Composition of the fresh leaves and stems of Melissa officinalis and evaluation of skin irritation in a reconstituted human epidermis model

Article abstract of DOI:10.1021/np9003195

The composition of a centrifuged product obtained from the fresh leaves and stems of Melissa officinalis and skin irritation in the reconstituted human epidermis (Episkin model) have been investigated in comparison to the EtOH-H₂O (1:1) extract

Full text of DOI:10.1021/np9003195

1512
J. Nat. Prod. 2009, 72, 1512–1515
Composition of the Fresh Leaves and Stems of Melissa officinalis and Evaluation of Skin
Irritation in a Reconstituted Human Epidermis Model
Teresa Mencherini,† Patrizia Picerno,† Paola Russo,† Marisa Meloni,‡ and Rita Aquino*,†
Dipartimento di Scienze Farmaceutiche, UniVersity of Salerno, Via Ponte Don Melillo, 84084, Fisciano, Salerno, Italy, and VitroScreen SrL,
Piazza G. Cesare 17, 20145, Milan, Italy
ReceiVed May 26, 2009
The composition of a centrifuged product obtained from the fresh leaves and stems of Melissa officinalis and skin
irritation in the reconstituted human epidermis (Episkin model) have been investigated in comparison to the EtOH-H2O
(1:1) extract obtained by Soxhlet from the dried plant. Two new sulfated triterpenes (1 and 2) and two ionol derivatives
have been isolated for the first time from Melissa officinalis together with caffeic and rosmarinic acids. The structures
of compounds 1 and 2 were established by analysis of their spectroscopic data. Both the centrifuged material and its
major constituents neither affected cell viability nor caused the release of pro-inflammatory mediators or the decrease
of trans-epithelial electrical resistance (TEER) in the reconstituted human epidermis.
Lemon balm (Melissa officinalis L., Lamiaceae) is a very popular
aromatic perennial herb growing in the Mediterriaean region and is
used in the traditional medicine of Europe.1 The volatile oil1,2 has
antibacterial and antifungal activities.3,4 The hydroalcoholic extract
possesses antioxidative properties related to the phenolic content,
particularly of rosmarinic acid.5-7 In previous work on M. officinalis,
we reported the isolation of phenols and six new ursene and oleanene
derivatives from the 1:1 EtOH-H2O extract of the dried leaves and
stems obtained by Soxhlet extraction, as well as the significant free-
radical-scavenging and antimicrobial activities exhibited by the extract
and its major constituent, rosmarinic acid.8 Topical remedies prepared
from extracts, juice, and plant homogenates of fresh lemon balm herb
are used for its antiviral, antimicrobial, antioxidative, and antipruritic
activities, in addition to the prevention of dandruff.9-13 However,
information on the constituents and bioactivity of the fresh M. officinalis
plant is lacking. The present paper reports on the composition of the
centrifuged material obtained using a tumble-dryer from the fresh
leaves and stems of this species using EtOH-H2O (1:1) and on the
effect of its major constituents on tissue viability and release of pro-
inflammatory mediators in a reconstituted human epidermis (RHE)
model. In addition, two known ionol derivatives and two new sulfated
triterpenes, 1 and 2, have been isolated for the first time from M.
officinalis. The structures of 1 and 2 were elucidated using spectro-
scopic methods including 1D- (1H and 13C) and 2D-NMR (DQF-
COSY, HSQC, and HMBC) experiments as well as HRESIMS
analysis. Differences in the composition as well as in the antioxidant
and antimicrobial activities of the fresh centrifuged material with
respect to the extract from dried plant were also evaluated.
was dried in vacuo and then partitioned between water and n-BuOH.
The n-BuOH-soluble fraction was subjected to purification over
Sephadex LH-20 and by RP-HPLC to yield a new sulfated triterpene
glucoside (2) and an R-ionol glucoside from fraction I, an R-ionol
from fraction II, caffeic acid from fraction III, and rosmarinic acid
and a new sulfated triterpene (1) from fraction IV. The known
compounds were identified by comparison of their NMR data with
those from the literature as 3,5,6-trihydroxydehydro-R-ionol,14
3,5,6-trihydroxydehydro-R-ionol 9-O-ꢀ-D-glucopyranoside,15 caffeic
acid,16 and rosmarinic acid.16 Rosmarinic acid has been previously
isolated from the dried leaves and stems of the plant, but ionol
derivatives are reported for the first time from M. officinalis.
The total phenolic content of the centrifuged material, as
determined by the Folin-Ciocalteau method and expressed as a
rosmarinic acid equivalent, and the concentrations of rosmarinic
and caffeic acids, as determined by HPLC, were lower than in the
extract from the dried plant (21.0 µg/mg, 0.08% and 0.01% in the
fresh centrifuged product, compared with 250 µg/mg, 5.6% and
0.1%).8 In the DPPH test, the well-recognized free-radical-
scavenging activity of rosmarinic and caffeic acids was confirmed
(EC50 3.1 and 3.3 µg/mL, respectively), but no activity was shown
by the ionol derivatives or by sulfated triterpenoids 1 and 2, and
the antioxidant activity of the centrifuged material was low
compared to that of the extract from dried plant (EC50 678.3 µg/
mL vs 18.5 µg/mL, respectively) (see also Table S1, Supporting
Information). No antimicrobial activity was observed, for either
the centrifuged or isolated compounds, against several Gram-
positive and Gram-negative bacteria, a yeast, and a mold, as tested
by a broth microdilution method (see Table S2, Supporting
Information). These findings demonstrate that the centrifuged
material from M. offinalis possesses a rather low antioxidant and
no antimicrobial activity, which may be correlated to its low total
phenolic and rosmarinic acid content. Furthermore, the centrifuged
material contained a series of characteristic compounds, such as
ionol derivatives and sulfate triterpenes 1 and 2, not found in the
extract from the dried plant.
The structure elucidation of compounds 1 and 2 proceeded as
follows. The HRESIMS of 1 showed a major ion peak at m/z
583.2915 [M - H]-, consistent with the molecular formula
C30H48O9S (calcd for C30H47O9S, m/z 583.2941) and suggesting a
triterpene derivative having a sulfate group in the molecule. MS-
MS analysis of the ion peak showed a fragment ion at m/z 503.332
[M - H - 80]-, indicating the loss of a sulfate group. Acid
hydrolysis of 1, followed by treatment with BaCl2, gave a white
precipitate, confirming the presence of a sulfate residue.17 Inspection
of the 1H and 13C NMR data (Table 1), assigned by 1H-1H DQF-
Fresh stems and leaves of M. officinalis were centrifuged with
EtOH-H2O (1:1) using a tumble-dryer. The centrifuged product
* To whom correspondence should be addressed. Tel: ++39 089 969737.
Fax: ++39 089 969602. E-mail: aquinorp@unisa.it.
† University of Salerno.
‡ VitroScreen SrL, Milan.
10.1021/np9003195 CCC: $40.75  2009 American Chemical Society and American Society of Pharmacognosy
Published on Web 08/04/2009

Notes
Journal of Natural Products, 2009, Vol. 72, No. 8 1513
Table 1. 13C and 1H NMR Data of Compounds 1 and 2 in
of the pertinent hydrogen (H-23a and H-23b) and carbon (C-23)
signals.8 Full assignments of the proton and carbon resonances,
obtained by 1H-1H DQF-COSY, HSQC, and HMBC spectra (Table
1), indicated that 2 is a ꢀ-glucopyranosyl (anomeric proton signal,
δH 5.35, 1H, d, J ) 7.6 Hz) derivative of compound 1. The
ꢀ-glucopyranosyl unit could be located at C-28 on the basis of the
HMBC correlation observed between the anomeric proton signal
and the carbon resonance at δC 179.5 (C-28). The sugar was
determined to be D-glucose after hydrolysis of 2 with 1 N HCl and
GC analysis. Thus, the structure of 2 was assigned as the 23-sulfate
ester of 2R,3ꢀ,19R,23-tetrahydroxyurs-12-en-28-oic acid 28-O-ꢀ-
D-glucopyranoside (the 23-sulfate ester of niga-ichigoside F1).
To assess the safe topical use of the fresh centrifuged material
of the M. officinalis, this and compound 1 were each tested on
reconstituted human epidermis (RHE, Episkin-SM)18 in vitro, in
comparison to the extract from the dried plant and its major
constituent, rosmarinic acid, at concentrations higher than those
(0.01-1.00%) used in topical remedies.12,13 The Episkin model
has been reported to mimic morphologically and biochemically
living skin and has been validated to classify skin irritants able to
produce a decrease in cell viability, as evaluated by a MTT assay,
below defined threshold levels (e50%)18,19 and an increased release
of the pro-inflammatory mediator interleukin IL-1R, as evaluated
by an ELISA assay,18,19 after an acute exposure (15 min followed
by 42 h of recovery).18,19 To better discriminate nonirritant products,
a third end point was evaluated, trans-epithelial electrical resistance
(TEER), measuring the barrier function of the epithelium.19,20
Solutions (2%) of the centrifuged plant material and of the extract
from the dried plant, and 1% solutions of compound 1 (17.1 mM)
and rosmarinic acid (27.7 mM) in PBS (phosphate buffer) were
applied directly on the Episkin surface in comparison to PBS as
control and to a surfactant (sodium dodecyl sulfate, SDS 5%), as
an irritant reference compound. After topical exposure for 15 min
followed by 42 h of recovery, tissue viability, release of IL-1R,
and TEER values were determined. The epidermis viability,
measured as optical density at 570 nm by the MTT assay21 and
calculated as percentage of cytotoxicity compared to the control
(PBS-treated epidermis), was not significantly reduced after ap-
plication of all product tested (viability >75%, see Figure S1,
Supporting Information). No increased release of IL-1R, expressed
as pg/mL in the medium underneath, was observed in tissue treated
with all products at 42 h, compared to the control. Using the same
test conditions, SDS (5%) caused an elevation of the pro-
inflammatory mediator release to about 300 pg/mL (Figure 1).
Finally, no significant decrease of TEER values was observed in
epidermis treated with all products from M. officinalis, compared
to the control. The irritant surfactant, SDS, impairing the barrier
integrity of the epithelium, caused a reduction of -14 kΩ cm2 in
TEER (Figure 2). Although compound 1 possesses a surfactant-
like structure, it showed neither cytotoxic nor pro-inflammatory
effects or affected TEER reduction in the RHE. On the basis of
the absence of skin-irritant effects at the concentrations tested,
centrifuged lemon balm and compound 1, as well as the dried balm
extract and rosmarinic acid, appear to be safe for topical use.
CD3ODa
1
2
position δC
δH (J in Hz)b
δC
δH (J in Hz)b
1
47.8 0.91, 1.97, m
47.8 0.93, 1.97, m
2
69.1 3.71, ddd, (3, 9.2, 13.0) 69.2 3.72, ddd, (3, 9.2, 13.0)
3
4
77.3 3.43, d, (9.2)
43.3
77.5 3.42, d, (9.2)
43.3
5
6
7
8
48.1 0.94, m
18.7 1.42, 1.60, m
33.5 1.31, 1.69, m
40.7
48.2 0.95, m
18.8 1.41, 1.58, m
33.2 1.31, 1.68, m
40.8
9
48.4 1.81, m
38.5
24.9 2.04, 1.94, m
128.7 5.30, t (3.5)
140.9
48.1 1.81, m
38.6
24.9 2.04, 0.98, m
128.9 5.33, t (3.5)
140.0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
42.5
42.3
29.5 0.98, 1.94, m
26.7 2.46, 1.54, m
49.4
55.6 2.61, s
73.7
42.9 1.39, m
27.3 1.18, 1.72, m
39.4 1.63, 1.81, m
70.2 3.76, d (9.6)
3.93, d (9.6)
13.6 0.78, s
17.3 1.06, s
18.1 0.88, s
24.5 1.34, s
177.7
29.1 0.96, 1.86, m
26.6 2.65, 1.65, m
49.1
54.8 2.54, s
72.2
42.9 1.35, m
27.0 1.19, 1.62, m
38.3 1.63, 1.87, m
70.3 3.76, d (9.6)
3.94, d (9.6)
24
25
26
27
13.6 0.76, s
17.5 1.05, s
17.4 0.79, s
24.5 1.35, s
28
179.5
29
30
27.3 1.22, s
16.5 0.95, d (6.5)
26.8 1.23, s
16.7 0.95, d (6.5)
95.3 5.35, d (7.6)
73.6 3.34, dd (9.0, 7.6)
78.2 3.36, t (9.0)
71.0 3.40, t (9.0)
77.8 3.27, m
62.2 3.71, dd (12.0, 4.0)
3.83, dd (12.0, 2.5)
Glc-1′
Glc-2′
Glc-3′
Glc-4′
Glc-5′
Glc-6′
a Assignments confirmed by 2D COSY, HSQC, and HMBC exper-
b
iments. 1H-1H coupling constants (Hz) were measured from the COSY
spectra.
COSY, HSQC, and HMBC experiments, and comparison with
chemical shift values of related glycosides isolated from the dried
plant extract suggested a 23-hydroxytormentic acid derivative.8 The
hydrogen and carbon resonances were superimposable on those of
the 3,23-disulfate ester of 23-hydroxytormentic acid,8 except for
the H-3ax and H-2ax (δH 3.43, d, J ) 9.2 Hz and δH 3.71, ddd, J
) 3, 9.2, 13.0 Hz, respectively) and C-3 (δC 77.3) signals. The
upfield shifts observed for these signals with respect to the model
compound8 suggested the absence of the sulfate group at C-3.
1H-1H DQF-COSY experiments confirmed the proton sequence
H-1a (δH 1.97), H-1b (δH 0.91), H-2 (δH 3.71), and H-3 (δH 3.43).
The chemical shifts of the H-23a (δH 3.93) and H-23b (δH 3.76)
signals and the HMBC correlations observed between the H-23a
and H-23b resonance and C-4 (δC 43.3), C-3 (δC 77.3), C-5 (δC
48.1), and C-24 (δC 16.6) confirmed the placement of the sulfate
group at C-23.8 Thus, compound 1 was identified as the 23-sulfate
ester of 2R,3ꢀ,19R,23-tetrahydroxyurs-12-en-28-oic acid (23-sul-
fate ester of 23-hydroxytormentic acid).
Experimental Section
General Experimental Procedures. Melting points were determined
using a Mettler-Toledo DSC822e apparatus. Optical rotations were
measured on a JASCO DIP-1000 digital polarimeter equipped with a
sodium lamp (589 nm) and a 10 cm microcell in MeOH solution. For
NMR experiments, a Bruker DRX-600 NMR spectrometer was used,
The HRESIMS of 2 (m/z 745.3502 [M - H]-, calcd for C36H57O14S,
745.3469) supported the molecular formula C36H58O14S, suggesting
a triterpene sulfate with an additional hexose unit in the molecule
with respect to compound 1. The MS-MS showed the most intense
ion at m/z 583.289 [M - H - 162]-, ascribable to the loss of a
hexose unit. As in compound 1, acid hydrolysis of 2, followed by
treatment with BaCl2, again demonstrated the presence of a sulfate
residue, and its position at C-23 was deduced by the chemical shifts
1
operating at 599.2 MHz for H and at 150.9 MHz for 13C and using
the UXNMR software package; chemical shifts are expressed in δ (parts
per million) referring to the solvent peaks δH 3.34 and δC 49.0 for
CD3OD; coupling constants, J, are in hertz. 1D- and 2D-NMR
experiments were carried out using conventional pulse sequences.
ESIMS was performed on a Finnigan LC-Q Deca instrument (Ther-
moquest, San Jose, CA) equipped with Xcalibur software. Exact masses
(HRESIMS) were measured by a Q-TOF Premier (Waters) triple-

1514 Journal of Natural Products, 2009, Vol. 72, No. 8
Notes
g). A portion (2.5 g) of the n-BuOH extract was chromatographed over
a Sephadex LH-20 column (1 m × 3 cm i.d.) using MeOH as eluent
(flow rate 0.5 mL min-1). Fractions (8 mL each) were collected and
checked by TLC (silica gel, using n-BuOH-AcOH-H2O (60:15:25)
and CHCl3-MeOH-H2O (7:3:0.3) as solvents. Fractions with similar
Rf values were combined, giving four major fractions (I-IV), which
were further purified by RP-HPLC on a 30 cm × 7.8 mm i.d.
µ-Bondapack column (flow rate 2.0 mL min-1). Fraction I (413 mg)
was chromatographed with MeOH-H2O (35:65) as mobile phase to
yield 3,5,6-trihydroxydehydro-R-ionol 9-O-ꢀ-D-glucopyranoside (4.0
mg) and compound 2 (4.6 mg). Fractions II (704 mg) and IV (223
mg) were purified using MeOH-H2O (3:7) as mobile phase. Fraction
II yielded 3,5,6-trihydroxydehydro-R-ionol (2.8 mg). Fraction IV gave
rosmarinic acid (120 mg) and compound 1 (5.1 mg). Fraction III (316
mg) was purified with MeOH-H2O (4:6) as the eluent to afford caffeic
acid (4.0 mg).
Compound (1): white powder; mp 211 °C; [R]31D +19.6 (c 0.066,
MeOH); 1H (CD3OD, 600 MHz) and 13C NMR (CD3OD, 150.9 MHz)
data are reported in Table 1; HRESIMS m/z 583.2915 [M - H]- (calcd
for C30H47O9S, 583.2941).
Compound (2): white powder; mp 237 °C; [R]31 +5.9 (c 0.113,
D
MeOH); 1H (CD3OD, 600 MHz) and 13C NMR (CD3OD, 150.9 MHz)
data are reported in Table 1; HRESIMS m/z 745.3502 [M - H]- (calcd
for C36H57O14S, 745.3469).
Figure 1. Effect of the centrifuged material (2%), the extract from
dried plant (2%, dried plant ext.), rosmarinic acid (1%, 27.7 mM),
and compound 1 (1%, 17.1 mM) on Il-1R release (pg/mL of
medium), estimated by an ELISA assay, in reconstituted human
epidermis (Episkin-SM) after 15 min of exposure followed by 42 h
of recovery compared to a control (PBS) and a toxic reference
compound, SDS (5%). Values are expressed as mean ( SD of three
independent experiments. Comparison were performed using a one-
way ANOVA test, **p < 0.01 calculated vs control (PBS).
3,5,6-Trihydroxydehydro-r-ionol: 1H and 13C NMR data were
consistent with those previously reported;14 ESIMS m/z 243 [M - H]-.
1
3,5,6-Trihydroxydehydro-r-ionol 9-O-ꢀ-D-glucopyranoside: H
and 13C NMR data were consistent with those previously reported;15
ESIMS m/z 405 [M - H]-.
Caffeic acid: 1H and 13C NMR data were consistent with those
previously reported;16 ESIMS m/z 179 [M - H]-.
Rosmarinic acid: 1H and 13C NMR data were consistent with those
previously reported;16 ESIMS m/z 359 [M - H]-.
Acid Hydrolysis. Compound 2 (0.8 mg) in 1 N HCl (0.25 mL) was
subjected to hydrolysis, derivatized in 1-(trimethylsilyl)imidazole and
pyridine (0.1 mL), and analyzed by GC using an L-Chirasil-Val column
(0.32 mm × 25 m), according to a method previously reported.8 The
peak from the hydrolysate of 2 was detected at 14.74 min (D-glucose).
The peak for standard D-glucose was detected at 14.71 min.
Sulfate Group Detection. Both compounds 1 and 2 (1 mg of each)
were refluxed with 10% HCl (4 mL) for 4 h and then extracted with
Et2O. An aliquot of the aqueous layer of each was treated with 70%
BaCl2 to give a white precipitate (BaSO4).17
Quantitative HPLC Analysis of the Extracts. Quantitative HPLC
was conducted using a 150 × 3.9 mm i.d. C18 µ-Bondapack column.
The solvents, elution gradient, and all experimental conditions were
as reported in a previous paper.8 Analysis was carried out in triplicate.
Reference standard solutions of rosmarinic acid were prepared at three
concentration levels in the range 0.25-2.00 mg/mL. The peak
associated with rosmarinic acid was identified by retention time and
spectroscopic UV and MS comparison with a standard, and confirmation
was by co-injection. The centrifuged product and its n-BuOH-soluble
portion were redissolved in MeOH and analyzed under the same
chromatographic conditions.
Quantitative Determination of Total Phenols. The centrifuged
material and its n-BuOH-soluble portion, dissolved in MeOH, were
analyzed for their total phenolic content according to the Folin-Ciocalteau
colorimetric method.8 Total phenols were expressed as rosmarinic acid
equivalents.
Bleaching of the Free Radical 1,1-Diphenyl-2-picrylhydrazyl
(DPPH Test). The antiradical activities of the centrifuged plant material,
rosmarinic acid, and R-tocopherol were determined using the stable
1,1-diphenyl-2-picrylhydrazyl radical (DPPH) method, according to the
procedure previously described by Mencherini et al.8 Briefly, an aliquot
(37.5 µL) of the MeOH solution containing different amounts of the
centrifuged material and its n-BuOH-soluble portion, or rosmarinic acid,
was added to 1.5 mL of DPPH solution (0.025 g/L in MeOH) daily
prepared; the maximum concentration employed was 200 µg/mL. An
equal volume (37.5 µL) of the vehicle alone was added to control tubes.
Absorbances at 515 nm were measured on a Shimadzu UV-1601
UV-visible spectrophotometer 10 min after starting the reaction.
R-Tocopherol (EC50 10.1 ( 1.3 µg/mL) was used as a positive control
in the test. All experiments were carried out in triplicate, and the mean
Figure 2. Effect of the centrifuged material (2%), the extract from
dried plant (2%, dried plant ext.), rosmarinic acid (1%, 27.7 mM),
and compound 1 (1%, 17.1 mM) on TEER (trans-epidermal electric
resistance, kΩ cm2), estimated using Millicel-ERS (resistance range
0-20 kΩ) volt/ohm-meter, in reconstituted human epidermis
(Episkin-SM) at time 0 and after 15 min of exposure followed by
42 h of recovery compared to PBS and toxic reference compound
SDS (5%). Values are expressed as mean ( SD of three inde-
pendent experiments. Comparisons were performed using a one-
way ANOVA test, **p < 0.01 calculated vs control (PBS).
quadrupole orthogonal time-of-flight (TOF) instrument having an
electrospray ionization source. HPLC separations were performed with
a Waters 590 series pumping system equipped with a Waters R401
refractive index detector, a µ-Bondapak C18 column (300 × 7.8 mm
i.d.), and a U6K injector. Quantitative HPLC analysis was carried out
on an Agilent 1100 series system equipped with a model G-1312 pump,
a Rheodyne model G-1322A loop (20 µL), and a DAD G-1315 A
detector. Peak areas were calculated with an Agilent integrator.
Plant Material. The aerial parts (stems and leaves) of M. officinalis
L. were collected at Trezzo sull’Adda, Italy, in May 2005 and identified
by Dr. A. Facchini, Respharma. A voucher sample (RES 0049 MM)
was deposited at the Herbarium of Respharma s.r.l., Trezzo sull’Adda,
Italy.
Extraction and Isolation. Fresh aerial parts (stems and leaves) of
M. officinalis (500 g) were centrifuged in a ALC-PK 120 (Thermo
Electron Corporation) tumble-dryer apparatus with EtOH-H2O 1:1 (1
L). The centrifuged material was dried in vacuo and then partitioned
between n-BuOH and H2O to afford a n-BuOH-soluble portion (5.0

Notes
Journal of Natural Products, 2009, Vol. 72, No. 8 1515
effective scavenging concentrations (EC50) were calculated using the
Litchfield and Wilcoxon protocol22 (see Table S1, Supporting Informa-
tion).
Statistical Analysis. Data on cell viability are expressed as percent-
age of viability versus negative control (PBS-treated epidermis). Data
on IL-1R release are expressed in pg/mL in the medium underneath.
TEER values are expressed in kΩ cm2. All data were calculated from
mean ( SD values of three independent determinations. Statistical
analysis was performed by ANOVA test and multiple comparison by
a Bonferroni test.22 All experiments were made at least three times,
each time with three or more independent observations.
Antimicrobial Activity. The centrifuged plant material, its n-BuOH-
soluble portion, and rosmarinic acid were tested for antimicrobial
activity using the broth microdilution method in 96-multiwell microtiter
plates, in duplicate, as reported by Mencherini et al.8 and as recom-
mended by the National Committee for Clinical Laboratory Standards
(NCCLS, 2001),23 using Gram-positive and Gram-negative bacteria, a
yeast, and a mold, all from the American Type Culture Collection
(ATCC). The lowest concentrations of the products at which microbial
growth was inhibited after 24 h (MIC) and at which survival of any
microbial cell was not possible after incubation for 48 (bacteria strains)
and 5 days (yeasts and molds) (MBC) were determined and are reported
in the Supporting Information (Table S2).
Toxicological Methods. Materials and Apparatus. 3-(4,5-Di-
methylthiazolyl-2-yl)-2,5-diphenyltetrazolium bromide (MTT), phos-
phate buffer solution (PBS), and 2-propanol were from Sigma Chemical
Co. (Milan, Italy). The reconstituted human epidermis (RHE) model
(Episkin-SM, Episkin SNC, Lyon, France) of 0.63 cm2 dimensions was
used for skin-irritation testing according to the validated method (ESAC
Statement 2007).18,24
Acknowledgment. The authors are grateful to ResPharma srl for
financial support.
Supporting Information Available: 1D and 2D NMR spectra of
the new compounds 1 (1H, HSQC, and HMBC) and 2 (1H, TOCSY,
HSQC, and HMBC), as well as tables with the DPPH, the antimicrobial,
and the MTT assay results. This information is available free of charge
via the Internet at http://pubs.acs.org.
References and Notes
(1) Carnat, A. P.; Carnat, A.; Fraisse, D.; Lamaison, J. L. Pharm. Acta
HelV. 1998, 72, 301–305.
(2) Adzet, T.; Ponz, R.; Wolf, E.; Schulte, E. Planta Med. 1992, 58, 562–
564.
(3) Mimica-Dukic, N.; Bozin, B.; Sokovic, M.; Simin, N. J. Agric. Food
Chem. 2004, 52, 2485–2489.
(4) Larrondo, J. V.; Agut, M.; Calvo-Torras, M. A. Microbios 1995, 82,
171–172.
For quantification of the IL-1R release in the medium underneath,
an Elisa kit Quantikine DLA-50 (R&D Systems, San Diego, CA) and
a M-200 Infinite microplate autoreader (Tecan) were used. For TEER
(trans-epidermal electric resistance) determination), a Millicel-ERS
(resistance range 0-20 kΩ) volt/ohm-meter (Millipore, Billerica, MA),
able to measure membrane potential and resistance of epithelial cells
grown on microporous membranes, was employed, equipped with a
silver/silver chloride (Ag/AgCl) electrode to measure trans-epithelial
voltage and the resistance of cells.
Skin Irritation on the Episkin Model. The extract of the dried
plant prepared as reported by Mencherini et al.,8 the centrifuged material
from the fresh plant, rosmarinic acid, and compound 1 were dissolved
in PBS to obtain 2% (extract) and 1% (pure compound) concentrations,
respectively. PBS solutions (10 µL) of the test products were applied
in duplicate, using a micropipet, to the surface of epidermis in a single
well treated with 300 µL of its specific maintenance medium. Tissues
were maintained at room temperature for 15 min, washed three times
with PBS (1 mL), and then incubated at 37 °C in a 5% CO2 atmosphere
for 42 h. The culture medium was changed daily. The same volumes
of PBS at pH 7.2 and sodium dodecyl sulfate (SDS 5%) were applied
to control wells. After 42 h of incubation, three complementary end
points were evaluated: (i) cellular parameter, the cytotoxicity by MTT
assay; (ii) molecular parameter, the release of IL-1R; (iii) electrical
resistance parameter, the TEER value. The tests were determined in
triplicate.
(5) To´th, J.; Mrlianova´, M.; Tekei’ova´, D.; Korenˇova´, M. Acta Fac. Pharm.
UniV. Comenianae 2003, L, 139–145.
(6) Lamaison, J. L.; Petitjean-Freytet, C.; Carnat, A. Pharm. Acta HelV.
1991, 66, 185–188.
ˇ
ˇ
(7) Herodezˇ, S. S.; Hadolin, M.; Skerget, M.; Knez, Z. Food Chem. 2003,
80, 275–282.
(8) Mencherini, T.; Picerno, P.; Scesa, C.; Aquino, R. J. Nat. Prod. 2007,
70, 1889–1894.
(9) Lugasi, A.; Hovari, J.; Hagymasi, K.; Jakoczi, I.; Blazovics, A. Acta
Aliment. 2006, 35, 85–97.
(10) Mori, S. Jpn. Kokai Tokkyo Koho JP 10175842, 1998.
(11) Varga, I. S.; Szollosi, R.; Bagyanszki, M. Curr. Top. Biophys. 2000,
24, 219–224.
(12) Mori, S.; Imahori, A. Jpn. Kokai Tokkyo Koho JP 10194915, 1998.
(13) Tsujiura, S.; Kamei, A.; Kawanishi, H.; Fujiwara, N. Jpn. Kokai
Tokkyo Koho JP 06247831, 1994.
(14) Nascimiento, M.; Arruda, A. C.; Arruda, M. S. P.; Mu¨ler, A. H.;
Yoshioka, C. Y. Fitoterapia 1999, 70, 628–629.
(15) Rastrelli, L.; Aquino, R.; Abdo, S.; Proto, M.; De Simone, F.; De
Tommasi, N. J. Agric. Food Chem. 1998, 46, 1797–1804.
(16) Kelley, C. J.; Harruf, R. C.; Carmack, M. J. Org. Chem. 1976, 41,
449–455.
(17) Perrone, A.; Masullo, M.; Bassarello, C.; Hamed, A. I.; Belisario,
M. A.; Pizza, C.; Piacente, S. J. Nat. Prod. 2007, 70, 584–588.
(18) Cotovio, J.; Grandidier, M. H.; Lelie`vre, D.; Roguet, R.; Tinois-
Tessonneaud, E.; Leclaire, J. Altern. Anim. Test. Exp. 2007, 14, 351–
358, and references therein.
(19) Robinson, M. K.; Cohen, C.; de Brugerolle de Fraissinette, A.; Ponec,
M.; Whittle, E.; Fentem, J. H. Food Chem. Toxicol. 2002, 40, 573–
592.
(20) ECVAM TP580. Statement of the Scientific Validity of the Rat Skin
Transcutaneous Electrical Resistence (TER) Test (an in Vitro Test for
Skin CorrosiVity), March 1998 (http://ecvam.jrc.it/).
(21) Picerno, P.; Autore, G.; Marzocco, S.; Meloni, M.; Sanogo, R.; Aquino,
R. P. J. Nat. Prod. 2005, 68, 1610–1614.
(22) Tallarida, R. J.; Murray, R. B. Manual of Pharmacological Calcula-
tions; Springer-Verlag: New York, 1984.
(23) National Committee for Clinical Laboratory Standards. Performance
Standards for Anti-microbial Susceptibility Testing: Eleventh Infor-
mational Supplements. NCCLS: Wayne, PA, document M100-S11,
2001.
(24) Tinois, E.; Tiollier, J.; Gaucherand, M.; Dumas, H.; Tardy, M.;
Thivolet, J. Exp. Cell Res. 1991, 193, 310–319.
Cell Viability. Cell viability was assessed according to Picerno et
al.,21 using the MTT assay, as previously reported.21 Optical density
was read at 570 nm against a blank of 2-propanol.
Release of Interleukin IL-1r. Underlying culture media were
collected at 42 h after product application in triplicate. The release of
IL-1R in the medium of each well was quantified by an Elisa Quantikine
DLA-50 (R&D Systems) kit, as reported in a previous paper.21
Trans-Epidermal Electric Resistance (TEER). The resistance was
measured (in kΩ) on the same tissues in triplicate at time 0 h (nontreated
tissue Rt0) and after the treatment with the test products at 42 h (Rt42).
Owing to the variability within tissues, the measurement performed at
t 0 h has been taken as a basal value and the reference of each single
tissue. Rt0 and Rt42 were calculated as follows: R ) Rsample - Rblank
where Rblank ) R of the culture medium (PBS solution).
Since the resistance is inversely proportional to the area of the tissue,
TEER in kΩ cm2 was calculated, correcting for the area covered by
the tissue, as the product of the resistance found in the experiments
and the area of effective membrane diameter.
NP9003195