l. Some Studies in Enzyme Kinetics. II. The Oxidation of 3-Indoleacetic Acid by Plant Enzymes. III. The Synthesis of Some α-Alkyl α-Amino Acids and Their Derivatives Citation Manning, David Treadway (1955) l. Some Studies in Enzyme Kinetics. II. The Oxidation of 3-Indoleacetic Acid by Plant Enzymes. III. The Synthesis of Some α-Alkyl α-Amino Acids and Their Derivatives. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/PC3Z-A208. https://resolver.caltech.edu/CaltechETD:etd-01142004-142952 Abstract NOTE: Text or symbols not renderable in plain ASCII are indicated by [...]. Abstract is included in .pdf document. The kinetics of the alpha-chymotrypsin catalyzed hydrolysis of N-carboethoxy-L-tyrosinamide have been determined in aqueous solution at 25[degrees]C and at both pH 7.9 and pH 8.2. The enzyme-inhibitor dissociation constants of N-carboethoxy-D-tyrosinamide, N-carboethoxy-D-tyrosinmethylamide, and N-carboethoxy-L-tyrosinmethylemide have been evaluated in aqueous solution at 25[degrees]C, the former at pH 7.9 and the latter two at pH 7.6. The L isomers of each enantiomorphic pair appear to have a greater affinity for the enzyme than do their respective D-isomers. An investigation into the nature of the enzymatic oxidation products of 3-Indoleacetic acid (IAA) has been made. 3-Indolealdehyde, o-formamidoacetophenone, o-aminoacetophenone, and 4-hydroxyquinoline have been eliminated as possible major products of the reaction. A study of synthetic routes to the alpha-alkyl alpha-amino acids has been made. DL-alpha-Methlphenylalanine and DL-alpha-methyltyrosine have been synthesized. All attempts to resolve these compounds by enzymatic means have failed. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Chemistry and Biology) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Minor Option: Biology Thesis Availability: Public (worldwide access) Research Advisor(s): Niemann, Carl G. Thesis Committee: Unknown, Unknown Defense Date: 1 January 1955 Record Number: CaltechETD:etd-01142004-142952 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-01142004-142952 DOI: 10.7907/PC3Z-A208 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 165 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 16 Jan 2004 Last Modified: 06 Jul 2023 18:01 Thesis Files Preview PDF (Manning_dt_1955.pdf) - Final Version See Usage Policy. 6MB Repository Staff Only: item control page

I. SOME STUDIES IN ENZYME KINETICS, II. THE OXIDATION OF 3=INDOLEACETIC ACID BY PLANT ENZYMES. ITI. THE SYNTHESIS OF SOME <-ALKYL <<AMINO ACIDS AND THEIR DERIVATIVES. Thesis by David Treadway Manning In Partial Fulfillment of The Requirements for The Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1955 -~iie- ACKNO#LEDGEMENTS To my research supervisor Professor Carl G. Niemann I wish to express my sincere appreciation for his guidance and instruction during my research program, for the wide variety of research problens made available to me, and for the freedom given me in pursuing certain lines of investigation apart from the major research program, I wish also to thank Dr. BR. Buchman and Professor H.J. Lucas for their invaluable counsél and friendly interest in the various synthetic problems arising during the present work. I am grateful to Professor Arthur Galston for an extremely interesting minor research project given me and for the privilege of working with him on the problem. During the course of this work I also enjoyed the interest and advice of Dr. S.M, Siegel. Throughout my graduate residence I have been continually grateful for the support given me by the California Institute of Technology through Graduate Teaching Assistantships and for the receipt of fellowships from the Shell 011 Company and the U.S. Public Health Service which enabled me to continue and complete my graduate training, - iii - ABSTRACT The kinetics of the alpha-chymotrypsin catalyzed hydrolysis of N-carboethoxy-L-tyrosinamide have been determined in aqueous solution at 25°C and at both pH 7.9 and pH 8.2, The enzyme-inhibitor dissociation constants of Necarboethoxy-D—-tyrosinamide, N~carboethoxy— D-tyrosinmethylamide, and N-carboethoxy~L|-tyrosinmethylamide have been evaluated in aqueous solution at 25°C, the former at pH 7.9 and the latter two at pH 7.6. The L isomers of each enantiomorphic pair appear to have a greater affinity for the enzyme than do their respective D-isomers. An investigation into the nature of the enzymatic oxidation products of 3-indoleacetic acid (IAA) has been made, 3=Indolealdehyde, o~formamidoacetophenone, o-aminoacetophenone, and 4-hydroxyquinoline have been eliminated as possible major products of the reaction. A study of synthetic routes to the alpha-alkyl alpha-amino acids has been made, DL~alpha~Methyl phenylalanine and DL~alpha= methyltyrosine have been synthesized. All attempts to resolve these compounds by enzymatic means have failed. _ TABLE OF CONTENTS Part | Page , ACKNOWLEDGEMENTS OPTS TUTE TTT TTT TTT TT eTTTTTeeTeTTTyT ey ii ABSTRACT FSESSCHSSSESSSHSSHS SSS SHSSSSSOSe Sse SesessesSseseesSseseeesessas ili SECTION I I THE KINETICS OF INTERACTION OF THE ENANTIOMORPHIC N-CARBOBTHOXYTYROSINAMIDES WITH ALPHA-CHYMOTRYPSIN IN AQUEOUS SOLUTION AT 25°% SSEeSSSSeHseeeeesesesee eee eee 1 Introduction cecocccccccscccccsecevccesesncccesevecece 2 Treatment of Kinetic Data ceccoceccvecrecesecccccoccece 4 N-Carboethoxy-D~tyrosinamide as an Inhibitor Against Acetyl1-L-tyrosoinhydroxamide SOCOH HOSEL OLED EEO ELEe 27 Discussion of Results cecocscccsccescvececcceseescecee 3 The pH Activity Relationship for the Substrate ....e.. 37 The Redetermination of the Kinetic Constants of Ne Carboethoxy~L-tyrosinamide at pH 8.10=-8.25 sscccove 40 Discussion of Results SCOCSSHE OH SCESHHOHHESEHOCEH EOS OCOOOE 49 IT THE DETERMINATION OF THE ENZYME-INHIBITOR DISSOCIATION CONSTANTS FOR THE ENANTIOMORPHIC N-CARBOETHOXYTYROSIN~ METHYLAMIDES AT pH Te6 and 2500 scccccccccccescccccese 50 Introduction SCCHOS HOES SEE SH CHEO HS SHH EEO ESEHOH ED DOEOE 51 Procedure and Treatment of the Deta SOoecereeeonscesesed 52 Discussion of RESULtS coscccccccececesecscesccsvesecee 61 Experimental SPORE SHCOHHHH HEHEHE HOSS E TES OOOSEHLECESEOE 65 References SPOCHOHHS HEHE SHHECESSOHEHOEHHEHHSCEH EO EHLETEOE T2 SECTION II EXPERIMENTS ON THE OXIDATION OF 3-INDOLEACETIC ACID BY PLANT ENZYMES SSCHSSSSSSSSCSHSHSSSSSHSSESHSHSHSETEHSHHOHAESEE 73 Introduction SOHSS HEHEHE H SH SH SSOHSEHS ESCH EHLESOHELESEEO Th Meterials and Methods POSSSHHSHSHHSSS LEH HHEO LE SEESOEO 79 Experimental Results SCOHSSHSSHSHSSH SEES TE HOSTESASEOCHSOTS 8s Discussion COOCOCHCESEEHE OHS HOSES EH OOOO DE 102 Summary of Conclusions SOCHHS HSH eHOR EHO HE SeHeeceeeseese 107 References CORSO SCOSHE SSH OOOOH SSCS SOEETECOEE SHOE DOSE THEO 108 continued sceccece Part TABLE OF CONTENTS (cont.) SECTION IIT THE SYNTHESIS OF SOME oC -ALKYL oC AMINO ACIDS AND THEIR DERIVATIVES SSCOHSSHSHESHSSSSSSSHSHSLESESHSEHHTOHRESERED Tntroducbion sccccccccceccecccereessecsseneesseceese Synthetic METhOdS coccccccvccensccseseeesesesesseeece Conclusions Regarding the Synthetic Methods ..ceccces Attempted Resolutions of the DL-alpha~Alkyl alpha- amino Acid Derivatives sccsescccccesscncccssecesee Experimental COCOSHHSSSHEHEHHHESEHSHEHESEHSEHERESEHOE RELEVENCSS secocencccesesesesessesesesesesevesssssecs Propositions POHHSOHOSES OHH SESHEH ESSE THEHSETHESL OOH RESDO é 110 111 114 136 139 141 175 178 SECTION TI PART I THE KINETICS OF INTERACTION OF THE ENANTIOMORPHIC N-CARBOETHOXY= TYROSINAMIDES WITH ALPHA-CHYMOTRYPSIN IN AQUEOUS SOLUTION AT 25°C. e 2a troduction Investigations concerning the structural requirements of alpha-amino acid derivatives of the general formula R,CHRAR, capable of interacting with the proteolytic enzyme alpha-chymotrypsin have led to the view''4) that such derivatives combine with the enzyme at its catalytically active site by virtue of specific interactions between the three principal structural features, Ry» Ro» and Ry of the amino acid derivative and three complémentary centers, Py Pos and fy assumed to exist at the active site of the enzyme. In customary terminology the structural characters R,, R 1° “2° the alpha-acylamino group, the alpha-amino acid side chain, and the and Ry represent functional derivative of the carboxyl group of the amino acid derivative, respectively (2), To gain insight into the nature of the forces involved in enzyme-substrate and enzyme~inhibitor interactions it has been of interest to investigate the effect upon the magnitudes of these forces, as revealed by the effect upon the various kinetic constants involved, of varying these structural features. According to the classical formations! 256) the interaction of a substrate with alpha-chymotrypsin and its subsequent hydrolysis in aqueous solution may be represented by k. (1) {&] + [s] “> [Es] {2] + products where the Michaelis constant K. is given by (2) K. = ao = 3 = and ky “represents the specific rate constant of the hydrolytic step. _If, in addition to the substrate, an added inhibitor is present in the system then the following equilibrium must also be considered; z! (3) fe] + (1] ——— [er] 5 for which the equilibrium constant is termed Kye While certain requirements''™4) must be met by all of the structural characteristics of an amino acid derivative to render it a specific substrate for alpha=chymotrypsin, considerable variation may be made in the nature of the alpha-acylamino group R, without serious reduction of this property (7 ), A number of alpha-amino acid derivatives differing only in the nature of the acyl group upon the alphe-amino nitrogen have therefore been investigated in the hope of correlating known electrostatic, steric, or other property of such a group with either the affinity of the enzyme for the substrate or with the susceptibility of the Ry carboxyl function toward hydrolysis. Since previous determinations of the kinetic constants of the acety1{28210) trifluoroacety1 $7) chloroacety1 (2 ) and nicotinyl (8-11) derivatives of various alpha-amino acid amides have indicated a possible relation between their K, values and corresponding molecular volumes of these groups? ) the question arose as to the effect of the bulky carbosthoxy group upon the K, and ky values. Accordingly, the enantiomorphic carboethoxytyrosinamides were prepared with the purpose of testing the i isomer as a specific substrate and the D isomer as a competitive inhibitor of alpha=chymotrypsin. In the case of the iL isomer it was -i- felt that any bulkeattraction or electrostatic influence of the Necarboethoxy group might be manifested in the value of the breakdown constant ky or in the steady state constant K 8 which may be con= sidered, in the case of the acylated amino acid amide substrates, as a measure of affinity of the substrate for the enayme(7?8299!9) re, as was expected, the D isomer proved to be a competitive inhibitor of alpha-chymotrypsin then its affinity for the enzyme would be expressed by Ky» the true equilibrium constant for the enzyme-inhibitor complex (equation (3)). Treatment of Kinetic Data Returning to the previously stated formulation for a system of enzyme, substrate, and added competitive inhibitor, (1) {e] + (s] == {Es] ———> [8] + products ae %, (3) {e]) + (1) —— [Et], Ks and making the following assumptions$!?*©910) that [S] and [I] are much greater than [BE] and that a steady state is rapidly attained in which afs] ,, alzs] afs] .. afer 4) at” at and at at it follows that the rate of enzymatic hydrolysis of a substrate in the a5 «= presence of an added inhibitor is given by . [e) [s] (5) v=- dsl. k, {zs] 3 7K, When [8S] >> K,» V approaches the value k,[E] and at the maximm concentration of enzyme=substrate complex ([ES] = [E]), when the enzyme is saturated with substrate, the velocity becomes (6) v2 k,{Es],, = k, [2] Substituting Van for k,{E] into equation (4) and inverting leads to the useful form eM 1.4 (7) a a (i This equation suggests that at a constant concentration of added inhibitor [I], the reciprocal of the velocity of a substrate obeying the above kinetic formilation bears a linear relationship to the reciprocal of the substrate concentration at any time +t. In the widely used Lineweaver-Burk treatment (6) initial velocities are estimated from constructed tangents at the t = 0 points of plots of [S] versus + or of i1n{S] versus +t. At time zero equation (7) becomes (1 + Lil), 8 (8) L 2 Bf Ps Oe + a Yo Vay s ° Vin and with no added inhibitor present ({I] = 0) a plot of the reciprocals of these estimated initial velocities against the corresponding a This is the inverted form of the Michaelis-Menten‘ '2) equation. =-6 = reciprocal initial substrate concentrations results in a straight K ' line having a slope of 7 and an intercept of + from which these m m constants may be obtained. Since the intercept is independent of added competitive inhibitors, the plotting of ty VS. 7 data for a series of such experiments conducted with increasing amounts of inhibitor present will result in a family of straight lines having an intercept common with that of the line representing uninhibited hydrolysis but having slopes greater than the [I] = 0 slope by factors of (1 + re e The advantage in estimation of instantaneous velocities at time zero arises from the fact that during the hydrolysis of an acylated amino acid amide substrate the hydrolysis product, i.¢e, the acylated amino acid, accumilates, Since such compounds are, in general, competitive inhibitors of alpha-chymotrypsin the velocity at any time +t greater than zero will be less than it would be in its absence and this velocity retarding effect increases throughout the reaction”?! 0) Hence, the estimation of initial velocities from re- actions limited to about 40 per cent total hydrolysis has been the general practice. Otherwise it would be feasible to determine an entire series of TT and + values from one kinetic experiment beginning at an appropriately high initial substrate concentration. While equations (7) and (8) are in theory entirely adequate for determining the desired kinetic constants, the practical difficulty arises of estimating initial velocities from the kinetic data, i.e., from the substrate concentration measured at time t. This situation is best illustrated by integration of equation (7) to give a | (s] (9) k,[E}t = (14 Ee, In Tay? + [s], - (s] which reduces to [s], (10) k[B]t = K, In yey" + [s], - [8] in the absence of added competitive inhibitors. Equation (10) relates, implicitly, substrate concentration to any explicit value of t for a given set of reaction constants and conditions. The two right hand members of this equation may be interpreted as first and zero order terns respectively §” ) Since in the general case the reaction kinetics are intermediate between zero and first order with respect to the substrate concentration a plot of either [S] or of in{S] versus the corresponding time values will give a curving line whose rate of departure from linearity depends upon the relative magnitudes of K, and the substrate concentrations [S]. It may be seen from equation (6) that in the range of large [S] 9 Values, reactions conducted over intervals of low total fractional hydrolysis will receive little con- tribution from the logarithmic term of equation (10) and approximate zero order kinetics. Under these conditions equation (10) becomes , [s], (11) k,[B}t = [s], ~ [s] and In —Tgy 7 in 1 = 9 At low substrate concentrations where the fraction of hydrolysis becomes appreciable the logarithmic term predominates and the kinetics approach the first order extremity of [s], (12) k,[E}t = kK, I-Ts} - & = A high value of K, may also increase the first order character of the kinetics as is indicated by this constant miitiplying the first order - term. A clearer exposition of the interplay between [S] and K, in determining the position of the reaction order between the two extremes is given by the equation * Ls (13) ve iar which derives from equation (5) when [I] = 0. As seen from equation (13) when K > {S] the velocity of the reaction approaches = - {s], or first order in [S], while in the region [Ss] > K, the expression reduces to v = Vin which is to say that when the enzyme is saturated with substrate the velocity is independent of the substrate concentration and is equal to the maximim velocity, i.e., the true zero order velocity. Thus, while a tangent constructed at the t = 0 point of either the [S] vs. t (zero order) or 1n{S] vs. t (first order) plot will yield, theoretically, the true initial velocity (v,) value associated with a particular value of ([S] o? the practical difficulty of drawing such a tangent line increases with the tendency of the plot to curve away from a straight line. This difficulty has, in the past, been overcame to some extent by the estimation of initial velocities from [S] vs. t plots when dealing with initial substrate concentrations higher than the value of K, and from 1n[S] vs. + plots when operating in the region K,> {s] °° For a value of ([s] o ~X, both types of plots depart from linearity with equal rapidity and at such a concentration neither plot is preferable to the other in velocity estimation. -9-= The alphe<chymotrypsin catalyzed hydrolysis of N-carboethoxy~ L-tyrosinamide was followed by the titrimetric method developed by Huang and Niemann!) in which aliquots of reaction mixture are removed at _guecessive time intervals and analyzed for content of hydrolysis product by a formol titration, Fifty-six such experiments were performed at pH 7.9 with the aqueous system buffered by a tris-(hydroxymethyl)-amino~ methene-hydrochloric acid buffer at a concentration of 0.02 F with respect to the amine component. In all experiments the enzyme concentration was 0.208 mg. protein<nitrogen per ml, and the initial substrate concentrations varied from 5.0 x 107? M to 20.0 x 107 M. ne experiment each was per- formed at [8], = 25.0 x 107? M and 30.0.x 107? M and as there was some indication that the solubility limit of the substrate had been reached further runs at concentrations in excess of 20 x 10° M were not attempted. In the case of the present hydrolysis studies, plots of in{s] ve. t appeared virtually linear over a sixty minute experimental interval while the zero order plota invariably curved abruptly below a straight line, The linearity of the first order and extreme downward curvature of the zero order plote (Figures 1 and 2) are explained by the generation, throughout the course of the reaction, of the inhibiting split product N=carboethoxy-L-tyrosine. This enhancement of the importance of the first order term (of equation (10)) by split product inhibition had been noted previously by Huang and Niemann‘'©) in their study of the hydrolysis of nicotinyl-L-tryptophanamide., While the effect upon a plot of ({s] 7 [S]) va. t is to increase further the inherent negative curvature from a straight line, the effect upon the log plot is to compensate for the downward deviation. Sn] 20 \ \ 1.4 . © 1.3} L7 oO 10 20 30 40 50 Time (Min) Fig. 1. First Order Plot of Run No. 42. [s], =10x107 M, [KE] = 0,208 mg. protein-nitrogen/ml, Vyg (LS]g-(5]) #133 - 11. 1.3 fo) 10 FO go 40 50 TIME (MIN.) Fig. 2. Zero Order Plot of Run No. 42. [s}, =10x107 My {E] = 0.208 mg. protein-nitrogen/ml. - 12 = To remedy the difficulty of estimating the true initial velocity of an enzymatic hydrolysis it was decided to employ the velocity correction “method developed by Jennings and Niemann‘!3) tm this method, the individual time values (t), corresponding to measured ([8), - [8]) or ln[S] values of an enzymatic hydrolysis are reduced to new parameters (t') which lie upon the initial tangent line. In the case of the first order logarithmic plots the procedure is as follows: for a plot of 1n[{s] vs. t, the equation Veo tt (4) nfs] = psy- ~ + infs], 8 may be written, representing the line tangent to the i1n[S] = f(t) curve at t= 0. (See Fig. 3.) When multiplied by om the slope of this tangent line gives the true initial velocity Yoo It should be mentioned that the experimental curve depicted in the above diagram is meant to represent the ideal case in which split product inhibition does not tend to reduce the downward curvature, By combining equations (14) and (10), a first order correction factor may be derived which, when divided into the time intervals (t) corresponding to the experimentally determined values of 1n[{S] on the lower curved line (Fig. 3), givestime parameters (t') lying on the tangent line. K (s]. = {s] (15) ee Be + TSI, (x, + (s],) 17g -13- ku [S] Fig. 3. Hypothetical first-order plot illustrating velocity correction method: open circles are experimental points; solid circles represent these values shifted to initial tangent (Eq. 14) by correction factors, -14- In a similar manner +t/t' correction factors may be derived to convert values on the zero order, [S] = f(t) plots to new values defining initial tangent Lines{"3 ) The usual procedure in determining initial _ velocities by this method is as follows. Initial velocities are first estimated from zero and first order plots and from these an approximate K, is calculated according to the Lineweaver~Burk method (equation (8)). This value of K, is then inserted into equation (15) (or in the corresponding equation for zero order correction factors) from which correction factors may be obtained. The corrected values of the initial velocities may then be used to determine an improved value of Ke Since the true initial velocity of an enzymatic hydrolysis obeying kinetics intermediate between zero and first order will, in general, fall between the velocities estimated from the two non-linear plots it is evident that one of the groups of correction factors mist be somewhat less than and the other sanewhat greater than unity with the result that, in a given determination, one of the estimated velocities will be decreased and the other increased by their respective correction factors to give the same intermediate corrected velocity. In the enzymatic experiments with N-carboethoxy~L-tyrosinamide at pH 7.9 initial velocities were estimated from both zero order and first order plots and their reciprocals plotted on separate Lineweaver- Burk plots(®) From these seperate plots two approximate values of K, were obtained by a least squares treatment of the data (see Table I). From these values of Ko» separately applied, zero order and first order - correction factors were calculated and applied to the correction of the first order and zero order initial velocities. As may be seen from -15 = Table II the agreement between corrected zero and first order velocities is good for a majority of the fifty-six experimental determinations, In cases where the discrepancy between corrected velocities was about 10 per cent or greater the ({s], - {S]) vs. t' and iIn{s) vs. t' plots for the run in question were checked for possible errors in velocity estimation. In runs no. 11, 135 215 225 255 30, 325 35, 39, 46, and 48 it was found possible to reestimate either the corrected zero or first order velocity in a manner bringing them closer to the same value. Finally the corrected zero order and first order initial velocities were averaged (last column of Table I) and their reciprocals plotted versus their corresponding reciprocal initial substrate concentrations (see Fig. 4). From these values K, and V were obtained by a least squares determination (Table III). The enzyme~inhibitor dissociation constant Ky of Necarboethoxy-D-tyrosinamide was next determined, Sixteen hydrolysis experiments with N-carboethoxy-L-tyrosinamide as the substrate were performed with the D isomer present as added inhibitor in concentrations of either [1] = 5.0x10°M or [I] = 10.0x 10 M. When the reciprocal initial velocities, estimated from first order plots, were plotted in the Lineweaver-Burk manner two straight lines corresponding to the two concentrations of added inhibitor resulted, The behavior of these lines suggested competitive inhibition by extrapolating to near the ordinate (1-) intercept for uninhibited hydrolysis and the slopes of the lines ineveased in the direction of increasing inhibitor con- centration, TABLE I Kinetic Constants for N-Carboethoxy-L-tyrosinamide from Uncorrected Initial Velocity Data” Kinetic Order Kx? Mg 8 2 0 7423 0.66 1° 5.35 0.69 All determinations at pH 7.9 and {E] = 6.208 mg. protein-nitrogen per ml. > in units of 107 M The first order kinetic constants were actually determined from a total of only 23 runs. In units of 107° nole/min/mg. protein-nitrogen/ml. -17 « TABLE II Initial Velocity Data for N-Carboethoxy-L-tyrosinamide” ‘Run [s] > Uncorr. vie Corr. vie? Aver. Corr. vise cord. 1 ora, cord, 1% ora 1 5.0 056, 063 059, «060, 059, 2 5.0 2060, 0469 2064, 065 26h, 3 5.0 0056, 062 0058, 058, 058, 4 5.0 6064, «069 2064, «065 2064, 5 5.0 2061, 2068 0062, 065 2063» 6 5.0 1050, 40455 050, - 2050, 7 5.0 -050 6059 2053 6056 20545 8 5.0 2060, 060, 2064, 056, 060, 9 5.0 060, 2068 2063 2065, +064, 105.70 0064, 072, 064, 068 +066, 11 5.70 061 20716 0063, 4065, 26h, 12 5.70 073-082 21h, = 077 2075, 13 5.70 6060, 0077 0065, .0705 -068, 14 5070 6063, 071 2066, 067, +066, 15 5.70 07, «076 2073, 6076, 075, 16 5670 6072, 4 084 077, 078 077, 17 5.70 070,081 076, 076, 076, 18 5.70 056, =. 068 2057, «065 061, 19 5.70 = 055, «072 2060, 06/45, 2062, 20 5.70 062, «067, 2064, 4063 063, = 18 « TABLE II (cont.) Run [s] > _—sUncorr. vig. Corr. vis? ‘Aver. Corr. vis? Oord. 18% ora. ord. 15° ora, ¢ 21. 5.70. +063, O77 066, 0716 2068, 23 =5.70 2057 2068 0058, 062, 2060, 24 5-70 2060, 2071, 2065, 068, 067, c 25 6.66 0060, e071 + 066, 2069, +068, 26 6.66 2081, 081 2081, +074, 078, 27 ~=—6.66 073, 2081 2075, 2076 2075, 28 6,66 O75, 0079 O77 2075 076, 29 6.66 00726 0076 O72 072, 072, c 31 6.66 2065, 0075 2068, 072, 070, ce 32 ~=8.00 0755 2087 Oh, +082, 078, 33 = 8.00 he 078 075, 0075 0075, . e 35 8.00 067, 080 070, .0735 07, c 39 8.33 © 072,089 s07h, «080 077, -19 - TABLE II (cont.) Run [s], Uncorr. vi” Corr. via" Aver. Corr. vis” coord, 1% ora, cord. 1% ora, 40 10.0 083, 093 087 083 2085 41 10,0 087%, 4093 087, 088 087, 42 1000 092, 092 -092, .090 091 43 12.50 2087 2095, 2089, 2091 2090, Ah 12.65 085, 096 108g 091, 089, 45 13.75 096, 2099 093, 2097, 2095, 46 = 13,15 0101 e103 097, 0%? 0077, 47 15.15 6089. 100 091, 096 093, 48 17.63 097, 6113, 071, «109, 103, 49 20.0 2103 0108 0103 0102 +102, 50 =—.20.0 e110 0106 e112 2101, 0106, 51 =20.0 2107, e111 2106, 0107 2106, 52 20.0 2108, 0112 0106, 0106 2106, 53 20.0 2100, 0104 2102, - 2102, 54 20.0 105, 0108 2107, Pare) 01056 55 20.0 +104, 0106 e102 2101 101, 56 20.0 0105, 0106 2105, 2101 0103, All determinations made at pH 7.9 and [BE] = 0.208 mg. protein-nitrogen per ml. Velocity values ere in units of 107° Corrected initial velocity reestimated. 1 Menin, K\- 20 14 18 17 16 15 i 13 l2, fi} 10 ‘) CLD OOO CO Fig. he -O5— /O AS” 20 i [S]o Lineweaver=Burk treatment of the average corrected velocity data for N-Carboethoxy-L-tyrosinamide at PH 7.9 and [E] = 0.208 mg. protein-nitrogen/ml. = 21 = TABLE IIT | Summary of The Kinetic Constants of The Enantiomorphic N-Carboethoxy=- tyrosinamides at pH 7.9 Ke Isomer Ke b I [1] = 5.0°% [1] = 10.0 L 6.43 + 050° 0.68 + 04° D 21.07 20.34 (Mean = 20.7 + 2.5) &@ tn units of 107° M b im units of 107 mole/min/mg. proteinenitrogen/ml. ¢ Standard deviations. ~ 22 = ; 1 1 As indicated by equation (8) a plot of 7 (VB. TI, for a ° . system containing an added competitive inhibitor will be a straight line having an "apparent K." equal to : = ft] (16) "apparent K." K(1 + Ky Knowing K, and using a tentative value of Kis a value for the "apparent kK," may be calculated and then applied to the correction of the estimated initial velocities of runs made in the presence of added inhibitor in exactly the same manner as K. is used to determine correc~ tion factors for uninhibited hydrolysis. Bnploying a K 8 of approximately 6 x 107? M and an approximate value of 15 x 10° M for K, (suggested by uncorrected first order data), a series of first order correction factors were calculated for the runs performed at each of the two inhibitor concentrations and the reciprocals of these corrected first order velocities plotted as before. Employing the final value of 6043 x 107° M for K, (Table III) and estimating by eye the best straight lines intercepting the L axis at the predetermined value of 7,10 and passing through the pointe corresponding to [I] = 5.0x 10°? M and {t1] = 10.0 x 1077 M, an average value of 23.1 x 10° M for Ky (for the two inhibitor concentrations) was obtained. Using this improved value of Ky and the final value of Kos a recalculation of the first order correction factors was then effected, The correction factors estimated from this improved value of K. differed from the original I correction factors by amounts of only 0-0.75 per cent and it was not possible therefore to refine further the originally corrected velocities TABLE IV ‘Initial Velocity Data for N-Carboethoxy-L-tyrosinamide Inhibited by N-Carboethoxy=D-tyrosinamide®* [I] = 5.0 x 107 M Run [s], x 10° wn 155 order w : Uncorr. Corr, 1 5.00 2057 20554 2 5.70 2062, 005% 3 6.66 067, 065, h 6.66 2064 20616 5 7.82 2072 20685 6 10.00 .081 078, 7 14.28 094. ° 090, [1] = 10,0 x 1077 M Run [s], x 10° 4 1 order v? Uncorr, _ Corrs 1 4.00 42, 2041, 2 5.00 048, 048 3 5.70 4 5.70 055, 20530 5 5.70 20555 20542 6 6.66 2057, 2056, 7 7.82 -063, 2061, g 10,00 2070, 068 9 10.00 2076 0736 All determinations made at pH 7.9 and [E] = 0.208 mg. protein~ nitrogen per ml. Velocity values are in units of 107° Memin Corrected initial velocity reestimated. Erratic run rejected, -aA<- by the recalculated correction factors, A summary of this velocity data is given in Table IV. As indicated in Table IV, however, the estimations of corrected velocities were, in a few cases, revised before being entered inte the final Lineweaver-Burk plot (Fig. 5). It should be noted that these reestimations were merely a matter of rejudgement upon the originally calculated values of t' vse. 1n{S] and did not result from any recalcula- tion of correction factors. The corrected reciprocal first order initial velocities and reciprocal initial substrate concentrations for each inhibitor concentration were plotted as before and values of Ky determined for each line by a least squares treatment. The lines corresponding to the two inhibitor concentrations were first made to pass through the predetermined ordinate intercept by giving this value of 1 the high m I for the two values of inhibitor concentration were found to be in excellent arbitrary statistical weight of n = 100. The values of K, thus determined agreement (Table III). The exclusive use of corrected first order initial velocities in determining K, was considered sufficiently accurate especially in view of the 25 per cent = 50 per cent enhancement” of the first order term (see equation (9)) by the miltiplier (1 + {1 , It % should be noted that in this determination of K, competitive inhibition I has been assumed to occur by deliberately causing the lines for inhibited hydrolysis to pass through the predetermined intercept. In Table V are shown the intercepts for the two inhibitor concentrations which have been calculated on the basis of inhibited hydrolysis data alone. The intercept for the substrate alone and its standard deviation are shown for comparison. The overlap of error limits indicates that, within the ® this was also noted by Huang and Niemann‘! 0}. TABLE V Data for N-Carboethoxy=D-tyrosinamide as an Inhibitor Disregarding The Intercept Predetermined from Uninhibited Hydrolysis. (L isomer included for comparison) Isomer 1 (ordinate intercept) Stan Dev. in 1/V L 7.10 + 0.38 D (at 5x10 M) 7.40 -> D (at 10x 107? M) 7.53 + 0.66 a b Units of 10? sec. M Not calculated since its difference in value from that of the L is less than the error of the latter, - 26 -— {1] = 10 [rz] il Wr a eO“m7". ry (a) n oO 9 OF 10 AS -2O 25° Fig. 5. Lineweaver-Burk plot of N-Carboethoxy-D-tyrosinamide as a Competitive Inhibitor of its L Antipode. =~ 27 = linits of experimental error, the three lines extrapolate to the same intercept. Hence, within these experimental limits, competitive inhibition is indicated. Assuming competitive inhibition, the use of the more precise intercept determined from the substrate alone is desirable con= sidering the great sensitivity of K, to the intercept and the good agreement between K, values noted in Table III. The justification of I this assumption awaits the independent evaluation of K for Necarboethoxy= D-tyrosinamide against a second substrate. N-Carboethoxy-D-tyrosingmide as an Inhibitor Against Acety]-L-tyrsoin- hydroxamide Six experiments” were performed in which N-carboethoxy- D-tyrosinamide, at a concentration of 10 x 1072 M, inhibited the chymotryptic hydrolysis of acetyl-L-tyrosinhydroxamide. The kinetics were followed by the spectrophotometric technique described in a later section (page 52). From an approximate apparent K, of 65 x 10°? M, first order correction factors were calculated and from these reciprocal corrected velocities obtained, In Table VI is a summary of the uncorrected and corrected initial velocity data. An ordinate intercept was first calculated from the six points using a least squares treatment. Its value of 0.68 is somewhat below the expected value of 1.11 calculated from the data of Foster and Niemann‘?7) , A second least squares treatment was then performed causing the line to pass through the intercept determined for acetyl- L-tyrosinhydroxamide alone (Fig. 6). While the resultant K, of = es In these experiments, [E] = 0.0266 mg. protein=nitrogen/ml. TABLE VI N-Carboethoxy-D-tyrosinamide as Inhibitor Against Acetyl -L-tyrosinhydroxamide [1] = 10x 107 4 Run (s]¢ Uncorr. 8 ord. vw Corr. 15% ord. ve 1 5.00 0064 0064 2 5,00 2062 062 3 6.02 076 2075 4 7.50 0094 2092 5 10.00 0125 121 6 10.00 0123 0120 a In units of 107? M > in units of 107 M/min, bw FH oN & XS NS} o OS co) AS 0 | [Slo Fig. 6. N-Carboethoxy-D-tyrosinamide as an Inhibitor against Acetyl-L-tyrosinhydroxamide. - 30-< 17.3 x 107? M 4s below the mean value of 20.7 x 10° M found against N-carboethoxy-L-tyrosinamide , it lies within the limit of error of this | latter value. This error derives from the uncertainty in XK _ (6.43 + 0.5) of the carboethoxy substrate which figures into the calcu- lation of K, by equation (16), Table VII summarizes the calculations I based upon the data of Table VI. - 31 = TABLE VII Least Squares Data for N-Carboethoxy=D-tyrosinamide as Inhibitor Against Acetyl-L-tyrosinhydroxamide Ordinate Intercept 0.68" + 0,31" K, (assuming competitive inhibition) 17.3 x 107 +1.2x 107? 4 ® sn units of 10° M7”! omin, Standard deviation. For calculation, see Youden, "Statistical Methods for Chemists", 1951, John Wiley and Sons, Inc. Discussion of Results The discovery that, in the case of the enantiomorphic N-carboethoxytyrosinamides, the K, of the D isomer (19.0 x 107 m)* I has a larger value than K, of its L antipode (6.4 x 107 M) was unexpected in view of the fact that in the case of every other D and L pair of acylated alpha~amino acid amides studied Ky has always been found to be less than Ko Since it is generally assumed (2981910) that in the case of the acylated=-Leamino acid amide substrates K, is approximately equal to 2, i.e., is a measure of the affinity of enzyme for substrate in the ES complex, the usually observed order for BD and ZL pairs of K Re K, has been interpreted as signifying that, in such I pairs, the 2 inhibitors have greater affinities for the enzyme than do their corresponding L isomers 7) This apparently lower stability of the ES complex has been considered to reflect a strained configuration of the substrate molecule in its act of meeting the geometrical require- ments of the ES union. The EI complex has been considered to exist without this strain(2*'4) According to one view, the specificity of alpha-chymotrypsin (and of proteolytic enzymes in general) for L amino acid (naturally occurring) derivatives reaults from the residence of this strain, in part, in the carboxylic acid function Ry thereby lowering its activation energy toward nucleophilic attack('4) This strain would result from the interference offered by the alpha-hydrogen atom toward RY, » Ros R,/, interaction during the act of duplicating the Averaging the value of K, obtained with its L antipode as the sub- strate with the value of ky determined against acetyl-L-tyrosin~ hydroxamide as the substraté, strain free EI type union in the inverted fashion, Another possible explanation for the specific hydrolysis of L derivatives is, of course, that the Ry group in EI may be unfavorably oriented with respect to subsequent attack(") In agreement with the concept of similar mode of combination in ES and EI is the observation® that for three pairs of acylated alpha-amino acid amides, differing (within each pair) only in the nature of the acyl group (Ry)» the ratios of the K 38 of the L members of each pair are roughly equal to the corresponding ratios of the K,'s of the D members. This ratio is approximately independent of I the nature of Ry the cases investigated (2) For these observations to suggest the same (the amino acid side chain) and is close to two in mode of combination for substrate and inhibitor requires that they also suggest (17) K. = i r 1.Gc,y k, >> k, for the acylated alpha-amino acid amides. In Table VIII, the K, and ky values of Necarboethoxy-L-tyrosinamide are compared with those of several other acyl~L-tyrosinamides. It will be noted that the K, values fall into two general groups; the high values of the acetyl, chloroacetyl, and trifluoroacetyl derivatives and the lower values of the bulkier benzoyl, nicotinyl, and isonicotinyl substrates, With a K, of ° Specifically, the acetyl and nicotinyl derivatives of the enantio= morphic tyrosinamides, phenylalaninamides, and tryptophanamides, These ratios are not equal in the cases of some chloroacetyl and trifluoroacetyl derivatives, suggesting the possibility of alternate modes of combination in these cases, ~ 34> . 6.4 x 107 M the carbcethoxy substrate lies in this latter category. The significance of the relation of the nature of the acyl group to K, has been discussed by Shine and Niemann‘) for the first four members of Table VIII. Evidently, the electronic character of the acyl group has little influence on the magnitudes of the kinetic constants, On the other hand, the bulk of the acyl group, expressable as its molecular refractivity, seems to be related to the size of Ke» the lower K 3° (higher affinities) being associated with the more voluminous acyl groups. The position of the formyl derivative in this apparent eategory is not understood. The possibility has been suggested that the high affinity (low K,) of the carboethoxy derivative may result from a cyclic configuration (equation (18)) of the carboethoxy group, assumed upon the occasion of ES interaction, in which it may approximate the molecular volume of a larger group, f.e., benzoyi{! 5) —_ (18) onl oe N | C=O This relation of size of the acyl group to affinity for enzyme is considered in agreement with the present concept that the forces responsible for the formation of the ES complex are primarily van der . Waals forces (3 116) - 35 TABLE VIII _ Kinetic Constants for a Number of Acylated L-tyrosinamides at pH 7.75-8.0. Substrate Molec. Ke ie Refr. of Acyl 8 3 Acetyl-L-tyrosinamide 10.3 32 204 Chloroacetyl-L-tyrosinamide 15.1 27 4.0 Trifluorcacetyl-L-tyrosinamide 10.2 26 2.6 Nicotinyl-L-tyrosinamide 25.8 12 5 Isonicotinyl-L-tyrosinamide 9 6.4 Benzoyl-L,-tyrosinamide 205 4.0 Fornyl1~L-tyrosinamide 12 oA5 N-Carboethoxy=L=-tyrosinamide 6.4 268 a 3 In units of 10 ~ M. b 3 In units of 10” mole/min/mg. protein-nitrogen/ml. ~- 3% - The right hend column of Table VIII indicates that ky may also be related to the size of the acyl group, the magnitude of this constant generally increasing with the bulk of the acyl group. With . this in mind, the low value of 0.45 for the formyl derivative is under= standable while the relatively low value for N-carboethoxy-L-tyrosinamide is surprising. It seems not unreasonable to assume that this low value of ky for the carboethoxy substrate is a manifestation of an unusually unstrained (and therefore stable) ES complex originally suggested by the "inverse" order of affinities of the D and L isomers. - 37 - Since the pH optima for the chymotryptic hydrolysis of all | previously investigated acyl derivatives of Letyrosinamide have been .found to be within the region pH 7.75-8.0, the kinetic constants of N=carboethoxy=L-tyrosinamide were determined at the usual pH of 7.9 before making a careful investigation of the pH-activity curve. The pH-activity relationship for the system N-carboethoxy= L-tyrosinamide-alpha-chymotrypsin in aqueous solution at 25°C was determined by performing a lerge number of short-term hydrolysis experi- ments buffered at the desired pH with a tris-(hydroxymethyl)- aminomethane«hydrochloric acid buffer at a concentration of 0.02 M with respect to the amine component. With this buffer it was possible to work over a range of pH 7.25 to pH 8.7, limiting the reaction time to the first 20 minutes for an enzyme concentration of 0.266 mg. protein-nitrogen per ml, The usual procedure was to determine, titrimetrically, the increase in liberated acid over the 20 minute reaction interval, taking the total change in end point for this period as @ measure of activity. The uncorrected zero order initial velocities (essentially, these were determined) were, in general, estimated from six total experimental points, three in the 1-25 minute range and three at 18-195 minutes, It was felt that limiting the estimation of end points change to this brief hydrolysis period adequately eliminated the complication of a significant pH change during reactions performed near the extremes of the pH range studied and also minimized the problem of split-product inhibition, - 38 - In Figure 7, the twenty minute hydrolysis increments, expressed as fractions of the maximm hydrolysis increment observed, are plotted versus the corresponding p's of the determinations. In the _ usual case where the pH varied over an interval of 0.02-0.10 units fron beginning to end of the run the initial and final pH values were averaged and these values plotted vs. the relative hydrolysis increments. Figure 7 indicates that the pHeactivity curve of N-carboethoxy- Letyrosinemide possesses a maximum near pH 8.25, differing in this respect from all previously investigated derivatives of L-tyrosinamide, the majority of which pass through a maximum at or near pH 7.9, At this pH and at the experimental initial substrate concentration of 10 x 10° M, Necarboethoxy-L-tyrosinamide is hydrolyzed at a rate of approximately 95 per cent of that at its maxim, In Figure 6, the PpHeactivity curve of acetyl-L-tyrosinamide (7) is shown for comparison. While the curves for the two derivatives are seen to be of the same general shape, that for the carboethoxy derivative is shifted toward the higher pH's by a value of approx. 0.35-0.40 pH unit. ‘The possession of a rather high pH maximum seems to be a further manner in which N-carbosthoxy-L-tyrosinamide departs from the typical behavior of the acyl-L-tyrosinamide substrates, 1.0 » F 2 - WY < 09 WwW 2 = x i / uw / pag J / 07 rae) 7S 8.0 os 7.0 “PH Fig. 7. N-Carboethoxy-L-tyrosinamide [s]}, = 10x 1073 M3 ee ee ee ee ["] = 0.266 mg. protein-ntirogen per ml.; 0.02 M-tris-(hydroxymethyl )~aminomethane-hydrochloric Acid Buffer, For Comparison, AcetytL-tyrosinamide, [s], = 20x 107M; (E] = 0.125 mg. protein=nitrogen/ml. After Thomas, MacAllister, and Niemann? . The Redetermination of The Kinetic Constants of N-Carbosthoxy- I-tyrosinamide at pH 8,108.25 While the knowledge of K, and ky . Latyrosinamide at pH 7.9 may be considered of interest for the sake of for N-carboethoxy= comparison with the kinetic constants of other acyl-L-tyrosinamides at the same pH, it was nevertheless desirable to have information regarding the carboethoxy substrate at its optimum pH of hydrolysis. Accordingly it was decided to redetermine the constants at a pH near the optimum value of 8.25. | To obviate the tedious process of estimating initial velocities from two types of plots and correcting these estimates to values for use in the Lineweaver-Burk treatment, it was decided to employ the recently developed method of Foster and Niemann‘17) which does not require the separate evaluation of initial velocities. For a system consisting of enzyme and substrate with no added inhibitor but in which competitive inhibition by one of the hydrolysis products occurs throughout the reaction, the following representation may be given! 256) (19a) [e] + [s] 5 {es} (B] + [P] (19b) [e] + [P] ~~ [EP] The course of such an enzymatic hydrolysis, inhibited by liberated re- action product, is given by the integrated rate expression -41- {s], {s}, K, (20) k felt = (1 +") KL In-pey + (1 - =) CES], - [s]) P B - where Ko is given by k, (21) Ky 7 Ke From equation (20) it is evident that for an experiment conducted at 1 ° particular value of [Ss] o? @ Plot of = (fs), - ee ts) () t in Ts] will be a straight line of slope < K, ho) « Since a straight line drawn through the origin of such a plot will have a slope equal to [S] o? that particular straight line drawn through the origin having a slope equal to the initial substrate concentration of a given experiment will intersect the line defined by the data of that experiment at an ordinate value equal to the corresponding initial velocity Yoo If then a straight line is drawn through a series of these intersections derived from a series of experiments conducted at the same enzyme concentration but differing in initial substrate concentration it will have a slope of WK, » an ordinate intercept of k,[B] and an abcissa intercept of k,[E]/K,- Besides eliminating the necessity of a preliminary evaluation of initial velocities, this treatment has the further advantage of offering a method for determining the value of K. P Of course, since the intersections of the experimentally determined lines with the constructed lines of slope equal to [s] } yield values of the true initial velocity vy» K, and ky may be alternatively determined by a Lineweaver=Burk treatment of these velocities. A disadvantage in this latter method lies in the fact that @W42= K, thus determined is a function not only of the slope of the 1 VS. tr line but also of its intercept and therefore inherits the un- cortainty of determination of this latter quantity. A total of eighteen experiments, buffered by 0.02 M THAM at pH 8.10-8.25, were performed, following the kinetics by the titrimetric method as before, In these runs the enzyme was an Armour preparation of alpha-chymotrypsin at a concentration of 0,532 mg. protein-nitrogen per ml. Since the Foster-Niemann method of analysis enables the direct interpretation of experimental data taken at high total hydrolysis, the reactionswere allowed to proceed to 63 per cent ~ 94 per cent hydrolysis over time intervals of approximately 50 minutes, By studying the reaction system at intermediate and later stages of the reaction, the experimental error is decreased and it becomes possible to obtain an estimate of Ky from the experimental lines having slopes equal to “K,(K,, + [s] AK, - x,)$17) In general the calculated values [s] t in Ts] behaved erratically over the first 10+20 minutes of each of reaction, At these initial stages of the reaction the values of In[S], - 1n{S] correspond to small differences between rather large values and the error is therefore quite large, The experimental data was plotted both according to the Foster=Niemann method (Fig. 7) and in the Lineweaver-Burk manner (Fig. 8) using a least squares treatment in each case to determine the best straight line. In the case of the latter plot, the L values are the ° reciprocals of the + ({s], - (S]) values obtained from the Foster= Niemann plot intersections. In Table IX these velocity values are - 43 = summarized, By omitting the data from runs 1, 2, 3, 6, 10, 12, which fell rather far from the lines defined by the majorities of the points of both plots, values of K, and k., in excellent agreement between - plots, were obtained (Table xX). From the expression | si - [s], (2) Xk Ee where Sl is the slope of a line defined by the experimental points, values of Ky were calculated and tabulated in Table XI. From equation (22) it is evident that at low values of Sl, i.e., close to K,» the denominator becomes a small difference between two relatively large numbers (since Sl is negative in sign) with a large resultant un- certainty in the calculated K p This is further complicated by the fact that at low values of ([S] o? the slope was often very hard to (} 1 1 = in tal ¢ ({s], - [S]) points hed a tendency to fluctuate rather badly. In theory, the slopes of these lines mst determine as the be greater in absolute magnitude than K, for Ky to have a finite positive value. Table XI indicates that, at the lower (s], values, the estimation of a reasonable value for K, was not possible. In most previous cases studied, K, is roughly three times the value of K, of the parent substrate. As runs 4, 5, 7, 8, 9, 13, and 14 gave experimental points the majority of which suggested straight lines of Slope greater than ~K,» the value of K, estimated from these runs were averaged giving a mean value of 32.3 + 13.4. Setting K, equal to this value and K, equal to 7.27, the expression for the slope becomes TABLE IX | ‘Initial Velocity Values of N-Carboethoxy-L-tyrosinamide at pH 8.10-8,25 and 25°C {&] = 0.532 mg. protein-nitrogen/ml. | a b 1 (slo Run . {s], ‘5 ¢ in “TsT 1/{s], /y, Comment 1 13.15 0246 20187 0761 4206 reject 2 5.0 2161 «0320 «200 6.21 reject 3 6.66 2189 20283 0150 5.29 reject 4 10.0 e272 0272 2100 3.68 5 20.0 0337 20169 050 2.97 6 2.0 - - - - reject 7 10.0 0253 00254 2100 3.95 8 5.70 2187 20327 01754, 5.35 9 20.0 0327 20163 2050 3.06 10 2.0 0084 - +500 11,83 reject 11 3.33 0145 ©0437 +300 6.90 12 8.00 - - - - reject 13 8.00 0242 -0303 0125 4e12 14 6.66 0216 +0323 21501 4.63 15 5.70 2202 20352 01754 495 16 5.70 200 20352 01754 5.00 17 5-70 0187 20329 01754 5.33 18 5.70 0192 20337 01754 5.19 ®@ in units of 107° M b in units of 107? M/min. © In units of mint! -~145- ¢ ({STe— [S]) rs ° .O0/ 08 -03 .O# OS + 2a Bde t {s] Fig, 8. Foster-Niemann Plot for N-Carboethoxy-L-tyrosinamide at pH = 8.10-8.25. ~ eqoTd UNBUETN-teyso wory 84°A "Gc°S-OT'S = Hd 2e epTueuyTsos£4—T-fxoyjeOQTBeN JOS 3OTd YMgeTaavesouTT °6 *STa *fS] { : OE" T-4 OZ: SI or: 50° O - 47 « TABLE X N-Carboethoxy~L-tyrosinamide Dependence of Kinetic Constants upon pH Type of Data Analys is ~-Genstent ___ K k 8 3 7 85=7.9 Lineweaver-Burk 6.43 0.68 +.50° +.04° 8.10=8,25 Lineweaver=Durk 7.17 0.84 8.10-8,25 Foster=Niemann 7.27 0.84 +.60° +.03° 8.10~8.25 (mean) 7.22 0.84 a 3 In units of 10°” M. b 3 In units of 10° mole/min/mg. protein-nitrogen/ml. ¢ All error limits are standard deviations, . Estimates of & TABLE XI for N-Carboethoxy-L-tyrosine at pH 8,108.25 and 25°C Run (s}¢ Slope* Cale. Comment 1 13.15 K, 0 Rej. in K, det. 2 5.0 K, 0 ft 3 6.66 8.14 124 t 4 10.0 22.45 16.5 " 5 20.0 “47 oh 12.2 6 2.0 - - 7 10.0 12.77 30.1 a 8 5.70 ~$ 66 1502 9 20.0 4267 12.8 10 2.0 K,, 0 " 11 3233 K, fe) L] 12 8.00 - - r 13 8.00 -18,69 2323 14 6.66 9.34 56.2 15 5.70 | K. 0 16 5.70 K, ) 17 5670 -7.48 456 18 5.70 K, 0 ® tn units of 107 M, -A9- (23) | Sl = -9.38 - 0.29[s], _ Since (23) would predict a slope of only -15.2 for an [S], value of 20 x 107? M (where virtually all of the experimental points fell on lines of slopes -42 to -47) it seems fairly certain that K,, for N-carboethoxy-L=tyrosine is less than this average value, probably in the vicinity of 12-18 x 107° M. Discussion of Results The value of K, for N-carboethoxy-L-tyrosine appears to lie approximately in the range expected from the K, of its parent substrate, The behavior of the carboethoxy substrate observed upon changing the pH of the hydrolytic reaction by approximately 0.35 pH unit is somewhat unexpected. Such changes in K, with pH are more characteristic of a charged, rather than of an uncharged, substrate (18) Somewhat surprising also is the decrease in affinity (higher K,) at the pH optimum over that observed at the lower pH. While this difference in K 3 is doubtless of significance it should be emphasized that the two values differ by an amount only slightly greater than the standard deviations in these values. This decrease in affinity at the maximum is more than offset by the increase in ky which is related to K, and to the initial velocity by {z] [s] - 50 = SECTION I PART IT THE DETERMINATION OF THE ENZYMR=INHIBITOR DISSOCIATION CONSTANTS FOR THE ENANTIOMORPHIC N-CARBOETHOXYTYROSINMETHYLAMIDES AT pH 7.6 AND 25°C. Introduction The observation of the so-called inverse" order of K, and Ky for the enantiomorphic Necarboethoxytyrosinamides prompted specu- lation concerning its significance. While it would seem to indicate the existence of a case in which a substrate is more tightly bound to the enzyme than is its D antipode, nothing certain is lnown about the magnitudes of the constants k, and k, and hence no definite con~ clusions may be drawn. only if it could be taken for granted that in the case of Necarboethoxy-L~tyrosinamide k, >> ky so that K, would represent, for all practical purposes, a true equilibrium constant would the suggested order of affinities be a certainty. An alternate approach is to study compounds sufficiently similar to N-carboethoxytyrosinamide in structure to permit comparison, but differing from that compound in some minimam structural detail rendering both D and L isomers competitive inhibitors of chymotrypsin. A similar "inverse" order of K's in I such a case would give force to the idea that the K 8 to K, relation=- I ship of the enantiomorphic Necarboethoxytyrosinamides represents an order of affinities, Since N-carboethoxy~L-tyrosinmethylamide is hydrolyzed only very slowly’ by alpha-chymotrypsin and therefore approxi- mates a true inhibitor (k, = 0) in behavior it was decided to determine the Ky of this compound and of its D antipode and compare the two. Suggested by Dr. Niemann, this approach originated in a proposition by D.W. Thomas, Ph.D. Thesis, Calif. Inst. of Technology, 1951, b Actually inferred from an unpublished experiment with acetyl-L- tyrosinbutylamide in these laboratories, - 52- Procedure and Treatment of Data The enzyme~inhibitor dissociation constanta of D and I N-carboethoxytyrosinmethylanide were evaluated against acetyl-L- tyrosinhydroxamide as the specific substrate at 25°C in aqueous solution buffered at pH 7.6 by tris-(hydroxymethyl)-aminomethane- hydrochloric acid (THAM) with the amine component at a concentration of 0.27 F. The hydrolyses were followed by an extension of the method of Hogness and Niemann‘ '9) which depends upon the colorimetric estimation of the colored complex formed between the hydroxamic acid and ferric ion. In the present experiments the absorbance of residual acetyl-L-tyrosinhydroxamide, as its ferric complex, was measured at a wavelength of 515m with a Beckman Model B Spectrophotometer, Since the Beer's Law expression (25) [Ss], = KxA where [S] + represents the concentration of the hydroxamide=ferric complex (and hence of the hydroxamide itself) and A is the absorbance, is known to apply to this system at ) = 5id5mu, a plot of In A versus. time has an initial slope which, upon multiplication by [Ss] °° yields the first order initial velocity. In the experiments with the methylamides values of ln A were plotted versus time and the reciprocal first order initial velocities estimated from these plots entered into Lineweaver=Burk plots. In these preliminary plots all lines for inhibited hydrolysis were drawn through the intercept for acetyl~L-tyrosinhydroxamide determined by Foster and Niemann‘"”) , From these plots provisional values of apparent - 53 = K 3 vere estimated and these used to calculate first order velocity correction factors. Tables XII and XIII present a summary of the first order velocity data for the methyl amides each of which was - tested at concentrations of [I] = 10x 107M and [I] = 20x 107M. Due to the high K, of acetyl-[-tyrosinhydroxamide (42 x 107? M) and the enhancement of the first order terms by the added inhibitors the hydrolysea were close to first order kinetically and the first order correction factors close to unity. Plotting the reciprocal corrected velocities in the Lineweaver- Burk manner, ordinate intercepts were calculated by a least squares treatment. These intercept values and their standard deviations are shown in Table XIV compared with the intercept for uninhibited hydrolysis. In a second series of least squares determinations, com= petitive inhibition was assumed to occur by causing the lims to pass through the ordinate (1/V) intercept for plain hydrolysis of acetyl- L-tyrosinhydroxamide. The values of K, assuming competitive | inhibition are presented in Table XV. Table XIV indicates that while the ordinate intercept values for the two concentrations of N-carboethoxy= D~tyrosinmethylamide differ from the substrate intercept by amounts somewhat greater than the combined calculated standard deviations of both intereepts there is no progressive increase of intercept with increasing inhibitor concentration that would suggest the occurrence of inhibition of a type other than competitive. Furthermore, the values of K. calculated upon the assumption of competitive inhibition I at the two inhibitor concentrations are in good agreement with each other. - 54 = TABLE XII Initial Velocity Data for Acetyl-L-tyrosinhydroxamide Inhibited By N-Carboethoxy-D-tyrosinmethylamide® [1] = 10x 107 w Run [s]° First order vw Unecorr. Corr. 2 6.02 082, 2081, 3 7.50 2100 2098) 4 7.50 2100 2098) 5 10.00 0133 +130, 6 15.00 2173, 170, {I] = 20x 10774 Run (s]¢ First order ve Uncorr,. Corr. 1 5.00 052, 0052, 2 8.00 2082, 081 6 3 15.00 | 0148, 0145) All determinations made at pH 7.6 and [E] = 0.0266 mg. protein- nitrogen per ml, Velocity values are in units of 107? Meminz' In units of 107? M, - TABLE XIII Initial Velocity Data for Acetyl-L-tyrosinhydroxamide Inhibited By . N-Carboethoxy-L-tyrosinmethylamide® [1] = 10x 1077 Run {s]¢ First order vw Uncorr. Corr. 1 5.00 053, 2053 2 5.00 0514 051 3 3 6.02 +062, 2061 Ps 5 10.00 ol 00, 2098, [1] = 20x 107 M Run (s}¢ First order we Uncorr, Corr, 2 15.00 +089, 2088, 3 20.00 2120 el 18, 5 40.00 2205 2201 All determinations made at pH 7.6 and [E] = 0.0266 mg. protein- nitrogen per ml, Velocity values are in units of 107? Memins! In units of 107° Me ~- 5% « ; TABLE XIV Data for The Enantiomorphic N-Carboethoxytyrosinmethylamides as Inhibitors Disregarding the Intercept Predetermined” from Uninhibited Hydrolysis. (1/v* for acetyl-L-tyrosinhydroxamide = 1.106 + 065)” Isomer Conc. 1/v* Stan. Dev. (Ordin. int.) in 1/V D 10x 107M 1.40 + 0.23 D 20 x 107 M .70 + 0.27 L 10 x 107? M 1.27 + 0,32 L 20 x 107? M 1.61° + 0.24 L 20 x 107 M 1.39% + 0,37 In units of 10° M”’ sec. Determined by Foster and Niemann‘"7) , Including the experimental point as [8], = 10 x 107? M. Quitting the point at [S], = 10 x 1077 M. ~ 57 = ' FABLE XV . Summary of the Enzyme-Inhibitor Complex Dissociation Constants of The Enantiomorphic N-Carboethoxytyrosinmethylamides at pH 7.6 and 25°C Isomer Kr (assuming competitive inhibition) were (1]=10x107%mM [I] = 20x10 D 22.2 + 1.69 21.4 + 0.9% Mean = 21.8 + 1.6% L 10.5 + 0.5% 8.8° pt 0054 9.1 Mean = 9.5 + 1,2¢ ® mn units of 107 M, b Including the experimental point at [Ss], = 10 x 107? Me Q Omitting the point at [s] 9 7 10x 107° Me for) Estimated by eye from Fig. 10 and Fig. 11. Hence the assumption of competitive inhibition seems justified. The case of the L-methylamide is less certain, Here the [I] = 10 x 107? line has an intercept within standard deviation range of the substrate intercept while the [S], = 10x 10°? M point must be omitted to cause the [I] = 20x 1077 M line to intersect the ordinate axis within thia range. There appears to be a slight increase in intercept with inhibitor concentration. Moreover the assumption of competitive inhibition yields values of Ky at the two inhibitor concentrations which differ by an amount greater than the combined estimated errors with the lower value of K. at the higher inhibitor concentration (Table XV). Such I an apparent decrease in K, with increasing {I] could result from overestimating the slope of the line by drawing the intercept tolow, ieee, through the substrate intercept. It is clear that the few experimental points of Figure 11 and the rather arbitrary significance of standard deviation do not allow any final judgement as to the behavior of the intercept with changing inhibitor concentration, This apparent behavior does, however, indicate the desirability of further experiments at high substrate concentrations using some method of plotting (such as a plot of v, versus v/{s),) °°) better suited to revealing possible differences in intercept. is 4 13 3 i lo q 8 7 6 5 4 3 e i °6 OF 10 15 20 Slo Fig. 10. N-Carboethoxy-D-tyrosinmethylamide as an Inhibitor against Acetyl-L-tyrosinhydroxamide. - 60 = O OF 10 AS .20 | [S]o Fig. ll. N-Carboethoxy-L-tyrosinmethylamide as an Inhibitor against Acetyl=L-tyrosinhydroxamide. - 61 = ' Biseugsion of Results While the occurrence of purely competitive inhibition in the sense of competitive three point attachment of the inhibitor cannot be considered established by the above data it is reasonable to assume that the Ky values calculated for the methylamides on the assumption of competitive behavior indicate magnitudes of enzyme-inhibitor affinity in the EI complexes. The occurrence of an actual increase in intercept with increasing (t] in the case of the L isomer would result if a certain fraction of trimolecular IES interactions took place in the system. Such interactions could be of the noncompetitive or uncom= petitive Tes type or, alternatively, a simultaneous attachment of substrate and inhibitor at the same active site. This type of inter- action was found to occur in the case of the inhibition of the hydrolysis of the bifunctional substrate, methyl hippurate, by indore 4), It was noted in the previous part of this chapter that N~carboethoxy- L-tyrosinamide has a surprisingly high apparent affinity for the enzyme and at the same time a low kg The same is true of the affinity of the Lemethylamide. A possible explanation of this behavior is that in the cases of the L isomers other modes of combination with the enzyme may be possible than that in which the L molecule duplicates the combining mode of its D isomer in the inverted manner, Such might be the case if the R, carboethoxy group, in molecules of the L configuration, were able to interchange roles with the Ro para-hydroxybenzyl side chain by assuming the cyclic hydrogen bridged configuration mentioned earlier in this chapter. The occurrence of a certain percentage of such ~ 62 = alternate type interactions could be responsible for the low value of ky of the L amide provided that such alternately oriented ES complexes were unproductive from the standpoint of hydrolysis. Another possible explanation for the low ky is the one stated previously, i.e., that it is a manifestation of the relatively unstrained ES complex suggested by the low K s° It is possible that a predisposition toward such alternative modes of combination in the case of the Lemethylamide might render it a potential monofunctional inhibitor in a certain fraction of its interactions, The L-methylamide would then be capable of participating in the trimolecular type IES complex noted in the methyl hippurate=indole case. Such behavior would result in an increase in intercept with increasing inhibitor concentration on the Lineweaver= Burk plot and this possibility emphasizes the necessity of a more thorough investigation of the possible influence of the Le-methylamide on the substrate intercept. The results summarized in Table XVI indicate that the "inverse" order of K, and K, observed with the D and L amides also obtains in the case of the K, values of the methylamides. Since the K I I values of the methylamides may be translated into affinities of enzyme for inhibitor the indication is that K, of Necarboethoxy-L-tyrosin- amide is indeed a measure of its affinity for chymotrypsin and, that in this enantiomorphic substrate~inhibitor pair, the substrate is bound more tightly in ES than is the inhibitor in EI. In Table XVI is a summary of the affinity data for the four compounds including the free energies of dissociation of their respective complexes with chymotrypsin. - 63 = TABLE XVI _ Summary of the ES and EI Dissociation Constants of The Enantiomorphic N-Carboethoxytyrosinamides® and N-Carboethoxytyrosinmethylamides” at 25°C Compound Ke Ky AF = -RT ln K Leamide 6.43 + 2987 Deamide 19 + 2345 L-methylamide 9.5 + 2756 Demethylamide 21.8 + 2264 Determined at pH 7.9. Determined at pH 7.6, Averaging the value of Ky obtained against its L antipode as the substrate with the value of K, determined with acetyl- Letyrosinhydroxamide as the substrate. In units of 1073 M. - 64 - A certain degree of independence of relative binding strengths of the members of the DL pairs from Ry is indicated by comparing the differences in AF's of the amides (A(AF) = 642) to A(AF) = 492 for the methylamides. The ratio of the K, value of the L-methylamide I to the K, value of the Leamide is ca. 1.48 while the ratio of the Ky values of the corresponding D isomers is ca. 1.15. This small difference in relative binding strengths is in the wrong direction to suggest any influence of ky on the size of K, of the L~-amide (known to be small anyway) and is possibly a reflection of alternate modes of combination, which as suggested earlier, may be open to the & isomers alone. - 65 = Experimental L~Tyrosine Ethy] Ester - Prepared according to the method of MacAllister? A suspension of 20.0 g. of Letyrosine (Lemke) in 100 ml. of absolute ethanol was saturated with anhydrous hydrogen chloride (Matheson) at 0-5°. The mixture was heated under refluxing conditions for 26 hours and the excess solvent removed by evaporation in vacuo to yield a solid cake of the ester hydrochloride. Neutralization by aqueous potassium carbonate gave the crude ester which was recrystallized once from ethyl acetate to give 11.6 g. of small stout prisms, mp. 101- 102.3° Found by MacAllister, Mepe 105—1 06° (after several recrystall~ izations). L~-Tyrosinamide ~ Prepared according to the method of MacAllister® A solution of 9.8 g. of L-tyrosine ethyl ester in 60 ml. of absolute methanol was saturated with anhydrous ammonia (Dow) at 0-5°, After warming slowly to room temperature the mixture was allowed to stand for 46 hours and the volatile material was removed by evaporation in vacuo. The resulting crude material was recrystallized from methanol- water to yield 3.34 g. of chunky prisms, m.p. 154.31 54.9°. Found by MacAllister, mop. 155—1 56° N-Carboethoxy-L-tyrosinamide - A solution of 3.34 g. of Letyrosinamide in 23 ml. dioxane-water was stirred with a 10 per cent excess of ethyl chlorocarbonate (Eastman) at 0~59 with concurrent addition of aqueous potassium carbonate to react with the liberated hydrogen chloride. & R.V. MacAllister, Ph.D. Thesis, Calif. Inst. of Technology, 1950. - 66 = The crude N-carboethoxy-[.-tyrosinamide was recrystallized from water to give fine white needles, mp. 156.6-1 58.1°, rat (2") , Mp. 155-1 57°, [a]*> = + 1.90 (C * 5 per cent in methanol). Anal, caled. for Cy» Hyg 0, No (252.3) : Cy 57.13 Di-Tyrosine Ethyl Ester - Prepared in the same manner as the 1, ester. Treatment of 100 g. of Di-tyrosine (Dow) with ethanolic-hydrogen chloride yielded a crude crop of ester weighing 65.7 g., mop. 104.91 07.92 The material was used without further purification, N-Carboethoxy-Di-tyrosine Ethyl Ester - To a solution of 100 g. of DL-tyrosine ethyl ester in 500 ml. of pyridine, maintained at 5-1 0°, 58 g. of ethyl chlorocarbonate (Eastman) was added dropwise with mechanical stirring over a period of 50 min, The pyridine was removed by evaporation in vacuo and the resulting syrup, dissolved in 300 ml. of chloroform, extracted in the cold with dilute hydrochloie acid to remove unreacted ester and pyridine. Evaporation of the chloroform yielded 91.0 g. of small clear prisms, mp. 8742=90.3°. Enzymatic Resolution of N-Carboethoxy-DL-tyrosine Ethyl Ester - 50 mg. alpha-chymotrypsin (Armour lot no. 10705) was added to a mechanically stirred solution” of 75 g. of N-carboethoxy-DL-tyrosine ethyl ester in a mixture of 422 ml. methanol and 1445 ml. water, maintained at 30-35°, and containing sufficient phenol red indicator (0.2 per cent Braun) & Some ester actually remained undissolved and had to be separated from the main reaction mixture and treated with additional enzyme. - 67 = to follow visibly the pH of the reaction. As the asymmetric hydrolysis proceeded, 5.46 g. of sodium hydroxide in 336 ml. of aqueous solution was added at a rate which mainteined the pH of the reaction mixture within the range, pH 7-8. From time to time it was necessary to re= place methanol lost by evaporation. After 30 min, an additional 25 mg. of chymotrypsin was added, followed by an occasional small portion of a water suspension of "Viobin", After By hours the reaction was essentially complete and the solution was reduced to ca. 400 ml. by evaporation in vacuo at 30-40°, The reaction mixture was bronght to pH 8.5 and extracted with 4-200 ml. portions of chloroform. Evaporation of the chloroform yielded the syrupy Deester which could not be erystallized, H-Carboethoxy~D-tyrosinamide - A syrupy preparation of N-carboethoxy- D-tyrosine ethyl ester obtained from an enzymatic resolution of 55.0 g. of N-carboethoxy-DL-tyrosine ethyl ester was dissolved in 150 ml. of absolute methanol and the solution saturated with anhydrous ammonia (Dow) at 0-5°, After standing at room temperature for 2 days the volatile material was removed by evaporation in vacuo and the resulting syrup slowly crystallized. Repeated recrystallization from water yielded 7.6 g. of fine white needles, mop. 156.6-158.1°, (a]?° = -19.4. (C = 5 per cent in methanol). Anal, caled. for 6), Hyg 0, Np (252.3) 2 Cy 57.13 ~ 68 = N-Carboethoxy-D~Tyrosinmethylamide = A syrupy preparation of N-carboethoxy- Detyrosine ethyl ester obtained from an enzymatic resolution of 75.0 g. of N-carboethoxy-DL-tyrosine ethyl ester was dissolved in 250 ml. of absolute methanol and cooled to 0-5°, Methylamine (Matheson) was passed into the solution until it had increased to about 400 ml. in total volume and the mixture was allowed to warm slowly to room temperature and stand undisturbed for 70 hours, Upon removal of volatile substances by evaporation, a syrup was obtained which rapidly crystallized in bunches of silky white needles, These were recrystallized twice from water to give 10.1 g. of fine white needles, mp. 1524-15449 | [a]?? = 20.9 (C = 5 per cent in methanol). Anal, caled. for C13 Wig 0, N, (266.3) : C, 58.6; Hy, 6.83 N, 10.5. Found: C, 58.6; H, 6.8; N, 10.4 L-Tyrosinemethylamide - Methylamine (Matheson) was passed into a solution of 40.5 g. of L-tyrosine ethyl ester in 300 ml. of absolute methanol at 0-5° until the total volume had increased to ca. 500 ml., and the resulting solution was allowed to warm slowly to room temperature and stand undisturbed for 4 days. Removal of volatile substances left an unerystallizable syrup which was, however, suitable for conversion to the carboethoxy derivative. N-Carboethoxy-L-Tyrosinmethylamide = The syrupy L-tyrosine methylamide from above was dissolved in 300 ml. of ethyl acetate-acetone and placed in a 1000-ml. Florence flask with a second phase consisting of 300 ml. of saturated aqueous potassium carbonate. The biphasic system was - 69 - cooled to 6° and 40.6 ml. of ethyl chlorocarbonate (Eastman) was added in small portions over a 20 minute period with continuous shaking and cooling. The precipitate which formed was recrystallized repeatedly from water to give 7.0 g. of white needles, m.p. 151.5-1 53.5°, [a}* = #19.2 (GC = 5 per cent in methanol), Anal, caled. for C,, Hyg 0, Np (266.3) +: Cy 58.6; Acetyl~L-Tyrosinhydroxamide - This material was prepared by R.R. Jennings. It was recrystallized once from isopropanol and the resulting fine white solid washed repeatedly with cold isopropanol and tried in vacuo. [a)2° = +39,0° (C = 5 per cent in water). rit"), [0]*? = +38,3° (C = 5 per cent in water). Enzyme Solutions - Two crystalline preparations of alpha—chymotrypsin were used in these studies. In all experiments utilizing N-carboethoxy= L-tyrosinamide as the substrate at pH 7.9, including those in the presence of its 2D antipode, the enzyme was an Armour preparation, lot no. 90402, containing 10.4 per cent protein=nitrogen, In the pH optimum study, in the hydrolyses of N-carboethoxy-L-tyrosinamide at pH $.10-8.25, and in the evaluation of all inhibitors against acetyl- L-tyrosinhydroxamide, the preparation was likewise an Armour one, lot no. 10705, containing 13.3 per cent protein-nitrogen. Both preparations were of bovine origin and contained added magnesium sulfate. Aqueous stock solutions of enzyme were prepared daily and were kept in the refrigerator between the intervals in which they were brought to 25° prior to introduction into the substrate reaction mixtures. Buffer Solutions - All hydrolysis experiments were buffered by the tris- (hydroxymethy])-aminomethane<hydrochloric acid buffer described by Iselin and Wiemann$72) The amine component of the buffer was recrystallized three times from water and made up to buffer solutions of the desired pH and concentration by titration with solutions of hydrochloric acid of the appropriate concentration. These adjustments of pH were followed by means of a Beckman pH meter. Formaldehyde = Merck reagent grade formaldehyde (36=38 per cent in aqueous solution) adjusted to pH 7-8 was used in the titrimetrically analyzed experiments. Ferric Chloride Solution - The ferric chloride solution used for the spectrophotometric determination of acetyl-L-tyrosinhydroxamide was prepared as follows. 54.0 g. of reagent ferric chloride hexahydrate was dissolved in 500 ml. of water and 16.3 ml. of concentrated hydrochloric acid (assay 37.5 ver cent HCl) added. This mixture was then made up to 1000 ml. with absolute methanol. The resulting solution was turbid when first prepared but became clear after filtering through a Whatman No. 1 paper and standing about 2 days. Enzyme Experiments - All enzymatic hydrolyses were conducted at 25° in aqueous solutions containing 4 tris~(hydroxymethy1)-aminomethane= hydrochloric acid buffer in the manner described by Huang and Wiemann{?0) The titrimetric method of analysis used with the runs involving N-carboethoxy-L-tyrosinamide was identical with that described by Huang and Niemann“! 0) and earlier Iselin and Niemann §22) -~71- In the experiments involving acetyl-L-tyrosinhydroxamide as the specific substrate the hydrolyses were followed by an extension of the method of Hogness and Niemann{'9) In the present experiments, ‘the abserbance, at any time +t, of unhydrolyzed acetyl-[-tyrosinhydrox- amide, as its ferric complex, was measured at a wavelength of 515m ua with a Beckman Model B Spectrophotometer. From the reaction mixtures buffered at pH 7.6 by tris-(hydroxymethyl )-aminomethane-hydrochloric acid, 0.27 F with respect to the amine component, 1l-ml. aliquots were pipetted out and introduced at the chosen time intervals into 10-ml, G.S. volumetric flasks containing 7.5 ml. of absolute methanol and 1 ml. of the ferric chloride solution, stoppered and mixed. The absorbance of the colored complex was then measured with the spectrophotometer set at 100 per cent transmission against a blank, prepared in the same manner as the test samples, but containing, instead of 1 ml. of reaction mixture, 1 ml. of a blank solution consisting of enzyme, buffer and water. In the case of determinations made in the range [S] 9 715% 107? M te [s], = 40x 107? M it was necessary to increase the dilution of the colored ferric complex to enable its spectrophotometric measurement, This was achieved by increasing the 10-ml, sample scale to 25 ml., intro- ducing a reaction aliquot into a mixture of 1 ml. of the ferric chloride solution and 22.5 ml. methanol, contained in a 25-ml. G.S. volumetric flask, and adding methanol to 25 ml, (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) - 72 - REFERENCES Neurath H., and Schwert, G.W., Chem. Revs. 46, 69 (1950). Shine, H.J., and Niemann, C., JeA.C.S. Thy 97 (1952). Huang, HeT., and Niemann, C., JoA.C.S. 75, 1395 (1953). Haldane, J.B.S., "Enzymes", Longmans, Green and Company, London New York, Toronto, 1930. Thomas, D.W., MacAllister, R.V., and Niemann, C., J.A.C.S. 735 1548 (1951). Huang, H.T., MacAllister, ReVey Thomas, DeW., and Niemann, Coy TeAeCoSe 73, 3231 (1951). Huang, H.T., and Niemann, C., J.A.C.S. 73,1541 (1951). Huang, H.T., and Niemann, Cry JoAcCeSe Thy 4713 (1952). Michaelis, Le, and Menten, M.Le, Biochem, Zeit. 49, 333 (1913). Jennings, R.R., and Niemann, C., JeACeS. 75, 4687 (1953). Huang, HoI., and Niemann, C., J.A.C.S. 745 4634 (1952). Suggested by Prof. Niemann, Huang, H.T., and Niemann, C., J.A.C.S. 74, 101 (1952). Foster, ReJ., and Niemann, C., Proc. NeA.Se 39, Now 10, 999 (1953). Jennings, RoR., and Niemann, C., to be published, Hogness, D.S., and Niemann, C., J.A.C.A. 75, 884 (1953). Augustinssen, K., Acta Physiol. Scand, 15, suppl. 52 (1948). Koenigs, E., and Mylo, B., Ber. 41, 4427 (1908). Iselin, B.M., and Niemann, C., Jour. Biol, Chem. 182, 821 (1950). - %3- SECTION II EXPERIMENTS ON THE OXIDATION OF 3-INDOLEACETIC ACID BY PLANT ENZYMES - = Introduction That 3-indoleacetic acid occurs widely in plants where it functions as the chief growth regulating hormone (auxin) has been established by a long series of investigations’'~>), This substance is biosynthesized in the apical meristems and young leaves of plants, from which it is translocated downward to control a wide variety of plant physiological responses, the most studied of which is that of cell elongation, 3-indoleacetic acid is formed in the plant from tryptophane, a fact indicated by the existence in plants of an enzyme system capable of this conversion and apparently distributed in such a way that its highest concentrations are found in those tissues known to synthesize and export auxin, In addition to this synthetic potentiality another enzyme system, which functions in the oxidative destruction of IAA, has been shown to be present in a number of plant tissues. The level of TAA within a given plant tissue may be determined consequently by the simultaneous operation of two enzyme systems, one capable of synthesizing this auxin and the other responsible for its oxidative degradation into inactive products. Since the typical physiological and morphological behavior of a growing plant are believed to depend upon the distribution of auxin among its various organs, this behavior may be related to a regulatory balance between the functioning of these two enzyme systems. For example, while IAA promotes the elongation of stem cells in the area immediately below the meristematic area, the same or somewhat higher concentrations are known to inhibit the elongation of root celis(®) Tang and Bonner ‘7) have suggested that the relatively high concentration of the IAA-oxidase system in the roots of many species may be related to the necessity of preventing an excessive accumlation of growth substance in this organ. The suggestion that the function of TAA-oxidase is to keep in check the IAA level produced by the synthetic system has been advanced by other vorkera‘®) Beside occurring widely in roots, the IAA-oxidase system is known to occur in the etiolated tissues of many seedlings. Tang and Bonner ‘?) found breis of etiolated pea epicotyls to be an excellent practical source of this enzyme system. They showed the inactivation process to be an oxidative decarboxylation in which one molecule of oxygen is apparently taken up and one molecule of carbon dioxide liberated for every molecule of IAA inactivated (Equation (1)) (1) o, + Tan SSh oxidase. oxidized IAA + CO, Wagenimecht and purris(8) report the same stoichiometry for the des=_ truction of TAAby the oxidase present in root homogenates of the yellow wax bean. They also found that 3-indolepropionic acid and 3-indolebutyric acid undergo oxidation in the presence of this enzyme but at rates much lower than that of IAA. Galston, Bonner, and Baker ‘| 0) have presented evidence that the IAA-oxidase system consists of a peroxidase coupled to a hydrogen peroxide generating flavoprotein in the following manner (2a) Red. flavoprotein + 0, —---> Oxid. flavoprotein + H,0 22 BR. unknown _” red.agent roxidage (2b) HO, + Ian PRESRIOSSS_» oxid. TAA 460, + HA0 This oxidation 1s accelerated by ight'!") by optimal amounts of manganous ton{®) and by certain phenolic type co-factors, particularly 2,4~dichlorophenol! 2) A somewhat similar enzyme system studied by Kenten and Mann‘'?"'5) snatcates a possible partial explanation for the actions of the two latter factors. The hydrogen peroxide-peroxidase component of their system was capable of oxidizing manganous ion via coupled oxidations of phenolic type co-factors. The high valency (+ 3, or + 4) manganese ion resulting is presumably a potential agent for further substrate oxidation and could be one of the responsible agents in the breakdown of IAA. Recently Siegel and Galston‘!©) have produced evidence that light, 2,4-dichlorophenol, and manganous ion enhance the peroxide producing capacity of pea tissue and also show that IAA itself is a peroxygenic agent. These agents can, therefore, boost the overall oxidation by intervening at an earlier stage of the oxidation chain as well. While the general flavoproteinperoxidase nature of the TAA-oxidase system seems fairly certain, the actual chemical trans=- formations involved in the decarboxylative oxidation are obscure. Nothing definite has been disclosed regarding the fate of the IAA carbon skeleton. Tang and Bonner ‘9? have stated that the indole nucleus remains intact as evidenced by the application of the Hopkins-Cole test to the reaction products and furthermore that it is a neutral substance capable of extraction from the reaction mixture by ether at neutral or alkaline pH's. Wagenknecht and purris(8) arrived at the same conclusions, basing their decision regarding the indole ring on the observation that ether extracts of the reaction mixture gave a red color with Ehrlich's reagent. They also state that the product is a carbonyl compound as evidenced by its formation of a 2,4-dinitrophenylhydrazone derivative. Neither the method of preparation nor the physical properties of this derivative are described by these authors. This evidence, particularly the observed stoichiometry of the degradation prompted both of these groups to postulate the formation of 3-indolealdehyde as the oxidation product. IAA CHCO2H OXIDASE CHO (3) +o, Snes Cw ° 0 4 N + COr Wiltehire''7) agreed with this hypothesis on the basis of spectral evidence. This compound, resulting from exclusive attack on the side chain, is known, moreover, to exhibit no growth promoting action. An alternate possibility is initial oxidative attack upon the indole nucleus. This hypothesis derives support from two comparable instances of peroxidase-catalyzed tryptophane oxidation, Knox and Mehier ‘'8) observed that a peroxidase-containing enzyme system of liver was capable of opening the indole ring of L-tryptophane to yield formyl kynurenine fe] LIVE rT CHa- CH-CO2H PEROXIOASE 6-CHa- CH-COaH (4) | Nie + 03, Nu y NHCHO + COr Like the IAA-oxidase of pea seedlings $?) this oxidase is an adaptive enzyme capable of increasing in response to added substrate, Wiltshire ('7) has reported the oxidation of tryptophane by. a pea seedling enzyme which is identical, apparently, with the IAA-oxidase preparation used in the present investigation. One of his products was tentatively identified as 3-hydroxykynurenine apparently resulting from oxidation of both pyrrol and benzene nuclei. These reports, in combination with the above mentioned stoichiometry, led us to hypothesize the following tentative reaction sequence for IAA oxidation 9 CHy-CO2.H — otis se. Bt oH en CH (5) NHCHO The o-formamidoacetophenone (OFA) resulting from muclear oxidation might appear in the reaction mixture or it might be deformylated in the presence of a formylase in a manner parallel to that observed by Knox and Mehler‘'?) with formylkynurenine. Alternatively, the OFA could undergo ring-closure to eee ‘-hvdronrauinchine’. oc _ 10 In the hope of gaining information regarding the nature of the oxidation product (or products) it was decided to prepare a number of these proposed products and compare them with the actual products obtained from the oxidation of IAA by preparations of pea- epicotyl oxidase. aA suggestion by Dr. D.G. Crosby of the Carbide and Carbon Chemicals Company. Materials and Methods Plant Material and Crude Enzyme = Alaska peas were planted in flats of moist vermiculite and allowed to germinate and grow in a darkroom maintained at 25 + 1° for a period of 7-9 days. The yellowish-white epicotyls were harvested, by cutting approximately one inch above the ground line, the necessary illumination being supplied by a wak green light? ‘The epicotyls were cut into small sections and ground with Sorensen's phosphate buffer, pH 6.6, in a Waring blendor at 0~2°%. It was found convenient to use ca, 60=100ml. of the buffer solution per flat of peas planted in this operation, The resulting brei was pressed through cheesecloth to remove the larger particles and the filtrate centrifuged in the cold for 30 minutes at 10,000 x g. The milky supernatant was used as the crude enzyme preparation and was stored in a deep-freeze until such time as it was fractionated and employed in the oxidation experiments. Fractionation of the Enzyme = In a typical fractionation 138 ml. of the crude enzyme preparation was carefully thawed and centrifuged in the cold for thirty minutes at 10,000 x g, discarding the sediment, To the supernatant, maintained at 0-5°C, was added solid reagent grade ammonium sulfate to effect the precipitation of protein fractions of decreasing average molecular weight. These fractions were allowed to precipitate over periods of ca. 5—12 hours at 0-2° after which they were collected by centrifugation at 10,000 x g for 20=30 minutes. They were then resuspended, each in 50 ml. of phosphate buffer, pH 6.6, a Photomorphogenically inactive. and stored in the refrigerator at 0-2°C, such preparations showed no deterioration in the cold over a period of at least two weeks. The IAA-oxidase activity of each fraction was determined by . incubation of a l«ml,. aliquot with 3—indoleacetic acid (Eastman) at an initial concentration of 130 ¥/ml. for an interval of ca. 65 minutes after which the remaining IAA was determined by the colorimetric method with Salkowski's reagent described by Tang and Bonner ‘?), In every case the reaction mixture contained 2,4-dichlorophenol as added co-factor at a concentration of 107° m{!2) Table I summarizes the activity data for a typical series of enzyme fractions. In all cases the highest activity was found associated with the final supernatant remaining after saturation with ammonium sulfate. Although this was somewhat surprising in view of the known molecular weight of 44,000 of horseradish peroxidase 2°), it is in agreement with the observation of Wiltshire‘'”) that 37 per cent, or the major portion, of the tryptophane oxidase activity of pea= seedling brei remained in the final supernatant, 2/3 saturated with respect to ammonium sulfate. Oxidation Experiments - Although no careful study was made to compare the effects of the different enzyme fractions on IAA, there was no indication that different fractions of the enzyme led to different oxidation products? In the oxidation experiments, therefore, either whole unfrozen enzyme, centrifuged 30 mirmtes at 10,000 x g to remove inactive fragments, or various combinations of the salt fractions were used. & A difference was actually noted by Wiltshire ('7), TABLE T Ammonium Sulfate Fractionation of IAA-oxidase from Breis of Etiolated Pea Epicotyls. Frac. Degree of Total Reac. Colorin, TAA No. “gatn.@s> galt Added ‘Mme Read.® Destroyed” grams min. Yul. 1 0-00.33 32.1 66 135 51.5 3 0.500. 58, 57.1 66 87 79.0 4 058 .-0.65 63.3 655 119 60.0 5 0.65-0.75, 13.3 65 137 50.0 Final. rantl 0.75, 73.3 65 70 89.0 a These figures are with respect to pure water at oc from the known solubility of ammonium sulfate. Actually the degrees of saturation are higher than above since the solubility of the salt is decreased by the presence of the protein. Total volume of original enzyme solution = 138 ml. Method of Tang and Bonner?) It must be understood that the activities represented by these figures are total activities and do not represent enzymic activity per unit protein-nitrogen. ~ 82 = In a typical oxidation, 50-100 ml. of an aqueous solution®™ _of IAA at an initial concentration of 500 Y/ml. was mixed with an equal volume if the enzyme solution in a 500-ml. Erlenmeyer flask and the mixture incubated at room temperature with gentle” shaking on an Eberbach shaker. In most instances the reaction mixtures contained 2,4-dichlorophenol at a concentration of 107 (12) The oxidations were allowed to proceed to about 75 per cent completion, as determined by colorimetric assay of withdrawn aliquots for 1aa‘?), The time intervals required for this amount of reaction varied from ca. 30 to ca. 300 minutes. At the end of the incubation, such a reaction was stopped by bringing it to pH 3 or pH 9, by the addition of catalase’, or by simply extracting the unreacted IAA and oxidation products away from the enzyme with reagent grade chloroform. Extraction and Chromatography of the Oxidation Products — Reagent grade chloroform proved to be the best solvent for extraction of the oxidation products, Usually it was necessary to apply low speed (1500-2000 RPM) centrifugation to cause separation of the chloroform phases from the protein containing aqueous portions following equilibration in a separatory funnel, Contrary to the reports of previous workers (8+), ether did not appear to be a satisfactory solvent for the oxidation preducts. The combined chloroform extracts from several equilibrations were partially dried by chilling to O-5°C. and filtering through a ® Containing a stoichiometric equivalent of sodium bicarbonate to dissolve the IAA. To avoid foaming and consequent denaturation of the enzyme, Armour (No. 30) Catalase was used. - 83 = chloroform wet filter paper and concentrated by evaporation to a few ml. volume in vacuo. Chromatographic separation of the oxidation products was performed upon ascending chromatograms of Whatman No. 1 filter paper. The most satisfactory procedure was the following one. Rectangles of the paper, 29 cm. x 33 cme, were cut and a pencil line drawn parallel to the long dimension and ca. 3/4" from one edge. The oxidation product concentrates were spotted at intervals of a few inches along the line, each spot being made by repeated applications of the chloroform concentrate from a glass capillary, 0.5~1.0 mm, in diameter. Care was exercized to keep the size of the spots as small as possible. Synthetic compounds of known composition were similarly applied at different points along the line. It was found convenient to hasten the evaporation of solvent between each capillary application by means of a gentle compressed<-air stream. A chromatogram thus prepared was wound into a cylinder, 29 em. in length, and fastened at each end and at the central junction by stapling. Such a cylindrical chromatogram could be con- veniently contained in a cylindrical glass developing tank, 15 cm. in diameter and 30 cm. high. The prepared chromatogram was then suspended over the developing solvent contained in the bottom of the tank. A rigid iron wire, bisecting the circular section of the chromatogram near the top, was caused to hold the paper cylinder firmly against the lower side of a glass plate, covering the open upper end of the tank, by means of a permanent magnet placed upon its upper surface. After the paper had been allowed to equilibrate in the solvent atmosphere for periods varying from 90 to 180 minutes, the magnet was removed, causing the paper cylinder to drop gently into the developer. The chromatogram was allowed to remain in the developer until the solvent front had nearly reached the upper edge of the paper and was then removed and hung in the air to dry. The dried developed paper was then sprayed with an appropriate spray reagent and given a brief heat treatment to accelerate the development of the colored spots. Developing Sovents = Two mixtures containing isopropanol, concentrated aqueous ammonia, and water were used, With a 1021:1 volume ratio(2") the solvent front moved rapidly, nearing the upper end of the cylinder in 6 hours. A 2:1:1 volume ratio, however, proved superior for the separation of the oxidation products and was used in the development of the chromatograms discussed in the next section. This developer moved slowly, 15 hours being a common time for development, Spray Reagents - Two spray reagents were used. A solution of Ehrlich! s reagent (22) (p~dimethylaminobenzaldehyde), 1 per cent in 1N hydrochloric acid was found to be an effective color reagent for indole-type com pounds and aromatic amines. A solution of ferric chloride, 1 per cent in water, was used as a spot reagent for 4-hydroxyquinoline and was later found to give color reactions with some of the other compounds involved. Hopkins-Cole Reagent - This reagent§7? ) containing magnesium glyoxylate as the active agent, was prepared according to Benedict (74) - 85 - The test for indolic compounds is conducted as follows (*?) To a small amount of the material to be tested is added 1 ml. of the reagent plus 0.2 ml. of 0.01 M aqueous cupric sulfate and the constituents well . Mixed. 5 ml. of concentrated sulfuric acid is then added cautiously and finally mixed thoroughly with the original solution. The formation of a red to blue-violet color is characteristic of many compounds containing the indole mcleus,. Chromatography and Preparation of The Oxidation Products for U,V. Spectral Analysis - The entire concentrated chloroform extract of the particular fraction to be analyzed was applied in a strip along the bottom of a cylindrical paper chromatogram, the sides of which were graduated in centimeter-unit scales. Such a chromatogram was then developed and dried in the usual manner, It was then opened, placed flat upon a flat surface and parallel pencil lines drawn connecting the opposite corresponding centimeter marks. The horizontal zones containing the oxidation products and unreacted IAA were located by cutting two thin vertical sample strips from near the ends of a chromatogram and spraying them with Ehrlich's reagent. The entire chromatogram was then cut horizontally into the 19cm, wide strips defined by the pencil lines, each strip being labelled with the number corresponding to the distance in centimeters of its upper edge from the bottom of the chromatogram. Each strip was then cut into small fragments and these washed with 3 ml. of reagent chloroform in a labelled test tube to elute any soluble material present. 0.1 ml. of each chloroform eluate was added to 5 ml. of absolute methanol and the resulting solution analyzed with a Beckman U.V. Spectrophotometer, -_ 86 = 3-Indoleacetic Acid -— Bastman No.4271 was used without further purification. 3~Indolealdehyde - A sample donated by Mr. R.A. Rernhard was employed. It was recrystallized twice from water to give white blades, insoluble in aqueous soidum hydroxide, mp. 193.5~195°, rit!) MePeo 194°, o-Formamidoacetophenone =- Prepared by the method of Witkop solution of 4.3 g. of C.P. skatole (Eimer and Amend) in 50 ml. of formamide, contained in a 250-m1l. round bottom flask, was chilled to 4°c by a beaker of water surrounded by an ice bath. A stream of oxygen- diluted ozone® was passed into the solution for a period of nine hours at which time the exit gas contained considerable ozone as evidenced by its ability to liberate iodine from an aqueous boric acid = sodium iodide solution, The dark solution was extracted with four 50-ml. portions of ethyl ether which were then combined and washed with water. The yellow ether solution was dried over anhydrous potassium carbonate and the ether removed by evaporation, The total weight of the crude product thus obtained was 3,11 g. Recrystallization of it from ether containing a@ smell amount of ligroin gave sparkling prisms. m.p. 18.9-79.5°. rit(?”) mip, 78°, e-fminoacetophenone = 0.25 g. of crude o-formamidoacetophenone was heated under refluxing conditions with 20 ml. of 3N hydrochloric acid for 45 minutes. Upon cooling, the mixture was neutralized by the addition of 5-6 g. of sodium bicarbonate which caused the separation of o~amine= . acetophenone as a dark oil. The oil was co-distilled from the neutral 2 I am indebted to Professor A.J. Haagen-Smit for the use of his ozone generator. - 87 =» mixture with water and separated from the latter by ether extraction. Removal of ether upon the steam bath left the ceaminoacetophenone as a yellow, pleasant smelling oil. 4~-Hydroxyquinoline ~ Prepared according to the procedure of camps(7®) 1.0 g. of o-formamidoacetophenone was dissolved in a boiling mixture of 71,5 ml. of water and 6.4 ml. of 95 per cent ethanol, To this was added a solution of 0.43 g. of sodium hydroxide in 4.0 ml. of water and the yellow mixture heated under refluxing conditions for 3 hours. O-aminoacetophenone, liberated as a hydrolysis by-product, was removed by repeated co-distillation with water and the volume of the reaction mixture was reduced to 10 mle, allowed to cool, and filtered to remove the dark side-product, o~flavaniline (bimolecular condensation). 6N hydrochloric acid was added to the clear filtrate to neutrality and the precipitate which formed was collected by suction filtration, Gradual evaporation of the filtrate caused the formation of beautiful flat prismatic crystals of 4—hydroxyquinoline which were combined with the original precipitate, ground to a powler, and washed with ether to remove traces of o~aminoacetophenone. MePe 198~205°, rit(?8) MePe 201°, Color with ferric chloride in aqueous solution, deep red at high con- centrations, orange in dilute solution, pit (78) Color, Carmine. ~ 88 = Experimental Results Color Reactions of Known Compounds It was hoped that by means of parallel chromatography and the use of color reagents, the oxidation product (or products) would be identified among the test compounds employed. For this purpose, three color reagents were used, The Ehriich's reagent and the ferric chloride solution®™ were sprayed separately on dried paper chromatograms, while the Hopkins~ Cole tests” were carried out in test tubes, Table II summarizes the average Rp values and colors given by the various known compounds with the different reagents. The observa- tion that Hopkins-Gole reagent gives a pale yellowish color with 3— indolealdehyde indicates that this compound alone could not have accounted for the purple color observed by Tang and Bonner ‘2) With the other indolic substances tested, this reagent gave various shades of red and purple. Apparently the presence of the indole nucleus is a necessary, but not sufficient, condition for an intense color reaction with this reagent, Table II indicates that the non=<indolic compounds were negative to this test. With respect to Ehrlich's reagent, the compounds tested fall into two general categories. Compounds containing the indole mucleus gave colors ranging from pink through lavender to blue-violet and required a definite development time characteristic of each substance, IAA, indole, and skatole gave their respective colors within about See section on materials and methods. - 8 - TABLE IT Rp Values and Color Reactions of Various Compounds Related to The Enaymic Oxidation of IAA Color With Substance Rp 1% gb H-Cole. Ehrlich FeCl, TAA 0.35 0.79 violet violet in reddish ca. 5 min. (slow) 3-indole~ pale violet inca. reddish aldehyde 0.90 0.91 yellow 10-15 min, (very slow) OFA 0.87 0.92 none immediate none orange= yellow OAA 0.89 0.93 none immediate none orange= yellow 4-hydroxy immediate quinoline 0.83 none orange skatole 0.87 bl-violet in 5 min, indole 0.92 pink in ca, 5 min, kynurenine® yellow anthrenilic immediate pale yellow acid orange= (slow) yellow * 102121 developer. © 28131 developer. © Becomes bluish on standing. d (18) From Knox and Mehler, - 90 - 4-6 minutes of drying with gentle heat while 3-indolealdehyde was much ‘Slower to appear, requiring 10-12 mimtes to give a perceptible violet, Compounds containing an aromatic amino group give an immediate color with Ehrlich's reagent due to Schiff base formation. ‘Thus o-amino= acetophenone (QAA) and o-formamidoacetophenone (OFA) both gave immediate intense orange-yellow coloration persisting indefinitely in the case of OAA and changing slowly to blue in the case of OFA, It seems probable that the immediate yellow given by OFA results from cleavage of the formyl group by the acid present to give OAA and that the blue may be characteristic of OFA itself, Knox and venter (18) found kynurenine to give a similar yellow color with Ehrlich's reagent. Hopkins-Cole Test upon the Oxidation Products. It is not clear from the report of Tang and Bonner ‘'9) whether they carried out their Hopkins-Cole reactions upon aliquots of reaction mixture or upon ether extracts of the same. Since their results in this text are compared with the Salkowski assays, however, and the latter are performed directly upon the reaction mixture it is presumed that their Hopkins-Cole tests were performed in the same way. In the present experiments Hopkins-Cole tests were carried out on both a reaction mixture and upon a chloroform extract of the same at pH 9. Blank determinations upon an enzyme solution which had been treated as in an oxidation run but containing no IAA were also performed. The results of these tests are shown in Table III. When performed directly upon a reaction mixture, the Hopkins-Cole test gave a pinkish=-brown color regardless of the presence or absence of IAA or of its oxidation products. -91 - TABLE III Hopkins-Cole Behavior of Various Components of an Oxidation Reaction Mixture, Tested Result Directly on reaction mixture” pinkish-brown Directly on enzyme preparation pinkish-brown Upon a chloroform extract of a reaction mixture at pH 9 purple Upon a chloroform extract oft the enzyme preparation pale-yellow Upon a chloroform extract of an TAA solution at pH 9 nothing & 407? w in 2,4~dichlorophenoi \!2) The preparation had been incubated in the manner of an oxidation run. ~ 92 - This color probably results from the high "blank" of tryptophane-contain- ing proteins in the enzyme preparation. Table III indicates that the . oxidation reaction mixture contained a substance, presumably indolic which could be extracted into chloroform™ and gave a Hopkins-Cole color.. This was shown to bermsither present in nor formed from the enzyme preparation. That this was not unreacted IAA was indicated by a negative Hopkins- Cole test given by a chloroform extract of an IAA solution at a comparable concentration and pH 9. Chromatographic Separation of the Oxidation Products Most of the knowledge regarding the nature of the oxidation products was gained by paper chromatography of the chloroform extracts of oxidation reaction mixtures. A large number of paper chromatograms from the material of eight oxidation experiments indicated the presence of at least two principal oxidation products. These appeared, before any spray treatment, as yellowish spots having average Ra values of 0.94 and 0.87, respectively.” Upon spraying with Ehrlich's reagent both spots were immediately colored orange-yellow, resembling OFA and OAA in color and behavior, Upon standing, however, both spots, unlike either OFA or OAA, began to develop reddish tinges, the spot at Re 0.94 becoming pinkish-orange and the lower one dark pinkish-orange to light red. These colors were dependent upon the concentrations of the oxidation products and deepened toward red with increasing concentration. That this color ripening did not result from a superposition of an aromatic amine TAA was indicate the presence of JAA as a clearly defined a It was found necessary to remove the chloroform prior to the test as it gives an interfering dark color, b These values varied between chromatograms. This variation seems to be due in part to time of equilibration before development and in part to other factors. The composition of the developer had little effect on these Ry’ Se - 93 = violet spot below the oxidation products, Treatment with the ferric chloride spray revealed both products as rose colored spots, which appeared within three to five minutes and intensified with time, particularly the Rp 0.94 spot, which became deep rose upon standing. This is unlike the behavior of 4=hydroxy= quinoline which gives an immediate and unchanging orange spot with ferric chloride. This and the fact that 4-hydroxyquinoline gives no color with Ehrlich's reagent eliminate it as an agent responsible for the observed spots. Beside failing to give Ehrlich tests that correspond to those of the oxidation products, 3~indolealdehyde, OFA, and OAA behave differently with ferric chloride as well, OFA and OAA give no color with this reagent even on standing and 3—indolealdehyde gives only a faint reddish spot on standing ten to twenty minutes (Table II). The Accumulation of Oxidation Products with Time To confirm the identity of the above spots as authentic reaction products, and to detect any possible change in the nature of these products or in the approximate ratios of their concentrations with time, a run was performed in which one-third of the reaction mixture was with- drawn at each of three successive time intervals and extracted with chloroform to remove the oxidation products. The results of a paper chromatogram of these chloroform extracts are indicated in Table IV. With each aliquot, three spots appeared upon treatment with Ehrlich spray, An upper pair, at RE values of 0.94 and 0.91, appeared immediately as yellow spots which ripened slowly to orange in the case of Rp 0.94 & the extractions were performed upon the reaction mixture portions without any prior adjustment of pH. The natural pH of the reaction mixtures was close to 7. -994< TABLE IV Aecumilation of The Oxidation Products with Time* Color devel, Order of color inten. I Spot Re with Ehrlich of time aliquot spots TAA 0.80 violet in 5 min. Ig5 ‘>> I, 00? Log Ox. prod. I 0.91 yellow > red 165 < Loo << tong OX pred, II 0.94 yellow + orange Tos < Lao < Tong a Time in minutes. - 95 « and bright red at Re 0.91. Unreacted TAA appeared gradually as a violet spot at Re 0.80. It was clear that while the intensities of the TAA spots decreased regularly with time, the intensities of the two upper apots, and in particular the one at Rp 0.91, increased in the game order. The same evidence was obtained by spraying with ferric chloride. This indicates that as IAA is destroyed at least two oxidation products accumulate in the reaction mixture. Acidic or Basic Natures of the Oxidation Products. In the hope of gaining information about the natures of the oxidation products from their pH=-solubility relationships, an oxidation run was performed following which one-half of the reaction mixture was adjusted to pH 3 and the other half to pH 9, and both ex= tracted in identical manners with chloroform. The first indication of a difference in behavior of the oxidation products was given by the colors of the two chloroform extracts. The extract from aqueous phase at pH 3 was bright yellow and was found to contain a dark yellow-brown oily pigment upon evaporation of the chloroform. The extract from pH 9 solution was virtually colorless and contained a small amount of a yellowish-white solid. Table V summarizes the chromatographic data of this experiment. The chromatogram of the pH 3 extract showed, when sprayed with Ehrlich's reagent, two spots. At Rp 0.93 an immediate orange color appeared which ripened to red in twenty-four hours, At Ry 0.82 an immediate carmine spot appeared which faded over a twenty- TABLE V pH-Solubility Behavior of The Oxidation Products Spot Rp Ext. at pH 3 Ext. at pH 9 Ehr. color FeCl Bhr,. color FeCl colot colof violet in pale none none TAA 0-74 = 510 ming — violet (v.slow) Ox. prod. I 0.82 carmine none none none Ox. prod. II 0.93 orange-red rose-pink orange-red vivid rose -97 - four hour period, ‘IAA appeared below at R, 0.74 as a violet spot. Ehrlich treatment of the chromatographed extract from pH 9 solution yielded only the spot at Rp 0.93 which vas practically identical with the upper spot on the other chromatogram. Only the compound responsible for the spots at Rp 0.93 gave a color with ferric chloride, and, of these, the roge color of the spot from the pH 9 extraction was consider- ably more intense than that of the corresponding spot on the pH 3 chromatogram, To check the possibility that any of these spots arose from the enzyme preparation, enzyme blanks were incubated in the usual manner, made pH 3 and pH 9, extracted as before and the concentrated extracts chromatographed. Neither Ehrlich's reagent nor ferric chloride indicated the presence of any interfering compounds in the enzyme preparation, | There is a possibility that the product causing the lower spot (Ry 0.82) may not be identical with the oxidation product observed at the lower R's of pairs seen on previous chromatograms, This is suggested by its Rp value of 0.82 which is considerably lower than the values of ca. 0.91 usually observed, The apparently neutral compound appearing at Rp 0.93 may actually, then, consist of two unresolved spots that have appeared as a resolved pair on some of the previous chromatograms. OFA and 3-Indolealdehyde as Possible Oxidation Intermediates Although the observations described in this section cannot be explained by the presence of either OFA or 3=indolealdehyde as such in the oxidation reaction mixtures, the possibility remained that either ~ 98 = or both of these compounds are initially formed from IAA and then con- verted into other final products, Thus, while neither of these substances _ may be the end product of the reaction, their occurrence as intermediates would be in accord with the observed stoichiometry provided that the remaining steps in the overall reaction did not involve 05 ~CO, gas exchange, To test this possibility, OFA and 3—indolealdehyde were incubated with preparations of IAA~oxidase in separate experiments, The conditions employed were the same as previously described. In each experiment the concentration of the hypothetical intermediate was comparable to the initial concentrations of IAA used in the other runs, and in each case chloroform extraction was performed at the natural pH of the reaction mixture following the incubation period. The Ehrlich's reagent and ferric chloride treatment of the chromatograms made from the chloroform extracts indicated clearly that neither OFA nor 3-indole=- aldehyde is an intermediate responsible for the observed oxidation products. UeVe Spectra of The Oxidation Products To ascertain the possible presence of oxidation products having R's below ca. 0.75 and not responding to either of the spray reagents employed, the optical densities of the methanol solutions pre=- pared from the paper strips (see page 85) cut from this region were determined at wavelengths of 220 mM, 250 mu, and 280 mu. In Figure 1 the optical densities at these three wavelengths are plotted versus the R, values corresponding to the sample strips. It is evidept a A suggestion of Prof. HK. Mitchell, -~ 99 = that below the Re region responsible for the color reactions observed the amount of material absorbing at these chosen wavelengths is negligible, In Fig. 2 are shown the absorption spectra of the two oxidation products extracted from the reaction mixture at pH 3 and of the product extracted at pH 9. The product of Rp 0.93 extracted at pH 9 possesses maxima at 224 m% and 280 mA and a minimum at 255 mu, In contrast, the material responsible for the Rp 0.93 spot extracted at pH 3 lacks the 255 mt minimum and the maximum at 280 mu. This difference is probably due to the presence of a second product, in one of the Rp 0.93 fractions, whose spectrum is superposed upon that of a common product. Of course, this spectral modification could also result from the presence of a compound (such as a phenolic co-factor) in the original brei possessing a higher chloroform solubility at the lower pH. This observation is, however, more in agreement with the possibility stated earlier that the spot appearing at Rp 0.93 represents two unresolved oxidation products. The U.V. spectrum of the acidic product (Ry 0.82) appears to have only the one maximum at ca, 218 ms and a point of inflection near 270 mw . It is possible, however, that the spectrum at this Re is due, in part, to the presence of some JAA. Although the spectrum of the material extracted at pH 9 (R, 0.93) resembles that of the added co-factor, 2,4-dichlorophenol, the concentration of this substance which might be present in the final methanol solution (at the most, 3 x 1079 M) is insufficient to account for the observed optical densities. DEWNS/TY OPTICAL 30 25 .20} AS -/0} 05, = 10 - Fig. le Distribution of Material on a Chromatogram According to Absorption at Three Chosen Wavelengths. "VVI Jo syonpory uoTyEPTXO ey} JO eWlog Jo eAqoedg uoPAdsosqy “A *N °@ ° BT ALISNIG TWI/LdIO (ww) ¥ OSE OEE OVE 062 OL2 Os2 oeZ Olé, ~— 20 — +0 § 5 90 ' \, \ 80 Ne o7 \ (q0823xe OFPTOY) \ 31 ze*o = “y qy-—-— at (908219xe OFpTow wor;) (79819xX9 oT eEq WoIz) c6%0 = “a ay ----- coo = ty ow 91 8 - 102 - Discussion The evidence obtained in this investigation indicates that IAA undergoes oxidative degradation in the presence of the TAA-oxidase system, giving rise to several oxidation products. That at least one of these products contains an indole ring is suggested by the red Hopkins=-Gole color given by a chloroform extract of a reaction mixture. That this color does not result, however, from the presence of 3~indole- aldehyde (as was originally proposed 897) ) is shown by the failure of this compound to give a similar color when treated with Hopkins-Cole reagent. At the same time, there is evidence (immediate formation of an orange-yellow color with Ehrlich's reagent) that at least two of these products contain aromatic amino groups, these necessarily resulting from cleavage of the indole nucleus. In the case of one experiment, one of the products (R, 0.82) contained, in addition to a potential amino group, an acidic group causing it to be extractable into chloroform from aqueous solution at pH 3, but not at pH 9. Although OFA, OAA, 3-indolealdehyde, and 4-hydroxyquinoline all possess R,'s within the range of the oxidation products, their behaviors with the spray reagents employed differ from those of the actual oxidation products. It does not seem likely that a combination of OAA (or OFA) and 3-indolealdehyde at the same position is responsible for the immediate orange-yellow ripening toward pinkish-orange which is observed on treatment with Ehrlich's reagent. In comparative chromato grams, the violet color of 3-indolealdehyde appeared at least five minutes after the color change of the oxidation products began. Moreover, this color ripening occurred in every case with not just one, but with all spots representing oxidation products. 4-hydroxyquinoline is definitely excluded as an oxidation product due to the failure of any -product to give an immediate permanent orange color with ferric chloride. It is further evident that neither 3-indolealdehyde nor OFA participate as intermediates inthe formation of the observed products. It is possible, however, that they are present to some extent and that their typical color reactions are masked by the presences of the major products. The absence of interference with these results by either the added co- factor, 2,4-dichlorophenol, or by any substances occurring naturally in the enzyme preparations was established by performing blank runs, identical with the oxidation runs, but containing no added IAA. Chromato~ grams of extracts from such blank reaction mixtures showed no color with either spray reagent provided that fresh enzyme preparations were used. Qn only one occasion, a chloroform extract of a brei which had been stored in the deep-freeze for several months yielded, at Rp 0.9, a slowly developing faint pink spot upon Ehrlich spray treatment. While the present indications that the indole ring of IAA is, at least in part, destroyed are in opposition to the conclusions of previous vorkers (829917) who have studied the IAA-oxidase system, some of the observations of these workers seem, nevertheless, to favor the concept of ring opening. The destruction of the indole ring of tryptophane reported by Knox and Mehter('8) ang by wiltshire’'?) has already been mentioned. Wagenknecht and purris(®) report that whereas the enzymic degradation of IAA results in the liberation of one molecule of carbon dioxide per molecule of oxygen consumed, the oxidation of -1%- 3eindolepropionic ‘acid occurs with an R.Q. of only 0.25. Such a _ reduced yield of carbon dioxide in the case of 3-indolepropionic acid could be explained by an opening of the indole ring to give a gamma- keto acid, which would be more resistant toward decarboxylation than the corresponding beta~keto acid resulting from Tas? Whereas the liver tryptophane oxidase studied by Knox and Mehler (28) was specific for the naturally occurring L isomer, the pea-seedling oxidase employed by Wiltshire (17) (possibly identical with our IAA oxidase) brought about the oxidation of both D and L tryptophane at the same rate, It is possible that such behavior could be related to the occurrence of an enzyme substrate complex involving modes of combination specific for the indole nucleus (or for its general structural outline) alone and more or less independent of the nature of the side chain, Although the hypotheses of OFA and 3-indolealdehyde do not appear adequate in the light of the present experimental findings, it should be noted that other possible explanations for the observed stoichiometry exist. For example, the formation of any isomeric dihydroxy skatole would occur with the uptake of one molecule of oxygen and the Liberation of one molecule of carbon dioxide per molecule of IAA activated. A compound of this type could not, of course, give a Schiff's base color with Ehrlich's reagent. Alternatively (or additionally), IAA inactivation may proceed via oxidative cleavage of the indole nucleus, and ® 4 proposition (“his Thesis). - 105 = the color reactions. described earlier in the present paper seem to indicate that this process occurs, Excluding OFA (or anything arising from it with no further gas exchange) as an oxidation product, it is difficult to devise a hypothetical scheme involving ring opening and in harmony with the 1:1:1 stoichiometry. The present experiments have demonstrated, however, that the IAA-oxidase catalyzed degradation of IAA leads to not one, but to several products. Unless the formation of one of these products alone may be considered to be the predominant course of the inactivation, it is pointless to hypothesize a single product capable of explaining the stoichiometry when the net gas exchange observed is certainly the resultant of a number of degradation pathways. Such a multi-pathway destruction of IAA is hardly surprising in view of the known complexity of the enzyme system involved and it is particularly suggested by the fact that IAA plays a dual role in the enzymic sequence, Not only is it the substrate for the H,0,~peroxidase complex, but it may also react in some peroxygenic capacity to produce peroxide for its own further destruction, If the stoichiometry of the inactivation process need not be accounted for by the formation of a single product it becomes possible to indulge in further speculation as to possible products resulting from oxidative cleavage of the indole nucleus. One possibility involves, in addition to opening of the pyrrol ring, hydroxylation of the benzene ring as well. Such a hypothesis derives some support from the apparent formation of 3~hydroxykynurenine from tryptophane by the same pea- seedling enzyme systen{'7) Moreover, tryptophane is known to be converted - 106 = into 3<hydroxyanthranilic acid in the rat (29) and in Neurospora‘? 0), If a hydroxy OFA (or OAA) were to arise from the IAA-oxidase mediated degradation of IAA, the hydroxylation step would necessarily (since . OFA is know not to be an intermediate) occur at some point before the actual ring opening as in the following sequence. » Se CO,H oy COzH 05, a eS sd NHCHO CHg Pin er '~CHy +s LES DEFORMYL. NHCHO A final product such as hydroxy OAA would have the advantages of being a near neutral compound (as is the product of RE 0.93) capable of giving a Schiff color reaction and probably (by virtue of the phenolic hydroxyl) a color with ferric chloride as well. Despite the mamerous possibilities for further investigation of the chemistry of this degradation along the lines presently taken, the way appears open for a more direct assault upon the problem. Since the possibility of chromatographic separation of the oxidation products has been demonstrated it would seem advisable to carry out a large scale conversion of IAA to its oxidation products, separate these chromato~ graphically, and attack the problem of their identities by the standard procedures of organic chemical characterization, a Actually, the hydroxylation would not have to occur as the first atep, but it would probably have to occur before the ring opening. Summary of Conclusions In the presence of pea-seedling IAA-oxidase, IAA undergoes oxidative conversion into at least two, probably more, products. Application of chromatographic techniques and color reagents indicates that some of the more obvious hypothetical products cannot account for the observed behavior of the authentic oxidation products. The compounds tested were 3-indolealdehyde, OFA, OAA, and 4~hydroxy- quinoline. It is also found that neither OFA nor 3-indolealdehyde participates as intermediates in the formation of the final oxidation products. Color reactions with Ehrlich's reagent and ferric chloride solution further indicate that the two most prominent oxidation products contain aromatic amino groups resulting from the oxidative cleavage of the indole nucleus. At the same time, the additional presence of a compound containing an indole ring is suggested by the Hopkins-Cole reaction. While the behavior of OFA with the various color reagents employed is not identical with the behaviors of the actual oxidation products there are, at the same time, certain similarities and it is felt that at least some of these products may be close to OFA in structure. It is suggested that the neutral oxidation product of R, 0.93 may be ea hydroxy-o-aminoacetophenone and a reaction sequence is proposed which involved hydroxylation of the benzene ring, followed by oxidation of the pyrrol nucleus in a manner similar to the ring openings of tryptophane observed by Knox and venter‘18) and Wiltshire !'7), (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) = 108 = REFERENCES Went, F.W., Rec. trav. bot. neerland 25, 1 (1928). Kogl, Fey Haagen-Smit, Aedes and Erxleben, Hes Ze physiol. Chem., 228, 90 (1934). Thimann, K.V., J. Biol, Chem., 109, 279 (1935). Haagen=Smit, A.J. et alo, am. J. Botany 33, 118 (1946). Wildman, S.G., and Bonner, J., Am. J. Botany 35, 740 (1948). Kogl, F., Haagen-Smith, A.J., and Erxleben, H., 2. physiol. Chem. 228, 104 (1934). Bonner, J., and Koepfli, J.B., Am. J. Botany 26, 557 (1939). Tang, Y.W., and Bonner, J., Am. J. Botany 35, 570 (1948). Wagenlmecht, A.C., and Burris, R.H., Arch. Biochem., 25, 30 (1950). Tang, Y.W., and Bonner, J., Arch, Biochem., 13,11 (1950). Galston, A.W., Bonner, J., and Baker, R.S., Arch. Biochem., 42, 456 (1953). Galston, A.W., and Baker, ReS.y Ame Je Bot. 38, 190 (1951). Goldacre, P.Le, Galston, A.W., and Weintraub, R.Le, Arch, Biochem. 43, 358 (1953). Kenten, R.H., and Mann, P.JeG., Biochem. Jeg 455 255 (1949) Kenten, R.He, and Mann, P.J.G., Biochem. J. 46, 67 (1950). Kenten, R.He, and Mann, P.J.G., Biochem. Je, 50, 360 (1952). Siegel, S.M., and Galston, A.W., in press. Wiltshire, G.H., Biochem. J., 55, 408 (1953). Knox, E.W., and Mehler, A.H.,y J. Biol. Chem. 187, 419 (1950). Knox, E.W., and Mehler, A.He, J. Biol. Chem. 187, 431 (1950). Theorell, He, Arkiv. Kemi, Mineral., Geol., 14B (20), 1, 1940; 15B (24), 1, 1942; 164 (2), 1943. (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) ~109 = REFERENCES (cont.) Bennett-Clark, T.A., Tambiah, M.S., and Kefford, N.P., Nature, 169, 452 (1952). Rohde, Ee, Ze physiol. Chem., 44, 161 (1906), Steensma, F.As, Z. physiol. Chem., 47, 25 (1906). Salkowski, E., Biochem. Z., 97, 123 (1919), Stowe, B., and Thimann, K.V,, abs. A.I.B.S. meeting, Madison, Wis., 1953. Hopkins, F.G., and Cole, S.W., Proc. Roy. Soc., 68, 21 (1901). Hopkins, F.G., and Cole, S.We, J. Physiol., 27, 418 (1902). Benedict, S.R., J. Biol. Chem, 6, 51 (1909), Galston, A.W., unpublished, Majima, R., and Kotake, M., Ber. 63, 2237 (1930). Witkop, Ann. 556, 103 (1944). Camps, Re, Ber. 34 (Vol. II), 2709 (1901), Heidelberger, C. et ale, J, Biol. Chem., 179, 151 (1949). Haskins, FeAe, and Mitchell, H.K., Proc. Natl. Acad. Sci., 22s 500 (1949). ~ 110 = SECTION IIT THE SYNTHESIS OF SOME oC =ALKYL oC -AMINO ACIDS AND THEIR DERIVATIVES Introduction In Section I of this thesis was described the working hypothesis that alpha-acylamine acid amides, of the general formula R, CHR, capable of interacting with the enzyme alpha-chymotrypsin as specific Ras substrates or competitive inhibitors do so by virtue of individual interactions of the three main structural features (R, » Ro» and Rg) of the molecule with three complimentary centers, Py Pos and 39 presumed to exist at the active enzymic site. Since both members of an enantiomorphic pair of such an amino acid derivative appear, in general, to combine with the enzyme at the same active site, the question arises as to what particular characteristics of those combinations confer upon the % isomer alone the behavior of a substrate. One possible explanation ig that whereas both BD and L isomers of an alpha-acylamino acid amide are capable of meeting the active site in a three-point attachment, the failure of the D isomer to undergo hydrolysis may result from its inability to attain an orientation in which the hydrolyzable bond of Ry is in a favorable position with respect to a particular enzymic entity presumed responsible for hydrolysis of the EL substrates. Certain lines of evidence described in Section I of this thesis have led us to believe, however, that both 2D and £ isomers of a great number of alpha~acylamino acid amides are capable of combination with the active site of alpha- chymotrypsin by virtue of the same Ry ~Pys Ry ~/o9 Ry - Py interaction. If this is the case, then we are required to explain the selective hydrolysis of the L isomers knowing that the carbamido group of the 3B isomer may enter into the same Ry “43 relationship. A consequence of - 112 = this hypothesis of similar modes of combination offers, however, a possible anawer to the problem. Let us consider the Ry Ro» and R groups of an alphaeacylamino acid amide as situated at three of the vertices of a tetrahedron having the asymmetric alpha carbon at its center and the alpha-hydrogen atom at the remaining vertex. It is then clear that if both D and |, isomers of such a derivative enter into einilar modes of combination (R, ~/y» Rk, - Pos R, = P3)s the alpha= hydrogen "vertex" of one isomer mist be directed away from, and the other toward the enzymic site. Within the realm of this difference - therefore, we may seck the reason for the difference in behavior of the D and |, isomers. Comparison of the enzyme~substrate and enzyme= inhibitor dissociation constants for a large number of enantiomorphic alpha=acylamino acid amides has shown that in every case studied, with but one exception, the D isomer of an enantiomorphic substrate-inhibitor pair is bound more tightly to the enzyme in its EI complex than is the isomer in ES, From this, it is reasonable to assume that if both isomers of such a pair undergo similar modes of canbination with the enzyme then the alpha hydrogen "vertex" is likely directed away from the enaymic site in the interaction of the D isomer (EI) and toward the site (impeding the union) in the case of the L isomer, The susceptibility to hydrolysis of the L isomers could then result from the introduction of strain into the amide linkage by the impeding orientation of the alpha-hydrogen., In the light of this interpretation the alpha~hydrogen atom of an |, substrate plays a curious and vital role in supplying an amount of steric strain sufficient to provoke the - 113 = hydrolytic reaction without preventing the ES union altogether. If this interpretation is correct it would seem likely that replacement of the alpha~hydrogen of an L specific substrate by a bulkier group would render it incapable of forming an ES complex and hence preclude its subsequent hydrolysis. Such a "deformed" substrate molecule might, however, be able to function as a bifunctional competitive inhibitor making use of the strong R, ~/“, interaction and the remaining R group properly oriented. On the other hand, substitution of a bulky group for the alpha-hydrogen of a BD inhibitor might not be expected to change, significantly, its K, value fran that of the original compound. I It was the purpose of the present investigation to test the adequacy of the above hypothesis by studying the behavior with chymotrypsin of several D and |, alpha=acylamino acid amides, substituted in the alpha position by bulky alkyl groups. Specifically, it was decided to prepare the alpha-methyl and alpha-isopropyl derivatives of DL tyrosine, DL phenylalanine, and DL tryptophane, resolve these into their B and L isomers, and then to determine the enzymic interaction properties of these isomers (in the form of their acyl amides) by the ‘methods of enzyme kinetics, - 114 = Synthetic Methods Relatively little attention has been given, in the past, to the synthesis of alphae-alkyl alpha-amino acids. This is due in part, of course, to the greater importance of and interest in the naturally occurring amino acids having hydrogen in the alpha position which has stimlated the work of synthetic chemists along this line. To a lesser degree it is due, undoubtedly, to the somewhat greater difficulties attending the syntheses of the alkyl acids, particularly when the substituent alkyl group is larger than methyl. Not all of the standard methols of synthesis of amino acids containing hydrogen in the alpha position are applicable to the preparation of their alpha~ alkyl homologues. Thus, the malonic ester type syntheses are ruled out due to the fact that alpha-hydrogen is automatically introduced during the final step of decarboxylation. The aldehyde condensation syntheses, including the hydantoin, diketopiperazine, and Erlenmeyer azlactone syntheses, all depend upon a Perkin type condensatimat an active methylene group followed by elimination of water to give a etabile un« saturated conjugated intermediate which is then reduced and hydrolyzed to the amino acid. The use of an active methylene component substituted at the alpha position by alkyl would necessarily lead to a beta~hydroxy compound, illustrated by the azlactone which would be incapable of R Ars} — <0 OH NWO RA! (I) losing water to give the unsaturated intermediate. Apparently, little ‘is known concerning the possibility of preparing such compounds as (I) by this method and converting them into alpha~alkyl alpha—<amino acids; a recent review" on the subject of azlactone chemistry mentions no such attempts. ‘The synthetic approaches to these alpha-alkyl compounds which have proved most successful are of two general types. Of these, the Strecker reaction and a modification of it in which the intermediate amino nitrile is converted into a 5,5=dialkyl hydantoin prior to the final hydrolytic step have been most widely employed. Thus, Herbst and Johnson") obtained a DL alpha-methylphenylalanine by vigorous hydrolysis” of 5—methyl-5'-benzylhydantoin. Their hydantoin was obtained by treating the amino nitrile hydrochloride of benzyl methyl ketone with potassium cyanate followed by a mild acid treatment to close the ring. Examples of the use of the conventional Strecker reaction include the conversion of the propiophenone into Dimal - ethylphenylalanine by treatment with ammonium cyanide and hydrolysis of the amino nitrile’) end a similar synthesis of Di-alpha~phenylalanine from acetophenone (2). Henze and Speer 7 ) have successfully converted a large number of ketones, including highly hindered ones, into the corresponding 5,5' dialkylhydantoins by treating the ketones in alcoholic solution with sodium cyanide and ammonium carbonate according to the method of Bucherer and tieo4), All of these hydantoins are “Carter, H.E., Org. Reactions ITI, 198 (1947). b By heating with aqueous barium hydroxide under reflux for thiry hours. - 116 = potential sources of their corresponding alpha~alkyl alpha-amino acids. A second group of synthetic methods for alphamalkyl alphae amino acids may be considered as a class of molecular rearrangements involving electronically deficient nitrogen atoms. Thus the Curtius rearrangement of acid azides, the Schmidt reaction upon ketones, the Hoffman degradation of amides, and the Lossen rearrangement of hydroxamic acid derivatives are all potential methods of replacing the carbonyl function of (II) by an amino group to give the alpha-emino acid skeleton, A Re v (A) C ~N\ RZ ~~ ¢-(8) oO (x1) In (II), R, and R, are the potential alpha-emino acid sidechain and ‘alpha-alkyl group respectively, —¢-(8) represents the carbonyl ° ’ function (-¢-N3 4 ~¢C-CH3 Oo ° fo) for the Curtius, Schmidt, Hoffman, and Lossen rearrangements, respec= —o-NH» and ~ Go NH- O-COR tively) to be replaced by amino, and (A) is a hydrolyzable carboxyl function ( -C=N, ,-CO2R etc.) which is inert toward the conditions of the particular method employed and may be hydrolyzed to the amino acid carboxyl group following the rearrangement. In each of these methods the amino acid side chain and alpha~alkyl substituent are initially introduced, generally as the halide, into ethyl acetoacetate or ethyl cyanoacetate. Lin, Li, and Huang ‘>) have reported the - 117 = syntheses of several alpha~alkyl alpha-amino acids, employing the - Hoffman rearrangement of the appropriately substituted half—amides of malonic ester. The amides were obtained by the partial hydrolysis of the corresponding dialkyl cyanoacetic esters with concentrated sul- furic acid at 25+100°, ' In the present investigation, both of the above mentioned classes of synthetic methods were explored as possible approaches to the desired alphamalkyl alpha-amino acids. In the following description of the synthetic procedures, the methods employed for the various amino acids will be outlined, reserving the more detailed gccounts of the important reactions for the experimental section, Also included are several interesting and rather anomalous reactions which complicated the experimental work and are considered worthy of mention. The Strecker reaction (Equation (1)) proved to be a practical route to the alpha-methyl derivatives of Di-phenylalanine, and Di- tyrosine, The appropriate methyl ketone, in ethanol-water solution, was treated with sodium cyanide, ammonium chloride, and ammonium hydroxide in a pressure bottle at 40~50° for a two hour period, Following acid hydrolysis and neutralization, the crude amino acid was obtained in good yield. | CHs NACH o* Mite R=#, cho In the case of the tyrosine derivative, cleavage of the nuclear methoxyl was accomplished with 30 per cent solution of hydrobromic acid ~ 118 - in glacial acetic acid and introduced no particular difficulties. An attempt to convert p-«methoxybenzyl isopropyl ketone into DL-O-nethy1-al phe«isopropyltyrosine by the Strecker reaction failed, the ketone being recovered, apparently unchanged, at the end of a ten hour reaction period. In the case of this more sterically hindered ketone, however, it was found possible to convert it to 5=isopropyl= 5' «p-methoxy benzylhydantoin, in 74 per cent yield, by the method of Henze and speer (7) Here the formation of the stable 5 membered ring overcomes the steric resistance of the ketone and drives the first step of hydantoin formation to completion. CH CH CHa CHs C- wna CH —1=0_ on ceo H The hydantoin thus obtained was heated with aqueous barium hydroxide under refluxing conditions in an attempt to convert it to DL-O-methyl= alpha-isopropyltyrosine. Only the unreacted hydantoin was isolated at the end of this treatment despite an attempt to recover the amino acid by acetylation, During the present work, considerable attention was given to the preparation of p-methoxybenzyl methyl ketone. Numerous accounts of the synthesis of this ketone are present in the early literature, most of which involve anethole as the starting material, p-Methoxybenzyl methyl ketone has reportedly been prepared by the dehydration of the isomeric glycols of anethole with 20 per cent sulfuric acid with sinc chloride (648) and by treatment of the epoxide of anathole with heat or acia§?) The epoxide of anethole is isomeric with p-methoxy~ benzyl methyl ketone and may be converted into it by a proton transfer mechanism, This method of preparing the ketone hag been employed as recently as 1938 by Auwers, Pdtz, and wo (1) While evidence indicating the formation of p=-methoxybenzyl methyl ketone (III) from the glycols of anethole has been adduced by at least one of the early workers? somewhat similar syntheses beginning with anethole have been described as leading to the formation of the isomeric compounds p-methoxypropiophenone (IV) and p-methoxyhydratropaldehyde (V). Chp fe) fo} (IIT) (Iv) (v) Specifically, the preparation of p-methoxypropionphenone (IV) has been reported by the treatment of anethole dibromide with alcoholic potassium hydroxide ('!) and by treatment with sodium methoxide followed by hydrolysie''?), Im this latter instance Wallach and Pond‘'@) supported their conclusion as to structure by performing a permanganate oxidation and identifying, among the oxidation products, both propionic acid and p-methoxyphenylglyoxalic acid® While a quantitative study of the oxidation products was not made and it might be argued that the presence & iffeneau and Daufresne”) reported the conversion of their product by hypiodite into iodoform and homoanisic acid, in agreement with the structure of p-methoxybenzyl methyl ketone. Each of these products could arise only from p-methoxyproplophenone and their empirical formulae were confirmed by analyses, - 120 - of pemethoxypropiophenone in a mixture of products could have accounted for their oxidation results, some indication of purity of their product was given by its melting point of 26-27°, In Table I is a list of the reported melting points of the various solid derivatives of the four isomeric p<methoxyphenyl-oxo-propanes. It is evident that on the basis of the oximes and semicarbazones (the only solid derivatives reported for both compounds prior to the present communication) alone, pemethoxy benzyl methyl ketone cannot be distinguished from p-methoxy=- propiophenone? p-Methoxyhydratropaldehyde has been reportedly prepared by treatment of anethole dibromide with aqueous sodium hydroxide followed by treatment of the resulting bromohydrin with hot alcoholic potassium hydroxide (13 ), by boiling the dibromide with an acetic acid solution of zinc acetate”), by heating the fodohydrin of anethole with aqueous potassium hydroxide''4), and by oxidising anethole with a mixture of iodine and mercuric oxide‘'>) able I indicates that the melting points of its known derivatives differ sufficiently from those of p=methoxybenzyl methyl ketone to prevent confusion of the two isomers, The same may be said for p-methoxyhydrocinnamaldehyde; moreover, neither of these compounds would be capable of the haloform reaction indicative of a methyl ketone. &n interesting synthesis of p-methoxybenzyl methyl ketone has been reported by Wallach and Matter (©) Anethole was treated with nitrous acid and the resulting intermediates converted into & Unless, of course, samples of eachwre available for mixed melting point determination of identity or non—identity. 7 121 = _ TABLE TI Melting Points of The Solid Derivatives of The Isomeric p-Methoxy phenyl- oxoO=propanes. Compound Mepe Oxine MeDe MeDe 294-DNP Semicarbazone p-methoxypropio- 714° ® 173-40 © 192=3° © -phenone oa 172=3 oc oc Ce) oi p-methoxybenzyl Th 175 107.8°=108.2 methyl ketone 718-9° a 175° a 108.6°=109.5° j 61=2° 175~6° © 56 61° g h 182° f T7 5°78 20 ok Oo p-methoxyhydra~ 96 135° 4b tropaldehyde 207-8 pemethoxyhydro— oe ol cinnemaldehyde 175-6 184 & Wallach and pona{'2) € Hoover and Hees("”) b Auwers, Pétz, and yor (1) h Winstein{'®) © mitfenean and Daufresne{”? 1 vert, to be published, a Wallach and Mutter §1©) J Manning, This Thesis. $ Balbiano and paoling (®) k Bougeuit ('?) £ péhel and Tiffeneau)4) 1 Forneau and Tiffenean (2°) 122 = 1-(p-methoxyphenyl)-2-nitro~L-propene by treatment with acetyl chloride. On CH-CHy HOMO, ¢, woh None sty SO w{ \cree- CH; No NO, "he Hot - CHO ChyC-CH, <= CHO Char S- CH 3 NOH The unsaturated nitro compound, obtained in 90 per cent yield, was reduced to the oxime which was then hydrolyzed to the ketone. Wallach and Muller concluded that this ketone was p-emethoxybenzyl methyl ketone rather than p-methoxypropiophenone on the basis of a hypobromite oxidation which yielded bromoform, anisic acid, and homoanisic acid. Further support of the conclusion of Wallach and Muller has been given by Hoover and Haes('”) in their synthesis of p-methoxy=- benzyl methyl ketones from anisaldehyde and nitroethane. Here, the l-(p-methoxyphenyl1)-2-nitro=-l-propene obtained from the initial con- densation, melting at 43-445 appears to be identical with the unsaturated nitro product of Wallach and Muller, On” (4) NOw No, "| Fe + fe) NOH The most recently reported synthesis of p-methoxybenzyl methyl ketone is that of Winstein et ail! 8) Anethole was treated with Nebromosuccinimide in aqueous solution and the resulting product, pre=- sumably the bromohydrin, converted to the epoxide by treatment with sodium hydroxide, ‘This was then rearranged to the ketone by boiling ‘with a solution of petoluenesulfonic acid. While this synthesis may possibly be criticized with regard to the uncertainty in the manner of ring-opening in the final step, the resulting ketone has been shown to give the bromoform reaction, to give a solid compound with sodium bisulfite, and possesses the IR absorption maximmat 5.83 microns characteristic(2") of aliphatic ketones. Moreover, the possible presence of p-methoxyhydratropaldehyde as a side product was legislated against by the failure of any part of the product to react with dimethone? That p-methoxybenzyl methyl ketone and not p-methoxy- propiophenone” results from all syntheses involving the acid catalyzed rearrangement of anethole oxide has been further indicated by an independent synthesis of p-methoxypropiophenone by Auwers, Pétz, and wou ‘1°) Phenyl propionate was converted into p-hydroxypropiophenone by the Fries rearrangement with boron trifluoride and the hydroxy ketone methylated to give the methoxy product. The melting point of their 2,4-dinitrophenylhydrazone’ 192-193°, is in contrast with the value of 107.8108.2! for the 2,jdinitrophenylhydrazone of the ketone prepared by Winstein's method. a ® A consideration of the reletive energies of the two transition states + - - + oot iu Si, CH and cwol_\-gandncny would predict formation of the methyl ketone, ° Enpirical formula determined by analysis!" 0) d Fort, A., and Roberts, J.D. to be published, - 12% = ‘In the present investigation, several methods for the Synthesis of p-methoxybenzyl methyl ketone were studied. A report by Mason and terry (22) of the condensation of chloroacetone with benzene in the presence of anhydrous aluminum chloride to give benzyl methyl ketone suggested the possibility of replacing the benzene with anisole to obtain the methoxy ketone, AICI (5) wo \sn + CICHa-¢~ CHa Met wok \rctig-c o fe) This condensation was attempted employing both aluminum chloride and boron trifluoride as catalysts, with various solvent systems, and at temperatures from 0~1 00°C. In all attempts an intractable resinous product resulted, Of the ketone syntheses attempted, the Darzens glycidic ester method proved the most successful and was used to prepare pemethoxybenzyl methyl ketone and p-methoxybenzyl isopropyl ketone according to the following reaction sequence, R { R crof \-cHo + CH(Br)-COrcaHs -NeOF? oy, New $- costs fo) (6) . -~COa f CHy CHER? <=, 0- CH~€—~ COaH R= CHss iso-C5H, 2 125 - Dar zens ‘73 ) has reported the preparation of p-methoxybenzyl methyl ketone by this method, and, although this report does not contain the properties of the ketoné's derivatives, the ketone prepared in the present study using this method was found to be identical®?” with the ketone obtained by the method of Winstein''®) ana identified as pemethoxybenzyl methyl ketone, The Grignard synthesis was investigated as a possible route to some of the ketones, the intention being to react a Grignard reagent of the desired alpha-alkyl group (methyl or isopropyl) with the nitrile of the amino acid side chain radical, (7) AR-CHeCN + RrMg-X ~ ho ClrG R R= CH3 4 (S0-C3 Ay RamarteLucas (24) has reported the preparation, in 60 per cent yield, of benzyl methyl ketone from phenylacetonitrile and methyl magnesium bromide. | When phenylacetonitrile was reacted with a 20 per cent excess of isopropylmagnesiumbromide and the reaction mixture treated with aqueous acid, however, benzyl isopropyl ketone was not obtained. Instead, a syrupy high boiling mixture resulted from which, on prolonged standing, a substance crystallized out in compact lemon<yellow prisms, mp. 179- 181,2°, This substance is the trimer of phenylacetonitrile and was obtained by Rondou'>) by the action of methylmagnesium bromide on the monomer as a yellow crystalline material melting at 178.5-1 79°, a b See experimental section for details. I am indebted to A. Fort for a sample of p-methoxybenzyl methyl ketone prepared by the Winstein method. = 126 - Rondou ‘29) also isolated the nitrile dimer, mep. 114.5=115°, which had previously been obtained by von Meyer (2°) from the action of sodium metal on phenylacetonitrile. By treating the dimer with hydroxylamine (26) and Rondou'29) obtained an under aqueous conditions, von Meyer oxime, MePe 107°, apparently resulting from the oximation of benzyl alpha-cyanobenzyl ketone (VII), the hydrolysis product of the imine dimer (VI). NH CN & CN While the crystalline dimer (VI) was not isolated in the present study, treatment of the undistillable residue with hydroxy- lamine hydrochloride gave the oxime, mep. 107.2-1 08.9°. A further example of the polymerization of nitriles under the conditions of the Grignard reaction has been given by Baerts‘2") When propionitrile was treated with ethylmagnesium bromide, diethyl ketone was obtained in low yield accompanied by the dimer and trimer of propionitrile. Ivanov and Paounoft'®) nave demonstrated the ability of organomagnesium halides to react with an active alpha-methylene hydrogen of phenylacetonitrile producing a new organomagnesium compound, (9) € N-crcen + R-Mg-8t ——> € Now CN +RH Mg Br A= CLHs 4 180-Cahh , a - 127. = When the nitrile was reacted with a 25 per cent excess of the Grignard -reagent and the resulting reaction mixture treated with carbon dioxide, phenyl cyanoacetic acid was obtained in 40 per cent yield. CN + CN 10 cHt + CO, —= 189, chZ (19) ( \ MgB On \coaH In the absence of added electrophilic reagents, the primarily formed cyano Grignard reagent may react with the cyano group of the nitrile itself to form the imino dimer, This condensation may be con= sidered as an example of the Thorp reaction in which the aldol type dimerization of nitriles is prompted by, in addition to organometallic compounds, sodium, sodium alkoxides, and scdium and lithium amide(*?) While the tendency of nitriles to undergo polymerization in the Grignard reaction may be decreased by the addition of a secondary reactant such as carbon dioxide (28) it may also be controlled by the presence of a large excess of the organomagnesium halide. Under these conditions, nitriles containing no extremely active methylene groups may react with the Grignard reagent at the cyano group to give a ketimine complex, convertible to the ketone by hydrolysis. Thus, whereas acetonitrile has a rather great tendency to undergo the polymerization", it, and a number of higher aliphatic nitriles, have been successfully converted into the corresponding phenyl alkyl ketones by treatment with a 300 per cent” excess of phenylmagnesiun bromide (30) a Presumably due to the steric freedom at the methyl carbon, b Higher excesses of Grignard reagent lowered , ¢ e yields of ketone due to formation of the tertiary carbinol - During the present study, several attempts were made to ‘prepare p=methoxybenzyl methyl ketone by reacting homoanisonitrile with a 300 per cent excess of methylmagnesium iodide. Instead of the expected product, a ketonic substance was obtained which resembled pemethoxybenzyl methyl ketone in some respects but was cléarly a different compound as indicated by the melting points of its derivatives and by its failure to react with aqueous ethanolic sodium bisulfite. In Table II is given a summary of its properties, including the melting points of its solid derivatives, compared with those of p-methoxybenzyl methyl ketone. _& comparison of the IR spectra of the two substances was made and it was found that while both were essentially identical in the region below ca. 9 microns, including the aliphatic carbonyl peak at 5.83 microns, there were certain dissimilarities at the higher wavelengths® The information presented in Table I and Table II indicates that the carbonyl compound arising from the Grignard synthesis cannot be any of the isomeric p-methoxyphenyl-oxo~propanes. | There is strong, if not conclusive, evidence favoring the identification of the principal carbonyl compound” resulting from the Grignard synthesis as 3~(p-methoxyphenyl)~2-butanone (VIII). This ketone could be formed during the Grignard synthesis according to the following sequences of reactions, * Of course, this might be attributable to impurities. 5 The product as a whole appears to be quite impure as judged by the difficulty in obtaining even approximately pure derivatives. - 129 = - TABLE IT Property or Ketone fron p-Methoxybenzyl Derivative Grignard methyl ketone Depe 95-100. 5°/650-800 101-104°/1500 * 88=92° /4,50-500 88-93° /S500 ° No 1.517674, 1.5185%4 1.522379*©, 4 503923 1.523672" a oxime 118119,2° 78-790 » nor 2 me 61-62 forms 2,4~DNP mp. 135,5~136.8° 108,6=109.5° * 107.8108.2° ° semicarbazone ° ob reaction with immediate solid bisulfite negative formation IR spectrum max. at 5.83 microns max. at 5.83 microns reaction with dimethone iedoform re action negative positive negative positive’?* @ Dargens synthesis. b c Wallach and Muller {'®) Winstein synthesis, = 130 - (11) uot CHeCN = + CHa Mg-L ——> amok gin Mal + CHy col \- cH-ow + CH;-LO ——> Hod \-gr-cw + Mgr cH, Mal, Hor ¢ Ms C CH-CN + CHa Mg-T —> => ¢ CH~C-CHs CHs ‘e) In view of the known tendencies of benzyl type nitriles to undergo the replacement of hydrogen at the active methylene carbon, and considering the present experimental conditions employed, 3-(p-methoxyphenyl)= 2-butanone (VIII) indeed would seem to be the expected product. Thus, in the reaction mixtures, the components; nitrile, methylmagnesium iodide, and methyl iodide were present in the approximate molar pro= portions of 1:43:1, the excess of methyl iodide over magnesium metal (Ca. 5:4) being sufficient to convert the cyano magnesium halide formed in the first step into the alpha=methyl nitrile (coupling reaction). 3~(p=methoxypheny1)-2-butanone is a known compound, having been prepared by sosa‘' 5) by an independent method. From Table IIT, in which the known properties and derivatives of Sosa's ketone are com= pared with those of the present Grignard product, it is evident that there is considerable similarity between the melting points of the oximes* and semicarbazones. The observation that 3«(p-methoxyphenyl )-2=butanone does not give a bisulfite compound (in agreement with the behavior of the Grignard product) is interesting and appears to reflect the increased steric hindrance introduced by the additional methyl group. In a The lower mep. of the presently obtained oxime may be due to the presence of other ketone oximes. ~ 131 « TABLE III Property or Ketone from 3=(p=-methoxyphenyl)= Derivative Grignard 2=butanone No 1.517674 1.528770 oxime Mep. 118=11942° 122=123° (125°) semicarbazone 172,8-1 hed” 170° MeDeo reaction with bisulfite negative negative - 132 = comparative, small scale (4 drops of ketone) iodoform tests, the product from the Grignard synthesis yielded much more iodoform than did pemethoxybenzyl methyl ketone. This would appear to be in agreement with the structure of 3—(p-methoxyphenyl)-2=butanone in which the methyl group upon the methylene carbon would block the iodination at this position with the concomitant unproductive (of iodoform) cleavage which occurs in the case of p-methoxybenzyl methyl ketone”, Further support for the identification as VIII of the Grignard product was given by analyses performed upon samples of its 2,4-dinitrophenylhydrazone, The analytical data is in fair agreement with that calculated for the butanone derivative (see experimental section). The data for the oxime, however, does not seem to correspond to that of any likely product and may result from the presence of impurities. An attempted Strecker reaction upon a sample of the Grignard carbonyl product resulted in a very low yield of an amino acid which was isolated and purified as the acetyl derivative (experimental section), Although this acetyl derivative was found to possess analytical data corresponding to N-acetyl-O-methyl-alpha-methyltyrosine, its melting point (222=224°) is considerably higher than that of the alpha=methyltyrosine derivative (178-181° ) and does not seem to represent a mere difference in crystalline modification, At the same time, the failure of the unknown acetyl derivative to give the ninhydrin test is some cause for believing that it is an alpha-alkyl amino acid. That this occurs in the case, pf p-=methoxybenzyl methyl ketone was shown by Wallach end Muller who identified anisic acid as one of the reaction products, =- 133 = It seems likely that this amono acid results from some minor ketonic constituent in the Grignard product, especially im view of the low - yield obtained. | | The homoanisonitrile, employed in the Grignard reaction, was best prepared by a method similar to that of shriner and Huli{?") reacting p=-methoxybenzyl chloride with cyanide ion in boiling aqueous dioxane. (12) cio Ven + KCN BO bey cy cen + Ka When the dioxane-water solvent was replaced with aqueous ethanol, pemethoxybenzyl ethyl ether was formed as a major side=product. The yields of this ether were found to increase (at the expense of the nitrile) as the ratio of ethanol to cyanide ion was increased (see experimental section), In as mich as aqueous ethanolic media have been successfully employed ‘> ) for the conversion of benzyl chloride to benzyl ¢yanide in high yields this occurrence in the case of the p-methoxy derivative is somewhat surprising. A probable explanation lies in the enhanced reactivity (in either or both the sy, and Ss, sense ) of the chloride due to p~methoxy substitution with a resultant levelling effect upon its selectivity toward reaction with various nucleophilic agents. Two methods were employed in the preparation of p-methoxy- benzyl chloride. - 134 < (134) io \ + CHO + HCI i> cKO PON, ces (13B) cok \-cron + HCI a Hod \nene . ~ #12 Of these methods, (13B) proved more satisfactory from the standpoint of yield than the chloromethylation procedure (134) and since the quantitative oxidation of the anisaldehyde, used to prepare the alcohol, to anisic acid (see experimental section) left no doubt as to its isomeric purity method (13B) was used exclusively for the preparation of the p-methoxybenzyl chloride used in the syntheses of the presently described amino acid derivatives. Preliminary investigations were made concerning the possibility of preparing the alpha~alkyl alpha-eamino acids via the Curtius rearrangement of dialkylcyanoacethydrazides,. RCN HONO RN _EtoH, RN RZ ~CONHNH. R.” ‘CONa R ~NHCO.Et q 4) R COsH HYOROL. rae Rx Wis While this method has been successfully employed (2 3) in the preparation of a mumber of amino acid derivatives unsubstituted at the alpha-carbon, there is no report of this reaction having been performed upon dialkyleyanoacethydrazides, In the present study attempts were made to convert lsopropyl-p-methoxybenzylcyanoacethydrazide and methyl-3- indolylmethyleyanoacethydrazide into their respective amino acid = 135 = derivatives. In each case the reaction products appeared unpromising and the attempts were not pursued further. Nevertheless, it is felt that these two failures are insufficient to condemn this method and _ that a further study of the Curtius reaction as applied to these com- pounds would likely reveal more suitable experimental conditions, Outside of the difficulties encountered in the Curtius rearrangement, a second problem exists in obtaining the dialkyle cyanoacethydrazides themselves, Thus, whileethyl methylcyancacetate was found to react readily enough with hydrazine, and Gagnon, Gaudry, and King ? 3) have reported a similar reactivity for the isopropyl derivative, the substitution of cyanoacetic ester by pairs of bulky groups (such as benzyl and isopropyl) brings about great resistance toward hydrazinolysis of the ester group, After eight days of treatment with a large excess of anhydrous hydrazine at 120+170°, ethyl isopropyl- pemethoxybenzylcyanoacetate underwent only 18.4 per cent conversion to the hydrazide, The synthesis of alpha-alkyl alpha-emino acids via the dialkyleyancacethydrazides appears to have great potential possibilities for the preparation of the alphasealkyl tryptophanes, When, in the present study, an equimolar mixture of gramine and ethyl methyl cyanoacetate was heated to 130-150° condensation occurred with the liberation of an amount of dimethylamine equivalent to a 61.6 per cent conversion of the gramine. - 136 = CN SN 3 \ ee CO,CG.He N CO,C2Hs N am When the difficulty of reacting hydrazine with highly hindered dialkyleyanoacetic ester derivatives became evident, the possibility of introducing the 3-indolylmethyl group into preformed monalkyl hydrazides was investigated. It was found that gramine and methylcyanoacethydrazide will also undergo condensation, upon fusion, liberating dimethylamine and forming the crude disubstituted hydrazide in high yield. C Lo cH nw” pn mn (16) [7 Ncw + CH(CH) ASS Cw \ CONHNH,, N CONH NA, N H #*{CHS), NH Conclusions Regarding the Synthetic Methods “he synthetic studies deseribed in the present paper enable certain conclusions to be drawn regarding the methods of preparation of the alpha-alkyl alpha-amino acids. While the Strecker reaction per se is of limited scope in the preparation of these compounds, modification of the Strecker reaction by the addition of ammonium carbonate to form the 5,5'-dialkylhydantoin should greatly increase its synthetic capabilities. Thus, the application of the Strecker reaction is limited to ketones which are unbranched in the two positions adjacent to the keto group. Moreover, the remainder of the molecule mist not be sensitive to the vigorous acid treatment used to hydrolyze the amino nitrile. More sterically hindered ketones - 137-- may be converted into their corresponding alpha-elkyl alpha-amino acids ‘by preliminary formation of the 5,5'-dialkylhydantoins followed by hydrolysis: In this case treatment with acid is avoided, the hydrolytic cleavage being effected with aqueous barium hydroxide. The Curtius rearrangement remains an unproved but promising possibility for the synthesis of alpha-alkyl alpha~-amino acids. While it is difficult to prepare the hydrazides of the highly hindered dialkyl derivatives of cyanoacetic ester it may be possible to improve the yields and convenience by carrying out the hydrazinolyses in sealed tubes at temperatures of ca, 1301 60°, The syntheses of the alpha=alkyl derivatives of DL-tryptophane involve special problems peculiar to the 3-indolylmethyl group. At the same time the properties of this group in the form of the Mannich~ type base gramine open new synthetic possibilities. Gramine may be easily condensed with monoalkyicyanoacethydrazides, thus avoiding the problem of reacting hydrazine with a highly hindered disubstituted derivative, It seems possible that other active methylenic compounds will be found capable of condensing with gramine to give intermediates which can then be converted to the Di-alpha-alkyltryptophanes by methods superior to the Curtius rearrangement, It would be of interest to attempt the condensation of gramine with certain 4-alkyl oxazolones to give 4,4' disubstituted exazolones convertible to the alpha-alkyl a No amino acid was isolated, however, in the present attempt to hydrolyze 5-isopropyl-5' -(P-methoxybenzyl)<hydantoin (pages 118, 161-2). Nevertheless, it may be possible to obtain better results by conducting the hydrolysis in a sealed vessel at higher temperature. = 138 = alpha-amino acids by direct hydrolysis? _ Sur mC eur? xa S wD Ps oun: CO2H The use of p-nitrophenyl for R' would be expected to. promote the HYDROL., condensation by increasing the lability of the active hydrogen atom. 4n additional possibility for the preparation of the DL-alpha~ alkyltryptophanes is, of course, by way of the Strecker reaction upon the appropriate 3-indolylmethyl alkyl ketones. These ketones may be prepared by the decarboxylative acylation of 3—indoleacetic acid with acid anhydrides in the presence of sodium acetate or pyridine, Brown, Henbest, and Jones ‘24) have prepared 3—indolylacetone by the prolonged action of acetic anhydride and fused sodium acetate upon 3-indoleacetic acid at 135<140°, The resulting N-acetyl-3-indolylacetone was saponified by treatment with methanolic sodium methoxide followed by aqueous acid? CH: COaH CH3sCO. Na OAc CH G-CHy Cw! * CH,co” (30-40 } 8 H , (18) CO-CHs : we? “hor While there may be some question regarding the ability of the indole nucleus to withstand the acidic treatment involved in the Strecker a Proposition, This Thesis. b A 24.6 per cent yield of 3-indolylacetone was obtained from 2.0 g. of 3<indoleacetic acid, (34) - 139 « reaction, it should be possible to employ the hydantoin modification with success. This modification will, of course, be the necessary route to the more sterically hindered 3<indolylmethyl alkyl ketones, Attempted Resolutions of The Dimalpha-Alky] alpha-amino Acid Derivatives. Several enzymatic methods were employed in the attempted resolutions of the various derivatives of DL~alpha-methyiphenylalanine and DL~alpha-methyltyrosine. Asymmetric hydrolyses attempted were? (1) N-acety1-DL-O-methyl~alpha-methyltyrosine methyl ester by chymotrypsin in aqueous methanolic and in water solutions, (2) O-methyl- DI-alpha-methyltyrosine methyl ester by “Viobin" in an aqueous paste, and (3) N-acetyl-DL-alpha-methylphenylalanine methyl ester by chymotrypsin in aqueous methanol and in water. Also attempted was the papain catalyzed asymmetric synthesis of N-acetyl-L-alpha-methyltyrosine- p-toluidide. In only one instance, namely the incubation of N~acetyl- DL~alphaemethylphenylalenine methyl ester with an approximately equal weight of chymotrypsin in water solution, was there evidence of a small amount of ester hydrolysis, This evidence suggests that, at best, the enzymatic resolution of these compounds would be a very slow process requiring prohibitively costly amounts of chymotrypsin to resolve gram scale quantities of the alkyl amino acidss A complication seems to exist, moreover, in the use of high concentrations of enzyme, namely the autolysis of the enzyme over long reaction periods with consequent con- tamination of the desired products by the products resulting from the breakdown of the enzyme. a This requires the assumption, not too well justified, that only the Lealpha-alkyl esters are split in the process. The failure to effect enzymic resolutions of the alpha-alkyl esters led to the consideration of other separation methods. An attempt was made to fractionally crystallize the diestereomeric L-brucine salts of D and Neacetyl—O-methylealpha-methyltyrosine, without success. Nevertheless, it is felt that future attempts to resolve the enantiomorphic alpha-alkyl alpha~amino acids should be directed along the lines of fractional crystallization of such diastereomeric mixtures, Particularly promising is the recent success of Ingersoll et a1 $29) in resolving a number of acetylamino acids by the use of alpha-fenchylamine, and it would be worth while to attempt the application of this method to the present problem, The resistance of the Lealpha-alkyl-alpha~acetylamino acid esters to enzymic hydrolysis is in great contrast to the behavior of Lealpha~-acylamino acid esters possessing hydrogen in the alpha position, Despite the present inability to obtain the individual isomers of the alpha-alkyl alpha~amino acids for enzyme kinetic study it is clear that the presence of alpha-alkyl substituents in the alphaeacylamino acid derivatives impedes the ES interaction necessary for the occurrence of the hydrolytic and synthetic reactions. This observation is in accord with the hypothesis presented in the introduction to this section. - 41 - Experimental Di-alphe-Methyiphenylalanine -~ Adapted from the method of Steiger, Org. Synthesis 24, 9 (1944). Te a solution of 12.5 &- of sodium cyanide (powdered 96 per cent egg) and 14.7 g. of ammonium chloride in 60 ml. of water contained in a 500 mle glass stoppered Erlenmeyer flask was added 33.5 ml. of concentrated aqueous ammonia (Sp. Gr. ca. 0.9) and the flask stoppered. To this was added a solution of 33.5 g. of benzyl methyl ketone (Eastman practical) in 75 ml, of 95 per cent ethanol, adding the solution in four portions with shaking. The stopper of the flask was then wired down and the mixture warmed in a water bath at 40-50° for a period of two and one quarter hours. After standing overnight at room temperature the mixture was chilled to 0-5° and poured into 200 ml, of cold 12 N hydrochloric acid contained in a 1000 ml. 3-neck flask equipped with inlet thtble tube and reflux condenser, and then saturated, at 0-5°, with anhydrous hydrogen chloride, After standing overnight, the reaction mixture was diluted with 250 ml. of water and heated under reflux for two and one-half hours in the hood. Water was then removed by vacuum evaporation and the resulting solid collected and stored over sodium hydroxide in a vacuum desiccator, A second crop of solid was obtained by evaporating the filtrate to dryness on the steam bath. The dry solid, freed of excess hydrogen chloride, was dissolved in 150 ml. of warm absolute ethanol, ~ 142 = the resulting solution cooled, and the precipitate of ammonium chloride removed by suction filtration. DL-alpha-~methylphenylalanine crystallized from the filtrate upon the addition of 40 ml, of pyridine, chilling the mother liquor in an ice bath during the crystallization. The first crop weighed 23.0 ge, Mep. 283-285° with decomposition, Reported!) MeDe 293=294°. Three additional crude crops of total weight 14.1 2g. were recovered from the mother liquor. Acety1-DI.~alpha-methylphenylalanine = A solution of 3.0 g. of crude DL~alpha-methylphenylalanine in 70 ml. of water was chilled to 0-5° in an ice bath and treated with 3.4 g. of acetic anhydride and 1.34 g. of sodium hydroxide in aqueous solution, added in small portions with contimied shaking and cooling in such e manner that the reaction mixture remained strongly basic throughout the addition. The reaction mixture was evaporated in vacuo to 40 mle, cooled to 0-5°, and acidified with hydrochloric acid. Acetyl<DL-alpha~methylphenylalanine was allowed to erystallize in the cold for several hours and was then collected and dried in a vacuum desiccator. Recrystallization from water gave 1.52 g. of white needles mp. 194.8-197.7°, Anal, Caled. for C1 oH 505 : Cy 65.13 H, 6.83 N, 6.3. Found: C, 65.03 Hy 6.93 Ny 6.3. DL~alpha-Methyl phenylalanine Methyl Ester - A suspension of 5.0 g. of Dipalpha-methylphenylalanine in 100 ml. of absolute methanol was cooled to 0-5° and saturated with anhydrous hydrogen chloride. The resulting solution was heated under reflux for 16 hours and the excess methanol - 143 = then removed by evaporation jin vacuo. The ester hydrochloride was neutralized with aqueous potassium carbonate and the resulting solution allowed to stend for four days in the refrigerator. The stout prismatic erystals which had formed by this time were separated from the mother liquor and the latter was evaporated to dryness under an air stream, The resulting solid was dried in a vacuum desiccator and extracted with hot acetone to recover the ester. 4.8 g. of crude ester was obtained, mep. 108, 5=122°, Acotyl~DL-alpha-methylphenylalanine Methyl Ester — A solution of 4.0 g of crude DL-alpha-methylphenylalanine methyl ester in 60 ml. of 1031 water=acetone was cooled to 0-5? in an ice bath. To this was added 6.3 g. of acetic anhydride in small portions, shaking and cooling after. each addition, together with sufficient aqueous sodium hydroxide to keep the reaction mixture above pH $ during the addition of the first 4.2 g. of acetic anhydride, Addition of the remainder of the anhydride caused the acetyl ester to precipitate. Recrystallization from aqueous ethanol gave 1.3 g. of small lustrous blades, mp. 118.1-119.0°. Anal. Caled. for Cy gHyy04N Cy 66.43 Hy 7233 Ny 209g Found: Cy, 66.23 Hy 7.23 Ny 5.8. peMethoxybenzyl Alcohol = A procedure adapted from Org. Synthesis Coll. Vol. II, page 590 was employed. To a solution of 434 g. of potassium hydroxide in 680 ml. of absolute methanol in a 3000-ml. 3eneck flask equipped with reflux - 144 =< condenser and mechanical stirrer was added, with stirring, a solution of 260 ml. of aqueous formaldehyde (37 per cent Merck), 354.3 g. of redistilled anisaldehyde* (beps 129/18 mm.), and 270 ml. of absolute . methanol over a period of 30 minutes. The temperature, initially at 58°, rose to 60-70° at the beginning of the addition and was maintained at. 64-67° for three hours, 820 ml. of methanol was then removed from the reaction mixture by distillation, and 780 ml. of cold water added causing an upper organic phase to separate. The aqueous layer was extracted with benzene and the benzene extract combined with the original organic layer. The organic phase was washed with water and dried over anhydrous sodium sulfate. After removal of the benzene, the p-methoxy=- benzyl alcohol was distilled at reduced pressure through a short Vigreau colum, Yield, 322 g. (90 per cent of theoretical) bap. 91-93°/500A4. 22-9 = 1,5440. For an authentic sample(Matheson ) 21 My 1.5450. p-Methoxybenzyl Chloride Method J. = Adapted from Org. Reactions Vol. I, page 63. 216.5 g. of anisole, 194.0 g. of aqueous formaldehyde (37 per cent Merck), and 1200 ml. of concentrated hydrochloric acid (assay 37.6 per cent) were cooled to 0-1 0°, and shaken for 14 hours in a 2000-ml. RB flask, The shakingwas interrupted and cooling in an ice bath applied at intervals to maintain a temperature below 15°, The reaction mixture was diluted with water and extracted with ether and the ether extract dried over anhydrous sodium sulfate. & Basic permanganate oxidation of 1.0 g. of the aldehyde see 1.3 g. of crude anisic acid, mep. 179.5-181° (Reported MePe 184°). - 145. = After removing the ether by vacuum evaporation, hydrogen chloride was ‘removed by pumping at room temperature with a vacuum pump for 15 min, The product was distilled rapidly through a short path distilling head into a fraction cutterecondenser, The fraction boiling at 100.5-1 83°/6-123 mm, weighed 128 g. (41 per cent of theoretical). During the distilla- tion hydrogen ehloride was evolved with the resulting pressure increase noted. A higher boiling fraction, bepe 176-180°/5=6.5 mms, crystallized in the receiver, Recrystallization of this material from 95 per cent ethanol gave plate like crystals, m.p, 46.6-50.8°, Tentatively identified as bis-(p-methoxypheny1)-methane® The p-methoxybenzyl chloride yielded, upon permanganate oxidation, anisic acid in fine yellowish-white needles, mp. 180-182,2°. iit?) mop, 184°, Method II ~ Essentially the method of Shriner and pu (3!) A solution of 210 g. of p=methoxybenzyl alcehol in 725 ml, of absolute ether was cooled to 0~5° and saturated with anhydrous hydrogen chloride. After standing in the cold for 20 hours, excess hydrogen chloride and ether were removed by vacuum evaporation, and the organic phase washed successively with water, aqueous sodium bicarbonate, dilute aqueous sodium chloride, and saturated aqueous sodium chloride. The organic phase was dried over anhydrous sodium sulfate and the ether removed by evaporation in vacuo, Prior to distillation the organic r In the preparation of pemethoxybenzyl chloride by this method Sosa(15) reported the formation of this compound (mp. 52-53°) as a side product, - 146 = material was pumped on with a vacuum pump for two hours to remove hydro= gen chloride. The p-methoxybenzyl chloride was then distilled rapidly through a short path system in the manner described above, The fraction collected at 117~1 23°/16 mm. weighed 205 g. (86 per cent of theoretical)* y 26 = 1.5461. reported ‘'9) no = 1.54916 peMethoxybenzyl chloride, obtained by either method, is unstable, undergoing eventual polymerization with concurrent liberation of hydrogen chloride. This decomposition is accelerated by heat, hydrogen chloride, ground glass surfaces, and ferric ion, and may occur very rapidly during distillation, The liberation of hydrogen chloride during distillation (particularly of the crude product) had the adverse effect of increasing the pressure in a system evacuated by a vacuum pump and thereby increasing the temperature and rate of decomposition. This difficulty was overcome to a large extent by evacuating the systems with a water aspirator during distillation, “lthough incapable of evacuation much below 15 mm., a water aspirator prevents the hydrogen chloride accumulation. Pure redistilled (b. p.- 91-93°/500.4 ) p-methoxybenzyl chloride decomposes slowly and it is necessary to add solid calcium carbonate occasionally to preserve a sample over a period of several months. Homosnisonitrile - A method adapted from Shriner and Hul1'?") was 2 gosa‘'?) obtained a similar yield (80 per cent) by this method. - 147 = To a boiling solution of 182 g. of potassium cyanide (Mallinckrodt Analytical Reagent) in 153 ml. of water, contained in a 3000-ml. 3neck flask equipped with reflux condenser and mechanical stirrer, was added, with stirring, a solution of 109.4 g. of p=methoxy- benzyl chloride in 220 ml. 1,4 dioxane (redistilled from sodium) over a 45 minute period. Refluxing and stirring were continmed for an additional three hours after which the dark reaction mixture was cooled, filtered through “hyflo" by water aspiration, and the excess solvents removed by evaporation in vacuo. The dark reaction mixture was again filtered, then washed with water and dried over anhydrous calcium chloride. The erude nitrile was distilled first at reduced pressure from a Claisen flask containing a few em. of packing (Raschig rings ). The pale yellow fraction boiling at 101-115°/500-1200 was collected and re- fractionated through a short Vigreau colum. Yield, 79.8 g. (77.5 per cent of theoretical) of homoanisonitrile. bsp. 101-106°/8004 . 2165 = 1.5320, Reported'%) 16-8 = 4.5317, | 5.0 g. of the above nitrile was hydrolyzed by heating under reflux for ca. 50 hours with concentrated aqueous sodium hydroxides After filtering the cooled reaction mixture, precipitation of the acid by acidification, and reerystallization from hot water, 1.82 g. of lustrous platy crystals, mop. 85.9-87.9°, was obtained. Reported (?7) for homoanisic acid, mep. 85-87°, a In performing the basic hydrolysis of homoanisonitrile, sosa‘!? ) obtained only 40 per cent of the expected homoanisic acid. <A side reaction was dimerization of the nitrile and hydrolysis of the dimer to give Attempts to convert p-methoxybenzyl chloride to the nitrile _ by treatment with potassium cyanide in aqueous ethanol led to the formation of pemethoxybenzyl ethyl ether as a major by product. The yield of the ether was found to increase (and that of the nitrile to decrease) as the ratio of ethanol to cyanide ion in the reaction mixture was increased. This is illustrated by theresults of the following two experimenta,. Trial Molar ratio Yield (on basis of chloride) (CNY Keron) of ether of nitrile 1 0.01 2262 per cent 31.3 per cent 2 0.0028 40.1 per cent insig. For p-methoxybenzyl cyanide obtained from Trial I, beps 97=100.5°/620= 6504, 22? = 1,530,. Anal Caled. for CoHyO Nz Cy 713.43 Hy 6024 founds CG, 73.13 Hy 6.3. For pemethoxybenzyl ethyl ether obtained from Trial II, Deps 75+5-76.2°/930K , NA! = 1.5060 Anal. Caled. for Cy Hy 4023 Cy 72.33 Hy 8.5. Found: C, 72.33 Hy 8.56 Carbonyl Product from the Reaction of Homoanisonitrile with Methyl— magnesiumiodide £92 g. of ether washed magnesium turnings (99.8 per cent Mg) was placed under 60 ml, of dry absolute ethyl ether in a dry 500-ml. - 149 - 3-neck flask equipped with mechanical stirrer, reflux condenser, and . dropping funnel, A few mle of methyl iodide (Matheson ) was added to start the reaction and, with stirring, a mixture of 34.3 g. of methyl iodide and 100 ml. of dry absolute ethyl ether was added dropwise over a 45~minute periods At the end of this time thereaction elackened and a warmvwater bath was applied for 15 min. to effect complete reaction of the magnesium. To the reaction mixture, cooled to < 0° in an ice-sodiun chloride mixture, was added a solution of 7.43 g. of homoanisonitrile in 20 ml. of dry absolute ethyl ether with vigorous stirring over a period of 15 minutes, Following addition of the nitrile, stirring was continued for one hour and the reaction mixture was allowed to stand overnight. The reaction mixture was carefully poured onto a mixture of 60 ml. of concentrated hydrochloric acid and 100 g. of ice and the resulting organic and aqueous phases separated, The acidic aqueous phase was heated under reflux for one hour and the resulting dark oil extracted with ether, The ether extract was washed first with aqueous sodium bicarbonate, then with aqueous sodium thiosulfate to remove the iodine liberated during the reflux. The original ethereal phase was combined with the ether extract and the ether removed by evaporation. To insure recovery of any ketone converted to the bisulfite compound by the thiosulfate treatment, the organic phase was steam distilled with the aqueous thiosulfate extracts (which had been acidified with a This rate of addition provided a rapid but controlled reaction rate as evidenced by the rate of reflux, ~ 150 = dilute hydrochloric acid) and the nearly colorless oily component of the distillate recovered by other extraction, washed with aqueous bicarbonate, and dried over anhydrous sodium sulfate, Following removal of ether by vacuum evaporation, the product was distilled from a Claisen flask. A yellow fraction boiling at 97=105°/700-800 A was collected and refractionated to give 3.8 g. pale yellow oil, bape 95-100.5°7/650-8004 , 285 = 1.5190, 2,4~Dinitrophenylhydrazone Sample A.=- Prepared according to the procedure of Lucas and Preasman’""), Reorystallization 3 times from absolute ethanol and twice from 60-70° ligroin gave golden needles, mp. 135.5=1 36.8°. Sample B.- Prepared according to the procedure of Shriner and Fuson ‘9? ), 6 recrystallizations from absolute ethanol and ligroin- ethanol mixtures gave flattened golden needles, mp. 133.4=136.1°, Anal, Caled. for C,H, 605N 1, (Pomethoxybenzy1 methyl ketone 2,4~DNP) Cy 55683 Hy 4.73 Ny 16.3. Caled. for C,H, gos As 3-(prme thcaypheny1)-2-butanone 9 a Cy 57.03 Hy 5013 Ny 15.6. Found : Sample A - Cy 57.93 Hy 5.03 Ny 15.4. Sample B = C, 56.8; Hy 5.03 Ny 16.0. QOxime - A solution of 0.5 g. of the ketone, 0.7 g. of hydroxylamine hydrochloride, and 20 ml. of 1:1 ethanol-water (v per v) was made pH 7 - 151 = with aqueous sodium hydroxide and heated under reflux for 10 mimtes. Evaporation of the ethanol and cooling led to the separation of the oxime as an oil which crystallized upon seratching. 4 recrystallizations from 60~70 ligroin gave colorless needles, m.p. 118=119,2°, Anal. Caled. for Cy qty gO" (p-methoxybenzyl methyl ketoxime) Caled. for C,,H, 50H (3—(p-methoxyphenyl)-2=butanone oxime ) - Found: Behavior with Sodium Bisulfite - Under conditions in which benzyl methyl ketone and p-methoxybenazyl methyl ketone gave rapid formation of solid derivatives the Grignard product failed to react to any extent with saturated aqueous-ethanolic sodium bisulfite. Dimethone Test - 0.1 g. of the product was boiled for 5-10 minutes with an excess of dimethone in aqueous ethanol. Upon cooling in the refrigerator, unreacted dimethone, mp. 144.7-150.9°, was obtained, Iodoform Test - Applied as directed by Shriner and Fuson’2?), the Grignard product gave a relatively large amount of golden lamellar jodoform, melting point of crude product, ca. 110-117° with charring. An equal amount of p=-methoxybenzyl methyl ketone gave a much smaller yield of iodoform,. ~ 152 = Strecker Reaction upon the Grignard Ketonie Product 20.1 g. of the ketone was treated with potassium cyanide, ammonium chloride, and. ammonium hydroxide under the same conditions as . described for the preparation of DL-O-methyl<alpha-methyltyrosine. After the hydrochloric acid treatment, the pH of the solution was adjusted to ca. 7 causing the separation of 1.05 g. of a brownish-white solid. . The product was then acetylated with acetic anhydride in the usual manner. 0.83 g. of brownish crystals were obtained and these were recrystallized from ethanol-water to give 0.44 g. of short white needles, m.pe. 222m22/,°, Ninhydrin Test - A small sample of the acetyl derivative was heated with dilute hydrochloric acid at 90-100° for ca. 16 minutes to liberate the free amino acid, the pH adjusted to ca. 7, and 4-5 drops of 1 per cent ninhydrin solution added. No color was produced even after 25 minutes of heating at 90-100°. Similar treatment of acetyl-DL~tyrosine produced an intense blue-violet coloration in 5 minutes, Reaction of Isopropylmagnesium Bromide with Phenylacetonitrile Procedure adapted from Org. Synthesis 21, 79 (1941). To 14.6 g. of clean magnesium turnings in a dry 1000-ml. 3eneck flask equipped with reflux condenser, mechanical stirrer, and dropping funnel was added a small amount of isopropyl bromide to start - 153.- the reaction followed by 125 ml. of dry absolute ethyl ether. A solution of 61.5 ge of isopropyl bromide in 200 ml. of absolute ether was then added dropwise, with stirring, at a rate causing the mixture to reflux gently. When the action slackened, heating under reflux was applied for an additional hour. ‘The reaction mixture was cooled in an ice=-salt bath and, with stirring, a solution of 48.8 g. of redistilled phenylacetonitrile (b.p. 108-108.9°/14 mm.) in 100 ml. of absolute ethyl ether was added dropwise. After the addition of the nitrile the reaction mixture was allowed to stand for two hours. The reaction mixture was decomposed in the cold with excess dilute sulfuric acid, the mixture extracted with ether, and the ether extract wakhed with aqueous potassium carbonate, water, and saturated aqueous sodium chloride, and dried over anhydrous sodium sulfate, After removing the ether by vacuum evaporation an attempt was made to distill the syrupy remainder at reduced pressure. When it became evident that no significant amount of material boiling in the range expected for benzyl isopropyl ketone was present, the attempt at distillation was abandoned and the high boiling material stored in the refrigerator. In the hope of identifying any ketonic materials present in the undistillable material, a small amount of it was treated with hydroxylamine hydrochloride according to the directions of Shriner and Fuson (9? ) for the preparation of oximes. A white solid product was obtained which was recrystallized from aqueous ethanol to give compact prisms melting at 104.6-107.4°. Two reerystallizations from absolute ~ 154 = ethanol gave flat colorless prisms, MePe 107.2-108.9°. Reportea (29 ) for the oxime of dimeric phenylacetonitrile mp. 107°, Anal. Caled. for Cy gHx0 Nps C, 77.13 Hy 5.33 N, 11.2. Found s C, 76.93 Hy 5.53 Ny 11.1. Upon standing in the cold for two weeks a yellow crystalline material, mep. 178=180.2°, separated from the undistillable reaction product, Reerystallization from absolute ethanol gave compact lemon- yellow prisms, mep. 179~181.2°, Reporteal25) for the trimer of phenylacetonitrile, mep.'178.5-179°, Anal. Caled. for Co, Ho Ng : Cy 82.03; H, 6.03 N, 12.0. p-Methoxybenzyl Methyl Ketone ~ By the Darzens glycidic ester method, adapted from Org. Reactions Vol. V, page 427. 59.8 g. of ethyl alpha-bromopropionate (Eastman) and 45.1 g. of redistilled anisaldehyde (129°/18 m.) were placed in a dry 500-ml. 3eneck flask equipped with reflux condenser, mechanical stirrer, and dropping funnel, The mixture was cooled to ca. -10-11° in an ice-salt bath and the system was swept out with a stream of dry nitrogen, Then, with stirring and contimel cooling a solution of 7.6 g. of sodium metal in 160 ml. of dry absolute ethanol was added dropwise over a three hour period. Stirring at the low temperature was continued for 12 hours and then continued, warming to room temperature, for two hours longer. Finally, the mixture was heated under reflux, with stirring, for a period of six hours. - 155. = The reaction mixture was treated with cold water, acidified with acetic acid, and solid scedium chloride added to cause separation of the organic phase. The glycidic ester was extracted with ether and the ether extract washed with aqueous sodium bicarbonate, water, and saturated aqueous sodium chloride , and then dried in two stages, first with anhydrous scdium sulfate then with drierite., after removal of ether by vacuum evaporation the material was distilled at reduced pressure from a Glaisen flask. 47.2 g. (60.4 per cent of theoretical) of glycidic ester was collected over the b.p. range 120=135°/50044 « 12667 = 1.5114 (for a sample prepared in a similar preparation as the above and boiling at 122-131°/500 4). With stirring, the glycidic ester was then added to a solution of 4.8 g. of sodium metal in 100 ml, of dry absolute ethanol and the resulting deep yellow solution cooled in an ice bath. Maintaining the temperature at 10-20°, 5 ml. of water was carefully added to precipitate the sodium salt of the glycidic acid, after standing over= night in the refrigerator the sodium salt was collected by suction filtra- tion and washed with 15 ml. of cold ethanol and 35 ml. of ether successively. The sodium salt was heated under reflux with a mixture of 93.5 ml. of water and 17.5 ml. of concentrated hydrochloric acid for two and one-half hours. After cooling, the yellow oily ketone, liberated by the acidic reflux, was extracted with benzene and the benzene extract washed successively with aqueous sodium bicarbonate, water, and saturated with aqueous sodium chloride and dried over - 156 - anhydrous sodium sulfate. after removing the benzene by vacuum evaporation, the ketone was distilled under reduced pressure from a Claisen flask. The fraction boiling at 88-93°/500.4. weighed 17.6 g. (32.5 per cent of theoretical for overall conversion). nz = 1.5239. Rgportea (16) ne = 1.5253. A somewhat purer sample prepared in 4 subsequent attempt boiled at 101=104,°/1500 AL with 1) 25.6 5 1.5223. D The ketone is a faintly yellowish oil with a slight butterlike odor. thoxybengyl Methyl Ketone 2,4-Dinitrophe razone Prepared according to the directions of Shriner and Fuson (7 9), 0.5 ge of the ketone yielded 0.66 g. of crude product which was recrystallized twice from absolute ethanol to give golden-crange needles mep. 108.6=109.5°. Reportea® MeDe 107,8-108.2°, A mixed melting point with a sample of the 2,4-—DNP of the ketone prepared by the method of Winstein 29) showed no depression (mp. 108.6-109.4°)« Found: Cy 55673 Hy 4083 Ny 16.2. In several instances when the 2,4-DNP's of the ketone samples were prepared by an alternative procedure in glacial acetic acid solution containing a few drops of concentrated hydrochloric acid, yellow needles melting at 198.7=200° were isolated in addition to the 2,4-dinitrophenylhydrazones. This compound, identified as acetyl~ 2, 4~dinitrophenylhydrazine, was also obtained by strain (40) during the preparation of 2,4-DNP's in alcoholic solutions containing sulfuric a Fort, Ae, and Roberts, J.D., to be published. and acetic acid. Strain reports a melting point of 199-201°, yd DL-O-Methyl-al pha»-methyltyrosine = was prepared from p-methoxybenzyl methyl ketone by the Strecker reaction in the seme manner as described for DL-alpha-methylphenylalanine, The initial reaction mixture con= tained, in addition to the usual components, ca. 1 ml. of methanol per gram of ketone employed. Following hydrolysis of the amino nitrile, the amino acid was precipitated by adjusting the aqueous solu- tion of its hydrochloride to pH 7 with sodium hydroxide. Nine days standing in the refrigerator was necessary to completely crystallize the amino acid. 56.6 g. of the ketone yielded 63.9 g. of crude amino acid. DL-alphastiethyltyrosine = To 12.6 g. of crude DL~-0-methyl-alpha~ methyltyrosine in a pressure bottle was added 50 ml. of a 3032 per cent solution of hydrobromic acid in glacial acetic acid solution, the bottle stoppered, and the mixture heated in a boiling water bath for one hour, After allowing the reaction mixture to cool and stand overnight, the excess acids were removed by evaporation in vacuo and the mixture was diluted to ca. 300 ml. with water. Upon cooling in an ice bath and adjusting the pH to ca. 7 with aqueous sodium hydroxide, a grayish precipitate of DL-alpha~methyltyrosine was formed. The dry weight of the crude material was 10.4 g. - 153 - N-Acety1~DI-al phasmethyl tyrosine ~ 10.0 g. of crude DL-alpha-methyl= tyrosine was acetylated with 13 ml. of acetic anhydride in a manner essentially the same as described for N-acety1-DL~alpha-methylphenylalanine. Upon acidification of the reaction mixture, 5.9 gs of the crude acetyl derivative was obtained. The acetyl derivative proved to be very difficult to crystallize from aqueous ethanol as it entered solution slowly and then tended to supersaturate. 1244 g. of smal) flat needles were obtained from ethanol-water, mep. 207.5-211.5° with decomposition, N-Aceotyl-DL~alpha«msthyltyrosine Methyl Ester - A solution of 1.37 ge of acetyl-DL-alpha-methyltyrosine in 20 ml. of absolute methanol was cooled in an ice-salt bath and saturated with anhydrous hydrogen chloride (Matheson). After warming to room temperature and standing for four days the mixture was again chilled and saturated with hydrogen chloride. After standing overnight at room temperature the volatile material was removed by vacuum evaporation and the remaining acid removed from the resulting brown syrup by vacuum dessication over sodium hydroxide. Upon dissolving the syrup in a little methanol, adding water, and cooling in an ice bath, 0.45 g. of crystals were obtained, Recrystallization from water containing a little ethanol gave fine flat prisms, m.p, 121-122°, : | -159 = N-Acety1-DL-O-methyl~alpha-methyltyrosine - 58.8 g. of the crude amino ' acid, acetylated with acetic anhydride in the usual manner, yielded 46.7 g. of the crude acetyl derivative melting at 176-178°. Recrystallization of 10 g. of the above acetyl derivative from aqueous ethanol gave 8.6 g. of flat prisms, mp. 178-181°. Anal. Caled. for Cj3H)70,N : C, 62.2; H, 6.8; N, 5.6. Acety1~DL~0~methy1—alpha-methyltyrosine Methyl Ester < A suspension of 10.0 g. of acetyl~DL~O~methyl-alpha~ methyltyrosine in 200 ml, of absolute methanol was cooled to 0-5° and saturated with anhydrous hydrogen chloride. After standing overnight, volatile material was removed by vacuum evaporation and remaining traces of water removed by repeated evaporation under benzene. The resulting dark syrup was again treated with methanolic hydrogen chloride and the benzene evaporation process repeated. Upon stirring the resulting syrup with aqueous methanol, 7.3 g. of a crystalline product was obtained. One recrystallization from aqueous ethanol gave 4.7 g. of fine stunted prisms, mp. 120.7~121.7°. Anal, Caled. for C) Hy g0,N : C, 63.45 Hy 7.25 N, 5.36 Found: C, 63.2; H, 7.23 N, 5.3. po-Methoxybenzyl isopropyl Ketone ~ Via the Darzens glycidic ester condensation. Equimolar amounts of anisaldehyde, ethyl alpha~bromo- isovalerate (Matheson), and sodium ethoxide were reacted in a manner - 160 = entirely similar to that described for the preparation of the methyl ketone (see page 157). In this case, 135 g. of anisaldehyde and 207. 5 g. of the bromo ester yielded 147.6 ge (56.4 per cent of theoretical) of the glycidic ester, bsp. 12345=127°/500-600 4 | Saponification of the glycidic ester gave 112.2 g. of the sodium salt (77.8 per cent on the basis of the glycidic ester). The sodium glycidate was acidified and decarboxylated to give 67.9 ge (81.3 per cent conversion of the sodium salt; 45.8 per cent for the overall process) of p-methoxybenzyl isopropyl ketone, b.p. 95~96°/ 55044 n2 = 1.5115. The ketone is a pale yellow, nearly colorless liquid having a delicate anise odor, 5-isopropy1-5' ~(p-methoxybenzy1)-hydantoin- Adapted from the procedure of Henze and speer (3) To a suspension of 13.2 g. of ammonium carbonate and 2.8 g. of sodium cyanide (powdered egg) in a mixture of 46 ml. of water and 10 ml. of 95 per cent ethanol, contained in a 300-ml. RB flask, was added a solution of 4.7 g. of p-methoxybenzyl isopropyl ketone in 36 mi. of 95 per cent ethanol, ‘The resulting mixture was heated for two hours in a water bath maintained at 60° + 2°. Upon cooling, 50 ml, of water was added to the yellow reaction mixture whereupon a certain amount of unreacted ketone separated out as an oil, 2.0 g. of potassium cyanide (Mallinckrodt AR) and 50 ml. of 95 per cent ethanol were then added and the heating resumed and continued for two hours. Removal of ethanol by vacuum evaporation - 161 - eaused the separation of the hydantoin as a white solid. The crude - product weighed 4.74 ge (14, per cent of theoretical). 5-isopropyl- 51=(p-methoxybenzyl)=hydantoin crystallized from water-ethanol in bunches of colorless needles. mp. 218.8°~221°, Attempted Conversion to N-Acetyl-O-methyl~alpha-isopropyltyrosine ~ 4eO0 ge Of crude 5<isopropyl=5'=(p-methoxybenzyl)-hydantoin was heated under reflux with a solution of 20 g. of barium hydroxide in 40 ml. of water for 52 hours, The mixture was cooled and an excess of dilute sulfuric acid added to precipitate barium sulfate. The latter was allowed to precipitate over ea 40 minute period at ca. 70° and was then removed by suction filtration. When, after bringing the filtrate to pH 5-7 and allowing to stand, it appeared that the amino acid would not crystallize it was decided to recover it by converting it into its acetyl derivative, The solution was reduced in volume to 100 ml., made basic, cooled in an ice bath, and treated with 7 ml. of acetic anhydride under basic con= ditions in the usual manner. The reaction mixture was filtered and the filtrate brought to pH 3 and stored in the refrigerator, When, on standing, the acidic filtrate failed to yield crystale it was evaporated to dryness, combined with the insoluble residue formed during the acetylation, and extracted with hot acetone in a Soxhlet extractor. Evaporation of acetone left a solid which was recrystallized 3 times - 162 = from aqueous ethanol to give colorless blades, mp. 218.2-220.1. Mixed mep. with 5=isopropyl=5' —(p-methoxybengy1)-hydantoin 218.7=220.2°. Apparently, unreacted hydantoin has been isolated. Ethyl isopropyleyanoacetate = Procedure adapted from the method of 41) Hessler and Henderson ‘4 To a solution of 12.2 g. of sodium metal in 500 ml. of absolute ethanol, contained in a 3000-ml. 3eneck flask equipped with reflux condenser and mechanical stirrer, was added, rapidly and with stirring, 5909 ge of ethyl cyanoacetate (Matheson), To the resulting mixture, isopropyl iodide (Matheson) was added dropwise, with continued stirring, over a one hour period, The reaction mixture was then heated under reflux for ca. two hours until it had become ca, pH 7 at which time 350 ml. of ethanol was removed from the cherry-red mixture by distillation. After removing the remaining ethanol by vacuum evaporation, the reaction mixture was equilibrated with ether and water and the ether layer separated and shaken, first with dilute aqueous sodium thiosulfate, then with saturated aqueous sodium chloride and dried over anhydrous sodium sulfate. The ether was then removed by vacuum evaporation and ethyl isopropylcyanoacetate distilled through a short Claisen colum packed with Raschig rings. The fraction collected at 68~78°/2200-3100.U weighed 48.9 g- (60 per cent theoretical). = 162 = from aqueous ethanol to give colorless blades, mp. 218.2-220.1. Mixed mp. with 5eisopropyl~5!' —(p-methoxybenzyl)<-hydantoin 218.7=220.2°,. Apparently, unreacted hydantoin has been isolated. Ethyl isopropylcyanoacetate - Procedure adapted from the method of Hessler and Henderson (42) To a solution of 12.2 g. of sodium metal in 500 ml. of absolute ethanol, contained in a 3000-ml. 3«neck flask equipped with reflux condenser and mechanical stirrer, was added, rapidly and with stirring, 59.9 g. of ethyl cyanoacetate (Matheson), To the resulting mixture, isopropyl iodide (Matheson) was added dropwise, with continued stirring, over a one hour period, The reaction mixture was then heated under reflux for ca. two hours until it had become ca, pH 7 at which time 350 ml. of ethanol was removed from the cherry-red mixture by distillation. After removing the remaining ethanol by vacuum evaporation, the reaction mixture was equilibrated with ether and water and the ether layer separated and shaken, first with dilute aqueous sodium thiosulfate, then with saturated aqueous sodium chloride and dried over anhydrous sodium sulfate. The ether was then removed by vacuum evaporation and ethyl isopropylicyanoacetate distilled through a short Claisen colum packed with Raschig rings. The fraction collected at 68~78°/2200-3100 4 weighed 48.9 g. (60 per cent theoretical), ~- 163 = Ethyl isopropyl=-benzyleyanoacetate - To a solution of 5.9 g. of “sodium metal in 370 ml, of dry absolute ethanol, contained in a 1000-m1, 3eneck flask equipped with reflux condenser and stirring motor was added 39.7 ge of ethyl isopropylcyanoacetate over a ten minute period. 36.7 g. of benzyl chloride was then added with stirring over a 15-20 minute period, As the stirring continued the reaction mixture began to reflux gently and after one and three~ quarter hours, when the action had begun to slacken, heat was applied and the mixture stirred under reflux for an additional four hours, 300 ml, of ethanol was then removed by distillation and, after cooling, the reaction mixture was shaken with ether and ice water. The ether layer was dried, first over anhydrous sodium sulfate then with anhydrous calcium chloride and the ether removed by evaporation in vacuo. Ethyl isopropyl-benzylcyanoacetate was distilled through a 30 em. Claisen column without packing. The fraction collected boiled at 130-136°/1300-1500.4 and weighed 43.3 g. (69 per cent of theoretical). Reaction of Ethyl Isopropy]~Benzylcyanoacetate with Hydragine Hydrate A mixture of 43.3 g. of ethyl isopropyl~benzyleyanoacetate, 12.2 g. of 100 per cent hydrazine hydrate (Matheson), and 150 ml. of n-hexanol was heated under reflux for 5 days. At the end of this time the reaction mixture was reduced to a small volume by vacuum evapora~ tion and placed in the refrigerator. Upon standing in the cold for 12 days a small amount of a white precipitate formed and was collected by suction filtration. This substance was amphoteric in behavior, soluble in both acid and base and could be precipitated out of aqueous solution at pH 6-7. The material was dissolved in 3N HCl and the solution extracted with ether to remove the suspended high boiling liquids. The acidic aqueous phase was then filtered and the pH adjusted to ca. 6 with aqueous sodium hydroxide. After standing overnight in the ice box, the amphoteric material was collected and dried over sodium hydroxide in vacuo. The compound melted at 149.2-150.2°. Anal. Found: C, 67.6 3; H, 7.6 3; N, 12.5. Ethyl isopropyl~p-methoxybenzylcyanoacetate— 48.9 g. of ethyl isopropylcyanoacetate was converted to its sodium derivative with a solution of 7.25 g. of sodium metal in 400 ml. of dry absolute ethanol, and reacted with 49.4 g. of p-methoxybenzyl chloride under conditions identical with those described for the preparation of ethyl isopropyl-benzylcyanoacetate. Following removal of ether = 165 = the ethyl isopropyl~p-methoxybenzylcyanoacetate was distilled, col- lecting 52.7 g. (60.8 per cent of theoretical) over the range 138.5-146 / 300-4004. "Anhydrous Hydrazine- Prepared according to the directions of audrieth‘42), 200 g. of 100 per cent hydrazine hydrate was stored with 200 g. of sodium hydroxide (C. P. pellets) under a nitrogen atmosphere for a period of 7 days. At this time the mixture, which had become a solid mass, was warmed in a water bath causing it to melt and form two liquid phases. The hydrazine rich upper phase was carefully decanted and distilled under a nitrogen atmosphere at reduced pressure. The most volatile material, boiling over a one degree range, was collected and used as the hydrazine preparation. A second fraction boiling over aca. 7 degree range was collected, treated with sodium hydroxide as before, and fractionated to give an additional batch of nearly anhydrous hydrazine. Isopropyl-p-methoxybenzylcyanoacethydrazide- 27.5 g. of ethyl isopropyl~p-methoxybenzylcyanoacetate and 6.4 g. of anhydrous hydra-~ zine were placed in a 100-ml. 3-neck flask equipped with reflux con- denser and nitrogen inlet tube. The mixture was heated in an oil bath, initially maintained at 145-155°, while passing a slow stream of nitrogen through the system. After heating for ca. 2 hours, the mixture was allowed to cool, 4.0 g. of anhydrous hydrazine and 30-40 ml. of n-hexanol added, and the heating resumed. The mixture was = 166 - heated intermittently for a period of approximately 8 days at a bath temperature of 120-170°, cooling from time to time to add small portions of anhydrous hydrazine and sweep the system out with nitro- gen (the continuous nitrogen stream was abandoned since it tended to carry hydrazine out of the reflux condenser). Over the & day period, ca, 10 g. of additional hydrazine was added. Finally the bulk of the volatile material was removed by evaporation in vacuo and the oily residue stored overnight in the refrigerator. A white solid separated out and was collected by suction filtration, washed with ether and dried. The crude product thus obtained weighed 5.0 g. and was fairly water soluble. The solid behaved in a peculiar manner upon heating in a capillary tube, clumping (without becoming moist) at 110° and evolving a volatile substance and finally melting over the range 147.5-161°, Upon reerystallizing from 95 per cent ethanol 4.8 g. (18.4 per cent of theoretical) of material was obtained, mp. 176.5-180.9°, without the preliminary gas evolution. Anal. Caled. for Cy,Hi9g02 Ng: Cy, 64.4 ; Hy, 7.3 3 N, 16.1. Found: C, 63.5 3; H, 7.33 N, 16.7. Attempted Curtius Rearrangement of Isopropyl-p-methoxybenzyl— cyanoacethydrazide A solution of 4.8 g. of the hydrazide in 5 ml. of concen- trated hydrochloric acid and 50 ml. of water was placed in a 200-ml. 3-neck flask equipped with mechanical stirrer, thermometer, and dropping - 167 - ‘funnel. 40 ml. of ether was then added, and the two phase system cooled to 0~5° in an ice bath. With vigorous stirring, a solution of 1,28 g. of sodium nitrite in a few ml. of water was run into the mixture over a period of 5 minutes. The temperature of the mixture increased by ca. I, and a solid separated out. As soon as the addi-~ tion of the nitrite was completed the stirring was ceased, and the yellow ether layer separated. The aqueous phase was extracted with ether and this extract combined with the original ether layer. The ether extract was washed with cold aqueous sodium bicarbonate, dried over anhydrous magnesium sulfate and added to a 300 ml. RB flask containing 50 ml. of absolute ethanol. A distilling head and con- denser were attached to the flask and the ether was removed from the mixture by distillation. 100 ml. of absolute ethanol was then added and the solution heated under reflux for one hour. Finally, the ethanol was removed from the residue by vacuum evaporation. 100 ml. of concentrated hydrochloric acid was added to the residue and the mixture heated under reflux, After 15 minutes, a dark oil had begun to separate out. After refluxing for ca. one hour, the mixture was cooled in an ice bath, saturated with anhydrous hydrogen chloride, and heating under reflux recommenced and continued for two and one-half hours. Upon reducing the volume of the reaction mixture and adjusting the pH to 7 a precipitate began to form, After standing in the cold room overnight the precipitate was collected by suction and dried. The yield of the crude material was extremely small. - 168 - The solid material appearing during the addition of the nitrite solution weighed 1.1 g. and is presumably the symmetrical dihydrazide. Gramine- Prepared according to the method of Kuhn and stein(43) , A mixture of 60.8 g. of 33 per cent aqueous dimethylamine and 62.4 g. of glacial acetic acid was chilled to 0-5°C and added to 46.5 g. of indole (Eastman) contained in a 2000-ml. beaker. As the reaction mixture became warm, cooling was applied by a water bath to keep the temperature below 45°. The clear light yellow mixture was allowed to stand at room temperature for one hour and then made strongly basic with aqueous sodium hydroxide causing the gramine to precipitate. The product was collected by suction, washed thoroughly with water, and dried over sodium hydroxide in vacuo. The crude material weighed 63.0 g. Ethyl methylcyanoacetate- 103.0 g. of ethyl cyanoacetate (Matheson) was converted into the sodium salt by treatment with a solution of 20.9 g. of sodium metal in 1100 ml. of dry absolute methanol and alkylated with 142.2 g. of methyl iodide in the same manner as previously described for isopropyleyanoacetic ester. In this case, however, the reaction was run entirely at room temperature, cooling in early stages with a water bath. As in the previously described alkylation with isopropyl iodide it was important to remove the iodine from the organic phase prior to distillation by washing with aqueous sodium thiosulfate. The ethyl methylcyanoacetate was collected over the range 43-50°/ 100044 and weighed 76.0 g. (65.7 per cent of theoretical). Condensation of Gramine with Ethyl methylcyanoacetate 17.4 ge of gramine and 12.7 g. of ethyl methylcyanoacetate were placed in a 200-ml. RB flask equipped with a side arm. From a one-hole rubber stopper in the mouth of the flask a pyrex tube led to a simple trap consisting of a 250-ml. Erlenmeyer flask equipped with a 2-hole rubber stopper. From the trap a similar tube led to a third flask where it terminated below the surface of a solution of picric acid in benzene. A slow stream of nitrogen was started through the apparatus and the reaction mixture was heated by an oil bath main- tained at 130-160°, The reaction mixture liquefied and as the reac- tion proceeded the liberated dimethylamine was swept into the picric acid solution where it was converted instantly into its yellow crystal~ line picrate, The heating was continued for ca. 3 hours after which the reaction mixture was allowed to cool and the dimethylamine picrate collected and dried in vacuo. The weight of the picrate, 16.9 Be, corresponded to 61.6 per cent conversion of the gramine. The cold reaction mixture was taken up in ether and extracted with cold dilute hydrochloric acid to remove unreacted gramine and then with dilute aqueous potassium carbonate and water. After removing the ether by evaporation, unreacted ethyl methylcyanoacetate (1.6 g.) was removed by distillation at reduced pressure, The product, pre- sumably ethyl 3-indolylmethyl-methylcyanoacetate, remained as a dark viscous syrup. Methyleyanoacethydrazide - To a solution of 35.4 g. of ethyl methylcyanoacetate in 100 ml. of n-amyl alcohol, contained in a 300-ml, - 170 = RB flask equipped with reflux condenser, was added 16 ml. of anhydrous hydrazine and the system swept out with nitrogen. The mixture was heated under reflux for ca. four and one-half hours and at the end of this time the excess solvent was removed by evaporation in vacuo. Upon cooling the residue, the hydrazide separated as a dark solid and was collected by suction filtration, washed with ether, and dried in vacuo. Remaining traces of solvent were removed from the hydrazide after several days storage in vacuo by grinding it to a fine powder and washing it repeatedly with ether. The dry dark red crude product weighed 29.5 g. Condensation of Gramine with Methylcyanoacethydrazide 20.3 ge of gramine and 13.2 g. of crude methylcyano- acethydrazide were ground together in a mortar and the mixture placed in a 250-ml. wide mouth Erlenmeyer flask. Upon heating the mixture in an oil bath maintained at 135°+ 3°, dimethylamine was vigorously evolved and the heating was continued for two and one~ half hours until the evolution of dimethylamine had virtually ceased. The cool solidified mixture was ground to a powder and washed several times with water to remove unreacted methylcyanoacethydrazide, then with cold dilute hydrochloric acid to remove gramine. Finally it was washed with aqueous sodium bicarbonate and water and dried in vacuo. The crude product, presumed to be 3-indolylmethyl~methy1- cyanoacethydrazide, weighed 21.0 g. - 171 - Attempted Curtius Rearrangement of 3-Indolylmethyl- methylcyanoacethydrazide To a solution of 9.13 g. of the crude hydrazide in 50 ml. of cold glacial acetic acid, contained in a 500-ml. 3-neck flask equipped with dropping funnel and mechanical stirrer, was added a cold mixture of 6.7 ml. of concentrated hydrochloric acid and 50 ml. of water, Addition of this solution caused at least partial pre- cipitation of the hydrazide. Immediately following this, a solution of 2.76 g. of sodium nitrite in 7 ml. of water was rapidly added. A tarry substance separated and the reaction mixture was extracted with chloroform and the chloroform extract washed with aqueous sodium bicarbonate and dried over anhydrous magnesium sulfate. The extract was then added to 200 ml. of ethanol and the mixture refluxed for six and one-half hours, Removal of the volatile material left a dark red syrupy substance, The tar originally formed in the reaction mixture had been observed to evolve a gas and change into a porous solid. The original aqueous acidic phase, containing the tar, was made neutral and heated under reflux for 4-5 hours. The porous solid, resembling peat moss, was collected, washed with water, and dried. This substance weighed 7.33 ge and appears to be the major reaction product. Neither the porous product, nor the red syrup obtained from the chloroform phase after the ethanol treatment was identified. @- 172 = Action of Chymotrypsin upon N-Acetyl-DL-alpha- methylphenylalanine Methyl Ester I- 0.0235 g. of N-acety1~DL-alpha~methyl phenylalanine methyl ester and 0.020 g. of al pha-chymotrypsin (Armour lot no. 10705) were incubated for a period of one hour at 25° in 10 ml. of aqueous solution. A formol titration® with aqueous 0.01 F sodium hydroxide indicated that approximately 14.5 per cent of the total ester had been hydrolyzed in this time interval. II- 7.08 g. of N~acety1-DL~alpha~methylphenylalanine methyl ester was dissolved in ca. 500 ml. of 30 per cent aqueous methanol and treated with 50 mg. of alpha-chymotrypsin and a small amount of "Viobin" paste at 30-35°. Following a small initial drop in pH upon the addition of the enzyme, no significant additional hydrolysis occurred over a 6-hour period, The methanol was then removed from the reaction mixture by evaporation and the attempt at resolution continued in 1600 ml. of aqueous solution. 8 days of treatment with "Viobin", added at intervals, caused no apparent hydrolysis beyond that accounted for by the autolysis of the enzyme. Attempted Papain Catalyzed Synthesis of N-Acetyl-L-alpha~ methyltyrosine-p—toluidide- The method employed was that of Huang (At) and Niemann . To a solution of 0.73 g. of N-acety1—-DL-alpha-methyl~ tyrosine in 28 ml. of 0.5 M acetic acid-0.5 M sodium acetate buffer was added 0.037 g. of L-cysteine hydrochloride and 0.33 g. of &® See Section I, experimental, of this thesis. ~ 173 = p-toluidine (Merck). The resulting solution was warmed to ca. 40°, 0.75 g. of an aqueous papain solution! 44) added, and the mixture “incubated in an oven maintained at 40°, After 7 days of storage at this temperature no significant amount of solid material had been formed. As a check, a control experiment under the above conditions was performed upon N-acetyl-DL-phenylalanine. After standing one day at 40° a erystalline precipitate of N-acetyl~L-phenylalanine—p~ toluidide was clearly evident. Attempted "Viobin" Resolution of Q-Methyl-DL-alpha- methyltyrosine Methyl Ester 2.73 g. of O-methyl-DL-alpha-methyltyrosine methyl ester, 0.2 g. of "Viobin", and 0.22 g. of water were mixed into a paste, stoppered in a 50-ml. flask, and stored at room temperature. After standing for 5 days, the mixture was apparently unchanged and tested pH 7 with indicator paper, Attempted Chymotryptic Resolution of N-Acety1-DL-0~ methyl-alpha-methyltyrosine Methyl Ester I- 2.65 g. of the DL ester was partially dissolved in a mix- ture of 45 ml. of methanol and 150 ml. of water, the mixture adjusted to pH 8, and treated with 100 ml. of alpha-chymotrypsin (Armour Lot “no. 10705) at 30°. No significant amount of hydrolysis occurred on 12-15 hours standing. ~ 174 < A solution of 1.0 g. of papain and 0.3 g. of L-cysteine hydrochloride in 5 ml. of water was added to the reaction mixture and the pH adjusted to 5.5. Five hours of incubation at 25-30° failed to effect any resolution. II~ 0.25 g. of the DL ester and 0.25 g. of Armour alpha~ chymotrypsin, in 78 ml. of aqueous solution, were incubated over-— night at room temperature. No decrease in the pH of the reaction mixture was observed to occur. Attempted Resolution of N-Acetyl-DL-O-methyl~alpha~ methyltyrosine by Fractional Crystallization of the L-Brucine Salts The method of Sealock(4>) was employed. 10.0 g. of N~acety1~DL~O-methyl—alpha-methyltyrosine and 15.7 g- of I-brucine were dissolved in 90 ml. of boiling absolute ethanol and the resulting solution filtered. The filtrate was then stored in the refrigerator. No crystallization had occurred after standing for four days. Nor did replacing the ethanol solvent by ethanol-ether, ethanol-water, or butanol-chloroform cause the crystallization of any material. (2) (2) » (3) (4) (5) (6) (7) (8) (9) (10) (a2) (12) (13) (14) (15) (16) (17) (18) (19) — (20) - 175 = ' REFERENCES Herbst, R.M., and Johnson, T.B., J.A.C.S. 5h, 2463 (1932). Jawelow, J., Ber. 39, 1199 (1906). Henze, H.R., and Speer, R.J., J.A.C.S. 6h, 522 (1942). Bucherer, H., and Lieb, V., J. Prak. Chem. 141, 5 (1934). Lin, K., Li, L., and Huang, Y., Sci. and Tech. China 1, 5 (1948). Chem. Abst. 43, 2166 (1949). Balbiano, L., Atti accad. Lincei, 17, II, 259 (1908); 20, II, 239 (1911); 22, II, 93 (1913). Tiffeneau, M., and Daufresne, M., Comp. rend. 144, 1354-56 (1907). Balbiano, L., and Paolini, V., Gazz. 36, I, 293 (1906). Hoering, P., Ber. 38, 2296, 3480 (1905). Auwers, V., Pdtz, H., and Noll, W., Ann. 535, 219 (1938). Giuliano, R., Ann. Chim. Appl. 30, 476 (1940). Wallach, 0., and Pond, F.J., Ber. 28, 2714 (1895). Tiffeneau, M., and Daufresne, M., Comp. rend. 144, 924 (1907). Béhal, A., and Tiffeneau, M., Comp. rend. 141, 596 (1905). Sosa, A., Ann, Chim. 14, 5 (1940). Wallach, 0., and Muller, H., Ann. 332, 323 (1904). Hoover, F.W., and Hass, H.B., J. Org. Chem. 12, 501 (1947). Winstein, S., et al, J.A.C.S. 74, 1140 (1952). Bougault, Ann. Chim. 7, 25, 518 (1902). Forneau, E., and Tiffeneau, M., Comp. rend. 141, 662 (1905). (21) (22) - (23) (21,) (25) © (26) (27) (28) (29) (30) (32) (32) (33) (3h) (35) (36) (37) ~ 176 = Fuson, N., Josien, M., and Shelton, E.M., J.A.C.S. 76, 2526 (1954). Mason, J.P., and Terry, L.I., J.A.C.S. 62, 1622 (1940). Darzens, G., Comp. rend. 142, 214 (1906). Ramart-Lucas, M., and Guerlain, J., Bull. soc. Chim. 49, 1860 (1931). Rondou, A., Bull. Soc. Chim. Belg. 31, 231 (1922). von Meyer, J. Prak. Chem. (2) 62, 81-117 (1895). Baerts, F., Bull. Soc. Chim, Belg. 31, 184 (1922). Ivanov, D., and Paounoff, I., Comp. rend. 197, 923 (1933). Fuson, R.C., "Advanced Organic Chemistry", Wiley, New York, 1950. Shriner, R.L., and Turner, T.A., J.A.C.S. 52, 1267 (1930). Shriner, R.L., and Hull, C.J., J. Org. Chem. 10, 228 (1945). Org. Synthesis. Coll, Vol. I, 107 (1941). Darapsky, A., and Steuernagel, E., J. Prak. Chem. 146, 250 (1936). Gagnon, P.E., Gaudry, R., and King, F.E., J. Chem. Soc. 13, (1944). Brown, J.B., Henbest, H.B., and Jones, E.R.H., J. Chem. Soc. 3172 (1952). Ingersoll, A.W., et al, J.A.C.S. 73, 3363 (1951). Heilbron, "Dictionary of Organic Compounds", Oxford, 1953. Cain, J.C,, Simonsen, J.L., and Smith, C., J. Chem. Soc. 103, 1035 (1913). (38) (39) (40) (41) (42) (43) (44) (45) -177 = Lucas, H.J., and Pressman, D., "Principles and Practice in Organic Chemistry", Wiley, New York, 1949. Shriner, R.L., and Fuson, R.C., "The Systematic Identification of Organic Compounds", Wiley, New York, 1948. Strain, H.H., J.A.C.S. 57, 758 (1938). Hessler, J.C., and Henderson, W.F., J.A.C.S. 43, 674 (1921). Audrieth, L.F., and Ogg, B.A., "The Chemistry of Hydrazine", Wiley, New York, 1951. Kuhn, H., and Stein, 0., Ber. 70B, 567 (1937). Huang, H.T., and Niemann, C., J.A.C.S. 73, 475 (1951). Sealock, R.R., J. Biol. Chem. 166, 1 (1946). ~ 178 = Propositions 1. 4lphawamino nitriles undergo decomposition in the presence of base to give a carbonyl compound having one less carbon atom, Since alpha-amino acids may be converted into their corresponding alpha~amino nitriles by conversion first to the Nephthaloyl derivatives, reaction with thionyl chloride followed by ammonia to give the amides, dehydration of these to the nitriles by phosphorus pentoxide, and finally removal of phthaloyl with hydrazine, I propose the use of this reaction sequence as a method of degradation of alpha-amino acids into their corresponding carbonyl products of one less carbon atoms. Lucas, HeJes and Pressman, D., "Principles and Practice in Organic Chemistry", Wiley, New York, 1950. Sheehan, J.C., and Franky VsSey JeAcCeS. 71, 1856 (1949). 26 fryptamine has been suggested as a possible intermediate in the conversion of tryptophane to IAA in plants. Recently, Jones et al, have isolated 3-indoleacetonitrile from cabbage leaves, Since the nitrile may arise from the same imino precursor proposed for 3-indole=- acetaldehyde and IAA, I propose that the occurrence of the nitrile in plents strengthens the hypothesis that IAA may arise from tryptophane via tryptamine, Bonner, Je, "Plant Biochemistry", Academic Press, New York, 1950. Jones, EeReH.,y ot ale, Nature, 169, 485 (1952). - 179 = 3. Bergmann and Stern have proposed the following mechanism for . the dehydrohalogenation of an alphs-(alpha'-haloacyl)-amino acid by R-CHar CH-CO2H = RCH =o _ _ acetic anhydride. Fee Acso TT Rvp — Fe / ——> _~+N fe) Co-CHR PYAR. NA va CHR CHaR I propose that insight into the intermediates involved in this reaction might be gained by treating the following types of compounds with acetic anhydride and pyridine, CHs (2) € Nea-cox and (2) € \-ct-€ com NH COCHLX NHCOCH2X Bergmann, M., and Stern, Fe, Ann. 448, 20 (1926). he BD alphe-acylamino acid esters and amides have been considered to combine with alpha-chymotrypsin in the same mode as do their ZL enantiomorphs. This implies that the hydrolyzable bond of a D ester or amide makes a potentially hydrolytic interaction in the EI complex. I propose that this hypothesis be further investigated by studying, as possible substrates of chymotrypsin, D and L alpha-acyl amino acid esters substituted at the alpha-carbon by strongly electron<attracting Broupse Huang, H.T. and Niemann, Coy JeAcl.S. Thy 4634 (1952). 5. Wagenknecht and Burris report that IAA is oxidatively degraded by IAA oxidase, releasing one molecule of carbon dioxide per molecule of oxygen taken up whereas 3—indolepropionic acid releases only - 180 ~ one-quarter mole of carbon dioxide per mole of oxygen. While these authors maintain that the oxidation is limited to the side chain in | IAA, I propose that the above observations are in better agreement with an opening of the indole nucleus to give, in the case of IAA, a beta-keto acid. Wagenkmecht, A.C.» and Burris, ReH., Arch. Biochem. 25, 30 (1950). 6. ‘Two principal mechanisms have been advanced for hemolytic aromatic substitution. (I) Rt wna ——~> R R' +H (II)}-A Re + at \a __ < \e + PH wm a Opa — ne In the case of R = CH, , pathway II would preceed with formation of methane. I ‘propose that a measurement of such evolved hydrocarbon night be used to estimate the relative contributions of the two mechanisms, The formation of methane by coupling of R- and H- could be eliminated by the proper experimental conditions, Hey, DeHes Je Chem, Soc, 1974 (1952). Johnson, G.ReAe,y Stein, Ge, and Weiss, Jey J. Chem. Soc. 3275 (1951). Te Nitrobenzene, chlorobenzene, and phenol have been hydrozylated in aqueous solution by either the action of lonizing radiations or by an acidic ferric ion-hydrogen peroxide solution. The isomer dis=« tributions resulting from these hydroxylations have been found to - 181 « depend upon (1) the method of hydroxylation, and (2) the pH of the reaction mixture. I propose that this behavior is inconsistent with the simple free radical mechanism accepted by these authors, Loeb, Hey Stein, G., and Weiss, J., J. Chem. Soc. 2074 (1949); | 2704 (1950). Johnson, G.R.A., Stein, G., and Weiss, J., J. Chem, Soc. 3275 (1951). Stein, Gey and Weiss, Je, J. Chem. Soc. 3265 (1951). | &. I propose that the withdrawal symptoms characteristic of narcotic habituation may be the result of an adaptive enzyme mechanism. 9. Gramine may be condensed with methyleyanoacethydrazide and ethyl methylcyanoacetate to give the corresponding 3~indolylimethyl substituted compounds, I propose that an attempt be made to condense gramine with the 4-alkyl derivatives of 2=(p-nitrophenyl)-oxazolone-5 as a direct route to the DLealpha-alkyltryptophanes, The possibility of generalizing the use of these oxazolones to prepare other alpha= alkyl amino acids should also be investigated, This Thesis, pages 137-138. 10. Sheehan and Frank, Barber and Wragg, and Ing and Manske give the following structure for phthalhydrazide [o) i ‘ NH - 182 = I propose that the acidic nature of this compound as evidenced by - its solubility in dilute aqueous base and by its ability to form a salt with n-propylamine are in better accord with the tautomeric structure Ov “TY we aN H Moreover, the formation of such a resonance stabilized compound would account for the great tendency for hydrazine to remove the phthaloyl group from phthalimido acid type compounds, Ing, HeR., and Manske, ReHeFey Je Chem, Soc. 2348 (1926). Barber, HeoJ., and Wragg, W.R., Nature 158, 514 (1946).