Communication
stage. Next, osmium-catalyzed dihydroxlation[21] of 20 gave
a mixture of diol 21 and its minor diastereomer (d.r.=4.3:1)
from which 21 was isolated in 75% yield. It is worth noting
that, without quinuclidine as additive,[21] this dihydroxylation
required prolonged reaction time and higher catalyst loading
and afforded 21 in lower yield. Deprotection of the MOM ether
of 21[22] followed by selective protection of the cis-diol as the
acetonide afforded alcohol 22. Swern oxidation of the secon-
dary alcohol of 22 provided desired ketone 23, which was
then converted to enone 10 following a well-established
procedure[9b] that involves: 1) enolization of the ketone of 23
followed by addition of allyl chloroformate to form the corre-
sponding allyl enol carbonate; and 2) a Saegusa-type palladi-
um-mediated oxidation[23] of this carbonate to the enone 10.
Next, a Hauser annulation[9c,11] between cyanophthalide 9 and
complex enone 10, employing lithium tert-butoxide as a base,
gave rise to the desired tricyclic hydroquinone, which was
found to be unstable and immediately methylated to afford tri-
cyclic compound 24. Due to the instability of the tricyclic hy-
droquinone intermediate and difficulty in handling, tricyclic
compound 24 was produced in a range of yields from 60–87%
for two steps. Removal of the TBDPS ether of 24 followed by
Swern oxidation of the resulting primary alcohol afforded the
aldehyde 25, which was quickly subjected to Pedersen-modi-
fied pinacol coupling[24] and subsequent Swern oxidation of
the resulting secondary alcohol to provide tetracycle 26. It is
worth noting that our attempts to directly convert 25 to 26
using N-heterocyclic carbene-catalyzed intramolecular alde-
hyde–ketone benzoin condensation[25] were unfruitful. Finally,
deprotection of the acetonide of 26 gave rise to the complex
tetracyclic aryliodide 8. The structure of tetracycle 8 was unam-
biguously confirmed by single-crystal X-ray crystallographic
analysis (Scheme 3).[26]
Scheme 1. Retrosynthesis of derhodinosylurdamycin A (4).
Scheme 2. Reagents and conditions: a) MgBr2, ether/THF, 87%; b) AlCl3, NIS,
CH2Cl2, RT; c) MOMCl, iPr2NEt, CH2Cl2, 56% for two steps; d) MgBr2, ether/
THF, 90%; e) K2CO3, BnBr, Acetone, 95%; f) TMSCN, KCN, 18-C-6, CH2Cl2, then
AcOH, 92%. NIS=N-iodosuccinimide; MOMCl=chloromethyl methyl ether.
inseparable mixture that contained the desired o-iodophenol
as the major product together with undesired p-iodophenol
and o,p-bis-iodophenol, as well as some recovered starting ma-
terial. Gratifyingly, converting this mixture of phenols to their
corresponding MOM ethers gave a separable mixture[16] from
which desired product 13 was isolated in 56% yield over two
steps. Similarly, magnesium bromide mediated deprotection of
methoxymethyl ether of 13, followed by reprotection of the re-
sulting phenol as its benzyl ether, provided desired product
14. Finally, cyanophthalide 9 was prepared from 14 in 92%
yield following a known two-step procedure.[17]
With tetracyclic aryliodide 8 in hand, we sought to prepare
its cross-coupling partner, glycal stannane 7. As shown in
Scheme 4, known d-rhamnal-derived glycal stannane 27[13a]
was subjected to silyl ether deprotection followed by regiose-
lective silyl protection to provide mono-silyl ether 28. Benzyla-
tion of the C4-OH of compound 28 afforded desired glycal
stannane 7 in 70% yield over three steps. In order to obtain
disaccharide donor 5, stereoselective glycosylation[27] between
2,6-dideoxy-2-iodo-d-glucopyranosyl
trifluoroacetimidate
Next, complex enone 10 was prepared through a modifica-
tion of a previously reported synthetic approach.[9b] As shown
in Scheme 3, halogen–lithium exchange of alkenyl bromide 16,
derived from known chiral (R)-2-bromo-cyclohex-2-en-1-ol
15,[18] provided desired alkenyl lithium, which subsequently re-
acted with known chiral epoxide 17[19] in the presence of
boron trifluoride diethyl etherate to afford tertiary alcohol 18
in 71% yield.[20] Deprotonation of the tertiary alcohol of 18 by
potassium hydride in DMF, followed by addition of benzyl bro-
mide and a catalytic amount of tetra-n-butylammonium iodide,
provided corresponding benzyl ether 19, which underwent de-
protection of p-methoxyphenyl (PMP) ether and reprotection
of the resulting primary alcohol to furnish 20. This change of
protecting group was necessary to avoid the complication of
oxidative deprotection of p-methoxyphenyl ether at a late
donor 29[28,29] and l-rhodinose-derived acceptor 30[30] was first
carried out to furnish desired b-linked disaccharide 31 in 93%
yield. Next, radical-mediated reductive deiodination of di-
saccharide 31 followed by DDQ-mediated oxidative removal of
PMB ether and subsequent acetylation of the anomeric
hydroxy group afforded desired disaccharide donor 5.
With all the fragments in hand, we executed the remaining
studies for the completion of synthesis of derhodinosylurdamy-
cin A. Thus, standard Stille coupling of glycal stannane 7 and
aryl iodide 8 followed by mesylation of the secondary alcohol
afforded desired product 32 (Scheme 5). Next, a one-pot ste-
reoselective reduction[13] of the cyclic enol ether of 32 using
NaCNBH3 and HCl under carefully controlled pH and concomi-
tant removal of the TBS ether afforded desired complex b-C-
arylglycoside acceptor 6.[31] Next, tert-butyldimethylsilyl trifluor-
Chem. Eur. J. 2015, 21, 13553 – 13557
13554
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim