Carbohydrate based IMDA/aldol strategy towards the densely functionalized trans-decalin subunit of azadirachtin

Article abstract of DOI:10.1039/a806400h

L-Rhamnal is converted into an hex-2-en-4-ulo C-glycopyranoside in which properly configured diastereomeric centers, asymmetric as well as geometric, are developed in the pendant anomeric substitutent via a Claisen aldol addition, and the resulting product proceeds via an IMDA reaction to provide a highly functionalised terpene AB ring system.

Full text of DOI:10.1039/a806400h

Carbohydrate based IMDA/aldol strategy towards the densely functionalized
trans-decalin subunit of azadirachtin
Dieter Haag, Xiao-Tao Chen and Bert Fraser-Reid*
Natural Products and Glycotechnology Research Institute Inc, 4118 Swarthmore Road, Durham, NC 27707 USA.
E-mail: dglucose@aol.com
Received (in Corvallis OR, USA) 10th August 1998, Accepted 14th October 1998
L
-Rhamnal is converted into an hex-2-en-4-ulo C-glycopyr-
The trans-decalin subunit of azadirachtin 1, as well as of
related terpenoids, invariably possesses hydroxy and carbon
substituents at C3 and C4 respectively,6 positions that correlate
with C8 and C9 of the sugar IMDA precursor. These
functionalities could conceivably be furnished by an aldol
reaction. Accordingly, the aldehyde group of 6c was first
protected by acetalization. Cleavage of the terminal double
bond via OsO4 induced dihydroxylation followed by periodate
cleavage in two discrete steps, but not in the one-pot Lemieux–
Johnson protocol,11 gave aldehyde 6e in 85% yield.
anoside in which properly configured diastereomeric cen-
ters, asymmetric as well as geometric, are developed in the
pendant anomeric substitutent via a Claisen aldol addition,
and the resulting product proceeds via an IMDA reaction to
provide a highly functionalised terpene AB ring system.
The development of synthetic routes from carbohydrates to
densely functionalized carbocycles has been a sustained area of
interest in our research group.1,2 The intramolecular Diels–
Alder (IMDA) strategy depicted in Scheme 13,4 takes advantage
of the preferred b-
2-ene-4-ones. The desired (S)-configuration exists in b-d- and
a- -derivatives, but either may be used since the sugar’s C5
stereocenter, whereby or is designated, does not survive in
the advanced retron 2. Thus a regioselective Baeyer–Villiger
oxidation of 3 permits its ready removal. This simplifying
retroanalysis also makes provisions for an array of functional
groups such as is found in the clerodane family of terpenes,5
exemplified by the ‘western half’ of the azadirachtins 1.6 An
IMDA approach, based on 4 with appropriate synthons at P, Q,
R, S, T and U, for this family of potent pharmacologically active
compounds,7 is therefore desirable.8
D-face approach in reactions of pyranosidic
Y
L
iii
O
D
L
AcO
i, ii
O
RO
76%
AcO
X
X
5a X = H
b X = Br
6a R = Ac, X = Br, Y = CH2
iv,v
quant.
72–76%
b R = TBDMS, X = Br, Y = CH2
c R = TBDMS, X = CHO, Y = CH2
vi
97%
vii
94%
d R = TBDMS, X = CH(OMe)2, Y = CH2
e R = TBDMS, X = CH(OMe)2, Y = O
viii,ix
85%
Dibromination of diacetyl -rhamnal 5a, followed by dehy-
L
drobromination, gave the 2-bromo glycal 5b in 72–76% yield
(Scheme 2). Danishefsky’s version9 of the Ferrier rearrange-
MO
EtO
ment10 gave a 76% yield of a-
-C-glycoside 6a (along with
L
+
A: M = Li, THF/HMPA, –78 °C
B: M = BBu2, CH2Cl2, –78 °C
19% of the corresponding b-anomer). The acetyl group was
replaced by TBDMS and a THF solution of the latter was mixed
with 5 equiv. of DMF, and then 6 equiv. of ButLi were added
slowly (ca. 2 h) at 278 °C to furnish, after aqueous work-up,
enal 6c in almost quantitative yield.
OH
8S
O
OH
8R
O
OEt
OEt
O
O
+
TBDMSO
TBDMSO
R4
H
H
O
R5
O
R1O
O
O
OH
7
8
1
MeO OMe
a (syn) + b (anti)
MeO OMe
P
14
U
10
8
8
a (syn) + b (anti)
3
O
O
4
T
Q
R2O
OH
H
H
S
R
R3
x–xii
x–xii
O
2
1
OAc
O
O
OAc
O
terpene
numbering
OEt
OEt
O
O
X
Y
O
O
O
X
O
O
4
5
9
10
T
H
H
1
3
P
8
1
3
a (syn) 76%, b (anti) 77%
a (syn) 74%, b (anti) 77%
U
U
2
13
Y
S
9
8
P
9
Scheme 2 Reagents and conditions: i, Br2, CH2Cl2; ii, A: one-pot, DBU,
76% + 14% of 5a, or B: work-up, then DBN, toluene–DMSO, 72% + 4% of
5a; iii, allyltrimethylsilane, BF3·Et2O, CH2Cl2, 0 °C to room temp.; iv,
NaOMe, MeOH; v, NaH, TBDMSCl, THF; vi, THF, DMF (5.0 equiv.),
278 °C, then ButLi (6.0 equiv.); vii, CSA, Me2C(OMe)2, reflux; viii, NMO,
1% OsO4, THF–H2O; ix, NaIO4, THF–H2O; x, Ac2O, Py, cat. DMAP,
CH2Cl2, 0 °C to room temp.; xi, HF·Py, THF, room temp.; xii, PCC, SiO2,
CH2Cl2, room temp.
T
Q
H
H
Q
R
S
R
3
X = CH2OR,
Y = H (β-D)
Y = CH2OR,
X = H (α-L)
4a
b
sugar
numbering
Scheme 1
Chem. Commun., 1998, 2577–2578
2577

Table 2 Diastereomeric product distribution in the aldol addition of 9b
Our next task was to elaborate the anomeric C-substituent
into the required dienic tether. A Claisen-type aldol addition
with ethyl sorbate seemed an ideal approach. Indeed that an
8-hydroxy-9-ethoxycarbonyl addition product had formed was
apparent from NMR examination of the crude material.
Separation of the four diastereomeric aldol adducts was
achieved by (repeated) flash chromatography, and syn- and
anti-configuration were easily distinguished on basis of the
coupling constant between the protons attached to C8 and C9
Yield (%)
Enolate/
Method equiv.
Yield of
6e (%)
t/min 7a
7b
8a
8b
7:8
A
A
A
A
B
2.0
2.0
2.0
4.0
2.5
2
20
60
2
29
24
24
35
29
26
27
27
28
6
12
17
18
14
58
12
11
11
13
5
2.3:1 15
1.8:1 15
1.8:1 13
2.3:1
1:1.8
6
—
300
3
3
(syn: J ~ 5.0 Hz, anti: J ~ 9.0 Hz). However, the absolute
configuration at C8 and C9 was assigned in retrospect upon
isolation of IMDA products 11–14 (Table1).
syn:anti ratios were generally poor. We therefore examined the
syn-selective protocol of Evans12 (conditions B) on 6e, and
were rewarded with a 61% yield of 8a+b in an 11:1 ratio.
Steric factors on the tether can substantially affect the course
of IMDA reactions.4,13 We were therefore pleased to see, from
the formation of 13 and 14 (Table 1), that the required (R)-
configuration at C8 of precursors 10a+b presents no obstacle to
the success of the IMDA reaction. Conceptually, either
configuration at C9 is acceptable in view of the future
quaternization at C4 of compounds 13 and 14. Furthermore, the
results in Table 1 suggest that selective formation of the IMDA
products may be possible by judicious choice of temperature
and duration of the reaction. The bulk of the C8-hydroxy
protecting group might also have a salutary effect.
For proof-of-concept, the major component (assigned in
retrospect as 7a, vide infra) was acetylated and treatment of the
crude material with HF–pyridine in THF for one day at ambient
temperature effected desilylation as well as cleavage of the
rather acid-sensitive dimethyl acetal. Final activation was
accomplished by oxidation with PCC on silica gel to furnish the
corresponding aldehyde enone 9a in 76% yield for the whole
four-step sequence. Heating of 5a in toluene at reflux for 30 h
afforded a 23% yield of 11, the structure being confirmed
independently by NMR (Table 1) and X-ray analyses.
Conceivably, the efficiency of the IMDA step, as well as the
correct C3 configuration in the resultant product, could both be
ensured by fine-tuning the preparative procedures. Thus under
conditions A (Scheme 2) variations in the equivalents of
enolate, and the duration of the reaction revealed that rapid
equilibration was occuring at 278 °C (Table 2). Our studies
also showed (a) that compounds 7a+b, having the undesired
(8S)-configuration, were kinetically favored, and (b) that the
These and other refinements for this IMDA/aldol approach to
compounds such as 1 are underway.
We are grateful to Professor A. T. McPhail of Duke
University for X-ray structure determination of 14, and to Dr
Ken Henry for insight and helpful suggestions. D. H. thanks the
Alexander von Humboldt Foundation for a Feodor Lynen
Fellowship. Partial support from the National Institutes of
Health (GM 51237) is acknowledged. Dedicated to the memory
of Professors H.-D. Scharf and D. H. R. Barton.
Table 1 Influence of the configuration of C8 and C9 in the dienic tether on
the course of the IMDA reaction (E = CO2Et)
IMDA
precursor Conditions Yield Product
3J/Hz
O
H
H
H
H
H
7
Notes and references
7,8: 12.4
E
toluene
H
7',8: 5.0
8,9: 6.0
1 B. Fraser-Reid and R. C. Anderson, Prog. Chem. Org. Nat. Prod., 1980,
39, 1; B. Fraser-Reid and R. Tsang, in Strategies and Tactics in Organic
Synthesis, ed T. Lindberg, Academic Press, New York, 1989, vol. 2, pp
123–162.
2 J. C. López and B. Fraser-Reid, Chem. Commun., 1997, 2251.
3 R. Tsang and B. Fraser-Reid, J. Org. Chem., 1992, 57, 1065.
4 J. C. López, A. M. Gómez and B. Fraser-Reid, J. Chem. Soc., Chem.
Commun., 1993, 762.
9a
23%
85%
70%
44%
AcO
AcO
H
reflux, 30 h
9
8
O
H
9,10: 5.6
O
O
O
O
H
H
11
O
H
7
H
7,8: 11.5
7',8: 4.7
benzene
H
8
5 I. R. Hanson, Nat. Prod. Rep., 1998, 15, 93.
6 S. V. Ley, A. A. Denholm and A. Wood, Nat. Prod. Rep., 1993, 10,
109.
9b
reflux, 5 h
9
O
E
8,9: 10.6
9,10: 12.6
H
H
7 M. Jacobson, Pharmacology and Toxicity of Neem, in Focus on
Phytochemical Pesticides, The Neem Tree, ed. J. Jacobson, CRC Press,
1989; vol. 1, p 133.
8 For some examples of synthetic approaches to the azadirachtins, see:
S. V. Ley, Pure Appl. Chem., 1994, 66, 2099; H. Watanabe, T.
Watanabe, K. Mori and T. Kitahara, Tetrahedron Lett., 1997, 38, 4429;
N. Konah, J. Ishihara and A. Murai, Synlett, 1997, 737.
9 S. J. Danishefsky and J. F. Kerwin Jr., J. Org. Chem., 1982, 47,
3805.
10 R. J. Ferrier and S. Middleton, Chem. Rev., 1993, 93, 2779.
11 R. Pappo, D. S. Allen Jr., R. U. Lemieux and W. S. Johnson, J. Org.
Chem., 1956, 21, 478.
12 D. A. Evans, J. V. Nelson, E. Vogel and T. R. Taber, J. Am. Chem. Soc.,
1981, 103, 3099.
12
O
H
7
H
7,8: 2.9
7',8: 3.8
8,9: 3.6
toluene
reflux, 18 h
H
8
10a
9
O
E
AcO
9,10: 12.6
H
13
O
H
H
E
9
7,8: 3.2
7',8: 3.8
8,9: 3.1
9,10: 2.6
7
xylenes
reflux, 24 h
H
10b
H
13 D. F. Taber, Intramolecular Diels–Alder and Alder Ene Reactions,
Springer-Verlag, 1984.
8
O
AcO
H
14
Communication 8/06400H
2578
Chem. Commun., 1998, 2577–2578