Content uploaded by Eva Chamorro
Author content
All content in this area was uploaded by Eva Chamorro on Jan 23, 2019
Content may be subject to copyright.
Photochemistry and Photobiology, 2013, 89: 468–473
Effects of Light-emitting Diode Radiations on Human Retinal Pigment
Epithelial Cells In Vitro
Eva Chamorro*
1
, Cristina Bonnin-Arias
1
, María Jesús Pérez-Carrasco
2
, Javier Muñoz de Luna
3
,
Daniel Vázquez
3
and Celia Sánchez-Ramos
1,2
1
Neuro-Computing and Neuro-Robotics Research Group, Universidad Complutense de Madrid, Madrid, Spain
2
Optometry and Vision Department, Escuela Universitaria de Óptica, Madrid, Spain
3
Optics Department, Escuela Universitaria de Óptica, Madrid, Spain
Received 26 June 2012, accepted 30 August 2012, DOI: 10.1111/j.1751-1097.2012.01237.x
ABSTRACT
Human visual system is exposed to high levels of natural and
artificial lights of different spectra and intensities along
lifetime. Light-emitting diodes (LEDs) are the basic lighting
components in screens of PCs, phones and TV sets; hence it
is so important to know the implications of LED radiations
on the human visual system. The aim of this study was to
investigate the effect of LEDs radiations on human retinal
pigment epithelial cells (HRPEpiC). They were exposed to
three light–darkness (12 h/12 h) cycles, using blue-468 nm,
green-525 nm, red-616 nm and white light. Cellular viability
of HRPEpiC was evaluated by labeling all nuclei with DAPI;
Production of reactive oxygen species (ROS) was determined
by H2DCFDA staining; mitochondrial membrane potential
was quantified by TMRM staining; DNA damage was deter-
mined by H2AX histone activation, and apoptosis was evalu-
ated by caspases-3,-7 activation. It is shown that LED
radiations decrease 75–99% cellular viability, and increase
66–89% cellular apoptosis. They also increase ROS produc-
tion and DNA damage. Fluorescence intensity of apoptosis
was 3.7% in nonirradiated cells and 88.8%, 86.1%, 83.9%
and 65.5% in cells exposed to white, blue, green or red light,
respectively. This study indicates three light–darkness (12 h/
12 h) cycles of exposure to LED lighting affect in vitro
HRPEpiC.
INTRODUCTION
Biologic chromophores of retinal pigment epithelium (RPE) cells
can absorb the electromagnetic visible light radiation (380–780 nm).
But this luminous energy, necessary for the visual process, can cause
a toxic effect, especially the most energetic radiations of the visible
spectrum: the violet and blue (400–500 nm) (1). Short wavelength
light can penetrate through tissues to the cells and their organelles,
inducing the generation of reactive oxygen species (ROS) in RPE
mitochondria and even apoptosis, potentially caused by ROS-dam-
aged mitochondrial DNA (2).
The human visual system is exposed to a limited number of
natural and artificial lights of different spectra and intensities.
Light pollution is increasing exponentially, and energy-efficient
light sources as light-emitting diodes (LEDs) have been devel-
oped as an option to replace the traditional light bulbs. In the
coming years, incandescent light sources will be progressively
replaced by LEDs, and it is estimated that by 1 September 2016
there will be no incandescent lights in Europe (3).
White LEDs present specific spectral and energetic character-
istics compared with that of other domestic light sources, so the
potential risks of these new light sources need to be explored to
answer whether they could be eventually harmful for the eye (3).
The purpose of the work was to study the effects of LED
lighting on RPE cells. Outcome measures included cell viability,
oxidative stress, mitochondrial membrane potential, DNA
damage and apoptosis.
MATERIALS AND METHODS
Cell culture of human RPE. The human retinal pigment epithelial cell
line HRPEpiC (ScienceCell Research Laboratories), was grown in a low-
serum epithelial cell culture medium (ScienceCell Research Laboratories).
After the primary cultures became confluent, the cells were detached
from the culture dish with the use of the Trypsin/EDTA solution (Sigma-
Aldrich). Cells were plated in a 96-well, black clear Imaging Plate
(Becton, Dickinson and Company) with Poly-L-lysine (Sigma-Aldrich)
Coating (density =5000 cells/well). The cells were incubated in a
humidified atmosphere of 5% CO
2
and 95% air at 37°C, and the culture
medium was changed every 24 h, following each light phase.
Light exposure. Illumination was produced by a LED-based system.
The cells plated in the imaging plate were exposed to three light–dark-
ness (12 h/12 h) cycles, using blue light (468 nm), green light (525 nm),
red light (616 nm) or white light in well chambers (light intensity was
5mWcm
2
). Measures of phototoxicity were taken after the last
darkness phase of the total exposure cycle.
Although this value is not very frequent in daylife situations, it can be
found in several cases. Moreover, we chose this value to compare our
results with other studies about this subject (4,5). This value implies
34.150 lux for an incandescent light source or 33.446 lux for a D65
(skylight) light source. It is similar to the horizontal irradiance for a lying
person looking upwards in a clear sky day when the sun is around 37.5°
(6) or a person at 20 cm of a 100 W incandescent lamp (7).
The control group consisted of RPE cells kept in darkness. Figure 1
shows a schematic diagram of the LED lighting irradiation system, and
the spectral irradiance of LED lighting.
Cell viability. The cell nuclei were labeled by incubating the cells
with the nuclear stain 4′6-diamidine-2-phenylindole dihydrochloride,
DAPI (Sigma-Aldrich), for 1 h. The viable cells were counted under
a BD Pathway 855 fluorescence microscope (Becton, Dickinson and
Company), and an analysis of the image data was performed using
Attovision software (Becton, Dickinson and Company) (Table 1).
Measurement of intracellular ROS production. Oxidative stress was
measured using the dye (5-(and-6)chloromethyl-2′,7′-dichlorodihydrofluo-
*Corresponding author email: eva.chamorro@opt.ucm.es (Eva Chamorro)
© 2012 Wiley Periodicals, Inc.
Photochemistry and Photobiology © 2012 The American Society of Photobiology 0031-8655/13
468
rescein diactate acetyl ester H2DCFDA (Invitrogen, Germany) at a final
concentration of 1:1000 for 30 min at 37°C in darkness. Excess dye was
removed by washing in PBS. Fluorescence intensity was measured in a
BD Pathway 855 Bioimager (Becton, Dickinson and Company) using an
excitation band pass filter at 492–495 nm and an emission cutoff filter at
517–527 nm.
Measurement of mitochondrial membrane potential (MMP). Mito-
chondrial damage was assessed by using the dye Tetramethylrhodamine,
methyl ester, TMRM (Invitrogen) at a final concentration of 1:1000 for
30 min at 37°C in darkness. Excess dye was removed by washing in
PBS. Fluorescence intensity was measured in a BD Pathway 855 Bioim-
ager (Becton, Dickinson and Company) using an excitation band pass fil-
ter at 549 nm and an emission cutoff filter at 572 nm.
Immunocytochemical detection of histone H2AX and activated cas-
pase-3 and -7. DNA damage and apoptosis were evaluated by immuno-
cytochemistry, evaluating the activation of histone H2AX and caspases-3
and -7. At given time periods, cells were washed with phosphate-buffered
saline, (PBS; Sigma-Aldrich) and fixed with 4% paraformaldehyde
(Sigma-Aldrich) for 1 h. Cells were suspended in 0.3% Triton X-100-
PBS (Sigma-Aldrich) in a 3% bovine serum albumin (BSA; Sigma-
Aldrich) 1% (wt/vol) in PBS for 30 min to suppress. The cells were then
incubated in 2.5% PBS +BSA containing either a combination of 1:400
diluted antiphospho-histone H2AX (Abcam, UK) and 1:400 anticaspase-3
rabbit antibody (Cell Signaling Technology). The cells were then incu-
bated for 1 h and washed twice with PBS, and resuspended in 1:400
diluted goat antimouse Alexa Fluor 633 conjugated (Invitrogen) and
1:400 diluted goat antirabbit Alexa Fluor 488 (Invitrogen) for 30 min at
room temperature in darkness. After three washing steps, the fluorescence
of the samples was measured in the Pathway 855 automated fluorescence
microscope (Becton, Dickinson and Company) using an excitation band
pass filter at 632 nm and an emission cutoff filter at 647 nm for caspase-
3, -7 detection. For histone H2AX detection, an excitation band pass filter
at 488 nm and an emission cutoff filter at 594 nm were used.
Statistical analysis. Every experiment was repeated three times. The
values were given as mean ±SD. Data were analyzed using an unpaired
two-tailed t-test by Statgraphics version Centurion XVI.I. A Pvalue less
than 0.05 was considered statistically significant.
RESULTS
Cell viability
Nonirradiated RPE cells grew properly, but the irradiation inhib-
ited the growth of RPE cells. The difference in the cell number
of RPE cells irradiated by blue, green or white LED lighting and
nonirradiated was statistically very significant (P<0.01). Maxi-
mum damage was observed in cells exposed to blue LED light-
ing. In the experiments, 99%, 88% and 75% of the irradiated
cells became nonviable after their exposure to blue, green or
white light. Red light caused a slight decrease of number of RPE
cells. However, the difference in cell number of RPE cells irradi-
ated by red light and not irradiated was statistically insignificant
(Figs. 2A, 3A and 4A).
Measurement of intracellular ROS production
Low level production of reactive oxygen species was observed
in RPE cells maintained in darkness. However, a significant
increase in the level of ROS was observed after three light–dark-
ness cycles (12 h/12 h) with blue light, green light or red light.
Nonincrease of cellular cytoplasm fluorescence was detected in
cells exposed to white LED lighting in comparison with non
irradiated cells (Figs. 2B, 3B and 4B).
Measurement of mitochondrial membrane potential
After three light–darkness cycles of irradiation, no significant
effect on mitochondrial membrane potential was detectable
compared to control cells for any of the different LED lighting
(Figs. 2C, 3C and 4C).
Effects of light on DNA damage of RPE
Significant DNA damage was observed for light-exposed RPE
cells. The fluorescence microscopic data for all irradiated RPE
Figure 1. Schematic diagram of the LED lighting irradiation system and
spectral irradiance of the different LED lighting sources: blue, green, red
and white light.
Table 1. Cell viability, reactive oxygen species ROS production, mitochondrial membrane potential, DNA damage and apoptosis of cultured RPE irradi-
ated with blue, green, red and white LED lighting. Values indicate fluorescence intensity, mean ±SD.
Control Blue light Green light Red light White Light
Viability (FU) 855 ±403 10 ±2* 99 ±114* 339 ±1 217 ±108*
ROS (FU) 593 ±78 737 ±19* 855 ±30* 1004 ±49* 656 ±26
MMP (FU) 634 ±19 620 ±39 823 ±30 780 ±128 770 ±18
DNA damage (FU) 131 ±41 2537 ±589* 2258 ±738* 1920 ±286* 2697 ±493*
Apoptosis (%) 3.7 ±0.02 86.1 ±0.03* 83.9 ±0.05* 65.5 ±0.07* 88.8 ±0.02*
*P<0.05 compared to the control.
Photochemistry and Photobiology, 2013, 89 469
cells show the increased degradation of nucleic acids in compari-
son with the control cells. Maximum damage was showed to
cells exposed to blue LED lighting (Figs. 2D, 3D and 4D).
Detection of apoptosis
The percentage of apoptotic cells was increased on light-exposed
RPE cells in comparison with RPE cells maintained in darkness.
The death of nonirradiated RPE cells reached a frequency of
3.7%. However, cell death was 86%, 84%, 66% and 89% for
blue, green, red and white-irradiated RPE, respectively (Figs. 2E,
3E and 4E).
DISCUSSION
Epidemiological studies suggest an association between visible
light exposure and increased risk of advanced age-related macu-
lar degeneration (AMD). Visible light can affect the retina and
RPE by photochemical, thermal and mechanical mechanism (8).
Photochemical damage occurs when the incident radiation has
a wavelength in the high energy portion of the visible spectrum.
An electron in an excited state can return to the inhibited state
dissipating the extra energy. One way to dissipate this energy is
to break a bond in another molecule through a direct exchange
of electron or direct exchange of hydrogen producing reactive
oxygen species (ROS) (2,4,9). A proposed mechanism of cell
damage induced by light is the oxidative process (10). The outer
layers of the retina are continuously exposed to high levels of
oxygen due to the abundant blood supply of the choriocapillaries
(1,3). The formation of ROS at the level of the RPE leads to cell
damage with the subsequent degeneration of photoreceptors (11).
Noell (1980) was the first to observe that the action spectrum of
retinal damage induced by light was similar to the action spec-
trum of rhodopsin (scotopic sensitivity), thus suggesting that rho-
dopsin or its photoproducts were acting as mediators in the
retinal damage (8,12). Subsequent studies have supported this
mechanism (13,14).
Experimental evidence has demonstrated that the retina and
RPE are much more sensitive to blue light damage than red or
green light (9,15,16). Most research works have been focused
on evaluating the response of the retina to light from conven-
tional lighting sources as halogen or fluorescent. It has been
speculated that LED lighting radiation may cause ocular damage
(3), however, the potential risks of these new light sources has
not been explored. In this study, we have demonstrated that
LED lighting can damage RPE cells. The results of this study
clearly show that LED lighting radiation decreases by 75–99%
the cellular viability and increases by 66–89% the cellular apop-
tosis, as well as there is an increase in the production of ROS
and DNA damage.
These results are consistent with previous reports that sug-
gest that visible light of conventional light sources could cause
Figure 2. Representative images of effects of LED lighting on human retinal pigment epithelial cells in vitro. HRPEpiC cells were exposed to blue,
green, red and white LED lighting (irradiated cells) or maintained in darkness (control) for three light–darkness cycles (12 h/12 h). (A) Cellular viability
of HRPEpic cells determined by labeling all nuclei with DAPI. (B) ROS production determined by the H2DCFDA staining and fluorescence micros-
copy; an increase of fluorescence in cells indicates oxidative stress. (C) Mitochondrial membrane potential determined by the TMRM staining and fluo-
rescence microscopy. Reduction or absence of fluorescence indicates decrease of MMP. (D) DNA damage determined by the activation of H2AX
histone. (E) Apoptosis determined by the activation of caspases-3,-7. The white arrows indicate apoptotic cells.
470 Eva Chamorro et al.
Figure 3. Effects of monochromatic LED lighting on human retinal pig-
ment epithelial cells in vitro. HRPEpiC cells were exposed to blue, green
and red LED lighting (irradiated cells) or maintained in darkness (con-
trol) for three light–darkness cycles (12 h/12 h). The graph displays
mean fluorescence intensity radios of irradiated cells versus non irradiated
controls. Bars represent mean ±SD from n=3–5 experiments. The
asterisk (*) indicates significant differences as compared to controls
(P<0.05, Student’st-test). (A) Cellular viability of HRPEpic cells deter-
mined by labeling all nuclei with DAPI. (B) ROS production determined
by the H2DCFDA staining and fluorescence microscopy. (C) Mitochon-
drial membrane potential determined by the TMRM staining and fluores-
cence microscopy. (D) DNA damage determined by the activation of
H2AX histone. (E) Apoptosis determined by the activation of caspases-3,
-7 is observed as a pink coloration around DAPI-stained cells.
Figure 4. Effects of white LED lighting on human retinal pigment epithe-
lial cells in vitro. HRPEpiC cells were exposed to white LED lighting (irra-
diated cells) or maintained in darkness (control) for three light–darkness
cycles (12 h/12 h). The graph displays mean fluorescence intensity radios
of irradiated cells versus non irradiated controls. Bars represent mean ±SD
from n=3–5 experiments. The asterisk (*) indicates significant differences
as compared to controls (P<0.05, Student’st-test). (A) Cellular viability
of HRPEpic cells determined by labeling all nuclei with DAPI. (B) ROS
production determined by the H2DCFDA staining and fluorescence micros-
copy. (C) Mitochondrial membrane potential determined by the TMRM
staining and fluorescence microscopy. (D) DNA damage determined by the
activation of H2AX histone. (E) Apoptosis determined by the activation of
caspases-3, -7 is observed as a pink coloration around DAPI-stained cells.
Photochemistry and Photobiology, 2013, 89 471
cell damage. Sparrow et al. (5) analyzed human RPE cells irra-
diated with blue light (430 nm, 8 mW cm
2
), green light
(550 nm, 8 mW cm
2
) and white light (246 mW cm
2
). The
light was delivered from a tungsten halogen source for 20 min
and it was observed that illuminated RPE cells remained viable.
In another study, Godley et al. exposed confluent cultures of
human primary retinal epithelial cells to visible light (390–
550 nm at 2.8 mW cm
2
) of a metal halide lamp for 0–9h
and analyzed cell viability and ROS production. Cells main-
tained in the absence of blue light exposure showed no
decrease in viability, no mitochondrial or nuclear DNA damage
and low level production of ROS; however, blue light-irradiated
cells showed an increasing loss of viability (ca 10%), time-
dependent increase in the levels of ROS and maximal mito-
chondrial DNA damage 3 h after exposure with evidence of
some repair mechanism (1).
On the other hand, Chu et al. studied changes on viability of
RPE as a result of blue and red halogen light irradiation. Early
passages of human RPE cells were exposed to blue light
(460 nm, 0.4 mW cm
2
) and red light (640 nm, 1 mW cm
2
)
for 48 h. Cell viability was not significantly affected by blue-
light irradiation or red-light irradiation at low doses (17). Later
on, Youn et al. investigated light-induced retinal damage in
human RPE cells exposed to specific narrow wavebands of blue
light obtained using interference filters and an arc lamp system
(400 nm at an irradiance of 1.555 mW cm
2
, 420 nm at an irra-
diance of 1.466 mW cm
2
and 435.8 nm at an irradiance of
1.351 mW cm
2
) for 3–12 h. Cells exposed to 400 nm light
showed decrease in cell viability, degradation of mitochondria
and nucleic acids damage; however, no alterations were observed
for 420 and 435.8 nm light-exposed RPE cells (18).
Of relevance is the research carried out by Roehlecke et al. in
which they evaluated the in vitro response of RPE cells exposed
to blue LED lighting. Cells were irradiated with 405 nm light at
an output power of 0.3 or 1 mW cm
2
for 3, 24 or 72 h. The
data showed a significantly stimulated ROS production and a
decrease of mitochondrial membrane potential after 24 h of
exposure to blue light, but no apoptosis or viability changes were
evidenced. They used low doses of light for up to 72 h without
a repair time, to establish an in vitro model system in which light
irradiation induced mild stress without causing cell death (2).
It has been suggested that cells may adapt themselves to the
light-induced stress and therefore survive (2) so in the present
study we have exposed cells to three light–darkness cycles
(12 h/12 h) instead of continuous light.
It is relevant that ROS production was the highest in cells
irradiated with the red light, not correlating with DNA damage
or apoptosis where blue and green light produce more phototoxic
effects.
From this we can infer that ROS are not the only elements
responsible for cell damage and apoptosis. Other photosensitive
molecules have been studied, mitochondrial respiratory chain
enzymes (13,19,20), melanin (21) and products of intermediate
intermedia genes (10,11,14). However, our results regarding
ROS must be considered with caution since the presence of
photosensitizers such as riboflavin in the culture medium can
influence on light-dependent ROS generation (22–24).
Summing up, three light–darkness cycles (12 h/12 h) expo-
sure to LED lighting affect the growth of RPE cells, produce
cellular stress increasing ROS levels accompanying an increase
of DNA damage and apoptotic cells. Future investigation will
determine the intensities and wavelengths of LED lighting which
are lethal and nonlethal for ocular tissues, as well as the effect of
optical filters in RPE cell protection. This information will be
necessary to develop appropriate normative for this growing
industry field.
Acknowledgement—This work has been supported in part by Fundación
Mapfre (Spain).
REFERENCES
1. Godley, B. F., F. A. Shamsi, F. Q. Liang, S. G. Jarrett, S. Davies
and M. Boulton (2005) Blue light induces mitochondrial DNA dam-
age and free radical production in epithelial cells. J. Biol. Chem.
280, 21061–21066.
2. Roehlecke, C., A. Schaller, L. Knels and R. H. Funk (2009) The
influence of sublethal blue light exposure on human RPE cells. Mol.
Vis. 15, 1929–1938.
3. Behar-Cohen, F., C. Martinsons, F. Vienot, G. Zissis, A. Barlier-
Salsi, J. P. Cesarini, O. Enouf, M. Garcia, S. Picaud and D. Attia
(2011) Light-emitting diodes (LED) for domestic lighting: any risks
for the eye? Prog. Retin. Eye Res. 30, 239–257.
4. Hui, S., L. Yi and Q. L. Fengling (2009) Effects of light exposure
and use of intraocular lens on retinal pigment epithelial cells in vitro.
Photochem. Photobiol. 85, 966–969.
5. Sparrow, J. R., A. S. Miller and J. Zhou (2004) Blue light-absorbing
intraocular lens and retinal pigment epithelium protection in vitro.
J. Cataract Refract. Surg. 30, 873–878.
6. Vazquez, D. and E. Bernabeu (1997) Quantitative estimation of clear
sky light in Madrid. Energy Build. 26, 331–336.
7. ORAM (2005) General Light Brochure. OSRAM, Spin.
8. Wu, J., S. Seregard and P. V. Algvere (2006) Photochemical damage
of the retina. Surv. Ophthalmol. 51, 461–481.
9. Ham, W. T., H. A. Mueller and D. H. Sliney (1976) Retinal
sensitivity to damage from short wavelength light. Nature 260, 153–
155.
10. Wenzel, A., C. Grimm, A. Marti, N. Kueng-Hitz, F. Hafezi, G. Nie-
meyer and C. E. Reme (2000) c-fos controls the “private pathway”
of light-induced apoptosis of retinal photoreceptors. J. Neurosci. 20,
81–88.
11. Grimm, C., A. Wenzel, F. Hafezi and C. E. Reme (2000) Gene
expression in the mouse retina: the effect of damaging light. Mol.
Vis. 6, 252–260.
12. Noell, W. K. (1980) Possible mechanisms of photoreceptor damage
by light in mammalian eyes. Vision. Res. 20, 1163–1171.
13. Pautler, E. L., M. Morita and D. Beezley (1990) Hemoprotein(s)
mediate blue light damage in the retinal pigment epithelium. Photo-
chem. Photobiol. 51, 599–605.
14. Reme, C. E. (2005) The dark side of light: rhodopsin and the silent
death of vision the proctor lecture. Invest. Ophthalmol. Vis. Sci. 46,
2671–2682.
15. Ham, W. T., H. A. Mueller, J. J. Ruffolo and A. M. Clarke (1979)
Sensitivity of the retina to radiation-damage as a function of wave-
length. Photochem. Photobiol. 29, 735–743.
16. Dorey, C. K., F. C. Delori and K. Akeo (1990) Growth of cultured
RPE and endothelial cells is inhibited by blue light but not green or
red light. Curr. Eye Res. 9, 549–559.
17. Chu, R., X. Zheng, D. Chen and D. N. Hu (2006) Blue light irradia-
tion inhibits the production of HGF by human retinal pigment epi-
thelium cells in vitro. Photochem. Photobiol. 82, 1247–1250.
18. Youn, H. Y., B. R. Chou, A. P. Cullen and J. G. Sivak (2009)
Effects of 400, 420, and 435.8 nm radiations on cultured human
retinal pigment epithelial cells. J. Photochem. Photobiol. B 95,
64–70.
19. Beatty, S., H. Koh, M. Phil, D. Henson and M. Boulton (2000) The
role of oxidative stress in the pathogenesis of age-related macular
degeneration. Surv. Ophthalmol. 45, 115–134.
20. Suter, M., C. Reme, C. Grimm, A. Wenzel, M. Jaattela, P. Esser, N.
Kociok, M. Leist and C. Richter (2000) Age-related macular degen-
eration. The lipofusion component N-retinyl-N-retinylidene ethanol-
amine detaches proapoptotic proteins from mitochondria and induces
472 Eva Chamorro et al.
apoptosis in mammalian retinal pigment epithelial cells. J. Biol.
Chem. 275, 39625–39630.
21. Marshall, J. (1985) Radiation and the ageing eye. Ophthalmic Phys-
iol. Opt. 5, 241–263.
22. Grzelak, A., B. Rychlik and G. Bartosz (2001) Light-dependent gen-
eration of reactive oxygen species in cell culture media. Free Radic.
Biol. Med. 30, 1418–1425.
23. Wang, R. J. (1976) Effect of room fluorescent light on the deteriora-
tion of tissue culture medium. In Vitro 12,19–22.
24. Mahns, A., I. Melchheier, C. V. Suschek, H. Sies and L. O. Klotz
(2003) Irradiation of cells with ultraviolet-A (320–400 nm) in the
presence of cell culture medium elicits biological effects due to
extracellular generation of hydrogen peroxide. Free Radic. Res. 37,
391–397.
Photochemistry and Photobiology, 2013, 89 473