Multicomponent synthesis of 6 H-dibenzo[b,d]pyran-6-ones and a total synthesis of cannabinol

Article abstract of DOI:10.1021/ol2030636

A multicomponent domino reaction that affords 6H-dibenzo[b,d]pyran-6-ones is reported. The overall transformation consists of six reactions: Knoevenagel condensation, transesterification, enamine formation, an inverse electron demand Diels-Alder (IEDDA) reaction, 1,2-elimination, and transfer hydrogenation. Both the diene and dienophile for the key inverse electron demand Diels-Alder (IEDDA) step are generated in situ by secondary amine-mediated processes. In most cases, the yields (10-79%) are considerably better than those obtained using a stepwise process. This methodology is employed in a concise total synthesis of cannabinol.

Full text of DOI:10.1021/ol2030636

ORGANIC
LETTERS
2012
Vol. 14, No. 1
310–313
Multicomponent Synthesis of
6H-Dibenzo[b,d]pyran-6-ones and a Total
Synthesis of Cannabinol
Penchal Reddy Nandaluru and Graham J. Bodwell*
Department of Chemistry, Memorial University, St. John’s, NL, Canada, A1B 3X7
gbodwell@mun.ca
Received November 14, 2011
ABSTRACT
A multicomponent domino reaction that affords 6H-dibenzo[b,d]pyran-6-ones is reported. The overall transformation consists of six reactions:
Knoevenagel condensation, transesterification, enamine formation, an inverse electron demand DielsÀAlder (IEDDA) reaction, 1,2-elimination,
and transfer hydrogenation. Both the diene and dienophile for the key inverse electron demand DielsÀAlder (IEDDA) step are generated in situ by
secondary amine-mediated processes. In most cases, the yields (10À79%) are considerably better than those obtained using a stepwise process.
This methodology is employed in a concise total synthesis of cannabinol.
Multicomponent reactions (MCR) are highly valuable
transformations due to their ability to incorporate three or
more substrates into a single target in one operation.1 MCRs
typically achieve a substantial increase in molecular complex-
ity and offer opportunities for diversity-oritented synthesis.
They have proven to be valuable in drug discovery,2 as well as
in the total synthesis of natural products.3
Cannabinoids form a class of ∼70 natural products
that have been isolated from the plant Cannabis sativa.4
Cannabinol (1), Δ9-tetrahydrocannabinol (THC) (2), and
cannabinodiol (3) are prominent members of this family
(Figure 1). G-protein coupled cellular receptors, CB1 and
CB2, are the targets of the cannabinoids.5 While the CB1
receptor is widely present in the central nervous system
(CNS), especially the brain, the CB2 receptor is less widely
distributed. The CB2 receptor is present in organs and
tissues of immune-related systems, such as the spleen,
thymus, bone marrow, and B lymphocytes. Hence, can-
nabinoid agonists that selectively bind to one of the
receptors are desirable in that side effects associated with
the expression of the other receptor would be minimized.6
Several strategies for the synthesis of cannabinol (1)
and its derivatives have been reported. These can be
classified according to the key steps involved: (1) aroma-
tization of tetrahydrocannabinols,7 (2) a nucleophilic
aromatic substitution/lactonization/Grignard reaction
sequence,8 (3) a Suzuki coupling/lactonization/Grignard
reaction sequence,9 (4) Ru-catalyzed cyclotrimeriza-
tion followed by a Grignard reaction.10 The latter three
(6) (a) Marriot, K. S.; Huffman, J. W. Curr. Top. Med. Chem. 2008, 8,
187. (b) Pertwee, R. G. Pharmacol. Ther. 1997, 74, 129. (c) Raitio, K. H.;
Salo, O. M.; Nevalainen, T.; Poso, A.; Jarvinen, T. Curr. Med. Chem.
2005, 12, 1217. (d) Huffman, J. W. Curr. Pharm. Des. 2000, 6, 1323. (e)
Malan, T. P., Jr.; Ibrahim, M. M.; Deng, H.; Liu, Q.; Mata, H. P.;
Vanderah, T.; Porreca, F.; Makriannis, A. Pain 2001, 93, 239.
(7) (a) Adams, R.; Baker, B. R.; Wearn, R. B. J. Am. Chem. Soc.
1940, 62, 2204. (b) Ghosh, R.; Todd, A. R.; Wright, D. C. J. Chem. Soc.
1941, 137. (c) Meltzer, P. C.; Dalzell, H. C.; Razdan, R. K. Synthesis
1981, 985. (d) Rhee, M.-H.; Vogel, Z.; Barg, J.; Bayewitch, M.; Levy, R.;
Hanus, L.; Breuer, A.; Mechoulam, R. J. Med. Chem. 1997, 40, 3228.
(e) Mahadevan, A.; Siegel, C.; Martin, B. R.; Abood, M. E.; Beletskaya,
I.; Razdan, R. K. J. Med. Chem. 2000, 43, 3778.
(1) Reviews: (a) Tejedor, D.; Garcia-Tellado, F. Chem. Soc. Rev.
ꢀ
2007, 36, 484. (b) Zhu, J., Bienayme, H., Eds. Multicomponent Reactions;
Wiley-VCH: Weinheim, 2005. (c) Orru, R. V. A.; Hulme, C.; Oddon, G.;
€
Schmitt, P. Chem.;Eur. J. 2000, 6, 3321. (d) Domling, A.; Ugi, I. Angew.
Chem., Int. Ed. 2000, 39, 3168.
(2) (a) Blackwell, H. E. Curr. Opin. Chem. Biol. 2006, 10, 203.
€
€
(b) Domling, A. Chem. Rev. 2006, 106, 17. (c) Brauer, S.; Almstetter,
M.;Antuch, W.;Behnke,D.;Taube, R.;Furer, P.;Hess, S. J. Comb. Chem.
€
2005, 7, 218. (d) Ugi, I.; Werner, B.; Domling, A. Molecules 2003, 8, 53.
ꢀ
(3) Review: Toure, B. B.; Hall, D. G. Chem. Rev. 2009, 109, 4439.
(4) ElSohly, M. A.; Slade, D. Life Sci. 2005, 78, 539.
(5) (a) Matsuda, L. A.; Lolait, S. J.; Brownstein, M. J.; Young, A. C.;
Bonner, T. I. Nature 1990, 365, 61. (b)Munro, S.;Thomas, K.L;Abu-Shaar,
M. Nature 1993, 365, 61.
ꢀ
(8) (a) Novak, J.; Salemink, C. A. Tetrahedron Lett. 1982, 23, 253.
(b) Hattori, T.; Suzuki, T.; Hayashizaka, N.; Koike, N.; Miyano, S. Bull.
Chem. Soc. Jpn. 1993, 66, 3034.
r
10.1021/ol2030636
Published on Web 12/15/2011
2011 American Chemical Society

The observation that a 2° amine plays a catalytic role in
both the formation of the electron-deficient diene
(Knoevenagel condensation) and the electron-rich dieno-
phile (enamine formation and subsequent elimination)
prompted us to investigate the possibility of generating
both IEDDA partners in situ.15 If successful, this would
be a multicomponent domino reaction consisting of six
steps: Knoevenagel reaction, transesterification, enamine
formation, IEDDA reaction, 1,2-elimination, and trans-
fer hydrogenation. The anticipated dual catalytic func-
tion of the 2° amine presented an opportunity to perform
auto-tandem organocatalysis.16 The existence of prece-
dence for the simultaneous in situ generation of both the
diene and dienophile in the normal DielsÀAlder reaction15
and the application of the IEDDA reaction in MCRs17
augured well for the success of the proposed MCR.
Figure 1. Prominent cannabinoids 1À3 and 6H-dibenzo[b,d]-
pyran-6-one (4).
A mixture of salicylaldehyde (5), dimethyl glutacon-
ate (6), and cyclopentanone (10) with morpholine as the
base and toluene as the solvent was chosen for initial
experiments. DBP 11 was obtained from the outset, and
through variation of the relative amounts of the reactants
and base, it was found that a maximum yield of 50% was
obtained when a 1:2:5:2.5 ratio of 5/6/10/morpholine was
used. While holding this ratio constant, the solvent and
base were varied (Table 1), and a maximum yield of 69%
was obtained using pyrrolidine as the base and 1,4-dioxane
as the solvent (Table 1, entry 10). Only with pyrrolidine as
the base was any reaction at rt observed (tlc analysis).
Although progress at rt was minimal, slightly better yields
were obtained when reactions were first stirred at rt for 2 h
before heating at reflux. Reactions conducted in 1,4-
dioxane were somewhat slower than those in other solvents,
so they were heated for 24 h instead of 12 h. Conditions
have not yet been identified, under which substoichio-
metric amounts of base afford competitive yields of 11.
categories all involve the intermediacy of a derivative of
6H-dibenzo[b,d]pyranone (DBP) (4).
In connection with our ongoing studies of the inverse elec-
tron demand DielsÀAlder (IEDDA) reaction,11 our group
has developed an IEDDA-based route to DBPs. In its orig-
inal form, an electron-deficient diene such as 7(the product of
a reaction between salicylaldehyde (5) and dimethyl glutaco-
nate (6)) was reacted with an enamine, e.g. 8, to afford the
corresponding DBP, e.g. 11 (Scheme 1).12 Subsequently, it
was found that an electron-rich dienophile (the enamine)
could be generated in situ from a secondary amine and a
ketone, e.g. 9 and 10.13 The formation of DBP 11 was
explained by a domino sequence consisting of an IEDDA reac-
tion, a 1,2-elimination of the secondary amine, and a dehy-
drogenation (most likely a transfer hydrogenation to a hydro-
gen acceptor, e.g., the enamine). This methodology was applied
to the synthesis of a metabolite of ellagic acid, urolithin M7.14
Scheme 1. Stepwise IEDDA-Based Approaches to DBPs
Table 1. Optimization of the MCR
isolated yield
entry
base
solvent
toluene
of 11 (%)
1
morpholine
piperidine
pyrrolidinea
morpholine
L-proline
50
53
60
54
43
57
56
49
52
69
50
45
58
2
toluene
3
toluene
4
ethanol
5
ethanol
6
piperidine
pyrrolidinea
morpholine
piperidine
pyrrolidinea
pyrrolidinea
pyrrolidinea
pyrrolidinea
ethanol
7
ethanol
(9) Khanolkar, A. D.; Lu, D.; Ibrahim, M.; Duclos, R. I., Jr.; Thakur,
G. A.;Malan, T. P., Jr.;Porreca, F.;Veerappan, V.; Tian, X.;Parrish, D. A.;
Papahatjis, D. P.; Makriyannis, A. J. Med. Chem. 2007, 50, 6493.
(10) Teska, J. A.; Deiters, A. Org. Lett. 2008, 10, 2195.
8
1,4-dioxaneb
1,4-dioxaneb
1,4-dioxaneb
chloroform
tetrahydrofuran
acetonitrile
9
10
11
12
13
(11) (a) Bodwell, G. J.; Z, Pi. Tetrahedron Lett. 1997, 38, 309.
(b) Bodwell, G. J.; Hawco, K. M.; da Silva, R. P. Synlett 2003, 179.
(c) Bodwell, G. J.; Hawco, K. M.; Satou, T. Synlett 2003, 879. (d) Dang,
A.-T.; Miller, D. O.; Dawe, L. N.; Bodwell, G. J. Org. Lett. 2008, 10, 233.
(12) Bodwell, G. J.; Pi, Z.; Pottie, I. R. Synlett 1999, 477.
a The reaction mixture was stirred at room temperature for 2 h prior
to heating at reflux. b The reaction mixture was heated at reflux for 24 h.
(13) Pottie, I. R.; Nandaluru, P. R.; Benoit, W. L.; Miller, D. O.;
Dawe, L. N.; Bodwell, G. J. J. Org. Chem. 2011, 76, 9015.
(14) Pottie, I. R.; Nandaluru, P. R.; Bodwell, G. J. Synlett 2011, 2245.
Org. Lett., Vol. 14, No. 1, 2012
311

Using the best conditions for the synthesis of 11, a series
of salicylaldehydes18 was reacted with dimethyl glutaco-
nate (6) and cyclopentanone (10) to afford a set of A-ring
substituted DBPs, most of which had been previously
synthesized using a stepwise approach (Table 2).12,13 The
yields ranged from 0% to 79% and, where comparisons
could be made, were superior (by 1À44%, Table S1) to
those obtained using stepwise syntheses.
entries 4 and 8À11). However, the drop in yield only became
drastic when a nitro group was present. This is presumably
due to the preferential formation of an isomeric 2H-chro-
mene over the desired nitrodiene.13
Table 3. Synthesis of C-Ring Substituted DBPs Using the MCR
Table 2. Synthesis of A-Ring Substituted DBPs Using the MCR
salicyl-
6H-dibenzo-
isolated
entry
aldehyde
substituent
[b,d]pyran-6-one
yield (%)
1
5
none
11
13
15
17
19
21
23
25
27
29
31
69
57
64
79
0
2
12
14
16
18
20
22
24
26
28
30
R1 = OMe
R2 = OMe
R3 = OMe
R4 = OMe
R1 = Me
R2 = Me
R3 = Me
R3 = Br
3
4
5
6
48
62
68
67
51
10
7
8
9
10
11
R3 = CO2Me
R3 = NO2
Only 6-methoxysalicylaldehyde (18) failed to afford
any of the desired DBP (Table 2, entry 5). Excluding
cyclopentanone (10) from the reaction mixture led to the
formation of the corresponding methoxydiene (cf. 7),19
and it was unreactive toward in situ generated enamine 8.
Presumably, steric hindrance at the transition state of the
cycloaddition inhibits the reaction. The other methoxy-
substituted salicylaldehydes (12, 14, 16) and the corre-
sponding methyl-substituted salicylaldehydes (20, 22, 24)
reacted smoothly to afford the respective DHPs (48À79%).
In both series, the yield for the 5-substituted system was
the best, followed by the 4- and 3-isomers (Table 2, entries
2À4 and 6À8). For the various 5-substituted salicylalde-
hydes (16, 24, 26, 28, 30), the yields were good until the
substituent became strongly electron-withdrawing (Table 2,
a 3 equiv of ketone were used instead of 5.
The ability of the MCR to generate C-ring-substituted
DBPs was then probed by conducting it with a series of
ketones, several of which had previously been used in
stepwise DBP syntheses (Table 3).12,13 Methyl ketones
(32, 34, 36, 38) reacted to afford the corresponding
9-substituted DBPs 33 (71%), 35 (36%), 37 (45%), and
39 (39%), respectively (Table 3, entries 1À4). In the case
of butanone (38), nonaromatized byproduct 40 was obtained
in 12% yield. This compound arises from IEDDA reaction of
diene 7 with the more highly substituted enamine derived
from 38. As previously observed for systems bearing a
substituent (i.e., one that is not part of a e5-membered fused
ring) at the 10 position of the DBP skeleton, dehydrogenation
did not occur.13 However, reaction of 40 with DDQ afforded
the corresponding DBP 50 in 85% yield (Scheme 2).
(15) Ramachary, D. B.; Barbas, C. F., III. Chem.;Eur. J. 2004, 10,
5323.
(16) Shindoh, N.; Takemoto, Y.; Takasu, K. Chem.;Eur. J. 2009,
15, 12168.
(17) (a) Kudale, A. A.; Kendall, J.; Miller, D. O.; Collins, J. L.;
Bodwell, G. J. J. Org. Chem. 2008, 73, 8437. (b) Kudale, A. A.; Miller,
D. O.; Dawe, L. N.; Bodwell, G. J. Org. Biomol. Chem. 2011, 9, 7196.
(18) Salicylaldehydes 20, 22, and 24 were synthesized using Skattebøl
formylation: (a) Hofsløkken, N. U.; Skattebøl, L. Acta Chem. Scand.
1999, 53, 258. Salicylaldehyde 28 was synthesized using Duff formyla-
tion: (b) Duff, C. J. J. Chem. Soc. 1941, 547. (c) Lindoy, L. F.; Meehan,
G. V.; Svenstrup, N. Synthesis 1998, 1029. The remaining salicylalde-
hydes were purchased.
In line with the stepwise DBP syntheses,12,13 small
cyclic ketones (e5-membered) 41 and 10 reacted to afford
DBPs 42 (35%) and 11 (69%), whereas larger cyclic
(19) This diene was obtained as a ca. 1:1 mixture with an unidentified
byproduct.
312
Org. Lett., Vol. 14, No. 1, 2012

Scheme 2. Synthesis of DBP 50
Scheme 3. Synthesis of Cannabinol (1)
ketones (g6-membered) 43 and 46 afforded nonaromatized
products (Table 3, entries 5À8). Cyclohexanone (43) gave
a mixture of cyclohexadienes 44 and 45 (60% combined,
57:3 by 1H NMR analysis), and cycloheptanone (46) gave
only 1,4-cyclohexadiene 47 (48%). The aromatization
of 44/45 and 47 using DDQ was reported earlier.13
2-Methylcyclopentanone (48) afforded only DBP 49,
which arises from reaction of the less substituted enamine
(Table 3, entry 9). Where comparisons are available, the
yields of the MCR are mostly better than those of the
corresponding stepwise syntheses (Table S2). Exceptions
are acetophenone (34), cyclohexanone (43), and cyclo-
heptanone (46).
6-Methoxysalicylaldehyde (18), which had failed to
afford DBP 19 in an MCR with cyclopentanone (10),
reacted with 6 and acetone (34) to provide DBP 52 (47%)
(Scheme 3). The MCR clearly tolerates one substituent,
but not two, in the bay region of the DBP framework.
Salicycladehyde 5120 also reacted well with 6 and 32,
affording DBP 53 (48%) on a 1.2 g scale. This product
was converted into cannabinol (1) by two different four-
step pathways (Scheme 3). Hydrolysis of 53 afforded acid
54 (90%). Treatment of 54 with MeLi (8 equiv), followed
by reaction of the crude product with p-TsOH, brought
about simultaneous conversion of the acid group to a
methyl ketone and the pyranone system to a dimethylpy-
ran unit. Alkene 55 (14%) was consistently obtained along
with the intended product 56 (42%). Alternatively, alkene
55 could be accessed by a Grignard reaction of 53 with
MeMgBr, followed by treatment of the crude product
with p-TsOH (87%, 2 steps). Oxidative cleavage of the
terminal alkene then afforded methyl ketone 56 (57%).
The synthesis of cannabinol (1) was then completed by
reacting 56with HI/Ac2O, which effectedboth demethyla-
tion and deacylation in high yield (95%). This seldom-used
retro-FriedelÀCrafts acylation relies upon the presence of
an adjacent methyl group.21
In conclusion, previously reported stepwise syntheses
of DBPs have been combined to afford a multicomponent
domino reaction that performs substantially better than the
stepwise approaches in most cases. Six reactions (Knoevenagel
reaction, transesterification, enamine formation, IEDDA re-
action, 1,2-elimination, transfer hydrogenation) occur during
the MCR, in which both IEDDA components are generaterd
in situ and pyrrolidine mediates two separate processes
(Knoevenagel reaction and enamine formation). This chem-
istry has been applied in the total synthesis of cannabinol (1).
Acknowledgment. Financial support of this work from
the Natural Sciences and Engineering Research Council
(NSERC) of Canada is gratefully acknowledged.
Supporting Information Available. Experimental pro-
1
cedures and characterization data. H and 13C NMR
€
(20) (a) Four steps from 3,5-dimethoxybenzoic acid: Baeckstrom, P.;
€
€
Sundstrom, G. Acta Chem. Scand. 1970, 716. (b) Torang, J.; Nieger, M.;
spectra. Tables of yields for stepwise and multicompo-
nent syntheses of DBPs. This material is available free of
charge via the Internet at http://pubs.acs.org.
€
Brase, S. Synthesis 2005, 1888.
(21) Olah, G. A.; Laali, K.; Mehrotra, A. K. J. Org. Chem. 1983, 48,
3360.
Org. Lett., Vol. 14, No. 1, 2012
313