Cholinergic Effects of Molecular Segments of Apomorphine and

pamine-lesioned rats, and (5) activation of adenylate cyclase in homogenates of mouse caudate nuclei. Instead of dopaminergic effects l-(3' ...
0 downloads 0 Views 783KB Size
1194

Journal of Medicinal Chemistry, 1975, Vol. 18, No. 12

(12) J. B. Hynes and W. T. Ashton, J . Med. Chern., 18,263 (1975). (13) J. H. Freisheim, C. C. Smith, and P. M. Guzy, Arch. Biochem.

Biophys., 148, l(1972).

B.Hynes, J . Med. Chem., 16,694 (1973). (15) W. E. Richter, Jr., and J. J. McCormack, J . Med. Chem., 17, 943 (1974). (14) W. T. Ashton, F. C. Walker, 111, and John

Ginos et al. (16) R. Ferone, J. 3. Burchall, and G. H. Hitchings, Mol. Pharmacol., 5,49 (1969). (17) R. I. Geran, N. H. Greenberg, M. M. MacDonald, A. M. Schumacher, and B. J. Abbott, Cancer Chemother. Rep., 3 (Part 3), 7 (1972). (18) T. S. Osdene, P. B. Russell, and L. Rane, J . Med. Chem., 10, 431 (1967).

Cholinergic Effects of Molecular Segments of Apomorphine and Dopaminergic Effects of N,N-Dialkylated Dopaminest James Z. Ginos,la George C. Cotzias,lb Eduardo Tolosa,la Lily C. Tang,la a n d Anthony LoMontela Medical Research Center, Brookhaven National Laboratory, Upton, New York 11973, and the Department of Neurology, Cornell Medical College, New York, New York 10021. Received April 2,1975 The hydrochlorides of molecular segments of apomorphine [2-(3’,4’-dihydroxybenzyl)-1,2,3,4-tetrahydroisoquinoline, 2-(3’,4’-dihydroxybenzyl)piperidine, and 1,2,3,4-tetrahydroisoquinolinewith their respective N-methyl and N-n-propyl homologs] and N,N-dialkylated dopamine compounds were synthesized and studied for ( 1 ) LDso in intact mice; (2) stereotypy in intact mice; (3) curving of the body in unilaterally caudectomized mice; (4) rotation in B-hydroxydopamine-lesioned rats, and (5) activation of adenylate cyclase in homogenates of mouse caudate nuclei. Instead of and 2-methyl-1,2,3,4-tedopaminergic effects l-(3’,4’-dihydroxybenzyl)-2-methyl-1,2,3,4-tetrahydroisoquino~ine trahydroisoquinoline showed cholinergic ones. These effects were blocked in atropine-pretreated animals. Of the N,N-dialkylated dopamine compounds synthesized, the N-n-propyl-N-n-butyldopamine ranked in all tests as the strongest dopamine-receptor agonist and N-methyl-N-n-propyldopamineas the weakest. In contrast, N,N-dimethyldopamine and 1-(3,4-dihydroxyphenylethyl)piperidineshowed no dopaminergic effects. The effectiveness of the dopaminergic agonists depended on the length of the N-alkyl substituents suggesting interactions with hydrophobic regions of the receptor site. An a t t e m p t to develop drugs effective against Parkinson’s disease b u t having side effects different from those of t h e dopamine (DA) precursor I-Dopa2a led t o studies of apomorphine l b because of some structural similarities between l b a n d DA.2b T h e dopaminergic effects of l b seem to be explained by these similarities, b u t some opposing effects encountered during these studies have been only tentatively e ~ p l o r e d T. ~h e potentiation of t h e therapeutic effects of oral !-Dopa by injected lb is compatible with a dopaminergic function, b u t t h e diminution of some I-Dopa side effects suggests antidopaminergic or even cholinergic properties in lb. T o clarify these effects we have synthesized molecular segments of l b a n d determined their LDjo in mice, a n d we have studied their effects o n unilaterally caudectomized mice,4 on nigral-lesioned rats: a n d , when indicated, on t h e dopamine-activated adenylate cyclase activity of homogenized mouse caudate In t h e present study, elimination of rings A a n d C from lb with retention of t h e catechol (ring D) a n d t h e piperidine (ring B) resulted in loss of recognizable dopaminergic or cholinergic effects (17a-c). W h e n t h e bond joining ring D to ring A was eliminated (21a-c), dopaminergic effects disappeared a n d cholinergic ones appeared so that it was necessary to dissociate t h e tetrahydroisoquinoline from t h e piperidine moieties. After synthesizing a series of 1,2,3,4tetrahydroisoquinolines (19a-c) a n d piperidines we found that some of t h e former had central cholinergic activities. T o preserve dopaminergic while eliminating cholinergic activity we synthesized a n d investigated N,N-dialkyl-substituted dopamines (DA) (9, 22-25). When we found t h a t three relatively nontoxic alkyl-substituted DA h a d dopaminergic activity in lesioned mice a n d rats, we deemed t h e m worthy of testing on DA-activated adenylate cyclase in homogenized mouse caudate nuclei”.’ t o confirm this activity. *This work was supported by the National Institutes of Health (Grant NO 111’31) and the Energ\ Research and Development Administration

I

I

R

R

la,b,c

Ho&

17a,b,c

I

0 RI

R 2 l a , b , ca = - H ; b = - C H 3 ; c = - C H19 ~ a,C b,c H~CH~

Results and Discussion Chemistry. l-Methyl-2-(3’,4’-dihydroxybenzyl)piperidine hydrochloride (17b) was synth_esized by two different routes as illustrated by Schemes I a n d 11. In Scheme I, hydrogenolysis in s t e p B enl loved both the O-benzyl protective groups a n d t h e benzylic hydroxyl,* while in Scheme I1

Journal of Medicinal Chemistry, 1975, Vol. 18, No. 12

Apomorphine and N,N-Dialkylated Dopamines

1195

T a b l e I. Preparation a n d Properties of Alkyl Iodide Derivatives

Compound

Q

MP (lit.), “C

Recrystn solvent

87

185-1 86

EtOH-EtOAc

78

149-1 5 0

Et OH -Et OAc

15H 1 81N02

84

166-168

Et OH -E OAc

lsH1 81N02

86

150-152

2 -Propanol

Cl?HdN02

202-203 (139-140b) 205-206

EtOH

C19H201N02

EtOH

C21H241N02

15b

1-Methyl -2 -(3‘,4’-dihydroxybenzy1)pyridinium iodide 1-n-Propyl-2 -(3’,4’-dihydroxybenzyl)pyridinium iodide 1-Methyl -2 -(3’,4’-dimethoxybenzy1)pyridinium iodide 1-Propyl-2 -(3’,4’-dimethoxybenzy1)pyridinium iodide 1-(3’,4’-Dimethoxybenzyl)-2-methylisoquinolinium iodide 1-(3’,4’-Dimethoxybenzyl)-2 -=-propylisoquinolinium iodide 2 -Methylisoquinolinium iodide

EtOH-ether

15c 16c

2 -n -Propylisoquinolinium iodide 1-n -Propylpyridinium iodided

161-1 62 (16 1-1 63‘) 124-12 5 68-72

Yield,

12b 12c 13b 13c 14b 14c

97 91 98 98 75

Formulaa Ct3H14N021

EtOH-ether

C12H14m

aAll were analyzed for C, H, I, and N and results were within 0.4% of theory. b J . L. Neumeyer, M. McCarthy, S. P. Battista, F. J. Rosenberg, and D. G. Teiger, J. Med. Chem., 16, 1227 (1973). cv. E. Schlittler and J. Muller, Helu. Chim. Acta, 31, 914 (1948). dCould not be crystallized (hygroscopic).

Scheme I

S c h e m e I1

PhCHpO

CH $0

I p-cymene

CHO

COOH

A

A PhCHzO

” -



CHll

cH30351.J3 -

~

SOCl2

B

OH

OH

B

HO

C

1

CH~O

C HIO

HO I

I

11

21

the 0 - m e t h y l protective groups were removed in t h e last step. T h e overall yield (12%) in Scheme I1 was twice t h a t obtained by Scheme I (5-6%) (see Tables 1-111). T h e synthesis of t h e doubly 0-methylated homologs of 21a-c is described e l ~ e w h e r e .T~h e subsequent hydrogenation of t h e alkyl iodides a n d hydrochlorides of t h e isoquinoline and pyridine derivatives t o their corresponding tetrahydroisoquinolines a n d piperidines gave yields in excess of 90% with two notable exceptions. T h e hydrogenation of t h e hydrochlorides of 1-(3’,4’-dimethoxybenzyl)isoquinoline a n d isoquinoline gave respectively yields of 45 a n d 66% of the corresponding 1,2,3,4-tetrahydroisoquinolines a n d 55 a n d 44% of t h e 5,6,7,8-tetrahydroisoquinoIines. T h e effect of experimental conditions on t h e ratio. of these two isomers is discussed e l ~ e h w e r e . Compounds ~ 21a-c (Table IV) were found t o be hygroscopic solids, noncrystallizable, with broad melting point ranges and were therefore identified by TLC, lH NMR, a n d elemental analysis after precipitation from ethanol with ether. Compounds 9 a n d 22-25 (Table IV) were synthesized according t o Scheme 111.

13 CHIO

CH i

H C)

P h a r m a c o l o g y . T h e LD50 (pg/g of body weight) was determined for each analog (Table V) so t h a t well-defined, nonlethal dosages could be chosen for screening behavioral effects in animals. T h e hydrochlorides of lb, N-n-propylnorapomorphine ( I C ) , DA, a n d N-methyl-DA were included in t h e toxicity studies as reference substances. An analysis of variancelo of t h e LD50 values by computer showed that, except in the case of the piperidine homologs, the N methyl tertiary amines were, in general, the most toxic. Among t h e piperidine homologs, this analysis showed N n-propylpiperidine t o be ten times as toxic as N-methylpiperidine.

1196 Journal of Medicinal Chemistry, 1975, Vol. 18, No. 12

Ginos et al.

T a b l e 11. Preparation and Properties of the Hydrochlorides of 1,2,3,4-Tetrahydroisoquinolineand Piperidine ~

Yield, 0/o Compound 17a 17b 17c 18a

18b 18c 19a 19b 19c

20c

(Scheme)

2 -(3 ',4' -Dihydroxybenzyl)piperidine 1 -Methyl-2 -(3',4'-dihydroxybenzyl)piperidine 1 -n -Propyl-2 -(3',4'-dihydroxybenzyl)piperidine 1 -(3',4'-Dimethoxybenzyl)-l,2,3,4-tetrahydroisoquinoline 1 -(3',4'-Dimethoxybenzyl)-2-methyl 1,2,3,4-tetrahydroisoquinoline 1-(3',4' -Dimethoxybenzyl)-2 -n -propyl 1,2,3,4-tetrahydroisoquinoline 1,2,3,4-Tetrahydroisoquinoline

MP (lit.), "C

Recrystn solvent

212-213 22 6-2 27

MeOH + EtOAc Me OH -Et OA c

230-231

Et OH-EtOAc

227 -2 28

EtOH

233 -2 3 5 (227-230 *) 203 -2 04

EtOH

197-198

Formula"

MeOH-ether EtOH

( 195-197d)

2 -Methyl - 1,2,3,4-tetrahydrois o quinoline 2 -n -Propyl -1,2,3,4-tetrahydroisoquinoline 1 -Propylpiperidine

225-2 26

EtOH-EtOAc

23 8.5-2 39.5

EtOH-ether

224-2 26 (225')

EtOH-ether

aAll were analyzed for C, H, C1, and N and results were within 0.4% of theory. b J . L. Xeumeyer, M. McCarthy, S. P. Battista, F. J. Rosenberg, and D. G. Teiger, J. Med. Chem., 16, 1227 (1973).CH.W.Magnusson and E. R. Schierz, Uniu. Wyo. Publ., 7, 1 (1940).dL. Helfer, Helu. Chim. Acta, 6 , 785 (1923).

T a b l e 111. Preparation and Properties of the Hydrochlorides of N ,N -Dialkyl-/3-(3,4-dimethoxyphenyl)ethylamine Compound

11 8

S,iV-Dimethyl-p-(3,4 -dimethoxypheny1)ethylamine N-Methyl-N-n-propyl-P-(3,4 -dime thoxypheny1)ethylamine ,V-Methyl-N-n-butyl-p-(3,4 -dimethoxypheny1)ethylamine Al"-n-Propyl-N-n-butyl -p-(3,4dimethoxypheny1)ethylamine 1 -(3,4-Dimethoxyphenylethyl)piperidine

Yield. $h

Mp (lit.), 'C

1 95-1 96 (19'lb) 142-143

88 87

13 0.5 -13 1

88 81

76-78

Recrystn solvent

For mula"

EtOH

C1zH1 xC1NOz

C H 2 C 12 -Et OA c

C14H24C1N02

CH,OH-Et,O

C15H26C1N02

CHSOH-Et20

C1'7H30C1N02

EtOH-EtOAc

15H24C1N02

2 14-2 15 90

(149-152,' 210-211d)

O A l l were analyzed for C, H, C1, and N and results were wjthin 0.4% of theory. %eference 20. c Z . Arnold and K. Hejno, Collect Czech Chem Commun , 20,567 (1955).dL. Dbbravkovi, I. Jeko, P. Seftovit. and Z . VotickL, Chern. Zuesti, 9,541(1955).

Two animal tests were used in assessing these drugs for dopaminergic or cholinergic properties. (1) Behavioral effects on caudectomized Sprague-Dawley a n d Swiss albino mice. These mice had the right caudate nucleus partially ablated4 and, starting 3 weeks later, reacted t o l b a n d other dopaminergic agents by curving t h e body toward the side with the lesion and to strong cholinergic agents (oxotremorine) by curving opposite t o the side with the lesion.22 (2) Behavioral effects on nigral-lesioned rats (SpragueDawley). These rats had 6-hydroxydopamine injected stereotactically in t h e right substantia nigra by the method of Ungerstedt e t al.,5 and, starting 4 weeks later, injection of l b (1 pg/g) caused them t o rotate in the direction away from the lesion" whereas cholinergic agents caused them t o rotate toward the lesion, and t h e number of turns was measured in rotometers5 (Table V). Nigral-lesioned rats were far more sensitive t h a n caudectomized mice. Some of the analogs were screened for DA-stimulated adenylate cyclase activity in mouse caudate h o m o g e n a t e ~ ~ , ~ (Table VI).

Except for N,N-alkylated DA (9, 23, and 24) none of t h e synthesized analogs (Table V) showed dopaminergic activity in either nigral-lesioned rats or caudectomized mice. In contrast, compound 21b (50 wg/g ip) produced responses in the nigral-lesioned rats t h a t were t h e opposite of those to dopaminergic agents b u t like those t o oxotremorine" and were blocked by atropine pretreatment (0.5 fig/g). Since others have demonstrated t h a t N-ethyltetrahydroisoquinoline derivatives are cholinesterase inhibitors,12 it was possible t h a t the cholinergic activity of 21b was contributed by the tetrahydroisoquinoline moiety. Indeed, when 19b a n d the tetrahydroisoquinoline and piperidine derivatives (Table V). were tested, only 19b produced in both caudectomized mice and nigral-lesioned rats responses like t h a t t o oxotremorine, which were blocked by atropine pretreatment (5 wg/g). T h e above findings suggested t h a t l b a n d perhaps its homolog IC and even 0 - m e t h y l metabolites of the latter13J4 have antidopaminergic or cholinergic properties, which might explain some of the antagonistic effects t o l-Dopa.

Journal of Medicinal Chemistry, 1975, Vol. 18,No. 12 1197

Apomorphine and N,N-DialkylatedDopamines

T a b l e IV. Preparation of the Hydrochlorides of Catechol Derivatives by Demethylation a n d Their Properties Yield, (k (Scheme)

Compound 1-Methyl -2 -(3‘,4’-dihydroxybenzyl)piperidine 1-(3’,4’ -Dihydroxybenzyl) -1,2,3,4 tetrahydroisoquinoline 1-(3 ’,4 ‘-Dihydroxybenzy1)-2 -methyl 1,2,3,4 -tetrahydroisoquinoline 1-(3 ’,4’ -Dihydroxybenzyl) -2 -n -propyl 1,2,3,4 -tetrahydroisoquinoline N,N-Dimethyl -p -(3,4 -dihydroxyphenyl)ethylamine N-Methyl -N-n -propyl -/3-(3,4 d i h y d r o x y pheny1)ethy lamine N-Methyl -N-n-buty 1 - p - (3,4 -dihydr oxypheny1)ethylamine N-n -Propyl-N-n-butyl-p-(3,4-di hydroxypheny1)ethylamine 1- (3,4 -Dih ydrox yphenylethy1)piperidine

1% 21a

2 lb 2 IC 22 9

23 24

25

(n)

Recrystn solvent

Mp (lit.),“C

Formula‘

22 8-2 29

Me OH-E t OAc

C13H20C1N02

98

Broad range

EtOH-Et20

C16H18C1N02

96

Broad range

EtOH-Et20b

C iTH20ClN02

94

Broad range

EtOH-Et20b

Ci9H27C1N02

88

EtOH-Et20

89

123-125 (127)c 137 -1 38

Et OH -Et OA c

c 12%

94

146-1 4 8

EtOH-EtOAc

C13H22C1N02

87

136-1 3 8

EtOH-Et20

70

241-242

86

10H16C1N02

15H26C1N02

MeOH-Et,O ~

~~~~

0 C 1NO2

13H20C1N02 ~~~

~~

aAll were analyzed for C, H, C1, and N and results were within 0.4%. *The compounds were hygroscopic and could not be recrystallized. They were precipitated from ethanol with ether. CReference20.

S c h e m e I11

R’ CH30

C HO

I R’

D

I

R’

This led us t o synthesize t h e N,N-dialkyl-DA analogs in which t h e tetrahydroisoquinoline segment (rings A and B) was eliminated. Of these, 9, 23, a n d 24 showed dopaminergic properties in tests with rats a n d mice b u t t o a far lower degree t h a n 1 b.“ I n the more sensitive nigral-lesioned rats as little as 3 pg/g of 24 produced significant rotation t o t h e left (234 f 54 turns130 min) b u t 5 Hg/g of 9 or 23 did not produce rotation significantly greater t h a n t h a t of waterinjected controls. For comparison, 1 Wglg of l b produced 535 f 61 turns130 min. T h e greater effectiveness of 24 t h a n of 9 a n d 23 as a DAreceptor agonist was paralleled by its degree of in vitro stimulation of adenylate cyclase (Table VI). T h e increase in cAMP production (in excess of nonstimulated production) due t o 24 was equal t o t h a t due t o a n equimolar amount of DA, whereas t h e increases due t o 9 a n d 23 were respectively 24 a n d 70% of t h a t due t o DA under identical conditions. N-Methyl-DA, a compound t h a t showed no dopaminergic activity in mice (50-150 pglg) a n d only a weak one in nigral-lesioned rats at lethal dosages (25 pg/g), stimulated cAMP as much as did DA under t h e same conditions. In contrast, 22 showed no dopaminergic effects in any of these tests.

Further studies are required t o determine t h e dependence of dopaminergic effects on the length a n d the degree of branching of the N-alkyl chains. T h e importance of the stereochemical arrangement of the N,N-alkyl substituents with respect t o interaction with t h e receptor site is suggested by the failure of 25, which (unlike 24) has its two N alkyl substituents constrained into a piperidine ring, t o have a n effect in caudectomized mice a n d nigral-lesioned rats. Although steric effects would be expected t o have increasing importance with lengthening of the N-alkyl chains, with adverse results, this was not borne out by our findings. Perhaps interaction with the hydrophobic regions of t h e receptor site15 is t h e predominant factor in shaping the “biologically active” molecular conformation. Our findings t h a t some N,N-dialkyl-DA analogs have low toxicity as well as dopaminergic properties suggest t h a t one or more of them may be of potential use in the study of parkinsonism and related diseases. These compounds would offer the following advantages over l-Dopa and the other dopaminergic agonists (lb, IC, a n d bromocriptine16). 1. They are easy t o synthesize in high yields and in a relatively short time from commercially available starting materials. 2. They have no optical enantiomers necessitating the separation of biologically active from biologically inactive components. 3. Unlike DA, they cross the blood-brain barrier because being tertiary amines they are resistant t o deamination by monoamine oxidase (MAO).17 Experimental Section Uncorrected melting points were determined on a Thomas-Hoover apparatus. ‘H NMR spectra were recorded on a T-60 Varian in CDC13 or Me&O-d,j with (Me)&i as the internal standard. Eastman chromagram sheets (6060 silica gel with fluorescent indicator) were used for TLC. The following TLC solvent systems were used: (A) C,jH&IeOH, 9:l; (B) 4:l; (C) 3:2; (D) 1-butanol-HzO-AcOH, 4:l:l; (E) I-butanol-H20-AcOH-pyridine, 15:12:3:10; (F) cyclohexane-EtOAc, 1:l; (G) 1:4;(H) 41; (I) CH&l-MeOH-AcOH, 17: 2:l; (J) CHCls-MeOH, 99:5; (K) n-PrOH-HZO, 89:28. Infrared spectra were obtained on a Perkin-Elmer 337 grating spectrometer, and elemental analyses were done by Galbraith Lab., Inc., and Schwarzkopf Microanalytical Lab. Materials. 3,4-Dimethoxybenzaldehyde, 3,4-dibenzyloxybenz-

1198 Journal of Medicinal Chemistry, 1975, Vol. 18,No. 12

Ginos et al.

Table V. Behavioral Effects on Caudectomized Mice and Nigral-Lesioned Rats and LD50 on Normal Micea

Compound

17ad 17bd 17ca 19a 19b 19c

21ad 21bd

Mice conjugate c u r v a tureb (eff dose, P d g )

Nigral-lesioned ratsC Dose, Direction Pg/g (turns f SEM/min)

0 0

0 0

19

(66)

21cd

20c Dopamine N-Methyldopamine

22 9 9

23 23 23 24

0 0 0

37

0 0 0 0 0 0 0

-

1oe

-

(120)

-

10 25 10 5 25

(90)

--

(30)

la, (-)-norapomorphine l b , (-)-apomorphine

IC, N - n -propyl -(-) -nor apomorphine Oxotremorineg

5 3

i

(475 (161 (732 (295

i i-

f i-

132 (129-136) 500 ( 4 3 3 4 7 7 ) 133 (125-142) 162 416 465 54 1978 212 240 295

108170) 48/30) 83/70) 50/30)

C G

G C

(155-168) (389445) (439474) (50-58 (1810-2162) (178-253) (226-255) (284-307)

C C C C C C C C

219 (209-230)

C

129 (123-136)

C

213 (200-221) 145 (138-150)

Q

323 (309-338)

Q

0

(875 i- 35/70)/ (266 i- 24/30)f (234 i 54/30)

0 0

-

C R and C C G

4.2/90)

0 0

1oe 25

24 24 25

(271-295) (120-132) (110-324) (292-314 (128-131)

Manner

of death

(14 i 3.5/139)

(14.3 Piperidine N-Methylpiperidine

283 126 189 303 131

0

-

LD50

(95% confidence l i m i t s ), pg/g

C

1

(1)

(964

i-

(0.25) (0.3)

0..3

C

132/70)

____,

C

(10.3 i- 3.4161)

chemicals were given ip as the hydrochloride salts with the exception of oxotremorine. b A total of eight mice tested for each compound. For method of testing and determining dose, see text: Tests with Caudectomized Mice. ?Each compound tested in five rats. For method of testing, see text: Tests with Nigral-Lesioned Rats. dRacemic mixtures. eHigher doses than 10 rg/g produced acute toxic effects followed often by death. No turns noted. fAccompanied by intense gnawing, tail biting, and licking. gPotent cholinergic agent. = conjugate curving and turning toward side of lesion; = conjugate curving and turning opposite to side of lesion; C, convulsion; R, rearing: G, gasping; Q, quietly.

-

aldehyde, 3,4-dimethoxyphenylacetic acid, picolinic acid, piperidine, N-methylpropylamine, N-methylpiperidine, isoquinoline, and p-cymene were obtained from Aldrich Chemical Co.; dimethylamine, N-methyl-N-n-butylamine, 1-iodopropane, and iodomethane from Eastman Organic Chemicals; (-)-apomorphine hydrochloride from Merck & Co.; dopamine hydrochloride from Calbiochem; N-n-propyl-(-)-norapomorphine hydrochloride from Sterling Winthrop Research Institute; N-methyldopamine hydrobromide and (-)-norapomorphine hydrochloride were gifts from Hoffmann-La Roche Inc.; oxotremorine from Nutritional Biochemical Corp.; N-n-propyl-N-n- butylamine from ICN-K&K Labs, Inc. 3,4-Dibenzyloxyphenyl(2-pyridyl)carbinol Hydrochloride (2). 2 was prepared as described elsewheres with the following modifications. N2 was bubbled through the reaction mixture. In the extraction with aqueous 2 N HCl, the lowest two layers were combined and basified with NHdOH, and the free base, a dark amber oil, was extracted with CHC13. The breaking of the ensuing emulsion was facilitated by warming of the separatory funnel. After removal of the CHC13 and pyridine under vacuum, the dark oily residue was taken up in ether and treated with HC1-saturated

ether. The semisolid salt was triturated repeatedly with ether until it solidified completely. The crude product was dissolved in 40 ml of MeOH and diluted with 90 ml of ether and was allowed to crystallize out at room temperature and then overnight at 4' following the addition of more solvent. The white needles were collected on a filter: wt 21.7 e (32%): mo 127-128'. The 'H NMR (Me?SO-dn) spectrum was- consisten't with the structure o f 2.- Anal. ( C Z ~ H Z ~ N OC,&H, ~ )C1, N. 3,4-Dihydroxyphenyl(2-pyridyl)methanehydrochloride (3) was prepared as described by otherss but recrystallized twice from EtOH-EtOAc before a mp of 149-151' (lit. 152-154') was attained. TLC (B, D, I) yielded one spot visualized with uv and 12 vapors. The spot darkened on standing. Its 'H NMR spectrum was consistent with its structure. 3,4-Dimethoxyphenyl(2-pyridyl)carbinol (4). 4 was prepared essentially as described by Sankey et al! The BuLi used was obtained from a commercial source. The crude product was recrystallized from 16 ml of acetone and 30 ml of heptane: mp 92-93' (lit.8 93-94'). 3,4-Dimethoxyphenyl(2-pyridyl)methane Hydrochloride (5). 5 was prepared from 4 by the method of Sperber et al.ls 5 was

Journal of Medicinal Chemistry, 1975, Vol. 18, No. 12 1199

Apomorphine and N,N-Dialkylated Dopamines

AczO, and 2.25 g (8.25 mmol) of 8 was refluxed for 0.75 hr under N2 and then the solvent was removed under reduced pressure. The viscous residue was redissolved in 7 ml of absolute EtOH and evaporated again under high vacuum. The pale yellow residue was redissolved in 7 ml of absolute EtOH, diluted with 25 ml of EtOAc, Net c y and induced to crystallize by scratching. The crystalline product clic AMP, was filtered off, dissolved in 30 ml of water, and basified with pmol (mean p NaHC03, and the free amine was extracted exhaustively with Compound i SEM) valued EtOAc. The solid residue obtained after removal of the solvent was dissolved in 8 ml of absolute EtOH and treated with HC1-saturated absolute ether. This yielded a colorless viscous oil as a precipiDopamine (7)e 21.0 i 2.8