SYNTHESIS OF HINDERED QUINOLETHERS
April 20, 1960
2055
ylglycolic acid is present in fairly large amounts in the reaction mixture from the incubation of 3,5diiodo-4-hydroxyphenylaceticacid. Furthermore, in order to explain the formation of hydracrylic acid from 3-(3,5-diiodo-4-hydroxyphenyl)-hydracrylic acid one would have to assume that the aliH + c I* o p R phatic side chain is eliminated as a carbanion -CH(OH)CH2COOH, a reaction mechanism that is difficult to conceive. The much more plausible mechanism of Johnson and Tewkesbury would lead not to hydracrylic acid but to the semi-aldeHOfiO-&R hyde of malonic acid. Decarboxylation of this 1 1)" IY unstable compound followed by oxidation would LROCH~CH ~acetic CO O Hamount of acetic yield a~id.~ aThe small R = CHKHzCOOH acid found in fraction C-2 may have been formed in adjacent to the aromatic ring to -CH(OH)-. They this manner since hydracrylic acid does not give based this suggestion on a model reaction in which rise to even traces of acetic acid under the conditetraiodothyroacetic acid is formed when 3,5- tions of the incubation experiment. Only a small diiodo-4-hydroxyphenylacetic acid is incubated. fraction, if any, of the incubated diiodophloretic The reaction mixture contained glyoxylic acid, the acid follows this pathway. Reduction of an aliquot formation of which is easily explained by an oxida- of fraction C-2 by diphosphopyridine nucleotide in tion of the starting material to 3,5-diiodo-4-hy- the presence of alcohol dehydrogenase from horse droxyphenylglycolic acid, followed by a hydroxyla- liver showed that only a minute amount (about tion of the aliphatic side chain according to John- 0.01 mmole in the total fraction) of a carbonyl son and Tewkesbury's mechanism. However, the compound reducible by this system, presumably choice of the acetic acid analog of diiodotyrosine acetaldehyde, was present. It can be concluded from the present model exas a model for the study of the fate of the "lost side chain" in the synthesis of thyroxine is not a fortu- periment that in the synthesis of thyroxine from nate one. In contrast to diiodotyrosine and to its diiodotyrosine the alanine side chain which is elimpropionic acid analog which we chose as a model, the inated is probably converted to a hydroxylated acetic acid analog contains an active methylene compound (either serine or hydroxypyruvic acid), group that is very susceptible to oxidation. While not to alanine or dehydroalanine (pyruvic acid and 3- (3 5-diiodo-4-hydroxyphenyl) -hydracrylic acid ammonia). It is not likely that the elimination is could not be detected with certainty in the reaction preceeded by a hydroxylation of the side chain. mixture from the incubation of diiodophloretic butanol-pyridine-water.1 I t gives a hrownish-yellow spot on paper acid,37the homologous 3,5-diiodo-4-hydroxyphen-chromatograms when sprayed with reagent 2, which indicates that an
'+
(37) A small amount of an unknown substance was found which may be this acid. It had Rf 0.05 in 1-butanol-2 iV ammonia and 0.11 in 1-butanol-dioxane-ammonia.1 It does not seem to be 3,5-diiodo-4hydroxybenzoic acid since it has a somewhat different Ry-value in 1 -
oxygen function is probably present on the carbon atom adjacent t o the aromatic ring.1 (38) Cf. H. D. Dakin, J. B i d . Chem., 6 , 409 (1909).
BETIIESDA14, MD. ~~~
[CONTKIBUTION FKOM TIIB
~
~~~
NATIONAL INSTITVTE OP ARTHRITISAND META5OLIC DISEASES, NATIONAL INSTITUTES OF HEALTH]
Model Reactions for the Biosynthesis of Thyroxine. 111. The Synthesis of Hindered Quinol Ethers and their Conversion to Hindered Analogs of Thyroxine'i2 B Y TERUO hfATSUURA3AND H. J. CAIINMANN RECEIVED AUGUST27, 1959 Model reactions :ire prescuted for the biosynthetic scheme (equations 1-3) proposed by Johnson and Tewkesbury for the formation of thyroxine from diiodotyrosine. Sterically hindered analogs of the quinol ether 111 have been synthesized in a sequence of free radical reactions and then converted to the corresponding analogs of thyroxine. Some of the properties of the quinol ethers and of the analogs of thyroxine are described.
The mechanism of the reaction in which thyroxine is formed from diiodotyrosine in vivo or in z&04 is still obscure. Various hypotheses concerning this mechanism have been proposed.6 The present report deals with the mechanism that was first supgested by Johnson and Tewkesbury6and elaboratgd (1) Paper 11, H. J. Cahnmann and T. Matsuura, THIS JOURNAL, 82, 2050 (1960). (2) A preliminary report of this work has been presented at the 134th Meeting of the American Chemical Society, September, 1958, Chicago, 111. (3) Visiting Scientist from Osaka City University, Japan. (4) P. van Mutzenbecher, Z . phrsiol. Chem., 261, 253 (1939). (6) For references cf. paper I of this series.1a
by Harington.' This mechanism is based on the extensive studies of Pummerer and his co-workers8a-k who have found that many oxidations of (6) T.B. Johnson and L. B. Tewkesbury, Jr., PYOC. Natl. A c a d . S c i . W 7 3 (1942). (7) C. K. Harington, J . Chem. Soc., 193 (1944). (8) (a) R. Pummerer, H. Puttfarcken and P. Schopflocher, Bcr., 68, 1808 (1925); see also (b) R. Pummerer and F. Frankfurter, ibid., 47, 1472 (1914); (c) 62, 1416 (1919); (d) R. Pummerer and E. Cherbuliez, ibid., 47, 2957 (1914); (e) 62, 1392 (1919); (f) R. Pummerer, ibid., 62, 1403 (1919); (9) R. Pummerer, D. Melamid and H. Puttfarcken. ibid., 66, 3116 (1922); (h) R. Pummerer and A. Rieche, ibid., 69, 2161 (1926); (i) R. Pummerer and F. Luther, ibid., 61, 1102 (1928); 6 ) R. Pummerer, G. Schmidutz and H. Seifert, Chem. Bcr., 86, 635 (1952); (k) R. Pummerer and I. Veit, i b i d . , 86, 412 (1953).
u. S.,
2056
TERUO MATSUURA AND H. J. CAHNMANN
Vol. 82
phenols take place via a free radical mechanism. That the first step of most, if not all, oxidations of phenols is the formation of a free radical was borne out by the classical work of Michaelis on the mechanism of the oxidation of hydroquinones to quiand by many subsequent investigations. According to the hypothesis of Johnson and Tewkesbury, the first step in the synthesis of thyroxine from diiodotyrosine (I) is an oxidation of the latter to a free radical 11.
version to diphenyl cthcrs, a reaction that is malogous to the last step in Johnson and Tewkesbury’s reaction scheme, has not been investigated. In the present paper we report the synthesis of several analogs of the quinol ether I11 in a sequence of free radical reactions closely resembling the first two steps of Johnson and Tewkesbury’s mechanism. We further report the conversion of some of the quinol ethers thus synthesized to the corresponding analogs of thyroxine. From what is known about the relationship between the stability of free radicals and their chemical the free radical 11, if formed at all in the course of the synthesis of thyroxine, must be expected to be unstable. In contrast, the analogous free radical I1 (R1and R1 = C(CH3)B) is known to be quite stable in the absence of oxyI g e 1 1 ~ ~due ~ to , ~the ~ resonance ~ ~ ~ , ~stabilization and steric hindrance provided by the three t-butyl / groups. Therefore, the corresponding phenol I? ; 2,4,6 - tri - t- butylphenol (I, R1 and RZ= C(CH3)3) has been selected as a starting material for the ----v-----synthesis of several analogs of the quinol ether 111. I1 The first step in this synthesis is the removal of RI I; R2 = CHzCH(NHz)COOH an electron from this phenol leading to the correThe second step is a dimerization of two molecules sponding free radical (reaction 1; R1 and RZ = (Jf the free radical to form a quinol ether 111. C(CH3)3). This reaction was carried out with potassium ferricyanide as the electron-removing agent.*a-ls14s1i Aisolution of an analog of tyrosine was then addcd to the blue solution of the stable free radical. The analogs of tyrosine used for this purpose were of the general type V characterized by a free or esterified aliphatic acid side chain in thc I11 $-position to the phenolic hydroxyl and by o-posi121 = I ; Ra = CH*CH(NH*)COOH tions that are either free or substituted with broFilially, the loss of the side chain 122 attached to mine or iodine. 911 electron transfer takes place the dienone ring leads to thyroxine (IV). resulting in the oxidation of the added analog of tyrosine to a free radical VI and the concomitant reduction of the blue free radical to tri-t-butylphe-
-03JR2
1101.
111 = I; Rz = CH2CII(NH2)COOH
>llthough this hypothetical mechanism appears plausible, i t has not previously been substantiated cxperimentally. No attempts to demonstrate the forniation of a quinol ether intermediate in the synthesis of thyroxine or its analogs5 have been reported. Furthermore, no aromatic quinol ethers of the type I11 carrying an aliphatic acid side chain in p-position to the cther bridge are known. Only a few other aromatic $-cluinol ethers have been syiithesizecl.8j~’1~1~ The possibility of their con(!l; (a) I,. 1Rlicliaclis, Coid S ~ Y ~Harbor. ~ Z ESyviposia Oiaanl. B i d . , 7 , 3 2 (1039); (11) A ~ z n .iY. Y. A c a d . Sci., 40, 39 (1910). ( 1 0 ) For leading references concerning t h e mechanism of biological
~ixidationsof phenols see D. H. R . Barton and T.Cohen, “Festschrift A . Stull, Birkhauser, Basel, 1967, p. 117, and H. Erdtman atid C. A . Wachtmeister, ibid., p. 144. Extensive studies on t h e mechanism of the i,z ritvo oxidation of phenols have been made by Pummerer and co-wixkers8a-k and by Goldschmidt and co-workers [first contribution: St. Goldschmidt, Bey., 55, 3194 (1922); last contribution: St. Goldschmidt, E . Schulz and H. Bernard, A m . , 418, 1 (193Oj1. (11) E. Muller, K. Ley and G. Schlcchte, Chem. Be?.. 90, 2660 ( 1 !4:57), (12) ’The transient fortiiation o f a n aruinatic p-cjuinol ether in the coursc of the oxidation of p-cresol to “Punimerer’s kctone” reported
H or Br or I ; COOClHs or ( CH?)zCOOC2H6or (CH2)zCOOH
Iil aiid Ka = C(C!Hs)~; R J =
Rq
= CHn-
In the course of the addition of the analog o f tyrosiiie the bluc solution beconies colorless or almost by I-l3,5dibromo-4-hydroxyphenylacetate in 50 ml. of benzene was added within a few minutes. After 15 minutes standing, the solution, which had become light blue, was subjected in the usual manner to a second and then to a third oxidation, each of which was followed by the addition of ethyl 3,sdibromo-4-hydroxyphenylacetate. After each addition of bromoester, the solution was permitted to stand for 15 minutes. After the second oxidation, 1.70 g. ( 5 mmoles) of bromoester was added, after the third oxidation 1.20 g. (3.6 mmoles). The total amount of ester used was 6.28 g. (18.6 mmoles) . Ice-cold methanol (25 ml.) was added to the crystalline residue obtained after evaporation of the light blue solution. The mixture was kept overnight at -25", then filtered. Yellow crystals, m.p. 88-90' (8.50 g., 85% based on the bromoester), were obtained which, after recrystallization from n-pentane, gave yellow prisms melting at 89.5-90.5'; yield 6.70 g. (67%). The fusion was greenish-blue a t the melting point; evolution of isobutene began a t about 145'. Anal. Calcd. for C28H38Br204: C, 56.20; H , 6.40; Br, 26.71. Found: C, 55.95; H, 6.25; Br, 26.88. The infrared spectrum (KBr) shows bands a t 5.74 (carbon>-1 of COOCzH6), 5.99 and 6.07 (carbonyl and double bonds of the quinol ring), 8.03 and 10.09 p (aromatic quinol ether) .30 Ethyl 3-(3,5-Dibromo-4-hydroxyphenyl)-propionate (Ethyl Ester of 3,s-Dibromophloretic Acid) (V, R3 = Br, Ra = (CHZ)zCOOCSH,).-A solution of 3.09 mi. (60 milliatoms) of bromine in 20 ml. of chloroform was added slowly with stirring and cooling (ice-bath) to a solution of 5.82 g. (30 mmoles) of the ethyl ester of phloretic acid31 in 50 ml. of chloroform. The solution was then allowed t o stand overnight a t room temperature. The residue obtained after evaporation of the solvent was distilled in vucuo. A colorless viscous liquid, b.p. 163-164' (1 mm.), was obtained which became crystalline on standing; yield 8.22 g. (78'70). This material was sufficiently pure for the preparation of the corresponding quinol ether. On recrystallization from ether-n-pentane i t gave colorless plates, 1n.p. 41-48'. Anal. Calcd. for CllHl2Br2O3: C, 37.53; H , 3.44; Br, 45.40. Found: C , 37.23; H , 3.58; Br, 45.38. 4- [2,6-Dibromo-4-( 2-Carbethoxyethy1)-phenoxy]-2,4,6tri-t-butyl-2,5-cyclohexadien-l-one(VII, RB = Br, RI = (CHz)zCOOC?H:).-Thjs quinol ether was prepared in the same manner as the quinol ether VI1 (R3 = B r , Ra = CHZCOOCzH6) except that the ethyl ester of 3,5-dibromophloretic acid was used instead of ethyl 3,5-dibrotno-4-hydroxyphenylacetate. X total amourit of 6.28 g. (17.8 mmoles) of brornoester was used; 3.52 g. (10 mmoles) in 30 nil. of benzene after the first oxidation, 1 . X g. ( 5 mmoles) in 20 ml. of benzene after the second oxidation, and 1.00 g. (2.8 mmoles) in 20 ml . of benzene after t h e third oxidation. The mixture of the crude quinol ether and methanol was permitted to stand a t -25" for several hours, then filtered. Yellow crystals, m.p. 1OO-10lo, were obtained; yield 10.0 g. (9270 based on the bromoester). Recrystallization from n-pentane gave yellow prisms, m.p. 103-104O; yield 8.12 g. (74%). Anal. Calcd. for CzeH40Brz04: C , 56.87; H , 6.58; Br, 26.10. Found: C,57.46; H,6.61; Br,25.83. The infrared spectrum ( K B r ) shows bands at 5.76 (carbonyl of COOC2Hb), 5.97 and 6.06 (carbonyl and double
2060
TERUO MATSUURA AND H.J. CAHNMANN
VOl. 82
bonds of the quinol ring), 8.01 and 10.07 p (:iromatic quinol Ethyl 3-(3,5-Diiodo-4-hydroxyphenyl)-propionate (Ethyl ethe~-).~O Ester of 3,s-Diiodophloretic Acid) (V, R3 = I, R4 = (CH2)23-(3,5-Dibromo-4-hydroxyphenyl)-propionic Acid (3,s- COOH).-A suspension of 16.0 g. (38 mmoles) of diiodoDibromophloretic Acid) (V, R, = Br, R, = (CH2)2COOH).phloretic acid35 in 75 ml. (1.3 moles) of absolute ethanol A solution of 3.09 ml. (120milliatoms) of bromine in 40 ml. of was saturated with dry HC1 in an ice-bath. .lfter standing a 2070 (w./v.) aqueous solution of KBr was added dropwise for one hour a t room temperature, the reaction mixture was with cooling (ice-bath) to a stirred solution of 4.98 g. (30 uoured into ice-water. Recrystallization of the precipitate mmoles) of phloretic acid' in 150 ml. of water. The precipifrom aqueous ethanol gave I S . Z ~gd~ (su(,~) of colorless tate formed was filtered and dried, m.p. 102-103'; yield 8.16 prisms, m . p . 86-87", lit.3j m.p. 86-87 . g. (847,). Recrystallization from chloroform-benzene 4- [Z,6-Diiodo-4-(2-carbethoxyethyl)-phenoxy] -2,4,6-tri tgave 7.33 g. (757,) of colorless prisms, m.p. 106-108', lit. butyl-2,5-cyclohexadien-~-one (VII, Ra = I, R4 = (CH,),1n.p. 107-108°33 and 106-108".34 COOCJ&).--l free radical solution was prepared froin 4- [Z,6-Dibromo-4-( 2-carboxyethy1)-phenoxy]-2,4,6-tri-t- 5.25 g. (20 mmoles) of 2,4,6-tri-t-butylphenol as described for butyl-2,5-cyclohexadien-l-one(VII, R3 = Br, R, = (CH,),- the synthesis of the quinol ether VI1 (R3 = H , It, = (CHz)yCOOH).--A free radical solution was prepared from 10.5 g. COOCZHB). .l solution of 4.6 g. (10 mmoles) of the ethyl (40 mmoles) of 2,4,6-tri-t-butylphenol as described for the ester of 3,5-diiodophloretic acid in 50 ml. of ethyl acetate synthesis of the quinol ether VII(R8 = H, R1 = (CH,)?- was then added within a few minutes. After standing for 10 COOC2H6). -4 solution of 6.48 g. (20 mmoles) of 3,5-d1- minutes, the slightly brownish-blue solution was evaporated bromophloretic acid in 50 ml. of ether was then added within in ' i v ~ z au ~t low temperature (< 0"). The light browti color a few minutes. After standing 15minutes the light blue solu- of the distillate indicated t h a t some iodine had been libertion was evaporated in vacuo. The crystalline residue was ated. The viscous residue was well mixed with 25 ml. of methanol in the cold (Dry Ice-acetone-bath). The mixture well mixed with a small amount of ice-cold n-pentane. .lfter filtration, 9.69 g. (83%) of yellow crystals, m.p. 115' dec., was then permitted to stand a t -25' for 30 minutes. Th: were obtained. Recrystallization from 90% methanol gave yellow crystals formed were collected bl- filtration; n1.p. '37 yellow prisms which melted a t 119' with decomposition dec.; yield 6.39 g. (90yc). Brief digestion with n-pentaiie (evolution of isobutene); yield 8.00 g . (6857,). in a Dry Ice-acetone-bath, followed bl- filtration, gave 5.9(j Anal. Calcd. for C2?H36Br2O4:C, 55.49; H, 6.21; Br, g: (84 %) of yellow crystals, m.p. 97' dec. The fusion begins to bubble a t about 125". 27.35. Found: C, 55.56; H , 6.42; Br, 27.22. The infrared spectrum (Fluorolube S from 2-7.4 p , Sujol Anal. Calcd. for C2sHao1204~C, 49.30; H , 5.71; I , from 7.4-15 p ) showed bands a t 5.83 (carbonyl of COOH), 35.93. Found: C, 50.60; H , 5 . ( ,; I , 34.39. 5.97 and 6.06 (carbonyl and double bonds of the quinol Digestion of the crystals with methanol, followed by ring), 8.00 and 10.11 p (aromatic quinol ether).30 another digestion with n-pentane, both with cooliug (Dry Ethyl 3,5-Diiodo-4-hydroxyphenylacetate.--X suspetision of 14.70 g. (36.4 mmoles) of 3,5-diiodo-4-hydroxyphenylace- Ice-acetone-bath), did not raise the melting point. The infrared spectrum (KBr) shows bands a t 5.77 (cartic acid1@in 125 i d . (2.2 moles) of absolute ethanol was saturated in a n ice-bath with dry HCl. First a clear solution was bonyl of COOC2Hj), 6.00 and 6.08 @ (carbonyl and double obtained, then a precipitate of needles formed. .lfter stand- bonds of the quinol ring), 8.06 and 10.10 p (aromatic quiilol ing at room temperature for one hour, the mixture was di- ether).30 4- [Z ,6-Diiodo-4-( 2-carboxyethy1)-phenoxy]-2,4,6-tri-t-butylluted with water. The crystals were collected by filtration, then recrystallized from aqueous ethanol. Colorless fine 2,5-cyclohexadien-l-one (VII, R3 = I, Ra = (CH?)..COOH). needles, n1.p. 121-122", were obtained; yield 14.76 g. -A free radical solution was prepared from 5.25 g. (20 inmoles) of 2,4,6-tri-t-butylphenol as described for the SY11(94%). thesis of thequinol ether (1711, R3 = H , K4 = (CH?)&OOCzAnal. Calcd. for CloHlJ203: C , 27.80; H, 2.33; I , He). A solution of 4.20 g. (10 mmoles) of 3,5-d11od?phlo58.75. Found: C, 27.50; H, 2.65; I, 59.17. retic acid35in 60 ml. of ethyl acetate was then sdded wlthm :i 4- [Z,6-Diiodo-4-( carbethoxymethy1)-phenoxy] -2,4,6-tri- few minutes. After standing for 10 minutes, the light blue t-butyl-2,5-cyclohexadien-l-one(VII, RJ = I, Ra = CH2- solution was evaporated in aacuo a t low temperature (0;). COOCzH6).--4 free radical solution was prepared from 5.25 The light brown color of the distillate showed t h a t some 10g. (20 mmoles) of 2,4,6-tri-t-butylphenol as described for the dine had been liberated. The crystalline residue was well synthesis of the quinol ether VI1 (Ra = H , Rq = (CH2)z- mixed with methanol (Dry Ice-acetolle-bath). After standCOOCZHS). A suspension of 4.32 g. (10 mmoles) of ethyl ing overnight a t -25' the yellow crystals were collected by 3,5-diiodo-4-hydroxyphenylacetate in 50 ml. of benzene filtration, then washed with a small amount of cold inethwas then added within a few minutes. After 15 minutes mol; m.p. 111' dec.; yield 2.60 g. (38y0 based on diiodostanding the greenish-blue solution was evaporated in vaczio phloretic acid). Digestion with 15 ml. of n-pentalle in the a t low temperature (about 5'). The sticky residue was cold (Dry Ice-acetone-bath), followed by filtration, raised rapidly mixed in a n ice-bath with 30 nil. of n-pentane. In- the melting point to 114' dec.; yield 1.80 g. (%yo).Evolusoluble white needles of recovered iodoester were removed tion of gas occurs already a t the melting point. by filtration (1.10 g,). The filtrate was evaporated in Anal. Calcd. for C2iH361201:C, 47.80; H , 5.35; 1, VQCUQ a t low temperature (< 0'). The crystalline residue 37.42. Found: C, 47.84; H, 5.35; I , 37.15. was rapidly digested in the cold (ice-bath) with 30 ml. of methanol to remove tri-t-butylpheuol. Insoluble yellow The infrared spectrum ( K B r ) shows bands a t 5.86 (Carcrystals of the quinol ether were collected by filtration; 11i.p. bonyl of COOH), 6.00 and 6.09 (carbonyl and double b [ ) ~ d s 89-90" dec.; yield 4.31 g. (62 YObased on the amount of iodo- of the quinol ring), 8.06 and 10.10 p (aromatic quinol ether).30 ester and 83% if the recovered iodoester is taken into ac- The infrared spectrum iu Fluorolube S (2-7.4 p)-Xuj[)l count). (7.3-15 p ) show the same bands. The filtrate, upon standing a t -25', deposited yellow Sodium 3',5'-Di-t-butylthyroacetate36 (VIII, R3 = H, Rq = crystals, m.p. 144' dec., which did not depress the m.p. CH2COONa).-The quinol ether VI1 (I& = H, Ra =, CHzE.,3 inmoles) \vas heated at 160-190 until (146') of bis-(4-oxo-1,3,5-tri-t-butyl-2,5-cyclohexadien-l-COOCzH,) . .. (1.76 . yl) peroxide.14b bubbling ceased (%bout 30 minutes). The crystals of the quinol ether were rapidly digested in The reddish-brown reaction niixture weighed I .55 g . the cold (ice-bath) with 10 nil. of n-pentane. The residual (weight loss, 12%; calcd. fur the elitniiiatiotl of one l-butyl crystals, isolated by filtration, melted at 89-90' dec., yield group, 13%). Short path distillation itz L'UCUO (about 0.1 inin.) 3.20 g. (467,; 62% if the recovered iodoester is taken into gave two fractions. Fractiou 1 (bntli temperature < 1 8 O 0 ) , account). The fusion begins to bubble at about 150". 0.35 g. of pale yellow crystals; fraction 2 (bath temperature Anal. Calcd. for C~8H3.&04: C , 48.57; H , 5.53; I, 180-210°), 0.92 g. (say0)of a viscous orange liquid. Recrystallization of fraction 1 from methanol gave color36.66. Found: C, 48.70; H, 5.46; I , 36.79. plates of tri-t-butylpllen[,l, m.p. and mixed 11i.p. 130The infrared spectrum (KBr) shows bands at 5.75 (car- less 131". bonyl of COOC2HS), 5.97 and 6.06 (carbonyl and double bonds of the quinol ring), 8.04 and 10.10 p (aromatic quinol (35) J . H. Barnes, IC. T. Borrows, J , Elks, B. A . Hems and A. G. ether).30 Long, J. Chein. Soc., 282-4 (1B50). (33) C . Stohr, Ann., 236, 57 (1884). (34) G. Habild, Z. physiol. C h c n . , %86, 127 (1Y50).
( 3 6 ) Thyroacetic acid, p-(p-hydroxyphenoxy)~l,henylacetic acid, thyropropionic acid, 3-[p-(D-hydroxyphenuxy)-phenyl]-propionic acid.
April 20, 1960
SYNTHESIS O F
HINDERED QUINOLETHERS
2061
Part of fraction 2 (0.66 g.) wits refluxed for one hour with a the reaction mixture was extracted several times with ligroin mixture of 2 ml. of 2 N NaOH and 10 ml. of ethanol. T h e (b.p. 66-75"). Evaporation of the ligroin extract gave 1.15 ethanol was evaporated and 20 ml. of water was added t o the g. of crude 2,4,6-tri-t-butylphenol. The aqueous layer was residue. acidified with 1 N HC1, then extracted with ether. The The slightly colored crystals formed were collected by ether extract was evaporated and the residue dissolved in 50 filtration; yield 0.55 g. (5lY0 based on the quinol ether). ml. of 1-butanol. The butanol layer was washed 20 ml. of Recrystallization from aqueous ethanol gave colorless plates 2 N KaOH and then with 20 ml. of water.19,3' Evaporation which contained water of crystallization. This was elimi- in zracuo of the butanol layer, followed by acidification of the nated by drying a t 80" i n vacuo. On heating, the dried crys- residue with dil. HC1, gave 0.75 g. of a sticky brown material tals turned brown a t about 230°, without melting. which was recrystallized from benzene; yield 0.28 g. (11%). Recrystallization from benzene gave 0.23 g. (97,) of colorAnal. Calcd. for C22H270INa: C, 69.82; H , 7.19. less needles, m.p. 169-171'. Found: C , 69.57; H, 7.31. Anal. Calcd. for C23H28Br204:C, 52.29; H , 5.34; Br, The infrared spectrum (KBr) shows bands at 2.84 (phenolic OH); 6.29-6.34 (carbonyl of COO-); 7.71,8.01,8.22, 8.35. 30.26. Found: C, 52.19; H , 5.37; Br, 30.08. The infrared spectrum (Fluorolube S for 2-7.4 w ; Kujol 10.36 fi (several of these bands are caused by the diphenyl for 7.4-15 p ) shows bands a t 2.82 (phenolic OH); 5.82 ether linkageiE)). dcidification of the sodium salt gave the free acid in the (carbonyl of COOH); 7.72, 7.95, 8.24, 8.38,10.41 p (several of these bands are caused by the diphenyl ether linkagei8). form of a viscous oil that could uot be crystallized. The Rf-values in paper chromatograms are 0.85 (solvent The Rr-values in paper chromatograms are 0.83 (solvent l ) , 0.82 (solvent 2 ) and 0.78 (solvent 3). l ) , 0.75 (solvent 2) and 0.67 (solvent 3 ) . ilcidification of the combined aqueous layers obtained in 3,5-Di-t-butylthyropropionicAcid36 (VIII, R3 = H, R4 = (CH,)&OOH).-A. By Pyrolysis.-The quinol ether (VI1 the butanol extraction with dil. HC1 gave 0.91 g. of a redRJ = H , Ra = (CHz)2COOH) (2.13 g., 5 mmoles) was heated dish-brown solid. Recrystallization from benzene gave 0.40 g. of colorless prisms, m.p. 105-108O, which were shown by a t 165' for 10 minutes, then a t 190" until bubbling ceased (about 20 minutes). The brownish-yellow reaction mixture mixed m.p. and by paper ~hromatography~g to be 3,j-dibroweighed 1.88 g. (weight loss, 127,; calcd. for the elimination mophloretic acid. B. By Acid Catalysis.-A solution of 1.O g. (1.7 mmoles) of one t-butyl group 137c). I t was dissolved in 50 ml. of 0.1 of the quinol ether VI1 (R3 = Br, Ra = (CH2)2COOH)and *V NaOH and this solution was extracted with n-pentane. Evaporation of the pentane extract gave brownish crystals of 0.03 g. of @-naphthalenesulfonic acid in 17 ml. of ethyl acetate was refluxed for 20 minutes. The brownish-green (0.30 g.) which, upon recrystallization from methanol, yielded colorless plates of 2,4,6-tri-t-butylphenol, m.p. and solution was evaporated in vacuo t o dryness, and the residue mixed m.p. 130-131'. was taken up in benzene. Insoluble @-naphthalenesulfonic The aqueous layer was acidified with 5 ml. of 1 N HC1. acid was removed by filtration and the filtrate was evapoThe light brown precipitate formed (1.30 g., 707,) gave, rated in vacuo. T h e residue was dissolved in 15 ml. of l-buupon two recrystallizations from benzene-isooctane, 0.65 g. tanol. This solution was treated with 2 N NaOH and worked (35%) of colorless needles, m.p. 187-189". up as described in procedure A . Needles (0.06 g., 77,) were Anal. Calcd. fur CS3Hd0O4:C, 74.56; H , 8.16. Found: obtained which, upon recrystallization, melted a t 168-170". A mixture with 3,5-dibromo-3',5'-di-t-butylthyropropionic C, 74.80; H , 8.08. acid prepared by pyrolysis (procedure A ) had the same The infrared spectrum ( K B r ) showed bands at 2.84 (phen- melting point. olic O H ) ; 5.87 (carbonyl of COOH); 7.72, 7.97, 8.19, Acidification of the combined aqueous layers from t h e bu8.38, 10.37 p (several of these bands are caused by the di- tanol extraction again yielded 3,5-dibromophloretic acid (see phenyl ether linkagelu). procedure A). The Rf-values in paper chromatograms are 0.80 (solvent Decomposition of the Quinol Ether VI1 (R3 = I, Ra = l ) , 0.74 (solvent 2 ) and 0.67 (solvent 3 ) . (CH,),COOH). A. By Pyrolysis.-The quinol ether VI1 (Ra Paper chromatography of the mother liquors from the re= I , Ra = (CH212COOH)(0.3441 E,.0.51 mmole) was heated crystallization of the di-t-butylthyropropionic acid followed at 125--13jb for-gminutesi weight-loss 3.9 mg. (1.1%; calcd. by spraying with diazotized ~i,~i-diethylsulfanilaniideigfor the elimination of one t-butyl group, 8.2y0). The reacrevealed the presence of phloretic acid. tion mixture was then heated at 155-160' until bubbling B. By Acid Catalysis.--.% solution of 0.21 g. (0.5 mmole) ceased (2 minutes). The total weight loss was i . 8 mg. (2.3 of the quinol ether VI1 (R3 = H , R4 = (CH2)?COOH)and of 7,of the starting material or 27yG of the theory). The 0.01 g. of @-naphthalenesulfonicacid in 5 ml. of ethyl acetate dark brown reaction mixture was dissolved in 20 nil. of 1was refluxed for 20 minutes. The originally yellow solution butanol. This solution was washed with 15 ml. of 2 -V Eabecame first green, then brownish-yellow. The residue ob- OH and with 10 ml. of water, then evaporated in F U C U O to tained after the removal of the solvent in aucuo was dis- dryness. T h e residue was acidified with aqueous HCl and solved in benzene. The solution was filtered to remove in- the acid mixture again evaporated. Extraction of the resisoluble @-naphthalenesulfonic acid and the filtrate was evap- due with benzene, followed by evaporation of the extract, orated in DQCUO. T h e residue was recrystallized from ben- gave 0.20 g. of a sticky dark brown mass. Attempts t o obzene-isooctane. The less soluble fraction (0.02 g.), upon tain crystals failed. T h e paper chromatogram (solvent 21, recrystallization from benzene-ethyl acetate, gave colorless after spraying with FeC13-K3 [Fe(CN)s]followed by washing plates, m.p. 118-122', which were identified by paper chro- with dil. HC1 and water,*O showed no distinct spots between matography'g as impure phloretic acid. The more soluble Rt 0.8 and 0.9, where 3,5-diiodo-3',5'-di-t-butylthyroprofraction (0.09 g., 48%), m.p. 178-183', upon recrystalliza- pionic acid must be expected. (There was a spot with a n Rition from benzene-ethyl acetate and then from aqueous value of 0.97 which is probably too high for the expected ethanol, gave colorless needles of 3,5-di-t-butylthyroproanalog of thyroxine.) pionic acid, m.p. 187-189". X mixture of this acid and the The combined aqueous layers from the butanol extraction one prepared by pyrolysis (procedure A ) , also melts a t 187were acidified with dil. HCl, then extracted with ether. 189'. Evaporation of the ether gave a sticky mass in which phloIn a blank experiment in which a solution of the quinol retic acid, 3-iodophloretic acid and 3,5-diiodophloretic acid ether in ethyl acetate was refluxed without @-naphthalene- were identified by paper c h r ~ m a t o g r a p h y . ' ~ sulfonic acid, no 3,j-di-t-but)-lthyropropionicacid could be B. By Acid Catalysis.-The quinol ether 171 (R3 = I , detected in the reaction mixture that remained green. Most Ra = (CH%)*COOH)(0.20 g.) and @-naphthalenesulfonic of the starting material was recovered unchanged. acid (0.02 g . ) were dissolved in 5 ml. of ethyl acetate. The 3,5-Dibromo-3',5'-di-t-butylthyropropionic Acid36 (VIII, solution which was first blue slowly became dark brown. R3 = Br, R4 = (CHs),COOH). A. By Pyrolysis.-The After standing at -25' for 18 days, the reaction mixture quinol ether 1'11 (It3 = Br, R1 = (CH2),COOH) (2.92 g., 5 was evaporated. The residue was extracted with methanol. mmoles) was heated a t 135140' for 5 minutes, then at 165Yellow crystals (0.05 g.), m.p. 144' dec., which did not dis170' for another 5 minutes. Vigorous bubbling ceased solve in methanol, proved to be bis-(4-oxo-l,3,5-tri-t-butylafter the first 3 minutes a t 135140'; thereafter, only little 2,5-cyclohexadien-l-yl) p e r 0 ~ i d e . l ~ Paper ~ chromatography evolution of gas took place. The dark brown reaction mix- of the methanol extract carried out as described in procedure ture weighed 2.74 g. (weight loss, 6.27,; calcd. for the elim- A showed the presence of the same substances that were obination of one t-butyl group, 9.6yo). ;Ifter the addition of a mixture of 25 ml. of 0.2 1%' NaOH and 15 ml. of ethanol, (37) J. P. Leland and G . L.Faster, J . Bid. Chein., 95, 105 (1932).
2062
CHOHHAOLI, EUCENSCMNABEL ANI) DAVID CIIVKC
tained during pyrolysis (procedure A ) and in addition a cotnpound with a n Rr value of 0.82. As this compound reacted not only with FeCls-K3.[Fe(CN)s] but also with diazotized N',~'-diethylsulfanilamide (brownish-yellow spot), i t cannot be a 3',5'-di-t-butpl analog of thyroxine. It is suspected to be 2,4,6-triiodophenol.l* Spontaneous Decomposition of the Quinol Ether VI1 (R3 = I, Rq = CH2COOC2H5).-The following is an example for the spontaneous decomposition of solutions of the iodinated quinol ethers of t h e type V I I , in the absence of a n acidic catalyst. T h e quinol ether VI1 (R3 = I , IZI = CH~COOCZH,) (0.80 g., 1.2 mmoles) was dissolved at room temperature in 30 ml. of n-pentane. T h e resulting light-blue solution began t o turn brownish-red after a few minutes. After standing 3
[CONTRIBUTION FROM THE
Vol. s2
days at rooni temperature the reaction mixture was filtered t o remove an amorphous precipitate t h a t had formed (0.13 g . ) . The filtrate was titrated with 0.1 N sodium thiosulfate. This titration showed t h a t XY0 of the iodine of the quinol ether had been liberated as iodine. T h e pentane layer was concentrated, and t h e concentrate filtered t o remove some more amorphous precipitate (0.18 g.). -ittempts t o crystallize the two batches of amorphous powder (total weight 0.31 g.), failed. T h e powder seems t o consist of a mixture of polymerized substances. It was apparentl5- formed by free radical-induced pc11ymeriz:ition. The filtrate, upon standing at - 2 5 " , deposited yellow crystals of bis-(l-oKo-1,3,j-tri-t-butyl-~,j-cycl~ilie~adien-lyl) peroxide,IAbm . p . 14-1.' dec. BETHESDA,14 MD.
HORMONE RESEARCH LABORATORY, UNIVERSlrY
OF CALIFORNIA, BERKELEY]
The Synthesis of L-Histidyl-L-phenylalanyl-L-ornithyl-~-~yptophyl-glycine and LHistidyl-D-phenylalanyl-L-ornithyl-L-tryptophyl-glyci~e and their Melanocytestimulating Activity BY CHOHHAOLI, EUGEN SCIINABEL AND DAVIDCI-IUKG RECEIVED AUGUST7, 1959 The synthesis of peptides L-histidyl-L-phenylalanyl-L-ornithyl-L-trpptophyl-glycine and L-histidyl-D-pheny~alanyl-L-ortlithyl-L-trpptophyl-glycine is described. These two synthetic peptides have been shown by bioassay to possess melanotropic activities identical to their p phenyl alanyl- and L-arginyl analogs.
In the course of the purification and isolation of adrenocorticotropins (ACTH) from pituitary glands of various species, it has been observed by a number of investigators t h a t every adrenocortically-active fraction always possesses melanocyte-stimulating activity.' It is now well established that the melanocyte-stimulating activity found in the adrenocorticotropins is an intrinsic biologic property of the hormone.1f2 When the chemical structure of the melanotropins (MSH)3s4is compared with that of the adrenocorticotropins,'J it is evident that a heptapeptide sequence, L-methionylglutamyl-L-histidyl-L-phenylalanyl-L-arginyl-L-tryptophyl-glycine, occurs in all preparations of these two hormones. Indeed, synthetic peptides6.'J related to this heptapeptide have recently been shown to possess melanocyte-stimulating activity. One of the outstanding features of these natural and synthetic peptides is the effect on them of alkaliheat treatment. For example, when a solution of L - histidyl - L - phenylalanyl - L - arginyl - L tryptophyl-glycine (1 mg. per nil.) in 0.1 Af NaOH was kept in a boiling water-bath for 15 minutes, the activity which is responsible for darkening the skin of hypophysectomized . frogs was greatly p r ~ l o n g e d . ~It, ~was suspected from unpublished observations in this Laboratory that either the (1) C . H. Li. A d u . Protein Chem., 11, 101 (1956). ( 2 ) C. H . Li, P. F#nss-Bech, I. I. Geschwind, T. Hayashida, G. IIungerford, A. J. Lostroh, W.R . Lyons, H . D. M o o n , W. 0. Reinhardt and bf. Sideman, J. E x p l . Med., 106, 335 (1957). (3) C. H . Li, A d v . Prolein Chem., 12, 269 (1957). (4) J. I. Harris and A. B. Lerner, Nafuve, 179, 1346 (1957). ( 5 ) C. H . Li, J. S. Dixon and D. Chung, THISJ O U R N A L , 80, 2587 (1958). (6) K. Hofmann, T. A. Thompson a n d E. T. Schwartz, ibid., 79, 6087 (1957). ( 7 ) R. Schivyzer and C. H . Li, .Vafure, 182, 1669 (1958). ( 8 ) K. Hofmann, M. E. Woolner, H. Yajima, G. Spiokler, T. A. Thompson and E. T. Schwartz, THISJOURNAL, 80, 6458 (1958). (9) C. €I. Li, Laboratory Inoestigation. 8, 574 (1959).
conversion of arginine t o ornithine or the racemization of L-phenylalanine, or both, may be responsible for the "prolongation" effect of the alkali-heat treatment; consequently, the synthesisof L-histidyl-L-phenylalanyl-L-ornithyl - L - tryptophy glycine (I) and L-histidyl-D-phenylalanyl-L-ornithylL-tryptophyl-glycine (11) was undertaken. The synthesis of I was accomplished in two different ways, which differed only in the blocking of the iinidazo group of histidine. I n one instance, the benzyl group was used to protect the imidazo group, whereas in the other the imidazo group remained free. Carbobenzoxy-L-histidine azide was coupled with L-phenylalanine methyl ester and converted to the carbobenzoxy-dipeptide hydrazide in the manner described by Hofmann, et a l . l o The azide was only slightly soluble in ethyl acetate and chloroforni ; therefore, after the aniorphous material was filtered, it was suspended in an ethyl acetate solution of 6-carbobenzoxy-L-ornithine methyl ester" and kept in the refrigerator for 3 days with occasional shaking. The carbobenzoxy-L-histidyl-L-phenylalanyl - 6-carbobenzoxy L-ornithine methyl ester was contaminated with some histidyl carbobenzoxy-L-histidyl-L-phenylaIanine'O and material formed by Curtius degradation, and could be purified only by recrystallization from a mixture of boiling water and ethyl acetate. The crystalline product was saponified with Ba(OH)s, and carbobenzoxy-L-histidyl-L-phenylalanine-6-carbobenzoxy-L-ornithine (111) was obtained in crystalline form. The condensation of I11 with L-tryptophyl-glycine benzyl ester (IV) was accomplished by means of the dicyclohexylcarbodiimide (DCCI) method.12 The product V was (IO) K. Hofmann, H Kappeler, A E Furlenmeier, h.1 E Wooner, E T Schwartz and T A. Thompson, THISJ O U R N A L , 79, 1641 (1957) (11) R L Synge, Bzochein J 42, 99 (1948). (12) J C Sheehan and G P. Hess, THISJOURNAL, 77, 1067 (1955)
