Int. J, Peptide Protein Res. 13, 1979,137-141 Published by Munksgaard, Copenhagen, Denmark No part may be reproduced by any process without written permission from the author(s)

SYNTHESIS O F OXYTOCIN USING IODINE FOR OXIDATIVE CYCLIZATION A N D SILICA GEL ADSORPTION CHROMATOGRAPHY FOR PURIFICATION G. FLOURET, S. TERADA,' T. KATO,* R. GUALTIERI and A. LIPKOWSKIt

Department o f Physiology, Northwestern University McGaw Medical Center, Chicago, Illinois, U.S.A.

Received 3 July, accepted for publication 7 September 1978 Oxytocin (OT) was synthesized employing the solid phase method. Resins made of copolymers o f polystyrene-I %-crosslinked with divinylbenzene gave better yields (73- 95%) o f Z - Cys (Bzl) - Tyr (Bzl)-Ile - Gln - A m -Cys(Bzl)-Pro -Leu - Gly -

N H 2 (I) than 2%-crosslinked resins (10-56%). Reduction o f I with Na-liq.NH3 and oxidation with Iz-MeOH at -40' minimized dimer and polymer formation, and resulted in good yields (49-54%) o f OT. The large volumes o f MeOH required when several grams of I are reduced and then oxidized were rapidly evaporated in vacuo, and the residue was desalted b y dissolving the peptide in a small volume o f glacial acetic acid and filtering t o remove the salt. OT was purified b y adsorption chromatography on a silica gel column with combinations of MeOH-CHCI3 o f graded polarity. Oxytocin elutes with 33%Me0H-CHCl3. After t w o purification steps b y adsorption chromatography, the resulting OT was found to be homogeneous. The hormone was characterized chemically and found to be active biologically. Key words: iodine; oxidative cyclization; oxytocin, synthesis; silica gel adsorption chromatography.

In connection with receptor studies the need arose for preparation of oxytocin (OT) in 0.5-1 .O-g batches, in order to avoid repetitive

* Current address: Laboratory of Biochemistry, Faculty o f Science, Kyushu University, Fukuoka, Japan. t Visiting investigator from Warsaw University, Department of Chemistry, Poland. Abbreviations employed follow the recommendation of IUPAC-IUB Commission on Biochemical Nomenclature (in Biochem. J. (1972) 126, 773-780). Additionally, the following abbreviations were used: Z = benzyloxycarbonyl: Boc = tert-butyloxycarbonyl; Bzl = benzyl; TFA = trifluoroacetic acid; Py = pyridine; OT = oxytocin. 0367-8377/79/020137-05

syntheses, characterizations and bioassays of smaller batches. Previous syntheses of OT have been mostly accomplished by the stepwise active ester method (Bodanszky & du Vigneaud, 1959), condensation of fragments (Schrijder & Liibke, 1966), or the solid-phase technique (Bayer & Hagenmaier, 1968; Manning, 1968; Ives, 1968; Beyerman & In't Veld, 1969), employing a variety of functional groups and coupling reagents. More recently, solid-phase syntheses, using active esters and 1-hydroxybenzotriazole (Kahn & Sivanandaiah, 1976) and benzhydrylamine resins (Hruby et d., 1977; h v e et ~ l . , 1977) and solution syntheses (Hase & Walter, 1973; Kisfaludy & Schon 1975) have been described. However,

$02.00/0 0 1979 Munksgaard, Copenhagen

137

G. FLOURET ET AL.

these syntheses are convenient for the preparation of about 50mg of OT. We report here synthetic and chromatographic modifications which allow the synthesis of OT, in amounts more than 10-fold higher than obtained by usual methods. The procedure of Manning (1968) for the solid-phase synthesis of Z Cy s(Bzl)Tyr(Bzl)Ile -Gln - Asn - Cys(Bzl)- ProLeu-Gly-NH, (I) were followed with only minor modifications (Flouret et ul., 1973). The 1%crosslinked resin gave the highest yields of I (15-17 g, 73-95% yield), consistently higher than 2%-crosslinked resins (10-56%). After reduction of I with Na in liquid NH3 (Sifferd & du Vigneaud, 1935), and evaporation of the liquid ammonia,, oxidative cyclization of residual oxytoceine was accomplished in MeOH solution, in acid medium, by adding 0.1 M 12MeOH either at room temperature or preferably at -40" in a dry ice acetone bath. The low temperatures were expected to lower the lunetic energy of the peptide molecules and thus minimize intermolecular collisions leading to dimer and polymer. The uptake of I2 was instantaneous and quantitative. Iodination of the Tyr residue was not detected, alkaline conditions being required for rapid iodination of OT (Flouret e l al., 1977). Similar conditions have been employed by Kamber & Rittel (1968), who used 12-MeOH (at room temperature for 1h) for removal of S-trityl groups and oxidative cyclization of a nonapeptide fragment of calcitonin, as well as by Jones el ul. (1973) who used 12-80%AcOH for removal of S-trityl groups and oxidative cyclization in the synthesis of [Arg' ]vasopressin. The removal of 1-21 MeOH in large reaction batches was accomplished very rapidly (0.51 h), in contrast with the long periods of time (several hours, or overnight if lyophilized) required to remove water in methods requiring aqueous medium.

-

* Attempts to evaporate the liquid ammonia solutions in the presence of AcOH by sweeping the ammonia with a stream of N, (Wdti & Hope, 1973) followed by oxidative cyclization of a dilute aqueous solution of residual oxytoceine with K:Fe(CN), , gave average yields (29-3576) of OT and we could not reproduce their 62% yield. Dr. Hope confumed that he also repeatedly failed to reproduce the high yield reported (private discussions, 1975). 138

FRACTiON NO.

FIGURE 1 Elution profile of second step in purification, by the procedure of Manning et 01. (1 968): a) Of synthetic OT obtained employing K,Fe(CN), in aqueous solution for the oxidative cyclization step. Nonapeptide I (178mg) was converted to OT by reduction with Na in liq.NH, followed by oxidation with K,Fe(CN),; desalting of product was accomplished as described by Manning, on Sephadex G-15 column (2.6 X 100cm) employing 50% AcOH as the eluent and all peptide materials, detected by the Folin-Lowry method, were combined and lyophilized; the lyophilisate was subjected to the second gel fiitration step, shown in this figure, on the same Sephadex G-15 column with 0.2 N AcOH as the eluent (peaks I, polymer; 11, dimer; 111, OT). The yield of OT after lyophilization was 37 mg (29%). b) Of synthetic OT obtained employing I, in methanolic solution at -40" for oxidative cyclization. The amount of starting peptide 1 and the purification steps are otherwise identical to those described in Fig. la The yield of OT after lyophilization was 6 8 mg (54%).

We compared the oxidative cyclization of small batches of I (after Na-liq.NH3 reduction) employing K3Fe(CN), (du Vigneaud e l al., 1960) or 12-MeOHat room temperature, or at -40". The product obtained after cyclization,

SYNTHESIS OF OXYTOCIN

in each case, was purified by the procedure of Manning et al. (1968) employing a desalting step on a Sephadex G-15 column, with 50% AcOH, and a second gel permeation chromatography step on the same column, with 0.2N AcOH as the eluting solvent to analyze the peptides (polymer, dimer, oxytocin) obtained in the first column. The elution profile (Fig. 1) shows that oxidation with I2 -MeOH at -40”, or at room temperature, gives higher yields (54 and 49% respectively after isolation) than with K3Fe(CN), (29%) and minimizes dimer and polymer formation. A yield of 30% in the synthesis of OT or derivatives employing K3 Fe (CN), is a “typical” yield (Hruby et al., 1977). The residue obtained after reductionoxidative cyclization of large batches of I was desalted by selectively dissolving the peptide in a few milliliters of AcOH and filtering to remove the salt; this solvent is then rapidly removed in vacuo at mild temperatures. This procedure makes it unnecessary to desalt by gel filtration on Sephadex G-15 as previously described by Manning et al. (1968). For the purification of larger batches of OT, we employed silica gel adsorption chromatography, since silica gel has high adsorption capacity (e.g. it has been used to purify a 22-g batch of thyrotropin releasing hormone by Flouret et ~ l . ,1972). Elution with MeOHCHC13 combinations of graded polarity, results in the separation of fast moving impurities followed by OT, with dimers, polymers and residual salts remaining in the column. To insure purity, OT thus obtained (average about 50% yield) was subjected to adsorption chromatography on a second silica gel column. After the second purification, usually completed in 5-8 h, yields of 250 mg (33%), 205 mg (27%), and 670mg (31%) of OT were obtained from 1.0, 1.0 and 2.8g, respectively, of protected nonapeptide I. These yields could have been optimized by increasing the ratio of silica gel used as an adsorbent, although this would require larger volumes of solvent to elute OT. Nevertheless, whereas the percentage yields of OT obtained by adsorption chromatography are average, the total yields are 10 to 20 times those attainable (and reported t o date) by other conventional methods of purification usually

yielding on the average 30-60mg of OT. In principle, the adsorption chromatography methods described can be escalated to yield 10 g (or more) of OT in the average laboratory. OT thus prepared had the same rotation, electrophoretic and chromatographic mobilities as an OT standard, as well as correct elemental and amino acid analyses. A sample of synthetic OT possessed an avian depressor* activity (Munsick el al., 1960) of 512 ? 25 U/mg when compared to a synthetic standard which had 5 10 k 23 U/mg. An oxytocic assay (Holton, 1948) employing van Dyke-Hastings solution containing 0.5 mM Mg++ (as modified by Munsick, 1960) showed this preparation and the OT standard to have identical uterotonic activity (500 U/mg). An adenylate cyclase assay employing bovine renal medullary membranes (Hechter el al., 1975) showed identical potency as the standard. EXPERIMENTAL PROCEDURES

Optical rotations were measured in 1-dm polarimeter tubes with a Rudolph polarimeter reading with a precision o f f 0.01”. Amino acid analyses were performed with a Durrum automatic analyzer on samples obtained by hydrolysis of the peptide in 6 N HCI at 110” for 2022 h. Gel permeation chromatography was conducted on Sephadex (3-15 or G-25 Pharmacia). Adsorption chromatography was conducted on silica gel 60 (Brinkman Instruments), particle size 0.063-0.2 mm. Column fractions were visualized with the aid of an ultraviolet monitor (LKB) at 280nm or by the Folin-Lowry color reaction, reading absorbance at 700nm in a Coleman Junior Spectrophotometer. Thin-layer chromatography and electrophoresis were performed on Eastman Chromagram silica gel thin layer sheets. For thin-layer electrophoresis a BrinkmanDesaga apparatus was used, at 4 0 0 V for 2 h . All solvents were of reagent grade. Boc-gIycine-resin

Boc-Gly was attached t o resins through a * This assay was performed in the laboratories of Drs. R. Walter and 1. L. Schwartz at Mount Sinai Medical School, New York City.

139

G . FLOURET ET AL.

benzyl ester bond as previously described 500 ml of 33% MeOH-CHC13; dimers and poly(Flouret et al., 1973). The degree of substitu- mers can be eluted with 40-50% MeOH-CHCl, . The desired OT-containing fractions were tion was 0.32-0.76 mmol/g. monitored on t.1.c. with silica gel thin-layer sheets and detected with Pauly and/or C12Z-Cys (Bz1)-Tyr(Bzll-ile-Gln-Asn-Cys(BzI)-Protoluidine color reactions (Rf = 0.4 in MeOH: Leu-Gly-NHz (I) The procedures described by Manning (1968) CHC13:AcOH:Hz0, 38:62:2:2). The yield of for solid phase peptide synthesis of OT were OT was 350mg (49%). To insure purity, OT followed except for minor modifications thus obtained was subjected to adsorption (Flouret et al., 1973). Thus, in most syntheses chromatography on a second silica gel column, 25% TFA-CHZCl2 was used and only in a few removing any traces of fast or slow-moving cases was 4 M HC1-dioxane used. Completion of by-products, and yielding 250mg of highly the coupling step was monitored by means of purified OT. Each column purification of OT the Kaiser test (Kaiser et al., 1970) Ammono- was usually completed in about 5-8 h. Yields lysis as described by Manning (1 968) yielded I. of 250mg (33%), 205mg (27%), and 670mg (31%) were obtained from 1.0, 1.0, and 2.8g, respectively, of protected nonapeptide I . OT Oxy tocin Protected nonapeptide I (1 .O g) was reduced thus prepared had [a],"- 22.0" (c 0.5, 1 N with Na in liq.NH3 (300ml) freshly distilled AcOH). Expected (Ressler & du Vigneaud, from Na until a faint blue color was detected 1957) [ L Y ] ~- 23.1" (C 0.5 1 , 1 N AcOH). (disappearing in 15-30s). The solvent was Anal. calc. for C45H74N12016S2.l C ~ H 4 0 2 , removed in vacuo and the residue was dissolved 2 H 2 0 : C, 49.3; H, 6.88; N, 14.9. Found: in 1,000 ml of MeOH and 6 N HCl was added C, 49.0; H, 6.75; N, 15.2. Amino acid analysis (Spackman et a[., dropwise until the color of wet hydrion paper 1964) showed the following molar ratios changed to a color corresponding to pH 4-5. This solution was cooled to -40" in a dry ice with Gly = 1.0: C y s 2.0; Tyr 0.9, Ile 0.95, acetone bath and it was treated with 0.1 M I2 Glu 0.90, Asp 1.0, Pro 1.05, Leu 1.0 and Gly in MeOH until a yellow color of excess iodine 1.O. The sample was homogeneous on t.1.c. with developed (6.9 ml). The uptake was approxi- MeOH:CHC13 :HzO:AcOH (62:38:2:2) and mately 95-100% of theory. This solution was with BuOH:AcOH:H,O, (4:1:5) and on t.l.c. passed through ion exchange resin AG 1-X2 (Cl-) (with 0.1 N Py-AcOH, pH 5.6) showing motilito remove iodide and the solution was evapor- ties identical to a synthetic standard of OT. ated to dryness in vucuo. The residue was triturated with 5 ml glacial acetic acid containing ACKNOWLEDGEMENTS 1% MeOH and the peptide solution was separated from the undissolved salt by filtration. We are indebted to Ms. Hildegard Beck for elemental After washing the salt precipitate with ad- analyses, to Dr. Arthur Veiss for amino acid analyses, ditional AcOH the combined filtrate was to Ms. T. Dixon for technical assistance in the oxyevaporated to dryness. The residue was dis- tocic assays, to Drs. R. Walter and 1. Schwartz for solved in MeOH and monitored by applying avian depressor assays, and to MI.Richard M. Simpson (Rohm and Haas, Co., Philadelphia, Pa.) for generous heavily to silica gel sheets and developing with samples of macroreticular resins. We also acknowledge n-Bu0H:PrOH: benzene: 1% aqueous AcOH support of this research by the National Institutes of containing 0.5% Py (12:1:1:16). Standards of Health Grants HD06237 and HD-08869. OT and monoiododooxytocin revealed n o detectable iodination in samples containing REFERENCES about 50pg of OT (Flouret et al., 1977). The residual oxytocin was dissolved in 5-1Oml of Bayer, E. & Hagenmaier, H. (1968) Tetrahedron Lett. 33%MeOH-CHC13 and applied to a column con17,2037-2039 taining 40 g of silica gel; after elution with 25% Beyerman, H. C . & In't Veld, R. A. (1969) Rec. Truv. MeOH-CHC13 (200 ml) to remove small amounts Chim. 88,1019-1027 of fast moving components, OT was eluted with Bodanszky, M. & du Vigneaud, V. (1959) J. Am. 140

SYNTHESIS O F OXYTOCIN

Chem. Soc. 81,5688-5691 du Vigneaud, V., Winestock, G., Murti, V. V. S., Hope, D. B. & Kimbrough, R. D., Jr. (1960) J. Biol. Chem. 235, PC 64-66 Rouret, G., Alter, A. & Gendrich, R. (1972) J. Labelled Compd. 8,53-61 Flouret, G . R., Arnold, W. H., Cole, J. W., Morgan, R. L., White, W. F., Hedlund, M. T. & Rippel, R. H. (1973) J. Med. Chem 16,369-373 Flouret, G . , Terada, S., Yang, F., Nakagawa, S. H., Nakahara, T. & Hechter, 0. (1977) Biochemisrry 16,2119-2124 Hase, S. & Walter, R. (1973) Int. J. Peptide Protein Res. 5,283-288 Hechter, O., Kato, T., Nakagawa, S. H., Yang, F. & Floutet, G. (1975) Proc. Natl. Acad. Sci. U.S.A. 72,563-566 Holton, P. (1948) J. Pharmacol, 3,328-334 Hruby, V. J., Upson, D. A. & Agarwal, N. S. (1977) J. Org. Chem. 42,3552-3556 Ives, D. A. J. (1968)Can. J. Chem. 46,2318-2320 Jones, D. A., Mikulec, R. A. & Mazur, R. H. (1973) J. Org. Chem. 38,2865-2869 Kahn, S. A. & Sivanandaiah, K. M. (1976) Terrahedron Left., 119-200 Kaiser, E., Colescott, R. L., Bossinger, C. D. & Cook, P. I. (1970) Anal. Biochem. 34,595-598 Kamber, B. & Rittel, W. (1968) Helv. Chim. Acra 51, 206 1 -2064 Kisfaludy, L. & Schon, I. (1975) Acta Chim. Acad. Sci Hung. 84,227-228

Live, D. H., Agosta, W. C. & Cowburn, D. (1977) J. Org. Chem 42,3556-3561 Manning, M. (1968) J. Am. Chem. SOC. 90, 13481349 Manning, M., Wuu, T. C. & Baxter, J. W. M. (1968) J. Chromatogr. 38, 396-398 Merrifield, R. B. (1963) J. Am. Chem. SOC.85, 21492154 Munsick, R. A. (1960) Endocrinology 66,451 -457 Munsick, R. A., Sawyer, W. H. & Van Dyke, H. B. (1960) Endocrinology 66,860-871 Ressler, C. & du Vigneaud, V. (1957) J. Am. Chem. Soc. 7 9 , 4 5 1 1 4 5 1 5 Schroder, E. & Lubke, K . (1966) in The Pepfides, vol. 2 , p. 281, Academic Press, New York Sifferd, R. H. & du Vigneaud, V. (1935) J. Biol. Chem. 108,753-761 Spackman, D. H., Stein, W. H. & Moore, S. (1958) Anal. Chem. 30, 1190-1206 Wdti, M. & Hope, D. B. (1973) Experientia, 389

Address: Dr. George E'louret Department of Physiology Northwestern University Medical School 303 East Chicago Avenue Chicago, Illinois 606 1 1 U.S.A.

141