Exploratory Studies on the in Vitro Anti-inflammatory Potential of Two Herbal Teas (Annona muricata L. and Jasminum grandiflorum L.), and Relation with Their Phenolic Composition

Article abstract of DOI:10.1002/cbdv.201700002

The need of new anti-inflammatory drugs has led to the search for safer and more potent molecules in distinct sources, such as natural products. This work aimed to explore the anti-inflammatory potential of aqueous extracts from two herbal teas (Annona muricata L. and Jasminum grandiflorum L.) in RAW 264.7 macrophages cells and in cell-free assays. Furthermore, the phenolic composition of both extracts and of their hydrolysates was characterized by HPLC-DAD, in order to establish possible relationships with the biological activity. In a general way, A.?muricata displayed a stronger capacity to inhibit nitric oxide (NO) production and the activity of phospholipase A₂ (PLA₂), displaying an IC₅₀ value of 142?μg/ml against this enzyme. A deeper look at phenolic compounds revealed that aglycones had more capacity to inhibit NO and PLA₂ than their corresponding glycosides, quercetin being clearly the most potent one (IC₅₀?=?7.47 and 1.36?μm, respectively). In addition, 5-O-caffeoylquinic acid, at 1.56?μm, could also inhibit PLA₂ (ca. 35%). Our findings suggest that the consumption of both herbal teas may be a preventive approach to inflammatory disorders.

Full text of DOI:10.1002/cbdv.201700002

Received Date: 04-Jan-2017
Revised Date: 11-Feb-2017
Accepted Date: 24-Feb-2017
Article Type: Full Paper
Exploratory Studies on the in Vitro Anti-inflammatory Potential of Two
Herbal Teas (Annona muricata L. and Jasminum grandiflorum L.), and
Relation with Their Phenolic Composition
Andreia P. Oliveira, Ivone Sá, David M. Pereira, Rui F. Gonçalves, Paula B. Andrade and Patrícia
Valentão*
REQUIMTE/LAQV, Laboratório de Farmacognosia, Departamento de Química, Faculdade de
Farmácia, Universidade do Porto, R. Jorge Viterbo Ferreira, n.º 228, 4050-313 Porto, Portugal, e-
mail:valentao@ff.up.pt
The need of new anti-inflammatory drugs has led to the search for safer and more potent molecules
in distinct sources, such as natural products. This work aimed to explore the anti-inflammatory
potential of aqueous extracts from two herbal teas (Annona muricata L. and Jasminum grandiflorum
L.) in RAW 264.7 macrophages cells and in cell-free assays. Furthermore, the phenolic composition
of both extracts and of their hydrolysates was characterized by HPLC-DAD, in order to establish
possible relationships with the biological activity. In a general way, A. muricata displayed a stronger
capacity to inhibit nitric oxide (NO) production and the activity of phospholipase A
an IC50 of 142 µg/mL against this enzyme. A deeper look at phenolic compounds revealed that
aglycones had more capacity to inhibit NO and PLA than their corresponding glycosides, quercetin
2
(PLA ), displaying
2
2
being clearly the most potent one (IC50 = 7.47 and 1.36 µM, respectively). In addition, 5-O-
caffeoylquinic acid, at 1.56 µM, could also inhibit PLA (ca. 35%). Our findings suggest that the
2
consumption of both herbal teas may be a preventive approach to inflammatory disorders.
Keywords: Anti-inflammatory activity • RAW 264.7 • Phospholipase A • Annona muricata •
2
Jasminum grandiflorum
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/cbdv.201700002
This article is protected by copyright. All rights reserved.

Introduction
Inflammatory diseases are among the most prevalent pathological conditions worldwide, frequently
resulting in some degree of disability or loss of life quality. For example, the years lived with
disability for musculoskeletal disorders, in which inflammation plays a pivotal role, has increased
[
1]
over 60% in the period 1990-2013. Several signalling pathways that regulate the expression of both
pro- and anti-inflammatory mediators, in resident tissue cells and leukocytes recruited from blood,
[
2]
are activated in the inflammatory response. During the inflammatory process, pro-inflammatory
cytokines, such as interleukin-1/6 and tumour necrosis factor α are released, as well as other
[
3, 4]
inflammation mediators like nitric oxide (NO).
In the continuous search for new drugs, natural products are to be referred, playing an important
role in the treatment of inflammatory conditions. During the last decades, some works have
demonstrated that the modulation of inflammatory response by bioactive natural compounds plays
an important role in the prevention and treatment of many diseases and conditions, an inverse
relationship between natural products intake and prevalence of inflammatory conditions being
[
5-7]
observed. Among natural products, phenolic compounds, such as flavonoids, are of particular
interest given the several biological activities that have been attributed to them, including anti-
[
5-13]
inflammatory.
Medicinal herbs are widely consumed in the Mediterranean countries, mainly as herbal teas. Their
[
14- 17]
phytochemicals are known to possess several biological activities.
Annona muricata L. and
Jasminum grandiflorum L. can be highlighted due to their medicinal application. A. muricata
commonly known as “graviola” is a species usually found in the tropical regions of Central America
and South America, West Africa and Southeast Asia. The aqueous extract of A. muricata leaves is
already known to possess several biological properties, including hepatoprotective, gastroprotective,
[
18]
anti-tumorigenic, hypoglycemic, hypotensive, antioxidant, antibacterial and insecticidal ones.
J.
grandiflorum, also known as “jasmim”, is a native species from Asia, largely recognized to possess
spasmolytic, anti-inflammatory, antimicrobial, antiulcer, antioxidant, cytoprotective,
[
19]
chemopreventive and anti-acne activities. Aqueous extracts from J. grandiflorum flowers are
[
19]
useful for skin diseases, ulcers and psychiatric disorders. The anti-inflammatory potential of the
ethanolic and hydroalcoholic extracts from A. muricata leaves in animal models has already been
[
20, 21]
described,
as well as the anti-inflammatory capacity of the methanolic extract of J.
[
22]
grandiflorum leaves on rats peritoneal macrophages. Nevertheless, as far as we know, there is no
study describing the anti-inflammatory potential of the aqueous extracts from either species, which
are of special interest as they mimic the most common form of ingestion, namely as herbal tea.
In this study we assessed the toxicity and the anti-inflammatory potential of herbal teas prepared
from A. muricata and J. grandiflorum in lipopolysaccharide (LPS)-stimulated RAW 264.7 cells, by
determining NO levels, and also evaluated their capacity to inhibit phospholipase A
2
(PLA ) in a cell-
2
free assay. Furthermore, as phenolic compounds are plant metabolites with well-described anti-
inflammatory activity, the phenolic compounds existing in both herbal teas were identified and
This article is protected by copyright. All rights reserved.

quantified by HPLC-DAD, in order to establish possible relationships with the biological properties. In
addition, given the fact that genines are more active than their respective heterosides, hydrolysed
herbal teas were also analysed, to try to find the real influence of the different compounds in the
activity of the aqueous extracts.
Results and Discussion
Phenolic compounds
Identification and quantification of phenolic compounds in A. muricata and J. grandiflorum aqueous
lyophilized extracts/acid hydrolysates were carried out in order to establish possible relationships
with the anti-inflammatory potential.
A. muricata
The aqueous lyophilized extract was characterized by the presence of one hydroxycinnamic acid [5-
O-caffeoylquinic acid (1)] and four flavonols [quercetin-3-O-galactoside (2), quercetin-3-O-glucoside
(
3), quercetin-3-O-rutinoside (4) and kaempferol-3-O-rutinoside (5)] (Fig. 1) already described in the
[
23]
aqueous extract of this species. In the corresponding acid hydrolysate, the hydroxycinnamic acid
(
(
1) and the aglycones resulting from the breakdown of the glycosides, quercetin (6) and kaempferol
7), were identified (Fig. 1).
A
0,3
0.3
2
3
0
0
,
.2
2
1
4
0,1
0.1
5
0
0
00
1
11
0
00
2
22
0
0
3
30
0
0
4
44
0
0
5
50
0
60
0
Time (min.)
0.3
0,3
B
0
0
.
,
2
2
6
7
0
0. 1, 1
1
0
0
0
10
10
20
20
30
30
40
40
50
50
60
60
0
Time (min.)
Figure 1. HPLC-DAD phenolic profile of A. muricata leaves at 350 nm. (A) Aqueous extract and (B) acid hydrolysate. Compounds´ identity as in Table 1.
This article is protected by copyright. All rights reserved.

Calibration curves were obtained with standard solutions at five different concentrations,
selected as representative of the range of compounds’ concentrations in the aqueous extract and
acid hydrolysate of A. muricata leaves. The calibration plots showed a good correlation, as indicated
by correlation coefficient values higher than 0.99 for all compounds (Table 1). The LOD and LOQ for
the analysed phenolic compounds are shown in Table 1.
In quantitative terms, 5-O-caffeoylquinic acid was clearly the main compound of the aqueous
extract, corresponding to ca. 51.5 % of total determined phenolics, followed by quercetin-3-O-
rutinoside (ca. 32.1%) (Table 1). Differently, the acid hydrolysate was mainly composed by quercetin
and 5-O-caffeoylquinic acid, which represented ca. 56.7% and 35.0% of total phenolic content,
respectively (Table 1).
a
Table 1. Linearity, detection and quantification limits, and phenolic composition of A. muricata.
LOD
LOQ
mg/g extract
2
Compounds
Regression equation
R
(
μg/mL)
(µg/mL)
AE
H
8
6
1
2
3
4
5
6
7
5-O-CQA
Q-3-O-galact
Q-3-O-gluc
Q-3-O-rut
K-3-O-rut
Q
y= 9 x 10 x – 2 x 10
0.9999
0.9997
0.9999
0.9985
0.9997
0.9995
0.9998
1.22
3.70
2.69
5.88
0.52
2.06
3.94
3.71
11.20
8.16
3.30 ± 0.14
0.69 ± 0.05
0.11 ± 0.01
2.06 ± 0.02
0.25 ± 0.00
nd
0.42 ± 0.04
8
y= 7 x 10 x + 727080
nd
nd
8
y= 9 x 10 x + 885697
8
6
6
6
6
y= 5 x 10 x + 1 x 10
17.83
1.58
nd
8
y= 6 x 10 x – 1 x 10
nd
9
y= 1 x 10 x + 2 x 10
6.26
0.68 ± 0.06
0.10 ± 0.00
1.20 ± 0.10
8
K
y= 8 x 10 x – 2 x 10
11.95
nd
Total
6.41± 0.22
a
Results are expressed as mean ± standard deviation of three determinations and correspond to mg/g of dry weight. LOD, Limit of
detection; LOQ, limit of quantification; AE, aqueous extract; H, acid hydrolysate; CQA, caffeoylquinic acid; Q, quercetin; galact,
galactoside; gluc, glucoside; rut, rutinoside; K, kaempferol; nd, not detected.
J. grandiflorum
The aqueous extract of J. grandiflorum flower buds contained two hydroxycinnamic acids [5-
dihydrocaffeoylquinic (8) and 5-dihydromethoxycaffeoylquinic acids (9)], two quercetin derivatives
[quercetin-3-O-(2,6-di-rhamnosyl)glucoside (10) and quercetin-3-O-rutinoside (4)] and one
kaempferol glycoside [kaempferol-3-O-(2,6-di-rhamnosyl)glucoside (11)] (Fig. 2), which had already
[
24]
been identified in this species. The two phenolic acids (8 and 9), quercetin (6) and kaempferol (7)
were detected in the acid hydrolysate (Fig. 2).
This article is protected by copyright. All rights reserved.

2
1
A
1
0
8
11
9
4
0
2
0
5
10
15
20
25
30
35
40
Time (min.)
B
6
1
0
7
8
9
0
5
10
15
20
25
30
35
40
Time (min.)
Figure 2. HPLC-DAD phenolic profile of the J. grandiflorum flower buds at 350 nm. (A) Aqueous
extract and (B) acid hydrolysate. Compounds´ identity as in Table 2.
The calibration plots showed a good correlation, as indicated by correlation coefficient values
higher than 0.99 for all phenolic compounds (Table 2). LOD and LOQ values obtained for the
analysed compounds are displayed in Table 2.
The quantity of phenolic compounds found in both aqueous extract and acid hydrolysate of J.
grandiflorum was higher than the verified for the A. muricata ones (Tables 1 and 2). J. grandiflorum
aqueous extract mainly contained flavonoids, namely quercetin-3-O-(2,6-di-rhamnosyl)glucoside
(
65.1% of total phenolic content) (Table 2), in opposition to the results obtained for A. muricata
aqueous extract, in which similar amounts of phenolic acids and flavonoids were found (Table 1). In
the corresponding acid hydrolysate, hydroxycinnamic acids were present in vestigial amounts,
quercetin being the main compound (ca. 73.0% of total phenolic content) (Table 2).
This article is protected by copyright. All rights reserved.

a
Table 2. Linearity, detection and quantification limits, and phenolic composition of J. grandiflorum.
LOD
LOQ
mg/g extract
2
Compounds
Regression equation
R
(
μg/mL)
0.95
0.95
0.48
2.00
0.48
3.76
3.16
(µg/mL)
AE
H
9
6
6
6
8
9
5- diHCQA
y= 1 x 10 x – 1 x 10
0.9999
0.9999
0.9983
0.9999
0.9983
0.9981
0.9983
2.89
2.89
1.44
6.06
1.44
11.38
9.56
3.38 ± 0.15
0.35 ± 0.02
23.76 ± 0.76
5.42 ± 0.34
3.56 ± 0.25
nd
nq
nq
9
5- diHMetCQA
y= 1 x 10 x – 1 x 10
8
1
0
1
Q-3-O-(2,6-di-rhmn)gluc
y= 5 x 10 x + 1 x 10
nd
8
1
K-3-O-(2,6-di-rhmn)gluc
y= 6 x 10 x - 386572
nd
8
6
7
6
4
Q-3-O-rut
y= 5 x 10 x + 1 x 10
nd
8
6
7
Q
K
y= 9 x 10 x + 2 x 10
2.79 ± 0.01
1.03 ± 0.02
3.82 ± 0.03
8
y= 7 x 10 x + 4 x 10
nd
Total
36.47 ± 1.52
a
Results are expressed as mean ± standard deviation of three determinations and correspond to mg/g of dry weight. LOD, Limit of detection;
LOQ, limit of quantification; AE, aqueous extract; H, acid hydrolysate; diHCQA, dihydrocaffeoylquinic acid; diHMetCQA,
dihydromethoxycaffeoylquinic acid; Q, quercetin; rhmn, rhamnosyl; gluc, glucoside; rut, rutinoside; K, kaempferol; nd, not detected; nq, not
quantified.
Anti-inflammatory potential
The anti-inflammatory potential of both A. muricata and J. grandiflorum extracts and respective acid
hydrolysates was assessed using the RAW 264.7 macrophage cell line. The toxicity of the several
extracts was preliminarily screened by the MTT assay. Thereafter, and taking into account that NO
overproduction is a trait of activated macrophages, the capacity of the extracts, at non-cytotoxic
concentrations, to reduce NO levels in the supernatant of the cell culture was evaluated.
Subsequently, their potential to inhibit PLA was also studied. The major compounds found in the
2
extracts were also tested to assess their contribution for the activity observed.
Cell viability
As can be seen in Fig. 3, the aqueous extract of A. muricata and the hydrolysate of the extracts of
both species affect cell viability in a dose-dependent way. On the other hand, the aqueous extract of
J. grandiflorum reduced the cell viability only for the higher concentration tested (Fig. 3). The
aqueous extract and acid hydrolysate from J. grandiflorum did not reveal the same behaviour (Fig.
3
B). This may be due to the presence of other non-determined compounds, which could be toxic in
the free form. For example, iridoid glycosides and triterpenoid saponins were already described in J.
[
19]
grandiflorum flowers, and the presence of glycosidic groups may promote the extraction of these
This article is protected by copyright. All rights reserved.

compounds with the solvent used in our experimental procedure. In addition, as a consequence of
[
25, 26]
the hydrolysis, and taking into account that iridoids and saponins are known to be cytotoxic
the resulting aglycons could contribute for the observed effects.
,
1
00
100
50
0
*
****
5
0
0
****
****
[
Aqueous extract of A. muricata] (μg/mL)
[Aqueous extract of J. grandiflorum] (μg/mL)
1
00
100
*
5
0
0
**
50
0
**
****
****
***
[
Hydrolysate of A. muricata] (μg/mL)
[Hydrolysate of J. grandiflorum] (μg/mL)
Figure 3. Viability of RAW 264.7 cells treated for 24 h with aqueous extracts and their acid hydrolysates: (A) A. muricata and (B) J. grandiflorum.
Results are presented as mean ± standard error of the mean of four experiments, performed in triplicate. *P<0.05, **P<0.01, ***P<0.001 and
****P<0.0001 compared to the control.
Concerning the pure phenolic compounds, no compound revealed to be cytotoxic in the concentration range assayed (Fig. 4).
Figure 4. Viability of RAW 264.7 cells treated for 24 h with quercetin, quercetin-3-O-rutinoside, kaempferol, kaempferol-3-O-rutinoside and 5-O-
caffeoylquinic acid. Results are presented as mean ± standard error of the mean of four experiments, performed in triplicate. *P<0.05, **P<0.01
and ****P<0.0001 compared to the control.
This article is protected by copyright. All rights reserved.

NO levels
As the exposure of macrophages to lipopolysaccharide (LPS) increases the production of NO by
induction of inducible nitric oxide synthase (iNOS), the ability of the samples to reduce NO levels was
assessed using LPS-stimulated RAW 264.7 macrophages. Aqueous extracts from both species and the
acid hydrolysate of J. grandiflorum were able to decrease NO levels, the aqueous extract of A.
muricata revealing higher inhibitory potential (Fig. 5). Despite its richness in phenolic compounds,
the aqueous extract of J. grandiflorum was the less active (Table 2, Fig. 5). The lower activity of this
aqueous extract may be related with its main compound, the tri-glycosylated flavonol, quercetin-3-
O-(2,6-di-rhamnosyl)glucoside (10) (Table 2). In fact, it is well known that the increased
hydrophilicity may hinder the entrance of those compounds in the cell. Additionally, it may also lead
to a steric hindrance compromising the scavenging capacity of NO and/or the enzyme’
[
27, 28]
modulation.
B
C
1
00
100
**
**
*** ***
*
50
0
5
0
0
A
1
00
**
**
**
5
0
0
[
Aqueous extract of A. muricata] (μg/mL)
[Aqueous extract of J. grandiflorum] (μg/mL)
1
00
100
50
0
Dexamethasone (μM)
**
5
0
0
[
Hydrolysate of A. muricata] (μg/mL)
[Hydrolysate of J. grandiflorum] (μg/mL)
Figure 5. NO levels on RAW 264.7 cells pre-treated for 2 h with the positive control dexamethasone (A), aqueous extracts and respective acid
hydrolysates obtained from A. muricata (B) and J. grandiflorum (C), followed by 22 h co-treatment with 1 µg/mL LPS. Results are presented as
mean ± standard error of the mean of four independent experiments, performed in triplicate. *P<0.05, **P<0.01 and ***P<0.001 compared to the
LPS control.
With the exception of 5-O-caffeoylquinic acid, all phenolic compounds tested revealed capacity to
lower NO levels in a concentration-dependent way (Fig. 6). Quercetin was the most active
compound, followed by kaempferol (IC values of 7.47 µM and 15.07 µM, respectively). In addition,
5
0
[
29, 30]
and as expected,
the heterosides of quercetin and kaempferol were less active than the
respective aglycones (IC50 values of 17 µM and 33.40 µM, respectively) (Fig. 6). The capacity of
phenolic compounds, mainly flavonoids, to lower NO levels is not surprising. In fact, there are
several studies reporting the capacity of these metabolites, namely flavonoids, to either scavenge
NO or inhibit iNOS activity and/or expression. Amongst flavonoids, quercetin is clearly one of the
[
28-33]
most studied and one of the most active flavonols, which is in agreement with our results.
This article is protected by copyright. All rights reserved.

Nevertheless, the IC values found in the literature for the inhibition of LPS-induced NO production
5
0
by quercetin can be very different (for example, it can be found IC50 values as different as 17 µM and
1
07 µM), which may be explained by the different conditions used in the assays, namely concerning
[28, 30]
to cell density and incubation period.
Regarding 5-O-caffeoylquinic acid, it was only observed ca. 13% reduction of NO levels at 100 μM,
[
34]
which is not in accordance with that previously reported by Hwang and collaborators. Those
authors provided evidence of a reduction of LPS-induced NO production in the same cell line by this
hydroxycinnamic acid, at 5 and 10 μM, albeit working with different cell density and incubation
periods.
Therefore, the capacity of both aqueous extracts to reduce NO levels may be partially related with
the presence of quercetin-3-O-rutinoside, which is found in A. muricata and J. grandiflorum aqueous
extracts at 3.3 µM and 5.8 µM, respectively. Although quercetin, kaempferol, kaempferol-3-O-
rutinoside and 5-O-caffeoylquinic acid are present in both aqueous extracts and acids hydrolysates
at concentrations lower than those studied, we cannot ignore the possible existence of sinergic
and/or antagonic effects amongst the different compounds, which could also interfere with the
potential of the extracts to lower NO levels.
1
00
IC50 = 7.47 μM
100
50
0
IC50 = 17 μM
*
*
5
0
0
*
*
**
**
**
**
1
00
[
Quercetin] (μM)
[
Quercetin-3-O-rutinoside] (μM)
IC50 = 33.40 μM
5
0
0
1
00
IC50 = 15.07 μM
100
50
0
**
5
0
0
*
***
***
*
[
5-O-Caffeoylquinic acid] (μM)
***
[
Kaempferol] (μM)
[Kaempferol-3-O-rutinoside] (μM)
Figure 6. NO levels on RAW 264.7 cells pre-treated for 2 h with quercetin, quercetin-3-O-rutinoside, kaempferol, kaempferol-3-O-rutinoside and 5-O-
caffeoylquinic acid, followed by 22 h co-treatment with 1 µg/mL LPS. Results are presented as mean ± standard error of the mean of four independent
experiments, performed in triplicate. *P<0.05, **P<0.01 and ***P<0.001 compared to the LPS control.
This article is protected by copyright. All rights reserved.

PLA2
Phospholipases are enzymes relevant in the inflammatory process, as they cleave phospholipids at
[
35-38]
the sn-2 position, thus releasing arachidonic acid, a ω6-polyunsaturated fatty acid.
After this
step, the free fatty acid is available to act as substrate for enzymes like cyclooxygenase or
lipoxygenase, thus originating mediators, such as eicosanoids, namely prostaglandins,
[
35-38]
thromboxanes, leukotrienes or hydroxyeicosatetraenoic acids.
Therefore, the capacity of both
extracts and hydrolysates to inhibit PLA activity was also evaluated. A. muricata aqueous extract
2
and its hydrolysate inhibited PLA activity, in a dose-dependent way, the hydrolysate being more
2
active (IC50 = 142 µg/mL) (Fig. 7). Concerning to J. grandiflorum aqueous extract and its acid
hydrolysate, although for some concentrations they were more active than the A. muricata ones, a
concentration-dependent effect was not observed (Fig. 7).
B
C
1
00
100
***
*** ***
*
**
**
***
**
5
0
0
***
50
0
**
A
1
00
*
5
0
0
[
Aqueous extract of A. muricata] (μg/mL)
[Aqueous extract of J. grandiflorum] (μg/mL)
1
00
100
**
p-Bromophenacyl bromide ( μM)
*** ***
***
5
0
0
50
*
***
0
[
Hydrolysate of A. muricata] (μg/mL)
[Hydrolysate of J. grandiflorum] (μg/mL)
Figure 7. Activity (%) of the positive control p-bromophenacyl bromide (A), aqueous extracts and respective acid hydrolysates obtained from A. muricata (B)
and J. grandiflorum (C) on the enzyme PLA . Results are presented as mean ± standard error of the mean of at least three experiments, performed in triplicate.
P<0.05, **P<0.01 and ***P<0.001 compared to the respective control.
2
*
Regarding the effect of pure phenolic compounds, quercetin was clearly the most active one with an
IC value of 1.36 µM, followed by 5-O-caffeoylquinic acid (Fig. 8). Some works described the
5
0
capacity of phenolic compounds to inhibit this enzyme. For example, it was reported that quercetin-
-O-rutinoside dose-dependently inhibits PLA activity from human synovial fluid, with an IC50 value
of 250 µM. Lättig and co-workers performed computational and experimental studies to
demonstrate that quercetin was able to inhibit the same PLA , and that the presence of the double
bond between C2 and C3 has a pivotal role in the binding of flavonoids to PLA . These authors also
(IC50 = 7.1 µM). Nevertheless, in our work, the
3
2
[
39]
[40]
2
2
[
40]
found that kaempferol was able to inhibit PLA
2
assay conditions and the origin of the enzyme are different and kaempferol did not reveal capacity
This article is protected by copyright. All rights reserved.

to inhibit PLA in a concentration-dependent way. On the other hand, concentrations above 3.12 µM
2
could not be tested because of solubility issues in the wells, ca. 11% of inhibition being observed for
this concentration (data not shown). Taking into account the activity observed for the tested
compounds, the inhibitory capacity of both A. muricata and J. grandiflorum aqueous extracts can be
partially related with the presence of 5-O-caffeoylquinic acid (9.3 µM and 9.5 µM, respectively). In
addition, quercetin-3-O-rutinoside could contribute for the activity of J. grandiflorum and A.
muricata aqueous extracts (5.8 µM and 3.3 µM, respectively) (Fig. 8, Tables 1 and 2). On the other
hand, despite the higher amounts of quercetin (the most active flavonoid) found in J. grandiflorum
acid hydrolysate, the inhibitory capacity of this extract was lower than that observed for A. muricata
one. This fact may be due to the presence of 5-O-caffeoylquinic acid in vestigial amounts (Table 2,
Fig. 8). In addition, the existence of possible sinergisms and/or antagonisms amongst the different
identified and non-identified compounds could influence the observed activities.
IC50 = 1.36 μM
1
00
100
50
0
*
**
**
***
***
***
***
5
0
0
***
*
** ***
[
Quercetin] (μM)
[Quercetin-3-O-rutinoside] (μM)
1
00
100
50
0
***
*
** *** ***
***
5
0
0
[
Kaempferol-3-O-rutinoside] (μM)
[5-O-Caffeoylquinic acid] (μM)
2
Figure 8. Activity (%) of quercetin, quercetin-3-O-rutinoside, kaempferol-3-O-rutinoside and 5-O-caffeoylquinic acid on the enzyme PLA . Results are
presented as mean ± standard error of the mean of at least three experiments, performed in triplicate. *P<0.05, ** P<0.01 and ***P<0.001 compared to the
respective control.
Conclusions
The purpose of the present study has been to outline, for the first time, the anti-inflammatory potential of aqueous extracts from two
worldwide herbal teas (A. muricata and J. grandiflorum). Based on our results, we can conclude that extracts and hydrolysates from both
2
species were able to decrease NO levels and to inhibit PLA activity at non-cytotoxic concentrations. Furthermore, it was also observed
that aglycones were more active than their respective glycosides, and that catechol group in ring B seems to be essential for the observed
activities. Moreover, the hydrolysed extracts showed that, besides flavonoids, 5-O-caffeoylquinic acid could also have influence in the
This article is protected by copyright. All rights reserved.

2
inhibitory capacity upon PLA . The biological potential of both species could be partially attributed to the determined compounds.
Regarding phenolic profiling, a good linearity for all compounds was obtained in both analytical methods, and J. grandiflorum was the
herbal tea with higher content of these metabolites. Nevertheless, as being complex mixtures, the influence of non-determined
compounds in the extracts cannot be excluded. Overall, A. muricata and J. grandiflorum may be interesting sources of nutraceuticals,
providing beneficial anti-inflammatory effects.
Experimental Section
Standards and Reagents
Quercetin, quercetin-3-O-rutinoside and kaempferol were from Sigma (St. Louis, MO, USA). Kaempferol-3-O-rutinoside, quercetin-3-O-
glucoside, quercetin-3-O-galactoside and 5-O-caffeoylquinic acid were from Extrasynthèse® (Genay, France). Dulbecco’s Modified Eagle
Medium (DMEM), Dulbecco’s phosphate buffered saline (DPBS), heat inactivated fetal bovine serum (FBS) and Pen Strep solution
(
Penicillin 5000 units/mL and Streptomycin 5000 μg/mL) were purchased from Gibco, InvitrogenTM (Grand Island, NY, USA). LPS from
Escherichia coli, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT), sulphanilamide, N-(1-naphthyl) ethylene diamine,
dimethyl sulfoxide (DMSO), PLA from honey bee venom (Apis melífera L.), 5-lipoxygenase (5-LOX) from soybean (Glycine max (L.) Merr.),
,2-dilinoleoyl-sn-glycero-3-phosphocholine (DL-PC), tris-HCl and sodium deoxycholate were from Sigma-Aldrich (St. Louis, MO, USA).
Acetic acid, phosphoric acid and methanol were obtained from Merck (Darmstadt, Germany). Formic acid was acquired from ChemLab NV
Zedelgem, Bélgica). Water was deionised in a Milli-Q water purification system (Millipore, Bedford, MA, USA). The Chromabond C18 SPE
2
1
(
columns (6 mL/1000 mg) were purchased from Macherey-Nagel (Duren, Germany).
Plant Material
A. muricata leaves (lot 3611.133.15) and J. grandiflorum dried flower buds (lot 3574.119.15) were purchased from Morais e Costa & CA.
Lda (Porto, Portugal). The plant material was powdered to a mean particle size below 910 μm and stored at room temperature in plastic
flasks protected from light, until extraction. Voucher specimens are deposited at the Laboratory of Pharmacognosy of the Faculty of
Pharmacy of Porto University (AM-062015 and JG-062015).
Extraction
Herbal Teas
Aqueous extracts of A. muricata leaves and J. grandiflorum dried flower buds were prepared by boiling ca. 3 g of powdered material for 20
min, in 500 mL of water. The resulting extracts were filtered using a Büchner funnel and then freeze-dried in a Virtis SP Scientific Sentry 2.0
apparatus (Gardiner, NY 12525, USA). The lyophilized extracts were kept in a desiccator in the dark until analysis.
Acid Hydrolysis
Each aqueous lyophilized extract (ca. 80 mg) was hydrolysed by refluxing with 16 mL of HCl 2N, for 30 min. After cooling to room
temperature, the solution obtained was passed through a SPE C18 column, previously conditioned with 3 mL of methanol and 6 mL of
acidic water (pH 2 with HCl). The phenolic fraction retained in the column was then eluted with methanol until negative reaction to 20%
NaCl (ca. 5 mL). The resulting solution was evaporated at reduced pressure (Rotavapor® R-215, BÜCHI Labortechnik, Flawil, Switzerland)
until dryness, at 40 ºC. The dry hydrolysed extracts were kept at -20 ºC and protected from light until analysis.
Phenolic Compounds Analysis
A. muricata
Phenolic compounds of A. muricata aqueous extract and its acid hydrolysate were determined by HPLC-DAD system (Gilson), using a C18
[
23]
Spherisorb ODS2 (25.0 × 0.46 cm; 5 μm particle size) column from Waters (Ireland), according to a previously described procedure. The
mobile phase consisted of two solvents: formic acid (5%) (A) and methanol (B). Elution started with 5% B and a gradient was used to
obtain 15% B at 3 min, 25% B at 13 min, 30% B at 25 min, 35% B at 35 min, 45% B at 39 min, 45% B at 42 min, 55% B at 47 min, 75% B at 56
min and 100% B at 60 min. The injection volume was 20 µL and the flow rate 0.9 mL/min. Chromatograms were recorded at 320 nm
This article is protected by copyright. All rights reserved.

(hydroxycinnamic acids) and 350 nm (flavonoids). Quantification of phenolic compounds was achieved by the peak areas recorded in the
chromatograms relative to that registered for known concentrations of external standards.
J. grandiflorum
Phenolic compounds in both J. grandiflorum aqueous extract and respective acid hydrolysate were
determined by HPLC-DAD in the same system, but using a Luna C18 (2) 100A (25.0 × 0.46 cm; 5 μm
particle size) column from Phenomenex (Macclesfield, UK), following the procedure described by
[
24]
Ferreres and collaborators. The mobile phase consisted of two solvents: acetic acid (1%) (A) and
methanol (B). Elution started with 5% B and a gradient was used to obtain 45% B at 20 min, 75% B at
3
5 min and 80% B at 50 min. The injection volume was 20 µL and the flow rate 0.8 mL/min.
Chromatograms were recorded at 320 nm (hydroxycinnamic acids) and 350 nm (flavonoids).
Compounds 8 and 9 were quantified as 5-O-caffeoylquinic acid, compound 10 as quercetin-3-O-
rutinoside and compound 11 as kaempferol-3-O-rutinoside. The other compounds were quantified
as themselves.
Methods Validation
Linearity
The linearity range of both methods was assessed by building calibration curves using five different
concentration levels of the phenolic compounds, according to the range of concentrations present in
A. muricata and J. grandiflorum (Tables 1 and 2).
Limits of Detection and of Quantification
The limits of detection (LOD = 3.3*S0/b) and of quantification (LOQ = 10*S0/b) (where S0 is the
standard deviation of signal-to-noise ratio of a low concentration standard and b is the slope of the
calibration plot) for the analysed compounds were determined from calibration curve data (Tables 1
and 2).
Anti-inflammatory Activity
Cell Culture
RAW 264.7 macrophage cells were from the American Type Culture Collection (LGC Standards S.L.U.,
Barcelona, Spain) and were maintained in DMEM supplemented with 10% FBS and 1%
penicillin/streptomycin at 37 °C, in a humidified atmosphere of 5% CO2, as previously reported [26].
Cells were seeded in 96-well plates (35,000 cells/well) and allowed to attach for 24 h. Then, they
were pre-treated with different concentrations of aqueous extracts/acid hydrolysates/isolated
compounds dissolved in either medium with 0.5% DMSO or medium, for 2 h. Afterwards they were
treated with 1 µg/mL LPS, for 22 h at 37 °C, in a humidified atmosphere of 5% CO2.
This article is protected by copyright. All rights reserved.

Cell Viability
[
41]
Cell viability was evaluated by the MTT reduction assay. After incubation (24 h), MTT (final
concentration 0.5 mg/mL) was added to each well and the plate was incubated for 2 h at 37 °C. The
formazan was dissolved by addition of a DMSO:isopropanol mixture (3:1) and quantified
spectrophotometrically at 560 nm. Results correspond to the mean of four independent
experiments performed in triplicate, and are expressed as percentage vs control cells.
NO Levels
The nitrite resulting from the conversion of NO in the culture medium was measured by the Griess
[
41]
reaction, spectrophotometrically, at 562 nm. Four independent experiments were performed in
triplicate. Dexamethasone was used as positive control. Control values were obtained in the absence
of aqueous extracts/ acid hydrolysates/ isolated compounds.
PLA activity
2
[
42]
The activity of PLA
based on a coupled enzymatic assay using DL-PC as PLA
6-well plates. Briefly, each well contained the mixture of DL-PC (65 μM), 5-LOX (0.23 µg/mL), PLA
0.25 μg/mL) and different concentrations of aqueous extracts/ acid hydrolysates/ reference
2
was determined spectrophotometrically according to a described procedure,
2
substrate and 5-LOX as coupling enzyme, in
9
(
2
compounds in buffer [Tris-HCl (50mM) with sodium deoxycholate (3mM), pH 8.5]. The plate was
incubated at 37 ºC, for 25 min. The absorbance was read at 234 nm in a Multiscan GO (Thermo
scientific, Finland) equipped with Skanlt 4.1 software. Three independent experiments were
performed in triplicate. p-Bromophenacyl bromide was used as positive control. Results are
expressed as percentage relatively to control.
Statistical Analysis
Statistical analysis was performed using GraphPad Prism 6 Software (San Diego, CA, USA). The
analysis of variance t test was used to compare the existence of significant differences relatively to
control. In all cases, p values lower than 0.05 were considered statistically significant.
Acknowledgements
This work received financial support from National Funds (FCT/MEC, Fundação para a Ciência e Tecnologia/Ministério da Educação e
Ciência) through project UID/QUI/50006/2013, co-financed by European Union (FEDER under the Partnership Agreement PT2020). To all
financing sources the authors are greatly indebted. Andreia P. Oliveira (SFRH/BPD/96819/2013) thanks FCT for the grant.
This article is protected by copyright. All rights reserved.

Author Contributions
Conceived and designed the experiments: P.V.; P.B.A. Performed the experiments: A.P.O.; I.S.; R.F.G.
Analyzed the data: A.P.O.; I.S.; P.V.; P.B.A.; R.F.G.; D.M.P. Contributed
samples/reagents/materials/analysis tools: P.V.; P.B.A. Wrote the paper: A.P.O; D.M.P.; R.F.G.;
P.B.A.; P.V.
References
[
1]
K. Moesgaard Iburg, T. Voss, J. JL. Salomon, C. Murray, Lancet 2015, 386, 743 – 800.
T. Lawrence, Cold Spring Harb. Perspect. Biol. 2009, 1, a001651.
[
[
[
[
[
[
[
[
[
[
[
2]
3]
4]
5]
6]
7]
8]
9]
N. Fujiwara, K. Kobayaski, Curr. Drug. Targets Inflamm. Allergy 2005, 4, 281 – 286.
T. Lawrence, D. A. Willoughby, D. W. Gilroy, Nat. Rev. Immunol. 2002, 2, 787 – 795.
Jr. E. Middleton, C. Kandaswami, T. C. Theoharides, Pharmacol. Rev. 2000, 52, 673 – 751.
M. P. Nair, S. Mahajan, J. L. Reynolds, R. Aalinkeel, H. Nair, S. A. Schwartz, C. Kandaswami, Clin. Vaccine Immunol. 2006, 13, 319 – 328.
D. M. Pereira, P. Valentão, J. A. Pereira, P. B. Andrade, Molecules 2009, 14, 2202 – 2211.
M. -H. Pan, C. -S. Lai, C. -T. Ho, Food Funct. 2010, 1, 15 – 31.
S. Martínez-Flórez, M. B. Gutiérrez, S. Sánchez-Campos, J. González-Gallego, M. J. Tuñón, J. Nutr. 2005, 135, 1359 – 1365.
10] T. -J. Chen, J. -Y. Jeng, C. -W. Lin, C. -Y. Wu, Y. -C. Chen, Toxicology 2006, 223, 113 – 126.
11] D. Ribeiro, M. Freitas, J. L. F. C. Lima, E. Fernandes, Med Res Rev. 2015, 35, 877 – 936.
12] M. Masullo, P. Mointoro, A. Mari, C. Pizza, S. Piacente, J. Pharm. Biomed. Anal. 2015, 113, 189 – 211.
[13] A. Manosroi, P. Jantrawut, E. Ogihara, A. Yamamoto, M. Fukatsu, K. Yasukawa, H. Tokuda, N. Suzuki, J. Manosroi, T. Akihisa, Chem. Biodivers. 2013, 10,
1
448 – 1463.
[
14] D. A. A. Kogiannou, N. Kalogeropoulos, P. Kefalas, M. G. Polissiou, A. C. Kaliora, Food Chem. Toxicol. 2013, 61, 152 – 159.
15] S. Dewanjee, T. K. Dua, R. Sahu, Food Chem. Toxicol. 2013, 59, 514 – 520.
[
[
16] F. Z. Benomari, N. Djabou, A. Medbouhi, A. Khadir, M. Bendahou, C. Selles, J. -M. Desjobert, J. Costa, A. Muselli, Chem. Biodivers. 2016, 13, 1559 –
572.
1
[17] D. Yu, W. Pu, D. Li, D. Wang, Q. Liu, Y. Wang, Chem. Biodiversity 2016, 13, 1140 – 1148.
[
[
[
[
[
[
[
18] A. V. Coria-Téllez, E. Montalvo-Gónzalez, E. M. Yahia, E. N. Obledo-Vázquez, Arab. J. Chem. 2016, doi.org/10.1016/j.arabjc.2016.01.004.
19] S. Sandeep, P. M. Paarakh, Pharmacologyonline 2009, 2, 586 – 595.
20] O. V. de Sousa, G. D. -V. Vieira, J. J. R. G. de Pinho, C. H. Yamamoto, M. S. Alves, Int. J. Mol. Sci. 2010, 11, 2067 – 2078.
21] C. P. Foong, R. A. Hamid, Rev. Bras. Farmacogn. 2012, 22, 1301 – 1307.
22] A. P. Chaturvedi, Y. B. Tripathi, Inflammopharmacol. 2011, 19, 273 – 281.
23] C. Grosso, P. Valentão, C. Andrade, P. B. Andrade, Microchem. J. 2015, 119, 176 – 182.
24] F. Ferreres, C. Grosso, A. Gil-Izquierdo, P. Valentão, P. B. Andrade, J. Pharm. Biomed. Anal. 2014, 88, 157 – 161.
This article is protected by copyright. All rights reserved.

[
[
[
[
[
[
[
25] E. L. Ghisalberti, Phytomedicine, 1998, 5, 147 – 163.
26] I. Podolak, A. Galanty, D. Solobewska, Phytochem. Rev. 2010, 9, 425 – 474.
27] S. Y. Namgoong, K. H. Son, H. W. Chang, S. S. Kang, H. P. Kim, Life Sci. 1994, 54, 313 – 320.
28] H. K. Kim, B. S. Cheon, Y. H. Kim, S. Y. Kim, H. P. Kim, Biochem. Pharmacol. 1999, 58, 759 – 765.
29] J. Wang, G. Mazza, J. Agric. Food Chem. 2002, 50, 850 − 857.
30] Y. C. Chen, S. C. Shen, W. R. Lee, W. C. Hou, L. L. Yang, T. J. Lee, J. Cell. Biochem. 2001, 82, 537 – 548.
31] J. -C. Chen, F. -M. Ho, P. -D. L. Chao, C. -P. Chen, K. -C. G. Jeng, H. -B. Hsu, S. -T. Lee, W. T. Wu, W. -W. Lin, Eur. J. Pharmacol. 2005,
5
21, 9 – 20.
[32] M. Hämäläinen, R. Nieminen, P. Vuorela, M. Heinonen, E. Moilanen, Mediat. Inflamm. 2007, 2007, 45673.
[33] V. García-Mediavilla, I. Crespo, P.S. Collado, A. Esteller, S. Sánchez-Campos, M. J. Tuñón, J. González-Gallego, Eur. J. Pharmacol. 2007,
5
57, 221 – 229.
[
[
[
[
[
[
[
34] S. J. Hwang, Y. M. Kim, Y. Park, H. J. Lee, K. W. Kim, Inflamm. Res. 2014, 63, 81 – 90.
35] J. E. Burke, E. A. Dennis, J. Lipid Res. 2009, 50, S237 – S242.
36] P. C. Calder, Biochem. Soc. Trans. 2005, 33, 423 – 427.
37] P. C. Calder, Am. J. Clin. Nutr. 2006, 83, 1505S – 1519S.
38] C. S. Williams, M. Mann, R. N. DuBois, Oncogene 1999, 18, 7908 – 7916.
39] M. Lindahl, C. Tagesson, Inflammation 1997, 21, 347 – 356.
40] J. Lättig, M. Böhl, P. Fischer, S. Tischer, C. Tietböhl, M. Menschikowski, H. O. Gutzeit, P. Metz, M. T. Pisabarro, J. Comput. Aided Mol. Des.
2
007, 21, 473 – 483.
[41] D. M. Pereira, G. Correia-da-Silva, P. Valentão, N. Teixeira, P. B. Andrade, PLoS ONE 2014, 9, e88341.
[42] M. Jimenez, J. Cabanes, F. Gandía-Herrero, J. Escribano, F. García-Carmona, M. Pérez-Gilabert, Anal. Biochem. 2003, 319, 131 – 137.
This article is protected by copyright. All rights reserved.