Cell Tissue Res (2007) 327:529537

DOI 10.1007/s00441-006-0340-z

REGULAR ARTICLE

Fibronectin and laminin induce expression of islet cell

markers in hepatic oval cells in culture
Adriana Ribeiro Leite & Maria Lcia Corra-Giannella &
Maria Lcia Zaidan Dagli &
Maria Angela Zanela Fortes &
Vanina Monique Tucci Vegas & Daniel Giannella-Neto

Received: 4 April 2006 / Accepted: 31 August 2006 / Published online: 6 December 2006
# Springer-Verlag 2006

Abstract Hepatic oval cells (OC) are considered to be

facultative liver stem cells and, because they may undergo
differentiation into a variety of cell lineages, they might
have the potential to be used in cellular therapy. Signals
delivered by extracellular matrix (ECM) proteins take part
in cellular differentiation in cooperation with signals from
growth factors; indeed, some ECM proteins, such as
laminin (LAM) and fibronectin (FN), have been shown to
contribute to -cell differentiation and islet development,
respectively. Since no previous studies have investigated
the effect of ECM proteins on the expression of islet cell
markers by cultured OC, the purpose of the present study
was to evaluate whether FN and LAM modulate the
expression of genes related to the endocrine pancreas in
these liver cells. OC proliferation was induced in Wistar
rats by prolonged treatment with 2-acetoaminofluorene/
allyl alcohol and confirmed by reverse transcription/
polymerase chain reaction and hepatic immunocytochemical and histopathological analyses. OC isolation was

This study was supported by FAPESP (Process number 02/13098-4),

So Paulo, Brazil.
A. R. Leite : M. L. Corra-Giannella : M. A. Z. Fortes :
V. M. T. Vegas : D. Giannella-Neto (*)
Laboratory for Cellular and Molecular Endocrinology (LIM-25),
University of So Paulo School of Medicine,
Av. Dr. Arnaldo, 455 s/4307,
01246-903 So Paulo, Brazil
e-mail: dgiannella@terra.com.br
M. L. Z. Dagli
Institute of Veterinary Medicine, University of So Paulo,
So Paulo, Brazil

performed by Ficoll gradient and magnetic-activated cell

sorting. OC were cultured for 1 and 2 months under several
conditions with specific growth factors, over a FN or LAM
substrate or in high glucose, nicotinamide and fetal calf
serum. OC cultured on FN substrate expressed Pdx-1, Pax-6,
insulin 2 and glucagon. LAM also induced the expression
of Pdx-1, insulin 1 and insulin 2, glucagon, somatostatin
and GLUT-2. Our results suggest that these ECM
proteins can be used in protocols of OC transdifferentiation aimed at reducing the period necessary for complete
transdifferentiation.
Keywords Hepatic oval cell . Differentiation . Fibronectin .
Laminin . Endocrine pancreas . Rat (Wistar, male)

Introduction
Islets of Langerhans are composed of four main types of
endocrine cells that synthesize insulin (-cells), glucagon
( cells), somatostatin ( cells) and pancreatic polypeptide
(Pictet and Rutter 1972). All four types of endocrine cell
are believed to originate from a common multipotent
precursor that expresses Pdx-1 (Guz et al. 1995), a
transcription factor required for pancreas development
(Jonsson et al. 1994; Offield et al. 1996) and normally
expressed in mature -cells, being considered a key
molecular marker of this cell type (Edlund 2002). With
regard to some transcription factor genes expressed more
downstream in the developmental cascade, Pax-6 is
required in the differentiation of all four endocrine-cell
types (Sander et al. 1997), Pax-4 is involved in the

530

differentiation of - and -cells (Sosa-Pineda et al. 1997),

whereas Nkx-6.1 is involved in -cell differentiation.
The exact nature of the pancreatic stem cell is still not
well defined (Bern et al. 2001) and neither are the factors
that regulate lineage differentiation of islets cells. None of
the in vitro strategies used so far have been able to promote
the complete differentiation of pancreatic ductal cells or
embryonic stem cells to a -cell phenotype. Other cell
types have therefore been considered as potential alternative sources of insulin-secreting cells to approach cellular
therapy of type 1 diabetes mellitus.
Because of the close relationship between liver and
pancreas during embryogenesis, these two organs are
proposed to be derived from a common progenitor cell
(Deutsch et al. 2001) and hepatic cells have been
experimentally transdifferentiated into endocrine-like cells
(Ferber et al. 2000; Horb et al. 2003; Ber et al. 2003;
Kojima et al. 2003). Hepatic oval cells (OC) are thought to
be facultative liver stem cells that arise as a result of severe
hepatic injury (Fausto et al. 1993; Petersen et al. 1998a,b)
and, because they may undergo differentiation into a variety
of cell lineages (Crosby et al. 2001; Qin et al. 2004), they
might have the potential to be used in cellular therapy. Yang
et al. (2002) have demonstrated that highly purified rat OC
can transdifferentiate into pancreatic endocrine hormoneproducing cells when maintained for more than 8 months in
culture (6 months in the presence of specific growth factors,
2 months in high glucose media and 7 days in the presence
of nicotinamide).
Signals delivered by extracellular matrix (ECM) proteins
take part in cellular differentiation in cooperation with
growth factors and hormones (Lelievre et al. 1996) and
some ECM proteins, such as laminin (LAM; Schmied et al.
2000) and fibronectin (FN; Cirulli et al. 2000) have been
shown to participate in the differentiation of -cells and in
islet development, respectively. Since no previous studies
have investigated the effect of ECM proteins on the
expression of islet cell markers by cultured OC, the purpose
of the present study was to evaluate whether LAM and FN
modulate the expression of such genes in these cells. We
also describe an alternative method of OC isolation based
on the purification of hepatic non-parenchymal cells (NPC)
by density gradient centrifugation in Ficoll followed by
magnetic-activated cell sorting (MACS) with the Thy1.1
antibody conjugated to magnetic beads.

Materials and methods

Animal experiments
Male Wistar rats (250 g body weight) were obtained from
the specific pathogen-free biotery, University of So

Cell Tissue Res (2007) 327:529537

Paulo School of Medicine. The rats were fed standard

laboratory chow ad libitum. The institutional animal care
committee had previously approved all animal care and
procedures.
Reagents
Allyl alcohol (AA) and 2-acetoaminofluorene (2-AAF) were
purchased from Sigma (St. Louis, MO, USA). For liver
digestion and further isolation steps, collagenase type IV was
purchased from Invitrogen (Carlsbad, CA, USA) and Ficoll
reagent (Sigma) and Thy1.1 Dynalbeads (Dynal, Wirral, UK)
were used together with a magnetic particle concentrator
(MPC; Dynal). Dulbeccos Iscoves medium (DMEM) for
cell culture was purchased from Invitrogen and was
supplemented with fetal calf serum (FCS; Cultilab, Campinas, So Paulo, Brazil) and with leukemia inhibitory factor
(LIF), interleukin 3 (IL-3), stem cell factor, rat recombinant
ligand -3 (Flt-3 ligand), nicotinamide and grown-over rat FN
or rat LAM (type V), all from Sigma. For hepatic
histopathological analysis, methacarn fixative was used
before paraffin embedding. Methanol, clorophorm and
glacial acetic acid (for methacarn) were purchased from
Sigma. Hepatic immunocytochemical analysis involved the
use of monoclonal mouse anti-human cytokeratin 19 (CK19)
clone AE1-AE3 from DAKO (Carpinteria, CA, USA) and
the detection of the primary antibody was performed with the
LSAB system - HRP (DAKO). Paraformaldehyde used in the
immunocytochemistry of enriched populations of OC was
purchased from Sigma. The primary antibody, anti-mouse
alfa-fetoprotein (AFP), was purchased from DAKO (Hamburg, Germany) and the secondary antisera biotinylated antirabbit IgG and peroxidase-conjugated streptavidin were from
Maxim Biotech (San Francisco, CA, USA). Fluorescent
mounting medium was obtained from Kirkegaard and Perry
Laboratories (Gaithersburg, MD, USA). Trizol reagent used
for total RNA isolation and SuperScript II Reverse Transcriptase and random primers used for cDNA synthesis were
purchased from Invitrogen.
Induction and characterization of oval cell proliferation
Pellets of 2-AAF (25 mg) were inserted into rats by
intraperitoneal injection for 2, 4, 7 and 9 days to stimulate
the expansion of OC (Petersen et al. 1998a,b). At 2 h before
sacrifice, a single dose of 1 ml/kg body weight of AA was
administered by intraperitoneal injection to induce final
hepatic injury, according to Petersen et al. (1998a,b) with
modifications. In order to characterize the time-dependent
increase on OC proliferation after 2-AAF, total RNA was
extracted from liver samples of treated and untreated
animals by using Trizol reagent according to the manufacturers protocol. Total RNA (5 g) was used for the

Cell Tissue Res (2007) 327:529537

531

synthesis of first-strand cDNA (employing SuperScript II

Reverse Transcriptase and random primers according to
manufacturers recommendations) and subsequent PCR
amplification with specific primers for Thy1 and -actin
(Actb). The primers (shown in Table 1) were designed with
the primer3_www.cgi v 0.2 program (Rozen and Skaletsky
2000; Table 1).
Livers from control and treated animals were removed,
sliced (45 mm thickness) and fixed for 8 h in methacarn
(60% methanol [vol:vol], 30% clorophorm and 10% glacial
acetic acid). After routine processing, the liver slices were
embedded in paraffin wax and sectioned at 5 m. Histopathological examinations were performed on liver sections
stained with haematoxylin-eosin (H&E).
The 5 m thick histological sections were positioned on
silanized slides, deparaffinized, re-hydrated, treated with a
solution of methanol and hydrogen peroxide (10%) for
30 min and incubated with monoclonal mouse anti-human
CK19. The detection of the primary antibody was
performed with the LSAB system - HRP. Anti-mouse/antirabbit biotinylated antibody (1:100 in phosphate-buffered
saline; PBS) was added, as the link, for 1 h in a humidified
chamber. Amplification of the signal was carried out with
horseradish peroxidase linked to streptavidin and biotin
(1:100 in PBS). The reaction was detected with a solution
containing diaminobenzidine and hydrogen peroxide. The
sections were counterstained with haematoxylin. Negative
controls were obtained, on slides, by suppressing the
primary antibody.
Oval cell isolation

sacrifice, were minced in cold PBS. Fragments digestion

were performed with 5 ml PBS containing 15 mg
collagenase type IV (for 25 min at 37C in a periodic
shaking bath) and interrupted by addition of 5% pure cold
FCS to the cell suspension. After centrifugation at 50 g for
5 min at 4C, the pellet containing the majority of
hepatocytes was discarded and the supernatant of NPC
(the fraction known to contain OC) was collected and recentrifuged at 350 g for 10 min at 4C. The pellet was
suspended in 2 ml DMEM and a first purification was
performed by Ficoll density gradient, as follows: 5 ml
1.20 g/ml Ficoll, 5 ml 1.12 g/ml Ficoll, cell suspension,
2 ml DMEM. After centrifugation at 350 g for 20 min at
4C, the interface between the 1.12 and 1.20 g/ml Ficoll
was decanted and re-centrifuged at 350 g for 5 min at 4C.
The pellet was resuspended in 5 ml DMEM and the cells
were counted. Approximately, 6.0 108 cells/rat were
obtained.
A volume of 25 l containing 1107 Thy 1.1 magnetic
Dynalbeads was pre-washed three times in PBS plus 0.1%
bovine serum albumin (BSA) before use. Aliquots of 1
107 cells were then incubated on ice for 20 min with 25 l
Thy1.1 Dynalbeads. The tubes containing cells were
coupled to an MPC for 23 min and the volume containing
unbound cells was removed. Positive cells attached to MPC
were collected by addition of 1.0 ml DMEM. Approximately 50% of the cells are recovered after magnetic sorting
(>95% viable by trypan-blue exclusion). When a large
number of cells is needed, the use of all cells obtained after
Ficoll gradient for MACS required a volume of approximately 1,500 l Dynalbeads and yielded around 3.0108
cells/rat.

Liver tissues, harvested from rats that were treated for

9 days with 2-AAF and with a single dose of AA 2 h before
Table 1 Primers used for RT-PCR
Accession number

Gene

Forward primers

Reverse primers

Product length, bp

NM_012673
NM_022264
NM_199498
NM_012493
MN_25823
NM_031691
NM_019130
NM_019129
NM_012707
NM_031693
NM_013001
NM_012659
NM_012879
NM_031799
AF_004431
NM_007393

Thy1
c-kit
CK19
AFP
CD45
CD11c
Insulin 2
Insulin 1
Glucagon
Pdx-1
Pax-6
Somatostatin
GLUT-2
Pax-4
Nkx-6.1
Actb

AGACCCAGGACGGAGCTATT
AGCAAGAGTTAACGATTCCGGAG
GAGCTGGAGGTGAAGATTCG
ACCTGACAGGGAAGATGGTG
CTGCCTCAAACTGACCCTTT
GGTGGAGTTGTGATCCTCCT
TGTGGTTCTCACTTGGTGGA
GTACCTGGTGTGTGGGGAAC
GAATTCATTGCTTGGCTGGT
GGTGGCACATTGCTGATAGA
TCCCAGGGATCTGAGAATTG
TCAAGCTCGGCTGTCTGAG
TTCAGATTGCTGGCCTCAG
GACGGTCTCAGCAGTGTGAA
TTCTTCCTCCTCCTCCTCGT
TGAGCGCAAGTACTCTGTGTGG

CAAATATCTCAAAACGCGGG
AGGAGGCACTTACACCTTTCTGG
CAGATGACTTCCGGACCAA
ACTCCGGTATCAACCACTGC
CTGACCTCACCACACTCACG
GACTGTGGTGCAGTTTGGTG
ACCTTCAGACCTTGGCACTG
CTGCTCCCTCTACCAACTGG
CAACTGGCTGATTCAAACCA
CAGAGCACGCAGAAAACATG
TCAGCTTGGTGGTGTCTTTG
AGGTCTGCCAACTCGAACC
GAGGAAGTCAGGGCAAAGAA
GAGTCCTTCGGGCACTACAG
CTGAAAAAGCACAATCCAGC
GAAGTCCCTCACCCTCCCAAA

621
343
185
154
265
143
155
199
172
168
406
307
165
386
660
504

532

Oval cell characterization

In order to characterize the cellular components sorted after
each isolation step, total RNA was isolated from cell
extracts and 5 g was used for the synthesis of first-strand
cDNA and the subsequent Thy1, CD45, CD11c and Actb
amplification (primers shown in Table 1). In the cell
extracts obtained after Thy1.1 magnetic sorting (positive
separation), RT-PCR was also performed for c-kit, AFP
and CK19, to improve the characterization of the freshly
isolated OC. The PCR products were separated by
electrophoresis in a 2% agarose gel and the signals related
to the bands were analysed by laser densitometry
(Amersham Biosciences, Buckinghamshire, UK). Gene
expression data were representative of three independent
experiments presented by the median and semi-interquartile range of arbitrary units of optical density in relation to
the endogenous internal control.
Freshly isolated OC were fixed with 4% paraformaldehyde
in PBS. Cells were blocked with 3% BSA for 30 min at room
temperature and incubated with monoclonal anti-mouse AFP
(1:150 dilution) overnight at 4C. The cells were then
incubated with biotinylated anti-rabbit IgG (1:50 dilution)
for 45 min, followed by peroxidase-conjugated streptavidin
(1:50 dilution). Slides were then washed with PBS and
coverslipped with fluorescent mounting medium. Fluorescent
images were obtained by using a Zeiss epifluorescence
microscope equipped with an Olympus camera.

Cell Tissue Res (2007) 327:529537

were gently rinsed three times with PBS before OC were

allowed to adhere. Under ADJ conditions, the following
adjuvant factors were added to RPMI 1640 medium: 10%
FCS, 23 mM glucose and 10 mM nicotinamide.
After the end of each experiment, total mRNA samples
were analysed by RT-PCR for Thy1, c-Kit, AFP, CK19,
insulin 1, insulin 2, glucagon, Pdx-1, Pax-4, Pax-6, Nkx6.1,
GLUT-2, somatostatin and Actb (primers shown in Table 1).
The PCR products were separated by electrophoresis in 2%
agarose gels.
Statistical analysis
All statistical comparisons were performed by the KruskalWalis non-parametric test supplemented by the Sheff test.
Statistical significance was fixed at P<0.05.

Oval cell cultures

Purified Thy1-positive OC were cultured under 10 different
experimental conditions: (1) for 30 days only in serum-free
DMEM medium (CTRL 30 D); (2) for 60 days only in
serum-free DMEM medium (CTRL 60 D); (3) in DMEM
supplemented with growth factors for 30 days (GF 30 D); (4)
in DMEM supplemented with growth factors for 60 days (GF
60 D); (5) for 30 days in DMEM supplemented with growth
factors and for an additional 30 days in RMPI 1640 with
adjuvant factors (GF 30 DADJ 30 D); (6) for 30 days in
RMPI supplemented with adjuvant factors (ADJ 30 D); (7)
on a FN substrate for 30 days in serum-free DMEM (FN 30
D); 8) on a LAM substrate for 30 days in serum-free DMEM
(LAM 30 D); 9) on a FN substrate in DMEM supplemented
with growth factors for 30 days (FN+GF 30 D); (10) on a
LAM substrate in DMEM supplemented with growth factors
for 30 days (LAM+GF 30D).
Under GF conditions, the following growth factors were
added to serum-free DMEM: LIF (10 ng/ml), IL-3 (10 ng/
ml), SCF(10 ng/ml), and Flt-3 (10 ng/ml). Under FN or
LAM conditions, 75-cm2 culture flasks were coated with
FN or LAM (type V). After incubation of the FN or LAM
solution (12 g/ml) for 1 h at room temperature, the flasks

Fig. 1 Representative agarose gel from a RT-PCR analysis showing

the time-dependent increase in Thy1 expression following 2-AAF
treatment for 2, 4, 7 and 9 days. The data from each experimental
condition is shown as abitrary units of optical density (A.U.) and is
presented as median and semi-interquartile ranges. The result shown is
representative of three independent experiments

Cell Tissue Res (2007) 327:529537

533

Results and discussion

Induction and characterization of oval cell proliferation
Figure 1 depicts the time-dependent increase of Thy1 RNA
expression in liver samples from animals treated with 2-AAF
for 2, 4, 7 and 9 days and suggests a steady increase of OC
proliferation after prolonged treatment. The histopathological changes in liver sections from rats exposed to 2-AAF for
9 days, followed by a single dose of AA 2 h before sacrifice
are shown in Fig. 2. Proliferating OC occurred as spindlelike cells radiating from the periportal region, in close
proximity to proliferating bile ducts and in areas of ductular
proliferation (Fig. 2b). Small oval cells stained positively
for CK19 (Fig. 2d, arrows), an oval cell marker (Weiss and
Strick-Marchand 2003). Neither OC proliferation nor CK19
staining was detected in the liver of untreated animals
(Fig. 2a,c).
Oval cell isolation and characterization
To confirm that the proposed methodology would result in
an enriched population of OC versus other NPC, especially haematopoietic cells (HC), which are known to
express Thy1 and likely to be present in the same Ficoll
fraction as OC, mRNA expression for Thy1, CD45 and
CD11cs was evaluated by RT-PCR in all cell extracts;
CD45 expression demonstrates possible HC contamination in OC preparations (Wang et al. 2003), whereas
CD11c is expressed on monocytes, granulocytes, dendritic
Fig. 2 Histopathological (H&E)
and immunohistochemical (for
CK19) analyses of rat liver
sections obtained from 2-AAF/
AA-treated (b, d) and control
(a, c) animals. Small oval cells
(arrows) were stained positively
for CK19 (d) and lay in close
proximity to proliferating bile
ducts and in areas of ductular
proliferation radiating from the
periportal region. No OC proliferation nor CK19 staining was
detected under control conditions (a, c). 20

Fig. 3 Representative agarose gel from an RT-PCR analysis of Thy1,

CD45 and CD11c mRNA expressions in cell extracts obtained after
Thy1 magnetic bead isolation (positive separation) and in flowthrough (negative separation) cell extracts. The data from each
experimental group are shown as arbitrary units of optical density
(A.U.) and are presented as median and semi-interquartile ranges
(*P<0.05)

cells and activated B lymphocytes (Lopez-Cabrera et al.

1993). As shown in Fig. 3, in the cell extracts obtained
after Thy1.1 magnetic sorting (positive separation), Thy1
mRNA expression was confirmed [median and range:
0.546 (0.2830.665)] and low levels of CD45 mRNA
expression [0.053 (0.0380.098)] and of CD11c mRNA
expression [0.053 (0.0360.073)] were detected, whereas

534

cell extracts that did not bind to the beads (negative

separation) presented higher levels of CD45 [0.424
(0.2250.571)] and CD11c [0.118 (0.1030.120)] mRNA
expression in comparison with that for Thy1 [0.029
(0.0240.146)] whose expression was barely detected. A
statistically significant difference was observed between
positive and negative separations for the three mRNA
studied (P<0.05). These results suggest that this methodology is highly specific for Thy1-positive cells.
Data presented in Fig. 4 corroborate the identity of these
cells as OC since the known OC markers c-kit, AFP and
CK19 (Weiss and Strick-Marchand 2003; Bisgaard et al.
1994) were detected by RT-PCR (Panel A) and positive
immunostaining for AFP was observed in freshly isolated
cells (Panel B).
The main features of the present isolation method are its
simplicity and efficiency, allowing the obtention of
enriched OC preparations, as shown by the low mRNA
expression of the HC markers. The combination of (1) the 2AAF/AA treatment providing OC proliferation without
hepatectomy, a procedure commonly used to induce OC
proliferation and which requires special technical expertise,
(2) a simpler one-step collagenase digestion, instead of the
two-step digestion described by other authors (Petersen et al.
1998a,b; Golding et al. 1995) and (3) OC isolation by
MACS, with no need of expensive equipment, make this
reproducible protocol an economical alternative to other
established methods based on fluorescence-activated cell
sorting.
Oval cell cultures
mRNA expression under the various experimental conditions is demonstrated in Fig. 5. After 30 and 60 days of
culture in the presence of serum-free media (CTRL 30 D
and CTRL 60 D, respectively), expression of the OC
markers Thy1, c-Kit, AFP and CK19 mRNA was detected,
whereas none of the islet cell markers was present,
suggesting the maintenance of the OC phenotype in the
Fig. 4 Representative agarose
gel from a RT-PCR analysis of
Thy1, c-kit, AFP and CK19 in
freshly isolated OC (a). Immunocytochemical staining for
AFP demonstrating cells coloured in light red (b)

Cell Tissue Res (2007) 327:529537

absence of growth factors and ECM proteins. When OC

were cultured in the presence of growth factors for 30 days
(GF 30 D), no expression of Thy1 was detected, but the
cells maintained the expression of the other OC markers
c-Kit, AFP and CK19 simultaneously with insulin 2
expression, suggesting the beginning of the transdifferentiation process. This process seemed to progress after
culturing for 60 days in the presence of growth factors
(GF 60 D), when the expression of insulin 2, insulin 1,
glucagon, Pdx-1 and Pax-6 was observed in the absence of
c-kit and AFP expression. Interestingly, CK19 mRNA
expression was maintained under all experimental conditions, including those in which diverse islet cell markers
were detected. Ductal cells express CK19 and, under some
experimental conditions, co-express islet cell markers
(Bonner-Weir et al. 2000). Because ductal cells have been
considered as one of the main candidates for pancreatic
progenitor/stem cells (Bonner-Weir and Sharma 2002), the
finding of CK19 expression under culture conditions
characterized by the expression of islet cell markers in the
absence of other OC markers does not seem to be related to
the maintenance of an OC phenotype.
OC cultured in the presence of adjuvant factors (FCS, high
glucose and nicotinamide) for 30 days showed insulin 2 and
glucagon mRNA expression. However, 30 days of exposure
to growth factors followed by 30 days of exposure to the
adjuvant factors nicotinamide and high glucose, which are
known as inducers of -cell differentiation (Otonkoski et al.
1993; Sjoholm et al. 1990), stimulated insulin 1 and Pdx-1
mRNA expression in addition to insulin 2 and glucagon. The
presence of FN and LAM induced the expression of markers
of endocrine differentiation in the absence of growth factors
and adjuvant factors after 30 days of culture (FN 30D and
LAM 30 D). FN induced the expression of insulin 2,
glucagon, Pdx-1 and Pax-6, whereas LAM stimulated the
expression of insulin 1, insulin 2, glucagon, Pdx-1, somatostatin and GLUT-2. The addition of growth factors to OC
grown on a FN substrate inhibited the expression of Pax-6,
whereas the expression profile of OC cultured in the

Cell Tissue Res (2007) 327:529537

535

Fig. 5 a Representation of the

various OC culture conditions
(CTRL 30 D OC cultured for
30 days only in serum-free
DMEM medium, CTRL 60 D
OC cultured for 60 days only in
serum-free DMEM medium, GF
30 D OC cultured in DMEM
supplemented with growth factors for 30 days, GF 60 D OC
cultured in DMEM supplemented with growth factors for
60 days, GF 30 DADJ 30 D
OC cultured for 30 days in
DMEM supplemented with
growth factors and for additional
30 days in RMPI 1640 plus
adjuvant factors, ADJ 30 D OC
cultured for 30 days in RMPI
supplemented with adjuvant
factors, FN 30 D OC cultured on
a FN substrate for 30 days in
serum-free DMEM, LAM 30 D
OC cultured on a LAM substrate
for 30 days in serum-free
DMEM, FN+GF 30 D OC
cultured on a FN substrate in
DMEM supplemented with
growth factors for 30 days,
LAM+GF 30D OC cultured on a
LAM substrate in DMEM supplemented with growth factors
for 30 days. b Representative
agarose gel from RT-PCR
analyses of Thy1, c-kit, AFP,
CK19, insulin 2, insulin 1,
glucagon, Pdx-1, Pax-6,
somatostatin, GLUT-2, Pax-4,
Nkx-6.1 and Actb mRNA
expression in OC cultured under
the various experimental
conditions

presence of LAM and growth factors was not different from

that observed in OC grown only in the presence of LAM.
Except under control conditions, insulin 2 expression
was detected in all other experimental groups. As

suggested by Edlund (2002), expression of the insulin gene

in the absence of Pdx-1 expression, which is considered the
key -cell transcription factor, probably does not reflect the
capability for the production and secretion of insulin,

536

according to previous observation on cells of neural origin

(Lumelsky et al. 2000). Interestingly, the insulin 1 gene was
expressed only under conditions when the Pdx-1 gene was
also detected. In mice, insulin transcripts found in extrapancreatic tissues are derived from the insulin 2 gene,
indicating that the two insulin genes are regulated differently in extra-pancreatic locations (Lamotte et al. 2004).
This concept may apply to OC and, for this cell type,
insulin 1 gene expression might be a more specific marker
than insulin 2 for endocrine cell transdifferentiation.
Yang et al. (2002) have shown that the transdifferentiation of OC depends on two factors: the removal of LIF (a
factor known to inhibit stem cell differentiation) and a
concomitant increase in the concentration of glucose. Under
the experimental conditions employed in this study, with
shorter durations of treatment in relation to the experiments
performed by Yang et al. (2002), we have not been able to
achieve OC transdifferentiation to endocrine pancreatic
cells; however, some degree of differentiation in the
direction to endocrine cells has been observed, even in the
presence of LIF, as observed under GFADJ 30D, GF
60D, FN+GF 30D and LAM+GF 30D conditions when
Pdx-1 gene expression is detected concomitantly with the
expression of one or both insulin genes.
The culture of OC on FN and LAM substrates induces
the expression of several endocrine markers, even after
30 days of culture in the absence of adjuvant factors.
Unlike FN, LAM induces the expression of somatostatin, a
marker of -cells, and of GLUT-2, a glucose transporter
expressed in the liver, small intestine, kidney and pancreatic
-cells (Ashizawa et al. 2004). Additional studies are
necessary to clarify whether the presence of LAM increases
the differentiation of OC to a - or -cell precursor.
Although the absence of expression of other transcription
factors required for -cell differentiation, such as Pax-4 and
Nkx-6.1, demonstrates that the experimental conditions
tested do not promote the full differentiation to a endocrine
phenotype, the results obtained with FN and LAM suggest
that these ECM proteins can be used successfully in
protocols of OC transdifferentiations aimed at reducing the
period necessary for complete transdifferentiation.
Acknowledgments We are greatful to Vernica Gonalvez and
Edith Amaral for performing the immunocytochemical analysis.

References
Ashizawa S, Brunicardi C, Wang XP (2004) Pancreas 28:109120
Ber I, Shternhall K, Perl S, Ohanuna Z, Goldberg I, Barshack I,
Benvenisti-Zarum L, Meivar-Levy I, Ferber S (2003) Functional,
persistent, and extended liver to pancreas trans-differentiation.
J Biol Chem 278:3195031957

Cell Tissue Res (2007) 327:529537

Bern G, Lon-Quinto T, Ensent-Waser R, Montanya E, Martn F,
Soria B (2001) Stem cells and diabetes. Biomed Pharmacother
55:206212
Bisgaard HC, Nagy P, Ton PT, Hu Z, Thorgeirsson SS (1994)
Modulation of keratin 14 and alpha-fetoprotein expression during
hepatic oval cell proliferation and liver regeneration. J Cell
Physiol 159:475484
Bonner-Weir S, Sharma A (2002) Pancreatic stem cells. J Pathol
197:519526
Bonner-Weir S, Taneja M, Weir G, Tatarkiewicz K, Song K,
Sharma A, ONeil J (2000) In vitro cultivation of human islets
from expanded ductal tissue. Proc Natl Acad Sci USA
97:79998004
Cirulli V, Beattie GM, Klier G, Ellisman M, Ricordi C, Quaranta V,
Frasier F, Ishii JK, Hayek A, Salomon DR (2000) Expression and
function of alpha(v)beta(3) and alpha(v)beta(5) integrins in the
developing pancreas: roles in the adhesion and migration of
putative endocrine progenitor cells. J Cell Biol 18 150:14451460
Crosby HA, Kelly DA, Strain AJ (2001) Human hepatic stem like
cells isolated using c-kit or CD34 can differentiate into biliary
epithelium. Gastroenterology 120:534544
Deutsch G, Jung J, Zheng M, Lora J, Zaret KS (2001) A bipotential
precursor population for pancreas and liver within the embryonic
endoderm. Development 128:871881
Edlund H (2002) Pancreatic organogenesisdevelopmental mechanisms and implications for therapy. Nat Rev Genet 3:524532
Fausto N, Lemire JM, Shiojiri N (1993) Cell lineages in hepatic
development and the identification of progenitor cells in normal
and injured liver. Proc Soc Exp Biol Med 204:237241
Ferber S, Halkin A, Cohen H, Ber I, Einav Y, Goldberg I, Barshack I,
Seijffers R, Kopolovic J, Kaiser N, Karasik A (2000) Pancreatic and
duodenal homeobox gene 1 induces expression of insulin genes in
liver and ameliorates streptozotocin-induced hyperglycemia. Nat
Med 6:568572
Golding M, Sarraf CE, Lalani EN, Anilkumar TV, Edwards RJ, Nagy
P, Thorgeirsson SS, Alison MR (1995) Oval cell differentiation
into hepatocytes in the acetylaminofluorene treated regenerating
rat liver. Hepatology 22:12431253
Guz Y, Montminy MR, Stein R, Leonard J, Gamer LW, Wright CVE,
Teitelman G (1995) Expression of murine STF-1, a putative
insulin gene transcription factor, in cells of pancreas, duodenal
epithelium and pancreatic exocrine and endocrine progenitors
during ontogeny. Development 121:1118
Horb ME, Shen CN, Tosh D, Slack JM (2003) Experimental
conversion of liver to pancreas. Curr Biol 13:105115
Jonsson J, Carlsson L, Edlund T, Edlund H (1994) Insulin-promoterfactor 1 is required for pancreas development in mice. Nature
371:606609
Kojima H, Fujimiya M, Matsumura K, Younan P, Imaeda H, Maeda
M, Chan L (2003) NeuroD-betacellulin gene therapy induces
islet neogenesis in the liver and reverses diabetes in mice. Nat
Med 9:596603
Lamotte L, Jackerott M, Bucchini D, Jami J, Joshi RL, Deltour L
(2004) Knock-in of diphteria toxin A chain gene at Ins2 locus:
effects on islet development and localization of Ins2 expression
in the brain.Transgenic Res 13:463473
Lelievre S, Weaver VM, Bissel M (1996) Extracellular matrix signalling
from cellular membrane skeleton to the nuclear skeleton: a model
of gene regulation. Recent Prog Horm Res 54:417432
Lopez-Cabrera M, Nueda A, Vara A, Garcia-Aguilar J, Tugores A,
Corbi A (1993) Characterization of the p150,95 leukocyte
integrin a subunit (CD1 IC) gene promoter. J Biol Chem
268:11871193
Lumelsky N et al (2000) Differentiation of embryonic stem cells to
insulin-secreting structures similar to pancreatic islets. Science
292:13891394

Cell Tissue Res (2007) 327:529537

Offield MF, Jetton TL, Labosky PA, Ray M, Stein RW, Magnuson
MA, Hogan BLM, Wright CVE (1996) PDX-1 is required for
pancreatic outgrowth and differentiation of the rostal duodenum.
Development 122:983995
Otonkoski T, Beattie GM, Mally MI, Ricordi C, Hayek A (1993)
Nicotinamide is a potent inducer of endocrine differentiation in
cultured human fetal pancreatic cells. J Clin Invest 92:14591466
Petersen BE, Goff JP, Greenberger JS, Michalopoulos GK (1998a)
Hepatic oval cells express the hematopoietic stem cell marker
Thy-1 in the rat. Hepatology 27:433445
Petersen BE, Zajac VF, Michalopoulos GK (1998b) Hepatic oval cell
activation in response to injury following chemically induced
periportal or pericentral damage in rats. Hepatology 27:1030
1038
Pictet RL, Rutter WJ (1972) Development of the embryonic pancreas.
In: Steiner DF, Freinkel N (eds) Handbook of physiology, section
7: endocrinology. American Physiological Society, Washington
DC, pp 6776
Qin AL, Zhou XQ, Zhang W, Yu H, Xie Q (2004) Characterization
and enrichment of hepatic progenitor cells in adult rat liver.
World J Gastroenterol 10:14801486
Rozen S, Skaletsky HJ (2000) Primer3 on the WWW for general users
and for biologist programmers. In: Krawetz S, Misener S (ed)

537
Bioinformatics methods and protocols: methods in molecular
biology. Humana, Totowa New York, pp 365386
Sander M et al (1997) Genetic analysis reveals that PAX6 is required
for normal transcription of pancreatic hormone genes and islet
development. Genes Dev 11:16621673
Schmied BM, Ulrich A, Matsuzaki H, Ding X, Ricordi C, Moyer MP,
Batra SK, Adrian TE, Pour PM (2000) Maintenance of human
islets in long-term culture. Differentiation 66:173180
Sjoholm A, Welsh N, Sandler S, Hellerstrom C (1990) Role of
polyamines in mitogenic and secretory responses of pancreatic
beta-cells to growth factors. Am J Physiol 259:C828C833
Sosa-Pineda B, Chowdhury K, Torres M, Oliver G, Gruss P (1997)
The pax4 gene is essential for differentiation of insulin-producing
cells in the mammalian pancreas. Nature 386:399402
Wang X, Foster M, Al-Dhalimy M, Lagasse E, Finegold M, Grompe
M (2003) The origin and liver repopulating capacity of murine
oval cells. Proc Natl Acad Sci USA 100 (Suppl 1):1188111888
Weiss MC, Strick-Marchand H (2003) Isolation and characterization of
mouse hepatic stem cells in vitro. Semin Liver Dis 23:313324
Yang L, Li S, Hatch H, Ahrens K, Cornelius JG, Petersen BE, Peck
AB (2002) In vitro trans-differentiation of adult hepatic stem
cells into pancreatic-endocrine hormone-producing cells. Proc
Natl Acad Sci USA 99:80788083