WO1990002756A1

WO1990002756A1 - Oxytocin antagonist

Info

Publication number: WO1990002756A1
Application number: PCT/US1989/003792
Authority: WO
Inventors: George Flouret; Laird Wilson
Original assignee: Northwestern University
Priority date: 1988-09-02
Filing date: 1989-09-01
Publication date: 1990-03-22
Also published as: AU4211089A

Abstract

A novel compound is disclosed for inhibiting the effects of oxytocin in a female mammal. As an analog of oxytocin the compound is named PMP?1-D-Trp2-Arg8¿-oxytocin, wherein PMP is β, β-cyclopentamethylene-β-mercapto-propionic acid. This compound can be administered to pregnant women to arrest premature labor while avoiding unwanted side effects due to antagonism of the antidiuretic hormone, vasopressin.

Description

OXYTOCIN ANTAGONIST

FIELD OF INVENTION

The present invention relates to a novel compound which is highly active as an oxytocin antagonist and which exhibits slight antagonism for

vasopressin.

BACKGROUND OF INVENTION

Preterm labor is the major cause of prenatal morbidity and mortality in the United States. Current methods of inhibiting preterm labor are not always successful and are often associated with significant side effects. Since the uterus is a target organ for oxytocin, and assuming that oxytocin is an important contributing factor to preterm labor, the development of a potent oxytocin antagonist would result in successful inhibition of preterm labor with few associated side effects.

Structurally, oxytocin (OT) and antidiuretic hormone (ADH), also called vasopressin, are similar. Their comparative structures are illustrated below.

Various investigations in the literature

have reported the synthesis of antagonists to ADH for the treatment of hypertension and the synthesis of antagonists to oxytocin. In 1960, Law, H.D. and V.

DuVigneaud, J. Am. Chem. Soc., 82:4579, reported the first synthesis of an oxytocin antagonist ( 2-0-methyltyrosine-OT). In 1967, Chan, Fear and DuVigneaud,

Endocrinology, 81:1267, reported the synthesis of

1-L-Penacillamine-oxytocin and 1-deamino-penacillamineoxytocin. This was the first study to show an in vivo inhibitory effect of an oxytocin antagonist on uterine contractions and response to oxytocin in the anesthetized rat.

In 1980, Sawyer, et al., Endocrinology,

106;81, reported the synthesis of an oxytocin antagonist that combined the two important features of the Law and DuVigneaud antagonist and of the Chan, et al. antagonist. The new antagonist was (1-deamino-penacillamine, 2-0-methyltyrosine) oxytocin. The new antagonist had a pA₂ of 7.8 as determined by the oxytocic bioassay. The pA₂ is the negative logarithm of the molar concentration of the antagonist that reduces the response to the antagonist by 1/2. It is defined by Schild,. British J. Pharmacology, 2:189 (1947).

In 1983, Manning, et al., J. Med. Chem.,26:1607 reported the synthesis of a number of antagonists to ADH. One of these antagonists proved to have potential anti-oxytocic activity (1-(ß-mercapto-ß,ß,- cyclopentaethylenepropionic acid),-D-Phe²,lie⁴)

arginine vasopressin with a pA₂ of 8.2, or in other words, 2.5 times more potent than the antagonist reported by Sawyer, et al. in 1980. This oxytocin antagonist can be called [PMP¹-D-Phe²-Phe⁴-Ile⁴- Arg⁸] oxytocin. It was not tested for in vivo activity in the rat. A related oxytocin antagonist, PMP¹-D- Trp²-Ile⁴-Arg⁸-oxytocin was disclosed by Wilson and Flouret, Abstract for Society for the Study of Reproduction Meeting July 15-17, 1986.

In 1981, Melin, et al., Endocrinology, 88:173, developed an oxytocin antagonist for inhibiting preterm labor. They synthesized 1-deaminoethyl-oxytocin which had a pA₂ of 7.2. They also showed that this compound inhibited uterine contractions in rats in vivo and in humans in vitro and in vivo (Akerland, et al., Obstet. and Gynecol.,

62:309, 1983). In 1985, Akerland, et al., Obstet. and Gynecol. Scand., 64:499, reported the synthesis of 1-deamino[2-D-Tyr(OET)²4-Thr8-ORN] oxytocin with a pA₂ of 8.3. They have tested this compound in vitro on human uterine tissue and have shown it to inhibit uterine contraction.

United States Patent 4,597,901 discloses the class of vasopressin antagonists in which cys- teine-1 is present in both oxytocin and vasopressin and substituted with ß ,ß-cylopentamethylene-ß- mercaptopropionic acid.

Other amino acids of vasopressin are substituted. The resulting class of compounds is said to be vasopressin antagonists with biological activity is manifested as a water diuresis.

SUMMARY OF INVENTION

The present invention comprises an oxytocin antagonist which is an analog of oxytocin. In the compound of this invention, cysteine-1 of oxytocin is substituted with ß,ß-cyclopentamethylene-β-mercaptopropionic acid. In addition, L- tyrosine-2 is substituted with D-tryptophan, and L-arginine is substituted in the 8 position for

L-leucine. The resulting compound [PMP¹-,

D-Trp², Arg⁸] oxytocin is believed to be novel and has been found to have remarkable properties. It is highly active as an oxytocin antagonist. At the same time, and although it is structurally similar to vasopressin and vasopressin antagonists described in the literature, the new compound exhibits minimal ADH antagonism. When these two antagonisms are expressed as a ratio, the compound of this invention has a very high oxytocin/ADH activity ratio. This combination of properties is highly advantageous for therapeutic use. Effective anti-oxytocin action can be obtained with minimal anti-ADH side effects. The compound of this inven- tion is therefore adapted for inhibiting contraction of the uterine muscle in response to bodily oxytocin, and can be used to suppress preterm labor.

DESCRIPTION OF THE INVENTION

The oxytocin antagonist of this invention is represented by the formula

The remarkable properties of the novel compound of this invention are shown by bioassays, which will now be described.

Oxytocin Bioassay

The protocol used for the oxytocin bioassay procedure is derived from procedures described in a paper by Sawyer, et al., Endocrinology, 106:81 (1980), which in turn was based on reports of Munsick, Brit. J. Pharmacol., 3:328 (I960), and Hoiton, Brit. J. Pharmacol., 3:328 (1948). The assay calculations for the pA₂ estimates are described by Schild, British J.Pharmacology, 2:189 (1947). The major difference in the present procedure from those reported by others in the field is that the area under the contraction is integrated where most other techniques calculate the amplitude. Integration provides much more consistent and reliable results although the pA₂ estimates are approximately a factor lower than those reported using amplitude of the contraction as the endpoint.

Method:

1. Animals - a 1.5 cm piece of uterus from a virgin rat (Holtzman) in natural estrus is used for the assay.

2. Buffer/Assay Bath - The buffer used is Munsicks. This buffer contains 0.5 mM Mg⁺⁺ which reduces the pA₂ estimates, but the results are reported to correlate better with in vivo data (Sawyer, et al., 1980). The buffer is gassed continuously with 95% oxygen; 5% carbon dioxide giving a pH of 7.4. The temperature of the assay bath is 37°C. A 10 ml assay bath is used that contains a water jacket for maintaining the temperature and inlet and outlet spikets for adding and removing buffer.

3. Polygraph/transducer - The piece of uterine tissue used for the assay is connected to a Statham Strain Gauge Force Transducer which in turn is attached to a Grass Polygraph Model 79 for monitoring the contractions.

4. Assay Protocol. (a) The tissue is equilibrated in the assay bath for one hour with washing with new buffer every fifteen minutes. One gram of tension is kept on the tissue at all times.

(b) The tissue is stimulated initially with oxytocin at 10 nM to "acclimate" the tissue and with 4 inM KCL to determine the maximum contractile response.

(C) A cumulative dose response curve is then done with oxytocin and a concentration of oxytocin equivalent to approximately 90% of the maximum is used for estimating the pA₂ of the anbagonist.

(d) The tissue is exposed to oxytocin (Calbiochemical, San Diego, California) for one minute and washed out. There is a three minute interval before addition of the next dose of the agonist or antagonist. When the antagonist is tested, it is given five minutes before the

agonist. The agonist is given for one minute. All responses are integrated using a 7P10 Grass Integrator. This is the major difference between the present protocol and others in the literature which usually measure amplitude of the contractions as the response. A single concentration of oxytocin, equal to 80% of the maximum response, is used to test the antagonist. Three different concentrations of antagonists are used, two that will reduce the response to the agonist by less than 50% and one that will reduce the response greater than 50%

(ideally this relation would be 25%, 50% and 75%). This is repeated three times for each dose of antagonist for a three point assay.

(e) Calculations for pA₂: The dose-response (DR) ratios are calculated for

antagonist and a Schild's Plot is performed by plotting the Log (DR-1) vs. Log of antagonist concentration. The line plotted is calculated by least squares regression analysis. The pA₂ is the concentration of antagonist at the point where the regression line crosses the 0 point of the Log (DR-1) ordinate. The pA₂ is the negative Log of the concentration of antagonist that will reduce the response to the agonist by one-half.

As an analog of oxytocin, the novel compound of this invention may be designated as PMP¹-D-Trp²-Arg⁸-oxytocin. When this compound was tested by the above-described assay for competitive antagonism with oxytocin, in an average of ten assays, the pA₂ value was found to be 7.73.

ADH-Bioassay

The above compound was also tested for antagonism to vasopressin. Anti-ADH activity can be determined by measuring the alteration in urine output due to ADH in the presence and absence of the antagonist. A suitable ADH-assay is described in Sawyer, et al., Endocrinology, 63:694 (1958). When tested by this method, it was found that the compound [PMP¹,D-Trp²,Arg⁸] oxytocin exhibited very low activity as a vasopressin antagonist. The ratio of oxytocin antagonism to ADH antagonism was very high, viz. over 200.

By virtue of its oxytocin antagonist activity with minimal vasopressin antagonism, the compound of this invention will be useful in treating symptoms requiring an oxytocin antagonist in humans and animals. It can be used to inhibit uterine contractions and milk letdown as well as to inhibit preterm labor. Although the structure of the compound resembles both oxytocin and vasopressin, it exhibits not only increased anti-oxytocin activity but also greatly decreased anti-ADH

activity. This compound might also be useful for inhibiting dysmenorrhea or serving as an antidote for overstimulation of uterine contraction during labor induction with oxytocin or for treating hypertension.

The compound of this invention can be administered to women by various known routes of administration. For hospital use, intravenous infusion will usually be the administration route of choice. However, the compound may also be administered intraperitoneally, subcutaneously, or intramuscularly. Oral administration may also be feasible. If required, tablets or capsules for oral use may be provided with an enteric coating protecting the compound from destruction in the stomach while permitting its release in the intestinal tract.

An effective but nontoxic quantity of the compound is employed in treatment. The dosage regimen for preventing or treating symptoms by the compound of this invention is selected in accordance with a variety of factors including the type, age, weight, sex and medical condition of the woman, the severity of the symptoms and the route of administration of the compound. An ordinary medical practitioner can determine and prescribe the effective amount based on the route of administration of the oxytocin antagonist to prevent or arrest the

progress of the condition to be inhibited. For example, an effective dose range may range from 0.01 to 100 milligrams per kilogram of body weight per day using administration by the intravenous route, such as in sterile normal saline. The compound of this invention may be prepared by a novel method. The substitution of tryptophan in peptides may have been avoided in the past because Trp-peptides are acid sensitive.

Bodanszky, et al., J. Med. Chem., 23:1258-1261 (1980) and Sawyer, et al., J. Med. Chem., 1258-1261 (1980) made [Trp-⁸] oxytocin by more difficult indirect methods, in order to avoid acid treatment of the Trp-peptide.

Chart A illustrates the method of forming the peptide intermediate and its reaction with PMP to form the compound of Formula 1.

CHART A

F

ABBRBVIATIOICS:

Boc = Tertiary-butyloxycarbonyl TFA = trifluoroacatic acid DCC = dicyclohexycarbodimiide Tos = para-toluenesulfonyl HOBt = 1-hydroxybenzotriazole DMF = dimethylformanids DIEA = Diisopropyletbylaains Meb = methylbenzyl

Bzl = benzyl The method of the present invention therefore involves the preparation of the oxytocin antagonist using a mild acidolytic method at room temperature which provides for the addition of tryptophan in the penultimate step of the synthesis. Only one acid addition step is needed to prepare the compound.

The peptide was assembled starting with tertiary butyloxycarbonyl-glycyl-PAM-resin ester by the solid phase method of synthesis. The peptide was removed from the resin by aminolysis, and the antagonist was obtained by deprotection with sodium in liquid ammonia and oxidation with potassium ferricyanide. The peptide can be isolated by standard chromatography procedures. The pure antagonist was obtained by preparing high performance liquid chromatography.

The invention will appear more fully from the examples which follow.

EXAMPLE I

Example of Coupling Cycle Yielding

Boc-Ile-Gln-Asn-Cys(Bzl)-Pro-Arg(Tos)-Gly-Resin

Tertiary-butyloxycarbonyl-glycyl-PAM (Phenyl-acetamidomethyl-resin ester, BoC-Gly-PAM- resin) was purchased from Applied Biosystems (Foster City, California). The Boc-Gly-PAM-resin (0.7 mmol/g) 0.72 g (0.5 mmol) was reacted at room temperature in an Applied Biosystems Peptide Synthesizer, Model 430A through one coupling cycle of a solid phase method of synthesis (see Chart A) with an excess of the following amino acid residues in succession: Boc-Arg (Tos) (0.86 g. 2 mmole)

Boc-Pro (0.43 g. 2 mmole)

Boc-Cys (Bzl) (0.52 g. 2 mmole)

Box-Asn (0.46 g. 2 mmole) wherein Tos is the tosyl group (para-toluene- sulfonyl), and Bzl is the benzyl.

Specifically, Boc-Gly-PAM-resin is reacted according to the preset standard program provided by Applied Biosystems (Foster City, California) as described in the Manual for Peptide Synthesizer 430A (1987).

All the Double Couple Cycles Use the Following Pattern:

1 33% TFA in DCM for 80 seconds;

2 50% TFA in DCM for 18.5 minutes;

3 Three DCM washes;

4 10% DIEA in DMF for about 1 minute;

5 10% DIEA in DMF for about 1 minute;

6 Five DMF washes;

7 First coupling period,. 42 minutes;

8 Three DMF washes;

9 10% DIEA in DMF for 45 seconds;

10 One DMF wash; and

11 Three DCM washes.

- End of First Half of Cycle -

12 Second coupling period, 26 minutes;

13 One DMF wash; and

14 Give DCM washes. Boc-Gln was also coupled with this pattern.

All the Single Couple Cycles

Conform to the Following Pattern:

1) 33% TFA in DCM for 80 seconds;

2) 50% TFA in DCM for 18.5 minutes;

3) Three DCM washes;

4) 10% DIEA in DMF for about 1 minute;

5) 10% DIEA in DMF for about 1 minute;

6) Five DMF washes;

7) Coupling period, 26 minutes; and

8) Five DMF washes.

Boc-Ile was coupled with this pattern. Solvents are delivered in volumes of 3 - 8 ml to make a suspension of the resin.

The cycle for Boc-Pro is performed by means of a single couple cycle which is identical as for Boc-Arg (Tos) from steps 1 through 6. However, Boc-Pro (0.43 g, 2 mmole) is dissolved in 3 ml of dichloromethane and treated in the activator vessel with 1 mmole of DCC. After eight minutes, the Boc- Pro symmetric anhydride formed is transferred to the concentrator vessel where most of the dichloromethane is removed at 15°C by purging with nitrogen gas.

After addition of two 2 ml portions of DMF, the reagent is transferred to the reaction vessel and is allowed to react for 26 minutes. When the resin is finally washed with five dichloromethane washes, and when the Kaiser test is negative, the Boc-Pro-Arg (Tos)-Gly-PAM-resin was processed further. The cycle for Boc-Cys(B21) is performed by a single couple cycle similar to that for Boc-Pro. The cycles for Boc-Asn and Boc-Gln are performed by means of a double couple cycle as for Boc-Arg(Tos). However, 4.0 of HOBt and 0.3 ml of DCM are used to dissolve the Boc amino acid prior to activation.

The Boc group of Boc-D-Trp-Ile-Gln-Asn- Cys(Bzl)-Pro-Arg(Tos)-Gly-polymer had to be removed under special conditions to prevent oxidative

destruction of D-tryptophan by air under conditions of acid cleavage. The resin was treated with 10 ml of 30% TFA in methylene chloride, containing 1% mercaptoethanol, an inhibitor of air oxidation which may also act as a scavenger for potentially damaging tertiarybutylcarbonium ions which can react with the indole ring of tryptophan. After shaking for 2.5 minutes, the reaction vessel was drained of acid reagent. Fresh acid reagent was added (second time) and shaking was continued for another 2.5 minutes. The reaction vessel was drained once again, and fresh acid reagent was added (third time), and shaking was allowed to continue for an additional 25 minutes (total treatment lasts 30 minutes). The reaction vessel is drained and the cycle of neutralization coupling is continued as usual. Most of the removal of the t-Boc group probably occurs in the first few minutes of the reaction, so that the two initial drainings rapidly remove t-butyl σarbαnium ions and prevent further damage to the Trp residue over the acidolytic period (normally performed as one

30-minute step).

A fraction of the final product

PMP(Bzl)-D-Trp-Ile-Gln-Asn-Cys (B21)-Pro-Arg(Tos)-Gly- resin (0.5 mmol) was suspended in 50 ml of methanol, treated with 50 ml liquid ammonia, and the suspension was stirred for three days. The suspension was then filtered, and the resin was dissolved in three portions of hot dimethylforamide, 10 ml each, and the pooled dimethylforamide was added to about 200 ml of water. The gelatinous product was collected, washed with water, air dried and comminuted with mortar and pestle to 363 mg of powder of the protected PMP(Bzl)- octapeptide amide,PMP(Bzl)-D-Trp-Ile-Gln-Asn-Cys ( Bzl)-Pro-Arg(Tos)-Gly-NH₂.

The protected PMP(Bzl)-oCtapeptide amide in the amount of 150 mg was dissolved in 200 ml of liquid ammonia and treated with sodium until a permanent bluish color was obtained for a relatively short period of time, about ten to about twenty seconds. The ammonia was removed in a vacuum and the solid residue containing the deprotected peptide was dissolved in 20 ml of 50% acetic acid. The solution was added to 1.5 liters of water and was neutralized to pH 7.0 by the addition of 16.5 ml of concentrated ammonium hydroxide. The cyclic disulfide was formed by oxidation, by the addition of 15 ml of 0. 01 M potassium ferricyanide. An additional 5 ml of potassium ferricyanide was added to insure the oxidation of the sulfhydryl groups. The solution was stirred for 20 minutes. The ferro and ferricyanide ions formed during reaction were, removed by treating the solution with 10 g (20 mmole) of ion exchange AG3-X4A (chloride form), manufactured by Bio-Rad, and

stirring the suspension for 20 minutes. The suspension was then added to a column containing 20 grams of the same ion exchange resin for a complete exchange. The eluent was pumped into a uBαndpak C18 preparative column (2.14 cm x 20 cm) (Waters

Associates) and eluted with a linear gradient of 0%-60% solvent B (100%-40% solvent A) for five minutes at a rate of 12 ml/min, monitoring at a wavelength of 220 nanometers. The linear gradient was delivered by combining solvents delivered from pump A (pumping solvent A - 0.05% trifluoroacetic acid) and pump B (pumping solvent B - 60% acetonitrile, 40% of 0.05% trifluoroacetic acid) in a Gilson Model 811 dynamic mixer, with a Gilson 803C Manometric Model.

The highly purified antagonist thus obtained [PMP¹, D-Trp², Arg⁸] oxytocin in the amount of 74 mg had only one peak on the HPLC, using a C18 uBondpak analytical column 0.4 x 30 cm, using 60% solvent B, 40% solvent A as the eluent where solvent B = 60% acetonitrile and 40% of 0.05% trifluoroacetic acid and solvent A comprised 0.05% trifluoroacetic acid.

EXAMPLE II

For comparison with [PMP¹, D-Trp^a,

Arg⁸] oxytocin, two related compounds were synthesized. One of these was the compound described by

Manning, et al., J. Med. Chem., 26:1607-1613 (1983).

This compound can be called [PMP¹,D-Phe²,Phe³,

Ile⁴,Arg⁸] oxytocin. The other compound was

[PMP¹,-Trp²,Phe³,Ile⁴,Arg⁸] oxytocin. The

three compounds were comparatively studied in bioas- says. The protocol used for the oxytocin bioassay procedure is derived from procedures described in a paper by Sawyer, et al., Endocrinology, 106;81 (1980), which in turn was based on reports of Munsick, Brit.

J. Pharmacol., 3:328 (1960), and HoIton, Brit. J.

Pharmacol., 3:328 (1948). The assay calculations for the pA₂ estimates are described by Schild,

Brit. J. Pharmacol. (1947). The major difference in procedure from those previously reported was

the integration of the area under the contraction instead of merely calculating the amplitude. Integration provides more consistent and reliable

results, although the pA₂ estimates are about

a factor lower than those reported using amplitude of the contraction as the endpoint.

Method: Animals. A 1.5 cm piece of uterus from a virgin rat (Holtzman) in natural estrus is used for the assay. Buffer/Assay Bath. The buffer used is Munsick's. This buffer contains 0.5 mM Mg++ which reduces the pA₂ estimates, but the results are reported to correlate better with the in vivo data (Sawyer, et al., 1980). The buffer is gassed continuously with 95% oxygen: 5% carbon dioxide giving a pH of 7.4. The temperature of the assay bath is 37°C. A 10 ml assay bath is used that contains a water jacket for maintaining the

temperature and inlet and outlet spikets for adding and removing buffer.

Polygraph/Transducer. The piece of uterine tissue used for the assay is connected to a Statham Strain Gauge Force Transducer which in turn is attached to a Grass Polygraph Model 79 for

monitoring the contractions.

Assay Protocol. (a) The tissue is equilibrated in the assay bath for one hour with washing with new buffer every 15 minutes. One gram of tension is kept on the tissue at all times.

(b) The tissue is stimulated initially with oxytocin at 10 nM to "acclimate" the tissue and with 4 mM KCL to determine the maximum contractile response.

(c) A cumulative dose response curve is then done with oxytocin and a concentration of oxytocin equivalent to approximately 80% of the maximum used for estimating the pA₂ of the

antagonist.

(d) The tissue is exposed to oxytocin (Calbiochemical) for one minute and washed out. There is a three minute interval before addition of the next dose of the agonist or antagonist. When the antagonist is tested, it is given five minutes before the agonist. The agonist is given for one minute. All responses are integrated using a 7P10 Grass integrator. This is the major difference between our protocol and others in the literature who usually measure amplitude of the contractions as the

response. A single concentration of oxytocin, equal to 80% of the maximum response, is used to test the antagonist. Three different concentrations of antagonists are used, two that will reduce the response greater than 50% (ideally this relation would be 25%, 50% and 75%). This is repeated three times for each dose of antagonist for a three point assay.

The anti-ADH activity is measured by the alteration in urine antagonist by ADH in the presence and absence of the antagonist to determine the specificity of the antagonist. The anti-ADH assay is described in Sawyer, et al., Endocrinology, 63:694 (1958).

Results of the comparative study are shown in Table A.

Compound C, comprising the novel compound of this invention, demonstrated a higher anti-oxytocic activity than either of the other compounds. Further, it had a much lower anti-ADH activity. The ratio of anti-OT/anti-ADH for

Compound C is approximately 243, while this ratio for Compound A was 1.8 and for Compound B 0.7. The data therefore indicates that Compound C can be expected to produce less anti-ADH side effects when administered as an oxytocin antagonist then either Compounds A or Compound B.

Claims

CLAIMS We claim :

1 . The oxytocin antagonist represented by the formula

PMP-D-Trp-Ile-Gln-Asn-Cys-Pro-Arg-Gly-NH₂; S S

wherein PMP is ß,ß-cyclopentamethylene-ß-mercaptopropionic acid, D-Trp is the D form of tryptophan, and lie. Gin, Asn, Cys, Pro, Arg, and Gly are, respectively, the L forms of isoleucine, glutamine, asparagine, cysteine, proline, arginine, and

glycine.

2. The method of inhibiting the effect of oxytocin in a female mammal, comprising administering to said female mammal the oxytocin antagonist of claim 1 by an effective route and in an effective amount to function as an oxytocin antagonist.

3. The method of claim 2 in which said mammal is a pregnant woman.