Total Synthesis of (?)-Histrionicotoxin through a Stereoselective Radical Translocation–Cyclization Reaction

Article abstract of DOI:10.1002/anie.201609941

Stereoselective total syntheses of (?)-histrionicotoxin and (?)-histrionicotoxin 235A are described. The 1-azaspiro[5.5]undecane skeleton was constructed diastereoselectively by a radical translocation–cyclization reaction involving a chiral cyclic acetal; the use of tris(trimethylsilyl)silane was crucial for the high diastereoselectivity. The cyclization product was converted into (?)-histrionicotoxin 235A through a one-pot partial-reduction–allylation reaction of a derivative containing an unprotected lactam. Finally, two terminal alkenes were transformed into enynes with the 1,3-amino alcohol protected as an oxathiazolidine oxide to complete the total synthesis of (?)-histrionicotoxin.

Full text of DOI:10.1002/anie.201609941

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201609941
German Edition: DOI: 10.1002/ange.201609941
Natural Products Synthesis
Total Synthesis of (ꢀ)-Histrionicotoxin through a Stereoselective
Radical Translocation–Cyclization Reaction
Manabu Sato+, Hiroki Azuma+, Akihiro Daigaku+, Sota Sato, Kiyosei Takasu, Kentaro Okano,
and Hidetoshi Tokuyama*
Abstract: Stereoselective total syntheses of (ꢀ)-histrionico-
toxin and (ꢀ)-histrionicotoxin 235A are described. The 1-
azaspiro[5.5]undecane skeleton was constructed diastereose-
lectively by a radical translocation–cyclization reaction involv-
ing a chiral cyclic acetal; the use of tris(trimethylsilyl)silane
was crucial for the high diastereoselectivity. The cyclization
product was converted into (ꢀ)-histrionicotoxin 235A through
a one-pot partial-reduction–allylation reaction of a derivative
containing an unprotected lactam. Finally, two terminal
alkenes were transformed into enynes with the 1,3-amino
alcohol protected as an oxathiazolidine oxide to complete the
total synthesis of (ꢀ)-histrionicotoxin.
been reported previously on the basis of a variety of strategies
for the construction of the azaspirobicyclic system. However,
no strategy based on radical cyclization[6,7] has been reported
so far for the stereocontrolled construction of the spirocyclic
system in an optically active form. Herein we report the total
synthesis of (ꢀ)-HTX (1) and (ꢀ)-HTX 235A (2) through the
stereoselective construction of the 1-azaspiro[5.5]undecane
core by a diastereoselective radical translocation–cyclization
cascade.
We planned to synthesize (ꢀ)-HTX (1) from (ꢀ)-
HTX 235A (2)[4a,b] by conversion of the two terminal olefins
into the enyne side chains (Scheme 2).[4] We decided to
H
istrionicotoxin (1, HTX), isolated by Daly, Witkop, and
co-workers from the poison-arrow frog Dendrobates histri-
onicus, possesses activity as a noncompetitive blocker of
nicotinic acetylcholine receptors (Scheme 1).[1,2] The struc-
Scheme 1. Structures of HTX (1) and HTX 235A (2).
ture of 1, consisting of a 1-azaspiro[5.5]undecane core with
two acid-labile Z enyne side chains, has attracted consider-
able interest from the synthetic community as a platform to
demonstrate synthetic strategies for the construction of an
azaspirobicyclic system.[3] Three total syntheses of optically
active HTX (1)[4] and two total syntheses of racemic 1[5] have
Scheme 2. Retrosynthetic analysis of HTX (1) and HTX 235A (2).
Bn=benzyl.
[*] M. Sato,[+] Dr. H. Azuma,[+] A. Daigaku,[+] Prof. Dr. K. Takasu,
Dr. K. Okano, Prof. Dr. H. Tokuyama
Graduate School of Pharmaceutical Sciences, Tohoku University
Aoba 6-3, Aramaki, Aoba-ku, Sendai 980-8578 (Japan)
E-mail: tokuyama@mail.pharm.tohoku.ac.jp
introduce the allyl group diastereoselectively onto the d-
lactam 5 by developing a one-pot partial-reduction–allylation
protocol via imine 4. For the construction of the 1-azaspiro-
[5.5]undecane core, we planned to apply the radical translo-
cation–cyclization reaction developed in our group[8] to d-
lactam 9 bearing an a,b-unsaturated ester side chain. Thus, sp2
radical 8, generated from iodide 9 under radical conditions,
should undergo 1,5-radical translocation to provide a more
stable sp3 radical 7.[8] Then, 6-exo-trig cyclization via a chair-
like conformation 7 bearing an equatorial OP group should
proceed to furnish an azaspirobicyclic compound 6 possessing
Dr. S. Sato
ERATO, Isobe Degenerate p-Integration Project and Advanced
Institute for Materials Research (AIMR)
2-1-1 Katahira, Aoba-ku, Sendai, 980-8577 (Japan)
and
AIMR and Department of Chemistry, Tohoku University
Aoba 6-3, Aramaki, Aoba-ku, Sendai 980-8578 (Japan)
[+] These authors contributed equally.
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201609941.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
the desired stereostructure, including the C6 spiro center, the
substituent at the C7 position, and the hydroxy group at the
C8 position. Substrate 9 for the radical reaction would be
readily assembled by conventional reactions starting from
(+)-(R)-glycidol (10).
We initiated the project by the stereoselective preparation
of substrate 11 from (+)-(R)-glycidol (10)[9] and subjected 11
to radical cyclization conditions (Scheme 3). Although the
expected product 12 of cyclization by the 6-exo-trig mode was
obtained as a single diastereomer,[10] the yield was just 23%.
Scheme 4. Working hypothesis for the radical reaction of a substrate
containing a chiral acetal.
and 6-exo-trig cyclization via the sterically favorable con-
formation.
With this idea in mind, we developed a facile and scalable
assembly of substrate 26 bearing a chiral cyclic acetal
(Scheme 5). Reductive amination[12] of the readily available
ketodiester 21 and benzylamine 22[13] gave d-lactam 23, which
was then converted into aldehyde 24 through reduction and
oxidation of the resultant primary alcohol. Subsequent 1,2-
addition of a vinyl Grignard reagent, followed by cross-
metathesis with methyl acrylate afforded the unsaturated
ester 25. Finally, oxidation with MnO2 and acetal formation of
the resultant enone with (+)-(2S,4S)-pentanediol by the
Noyori protocol[14] provided 26 as a 1:1 diastereomeric
mixture at the C6 position.
Scheme 3. Initial attempts at the radical translocation–cyclization
reaction. AIBN=azobisisobutyronitrile, TBDPS=tert-butyldiphenylsilyl.
Having synthesized the requisite substrate 26 bearing
a chiral cyclic acetal, we tested the radical cyclization.
Treatment with nBu3SnH in benzene at reflux afforded the
We also examined the reaction of 14, which is the C6 epimer
of 11. Interestingly, the reaction of 14 also afforded product 12
as a single diastereomer, thus indicating that inversion of
configuration occurred after 1,5-radical translocation, and
subsequent 6-exo-trig cyclization proceeded via 13 to avoid
1,3-diaxial interactions in 16. It can be concluded that the
configuration of the spiro center and the C7 center bearing
the side chain was only controlled by the stereogenic C8
center and did not depend on the configuration of the C6
position of the substrates. We reasoned that the low yield was
due to an undesired second 1,5-radical translocation to
provide an O-stabilized allyl radical, which would decompose
during the reaction.
On the basis of these results, we revised the strategy for
the construction of the 1-azaspiro[5.5]undecane core
(Scheme 4). To prevent the undesired 1,5-radical transloca-
tion while maintaining the chiral environment, we designed
a substrate of type 17 bearing a chiral cyclic acetal[11] in the
side chain. We hoped that radical cyclization of a substrate
containing a chiral cyclic acetal derived from (+)-(2S,4S)-
pentanediol or its enantiomer would proceed via a stable
chair-like conformation 19a to furnish the desired azaspiro-
bicyclic product 20a. According to the results with substrates
11 and 14 (Scheme 3), we expected that a 1:1 mixture of
diastereomers with respect to the C6 position would provide
one product through rapid inversion of the sp3 radical center
Scheme 5. Preparation of chiral acetal 26 for the key reaction.
K-selectride=potassium tri-sec-butylborohydride, TEMPO=2,2,6,6-tet-
ramethylpiperidine 1-oxyl, TMS=trimethylsilyl, Tf=trifluoromethane-
sulfonyl.
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Diastereoselectivity of the radical translocation–cyclization
reaction of substrate 26 containing a chiral acetal.
Entry
Reagent
Yield [%]
27/28[a]
1
2
3
nBu3SnH
Ph3SnH
(TMS)3SiH
79
84
3:1
1.7:1
38:1
68 (66[b])
1
[a] The ratio was determined on the basis of H NMR spectroscopy.
[b] The reaction was carried out on a 13 g scale.
Scheme 6. Rationale for the observed diastereoselectivity and stereo-
chemical inversion.
cyclization product 27 in good yield with moderate diaste-
reoselectivity (Table 1, entry 1). The structure of product 27
was determined unambiguously after conversion into the
bromide 29. X-ray crystallographic analysis revealed that 29
possessed the desired absolute and relative configuration for
the synthesis of natural 1 and 2 (Figure 1). The configuration
of one acetal methyl group of the major product was inverted
in the reaction,[15] whereas that of the minor diastereomer was
retained (see below). Furthermore, we observed significant
effects of the H-atom source on the diastereoselectivity. Thus,
a reaction with Ph3SnH gave a mixture of 27 and 28 in a lower
diastereomeric ratio (Table 1, entry 2). Surprisingly, the ratio
was dramatically improved when using bulky (TMS)3SiH
(entry 3). Eventually, we found that we could carry out the
reaction on more than a 10 g scale with (TMS)3SiH.
high diastereoselectivity of the reaction with less active[16] and
bulky (TMS)3SiH could be due to smooth H-atom transfer to
the radical species C, which is more readily accessible than
other radical species in the equilibrium between A and E. In
contrast, H-atom transfer from active[16] and sterically less
hindered H-atom sources would be less selective.
The facial selectivity of the 6-exo-trig radical cyclization
was determined after removal of the acetal moiety
(Scheme 7).[15] Deacetalization was conducted by the treat-
ment of 27 with FeCl3 and silica gel in acetone[17] to provide
keto lactam 30. Surprisingly, compound 30 showed an
Scheme 7. Removal of the acetal moiety to determine facial selectivity.
excellent ee value (94% ee), which was independent of the
H-atom source. Although we currently do not have a clear
explanation for this excellent selectivity, it is conceivable that
facial selectivity is controlled by the relative rate of the 6-exo-
trig cyclization step and is independent of the final H-
abstraction step for two reasons (Scheme 8). One is that the
high ee value was independent of the diastereomeric ratio
between 27 and 28. The second is that after configurational
inversion of the chiral acetal moiety through the second 1,5-
radical translocation (see above) of d, the minor product
formed by the opposite facial selectivity became the enantio-
mer of the major product 27. Therefore, there should be no
rate difference for the terminal H-abstraction steps. As for the
cyclization step, cyclization via the conformer a would be
favorable since there would be steric repulsion between one
methyl group on the cyclic acetal moiety and a hydrogen atom
on the chain in conformer c, which leads to ent-27.
Figure 1. X-ray crystallographic structure of the azaspirocyclic com-
pound 29, which was prepared from 27 by bromination of the benzene
ring.
The dependence of the diastereoselectivity on the H-atom
source and the stereochemical inversion of the acetal moiety
of the major isomer can be rationalized as shown in Scheme 6.
The minor isomer 28 would be formed by H abstraction from
the relatively hindered radical species D or E. On the other
hand, the major product 27 would be obtained by H abstrac-
tion from the readily accessible radical species C, which would
be formed by conformational flipping of the acetal ring of
intermediate A to B and subsequent 1,5-radical translocation
and reflipping of the acetal moiety. We considered that the
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Finally, we completed the total synthesis of (ꢀ)-HTX (1)
by conversion of the two terminal alkenes into Z enyne chains
(Scheme 9).[4] According to the past synthetic studies on
(ꢀ)-HTX (1) and our preliminary studies, protection of the
secondary amine of the azaspirobicyclic system was a difficult
task, as was the elongation of the two terminal alkenes
without protection of the nitrogen atom.[4a] We overcame this
problem by using thionyl chloride for the problematic
protection of the nitrogen atom. Thus, the treatment of (ꢀ)-
HTX 235A (2) with thionyl chloride afforded oxathiazolidine
oxide 34 as a 4.7:1 mixture of diastereomers. The two terminal
alkenes were then converted into enyne side chains according
to the protocol developed by the research groups of Stork[4a,24]
and Holmes.[4b] After separation of the two isomers,[25] the
major isomer 34[26] was ozonized to provide a dialdehyde,
which was subjected to Z-selective Wittig olefination fol-
lowed by Sonogashira coupling[4,27] with trimethylsilylacety-
lene to provide 35 with two Z enyne units. Finally, the
protecting groups of the amino alcohol and the terminal
acetylene were removed with LiAlH4 to complete the total
synthesis of (ꢀ)-HTX (1).[4b,c]
Scheme 8. Rationale for the facial selectivity of the 6-exo cyclization.
Having established a highly stereoselective and scalable
method for the construction of the azaspirobicyclic skeleton,
we focused our attention on the introduction of two enyne
side chains (Scheme 9). The optical purity of 30 was increased
to > 99% ee by recrystallization. Reduction of ketoester 30[5a]
In conclusion, we have completed the total synthesis of
(ꢀ)-HTX (1) and (ꢀ)-HTX 235A (2) through a strategy
featuring a highly stereoselective
radical translocation–cyclization
cascade for the construction of
the 1-azaspiro[5.5]undecane skel-
eton. For the late-stage construc-
tion of the two enyne side chains,
we developed a one-pot partial
reduction–allylation of an unpro-
tected lactam as well as a unique
protecting group for the amino
alcohol.
Scheme 9. One-pot allylation of the d-lactam and total synthesis of (ꢀ)-HTX (1) and (ꢀ)-HTX 235A (2).
Reagents and conditions: a) Li, liq. NH3, tBuOH/THF, ꢀ788C; b) 2-nitrophenyl selenocyanate, nBu3P,
THF, room temperature, 44% (2 steps); c) p-TsOH·H2O, toluene, reflux; d) TBDPSCl, imidazole, CH2Cl2,
room temperature, 75% (2 steps); e) m-CPBA, CH2Cl2, 08C!RT, 83%; f) Ti(OiPr)4, Et2SiH2, THF,
reflux; allylmagnesium chloride, cat. ZnCl2, ꢀ788C!RT, 71%; g) TBAF, THF, reflux, 93%; h) SOCl2,
imidazole, CH2Cl2, 08C, 83% (d.r. 4.7:1); i) O3; Me2S, CH2Cl2/MeOH, ꢀ788C, 76%; j) iodomethyltriphe-
nylphosphonium iodide, KHMDS, THF, ꢀ788C, 40%; k) trimethylsilyl acetylene, cat. [Pd(PPh3)4], CuI,
Et2NH, room temperature, 77%; l) LiAlH4, THF, 08C!RT, 49%. p-Ts=p-toluenesulfonyl, m-CPBA=m-
chloroperoxybenzoic acid, KHMDS=potassium bis(trimethylsilyl)amide, TBAF=tetrabutylammonium
fluoride.
Acknowledgements
This research was supported
financially by the Cabinet
Office, Government of Japan
through its “Funding Program
for Next-Generation World-
Leading Researchers” (LS008),
JSPS
KAKENHI
Grants
gave the corresponding diol as a single diastereomer, which
was converted into selenide 31.[18] A sequence comprising the
removal of the 3,4-dimethoxybenzyl group under acidic
conditions, silylation of the secondary alcohol, and selenoxide
elimination[19] then gave spirolactam 32. At this point, we
established a one-pot partial-reduction–allylation protocol
for the unprotected lactam. Thus, cyclic imine 33 was obtained
by modification of the protocol described by Buchwald and
co-workers.[20] For the subsequent allylation in the one-pot
JP16H01127 in Middle Molecular Strategy and JP16H00999
in Precisely Designed Catalysts with Customized Scaffolding,
and a Grant-in-Aid for Scientific Research (A) (26253001).
We are grateful to Professor Hiroyuki Isobe (Tohoku
University) for determining the structure of intermediates
29 and 34.
Conflict of interest
process, we found that a Grignard reagent in the presence of
[21]
catalytic ZnCl2
was effective. The desired product was
The authors declare no conflict of interest.
obtained in 71% yield as a single diastereomer.[22] The
configuration of the allylation product was established by
conversion into (ꢀ)-HTX 235A (2).[23]
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[11] For an example with a chiral diol, see: D. L. Comins, Y. Zhang,
Keywords: alkaloids · diastereoselectivity · radical cyclization ·
X. Zheng, Chem. Commun. 1998, 2509 – 2510.
[12] A. F. Abdel-Magid, B. D. Harris, C. A. Maryanoff, Synlett 1994,
81 – 83.
spiro compounds · total synthesis
[13] The 2-bromo-4,5-dimethoxybenzyl group was used to enable
deprotection of the lactam nitrogen atom under milder con-
ditions.
[14] T. Tsunoda, M. Suzuki, R. Noyori, Tetrahedron Lett. 1980, 21,
1357 – 1358.
[15] Owing to the inversion of configuration, the cyclic acetal moiety
lost its chirality. Therefore, the product obtained through the
opposite facial selectivity of the cyclization became the enan-
tiomer of 27. Thus, the facial selectivity could not be determined
on the basis of the diastereomeric ratio. Furthermore, the
ee value of 27 could not be determined at this point because of
difficulties in synthesizing racemic 27.
[16] a) D. J. Carlsson, K. U. Ingold, J. Am. Chem. Soc. 1968, 90, 7047 –
7055; b) C. Chatgilialoglu, J. Dickhaut, B. Giese, J. Org. Chem.
1991, 56, 6399 – 6403.
[17] K. S. Kim, Y. H. Song, B. H. Lee, C. S. Hahn, J. Org. Chem. 1986,
51, 404 – 407.
[18] P. A. Grieco, S. Gilman, M. Nishizawa, J. Org. Chem. 1976, 41,
1485 – 1486.
[19] K. B. Sharpless, M. W. Young, J. Org. Chem. 1975, 40, 947 – 949.
[20] S. Bower, K. A. Kreutzer, S. L. Buchwald, Angew. Chem. Int. Ed.
Engl. 1996, 35, 1515 – 1516; Angew. Chem. 1996, 108, 1662 – 1664.
[21] M. Hatano, S. Suzuki, K. Ishihara, J. Am. Chem. Soc. 2006, 128,
9998 – 9999.
[22] E. J. Corey, J. F. Arnett, G. N. Widiger, J. Am. Chem. Soc. 1975,
97, 430 – 431.
[23] T. F. Spande, H. M. Garraffo, J. W. Daly, T. Tokuyama, A.
Shimada, Tetrahedron 1992, 48, 1823 – 1836.
[24] G. Stork, K. Zhao, Tetrahedron Lett. 1989, 30, 2173 – 2174.
[25] The minor isomer was recycled by conversion into compound 2
with LiAlH4 in 86% yield (see the Supporting Information for
details).
[26] The structure of the major isomer was determined by X-ray
crystallographic analysis (see the Supporting Information).
[27] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975,
16, 4467 – 4470.
[1] J. W. Daly, I. Karle, C. W. Myers, T. Tokuyama, J. A. Waters, B.
Witkop, Proc. Natl. Acad. Sci. USA 1971, 68, 1870 – 1875.
[2] a) K. Takahashi, B. Witkop, A. Brossi, M. A. Maleque, E. X.
Albuquerque, Helv. Chim. Acta 1982, 65, 252 – 261; b) W.
Gessner, K. Takahashi, B. Witkop, A. Brossi, E. X. Albuquer-
que, Helv. Chim. Acta 1985, 68, 49 – 55.
[3] For a review, see: A. Sinclair, R. A. Stockman, Nat. Prod. Rep.
2007, 24, 298 – 326.
[4] For the total synthesis of (ꢀ)-HTX (1), see: a) G. Stork, K. Zhao,
J. Am. Chem. Soc. 1990, 112, 5875 – 5876; b) G. M. Williams,
S. D. Roughley, J. E. Davies, A. B. Holmes, J. Am. Chem. Soc.
1999, 121, 4900 – 4901; c) Y. Adachi, N. Kamei, S. Yokoshima, T.
Fukuyama, Org. Lett. 2011, 13, 4446 – 4449.
[5] For syntheses of (ꢁ)-HTX (1), see: a) S. C. Carey, M. Aratani, Y.
Kishi, Tetrahedron Lett. 1985, 26, 5887 – 5890; b) M. S. Karatho-
luvhu, A. Sinclair, A. F. Newton, M.-L. Alcaraz, R. A. Stock-
man, P. L. Fuchs, J. Am. Chem. Soc. 2006, 128, 12656 – 12657.
[6] a) S. Van der Jeught, N. De Vos, K. Masschelein, I. Ghiviriga,
C. V. Stevens, Eur. J. Org. Chem. 2010, 5444 – 5453; b) H. Zhang,
E. B. Hay, S. J. Geib, D. P. Curran, J. Am. Chem. Soc. 2013, 135,
16610 – 16617; c) D.-S. Hsu, C.-H. Chen, C.-W. Hsu, Eur. J. Org.
Chem. 2016, 589 – 598.
[7] a) N. A. Porter, B. Lacher, V. H.-T. Chang, D. R. Magnin, J. Am.
Chem. Soc. 1989, 111, 8309 – 8310; b) D. P. Curran, W. Shen, J.
Zhang, T. A. Heffner, J. Am. Chem. Soc. 1990, 112, 6738 – 6740;
c) Q. Zhang, R. M. Mohan, L. Cook, S. Kazanis, D. Peisach,
B. M. Foxman, B. B. Snider, J. Org. Chem. 1993, 58, 7640 – 7651;
d) M. Nishida, E. Ueyama, H. Hayashi, Y. Ohtake, Y. Yamaura,
E. Yanaginuma, O. Yonemitsu, A. Nishida, N. Kawahara, J. Am.
Chem. Soc. 1994, 116, 6455 – 6456.
[8] a) K. Takasu, H. Ohsato, M. Ihara, Org. Lett. 2003, 5, 3017 –
3020; b) M. Fujitani, M. Tsuchiya, K. Okano, K. Takasu, M.
Ihara, H. Tokuyama, Synlett 2010, 822 – 826.
[9] For details of the preparation of 11 and 14, see the Supporting
Information.
[10] The structure of cyclization product 12 was determined by an
NOE experiment (see the Supporting Information).
Manuscript received: October 11, 2016
Final Article published: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Natural Products Synthesis
M. Sato, H. Azuma, A. Daigaku, S. Sato,
K. Takasu, K. Okano,
H. Tokuyama*
&&&& — &&&&
Total Synthesis of (ꢀ)-Histrionicotoxin
through a Stereoselective Radical
Translocation–Cyclization Reaction
A powerful protection strategy enabled
the efficient synthesis of (ꢀ)-histrionico-
toxin. The diastereoselective construction
of the 1-azaspiro[5.5]undecane skeleton
by a radical translocation–cyclization
reaction was made possible by a chiral
cyclic acetal (see scheme), whereas the
unprotected lactam underwent reduction
and allylation, and the alkene groups were
converted into enynes following novel
protection of the 1,3-amino alcohol as an
oxathiazolidine oxide.
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!