ORGANIC CHEMISTRY.1 . INTRODUCTION.THE subjects chosen for review include an account of recent work on organiccompounds containing fluorine, a comparatively new field of study on whichgreat progress has been made within recent years, and from which importantresults are being obtained of both theoretical and practical interest. Somerecent developments in aromatic chemistry are reviewed, such as Reppe'swork on cyclooctatetraene, the controversy over the constitution of thediazocyanides, and the mechanism of the Bucherer reaction. Recent workon acyclic terpenes containing isoprene units in irregular union and onsesquiterpenes is reviewed. Further developments in the steroid field arereported, which include a reference to Miescher's total synthesis of oestroneand to recent progress towards the total synthesis of sterols themselves. Thechemistry of marrianolic and doisynolic acids and related compounds is alsodiscussed.Two methods are availablc for the formation of fully fluorinated hydro-carbons.Fluorination can be carried out either witrh fluorine in the presenceof a catalyst such as silver or gold under conditions which dissipate thetremendous heat of formation, or with an inorganic fluoride such as cobaltfluoride. Both methods replace all hydrogen atoms by fluorine andsaturate any double bonds. The products formed are called ~fluorocarbons,and their carbon " skeletons " are believed to correspond to those of theoriginal hydrocarbon. By the application of these methods benzene isconverted into perfluorocycbhexane CGF12, n-heptane into perfluoro-n-heptane %C,F16, and toluene into perfluoromethylcychhexane CF3*C6Fl1.In similar manner naphthalene, tetrahydronaphthalene, and decahydro-naphthalene are converted into perfluoronaphthalene C,,Fl,.Fluoro-carbons are unique among organic substances in their stability towardschemical reagents, and arc decomposed only when heated with fluorineitself or with sodium at 300400". Prolonged boiling with concentratedmineral acids or alkalis is without effect upon them, and they seem to beinert towards all other chemical reagents. Because of their stability,fluorocarbons are best characterised by their physical constants. Theyhave low boiling points, low surfacc tensions, low refractive indices, highdensities, and high viscosities.For cornpounds of relatively high molecularweight fluorocarbons have remarkably low boiling points ; with the exceptionof CF,, C,F6, C3F,, and C,F,, the fluorocarbons have boiling points lower thanthe parent hydrocarbons, a phenomenon which is believed to be due to thelack of hydrogen bonding.Partly fluorinated organic compounds may be obtained by a number ofmethods. Controlled fluorination by fluorine itself is not yet a procedureof wide use, but replacement of one or more halogens (Cl, Br) byfluorine can be effected by the use of an inorganic fluoride or a8BAKER AND HEY : INTRODUCTION. 83mixture of fluorides (KF, HF, SbF,, SbF,Cl,, SbF,, HgO + HF), e.g.,EX, --+ XF,. Replacement of halogen and addition of hydrogen fluorideto a double bond can also be carried out, e.g., -CX=CH- --+ -CF,-CH,-.Another method involves the addition of hydrogen fluoride to acetyleniccompounds, e.g., R-CiCH --+ R*CF,-CH, (R = alkyl or halogenatedalkyl).Addition of fluorine to a double bond can also be accomplished bythe agency of lead tetrafluoride (prepared in situ from PbO, and HF), e.g.,2HF> C = K 3 >y-y<F F aThe effect of the >CF, )CF, and -CF, groups on such processes aschlorination and polymerisation is discussed, and reference is made to theprofound effect which one or more of these fluorinated groups has uponthe properties of unsaturated systems and upon the function of such groupsas H, OH, X (halogen), NH,, CO,H, and C,H,. The suggestion is putforward that research in this new branch of organic chemistry will furnishresults which should either confirm some present ideas of the electronicinterpretation of organic reactions or necessitate the development of new ones.Reference is also made to the industrial importance of organic fluorides andtheir derivatives. The solid polymers (CF,*CF,), and (CF,*CFCl),, stableto boiling concentrated mineral acids, are excellent materials from which tomake apparatus required to resist corrosive chemicals.These solid polymershave good insulation properties, and the intermediate polymeric fluoro-carbon oils are useful as heat-transfer agents, transformer oils, and lubricatingoils.I n spite of the vast amount of work which has been carried out onaromatic compounds, much of our knowledge is still unsatisfactoryand incomplete, but, by the application of new methods, improvedtechnique, and modern theory, considerable advances have been madewithin recent years, and many important problems cleared up.It issurprising that, although cyclooctatetraene is a compound of unusualtheoretical interest and considerable doubt has been expressed as to whetherin fact it had been synthesised, some thirty years passed without any attemptbeing made to repeat Willstatter’s work. This state of affairs has now beenremedied, and as the result of the work of W. Reppe cyclooctatetraene can beobtained in quantity and has been thoroughly investigated by chemical andphysical methods. Willstatter’s work has been confirmed and extended,and cyclooctatetraene has been shown beyond all doubt to be a highlyunsaturated cyclic polyene.Its chemical properties are even more variedand interesting than had been suspected, and its ease of aromatisationopens up considerable possibilities for industrial as well as laboratorysynthesis. There is still controversy as to its fine structure, the balanceof the evidence being in favour of a non-planar slightly resonating moleculewith alternate single and double bonds.Our knowledge of the fine structure of aromatic compounds has beenincreased by the quantitative ozonisation of o-xylene, 2 : 3-dimethyl-naphthalene, and similar compounds, by the X-ray determination of the bon84 ORGANIC CHEMISTRY.lengths in pyrene, 1 : 2 : 5 : 6-dibenzanthracene, and other hydrocarbons,and by the application of theoretical considerations and physical measure-ments to indrtne and its derivatives : the results are interpreted in terms ofthe theory of resonance.Other notable work on aromatic hydrocarbonsincludes ring-closure by cyclo-dehydrogenation and cyclo-dehydration, anddetailed investigations on the reduction of the polycyclic compoundsanthracene, phenanthrene, and pyrene. The structure of the diazocyanideshas been studied by means of their infra-red spectra, and strong evidenceis adduced to show that both the cis- and the trans-forms are true nitriles andnot isonitriles. Structural isomerism is thus excluded, and the Hantzschtheory would appear to be confirmed.The mechanism of reactions continues to attract attention. TheBucherer reaction has been investigated kinetically, and the results obtainedwere found to be in harmony with the commonly accepted mechanism ofFuchs and Stix.Structures have been assigned to the intermediates formedby the action of a bisulphite on sodium naphthionate. The homogeneousCannizzaro reaction is found t.0 be not influenced by t'he addition of peroxidesor peroxide inhibitors, and objections to the postulated intermediat'eformation of benzyl benzoates have been removed. Aldehydes which do notundergo the change under the usual conditions do so in presence of an activesilver catalyst. The synthetic possibilities of the Stobbe reaction have beendemonstrated, and attention has been drawn to the frequent occurrence ofdimeric products in the Clemmensen reduction.Nearly ten years have passed since the mono- and sesqui-terpenes werelast reviewed.Although this period includes the war and immediate post-war years, there has been so much activity in this field that a comprehensivereview is not possible. The monoterpene section has therefore beenrestricted to acyclic terpenes containing isoprene chains in irregular union.A new alcohol, lavandulol, has been isolated from oil of lavender ; it closelyresembles an isomer of geraniol previously synthesised by L. Ruzicka andhis collaborators. The structure of lavandulol has been put on a firm basisby an unequivocal synthesis of the ( f)-alcohol. The irregular structure ofartemisia ketone has recently been confirmed by synthesis.The out-standing problem of the structure and synthesis of irone, the perfume fromorris root, a t long last seems to have been solved. L. Ruzicka, and hiscollaborators, having disposed of the cycloheptene skeleton, established the6-methylionone structure by extensive degradative experiments, followed bya synthesis of (f)-a-irone. hone is regarded as being a mixture of a-, P-,and y-isomers, the y-isomer, coilfaining an exocyclic methylene group,predominating in natural irone.Much interesting work has been accomplished in the sesquiterpene group.The structures of several cadinene hydrocarbons have been revised, followingthe application of a new method for locating the double bonds. Thisconsists in treating the sesquiterpene oxide with inethylmagnesium iodidefollowed successively by dehydration and dehydrogenation.The positions ofthe additional methyl groups in the resulting homocadalene are found bBARER AND HEY : INTRODUCTION. 85degradation and synthesis, thereby locating the positions of the double bonds.The eudslenic sesquiterpene ketone, eremophilone, has been shown by J. 1;.Simonsen and his collaborators to possess the 1 : 9-dimethyl-7-isopropyldecalinskeleton. Despite much study, the structure of caryophyllene has not yet beencomplebly elucidated, although the bicycZ45 : 2 : Olnonane skeleton seemsestablished. The adduct with maleic anhydride now appears to be asubstitution product at an ally1 position. Similarly, a bridged bicyclo-[5 : 3 : Oldecane structure appears likely for cedrene.A comprehensivereview of recent work on the azulenes is given. Following the establishmentof the structures of guaiazulene and vetivazulene, P. A. Plattner and hiscollaborators have elucidated the structures of guaiol and vetivone,representatives of the bicyclo[5 : 3 : Oldecane sesquiterpenes.In the steroid field the total synthesis of the natural estrogenic hormone,(+)-estrone, has been reported by G . Anner and K. Miescher, and anadvance which holds promise of further achievements is the total synthesisby J. W. Cornforth and Sir Robert Robinson of the tricyclic diketone (I),obtainable as a by-product from the oxidative degradation of methyl diacetyl-deoxycholate and of cholesteryl acetate ilibromide.The addition of ring Dpresents no difficulties in principle, and the way is now open to the totalsynthesis of saturated and unsaturated derivatives of androstane andztiocholane, such as androsterone, testosterone, progesterone, corticosterone,lithocholic acid, and cholesterol. A direct proof ofthe (@)-orientation of the hydroxyl group in cholesterolhas been given by C. W. Shoppee; this applies to all3-hydroxy-steroids which, directly or indirectly, havebeen correlated with cholesterol in respect of configur-ation a t C3, and is subject to qualification only in so faras future determination of the absolute configuration at some centre ofasymmetry may show the actual stereochemical arrangement of the steroidniicleus to be the mirror-image of that at present accepted by convention.A method of possibly general application for the introduction of radioactivecarbon into the steroid skeleton at position 3 has been worked out by R.B.Turner in the case of cholest-4-en-3-one.Recent syntheses of (+)-oestrone, " o! "-estradiol, equilenin, and iso-equilenin are summarised, and the available evidence indicating that theC/D-ring union in the natural estrogenic hormones is of the same type(trams) is discussed. The physiologically active cardiac glycosides and thephysiologically inactive albglycosides both possess cis-C/D-ring unions, andhave been shown by T. Reichstein to differ only in configuration at Cl,. Thebile acids constitute one of the most important sources of starting materialfor steroid partial syntheses; a summary has therefore been given of theN-bromosuccinimide method introduced by K.Miescher for the degradationof the side-chain with removal of three carbon atoms in one stage.Results of the highest chemical and biological iniportance have accruedfrom the work of K. Miescher and his co-workers on the marrianolic anddoisynolic acids. Biologically, the most striking results are the productionc\( I . ) CI i386 ORGANIC CHEMISTRY,from (+)-equilenin and by total synthesis of a compound, (-)-" ct "-bisde-hydrodoisynolic acid, and the total synthesis of a racemic doisynolic acid,the estrogenic activities of which are about ten times greater in the rat thanthat of diethylstilboestrol. Dimethylethylallenolic acid, which shows someformal structural resemblance to the doisynolic acids, is reported to bealmost as active as the foregoing intensely estrogenic acids.In an attemptto discover some relation between estrogenic potency and constitution, avery large number of analogues of the doisynolic acids has been described;the results obtained have been summarised and the methods used for theirsynthesis reviewed.W. B.D. H. H.2. ORGANIC COMPOUNDS OF FLUORINE.Introduction.The chemistry of organic and inorganic fluorides has undergone anunexpected expansion during the last seven or eight years, mainly becauseof the exigencies of war. This expansion, which is still continuing, requireda plentiful supply of good-quality fluorine. Once a laboratory curiosityand regarded as a dangerous substance, it is now readily available for eitherlaboratory or manufacturing purposes.Though still a dangerous chemical,its properties are so well known that it can be handled with relative ease,and it is used a t the present time for the production of a large variety oforganic substances which serve as lubricants, insulators, corrosive-resistantapparatus, and media for the transmission of power.The unique character of fluorine is due to the fact that it is the mostelectro-negative element. It is also remarkable because the fluorine atomnot only acquires and holds an electron with tenacity but shows a markedtendency to attract still more. Both these characteristics of fluorine playan important r6le in the preparation and the properties of organic fluoro-compounds.The electrophilic effect of fluorine atoms on the propertiesof organic compounds is so pronounced t h a t their study is likely to providefurther information concerning the theory of the electronic interpretationof organic reactions. Fluorine compounds show extremes of chemicalbehaviour: some are very reactive and unstable, and others display aninertness comparable only with that of the so-called rare gases. Someorganic fluoro- and chlorofluoro-compounds can be inhaled without any illeffect, while others, of which some are completely fluorinated, are extremelytoxic, and there is reason to believe that some of these perfluoro-compoundsmay display anesthetic properties.Events during the recent war were largely responsible for the develop-ment of this vast and comparatively new field of chemistry. From whatmay be said now of this progress, it may seem to the reader that thework done by the pioneer workers in fluorine chemistry is insignificant.Nothing could be further from the truth, for the recent rapid progresSMTTH : ORGANIC COMPOUNDS OF F1,IJC)RTNE 87could not have been made had it not been for the early work of Moissan.Ruff, Swarta, and others in a relatively difficult field.Fluorocarboit.9.The availability of large supplies of good-quality fluorine and theknowledge that it can be handled in aluminium, magnesium, iron, silver,copper, and nickel apparatus a t temperatures up to about 50O0, and insilica or “ Pyrex ” glass a t room temperature has enabled much progressto be made in studies of direct fluorination.It is in this field of directfluorination that perhaps the greatest advance has been made recently.The indirect method of fluorination is dependent upon the fact that aninorganic fluoride such as antimony trifluoride will replace the halogen inorganic halides by fluorine, a reaction often named after F. Swarts, ttdistinguished pioneer in this field. Except in the case of a few inorganicfluorides like CoF,, AgF,, MnF,, or CeF,, which require elementary fluorinefor their preparation,21 the metal fluorides by themselves do not give riseto completely fluorinated substances. From an examination of the experi-mental facts there is reason to believe, however, that a combination of theindirect with the direct method of fluorination has great possibilities forthe production of fully fluorinated substances of all types on a large scale.Fluorine, having a standard electrode potential of -2.85 volts (chlorinehas a standard electrode potential of - 1 ~ 3 6 volts) can be expected to bemuch more reactive than other electronegative substances, and this differenceis likely to lead not only to unique reactions but to unique products.Becauseof the great difference in reactivity between fluorine and chlorine theirbehaviour in organic reactions is so different that a knowledge of the reactionsof chlorine is of little use in predicting those of fluorine. Fluorine combineswith all the elements except the inert gases.The molecule of fluorine isextremely stable, but under some conditions it is highly reactive, so much sothat once it enters into a reaction, the combination may proceed either at anuncontrollable rate or even with extreme violence.5* 6* Many of itsreactions are accompanied by the liberation of a great deal ‘of energy, andthey involve high activation energy. Such is the case when fluorine combineswith organic substances. Normally there is complete decomposition, theonly identifiable product in any quantity being CF4.** The reason forH. J. Emelbus, J . , 1942,441 ; S. G. Osborne and M. M. Brandegee, Ind. Eng. Chem.,1947, 39, 273; J. F. Froning et at., ibid., 1947, 39, 275; A. V. Grosse and H. F. Priest,ibid., p. 279 ; R. Landau and R.Rosen, ibid., p. 281 ; A. F. Benning et al., ibid., p. 286.R. Keim and 0. Ruff, 2. anorg. Chem,., 1931, 201, 245; cf. B. Brauner, Ber., 1881,14, 1944; J . , 1882, 41, 68; J . , 1894, 65, 392; Z. arwrg. Chein., 1894, 7, 1.3 R. D. Fowler et al., Id. Eng. Ckem., 1947, 39, 343.4 W. M. Latimer and J. H. Hildebrand, “ Reference Book of Inorganic Chemistry,”’6 H. Moissan, Ann. Chim. Phys., 1891, 24, 224.6 H. Moissan and G. Chavanne, Compt. rend., 1905, 140, 407.7 M. M. Meslans, Ann. Chim. P h p . , 1894, 1, 346.8 B. Humiston, J . Physical Chem., 1919, 23, 572.9 A. Damiens and P. Lebeau, Compt. rend., 1926, 182, 1340 ;p. 474, Macmillan, 1940.1930, 191, 939When the 3-hydroxy-7 : 17-diketone (LXIII) is acetylated, there isobtained the enol-diacetate (LXVII), in which the position of the new doublebond between c6 and C, is established by the absorption spectrum, and whichby hydrogenation with palladium black in acetic acid and subsequentalkaline hydrolysis gives the 17-keto-3 : 'I-diol (LXVI; R = H).52 This isconverted by treatment as the 3-monobenzoate with phosphorus penta-chloride (1 mol.) in the presence of calcium carbonate into a 7-chloride,dehydrochlorinated by use of sodium iodide in pyridine to give, after removalof the benzoyl group by alkaline hydrolysis, A6-cestrone (LXV).64 Theposition of the new double bond in (LXV) is established by the absorptionspectrum, and hydrogenation with palladium black in ethanol a t 20" gives(+)-oestrone as the sole product.% This sequence of reactions avoids thehigh temperature Wolff-Kishner reduction, and again links (+ )-cestrone(XLIVa) with (+)-equilenin (LXI) in respect of configuration a t both C,,and C14.Transformations analogous to the foregoing have recently been carriedout 55 on " E "-dihydroequilin (LX with -OH for :O at C,,), which as the3 : 17-" a "-diacetate is hydroxylated by treatment with osmium tetroxideto a 3 : 7 : 8 : 17-" a "-tetrol (LXIV with -OH for 10 a t C17) also obtainedby reduction of (LXIV) with sodium-ethanol.The tetrol either bydistillation a t 200" in a high vacuum or by treatment with hot 8% sulphuric54 W. H. Pearlman and 0. Wintersteiner, J . Biol. Chem., 1940, 132, 605.5 5 E. Schwenk, E. Bloch, and B. Whitman, U.S.P. 2,418,603, abstracted in Chem.Abs., 1947, 41, 4518182 ORGANIC CHEMISTRY.acid undergoes pinacolic dehydration to give a good yield of a 7-keto-3 : 17." a "-diol (cf.LXIII), which by Wolff-Kishner reduction or by removal ofthe 7-carbonyl group by the sequence involving a 7-chloride, gives" a "-cestradiol (XLIVa, with -OH for :O a t C,,), a reduction product of( + ) - ~ s t r o n e . ~ ~From compounds recently prepared by Ruzicka et aZ.,57 Reichsteinet aE.,S* and K. M e ~ e r , ~ ~ the molecular rotation difference for inversion ofconfiguration at C14, A14i,qr, - 14-nomaz, is about + 200".60 If we accept theabove evidence that (+ )-cestrone and (+ )-equilenin possess correspondingconfiguration at C,, and C1, we can writ,e the following set of formulse for(+)-equilenin and (+)-14&oequilenin, and for (-)-equilenin and (-)-14-i~0-equilenin :( + )-Equilenin (+)-14-isoEquilenin ( -)-Equilonin ( -)-14-iso-Equilenin(LXIU.) (LXVIIIa.) (LXIb.) (LXVIIIb.)[MID f 224" + 391" - 224" - 391" + 224" t 224"A14iso - I4nOrmal + 167" A13880 - 13mrmal - 615"_.-It will be seen that the molecular rotation difference for inversion at C,, is& 167", which is not far from 200", whilst for inversion at CI3,AI3-bo - 13normal = 615", which is in good agreement with the figure of 570"given above.A new andingenious synthesis of (&)-equilenin and (&-)-14-isoequilenin hasbeen described by W.S. Johnson, J. W. Petersen, and C. D. Gutsche,G1 theprincipal stages of which are given below. The hydroxymethylene-ketone(LXIX) with hydroxylamine gives 96% of the crystalline isooxazole (LXX),which by treatment with sodium methoxide and methyl iodide undergoesfission, ketonisation, and methylation to give 92% of the crystallinecyano-ketone (XXI).Condensation with methyl succinate furnishes by aThorpe reaction the unsaturated 17-imino-ester-acid (LXXII), which losescarbon dioxide and ammonia to yield 83% of the 17-keto-ester (LXXIII);this by hydrolysis with barium hydroxide yields 98% of the keto-acid(LXXIV) , decarboxylated with pyridine hydrochloride-hydrochloric acid5 6 E. Schwenk and F. HiIdebrandt, Naturwiss., 1933, 21, 177; B. Whitman,0. Wintersteiner, and E. Schwenk, J . Bid. Chem., 1937,118, 789.5 7 P. A. Plattner, L. Ruzicka, H. Heusser, J.Pataki, and K. Meier, Helv. Chim. Acta,1946, 29, 942, 949; 1947, 30, 1342.5 8 J. von Euw and T. Reichstein, ibid., 1944, 27, 1851 ; P. Speiser and T. Reichstein,ibid., 1947, 30, 2143.58 Ibid., 1946, 29, 718; 1947, 30, 1977, 2025.6o W. Klyne, Nature, 1948, 181, 434.61 J . Amer. Chem. s'oc., 1945, 67, 2274; 1947, 69, 2942SROPPEE : STEROIDS AND RELATED COMPOUNDS. 183to give 66% of A14(1S)-equilenin methyl ether (LXXV). Reduction of theC,,,,,,-double link affords 32% of ( j - ) - 14-isoequilenin methyl etherNaOMe(LXXI.)I(LXIX .) (LXX.)(LXIa. ) (LXIb.) 10+J(LXVIIIb.)' (LXXIII.)(LXXV.) (LXXIV.)(LXVIIIa + LXVIIb; R = Me) and 63% of (&)-equilenin methyl ether(LXIa + LXIb : R = Me), demethylated with hydrochloric acid-aceticacid in 98% yield to (-j-)-equilenin (LXIa + LXIb ; R = H).The parent hydrocarbons of ( j - ) -equilenin and ( & ) - 14-isoequilenin, termed-equilenane, have been prepared from the " p "- and" a "-equilen-l'l-ones by Clemmensen or Wolff-Kishner reduction ; the" p "-hydrocarbon is also obtained from (-j-)-equilen-l6-0ne.~~ A single(&)-2-hydroxy-analogue of (A)-equilenin or (-J-)-14-isoequilenin has been~ynthesised.~~Oxidation of ( + )-cestrone acetate (XLIVcc) with alkaline hydrogenperoxide or peracetic acid in acetic acid 64 gives a 3-acetoxy-lactone(LXXVII), which by alkaline hydrolysis gives the sodium salt of a 3-hydroxy-acid.Acidification to pH5 gives the free 3-hydroxy-acid, called oestrolicC < p I ) _ and C < a Y Y0(XLIVa.) (LXXVI.) (LXXVII.( LXXVIII. )acid (LXXVI), whilst excess of acid and heating leads to the3-hydroxyl-actone (LXXVII). Methyl estrolate is readily hydrolysed by62 A. L. Wilds, L. W. Beck, and T. L. Johnson, J . Amer. Chem. SOC., 1946,68,2161.63 W. E. Bachmmn and W. J. Horton, ibid., 1947, 69, 58.64 R. P. Jacobsen, J . Bwl. Chem., 1947, 171, 61184 OKC; .4 N IC' CH EM TS'I'KV.warin sodium hydrogen carbonate to the 3-hydroxy-lactone7 so tlhat formula(LXXVII) is preferred to (LXXVIII) involving a tertiary carboxyl group.The lactone (LXXVII) and its derivatives h a ~ e in. p.s closely similar tothose of the h ydroxy-lactone and its derivatives obtained byW. W. Westerfeld 65 from (+)-@shone by oxidation with hydrogen peroxide,but their identity is unlikely because Westerfeld's hydroxy-lactone is sixtimes more cestrogenically potent than (LXXVII) and stimulates thepituitary, whilst cestrolic acid and its lactone inhibit secretion of gonadotropichormone by the pituitary.Analogous acids and lactones have beenobtained 66 corresponding to androsterone, isoandrosterone, dehydroiso-androsterone, testosterone, and androstane-3 : 17-dione ; they all lackandrogenic activity but stimulate pituitary growth and secretion. Ahitherto unknown polymorph of isoandrosterone acetate, m. p. 116-5-1 17')was encountered.66 Finally a bisdehydro-aestrolic acid and lactone havebeen prepared 67 from (+)-equilenin acetate.The classical method associated with the names of Barbier 68 andWieland 69 for the stepwise degradation of the bile-acid side-chain has beenthe subject of several proposed modifications with a view to improving theoverall yield from C,,-acid to C,,-ketone.A new method used byT. F. Gallagher and V. P. Hollander 70 eliminates two carbon atoms a t onetime :R-CHMe-CH,*CH,-CO,H synthe4is -+ R*CHMe*CH,-CH,*COGH,N, diazoketone kIR*C!HMe*CH,*CH,*CO*CH, +-"- RXHMe*CH,*CH,*CO*CH,ClBr2 - U R rCrO, .1R*CHMe*CH=CH*CO*CH, ---3 R*CHMe*CO,H .Another new and elegant method devised by Miescher et aZ.71 eliminatesthree carbon atoms a t one time :PhNgBr R*CHMe*CH,*CH,*CO,Me ---+ R*CHMe*CH,*CH,*CPh,*OH20 21 '32 '33 24I 4- X-homo- R*CHMe*CHBr*CRrCPh, +----,z-- R*CHMe*CH,*CHXPh,22ICrO R-CHMeX!H*CHrCPh, --:* R-COMe -1- OCH*CH=CPh,.'Lo "1(iL .I.Biol. Chem., 1942. 143, 177.( i 7 It. P. Jacobsen, C. 31. Picha, and H. Levy, ibid., p. 81.68 P. Barbier and R. Locquin, Compt. ?*end., 1913, 156, 1443.8u H. Wieland, 0. Schlichting, and R. Jacobi, 2. physiol. CIwm., 1926, 161, SO.70 J . Biol. Chem., 1946, 162, 549.71 C. Meystre, H. Frey, A. Wettstein, and K. Mioscher, Helv. Chim. -4cta, 1944,27,18 15.(jG R. P. Jacobsen, ;bid., 1947, 171, 71Bromination of the olefin from deoxycholic acid with N-bronio-succiiiimide gives a 38% yield of the 22-bromo-olefin, increased by irradiationto 75% ; 72 N-bromophthalimide may also be used. The overall yield frommethyl deoxycholate to 3(a) : 12(a)-dihydroxypregnan-20-one (6 stages) is186% ; 'il using the Barbier-Wieland procedure (17 stages), the overallyield is from 3.8 73 to 7.1%.7* The recognition of diphenylacraldehyde asthe other product of the final oxidative fission is interesting theoretically, andof practical importance because its removal as the bisulphite compoundfacilitates the isolation of the steroid methyl ketone in a state of p~rity.7~The method has been applied to cholic acid 76 which furnishes a 30% yield of3(a) : 7(a) : 12(a)-pregnan-20-one, whilst 5-aZZocholanic acid,76 which hasnot previously been degraded by the Barbier-Wieland procedure,gives 5 -aZZopregnan-2O-one.3 ( p ) -Hydroxy- 5-aZEocholanic acid 77 furnishes3( p)-hydroxy-5-aZEopregnan-20-oiie, a constituent of the corpus luteurn ;this is reduced by sodium-ethanol to 5-aZZopregnane-3( p) : 20-a-diol, the20-p-epimeride of which has been identified as a component of ox bile.783( a)-Hydroxycholanic acid (lithocholic acid) 77 gives 3( a)-hydroxypregnan-20-0ne, previously isolated &om pregnancy urine, and reduced by sodium-ethanol to a mixture of the pregnane-3(a) : 20-a-diol of G.F. Marrian 79 andthe pregnane-3(a) : 20-p-diol of R. E. Marker et aLa0 3(a) : 6(a)-Dihydroxy-cholanic acid (" a "-hyodeoxycholic acid) a1 is degraded by the Barbier-Wieland method as used by W. M. Hoehn and H. L. Mas0n,7~ and byMiescher's new method to the same 3( a) : 6( P)-dihydroxypregnan-20-onewhich differs from the compounds previously stated 8 2 ~ 83 to have this(LXXIX.) (LXXX.) (1XXXI.)structure. Use of' 3(~)-hydroxychol-5-enic acid,84 wherein the double bondhas been protected by addition of hydrogen chloride to give (LXXIX) leads72 C.Meystre, L. Ehmann, R. Neher, and K. Miescher, Helo. China. -4cta, 1945, 28,'3 P. Hegner and T. Reichstein, ibid., 1943, 26, 715.SB K. Xescher and J. Schmidlin, Helv. Chim. Acta, 1947, 30, 1405.7 6 C:. Meystre and K. Miescher, ibid., 1945, 28, 1497.7; C. lieystre and K. Miescher, ibid., 1946, 29, 33.7 8 W. H. Pearlman, J . Riol. Cheni., 1946, 166, 473.7 y Hiochem. J . , 1947, 41, 193; 1929, 23, 1090.IU J . A4rner. Chem. ~Soc., 1937, 59, 2291.81 R. B. Moffet, J. E. Stsfford, J. Linsk, and W. M. Hoehn, ibid., 1046.68, 1857.8" T. Kimura and G. Sugiyama, J . Biochenz. Japan, 1939,29,409.B3 1:. 3:. Marker and J. lirueger, J . Amer. Chern. Xoc., 1940, 62, 79.1252.W.M. Hoehn and H. L. Mason, J . Amer. Chern. SOC., 1938, 60, 1195186 ORGANIU UHEMISTRY.to pregnenolone (LXXX) and thence to progesterone (LXXXI) ; threedifferent routes are elab0rated.8~By bromination with N-bromosuccinimide, 3-keto-5-do-steroids yieldthe 2-bromo-derivatives, and 3-keto-steroids give the 4-bromo-derivati~es,~~whilst 3-keto-A4-steroids (e.g., testosterone, progesterone) yield the 6-bromo-compounds ; 86 it has, however, been found that by use of N-bromosuccinimidewith irradiation bromination at C,, of 24 : 24-diphenylchol-23-enes proceedsso rapidly that 3-ketones and A4-unsaturated 3-ketones are unaffected.87Thus (LXXXII) gives the 22-bromo-compound (LXXXIII), dehydro-brominated to (LXXXIV), which by oxidation gives progesterone (4throute) .88 Alternatively, the acetoxychloro-diene (LXXXV) can behydrolysed by hydrochloric acid to the hydroxychloro-diene (LXXXVI),oxidised to the 5-chloro-diketone (LXXXVII), which affords progesterone(LXXXI) by simple treatment with potassium carbonate (5th route) ; thesequence of hydrolysis and oxidation can be reversed via (LXXXVIII)(6th(LXXXII.) (LXXXIII.) (LXXXIV.)(LXXXI)\/\I1 HO C1 (LXXXVI.) CPh,AAC(LXXXV.) ciyH--' (LXXXVII.)ACO' C1 (LXXXVIII.)When the 24 : 24-diphenylchola-20 : 23-diene (LXXXIX) is treated withN-bromosuccinimide in either the absence or the presence of light,substitution of the C,, -methyl group occurs to give the 21-bromo-compound(XC) ; this is readily hydrolysed t o the 21-hydroxy-compound (XCI) whichcan be reconverted into the parent bromide.*', 89 By treatment withC.BTeystre, H. Frey, R. Neher, A. Wettstein, and K. Miescher, Helv. Chirn. Acta,1946, 29, 627.8 5 C. Djerassi and C. R. Scholz, Experieiztia, 1947, 3, 107.8 6 C. Meystre and A. Wettstein, ibid., 1946, 2, 408.88 C. Meystre, A. Wettstein, and K. Miescher, Helv. Chim. Acta, 1947, 30, 1022.6 R Idem, ibid., p. 1037.8 7 Idem, ibid., 1947, 3, 186SBOPPEE : STEROIDS AND. REIIATED COMPOUNDS. 187potassium acetate or sodium ethoxide, the bromo-compound furnishes bhe21-acetoxy-diene (XCII) or the 21-ethoxy-diene (XCIII) from which theside-chains may be removed by oxidation. Less satisfactorily, the side-chain can be eliminated first by oxidation of the 21-bromo-diene (XC) to givethe 21-bromo-20-ketone (XCIV), in which the bromine atom can be replacedby acyloxy- or alkyoxy-groups.I n this way deoxycholic acid is readilytransformed into the 3(a) : 12(a) : diacetoxy-20 : 21-ketol acetate (XCV)prepared previously by H. G. Fuchs and T. Reichstein and by L. Ruzicka,P. A. Plattner, and J. Pataki ; 91 the analogous 3-keto-compound and thecorresponding 21 -ethyl ether (XCVI) have also been obtained.89AcO'(LXXXIX.)AcO CrO,\ (XCI.)(XCV.) (XCIV.) (XCVI.)The method has been extended to the production from 3(P)-hydroxychol-5-enic acid 92 of 11 -deoxycorticosterone acetate and 11 -deoxycorticosterone21-methyl ether, and from 3(a)-hydroxy-ll-ketochoIanic acid 93 of ll-keto-progesterone and 11 -dehydrocorticosterone acetate.(XCVII.) (XCVIII.) (XCIX.) (C.)90 Helv.Chim. Acta, 1943, 26, 511.e2 C. Meystre and A. Wettstein, ibid., 1947, 30, 1256.O 1 Ibid., 1944, 27, 988.93 Idem, ibid., p. 12621 HH ORGANIC!. CHE'MISTRY.By contrast with the ready 22-bromination undergone by 24 : 24-di-phenylchol-23-enes (as XCVII), 20-bromination of the 23 : 23-diphenyl-norchol-22-ene (XCVIII) does not occur ; 72 again, whilst 21-brominatlion of24 : 24-diphenylchola-20 : 23-dienes (as XCIX) occiirs readily, 21-brominationof tjhe 22 : 22-diphenylbisnorchol-20-ene (C) docs not take place.89 When22- bromo- 3 ( p ) -acetoxy-24 : 24 -diphenyl-5-aZZochol-23-ene (CI) is oxidisedwith chromium trioxide, the expected 22-bromo-noracid. (CIII) isaccompanied by the broinine-free bisnor-acid (CII) which may arise byanionotroyy (CI + CIV) followed by oxidation, or by hydrolysis of (CI) to(CV) followed by oxidation.94 With sodium-potassium hydroxide a t 200",the bromo-ncicl (CIII) gives 3( ~)-h~ciroxy-ri-aZZonorchol-20-enic acid (CVI) .94coo H Br Hr(CII.) (CI.1 (CIII. )OH\/\\I,/$ CBrYh, I ,,TL ,I-- /;- /;-H (CIV.) H (CV.) H (CVI.)Another method for shortening side-chains is that of H. HunsdieckerR*CH,*C02Ag -'A R*CH2Br ---+ R*Meand C. Hunsdiecker : 95Zn,AcOHAccording to I>. 31. Clark, N. G. Brink, and E. S. the diacetates ofdeoxycholic, nor-, and bisnor-deoxycholic acids as the dry silver salts giverespectively 25%, 40%, and 65% of the corresponding bromides, which arereduced in 90yo yields t o 3(a) : 12(a)-diacetoxynor-, -bisnor-, and -trisnor-cholanes.In the hands of B. Koechlin and T. Reich~tein,~' the Hunsdieckerreaction failed with 3( ~)-acetoxyaetio-5-aZZocholanic acid.Deoxycholic acid is an important starting material for partial syntheses ;attention may therefore be directed to papers dealing with its preparation,98purification,S9 and chromatographic separation from cholic acid.lO0Conversion of cholic acid into lithocholic acid,lol and the preparation of5-rtllocholanic acid froin 3( p)-hydroxychol-5-enic acid, have been described. 7gy4 Y. Wieland and K. Mieseher, Hclz,. Ghim. Acta, p. 1576.9 L Ber., 1942, 75, 291.y i Helv. Chinz. Acta, 1944, 27, 549.9s V. P. Basu and P. Gupta, J . Sci. I d . Res. India, 1946, 5, B, 83.99 R.Charonnat and B. Gauthier, Compt. rend., 1946, 223, 1009; 1947, 224, 279.100 H. Silbermaii and S. Silberman-Martyncewa, J . Biol. Chent., 1946, 165, 359.lU1 H. Heusser and H. W7uthier, Helti. C'him. Acfa, 1947, 30, 2165.O 6 .7. Riol. Clzem., 1946, 162, 095SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 189Steroid keto-groups can readily be eliminated by reduction of the mercaptalswith Raney nickel. Ethylenedithiol condenses with keto-groups at C,, C,,CI2, and C1,, but monothiols react only at C,; 3 : 7 : 12-triketocholanic(dehydrocholic) acid can thus give either cholanic acid or 7 : 12-diketo-cholanic acid, whilst 3( a)-bydroxy-7 : 12-diketocholanic acid gives3(a)-hydroxycholanic (lithocholic) acid.lo2 A new method for reduction ofthe carboxyl group : -CO,H 4 -COCl~-- -CO*SMe 5 -CH,*OH, hasbeen described by V.Prelog et aZ.lo3 3-Iodo-7 : 12-diketocholanic acid hasbeen prepared 1°4 for use as a radio-opaque substance for X-ray diagnosticvisualisation of the gall-bladder, but is ineffective because unabsorbed.D-Homotestosterone and its 17a-epimeride have been synthesised ; lo5the former is 15-20 times more active than the latter, and two-thirds asactive as testosterone, which itself is 30 times more active than its17-epimeride ; D-homoandrost-4-ene-3 : 17a-dione is half as active astestosterone. The codguration of the hydroxyl group attached to ring Dis clearly more important in regard to biological activity than the size oft'his ring, and it has been suggested lo5 that in testosterone the hydroxylgroup has the (p)- rather than the (a)-configuration formerly assigned.The influence of configuration on physiological activity is further exemplifiedby the active cardiac glycosides and the inactive isomeric alloglycosides.That these differ only in configuration a t C,, was originally suggested byR. Tschesche,lo6 was rendered highly probable by L.Ruzicka's synthesis of~Zlouzarigenin,~~~ and proved by T. Reichstein's degradation of periplogenin(CVIII) and albperiplogenin (CIX) to ~tio-5-aZlocholanic acid (CVII) and14-iso-17-iso-~tio-5-~ZZocholanic acid (CX) respectively ; lo8 in periplogeninall the functional groups are thus ((3)-orientated.R-SIT(CVIL) (CVIII. ) (CIX.) (CX,)11~2 H. Hauptmann, J . Anher. C%em. Soc., 1947, 69, 663.l o 3 0.Jeger, J. Norymberski, S. Szpilfogel, and V. Prelog, Helv. C'hitti. Actu, 1948,lo4 R. P. Jacobsen, G . M. Yicha, C. Weinstein, and L. Ittomanoff, J . Bid. C ' l ~ c ) t ~ . ,1°5 M. W. Goldberg, J. Sice, H. Robert, and P. A. Plattner, Helv. (?him. Acla, 1947,106 R. Tschesche and K. Bohle, Ber., 1938, 71, 654; R. Tschesche, K. Bohle, and10' P. A. Plattner, L. Ruzicke, H. Heusser, and E. Angliker, HeEv. Chim. Acta, 1947,108 P. Speiser and T. Reichstein, ;bid., p. 2143.29, 683.1947, 171, 87.30, 1441.W. Neumann, ibid., p. 1927.30, 1073190 ORGANIC CHEMISTRY.An adequate account of the synthetic work from the Ziirich laboratoryand of the degradational work from the Basle laboratory on cardiacglycosides and aglycones would require a separate article and must bedeferred.Eight hydrocarbons, C21H36, epimeric at C,, C14, and C,, should exist;four of these are now known-pregnane (CXI), 5-allopregnane ((3x11) ,14-iso-17-isopregnane 109 (CXIII), and 14-iso-17-iso-5-allopregnane l10 (CXIV)which is identical with the hydrocarbon, diginane, obtained byC.W. Shoppee ll1 by degradation of the physiologically inactive cardiacaglycone, diginigenin. The identity of R. E. Marker's urane,l12 C2,H,,,m. p. 127-128", remains obscure, but it may prove to be one of the fourremaining isomeridea.(CXII.) (CXIII.) (CXIV. ) ( C X I . )m. p. 83", rn. p. 84", m. p. 103 and 105" m. p. 76",[ a ] ~ + 19.6" [.ID f 12.7" [.ID + 19" [alu + 24"In conclusion reference should be made to the application of infia-redspectroscopy to steroid chemistry; 113 by this means, in combination withchromatographic analysis, no fewer than 42 steroids, of which 28 have beenidentified, have been detected in the keto-steroid fraction from normal andpathological urines.l14 Some of these are steroids with oxygen (:O or -OH)at Cll, or A9(11)-steroids clearly derived from cortical precursors, andemphasise the importance of the adrenal gland in the bodily economy inhealth and in disease.Because of lack of space an account of recentdevelopments in the synthesis of 11 -oxygenated steroids and adreno-corticalhormones must be held over.Marrianolic Acids, Doisynolic Acids, Allenolic Acids and Related Compounds.by fusion of cestriolwith potassium hydroxide obtained a dicarboxylic acid, C, 8H2205, alsoisolated by D.W. MacCorquodale, S. A. Thayer, and E. A. DoisyY2 and nowIn 1932, G. F. Marrian and G. A. D. Haslewoodlog K. Rieyer, Helv. Chim. Acta, 1947, 30, 2024.110 Idem, ibid., p. 2127.112 It. E. Marker, 0. Kamm, T. S. Oakwood, E. L. Wittle, and E. J. Lawson, J . Awer.C'hesri. SOC., 1938, 60, 1061 ; W. Klyne and J. Y. F. Paterson, Biochenz. J., 1948, 42,Yroc. ii.llS R. F. Furchgott, H. Itosenkraritz, and E. Shorr, J . Biol. Chem., 1946, 163, 3751946, 164, 621 ; 1947,167, 627 ; 1947, 171, 523.114 K. Dobriner, 112th Meeting of the Ainerican Chemical Socioty, New York City,September 1947; K. Dobiiner, S. Lieberman, C. P. Rhoads, B. R. Hill, L. F. Fieser,It. N. Jones, V. Z. Williams, and R. B. Barnes, J .BioZ. Chenz., 1948,172, 241, 263, 297.Ibid., 1944, 2'7, 246.J . SOC. Chm. Id., 1932, 51, 2 7 9 ~ . a J . Bwl. Chem., 1933,9@, 327SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 191known as marrianolic acid. In 1933, D. W. MacCorquodale, L. Lewin,S. A. Thayer, and E. R. Doisy 3 subjected estrone to fusion with potassiumhydroxide and obtained a moiiocarboxylic acid, C,,H,,O,, now known asdoisynolic acid, which appears also to have been isolated in an impure stateby H. A. Weidli~h.~ The American workers claimed3 that these acidspossessed biological activity several times greater than that of cestrone in theestrus test in rats, but this claim was subsequently withdrawn.6 In 1937and 1939, W. Hohlweg and H. H. Inhoffen in association with Schering A.-G.filed patents 6 describing the production by fusion with potassium hydroxideof cestradiol, dihydroequilin, equilin, or equilenin of the same acid as hadbeen obtained from cestrone, or of analogous monocarboxylic acids.Theseacids were regarded as hydrophenanthryl-1 -acetic acids, and it was claimedthat they possessed threshold activities by oral administration in the estrustest in rats of 1-67, whereas the corresponding figure for oestrone is 20-3Oy.The chemical and biological aspects of the matter have now beeninvestigated and extended by K. Miescher and his collaborators; the moststriking results are the production from (+)-equilenin and " 01 "-dihydro-equilenin, and by total synthesis, of a compound, (-)-'< cc "-bisdehydro-doisynolic acid, and the total synthesis of a racemic doisynolic acid, theestrogenic activities of which both by a single subcutaneous application andby a single oral administration are about 10 times greater than that ofdiethylstilbestrol.Porrrzution, Stereochemistry, and Physiological Activity.-In addition to theoriginal preparation from cestriol by potash fusion, alternative routes to(+)-marrianolie acid (IVu) as the 7-methyl ether (IVb) have been describedinvolving permanganate oxidation of astrio1 3-methyl ether (Ib) 3 andconversion of (-J-)-cestrone 3-methyl ether (IIb) into the 16-oximino-derivativeand subsequent ring-fi~sion.~ An improved preparation has now beendescribed,8 whereby (+) -estrone 3-benzyl ether (IIc) is oxidised withpotassium hypoiodite at 20" to give (+)-7-benzylmarrianolic acid (IVc), whichis readily converted into (+)-marrianolic acid (IVa) by hydrogenation withnickel at 20".(+)-Marrianolie acid (IVa) passes into the anhydride at 180"/0-15 mm.;esterification with diazomethane gives the dimethyl ester, one ester group ofwhich is very readily hydrolysed and the other very resistant to hydrolysis[cf. the dimethyl ester of (IVb) '1. The pure dimethyl ester benzoate giveswith tetranitromethane a faint but distinct yellow coloration which must beattributed to the aromatic double-bond system. The constitution isestablished by dehydrogenation with selenium to 7-hydroxy-1 : 2-dimethyl-phenanthrene (V; R = OH), converted by distillation with zinc intoJ. Bid. Chem., 1933, 101, 753.S.A. Thayer, D. W. MacCorquodale, and E. A. Doisy, J . Pharm. Exp. Ther.,D.R.-PP. 705,862 (Dee. 19, 1937), 719,578 (Jan. 20, 1939).F. Litvan and R. Robinson, J., 1938, 1997.J. Heer and K. Miescher, Helv. Chirn. Actu, 1945, 28, 156.Dissertation, Gottingen, 1934.1937, 59, 48192 ORGANIC CHEMISTRY.1 : 2-dimethylphenanthrene (V ; R = H).9 (+)-Marrianolic acid by oral orsubcutaneous administration is completely inactive in doses of 1 mg. in theOHBOH(IVa; R = H.)(IVC; R = CR,Ph.)(IVb; R = Me.)(VIa; R1= R2 = H.)(VIb; R1= H, R2 = Me.)(VIc; R1 = Bz, R2 = Me.)1tV.1 tm-)estrus and uterus test in rats; lo this confirms the revised results of Doisyet a1.5(+)-l)oisynolic acid (VIa) is obtained in SOYo yield by fusion of either(+)-oestrone (IIa) or " o! "-oestradiol (111) with potassium hydroxide at275" ; its constitution as a 7-hydroxy-2-methyl-l-ethyloctahydro-phenanthrene-2-carboxylic acid follows from its dehydrogenation withpalladised charcoal in acetone at 320" in 54% yield to 7-hydroxy-2-msthyl-l-ethylphenanthrene 8 (VII; R = OH), converted by distillation with zincinto 2-methyl-l-ethylphenanthrene l1 (VII; R = H).-The acid (VIu)cannot be esterified with alcoholic hydrochloric acid, but with diazomethanegives the methyl ester (VIb) characterised as the benzoate (VIc) ; (VIc) withhot 2N-alcoholic potassium hydroxide gives (VIb), which can only behydrolysed with potassium hydroxide at 180". The pure ester-benzoate(VIc) gives a pale yellow oolour with tetranitromethane, but is unaltered bytreatment with osmium tetroxide or monoperphthalic acid.The ultra-violetabsorption spectrum of (VIb) is practically identical with that of dimethylA. Butenandt, H. A. Weidlich, and H. Thompson, Ber., 1933, 66, 601; J . SOC.Chern. Id., 1933,52,289~.lo J. Eeer, J. R. Billeter, and K. Mie~cher, Hdv. Chirn. A&, 1946,28, 991.l1 R. D. Haworth, J., 1934, 460SHOPPEE STEROIDS AND RELATED COMYOUSDS. 19::(+)-marrianolate, and the absence of unsaturation, apart from the aromaticsystem, is shown by its resistance to hydrogenation with platinum oxide inacetic acid; under these conditions, the free acid (VIa) gives n mixture ofhexahydro-acids which is biologically inactive.( +)-Doisynolic acid is highly estrogenic ; l2 this confirins the originalclaim of Doisy et aZ.3 which was later ~ i t h d r a w n .~ By subcutaneousadministration only 0.7-1 y is required, so that the acid is practically asactive as (+)-cestrone; given orally the threshold dose is 1.5 y, in goodagreement with the value (2.5 y ) reported by Hohlweg and Inhoffen.6 rH PI10II /\I. * ; ?\ I /,A- /\'\4RO{A/ p HOk>{/i H (IX.) ' --I KOH(VIIIa; P, = H.)(VIIIb; R == CH2Yh.)(VIIIc; It == Me.)(XU; It = H.)( X b ; R = CH,Ph.)(Xc; R = Ale.)(XIU; R'= H2 = H.)(XIb; I t 1 -- Rz = Me.)(XTc; R1 = Me, HZ = H.)(XIIa; It' = R2 = H.)(XIIb; R1 = R2 = Me.)(XIIc; R' = Me, It2= H.)rd-c i11 ~ , C O at 3200 \ A / A;:,. ""'fH d \,A/(VII.)( +)-Equilenin benzyl ether (VIIIb) by oxidatioii with potassiumhypoiodite at 20" gives ( +)-7-benzylbisdehydromarrianolic acid (Xb), whichwith acetic anhydride-pyridine at 20" readily affords the anhydride( X I I I ; X = 0), from which the imide ( X I I I ; X = NH) is obtained.Hydrogenolysis of (Xb) with palladium-calcium carbonate gives an 89 yoyield of ( +)-bisdehydromarrinnolic acid 10 ( X a ) .(+)-Equilenin methylcther (VIIIc) 137 l4 with potassium hypoiodite similarly gives the inethylcther of ( +)-bisdehydromarrianolic acid (Xc), demethylated by pyridinchydrochloride l5 a t 170" to (+)-bisdehydromarrianolic acid ( X u ) , which bycontrast wit'h its henzyl ether ( X b ) does not furnish an anhydride butI f K. Miesclier, Hrliv. P h i t ) t . A c t u , 1916, a, 1727.l 3 (:. Sandulescw.\Ir. \Ir. rrschlIng. and .A. Giratrd, ( ' U / t ! ] J t . ~ d . , 1933. 196, 13'7.'I \V. E. Bachmann, LV. C'ole. and .-\. I,. LViIdq, J . Awer. Chem. SOC., 1940. 62, 824.l 5 V. Prey, Ber., 1941, 74, 1219.1tEP.-VOL. XLTV. 194 ORGANIC CHEMISTRY.decomposes when sublimed in a high vacuum. The acids ( X a and Xc) withdiazomethane give dimethyl esters, one carbomethoxy-group of which ishydrolysed with extreme ease. ( +)-Bisdehydromarrianolic acid by sub-cutaneous or oral administration is completely inactive in doses of 1 mg. inthe estrus and the uterus test in rats.10( +)-7-Methylbisdehydromarrianolic acid ([.ID + 102") (Xc) is clearlyone of the four enantiomorphs comprised by the racemic " a "- and" p "-7-methylbisdehydrornarrianolic acids synthesised by Bachmann, Cole,and Wilds.14 Since the oxidation (VIII--+ X) occurs under mildconditions, and presumably without disturbance of the stereochemicalarrangement, configuration a t C, and C, in " natural " (+)-bisdehydro-marrianolic acid (Xa) should correspond with configuration a t C,, and C,,in ( +) -equilenin (VIIIa). A similar configurational relationship should holdbetween " natural '' (+)-marrianolie acid (IVa) a t C, and C, and (+)-estrone(IIa) a t C,, and C1,, and it has been shown l6 that (Xc) is in fact t,he( +)-enantiomorph of the racemic " p "-7-methylbisdehydromarrianolic~acid, m.p. 213", of Bachmann et aZ.,14 from which (-J-)-equilenin wassynt hesised.(+)-Equilenin (VIIIa) or '' a "-dihydroequilenin (IX),17 m. I>.24X",by fusion with potassium hydroxide a t 275" both yield 70--80% of a mixtureof acids from which (normal) * (-)-" a "-bisdehydrodoisynolic acid (XIa)and (iso) * (+)-" "-bisdehydrodoisynolic acid (XIIa) are obtained in 10%yield.10 After removal of phenolic material, the acids are isolated byfractional precipitation of the sodium salts a t given pH. The crude acid(XIa) is highly estrogenic (threshold value in rats by oral application,0.1-0.15 y).; it is purified by conversion with diazomethane and methylsulphate into the methyl ether-methyl ester (XIb). This, after chromato-graphic purification, is hydrolysed with potassium hydroxide at 180" to themethyl ether (XIc), which cannot be esterified with alcoholic hydrochloricacid, and is demethylated by pyridine hydrochloride at 180" to give(-)-" u "-bisdehydrodoisynolic acid (XIa).Similar treatment of the crudeacid (XIIa) gives (XIIb), (XIIc), and finally (+)-" p "-bisdehydrodoisynolicacid (XIIa). " a "-Dihydroequilenin l7 by fusion with potassium hydroxidea t 275" gives the same acids (XIa, XIIa), which are formed in roughly equalproportion and in overall yields of S-lO%. The constitutions of the acidsJ. Heer and K. Miescher, Helv. Chin$. Acla, 1946, 29, 1895.1' L. Ruzicka, P. Muller, and E. Morgeli, ibid., 1938, 21, 1394; R. E. Marker el al.,J . Auaer. Chem SOC., 1937, 59, 768.* I n Parts I-XI1 inclusive of the series entitled " f:ber aestrogene Carbonsauren "(Helv. Chim. Acta, 1944, 27, 1727 et seq.) these acids were distinguished by the prefixesiwrnaal and is0 because it was thought that, despite contrary rotatory powers, thebiologically active acid corresponded configurationally with the biologically activeequilenin, and the physiolopic.ally inactive acid with the physiologically inactiveli-isoequilenin. Tn Part XIIL (ibid., 1946, 29, 1x95; ref.16) it was shown that thisassumption waa untenable ; the terminology previously used was abandoned for thetrivial indices " a "-(= normal) and " /I "-(= iao), and the formulR previously employedwere interchangedSHOPPEE : STEROIDS AND RELATED COMPOUPITDS. 195follow from their conversion by dehydrogenation with palladium-charcoalinto the phenanthrol (VII), and have been proved by total synthesis; forconfigurations, see below.(+)-" p "-Bisdehydrodoisynolic acid (XIIa) is inactive by oraladministration in the estrus test in rats in doses of 5OOy, but (-)-" a "-bis-dehydrodoisynolic acid (XIa) is fully active in rats a t the threshold dose of0.05~; it is thus the most potent estrogen known,1° and is only equalledby one of the synthetic (-+)-doisynolic acids (see below).The (-)-7-methyl-" a "-acid (XIc) is equally active.1°By contrast with the marrianolic acids, the " natural " doisynolic acidsare obtained from (+)-oestrone, " a "-estradiol, (+)-equilenin, and" rx "-dihydroequilenin by very violent treatment-potassium hydroxide a t275" ; inversion of ( +)-equilenin and " a "-dihydroequilenin must occur a tone of the asymmetry centres C,, or C,,, because two diastercoisomerides-the (-)-" a "- and (+)-.' 3 "-bisdehydrodoisynolic acids-are produced.Itmay be noted that tlhe action of potassium hydroxide a t 275" on estriolproceeds without disturbance of configuration a t C,, or CI4, because only asingle (+)-marrianolie acid (IVa) is obtained, which is identical with thatproduced from (+)-estrone by oxidative fission at 20" ; likewise,(+)-estrone and " a "-cmtradiol with potassium hydroxide a t 275" give onlya single ( +) -doisynolic acid (VIa), which should correspond stereochemicallywith (+)-oestrone." Natural " (+)-marrianolic acid (IVa), as the 7-methyl ether-2-methylester, with oxalyl chloride gives the 1-acid chloride reduced by theRosenmund method with palladium-charcoal to the 1-aldehyde (XIV),which is converted by low pressure Wolff-Kishner reduction l8 into( +)-7-methyldoisynolic acid, also obtained less satisfactorily from (XIV) byClemmensen reduction ; this by complete hydrolysis yields ( +)-doispolicacid (VIn),lG identical wit,h the " natural " acid from (+)-estrone (IIa) or" a "-estradiol (111).1Similarly, the 3-methyl ether of (-)-lumioestrone (XV), obtained in 33%yield by irradiation * of (+)-oestrone methyl ether,l6? l9 has been oxidisedM.D. Soffer, &I. B. Soffer, and K. \Ir. Sherk, J . Amer. CIhtttL. Soc., 1945, 67, 1435.* Irradiation of the 3-methyl ether of (+)-equilenin gives only traces of the 3-methylether of ( -)- 14-isoequile~i~, 1196 ORGAN 1C CHEMISTRY,with potassium hypoiodite a t 20" to a (+)-lumimarrianolic acid (XVI);this, by conversion into the 1-aldehyde, Wolff-Kishner reduction of this, andfinally complete hydrolysis, gives ( + ) -1umidoisynolic acid (XVII) .16The doisynolic acids (VIa, XVII), as the 7-methyl ether-methyl esters,by dehydrogenation with palladium-charcoal in acetone a t 250°, giverespectively the 7-methyl ether-methyl esters of (+)-" p "-bisdehydro-doisynolic acid (XIIa) and ( +)- " u "-bisdehydrodoisynolic acid (XVIII) .,O(XIIa.) (XVIII.)(+)-14-isoEquilenin (XIX), which was prepared 2o from ( +)-oestroiieby dehydrogenation with palladium-charcoal, so confirming the original'observation of Butenandt,19 as the 3-methyl ether by oxidation withpotassium hypoiodite a t 20" gives the 7-methyl ether of (-)-" u "-bis-dehydromarrianolic acid (XX),,O which was also obtained 2o by resolution ofthe 7-methyl ether of the (*)-" u "-bisdehydromarrianolic acid of Bachmannct ~ 1 .~ ~ from which (&)-14-isoequilenin was synthesised. Application of thereaction sequence : -CH,*CO,H + -CH2*COC1+ -CH,*CHO -+ -Et,to the marrianolic acid (XX) gives the 7-methyl ether of (-)-" a "-bis-dehydrodoisynolic acid (XIa), so that ( +)-equilenin with potassiumhydroxide a t 275" and (+)-14-isoequilenin by stepwise degradation afford thesame (-)-" u "-bisdehydrodoisynolic acid.20 (&)- 14-isoEquilenin has alsobeen converted 16~ 2o via (-J-)-'' u "-bisdehydromarrianolic acid into (-&)-" a "-bisdehydrodoisynolic acid ; in effect, therefore, it is known that (-)-( 14)-iso-equilenin (XXI) is transformed via ( +)-" a "-bisdehydromarrianolic acid(XXII) into (+)-" o! "-bisdehydrodoisynolic acid (XVIII), so that the enantio-inorphs of 14-isoequilenin give, each to each, the rotationally oppositeenantiomorphs of '' u "-bisdehydrodoisynolic acid [( $)-M-iso --+ (-)-" a ' I ,Finally, the natural ( +)-bisdehydromarrianolic acid (Xu), obtained bymild oxidative fission of (+)-equilenin,lo has been converted by the reactionsequence : -CH,*CO,H -+ -CH,*COCl-+ -CH,CHO ---% -Et,* into the( - ) - i 4 - i ~ --+ ( + I - ' ~ oL y .2 0A . Butenantlt et al., B e y . , 1941,74, 1308; 1942, 75, 1931 ; 1944, n, 393, 394."O J . Heer and K. Rlieschcr. t f d t 9 . C?~;M. --ldu, 1917, 30, 350.* Since this report was writteii. this conversion h:is nlso hscn ~cliieved, n\wicling theI1ig.11-temperature Wolfl-Kishner reduction stage, by hyclrogenation of the benzylinercaptttl : -CH;OH --+ CH;CH(S-CH,Pli), -+ Et, with Raiiey nickel in itqtleorlsethanol a t 100" (k.Heer and K. Jliescher, Helv. C'?L~/~L. .4ctu, 1918, 31, 405)SITOFPEE : S'I'EROT1)S AND RELATED COhlPOUN DS. 197biologically inactive enantiomorph ( +)- " p "-bisdehydrodoisynolic acid(XIIa) .I6 Clearly, (-)-equilenin (XXIII) should give a (-)-bisdehydro-inarrianolic acid (XXIV), which in turn should furnish the (-)-" p "-bis-dehydrodoisynolic acid (XXV), which is already available in principlesince the (&)-" p "-acid has been prepared by total synthesis (see below) ;this conversion has not yet been accomplished but is included for the sake ofcompleteness in the following set of formulae, which give a complete andconsistent picture of the foregoing transformations 21 on the basis that(-t)-cest,rone and ( f-)-equilenin possess the same configuration a t C,, and9(- )-Equilenin0I1(XXIII.)HOI II I €3 \A/( + )-EquileninRisdahydromarrianolic Hisdehydrodoisynolicacids.ncids.CO,H(+I-" B "(.-)-C' p "(XXV.)f (VIIIU.) (XU.)(XIX.) WX.1 (XIU.)0C0,H C0,H3( - ) - 1 1 ~ iso Eqii i leriin (L)-'. Q " (+)-" a "(XXI.) (XXTI.) (XVIII. )The production of the diastereoisomeric acids (XIa) arid (XIIa) from(-1-)-equilenin (VIIIa) by potash fusion may involve inversion a t C14 ; theobservation 2o that the acids ( X u ) and (XIIa) undergo interconversioni n the presence of potassium hydroxide at 295" or of the sodium derivative21 C.W. Shoppee, Xature, 1947, 160, 64. 22 Idem, ibid., 1948, 161, 20719s ORGANIC CHEMISTRY.of ethylene glycol at 190" (eOCH,-CH,Oe, cf. CH,Oe, being more effectivethan OHe for extraction of He) suggests, however, that the primary reactionundergone by ( +)-equilenin with potassiuni hydroxide a t 275" is hydrolyticfission to the (+)-" p "-bisdehydrodoisynolic acid (XIIn), which thenundergoes reversible conversion into the diastereoisoineric ( -)-" a "-acid( X u ) with inversion of configuration at C,.Support for the relationships given in the foregoing table is provided 23by conversion of the 7-methyl ethers of (&)-" a "-bisdehydrodoisynolicacid * (XIa + XVIII) and of (&)-" p "-bisdehydrodoisynolic acid *(XLIa + XXV) via the sequence --COCl+ -CHO --+ -Me, intothe same (&)-7-methoxy-2 * 2-dimethyl-l-ethyl-1 : 2 : 3 : 4-tetrahydrophen-4- (XXV.)(XIIa.)anthrene (XXVI + XXVII).If corresponding transformations were carriedout with the diastereoisomeric (-)-" a "- (XIu) and (+)-" p ')-acids (XIIa)and yielded respectively the enantiomorphs (XXVI) and (XXVII), thiswould prove the postulated occurrence of inversion of configuration a t C,.(&)-" ct "-Bisdehydrodoisynolic acid (XIa + XVIII) is also obtained 24when the (&)-" a "-monodehydrodoisynolic acid (XXVIII and its mirror-image) (prepared by total synthesis, see p. 210) is dehydrogenated as theinethyl ether-methyl ester ; when the 11 : 12-double bond of' this(&)-" a "-monodehydrodoisynolic is reduced, there are obtained two (&)-7-methyldoisynolic acids, (XXIX and mirror-image) and (XXX and mirror-image), which differ only in configuration a t C,, and C,,.24 Both undergodehydrogenation to give ( &)-" a "-bisdehydrodoisynolic acid (XIa $- XVIII),so that the stereochemical arrangement, C,-Et/C,-CO,H : cis, is assured ;the acid (XXIX), m.p. 187-188", is intensely estrogenic (threshold dose0*05y, oral or subcutaneous), whereas the acid (XXX), m. p. 213", isrelatively so inactive (threshold dose, 1OOy) that this small activity may bedue to mere traces of the isorueride.%The liquid keto-ester (XXXI) of R. Robinson and J. Walker 25 andz3 J. Heer and K. Miescher, Helv.Chim. Acta, 1947, 30, 777.24 Idem, ibid., p. 1422.* In the original paper, formula (XVIII), which actually represents the (4- )-'' a "-enantiomorph, is used t o represent the (&)-" a "-acid; likewise, formula (XIIa),corresponding to the (+)-" j? "-enantiomorph, is used to represent the (&)-" j? "-acid.2s J., 1936, 747SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 199Bachman et u E . , ~ ~ with 3 asymmetric centres, should exist in 4 racernicmodifications (XXXI A, B, C, and D) of which 3 have now been obtained(XXVIII.) (XXIX.) m. p. 187" (XXX.) m. p. 713"Pd-C in bleOH ttt 290" or Ni in MeOH a t 110"/41) atms.A XI^ + 'XVIII)~rystalline,~~ and one (XXXI A) has been converted into (+)-oestrone.aThese racemic keto-esters have been converted (for details see p.212) into6 crystalline (j-)-7-methyldoisynolic acids (XXXII A " a ", A " p " ;XXXII B " a " ; XXXII C " a ", C " p ").-4, m. p. 133-135" B " fi ", ----B, m. p. 87-89' D, - A " fi ", m. p. 227" C " u ", m. p. 179"C " fi ", m. p. 189"C , m. p. 127-128" A " a ", m. p. 187"B " u ", m. p. 213"The doisynolic acid structure (XXXII) with 4 asymmetric centres(C2, C,, CI1, C12) can exist in 8 racemic forms corresponding to 16 opticalc, Cl C l l c12CO,H Et H H(XXXIIB.)I ''f" I I iI I II t IIIIIL L > >"' B "Ic2 c, Cl, Cl,CO,H Et H HI I II I (Me) IIIIII (Me)I I 12 . l ~ G . Anner and K. Miescher, Experientia, 1948, 4, 25.20 W. E. Bachmann, S. Kushner, and A. C. Stevenson, J . Anaer.Chem. SOC., 1942,64, 974200 ORCiANIC CHEMISTRY.isomerides; these are shown in horizontal pairs in the accompanyingdiagrams, in which the usual convention of broken and full lines for represent-C', (3, CIl C12CO,H Et H HII (Ale)I II II I1 I (Me)I '1 'ation of (a)- and ($)-orientated groups is employed. As a convenientreference point, the angular methyl group at C, (=C,, of the originalsteroid skeleton) is also shown." Natural " ( +)-7-methyldoisynolic acid is represented by the left-handmember of the enantiomorphous pair (XXXII A " p "), and the synthetic7-methyldoisynolic acid, m. p. 227", is regarded as the racemate (XXXII A" p "). The intensely active (-J-)-7-methyldoisynolic acid, m. p. 187"(XXIX), is identical with the racemate represented by (XXXII A " cc ") ;it thus possesses the stereochemical arrangement at C,, and CI2 shown in(XXXIII), corresponding to the trans-B/C ring-union present at C, and C, inthe natural steroids (as XXXIV).CO,HThese configurational assignments are confirmed by dehydrogenation ;(XXXII A " a ") gives the intensely estrogenic (-J-)-" a "-7-methylbisdehydro-doisynolic acid (XIa + XVIII), whilst (XXXII A " p ") gives the relativelynon-cestrogenic (A)-" p "-7-methylbisdehydrodoisynolicacid (XIIa + XSV) .2*The (&-)-7-methyldoisynolic acid (XXX), in.p. 213", is identical with theracemate (XXXII B " a "), which appears to be that of (+)-7-methyl-umidoisyiiolic acid (XVII), which is itself the right-hand member of thSHOPPEE STEROlDS AND RELATED COMPOUNDS.20 1enantiomorphous pair. The (-J-)-7-methyldoisynolic acid, m. p. 189",by dehydrogenation gives (-J-)-" p "-7-methylbisdehydrodoisynolic acid(XIIa + XXV), and is therefore to be represented by either the racemate(XXXII C " p ") or the racemate (XXXII D " p "); these differ only inorientation at C,, and CI2, which would correspond in either case with acis-B/C steroid ring-union.Compound. M. p.( +)-" a "-Bisdehydrodoisynolic acid (syn-thetic), ..........................................(-)-" a -Bisdehydrodoisynolic acid (syn-thetic!, ..........................................( -)-" a -Bisdehydrodoisynolic acid[from ( + ) -equilenin] .....................( +)-" /3 "-Bisdehydrodoisynolic acid[from ( +)-equilenin] .....................( 4-) -" a "-Bisdehydrodoisynolic acid (syn-thetic) ..........................................( f )-" "-Bisdehydrodoisynolic acid (syn-thetic!,. .........................................( + ) - " a -7-Methylbisdehydrodoisynolicacid (synthetic) ..............................( - ) -" a "- 7-Methylbisdehydrodoisynolicacid (ynthetic) ..............................( - )-" a -7-Methylbisdehydrodoisynolicacid [from (+)-equilenin] ...............( f )-" a "-7-Methylmonodehydrodoisynolicacid (synthetic) ..............................( &- )-" /3 "-7-Methylmonodehydrodoisynolicacid (synthetic ) ..............................(-/-)-D,$synolic acid [from (+)-cestrone or(+)-Lumidoisynolic acid [from (-)-lumi-cestrone J .......................................( f )-7-Methyldoisynolic acid (XXIX) (syn-thetic) ..........................................( f)-7-Methyl-lumidoisynolic acid (XXX)(synthetic) ....................................( + )-Bisdehydromarrianolic acid [from( + ) -equilenin] ..............................(+)-Marrianolie acid (from (+)-aestroneor cestriol) ....................................(+)-(Estrone ...................................." a "-CEstradiol .................................Diethylstilbaestrol ..............................a -cestradiol] ..............................159-1 61 O159-160161-1622 5 6-2 5 8204238-240220-2212 19-22 12 1 9-22 1168-170171-173199-200152-154187213243-245223-224 -ICEstrus thresholdvalue in rats.BY a[ a l y 2 "in single sub- By oralEtOH.cutaneous adminis-adminis- tration :tration : y. y.+ 115"-116-117+ 27--+ 100-5- 99.5- 103--+ 102+ 70--+ 103100.050.05200-3000.1-0.15> 100150.050.05--0.7-1.02500.05100> 1000+ 90 >loo0- 0-7- 0.3-0.4- 0.3-0.4100.050.05-0.1-0.15> 500150.050-050.131.5-0-05100> 1000> 100020-3020-300.7-1.0Compiled mainly in respect of biological data from refs. 10, 12, 23, 24, 27, and 28;cf. also E. Tschopp, Schweiz. med. Woch., 1944, 74, No. 51, 11; HeEv. Physiol. Actu,1946, 4, 271, 401.Following alteration of the nomenclature (ref. 16), n became " a ", {S~O became " /3 ' 7 .The Table summarises the chemical and biological properties of theniarrianolic and doisynolic acids ; for comparison, the threshold activitiesof (+)-cestrone, " a "-cestradiol, and diethylstilboestrol are included.The2 7 R. Rometsch and K. Miescher, Helv. Chim. Actu, 1946,29, 1231.28 G. Anner and K. Miescher, ibid., p. 1889; Experientia, 1946, 2, 409203 ORGANIC CHEMISTRY.bisdehydrodoisynolic acids are notable in possessing the same activity in ratswhether administered orally or subcutaneously ; moreover a single dosegives the full effect. The sodium salt of racemic " a "-7-methylbisdehydro-doisynolic acid has been introduced for clinical use under the name" fenocyclin," but up to now the required dosage has not been definitivelysettled.R2 CO,H "/ K2 CO,H /\I.*' ItL CO,H p y o z H /Jg:.../yy,,,-l<l q . ) \ i g R 1 #A/\/ II I H '. *I3'RO\?/\//(-)-" fl "R d \/v Rok,\/ l%,,/(-)-'' 0. "- (+)-" 0. " (+)-" fl ''Bisdehydrodoisynolic acids. Bisdehydrodoisynolic acids.In the bisdehydro-series, high estrogenic activity appears to depend oncis-orientation of C1-Et/C,-C02H ; it is, however, a most striking fact that,in addition, such a cis-relationship must involve (a)- and not (@)-orientatedgroups. Thus the (-)-" a "-bisdehydrodoisynolic acid (which correspondsin regard to configuration a t C, with the configuration a t C,, in aEE the steroidhormones) is some 200 times more potent than the enantiomorphous(+)-" a "-acid. The homologous racemic " a "-bisdehydrodoisynolic acidsin which R1 = R2 = Me, and R1 = Et, R2 = Pra, are almost as active as racemic" a "-bisdehydrodoisynolic acid itself (R1 = Et, R2 = Me), but the activityfalls to about one half when R1 = R2 = Et.29 If alkyl groups are absent fromthe 1- and 2-positions, or only from the l-position, estrogenic powerlargely disappears,29 so that activity appears also to be connected with thepresence of trans-orientated alkyl groups. Inversion of configuration atposition 1, from CL-Et/C,-CO,H : cis in (-j-)-" CL "-bisdehydrodoisynolic acid toC,-Et/C2-C02H ; trans in ( j-)-" p "-bisdehydrodoisynolic acid is here sufficientto depress activity by a factor of more than 1000; of the homologous(5)''' p "-bisdehydrodoisynolic acids, only that with R1= R2 = Me showsappreciable activity (oral threshold value, 7 0 ~ ) .~ ~ The (&)-7-methyl-l : 2-cyclopentano- (XXXV) and - 1 : 2-cycEohexano-bisdehydrodoisynolic acids(SXXVI), which correspond stereochemically with the (&)-" p "-bisdehydro-acids in that the cis-orientated alkyl groups of the latter (R1 = Et ; R2 = Me, orR1 = R2 = Et) have become united in a ring, have little potency (thresholdvalue 200-700yy by a single subcutaneous injection) .30/\I-CO,H /)A(&--bX/ (XXXVII * )f/ sz ,, \(' /V\A p 5 31eol 11 I \/\/(XXXVI.)Numerous derivatives of ( &)-" a "-bisdehydrodoisynolic acid have beenprepared; 31 these include the 7-Me, -Et, -Pr', -Bus, -CH,*CH:CH,,29 J. Heer and K. Miescher, Helv. Chim. Acta, 1945, 28, 1504.3O L. Ehmanri and K.Miescher, ibid., 1947, 30, 413.31 G . Anner, J . Heer, and K. Miescher, ibid., 1946, 29, 1071.If0*\A/(XXXV.SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 203-isoamyl, and -CH,*CH,*NEt2 ethers and the corresponding methyl esters,the Me, Et, Pray Bus, and -CH,*CH,*NEt, esters of the 7-methyl ether,and the P-diethylaminoethyl ester of the 7-p-diethylaminoethyl ether.Increasing chain-length of the group R in RO- (at position 7) or of thegroup R in -CO,R (at position 2) leads to decreasing activity; in the basicderivatives (R = CH,*CH,*NEt,) activity is much decreased. Finally, deriv-atives of ( &)-" a "-7-methylbisdehydrodoisynolic acid containing variantsof the carboxyl group -CO*R, where R = C1, Me,* CHN,, CH,*OAc,f. 31 orNH2,23 and the analogues in which this group is replaced by -CN23 and-CH,*CO,H,31 exhibit activity only at very high dosages ; on the other handthe derivatives in which the carboxyl group is modified to -CHO, -CH,*OH,and -CH,*OAc are almost as active by oral or subcutaneous application asthe parent acid, and double the duration of the estrus state.= Even the(&)-phenanthrol methyl ether (XXVI + XXVII), in which the originalcarboxyl group has become a methyl group, is fairly active (threshold value5-2Oy, by oral or subcutaneous application) with a strikingly protractedaction (> 55 days after injection of 1007, cf.11-13 days after injection of1 0 9 of (&)-" a "-bisdehydrodoisynolic acid).= The analogous aldehyde andalcohol from ( &) -" p "-7-methylbisdehydrodoisynolic acid are practicallyinactive.,3In the (-J-)-" CL "-bisdehydro-series, the 7-methyl ethers possess the sameactivities, by oral or by subcutaneous administration, as do the freehydroxy-acids; this is in striking contrast to the activity ratios of(+)-oestrone, " a "-oestradiol, and (+)-equilenin to their 3-methyl ethers(1 : 14, 1 : 29, and 1 : 3.3 respectively). In fact, the presence of a 7-hydroxylor 7-methoxyl group appears t o be less important than that of the properalkyl groups at C, and C, properly orientated [ ( a ) ] ; $ thus the synthetic(-j-)-7-deoxybisdehydrodoisynolic acid (XXXVII) 33 is active at a level of57, i.e., is only 50 times weaker than (&)-" cc "-bisdehydrodoisynolic acid,and is provisionally assi&ed to the " a "-series.In the monodehydro-series, the (-+)-" a "-7-methyl acid (XXVIII)(CI-Et/C,-CO,H : cis) is active at a level of 0.17, whilst the (j-)-" p "-7-methylisomeride (C,-Et/C,-CO,H : tram) is active at a level of 37, so that hereinversion of configuration at C, only decreases activity by a factorof 30.Various analogues of the bisdehydrodoisynolic acids with modifiedring-systems have been prepared and tested for biological activity. Thetwo stereoisomeric ( j-)-C-norbisdehydrodoisynolic acids represented by(XXXVIII) are both active by a single subcutaneous injection, but only at32 W.E. Bachmann and A. L. Wilds, J . Amer. Chem. SOC., 1940, 62, 2084.33 G. h e r and K. Miescher, Helv. Chim. Acta, 1946, 29, 586.* This ketone has no progestational activity.*f This ketol-acetate has no cortical activity.Racemic 3-deoxyequilenin is inactive at a level of 500y; 32 the activity of( &)-equilenin has apparently not been determined, but may be roughly estimated atitbout 6% from the known activities l4 of the enantiomorphs [( +)-equilenin, 30y;( -)-equilenin, 40OyI204 ORGANZC OHEMTSTRY.a level of about 1000y.34 Surprisingly the ( &) -B-normonodehydrodoisynolicacid (XXXIX), which may also be formulated as (XL), is highly active bothas the hydroxy-acid and as the 7-methyl ether at a level of about ly; thelower homologue (XXXIX with Me for Et) is also active at a level ofabout 7y.35(XXXVIII.) (XXXIX.) (XL.)Attempts to simplify the tricyclic structure generally lead to inactivecompounds. The two stereoisomeric ( 5)-tetrahydronaphthalenecarboxylicacids represented by (XLI) are both inactive at a level of 1oOOy,33* 363 37whilst the isomeric (&)-acid (XLII) is inactive at a level of 1OOy.On the otherhand, (&-)-dimethylethylallenolic acid (XLIII; R1 = R2 = Me, R3 = Et),which bears a formal resemblance to G-norbisdehydrodoisynolic acid(XXXVIII), and its methyl ether are highly active (threshold value1--1*5y, by single subcutaneous or oral application), and evoke the samephysiological responses as the natural oestrogenic hormone^.^' The methylethers of dimethyl-, trimethyl-, and methylethyl-allenolic acids (XLIII ;R1= R2 = Me, R3 = H ; and R1= H, R2 = Me,R3 = Et) are inactive at IOOOy, active at 1Oy, and very active at lOOyrespectively, whilst the unsaturated analogue (XLIII; R1 = R2 = Me,R3 = XHMe) is also active at a level of 25-50~.3~R1= R2 = R3 = Me;Rl-.:CO, /yyH-R3 C ; - C 0 2 H/ --Et>red\, II H I /I 13(XLIII.)A??(XLI.) (srm.)In the doisyliolic acid series, information bearing on the relationshipbetween structure and biological activity is less extensive.“ Natural ”(+)-doisynolic acid (VIa or XXXII A “ p ”; left-hand member ofenantiomorphous pair) is highly active (threshold value about l y ) , despitepossession of the unfavourable C,-Et /C,-CO,H : trans-arrangement . Theorientations of the hydrogen atoms at C,, and CI2, however, correspond withthose at C, and C, of the natural steroids. In (+)-lumidoisynolic acid(XVII or XXXII B “ o! ” ; right hand member of enantiomorphous pdr)(threshold value, 250y), the favourable C1-EtjC2-C0,H : cis configuration ispresent but involves the unfavourable ($)-orientation ; again configurationat C,, and C,, corresponds with that at C, and C, in the natural steroids.34 J.R. Billeter and K. Miescher, Helv. CI~irn, Acta, 1946, 29, 859.35 G. Anner and K. Miescher, ibid., 1027, 30, 544.36 J. H. Hunter and J. Kormrtn, J . L4??ier. Ckem. SOC., 1947, 69, 21%4.31 R . Courrier, A. Horeau, and J. Jncques, Compt. rend., 1946, 222, 961, 1113; 1946,224, 862, 1401 ; Compt. rend. SOC. Biol.. 1947, 141, 159, 747SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 205Synthetic (&)-7-methyl-lumidoisynolic acid (XXX or XXXII B " a ") isactive at a level of lOOy ; if this activity is really attributable to the racemicIumi-acid (cf.p. 198) it suggests that the as yet unknown (-)-lumidoisynolicacid (XXXII B '( a " ; left-hand member of the racemic pair) must be activea t a level of about 50y. This seems unlikely, because although the( - )-lumiacid has the favourable C,-Et/C,-CO,H : &-arrangement with thoadvantageous (a)-orientation, configuration a t C,, and C,, is enantiomorphouswith that present a t C, and C, in the natural steroids.(XXXIIA " /3 ".) (XXXII B " a ¶'.)Left-hand member Left -hand member Right-hand memberof raceinic pair of racemic pair of racemic pairThe synthetic (~)-7-inethyldoisynolic acid (XXIX or XXXII A " a "),m. p. 187", is active a t a level of 045y and so equds (-)-" 0: "-bisdehydro-doisynolic acid as the most potent estrogen yet discovered; both theenantiornorphs possess the favourable C,-Et /C,-CO,H : cis-arrangement,but one has the favourable (a)-orientation and the other tho unfavourablc(P)-orientation.If, as is suggested by the available evidence, configurationa t C,, and C,,, corresponding with that present at C, and C, in the naturalsteroid hormones, is important for high oestrogenic activity, then it seemsprobable that the left -hand member of the enantiomorphous pair(XXXII A " a ") will be found to be active at a level of about 0*025y, whilstthe right-hand member will be found to be relatively inactive.(XXXII A " a ".)Reduction of the 7-methyl ether of (+)-doisynolic acid (Vlu or XXXIl A" p " ; left-hand member of the enantiomorphous pair) to the correspondingaldehyde and alcohol leads to decreased activity (threshold doses 4-77 and40-6Oy respectively23).It is not known whether a 7-hydroxyl group isessential for oestrogenic activity ; the synthetic ( &)-" 01 "-7-deoxydoisynolicacid 24 represented by (XLIV) should be active, whilst the isomeric(-+)-7-deoxydoisynolic acid 24 (XLV) may bc expected to be inactive.(XLIV.) /ykJH""t 4 'Polyhydro-analogues of the inarrianolic and doisynolic acids l1ai.e alsobeen prepared ; dehydroisoandrosterone (XLVq and isoandrosteron206 ORGANIC CHEMISTRY.(XLVII) by oxidation with potassium hypoiodite give the polyhydro-marrianolic acids (XLVIII) and (XLIX) which, after conversion into the7 -acetoxydime t hyl esters, hydrolysis to the 7 - hydroxy-2 -met hyl esters, andreacetylation, are converted by the reaction sequence a t C, : -GH,*CO,H --+-CH,*COCl _f -CH,*CHO -+ -Et, into the polyhydrodoisynolic acids(L) and (LI), from which the related acids (LII), (LIII), and (LIV) areobtained.The acids (L-LIV) are all androgenically inactive (comb test),and the acid (L) has also no ceshogenic a~tivity.~80 0KIOI1 \/.:CO,H/i-CH2*C0,H +H(XLVIII.) (XLIX. )(XLVII.)HO(XLVI.), HO HO H HO H(L.) (LIII.) w.1(LII.) (LIV.)Syntheses.-The total synthesis of the bisdehydrodoisynolic acids 39commences with Cleve's acid (LV) which is converted via l-iodo-6-methoxy-naphthalene (LVI) and 6-methoxy-l-(2-bromoethyl)naphthalene (LVII)in six stages into the keto-ester (LVIII) of Bachmann, Cole, and Wilds,l* thecarbonyl group only of which reacts with ethylmagnesium bromide to give in80% yield a mixture of the l-epimeric carbinols (LXa) and (LXb) in which(LXa) largely predominates.An alternative route involving addition ofsodium acetylide in liquid ammonia to (LVIII) gives, in the ratio of about20 : 1, the epimeric carbinols (LIXa) and (LIXb), which are convertedrespectively into (LXa) and ( L X b ) by hydrogenation with platinum inethanol.Iodine in chloroform or, better, 90% formic acid for a few minutes a t100" dehydrates the carbinol (LXa) to a mixture of stereoisomeric olefins38 J. Heer and K. Miescher, Helv. Chint. .4cta, 1947, 30, 786.39 J. Hem, J. It.Billeter, and K. Miegcher, ibid., 1945, 28, 1342SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 207(LXIa) and (LXIB); either pure individual by treatment with 90% formicacid at 100" gives a mixture containing approximately equal amounts ofboth isornerides. The olefins (LXIa) and (LXIb) are hydrolysed bypotassium hydroxide at 160" to the corresponding acids (LXIIa) and (LXIIb),either of which as the sodium salt by hydrogenation with nickel in aqueous-alkaline solution furnishes a mixture of (&)-" a "- and (&)-'' p "-7-methyl-bisdehydrodoisynolic acids (XIc) * and (XIIc); * hydrogenation in thepresence of excess of sodium hydroxide leads to the (-J-)-" u "-acid almostexclusively, whilst in sodium carbonate solution the (-+)-" u "- and (&)-" p "-/CH,Br'?HZ I NH,/\/\ /\A /\AM ~ O I 11 I + JIeol 11 I + HO,SI 11 1 \A/ \A,/ \/\/(LVII.) (LVI.) GV-1L5 t 11 g I%r.I.A --C'(?,Me pp,&3 : I //\/ \fo* ---CGCH // /\JIed 11 \/\/11 6 stitgesI ')I f O H31eo\/\/\/\/ I H H 0(XIC.) (XIIC.) -> (XIIa.) (Xla.)ROH at 'Loo 1 or Iyridine hydrochloride at 1 8i0° +acids are formed in the ratio of 2 : 3.The mixture of acids obtained byhydrogenation in excess of sodium hydroxide solution is dernethylated byenantiomorph in racemates.* In this section a single formula will be used to represent the individual and it208 ORGANIC CHEMISTRY.potassium hydroxide at 200" or by treatment with pyridine hydrochloride at180" t o give, after removal by fractional precipitation of traces of (&)-" p "-bisdehydrodoisynolic acid (XIIa), an 84% yield of (3)-" a "-bisdehydro-doisynolic acid (XIa), the ultra-violet absorption spectrum of which isidentical with that of the (-)-enantiomorph from (+)-equilenin or" O! "-dihydroequilenin and resembles that of (+)-equilenin and ofp-naphthol.The pure racemates (XIc) and (XIIc) have also beendemethylated to (*)-" a "- (XIa) and (*)-" p "-bisdehydrodoisynolic acid(XIIa) respectively ; the reverse transformations are accomplished byesterification with diazomethane, methylation, and hydrolysis of the tertiarycarbomet'hoxy-group with potassium hydroxide at 160".An alternative route proceeds by combined hydrolysis and demethylationof the mixed olefinic methoxy-esters (LXIa) and (LXIb) with ethanolicpotassium hydroxide under pressure a t 200", whereby only a single olefinichydroxy-acid (LXIII) can be isolated.This acid is also the sole product ofcorresponding treatment of a mixture of the olefinic methoxy-acids (LXIIa)and (LXIIb); it is related in respect of configuration to the olefins (LXIb)and (LXIIb), since, by methylation and esterification (diazomethane), theolefinic methoxy-ester (LXIb) is regenerated. Hydrogenation of the acid(LXIII) with nickel in excess of sodium hydroxide gives an 86% yield ofalmost pure (5)'" a "-bisdehydrodoisynolic acid (XIa).(LVII.) (LXIV.) (LXn, LXb.)A simplified synthesis has been described by Anner and Miescher;6-methoxy-l-(2-bromoethyl)naphthalene (LVII) by treatment witha-propionylpropionic ester gives an 80% yield of the keto-ester (LXIV),which in 800/, sulphuric acid a t 0" undergoes an intramolecular additionreaction to give both the epimeric carbinols (LXa) and (LXb) with smallquantities of the corresponding olefins (LXIa) and (LXIb). The reaction isnotable as furnishing principally the primary product of cyclisation ; neitherketonic nor acid hydrolysis of the keto-ester (LXIV) occurs to any appreciableextent.For preparative purposes, the crude cyclisation product isdehydrated and hydrolysed to the acids (LXIIa) and (LXIIb), and themixed acids are reduced with nickel in strongly alkaline solution to givemainly the highly estrogenic ( &)-" 01 "-7-methylbisdehydrodoisynolic acid(XIc) accompanied by a little of the biologically inactive (A)-'' /3 "-7-methyl,The (3)-" a "-bisdehydrodoisynolic acid (XIa) obtained synthetically hasan estrogenic threshold value of 0 - l y by oral administration in rats, i.e., itsactivity is about one-half of that of the (-)-enantiomorph (threshold value0.05y) obtained from (+)-equilenin.It seemed probable that almost thewhole activity of the racemic mixture is dire to the (-)-enantiomorph ; this,acid (XIIC)SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 200has been shown to be so by resolution of the synthetic product.27 (A)-'' a "-7-Methylbisdehydrodoisynolyl chloride 31 is converted by being heated withL-menthol a t 100-110" in nitrogen into the L-menthyl esters, the solubilitiesof which in acetone differ by a factor of 20. The (+)- and the (-)-L-menthylesters are hydrolysed by n-propyl-alcoholic potassium hydroxide at 170" and155" respectively without racemisation ; the (-)-" a "-7-methylbisdehydro-doisynolic acid so obtained is identical with the 7-methyl compound (XIc)from ( +)-equilenin. Demethylation of the (+)- and (-)-" ct "-7-methyl-bisdehydrodoisynolic acids with hot hydrobromic-acetic acid gives the( +)- and (-)-" a "-bisdehydrodoisynolic acids, of which the latter is identicalwith the acid (XIa) from natural (+)-equilenin.By using the route [LVII + LVIII --+ LX + (XIa + XIIa)], andcombining the modified keto-esters (LXV; R2 = Me, Et, Pr"; cf.XVII)with alkylmagnesium bromides (R1 = Me, Et), Heer and Miescher 29 havesynthesised various homologues, (LXVI) and (LXVII), of the (&-)-" a "- and" P "-bisdehydrodoisynolic acids.The lower homologues (LXVIII), inwhich there is no substituent a t the 1-position so that the " a "- and" p "-series coalesce, are obtained by Clemmensen reduction of the keto-esters (LXV), hydrolysis, and demethylation.R2Zn-Hg, HCl, etc. ___, //\/\/(LXV.) (LXVIII.)MeOl 11 I (3 stages)\/\/R'MgBr, eta.(5 stages)R2 = H or Me(LXVI.) (LXVII.)R1= = Me or Et; R1= Et, R2 = PraSimilarly, by use of the simplified route [LVII + LXIV --+ LX ---+(XIa + XIIa)], Anner and Miescher, employing ethyl acetoacetate, obtainedthe keto-ester (LXIX; R1 = Me), which undergoes cyclisation withsimultaneous hydrolysis to give the acid (LXXI), although some isdehydrogenated to the phenanthrenecarboxylic ester (LXX) ; the acid(LXXI) by reduction in strongly alkaline solution gives a single acid(LXXII ; R = Me) demethylated to yield the (5)-bisnorbisdehydrodoisynolicacid (LXXII ; R = H).A further variant used by Anner and Miescher 33utilises the keto-esters (LXIX), which, after introduction of the " angular "methyl group, are treated with alkyl- or aryl-magnesium bromides to yieldcarbinols (LXXIII) ; these by cyclisation give directly (&)-l : 1-disubstituted7 -met hylbisdehydrodoisynolic acids (LXXIV) 210 ORGANIC CHEMISTRY,Miescher and Anner also describe the preparation from 1-(2-bromo-ethy1)naphthalene (LXXV) of a single ( 5 ) -7-deoxybisdehydrodoisynolic acid(LXXVI) thought to belong to the " a "-series, and, from m-methoxyphenyl-bromide (LXXVII), of two stereoisomeric ( &)-tetrahydronaphthalene-carboxylic acids represented by (LXXVIII) .The (A)-" a "- and (-J-)-"p' "-monodehydrodoisynolic acids (XXVIII,LXXXI; R = H) have been synthesised by Anner and Miescher; 28 theketo-ester 26 (LXXIX) with ethylmagnesium iodide gives a single olefin(LXXX; R = Me), also obtained by dehydration of the carbinol(LXXXII) ; the olefin by hydrogenation with nickel in an alkaline mediumyields the highly estrogenic (&)-" a "- (XXVIII ; R = Me) and relativelyinactive (A)-'' p "-7-methylmonodehydrodoisynolic acid (LXXXI ; R = Me) ;these are demethylated by pyridine hydrochloride at 170" to the respectivehydroxy-acids (XXVIII, LXXXI; R = H).Hexahydrophenant hrene compounds with an unsaturated side chain arefound to pass very readily into tetrahydrophenanfhrene derivatives withmigration of the extracyclic double bond into the ring system.Thus theketo-ester (LXXIX) with ethylmagnesium bromide gives not only thSHOPPEE : STEROIDS AND RELATED COMPOUNDS. 21 1carbinol (LXXXII) [also obtained from (LXXIX) by addition of acetyleneand reduction of the resulting ethinyl compound] and the olefin (LXXX;R = Me), but also themethyl esters of both (-+)-" a "- and (&)-" c3 "-7-methyl-bisdehydrodoisynolic acids (XIb) and (XIIb). Attempts to dehydrate thecarbinol (LXXXII) invariably involve migration of the resulting unsaturatedlinkage, and the rearranged compound may form the main product ; thus thecarbinol (LXXXII) and the olefin (LXXX; R = Me) by treatment witha trace of iodine in boiling chloroform solution give exclusively (afteralkaline hydrolysis) (&)-" a "- and (-J-)-" p "-7-methylbisdehydrodoisynolicacids (XIc) and (XIIc). Similarly, although demethylation of the methoxy-olefin (LXXX; R = Me) with pyridine hydrochloride a t 170" gives thehydroxy-olefin (LXXX; R = H) without rearrangement of the extracyclicunsaturated linkage, alkaline hydrolysis a t 170" induces isomeric change togive mainly (&)-" a "-bisdehydrodoisynolic acid (XIa).The acid, m. p.218", obtained in this way by Hunter and Hogg40 is probably a mixture of(&)-" a "- (m. p. 204") and (*)-" p "-bisdehydrodoisynolic acid (m. p. 238").(LXXXII.) (XI-) (LXXXIII.)Catalytic reduction of the 11 : 12-double bond in (LXXXI) is difficultbut proceeds by means of palladium-acetic acid to give a non-crystallinemixture of stereoisomeric acids (LXXXIII; R = Me) together with some(&)-" a "-7-methylbishydrodoisynolic acid (XI ; R = Me) ; the amorphousmixture of doisynolic acids (LXXXIII ; R = H) obtained by demethylationis highly estrogenic, but this activity may be due to contamination by smallamounts of (&)-" a "-bisdehydrodoisynolic acid (XI ; R = H).The synthesisof (LXXXIII; R = H) by hydrogenation of the olefin (LXXX; R = Me)and subsequent complete hydrolysis has also been reported by Hunter andHogg; 4o the high estrogenic activity of their amorphous product couldsimilarly be due to the presence of (j-)-" a "-bisdehydrodoisynolic acid (XT).'O J.H. Hunter and J. A. Hogg, J . Arner. Chenz. SOC., 1946, 68, 1676212 ORGANIC CHEMISTRY.Reduction of the 11 : 12-double bond of the methyl ester of (&)-" a "-7-methylmonodehydrodoisynolic acid (XXVIII) with sodium in ethanol-liquid ammonia a t - 40" leads, after hydrolysis of the neutral reactionproduct with potassium hydroxide at 170", to the intensely estrogenic(&)-" a "-7-methyldoisynolic acid (XXIX), m. p. 187", and the less active(&)-'' p "-7-methyldoisynolic acid, m. p. 213" (XXX) : 24(XXVIII.) (XXIX .) (XXX.)These acids are also obtained, inter alia, when the crystalline (&)-7-methoxy-keto-esters (XXXI A, B, and C) are converted into (&)-7-methyldoisynolicacids (XXXII) by the following route, (XXIX) being identical with(XXXII A " a ") and (XXX) with-(XXXII R " a ") : 24(XXXI A, B, and C.)I,,,,(,~=cHM~ A ' - - C O , H -+ lT,,Pt1(XXXII ,4 " a " , m.p. 1X7".)(XXXII A " fly', m. p. 217".)(XXXII B " a " , m. p. 213'.)(XXXIIC " a " , m. p. 179'.)(XXXII C '* p", m. p. IW".)Reduction of ( &)-" a "-7-methylmonodehydrodoisynolic acid (XXVIII)with Rupe nickel a t 80"/15 atms. in presence of sodium hydroxide leads toloss of the methoxyl group with production of (&)-'' a "-7-deoxydoisynolicacid (XLIV) ; 24 under the same conditions the methyl ether of " natural "( +)-doisynolic acid gives a (+)-7-deoxydoisynolic acid (XLV).= Moredrastic conditions (Rupe nickel a t 110"/48 atms.; absence of alkali)dehydrogenate (XXVIII) to (&)-" a "-7-methylbisdehydrodoisynolic acid6-Methoxy- 1 - (2-bromoethy1)naphthalene (LVII) by condensation withthe potassium derivative of ethyl cyclopentan-%one- or cyclohexan-%one-1 -carboxylate gives the keto-esters (LXXXIV) and (LXXXV), which hgcyclo-dehydration in ether with 80% sulphuric acid at - lo", hydrolysis,and subsequent hydrogenation with platinum in acetic acid, give respectively( rf)-l : 2-cyclopentano- and -1 : 2-cyclohexano-7-methylbisdehydrodoisynolic(XI) .SHOPPEE : STEROIDS AND REIJATED COhIPOUNDS.213acids (XXXV) and (XXXVI) ; hydrolysis with aqueous-methanolicpotassium hydroxide a t 210" gives the corresponding hydroxy-acids.mCO,Xti C0,Et CO,Ho='- '1 f\il SO% H,SO,at - loo -- + Me01 II I \V\/(XXXV.)COzEt(LXXXV.) (XXXVI.)A single ( -+)-B-normonodehydrodoisynolic acid (XXXIX or XL) hasobtained35 by application of the method of Heer, Billeter, andMiescher 39 to methyl 1-keto-7-methoxy-2-methyl-1 : 2 : 3 : 4-tetrahydro-fluorene-2-carboxylate (LXXXVI), which was synthesised from ethylm-met hoxyphenylmet hylmalonat e and the ethyl ester-chloride of glutaricacid.35 The lower homologue (XXXIX or XL wit'h Me for Et) has alsobeen prepared.(LXXXVI,) (XXXIS.) (XL.)Proin 6-methoxy-1 -chloromethylnaphthalene (LXXXVII) Billeter andMiescher a have obtained the pair of stereoisomeric ( +) -P-norbisdehydro-doisynolic acids represented by (XXXVTU).H,CI(LXXXVII.) (LXXXVIII.) (LXXXIX.)(XC.) (XCI.) (XXXVIII.)The ketone (LXXXVIII) from (LXXXVII) gave the glyoxylate(LXXXIX), but this could not be caused to lose carbon monoxide214 ORGANIC CHEMTSTRY,Condensation of (LXXXVII) with methyl a-propionylpropionate yielded(XC), cyclised by sulphuric acid with simultaneous dehydration to twostereoisomeric olefins (XCI) ; hydrolysis, reduction, and demethylation ofthese gave respectively the two (A) -C-norbisdehydrodoisynolic acidsrepresented by (XXXVIII). The same pair of olefins (XCI) have alsobeen prepared from l-keto-7-methoxy-1 : 2 : 3 : 4-tetrahydrophenantlirene(XCII) ; 149 4 1 ring fission of this cannot be achieved with hypoiodite,permanganate, or chromium trioxide, but is smoothly accoinplished byconversion into the hydroxymethylene compound (XCIII) and oxidation ofthis with alkaline hydrogen peroxide.The resulting acid as the dimethylester (XCIV) by the Dieckmann reaction gives the keto-ester (XCV). Acidhydrolysis of this gives the ketone (LXXXVIII) from which the keto-ester(XCV) can be regenerated by condensation with methyl carbonate.42Angular methylation of (XCV) yields the keto-ester (XCVI) from which bythe use of ethylmagnesium bromide, or, better, by condensation with sodiumacetylide in liquid ammonia a t - 60" and hydrogenation with platinum, thecarbinol (XCVII) is obtained. Dehydration of (XCVII) with hot 90%formic acid gives the two stereoisomeric olefins (XCI).(XCII.) (XCIII.) / (XCIV.)Coumarones analogous to C-norbisdehydrodoisynolic acid (XCVII) havebeen synthesised ; 43 2-( 1 -hydroxynaphthyl) ethyl ketone condenses withethyl a-bromopropionate to give the ester (XCVIII) which during distillationpasses into the ketone (XCIX); this is cleaved by hot alcoholic potassiumhydroxide to yield the olefin (C) which does not yet appear to have beenreduced to the acid (CI)./\ J+gY2" \/id0~-~-CO,Et I1 ' y = C H M e o--I- CO,H-CHMe ?--[A_j ,\,==I -+ -+ ( I 1 I H/\A,/ -I \/\/(XCIX.) (C.1 (CI.1I II I\A/(XCVIII.)A. Butenandt and G. Schramm, Ber., 1935, 68, 2083; G. Huberland, ibid., 1937,70, 169.42 N. A. Preobrajenski, N. N. Schtschukina, and R. A. Lapina, ibid., 1936,69, 161.5.43 D. Molko and C. Mentzer, Compt. rend., 1946,223, 333SHOPPEE : STEROIDS AND RELATED COMPOUNDS. 215The phenanthrene-1 : 2-dicarboxylic acid anhydrides (CII, CIII, CIV ;ft = H) were stated 44 to possess some degree of estrogenic activity, but thisclaim was later withdrawn ; 45 (CIV ; R = H) wa,s also found to be inactiveby Cohen and Warren.46 The methoxy-anhydrides (CII, CIII, CIV ;R = OMe) and the hydroxy-anhydride (CIV; R = OH) have now beenprepared and all found to be inactive a t a level of 1 mg. by subcutaneousapplication in the Allen-Doisy test in r a t ~ . ~ 1/v\(CII.) (CIII.) (CIV.)(+)-Dimethylethylallenolic acid (XLIII; R = H, R1 = R2 = Me,R3 = Et), which exhibits intense estrogenic activity and bears a formalresemblance to C-norbisdehydrodoisynolic acid (XCVII), is prepared from2-cyano-6-methoxynaphthalene (CV) which by the Reformatsky reactionwith ethyl a-bromoisobutyrate (1 mol.) gives the keto-ester (CVI), convertedby ethylmagnesium bromide into the carbinol-ester (CVIII), also obtaineddirectly from 2-(6-methoxynaphthyl) ethyl ketone (CVII) by theReformatsky reaction. Dehydration of (CVIII) furnishes the olefin (CIX),which by hydrogenation and subsequent hydrolysis gives the methyl etherof (&)-dimethylethylaIlenoIic acid (XLIII; R = R1 = R2 = Me, R3 = Et),demethylated by hot pyridine hydrochloride to the free hydroxy-acid.By suitable modification of this procedure methyl ethers of homologous(&)-allenolic acids (XLIII ; R1= R2 = Me, R3 = H ; R1= R2 = R3 = Me ;and R1 = H, R2 = Me, R3 = Et) are obtained3'/\ - m d I( 1 I___,>1ed C:Me,Br*('O,Et \,A/WV.) (CVI.) (XLIII.)EtMgBr(CIX.)/\/\- I, I COEt Meo4A//(CVII.) (CVIII.)On account of the intense estrogenic activity of the doisynolic acids,carboxylic acids of the cxp-diethylstilbene series have been prepared.474 4 L. F. Fieser and E. B. Herschberg, J. Arner. Chern. SOC., 1936, 57, 1508, 1867.4 5 Idem, ibid., 1936, 58, 2315.4 7 R,. Neher and K. Miescher, HeLv. Chim. Acta, 1946, 29, 449.46 J., 1937, 1315216 ORGANIC CHEMISTRY.The acid (CX; R1 = R2 = Et), regarded as the truns-form, possesses someactivity [threshold value 10-20 y (subcutaneous), 20-30 y (oral)]. Varioushomologues of (CX) [R1 = R2 = H (cis- and trans-forms), Me, or Pr";R1 = H, R2 = E t ; R1= Et, R2 = HI have been synthesised; thedi-n-propyl acid exhibits slight activity ; the dimethyl acid is inactive, butits methyl ester shows a weak but protracted action. The m-isomeride of(CX ; R1 = R2 = Et) and the triphenylethylene analogue (CXI) arep -HO C6H4*CR1~CR2*C6H ,*C02H ( p ) ( CX . )(p-MeO~C6H,-)2C~CR~C6H4*C02H ( p ) (CXI.)completely inactive. Replacement of a p-hydroxyl group by a carboxylgroup thus leads to aconsiderable decrease in estrogenic potency.c. w. s.W. BAKER.N. CAMPBELL.S. H. HARPER.D. H. HEY.C. W. SHOPPEE.F. SMITH