Intense Physiological Light Upregulates Vascular Endothelial
Growth Factor and Enhances Choroidal Neovascularization
via Peroxisome P
­ roliferator-­Activated Receptor g
Coactivator-1a in Mice
Takashi Ueta, Tatsuya Inoue, Kentaro Yuda, Takahisa Furukawa, Yasuo Yanagi, Yasuhiro Tamaki
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
Objective—Toxicity of intense light to facilitate the development of neovascular a­ ge-­related macular degeneration has been
a health concern although the mechanism remains unclear.
Methods and Results—Effects of intense, but within physiological range, light on retinal pigment epithelium, a major
pathogenic origin of ­age-­related macular degeneration were studied in mice. Intense physiological light upregulated
vascular endothelial growth factor (VEGF) expression in retinal pigment epithelium, independent of circadian rhythm,
which resulted in enhancement of choroidal neovascularization. In rd1/rd1 mice or Crx2/2 mice that do not possess
outer segment structure, light exposure did not induce VEGF, indicating that VEGF upregulation by light depended on
increased outer segment phagocytosis by retinal pigment epithelium. In retinal pigment epithelium cells phagocytosing
increased amount of outer segment, peroxisome ­proliferator-­activated receptor g coactivator-1a (PGC-1a) not ­hypoxia­inducible factor-1a was induced, leading to VEGF upregulation. The VEGF upregulation and choroidal neovascularization
enhancement were abrogated in PGC-1a2/2 mice and e­ strogen-­related receptor-a2/2 mice, indicating the involvement of
PGC-1a/­estrogen-­related receptor-a pathway.
Conclusion—Intense physiological light is involved in choroidal neovascularization through excess outer segment phagocytosis and VEGF upregulation mediated by PGC-1a in vivo. (Arterioscler Thromb Vasc Biol. 2012;32:1366-1371.)
Key Word: angiogenesis  choroidal neovascularization  retinal photoreceptor cell outer segment
 Ppargc1a protein, mouse
­A
been difficult to discuss the association of light exposure and
AMD. For instance, mice entrain circadian rhythm with 0.1
lx light13,14 and are maintained at 10 to 20 lx in experimental
conditions; conversely, for humans, sunlight, that is around 50
000 lx, is just within physiological range, although it is often
unpleasant. In the present study, we developed a mouse model of
intense physiological light (IPL) exposure and studied its effects
on RPE and CNV, pathogenic origins of neovascular AMD.
ge-­related macular degeneration (AMD) is a prominent cause of legal blindness worldwide.1 Although it
is difficult to measure light exposure in human subjects, several ­population-­based studies have indicated the causality
of intense light exposure, such as sunlight exposure, for the
development of late AMD,2–5 especially neovascular AMD
that is choroidal neovascularization (CNV) in the elderly.2,3
Recent experimental studies have also suggested the association of light exposure and neovascular AMD,6–9 although they
have been based on a hypothesis that the reaction between the
light and the increased amount of bisretinoids in retinal pigment epithelium (RPE) might be sufficient to induce VEGF
upregulation or inflammation.
A conventional animal model for ­light-­induced toxicity on
the retina has been a rodent model exposed to visible light with
intensities with several thousand lux (lx),10,11 that causes acute
loss of photoreceptors. Recently, 3000 lx light was reported to
induce CNV besides retinal degeneration12; however, a considerable problem of the existing animal model of light toxicity is
that the light intensities are unphysiologically strong, and it has
Research Design and Methods
All experiments in the present study were conducted according to
ARVO Statement for the Use of Animals in Ophthalmic and Vision
Research and were approved by the Institutional Animal Research
Committee of the University of Tokyo.
Adult male mice 8 to 12 weeks old were used in all experiments,
except Crx2/2 mice which was used at the age of 8 weeks. rd1/rd1
mice were purchased from Saitama Experimental Animals Supply
Co Ltd. Peroxisome ­proliferator-­activated receptor g ­coactivator-1a
(PGC-1a)2/2 mice were purchased form Jackson Laboratory.
­Estrogen-­related receptor-a (ERR-a)2/2 mice were kindly provided
by Novartis Institutes for BioMedical Research Inc. All the mice used
in this study were raised in a C57BL/6 background. Mice were kept
under 5- to 10-lx (12 hours) light/dark (12 hours) cycle at 22°C, if not
Received on: February 17, 2012; final version accepted on: March 29, 2012.
From the Department of Ophthalmology, University of Tokyo School of Medicine, ­Bunkyo-­ku, Tokyo, Japan (T.U., T.I., K.Y., Y.Y., Y.T.); and
Department of Developmental Biology, Osaka Bioscience Institute & JST, CREST, Suita, Osaka, Japan (T.F.).
The o­ nline-­only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.112.248021/-/DC1.
Correspondence to Takashi Ueta, Department of Ophthalmology, University of Tokyo School of Medicine, 7-3-1 Hongo, B
­ unkyo-­ku, Tokyo 113–8655,
Japan. E
­ -­mail [email protected]
© 2012 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.orgDOI: 10.1161/ATVBAHA.112.248021
1366
Ueta et al Light and Choroidal Neovascularization 1367
mentioned otherwise. Light exposure up to weak 15 lx light did not
change the temperature inside the cages.
Details of the methods including light exposure experiments,
VEGF mRNA and protein measurements, Western blotting, antibodies, primers, and statistics are provided in the online-only Data
Supplement.
Results
Development of a Mouse Model of IPL Exposure
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
First, we defined the intensity of IPL in C57BL/6J ­wild-­type
(WT) mice. Because generally in experiments using mice,
they are kept under 10 to 20 lx light at most, we first addressed
whether light 20 lx can cause discomfort for WT mice. A
WT mouse was placed in a cage illuminated with varying
intensities of visible light (1−20 lx) or in a cage with complete darkness. One mouse was placed in 1 cage at a time and
the test was repeated 7 to 8 times for each cage. The location
of the mouse in each cage was recorded 4 hours later. In the
illuminated cage, no mice chose to stay in the area 10 lx
of light while in the cage under darkness, mice did not show
any preference to a specific area of the cage (Figure 1A). We
also confirmed that 15 lx light was physiological for mice;
15-lx light (12 hours)/dark (12 hours) cyclic light for 3 weeks
neither damaged photoreceptors or RPE cells nor changed the
RPE65 protein level or zonula occludens-1 protein expression
in RPE cells (Figure 1B). In this study, therefore, 15 lx was
chosen as a representative of IPL for mice.
IPL Upregulates VEGF Expression in
RPE of WT Mice
Reportedly, the expression of VEGF increases in RPE cells
by treatment with outer segment (OS) in vitro.15 At the same
time, it has been well established that physiological light
exposure is implicated in the initiation and promotion of OS
shedding by photoreceptor cells, and the shed OS are phagocytosed by RPE cells.16–21 Based on this previous evidence,
we speculated that exposure to IPL could increase VEGF
expression in mouse RPE in vivo. VEGF protein produced in
RPE is released basolaterally into the choroidal vasculature,
an event implicated in both the maintenance of the choroidal vasculature16,22 and the pathological CNV.23–25 Indeed, the
increased level of ­VEGF-­A protein was detected in the RPE/
choroid of WT mice exposed to 15 or 150 lx, but not 0.5 lx of
light, compared with those kept in the dark (Figure 2A). The
­VEGF-­A upregulation followed a circadian pattern, increasing during the day and decreasing during the night (Figure
2B). To understand the ­circadian-­independent, direct effect
of light on the VEGF upregulation, the unilateral eyes of
WT mice were closed (the unilateral ­closed-­eye model) and
the mice were maintained for 24 hours in the dark. The mice
were then exposed to 0.5 or 15 lx of light at night. Again,
an increase in ­VEGF-­A protein level was observed in open
eyes of mice under IPL at night, indicating that the VEGF
upregulation occurred as a result of the direct effect of IPL on
the eyes and independent of circadian rhythm (Figure 2C). To
confirm the source of the increased VEGF protein in the RPE/
choroid tissues, V
­ EGF-­A mRNA expression was evaluated in
RPE and choroid separately. The increased V
­ EGF-­A mRNA
was detected in RPE, but not in the choroid, of the WT mice
exposed to 15 lx of light (Figure 2D). In addition, we evaluated
the increased OS phagocytosis by RPE on IPL in WT mice,
which occurs in a ­circadian-­independent manner (Figure I in
the ­online-­only Data Supplement). Furthermore, in mice that
lack OS structure (ie, rd1/rd1 and Crx2/2 mice26,27), ­VEGF-­A
protein level in RPE/choroid or V
­ EGF-­A mRNA in RPE did
not change on IPL (Figure 2E and Figure II in the ­online-­only
Data Supplement).
VEGF Upregulation in RPE Cells Treated
With OS Is Mediated by PGC-1/ERR-
Pathway In Vitro
Next, we explored the mechanism of the VEGF upregulation
in RPE cells phagocytosing increased volume of OS, using the
cultured human RPE cell line, ARPE-19 cells. Consistent with
our findings in vivo and a previous report,15 the expression of
both V
­ EGF-­A mRNA in ARPE-19 cells and ­VEGF-­A protein
released in the medium increased, depending on the amount
of OS to treat the cells (Figure 3A and 3B). In ARPE-19 cells
transfected with a reporter plasmid containing human VEGF
promoter region (from 21180 to 1338) driving the luciferase
gene,28 the treatment with OS led to a 3-fold increase in the
induction of luciferase activity (Figure 3C), indicating that
OS treatment on RPE cells resulted in the activation of VEGF
transcription.
In RPE cells, transcriptional regulation of VEGF has
been well characterized by h­ ypoxia-­inducible factor-1a
(HIF-1a).29,30 However, recent studies have revealed the
importance of the ­HIF-­independent, PGC-1a/ERR-a
Figure 1. Definition of intense physiological light for mice. A, A ­wild-­type (WT) mouse was placed in the cage with varying intensities of
light or in the same cage with complete darkness. One mouse was placed in the cage at a time, and the test was repeated 7 to 8 times
for each cage. Location of each mouse was marked as “X.” Preference for area 1 was compared between the 2 cages (*P0.05, by
Fisher exact test). B, Eyes of WT mice maintained under 15-lx light/dark cycle or 0.5-lx light/dark cycle for 3 weeks. The integrity of photoreceptor cells was evaluated by the thickness of outer nuclear layer (ONL). The distribution of zonula occludens (ZO)-1 protein in RPE
was evaluated on flatmount, and RPE65 protein level in RPE on Western blotting (scale bar, 50 mm).
1368 Arterioscler Thromb Vasc Biol June 2012
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
Figure 2. Vascular endothelial growth factor (VEGF) expression increases in retinal pigment epithelium (RPE) of w
­ ild-­type (WT) mice when
exposed to intense physiological light (IPL). A, Total ­VEGF-­A protein (pg/mg extract) in RPE/choroid was measured in WT mice under 0,
0.5, 15, or 150 lx for 3 or 9 hours (meanSEM; n8–15 per group; ns, not significant; ***P0.001, by ANOVA, followed by post hoc
Dunnett test in comparison with the 0 lx group). B, Circadian change in ­VEGF-­A protein level in RPE/choroid of WT mice under 0.5-lx light
(12 hours)/dark (12 hours) cycle and 15-lx light (12 hours)/dark (12 hours) cycle. C, Comparison of the total ­VEGF-­A protein (pg/mg) in
RPE/choroid between 2 eyes of the same WT mice using the unilateral c
­ losed-­eye model. Mice were exposed to 0.5, 15, or 150 lx for 3
hours (n5; **P0.01, by paired t test). D, Relative ­VEGF-­A mRNA expression was measured in RPE or choroid of WT mice by ­real-­time
reverse transcriptase polymerase chain reaction. The effect of IPL was examined in the unilateral c
­ losed-­eye model (n5, **P0.01, by
paired t test). E, Total V
­ EGF-­A protein (pg/mg extract) in RPE/choroid was measured in rd1/rd1 and Crx2/2 mice under 0, 0.5, 15, or 150 lx
of light for 3 or 9 hours (meanSEM; n8 per group; ns, not significant by ANOVA; INL, inner nuclear layer; ONL, outer nuclear layer).
pathway in the transcriptional regulation of VEGF.31 OS
consists of a pile of uniquely folded cell membrane disks
that are degraded into fatty acids in RPE cells following
phagocytosis.17 Given that PGC-1a/ERR-a signaling is
critical in the metabolism of fatty acids, we speculated that
this pathway might be involved in the VEGF transcription in RPE cells phagocytosing OS. There has also been
a report that OS phagocytosis activates the expression of
peroxisome ­proliferator-­activated ­receptor-­activated receptor-g, a target of PGC-1a, in a rat RPE primary culture.32
Indeed, in ARPE-19 cells treated with OS, expression of
PGC-1a mRNA, but not HIF-1a mRNA, increased in a
d­ ose-­dependent manner (Figure 3D). OS treatment did not
change HIF-1a protein stability; meanwhile, it induced
expression of PGC-1a protein (Figure 3E). In addition, OS
treatment increased the expression of other key target genes
in the PGC-1a/ERR-a pathway, including genes for fatty
acid utilization (­medium-­chain ­acyl-­CoA dehydrogenase
and pyruvate dehydrogenase kinase 4) and fatty acid transport (CD36) (Figure III in the ­online-­only Data Supplement).
We investigated the effect of the PGC-1a/ERR-a pathway
on the ­VEGF-­A gene using siRNAs designed to specifically
knock down PGC-1a and ERR-a in ARPE-19 cells (Figure
IV in the o­ nline-­only Data Supplement). Knockdown of
these transcriptional regulatory factors diminished the
induction of ­VEGF-­A mRNA in ARPE-19 cells treated with
OS (Figure 3F).
PGC-1/ERR- Pathway Is Implicated
in the VEGF Upregulation in RPE Stimulated
by IPL In Vivo
We further tested the implication of PGC-1a/ERR-a pathway in the induction of VEGF in RPE stimulated by IPL in
vivo. Using the unilateral ­closed-­eye model, we confirmed
an increase in the level of PGC-1a protein, but not HIF-1a
protein, in the eyes of WT mice exposed to 15 lx of light in
vivo (Figure 4A). We also used PGC-1a2/2 and ERR-a2/2
mice to confirm the involvement of PGC-1a-ERR-a pathway
in the VEGF upregulation in RPE cells by IPL. Both PGC1a2/2 and ERR-a2/2 mice possess a normal retinal phenotype
harboring an intact OS structure (Figure V in the o­ nline-­only
Ueta et al Light and Choroidal Neovascularization 1369
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
Figure 3. Vascular endothelial
growth factor (VEGF) upregulation in retinal pigment epithelium
(RPE) cells treated with outer
segment (OS) is mediated by
PGC-1a/ERR-a pathway in vitro.
A, V
­ EGF-­A mRNA in ARPE-19
cells and (B) ­VEGF-­A protein level
in the medium with or without
OS treatment for 3−6 hours were
measured by ­real-­time reverse
transcriptase polymerase chain
reaction and ELISA, respectively
(meanSEM; n4 per group;
*P0.05, **P0.01, ***P0.001,
by ANOVA, followed by Tukey
post hoc test). C, Luciferase
activity was measured in
ARPE-19 cells, transfected with
VEGF or control promoter constructs, with or without OS treatment for 3 hours (meanSEM;
n4; ns, not significant,
***P0.001 by Student t test).
D, HIF-1a and PGC-1a mRNA
in ARPE-19 cells was measured,
with or without OS treatment
with OS for 6 hours (meanSEM; n4 per group; ***P0.001, by ANOVA, followed by Tukey post hoc test; ns, not significant, by ANOVA).
E, Lysates of ARPE-19 cells, with or without OS treatment for 6 hours, were immunoblotted for HIF-1a and PGC-1a, and relative protein
abundance was measured. F, V
­ EGF-­A mRNA was measured in ARPE-19 cells transfected with scramble siRNA or siRNAs for PGC-1a and
ERR-a. ARPE-19 cells were treated with or without OS for 3 hours (meanSEM; n4 per group; ***P0.001, by Student t test). ERR-a,
­estrogen-­related receptor-a; HIF, h
­ ypoxia-­inducible factor; PGC-1a, peroxisome p
­ roliferator-­activated receptor g coactivator-1a.
Data Supplement); however, neither ­VEGF-­A protein level
in RPE/choroid nor ­VEGF-­A mRNA expression in RPE was
changed by IPL in these mice (Figure 4B and Figure VI in the
­online-­only Data Supplement).
Pathological CNV Is Enhanced by IPL Through
PGC-1/ERR- Pathway
Figure 4. PGC-1a/ERR-a pathway is implicated in the vascular
endothelial growth factor (VEGF) upregulation in retinal pigment epithelium (RPE) cells stimulated by intense physiological
light in vivo. A, Lysates of RPE/choroid of ­wild-­type mice were
immunoblotted to compare HIF-1a and PGC-1a protein abundance between the 2 eyes of the same mice using the unilateral
­closed-­eye model. The mice were exposed to 0.5 lx or 15 lx of
light for 6 hours. The result was confirmed by 3 independent
tests. B, Both PGC-1a2/2 and ERR-a2/2 mice have normal retinal
phenotype harboring outer segment (red). ­VEGF-­A protein level
in RPE/choroid was compared between the 2 eyes of the same
PGC-1a2/2 or ERR-a2/2 mice using the unilateral c
­ losed-­eye
model. Mice were exposed to 15 lx of light for 3 hours (n5;
ns, not significant, by paired t test). ERR-a, ­estrogen-­related
receptor-a; HIF, ­hypoxia-­inducible factor; PGC-1a, peroxisome
­proliferator-­activated receptor g coactivator-1a.
Finally, to elucidate the role of exposure to IPL in pathological CNV formation, the l­aser-­induced CNV model was
used. After laser treatment for induction of CNV on day 0,
WT, PGC-1a2/2 and ERR-a2/2 mice were kept under either
a 0.5-lx light (12 hours)/dark (12 hours) cycle; a 15-lx light
(12 hours)/dark (12 hours) cycle; or continuous darkness for
1 week before the measurement of CNV size (Figure 5A). In
WT mice kept under the 15-lx cyclic light, the area of CNV
was larger than WT mice kept under 0.5-lx cyclic light or
in constant darkness. Meanwhile, in PGC-1a2/2 and ERRa2/2 mice, the 15-lx cyclic light did not increase CNV size
(Figure 5B). Furthermore, when WT mice were treated with
an anti-­VEGF-­A neutralizing antibody, the CNV size was not
changed by the 15-lx light exposure (Figure 5B).
Discussion
These results indicate the mechanism by which IPL could
cause toxicity in RPE and choroidal vasculature; exposure to
intense light increases photoreceptor OS phagocytosis by RPE
cells, which in turn induces overactivation of the PGC-1a/
ERR-a pathway, upregulating VEGF in RPE and enhancing
CNV (Figure 6).
PGC-1a is ­well-­known as a powerful transcriptional coactivator facilitating adaptation to physiological stimuli, including
adaptive thermogenesis,33 exercise,34 and fasting/feeding.35
1370 Arterioscler Thromb Vasc Biol June 2012
Figure 5. Pathological choroidal neovascularization (CNV) is enhanced by intense physiological light through PGC-1a/ERR-a pathway.
A, L
­ aser-­induced CNV model was used in ­wild-­type (WT), PGC-1a2/2 and ERR-a2/2 mice, and WT mice treated with a neutralizing antivascular endothelial growth factor (VEGF)-A antibody at day 1. Mice were maintained under constant darkness; 0.5-lx (12 hours) light/
dark (12 hours) cycle or 15-lx (12 hours)/dark (12 hours) cycle. B, CNV size was measured on day 7 (meanSEM; n6–8 per group;
*P0.05, by ANOVA, followed by Tukey post hoc test; ns, not significant, by Student t test; scale bar, 100 mm). ERR-a, ­estrogen-­related
receptor-a; PGC-1a, peroxisome ­proliferator-­activated receptor g coactivator-1a.
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
Although PGC-1a also mediates circadian rhythm,36,37 PGC-1a
induction by IPL in the present study was considered as the
direct effect of light, not through a change in circadian rhythm
because PGC-1a was induced only in the open eyes, not in
the closed eyes of WT mice under IPL, and the increased
VEGF expression by IPL was also observed during the night.
Our results disclosed a novel function of PGC-1a to mediate
VEGF induction in RPE cells by light through OS phagocytosis. We consider that PGC-1a-mediated VEGF expression in
RPE should be physiologically important to maintain healthy
choroidal vasculature. However, our results indicated that the
overactivation by the excess OS burden seems to render a
pathogenic role.
OS shedding/phagocytosis has been a w
­ ell-­known physiological phenomenon but its physiologic and pathological
roles have not been fully understood. As supportive roles of
OS phagocytosis, it reportedly protects RPE cells against oxidative s­ tress-­induced apoptosis in vitro.38 In mice whose OS
phagocytosis by RPE was disturbed by CD36 abrogation, an
increased avascular area in choriocapillaris was observed.15 In
contrast, the deleterious effects of excessive OS phagocytosis
by RPE cells have been commonly discussed16,39,40 because the
uptake of photooxidized OS is considered to render oxidative
stress to RPE cells,39 and it is deemed responsible for lipofuscin
accumulation in RPE cells which is considered to contribute to
the pathogenesis of AMD.39,40 In agreement with the reported
deleterious effects of excessive OS phagocytosis, our results
indicate that excessive OS phagocytosis induced by IPL could
enhance pathological CNV in vivo. Based on our results and
previous literature, VEGF secretion mediated by OS phagocytosis is essential for the development and maintenance of healthy
choroidal vasculature15 while VEGF upregulation of excessive
OS phagocytosis could have a pathogenic role. Considering
that the declined defense against antioxidative stress is a cardinal feature of aging, OS of photoreceptor cells might be more
readily oxidized by light exposure in the aged, which might lead
to a further increase in the volume of phagocytosed OS41 and
VEGF expression. In contrast, a previous study reported that
monocyte chemotactic protein 1 and IL-8, angiogenic cytokines were upregulated in human ARPE-19 cells by oxidized
OS.40 However, the upregulation was observed only when OS
were irradiated by ultraviolet light although ultraviolet light is
filtered at the anterior part of the eye including cornea and lens,
and does not reach RPE in human eyes.42,43
In literature, light toxicity to facilitate CNV has been
discussed from the viewpoint of fluorescent bisretinoids, constituents of lipofuscin in RPE cells, including A2E, a ­by-­product
of the visual cycle, because they could be activated by visible
light exposure.6 In an in vitro study, blue ­light-­stimulated A2E
could enhance the generation of lipid peroxidation products
and advanced glycation end products in RPE cells, which could
lead to the increased expression of VEGF.7 Alternatively, photooxidized A2E could activate the complement system,8 which
may promote CNV.9 Because the visual cycle is coupled to OS
shedding/phagocytosis,16,44 and excess visual cycle results in
the increased A2E synthesis in RPE cells,45 these hypotheses
involving bisretinoids may be associated with the mechanism
proposed by our study. In addition, a limitation of the present
study is that although our data indicated IPL could enhance
CNV in vivo, it is still not clear how IPL could facilitate the de
novo development of CNV.
Sources of Funding
Figure 6. A schematic illustration of the mechanism by which
intense physiological light facilitates choroidal neovascularization through outer segment (OS) phagocytosis and PGC-1a/
ERR-a pathway in retinal pigment epithelium (RPE) cells. ERR-a,
­ strogen-­related receptor-a; VEGF, vascular endothelial growth
e
factor.
T. Ueta was supported by DC2 Research Fellowship for Young
Scientists from Japan Society for the Promotion of Science. This
work was supported by Grants-­in-­Aid for Scientific Research from
Japan Society for the Promotion of Science to Y. Tamaki (20390446
and 23659806).
Disclosures
None.
Ueta et al Light and Choroidal Neovascularization 1371
References
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
1. Jager RD, Mieler WF, Miller JW. ­Age-­related macular degeneration.
N Engl J Med. 2008;358:2606–2617.
2. Fletcher AE, Bentham GC, Agnew M, Young IS, Augood C, Chakravarthy
U, de Jong PT, Rahu M, Seland J, Soubrane G, Tomazzoli L, Topouzis F,
Vingerling JR, Vioque J. Sunlight exposure, antioxidants, and ­age-­related
macular degeneration. Arch Ophthalmol. 2008;126:1396–1403.
3. Cruickshanks KJ, Klein R, Klein BE. Sunlight and ­age-­related macular degeneration. The Beaver Dam Eye Study. Arch Ophthalmol.
1993;111:514–518.
4. Mitchell P, Smith W, Wang JJ. Iris color, skin sun sensitivity, and
­age-­related maculopathy. The Blue Mountains Eye Study. Ophthalmology.
1998;105:1359–1363.
5. Taylor HR, West S, Muñoz B, Rosenthal FS, Bressler SB, Bressler
NM. The ­long-­term effects of visible light on the eye. Arch Ophthalmol.
1992;110:99–104.
6. Sparrow JR, Fishkin N, Zhou J, Cai B, Jang YP, Krane S, Itagaki
Y, Nakanishi K. A2E, a byproduct of the visual cycle. Vision Res.
2003;43:2983–2990.
7. Zhou J, Cai B, Jang YP, Pachydaki S, Schmidt AM, Sparrow JR. Mechanisms
for the induction of HNE- MDA- and ­AGE-­adducts, RAGE and VEGF in
retinal pigment epithelial cells. Exp Eye Res. 2005;80:567–580.
8. Zhou J, Jang YP, Kim SR, Sparrow JR. Complement activation by photooxidation products of A2E, a lipofuscin constituent of the retinal pigment
epithelium. Proc Natl Acad Sci U S A. 2006;103:16182–16187.
9. Nozaki M, Raisler BJ, Sakurai E, Sarma JV, Barnum SR, Lambris JD,
Chen Y, Zhang K, Ambati BK, Baffi JZ, Ambati J. Drusen complement
components C3a and C5a promote choroidal neovascularization. Proc
Natl Acad Sci U S A. 2006;103:2328–2333.
10. de Jong PT. ­Age-­related macular degeneration. N Engl J Med.
2006;355:1474–1485.
11. Ham WT Jr, Mueller HA, Sliney DH. Retinal sensitivity to damage from
short wavelength light. Nature. 1976;260:153–155.
12. Albert DM, Neekhra A, Wang S, Darjatmoko SR, Sorenson CM,
Dubielzig RR, Sheibani N. Development of choroidal neovascularization
in rats with advanced intense cyclic ­light-­induced retinal degeneration.
Arch Ophthalmol. 2010;128:212−222.
13. Altimus CM, Güler AD, Alam NM, Arman AC, Prusky GT, Sampath AP,
Hattar S. Rod photoreceptors drive circadian photoentrainment across a
wide range of light intensities. Nat Neurosci. 2010;13:1107–1112.
14. Ebihara S, Tsuji K. Entrainment of circadian activity rhythm to the
light cycle: effective light intensity for Zeitgeber in the retinal degenerate C3H mouse and the normal C57BL mouse. Physiol Behavior.
1979;24:523−527.
15. Houssier M, Raoul W, Lavalette S, Keller N, Guillonneau X, Baragatti
B, Jonet L, Jeanny JC, ­Behar-­Cohen F, Coceani F, Scherman D,
Lachapelle P, Ong H, Chemtob S, Sennlaub F. CD36 deficiency leads to
choroidal involution via COX2 ­down-­regulation in rodents. PLoS Med.
2008;5:e39.
16. Strauss O. The retinal pigment epithelium in visual function. Physiol Rev.
2005;85:845–881.
17. Bazan NG. Homeostatic regulation of photoreceptor cell integrity: significance of the potent mediator neuroprotectin D1 biosynthesized from
docosahexaenoic acid: the Proctor Lecture. Invest Ophthalmol Vis Sci.
2007;48:4866–4881; biography 4864.
18. LaVail MM. Rod outer segment disk shedding in rat retina: relationship to
cyclic lighting. Science. 1976;194:1071–1074.
19. Bobu C, Hicks D. Regulation of retinal photoreceptor phagocytosis
in a diurnal mammal by circadian clocks and ambient lighting. Invest
Ophthalmol Vis Sci. 2009;50:3495–3502.
20. Basinger S, Hoffman R, Matthes M. Photoreceptor shedding is initiated
by light in the frog retina. Science. 1976;194:1074–1076.
21. Besharse JC, Hollyfield JG, Rayborn ME. Photoreceptor outer segments:
accelerated membrane renewal in rods after exposure to light. Science.
1977;196:536–538.
22.­Saint-­Geniez M, Kurihara T, Sekiyama E, Maldonado AE, D’Amore PA.
An essential role for ­RPE-­derived soluble VEGF in the maintenance of
the choriocapillaris. Proc Natl Acad Sci U S A. 2009;106:18751–18756.
23. Witmer AN, Vrensen GF, Van Noorden CJ, Schlingemann RO. Vascular
endothelial growth factors and angiogenesis in eye disease. Prog Retin
Eye Res. 2003;22:1–29.
24. Spilsbury K, Garrett KL, Shen WY, Constable IJ, Rakoczy PE.
Overexpression of vascular endothelial growth factor (VEGF) in the
retinal pigment epithelium leads to the development of choroidal neovascularization. Am J Pathol. 2000;157:135–144.
25. Schwesinger C, Yee C, Rohan RM, Joussen AM, Fernandez A, Meyer
TN, Poulaki V, Ma JJ, Redmond TM, Liu S, Adamis AP, D’Amato RJ.
Intrachoroidal neovascularization in transgenic mice overexpressing vascular endothelial growth factor in the retinal pigment epithelium. Am J
Pathol. 2001;158:1161–1172.
26. Morrow EM, Furukawa T, Raviola E, Cepko CL. Synaptogenesis and
outer segment formation are perturbed in the neural retina of Crx mutant
mice. BMC Neurosci. 2005;6:5.
27. Furukawa T, Morrow EM, Li T, Davis FC, Cepko CL. Retinopathy
and attenuated circadian entrainment in C
­ rx-­deficient mice. Nat Genet.
1999;23:466–470.
28. Akiyama H, Tanaka T, Maeno T, Kanai H, Kimura Y, Kishi S,
Kurabayashi M. Induction of VEGF gene expression by retinoic acid
through Sp1-binding sites in retinoblastoma Y79 cells. Invest Ophthalmol
Vis Sci. 2002;43:1367–1374.
29. Forooghian F, Razavi R, Timms L. ­Hypoxia-­inducible factor expression
in human RPE cells. Br J Ophthalmol. 2007;91:1406−1410.
30. Fernandes AF, Guo W, Zhang X, Gallagher M, Ivan M, Taylor A, Pereira
P, Shang F. ­Proteasome-­dependent regulation of signal transduction in
retinal pigment epithelial cells. Exp Eye Res. 2006;83:1472–1481.
31. Arany Z, Foo SY, Ma Y, Ruas JL, B
­ ommi-­Reddy A, Girnun G, Cooper
M, Laznik D, Chinsomboon J, Rangwala SM, Baek KH, Rosenzweig
A, Spiegelman BM. H
­ IF-­independent regulation of VEGF and angiogenesis by the transcriptional coactivator PGC-1alpha. Nature.
2008;451:1008–1012.
32. Ershov AV, Bazan NG. Photoreceptor phagocytosis selectively activates
PPARgamma expression in retinal pigment epithelial cells. J Neurosci
Res. 2000;60:328–337.
33. Puigserver P, Wu Z, Park CW, Graves R, Wright M, Spiegelman BM. A
c­ old-­inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell. 1998;92:829–839.
34. Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss O, Michael LF, Puigserver
P, Isotani E, Olson EN, Lowell BB, B
­ assel-­Duby R, Spiegelman BM.
Transcriptional ­co-­activator PGC-1 alpha drives the formation of
­slow-­twitch muscle fibres. Nature. 2002;418:797–801.
35. Yoon JC, Puigserver P, Chen G, Donovan J, Wu Z, Rhee J, Adelmant
G, Stafford J, Kahn CR, Granner DK, Newgard CB, Spiegelman BM.
Control of hepatic gluconeogenesis through the transcriptional coactivator PGC-1. Nature. 2001;413:131–138.
36. Huang W, Ramsey KM, Marcheva B, Bass J. Circadian rhythms, sleep,
and metabolism. J Clin Invest. 2011;121:2133–2141.
37. Liu C, Li S, Liu T, Borjigin J, Lin JD. Transcriptional coactivator PGC1alpha integrates the mammalian clock and energy metabolism. Nature.
2007;447:477–481.
38. Mukherjee PK, Marcheselli VL, de Rivero Vaccari JC, Gordon WC,
Jackson FE, Bazan NG. Photoreceptor outer segment phagocytosis attenuates oxidative ­stress-­induced apoptosis with concomitant neuroprotectin
D1 synthesis. Proc Natl Acad Sci U S A. 2007;104:13158–13163.
39. Beatty S, Koh H, Phil M, Henson D, Boulton M. The role of oxidative
stress in the pathogenesis of ­age-­related macular degeneration. Surv
Ophthalmol. 2000;45:115–134.
40. Higgins GT, Wang JH, Dockery P, Cleary PE, Redmond HP. Induction
of angiogenic cytokine expression in cultured RPE by ingestion of
oxidized photoreceptor outer segments. Invest Ophthalmol Vis Sci.
2003;44:1775–1782.
41. Sun M, Finnemann SC, Febbraio M, Shan L, Annangudi SP, Podrez EA,
Hoppe G, Darrow R, Organisciak DT, Salomon RG, Silverstein RL, Hazen
SL. L
­ ight-­induced oxidation of photoreceptor outer segment phospholipids generates ligands for CD36-mediated phagocytosis by retinal pigment
epithelium: a potential mechanism for modulating outer segment phagocytosis under oxidant stress conditions. J Biol Chem. 2006;281:4222–4230.
42. Moses RA. Adler’s Physiology of the Eye. 7th ed. London, UK: Mosby.
1981:357–369.
43. Lerman S. Radiant Energy and the Eye. New York: Macmillan Pub Co
Inc; 1980:131–133.
44. Kim SR, Nakanishi K, Itagaki Y, Sparrow JR. Photooxidation of A2-PE,
a photoreceptor outer segment fluorophore, and protection by lutein and
zeaxanthin. Exp Eye Res. 2006;82:828–839.
45. Golczak M, Maeda A, Bereta G, Maeda T, Kiser PD, Hunzelmann S, von
Lintig J, Blaner WS, Palczewski K. Metabolic basis of visual cycle inhibition by retinoid and nonretinoid compounds in the vertebrate retina. J Biol
Chem. 2008;283:9543–9554.
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
Intense Physiological Light Upregulates Vascular Endothelial Growth Factor and Enhances
Choroidal Neovascularization via Peroxisome Proliferator-Activated Receptor γ
Coactivator-1 α in Mice
Takashi Ueta, Tatsuya Inoue, Kentaro Yuda, Takahisa Furukawa, Yasuo Yanagi and Yasuhiro
Tamaki
Arterioscler Thromb Vasc Biol. 2012;32:1366-1371; originally published online April 19, 2012;
doi: 10.1161/ATVBAHA.112.248021
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 2012 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/32/6/1366
Data Supplement (unedited) at:
http://atvb.ahajournals.org/content/suppl/2012/04/19/ATVBAHA.112.248021.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is
being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online at:
http://atvb.ahajournals.org//subscriptions/
Supplement Material
Supplemental Methods
Light Exposure Experiments− The experiments were performed in a closed room
without a window. Source of light used in this study was commercially available
fluorescent bulbs (Panasonic, Inc). Mice were exposed to indirect illumination with
different intensities (i.e., 0.5 lx, 15 lx, or 150 lx). For complete darkness (i.e., 0 lx), the
source of light was turned off. The light intensity was confirmed by a light meter
(Lutron Electronics, Inc.). The IPL for mice defined in this study (15 lx) was too weak
to influence on the temperature in the cage. Mouse pupils were not pharmacologically
manipulated. To evaluate direct effects of light on RPE cells, unilateral closed-eye
model was used, where unilateral eyes of the same mice were closed, the mice were
maintained for 24 h in the dark, and followed by exposure to various intensities of light
for 3−9 h.
Measurement of VEGF-A protein level in RPE/choroid− Extraocular tissues of the
enucleated mouse eyes were removed. Eye-cups were made in a standard manner by
making incision behind the corneal limbus and en bloc resection of anterior parts of the
eyes. For the separation of the tissues including the retina, RPE and choroid, incubation
in dispase solution was used. RPE/choroid was separated from retina, and was
homogenized and resolved in 100 μl RIPA buffer with a protease inhibitor. The total
VEGF-A protein (pg) per mg of the extracted RPE/choroid tissue was calculated using
ELISA development kits (R&D Systems, Minneapolis, MN).
Real-time RT-PCR− RNA from homogenized samples of RPE or choroid was extracted
using Trizol reagent (Invitrogen). cDNA was prepared using Superscript III for RT-PCR
(Invitrogen, Carlsbad, CA). Real-time PCR was carried out by LightCycler 480 Sybr
Green (Roche Applied Science). Values for each gene were normalized to expression
levels of GAPDH. The sequences of the primers used for the real-time RT-PCR were
designed to cross introns (Supplemental Table Ⅰ).
Western blot analysis− Total protein extracts from RPE/choroid samples were
separated on SDS-PAGE gels, and transferred onto nitrocellulose membranes that were
blocked with 5% non-fat dry milk in TBS-T buffer (Tris buffer saline plus 0.1%
Tween-20). Incubation with primary antibodies was performed for 1 hr in TBS-T buffer
containing 5% non-fat dry milk. Primary antibodies used were mouse antibody to
β-actin (SIGMA, A5316), RPE65 (Santa Cruz sc-53489), rhodopsin (Santa Cruz
sc-57433), HIF-1α (R&D Systems, MAB1536), and PGC-1α (Calbiochem, ST1202).
The membranes were incubated with anti-mouse-horseradish peroxidase labeled
secondary antibody (Amersham Biosciences) for 1 hr. The membranes were developed
with ECL Plus Western Blotting Detection Reagents (GE Healthcare). The protein level
of interest was calculated by normalization to the level of β-actin.
OS Isolation and Treatment on ARPE-19 Cells− OS was isolated from normal bovine
eyes under the dim red light as described previously1 and stored in 10mM Na-phosphate,
pH 7.2, 0.1 M NaCl, 2.5% sucrose at -80℃. ARPE-19 cells were treated with OS by the
concentrations of 1x105, 1x106 or 1x107 outer segments/ml of growth medium to test
dose dependency.
Luciferase Assay− VEGF promoter activity was tested with a Luciferase
assay system (Promega) according to the manufacturer’s instructions.
Plasmids containing human VEGF promoter sequence between -1180 and
+338 relative to the transcription start site were a kind gift from Akiyama
H2 .
Knockdown of PGC-1α and ERR-α− siRNAs designed to specifically
knock down PGC-1α and ERR-α (B-Bridge International) were transfected
to ARPE-19 cells using X-tremeGENE siRNA transfection reagent (Roche)
according to the manufacturer’s instruction.
Laser-induced CNV Model− The CNV lesions were induced by laser photocoagulation
as has been previously described3 with modifications. A glass cover served as a contact
lens. Diode-laser (DC-3000; NIDEK, Osaka, Japan) irradiation was delivered through a
slit lamp (SL150; Topcon, Tokyo, Japan). Spot size of 200 µm, power of 200 mW, and
exposure duration of 20 msec was delivered to the mouse fundus between the major
retinal vessels. Breaking of the Bruch's membrane was confirmed by central bubble
formation immediately after the photocoagulation. After the laser treatment, the mice
were maintained under cyclic light of various intensities or constant dark. On day 7 the
eyes were enucleated after systemic perfusion with fluorescein isothiocyanate dextran in
PBS. RPE/choroid flatmounts were made and CNV area size was measured. An
anti-VEGF-A neutralizing antibody (2 ng, R&D Systems) was administered
intravitreously on day 1 after CNV induction.
Statististics− All statistical analyses were carried out by JMP 9 software (SAS Institute
Inc.). For comparison between unpaired 2 groups, Student’s t-test and Fisher’s exact
test were used. Paired t-test was used for comparison between paired 2 groups. For
comparison among 3 or more groups, one-way analysis of variance (ANOVA) was
conducted, followed by Tukey’s test or Dunnett’s test. P < 0.05 meant statistical
significance.
Supplemental References
1. Molday RS, Hicks D, Molday L. Peripherin. A rim-specific membrane protein of rod
outer segment discs. Invest Ophthalmol Vis Sci. 1987;28:50-61.
2. Akiyama H, Tanaka T, Maeno T, Kanai H, Kimura Y, Kishi S, Kurabayashi M.
Induction of VEGF gene expression by retinoic acid through Sp1-binding sites in
retinoblastoma Y79 cells. Invest Ophthalmol Vis Sci. 2002;43:1367−1374.
3. Iriyama A, Inoue Y, Takahashi H, Tamaki Y, Jang WD, Yanagi Y. A2E, a
component of lipofuscin, is pro-angiogenic in vivo. J Cell Physiol. 2009;220:469−475.
Supplemental Table Ⅰ. Primers used in this study
Gene
mouse GAPDH
Primer sequence
Forward 5’-CACATTGGGGGTAGGAACAC-3’
Reverse 5’-AACTTTGGCATTGTGGAAGG-3’
mouse VEGF-A
Forward 5’-GCCAGCACATAGAGAGAATGAGC-3’
Reverse 5’-CAAGGCTCACAGTGATTTTCTGG-3’
human GAPDH
Forward 5’-TTGATTTTGGAGGGATCTCG-3’
Reverse 5’-GAGTCAACGGATTTGGTCGT-3’
human VEGF-A
Forward 5’-CCCTGATGAGATCGAGTACATCTT-3’
Reverse 5’-AGCAAGGCCCACAGGGATTT-3’
human HIF-1α
Forward 5’-TCCATGTGACCATGAGGAAA-3’
Reverse 5’-CCAAGCAGGTCATAGGTGGT-3’
human PGC-1α
Forward 5’-GTGAAGACCAGCCTCTTTGC-3’
Reverse 5’-TCACTGCACCACTTGAGTCC-3’
human MCAD
Forward 5’-TTGAGTTCACCGAACAGCAG-3’
Reverse 5’-TCCAAGTCCAAGACCTCCAC-3’
human PDK4
Forward 5’-CCTTTGGCTGGTTTTGGTTA -3’
Reverse 5’-CCTGCTTGGGATACACCAGT-3’
human CD36
Forward 5’-CCTGCAGCCCAATGGTGCCA-3’
Reverse 5’-ACCAACTGTGGTAGTAACAGGGTACGG-3’
Supplemental Figure Ⅰ. OS phagocytosis by RPE increases upon IPL. (A)
RPE/choroid lysates were immunoblotted for rhodopsin to compare the degree of OS
phagocytosis by RPE. Following preconditioning with 0.5-lx cyclic light, WT mice
were exposed to 0.5 or 15 lx of light for 3 h in the morning. (B) To understand the
circadian-independent effect of IPL, WT mice underwent unilateral eye closure,
followed by maintenance in the dark for 24 h. Then, mice were exposed to 0.5 or 15 lx
of light during the night for 3 h. Lysates of the RPE/choroid were immunoblotted for
rhodopsin to compare the degree of OS phagocytosis by RPE cells between the open
and closed eyes of the same mice.
Supplemental Figure Ⅱ. VEGF-A mRNA expression was not induced by IPL in
RPE of rd1/rd1 or Crx−/− mice. The effect of IPL was examined in the unilateral
closed-eye model (n = 5; ns, not significant by paired t-test).
Supplemental Figure Ⅲ. Induction of PGC-1α/ERR-α pathway target genes in
ARPE-19 cells treated with OS. PDK4, MCAD and CD36 mRNA in ARPE-19 cells
treated with or without OS (mean ± SEM; n= 3 per group; *P < 0.05, **P < 0.01, by
Student’s t-test).
Supplemental Figure Ⅳ. Knockdown efficiency of siRNAs for PGC-1α and ERR-α
was evaluated by mRNA level in ARPE-19 cells (mean ± SEM; n = 4 per group; ***p
< 0.001, by Student’s t-test).
Supplemental Figure Ⅴ. OS structure is intact and OS phagocytosis increases upon
IPL in PGC-1α−/− or ERR-α−/− mice. On immunohistochemistry, intact OS structure
was depicted by an antibody for rhodopsin. On western blotting, RPE/choroid lysates
were immunoblotted for rhodopsin and β-actin. The samples were from the open and
the closed eyes of the unilateral closed-eye model mice upon IPL (GCL, ganglion cell
layer; INL, inner nuclear layer; ONL, outer nuclear layer).
Supplemental Figure Ⅵ. VEGF-A mRNA expression was not induced by IPL in
RPE of PGC-1α−/− or ERR-α−/− mice. The effect of IPL was examined in the unilateral
closed-eye model (n = 5; ns, not significant by paired t-test).