NOTES
Oct., 1943
2039
ethylidene bond of isoquinine takes place more material. The specific rotation of the iodohydroquinine slowly than similar addition to the vinyl group of base in alcohol was -150’. A n d . Calcd. for ctOH&TzOIIGHc: I, 24.0. Found: quinine. I, 24.4. Although the reduction of isoquinine has been On treating this sample of iodohydroquinine with alcostudied,’ the use of an unsymmetrical molecule holic potassium hydroxide, isoquinine and a small amount as an addendum has not been described. From of niquine were obtained. the material available we were not able to identify CONTRIBUTION FROM THE D I ~ A R T M OF~ T IN PURECHEXISTRY epimeric C-3 derivatives as Henry and co-workers’ RESEARCH MELLON INSTITUTE OF INDUSTRIAL R~EARCH did with dihydroquinine. Subsequent removal PITTSBURGH, PENNA. RECEEIVED JUNE29, 1943 -. of halogen from iodohydroquinine (prepared The Role of Glycine in Protein Structure from isoquinine) regenerated isoquinine and gave BY HANSNEURATH some niquine. The separation of the a and a’ isomers2 of 10Recent discussions of protein structure have iodohydroquinine by recrystallization from ben- emphasized the importance of the amino acid side zene had been carried out in this Laboratory chains in determining the structure’of fibrous and prior to the publication of Reyman and Suszko’; globular The position of the side the specific rotations observed (-218’ and 19’) chains along a polypeptide chain is determinant were practically identical with the values reported for the types of lateral bonds and the extent of by the Polish investigators. [a] - 19’ is the internal rotation of polypeptide chains. A skelevalue for a’-iodohydroquinine, having one mole of tal polypeptide chain, stripped of side chains, is benzene of crystallization. When benzene was endowed with a considerable degree of internal removed by evaporating an ether solution to dry- rotation about single valence bonds of the conness, the product melted a t 130’ and [a]D was stituent carbon and nitrogen atoms. However, -22.3’. The -218’ fraction was most advan- in a genuine polypeptide chain, equipped with tageously crystallized from acetone or alcohol for the full complement of side chains, internal rotafinal purification. tion is greatly restricted due to the space requireThe strongly levorotatory a-isomer loses hydro- ments of the amino acid residues. Various congen iodide to give predominantly niquine, and the figurations, previously proposed for folded polya’-isomer gives predominantly i s ~ q u i n i n e . ~ In~ ~ peptide chains and chain net-works, had, accordour experience it was not possible to get the ex- ingly, to be excluded, since they resulted in an clusive transformation Suszko reported. During excessive crowding of component atoms and storage for a number of months, the more levo of group^.^*^ Such steric limitations apply equally the isomeric iodohydroquinines appears to be the to fully extended polypeptide chains, if they less stable, and this fact has been noted previously be made up of a combination of alternating d- and for the quinidine series in the case of bromodi- 1-amino acids, but not to polypeptides comprised hydroquinidine.b of less than four amino acid residues. In the Iodohydroquinine from 1soquinine.-Ten-gram por- latter, terminal side chains are free to branch out tions of isoquinine, heated on a water-bath for two hours and, by virtue of rotation about the alp carbon with 60 g. of hydriodic acid6 (d. 1.7), gave only a 35y0 bond, to orient themselves as demanded by their yield of the yellow crystalline iodohydroquininedihydrio- space requirements. dide. Continued heating for thirty minutes increased the The space requirements of amino acids depend yield to 5Oy0. Unchanged isoquiniine, recovered from the on their average chain length and their crossfinal liquor, accounted for 25-3070 of the original isquinine. This slow reaction rate is in contrast to the 70% sectional area beyond the a-carbon atom.‘ yield obtained in two hours when quinine was the starting The amino acid with the smallest cross-sectional area is glycine, since a t the place where other (1) Henry, Solomon and Gibbr, J . Chcm. Soc., 692 (1937). (2) In the absence of a uniform scheme of nomenclature, we have amino acids carry a side chain, glycine has merely followed the method of Goodson, who arbitrarily used the prefix a a hydrogen atom of about 4 sq A. cross-sectional for the isomer of higher rotation and a’ for the other isomer: Good-
-
son, i b i d . , 1094 (1935). (3) Reyman and Surzko, Bull. I n f c m . Acad. Polonaisc, A, 360 (1935). (4) Podlewski and Suszko, Rcc. frar. chim., I S , 882 (1936). ( 5 ) Gibbs and Henry, J. Chcm. SOC.,240 (1939). (6) Skraup, Monafsh., 14, 428 (1893): Rosmmund and Kittler. Arch. Pharm., 262, 18 (1924).
(1) Neurath, 1.Phys. Chcm., 44, 296 (1940). (2) Bull, “Advances in Enzymology,” Vol. I , Interscience Publishers, New York, N . Y., p. 1 el scq. (3) Astbury and Bell, Nature, 147, 696 (1941). (4) Chibnall, Proc. Roy. Soc. (London), B181, 136 (1942). (5) Mack, Ohro 1.Sci., 41, 183 (1941): Huggins, Ann. Rev. Biochcm.. 11, 32 (1942).
2040
NOTES
Vol. 65
area (as compared to about 11.5 sq. A. for the the fiber axis although a repeating pattern a t methyl group of alanine, and about 35 sq. fi. about 16 X 3.5 A. would have to be expected if for the benzene nucleus of phenylalanine). I t the residues in this protein, of which about 44% occurred to us that the presence of glycine residues are glycine and 25% alanine, were distributed in may confer upon a polypeptide chain a consider- a regular fashion.8 This could be explained by able degree of flexibility and internal rotation, ascribing to glycine a position, determined by a since, wherever a glycine residue occurs, a hydro- specific stereochemical function, in preference to gen atom is taking the place of a more complex the position required by a periodical arrangement residue R. Hence, free rotation about -C(HR)of amino acid residues. N(H)- bonds should be possible to nearly the While glycine thus facddates folding of polypepsame extent as if a polypeptide chain were devoid tide chains, more reactive amino acid residues of a side chain. Accordingly, the presence of probably determine the mode of folding, by mutual glycine residues, and their distribution along the interaction. chain, may be factors influential in determining the The present hypothesis rests on the assumption specific pattern of folding of polypeptide chains. that glycine is a natural constituent of all proI t is of interest to examine certain experimental teins. While certain proteins have been reported and theoretical aspects of this hypothesis in rela- to be devoid of g l y ~ i n eit, ~has to be recalled that tion to the problem of protein structure: most of these analyses were carried out with 1. The proper distribution of glycine residues methods which would fail to reveal a glycine conwould permit any combination of side chains tent of several per cent.l0J1 Therefore, this arguwhich, otherwise, would have to be excluded be- ment remains open pending further analytical data. cause of steric hindrance. For instance, while I t may be of significance, however, that the large i t would be extremely difficult to depict a situa- space requirements of proline and hydroxyproline tion where any two aromatic or heterocyclic in the gelatin molecule (about 32%) are counteramino acids, separated by any third amino acid, balanced by a comparable content of glycine 5,could follow each other along a folded poly- (about 25%). Similarly, elastin contains 29% of peptide chain without causing steric interference, glycine as compared to 15% of proline and 2% such a combination could occur if X is glycine. of hydroxyproline. l 2 i n such an instance, free rotation would allow The present hypothesis does not necessarily them to assume juxtapositions with respect to the conflict with the Bergmann-Niemann hypothesis13 plane of the main chain from which they protrude. of the periodicity of occurrence of amino acids in I t would also be conceivable that natural and polypeptide chains. Yet, a very high degree of unnatural amino acids might be incorporated selectivity in the synthesis of proteins would have together in a polypeptide chain if glycine residues to be evoked in order to place amino acids in the are interdispersed to provide internal rotation.6 positions called for by a combined demand of 2 . The concept of internal rotation due to periodicity and stereochemical space requirements. glycine might account for the selective orientaThe unique stereochemical conditions caused tion of side chains in monomolecular protein by the presence of glycine in a peptide chain films-polar side chains toward the water, non- have already been recognized by Bergmann in polar side chains toward the air phase.7 studying the antipodal specificity of proteolytic It also obviates the necessity for concluding that enzymes.14 Optically active amino acids, when certain proteins such as keratin and myosin con- present in the d-form, inhibit through steric sist of an even number of polar and non-polar side hindrance the action of enzymes that are adapted chains, arranged alternately along the main chains.8 to substrates containing 1-amino acids. Row3, I t was noted that silk fibroin is devoid of a (9) Cf.Patton, J . B i d . Chcm., 108, 267 (1935). diffraction period Ionger than 2 X 3.5 A. along (lo) Personal communication of Dr. Max Bergmsnn. (6) Pauling (Taur JOURNAL, 61, 2643 (1940)), cognizant of the large space requirements of proline and hydroxyproline, suggested that these amino acid residuea occupy terminal chain positions in normal and innuno-serum globulins. As an alternative hypothtsis one could assume that they occupy positions adjacent to glycine residues. (7)Neulrth and Bull, Chem. Reo., OS, 891 (1938). (8) Astbury, “Advances in Enzymology,” Vo1 111, Interscience I’i~blishers.N e w York, N Y , p 63 ff.
(11) Egg albumin, for instance. has been reported to bc devoid of glycine whereas the solubility method of Bergmann and Stein revealed a glycine content of about 3.1% (Stein, Proc. Piftk Ann Meeting A m . SOC.Brewlng Chcm., May 25-27 (1942)),equal to 18-19 residues per protein molecule (46,000 molecular weight). (12) Stein and Miller, J . B i d . Chcm., 126,599 (1938). (13) Bergmann and Niemann, ibid., 116, 77 (1936). (14) Bergmann. Science, ?S, 479 (1934); Bergmann and Fruton. J B i d Chcm , 117, 189 (1937).
Oct., 1943
2041
NEWBOOKS
ever, glycine-containing substrates are split under these conditions since glycine contains no side chain in such a spatial arrangement that i t could prevent an approach of the enzyme. It is a pleasure to acknowledge many valuable suggestions which Dr. Max Bergmann has offered in a discussion of this problem.
action of 4-phenylphenol and butyryl chlcu-ihl in pyridine solution with 1,4-dioxane as diluent.1 The crude product was dissolved in benzene, and the resulting solution was washed with dilute hydrochloric acid and sodium hydroxide solution and decolorized with Norite. After removal of the benzene on the steam-bath,. the product was crystallized four times from 30-60" ligroin; colorless platelets resulted; m. p. 6940.3".
DEPARTME,NT OF BIOCHEMISTRY SCHOOL OF MEDICINE DUKE UNIVERSITY JULY 23, 1943 NORTHCAROLINA RECEIVED DURHAM,
A d . Calcd. for CI~XMOS:C, 80.0; H, 6.67. Found: C, 79.83; H, 6.85. (1) Gilman and Blatt, ''Organic Syntheses," John Wiley and Sons, Inc., New York, N. Y.,Coll. Vol. I. 2d ad., 1941, p. 147. (2) Harlct, Hensley and Jass, THISJOURNAL, 64, 2449 (1942).
NEW COMPOUND 4-Phenylphenyl Butyrate This compound was prepared in 81% yield by the inter-
DEPARTMENT OF CHEMISTRY E. HAZLBT STATE COLLEGE OF WASHINGTON STEWART PULLMAN, WASHINGTON LEE C. HENSLEY RECEIVED JULY 20, 1943
COMMUNICATION T O T H E EDITOR THERMAL PROPERTIES OF ISOPENTANE
Sir: Anyone not completely familiar with the field, on reading the paper by Guthrie and Huffman on page 1143 of the June, 1943, issue of THISJOURNAL might receive a wrong impression. We have carefully reviewed our work published on isopentane in THISJOURNAL and find no reason to doubt the data there reported. The facts are that two independent groups of workers using different calorimeters have observed an anomaly in the thermal behavior of isopentane. A third group of workers using a different calorimeter
have failed to observe this phenomenon. The second independent series of measurements was made in our laboratory but in a different calorimeter (Gold calorimeter C). The work was done by M. L. Sagenkahn and H. F. Zuhr. The third independent series of measurements is that of Guthrie and Huffman. After the present emergency we shall repeat again the work on isopentane using the Huffman type calorimeter. SCHOOL OF CHEMLSTRY AND PHYSICS THEPENNSYLVANIA STATE COLLEGE PA. STATECOLLEGE, RECEIVED AUGUST 6, 1943
J. G. ASTON
NEW BOOKS Volume 23. LEE IRVINSM~TH, members of the publishing Board have taken part in the Editor-in-Chief, HOMER ADKINS,C. F. H. ALLEN,W. development and construction of the experimental techE. BACHMANN, NATHANL. DRAKE,C. S. HAMILTON, nique described. The techniques proposed have been R. L. SI~RINER, H. R.'SNYDERAND A. H. BLATT,Secre- checked in each experiment by two independent workers. tary to the Board. John Wiley and Sons, Inc., 440 A short literature review accompanies each preparation Fourth Avenue, New York, N. Y., 1943. 124 pp. and in many cases useful notes are inserted which serve 15 X 23.6 cm. Price, $1.75. to guide the experimenter in applying the experimental procedure recommended. All the procedures are clearly In this last volume of this important series of Organic Synthues are recorded specific directions for preparing written, and the book should find a most useful service thirty-nine different organic compounds These embrace in every laboratory where organic synthesis is being'aprepresentatives of the aliphatic, aromatic and hetero- plied and practiced. TREAT B. JOHNSON cyclic series; and fifty-four contributors other than Organic Syntheses.
