Sovember 1969
DIURETIC AGEXTS
977
4 : l ) followed by crystallization (H20) yielded 100 mg of 12, nip apparently accompanied by loss of HSS. hiass spectrographic 178-182". Ir, mass spectra, and nmr data indicated identity examination of the crystals showed the expected >I+. This with 12 pi,epared by me1 hod A. product was wed in the nest st,ep where separation of a pnre Thioamidolincomycin Tetraacetate Hydrochloride (17 'Ha).prodrict, proved less vompIic*a1.ctl. 7-Deoxy-7(R)-thiolincomycin(10). --Jlet hyl 7-tleosj-7 ( R ) - Jincomycin tetraacetate (19.5 g), prepared by acylation of lincomycin wit,h Ac$3-C,HiN, was treated with 7 g nf PSIOin aqueous I hinlincosaminide (9) (.5.2 g ) w a l ~lrmted with 4.14 g of trons-4-ndioxane as previously described. Conversion of t,he crude prodI)ropyl-l.-proliIir::~1 o give ti.26 g of crude prodiict. Chroniatoguct tCJ the hydrochloride yielded 24.6 g of 17.IICl which gave 9.0 raphy (CIICI4-lIe0ll,4: 1) gave 5.M g of 10, [ a ] uf123' (HzO), as an amorphoLis solid. The hydrochloride salt also wah not g of 17.IIC1, mp 207-214", from AcCH,-H?O. Recrystallization gave 17.HCI, mp 221-225" (hygroscopic). And. (C2BH43crystalline. Anal. (C,8H8,SZO&) C, 11, iY. Methyl 7-Deoxy-6-deamino-6,7-aziridinothiolincosaminide ClNPOgSP) C, H, N, S, C1. Thioamidolincomycin (IS).-Five grams of 17.HCl was treated (13).--hmiIlo sugar 4 (1 g ) was treated with &CO3 in boiling with 0.5 S 50% aqueous methanolic S a O H a t 26' for 1 hr. The I)MF. Crystallization of the residne from NeOH after distillacrude product (3.5 g) was obtained by extraction with MeCL. tion of the DMF afforded 370 mgof 13, mp 192-198', and 160 mg, Crystallization (aqueous MeOH) afforded 300 mg of 18, mp 221mp 182-189'. Two recrystallizations (1IeOH-AcCH3) gave 226", and b second crop of 530 mg of 18, mp 195-205'. The a h , mp 20:1-220D der, [ L Y ] D f320' (1)lIPo). Only one analytical s,imple was further purified by chromatography (Etspot was noted on tlc and the melting point was unchanged for OA4c-Ac31e-HP0, 8:5:1). Anal. (ClsHz4N20&) C , H, N, S. further recryitallizationr. ;Lnal. (Cc,H,iS04P)C, H, iY. Thioamidoclindamycin Triacetate (19).-1n t,he manner deMethyl 7-Deoxy-7(S)-thiolincosaminide (14).--A suspension cribed above, clindamycin was converted to the triacetate and of 1.5 g of aziridiue 13 i n 30 ml of i-PrOH saturated with H2S a t treat.ed with P&O. After twice chromatographing (CsHiP-Ac0" was heated in a glass bomb on a steam bath for 5 hr. FiltraMe, 2 : l ) 120 mg of 19, mp 178-181", was obtained from 4 g of tion of the cooled reaction mistmureyielded 1.6 g of 14, mp 190t'riacetat'e. A n a l . (C2rH39C1K~0& C, H, N, S. 197", which when recrystallized (Et'OH) melt'ed at 193-198". 7-Deoxy-7(S)-chlorothiamidolincomycin (ZO).-Triacetate 19 .tna/. (C,H,,SO,S,) C, H, X, S. (12.9 mg) was treated wit,h dilute alkali in aqueous AcMe for 25 The pentaaretate (14a), mp 266-268", was prepared by acylamin. Only one spot was noted on tlc moving about \There extion of 14 with AryO-CjH:N in the Iisiial manner. Anal. ((319pected. The solution was acidified and lyophilized. PnrifiraII?,SO,S,) c, H, N, s. tion was not attempted. 7-Deoxy-7(S)-thiolincomycinHydrochloride (15).-In the Solvolysis of 19.--9 solution of 10 mg of 19 in 5070 aqueous manner previously described,Zb 6.8 g of 14 was condensed with 5.2 DMF was heated a t reflux for 17 hr. A more polar product g of trans-4-n-propyl-L-proline to yield 10.2 g of crude product. which reacted very rapidly with Lemieux reagent (characteristic Chromat,ography (CHCI3-MeOH, 4 : l), followed by conversion of thiols) was noted on tlc. The p H --as adjusted t o 10.5 and after t o t,he hydrochloride, afforded 4.35 g of amorphous solid, [.ID 1 hr evaporated in vacuo. The major spot on tlc (CsH12-AcMe, -161" (HyO). d n a l . (C,8H3sC1NaO&) C, H, N. Iiechro2 : l ) (CHC13-MeOH, 1 0 : l ) moved with and was not separated mat.ography of the free ba,qe gave an analyt'ical sample. Anal. from 10 on &in. tlc plates. For visualization Lemieux reagent, (C,,H~,NPO&)C, H, S, S. 1 2 , and bioautograph us. S. lutea were employed. Methyl 6-Deamino-7-deoxythiolincosaminide(12). Method A,-Aziridine 13 (235 mg) was treated with HON07 followed by Acknowledgment.-The authors are indebted t,o Dr. catalyt,ic hydrogenation over 10% Pd-C in MeOH t,o give 60 mg D. J. Mason for in vitro antibacterial t'esting, t'o C. Lewis of 12 (,lcCH1), mp 189-190". Anal. (CgH,,O$) C, H, S. for in vivo assays, and to R . J. Reid for technical Method B.--l)isiilfide 11 (1.4 g ) was heated at reflux for 18 hr assist'ance. n-ith hydrazine hydrate. Chromat,ographjr (CHCI3-MeOH,
The Relationship between Structure, Stereochemistry, and Diuretic Activity in the 2-Amino-a-phenylcyclohexanemethanol Series, a New Class of Diuretic Agents. 111
L. L. SKALETZKY, B. E. GRAHAM, AND J. SZJIUSZKOVICZ Research Laboratories of The Upjohn Company, Kalanaazoo, Michigan
@OOl
Received M a y 29, 1969
A series of lJ3-amino alcohols and lJ3-amino ethers were made for diuretic t'esting. The four diastereoisomers of both the alcohols and methyl et,hers (I, R = H and CH,, respectively) were synthesized and the relationship of stereochemistry to diuretic activity is discussed. 1- [2-(p,aY-Dimethoxybenzyl)cgclohexyl]piperidine (2) was resolved and t,he d enantiomorph (3) was chosen for further pharmacologic evaluation.
I n this paper, a new class of diuretic agents2 illustrated by structure I is reported. Compounds of structure type I have three asymmetric centers and, in the case of both the alcohols and methyl ethers (I, R = H and CH3, respectively), the four possible diastereoisomers or racemates have been synthesized. Stereochemistry is a prime factor in the correlation of structure with biological activity and in the study of
receptor surface^.^ We therefore undertook a detailed study of the relationship between stereochemistry and diuretic activity in this class of compounds.
(1) Paper I : J. Szmuszkovicz and L. L. Skaletzky, J . Org. Chem., S I , 3300 (1967). (2) For a review of diuretic agents: (a) H . J . Hess in "Annual Reports in
I
Medicinal Chemistry, 1967," C. K. Cain, Ed., Academic Press Inc.. New York, K. Y.. 1968, p 62,and previous Annual Reports; (b) G. de Stevens, "Diuretica," Academic Press, Inc., New York, N. Y., 1863.
(3) (a) E . J. Ariens, M o l . Pharmacal., 1, 232 (1964); (b) P. 9. Portoghese, A. A. Mikhail, and H. J. Kupferberg, J . M e d . Chem., 11, 219 (1968), and references noted therein.
The diastereoihonieric itlcohols atid inethyl ethers of structure I (R = H aiid CHJ, respectively) i n the cirerythro and trans-erythro4 series were active diuretic agents in rats while the cis- arid trans-threo4 isomerh were less active :it comparable doses (see Tables I and 11). 1he czs-erythro methyl ether (2) was resolved and the cl enantiomorph (3) was chosen for further pharmacologic evaluation. The conformations of the four diastereoisomeric alcohols (11) have been unequivocally established.' -' Each of these isomeric alcohols exist in solution a.i n unique conformation due to two iniportant f:tctors: i:~) hydrogen bonding between the hydroxyl nntl piperidine nitrogen, and (b) relative "effective" size of the benzylic side chain and piperidine group on the cyclohexane (chair) ring.' I n the cis alcohols, the conformational studies have established that rzs-threo alcohol (IIA) has the piperidine group axial and the benzylic side chain equatorial on thc cyclohexane ring nhilc the cis-erythro isomer (IIB) has the piperidine group equatorial and the side chain axial. The difference in diuretic activity of the two ris dcohols ha.; ( J i t hcw co ri formatio na 1 been ratiocalized on the h differences (see Discussion). Chemistry.--The synt hesiy and elucidation of stereochemistry of the four racemic anlino alcohols I1 and methyl ethers I11 have been described.' The synthesis of the two cis methyl etherq IIIA6and IIIB is outlined in Chart I arid the two trans methyl ethers IIIC and IIID were made in :I similar mariner. The two cis racemates (or trans racemates) have the same relative ( 4 ) T h e terms erythro and thrro refer t o the conhguration a t t h e Lienavli< side-chain carbon For the definitions of erythro and threo, see ref I c i s and trans refer t o t h e stereochemiatrq of groups on t h e cgclohexane ring ( 5 ) Recent nmr conformation studies on cas-threo alcohol (I14) deuterated in t h e cbrlohexane ring has given t h e folloning results T h e free liase of I1 4 exists in yolution as a single conformation n i t h t h e piperidine group axial and ben731ic side chain equatorial on the cyclohexane (chair) r ~ n g . I n ~ c i n t r a ~tlie t , ti) drochloride of I I A in water has t h e piperidme equatorial and tiit, hrnz) lie side chain axial T h i q striking difference in conformation betrreen the free hase a n d t h e hldrochloridr is primarily d u e t o t h e importance of hldrogen bonding in t i i t s former T h e ~ econformailon studieLi rill be published elseir here (6) Roman numerals are used t o represent t h e general qtructure and letter.\, U , C, a n d D t o represent t h e sprrlfic racemate. Arahir numeral- r p f r r t o coinnoiinds listed in t h e tables.
y ;it tlic cy(*lohcs:i~~~ ring ,junctio~i*I ) i i t differ in tht. c~onfigui.ntioii:it the side-chain htarizylict wrbon. .\Iethyl ether lIIH was resolvtd; the four raceniic tnethyl ethers IIIA, IIIH, IIIC, and 1111) arid optical isomers of IIIB are listed in Table I. A series of analogs of 111 is listed in Table 11. ' h e (+hers were made by methods a and b of Chart I.
The majority of cis amino alcohols required in the preparation of the ethers (Tables I and 11) were made by the general procedures c-f of Chart I1 and art& listed in Tables I1 (8-11) and 111. Several of the cis amino alcohols were made by two or more methods,
Sovember 1969
DIURETIC AGENTS
thereby interrelating the stereochemistry of compounds prepared by these procedures. I n the cases where the stereochemistry was not established, compounds made by the same methods were assumed to have similar stereochemistry. The cis--4 series of amino alcohols was made by general methods c-e of Chart I1 and representative examples follow. Method c.-Condensation of dione IT' with piperidine gave the vinylogous amide V which was reduced catalytically n-ith 2 moles of hydrogen to amino alcohol IIA. Catalytic reduction of V with 1 mole of hydrogen gave the intermediate cis ketone JrI. Method d.-Several cis-A alcohols were made as illustrated in the case of 69. The benzyloxy ether (65) was made by method c. Methylation of phenol 67 gave methyl ether IIA thereby establishing the stereochemistry of 67 and related derivatives. Method e.-The preparation of 2-piperidino-a(42, (3,4,5-trimet hoxyphenyl) cy clohexanemethanol Table 111) is illustrated. Compound 42 was also prepared by method c and the two materials uere identical. The amino alcohol (42) has diuretic activity and a series of related analogs was made (Table 111). Method f.-The best method of preparation of the cis-B amino alcohols was the isomerization of the cis-A amino alcohols (obtained by methods e-e) with trifluoroacetic acid. Method g.-The synthesis of cis-2-piperidino-a- (ohydroxypheny1)cyclohexanemethanol (62) is given in Chart 111. The stereochemistry of 62 was established by methylation to 63 (prepared by method c). Salicylaldehyde was treated with 1-cyclohexen-1-ylpiperidine in Skellysolve B and the resulting adduct VI1 was hydrogenolyzed with PtOz in AcOH to 62. Amino alcohol (62) was treated with anhydrous HC1 in CHsOH (method b) to give methyl ether 34 whose stereochemistry was established as shown. Esters of the amino alcohols were prepared by treating the alcohol with an acid anhydride and are listed in Table I1 (12-17). A number of his (a,a-phenyl)-2-piperidinocyclohexanemethanols were synthesized as illustrated in Chart IV and are listed in Table IV and these compounds have good diuretic activity at the doses shown. Relationship of Stereochemistry to Diuretic Activity. -The amino alcohols (11) and ethers (111) have three asymmetric centers in which the ring junction can be either cis or trans and the side-chain configuration either erythro or threo. However, on the basis of the established stereochemistry and structure-activity relationships for these compounds, the cyclohexylpiperidine moiety is suggested to be the principal structural feature required for diuretic activity. I n general, for the amino alcohols and ethers, the erythro racemates in both the cis and trans series possess the diuretic activity (i.e., cis-B and trans-C are more active than cis-A and trans-D racemates). Examination of Table I indicates that two of the 1-[2-(p,adimethoxybenzyl)cyclohexyl]piperidines, racemates I I I B and IIIC, are potent diuretic agents in rats while the other two racemates, IIIA and I I I D , are less active a t comparable doses. The diastereoisomeric amino alcohols I1 (8-11, Table 11) also show a similar spectrum of diuretic activity.
979
The activities of various other compounds (83-98) are given in Table V. X number of factors may be responsible for this spectrum of diuretic activity of the four racemates of I1 and I11 including possible differences in metabolism and absorption of these compounds. However, a consistent picture based on consideration of the established stereochemistry of these racemates may be proposed to explain the diuretic results. Elucidation of the stereochemistry of methyl ether (111) and amino alcohol (11) racemates has been reand a summary of this stereochemical information for the amino alcohols (11)is given in Table VI. I n the amino alcohols and the ethers, the active erythro-cis and -trans racemates have the piperidine group equatorial on the cyclohexane ring (see Table VI). I n contrast, the piperidine ring in the cis-threo alcohol (IIA) is axial. The conformation of the piperidine substituent in the inactive threo ether (IIIA) has not been established. Since the ethers cannot form a hydrogen bond, the conformation of 111-4 with the piperidine group equatorial may be in predominance.s I n summary, two important stereochemical factors appear to influence diuretic activity in this class of compounds : (a) configuration at the side-chain carbon -erythro isomers are most active diuretic agents a t the doses shown; and (b) conformation of the piperidine substituent on cyclohexane ring-racemates with an equatorial piperidine are most active. To test the importance of an equatorial piperidine for diuretic activity, 1-cyclohexylpiperidine hydrochloride (Table V, 96), cis-2-piperidinocyclohexanemethanol hydrochloride (97), and tra?ls-P-piperidinocyclohexanemethanol hydrochloride (98) were tested for diuretic activity. These three compounds would be expected to have the large piperidine group essentially equatorial, although cis-2-piperidinocyclohexanemethano1 hydrochloride (97) may exist as an equilibrium mixture of conformations with axial and equatorial trans-2-Piperidinocyclohexanemethanol piperidines. hydrochloride (98) and 1-cyclohexylpiperidine hydrochloride (96) were found to have significant diuretic activity while cis-2-piperidinocyclohexanemethanolhydrochloride (97) was less active a t comparable doses. The suggestion that the equatorial piperidine portion of the molecule of I11 may be responsible for part of the diuretic activity is consistent with the structureactivity relationships for analogs of I11 (see Tables I and 11). For example, dZ-IIIB and optical isomers cl-IIIB and I-IIIB (Table I, 2-4) have approximately the same diuretic activity, but the cZ isomer is much less anticholinergic than the 1 or dl isomers. Also, variation in the size of the benzylic ether in the active erythro (B racemate) series appears to have only a small effect on diuretic activity (cf. 2, 28, and 29). I n conclusion, one basic structural requirement for diuretic activity of amino alcohols and amino ethers related to I1 and 111 may be the 1-cyclohexylpiperidine moiety (Le., it has both affinity and intrinsic activity38). This point will require further substantiation. Structure-Activity Relationships.-In analogs of the amino ethers (ie., analogs of IIIB), the relative configuration of the side-chain benzylic carbon is very important as discussed before, the erythro (B and C ) isomers being the most active. For maximal diuretic
080
i
X
Y
DIURETIC AGEXTB
November 1969
981
activity the benzylic carbon should have an ether substit.uent', ie., replacement of the benzylic methyl ether hy H, OH, or *COR resulf'erl i n dewcased a,ct,ivit,y (cf. 2, 7, 9, 13, Tables I and 11). The amine may be pyrrolidine, piperidine, or 3-azabicyclo [3.2.2]nonane; a secondary amine (26, Table 11) and the primary amine (27) were lcss act,ive. The aromatic ring may be substit'uted by 0- or p-methoxy or an o-hydroxy group (qf. 2, 34, 35, Tables I and 11); methyl subst'itution on the aromatic ring appears t o increase activity (cf. 2, 36, 37). Introduction of t'wo methyl groups para to the piperidine substituent in the cyclohexane ring increases activit'y (83, Table V). The N-oxide of I I I B was less active (40, Table 11). A large number of 1,3-aniino alcohols were made and the majority (Table 111; Table 11, 8-11) were less active than the corresponding methyl ethers at the doses shown. Rlaxinixl diuretic activity in t.he amino alcohols was found in compounds (41, 42, 53, 54, Table 111). I n the bis aromat'ic series, amino alcohols (78,80, 81, Table IV) were the most active.
Experimental Section
ouou 3330
Biological Methods. Diuretic Assay (Rats).-Modifications of the method of Lipschita, et al.,' were employed for diuretic assays. Upjohn Sprague-Dawley male rats (200 f10 g on test day) were hydrated orally with physiological saline (25 cc/kg) and starved for 18 hr. Water was allowed ad lib. Water was removed 1.5 hr before test time. At test time individually weighed rats were again hydrated (25 cc/kg) with saline and the test, comporind was incorporat,ed in this load. Groups of seven rat,s (in individual metabolism cages) were routinely used a t each dose level of each test, compound. Seven untreated control rat,s (saline only) were used for each assay. Diuretic responses for the 5-hr test period \\-ere calculated in terms of per cent increase in urine excretion in excess of the controls. St'atistical analysis of excretion data on 1280 cont>rol rats shows the mean excretion (per cent of load) to be 83.51 =t19.23 SD. Sodium, potassium, chloride, and pH determinations were done on interesting compounds and dose-response relationships were determined. All diuret,ic studies were carried out in a temperature (24.3") and hnmidity (487:) con1 rolled laboratory. Diuretic Assay (Dogs).-Compounds active in rats were often furt,her evahiatetl orally i n six trained purebred female beagle dogs. Assay procediire for dogs \vas identical wit,h that for rats except their bladders were dtaiiierl (by cathet!er) 1 hr after the saline load, j i i s t prior to drng adniit~istration. This urine colleo tion was disoxtded atid not, iised in c~dciilatioiisof diiiretic response. Each dog served as her own contrul. Uiine volumes, electrolytes (Na, K, Cl), and pN values on drug days were compared to those on control days. Top 95co confidence limits have been established for each dog for urine volume and electrolytes. These top limits in each parameter were used as a control basis for significant diuretic activity and electrolyte excretion when testing drugs in these dogs. Pharmacology-Represeiit,ative ronipoiitids of the amino ethers, amino alcohols, and bis nroniat ic amino alcohols were further evaluated as diiirelic agents. 011the hasis of diuretic activity and electrolyte excretion patteriis iu the rat and dog, toxicity in mice and rats, and atiticholinergic: activity (isolated gut) d-IIIB (Table I, 3) was rhoyen for clinical evalrilat,ion. This compound was foinicl to have atropine itidex o f to l/lWO on the rabbit ileum ivhile the 1 arid dl isomers for compaiison had anticholinergic activity '/?wth of atropine. The rat st,udy shown in Table VI1 shows ti-IIIB to be an orally ell'ective diureticsaluretic agent wit,h minimal efl'ects on potassium excretion. The dog study (Table VIII) shows d-IIIB to be an orally effective diuretic-saluretic agent, in all dogs wit,h a significant kaliuretic effect in three of the six dogs. General Pharmacology-il general pharmacologic profile xvas (7) W. L. Lipsohits, Z. Hadidian, and A. Kerposar, J . Pharmacol. Ezp. Thsrap., 79, 97 (1943).
!>82
November 1969
DIURETIC AGEXTS
m m
-
Ei. a- x^ E- E- E4 u a-u-u-a-u-o-a4
u-wCI
"c
h
U 9
u
i
ci"
983
obtained on (E-IIIB as follows. Intravenously in anesthetized dogs small doses were ineffective on the mean arterial blood pressure while larger doses (4-8 mg/kg) produced prolonged depressor responses. No anticholinergic, anticholinesterase, adrenergic, adrenolytic, histaminic, or antihistaminic actions were observed. Epinephrine responses were potentiated. The suggestion of hypotensive activity was not borne out in aortic intubated8 unanesthetized normotensive rats. I n anesthetized cats (pentobarbital or chloralose) and a decerebrate cat (ether anesthesia) d-IIIB did not function as a neuromuscular blocking agent. I n the rat hind paw edema (carageenin induced) assay d-IIIB exhibited antiinflammatory activity equal to phenylbutazone but only a t high doses bordering toxic levels and therefore cannot be considered specifically antiinflammatory. X o phototoxicity was observed in rats following oral doses of 38.7 mg, kg. I n the mouse d-IIIB did not antagonize electroshock convulsions or aid in their induction nor did it afford protection against thiosemicarbazide or strychnine lethality. Ambient temperatures tested had no effect on the lethality of d-IIIB, suggesting it is not a sympathomimetic agent. Intraperitoneally in mice the LDSOwas found to be 40 mg/kg (35.9-44.7).9 Orally in rats the LDbo was 91.6 mg/kg (76.6109.5).8 Chemical Methods. lo 2- [ (p-Benzyloxy )benzoyl]cyclohexanone u as prepared from 1-cyclohexen-1-ylpyrrolidineand p-benzyloxybenzoyl chloride as previously described in the case of IV.1 The crude diketone did not require purification vzu the copper complex. It was triturated with Et20 to obtain material suitable for further use: 737, yield, mp 112-117". A sample was recrystallized for analyses from CsH6: mp 111-111.5°. Anal. (C2OH2003) C, H. cis-2-Piperidino-a- (p-benzyloxypheny1)cyclohexanemethanol (65) was prepared in a manner analogous to that described for the preparation of 8 (method c).' The product precipitated from solutioii during the hydrogenation step and was separated from the catalyst by addition of sufficient EtOH, heating to boiling, and filtering. On cooling, the product separated from solution: yield 53%, mp 146.5-147.5. The analytical sample melted a t 148.5-149.5 (EtOH). cis-2-Piperidino-a- (p-hydroxypheny1)cyclohexanemethanol Hydrochloride (67).-A suspension of 10 g (0.0264 mole) of 65 in 300 ml of CHIOH was hydrogenated in the presence of 1.0 g of 10% Pd-C for 24 hr a t an initial hydrogen pressure of 3.5 kg/cma. The mixture was filtered, the filtrate was evaporated to dryness, and residue was crystallized from EtOH-Hz0 (1: 1): 7.0 g (91.S70), mp 177-177.5. Recrystallization from EtOH raised the melting point to 179-180". Anal. (ClgH27r\;02)C, H, N. The hydrochloride was prepared and crystallized from EtOHEtrO: mp 204-204.5". cis-2-Piperidino-or- (p-ethoxypheny1)cyclohexanemethanol Hydrochloride (69).-A mixture of 2.85 g (0.01 mole) of 67 and 0.45 g of 53.3% KaH mineral oil dispersion in 50 ml of dry DMSO mas stirred for 1 hr. A sulution of 2.0 g (0.01 mole) of ethyl p toluenesulfonate in 15 ml of Eta0 was added over 30 min, and the mixture was stirred for 5 hr and poured into ice-water. The product was extracted into Et20 which was washed with HzO, dried (Sa2S0,), and evaporated to an oil. A hydrochloride was prepared with ethereal HC1 and crystallized from EtOH-EtzO: 2.2 g (637,), mp 218-219". The analytical sample melted a t 221-222' (EtOH-EtaO). 1,2,3,4,4a,Sa-Hexahydro-4a-piperidinoxanthen-9-ol (89) was prepared by a modification of the procedure of Paquette.11 -4 solution of 61 g (0.5 mole) of salicylaldehyde and 83 g (0.5 mole) of distilled 1-cyclohexen-1-ylpiperidinein 500 ml of Skellysolve B was allowed to stand 24 hr a t room temperature and then cooled ' for 1-4 days. The solid which separated was filtered and at 0 washed with Skellysolve B to give 102 g of cnide product, mp 84-90'. The yellowish solid was crystallized from Skellysolve (8) hlodifications of tlle metliod of J. R. Weeks and J. A. Jones, Proc. SOC.Exp. B i d . M e d . , 104, 646 (1960). (9) Spearman-Karber method (D. J. Finney, "Statistical Method in Biological Assay," Hafner Publishing Co., New York. N. Y., 1952, p 524). (10) hlelting points were taken in a Thomas-lioover capillary melting point apparatus a n d are uncorrected. T h e ir, uv, and nmr spectra were compatible with the assigned structures. Where analyses are indicated only by symbols of t h e elements, analytical results obtained for these elements were within &0.4Y0 of the theoretical values. (11) L. A. Paquette a n d H. Stucki. J . Org. Chem., 31, 1232 (1966).
Method c
\'
IIA ( 8 ) 4
\'I (90).cisketoiie
I
65
69
Method e
+
P-
+
o
cisB .4lcohols
Method f
U Method g
c,
Sovcmber 19G9
985
DIURETIC iiGENTS
bCH,
WI (91). trans ketone
78
Method i OCHj
e
COOR
+
2BrMg
\ /
OCH,
-
w
IX
80
BIS( a,a-PHEKTL)-2-PIPERIDIKOCYCLOHES.\SE>fETH.WOLS
No.
Ring junction
Yield,b
R1
R2
Rs
Method
bfp, 'C
Recrystn" solvent
Formula
.4nalyses
Diuretic act.c yo increase in urine excretion over controls Dose, rng/kg PO 5 20
OCH3 h 6jd 135-136 H H cis E-Pe 77 C, H, N 25 71 C2~H3&03 OCH3 h 51 111-113 H H 78 trans S 46 96 C2~H3jN03 C, H, X H OCHI H 1 79 cis C Z ~ H ~ ~ N C, O ~H, S 4 29 5.5 145-146 S 80 cis OCHI H H i 60 197-19s 1\Ie C46H35S03 C, H , N 70 133 81 cis OCH,CH, H H i 63 185-157 Me C28H39N03 C, H, N 48 124 X2 cis H H OCHICHp i C28H3gN03 C, H, li 2 1 5.5 164-165 8 a See Table 11, footnote a. * The yields are calculated on the basis that compound IX is the ethyl ester. Compound IX is a mixture of methyl and ethyl esters; see Experimental Section. See Table I, footnote b. Yield of 77 by method i was 2157.
B to give 66 g (46%) of 89 as an off-whit,e solid melting a t 84-86'. Recrystallization from Skellysolve B gave the analytical sample melting at 86-87'. cis-2-Piperidino-a- (o-hydroxyphenyl )cyclohexaneme thanol (62).--h solution of 25.8 g (0.090 mole) of 89 in 200 mi of AcOH was hydrogenated in the presence of 1.0 g of PtO2 for 4 hr a t an initial hydrogen pressure of 3.5 kg/cm2. The mixture was filtered, t.he filtrate was evaporated in vucw), and the solid residue was triturated with Et20 to give 30.6 g of the AcOH salt of 62, mp 199-201" (softens -190'). The AcOH salt was converted t'o the free base with NaEICO3. The product was extracted into C H d X which was washed with NaCl solution, dried ( N a O , ) , and evaporated to a pink solid. T h k solid was crystallized from Skellysolve B: 20.6 g (SO4;), nip l19-12Lo. lteciya~allixttlioiiTrom Skellyaolve B raised the melting point of 62 to 122-1225'. cis-2-Piperidino-a- (0-methoxyphenyl)cyclohexanemethanol (63).-A mixture of 5.8 g (0.02 mole) of 62 and 0.90 g of 53.3y0
S a H mineral oil dispersion in 50 ml of dry DMF was stirred for 15 min a t room temperature and then cooled to 0-5". -4solution of 3.8 g (0.02 mole) of methyl p-toluenesulfonate in 5 ml of DMF was added over 5 min. The mixture was stirred for 45 min at 0-lo", t'hen for 5 hr at room temperature, and allowed to stand 16 hr. The mixture was evaporated in I.UCUO and the residue dissolved in lM)-H20. The Et20 layer was washed with H20 aiid NaCl solutioii, dried (M~SOJ),aiid evaporated to a gum which was crystallized from pentane in two crops: 4.05 g, mp 98-100'; 0.75 g, mp 98.5-100.5; the combined yield was 80%. Mixture melting point nit,h authentic 63 prepared by method c showed no depression and the ir spectra were identical. n's-1 [2-~o-Hydroxy-~-methoxybenzyl)cyclohexyl] piperidine (a).-:solrit,ioii \ o f 29 g (0.10 mole) o f 62 i i i 11. t ~ CII3OH f containing 50 g of anhydrous HC1 was heated at reflux for 16 hr. The solution was evaporated in oucuo. The residue was dissolved in cold water, basified with lOy0 Na,COa, and extracted
-
atereu-
Struct,ure
?lieinistry
rS
T
r'
A
IT '1
:itJ
987
Diuretic a c t . L
yo increase in
No.
Stereochemistry
Structure
Yiehl,
1111,Recrystnu OC solvent
yo
hletliod
Formula
urine excretion over controls Dose, mg/kg PO 5 20
.illal\.Jt.a
9
77
I0
74
cis
0
