3 c y s a ' le r treOlmenl wltn 535-3cvc

Fig. 1.---Radioactivity (CPl I ) of l>lood, urine arid feces crf rat treated intraperitoneally with S35-I>CVC.

essentially the saiiie iii botli rats. In coiitrast, tlic total urinary excretion of S15 by the orally treated rat was, during the first 24 aiid 72-hour intervals, only 40% of that excreted by the pareiiterally treated animal. The radioactive coinpoileiits iii tlie urine included in- organic sulfate, DCYC and its S-acetyl derivative as reported elsewhere.13 The fact that in spite of a high rate of clearance of I3CI7C through the kidneys, excre- tion of the S35 isotope coiitiiiued for a long period i l l -

dicates that DCT-C or a sulfur conipound derived from it combined with c o i n p o i w i l t s of tissues and thus es- caped the initial rapid clearaiicc froin the blood streaiii.

Experimectal

S3j-Benzyl Mercaptan (I).-Tlie proc,ediire of Tarver :mi Schmidt" was adapted for use of Bas35 (30 mc.)l* and scaled down to obtain preparations of high specific activity. The reac- tions were carried out under a slow stream of nitrogen in a closed system from which emerging gas passed thiough a solution of ethanolic 2y0 HgCh.

S3j-Benzyl-~-cysteine (III).-l-rethyl 2-aniin11-3-i~hloro-~-pro- pionate was prepared according to Fischcr and RaskeI3 except that diethyl ether was suhstituted for petroleum Ether to tvasli the product. This ester (800 mg.) wva6 hydrolyzed with 135 nil. of 5.8 N HCI and the solution concentrated to dryness in uucuo. The residue of 2-amino-3-chloro-1.-propionic acid hydrochloride (11) was dissolved in 5 ml. of absolute ethanol and injected with a long needle and syringe through the vented stopper into t'he flask containing S35-benzyl mercaptan ( I ) . I 4 The mixture \vas stirred by a continuous flow of nitrogen which escaped t,hrough tho vent into a 2yo.ethanolic solution of HgC1,. Sodium ethoxide (0 .5 g. of Na/lO ml. of absolute ethanol) as added dropwise until the thymolphthalein indicator turned blue and 0.4 nil. more was added. The flask \vas then heated a t 72" for '2 hr. and more sodium ethoside and/or ethanol added as needed. Aftcr cooling, the precipitates of YaC1 and KCl were centrifuged, the supernatant removed, and the precipitate washed 3 times \vith ahsolute ethanol. The supernatant and washings were evaporated

(10) For identification of tlir riiajor S"'-i:ont,zinine cornpounds as inorganic, siilfatc. DC\'TC and K-acetyl IICVC, pabier clirninatography. paper electro- Iihoresis and dilution analysis of tlic eluted fractions were used. I n addition, inorganic sulfate was identified t11ri)uaIi preripitation as barium sulfate and henxidine sulfate. The 2,Pdinitrolitienyl derivative of DCVC-SaS from urine w a s identical with an authentic sperinien (rn.11. 160°). Authentic specimens of N-:icetyl DCVC (rn.p. 108-109°) and its m-toluidide (rn.p. 1,j0-161°) sr . rv~i1 to identify N-arrtyl-IICVC-S36 from urine. Details of the methods i i a c d are described elseahere (R. F. Ilerr , Pti.11. Thesis, University of Min- nesota, 1'360; R. F. Derr and &I. 0 . Scliultze, Biochem. Pharmaco?.. in press).

(11) H. l 'arrer and C . L. A . Schmidt. J . Bid. Chem., 130, 67 (1939). 112) Purchased from Union CarbideNiiclear Company, Oak Ridge, Tenri.,

(1:O E. Fisclier and K. Raske, Bel. . , 40, 3717 rl'd07). ( 1 4 ) J. L. Woodand I.. Van Middlesworth, J . Bioi. Chem., 179, 529 (1'34'3J,

i i r i i l r r .\KC I k e n s e 22-187-1.'3.

N - ( y - I,- Glu tamy1)aminobenzoic Acids

Of a group of ?;-~pheii~lalk~l)gl~itamiiics T\ Iiicli ~vcrt' recently prepared arid c-amiiicd for antimicrohial activity, S-benzyl-L-glutamiile n as ail effectiw uiitilg- onist of the natural amiiio acid.? In additioll, 0tht.r

i 1 ) Predoctoral l'rilow (C1'-10,027) Sutiunal Carlc?r insti tute, Vrlited States Public Health Service.

March, 1963

y-glutamyl derivatives have been isolated either from natural source^^,^ or prepared synthetically and found to inhibit the growth of microOrgani~ms.5~~ In an effort to study the effect of the presence of moderately reactive groupings in augmenting the toxicity of this type of metabolite analog, a series of S-phenyl derivatives of qlutamine were prepared which contain an 0-, m- and p-carboxyl group on the aromatic nucleus. In addition, because of the presence of the naturally occurring o- and p-aminobenzoic acid moieties in the corresponding glutamine condensation products, these two analogs were also examined as potential intermediary metabo- lites in a number of assay systems. This was further suggested by the fact that glutamine contributes a nitrogen in the biosynthesis of both anthranilic acid and p-aminobenzoic acid from shikimic acid.'

As indicated in the equations, the syntheses of the K-( y-L-glutamy1)aminobenzoic acids were accomplished through a series of reactions involving the initial con- densation of K-carbobenzoxy- glutamic anhydride with the appropriate methyl aminobenzoate to give the corresponding methyl ester of N-carbobenzoxy-y-L- glutamylaminobenzoic acid. In order to obtain an acceptable yield of the meta derivative, it was necessary to extend the reaction time for this isomer approxi- mately eight-fold over that required for the ortho and para analogs. Hydrogenolysis of the condensation products yielded the corresponding 7-L-glutamylamino- benzoic acid methyl esters which were then hydrolyzed with barium hydroxide to form the desired products. Using an ascending paper chromatographic technique, and developing the chromatograrlls with ninhydrin reagent, each of the products was demonstrated to be chromatographically pure in two different solvent systems.

Since the interaction of N-carbobenzoxyglutamic anhydride with the various aminobenzoic acids could yield the amide linkage through either of the two car- boxyl groupings in the glutamyl moiety, the chemical structures of these analogs were subsequently demon- strated through the formation of hydantoin derivatives. Csing the general procedure of Warel8 each of the glu-

,cO--CH2 CH? \

\ / p-HiSCaH4CO2CH3 + 0

CO-CH I

I

SHCbz - CH,02CC6H4NHCO(CH2),CHC02H I . H~ (pt)

SHCbz 2. Ba(OH)z --

NOTES 209

CsHaSCO

H + HO?CCeH?THCO( CH,)?CH( ;\",)COZH -

HOsCC6H,NHCO( CH?]zCH- CO I

0

(2) J. Edelson, C. G. Skinner, and W. Shire. J . Med. Pharm. Chem., 1,

(3) Y . Sakato, J . Agtic . Chem. Soe. Japan, 23, 262 (1950); through

(4) E. D. Schilling and F. M. Strong, J . A m . Chem. Soc. , 77, 2843 (1955). (5 ) N. Lichtenstein, {bid., 64, 1021 (1942). (6) N. Lichtenstein and N. Grossowicz, J . B i d . Chem., 171, 387 (1947). (7) P. R. Srinivasan, J . A m . Chem. Soc., 81, 1772 (1959). (8) E . Ware, ibid., 60, 2653 (1938).

165 (1959).

Chem. Ahstr.. 45, 3628 (1951).

tamylaminobenzoic acids interacted with phenyl isocy- anate, and in each case a hydantoin was formed, indicat- ing the presence of a free a-amino group adjacent to a carboxyl moiety.

The biological activities of the glutamylaminobenzoic acids were uniformly disappointing. In contrast to the inhibitory activity of N-benzylglutarnine,2 none of the three analogs was toxic to growth of Streptococcus lactis. Lactobacillus aTabinosus or Escherichia coli, a t their limits of solubility in the assay media (about 500 pg./ ml,). Because of the presence of the naturally occurring anthranilic and p-aminobenzoic acid moieties in the condensation products, the compounds were also ex- amined for growth promoting activity. A number of studies were carried out using L. arabinosus in an at- tempt to replace a tryptophan response in this organ- ism by S-( y-glutamy1)-o-aminobenzoic acid; however, no growth-promoting effects mere observed. Sub- sequently, using an E. coli mutant which requires the aromatic amino acids plus p-aminobenzoic acid, un- successful attempts mere made to replace the trypto- phan and p-aminobenzoic acid requirement by N-(y- L-glutamy1)-o-aminobenzoic and N-(7-L-glutamy1)-p- aminobenzoic acids, respectively. The inhibitory and growth-promoting effects of these derivatives were also examined in some botanical systems, e.g., using floating sections of both apical and subapical segments of Avena sativa L.9 The 0- and p-(y-L-glutamy1)aminobenzoic acids were relatively ineffective ; however, the N-(~-L- glutamy1)-m-aminobenzoic acid was somewhat inhibi- tory at about 0.1 mg./ml.

The lack of any significant biological activity of these compounds may be due to their inability to pene- trate the cell wall; or, if they are structurally compa- rable to a natural intermediate, the normal biosynthetic route must involve a more complex derivative of the condensation product.

Experimental 10

Since these derivatives were synthesized through essentially identical experimental procedures, specific details will be given for only one isomer; data on the other analogs will be pre- sented in Table I. In each case, the N+y-glutamy1)aminoben- zoic acid methyl esters as well as the free acids were demon- strated to be homogeneous by the ascending paper chromato- graphic technique in two solvent systems using ninhydrin reagent to develop the chromatograms.

o-(N-Carbobenzoxy-7-L-glutamyl)aminobenzoic Acid Methyl Ester.-A solution of 11.8 g. (0.045 mole) of N-carbobenzoxy-L- glutamic anhydride" and 8.1 g. (0.054 mole) of methyl anthrani- late in 80 ml. of ethyl acetate was heated under reflux using anhydrous conditions for 6 hr. After cooling to room temperature, the reaction mixture was extracted with 57, sodium hydroxide solution. The resulting alkaline aqueous phase was acidified to p H 3 with concd. hydrochloric acid and then extracted with ether. After drying with anhydrous sodium sulfate, the ether was evapo- rated zn vucuo, and the viscous residue was crystallized from ethanol and water. The product was filtered, washed with water and dried, yielding 16.8 g. (767,), m.p. 118-120".

o-(7-L-Glutamyl)aminobenzoic Acid Methyl Ester.-0-(N- Carbobenzoxy-y-L-g1utamyl)aminobenzoic acid methyl ester (6.5 g.) was dissolved in 200 ml. of ethyl alcohol, and hydrogen was passed through the solution for 3 hr. a t room temperature and

(9) A. R. Shrank, Plant Physzol.. 35, 735 (1960). (10) All melting points arc cqrrected and were determined by the capillary

technique using a well.stirr?d liquid bath. The authors are indebted t o R. Johle and ,J. D. Glass for the elemental analyses. and t o Dr. J. M. Ravel, Mrs. Jean Humphrirs, and hIrF. Paula Ables for da ta on the microbial and enzymatic assays, and t o Professor A. R. Shrank for examining the botanical activity of these analogs.

(11) hI. Brrgmann and L. Zervas, Ber., 66, 1192 (1932).

"0

Rrflua tiine, D hr. RiAflrix tinic., -18 hr.

at atniospheric presstire i n the presence of 0.5 g . of palladirini Mack catalyst. The reaction mixture was filtered to remove the c:it- alyst, the filtrate was evaporated to dryness, arid the residue way crystallized from ethanol and water. There was ohtained 3 .5 g . (goo/,,) of product, m.p. 135-156'.

o-(7-L-G1utamyl)aminobenzoic Acid.-A solution IJf 1 .0 g. of o-( 7-L-glutamy1)aniinobensoic acid methyl ester in 100 nil. ( i f

water was adjusted to pH 12 with saturated barium hydroxide solution, and heated on a steam bath for about 1 hr. The reaction mixture was kept at pH 12 by the addition of more barium hy- droxide as required. After the pH of the reaction mixture became constant, it was cooled to room temperature, and carbon dioxide was passed through the solution until i t hecanie acid to litmiis paper. Precipitation of the remaining hariurn ions was rom- pleted by the addition of 1 0 5 sulfuric acid. The resulting niix- ture was adjusted to pH 8 with 10' ; sodium h>-droxide solutii~n and filtered through a Celite mat. The filtrate was acidified t i i pH 5 with sulfuric acid, cooled, arid the rrystallixed pri~duct whicli formed was filtered arid dried ovrr phosphorus pentoside to give (J.7 g. (745;) of niateri:tl which was rerryst:tllizecl f r i ) i i i

water, m.p. 258-2551' dec. N- (2,4-Dioxo-3-phenyl-5-imidazolidine-~-propanoyl)-o-amino-

benzoic Acid [3-Phenylhydantoin of o-(r-L-Glutamyl)amino- benzoic Acid] ,-According to the general procedure of Wares for the preparation of hydantoins, 148 mg. (1006 molar exvess) ( J (

phenyl isocyanate was added dropwise t o a hot stirred soliit,ion i d 300 mg. of o-( 7-L-glutamy1)aminohenzoic acid and I20 m g . of sodium carbonate in 15 nil. of dioxane and water i 1 : I ) . The re:tction mixture was heated on a steam h t h for 30 niin. :tiid kept a t about pH 8 through the addition of sodium carbonate. Suffi- cient concd. hydrochloric acid was added t o the cooled react,ion mixture t<J give a solution of about, p H 1, and this was then heated on it steani bath for an additional 15 min. After cooling in tin ice bath, the precipitated product was filt,ered, washed with water and dried. The resulting mat,erinl was recrystallized from ethanol and water t o give 210 mg. of product, m.p. 1 9 4 - 1 ! 1 3 O dec., which gave a negative reaction to ninhydrin reagent.

Microbial Assays.-Previously reported procedures1*, I s were used for general toxicity studies with 5'. lactis 8039 and L. nrnbinosus 17-5. The assays with E. coli 9723 were carried orit rising an inorganic salts-glucose niedirini l 4 and a previously re- ported assay technique.15 In addition t ~ i tbe inhibition studies, the ability of o-( y-i,-glut arn~l)aniinoberixiiic acid to replace an- thranilic acid or tryptophan for I,. urnhinosus I i -5 was examined in a mediumi2 which w w modified h y omitting trypt,oph:tn and

( 1 2 ) .I. 31. Kair l , I , . M'ouilr, B. I'elsirin. atxl U'. Sllivta, .I. Bioi . ('hrii i . ,

113) J. RI. Ra.\-c.l, T. .I. 1Ii ,<' i , txl , ( ' . 1;. Skinner , ancl M'. Shiv?, ? h i d . .

(14) E. EI. . indc~rson. / ' N c . .\'ut/. .4~ a d . Sc,i., 32. I 2 0 (lIl46). (1.5) I?. \V. 1 ) u n n . .I. 17. Rave l . a n ~ l W Sliivi.. . I . Bioi. Chrr t i . . 219, 8l)!4

~~. ~

206, 391 (1934).

asa, 159 (IYW.

( l Y . ? ( i ) .

which h:id :I t1~1,:d g~iit:inii(~ :wid conceritr:ttion [if 20 ~ K . / I I I [ . 111 cc~nipar:~l~le studies with the E . coli intitant) 83-I,l6 the lins:~I mediuni \vas suppleniented with I O pg./ml. w c h of ~-pheiiyl- danine, 1.-tyrosine nrid L-tryptophan, and various coricentratiori levels of p-( -, -i.-glutninyl)aniinc,beiizuic acid were added in an effort t o induce growth. Alternately, I,-phenylalanirie, ],-tyrosine and p-aminobenzoic acid (0.01 p*g./ml.) were added to be Iiasal medium, and the effect of increasing ccrncentration levels ( i f

o-( ~,-i.-glutamyl)aininobenzoic acid (which could pot en ti all^. swvr :LS a precursor i f tr)-ptophan) wits determined. . ..~ ~

i 1 A ) H. H. I ) a \ i s . i b i i f , , 191, r41.j (1931).

Agents Affecting Lipid Metabolism. 11. Analogs of Mevalonic Acid

Sinre the discovery that mevaloiiic acid i h o i i c of t l w kcy iiitermediateh i i i the biosyritliesis of clwleatt~rol. many attempts have been made to find aiitimetabolitw of this compouiid.2 I? With the exception of thc no rk of Ilaeniker and I)ruey,3 1101ie of these poteiitial ailti-

mf>taholites has iiivolved modification of the futirtioiial