JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICS

Volume 26, Number 1, 2010

© Mary Ann Liebert, Inc.
DOI: 10.1089/jop.2009.0035

Ocular Safety of Infliximab in Rabbit and Cell Culture Models

Fabrizio Giansanti,1 Laura Papucci,2 Sergio Capaccioli,2 Daniela Bacherini,1 Lorenzo Vannozzi,1
Ewa Witort,2 and Ugo Menchini1

Abstract
Purpose: To undertake the safety testing of infliximab in animal and cell culture models.
Methods: Sixteen New Zealand albino rabbits were divided into 4 groups of 4 animals. Each group received 2
mg of infliximab in the right eye and saline solution in the left eye. Clinical examination and retinal histology
were performed at months 1, 2, and 3 for groups A, B, and C, respectively. Electroretinography (ERG) recordings
were made before the injection, and at months 1 and 2 for group D. The assessment of the safety of infliximab in
retinal pigment epithelial (RPE) and retinal ganglion (RGC-5) cells was performed by Western blot analysis of
caspase-3 involved in apoptosis pathway, the analysis of cytotoxicity of infliximab by WST-1 proliferation assay,
and time-lapse video-microscopy to register the morphology of treated cells in time-lapse.
Results: Clinical examination of the eyes, histological study of the retina, and ERG recordings showed no sign
of ocular toxicity after a single intravitreal injection of 2 mg of infliximab in the rabbit model. RPE and RGC via-
bility was not reduced either in the range of concentration (up to 3 mg) or during the follow-up interval.
Conclusion: This study suggests that infliximab has no direct retinal toxicity using rabbit (2 mg) and cell culture
(2 and 3 mg) models.

Introduction
choroidal neovascularization in an animal model11 and ben-

T umor necrosis factor alpha (TNF-α) is a 17-kDa pep-

tide cytokine playing a key role in inflammation and
immunity.1 A number of confirmations from animal and
human studies have indicated that TNF-α is a potentially
important mediator in uveitis.2,3
TNF-α is mainly released by macrophages, T lymphocytes,
and natural killer cells in response to different inflamma-
tory agents.1 During inflammation, TNF-α mediates leu-
kocytic infiltration, dendritic cells maturation, macrophages
activation, and drives T-helper Type 1 cell responses within
tissues in experimental autoimmune uveitis (EAU).4
Increasing evidence suggests that diabetic retinopa-
thy has an underlying inflammatory element manifested
by leukocyte involvement, adhesion to the retinal vessels,
and up-regulation of inflammatory genes.5,6 Even though
diabetic macular edema is mainly the result of fluid store
due to capillary leakage, the etiology of diabetic macular
edema is not wholly understood. An inflammatory ele-
ment exists, and some studies have pointed out an altered
local expression of TNF-α in this disease.6–10 The role of
TNF-α in retinal angiogenesis is under investigation at the
moment. Previous studies demonstrated the regression of
efits to patients affected with neovascular age-related macu-
lar degeneration12 after anti-TNF-α therapy.
A drug that specifically targets TNF-α may represent
an interesting therapeutic approach for patients with ocu-
lar inflammation.13 After the initial report of the favorable
effects of infliximab in the sight-threatening panuveitis of
patients with Behcet’s disease14 was released, many other
studies of its off-label use, in a series of patients with various
forms of uveitis, reported beneficial results with anti-TNF
treatment.3
Infliximab (Remicade®, Roche, GB), is a human–mouse
chimaeric IgG1 antibody, administered systemically, that
inhibits the active soluble TNF-α implicated in the inflam-
matory process. Serious side effects have also been described
when infliximab is administered systemically, including
opportunistic intracellular infection, infusion reaction, neu-
tropenia, thrombocytopenia, and increased malignancy
risk. The intravitreal administration of infliximab could
decrease the systemic adverse side effects and improve the
safety and efficacy of the drug in selected patients affected
with ocular inflammatory disease. Recently, Theodossiadis
and colleagues15 reported their favorable preliminary

1
Department of Oto-Neuro-Ophthalmological Surgical Sciences, Eye Clinic, and 2Department of Experimental Pathology and
Oncology, University of Florence, Florence, Italy.

65
66 GIANSANTI ET AL.

experience of intravitreal administration of 2 mg infliximab
in patients affected with neovascular age-related macular
degeneration.
Retinal safety is of primary importance when using an
intravitreal drug. The aim of our experimental study was to
confirm the retinal safety of intravitreal infliximab at 2 mg
in a rabbit model and to undertake safety testing of different
doses of infliximab using a cell culture model (retinal gan-
glion (RGC-5) and retinal pigment epithelial (RPE) cell lines).
Methods
Experiment 1: Animal model
Animals. Sixteen New Zealand albino rabbits weighing
between 1.8 and 2.3 kg were selected for this study. The rab-
bits were divided into 4 groups of 4 animals. Groups A, B, C,
and D were used to evaluate ocular safety at different times
after intravitreal infliximab injection. The animals were
treated according to the Association for Research in Vision
and Ophthalmology guidelines. Slit-lamp examination and
indirect ophthalmoscopy were performed on all eyes before
the study. Animals showing corneal or lens opacity or reti-
nal damage before the study were excluded.
Rabbits were anesthetized with a mixture of ketamine
hydrochloride (50 mg/kg) and xylazine hydrochloride
(5 mg/kg). Benoxinate hydrochloride (0.4%) was applied
for topical anesthesia. Eyes were dilated with topical tropi-
camide (1%) applications. Fluoroquinolone antibiotic was
used topically before and after all surgical procedures.
Intravitreal injection. Infliximab (Remicade, Roche, GB)
was reconstituted with sterile saline solution. A single inflix-
imab dose of 2 mg was tested on all groups, injected into the
right eye. Left eyes were treated with sterile saline solution.
In all cases, the injection volume was 0.1 mL. Intravitreal
injection was performed ~2 mm posterior to the limbus
with a 30-gauge needle attached to a tuberculin syringe. A
few seconds later, 0.1 mL of aqueous humor was withdrawn
from the anterior chamber via a paracentesis performed by a
12 o’clock limbal insertion of a 30-gauge needle connected to
a 1-mL syringe. The rabbits were kept for 12 weeks in ambi-
ent light on a 12-h light/dark schedule.
Clinical examination. Slit-lamp examination and indirect
ophthalmoscopy were used to evaluate changes in the cor-
nea, lens, vitreous, retina, and optic nerve. Such observations
were performed on all eyes immediately after the injections
and at months 1, 2, and 3. Pupils were dilated prior to the
ophthalmic examination. Experimental and control eyes
from all animals were graded for alterations using a quanti-
tative clinical grading scale16 (see Table 1) at months 1, 2, and
3. In most cases, gradings were carried out by 2 examiners.
Histological examination. Groups A, B, and C were eutha-
nized with an overdose of pentobarbital 1, 2, and 3 months
after injection, respectively. The eyes were immediately
enucleated, and the sclera incised 2 mm posterior to the
limbus and fixed in a solution of 10% formalin. Eyes were
sectioned, the tissues processed and embedded in paraffin,
sectioned at a thickness of 5 μm, and stained with hematox-
ylin–eosin. Histological examination was carried out using
light microscopy and was performed by a blinded patholo-
gist. Thickness measurements of the retinal layer and retinal
ganglion cells counts were performed in the experimental
and control eyes. Signs of focal necrosis, vacuolization, and
inflammatory cell infiltration were also investigated.
Electroretinography. The eyes of group D were evaluated
by electroretinography (ERG) before injection and 4 and 8
weeks after injection. Pupils were dilated with tropicamide,
and the animals were dark-adapted for at least 2 h. Standard
ERGs were recorded in both eyes using corneal electrodes
(CSO Ophthalmic Instruments, Florence, Italy). Reference

Table 1. Quantitative Clinical Grading Scale

Tissue response Score 0 Score 1 Score 2 Score 3 Score 4

I. Cornea
a. Transparency opacity Clear Mild Moderate (iris Severe (bare Opaque (no iris
visible) iris detail) view)
b. Vessels None <1 mm 1–3 mm from 3–4 mm from Central
limbus limbus involvement
c. Abscess None <1 mm 1–2 mm 3–4 mm >5 mm
II. Anterior chamber
a. Protein flare None Trace Mild Moderate Severe
b. Inflammatory cells None Trace Mild Moderate Severe
c. Fibrin None Mild strand Moderate Severe strands No iris details
strands <15% >15%
d. Hemorrhage None Mild Moderate >50% 8-ball
III. Iris
a. Blood vessels None Mild Moderate Severe Neovascular
IV. Vitreous
a. Protein flare None Trace Mild Moderate Severe
b. Opacities None Cells Multiple clumps Red reflex Opaque
V. Retinal detachment None 25% 50% 75% Total
VI. Optical media Clear View of vessels Sharp red reflex Dull red reflex Totally opaque
outlines only

Source: From Meredith TA et al. (1990).16

OCULAR SAFETY OF INFLIXIMAB
and ground electrodes were made of stainless steel surgical
needles and were inserted into the ears. ERG signals were
recorded using Retimax (CSO Ophthalmic Instruments,
Florence, Italy). Dark-adapted scotopic responses and sco-
topic flash responses were recorded simultaneously in both
eyes. Flashes varied in intensity from −2.50 to +0.5 log
scot cd s/m2. At each scheduled examination, ERGs were
repeated 3 times for each rabbit. The amplitudes of a- and
b-waves were calculated for each recording.
Experiment 2: Cell culture model
RPE cell line. Cells of the human RPE line ARPE-19 were
obtained from the American Type Culture Collection (ATCC;
Manassas, VA). ARPE-19 is an immortalized cell line that is
spontaneously produced by the culture of human RPE.17 The
cells were cultured in a humidified incubator at 37°C in 5%
CO2 in 10% fetal bovine serum-defined minimal essential
(FBS-DMEM)-F12 medium supplemented with 100 U/mL
penicillin G and 100 μg/mL streptomycin.
RGC-5 cell line. Retinal ganglion cell (RGC)-5 cell line,18–20
developed by transformation of RGCs post-natal Sprague-
Dawley rats, was grown in Dulbecco’s modified Eagle’s
medium (DMEM; Gibco, Carlsbad, CA) supplemented with
10% heat-inactivated FBS (Gibco) and 100 U/mL of penicil-
lin and 100 μg/mL of streptomycin (Gibco). Cells were pas-
saged every 2–3 days, and the cultures incubated at 37°C in
5% CO2. For all experiments cells were used at an 80% of cell
confluence.
Western blot analysis. The assays were performed accord-
ing to standard conditions. In short, following treatments
with infliximab (2 and 3 mg), cells were washed twice in
ice-cold PBS, re-suspended in 100 μL of ice-cold RIPA lysis
buffer, vortexed for 3 s, and incubated on ice for 30 min. The
protein lysates were obtained by centrifugation at high speed
for 20 min at 4°C to separate nonsoluble cell debris. Proteins
(20 μg/lane) were separated by NuPAGE® 12% Bis–Tris Gel,
electroblotted (Trans-Blot® Semi-Dry apparatus; Bio-Rad,
Hercules, CA) onto Protran® nitrocellulose transfer mem-
brane (Schleicher & Schuell, Germany). The blots were probed
with the mouse monoclonal anti-caspase-3 antibody (Santa
Cruz Biotechnology, Santa Cruz, CA) for RPE cell line or
with the rabbit polyclonal anti-caspase-3 (Upstate-Millipore,
Billerica, MA) for RGC-5 cell line. The mouse monoclonal
anti-α-tubulin antibody (Sigma-Aldrich, St. Louis, MO) was
used as protein loading control. Following treatment with
the secondary antibody, goat anti-mouse or mouse anti-rab-
bit IRDye® 700CW (Li-Cor Bioscience, Lincoln, NE), respec-
tively, according to the instruction manual, protein bands
were analyzed by Odyssey Infrared Imaging System (Li-Cor)
using the software for protein quantification.
Cell viability assay: WST-1. The viability of ARPE-19 and
RGC-5 cells was determined by a cell proliferation assay
using WST-1 reagent (Roche, Milano, Italy). WST-1 is a water-
soluble sulfonated tetrazolium salt that is cleaved by cellu-
lar succinate hydrogenases in living cells, yielding dark blue
formazan. Damaged or dead cells exhibit reduced or no
dehydrogenase activity. In brief, RPE and RGC-5 cells were
plated onto a 96-multiwell plate in quadruple and treated or
not with infliximab (2 and 3 mg) for 24, 48, or 72 h. WST-1
solution/culture medium (5 mmol/L, 1:9) was added to
each well. Following 2-h incubation at 37°C, absorbance at
450 nm (reference at 630 nm) was measured by a Multiskan
67
JX microplate reader. Percentage of cell viability was calcu-
lated based on the absorbance measured relative to that of
the untreated control cells maintained in culture medium
alone.
Time-lapse video-microscopy. Each cell line was treated
with or without infliximab (2 and 3 mg) and the apoptotic
events were registered for 72 h and counted by time-lapse
video-microscopy using a Zeiss-inverted phase-contrast
microscope equipped with a 10× objective, Panasonic CCD
cameras and JVC BR9030 time-lapse video recorders, as
previously reported.21 An apoptotic event was registered at
the moment the cell shrank completely and split into apop-
totic bodies.
Results
Experiment 1
Clinical observation. One rabbit from group A showed
trace of protein flare in the anterior chamber of both eyes.
The average score of group A at month 1 was 0.25 in the
experimental eyes and 0.25 in the control eyes.
One rabbit from group B showed trace of protein flare
in both eyes at month 1. The average score of Group B was:
0.25 at month 1 in the experimental eye and 0.25 in the con-
trol eye; 0 at months 2 and 3 in the experimental and control
eyes.
A rabbit from group D showed trace of protein flare in
the vitreous humor of the experimental eye at month 1. The
average score of group D was: 0.25 at month 1 in the exper-
imental eye and 0 in the control eye; 0 at months 2 and 3 in
both eyes.
Three rabbits showed subconjunctival hemorrhage
observed immediately after the injection.
In summary, there was no clinical evidence of ocular
damage in rabbits from any of the groups.
Histological findings. Retinal histology was examined by
light microscopy. There were no differences in histological
Experimental Eye Control Eye
Month 1
Month 3
FIG. 1. Light microscopy examination of the retina per-
formed after intravitreal injection of 2 mg of infliximab and
sterile saline solution into the experimental eye and into the
control eye after 1 and 3 months, respectively. No differ-
ence between experimental and control eyes was detected.
Magnification, 400×; bar, 50 μm.
68 GIANSANTI ET AL.

appearance between the eyes injected with infliximab and
those injected with sterile saline solution in groups A, B,
and C (Fig. 1). No signs of focal necrosis, vacuolization, or
inflammatory cell infiltration were detected in the experi-
mental and control eye.
The statistical significance of baseline differences
between experimental and control eye (retinal ganglion cell
number and differences in retinal thickness) as well as of
change from baseline to months 1 and 2 was computed tak-
ing into account within-subject correlation in a linear-mixed
model with rabbit as a random effect. Stata 11.0 software was
used for calculations (StataCorp, College Station, TX).
The retinal ganglion cell (RGC) number did not differ
between control and experimental eye at month 1 (difference
−0.45, 95% CI: −2.35 to 1.44, P = 0.69), month 2 (difference in
RGC number −0.38, 95% CI: −2.13 to 1.36, P = 0.66) or month
3 (difference 0.43, 95% CI: −1.95 to 2.81, P = 0.72).
The retinal thickness did not differ between control and
experimental eye at month 1 (difference: 1.49 μm, 95% CI:
−6.64 to 9.64; P = 0.71), month 2 (difference: −3.79 μm, 95%
CI: −10.84 to 3.25, P = 0.29), or month 3 (difference: −2.77
μm, 95% CI: −12.74 to 7.19, P = 0.58).
The mean retinal thickness was 152.4 μm (95% CI:
138.4–166.4) in the experimental eyes and 152.4 μm (95% CI:
136.9–164.9) in the control eyes at month 1; 131.2 μm (95%
CI: 119–143.3) in the experimental eyes and 135 μm (95% CI:
122.8–147.1) in the control eyes at month 2; 124.6 μm (95% CI:
107.4–141.8) in the experimental eyes and 127.4 μm (95% CI:
110.2–144.5) in the control eyes at month 3.
Electroretinography. ERG changes were considered sig-
nificant if the follow-up differences in amplitude (a- and
b-waves) were lower than 30% compared to pre-injection
values.22 Figure 2 displays the percentage of variation of a-
and b-amplitude differences in all experimental eyes (group
D) with respect to preinjection. In all cases, signal varia-
tions were not higher than 30%. The statistical significance
of baseline differences between experimental and control
eyes (a- and b-amplitude difference) as well as of change
from baseline to months 1 and 2 was computed taking into
account within-subject correlation in a linear-mixed model
with rabbit as a random effect. The baseline differences

A ERG Results With 0.5 log Stimulus

30 A ERG 0.5 log Stimulus
180
Difference in (a and b) Amplitude
20
Amplitude Difference (%)

170
10
160
(microV)

0
150

–10
140

–20
130 Experimental Eye
R1 R2 R3 R4
–30 Control Eye
120
–2 0 2 4 6 8 0 4 8
Time (Weeks) Weeks

B ERG Results With –2.5 log Stimulus B ERG –2.5 log Stimulus
30 150
Difference in (a and b) Amplitude

20 140
Amplitude Difference (%)

10
(microV)

130

0
120
–10
110 Experimental Eye
–20 Control Eye
R1 R2 R3 R4
–30 0 4 8
–2 0 2 4 6 8 Weeks
Time (Weeks)
FIG. 3. Electroretinography (ERG) amplitude variation
FIG. 2. Electroretinography (ERG) amplitude variation in between experimental and control eyes from all rabbits
the experimental eyes with respect to pretreatment with a stimulated with (A) +0.5 log units of cd s/m2 and (B) −2.5
stimulus of +0.5 log units of cd s/m2 (A) and −2.5 log units of log units of cd s/m2. The y-axis reports a- and b-amplitude
cd s/m2 (B). Individual rabbit ERG amplitudes are plotted. The difference (μV) at each measurement time. The first point
y-axis reports the percentage difference of a- and b-amplitude (before zero) is the pretreatment measure. Dots and squares
of posttreatment recordings with respect to pretreatment. The indicate mean values; error bars represent standard error
first point (before zero) is the pretreatment measure. from the mean.
OCULAR SAFETY OF INFLIXIMAB
between experimental and control eyes (a- and b-amplitude
difference) were not statistically significant (P = 0.924). The
changes from baseline between experimental and control
eyes to months 1 and 2 were nonstatistically significant
(P = 0.735 and 0.889, respectively). Figure 3 displays the a-
and b-amplitude difference in all experimental eyes with
respect to control eyes at each time point.
Experiment 2
Cellular apoptotic death of ARPE-19 and RGC-5 cells,
involving the intrinsic death pathway initiated at the level
of the mitochondrion, was detected through activity of the
effector caspase-3. Caspase-3 has been reported to activate
death effector molecules resulting in the fragmentation of
genomic DNA in association with structural and morpho-
logical changes characteristic of apoptosis. The enzymatic
activity of caspase-3 is usually detected as a reduction in the
amount of pro-enzyme and/or appearance of smaller pep-
tides, products of pro-caspase cleavage. In protein lysates of
both ARPE-19 and RGC-5 cell lines treated with infliximab
(2 and 3 mg) analyzed by WB analysis with anti-caspase-3
specific antibody, the levels of pro-caspase-3 observed as a
protein band of 32 kDa remained stable and comparable to
untreated control cell lines (Fig. 4). This shows that inflix-
imab, even at a relatively high dose, did not induce apoptotic
death program in both tested cell lines.
The cell proliferation reagent WST-1 (Roche) was used to
analyze possible cytotoxic activity of 2 or 3 mg/mL of inf-
liximab in RPE and RGC-5 cell lines by the spectrophoto-
metric quantification of cell proliferation, using the 96-well
format as described in the manufacturer’s protocol. At least
3 experiments were performed and each sample’s absor-
bance at 450 nM was analyzed in quadruple by Multiskan JX
microplate reader. The results presented in Figure 5 demon-
strate the absence of any statistically significant cytotoxicity
A RPE
0 24 48
α-tubulin
Caspase-3
Infliximab 3 mg/mL
B RPE
0.25
Caspase 3/Tubulin (ratio)
0.20
0.15
0.10
0.05
0.00
0 24 48 72
69
of the infliximab administered to both cell lines at the dose
of 3 mg/mL tested.
The possibility that infliximab could lead to enhance-
ment of the apoptotic events has been further evaluated by
time-lapse video-microscopy employing microscopic video
registration. The number of apoptotic events occurring in
ARPE-19 and RGC-5 cells cultured in a medium supple-
mented with infliximab (2 and 3 mg) have been quantified
during 24, 48, and 72 h of microscopic video registration
(Fig. 6). An apoptotic event was considered positive at the
moment the cell detached from the substrate, shrank, and
blebbed. The ARPE-19 and RGC-5 cells cultured in normal
growth medium served as untreated control. The number
of apoptotic events registered did not differ between cells
treated with infliximab and nontreated controls, and never
exceeded the physiological number of apoptotic cells even
after 72 h of treatment.
Discussion
TNF-α is implicated in the pathogenesis of a variety of
inflammatory processes and, recently, anti-TNF therapy has
also been able to arrest experimental choroidal neovascu-
larization and reduce macular edema.10,11 In ophthalmology,
the clinical usefulness of anti-TNF therapy is under investi-
gation. If its efficacy is confirmed, it could be advantageous
in blocking only the intraocular TNF-α so as to minimize
the adverse systemic events. Recently, Theodossiadis et al.15
reported their preliminary results of intravitreal administra-
tion of infliximab into 3 patients affected with neovascular
age-related macular degeneration. Two intravitreal injec-
tions of 2 mg of infliximab were performed on each patient.
The patients were followed up for 7 months and neither local
nor systemic adverse events were noted.
Safety of intraocular infliximab up to 1.7 mg was previ-
ously investigated in a pilot study23 in a rabbit model by our
RGC-5
72 0 24 48 72 hours
FIG. 4. Analysis of caspase-3 pro-
tein expression. (A) Western blot anal-
ysis. Protein extracts were separated
by 12% SDS-PAGE, blotted onto a
nitrocellulose membrane, and probed
with mouse monoclonal anti-cas-
pase-3 antibody for retinal pigment
epithelial (RPE) cells and with rab-
RGC-5 bit polyclonal anti-caspase-3 for reti-
nal ganglion (RGC-5) cells. A mouse
monoclonal anti-α-tubulin antibody
was used as protein loading control.
(B) Densitometric analysis of cas-
pase-3 activity. Bands were analyzed
by Odyssey Infrared Imaging System
(Li-Cor) using the software for pro-
tein quantification.
0 24 48 72 hours
70 GIANSANTI ET AL.

A WST-1 % of Living Cells B WST-1 Absorbance

RPE RGC-5
100 0.25

80 0.20
% of Living Cells

Absorbance (450 nm)

60 0.15

40 0.10
RPE Untreated
0.05 RPE +Infliximab 3 mg/mL
20
RGC-5 Untreated
RGC-5 +Infliximab 3 mg/mL
0 0.00
24 48 72 24 48 72 0 10 20 30 40 50 60 70 60
Untreated Infliximab 3 mg/mL Hours

FIG. 5. Effect of treatment of 3 mg/mL of infliximab on cell proliferation of retinal ganglion (RGC-5) and ARPE-19 cell
lines. The cell proliferation reagent WST-1 was used to analyze cytotoxic activity of infliximab. Data were expressed as
means ± SD from 4 experiments. P > 0.05 vs. non-treated cells. Percentage of living cell (A) and absorbance (B).

group. In that study, 3 groups of 3 rabbits received a dose of Two-milligram dose was selected because it was used in
1, 1.7, and 3.3 mg, respectively. We detected signs of retinal humans.15 Our results indicate that 2 mg infliximab admin-
toxicity only in the histological samples at a dose of 3.3 mg. istered intravitreally in rabbit eyes did not display ocular
In addition, Theodossiadis and colleagues24 demonstrated toxicity when examined by histology or slit-lamp biomicros-
the safety of a single intravitreal injection of infliximab up copy, indirect fundoscopy, and ERG during the follow-up.
to a dose of 2 mg, in 2 groups of 4 rabbits treated, respec- To our knowledge an analysis of the safety of infliximab
tively, with 1 and 2 mg of infliximab and followed up for 45 in RPE and RGC cells has never been reported. In our study
days. Doses of 5 mg or higher revealed morphological and (experiment 2), we reported the effects of the infliximab
functional changes of the retina. exposure on 2 cell lines, ARPE-19 and RGC-5, with differ-
In the present study (experiment 1), we have aimed to ent doses (2 and 3 mg), and demonstrated the absence of
confirm the safety profile of 2 mg of intravitreal infliximab any type of cytotoxic activity of infliximab in our cellular
by studying 4 groups (A, B, C, and D) of 4 rabbits followed model. To demonstrate this we used Western blot analysis
up for 3 (groups A, B, and C) and 2 (group D) months. of caspase-3 involved in apoptotic death pathway,25 analysis
of cytotoxicity of infliximab by WST-1 proliferation assay,
and time-lapse video-microscopy to register morphology of
Infiximab 3 mg/mL treated cells in time-lapse.
25
In our study we could not detect any proteolytic activity
RGC-5 + Infliximab of caspase-3, the well-known effector of the apoptotic pro-
20 RGC-5 Untreated gram. The cleaved subunits were not detected by western
Cumulative Apoptotic Cells

RPE + Infliximab
RPE Untreated
15
10
5 detectable proteolytic activity as a result of the absence of
0
0 10 20
FIG. 6. Time-lapse video-microscopic analysis of inflix-
imab administration on cell viability. The apoptotic events
were registered for 72 h and counted by 3 independent oper-
ators both in infliximab-treated and non-treated controls.
Data were expressed as means ± SD from at least 3 experi-
ments. P > 0.05 vs. non-treated cells.
blot analysis with the antibodies used. These observations
suggest that our antibodies were incapable of detecting
these cleaved fragments; this could be due to an insufficient
amount of cleaved material present at the time point in which
the cells were harvested for preparation of protein lysates,
or the cells might clear these fragments rapidly, as has been
observed in other studies,26 or most probably, there was no
an activated apoptotic program. Since we observed a close
correlation between the time-lapse results that showed no
significant number of apoptotic cells in both cell lines, this
30 40 50 60 70 confirms the failure of infliximab to induce apoptotic death.
Hours Infliximab was also proved safe in our cellular model since
we could not detect any significant cytotoxicity or inhibition
of cell proliferation between treated and nontreated controls
with WST-1 assay.
In conclusion, our experimental findings confirm
the safety of intravitreal infliximab when used at 2 mg.
Furthermore, infliximab showed no nondesired or unex-
pected effects on cellular physiology in our cellular model of
OCULAR SAFETY OF INFLIXIMAB
RGC-5 and ARPE-19 lines up to a dose of 3 mg. These results,
with those from earlier studies, provide the safety dosage
(up to 2 mg) for the use of intravitreal infliximab in future
clinical studies.
Acknowledgments
The authors thank Dr. N. Agarwal who kindly gifted
RGC-5 cell line and Prof. G. Virgili for assistance with sta-
tistics. Supported by a grant from the Cassa di Risparmio di
Firenze Foundation, Florence, Italy.
Author Disclosure Statement
No competing financial interests exist.
References
1. Bazzoni, F., and Beutler, B. The tumor necrosis factor ligand and
receptor families. N. Engl. J. Med. 334:1717–1725, 1996.
2. Ooi, K.G., Galatowicz, G., Calder, V.L., et al. Cytokines and
chemokines in uveitis: is there a correlation with clinical phe-
notype? Clin. Med. Res. 4:294–309, 2006.
3. Theodossiadis, P.G., Markomichelakis, N.N., and Sfikakis, P.P.
Tumor necrosis factor antagonists: preliminary evidence for an
emerging approach in the treatment of ocular inflammation.
Retina. 27:399–413, 2007.
4. Dick, A.D., Forrester, J.V., Liversidge, J., et al. The role of tumour
necrosis factor (TNF-alpha) in experimental autoimmune uveo-
retinitis (EAU). Prog. Retin. Eye Res. 23:617–637, 2004.
5. Joussen, A.M., Murata, T., Tsujikawa, A., et al. Leukocyte-
mediated endothelial cell injury and death in the diabetic ret-
ina. Am. J. Pathol. 158:147–152, 2001.
6. Joussen, A.M., Poulaki, V., Le, M.L., et al. A central role for
inflammation in the pathogenesis of diabetic retinopathy.
FASEB J. 18:1450–1452, 2004.
7. Limb, G.A., Webster, L., Soomro, H., et al. Platelet expression of
tumour necrosis factor-alpha (TNF-alpha), TNF receptors and
intercellular adhesion molecule-1 (ICAM-1) in patients with pro-
liferative diabetic retinopathy. Clin. Exp. Immunol. 118:213–218,
1999.
8. Limb, G.A., Hollifield, R.D., Webster, L., et al. Soluble TNF recep-
tors in vitreoretinal proliferative disease. Invest. Ophthalmol.
Vis. Sci. 42:1586–1591, 2001.
9. Doganay, S., Evereklioglu, C., Er, H., et al. Comparison of serum
NO, TNF-alpha, IL-1beta, sIL-2R, IL-6 and IL-8 levels with
grades of retinopathy in patients with diabetes mellitus. Eye.
16:163–170, 2002.
10. Sfikakis, P.P., Markomichelakis, N., Theodossiadis, G.P., et al.
Regression of sight-threatening macular edema in type 2 dia-
betes following treatment with the anti-tumor necrosis factor
monoclonal antibody infliximab. Diabetes Care. 28:445–447,
2005.
11. Shi, X., Semkova, I., Müther, P.S., et al. Inhibition of TNF-alpha
reduces laser-induced choroidal neovascularization. Exp. Eye
Res. 83:1325–1334, 2006.
12. Markomichelakis, N.N., Theodossiadis, P.G., and Sfikakis, P.P.
Regression of neovascular age-related macular degeneration
71
following infliximab therapy. Am. J. Ophthalmol. 139:537–540,
2005.
13. Nussenblatt, R. Treating intraocular inflammatory disease in
the 21st century. Arch. Ophthalmol. 123:1000–1001, 2005.
14. Sfikakis, P.P., Theodossiadis, P.G., Katsiari, C.G., et al. Effect of
infliximab on sight-threatening panuveitis in Behçet’s disease.
Lancet. 358:295–296, 2001.
15. Theodossiadis, P.G., Liarakos, V.S., Sfikakis, P.P., et al. Intravitreal
administration of the anti-tumor necrosis factor agent inflix-
imab for neovascular age-related macular degeneration. Am. J.
Ophthalmol. 147:825–30, 830.e1, 2009.
16. Meredith, T.A., Trabelsi, A., Miller, M.J., et al. Spontaneous
sterilization in experimental Staphylococcus epidermidis
endophthalmitis. Invest. Ophthalmol. Vis. Sci. 31:181–186, 1990.
17. Dunn, K.C., Aotaki-Keen, A.E., Putkey, F.R., et al. ARPE-19, a
human retinal pigment epithelial cell line with differentiated
properties. Exp. Eye Res. 62:155–169, 1996.
18. Krishnamoorthy, R.R., Agarwal, P., Prasanna, G., et al.
Characterization of a transformed rat retinal ganglion cell line.
Brain Res. Mol. Brain Res. 86:1–12, 2001.
19. Maher, P., and Hanneken, A. The molecular basis of oxidative
stress-induced cell death in an immortalized retinal ganglion
cell line. Invest. Ophthalmol. Vis. Sci. 46:749–757, 2005.
20. Agar, A., Li, S., Agarwal, N., et al. Retinal ganglion cell line apop-
tosis induced by hydrostatic pressure. Brain Res. 1086:191–200,
2006.
21. Luzi, E., Papucci, L., Schiavone, N., et al. Downregulation of
bcl-2 expression in lymphoma cells by bcl-2 ARE-targeted mod-
ified, synthetic ribozyme. Cancer Gene Ther. 10:201–208, 2003.
22. Manzano, R.P., Peyman, G.A., Khan, P., et al. Testing intravitreal
toxicity of bevacizumab (Avastin). Retina. 26:257–261, 2006.
23. Giansanti, F., Ramazzotti, M., Vannozzi, L., et al. A pilot study
on ocular safety of intravitreal infliximab in a rabbit model.
Invest. Ophthalmol. Vis. Sci. 49:1151–1156, 2008.
24. Theodossiadis, P.G., Liarakos, V.S., Sfikakis, P.P., et al. Intravitreal
administration of the anti-TNF monoclonal antibody infliximab
in the rabbit. Graefes Arch. Clin. Exp. Ophthalmol. 247:273–281,
2009.
25. Tempestini, A., Schiavone, N., Papucci, L., et al. The mecha-
nisms of apoptosis in biology and medicine: a new focus for
ophthalmology. Eur. J. Ophthalmol. 13(Suppl 3):S11–S18, 2003.
26. Zhang, Y., and Bhavnani, B.R. Glutamate-induced apoptosis in
primary cortical neurons is inhibited by equine estrogens via
down-regulation of caspase-3 and prevention of mitochondrial
cytochrome C release. BMC Neurosci. 6:13, 2005.
Received: April 8, 2009
Accepted: October 12, 2009
Address correspondence to:
Dr. Fabrizio Giansanti
Department of Oto-Neuro-Ophthalmological Surgical Sciences
Eye Clinic
University of Florence
Viale Morgagni 85
Florence 50134
Italy
E-mail: fabrizio.giansanti@unifi.it