摘要:

1977 1905Formation and Degradation of Urea Derivatives in the Azide Method ofPeptide Synthesis. Part 1. The Curtius Rearrangement and UreaFormation tBy Ken Inouye,' Kunio Watanabe, and Masaru Shin, Shionogi Research Laboratory, Shionogi & Co., Ltd.,The azide method of peptide synthesis (R'CON, + WNH, RICO.NHR*) has been investigated with respectto the effect of reaction conditions especiallyon the formation of urea derivatives ( RINH.CO.NH R2) as side products.Z-Gly-Phe-N, and H-Gly-OBut were used as model compounds for RlCON, and R2NH,, respectively, and therate of formation of peptide was compared with that of urea under various conditions. Peptide formation wasestimated from the consumption of R2NH, and urea formation by the extent of the rate-determining Curtius re-arrangement (RlCON, + RlNCO + N2).Data have also been obtained for other azides and amines. Theresults show that the conditions currently used in azide couplings (ca. 0.1 M-solutions and 0-5 "C) are generallyadequate for minimising the side reaction. On the other hand, some urea derivatives, including Boc-Gly-Tyr-Ser-N H *CH (CH,.CH,-SMe) .N H CO- Glu (OBut) -His- Phe- Arg-frp- Gly- OH (6) and Z- Lys( Boc) - Pro-Val- Gly- N H .CH-([CH,]4.NH-Boc)*NH.CO-Lys(Boc)-Arg-Arg-NH, (5) can be synthesisedfrom RlNCO and R2NH2. These ureashave been compared with the corresponding peptides in terms of chromatographic behaviour; the results suggestthatwhen Rland R2are fairly large and complex as in (5) and (6) the separation of peptide from urea will presentconsiderable difficulties.Fukushima-ku, Osaka 553, JapanTHE azide method of peptide synthesis, introduced byCurtius over 70 years ago,l is still one of the majorcoupling procedures in current use.One reason for itscontinued use is that it permits coupling with minimumprotection on amino-acid side chains. This is especiallyadvantageous for the synthesis of long-chain peptides.t The results described in this paper were presented a t the 1 l t hSymposium on Peptide Chemistry, Kanazawa, Japan, October1973 (Proceedings of the 11th Symposium on Peptide Chemistry,ed. H. Kotake, Protein Research Foundation, Osaka, Japan,1974, p. 30).All the amino-acid residues mentioned are of the L-configur-ation. Abbreviations used are those recommended by theIUPAC-IUB Commission on Biochemical Nomenclature (Bio-chem.J . , 1967, 102, 23; 1967, 104, 17; 1972, 128, 733);Z = benzyloxycarbonyl, Boc = t-butoxycarbonyl, DCC =dicyclohexylcarbodi-imide, HOSu = N-hydroxysuccinimide,HOBt : 1-hydroxybenzotriazole, TFA = trifluoroacetic acid,DMF = NN-dimethylformamide, ACTH = adrenocorticotrophichormone (corticotrophin) .Since Weygand et al.3 found that azides do give racemicproducts in the presence of an excess of base, racemisation inazide couplings has been observed in a number of instance^.^However, the risk of racemisation seems to be generally lowerthan with other coupling methods, including the DCC-HOBtprocedure,S as Young has noted recently.gAnother reason, probably the most important, is that themethod carries the least risk of racemisation.21 Becauseof this the azide method has been employed ratherroutinely for peptide-petide couplings, in spite of thewell known fact that various kinds of undesirableT.Curtius, Bey., 1902, 35, 3226.N. A. Smart, G. T. Young, and M. W. Williams, J . Chem.SOG., 1960, 3902.F. Weygand, A. Prox, and W. Konig, Chem. Ber., 1966, 99,1451.G. W. Anderson, J. E. Zimmerman, and F. M. Callahan,J . Amer. Chem. SOG., 1966, 88, 1338; P. Sieber, M. Brugger, andW. Rittel, in ' Peptides 1969,' Proceedings of the 10th EuropeanPeptide Symposium, ed. E. Scoffone, North-Holland, Amsterdam,1971, p. 60; P. Sieber, B. Riniker, M. Brugger, B. Kamber, andW. Rittel, Helv. Chim. Acta, 1970, 53, 2135; D.S. Kemp, 2.Bernstein, and J . Rebek, jun., J. Amer. Chem. SOC., 1970, 92,4756; D. S. Kemp, S. W. Wang, G. Busby 111, and G. Hugel,ibid., p. 1043; M. Dzieduszycka, M. Smulkowski, and E.Taschner, in ' Peptides 1972,' Proceedings of the 12th EuropeanPeptide Symposium, eds. H. Hanson and H. D. Jakubke, North-Holland, Amsterdam, 1973, p. 103.Peptides 1972,' Proceedings of the 12thEuropean Peptide Symposium, ed. H. Hanson and H. D. Jakubke,North-Holland, Amsterdam, 1973, p. 132.6 W. Konig and.R.,Geiger, Chem. Ber., 1970, 108, 788.6 G. T. Young, i1906 J.C.S. Perkin Iside reactions' tend to occur during the course of thecoupling procedure.The important side reactions associated with theazide method are shown in the Scheme. Among themRICO*NH*NH,{HNOa- N1ORlCO*NH*NH*NO -t RICO*NH, (amide formation)1-HIOR'NH,R'CON, -)c R1CO-NHR2 (peptide formation)I(RlNH),CO (sym.urea formation) /-Nafl,oR~NCOR'NHlR1NH*CO*NHR2 (urea formation)SCHEME\the formation of amide, RICO*NH,, first observed byPrelog and Wieland 8 has been reported most frequently.Honzl and Rudinger found that amide formation islargely suppressed when the preparation of the azidefrom the hydrazide is performed in an anhydrous andhomogeneous solution at low temperatures and lowacidity with an alkyl nitrite ( e g . t-butyl nitrite) as nitro-sating agent.9 It is generally accepted that the amidemay not be a product from the azide,71v*10 whereas theureas are derived from isocyanate, RlNCO, whichis formed from azide, RlCON,, by the Curtius re-a1-rangement.7,~~ The isocyanate produces a urea,R1NHCO*NHR2, when it reacts with an amine com-ponent, R2NH2.The isocyanate can also produce asymmetrical urea (RINH),CO on hydrolysis followed byreaction with unhydrolysed isocyanate. We will nowconsider the formation of the urea R1NH*CO*NHR2,which structurally resembles the desired peptide R1-CO*NHR2 most closely. The degree of resemblanceincreases with increasing size and complexity of thegroups R1 and R2. For fairly large peptides, it couldbecome extremely difficult to detect the presence of sucha closely related by-product and to separate it from thedesired compound. Thus urea formation is a seriousproblem in azide couplings, even though its occurrencehas been reported less frequently.12 In a search for a* The urea undergoes acidic hydrolysis in the following man-ner :73HI0R1CO*NHCHR2*NH*CO*NHR3 -----tRICO,H + RWHO + R3NH, + 2NH3 + CO,In the case of compound (l), however, the amounts of glycineand ammonia released upon hydrolysis ( ~ M - H C ~ ; 110 OC; 24 h)were much smaller than the theoretical values.This probablyindicates the liberation of hydantoic acid (NH,CO.NHCH,.CO,H) as a stable intermediate (see also following paper).E. Schnabel, AnnuZen, 1962,659, 168.8 V. Prelog and P. Wieland, Helv. Chim. Actu, 1946, 29, 1128;8 J. Honzl and J. Rudinger, CoZZ. Czech. Chem. Comm., 1961,10 Y . S. Klausner and M. Bodanszky, Synthesis. 1974, 649.for other observations see references cited in ref.9.26, 2333.solution of this problem we have investigated the rate offormation of urea in relation to the rate of formation ofthe peptide bond and have compared the resulting ureaswith the related peptides in terms of some chemicaland physicochemical properties. These studies wereperformed with Z-Gly-Phe-NH*NH2 and Z-Gly-Tyr-NH-NH, as model compounds and then with some pep-tide hydrazides related to an active fragment of cortico-trophin, [GlylJ- ACTH ( 1-1 8) -NH2.13The conversion of hydrazide into azide was carried outin aqueous tetrahydrofuran in the presence of 2-3equiv. of hydrochloric acid. The azide farmed uponaddition of sodium nitrite was extracted at pH 9 withethyl acetate. The resulting azide solution was quicklydried over magnesium sulphate prior to immediate useand, if necessary, the solvent was removed in vacuo.Throughout these operations the temperature wasmaintained at 0 "C.PreParatioN of Ureas.-The azide once formed spon-taneously undergoes the Curtius rearrangement to pro-duce the isocyanate.The azide and the isocyanate arecharacterised by i.r. bands at 4.75 and 4.5 pm, respect-ively.l* Therefore, if an amine component is added whenthe azide band has disappeared, urea formation takesplace rapidly, with concomitant disappearance of theisocyanate band. For example. when Z-Gly-Tyr-N, inethyl acetate was kept at 25 "C overnight and H-Gly-OBut was then aded, prompt formation of the ureaZ-Gly-NH*CH(CH,*C,H,*OH)*NH*CO-Gly-OBut (1) re-sulted.Compound (1) was isolated crystalline afterchromatography on silica gel. Amino-acid analysisrevealed the absence of tyrosine, as expected,* In thismanner, some other urea derivatives were synthesised.These ureas and the corresponding peptides are shown inTable 1, which also includes a symmetrical urea (3) whichmay be compared with the amide Z-Gly-Phe-NH,.Compound (3) was formed when a solution of Z-Gly-Phe-N, in wet ethyl acetate was kept at room temper-ature for a few days. Compounds (4) and ( 5 ) are relatedto the amino-acid sequence 11-18 of corticotrophin(ACTH) l6 and (6) to the sequence 1-10 of the hormone.16The ureas (l), (2), and (4) were clearly distinguished fromthe corresponding peptides by t.l.c., whereas (5) and (6)showed no separation from the peptides.These factssuggest that when the molecule is relatively small[as in (1) and (2)] or when polar side chains are fullyl1 T. Curtius, J. prakt. Chem., 1904, 70, 57; T. Curtius and H.Curtius, ibid., p. 158.la (a) M. A. Nyman and R. M. Herbst, J. Org. Chem., 1950, 15,108; (b) J. M. Hinman, E. L. Caron, and H. N. Christensen, J.Amer. Chem. SOC., 1950,72, 1620; (c) K. C. Hooper, H. N. Rydon,J. A. Schofield, and G. S. Heaton, J. Chern. SOC., 1956, 3148;( d ) G. L. Tritsch and D. M. Woolley, J. Amer. Chern. SOC., 1960,82, 2787; (e) K. Hofmann, T. A. Thompson, H. Yajima. E. T.Schwartz, and H. Inouye, ibid., p. 3715.18 H. Otsuka, M. Shin, Y. Kinomura, and K. Inouye, Bull.Chem. SOC. Japan, 1970, 48, 196.l4 R.Schwyzer and H. Kappeler, Helv. Chim. Actu, 1961, 44,1991.l6 H. Otsuka, K. Inouye, M. Kanayama, and F. Shinozaki,Bull. Chem. SOC. Japan, 1966,89, 882.l6 H. Otsuka, K. Inouye, F. Shinozaki, and M. Kanayama,Bull. Chem. SOC. Japan, 1966, 89, 11711977 1907blocked [as in (4)] the separation of urea from peptide determines the rate of urea formation. The rate of themay be possible,* whereas it must be difficult in more Curtius rearrangement (decomposition of azide intogeneral cases such as ( 5 ) and (6). For all these urea isocyanate and nitrogen) was determined by measuringderivatives, recovery of the amino-acid whose a-amino- the volume of nitrogen evo1ved.l' The rate of peptidegroup had been involved in the formation of the urea formation was determined by measuring the consumptionbond was generally low: Gly 20% [in compound (l)], of amine component by the ninhydrin method.l*30% ( 2 ) ; Lys 70% (4), 80% ( 5 ) ; Glu 70% (6).This is The results obtained from the Curtius rearrangementTABLE 1Some urea derivatives and the corresponding peptidesUrea : Z-Gly-NH*CH(CH,*C,H,*OH).NHCO-Gly-OBut (1)Peptide : 2-Gly-Tyr-Gly-OButtlrea : Z-Gly-NHCH(CH,Ph).NHCO-Gly-OBut (2): 2-Gly-Phe-Gly-OButUrea (sym) : [Z-Gly-NHCH(CH,.Ph)*NH],CO (3)Peptide : 2-Gly-Phe-NH,Urea : 2-Lys(Boc) -Pro-Val-Gly-NH*CH( [CH,],*NH-Boc) *NH.CO-Lys( Boc) -Lys( Boc) -Lys(Boc) -NH, (4)Urea : 2-Lys( Boc) -Pro-Val-Gly-NH-CH ([CH,] ,*NH-Boc) *NH*CO-Lys( Boc) -Arg-Arg-NH, (5)Peptide : 2-Lys(Boc)-Pro-Val-Gly-Lys(Boc) -Lys(Boc)-Arg-Arg-NH, 15Urea : Boc-Gly-Tyr-Ser-NH-CH(CH,CH,.SMe) .NHCO-Glu (OBut ) -His-Phe- Arg-Trp-Gly-OH (6) { Pep tide : Boc- Gly-Tyr-Ser-Me t-Glu ( OBut ) -His-Phe- Arg-Trp-Gly-OH l 6: 2-Lys(Boc) -Pro-Val-Gly-Lys(Boc)-Lys(Boc)-Lys(Boc)-Lys(Boc)-NH,probably because a relatively stable carbamoylamino-acid, NH,*CO*NH*CHR*CO,H, is formed as an inter-mediate upon hydrolysis of the urea.0 2 4 6 8t / hFIGURE 1 Curtius rearrangement of 2-Gly-Phe-N, in ethylacetate: Vt = volume of evolved nitrogen at time t ; VVOO =total volume collectedComparison of Peptide aizd Urea in Terms of Rate ofFormation and Kinetics.-Attempts were made to com-pare the rates of formation of urea and peptide undervarious conditions.The formation of urea was esti-mated by the extent of the Curtius rearrangement, which* The reaction of an azide-isocyanate mixture, prepared from2-Gly-Phe-N,, with H-Gly-OBut led to 2-Gly-Phe-Gly-OBut andthe urea (2) as the two major products.The peptide and theurea could bc separated by chromatography on a silica gel. column with chloroform-methanol as solvent. The combinedyield was ca. 60%. In a similar manner, pure samples of2-Gly-Tyr-Gly-OBut and the urea (1) were obtained from amixture in which these compounds were the two major com-ponents. For details see Experimental section.of the model compound 2-Gly-Phe-N, are shown inFigure 1 and Tables 2 and 3. Figure 1 demonstrates thatTABLE 2Kates of Curtius rearrangement of2-Gly-Phe-N, in ethyl acetate( "C) min-1 t)/min2 40010 7.5 93020 27 26030 100 69Temp.104kl/5 2.9TABLE 3Kates of Curtius rearrangement of2-Gly-Phe-N, in various solventsTemp.( "C)Tetrahydrof uran 20Ethyl acetate 20Dimethylf ormamide 20Solvent303030Benzene 30Chloroform 30103kl/min-11.97.22.73.510101117the decomposition of the azide follows first-order kinetics.The reaction should, therefore, be independent ofconcentration, but it depends on conditions such astemperature and solvent. The remarkable temperaturedependence of rate constant k, and half-life tg of thisfirst-order reaction can be seen in Table 2. Table 3shows values of k, measured in five different solvents,indicating that the nature of the solvent has much lesseffect on the reaction rate than the temperature.Ratesof decomposition were also measured for a variety ofacyl azides: the effects of changing the acyl groups weremuch less marked than that of temperature (Table 4).The aminolysis of 2-Gly-Phe-N, with an equimolar17 Y. Yukawa and Y. Tsuno, J . Amer. Chem. SOC., 1957, 79,l8 W. H. Stein and S. Moore, J . Biol. Chem., 1948, 176, 367.5530; 1959,81, 20071908 J.C.S. Perkin Iamount of H-Gly-OBut in ethyl acetate was studied as amodel system for peptide formation. The resultsobtained are shown in Figure 2 and Table 5. Figure 2TABLE 4Rates of Curtius rearrangement of acyl-peptideazides in ethyl acetateTemp. 1O3k1/Azide ("C) min-lZ-Gly-Phe-N, 20 2.730 10Z- G1 y- T yr-1. , 20 2.530 9.2Boc Boc20 2.4 Z-Lys-Pro-Val-Gly-Lys-N,30 11Boc-Ser-N, l6 30 6.1Boc-Ser-Tyr-N, l6 30 7.8I IBoc-Gly-Tyr- Ser-Met-N, 30 5.30 4 0 20 40 60t J minFIGURE 2 Aminolysis of Z-Gly-Phe-N, with H-Gly-OBut inethyl acetate a t 5 O C ; a = initial concentration; x = con-centration of H-Gly-OBut a t time t ; a = 10 m M for both reac-tantsTABLE 5Rates of aminolysis of 2-Gly-Phe-N, withH-Gly-OBut in ethyl acetateInitialconc. (mM)10Temp.k,/l mol-l(;)min-l16.7 533.3 5 2.866.7 516.7 5 2.816.7 10 3.416.7 20 4.116.7 30 5.9ti /min422210221815104.7is a plot of x/[a(a - x ) ] against time (where a and x areconcentrations of amine component at time 0 and time t ,respectively). The satisfactory linearity confirms thatthe reaction follows the second-order law. Table 5shows values of the rate constant k, and the half lifet ) of this second-order reaction under various conditions.These data indicate that the rate depends very much onthe initial concentrations and to a relatively small extenton the temperature.In this respect the peptide form-ation offers a striking contrast to the first-order decom-position of azide. Table 6 summarises the resultsTABLE 6Rates of aminolysis of 2-Gly-Phe-N, with variousamino-acid t-butyl esters in ethyl acetateConc. Temp. k,/l mol-'Ester ( m M ) ("C) min-' t)/minH-Gly-OBut 33.3 5 2.8 10H-Val-OBut 33.3 5 0.37 81H-Glu(OBut)-OBut 33.3 5 0.23 134H-Phe-OBut 33.3 5 0.15 200obtained in the aminolysis of 2-Gly-Phe-N, with variousamino-acid t-butyl esters under the same conditions.The reaction rate varies over a wide range depending onthe nature of the amine component; e.g.k, for H-Phe-OBut is ca. 20 times smaller than that for the glycine ester.The half-life values ti in Tables 2 and 5 indicate thatfor 16.7m~-solution and at 20 "C the aminolysis of Z-Gly-Phe-N, with H-Gly-OBut is only 17 times faster thanthe decomposition of the azide, whereas at 66.7m~ and5 "C the aminolysis proceeds 500 times faster than thedecomposition. The conditions in the latter case aresimilar to those of azide couplings currently used inpeptide synthesis, in which the initial concentration ofreactants is usually ca. 0 . 1 ~ and the temperature 0-5 "C.These conditions may generally be adequate for minimis-ing the Curtius rearrangement. However, aminolysisby some amino-acid esters is not so fast as that by H-Gly-OBut (Table 6).In addition, the rate may alsodepend on the nature of the azide; when the azide issterically hindered (e.g. valine azide lk~a) aminolysis maybe much slower. In such cases the side reaction is likelyto occur to a considerable extent. Since the Curtiusrearrangement is a unimolecular reaction and the ureaformation is thus inevitable, it should be borne in mindthat peptide prepared by the azide method may be con-taminated with urea if this side product is closelysimilar to the peptide in chemical and physicochemicalproperties.EXPERIMENTALT.1.c. was performed on precoated silica-gel plates(Merck Kieselgel 6OF,,,) with the following solvent systems(ratios by volume) : A, chloroform-methanol-acetic acid(95 : 5 : 3) ; B, chloroform-methanol-acetic acid (90 :10 : 3) ; C, chloroform-methanol-acetic acid (80 : 15 : 10) ;D, ethyl acetate-acetic acid-water (3 : 1 : 1) ; E, butan- l-ol-acetic acid-water (4 : 1 : 2).For detection the plate wasfirst sprayed with hydrobromic acid (sp. gr. 1.48), thenheated a t 150 "C for a few minutes (to dryness), and finallysprayed with ninhydrin (HBr-ninhydrin test). Kieselgel60 (Merck) was used for silica gel column chromatographythroughout. Samples for amino-acid analysis were hydro-lysed with constant-boiling hydrochloric acid in evacuatedsealed tubes a t 110 "C for 24 h.2-Gly-Tyr-NH*NH,.-2-Gly-OH (9.62 g, 46 mmol) andH-Tyr-OMe (free base; 8.97 g, 46 mmol) were coupled wit1977 1909DCC (9.48 g, 46 mmol) in DMF-ethyl acetate (1 : 5; 240 ml)to give 2-Gly-Tyr-OMe as an amorphous solid.This wasthen treated with hydrazine hydrate (5.75 ml, 115 mmol) inethanol (100 ml) a t 25 "C overnight to afford the crystallinehydrazide (16.1 g, 93y0), m.p. 202-204" (decomp.), [a]D23.6+10.9 f 0.5" (c 1 in M-HC1) (Found: C, 59.3; H, 5.7;N, 14.45. C,,H,,N,O, requires C, 59.05; H, 5.75; N,14.5%).2-Phe-NH-NH-Boc.-2-Phe-OH (2.99 g, 10 mmol) andt-butyl carbazate (1.32 g, 10 mmol) were coupled with DCC(2.06 g, 10 mmol) in ethyl acetate (10 ml) at 4 "C overnight.The crude product was recrystallised from ether-petroleumto give the desired compound (3.92 g, 95%), m.p.105-llO",[aID26 -27.0 f 0.7" (c 1.0 in MeOH) (Found: C, 63.9; H,6.6; N, 10.15. C,,H,,N,O, requires C, 63.9; H, 6.6; N,10.15%) ; homogeneous (HBr-ninhydrin) by t.1.c. insystem B.2-Gly-Phe-NH*NH,,HCl.-2-Phe-NH-NH-Boc (8.27 g,20 mmol) was hydrogenolysed over palladium in methanoland the product was coupled with 2-Gly-OH (4.18 g, 20mmol) by the DCC method in the usual manner to give2-Gly-Phe-NH*NH-Boc as an oil. This was then treatedwith M-hydrogen chloride in acetic acid (70 ml) a t 25 "C for60 min. The crystalline precipitate was filtered off, washedwith acetic acid and ether, and dried in vacuo (5.7 g).Concentration of the filtrate gave more of the desiredproduct (2.1 g). The two crops were combined and recry-stallised from ethanol; yield 6.46 g (79%); m.p.152-154"(decomp.); [aIDz4 +3.5 f 0.3" (c 2.0 in 0.5~-HCl) (Found:C, 55.9; H, 6.0; C1, 8.95; N, 13.7. C,,H2,N404,HClrequires C, 56.1; H, 5.7; C1, 8.7; N, 13.75%).Preparation of A zides (General Procedure) .-The azideswere prepared from the corresponding hydrazidesimmediately before use ; the temperature was maintaineda t 0 "C throughout and the reagents had previously beenchilled in an ice-bath.To a solution of the hydrazide (1 mmol) in 1 : 1 water-tetrahydrofuran (4 ml) were added M-hydrochloric acid(2.5 ml) and 2hl-sodium nitrite (0.55 ml), successively, andthe mixture was stirred for 4 min. After addition of 50%potassium carbonate (3 ml) the mixture was extracted withethyl acetate (7 ml x 2).The organic extracts were com-bined, dried (MgSO,), and evaporated in vacuo to leave theazide.Rate Determinations.-(a) Curtius rearrangement. Therate was determined by measuring the volume of nitrogenevolved. l7 The apparatus consisted of a round-bottomedflask and a gas burette with a three-way stop-cock and adrying tube in the line connecting the flask and the burette.The reaction mixture was not stirred, but a piece of sinteredglass was added to prevent supersaturation by the nitrogenevolved.The flask containing the azide, prepared from the cor-responding hydrazide (1 mmol) as described above, wasimmersed in the bath a t the appropriate temperature(k0.1 "C). To the azide was added the solvent (15 ml),previously equilibrated a t the same temperature, and themixture was swirled to effect dissolution.The stop-cockwas then opened and the nitrogen evolved was collected.The volume of nitrogen was measured a t intervals appro-priate to the rate of the reaction. The reaction wasfollowed to 80-90~0 completion. The bath was thenremoved and the flask was kept a t room temperature over-night, then immersed again in the bath. The volumereading at this stage was taken as Vm. In most cases, theobserved values of Vm were 88-99y0 of the theoreticalones. The rate constant k , was calculated from equation(i), where V, is the volume of nitrogen a t time t . The half-life ta is expressed as (In 2)/k,.k,t = 1nCVmWm - Ql (1)(b) Peptide formation. The rate was determined bymeasuring the consumption of an amine component by theninhydrin method.The flask containing 2-Gly-Phe-N,, prepared from Z-Gly-Phe-NH-NH,,HCl (0.407 g, 1 mmol) as described above,was immersed in the bath a t a given temperature ( fO.1 "C).A solution of an amino-acid t-butyl ester (amine component;1 mmol) in ethyl acetate, which had been equilibrated a tthe same temperature, was then added.The concentrationof the reactants was controlled by adjusting the volume ofethyl acetate used. At intervals samples were withdrawnfrom the mixture and immediately mixed with 0.5 ml of aninhydrin solution 18 which had been kept in an ice-bath.The mixture was then heated on a boiling water bath for15 min and cooled; after appropriate dilution with 50%ethanol, the absorbance a t 570 nm was measured.Asample with no azide was also submitted to the ninhydrinreaction and the absorbance measured was taken as thevalue a t time 0. The second-order rate constant k , wascalculated from equation (ii), where a and x are the con-centrations of amine component a t times 0 and t, respect-ively; half life l/(k,a).k,t = x/[a(a - x ) ] (ii)Formation of Peptide and Urea.-(a) Z-Gly-Tyr-Gly-OBut and Z-Gly-NH*(CH)CH,*C6H4*OH) *NH*CO-Gly-OBut(1). A solution of 2-Gly-Tyr-N,, derived from the cor-responding hydrazide (0.77 g, 2 mmol), in ethyl acetate (32ml) was divided into four equal portions. The first portionwas, after addition of H-Gly-OBut (0.10 g, 0.75 mmol),kept a t 4 "C for 2 days (A). The second was evaporatedin vacuo at a bath temperature of 0 "C and the residue wasdissolved in DMF (10 ml).To this was added H-Gly-OBut (0.10 g) and the mixture was kept a t 4 "C for 2 days(B). The third was kept at 4 "C for 21 h, after which H-Gly-OBut (0.10 g) was introduced. The mixture was thenkept a t 4 "C overnight (C). The last portion was treatedin the same manner as the third, but the temperature wasmaintained at 25 "C throughout (D). T.1.c. in system Bshowed little or no urea formation in reaction mixtures (A)and (B). In (C), both peptide and urea were major pro-ducts, but no peptide was detected in (D).Reaction mixtures (C) and (D) were combined andchromatographed on a silica gel column (20 g) with 5%methanol in chloroform as solvent.The fractions contain-ing peptide (I) and those containing urea (11) as main com-ponent were combined. Fraction (11) was evaporated andthe residue was crystallised from methanol-ether to givethe urea (0.19 g), m.p. 90-91", [aIDz4 -6.7 + 0.5" ( c 1.OinMeOH) [Found: C, 59.8; H, 6.2; N, 11.05. Calc. forcompound (1) (C,,H3,N407): C, 60.0; H, 6.45; N, 11.2%];amino-acid analysis (theoretical values in parentheses) :Gly 1.31 (2), NH, 1.17 (2), Tyr not detected (0).Fraction (I) combined with reaction mixtures (A) and(B) was chromatographed on a silica gel column (20 g) with2% methanol in chloroform as solvent. The fractionscontaining a major product as a single component werecollected and evaporated in vucuo to give an amorphou1910 J.C.S.Perkin Isolid (0.51 g ) ; [a&,,* -7.0 f 0.5" (c 1.0 in MeOH) [Found:C, 61.45; H, 6.55; N, 8.65. Calc. for Z-Gly-Tyr-Gly-OBut(C,,H,,N,O,): C, 61.85; H, 6.45; N, 8.65y0].(b) Z-Gly-Phe-Gly-OBut and Z-Gly-NH*CH(CH,Ph) *NH*CO-Gly-OBut (2). A solution of the azide, prepared fromZ-Gly-Phe-NH*NH,,HCl (0.81 g, 2 mmol), in ethyl acetate(32 ml) was divided into four equal portions. To the firstwas added H-Gly-OBut (0.10 g, 0.75 mmol) and the mixturewas kept at 4 "C for 2 days (A). The remaining portionswere evaporated in vacuo a t a bath temperature of 0 "C andthe residues were dissolved in chloroform (8 ml each). Toone of these was immediately added H-Gly-OBut (0.10 g)and the mixture was kept a t 4 "C for 2 days (B). Anotherwas kept a t 4 "C for 21 h, after which H-Gly-OBut (0.10 g)was added.The mixture was then kept a t 4 "C overnight(C). The last one was kept a t 25 "C for 23 h before additionof H-Gly-OBut (0.10 g) and the mixture was kept a t 25 "Covernight (D). The mixtures (A)-(D) were examined byt.1.c. (in system B; HBr-ninhydrin). (A) containedpeptide and no detectable urea; (B) peptide and a smalleramount of urea; (C) urea and a smaller amount of peptide;(D) urea and no detectable peptide. These mixtures werethen combined and evaporated in vacuo. The residue waschromatographed on a column of silica gel (50 g) with 3%methanol in chloroform as solvent. Fractions (7 ml) werecollected and examined by t.1.c. in system B. The fractionscorresponding to two major components were collectedseparately and evaporated in vacuo; tubes 25-38 and39-60 afforded 0.33 and 0.40 g, respectively, of syrupyresidue.The material derived from tubes 39-60 wascrystallised from methanol-ether and recrystallised fromethyl acetate-ether; yield 0.30 g; m.p. 165-167"; [aID23*5-7.6 f 0.5" (c 1.0 in MeOH) [Found: C, 61.9; H, 6.9;N, 11.5. Calc. for compound (2) (C,,H,,N,O,): C, 61.95;H, 6.65; N, 11.55%]; amino-acid analysis: Gly 1.20 (2),NH, 0.85 (2), Phe not detected (0).The material derived from tubes 25-38 was treated withtrifluoroacetic acid (TFA) at 25 "C for 60 min; the mixturewas then evaporated in vacz~o and the residue was crystal-lised from ether and recrystallised from ethyl acetate-ether ;yield 0.20 g; m.p. 156-158'; -15.0 f 0.5" (c 1.0in EtOH) [Found: C, 60.9; H, 5.75; N, 10.2.Calc. forZ-Gly-Phe-Gly-OH (C,,H,,N,O,) : C, 61.0; H, 5.6; N,10.15~0]. This product was identical with authentic Z-Gly-Phe-Gly-OH. *(c) [Z-Gly-NH.CH(CH,Ph)-NH],CO (3). An ethyl ace-tate extract of Z-Gly-Phe-N,, prepared from the corres-ponding hydrazide (0.81 g, 2 mmol) was kept, withoutdrying, a t 25 "C for 3 days. The crystalline precipitate(0.44 g ) was recrystallised from ethanol; yield 0.17 g ;m.p. 185-190" (decomp.); [ct]D23.5 -3.1 f 0.5" (c 1.0 inDMF) (Found: C, 65.0; H, 5.95; N, 12.05. C3,H,oN,0,requires C, 65.3; H, 5.9; N, 12.35%). This product wasclearly distinguished from Z-Gly-Phe-NH, t on t.1.c. insystem B.Preparation of Ureas.-(a) Z-Lys( Roc) -Pro-Val-Gly-NH*CH ([CH,] ,*NH-Boc) .NH*CO-Lys( BOC) -LYs( BOC) -LYs( Boc)-* Z-Gly-Phe-Gly-OH was derived from the correspondingtripeptide t-butyl ester, which was synthesised stepwise from theC-terminal by the DCC procedure; m.p.159-160", [aIDz4-15.3 0.7" (c 0.9 in EtOH) {lit. m.p. 162.5-163", [aIDZ1- 14.6 -f 0.3" (c 1.3 in EtOH) (D. W. Clayton, J. A. Farrington,G. W. Kenner, and J. M. Turner, J . Chem. Soc., 1957, 1398)).Z-Gly-Phe-NH, was synthesised from Z-Gly-OH and H-Phe-NH, by the DCC method; m.p. 145-147"; [ a ] ~ ~ ~ 0.0 & 1.0"(c 0.5 in DMF), +6.0 f 1.0" (c 0.5 in MeOH).NH, (4). Z-Lys(Boc)-Pro-Val-Gly-Lys(Boc)-NHONH,(0.24 g, 0.27 mmol) was converted into the correspondingazide, which was dissolved in DMF-chloroform ( 1 : 6;35 ml) and kept a t 25 "C for 4 h, during which period theazide i.r.band at 4.75 pm had disappeared. To this wasthen added a solution of H-Lys(Boc)-Lys(Boc)-Lys(Boc)-NH, (free base; 0.16 g, 0.23 mmol): in DMF (1 ml) and themixture was, after concentration in vucuo to remove chloro-form, allowed to stand a t 25" C for 20 h. A gelatinousprecipitate which had separated was filtered off (0.25 g)and crystallised from methanol (0.19 g, 54%) ; m.p. 223-225" (decomp.), [aiJD2* -30.5 f 0.8" (c 1.0 in MeOH) (Found:C, 57.8; H, 8.25; N, 13.25. C,,H1,,N1,O,, requires C,57.7; H, 8.35; N, 13.45%); amino-acid analysis: Lys3.73 (4), NH, 2.96 (3), Pro 0.97 (l), Gly 1.00 (l), Val 1.00 (1).The urea thus obtained could be distinguished from thecorresponding peptide (Table I ) $ by t.1.c.in system A.Z-Lys( Boc)-Pro-Val-Gly-NH*CH( [ CH,],*NH-Boc)NHCO-Lys(Boc)-Arg-Arg-NH, (5). A solution of Z-Lys( Boc) -Pro-Val-Gly-Lys( Boc) -N, prepared from the cor-responding hydrazide (0.44 g, 0.5 mmol) n~ in DMF-chloroform (1 : 6; 7 ml) was kept a t 25 "C for 7 h. To thiswere then added H-Lys(Boc)-Arg-Arg-NH, acetate (0.32 g,0.42 mmol) Is and triethylamine (0.20 ml) with DMF (2 ml)as solvent. The mixture was kept at 25 "C overnight andevaporated in vucuo. To the residue were added ethylacetate (30 ml) and M-acetic acid (30 ml), and the mixturewas shaken vigorously. The organic phase was extractedwith M-acetic acid (20 ml x 3). The aqueous solutionswere combined, washed with ethyl acetate (20 ml x 3),concentrated in vucuo to ca.20 ml, and extracted with water-saturated butan-1-01 (20 ml x 3). The extracts werecombined, washed with M-acetic acid (20 ml x 3), andevaporated in vacuo, and the residue was precipitated frommethanol-ether to give the desired urea in pure form (0.40 g,66%); m.p. 148-150"; [a]D23-5 -31.1 & 0.8" (c 1.0 inMeOH) (Found: C, 53.0; H, 8.2; N, 17.8. C,5Hl13Nl,01,,-3H20 requires C, 53.1; H, 8.15; N, 18.1%); amino-acidanalysis: Lys 1.81 (2), NH, 2.21 (3), Arg 1.89 (2), Pro 1.03(l), Gly 0.98 (l), Val 1.00 (1). The product was homo-geneous (HBr-ninhydrin and Sakaguchi reagent) on t.1.c.in system E, but could not be distinguished from the cor-responding peptide (Table 1) l5 in the same solvent system.(c) Boc-Gly-Tyr-Ser-NHCH (CH,*CH,*SMe) *NH*CO-Gly-OBut .Boc-Gly-Tyr-Ser-Met-NH*NH, (0.58 g, 1mmol) 16 was converted into the corresponding azide. Theresulting solution in ethyl acetate was kept a t 30 "C for7.5 h (for kinetic measurements, see Table a), then set asidea t room temperature overnight. A small amount ofprecipitate was filtered off. To the filtrate was added H-Gly-OBut (0.13 g, 1 mmol) and the mixture was kept a troom temperature overnight. The resulting precipitate(0.42 g) was chromatographed on a silica gel column (20 g)(b)$ This compound was obtained as an intermediate in the syn-thesis of Z-Lys(Boc)-Pro-Val-Gly-[Lys(Boc)],-NH, (Table 1).Z-Lys(Boc)-NH,, derived from Z-Lys(Boc)-OSu l 5 by treatmentwith ammonia, was coupled with Z-Lys(Boc)-OSu repeatedly incombination with the removal of the 2 group by catalytic hydro-genolysis to give H-[Lys(Boc)],-NH,.This was then combinedwith Z-Lys(Boc)-Pro-Val-Gly-OH lS by the DCC method to affordthe octapeptide derivative; m.p. 221-223", [ a ] ~ ~ , -- 36.6 & 0.8"(c 1.0 in MeOH) (K. Inouye, Y. Sumitomo, and M. Shin, un-published data) {lit. m.p. 219-221", [ a ] D 2 0 -38" (c 1 in MeOH)(B. Riniker and W. Rittel, Helv. Chim. Acta, 1970, 53, 513)).l8 H. Otsuka, K. Inouye, andY. Jono, Bull. Chem. SOC. Jafian,1964,87. 14711977 1911with methanol-chloroform (2 : 98, 100 ml; 10 : 90, 200ml) as solvent. The fractions containing the main productas a single component were collected and evaporated invacuo and the residue was solidified by treatment with ethylacetate (0.25 g, 36.5%) ; m.p. 156-158" (decomp.),-2.1 f 0.5" ( G 1.0 in MeOH) (Found: C, 52.6; H, 7.2;N, 11.9; S, 4.75. C,oH48N,0,,S requires: C, 52.6; H,7.05; N, 12.25; S, 4.7%). The product was homogeneouson t.1.c. in system C.(d) Boc-Gly-Tyr-Ser-NHCH(CH,*CH,*SMe)*NH*CO-Glu(0But)-His-Phe- Arg-Trp-Gly-OH (6). The azide pre-pared from Boc-Gly-Tyr-Ser-Met-NH.NH, (0.29 g, 0.5mmol) was allowed to undergo rearrangement as describedabove. To the resulting isocyanate were added H-Glu-(OBut)-His-Phe-Arg-Trp-Gly-OH (0.20 g, 0.2 mmol) 2O andtriethylamine (0.07 ml, 0.5 mmol) with DMF (5 ml) assolvent, and the mixture was kept a t room temperature.After several hours it was evaporated in vucwo and thegelatinous residue was solidified by trituration with ether(0.40 g). A portion (0.20 g) of this crude product waschromatographed on a column (3.4 x 150 cm) of SephadexG-25 (medium) with butan-l-ol-acetic acid-water (4 : 1 : 2)as solvent. Fractions (10 ml) were collected and theirabsorption at 280 nm was measured, Tubes 41-50 cor-responding to a main peak were combined and evaporatedin V ~ C W O and the residue was precipitated from methanol-ether to yield the desired urea (0.16 g ) ; amino-acid analysis :Ser 0.99 (I), Glu 0.70 (I), Gly 2.00 (2), Met 0.02 (0), Tyr 1.01(l), Phe 0.99 (l), His 1.00 (l), Arg 0.98 (1). The productwas homogeneous (HBr-ninhydrin and Ehrlich reagent) ont.1.c. in systems D and E, but was inseparable from thecorresponding peptide (Table 1) l6 in the same solventsystems. Amino-acid analysis revealed a small amountof methionine, indicating that the product was contaminatedwith some peptide.[7/221 Received, 8th February, 19771$0 K. Inouye, Bull. Chem. SOC. Japan, 1965, 38, 1148

ISSN:1472-7781    

DOI:10.1039/P19770001905

出版商:RSC    

年代:1977

数据来源: RSC