Content uploaded by Paul E Milbury
Author content
All content in this area was uploaded by Paul E Milbury
Content may be subject to copyright.
Bilberry (Vaccinium myrtillus) Anthocyanins Modulate
Heme Oxygenase-1 and Glutathione S-Transferase-pi
Expression in ARPE-19 Cells
Paul E. Milbury,
1
Brigitte Graf,
2
Joanne M. Curran-Celentano,
3
and Jeffrey B. Blumberg
1
PURPOSE. To determine whether anthocyanin-enriched bilberry
extracts modulate pre- or posttranslational levels of oxidative
stress defense enzymes heme-oxygenase (HO)-1 and glutathi-
one S-transferase-pi (GST-pi) in cultured human retinal pigment
epithelial (RPE) cells.
M
ETHODS. Confluent ARPE-19 cells were preincubated with
anthocyanin and nonanthocyanin phenolic fractions of a 25%
enriched extract of bilberry (10
⫺6
–1.0 mg/mL) and, after phe
-
nolic removal, cells were oxidatively challenged with H
2
O
2
.
The concentration of intracellular glutathione was measured
by HPLC and free radical production determined by the dichlo-
rofluorescin diacetate assay. HO-1 and GST-pi protein and
mRNA levels were determined by Western blot and RT-PCR,
respectively.
R
ESULTS. Preincubation with bilberry extract ameliorated the
intracellular increase of H
2
O
2
-induced free radicals in RPE,
though H
2
O
2
cytotoxicity was not affected. By 4 hours, the
extract had upregulated HO-1 and GST-pi protein by 2.8- and
2.5-fold, respectively, and mRNA by 5.5- and 7.1-fold, respec-
tively, in a dose-dependent manner. Anthocyanin and nonan-
thocyanin phenolic fractions contributed similarly to mRNA
upregulation.
C
ONCLUSIONS. Anthocyanins and other phenolics from bilberry
upregulate the oxidative stress defense enzymes HO-1 and
GST-pi in RPE, suggesting that they stimulate signal transduc-
tion pathways influencing genes controlled by the antioxidant
response element. (Invest Ophthalmol Vis Sci. 2007;48:
2343–2349) DOI:10.1167/iovs.06-0452
O
xidative stress is implicated in the pathogenesis of age-
related macular degeneration (AMD), including impair-
ment and death of retinal pigment epithelial (RPE) cells.
1
Hu
-
man studies show that dietary intake and supplementation
with antioxidant nutrients, including

-carotene, vitamins C
and E, and zinc, are associated with reduced risk of AMD.
Anthocyanins are potent antioxidants in vitro
2
and may also
affect visual function in vivo,
3
but little information is available
regarding their action in the retina. The low bioavailability and
cellular concentrations of anthocyanins suggest that their bio-
activity may result more from modulating redox regulation
than from direct quenching of reactive oxygen species (ROS).
4
Anthocyanins are absorbed and excreted both in unmetabo-
lized glycosylated forms
5
and as glucuronidated and methyl
-
ated derivatives.
6
Although little information is available re
-
garding the form, distribution, or retention of anthocyanins in
the retina, data on glycosylated anthocyanin activity in RPE
cells in vitro may provide a biological plausibility for its puta-
tive role in promoting visual function. Bilberries, an anthocya-
nin-rich fruit, have long been used in traditional herbal thera-
pies for the treatment of eye disorders. Recently, Fursova et al.
7
found that supplementation with bilberry extract decreased
serum and retinal lipid peroxides and slowed the development
of cataract and macular degeneration in OXYS rats, a hyper-
tensive strain that presents with a shortened lifespan and early
phenotypes of age-related disorders.
Upregulation of stress proteins is a universal protective
response to adverse conditions, including oxidative stress.
8
Heme-oxygenase (HO)-1, a heat-shock protein (Hsp32), cata-
lyzes the rate-limiting first step in heme catabolism to carbon
monoxide (CO), free Fe
2⫹
, and biliverdin that is converted to
the antioxidant bilirubin by biliverdin reductase.
9
Bilirubin also
modulates cell signal transduction pathways relevant to inflam-
mation.
9
Free Fe
2⫹
rapidly induces ferritin expression and the
ATPase Fe
2⫹
-secreting pump to decrease Fe
2⫹
, thereby limit
-
ing oxidative damage created via the Fenton reaction.
10
CO,
the third product of HO-1 activity, stimulates cell signaling
similar to nitric oxide, but absent its radical activity. For exam-
ple, CO mediates vasodilation, inhibits platelet aggregation,
and suppresses cytokine production, all factors associated with
the amelioration of AMD pathophysiology.
11
Thus, HO-1 induc
-
tion may confer protection in the retina by increasing resis-
tance to oxidative stress, inflammation, and apoptosis.
12
Two HO isoforms, inducible HO-1 and constitutive HO-2,
have been found in human RPE cells.
13
Exposure to light
upregulates retinal HO-1 and HO-1 protein is increased in RPE
by neovascular AMD.
14
Although the impact of these changes
in RPE is not fully elucidated, HO-1 upregulation is generally
thought to protect the cells.
15
HO-1 shares genomic regulatory
mechanisms with other protective enzymes, including gluta-
thione S-transferase (GST) and NAD(P)H:quinone oxidoreduc-
tase (NQO1), enzymes that detoxify byproducts of oxidative
stress.
16
Inducing enzymes involved in glutathione (GSH) bio
-
synthesis also enhances cellular antioxidant defenses.
17
Age-
related declines in GSH are associated with increased risk of
AMD.
18
The purpose of this study was to investigate the effects
of anthocyanin and other phenolic compounds from bilberry
(Vaccinium myrtillus) on quiescent and oxidatively stressed
RPE cells in vitro.
From the
1
Antioxidants Research Laboratory, Jean Mayer USDA
Human Nutrition Research Center on Aging, Tufts University, Boston,
Massachusetts; and the
3
Department of Animal and Nutritional Sci
-
ences, University of New Hampshire, Durham, New Hampshire.
2
Current affiliation: Unilever Food and Health Research Institute,
Vlaardingen, The Netherlands.
Supported by the USDA Agricultural Research Service under Co-
operative. Agreement No. 58-1950-4-401. The contents of this publica-
tion do not necessarily reflect the views or policies of the USDA nor
does mention of trade names, commercial products, or organizations
imply endorsement by the U.S. Government.
Submitted for publication April 20, 2006; revised July 11 and
October 18, 2006, and January 11, 2007; accepted March 9, 2007.
Disclosure: P.E. Milbury, None; B. Graf, None; J.M. Curran-
Celentano, None; J.B. Blumberg, None
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be marked “advertise-
ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Corresponding author: Paul E. Milbury, Antioxidants Research
Laboratory, Jean Mayer USDA Human Nutrition Research Center on
Aging, Tufts University, 711 Washington Street, Boston, MA 02111;
paul.milbury@tufts.edu.
Investigative Ophthalmology & Visual Science, May 2007, Vol. 48, No. 5
Copyright © Association for Research in Vision and Ophthalmology
2343
MATERIALS AND METHODS
RPE Cell Culture
ARPE-19 cells were obtained from the American Type Culture Collec-
tion (Manassas, VA) and propagated according to methods described
by Dunn et al.
19
Cells were maintained in Dulbecco’s modified Eagle’s
medium and Ham’s F12 medium, supplemented as described with the
exception that, after confluence, fetal bovine serum was decreased to
5% to promote differentiation.
Bilberry Extract
An extract of bilberry (25% anthocyanin enriched) used commercially
in dietary supplements was a gift from Artemis International Inc. (Fort
Wayne, IN). Concentrations of the extract are expressed as dry weight
per milliliter of medium.
Total Phenols Assay and Total
Anthocyanins Assay
Total phenols were determined colorimetrically by the Folin-Ciocalteu
assay.
20
Total anthocyanins were estimated by the pH differential
absorbance method.
21
Absorbance was measured in a spectrophotom
-
eter (UV-1601; Shimadzu, Columbia, MD) and results expressed as
grams cyanidin-3-glucoside equivalents (CGE) per 100 grams dry
weight.
Anthocyanin Chromatography
To determine whether the observed effects of bilberry were due to
anthocyanins or to other phenolic constituents of the bilberry, the
extract was fractionated by semipreparative methanolic gradient elu-
tion from a reversed-phase C18, silica bonded, low-pressure chroma-
tography column (15 ⫻ 300 mm). Fractions were analyzed and com-
bined into two pools, one containing only anthocyanins and the other
containing all the remaining phenolic compounds from the extract.
Anthocyanin components were analyzed by HPLC with electrochemi-
cal detection (ECD; ESA Inc., Chelmsford, MA) using anthocyanin
standards as described by Milbury
22
with identities verified by liquid
chromatography-tandem mass spectrometry (LC-MS/MS) using an
HPLC system (model 1100; Agilent, Palo Alto, CA) fitted with a pho-
todiode array (PDA; UV G1315A; Agilent) and a Bruker ion trap MS/MS
detector and electrospray interface (Esquire; Brucker Daltonics, Bil-
lerica, MA).
Cell Treatment and Viability Assays
The 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyl tetrazolium bromide
(MTT) assay, an assays of early mitochondrial dysfunction, was used to
determine cell viability.
23,24
MTT was obtained from Invitrogen-Molec
-
ular Probes, Inc. (Eugene, OR).
Briefly, cells grown in 96-well plates were washed twice with PBS
before treatment. Cells were exposed to serum-free medium with or
without bilberry extract for 4 hours, rinsed free of bilberry medium,
and then challenged with either control media or H
2
O
2
for 2 hours.
Stock solutions of H
2
O
2
(30%; Fisher Scientific, Pittsburgh, PA) or
bilberry extracts were made in water and final dilutions prepared in
serum-free medium (without phenol red) immediately before the ex-
periments. After treatment, cells were allowed to recover in phenol
red–free preconditioned medium for 1 hour. MTT was added during
the entire recovery period. When viability was assessed 24 hours after
treatment, MTT was incorporated for the last hour. Absorbance and
fluorescence measurements were performed on a plate reader (Fluo-
Star Optima; BMG Labtech GmbH, Offenburg, Germany). Once pre-
liminary studies demonstrated clear protein upregulation by 24 hours
after treatment and detectable upregulation by 4 hours, cells were
harvested at 4 hours after treatment for both protein and mRNA
determinations. Butylated hydroxytoluene (BHT) and

-naphthofla-
vone (BNF) were obtained from Sigma-Aldrich (St. Louis, MO).
Dichlorofluorescein Assay
Intracellular free radical production was determined using the dichlo-
rofluorescein (DCF) assay described by Sohn et al.
25
DCFH-DA (2⬘,7⬘-
dichlorofluorescin-diacetate) was acquired from Invitrogen-Molecular
Probes, Inc. (Eugene, OR). After fluorescence measurements were
completed, excess H
2
O
2
(500 mM) was added to assess maximum
assay fluorescence and assure equivalent DCFH-DA loading.
Protein Analysis by Western Blot
HO-1, GST-pi, and

-actin protein levels were determined by Western
blot analysis. Extraction reagent (CytoBuster Protein; Novagen-Calbio-
chem, La Jolla, CA) containing protease inhibitors (Protease Inhibitor
Cocktail Set III; Calbiochem, La Jolla, CA) was used to extract protein
and samples were stored at ⫺80°C until electrophoresis.
Protein concentration was determined by the bicinchoninic (BCA)
protein assay (Pierce Biotechnology Inc., Rockford, IL). After resolu-
tion, proteins were electrotransferred onto polyvinylidene difluoride
(PVDF) membranes (Invitrogen, Carlsbad, CA). Primary anti-human
HO-1 and GST-pi IgG1 mouse isotypes were obtained from BD Bio-
science (San Jose, CA) and housekeeping gene

-actin mouse mono-
clonal antibody was obtained from Novus Biologicals (Littleton, CO).
Visualization was accomplished using horseradish peroxidase-conju-
gated sheep anti-mouse IgG (GE Healthcare, Arlington Heights, IL) and
chemiluminescent substrates (SuperSignal West; Pierce). Exposed x-ray
film was quantified by densitometry analysis (Quantity One software;
Bio-Rad Laboratories, Hercules, CA).
RNA Extraction and RT-PCR Analysis
Cell mRNA was extracted (RNeasy Kits; Qiagen Inc., Valencia, CA) and
stored at ⫺80°C until quantitative RT-PCR analysis. A reverse transcrip-
tion kit (Script III; Invitrogen Corp., Carlsbad, CA) was used to synthe-
size cDNA for amplification. RT-PCR was conducted in a PCR system
(Prism 7000; Applied Biosystems, Foster City, CA) using certified
housekeeping fluorogenic primers (GAPDH and

-actin) and custom
gene primers (D-LUX, Invitrogen) described in Table 1. Data analysis
was performed using the 2
⫺⌬⌬CT
method
26
normalized to GAPDH
mRNA and expressed relative to the control subjects.
Data Analysis and Statistics
Data are expressed as a percentage of control data or as increases
(x-fold) over values obtained under control conditions and are pre-
sented as mean ⫾ SD of results in three or more independent exper-
TABLE 1. Heme-oxygenase and Glutathione S-transferase-pi Lux Primer Definitions and Sequences
Gene Primer Strand 3ⴕ Loc Sequence
HO-1 X06985.1_633RL Reverse 633 cgcatATCTCCAGGGAGTTCATGcG
HO-1 X06985.1_633RL/611FU Forward 611 ACATTGCCAGTGCCACCAAG
GST-pi NM_000852.2_135FL Forward 135 cggtcGAAGGAGGAGGTGGTGACcG
GST-pi NM_000852.2_135FL/155RU Reverse 155 TAGGCAGGAGGCTTTGAGTGAG
GAPDH NM_002046 Certified LUX Primer Set*

-Actin NM_001101 Certified LUX Primer Set*
* Invitrogen, Carlsbad, CA.
2344 Milbury et al. IOVS, May 2007, Vol. 48, No. 5
iments, each performed using triplicate cell culture plates with a
minimum of eight wells per condition. Statistical analysis was per-
formed using ANOVA followed by the Dunnett or Bonferroni tests. P ⱕ
0.05 was considered to show statistic significance.
RESULTS
The bilberry extract contained 28.1% CGE as determined by
pH differential spectroscopic analysis of anthocyanins.
HPLC-UV (photodiode array) and HPLC-ECD analysis indicated
total anthocyanin content as 27.7% and 27.9% CGE, respec-
tively. The total phenols content was 61.8 g catechin equiva-
lents/100 g. Fifteen anthocyanins were identified by HPLC-ECD
and by LC-MS/MS, using authentic standards of delphinidin
3-galactoside, delphinidin 3-glucoside, cyanidin 3-galactoside,
delphinidin 3-arabinoside, cyanidin 3-glucoside, petunidin
3-galactoside, cyanidin 3-arabinoside, petunidin 3-glucoside,
peonidin 3-galactoside, petunidin 3-arabinoside, peonidin
3-glucoside, malvidin 3-galactoside, peonidin 3-arabinoside,
malvidin 3-glucoside, and malvidin 3-arabinoside. Using low-
pressure, semipreparative chromatography, we made sample
pool from fractions 55 to 70 that contained polyphenols and
phenolic acids but was free of anthocyanins. The constituent
phenolics included trans-resveratrol, ferulic acid, p-coumaric
acid, quercetin, and caffeic acid; chlorogenic acid, which was
the most abundant phenolic acid, was present at approxi-
mately 60
g/g in the original extract. A sample pool created
from fractions 70 to 75 included all anthocyanins present in the
bilberry extract.
After growing for 10 days, confluent ARPE-19 cells dis-
played the cobblestone morphology described by Dunn et al.
19
and exhibited a robust defense against H
2
O
2
with a 50% lethal
dose (LD
50
)of⬃500
MH
2
O
2
. Treatment of ARPE-19 cells
grown for ⬍10 days with 500
MH
2
O
2
resulted in ⬎70% cell
death as illustrated in 9-day-old cultures (Fig. 1). At 3 days in
culture, although cells appeared confluent microscopically, the
LD
50
for H
2
O
2
by the MTT assay was 41⫾14
M.
The effect of bilberry preincubation on H
2
O
2
-induced RPE
cell cytotoxicity is illustrated in Figure 1. In 9-day cultures,
lower dose challenges of H
2
O
2
showed stimulation of addi
-
tional proliferation that was inhibited by bilberry extract. Bil-
berry extract preincubation with doses between 10
⫺5
and 1.0
mg/mL did not shift viability curves to the right (i.e., increase
resistance to H
2
O
2
). Similar experiments performed at 10 days
in culture also revealed no shift in viability curves, suggesting
no increase in ARPE19 cell resistance to H
2
O
2
-induced cyto
-
toxicity (data not shown).
DCF assays were conducted to determine whether bilberry
could be transported into RPE cells to affect intracellular rad-
ical production. Preincubation with 10
⫺5
and 10
⫺1
mg/mL
bilberry extract decreased intracellular radical formation by
18% and 65%, respectively (Fig. 2). These results suggest that
bilberry preincubation reduced intracellular radicals by radical
quenching or modulation of cellular redox status to near basal
levels; nevertheless, treatment with 500
MH
2
O
2
induced
cellular damage sufficient to cause cell death even in the
presence of intracellular anthocyanins (Fig. 1).
When assessed at 24 hours after treatment, no increase in
HO-1 protein expression was observed in 10-day RPE cells
exposed to 200
MH
2
O
2
for 2 hours, suggesting that antiox
-
idant defenses were adequate. In contrast, cells exposed to 500
MH
2
O
2
had a 10-fold increase in HO-1 protein expression 24
hours after treatment (Figs. 3A, 3B), a reflection of the induc-
tion of oxidative stress. Upregulation of HO-1 protein was not
prevented by a 4-hour preincubation with 1.0 mg/mL bilberry
extract before H
2
O
2
exposure. Increased HO-1 protein expres
-
sion was observed in both control cells and those exposed to
200
MH
2
O
2
(Fig. 3B), demonstrating that bilberry alone can
upregulate HO-1. HO-1 protein levels were then assessed im-
mediately after a 4-hour exposure to bilberry extract (10
⫺6
–1.0
0
20
40
60
80
100
120
0 10 50 100 500 1000
H
2
O
2
(µM)
Viability by MTT Reduction
24h After H
2
O
2
Exposure
(Percent of Control)
0 Extract
1 mg/mL
10-1 mg/mL
10-2 mg/mL
10-3 mg/mL
10-4 mg/mL
10-5 mg/mL
Bilberry Extract
∗
∗
0
20
40
60
80
100
120
0 10 50 100 500 1000
H
2
O
2
(µM)
Viability by MTT Reduction
24h After H
2
O
2
Exposure
(Percent of Control)
0 Extract
1 mg/mL
10-1 mg/mL
10-2 mg/mL
10-3 mg/mL
10-4 mg/mL
10-5 mg/mL
Bilberry Extract
∗
∗
FIGURE 1. H
2
O
2
-induced cell death in RPE cells grown 9 days in
culture is not prevented by preincubation with bilberry extract. RPE
cells were either untreated or preincubated for 4 hours with bilberry
extract (10
⫺5
–1.0 mg/mL) in serum-free medium. The cells were chal
-
lenged with H
2
O
2
for 2 hours before being rinsed with serum-free
medium and left to recover in medium containing 5% serum for 24
hours. During the last hour of the recovery period, MTT was added to
the media to assess cell viability. Cell tolerance to H
2
O
2
varied in this
assay, depending on the time in culture of the RPE cells. Cells cultured
for 3, 9, 10, or 35 days exhibited H
2
O
2
LD
50
sof41⫾ 14, 342 ⫾ 15,
566 ⫾ 14, and 583 ⫾ 14
M, respectively. Low levels of H
2
O
2
demonstrated mild proliferative stimulation in 9-day cultures that was
not significant in cultures grown 10 days or more. *P ⬍ 0.05 versus
control (no H
2
O
2
). All data points at 500 and 1000
MH
2
O
2
are
decreased (P ⬍ 0.05) versus all control cultures. Data represent the
mean, n ⫽ 8 cultures for each condition. Standard deviation bars are
omitted for clarity.
0
20
40
60
80
100
120
0 0.00001 0.001 0.1
Bilberry (mg/ml)
DCF Fluorescence
(% of Control)
a
a
a b
Bilberry Extract (mg/ mL )
DCF Fluorescence
(% of H2O2 Treated Control)
0
20
40
60
80
100
120
0
10
-5
Bilberry (mg/ml)
DCF Fluorescence
Bilberry Extract (mg/mL)
DCF Fluorescence
(% of H
2
O
2
Treated Control)
10
-3
10
-1
?
?
?
0
20
40
60
80
100
120
0 0.00001 0.001 0.1
Bilberry (mg/ml)
DCF Fluorescence
(% of Control)
a
a
a b
Bilberry Extract (mg/ mL )
DCF Fluorescence
(% of H2O2 Treated Control)
0
20
40
60
80
100
120
0
10
-5
Bilberry (mg/ml)
DCF Fluorescence
Bilberry Extract (mg/mL)
DCF Fluorescence
(% of H
2
O
2
Treated Control)
10
-3
10
-1
*
*
*
FIGURE 2. Intracellular radical quenching in 10-day cultured RPE cells
treated with bilberry extract assessed by the DCF assay. RPE cells were
either untreated or were preincubated for 4 hours with 0, 10
⫺5
,10
⫺3
or 10
⫺1
mg/mL bilberry extract in serum-free medium. The cells were
then challenged with 500
MH
2
O
2
for 2 hours before being rinsed
with PBS and loaded with DCF for 30 minutes before the fluorescence
was read. Cells not treated with H
2
O
2
showed a basal intracellular DCF
fluorescence level that was approximately 30% of the level measured in
cells treated with 500
MH
2
O
2
and not significantly different from the
H
2
O
2
-treated cells that were preincubated with 10
⫺1
mg/mL bilberry
extract. *P ⬍ 0.05 versus control cells (no bilberry preincubation and
treated with H
2
O
2
). †P ⬍ 0.05 versus cells preincubated with 10
⫺5
and
10
⫺3
mg/mL bilberry extract and then treated with H
2
O
2
. Data repre
-
sent the mean ⫾ SD of results in three independent experiments each
performed with triplicate wells.
IOVS, May 2007, Vol. 48, No. 5 Bilberry and Oxidative Stress Defense 2345
mg/mL). Collected data from different experiments show a
dose-dependent increase in HO-1 protein with bilberry treat-
ment (Fig. 4).
BHT, an antioxidant response element (ARE) agonist, was
used as a positive control at 300 and 1000
M and upregulated
HO-1 protein expression by 1.6- and 2.3-fold at 4 hours, re-
spectively (Fig. 5). Although the increase induced by 300
M
BHT was not statistically significant in the Western blot analysis
of GST-pi protein expression, bilberry extract and H
2
O
2
in
-
duced significant increases (Fig. 6).
RT-PCR analysis showed upregulation of HO-1 and GST-pi
mRNA by bilberry extract (Fig. 7). In addition, BNF, a more
potent ARE agonist than BHT, also induced increases in HO-1
and GST-pi mRNA (Fig. 7). BHT (300
M) significantly in-
creased GST-pi mRNA, while an upward trend was observed
for HO-1 mRNA. In other experiments, anthocyanin and poly-
phenolics–phenolic acid pools from the bilberry extract, as
well as 300
M BHT, induced significant increases in HO-1
mRNA. The bilberry component pools showed comparable
potencies (Fig. 8), which appear to account for the mRNA
levels observed with the intact extract (Fig. 7).
DISCUSSION
HO-1 and GST-pi are found in the retina and can be induced as
a defense mechanism against oxidative stress.
8
HO-1 and
GST-pi mRNA and protein levels are elevated by stressors, such
as oxidant and xenobiotic compounds, and are protective of
human neuronal and RPE cells in culture.
27,28
HO-1 is upregu
-
lated in rat retina exposed to visible light and is present at
higher levels during daylight hours, suggesting a protective
role in limiting light-induced radical damage.
29
With advanced
age and in AMD, the ability to upregulate HO-1 and GST-pi
diminishes and oxidative stress increases in human RPE.
14,30
Further evidence of an inverse relationship between HO-1
expression and oxidative stress is shown in Asian Indian pa-
tients with type 2 diabetes. These patients have increased
oxidative damage and microangiopathy that is associated with
increased NADPH oxidase expression and lower HO-1 gene
expression than levels found in a comparable healthy popula-
FIGURE 3. (A, B) Upregulation of HO-1 protein expression by 500
M
H
2
O
2
and by 1.0 mg/mL bilberry extract. RPE cells cultured for 10 days
after plating were preincubated for 4 hours with either serum-free
medium along or containing 1.0 mg/mL extract. The medium contain-
ing extract was twice washed from the cells with fresh medium, and
then the cells were incubated an additional 2 hours with serum- and
phenol-free medium containing either 200 or 500
MH
2
O
2
. The cells
were then refed medium containing 2% serum for 24 hours and the
protein extracts harvested. Protein was separated by SDS-PAGE fol-
lowed by immunoblot analysis using anti-human HO-1 and anti-human

-actin antibodies. Quantification was performed by densitometry anal-
ysis, and the values were adjusted to the corresponding

-actin values
obtained from the same gel. The results are expressed as relative units.
*P ⬍ 0.05 versus control cells. †P ⬍ 0.05 versus cells treated with 200
MH
2
O
2
. Values are mean ⫾ SD of results of three independent
experiments, each performed with three or more cultures.
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
1
(Fold Increase Over Control)
R
2
=0.6056
P<0.0001
0
1
2
3
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
1
HO-1 Protein
4 h After Bilberry Extract Exposure
0
1
2
3
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
1
Bilberry Extract
(mg/mL)
R
2
=0.6056
P<0.0001
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
1
(Fold Increase Over Control)
R
2
=0.6056
P<0.0001
0
1
2
3
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
1
HO-1 Protein
4 h After Bilberry Extract Exposure
0
1
2
3
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
1
Bilberry Extract
(mg/mL)
R
2
=0.6056
P<0.0001
FIGURE 4. Dose–response effect of bilberry extract on HO-1 protein
expression, as determined 4 hours after exposure to bilberry extract.
Each depicted data point represents the mean of at least three data
points (usually eight wells) for a given dose of extract, and each point
is collected from independent culture experiments. The scattergram
indicates a dose-dependent increase in HO-1 protein in response to
bilberry treatment, determined by Western blot analysis.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
HO-1 Protein
(Fold Increase Over Control)
(mg/ml)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Bilberry Extract
10
-3
10
-2
10
-1
Control
300 1000 500
BHT
(
µ
M)
H
2
O
2
(
µ
M)
∗
∗
∗
∗
∗
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
HO-1 Protein
(Fold Increase Over Control)
(mg/ml)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Bilberry Extract
10
-3
10
-2
10
-1
Control
300 1000 500
BHT
(
µ
M)
H
2
O
2
(
µ
M)
∗
∗
∗
∗
∗
FIGURE 5. Increased HO-1 protein by Western blot analysis in re-
sponse to treatment for 4 hours with media containing bilberry extract,
300 or 1000
M BHT, or 500
MH
2
O
2
.*P ⬍ 0.01 versus control. Data
represent results of a typical experiment (mean ⫾ SD), with five wells
used for each condition.
2346 Milbury et al. IOVS, May 2007, Vol. 48, No. 5
tion.
31
Similarly, RPE of control donor eyes exhibited 20-fold
higher HO-1 mRNA than that of donors with diabetes, suggest-
ing that diabetics have an impairment of endogenous defense
mechanisms and a vulnerability to oxidative stress, especially
in the neuroretinal cells.
32
There is a heterogeneous mosaic pattern and an age-related
decline in HO-1 protein and gene expression in the retina.
14,30
Despite general age-related declines in gene expression, trans-
lation, and transcription, the levels of most enzymes and pro-
teins remain relatively constant.
33
This paradox has been at
-
tributed to age-associated declines in protein degradation that
could result in an accumulation of damaged proteins and a
decrease in the response of inducible enzymes to stimuli.
34
Potentially, interventions that stimulate upregulation of HO-1
or influence glutathione metabolism may be useful in the
prevention or treatment of age-related retinopathies such as
AMD.
The 12th-century German herbalist Hildegarde von Bingen
(1098–1179 CE) indicated the use of bilberry for the treatment
of eye disorders.
35
Although bilberry has not been included as
an ingredient in clinical trials of AMD, bilberry extracts have
shown some efficacy in preclinical studies as an antioxidant,
anti-inflammatory, vasoprotectant, hypoglycemic, and lipid-
lowering agent—actions relevant to risk factors for AMD. Of
note, using rodent models, Joseph et al.
36
found that diets rich
in blueberries, an anthocyanin-rich Vaccinium berry related to
bilberries, significantly reversed age-related declines in neuro-
nal signal transduction, cognition, and motor behavioral defi-
cits. While investigations of anthocyanins in human studies are
limited, approximately 30 trials published during the past 45
years suggest that they possess activity regarding visual adjust-
ments to light.
37
Although night vision improvements were
found in eight studies, four randomized controlled trials
yielded null outcomes. Nonetheless, the effect of bilberry on
visual function in subjects with impaired night vision or eye
disease has yet to be investigated.
Our results show that 3-day, undifferentiated, confluent RPE
cells can be stimulated to divide further with a modest pro-
oxidant challenge. These cells are less resistant to higher con-
centrations of H
2
O
2
than is more developed RPE cells. These
findings are consistent with Bailey et al.
38
and Wada et al.
39
who also found increased resistance to oxidative stress with
increasing age and differentiation in ARPE-19 cells. We found
little additional antioxidant defense benefit by extending cul-
tures beyond 10 to 12 days to several weeks in culture. These
observations underscore the importance of culture conditions
for RPE cells to defining results that are dependent on the state
of differentiation.
Alizadeh et al.
40
demonstrated that the oxidants tert-butyl
hydroperoxide (tBH) and H
2
O
2
upregulates HO-1 and fibro
-
blast growth factor receptor (FGFR1) mRNA in differentiated
ARPE-19 cells with concurrent downregulation of specific
genes, including cellular retinaldehyde-binding protein (CRALBP)
0
1
2
3
4
5
6
10
-2
10
-1
Control
300
500
(mg/
Bilberry Extract BHT
(
µ
M)
H
2
O
2
(
µ
M)
GST-pi Protein
(Fold Increase Over Control)
∗
∗
∗
0
1
2
3
4
5
6
10
-2
10
-1
Control
300
500
(mg/mL)
Bilberry Extract BHT
(
µ
M)
H
2
O
2
(
µ
M)
GST-pi Protein
(Fold Increase Over Control)
∗
∗
∗
FIGURE 6. Increase in GST-pi protein by Western blot analysis in
response to treatment for 4 hours with media containing bilberry
extract (10
⫺2
and 10
⫺1
mg/mL), 300
M BHT, or 500
MH
2
O
2
.*P ⬍
0.01 versus control. Data represent a typical experiment showing
mean ⫾ SD with four wells for each condition.
FIGURE 7. Effects of bilberry extract on HO-1 and GST-pi mRNA in
RPE cultures. Cells were preincubated with increasing doses ranging
from 10
⫺3
to 10
⫺1
mg/mL bilberry extract, 300
M BHT, or 10
M BNF
for 4 hours. Total RNA was extracted and analyzed by RT-PCR for
human HO-1 mRNA and GST-pi and then normalized to

-actin or
GAPDH. Data are expressed as x-fold increases over untreated controls
using the ⌬⌬
CT
method. *P ⬍ 0.01 compared with control. Data are the
mean ⫾ SD of results in three independent experiments, each per-
formed in three or more cultures.
FIGURE 8. HO-1 mRNA upregulation in RPE cells in response to treat-
ment with a pool of anthocyanin fractions (70 –75) or a pool of
phenolic acids and other compounds (fractions 55–70) from bilberry
extract. RPE cells grown 10 days in culture were preincubated with the
respective pools at concentrations of phenolics or anthocyanins rep-
resenting the original amounts in 10
⫺3
,10
⫺2
, and 10
⫺1
mg/mL extract
for 4 hours. Total RNA was extracted and RT-PCR conducted for
human HO-1 mRNA and housekeeping gene GADPH mRNA. HO-1
mRNA was normalized to GAPDH and expressed relative to that of
untreated control culture samples using the ⌬⌬
CT
method. *P ⬍ 0.01
compared to control. Each pool contributed approximately half of the
upregulation observed with the complete mixture in Figure 7. Data are
mean ⫾ SD of results in three independent experiments, each per-
formed in three or more cultures.
IOVS, May 2007, Vol. 48, No. 5 Bilberry and Oxidative Stress Defense 2347
and the RPE-specific 65-kDa protein (RPE65). Similarly, in our
study, low doses of H
2
O
2
induced a proliferative response in
ARPE-19 cultures where nonquiescent cells may remain (9
days), whereas higher doses induced cell death. It is well
established that markedly different responses to oxidative
stress by proliferating and nonproliferating mammalian cells
are dependent on the degree of cellular differentiation and on
the magnitude of the stress stimulus.
41,42
In this regard, al
-
though the threshold of viability to the H
2
O
2
challenge in fully
differentiated RPE cells in our study was relatively high, the
slope of the viability curves were quite steep. These experi-
ments show that bilberry anthocyanins can be preloaded into
cells and quench intracellular radicals, albeit at supraphysi-
ologic levels; however, this radical quenching activity was
insufficient to prevent cell death caused by a 500
MH
2
O
2
challenge. High levels of H
2
O
2
may render catastrophic oxi
-
dant damage to surface proteins and receptors representing an
insult disproportionate to in vivo conditions. Lethal cellular
injury mediated by acute H
2
O
2
-induced oxidative stress in
-
volves an elevation of intracellular cytosolic Ca
2⫹
associated
with collapse of the mitochondrial membrane potential. The
viability assay chosen for these experiments is an indicator of
mitochondrial dysfunction leading to failure of reductive po-
tential in the cell. Sequestration of Ca
2⫹
in the mitochondrion
depends on NADPH and GSH, both of which are consumed in
the metabolism of H
2
O
2
. GSH was depleted rapidly by 500
M
H
2
O
2
in our 10-day RPE cultures (data not shown). Cata
-
strophic oxidative defense failure, observed in in vitro exper-
iments where it is necessary to use high levels of H
2
O
2
, may
not adequately mimic in vivo oxidative stress associated with
aging. These experiments should be repeated in an optimal in
vivo model over a longer time period with lower levels of
oxidative stress to test the impact of long-term nutrient intakes
on upregulation of protective enzymes and potential reduction
in risk of chronic diseases like AMD.
Hanneken et al.
43
showed that, unlike other flavonoids,
cyanidin, malvidin, and peonidin (all at 50
M) were ineffec-
tive in protecting 1-day ARPE-19 cells from 250
MH
2
O
2
. They
also found that other flavonoids, including quercetin, fisetin,
and galangin, protect retinal ganglion and RPE cells by stimu-
lation of ARE responsive genes.
43,44
Neither anthocyanindins
nor anthocyanins were tested for the ability to upregulate HO-1
or GST-pi. Delphinidin and cyanidin do influence expression of
HO-1 in human vascular endothelial cells in vitro,
45
but it is
unlikely that these aglycones (anthocyanidins) are a predomi-
nant form in vivo.
6
To our knowledge, this is the first study
showing modulation of oxidative stress defense enzymes HO-1
and GST-pi by bilberry anthocyanins in RPE in vitro, suggesting
the need for confirmatory studies in situ and in vivo.
References
1. Cai J, Nelson KC, Wu M, Sternberg P Jr, Jones DP. Oxidative
damage and protection of the RPE. Prog Retin Eye Res. 2000;19:
205–221.
2. Rice-Evans CA, Miller NJ, Paganga G. Structure–antioxidant activity
relationships of flavonoids and phenolic acids. Free Radic Biol
Med. 1996;20:933–956.
3. Matsumoto H, Nakamura Y, Tachibanaki S, Kawamura S, Hirayama
M. Stimulatory effect of cyanidin 3-glycosides on the regeneration
of rhodopsin. J Agric Food Chem. 2003;51:3560 –3563.
4. Kong JM, Chia LS, Goh NK, Chia TF, Brouillard R. Analysis and
biological activities of anthocyanins. Phytochemistry. 2003;64:
923–933.
5. Milbury PE, Cao G, Prior RL, Blumberg J. Bioavailability of elder-
berry anthocyanins. Mech Ageing Dev. 2002;123:997–1006.
6. Kay CD, Mazza GJ, Holub BJ. Anthocyanins exist in the circulation
primarily as metabolites in adult men. J Nutr. 2005;135:2582–
2588.
7. Fursova AZh, Gesarevich OG, Gonchar AM, Trofimova NA, Ko-
losova NG. Dietary supplementation with bilberry extract prevents
macular degeneration and cataracts in senesce-accelerated OXYS
rats. Adv Gerontol. 2005;16:76 –79.
8. Jaattela M. Heat shock proteins as cellular lifeguards. Ann Med.
1999;31:261–271.
9. Stocker R, Yamamoto Y, McDonagh AF, Glazer AN, Ames BN.
Bilirubin is an antioxidant of possible physiological importance.
Science. 1987;235:1043–1046.
10. Baranano DE, Wolosker H, Bae BI, Barrow RK, Snyder SH, Ferris
CD. A mammalian iron ATPase induced by iron. J Biol Chem.
2000;275:15166–15173.
11. Wu L, Wang R. Carbon monoxide: endogenous production, phys-
iological functions, and pharmacological applications. Pharmacol
Rev. 2005;57:585– 630.
12. Kapitulnik J. Bilirubin: an endogenous product of heme degrada-
tion with both cytotoxic and cytoprotective properties. Mol Phar-
macol. 2004;66:773–779.
13. Frank RN, Amin RH, Puklin JE. Antioxidant enzymes in the macular
retinal pigment epithelium of eyes with neovascular age-related
macular degeneration. Am J Ophthalmol. 1999;127:694–709.
14. Miyamura N, Ogawa T, Boylan S, Morse LS, Handa JT, Hjelmeland
LM. Topographic and age-dependent expression of heme oxygen-
ase-1 and catalase in the human retinal pigment epithelium. Invest
Ophthalmol Vis Sci. 2004;45:1562–1565.
15. Ryter SW, Alam J, Choi AM. Heme oxygenase-1/carbon monoxide:
from basic science to therapeutic applications. Physiol Rev. 2006;
86:583–650.
16. Nguyen T, Sherratt PJ, Pickett CB. Regulatory mechanisms control-
ling gene expression mediated by the antioxidant response ele-
ment. Annu Rev Pharmacol Toxicol. 2003;43:233–260.
17. Hayes JD, McLellan LI. Glutathione and glutathione-dependent
enzymes represent a co-ordinately regulated defence against oxi-
dative stress. Free Radic Res. 1999;31:273–300.
18. Samiec PS, Drews-Botsch C, Flagg EW, et al. Glutathione in human
plasma: decline in association with aging, age-related macular
degeneration, and diabetes. Free Radic Biol Med. 1998;24:699 –
704.
19. Dunn KC, Aotaki-Keen AE, Putkey FR, Hjelmeland LM. ARPE-19, a
human retinal pigment epithelial cell line with differentiated prop-
erties. Exp Eye Res. 1996;62:155–169.
20. Singleton VL, Orthofer R, Lamuela-Ravent RM. Analysis of total
phenols and other oxidation substrates and antioxidants by means
of Folin-Ciocalteu reagent. Methods Enzymol. 1999;299:152–178.
21. Cheng GW, Breen PJ. Activity of phenylalanine ammonia-lyase
(PAL) and concentrations of anthocyanins and phenolics in devel-
oping strawberry fruit. J Am Soc Hort Sci. 1991;116:865– 869.
22. Milbury PE. Analysis of complex mixtures of flavonoids and poly-
phenols by high-performance liquid chromatography electrochem-
ical detection methods. Methods Enzymol. 2001;335:15–26.
23. Ballinger SW, Van Houten B, Jin GF, Conklin CA, Godley BF.
Hydrogen peroxide causes significant mitochondrial DNA damage
in human RPE cells. Exp Eye Res. 1999;68:765–772.
24. Vrabec JP, Lieven CJ, Levin LA. Cell-type-specific opening of the
retinal ganglion cell mitochondrial permeability transition pore.
Invest Ophthalmol Vis Sci. 2003;44:2774 –2782.
25. Sohn JH, Han KL, Lee SH, Hwang JK. Protective effects of pandu-
ratin A against oxidative damage of tert-butylhydroperoxide in
human HepG2 cells. J Biol Pharm Bull. 2005;28:1083–1086.
26. Livak KJ, Schmittgen TD. Analysis of relative gene expression data
using real-time quantitative PCR and the 2(-Delta Delta C(T))
method. Methods. 2001;25:402– 408.
27. Maeda A, Crabb JW, Palczewski K. Microsomal glutathione S-
transferase 1 in the retinal pigment epithelium: protection against
oxidative stress and a potential role in aging. Biochemistry. 2005;
18:480–489.
28. Kutty RK, Nagineni CN, Kutty G, Hooks JJ, Chader GJ, Wiggert
BJ. Increased expression of heme oxygenase-1 in human retinal
pigment epithelial cells by transforming growth factor-beta. J Cell
Physiol. 1994;159:371–378.
29. Organisciak DT, Darrow RM, Barsalou L, Kutty RK, Wiggert B.
Circadian-dependent retinal light damage in rats. Invest Ophthal-
mol Vis Sci. 2000;41:3694 –3701.
2348 Milbury et al. IOVS, May 2007, Vol. 48, No. 5
30. Naash MI, Nielsen JC, Anderson RE. Regional distribution of glu-
tathione peroxidase and glutathione-S-transferase in adult and pre-
mature human retinas. Invest Ophthalmol Vis Sci. 1988;29:149 –
152.
31. Adaikalakoteswari A, Balasubramanyam M, Rema M, Mohan V.
Differential gene expression of NADPH oxidase (p22phox) and
hemoxygenase-1 in patients with type 2 diabetes and microangi-
opathy. Diabet Med. 2006;23:666 –674.
32. da Silva JL, Stoltz RA, Dunn MW, Abraham NG, Shibahara S.
Diminished heme oxygenase-1 mRNA expression in RPE cells from
diabetic donors as quantitated by competitive RT/PCR. Curr Eye
Res. 1997;6:380 –386.
33. Rattan SI, Clark BF. Intracellular protein synthesis, modifications
and aging. Biochem Soc Trans. 1996;24:1043–1049.
34. Richardson A, Cheung T. Current concepts: I. The relationship
between age-related changes in gene expression, protein turnover,
and the responsiveness of an organism to stimuli. Life Sci. 1982;
31:605–613.
35. Morazzoni P, Bombardelli E. Vaccinium myrtillus L. Fitoterapia.
1996;67:3–29.
36. Joseph JA, Shukitt-Hale B, Denisova NA, et al. Reversals of age-
related declines in neuronal signal transduction, cognitive, and
motor behavioral deficits with blueberry, spinach, or strawberry
dietary supplementation. J Neurosci. 1999;19:8114 –8121.
37. Canter PH, Ernst E. Anthocyanosides of Vaccinium myrtillus (bil-
berry) for night vision: a systematic review of placebo-controlled
trials. Surv Ophthalmol. 2004;49:38 –50.
38. Bailey TA, Kanuga N, Romero IA, Greenwood J, Luthert PJ,
Cheetham ME. Oxidative stress affects the junctional integrity of
retinal pigment epithelial cells. Invest Ophthalmol Vis Sci. 2004;
45:675–684.
39. Wada M, Gelfman CM, Matsunaga H, et al. Density-dependent
expression of FGF-2 in response to oxidative stress in RPE cells in
vitro. Curr Eye Res. 2001;23:226 –231.
40. Alizadeh M, Wada M, Gelfman CM, Handa JT, Hjelmeland LM.
Downregulation of differentiation specific gene expression by
oxidative stress in ARPE-19 cells. Invest Ophthalmol Vis Sci. 2001;
42:2706–2713.
41. Kim BY, Han MJ, Chung AS. Effects of reactive oxygen species on
proliferation of Chinese hamster lung fibroblast (V79) cells. Free
Radic Biol Med. 2001;30:686 –698.
42. Davies KJ. The broad spectrum of responses to oxidants in prolif-
erating cells: a new paradigm for oxidative stress. IUBMB Life.
1999;48:41–47.
43. Hanneken A, Lin FF, Johnson J, Maher P. Flavonoids protect hu-
man retinal pigment epithelial cells from oxidative-stress-induced
death. Invest Ophthalmol Vis Sci. 2006;47:3164–3177.
44. Maher P, Hanneken A. Flavonoids protect retinal ganglion cells
from oxidative stress-induced death. Invest Ophthalmol Vis Sci.
2005;46:4796–4803.
45. Lazze MC, Pizzala R, Perucca P, et al. Anthocyanidins decrease
endothelin-1 production and increase endothelial nitric oxide syn-
thase in human endothelial cells. Mol Nutr Food Res. 2006;50:44 –
51.
IOVS, May 2007, Vol. 48, No. 5 Bilberry and Oxidative Stress Defense 2349
