Content uploaded by Luana Calabrone
Author content
All content in this area was uploaded by Luana Calabrone on Nov 04, 2015
Content may be subject to copyright.
Food &
Function
PAPER
Cite this: DOI: 10.1039/c5fo00475f
Received 30th April 2015,
Accepted 8th September 2015
DOI: 10.1039/c5fo00475f
www.rsc.org/foodfunction
Anti-inflammatory activity of horseradish
(Armoracia rusticana) root extracts in
LPS-stimulated macrophages
Stefania Marzocco,*
a
Luana Calabrone,
b
Simona Adesso,
a
Marilena Larocca,
b
Silvia Franceschelli,
a
Giuseppina Autore,
a
Giuseppe Martelli
b
and Rocco Rossano
b
Horseradish (Armoracia rusticana) is a perennial crop belonging to the Brassicaceae family. Horseradish
root is used as a condiment due to its extremely pungent flavour, deriving from the high content of
glucosinolates and their breakdown products such as isothiocyanates and other sulfur compounds.
Horseradish also has a long history in ethnomedicine. In this study the anti-inflammatory potential of
three accessions of Armoracia rusticana on lipopolysaccharide from E. coli treated J774A.1 murine macro-
phages was evaluated. Our results demonstrate that Armoracia rusticana reduced nitric oxide, tumor
necrosis factor-αand interleukin-6 release and nitric oxide synthase and cyclooxygenase-2 expression in
macrophages, acting on nuclear transcription factor NF-κB p65 activation. Moreover Armoracia rusticana
reduced reactive oxygen species release and increased heme-oxygenase-1 expression, thus contributing
to the cytoprotective cellular effect during inflammation.
Introduction
Inflammation is a biological response induced by microbial
infection or tissue injury that involves an enormous expendi-
ture of metabolic energy, damage and destruction of host
tissues, even involving risk of sepsis, multiple organ failure
and death. In principle, inflammation is an essential response
to eliminate aggressors but can be deleterious when the initial
reaction is not limited. In these cases, anti-inflammatory com-
pounds are therapeutically useful and are administered to
control the inflammation response.
During inflammation macrophages play a major role in
defending the host; however, their excessive activation may
contribute to extensive tissue damage. Macrophage activation
by bacterial cell wall components such as lipopolysaccharide
from E. coli (LPS), a component of the Gram-negative bacteria,
promotes the synthesis and release of large amounts of
mediators involved in the inflammatory onset such as cyto-
kines (e.g. Tumor Necrosis Factor-α, TNF-α, and Interleukin-6,
IL-6), nitric oxide (NO), pro-inflammatory enzymes (e.g.
cyclooxygenase-2, COX-2) and reactive oxygen species
1
(ROS).
Transient activation of the nuclear transcription factor NF-κB
constitutes an important step in the course of inflammatory
responses and plays a key role in the regulated expression of
several pro-inflammatory mediators including cytokines and
pro-inflammatory enzymes
2
(e.g. nitric oxide synthase, iNOS)
and COX-2. Because of this pivotal role NF-κB is a relevant
target for the pharmacological action of anti-inflammatory
molecule activation in a variety of inflammatory conditions.
2
Also the oxidative response regulates many physiological
responses in human health and, as inflammation, if not pro-
perly regulated it could also lead to a number of deleterious
effects mediating many aspects of inflammatory-induced tissue
damage and dysfunctions.
3,4
Nowadays, the study of oxygen-con-
taining free radicals in humans and their role has been of
growing interest among scientists. Synthetic antioxidants are
known to have free radical inhibition properties in the human
body but these compounds could also be toxic as they present
hazards to the human body as well. The most important and
useful source of such inhibitors, with both anti-inflammatory
and anti-oxidant properties, is the area of natural products.
Horseradish (Armoracia rusticana Gaertn) is a perennial
crop belonging to the Brassicaceae family. Due to its extremely
pungent root, horseradish is used freshly grated as a condi-
ment for meat and fish products or in sauces. The ethno-
medical uses of Armoracia rusticana leaves and roots have a
long history. Horseradish is rich in glucosinolates (GLSs) that
provide the characteristic flavor
5
and aroma as a result of their
breakdown into isothiocyanates (ITCs) and other sulfur com-
pounds. Horseradish, as well as the other members of the
Brassicaceae family, represents a rich source of health-promot-
a
Department of Pharmacy, University of Salerno, Via Giovanni Paolo II 132, 84084
Fisciano (Salerno), Italy. E-mail: smarzocco@unisa.it; Tel: +39 089969250
b
Department of Sciences, University of Basilicata, Viale dell’Ateneo Lucano, 10,
I-85100 Potenza, Italy
This journal is © The Royal Society of Chemistry 2015 Food Funct.
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
View Journal
ing phytochemicals. Their beneficial effects have been princi-
pally attributed to the anticancer properties of GLSs and their
ITC derivatives,
6
complex mixtures of phenolic compounds
possessing antioxidant activity
7,8
and vitamins.
9
Several
studies have been conducted on the allyl-isothiocyanate (AITC)
effects, its main ITC resulting from the hydrolysis of the
corresponding glucosinolate sinigrin by myrosinase enzyme,
highlighting its anticancer properties.
6,10,11
Several authors reported on the AITC action on inflamma-
tory mediator production in macrophage cells.
12–15
Also, some
authors
16
reported on the tumor cell proliferation inhibition
by monogalactosyl diacylglycerides isolated from horseradish
rhizomes.
In this study we investigated the anti-inflammatory effect of
root horseradish extracts, deriving from three accessions, on
LPS-induced inflammation in J774A.1 macrophages. After
identifying the most active accession we investigated, in par-
ticular, its anti-inflammatory effect and the myrosinase contri-
bution to (1) NO production; (2) iNOS and COX-2 expression;
(3) TNF-αand IL-6 release; (4) p65 NF-κB nuclear translocation;
(5) ROS production and (6) the cytoprotective enzyme heme-
oxygenase (HO-1) expression.
Results and discussion
Myrosinase activity and sinigrin content
Plants such as the Brassicaceae, which are always able to syn-
thesize glucosinolates, also possess a β-thioglucoside gluco-
hydrolase commonly known as myrosinase (EC 3.2.3.1).
Myrosinase represents a group of isoenzymes present in all
Brassicaceae species examined and is also found in 14 other
plant families and in the colon microflora.
5
Glucosinolates are
hydrolyzed by myrosinase leading to the generation of several
compounds such as isothiocyanates, thiocyanates, sulfate,
glucose, nitriles and epithioalkanes.
17
Among the three horse-
radish samples analyzed (Fig. 1A), the extract obtained from
the accession GUA showed the highest myrosinase activity
(1.02 ± 0.14 U per mg protein). Regarding sinigrin content, the
predominant glucosinolate in the horseradish plant, in our
study the sinigrin concentration ranged from 52.71 to
66.58 μmol per g DW, with the sample TRI showing the
highest value (Table 1).
GUA, MIN, and TRI do not affect J774A.1 macrophage cell
proliferation
Cell viability of J774A.1 macrophages treated with GUA, MIN,
and TRI, also in the presence of myrosinase, was not signifi-
cantly different from untreated cells. Cell viability, expressed
as a percentage of viability vs. the control, for GUA, MIN, TRI
and TRI plus myrosinase was 94.9 ± 0.9, 95.1 ± 1.2, 94.8 ± 1.9,
94.1 ± 1.5 respectively (200 µg mL
−1
); 96.1 ± 1.3, 97.1 ± 0.8,
94.2 ± 0.9, 96.3 ± 1.3 respectively (150 µg mL
−1
); 95.2 ± 1.5,
96.4 ± 1.4, 96.5 ± 1.7, 96.8 ± 1.6 respectively (100 µg mL
−1
);
97.2 ± 1.3, 97.6 ± 1.8, 95.7 ± 0.8, 95.1 ± 1.9 respectively (50 µg
mL
−1
); and 98.1 ± 1.5, 96.9 ± 1.9, 98.4 ± 1.0, 98.3 ± 1.3 respect-
ively (25 µg mL
−1
;P> 0.05 vs. control). Thus the IC
50
for all the
tested extracts on J774A.1 macrophages was >200 µg mL
−1
. The
absence of antiproliferative activity under our experimental
conditions indicated that all the evaluated parameters to
assess the anti-inflammatory potential of the extracts were
uninfluenced by the cytotoxic activity on J774A.1 macrophages.
GUA, MIN, TRI reduce NO release and iNOS and COX-2
expression in LPS-treated J774A.1 macrophages
LPS induces an inflammatory response that culminates in the
release of pro-inflammatory mediators such as NO, iNOS,
COX-2 and ROS associated with mechanisms aimed at protect-
ing the cell, such as HO-1 enzyme expression. NO is a pleiotro-
pic mediator that acts on a variety of physiological and
pathophysiological processes. This molecule is produced from
the oxidation of L-arginine by the NOS enzyme, which occurs
in the form of two major classes: one is constitutive (including
endothelial and neuronal isoforms) and the other is inducible
(including macrophagic isoform). iNOS may be expressed in
different cell types (e.g. macrophages, smooth muscle cells,
epithelia) by various proinflammatory agents such as LPS.
When macrophages are activated by LPS or IFN-γ, iNOS is sig-
nificantly expressed, and massive amounts of NO are produced
to exert a nonspecific immune response. Induced NO, in
addition to being a “final common mediator”of inflam-
mation, is essential for the up-regulation of the inflammatory
response. Furthermore, NO contributes to tissue damage both
directly via the formation of peroxynitrite, with its associated
toxicity, and indirectly through the amplification of the inflam-
matory response. In our experiments, LPS induced in
J774A.1 macrophages a marked increase in NO release associ-
ated with an increased iNOS expression: GUA, MIN, TRI
(200–50 µg mL
−1
) significantly reduced NO release (200–150 µg
mL
−1
;P< 0.001 vs. LPS alone; Fig. 1B). iNOS is the main NOS
isoform involved in NO during inflammation and, under the
same experimental conditions, GUA, MIN, and TRI (200–50 µg
mL
−1
) significantly reduced iNOS expression too (P< 0.01 vs.
LPS alone; Fig. 1C). An interaction between iNOS and COX
pathways represents an important mechanism for the modu-
lation of the inflammatory response. COX-2 is a well known
pro-inflammatory enzyme triggered by agents such as LPS, it is
involved in macrophage response and its expression was also
influenced by NO.
18
Thus, we evaluated the effect of GUA,
MIN, TRI (200–50 µg mL
−1
) on COX-2 expression. Our data
show that, similarly to NO and iNOS, also COX-2 protein
expression was significantly inhibited by GUA, MIN, TRI,
further contributing to the reduction of LPS-induced inflam-
mation in J774A.1 macrophages (250–50 µg mL
−1
;P< 0.01 vs.
LPS alone; Fig. 1D). Among the tested samples TRI showed the
highest anti-inflammatory potential (Fig. 1).
TRI, in the presence of myrosinase, further reduces NO and
ROS release and iNOS and COX-2 expression in LPS-stimulated
macrophages
Because myrosinase is a key enzyme in ITC production, we
evaluated the effect of TRI on LPS-stimulated J774A.1 macrophages
Paper Food & Function
Food Funct. This journal is © The Royal Society of Chemistry 2015
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
in the presence of myrosinase, using sinigrin as the standard
compound. To investigate if TRI influences NO production
and iNOS and COX-2 expression, even in the presence of myro-
sinase, in another set of experiments J774A.1 macrophage
were treated with TRI (200–25 µg ml
−1
) or sinigrin (30 µM) for
1 h and in combination with LPS (1 μgml
−1
) for 24 h. In some
experiments myrosinase (0.1 U ml
−1
) was added 30 minutes
before TRI or sinigrin treatments. The presence of myrosinase
significantly increased the inhibitory effect of TRI on NO
release (Fig. 2B; P< 0.001 vs. TRI alone), on iNOS (Fig. 2D; P<
0.001 vs. TRI alone) and COX-2 expression (Fig. 3B; P< 0.001
vs. TRI alone) compared to TRI alone.
These results clearly showed that the major anti-inflamma-
tory activity of sample TRI when myrosinase was added could
be due to the increased AITC availability. In this regard, AITC
has been shown to exert an anti-inflammatory action,
19
research has shown that hydrolyzed form of sinigrin sup-
presses the formation of NO in macrophages.
20
Also, regarding
Fig. 1 Effect of GUA, MIN, and TRI on NO release and iNOS and COX-2 expression. Representative root samples of the three horseradish acces-
sions. (Panel A) NO release, evaluated as NO
2
−
(µM), by macrophages J774A.1 stimulated with LPS. (Panel B) Representative western blot of iNOS;
(Panel C) COX-2; (Panel D) densitometric analysis of the concentration dependent effect of GUA, MIN, TRI (200–50 μgml
−1
) on LPS-induced iNOS
and COX-2 in J774A.1 macrophages. Values, means ± S.E.M., are expressed as NO
2
−
(µM) release (% of inhibition vs. LPS), iNOS expression (% of inhi-
bition vs. LPS), and COX-2 expression (% of inhibition vs. LPS). Comparisons were made using one-way analysis of variance, and multiple compari-
sons were made by Bonferroni’s test. ***, ** denote P< 0.001 and P< 0.01 vs. LPS.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2015 Food Funct.
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
the anti-inflammatory activity, in a previous study
8
we have
reported on the presence of flavonoids in the extract obtained
from the horseradish roots. It is assumed that the biological
activity of the TRI extract could be due to synergistic effects
among flavonoids and AITC.
ROS, as well as NO, generation in the inflammatory site is
typically induced as part of a defensive reaction intended to
clear infectious and environmental threats, including
microbial agents and particulate material. Alternatively, ROS
activation could act as a significant and adverse participant in
abnormal inflammatory disease. LPS (1 μgml
−1
) induced sig-
nificant ROS production in macrophages after 24 h. Treatment
with TRI, at all tested concentrations (200–25 μgmL
−1
),
reduced ROS production in macrophages (P< 0.001 vs. LPS
alone; Fig. 3C), thus indicating its antioxidant effects. The
simultaneous presence of TRI and myrosinase further reduced
Table 1 Sinigrin content and myrosinase activity in horseradish roots
Samples
Myrosinase activity
a
(U per mg protein)
Sinigrin content
b
(μmol per g DW)
MIN 0.71 ± 0.11 50.77 ± 3.18
TRI 0.73 ± 0.07 66.58 ± 4.42*
GUA 1.02 ± 0.14* 52.71 ± 3.57
a
Myrosinase activity was determined by the decrease in absorbance of
sinigrin at 227 nm. One unit of activity (U) is defined as the amount of
myrosinase that catalyzes the hydrolysis of 1 μmol of sinigrin per
minute (pH = 6.0; T= 37 °C).
b
Sinigrin was desulfated and determined
using HPLC; results were reported as micromoles of sinigrin per gram
of dry weight. Values are reported as means ± SD of three independent
experiments. *One-way ANOVA was carried out to test for significant
differences and results were considered to be statistically significant at
p< 0.05 (Tukey’s test).
Fig. 2 Effect of TRI and sinigrin, alone and in combination with myrosinase, on NO release and iNOS expression. NO, evaluated as NO
2
−
(µM), by
macrophages J774A.1 stimulated with LPS (Panel A) and stimulated with LPS and myrosinase (0.1 U ml
−1
; Panel B). Representative western blot of
iNOS and densitometric analysis of the concentration dependent effect of TRI (200–25 μgml
−1
) on J774A.1 macrophages stimulated with LPS (Panel
C) and stimulated with LPS and myrosinase (0.1 U ml
−1
; Panel D). Values, means ± S.E.M., are expressed as NO
2
−
(µM) release (% of inhibition vs. LPS),
and iNOS expression (% of inhibition vs. LPS). Comparisons were performed using one-way analysis of variance and multiple comparisons were
made by Bonferroni’s test. ***, ** and * denote P< 0.001, P< 0.01 and P< 0.05 vs. LPS.
###
denotes P< 0.001 vs. TRI alone.
+++
denotes P< 0.001
vs. sinigrin.
Paper Food & Function
Food Funct. This journal is © The Royal Society of Chemistry 2015
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
ROS production in LPS-treated J774A.1 macrophages (P< 0.001
vs. TRI alone; Fig. 3D).
TRI reduces LPS-induced TNF-αand IL-6 in
J774A.1 macrophages
TNF-αand IL-6 are pro-inflammatory cytokines elevated in
sepsis. Moreover TNF-αitself induces iNOS expression and
large amounts of NO production.
21
Although cytokine pro-
duction is necessary to protect against pathogens and promote
tissue repair, excessive release or decreased clearance, or both,
can lead to organ failure and premature death. In our experi-
mental model TRI (200–25 µg ml
−1
) added 1 h before and
simultaneously to LPS significantly reduced TNF-αrelease (P<
0.001 vs. LPS alone; Fig. 4A) and IL-6 (P< 0.01 vs. LPS alone;
Fig. 4C) at all tested concentrations in macrophages. The pres-
ence of myrosinase further reduced TNF-αrelease (P< 0.001
vs. TRI alone; Fig. 4B) and IL-6 (P< 0.05 vs. TRI alone; Fig. 4D)
with respect to TRI alone.
TRI inhibits p65 NF-κB nuclear translocation in LPS-treated
macrophages
LPS is known to activate the pro-inflammatory transcription
factor NF-κB, also regulated via a number of second messen-
gers, including ROS.
22
One of the primary physiological func-
tions of NF-κB is the regulation of immune responses,
including pro-inflammatory enzyme production (e.g. iNOS,
COX-2), antigen presentation, pattern recognition and phago-
cytosis. Following p65 phosphorylation, the free NF-κB dimers
translocate into the nucleus and bind to specific sequences to
regulate the downstream gene expression.
23
So we labelled p65
with a green fluorescence to track the influence of TRI tested
at two medium concentrations of TRI (150–100 μgmL
−1
) and
added 1 h before LPS (1 μgmL
−1
) on NF-κB translocation. As
shown in Fig. 5, nuclear NF-κB p65 was increased after
15 minutes by LPS. NF-κB translocation is reduced by TRI in
J774A.1 treated macrophages compared to LPS alone. The pres-
ence of myrosinase increased the inhibitory effect of TRI on
Fig. 3 Effect of TRI and sinigrin, alone and in combination with myrosinase, on COX-2 expression and ROS release. Representative western blot of
COX-2 and densitometric analysis of the concentration dependent effect of TRI (200–25 μgml
−1
) on LPS-induced COX-2 on J774A.1 macrophages
stimulated with LPS (Panel A) and stimulated with LPS and myrosinase (0.1 U ml
−1
; Panel B). Effect of TRI (200–25 μgml
−1
) on ROS formation, evalu-
ated by means of the probe 2’,7’-dichlorofluorescein-diacetate (H2DCF-DA), on J774A.1 macrophages stimulated with LPS (Panel C) and treated
with LPS and myrosinase (0.1 U ml
−1
; Panel D). Values, means ± S.E.M., are expressed as COX-2 expression (% of inhibition vs. LPS) and as ROS (% of
inhibition vs. LPS). Comparisons were performed using one-way analysis of variance and multiple comparisons were made by Bonferroni’s test. ***,
** denote P< 0.001, P< 0.01 vs. LPS.
###
denotes P< 0.001 vs. TRI alone.
+++
denotes P< 0.001 vs. sinigrin.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2015 Food Funct.
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
p65 translocation. Our evidence indicates that in the presence
of LPS, TRI inhibits the nuclear translocation of the p65 sub-
units, thus reducing NF-κB activity in J774A.1 macrophages
(Fig. 5). These results indicate an activity of TRI in the early
steps of the inflammatory response.
TRI increases HO-1 expression in LPS-treated macrophages
In order to protect themselves against inflammatory and oxi-
dative injury, cells, such as macrophages, up-regulate some
defence mechanisms such as HO-1 expression. HO-1, the rate-
limiting enzyme in heme degradation, catalyzes the oxidation
of heme to generate several biologically active molecules such
as carbon monoxide (CO), biliverdin, and ferrous ions.
24
HO-1
can increase the cellular anti-oxidant status by generating anti-
oxidants such as bilirubin, which can inhibit iNOS protein
expression and suppress NO production.
25
Moreover, CO, a
major product of HO-1 activity, was shown to inhibit COX-2
expression. CO was also shown to inhibit iNOS enzymatic
activity, thus decreasing NO production.
26
HO-1 is normally
expressed at low levels in most tissues/organs except for the
spleen; however, it is highly inducible in response to a variety
of stimuli, such as LPS, to protect cells against oxidative and
inflammatory injury.
24
Considering the beneficial role of HO-1
in controlling various inflammatory mediators, we evaluated
whether its expression was influenced by TRI. HO-1 expression
in J774A.1 macrophages was increased by LPS, and TRI treat-
ment further increased HO-1 (P< 0.05 vs. LPS alone; Fig. 6A)
mostly in the presence of myrosinase (P< 0.05 vs. TRI alone;
Fig. 6B).
Thus during LPS-induced inflammation in macrophages,
TRI on the one hand inhibits pro-inflammatory mediators and
on the other stimulates a cytoprotective response, with respect
to TRI alone.
Experimental
Reagents
Unless stated otherwise, all reagents and compounds were pur-
chased from Sigma Chemical Company (Sigma, Milan, Italy).
Fig. 4 Effect of TRI and sinigrin, alone and in combination with myrosinase, on TNF-αand IL-6 production. TNF-αproduction was measured in the
supernatants of J774A.1 cells treated with TRI (200–25 μgml
−1
) and LPS (Panel A) and treated with LPS and myrosinase (0.1 U ml
−1
; Panel B) by the
ELISA assay. Effect of TRI (200–25 μgml
−1
) on LPS-induced IL-6 production in J774A.1 macrophages. IL-6 production was measured in the super-
natants of J774A.1 cells treated with TRI (200–25 μgml
−1
) and LPS (Panel C) and treated with LPS and myrosinase (0.1 U ml
−1
; Panel D) by the ELISA
assay. Values, means ± S.E.M., are expressed as TNF-α(% of inhibition vs. LPS) and IL-6 (% of inhibition vs. LPS). Data were analyzed by ANOVA test,
and multiple comparisons were made by Bonferroni’s test. *** and ** denote P< 0.001 and P< 0.01 vs. LPS.
###
,
##
and
#
denote P< 0.001, P< 0.01
and P< 0.05 vs. TRI alone.
+++
denotes P< 0.001 vs. sinigrin.
Paper Food & Function
Food Funct. This journal is © The Royal Society of Chemistry 2015
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
Plant material
Three different accessions of horseradish (Armoracia rusticana,
Gaertn) samples named GUA (Acc1), MIN (Acc2), and TRI
(Acc3) were harvested in December 2012 from a private crop
grown in Southern Italy (Accettura –latitude 40° 29′N, longi-
tude 16° 9′E–Basilicata, Italy).
For each sample, roots (35–40 cm long and 2.5–3.0 cm in
diameter) were collected from the middle of the plants,
cleaned with distilled water, dried with paper towels and
quickly frozen in liquid nitrogen. Samples were then lyophi-
lized, ground into a fine powder and stored at −70 °C.
Myrosinase activity
Aliquots of root powder were dissolved (1 : 5, w/v) in a cold
2 mM DTT, 5 mM EDTA, phosphate buffer (pH 6.0) for 1 h
(vortexing 30 s every 5 min), the resuspended samples were
centrifuged at 12 000gfor 15 min at 4 °C and the supernatants
were assayed for myrosinase activity. Enzyme activity was deter-
mined spectrophotometrically by the decrease in absorbance
at 227 nm deriving from the hydrolysis of sinigrin.
27
Briefly,
100 µL of the enzymatic extract was added to the reaction
mixture (pre-equilibrate at 37 °C) consisting of 2.3 mL de-
ionised water, 400 μL of 100 mM phosphate buffer pH 6.0, 20 μL
of 100 mM ascorbic acid and 10 μL of 40 mM sinigrin. The
decrease in absorbance as a result of sinigrin breakdown was
plotted at 227 nm over 4 min. Activity was determined from
the linear slope representing the disappearance of sinigrin in
time. One unit of activity (U) is defined as the amount of myro-
sinase that catalyzes the hydrolysis of 1 μmol of sinigrin per
minute at 37 °C and pH 6.0. Specific activity was expressed as
units per milligram of protein. Protein content of the extracts
was determined according to the method of Bradford (1976),
28
using the Bio-Rad reagent and bovine serum albumin as the
standard protein.
Sinigrin determination
Sinigrin was extracted with hot methanol from freeze-dried
samples followed by an enzymatic desulfation. Briefly, for each
sample, 500 mg of freeze-dried samples was added to 0.3 mL
of 2 μM benzylglucosinolate (internal standard), extracted in
4 mL of 100% methanol at 70 °C and heated for 20 min on a
heating block. The extracting procedure was repeated twice
and the supernatants were combined. After centrifugation
(10 000gfor 8 min) 1 mL of the supernatant was added to a
Fig. 5 Effect of TRI and sinigrin, alone and in combination with myrosinase, on NF-κB activation. p65 NF-κB nuclear translocation in
J774A.1 macrophages was induced by LPS (Panel A). TRI alone (A) and in the presence of myrosinase (0.1 U ml
−1
; B) reduced p65 NF-κB nuclear
translocation. Nuclear translocation of NF-κB p65 subunit was detected using the immunofluorescence assay by confocal microscopy. Scale bar,
10 µm. Blue and green fluorescence indicates localization of the nucleus (DAPI) and p65 NF-κB respectively. Analysis was performed by confocal
laser scanning microscopy.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2015 Food Funct.
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
Sephadex A25 column and desulfated (for 16 h) using 0.5 mL
of aryl sulfatase (20 U mL
−1
; Sigma, St. Louis, MO). Desulfo-
sinigrin was eluted with 2 mL of deionised water, filtered
through a 0.2 μm filter and separated with an Alliance Waters
2695 HPLC (Waters, Milford, MA) equipped with a UV detector
set at 229 nm and a Spherisorb ODS2 column (150 × 4.6 mm,
3μm, Waters). Elution was performed by a linear gradient
mobile phase from 100% A (0.5% trifluoroacetic acid in water)
to 30% B (acetonitrile) in 30 min, at 1 mL min
−1
. Desulfosini-
grin was identified and quantified by comparison of the HPLC
retention time with that of a sinigrin standard (Sigma) after
on-column sulfatase treatment as described above. Samples
were independently analyzed in triplicate, and the results were
reported as micromoles of sinigrin per gram of dry weight
(μmol per g DW).
Preparation of extracts
For each sample ten grams of root powders were extracted for
12 h at 30 °C in a conical flask at 150 rpm with 100 mL of
methanol. The extracts were recovered after centrifugation
(18 000gfor 4 min at 20 °C), filtered through filter paper and
freed of solvent in a rotary vacuum evaporator. The dried crude
extracts were weighed to calculate the yield and stored at −70 °C.
Cell culture
J774A.1 murine monocyte macrophage cell line (American
Type Culture Collection, Rockville, MD) was grown adherent to
Petri dishes with Dulbecco’s modified Eagle’s medium
(DMEM) supplemented with 10% foetal calf serum (FCS),
25 mM HEPES, 2 mM glutamine, 100 U mL
−1
penicillin and
100 mg mL
−1
streptomycin at 37 °C under a 5% CO
2
atmosphere.
Antiproliferative activity
J774A.1 macrophages (5 × 10
4
per well) were plated on 96-well
plates and allowed to adhere for 4 h. The medium was then
replaced with fresh medium alone or containing serial
dilutions of GUA, MIN, TRI (200–50 μgml
−1
), also in the pres-
ence of myrosinase, and cells were incubated for 24 h. Cell via-
bility was assessed using the MTT assay as previously
reported.
29,30
Briefly, 25 µL of MTT (5 mg mL
−1
) were added
and cells were incubated for an additional 3 h. Thereafter,
cells were lysed and the dark blue crystals solubilised with
100 mL of a solution containing 50% (v/v) N,N-dimethyl-
formamide, 20% (w/v) SDS with an adjusted pH of 4.5. The
optical density (OD) of each well was measured with a micro-
plate spectrophotometer (Titertek Multiskan MCC/340-DASIT)
equipped with a 620 nm filter. Macrophage viability in
response to treatment with the extracts was calculated as
% dead cells = 100 × (OD treated/OD control) × 100], as pre-
viously reported.
31
Cell treatment for the anti-inflammatory activity evaluation
Macrophages J774A.1 were plated in P60 (1.8 × 10
6
) and
allowed to adhere for 4 h. Thereafter, the medium was
replaced with fresh medium alone or containing serial
dilutions of GUA, MIN, TRI (200–50 μgml
−1
) for 1 h and then
co-exposed to LPS (1 μgml
−1
) for a further 18–24 h for NO
detection and iNOS and COX-2 expression.
In another set of experiments TRI (200–25 μgml
−1
), activity
was compared to sinigrin (30 µM) also in the presence of myro-
sinase (0.1 U ml
−1
) added 30 minutes before TRI or sinigrin
treatment.
Nitrite determination and western blot analysis for iNOS,
COX-2 and HO-1 expression
NO generation was measured as nitrite (NO
2
−
), the index of
NO released by cells, in the culture medium 24 h after LPS
Fig. 6 Effect of TRI and sinigrin, alone and in combination with myrosi-
nase, on HO-1 expression. Representative western blot of HO-1 enzyme
expression and densitometric analysis of the concentration dependent
effect of TRI (200–25 μgml
−1
) on J774A.1 macrophages stimulated with
LPS (Panel A) and stimulated with LPS and myrosinase (0.1 U ml
−1
; Panel
B). Values, means ± S.E.M., are expressed as HO-1 expression (% of inhi-
bition vs. LPS). Comparisons were made using one-way analysis of vari-
ance and multiple comparisons were made by Bonferroni’s test. ***, **
and * denote P< 0.001, P< 0.01 and P< 0.05 vs. LPS.
###
,
##
and
#
denote P< 0.001, P< 0.01 and P< 0.05 vs. TRI alone.
+++
denotes P<
0.001 vs. sinigrin.
Paper Food & Function
Food Funct. This journal is © The Royal Society of Chemistry 2015
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
stimulation by Griess reaction, as previously reported.
32
Briefly, 100 mL of cell culture medium were mixed with
100 mL of Griess reagent –equal volumes of 1% (w/v) sulpha-
nilamide in 5% (v/v) phosphoric acid and 0.1% (w/v) naphthyl-
ethylenediamine-hydrogen chloride and incubated at room
temperature for 10 min, and then the absorbance was
measured at 550 nm on a microplate reader Titertek (Dasit,
Cornaredo, Milan, Italy). The amount of NO
2
−
,asμM concen-
tration, in the samples was calculated by using a sodium NO
2
−
standard curve. iNOS and COX-2 protein expression in
J774A.1 macrophages was assessed by western blot analysis. As
previously reported,
33
after NO
2
−
determination, cells were
scraped off, washed with ice-cold phosphate-buffered saline
(PBS), and centrifuged at 5000gfor 10 min at 4 °C. The cell
pellet was lysed in a buffer containing 20 mM Tris HCl (pH
7.5), 1 mM sodium orthovanadate, 1 mM phenylmethyl-
sulfonyl fluoride, 10 μg per ml leupeptin, 10 mM sodium fluor-
ide, 150 mM sodium chloride, 10 mg per ml trypsin inhibitor,
and 1% Tween-20. Protein concentration was estimated by
using the Bio-Rad protein assay using bovine serum albumin
as the standard. Equal amounts of protein (50 μg) were dis-
solved in Laemmli’s sample buffer, boiled, and run on a SDS
polyacrylamide gel electrophoresis (SDS-PAGE) minigel (8%
polyacrylamide) and then transferred to a hybond polyvinyli-
dene difluoride membrane for 40 min at 5 mA cm
2
. Mem-
branes were blocked for 40 min in PBS and 5% (w/v) nonfat
milk and subsequently probed overnight at 4 °C with mouse
monoclonal anti-iNOS, anti-COX-2 antibody (BD Laboratories)
and anti-HO-1 antibody (Santa Cruz Biotechnology, Inc.) in
PBS, 5% w/v nonfat milk, and 0.1% Tween-20. Blots were then
incubated with horseradish-peroxidase conjugated goat anti-
mouse immunoglobulin (I
g
)G (1 : 5000) for 1 h at room temp-
erature. Immunoreactive bands were visualized using the
electro-chemiluminescence assay (ECL) detection system
according to the manufacturer’s instructions and exposed to
Kodak X-Omat films. The protein bands of iNOS, COX-2, HO-1
and tubulin on X Omat films were quantified by scanning
densitometry (Imaging Densitometer GS-700, Bio-Rad, U.S.A.).
Data are normalized with tubulin expression, used as the refer-
ence protein, and are expressed as arbitrary densitometric
units as previously reported.
34
TNF-αand IL-6 determination
TNF-αand IL-6 concentrations were assessed by an Enzyme-
Linked Immuno Sorbent Assay (ELISA) using a commercial kit
for murine TNF-αor IL-6 according to the manufacturer’s
instructions (e-Biosciences, CA, USA) in J774A.1 culture
medium stimulated for 18 h with TRI (200–25 μgml
−1
) and/or
sinigrin (30 µM) for 1 h and in combination with LPS (1 μg
ml
−1
) for 24 h and, in other experiments, myrosinase (0.1 U
ml
−1
) was added 30 minutes before treatment.
Immunofluorescence analysis with confocal microscopy
For the immunofluorescence assay, J774A.1 cells (3 × 10
5
per
well) were seeded on coverslips in 12 well plates and treated
with TRI using two medium concentrations (150–100 µg ml
−1
)
for 1 h and then simultaneously with LPS (1 µg ml
−1
) for
20 min for NF-κB. Then the cells were fixed with 4% para-
formaldehyde in PBS for 15 min and permeabilized with 0.1%
saponin in PBS for 15 min. After blocking with BSA and PBS
for 1 h, cells were incubated with rabbit anti-phospho p65 anti-
body (Santa Cruz Biotechnologies) for 1 h at room tempera-
ture. The slides were then washed three times with PBS and
fluorescein conjugated secondary antibody (FITC) was added
for 1 h, and DAPI was used for counterstaining of nuclei.
Coverslips were finally mounted in a mounting medium and
fluorescent images were taken under a Laser Confocal Micro-
scope (Leica TCS SP5) as previously reported.
35
Measurement of intracellular ROS
ROS formation was evaluated by means of the probe 2′,7′-
dichlorofluorescein-diacetate (H
2
DCF-DA). H
2
DCF-DA is a
non-fluorescent permeant molecule that passively diffuses into
cells, where the acetates are cleaved by intracellular esterases
to form H
2
DCF and thereby trap it within the cell. In the pres-
ence of intracellular ROS, H
2
DCF is rapidly oxidized to the
highly fluorescent 2′,7′-dichlorofluorescein (DCF). Briefly,
J774A.1 cells were plated at a density of 3 × 10
5
cells per well
into 24-well plates. Cells were allowed to grow for 24 h; there-
after, the medium was replaced with fresh medium and cells
were stimulated with TRI (200–25 μgml
−1
) and/or sinigrin (30
µM) for 1 h and in combination with LPS (1 μgml
−1
) for 24 h.
In other experiments myrosinase (0.1 U ml
−1
) was added
30 minutes before treatment. Cells were then collected,
washed twice with phosphate buffer saline (PBS) and then
incubated in PBS containing H
2
DCF-DA (10 µM) at 37 °C. After
15 minutes, cell fluorescence was evaluated using fluo-
rescence-activated cell sorting (FACScan; Becton Dickinson)
and elaborated with Cell Quest software as previously
reported.
36
Data analysis
Data are reported as the mean ± standard error of the mean
(S.E.M.) values of at least three independent experiments, each
performed in triplicate. Statistical analysis was performed by
the analysis of variance test, and multiple comparisons were
made by Bonferroni’s test. A P-value less than 0.05 was con-
sidered significant.
Conclusions
In the last few years, the search for natural products from folk
medicine has been widely investigated for their potential anti-
inflammatory activity. The ethno-medical uses of Armoracia
rusticana leaves and roots have a long history; juice extracted
from horseradish has been used to treat sinusitis, urinary and
respiratory tract infections. In this study we have demonstrated
the anti-inflammatory potential of three accessions of Armoracia
rusticana on LPS-treated J774A.1 murine macrophages. Our
results indicate that horseradish reduced NO, iNOS and COX-2
expressions, in particular for the TRI extract. This result was
Food & Function Paper
This journal is © The Royal Society of Chemistry 2015 Food Funct.
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
probably due to the highest content of sinigrin than the other
accessions that was converted to AITC.
TRI extract was also able to reduce ROS and cytokine
release acting on NF-κB p65 activation, a pivotal transcription
factor in chronic inflammatory diseases.
37
Moreover Armoracia
rusticana TRI extract increased HO-1 expression, thus contribut-
ing to the cytoprotective cellular effect during inflammation.
TRI anti-inflammatory effects were further enhanced by the
presence of myrosinase presumably due to the increase of allyl-
isothiocyanate (AITC) concentration, derived from the action of
myrosinase on sinigrin, the predominant glucosinolate in
horseradish.
5
These results indicate that horseradish has inhibitory
effects on the production of pro-inflammatory mediators, such
as many anti-inflammatory drugs, and are consistent with
several studies on the biological effects of herbal medication
extracts on LPS-induced inflammation in macrophages.
38–41
Conflict of interest
The authors declare no competing financial interest.
Acknowledgements
This study was supported by a grant from University of Basili-
cata (Ricerca Interesse Locale, 2011) and by a grant from Basili-
cata Region (Mis 214 Azione 5 “Agrobiodiversità: Biodiversità
di specie orticole ed areali lucani: da patrimonio a strumento
di sviluppo”). The authors are grateful to Mr Paolo Giannelli
for editing the manuscript.
References
1 J. MacMicking, Q. W. Xie and C. Nathan, Annu. Rev.
Immunol., 1997, 15, 323–350.
2 P. A. Baeuerle, Cell, 1998, 95, 729–731.
3 J. Pike and R. K. Chandra, Int. J. Vitam. Nutr. Res., 1995, 65,
117–121.
4 A. Popolo, G. Autore, A. Pinto and S. Marzocco, Free Radical
Res., 2013, 47, 346–356.
5 X. Li and M. M. Kushad, J. Agric. Food Chem., 2004, 52,
6950–6955.
6 Y. Zhang, Mol. Nutr. Food Res., 2010, 54, 127–135.
7 A. Majewska, B. Bałasińska and B. Dąbrowska, Folia Hort.,
2004, 16,15–22.
8 L. Calabrone, M. Larocca, S. Marzocco, G. Martelli and
R. Rossano, Food Nutr. Sci., 2015, 6,64–74.
9 G. Mazza, Can. Inst. Food Sci. Technol. J., 1984, 17,18–23.
10 L. Tang, G. Li, L. Song and Y. Zhang, Anti-Cancer Drugs,
2006, 17, 297–305.
11 A.Bhattacharya,Y.Li,K.L.Wade,J.D.Paonessa,J.W.Fahey
and Y. Zhang, Carcinogenesis, 2010, 31,2105–2110.
12 K. Ippoushi, H. Itou, K. Azuma and H. Higashio, Life Sci.,
2002, 71, 411–419.
13 P. Thejass and G. Kuttan, Nitric Oxide, 2007, 16, 247–257.
14 A. E. Wagner, C. Boesch-Saadatmandi, J. Dose,
G. Schultheiss and G. Rimbach, J. Cell Mol. Med., 2012, 16,
836–843.
15 S. J. Kim, H. J. Park, H. J. Shin, D. H. Shon, H. Kim do and
H. S. Youn, Int. Immunopharmacol., 2012, 13, 403–407.
16 M. J. Weil, Y. Zhang and M. G. Nair, J. Agric. Food Chem.,
2005, 53, 1440–1444.
17 A. M. Bones and J. T. Rossiter, Physiol. Plant., 1996, 97,
194–208.
18 N. Ahmad, L. C. Chen, M. A. Gordon, J. D. Laskin and
D. L. Laskin, J. Leukocyte Biol., 2002, 71, 1005–1011.
19 H. M. Woo, J. H. Kang, T. Kawada, H. Yoo, M. K. Sung and
R. Yu, Life Sci., 2007, 80, 926–931.
20 D. K. Patel, K. Patel, M. Gadewar and V. A. Tahilyani,
AsianPac. J. Trop. Biomed., 2012, S446–S448.
21 C. Thiemermann, C. C. Wu, C. Szabó, M. Perretti and
J. R. Vane, Br. J. Pharmacol., 1993, 110, 177–182.
22 D. M. Brown, K. Donaldson, P. J. Borm, R. P. Schins,
M. Dehnhardt, P. Gilmour, L. A. Jimenez and V. Stone,
Am. J. Physiol.: Lung Cell. Mol. Physiol., 2004, 286, 344–353.
23 P. P. Tak and G. S. Firestein, J. Clin. Invest., 2001, 107,7–11.
24 M. L. Wu, Y. C. Ho, C. Y. Lin and S. F. Yet,
Am. J. Cardiovasc. Dis., 2011, 1, 150–158.
25 W. W. Wang, D. L. Smith and S. D. Zucker, Hepatology,
2004, 40, 424–433.
26 C. Thorup, C. L. Jones, S. S. Gross, L. C. Moore and
M. S. Goligorsky, Am. J. Physiol., 1999, 277, F882–F889.
27 S. Palmieri, R. Leoni and R. Iori, Anal. Biochem., 1982, 123,
320–324.
28 M. Bradford, Anal. Biochem., 1976, 72, 248–254.
29 G. Autore, S. Marzocco, C. Formisano, M. Bruno,
S. Rosselli, M. B. Jemia and F. Senatore, Molecules, 2015, 20
(1), 571–1578.
30 G. Cioffi, R. Sanogo, A. Vassallo, F. Dal Piaz, G. Autore,
S. Marzocco and N. De Tommasi, J. Nat. Prod., 2006, 69(4),
625–635.
31 M. Ben Jemia, M. E. Kchouk, F. Senatore, G. Autore,
S. Marzocco, V. De Feo and M. Bruno, Chem. Cent. J., 2013,
7(1), 47.
32 G. Bianco, R. Russo, S. Marzocco, S. Velotto, G. Autore and
L. Severino, Toxicon, 2012, 59, 644–650.
33 E. Sommella, G. Pepe, F. Pagano, G. C. Tenore,
S. Marzocco, M. Manfra, G. Calabrese, R. P. Aquino and
P. Campiglia, J. Funct. Foods, 2014, 7, 641–649.
34 F. Cianchi, S. Cuzzocrea, M. C. Vinci, L. Messerini,
C. E. Comin, G. Navarra, G. Perigli, T. Centorrino,
S. Marzocco, E. Lenzi, N. Battisti, G. Trallori and E. Masini,
Dig Liver Dis., 2010, 42(1), 20–27.
35 M. Del Regno, S. Adesso, A. Popolo, A. Quaroni, G. Autore,
L. Severino and S. Marzocco, Toxicol. Appl. Pharmacol.,
2015, 285(2), 118–127.
36 S. Adesso, A. Popolo, G. Bianco, R. Sorrentino, A. Pinto,
G. Autore and S. Marzocco, PLoS One, 2013, 8, e76778.
37 P. J. Barnes and M. Karin, N. Engl. J. Med., 1997, 336, 1066–
1071.
Paper Food & Function
Food Funct. This journal is © The Royal Society of Chemistry 2015
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
38 Y. C. Oh, Y. H. Jeong, W. K. Cho, J. H. Ha, M. J. Gu and
J. Y. Ma, Int. J. Mol. Sci., 2015, 16(1), 1232–1251.
39 Y. C. Oh, Y. H. Jeong, J. H. Ha, W. K. Cho and J. Y. Ma,
BMC Complementary Altern. Med., 2014, 14, 242.
40 D. J. Kwon, S. M. Ju, G. S. Youn, S. Y. Choi and J. Park, Food
Chem. Toxicol., 2013, 58, 479–486.
41 A. Braca, F. Dal Piaz, S. Marzocco, G. Autore, A. Vassallo and
N. De Tommasi, Curr. Drug Targets, 2011, 12(3), 302–321.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2015 Food Funct.
Published on 15 September 2015. Downloaded by Universita della Basilicata on 06/10/2015 11:34:29.
View Article Online
