FE 200 440: a selective oxytocin antagonist onthe term-pregnant human uterus

Lotta Nilssona, Torsten Reinheimerb, Margareta Steinwalla, Mats Akerlunda,*

Objective To compare a newly developed oxytocin antagonist, FE 200 440, with atosiban and ANTAG III, asto potency and selectivity of inhibitory effects on oxytocin- and vasopressin-induced myometrial responses.FE 200 440 has high affinity for the human cloned oxytocin receptor, approximately 300-fold that for thevasopressin V1a receptor, whereas atosiban binds well to both receptors.

Design In vitro study of human myometrial contractility.

Setting The Research Laboratory of the Department of Obstetrics and Gynecology, Lund University Hospital,Sweden.

Participants Forty-seven term-pregnant women not in labour who were delivered by caesarean section.

Interventions Concentration–response curves with oxytocin and arginine vasopressin on isolated myometrialstrips were recorded in control experiments, in the presence of atosiban in concentrations of 25, 250 and 750nmol/L, and after pretreatment with ANTAG III and FE 200 440, both in concentrations of 2.5, 25 and 250nmol/L.

Main outcome measure pA2 values (i.e. an index of inhibitory action).

Results With oxytocin as the agonist, the median pA2 values for atosiban, ANTAG III and FE 200 440 were10.6, 9.5 and 8.3, respectively. With vasopressin as the agonist, the pA2 values for atosiban and ANTAG IIIwere 8.8 and 8.7, whereas no inhibitory effect of FE 200 440 was seen in five out of six experiments.

Conclusion The new analogue FE 200 440 is a selective oxytocin antagonist and, in contrast to atosiban andANTAG III, has practically no effect on vasopressin-induced contractions of isolated term-pregnant humanmyometrium.

INTRODUCTION

Oxytocin stimulates the human uterus preterm and at

term via a specific oxytocin receptor1 – 4, which is also

found in the non-pregnant condition5. The uterus also

contains vasopressin V1a receptors, which may play a role

in uterine activation2,3,5. In late pregnancy, the densities of

myometrial oxytocin and vasopressin V1a receptors are

approximately equal3, whereas in the non-pregnant condi-

tion vasopressin V1a receptors occur in about fivefold

higher concentration5.

A new therapeutic approach to the problem of preterm

labour is to block uterine oxytocin and vasopressin V1a

receptors. The analogue 1-deamino-2-D-Tyr(OEt)-4-Thr-8-

Orn-oxytocin ¼ atosiban (Ferring Pharmaceuticals, Copen-

hagen, Denmark) was shown to have a high affinity to both

receptors6, and to inhibit the uterine actions of the peptides

in animal experiments and in the human in vitro and

in vivo7 – 10. This analogue was also shown to reduce the

contraction rate in preterm labour11,12 and is now registered

in Europe on the indication of delaying imminent preterm

birth (TRACTOCILE, Ferring). Recently, an oxytocin antag-

onist was developed, FE 200 440 (Ferring), which in various

animal experiments has shown up to several fold increased

selectivity as an oxytocin inhibitor compared with atosiban.

A comparison of the potency and selectivity of the inhibitory

actions of these analogues on responses to oxytocin and

vasopressin is of interest not only for drug development, but

also for obtaining further information on the interaction

between oxytocin and vasopressin with their receptors in

the human uterus. The formal hypothesis of the investigation

was that FE 200 440 is a selective oxytocin receptor

antagonist in the human pregnant uterus, while atosiban

and another reference compound, ANTAG III13 are not.

METHODS

Myometrial tissue for this study was obtained from 47

women, aged 18–42 years (mean 38 years) and delivered

by elective caesarean section. The median gestational

length at the operation was 38 weeks (range 36–41 weeks).

BJOG: an International Journal of Obstetrics and GynaecologyNovember 2003, Vol. 110, pp. 1025–1028

D RCOG 2003 BJOG: an International Journal of Obstetrics and Gynaecology

PII: S 1 47 0 - 0 3 2 8 ( 0 3 ) 2 0 5 40 - 1 www.bjog-elsevier.com

aDepartment of Obstetrics and Gynecology, Lund

University Hospital, SwedenbDepartment of Clinical Pharmacology and Experimental

Medicine, Ferring Pharmaceuticals A/S, Copenhagen,

Denmark

* Correspondence: Professor M. Akerlund, Department of Obstetrics and

Gynecology, University Hospital, S-221 85 Lund, Sweden.

The indications for the operation were psychosocial (n ¼13), breech position (n ¼ 10), earlier section (n ¼ 8), pelvic

disproportion (n ¼ 4), placenta praevia (n ¼ 3), bad

obstetric history (n ¼ 2), heart disease baby (n ¼ 2),

intrauterine growth retardation (n ¼ 2), fetal malformation

(n ¼ 2) and oligohydramnios (n ¼ 1). The subjects were

informed about the purpose and procedure of the investi-

gation and gave their verbal consent to the procedure. The

study was approved by the local Ethics Committee.

The experimental procedure used for in vitro recordings

was described in detail recently6. In short, at the caesarean

section, a myometrial tissue sample of 7 � 7 � 15 mm was

excised from the upper rim of the transverse incision of the

isthmic part of the uterus. The tissue piece was stored in

ice-cold, non-perfused Krebs–Ringer solution aerated with

carbogene (5% CO2 in O2), and used within 24 hours14.

The tissue pieces taken from each uterus were divided into

six to seven myometrial strips of a standardised size of 2 �2 � 10 mm, all with the same muscle fibre direction. Final

dissection was performed under a microscope with 10-fold

magnification. The strips were mounted in organ baths

containing 10 mL of non-perfused Krebs–Ringer solution

at pH 7.4, at a temperature of 37jC and aerated with

carbogene. Isometric contractions were recorded at a rest-

ing tension of 10 mN by a Grass force transducer (FT 03).

The recorded signals were stored in a computer (IBM

35SX), which was provided with a system for online

integration of the recording curve (Synectics, Sweden).

Each myometrial preparation was initially subjected to a

test, using oxytocin and vasopressin in a concentration of

1.0 nmol/L for 7 minutes, to ensure that the tissue was

viable. Maximally, one preparation for each patient had to

be exchanged in order to obtain six viable strips for the

recording experiments. The three antagonists tested in the

present study are analogues of oxytocin. The binding

affinities of FE 200 440 for the human cloned oxytocin

and vasopressin V1a receptors are 0.31 and 85.3 nmol/L,

respectively, while that of atosiban are 11.3 and 0.43 nmol/L.

In experiments on myometrial pieces from the first 11

women, it was verified that concentration–response experi-

ments with the agonists, and washing in between gave

unchanged results. It was also found that antagonists were

difficult to wash out completely, and therefore, parallel

recordings with only one administration of antagonist for

each preparation were used.

Strips from 18 women were used in experiments with

oxytocin as agonist, and strips from another 18 women were

used with vasopressin. The concentrations of agonists and

antagonists were chosen on the basis of results from experi-

ments performed during method development. At first,

concentration–response curves were constructed. Regard-

ing oxytocin, 13–14 concentrations were used in the range

of 0.0625 to 256 nmol/L, and for vasopressin, 16 concen-

trations in the range of 0.00098 to 32 nmol/L were used. The

agonists were added to the organ chamber every 7 minutes.

Thereafter, the preparations were repeatedly washed.

In antagonist testings, the six myometrial strips from

each uterus were used for only one of the compounds.

Three different concentrations of antagonist were studied

with two strips employed for each concentration. Atosiban

was added in concentrations of 25, 250 and 750 nmol/L,

whereas FE 200 440 and ANTAG III were added in

concentrations of 2.5, 25 and 250 nmol/L. The compound

was given 5 minutes before cumulative addition of oxyto-

cin or vasopressin. In these experiments with antagonists,

the oxytocin concentrations used were 1 to 10240 nmol/L

and those of vasopressin were 0.1563 to 10240 nmol/L.

The power calculation was done on the basis of results in

animal experiments showing that FE 200 440 in those

species was highly selective for the oxytocin receptor.

Thus, a yes or no answer was expected in the testings on

human uterine strips, making the number of subjects, six

per each antagonist with oxytocin and six for each with

vasopressin appropriate.

Myometrial activity was measured as area under the

recording curve (AUC) over 7-minute periods. In all experi-

ments, values for individual strips were calculated and the

mean of the two strips for each antagonist concentration

Fig. 1. Representative recordings of the inhibitory effect of FE 200 440 are

shown on oxytocin-induced contractions of isolated myometrial strips. The

cumulative effect of oxytocin (OT) per se (0.0625 to 256 nmol/L, starting

at the arrow) was first recorded and then after washing (W), the effect of

the same oxytocin concentrations (starting as indicated) was measured

after pre-incubation with the antagonist in concentrations of 2.5 (A1), 25

(A2) and 250 (A3) nmol/L.

1026 L. NILSSON ET AL.

D RCOG 2003 Br J Obstet Gynaecol 110, pp. 1025–1028

was taken as a representative result. The data were

analysed by non-linear regression, using sigmoidal

dose–response (variable curve), with the software Graph-

Pad Prism (GraphPad software, San Diego, USA) and the

maximal contractile response (Emax) was calculated by

this program. In this calculation, the minimum value of

the regression curve of the program was assumed to be

�0. The pA2 values were also estimated15. pA2 is the

negative decimal logarithm of the antagonist concentration

in the presence of which the agonist concentration has to

be doubled in order to reach again 50% of the agonist

maximal effect. In calculating the pA2 value, the concen-

tration–response experiments at the beginning, performed

on two representative strips, were chosen as controls. The

median pA2 values for the three inhibitors with oxytocin

and vasopressin as agonists were compared using one-way

ANOVA.

RESULTS

Representative recordings illustrating the technique used

in the experiments done for comparison of the antagonists,

are shown in Fig. 1. All strips exhibited spontaneous

regular contractions at a frequency of 10–20 per hour.

After adding an antagonist, total inhibition of spontaneous

activity was frequently observed; the length of this inhibi-

tion tended to be dose dependent as shown in Fig. 1.

The pA2 values for all three antagonists with oxytocin

and vasopressin as agonists are shown in Table 1. All

antagonists inhibited the effect of oxytocin, with median

pA2 values varying between 8.3 and 10.6. The inhibitory

effect of the antagonists did not differ significantly. With

vasopressin as agonist, FE 200 440 showed no inhibition

in five out of six experiments, whereas the analogues

atosiban and ANTAG III both inhibited vasopressin

effects, with median pA2 values of 8.8 and 8.7, respec-

tively (Table 1).

DISCUSSION

The present study demonstrated an inhibition of oxy-

tocin effect by the three analogues, FE 200 400, ANTAG

III and atosiban. FE 200 440 caused no inhibition of

vasopressin effects, whereas the other two analogues

were potent inhibitors. This demonstrates the possibility

of obtaining a separate inhibition of these two closely

related receptors. The high selectivity of FE 200 440 as

inhibitor of oxytocin effects compared with previous

peptide and non-peptide analogues tested in the human

in vitro and in vivo6,16 – 18 is a novel finding. The non-

peptide compound L-371,257 appears to be highly se-

lective for the oxytocin receptor in vitro19, however,

the clinical development of this compound has been

stopped.

The effects of the antagonists on spontaneous uterine

activity were not specifically studied in the present

investigation, but it was observed that following addition

of the antagonists, a dose-dependent inhibition of sponta-

neous contractions occurred. At present, there is some

debate as to what is the role of oxytocin in human term

and preterm labour with regard to the hypothesis of

spontaneous contractions being related to a local uterine

production of oxytocin19. In fact, it was recently shown

that an oxytocin antagonist inhibited not only oxytocin-

induced contractions, but also spontaneous contractility of

isolated myometrial strips from pregnant humans19.

The inhibitory actions of the three analogues on oxyto-

cin responses were pronounced and similar in magnitude.

Regarding FE 200 440 and atosiban, this observation is in

agreement with the high and similar binding affinities of

the two compounds to the human oxytocin receptor.

Regarding inhibition of vasopressin effects, the observa-

tions from the present study are also in agreement with

binding affinity data. FE 200 440 in contrast to atosiban

has a poor affinity for the vasopressin V1a receptor,

approximately 300-fold lower than that for the oxytocin

receptor. It may have been expected that some reduction of

effect of vasopressin on contractile activity would appear

with FE 200 440, in view of the cross-reactivity of the

agonist, vasopressin binding to some extent to the oxytocin

receptor as well. However, such an effect was not seen in

the present study.

In the non-pregnant condition, vasopressin is probably a

more important uterine agonist than oxytocin, considering

the five times higher myometrial content of this receptor

as well as its corresponding gene, and a much higher

potency5,20. The myometrial hyperactivity and reduced

blood flow in women with primary dysmenorrhoea are

probably caused by an action of vasopressin rather than by

oxytocin stimulation21 – 23.

Whether or not the increased selectivity of FE 200

440 to the oxytocin receptor with practically no antag-

onism of vasopressin effects may be a clinical advantage

in the treatment of preterm labour remains to be studied.

However, studies with a selective oxytocin receptor

antagonist in preterm labour would certainly be useful

in delineating the role of the two respective receptors in

that condition.

Table 1. Inhibitory effects expressed as pA2 values of atosiban, ANTAG

III and FE 200 440 on oxytocin- and vasopressin-induced contractions of

isolated, human, term-pregnant myometrium. The results are given as

medians (range) of experiments with myometrium from a total of 36

patients, each contributing to the testing with six myometrial strips. The

results for each analogue with one agonist are from a total of six patients.

Substance Oxytocin as agonist Vasopressin as agonist

Atosiban 10.6 (9.2– 14.9) 8.8 (8.3– 10.3)

ANTAG III 9.5 (8.6– 17.5) 8.7 (8.5– 9.5)

FE 200 440 8.3 (5.0– 15.1) –

FE 200 440: A SELECTIVE OXYTOCIN ANTAGONIST 1027

D RCOG 2003 Br J Obstet Gynaecol 110, pp. 1025–1028

References

1. Fuchs AR, Fuchs F, Husslein P, Soloff MS, Fernstrom MJ. Oxytocin

receptors and human parturition: a dual role for oxytocin in the ini-

tiation of labor. Science 1982;215:1396– 1398.

2. Maggi M, Del Carlo P, Fantoni G, et al. Human myometrium during

pregnancy contains and respond to V1 VP receptors as well as oxy-

tocin receptors. J Clin Endocrinol Metab 1990;70:1142–1154.

3. Bossmar T, Akerlund M, Fantoni G, Szamatowicz J, Melin P, Maggi M.

Receptors for and myometrial responses to oxytocin and vasopressin in

preterm and term human pregnancy. Effects of the oxytocin antagonist

atosiban. Am J Obstet Gynecol 1994;171:1634–1642.

4. Wathes DC, Borwick SC, Timmons PM, Leung ST, Thornton S.

Oxytocin receptor expression in human term and preterm gestational

tissues prior to and follow in the onset of labour. J Endocrinol

1999;161:143– 151.

5. Bossmar T, Akerlund M, Szamatowicz J, Laudanski T, Fantoni G,

Maggi M. Receptor-mediated uterine effects of vasopressin and

oxytocin in non-pregnant women. Br J Obstet Gynaecol 1995;

102:907– 912.

6. Akerlund M, Brouard R, Bossmar T, et al. Receptor binding of oxy-

tocin and vasopressin antagonists and the inhibitory effects on isolated

myometrium from preterm and term pregnant women. Br J Obstet

Gynaecol 1999;106:1047– 1053.

7. Akerlund M, Carlsson AM, Melin P, Trojnar J. The effect on the

human uterus of two newly developed competitive inhibitors of oxy-

tocin and vasopressin. Acta Obstet Gynecol Scand 1985;64:499– 504.

8. Melin P, Trojnar J, Johansson B, Vilhardt H, Akerlund M. Synthetic

antagonists of the myometrial response to oxytocin and vasopressin.

J Endocrinol 1986;111:125– 131.

9. Akerlund M, Hauksson A, Lund S, Melin P, Trojnar J. Vasotocin

analogues which competitively inhibit vasopressin stimulated uterine

activity in healthy women. Br J Obstet Gynaecol 1986;93:22– 27.

10. Hauksson A, Akerlund M, Melin P. Uterine blood flow and myome-

trial activity at menstruation, and the action of vasopressin and a

synthetic antagonist. Br J Obstet Gynaecol 1988;95:898– 904.

11. Akerlund M, Stromberg P, Hauksson A, et al. Inhibition of uterine

contractions of premature labour with an oxytocin analogue. Results

from a pilot study. Br J Obstet Gynaecol 1987;94:1040– 1044.

12. Worldwide Atosiban versus h-Agonist Study Group. An analysis of

the pooled data from three randomised double-blind controlled trials

of the oxytocin antagonist atosiban versus the h-adrenergic agonists

ritodrine, terbutaline and salbutamol in the treatment of preterm

labour. Br J Obstet Gynaecol 2001;108:133– 142.

13. Fejgin MD, Pak SC, Flouret G, Parsons MT, Wilson Jr L. Comparison

of the in vivo activity of different oxytocin antagonists in the pregnant

baboon. J Soc Gynecol Investig 1998;5:251– 254.

14. Wilson RJ, Allen MJ, Nandi M, Giles H, Thornton S. Spontaneous

contractions of myometrium from humans, non-human primate and

rodents are sensitive to selective oxytocin receptor antagonism in

vitro. Br J Obstet Gynaecol 2001;108:960– 966.

15. Eggena P, Schwartz IL, Walter R. A sensitive hydroosmotic toad

bladder assay. Affinity and intrinsic activity of neurohypophyseal

peptides. J Gen Physiol 1968;52:465 –481.

16. Akerlund M, Haukson A, Lundin S, Melin P, Trojnar J. Vasotocin

analogues which competitively inhibit vasopressin stimulated uterine

activity in healthy women. Br J Obstet Gynaecol 1986;93:22– 27.

17. Bossmar T, Brouard R, Doberl A, Akerlund M. Effects of SR 49059,

an orally active vasopressin V1a receptor antagonist on vasopressin-

induced uterine contractions in women. Br J Obstet Gynaecol 1997;

104:471– 477.

18. Steinwall M, Bossmar T, Gaud C, Akerlund M. Inhibitory effects of

SR 49059 on oxytocin- and vasopressin-induced uterine contractions

in non-pregnant women. Acta Obstet Gynaecol Scand. In press.

19. Wilson RJ, Allen MJ, Nandi M, Giles H, Thornton S. Spontaneous

contractions of myometrium from humans, non-human primate and

rodents are sensitive to selective oxytocin receptor antagonism in vitro.

Br J Obstet Gynaecol 2001;108:960– 966.

20. Helmer H, Hackl T, Schneeberger C, et al. Oxytocin and vasopressin

1a receptor gene expression in the cycling or pregnant human uterus.

Am J Obstet Gynecol 1998;179:1572– 1578.

21. Akerlund M, Andersson K-E, Ingemarsson I. Effects of terbutaline on

myometrial activity, endometrial blood flow, and lower abdominal

pain in women with primary dysmenorrhoea. Br J Obstet Gynaecol

1976;83:673 –678.

22. Akerlund M, Stromberg P, Forsling M-L. Primary dysmenorrhoea

and vasopressin. Br J Obstet Gynaecol 1979;86:484 –487.

23. Stromberg P, Akerlund M, Forsling ML, Granstrom E, Kindahl H.

Vasopressin and prostaglandin in premenstrual pain and primary dys-

menorrhoea. Acta Obstet Gynecol Scand 1984;63:533– 538.

Accepted 26 June 2003

1028 L. NILSSON ET AL.

D RCOG 2003 Br J Obstet Gynaecol 110, pp. 1025–1028