Journal Pre-proof

Phenylephrine Induces Necroptosis and Apoptosis in Corneal Epithelial

Cells Dose- and Time-Dependently

Xin-Guo You, Ting-Jun Fan, Guo-Jian Jiang

PII: S0300-483X(19)30262-8
DOI: https://doi.org/10.1016/j.tox.2019.152305
Reference: TOX 152305

To appear in: Toxicology

Received Date: 20 July 2019

Revised Date: 13 September 2019
Accepted Date: 2 October 2019

Please cite this article as: You X-Guo, Fan T-Jun, Jiang G-Jian, Phenylephrine Induces
Necroptosis and Apoptosis in Corneal Epithelial Cells Dose- and Time-Dependently,
Toxicology (2019), doi: https://doi.org/10.1016/j.tox.2019.152305

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.

© 2019 Published by Elsevier.

Phenylephrine Induces Necroptosis and Apoptosis in Corneal
Epithelial Cells Dose- and Time-Dependently

Xin-Guo You, Ting-Jun Fan, Guo-Jian Jiang *

Laboratory for Corneal Tissue Engineering, College of Marine Life Sciences, Ocean University

of China, Qingdao, Shandong Province 266003, P.R. China

of
ro
* Correspondence to: Guo-Jian Jiang, PhD
-p
Address: Laboratory for Corneal Tissue Engineering, College of Marine Life Sciences,
re
Ocean University of China, Yushan Road No.5 Qingdao, Shandong Province

266003, P.R. China

lP

E-mail: gjjiang@ouc.edu.cn
na

Tel: +8613687641199

Fax: +8653282031793
ur

Highlight
Jo

 10% (clinical dosage) to 1.25% phenylephrine induce necroptosis in in vitro cultured

human corneal epithelial cells through via RIPK1, RIPK3 and MLKL pathway.

 0.625% phenylephrine induce apoptosis in in vitro cultured human corneal epithelial

cells through activation of caspase-2, -8, -9 and -3 as well as down-regulation of Bcl-

1
 2, up-regulation of Bad, ΔΨm disruption and release of cytochrome c and AIF into

cytosol.

 10% phenylephrine induced destruction of the corneal epithelia and apoptosis of

corneal epithelial cells in rabbit corneas.

Abstract

In the present study, the toxicity of phenylephrine, a selective α1-adrenergic receptor agonist,

of
in corneal epithelial cells and its underlying mechanisms were investigated using an in vitro

ro
model of human corneal epithelial cells (HCEPCs) and an in vivo model of New Zealand white

rabbit corneas. The HCEPCs treated with phenylephrine at concentrations from 10% to
-p
0.078125% displayed abnormal morphology, decline of cell viability and elevation of plasma
re
membrane permeability time- and dose-dependently. Moreover, 10% to 1.25% phenylephrine

induce necrosis characteristics of marginalization and uneven distribution of chromatin through

lP

up-regulation of RIPK1, RIPK3 and MLKL along with inactivation of caspase-8 and caspase-
na

2, whereas 0.625% phenylephrine induced condensed chromatin, S phase arrest,

phosphatidylserine externalization, DNA fragmentation and apoptotic body formation in the

ur

HCECs through activation of caspase-2, -8, -9 and -3 as well as down-regulation of Bcl-2, up-

regulation of Bad, ΔΨm disruption and release of cytochrome c and AIF into cytosol. At last,
Jo

10% phenylephrine induced destruction of the corneal epithelia and apoptosis of corneal

epithelial cells in rabbit corneas. In conclusion, 10% to 1.25% phenylephrine cause necroptosis

via RIPK1-RIPK3-MLKL axis and 0.625% phenylephrine induce apoptosis via a

mitochondrion-dependent and death receptor-mediated signal pathway in HCEPCs.

2
Keywords: Phenylephrine; Toxicity; Human corneal epithelial cells; New Zealand white rabbit;

Necroptosis; Apoptosis.

1. Introduction

The human corneal epithelium (HCEP) constructed with multi-layers of superficial flat

squamous cells and suprabasal cuboid wing cells, and a monolayer of basal column-shaped

cells located at the anterior surface of the cornea. Functionally, the HCEP is the first defense

of
barrier of the cornea against pathogens and toxic substance, which plays a key role in

maintaining corneal transparency and absorption of nutrients and oxygen (Kinoshita et al.,

ro
2001). Thus, the HCEP is vulnerable to the damage of drug of eye drop, trauma and infection

-p
of bacteria or viruses due to its outermost location of the ocular surface. Among them the toxic

effects of the ophthalmic solution on corneal epithelium is giving rise to the attention of clinical
re
administration, because documented reports indicated that routine clinical administration of eye
lP

drop caused the corneal epithelial abnormality (Raizman et al., 2017).

Phenylephrine (PE), a selective α1-adrenergic receptor agonist belonging to

na

phenethylamine class, is commonly used decongestant for mydriasis prior to cataract surgery

and applied in the treatment of open angle glaucoma (Lay Suan et al., 2017; Guthrie et al.,
ur

2018). However, cumulative evidence demonstrated that PE is toxic to corneal epithelium,

Jo

whereas its underlying mechanism is still elusive, owing to lack of suitable research model

(Edelhauser et al., 1979; Lapalus et al., 1990; Pless & Friberg, 2003). Up to now, cell cultures

have become gold standard for preclinical studies to explore important findings in drug toxicity

studies (Jaroch et al., 2018; Vidmar et al., 2017). A non-transfected human corneal epithelial

cell (HCEPC) line provides us an ideal in vitro cell model to study the cytotoxicity of PE and

3
the possible cellular and molecular mechanisms, owing to its normal genotype and inherent

properties (Fan et al., 2011). Therefore, the present study was designed to evaluate the

cytotoxicity of PE in HCEPCs and investigate its underlying cytotoxic mechanisms with the in

vitro model of non-transfected HCEPCs and an in vivo model of rabbit corneas to offer

theoretical basis on developing a reasonable strategy for ocular topical administration of PE.

2. Materials and Methods

of
2.1. Test solution and eye-drop

The PE stock solution was prepared at the concentration of 20% (m/v%) by dissolving PE

ro
(Bailingwei Technology, Beijing, China) into serum-free Dulbecco’s modified Eagle medium:

-p
Ham’s nutrient mixture F-12 medium (1: 1) (DMEM/F12; Invitrogen). A serial test solutions

ranging from 10% to 0.0390625% were prepared by double diluting the stock solution with
re
DMEM/F12 medium containing 20% (v/v) fetal bovine serum (FBS; Gibco, NY, USA) for in
lP

vitro experiment. PE eye-drop for in vivo experiment was prepared at a concentration of 10%

(clinic concentration) by dissolving PE into physiological saline (0.9% NaCl).

na

2.2. Experimental animals

ur

Twelve healthy male New Zealand white rabbits, with a body weight ranging from 2.0 to

2.5 kg were obtained from the Qingdao Kangda Bio-Tech Co., Ltd (Qingdao, China), with a
Jo

certificate number of SYXK-(SD)-2016-0002. They were acclimated for 1 week to the

conditions of our lab before experiments. The animal experiment procedures in the present

study were conducted in accordance with the National Institutes of Health guide for the care

and use of Laboratory animals (NIH Publications No. 8023, revised 1978) and approved by the

4
institutional Ethics Committee of Animal Care and Experimentation (approval no. SD-SYKY-

2014-021). All of animal protocols were adhering to the guidelines in the Association for

Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in

Ophthalmic and Vision Research.

2.3. HCEPC culture and PE treatment

The non-transfected HCEPCs at passage121 with similar genotype and function to in vivo

of
HCEPCs and without tumorigenicity were cultured in DMEM/F12 medium containing 10%

FBS as described previously (Fan et al., 2011). As the cells grew into the logarithmic phase, the

ro
culture medium was replaced completely with fresh medium containing PE at concentrations

-p
from 10% to 0.0390625%. The HCEPCs cultured in the same medium without supplementing

PE were used as controls.

re
2.4. Microscopic observation of the morphology of HCEPCs
lP

HCEPC morphology and growth status were monitored with an Eclipse TS100 inverted

microscope (Nikon, Tokyo, Japan) every 2–4 h after exposure to PE at a serial concentrations
na

from 10% to 0.0390625%.

ur

2.5. MTT assay

The HCEPC viability was detected using MTT assay as described previously (Carmichael
Jo

et al., 1987). Briefly, HCEPCs were inoculated into 96- well culture plates at a density of 1×104

cells per well. After cultured in DMEM/F12 with 10% FBS for 12 h, the cells were treated with

PE at the concentrations from 10% to 0.0390625% and the cells without treating with PE at the

same time were used as controls. At a 2–4 h interval for 28h, the culture medium was discarded

5
and the cells were incubated with 20 μl of 5 mg/ml MTT (Sigma-Aldrich) at 37 oC in dark for

4 h. After adding 150 μl DMSO (Sigma- Aldrich), the absorbance at 490 nm was measured

with a Multiskan GO microplate reader (Thermo Scientific) and the cell viability was calculated

A (490) 𝑜𝑓 𝑃𝐸 𝑡𝑟𝑒𝑎𝑡𝑒𝑑 𝑐𝑒𝑙𝑙𝑠

as: 𝐶𝑒𝑙𝑙 𝑣𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦 (%) = A(490) 𝑜𝑓 𝑐𝑜𝑛𝑡𝑟𝑜𝑙
× 100%.

2.6. AO/EB double staining

Plasma membrane (PM) permeability was assayed by AO/EB double staining as described

of
previously (Li et al., 2015). Briefly, HCEPCs at the logarithmic phase were treated with PE as

described above. At a 2–4 h interval for 28h, the cells were collected, suspended in 0.1 ml of

ro
serum-free DMEM/F12 medium, and subsequently stained at 25oC by adding 4 μl of 100 μg/ml

-p
AO/EB solution (1:1; Sigma-Aldrich) for 1 min. The stained cells were observed with a Ti-S

fluorescent microscope (Nikon), the dead cells with red or orange nuclei were designated as
re
their PM permeated, while the living cells with green nuclei. At least 400 cells of one sample
lP

were counted randomly and the plasma membrane permeability was calculated

𝑐𝑒𝑙𝑙𝑠 𝑤𝑖𝑡ℎ 𝑟𝑒𝑑 𝑜𝑟 𝑜𝑟𝑎𝑛𝑔𝑒 𝑛𝑢𝑐𝑙𝑒𝑖

as: PM permeability (%) = total cells
× 100%.
na

2.7. DNA electrophoresis

ur

DNA damage of HCEPCs was determined with agarose gel electrophoresis as described

previously (Fan and Fan, 2017). HCEPCs cultured in 25 cm2 culture flasks were treated with
Jo

PE and collected as described above. The genomic DNA was isolated with a Quick

Tissue/Culture Cells Genomic DNA Extraction Kit (CWbiotech, Beijing, China) according to

the instruction of the manufacturer. The DNA extract was separated in a 1% agarose gel (200

mA, 270 min), the electrophoresed gel was stained in 0.5 μg/ml EB solution for 10 min and

6
imaged with an EC3 Imaging System (UVP, Upland, CA, USA).

2.8. Transmission electron microscopy (TEM) observation

The ultrastructure of HCEPCs post-exposure to 0.625% PE was observed with TEM.

HCEPCs were incubated in 25 cm2 culture flasks, treated with 0.625% PE and harvested at 4,

8, and 12 h. The cells were fixated with 4% glutaraldehyde at 4oC overnight and 1% osmium

tetroxide for 1.5 h at 25oC successively. Following routine dehydration, the samples were

of
embedded in epoxy resin. Then the ultrathin sections were stained with 2% uranyl acetate-lead

citrate and observed with a H700 TEM (Hitachi, Tokyo, Japan).

ro
2.9. Flow cytometry assay
-p
The cell cycle progression, PS externalization and mitochondrial transmembrane potential
re
(ΔΨm) were determined by Flow cytometry (FCM). Firstly, HCEPCs were incubated in 6-well

culture plates, treated with 0.625% PE and harvested at 4, 8 and 12 h. As for cell cycle assay,
lP

the cells were fixated with 70% alcohol at 4oC overnight. Then the fixated cells were stained
na

with 500 μl PI and RNase solution (BD Biosciences, San Jose, CA, USA) for 30 min in dark.

The samples for PS externalization assay were prepared to stain the cells in 500μl suspension
ur

with FITC-labeled Annexin-V and PI using FITC-Annexin V Apoptosis Detection Kit I (BD

Biosciences) according to the manufacture’s instruction. For ΔΨm assay, the cells in 500 µl
Jo

suspension were stained with 5 µl of 10 µg/ml 5, 5’, 6, 6’-tetrachloro-1, 1’, 3, 3’-

tetraethylbenzimidazolocarbocyanine iodide (JC-1, Sigma-Aldrich) in dark for 15 min. The

stained cells were assayed with a FACScan FCM and analyzed with CXP analysis software (BD

Biosciences) (Shan and Fan, 2016).

7
2.10. ELISA assay

Caspase activation in HCEPCs was assayed by enzyme-linked immunosorbent assay

(ELISA) as described previously (Tian et al., 2015). The HCEPCs were inoculated into 24-well

culture plates, treated with 10%, 5%, 2.5%, 1.25% PE and harvested at 1h, 2h, 3h, 4h or treated

with 0.625% PE and harvested at 2h, 4h, 6h, 8h, 10h, 12h, 14h, 16h. Whole-cell protein extracts

were prepared with radio-immunoprecipitation assay (RIPA) lysis buffer RIPA lysis buffer

supplemented with phenylmethylsulfonyl fluoride (PMSF) (Biotime, Beijing, China) according

of
to the manufacturer’s instructions. Subsequently 100 μl protein extracts were and coated into a

ro
high binding 96-well microtiter plate overnight at 4oC and washed 3 times with PBS containing

0.05% Tween-20 (PBST). Then each well was blocked with 5% non-fat milk (BD Bioscience,
-p
New Jersy) and incubated with 100 µl rabbit anti-human active caspase-2, -3, -8, and -9
re
monoclonal antibodies (1: 500; Biosynthesis biotechnology, Beijing, China) at 37 oC for 90 min

and 100 μl goat anti-rabbit secondary antibody conjugated with horseradish peroxidase (HRP)
lP

(1: 3000; ComWin Biotechnology) at 37 oC for 2h. After 3 washes, 100 μl chromogenic
na

substrate (1% tetramethylbenzidine) were added into each well and incubated for 25 min at

25oC in dark to induce colorimetric reaction. The color development was stopped by adding 50
ur

µl of 0.5 M H2SO4 solution. The absorbance at 490 nm of each well was assayed with a

Multiskan GO microplate reader (Thermo Scientific).

Jo

2.11. Western blotting

Expression of necroptosis and apoptosis related proteins of RIPK1, RIPK3, MLKL and

Bcl-2 family proteins, and the amount of cytoplasmic mitochondrion-released pro-apoptotic

proteins were quantified by Western blotting as described previously (Mulay et al., 2016; Fan

8
et al., 2017). HCEPCs were cultured in 25 cm2 culture flasks, treated with 0.625% PE and

harvested at 4, 8 and 12h or treated with 10%, 5%, 2.5% and 1.25% PE and harvested at 4h.

Whole-cell protein extracts and cytoplasmic protein extracts were prepared as above in the

section “ELISA assay” and cytoplasmic extracts from the same number of cells were prepared

with mitochondrial and cytoplasmic protein extraction kit according to the manufacture’s

instruction (Sangon biological engineering, Shanghai, China) for quantitative assay of

cytoplasmic pro-apoptotic proteins released from mitochondria such as Cyt.c and AIF. Then

of
whole-cell and cytoplasmic protein extracts were electrophoresed in 10% SDS-PAGE,

ro
transferred to PVDF membranes, blocked by 5% nonfat milk for 1 h and washed 3 times. Then,

the membranes were incubated with primary antibody of rabbit anti-human monoclonal
-p
antibody (all of the primary antibodies purchased from Abcam, Cambridge, UK) against β-actin
re
(1:5000), RIPK1 (1:1000), RIPK3 (1:1000), MLKL(1:1000), Bcl-xL (1:1000), Bcl-2 (1:1000),

BAD (1:1000), Bax (1:2000), Cyt. c (1:5000), AIF (1:1000) at 4 oC overnight. After the
lP

membranes were incubated with 100 μl HRP- conjugated goat anti-rabbit IgG monoclonal
na

antibody (1:5000; Proteintech) for 1.5 h at 25oC and immersed in chemiluminescence reagents

(Pierce, Rorkford, IL, USA), the protein bands were detected using a cECL Western Blot Kit
ur

(CW biotech) according to the manufacturer’s instructions and were observed under an EC3

Imaging System (UVP). The optical density of each band was quantified with image J analysis
Jo

software (NIH, NY, USA) using β-actin as an internal control.

2.12. Toxic effects of PE on in vivo corneal epithelium

The toxic effect of PE on corneal epithelium was evaluated using New Zealand rabbit

corneas as an in vivo model. The corneas of the right eyes of four rabbits were instilled with 2

9
drops of 10% PE eye drops 5 times a day for 7 days in total, and all corneas of the left eyes

were instilled with 2 drops of physiological saline in parallel as controls. 7 days later, the rabbits

were euthanized with ether and their corneas were dissected for ex vivo examination. Samples

of 7 μm-thick paraffin sections of the corneas were made for routine hematoxylin and eosin

(H&E) staining to detect the impact of PE on the tissue structure of the rabbit cornea with a

E200 light microscope (Nikon). The 7 μm-thick paraffin sections were also used for terminal

deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL) staining with

of
fluorescein isothiocyanate (FITC), using a one-step TUNEL in situ cell apoptosis detection kit

ro
(Kegen Biotech, Nanjing, China) to detect the apoptotic effect of PE on the rabbit corneal

epithelial cells in vivo under an E80i fluorescence microscope (Nikon)(Fan and Fan, 2017).
-p
Furthermore, the cornea epithelia from the rabbits were examined by scanning electron
re
microscopy (SEM; JSM-840; JEOL, Tokyo, Japan) and 60 nm-thick ultrathin sections were

prepared for TEM observation under an a H700 transmission electron microscope (Hitachi,
lP

Tokyo, Japan) as described previously (Wen et al., 2015).

na

2.13. Statistical analysis

All data is shown as mean ± standard deviation (SD) from three independent experiments
ur

and analyzed for statistical significance by one way analysis of variance (ANOVA) with SPSS
Jo

17.0 software (SPSS Inc., Chicago, IL, USA). Differences to controls at P<0.05 were

considered statistically significant.

3. Results

3.1. The morphology, growth status and viability of HCEPCs

10
 The growth status and morphological characteristics of PE-exposed HCEPCs were

monitored under a light microscopy. The result exhibited that the HCEP treated with PE in a

concentration from 10% to 0.039125% showed growth retardation, shrinkage, vacuolation and

detachment from culture matrix in a dose-and time-dependent fashion (Fig. 1A). Especially,

HCEPCs exposed to PE at concentrations of more than 0.625% were ablated in a large area

within a short time. Alternatively, the results of MTT indicated that the viabilities of HCEPC

sharply declined to 0 post-exposure to 10%, 5%, 2.5% and 1.25% of PE at 2h, 4h and 8h,

of
respectively (Fig. 1B). Although there was no significant difference between the HCEPCs

ro
treated with PE at concentrations below 0.039125% and control (p>0.05), the PE at

concentrations from 0.625% to 0.078125% caused significant decrease in the HCEPC

-p
viabilities dose- and time-dependently (p<0.05) (Fig. 1B). Thus, the PE at concentrations from
re
10% to 0.078125% has toxic effects on HCEPCs in a dose-and time-dependent fashion.
lP

3.2. PM permeability and DNA damage of PE-treated HCEPCs

The results of AO/EB double staining demonstrated that the PE at concentrations equal to
na

or more than 0.078125% could induce a dose- and time-dependent elevation of PM

permeability (P <0.01 or 0.05) compared with controls (Fig. 2B). When the HCEPCs exposed
ur

to more than or equal to 1.25% PE, their PM permeability increased dramatically with necrosis
Jo

features of marginalization and uneven distribution of chromatin (Jabłońska-Trypuć et al.,

2019). While at concentrations from 0.625% to 0.078125%PE induced increase of PM

permeability of HCEPCs in a dose- and time-dependent manner with apoptosis features of

visible chromatin condensation and apoptotic body (p<0.01; Fig.2A and 2B). Moreover, the

DNA eletroporesis demonstrated that 0.625% to 0.078125% PE caused genomic DNA of

11
HCEPCs highly fragmented with obvious DNA ladders which is the typical feature of apoptosis,

while at concentrations more than or equal to 1.25%, PE caused genomic DNA of HCEPCs

irregularly fragmented (Fig. 2C). Therefore, 0.625% to 0.078125% PE caused apoptosis of

HCEPCs and that more than or equal to 1.25% caused necrosis of HCEPCs.

3.3. Investigation of necroptosis-associated molecules in PE treated HCEPCs

The necrosis mechanism in PE-treated HCEPCs was explored by analysis of caspase-2

of
and -8 activation with ELISA and quantitative alterations of RIPK1, RIPK3 and MLKL with

Western blotting. As shown in Fig. 3, the activation of caspase-2 and -8 in HCEPCs decreased

ro
significantly after exposure to 10%, 5%, 2.5% and 1.25% PE from 1 to 4h in comparison with

-p
controls (Fig. 3C and 3D). Accordingly, the expression level of RIPK1, RIPK3 and MLKL

increased significantly after exposure to 10%, 5%, 2.5% and 1.25% PE for 4h in comparison
re
with controls (Fig. 3A and 3B, P <0.01). Therefore, PE at concentrations from 10% to 1.25%
lP

induces necroptosis of HCEPCs.

na

3.4. Evaluation of HCEPC cycle progression

According to the results of above, PE at concentrations from 0.625% to 0.078125%

ur

induces apoptosis of HCEPCs and 0.625% is the maxium concentration of PE to induce

apoptosis of HCEPCs in the present study. Thus, the concentration of 0.625% is appropriate to
Jo

study the PE induced apoptosis mechanism of HCEPCs.

The cell cycle progression of HCEPCs exposed to PE was detected by FCM using PI

staining to examine the effects of PE on growth and proliferation of HCEPCs. After exposure

to 0.625% PE for 4, 8, and 12 h, the percentage of HCEPCs at S phase increased and decreased

12
at the G2 phase significantly compared to controls (P<0.01 or 0.05; Table 1 and supplementary

material 2). While at same sampling time, the percentage of HCEPCs is more at S phase and

less at G2/M phase than control significantly (P<0.01), which implicated PE-induced HCEPC

S phase arrest.

3.5. Analysis of PS externalization and ultrastructure of PE treated HCEPCs

The results of FCM with Annexin V/PI staining indicated that the number of PS-

of
externalized cells (positive for Annexin V) increased with time after exposure to 0.625% PE

for 4, 8, and 12h (P<0.01) (Fig. 4B and supplementary materials 3). Furthermore, results of

ro
TEM observation of HCEPCs exposed to 0.625% PE for 4, 8 and 12h exhibited typical

-p
ultrastructural characteristics of apoptosis such as cytoplasmic vacuolation, structural

disorganization, chromatin condensation, swollen mitochondria with disrupted cristea and

re
apoptotic bodies in correspondence with normal ultrastructure of controls (Fig. 4A). Thereby,
lP

it is suggested that PE at the concentration of 0.625% have an apoptotic effect on HCEPCs in

vitro.
na

3.6. Investigation of PE induced intrinsic apoptosis pathway in HCEPCs

ur

Investigation of the intrinsic apoptosis pathway in 0.625% PE treated HCEPCs by analysis

of ΔΨm disruption with FCM and quantitative alterations of Bcl-2 family proteins and
Jo

mitochondria associated apoptosis regulators with Western blotting. The results of JC-1 staining

FCM showed that the number of ΔΨm disrupted cells (positive for JC-1) increased significantly

from 4h to 12h (Fig. 5E and supplementary materials 4). Accordingly, Western blot results

displayed that the amount of cytoplasmic Cyt. c, AIF, Bad and Bax was up-regulated, while that

13
of Bcl-2 and Bcl-xL was down-regulated in comparison with controls at 4,8 and 12h,

respectively (Fig. 5A -5D). All these results indicated that the PE induces apoptosis of HCEPCs

through a mitochondria-dependent pathway via the disruption of ΔΨm to release Cyt.c and AIF

into cytoplasm along with upregulation of pro-apoptotic proteins such as Bad and Bax as well

as downregulation of anti-apoptotic proteins such as Bcl-2 and Bcl-xL.

3.7. Analysis of PE induced caspase activation in HCEPCs

of
The apoptotic signaling pathway of HCEPCs triggered by PE was investigated through

activation of caspase-2, -3, -8 and -9. The results demonstrated that the values of the activation

ro
of caspase-2 in 0.625% PE-treated HECPCs increased significantly from 0 to 2h with a

-p
maximum value of (208.44±11.13) % of control at 2h (P<0.01). While caspase-8,-3 and-9 in

0.625% PE-treated HECPCs increased significantly from 0 to 6h with their maximum value of
re
(183.04±19.19)%, (170.51±10.97)% and (188.17±8.86)% at 6h (p<0.01) respectively, in
lP

comparison with controls (Fig. 5F). Thereby, PE induced apoptosis of HCEPCs through the

activation of caspase-2, -8, -9, and -3 successively in HCEPCs.

na

3.8. Investigation of cytotoxicity of PE in corneal epithelial cells in vivo

ur

The toxic effect of PE on corneal epithelial cells was investigated using an in vivo model

of rabbit corneas post-exposure to 10% PE for 7 days. The results of H&E staining and
Jo

observation of SEM and TEM displayed the loose structure of the rabbit corneal epithelium

with reduction of corneal epithelial layers and disorganized squamous cells, destruction of cell

junction, reduction and disappearance of microvilli of squamous cells and atrophy chromatin

(Fig. 6A-6C). Moreover, the TUNEL positive cells in the rabbit corneal epithelium means their

14
nuclear DNA was fragmented (Fig. 6D). These results indicated that PE at its clinical

concentration induces apoptosis of rabbit corneal epithelial cells in vivo.

4. Discussion

It has been reported that topical administration of PE eye drop caused abnormal cell

morphology and pathogenic symptom of corneal epithelium, which gave rise to more attention

for clinical safety (Raizman et al., 2017). Therefore, study on the toxicity of PE in corneal

of
epithelium and the underlying mechanism plays a pivotal role in making a safe and effective

strategy for clinical administration of PE eye drop. Herein, we investigated the cytotoxic effects

ro
and its underlying mechanisms of PE on corneal epithelial cells both in vivo and in vitro to offer

references for its safe medication in eye clinic.

-p
As for exploration of the toxic efficacy of PE to HCEPCs and the underlying mechanism,
re
an ideal in vitro cell model is necessary (Jaroch et al., 2018; Vidmar et al., 2017). Thereby we
lP

applied the in vitro cultured HCEPCs from non-infected HCEPC line in the present study. Our

results demonstrated that PE at concentrations from 0.078125% to 10% is toxic to HCEPCs by

na

inducing morphological abnormality, growth retardation, PM permeability increase and

viability decline of HCEPCs in a dose- and time-dependent manner. In addition, the results of
ur

AO/EB and DNA electrophoresis exhibited that PE at concentrations from 10% to 1.25%
Jo

caused necrotic characteristics in HCEPCs such as marginalization and uneven distribution of

chromatin and genomic DNA irregular fragmentation (Jabłońska-Trypuć et al., 2019; Fig. 2C).

While at concentrations from 0.625%to 0.078125%, PE induced apoptotic features of

condensed chromatin, apoptotic body, increased PM permeability and fragmented genomic

DNA with obvious DNA ladders in HCEPCs. It is suggested that the toxic degree in HCEPCs

15
caused by PE is time- and dose-dependent. The PE at concentrations ranging from 0.625% to

0.078125% caused apoptosis and that above 1.25% caused necrosis dose- and time-dependently

(Fig. 2A and 2B, p<0.01).

Necroptosis, one type of the necrosis, is regulated by the activity of receptor-interacting

protein kinase (RIPK) 1 and 3 (Galluzzi and Kroemer, 2008). Once RIPK1 is activated, it binds

to RIPK3 to form the necrosome complex. Then, the necrosome recruits and enhances mixed

lineage kinase domain-like (MLKL) phosphorylation (Rickard et al., 2014). Finally, the

of
activated MLKL oligomerizes, translocates from the cytosol to membranes, binds to the

ro
membrane phospholipids and subsequently leads to membrane disintegration and necrosis

(Mulay et al., 2016). Therefore, RIPK1 and RIPK3 are central regulators for initiating
-p
necroptosis (Hitomi et al., 2008; Zhou and Yuan, 2014;Moriwaki, et al., 2015).
re
Necroptosis and apoptosis can be induced by the upstream elements of same signaling

pathway under different stresses. In TNF-α signaling pathway, TNF-α binds to and activates
lP

TNFR1. The activatedTNFR1 recruits TNF receptor-associated death domain (TRADD),

na

RIPK1 and several ubiquin E3 ligases to form complex I in which the RIPK1 is

polyubiquinated and subsequent deubiquitinated to form complex IIa under mild stress and
ur

complex IIb under severe stress (Wertz et al., 2004; D'Arcy, 2019; Christofferson et al., 2014).

When caspase-8 is inhibited, complex IIb is formed to activate RPK1-RPK3-MLKL pathway

Jo

(Wang et al., 2014). In addition, recent research reported that caspase-2 is also the negative

regulator of necroptosis (Jabłońska-Trypuć et al., 2019). Therefore, our results of the up-

regulation of RPK1, RPK3 and MLKL in combination with inhibition of the activity of

caspase-8 and caspase-2 indicated that PE at concentrations from 10% to 1.25% caused

16
 necroptosis of HCEPCs via RIPK1-RIPK3-MLKL signaling pathway. In complex IIa, initiator

caspase-8 activated. The activated caspase-8 subsequently activates the executioner such as

caspase-3 etc. and result in apoptosis.

In general, there are two apoptotic signaling pathways, namely, the death receptor-

mediated extrinsic pathway and the mitochondrion-mediated intrinsic pathway (Matthews et

al., 2012). The extrinsic pathway activated by the death ligands acting on the death receptors

leads to activations of initiator caspase-2, -8, and/or -10, and the intrinsic pathway is switched

of
through mitochondria such as release of Cyt.c to activate initiator caspase-9 (Fan et al., 2005;

ro
Jin et al., 2005). Subsequently executioner caspases -3, -6, and -7 are activated in both extrinsic

and intrinsic pathway. In the present study, the results of the typical apoptotic features of cell
-p
cycle arrest, PS externalization, chromatin condensation and apoptotic body formation in
re
combination with the successive activation of capspase-2, -8, -9 and -3 indicated that 0.625%

PE induces the apoptosis in the HCEPCs through both a death receptor-mediated and a
lP

mitochondrion dependent pathway.

na

The apoptosis is generally regulated via the concordant cooperation of caspases, pro-

apoptotic Bcl-2 family proteins of Bax and Bad, anti-apoptotic Bcl-2 family proteins of Bcl-2
ur

and Bcl-xL and the mitochondrion-sequestered apoptosis-triggering proteins such as Cyt. c, AIF,

and Smac/Diablo (Fan et al., 2005; Garrido et al., 2006; Cai et al., 1998; Natarajan et al., 2012;
Jo

Narita et al., 1998; Czabotar et al., 2014; Kluck et al., 1997; Rodriguez-Enriquez et al., 2004;

Sedlackova and Korolchuk 2019; Napoletano et al. 2019; Santucci et al. 2019; McArthur and

Kile 2018; Knight et al. 2019). Apoptosis occurs as up-regulation of pro-apoptotic proteins and

down-regulation of anti-apoptotic proteins caused reduction of ΔΨm and subsequent release of

17
Cyt. c, AIF and Smac/Diablo from mitochondria into cytosol. Cyt. c in cytosol combined with

procaspase-9 and apoptotic proteinase activating factor-1 (Apaf1) assemble apoptosome which

is essential for activating caspase-9, while AIF induces caspase-independent peripheral

chromatin condensation and large-scale DNA fragmentation, and Smac/Diablo activates

caspase by inhibiting the inhibitor of apoptosis proteins. In the present study, the results of the

down-regulated anti-apoptotic protein Bcl-2 and up-regulated pro-apoptotic protein Bad in PE

treated HECPCs caused ΔΨm disruption and subsequently released Cyt. c and AIF from

of
mitochondria into cytosol to activate intrinsic apoptosis pathway resulting in chromatin

ro
condensation and DNA fragmentation. Altogether, PE induces apoptosis of HCEPCs is

probably via death receptor-mediated and mitochondrion-dependent signaling pathway.

-p
The toxicity of PE in clinical concentration in rabbit corneal epithelium was also
re
investigated using an in vivo model of New Zealand white rabbit cornea. Our results indicated

that 10% PE caused damage to both corneal epithelial tissue such as loose structure with
lP

reduced corneal epithelial layers, disorganized squamous cells etc. and corneal epithelial cells
na

such as destruction of cell junction, reduction and disappearance of the microvilli on the surface

of squamous cells, condensed chromatin and fragmentation of nuclear DNA (Fig. 7). Thus, PE
ur

at its clinical concentration is toxic to corneal epithelium and induces apoptosis of rabbit corneal

epithelial cells in vivo, which offers basic data for making safe and effective strategy for clinical
Jo

administration of PE. Moreover, the results of the present study indicated that the in vitro cell

model is more sensitive to PE than in vivo animal model, due to the in vitro cell model is

deficient in ability of mimicking the complexities and numerous physiological parameters in

the tissue microenvironment of cornea (Rönkkö et al., 2016). Thus, the data from the in vitro

18
cell model cannot predict the potential risks and infer clinical safety without in vivo experiments,

but animal experiments have been widely criticized for ethical reasons (Rönkkö et al., 2016).

Moreover, there are other drawbacks in animal experiments such as long raising periods, high

breeding cost and individual differences between animals. Therefore, tissue-engineered human

cornea is a promising replacement for human cornea in real-time toxicological and fundamental

biological study (Shamir et al., 2014). Together with the finding of the present study, we will

construct an in vitro 3 dimension human corneal epithelial equivalent with inherent features of

of
native human corneal epithelia for safety evaluation of ophthalmic drugs and histopathological

ro
study of human corneal epithelium in our further research.

In conclusion, PE at concentrations from 10% to 1.25% caused necroptosis of the cells

-p
from HCEPC line by activating RIPK1-RIPK3-MLKL signal pathway and 0.625% to 0.078125%
re
PE induces apoptosis of HCEPCs probably via death receptor-mediated and mitochondrion-

dependent signaling pathway. While, 10% PE caused disorganization of rabbit corneal epithelia
lP

and apoptosis of rabbit corneal epithelial cells in vivo. Therefore, the in vitro cell model is
na

suitable to explore in depth the mechanism of cytotoxicity of drugs based on its stable status

and the in vivo model is suitable to evaluate the clinical safety and efficacy of medicine. These
ur

findings opened a new avenue for exploring the mechanism of acute and chronic cytotoxic

effect of ophthalmic drugs.

Jo

Conflict of interest

The authors declare that they have no conflict of interest.

Acknowledgements

This study was supported by the National Key Research and Development Program of

19
China (2018YFC1106000/2018YFC1106001) from the Ministry of Science and Technology of

the People’s Republic of China.

References

Cai, J., Yang, J., Jones, D.P., 1998. Mitochondrial control of apoptosis: the role of cytochrome

c. Biochim. Biophys. Acta 1366, 139–149.

Carmichael, J., DeGraff, W.G., Gazdar, A.F., Minna, J.D., Mitchell, J.B., 1987. Evaluation of a

of
tetrazolium-based semiautomated colorimetric assay: assessment of chemosensitivity testing.

Cancer Res. 47(4), 936–942.

ro
Christofferson, D.E., Li, Y., and Yuan, J., 2014. Control of Life-or-Death Decisions by RIP1

-p
Kinase. Annu. Rev. Physiol. 76, 129–150. doi: 10.1146/annurev-physiol-021113-170259.

Czabotar, P.E., Lessene, G., Strasser, A., Adams, J.M., 2014. Control of apoptosis by the BCL-
re
2 protein family: implications for physiology and therapy. Nat. Rev. Mol. Cell Bio. 15, 49–
lP

63.

D'Arcy, M.S., 2019. Cell death: a review of the major forms of apoptosis, necrosis and
na

autophagy. Cell Biol. Int. 43(6), 582-592. doi: 10.1002/cbin.11137.

Edelhauser, H.F., Hine, J.E., Pederson, H., Van Horn, D.L., Schultz, R.O., 1979. The effect of
ur

phenylephrine on the cornea. Arch Ophthalmol. 97(5), 937-947.

Jo

Fan, D., Fan, T.J., 2017. Clonidine Induces Apoptosis of Human Corneal Epithelial Cells

through Death Receptors-Mediated, Mitochondria-Dependent Signaling Pathway. Toxicol.

Sci. 156(1), 252–260. doi: 10.1093/toxsci/kfw249

Fan, T. J., Xu, B., Zhao, J., Yang, H. S., Wang, R. X., and Hu, X. Z., 2011. Establishment of an

untransfected human corneal epithelial cell line and its biocompatibility with denuded

20
 amniotic membrane. Int. J. Ophthalmol. 4(3), 228–234. doi: 10.3980/j.issn.2222-3959.

Fan, T., Han, L., Cong, R., Liang, J., 2005. Caspase family proteases and apoptosis. Acta

Biochim. Biophys. Sin. (Shanghai) 37, 719–727.

Fan, W.Y., Wang, D.P., Wen, Q., Fan, T.J., 2017. The cytotoxic effect of oxybuprocaine on

human corneal epithelial cells by inducing cell cycle arrest and mitochondria-dependent

apoptosis. Hum Exp. Toxicol. 36(8), 765-775.

Galluzzi, L., Kroemer, G., 2008. Necroptosis: a specialized pathway of programmed necrosis.

of
Cell 135 (7), 1161–1163.

ro
Garrido, C., Galluzzi, L., Brunet, M., Puig, P.E., Didelot, C., Kroemer, G., 2006. Mechanisms

of cytochrome c release from mitochondria. Cell Death Differ. 13, 1423–1433.

-p
Guthrie, S., Jensen, T., Hartley, R.C., Ramaesh, K., Lockington, D., 2018. Assessing the
re
accuracy of intracameral phenylephrine preparation in cataract surgery. Eye (Lond) 32(10),

1615-1620. doi: 10.1038/s41433-018-0143-y.

lP

Hitomi, J., Christofferson, D.E., Ng, A., Yao, J., Degterev, A., Xavier, R.J., Yuan, J., 2008.
na

Identification of a molecular signaling network that regulates a cellular necrotic cell death

pathway, Cell 135 (7), 1311–1323.

ur

Jabłońska-Trypuć, A., Krętowski, R., Wołejko, E., Wydro, U., Butarewicz, A., 2019. Traumatic

acid toxicity mechanisms in human breast cancer MCF-7 cells. Regul. Toxicol. Pharmacol.
Jo

106, 137-146. doi: 10.1016/j.yrtph.2019.04.023.

Jaroch, K., Jaroch, A., Bojko, B., 2018. Cell cultures in drug discovery and development: The

need of reliable in vitro-in vivo extrapolation for pharmacodynamics and pharmacokinetics

assessment. J. Pharm. Biomed. Anal. 147, 297-312. doi: 10.1016/j.jpba.2017.07.023.

21
Jin, Z., and El-Deiry, W.S., 2005. Overview of cell death signaling pathways. Cancer Biol. Ther.,

4,139–163.

Kinoshita, S., Adachi, W., Sotozono, C., Nishida, K., Yokoi, N., Quantock, A.J., Okubo, K.,

2001. Characteristics of the human ocular surface epithelium. Prog. Retin. Eye Res. 20(5),

639-673. PubMed PMID: 11470454.

Kluck, R.M., Bossy-Wetzel, E., Green, D.R., Newmeyer, D.D., 1997. The release of

cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis. Science

of
275, 1132–1136.

ro
Knight, T., Luedtke, D., Edwards, H., Taub, J.W., Ge, Y., 2019. A delicate balance - The BCL-

2 family and its role in apoptosis, oncogenesis, and cancer therapeutics. Biochem. Pharmacol.

162, 250-261.
-p
re
Lapalus. P., Ettaiche, M., Fredj-Reygrobellet, D., Jambou, D., Elena, P.P., 1990. Cytotoxicity

studies in ophthalmology. Lens Eye Toxic. Res. 7(3-4), 231-242.

lP

Lay Suan, A.L., Hamzah, J.C., Ken, T.S., Mansurali, V.N., 2017. Intracameral mydriatics versus
na

topical mydriatics in pupil dilation for phacoemulsification cataract surgery. J. Cataract

Refract Surg. 43, 1031–1035.

ur

Li, Y., Wen, Q., Fan, T., Ge, Y., Yu, M., Sun, L., Zhao, Y., 2015. Dose dependent cytotoxicity

of pranoprofen in cultured human corneal endothelial cells by inducing apoptosis. Drug

Jo

Chem. Toxicol. 38, 16–21.

Matthews, G.M., Newbold, A., Johnstone, R.W., 2012. Intrinsic and extrinsic apoptotic

pathway signaling as determinants of histone deacetylase inhibitor antitumor activity. Adv.

Cancer Res. 116, 165–197.

22
McArthur, K., Kile, B.T., 2018. Apoptotic Caspases: Multiple or Mistaken Identities?. Trends

Cell Biol. 28(6), 475-493.

Moriwaki, K., Bertin, J., Gough, P.J., Orlowski, G.M., Chan, F.K., 2015. Differential roles of

RIPK1 and RIPK3 in TNF-induced necroptosis and chemotherapeutic agent-induced cell

death. Cell Death Dis. 6, e1636. doi: 10.1038/cddis.

Mulay, S.R., Desai, J., Kumar, S.V., Eberhard, J.N., Thomasova, D., Romoli, S., Grigorescu,

M., Kulkarni, O.P., Popper, B., Vielhauer, V., Zuchtriegel, G., Reichel, C., Bräsen, J.H.,

of
Romagnani, P., Bilyy, R., Munoz, L.E., Herrmann, M., Liapis, H., Krautwald, S., Linkermann,

ro
A., Anders, H.J., 2016. Cytotoxicity of crystals involves ripk3-mlkl-mediated necroptosis.

Nat. Commun. 7 (2016), 10274.

-p
Napoletano, F., Baron, O., Vandenabeele, P., Mollereau, B., Fanto, M., 2019. Intersections
re
between Regulated Cell Death and Autophagy. Trends Cell Biol. 29(4), 323-338.

Narita, M., Shimizu, S., Ito, T., Chittenden, T., Lutz, R.J., Matsuda, H., Tsujimoto, Y., 1998.
lP

Bax interacts with the permeability transition pore to induce permeability transition and
na

cytochrome c release in isolated mitochondria. Proc. Natl. Acad. Sci. USA 95, 14681–14686

Natarajan, S.K., Becker, D.F., 2012. Role of apoptosis-inducing factor, proline dehydrogenase,
ur

and NADPH oxidase in apoptosis and oxidative stress. Cell Health Cytoskelet. 2012 (4), 11–

27.
Jo

Pless, M., Friberg, T.R., 2003. Topical phenylephrine may result in worsening of visual loss

when used to dilate pupils in patients with vaso-occlusive disease of the optic nerve. Semin.

Ophthalmol. 18(4), 218-221.

Raizman, M.B., Hamrah, P., Holland, E.J., Kim, T., Mah, F.S., Rapuano, C.J., Ulrich, R.G.,

23
 2017. Drug-induced corneal epithelial changes. Surv. Ophthalmol. 62(3), 286-301. doi:

10.1016/j.survophthal.2016.11.008.

Rickard, J.A., O'Donnell, J.A., Evans, J.M., Lalaoui, N., Poh, A.R., Rogers, T., Vince, J.E.,

Lawlor, K.E., Ninnis, R.L., Anderton, H., Hall, C., Spall, S.K., Phesse, T.J., Abud, H.E.,

Cengia, L.H., Corbin, J., Mifsud, S., Di Rago, L., Metcalf, D., Ernst, M., Dewson, G., Roberts,

A.W., Alexander, W.S., Murphy, J.M., Ekert, P.G., Masters, S.L., Vaux, D.L., Croker, B.A.,

Gerlic, M., Silke, J., 2014. Ripk1 regulates ripk3-mlkl-driven systemic inflammation and

of
emergency hematopoiesis, Cell 157 (5), 1175–1188.

ro
Rodriguez-Enriquez, S., He, L., Lemasters, J.J., 2004. Role of mitochondrial permeability

transition pores in mitochondrial autophagy. Int. J. Biochem. Cell. Biol. 36, 2463–2472.
-p
Rönkkö, S., Vellonen, K.S., Järvinen, K., Toropainen, E., Urtti, A., 2016. Human corneal cell
re
culture models for drug toxicity studies. Drug Deliv. Transl. Res. 6, 660-675.

Santucci, R., Sinibaldi, F., Cozza, P., Polticelli, F., Fiorucci, L., 2019. Cytochrome c: An
lP

extreme multifunctional protein with a key role in cell fate. Int. J. Biol. Macromol. 136, 1237-
na

1246.

Sedlackova, L., Korolchuk, V.I., 2019. Mitochondrial quality control as a key determinant of
ur

cell survival. Biochim. Biophys. Acta Mol. Cell Res. 1866(4), 575-587. doi:

10.1016/j.bbamcr.
Jo

Shamir, E.R., Ewald, A.J., 2014. Three-dimensional organotypic culture: experimental models

of mammalian biology and disease. Nat. Rev. Mol. Cell Biol. 15(10), 647-664. doi:

10.1038/nrm3873.

Shan, M., Fan, T.J., 2016. Cytotoxicity of carteolol to human corneal epithelial cells by inducing

24
 apoptosis via triggering the Bcl-2 family protein-mediated mitochondrial pro-apoptotic

pathway. Toxicol. In Vitro 35, 36-42.

Tian, C.L., Wen, Q., Fan, T.J., 2015. Cytotoxicity of atropine to human corneal epithelial cells

by inducing cell cycle arrest and mitochondrion-dependent apoptosis. Exp. Toxicol. Pathol.

67(10), 517-524.

Vidmar, J., Chingwaru, C., Chingwaru, W., 2017. Mammalian cell models to advance our

understanding of wound healing: a review. J. Surg. Res. 210, 269-280. doi:

of
10.1016/j.jss.2016.10.016.

ro
Wang, H., Sun, L., Su, L., Rizo, J., Liu, L., Wang, L.F., Wang, F.S. and Wang, X., 2014. Mixed

lineage kinase domain-like protein MLKL causes necrotic membrane disruption upon
-p
phosphorylation by RIP3. Molecular cell, 54(1), 133-146.
re
Wen, Q., Fan, T.J., Bai, S.R., Sui, Y.L., 2015. Cytotoxicity of proparacaine to human corneal

endothelial cells in vitro. J. Toxicol. Sci. 40, 427–436.

lP

Wertz, I.E., O'Rourke, K.M., Zhou, H., Eby, M., Aravind, L., Seshagiri, S., Wu, P., Wiesmann.
na

C., Baker, R., Boone, D.L., Ma, A., Koonin, E.V., Dixit, V.M., 2004. De-ubiquitination and

ubiquitin ligase domains of A20 downregulate NF-kappaB signalling. Nature 430 (7000),
ur

694-699.

Zhou, W., Yuan, J., 2014. Snapshot: necroptosis. Cell 158 (2), 464-464.e1. doi: 10.1016/j.cell.
Jo

25
Table Legend

Table 1. Variations of cell cycle phases in HCEPCs post-exposure to 0.625% phenylephrine

Note: The number of cells in each cell cycle phase is calculated as percentage of the total

number of cells (mean ± SD, n=3). The difference of the number of HCEPCs between the 0.625%

phenylephrine-treated cells and blank control is considered as significance when ** P<0.01 and

of
*P< 0.05.

ro
Figure legends

-p
re
lP

Fig.1. Phenylephrine induced morphological changes and variations of the viability of HCEPCs.
na

The HCEPCs at passage of 121 from a non-transfected HCEPC line were treated with a serial

concentrations from double-diluting of 10% Phenylephrine. (A) The cell morphology was
ur

observed with light microscopy. Scale bar: 50 µm, ♯: shrinkage, v: vacuolation; (B) the cell
Jo

viability was assay with MTT and is expressed as percentage of absorbance at 490 nm to blank

controls (mean±SD，n=3)., *P<0.05, **P<0.01.

26
 of
ro
Fig. 2. Variations of plasma membrane permeability and DNA damage in phenylephrine-treated

HCEPCs. (A) The plasma membrane permeability of HCEPCs was detected by acridine orange
-p
(AO)/ethidium bromide (EB) double staining after exposed to phenylephrine in concentrations
re
from 10% to 0.0390625% at different time. The nucleus of the HCEPCs are swollen with

marginalized and uneven distributed chromatin post-exposure to phenylephrine from 10% to

lP

1.25%, while in concentrations equal to or less than 0.625% phenylephrine induced

na

condensation of the chromatin. (B) The plasm membrane permeability in each group is

expressed as percentage of the number of cells with red staining over total number of cells
ur

(means ± SD, n=3), *P<0.05, **P<0.01 compared to control. Scale bar: 20μm. (C) DNA

electrophoresis. Genomic DNA from the HCEPCs treated with or without phenylephrine at the
Jo

indicated concentration and exposed time was electrophoresed in a 1% agarose gel. The

damaged DNA exhibits as fragmented ladders and irregular diffusion.

27
 of
Fig. 3. Alteration of the expression and activation of necroptosis related proteins in HCEPCs.

ro
HCEPCs treated with or without phenylephrine in concentrations of 10%, 5%, 2.5%, 1.25% for

-p
the indicated time, and the expression of RIPK1, RIPK3 and MLKL were examined by western

blotting and activation of caspase-2 and -8 were assayed by ELISA. (A) Western blotting image.
re
(B)The relative level of protein expression was expressed as percentage of protein band density
lP

compared to that of β-actin at the same time point. (C, D) The activation of caspase-2 and

caspase-8 are expressed as percentage of control (means ± SD, n = 3), *P <0.05; **P <0.01 vs
na

control.
ur
Jo

28
 of
Fig. 4. The ultrastructural alterations and PS externalization of HCEPCs exposed to 0.625%

ro
phenylephrine for 4, 8 and 12h (n=3). (A) TEM microphotograph exhibited swollen

mitochondria, cytoplasmic vacuolation, condensed chromatin and apoptotic body; N: nucleus;

-p
c: condensed chromatin; m: mitochondrion; v: vacuole; a: apoptotic body. (B) PS-
re
externalized cells were assayed by FCM using Annexin V/propidium iodide (PI) staining. PS-

externalized cells in each group is expressed as percentage (mean ± SD) of the total number of
lP

cells (n=3). **P<0.01 versus the control.

na
ur
Jo

Fig. 5. 0.625% phenylephrine induced intrinsic apoptosis pathway and activation of caspases

29
in HCEPCs. The intrinsic apoptosis pathway is related with variations of the expression of Bcl-

2 family proteins, cytoplasmic translocation of Cyt. c and AIF, and ΔΨm disruption. (A, B)

Western blot images and densitometry analysis of altered expression of Bcl-2 family proteins;

(C, D) Western blot images and densitometry analysis of cytoplasmic translocation of Cyt. c,

and AIF. The relative level of protein was expressed as percentage (mean ± SD) of protein band

density compared to the internal control of β-actin (n=3). Ctrl: control group, PE:

phenylephrine-treated group, **P<0.01 versus control. (E) ΔΨm disruption was assayed by

of
FCM using JC-1 staining based on JC-1 in mitochondria with depolarized ΔΨm maintains

ro
monomers in green fluorescence, while that in mitochondria with normal ΔΨm incorporates

into aggregates in red fluorescence. The number of JC-1 positive cells (cells emitting green
-p
signals are depolarized mitochondrial membrane) was calculated as percentage (mean ± SD) of
re
the total number of cells (n=3). **P<0.01 versus control. (F) Caspase activation was

measured by ELISA using monoclonal antibodies against the active form of caspase-2, -3, -8
lP

and -9. The activation ratio was expressed as percentage (mean ± SD) compared to its
na

corresponding control based on the absorbance at 490nm (n=3). *P<0.05, **P<0.01 versus

control.
ur
Jo

30
 of
ro
Fig. 6. Ex vivo microscopic photographs of the corneal epithelial cells and corneal structure of
-p
New Zealand white rabbits 7 days after exposed to 10 % phenylephrine (the clinical
re
concentration). One representative photograph from four rabbits was shown. (A) Scanning

electron microscopy images display impaired cell junction, reduction and disappearance of
lP

microvilli of squamous cells, mv: microvilli, white arrow: impaired cell junction; (B)
na

Transmission electron microscopy images display the disorganized squamous cells and

condensed chromatin. SEM: scanning electron microscopy. TEM: transmission electron

ur

microscopy. (C) Hematoxylin and eosin (H&E) staining image exhibits loose structure and

reduced layers of 10 % phenylephrine-treated corneal epithelium; (D)The in situ DNA

Jo

fragmentation was examined by TUNEL using FITC fluorescent dye, with nuclei

counterstained with DAPI. The FITC-positive cells, DNA-fragmented cells, were shown (n=3).

Ctrl: control, PE: 10 % phenylephrine-treated group.

31
Table 1. Variations of cell cycle phases in HCEPCs post-exposure to 0.625% phenylephrine

Number of Cells (%)

G1 phase S phase G2/M phase
Control
0h 37.04±0.5 44.88±0.19 17.08±0.29
4h 37.03±1.19 46.18±0.31 16.63±1.35
8h 37.69±0.94 46.31±0.56 15.86±0.29
12h 38.31±1.4 47.75±1.24 13.79±0.54
0.625% phenylephrine
4h 35.47±0.47 49.32±0.4* 15.16±0.7
8h 37.85±0.68 52.17±0.78** 10.16±0.5*
12h 40.63±0.75 58.04±0.6** 1.28±0.27**

of
Note: The number of cells in each cell cycle phase is calculated as percentage of the total

ro
number of cells (mean ± SD, n=3). The difference of the number of HCEPCs between the 0.625%

phenylephrine-treated cells and blank control is considered as significance when ** P<0.01 and

*P< 0.05.
-p
re
lP
na
ur
Jo

32