydro-15-keto prostaglandin F2„ during the Schams, D., Schallenberger, E., Walters, D.L., Meyer, oestrous cycle and early pregnancy in the ewe. H.H.D. & Bullermann, . (1983) Evidences for Prostaglandins 22, 443-453. ovarian oxytocin during the oestrous cycle in cattle. Acta endocr., Copenh. Suppl. 256, 261—Abstract. Received 19 April 1984

Downloaded from Bioscientifica.com at 09/29/2021 04:30:39AM via free access