spheroid cultures promote the stemness of corneal stromal cells

30

Click here to load reader

Upload: jiansu

Post on 07-Apr-2017

222 views

Category:

Documents


6 download

TRANSCRIPT

Page 1: Spheroid cultures promote the stemness of corneal stromal cells

Accepted Manuscript

Title: Spheroid Cultures Promote the Stemness of CornealStromal Cells

Author: Hongyang Li Ying Dai Jianchang Shu Rongjie YuYonglong Guo Jiansu Chen

PII: S0040-8166(14)00116-5DOI: http://dx.doi.org/doi:10.1016/j.tice.2014.10.008Reference: YTICE 903

To appear in: Tissue and Cell

Received date: 21-9-2014Revised date: 31-10-2014Accepted date: 31-10-2014

Please cite this article as: Li, H., Dai, Y., Shu, J., Yu, R., Guo, Y., Chen, J.,SpheroidCultures Promote the Stemness of Corneal Stromal Cells, Tissue and Cell (2014),http://dx.doi.org/10.1016/j.tice.2014.10.008

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

Page 2: Spheroid cultures promote the stemness of corneal stromal cells

Page 1 of 29

Accep

ted

Man

uscr

ipt

Highlights

Rabbit and mouse CSCs could form suspension spheroids in methylcellulose

medium.

Rabbit CSCs could form adherent spheroids by proteins of Oct4, Klf4, Sox2

and VPA.

CSC suspension and adherent spheroids were positive for vimentin, CD34 and

nestin.

The gene expression of nanog was positive for CSC adherent spheroids by

proteins

CSC suspension and adherent spheroids possess mesenchymal and stem cell

phenotypes.

*Research Highlights

Page 3: Spheroid cultures promote the stemness of corneal stromal cells

Page 2 of 29

Accep

ted

Man

uscr

ipt

Spheroid Cultures Promote the Stemness of Corneal Stromal Cells

Hongyang Lia, b

, Ying Daic, Jianchang Shu

d, Rongjie Yu

e, Yonglong Guo

c, Jiansu Chen

a, c, f

aOphthalmology Department, First Affiliated Hospital of Jinan University, Guangzhou, China,

bOphthalmology Department, Guangzhou Red Cross Hospital Affiliated to Medical College of

Jinan University, Guangzhou, China, cKey Laboratory for Regenerative Medicine of Ministry

of Education, Jinan University, Guangzhou, China, dGastroenterology Department, Guangzhou

Red Cross Hospital Affiliated to Medical College of Jinan University, Guangzhou, China,

eBioengineering Institute of Jinan University, Guangzhou, China,

fInstitute of Ophthalmology,

Medical College, Jinan University, Guangzhou, China.

Correspondence to: Jiansu Chen, Institute of Ophthalmology, Medical College, Jinan

University, 601 West Huangpu Avenue, Guangzhou 510632, China.

E-mail:[email protected]

Authors' contributions:

Conceived and designed the experiments: Hongyang Li, Jiansu Chen;

Performed the experiments: Hongyang Li, Ying Dai, Jianchang Shu, Jiansu Chen;

Analyzed the data: Hongyang Li, Ying Dai, Rongjie Yu, Jianchang Shu, Jiansu Chen;

Contributed reagents/materials/analysis tools: Jiansu Chen;

Wrote the manuscript: Hongyang Li, Jiansu Chen, Ying Dai, Yonglong Guo.

*Manuscript

Page 4: Spheroid cultures promote the stemness of corneal stromal cells

Page 3 of 29

Accep

ted

Man

uscr

ipt

Abstract

Several culture methods generated spheroids of rabbit and mouse corneal stromal cells

(CSCs) in vitro. In this study, rabbit CSC spheroids were positively expressed the mesenchymal

and stem cell phenotypes, which contained immunopositive for vimentin (a mesenchymal cell

marker) and CD34 (a stem cell marker), as well as mRNA expression of nestin (a neural stem

cell marker) and Nanog (a stem cell marker), in suspension or adherent cultures that were

induced by methylcellulose, a rotary cell culture system (RCCS) or reprogramming proteins

and VPA. Mouse CSCs showed poor growth and hardly formed spheroids after treatment with

methylcellulose or reprogramming proteins and VPA. Our work has laid a promising

foundation to elucidate CSCs and the further use of CSC spheroids for reprogramming,

bioprinting and tissue engineering.

Keywords: Spheroid culture; Corneal stromal cells; Reprogramming protein; Simulated

microgravity culture; Valproic acid

Introduction

The corneal stroma mainly consists of a dense and regularly packed collagen fibril

extracellular matrix (ECM) deposited by corneal stromal cells (CSCs) during late embryonic

development (Hassell and Birk, 2010). According to specific environmental condition and

signals, CSCs possess at least three different phenotypes, including quiescent dendritic

keratocytes, fibroblasts and myofibroblasts (Jester and Ho-Chang, 2003). The cues presented to

CSCs are the major determinants of their phenotypes. Moreover, there is a small population of

corneal stromal stem cells (CSSCs), progenitors or precursors in the corneal stroma, which is

largely located in the peripheral stroma, and CSCs represent the default lineage. These CSSCs

have a role in corneal stromal wounds healings (Pinnamaneni and Funderburgh, 2012, Mimura

et al., 2008). Isolation of CSSCs by sphere forming assay has been reported (Funderburgh et al.,

2005, Yoshida et al., 2005). These cells formed spheres in culture, showed side population

characteristics, were multipotent and expressed various adult stem cell markers. Normally, cells

in spheroid culture exhibit some properties that are distinct from monolayer cells. For example,

they grow with similar characteristics to in vivo tissue and can simulate native tissue behaviors

much more accurately than two-dimensional (2-D) cultures. Furthermore, stem cells with

Page 5: Spheroid cultures promote the stemness of corneal stromal cells

Page 4 of 29

Accep

ted

Man

uscr

ipt

self-renewing capacity usually possess spheroid forming capacity (Dontu et al., 2003, Page et

al., 2013). Thus, spheroid cultures can be used to isolate stem cells from tissue. When grown as

spheres, cells have increased cell viability and functional performance compared with

monolayer cultures (Lin and Chang, 2008). Cultured spheroids have advantages provided by

the three-dimensional (3-D) microenvironment, which maintains cell-to-cell interactions and

allows for engraftment (Hattorietal, 2010). Spheroid cultures also provide favorable conditions

for tissue engineering using bioprinting or other reconstruction techniques (Page et al., 2013,

Lin and Chang, 2008, Takács et al., 2009, Jo YH et al., 2014).

The simulated microgravity (SMG) conditions of a rotary cell culture system (RCCS) allow

cells to proliferate in under rotating conditions but with low shear stress and in a low turbulence

environment. Our previous studies showed that rabbit CSCs in SMG culture tended to

aggregate. CSCs under RCCS conditions on scaffolds of decellularized bovine cornea grew in

spheres for 19 days, but in static culture, they grew as a 2-D monolayer (Chen et al., 2007). We

also found that rabbit CSCs in SMG culture were round in shape with many prominences and

were more likely to aggregate and grow into the pores of the decellularized cornea carriers

when supplemented with valproic acid (VPA), vitamin C (VC) and 10% fetal bovine serum

(FBS). However, rabbit CSCs in static plastic culture conditions only displayed a spindle shape

and were rarely interconnected (Dai et al., 2007).

Growing cells in a 3-D environment generates important differences in cellular

characteristics and behavior compared with 2-D environments. 3-D cell culture represents an

important bridge for linking our current knowledge of cell structure and metabolism to the

extensive complexity of tissues and organs (Page et al., 2013). In this work, we investigated

CSC 3-D spheroid culture in suspension and aggregated growth induced by methylcellulose,

the recombinant cell-penetrating reprogramming of proteins PTD-Oct4, PTD-Klf4, and

PTD-Sox2 (PTD-Oct4/Klf4/Sox2), VPA and RCCS to further understand characteristics in

various environmental conditions.

Materials and Methods

Materials

Culture reagents were purchased from Gibco (Grand Island, NY, USA). Unless otherwise

stated, all the other reagents were from Sigma (St. Louis, MO, USA). VPA was from Energy

Chemical (Guangzhou, China), and B27 was purchased from Invitrogen (CA, USA). Epidermal

Page 6: Spheroid cultures promote the stemness of corneal stromal cells

Page 5 of 29

Accep

ted

Man

uscr

ipt

growth factor (EGF) and basic fibroblast growth factor (bFGF) were purchased from Pepro

Tech (CA, USA). The monoclonal anti-CD34 antiboby was purchased from Beijing

Biosynthesis Biotechnology Co., Ltd (Beijing, China), and the monoclonal anti-vimentin

antiboby (NeoMarkers) was from Lab Vision Corp (Fremont, MO, USA). The EZgene TM

Tissue RNA Miniprep Kit was from Biomiga (San Diego, CA, USA), and the ReverTra Ace

qPCR RT Kit, Blend Taq® and Blend Taq

®-Plus were purchased from Toyobo (Osaka, Japan).

Primers were synthetized by BGI (Beijing, China).

Preparation and activity of reprogramming proteins

The reprogramming proteins, including Oct4, Klf4 and Sox2, were expressed and purified

as fusion proteins with an N-terminally linked protein transduction domain (PTD) with an

amino acid sequence of YGRKKRRQRRR and a 6-His purification tag at the C-terminal,

respectively. In brief, the genes encoding the fusion proteins were cloned into the expression

vector pKYB to construct the recombinant expression vectors. After the recombinant vectors

were transformed into the ER2566 E. coli strain, the fusion proteins, such as PTD-Oct4,

PTD-Klf4 and PTD-Sox2, were expressed and purified by Ni-affinity chromatography. The

fluoresceinisothiocyanate (FITC) labeled PTD-Oct4, PTD-Klf4 and PTD-Sox2 were used to

investigate the penetrating ability of the fusion protein into Chinese hamster ovary (CHO) cells

as previously described (Li et al., 2011, Su et al., 2011, Liu et al., 2011). Briefly, the fusion

proteins were labeled with FITC (excitation 490 nm, emission 525 nm) using a FITC labeling

kit (Xirun. Bio. China). CHO cells were grown to confluence on a 24well plate, and then, the

culture medium was replaced with 200 μL FITC labeled PTD-Oct4, PTD-Klf4 or PTD-Sox2

solution for 1 hour. The cells were thoroughly washed with phosphate-buffered saline (PBS)

four times and then were imaged using an inverted fluorescence microscope. The FITC labeled

fusion protein maxadilan (MAX) was used as control, which has no penetrating ability (Zeng et

al., 2009). The rate of proteins labeled with FITC was calculated with the following formula:

F(FITC)/P(proteins)=-3.053×OD(495 nm) /OD(280 nm)-0.225×OD(495 nm). After 1 hour

incubation with FITC labeld proteins, CHO cells were treated with RIPA cell lysis buffer. A

fluorescence detector was used to observe the penetrating FITC in the cell lysis solution at an

OD of 495nm, and the protein transduction into CHO cells =OD 495nm/rate of FITC proteins

labeled. The transmembrane protein crossing efficiency was calculated as below: protein

transduction into CHO cells/total content of FITC labeled proteins in each well.

Page 7: Spheroid cultures promote the stemness of corneal stromal cells

Page 6 of 29

Accep

ted

Man

uscr

ipt

The binding activities of the reprogramming proteins, which have target sequences, were

identified using fluorescence resonance energy transfer (FRET) assays. Briefly, two

single-stranded oligo nucleotides sequences from Oct4, Klf4 and Sox2 were produced by

chemical synthesis, which connected the anthocyan dye (CY3) (excitation wavelength of 550

nm, emission wavelength of 575 nm) at the 5'end. The specific sequences of Oct4, Klf4 and

Sox2 were shown in table 1. Each double stranded DNA sequence was obtained by annealing

two reverse compliment single DNA strands, which were synthesized by Invitrogen

(Guangzhou, China). The Cy3-labeled double-stranded target DNA sequences specific binding

of Oct4, Klf4 and Sox2 were obtained by denaturing annealing (95 °C 5 min, 37 °C 2 min, 0 °C

2 min). The reprogramming proteins PTD-Oct4, PTD-Klf4 and PTD-Sox2 were labeled with

FITC (excitation 490 nm, emission 525 nm) using a FITC labeling kit (Xirun. Bio., China). The

binding of the reprogramming proteins with their target sequences Oct4, Klf4 and Sox2

resulted in energy transferring from FITC to Cy3. Fluorescence emission energy scanning from

FITC labeled reprogramming proteins following the addition of its Cy3 labeled target DNA

sequences was performed with a multiple function scanner (Perkin Elmer, German) using an

non-target DNA sequence as negative control. Moreover, the variation in the emission spectrum

was measured to confirm the fluorescence resonance energy transfer, which represents the

binding of the reprogramming proteins to their target sequences.

Animals and CSC isolation

Primary cultures were established from the corneas of the New Zealand white rabbit (4

eyes), which were aged 3-4 months with a weight range of 2-2.5 kg and the corneas of C57

mice that were 7-8 weeks old. Animals were handled in accordance with the ARVO Statement

on the Use of Animals in Ophthalmic and Vision Research. The protocol was approved by the

Institute Animal Care and Use Committee of Jinan University. All surgeries were performed

under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.

Eyes from New Zealand white rabbits were obtained and corneas were excised for CSCs.

Rabbit CSCs were isolated and expanded by explant culture as previously described (Choong et

al., 2007). Briefly, the connective tissues and external muscles of eyes were removed. The

corneas were rinsed with saline containing an antibiotic solution (prepared with 100 U/mL

penicillin G sodium and 100 mg/mL streptomycin sulfate). Corneas stripped of both endothelial

and epithelial tissues were rinsed with saline containing an antibiotic solution three times,

Page 8: Spheroid cultures promote the stemness of corneal stromal cells

Page 7 of 29

Accep

ted

Man

uscr

ipt

minced into fine pieces, and explanted onto the tissue culture plates in the CSC culture medium

which was composed of DMEM supplemented with 10% FBS. Cells were incubated at 37°C in

a 5% CO2 incubator. After three days, CSCs emerged from the small explants of corneal

stromal tissue. The explants were removed on day 7, and the CSCs were grown under adherent

conditions on tissue culture plates until 80% confluent before further passaging further. The

culture medium was changed every second day.

Mouse CSCs were enzymatically isolated from corneal stromal tissues. Briefly, eyes from

C57 mice were obtained and corneas were excised for CSCs. Connective tissue and external

muscles were then removed. The eyes were rinsed with saline containing an antibiotic solution

and placed into DMEM medium supplemented with 5 mg/mL dispase overnight at 4°C. Under

aseptic conditions, corneas were isolated and stripped of both endothelial and epithelial tissues.

Corneas were placed in a solution of 2.5 mg/mL typeⅡcollagenase in culture medium for 2

hours at 37 °C. Mouse CSCs were then rinsed in DMEM, centrifuged (1800rpm, 200g, 5 min),

and suspended at a concentration of 1×104

cells/mL in CSC culture medium supplemented with

10% FBS. The cells were seeded onto tissue culture plates and incubated at 37 °C in a 5% CO2

incubator. The culture medium was changed every second day.

CSC spheroid formation in suspension induced by methylcellulose in static and RCCS

conditions

For CSC spheroid formation in suspension under static conditions, primary rabbit and

mouse CSCs were trypsinized (TrypLETM

, Invitrogen) and seeded in 24-well uncoated tissue

culture plates (TCPS) at a concentration of 5×104 cells/mL in serum-free medium with

methylcellulose, which was composed of DMEM/F12 medium and a 0.8% methylcellulose gel

matrix supplemented with B27, 20 ng/mL EGF and 40 ng/mL bFGF. Then, cells were

incubated at 37 °C in a 5% CO2 incubator. Changes in CSCs were observed by inverted

microscope. The number and the area of rabbit CSC spheroids were measured with the Image J

software on days 3, 5 and 7, respectively.

For CSC spheroid formation in suspension under RCCS conditions, primary rabbit CSCs

were trypsinized and cultured in the SMG culture system. The first step was to slowly inject 10

mL serum-free medium with methylcellulose into the 25 mL vessel for RCCS. Then, medium

with methylcellulose containing rabbit CSCs was placed into this vessel. Finally, serum-free

medium with methylcellulose was used to fill the vessel which cells at a concentration of 5×104

Page 9: Spheroid cultures promote the stemness of corneal stromal cells

Page 8 of 29

Accep

ted

Man

uscr

ipt

cells/mL and was supplemented with B27, 20 ng/mL EGF and 40 ng/mL bFGF. Gas bubbles in

the RCCS vessel were removed. The vessel was placed into the incubator, and the rotational

speed was set at 10, 15, and 20 rpm on the first, third, seventh days of culture, respectively.

Changes in CSCs were observed with an inverted microscope. The number and the area of

rabbit CSC spheroids were measured with the Image J software on days 3, 5 and 7,

respectively.

Adherent aggregated growth of CSCs induced by reprogramming proteins and VPA

Rabbit CSCs at passage 2 (P2) with concentration of 5×104 cells/mL were treated with

reprogramming proteins ( PTD-Oct4/Klf4/Sox2) (0.2mg/mL, respectively) and VPA (0.8μg/mL)

for 16 hours, followed by replacing the same medium with medium lacking reprogramming

proteins and VPA, and culturing for an additional 56 hours before the next treatment cycle. This

treatment was repeated for seven cycles. Mouse CSCs at P2 with concentration of 5×104

cells/mL and adherent aggregated spheroids were treated with PTD-Oct4/Klf4/Sox2

(0.2mg/mL, respectively), VPA(0.8μg/mL), 2000 units/mL of leukemia inhibitory factor (LIF)

(Pepro Tech, CA, USA), and an equal volume of DMEM/F12 with 1% FBS (1:1) for four

cycles. The medium was changed every other day. The changes in spheroids were observed

with an inverted microscope.

Histological analysis

Rabbit and mouse CSCs were stained with hematoxylin-eosin (H&E) and imaged with

light microscopy. For staining, the samples were placed into 95% ethanol for 15 min after being

rinsed with PBS buffer three times, washed with tap water twice for 1 min each, stained in

small amounts of hematoxylin for 1 min, and again rinsed with tap water.

Immunofluorescent assay

Rabbit CSCs were identified with an immunofluorescence assay. Briefly, after fixation in

4% paraformaldehyde for 30 min at room temperature, CSCs were permeabilized with 0.1%

Triton X-100 in PBS for 15 min at room temperature, washed three times with PBS and

incubated with PBS containing 10% FBS for 30 min at room temperature. The cells were

incubated with the anti-vimentin (1:500) and, anti-CD34 (1:200) monoclonal antibodies for 60

min, and then with secondary antibodies for 60 min at room temperature. The cells were rinsed

with PBS twice for 3 min. Then, the samples were incubated in a moist chamber for 15 min

with DAPI to stain the nucleus. Finally, the samples were washed, and the cells were examined

Page 10: Spheroid cultures promote the stemness of corneal stromal cells

Page 9 of 29

Accep

ted

Man

uscr

ipt

by fluorescence microscopy (Olympus, Japan).

Reverse transcription-polymerase chain reaction (RT-PCR) analysis

The expression of GAPDH, nestin and CK3 in rabbit CSCs spheroids, which were

cultured under static and SMG conditions, were measured by RT-PCR. The expression of

GAPDH, Nanog, vimentin and nestin in rabbit CSCs, which were treated with reprogramming

proteins and VPA were measured by RT-PCR. Total RNA from CSCs was isolated using the

Tissue RNA Miniprep Kit, and the resulting RNA samples were quantified by measuring the

OD at 260 nm. The OD 260/280 ratios for all RNA samples were between 1.8 and 2.1. Total

RNA (1 μg) was reverse transcribed in a 10μL reaction mixture containing 2μL 5× RT Buffer,

0.5μL RT Enzyme Mix, 0.5μL Primer Mix, and 6μL nuclease-free water at 42 °C for 1 h. One

tenth of the RT product was used for subsequent PCR with the final concentration of PCR

reaction at 1× RT Buffer, 0.2 mM dNTPs, and 1.25 U Blend Taq® in a total volume of 50 μL,

using primers shown in Table 2. The PCR mixture was first denatured at 94 °C for 2 min then

amplified for 30 cycles (94 °C, 30 sec; Tm-5 °C, 30 sec; 72 °C, 1 min) using an authorized

thermal cycler (Eppendorf, Hamburg, GER). After amplification, 5 μL of each PCR product

and 1μL of 6× loading buffer were mixed and electrophoresed on a 1.5% agarose gel in 0.5×

Tris-boric acid-EDTA containing 0.5 μg/mL ethidiumbromide. Gels were photographed and

scanned.

Statistical analysis

Statistical analysis was performed with a software package (SPSS 16.0, Inc., Chicago, IL,

USA). Statistical significance that compared multiple sample sets with the control was

analyzed with repeated measures MANOVA. Data were presented as the mean ± SD.A p-value

of less than 0.001 was considered statistically significant.

Results

The transmembrane crossing efficiency of reprogramming proteins and the identification

of their binding activities with their target DNA sequences

The recombinant vectors (PKYB-PTD-Oct4/Klf4/Sox2-6His) were successfully

constructed. After transformation into ER2566 E. coil, the fusion PTD-Oct4, PTD-Klf4 and

PTD-Sox2 were expressed and purified by Ni-affinity chromatography. The imidazole gradient

concentration was set to obtain an optimal elution concentration of 60 mmol/L. CHO cells

Page 11: Spheroid cultures promote the stemness of corneal stromal cells

Page 10 of 29

Accep

ted

Man

uscr

ipt

treated with a FITC-labeled PTD-Oct4, PTD-Klf4 or PTD-Sox2 solution revealed positive

green fluorescent staining, which demonstrated that FITC-labeled PTD-proteins could enter

cells, but control cells treated with FITC-labeled MAX displayed no staining (Fig. 1 up). The

transmembrane crossing efficiency of these proteins was 28.3±2.4% (PTD-Oct4) and 7.6±1.9%,

(MAX); 22.29±2.1% (PTD-Klf4) and 6.5±1.9% (MAX); 40.86±1.97% (PTD-Sox2) and

2.7±1.41% (MAX), respectively (Fig. 1 middle). FRER analysis revealed that after the

respective addition of the Oct4, Klf4 and Sox2 target sequences to the recombinant proteins,

the fluorescence emission intensity at 565 nm, 570 nm and 570 nm increased for Oct4, Klf4

and Sox2, respectively (Fig. 1 down A), however, there was no significant fluorescence

emission intensity increase promoted by non-target DNA sequences (Fig. 1down B). The FRET

results showed that the reprogramming proteins PTD-Oct4, PTD-Klf4 and PTD-Sox2 had

specific activity that recognized and bound to their target DNA sequences.

Morphological characterization of rabbit and mouse CSCs

Inverted light microscopy imaging showed that proliferating fibroblast-like cells migrated

from the periphery of corneal stromal explants after 3 days in culture (Figure 2A). The primary

rabbit or mouse CSCs in the CSC culture medium displayed spindle and dendritic shapes

(Figure 2A, 2C). H&E staining also revealed that spindle cells were irregularly interconnected

or disconnected with one another (Figure 2B, 2D). CSCs continued to grow as a monolayer

until they reached 80% confluency after 7 days in the culture.

Rabbit CSC spheroid formation in suspension induced by methylcellulose in static and

RCCS conditions

When rabbit CSCs were cultured in serum-free medium with methylcellulose under static

or SMG conditions for 7 days, viable spheroids experienced healthy growth. Images of

representative spheroids on days 1, 3, 5, and 7 in culture were shown in Figure 3. Over time,

rabbit CSCs spheroids gradually increased in size. The cellular spheroids of the static group

were markedly larger compared with the spheroids of the SMG group. The area (Figure 3A)

and number (Figure 3B) of spheroids in the static group were significantly increased compared

with the SMG group on days 3, 5, and 7.

The immunofluorescence assay showed that rabbit CSC spheroids under static conditions

were positively stained with vimentin (a mesenchymal cell marker) and CD34 (a quiescent

keratocyte marker and hematopoietic stem cell marker) on day 7. However, small spheroids

Page 12: Spheroid cultures promote the stemness of corneal stromal cells

Page 11 of 29

Accep

ted

Man

uscr

ipt

cultured in SMG conditions were not stained with vimentin and CD34. Rabbit CSCs cultured

with DMEM-10% FBS on tissue culture plates were used as the control group. The

immunofluorescence assay showed that control CSCs stained positive for vimentin but negative

for CD34 (Figure 4). The expression of nestin, CK3 and GAPDH transcripts in rabbit CSCs

spheroids was confirmed using RT-PCR. Spheroids cultured in static conditions (Figure 4B)

expressed the nestin (a neural stem cell marker) transcript, but nestin mRNA was not detected

in the SMG (Figure 4C) and control (Figure 4A) groups. CK3 (corneal epithelial cell marker)

gene expression was not detected in any cells, but GAPDH mRNA was expressed in all groups.

Adherent aggregated growth of rabbit CSCs induced by reprogramming proteins and

VPA

Rabbit CSCs were treated with 0.2 mg/mL reprogramming proteins (PTD-Oct4/Klf4/Sox2)

and 0.8 μg/mL VPA for seven cycles (Figure 5). The adherent rabbit CSCs were spindle shaped

after the 0 and 16 hour treatments (Figure 5A, 5B). However, after three treatment cycles,

rabbit CSCs adhesion on tissue culture plates detached slightly. The spindle and dendritic

shapes of CSCs gradually changed to a short and round shape (Figure 5C). CSCs tended to

densely aggregate after five treatment cycles (Figure 5D). Moreover, CSCs were passaged at

1:6 using 0.25% trypsin. These CSC subculture also rapidly displayed adherent spheroid

aggregates on day 1 (Figure 5E). CSCs were continuously treated with reprogramming proteins

and VPA. CSCs still showed spheroid aggregates after six and seven treatment cycles (Figure

5F, 5G). However, rabbit CSCs cultured without reprogramming proteins and VPA for 21 days

displayed diverse cellular morphology and lacked aggregate formation (Figure 5H).

Immunofluorescence identification revealed that vimentin and CD34 were expressed in

rabbit CSC spheroid aggregates after seven treatment cycles with reprogramming proteins and

VPA (Figure 6A, B). Rabbit CSCs without reprogramming proteins and VPA were positively

stained for vimentin but not for CD34 (Figure 6C, D). RT-PCR analysis showed that the gene

expression of vimentin (Figure 6a), nestin (Figure 6c) and Nanog (Figure 6e) were expressed in

rabbit CSC spheroids after seven treatment cycles, but rabbit CSCs in CSC culture medium

without the treatment expressed vimentin (Figure 6b), barely any nestin (Figure 6d), and no

Nanog (Figure 6f). GAPDH was expressed in all CSCs.

Mouse CSC adherent aggregated growth and spheroid formation in suspension

For cell adherent aggregated growth induced with reprogramming proteins and VPA or

Page 13: Spheroid cultures promote the stemness of corneal stromal cells

Page 12 of 29

Accep

ted

Man

uscr

ipt

spheroid formation in suspension with methylcellulose, mouse CSCs always showed reduced

growth compared with rabbit CSCs under the same conditions. Normal spindly adherent mouse

CSCs on tissue culture plates (Figure 7A) gradually shrank and died with reprogramming

proteins and VPA treatment (Figure 7B). After four treatment cycles, there were only a few

adherent mouse CSCs that remained on the plastic dish (Figure 7C). For CSC spheroid

formation in suspension induced by methylcellulose under static conditions, mouse CSC

spheroids also were reduced in size compared with their rabbit counterparts. The images of

mouse CSC spheroids on days 1, 4, and 7 of culture were respectively shown in Figure 7D, 7E

and 7F. However, when these suspension-cultured mouse CSC spheroids were placed in

adherent culture on day7 and then treated with 0.2 mg/mL reprogramming proteins

(PTD-Oct4/Klf4/Sox2) and 0.8 μg/mL VPA, they readily attached to the surface of the culture

plates on day 1 and had healthy growth (Figure 7G). Following attachment, the adherent

spheroids maintained and generated cells that eventually repopulated into as a confluent

monolayer in medium that included reprogramming proteins and VPA on day 4 (Figure 7H) and

day 7 (Figure 7I). However, without reprogramming proteins or VPA, the adherent CSC

spheroids rapidly disappeared on day 1, CSCs were scarce.

Discussion

Cells within the mammalian body always interact with adjacent cells and the ECM. These

interactions among in the 3-D architecture often form a complex communication network in

vivo, which is critical for normal cellular and tissue physiology. Loss of tissue specific

properties is common for cells grown in 2-D monolayer cultures (Lin and Chang, 2008). 2-D

monolayer cell cultures supply an unnatural environment and generate unnatural cellular

morphology, behaviors and function. Many studies have developed 3-D cell cultures order to

reduce experimental uncertainties arising from monolayer cultures. Spheroid cultures are one of

the most common models used as 3-D cultures in research, including adherent and

non-adherent conditions. Spheroids can be generated in several ways, such as hanging-drop

culture, spinner flask or NASA rotary cell culture systems, pellet culture, etc (Page et al., 2013,

Lin and Chang, 2008, Mueller-Klieser, 1997). In the 1990s, Reynolds and Weiss first cultured

cells that exhibited stem cell properties as free-floating spheroids, called neurospheres, from the

adult brain. They enzymatically dissociated striatal tissue to single cells and plated them in

Page 14: Spheroid cultures promote the stemness of corneal stromal cells

Page 13 of 29

Accep

ted

Man

uscr

ipt

non-adherent conditions in serum-free medium and in the presence of EGF (Reynolds and

Weiss, 1992). Thereafter, spheroid forming assays are widely used in stem cell isolation and

research, which includes corneal limbal stem cells, CSSCs, corneal endothelial stem cells,

retinal stem cells and other tissue stem cells (Pastrana et al., 2011, Chang et al., 2011,

Yamagami et al., 2007, Tropepe et al., 2000). Mimura et al. produced spheroids from rabbit and

human corneal stroma (Mimura et al., 2008, Uchida et al., 2005). Their experiment confirmed

that rabbit CSC spheroids contained mesenchymal and neuronal bi-potential precursors.

In this experiment, we also showed similar results, and most cells in our rabbit spheroid

static cultures were immunopositive for vimentin (a mesenchymal cell marker) and stemness

markers of CD34 and nestin mRNA expression. We found that in serum-free medium with a

0.8% methylcellulose gel matrix supplemented with B27, EGF and bFGF (also known as

neurosphere medium), both rabbit and mouse CSCs could form spheroids in suspension. Rabbit

spheroids were markedly larger compared with mouse spheroids. We first found that the areas

and numbers of rabbit spheroids in static culture conditions were significantly increased

compared with a SMG environment. In this situation, the immunfluorescent stains for both

vimentin and CD34 were negative in the SMG group. In our previous SMG experiments and, in

the presence of decellularized bovine cornea scaffolds, rabbit CSCs were inclined to grow in

aggregate and spheroids in CSC culture medium of (DMEM and 10% FBS), and positive

immunostaining was observed for vimentin, keratocan and lumican (Chen et al., 2007, Dai et

al., 2012). According to the above mentioned results, we presumed that it is difficult to form

CSCs spheroids under dynamic SMG conditions without a scaffold.

Cell fate and function can be regulated and reprogrammed by intrinsic genetic programs

and extrinsic factors. Studies have revealed that reprogramming proteins, such as Oct4, Klf4,

Sox2, and c-Myc, with cell-penetrating peptides (CPPs) or protein transduction domains (PTDs)

can regulate cell states (Thier et al., 2010). We also found that the Oct4, Klf4 and Sox2 genes

could be expressed as reprogramming fusion proteins combined with PTD using the

prokaryotic expression vector PKYB, which successfully entered CHO cells and localized to

the nucleus. These recombinant cell-penetrating reprogramming proteins had specific binding

capacities with their target DNA sequences as assays with FRET. The transmembrane crossing

efficiencies of our reprogramming proteins were 28.3% for PTD-Oct4, 22.29% for PTD-Klf4

and 40.86% for PTD-Sox2. We demonstrated that rabbit CSCs tended to grow in aggregates

Page 15: Spheroid cultures promote the stemness of corneal stromal cells

Page 14 of 29

Accep

ted

Man

uscr

ipt

after three treatment cycles of PTD-Oct4/Klf4/Sox2 and VPA, but CD34 immunostaining was

negative (data not shown). The adherent rabbit CSCs spheroids were apparent after seven

treatment cycles. Such spheroids were immunopositive for vimentin and CD34. The stemness

markers of nestin and Nanog mRNA expression were revealed.

Compared with rabbit CSCs, mouse CSCs had different characteristics. For example, the

3G5 antigen was constitutively expressed on cultured rabbit CSCs, but mouse CSCs cultures

did not express it. 3G5 is associated with cell adhesion and involved in the regulation and

maintenance of cell shape (Stramer et al., 2004). In this study, we also found differences in

CSC spheroid formation between these two species. For instance, rabbit spheroids in

suspension were larger compared with mouse spheroids in the same serum-free medium

containing a 0.8% methylcellulose gel matrix supplemented with B27, EGF and bFGF.

Adherent rabbit aggregated spheroids occurred after reprogramming protein and VPA treatment,

but it was difficult for mouse CSCs to grow in aggregate under the same conditions. Whether

these phenomena are related to the presence of the 3G5 antigen or other elements in mouse

CSCs remains to be determined. Another study has also shown that there were species-specific

differences in spheroid formation of corneal CSCs. Bovine CSCs produced spheroids under

adherent or low attachment conditions, but human CSCs only produced spheroids under low

attachment conditions (Scott et al., 2011).

CSCs isolated from the normal cornea can be induced to grow and differentiate into

keratocytes, fibroblasts and myofibroblasts in culture. CSCs have ability to change phenotype,

which is controlled by specific environmental signals (Jester and Ho-Chang, 2003). CSCs are

specialized neural crest-derived mesenchymal corneal fibroblasts that have a bipotential nature

(Hassell and Birk, 2010, Ruberti et al., 2008). Our results showed that rabbit CSCs in

suspension and adherent spheroid cultures expressed vimentin (immunofluorescent staining)

and nestin (RT-PCR), confirming their mesenchymal and neural crest origins. Spheroid

culturing of cell aggregates is similar to the cell-cell contacts normally present in tissues. Thus,

the formation of spheroids has been widely used to study cellular and tissue properties. CSCs

spheroids can display keratocyte, mesenchymal, neuronal, stem cell phenotypes (Mimura et al.,

2008, Yoshida et al., 2005, Scott et al., 2011, Chen et al., 2009, Funderburgh et al., 2008).

Mimura et al (Mimura et al., 2008) reported that rabbit CSCs spheroids could seed onto gelatin

hydrogels to reconstruct engineered corneal stromal sheets tissues. The transplantation of such

Page 16: Spheroid cultures promote the stemness of corneal stromal cells

Page 15 of 29

Accep

ted

Man

uscr

ipt

engineered sheets into the cornea on day 28 revealed that the CD34 positive and nestin positive

cells localized to the transplanted gelatin hydrogels, which showed that CSC precursors form

seeding spheroids with a greater self-renewal potential and continued to proliferate even after

transplantation.

Cells grown in a 3-D environment produce differences of morphology, differentiation,

proliferation, viability, gene expression, and responses to stimuli compared with a 2-D

environment (Page et al., 2012). Cell spheroids have provided many advantages in the fields of

cellular biology, regenerative medicine and cell therapy. Spheroid cell survival has been shown

to be enhanced, and spheroid melanocytes had superior survival compared to monolayered

dendritic melanocytes (Lin et al., 2006). The differentiation capabilities of adipose-derived

stem cells (ASCs) were significantly enhanced after spheroid formation. ASCs cultured as

spheroids on chitosan films increased their therapeutic potentials (Cheng et al., 2012). In this

study, we studied rabbit and mouse CSC spheroid formation in suspension induced by

methylcellulose in static and RCCS conditions, as well as adherent aggregated growth of CSCs

with the reprogramming proteins PTD-Oct4/Klf4/Sox2 and VPA. We found that rabbit CSCs in

static spheroid culture conditions experienced increased growth compared with the SMG

system. Reprogramming proteins and VPA treatment was favorable to the formation and

maintenance of adherent CSC aggregated spheroids. There are still many problems prompting

us to explore CSC spheroid cultures in our next study. We will more deeply study the effects of

reprogramming proteins and the SMG system on CSC spheroid culture to elucidate the

relationship of CSC spheroids and CSC reprogramming changes. For future clinical

consideration, further study on cultured human spheroid corneal stromal cells will be evaluated.

In addition, other spheroid techniques, which can promote the survival and function of cells,

such as co-culturing spheroids, will be used in our future research (Jun et al., 2014).

Acknowledgements: We would like to thank Prof. Jintang Xu for his helpful insights in the

study of corneas, and Prof. Hongwei Pan and, Prof. Dongqing Cai for their guidance and help

in for the spheroid cultures. We thank Xiaoxia Li, Xiaofei Liu and Hang Su for their help with

the reprogramming proteins. We thank Jingxiang Zhong, Jian Chen, Jing Wu, and Yong Ding

for their help with the experiment. We also thank Zhijie Li for his help in the revision of the

manuscripts.

Page 17: Spheroid cultures promote the stemness of corneal stromal cells

Page 16 of 29

Accep

ted

Man

uscr

ipt

Funding

This work was supported by the National Natural Science Foundation of China (81371689),

and the Medical Foundation of Bureau of Health Guangzhou Municipality of China

(20141A010017).

Conflicts of interest: The authors report no conflicts of interest.

References

Chang, C.Y., McGhee, J.J., Green, C.R. and Sherwin, T. (2011): Comparison of stem cell

properties in cell populations isolated from human central and limbal corneal epithelium.

Cornea., 30, 1155-1162.

Chen, J., Chen, R. and Gao, S. (2007): Morphological characteristics and proliferation of

keratocytes cultured under simulated microgravity. Artif Organs., 31, 722-731.

Chen, Y.H., Wang, I.J. and Young, T.H. (2009): Formation of keratocyte spheroids on

chitosan-coated surface can maintain keratocyte phenotypes. Tissue Eng Part A., 15,

2001-2013.

Cheng, N.C., Wang, S. and Young, T.H. (2012): The influence of spheroid formation of

human adipose-derived stem cells on chitosan films on stemness and differentiation capabilities.

Biomaterials., 33, 1748-1758.

Choong, P.F., Mok, P.L., Cheong, S.K. and Then, K.Y. (2007): Mesenchymal stromal cell-like

characteristics of corneal keratocytes. Cytotherapy. , 9, 252-258.

Dai, Y., Chen, J., Li, H., Li, S., Chen, J., Ding, Y., Wu, J., Wang, C. and Tan, M. (2012):

Characterizing the effects of VPA, VC and RCCS on rabbit keratocytes onto decellularized

bovine cornea. PLoSOne, 7, e50114.

Dontu, G., Abdallah, W.M., Foley, J.M., Jackson, K.W., Clarke, M.F., Kawamura, M.J. and

Wicha,M.S.(2003):In vitro propagationand transcriptional profilingof humanmammarystem/pro

genitorcells. Genes Dev, 17, 1253-1270.

Funderburgh, M.L., Du, Y., Mann, M.M., SundarRajN and Funderburgh, J.L. (2005): PAX6

expression identifies progenitor cells for corneal keratocytes. FASEB J, 19, 1371-1373.

Page 18: Spheroid cultures promote the stemness of corneal stromal cells

Page 17 of 29

Accep

ted

Man

uscr

ipt

Funderburgh, M.L., Mann, M.M. and Funderburgh, J.L. (2008): Keratocyte phenotype is

enhanced in the absence of attachment to the substratum. Mol Vis, 14, 308-317.

Hassell, J.R. and Birk, D.E. (2010): The molecular basis of corneal transparency. Exp Eye Res,

91, 326-335.

Hattori, F., Chen, H., Yamashita, H., Tohyama, S., Satoh, Y.S., Yuasa, S., Li, W., Yamakawa,

H., Tanaka, T., Onitsuka, T., Shimoji, K., Ohno, Y., Egashira, T., Kaneda, R., Murata,

M., Hidaka, K., Morisaki, T., Sasaki, E., Suzuki, T., Sano, M., Makino, S., Oikawa, S. and

Fukuda, K. (2010): Non-genetic method for purifying stem cell derived cardiomyocytes. Nat

Methods, 7, 61-66.

Jester, J.V. and Ho-Chang, J. (2003): Modulation of cultured corneal keratocytephenotype by

growth factors/cytokines control in vitro contractility andextracellular matrix contraction. Exp

Eye Res, 77, 581-592.

Jo, Y.H., Jang, I.J., Nemeno, J.G., Lee, S., Kim, B.Y., Yang, W., Lee, K.M., Kim, H., Takebe, T.,

Kim, Y.S. and Lee, J.I. (2014): Artificial islets from hybrid spheroids of three pancreatic cell

lines. Transplant Proc, 46, 1156-1160.

Jun, Y., Kang, A.R., Lee, J.S., Park, S.J., Lee, D.Y., Moon, S.H. and Lee, S.H. (2014):

Microchip-based engineering of super-pancreaticis lets supported by adipose-derived stem cells.

Biomaterials, 35, 4815-26.

Li, H.Y., Li, X.X., Yu, R.J., Liu, X.F., Wang, J.J., Su, H. and Chen, J.S. (2011): Preparation of

recombinant OCT4 with penetrating peptide PTD and activity assay. [Article in Chinese].

Chinese Journal of Pathophysiology, 27, 1790-1795.

Lin, R.Z. and Chang, H.Y. (2008): Recent advances in three-dimensional multicellular spheroid

culture for biomedical research.Biotechnol J, 3, 1172-1184.

Lin, S.J., Jee, S.H., Hsiao, W.C., Yu, H.S., Tsai, T.F., Chen, J.S., Hsu, C.J. and Young, T.H.

(2006): Enhanced cell survival of melanocyte spheroids inserum starvation condition.

Biomaterials, 27, 1462-1469.

Liu, X.F., Su, H., Wang, J.J., Su, H., Lin, H. and Chen, J.S. (2011): Preparation and activity

identification of recombinantPTD-Sox2. [Article in Chinese]. Biotechnology, 19, 23-29.

Mimura, T., Amano, S., Yokoo, S., Uchida, S. and Yamagami, S. (2008): Isolation and

distribution of rabbit keratocyte precursors. Mol Vis, 14, 197-203.

Mimura, T., Amano, S., Yokoo, S., Uchida, S., Yamagami, S., Usui, T., Kimura, Y. and Tabata,

Page 19: Spheroid cultures promote the stemness of corneal stromal cells

Page 18 of 29

Accep

ted

Man

uscr

ipt

Y. (2008): Tissue engineering of corneal stroma with rabbit fibroblast precursors and gelatin

hydrogels. Mol Vis, 14, 1819-1828.

Mueller-Klieser, W. (1997): Three-dimensional cell cultures: from molecularmechanisms to

clinical applications. Am J Physiol, 273, C1109-1123.

Page, H., Flood, P. and Reynaud, E.G. (2013): Three-dimensional tissue cultures: current trends

and beyond. Cell Tissue Res, 352, 123-131.

Pastrana, E., Silva-Vargas, V. and Doetsch, F. (2011): Eyes wide open: a critical review of

spheroid -formation as an assay for stem cells. Cell Stem Cell, 8, 486-498.

Pinnamaneni, N. and Funderburgh, J.L. (2012): Concise review: Stem cells in the corneal

stroma. Stem Cells, 30, 1059-1063.

Reynolds, B.A. and Weiss, S. (1992): Generation of neurons and astrocytes from isolated cells

of the adult mammalian central nervous system. Science, 255, 1707-1710.

Ruberti, J.W. and Zieske, J.D. (2008): Prelude to corneal tissue engineering gaining control of

collagen organization. Prog Retin Eye Res, 27, 549-577.

Scott, S.G., Jun, A.S. and Chakravarti, S. (2011): Spheroid formation from corneal keratocytes

and phenotype specific markers. Exp Eye Res, 93: 898-905.

Stramer, B.M., Kwok, M.G., Farthing-Nayak, P.J., Jung, J.C., Fini, M.E. and Nayak, R.C.

(2004): Monoclonal antibody (3G5)-defined ganglioside: cell surface marker of corneal

keratocytes. Invest Ophthalmol Vis Sci, 45, 807-812.

Su, H., Yu, R.J., Liu, X.F., Wang, J.J., Li, X.X. and Chen, J.S. (2011): Preparation of

recombinant PTD-KLF4 with penetrating peptide and activity Assay. [Article in Chinese].

China Biotechnology, 31, 48-54.

Takács, L., Tóth, E., Berta, A. and Vereb, G. (2009): Stem cells of the adult cornea: from

cytometric markers to therapeutic applications. Cytometry A, 75, 54-66.

Their, M., Münst, B. and Edenhofer, F. (2010): Exploring refined conditions for reprogramming

cells by recombinant Oct4 protein. Int J DevBiol, 54, 1713-1721.

Tropepe, V., Coles, B.L., Chiasson, B.J., Horsford, D.J., Elia, A.J., McInnes, R.R. and Van der

Kooy, D. (2000): Retinal stem cells in the adult mammalianeye. Science, 287, 2032-2036.

Uchida, S., Yokoo, S., Yanagi, Y., Usui, T., Yokota, C., Mimura, T., Araie, M., Yamagami, S.

and Amano, S. (2005): Spheroidformation and expression ofneural proteins byhuman corneal

stromal cells in vitro. Invest Ophthalmol Vis Sci, 46, 1620-1625.

Page 20: Spheroid cultures promote the stemness of corneal stromal cells

Page 19 of 29

Accep

ted

Man

uscr

ipt

Yamagami, S., Yokoo, S., Mimura, T., Takato, T., Araie, M. and Amano, S. (2007): Distribution

of precursors in human corneal stromal cells and endothelialcells. Ophthalmology, 114,

433-439.

Yoshida, S., Shimmura, S., Shimazaki, J., Shinozaki, N. and Tsubota, K. (2005): Serum-free

spheroid culture of mouse corneal keratocytes. Invest Ophthalmol Vis Sci, 46, 1653-1658.

Zeng, L., Yu, R., Xu, M., Chen, J., Wang, J. and Li, J. (2009):

Preparation and identification of recombinant PTD-maxadilan. [Article in Chinese]. Sheng Wu

Gong Cheng XueBao, 25, 1739-1745.

Page 21: Spheroid cultures promote the stemness of corneal stromal cells

Page 20 of 29

Accep

ted

Man

uscr

ipt

Figure caption

Figure 1: The transmembrane efficiency of reprogramming proteins and the

identification of their binding activities with their target DNA sequences. CHO cells

treated with FITC labeling PTD-Oct4, PTD-Klf4 and PTD-Sox2 revealed positive green

fluorescein staining, while cells in control treated with FITC labeling MAX displayed negative

staining (up). The transmembrane efficiency of proteins was 28.3±2.4% (PTD-Oct4) and

7.6±1.9%, (MAX); 22.29±2.1% (PTD-Klf4) and 6.5±1.9% (MAX); 40.86±1.97% (PTD-Sox2)

and 2.7±1.41% (MAX) respectively (middle). There were significant FRET signals (down) on

565nm (PTD-Oct4), 570nm (PTD-Klf4) and PTD-Sox2 (570nm) (A, indicated by arrow heads),

while no FRET signal between reprogramming proteins and non-target sequence (B).

Figure 2: The morphological characterization of rabbit and mouse CSCs. Primary rabbit

CSCs migrated out from periphery of corneal stromal explants for 3 days of the culture (A).

Primary mouse CSCs were cultured on plates on day 3. (C). H&E staining showed that rabbit

(B) and mouse CSCs (D) on plastic in conventional culture.

Figure 3: Spheroid formation from rabbit CSCs in static and SMG conditions on day1, 3,

5 and 7. With time, rabbit CSC spheroid gradually increased. Rabbit CSCs formed larger

spheroids in static culture than that under SMG culture. The areas of rabbit spheroid derived

from samples of the static condition were significantly higher than that for samples of the SMG

condition (A). The numbers of spheroid obtained from samples of the static condition were

significantly higher than that for samples of the SMG condition (B). (MANOVA of repeated

measuring was performed, p=0.0001, n=10)

Figure 4: The immunofluorescence staining and RT-PCR analysis for rabbit CSC

spheroid in three groups on day 7 of culture. Rabbit CSC spheroids were positively stained

for immunofluorescence of vimentin and CD34 in static condition, while did not detect CD34

expression in control group and SMG group. RT-PCR analysis showed that rabbit CSC

spheroids cultured in static condition (B) expressed nestin transcript. However, nestin was not

expressed in rabbit CSCs cultured on tissue culture plates (A) or in SMG condition (C). No

Page 22: Spheroid cultures promote the stemness of corneal stromal cells

Page 21 of 29

Accep

ted

Man

uscr

ipt

gene expression of K3 was detected in all groups. GAPDH were expressed in the cells from all

groups.

Figure 5: The cellular morphology of adherent rabbit CSC spheroids by the treatment of

reprogramming proteins PTD-Oct4/Klf4/Sox2 and VPA. The adherent rabbit CSCs were

culture after 0 hour (A) and 16 hour (B) treatments. Rabbit CSCs became round after three

cycles treatment(C) and cells showed aggregate tendency after the fifth cycle (D). The

aggregations became spheroids after the fifth cycle treatment and subculture (E). The spheroid

aggregates maintained after subculture and the sixth cycle (F) and seventh cycle (G) treatment.

Rabbit CSCs cultured without reprogramming proteins and VPA for 21 days displayed diverse

cellular morphology and lack of aggregate formation (H).

Figure 6: The immunofluorescence staining and RT-PCR analysis of rabbit CSCs after

treatment of reprogramming proteins and VPA. Rabbit CSC spheroids were positively

stained for vimentin (A) and CD34 (B) after the seventh cycle treatment. Rabbit CSCs in the

CSCs culture medium positively stained for vimentin but negatively stained for CD34. The

gene expressions of vimentin (A), nestin (C) and nanog (E) of rabbit CSCs were positively

displayed after the seventh cycle treatment of reprogramming proteins and VPA. But rabbit

CSCs in CSC culture medium without the treatment expressed vimentin (B), scarcely any

nestin (D), and did not expressed nanog (F). GAPDH were expressed in all CSCs.

Figure 7: Light microscopic observation of mouse CSC spheroids. Under reprogramming

proteins and VPA, normal spindle adherent mouse CSCs (A) showed poor growing status (B).

After four cycle treatment of reprogramming proteins and VPA, there were only a few of

adherent mouse CSCs (C). Suspension mouse CSC spheroids were showed in static spheroid

culture on day 1 (D), day 4 (E) and day 7 (F). Such suspension cultured mouse CSC spheroids

on day 7 placed into adherent culture and then treated with reprogramming proteins and VPA,

they readily attached to the surface of culture plates and grew well on day 1 (G), day 4 (H) and

day 7 (I).

Page 23: Spheroid cultures promote the stemness of corneal stromal cells

Page 22 of 29

Accep

ted

Man

uscr

ipt

Table 1. List of the specific sequences of OCT4, Klf4 and Sox2

Oct4-F: 5'-cy3- ATGCATGCAAATATGCAAAT-3'

Oct4-R: 5'-cy3- CAGT ATTTGCATATTTGCAT-3'

Klf4-F: 5'-cy3-ATGCACCCCAGTCACCCTAGC-3'

Klf4-R: 5'-cy3-TCTAGGGTGATAGGGTGCAT-3'

Sox2-F: 5'-cy3-CAGTCAAACAAAGACAAACAAAGAGCAT-3'

Sox2-R: 5'-cy3-ATGCACTTTGTTTGTCTTTGTTTGACTG-3'

Table 2: List of primers of Vimentin, Nestin, Cytokeratin3, Nanog

Primers Sequences (5’to 3’)

Vimentin(sense) 5′- CTT CTC AGC ATC ACG ATG ACC −3′

Vimentin(antisense) 5′- ATC TAT CTT GCG CTC CTG −3′

Nestin(sense) 5′- TTG AGA C(A/T)C CTG TG(C/A) CAG CCT −3′

Nestin(antisense) 5′- CTC TAG AC (T/C) CAC (T/C)GG ATT CT−3′

Cytokeratin3(sense) 5′- GCA GCAGCA GGA TGA GCT G −3′

Cytokeratin3(antisense) 5′- GTT GAG GGT CTT GAT CTG−3′

GAPDH(sense) 5′- CAT CAC CAT CTT CCA GGA GC −3′

GAPDH(antisense) 5′- ACA ATG CCG AAG TGG TCG −3′

Nanog(sense) 5′- AGG CAC CCC TGG TGG TAA GCA −3′

Nanog(antisense) 5′- ACC ACT CCA ACA CAG GCA GTT CC−3′

Table