Cyclic Diarylheptanoids from Corylus avellana Green Leafy Covers: Determination of Their Absolute Configurations and Evaluation of Their Antioxidant and Antimicrobial Activities

Article abstract of DOI:10.1021/acs.jnatprod.6b00703

The methanol extract of the leafy covers of Corylus avellana, source of the Italian PGI (protected geographical indication) product Nocciola di Giffoni , afforded two new cyclic diarylheptanoids, giffonins T and U (2 and 3), along with two known cyclic diarylheptanoids, a quinic acid, flavonoid-, and citric acid derivatives. The structures of giffonins T and U were determined as highly hydroxylated cyclic diarylheptanoids by 1D and 2D NMR experiments. Their relative configurations were assigned by a combined quantum mechanical/NMR approach, comparing the experimental ¹³C/¹H NMR chemical shift data and the related predicted values. The absolute configurations of carpinontriol B (1) and giffonins T and U (2 and 3) were assigned by comparison of their experimental electronic circular dichroism curves with the TDDFT-predicted curves. The ability of the compounds to inhibit the lipid peroxidation induced by H₂O₂ and H₂O₂/Fe²⁺ was determined by measuring the concentration of thiobarbituric acid reactive substances. Furthermore, the antimicrobial activity of the methanol extract of leafy covers of C. avellana and of the isolated compounds against the Gram-positive strains Bacillus cereus and Staphylococcus aureus and the Gram-negative strains Escherichia coli and Pseudomonas aeruginosa was evaluated. Carpinontriol B (1) and giffonin U (3) at 40 μg/disk caused the formation of zones of inhibition.

Full text of DOI:10.1021/acs.jnatprod.6b00703

Article
pubs.acs.org/jnp
Cyclic Diarylheptanoids from Corylus avellana Green Leafy Covers:
Determination of Their Absolute Configurations and Evaluation of
Their Antioxidant and Antimicrobial Activities
Antonietta Cerulli,†,‡,∥ Gianluigi Lauro,†,∥ Milena Masullo,† Vincenza Cantone,†,‡ Beata Olas,§
Bogdan Kontek,§ Filomena Nazzaro,⊥ Giuseppe Bifulco,*,† and Sonia Piacente*,†
†Dipartimento di Farmacia and ‡Ph.D. Program in Drug Discovery and Development, Universita
Paolo II 132, 84084 Fisciano (SA), Italy
̀
degli Studi di Salerno, Via Giovanni
§Department of General Biochemistry, Institute of Biochemistry, Faculty of Biology and Environmental Protection, University of
Lodz, Pomorska 141/3, 90-236, Lodz, Poland
⊥Istituto di Scienze dell’Alimentazione CNR-ISA, Via Roma 64, 83100 Avellino, Italy
S
* Supporting Information
ABSTRACT: The methanol extract of the leafy covers of Corylus
avellana, source of the Italian PGI (protected geographical indication)
product “Nocciola di Giffoni”, afforded two new cyclic diary-
lheptanoids, giffonins T and U (2 and 3), along with two known
cyclic diarylheptanoids, a quinic acid, flavonoid-, and citric acid
derivatives. The structures of giffonins T and U were determined as
highly hydroxylated cyclic diarylheptanoids by 1D and 2D NMR
experiments. Their relative configurations were assigned by a combined
quantum mechanical/NMR approach, comparing the experimental
13C/1H NMR chemical shift data and the related predicted values. The
absolute configurations of carpinontriol B (1) and giffonins T and U (2
and 3) were assigned by comparison of their experimental electronic
circular dichroism curves with the TDDFT-predicted curves. The
ability of the compounds to inhibit the lipid peroxidation induced by H2O2 and H2O2/Fe2+ was determined by measuring the
concentration of thiobarbituric acid reactive substances. Furthermore, the antimicrobial activity of the methanol extract of leafy
covers of C. avellana and of the isolated compounds against the Gram-positive strains Bacillus cereus and Staphylococcus aureus and
the Gram-negative strains Escherichia coli and Pseudomonas aeruginosa was evaluated. Carpinontriol B (1) and giffonin U (3) at 40
μg/disk caused the formation of zones of inhibition.
The kernel is the nut of commerce, while the skin, the hard
shell, and the green leafy cover as well as the tree leaf may be
considered waste products of the food industry.2 Although
some reports have been published regarding the antioxidant
activity of some byproducts of C. avellana,1,2,5−7 there is limited
information on the chemical composition of byproducts that
could represent a source of natural products with potential
biological activities. Previously, we reported 16 diarylhepta-
noids, giffonins A−P, from the leaves of the C. avellana cultivar
“Tonda di Giffoni”6,8 and three diarylheptanoids, giffonins Q−
S, from the flowers of the C. avellana cultivar “Tonda di
Giffoni”,9 some of which prevented oxidative damage of human
plasma lipids, induced by H2O2 and H2O2/Fe2+. To date, no
studies are reported on the chemical composition of green leafy
covers of C. avellana. Nuts develop in clusters of 1−12, each
separately enclosed in an involucre made up of two
overlapping, leafy bracts (modified leaves) that vary consid-
azelnut (Corylus avellana L.), belonging to the Betulaceae
Hfamily, is a well-known nut of which production on a
worldwide basis ranks second after almond.1 Hazelnuts are a
source of fats, proteins, and vitamins and are widely used in
dairy, bakery, confectionery, candy, and chocolate products.2
Italy is the second largest hazelnut producer in the world after
Turkey, and 98% of its production is due to four regions:
Latium, Piedmont, Sicily, and Campania. In the Campania
region the cultivars mainly used by the food industry are
“Mortarella” (38%) and “San Giovanni” (37%), along with
“Tonda di Giffoni” (12%), which has been awarded the
protected geographical indication (PGI) mark by the European
Union as “Nocciola di Giffoni”.3
Despite its wide cultivation for nut collection, hazelnut tree
leaves are also consumed as an infusion. They are used in folk
medicine for the treatment of hemorrhoids, varicose veins,
phlebitis, and edema, as a consequence of their astringent,
vasoprotective, and antiedema properties and also for their mild
antimicrobial effects.1,4
Received: July 29, 2016
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Chart 1
erably across the Corylus species in terms of length, constriction
around the nut, indentation and serration at the apex, and
thickness at the base. In C. avellana this cup of green leafy cover
encloses about three-quarters of the nut. The hazelnut green
leafy covers are removed from the nuts soon after harvesting
and have no current commercial value, but are occasionally
used as fertilizer for the hazelnut trees upon composting.10 On
the basis of the occurrence of phenolic antioxidants with
potential health benefits in the leaves of C. avellana, a
phytochemical investigation of the green leafy covers of
C. avellana cultivar “Tonda di Giffoni” has been carried out.
Herein, the isolation and structural elucidation of two new
cyclic diarylheptanoids (2 and 3), named giffonins T and U, are
reported. In order to determine the absolute configurations, the
relative configurations of compound 1, carpinontriol B,11 and
compounds 2 and 3 have been established by a combined QM
(quantum mechanical)/NMR approach, using a comparison of
the experimental 13C/1H NMR chemical shift data and the
related predicted values. To establish the absolute config-
urations of compound 1 and giffonins T and U (2 and 3),
comparison of their experimental electronic circular dichroism
(ECD) spectra with the TDDFT-predicted curves was carried
out. Furthermore, to explore the antioxidant ability of
C. avellana green leafy covers, the isolated compounds were
B
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
Table 1. H NMR and 13C NMR Data of Compounds 1−3 (Methanol-d4)
1
2
3
δC
δH (J in Hz)
δC
δH (J in Hz)
δC
δH (J in Hz)
1
2
3
4
5
6
7
127.1
127.0
153.3
117.1
129.2
130.7
37.1
128.4
127.7
151.9
117.3
130.6
130.2
37.1
126.8
128.4
153.1
116.8
130.6
130.6
39.5
6.78, d (8.1)
6.81, d (8.1)
6.81, d (8.2)
7.02, dd (8.1, 1.8)
7.05, dd (8.1, 1.8)
7.09, dd (8.2, 1.8)
2.91, d (15.8, 11.3)
3.07, dd (15.8, 4.2)
4.74, dd (11.3, 4.2)
3.93, d (10.1)
2.88, dd (15.8, 12.2)
3.09, dd (15.8, 4.4)
4.67, dd (12.2, 4.4)
3.74, d (10.1)
2.94, dd (16.3, 9.6)
3.01, dd (16.3, 3.8)
4.22, dd (9.6, 3.8)
4.41, br s
8
68.5
69.7
68.5
69.5
71.9
70.1
9
10
11
12
78.6
4.25, d (10.1)
78.4
4.21, d (10.1)
77.6
4.35, d (5.6)
215.0
37.1
215.3
37.0
81.7
4.27, d (5.6)
2.97, ddd (19.6, 5.2, 2.0)
3.54, ddd (19.6, 12.6, 2.0)
2.87, ddd (16.6, 5.2, 2.0)
3.18, ddd (16.6, 12.6, 2.0)
2.95, m
216.9
3.55, m
2.94, m
13
24.9
24.7
43.7
3.13, d (12.0)
4.74, d (12.0)
3.15, ddd (17.2, 12.6, 2.0)
14
15
16
17
18
19
1
130.7
129.0
117.1
152.7
134.6
134.4
132.8
129.7
115.4
152.6
135.9
129.9
130.1
116.8
153.1
136.9
135.6
7.08, dd (8.1, 1.8)
6.81, d (8.1)
7.23, dd (8.1, 1.8)
7.20, d (8.1)
7.06, dd (8.1, 1.8)
6.83, d (8.1)
6.39, d (1.8)
6.68, d (1.8)
6.30, d (1.8)
6.54, d (1.8)
7.25, d (1.8)
6.68, d (1.8)
135.8 β-Glc (at C-17)
102.3
74.6
78.0
71.2
77.9
62.3
5.12, d (7.6)
2
3.48, dd (9.0, 7.6)
3.54, dd (9.0, 9.0)
3.41, dd (9.0, 9.0)
3.52, m
3
4
5
6
3.73, dd (12.0, 4.5)
3.93, dd (12.0, 2.5)
evaluated for their inhibitory effects on human plasma lipid
peroxidation induced by H2O2 and H2O2/Fe2+, by measuring
the concentration of TBARS (thiobarbituric acid reactive
substances). On the basis of the antimicrobial activity reported
for hazelnut tree leaves and for some diarylheptanoid
derivatives,1,12,13 the antimicrobial activity of the methanol
extract of C. avellana leafy covers and of isolated compounds
against Bacillus cereus, Staphylococcus aureus, Escherichia coli, and
Pseudomonas aeruginosa was evaluated.
carbon signals, typical of a cyclic diarylheptanoid.6,8 HMBC and
COSY cross-peaks permitted the assignment of the three
hydroxy groups to C-8 (δ 68.5), C-9 (δ 69.7), and C-10 (δ
78.6). The position of the carbonyl group was determined by
2D NMR data; in particular, the HMBC cross-peaks between
the signals of H-9 (δ 3.93), H-10 (δ 4.25), H2-12 (δ 3.54, 2.97),
and H2-13 (δ 3.18, 2.87) and the carbon resonance at δ 215.0
allowed the carbonyl group to be located at C-11 of the heptyl
moiety. The ROESY experiment showed correlations of H-19
(δ 6.68) with H-18 (δ 6.39), H-8 (δ 4.74), and H-9 (δ 3.93), as
well as correlations of H-18 (δ 6.39), H2-12 (δ 3.54), and H2-13
(δ 3.18). Further correlations of H-8 (δ 4.74) with H-9 (δ 3.93)
and H-10 (δ 4.25) were observed. However, ROESY
correlations cannot define the orientations of the hydroxy
groups on the heptyl moiety chain because of its conforma-
tional mobility. On the basis of the aforementioned data, the
2D structure of compound 1 was elucidated as depicted.
RESULTS AND DISCUSSION
■
The MeOH extract of the leafy covers of C. avellana was
fractionated by size-exclusion chromatography, and the
resulting fractions were purified by semipreparative HPLC to
afford compounds 1−3. Their 2D structures were defined by
1D and 2D NMR experiments in combination with HRESIMS
analyses.
1
The H and 13C NMR spectra of compounds 2 and 3, in
The HRESIMS data of 1 (m/z 343.1188 [M − H]−, calcd for
C19H19O6, 343.1182) in combination with the 13C NMR data
showed the molecular formula C19H20O6. In the IR spectrum
an absorption maximum at 1720 cm−1 due a ketocarbonyl
group was evident. The 1H NMR data showed signals
ascribable to two 1,2,4-trisubstituted aromatic rings at δ 7.08
(dd, J = 8.1, 1.8 Hz), 7.02 (dd, J = 8.1, 1.8 Hz), 6.81 (d, J = 8.1
Hz), 6.78 (d, J = 8.1 Hz), 6.68 (d, J = 1.8 Hz), and 6.39 (d, J =
comparison with those of 1, suggested they were also
diarylheptanoid derivatives.
The molecular formula of 2 was established as C25H30O11 by
HRESIMS (m/z 505.1717 [M − H]−, calcd for C25H29O11,
505.1710) and the 13C NMR data. The NMR data of 2 revealed
that it differed from 1 by the presence of a β-glucopyranosyl
unit (δ 5.12) (Table 1). The D-configuration of the glucose unit
was established via hydrolysis of 2 with 1 N HCl,
trimethylsilation, and GC analysis.14 The linkage site of the
sugar unit on the diaryl moiety was obtained from the HMBC
spectrum, which showed a cross-peak between H-1glc (δ 5.12)
1
1.8 Hz) (Table 1). The H NMR spectrum showed further
signals corresponding to three oxymethine protons at δ 4.74
(dd, J = 11.3, 4.2 Hz), 4.25 (d, J = 10.1 Hz), and 3.93 (d, J =
10.1 Hz). The 13C NMR spectrum of compound 1 showed 19
C
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
e
Table 2. 13C/1H MAE (ppm) Values and DP4+ Data Reported for the Possible Stereoisomers of Compounds 1 and 3
a
13C MAE = (∑[|(δexp − δcalcd)|])/n, summation through n of the absolute error values (difference of the absolute values between corresponding
experimental and 13C chemical shifts), normalized to the number of chemical shifts. Chemical shift data reported were produced using the
“multistandard” approach, using TMS as reference compound for sp3 13C atoms and benzene for sp2 13C atoms; the related data are reported in
b
Tables S5, S7, S9, S10, Supporting Information. 1H MAE = (∑[|(δexp − δcalcd)|])/n, summation through n of the absolute error values (difference of
the absolute values between corresponding experimental and 1H chemical shifts), normalized to the number of chemical shifts. Chemical shift data
reported were produced using the “multistandard” approach, using TMS as reference compound for sp3 1H atoms and benzene for sp2 1H atoms; the
c
related data are reported in Tables S6, S8, S9, S10, Supporting Information. DP4+ probabilities related to the set of data reported in Tables S1, S3
(13C chemical shift set of data), Tables S2, S4 (1H chemical shift set of data), Supporting Information. This set of data was produced using only TMS
as reference compound, and then sp3/sp2 atoms were differently treated following the “multistandard” approach flagging them in the DP4+ Excel file.
d
The absolute configurations of compounds 1−3 were determined after comparison of the experimental and predicted ECD spectra (see Figure 1).
Starting from the identified absolute configurations of 1−3, we also predicted the ECD spectra of aR,8R,9S,10R-1 and aR,8R,9R,10R,11R-3, namely,
the atropisomers differing from (aS,8R,9S,10R)-1 and (aS,8R,9R,10R,11R)-3 for the inversion of the axis of chirality. Their predicted CD spectra
showed a poor superposition with the experimental curves and an opposite behavior compared with those of aS,8R,9S,10R-1 and aS,8R,9R,10R,11R-3
(see Figure S15, Supporting Information), suggesting that the ECD transitions are strongly affected by the geometry of the biphenyl moiety and
e
corroborating the aS axial chirality for the investigated compounds. Stereochemistry of compound 2 was considered the same as 1, since they differ
by the presence of a β-D-glucopyranosyl unit. In the last two columns, the predicted absolute configurations and the related chemical structures of 1−
3 are reported.
and C-17 (δ 152.6). Thus, the 2D structure of giffonin T (2)
was established as shown.
(δ 4.41, 4.35, 4.27, 4.22) proton resonances correlating to C-9
(δ 70.1), C-10 (δ 77.6), C-11 (δ 81.7), and C-8 (δ 71.9),
respectively (Table 1). The carbonyl group was located at C-12
(δ 216.9), on the basis of the HMBC cross-peaks between H2-
13 (δ 4.74, 3.13), H-10 (δ 4.35), and H-11 (δ 4.27) and the
carbonyl resonance at δ 216.9. In the ROESY spectrum
correlations of H-18 (δ 7.25) with H-19 (δ 6.68) and H-10 (δ
4.35) and of H-8 (δ 4.22) with H-9 (δ 4.41), H-10 (δ 4.35),
and H-7 (δ 2.94) were observed. The 2D structure of giffonin
U (3) was thus established as shown.
The 13C NMR and HRESIMS data of 3 (m/z 359.1136 [M
− H]−, calcd for C19H19O7, 359.1131) suggested a molecular
formula of C19H20O7. A carbonyl group was evident in the IR
spectrum at 1730 cm−1. Analysis of the NMR data showed that
compound 3 differed from 1 regarding the presence of an
additional secondary hydroxy group and the location of the
carbonyl function (Table 1). In particular, HSQC data
confirmed a diarylheptanoid core structure with oxymethine
D
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
In our previous investigations of the leaves of C. avellana, 16
cyclic diarylheptanoids, named giffonins A−P, were isolated.
Chemically, giffonins A−I are characterized by the presence of
only one stereogenic center on the heptyl moiety; for these
compounds the absolute configuration was established through
the application of the modified Mosher’s method.6 For giffonins
J−P, possessing at least two stereogenic centers on the heptyl
unit, a combined QM/NMR approach was used to establish the
relative configurations.15 In the present investigation, the QM/
NMR approach was employed to determine the relative
configurations of compounds 1−3 through the comparison of
the experimental 13C/1H NMR chemical shift data and the
related predicted values. The latter data were computed on all
the possible diastereoisomers,16−21 also taking into account the
atropisomers arising from the hindered rotation along the biaryl
axis. Specifically, QM/NMR calculations were performed for
compounds 1 and 3, since the experimental data revealed that 2
differs from 1 by the presence of a β-D-glucopyranosyl unit,
while maintaining the same relative configuration. First, the
experimental NMR data for compound 1 showed a high
similarity to those reported for carpinontriol B from Carpinus
cordata.11 In this case, the QM/NMR combined approach was
used for confirming the established relative configuration, while
additional stereochemical information arising from the
preferred atropisomeric forms was reported. Moreover, the
absolute configurations of 1 and 3 were established by
comparing the calculated and experimental ECD spectra.
As previously described for analogous compounds,8 an
extensive conformational search related to all the possible
diastereoisomers of 1 and 3 was required for the subsequent
phases of computation of the NMR parameters. First, the
conformational search was performed at the empirical level
(molecular mechanics, MM), combining Monte Carlo molec-
ular mechanics (MCMM), low-mode conformational sampling
(LMCS), and molecular dynamics (MD) simulations (see the
Experimental Section). The accurate analysis of these
representative conformers highlighted a significant conforma-
tional variability mainly due to the flexible heptyl moieties, and
the distribution of intramolecular H-bonds between the
hydroxy groups on adjacent carbons rather influenced the
energy of the related conformers weighted in the Boltzmann
distribution. Also, the hypothesized hindered rotation along the
o-disubstituted biaryl axis determines one preferential arrange-
ment of the two phenyl moieties (aR or aS absolute
configuration) giving rise to atropisomerism.22,23 Importantly,
the starting MM sampling generated conformers for both
groups of possible atropisomers for each diastereoisomer,
which were subsequently submitted to geometry and energy
optimization steps at the density functional level of theory
(DFT). After the optimization of the geometries, the
conformers were visually inspected in order to avoid further
possible redundant conformers. DFT calculations were
employed for predicting the rotational energy barrier related
to the interconversion between the atropisomers (Experimental
Section), specifically on the aR*/aS* atropisomers of
8S*,9R*,10R*-1 as a representative system for all the
investigated compounds. The difference between the energy
of the most energetically favored conformer and the related
energy of the transition state indicated ΔG⧧ = 20.7 kcal/mol,
thus suggesting hindered rotation about the biphenyl bond.22
Specifically, this parameter was calculated for the
aR*,8S*,9R*,10R* (1a)/aS*,8S*,9R*,10R* (1b) atropisomers
of compound 1 as a representative system for all the
investigated diastereoisomers and showed ΔG⧧ = 20.7 kcal/
mol for 1a and ΔG⧧ = 16.6 kcal/mol for 1b (difference
between the energy of the most energetically favored
conformers of 1a and 1b and the related energy of the
transition state). The high rotational barrier calculated for 1a
(ΔG⧧ = 20.7 kcal/mol) indicated the hindered rotation about
the biphenyl bond,22 corroborating the presence of atropiso-
merism and the related hindered interconversion to the 1b
isomer. On the other hand, 1b showed a lower predicted
energy barrier (ΔG⧧ = 16.6 kcal/mol), which would allow its
interconversion to the energetically more favored and stable 1a
isomer.
For each considered compound, the NMR chemical shift
data were computed for all the possible diastereoisomers
featuring a specific relative configuration at the stereogenic
centers and at the biaryl axis (aR*/aS* atropisomeric forms)
(Charts S1 and S2, Supporting Information). For each of them,
1
the weighted averages of the predicted 13C and H NMR
chemical shifts were computed at the density functional level of
theory, accounting for the energies of the sampled conformers
on the final Boltzmann distribution (Tables S1−S8, Supporting
1
Information). Also, a first set of 13C and H NMR chemical
shift data was obtained with tetramethylsilane (TMS) as
reference compound (Tables S1−S4, Supporting Information);
afterward, the “multistandard” approach24,25 was also em-
ployed, obtaining a second set of values, using TMS as
reference only for sp3 13C and 1H atoms and benzene as
reference for sp2 13C and 1H atoms (Tables S5−S8, Supporting
Information). For each atom of the investigated molecules,
1
experimental and calculated 13C and H NMR chemical shifts
were compared, and afterward the mean absolute errors
(MAEs) for all the possible diastereoisomers were computed
(Table 2, Tables S1−S10, and Figures S16 and S17, Supporting
Information). The results highlighted the slight accordance
between 13C/1H MAEs related to the possible isomers of 1 and
3 also when using the data arising from the “multistandard”
approach, then determining the uncertainty of unambiguously
assigning the relative configurations (Table 2, Tables S1−S10,
and Figures S16 and S17, Supporting Information). For these
reasons, we also relied on the recently introduced DP4+
approach,26 which emerged as a new powerful tool for the
correct stereochemical assignment of organic compounds. In
particular, the relative configurations of compounds 1 and 3
were predicted selecting the stereoisomers with the highest
DP4+ probability (all data DP4+, namely, combining both
13C/1H NMR chemical shift data) (Table 2). The results
confirmed the 8S*,9R*,10S* configuration for compounds 1
and 2, the same as was reported for carpinontriol B.11 This was
corroborated by the perfect agreement between experimental
3
and calculated J9,10 (Jexp = 10.1 Hz, Jcalc = 10.2 Hz). The data
further suggest the axial chirality, permitting assignment of the
aR*,8S*,9R*,10S*-1 and -2 relative configurations (isomers 1c
and 2c, Supporting Information).
1
Moreover, the 1D H NMR spectra at various temperatures
of compound 1, in DMSO-d6, have been acquired, focusing on
the protons of the diphenyl moieties. As can be observed in the
1H NMR spectra of carpinontriol B (1) (Figure S18,
Supporting Information), the resonances of H-4, H-5, H-15,
H-16, H-18, and H-19 showed no significant changes in their
chemical shifts over a range of temperatures (298−373 K), a
finding in agreement with a high rotational barrier, hence
confirming the presence of atropisomers.
E
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. Comparison of the experimental ECD spectra with the TDDFT-predicted curves of compounds 1 and 3.
Figure 2. Effects of the tested compounds (1−11) (0.1−100 μM; 30 min) and curcumin (0.1−100 μM; 30 min) on plasma lipid peroxidation
induced by H2O2. The results are representative of 5−9 independent experiments and are expressed as means SD. The effect of five different
concentrations of tested compounds (0.1, 1, 10, 50, and 100 μM) was statistically significant according to the ANOVA I test, p < 0.05; n.s. p > 0.05.
Concerning compound 3, this procedure led to identification
of the 3o isomer, showing the lowest 1H MAE and the highest
DP4+ probability. We then proposed aR*,8S*,9S*,10S*,11S*-3
as the relative configuration, which was further corroborated by
the good agreement between experimental and predicted 3J10−11
(Jexp = 5.6 Hz, Jcalc = 5.1 Hz).
Once the most probable relative configurations of 1 and 3
(1c, 3o, and accordingly 2c as analogue of 1c, Charts S1 and
S2, Supporting Information) were identified, the absolute
configurations were assigned by comparing the calculated and
experimental ECD spectra of the two enantiomers.27−29
Starting from the selected conformers related to the isomers
of 1 and 3 (1c, 3o), QM calculations at the TDDFT
MPW1PW91/6-31G(d,p) level were performed in EtOH
IEFPCM to reproduce the experimental solvent environment.
As shown in Figure 1, the comparison of calculated and
experimental ECD curves permitted assignment of the absolute
configurations of 1−3 as (aS,8R,9S,10R)-1 and -2 and
(aS,8R,9R,10R,11R)-3.
In addition, giffonin I (4),6 citric acid (5),30 1-methylcitrate
(6),30 trimethylcitrate (7),30 kaempferol 3-O-rhamnopyrano-
side (8),8 3,5-dicaffeoylquinic acid (9),31 myricetin 3-O-
rhamnopyranoside (10),8 and kaempferol 3-O-(4″-trans-p-
coumaroyl)rhamnopyranoside (11)8 were isolated from the
green leafy covers of C. avellana.
Interestingly, the aglycone moiety of giffonin I (4′) shows no
stereogenic centers in the macrocyclic ring, but its biphenyl
moiety also possesses an axis of chirality. After performing an
extensive conformational search at the MM level, the
conformers were submitted to an optimization of the
geometries at the QM level. In order to assess the possible
interconversion between the aR and aS atropisomers of 4′ (4′a
F
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
and 4′b, respectively; Chart S3, Supporting Information), the
rotational barrier predicting the energy of the transition state
was computed and showed a ΔG⧧ = 17.1 kcal/mol, which may
be compatible with an interconversion between the two
atropisomers.22 These data were corroborated by the
experimental ECD spectrum of 4′ with the absence of Cotton
effects near 220 nm (Figure S19, Supporting Information).
On the basis of the antioxidant activity reported for the green
leafy covers of C. avellana5 and for giffonins A−H,6,9 the
isolated compounds were evaluated for their potential
protective properties against oxidative damage (lipid perox-
idation) induced by H2O2 and H2O2/Fe2+ in human plasma.
The activity of compounds 1−11 was compared to that of the
well-known antioxidant curcumin. Compounds 1−11 and
curcumin were tested at doses ranging from 0.1 to 100 μM.
Compounds 10 and 11 at 10 μM reduced both H2O2- and
H2O2/Fe2+-induced lipid peroxidation by more than 30%,
hence were more active than curcumin, while compounds 1 and
3 reduced H2O2-induced lipid peroxidation by more than 30%
at 10 μM (Figure 2, Table 3). Compound 9 exhibited a
Table 3. Inhibitory Effects of Compounds 1−11 (10 μM; 30
min) and Curcumin (10 μM; 30 min) on Plasma Lipid
Peroxidation Induced by H2O2 or H2O2/Fe2+
inhibition of lipid
peroxidation induced by
H2O2 (%)
inhibition of lipid peroxidation
induced by H2O2/Fe2+ (%)
compound
1
32.2 7.7 (p < 0.05)
14.7 4.3 (p < 0.05)
33.3 9.9 (p < 0.05)
12.6 4.5 (p < 0.02)
24.4 6.7 (p < 0.05)
23.3 7.9 (p < 0.05)
14.4 6.1 (n.s.)
18.1 4.8 (p < 0.05)
n.s.
2
3
4
a
5
n.s.
6
n.s.
n.s.
7
n.s.
n.s.
8
5.7 3.2 (p > 0.05)
24.2 6.7 (p < 0.05)
44.4 8.4 (p < 0.05)
31.8 9.9 (p < 0.02)
21.1 7.9 (p < 0.05)
20.1 7.2 (p < 0.05)
25.5 9.9 (p < 0.05)
34.1 6.9 (p < 0.05)
39.7 8.2 (p < 0.02)
22.9 5.1 (p < 0.05)
9
10
11
curcumin
a
Not significant.
protective action against oxidative stress induced by H2O2 or
H2O2/Fe2+, similar to that shown by curcumin, while weaker
activity was displayed by compounds 2, 4, and 8.
The cytotoxic activity of compounds 1−11 was tested against
two cancer cell lines, namely, A549 (human lung adenocarci-
noma) and DeFew (human B lymphoma). In the range 10−
100 μM, the compounds did not show a significant reduction of
the cell number (data not shown), in agreement with the
absence of cytotoxicity previously reported for giffonins J−P.15
The antimicrobial assays of the methanol extract of leafy
covers of C. avellana and of compounds 1−11 were performed
against the Gram-positive strains Bacillus cereus and Staph-
ylococcus aureus and the Gram-negative strains Escherichia coli
and Pseudomonas aeruginosa. The results are shown in Tables 4
and 5. The most active compounds were 1 and 3. In particular,
carpinontriol B (1) proved effective at 10 μg/disk, with the
exception of S. aureus, which required 40 μg/disk to obtain an
inhibition halo (Tables 4 and 5). At 40 μg/disk both
compounds caused the formation of zones of inhibition
completely comparable to those obtained with tetracycline at
7 μg/disk used as positive control (Table 4). Giffonin T (2)
G
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
the cyclic diarylheptanoids carpinontriol B and giffonin U are
the most effective against the tested strains.
Table 5. Antimicrobial Activity of Compounds 4−10 at 40
a
μg/disk
EC
BC 4384 BC 4313
PA
SA
EXPERIMENTAL SECTION
e
b
b
b
b
b
b
b
b
b
e
■
4
10 (0)
0(0)
0(0)
0(0)
0(0)
5(0)
0(0)
0(0)
5 (0)
4.67
9.33
10 (0)
e
f
f
e
e
f
General Experimental Procedures. Optical rotations were
obtained on an Autopol IV (Rudolph Research Analytical) polar-
imeter. IR data were measured on a Bruker IFS-48 spectrometer. NMR
spectra were recorded in methanol-d4 (99.95%, Sigma-Aldrich) on a
Bruker DRX-600 spectrometer (Bruker BioSpin GmBH, Rheinstetten,
Germany) equipped with a Bruker 5 mm TCI CryoProbe at 300 K.
Data processing was carried out with Topspin 3.2 software. The
5
10.67 (0.57)
10 (0)
10.67 (0.57)
10.67 (0.57)
11.33 (0.57)
10.33 (0.57)
e
b
b
6
10(0)
5.67
0(0)
b
b
7
11.33 (0.57)
0(0)
6.67
4(0)
0(0)
e
b
b
b
e
8
10(0)
0(0)
d
b
c
9
8.33 (0.57)
0(0)
6.67 (0.57)
f
b
b
c
10
9.67 (0.57)
0(0)
5 (0)
6.33 (0.57)
a
1
ROESY spectra were acquired with tmix = 400 ms. H NMR spectra
Data are expressed in mm. Results are shown as the mean SD (n =
were acquired in DMSO-d6 (Sigma-Aldrich, 99.95%) on a Bruker 600
MHz instrument at different temperatures (298−373 K). HRESIMS
data were acquired on an LTQ Orbitrap XL mass spectrometer
(Thermo Fisher Scientific, San Jose, CA, USA) operating in negative
ion mode. GC analysis was performed on a Thermo Finnigan Trace
GC apparatus. Size exclusion chromatography was performed over
Sephadex LH-20 (Pharmacia). HPLC separations were carried out on
a Waters 590 HPLC system.
3). Means followed by different letters in each column differ
significantly to Dunnett’s multiple comparisons test, at the significance
level of p < 0.05. EC: Escherichia coli; BC 4384: Bacillus cereus DSM
4384; BC 4313: Bacillus cereus 4313; PA: Pseudomonas aeruginosa; SA:
Staphyloccus aureus. Tetracycline (7 μg/disk) and DMSO were used as
b
c
d
positive and negative control, respectively. p < 0.0001. p < 0.001. p
e
f
< 0.001. p < 0.005. p < 0.05.
Plant Material. The green leafy covers of C. avellana L., cultivar
“Tonda di Giffoni”, were collected at Giffoni, Salerno, Italy, in August
2014 and identified by V. De Feo (Department of Pharmacy,
University of Salerno, Italy). A voucher specimen (No. 138) has been
deposited in this Department.
showed weaker antimicrobial activity and only against B. cereus
(4313), P. aeruginosa, and S. aureus, at 40 μg/disk (Table 5).
The different behavior of compounds 1−3 was confirmed by
evaluation of the minimal inhibitory concentration (MIC)
(Table 6), which evidenced a stronger activity of carpinontriol
Extraction and Isolation. The green leafy covers of C. avellana L.,
cultivar “Tonda di Giffoni” (535 g), were dried and extracted with n-
hexane (2 × 4.4 L, 3 days each), CHCl3 (2 × 4.6 L, 3 days each), and
MeOH (3 × 4.6 L, 3 days each), to obtain 23.8 g of crude MeOH
extract. The dried MeOH extract (3 g) was fractionated on a Sephadex
LH-20 (Pharmacia) column (100 × 5 cm), using MeOH as mobile
phase, to afford 65 fractions (8 mL), monitored by TLC. Some of
these fractions were further chromatographed by semipreparative
HPLC using MeOH−H2O (7:13) as mobile phase (flow rate 2.5 mL/
min). Fractions 15 and 16 (37.0 mg) were purified by HPLC using
MeOH−H2O (9:11) to yield giffonin I (4) (32.5 mg, tR = 16.2 min).
Fractions 19 and 20 (23.9 mg) were purified by HPLC using MeOH−
H2O (7:13) to yield citric acid (5) (8.1 mg, tR = 2.8 min), 1-
methylcitrate (6) (2.5 mg, tR = 4.0 min), and trimethylcitrate (7) (1.6
mg, tR = 4.2 min). Fractions 41−44 (77.8 mg) were purified by HPLC
using MeOH−H2O (2:3) to obtain compounds 2 (1.3 mg, tR = 6.4
min), 3 (5.1 mg, tR = 9.8 min), carpinontriol B (1) (3.2 mg, tR = 20.8
min), and kaempferol 3-O-L-rhamnopyranoside (8) (1.8 mg, tR = 26.7
min). Fractions 47 and 48 (17.2 mg) were purified by HPLC using
MeOH−H2O (2:3) to obtain 3,5-dicaffeoylquinic acid (9) (1.6 mg, tR
= 8.0 min) and myricetin 3-O-L-rhamnopyranoside (10) (2.1 mg, tR =
24.2 min). Fraction 55 (6.0 mg) corresponded to kaempferol 3-O-(4″-
trans-p-coumaroyl)-L-rhamnopyranoside (11) (17.1 mg).
Table 6. Minimal Inhibitory Concentration (MIC, μg/disk)
of Compounds 1−3 and of MeOH Extract of Leafy Covers of
C. avellana Cultivar “Tonda di Giffoni”
microorganism
1
2
3
MeOH extract
Bacillus cereus 4313
Bacillus cereus 4384
Escherichia coli
4 μg
4 μg
30 μg
50 μg
50 μg
30 μg
30 μg
5 μg
5 μg
50 μg
50 μg
100 μg
30 μg
30 μg
10 μg
10 μg
30 μg
10 μg
10 μg
30 μg
Pseudomonas aeruginosa
Staphylococcus aureus
B (1) and giffonin U (3) on almost all strains tested (except
against S. aureus). Giffonin I (4) was effective against almost all
the tested strains at 40 μg/disk. The antimicrobial activity
shown by the kaempferol derivatives 8 and 11 as well as by
myricetin (10) was in agreement with those reported by
Cushnie and Lamb.32 The antimicrobial activity of citric acid
(5), 1-methylcitrate (6), and trimethylcitrate (7) confirmed the
capability of this class of compounds to inhibit the growth of
microorganisms such as E. coli.33
This is the first report of the chemical composition of the
green leafy cover of C. avellana. This study led to the isolation
of 11 compounds, including two new cyclic diarylheptanoids.
Diarylheptanoid derivatives occur frequently in plants belong-
ing to the Betulaceae family, while the cyclic diarylheptanoids
are reported in only a few species such as Carpinus cordata,11
Alnus sieboldiana,34,35 and Ostryopsis nobilis.36 In the Corylus
genus cyclic diarylheptanoids are so far reported only in
C. sieboldiana37 and C. avellana.6,15 For the first time, the
determination of the absolute configuration of highly
substituted cyclic diarylheptanoid has been defined by a
combined QM/NMR approach, followed by comparison of
the experimental ECD curves with the TDDFT-predicted data.
The evaluation of the biological activity of the isolated
compounds confirmed the antioxidant activity of cyclic
diarylheptanoids and provided insight into the antimicrobial
activity of the MeOH extract of the leafy covers, showing that
Giffonin T (2): amorphous, white solid; [α]25 −6 (c 0.1 MeOH);
D
IR (KBr) max 3425, 2930, 1713, 1665 cm−1; 1H and 13C NMR
(methanol-d4, 600 MHz) data, Table 1; HRESIMS [M − H]− m/z
505.1717 (calcd for C25H29O11, 505.1710).
Giffonin U (3): amorphous, white solid; [α]25D −21 (c 0.1 MeOH);
IR (KBr)
3425, 2930, 1730, 1665 cm−1; 1H and 13C NMR
max
(methanol-d4, 600 MHz) data, Table 1; HRESIMS [M − H]− m/z
359.1136 (calcd for C19H19O7, 359.1131).
Compound 4′: amorphous, white solid after hydrolysis of 4 with 1
N HCl; [α]25 0 (c 0.1 MeOH).
D
Determination of the Sugar Configuration. The configuration
of the sugar unit of compound 2 was established after hydrolysis of 2
with 1 N HCl, trimethylsilation, and determination of the retention
times by GC operating under the reported experimental conditions.38
The peak of the hydrolysate of 2 was detected at 14.75 min (D-
glucose). The retention time for an authentic sample after being
treated in the same manner with 1-(trimethylsilyl)imidazole in
pyridine was detected at 14.71 min (D-glucose).
Computational Details. Maestro 10.239 was used for generating
the starting 3D chemical structures of all possible diastereoisomers of
compounds 1, 3, and 4. Calculations were not performed for
H
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
compound 2, since experimental data revealed that it differs from 1
only by the presence of a β-D-glucopyranosyl unit. Optimization of the
3D structures was performed with MacroModel 10.239 using the OPLS
force field40 and the Polak−Ribier conjugate gradient algorithm
(PRCG, maximum derivative less than 0.001 kcal/mol).
In particular, for compound 1, which has three stereogenic centers
and an axis of chirality, eight possible diasteroisomers were considered:
1a (aR*,8S*,9R*,10R*), 1b (aS*,8S*,9R*,10R*), 1c
( a R * , 8 S * , 9 R * , 1 0 S * ) , 1 d ( a S * , 8 S * , 9 R * , 1 0 S * ) , 1 e
(aR*,8S*,9S*,10R*), 1f (aS*,8S*,9S*,10R*), 1g (aR*,8S*,9S*,10S*),
1h (aS*,8S*,9S*,10S*) (Chart S1, Supporting Information).
For compound 3, possessing four stereogenic centers and an axis of
chirality, 16 diastereoisomers were considered: 3a
(aR*,8S*,9R*,10R*,11R*), 3b (aS*,8S*,9R*,10R*,11R*), 3c
(aR*,8S*,9R*,10R*,11S*), 3d (aS*,8S*,9R*,10R*,11S*), 3e
(aR*,8S*,9R*,10S*,11R*), 3f (aS*,8S*,9R*,10S*,11R*), 3g
(aR*,8S*,9R*,10S*,11S*), 3h (aS*,8S*,9R*,10S*,11S*), 3i
(aR*,8S*,9S*,10R*,11R*), 3j (aS*,8S*,9S*,10R*,11R*), 3k
(aR*,8S*,9S*,10R*,11S*), 3l (aS*,8S*,9S*,10R*,11S*), 3m
(aR*,8S*,9S*,10S*,11R*), 3n (aS*,8S*,9S*,10S*,11R*), 3o
(aR*,8S*,9S*,10S*,11S*), 3p (aS*,8S*,9S*,10S*,11S*) (Chart S2,
Supporting Information).
(8S*,9R*,10R*)-1 and the transition state confirmed the hindered
rotation about the biphenyl axis (Results and Discussion). The
different atropisomers could be differently treated for the subsequent
calculations of the NMR parameters and for the computation of the
CD spectra.
Following the same procedure, the energy of the transition state
associated with the interconversion between (aR) and (aS)
atropisomers of 4′ (4′a and 4′b, respectively; Chart 3, Supporting
Information) was computed.
The computation of the 13C and 1H NMR chemical shifts was
performed on all the selected conformers for the different
diastereoisomers of compounds 1 and 3, using the MPW1PW91
functional, the 6-31+G(d,p) basis set, and MeOH IEFPCM. Final 13C
and 1H NMR spectra for each of the diastereoisomers were built
considering the influence of each conformer on the total Boltzmann
distribution taking into account the relative energies. Calibrations of
1
calculated 13C and H chemical shifts were performed following the
multistandard approach.24,25 In particular, sp2 13C and 1H NMR
chemical shifts were computed using benzene as reference
compound,24,25 while TMS was used for computing sp3 13C and H
1
chemical shift data.
A further set of data was produced using only TMS as reference
compound, and it was subsequently used for the computation of the
DP4+ probabilities.
Furthermore, for compound 4′, possessing an axis of chirality, two
atropisomeric forms, 4′a (aR) and 4′b (aS), are reported in Chart S3,
Supporting Information.
First, experimental and calculated 13C and 1H NMR chemical shifts
were compared computing the Δδ parameter (Tables S1−S8,
Supporting Information):
Starting from the obtained 3D structures, exhaustive conformational
searches at the empirical molecular mechanics level with the MCMM
method (50 000 steps) and LMCS method (50 000 steps) were
performed, in order to allow a full exploration of the conformational
space. Furthermore, molecular dynamics simulations were performed
at 450, 600, 700, and 750 K, with a time step of 2.0 fs, an equilibration
time of 0.1 ns, and a simulation time of 10 ns. A constant dielectric
term of methanol, mimicking the presence of the solvent, was used in
the calculations to reduce artifacts.
Δδ = |δexp − δ
|
calc
where δexp (ppm) and δcalc (ppm) are the 13C/1H experimental and
calculated chemical shifts, respectively.
The mean absolute errors for all the considered diastereoisomers
were computed using the following equation:
For each diastereoisomer, all the conformers obtained from the
conformational searches were minimized (PRCG, maximum derivative
less than 0.001 kcal/mol) and compared. The “Redundant Conformer
Elimination” module of Macromodel 10.239 was used to select
nonredundant conformers, excluding those differing by more than 21.0
kJ/mol (5.02 kcal/mol) from the most energetically favored
conformation and setting a 0.5 Å RMSD (root-mean-square deviation)
minimum cutoff for saving structures. For compounds 1, 3, and 4′,
MM conformational searches produced both sets of atropisomers,
which were manually separated after visual inspection once the
hindered rotation along the biaryl axis was assessed by means of
quantum mechanical calculations (vide infra). All the QM calculations
were performed using Gaussian 09 software.41
∑ (Δδ)
MAE =
n
defined as the summation (∑) of the n computed absolute error
values (Δδ), normalized to the number of chemical shifts considered
(n) (Table 2 and Tables S1−S10, Supporting Information).
Furthermore, DP4+ probabilities related to all the stereoisomers of
1 and 3 were computed considering both 13C and H NMR chemical
1
shifts and comparing them with the related experimental data. In
particular, since the available DP4+ Toolbox (Excel file) for the DP4+
computation allows the setting of sp3/sp2 atoms following the
“multistandard” approach, we used the chemical shift data set obtained
using TMS as reference compound (Table 2).
The conformers were optimized at the QM level using the
MPW1PW91 functional and the 6-31G(d) basis set.42 Experimental
solvent effects (MeOH) were reproduced using the integral equation
formalism version of the polarizable continuum model (IEFPCM).43
After this step at the QM level, the optimized geometries were visually
inspected in order to remove redundant conformers.
For compounds 1c and 3o, identified as the most probable
diastereoisomer of 1 and 3, respectively, Boltzmann-weighted
prediction of J values was performed for the most energetically
favored conformers [energies associated with the QM optimization
step, MPW1PW91/6-31G(d)], performing a two-step spin−spin
calculation (mixed keyword for Gaussian calculations) using the
MPW1PW91 functional and the 6-311+G(d,p) basis set.
Once the relative configurations of 1 and 3 were obtained, the
prediction of ECD spectra was performed using all the conformers
obtained from the DFT calculations and performing QM calculations
at the TDDFT (NStates = 40) MPW1PW91/6-31G(d,p) level, in
EtOH IEFPCM to reproduce the experimental solvent environment.
The final ECD spectra for both the enantiomers related to the
predicted stereoisomers of 1 and 3 (1c and 3o) were calculated
considering the influence of each conformer on the total Boltzmann
distribution taking into account the relative energies and were
graphically plotted using SpecDis software.44 In order to simulate
the experimental ECD curve, a Gaussian band-shape function was
applied with the exponential half-width (σ/γ) of 0.20 eV.
The perceived atropisomerism arising from the hindered rotation
about the biphenyl axis was evaluated by computing the rotational
energy barrier required for the interconversion between the
(aR*,8S*,9R*,10R*) and (aS*,8S*,9R*,10R*) atropisomers of com-
pound 1 (1a and 1b; respectively, Chart 1, Supporting Information),
assuming that this system could be considered representative of all
diastereoisomers. Specifically, the starting geometry model represent-
ing the transition state was built with the two phenyl moieties
occupying the same plane, which was subsequently optimized at the
QM level using the Berny algorithm, the MPW1PW91 functional, and
the 6-31G(d) basis set followed by vibrational frequency calculations
(TS, CalcAll, Freq keywords for Gaussian calculations). Analysis of the
vibrational frequencies showed that the optimized structure was
correctly associated with the transition state, since the two phenyl
moieties slightly move along the biaryl axis, producing the two
different atropisomeric forms at each oscillation. Comparison of the
energies between the lowest energy-associated conformer found for
Lipid Peroxidation Measurement. Stock solutions of the
compounds and plant extract were made in 50% DMSO. The final
concentration of DMSO in the samples was lower than 0.05%, and in
all experiments its effects were determined.
I
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Fresh human plasma was obtained from medication-free, regular
donors at the blood bank (Lodz, Poland). Samples of human plasma
were incubated for 30 min at 37 °C with compounds 1−11 and
curcumin (0.1−100 μM) alone and with 2 mM H2O2, and with
compounds 1−11 and curcumin at 10 μM plus 4.7 mM H2O2/3.8 mM
Fe2SO4/2.5 mM EDTA. Pure compounds were tested by using the
TBARS assay as previously reported.6
calculated NMR chemical shifts of all the possible
stereoisomers of compounds 1 and 3; ECD spectrum of
1
compound 4′; H NMR spectra of 1 acquired over a
range of temperatures; effects of compounds 1−11 and
curcumin on plasma lipid peroxidation induced by
H2O2/Fe2+ (PDF)
Cell Culture and Viability Assay. The A549 cell line was
maintained in DMEM medium supplemented with 10% fetal bovine
serum (Invitrogen), 1% penicillin/streptomycin, and 2 mM L-
glutamine, and the DeFew cell line was maintained in RPMI 1640
medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10%
fetal bovine serum, 1% penicillin/streptomycin, and 2 mM L-
glutamine. Both cell lines were cultured at 37 °C in a 5% CO2
humidified atmosphere.
AUTHOR INFORMATION
Corresponding Authors
*Tel: +39 089969741. Fax: +39 089969602. E-mail: bifulco@
unisa.it.
*Tel: +39 089969763. Fax: +39 089969602. E-mail: piacente@
unisa.it.
ORCID
■
Cell viability was determined using the MTT assay45 as previously
reported.46
Sonia Piacente: 0000-0002-4998-2311
Antimicrobial Activity. The antibacterial activity of the MeOH
extract of the leafy covers of C. avellana, cultivar “Tonda di Giffoni”,
and of compounds 1−11 was assayed by the inhibition halo test on
agar plates47 against Bacillus cereus (DSM 4313 and DSM 4384),
Staphylococcus aureus DSM 25923, Escherichia coli DSM 8579, and
Pseudomonas aeruginosa DSM 50071, provided by Deutsche Sammlung
von Mikroorganismen and Zellkulturen GmbH (DSMZ, Braunsch-
weig, Germany). Each strain was incubated at 37 °C for 18 h in TY
broth (Sigma-Aldrich, Milano, Italy). The microbial suspensions at 1 ×
108 colony forming units (cfu)/mL were uniformly spread on the solid
media plates (⦶ = 90 mm dishes). Sterile Whatman grade 1 paper
filter disks (⦶ = 5 mm), previously impregnated with samples (final
amount ranging from 10 to 40 μg/disk) and dried at room
temperature for 60 min, were individually placed on the inoculated
plates. Plates were then incubated at 37 °C for 24−48 h, depending on
the strain. The activity of compounds was evaluated by measuring the
diameter (in mm) of the inhibition zones around the disks. Sterile
DMSO was used as negative control. Tetracycline (7 μg/disk; Sigma-
Aldrich, Milano, Italy) was the reference sample. The samples were
tested in triplicate, and the results are expressed as the mean values
standard deviations.
Minimum Inhibitory Concentration. The MIC values for the
MeOH extract, carpinontriol B (1), and giffonins T and U (2 and 3)
were evaluated by the resazurin microtiter-plate assay modified from
Sarker and co-workers (2007),48 as previously reported.49 It is based
on the capability of microorganisms to lower the redox potential of the
medium in which they are located as a result of their growth and
metabolic activities. The addition of resazurin as a redox indicator
displays, through its color change (dark purple → colorless), the state
of oxidation or reduction, thus the growth of bacteria. Samples,
dissolved in DMSO, were pipetted in a multiwell plate with different
volumes of Muller-Hinton broth (Sigma-Aldrich, Milano, Italy). Two-
fold serial dilutions were performed such that each well had 50 μL of
the test material in serially descending concentrations. A 35 μL
amount of 3.3× strength isosensitized broth and 5 μL of resazurin
indicator solution were added to a final volume/well of 240 μL.
Finally, 10 μL of bacterial suspension was added to each well to
achieve a concentration of about 5 × 105 cfu/mL. Ciprofloxacin (1
mg/mL in DMSO, Sigma) and DMSO were used as positive and
negative controls, respectively. Plates were prepared in triplicate and
incubated at 37 °C for 24 h. The lowest concentration at which a color
change occurred indicated the MIC value.
Author Contributions
∥A. Cerulli and G. Lauro contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank “Consorzio di tutela Nocciola di Giffoni
I.G.P.” for the useful information on Corylus avellana, cultivar
“Tonda di Giffoni”. Part of this work (evaluation of the
antioxidant activity) was supported by grant 506/1136 from the
University of Lodz.
REFERENCES
■
(1) Oliveira, I.; Sousa, A.; Valentao, P.; Andrade, P. B.; Ferreira, I. C.
F. R.; Ferreres, F.; Bento, A.; Seabra, R.; Estevinho, L.; Pereira, J. A.
Food Chem. 2007, 105, 1018−1025.
(2) Shahidi, F.; Alasalvar, C.; Liyana-Pathirana, C. M. J. Agric. Food
Chem. 2007, 55, 1212−1220.
(3) Petriccione, M.; Ciarmiello, L. F.; Boccacci, P.; De Luca, A.;
Piccirillo, P. Sci. Hortic. (Amsterdam, Neth.) 2010, 124, 153−158.
(4) Riethmuller, E.; Alberti, A.; Toth, G.; Beni, S.; Ortolano, F.; Kery,
A. Phytochem. Anal. 2013, 24, 493−503.
(5) Piccinelli, A. L.; Pagano, I.; Esposito, T.; Mencherini, T.; Porta,
A.; Petrone, A. M.; Gazzerro, P.; Picerno, P.; Sansone, F.; Rastrelli, L.;
Aquino, R. P. J. Agric. Food Chem. 2016, 64, 585−595.
(6) Masullo, M.; Cerulli, A.; Olas, B.; Pizza, C.; Piacente, S. J. Nat.
Prod. 2015, 78, 17−25.
(7) Alasalvar, C.; Hoofman, A. M.; Shahidi, F. Nutraceutical Sci.
Technol. 2009, 9, 215−235.
(8) Iranshahi, M.; Chini, M. G.; Masullo, M.; Sahebkar, A.; Javidnia,
A.; ChitsazianYazdi, M.; Pergola, C.; Koeberle, A.; Werz, O.; Pizza, C.;
Terracciano, S.; Piacente, S.; Bifulco, G. J. Nat. Prod. 2015, 78, 2867−
2879.
(9) Masullo, M.; Mari, A.; Cerulli, A.; Bottone, A.; Kontek, B.; Olas,
B.; Pizza, C.; Piacente, S. Phytochemistry 2016, 130, 273−281.
(10) Alasalvar, C.; Karamac, M.; Amarowicz, R.; Shahidi, F. J. Agric.
Food Chem. 2006, 54, 4826−4832.
(11) Lee, J. S.; Kim, H. J.; Park, H.; Lee, Y. S. J. Nat. Prod. 2002, 65,
1367−1370.
(12) Lv, H.; She, G. Nat. Prod. Commun. 2010, 5, 1687−1708.
(13) Liu, F.; Zhang, Y.; Sun, Q.-Y.; Yang, F.-M.; Gu, W.; Yang, J.;
Niu, H.-M.; Wang, Y.-H.; Long, C.-L. Phytochemistry 2014, 103, 171−
177.
(14) Gulcemal, D.; Masullo, M.; Bedir, E.; Festa, M.; Karayildirim, T.;
Alankus-Caliskan, O.; Piacente, S. Planta Med. 2012, 78, 720−729.
(15) Masullo, M.; Cantone, V.; Cerulli, A.; Lauro, G.; Messano, F.;
Russo, G. L.; Pizza, C.; Bifulco, G.; Piacente, S. J. Nat. Prod. 2015, 78,
2975−2982.
(16) Bifulco, G.; Dambruoso, P.; Gomez-Paloma, L.; Riccio, R. Chem.
Rev. 2007, 107, 3744−3779.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.6b00703.
1H, HSQC, HMBC, COSY, and ROESY spectra of
compounds 1−3; chemical structures and tables of
13C/1H NMR data for compounds 1 and 3 and
J
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(17) Di Micco, S.; Chini, M. G.; Riccio, R.; Bifulco, G. Eur. J. Org.
Chem. 2010, 2010, 1411−1434.
(44) Bruhn, T.; Schaumloffel, A.; Hemberger, Y.; Bringmann, G.
Chirality 2013, 25, 243−249.
(45) Mosmann, T. J. Immunol. Methods 1983, 65, 55−63.
(46) Horo, I.; Masullo, M.; Falco, A.; Senol, S. G.; Piacente, S.;
Alankus-Caliskan, O. Phytochem. Lett. 2015, 14, 39−44.
(47) Fratianni, F.; Sada, A.; Cipriano, L.; Masucci, A.; Nazzaro, F.
Open Food Sci. J. 2007, 1, 10−16.
(48) Sarker, S. D.; Nahar, L.; Kumarasamy, Y. Methods 2007, 42,
321−324.
(49) Aliberti, L.; Caputo, L.; De Feo, V.; De Martino, L.; Nazzaro, F.;
Souza, L. F. Molecules 2016, 21, 124410.3390/molecules21091244.
(18) Barone, G.; Duca, D.; Silvestri, A.; Gomez-Paloma, L.; Riccio,
R.; Bifulco, G. Chem. - Eur. J. 2002, 8, 3240−3245.
(19) Barone, G.; Gomez-Paloma, L.; Duca, D.; Silvestri, A.; Riccio,
R.; Bifulco, G. Chem. - Eur. J. 2002, 8, 3233−3239.
(20) Bassarello, C.; Bifulco, G.; Montoro, P.; Skhirtladze, A.;
Kemertelidze, E.; Pizza, C.; Piacente, S. Tetrahedron 2007, 63, 148−
154.
(21) Plaza, A.; Piacente, S.; Perrone, A.; Hamed, A.; Pizza, C.;
Bifulco, G. Tetrahedron 2004, 60, 12201−12209.
(22) Bringmann, G.; Mortimer, A. J. P.; Keller, P. A.; Gresser, M. J.;
Garner, J.; Breuning, M. Angew. Chem., Int. Ed. 2005, 44, 5384−5427.
(23) Smyth, J. E.; Butler, N. M.; Keller, P. A. Nat. Prod. Rep. 2015, 32,
1562−1583.
(24) Sarotti, A. M.; Pellegrinet, S. C. J. Org. Chem. 2009, 74, 7254−
7260.
(25) Sarotti, A. M.; Pellegrinet, S. C. J. Org. Chem. 2012, 77, 6059−
6065.
(26) Grimblat, N.; Zanardi, M. M.; Sarotti, A. M. J. Org. Chem. 2015,
80, 12526−12534.
(27) Nadmid, S.; Plaza, A.; Lauro, G.; Garcia, R.; Bifulco, G.; Mueller,
R. Org. Lett. 2014, 16, 4130−4133.
(28) Masullo, M.; Bassarello, C.; Bifulco, G.; Piacente, S. Tetrahedron
2010, 66, 139−145.
(29) Gong, J.; Sun, P.; Jiang, N.; Riccio, R.; Lauro, G.; Bifulco, G.; Li,
T. J.; Gerwick, W. H.; Zhang, W. Org. Lett. 2014, 16, 2224−2227.
(30) Talapatra, B.; Bhaumik, A.; Talapatra, S. K. Indian J. Chem., Sect.
B 1993, 32B, 1292−1294.
(31) Dall’Acqua, S.; Viola, G.; Piacente, S.; Cappelletti, E. M.;
Innocenti, G. J. Nat. Prod. 2004, 67, 1588−1590.
(32) Cushnie, T. P. T.; Lamb, A. J. Int. J. Antimicrob. Agents 2005, 26,
343−356.
(33) Eswaranandam, S.; Hettiarachchy, N. S.; Johnson, M. G. J. Food
Sci. 2004, 69, M79−M84.
(34) Chiba, K.; Ichizawa, H.; Kawai, S.; Nishida, T. J. Wood Chem.
Technol. 2013, 33, 44−51.
(35) Ibrahim, S. R. M.; Mohamed, G. A.; Khedr, A. I. M.; Aljaeid, B.
M. Rec. Nat. Prod. 2016, 10, 362−368.
(36) Zhang, Y. X.; Xia, B.; Zhou, Y.; Ding, L. S.; Peng, S. L. Chin.
Chem. Lett. 2013, 24, 512−514.
(37) Watanabe, N.; Sasaya, T.; Sano, Y. Mokuzai Gakkaishi 1994, 40,
1219−1225.
(38) Pereira do Amaral, F.; Napolitano, A.; Masullo, M.; Campaner
dos Santos, L.; Festa, M.; Vilegas, W.; Pizza, C.; Piacente, S. J. Nat.
Prod. 2012, 75, 547−556.
(39) Maestro 10.2; Schrodinger LLC: New York, NY, 2013.
̈
(40) Jorgensen, W. L.; Tirado-Rives, J. J. Am. Chem. Soc. 1988, 110,
1657−1666.
(41) Frisch, M. J. T., G, W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.
(42) Cimino, P.; Gomez-Paloma, L.; Duca, D.; Riccio, R.; Bifulco, G.
Magn. Reson. Chem. 2004, 42, S26−S33.
(43) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999−3093.
K
DOI: 10.1021/acs.jnatprod.6b00703
J. Nat. Prod. XXXX, XXX, XXX−XXX