ARTICLE IN PRESS

Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245

Contents lists available at ScienceDirect

Prostaglandins, Leukotrienes and

Essential Fatty Acids
journal homepage: www.elsevier.com/locate/plefa

Antiproliferative activity of guava leaf extract via inhibition of prostaglandin

endoperoxide H synthase isoforms
Yuki Kawakami a, Tomomi Nakamura a, Tomoko Hosokawa a, Toshiko Suzuki-Yamamoto a,
Hiromi Yamashita a, Masumi Kimoto a, Hideaki Tsuji a, Hideki Yoshida b,
Takahiko Hada b, Yoshitaka Takahashi a,
a
Department of Nutritional Science, Faculty of Health and Welfare Science, Okayama Prefectural University, 111 Kuboki, Soja, Okayama 719-1197, Japan
b
Research Center, Bizen Chemical Co., Ltd., 363 Tokutomi, Akaiwa, Okayama 709-0716, Japan

a r t i c l e in fo abstract

Article history: Prostaglandin endoperoxide H synthase (PGHS) is a key enzyme for the synthesis of prostaglandins
Received 30 October 2008 (PGs) which play important roles in inflammation and carcinogenesis. Because the extract from Psidium
Received in revised form guajava is known to have a variety of beneficial effects on our body including the anti-inflammatory,
3 April 2009
antioxidative and antiproliferative activities, we investigated whether the extract inhibited the catalytic
Accepted 23 April 2009
activity of the two PGHS isoforms using linoleic acid as an alternative substrate. The guava leaf extract
inhibited the cyclooxygenase reaction of recombinant human PGHS-1 and PGHS-2 as assessed by
Keywords: conversion of linoleic acid to 9- and 13-hydroxyoctadecadienoic acids (HODEs). The guava leaf extract
Prostaglandin endoperoxide H synthase also inhibited the PG hydroperoxidase activity of PGHS-1, which was not affected by nonsteroidal anti-
Cyclooxygenase
inflammatory drugs (NSAIDs). Quercetin which was one of the major components not only inhibited the
Prostaglandin hydroperoxidase
cyclooxygenase activity of both isoforms but also partially inhibited the PG hydroperoxidase activity.
Psidium guajava
Overexpression of human PGHS-1 and PGHS-2 in the human colon carcinoma cells increased the DNA
synthesis rate as compared with mock-transfected cells which did not express any isoforms. The guava
leaf extract not only inhibited the PGE2 synthesis but also suppressed the DNA synthesis rate in the
PGHS-1- and PGHS-2-expressing cells to the same level as mock-transfected cells. These results
demonstrate the antiproliferative activity of the guava leaf extract which is at least in part caused by
inhibition of the catalytic activity of PGHS isoforms.
& 2009 Elsevier Ltd. All rights reserved.

1. Introduction drugs (NSAIDs). They produced reversible or irreversible inhibi-

Prostaglandin endoperoxide H synthase (PGHS) is a rate
limiting enzyme for the production of prostaglandins (PGs) and
thromboxane from arachidonic acid. The enzyme catalyzes both
bisdioxygenation of arachidonic acid (cyclooxygenase activity)
producing PGG2 and conversion of PGG2 to PGH2 (PG hydroper-
oxidase activity) [1]. There are two isoforms, PGHS-1 which is
constitutively expressed and PGHS-2 which is inducible [2,3].
Recent studies using PGHS-1-deficient and PGHS-2-deficient mice
as well as isoform-specific inhibitors revealed physiological
functions of the each enzyme. For example, PGHS-1 but not
PGHS-2 played an essential role in platelet aggregation, whereas
only PGHS-2 was necessary for ovulation and implantation [4].
Both isoforms functioned coordinately in carcinogenesis and
inflammation [4]. The isoforms are pharmacological targets of
aspirin, indomethacin and other nonsteroidal anti-inflammatory
 Corresponding author. Tel./fax: +81866 94 2155.
E-mail address: ytaka@fhw.oka-pu.ac.jp (Y. Takahashi).
tion of cyclooxygenase by competing with substrate arachidonic
acid for the active site of the enzyme, leaving the PG hydro-
peroxidase activity of the enzyme unaffected [2,3]. NSAIDs also
contributed to a reduction in mortality from colorectal cancer in
individuals [5,6]. We previously found that indomethacin sup-
pressed the stimulated growth, the increased DNA synthesis, and
the induction of epidermal growth factor receptor in human colon
carcinoma cells overexpressing either PGHS-1 or PGHS-2 [7].
The extract from guava leaves which contained a variety of
polyphenols was commonly taken in Japan and subtropical
countries as a dietary supplement showing various pharmacolo-
gical effects including anti-inflammatory, antioxidative and anti-
proliferative activities [8–12]. A number of reports showed that
flavonoids and related compounds in plants reduced PG release by
suppressing PGHS-2 transcription [13–15]. In fact, it was reported
that the fermented guava leaf extract suppressed lipopolysacchar-
ide-induced PGHS-2 and iNOS expression in mouse macrophage
cells by inhibition of transcription factor NF-kappaB [16]. How-
ever, only limited studies demonstrated the modulation of the
enzyme activity of PGHS isoforms by flavonoids. We previously

0952-3278/$ - see front matter & 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.plefa.2009.04.006

Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
 ARTICLE IN PRESS
240 Y. Kawakami et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245
used [1-14C] arachidonic acid as a substrate to assay the enzyme
activity by quantification of radioactive products on thin layer
chromatography [7,17,18]. In this study, we developed the method
for the PGHS activity assay using non-radioactive linoleic
acid as an alternative substrate. Linoleic acid was oxygenated by
the cyclooxygenase activity of the both PGHS isoforms to 9- and
13-hydroperoxyoctadecadienoic acids (HPODEs) followed by
reduction of the produced HPODEs to 9- and 13-hydroxy-
octadecadienoic acids (HODEs) by the PG hydroperoxidase
activity of the same enzyme [19,20]. The produced 9- and
13-HODEs were easily quantified by monitoring UV absorption
derived from the conjugated dienes of the products after their
separation by HPLC [21,22]. Using this method we demonstrated
the inhibition of the catalytic activity of recombinant human PGHS
isoforms by the guava leaf extract. We also reported inhibitory
effects of the guava leaf extract on the proliferation rate of the
human colon carcinoma cells overexpressing PGHS-1 or PGHS-2.
2. Materials and methods
2.1. Materials
Arachidonic acid and linoleic acid were obtained from Nu-Chek
Prep (Elysian, USA), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl
tetrazolium bromide (MTT), ellagic acid, geneticin, RPMI 1640
medium, and soybean lipoxygenase (type I) from Sigma (St. Louis,
USA), fetal bovine serum (FBS) from JRH Biosciences (Lenexa,
USA), lipofectamine from Invitrogen (Carlbad, USA), a PGE2 high
sensitivity immunoassay kit from R&D Systems (Minneapolis,
USA), ovine PGHS-1 (45,000–55,000 units/mg of protein, 1 unit of
enzyme consumed 1 nmol of oxygen per min at 37 1C) from
Cayman Chemical (Ann Arbor, USA), quercetin dihydrate from
Wako (Osaka, Japan), and quercetin-3-O-glucoside from Extra-
synthese (Genay, France). 13-HODE and 15-hydroxyeicosatetrae-
noic acid (15-HETE) were prepared by incubation of soybean
lipoxygenase type I with linoleic acid and arachidonic acid,
respectively, followed by purification with HPLC as described
[23]. 13-Hydroperoxyoctadecadienoic acid was synthesized by
incubation of linoleic acid with soybean lipoxygenase and
subjected to thin layer chromatography with a solvent mixture
of diethyl ether/petroleum ether/acetic acid (85/15/0.1, v/v) at
4 1C. 13-HPODE was located by UV illumination, eluted from the
silica gel with ethyl acetate, and dissolved in a small volume of
ethanol. A molecular extinction coefficient of 27,000 was used to
determine the concentration of the hydro(pero)xy acids [24]. The
expression plasmids pBOSNeoPGHS1 and pBOSNeoPGHS2 in
which the transcription of human PGHS-1 [25] and human
PGHS-2 [26] were driven by a powerful elongation factor-1a
promoter were constructed as described [17]. All the other
reagents and chemicals were commercially available extra-pure-
grade products.
2.2. Preparation of guava leaf extract
Leaves of guava (Psidium guajava) were obtained from Fukuda
Ryu Co., Ltd. (Osaka, Japan). The dried guava leaves (150 g) were
extracted with 1000 ml of 50% ethanol (v/v), and the extracted
solution was filtrated and evaporated. The residue was dissolved
in water and subjected to hydrophobic chromatography using a
Diaion HP-20 column (Mitsubishi Chemical Co., Tokyo, Japan).
After washing the column with 750 ml of water, a polyphenol-rich
fraction was eluted with 150 ml of 50% ethanol (v/v). Finally, the
eluted fraction was concentrated by freeze-drying to obtain the
powder (9.8 g) and used as the guava leaf extract. The content of
Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
polyphenol was 70.971.9 g per 100 g of the extract as measured
by the Folin–Denis method [27].
2.3. Cell culture and plasmid transfection
A human colon adenocarcinoma cell line, COLO320DM, was a
gift from Dr. T. Sasaki (Kanazawa University). The cells did not have
any enzyme which metabolized arachidonic acid [7]. Transfection
of the cells with the pBOSNeoPGHS1 and pBOSNeoPGHS2 as well
as the parental pBOSNeo vector (mock transfection) and subse-
quent isolation of clones were carried out as described previously
[7]. The cells were cultured in RPMI 1640 medium supplemented
with 10% FBS, 100 U/ml penicillin, 100 mg/ml streptomycin and
500 mg/ml geneticin. The cells were maintained at 37 1C in a
humidified atmosphere of 95% air and 5% CO2, and subcultured
every 3–4 days using a standard trypsin protocol.
2.4. Cyclooxygenase assay
The cyclooxygenase reaction was carried out as described
previously [7] except that linoleic acid was used as a substrate.
Human PGHS-1 and PGHS-2 were prepared from COLO320DM
cells overexpressing the enzyme. Briefly, the cultured cells were
sonicated in 50 mM Tris–HCl buffer at pH 7.4 containing 1 mM
EDTA by a Branson sonifier model 250 (Danbury, USA). Mem-
branes were pelleted by centrifugation at 150,000g for 1 h at 4 1C
and the washed membranes were resuspended with a Potter
homogenizer in 50 mM Tris–HCl buffer at pH 7.4 containing 1 mM
EDTA. Human PGHS-1 and PGHS-2 were solubilized by adding
Tween 20 to 1% (v/v) and stirring for 30 min at 4 1C. Residual
materials were removed by another round of centrifugation and
the supernatant liquid containing solubilized human PGHS-1 or
PGHS-2 was used in the cyclooxygenase assay. Alternatively,
commercially available purified ovine PGHS-1 was used. The
enzyme was preincubated at 24 1C for 5 min with indicated
concentrations of the guava leaf extract, ellagic acid, quercetin
or quercetin-3-O-glucoside in a standard 200-ml reaction mixture
containing 100 mM Tris–HCl buffer (pH 7.4), 2 mM hematin and
5 mM tryptophan. The enzyme reaction was started by addition of
25 mM linoleic acid (5 nmol/5 ml ethanol solution) to the mixture
and continued for 5 min (with human PGHS-1 and PGHS-2) or 30 s
(with ovine PGHS-1) at 24 1C with constant mixing. The
reaction was quenched by addition of 0.05 M HCl, and 1 nmol of
15-HETE was added as an internal standard. The products
extracted with ice-cold diethyl ether were analyzed by reverse-
phase HPLC using a COSMOSIL 5C18-MS-II column (5 mm particle,
4.6  250 mm, Nacalai Tesque, Kyoto, Japan) with a solvent system
of methanol/water/acetic acid (80:20:0.01, v/v) at a flow rate of
1 ml/min as described [23]. Absorption at 235 nm was continu-
ously monitored. The produced 13-HODE and 9-HODE which
cochromatographed on reverse-phase HPLC as a single peak were
quantified by the comparison of the area of the peak with that of
an internal standard. Linear relationship was confirmed by the
calibration curves (five-point measurements) for the conjugated
dienes of 13-HODE and 15-HETE. In some experiments the
enzymes were incubated with 25 mM arachidonic acid in the
standard reaction mixture as described above and the produced
PGE2 were quantified with a PGE2 high sensitivity immunoassay
kit. Protein concentration was determined using a BCA protein
assay kit with bovine serum albumin as a standard.
2.5. PG hydroperoxidase assay
The purified ovine PGHS-1 (20 units) was preincubated with
indicated concentrations of the guava leaf extract or quercetin at
 ARTICLE IN PRESS
Y. Kawakami et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245 241

24 1C for 5 min in a 200-ml reaction mixture. The mixture

contained 100 mM Tris–HCl buffer at pH 7.4, 2 mM hematin,
5 mM tryptophan, and 3.75 mg/ml bovine serum albumin which
stabilized unstable 13-HPODE as a substrate. The enzyme reaction
was started by addition of 50 mM 13-HPODE (10 nmol/10 ml
ethanol solution) and continued for 30 s at 24 1C with constant
mixing. The reaction was quenched by addition of 0.05 M HCl and
3 nmol of 15-HETE was added as an internal standard. The
products were extracted with ice-cold diethyl ether and analyzed
by straight-phase HPLC using a Nova-Pak silica column (5-mm
particle, 3.9  150 mm, Waters, Milford, USA) with a solvent
system of hexane/2-propanol/acetic acid (100:0.5:0.1, v/v) at a
flow rate of 1 ml/min as described [23]. Absorption at 235 nm
was continuously monitored, and produced 13-HODE as well as Fig. 1. Reverse-phase HPLC analysis of the reaction products of PGHS-1 and PGHS-
2 from linoleic acid. Human PGHS-1 (A) and PGHS-2 (B) prepared from
the substrate 13-HPODE was quantified by the comparison of the
COLO320DM overexpressing the enzyme (80 mg each of protein) were incubated
area of the peak with that of an internal standard. Linear with 25 mM linoleic acid at 24 1C for 5 min in the standard cyclooxygenase reaction
relationship was confirmed by the calibration curve (five-point mixture. The same preparation from mock-transfected cells was also incubated (C).
measurements) for the conjugated dienes of 15-HETE, 13-HODE The products were extracted and subjected to reverse-phase HPLC monitoring at
and 13-HPODE. 235 nm. 15-HETE at 1 nmol was included as an internal standard. Arrows indicate
the elution positions of authentic standards.

2.6. Cell proliferation assay

Cell proliferation was assessed by using a BrdU cell prolifera-

tion assay kit (Exalpha Biologicals, Watertown, USA) according to
the manufacturer’s instructions. In brief, the cells overexpressing
PGHS-1 or PGHS-2 were seeded into a 96-well plate at a density of
5  104 cells/well in RPMI 1640 containing 10% FBS and 500 mg/ml
of geneticin and preincubated at 37 1C for 12 h. After addition of
the guava leaf extract at 0, 20 and 100 mg/ml, the cells were
incubated for 4 h, followed by treatment for 1 h with 60 mM
bromodeoxyuridine (BrdU), fixation for 30 min, incubation for 1 h
with an anti-BrdU antibody, and incubation for 30 min with a goat
anti-mouse IgG-peroxidase conjugate. After washing, tetramethyl-
benzidine was added and incubated for 30 min at room tempera-
ture in the dark. The reaction was quenched with 1.2 M sulfuric
acid. Absorbance at 450 nm was measured with a reference
wavelength of 570 nm by using a Bio-Rad microplate reader Model
680 (Hercules, USA). The number of viable cells was determined
simultaneously by the MTT assay as described [7] and used to
normalize the data of the cell proliferation assay.

3. Results

3.1. Inhibition of cyclooxygenase activity by guava leaf extract

We measured the cyclooxygenase activity of PGHS using

linoleic acid as an alternative substrate because the products
could be easily quantified by HPLC monitoring UV absorption. The
human PGHS-1 and PGHS-2 prepared from COLO320DM cells
overexpressing the enzyme were incubated with linoleic acid in
the standard reaction mixture and the products were analyzed by
HPLC. A single peak which cochromatographed with authentic
13-HODE on reverse-phase HPLC was observed (Fig. 1A and B).
The peak was separated on straight-phase HPLC into two
peaks with approximately the same area and the peaks
cochromatographed with authentic 13-HODE and 9-HODE (data
not shown). The specific activities were 6.2 and 4.6 nmol of total
HODEs/5 min/mg of protein for the human PGHS-1 and PGHS-2
preparations, respectively. The Km values of PGHS-1 and PGHS-2
for linoleic acid were 5.5 and 6.8 mM, respectively. These were
in the same order of magnitude as reported Km values for
arachidonic acid [20]. The same preparation from mock-
transfected cells did not convert linoleic acid (Fig. 1C).
Fig. 2. Inhibition of the cyclooxygenase activity by the guava leaf extract: (A)
human PGHS-1 and PGHS-2 (80 mg protein) were incubated with 25 mM linoleic
acid at 24 1C for 5 min in the standard cyclooxygenase reaction mixture in the
presence of the guava leaf extract at various concentrations and the products were
quantified using reverse-phase HPLC. The data represent means7SD of three
separate experiments. (B) The same enzyme preparations were incubated with
25 mM arachidonic acid at 24 1C for 5 min in the standard cyclooxygenase reaction
mixture in the presence of the guava leaf extract at various concentrations, and
produced PGE2 was quantified using a PGE2 high sensitivity immunoassay kit.
Relative enzyme activities as compared with the activity without inhibitors are
shown. Data represent means7SD of triplicate experiments.
The effect of the guava leaf extract on the enzyme activity of
human PGHS isoforms was investigated. As shown in Fig. 2A, the
guava leaf extract inhibited the cyclooxygenase activity of both
isoforms in a concentration-dependent manner. The IC50 values
for PGHS-1 and PGHS-2 were 55 and 560 mg of dry extract/ml,
respectively. As control experiments using known inhibitors for

Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
 ARTICLE IN PRESS
242 Y. Kawakami et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245
PGHS isoforms, indomethacin at 0.1 mM inhibited the PGHS-1 and
PGHS-2 reaction by 46% and 7%, respectively, whereas 1 mM of
NS-398, a specific PGHS-2 inhibitor, inhibited the PGHS-1 or
PGHS-2 reaction by 5% and 64%, respectively (data not shown).
We incubated the enzyme preparations with arachidonic acid in
the presence of the guava leaf extract and the produced PGE2 was
quantified by an enzyme immunoassay (Fig. 2B). The rate of PGE2
production was 3.8 and 1.3 nmol/5 min/mg of protein for PGHS-1
and PGHS-2, respectively. The guava leaf extract inhibited both
isoforms with the IC50 values of 44 and 420 mg of dry extract/ml
for PGHS-1 and PGHS-2, respectively.
3.2. Inhibition of PG hydroperoxidase activity by guava leaf extract
We investigated whether the guava leaf extract inhibited the
PG hydroperoxidase activity of PGHS. To minimize non-enzymatic
degradation of unstable 13-HPODE as a substrate during the
enzyme reaction, we used purified ovine PGHS-1, added bovine
serum albumin in the reaction mixture and shortened the
incubation period to 30 s. The products were readily extracted
with the substrate 13-HPODE and quantified by straight-phase
HPLC (Fig. 3A–C). The amount of produced 13-HODE
was corrected by the total amount of recovered 13-HPODE and
13-HODE. The recovery rate was 60–70% in this assay condition.
The cyclooxygenase activity was measured with the purified ovine
PGHS-1 using 25 mM linoleic acid as a substrate in the same
condition as that with the PG hydroperoxidase activity except that
bovine serum albumin was excluded from the reaction mixture.
The specific activities of cyclooxygenase and PG hydroperoxidase
were 0.9 and 0.5 mmol/30 s/mg of protein, respectively. As shown
Fig. 3. Inhibition of the PG hydroperoxidase activity by the guava leaf extract. The
purified ovine PGHS-1 was incubated with 50 mM 13-HPODE at 24 1C for 30 s in the
standard PG hydroperoxidase reaction mixture in the absence (A) or presence of
100 mg/ml the guava leaf extract (B) or 20 mM indomethacin (IM) (C). The products
were analyzed using straight-phase HPLC monitoring at 235 nm. 15-HETE at
3 nmol was included as an internal standard. Arrows indicate the elution positions
of authentic standards. Cyclooxygenase (D) and PG hydroperoxidase (E) activities
were measured using purified ovine PGHS-1 as described in Materials and
methods in the presence of indicated concentrations of the guava leaf extract or
IM. Relative enzyme activities as compared with the activity without inhibitors are
shown. Data represent means7SD of three separate experiments. *Po0.05 from
the enzyme activity without the guava leaf extract using one-way analysis of
variance with Dunnet’s post hoc test.
Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
in Fig. 3D and E, the guava leaf extract dose-dependently inhibited
the PG hydroperoxidase activity (IC50 ¼ 5.1 mg of dry extract/ml)
as well as the cyclooxygenase activity (IC50 ¼ 4.5 mg of dry extract/
ml) of purified ovine PGHS-1. The difference of the IC50 values for
the cyclooxygenase activity between human and ovine PGHS-1
might be due to species differences of the enzyme [28,29].
Indomethacin at 20 mM inhibited the cyclooxygenase activity
almost completely, whereas it did not affect the PG hydro-
peroxidase activity (Fig. 3D and E).
3.3. Inhibition of cyclooxygenase and PG hydroperoxidase activities
by components contained in guava leaf extract
Because our preliminary experiments showed that ellagic
acid, quercetin and quercetin glycosides were major polyphenols
in the guava leaf extract, we examined whether these compounds
inhibited PGHS isoforms using human PGHS-1 and PGHS-2
preparations. As shown in Fig. 4A and B, quercetin dose-
dependently inhibited the cyclooxygenase activity of human
PGHS-1 and PGHS-2 with IC50 values of 5.3 mg/ml (18 mM)
and 26 mg/ml (86 mM), respectively. Quercetin-3-O-glucoside
inhibited the cyclooxygenase activity of both isoforms at
higher concentrations. Ellagic acid hardly inhibited the
cyclooxygenase activity, although the concentration at more
than 1000 mg/ml was not tested because of the limited solubility
of the compound. Quercetin significantly inhibited the PG
hydroperoxidase activity of purified ovine PGHS-1, although
more than 50% inhibition was not observed at the concentration
where the cyclooxygenase activity was almost completely
inhibited (Fig. 5A and B).
Fig. 4. Effect of ellagic acid, quercetin and quercetin-3-O-glucoside on the
cyclooxygenase activity of PGHS isoforms. Human PGHS-1 (A) and PGHS-2 (B)
were incubated with 25 mM linoleic acid at 24 1C for 5 min in the standard
cyclooxygenase reaction mixture in the presence of various concentrations of
ellagic acid, quercetin or quercetin-3-O-glucoside. Relative enzyme activities as
compared with the activity without inhibitors are shown. Data represent
means7SD of three separate experiments.
 ARTICLE IN PRESS
Y. Kawakami et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245
3.4. Reversibility and kinetic analysis of cyclooxygenase inhibition
To test the reversibility of the cyclooxygenase inhibition
by the guava leaf extract, we preincubated the purified ovine
PGHS-1 for 5 min with 100 mg/ml of the guava leaf extract
which almost completely inhibited the enzyme activity. The
preincubation mixture was 200-fold diluted 0, 5 and 30 min
before starting the enzyme reaction under the standard condition
for the cyclooxygenase assay. As shown in Fig. 6A, the
cyclooxygenase activity was time-dependently recovered after
dilution of the guava leaf extract, and the inhibition was hardly
observed after 30 min. The Lineweaver–Burk plot analysis of the
apparent initial velocity as measured by 30-s reaction of purified
ovine PGHS-1 with linoleic acid in the presence of various
concentrations of quercetin suggested the inhibition of the
enzyme by quercetin was apparently competitive (Fig. 6B). The
Ki value was estimated at 3.8 mM by Dixon plot. The analysis using
the guava leaf extract also showed apparently competitive
inhibition (data not shown), indicating that the major active
components contained in the guava leaf extract were apparently
competitive inhibitors.
Fig. 5. Inhibition of cyclooxygenase and PG hydroperoxidase activities by
quercetin. Cyclooxygenase (A) and PG hydroperoxidase (B) activities were
measured using purified ovine PGHS-1 as described in Fig. 4 in the presence of
indicated concentrations of quercetin or indomethacin (IM). Relative enzyme
activities as compared with the activity without inhibitors are shown. Data
represent means7SD of three separate experiments. *Po0.05 from the enzyme
activity without quercetin using one-way analysis of variance with Dunnet’s post
hoc test.
Fig. 6. Reversibility and kinetic analysis of cyclooxygenase inhibition. (A) The
purified ovine PGHS-1 was preincubated with 100 mg/ml the guava leaf extract at
24 1C for 5 min, and then the mixture was 200-fold diluted in the standard
cyclooxygenase reaction mixture. After 0, 5 and 30 min, the mixtures were
incubated with 25 mM linoleic acid at 24 1C for 30 s. The reaction without dilution
of the extract (no dilution) was also carried out. The relative enzyme activities as
compared with the control reaction carried out without the extract are shown.
(B) The purified ovine PGHS-1 was incubated with various concentrations of
linoleic acid at 24 1C for 30 s in the standard cyclooxygenase reaction mixture in
the absence or presence of indicated concentrations of quercetin. A double
reciprocal plot of the rate of HODE production (nmol/30 s) versus the substrate
concentration (mM) is shown.
Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
243
3.5. Growth suppression of human colon carcinoma cells
overexpressing PGHS by guava leaf extract
Because the guava leaf extract inhibited the enzyme activity of
PGHS isoforms, we investigated the effect of the guava leaf extract
on the growth of human colon carcinoma COLO320DM cells
overexpressing human PGHS-1 or PGHS-2. When the cells were
treated with 25 mM arachidonic acid, PGHS-1- and PGHS-2-
expressing cells produced 2.9 and 1.2 pmol of PGE2/104 cells,
respectively, as measured by an enzyme immunoassay. As shown
in Fig. 7A, PGE2 synthesis by the cells was inhibited by the
treatment with the guava leaf extract. We previously
demonstrated that the DNA synthesis rate as well as the growth
rate of PGHS-1- or PGHS-2-expressing cells was increased as
compared with mock-transfected cells which did not express
PGHS [7]. In the present experiment, we confirmed that
overexpression of PGHS-1 and PGHS-2 in the human colon
carcinoma COLO320DM cells increased the DNA synthesis rate
by 1.3- and 1.9-fold, respectively, as compared with mock-
transfected cells as examined by the BrdU incorporation
assay. As shown in Fig. 7B, treatment of the PGHS-1- and
Fig. 7. Growth suppression of COLO320DM cells overexpressing PGHS-1 and
PGHS-2 by the guava leaf extract. (A) The human PGHS-1- and PGHS-2-expressing
cells were cultured in a 96-well plate at a density of 5  104 cells/well. The cells
were pretreated for 30 min with the guava leaf extract at indicated concentrations,
followed by addition of 25 mM arachidonic acid and incubation at 37 1C for 15 min.
The medium was collected and subjected to a PGE2 enzyme immunoassay. Relative
amounts of produced PGE2 as compared with that without the extract are shown.
Data represent means7SD of triplicate experiments. (B) The human PGHS-1- and
PGHS-2-expressing cells as well as mock-transfected cells were cultured, and BrdU
incorporation was determined as described in Materials and methods. Results
were normalized based on the numbers of viable cells which were simultaneously
determined by the MTT assay. Data represent means7SD (n ¼ 6). *Po0.05 from
the cells without the guava leaf extract using one-way analysis of variance with
Dunnet’s post hoc test.
 ARTICLE IN PRESS
244 Y. Kawakami et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245
PGHS-2-expressing cells with 100 mg/ml of the guava leaf extract
significantly suppressed the DNA synthesis rate measured by a
BrdU incorporation to the level of that of mock-transfected cells
(*Po0.05). In contrast, the guava leaf extract did not affect the
DNA synthesis rate of mock-transfected cells. These results
indicated that treatment of human colon carcinoma cells with
the guava leaf extract suppressed the cell growth by inhibiting
PGHS isoforms.
4. Discussion and conclusion
In the present study, we demonstrated that the guava leaf
extract inhibited the enzyme activity of human PGHS-1 and
PGHS-2. Quercetin appeared to be included as an active
ingredient. In accordance with these observations, Banerjee
et al. reported that quercetin at 50 mM inhibited PGE2 biosynthesis
in A549 human lung adenocarcinoma cells very strongly with a
little effect on PGHS-2 mRNA and protein expression [30]. On the
other hand, quercetin suppressed PGE2 production in a OE33
human oesophageal adenocarcinoma cell line without any direct
inhibition of the PGHS-1 or PGHS-2 isoforms, although they used
a cyclooxygenase inhibitor screening assay kit with which
antioxidant might interfere with the assay [31]. Recently it was
reported that PGHS-1 and PGHS-2 were activated by dietary
bioflavonoids including quercetin [32]. However, they carried out
the enzyme reaction in the reaction mixture which did not
contain aromatic compounds such as phenol or tryptophan
required not only for the maximal PG hydroperoxidase activity
but also for the enzyme stability [33,34]. They added glutathione
which did not stimulate the activity [34], but rather changed the
degradation profiles of the produced PGH2 to the preferential
formation of 12-hydroxyheptadecatrienoic acid in the reaction
mixture [35]. It might be possible that the added flavonoids
compensated the role of tryptophan in the reaction and stabilized
the enzyme.
We used linoleic acid as a substrate for the activity assay
of cyclooxygenase. When the enzyme preparations from
COLO320DM cells overexpressing PGHS-1 or PGHS-2 were
incubated with 25 mM [1-14C] arachidonic acid at exactly the
same condition as employed in the present experiments, the
produced PGH2 was degraded into PGE2, PGD2, PGF2a and
12-hydroxyheptadecatrienoic acid, and PGE2 was calculated to be
25% of the total degradation products [7]. We incubated the same
enzyme preparations with arachidonic acid and quantified the
produced PGE2 by an enzyme immunoassay to calculate the
amount of PGH2 produced by PGHS isoforms. We found that
reactivities of linoleic acid with PGHS-1 and PGHS-2 were
approximately 41% and 88%, respectively, as compared with those
of arachidonic acid. Furthermore, the IC50 values of the guava leaf
extract to PGHS isoforms calculated with the two assay methods
were in the same order of magnitude (55 versus 44 mg dry
extract/ml for PGHS-1, 560 versus 420 mg dry extract/ml for PGHS-2).
The difference of the specific activity of the enzyme between
PGHS-1 and PGHS-2 might be caused by the different copy
number of the cDNAs integrated into the genomic DNA in the
transfected cells [17].
NSAIDs inhibit the cyclooxygenase activity but not the PG
hydroperoxidase activity [2,3]. We found that the guava leaf
extract also inhibited the PG hydroperoxidase activity of purified
ovine PGHS-1. It would be possible that more than one compound
in the guava leaf extract coordinately inhibited the PG hydro-
peroxidase activity. In fact, it was reported that active sites for the
cyclooxygenase activity and the PG hyderoperoxidase activity
were physically and functionally separate [1]. Thus, it is notable
that quercetin inhibited both cyclooxygenase and PG hydroper-
Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
oxidase activities. So far as we know, quercetin is the first
component inhibiting both activities of the enzyme. Further
investigation will be necessary to elucidate the mechanism on
such dual inhibition of PGHS by a single compound.
Our results indicated that the guava leaf extract exhibited the
growth inhibitory effect on human colon carcinoma cells over-
expressing PGHS isoforms by inhibiting the enzyme activity.
Recently, Manosroi et al. reported that the guava leaf oil showed
the anti-proliferative activity on the human mouth epidermal
carcinoma KB cells and murine leukemia P388 cells [8]. Chen et al.
indicated that the aqueous extract of P. guajava inhibited the
proliferation rate of the prostate cancer DU-145 cells [9]. However,
they did not show the mechanism of these effects. Fractionation
and identification of more effective components inhibiting PGHS
isoforms are currently in progress in our laboratory to elucidate
the detailed mechanism of the antiproliferative activity of the
guava leaf extract.
Acknowledgements
This work was supported by grants from the Ministry of
Education, Culture, Sports, Science and Technology of Japan, the
Japan Foundation for Applied Enzymology, and the Iijima
Memorial Foundation for the Promotion of Food Science and
Technology.
References
[1] W.L. Smith, L.J. Marnett, Prostaglandin endoperoxide synthase: structure and
catalysis, Biochim. Biophys. Acta 1083 (1991) 1–17.
[2] W.L. Smith, R.M. Garavito, D.L. DeWitt, Prostaglandin endoperoxide H
synthases (cyclooxygenases)-1 and -2, J. Biol. Chem. 271 (1996) 33157–33160.
[3] J.R. Vane, Y.S. Bakhle, R.M. Botting, Cyclooxygenases 1 and 2, Annu. Rev.
Pharmacol. Toxicol. 38 (1998) 97–120.
[4] W.L. Smith, R. Langenbach, Why there are two cyclooxygenase isozymes,
J. Clin. Invest. 107 (2001) 1491–1495.
[5] G.N. Levy, Prostaglandin H synthases, nonsteroidal anti-inflammatory drugs,
and colon cancer, FASEB J. 11 (1997) 234–247.
[6] L. Rosenberg, C. Louik, S. Shapiro, Nonsteroidal antiinflammatory drug use
and reduced risk of large bowel carcinoma, Cancer 82 (1998) 2326–2333.
[7] T. Kinoshita, Y. Takahashi, T. Sakashita, H. Inoue, T. Tanabe, T. Yoshimoto,
Growth stimulation and induction of epidermal growth factor receptor by
overexpression of cyclooxygenases 1 and 2 in human colon carcinoma cells,
Biochim. Biophys. Acta 1438 (1999) 120–130.
[8] J. Manosroi, P. Dhumtanom, A. Manosroi, Anti-proliferative activity of
essential oil extracted from Thai medicinal plants on KB and P388 cell lines,
Cancer Lett. 235 (2006) 114–120.
[9] K.C. Chen, C.L. Hsieh, C.C. Peng, H.M. Hsieh-Li, H.S. Chiang, K.D. Huang,
R.Y. Peng, Brain derived metastatic prostate cancer DU-145 cells are
effectively inhibited in vitro by guava (Psidium gujava L.) leaf extracts, Nutr.
Cancer 58 (2007) 93–106.
[10] Y. Deguchi, K. Osada, K. Uchida, H. Kimura, M. Yoshikawa, T. Kudo, H. Yasui,
M. Watanuki, Effects of extract of guava leaves on the development of
diabetes in the db/db mouse and on the postprandial blood glucose of human
subjects, Nippon Nogeikagaku Kaishi 72 (1998) 923–931 (in Japanese).
[11] J.A. Ojewole, Antiinflammatory and analgesic effects of Psidium guajava Linn.
(Myrtaceae) leaf aqueous extract in rats and mice, Methods Find. Exp. Clin.
Pharmacol. 28 (2006) 441–446.
[12] T.T. Mai, N.N. Thu, P.G. Tien, N. Van Chuyen, Alpha-glucosidase inhibitory and
antioxidant activities of Vietnamese edible plants and their relationships
with polyphenol contents, J. Nutr. Sci. Vitaminol. (Tokyo) 53 (2007) 267–276.
[13] G.M. Raso, R. Meli, G. Di Carlo, M. Pacilio, R. Di Carlo, Inhibition of inducible
nitric oxide synthase and cyclooxygenase-2 expression by flavonoids in
macrophage J774A.1, Life Sci. 68 (2001) 921–931.
[14] T.Y. Lau, L.K. Leung, Soya isoflavones suppress phorbol 12-myristate
13-acetate-induced COX-2 expression in MCF-7 cells, Br. J. Nutr. 96 (2006)
169–176.
[15] V. Garcia-Mediavilla, I. Crespo, P.S. Collado, A. Esteller, S. Sanchez-Campos,
M.J. Tunon, J. Gonzalez-Gallego, The anti-inflammatory flavones quercetin
and kaempferol cause inhibition of inducible nitric oxide synthase, cycloox-
ygenase-2 and reactive C-protein, and down-regulation of the nuclear factor
kappaB pathway in Chang liver cells, Eur. J. Pharmacol. 557 (2007) 221–229.
[16] S.Y. Choi, J.H. Hwang, S.Y. Park, Y.J. Jin, H.C. Ko, S.W. Moon, S.J. Kim, Fermented
guava leaf extract inhibits LPS-induced COX-2 and iNOS expression in mouse
macrophage cells by inhibition of transcription factor NF-kappaB, Phytother.
Res. 22 (2008) 1030–1034.
 ARTICLE IN PRESS
Y. Kawakami et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 80 (2009) 239–245
[17] Y. Takahashi, F. Kawahara, M. Noguchi, K. Miwa, H. Sato, M. Seiki, H. Inoue,
T. Tanabe, T. Yoshimoto, Activation of matrix metalloproteinase-2 in human
breast cancer cells overexpressing cyclooxygenase-1 or -2, FEBS Lett. 460
(1999) 145–148.
[18] K. Kadoyama, Y. Takahashi, H. Higashida, T. Tanabe, T. Yoshimoto, Cyclooxygen-
ase-2 stimulates production of amyloid beta-peptide in neuroblastoma x glioma
hybrid NG108-15 cells, Biochem. Biophys. Res. Commun. 281 (2001) 483–490.
[19] S.H. Lee, M.V. Williams, R.N. DuBois, I.A. Blair, Cyclooxygenase-2-mediated
DNA damage, J. Biol. Chem. 280 (2005) 28337–28346.
[20] O. Laneuville, D.K. Breuer, N. Xu, Z.H. Huang, D.A. Gage, J.T. Watson, M. Lagarde,
D.L. DeWitt, W.L. Smith, Fatty acid substrate specificities of human prostaglandin
endoperoxide H synthase-1 and -2, J. Biol. Chem. 270 (1995) 19330–19336.
[21] W.C. Hubbard, A.J. Hough, A.R. Brash, J.T. Watson, J.A. Oates, Metabolism of
linoleic and arachidonic acids in VX2 carcinoma tissue: identification of
monohydroxy octadecadienoic acids and monohydroxy eicosatetraenoic
acids, Prostaglandins 20 (1980) 431–447.
[22] K. Varvas, I. Jarving, R. Koljak, K. Valmsen, A.R. Brash, N. Samel, Evidence of a
cyclooxygenase-related prostaglandin synthesis in coral. The allene oxide
pathway is not involved in prostaglandin biosynthesis, J. Biol. Chem. 274
(1999) 9923–9929.
[23] T. Sakashita, Y. Takahashi, T. Kinoshita, T. Yoshimoto, Essential involvement of
12-lipoxygenase in regiospecific and stereospecific oxidation of low density
lipoprotein by macrophages, Eur. J. Biochem. 265 (1999) 825–831.
[24] M. Claeys, G.A. Kivits, E. Christ-Hazelhof, D.H. Nugteren, Metabolic profile of
linoleic acid in porcine leukocytes through the lipoxygenase pathway,
Biochim. Biophys. Acta 837 (1985) 35–51.
[25] Y. Takahashi, N. Ueda, T. Yoshimoto, S. Yamamoto, C. Yokoyama, A. Miyata,
T. Tanabe, I. Fuse, A. Hattori, A. Shibata, Immunoaffinity purification and
cDNA cloning of human platelet prostaglandin endoperoxide synthase
(cyclooxygenase), Biochem. Biophys. Res. Commun. 182 (1992) 433–438.
[26] T. Kosaka, A. Miyata, H. Ihara, S. Hara, T. Sugimoto, O. Takeda, E. Takahashi,
T. Tanabe, Characterization of the human gene (PTGS2) encoding prostaglandin-
endoperoxide synthase 2, Eur. J. Biochem. 221 (1994) 889–897.
Descargado para Anonymous User (n/a) en Monterrey Institute of Technology and Higher Education de ClinicalKey.es por Elsevier en noviembre 08, 2021.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.
245
[27] O. Folin, W. Denis, A colorimetric method for the determination
of phenols (and phenol derivatives) in urine, J. Biol. Chem. 22 (1915)
305–308.
[28] N.V. Chandrasekharan, H. Dai, K.L.T. Roos, N.K. Evanson, J. Tomsik, T.S. Elton,
D.L. Simmons, COX-3, a cyclooxygenase-1 variant inhibited by acetamino-
phen and other analgesic/antipyretic drugs: cloning, structure, and expres-
sion, Proc. Natl. Acad. Sci. USA 99 (2002) 13926–13931.
[29] A.P. Curnock, P.A. Robson, C.M. Yea, D. Moss, S. Gadher, T.A. Thomson,
R. Westwood, E. Ruuth, R.A. Williamson, Potencies of leflunomide and HR325
as inhibitors of prostaglandin endoperoxide H synthase-1 and -2: comparison
with nonsteroidal anti-inflammatory drugs, J. Pharmacol. Exp. Ther. 282
(1997) 339–347.
[30] T. Banerjee, A. Van der Vliet, V.A. Ziboh, Downregulation of COX-2 and iNOS
by amentoflavone and quercetin in A549 human lung adenocarcinoma cell
line, Prostaglandins Leukot. Essent. Fatty Acids 66 (2002) 485–492.
[31] E. Cheong, K. Ivory, J. Doleman, M.L. Parker, M. Rhodes, I.T. Johnson, Synthetic
and naturally occurring COX-2 inhibitors suppress proliferation in a human
oesophageal adenocarcinoma cell line (OE33) by inducing apoptosis and cell
cycle arrest, Carcinogenesis 25 (2004) 1945–1952.
[32] H.-W. Bai, B.T. Zhu, Strong activation of cyclooxygenase I and II catalytic
activity by dietary bioflavonoids, J. Lipid Res. (2008).
[33] T. Miyamoto, N. Ogino, S. Yamamoto, O. Hayaishi, Purification of prostaglan-
din endoperoxide synthetase from bovine vesicular gland microsomes, J. Biol.
Chem. 251 (1976) 2629–2636.
[34] N. Ogino, S. Ohki, S. Yamamoto, O. Hayaishi, Prostaglandin endoperoxide
synthetase from bovine vesicular gland microsomes. Inactivation and
activation by heme and other metalloporphyrins, J. Biol. Chem. 253 (1978)
5061–5068.
[35] J.H. Capdevila, J.D. Morrow, Y.Y. Belosludtsev, D.R. Beauchamp, R.N. DuBois,
J.R. Falck, The catalytic outcomes of the constitutive and the mitogen
inducible isoforms of prostaglandin H2 synthase are markedly affected by
glutathione and glutathione peroxidase(s), Biochemistry (Mosc) 34 (1995)
3325–3337.