Regioselective single pot C3-glycosylation of strophanthidol using methylboronic acid as a transient protecting group

Article abstract of DOI:10.1038/s41429-019-0172-1

This manuscript describes a single pot protocol for the selective introduction of unprotected sugars to the C3 position of the cardiotonic steroid strophanthidol. These reactions proceed with high levels of regiocontrol (>20:1 rr) in the presence of three other hydroxyl functionalities including the C19 primary hydroxyl group and could be applied to different sugars to provide the deprotected cardiac glycosides upon work up (5 examples, 77–69% yield per single operation). The selective glycosylation of the less reactive C3 position is accomplished by the use of traceless protection with methylboronic acid that blocks the C5 and C19 hydroxyls by forming a cyclic boronic ester, followed by in situ glycosylation and a work up with ammonia in methanol to remove the boronic ester and the carbohydrate ester protecting groups.

Full text of DOI:10.1038/s41429-019-0172-1

The Journal of Antibiotics
https://doi.org/10.1038/s41429-019-0172-1
SPECIAL FEATURE: ARTICLE
Regioselective single pot C3-glycosylation of strophanthidol using
methylboronic acid as a transient protecting group
1
Jia-Hui Tay1 Valentin Dorokhov2 Sibin Wang1 Pavel Nagorny
●
●
●
Received: 23 January 2019 / Revised: 9 February 2019 / Accepted: 18 February 2019
© The Author(s), under exclusive licence to the Japan Antibiotics Research Association 2019
Abstract
This manuscript describes a single pot protocol for the selective introduction of unprotected sugars to the C3 position of the
cardiotonic steroid strophanthidol. These reactions proceed with high levels of regiocontrol (>20:1 rr) in the presence of
three other hydroxyl functionalities including the C19 primary hydroxyl group and could be applied to different sugars to
provide the deprotected cardiac glycosides upon work up (5 examples, 77–69% yield per single operation). The selective
glycosylation of the less reactive C3 position is accomplished by the use of traceless protection with methylboronic acid that
blocks the C5 and C19 hydroxyls by forming a cyclic boronic ester, followed by in situ glycosylation and a work up with
ammonia in methanol to remove the boronic ester and the carbohydrate ester protecting groups.
Introduction
neural outgrowth differentiative [17], anti-inflammatory [18],
and anti-hypertensive [19], cardiotonic steroids have been
evaluated as potential therapeutic agents in a variety of recent
studies. An overwhelming amount of data (including several
clinical trials) indicates that cardiotonic steroids hold great
potential as the therapeutics for treating cancer [20]. It is
suggested that the steroid interaction with Na+/K+-ATPase
results in the activation of various intracellular signaling
cascades that culminates in different types of cell death. It is
known that cardenolides can induce apoptosis, autophagy,
anoikis, and immunogenic cell death depending on the cell
type and steroid chemical features [21–23].
Strophanthidol (4a) glycosides have been isolated from
various natural sources, and are among the most potent
cardenolide-based anticancer agents with the in vitro
activity in low nM range for a variety of different human
cancer cells (cf. Figure 1b) [24–26]. The sugar moiety of
these compounds plays an important role in defining their
anticancer activity, and without the sugar moiety, aglycone
4a is significantly less active. In addition to α-L-rhamno-
side 4b, other potent strophanthidol glycosides such as 4c
have been identified and investigated. These observations
as well as other prior studies [27–29] suggest that further
optimization of the sugar moiety of strophanthidol (4a)
glycosides may lead to even more potent analogs. How-
ever, the glycosylation of polyoxygenated cardiotonic
steroids is not trivial, primarily due to the regioselectivity
issues. In most syntheses of cardiac glycosides, multistep
protection/deprotection sequence is commonly used to
Cardiotonic steroids represent a large family of natural pro-
ducts with a diverse range of biological activities [1–7].
These molecules have been considered of great importance
for human and animal health due to their involvement in the
regulation of such crucial functions as renal sodium transport
and arterial pressure, as well their roles in modulation of cell
growth, differentiation, apoptosis, fibrosis, immunity and
carbohydrate metabolism, and the control of various central
nervous functions and behavior [8–13]. Due to their wide
spectrum of biological activities including cardiotonic,
anticancer [14], anti-viral [15], immunoregulatory [16],
Dedication: This article is dedicated to Professor Samuel J.
Danishefsky for his great contributions to the areas of total synthesis of
complex natural products and glycoscience.
Supplementary information The online version of this article (https://
doi.org/10.1038/s41429-019-0172-1) contains supplementary
material, which is available to authorized users.
* Pavel Nagorny
nagorny@umich.edu
1
Department of Chemistry, University of Michigan, 930N.
University Avenue, Ann Arbor, MI 48109, USA
2
Higher College of the Russian Academy of Sciences, Mendeleev
University of Chemical Technology, Miusskaya Square 9,
Moscow, RU 125047, Russia

J.-H. Tay et al.
Fig. 1 Examples of cardenolides
ensure the selective delivery of a sugar unit at the desired
C3 position.
Taylor, and others have previously shown the utilities of
various aryl boronic acids as protecting groups in the
synthesis of polysaccharides [36–39]. Beyond the synthesis
of simple carbohydrate derivatives, boronic acid has not
been widely used to introduce sugar moiety into natural
products. One of the few examples was reported by Pis and
coworkers in 1994 (Scheme 2) [40]. In their synthesis of 20-
hydroxyecdysone glycosides, phenylboronic acid was used
to selectively protect the vicinal-diol moiety of the natural
product 11 to provide ester 12. Thus formed phenylboronic
acid ester was isolated and subjected to glycosylation
resulting in the mixture of products 13 and 14 with intact
phenylboronic acid ester. Recently, our group developed a
one-pot regioselective glycosylation procedure that allows
selective glycosylation of the C11 hydroxyl of 14-
membered macrolactone such as 6-deoxy-erythronolide B
(6-dEB) (15) [41]. The solution of 6-dEB was treated with
phenylboronic acid that served as the transient protecting
group for the 1,3-diol moiety of the macrolactone. The
in situ formation of the boronic acid ester 16 was followed
by the addition of the glycosyl donor and promoter that
provided glycoside 17, and the subsequent aqueous work up
was used to cleave the boronic acid ester and provide the
C11 glycosylated product 18 in a single operation. Having
long-term interests in improving the synthesis of cardiotonic
steroids [34, 41–45] and building on these prior studies, we
surmised that this strategy might be particularly useful for
the glycosylation of steroids 1–4a, and, in particular, stro-
phanthidol (4a). This molecule possesses hydroxyl group at
This is certainly the case for the C19-hydroxylated
steroids such as strophanthidol (4a), sarmentologenin (3), or
ouabagenin (1) [30–35]. The primary C19-hydroxyl func-
tionality of such steroids can undergo competitive glyco-
sylation in the presence of the unprotected C3 hydroxyl, and
the products of C19 glycosylation (i.e., 9, Scheme 1) or bis-
glycosylation (i.e., 6 and 10, Scheme 1) are typically
observed [30, 34]. To circumvent this, strategies requiring
C19 protection, glycosylation and following deprotection
have been developed [34, 35]. However, these strategies
require careful selection of the C19 protecting group as the
competitive decomposition of the butenolide moiety can be
observed upon deprotection [34]. Altogether a more effi-
cient strategy for the direct introduction of the sugar moiety
at the C3 position of these molecules is highly desired. To
streamline the synthesis of glycosylated strophanthidol (4a)
analogs as well as other C3-glycosylated cardiotonic ster-
oids possessing C3/C5/C19 hydroxylation pattern, we set
out to develop a direct single pot method that is based on
the use of boronic acid esters as the traceless protecting
group.
The key design element present in our new protocol is
the use of boronic acid as the transient protecting group for
the C5/C19 diol moiety. One of the biggest advantages of
using boronic acid as a protecting group is the operational
simplicity and mild conditions associated with the instal-
lation and removal of the boronic acid esters. Madsen, Kaji,

Regioselective single pot C3-glycosylation of strophanthidol using methylboronic acid as a transient. . .
Scheme 1 Challenges with the regioselective glycosylation of C19-hydroxylated steroids
Ph
Ph
Pis and coworkers (Ref. 40)
B
B
O
Me
24
O
OH
O
O
OH
OH
Me
24
Me
Me
Me
O
Me
Me
24
Me
OH
Me
Me
21
R
21
21
H
H
Me
H
Me
Me
H
AcO
AcO
AcO
Me
H
Me
H
2
AcO
HO
2
O
AcO
O
2
HO
HO
1) PhB(OH)2, MeOH
2) Ac2O, Py
AcO
AcO
AcO
OH
12
OH
OH
AcO
O
R
13, R1 =
Br
R2 = H
3
11
3
3
AcO
AcO
AcO
O
Ag-silicate
72% yield, 2 steps
O
O
27% of 13
20% of 14
O
R2 =
R = H,
1
14,
AcO
AcO
Nagorny and coworkers (Ref. 41)
O
Me
Me
BnO
BnO
BnO
BnO
O
O
Me
Me
5
O
O
O
Me
O
BnO
Me
OBn
O
Me
CCl
BnO
BnO
O
Me
Me
OH
Me
Me
11
Me
PhB(OH)2
4Å MS
PhMe, r.t.
O
Me
Me
TMSOTf
NH
Me
11
BnO
11
Me
Me
Et
Me
OH
11
Et3N, MeOH
Me
Et
Me
Et
O
OH
Me
OH
5
Me
3
5
5
Et
O
O
B
Step 3: in situ
work up
Step 1: diol
protection
Me
O
OH
Step 2: in situ
glycosylation
O
O
O
B
3
3
Me
3
18
O
O
Ph
O
OH
Ph
O
O
r.r. > 99:1
β:α= 3.7:1
62% (98% brsm)
Me
Me
Me
17
16
6-dEB (15)
Current study
19
O
19
5
HO
O
O
Me
OH
Me
selective
protection
Me
19
19
5
B
HO
O
OH
2.6 kcal/mol
R
O
O
5
O
B
O
Me
B
HO
5
O
DFT, B3LYP, 6-31G*
OH
O
HO
20
R
B
19
3
3
HO
HO
OH
strophanthidol (4a)
Scheme 2 Traceless boronic acid protection strategy in site-selective glycosylation of natural products
the C3, C5, C14, and C19 positions, and the boronic acid
protection could result in two isomeric products protected
either at the C3 and C5 positions (exemplified by compound
19) or at the C5 and C19 positions (exemplified by com-
pound 20). Our computational studies indicate that 20 is 2.6
kcal/mol more stable than 19 (DFT, B3LYP, 6–31 G*), and
therefore the application of the traceless protection/glyco-
sylation strategy to strophanthidol (4a), should lead to the
selective formation of the C3 glycosides. In this paper, we
describe the development of a single step protocol to directly
generate unprotected C3 glycosides of strophanthidol using
methylboronic acid as a transient protecting group.

J.-H. Tay et al.
Results and discussion
diastereoselectivity (57% yield, 1:2.5 d.r.). This result
supports our initial hypothesis that the formation of cyclic
boronate at the C5 and C19-hydroxyl groups of stro-
phanthidol is favored exclusively. Remarkably, the less
reactive tertiary C14 hydroxyl did not undergo glycosyla-
tion under these conditions. However, the glycosylation
yield and diastereoselectivity had to be optimized, and the
traceless protection/glycosylation/deprotection sequence
had to be executed in a single operation without isolating
the intermediates such as 21.
To control the diastereoselectivity, we switched the
protecting group on the glycosyl donor from benzyl to
benzoate, which is a well-known participating group that
allows to direct the glycosylation reactions (cf. Table 1).
After the in situ formation of the protected strophanthidol
21, L-rhamnose-derived glycosyl donor 7 and reaction
promoter were added to the same reaction vessel to initiate
the glycosylation reaction. We tested several general con-
ditions for the activation of trichloroacetimidate donor 7,
and we discovered that some of the common Lewis acid
activators were inefficient in promoting the glycosylation
reaction leading to 24 (entry 1–3). When TfOH was used as
the activator, we were able to obtain significantly
better conversion (57%) and isolated yield (37%) for the
Our studies commenced with identifying the conditions for
strophanthidol protection with methylboronic acid and
evaluating the regioselectivity of this reaction. We surmised
that methylboronic acid would preferably form cyclic ester
with the C5 and less sterically hindered C19-hydroxyl
group based on both the thermodynamic (vide supra) and
kinetic reasons. The methylboronic acid was selected due to
its lower stability of the corresponding ester toward
hydrolysis in comparison to phenylboronic acid that is more
commonly employed for the protection of 1,3-diols. Indeed,
after we mixed strophanthidol (4a) with methylboronic acid
in the presence of 4 Å molecular sieves in dichloromethane,
we were able to obtain a single compound (21) containing
methylboronic ester moiety with the quantitative mass
recovery after
a
simple filtration through Celite®
(Scheme 3). We subsequently subjected thus isolated
boronic ester intermediate 21 to the glycosylation condition
with benzyl protected fucose trichloroacetimidate donor 22
and TMSOTf as activator. Upon work up with MeOH to
remove the methylboronic acid masking group, we
observed exclusive glycosylation at the C3 position to form
glycoside 23, albeit with moderate yield and low
O
Scheme 3 Protection of
strophanthidol (4a) with
methylboronic acid and
subsequent glycosylation
O
O
Me
19
O
OH
Me
OH
Me
O
B
MeB(OH)2
19
HO
O
H
5
4Å M.S.
DCM, r.t.
2 h
3
HO
21
H
5
3
HO
>98% mass recovery
single boronate compound by 1H NMR
OH
strophanthidol (4a)
NH
O
CCl3
Me
O
OBn
22
OBn
BnO
O
TMSOTf (5 mol%),
4Å M.S., DCM, –20 oC
then azeotrope with MeOH
O
Me
OH
HO
H
H
3
O
Me
O
23
OBn
OH
OBn
BnO
57% yield
(α : β= 1 : 2.5)

Regioselective single pot C3-glycosylation of strophanthidol using methylboronic acid as a transient. . .
Table 1 Optimization of glycosylation of strophanthidol (4a)
MeB(OH)2 (1.1 equiv)
4Å M.S.
O
DCM, r.t., 2 h
O
strophanthidol
Me
OH
NH
(4a)
then
HO
7
O
CCl
H
Me
BzO
O
H
BzO
3
OBz
O
OH
conditions
Me
BzO
O
then azeotrope with
MeOH
24
BzO
OBz
conversion (%)a
/yield (%)
entry
conditions
o
1
2
3
24
4
TMSOTf (10 mol%), 7 (2 equiv.), –78 C to r.t., 2.5 h
o
AgOTf (100 mol%), 7 (1.1 equiv.), –40 C, 1 h
NOBF (120 mol%), 7 (2 equiv.), –15 C, 1.5 h
TfOH (20 mol%), 7 (3 equiv.), r.t., 3 h
TfOH (20 mol%), 7 (3 equiv.), r.t., 3 h
TfOH (30 mol%), 7 (2 equiv.), 0 C to r.t., 5 h
o
10
57 (37)
88 (76)
4
4
5b
6b
o
85 (81)
1
aConversion by H NMR analysis
bDonor was added portionwise
C3-glycosylated product 24. (entry 4). We also noticed that
the isolated yield was increased two-fold, from 37 to 76%,
when the glycosyl donor 7 was added to the reaction por-
tionwise at 30 min intervals (entry 5). Through further
optimization, we were able to obtain 24 in 81% yield
exclusively as the α-diastereomer (entry 6).
To further enhance the efficiency of our protocol, we
optimized the reaction work up to accomplish the depro-
tection of both the methylboronic ester and benzoate pro-
tecting groups on the sugar at the end of the reaction (cf.
Scheme 4). After the glycosylation leading to glycosylated
steroid was completed as judged by TLC, the reaction was
diluted with MeOH, and the solution was saturated with
ammonia by bubbling NH3 gas through it. The resultant
reaction mixture was left stirring at room temperature
overnight, concentrated, and purified by flash chromato-
graphy on silica gel. This simple single pot protocol resulted
in unprotected strophanthidol-3-O-α-L-rhamnoside (4b) in
73% yield (entry 1).
Using this optimized procedure, we also synthesized
strophanthidol glycoside 26a and 26b with unprotected
L-fucose and D-fucose at the C3 position in 77 and 72%
yield correspondingly (entries 2 and 3). However, when we
applied this strategy to generate the D-glucose and
D-mannose derivatives 26c and 26d, we noticed significant
formation of the ortho-ester by-products. The formation of
these side-products significantly reduced the overall yield
and complicated the purification of 26c and 26d. However,
we surmised that this undesired side-reaction might be
easily circumvented without changing our strategy if the
acyl groups that are used for the protection of sugar donors
are slightly modified. In particular, Kihlberg and coworker
[46] have noticed that the use of fluorine-substituted
benzoates as the protecting groups on sugar donors could
significantly minimize the formation of ortho-esters during
the glycosylation. Indeed, when we used 2-fluorobenzoate
protected D-glucose and D-mannose trichloroacetimidate
donors (24c and 24d) in our reaction, the formation of
ortho-esters was suppressed, and glycosides 26c and 26d
were formed in 75 and 69% yield correspondingly. All in
all, our one-pot protocol has allowed us to efficiently and
rapidly generate five glycosylated derivatives of stro-
phanthidol, three of which are new glycosides (26a, 26b,
and 26d) that have not been reported in the literature to date.
In summary, an efficient regioselective glycosylation of
strophanthidol (4a) to generate five different unprotected
glycosides in a single pot (77–69% yield) has been devel-
oped. This transformation was enabled by development of
regioselective protection of the C5/C19 diol moiety with
methylboronic acid. This method could be generally applied
to regioselective glycosylation of other C19-hydroxyl con-
taining steroids, and the use of traceless boronic acid pro-
tection to accomplish regioselective glycosylation of steroids
1–3 is the subject of ongoing studies in our laboratories.

J.-H. Tay et al.
O
O
MeB(OH)2 (1.1 equiv)
4Å M.S.
Scheme 4 Regioselective single
pot synthesis of C3-glycosylated
strophanthidol derivatives 4b
and 26. Isolated yield. Average
of 2 runs
O
O
DCM, r.t., 2 h
1
Me
OH
Me
H
HO
then Glycosyl Donor
(2 equiv)
TfOH (30 mol%), 0 °C
to rt, 5 h,
then NH3 in MeOH
r.t., 18 h
HO
3
H
H
H
OH
HO
SugarO
OH
OH
strophanthidol (4a)
26
entry
donor
product
yield1
HO
3
NH
7
4b
O
CCl3
Me
BzO
O
73%
77%
O
1
OH
BzO
Me
HO
O
OBz
NH
HO
O
OH
HO
3
25a
O
CCl3
26a
Me
O
2
3
OBz
Me
O
OBz
OH
OH
BzO
OH
HO
BzO
Me
HO
3
25b
26b
O
BzO
HO
BzO
Me
72%
75%
O
CCl3
O
O
O
HO
OH
OH
NH
HO
HO
3
RO
RO
RO
25c
26c
O
HO
HO
HO
O
RO
O
CCl3
4
R= 2-F-Bz
HO
NH
25d
26d
RO
RO
O
HO
3
RO
RO
HO
HO
HO
HO
O
69%
5
O
CCl3
R= 2-F-Bz
O
NH
OH
Experimental section
purchased from Glentham Life Sciences Limited. 4 Å
molecular sieve was activated prior to use by heating under
reduced pressure. Cooling was achieved by use of cryocool
machine or ice bath. Deionized water was used in the pre-
paration of all aqueous solutions and for all aqueous
extractions. Solvents used for extraction and chromato-
graphy were ACS or HPLC grade. Purification of reactions
General methods
All reagents and solvents were purchased from Sigma-
Aldrich, Fisher Scientific, and were used as received with-
out further purification unless specified. Strophanthidin was

Regioselective single pot C3-glycosylation of strophanthidol using methylboronic acid as a transient. . .
mixtures was performed by flash chromatography using
SiliCycle SiliaFlash P60 (230–400 mesh).
H, m), 1.86 (1 H, dtd, J = 14.4, 9.3, 5.4 Hz), 1.79–1.63 (4
H, m), 1.64–1.45 (5 H, m), 1.41–1.31 (2 H, m), 1.25–1.17
(2 H, m), 0.85 (3 H, s). 13C NMR (176 MHz, CDCl3) δ
174.8, 174.7, 117.9, 85.5, 78.2, 73.7, 67.9, 65.7, 50.7, 49.6,
42.3, 40.6, 40.2, 39.0, 36.9, 35.5, 32.7, 27.7, 26.9, 24.0,
21.4, 19.2, 15.9.
1H NMR spectra were recorded on Varian vnmrs 700
(700 MHz), Varian vnmrs 500 (500 MHz), Varian INOVA
500 (500 MHz), or Varian MR400 (400 MHz) spectro-
meters and chemical shifts (δ) are reported in parts per
million (ppm) with solvent resonance as the internal stan-
dard (CDCl3 at δ 7.26, CD3OD at δ 3.31, and C6D6 at δ
7.16). Data are reported as (br = broad, s = singlet,
d = doublet, t = triplet, q = quartet, m = multiplet; coupling
constant(s) in Hz; integration). Proton-decoupled 13C NMR
spectra were recorded on Varian vnmrs 700 (700 MHz),
Varian vnmrs 500 (500 MHz), Varian INOVA 500 (500
MHz), or Varian MR400 (400 MHz) spectrometers and
chemical shifts (δ) are reported in ppm with solvent reso-
nance as the internal standard (CDCl3 at δ 77.16, CD3OD at
49.0, C6D6 at 128.06). High resolution mass spectra
(HRMS) were recorded on Micromass AutoSpec Ultima or
VG (Micromass) 70–250-S Magnetic sector mass spectro-
meters in the University of Michigan mass spectrometry
laboratory. Infrared (IR) spectra were recorded as thin film
on a Thermo-Nicolet IS-50 spectrometer. Absorption peaks
were reported in wavenumbers (cm−1). Optical rotations
were measured in a solvent of choice on a JASCO P-2000
or Autopol III digital polarimeter at 589 nm (D-line).
Boronate 21
4a (8.0 mg, 0.020 mmol), MeB(OH)2 (1.3 mg, 1.1 equiv.),
and 30 mg of powered 4 Å molecular sieve were stirred in
100 μL of dry dichloromethane at room temperature. After
2 h, the reaction was filtered through Celite and con-
centrated under reduced pressure to obtain boronate inter-
mediate 21 ( > 98% mass recovery).
Infrared spectroscopy; IR vmax (film) cm−1 3470, 2938,
1780, 1737, 1620.
Optical rotation; ½αŠ25 + 21.8 (c 1.0, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (700 MHz,
CDCl3) δ 5.88 (1 H, s), 4.96 (1 H, d, J = 18.0 Hz), 4.79 (1
H, d, J = 18.1 Hz), 4.28 (d, 1 H, J = 11.7 Hz), 4.06–3.98 (1
H, m), 3.59 (d, 1 H, J = 8.3 Hz), 3.45 (d, 1 H, J = 11.6 Hz),
2.77
(dd,
1 H,
J = 9.7,
5.5 Hz),
2.20–2.15
(1 H, m), 2.10–2.02 (4 H, m), 1.95 (dd, 1 H, J = 13.9, 3.6
Hz), 1.88 (dtd, 1 H, J = 14.6, 9.4, 5.4 Hz), 1.79–1.60 (10 H,
m), 1.59–1.52 (2 H, m), 1.45–1.34 (3 H, m), 1.22 (dtd, 1 H,
J = 26.4, 13.6, 12.5, 5.8 Hz), 1.10 (qd, 1 H, J = 13.9, 13.4,
3.8 Hz), 0.86 (3 H, s). 13C NMR (176 MHz, CDCl3) δ
174.6, 174.3, 118.0, 85.3, 76.2, 73.6, 66.7, 65.3, 50.6, 49.6,
41.0, 40.2, 38.6, 38.1, 36.9, 36.8, 32.9, 27.6, 27.0, 23.8,
21.4, 19.8, 15.9. 10B NMR (75 MHz, CDCl3): δ 32.6 (br).
Strophanthidol (4a)
Strophanthidin (500 mg, 1.24 mmol) was dissolved in EtOH
(0.15 M) and cooled on an ice bath. NaBH4 (95 mg, 2.0
equiv.) was added, and the reaction was warmed up to room
temperature. After stirring at room temperature for 2 h, the
reaction was cooled in ice bath, diluted with 50 mL of
EtOAc, and quenched slowly with 50 mL of saturated
NH4Cl (aq). The biphasic mixture was stirred vigorously
overnight. Next, the organic layer was separated and con-
centrated under reduced pressure. The crude mixture was
purified by flash column chromatography on silica gel
(MeOH:CH2Cl2 15:1 to 10:1 v/v) to obtain strophanthidol
4a (430 mg, 85% yield) as white solid.
Glycoside 23
21 and trichloroacetimidate donor 22 (23.2 mg, 2.0
equiv.) was dissolved in 400 μL dry dichloromethane and
cooled to –20 °C. TMSOTf (0.18 μL, 0.05 equiv.) in 20
μL dry dichloromethane was added to the reaction. After
stirring for 16 h, the reaction was quenched with Et3N at
room temperature and filtered through a short pad of
Celite. The crude mixture was quenched and then azeo-
troped with MeOH multiple times and then subsequently
purified by flash column chromatography on silica gel
(EtOAc:Hexanes, 70:30 v/v) to obtain glycoside 23 as a
mixture of two diastereomers (9.4 mg, 57% yield). The
diastereomeric glycosides were separated by semi-prep
HPLC (18 mL/min, EtOAc:Hexanes, 70:30 v/v) and
characterized.
Mass Spectrometry; HRMS m/z 429.2247 (M + Na,
C23H34O6).
Infrared Spectroscopy; IR vmax (film) cm−1 3323, 2939,
1779, 1731, 1619, 1450, 1025.
Optical Rotation; ½αŠ25 + 30.6 (c 1.0, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (700 MHz,
CDCl3): δ 5.87 (1 H, s), 5.75 (1 H, s), 4.97 (1 H, d, J = 18.2
Hz), 4.80 (1 H, d, J = 17.9 Hz), 4.70 (1 H, d,
J = 7.5 Hz), 4.37 (1 H, d, J = 11.2 Hz), 4.20 (1 H, d,
J = 4.1 Hz), 4.07 (1 H, d, J = 4.6 Hz), 3.47 (1 H, dd, J =
11.4, 7.5 Hz), 2.76 (1 H, dd, J = 9.8, 5.5 Hz), 2.55 (1 H, td,
J = 14.7, 4.1 Hz), 2.15 (1 H, dt, J = 13.6, 10.0 Hz), 2.09 (1
H, dd, J = 15.0, 3.1 Hz), 2.05–1.98 (1 H, m), 1.97–1.91 (3
β-23
Mass spectrometry: HRMS m/z 845.4232 (M + Na,
C50H62O10) 845.4235.

J.-H. Tay et al.
Infrared spectroscopy; IR (film) cm−1 3441, 2935, 1780,
cooled in ice bath. TfOH (1.1 μL, 0.3 equiv.) in 20 μL of dry
dichloromethane was added to the reaction. At 1 h interval,
0.5 equiv. of sugar donor in 200 μL of dry dichloromethane
was added to reaction (repeated three times). After the last
portion of sugar portion was added, the reaction was stirred
in ice bath for 1 h before being warmed up to room tem-
perature and quenched with Et3N. Next, the reaction was
filtered through a short pad of silica gel/Celite and con-
centrated under reduced pressure. The crude mixture was
quenched and azeotroped with MeOH multiple times and
then purified by flash column chromatography on silica gel
(EtOAc:Hexanes, 60:40 to 70:30 v/v) to obtain 24 (27.9 mg,
81% yield) as white foamy solid.
1742, 1620, 1453, 1063.
Optical rotation; ½αŠ25 + 21.3 (c 0.31, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (700 MHz, CDCl3)
δ 7.38–7.27 (15 H, m), 5.87 (1 H, s), 5.01–4.93 (2 H, m),
4.87–4.81 (2 H, m), 4.79 (1 H, dd, J = 18.1, 1.8 Hz),
4.77–4.71 (2 H, m), 4.70 (1 H, d, J = 11.8 Hz), 4.67 (1 H, s),
4.38 (1 H, d, J = 7.7 Hz), 4.30 (1 H, d, J = 11.4 Hz), 4.24 (1
H, d, J = 10.2 Hz), 4.22 (1 H, s), 3.81 (1 H, dd, J = 9.7, 7.7
Hz), 3.57 (1 H, d, J = 2.9 Hz), 3.53 (1 H, dd, J = 9.7, 2.9 Hz),
3.50–3.43 (1 H, m), 3.42 (1 H, t, J = 10.8 Hz), 2.76 (1 H, dd,
J = 9.7, 5.4 Hz), 2.54 (1 H, td, J = 14.8, 3.8 Hz), 2.19–2.10
(2 H, m), 2.07–1.96 (2 H, m), 1.95–1.82 (3 H, m), 1.76–1.71
(1 H, m), 1.70–1.62 (2 H, m), 1.60–1.44 (6 H, m), 1.39–1.24
(6 H, m), 1.20 (1 H, dd, J = 13.7, 4.5 Hz), 1.16 (3 H, d, J =
6.4 Hz), 0.85 (3 H, s). 13C NMR (176 MHz, CDCl3) δ 174.5,
174.3, 138.5, 138.4, 138.3, 128.7, 128.6, 128.5, 128.4, 128.3,
127.8, 127.7, 118.0, 101.4, 85.7, 83.1, 79.1, 76.8, 75.9, 75.7,
74.9, 74.8, 73.5, 73.2, 70.9, 64.8, 50.7, 49.5, 42.6, 40.6, 40.2,
39.1, 36.2, 35.1, 32.8, 29.9, 26.8, 24.2, 23.9, 21.3, 19.1, 17.0,
15.8.
Mass spectrometry; HRMS m/z 887.3594 (M + Na, for
C50H56O13).
Infrared spectroscopy; IR vmax(film) cm−1 3485, 2935,
1780, 1729, 1601, 1451.
Optical rotation; ½αŠ25 + 9.3 (c 0.57, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (500 MHz,
CDCl3) δ 8.12–8.06 (2 H, m), 8.04–7.97 (2 H, m),
7.87–7.78 (2 H, m), 7.65–7.59 (1 H, m), 7.56–7.46 (3 H,
m), 7.46–7.36 (3 H, m), 7.27–7.24 (2 H, m), 5.88 (1 H, s),
5.71 (2 H, dd, J = 5.5, 2.4 Hz), 5.66–5.60 (1 H, m), 5.16 (1
H, d, J = 1.9 Hz), 4.98 (1 H, dd, J = 18.1, 1.8 Hz), 4.81 (1
H, dd, J = 18.1, 1.7 Hz), 4.40 (1 H, d, J = 11.4 Hz),
4.30–4.19 (4 H, m), 3.50 (1 H, t, J = 10.6 Hz), 2.78 (1 H,
dd, J = 9.6, 5.4 Hz), 2.60 (1 H, td, J = 14.9, 3.8 Hz), 2.20
(1 H, dd, J = 15.4, 3.3 Hz), 2.18–1.95 (5 H, m), 1.88 (1 H,
dtd, J = 14.3, 9.2, 5.4 Hz), 1.79 (1 H, s), 1.76–1.68 (3 H,
m), 1.64 (3 H, h, J = 4.3 Hz), 1.54 (3 H, tq, J = 11.9, 3.3
Hz), 1.36 (3 H, d), 1.35–1.31 (1 H, m), 1.28–1.15 (1 H, m),
0.87 (3 H, s). 13C NMR (125 MHz, CDCl3) δ 174.6, 174.4,
165.9, 165.7, 165.5, 133.8, 133.6, 133.3, 130.0, 129.9,
129.8, 129.2, 129.1, 129.1, 128.8, 128.6, 128.4, 117.9,
97.8, 85.5, 77.4, 77.3, 76.9, 73.6, 71.3, 71.1, 70.0, 68.1,
64.9, 50.7, 49.5, 42.7, 40.5, 40.2, 39.2, 35.8, 34.6, 32.7,
29.8, 26.8, 25.5, 24.1, 21.3, 19.2, 17.8, 15.9.
α-23
Mass spectrometry; HRMS m/z 845.4230 (M + Na,
C50H62O10).
Infrared spectroscopy; IR vmax (film) cm−1 3436, 2924,
1742, 1620, 1453.
Optical rotation; ½αŠ25 – 27.1 (c 0.11, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (700 MHz,
CDCl3) δ 7.41–7.27 (15 H, m), 5.88 (1 H, s), 5.21 (1 H, s),
4.99–4.94 (2 H, m), 4.83–4.77 (3 H, m), 4.77–4.71 (2 H,
m), 4.65 (3 H, dd, J = 17.9, 11.8 Hz), 4.36 (1 H, d, J = 11.3
Hz), 4.02 (1 H, t, J = 2.8 Hz), 4.00 (1 H, dd, J = 10.0, 3.9
Hz), 3.91–3.83 (2 H, m), 3.65 (1 H, d, J = 1.4 Hz), 3.42 (1
H, d, J = 9.7 Hz), 2.77 (1 H, dd, J = 9.7, 5.5 Hz), 2.54 (1 H,
td, J = 14.5, 4.4 Hz), 2.19–1.92 (6 H, m), 1.87 (1 H, dtd, J
= 14.6, 9.4, 5.5 Hz), 1.73–1.44 (8 H, m), 1.34 (2 H, m),
1.31–1.17 (4 H, m), 1.09 (3 H, d, J = 6.5 Hz), 0.86 (3 H, s).
13C NMR (176 MHz, CDCl3): δ 174.5, 174.2, 138.6, 138.6,
138.4, 128.6, 128.4, 128.4, 127.9, 127.8, 127.7, 127.7,
118.1, 96.4, 85.7, 79.4, 77.4, 77.1, 76.7, 75.6, 75.0, 74.0,
73.5, 73.1, 72.5, 67.2, 65.0, 50.7, 49.4, 42.4, 40.6, 40.3,
39.0, 35.3, 34.2, 32.8, 29.9, 26.8, 26.2, 24.2, 21.3, 19.4,
16.8, 15.8.
General procedure for one-pot glycosylation of
strophanthidol (4a)
Strophanthidol 4a (16.3 mg, 0.040 mmol), MeB(OH)2 (2.6
mg, 1.1 equiv.) and 60 mg of powered 4 Å molecular sieve
were stirred in 200 μL of dry dichloromethane at room
temperature. After 2 h, 0.5 equiv. of trichloroacetimidate
donor 7 or 25a-d in 1.8 mL of dry dichloromethane was
added to the reaction, and the mixture was cooled in ice
bath. TfOH (1.1 μL, 0.3 equiv.) in 20 μL of dry dichlor-
omethane was added to the reaction. At 1 h interval, 0.5
equiv. of trichloroacetimidate donor in 200 μL of dry
dichloromethane was added to reaction (repeated 3 times).
After the last portion of trichloroacetimidate donor was
added, the reaction was stirred in ice bath for 1 h before
Glycoside 24
4a (16.3 mg, 0.040 mmol), MeB(OH)2 (2.6 mg, 1.1 equiv.)
and 60 mg of powered 4 Å molecular sieve were stirred in
200 μL of dry dichloromethane at room temperature. After
2 h, 0.5 equiv. of sugar donor 7 in 1.8 mL of dry dichlor-
omethane was added to the reaction, and the mixture was

Regioselective single pot C3-glycosylation of strophanthidol using methylboronic acid as a transient. . .
being warmed up to room temperature. Next, the reaction
was diluted with 3 mL of MeOH and NH3 gas was bubbled
through the reaction for 30 min. After stirring at room
temperature for 18 h, the reaction mixture was filtered
through a short pad of silica gel/Celite and concentrated
under reduced pressure. The crude mixture was purified by
flash column chromatography (on pipette column) on silica
gel (starting with MeOH:CH2Cl2, 5:95 v/v + 0.5% of
AcOH, followed by a gradient of MeOH:CH2Cl2, 10:90 to
20:80 v/v) to obtain glycoside 4b, 26a-d.
CD3OD) δ 178.3, 177.2, 117.9, 102.4, 86.4, 77.3, 75.6,
75.3, 75.2, 73.0, 72.3, 72.1, 65.4, 51.9, 50.8, 44.1, 41.4,
41.1, 40.0, 37.4, 36.5, 33.0, 27.9, 25.1, 24.1, 22.7, 20.1,
16.7, 16.3.
Glycoside 26b
Compound 26b was prepared according to general proce-
dure for one-pot glycosylation of strophanthidol in 72%
yield after flash column chromatography.
Mass spectrometry; HRMS m/z (M + Na, C29H44O10)
575.2829.
Glycoside 4b
Infrared spectroscopy; IR vmax (film) cm−1 3354, 2940,
1733, 1666, 1624, 1449, 1024.
Compound 4b was prepared according to general procedure
for one-pot glycosylation of strophanthidol (4a) in 73%
yield after flash column chromatography.
Mass spectrometry; HRMS m/z (M + Na, C29H44O10)
575.2828.
Optical rotation; ½αŠ25 + 12.6 (c 1.0, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (500 MHz,
CD3OD) δ 5.90 (1 H, s), 5.03 (1 H, dd, J = 18.4, 1.8 Hz),
4.91 (1 H, dd, J = 18.4, 1.7 Hz), 4.36–4.28 (1 H, m), 4.20
(1 H, d, J = 11.3 Hz), 4.17 (1 H, s), 3.64 (1 H, m),
3.61–3.54 (2 H, m), 3.49–3.44 (2 H, m), 2.85–2.81 (1 H,
m), 2.32 (1 H, td, J = 14.5, 3.8 Hz), 2.22–2.09 (4 H, m),
2.03–1.81 (4 H, m), 1.80–1.61 (6 H, m), 1.58–1.34 (7 H,
m), 1.26 (3 H, d, J = 6.5 Hz), 1.24–1.16 (2 H, m), 0.88 (3 H,
s). 13C NMR (125 MHz, CD3OD) δ 178.3, 177.2, 117.9,
102.6, 86.4, 77.3, 75.8, 75.3, 75.3, 73.1, 72.3, 72.1, 65.6,
51.9, 50.8, 43.8, 41.4, 41.1, 40.0, 36.8, 35.3, 33.1, 27.9,
26.5, 25.2, 22.6, 20.5, 16.8, 16.3.
Infrared spectroscopy; IR vmax (film) cm−1 3342, 2938,
1732, 1449, 1021.
Optical rotation. ½αŠ25 – 3.68 (c 1.0, CHCl3).
D
Nuclear magnetic resonance: 1H NMR (500 MHz,
CD3OD) δ 5.90 (1 H, s), 5.03 (1 H, dd, J = 18.4, 1.8 Hz),
4.92 (1 H, dd, J = 18.4, 1.8 Hz), 4.85 (1 H, d, J = 1.8 Hz),
4.19–4.11 (2 H, m), 3.79 (1 H, dd, J = 3.4, 1.7 Hz),
3.69–3.59 (3 H, m), 3.40 (1 H, t, J = 9.5 Hz), 2.86–2.83 (1
H, m), 2.26–2.09 (5 H, m), 2.03–1.94 (1 H, m), 1.94–1.85
(2 H, m), 1.84–1.65 (6 H, m), 1.58–1.38 (9 H, m), 1.29 (1 H,
s) 1.26 (3 H, d, J = 6.2 Hz), 1.25–1.19 (1 H, m), 0.89 (3 H,
s). 13C NMR (125 MHz, CD3OD) δ 178.3, 177.2, 117.8,
100.9, 86.4, 77.4, 75.6, 75.3, 73.8, 72.5, 72.4, 70.7, 65.2,
51.9, 50.8, 44.1, 41.5, 41.1, 40.0, 36.2, 35.9, 33.0, 27.9,
26.4, 25.0, 22.8, 20.4, 18.0, 16.3.
Trichloroacetimidate donor 25c
D-Glucose (300 mg, 1.67 mmol) and DMAP (102 mg, 0.5
equiv.) were dissolved in dry pyridine and cooled on an ice
bath. 2-fluorobenzoyl chloride (1.4 mL, 7.0 equiv.) was
added to the reaction dropwise. After stirring at room
temperature overnight, the reaction was quenched with 2
mL of MeOH and stirred for 1 h. Next, the reaction was
diluted with 50 mL of DCM and washed with 2 N HCl (two
times), sat. NaHCO3 (two times), and brine solution. The
organic layer was dried over Na2SO3, filtered and con-
centrated under reduced pressure. The crude mixture was
used in next step without further purification.
The crude from previous step was dissolved in 5 mL of
DCM and 33% HBr in AcOH was added to the reaction
dropwise. After stirring at room temperature for 5 h, the
reaction was quenched slowly with sat NaHCO3 (aq). Next,
the mixture was diluted with DCM, and washed with sat
NaHCO3 (aq), followed by brine solution. The organic layer
was dried over Na2SO3, filtered and concentrated under
reduced pressure. The crude mixture was used in next step
without further purification.
Glycoside 26a
Glycoside 26a was prepared according to general procedure
for one-pot glycosylation of strophanthidol in 77% yield
after flash column chromatography.
Mass spectrometry; HRMS m/z (M + Na, C29H44O10)
575.2828.
Infrared spectroscopy; IR vmax (film) cm−1 3348, 2940,
1733, 1666, 1624, 1448, 1022.
Optical rotation; ½αŠ25 + 14.5 (c 1.0, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (500 MHz,
CD3OD) δ 5.90 (1 H, s), 5.03 (1 H, dd, J = 18.4, 1.8 Hz),
4.91 (1 H, dd, J = 18.4, 1.7 Hz), 4.36–4.27 (1 H, m), 4.21
(1 H, s), 4.18 (1 H, d, J = 11.4 Hz), 3.69–3.62 (1 H, m),
3.60 (1 H, d, J = 1.6 Hz), 3.58 (1 H, d, J = 11.4 Hz),
3.48–3.44 (2 H, m), 2.85–2.82 (1 H, m), 2.34–2.22 (2 H,
m), 2.21–2.08 (2 H, m), 2.03–1.82 (5 H, m), 1.77–1.58 (6
H, m), 1.56–1.35 (7 H, m), 1.26 (3 H, d, J = 6.4 Hz),
1.24–1.18 (2 H, m), 0.88 (3 H, s). 13C NMR (125 MHz,
The crude from previous step was dissolved in acetone/
H2O/DCM (10 mL: 1 mL: 2 mL) before Ag2CO3 (800 mg)
was added to the reaction. After stirring at room temperature

J.-H. Tay et al.
for overnight, the reaction was filtered through Celite and
concentrated under reduced pressure. The crude mixture was
purified by flash column chromatography on silica gel
(EtOAc: hexanes, 30:70 to 40:60 v/v) to obtain benzoate
protected hemiacetal, which (300 mg, 0.45 mmol) was dis-
solved in 3 mL of DCM (0.15 M) and cooled on an ice bath.
Trichloroacetonitrile (0.9 mL, 20 equiv.) was added to the
reaction, followed by DBU (34 μL, 0.5 equiv.). After stirring at
room temperature for 2 h, the reaction was concentrated under
reduced pressure. The crude mixture was purified by flash
column chromatography on silica gel (EtOAc:hexanes, 15:85
v/v) to obtain α-trichloroacetimidate donor 25c (250 mg, 68%
yield) as white foamy solid.
2.22–2.09 (3 H, m), 2.05–1.81 (4 H, m), 1.81–1.59 (6 H,
m), 1.57–1.32 (7 H, m), 1.32–1.20 (2 H, m), 0.88 (3 H, s).
13C NMR (125 MHz, CD3OD) δ 178.3, 177.2, 117.9, 102.0,
86.4, 78.2, 78.2, 77.3, 75.7, 75.3, 75.1, 71.6, 65.6, 62.7,
51.9, 50.8, 43.8, 41.4, 41.0, 39.9, 36.9, 35.2, 33.1, 27.9,
26.5, 25.2, 22.6, 20.5, 16.3.
Trichloroacetimidate donor 25d
Compound 25d was prepared from D-mannose following
the same synthetic procedures as for 25c. Mass Spectro-
metry; HRMS m/z (M + Na, C36H24Cl3F4NO10) 834.0303.
Infrared spectrometry; IR vmax (film) cm−1 1725, 1678,
1612, 1585, 1488, 1456, 1289, 1243.
Mass
spectrometry;
HRMS
m/z
(M + Na,
C36H24Cl3F4NO10) 834.0301.
Optical rotation; ½αŠ25 – 34.3 (c 1.0, CHCl3).
D
Infrared spectroscopy; IR vmax (film) cm−1 1729, 1677,
Nuclear magnetic resonance; 1H NMR (700 MHz,
CDCl3) δ 8.87 (1 H, s), 7.99 (2 H, q, J = 7.8 Hz), 7.82 (2 H,
q, J = 7.9 Hz), 7.59–7.56 (1 H, m), 7.54–7.46 (2 H, m), 7.44
(1 H, qd, J = 7.0, 6.5, 2.8 Hz), 7.19 (1 H, t, J = 7.6 Hz),
7.16–7.11 (3 H, m), 7.11–7.05 (3 H, m), 7.01 (1 H, dd, J =
10.6, 8.4 Hz), 6.57 (1 H, s), 6.21 (1 H, t, J = 9.7 Hz),
5.97–5.95 (2 H, m), 4.69–4.65 (1 H, m), 4.62–4.56 (2 H,
m). 13C NMR (176 MHz, CDCl3) δ 163.8, 163.8, 163.2,
163.2, 163.1, 163.1, 163.1, 163.0, 162.8, 161.7, 161.6,
161.6, 161.3, 159.9, 135.4, 135.3, 135.1, 135.1, 134.8,
134.7, 132.5, 132.5, 132.2, 132.1, 124.2, 124.2, 124.2,
124.1, 124.1, 124.0, 124.0, 118.4, 118.3, 117.9, 117.8,
117.7, 117.6, 117.6, 117.5, 117.3, 117.2, 117.2, 117.1,
117.1, 117.1, 117.0, 94.6, 90.7, 71.5, 70.1, 69.0, 66.4, 62.8
(complex multiplet in δ 164–117 due to C-F couplings).
1612, 1584, 1489, 1456, 1290, 1243.
Optical rotation; ½αŠ25 + 66.5 (c 1.0, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (700 MHz, CDCl3)
δ 8.67 (1 H, s), 7.96 (1 H, t, J = 7.5 Hz), 7.89 (1 H, t, J = 7.5
Hz), 7.83 (1 H, t, J = 7.4 Hz), 7.77 (1 H, t, J = 7.5 Hz),
7.54–7.48 (3 H, m), 7.44 (1 H, q, J = 7.1 Hz), 7.21 (1 H, t,
J = 7.5 Hz), 7.18–7.09 (4 H, m), 7.08 (2 H, t, J = 9.5 Hz),
7.02 (1 H, t, J = 9.5 Hz), 6.24 (1 H, t, J = 9.9 Hz), 5.80 (1 H,
t, J = 9.9 Hz), 5.60 (1 H, dd, J = 10.2, 3.7 Hz), 4.67–4.60 (2
H, m), 4.57 (1 H, dd, J = 12.7, 4.9 Hz). 13C NMR (176 MHz,
CDCl3) δ 163.9, 163.9, 163.2, 163.2, 163.2, 163.1, 163.0,
163.0, 162.9, 162.9, 162.8, 162.8, 161.7, 161.5, 161.3, 161.3,
160.7, 135.4, 135.4, 135.2, 135.2, 134.9, 134.9, 134.9, 134.8,
132.4, 132.3, 132.1, 131.9, 124.3, 124.2, 124.2, 124.2, 124.2,
124.1, 124.1, 124.1, 118.3, 118.2, 118.0, 118.0, 117.7, 117.7,
117.3, 117.2, 117.2, 117.1, 117.1, 117.1, 116.9, 93.1, 90.8,
70.8, 70.6, 70.4, 68.9, 62.7 (complex multiplet in δ 164–117
due to C-F couplings).
Glycoside 26d
Compound 26d was prepared according to general proce-
dure for one-pot glycosylation of strophanthidol in 69%
yield after flash column chromatography.
Mass spectrometry; HRMS m/z (M + Na, C29H44O11)
591.2780.
Glycoside 26c
Compound 26c was prepared according to general proce-
dure for one-pot glycosylation of strophanthidol in 75%
yield after flash column chromatography.
Mass sepctrometry; HRMS m/z (M + Na, C29H44O11)
591.2791.
Infrared spectroscopy; IR vmax (film) cm−1 3358, 2939,
1733, 1667, 1617, 1452, 1027.
Optical rotation; ½αŠ25 + 46.9 (c 0.98, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (500 MHz,
CD3OD) δ 5.88 (1 H, s), 5.01 (1 H, dd, J = 18.5, 1.8 Hz),
4.94–4.86 (2 H, m), 4.24 (1 H, s), 4.14 (1 H, d, J = 11.4
Hz), 3.86–3.78 (2 H, m), 3.72–3.55 (5 H, m), 2.82–2.80 (1
H, m, J = 9.0 Hz), 2.34 (1 H, dd, J = 15.3, 3.5 Hz),
2.22–2.02 (3 H, m), 2.00–1.79 (5 H, m), 1.77–1.59 (5 H,
m), 1.57–1.34 (8 H, m), 1.31–1.15 (2 H, m), 0.87
(3 H, s). 13C NMR (125 MHz, CD3OD) δ 178.3, 177.2,
117.9, 98.5, 86.4, 77.5, 75.4, 75.3, 72.8, 72.7, 72.4, 68.6,
65.2, 63.0, 51.9, 50.8, 44.3, 41.5, 41.0, 40.0, 37.7, 36.1,
33.0, 27.9, 25.0, 23.0, 22.8, 20.0, 16.3.
Infrared spectroscopy; IR vmax (film) cm−1 3341, 2941,
1734, 1668, 1450, 1019.
Optical rotation; ½αŠ25 + 10.1 (c 1.0, CHCl3).
D
Nuclear magnetic resonance; 1H NMR (500 MHz,
CD3OD) δ 5.90 (1 H, s), 5.03 (1 H, dd, J = 18.5, 1.8 Hz),
4.95–4.87 (1 H, dd, J = 18.5, 1.7 Hz), 4.40 (1 H, d, J = 7.9
Hz), 4.23 (1 H, s), 4.20 (1 H, d, J = 11.4 Hz), 3.86 (1 H, dd,
J = 11.9, 1.7 Hz), 3.69–3.63 (1 H, m), 3.58 (1 H, d, J =
11.4 Hz), 3.29–3.26 (2 H, m), 3.18 (1 H, dd, J = 9.2, 7.9
Hz), 2.85–2.82 (1 H, m), 2.32 (1 H, td, J = 14.5, 4.0 Hz),

Regioselective single pot C3-glycosylation of strophanthidol using methylboronic acid as a transient. . .
Acknowledgements This work was supported by NIH grant
R01GM111476 (PN). PN is the Sloan Foundation and Amgen Young
Investigator Fellow.
18. Orellana AM, Kinoshita PF, Leite JA, Kawamoto M, Scavone C.
Cardiotonic steroids as modulators of neuroinflammation. Front
Endocrinol. 2016;7:1–10.
19. Ferrari P, Ferrandi M, Valentini G, Bianchi G. Rostafuroxin: an
ouabain antagonist that corrects renal and vascular Na+-K+-
ATPase alterations in ouabain and adducin-dependent hypertension.
Am J Physiol Regul Integr Comp Physiol. 2006;290:R529–35.
20. Mijatovic T, Kiss R. Cardiotonic steroids-mediated Na+/K
+-ATPase targeting could circumvent various chemoresistance
pathways. Planta Med. 2013;79:189–98.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
21. Yang P, et al. Cellular location and expression of Na+, K+‐
ATPase α subunits affect the anti‐proliferative activity of olean-
drin. Mol Carcinolg. 2014;53:253–63.
Publisher’s note: Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
22. Goncalves-de-Albuquerque FC, Silva RA, da Silva IC, Castro-
Faria-Neto CH, Burth P. Na/K Pump and beyond: Na/K-ATPase
References
as
a modulator of apoptosis and autophagy. Molecules.
1. Wasserstrom JA, Aistrup GL. Digitalis: new actions for an
2017;22:578–96.
old drug. Am
J Physiol Heart Circ Physiol. 2005;289:
23. Simpson CD, et al. Inhibition of the sodium potassium adenosine
triphosphatase pump sensitizes cancer cells to anoikis and pre-
vents distant tumor formation. Cancer Res. 2009;69:2739–47.
24. Jiang M-M, et al. Cardenolides from Antiaris toxicaria as potent
selective Nur77 modulators. Chem Pharm Bull. 2008;
56:1005–8.
25. Shi L-S, et al. Cytotoxic cardiac glycosides and coumarins from
antiaris toxicaria. Bioorg Med Chem. 2014;22:1889–98.
26. Tian D-M, Cheng H-Y, Shen W-Z, Tang J-S, Yao X-S. Cardiac
glycosides from the seeds of Thevetia peruviana. J Nat Prod.
2016;79:38–50.
27. Langenhan JM, Peters NR, Guzei AI, Hoffmann M, Thorson JS.
Enhancing the anticancer properties of cardiac glycosides by neo-
glycorandomization. Proc Natl Acad Sci USA2005;102:12305–10.
28. Iyer AKV, et al. Direct comparison of the anticancer activities of
digitoxin MeON-neoglycosides and O-glycosides. ACS Med
Chem Lett. 2010;1:326–30.
29. Li X. et al. Synthesis of C3-neoglycosides of digoxigenin and their
anticancer activities. Eur J Med Chem. 2018;145:252–62.
30. Pal’yants N, Sh., Abubakirov NK. Glycosylation of cardioster-
oids. Khim Prir Soedin. 1984;5:515–29.
31. Hong B-C, Kim S, Kim T-S, Corey EJ. Synthesis and properties
of several isomers of the cardioactive steroid ouabain. Tetrahedron
Lett. 2006;47:2711–5.
32. Zhang H, Reddy MS, Phoenix S, Deslongchamps P. Total
synthesis of ouabagenin and ouabain. Angew Chem Int Ed.
2008;47:1272–5.
33. Reddy MS, Zhang H, Phoenix S, Deslongchamps P. Total synthesis
of ouabagenin and ouabain. Chem Asian J. 2009;4:725–41.
34. Bhattarai B, Nagorny P. Concise enantioselective synthesis of
oxygenated steroids via sequential Copper(II)-catalyzed Michael
addition/intramolecular Aldol cyclization reactions. Org Lett.
2018;20:154–7.
35. Urabe D, et al. Total synthesis and biological evaluation of 19-
Hydroxysarmentogenin-3-O-α-l-rhamnoside, trewianin, and their
aglycons. J Org Chem. 2018;83:13888–910.
H1781–H1793.
2. Wiesner K, Tsai TYR. Some recent progress in the synthetic and
medicinal chemistry of cardioactive steroid glycosides. Pure Appl
Chem. 1986;58:799–810.
3. Heasley B. Chemical synthesis of the cardiotonic steroid glyco-
sides and related natural products. Chem Eur J.
2012;18:3092–120.
4. Michalak M, Michalak K, Wicha J. The synthesis of cardenolide
and bufadienolide aglycones, and related steroids bearing a het-
erocyclic subunit. Nat Prod Rep. 2017;34:361–410.
5. Pessoa MTC, Barbosa LA, Villar JAFP. Synthesis of cardiac
steroids and their role on heart failure and cancer. Stud Nat Prod
Chem. 2018;57:79–113.
6. Ahmad VU, Basha A. Spectroscopic Data of Steroid Glycosides:
Spirostanes, Bufanolides, Cardenolides.. New York: Springer;
2006. p. 524–695.
7. Ahmad VU, Basha A. Spectroscopic Data of Steroid Glycosides:
Cardenolides and Pregnanes.. New York: Springer; 2006. p.
2089–759.
8. Nesher M, Shpolansky U, Rosen H, Lichstein D. The digitalis-like
steroid hormones: new mechanisms of action and biological sig-
nificance. Life Sci. 2007;80:2093–107.
9. Schoner W, Scheiner-Bobis G. Endogenous and exogenous car-
diac glycosides: their roles in hypertension, salt metabolism, and
cell growth. Am J Phsyiol Cell Physiol. 2007;293:C509–C536.
10. Prassas I, Diamandis EP. Novel therapeutic applications of cardiac
glycosides. Nat Rev. 2008;7:926–35.
11. Bagrov AY, Shapiro JI, Fedorova OV. Endogenous cardiotonic
steroids: physiology, pharmacology, and novel therapeutic targets.
Pharm Rev. 2009;61:9–38.
12. Babula P, Masarik M, Adam V, Provaznik I, Kizek R. From Na
+/K+-ATPase and cardiac glycosides to cytotoxicity and cancer
treatment. AntiCancer Agent Med Chem. 2013;13:1069–87.
13. Schneider NFZ, Cerella C, Simoes CMO, Diederich M. Antic-
ancer and immunogenic properties of cardiac glycosides. Mole-
cules. 2017;22:1932–47.
36. Fenger, TH; Madsen,
R Regioselective Glycosylation of
14. Cai H, et al. Digitoxin analogues with improved antic-
ytomegalovirus activity. ACS Med Chem Lett. 2014;5:395–9.
15. Akhtar N, Malik A, Ali SN, Kazmi SU. Proceragenin, an anti-
bacterial cardenolide from Calotropis procera. Phytochemistry.
1992;31:2821–4.
16. Jacob PL, et al. Immunomodulatory activity of ouabain in
Leishmania leishmania amazonensis-infected Swiss Mice. Para-
sitol Res. 2013;112:1313–21.
Unprotected Phenyl 1-Thioglycopyranosides with Phe-
nylboronic Acid as a Transient Masking Group. Eur. J. Org.
Chem. 2013;26:5923–33.
37. Kaji E, Nishino T, Ishige K, Ohya Y, Shirai Y. Regioselective
glycosylation of fully unprotected methyl hexopyranosides by
means of transient masking of hydroxy groups with arylboronic
acids. Tetrahedron Lett. 2010;51:1570–3.
38. Mancini RS, Lee JB, Taylor MS. Boronic esters as protective
groups in carbohydrate chemistry: processes for acylation, sily-
lation, and alkylation of glycoside-derived boronates. Org Biomol
Chem. 2017;15:132–43.
17. Yoneyama T, et al. Notch inhibitors from Calotropis gigantea that
induce neuronal differentiation of neural stem cells. J Nat Prod.
2017;80:2453–61.

J.-H. Tay et al.
39. Dimakos V, Taylor MS. Site-selective functionalization of
hydroxyl groups in carbohydrate derivatives. Chem Rev.
2018;118:11457–517.
43. Cichowicz N, Nagorny P. New strategy based on sequential
Michael/Aldol reactions for the asymmetric synthesis of carde-
nolides. Strat Tactics Organ Synth. 2016;12:237–67.
40. Pis J, Hykl J, Budesinsky M, Harmatha J. Regioselective synthesis
of 20-hydroxyecdysone glycosides. Tetrahedron. 1994;50:
9679–90.
41. Tay J-H, et al. Regiodivergent glycosylations of 6-deoxy-
Erythronolide B and Oleandomycin-derived macrolactones enabled
by chiral acid catalysis. J Am Chem Soc. 2017;139:8570–8.
42. Cichowicz N, et al. Concise enantioselective synthesis of oxyge-
nated steroids via sequential copper(II)-catalyzed Michael addi-
tion/intramolecular Aldol cyclization reactions. J Am Chem Soc.
2015;137:14341–8.
44. Kaplan W, Khatri HR, Nagorny P. Concise enantioselective total
synthesis of cardiotonic steroids 19-hydroxysarmentogenin and
trewianin aglycone. J Am Chem Soc. 2016;138:7194–8.
45. Lee J, Wang S, Callahan M, Nagorny P. Copper(II)-catalyzed
tandem decarboxylative Michael/Aldol reactions leading to the
formation of functionalized cyclohexenones. Org Lett.
2018;20:2067–70.
46. Sjolin P, Kihlberg J. Use of fluorobenzoyl protective groups in
synthesis of glycopeptides: β-elimination of O-linked carbohy-
drates is suppressed. J Org Chem. 2001;66:2957–65.