3164
A more critical test of the method was the synthesis storage, and should possess the suitable mechanical of leucylalanylglycylprolylphenylalanylarginine from properties. crystalline phenylalanylarginine prepared by the NCA To test the possible use of chemically reactive polymethod. The crude reaction mixture was purified on mers in acylation reactions we prepared insoluble, silica gel H to give a 60% yield of hexapeptide which high molecular weight active polyesters of acetic acid -62.4O moved as a single component by tlc, [a]5sg (Ha) and benzoic acid (IIb) by allowing the corre(c 1 %, 6 N HCl), and which showed the following amino sponding chlorides to react with poly-4-hydroxy-3acid ratio in its acid hydrolysate: Leuo.9sAlal,ol- nitrostyrene cross-linked with 4 % divinylbenzene Glyo.g8Prol.ooPheo.~~Arg~.o~. Sequential treatment of (I) in dimethylformamide (DMF), in the presence of the purified hexapeptide with a 5-8% excess of the pyridine. The high molecular weight active esters NCA's of glutamic acid, isoleucine, and proline, re(11) contained approximately 5 mmoles of acyl residues/ spectively, gave a product which was purified on g of insoluble polymer and could be stored at room Sephadex to give a 46% yield (based on hexapeptide) temperature in powder form without decomposition. of the nonapeptide, [aI5a9- 86.7" (c 1%, 6 N HCI). Treating IIa or IIb (1.0 g) in suspension in D M F The amino acid ratio after acid hydrolysis was: Pro2.04- (15 ml) with 0.5 mmole of tri-L-alanyl-p-nitrobenzyl I l e u o . g ~ G l u ~ . ~ ~ L e u ~ . ~ ~ A l a ~ , ~ ~ G l y ~ . ~ ~ P hester e ~ , (~L~-AAr ~g ~ ~. -~P~Nwas . B ) carried out at room temperaPhenylalaninamide, prepared from the NCA by ture for 4-5 hr. The polymer was removed by centreatment with NH3, was treated sequentially with the trifugation and the supernatants evaporated to dryness NCA's of aspartic acid, methionine, and tryptophan in uacuo. The solid residues were found to contain in a Waring Blendor. The crude product was precipia practically quantitative yield of the corresponding tated at pH 7 and purified on silica gel H to give a 30% acyl derivatives: acetyl-~-Ala~-PNB, mp 226", [ a ] * j ~ yield of the C-terminal l tetrapeptide sequence of -28.9" (c 0.58, DMF); benzoyl-L-Ala3-PNB, mp gastrin characterized as the crystalline hydrochloride, 216-218", [a]*'Df2.9" (c0.70, DMF). [crljSg - 17.5" (c 2%, methanol), -31.6' (c 1 %, DMF). Lccc/ u/u The amino acid composition, as determined by acid and LAP hydrolyses, was Tryo,g5Metl.ooAspl,ooPhel.ol+*(I2 H;"_ RCOHNR' 4- I NH4+0.95and Tryl,03Metl.05Aspo,g4Pheo.gg, respectively. The LAP cleavage left no peptide fragments detectable 0 0 by tlc. This tetrapeptide synthesis required a total I I of about 1 hr and the isolation procedures and crystalc=o H lization about 3 days. I I R Acknowledgment. We wish to thank Dr. Max Tishler I1 for guidance and encouragement throughout this in11: esters of: acetic acid (IIa); benzoic acid (IIb); 2-L-Phe vestigation. We also wish to acknowledge the excellent (IIc); Z-~-Ileu(IId); Z-L-PTO (IIe); N,S-Di-Z-L-Cys (IIQ; N-Z-a,. technical assistance of Messrs. Victor Garsky, John OBZ-L-G~U (IIg), where Z = C;H70C0 and Bz = C7H7 Sondey, and Jack Fabian. The successful preparation of active esters of N(12) J. C . Anderson, et al., Nature, 204, 933 (1964). blocked amino acids of type I1 enabled their utilization in peptide synthesis. Thus peptides with N- and Robert G. Denkewalter, H. Schwam, R. G . Strachan C-blocked terminal groups were obtained on coupling Thomas E. Beesley, Daniel F. Veber, Erwin F. Schoenewaldt H. Barkemeyer, William J. Paleveda, Jr. the insoluble active esters IIc to IIg with desired soluble Theodore A. Jacob, Ralph Hirschmann amino acid or peptide esters containing a free CYMerck Sharp and Dohme Research Laboratories amino group. Preferential removal of the N-blocking Division of Merck & Co., h e . group from the newly formed peptide enabled the Rahway, New Jersey repetition of the coupling reaction with an insoluble Received April 26, 1966 active ester of another N-blocked amino acid. Further repetitions of this set of reactions lead obviously to the elongation of the peptide chain and formation of Use of Polymers as Chemical Reagents. a peptide with a predetermined amino acid sequence. I. Preparation of Peptides The insoluble active esters IIc-g were derived from Sir: the following corresponding benzyloxycarbonyl amino acid derivatives: benzyloxycarbonyl-L-phenylalanine, Polymers of the type @-A containing covalently benzyloxycarbonyl-L-isoleucine, benzyloxycarbony1-Lbound groups A, which readily react with a low molecproline, N,S-dibenzyloxycarbonyl-L-cysteine,and Nular weight reagent B, can be used to synthesize combenzyloxycarbonyl-a-benzyl-L-glutamate, by their pound A-B according to eq 1. To increase yields I in DMF by the DCC coupling with polymer @-A + B +A-B + 0 method. 192 The insoluble polymers IIc to IIg contained and to facilitate the synthetic procedure insoluble polyper gram approximately 1.0-1.5 mmoles of amino mers of the above type may be added in large excess acid and could be stored at room temperature without decomposition, similarly to IIa and IIb. In suspento a solution of B in a suitable solvent. At the end of the reaction the insoluble polymer can be removed by sion in inert organic solvents polymers IIc to IIg filtration or centrifugation. The filtrate which is showed chemical behavior similar to that of the cordevoid of A should thus contain only A-B and un(1) J. C. Sheehan and G. P. Hess, J . Am. Chem. Soc., 77, 1067 (1955); M. Rothe and F. W. Kunitz, Ann., 609, 88 (1957); D. F. Elliott and D. reacted B. The most suitable polymers @A to be W. Russell, Biochem. J., 66,499 (1957). used as chemical reagents should contain a relatively (2) M. Fridkh, A. Patchornik, and E. Katchalski, J . Am. Chem. Soc., large amount of A, should show high stability on 87, 4646 (1965).
+m2
Journal of the American Chemical Society
/
88:13
/ Jury 5, I966
=
3165
responding low molecular weight amino acid active excess of insoluble active ester which can be readily esters. removed at the end of the reaction. It is thus possible N,S-Di-Z-L-Cys-Gly-OBz was obtained by allowing to increase yields and to shorten the time of the coupling IIf (1 g containing 1 mmole of cysteine) to react reaction. with glycine benzyl ester (0.5 mmole) in D M F (20 ml) The application of the new method for the synthesis with stirring for 5-8 hr at room temperature. The of various low and high molecular weight peptides is polymer was removed by centrifugation and washed under investigation. with DMF, and the combined D M F solutions were Mati Fridkin, Abraham Patchornik, Ephraim Katchalski, evaporated to dryness in vacuo. The residue was Department of Biophysics, The Weizmann Institute of Science dissolved in wet ethyl acetate and washed with 1 N Rehovoth, Israel HCl, 5 % NaHC03, and water. On evaporation of Received April 29, 1966 the ethyl acetate a chromatographically pure solid product was obtained; yield 98%, based on the Synthesis of Purine Cyclonucleoside Having a amount of glycine benzyl ester employed; mp 1168,2'-O-Anhydro Linkage 118", [a]25D -45.3" (C 1.3, DMF) [lit. mp 118-119°,3 [CY]?~D - 45.5" (c 2.0, DMF)3]. Z-L-Pro-Gly-OBz Sir : was obtained analogously from IIe and glycine benzyl Synthesis of various purine cyclonucleosides conester; yield 90%, mp 87-88" (lit. mp 88-8904). Ztaining S-anhydro linkages has been reported. L-Ileu-(L-Ala)5-PNB was obtained by coupling IId with These compounds were easily desulfurized to afford H2N-(L-Ala)5-PNB in D M F at room temperature; 2 '-deoxyadenosine, 3 '-deoxyadenosine (cordycepin), yield 71%, mp 258-260", [O!]~'D -91.1" (c 0.34, diand 5 '-deoxyguanosine. chloroacetic acid). An identical peptide was obtained In analogy to similar studies4 with pyrimidine cyclo~ (c by the DCC method; mp 261-265", [ a ] ? j-90.5" nucleosides, it would be of interest to investigate the 0.46, dichloroacetic acid). N,S-Di-Z-glutathione diphysical and chemical properties of the purine cyclobenzyl ester (N-Z-L-G~U-(~-OBZ)L-CYS- ( -S -Z) Glynucleosides containing 0-anhydro linkages. Although OBz), mp 156", [ a I z 5-35.5" ~ (c 1.0, methanol) [lit. ~ (c 1.0, m e t h a n ~ l ) ~ ] an earlier attempt5 to obtain an 8-hydroxypurine mp 158-159°,3 [ a I z 5 -35.5' was obtained in 85% yield from IIg and S-Z-L-CYS- nucleoside met with little success, a novel method for the introduction of a hydroxyl group in the 8 position Gly-OBz. The latter was derived from the N,S-diof preformed purine nucleoside has been developed Z-L-CYS-G~Y-OBZ preparation obtained from IIf and recently in our laboratory.6 Gly-OBz on treatment with HBr in glacial acetic acid. We wish to report the synthesis of 8,2'-anhydro-8Treating L-alanine p-nitrobenzyl ester with excess IIc, hydroxy-9-P-~-arabinofuranosyladenine(I), the first removal of the Z-group (by HBr-CH3COOH) from the purine cyclonucleoside having an 0-anhydro linkage. dipeptide formed (Z-L-Phe-L-Ala-PNB), neutraliza2',3'-O-Isopropylideneadenosine (11) was brominated tion with triethylamine in DMF, and treating the by the method of Holmes, et d.,'to give 8-bromo-2',dipeptide ester with a new excess of IIc gave Z-L3'-O-isopropylideneadenosine, mp 215-217 " (Anal. Phe-L-Phe-L-Ala-PNB in an over-all 84 % yield ; mp Calcd for C13H1604N5Br:C, 40.69; H, 4.18; N, 18.13. 159", [Q]?~D-26.8" (c 0.9, DMF). The same peptide prepared by the DCC method gave mp 160", [ a I z 5 ~ Found: C, 40.60; H, 4.35; N, 18.48); ultraviolet 0.60, (B) 0.73. ; OH- 264 mp; - 26.2" (c I .39, DMF). Z-L-Phe-L-Phe-L-Phe-NH2 A,,EtOH 265 mp, A,&E+ Acetylation of this compound gave 8-bromo-2',3 'was prepared analogously in 81 yield, mp 227', 0-isopropylidene-5 '-0-acetyladenosine (111), mp 158[ a I z 5-31.5" ~ (c 0.8, DMF). The same peptide prepared by the DCC method gave mp 226", [ a I z 5 ~ 159" (Anal. Calcd for C15HIBOjN6Br:C, 42.06; H, 4.47; N, 16.38. Found: C, 41.92; H, 4.19; N, 16.71); - 32.1 O (C 1.2, DMF). Rf(B) 0.82. The over-all yield of I11 from I1 was ca. A comparison of the method for peptide synthesis 30%. Compound I11 could be obtained also from described here with that of Merrifieldj reveals that 2',3'-0-isopropylidene-5'-0-acetyladen~sine~ by the whereas in the "solid-phase peptide synthe~is"~ it is the peptide which is bound to the insoluble carrier (1) M. Ikehara and H. Tada, J . Am. Chem. Soc., 85, 2344 (1963); 87, 606 (1965). and the N-blocked amino acid active ester is added (2) M. Ikehara, H. Tada, and K. Muneyama, Chem. Pharm. Bull. while in solution, in our case a solution of free peptide (Tokyo), 13, 639 (1965). ester is added to an insoluble N-blocked amino acid (3) M. Ikehara and H. Tada, Abstracts of papers presented at the 21st Meeting of the Pharmaceutical Society of Japan, 1965, p 258. active ester. Furthermore, purification of the inter(4) J. J. Fox and I. Wempen, Aduan. Carbohydrate Chem., 14, 283 mediate peptides formed during synthesis can be (1959); A. M. Michelson, "The Chemistry of Nucleosides and Nucleotides," Academic Press Inc., New York, N. Y . ,1963, p 15,68. readily effected in our method, since these peptides ( 5 ) Condensation of 8-hydroxyadenine chloromercury salt with a proare liberated into solution. In the Merrifield synthesis, tected ribosyl halide by Davoll's procedure gave only unidentified proon the other hand, peptide purification can be carried ducts (M. Ikahara and H. Tada, unpublished experiment). (6) M. Ikehara, H. Tada, and K. Muneyama, Chem. Pharm. Bull. out only after detachment of the final product from the (Tokyo), 13, 1140 (1965). polymeric carrier. In the procedure described here (7) R. E. Holmes and R. K. Robins, J . Am. Chem. SOC.,86, 1242 the reaction between peptide ester and active amino ( 1964). (8) Rf(A) stands for the R fvalue of the paper chromatography carried acid ester can be carried out in the presence of a large
-
(3) M. Sokolovsky, M. Wilchek, and A. Patchornik, J. Am. Chem. Soc., 86, 1202 (1964). (4) J. Kurtz, unpublished results. ( 5 ) R. B. Merrifield, J . Am. Chem. Soc., 85, 2149 (1963); R. B. Merrifield, Biochemistry, 3, 1385 (1964); R.B. Merrifield,J. Org. Chem., 29, 3100 (1964); G. R. Marshall and R. B. Merrifield, Biochemisfry, 4, 2394 (1965).
out by ascending technique in solvent A. Solvent A : 1-butanol-acetic acid-water, 4:1:5 (upper layer was used); solvent B: 1-butanol-water, 86: 14; solvent C: solvent %concentrated ammonia, 1OO:l; solvent D: water adjusted at pH 10 with ammonia; solvent E: l-propanolwater, 3 :1 : solvent G: 2-propanol-concentrated ammonia-water, .I., .* I .I .A. (9) D. M. Brown, J. L. Haynes, and A. R. Todd, J. Chem. Soc., 3999 ( 1950).
Communications to the Editor
