January, 1964
r\TOTES
aldehyde nitrogen mustard 11- and o-tolualdehyde nitrogen mustard Y were obtained commercially.' All melting points are corrected. N,N'-Bis(2,4-dichlorobenzylidene)-1,3-diaminopropane(VIc).h 44.9-g. (0.606 mole) portion of 1,3-diaminopropane was added to a refluxing solution of 216.0 g. (1.23 moles) of 2,4-dichlorobenzaldehyde in 400 ml. of a,bsolute ethanol. The addition took 'LO min. and a white oil separated from the solution during this period. The mixture was refluxed for 1 additional hr., then 150 nil. of absolute ethanol was added t o dissolve the oil and the solution was poured into a beaker. As t,he mixture cooled, a white oil again separated. However, this crystallized as the solution :tpproached room temperature. There was obtained 208.5 g. I(,;) of white solid, m.p. 103-104.5". This material was recnrystallized from absulute ethanol to give 206.0 g. (87.6%) of crystals, m.p. 104.0-104.5°. .4nal. Calcd. for CliH14C14S2: 5 , 7.22. Found: S , 7.40. N,N'-Bis(3,4-dichlorobenzylidene)-1,3-diaminopropane(VId). -This di-Schiff base was prepared in a manner similar to \-IC from 25.2 g. (0.34 mole) of 1,3-diaminopropane and 125.0 g. (0,715 mole) of 3,+dichlorobenzaldehyde. This furnished 110.5 g. (83.75;) of crude material, m.p. 74.5-76.5°. It was recrystallized 3 times from absolute ethanol to yield 94.0 g. (71.2%) of white solid, m.p. 76.5-77.5". ilnnl. Calcd. for C , i H , , C l S ~ :?jj 7.22. Found: N, 7.45.
Acknowledgment is due Drs. H. W. Bond, R. B. ROSS, and J. Leiter of the Cancer Chemotherapy Sational Service Center for their cooperation in obtaining the screening data. We wish to thank the Union Carbide Chemicals Company for the 1,3-diaminopropane which they supplied.
N,N'-Bis(2,4-dichlorobenzyl)-1,3-diaminopropane (IIIc).-A mixture of 79.0 g. (,0.203 mole) of X,?j'-bis(2,4-dichlorobenzglidene)-1,3-diaminopropane (T'Ic) and 1 1. of methanol was cooled in an ice-water bath. The di-Schiff base was relatively insoluble in the cold methanol. The mixture was stirred and 23.0 g. (0.61 mole) of solid sodium borohydride were added in small portions over a 10-min. period. The mixture was warnied until there was an evolution of gas. The external heating was discontinued, although stirring was continued until 1 hr. after all of the solids had dissolved. One-half of the solvent was distilled a t atmospheric pressure and when the solution had cooled to room temperature, 1 1. of a 1.2 JI solution of sodium hydroxide was added. The mixture was shaken briefly and then divided into two approxiniat,ely equal parts, each part was extracted 3 times with 55-3111. portions of diethyl ether. The ether extracts were combined and dried over sodium hydroxide pellets. The solution was filtered into a round-bottomed flask and the ether was removed wit,h a rotary evaporator. There was obtained 70.0 g. (78.7%) of a clear, colorless oil. Anal. Calcd. for CnH1&lSin: X, 7.15. Found: X, 7.29. A small portion of this oil was dissolved in acetonitrile and cooled overnight. A white solid separated and was recrystallized from acetonitrile to give material which melted a t 46.0-47.0'. N,N'-Bis(3,4-dichlorobenzyl)-1,3-diaminopropane (IIId) was prepared by a procedure like that used to prepare IIIc. .4 78.0-g. (0.201 mole) sample of YId was reduced using 50.0 g. (1.32 moles) of sodium borohydride. The product was 61.2 g. (77.7%) of a ~ An analytical sample was obclear, colorless oil, n 2 5 1.5880. tained by short-path distillation a t 170' (0.1 nim.). Anal. Calcd. for C,7H,&182: K, 7.15. Found: X, 7.32. 2-[4-(N,N-Bis(2-chloroethyI)amino)aryl]-1,3-bis(aralkyl)hexahydropyrimidines (Table I).-The synthesis of 2-[4-(N,Nbis( 2-chloroethyl)aniino)-o-tolyll-l,3-bis(p-chlorobenzyl)-hexahydropyriniidine ( X I ) was typical of these compounds. A 5.20-g. (0.02 mole j sample of o-tolualdehyde nitrogen mustard (Y)was dissolved in 25 ml. of warm absolute ethanol. To this solution was added 8.47 g. (0.02 mole) of N,X'-bis(p-chlorobenzyl)-1,3-diaminopropane, also in 25 ml. of absolute ethanol. The mixture was refluxed for 15 min., then 10 ml. of solvent was allowed to distil at atmospheric pressure over another 15-min. period. The solution was :illowed t o cool slowly to room temperature and then stored overnight in a refrigerator. This effected the separation of a light yellow solid. There was 5.3 g. of this material, n1.p. 10:3-130". This was recrystallized from 7.5 ml. of acetonitrile to furnish 4.05 g., n1.p. 123-131'. The filtrate from the reaction mixture was reduced to approximately ' / 3 of the original volume. Subsequent cooling of this solution yielded an additional 2.62 g. of white solid which melted at 129-131'. This was combined with the 4.05 g. of solid obtained from the initial crop of product and recrystallized twice from aretonitrile. This gave 4.73 g. (41.8c;'cj of crystals, m.p. 130.0133.5". Anat. Calcd. for C,,H,,ClS3: S , 7.43. Found: S , 7..iG. (7) Frinton Laboratories, South Vineland, N. J.
117
7-Aminodesacetoxycephalosporanic Acid and Its Derivatives
R.J. S T E D M A S , K. SWERED,A N D J. R. E. HOOVER Smith Kline and French Laboratories, Philadelphia, Pennsylvania
Received September 24, 1963
Catalytic hydrogenation of the antibiotic cephalosporin C (1a)l gave a product which we have identified as the desacetoxy derivative I I a S 2 This new antibiotic material was not isolated, but its structure was established by hydrolysis to 7-aminodesacetoxycephalosporanic acid (IIb). Although I I b had no appreciable antibacterial action, three derivatives (IId, IIe, and IIf) prepared from it showed modest activities. Since the completion of this work, Morin, et aLJ3have reported briefly the hydrogenation of a cephalosporin derivative; the structure given for their product is in accord with our findings.
a, R = HzNCH (CHZ)~CO
I
COOH
b, R = H C,
d, R=CsHsCHzCO
R = CeHt (NOz)z NHCH(CH2)sCO COOH I
e'
Ra m C H z C O
f, R=HzNCO
The hydrogenation of Ia was carried out a t low pressure using a large quantity of palladium catalyst. Paper electrophoresis showed that in addition to unchanged starting material there was present a new antibacterial substance (IIa) which gave a purple color with ninhydrin and had a strong ultraviolet absorption. The reaction mixture was treated with 2,4-dinitrofluorobenzene to give a mixture of ICand IIc, and the latter, after partial purification, was hydrolyzed with acid to afford a small yield of IIb, isolated by ionexchange chromatography. The dinitrophenyl group did not influence the hydrolytic cleavage of the side chain, but its introduction facilitated the manipulation of the intermediates and the isolation of the product.* (1) E. P. Abraham and G . G . F. Newton, Biochem. J . , 79, 377 (1961). (2) (a) E. P. Abraham and G. G. F. Newton, ibid., 69, 658 (1956), reported t h a t cephalosporin C absorbed hydrogen in t h e presence of a catalyst, but they did not characterize the product; (b) according t o t h e nomenclature recommended in ref. 5 . the hydrogenation product is a derivative of 7amino-3-n1ethyl-A~-cephem-4-carboxylic acid (IIb). (3) R. B. Morin, B. G . Jackson, R. A. Jfueller, E. R. Lavagnino, JV. B. Scanlon, and S. L. A n d r e w , J . Am. Chem. Soc., 85, 1896 (1963). (4) The hydrolysis and isolation procedures were similar t o those described for the preparation of I b in Belgian Patent 593.777 (1959).
January, 1964
NOTES
was isolated from the hydrolysate by absorption from solution a t pH 6-7 onto a column of Dowex 1-X8 (acetate form) and elution with S acetic acid. The fractions showing strong absorption a t 260 mp deposited crystals which were dissolved in water by bringing to pH 8 with sodium hydroxide and reprecipitated by adding hydrochloric acid to p H 3.7. The yield of colorless product was 26 mg., m.p. 241-242' dec. Anal. Calcd. for CRH10X203S:C, 44.85; H, 4.70; N, 13.08. Found: C,44.64; H, 4.94; N, 12.98. B.-7-Aminocephalosporanic acid (8.81 g., 0.025 mole) dissolved in 100 nil. of water containing 4.20 g. (0.05 mole) of sodium bicarbonate was hydrogenated for 1.5 hr. at 4 kg./cm.2 over 27 g. of 5yc palladium on barium ~u1fate.l~The catalyst was removed and washed with dilute sodium bicarbonate. To the dark solution were added 60 ml. of 6 S hydrochloric acid and 9 g. of charcoal. Filtration gave a clear yellow solution which was cooled in ice and brought to pH 3.7 with 2 X sodium hydroxide to precipitate 3.32 g. (62%) of light, tan crystals, sufficiently pure for the preparation of derivatives. A colorless product was obtained by dissolving in 0.5 S hydrochloric acid and reprecipitating by adding sodium hydroxide to pH 3.7. An analytical sample (reprecipitated as in method A ) was indistinguishable by m.p., infrared spectrum and electrophoretic mobility from material prepared by method A. Anal. Calcd. for C8H10S203S: C, 44.85; H, 4.70; N, 13.08. Found: C, 44.63; H,4.75; K, 13.03. 7-Phenylacetamidodesacetoxycephalosporanic Acid (IId).-A solution of 4.29 g. (0.02 mole) of I I b in 200 ml. of water containing 5.04 g. (0.06 mole) of sodium bicarbonate was mixed with 300 ml. of acetone and stirred a t -20" while 3.71 g. (0.024 mole) of phenylacetyl chloride was added. The reaction mixture was kept at -20" for 30 min. and then at 3" for 2 hr. Two washings with ether removed the acetone. The aqueous residue was acidified with 30 nil. of 6 S hydrochloric acid and extracted twice with ethyl acetate. The extracts were washed with a little 0.5 S hydrochloric acid and evaporated to dryness. The residual foam was triturated with ether to give 5.51 g. of white solid (IId as free acid). This was dissolved in 300 ml. of 1-propanol cont,aining sc,) of water and treated with 12 nil. of 1.65 N potassium 2-ethylhexanoate16 in 2-propanol to precipitate 5.14 g. (65yi) of the colorless potassium salt of the product. A sample recrystallized from water-1-propanol and air-equilibratedI6 had m.p. 212-213' dec.; lactani carbonyl stretching at 5.75 p (Sujolj. Anal. Calcd. for C,cH15KX,0aS.1.5H,O: C, 48.35; H, 4.56; N, 7.05. Found: C,48.18; H,4.43; S,7.14. 7-(2-Thienylacetamido)desacetoxycephalosporanic Acid (IIe). -A procedure similar to that described in the foregoing experiment was used to acylate 1.07 g. (0.005 mole) of IIb with 2-thienylacetyl chloride" and to isolate the product as the free acid (1.25 g. of white solid). This was dissolved in 80 nil. of 2-propanol containing 5'/c of water and 2.7 nil. of 1.8 ibf sodium 2-ethylhexanoate1*in 2-propanol was added to precipitate 1.17 g. (63(,.1) of the colorless sodium salt. A sample recrystallized from water-2-propanol and air-equilibrated had m.p. 231-232' dec.; lnctam carbonyl stretching a t 5.75 p (Nujol). Anal. Calcd. for C1aH13~~Sa04Sz.O.5H?O:C, 45.52; H, 3.82; N,7.58. Found: C,45.50; H, 3.66; S, 7.57. 7-Ureidodesacetoxycephalosporanic Acid (IIf).-A suspension of 2.14 g. (0.01 mole) of IIb in 20 ml. of water containing 0.811 g. (0.01 mole) of potassium cyanate was stirred for 7 hr. a t room temperature. ilcidification of the resulting clear solut,ion to p H 0.5 with hydrochloric acid gave 1.95 g. of white solid (IIf as free acid). This was suspended in 70 ml. of methanol and stirred lohexylamine was added to pH 7 . 3 . The solid dissolved and the solution was evaporated to a foam which was crystallized from water-tert-butyl alcohol to give 1.74 g. (459;) of the colorless cyclohexylammonium salt of the product, m.p. 173-174' dec. after air-equilibration; lactam carbonyl stret'ching a t 5.71 p (Sujol). (14) From Engelliard Industries, Inc., Newark, N . J. T h e catalyst was a h r o n n ponder. 4 gray-colored palladium on barium sulfate produced by the same company gave less satisfactory results, a8 did palladium on carbon. (15) German Patent 965,703 (1957). (16) Several weeks were required for equilibration. A shorter period gave material with propanol of crystallization. (17) 2-Thienylacetic acid (J. H. Ford, G. C. Prescott, and D. R. Colingsn o r t h , J . A m . Chem. Soc., 72, 2109 (1950)) was converted t o the acid chloride ( h . p . 32' a t 0.2 inin.) h y refluxing with thionyl chloride in benzene. (18) .i. Gourevitch, C. T . Holdrepe. C;. .1.Hunt, FV. F. Minor, C. C . Falnipan. I.. C. Cheney. and J. Lein, dntibiot. Chemotherapy. 12, 318 (1962).
119
Anal. Calcd. for C ~ H , , ~ ' , O ~ S . C ~ H ,C, J ~ 45.11; H ~ ~ : H, 7.00; S, 14.96. Found: C, 48.17; H, 6.89; 5 , 14.68.
Acknowledgments.-The authors are indebted to Mrs. D. A. Rolston and her staff for the microanalyses reported herein, to Dr. W. E. Thompson for valuable discussions of the interpretation of n.m.r. spectra, and to Dr. C. A. Simpson for the determination of pK, values. They also wish to thank Miss 31. Dolan and her staff for carrying out the microbiological work.
Hypocholesterolemic Agents. IV. 3a-39ethoxy-5a-androstaneDerivatives R. E. COCNSELL ASD P. L).I~LIMSTRA Division of Chemical Research of G . D . Searle and Co., Chicago 80, Illinois Receiced October 10, 1963
I n 1959 Hellman and associates2 demonstrated that the serum cholesterol levels of hypercholesterolemic patients could be reduced by the parenteral administration of androsterone (3a-hydroxy-5a-androstan17-one). Later studies by Cohen, et u Z . , ~ ~however, ~ showed that this hypocholesterolemic effect was not achieved when the steroid was given orally. As a consequence of these findings, a program was initiated in our Laboratories aimed at preparing an orally active androsterone analog having a good separation of hypocholesterolemic and androgenic properties. Since it had been previously shown4>jthat orally administered androsterone is rapidly conjugated in the liver and excreted predominantly as the 3-glucuronide, it was felt that etherification of the 3a-hydroxyl group mould inhibit this conjugation and lead to orally effective compounds. As a result, a series of 3a-methoxy-5aandrostane derivatives was synthesized and biologically evaluated.
fi
I I1
=o
I11
OH
s
8H I
L._._._.....__... H
CH3
- .
VI
C=CH
?H L
...............C,,H 5
The methyl ether of androsterone (I)6was readily prepared (74% yield) from epiandrosterone (3P-hy(1) Paper 111: R . E. Connsell, P. D. Klimstra, and R. E. Ranney, J M e d . P h a r m . Chem., 6 , 1224 (1962). (2) L. Hellman, H. L. Bradlow, R. Zunioff, D. K. Fukushima, and T. F. Gallagher, J. Clin. Endocrinol. Metab., 19, 936 (1959). (3) UT.D. Cohen, N. Higano, and R. W. Robinson, C i r c d a t t o n , 22, 659 (1960). (4) UT. D. Cohen, N. Higano, R. W. Robinson, and R. J. LeReau. J . Clin. Endocrinol. ,Metab., 21, 1208 (1961). ( 5 ) 0. CrBpy, RI. F. Jayle, and F. hIeslin, C o m p t . rend. SOC. biol.. 161, 234 (19.37). (6) Since completion of this work. this compound liar been prepxred in essentially the same manner by R. Gardi, R. Vitalli, and 1.Ercuii, GQZZ. Chim. Ital.. 92, 632 (1982).
