16-FLUORINATED CORTICOIDS. II. 16α-FLUOROPREDNISOLONE

Barney J. Magerlein, Robert D. Birkenmeyer, and Fred Kagan. J. Am. Chem. ... Hoffman , Henry M. Kissman , and Martin J. Weiss. Journal of Medicinal an...
0 downloads 0 Views 272KB Size
1252

1-01, s2

CONMUNICATIONS TO THE EDITOR

carbonyl), and [1,2-(trimcthylene)-cyclopentadie- compound WAS identical with that prcpai-d I J the ~ nyl]-manganese tricarbonyl, respectively. An in- osmium tetroxide hydroxylation of the Al6-2Odependent synthesis was undertaken to confirm the ketone (XVII), vide iizfra. The position and structures. stereochemistry of the new hydroxyl introduced into I was thus established as 1Ga. \Tith thionyl ---Mn(C0)3 W - - - M n ( C 0 ) 3 chloride-tributylamine, I1 yielded 20-chloro-21acetoxy-l1~-hydroxy-1,4,16-pregnatrien-3-one (V), I I1 m.p. 160-161", [ a ] D 65" ( ~ h f . ) ~Chromic . acid oxidation of the 1Ga,l7a-diol (VI), m.p. 223224", [ a ] D 2" (chi.), formed by treatment of V with osmium tetroxide, yielded 21 -acetoxy-20chloro - 3,11,17 - trioxo - 1G,17 - seco - 1,4 - pregna111, K = COCl VI1 dien-16-oic acid (VII), m.p. 238-241", 238 I \ - , R = CHO r n p (15,000). Titration with 0.1 N sodium hyY, R = CH=CH-COOH 1-1, R = CH,CH,COOH droxide quantitatively transformed V t o the 20,21Accordingly [ (chloroformyl) -cyclopentadienyl 1- oxide (VIIIa), m.p. 205-210", [ a ] ~ 125" (chf.). manganese tricarbony13 (111) was treated with In a similar manner 21-acetoxy-1ID-hydroxylithium tri-t-but~xyaluminohydride~ to give 4,17(20)-pregnadien-3-one (Ib)5 was converted to [(formyl)-cyclopentadienyl ]-manganese tricarbonyl 11P - hydroxy- 20,21-oxido -4,lG-pregnadien- 3 -one (IV). The IV was condensed with malonic (VIIIb), m.p. l54-155", [a]D+212" (chf.). Lithacid to give [(2-carboxyvinyl) - cyclopentadienyl]- ium aluminum hydride reduction of VIIIb and then manganese tricarbonyl (V). V was reduced over oxidation with nianganese dioxide led to the isoRaney nickel to give [ (2-carboxyethyl) -cyclo- lation of approximately equal amounts of 11,8,21(IX) m.p. 151pentadienyll-manganese tricarbonyl (VI). VI was dihydroxy-4,1G-pregnadien-3-one 153" and 1 1,R-hydroxy-4,1G-pregnadiene-3,20-dione cyclized with polyphosphoric acid t o yield [1,2(1 - oxo - trimethylene) - cyclopentadienyl] - man- (X),m.p. 171-173", XE::H241 ml.r (25,200). The latganese tricarbonyl (VII). VI1 was reduced with ter compound was oxidized t o the known 3,11,20zinc and hydrochloric acid to yield 11. Compari- trione6. This sequence established the structure son of this material with that obtained by reduction of the 20,2l-oxide, VIIIb, and by analogy the of I, by infrared, mixed melting point, vapor phase A1s4-analog,VIIIa. %'hen the oxide VIIIa was treated with hydrogen chromatography and X-ray diffraction showed them to be identical in every respect. Since there fluoride and the product acylated, a mixture was is only one possible position for the double bond in obtained from which 20-fluoro-21-acetoxy-llbliydroxy-l,4,16-pregnatrien-3-one XI, m.p. 173I , this also proves the structure of I. liso, [a]D+80° (chi.) and 1Ga-fluoro-21-acetoxy( 3 ) Prepared b y t h e method of J . Kozikowski, unpublished work. 1lp-hydroxy-l,4,1i(20)-pregnatrien-3-one XII, m.p. (4) H . C. Brown and R. F. RIcFarlin, Tins J O U R N A L , 80, 5372 " were isolated. Evi190-191", [ ~ r ] ~ + 5 9(chf.) (1958). E T H Y L CORPORATIOS T. 1%.COFF~ELDdence of the presence of the lG,&fluoro-isomer XI11 FERNDALE 20 I;.G. IIIRMAS was obtained, but this compound was not isolated. I)ETIIOIT, LIICII. Ii'. BURXS Ozonization of XI yielded the 16,17-seco-keto RECEIVED DECEMBER 28, 1959 acid XIV, m.p. 221-224' while similar treatment of XI1 afforded 10a-fluoro-1lp-hydroxy-4-androstene-3,li-dione XP, m.p. 197-198", [c~]~+135" 16-FLUORINATED CORTICOIDS. 11. (chf.). A mixture of 20- and 16-fluoro compounds 16~-FLUOROPREDNISOLONEAND was also obtained when the 20-chloro compound V 9a,16or-DIFLUOROPREDNISOLONEDERIVATIVES was treated with silver fluoride under a variety of Sir: conditions; however, the main product of this reThis report is a preliminary account of the syn- action was the 20-hydroxy compound XVI, m.p. thesis of 16-fluor0 cortical hormones, a unique class 194-195", [a]D+92' (chf.). hlangancse dioxide of biologically active steroids. oxidation of XVI afforded the 20-ketone XYII, 21-Acetoxy-1 1/3-hydroxy-l,4,1i(20)-pregnatrien- m.p. 208-209", [cu]D+146", 2.22 m p (23,750), %one (1a)l was oxidized by selenium dioxide in which when treated with osmium tetroxide with aqueous dioxane to form 21-acetoxy-l1/3,16a- subsequent acylation yielded the known tetrol (W). dihydroxy-1,4,17 (20)-pregnatrien-3-one (11) , m.p. The mixture of 16-fluorides XI1 and XI11 when 179-181", [ ~ J D+83" (chf.). The oily 10-ace- treated with N-methylmorpholine oxide-peroxide tate (111) obtained from 11, when treated with N- and a catalytic amount of osmium tetroxide gave methylmorpholine oxide-peroxide2 and a catalytic 1OQ-fluoro-1lp,17a-dihydroxy-21-acetoxy-1,4 - pregamount of osmium tetroxide afforded the known nadiene-3,20-dione XVITI, m.p. 2 19-220 " , and 16a,21-diacetoxy-11~,17a-dihydroxy-1,4 - pregna- small amounts of the IGP-isomcr XTX. m.p. 17.1diene-3,20-dione (IIT)3, m.p. 162-165". This 177", identical with that obtained by another

a

+

+

+

~

(1) J. A. Hogg, F. H . Lincoln, A. €1. LTathan, A. R. Hanze, B. J , Magerlein, W. P. Schneider, P. F. Beal and J. Korman, THISJ O U R N A L , 77, 4438 (1955). (2) W. P. Schneider and A. R. Hanze, U.S. 2,769,823 (November 6, 1956). (3) S Bernstein, R. H. Lenhard, U'. S Allen, M , Heller, R. Littell, S . If.Stolar, I.. I. Feldman und R. 1%.Blank, TITIS JOuRN.4L. 81, 1G89 ( 13.iO).

(1) Cf.R. E. Ireland, T. I. Wrigley and W. G. Young, ibid., 80, 4604 (1958), who also noted t h a t thiongl chloride-tributglamine, conditions t h a t usually favor a n SK2 process, in certain cases yieldcd the rearranged chloride by way of a n S S i ' mechanism. ( 5 ) J . A . Hogg, P. F. Beal, A. H . Xathan, F. H. Lincoln, \V. P Schneider, B. J. hlagerlein and R. \T. Jackson, ibid.,77, 4430 (195.51. (0) B. J. Magerlein, D. A . Lyttle, R. €I. Levin, J . O r g . Chcir?., 2 0 , 1709 (1855).

March 5 , 1960

~ O M M U N I C X T I O N STO THE

EDITOR

1253

process'. Dehydration of XVIII and introduc- 4.25, 4.08, [aIz5D-97", Rf 0.52).3 Analogies4J tion of the Sa-fluoro group by established methods8 to this electrophilic reaction on aromatic compounds yielded 9a,16a-difluoro-1lp, 17a-dihydroxy-21-ace- suggested substitution in the D ring, which was toxy-l,4-pregnadiene-3,20-dioneXX, m.p. 265- supported by the absorption spectra and chemical behavior of the product. Convincing evidence 268'. Preliminary anti-inflammatory assaysg showed for the exact assignment was obtained by carrying 16a-fluoroprednisolone acetate XVIII t o be about out the halogenation on 6-demethyl-6-deoxytetra16 times as active as hydrocortisone and the cycline labeled with tritium in the 7-positi0n.~ 9a-fluoro-derivative X X to be about 75 times as The displacement of tritium by bromine provided active as hydrocortisone'o. proof that the substituent occupied the 7-positi0n.~ Similarly, treatment of 6-demethyl-6-deoxytetra(7) D . E. Ayer and W. P. Schneider, THISJ O U R N A L , 83,1249 (1960). cycline (11) with N-iodosuccinimide gave 7-iodo-6( 8 ) J. Fried and E. Saho, rbid.. '76, 1455 (1954). (9) A. Robert and J. E. Neeamis, A d a Endocriizol., ZS, 105 (1957). demethyl-6-deoxytetracycline sulfate (IV, found (10) T h e authors are indebted t o W. E. Dulin, S. C. Lyster and for CzlH21N20~I.H&304.0.5H~0: C, 39.1; H, 4.2; associates for t h e biological d a t a , t o J. L. Johnson a n d W. A. Struck 230, 345 mp, log E 4.48, I, 19.2; ",';A and associates for elemental and spectral analyses and rotations, t o 4.12, [CX]~'D +383", & 0.91). Reaction of 6G. Slomp for n.m.r. d a t a a n d G . E. VandenBerg for technical assistance. deoxytetracycline (I) with either N-bromosucRESEARCH LABORATORIES BARNEY J . MAGERLEIN cinimide or N-iodosuccinimide yielded 7-bromo-6THEUPJOHNCOMPANY ROBERTD. BIRKEKMEYERdeoxytetracycline sulfate (V, found for C&23KALAMAZOO, MICHIGAX FREDKAGAN N2O7Br.HzSO4.H20:C, 42.3; H , 4.7; Br, 13.0; RECEIVED JASUARY 14, 1960 A;J:HC'268, 345 mp, log E 4.24, 4.10, [ff]"D -221', Rt 0.80) or 7-iodo-6-deoxytetracycline sulfate (VI, found for C?2H23N20,1.H2S04: N , 3.8; S, 4.8; I, 6-DEOXYTETRACY CLINES. I. CHEMICAL 19.5; 240, 260, 345 mp, log E 4.26, 4.22, MODIFICATION BY ELECTROPHILIC SUBSTITUTION 4.05, [ff]"D -282", Rf 0.91),respectively. Sir: In contrast to halogenation, nitration of 6demethyl-6-deoxytetracycline with potassium niThe chemical stability of the recently reported1,* broad spectrum antibiotics, 6-deoxytetracycline trate in concentrated sulfuric acid a t 0' gave two (I) and 6-dernethyl-6-deoxytetracyline(11), has mononitro isomers. Using the technique described permitted the study of a series of electrophilic above with tritium labeled starting material one substitution reactions under strongly acidic condi- of these isomers was proved to be 7-nitro-6tions. From two such reactions, nitration and demethyl-6-deoxytetracycline (VII, found for G1bromination, we have obtained several new deriva- H~iN309.2H20:C, 51.3; H, 5.8; N, 8.2;",',:A tives which possess antibacterial properties. These 262, 350 mp, log E 4.35, 4.27, [ f f I z 5 D -442", Rr properties are being evaluated and will be the 0.64). Since the groups attached to the aromatic ring of the molecule would direct electrophilic subject of a future communication. attack t o the 7 and 9 positions, it was assumed that the isomer which retains the tritium label was 9c,H3 5 H 3 nitro-6-demethyl-6-deoxytetracycline(VIII, found for C21H21N309: C, 55.1; H, 6.2; N, 9.0; ?:A: 263, 360 r n p , log E 4.42, 4.24, [ c ~ ] ? ~ -131°, D Rr 0.48). In a like manner, nitration of G-deoxytetracycline (I) gave two isomers, but the ratio of 7nitro-6-deoxytetrac cline (IX)8 to 9-nitro-6-deoxytetracycline (Xy was smaller than in the 61,R demethyl series and was attributed to the steric 11, R 111, R hindrance of the &methyl group. The 9-nitro-6IV, R deoxytetracycline (X, found for C2?H23N30g.C4V, R HgOH.HzS04: C, 47.8; H, 5.2; N, 6.9; VI, R 260, 365 mp, log E 4.43, 4.23, [ff]25D -268", Rr VII, R VIII, R 0.58) was purified in sufficient quantities to be IX, R used in subsequent reactions. S,R Catalytic reduction of the nitro compounds with XI, R platinum yielded the corresponding amino derivaXII, R SIII, R tives, 'T-amino-6-demethyl-6-deoxytetracycline (XI, XI\;, R found for CzlH23Na07.2HC1.3H20: C, 45.4; H , SV,R

Treatnient of (i-deinethyl-(5-deoxytetracyclinc ( I I) with N-brotnosuccinimide in concentrated sulfuric acid a t 0" yielded a single monobromo-(idemethyl-6-deosytetracycline sulfate (I I I , found C , 42.5; H , for C?1HzlN~0.iBr.H2S04CH30H: 4.7; OCH3, 5.0; Az,'>Hcl 270, 345 m p , log E (1) J. R. D. McCormick, E. R. Jensen, P. A. Miller and A. P. DoerschtLk, THISJOURNAL, in press. ( 2 ) C. R . Stephens, K. Rlurai. H. H. Retinhard, L. H. Conover and K. J Brunings, ibid., 80, 5324 (1958).

(3) All optical rotations mere determined a t a concentration of 0.1-

0 5"' ,c i.n 0 1 N sulfuric acid. Rfvalues were determined in the system I butanol/0.2 M phosphate buffer, fiH 2.

J. B. hfenke, THISJ O U R N A L . 44, 141 (1925). ( 5 ) H. Schmid, Nciv. Chim.A c t a , 29, 1144 (1946). (6) This material was prepared by the method of J . R . D. l l c Cormick, el al. (see ref. 1) using 6-demethyltetracycline-7H8 made b y t h e method of T. Andre and S. Ullberg. THIS JOURNAL, '79, 494 (1957). (7) T o our knowledge this is t h e first use of t.itium replacement as a structure proof and we are indebted t o Dr. E. F. Ullman for t h e suggestion of this elegant method. (8) This material was characterized by paper strip chromatography, and the isomer ratio was estimated from the mixture. (4)