Clinical Endocrinology (1990) 32, 307-3 13
THE EFFECT OF OXYTOCIN INFUSION ON ADENOHYPOPHYSEAL FUNCTION IN MAN
S. R. PAGE, V. T. Y. ANG, R. JACKSON, A. WHITE, S. S. NUSSEY AND
J. S . JENKINS
Division of Biochemical Medicine, Department of Cellular and Molecular Sciences, St. George’s Hospital Medical School, London and University of Manchester, Department of Medicine, Section of Clinical Biochemistry, Hope Hospital, Salford M6 BHD, UK (Received 14 April 1989; returnedfor revision 19 May 1989;finally revised 3 August 1989; accepted 22 August 1989)
SUMMARY
The responses of the adenohypophyseal hormones adrenocorticotrophin (ACTH), growth hormone (GH), thyroid stimulating hormone (TSH), prolactin, luteinizing hormone (LH) and follicle stimulating hormone (FSH) to sub-maximal doses of hypothalamic releasing factors were studied in six lean male volunteers (age 23-35 years) with and without infusions of oxytocin (OXT). OXT infusion (mean plasma concentration 133.6f 2.6 pmol/l) completely inhibited the plasma ACTH responses to corticotrophin releasing hormone (CRH) (saline, peak increment ACTH 1.61 f0.75 pmol/l; OXT, peak increment ACTH -0.04 f0.28 pmol/l; P < 0.05). OXT infusion had no significant effect on the GH response to growth hormone releasing hormone (GHRH), the TSH and prolactin responses to thyrotrophin releasing hormone (thyroliberin, TRH) or the LH and FSH responses to gonadotrophin releasing hormone (luteoliberin, GnRH). The data support a role for OXT in the modulation of ACTH secretion in man. OXT and arginine vasopressin (AVP) have been measured in rat hypophyseal portal blood in concentrations approximately fiftyfold higher than those in peripheral blood (Gibbs, 1984a); even higher concentrations of OXT and AVP have been reported in hypophyseal portal blood of primates (Plotsky, 1987). Both hormones are probably secreted from the external zone of the median eminence where OXT and AVP-containing nerve terminals have been identified near the primary portal capillaries (Vandesande & Dierickx, 1975). AVP has a generally accepted role in modulating adenohypophyseal function in man. It stimulates the secretion of GH (Gagliardino et al., 1967) and ACTH (Staub et al., 1973) and markedly potentiates the activity of CRH both in vitro (Gilles et al., 1982) and in vivo (Orth et al., 1985). Correspondence: Dr S. R. Page, Division of Biochemical Medicine, Department of Cellular and Molecular Sciences, St. George’s Hospital Medical School, Cranmer Terrace, London SW17 O R E , UK.
307
308
S. R. Page et al.
However, the evidence that OXT modulates adenohypophyseal function, particularly with regard to ACTH secretion, is controversial. Studies using rats in vitro (Antoni et al., 1983) and in vivo (Gibbs 1984b, 1986) suggest a stimulatory effect of OXT on ACTH secretion. In man however, Legros and co-workers reported an inhibitory effect of OXT on basal ACTH concentrations and also on the ACTH responses to insulin-induced hypoglycaemia and exogenous AVP (Legros et al., 1982, 1984). However, others have been unable to confirm these observations (Lewis & Sherman, 1985; Nussey et al., 1988). The reasons for these discrepancies remain unclear. There is evidence in man that OXT affects the secretion of prolactin (Coiro et al., 1987) and GH (Chiodera et al., 1984a). Furthermore, in-vitro studies in rats (Evans & Catt, 1987) suggest that OXT is able to stimulate gonadotrophin secretion. Hypoglycaemia is a potent stimulus to the secretion of a number of adenohypophyseal hormones and may overcome a weaker modulating effect of OXT. We reasoned that any effect of OXT on adenohypophyseal function may be more evident if the adenohypophysis was sub-maximally stimulated by hypothalamic releasing factors. We have therefore studied the effect of OXT infusion on the adenohypophyseal hormone responses to combined stimulation with appropriate doses of hypothalamic releasing hormones.
SUBJECTS AND METHODS All subjects gave informed consent and the study received approval from the Local Ethical Committee. Six healthy male volunteers (age 22-35 years), all within 10% ideal body weight, were recruited from Medical School Staff and Students. Following an overnight fast tests were performed on recumbent subjects using sub-maximal doses of hypothalamic releasing factors with or without OXT infusion. Tests were performed in random order with intervals of 2 weeks between tests. Indwelling cannulae were inserted into the left antecubital vein for blood sampling and a right forearm vein for administration of infusions. Patency of the sampling cannula was maintained with small volumes of isotonic heparinized saline. Baseline blood samples were taken 45 min later (0915 h for every test). An infusion of either OXT (1 11 mU/min for 4 h, Syntocinon, Sandoz Pharmaceuticals, Feltham, Middlesex, UK; conversion 1U= 2 nmol) or an equivalent volume of isotonic saline was commenced immediately following basal blood sampling. A further blood sample was taken at time 0 (0930 h in every case) and then CRH (Peninsula Laboratories, St Helens, Merseyside, UK) 50 pg, GHRH (1-29) NH2 50 fig (Kabi-vitrum Ltd, Uxbridge, Middlesex, UK), TRH 100 pg and GnRH 50 pg were injected intravenously. Where necessary, releasing factors were reconstituted with isotonic saline immediately prior to administration. Further blood sampleswere taken at 10-minintervals during the first hour and at 20-min intervals for the subsequent 3 h. All subjects experienced mild facial flushing following the bolus injection of the hypothalamic releasing factors. AVP, at a dose known to potentiate the ACTH response to CRH (10 U intramuscularly) was co-administered with releasing factors to three subjects during saline and OXT infusions to evaluate further the effect of OXT on the ACTH response to this combined stimulus. During all studies, pulse rate was measured by palpation and blood pressure was measured using a sphygmomanometer immediately prior to each blood sample.
OXT and adenohypophysealfunction
309
-- 200 \ ; i
1601
120
,
20
60
100
140
180
220
Time (min)
Fig. 1. Peripheral plasma concentrationsof 0,OXT and 0 ,AVP. Infusion ofOXT (1 11 mU/min) was commencedat time - 15 min and AVP (IOUi.m.)was administered at time 0 min. Results are expressed as meankSEM.
Blood samples were collected into chilled plastic tubes containing lithium heparin or EDTA (haematocrit). After mixing samples were centrifuged at 4°C for 10 min at 2000g. Aliquots of fresh plasma were removed for sodium and osmolality estimations and the remainder was snap frozen and stored at - 20°C prior to assay. Plasma osmolality was measured by freezing-point depression using an Advanced Digimatic Osmometer Model 3D. Haematocrit was measured in duplicate by a microcentrifuge technique. OXT and AVP were extracted and assayed as previously described (Jenkins et al., 1984). The inter and intra-assay coefficients of variation (CV) (at the 1-10 fmol/tube level) were 12.8 and 6.5% for OXT and 13 and 7.5% for AVP, with detection limits of 0.6 fmol/tube for OXT and 0.3 fmol/tube for AVP. Plasma GH, TSH, prolactin, LH and FSH were assayed by established in-house radioimmunoassays with an inter and intra-assay CV of 4-10% for all assays and detection limits of 0-5,0.5,20,1.0and 0.5 mU/1respectively. Plasma ACTH was assayed using a two-site immunoradiometric assay with a detection limit of 0.9 pmol/l and a CV of less than 10% over the range 4-220 pmol/l (White et al., 1987). Assays were performed by persons unaware of the experimental protocol and samples from each subject were assayed together. Results were plotted as changes from basal hormone concentrations. Statistical analysis was performed using analysis of variance (ANOVA) Tukey’s Honestly Significant Difference Test (Tukey, 1949). Results are plotted as mean If:SEM.
RESULTS
Plasma OXT There were no significant changes in pulse, blood pressure, plasma sodium, osmolality or haematocrit over the duration of any study. Plasma OXT concentrations rose from 2.7f0.3 to 133.6f2.6 pmol/l (mean+SEM of 42 estimations in six subjects) (Fig. 1). Plasma AVP rose from 4.6 & 2.1 pmol/l to a peak of 149.8 f9.5 pmol/l (mean k SEM of six estimations in three subjects) following intramuscular injection (Fig. 1).
S. R. Page et al.
310
-10
I
-20
I
I
I
I
I
I
I
0
20
40
60
80
100
120
Time (min)
Fig. 2. a, Change from basal concentrations of plasma ACTH (meanfSEM, n = 6 ) during infusion of 0, saline and @, OXT. OXT inhibited the ACTH response to 50 fig CRH (saline,peak ACTH 1.61 k0.75pmol/l; OXT peak ACTH -044k0.28 pmol/l, *P
THE EFFECT OF OXYTOCIN INFUSION ON ADENOHYPOPHYSEAL FUNCTION IN MAN
S. R. PAGE, V. T. Y. ANG, R. JACKSON, A. WHITE, S. S. NUSSEY AND
J. S . JENKINS
Division of Biochemical Medicine, Department of Cellular and Molecular Sciences, St. George’s Hospital Medical School, London and University of Manchester, Department of Medicine, Section of Clinical Biochemistry, Hope Hospital, Salford M6 BHD, UK (Received 14 April 1989; returnedfor revision 19 May 1989;finally revised 3 August 1989; accepted 22 August 1989)
SUMMARY
The responses of the adenohypophyseal hormones adrenocorticotrophin (ACTH), growth hormone (GH), thyroid stimulating hormone (TSH), prolactin, luteinizing hormone (LH) and follicle stimulating hormone (FSH) to sub-maximal doses of hypothalamic releasing factors were studied in six lean male volunteers (age 23-35 years) with and without infusions of oxytocin (OXT). OXT infusion (mean plasma concentration 133.6f 2.6 pmol/l) completely inhibited the plasma ACTH responses to corticotrophin releasing hormone (CRH) (saline, peak increment ACTH 1.61 f0.75 pmol/l; OXT, peak increment ACTH -0.04 f0.28 pmol/l; P < 0.05). OXT infusion had no significant effect on the GH response to growth hormone releasing hormone (GHRH), the TSH and prolactin responses to thyrotrophin releasing hormone (thyroliberin, TRH) or the LH and FSH responses to gonadotrophin releasing hormone (luteoliberin, GnRH). The data support a role for OXT in the modulation of ACTH secretion in man. OXT and arginine vasopressin (AVP) have been measured in rat hypophyseal portal blood in concentrations approximately fiftyfold higher than those in peripheral blood (Gibbs, 1984a); even higher concentrations of OXT and AVP have been reported in hypophyseal portal blood of primates (Plotsky, 1987). Both hormones are probably secreted from the external zone of the median eminence where OXT and AVP-containing nerve terminals have been identified near the primary portal capillaries (Vandesande & Dierickx, 1975). AVP has a generally accepted role in modulating adenohypophyseal function in man. It stimulates the secretion of GH (Gagliardino et al., 1967) and ACTH (Staub et al., 1973) and markedly potentiates the activity of CRH both in vitro (Gilles et al., 1982) and in vivo (Orth et al., 1985). Correspondence: Dr S. R. Page, Division of Biochemical Medicine, Department of Cellular and Molecular Sciences, St. George’s Hospital Medical School, Cranmer Terrace, London SW17 O R E , UK.
307
308
S. R. Page et al.
However, the evidence that OXT modulates adenohypophyseal function, particularly with regard to ACTH secretion, is controversial. Studies using rats in vitro (Antoni et al., 1983) and in vivo (Gibbs 1984b, 1986) suggest a stimulatory effect of OXT on ACTH secretion. In man however, Legros and co-workers reported an inhibitory effect of OXT on basal ACTH concentrations and also on the ACTH responses to insulin-induced hypoglycaemia and exogenous AVP (Legros et al., 1982, 1984). However, others have been unable to confirm these observations (Lewis & Sherman, 1985; Nussey et al., 1988). The reasons for these discrepancies remain unclear. There is evidence in man that OXT affects the secretion of prolactin (Coiro et al., 1987) and GH (Chiodera et al., 1984a). Furthermore, in-vitro studies in rats (Evans & Catt, 1987) suggest that OXT is able to stimulate gonadotrophin secretion. Hypoglycaemia is a potent stimulus to the secretion of a number of adenohypophyseal hormones and may overcome a weaker modulating effect of OXT. We reasoned that any effect of OXT on adenohypophyseal function may be more evident if the adenohypophysis was sub-maximally stimulated by hypothalamic releasing factors. We have therefore studied the effect of OXT infusion on the adenohypophyseal hormone responses to combined stimulation with appropriate doses of hypothalamic releasing hormones.
SUBJECTS AND METHODS All subjects gave informed consent and the study received approval from the Local Ethical Committee. Six healthy male volunteers (age 22-35 years), all within 10% ideal body weight, were recruited from Medical School Staff and Students. Following an overnight fast tests were performed on recumbent subjects using sub-maximal doses of hypothalamic releasing factors with or without OXT infusion. Tests were performed in random order with intervals of 2 weeks between tests. Indwelling cannulae were inserted into the left antecubital vein for blood sampling and a right forearm vein for administration of infusions. Patency of the sampling cannula was maintained with small volumes of isotonic heparinized saline. Baseline blood samples were taken 45 min later (0915 h for every test). An infusion of either OXT (1 11 mU/min for 4 h, Syntocinon, Sandoz Pharmaceuticals, Feltham, Middlesex, UK; conversion 1U= 2 nmol) or an equivalent volume of isotonic saline was commenced immediately following basal blood sampling. A further blood sample was taken at time 0 (0930 h in every case) and then CRH (Peninsula Laboratories, St Helens, Merseyside, UK) 50 pg, GHRH (1-29) NH2 50 fig (Kabi-vitrum Ltd, Uxbridge, Middlesex, UK), TRH 100 pg and GnRH 50 pg were injected intravenously. Where necessary, releasing factors were reconstituted with isotonic saline immediately prior to administration. Further blood sampleswere taken at 10-minintervals during the first hour and at 20-min intervals for the subsequent 3 h. All subjects experienced mild facial flushing following the bolus injection of the hypothalamic releasing factors. AVP, at a dose known to potentiate the ACTH response to CRH (10 U intramuscularly) was co-administered with releasing factors to three subjects during saline and OXT infusions to evaluate further the effect of OXT on the ACTH response to this combined stimulus. During all studies, pulse rate was measured by palpation and blood pressure was measured using a sphygmomanometer immediately prior to each blood sample.
OXT and adenohypophysealfunction
309
-- 200 \ ; i
1601
120
,
20
60
100
140
180
220
Time (min)
Fig. 1. Peripheral plasma concentrationsof 0,OXT and 0 ,AVP. Infusion ofOXT (1 11 mU/min) was commencedat time - 15 min and AVP (IOUi.m.)was administered at time 0 min. Results are expressed as meankSEM.
Blood samples were collected into chilled plastic tubes containing lithium heparin or EDTA (haematocrit). After mixing samples were centrifuged at 4°C for 10 min at 2000g. Aliquots of fresh plasma were removed for sodium and osmolality estimations and the remainder was snap frozen and stored at - 20°C prior to assay. Plasma osmolality was measured by freezing-point depression using an Advanced Digimatic Osmometer Model 3D. Haematocrit was measured in duplicate by a microcentrifuge technique. OXT and AVP were extracted and assayed as previously described (Jenkins et al., 1984). The inter and intra-assay coefficients of variation (CV) (at the 1-10 fmol/tube level) were 12.8 and 6.5% for OXT and 13 and 7.5% for AVP, with detection limits of 0.6 fmol/tube for OXT and 0.3 fmol/tube for AVP. Plasma GH, TSH, prolactin, LH and FSH were assayed by established in-house radioimmunoassays with an inter and intra-assay CV of 4-10% for all assays and detection limits of 0-5,0.5,20,1.0and 0.5 mU/1respectively. Plasma ACTH was assayed using a two-site immunoradiometric assay with a detection limit of 0.9 pmol/l and a CV of less than 10% over the range 4-220 pmol/l (White et al., 1987). Assays were performed by persons unaware of the experimental protocol and samples from each subject were assayed together. Results were plotted as changes from basal hormone concentrations. Statistical analysis was performed using analysis of variance (ANOVA) Tukey’s Honestly Significant Difference Test (Tukey, 1949). Results are plotted as mean If:SEM.
RESULTS
Plasma OXT There were no significant changes in pulse, blood pressure, plasma sodium, osmolality or haematocrit over the duration of any study. Plasma OXT concentrations rose from 2.7f0.3 to 133.6f2.6 pmol/l (mean+SEM of 42 estimations in six subjects) (Fig. 1). Plasma AVP rose from 4.6 & 2.1 pmol/l to a peak of 149.8 f9.5 pmol/l (mean k SEM of six estimations in three subjects) following intramuscular injection (Fig. 1).
S. R. Page et al.
310
-10
I
-20
I
I
I
I
I
I
I
0
20
40
60
80
100
120
Time (min)
Fig. 2. a, Change from basal concentrations of plasma ACTH (meanfSEM, n = 6 ) during infusion of 0, saline and @, OXT. OXT inhibited the ACTH response to 50 fig CRH (saline,peak ACTH 1.61 k0.75pmol/l; OXT peak ACTH -044k0.28 pmol/l, *P
