The Synthesis of Some Tertiary Amino Alcohols Related to Chelidonine

of substance be placed on a 28 X 8 cm. calcium hydroxide column.) The separation of phytofluene from some mem- bers of the stereoisomeric a-carotene s...
0 downloads 0 Views 221KB Size
SYNTHESIS OF TERTIARY AMINOALCOHOLS RELATED TO CHELIDONINE

Feb., 1946

201

with the bluish-white fluorescence of the vitamin A section and with the orange-yellow fluorescence of the anhydro compounds, the phytofluene zone appeared definitely greenish in ultraviolet light.

carotene has the tendency to partially overlap with the pigment. (About 200 mg. of phytofluene may be handled on a 24 X 4.8 cm. alumina column.) Somewhat greater becomes the difference in the adsorbabilities of phytofluene and a-carotene when alumina-calcium hydroxide mixtures are used. The phytofluene zone is then markedly broader and the overlapping is of lesser extent. Phytofluene spreads out still more on pure calcium hydroxide and a t the same time a differentiation of some stereoisomers may be observed. (We recommend that not more than 100 mg. of substance be placed on a 28 X 8 cm. calcium hydroxide column.) The separation of phytofluene from some members of the stereoisomeric a-carotene set is difficult. I n mixed chromatograms, using calcium hydroxide and petroleum ether, the following sequence was observed from top t o bottom : vitamin A, phytofluene, anhydrovitamin A, and isoanhydrovitamin A ; the phytofluene zone was flanked by broad, non-fluorescing interzones. Compared

Summary The isolation of an oily, colorless, in ultraviolet light strongly fluorescing polyene hydrocarbon (possibly C40H64) from commercial tomato paste is described. The compound is photo- and airsensitive and shows an unusually high adsorption affinity. Some of its spectral and structural characteristics are discussed as well as its role in the bio-synthesis of carotenoid pigments. PASADENA, CALIFORXIA

RECEIVED OCTOBER 15, 1945

-

I CONTRIBUTION FROM

THE

DEPARTMENT OF CHEMISTRY OF STANFORD UNIVERSITY ]

The Synthesis of Some Tertiary Amino Alcohols Related to Chelidonine BY C. R. NOLLERAND P. D. KNEELAND determine their pharmacological action. Compound I1 can be thought of as arising from cheliproperties resembling those of papaverine.2 The donine by the breaking of two bonds of the alicurrently accepted structural formula for chelido- cyclic rings. Moreover, the product of reaction nine (I) was proposed by Bruchausen and Berschea of safrole oxide with piperonylmethylamine, 111, In view of the fact that compounds related to differs from I1 only in having one less methylene papaverine, but having the nitrogen in an acyclic group between the amino group and the hydroxyl portion of the molecule, may have antispasmodic group, and a different linkage for one of the properties less than, equal to, or greater than that methylenedioxyphenyl groups. of p a p a ~ e r i n e ,i~t was considered desirable to synthesize similar analogs of chelidonine and to Chelidonine is an alkaloid occurring in celandine

(Chelidonium wajus), and has pharmacological

\~,~-cH~sHcH~

+

CHr-0

I Chelidonine H

CH2-0

-

I1

(1) Henry, "Plant .Alkaloids," P. Blakiston's Son and Company, Philadelphia, Penn., 1939, p. 173. (2) Hanzlik, J . Phwmacol., 7, 99 (1915); 18, 63 (1921); SS, 387 (1928). (3) Bruchausen and Bersch, Ber., 63, 2520 (1930); 64, 947 (1931). (4) Rosenmund and co-workers, ibid., T I , 19,2161 (1939); Wagner-Jauregg, Arnold and Born, ibid., 72, 1551 (1939).

Accordingly the compounds listed in Table I were synthesized by analogous reactions. Their pharmacological action was tested in the laboratory of Dr. P. J. Hanzlik of the Department of Pharmacology, Stanford University Medical School, who states that they show no antispasmodic action, but instead a fairly consistent stimulating action on smooth muscle of different organs. Furthermore, the compounds on intravenous injection depressed the heart sufficiently to be practically useless for therapeutic purposes. The constitution assigned to the compounds is based on the constitution of the reaction product of benzylethylene oxide and ammonia.5 (5) Castro and Noller, THISJOUBNAL, 68, 203 (1946).

202

C. R. NOLLERAND P. D. KNEELAND

,hI,VO

VOl. 68

TABLE I ALCOHOL HYDROCHLORIDES FROM BENZYLETHYLENE OXIDES AND BENZYLMETHYLAMINES ,-Analyses,c % '

Compound, RCH?N(CHs)CH*CHOHCH*R'.HC1 R = Phenyl, R' = 3,4-Methylenedioxyphenyl R and R' 3,4-Methylenedioxyphenyl R = 3,4-Dimethoxyphenyl, R' = 3,4-Methylenedioxyphenyl R Phenyl, R' = 3,4-Dimethoxyphenyl R = 3,4--Methylenedioxyphenyl,R' = 3,4-Dimethoxyphenyl R and R.' = 3,4-Dimethoxyphenyl

-

-

SI.p . , 'C. 131-133a 152-154' 148.4-149.4'

Molecular formula CiaHizClNOa CisHzzClN01 CmH2eClN06

165,8-167.4a

C21HmClPIIOe 6 1 . 2 3

Carbon

Hydrogen

Chlorine

-

. 7 r

Calcd. Found Calcd. Found Calcd. Found 6 4 . 3 7 64.15 6 4 . 1 0 6 . 6 1 6 . 6 1 6 . 6 7 1 0 . 5 7 10.40 60.08 60.19 60.29 5 . 8 4 5.84 5.75 9 . 3 3 9.21 6 0 . 6 7 6 0 . 6 4 6 0 . 6 4 6.62 6 . 7 0 6 . 7 4 8 . 9 6 8 . 8 4 8.76 150.4-151.4b Ci9H2aCINOa 6 4 . 8 5 6 4 . 8 3 6 4 . 9 0 7 . 4 5 7 . 4 7 7 . 1 0 1 0 . 0 8 9 . 8 9 151.9-154.1b CmH2sClN01 6 0 . 6 7 6 0 . 6 7 6 0 . 7 2 6 . 6 2 6 . 7 9 6 . 7 6 8 . 9 6 8.82 61.13 61.02 7 . 3 4 7 . 3 2 7.40

* Crystallized from isopropyl alcohol. * Crystallized from ethyl alcohol-ether mixture. by Dr. E. 1%'. I). Huffman, Denver, Colo.; chlorine analyses by Mr. A. J. Castro.

c

8.61

8.35

Carbon-hydrogen analyses

amine and sufficient water was stirred into the mixture so that a 2-cm. layer remained during the stirring. The mixBenzylmethy1amines.-To a solution of 0!5 mole of the ture was allowed t o stand for three days a t room temperaaromatic aldehyde in 0.5 mole of 33% aqueous methyl- ture, reemulsification with the water being brought about amine and 150 cc. of 95y0 alcohol was added 0.5 g. of by stirring each morning and evening. The excess water platinum oxide, and the mixture was shaken with hydrogen was decanted, the remaining water removed by distillation under an initial gage pressure of 1 5 pounds. Absorption under reduced pressure, and the residue distilled a t a pres; of hydrogen usually stopped a t about 90% of the cal- sure of less than 1 mm. The fractions boiling over a 10 culated amount after four hours, although occasionally the range amounted to 85-9OC; of the calculated amounts. catalyst required three to four hours to be reduced, and The free bases were thick viscous oils and were converted twenty-five t o thirty hours was required for complete re- to the hydrochlorides by dissolving in a minimum volume duction. After removal of the catalyst and solvent, an of absolute ether, cooling in an ice-bath, and adding a cold equal volume of water was added, and then 100 cc. of 6 N saturated solution of dry hydrogen chloride in absolute hydrochloric acid, and the solution extracted with three 75- ether until precipitation was complete. At this stage the cc. portions of benzene. The aqueous layer was made hydrochlorides were very hygroscopic and in the subsealkaline by the addition of 150 cc. of 6 Nsodium hydroxide quent purification were exposed to air as little as possible, solution, the oil separated, and the aqueous layer extracted and all solvents were kept strictly anhydrous. The mixture with three 50-cc. portions of benzene. The benzene was was allowed to warm to room temperature, and the precipiremoved and the residual oil distilled a t reduced pressure. tate became pasty. The ether was decanted and as much The yields of amines having the indicated boiling ranges more as possible worked out of the paste with a spatula. were as follows: ben~ylmethylamine,~ b. p. 68-70" a t 9 Addition of acetone, dried over anhydrous potassium mm., 82%; (3,4-methylenedioxybenzyl)-methylamine,icarbonate, first dissolved the paste and then caused preb. p. 118-122' a t 7 mm., 89%; (3,4-dimethoxybenzyl)- cipitation of the hydrochloride in a granular form which methylamine,8 b. p. 112-1-17" a t 9 mm., 80%. was not too hygroscopic t o be filtered. The hydrochlorides Benzylethylene Oxides.-A solution of 0.2 mole of were recrystallized from either anhydrous isopropyl alcosafrole or methyleugenol in 320 cc. of cold chloroform con- hol, or from a mixture of absolute ethyl alcohol and ether taining 0.2 mole of perbenzoic acid was allowed t o stand until pure, as indicated by halogen ana lyse^.^" Dca t 10" until all of the perbenzoic acid had reacted (ap- composition points and analyses for the individual prodproximately three days). The solution was freed of ucts are given in Table I. benzoic acid by extracting with lo%, aqueous sodium hydroxide, and the chloroform layer was washed thoroughly Summary with water and dried over anhydrous sodium sulfate. X number of amino alcohols that may be conRemoval of the solvent and fractional distillation under sidered to be related to the alkaloid, chelidonine, reduced pressure gave the following yields of oxides havin! the indicated boiling ranges: safrole oxide,g b. p. 116-118 have been synthesized by the reaction of benzyl; methyleugenol oxide,g b. p. 190-195' ethylene oxides with benzylmethylamines. In Some of the starting material always contrast to chelidonine, which has an antispasd based on the amount reacted, the modic action, these amino alcohols have a stimuyields were 80-85(r, of the calculated amounts. Amino Alcohols.-In a .25 x 200 mm. test-tube was lating action on smooth muscle. Moreover, they ;)laced ( 1 . 1 mole of the oxide, and 0.1 mole of the secondary are too toxic for therapeutic use.

Experimental

-~- ...

b:nirle. Ai.rh P h o y w . . 247, :iii4 (19091. i i , 'Anrlree. Nri. . 36, 420 (1902) 1x1 T i f i r n e a u . H i r l i . soc. c h i ~ ~ i[ ..I 9,]$120: (1911). 101 F o r n e a u an:l T i f f e n e a u , C o m p l i u i i d . . 141, fifi2 (1906). ilii

STAXFORD UNIVERSITY, CALIF.

RECEIVED J U L Y 2, 1915

(10) The authors are indebted to Mr, A . J. Castro, who purified t h e products for analysis and analyzed them for chlorine.