COMMUNICATIONS TO THE EDITOR
July 5, 1960
3483
ton transfer between imidazole and chlor phenol meric compounds derived from y-methyl-L-glutamic red (&), appears to be slightly smaller than the acid. The optical activities were determined in value expected for a diffusion controlled p r o ~ e s s . ~both random coil and helix-forming solvents. Weller* also found similar rates for reactions with However, association complicated interpretation 5 x 10' &I--1 sec.-l) which suggests of the results.','" Molecular weight determinations acridine (k that this rate either is characteristic of proton using short column equilibrium ultracentrifugation transfers involving a nitrogen atom or depends confirmed association.'l In this communication somewhat on the pK-values of the proton donor- we wish to report optical activity enhancements acceptor system. For such transfer processes, and abnormal rotatory dispersions for the oligodiffusion controlled rates ( k - 3 X l o 9 'If-' sec.-l) meric peptides in solvents where no association have been founds only when the proton was trans- occurs. When the oligomeric peptides are studied ferred between two oxygen atoms which were more in dimethylformamide and m-cresol, concentration acidic (pK--5) than the systems considered above. independent optical activity enhancements are The value of k,z should be compared to previous re- noted (Table 11). The lack of association was sults obtained for similar reactions between a cation confirmed by molecular weight determination in Table I contains a list and hydroxyl ion; these are listed in Table I. All dimethylf~rmamide.~~ these rate constants are characteristic of diffusion of the optical rotations and rotatory dispersion bo controlled processEs; the differences in rates are due values from the Moffitt equationl2 for polymers of to steric factors. Thus, ammonia is the fastest be- poly- y-methyl-L-glutamates in ditnethylformainide cause it has four sites for proton transfer. Tri- and dichloroacetic acid. methylamine ion, on the other hand, is sterically TABLE I hindered by the methyl groups and has a corDATAO N POLY-7-METHYL-L-GLUTAMATES respondingly slower rate of reaction. The fact OPTICALROTATION
-
Solvent
TABLE I Cation "a
+
Acridinium (CHa)a?iIIf
kir( M - 1 ser.-i)
3 . 0 X 1Olo 1 . 8 5 x 10'0 1 . 0 x 10'0
Reference
1 8 9
that the imidazole rate constant is twice as large as that for trimethylamine indicates that the reaction is able to take place a t both nitrogen atoms in the imidazoliurn ion ring. This occurrence is due either to resonance stabilization which makes the two nitrogens equivalent, or to a fast tautomerization of the ring after reaction a t the neutral nitrogen. Acridine would be expected to be less sterically hindered than trimethylamine, but should react more slowly than imidazole, which has two reactive nitrogen atoms. This fact is in accord with the intermediate value of the rate constant that was found. This investigation was carried out during the tenure of Postdoctoral Fellowships from the National Cancer Institute, United States Public Health Service (K.K.) ; and National Science Foundation (G.G.H.). ( 8 ) A . Weller, Z . Rieklrorhrin., 6 4 , 5: (19iiOi. (9) h l . Rigeri, 3 . S. Jhnson,E. K u \ t i n nnd A . E i t t i g , in prepn.
h l ~ PLASCK x INSTITL~T
;\rASFRED B I G E N (;ORUOS [.; IIAMMES
FL'ER P*IVSIKAI,ISCHE C H E R l I l i
G ~ E T T I N G EGERMASY S, RticErvsn
J ~ - N E2 ,
KENNETI-I KUSTIN 1060
CONFORMATIONAL ASPECTS O F SYNTHETIC POLYPEPTIDES. 11. CRITICAL RANGE FOR INTRAMOLECULAR HYDROGEN BONDING
Sir: In our previous report,' we employed optical activity to measure the conf~rmation*-~ of oligo(1) M. Goodmanand E. E. Schmitt, THIS J O U R N A L , 81.6507 (1959). (2) J. T. Y a n g a n d P. Doty, i b i d . , 79, 761 (1957). (3) P. D o t y , A. M . Holtzer, J. H. Bradbury. E. K.Blout, ibid., 76, 4493 (1954), (4) P. Doty and J. T. Yang, ibid., 78, 498 (1933). ( 5 ) P . D o t y and R. D. Lundberg, Proc. N u l l . Acad. Sci., 43, 213
(1957).
A/I
[ a 1"D
Dirnetliylformamide High +lo Dimethyliormamide 93 +lo Dimeth ylformamide 22 8 Dichloroacetic acid 104 -33 Corrected for refractive index of solvent; a 212 m p was used as XO.
+
boa
-544 -532 -4486 56 value of
+
Rotatory dispersion as a criterion for helix content has been established and examined by Xloffitt,'2 Doty,2 Cohen,13 Tinoco,14 Blout15 and Elliott.IG Moffitt expanded the Drude equation to include a higher order term.'? He suggested that a plot of [ a ](X2-Xr,2) us. ( X 2 - X 2 ~ ) - ' with other than a zero slope exhibits abnormal rotatory dispersion, indicating helical structures. Doty2 suggested the use of A," as a criterion of helicity for some synthetic and naturally occurring polypeptides. We determined the optical rotatory dispersion of the oligomeric compounds in helix and random coil solvents. In dichloroacetic acid Moffitt plots indicated material essentially of a random coil nature. In dimethylformamide, however, increasingly negative abnormal dispersions were observed in going from the trimer to the undecamer (Fig. 1) (Table 11). ( 6 ) A . R . I h m n i e , 4 . Elliott, W. I?. Hnnby a n d R . R . n l : c l c . ~ ~ l m , P r o i R o y . SOL.( r n v d o i i ) , A242, :12h (1957). ( 7 ) I. Tinoco, THISJOTJRNAL, 81, 1 5 4 1 (1959). ( 8 ) C . Cohen, N n t i w r , 1'75, 129 (1955). (9) E. K. Blout, P. Duty and J. T. Yang, T H E J C J U R N A I . . 7 9 , 719 (195i). (10) P. Doty, J , H. Bradbury nnd A . h l . IIoltzer, ibici., 78, !417 (195Ij). ( 1 1) 17. Yphantic, A n n . .\'. 1'. A r o d . Sei., to be piibliahed (12) (a) W hloffitt, J. C h e f s . Phys., 2 5 . 4R7 (195fj); (b) W. \loOift and J. T . Yang, Proc. T o l l . A c a d . S c i . , 4 2 , 5 O l i (1956); (c) W .hloffitt.
;bid., 42, 736 (1956); (d) W. Moffitt. D. D. F i t t s and J. G. Kirkwood, ibid.. 43, 723 (1957). (13) C. Cohen and A . G. Szent-Gyorgi, THIS J O U R N A L , 79, 218 (1957). (14) I. Tinoco and R . E'. Woody, J. Chenz. P h y s . , 3 2 , 461 (19fiOi. (15) E. R . Blout in C. Djerassi. "Optical Rotatory Dispersion." Chap. 17, hIcGraw-Hill Book Company, Inc., New Ymk, N. Xr.,19GO. (16) E. hl. Biadbury, L. Brown, A. R. Downie, A . Elliott, W. E. H a n h y and T. R. R. McDonald, Nature, 183, 1736 (19.59). (17) Derived from the square root of the slope obtained from a plot Of
[Q]
VS. xy[P].
COMMUNICATIONS TO THE EDITOR
3484
1
-'
-6
TRIPEPTICE
t
TETRAPEPTIDE
.st
HEXAPEPTIDE
HEPTAPEPTIDE
-6
I
I
b
I
/
6
1
e
I/(
Fig. 1.-The
,
l 10
,
I 12
K- A$)
I/(
P-A,:
Moffitt plots for peptides derived from y-methyl glutamate in dimethylformamide a t 25.0"
TABLE I1 SPECIFIC ROTATIOSSAXD ABSORMAL ROTATURY OF OLIGOMERIC PEPTIDES COEFFICIENTS -13.9 -50.0 +52 183mp -15.7 i49 1181rnp J -15.5 -55,O +4 212mp 6 -15.4 -118 219mp , -15.6 -50.0 -53 230rnpd 9 -10.5 -37.5 -191 286mp' 11 -4.5 -14.0 -251 320mpd Tcikeii i n tliiriet1i~lfornna:riitic. T:ikei1 i n ill-cresol. Corrected for index of refraction; :i value of 212 m p WLS used as A,, Slope was lint linear a t higher wave lengths. .\pprosirnate values Tvrre c~btxiiie(1 frcJtil t l l x t portion c i f the w r y ? xliich 11';~slineair.
3 4
-
'l
I'
'CVr feel that the abnornial optical rotatory dispersion and the optical rotation enhancement are excellent indications of stable helical forms. The critical size for helix formation in solution depends on the nature of the peptide, the solvent and the temperature.Ig For peptides derived from y-methyl-L-glutamate in dimethylformamide a t ?so,a critical range actually exists for the forniation of helical structures. (
VOl. 82
(18) C. G. Schellman a n d J. A . Schellman, C o i i z N . o v i s h e r p , . S P Y . ;him , 3 0 , 4H6 (19.58).
rpitd.
If optical activity enhancement is used as a criterion for helicity, the critical range appears to be between the hepta and nonapeptides. When the X, and bo values are used as the criteria, there seems to be a small but sudden change in these values going from the tetrapeptide t o the peiitapeptide while a much greater change occurs in going from the hepta- to the nonapeptide. These changes result from the folding of the peptide chain. In the region between the penta and nonapeptides intramolecular hydrogen bonds can be formed using either the carbonyl or amido groups but not both from the same amino acid residue. The large effect which cotiinierices at the nonapeptide may be attributed t o more stable helical forms. At this chain length residues appear which can intrainolecularly hydrogen bond both through the carbonyl and amido groups siinultaneously. Acknowledgment.-IVe gratefully acknowledge the support for this research given by the National Science Foundation grant G-8314. POLYTECHNIC IXSTITUTE OF BROOKLTS DEPARTMENT OF CHEXISTRT hILTRR.4Y GOODMAN EDWARD E. SCHMITT RROOKLYK,N. \-. ROCKEFELLER ISSTITUTE DAVIDY PHANTIS SEI?, \-(IRK, S.1..
t v n a . lob.
KFCRIVFP
S I A Y 4 . 1F)fiO
