The Active Principles Of Cannabis Indica Resin

Jacob Bell

New Member
BY THOMAS SPENCE WORK, FRANZ BERGEL AND
ALEXANDER ROBERTUS TODD
From the Biochemical Department, Lister Institute, London
(Received 24 November 1938)

THE resinous exudate of the female flowers of Cannabis indica (C. sativa) forms
the essential constituent of the drug variously known as hashish, bhang, charas,
ganja and marihuana according to the country of origin and mode of preparation.
Extracts of the flowers have been used in European medicine to some
extent but were found to be so variable in activity that C. indica has been
removed from the British Pharmacopoeia. The plant is, however, well known
through its use as a drug in oriental countries and recently, pnder the name of
" marihuana ", its use has assumed dangerous proportions in America. Cannabis
resin has been the subject of many investigations in the past but much of the
earlier work is contradictory. The active principle is contained in a high-boiling
resin and is not an alkaloid [Smith, 1857]. The active fraction isolated from the
crude resin by Wood Spivey & Easterfield [1896] and given by them the name
cannabinol was later found by the same workers [1899] to be a mixture, and the
name cannabinol was transferred to that portion of the active resin yielding a
crystalline acetate, M.P. 75°. A good deal of confusion was introduced by later
workers, who, although failing to obtain a crystalline acetate, nevertheless
applied the name cannabinol to their products. The situation was largely
clarified by Cahn [1931], who confirmed the observations of Wood Spivey &
Easterfield and established for cannabinol the formula C21H2,O2. The constitution
of cannabinol has been investigated by Cahn [1930-33] and by Bergel
[1932], and for it the former proposed structure (I) in which only the positions
of the hydroxyl and n-amyl groups are to be regarded as uncertain.

In his last communication Cahn states that " cannabinol is not the pharmacologically
active principle of Cannabis indica" [1933]. It seemed desirable
that further investigations on the resin should be made and we have taken up
the study of cannabis resin with a view to isolating the substance or substances
responsible for its pharmacological action.
The starting point in these investigations was material which corresponded
to the high-boiling resin (B.P. 265°/20 mm.) first described by Wood Spivey &
Easterfield [1896]. It was found that cannabinol can be removed almost quantitatively
as its crystalline p-nitrobenzoate (amounting to 25 % of the whole) on
p-nitrobenzoylating the resin. On hydrolysis the crystalline ester yields cannabinol
as a colourless oil, which gives the above-mentioned crystalline acetate
(M.P. 750) on acetylation. The non-crystalline portion of the p-nitrobenzoylated
resin gave on hydrolysis a colourless oil from which no crystalline acetate could
be obtained.
Active hashish preparations induce a characteristic cataleptic condition in
dogs, but this effect, described first by Fraenkel [1903] is not readily made the
basis of quantitative assay. Gayer [1928] showed that in various animals-e.g.
cats, rabbits, mice-hashish preparations induce corneal anaesthesia and that
this effect is characteristic of active fractions of the resin. This Gayer test has
been developed by Marx & Eckhardt [1933] using rabbits and has been employed
throughout the work described in this paper. The pure cannabinol prepared
from the crystalline p-nitrobenzoate was found to be highly toxic when injected
intravenously into rabbits but, unlike the original resin, did not produce corneal
anaesthesia. The cannabinol-free resin on the other hand was much less toxic
and always induced corneal anaesthesia in rabbits. This distinction is further
emphasized by the fact that an acetone solution of cannabinol was found to be
non-toxic after standing for 3 days in contact with air while only slight loss of
activity occurred on similar treatment of the cannabinol-free resin.
Attempts to fractionate further the cannabinol-free resin by distillation or
crystallization of derivatives failed completely but chromatographic analysis
gave promising results. Using activated aluminium oxide as adsorbent it was
possible to obtain fairly readily an oil having considerably greater activity in
the Gayer test than the starting material. Even better results were obtained by
submitting the oily p-nitrobenzoate mixture left after separating the cannabinol
p-nitrobenzoate to chromatographic analysis on activated aluminium oxide.
By this procedure remaining traces of cannabinol were removed and an oil was
obtained yielding on hydrolysis a product giving a positive result in the Gayer
test at a dose of 0'25 mg. per kg. body weight; this material has low toxicity as
compared with cannabinol and possesses none of the convulsant action of the
latter substance.
Further investigation of this highly active material is in progress and the
results will be reported later. We have not as yet obtained any crystalline
derivative from the most active preparation and suspect that it does not yet
represent the homogeneous active principle.
EXPERIMENTAL
Starting material. The starting material was the resin (B.P. 185-190o/0.6 mm.)
obtained on working up hashish of Indian origin in the manner described by
Bergel [1930]. The resin was nearly colourless when freshly distilled. It corresponds
to the "crude cannabinol" of Wood Spivey & Easterfield and possessed
the characteristic pharmacological properties of the original extract.
Isolation of cannabinol. The above resin (25 g.) was dissolved in pyridine
(110 ml.) and p-nitrobenzoyl chloride (36 g.) added. The mixture was refluxed
for 4 hr., then poured on a mixture of ice and sufficient H2SO4 to make the
resulting suspension acid to Congo red. The precipitate was collected, washed
with water, dried and refluxed with light petroleum (750 ml., B.P. 80-100°) for
1 hr. and filtered hot, the filter residue being treated in the same way with a
further quantity of light petroleum (250 ml.). The combined ifitrates were
washed with aqueous Na2C03, dried and concentrated to about 150 ml. On
standing cannabinol p-nitrobenzoate separated. Recrystallized first from alcohol then from light petroleum (B.P. 80-100°) it formed pale yellow needles M.P. 1600
(yield, ca. 65t g.). (Found: C, 73'0; H, 6-2; N, 3-2 %. Cg8H2905N requires C, 73X2;
H, 6X3; N, 3'2 %.) Since cannabinol p-nitrobenzoate is very sparingly soluble in
methyl alcohol the oil left on evaporating the original light petroleum mother
liquors may be largely freed of cannabinol by fractionation with this solvent.
Cannabinol p-aminobenzoate. Cannabinol p-nitrobenzoate (1 g.) dissolved in
alcohol (80 ml.) was hydrogenated using a platinum oxide catalyst. Absorption
of hydrogen ceased when 200 ml. had been absorbed (theoretical 190 ml.). The
resulting p-aminobenzoate crystallized from methyl alcohol in colourless needles
M.P. 149-150°. (Found: C, 78-2; H, 7-4%. C2H31O3N requires C, 78-3; H,
7.2 %.)
Cannabinol. Cannabinol p-nitrobenzoate hydrolysed by refluxing with
methyl alcoholic KOH (5 %) during 1 hr. gave cannabinol as an almost colourless
oil which, with acetic anhydride-pyridine gave in quantitative yield a
crystalline acetate M.P. 750 not depressed on admixture with a sample of cannabinol
acetate kindly supplied by Dr R. S. Cahn.
Pharmacological tets
The Gayer test was carried out on rabbits as described by Marx & Eckhardt
[1933], the substances being injected in acetone solution (0.5% wt./vol.).
(a) Cannabinol. Pure cannabinol prepared by hydrolysing the p-nitrobenzoate
and subsequent distillation in a high vacuum was used. In doses less
than 2 mg. per kg. body weight the material had no visible effect and the
corneal reflex remained normal. At any higher dosage the following sequence
of events was observed. For about 1 min. the animal behaved normally, but at
the end of this time it lay down and in a few seconds rolled over on its side and
became rigid, the corneal reflex remaining normal. In any time from a few
seconds to 1 min. later, depending on the size of the dose, the rabbit went into
violent convulsions terminating in death within about 30 sec.
(b) Material from non-crystalline p-nitrobenzoates. After separation of the
crystalline p-nitrobenzoate the mixture of oily esters from the resin was hydrolysed
and the product distilled in a high vacuum. The nearly colourless oil
obtained had no effect when injected into rabbits in doses less than 1 mg. per
kg. body weight. In a dose of 1 mg. per kg. body weight the drug had no effect
for about 3 min., after which time the animal's head began to nod gently and it
subsided to its normal sleeping position and remained so. During this period the
corneal reflex slowly disappeared until no response could be elicited. The animal
could be roused for a few seconds by violent shaking but when left undisturbed
quickly relapsed. In doses up to 5 mg. per kg. body weight exactly the same
effect was observed, the period of sleep or stupor extending from 30 min. up to
6 or 7 hr. after which the rabbit recovered completely. In doses of 5 mg. or
more the animal, although giving a positive Gayer test, died with convulsions
within the first 2 hr. after injection. The death after convulsions was presumably
due to the presence of some cannabinol in the injected material (cf. below).
Chromatographic analysis of hydrolysate of non-crystalline p-nitrobenzoates
A sample of the oil (2 g.) prepared by hydrolysis of the non-crystalline
fraction of the p-nitrobenzoylated resin and subsequent distillation in a high
vacuum was dissolved in light petroleum (300 ml.; B.P. 60-80°) and allowed to
percolate through a column of activated aluminium oxide (Merck), the chromatogram
being developed first with light petroleum (750 ml.; B.P. 60-80°), then
with a mixture (1 litre) of equal parts of light petroleum (B.P. 60-80') and ether.
When the column was viewed in ultraviolet light four distinct bands were
visible; from the top downwards these were: (1) yellow 6 cm., (2) colourless
8 cm., (3) yellow 8 cm., (4) blue fluorescent 6 cm. On elution with a mixture of
ether and methyl alcohol (4: 1) the oils from sections (1), (2) and (3) were
found to be inactive when tested on rabbits while that from section (4) gave a
positive Gayer test in a dose of 3 mg. per kg. body weight.
The oil from section (4) was therefore combined with that obtained by
evaporating the filtrate from the chromatogram, dissolved in light petroleum
(B.P. 60-80o) containing 5 % ether and re-adsorbed on a column (2 x 30 cm.) of
activated aluminium oxide (Merck), the chromatogram being developed with the
same solvent mixture. From the top downwards the column showed in ultraviolet
light the following bands: (1) purple 3 cm., (2) yellowish blue 10 cm.,
(3) deep blue 10 cm., (4) strongly blue fluorescent 2 cm., (5) yellow 3 cm. Tests
on rabbits showed that the oil from section (4) (50 mg.) was lethal in a dose of
2*5 mg. per kg. body weight and had the typical action of cannabinol. The oil
from section (2) (60 mg.) was active in the Gayer test in a dose of 1 mg. per kg.
body weight and when injected in a dose of 5 mg. per kg. body weight the
animal survived for 10 hr. without any trace of muscular rigidity and died in
sleep without any convulsions. The oil from section (3), like the starting material,
possessed both convulsant and sleep-producing properties.
Chromatographic analysi8 of non-crystalline p-nitrobenzoates. The oily residue
(8 g.) left on removing the crystalline cannabinol derivative from the p-nitrobenzoylation
product of the distilled resin, was dissolved in light petroleum
(B.P. 40-60') and subjected to adsorption on a column (5 x 45 cm.) of activated
aluminium oxide (Merck) previously washed with a solution of phenol in light
petroleum to reduce alkalinity. After developing with light petroleum (4 litres)
the chromatogram showed six distinct bands when viewed in ultra-violet light.
Each of these was separately eluted, hydrolysed, distilled and tested, the filtrate
from the column being evaporated and the residue similarly treated and tested.
The following table shows the results obtained, the bands in the chromatogram
being numbered from the top downwards.
Wt. of
Length eluate Millon Gayer M.L.D.
Colour cm. g. test test mg.
1 Yellow 3 03 - - -
2 Colourless 15 1-2 - - -
3 Bright blue 3 0-25 + - 6
4 Colourless 10 2-0 + - 5.5
5 Yellow 5 0.9 - - -
6 Colourless 10 1-2 + + (2 mg.) 9
7 Colourless Filtrate 1-5 + +(0*25 mg.) 5
The biological test results were reproducible in different animals, there being
very little variation either in the minimum active dose or in the minimum lethal
dose (M.L.D.). Sections (3) and (4) possessed the typical convulsion-producing
properties of cannabinol while sections (6) and (7) had no such action. It is
clear then from the table that the remaining traces of cannabinol were concentrated
in sections (3) and (4), while the material producing corneal anaesthesia
was concentrated in sections (6) and (7). The oil from sections (6) and (7) gave,
like cannabinol, a precipitate with Millon's reagent.
Relative stabilities of cannabinol and resin from fraction 7 (above). A solution
(0.5 %) of cannabinol in acetone was exposed to the air for 3 days. When tested
on rabbits at the end of this time it was found to be non-toxic. A solution of the resin from fraction 7 (above) after standing for 6 months under the same conditions
retained about 25% of its activity in the Gayer test.
SU1MMY
p-Nitrobenzoylation of the high-boiling pharmacologically active resin from
the female flowers of Cannabi8 indica yields crystalline cannabinol p-nitrobenzoate
and a mixture of resinous esters. Cannabinol is highly toxic and gives
a completely negative reaction in the Gayer hashish test on rabbits, while the
hydrolysis product of the resinous esters gives a strong positive reaction and is
less toxic than cannabinol. The material giving a positive Gayer test has been
fractionated by adsorption methods and a product obtained showing a positive
Gayer test in rabbits in a dose of 025 mg. per kg. body weight.
One of the authors (F. B.) desires to record his appreciation of the hospitality
accorded him in the University College and the Laboratory of the Public
Analyst, Colombo, and to thank the Director of the Royal Botanic Gardens,
Peradeniya, Ceylon, for his generous assistance.
REFERENCES
Bergel (1930). Liebig8 Ann. 482, 57.
- (1932). Liebig8 Ann. 493, 250.
Cahn (1930). J. chem. Soc. p. 986.
(1931). J. chem. Soc. p. 630.
(1932). J. chem. Soc. p. 1342.
(1933). J. chem. Soc. p. 1400.
Fraenkel (1903). Arch. exp. Path. Pharm. 49, 266.
Gayer (1928). Arch. exp. Path. Pharm. 129, 312.
Marx & Eckhardt (1933). Arch. exp. Path. Pharm. 170, 395.
Smith (1857). J. chem. Soc. 21, 47.
Wood Spivey & Easterfield (1896). J. chem. Soc. 69, 539.
- - (1899). J. chem. Soc. 75, 20.


Source: The Active Principles Of Cannabis Indica Resin
 
Back
Top Bottom