0013-7227/90/1264-2095$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society
Vol. 126, No. 4 Printed in U.S.A.
Oxytocin- and Vasopressin-Binding Sites in the Rat Uterus: Competition Binding and Inhibitory pA2 Studies with Oxytocin and Oxytocin Antagonists* W. Y. CHAN, LIN CAO, PATRICIA S. HILL, AND VICTOR J. HRUBY Department of Pharmacology, Cornell University Medical College (W. Y.C, L.C.), New York, New York 10021; and the Department of Chemistry, University of Arizona (P.S.H., V.J.H.), Tucson, Arizona 85721
ABSTRACT. Recent reports have presented evidence suggesting that there are distinct oxytocin (OT) and vasopressin (VP) receptors in the human and rabbit myometrium. In this study we have investigated whether OT and arginine vasopressin (AVP) activate the same or two different receptor systems in the rat uterus in producing their uterotonic action and whether the myometrial OT/VP receptors are similar to the V^ receptors in the vascular smooth muscle cells. We compared the doseresponse characteristics of OT and AVP by the in vitro cumulative dose-response curve technique. We determined the ligandreceptor binding characteristics of [3H]OT and [3H]AVP on uterine membrane fractions from nonpregnant and pregnant rats. Specific OT antagonists were used in competition receptor binding assays and in antioxytocic pA2bioassays against OT and AVP to determine whether OT antagonists can discriminate between OT- and AVP-binding sites in the myometrium. We also compared the in vitro antioxytocic (OT receptor-mediated action) and the in vivo antivasopressor (Vj receptor-mediated
T
action) potencies of a series of six OT antagonists. Our results show that OT- and AVP-binding sites in the nonpregnant rat uterus have similar binding characteristics and cannot be distinguished by the dose-response study, radioligand receptor binding assays, or OT antagonists in the competition binding and pA2 assays. However, in the term pregnant parturient uterus, the two binding sites can be clearly differentiated. OT receptor density, but not AVP, was markedly increased at term pregnancy. All six OT antagonists studied in this investigation were more potent in antagonizing the uterotonic response to OT than the vasopressor response to AVP. The antioxytocicrantivasopressor potency ratios, however, were different between the antagonists, ranging from nearly equal (0.91) to low (0.1). The results above suggest that there are distinct OT- and AVP-binding sites in the rat myometrium. The myometrial OT/ AVP receptors are similar to but not the same as the Vx receptors in the vascular smooth muscle cells. (Endocrinology 126: 20952101, 1990)
Recently, we have developed a new series of highly potent and long-acting OT antagonists (10, 11). In this study we investigated whether these specific OT antagonists can discriminate between OT- and VP-binding sites in the rat myometrium and whether these antagonists can delineate the receptors that mediate the uterotonic and vasopressor responses of OT and VP. Competition bindings between OT antagonist and [3H]OTbinding sites and [3H]AVP-binding sites were studied. In vitro antioxytocic assays and in vivo antivasopressor assays were performed. Preliminary data of some of this work have been reported (12).
HE NEUROHYPOPHYSIAL hormones oxytocin (OT) and vasopressin (VP) both stimulate myometrial contractions. OT receptors in the uterus have been characterized (1-4). It has also been shown that the contractile response to OT correlates closely with OT receptor concentrations (5, 6). However, it has not been established whether the uterotonic action of VP is mediated via VP receptors or via the cross-reactivity of VP on the OT receptors. A number of studies have suggested that there are distinct OT and VP receptors in the myometrium and smooth muscle cells in the male reproductive tract, and that the myometrial VP-binding sites are of the Vx subtype, which mediates the vasopressor response of VP, and not the V2 subtype, which mediates the antidiuretic response (7-9).
Materials and Methods
Received September 18,1989. Address requests for reprints to: Dr. W. Y. Chan, Department of Pharmacology, Cornell University Medical College, 1300 York Avenue, New York, New York 10021. * This work was supported in part by USPHS Grants HD-20839 (to W.Y.C.) and AM-17420 (to V.J.H.).
Wistar rats were used in this investigation. Male rats (200220 g), female rats (180-200 g), or dated pregnant rats were purchased from Hilltop Animals (Scottsdale, PA). OT and VP receptor binding assays Receptor binding assays were carried out on uterine plasma membrane fractions by the method described by Soloff and
2095
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2096
Swartz (1). Uterine horns from nonpregnant rats [nonestrous or diethylstilbestrol (DES)-treated] and day 23 term pregnant rats (fetal tissues removed) were homogenized in 3 vol ice-cold 10 mM Tris-HCl buffer, pH 7.4, containing 2 mM EDTA-2Na and 0.5 mM dithiothreitol. The homogenate was centrifuged at 1,000 X g for 20 min at 4 C. The resultant supernatant was centrifuged at 100,000 X g for 60 min at 4 C. The EDTA in the preparation buffer was present to dissociate OT from its binding sites and ensure the availability of all OT-binding sites for ligand-receptor binding in the binding assay (13). The pellet that contained the plasma membrane fraction was washed three times with assay buffer (50 mM Tris-maleate buffer, pH 7.6, containing 10 mM MnCl2 and 1% gelatin) to remove EDTA. This membrane-microsomal fraction has been shown to be richer in OT-binding sites than the 10,000 x g and 20,000 x g fractions. The washed pellet was then resuspended in assay buffer to yield a protein concentration of 20-25 mg/ml, determined by the method of Lowry et al. (14). The membranemicrosomal suspension was stored in liquid nitrogen until used for receptor binding assays. For binding assays, the crude membrane preparation was diluted to 10 mg/ml. The radioligand receptor binding assay was carried out with 1.0 mg/ml membrane protein, incubated with six concentrations of [3H]OT (0.5-15 nM) or [3H]AVP (0.5-20 nM) in a final volume of 250 fi\. The incubation was carried out at 22 C for 60 min and in the absence and presence of excess (1000-fold) unlabeled OT or AVP to distinguish nonspecific and specific binding. The incubation was terminated by adding 5 ml ice-cold gelatin-free assay buffer. Free and bound labeled ligands were separated by filtration through Whatman GF/F glass microfiber filters (Clifton, NJ), using a Brandel M-24R cell harvester (Gaithersburg, MD). The filter was rinsed with 2 ml ice-cold gelatin-free assay buffer. The filtration and rinse were completed in 5 sec. The filter was then dried, and the content of radioactivity was determined by liquid scintillation spectrometry. All assays were performed in duplicate. The binding data were evaluated by Scatchard analysis (15). A computer-assisted ligand binding data analysis program was used to calculate the dissociation constant (Kd) and the saturable binding sites (Bmax). Competition binding assays were carried out in the presence of 0.5 nM labeled ligands, [ 3 H]0T or [3H]AVP, and increasing concentrations of unlabeled ligands (1-20 nM for OT and 5100 nM for OT antagonist). The IC50 value of each peptide (concentration that causes 50% displacement of the labeled ligand from binding sites) was determined from competition binding data. In vitro oxytocic assays In vitro oxytocic assays were performed on isolated uteri from nonpregnant rats pretreated with DES (50 Mg» sc) on the day before assay, with the use of Mg2+-free van Dyke-Hastings solution (16). The antioxytocic potencies of OT antagonists against OT- or AVP-induced contractile response were determined in this in vitro system by the pA2 method of Schild (17). The pA2 is the negative log molar concentration of the antagonist that will reduce the response of 2X units of the agonist to
Endo • 1990 Vol 126 • No 4
X units of the agonist. Oxytocic activities of OT and AVP, ED50 and intrinsic activity, were determined by the cumulative dose-response curve technique, as described by van Rossum (18). Antivasopressor assays Antivasopressor potencies of OT antagonists against AVP were determined in urethane-anesthetized and phenoxybenzamine-treated male rats. Blood pressure was monitored via a carotid arterial catheter, and peptides were injected via a jugular venous catheter. The effective dose for the pA2 assay was determined; this was the dose of the antagonist that would reduce the pressor response of 2X units of AVP to that of X unit of AVP. Assuming a volume of distribution of 67 ml/kg, the effective dose is converted to molar concentration for the calculation of pA2 value as in the in vitro pA2 assay. In vivo pA2 values thus obtained are only approximate estimates, since the molar concentrations of the antagonist cannot be precisely determined. Statistical analysis of data All data were expressed as the sample mean ± SEM and analyzed by analysis of variance. Significant differences between sample means were analyzed by Student's t test, both tails at P 0.05 level. Materials The OT antagonists used in this study were synthesized in our laboratories. They were [Pen\Phe(Me)2,Thr4,Orn8]OT, [Pen1,Phe2,Thr4,A3-4-Pro7]OT, [Pen\D-Phe2,Thr4,Thr5,Om8]OT, [Pen1,D-Phe2,Thr4,Leu5,Orn8]OT, [Pen1,D-Phe2,Thr4,Asp5,Orn8] OT, and [PenSD-Phe'.Thr'.Ty^Orn^OT. OT and AVP were purchased from Peninsula Laboratories (Belmont, CA). Multilabeled [3H]OT (34-38 Ci/mmol) and [3H]AVP (68 Ci/mmol) were purchased from New England Nuclear (Boston, MA). Radiochemical purities were 97-98% for [ 3 H]0T and 99% for [3H]AVP, as certified by the manufacturer. TLC performed in our laboratory showed that greater than 90% of the radioactivity migrated with the OT or AVP standard.
Results In vitro oxytocic activity of OT and A VP AVP is 20-30 times less potent than OT in oxytocic activity. However, it is a full agonist. Figure 1 shows the in vitro oxytocic dose-response curves of OT and AVP. Cumulative dose-response curves were carried out in eight uterine preparations from four DES-treated rats. In each horn, at least one OT response curve and one AVP response curve were obtained. The maximal response to OT was taken as 100%. All other responses to OT or AVP were expressed as a percentage of this maximum. The OT curve and the AVP curve were parallel and had the same maximum. The ED50 for OT was 6.9 x 10"6 nM, and that for AVP was 1.8 X 10~4 /xM. The ratio of ED50 AVP/ED50 OT was 26.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2097
TABLE 1. OT- and AVP-binding sites in nonpregnant and pregnant rat uterine membranes
100
Bmax (fmol/mg membrane protein) Ligand w 90
§
[3H]OT [3H]AVP
80
Nonpregnant (nonestrous) 66 ± (1.01 ± 65 ± (0.89 ±
11 0.14) 18 0.17)
Nonpregnant (DES-treated) 251 ± 24 (1.40 ± 0.18) 242 ± 5 (0.83 ± 0.07)
Term pregnant (in labor) 356 ± (1.99 ± 183 ± (1.47 ±
17° 0.36) 16" 0.19)
Values are the mean ± SEM (n = 3-4); the Kd is in parentheses (nanomolar concentrations). ° Significantly different from DES-treated groups, P < 0.05. 6 Significantly different from pregnant OT group, P < 0.01.
70
trous membranes, with an average Bmax of 65 fmol/mg membrane protein. DES treatment increased the number of both OT- and VP-binding sites by 4-fold. Scatchard analysis yielded one high affinity binding site for either radioligand. The Kd range of 0.83-1.4 nM was not statistically different and not affected by DES treatment.
60
50
OT- and A VP-binding sites in term pregnant rat uterus
30
20
10
0.1
1.0
10
100
1000 -11. CUMULATIVE DOSE IN 10 M
10000
FIG. 1. Cumulative dose-response curves of the uterotonic responses to OT and AVP. In vitro oxytocic assays on uterine horns from DEStreated rats were performed. Each curve represents the mean ± SEM of at least eight bioassays from eight horns (four rats). In each bioassay at least one OT and one AVP curve were obtained from the same horn. The maximal response to OT was taken as 100%. All other responses were expressed as a percentage of this value.
OT- and A VP-binding sites in nonpregnant rat uterus 3
3
[ H]0T- and [ H]AVP-binding sites in uterine membranes from nonpregnant rats and nonpregnant DEStreated rats were determined. Three or four binding experiments were performed for each radioligand. The results are summarized in Table 1. [3H]OT-binding sites and [3H]AVP-binding sites in nonpregnant rat uterine membranes had similar binding characteristics and could not be distinguished from each other. Saturable binding sites for both ligands were low in nonpregnant nones-
[3H]OT- and [3H]AVP-binding sites in day 23 term pregnant (in labor) uterine membranes were determined. Three or four binding experiments were perfomed for each radioligand. Nonspecific binding of [ 3 H]0T and [3H]AVP in term pregnant rat uterine membrane was low, less than 10% for [3H]OT and 15% for [3H]AVP at Kd levels. Table 1 shows that the saturable binding sites for [3H]OT were 356 fmol/mg membrane protein, significantly higher than the Bmax of 251 fmol/mg protein for the DES-treated membranes. Term pregnancy, however, did not further increase [3H] AVP-binding sites. The Bmax (183 fmol/mg protein), though lower, was not statistically different from the value for the DES-treated membranes. The difference between the Bmax values for [3H] OT and [3H]AVP in the term pregnant membranes is highly significant. Scatchard analysis yields one high affinity binding site for either [3H]OT or [3H]AVP, with Kd values of 1.99 and 1.47 nM, respectively, not significantly different from the values for the nonpregnant membranes. Competition binding of OT antagonist for fHJOT- and fHJA VP-binding sites Competition binding of [Pen\Phe(Me)2,Thr4,Orn8] OT, a long-acting OT antagonist, for [3H]OT- and [3H] AVP-binding sites in uterine membranes of day 23 term pregnant rats were determined. Homologous competition between OT and [3H]OT and competition between OT and [3H]AVP were also determined. Five to seven competition binding experiments were performed for each pair of ligands. Linear displacements by increasing log
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2098
concentrations of the competing peptide were obtained in all four competition binding studies within the concentration range of the unlabeled peptides used. The displacement curves are shown in Fig. 2. All four curves were parallel to one another. The IC50 values calculated from the competition experiments are shown in Table 2. OT was equally potent in displacing [3H]OT or [3H]AVP from their binding sites. The IC50 for OT, however, was 2-3 orders of magnitude higher than its ED50 calculated from the in vitro oxytocic cumulative dose curve. [Pen\Phe(Me)2,Thr4,Orn8]OT was also equally potent in displacing [3H]OT and [3H]AVP from their binding sites. Its affinity for the receptor site, however, was significantly lower than that of the OT agonist. Unlike OT, the IC50 values for [Pen\Phe(Me)2,Thr4,Orn8]OT showed a good agreement with its respective molar concentrations for their in vitro antioxytocic pA2 values. In vitro antioxytocic potencies and in vivo antivasopressor potencies of OT antagonists In vitro antioxytocic and in vivo antivasopressor potencies of a series of specific OT antagonists were determined by the pA2 method. The antioxytocic potencies of [Pen\Phe(Me)2,Thr4,Orn8]OT and [Pen\Phe2,Thr4,A3-4Pro7]OT, were measured in the isolated uterus against OT and against AVP. The two OT antagonists yielded similar antioxytocic pA2 values whether assayed against OT or AVP on the isolated uterine preparations. The results are shown in Table 3. In addition, with four other OT antagonists, [Pen\D-
Endo • 1990 Voll26«No4
Phe2,Thr4,Thr5,Orn8]OT, [Pen OT, [Pen\D-Phe2,Thr4,Asp5,Orn8]OT, and [Pen\DPhe2,Thr4,Tyr5,Orn8]OT, the antagonists were assayed for their in vivo antivasopressor pA2 values against AVP in intact rats. The results are presented in Table 4. All of the OT antagonists tested were effective inhibitors in the antioxytocic and antivasopressor assays. All were more potent in their antioxytocic than their antivasopressor activities. However, each peptide exhibited a different antioxytocic:antivasopressor potency ratio. Discussion In this study we have investigated whether OT and AVP activate the same or different receptors in the rat myometrium and whether OT receptors in the myometrium can be differentiated from the Vx receptors in vascular smooth muscle cells. In vitro dose-response curve analysis shows that the OT curve and the AVP curve were parallel and had the same maximal response. As expected, OT was more potent than AVP in oxytocic activity. The ED50 for OT was 6.9 X 10~6 MM, and that for AVP was 1.8 X 10~4 fiM, yielding a OT:AVP ratio of 26. The oxytocic activity for OT is 450 U/mg, and that for AVP 17 U/mg (19), giving an OT:AVP ratio of 26. Conventional interpretation of the dose-response curves would suggest that OT and AVP have similar intrinsic activities and produce their uterotonic action by a similar receptor-effect mechanism. The higher ED50 for AVP would indicate a lower affinity of AVP for the receptor site.
100
o (X EH
§ 80 u
FIG. 2. Competition of OT and OT antagonists for [3H]OT- and [3H]AVPbinding sites in uterine membranes of term pregnant rats. The concentration of radiolabeled ligands used in each binding assay was 0.5 nM. Displacements of radioligand binding by the competing unlabeled peptides were expressed as a percentage of the specific binding in the control sample. The curves represent the mean values of five to seven competition binding experiments.
ft. O
60
40
OT v s OT v s
UH]0T [ H]AVP
20 OT-ANTAGONIST v s OT-ANTAGONIST v s
1.0
UH]0T [ H]AVP
10 CONCENTRATION OF COMPETING PEPTIDE, nM
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
100
UTERINE OT AND VP RECEPTORS
2099
TABLE 2. Competition of OT and OT antagonist for [3H]OT- and [3H]AVP-binding sites in rat uterine membranes and comparison with their in vitro oxytocic and antioxytocic activities Biological activities* (nM)
Competition binding, IC50° (nM)
Peptide
[3H]AVP site
[3H]OT site OT OT antagonist
6.7 ± 0.6 17.4 ± 1.2C
7.7 ± 0.7 32.1 ± 2.9C
[M]
ED 50
for pA2
3
6.9 X 10"
41.3 ± 3.1 (us. OT) 49.0 ± 5.5 (vs. AVP)
Competition binding experiments were carried out in membranes prepared from day 23 term pregnant rats with 1.0 mg membrane protein, 0.5 nM labeled ligand, and four different concentrations of unlabeled competing peptide. ° IC50> Concentration of unlabeled peptide obtained from the competition binding curve which caused 50% displacement of the labeled ligand from the binding sites. The OT antagonist was [Pen\Phe(Me)2,Thr4,Orn8]OT. 6 ED50, Concentration of agonist causing 50% of the maximal response obtained from Fig. 1. [M], Concentration of OT antagonist required for the pA2 assays. c Significantly different from the corresponding OT IC50 values, P < 0.05. TABLE 3. Comparisons of inhibitory activities of OT antagonists against OT or VP agonist In vitro antioxytocic potency OT antagonists
Vs. OT
Vs. AVP
4.13 x 10"8 ± 0.31 7.38 ± 0.03
4.90 X 10~8 ± 0.55 7.31 ± 0.05
3.45 x 10"8 ± 0.35 7.46 ± 0.03
3.99 X 10"8 ± 0.22 7.41 ± 0.02
[Pen\Phe(Me)2, Thr\Orn 8 ]OT [M] pA2
[Pen',Phe2,Thr4, A34,Pro7]OT [M] pA 2
[M], Molar concentration of the antagonist for the pA2. All values are the mean ± SEM (n = at least 4). Differences between antioxytocic PA2 values are not statistically significant.
However, [3H]OT and [3H]AVP binding experiments on nonpregnant rat uterine membranes reveal that the receptor densities for [3H]OT and [3H]AVP are similar and that receptor affinity is higher for [3H]AVP than [ 3 H]0T, although the difference is not statistically significant. Scatchard plots are essentially linear for both ligands, indicating a single class of high affinity binding site. In nonpregnant human myometrium, Guillon et al. (7) reported a single class of binding sites for [3H]0T, but two classes of sites for [3H]AVP. Maggi et al (9), using rabbit uterine membranes, found that [3H]OT and [3H]AVP both bound to only one class of sites. In these two studies, OT- and VP-binding sites were found to be distinct from each other. We were unable in our binding assays to distinguish OT-binding sites from VP-binding sites in the nonpregnant rat uterus. In the nonpregnant and nonestrous rat uterus, both OT- and VP-binding sites were similarly low. DES stimulated increases in OT- and VP-binding sites equally. Similar findings in rabbits have been reported by others (9). Although our receptor binding experiments on nonpregnant uterine membranes could not distinguish OTand VP-binding sites, distinct binding sites for OT and
VP were demonstrated in the parturient uterus. OT receptor formations undergo dramatic changes during pregnancy. At parturition, there is an abrupt increase in OT receptor concentrations (4, 6). In the present study we found that estrogen stimulation increased OT-binding sites, which were further increased at parturition. VP-binding sites in the nonpregnant uterus were also increased by estrogen treatment, but, unlike OT-binding sites, VP-binding sites did not show further increases at parturition. Maggi et al. (9) have reported similar findings in rabbit myometrium. In the parturient rat uterus, the difference in Bmax values between [3H]OT and [3H] AVP binding was highly significant. The two binding sites are, therefore, distinct and could be distinguished in the parturient uterus. We then investigated whether the myometrial OT receptors and VP receptors can be differentiated by competition displacement with unlabeled OT and an OT antagonist. Unlabeled OT displaced [3H]OT and [3H] AVP from binding sites with equal potency. The OT antagonist [Pen\Phe(Me)2,Thr4,Orn8]OT also competed for OT- and AVP-binding sites with equal potency. The displacement curves for the four competition binding experiments were all parallel, suggesting that the binding sites involved were similar. The OT antagonist, however, was 3-5 times less potent than the agonist OT. Fuchs et al. (20), using a different OT antagonist in their competition binding experiments, also found that the OT antagonist was 4.5 times less potent than OT. It is interesting to note that the IC50 of OT for [ 3 H]0T binding (6.7 nM) was 3 orders of magnitude higher than the ED50 of OT calculated from the uterotonic cumulative doseresponse curve. The IC50 of the OT antagonist, on the other hand, showed a good agreement with its biological antagonistic activity, pA2. A possible explanation for this difference is that in the agonist case, the biological activity (uterine contractions) is the ultimate response after a number of amplifications in the transduction step.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2100
Endo • 1990 Vol 126* No 4
TABLE 4. Antioxytocic and antivasopressor potencies of OT antagonists Potency ratio antioxytocic [M]/ antivasopressor [M]
In vitro antioxytocic potency
In vivo antivasopressor potency
pA2 [M]
7.46 ± 0.03 3.45 x 10' 8 ± 0.35
7.00 ± 0.05 1.00 X 10"7 ± 0.11°
0.34
pA2 [M]
7.39 ± 0.03 4.13 X 10"8 ± 0.31
7.11 ± 0.08 7.67 X 10~8 ± 1.23
0.53
pA 2 [M]
7.21 ± 0.06 6.47 x 10"8 ± 0.89
6.68 ± 0.08 2.23 X 10~7 ± 0.40*
0.29
7.16 ± 0.03 7.03 x 10~8 ± 0.48
6.20 ± 0.04 6.40 X 10~7 ± 0.71"
0.10
6.76 ± 0.06 1.74 X 10"7 ± 0.26
6.72 ± 0.10 1.91 x 10"7 ± 0.80
0.91
6.67 ± 0.04 2.24 X 10~7 ± 0.24
6.20 ± 0.03 6.35 X 10"7 ± 0.44°
0.35
OT antagonists [Pen'.Phe'.Thr'.A^.Pro'lOT [Pen\Phe(Me)2,Thr4,Orn8]OT [Pen\D-Phe2,Thr\Asp6,0rn8]0T [Pen\D-Phe2)Thr4>Thr6,Orn8]OT pA 2 [M]
[Pen'.D-Phe'.Thr'.Ty^.Orn^OT pA 2 [M] 1
2
4
5
8
[Pen ,D-Phe )Thr ,Leu )0m ]0T pA2 [M]
All potency values are the mean ± SEM (n = at least 4). [M], Molar concentration of the antagonist for the pA2 assay. Significantly different from its corresponding antioxytocic [M] value, P < 0.05. 6 Significantly different from its corresponding antioxytocic [M] value, P < 0.01.
a
Whereas in the case of antagonist, the biological antagonistic activity is principally a membrane phenomenon, as the ligand-receptor binding interaction. An alternative explanation is that the agonist response involves a spare receptor system. In a separate series of experiments, we examined whether by determining the antagonistic potencies of a series of OT antagonists against OT and AVP in the uterus and on the pressor response, we may be able to distinguish OT and AVP receptors in the myometrium and differentiate the myometrial receptors from the Vi receptors in the vascular smooth muscle cells. The two OT antagonists ([Pen\Phe(Me)2,Thr6,Orn8]OT and [Pen\Phe2,Thr4,A3-4,Pro7]OT) that were tested were unable to distinguish in the antioxytocic assays whether the agonist used was OT or AVP. The two OT antagonists together with four other 5substituted OT antagonists were measured for their antagonistic potencies in in vitro oxytocic assays (OT receptor-mediated action) and in in vivo vasopressor assays (Vi receptor-mediated action). All six OT antagonists were more potent in antagonizing the uterotonic response to OT than the vasopressor response to AVP, as reflected by their antioxytociciantivasopressor potency ratios. Significantly, each peptide exhibited a different potency ratio, ranging from a high of 0.91 to a low of 0.10. This lack of unity in the potency ratios among the OT antagonists suggests that the myometrial OT/VP receptors are not the same as the Vi receptors in the vascular smooth muscle cells. However, the antivasopressor assays were performed on the intact rat. If dif-
ferences in pharmacokinetic properties among the OT antagonists exist, they could differentially affect their absorption, distribution, and elimination and, hence, influence their in vivo antagonistic potencies accordingly. Although we could not rule out the pharmacokinetic contributing factors, analysis of the data does not lend support to this mechanism. If the differences were to be ascribed to pharmacokinetic properties, one would expect that the largest differences in these parameters would be for the ligands with the greatest differences in physicalchemical properties. From this perspective, one would expect the Asp5 (acidic side-chain) and Tyr5 (highly lipophilic side-chain) analogs to show the greatest differences. This was not the case. The analog that was most affected was the Thr 5 analog, which falls somewhere between lipophilic and hydrophilic in its properties. The Tyr5 analog, in fact, showed virtually no difference between the two receptors, with a potency ratio of 0.91. Taken together, we believe that the potency ratio differences are most readily explained by the differential requirements of the myometrial and vascular receptors. This is consistent with the structure-activity-relationship (SAR) studies of Manning et al. (21) showing that selective anti-OT and anti-i, analogs could be synthesized. Based on our findings presented here, we submit that there are distinct OT- and VP-binding sites in the term pregnant rat myometrium. In the nonpregnant rat, OTand AVP-binding sites have similar binding characteristics and cannot be distinguished by [ 3 H]0T, [3H]AVP binding studies or by competition displacement binding
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS with OT or OT antagonists. However, they can be differentiated by binding assays in the term pregnant parturient uterus. The mechanism responsible for this differentiation is not known. Since DES stimulated OT receptor and AVP receptor formation in the myometrium of nonpregnant rats with equal potency, the marked effect of parturition on myometrial OT receptor formation, but not AVP receptor, cannot be ascribed solely to elevated blood concentrations of estrogens. Prostaglandins appear to have a regulatory role in OT receptor formation (22), but their effect on VP receptors has not been investigated. Our experimental findings also indicate that the myometrial OT/VP receptors are similar to but not the same as the Vi receptors in vascular smooth muscle cells.
9.
10.
11.
12.
13.
Acknowledgments
14.
The authors wish to acknowledge the skillful technical assistance of Ms. Pearl Chua-Eoan and Mrs. Lisette Sookyung Kang.
15. 16.
References 1. Soloff MS, Swartz TL 1974 Characterization of a proposed oxytocin receptor in the uterus of the rat and sow. J Biol Chem 249:1376 2. Crankshaw DJ, Branda LA, Matlib MA, Daniel EE 1978 Localization of the oxytocin receptor in the plasma membrane of rat myometrium. Eur J Biochem 86:481 3. Alexandrova M, Soloff MS 1980 Oxytocin receptors and parturition. I. Control of oxytocin receptor concentration in the rat myometrium at term. Endocrinology 106:730 4. Soloff MS 1985 Oxytocin receptors and mechanisms of oxytocin action. In: Amico JA, Robinson AG (eds) Oxytocin Clinical and Laboratory Studies. Excerpta Medica, New York, p 259 5. Fuchs AR, Periyosamy S, Alexandrova M, Soloff MS 1983 Correlation between oxytocin receptor concentration and responsiveness to oxytocin in pregnant rat myometrium. Effects of ovarian steroids. Endocrinology 113:742 6. Riemer RK, Goldfien AC, Goldfien A, Roberts JM 1986 Rabbit uterine oxytocin receptors and in vitro contractile response: abrupt changes at term and the role of eicosanoids. Endocrinology 119:699 7. Guillon G, Balestre MN, Roberts JM, Bottari SP 1987 Oxytocin and vasopressin: distinct receptors in myometrium. J Clin Endocrinol Metab 64:1129 8. Maggi M, Malozowski S, Kassis S, Guardabasso V, Rodbard D
17. 18.
19.
20. 21.
22.
2101
1987 Identification and characterization of two classes of receptors for oxytocin and vasopressin in porcine tunica albuginea, epididymis, and vas deferens. Endocrinology 120:986 Maggi M, Genazzani AD, Giannini S, Torrisi C, Baldi E, di Tomasso M, Munson PJ, Rodbard D, Serio M 1988 Vasopressin and oxytocin receptors in vagina, myometrium, and oviduct of rabbits. Endocrinology 122:2970 Chan WY, Hruby VJ, Rockway TW, Hlavacek J 1986 Design of oxytocin antagonists with prolonged action: potential tocolytic agents for the treatment of preterm labor. J Pharmacol Exp Ther 239:84 Chan WY, Rockway TW, Hruby VJ 1987 Long-acting oxytocin antagonists: effects of 2-D-stereoisomer substitution on antagonistic potency and duration of action. Proc Soc Exp Biol Med 185:187 Chan WY, Cao L, Hill PS, Hruby VJ, Pharmacological studies of oxytocin and vasopressin binding sites in the rat uterus. 4th International Conference on the Neurohypophysis, Copenhagen, Denmark, 1989, p 6 (Abstract) Perlmutter AF, Soloff MS 1979 Characterization of the metal ion requirement for oxytocin-receptor interaction in rat mammary gland membranes. J Biol Chem 254:3899 Lowry OM, Rosenbough IN, Farr AL, Randall RJ 1951 Protein measurement with the Folin phenol reagent. J Biol Chem 193:265 Scatchard G 1949 The attractions of proteins for small molecules and ions. Ann NY Acad Sci 51:660 Chan WY, Fear R, du Vigneaud V 1967 Some pharmacologic studies on 1-L-penicillamine-oxytocin and 1-deaminopenicillamine-oxytocin: two potent oxytocin inhibitors. Endocrinology 81:1267 Schild HO 1947 pA, a new scale for the measurement of drug antagonism. Br J Pharmacol 2:189 van Rossum JM 1963 Cumulative dose-response curves. II. Technique for the making of dose-response curves in isolated organs and the evaluation of drug parameters. Archs Int Pharmacodyn Ther 143:299 Berde B, Boissonnas RA 1968 Basic pharmacological properties of synthetic analogues and homologues of the neurohypophysial hormones. In: Berde B (ed) Neurohypophysial Hormones and Similar Polypeptides. Handbook of Experimental Pharmacology. SpringerVerlag, New York, p 802 Fuchs AR, Vangsted A, Ivanisevic M, Demarest K 1989 Oxytocin antagonist (dTVT) and oxytocin receptors in myometrium and decidua. Am J Perinatol 6:205 Manning M, Kruszynski M, Bankowski K, Olma A, Lammek B, Cheng LL, Klis WA, Seto J, Haldar J, Sawyer WH 1989 Solidphase synthesis of 16 potent (selective and nonselective) in vivo antagonists of oxytocin. J Med Chem 32:382 Chan WY, Berezin I, Daniel EE 1988 Effects of inhibition of prostaglandin synthesis on uterine oxytocin receptor concentration and myometrial gap junction density in parturient rats. Biol Reprod 39:1117
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
Vol. 126, No. 4 Printed in U.S.A.
Oxytocin- and Vasopressin-Binding Sites in the Rat Uterus: Competition Binding and Inhibitory pA2 Studies with Oxytocin and Oxytocin Antagonists* W. Y. CHAN, LIN CAO, PATRICIA S. HILL, AND VICTOR J. HRUBY Department of Pharmacology, Cornell University Medical College (W. Y.C, L.C.), New York, New York 10021; and the Department of Chemistry, University of Arizona (P.S.H., V.J.H.), Tucson, Arizona 85721
ABSTRACT. Recent reports have presented evidence suggesting that there are distinct oxytocin (OT) and vasopressin (VP) receptors in the human and rabbit myometrium. In this study we have investigated whether OT and arginine vasopressin (AVP) activate the same or two different receptor systems in the rat uterus in producing their uterotonic action and whether the myometrial OT/VP receptors are similar to the V^ receptors in the vascular smooth muscle cells. We compared the doseresponse characteristics of OT and AVP by the in vitro cumulative dose-response curve technique. We determined the ligandreceptor binding characteristics of [3H]OT and [3H]AVP on uterine membrane fractions from nonpregnant and pregnant rats. Specific OT antagonists were used in competition receptor binding assays and in antioxytocic pA2bioassays against OT and AVP to determine whether OT antagonists can discriminate between OT- and AVP-binding sites in the myometrium. We also compared the in vitro antioxytocic (OT receptor-mediated action) and the in vivo antivasopressor (Vj receptor-mediated
T
action) potencies of a series of six OT antagonists. Our results show that OT- and AVP-binding sites in the nonpregnant rat uterus have similar binding characteristics and cannot be distinguished by the dose-response study, radioligand receptor binding assays, or OT antagonists in the competition binding and pA2 assays. However, in the term pregnant parturient uterus, the two binding sites can be clearly differentiated. OT receptor density, but not AVP, was markedly increased at term pregnancy. All six OT antagonists studied in this investigation were more potent in antagonizing the uterotonic response to OT than the vasopressor response to AVP. The antioxytocicrantivasopressor potency ratios, however, were different between the antagonists, ranging from nearly equal (0.91) to low (0.1). The results above suggest that there are distinct OT- and AVP-binding sites in the rat myometrium. The myometrial OT/ AVP receptors are similar to but not the same as the Vx receptors in the vascular smooth muscle cells. (Endocrinology 126: 20952101, 1990)
Recently, we have developed a new series of highly potent and long-acting OT antagonists (10, 11). In this study we investigated whether these specific OT antagonists can discriminate between OT- and VP-binding sites in the rat myometrium and whether these antagonists can delineate the receptors that mediate the uterotonic and vasopressor responses of OT and VP. Competition bindings between OT antagonist and [3H]OTbinding sites and [3H]AVP-binding sites were studied. In vitro antioxytocic assays and in vivo antivasopressor assays were performed. Preliminary data of some of this work have been reported (12).
HE NEUROHYPOPHYSIAL hormones oxytocin (OT) and vasopressin (VP) both stimulate myometrial contractions. OT receptors in the uterus have been characterized (1-4). It has also been shown that the contractile response to OT correlates closely with OT receptor concentrations (5, 6). However, it has not been established whether the uterotonic action of VP is mediated via VP receptors or via the cross-reactivity of VP on the OT receptors. A number of studies have suggested that there are distinct OT and VP receptors in the myometrium and smooth muscle cells in the male reproductive tract, and that the myometrial VP-binding sites are of the Vx subtype, which mediates the vasopressor response of VP, and not the V2 subtype, which mediates the antidiuretic response (7-9).
Materials and Methods
Received September 18,1989. Address requests for reprints to: Dr. W. Y. Chan, Department of Pharmacology, Cornell University Medical College, 1300 York Avenue, New York, New York 10021. * This work was supported in part by USPHS Grants HD-20839 (to W.Y.C.) and AM-17420 (to V.J.H.).
Wistar rats were used in this investigation. Male rats (200220 g), female rats (180-200 g), or dated pregnant rats were purchased from Hilltop Animals (Scottsdale, PA). OT and VP receptor binding assays Receptor binding assays were carried out on uterine plasma membrane fractions by the method described by Soloff and
2095
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2096
Swartz (1). Uterine horns from nonpregnant rats [nonestrous or diethylstilbestrol (DES)-treated] and day 23 term pregnant rats (fetal tissues removed) were homogenized in 3 vol ice-cold 10 mM Tris-HCl buffer, pH 7.4, containing 2 mM EDTA-2Na and 0.5 mM dithiothreitol. The homogenate was centrifuged at 1,000 X g for 20 min at 4 C. The resultant supernatant was centrifuged at 100,000 X g for 60 min at 4 C. The EDTA in the preparation buffer was present to dissociate OT from its binding sites and ensure the availability of all OT-binding sites for ligand-receptor binding in the binding assay (13). The pellet that contained the plasma membrane fraction was washed three times with assay buffer (50 mM Tris-maleate buffer, pH 7.6, containing 10 mM MnCl2 and 1% gelatin) to remove EDTA. This membrane-microsomal fraction has been shown to be richer in OT-binding sites than the 10,000 x g and 20,000 x g fractions. The washed pellet was then resuspended in assay buffer to yield a protein concentration of 20-25 mg/ml, determined by the method of Lowry et al. (14). The membranemicrosomal suspension was stored in liquid nitrogen until used for receptor binding assays. For binding assays, the crude membrane preparation was diluted to 10 mg/ml. The radioligand receptor binding assay was carried out with 1.0 mg/ml membrane protein, incubated with six concentrations of [3H]OT (0.5-15 nM) or [3H]AVP (0.5-20 nM) in a final volume of 250 fi\. The incubation was carried out at 22 C for 60 min and in the absence and presence of excess (1000-fold) unlabeled OT or AVP to distinguish nonspecific and specific binding. The incubation was terminated by adding 5 ml ice-cold gelatin-free assay buffer. Free and bound labeled ligands were separated by filtration through Whatman GF/F glass microfiber filters (Clifton, NJ), using a Brandel M-24R cell harvester (Gaithersburg, MD). The filter was rinsed with 2 ml ice-cold gelatin-free assay buffer. The filtration and rinse were completed in 5 sec. The filter was then dried, and the content of radioactivity was determined by liquid scintillation spectrometry. All assays were performed in duplicate. The binding data were evaluated by Scatchard analysis (15). A computer-assisted ligand binding data analysis program was used to calculate the dissociation constant (Kd) and the saturable binding sites (Bmax). Competition binding assays were carried out in the presence of 0.5 nM labeled ligands, [ 3 H]0T or [3H]AVP, and increasing concentrations of unlabeled ligands (1-20 nM for OT and 5100 nM for OT antagonist). The IC50 value of each peptide (concentration that causes 50% displacement of the labeled ligand from binding sites) was determined from competition binding data. In vitro oxytocic assays In vitro oxytocic assays were performed on isolated uteri from nonpregnant rats pretreated with DES (50 Mg» sc) on the day before assay, with the use of Mg2+-free van Dyke-Hastings solution (16). The antioxytocic potencies of OT antagonists against OT- or AVP-induced contractile response were determined in this in vitro system by the pA2 method of Schild (17). The pA2 is the negative log molar concentration of the antagonist that will reduce the response of 2X units of the agonist to
Endo • 1990 Vol 126 • No 4
X units of the agonist. Oxytocic activities of OT and AVP, ED50 and intrinsic activity, were determined by the cumulative dose-response curve technique, as described by van Rossum (18). Antivasopressor assays Antivasopressor potencies of OT antagonists against AVP were determined in urethane-anesthetized and phenoxybenzamine-treated male rats. Blood pressure was monitored via a carotid arterial catheter, and peptides were injected via a jugular venous catheter. The effective dose for the pA2 assay was determined; this was the dose of the antagonist that would reduce the pressor response of 2X units of AVP to that of X unit of AVP. Assuming a volume of distribution of 67 ml/kg, the effective dose is converted to molar concentration for the calculation of pA2 value as in the in vitro pA2 assay. In vivo pA2 values thus obtained are only approximate estimates, since the molar concentrations of the antagonist cannot be precisely determined. Statistical analysis of data All data were expressed as the sample mean ± SEM and analyzed by analysis of variance. Significant differences between sample means were analyzed by Student's t test, both tails at P 0.05 level. Materials The OT antagonists used in this study were synthesized in our laboratories. They were [Pen\Phe(Me)2,Thr4,Orn8]OT, [Pen1,Phe2,Thr4,A3-4-Pro7]OT, [Pen\D-Phe2,Thr4,Thr5,Om8]OT, [Pen1,D-Phe2,Thr4,Leu5,Orn8]OT, [Pen1,D-Phe2,Thr4,Asp5,Orn8] OT, and [PenSD-Phe'.Thr'.Ty^Orn^OT. OT and AVP were purchased from Peninsula Laboratories (Belmont, CA). Multilabeled [3H]OT (34-38 Ci/mmol) and [3H]AVP (68 Ci/mmol) were purchased from New England Nuclear (Boston, MA). Radiochemical purities were 97-98% for [ 3 H]0T and 99% for [3H]AVP, as certified by the manufacturer. TLC performed in our laboratory showed that greater than 90% of the radioactivity migrated with the OT or AVP standard.
Results In vitro oxytocic activity of OT and A VP AVP is 20-30 times less potent than OT in oxytocic activity. However, it is a full agonist. Figure 1 shows the in vitro oxytocic dose-response curves of OT and AVP. Cumulative dose-response curves were carried out in eight uterine preparations from four DES-treated rats. In each horn, at least one OT response curve and one AVP response curve were obtained. The maximal response to OT was taken as 100%. All other responses to OT or AVP were expressed as a percentage of this maximum. The OT curve and the AVP curve were parallel and had the same maximum. The ED50 for OT was 6.9 x 10"6 nM, and that for AVP was 1.8 X 10~4 /xM. The ratio of ED50 AVP/ED50 OT was 26.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2097
TABLE 1. OT- and AVP-binding sites in nonpregnant and pregnant rat uterine membranes
100
Bmax (fmol/mg membrane protein) Ligand w 90
§
[3H]OT [3H]AVP
80
Nonpregnant (nonestrous) 66 ± (1.01 ± 65 ± (0.89 ±
11 0.14) 18 0.17)
Nonpregnant (DES-treated) 251 ± 24 (1.40 ± 0.18) 242 ± 5 (0.83 ± 0.07)
Term pregnant (in labor) 356 ± (1.99 ± 183 ± (1.47 ±
17° 0.36) 16" 0.19)
Values are the mean ± SEM (n = 3-4); the Kd is in parentheses (nanomolar concentrations). ° Significantly different from DES-treated groups, P < 0.05. 6 Significantly different from pregnant OT group, P < 0.01.
70
trous membranes, with an average Bmax of 65 fmol/mg membrane protein. DES treatment increased the number of both OT- and VP-binding sites by 4-fold. Scatchard analysis yielded one high affinity binding site for either radioligand. The Kd range of 0.83-1.4 nM was not statistically different and not affected by DES treatment.
60
50
OT- and A VP-binding sites in term pregnant rat uterus
30
20
10
0.1
1.0
10
100
1000 -11. CUMULATIVE DOSE IN 10 M
10000
FIG. 1. Cumulative dose-response curves of the uterotonic responses to OT and AVP. In vitro oxytocic assays on uterine horns from DEStreated rats were performed. Each curve represents the mean ± SEM of at least eight bioassays from eight horns (four rats). In each bioassay at least one OT and one AVP curve were obtained from the same horn. The maximal response to OT was taken as 100%. All other responses were expressed as a percentage of this value.
OT- and A VP-binding sites in nonpregnant rat uterus 3
3
[ H]0T- and [ H]AVP-binding sites in uterine membranes from nonpregnant rats and nonpregnant DEStreated rats were determined. Three or four binding experiments were performed for each radioligand. The results are summarized in Table 1. [3H]OT-binding sites and [3H]AVP-binding sites in nonpregnant rat uterine membranes had similar binding characteristics and could not be distinguished from each other. Saturable binding sites for both ligands were low in nonpregnant nones-
[3H]OT- and [3H]AVP-binding sites in day 23 term pregnant (in labor) uterine membranes were determined. Three or four binding experiments were perfomed for each radioligand. Nonspecific binding of [ 3 H]0T and [3H]AVP in term pregnant rat uterine membrane was low, less than 10% for [3H]OT and 15% for [3H]AVP at Kd levels. Table 1 shows that the saturable binding sites for [3H]OT were 356 fmol/mg membrane protein, significantly higher than the Bmax of 251 fmol/mg protein for the DES-treated membranes. Term pregnancy, however, did not further increase [3H] AVP-binding sites. The Bmax (183 fmol/mg protein), though lower, was not statistically different from the value for the DES-treated membranes. The difference between the Bmax values for [3H] OT and [3H]AVP in the term pregnant membranes is highly significant. Scatchard analysis yields one high affinity binding site for either [3H]OT or [3H]AVP, with Kd values of 1.99 and 1.47 nM, respectively, not significantly different from the values for the nonpregnant membranes. Competition binding of OT antagonist for fHJOT- and fHJA VP-binding sites Competition binding of [Pen\Phe(Me)2,Thr4,Orn8] OT, a long-acting OT antagonist, for [3H]OT- and [3H] AVP-binding sites in uterine membranes of day 23 term pregnant rats were determined. Homologous competition between OT and [3H]OT and competition between OT and [3H]AVP were also determined. Five to seven competition binding experiments were performed for each pair of ligands. Linear displacements by increasing log
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2098
concentrations of the competing peptide were obtained in all four competition binding studies within the concentration range of the unlabeled peptides used. The displacement curves are shown in Fig. 2. All four curves were parallel to one another. The IC50 values calculated from the competition experiments are shown in Table 2. OT was equally potent in displacing [3H]OT or [3H]AVP from their binding sites. The IC50 for OT, however, was 2-3 orders of magnitude higher than its ED50 calculated from the in vitro oxytocic cumulative dose curve. [Pen\Phe(Me)2,Thr4,Orn8]OT was also equally potent in displacing [3H]OT and [3H]AVP from their binding sites. Its affinity for the receptor site, however, was significantly lower than that of the OT agonist. Unlike OT, the IC50 values for [Pen\Phe(Me)2,Thr4,Orn8]OT showed a good agreement with its respective molar concentrations for their in vitro antioxytocic pA2 values. In vitro antioxytocic potencies and in vivo antivasopressor potencies of OT antagonists In vitro antioxytocic and in vivo antivasopressor potencies of a series of specific OT antagonists were determined by the pA2 method. The antioxytocic potencies of [Pen\Phe(Me)2,Thr4,Orn8]OT and [Pen\Phe2,Thr4,A3-4Pro7]OT, were measured in the isolated uterus against OT and against AVP. The two OT antagonists yielded similar antioxytocic pA2 values whether assayed against OT or AVP on the isolated uterine preparations. The results are shown in Table 3. In addition, with four other OT antagonists, [Pen\D-
Endo • 1990 Voll26«No4
Phe2,Thr4,Thr5,Orn8]OT, [Pen OT, [Pen\D-Phe2,Thr4,Asp5,Orn8]OT, and [Pen\DPhe2,Thr4,Tyr5,Orn8]OT, the antagonists were assayed for their in vivo antivasopressor pA2 values against AVP in intact rats. The results are presented in Table 4. All of the OT antagonists tested were effective inhibitors in the antioxytocic and antivasopressor assays. All were more potent in their antioxytocic than their antivasopressor activities. However, each peptide exhibited a different antioxytocic:antivasopressor potency ratio. Discussion In this study we have investigated whether OT and AVP activate the same or different receptors in the rat myometrium and whether OT receptors in the myometrium can be differentiated from the Vx receptors in vascular smooth muscle cells. In vitro dose-response curve analysis shows that the OT curve and the AVP curve were parallel and had the same maximal response. As expected, OT was more potent than AVP in oxytocic activity. The ED50 for OT was 6.9 X 10~6 MM, and that for AVP was 1.8 X 10~4 fiM, yielding a OT:AVP ratio of 26. The oxytocic activity for OT is 450 U/mg, and that for AVP 17 U/mg (19), giving an OT:AVP ratio of 26. Conventional interpretation of the dose-response curves would suggest that OT and AVP have similar intrinsic activities and produce their uterotonic action by a similar receptor-effect mechanism. The higher ED50 for AVP would indicate a lower affinity of AVP for the receptor site.
100
o (X EH
§ 80 u
FIG. 2. Competition of OT and OT antagonists for [3H]OT- and [3H]AVPbinding sites in uterine membranes of term pregnant rats. The concentration of radiolabeled ligands used in each binding assay was 0.5 nM. Displacements of radioligand binding by the competing unlabeled peptides were expressed as a percentage of the specific binding in the control sample. The curves represent the mean values of five to seven competition binding experiments.
ft. O
60
40
OT v s OT v s
UH]0T [ H]AVP
20 OT-ANTAGONIST v s OT-ANTAGONIST v s
1.0
UH]0T [ H]AVP
10 CONCENTRATION OF COMPETING PEPTIDE, nM
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
100
UTERINE OT AND VP RECEPTORS
2099
TABLE 2. Competition of OT and OT antagonist for [3H]OT- and [3H]AVP-binding sites in rat uterine membranes and comparison with their in vitro oxytocic and antioxytocic activities Biological activities* (nM)
Competition binding, IC50° (nM)
Peptide
[3H]AVP site
[3H]OT site OT OT antagonist
6.7 ± 0.6 17.4 ± 1.2C
7.7 ± 0.7 32.1 ± 2.9C
[M]
ED 50
for pA2
3
6.9 X 10"
41.3 ± 3.1 (us. OT) 49.0 ± 5.5 (vs. AVP)
Competition binding experiments were carried out in membranes prepared from day 23 term pregnant rats with 1.0 mg membrane protein, 0.5 nM labeled ligand, and four different concentrations of unlabeled competing peptide. ° IC50> Concentration of unlabeled peptide obtained from the competition binding curve which caused 50% displacement of the labeled ligand from the binding sites. The OT antagonist was [Pen\Phe(Me)2,Thr4,Orn8]OT. 6 ED50, Concentration of agonist causing 50% of the maximal response obtained from Fig. 1. [M], Concentration of OT antagonist required for the pA2 assays. c Significantly different from the corresponding OT IC50 values, P < 0.05. TABLE 3. Comparisons of inhibitory activities of OT antagonists against OT or VP agonist In vitro antioxytocic potency OT antagonists
Vs. OT
Vs. AVP
4.13 x 10"8 ± 0.31 7.38 ± 0.03
4.90 X 10~8 ± 0.55 7.31 ± 0.05
3.45 x 10"8 ± 0.35 7.46 ± 0.03
3.99 X 10"8 ± 0.22 7.41 ± 0.02
[Pen\Phe(Me)2, Thr\Orn 8 ]OT [M] pA2
[Pen',Phe2,Thr4, A34,Pro7]OT [M] pA 2
[M], Molar concentration of the antagonist for the pA2. All values are the mean ± SEM (n = at least 4). Differences between antioxytocic PA2 values are not statistically significant.
However, [3H]OT and [3H]AVP binding experiments on nonpregnant rat uterine membranes reveal that the receptor densities for [3H]OT and [3H]AVP are similar and that receptor affinity is higher for [3H]AVP than [ 3 H]0T, although the difference is not statistically significant. Scatchard plots are essentially linear for both ligands, indicating a single class of high affinity binding site. In nonpregnant human myometrium, Guillon et al. (7) reported a single class of binding sites for [3H]0T, but two classes of sites for [3H]AVP. Maggi et al (9), using rabbit uterine membranes, found that [3H]OT and [3H]AVP both bound to only one class of sites. In these two studies, OT- and VP-binding sites were found to be distinct from each other. We were unable in our binding assays to distinguish OT-binding sites from VP-binding sites in the nonpregnant rat uterus. In the nonpregnant and nonestrous rat uterus, both OT- and VP-binding sites were similarly low. DES stimulated increases in OT- and VP-binding sites equally. Similar findings in rabbits have been reported by others (9). Although our receptor binding experiments on nonpregnant uterine membranes could not distinguish OTand VP-binding sites, distinct binding sites for OT and
VP were demonstrated in the parturient uterus. OT receptor formations undergo dramatic changes during pregnancy. At parturition, there is an abrupt increase in OT receptor concentrations (4, 6). In the present study we found that estrogen stimulation increased OT-binding sites, which were further increased at parturition. VP-binding sites in the nonpregnant uterus were also increased by estrogen treatment, but, unlike OT-binding sites, VP-binding sites did not show further increases at parturition. Maggi et al. (9) have reported similar findings in rabbit myometrium. In the parturient rat uterus, the difference in Bmax values between [3H]OT and [3H] AVP binding was highly significant. The two binding sites are, therefore, distinct and could be distinguished in the parturient uterus. We then investigated whether the myometrial OT receptors and VP receptors can be differentiated by competition displacement with unlabeled OT and an OT antagonist. Unlabeled OT displaced [3H]OT and [3H] AVP from binding sites with equal potency. The OT antagonist [Pen\Phe(Me)2,Thr4,Orn8]OT also competed for OT- and AVP-binding sites with equal potency. The displacement curves for the four competition binding experiments were all parallel, suggesting that the binding sites involved were similar. The OT antagonist, however, was 3-5 times less potent than the agonist OT. Fuchs et al. (20), using a different OT antagonist in their competition binding experiments, also found that the OT antagonist was 4.5 times less potent than OT. It is interesting to note that the IC50 of OT for [ 3 H]0T binding (6.7 nM) was 3 orders of magnitude higher than the ED50 of OT calculated from the uterotonic cumulative doseresponse curve. The IC50 of the OT antagonist, on the other hand, showed a good agreement with its biological antagonistic activity, pA2. A possible explanation for this difference is that in the agonist case, the biological activity (uterine contractions) is the ultimate response after a number of amplifications in the transduction step.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS
2100
Endo • 1990 Vol 126* No 4
TABLE 4. Antioxytocic and antivasopressor potencies of OT antagonists Potency ratio antioxytocic [M]/ antivasopressor [M]
In vitro antioxytocic potency
In vivo antivasopressor potency
pA2 [M]
7.46 ± 0.03 3.45 x 10' 8 ± 0.35
7.00 ± 0.05 1.00 X 10"7 ± 0.11°
0.34
pA2 [M]
7.39 ± 0.03 4.13 X 10"8 ± 0.31
7.11 ± 0.08 7.67 X 10~8 ± 1.23
0.53
pA 2 [M]
7.21 ± 0.06 6.47 x 10"8 ± 0.89
6.68 ± 0.08 2.23 X 10~7 ± 0.40*
0.29
7.16 ± 0.03 7.03 x 10~8 ± 0.48
6.20 ± 0.04 6.40 X 10~7 ± 0.71"
0.10
6.76 ± 0.06 1.74 X 10"7 ± 0.26
6.72 ± 0.10 1.91 x 10"7 ± 0.80
0.91
6.67 ± 0.04 2.24 X 10~7 ± 0.24
6.20 ± 0.03 6.35 X 10"7 ± 0.44°
0.35
OT antagonists [Pen'.Phe'.Thr'.A^.Pro'lOT [Pen\Phe(Me)2,Thr4,Orn8]OT [Pen\D-Phe2,Thr\Asp6,0rn8]0T [Pen\D-Phe2)Thr4>Thr6,Orn8]OT pA 2 [M]
[Pen'.D-Phe'.Thr'.Ty^.Orn^OT pA 2 [M] 1
2
4
5
8
[Pen ,D-Phe )Thr ,Leu )0m ]0T pA2 [M]
All potency values are the mean ± SEM (n = at least 4). [M], Molar concentration of the antagonist for the pA2 assay. Significantly different from its corresponding antioxytocic [M] value, P < 0.05. 6 Significantly different from its corresponding antioxytocic [M] value, P < 0.01.
a
Whereas in the case of antagonist, the biological antagonistic activity is principally a membrane phenomenon, as the ligand-receptor binding interaction. An alternative explanation is that the agonist response involves a spare receptor system. In a separate series of experiments, we examined whether by determining the antagonistic potencies of a series of OT antagonists against OT and AVP in the uterus and on the pressor response, we may be able to distinguish OT and AVP receptors in the myometrium and differentiate the myometrial receptors from the Vi receptors in the vascular smooth muscle cells. The two OT antagonists ([Pen\Phe(Me)2,Thr6,Orn8]OT and [Pen\Phe2,Thr4,A3-4,Pro7]OT) that were tested were unable to distinguish in the antioxytocic assays whether the agonist used was OT or AVP. The two OT antagonists together with four other 5substituted OT antagonists were measured for their antagonistic potencies in in vitro oxytocic assays (OT receptor-mediated action) and in in vivo vasopressor assays (Vi receptor-mediated action). All six OT antagonists were more potent in antagonizing the uterotonic response to OT than the vasopressor response to AVP, as reflected by their antioxytociciantivasopressor potency ratios. Significantly, each peptide exhibited a different potency ratio, ranging from a high of 0.91 to a low of 0.10. This lack of unity in the potency ratios among the OT antagonists suggests that the myometrial OT/VP receptors are not the same as the Vi receptors in the vascular smooth muscle cells. However, the antivasopressor assays were performed on the intact rat. If dif-
ferences in pharmacokinetic properties among the OT antagonists exist, they could differentially affect their absorption, distribution, and elimination and, hence, influence their in vivo antagonistic potencies accordingly. Although we could not rule out the pharmacokinetic contributing factors, analysis of the data does not lend support to this mechanism. If the differences were to be ascribed to pharmacokinetic properties, one would expect that the largest differences in these parameters would be for the ligands with the greatest differences in physicalchemical properties. From this perspective, one would expect the Asp5 (acidic side-chain) and Tyr5 (highly lipophilic side-chain) analogs to show the greatest differences. This was not the case. The analog that was most affected was the Thr 5 analog, which falls somewhere between lipophilic and hydrophilic in its properties. The Tyr5 analog, in fact, showed virtually no difference between the two receptors, with a potency ratio of 0.91. Taken together, we believe that the potency ratio differences are most readily explained by the differential requirements of the myometrial and vascular receptors. This is consistent with the structure-activity-relationship (SAR) studies of Manning et al. (21) showing that selective anti-OT and anti-i, analogs could be synthesized. Based on our findings presented here, we submit that there are distinct OT- and VP-binding sites in the term pregnant rat myometrium. In the nonpregnant rat, OTand AVP-binding sites have similar binding characteristics and cannot be distinguished by [ 3 H]0T, [3H]AVP binding studies or by competition displacement binding
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
UTERINE OT AND VP RECEPTORS with OT or OT antagonists. However, they can be differentiated by binding assays in the term pregnant parturient uterus. The mechanism responsible for this differentiation is not known. Since DES stimulated OT receptor and AVP receptor formation in the myometrium of nonpregnant rats with equal potency, the marked effect of parturition on myometrial OT receptor formation, but not AVP receptor, cannot be ascribed solely to elevated blood concentrations of estrogens. Prostaglandins appear to have a regulatory role in OT receptor formation (22), but their effect on VP receptors has not been investigated. Our experimental findings also indicate that the myometrial OT/VP receptors are similar to but not the same as the Vi receptors in vascular smooth muscle cells.
9.
10.
11.
12.
13.
Acknowledgments
14.
The authors wish to acknowledge the skillful technical assistance of Ms. Pearl Chua-Eoan and Mrs. Lisette Sookyung Kang.
15. 16.
References 1. Soloff MS, Swartz TL 1974 Characterization of a proposed oxytocin receptor in the uterus of the rat and sow. J Biol Chem 249:1376 2. Crankshaw DJ, Branda LA, Matlib MA, Daniel EE 1978 Localization of the oxytocin receptor in the plasma membrane of rat myometrium. Eur J Biochem 86:481 3. Alexandrova M, Soloff MS 1980 Oxytocin receptors and parturition. I. Control of oxytocin receptor concentration in the rat myometrium at term. Endocrinology 106:730 4. Soloff MS 1985 Oxytocin receptors and mechanisms of oxytocin action. In: Amico JA, Robinson AG (eds) Oxytocin Clinical and Laboratory Studies. Excerpta Medica, New York, p 259 5. Fuchs AR, Periyosamy S, Alexandrova M, Soloff MS 1983 Correlation between oxytocin receptor concentration and responsiveness to oxytocin in pregnant rat myometrium. Effects of ovarian steroids. Endocrinology 113:742 6. Riemer RK, Goldfien AC, Goldfien A, Roberts JM 1986 Rabbit uterine oxytocin receptors and in vitro contractile response: abrupt changes at term and the role of eicosanoids. Endocrinology 119:699 7. Guillon G, Balestre MN, Roberts JM, Bottari SP 1987 Oxytocin and vasopressin: distinct receptors in myometrium. J Clin Endocrinol Metab 64:1129 8. Maggi M, Malozowski S, Kassis S, Guardabasso V, Rodbard D
17. 18.
19.
20. 21.
22.
2101
1987 Identification and characterization of two classes of receptors for oxytocin and vasopressin in porcine tunica albuginea, epididymis, and vas deferens. Endocrinology 120:986 Maggi M, Genazzani AD, Giannini S, Torrisi C, Baldi E, di Tomasso M, Munson PJ, Rodbard D, Serio M 1988 Vasopressin and oxytocin receptors in vagina, myometrium, and oviduct of rabbits. Endocrinology 122:2970 Chan WY, Hruby VJ, Rockway TW, Hlavacek J 1986 Design of oxytocin antagonists with prolonged action: potential tocolytic agents for the treatment of preterm labor. J Pharmacol Exp Ther 239:84 Chan WY, Rockway TW, Hruby VJ 1987 Long-acting oxytocin antagonists: effects of 2-D-stereoisomer substitution on antagonistic potency and duration of action. Proc Soc Exp Biol Med 185:187 Chan WY, Cao L, Hill PS, Hruby VJ, Pharmacological studies of oxytocin and vasopressin binding sites in the rat uterus. 4th International Conference on the Neurohypophysis, Copenhagen, Denmark, 1989, p 6 (Abstract) Perlmutter AF, Soloff MS 1979 Characterization of the metal ion requirement for oxytocin-receptor interaction in rat mammary gland membranes. J Biol Chem 254:3899 Lowry OM, Rosenbough IN, Farr AL, Randall RJ 1951 Protein measurement with the Folin phenol reagent. J Biol Chem 193:265 Scatchard G 1949 The attractions of proteins for small molecules and ions. Ann NY Acad Sci 51:660 Chan WY, Fear R, du Vigneaud V 1967 Some pharmacologic studies on 1-L-penicillamine-oxytocin and 1-deaminopenicillamine-oxytocin: two potent oxytocin inhibitors. Endocrinology 81:1267 Schild HO 1947 pA, a new scale for the measurement of drug antagonism. Br J Pharmacol 2:189 van Rossum JM 1963 Cumulative dose-response curves. II. Technique for the making of dose-response curves in isolated organs and the evaluation of drug parameters. Archs Int Pharmacodyn Ther 143:299 Berde B, Boissonnas RA 1968 Basic pharmacological properties of synthetic analogues and homologues of the neurohypophysial hormones. In: Berde B (ed) Neurohypophysial Hormones and Similar Polypeptides. Handbook of Experimental Pharmacology. SpringerVerlag, New York, p 802 Fuchs AR, Vangsted A, Ivanisevic M, Demarest K 1989 Oxytocin antagonist (dTVT) and oxytocin receptors in myometrium and decidua. Am J Perinatol 6:205 Manning M, Kruszynski M, Bankowski K, Olma A, Lammek B, Cheng LL, Klis WA, Seto J, Haldar J, Sawyer WH 1989 Solidphase synthesis of 16 potent (selective and nonselective) in vivo antagonists of oxytocin. J Med Chem 32:382 Chan WY, Berezin I, Daniel EE 1988 Effects of inhibition of prostaglandin synthesis on uterine oxytocin receptor concentration and myometrial gap junction density in parturient rats. Biol Reprod 39:1117
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 October 2015. at 12:51 For personal use only. No other uses without permission. . All rights reserved.
