- Email: [email protected]
Retrosynthetic studies with forskolin
Tetrahedron Letters,Vo1.28,No.35,pp Printed in Great Britain
RETROSYNTHETIC
Bernard
Laboratoire
0040-4039/87 $3.00 + .OO Pergamon Journals Ltd.
4061-4064,1987
STUDIES
DELPECH
CNRS / ROUSSEL
WITH FORSKOLIN
and Robert LET1
IJCLAF - 102, Route de Noisy
B.P. 9 - 93230 ROMAINVILLE
(France)
ABSTRACT : New preparations of some dihydro- Y-pyrones derived from forskolin 1 are reported. ConJugate addition of cuprates is shown to occur on these compounds , in good yield, in the presence of BF -Et20 : dimethyl and vinyl cuprates add with major ~1 stereoselectivity, whereas dibuty 9 cuprate adds mostly B . Some other reaction conditions or copper reagents failed. Forskolin 1, forskohlii, Therefore
a labdane
diterpene
isolated
from the roots of the Indian herb Coleus
has been shown to have a very promising
a considerdble
interest
biological
and medical
has been found in its chemical
potential
modifications
(1).
and total
synthesis. Forskolin structural
is indeed a fascinating
features
rearrangements
target for total synthesis
often result in an unusual chemical
reactivity,
(2). The recent report of a stereocontrolled
forskolin 2 by IKEGAMI and coworkers the C ring elaboration,
retrosynthetic
4061
as shown by some remarkable
synthesis
(3) prompts us to disclose
from the following
since its very peculiar
of (+I-1,6,7-trideoxy
some of our results
scheme.
concerning
4062
0
0 -+
...o I
0
0 . ..I .-
g..
%
m
3 0
+
% 10
\O .“‘6 .. %X6
o
k-0 6.
..T
-0
0 +
0
-0
0
“.I
* .
.*
o...
%I
s” c
?f’
t
/ bl
406 3
Starting
from 1, we first prepared -3 according
to SAKSENA et al. (4). Our preliminary
experiments of vinyl copper reagents conjugate addition failed, possibly due to heterogeneous conditions
and thermal
instability
of these cuprates
(5, 6). Therefore,
we exai,linedthe
reaction
of 3 with dimethylcuprdte which is known to give homogeneous and quite stable solutions. However, attempts of conjugate addition with (CH3)2CuLi / Et20 (- 78" to
: only products of hydrolysis of the 1,9-carbonate
r.tp.) failed also with -3
78 -OAc, and 78 to 68 acetate Conjugate
cuprate
transposition,
additions
ones (9) are indeed generally control
of the conjugate
interaction
between
on Y-pyrones difficult.
addition
the incoming
thought
the bis-acetonide to facilitate
withdrdwing
For preparing
dihydro-y
followed
which probably possibly
B-unsaturated
Due to the low yield However,
ozonolysis
to the enolate;
bond is generated
(8) or thiin-4-
on the other hand
state in which a 1,4-diaxial
interaCtiOn
(10).
and 78 -0Ac migration
analogous
we met with -3, we next
(7) or thiin-4-ones
to that reported
compound
Reaction
material
eq,Et20,0")
of 1, we next tried to prepare
(17%),
with some chromones
8 -position
the protected
14% of g
was also recovered
reagent as observed
group in the
of 5 was
from 5 (RuCy3 /
only - and, quite unexpectedly, Starting
of 6 by the cuprate
ketones having a methyl
of 4 in CH2C12/CH30H
in 86% yield
of 6 with (CH3)2CuLi(3
but in 16% yield
(9).
for 3 (4), ozonolysis
6 was obtained
from a radical process.
due to some enolization
(7a) ora,
-pyrones
of -1 or 1, stereoelectronic creates a 1,3-diaxial
-pyrone -6. In 5, introduction of the l2-COOCH3 group was addition, as such an effect by an electron
However,
product1,
originated
state leading
and (or) the
of isolated compounds).
conjugate
by Cu(OAc)2/Pb(OAc)4).
gave the 1, 4-addition
bond formation
been shown for chromones
6 by a sequence
found to be unsatisfactory. NaI04,
hydrolysis
the cuprate
group has already
in the formation
transition
group and the C-9/0-9
In order to avoid the partial prepared
transition
gives a pre-twist
(72% yield
(7), dihydro-y
(chromones)
Moreover,
implies that8-CC
in the pre half-chair
cr-CC bond formation
were obtained
(11).
dihydro-Y-pyrone
-14.
gave the lactol 9 (94%) and not the expected
aldehyde -13. The lactol 9 was converted
into 10 with Pb(OAc)4/CaC03 (44%, 66% corrected reaction of -10 with (CH3)2CuLi was attempted to involve
from recovered !J) (12). Further 9cr-alkoxy cuprate
which,
if formed
formation
at C-13. However,
conjugate
addition
The aldehyde of pyridine,
-but,
facilitate
and direct an intramolecular
a CC bond
reaction
of -10 with (CH3)2CuLi (Et20, -78" to r.tp) gave no interestingly again, 5% of compound -12 were isolated.
13 was obtained
(87%) by ozonolysis
of 4 in CH2C12/CH30H
in the presence
of -14 by the sequence developed by SAKSENA et al (4). However, we found that 14 was also conveniently prepared (76%) in two steps from 4 (RuC13/NaI04 ; Cu(OAc)2/Pb (OAC)~), with no intermediate purification, like for the conversion
allowing
(13), might
the preparation
of 5 into -6. of conjugate
addition - with methyl,
with 14 in diethyl
ether or THF (-78" to r.tp).
Many attempts completely material
was alw&
In contrast,
dimethylcuprate
(15), reacted with 14 to yield with aqueous
in good yield
HCl in>HF
(from CH3Li/Et20 16 (56%).
-
reagents
- failed
In all these experiments,
starting
(90% to almost quant.)
the 1,4-addition
afforded
ethyl or vinyl copper (14).
and CuI), in the presence product
of BF3-Et20
15 (85%). Subsequent -
hydrolysis
at 20"
4064
Our next goal was then to determine a total synthesis in the foregoing
of forskolin. experiments,
the stereochemistry
of the reaction
As we were unable to achieve
we prepared 17
for the planning
vinyl or ethyl conjugate
(96%) by alkylation
of -14 (LDA/HMPT,
CH31 at
-78').Itethyl conjugate addition occurred on -17 in the same conditions,with
BF3-Et20,
yielding
was shown
18 (80%) and only traces of 19. The structure
unarabiguo&ly
to be 18, by cooparison\ith
of the major product
the hydrogenation
of
addition
product -19 of -4 (H2, Pd/C, EtOH,
r.tp). More recently, vinyl tri n-butyl obtained
we were also able to achieve tin and n-butyl lithium
addition
of vinyl cuprate
(from
and
-4 as the major product (59%) together with the C-13 epimer (14%).
To sum up, 1,4-addition BF3-Et20
the conjugate
in situ, and CuI), with added EF3-Et20,
of methyl or vinyl cuprate
and shown to occur stereoselectively
examples, transition
1,4-cuprate
addition
state. However,
dihydro-y-pyrones same conditions
appears
one has to be cautious
(13 - +
Ph3P=CH-CH2-CH3,
in the presence
an intermediate
in any generalization
of
at least in these twist type
to either other
: we observed that -14 reacts with nBu2CuLi in the
to afford the major E-1,4 addition
as shown by identification
Consequently,
to involve preferentially
or other copper reagents
(23%),
has been achieved
c1 at C-13.
with the compound
product
(61%) together
prepared
with the a-adduct
by the following
sequence
THF, r.tp; 1-12,Pd/C).
Acknowledgements : We thank Ur. P. POTIER, Dr. E. TOROFIANOFF, Dr. J. MARTEL and Dr. A. RAEIET for their interest in this work and fruitful discussions, and the Dspartement des Recherches et Applications Physiques at ROUSSEL-UCLAF (Romainville). We also thank HOECHST Frankfurt (Dr. TEICHMANN) for a gift of forskolin. REFERENCES AND NOTES 1) "Proceedings of the International Symposium on Forskolin : its Chemical, Biologicdl and Medical potential" Ed. : R.H. RUPP, N.J. de SOUZA, A.N. DOHADWALLA - HOECHST INDIA (1985) 2) a) A.K. SAKSCNA, t1.J. GREEN, H.J. SHUE, J.K. WONG, A.T. MC PHAIL, Tetrahedron Lett., 26, 551 (1985) b) Y. KHANDELWAL, G. MORAES, N.J. de SOUZA, H.W. FEHLHABER, E.F. PAULUS, Tetrahedron Lett., I, 6249 (1986) 3) S.I. HASHIMOTO, M. SONEGAWA, S. SAKATA, S. IKEGAllI, J.C.S. Chem. Comm., 24, (1987) 4) A.K. SAKSENA, M.J. GREEN, H.J. SHUE, J.K. WDNC, J.C.S. Chem. Comm., 1743 (1985) 5) H.O. HOUSE, M.J. UMEN, 3; Org. Chem; 38, 3893 (1973) 6) Noteworthy, attempts of conjugate addition with 3 and (CH =CH)2 CuCNLi were reported to fail completely (4), in contrast with the recent resul z s of S. IKEGi MI et al. for the synthesis of 2 (3). 7) a) T.W. WALLACE, Tetrahedron Lett., 25, 4299 (1984) b) P.D. CLARKE. A.O. FITTON. H. SUSCWZKY, T.W. WALLACE, H.A. DOWLATSHAHI. J.L. SUSCHITZKY, Tetrahedron-Lett., 27, 9i (1986) 8) a) M.B. YUNKER. U.E. PmUMANN. B. FRASER-REIU. Canad.J.Chem.. 55. 4002 (1977) bj T.E. GOODWIt?, M. CROWDER, i.B. WHITE, J.S.-SWANSON, F.E. Emi, W.L..MEYER, J. Org. Chem., 48, 376 (19B3) 9) a) R.J. BATTEN, J.D. COYLE, R.J.K. TAYLOR, Synthesis, 910 (1980) b) G. CASY, S. LANE, R.J.K. TAYLOR, J.C.S. PERKIN Trans 1, 1397 (1986) 10) E. TOROFIANOFF, Bull. Sot. Chim. Fr., 708 (1962) 11) H.O. HOUSE, C.Y. CHU, J.M. WILKINS, M.J. UMEN, J. Org. Chem., 40, 1460 (1975) 12) Compound 10 results from a quantitative 9 a to la -formyl migrxion in the reaction condition313) G.H. POSNER, C.E. WHITTEN, J.J. STERLING, J. Am. Chem. Sot., 95, 7788 (1973) 14) At the present time, it is not clear tiether these experiments failed due to heterogeneous conditions and (or) complexation with the oxygenated functions of forskolin derivatives, structurally and chemically more complex than 2. See also (4) 15) Y. YAMAMOTO, Angew.Chem. Int. Ed. Engl., 2, 947 (1986)
(Received in France 20 May 1987)
RETROSYNTHETIC
Bernard
Laboratoire
0040-4039/87 $3.00 + .OO Pergamon Journals Ltd.
4061-4064,1987
STUDIES
DELPECH
CNRS / ROUSSEL
WITH FORSKOLIN
and Robert LET1
IJCLAF - 102, Route de Noisy
B.P. 9 - 93230 ROMAINVILLE
(France)
ABSTRACT : New preparations of some dihydro- Y-pyrones derived from forskolin 1 are reported. ConJugate addition of cuprates is shown to occur on these compounds , in good yield, in the presence of BF -Et20 : dimethyl and vinyl cuprates add with major ~1 stereoselectivity, whereas dibuty 9 cuprate adds mostly B . Some other reaction conditions or copper reagents failed. Forskolin 1, forskohlii, Therefore
a labdane
diterpene
isolated
from the roots of the Indian herb Coleus
has been shown to have a very promising
a considerdble
interest
biological
and medical
has been found in its chemical
potential
modifications
(1).
and total
synthesis. Forskolin structural
is indeed a fascinating
features
rearrangements
target for total synthesis
often result in an unusual chemical
reactivity,
(2). The recent report of a stereocontrolled
forskolin 2 by IKEGAMI and coworkers the C ring elaboration,
retrosynthetic
4061
as shown by some remarkable
synthesis
(3) prompts us to disclose
from the following
since its very peculiar
of (+I-1,6,7-trideoxy
some of our results
scheme.
concerning
4062
0
0 -+
...o I
0
0 . ..I .-
g..
%
m
3 0
+
% 10
\O .“‘6 .. %X6
o
k-0 6.
..T
-0
0 +
0
-0
0
“.I
* .
.*
o...
%I
s” c
?f’
t
/ bl
406 3
Starting
from 1, we first prepared -3 according
to SAKSENA et al. (4). Our preliminary
experiments of vinyl copper reagents conjugate addition failed, possibly due to heterogeneous conditions
and thermal
instability
of these cuprates
(5, 6). Therefore,
we exai,linedthe
reaction
of 3 with dimethylcuprdte which is known to give homogeneous and quite stable solutions. However, attempts of conjugate addition with (CH3)2CuLi / Et20 (- 78" to
: only products of hydrolysis of the 1,9-carbonate
r.tp.) failed also with -3
78 -OAc, and 78 to 68 acetate Conjugate
cuprate
transposition,
additions
ones (9) are indeed generally control
of the conjugate
interaction
between
on Y-pyrones difficult.
addition
the incoming
thought
the bis-acetonide to facilitate
withdrdwing
For preparing
dihydro-y
followed
which probably possibly
B-unsaturated
Due to the low yield However,
ozonolysis
to the enolate;
bond is generated
(8) or thiin-4-
on the other hand
state in which a 1,4-diaxial
interaCtiOn
(10).
and 78 -0Ac migration
analogous
we met with -3, we next
(7) or thiin-4-ones
to that reported
compound
Reaction
material
eq,Et20,0")
of 1, we next tried to prepare
(17%),
with some chromones
8 -position
the protected
14% of g
was also recovered
reagent as observed
group in the
of 5 was
from 5 (RuCy3 /
only - and, quite unexpectedly, Starting
of 6 by the cuprate
ketones having a methyl
of 4 in CH2C12/CH30H
in 86% yield
of 6 with (CH3)2CuLi(3
but in 16% yield
(9).
for 3 (4), ozonolysis
6 was obtained
from a radical process.
due to some enolization
(7a) ora,
-pyrones
of -1 or 1, stereoelectronic creates a 1,3-diaxial
-pyrone -6. In 5, introduction of the l2-COOCH3 group was addition, as such an effect by an electron
However,
product1,
originated
state leading
and (or) the
of isolated compounds).
conjugate
by Cu(OAc)2/Pb(OAc)4).
gave the 1, 4-addition
bond formation
been shown for chromones
6 by a sequence
found to be unsatisfactory. NaI04,
hydrolysis
the cuprate
group has already
in the formation
transition
group and the C-9/0-9
In order to avoid the partial prepared
transition
gives a pre-twist
(72% yield
(7), dihydro-y
(chromones)
Moreover,
implies that8-CC
in the pre half-chair
cr-CC bond formation
were obtained
(11).
dihydro-Y-pyrone
-14.
gave the lactol 9 (94%) and not the expected
aldehyde -13. The lactol 9 was converted
into 10 with Pb(OAc)4/CaC03 (44%, 66% corrected reaction of -10 with (CH3)2CuLi was attempted to involve
from recovered !J) (12). Further 9cr-alkoxy cuprate
which,
if formed
formation
at C-13. However,
conjugate
addition
The aldehyde of pyridine,
-but,
facilitate
and direct an intramolecular
a CC bond
reaction
of -10 with (CH3)2CuLi (Et20, -78" to r.tp) gave no interestingly again, 5% of compound -12 were isolated.
13 was obtained
(87%) by ozonolysis
of 4 in CH2C12/CH30H
in the presence
of -14 by the sequence developed by SAKSENA et al (4). However, we found that 14 was also conveniently prepared (76%) in two steps from 4 (RuC13/NaI04 ; Cu(OAc)2/Pb (OAC)~), with no intermediate purification, like for the conversion
allowing
(13), might
the preparation
of 5 into -6. of conjugate
addition - with methyl,
with 14 in diethyl
ether or THF (-78" to r.tp).
Many attempts completely material
was alw&
In contrast,
dimethylcuprate
(15), reacted with 14 to yield with aqueous
in good yield
HCl in>HF
(from CH3Li/Et20 16 (56%).
-
reagents
- failed
In all these experiments,
starting
(90% to almost quant.)
the 1,4-addition
afforded
ethyl or vinyl copper (14).
and CuI), in the presence product
of BF3-Et20
15 (85%). Subsequent -
hydrolysis
at 20"
4064
Our next goal was then to determine a total synthesis in the foregoing
of forskolin. experiments,
the stereochemistry
of the reaction
As we were unable to achieve
we prepared 17
for the planning
vinyl or ethyl conjugate
(96%) by alkylation
of -14 (LDA/HMPT,
CH31 at
-78').Itethyl conjugate addition occurred on -17 in the same conditions,with
BF3-Et20,
yielding
was shown
18 (80%) and only traces of 19. The structure
unarabiguo&ly
to be 18, by cooparison\ith
of the major product
the hydrogenation
of
addition
product -19 of -4 (H2, Pd/C, EtOH,
r.tp). More recently, vinyl tri n-butyl obtained
we were also able to achieve tin and n-butyl lithium
addition
of vinyl cuprate
(from
and
-4 as the major product (59%) together with the C-13 epimer (14%).
To sum up, 1,4-addition BF3-Et20
the conjugate
in situ, and CuI), with added EF3-Et20,
of methyl or vinyl cuprate
and shown to occur stereoselectively
examples, transition
1,4-cuprate
addition
state. However,
dihydro-y-pyrones same conditions
appears
one has to be cautious
(13 - +
Ph3P=CH-CH2-CH3,
in the presence
an intermediate
in any generalization
of
at least in these twist type
to either other
: we observed that -14 reacts with nBu2CuLi in the
to afford the major E-1,4 addition
as shown by identification
Consequently,
to involve preferentially
or other copper reagents
(23%),
has been achieved
c1 at C-13.
with the compound
product
(61%) together
prepared
with the a-adduct
by the following
sequence
THF, r.tp; 1-12,Pd/C).
Acknowledgements : We thank Ur. P. POTIER, Dr. E. TOROFIANOFF, Dr. J. MARTEL and Dr. A. RAEIET for their interest in this work and fruitful discussions, and the Dspartement des Recherches et Applications Physiques at ROUSSEL-UCLAF (Romainville). We also thank HOECHST Frankfurt (Dr. TEICHMANN) for a gift of forskolin. REFERENCES AND NOTES 1) "Proceedings of the International Symposium on Forskolin : its Chemical, Biologicdl and Medical potential" Ed. : R.H. RUPP, N.J. de SOUZA, A.N. DOHADWALLA - HOECHST INDIA (1985) 2) a) A.K. SAKSCNA, t1.J. GREEN, H.J. SHUE, J.K. WONG, A.T. MC PHAIL, Tetrahedron Lett., 26, 551 (1985) b) Y. KHANDELWAL, G. MORAES, N.J. de SOUZA, H.W. FEHLHABER, E.F. PAULUS, Tetrahedron Lett., I, 6249 (1986) 3) S.I. HASHIMOTO, M. SONEGAWA, S. SAKATA, S. IKEGAllI, J.C.S. Chem. Comm., 24, (1987) 4) A.K. SAKSENA, M.J. GREEN, H.J. SHUE, J.K. WDNC, J.C.S. Chem. Comm., 1743 (1985) 5) H.O. HOUSE, M.J. UMEN, 3; Org. Chem; 38, 3893 (1973) 6) Noteworthy, attempts of conjugate addition with 3 and (CH =CH)2 CuCNLi were reported to fail completely (4), in contrast with the recent resul z s of S. IKEGi MI et al. for the synthesis of 2 (3). 7) a) T.W. WALLACE, Tetrahedron Lett., 25, 4299 (1984) b) P.D. CLARKE. A.O. FITTON. H. SUSCWZKY, T.W. WALLACE, H.A. DOWLATSHAHI. J.L. SUSCHITZKY, Tetrahedron-Lett., 27, 9i (1986) 8) a) M.B. YUNKER. U.E. PmUMANN. B. FRASER-REIU. Canad.J.Chem.. 55. 4002 (1977) bj T.E. GOODWIt?, M. CROWDER, i.B. WHITE, J.S.-SWANSON, F.E. Emi, W.L..MEYER, J. Org. Chem., 48, 376 (19B3) 9) a) R.J. BATTEN, J.D. COYLE, R.J.K. TAYLOR, Synthesis, 910 (1980) b) G. CASY, S. LANE, R.J.K. TAYLOR, J.C.S. PERKIN Trans 1, 1397 (1986) 10) E. TOROFIANOFF, Bull. Sot. Chim. Fr., 708 (1962) 11) H.O. HOUSE, C.Y. CHU, J.M. WILKINS, M.J. UMEN, J. Org. Chem., 40, 1460 (1975) 12) Compound 10 results from a quantitative 9 a to la -formyl migrxion in the reaction condition313) G.H. POSNER, C.E. WHITTEN, J.J. STERLING, J. Am. Chem. Sot., 95, 7788 (1973) 14) At the present time, it is not clear tiether these experiments failed due to heterogeneous conditions and (or) complexation with the oxygenated functions of forskolin derivatives, structurally and chemically more complex than 2. See also (4) 15) Y. YAMAMOTO, Angew.Chem. Int. Ed. Engl., 2, 947 (1986)
(Received in France 20 May 1987)