Pediat. Res. 11: 627-630 (1977) Fetal neurohypophysis vasotocin pituitary Fetal Neurohypophyseal Vasopressin and Arginine Vasotocin in Man and Sheep

W. RONALD AND DELIJERT A. FISHER Departmmts of Endocrinology. Veterans Administration llospiwl, Long Beach, and llarbor General/lospillll, Torrance, Ca/ifomia, USA

Summary (biopotcncy 150 units/mg) were obtained from Schwarz/Mann Laboratories (27) and were used for standards and radioiodina­ Immunoreactive arginine vasopressin (A VP) and arginine va­ tion. sotocin (A VT) were IJU:tntitated in 15 of 17 hum:m fetal pitui­ tary glands early in gestation ( 1 1-19 weeks) and in 8 of 9 ovine RADIOIMI\IUNOASSA Y PROCEDURES fetal pituitary glands late in gestation (109-137 days). In 14 of 15 human fetal glands, A VT content exceeded that of A VP. Antisera to A VP and A VT were produced in New Zealand There was a significant rise of A VP (as a percentage of total White female rabbits, using vasopressin conjugated to A VP plus A VT content) with gestatioi1al age over the period of bovine thyroglobulin as previously described ( 12, 22). Synthetic 12-19 Wl'l'ks (P < 00.01). The ovine fetal glands demonstrated A VP and A VT were iodinated, using a modification of the a preponderance of A VP over A VT. The mean A VP and A VT method of Hunter and Greenwood (24). Assay procedures were content in the ovine glands was 5.7 ± 2.9 and 0.8 ± 0.2 mU/mg carried out as described earlier (21, 24), and bound and free gland weight, respectively, compared with the values in the hormones were separated exclusively by the double antibody human fetaiJlitiutaries, 0.8 ± 0.2 and 1.2 ± 0.2 mU/mg gland method (13). All assays were carried out in duplicate, using both weight, respectively. The relative percentage of A VP and A VT antiserum R-71 (specific for A VP) and antiserum R-70 (which in the ovine fetal pituitaries was 76.0 ± 9.6% and 23.9 ± 9.6%, measured both A VP plus A VT). A VT measurements arc ex­ respectively, as contrasted to the human fetal glands, 36.7 ± pressed as the· difference between values measured with the two 2.7% A VP and 63.3 ± 2.7% A VT. The preponderance of A VT antisera. Using antiserum R-71, A VT showed negligible cross­ over A VP in the early gestational age mammalian fetus may reactivity; the ratio of A VT:A VP at 50% binding was 350:3, represent a primative first step in molecular evolution of the and (OXY) showed no significant cross-reaction with neurohypophyseal peptides. labelctl antigen. Antiserum R-70 measured A VP and A VT with similar cross-reactivity while OXY showed a somewhat parallel Speculation response with cross-reactivity demonstrated at concentrations of OXY:AVT greater than 5:1. The coefficient of variation (CV) The results suggest that fetal neurohypophyseal vasotocin, for measurement of A VP (using R-71) was 7.1 % within assay present ine:trly fetal life, del·reases with gestational maturation, and 14 .9% between assay. Using R-70, the within assay CV was with a corresponding increase in vasopressin. Further studies arc 8.8% whereas between assay CV was 12.9%. needed in the same mammalian species periodically throughout gestation to l'onfirm this duonologic change which suggests that PROCESSING OF TISSUE SAMPLES ontogeny recapitulates ph)·logcny. Human fetal pituitaries (11-19 weeks of gestation) were ob­ tained following induced therapeutic abortion via dilatation and With the advent of sensitive radioimmunoassay (RIA) systems curettage or, in a few instances, saline lavage. No significant or for peptide and steroid hormones, much information has been unusual maternal drug medication nor maternal illness was doc­ obtained regarding the hypothalamoneurohypophyseal physiol­ umented. Gestational age of the fetus was confirmed by crown­ ogy of the developing mammalian fetus . A VP has been quanti­ rump measurement after the stantlards of Streeter (25). The tated within the fetal mammalian pituitary (1, 2, 26) and kinetic pituitary gland was manually dissected from surrounding tissue studies have suggested autonomous fetal secretion at rates within 10 min of fetal evacuation, immediately weighed, dried in greatly exceeding maternal production (23). Recent reports acetone for 18 hr, and then dehydrated in vacuo. Each gland was have suggested the presence of the nonapeptide A VT in the fetal pulverized and homogenized in I .0 ml phosphate (50 M)­ mammalian pituitary (19, 20, 26) as well as in the mammalian buffered saline ( 154 mM, pH 7 .5) at 4° in a small glass homoge­ pineal gland ( 6, 10, 14, 16). A VT also has been measured in the nizer. The homogenatcs were stored at -20° until thawed and cerebral spinal lluid of adult mammals, including man (7, 15, diluted in buffer for RIA measurement. 17). Normal fetal sheep pituitaries were obtained at 109-137 days We have recently reported RIA systems for measurement of of gestation, following surgical uterotomy under spinal anesthe­ A VP (24) and A VT, using a differential antiserum protocol sia in 1-4-year-old Columbia or Columbia-Suffolk ewes date­ (21). The present study reports results applying this method to bred to Columbia or Suffolk rams obtained from a local source. quantitation of A VP and A VT in early gestational age human Gestational age of the fetus was again confirmed by crown-rump fetal pituitary glands and late gestational age ovinc fetal pituitary measurement. The tissue was obtained within I 0 min after glands. delivery, weighed, processed, and stored as described above until RIA measurement.

MATERIALS AND METHODS RESULTS All chemicals used were reagent grade except as indicated. Table 1 lists the immunologic A VP and A VT activity in the Synthetic A VP (biopotency 300 units/mg) aml synthetic A VT human and sheep fetal pituitaries; values arc expressed both as a 627 628 SKOWSKY AND FISHER

Table I . A rgininc vasopressin (;\ V P) and vasotocin (;\ VT) immunoreactive A VP as a percentage of total A VP plus A VT content in fetal human and sheep pituitary glands inmilliunits of content. there is a highly significant correlation between a rising ;\VI' and A VT per mg glandll'cight, percentage of;\ VP and percentage of A VP and gestational age over the period of 12-19 A VT, and A V P:A VT ratios 1 weeks (correlation coefficient= O.XOI. P < (l.Ol); this is illus­ -- trated in Figure 2. Conversely. the percentage of AVT demon­ mU/mg wet wt % strates a statistically significant decrease over the similar time period in the human fetal pituitaries. AYP: The values for immunoreactive A VT obtained in the present Age. weeks AVP AVT AVP AVT AVT study arc subject to a possible error in interpretation. As we !Iuman fetal pituitary glands have previously reported, our differential radioimmunoassay II N.D. N.D. procedure demonstrates a cross-reactivity of OXY when com­ 12 lUi 1.6 27.9 72.1 0.39 bined A VP-A VT content is measured. Antiserum R-70 shows a 12 0.3 I.I 2I. I 78.9 0.26 displacement of tracer-labeled antigen by OXY when the ratio of I3 0.6 1.3 30.9 69.I 0.45 OXY:AVT is greater than 5: I. The present series of studies did not incorporate an initial chromatographic separation of OXY 13 N.D. N.D. and. if OXY was present in large amounts in the tissue extracts. I4 O.I 0.3 27.8 72.2 0.39 this would have produced falsely elevated immunoreactive A VT I4 0.2 0.5 29.8 70.2 0.43 measurements. Presently, there is no evidence to suggest the I5 0.6 1.1 36.1 63.9 0.57 presence of OXY in the human fetal neurohypophysis at this I5 O.I 0.5 I7.7 82.3 0.22 gestational time period. Moreover. it is improbable that, if I6 1.6 2.9 35.6 64.4 0.55 synthesized OXY is stored within the tissue. it would be present 16 0.1 0.1 45.4 54.6 0.83 I7 0.7 0.9 42.4 0.74 57.6 30 17 2.2 2.0 52.0 48.0 I .08

I8 1.8 2.1 46.0 54.0 0.85 1- z 25 I8 2.8 2.8 49.6 50.4 0.98 UJ (:==J AVP c:::J AVT I9 0.8 0.9 45.2 54.8 0.82 0' u :::> ** NON-DETECTABLE 19 O.I 0.1 43.5 56.5 0.77 >- E 20 a: 1- 1-.. > :; .. Mean 0.8 1.2 36.7 63.3 0.62 1- 0 - z I 5 SEM 0.2 0.2 2.7 2.7 0.01 Q. Ovine fetal pituitary glands t:' :I: 05 I20 2.9 0.3 89.5 I ().5 8.53 I24 5.0 O.I 98.6 I.4 68.49 II l 12 13 14 15 16 17 18 129 4.2 0.3 94.2 5.8 16.24 19

135 3.4 0.5 87.2 I 2.8 6.80 GESTATIONAL AGE (WEEKS) 135 0.4 0.5 42.6 57.4 0.74 Fig. I. Immunoreactive A VP and A VT in the human fetal pituitary 137 25.4 2.4 9 I .4 8.6 I0.62 glands of 11-19 weeks of gestation. Results arc expressed as milliunits of I37 0.4 1.2 25.0 75.0 0.33 hormone per mg gland weight. Neither hormone was measurable in the 11-weck gestational age gland nor in one of the I 3-week gestational age Mean 5.7 0.8 76.0 23.9 14.46 glands. SEM 2.9 0.2 9.6 9.6 7.93 1 ND: nondetectablc. 60 40 percentage of total A VP and A VT content and as a relative ratio of AVP:AVT. In the human fetal pituitaries examined, A VP 50 50 1- 1- and A VT were measureable in 15 of 17 glands. Neither hor­ > > > -' > > pituitary. It is also possible that the gestational age estimation • • studied. > •

The 120-137-day ovinc fetal pituitaries demonstrated a pre­ n. 50 > ponderance of A VP over A VT content in six of eight glands these pituitaries had the smallest actual quantity of immunologic >- AVP of all the ovinc tissue studied (0.4 mU/mg wet weight), n. ..J 10 f------L-.------much less than would be predicted from similar gestational-aged - • • • I • w glands. ... • • Since the ovinc pituitaries were obtained from sheep at 120- 05 • • 137 days of gestation, which represents the final 2 weeks of • • • pregnancy (term 145 days), no attempt was made at chronologie • analysis. The mean A VP and A VT content in the eight ovinc • fetal glands was 5.7 ± 2.9 and 0.8 ± 0.2 mU/mg gland weight. respectively, as compared with the values in the human fetal 01 pituitaries, 0.8 ± 0.2 and 1.2 ± 0.2 mU/mg gland weight. II 12 13 14 15 IG 17 18 19 respectively. The relative percentage of A VP and A VT content GESTATIONAL AGE (WEEKS) in the eight ovine fetal pituitaries was 76.0 ± 9.6% and 23.9 ± Fig. 3. Fetal pituitary AVP:AVT ratios in the human and (lVine 9.6%, respectively, as constratcd to the human fetal glands, glands. A ratio 1.0 36.7 ± 2.7% AVP and 63.3 ± 2.7% AVT. indicates greater A VP. The ratio in 14 of 15 human fetal glands is I .0. early in mammalian fetal development (as demonstrated in the human fetal pituitaries) with an absolute decrease in A VT and concomitant rise in A VP content late in fetal life (as seen in the relative to arginine vasopressin and that this peptide ratio de­ ovinc fetal pituitaries). The relative amounts of AVP and AVT creases over the chronologie time period of 12-19 weeks. In the in early and late gestation can be quantified by expressing the ncar term mammalian ovine fetal pituitary gland, arginine vaso­ values as an AVP/AVT ratio (Fig. 3). An AVP/A VT ratio< 1.0 pressin is the predominant vasopressor peptide present with only indicates a quantity of AVP less than AVT and an AVP/A VT small amounts of arginine vasotocin. However, as these data ratio > I .0 suggests a quantity of A VP greater than A VT. The reflect studies in two different mammalian species, it would be ratios obtained from 14 of 15 human fetal glands with immuno­ presumptuous to regard the results as representing a continuity logically measureablc hormones arc less than 1 .0 with a rise in of neurohypophyseal development throughout mammalian ges­ the ratio with advancing fetal age. This chronologie increase in tation. the A VP:A VT ratio represents both an actual increase in the The presence of arginine vasopressin in the fetal mammalian percentage of A VP content as well as a measure able decrease in pituitary gland has been documented by other investigators (I, percent A VT content. The mean A VP:A VT ratio of all the 2, 22, 26). Recently, Burton and Forsling (2) detected AVP human fetal glands is 0.62 ± 0.01. immunoreactivity in the fetal guinea pig neurohypophysis after The ovinc fetal pituitaries demonstrate an A VP:A VT ratio 20 days of gestation (term= 64 days). There was an increase in > 1 .0 in six of eight glands examined, representing, for the most pituitary content with increasing age. Perks and Vizsolyi (20) part, an absolute percentage increase in AVP. The mean demonstrated bioassayable A VP in the fetal pituitaries of sheep AVP:AVT ratio in the fetal sheep glands is 14.46 ± 7.93. and seals at midgestation. It is probable that the OXY content of the ovine fetal pituitary Recent reports have suggested the presence of A VT within the glands is elevated at this late gestational time period and the fetal pituitary gland as well as in the fetal pineal gland. Bioassay­ A VP:OXY ratio has been reported to be about 5 in fetal sheep able A VT has been identified chromatographically shortly after (20, 26). Some percentage of the immunoreactive AVT may, midterm in the neurohypophysis of the fetal seal, sheep, and indeed, represent OXY cross-reactivity. However, the A VP guinea pig (20, 26). These investigators, however, reported bio­ values measureable (by R-71 antisera) do not represent RIA logic activities for A VT and A VP in the term fetal ovine neuro­ displacement by elevated OXY levels and suggest valid interpre­ hypophysis which were 10 times greater than our results. In ad­ tations. Moreover, as AVP:AVT ratios never rose above 120:1, dition, Pavel (19) has reported biologic activities for A VT and it is unlikely that small quantities of A VT were obscured by large A VP in 20-22-week human fetal glands over 50-fold higher than amounts of A VP. Thus, the observation in the ovine fetal pitui­ our data suggests. It is uncertain why this discrepancy exists. taries that the A VP:A VT ratios arc relatively high late ingesta­ Although the above studies utilized an acetic acid extraction tion appears to be valid. (rather than our method of acetone drying), we have been unable to demonstrate any differences in immunologic activity of A VP or A VT with either method (unpublished observations). It DISCUSSION is possible that both methods may alter immunoreactivity with We have utilized a recently developed differential RIA system minor effects on biologic activity. The possibility exists, however for quantitiating immunoreactive A VP and A VT in the early small, of a prohormone or neurophysin-peptide complex that is gestational age human fetal neurohypophysis and in the late not recognized by our antisera yet still possesses activity under gestational age ovine fetal neurohypophysis. The results suggest the conditions of the bioassay. This might also account for the that early in human fetal development, the posterior pituitary ability of Pavel ( 19) to detect biologically active A VT (albeit no gland contains a preponderance of stored arginine vasotocin A VP) in human fetal neurohypophysis of 8-9 weeks of gesta- 630 SKOWSKY AND FISHER lion- an earlier gestational time for the presence of A VT than REFERENCES AND NOTES our studies would suggest. If the reason for discrepancy of I. Alexander, D. P., 13ritton, II. G., Forsling, 1\L L., Nixon, D. A., and absolute values of A VP ami A VT between other laboratories Radcliff, J. G.: The release of cortk·otrophin and vasopressin in the foetal and ours is due to a constant factor, then the ratios of pcptidcs sheep in response to J. PhysioL, 213: 31P (1'!71). could still be valid. 2. Durton, A. l\1., and forsling, l\1. L.: llormonc content of the neurohypophysis in foetal, new-born and adult guinea-pigs. J. Physiol., 22/: 6P (I \172). Pavel has demonstrated the ability to synthesize a vasotocin­ 3. 13urton, A. 1\1., Illingworth, D. V., Challis, J. R. G., and McNeilly, A. S.: like peptide in vitro of fetal human pineal tissue of 95-115 days Placental transfer of oxytocin in the guinea pig and its release during of gestation (19) (and as early as 55-65 days). These studies parturition. J. Endoerinol., 60: 4'!'! (1'!74). suggest that the fetal neurohypophysis functions as a neurosecre­ 4. Chard, T.,13oyd, N. R.JI., Forsling,l\1. I., McNeilly, A. S .. and Landon, J.: The development of a radioimmunoassay for oxytocin: The extraction of tory organ as well as a storage site for A VT. The presence of oxytocin from plasma, anU its measurement during parturition in human and ependymal cells in the mammalian fetal posterior pituitary that goat blood. J_ Endocrinol., 48: 223 (1'!70). arc morphologically similar to those in the fetal pineal gland may 5. Chard, T.,Jiudson, C N., Edwards, C. R. W., and 13oyd, N. R. 11.: Release represent the site of synthesis of A VT. The fetal neurohypophy­ of oxytocin and vasopressin of the human foetus during labour. Nature, 234: 352 (1\171). seal ependymal cells (but not those in the pineal) disappear after 6. Chessman, D. W.: Structural elucidation of a gonadotropin-inhibiting sub­ birth, correlating with the disappearance of A VT from the pos­ stance from the bovine pineal gland. Biochim. 13iophys. Acta, 207: 247 terior pituitary gland of the adult mammal. (1'!70). The values for immunoreactive A VT in the present study arc 7. Coculescu, 1\1., and Pavel, S.: Arginine vasotocin-like activity of cerebrospinal fluid. J. Clin. EndocrinoL Metab., 31: 369 (1'!70). valid only if OXY is present in ratios less than 4:1. Studies in the 8. llcnderson, I. W., and Wales, N. A.l\1.: Renal diuresis and antidiurcsis after fetal guinea pig suggest undetectable levels of OXY until after injections of arginine vasotocin in the fresh water eel. J. Endocrinol., 61: 46 days (term= 64 days) (2). Data from fetal sheep, seals and 41!7 (I \174). pigs suggest extremely low OXY levels till well after midterm. 9. McNeilly, A. S., Martin, 1\1. J., Chard, T., and Hart, 1.: Simultaneous release of oxytocin and neurophysin during parturition in the goat. J. EndocrinoL, Indeed, even at parturition, the OXY content of the fetal neuro­ 52: 213 (1972). hypophysis is far below adult levels: 7-10% of adult values in 10. Mileau, S. 1\1., Pavel, S., and Neacsu, C: Biological and chromatographic the guinea pig (2), 30% in the pig, 35% in the seal, and 8% in characterization of a polypeptide with pressor and oxytocic activities iso­ the sheep (20, 26). Indeed, there is evidence that the increased lated from bovine pineal gland. Endocrinology, 72: 563 (I '!63). II. Morel, E., and Jard, S.: Actions and functions of the neurohypophysial levels of OXY in fetal blood at parturition in the goat ( 4, 9), man hormones and related peptides in lower vertebrates. In: Neurohypophysial (5), and guinea pig (3) reflect, to a major degree, placental hormones and similar poly-peptides. Jlandbuch der Experimcntellen Phar­ transport of maternal OXY. It therefore seems unlikely that makologie, VoL 23, p. 653 (Springer-Verlag, 13crlin, I

Copyright© 1977 International Pediatric Research Foundation, Inc. Printed in U.S.A.