Total synthesis and stereochemical revision of acortatarins A and B

Article abstract of DOI:10.1021/ol202121k

A first total synthesis of acortatarins A, B, and an enantiomer of the proposed structure of acortatarin B is described by using readily available d-sugars. This convergent total synthesis revealed the revision of the absolute configuration of acortatarin A and structural revision of acortatarin B. The key steps involved are regioselective epoxide opening with deprotonated 2,5-disubstituted pyrrole and spiroketalization.

Full text of DOI:10.1021/ol202121k

ORGANIC
LETTERS
2011
Vol. 13, No. 20
5452–5455
Total Synthesis and Stereochemical
Revision of Acortatarins A and B
Gangarajula Sudhakar,*,† Vilas D. Kadam,† Shruthi Bayya,† Gavinolla Pranitha,‡ and
Bharatam Jagadeesh‡
Organic Chemistry Division-II and Center for Nuclear Magnetic Resonance,
CSIR-Indian Institute of Chemical Technology, Hyderabad-500 607, India
gsudhakar@iict.res.in
Received August 4, 2011
ABSTRACT
A first total synthesis of acortatarins A, B, and an enantiomer of the proposed structure of acortatarin B is described by using readily
available D-sugars. This convergent total synthesis revealed the revision of the absolute configuration of acortatarin A and structural
revision of acortatarin B. The key steps involved are regioselective epoxide opening with deprotonated 2,5-disubstituted pyrrole and
spiroketalization.
Recently, Hou and co-workers isolated two novel spiro-
cyclic alkaloids, acortatarin A and B (Figure 1),1 with a
naturally unusual morpholine motif from the rhizome of
Acorus tatarinowii, which is used as a traditional Chinese
medicine for treating central nervous system disorders.2
They found that acortatarin A significantly inhibits reac-
tive oxygen species production in high-glucose-stimulated
mesangial cells in a dose- and time-dependent manner. The
relative stereochemistry of these compounds was proposed
on the basis of extensive spectroscopic analysis. In addi-
tion, the relative and absolute configurations of acorta-
tarin A were determined by using X-ray crystallographic
diffraction analysis and Mosher’s method, respectively.
The relative configuration of acortatarin B was assigned
via ROESY experiments and assumed that from the
biogenetic point of view it may have the same absolute
configuration as that of acortatarin A.1
The novel spiroalkaloids contain a unique tricyclic
structural skeleton in which the morpholine motif, a
common pharmacophore present in many inhibitors, is
embodied by a spiro fused sugar ring. The interesting
structure combined with an impressive biological profile
attracted us to develop a synthetic strategy that could be
used to access structural analogues in addition to the
Figure 1. Originally proposed and revised structures of acorta-
tarins A and B. Retrosynthetic analysis for 1 and 20.
† Organic Chemistry Division-II
‡ Center for Nuclear Magnetic Resonance
r
10.1021/ol202121k
Published on Web 09/28/2011
2011 American Chemical Society

larger quantities of natural products required for further
biological studies. We envisioned that the sugar moiety
present in the acortatarins could be accessed from the L-
sugars, but we found that the stereochemistry of acorta-
tarins A (I) and B (II) was misassigned when in the course
ofour total synthesis effortusing readily availableD-sugars
as starting materials to accomplish enantiomers of acorta-
tarins natural products resulted. The revised stereochemi-
cal assignments are shown as acortatarins A (1) and B (2)
(Figure 1).
Our retrosynthetic analysis for 1 and 20 was based on the
spiroketalization of suitably protected ketones 3 and 4,
respectively, followed by deprotection of benzyl ethers
(Figure 1). Ketones 3 and 4 could be prepared by reacting
epoxides 6 and 7, respectively, with 2,5-disubstituted pyr-
role 5 which in turn is, easily, accessible from pyrrole.
Epoxide fragments 6 and 7 could be derived from chiral
pool starting materials 2-deoxy-D-ribose and D-ribose,
respectively.
0.25 equiv of NaBH4. Because of the poor stability of the
diol 8,6 we have synthesized dicarbaldehyde 9 by following
another reported procedure7 and converted it to mono-
aldehyde 10.8 Subsequently, primary hydroxyl group of 10
was protected as THP ether to give the pyrrole fragment 5
in 96% yield.
The synthesis of appropriately protected epoxides 6 and
7 (Scheme 2) was started from the known 3,5-di-O-benzyl-
2-deoxy-D-ribofuranose (11) and 2,3,5-tri-O-benzyl-D-ri-
bofuranose (12), respectively, which in turn were prepared
from their corresponding sugars by following the reported
procedure in three steps.9 The lactols 11 and 12 were
reacted with methylenetriphenylphosphorane to give the
alkenes 1310 (63%) and 149d (31%), respectively, and the
resulting secondary hydroxyl group was transformed to O-
silyl ethers 15 (91%) and 16 (98%) by using TBSOTf in the
presence of 2,6-lutidine. Terminal olefins 15 and 16 were
treated with m-CPBA to give the anticipated epoxides 6
(73%) and 7 (70%), respectively.
Scheme 1. Synthesis of Pyrrole Fragment 5
Scheme 2. Synthesis of Fragments 6 and 7
Initial efforts were focused on the synthesis of the
pyrrole fragment 5 as shown in Scheme 1. The 2,5-bis-
(hydroxymethyl)pyrrole (8) was prepared from pyrrole
according to the known procedure.3 Controlled oxidation
of diol 8 using 1 equiv of MnO2 resulted pyrrole-2,5-
dicarbaldehyde (9) and 5-hydroxymethylpyrrole-2-carbal-
dehyde (10)4 in 29% and 51% yield, respectively. The
dicarbaldehyde 9 was reduced to 105 in 96% yield by using
With both key fragments in hand, after having examined
several conditions,11,8 the deprotonation of substituted
pyrrole12 5 with NaH in DMF followed by N-alkylation
(7) Knizhnikov, V. A.; Borisova, N. E.; Yurashevich, N.Ya.; Popova,
L. A.; Chernyad’ev, A. Yu.; Zubreichuk, Z. P.; Reshetova, M. D. Russ. J.
Org. Chem. 2007, 43, 855–860.
(1) Tong, X. G.; Zhou, L. L.; Wang, Y. H.; Xia, C.; Wang, Y.; Liang,
M.; Hou, F. F.; Cheng, Y. X. Org. Lett. 2010, 12, 1844–1847. (b) Tong,
X. G.; Zhou, L. L.; Wang, Y. H.; Xia, C.; Wang, Y.; Liang, M.; Hou,
F. F.; Cheng, Y. X. Org. Lett. 2011, 13, 4478–4478.
(2) Hu, J. G.; Gu, J.; Wang, Z. W. Pharmacol. Clin. Chin. Mater.
Med. 1999, 15, 19–21.
(3) (a) Taniguchi, S.; Hasegawa, H.; Nishimura, M.; Takahashi, M.
Synlett 1999, 73–74. (b) Taniguchi, S.; Hasegawa, H.; Yanagiya, S.;
Tabeta, Y.; Nakano, Y.; Takahashi, M. Tetrahedron 2001, 57, 2103–
2108.
(8) See the Supporting Information.
(9) (a) Alexander, V.; Choi, W. J.; Chun, J.; Kim, H. O.; Jeon, J. H.;
Tosh, D. K.; Lee, H. W.; Chandra, G.; Choi, J.; Jeong, L. S. Org. Lett.
2010, 12, 2242–2245. (b) Xu, G.; Moeller, K. D. Org. Lett. 2010, 12,
2590–2593. (c) Adamo, M. F. A.; Pergoli, R. Org. Lett. 2007, 9, 4443–
4446. (d) Hekking, K. F. W.; Moelands, M. A. H.; van Delft, F. L.;
Rutjes, F. P. J. T. J. Org. Chem. 2006, 71, 6444–6450. (e) Takahashi, T.;
Hirose, Y.; Iwamoto, H.; Doi, T. J. Org. Chem. 1998, 63, 5742–5743.
(10) Hossain, N.; Blaton, N.; Peeters, O.; Rozenski, J.; Herdewijn,
P. A. Tetrahedron 1996, 52, 5563–5578.
(4) Olsson, K.; Pernemalm, P. A.; Theander, O. Acta Chem. Scand.,
Sect. B 1979, 32, 249–256.
(11) Initial efforts were undertaken to furnish 17a and 18a from 5a
(P = TBS) to avoid complexity in the NMR specrtra of 17 and 18 (P =
THP) because of the diastereomeric mixture from the THP group in
addition to the secondary hydroxyl stereocenter. But 6 or 7 with 5a and
NaH under heating conditions in DMF or THF consistently gave 17a
and 18a in low yields (10ꢀ20%) due to the falling of TBS.
(5) (a) Cin, Y. W.; Lim, S. W.; Kim, S. H.; Shin, D. Y.; Suh, Y. G.;
Kim, Y. B.; Kim, Y. C.; Kim, J. Bioorg. Med. Chem. Lett. 2003, 13, 79–
81. (b) Chiou, W. F.; Shen, C. C.; Yu, H. J.; Chiang, C. H.; Chen, C. C.;
Chang, W.; Don, M. J. Heterocycles 2005, 65, 1215–1220.
(6) The diol 8 became unstable in our hands, leading to quick
polymerization or decompostion to unidentified compounds. Its worth
mentioning that when the diol 8 was converted to monoaldehyde 10 it
attained great stability, indicating that the aldehyde may play a key role
in the stability of C ring of these natural products.
(12) (a) Rudolph, A.; Rackelmann, N.; Savard, M. O. T.; Lautens, M.
J. Org. Chem. 2009, 74, 289–297. (b) Ludwig, J.; Bovens, S.; Brauch, C.;
€
Elfringhoff, A. S.; Lehr, M. J. Med. Chem. 2006, 49, 2611–2620. (c) Bos,
M.; Jenck, F.; Martin, J. R.; Moreau, J. L.; Sleight, A. J.; Wichmann, J.;
Widmer, U. J. Med. Chem. 1997, 40, 2762–2769.
Org. Lett., Vol. 13, No. 20, 2011
5453

via regioselective opening of terminal epoxides 6 and 7 at
55 °C led to the secondary alcohol in each case as a
inseparable diastereomeric mixture of 17 (45%) and 18
(43%), respectively, in moderate yield on the basis of recov-
ered starting material (Scheme 3). However, stereochemistry
of this hydroxyl group is inconsequential as it would be
oxidized to the corresponding ketone in the next step.
magnitude and the same sign as reported for acortatarin
A [[R]27D þ178.4 (c 0.4, MeOH)]. This result surprised us
as the synthesis was planned for the enantiomer of acorta-
tarin A (based on the proposed structure) from the D-
sugar. With this unexpected result, we synthesized and
compared spectral data of MTPA-esters of 18 with the
Mosher esters which were used in determining the absolute
configuration of acortatarin A in the isolation paper.1 We
found that the same absolute configuration of MTPA-Cl
was carried over to the MTPA-ester;17 as a result, the
absolute configuration of acortatarin A was misassigned.1
Thus, the compound synthesized is the natural enantiomer
and the revised absolute configuration of acortatarin A (1)
is C-2 (R), C-3 (S), and C-5 (R) (Figure 1).
Scheme 3. Total Synthesis of Acortatarin A (1) and Enantiomer
of the Proposed Structure of Acortatarin B (20)
In a similar manner, 20 and 20a were independently
subjected to debenzylation to give 20 (75%) and 2a0 (77%).
We expected that the NMR spectra of one of the 20 or 2a0
should be identical with the spectra of acortatarin B,
nevertheless, the synthesis was planned for the enantiomer
of acortatarin B (based on the proposed structure) from
1
D-ribose. Unfortunately, the H and 13C NMR spectral
data of 20 and 2a0 did not match the data reported for
acortatarin B,18,8 indicating that the structure was misas-
signed. The 1H NMR spectrum of 2a0 is similar to the one
reported for acortatarin B except for the major chemical
shift discrepancies at C-2 and C-3 protons. However,
extensive NMR experiments14,8 of 20 and 2a0 established
that 20 had an enantiomeric relationship with the structure
proposed to acortatarin B.
To find out the correct structure of acortatarin B by its
total synthesis, one of each of the possible eight enan-
tiomeric pairs needs to be synthesized as it has four
asymmetric centers. Formation of anomeric mixture at
spirocyclization stage is advantageous as it gives access
to both the diastereomers required for confirmation of
acortatarin B. Thus, the synthesis of another three diaster-
eomers of ketone 4 is, indeed, needed. These three diaster-
eomerscan be accessedfrom the sameD-sugarseries, arabi-
nose, lyxose, and xylose, as 4 was already derived from
D-ribose.
We have started with D-xylose owing to the major
chemical shift differences at C-2 and C-3 protons in the
1H NMR spectrum of 2a0 and acortatarin B. The lactol
21 (Scheme 4), synthesized from D-xylose following the
known procedure,19 was treated with methylenetriphenyl-
phosphorane to give alkene 22.20 Transformation of free
alcohol to O-silyl ether followed by epoxidation of the
terminal olefin with m-CPBA resulted epoxide 23 in 85%
yield over two steps. The key reaction between pyrrole
Oxidation of the secondary hydroxyl group of 17 and
18 with DMP13 furnished ketones 3 and 4 each in 88%
yield. Treatment of 3 and 4 with PTSA in CH2Cl2
effected, as expected, deprotection of the THP and TBS
groupsandsimultaneous intramolecular spiroketalization
to give chromatographically separable mixture of anom-
ers at the C-5 position. The ratio is 1.4:1 in the case of 19
and 19a in 75% combined yield and 1:1.3 for 20 and 20a in
70% combined yield (Scheme 3).14,8 Anomers 19 and 19a
were independently subjected to deprotection15 of benzyl
groups by using 1 M TiCl4 that accomplished,16 interest-
ingly, in both cases again anomeric mixture 1 and 1a in the
ratio of 9:1 in 80% combined yield. Comparison of the 1H
and 13C NMR spectra revealed that the 1 (major) repre-
sented the desired natural product acortatarin A.
However, the specific rotation of the synthesized com-
pound [[R]27 þ191.4 (c 0.27, MeOH)] is of similar
D
(17) (a) Hoye, T. R.; Jeffrey, C. S.; Shao, F. Nat. Protoc. 2007, 2,
2451–2458. (b) Joshi, B. S.; Pelletier, S. W. Heterocycles 1999, 51, 183–
184.
(18) We were unsuccessful in obtaining crystals of 2a0 (derived from
major isomer 20a) or its bromobenzoyl derivative.
(19) (a) Barker, R.; Fletcher, H. G., Jr. J. Org. Chem. 1961, 26, 4605–
4609. (b) Tejima, S.; Fletcher, H. G., Jr. J. Org. Chem. 1963, 28, 2999–
3004.
(20) Yu, C. Y.; Huang, M. H. Org. Lett. 2006, 8, 3021–3024.
(21) Pearson, W. H.; Hines, J. V. J. Org. Chem. 2000, 65, 5785–5793.
(22) Maryanoff, B. E.; Nortey, S. O.; Inners, R. R.; Campbell, S. A.;
Reitz, A. B.; Liotta, D. Carbohydr. Res. 1987, 171, 259–278.
(13) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155–4156.
(b) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277–7287.
(14) Compounds 19, 19a, and all spirocyclic compounds at the final
stage have been fully characterized by using 2D NMR experiments.
(15) Our attempts to deprotect benzyl groups by conventional hydro-
genolysis in the presence of Pd/C resulted no reaction or at higher
catalyst loading led to complex mixture of compounds.
(16) (a) Srihari, P.; Kumaraswamy, B.; Bhunia, D. C.; Yadav, J. S.
Tetrahedron Lett. 2010, 51, 2903–2905. (b) Gurjar, M. K.; Nagaprasad,
R.; Ramana, C. V. Tetrahedron Lett. 2003, 44, 2873–2875.
5454
Org. Lett., Vol. 13, No. 20, 2011

that neither 27 nor 27a represented the correct structure of
acortatarin B.
Scheme 4. Synthesis of 27, 27a, 34, and Total Synthesis of
Acortatarin B (2)
The additional hydroxyl group at C-4 stereocenter in
acortatarin B, which only differs from acortatarin A,
suggested that D-arabinose could be the better choice of
starting material than D-lyxose. Accordingly, terminal
alkene 2921 (Scheme 4) was subjected to a similar sequence
of reactions such as TBS protection22 and epoxidation to
deliver 30 in 70% yield over two steps. N-Alkylation of
the deprotonated pyrrole 5 with epoxide 30 resulted in
31 which was oxidized to the ketone 32 in 37% yield
over two steps. Exposure of the ketone to PTSA resulted
separable mixture of anomers 33 and 33a in the ratio
4.6:1 in 65% combined yield. Subsequent benzyl groups
deprotection of both the anomers independently resulted
34 (75%) and 2 (78%). To our pleasure and surprise the 1H
and 13C NMR spectra of 2, obtained from minor anomer
33a, are identical with the one reported for acortatarin B.
The specific rotation [[R]27D ꢀ94.4 (c 0.04, MeOH); lit.1b
[R]27 ꢀ92.7 (c 0.10, MeOH)] indicated the compound
D
synthesized is the natural enantiomer. The stereochemistry
at C-5 was established by using ROESY experiments and
found cross peaks H-3TH-15, H-4TH-10, and H-7TH-
15; hence, the revised structure and absolute configuration
of acortatarin B are as shown in Scheme 4.
In conclusion, the first total synthesis of acortatarins A,
B and an enantiomer of the proposed structure of acorta-
tarin B revealed the revision of stereochemical assignment
and suggested D-sugars could be the biosynthetic precur-
sors for these natural products. In addition, the developed
synthetic strategy is short, efficient, and practically appli-
cable, which is apparent from the synthesis of various
diastereomers to unambiguously reassign the stereochem-
istry of acortatarin B. The synthesis of further acortatarin
analogues and their biological evaluation is in progress in
our laboratory and will be reported in due course.
Acknowledgment. V.D.K., S.B., and G.P. thank the
Council of Scientific and Industrial Research (CSIR),
New Delhi, India, for the award of research fellowship,
and G.S. thanks Dr. V. V. N. Reddy, Head, Org-II, Dr. J.
S. Yadav, Director, IICT, and Dr. T. K. Chakraborty,
Director, CDRI, for their encouragement and support.
fragment 5 and epoxide 23 was executed under the same
conditions to give 24, which was oxidized to 25 in 37%
yield over two steps. Subsequent treatment with PTSA
resulted chromatographically separable anomers at C-5
position 26 and 26a in the ratio 1.6:1 in 63% combined
yield. Finally, cleavage of benzyl ethers of both the anom-
ers independently furnished 27 (75%) and 27a (76%).
Disappointingly, comparison of the NMR spectra of 27
and 27a with the one reported for acortatarin B revealed
Supporting Information Available. Experimental pro-
1
cedures and copies of H and 13C NMR spectral data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Org. Lett., Vol. 13, No. 20, 2011
5455