A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural assignment via a protocol combining NMR and DP4 plus application

Article abstract of DOI:10.1016/j.phytochem.2016.10.010

A growing body of evidence points to the useful roles of computational approaches in the structural characterization of natural products. Rhododendron brachycarpum has been traditionally used for the control of diabetes, hepatitis, hypertension, and rheumatoid arthritis and classified as an endangered species in Korea. A grayanotox-9(11)-ene derivative along with five known diterpenoids, were isolated from the MeOH extract of R.?brachycarpum. Extensive 1D and 2D NMR experiments were conducted to establish the 2D structure and relative configuration of the grayanotox-9(11)-ene derivative. Comparison of simulated and experimental ECD spectra resulted in an inconclusive outcome to assign its absolute configuration. Alternatively, gauge-including atomic orbitals (GIAO) NMR chemical shift calculations, with support by the advanced statistical method DP4 plus, and acid hydrolysis were employed to establish its absolute configuration. This work exemplifies how NMR analysis, combined with quantum mechanics calculations, is a viable approach to accomplish structural assignment of minor abundance molecules in lieu of X-ray crystallography or chiroptical approaches.

Full text of DOI:10.1016/j.phytochem.2016.10.010

Phytochemistry xxx (2016) 1e6
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum
and its structural assignment via a protocol combining NMR and DP4
plus application
,
b
,
,
d
,
e
,
,
Nguyen Quoc Tuan a 1, Joonseok Oh c
1, Hyun Bong Park d e, Daneel Ferreira c,
Sanggil Choe f, Juseon Lee f, MinKyun Na a
,
*
a College of Pharmacy, Chungnam National University, Daejeon 34134, Republic of Korea
b Phutho College of Pharmacy, Viettri City, Phutho Province, Viet Nam
c Department of BioMolecular Sciences, Division of Pharmacognosy, Research Institute of Pharmaceutical Sciences, School of Pharmacy, The University of
Mississippi, MS 38677, United States
d Department of Chemistry, Yale University, New Haven, CT 06520, United States
e Chemical Biology Institute, Yale University, West Haven, CT 06516, United States
f Forensic Toxicology Division, National Forensic Service, Wonju 220-170, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
A growing body of evidence points to the useful roles of computational approaches in the structural
characterization of natural products. Rhododendron brachycarpum has been traditionally used for the
control of diabetes, hepatitis, hypertension, and rheumatoid arthritis and classified as an endangered
species in Korea. A grayanotox-9(11)-ene derivative along with five known diterpenoids, were isolated
from the MeOH extract of R. brachycarpum. Extensive 1D and 2D NMR experiments were conducted to
establish the 2D structure and relative configuration of the grayanotox-9(11)-ene derivative. Comparison
of simulated and experimental ECD spectra resulted in an inconclusive outcome to assign its absolute
configuration. Alternatively, gauge-including atomic orbitals (GIAO) NMR chemical shift calculations,
with support by the advanced statistical method DP4 plus, and acid hydrolysis were employed to
establish its absolute configuration. This work exemplifies how NMR analysis, combined with quantum
mechanics calculations, is a viable approach to accomplish structural assignment of minor abundance
molecules in lieu of X-ray crystallography or chiroptical approaches.
Received 12 April 2016
Received in revised form
6 August 2016
Accepted 25 October 2016
Available online xxx
Keywords:
Rhododendron brachycarpum
DP4 plus
Gauge-including atomic orbitals NMR
chemical shift calculations
Stereochemical assignments
Endangered plant species
Electronic circular dichroism
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
successfully applied in structural assignment or revision of complex
natural products in cases where compounds were mass-limited
Erroneous assignment of natural product structures is not un-
common due to high molecular complexity, human errors, and
NMR correlation ambiguities (Grimblat et al., 2015). Even though
synthesis approaches and X-ray crystallography have played a role
in structural assignment and revision (Nicolaou et al., 2000), such
methods may oftentimes be unfeasible in terms of costs and un-
favorable molecular features. Alternatively, quantum mechanics-
based calculations of NMR parameters and chiroptical features
have emerged as useful tools for the full elucidation of chemical
structures (Grimblat et al., 2015). These techniques were
and/or conformationally flexible (Lodewyk et al., 2012; Oh et al.,
2013; Smith and Goodman, 2010; Waters et al., 2015). It is of
particular interest to define stereochemical details employing
gauge-including atomic orbitals (GIAO) NMR chemical shift calcu-
lations with support from advanced statistical tools such as CP3 and
DP4, and recently introduced DP4 plus, considering the versatility,
high accuracy, and relatively inexpensive investment of computa-
tional resources (Grimblat et al., 2015; Smith and Goodman, 2009,
2010).
Rhododendron brachycarpum G. Don, a member of the Ericaceae
family, is an evergreen broad-leaved shrub (Popescu and Kopp,
2013). The plant species, predominantly populated in the north-
ern parts of Korea and central Japan (Lee et al., 2002), has been
classified as an endangered and rare plant species in Korea due to a
* Corresponding author.
E-mail address: mkna@cnu.ac.kr (M. Na).
Both authors contributed equally to this study.
1
http://dx.doi.org/10.1016/j.phytochem.2016.10.010
0031-9422/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Tuan, N.Q., et al., A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural
assignment via a protocol combining NMR and DP4 plus application, Phytochemistry (2016), http://dx.doi.org/10.1016/j.phytochem.2016.10.010

2
N.Q. Tuan et al. / Phytochemistry xxx (2016) 1e6
rapid decrease in its population (Korea National Arboretum, 2009).
The leaves of R. brachycarpum have been traditionally used for the
treatment of cardiovascular conditions, hepatitis, hypertension,
rheumatoid arthritis, and headache. Some of these ethno-
pharmacological applications have been validated extensively (Jang
et al., 2005; Popescu and Kopp, 2013). Chemical investigations have
identified various flavonoids, oleanane- and ursane-type triterpe-
noids, and diterpenoids such as the structurally diverse grayano-
toxins as the major constituents (Choi et al., 1986, 1987; Youn and
Cho, 1991).
In ongoing endeavors to discover new chemistry from endan-
gered plant species, identification of the first grayanotox-9(11)-ene
derivative (1) and five known grayanotoxins, grayanoside B (2)
(Sakakibara et al., 1979), rhodomoside A (3) (Bao et al., 2003),
piersformoside (4) (Wang et al., 2000), grayanotoxin III (5)
(Masutani et al., 1979), and grayanotoxin I (6) (Lechtenberg et al.,
2014) from the MeOH extract of R. brachycarpum leaves is
described herein (see Fig. 1).
spectra showed close similarities to those of pieroside C from Pieris
japonica (Ericaceae), a grayanoside carrying grayanotox-9(10)-ene
as its aglycone moiety (Kaiya and Sakakibara, 1985), indicating
the presence of an A-nor-B-homo-ent-kaurene scaffold in 1 (Li
et al., 2013). Structural differences in 1 included a change in the
location and nature of the olefinic moiety and a molecular mass
increase of 16 atomic mass units. The latter implied that compound
1 carried an additional hydroxy group compared to pieroside C. The
HMBC cross-peaks from H-11 and H-20 to C-10, and from H-20 to C-
9 and C-1 located the olefinic and hydroxy moieties as depicted in
Fig. 2, confirming that compound 1 indeed possesses an unprece-
dented grayanotox-9(11)-ene framework. The connectivity of the
aglycone and the monosaccharide unit was established based on
the HMBC cross-peak between H-3 (dH 4.16) and C-10 (dC 106.6). The
7.7 Hz coupling constant of the anomeric proton indicated a
b-
glucosidic linkage and acid hydrolysis followed by LC analysis
verified the presence of a D-(þ)-glucopyranose moiety (Fig. S9).
The relative configuration of 1 was assigned by analysis of the
splitting patterns and coupling constants of diastereotopic protons
and ROESY correlations. The splitting pattern and coupling con-
2. Results and discussion
stants of H-2
moiety comprising the A-ring predominantly adopted a C5-endo
envelope conformation where H-2 and H-3 are eclipsed (Fig. 3A).
The ROESY correlation from H-1 to H3-18 established a trans-fused
A/B-ring junction. The respective - and -orientations of 6-OH and
10-OH were assigned via the ROESY correlations from H-1 to H-6,
and from H-2 to H-20, respectively (Fig. 3B). The ROESY correla-
tion from H-6 to H-14 led to the assignment of the fused mode of
the D-ring with H-13 being -oriented (Fig. 3B). The -orientation
a
(td, J ¼ 13.1, 7.3 Hz) indicated that the cyclopentane
Compound 1 was obtained as a white amorphous powder. Its
molecular formula was established as C26H42O10 based on the
HRESIMS sodium adduct ion at m/z 537.2670 (calcd [MþNa]þ, m/z
537.2676) and 13C NMR data (Table 1). The 1H NMR spectrum
a
b
a
exhibited resonances for one olefinic proton
(dH 5.35), one
b
anomeric proton (dH 5.04), two oxymethine protons (dH 4.14 and
4.17), and two secondary (dH 1.47 and 1.51) and tertiary methyl
group protons (dH 1.41 and 1.96). The 13C NMR spectrum displayed
characteristic resonances for a monosaccharide moiety (dC 106.6,
79.1, 78.9, 76.2, 72.5, and 63.6), two olefinic carbons (dC 117.6 and
156.5), three oxygenated tertiary carbons (dC 86.1, 80.8, and 77.6),
two oxygenated secondary carbons (dC 87.9 and 73.7), two qua-
ternary carbons (dC 51.2 and 46.2), and two tertiary (dC 20.8 and
28.8) and secondary methyl carbons (dC 28.3 and 30.4). Its 1D NMR
a
a
of the C-16 hydroxy group in the D-ring was established via the
ROESY correlation between H2-12 and H3-17 (Fig. 3B).
The assignment of the absolute configuration was attempted by
comparison of the experimental and calculated electronic circular
dichroism (ECD) spectra (Niu et al., 2016). An initial conformational
search was conducted using the MMFF force field in the Macro-
€
Model module (Schrodinger LLC) and the geometry of the identified
conformers were further optimized at the B3LYP/6-31þ(d) level
(Tables S1 and S2). The resulting conformers were subjected to ECD
calculations utilizing the identical theory level and the simulated
ECD spectra were Boltzmann-averaged based on the Gibbs free
energy of each conformer. The comparison of the experimental and
the computed ECD spectra, however, was inconclusive (Fig. 4),
presumably due to the presence of a weak chromophore in 1.
Alternatively, the full 3D structural assignment of 1 was established
employing calculations of shielding tensor values with support
from DP4 plus analysis capable of improving the accuracy of pre-
diction of sp2 carbon shielding tensor values (Grimblat et al., 2015).
Having established the D-glucopyranosyl moiety in 1, di-
astereomers I and II carrying enantiomeric aglycone moieties were
considered (Fig. 5). The major conformers of diastereomers I and II
were found and optimized using the aforementioned approaches
and the GIAO shielding tensors of the generated conformers were
computed at the mPW1PW91/6-31G(d,p) level utilizing the
conductor-like polarizable continuum solvation model (CPCM)
with a dielectric constant representing pyridine (Grimblat et al.,
2015). The calculated shielding tensors were Boltzmann-averaged
based on the corresponding Gibbs free energy of each conformer
and these averaged values were utilized for DP4 plus analysis
(Tables S3eS5). The DP4 plus application demonstrated the struc-
tural equivalence of diastereomer I with compound 1 with 99.95%
probability (Fig. S10). Moreover, stereochemical consistency of the
grayanane architecture originating from the ent-kaurane precursor
further supported the resulting DP4 plus analysis when considered
in conjunction with a similar biosynthesis pathway towards anal-
ogous A-nor-B-homo-diterpenoids (Li et al., 2013).
Table 1
1H (600 MHz) and 13C NMR (150 MHz) spectroscopic data for compound 1 in pyr-
idine-d5.
Position
1
dH, multi. (J in Hz)
dC
1.76, dd (13.9, 6.2)
2.80, td (13.1, 7.3)
50.0
36.0
2
2
3
4
5
6
7
7
8
9
a
b
2.51 (m)
4.17 (m)
e
87.9
51.2
86.1
73.7
45.8
e
4.14, d (11.0)
1.99, d (13.9)
a
b
3.72, dd (13.9, 11.0)
e
46.2
156.5
77.6
117.6
31.7
48.9
44.7
62.3
80.8
28.3
28.8
20.8
30.4
106.6
76.2
79.1
72.5
78.8
63.6
e
10
11
12
13
14
15
16
17
18
19
20
10
e
5.35 (br t)
2.30, d (18.4), 2.21 (m)
2.46 (br t)
2.53 (m), 2.21 (m)
2.59, br d (12.9), 2.22 (m)
e
1.51, s
1.41, s
1.96, s
1.47, s
5.04, d (7.7)
4.09, t (8.2)
4.30, t (8.7)
4.28, t (8.9)
20
30
40
50
4.05, ddd (8.6, 5.4, 2.4)
4.64, dd (11.8, 2.5), 4.44, dd (11.7, 5.5)
60
Please cite this article in press as: Tuan, N.Q., et al., A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural
assignment via a protocol combining NMR and DP4 plus application, Phytochemistry (2016), http://dx.doi.org/10.1016/j.phytochem.2016.10.010

N.Q. Tuan et al. / Phytochemistry xxx (2016) 1e6
3
Fig. 1. Structures of isolated compounds from R. brachycarpum.
spectroscopic data.
Diverse grayanoids and their glycosides displayed a wide range
of pharmacological relevance including cytotoxic, analgesic,
neurotoxicity, and sedative effects (Li et al., 2013). Previous inves-
tigation of Rhododendron plants also revealed significant antimi-
crobial activities (Choi and Rhim, 2011; Nisar et al., 2013). Based on
these studies, compounds 1e6 were evaluated for their growth
inhibitory activity against the indicator strains Gram-positive Ba-
cillus subtilis NCIB3610, Gram-negative Escherichia coli Nissle 1917,
and the fungus, Saccharomyces cerevisiae utilizing the MIC method
in triplicate. None of the tested compounds, however, were active
in this assay (MIC ꢀ 200
mg/mL).
Fig. 2. Key COSY (
were omitted for clarity of presentation.
) correlations and HMBC (/) cross-peaks in 1. Stereodescriptors
Fig. 3. Configurational and conformational analysis of 1 utilizing splitting patterns, coupling constants, and ROESY correlations (yellow dotted lines). The geometry of the ster-
eostructure was initially minimized at the MMFF force field and subsequently optimized at the B3LYP/6-31 þ G(d) level. Some protons were removed for a clearer presentation. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The known diterpenoids, grayanoside B (2) (Sakakibara et al.,
1979), rhodomoside A (3) (Bao et al., 2003), piersformoside (4)
(Wang et al., 2000), grayanotoxin III (5) (Masutani et al., 1979), and
grayanotoxin I (6) (Lechtenberg et al., 2014) were identified by
comparing the experimental and reported physical and
3. Conclusions
The current phytochemical study of the R. brachycarpum
metabolome demonstrates how new chemistry still remains un-
covered among many endangered plant species. This investigation
Please cite this article in press as: Tuan, N.Q., et al., A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural
assignment via a protocol combining NMR and DP4 plus application, Phytochemistry (2016), http://dx.doi.org/10.1016/j.phytochem.2016.10.010

4
N.Q. Tuan et al. / Phytochemistry xxx (2016) 1e6
cysteine methyl ester, phenyl isothiocyanate, and D- and L-glucose
were purchased from Sigma-Aldrich (St. Louis, MO) and used
without further purification. Sterilized 96-well plates (Costar 3595,
Corning, NY) were used for the minimum inhibitory concentration
(MIC) assay and compact microbiological incubators (Fisher Sci-
entific, Waltham, MA) and shaking incubator (Innova 4200, New
Brunswick Scientific, Enfield, CT) with corresponding temperatures
were utilized for bacterial growth.
4.2. Extraction and isolation
The collection and extraction of the plant material were elabo-
rated in our earlier communications (Zhou et al., 2013, 2014). The n-
BuOH-soluble fraction (320 g) was subjected to silica gel VLC and
eluted with a stepwise gradient of MeOH:CHCl3 (1:9, 2:8, 3:7, 4:6,
5:5; 5 L for each step) to generate six fractions (Fr. B1eB6). Fraction
Fig. 4. Overlaid experimental and average calculated ECD spectra of compound 1. The
ECD spectrum was simulated at the B3LYP/6-31þ(d) level in the gas phase.
Fig. 5. Two plausible diastereomers of 1 used for the DP4 plus application.
also validates that a protocol combining experimental NMR data
with the associated modern computational chemistry supported by
advanced statistics might be a viable and versatile tool in accom-
plishing full structural assignment of a minor secondary metabolite
in cases where conventional X-ray crystallography or chiroptical
approaches are not effective. Furthermore, this type of research
may justify the conservation of endangered plant species through
identification of the unique chemistry embodied in these plants.
B2 (19.8 g) was purified by MPLC, eluting with a gradient of MeOH
in H2O from 20 to 60% to produce 11 fractions (B2-1eB2-11).
Fraction B2-7 (360 mg) was subjected to silica gel column chro-
matography (CC) [1.5 cm (i.d.) ꢁ 60 cm (length)] and eluted under
isocratic conditions of n-hexane/acetone (1:1) to yield 5 (30 mg).
Fraction B2-11 (360 mg) was purified using the aforementioned
silica gel CC and eluted with MeOH/CHCl3 (12:88) to afford 4
(50 mg). Fraction B3 (80 g) was fractionated into 10 sub-fractions
(Fr. B3-1e3-10) utilizing MPLC with a mixture of MeOH/H2O
starting from 10 to 60% MeOH. Fraction B3-7 (3.1 g) was further
purified employing MPLC with a gradient of MeOH in CHCl3 from 5
to 20% MeOH to generate six fractions (B3-7-1eB3-7-6). Fraction
B3-7-3 (330 mg) was subjected to silica gel CC [2 cm (i.d.) ꢁ 75 cm
(length)] and eluted with a mixture of MeOH/EtOAc/n-hexane
(1:9:1) to acquire 3 (50 mg). Compound 2 (680 mg) was purified
from Fr. B3-9 (3 g) by MPLC eluting with an MeOH/CHCl3 gradient
from 5 to 20% MeOH. Fr. B4 (54 g) was divided using MPLC with a
MeOH/H2O gradient from 20 to 60% to acquire 10 fractions (B4-
1eB4-10). Fr. B4-10 (200 mg) was purified by silica gel CC [1.5 cm
(i.d.) ꢁ 60 cm (length)] and eluted with MeOH/CHCl3 (25:75) to
afford 1 (9 mg). The CHCl3-soluble fraction (185 g) was subjected to
silica gel VLC, eluting with a stepwise gradient of n-hexane/EtOAc
(9:1, 8:2, 7:3, 6:4, 5:5, 4:6, 3:7, 2:8, 1:9, 5 L for each step) and
separated into nine fractions (Fr. C1eC9). Fraction C9 (55 g) was
purified using VLC, eluting with a stepwise gradient of n-hexane/
EtOAc (9:1, 8:2, 7:3, 6:4, 5:5, 4:6, 3:7, 2:8, 1:9, 2 L for each incre-
ment) to give 10 factions (Fr. C9-1eC9-10). Fraction C9-10 (40 g)
was purified employing MPLC (KP-C18-HS) with an isocratic
mixture of MeOH/H2O (1:1) to generate nine fractions (Fr. C9-10-
1eC9-10-9). Fraction C9-10-2 (250 mg) was purified by HPLC using
4. Experimental procedures
4.1. General experimental procedures
Optical rotations were recorded on a JASCO DIP-1000 (Tokyo,
Japan) polarimeter. Vacuum liquid chromatography (VLC) was
performed utilizing silica gel (70e230 mesh, EMD Millipore,
Darmstadt, Germany). The experimental ECD spectrum was
measured using a Chirascan qCD (Applied Photophysics, UK). Me-
dium pressure liquid chromatography (MPLC) was accomplished
using Biotage Isolera™ with reversed-phase C18 SNAP cartridges
KP-C18-HS (340 g, Biotage, Charlotte, NC). Preparative reversed-
phase HPLC separation was carried out on a YMC C18 column
(250 ꢁ 20.0 mm, 5
For the identification of the monosaccharide, a Kinetex C18 column
m; Phenomenex, Torrance, CA) was used.
mm; Allentown, PA) at a flow rate of 5 mL/min.
(4.6 mm ꢁ 250 mm, 5
m
NMR spectra were recorded on a Bruker Avance III 600 MHz
spectrometer (Billerica, MA) equipped with a 5 mm direct detection
pulsed-field-gradient probe. HRESIMS data were generated by the
Korea Basic Science Institute (Chungbuk, South Korea) on a Synapt
mass spectrometer (Waters, Milford, MA). Chemicals including L-
Please cite this article in press as: Tuan, N.Q., et al., A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural
assignment via a protocol combining NMR and DP4 plus application, Phytochemistry (2016), http://dx.doi.org/10.1016/j.phytochem.2016.10.010

N.Q. Tuan et al. / Phytochemistry xxx (2016) 1e6
5
a gradient of MeOH/eH2O (0e10 min, 20% MeOH; 10e15 min,
20e40% MeOH; 15e80 min, 40e50% MeOH; 80e85 min, 50e100%
MeOH; 85e115 min, 100% MeOH) to furnish 6 (50 mg). While the n-
hexane- and CHCl3-soluble fractions, as well as the B1, B5, B6, and
C1eC8 fractions, were also analyzed by utilizing the HPLC and 1H
NMR approaches described for the identification of the grayano-
toxin derivatives, the latter types of compounds do not appear to be
present in significant quantities (Fig. S11). For this HPLC profiling
Boltzmann population in the associated Gibbs free energy
(Table S5) and utilized for DP4 plus analysis (Fig. S10) facilitated by
an Excel sheet provided by the original authors (Grimblat et al.,
2015).
4.4. Antimicrobial activity tests (MIC tests)
The bacterial strains, Bacillus subtilis and Escherichia coli were
grown in 5 mL of LB broth (yeast extract 5 g, tryptone 10 g, NaCl 10 g
per 1 L H2O), and fungal strain Saccharomyces cerevisiae was culti-
vated overnight in 5 mL of YM broth (yeast extract 5 g, malt 30 g per
1 L H2O) at 30 ꢄC. Each microbial culture broth was sub-cultured
into the according fresh media and grown until the cultivation
reaches an OD600 value ¼ 0.1 in a shaking incubator (250 rpm,
30 ꢄC). Compounds 1e6 and positive controls (ampicillin for the
bacteria and amphotericin B for the fungal strain) were dissolved in
process, a YMC-pack C18 column [5
m
, 120 Å, 4.6 mm (i.d.) ꢁ 150 mm
(length); YMC, Japan] was used with a programmed gradient
comprising 0e10 min, 20% MeOH; 10e15 min, 20e40% MeOH;
15e40 min, 40e50% MeOH; flow rate: 1 mL/min.
4.2.1. Compound 1
White amorphous powder; ½a D25
ꢂ
¼ ꢃ57:8 (c 0.1, MeOH); For 1H
NMR and 13C NMR spectroscopic data, see Table 1; HRESIMS m/z
537.2670 [MþNa]þ (calcd for C26H42O10Na, 537.2676).
DMSO (10
dilution was initiated from a maximum final concentration of
200 g/mL of the tested compounds and cells were dispensed over
mg/mL), and DMSO was used as a negative control. Serial
4.2.2. Acid hydrolysis of 1
m
96-well plates. The plates were incubated at 30 ꢄC for 24 h. The MIC
was determined at the point of complete inhibition of compounds
against the indicator strains and all experiments were performed in
triplicate.
Compound 1 (2 mg) was heated in a solution of 10% HCl (1 mL)
at 85 ꢄC for 6 h, under reflux conditions, and the residue was
extracted with EtOAc (3 mL). The aqueous layer containing the
monosaccharide moiety was dissolved in anhydrous pyridine
(1 mL) and heated with L
-cysteine methyl ester (6 mg) at 60 ꢄC for
1 h. Phenylisothiocyanate (0.1 mL) was added to the reaction
mixture and heated at 60 ꢄC for 1 h. The reaction mixture was
evaporated, dissolved in MeCN, and analyzed eluting with a mobile
phase of MeOH/H2O (25:75, flow rate of 1 mL/min). The derivatized
sugar moiety was detected at the UV wave length 250 nm.
Acknowledgments
This research was supported by the Basic Science Research
Program (2014R1A2A2A01006793) and Priority Research Centers
Program (NRF-2009-0093815) through the National Research
Foundation of Korea (NRF) grant funded by the Korean Govern-
ment. This work was also supported by National Forensic Service
(NFS2015DNT06), Ministry of the Interior, Republic of Korea.
Authentic
identical protocol. By comparing the retention time of the sugar
derivative obtained from 1 with those of the standard -glucose (tR
41.60 min) and -glucose (tR 41.90 min) derivatives, the carbohy-
D- and L-glucose were derivatized and analyzed using the
D
L
drate moiety of compound 1 was defined as
D-glucose (Fig. S9).
Appendix A. Supplementary data
4.3. Computational details
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.phytochem.2016.10.010.
4.3.1. ECD simulation
All conformational searches for ECD simulation and DP4 plus
analysis were accomplished employing Macromodel (Version 11.0,
Schrodinger LLC) program with “mixed torsional/low mode sam-
pling” in the MMFF force field. The searches were conducted in the
gas phase with a 50 kJ/mol energy window limit and 10,000
maximum number of steps to thoroughly probe into all low-energy
conformers. Minimization processes were achieved utilizing the
PolakeRibiere conjugate gradient (PRCG) method, 10,000
maximum iterations, and a 0.001 kJ (mol Å)ꢃ1 convergence
threshold on the root mean square gradient. The geometry of
conformers within 10 kJ/mol were further optimized at the B3LYP/
6-31 þ G(d) level with tight convergence (Tables S1 and S2) and
ECD calculations were performed at the identical theory level and
basis sets. The generated excitation energies and rotational
strengths were Boltzmann-averaged based on the calculated Gibbs
free energy of each conformer and used for the ECD visualization
utilizing SpecDis (Fig. 4) (Bruhn et al., 2011).
References
Bao, G.-H., Wang, L.-Q., Cheng, Y.-H., Li, X.-Y., Qin, G.-W., 2003. Diterpenoid and
phenolic glycosides from the roots Rhododendron molle. Planta Med. 69,
434e439.
Bruhn, T., Hemberger, Y., Schaumloffel, A., Bringmann, G., 2011. SpecDis, Version
1.51. University of Wuerzburg, Germany.
Choi, J.S., Young, H.S., Park, J.C., Choi, J.H., Woo, W.S., 1986. Flavonoids from the
Rhododendron brachycarpum. Arch. Pharm. Res. 9, 233e236.
Choi, J.S., Young, H.S., Park, J.C., Choi, J.H., Woo, W.S., 1987. Further study on the
constituents of Rhododendron brachycarpum. Arch. Pharm. Res. 10, 169e172.
Choi, M.-Y., Rhim, T.-J., 2011. Antimicrobial effects against food-borne pathogens
and antioxidant activity of Rhododendron brachycarpum extract. J. Korean Soc.
Food Sci. Nutr. 40, 1353e1360.
Grimblat, N., Zanardi, M.M., Sarotti, A.M., 2015. Beyond DP4: an improved proba-
bility for the stereochemical assignment of isomeric compounds using quantum
chemical calculations of NMR shifts. J. Org. Chem. 80, 12526e12534.
Jang, G.U., Chooi, U.S., Lee, K.R., 2005. Cytotoxic constituents of Rhododendron
brachycarpum. Yakhak Hoeji 49, 244e248.
Kaiya, T., Sakakibara, J., 1985. Occurrence of pieroside C, a grayanotox-9-ene de-
rivative, in Pieris japonica. Chem. Pharm. Bull. 33, 4637e4639.
Korea National Arboretum, 2009. http://www.kna.go.kr/upload/publish/19965/01/
1291686426891_deryn7msdM.html.
Lechtenberg, M., Dierks, F., Sendker, J., Louis, A., Schepker, A., Hensel, A., 2014.
Extracts from Rhododendron ferrugineum do not exhibit grayanotoxin I: an
analytical survey on grayanotoxin I within the genus Rhododendron. Planta
Med. 80, 1321e1328.
4.3.2. DP4 plus analysis
For the DP4 plus application, conformers of diastereomers I and
II within 5 kJ/mol found in the MMFF force field were considered
and the geometry of the conformers was optimized at the B3LYP/6-
31G(d) level in the gas phase (Tables S3 and S4) (Grimblat et al.,
2015). The shielding tensor values of the optimized conformers
were computed at the mPW1PW91/6-31G(d,p) level in the polar-
izable continuum solvation model with a dielectric constant rep-
resenting pyridine. These values were averaged based on the
Lee, S.W., Kim, Y.M., Jang, S.S., Chung, J.M., 2002. Genetic variation and structure of
Rhododendron brachycarpum D. Don, a rare and endangered tree species in
Korea. Silvae Genet. 21, 215e219.
Li, Y., Liu, Y.-B., Yu, S.-S., 2013. Grayanoids from the Ericaceae family: structures,
biological activities and mechanism of action. Phytochem. Rev. 12, 305e325.
Lodewyk, M.W., Soldi, C., Jones, P.B., Olmstead, M.M., Rita, J., Shaw, J.T., Tantillo, D.J.,
2012. The correct structure of aquatolide-experimental validation of
a
Please cite this article in press as: Tuan, N.Q., et al., A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural
assignment via a protocol combining NMR and DP4 plus application, Phytochemistry (2016), http://dx.doi.org/10.1016/j.phytochem.2016.10.010

6
N.Q. Tuan et al. / Phytochemistry xxx (2016) 1e6
theoretically-predicted structural revision. J. Am. Chem. Soc. 134, 18550e18553.
diastereoisomers using GIAO NMR shift calculation. J. Org. Chem. 74,
4597e4607.
Smith, S.G., Goodman, J.M., 2010. Assigning stereochemistry to single di-
astereoisomers by GIAO NMR calculation: the DP4 probability. J. Am. Chem. Soc.
132, 12946e12959.
Masutani, T., Kawazu, K., Uneyama, K., Torii, S., Iwasa, J., 1979. Assignment of 13C-
NMR spectra of grayanotoxin- I and - III. Agric. Biol. Chem. 43, 631e635.
Nicolaou, K., Vourloumis, D., Winssinger, N., Baran, P.S., 2000. The art and science of
total synthesis at the dawn of the twenty-first century. Angew. Chem. Int. Ed.
39, 44e122.
Wang, L.-Q., Chen, S.-N., Cheng, K.-F., Qin, C.-S., Qin, G.-W., 2000. Diterpene glu-
cosides from Pieris formosa. Phytochemistry 54, 847e852.
Nisar, M., Ali, S., Qaisar, M., 2013. Antibacterial and cytotoxic activities of the
methanolic extracts of Rhododendron arboreum. J. Med. Plants Res. 7, 398e403.
Niu, C.S., Li, Y., Liu, Y.B., Ma, S.G., Li, L., Qu, J., Yu, S.S., 2016. Analgesic diterpenoids
from the twigs of Pieris formosa. Tetrahedron 72, 44e49.
Oh, J., Bowling, J.J., Zou, Y., Chittiboyina, A.G., Doerksen, R.J., Ferreira, D.,
Leininger, T.D., Hamann, M.T., 2013. Configurational assignments of conforma-
tionally restricted bis-monoterpene hydroquinones: utility in exploration of
endangered plants. Biochim. Biophys. Acta Gen. Subj. 1830, 4229e4234.
Popescu, R., Kopp, B., 2013. The genus Rhododendron: an ethnopharmacological and
toxicological review. J. Ethnopharmacol. 147, 42e62.
Waters, A.L., Oh, J., Place, A.R., Hamann, M.T., 2015. Stereochemical studies of the
karlotoxin class using NMR spectroscopy and DP4 chemical-shift analysis: in-
sights into their mechanism of action. Angew. Chem. Int. Ed. 54, 15705e15710.
Youn, H., Cho, J.H., 1991. Isolation and structure elucidation of triterpenoidal con-
stituents from the leaves of Rhododendron brachycarpum. Korean J. Pharmacogn.
22, 18e21.
Zhou, W., Oh, J., Lee, W., Kwak, S., Li, W., Chittiboyina, A.G., Ferreira, D.,
Hamann, M.T., Lee, S.H., Bae, J.-S., Na, M., 2014. The first cyclomegastigmane
rhododendronside
A from Rhododendron brachycarpum alleviates HMGB1-
Sakakibara, J., Shirai, N., Kaiya, T., Nakata, H., 1979. Grayanotoxin-XVIII and graya-
noside B, a new A-nor-B-homo-ent-kaurene and its glucoside from Leucothoe
grayana. Phytochemistry 18, 135e137.
induced septic. Biochim. Biophys. Acta Gen. Subj. 1804, 2042e2049.
Zhou, W., Oh, J., Li, W., Kim, D.W., Lee, S.H., Na, M., 2013. Phytochemical studies of
Korean endangered plants: a new flavone from Rhododendron brachycarpum G.
Don. Bull. Korean Chem. Soc. 34, 2535e2538.
Smith, S.G., Goodman, J.M., 2009. Assigning the stereochemistry of pairs of
Please cite this article in press as: Tuan, N.Q., et al., A grayanotox-9(11)-ene derivative from Rhododendron brachycarpum and its structural
assignment via a protocol combining NMR and DP4 plus application, Phytochemistry (2016), http://dx.doi.org/10.1016/j.phytochem.2016.10.010