Human Cell 2010; 23: 141151 doi: 10.1111/j.1749-0774.2010.00096.

RESEARCH ARTICLE huc_96 141..151

Value of human amniotic epithelial cells in tissue

engineering for cornea
Simat Siti FATIMAH,1 Sook Luan NG,1 Kien Hui CHUA,2 Abdul Rahman HAYATI,1 Ay Eeng TAN3
and Geok Chin TAN1
Departments of 1Pathology, 2Physiology and 3Obstetrics and Gynaecology, Universiti Kebangsaan Malaysia, Kuala
Lumpur, Malaysia

Abstract
Human amniotic epithelial cells (hAECs) are potentially one of the key players in tissue
engineering due to their easy availability. The aim of the present study was to develop an
optimal isolation and transportation technique, as well as to determine the immunophenotype
and epithelial gene expression of hAECs. Amnion was mechanically peeled off from the chorion
and digested with trypsin-ethylenediaminetetraacetic acid. The isolated hAECs were cultured in
medium containing 10 ng/mL epidermal growth factor until P4. The epithelial gene expression,
cell surface antigen and protein expression of hAECs were analyzed by quantitative poly-
merase chain reaction, ow cytometry and immunocytochemistry. hAECs were also cultured in
adipogenic, osteogenic and neurogenic induction media. The best cell yield of hAECs was seen
in the digestion of 15 pieces of amnion (2 2 cm) and isolated 30 min after digestion with
trypsin. F12:Dulbeccos modied eagle medium was the best medium for short term storage
at 4 C. hAECs expressed CD9, CD44, CD73 and CD90, and negligibly expressed CD31, CD34,
CD45 and CD117. After serial passage, CK3, CK19 and involucrin gene expressions were
upregulated, while p63, CK1 and CK14 gene expressions were downregulated. Sustained gene
expressions of integrin b1 and CK18 were observed. At initial culture, these cells might have
stem-like properties. However, they differentiated after serial passage. Nonetheless, hAECs
have epithelial stem cell characteristics and have the potential to differentiate into corneal
epithelial cells. Further investigations are still needed to elucidate the mechanism of differen-
tiation involved and to optimize the culture condition for long term in vitro culture.

Key words: amniotic epithelium, cornea, placenta, regenerative medicine, stem cells.

INTRODUCTION eliminates waste from the fetus. Placenta is involved in

Placenta is a complex organ that connects mother to the
developing fetus through the umbilical cord. The moth-
ers blood provides nutrient and gas exchange, and
Correspondence: Dr Geok Chin Tan, Department of
Pathology, Universiti Kebangsaan Malaysia, Jalan Yaacob
Latif, Bandar Tun Razak, 56000 Cheras, Kuala Lumpur,
Malaysia. Email: tan_geok_chin@yahoo.com
Received 30 July 2010; accepted 22 September 2010
maintaining fetal tolerance and displays cells with
immunomodulatory properties. Many cell types can be
isolated from the placenta, which is a potential source of
cells for regenerative medicine.1
Amnion is a component of fetal membrane that is
composed of a layer of columnar epithelial cells, base-
ment membrane, an acellular compact layer with the
underlying fibroblast and a spongy layer.1 Human amni-
otic epithelial cells (hAECs) develop on the 8th day of
fertilization, long before gastrulation (days 1517) and,
hence, are believed to retain some or all of the epiblast

2010 The Authors

Human Cell 2010 Japan Human Cell Society 141
SS Fatimah et al.
pluripotency. Studies have shown that hAECs expressed
the stem cell genes and neurogenic genes that include
Oct 4, SOX-2, FGF-4, Rex-1, Nanog3, ABCG2,
Nestin, Vimentin and BST-1.2,3 Miki et al. and
Portmann-Lanz et al. reported that hAECs could be dif-
ferentiated to adipogenic, osteogenic, chondrogenic,
neurogenic, myogenic, hepatocyte and cardiomyocyte
-lineage cells.4,5
Human amniotic epithelial cells have therapeutic
potential in the treatment of Type I diabetes6 and in
tissue engineering of skin.7 Amnion has also been used
as temporary skin substitutes8 and as a supporting mate-
rial for the treatment of ocular surface disorders.9 More-
over, hAECs have immunoprivileged characteristics10
and therefore the risks of rejection upon transplantation
might be reduced. However, it is known that epithelial
cells have limited growth period in cell culture and their
long-term growth is difficult to sustain. Boyce and Ham
showed that human keratinocytes had a declined
growth rate and colony forming efficiency between the
third and seventh serial passage. This could be due to
the in vitro cell differentiation or maturation.11 In order
to make it possible for clinical usage, optimal culture
condition to generate large quantity of epithelial cells in
serial passage is required.
Despite the therapeutic potential of hAECs, under-
standing of their cell biology remains largely unclear. A
number of epithelial genes have been used in the study
of keratinocytes and cornea. Gene expression level of
epithelial stem cells marker, for example integrin b112
and CK19;13 some differentiated cytokeratin markers
such as CK1,14 CK3,15 CK1414 and CK18;16 transcription
factor namely p63;17 and involucrin,18 have been used to
assess the epithelial characteristics together with the
immunostaining analysis. These markers may be valu-
able to assess and understand the genotypic and phe-
notypic changes and differentiation potential of hAECs
after serial passage. The aim of this study was to develop
an optimal isolation and transportation technique, as
well as to determine the immunophenotype and epithe-
lial gene expression profile of hAECs.
METHODS
This research was approved by the University Research
and Ethics Committee (FF-272-2007). Informed,
written consent was obtained from each donor prior to
collection. Human term placentae were obtained after
uncomplicated, elective caesarean section from healthy
mothers aged between 20 and 40 years.
142
Isolation and storage of hAECs
Amnion was mechanically peeled off from chorion and
washed several times with sterile phosphate buffered
saline (PBS, Gibco/BRL, Grand Island, NY, USA). The
optimal yield in hAECs isolation was determined by
investigating the number of pieces (5, 10, 15, 20 and 25
pieces) of amniotic tissue (2 2 cm) needed for diges-
tion. To identify a suitable transport buffer for amnion
sample, 15 pieces of amnion (2 2 cm) were stored at
4 C in different transport buffers, which comprised
basal medium of nutrient mixture F12: Dulbeccos
modified eagle medium (F12: DMEM, Gibco/BRL)
(1:1), PBS and normal saline. hAECs were isolated at 6,
24 and 48 h after storage in the three different transport
buffers. These tissues were digested in 0.05% trypsin
containing 0.2 g/L of ethylenediaminetetraacetic acid
(EDTA)4 Na (Invitrogen, Calrsbad, CA, USA, Gibco/
BRL) and incubated at 37 C with constant agitation.
hAECs were harvested after 10 min, followed by the
next 30 min and 60 min of digestion, and centrifuge at
600 g. The trypsin reaction was inhibited by adding
F12:DMEM (1:1) supplemented with 10% heat-
inactivated fetal bovine serum (FBS, Hyclone, Waltham,
MA, USA). Number of hAECs was counted using
hemocytometer and tryphan blue dye exclusion (Gibco/
BRL) was used to determine the number of viable cells
isolated. The amnion before and after trypsin digestion
were fixed in 10% buffered formalin and stained with
hematoxylin & eosin (H&E) to assess the completion of
epithelial cells dissociation (Fig. 1).
Human amniotic epithelial cell culture
Human amniotic epithelial cells were plated on six-well
plates at 20 000 cells/cm2 seeding density in culture
medium of F12: DMEM (1:1) supplemented with 10%
FBS, 10 ng/mL epidermal growth factor (EGF, Pepro-
tech, Rocky Hill, NJ, USA), 1% antibiotic-antimycotic
(Invitrogen), 1% Glutamax (Invitrogen) and 1% Vitamin
C (Merck, Darmstadt, Germany). The culture plates
were incubated in a humidified incubator under stan-
dard conditions at 37 C and 5% CO2. hAECs in passage
0 (P0) were initially split at a ratio of 1:3 to passage 1
(P1). Subsequent passages until P4 were at a split ratio
of 1:2. Splitting was performed when cells reached
approximately 90% confluence.
Colony-forming efficiency
To determine the self renewal ability of hAECs at P2,
they were seeded in a 60-mm diameter Petri dish with
2010 The Authors
Human Cell 2010 Japan Human Cell Society

Figure 1 Amniotic membrane before
(a) and after (b) trypsin digestion
(200). The epithelial layer was
completely removed from the base-
ment membrane after treatment with
0.05% trypsin-ethylenediaminetetra-
acetic acid (EDTA). (BM, basement
membrane; E, epithelium; S, stromal
layer).
serial dilutions of seeding density of 50, 100, 150, 200
and 250 cells/cm2. Medium was changed every 23
days. Cultures were maintained for 21 days before they
were fixed in 10% formalin. The number of hAECs
colonies, comprising more than four cells, were counted
after being stained with May Grunwald Giemsa. Colony-
forming efficiency (CFE) was calculated using the
formula, CFE = (colonies formed/no. of cells inoculated
at the start of the growth period) 100.
In vitro differentiation analysis
The ability of differentiation of hAECs was assessed.
Cells were cultured in adipogenic, osteogenic and neu-
rogenic induction media.
Adipogenic induction
To induce adipocytes differentiation, hAECs at P2 were
split at a splitting ratio of 1:2 and cultured in DMEM
(high glucose) supplemented with 10% FBS, 200 mM
indomethacin, 10 mM insulin, 0.5 mM 3-isobutyl-1-
methylxanthine (IBMX) and 1 mM Dexamethasone.
Culture was maintained for 3 weeks. Medium was
changed every 23 days. The cells were fixed in 10%
formalin and stained with oil-red O to determine the
presence of lipids.
Osteogenic induction
To induce osteocytes differentiation, hAECs at P2 were
split at a splitting ratio of 1:2 and cultured in DMEM
(high glucose) supplemented with 10% FBS, 10 mM
b-glycerophosphate, 0.2 mM ascorbic acid, and 1
108 M dexamethasone. Medium was changed every 23
days. After 3 weeks, the cells were fixed in 10% formalin
and stained with Alizarin Red to identify the presence of
mineralization.
2010 The Authors
Human Cell 2010 Japan Human Cell Society
hAECs and tissue engineering
Neurogenic differentiation
To induce neurogenic differentiation, hAECs at P1 were
split at a splitting ratio of 1:3 and cultured in Neurobasal
medium (1, Invitrogen) supplemented with 5 105 M
all-trans retinoic acid (Sigma, St. Louis, MO, USA),
10 ng/mL FGF-4 (Peprotech), N-2 supplement (100X,
Gibco, Invitrogen) and B-27 supplements (50, Gibco,
Invitrogen). After 7 days, the cells were fixed in 4%
paraformaldehyde and immunocytochemistry was
carried out using one of the mouse monoclonal antibod-
ies (DAKO, Glostrup, Denmark): NSE (neurone specific
enolase, 1:175), GFAP (glial fibrillary acidic protein),
NFM (neurofilament) or vimentin.
Flow cytometry analysis
hAECs in serial passages (P0, P1, P2 & P4) were disso-
ciated using Accutase and suspended in 0.5% bovine
serum albumin (BSA, Sigma). Fluid sheath (100 mL) was
added to 5 105 cells and incubated in 10 mL of primary
antibody for 15 min in the dark at room temperature.
Fluorescein isothiocyanate (FITC)-conjugated antibod-
ies against CD34, CD44, CD45, CD90, HLA-A, B, C and
HLA-DR, DP, DQ; and phycoerythin (PE) conjugated
antibodies against CD9, CD31, CD73 and CD117 were
used (BD Bioscience, San Diego, CA, USA). After anti-
body incubation, the cells were washed with fluid
sheath and fixed in 500 mL of Cell Fix (BD Bioscience).
Negative control was incubated with fluid sheath
without primary antibody. The samples were analyzed
by fluorescence-activated cell sorting (FACS) Calibur
cytometer using CELL Quest software (BD Bioscience).
Gene expression analysis
Total RNA extraction
Total RNA was extracted from hAECs by using TRI-
Reagent (Molecular Research Centre, Cincinnati, OH,
143
SS Fatimah et al.
Table 1 Primers used in the polymerase chain reaction (PCR)
Gene Accession no.
Cytokeratin 1 NM_006121
Cytokeratin 3 NM_057088
Cytokeratin 14 NM_000526
Cytokeratin 18 NM_000224
Cytokeratin 19 NM_002276
Integrin b1 NM_004763
Involucrin NM_005547
P63 NM_003722
USA) according to the manufacturers instruction. 5 mL
of polyacryl carrier (Molecular Research Centre, USA)
was added to precipitate the total RNA. The RNA pellet
was washed with 75% ethanol and dried before dissolv-
ing in DNAse- and RNAse-free distilled water (Gibco/
BRL). The yield and RNA integrity was determined by
spectrophotometer (Bio-Rad, Hercules, CA, USA). The
extracted RNA was immediately stored at 80 C for
further analysis.
Reverse transcription (cDNA synthesis)
Five microliters of the total RNA was used to synthesize
cDNA using Superscript III First-Strand Synthesis kit
(Invitrogen) which contained reverse transcriptase
enzyme mix, 2 reverse transcription (RT) reaction mix
and Escherichia coli RNase H. The reaction was per-
formed according to the manufacturers instructions.
The cDNA was stored at 20 C for further analysis.
Real-time polymerase chain reaction
The resulting cDNA was subjected to real-time poly-
merase chain reaction (PCR) using IQ SYBR Green
Supermix (Bio-Rad). CK1, CK3, CK14, CK18, CK19,
Integrin b1, involucrin and p63 gene expression levels
were assessed. The forward and reverse primers were
obtained from NIH GenBank database (Table 1).
glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
was used as the housekeeping gene. PCR was performed
in an iCycler IQ5 (Bio-Rad) with activation of Taq DNA
polymerase at 95 C for 3 min and 30 s. PCR amplifica-
tion was carried out for 40 cycles, each cycle contained
95 C for 10 s and 61 C for 30 s. The PCR products
144
Primer sequences (5 3) Product size (bp)
F: ggg tgg tta tgg tcc tgt ctg 128
R: gct ccc ttt ctc gag act tca
F: agc ttc agg cca aag tgg at 154
R: atg atg ctg tcc agg tcc ag
F: cga gga atg gtt ctt cac ca 151
R: ttt cat gct gag ctg gga ct
F: ttg acc gtg gag gta gat gc 188
R: gtc gtc tca gca gct cca ac
F: aag gcc tga agg aag agc tg 104
R: gga atc cac ctc cac act ga
F: gcc ttg aag ggc cat tag ac 166
R: tag aga gca tgc ctg tgc aa
F: gaa aca gcc aac tcc act gc 123
R: cat tct tgc tca ggc agt cc
F: gaa acc aga gat ggg caa gtc 145
R: gct tcg tac cat cac cgt tct
specificity was further verified by 2% agarose (Invitro-
gen) gel electrophoresis stained with ethidium bromide
(Sigma) and visualized by UV transillumination (Vilber
Lourmat, Marne La Vallee, France).
Immunocytochemistry analysis hAECs at P0 and P4
were cultured on chamber slides and fixed with 4%
paraformaldehyde in PBS (pH7.2, Gibco/BRL) for
30 min at 4 C. The cells were permeabilized with
acetone for 5 min at 4 C. Then washed for 3 min and
incubated in 0.3% hydrogen peroxide (H2O2) to inhibit
the endogenous peroxidase activity, followed by rinsing
in Tris Buffered Saline (TBS, Lab Vision, Fremont, CA,
USA). The slides were then incubated in one of the
following primary mouse monoclonal antibodies:
mouse anti-CK14 (1:150; Lab Vision), mouse anti-
involucrin (1:150; Lab Vision) and mouse anti-CK18
(1:50, DAKO) for 30 min at room temperature. Samples
were then rinsed with TBS (Lab Vision) and incubated
for 30 min with horse radish peroxidase (HRP) labeled
polymer goat secondary antibody molecules against
mouse immunoglobulin (DAKO). To develop the signal,
3, 3-diaminobenzidine (DAB; DAKO) was added, fol-
lowed by hematoxylin nuclear counterstaining.
Statistical analysis
Quantitative data are presented as group mean stan-
dard error of the mean (SEM). The results obtained were
collected from six to eight samples. The data were ana-
lyzed for normality test and tested for homogeneity of
variances using Shapiro Wilks Statistics and Levenes
Test, respectively. To identify the significant difference,
2010 The Authors
Human Cell 2010 Japan Human Cell Society
 hAECs and tissue engineering

one-way between-groups ANOVA with a pairwise com-

parison using Dunnette test were used if the assump-
tions above were not violated. The statistical analysis
was performed using SPSS v.13.0 for Windows. P < 0.05
was considered to be statistically significant.

RESULTS
Isolation and storage conditions
of hAECs
One of the 15 amnion collected failed to isolate hAECs,
hence this isolation protocol has 93% successful rate.
On average, trypsin-EDTA enzymatic digestion yielded
5.09 107 hAECs per amnion. The first 10 min of
digestion yielded the least number of viable cells. The
most number of viable cells was isolated 30 min after
digestion. There was a positive linear correlation
between the total numbers of viable cells acquired with
the amount of tissue used during enzymatic digestion. Figure 2 (a) The total number of viable human amniotic
However, the increase became insignificant from 20 epithelial cells (hAECs) isolated in relation to the number of
pieces of amnion onward (Fig. 2a). The total numbers of pieces of aminotic tissue and time of isolation. Viable cells
viable cells obtained from 15, 20 and 25 pieces of isolated from 15, 20 and 25 pieces of amnion 2 2 cm were
amnion (2 2 cm) were significantly higher than five higher than five pieces of amnion at 30 (*P < 0.05) and
pieces of amnion at the next 30 and 60 min after diges- 60 min (**P < 0.05) after digestion. (b) The cell viability
tion (P < 0.05). We believe 15 pieces of amnion was the after storage at 4 C for 6, 24 and 48 h in different transport
buffer. Total number of viable cells isolated was significantly
minimal amount required to obtain the best cell yield.
reduced in phosphate-buffered saline (PBS) (*P < 0.05) and
Prior to its isolation, amnion was stored in three saline (**P < 0.05) when stored for 24 and 48 h compared
different buffers for transportation (PBS, normal saline with 6 h. Data presented as mean standard error of the
and F12/DMEM) to determine which buffer is the best mean (SEM); n = 6.
to transport the amnion from operation theatre to labo-
ratory for processing. There was no significant differ-
ence in the total numbers of viable cells isolated from
the three different transport buffers at 6 h. However, if
amnion was stored at 4 C for 24 and 48 h, the greatest Colony forming efficiency of hAECs
viable cell yield was observed in amnion stored in
The number of cell colonies was determined at P2 with
F12:DMEM when compared with PBS and saline (P <
a colony forming efficiency of 5.7% and frequency of
0.05) (Fig. 2b).
one colony per 23 HAECs plated.

Morphology of the hAECs

Isolated hAECs cultured in vitro at high seeding density
reached confluence in 23 days. 10 ng/mL EGF was
added to the culture medium to promote hAECs prolif-
eration. At the initial culture, hAECs were plastic adher-
ent and polygonal in shape (Fig. 3a). At P4, the cultured
hAECs reached confluence at 810 days of culture, and
displayed large, flattened morphology where their
nucleus to cytoplasm ratio was lower compared with
hAECs at P0 (Fig. 3b).
In vitro differentiation ability of hAECs
hAECs displayed the ability to differentiate into neuro-
ectodermal and mesodermal lineages.
Neuro-ectodermal differentiation
Morphologically, hAECs cultured in neurogenic
medium at P2 showed scanty cytoplasm with

2010 The Authors

Human Cell 2010 Japan Human Cell Society 145
SS Fatimah et al.

Figure 3 Morphology human amni-

otic epithelial cells (hAECs) in
F12:Dulbeccos Modified Eagle
Medium (DMEM) (1:1) (FD) + 10%
fetal bovine serum (FBS) + 10 ng/mL
epidermal growth factor (EGF). (a) At
P0, hAECs were small and polygonal
in shaped (100). (b) At P4, they
were larger with abundant cytoplasm
(100). (c) hAECs cultured in
neurogenic induction medium dis-
played cytoplasmic immunoreactivity
(neurone specific enolase [NSE],
100) and (d) (vimentin, 40). They
did not express glial fibrillary acidic
protein (GFAP) (40) (e) and neu-
rofilament (NFM) (40) (f). (g) Oil
red O positivity in hAECs after
growing in adipogenic induction
medium (arrow) (400). (h) Alizarin
red positivity in hAECs after growing
in osteogenic induction medium
(arrow) (100).

extension processes. Some hAECs showed spherical but negative for GFAP (Fig. 3e) and NFM (Fig. 3f).
cell bodies with multiple cell processes. The However, after culturing hAECs in neural diffe-
neurogenic-induced hAECs showed positive immu- rentiating medium, the number of cells decreased
noreactivity for NSE (Fig. 3c) and Vimentin (Fig. 3d), significantly.

2010 The Authors

146 Human Cell 2010 Japan Human Cell Society
 hAECs and tissue engineering

Figure 4 Flow cytometry analysis

of human amniotic epithelial cells
(hAECs) at P0. hAECs highly
expressed CD9, CD44, CD73, CD90
and HLA-A,B,C, and negligibly
expressed for CD31, CD34, CD45,
CD117 and HLA-DR,DP,DQ. Data
presented as mean standard error of
the mean (SEM); n = 7.

Mesodermal differentiation
hAECs showed positive staining for Oil Red O in adi-
pogenic induction medium (Fig. 3g) and Alizarin Red in
osteogenic induction medium after 3 weeks in culture
(Fig. 3h).

Immunophenotype analysis
hAECs were positive for the following surface antigen
markers: CD9, CD44, CD73 (Ecto-5-Nucleotidase) and
CD90 (Thy-1). The expression of CD44 and CD90
increased significantly at P1, P2 and P4, while CD73
remained to be highly expressed after serial passages. In
contrast, hematopoietic markers such as CD34, CD45
and CD117 (c-kit); and endothelial markers like CD31
(PECAM-1) were negligibly expressed.
In the aspect of immunology, hAECs were positive for
HLA-A, B, C (MHC I) and these expressions signifi-
cantly increased following serial passages. However,
hAECs showed negligible expression toward HLA-DR,
DP, DQ (MHC II) (Fig. 4).

Epithelial gene expression analysis

CK3, CK19 and involucrin expressions were increased
after serial passages. Expression for CK19 was signifi-
cantly higher at P2 and P4 (1.9-fold and 2.7-fold higher,
respectively) when compared with P0. On the other
hand, the gene expressions in CK1, p63 and CK14
decreased after serial passages. There was no difference
in CK18 and integrin b1 expressions from P0 to P4
(Fig. 5).
Immunocytochemistry analysis
Immunocytochemical staining of hAECs displayed posi-
tivity toward both CK14 and CK18 at P0 and P4.
Figure 5 Determination of epithelial gene expression of
human amniotic epithelial cells (hAECs) using quantitative
polymerase chain reaction (PCR) analysis: CK1, CK3, CK14,
CK18, CK19, Involucrin, Integrin b1 and p63. Gene expres-
sion relative to glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) at P1, P2 and P4 were compared with expression
of the same gene at P0 (*P < 0.05). Data presented as mean
SEM, n = 8.

2010 The Authors

Human Cell 2010 Japan Human Cell Society 147
SS Fatimah et al.
However, the percentage of hAECs exhibiting positivity
toward CK14 was reduced at P4. Involucrin staining was
negative at P0 but there was weak positivity at P4 (Fig. 6).
DISCUSSION
Effective isolation and storage for subsequent culturing
of hAECs are crucial to obtain cells for clinical applica-
tion. Maintaining higher cell viability during the trans-
port of the amnion collected from the operation theatre
to a cell culture laboratory is important. Here, we
showed that the amnion tissue can be effectively isolated
and transported in short-term hypothermic storage con-
ditions. The minimal amount of amnion required for
optimal hAECs isolation and expansion was 15 pieces of
148
Figure 6 Human amniotic epithelial
cells (hAECs) display CK14 (a) and
CK18 (b) cytoplasmic immunoreac-
tivity at P0 (1) and P4 (2). However,
CK14 expression reduced in P4.
Involucrin (c) was negative in P0 (1)
but after serial passage at P4 (2), some
of the cells began to exhibit mild cyto-
plasmic immunoreactivity (arrow)
(100).
amnion (2 2 cm). F12:DMEM was the best transport
medium for amnion at 4 C compared with other trans-
port buffers such as PBS and normal saline. F12:DMEM
is also the basal medium for the culture of hAECs that
provides basic nutrient such as glucose and glutamine.
Glutamine and glucose are the major carbon and energy
sources in cell culture media and these nutrients are
required for cell growth and metabolism.19 Further
investigation is needed to determine whether this short-
term hypothermic storage condition has any effects on
the cell integrity as a prerequisite to their function.
The morphological features of hAECs were similar to
other epithelial cells, like keratinocytes20 and respiratory
epithelial cells,21 which display plastic adherent prop-
erty and polygonal shape. Serial passage of hAECs
showed more homogenous epithelial cells features with
2010 The Authors
Human Cell 2010 Japan Human Cell Society

no observable spindle or fibroblastic shape cells. This
suggests that the isolated hAECs were pure and lack
contamination from the mesenchymal cells. After serial
passage, hAECs at P4 demonstrated larger, flattened
shapes with more abundant cytoplasm. In contrast to a
previous study by Bilic et al.,22 they showed that the
hAECs morphology changed towards mesenchymal
phenotype over several passages. However, we believed
hAECs had differentiated and therefore resulted in the
declining of growth rate in serial passages. This is similar
with the previous study that epithelial cells are difficult
to culture and have poor proliferation rates in in vitro
culture.23 We found that hAECs could only be subcul-
tured to P4-P6 in our in vitro culture method. Hence,
optimizing the culture condition is necessary to
promote the growth of hAECs in vitro while maintaining
their stem-like properties.
The hematopoietic and endothelial cell surface
markers such as CD31, CD34, CD45 and CD117 were
negative in hAECs. This showed that there was no con-
tamination of hematopoietic stem cells from cord blood.
hAECs were positive for CD9, CD44 and CD73, which
were common to both epithelial and mesenchymal stem
cells. hAECs were also positive for CD90, which was
found to be expressed in hepatic oval cells. These cells
exhibit properties of bipotent hepatic epithelial progeni-
tor cells.24 CD90 was also reported to be expressed by
the keratinocytes stem cells.25
The low expression of HLA-DR, DP, DQ (MCH II)
from P0 to P4 suggested that HAECs may be immuno-
privileged cells. This finding is supported by previous
studies where HAECs had been shown to have low
immunogenicity10 and secrete immunosuppressive fac-
tors,26 which suppressed the mixed allogeneic and xeno-
geneic lymphocytes reactions.27 However, there was an
increase in HLA-A, B, C (MCH I) expression in serial
passage.
We have previously shown that hAECs expressed
stemness and neurogenic genes.2,3 Here, we further
demonstrated the colony forming and differentiation
abilities of hAECs. This supports the notion that hAECs
may have stem-like characteristics. Our next plan is to
optimize the culture conditions for hAECs to achieve
long term in vitro culture.
CK18 is expressed in simple epithelium16 and during
embryonic development.28 Our study showed that cul-
tured HAECs expressed CK18 at the transcript as well as
protein levels. Integrin b1 and CK19 have been identified
as epithelial stem cell markers.12,13 Integrin b1 could
distinguish the different subpopulations of proliferating
keratinocytes, where integrin b1 was highly expressed in
2010 The Authors
Human Cell 2010 Japan Human Cell Society
hAECs and tissue engineering
epidermal stem cells.12 CK19 was suggested as a marker
for the epidermal stem cells in the hair follicles of the skin
and was noted to be absent in the interfollicular epider-
mis.13 hAECs showed high expression of integrin b1 and
an increasing level of CK19 expression in serial passage.
These results suggest that HAECs could have maintained
their epithelial stem cell properties.
p63 is a nuclear transcription factor that belongs to
the same family as p53, a tumor suppressor gene. Unlike
its homolog, p63 is associated with the initiation of the
epithelial stratification and maintenance of the prolifera-
tive potential of basal keratinocytes in mature epider-
mis.17 It has been identified as a putative epithelial stem
cell marker for both skin and cornea.29 We found that
p63 expression in hAECs declined after serial passage.
p63 might regulate the differentiation of hAECs towards
stratified epithelium.
Involucrin is a soluble protein precursor of the cross-
linked envelope in human stratified squamous epithe-
lium. It is a marker of the transition from proliferation
to differentiation. Involucrin is expressed in the differ-
entiated, superficial cell layer of the skin and ocular
epithelium.30 It is not expressed in the basal layer of
epithelium. In the present study, involucrin expression
increased after serial passage, supported by weak immu-
noreactivity of involucrin protein at P4. This suggests
that hAECs might have differentiated.
CK14 is a marker of stratified epithelium, expressed
by cells at the basal layer. In skin and limbal epithelium,
CK14 expression decreased as the cells differentiated
and migrated from the basal to suprabasal layer.14 We
observed a decrease in CK14 expression in serial
passage, accompanied by a reduction in CK14 immu-
noreactivity. This again suggests that hAECs might have
differentiated.
CK3 is absent from the basal layers of the limbal
epithelium, but is expressed in the mature corneal epi-
thelium.15 CK3 is regarded as a marker for the corneal
epithelial differentiation. In contrast, CK1 is highly
expressed by the suprabasal cells and unique to kerati-
nized epidermis.14 hAECs showed reciprocal trend in
CK1 (decreasing) and CK3 (increasing) expression in
serial passage. This suggests that hAECs might have
higher potential to differentiate into corneal epithelial
cells.
In conclusion, 15 pieces of amniotic epithelium (2
2 cm) is the best amount to obtain high yield of hAECs.
It can be temporarily stored in F12/DMEM at 4 C.
hAECs are positive for CD9, CD44, CD73 and CD90
surface markers. At initial culture, these cells might have
some stem-like properties. However, they differentiated
149
SS Fatimah et al.
after serial passage. Nonetheless, hAECs have epithelial
stem cell characteristics and have the potential to differ-
entiate into corneal epithelial cells.
ACKNOWLEDGMENTS
This study was funded by the Ministry of Science,
Technology and Innovations (02-01-02-SF0289) and
Universiti Kebangsaan Malaysia (FF-272-2007; FF-
051-2009). We would like to thank Mohd Harisfazal
Mohd Basri, Mazne Mahasin and Wirda Indah Farouk
for their technical expertise in immunocytochemistry.
REFERENCES
1 Parolini O, Alviano F, Bagnara GP et al. Concise review:
isolation and characterization of cells from human term
placenta: outcome of the first international workshop on
placenta derived stem cells. Stem Cells 2008; 26: 300
11.
2 Simat SF, Chua KH, Abdul Rahman H, Tan AE, Tan GC.
The stemness gene expression of cultured human amni-
otic epithelial cells in serial passages. Med J Malaysia
2008; 63 (A): 534.
3 Tan GC, Simat SF, Abdul Rahman H, Tan AE, Chua KH.
Quantitative RT-PCR approach to evaluate the neuro-
genic and gliagenic gene expression of cultured human
amniotic epithelial cells. Med J Malaysia 2008; 63 (A):
512.
4 Miki T, Lehmann T, Cai H, Stolz DB, Strom SC. Stem cell
characteristics of amniotic epithelial cells. Stem Cells
2005; 23: 154959.
5 Portmann-Lanz CB, Schoeberlein A, Huber A et al. Pla-
cental mesenchymal stem cells as potential autologous
graft for pre- and perinatal neuroregeneration. Am J
Obstet Gynecol 2006; 194: 66473.
6 Hou Y, Huang Q, Liu T, Guo L. Human amniotic epi-
thelial cells can be induced to differentiate into func-
tional insulin-producing cells. Acta Biochim Biophys Sin
(Shanghai) 2008; 40: 8309.
7 Ito A, Takizawa Y, Masashige S et al. Proliferation
and stratification of keratinocyte on cultured amniotic
epithelial cells for tissue engineering. J Biosci Bioeng
2003; 95: 58993.
8 Ravishanker R, Bath AS, Roy R. Amnion bank- the
use of long term glycerol preserved amniotic mem-
branes in the management of superficial and super-
ficial partial thickness burns. Burns 2003; 29:
36974.
9 Koizumi N, Inatomi T, Suzuki T, Sotozono C, Kinoshita
S. Cultivated corneal epithelial stem cell transplantation
in ocular surface disorders. Ophthalmology 2001; 108:
156974.
150
10 Akle CA, Adinolfi M, Welsh KI, Leibowitz S, McColl I.
Immunogenicity of human amniotic epithelial cells
after transplantation into volunteers. Lancet 1981; 2:
10035.
11 Boyce ST, Ham RG. Calcium-regulated differentiation of
normal human epidermal keratinocytes in chemically
defined clonal culture and serum-free serial culture. J
Invest Dermatol 1983; 81: 33s40s.
12 Jones PH, Watt FM. Separation of human epidermal
stem cells from transit amplifying cells on the basis of
differences in integrin function and expression. Cell
1993; 73: 71324.
13 Michel M, Trk N, Godbout MJ et al. Keratin 19 as a
biochemical marker of skin stem cells in vivo and in vitro:
keratin 19 expressing cells are differentially localized in
function of anatomic sites, and their number varies with
donor age and culture stage. J Cell Sci 1996; 109: 1017
28.
14 Sun TT, Eichner R, Nelson WG et al. Keratin
classes: molecular markers for different types of epithe-
lial differentiation. J Invest Dermatol 1983; 81: 109s
115s.
15 Schltzer-Schrehardt U, Kruse FE. Identification and
characterization of limbal stem cells. Exp Eye Res 2005;
81: 24764.
16 Debus E, Weber K, Osborn M. Monoclonal cytokeratin
antibodies that distinguish simple from stratified squa-
mous epithelia: characterization on human tissues.
EMBO J 1982; 1: 16417.
17 Koster MI, Kim S, Mills AA, DeMayo FJ, Roop DR.
p63 is the molecular switch for initiation of an
epithelial stratification program. Genes Dev 2004; 18:
12631.
18 Watt FM. Involucrin and other markers of keratinocyte
terminal differentiation. J Invest Dermatol 1983; 81:
100s103s.
19 Miller WM, Blanch HW, Wilke CR. A kinetic analysis of
hybridoma growth and metabolism in batch and con-
tinuous suspension culture: effect of nutrient concentra-
tion, dilution rate, and pH. Biotechnol Bioeng 1988; 32:
94765.
20 Mazlyzam AL, Aminuddin BS, Fuzina NH et al. Recon-
struction of living bilayer human skin equivalent
utilizing human fibrin as a scaffold. Burns 2007; 33:
35563.
21 Noruddin NA, Saim AB, Chua KH, Idrus R. Human
nasal turbinates as a viable source of respiratory epithe-
lial cells using co-culture system versus dispase-
dissociation technique. Laryngoscope 2007; 117:
213945.
22 Bilic G, Zeisberger SM, Mallik AS, Zimmermann R, Zisch
AH. Comparative characterization of cultured human
term amnion epithelial and mesenchymal stromal cells
for application in cell therapy. Cell Transplant 2008; 17:
95568.
2010 The Authors
Human Cell 2010 Japan Human Cell Society

23 Terada S, Matsuura K, Enosawa S et al. Inducing
Proliferation of human amniotic epithelial (HAE)
cells for cell therapy. Cell Transplant 2000; 9: 701
4.
24 Oertel M, Menthena A, Chen YQ, Shafritz DA. Compari-
son of hepatic properties and transplantation of Thy-1+
and Thy-1- cells isolated from embryonic day 14 rat fetal
liver. Hepatology 2007; 46: 123645.
25 Nakamura Y, Muguruma Y, Yahata Y et al. Expression of
CD90 on keratinocyte stem/progenitor cells. Br J Derma-
tol 2006; 154: 106270.
26 Li H, Niederkorn JY, Neelam S et al. Immuno-
suppressive factors secreted by human amniotic epithe-
lial cells. Invest Ophthalmol Vis Sci 2005; 46: 9007.
2010 The Authors
Human Cell 2010 Japan Human Cell Society
hAECs and tissue engineering
27 Bailo M, Soncini M, Vertua E et al. Engraftment poten-
tial of human amnion and chorion cells derived
from term placenta. Transplantation 2004; 78: 1439
48.
28 Waseem A, Alexander CM, Steel JB, Lane EB. Embryonic
simple epithelial keratins 8 and 18: chromosomal loca-
tion emphasizes difference from other keratin pairs. New
Biol 1990; 2: 46478.
29 Pelligrini G, Dellambra E, Golisano O et al. p63 identifies
keratinocyte stem cells. Proct Natl Acad Sci USA 2001;
98: 315661.
30 Chen Z, de Paiva CS, Luo L, Kretzer FL, Pflugfelder SC,
Li DQ. Characterization of putative stem cell phenotype
in human limbal epithelia. Stem Cells 2004; 22: 35566.
151