C-glucosidic ellagitannins from lythri herba (European Pharmacopoeia): Chromatographic profile and structure determination

Article abstract of DOI:10.1002/pca.2415

Introduction Lythri herba, a pharmacopoeial plant material (European Pharmacopoea), is obtained from flowering parts of purple loosestrife (Lythrum salicaria L.). Although extracts from this plant material have been proven to possess some interesting biological activities and its pharmacopoeial standardisation is based on total tannin content determination, the phytochemical characterisation of this main group of compounds has not yet been fully conducted. Objective To isolate ellagitannins from Lythri herba, determine their structures and develop chromatographic methods for their qualitative analysis. Results Five C-glucosidic ellagitannins - monomeric- vescalagin and castalagin together with new dimeric structures - salicarinins A-C, composed of vescalagin and stachyurin, vescalagin and casuarinin, castalagin and casuarinin units connected via formation of valoneoyl group, were isolated using column chromatography and preparative HPLC. Structures were determined according to ¹H and ¹³C-NMR (one- and two-dimensional), electrospray ionisation-time of flight (ESI-TOF), electrospray ionisation-ion trap (ESI-MSⁿ) and circular dichroism (CD) spectra, together with acidic hydrolysis products analysis. HPTLC on RP-18 modified plates and HPLC-DAD-MSⁿ on RP-18 column methods were developed for separation of the five main ellagitannins. Copyright 2012 John Wiley & Sons, Ltd. Composition of ellagitannins in Lythri herba (European Pharmacopoeia) was examined. Five C-glucosidic ellagitannins were isolated: monomeric vescalagin and castalagin together with not previously reported dimeric salicarinins A-C composed of vescalagin and stachyurin, vescalagin and casuarinin, castalagin and casuarinin units connected via formation of the valoneoyl group. Structures were determined according to ¹H- and ¹³C-NMR (one- and two-dimensional), ESI-TOF, MSⁿ and CD spectra together with acidic hydrolysis products analysis. HPLC-DAD-MSⁿ and HPTLC methods were developed for qualitative determination of ellagitannins in the examined plant extract. Copyright

Full text of DOI:10.1002/pca.2415

Research Article
Received: 12 September 2012,
Revised: 29 October 2012,
Accepted: 31 October 2012
Published online in Wiley Online Library: 28 December 2012
(wileyonlinelibrary.com) DOI 10.1002/pca.2415
C-glucosidic Ellagitannins from Lythri herba
(
European Pharmacopoeia): Chromatographic
Profile and Structure Determination
Jakub P. Piwowarski* and Anna K. Kiss
ABSTRACT:
Introduction – Lythri herba, a pharmacopoeial plant material (European Pharmacopoea), is obtained from flowering parts of
purple loosestrife (Lythrum salicaria L.). Although extracts from this plant material have been proven to possess some
interesting biological activities and its pharmacopoeial standardisation is based on total tannin content determination, the
phytochemical characterisation of this main group of compounds has not yet been fully conducted.
Objective – To isolate ellagitannins from Lythri herba, determine their structures and develop chromatographic methods for
their qualitative analysis.
Results – Five C-glucosidic ellagitannins – monomeric- vescalagin and castalagin together with new dimeric structures –
salicarinins A–C, composed of vescalagin and stachyurin, vescalagin and casuarinin, castalagin and casuarinin units
connected via formation of valoneoyl group, were isolated using column chromatography and preparative HPLC. Structures
1
13
were determined according to H and C-NMR (one- and two-dimensional), electrospray ionisation–time of flight (ESI–TOF),
n
electrospray ionisation-ion trap (ESI-MS ) and circular dichroism (CD) spectra, together with acidic hydrolysis products
n
analysis. HPTLC on RP-18 modified plates and HPLC–DAD–MS on RP-18 column methods were developed for separation
of the five main ellagitannins. Copyright © 2012 John Wiley & Sons, Ltd.
Keywords: HPLC–MS; TLC; ellagitannins; Lythri herba; Lythrum salicaria
and castalagin were detected together with other polyphenols
such as C-glucosidic flavonoids and phenolic acids. Ma et al.
Introduction
Purple loosestrife (Lythrum salicaria L.) is a plant belonging to
(1996) isolated dimeric ellagitannins from Lythrum salicaria, but
the Lythraceae family. Extracts from its herb have been used in
traditional medicine since ancient times as a strong astringent
and haemostatic agent in cases of diarrhoea, haemorrhaging
and haemorrhoidal disease. Lythri herba was introduced to the
European Pharmacopoeia in 2001. Studies conducted on cell,
cell-free and animal models revealed strong anti-oxidative,
anti-inflammatory, anti-nociceptive and haemostatic activity
the article contains suggested structures only, without any spectral
or chemical data provided.
Tannins are divided into two large groups: condensed and
hydrolysable tannins. Among the latter, gallotannins and
ellagitannins are distinguished. C-glucosidic ellagitannins are
a specific subclass characterised by the C–C linkage between
the C-1 of the open chain of D-glucose and a phenolic carbon.
Chromatographic analysis of ellagitannins raises many difficulties
due to their high molecular weight, high hydrophylicity and low
resistance to different solvents and high temperature. Many
HPLC–MS methods have been developed for the analysis of ellagi-
tannins (Quideau et al., 2005; Hager et al., 2008; Sanz et al., 2010;
Romani et al., 2012), but only a few TLC methods have been intro-
duced (Fecka and Cisowski, 2002; Bazylko et al., 2007; Fecka, 2009).
Furthermore, in the case of C-glucosidic ellagitannins no planar
chromatographic methods have been developed.
(Çoban et al., 2003; Tunalier et al., 2007; Pawlaczyk et al., 2011;
Piwowarski et al., 2011). Anti-microbial and fungistatic in vitro
activities were also established (Moskalenko, 1986; Rauha et al.,
2000; Saltan Çitoglu and Altanlar, 2003; Becker et al., 2005). Some
studies in the 1980s on animal models revealed hypoglycemic
effects (Cadavid Torres and Calleja Suarez, 1980; Lamela et al.,
1
986).
According to its monograph in European Pharmacopoeia
quantitative standardisation of Lythri herba is based on total
tannin content determination, which should be at least 5% per
dried drug. Despite ellagitannins being the main group of active
compounds present in extracts from Lythri herba, there had
been no studies conducted regarding their complex characteristic.
The majority of phytochemical examinations of Lythri herba had
been concerned on the quantitative analysis of the total amount
of tannins without taking into consideration the distinguishment
in their structures (Nagy and Rodinová, 2006; Humadi and Istudor,
Recent advances in the bioactivity research of ellagitannins,
underlining specific, structure-dependent modes of action (Coca
et al., 2009; Quideau et al., 2011), emphasise the necessity of
*
Correspondence to: J. P. Piwowarski, Department of Pharmacognosy and
Molecular Basis of Phytotherapy, Medical University of Warsaw, Faculty of Phar-
macy, Banacha St. 1, 02-097 Warsaw. E-mail: jakub.piwowarski@wum.edu.pl
2
009; Møller et al., 2009; Bencsik et al., 2011). Rauha et al. (2001)
conducted a screening of polyphenols from Lythri herba using
the HPLC–MS method and monomeric ellagitannins: vescalagin
Department of Pharmacognosy and Molecular Basis of Phytotherapy, Medical
University of Warsaw, Faculty of Pharmacy, Banacha St. 1, 02-097 Warsaw
3
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.

C-Glucosidic Ellagitannins from Lythrum Salicaria
their complex phytochemical examination in ellagitannin-rich
pharmacopoeial plant material such as Lythri herba.
The aim of this study was to isolate ellagitannins from Lythri
herba, determine their structures and develop chromatographic
methods for their qualitative determination.
NMR analysis
1
13
ꢀ
H- and C-NMR spectra were recorded at 25 C on a Bruker Avance II NMR
1
13
instrument (400 MHz for H- and 100.61 MHz for C-NMR). All NMR experi-
ments were conducted using acetone-d . Chemical shifts are given in d
(ppm) values relative to those of the solvent signal – acetone-d d 2.05;
6
6
H
C
d 29.84 and coupling constants (J) are reported in Hz. Two-dimensional
1
1
NMR experiments, H– H correlation spectroscopy (COSY), heteronuclear
single quantum correlation (HSQC) and heteronuclear multiple bond
coherence (HMBC) were conducted using conventional pulse sequence.
Experimental
Solvents and reagents
Chromatographic grade acetonitrile was purchased from Merck (Darmstadt,
Germany). Water was purified with the Millipore Simplicity System (Bedford,
Massachusetts, USA). All other solvents (analytical grade) used for TLC
Circular dichroism analysis
Circular dichroism (CD) analyses of isolated ellagitannins were performed
on a Jasco J-715 (Easton, Maryland, USA) spectrometer using methanol
as the solvent.
6
analysis were purchased from POCh (Gliwice, Poland). Acetone-d was
purchased from Sigma Aldrich (St. Louis, California, USA). Standard
substances vitexin, orientin, isoorientin and ellagic acid were purchased
from Carl Roth (Karlsruhe, Germany). Valoneic acid dilactone was isolated
by Kiss et al. (2012).
Hydrolysis
A solution of each compound (2 mg) in 1 M hydrochloric acid (2.0 mL) was
heated in a boiling water bath for 6 h. The reaction mixture was evapo-
rated under reduced pressure and dissolved in 0.5 mL of methanol. The
Plant material
n
HPLC–DAD–MS analysis of hydrolysis products was performed using
The Lythrum salicaria L. flowering herb was collected from its natural habi-
tat in Gietrzwałd in the Masuria region (Poland) in July 2010. Identity was
confirmed anatomically and morphologically based on its pharmacopoeial
monograph in the Department of Pharmacognosy and Molecular Basis of
Phytotherapy, Medical University of Warsaw, where voucher specimen
HE0806 is deposited.
HPLC conditions developed for ellagitannins analysis.
TLC analysis
Stationary phases: HPTLC silica gel 60 RP-18 WF254s (10 ꢁ 10 cm, 250 mm)
(
Merck, Darmstadt, Germany); RP-8 WF254s (5 ꢁ 10 cm, 250 mm);
HPTLC RP ꢂ 2 WF254s (5ꢁ 10 cm, 250 mm); Diphenyl-F reversed phase
5 ꢁ 10 cm, 250 mm) (Whatman, Clifton, New Jersey, USA); HPTLC NH
(10 ꢁ 10 cm, 200 mm) (Merck, Germany); HPTLC CN F254s (10ꢁ 10 cm, 200 mm)
Merck, Germany). Mobile phase: different proportions of tetrahydrofuran:water:
(
2 254
F s
Extraction and isolation
(
Plant material was shade dried and cut. A portion of 200g was divided into
two parts and each was extracted three times for 30 min with 1000mL of
methanol/acetone/acetonitrile, acidified with 0–5% of formic acid (v/v).
A 10 mg amount of extract and 1 mg of each ellagitannin were
dissolved in 1 mL of acetone:water (8:2, v/v), filtered through a 0.45 mm
syringe filter and applied to a plate (40 mL, 10 mm) using Linomat 5
ꢀ
water at a temperature of 40 C using an ultrasonic bath. Extracts were
filtered, combined and freeze dried. The amount of crude dry extract
obtained was 43.0g. Dry aqueous extract was dissolved in 500 mL of water
and subsequently purified with ethyl acetate (4ꢁ 500 mL) and n-butanol
(
Camag, Basel, Switzerland). Plates were developed in an ADC2
ꢀ
Automatic Developing Chamber (Camag, Switzerland) at 22 C with
(
4ꢁ 500mL). Water residue was freeze-dried (28.1g) and subjected to silica
humidity control (49.5 ꢃ 0.5% RH), with 10 min of tank saturation and
gel 60 RP-18 (40–63 mm) (Merck, Darmstadt, Germany) column and subse-
1
0 min of plate pre-conditioning. The migration path length was 85mm.
quently eluted with 500mL of methanol:water (0:1, 2:8, 1:0, v/v). Methanol
Plates were photographed in UV light at 254 nm using a TLC Visualizer
Camag, Switzerland), then derivatised using a Chromatogram Immersion
ꢀ
was immediately evaporated under reduced pressure at 40 C and eluates
(
were freeze dried. Fraction 2 (3.0g) was subjected to a Toyopearl HW-40F
Device III (Camag, Switzerland) with 3% iron (III) chloride solution in
methanol (w/v) and photographed in white light. The number of theoretical
(Tosoh, Tokyo, Japan) column and eluted with acetone:water (2:8, v/v),
which was increased to 3:7. Fractions containing target compounds were
plates was determined based on the United States Pharmacopoeia formula:
ꢀ
combined; acetone was evaporated under reduced pressure at 40 C and
2
N= 16(t
R
/W) . Resolution was determined according to Purnell’s equation:
freeze-dried. Dried fractions were purified using preparative HPLC
apparatus – Shimadzu LC-10AT equipment with a diode array detector
2 2
Rs = [(√N)/4)][(a ꢂ 1)/a][k /(1 + k )].
(
1
SPD-M10A). The column was a Nucleosil 7 C18 (250mmꢁ ½", particles size
n
0 mm) (Machery Nagel, Duren, Germany). Eluent A was 0.2% formic acid
HPLC–DAD–MS analysis
and B was 0.2% formic acid:acetonitrile (20:80, v/v). The gradient solvent
system used was: 0–30% B (60 min) at a flow rate of 4mL/min. UV-Vis
spectra were recorded in the range of 190–450nm and chromatograms
were acquired at 254 nm. Isolated pure compounds were: 1, 54.8 mg; 2,
Optimisation of HPLC separation was carried out using an UHPLC-3000
RS system (Dionex, Idstein, Germany) with a diode array detector (DAD)
coupled with an AmaZon SL ion trap MS with an ESI source (Bruker Dal-
tonics, Bremen, Germany). The LC eluate was introduced directly into the
ESI interface without splitting and compounds were analysed in negative
57.7 mg;, 3, 117.9 mg; 4, 95.0 mg; and 5, 62.6 mg.
and positive ion mode with the following settings: nebuliser pressure,
ꢀ
4
0 psi; drying gas flow rate, 9.0 L/min; nitrogen gas temperature, 300 C;
ESI–TOF–MS analysis
capillary voltage, 4.5 kV. The mass scan ranged from 100 to 2200 m/z.
The UV spectra were recorded in the range of 200–400 nm, and
chromatograms were acquired at 254, 280 and 350 nm. Analyses were
performed on a Zorbax SB C18-column (150 ꢁ 2.1 mm, 1.9 mm) (Agilent,
Santa Clara, California, USA) and Hypersil GOLD C18 (100 ꢁ 2.1 mm,
1.9 mm) (Thermo Scientific, Waltham, Miami, USA) with the column
Accurate mass measurements were conducted using a micrOTOF-Q
(Bruker Daltonics, Bremen, Germany) spectrometer equipped with an
electrospray ionisation (ESI) source by direct injection of aqueous
solutions of isolated compounds. The instrument parameters were set
as follows: polarisation mode, negative; scan range, 50–3000 m/z;
ꢀ
ꢀ
capillary voltage, +3.2 kV; drying gas temperature, 180 C; drying gas
temperature set to 25 C. The flow rate was 0.2 mL/min. The selection of
flow, 8.0 L/min; nebuliser gas pressure, 0.8 bar. The accurate mass data
of molecular ions were processed using Data Analysis 4.0 software
appropriate separation conditions was based on the modification of the
multistep gradient system consisting of mobile phase (A) 0.5% formic acid
in water (v/v) and (B) 0.1% formic acid in acetonitrile (v/v). Identification of
(Bruker Daltonics, Bremen, Germany), which provided possible elemental
n
formulae using the Smart Formula Editor.
compounds was based on comparison of retention times, UV and MS
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/pca

J. P. Piwowarski and A. K. Kiss
spectra with previously isolated substances. The separation parameters
were determined according to formulae presented in the TLC section.
The tailing factor (T ) was determined at 10% of peak height (W0.1).
f
determined by comparison with spectroscopic data presented in
the literature (Tanaka et al., 1997). Because assignment of glucose
carbons in dimeric ellagitannins can be based on the corresponding
carbon peaks in monomers (Yoshida et al., 1984), the isolated
structures of dimeric ellagitannins were deducted based on glu-
cose carbon signals of isolated monomers supported by literature
data (Okuda et al., 1982a; Hatano et al., 1988; Nonaka et al., 1990).
The structure of each monomeric part was confirmed by compar-
ison of glucose proton signals together with carbonyl and
aromatic carbons shifts.
Results and discussion
Structure determination of isolated compounds
In the present study five main ellagitannins were isolated from
Lythri herba aqueous extract (Figs 1 and 2). The molecular
formulae were determined by ESI–TOF–MS analysis. Full
assignment of sugar and aromatic protons and carbons was
achieved based on one- and two-dimensional NMR experiments
Further structure confirmation of isolated compounds was
n
based on ESI–MS analysis (Table 3). The presence of valoneoyl
and hexahydroxydiphenoyl (HHDP) moieties was determined by
(Tables 1 and 2). The structures of monomeric ellagitannins were
n
HPLC–DAD–MS analysis of acidic hydrolysis products, among
which valoneic acid dilactone and ellagic acid were detected
n
based on retention times, UV and MS spectra comparison with
standard substances (Table 4 and Fig. 3). The atropoisomerism of
valoneoyl and HHDP groups was determined by circular dichroism
spectra by diagnostic positive Cotton effect at around 235 nm
indicating an S-configuration (Okuda et al., 1982b).
Vescalagin (1) was isolated as an off-white amorphous powder.
The molecular formula was determined to be C41
H
26
O
26 from the
ꢂ
ESI–TOF–MS (933.0649 [Mꢂ H] ). The absence of a hemiacetal
carbon signal of sugar moiety in the region of d 90–100 and the
1
1
1
H-NMR spectrum assigned by H– H COSY exhibiting one set of
sugar protons suggested the C-glucosidic nature of this ellagitannin
Okuda et al., 1982a). The chemical shift and coupling constant of
the anomeric proton signal [d 4.91 (br s, 1H), H1] indicated b ori-
(
H
entation of the hydroxyl group at glucose C1 (Nonaka et al., 1990).
1
H-NMR exhibited one set of three aromatic proton signals due to
the presence of HHDP and a flavogallonyl unit. The full assignment
of sugar and aromatic protons and carbons (Tables 1 and 2) was
consistent with those reported in the literature for vescalagin
n
(
Tanaka et al., 1997). The ESI–MS analysis (Table 3) revealed
signals in both negative and positive ion mode. In negative mode
two peaks were present due to the formation of single and double
charged ions at m/z 933 and 466 respectively. Among fragmenta-
tion products, ions derived from the neutral loss of a HHDP unit
(302 amu) and/or CO with H O (62 amu) or H O itself (18 amu)
Figure 1. Structure of isolated monomeric ellagitannins: vescalagin
(
(
1) and castalagin (2) with selected heteronuclear multiple bond coherence
HMBC) correlations and fragmentation pathways.
2
2
2
Figure 2. Structure of isolated dimeric ellagitannins: salicarinins A–C (3, 4, 5) with selected heteronuclear multiple bond coherence (HMBC) correlations
and fragmentation pathways.
wileyonlinelibrary.com/journal/pca
Copyright © 2012 John Wiley & Sons, Ltd.
Phytochem. Anal. 2013, 24, 336–348

C-Glucosidic Ellagitannins from Lythrum Salicaria
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/pca

J. P. Piwowarski and A. K. Kiss
1
3
Table 2. C-NMR spectroscopic data of compounds 1–5 (100.61 MHz, acetone-d6)
a
Position
Compound 1 Compound 2 Compound 3 Compound 4 Compound 5
1
3
Glucosyl residue C (d
C
)
1
2
3
4
5
6
65.76
78.09
68.79
69.70
71.42
65.60
67.87
74.17
66.57
69.48
71.28
65.58
65.30
78.05
68.63
70.18
71.80
65.23
65.17
81.19
72.22
73.38
71.50
64.37
65.16
78.43
68.77
70.70
72.32
65.11
67.35
76.46
69.64
73.98
71.64
64.62
67.48
73.98
66.42
69.34
71.90
65.38
67.61
76.96
69.74
74.37
71.30
64.57
1
2
3
4
5
6
s
s
s
s
s
s
13
b
Carbonyl and aromatic C (dC) C = O 2 (A)
164.4
125.5
117.6
144.6
163.4
116.0
164.4
125.2
117.1
164.1
125.2
117.1
144.2
164.6
A 2
A 3
A 4
C = O 3 (B)
165.4
165.5
165.3
165.4
164.8
C = O 4 (C)
C 1
166.5
115.7
166.5
115.7
166.4
118.2
125.1
108.2
146.5
137.8
166.7
117.2
166.5
118.0
C 2
C 3
C 4
C 5
108.7
145.6
136.9
108.6
145.5
136.6
109.1
146.6
137.5
108.7
147.0
137.5
C = O 5 (D)
D 1
166.9
114.5
166.6
114.6
166.9
114.4
125.7
109.0
145.5
136.6
167.1
114.8
167.1
114.6
D 2
D 3
D 4
D 5
109.2
145.6
136.7
109.2
145.5
136.7
109.1
145.4
136.8
109.3
145.5
136.8
C = O 6 (E)
E 1
169.0
115.2
107.9
145.5
136.2
168.8
115.0
107.9
145.5
136.2
168.8
115.4
107.8
145.1
136.0
168.8
115.6
107.9
145.5
135.8
168.7
115.3
107.9
145.1
136.0
E 3
E 4
E 5
C = O 2 s (F)
165.4
122.9
119.2
164.4
120.7
116.0
164.7
F 2
F 3
C = O 3 s (G)
169.3
116.2
105.9
145.8
135.0
169.8
116.6
105.6
145.9
135.2
169.7
116.5
105.8
145.8
135.1
G 1
G 3
G 4
G 5
C = O 4 s (H)
168.6
116.0
125.3
108.8
145.5
136.9
168.9
116.5
168.8
116.4
H 1
H 2
H 3
H 4
H 5
108.8
145.5
137.3
108.2
145.3
137.1
CO 5 s (I)
165.7
115.3
110.0
140.2
138.0
144.0
166.2
165.8
I 2
I 6
I 3
I 4
I 5
110.2
139.9
137.4
144.5
109.8
140.7
137.4
143.5
(Continues)
wileyonlinelibrary.com/journal/pca
Copyright © 2012 John Wiley & Sons, Ltd.
Phytochem. Anal. 2013, 24, 336–348

C-Glucosidic Ellagitannins from Lythrum Salicaria
Table 2. (Continued)
a
Position
Compound 1 Compound 2 Compound 3 Compound 4 Compound 5
C = O 6 s (J)
168.7
115.4
107.9
145.3
136.5
168.9
116.3
108.6
145.9
137.2
168.8
115.5
106.7
145.5
136.7
J 1
J 3
J 4
J 5
a
See Figs 1 and 2 for numbering.
Carbonyl carbon attached to the glucose carbon of given number and ring of symbol given in parentheses.
b
Values in bold indicate significant downfield shifts attributed to the replacement of the hydroxyl group by an etheral phenyl substituent.
were present together with characteristic for the ellagitannins m/z
between sugar proton signals, carbonyl carbon and aromatic
proton signals (Fig. 2), leading to an almost full assignment of
carbonyl and aromatic carbon signals. Excluding significant down-
field shifts of signals attributed to aromatic carbons of ring C of
ꢂ
3
01 [HHDP ꢂ H] ion. In positive ion mode only the single charged
+
positive ion at m/z 935 [M + H] was present together with its so-
dium adduct at m/z 957 [M + Na] . Fragmentation products repre-
+
1
13
sented similar neutral losses to those in negative mode but no ion
HHDP unit, all H and C signals belonging to the vescalagin part
were almost superimposable with those determined for vescalagin
(Tables 1 and 2). Despite the absence of significant upfield shifts of
aromatic protons that would suggest the presence and orientation
of the valoneoyl group (Yoshida et al., 1992), the formation of
valoneic acid dilactone during acidic hydrolysis was established.
n
representing the HHDP unit was found. The HPLC–DAD–MS anal-
ysis of acidic hydrolysis products (Table 4) revealed the formation
ꢂ
of ellagic acid together with compound P1 (m/z 631 [M ꢂ H] )
n
(
Fig. 3), which can be identified based on the MS spectrum and
t
R
value as vescalin (Quideau et al., 2005). The S-configuration
13
of the HHDP group was confirmed by a positive Cotton effect at
This fact together with the remarkable downfield shifts of
signals among ring C aromatic carbons (including etheral carbon
C4 signal – d 146.5 ppm (Ito et al., 2002)) due to replacement of
C
236 nm in the CD spectrum (Table 3).
Castalagin (2) was obtained as an off-white amorphous
C
ꢂ
powder. Based on the [M ꢂ H] ion peak at m/z 933.0635 in
one of the hydroxyl group by an etheral phenyl substituent,
being the galloyl unit of the stachyurin part, led to the conclusion
that compound 3 is a dimeric ellagitannin composed of vescalagin
and stachyurin connected via formation of the valoneoyl group by
intermolecular C–O oxidative coupling of the galloyl unit of
stachyurin and ring C of the HHDP unit of vescalagin. The structure
ESI–TOF–MS, the molecular formula C41
It differed from compound 1 by the chemical shifts and elevated
coupling constant of an anomeric proton signal [d 5.71
d, J = 4.7 Hz, 1H), H1)], which indicated a orientation of the
26
H O26 was determined.
H
(
hydroxyl group at glucose C1 (Nonaka et al., 1990). All proton
and carbon signal assignments were consistent with those
reported in literature for castalagin (Tanaka et al., 1997). Despite
n
determined was supported by MS analysis (Table 3). The electro-
spray ionisation method, apart from the most abundant double-
n
2ꢂ
3ꢂ
MS fragmentation pattern and neutral losses being similar to
charged ion at m/z 933 [M ꢂ 2H] and m/z 622 [M-3H] , led to
the formation of fragment ions in the source. Among product
ions those representing the vescalagin part lacking the HHDP unit
those observed for compound 1, these two compounds can be
distinguished by their elution order (Table 3). The acidic hydrolysis
ꢂ
led to formation of ellagic acid and compound P2 (m/z 631
(m/z 631 [V ꢂ HHDP ꢂ H] ) or with an additional galloyl unit (m/z
ꢂ
ꢂ
[
Mꢂ H] ) (Table 4 and Fig. 3), which, based on its fragmentation
1101 [V + Gal ꢂ H] ) together with ions attributable to the loss of
ꢂ
pattern and t value, can be identified as castalin (Quideau et al.,
a galloyl unit from the stachyurin part (m/z 783 [S ꢂ Gal ꢂ H] )
R
2005). The positive Cotton effect at 238 nm in the CD spectra
were detected. Ionisation in positive ion mode was not
observed. Among acidic hydrolysis products (Table 4 and Fig. 3),
apart from already mentioned valoneic acid dilactone, ellagic
acid and vescalagin part-derived compound P1 were present.
The S-configuration at the HHDP and valoneoyl unit was estab-
lished by a diagnostic Cotton effect at 233 nm in the CD spectrum.
Salicarinin B (4) was obtained as an off-white amorphous powder
with molecular formula C H O determined by ESI–TOF
indicated the S-configuration of the HHDP group.
Salicarinin A (3) was obtained as an off-white amorphous
powder and assigned a molecular formula of C H O from
8
2 52 52
2
ꢂ
the [M ꢂ 2H]
ion peak at m/z 933.0642 in the ESI–TOF
spectrum. Based on the assignments of the H-NMR signals by
H– H COSY, it was deduced to be a dimeric ellagitannin
1
1
1
composed of two monomeric units having an open-chain glucose
core. Chemical shifts and coupling constants of two anomeric pro-
8
2 52 52
2
ꢂ
([M ꢂ 2H] at m/z 933.0644). Among two sets of sugar proton
H
ton signals [d 4.91 (d, J = 2.0Hz, 1H) H1 and 4.85 (d, J = 1.8Hz, 1H)
signals, one possessed a coupling constant value indicating b
H1 s] indicated b orientation of hydroxyl groups at both anomeric
orientation of the hydroxyl group at anomeric C1 [d
(d, J = 1.9Hz, 1H) H1], while the other coupling constant value
indicated a orientation [d 5.49 (d, J = 5.0Hz, 1H) H1]. The sugar
H
5.10
1
13
carbons. H– C HSQC spectra allowed assignment of two series of
sugar carbons, which together with sugar protons were closely
related to those determined for vescalagin and those reported in
the literature for stachyurin (Hatano et al., 1988; Nonaka et al.,
H
1
13
carbon assignment based on H– C HSQC spectra revealed that
the pattern of the former proton set together with the attributed
carbon signals are consistent with those determined for vescalagin.
In the case of the latter, comparison with literature data showed
the casuarinin to be the second dimer component (Okuda et al.,
1982a; Hatano et al., 1988). As in the case of compound 3, seven
1
990). In the aromatic region seven proton 1H singlets appeared
with only one attributable to the ortho proton of stachyurin’s galloyl
unit (d 7.08 s, 1H), which indicated that the single ortho carbon
H
from this unit takes part in formation of linkage with the vescalagin
part. The aromatic carbon signals were assigned based on the
1
aromatic proton signals were present in H-NMR spectra, with
1
13
H– C HMBC spectrum, which showed three bond correlations
one attributed to the casuarinin-derived galloyl group (Table 1).
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/pca

J. P. Piwowarski and A. K. Kiss
wileyonlinelibrary.com/journal/pca
Copyright © 2012 John Wiley & Sons, Ltd.
Phytochem. Anal. 2013, 24, 336–348

C-Glucosidic Ellagitannins from Lythrum Salicaria
The presence of valoneic acid dilactone among hydrolysis products
together with remarkable downfield shifts of signals attributed to
aromatic carbons of ring C led to the conclusion that compound 4
is a dimeric ellagitannin composed of vescalagin and casuarinin
connected via oxidative coupling of ring C of the HHDP of vescalagin
part with the galloyl unit of casuarinin, leading to the formation of
the valoneoyl group. These structural assignments were sustained
n
by ESI–MS analysis in negative ion mode (Table 3) and acidic
n
hydrolysis products HPLC–DAD–MS analysis (Table 4 and Fig. 3).
The chirality of the HHDP and valoneoyl groups was determined
to be S, based on the positive Cotton effect at 235 nm.
Salicarinin C (5) was isolated as an off-white amorphous
powder, the molecular formula of which was determined, based
2
ꢂ
on ESI–TOF (m/z 933.0643 [Mꢂ 2H] ) analysis, to be C H O .
8
2 52 52
The elevated coupling constants of two sugar anomeric proton
signals [d 5.50 (d, J = 4.6 Hz, 1H), H1 and d 5.60 (d, J = 5.0Hz, 1H),
H
H1 s] indicated that both sugar units posses hydroxyl groups at
anomeric C1 carbons in a orientation. Between two sets of signals
1
13
in the H and C spectra, one was almost superimposable with
that determined for castalagin and the other, based on literature
data, was consistent with casuarinin (Okuda et al., 1982a; Hatano
et al., 1988) (Tables 1 and 2). As in the case of compounds 3 and 4,
13
in compound 5 C spectra significant downfield shifts of signals
attributed to aromatic carbons of ring C were observed, and valoneic
n
acid dilactone was formed during acidic hydrolysis. The MS spectra
(Table 3) together with the hydrolysis products pattern supported
the determined structure. (Table 4 and Fig. 3). The orientation of
the valoneoyl and HHDP groups was determined to be S according
to the positive Cotton effect at 239 nm in the CD spectra.
Ma et al. (1996) previously isolated castalagin and four new
dimeric ellagitannins from Lythri herba: lythrines A, B, C and D,
which also contained stachyurin/casuarinin connected with
vescalagin/castalagin. In the suggested structures (lythrines A, B
and D) the galloyl group of stachyurin/casuarinin was attached
to ring E (connected to glucose carbon 6) of the HHDP group of
vescalagin/castalagin forming a valoneoyl interunit. Determined
for the first time in the present research, novel structures named,
after the plant latin name, salicarinins A–C, although composed
from the same C-glucosidic ellagitannins’ monomers, they differ
from structures proposed by Ma et al. (1996) by the valoneoyl
interunit orientation. In the case of salicarinins A–C the valoneoyl
interunit is formed by the galloyl group of stachyurin/casuarinin
attached to ring C (connected to glucose carbon 4) of the HHDP
group of vescalagin/castalagin (Fig. 2). Ma et al. (1996) also
described lythrin C, in which the monomeric parts were connected
by a C–C bond between the galloyl group and glucose carbon 1. In
the present research no ellagitannin with a corresponding dimeri-
sation mode was determined. Unfortunately due to the lack of
spectral and chemical data in Ma et al. (1996) article, discussion of
the significant structural differences raised cannot be conducted
comprehensively. However, it is possible that these differences in
the structures determined can be a consequence of a phytochemical
diversity of plant species from different habitats/continents.
TLC analysis
Optimisation of the separation of C-glucosidic ellagitannins was
performed on TLC and HPTLC plates pre-coated with silica gel
bonded phases of different polarities. The best separation condi-
tions were achieved on RP-18 HPTLC plates using a mobile phase
of water:acetonitrile:tetrahydrofuran:formic acid (80:5:15:2, v/v/v/v)
(Table 5 and Fig. 4). Separation was achieved for 1–3 and 3–2,
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/pca

J. P. Piwowarski and A. K. Kiss
but the resolution (Rs) values needed improvement. The propor-
tion of acetonitrile:tetrahydrofuran had the most significant influ-
ence on the R value of compound 3. The increased ratio of aceto-
f
nitrile:tetrahydrofuran caused elevation of the relative R value of
f
compound 3, which overlapped with compound 1 spot, but led
to a complete exposure of compound 2 spot. In contrast, when
the ratio of acetonitrile:tetrahydrofuran decreased, the relative Rf
value of compound 3 decreased and overlapped with compound
2
spot, but in these conditions compound 1 spot became more ex-
posed (Table 5). Satisfactory separation of ellagitannins was also
obtained using RP-2 plates with the mobile phase water:acetoni-
trile:tetrahydrofuran:formic acid (70:10:20:1, v/v/v/v). Unfortunately,
in these conditions the demixion zone appeared at the level of
compound 1 spot, even when the plate was pre-washed with
methanol, a saturation pad was used and pre-conditioning para-
meters were modified. Some separation was also achieved using
RP-8 and diphenyl modified plates, but the spots were not sharp
enough and Rs values were too low even for qualitative analysis
2
of the extract. In the case of NH - and CN-bonded stationary phases
different combinations of the polar mobile phase composition
resulted in migration of all ellagitannins with the solvent front.
The use of a normal phase approach did not cause any migration.
The developed HPTLC method on silica gel RP-18 plates can be
useful in quantitative and qualitative examination of C-glucosidic
ellagitannins from Lythri herba, especially as the separation of
target compounds can be easily enhanced with the appropriate
acetonitrile:tetrahydrofuran ratio. Until now only a few TLC methods
have been developed for ellagitannins determination (Fecka and
Cisowski, 2002; Bazylko et al., 2007; Fecka, 2009). Furthermore, in
the case of the C-glucosidic ellagitannins examined no planar
chromatography methods have been developed so far.
n
HPLC–DAD–MS analysis
Satisfactory separation was achieved on a Zorbax SB-C18 column
using a gradient system as follows: 0–5 min isocratic elution with
100% of mobile phase A; 5–10 min linear gradient from 0 to 5%
of B in A; 10–20 min 5–10% B; 20–30 min 10–20% B; 30–35 min
2
0–25% B; 35–40 min 25–60% B; 40–50 min 60–100% B.
HPLC–DAD–MS analysis revealed that the dominating
n
compounds in the extract are C-glucosidic ellagitannins: mono-
meric vescalagin (1) and castalagin (2); and dimeric salicarinin A,
B, C (3, 4, 5 respectively) (Fig. 4). Other polyphenolic compounds:
orientin, isoorientin, vitexin and ellagic acid were also detected
n
based on comparison of retention times, MS and UV spectra
of standard substances (Table 6). The separation parameters
obtained (Table 7) indicate that the HPLC method developed is
suitable for qualitative determination of C-glucosidic ellagitannins
in Lythri herba. Satisfactory separation was not achieved in the
case of the Hypersil GOLD C-18 column (Table 7).
The elution pattern of C-glucosidic ellagitannins on RP-18
modified stationary phases depended mostly on hydroxyl group
orientation at anomeric glucose carbon C1, and then on their size.
The retention times of C-glucosidic ellagitannins from Lythri herba
determined by HPLC analysis, together with the retention factors
values obtained on HPTLC plates, indicate that the b orientation
in the case of vescalagin (1) and both glucose cores of dimeric
salicarinin A (3) determined faster elution, then in case of
possessing a orientation castalagin (2), while salicarinin C (5)
with both parts with a orientation of hydroxyl groups had the
R f
highest t and the lowest R values. These observations are consistent
with the previously reported elution order for vescalagin and
wileyonlinelibrary.com/journal/pca
Copyright © 2012 John Wiley & Sons, Ltd.
Phytochem. Anal. 2013, 24, 336–348

C-Glucosidic Ellagitannins from Lythrum Salicaria
Figure 3. Structures of hydrolysis products of ellagitannins.
Table 5. Influence of acetonitrile:tetrahydrofurane ratio on ellagitannins’ 1–5 separation parameters on RP-18 modified HPTLC plates
Compound
Water:acetonitrile:tetrahydrofurane:formic acid ratio (v/v/v/v)
80:10:10:2
80:5:15:2
80:0:20:2
a
b
a
f
b
a
b
R
f
Rs
R
Rs
R
f
Rs
1
3
2
4
5
0.65
0.61
0.57
0.44
0.37
0.69
0.69
0.60
0.53
0.47
0.68
0.60
0.60
0.45
0.39
1.45
0.95
4.91
2.63
0.00
2.28
2.62
2.18
3.04
0.00
3.71
2.26
a
Mean values (n = 5).
b
Resolution (Rs) value depending on efficiency (N), selectivity (a) and retention parameter (k) according to Purnell’s formula.
castalagin (Zhentian et al., 1999). The UV-Vis chromatogram
revealed that the ellagitannins are the main compounds present
in the extract. As determined by phytochemical analysis (Fig. 4)
the level of flavonoids is low in comparison to the amount of
ellagitannins, which are probably the main determinants of
Lythri herba biological activity. Estimation based on HPLC–DAD
chromatogram is consistent with quantitative determinations
performed by Humadi and Istudor (2009) and Møller et al. (2009).
Rauha et al. (2001) determined the polyphenolic composition of
of their structural similarities and very high hydrophilicity, what
results in poor separation and extremely low retention times in
most of the commonly used reversed phase HPLC eluent
gradients. Taking into consideration the fact that structure–activity
relationships of ellagitannins have been firmly established
(Quideau and Feldman, 1996; Coca et al., 2009) and significant
differences of biological effects between monomeric and dimeric
ellagitannins (Feldman, 2005) have been observed, their separa-
tion during phytochemical and biological evaluation, especially
in the case of pharmacopoeial plant material, is necessary. Because
ellagitannins are made up from the same building blocks
organised into a large number of different structures possessing
characteristic but very often almost identical mass spectra, their
characterisation requires isolation and comprehensive spectro-
scopic analysis (Vrhovsek et al., 2012).
3
Lythri herba using the HPLC–DAD–MS method. The castalagin
and vescalagin were separated using HPLC. The presence of
vescalagin was inferred based on MS spectra and authors clearly
stated that isolation is required for deeper investigation of the
structure of the compounds detected.
This study presents the first detailed examination of the
composition of ellagitannins in Lythri herba, a tannin-rich pharma-
copoeial plant material. The presence of C-glucosidic ellagitannins,
vescalagin and castalagin, together with their dimers with
Conclusions
Common treatment of ellagitannins from Lythri herba as an
indistinguishable group of compounds is probably a consequence
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/pca

J. P. Piwowarski and A. K. Kiss
Figure 4. HPLC–DAD chromatogram of Lythri herba aqueous extract performed on Zorbax SB C18 (150 ꢁ 2.1 mm, 1.9 mm) column compared with a
RP-18 HPTLC chromatogram (mobile phase water:acetonitrile:tetrahydrofuran:formic acid (80:5:15:2 v/v/v/v)) derivatised with 3% iron (III) chloride in
methanol (w/v). O, orientin; IO, isoorientin; V, vitexin; EA, ellagic acid.
n
Table 6. Retention times, UV maxima and MS spectra of other polyphenolic compounds observed in Lythri herba extract
n
n
Compound
Isoorientin
t
R
m/z (ꢂ)
MS
m/z (+)
MS
UV, lmax (nm)
ꢂ
O ꢂ H]ꢂ
+
+
33.7
447 [M ꢂ H]
r
r
429 [M ꢂ H
2
449 [M + H] r
431 [M ꢂ H
2
O + H]
270, 350
ꢂ
+
8
95 [2 M ꢂ H] ꢂ
357 [M-90 ꢂ H]
383 [M ꢂ 66 + H]
353 [M ꢂ 96 + H]
+
3
27 [M-120 ꢂ H]ꢂ
+
3
29 [M ꢂ 120 + H]
ꢂ
357 [M-90 ꢂ H]ꢂ
+
+
+
Orientin
34.4
447 [M ꢂ H]
449 [M + H] r
431 [M ꢂ H
413 [M ꢂ H
2
2
O + H]
267, 350
8
95 [2 M ꢂ H]ꢂ
327 [M-120 ꢂ H]ꢂ
O + H]
+
3
3
3
83 [M ꢂ 66 + H]
+
53 [M ꢂ 96 + H]
+
29 [M ꢂ 120 + H]
+
ꢂ
ꢂ
+
Vitexin
36.0
36.4
431 [M ꢂ H]
r
r
413 [M ꢂ H2O ꢂ H]
433 [M + H] r
415 [M ꢂ H
2
O + H]
268, 336
254, 364
62 [2 M ꢂ H]ꢂ
341 [M ꢂ 90 ꢂ H]
ꢂ
367 [M ꢂ 66 + H]
337 [M ꢂ 96 + H]
+
8
+
3
11 [M ꢂ 120 ꢂ H]ꢂ
+
3
13 [M ꢂ 120 + H]
ꢂ
ꢂ H]ꢂ
+
+
Ellagic acid
301 [M ꢂ H]
257 [M ꢂ CO
229
83
2
627 [2 M + Na]
325 [M + Na]
303 [M + H] r
285 [M ꢂ H
2
O + H]
03 [2 M ꢂ H]ꢂ
+
259 [M ꢂ CO
+
6
2
+ H]
+
1
183
r, most abundant ion subjected to fragmentation.
wileyonlinelibrary.com/journal/pca
Copyright © 2012 John Wiley & Sons, Ltd.
Phytochem. Anal. 2013, 24, 336–348

C-Glucosidic Ellagitannins from Lythrum Salicaria
Table 7. Separation parameters of ellagitannins 1-5 on different HPLC RP-18 modified columns
Number
Zorbax SB C18 (150 ꢁ 2.1 mm, 1.9 mm)
Hypersil GOLD C18 (100 ꢁ 2.1 mm, 1.9 mm)
a
b
a
b
t
R
(min)
T
f
Rs
t
R
(min)
T
f
Rs
1
3
2
4
16.9
1.2
1.5
2.0
1.0
1.2
8.0
2.6
1.9
1.9
1.8
-
1
4
2
3
.53
.06
.90
.00
0.85
0.69
1.39
1.06
18.1
21.0
22.3
24.3
10.5
12.5
16.0
18.3
5
a
Mean values (n = 5).
b
Resolution (Rs) value depending on efficiency (N), selectivity (a) and retention parameter (k) according to Purnell’s formula.
stachyurin and casuarinin- salicarinins A–C, was firmly established.
The structures of monomeric and novel dimeric C-glucosidic
ellagitannins were determined according to their ESI–TOF–MS,
Hager TJ, Howard LR, Liyanage R, Lay JO, Prior RL. 2008. Ellagitannin
composition of blackberry as determined by HPLC–ESI–MS and
MALDI–TOF–MS. J Agric Food Chem 56: 661–669.
13
Hatano T, Yoshida T, Shingu T, Okuda T. 1988. C Nuclear magnetic
n
NMR, ESI–MS , CD spectra and hydrolysis products analysis.
resonance spectra of hydrolyzable tannins. III. Tannins having 1C4
glucose and C-glucosidic linkage. Chem Pharm Bull 36: 3849–3856.
Humadi SS, Istudor V. 2009. Lythrum salicaria (purple loosestrife). Medicinal
use, extraction and identification of its phenolic compounds. Farmacia
Structures of the isolated dimeric ellagitannins salicarinins A, B
and C have not been reported previously in literature. The LC,
HPTLC and HPLC methods were developed for separation and
qualitative evaluation of Lythri herba ellagitannins. The necessity
of determining ellagitannins composition in Lythri herba in
connection with their structure-dependent biological activities
reported in literature is emphasised.
5
7: 192–200.
Ito H, Yamaguchi K, Kim TH, Khennouf S, Gharzouli K, Yoshida T. 2002.
Dimeric and trimeric hydrolyzable tannins from Quercus coccifera
and Quercus suber. J Nat Prod 65: 339–345.
Kiss AK, Granica S, Stolarczyk M, Melzig MF. 2012. Epigenetic modulation of
mechanisms involved in inflammation: Influence of selected polyphenolic
substances on histone acetylation state. Food Chem 131: 1015–1020.
Lamela M, Cadavid I, Calleja JM. 1986. Effects of Lythrum salicaria extracts
on hyperglycemic rats and mice. J Ethnopharm 15: 153–160.
Ma X, Ji C, Wang Y, Zhang G, Liu Y. 1996. New tannins from Lythrum
salicaria L. J Chin Pharm Sci 5: 225.
Acknowledgements
The authors are grateful to Leszek Siergiejczyk from the Medical
University of Bialystok for performing the NMR experiments and
Magdalena Popławska from the Medical University of Warsaw for
Møller C, Hansen SH, Cornett C. 2009. Characterisation of tannin-containing
herbal drugs by HPLC. Phytochem Anal 20: 231–239.
n
performing the ESI–TOF–MS experiments. The HPLC–ESI–MS
and TLC experiments were conducted on apparatus at the Centre
for Preclinical Research and Technology (CePT).
Moskalenko SA. 1986. Preliminary screening of far-eastern ethnomedicinal
plants for antibacterial activity. J Ethnopharm 15: 231–259.
Nagy M, Rodinová K. 2006. Lythrum salicaria L. (purple loosestrife) – A
new plant in Slovak pharmacopoeia 1. Lythrum salicaria L. (vrbica
vŕbolistá) – Nová rastlina v Slovenskom liekopise 1 75: 280–283.
Nonaka G, Sakai T, Tanaka T, Mihashi K, Nishioka I. 1990. Tannins and
related compounds. XCVII. Structure revision of C-glycosidic ellagi-
tannins, castalagin, vescalagin, casuarinin and stachyurin, and related
hydrolyzable tannins. Chem Pharm Bull 38: 2151–2156.
References
Bazylko A, Kiss AK, Kowalski J. 2007. High-performance thin-layer chro-
matography method for quantitative determination of oenothein B
and quercetin glucuronide in aqueous extract of Epilobii angustifolii
herba. J Chromatogr A 1173: 146–150.
Becker H, Scher JM, Speakman JB, Zapp J. 2005. Bioactivity guided isolation
of antimicrobial compounds from Lythrum salicaria. Fitoterapia 76:
Okuda T, Yoshida T, Ashida M, Yazaki K. 1982a. Casuariin, stachyurin and
strictinin, new ellagitannins from Casuarina stricta and Stachyurus
praecox. Chem Pharm Bull 30: 766–769.
5
80–584.
Okuda T, Yoshida T, Hatano T, Koga T, Toh N, Kuriyama K. 1982b. Circular
dichroism of hydrolysable tannins-I ellagitannins and gallotannins.
Tetrahedron Lett 23: 3937–3940.
Bencsik T, Horváth G, Papp N. 2011. Variability of total flavonoid,
polyphenol and tannin contents in some Lythrum salicaria populations.
Nat Prod Commun 6: 1417–1420.
Cadavid Torres I, Calleja Suarez J. 1980. A preliminary study of hypoglyce-
mic activity of Lythrum salicaria. J Nat Prod 43: 559–563.
Çoban T, Çitoglu GS, Sever B, Işcan M. 2003. Antioxidant activities of
plants used in traditional medicine in Turkey. Pharm Biol 41: 608–613.
Coca A, Feldman KS, Lawlor MD. 2009. Immunomodulatory ellagitannin
chemistry. In Chemistry and Biology of Ellagitannins, Quideau S (ed.).
World Scientific Publishing: Singapore; 203–272.
Pawlaczyk I, Capek P, Czerchawski L, Bijak J, Lewik-Tsirigotis M, Pliszczak-Król
A, Gancarz R. 2011. An anticoagulant effect and chemical characterization
of Lythrum salicaria L. glycoconjugates. Carbohydr Polym 86: 277–284.
Piwowarski JP, Kiss AK, Kozłowska-Wojciechowska M. 2011. Anti-
hyaluronidase and anti-elastase activity screening of tannin-rich
plant materials used in traditional Polish medicine for external
treatment of diseases with inflammatory background. J Ethnopharm
137: 937–941.
Fecka I. 2009. Development of chromatographic methods for determina-
tion of agrimoniin and related polyphenols in pharmaceutical products.
J AOAC Int 92: 410–418.
Fecka I, Cisowski W. 2002. TLC determination of tannins and flavonoids in
extracts from some Erodium species using chemically modified
stationary phases. J Planar Chromatogr 15: 429–432.
Quideau S, Feldman KS. 1996. Ellagitannin chemistry. Chem Rev 96: 475–503.
Quideau S, Jourdes M, Lefeuvre D, Montaudon D, Saucier C, Glories Y,
Pardon P, Pourquier P. 2005. The chemistry of wine polyphenolic
C-glycosidic ellagitannins targeting human topoisomerase II. Chem
Eur J 11: 6503–6513.
Quideau S, Deffieux D, Douat-Casassus C, Pouységu L. 2011. Plant
polyphenols: Chemical properties, biological activities, and synthesis.
Angew Chem Int Ed 50: 586–621.
Feldman KS. 2005. Recent progress in ellagitannin chemistry. Phytochem
66: 1984–2000.
Phytochem. Anal. 2013, 24, 336–348
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/pca

J. P. Piwowarski and A. K. Kiss
Rauha JP, Remes S, Heinonen M, Hopia A, Kähkönen M, Kujala T, Pihlaja K,
Vuorela H, Vuorela P. 2000. Antimicrobial effects of Finnish plant
extracts containing flavonoids and other phenolic compounds. Int J
Food Microbiol 56: 3–12.
wood (Castanea crenata Sieb. et Zucc.). Chem Pharm Bull 45:
1751–1755.
Tunalier Z, Koşar M, Küpeli E, Çaliş I, Başer KHC. 2007. Antioxidant, anti-
inflammatory, anti-nociceptive activities and composition of
Lythrum salicaria L. extracts. J Ethnopharm 110: 539–547.
Vrhovsek U, Guella G, Gasperotti M, Pojer E, Zancato M, Mattivi F. 2012.
Clarifying the identity of the main ellagitannin in the fruit of the
strawberry, Fragaria vesca and Fragaria ananassa Duch. J Agric Food
Chem 60: 2507–2516.
Rauha JP, Wolfender JL, Salminen JP, Pihlaja K, Hostettmann K, Vuorela H.
2001. Characterization of the polyphenolic composition of purple
loosestrife (Lythrum salicaria). Z Naturforsch 56: 13–20.
Romani A, Campo M, Pinelli P. 2012. HPLC/DAD/ESI-MS analyses and anti-
radical activity of hydrolyzable tannins from different vegetal species.
Food Chem 130: 214–221.
Saltan Çitoglu G, Altanlar N. 2003. Antimicrobial activity of some plants
used in folk medicine. J Fac Pharm 32: 159–163.
Yoshida T, Hatano T, Okuda T, Memon MU, Shingu T, Inoue K. 1984.
1
3
Spectral and chromatographic analyses of tannins. I. C nuclear
magnetic resonance spectra of hydrolyzable tannins. Chem Pharm
Bull 32: 1790–1799.
Sanz M, Cadahía E, Esteruelas E, Muñoz ÁM, Fernández De Simón B,
Hernández T, Estrella I. 2010. Phenolic compounds in chestnut
Yoshida T, Hatano T, Kuwajima T, Okuda T. 1992. Oligomeric hydrolyzable
1
(
Castanea sativa Mill.) heartwood. Effect of toasting at cooperage.
tannins – Their H NMR spectra and partial degradation. Heterocycles
J Agric Food Chem 58: 9631–9640.
Tanaka T, Ueda N, Shinohara H, Nonaka GI, Kouno I. 1997. Four new
C-glycosidic ellagitannins, castacrenins D - G, from Japanese chestnut
33: 463–482.
Zhentian L, Jervis J, Helm RF. 1999. C-glycosidic ellagitannins from white
oak heartwood and callus tissues. Phytochem 51: 751–756.
wileyonlinelibrary.com/journal/pca
Copyright © 2012 John Wiley & Sons, Ltd.
Phytochem. Anal. 2013, 24, 336–348