1
The H and 13C NMR data exhibited by this diol 4 were
afford the anti-adduct 17 in 81% yield (95:5 dr, 95% ee).17
Following methyl ether formation (NaH, MeI), the terminal
alkene was oxidized to give 9 (71%)18 under modified19
Wacker conditions. By using (+)-Ipc2BOMe, the enantio-
meric methyl ketone ent-9 was prepared from 16 in an
analogous manner.
consistent with that reported by D’Auria and co-workers for
both the C17-C22 fragment obtained from the chemical
degradation of natural reidispongiolide A and that of a
synthetic sample of ent-4.7 The absolute configuration of the
degradation fragment was further confirmed as (18S,19S,-
21R) through comparison of the specific rotation values.12
Additionally, treatment of the diol 4 with p-nitrobenzoyl
chloride (Et3N, DMAP) gave the bis-ester 11, where single-
crystal X-ray analysis confirmed the all-syn stereochemistry.
The convergent assembly of the C23-C35 fragments
commenced with a substrate-controlled, boron-mediated aldol
reaction (c-Hex2BCl, Et3N)8a,10 between the ethyl ketone 813
and aldehyde 12 (Scheme 3). Upon standard oxidative
We next examined the coupling of the ketone 9 and the
aldehyde 18, obtained by Dess-Martin oxidation of 15
(99%). This pivotal aldol coupling step was best achieved
by generation of the lithium enolate of 9 (LDA, THF, -78
°C) and addition of a solution of 18 (0.5 equiv, 30 min) to
give the â-hydroxy ketone 19 in 85% yield, as an inconse-
quential 4:1 mixture of diastereomers. Subsequent elimina-
tion, through formation of the corresponding mesylate (MsCl,
Et3N) and in situ treatment with DBU, provided (E)-enone
20 (65%). Subjection of 20 to standard hydrogenation
conditions (H2, Pd/C) then effected both clean hydrogenolysis
of the PMP acetal and reduction of the alkene. Finally, ketone
reduction (NaBH4) and TBS ether removal (TBAF) provided
the alcohols 5, obtained as a 4:1 mixture of epimers at C31,
which were separated by flash chromatography.
Scheme 3a
1
Notably, the H NMR data (500 MHz, CD3OD) of the
alcohols 5 were in close agreement20 to that reported by
D’Auria and co-workers for the C23-C35 fragment obtained
in their degradation work,7 thus supporting this relation-
ship between the two remote stereoclusters at C24-C28 and
C31-C33 in reidispongiolide A. At the time, inconsistencies
between our 13C NMR data and that reported by the Naples
group led us to prepare 6 having the other anti stereorela-
tionship at C31-C32. This involved the analogous aldol
coupling between 18 and ent-9 to give 21 followed by
elaboration into 6, obtained as a 4:1 mixture of epimers at
C31. Spectroscopic analysis20 of the separated alcohols 6
revealed that these were clearly diastereomers of the
C23-C35 fragment obtained by ozonolysis of reidispongiolide
A. Subsequently, comparison of the 13C NMR data for
synthetic fragments 5 and 6 with the revised data provided21
for material obtained by chemical degradation allowed the
confident assignment of the relatiVe stereochemistry for the
C23-C35 sequence of reidispongiolide A.
a Conditions: (a)c-Hex2BCl,Et3N,Et2O,-78°C;(b)Me4NBH(OAc)3,
MeCN/AcOH; (c) DDQ, 4 Å MS, CH2Cl2; (d) NaH, MeI, THF;
(e) TBAF, THF.
workup, the expected anti-anti adduct 13, resulting from
the high level of π-face discrimination exercised by the
intermediate (E)-enolate,8 was obtained in 95% yield (95:5
dr). This adduct provided a suitable substrate for hydroxyl-
directed reduction to set up the C23-C29 stereopentad. By
employing Me4NBH(OAc)3,14 the desired 1,3-anti diol 14
was obtained cleanly (90%, >99:1 dr). With the five
contiguous stereocenters now secured in a concise manner,
the differentiation of the two hydroxyl groups was required.
Treatment of the diol 14 under DDQ-mediated oxidative
cyclization conditions15 resulted in the exclusive formation
of the corresponding six-membered PMP acetal, as a single
diastereomer. Finally, methylation of the free hydroxyl (NaH,
MeI) and removal of the TIPS ether (TBAF) afforded the
crystalline alcohol 15 in 80% overall yield from 14. At this
point, the relative stereochemistry of this C23-C29 subunit
was confirmed by X-ray crystallographic analysis of 15.
With the required stereopentad 15 in hand, we turned to
preparing the methyl ketones 9 and ent-9 to access both pos-
sible anti-relationships at C32 and C33 in the extended
C23-C35 fragment (Scheme 4). By using Brown’s methodol-
ogy,16 the aldehyde 16 was treated with the (E)-crotylborane
reagent derived from trans-butene and (-)-Ipc2BOMe to
(12) The low value of [R]20 recorded for synthetic 4 (+4.1, MeOH)
D
was in accord with that reported (ref 7) for the corresponding degradation
fragment. The bis-(S)-MTPA ester was also prepared for spectroscopic
comparison.
(13) (a) Paterson, I.; Norcross, R. D.; Ward, R. A.; Romea, P.; Lister,
M. A. J. Am. Chem. Soc. 1994, 116, 11287. (b) Paterson, I.; Florence, G.
J.; Gerlach, K.; Scott, J. P.; Sereinig, N. J. Am. Chem. Soc. 2001, 123,
9535.
(14) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc.
1988.
(15) Oikawa, Y.; Yoshioka, T.; Yonemitsu, O. Tetrahedron Lett. 1982,
23, 889.
(16) Brown, H. C.; Bhat, K. S.; Randad, R. S. J. Org. Chem. 1989, 54,
1570.
(17) The [R]20D value recorded for 17 (-8.4, CHCl3) was in accord with
that previously reported. McRae, K. J.; Rizzacasa, M. A. J. Org. Chem.
1997, 62, 1196.
(18) Accompanied by small amounts of the corresponding aldehyde (8%).
(19) Smith, A. B., III; Cho, Y. S.; Friestad, G. K. Tetrahedron Lett. 1998,
39, 8765.
(20) For tables of spectroscopic data (1H and 13C NMR) for 5 and 6, see
Supporting Information.
(21) Professor D’Auria has since provided us with revised 13C NMR
data for their C23-C35 degradation fragment which are in complete
agreement with the data we have acquired for the alcohols 5.
Org. Lett., Vol. 5, No. 11, 2003
1965