Preparation of the 4-hydroxytryptamine scaffold via palladium-catalyzed cyclization: A practical and versatile synthesis of psilocin

Article abstract of DOI:10.1021/ol0341039

(Matrix presented) The 4-hydroxytryptamine scaffold of psilocin was successfully prepared via palladium-catalyzed cyclization of protected N-tert-butoxycarbonyl-2-iodo-3-methoxyaniline and an appropriately substituted silyl acetylene. Removal of the protecting groups afforded psilocin in good yield.

Full text of DOI:10.1021/ol0341039

ORGANIC
LETTERS
2003
Vol. 5, No. 6
921-923
Preparation of the 4-Hydroxytryptamine
Scaffold via Palladium-Catalyzed
Cyclization: A Practical and Versatile
Synthesis of Psilocin
,†
Nicholas Gathergood†,‡ and Peter J. Scammells*
Department of Medicinal Chemistry, Victorian College of Pharmacy,
Monash UniVersity, 381 Royal Parade, ParkVille, Victoria 3052, Australia, and
School of Biological and Chemical Sciences, Deakin UniVersity,
Geelong, Victoria 3217, Australia
peter.scammells@Vcp.monash.edu.au
Received January 20, 2003
ABSTRACT
The 4-hydroxytryptamine scaffold of psilocin was successfully prepared via palladium-catalyzed cyclization of protected N-tert-butoxycarbonyl-
2-iodo-3-methoxyaniline and an appropriately substituted silyl acetylene. Removal of the protecting groups afforded psilocin in good yield.
4-Substituted indoles are an important class of alkaloids that
exhibit a wide range of activity.1 Psilocybin 1 and its
metabolite, psilocin 2, are reported to enter the central
nervous system through the gastrointestinal tract and cause
powerful psychotomimetic effects.2 There have been con-
siderable studies into the effects of substitution on the
4-hydroxytryptamine scaffold.3
economical and efficient synthesis of psilocin for use as a
reference standard.
Methods for the preparation of indoles substituted in the
3 position can be split into two categories. Either the desired
indole core is formed (e.g., 4-hydroxyindole) and then
modified at the 3 position or the appropriate ortho-haloaniline
structure is coupled with a silylated alkyne, directly giving
an indole product substituted at the 3 position. Synthesis of
4-hydroxyindoles from indole via 4-iodoindoles using thal-
lium acetate has been reported by Somei et al.,4 and this
route has been used to prepare psilocin.5 A synthesis of
(2) (a) Stoll, A.; Troxler, F.; Peyer, J.; Hofmann, A. HelV. Chim. Acta
1955, 38, 1452. (b) Hofmann, A. Experientia 1958, 14, 107. (c) Hofmann,
A.; Heim, R.; Brack, A.; Kobel, H.; Frey, A.; Ott, H.; Petrzilka, T.; Troxler,
F. HelV. Chim. Acta 1959, 42, 1557. (d) Troxler, F.; Seemann, F.; Hofmann,
A. HelV. Chim. Acta 1959, 42, 2073. (e) Downing, D. F. Quart. ReV.
(London) 1962, 16, 133. (f) Hofmann, A. Bull. Narc. 1971, 23, 3. (g)
Brimblecombe, R. W.; Pinder, R. M. Hallucinogenic Agents; Wright-
Scientechnica: Bristol, UK, 1975; pp 106-108.
(3) (a) Repke, D. B.; Ferguson, W. J.; Bates, D. K. J. Heterocycl. Chem.
1981, 18, 175. (b) Repke, D. B.; Ferguson, W. J. J. Heterocycl. Chem.
1982, 19, 845. (c) Repke, D. B.; Ferguson, W. J.; Bates, D. K. J. Heterocycl.
Chem. 1977, 14, 71. (d) Repke, D. B.; Grotjahn, D. B.; Shulgin, A. T. J.
Med. Chem. 1985, 28, 892.
To facilitate the development of an improved methodology
for the analysis of psilocin, our aim was to develop an
† Monash University.
‡ Deakin University.
(1) (a) Brown, R. T.; Joule, J. A.; Sammes, P. G. ComprehensiVe Organic
Chemistry; Barton, D. H. R., Ollis, W. D., Eds.; Pergamon Press: Oxford,
1979; Vol. 4, p 411. (b) Kutney, J. P. Total Synthesis of Natural Products;
ApSimon, J., Ed.; Wiley-Interscience: New York, 1977; Vol. 3, p 273.
10.1021/ol0341039 CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/15/2003

psilocin and psilocybin from 4-benzyloxyindole was reported
by Nichols and Frescas;6 however, the cost of this starting
material is considerable.7 The 4-hydroxyindole ring structure
has also been formed in a two-step process by the palladium-
catalyzed cross-coupling of ortho-iodoanilines and (trimeth-
ylsilyl)acetylene, followed by a cyclization in the presence
of potassium tert-butoxide.8 This method, however, is not
amenable to the direct preparation of the desired indole
framework present in tryptamine. An example of a pal-
ladium-catalyzed cyclization reaction of ortho-vinylanilines
to yield indoles has also been reported.9
Scheme 1. Synthesis of Silyl Acetylene Precursorsa
a Reagents and conditions: (a) TsCl, NEt3, CH2Cl2; (b) HNMe2
(40 wt % solution in water), rt, 16 h or HNBn2, MeOH, reflux, 48
h; (c) n-BuLi, Et2O, -10 °C and then TMSCl. Yield of 5a ) 88%,
5b ) 56% (both from 4).
Palladium-catalyzed cyclization of iodo-aromatics with
unsaturated fragments to yield indole products substituted
at the 3 position has been reported.10 Ujjainwalla and Warner
describe the synthesis of 5-, 6-, and 7-azaindoles derivatives
via palladium-catalyzed heteroannulation of 4-(triethylsilyl)-
3-butyn-1-ol and aminopyridines (e.g., 2-amino-3-iodo-
pyridine).11 Recently, triethylsilylalkynes were reacted with
ortho-iodoanilines to give substituted tryptophan ana-
logues.12,13 Sakagami and Ogasawara14 reported the prepara-
tion of psilocin in six steps from N-tert-butoxycarbonyl-2-
iodo-3-methoxyaniline 3.
alkyne in good yield (Scheme 1). Compound 5b was
prepared analogously using N,N-dibenzylamine.
The key palladium-catalyzed cyclization step (Scheme 2)
was attempted under a variety of conditions, and the best
Scheme 2. Synthesis of Psilocina
We now report a short preparation of psilocin, avoiding
the use of thallium salts, from inexpensive starting materials
that we believe is convenient for synthetic and analytical
chemists. Our approach is a concise, convergent synthesis
of psilocin from N-tert-butoxycarbonyl-2-iodo-3-methoxy-
aniline 3 in three steps. The key step is the formation of the
indole core via a palladium-catalyzed cyclization. The two
fragments required for the cyclization are 3 and alkyne 5a.
Compound 3 was prepared from Boc-protected 3-methoxy-
aniline, via directed lithiation15 and iodination.16
The preparation of 5a from 3-butyn-1-ol 4 has been
previously reported;13 however, no experimental procedure
or characterization data was included in this patent. Tosyla-
tion, substitution with N,N-dimethylamine,17 and treatment
with n-butyllithium, trimethylsilyl chloride gave the required
a Reagents and conditions: (a) Pd(OAc)2 (0.2 equiv), PPh3 (0.4
equiv), NEt4Cl (1 equiv), i-Pr2EtN (3 equiv), DMF, 80 °C (yield
of 6a ) 69%, 6b ) 77%); (b) neat TFA, 25 °C, 3 h (yield of 7 )
58%); (c) BBr3, CH2Cl2, from -78 to 25 °C (yield of 2 ) 61%);
(d) 20% Pd(OH)2/C, 40 psi H2 (yield of 6c ) 83%).
(4) (a) Somei, M.; Yamada, F.; Kunimoto, M.; Kaneko, C. Heterocycles
1984, 22, 797. (b) Yamada, F.; Tamura, M.; Hasegawa, A.; Somei, M.
Chem. Pharm. Bull. 2002, 50, 92.
(5) Yamada, F.; Tamura, M.; Somei, M. Heterocycles 1998, 49, 451.
(6) Nichols, D. E.; Frescas, S. Synthesis 1999, 6, 935.
(7) Aldrich: 4-benzyloxyindole (1 g), $A225; 4-hydroxyindole (1 g),
$A288 (1/2003). A patent describing an efficient synthesis of 4-hydroxy-
indoles from cyclohexane-1,3-dione, where the key step is the reaction of
oxochromancarboxylic acid derivatives with ammonia in methanol in an
autoclave, has been filed. The author notes that this may effect the price of
4-hydroxyindoles in the future. Matsuura, T. (Nippon Zeon Co., Ltd., Japan)
Jpn. Kokai Tokkyo Koho JP 2000044555, 2000.
(8) Kondo, Y.; Kojima, S.; Sakamoto, T. J. Org. Chem. 1997, 62, 6507.
(9) Krolski, M. E.; Renaldo, A. F.; Rudisill, D. E.; Stille, J. K. J. Org.
Chem. 1988, 53, 1170.
results were obtained using Pd(OAc)2, triphenylphosphine,
tetraethylammonium chloride, and N,N-diisopropylethyl-
amine in DMF at 80 °C for 48 h.18 When tri-2-furylphosphine
was used in place of triphenylphosphine, a significantly lower
yield of the desired indole was obtained (32%). Although
LiCl has been reported to improve the regioselectivity,
reproducibility, and yield of such cyclizations,11 the use of
LiCl and Na2CO3 in this case gave slightly inferior results.
(10) Larock, R. C.; Yum, E. K. J. Am. Chem. Soc. 1991, 113, 6689.
(11) Ujjainwalla, F.; Warner, D. Tetrahedron Lett. 1998, 39, 5355.
(12) Ma, C.; Liu X.; Li, X.; Flippen-Anderson, J.; Yu, S.; Cook, J. M.
J. Org. Chem. 2001, 66, 4525.
(13) Smith, A. L. Brit. UK Pat. Appl. GB2328941, 1999.
(14) Sakagami H.; Ogasawara, K. Heterocycles 1999, 51, 1131.
(15) Snieckus V. Chem. ReV. 1990, 90, 879.
(16) Preparation of 14 g of 3 was performed in our laboratories using
standard glassware.
(17) This substitution reaction was accomplished successfully using an
aqueous solution of N,N-dimethylamine with no formation of 3-butyn-1-ol
detected by NMR. Previous methods have used N,N-dimethylamine as a
gas or dissolved in an organic solvent; both of these sources of N,N-
dimethylamine are considerably more expensive or inconvenient to use.
(18) A dry flask was charged with 1-tert-butoxycarbonyl-2-iodo-3-
methoxyaniline (3.21 g, 9.16 mmol, 1 equiv), 4-(trimethylsilyl)-3-butyn-
1-dimethylamine (3.10 g, 18.3 mmol, 2 equiv), palladium(II) acetate (420
mg, 1.84 mmol, 0.2 equiv), triphenylphosphine (960 mg, 3.68 mmol, 0.4
equiv), tetraethylammonium chloride (1.52 g, 9.12 mmol, 1 equiv),
diisopropylethylamine (3.54 g, 4.8 mL, 27.4 mmol, 3 equiv), and DMF
(65 mL) under a nitrogen flush and heated to 80 °C for 48 h. After the
mixture was cooled, DMF and volatiles were removed by rotary evaporation
and then ethyl acetate (50 mL) and water (50 mL) were added to the residue.
The aqueous phase was extracted with ethyl acetate (3 × 25 mL); the organic
phase was washed with 5% NaHCO3 (25 mL) and brine (25 mL), and
solvents were removed by rotary evaporation. The crude was purified by
column chromatography (SiO2, 90:10:1 CHCl3/CH3OH/NH4OH) to give
the title compound as a light brown oil in 69% yield (2.45 g, 6.3 mmol).
922
Org. Lett., Vol. 5, No. 6, 2003

In all cases, several byproducts were present in the crude
reaction mixture (apparent by TLC and 1H NMR) and
column chromatography was required to obtain pure 6a. One
plausible route for the formation of these byproducts is the
cyclization of the dimethylamino group on the activated
vinylic-palladium bond. Although 6a was stable to purifica-
tion by column chromatography, the byproducts decomposed,
making their identification difficult.
has been prepared and that the extra steric bulk of the benzyl
groups on the nitrogen has not inverted the regiochemistry
of the cyclization. Psilocin prepared by the route shown in
Scheme 2 exhibited NMR data that was in agreement with
the literature,14 thus proving the regiochemistry of the
palladium-catalyzed cyclization of 5a and 3.
Modification of the alkyl groups in serotonin and related
compounds to alter the activity of these compounds has been
an active field of research.3 Compound 6b is also a versatile
intermediate for the preparation of analogues of psilocin with
modified amine substituents. The N-benzyl groups of 6b were
removed by catalytic hydrogenation to give 6c in good yield.
Compound 6c is amenable to conversion to psilocin ana-
logues with modified side chains. For example, reductive
alkylation of the terminal amino group of 4-benzyloxy-
tryptamine has been successfully completed by Yamada et
al.5
In conclusion, the carbon framework for psilocin can be
formed by the palladium-catalyzed cyclization of 3 with 5a.
Removal of the protecting groups leads to the target in good
yield. Using 5b in the cyclization step increased the yield
and generated a cleaner reaction. The benzyl-protected amino
group of 6b can then be selectively deprotected. This leads
to the useful intermediate 6c, which can be reductively
alkylated to a series of secondary and tertiary amines. This
approach has the flexibility to allow a short synthesis of
amino analogues by incorporation of the desired amine via
the alkyne fragment or by modification of a late-stage
intermediate. New methodology for analysis of psilocin and
its analogues will be reported in due course.
To complete the synthesis of psilocin, the Boc and
trimethylsilyl groups of 6a were cleaved by treatment with
neat TFA to afford 7 in good yield. O-Demethylation using
boron tribromide13,19 yielded psilocin 2.
To further explore the versatility of the palladium cycliza-
tion and increase the degree of derivatization of the route
presented, we studied the effect of preparing alkynes with
more sterically hindered amino fragments. Compound 6b was
prepared in the same manner as 6a. We were pleased to find
that the Pd cyclization of 3 with 5b gave a clean reaction to
6b in 77% yield. This suggests that the undesired cyclization
of the terminal amine is inhibited by the steric bulk of the
two benzyl groups.
Confirmation of the regiochemistry of the palladium-
catalyzed cyclization between the dibenzylamino-substituted
alkyne 5b and 3 was established by X-ray crystallography.
Figure 1 (below) clearly shows that the tryptamine scaffold
Acknowledgment. The authors thank Mr. Gary Fallon
for the crystal structure of compound 6b and the Australian
Research Council for financial support.
Supporting Information Available: Spectroscopic data
for psilocin 2 and X-ray data for the indole 6b. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL0341039
(19) (a) McOmie, J. W. W.; West, D. E. Org. Synth., Collect. Vol. V
1973, 412. (b) Peat, A. J.; Buchwald, S. L. J. Am. Chem. Soc. 1996, 118,
1028.
Figure 1. ORTEP of indole 6b (hydrogen omitted for clarity).
Org. Lett., Vol. 5, No. 6, 2003
923