Cassaine Analogs. II. 7-Deoxy Basic Esters - Journal of Medicinal

Synthesis and biological activity of semisynthetic digitalis analogs. Richard Thomas , John Boutagy , Alex Gelbart. Journal of Pharmaceutical Sciences...
0 downloads 0 Views 1MB Size
CASSAIXEANALOGS.I1

July 1967

583

J, Table 111). Although the mechanism of this isomerization is not proven, it probably involves carbonium ion formation under the influence of acid (Scheme 11).

SCHEMEI

SCHEME I1 OA

4

3,R=P-OH 4,R=O

5,R=@-OH 6,R=O

is not immediately evident since no influencing structures seem to lie near the chromophore. A discussion of this configurational assignment together with that in the C-10 oxo series and the cassaine series is given in the paper devoted to 10-oxophenanthreneacetic acid derivatives. Oxidation of the mixture of 3 and 5 with CrOB in pyridine afforded a mixture of ketones from which the trans isomer (4) could be obtained essentially pure by crystallization. It was then hydrolyzed to give the pure trans acid (7). The mother liquor from separation of the trans ester (4) was rich in cis ester (6)and hydrolysis of this mixture allowed separation of the cis acid (9). Esterification of the mother liquor from 9 with diazomethane furnished more of the cis isomer

Since the present work was directed toward determining the relationship between structure and activity, it was necessary to separate the cis and trans isomers in only a few confirmatory instances because the difference in cardiotonic activity of these isomers was minor. The principal inconvenience in working with such isomer mixtures lay in their broad melting points. An exceptionally useful technique for purifying many of the basic esters reported here was that described by Brown and Kupchan5 which involves partition chromatography of basic materials. A dye which is incorporated in the stationary phase indicates the degree of movement and separation of the bases on the column. This method did not, however, effect separation of our cis and trans isomers. Tables 1-111 summarize most of the alkyl esters (Wittig products), carboxylic acids, and basic esters presently reported and footnotes to these tables describe deviations from the standard procedures given in the Experimental Section. Unless otherwise noted, the reaction sequence involved ketone -+ alkyl ester -+ acid basic ester. The tricyclic ketones used in the preparation of the esters of Table I are described by Daum, et u Z . , ~ with the exception of the ketone precursors of compounds A and 0 of Table I which are described in the Experimental Section. Examination of Table I11 reveals the general scope of the investigation. Compound A has no oxygen function in either rings A or B. Compounds B-F and I have the same side chain at C-2 (phenanthrene numbering) but differ at C-7. Compounds G, H, and J-& fall into this same category except that they carry a C-ib methyl group. Compounds R-Z differ principally in the type of side chain at C-2. The amide Z is the only rionester listed. It should be noted that compound I' contains a A8 bond, compound XA has an additional methyl group at C-1, compound BB has two methyl groups at C-8, compounds CC and D D have rings A and B fused in a cis manner and, finally, that EE has rings B and C fused in a cis manner. Several additional basic esters are described later owing to their special mode of synthesis. Isolation of the primary aminoethyl ester IT of Table I11 was difficult on account of the propensity for such esters t o rearrange to P-hydroxy amide^.^ This

-

7,R=H

9,R=H

8, R = C H ~ C H L N ( C H ~ ) ~ 10, R = CHJ 11, R=CHzCHLN(CHs)z

as its methyl ester (10) and hydrolysis of 10 afforded the purest sample of cis acid 9 obtained in this work. Treatment of the sodium salts of isomers 7 and 9 u ith oxalyl chloride followed by dimethylaminoethanol produced trans (8)and cis (11) basic esters which formed nicely crystalline, water-soluble hydrochloride salts. The reaction of the sodium salts with oxalyl chloride is essentially complete within 5-10 min at room temperature. When a single isomer is at hand, it is necessary at this point to remove excess oxalyl chloride and to add the required amino alcohol as quickly as possible since allowance of the acid chloride to stand at room temperature for an extended period or heating it briefly with steam following solvent removal causes isomerization and production of a cis-trans mixture. The Experimental Section records an instance of complete equilibration of a trans acid (compound J, Table 11) to a 1: 1 mixture of cis and trans basic esters (compound ( 4 ) S J. Daum. hI. hf. Riano, P. E. Slian, and R. Chem 32 1435 (1967).

.

L. Clarke, J.

Org.

S.Drown and S. 11 Kupclian J . Chromatog 9, 71 (1962). (6) S J. Daum, P. E. Shaw. and R L. Clarke, J O r g C h e m , 32, 1 4 2 i (1967) ( 7 ) See G. Fodor. A h C h m Acod h i . H u n g 6, d7Y (1954) a n d espe01a11> p 389. (5) K.

I1 11 I1 II I1 CII, I1 11 I1 II II

CIIa CTI:, C1.1ij CJlj CJI, CII, C113

CZIT5 CiJIS ClI, ClIa

1:l 1:l 1: 1

ti:l I:(i

7:s 1:1 1:1 1: 1 1:l 1:l

irotevted tlirough its carbobenzoxy ter forni:itiori aiid the resulting ester readily c1e:ivcd n ith trifluoruacetic acid i i i '11hr at room temperature. So rearrangeiiient ocruretl uiider these acidic. c*oriditioris. I t proved powihle to 3trenl; thih amiiioethyl ester-trifluoroac~etate s i l t O I L 4lic.n-coated preparative chroniatoplates, develop :md elute quickly with tetrnhydrofuran (THF) (~)iit:iiiiiiigi.;oI)ropyl:iniiiie, a i i d coiivert the purified :iiiiiiioehtcr t o it. stnhle hydrochloride salt without seriou- (leconiposi t 1011. Surpri\iiigly, I t i:, possihk t ( J t ~ ) i i v e r t7-hydroxy uiis:~turntecl:wids iuch as 11, S,0 , and S of Table I1 tlircrtly t o T-hydroxy hasic eitcrs by the oxalyl chloride tecdhiiiquc. hlrnoit cbertaiiily :til oxalyl ester-arid chloride forms at C'-7 but treatment of this intermedi:it e 11 ith diniethylaiiiii~oethaliolin boiling benzene (u.u:il ti*e:it iiieiit in side-chain ester formation) results i n regcrieration of tlic free hydroxyl group at C-7. t h e Wittig reaction producing the esters of 'l':ihl(> I n o r i d l y procwdq r:ipidly with negligible \t:irtiiig material cienioii ihle in t h e crude products 1)y t IC, rcuc*iioii of triiiietliyly,hos~~lii~ii(~acetate with I\(+otic 12 : i I n x p r w d t e d i i i recovery of about 50% of

12

13

unchanged startirig material. This iwovercd material theii underneiit :I further 5OYc (wiis ci-iori uiirkr tlie m n e coiiditioris. IiCxteiidirig t h c rcavtioti tiiiic, clcvatiiig the temperature, or iIicwnsirig the proport H I I I i ) f Wittig reagent preseiit Euilcd t o 1)r(1duc~ further rcaction. It is pokrihlc that ;MI q u a t oi~:il methyl group :djacent t o the caarbonyl group ( 1 : ~ i i h wforinntioil of :in enol phosy1i:ite :is :L (~onywtingprotluct nliirli regeiierates stxrtiiig imiterial during n-ork-up. I'eriodir check of the reaction niixturo o i i silica tlc pl:~tesalways h h o n ed a spot represcntiiig this uiicliaiiged niatcrial. Thc iitmc partial JVittig re:wtioii 11 : i ~ o1)berved wheii :iii equ:%turialiiietliyl g in a precursor (13) t o methyl 14-epic when this methyl group was ;ixi:iI.x the Wittig reacatioii on 13, the pi'etlomili:i1it coniporiciit (707,) of thcl produrt from t h c re:icotjoii n i t h 12 was assigned the flnns-ctluatorial .t rucaturc. 'rhc niiiioi. product C ~ O U ~\I Cell~ h:lvcl hccti other t Ii:iii t h(1 CZPequatorial f ~ r i n . ~ A first attempt to prcynre t 1' 14 kJ)' ii \\-Itti% re:l('tion between triethyl cr-~~hosphoiiopro~)ioii:it e 16 and the requisite tricaycalic ketoiie failed (~oiiipletelywhcn cliinethoxyethane was used as a solveiit, cveii with a reflux period of 3 days. With dimethyl sulfoxide as thc solventJg 14 was isolated i l l 71% yield. Hydrolysis : ~ n d esterificatioii with ~-dimethylaniinoethallol ('on-

555

CASSAINE ANALOGS.I1

July 1967

TABLE I1 TRICYCLIC ~U,~-UNS.ITURATED ACIDS

R

KO

R?

R1

Compd

O=

H

R3

truns: cis ratio

R1

CH3 H H H I-I II

H H II H H H

1; 1 czs trans 1:1 1: 1 1:l

CH3

H

I1

Ultravioletn Mp,

oc

Lmp

%--

--Found,b

c

6

H

-

--1.,3 . 1 13.5

S

169-179C 220-222"e 224-22'7' e 173-189c 205-20iC 194-200'

222 221 222 223

15,800 15,400 16, 200 15,400

235

25,800

...

254-259h

221

16,600

353'

H

1:1

18i-220k

219

16,000

64.S1

8 . 0 19.0

I1

H

...

270-2ilk

221

16,700

92

54.0

6.7

CII, CH3 CH3

I1 H H

... trans cas

194-200" 181-184' 219-221c

221 221 221

16,400 16,200 16,100

23" 66 40

73.9

8.8

73.8

8.7

74.1

8.7

CH3

H

...

C H3

H H

-7

10.1 8.4 8.4 9 1

i1.I .>

72.6

18.0

15.8

Used crude

173-2 12? 222 16,000 -IO 73.0 9 3 223-2238 16,100 221 i 3 . i L '3.6 CH3 H 198-20OU 222 15,300 48 i3.3 9.5 CH3 . . . r s e d crude CH3 CH3 H 1: 1 186- 1YOk 222 16,500 43j i4 8 10.0 CH3 H 1: 1 185-200c H 72.8u 8.4 222 16,000 7 2 H 1: 1 Uaed crude H 1:l 208-232k 229 24, 300 @-OH H 61 Li o= AS 7.1 From rthvl acea JIeasured iii 95yr ethaiiol. All analvtical values are Rithiii 0.3% of those calculated uiiless otherwise noted. . tate. d This preparatioii is described iii detail in the Experimental Section. e 111 an evacuated capillary. j Yield for two steps from tricyclic ketoiie. 0 Prepared from C of Table I. From acetic acid. Neutralization equivalent (calcd 332.3). j Product pwified by chromatography 011 silica gel coated plates with development by 3: 47: 50 acetic acid-pentane-ether. k From acetone. I A4nalytical value 0.4% high. 'n Prepared from J of Table I. Yield for t'hree steps from tricyclic ketoiie through I and then J of Table I. 0 From From acetone-hexane. L of Table I. p From 11 of Table I. * From I of Table I . From N of this table by recrystallizatioii three Prepared from basic ester AI, Table 111,by alkalilie hydrol) times from acetone aiid once from acetoiiit'rile. ill the standard mallller, From ether. Rings A and B are fused iii a cis maliller (4b0, 8ap). Aiialytical value 0.5% low. Itiiigs B arid C are fused in a Analytical results were erratic over several per cent. TIC showed a single spot. CompoLuds gave the cis manner (4aa, 10aa). curresponding basic eater which showed correct analysis. 1:l

t runs cis

~

/CH,

\COOR

0 CH3

T

I

(CzH:0)2P-CH COOCzHe 16

HO 14, R = CzHs 15, R = CHnCHaN(CH.i)z

verted ester 14 to basic ester 15 as a mixture of cis and trans isomers. An analog of the above basic esters wherein ring A was aromatic (18) was prepared in the standard manner by a Wittig reaction on ketone 17 followed by hydroly-

~~

~~~

~

intent to prepare the corresponding dimethylaminoethyl ester. I n the process of treating the sodium salt of acid 22 with oxalyl chloride and dimethylaminoethanol, elimination of the hydroxyl group occurred with aroniatizatiori of ring A and formation of the phenolic basic ester 19. Hydrogenation of unsaturated ester 3 in the presence of Pd-C produced a mixture of compounds 23a which were epimeric at C-3. These were simply designated isomers h and B in the absence of a basis for configurational assignment. Isomer h crystallized spontaneously from the mixture. Hydrolysis of the mother liquor afforded isomer B in the form of its acid 23b. The corresponding pure isomeric. baqic esters A and B, 23c, were then readily prepared through their acid chlorides. Seither the isomers of 23a nor those of 23c

17, R = CH3; R ' = 0 20,R=O 18, R = C H I ;R ' = CHCOOCHZCH,N(CHS)~ R=CHC00CH3 22, R = CHCOOH 19, R = H ; R ' = C H C O O C H ~ C H I N ( C H I ) ,

sis and esterification with dimethylaminoethanol. The related phenolic ester 19 came from an unexpected direction. Ketone 204 mas converted through Wittig product 21 to a C-4b hydroxylated acid 22 with the

~

R = CzH5 b,R=H C, R CH,CH,N(CH,),

23a,

-- --

p

3

c

g

L

Qi5

-..- ... .. . . .:.2 .- +. .

-

e> s N

z

C,

N

N

z

A

--

43-

CAYSAINE ANALOGS.I1

July 1967

587

could be separated by tlc but those of 23b were separable, isomer B being the more polar. Biological Methods The initial test for cardiotonic activhy was performed in the isolated appendage of the rabbit heart. A more advanced evaluation of drug action was made in the dog. Rabbit.--Appendages were suspended in Tyrode solution a t 37" and stimulated by rect,angular pulses (supramaximal voltage, 180/min, 10 m.sec duration) via Pt electrodes. After 75 min the tyrode solution was replaced with fresh solution and 15 min later dr1ig.j were added. The sequence of drug concent,rat,ions (expressed i n terms of the base) was 0.01, 0.05, 0.1, 0.5, 1, 5, and 10 pg/ml wit.h a 10-min interval between each concent,rat,ion change. Cont,racstile force was recorded isomet,rically on a Grass polygraph b v means of a force-displacement transducer. Dog.-Cardiovascular activity in anesthetized (pentobarbital, 30 my/kg iv) spontaneously respiring dogs was recorded on a Grass polygraph. I3lood pressure in the abdominal aorta was measured by means of a Statham pressure transducer via a polyethylene cannula inserted into the femoral artery. Heart cont mct,ile force was measiired isomet.rically by means of a Walt,onBrodie strain gaiige arch sutured to the wall of t,he right ventric.le.10 Heart rate was calciilated from the force recordings and rardiac elect,ric.al activity was monitored via a lead I1 electrocardiogram. Drugs as salt,s were dissolved in distilled water atid administered into t.he femoral vein usually in t,he sequence 0.25, 0 . 5 , 1, 2, and 4 mg (calculated as free base)/kg of body weight,. Dose-response ciirves were fitted to the data b v eye and the dose producing a 20yr increase in cont,ractile force was estimated from t.he regression of tshe curve. Cumulative doses were iised in constract,ing the dose-response curve of ouabain. I n t.he infiision experiments a const'ant-speed infusion pump was used to deliver t.he drrigs to anesthetized animals (pentobarbital, 30 mg/kg iv; a-chloralose, 80 mg/kg iv).

Results Table IV prcsents the testing results in both the isolated appendage of the rabbit heart and in the intact dog. Discussion of cardiotonic activity relative to the structural modification of the cassaine molecule can be conveniently divided into two parts: (1) that dealing with alterations in the phenanthrene nucleus, and ( 2 ) that concerning changes in the side chain at C-2. Changes in the Phenanthrene Nucleus.-The results show that the substituent at C-3 [cassaine (1) numbering] can greatly influence the cardiotonic activity of the mole. The most active compounds contained an ethylenedisulfonyl group (I) and an amidinohydrazone group (H) at this position. Compounds containing 3a- or 3 &hydroxyl groups were about equiactive (E and IC). Furthermore, the substitution of a carbonyl group (D and J) for a hydroxy group (E) did not affect activity. However, replacement of the hydroxyl by a hydrogen atom resulted in a less active compound (A). So significant change in cardiotonic activity was ohserz ed upon addition of two methyl groups at C-4 (BB), an equatorial methyl group at C-14 (AA), or an axial methyl group at C-10 (compare D and J). I t should be noted that cassaine carries an axial methyl group at C-14. The preparation of 24 is described in ref 4. This "ring-demethylated" cassaine seems to be even less active than the cassaine analogs without a ring-B oxygen (see Table IV, 24). (10) T. G. Brown. Jr., and A. AI. Lands in "The Evaluation of Drug .\ctivitu," Yol. 1. D. R. Lawrence and A. L. Bacharach, Ed., Academic Prc-s Inc. New York, N. Y., 1961, Chapter 17.

24

The manner of fusion of the A/B arid B/C rings does not greatly alter cardiotonic activity. Compounds CC and DD, having an A/B cis fusion, are slightly less active than compound EE which has rings B and C fused in a cis manner. Changes in the Side Chain.-The question of relative activity when the side chain is in the cis us. trans configuration was answered by comparing coinpounds B with C and L with 34. I n both instances there is no major difference in activity. A mixture of B and C is represented by D and a mixture of L and 11 (4:G) is present as E(. Hydrogenation of the double bond of the side chain (23) caused a diminution in cardiotonic activity. Decreased activity was also noted with the substitution of a methyl group on the double-bonded C atom of the side chain (15). Replacement of the ester linkage by an amide (2) resulted in a compound which depressed contractile force. Elongation of the alcohol portion of the side chain from dimethylaminoethyl to dimethylaminobutyl (T) had no effect on cardiotonic activity. The effect of the size of the substituents on the aniino nitrogen was significant. Hydrogen atoms (V) could be substituted for methyl groups (J) with no loss of activity. However, the diethyl analog (W) was only weakly active and the isopropyl analog (X) depressed contractile force in the dog ventricle. Toxicity.-Table IV shows that low concentrations of the synthetic cassaine analogs possess positive inotropic activity in the isolated rabbit appendage. As the concentration of drug in the bath was increased, toxic effects were observed, i.e., arrhythmic beating, a negative inotropic action, and cessation of beating. These effects were also observed with ouabain and cassaine (see Table IV) and are typical of the cardiac glycosides. ,4similar sequence of effects was observed after intravenous administration of single doses of the synthetic compounds to the intact anesthetized dog (Table IV). Low doses produced an increase in ventricular contractile force, intermediate doses resulted in cardiac slowing, and high doses evoked toxic effects, ie., A-Ti block, ventricular ectopic beats, ventricular fibrillation, and respiratory arrest. The results of experiments in which L and cassaine w r e infused into the femoral vein of the dog are summarized in Table V. Tremors were observed after 15 min in only 1/3 dogs anesthetized with pentobarbital, whereas clonic-tonic convulsions were observed after 4 and 6 min in 2/3 dogs anesthetized with a-chloralose. The third dog in the latter group exhibited severe tremors after 15 min. It would, therefore, appear that convulsive activity due to the synthetic compound is readily suppressed by prior administration of barbiturates (but not by a-chloralose), a characteristic shared by cassaine." (11) hl. deV. Cotten, L. I. Goldberg, and K . P. Walton, J . Phurmacol. Ezptl. Therap.. 106, 94 (1952).

.jSS

\ < ) I . IO

CASSAIXE AXALOGS. 11

July 1967

R a t e of infusion, rngikgjmin

Compd

L, 2 5 sol

+

Time to give ventricular Time t o ectopic Time t o rate contractile sloving, beats, min force t , min min

Time t o give

conrulsions, min

Lethality Time, min mg 'kg

0.41

5.0

..

45

15

...

...

1.11

1.3

..

30

..

41.1

37

1.96 0.41

1.3 3.0

..

..

11.8

..

15

...

0.90

1.0

..

..

4

34.2

38

1.69

1.5

..

..

6

27.9

17

50

60

..

2.2

234

pentobarbital, 30 mg/kg iv

+

L, 27; sol a-chloralose, 80 mg/kg iv

+

Ca>saine, 0.1c.( sol pentobarbital, 30 mg/kg iv

0.0094

100

tated product was extracted with ether and the e x t r a m were washed wit.h brine and dried (Sa2S04). Removal of the ether gave an oily residue which partially crybtallized upon addit,ion of about 25 ml of ethanol. Dilution of this mixture with 300 ml of H20 aiid filtrat'ion afforded 10.4 g ( 1 0 0 9 , ) of a crystalline product., mp 89-10l0, which was shown by glpc to be a 1 : l mixtiire of cis aiid trans isomers together wit,h 1.6% of an impurity. These isomer;; were not separable on silica chromatoplates developed wit'h pure ether but t,his chromat.ographic process way w e d to remove the impurity. The isomer mixture allized from ether-hexaiie to give compound B (Table I ) . Ethyl dl-trans-3,4,4acu,4bp,5,6,7,8,8acu,9,lO,lOa~-Dodecahydrohreneacetate (4j.-A solution of 25.4 g .1 dl-3,4,4ao(,4bp,5,6,7,8,8acu,9,10,10ap-do-A2(1H),a-pheiiaiithreneacetate (B, Table I ) iii 220 ml of pyridiiie was added i n 2 min with stirring to a mixture of 21.9 g (0.2%mole) of CrOa aiid 220 ml of pyridine a t room temperatiire aiid the resiilt,iiig mixture mas &red overnight. Ethyl acet,ate (1.5 1.) was added, t.he mixt,nre was liltmeredand t.he filtrate was conceiit8rated t,o a residue by warming iinder reduced pressure. This residue was treated with 400 ml of ether aiid a furt,her insoluble mat,erial was removed by filtrat.ion. Coiiceiit,rat'ioii of the et,her solut'ion and addition of hexane aftorded 6.13 g of c r y s t a l h e solid, mp 85-90'. Recryst,allizat.ioli from elmher,with hexane added, gave 5.7 g of the trans isomer, mp 94-96". Reworking all of t,he mot8her liquors furnished aiiother 0.75 g of product,, mp 94-97' (total yield 26%). The analyt.ic.al sample, obt,aiiied from a similar experiment, is D (Table I ) . tll-cis-3,4,4a~,4bp,5,6,7,8,8aa,9,10,10a~-Dodecahydro-7-oxoA2(1H),a-phenanthreneaeetic Acid (9).-The mother liquor residues from the precediiig experiment were dissolved in 500 ml of (357, ethaiiol, 200 ml of 2 S aqueous YaOH was added, and the solutioii U-asrefluxed in a nitrogen atmosphere for 1.25 hr. The reactioii mixture was added to ice-water and neutralized with acetic acid, a i d the product was extracted with ether. The ether est,racts were extracted with 2 -YNaOH and t,hese extracts were aridified with 2 i Y HCl. The precipitated carboxylic acid was collect'ed and recrystallized from ethyl acetate to give 4.36 g of the cis acid, mp 206-214', " : :A 221 nip ( E 15,iOi)), and a second crop of 0.28 g, mp 202-213" (18yc). The aualytical sample of this acid (Table 11, B ) was obtained by hydrolysis of the piire cis methyl eder (10) using the general procedure for hydrolysis of alkyl est,ers. Methyl dI-cis-3,4,4aa,4bp,~,6,7,8,8aa,9,lO,lOa~-Dodecahydro7-0xo-A~[l~).~-phenanthreneacetate (lo).-The mother liquor residue:, from separation of the cis acid described immediately above contained 8.22 g (0.0313 mole) of the cis and trans unsaturated carboxylic acids. This solid, mp 180-195", was dissolved iii 250 ml of methanol, 0.10 mole of diazomethane iii ether was added, aiid the wlutioii was allowed to staiid overnight.

6 .

.

I

Remarks

Tremors, -1-V block, terminated at 170 miii after 69.5 mg/kg Ventricular fibril, resp arrest a t 33 min Resp arrest a t 2 miii Severe tremors, t,erminated a t 60 miii after 24.5 mg/kg Cloiiic-tonic convulsioiis, resp arrest, ventriciilar fibril Tremors, clonic-tonic convulsions, ventricular fibril .4-1-block, ventricular ectopic tachycardia, cessation of resp, ventricular fibril

The solvent, was removed and t,he cr .alline residue was recrystallized from ether by the addition of hexane t o give 2.87 g of material which melt,ed a t 100-130". Two furt.her recrystallizat,ions furnished 1.6 g of t,he cis met,hyl est.er 10 (Table I, E ) which was shown by glpc to be a single compotind. General Procedure for Hydrolysis of Alkyl Esters. dl-trans3,4,4a0,4bp, 5,6,7,8, Sam, 9,10,10ap-Dodecahydro-7-oxo-A2(1H)~aphenanthreneacetic Acid (7).--9 solution of 6.0 g (0.021 mole) of the ethyl ester (Table I, D ) of the title compound in 200 ml of 95% et'hanol was treated with 80 nil (0.16 mole) of 2 .Y aqueous Sa013 and the solrit,ioii was refluxed for 1.25 hr under S p . The react,ioii mixt,ure was cooled, acidified with acetic acid, and coiiceritrat,ed under rediiced pressure iuitil the ethanol was removed. The prodact, was ext,racted from the resulting mixtiire with et,her aiid then extracted from tmheet,her with 2 S aqueous PiaOH. Acidificatioii of this ext,ract wit.h coiiceiitrat,ed IlCl precipitated t,he product, which was collected and recryst,allized from ethyl acetate to give 4.2 g of the trans acid C (Table 11). General Procedure for Making Basic Esters. 2-Dimethylaminoethyl dl-trans-3,4,4aa,4b~,5,6,7,8,8aLu,9,lO,lOa~-Dodecahydr0-7-oxo-A~(l~)~~-phenanthreneaeetate (8).--A solution of 4.47 g (0.017 mole) of the trans acid 7 in 100 ml of TIIF was treated with 0.92 g (0.17 mole) of sodium methoxide and 1 ml of 1320. The solvent was then removed by aarmiiig under reduced pressure, 20 ml of absolute ethanol was added aiid evaporated i n the same manner and, finally, two 20-ml portions of dry benzene were added and evaporat.ed. The resriltiiig dry sodium salt, was suspended in 150 ml of dry benzene, 3.46 g (0.044 mole) of pyridine was added, the mixt,ure was immersed iii an ice bat,h, and 40 ml of oxalyl chloride was added iii a fast stream of dropr with btirriiig. The mixtiire was removed from t,he ice bat.h, * k e d for 10 min, and then conceiit,rated as rapidly as possible under rediiced pre,>siire iising a waber bat.h at 4 5 O . ilpplication of heat, was stopped as the last of the solvent evaporat,ed aiid 150 ml of beiizeiie was added followed by 40 ml of 2-dimethylamiiioethaiiol in a rapid rtream of drops wit,h stirriiig and cooliiig. Wheii additioii was complete, the mixtiire w a s heated on the steam bath for 5 miii, cooled, and dilrit,ed with 1 1. of ether and 600 ml of saturat,ed aqueous XarCOs. The layers were separated and the water layer was wahhed with et.her aiid discarded. The combined et.her layera were extracted wit,h t,wo 100-ml port,ions and one 50-ml portion of 2 HC1 and the combined ext,racts were made basic wit.h S a O H solution. This alkaline mixture was extract,ed wit.h ether and the ext,racts were washed with brine and dried (KaaS04). Removal of the ether afforded 4.9 g of a yellow oil which wah (by glpcj a trans-cis (96:4) mixture of isomers t,oget,her u-ith 12% of impurity. The product, was piirified by partit,ioii chromatography as described by Brown and Kupchan.6 The solvent system employed was a 12: 1: 2: 0.2 mixt,ure of hexarie-ethyleiie dichloridemethariol-I-TdJ. Supercel (300 g ) was wet,ted nit,h 225 ml of t.he polar phase containing 75 rng of brom cresol purple, t,he color

CASSAINE ANALOGS.I1

July 1967

59 1

NaOH. It was extracted twice with ether and the extract5 were washed twice with brine, dried, and concentrated to give 1.78 g and 8 ml (96 mmoles) of pyrrolidine was heated under reflux for 4.5 hr with a water separator attached to the system. This of a viscous, amber oil. The oily acetate was converted to 1.75 g of its hvdrochloride salt, mp 175-180'. Two recrystallizations solution was concentrated t,o a residue by warming under reduced pressure and the residue was treated with 50 ml of dry afforded N,Table 111. benzene and 3.5 ml (93 mmoles) of formic acid. The mixture 2-Dimethvlaminoethvl dl-3.4.4a~1.4b,5,6,7,S,Sa~t,S,lO,lOapwas heated under reflux for 30 min, cooled, and treated with Dodeeahydr0-7~-hydrox~-4bp-methyl-A~(~~)~~-phenanthrene1.5 ml of formic acid. Wat,er (60 ml) and ether (100 ml) were acetate 7-Benzoate (Table 111, O).-A solution of 0.90 g of the added and the layers were separated. The &her layer was base from K, Table 111, in 50 ml of dry benzene was treated R-ith exlravted once with 2 S HCl and discarded. Addition of the 0.5 ml of pyridine and 2.0 ml of benzoyl chloride and heated on acidic extract, to t,he aqueous portion of the reaction mixture the st,eam bath for 5 min. The solver~t,~ were removed by warmcaused precipitat,ion of the hydrochloride salt of the product,. ing under reduced pressiire and the residue was partit.ioned between 2 N aqiieoiis NaOH and CH2C12. The organic layer was Concent,rated HC1 (3 ml) and 10 ml of brine were added and the precipit,ate was collwted. It was washed well witah acet'onit,rile separated, washed with brine, dried (Na2S04),and concentrat,ed t,o give 1.3 g of a cryst,alline residue. This base was dissolved in aiid then ether to give 2.57 g of crude RI, Table 11, which was sriitable for conversion to a basic ester. hot acet,onit,rile,0.3 ml of 8 ,V alcoholic HC1 was added, and the 2-Dimethylaminoethyl dZ-3,4,4a~,4b,5,6,7,S,Saa,S,lO,lOa~- mixture was cooled t,o give t,he hydrochloride salt of the desired dodecahydro -4bp-methyl-7-oxo- AZ(lHJsa-phenanthreneaeetate product. One recrystallizat#ion from acet,onit,rile furnished 0 Amidinohydrazone (Table 111, H).-A solution of 5.60 g (0.0161 of Tahle 111. mole) of 2-dimethylaminoethyl dl-3,4,4aa,4h,5,6,7,8,8a~t,9,10,- 2-Dimethylaminoethyl ~Z-3,4,4aa,4b,5,6,7,8,8aa,S,10,10aplOap - dodecahydro-4bp-methyl-7- oxo - A2(lH)sa -phenanthreneDodecahydro-7p-hydroxy-4bp-methyl-~2([email protected] (Table 111, D ) and 1.2 ml of concentrated HCI in 25 ml acetate 7-Nitrate (Table 111, P).-Nitric acid (90%, 8 ml) was of met,hanol was added to 175 ml of 1 S methanolic HCl in which added slowlp with stirring at -10 t,o 0" to 50 ml of acetic anhad bee11 dissolved 6.0 g (0.044 mole) of aminoguanidine biohydride. Then a solution of 4.00 g (0.0115 mole) of the free c4arbonat.e. The solution was allowed to stand at room temperabase of K, Table 111, in 15 ml of CHC1, was added dropwise with tiire for 41 hr and was then treated with solid NaHC03 until st,irring at - 5 to - 10" in 15 min. This solution was kept, cold neiitral. The solvents were removed by warming under refor 1.5 hr and then poured into 400 ml of ice and wat.er. The diiced pressure and the residue was dissolved in 1 :4 acetic acidmixture was allowed t,o stand for 1 hr, made alkaline with concenwater. A small insoluble residue was removed by filtration and trat,ed SHdOH, and ext,racted twice with ether. The ext,racts the filtrate was cooled in an ice bath while being made strongly were washed wit.h brine and concentrated to a residue by warming alkaline with 35y0 aqueous S a O H . The precipitated product under reduced pressure. was collect,ed and dried by addition and evaporation of several The oily residue was chromatographed on silica gel coated portions of ethyl acetate. platmeswhich were developed with 1: 1:98 methanol-isopropylThe 6.8-g residue was dissolved in 110 ml of methanol and t,he amine-CHCla. The loading amounted tro about 0.4 g/20 X 40 solut,iori was treated with 2.8 ml of concentrated HC1 and 500 ml cm plate carrying a 1-mm coating of silica gel. The principal of ether. This mixture st.ood for 64 hr and was then filteied to hand from the plat,es afforded an oil whose infrared spectrum give 5.45 g of desired amidinohydrazone hydrochloride, mp showed no hydroxyl absorption. The oil was deeolvated under 180-200" dec. It was recrystallized by dissolving it in methanol reduced pressure at, 54", dissolved in 10 ml of ether, and treated and adding ether t o give perhaps a different polymorph bewith 2.0 ml of 6 N alcoholic HCI. The precipit.ated solid was caiise t,his material (4.07 g ) melt'ed a t 167" dec (Table 111, H). boiled wit,h 15 ml of acetone and the mixture was cooled and 2-Dimethylaminoethyl dl-3,4,4a~,4b,5,6,7,S,Sa~,S,lO,lOap- filtered. The crystalline salt, was t.hen recrystallized by dilut.ing Dodecahydro 7p hydroxy -4bp methyl A2(1H).a-phenanthrenea soliition of it in 15 ml of warm methanol with ether to t'he point acetate (Table 111, K).-A 3.65-g sample (10.5 mmoles) of of cloudiness. Colorless blades separated. This mixtiire was amorphous J, Table 111, was dissolved in 100 ml of methanol diluted with 100 ml of ether and filtered to give P, Table 111. and stirred while 0.53 g of NaBH4 was added in small port,ions. 2-(Carbobenzoxyamino)ethyl dl-3,4,4a~~,4b,5,6,7,S,Saa,S,lO,T h e solutio11 was allowed t,o stand overnight at room temperature, lOap-Dodecahydro-4bp-methyl-7.oxo-~ 2 ( 1.=-phenanthreneace~ J acidified wit,h 2 AV &Sod, and concent,rated by warming under tate (Table 111, V).-A solution of 4.84 g (0.0175 mole) of J, Table reduced pressure until the methanol was removed. The aqueous 11, in 30 ml of DLTSO was treated with 0.99 g (0.018 mole) of residue was diluted with 250 ml of HzO, washed with ether, and sodium methoxide followed by 4.74 g (0.0183 mole) of benzyl made alkaline with 2 S aqueous NaOH. T h e precipitated prod2-bromoethylcarbamate.~~This mixture was allowed to stand i i c t was ext,racted wit.h &her and the extracts were washed for 1.25 hr at room temperature, heated a t 100" for 4.25 hr, (H20, brine) and dried (AIgSO,). Concentration afforded 3.00 g cooled, diluted with 150 ml of H20,and ext,racted tn-ice with et.her. of colorles.;, oily tkle compound which was converted t.0 its Concentration of t,he extracts gave an oil (8 g) which was chrohydrochloride salt, (Table 111, K). matographed on 22 20 x 40 cm silica chromatoplat.es developed Separation of cis and trans Isomers (Table 111, L and M ) of with 1 :4 ethyl acetat,e-CHC13. The oily product, thus purified, 2 Dimethylaminoethyl dl 3,4,4aa,4b,5,6,7,8,8a(u,9,10,10ap -Doappeared by tlc to he bett,er than 98% pure. It showed"::A: deeahydro-7p-hydroxy-4bp-methyl-Az(~H~ ,a-phenanthreneacetate 213, 217, and 223 mp ( e 18,700, 18,600, and 17,000, respectively); (Table 111, K ) by Fractional Crystallization of the Methanesulfo: :A 2.99 (ms) (NH),5.85 (vs) and broad (C=O and ester), nate Salts.-The amorphous base obtained from K of Table 111 6.09 (s) ( e m C=C), and 6.55 p (s) ( K H deformat'ion); nmr (2.3 g, 6.6 mmoles) dissolved in 30 ml of ethyl acetate was treated 5.18 peaks (CDCl3) at, 7.28 (5 aromatic H), 5.5 (=CH-), with 0.53 g (5.5 mmoles and 17% less than theory through error) (N-H), 5.07 (benzvl CH,), 3.42-4.13 (OCH,), and 0.90 ppm of methanesulfonic acid dissolved in 20 ml of ethyl acetate. T h e (+ CCH,). The 6.13 g of oil (77%) was cleaved with trifluoroaceprecipitate which formed was collect'ed and recrystallized three t,ic acid as described below without, further characterizat'iori. times from acetonitrile t.0 give 0.74 g (2570) of solid, mp 232-236" 2-Aminoethyldl-3,4,4aa,4b,5,6,7,S,Sa~,S,lO,lOap-Dodecahydr0dec. One furbher recrystallization afforded t,he trans isomer 4bp-methyl-7-oxo-AZ(lH) la-phenanthreneacetate (Table 111, w). (Table 111, L). -A solution of 3.59 g (0.0079 mole) of the carbobetizoxyamino Concent,ration of each of the three mother liquors separately in ester described in the preceding experiment in 20 ml of commerthe above separat,ion of the trans isomer of mp 232-236' afforded cial trifluoroacetic acid was allowed to stand for 24 hr. T h e the more soluble cis isomer in crops of 0.20 g, mp 213-218"; solution was diluted with 30 ml of pent,ane, the mixture was 0.25 g, mp 218-221'; 0.20 g, mp 219-223"; total yield 0.65 g stirred thoroughly, and the supernatant liquid was decanted from (22%). A single recrystallization from acetonitrile furnished the an oily layer. This process was repeated four times. Then 5 pure cis isomer (Table 111, M). ml of et,her was added, the mixture was stirred, 50 ml of pent,arie 2-Dimethylaminoethyl di-trans-3,4,4aa,4b,5,6,7,8,8a~,S,lO,was added, and the supernatant liquid was decanted. This 1Oap Dodecahydro-7p-hydroxy-4bp-methyl-Aa(1H),aphenanprocess was repeated twice. The oily product was then diluted threneacetate 7-Acetate (Table 111, N).-Compound L, Table I11 with a few milliliters of acetone and st.reaked on ten 20 X 40 cm (2.00 g), was treated with 25 ml of H,O and 3 ml of 2 iV NaOH silica plat,es which were developed wit,h 3 : 3: 94 met,hanol-iso*elution and the liberated base was separated with ether. The propylamine-CHCl,. The principal hands were quickly scraped et'her was evaporat'ed and t,he residual oil was treat,ed with 10 ml off and eluted with 1:19 isopropylamine-THF. T h e eluate of pyridine and 5 ml of acetic anhydride for 16 hr. This mixture \\-:&a diluted with 125 ml of H 2 0 and made strongly basic with 2 S (13) E. Katchalski and I). B. Ishai, J . Org. Chem., 15, 1067 (1950).

A mixture of 3.10 g (11.2 mmoles) of J, Table 11, 75 ml of benzene,

~

-

-

-

-

-

-

-

--. ~~

~~

CASSAINE AXALOGS.I V

July 1967

593

dI-1,2,3,4,4a~,4bp,5,6,7,8,8aa,9,lO,lOa~-Tetradecahydro-7~fied by partitioil chromatography oii 300 g of Supercel as dehydroxy-2$-phenanthreneacetic Acid (23b) (Isomer B).-The scribed iii the general procedure. T h e major baiid was eluted mother liquor residues from the immediately preceding experiand the recovered oil was converted to 3.2 g (647,) of the hydroment Tvere hydrolyzed accordiiig to the general procedure to Recrystallizachloride salt of the title compouiid, mp 24i-251'. give the tit81e compouiid. Two recrystallizat,ioiis from ethyl acet#ateand oiie from acetone afforded 3.13 g (295,, 2 st,eps) of 23b, mp 164-166". Furt.her recryst,allizat,ion from acet,oiie afforded the aiialytical sample, mp 170-176". Tlc arialybis, done as with isomer A , iiidicated one conipuuiid was piebelit,, h ) 0.50, ~ d n d . Calcd for C 1 6 I T 2 6 0 3 : C, 7 2 . 2 2 ; I%, 9.83. Fouiid: C, 7 2 . 3 ; H , 10.1.

tion from methanol with ether added afforded the arialytical sample, mp 255-257'. The free base, liberated from t,his salt, could iiot, be distiiiguiuhed from isomer B base by glpc. Ana(. Calcd for C?oH,jN03.1-IC1: C, 64.24; H, 9.71; C1, 0.4s. F ~ i i d C,64.1; : H, 9.9; Cl,9.5. Dimethylaminoethyl dl-l,2,3,4,4aa,4bp,5,6,7,8,8aa,9,10,10~p-

cedure ittid the product was re tallized oiice from ethyl acetate to give 3.95 g ((J7(;>i) of the ti cumpomid, nip 214-216". Tlc aiialysis oii a silica gel plate uaiiig acetic acid-CIICls (3:97) for developmeiit, indicated t.hat. oiie compound was present, K f 0.56. Further recryst,allizatioii gave nip 214.5-21 Anal. Calcd f(ir c16112603: C, 72.2%; TI, C, 7 2 . 5 ; IT, 9.8.

raked thib iiieltiiig poiiit to 202-213' aiid it was uiichaiiged upoii further recrys t allizitt ioii. A n d . C d c d for C~OH~jN08.HCl:C, 64.24; H, 9.71; C1, 0.48. Foiuid: C, 64.4; 11, (3.8; C1, 9.4.

tetradecahydr0-7p-hydroxy-Z~-phenanthreneacetate(23c) (isodl-1,2,3,4,4a~,4b~,5,6,7,8,8a~,9,lQ,lOap-Tetradecahydro-7p-mer B) wits prepared from 2.2 g of 23b,isomer 13, iii the staiidard hydroxy-2E-phenanthreneacetic Acid (23b) (Isomer A).-The maiiiier to give 2.53 g (81c'li)of the title compouiid as its hydroester 23a (4.5 g ) was hydrolyzed accordiiig to the general prochloride salt, nip 196-%06°. Oiie recrystallizatioii from acetoiie

Acknowledgments.-Appreciation is expressed t o i\Irs. G. A. Snyder and Mrs. J . T. Dunn for technical Dimethylaminoethyl dI-l,2,3,4,4aa,4bp,5,6,7,8,8aa,9,10,lOa/3- assistance and to the Physical and Analyt~icalSect'ions Tetradecahydro-7p-hydroxy-2$-phenanthreneacetate(23c, Isoof t8heSterling-Winthrop Research Institute for spect'ral mer A).-This hasic ester was prepared from 3.5 g of 23b, isomer. arid analytical determinations. A, iii the stitiidard maiiiier to give 4.3 g of basic oil which was puri-

Cassaine Analogs. IV. Distant Analogs of Cassaine ROBERT L.

J. DAUM, PHILIP E. SHAW', THEODORE G . BROWX, JR., G. E. GROBLETVSKI, 4 S D W. T'. O'CONNOR

C L B R K E , SOL

Sterlzng-TVznthrop Research Znstztute, Rensselaei, S e w Y o r k

12144

Recezvecl December 1, 1966 ReGasecl Jlanuscrzpt Recezved J f a r c h 1, 1M? Basic esters have been made which bear a distaiit resemblance t o t,he Erythrophleuin alkaloid ca3saiiie. These monocyclic aiid bicyclic analogs were needed iii order to define the role of the skeleton and various substituents of cassaiiie in its cardiac actioii.

The Erythrophleum alkaloid cassaine (1) is reported to be quite similar to digitalis in its action as a cardiac stimulant.' Both of these drugs suffer from the disH

,COOCHzCHzN(CH3)z

2, R = O

3,

1

advantage of producing toxic symptoms in doses only slightly higher than those producing therapeutic effects. It was of particular interest to determine the role of the skeletal structure and the various substituents of cassaine in the cardiotonic activity and toxicity demonstrated by this alkaloid. I n papers 112and 1113of this series we have described a large number of basic esters which bear a rather close resemblance to cassaine. Presently we report some more distant analogs of this compound. The bicyclic analog 5 was prepared from the trans ketone 2. A Wittig reaction using trimethyl phos-

s.

(1) See F Eriaiec and .'Ldamilen-ski, a n d IJ-. V. O'Connor, J . M e d . Chem., 10, 582 (196i).

s.

(3) J. n a u m . AI. AI. ~ Chem. 32. 1485 ( 1 Y 6 i ) .

i p. E,~ Shaw,~ and ~R , L, , Clarke,

J,

Org,

R = CHCOOCH3

phonoacetate transformed 2 into the a,p-unsaturated ester 3 which was a roughly 1 : l mixture of cis and tmns isomers (about the double bond). Yo effort was made in the presently reported work to separate these isomers since it was found in the series of closer analogs2 that there were only slight differences in the cardiotonic activity of such cis arid trans isomers. Hydrolysis of ester 3 was accomplished with KaOH in aqueous ethanol to give carboxylic acid 4. Basic ester 5 was then formed by the reaction of 2-dimethylaminoethanol o n the acid chloride of 4. This acid chloride was best prepared by treating the sodium salt of 4 with excess oxalyl chloride in the presence of pyridine. Any intermediate function formed at C-6 from attack there by oxalyl chloride was decomposed by the treatment with 2-diniethylaminoethanol; the 6-hydroxy basic ester was isolated.2 Incidentally, the &acetate ester of 5 \vas also prepared. This same sequence of reactions vas used in the preparation of all of the basic esters reported here.