Advances in Bioscience and Biotechnology, 2013, 4, 47-55 ABB
http://dx.doi.org/10.4236/abb.2013.410A4005 Published Online October 2013 (http://www.scirp.org/journal/abb/)
Reduction of corneal scarring in rabbits by targeting the
TGFB1 pathway with a triple siRNA combination
Sriniwas Sriram1, Daniel Gibson2, Paulette Robinson2, Sonal Tuli3, Alfred S. Lewin4, Gregory Schultz1
1Department of Biomedical Engineering, University of Florida, Gainesville, USA
2Institute for Wound Research, Department of Obstetrics and Gynecology, University of Florida, Gainesville, USA
3Department of Ophthalmology, University of Florida, Gainesville, USA
4Department of Molecular Genetics and Microbiology, University of Florida, Gainesville, USA
Received 30 August 2013; revised 19 September 2013; accepted 5 October 2013
Copyright © 2013 Sriniwas Sriram et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Purpose: The transforming growth factor beta1
(TGFB1) pathway has been linked to fibrosis in sev-
eral tissues including skin, liver, kidney and the cor-
nea. In this study, a RNA interference-based approach
using siRNAs targeting three critical scarring genes,
TGFB1, TGFB receptor 2 (TGFBR2) and connective
tissue growth factor (CTGF), was tested for effects on
reducing alpha smooth muscle actin (SMA) and cor-
neal scarring (haze) in excimer laser ablated rabbit
corneas. Methods: Levels of TGFB1 and CTGF mRNAs
were measured using qRT-PCR in the epithelial and
endothelial cell layers of normal and excimer ablated
rabbit corneas at 30 minutes, 1 day and 2 days after
ablation. Two different scarring models were utilized
to assess the effects of the triple siRNA combination
on corneal scarring. In the first model, rabbit corneas
were unevenly ablated creating a mesh pattern then
treated immediately with the triple siRNA combina-
tion. After 1 day the ablated areas of corneas were
collected and levels of mRNAs for TGFB1, TGFBR2
and CTGF were measured. After 14 days, levels of
mRNA for SMA were measured and SMA protein im-
munolocalized in frozen sections. In the second model,
rabbit corneas were uniformly ablated to a depth of
155 microns followed by three daily doses of the triple
combination of siRNA. After 14 days, corneas were
photographed and images were analyzed using Image
J software to assess corneal scarring. Corneas were also
analyzed for levels of SMA mRNA. Results: In both
unwounded and wounded corneas, levels of TGFB1
and CTGF mRNA were always significantly higher in
endothelial cells than in epithelial cells (10 to 30 fold).
Thirty minutes after injury, levels of both TGFB1
and CTGF mRNAs increased approximately 20-fold
in both epithelial and endothelial cells, and further in-
creased approximately 60-fold in 2 days. In the first
therapeutic experiment with a single siRNA dose, two
of three rabbits showed substantial reductions of all
three target genes after 1 day with a maximum knock
down of 80% of TGFb1, 50% reduction of TGFBR2
and 40% reduction of CTGF mRNA levels and reduc-
ed SMA mRNA at day 14. In the second therapeutic
experiment with multiple doses of siRNA treatment,
both rabbits showed a ~22% reduction in scar forma-
tion at day 14 as calculated by image analysis. There
was also a corresponding 70% and 60% reduction of
SMA RNA expression. Conclusion: These results de-
monstrate that both TGFB1 and CTGF dramatically
increase in rabbit corneal epithelial and endothelial
cells after injury. Treatment of excimer ablated rab-
bit corneas with a triple combination of siRNAs effec-
tively reduced levels of the target genes and SMA, lead-
ing to reduced corneal scarring at 14 days, suggesting
tha t this triple siRNA combination may be an effective
new approach to reducing scarring in cornea and
other tissues.
Keywords: RNA Interference; siRNA Combination;
Corneal Scarring; TGFB1; CTGF
1. INTRODUCTION
Corneal scarring remains a serious complication that can
ultimately lead to functional vision loss. In an injured
cornea, a cascade of molecular events is initiated by pro-
longed, elevated levels of transforming growth factor
beta (TGFB1) which then combines with the Transform-
ing Growth Factor Receptor II (TGFBR2) inducing the
synthesis of Connective Tissue Growth Factor (CTGF)
causing excessive scarring (corneal haze) that impairs vi-

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 48
sion. The TGF-β system, has emerged as a key compo-
nent of the fibrogenic response to wounding by regulat-
ing the transformation of quiescent corneal keratocytes
into activated fibroblasts that synthesize ECM and into
myofibroblasts that contract corneal matrix (Chen et al.
2000; Jester, Petroll, and Cavanagh 1999). These myofi-
broblasts are filled with alpha smooth muscle actin (SMA)
that forms microfilaments that are the major source of
light scattering in corneal scars [1]. CTGF acting as a
downstream mediator of TGFB1, down regulates synthe-
sis of corneal crystallin proteins in quiescent keratocytes
and up regulates synthesis of collagen. Thus, the exces-
sive scattering of light that is clinically described as cor-
neal scar and haze results from the combination of colla-
gen laid down in irregular pattern in the wound and opa-
que activated fibroblasts and myofibroblasts that no lon-
ger synthesize the corneal crystallin proteins that keep
their cytoplasm transparent.
We have previously shown the effect of PRK on CTGF
levels in rat and mouse corneas and found that CTGF
was present in all cell layers of the cornea. The levels of
CTGF were found to continually rise from the time of
wounding up through 28 days post wounding [1]. How-
ever, in order to completely understand the role of TGFB1
in tissues repair and scarring, it is essential to understand
the timing and site of the synthesis of these growth fac-
tors so that the appropriate cell layer is targeted for nu-
cleic acid therapies. The experiments in this study are ex-
pected to reveal when and where TGFB1 and CTGF are
synthesized after a corneal injury so that the best mode
of action for an anti-fibrotic therapy can be chosen.
There currently are no FDA approved drugs that selec-
tively reduce the expression of genes causing corneal
scarring and haze. At present, the methods used to de-
crease corneal haze are topically applied steroids or anti-
metabolite drugs that target the cells capacity to respond
to signaling. Mitomycin C is used during some ocular sur-
geries, but it may have very damaging side effects, such
as epithelial defects, stromal melting, endothelial damage,
and conjunctival thinning [2]. Hence, there is a need to
develop a targeted approach that can nullify the specific
molecular pathways that give rise to a scar.
It is however difficult to achieve significant therapeu-
tic effect by employing a one-target, one-drug paradigm
on such a complex, multi-factorial signaling pathway.
Hence, using a multi-target approach can interrupt or act
on the complex signaling network at multiple points, and
affect the cell in ways that an individual component can-
not [3]. In this study, we have tested a siRNA triple com-
bination targeting TGFB1, TGFBR2 and CTGF. This tri-
ple siRNA combination was shown to be effective in re-
ducing the expression of target (TGFB1, TGFBR2 and
CTGF) and downstream mediators like Collagen-I and
α-Smooth Muscle Actin (SMA) in both in vitro cell cul-
ture system and ex vivo organ cultures [4].
Additionally, it is important to deliver these siRNA
combinations to the corneal layer where there is high lo-
calization of the target growth factors after wounding. Al-
though most of the targeted anti-fibrotic approaches tar-
get the stromal fibroblasts due to the eventual presence
of myofibroblasts in this region, there has not been any
research on the post-wounding localization ofTGFB1 and
CTGF in the epithelium and the endothelium. A delivery
method that targets the corneal layer with maximum post-
wounding growth factor localization is critical for an ef-
fective therapy.
Delivery of drugs to the cornea is a major challenge,
as the mechanical barriers that protect the cornea (multi-
layered epithelium, tight junctions) constrain ocular drug
delivery [5]. The principal properties governing corneal
drug absorption are its lipophilicity, partition coefficient
and molecular size [6]. Nanocarriers are the potential solu-
tion for targeted ocular drug delivery as they have been
shown to be non-immunogenic, have relatively low toxi-
city, be resistant to protein/serum absorption and be sta-
ble in an enzymatic environment [7]. We have previously
showed the high efficacy of the nanoparticle kit used in
this study in delivering fluorescently labeled siRNA to all
layers of the cornea including the endothelium in an ex
vivo organ culture model [8].
The overall goal of this study is to test and deliver a
previously optimized effective triple siRNA combination
to the appropriate corneal layer with high post wounding
localization of TGFB1 and CTGFso that there is a maxi-
mal reduction of scar formation in rabbits.
2. METHODS
2.1. Laser Ablation of Rabbits
Adult New Zealand Rabbits free of disease were used and
treated according to ARVO Statement for the Use of Ani-
mals in Ophthalmic and Vision Research. Excimer abla-
tion and collection of corneas was performed as previ-
ously described [9]. Briefly, rabbits were anesthetized
with isoflurane inhalation, and proparacaine eye drops
provided topical anesthesia. Laser ablations were perform-
ed to both eyes of each rabbit with a Summit SVS exici-
mer laser that is committed to animal vision research. In
this study, two different approaches were tested to obtain
the most intense scarring in the rabbits. In the first ap-
proach, using the laser in phototherapeutic keratectomy
mode, the central 6 mm diameter area of the cornea was
ablated at a dose of 160 mJ/cm2 to an initial depth of 80
microns to remove the epithelium and then the final 45
microns were ablated by placing a mesh over the cornea
to make an uneven ablation. In the second approach, us-
ing the same laser parameters, the central 6 mm diameter
area of the cornea was ablated to an even depth of 155

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 49
microns.
The eyes were then pretreated with 50 μM EDTA for
10 minutes. A total of 150 μl of the nanoparticle complex-
ed with the siRNA triple combination was added to one
of the eyes while the other was treated with the vehicle
control and was considered as a paired negative control.
The eyes were held open for 3 minutes to allow the nano-
particle to penetrate the stroma before being disturbed.
No postoperative topical steroid was used to ensure that
the wound healing process is not altered with anti-inflam-
matory agents. Corneas were collected at different time
points according to the experiment, homogenized in a
pestle with liquid nitrogen and then transferred to TRIzol.
The RNA was then extracted using a hybrid RNA extrac-
tion protocol with RNeasy spin columns [10].
2.2. Gross Corneal Dissection
Rabbits without observable corneal wounds were anes-
thetized, excimer ablated to 125 microns and euthanized
at the designated time points as described before. A scal-
pel was used to immediately scrape the epithelium off
with care taken to ensure that the scraped mass was re-
tained on the blade. The scraped epithelial mass was then
transferred to 350 μl of tissue lysis buffer (Qiagen, buffer
RLT) and the blade was rinsed with 250 μl of additional
lysis buffer. The cornea was then excised from the globe
by cutting with a fresh scalpel and scissors at the corneal/
scleral boundary. The cornea was placed face down and
yet another fresh scalpel was used to scrape off and re-
tain the endothelium as was done with the epithelium.
The endothelial mass was transferred to 350 μl of lysis
buffer and the blade rinsed with an additional 250 μl of
lysis buffer. Each grossly isolated cellular layer was then
subjected to ultrasonication on ice for further tissue dis-
ruption. The probe was rigorously washed, rinsed and
dried in between each sample. The homogenates were
then immediately loaded onto Qiagen gDNA removal co-
lumns and the RNA was purified in accordance with the
manufacturer’s provided protocol (Qiagen RNA easy, Qi-
agen, Inc., Cat. #74104).
The purified RNA was quantified via ultra-violet ab-
sorbance using a Nanodrop ND-1000 spectrophotometer
set for RNA quantification.
2.3. Preparation of siRNA-Nanoparticles
A commercially available nanoparticle kit called Invivo-
plex® was purchased from Aparnabio (Rockville, MD) and
used according to manufacture’s instructions. Briefly,
siRNA triple combination solution was made at a con-
centration of 0.9 mg/ml. 600 μL of this solution was add-
ed with 300 μL of the provided cargo buffer. This solu-
tion was added drop wise to 900 μL of the given nano-
particles over a magnetic stirrer. This preparation forms
siRNA nanoparticles <50 nm that are stable for a week.
2.4. Reverse Transcription-Polymerase Chain
Reaction
Total RNA was extracted using the Qiagen RNeasy mini
isolation kit (Qiagen, Inc., Valencia, CA) and used accor-
ding to the manufacturer’s directions. cDNA was synthe-
sized using the High Capacity cDNA Reverse Transcrip-
tion Kit (Applied Biosytems, Carlsbad, CA) according to
manufacturer’s procedure. The level of mRNA for TGF-
B1, CTGF and SMA were determined using the Real-
Time PCR TaqMan assay. The primers and probes for
each gene are defined in Table A1. The endogenous con-
trol, ribosoma18S RNA and GAPDH was used normalize
target genes. Primers, probes and cDNA were combined
with TaqMan Universal PCR Master Mix (Applied Bio-
sytems, Carlsbad, CA) and amplification was performed
by the Applied Biosystems 7300 HT Fast Real Time PCR
System (Carlsbad, CA). A few samples were run without
reverse transcriptase to measure the quantity of genomic
DNA (gDNA) present in the sample. The thermal cycling
conditions were as follows: 2 min at 50˚C, 10 min at
95˚C, 40 cycles of 15 sec at 95˚C, and 1 min at 60˚C. The
relative gene expression of the growth factors was calcu-
lated using the 2-ΔΔCt method.
2.5. Macrophotography
Prior to general anesthesia, each eye was topically anes-
thetized with proparacaine and each pupil was dilated
with phenylephrine 2.5% and tropicamide eye drops.
Each rabbit was then generally anesthetized with inhaled
isoflurane as described earlier. The eyelids were held open
and out of the way with either an eyelid speculum or a
pair of cotton swabs. A Nikon D7000, was outfitted with
a macro lens capable of native 1:1 reproduction (either a
100 mm Tokina or 60 mm Nikkor) and the Nikon R1C1
Creative Lighting System (CLS) flash system. The D40
was set to the “Normal” program, ISO 200, manual ex-
posure with a shutter speed of 1/500 second and f/16.
While the 7000 was set to the “Standard” program, ISO
100, manual exposure with a shutter speed of 1/250 sec-
ond and f/18. To visualize and measure haze, the flash
power was set manually (1/16th, D40, 1/6.4th D7000) and
neither the flash nor lens had a filter. For all images, the
lens was set to manual focus and pre-focused to a 1:1 re-
production ratio and the camera was focused by moving
the camera closer or further from the subject. Guide lights
on the flash heads were used to facilitate haze visualiza-
tion and focusing.
2.6. Immunohistochemistry
The rabbit corneas from the experiments were fixed
overnight in 4% paraformaldehyde. They were then bi-

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 50
within the same TGFB1 pathway-TGB1, TGFBR2 and
CTGF [4,8].
sected, fixed in OCT and then sectioned in 10 μm slides.
The slides were then washed with PBS and blocked in
horse serum for 1 hour. Finally, they were incubated with
SMA antibody-Cy3 (Sigma) for 1 hour at room tempera-
ture. The slides were mounted with DAPI and imaged
using fluorescence microscope.
3.1. The Effect of PRK on mRNA Levels of
TGFB1 and CTGF
The corneas of 11 rabbits were evenly ablated to 125 mi-
crons using an excimer laser. 3 rabbits were sacrificed at
30 mins after ablation, and 4 rabbits each for Day1 and
Day2 post-ablation. The corneas of three rabbits were
unablated and used as control. At the designated time
points, the corneas were collected and the epithelium and
endothelial layers were scrapped using a surgical scalpel
and collected in separate tubes. The expression levels of
TGFB1 and CTGF were analyzed using qRT PCR. As
shown in Figure 1, there is an initial spike in the levels
of TGFB1 and CTGF as early as 30 minutes after abla-
tion. There is then a decrease in the expressions at Day 1
followed by an exponential increase on Day2. The ex-
pression trend of TGFB1 and CTGF follows a similar
trend in both the epithelial and endothelial layers. Figure
1(e) plots the RNA level expressions of TGFB1 and CTGF
in terms of the epithelium. The expressions of both TGFB1
and CTGF were consistently higher in the endothelial lay-
er when compared to the epithelium particularly at Day1
when the expression of CTGF in endothelium was ~35
times that of the epithelium. All expressions were calcu-
lated with respect to the unablated corneas and were nor-
2.7. Statistical Analysis
All experiments were performed in triplicate and all sta-
tistical analyses were conducted using Graph Pad prism
(San Diego, CA). Student’s t test or Analyses of Varianc-
es (ANOVA) with Tukey’s post-hoc assessments were
accordingly used to test for significance between the
groups. Results were considered statistically significant
where p < 0.05.
3. RESULTS
The levels of TGFB1 and CTGF were analyzed at sev-
eral time points following ablation to find the best time
to dose with anti-scarring drugs. Two different models of
scarring were tested to find which of the two generated
the most intense scaring in rabbits. Also in this study, the
in vivo efficacy of a previously optimized triple siRNA
combination that was effective in reducing downstream
scarring genes (SMA and collagen-I) in both in vitro cul-
ture and ex vivo organ cultures was evaluated. The triple
siRNA combination targets three critical scarring genes
(a) (b)
(c) (d) (e)
Figure 1. Post-ablation expression timeline of TGFB1 and CTGF. The corneas of 11 rabbits were evenly ablated to 125 microns
using an excimer laser. 3 rabbits were sacrificed at 30 mins after ablation, and 4 rabbits each for Day1 and Day2 post-ablation. The
corneas of three rabbits were unablated and were used as the control. The corneas were collected and the epithelium and endothelial
layers were scrapped using a scalpel and collected in separate tubes. The expression levels of TGFB1 and CTGF were analyzed using
qRT PCR. All expressions were calculated with respect to the unablated corneas and was normalized using GAPDH as the house-
keeping gene. Figure (e) plots the RNA expressionsof TGFB1 and CTGF in endothelium in terms of the epithelium.

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 51
malized using GAPDH as the housekeeping gene.
3.2. The Effective Triple Combination (T1R2C1)
Inhibits Target mRNA Accumulation in
Rabbits
The corneas of 9 rabbits were unevenly ablated to 125
microns using an excimer laser. The right eye was treated
with 150 μL of the effective triple combination (T1R2C1)
complexed with nanoparticles and the left eye received
equal volume of the vehicle control. One day later, 3 rab-
bits were humanely sacrificed and total RNA was extract-
ed for analysis by RT-PCR. The effective triple combina-
tion (T1R2C1) gave an average of knockdown of 57%
for TGFB1, 25% for TGFBR2 and 24% for CTGF (Fig-
ure 2). One of the rabbits (rabbit 1) had a maximum
knockdown of 80% for TGFB1, 57% for TGFBR2 and
46% for CTGF indicating some the siRNA combination
was effectively delivered to the corneal stroma in this
animal. The knockdown percentages were calculated with
respect to the left eye, which received vehicle control with-
out the siRNA.
3.3. SMA Immunohistostaining in Triple siRNA
Treated Rabbit Corneas
6 out of the 9 treated rabbits from the above experiment
were used for a long-term experiment to observe scar for-
mation. After 14 days, the intensity of scarring in both the
treated and the control eyes was graded by a masked op-
hthalmologist and was also imaged using a digital camera.
The rabbits were then humanely sacrificed and three cor-
Figure 2. Short-term knockdown of target growth factors. The
corneas of 3 rabbits were unevenly ablated to 125 microns us-
ing an excimer laser. The right eye was treated with the effec-
tive triple combination (T1R2C1) complexed with nanoparti-
cles and the left eye received the vehicle control. One day later,
the rabbits were sacrificed and RNA was extracted for analysis
by qRT PCR. The figure gives the RNA level knockdown per-
centages of the target growth factors calculated with respect to
the left eye. All expressions were normalized to 18S rRNA.
neas were collected for SMA immunohistostaining. The
corneas were fixed overnight in 4% paraformaldehyde.
They were then bisected, fixed in OCT, sectioned in 10
μm slides and stained for SMA. The treated cornea of two
of the three rabbits selected for immunohistostaining had
a lower haze grading score when compared to the control
cornea. The control eye that was ablated and treated with
vehicle control shows SMA staining in the basal epithe-
lium and stroma while the triple siRNA treated right eye
shows reduction in SMA staining (Figures 3(b) and (d)).
Three corneas from the 6 treated rabbits from the
above experiment were collected for RNA level analysis
by qRT PCR. SMA knockdown percentage of >40% was
observed in 2 out of the 3 rabbits. The untreated left eye
in these rabbits had haze-grading scores of 2 and 3 re-
spectively. No knockdown was observed in the other rab-
bit, which had a haze grading score of one in the un-
treated left eye. The RNA level knockdown percentage
of SMA was calculated with respect to the untreated left
eye and all expressions were normalized to 18S rRNA.
3.4. Macrophotography Images of Reduction in
Scarring by Repeated siRNA Dosing
The corneas of 3 rabbits were evenly ablated to 155 mi-
crons using an excimer laser. One of the rabbits had a
pre-existing scar and spots of neovascularization on the
Figure 3. SMA immunohistostaining in triple siRNA treated
rabbit corneas. The corneas of 6 rabbits were unevenly ablated
to 125 microns using an excimer laser. The right eye was treat-
ed with the effective triple combination (T1R2C1) complexed
with nanoparticles and the left eye received the vehicle control.
14 days later, the rabbits were sacrificed and three corneas were
collected for SMA immunohistostaining. The corneas were
fixed overnight in 4% paraformaldehyde. They were then bi-
sected, fixed in OCT and then sectioned in 10 μm slides. To
stain for SMA, slides were blocked in horse serum and then in-
cubated with cy3 labeled SMA antibody (red). The control eye
that was ablated and treated with vehicle control shows SMA
staining in the basal epithelium and stroma while the triple
siRNA treated right eye show reduction in SMA staining.

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 52
right eye and hence had to be excluded. The right eye
was treated with the effective triple combination (T1R2C1)
complexed with nanoparticles for the first three days after
ablation while the left eye was left untreated. After 14
days, the corneas of both rabbits were imaged using the
macrophotography technique described in the methods
section. The haze grading scores of both the rabbits were
similar with Rabbit b showing a slight reduction in scar-
ring after treatment (Figure 4).
3.5. Quantification of Scar Reduction by the
siRNA Treatment
The digital images from the above experiment were sub-
jected to anti-red gray scale conversion by only using the
data in the blue channel. The contrast was increased by
automatic brightness correction in Image J. The wound-
ing region was split into two regions-Top (Region-I) and
Bottom (Region-II) (Figures 5(a) and (b)). The pixel in-
tensities of the regions of interest were normalized to that
of the transparent unwounded regions of the correspond-
ing corneas. The percentage of haze reduction was calcu-
lated by the reduction in pixel intensity of the scarring
region in the treated eye with respect to that of the un-
treated eye. The region-I of both the rabbits show an av-
erage of ~22% reduction in scarring due to the siRNA
treatment (Figure 5(c)). However there was no visible re-
duction in scar formation in region II of both the rabbits.
After imaging, the wounding region in the corneal tis-
sues was collected with an 8-mm biopsy punch for RNA
analysis. In the corneas treated with the triple siRNA com-
bination, both the rabbits show a corresponding reduc-
tion of 60% and 40% in the RNA level expression of SMA
Figure 4. Reduction in scarring by repeated siRNA dosing. The
corneas of 3 rabbits were evenly ablated to 155 microns using
an excimer laser. The right eye was treated with the effective
triple combination (T1R2C1) complexed with nanoparticles for
the first three days after ablation while the left eye was left un-
treated. Both eyes were imaged using a digital camera after 14
days.
Figure 5. Quantification of scar reduction. The digital images
from the above experiment were subjected to anti-red grayscale
conversion by using the data in the blue channel. The contrast
was increased by automatic brightness correction in ImageJ.
The wounding region was split into two regions-Top (Region-I)
and Bottom (Region-II). The pixel intensities of the regions of
interest were normalized to that of the transparent unwounded
regions of the corresponding corneas. The percentage of haze
reduction was calculated by the reduction in pixel intensity of
the scarring region in the treated eye with respect to that of the
untreated eye (Figure (c)). After imaging, the wounding region
in the corneal tissues was collected with a 6-mm biopsy punch
for RNA analysis. Figure (d) gives the RNA level knockdown
percentage of SMA in the corneas calculated with respect to the
untreated eye. All expressions were normalized to 18S rRNA.
(Figure 5(d)). The knockdown percentages of SMA in the
corneas were calculated with respect to the untreated eye.
All expressions were normalized to 18S rRNA.
4. DISCUSSION
Several papers discuss the importance of the TGFB1 path-
way acting through CTGF to activate quiescent corneal
keratocytes and transform them into fibroblasts that syn-
thesize collagen and further transform into myofibroblasts
[11-13]. However, there has been very little research on
which cell layers of the cornea synthesize high levels of
TGFB1 and CTGF. Our observation that both TGFB1
and CTGF mRNA levels are the highest in the corneal en-
dothelium is novel, and it has reshaped the thinking be-
hind the primary target cells and future delivery methods
for corneal anti-fibrotic therapies. The immediate increase
in the expressions of TGFB1 and CTGF as early as 30
minutes post ablation highlights also emphasized the im-
portance of an immediate treatment. The nanoparticle de-
livery system used in this study was able to successfully

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 53
deliver siRNA to all layers of the cornea including the en-
dothelium in an ex vivo organ culture model [8].
The results of the initial therapeutic experiment as-
sessing the efficacy of a single application of the triple
siRNA combination demonstrated the “proof of princi-
ple” that the siRNAs could be effectively delivered into
the target corneal cells. However, there was variability in
the level of knockdown among the three rabbits, suggest-
ing that the dosing was not optimized for in vivo experi-
ments. Factors that could contribute to the variability in-
clude the presence of a nictitating membrane in addition
to the eyelids and tears that tend to clear the surface of the
cornea and reduce drug exposure time. It is also possible
that the volume of the siRNA formulation used for the sin-
gle dose (150 μL) in this experiment may also have been
too high, which would have reduced the amount of the
siRNAs that penetrated the cornea and were instead wash-
ed away once the trephine was removed.
The reduction in scar formation, 14 days after a single
dose of triple siRNA treatment, also showed a positive
trend. Three out of six rabbits had a reduction in scar for-
mation as observed in the haze grading scores of a mask-
ed ophthalmologist. Both the digital image and the im-
munohistostaining staining for SMA in Figure 3 showed
a reduction in scarring in the treated corneas when com-
pared to the control. Both of these measurements were,
however, qualitative, and we were unable to accurately
quantify the exact reduction in scarring due to the lack of
a standard imaging technology in the field of corneal scar-
ring for those rabbits.
There was an interesting trend associated with the haze
grading scores and the RNA knockdown percentage of
SMA. We observed a RNA knockdown of 40% and 50%
in the two rabbits that had a haze grading score of 2 and
3 respectively. However, no SMA knockdown was ob-
served in the rabbit with a haze grading score of 1. This
suggests that the triple siRNA combination is effective in
knocking down SMA expressions in intense scars and is
less effective in case of mild scarring.
A key paper in the literature reports that deeper abla-
tions during PRK lead to more intense scarring in ani-
mals [9]. In the second therapeutic experiment, we crea-
teda deep corneal ablation of 155 microns and also re-
peated the dosing of the triple siRNA combination for the
first 3 days after ablation. Neither of the two rabbits in this
experiment, however, developed an intense scar in the
untreated control corneas. The haze grading scores of
both the rabbits were similar with the one showing a slight
reduction in scarring after siRNA treatment. However,
the image analysis on the digital images revealed an av-
erage of ~22% reduction in scarring in region 1 of both
the corneas treated with the siRNA combination (Figure
5). It is interesting to note that the siRNA treatment had
an effect in reducing scarring in Region 1, but not in Re-
gion 2. This could be due to the uneven transfection of
the siRNA-nanoparticle complex into the different cor-
neal layers. The RNA level expression of SMA was also re-
duced in both the rabbits by 60% and 40%, respectively.
In this study, we performed a series of pilot experi-
ments to better understand the dynamics involved in the
in vivo translation of an effective in vitro anti-fibrotic
drug. The most immediate problem associated with the
testing of an anti-fibrotic drug in the cornea of rabbits is
the generation of a consistent and intense scar. We tried
two different models of scarring in this study and al-
though the corneas develop mild to average hazing, there
is no consistent and intense scarring among animals. Since
the variation in intensity of scarring among animals is na-
tural and cannot be controlled, treating the corneas with
TGFB1 post ablation for the first two days may help in
the intensification of scar formation [14].
Another major hurdle in evaluating the efficacy of an-
ti-fibrotic drugs in the cornea is the lack of a standard
imaging technology to quantify the reduction in scarring.
The current clinical method of haze grading by a masked
ophthalmologist is qualitative and very subjective. Al-
though the macrophotography technique described in this
study is a lot more objective than haze grading, it also
has its disadvantages. The dual flash heads create an un-
wanted reflection on the cornea due to which we are un-
able to measure the entire wounding region. A potential
solution to this problem could be to image and analyze
the corneas after excision. This would help in uniform il-
lumination of the cornea without flash bias and the scar
can also be measured as a function of the transparency.
Finally, the dosing regimen needs to be optimized so
that maximum amount of inhibitory RNA can be deliv-
ered to the cornea. Lowering the volume and administer-
ing the siRNA cocktail for 2 - 3 days after the surgery
along with an agent to increase the viscosity of the drug,
so that it sticks to the surface of the cornea for a longer
duration, might help increase the drug exposure time [15].
Other options could also include iontophoretically driv-
ing the drug to different layers of the cornea [16,17].
Although, the triple siRNA treatment in this study did
not completely eliminate scarring, it did generate a strong
positive trend in the reduction of scarring. It is important
to understand and resolve the problems associated with
the generation of intense scarring, drug delivery optimi-
zation and scar imaging before investing in a major in
vivo experiment with many animals. Once these parame-
ters are optimized, the triple siRNA combination could
lead to a significant reduction of scarring in the cornea
and perhaps also in other tissues.
5. ACKNOWLEDGEMENTS
Supported by Grants from the US Army Medical Research Acquisition

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 54
[9] Netto, M.V., Mohan, R.R., Sinha, S., Sharma, A., Dupps,
W. and Wilson, S.E. (2006) Stromal haze, myofibroblasts,
and surface irregularity after PRK. Experimental Eye Re-
search, 82, 788-797.
http://dx.doi.org/10.1016/j.exer.2005.09.021
Activity W81XWH-10-2-0917, National Eye Institute grant EY000587,
National Eye Institute T32-EY07132 training grant, and the National
Eye Institute P30-EY021721 Vision Core Grant.
REFERENCES [10] Rodriguez-Lanetty, M. (2007) Trizol/RNeasy hybrid RNA
extraction protocol.
[1] Blalock, T.D., Duncan, M.R., Varela, J.C., Goldstein, M.H.,
Tuli, S.S., Grotendorst, G.R. and Schultz, G.S. (2003) Con-
nective tissue growth factor expression and action in hu-
man corneal fibroblast cultures and rat corneas after pho-
torefractive keratectomy. Investigative Ophthalmology &
Visual Science, 44, 1879-1887.
http://dx.doi.org/10.1167/iovs.02-0860
[11] Massague, J. (1990) The transforming growth factor-beta
family. Annual Review of Cell Biology, 6, 597-641.
http://dx.doi.org/10.1146/annurev.cb.06.110190.003121
[12] Grotendorst, G.R. (1997) Connective tissue growth factor:
A mediator of TGF-beta action on fibroblasts. Cytokine
Growth Factor Reviews, 8, 171-179.
http://dx.doi.org/10.1016/S1359-6101(97)00010-5
[2] Azar, D.T. and Jain, S. (2001) Topical MMC for subepi-
thelial fibrosis after refractive corneal surgery. Ophthal-
mology, 108, 239-240.
http://dx.doi.org/10.1016/S0161-6420(00)00393-6
[13] Jester, J.V., Petroll, W.M. and Cavanagh, H.D. (1999) Cor-
neal stromal wound healing in refractive surgery: The role
of myofibroblasts. Progress in Retinal and Eye Research,
18, 311-356.
http://dx.doi.org/10.1016/S1350-9462(98)00021-4
[3] Chou, T.-C. (2006) Theoretical basis, experimental de-
sign, and computerized simulation of synergism and an-
tagonism in drug combination studies. Pharmacological
Reviews, 58, 621-681.
http://dx.doi.org/10.1124/pr.58.3.10
[14] Desmoulière, A., Geinoz, A., Gabbiani, F. and Gabbiani, G.
(1993) Transforming growth factor-beta 1 induces alpha-
smooth muscle actin expression in granulation tissue myo-
fibroblasts and in quiescent and growing cultured fibro-
blasts. The Journal of Cell Biology, 122, 103-111.
http://dx.doi.org/10.1083/jcb.122.1.103
[4] Sriram, S., Schultz, G.S., Lewin, A., Robinson, P.M., Gib-
son, D.J. and Tuli, S.S. (2012) Combination siRNA Treat-
ment to Knockdown CTGF, TGF-ss1, TGF-ssR2 and Col-
lagen in Rabbit Corneal Fibroblasts. ARVO Meeting Ab-
stracts, 53, 2209.
[15] Tani, E., Katakami, C. and Negi, A. (2002) Effects of va-
rious eye drops on corneal wound healing after superfi-
cial keratectomy in rabbits. Nippon Ganka Gakkai Zasshi,
106, 135-142.
[5] Hao, J., Li, S.K., Kao, W.W.Y. and Liu, C.-Y. (2010) Gene
delivery to cornea. Brain Research Bulletin, 81, 256-261.
http://dx.doi.org/10.1016/j.brainresbull.2009.06.011 [16] Mohan, R.R., Tovey, J.C.K., Sharma, A. and Tandon, A.
(2012) Gene therapy in the cornea: 2005-present. Pro-
gress in Retinal and Eye Research, 31, 43-64.
http://dx.doi.org/10.1016/j.preteyeres.2011.09.001
[6] Lee, V.H. and Robinson, J.R. (1986) Topical ocular drug
delivery: Recent developments and future challenges. Jour-
nal of Ocular Pharmacology, 2, 67-108.
http://dx.doi.org/10.1089/jop.1986.2.67 [17] Hao, J., Li, S.K., Liu, C.-Y. and Kao, W.W.Y. (2009) Elec-
trically assisted delivery of macromolecules into the cor-
neal epithelium. Experimental Eye Research, 89, 934-941.
http://dx.doi.org/10.1016/j.exer.2009.08.001
[7] La Fuente de, M., Seijo, B. and Alonso, M.J. (2008) “Bi-
oadhesive hyaluronan-chitosan nanoparticles can trans-
port genes across the ocular mucosa and transfect ocular
tissue. Gene Therapy, 15, 668-676.
http://dx.doi.org/10.1038/gt.2008.16
[8] Sriram, S., Robinson, P., Lewin, A. and Schultz, G. (2013)
Nanoparticle vectored siRNAs reduce profibrotic gene
expression in wounded rabbit corneas. ARVO Meeting
Abstracts, 54, 3878.

S. Sriram et al. / Advances in Bioscience and Biotechnology 4 (2013) 47-55 55
APPENDIX
Table A1. TAQMAN™ RT PCR primers and probe sequences.
Growth Factor Species Accession Number
CTGF
Forward
Reverse
Probe
AGGAGTGGGTGTGTGATGAG
CCAAATGTGTCTTCCAGTCG
ACCACACCGTGGTTGGCCCT
TGFB1
Forward
Reverse
Probe
CCTGTACAACCAGCACAACC
CGTAGTACACGATGGGCAGT
CTCCAGCGCCTGTGGCACAC
GAPDH
Forward
Reverse
Probe
GAGACACGATGGTGAAGGTC
ACAACATCCACTTTGCCAGA
CCAATGCGGCCAAATCCGTT
SMA Forward AGAGCGCAAATACTCCGTCT
Reverse CCTGTTTGCTGATCCACATC
Probe CGGCTCCATCCTGGCCTCTC
18S rRNA Forward GCCGCTAGAGGTGAAATTCTTG
Reverse CATTCTTGGCAAATGCTTTCG
Probe ACCGGCGCAAGACGGACCAG