Dissertation final complete1

'Epigenetic Changes and Pancreatic Cell
Transdifferentiation for Hepatocyte-like Cells'
Patrick James Newton
Liver Research Group, Institute of Cellular Medicine, Newcastle University
1. Abstract
The B-13 pancreatic cell line is unique as it can readily transdifferentiate to hepatocyte-like B13/H cells after exposure to glucocorticoids
such as dexamethasone. However, the underlying mechanisms that take place during this process are not yet fully understood. Previous
research has shown important functions of the glucocorticoid receptor and WNT signalling pathways, as well as the intracellular messenger
β-catenin and SGK1 gene. Epigenetic changes are likely to trigger the transdifferentiation process, as a transient genomic DNA methylation
at 12 hours, followed by an extended demethylation, have been consistently recorded in B-13 cells. This study looks at the HPAC, human
pancreatic acinar cell line, to investigate the epigenetic changes that occur in its transdifferentiation to hepatocyte-like cells and, using
inhibitors, aims to further understand the overall process. These data collected in this study suggest a transient DNA methylation in HPACs
that peaks at 24 hours after dexamethasone treatment and shows a phenotypic change from pancreatic-like to hepatocyte-like cells after 7
days. This study also briefly looks into the sequence of the SGK1 promoter gene, a gene known to play an essential role in B-13
transdifferentiation, and shows the similarities between B13 and rat sequences. From the results of this study the therapeutic use of
laboratory grown hepatocytes to treat liver disease shows much potential and is more cost-effective than iPSC and ESC alternatives.
Key words: DNA methylation, Liver, HPAC, glucocorticoid receptor.
2. Introduction
The hepatocyte is the most common cell type of the liver and it
plays a vital role in the removal of toxins from the body as well as
having roles in the metabolism of hormones, carbohydrates and
lipids. Hepatocytes are bi-functional where they carry an exocrine
role during synthesis and secretion of bile constituents, and an
endocrine role in secretion of a multitude of blood proteins (1, 2).
Therefore when a liver becomes sufficiently damaged by an acute
injury or a chronic low level injury, cirrhosis and scarring mean
that transplantation is often the only treatment. It is however a
long and expensive process with many complications. A recent
study by Parker et al. predicted that about 20% of patients on the
NHS waiting list will die whilst waiting for a liver donor (3). A
number of things can cause liver cirrhosis, including but not
limited to, alcohol and drug abuse, viral hepatitis and metabolic
diseases (2).
Over the last decade a number of discoveries have given rise to
several possible solutions to overcome the restrictions of
transplantation. There is potential for fully differentiated cells to
revert back to progenitor cells through transdifferentiation or
genetic reprogramming to create induced pluripotent stem cells
(iPSCs). These iPSCs can then differentiate into a number of
different cells; theoretically any cell type of the individual the pre-
induced cell was taken from (4). Embryonic stem cells (ESCs) are
another source for transplantation as they are pluripotent and
have the ability to self-renew and differentiate in the long-term
when subjected to the right conditions. This means both iPSCs and
ESCs could potentially be an unlimited source of cells (5).
This study focuses on researching HPACs (Human pancreatic
acinar cells) and the epigenetic effects associated with their
transdifferentiation to hepatocyte-like cells in response to
glucocorticoid treatment, something which both iPSCs and ESCs
have been unable to do past foetal liver stage in vitro (6, 7).
Glucocorticoids are a class of steroid hormone secreted from the
adrenal glands and transdifferentiation is a form of metaplasia;
which is the irreversible switching of fully differentiated cells to
another cell type (8, 9). This research is based on the considerable
evidence of an observable phenotypic change in a pioneering
study by Shen et al. (2000) (10). The study used the AR42J-B13 rat
pancreatic cell line which was first isolated from a rat pancreatic
tumour in the 1970s (11). The AR42J-B-13 or just B13 cell, is a
clone from the original tumour cell line AR42J and the subject of
the study. The AR42J-B-13 is unique as it readily
transdifferentiates into B13/H hepatocyte-like cells, in vitro upon
glucocorticoid treatment and is one of the few cells known to do
this (10, 12).
The cost of producing hepatocytes derived from iPSCs or ESCs is
significantly more than the cost of deriving B-13/H cells, in the
region of 5 million times more expensive. Obtaining hepatocytes
from iPSCs or ESCs is not only more expensive but also more
complicated requiring a multiple stage protocol and a variety of
recombinant growth factors. This estimate doesn’t even take into
account the cost of laboratory equipment, culture media, and the
cost of failures to name but a few (6, 13). Therefore, it is much
more cost effective to use dexamethasone (a synthetic
glucocorticoid) to treat B13 cells and produce hepatocytes, and
much simpler as it only requires basic culture medium (14).
The epigenetic changes, those that affect the expression of DNA
without changing the DNA sequence itself, are thought to be a key
process leading to transdifferentiation. In vertebrates, DNA
methylation, the addition of a methyl group to the 5-position of
cytosine, is restricted to CpG dinucleotide palindromic sequences
which are often found in clusters called CpG islands (15). These
islands are highly methylated and found in about 60 percent of
human promoters (16). Importantly to this study, methylated CpG
islands are associated with transcriptional repression.

DNA is stored as chromatin in eukaryotes; which encompasses a
string of basic repeating units termed the nucleosome core
particle (NCP). Each NCP consists of two tightly wrapped
superhelical turns of DNA wrapped around an octamer core of the
four histone pairs (17). The protein histone pairs are H4, H3, H2A
and H2B and the amino-terminal tails of these histones are subject
to post translational modifications by methylation of lysine and
arginine; further histone modifications can be acetylation,
ubiquitination and phosphorylation (18). Acetylation is
determined by HATs (histone acetyltransferases) and HDACs
(histone deacetylases).
WNT/beta-catenin signalling has been known to play an
important role in liver development as well as regulating ‘zones’ of
hepatocyte gene expression (19, 20). In a study by Wallace et al.
(2010) WNT signalling activity was significantly repressed when
B13 cells were treated with glucocorticoid. The effect was a
transient loss of constitutive WNT3a expression, which has been
shown to be essential for the transdifferentiation towards B13/H
cells (8). Further research by Wallace et al. determined that this
transient repression is upstream of the induction of CCAAT-
enhancer-binding protein-β (C/EBP-β). The mechanisms were
confirmed by siRNA knockdown of β-catenin, the intracellular
messenger of the WNT signalling pathway (21). As WNT signalling
is further upstream of C/EBP-β siRNA knockdown of β-catenin still
resulted in C/EBP-β induction and transdifferentiation in B-13 cells
(22).
The mechanisms behind the effect of glucocorticoids on the WNT
pathway are still unknown. The induction of the serum- and
glucocorticoid-regulated kinase 1 (SGK1) gene to phosphorylate β-
catenin could be part of the crosstalk between glucocorticoid and
WNT signalling pathways (22, 23). When glucocorticoid is added,
SGK1 is distinctly induced, whereas when siRNA was used to
knockdown SGK1, glucocorticoid-dependent transdifferentiation
was inhibited (22).
Since unpublished work (figure.1) has shown methylation of DNA
in B13 after DEX treatment, this may also occur in HPACs.
Therefore, the aim of this study is to see what effects DEX
treatment has on methylation of HPAC DNA and see what effects,
if any, various inhibitors have on DNA methylation in the same cell
line. The SGK1c gene was also looked at to investigate its role in
signalling pathways.
3. Materials and Methods
Cell culture
HPACs (ATCC® CRL-2119) were prepared in vitro in 75cm2 culture
flasks under laminar flow hood. All treatments used the same
500ml of Dulbecco’s modified eagle’s medium (DMEM, Sigma-
Aldrich®) and 50ml of Foetal calf serum (FCS), 80 units/ml-1
streptomycin, 80µg/ml-1 penicillin and 2mM L-Glutamine were
added to each bottle. Flasks were always incubated at 37°C and at
5% CO2 in air. Media was changed every 2-3 days minimum and no
more than 80% confluence was permitted. Phosphate buffer
solution (PBS) (137mM NaCl, 2.7mM KCl, 10mM phosphate pH
7.4) was used to wash cells between media changes. When the
cells became too confluent they were split using about 2mls of 10x
Trypsin EDTA to detach the cells and then a combination of
incubation and gently tapping to encourage all cells to become
suspended. 8mls of original media was added to stop the action of
trypsin and suspension was centrifuged at 2000 rpm for 4
minutes. Supernatant was then removed and cells were re-
suspended in 10 ml of fresh media and distributed equally
amongst new 75cm2 culture flasks and topped up with media.
When a cell count of about 6x106 cells per flask was reached
treatment media was added. For experiment 1, a bottle of DMEM
was prepared as above with the addition of 1µM DEX (added from
1mM stock in ethanol). As a vehicle control the same was done
but 0.1% v/v of ethanol (final concentration) was used instead of
DEX as DEX stock was made up in ethanol. Flasks were then set up
and left for between 0 hours and 14 days (see figure 2.). There
were 3 samples of each treatment to establish some repeatability
and 33 flasks in total. The 0hr cells were just wild type HPACs with
no exposure to treatment.
Figure 2. Table showing treatments in 0hr to 14 day experiment.
VC is normal HPAC media and 0.1% ethanol, DEX is
dexamethasone media. 0hr is wild type HPACs without ethanol
or DEX.
Figure 1 data from unpublished paper titled ‘Trans-differentiation of a
pancreatic progenitor cell to hepatocytes is dependent on irreversible
glucocorticoid receptor-dependent epigenetic alterations’, showing the
percentage methylation of B13 cells treated with both Dexamethasone and
0.1% ethanol over a time course of 10 days. The 6hrs treatment had 10nM
DEX treatment for 6 hours only; then normal media without DEX was used.
10nM DEX had DEX treatment throughout the course of the experiment.
Methylation peaked at 90% within the first 24 hours and quickly dropped.
Data are courtesy of the Wright group, ICM (Newcastle University).
Media Time point Number of flasks
HPAC 0hr 3
HPAC VC 12hr 3
HPAC + DEX 12hr 3
HPAC VC 24hr 3
HPAC + DEX 24hr 3
HPAC VC 48hr 3
HPAC + DEX 48hr 3
HPAC VC 7 days 3
HPAC + DEX 7 days 3
HPAC VC 14 days 3
HPAC + DEX 14 days 3

Inhibitors
The same culture method was used with inhibitors over a period
of 48 hours treatment. The volumes in figure 3. were added at 0
hours and 24 hours to either DEX treated HPACs or ethanol
treated HPACs. However there were no repeats in this
experiment. Inhibitors were dissolved in DMSO and as such 10µl
DMSO with ethanol treatment was used as the control and
reference for methylation.
DNA isolation
At the end of each time point the cells were photographed then
washed with PBS and scraped from the flasks into eppendorfs.
They were then centrifuged at 13000 rpm for 4 minutes and
buffer removed. 200 µl genomic DNA preparation buffer (50mM
Tris-HCL, pH 8.0, 100mM NaCl, 10mM EDTA, 0.5% NP-40) and 20µl
of proteinase K solution was added and eppendorfs were
incubated at 55°C overnight. 50µg of RNase A was then added to
each sample before incubation at rtp. for 20mins. After this 200µl
of phenol was added and each eppendorf was vortexed and pulse
centrifuged to separate the top aqueous phase containing DNA
(~200µl which was kept). 20µl of 3M Na-acetate (pH5.2) and 200µl
of 100% ethanol were then added before placing on dry-ice for
20mins and centrifuging at 13000 for 10mins at 4°C. Supernatant
was then discarded and pellet washed with 70% ethanol before
the same freeze and centrifuge cycle was repeated. Ethanol was
removed and samples dried in air to get rid of any remaining
alcohol. DNA pellets were then re-suspended in between 10-40µl
of sterile water and taken to the nanodrop to quantify DNA
concentration using sterile water as a blank.
Methylation assay
Isolated DNA samples were diluted so that they were 100ng per
30µl as required for the Imprint® Methylated DNA quantification
kit; which was used in the way described by the manufacturer on
isolated DNA from each time point. Absorbance at 450nm was
read using the plate reader and the level of methylation calculated
compared to the wild type (0hr) HPACs.
SGK1c gene
SGK1c promoter gene was predicted to be 241bp through
interrogation of the rat genome sequence (Rn5). Within this
sequence, a functional glucocorticoid response element was
confirmed using AliBaba2.1. Upstream and downstream primers
were designed based on the sequence prediction. The promoter
sequence was amplified by PCR and then sent to DNA Sequencing
and Services™ (Dundee University) for sequencing of the SGK1c
promoter.
Constructs of B13 and rat DNA were prepared using Zero Blunt®
PCR cloning kit (Life Technologies™) and plasmid DNA purification
was achieved using QIAprep® Spin MIniprep kit. Both processes
followed manufacturers’ protocol.
Samples were restriction digested using Ecor1 enzyme and
analysed by gel electrophoresis along with a 100bp ladder and
uncut construct sample.
4. Results
Wild type HPACs can be seen under the microscope in figure 4 (a).
The change in morphology of the cell started to become apparent
by day 7 of the study, with the same features noted in B-13 cell
differentiation that were illustrated by Shen et al. The ‘flattening
onto the substratum and becoming more epithelial-like’
characteristics can be seen most clearly in figure 4. (b) in the 19
day DEX culture. The B-13 treated with DEX had maximum
differentiation of ~95% confluence after 2 weeks (10) but the
HPACs treated with DEX took a few days longer to reach full
confluence and differentiation, which was closer to the 19 day
mark.
The 0.1% ethanol treated vehicle control HPACs had a much
faster doubling rate than the 10 µM DEX treatment, so much so
that the 14 day control flasks required two cell passages in the
time frame and the DEX treated 14 day flasks didn’t need any cell
passage without becoming confluent. In fact DEX treated HPACs
were still alive after 28 days and one cell passage.
The results after both treatments to cells can be seen in figure 5.
A to J. All pictures are from the same cell line and cells had the
same observable wild type phenotype from 0 hours through to 48
hours in ethanol control A-C and DEX treated F-H. At 7 days
treatment there is a clear difference between D and I, and even
more so between E and J (14 days). We can therefore see that the
cells have differentiated into hepatocyte-like cells 7 days after
10µM DEX treatment.
Flask Inhibitor volume and media
1 10µl DMSO 10ml ethanol media
2 10µl RU486 stock 10ml ethanol media
3 10µl SGK1 10ml ethanol media
4 10µl ethanol 10ml Na-butyrate
5 10µl 5-AZA 10ml ethanol media
6 10µl trichostatin A 10ml ethanol media
7 200µl DMSO 10ml ethanol media
8 10µl DMSO 10ml DEX media
9 10µl RU486 stock 10ml DEX media
10 10µl SGK1 10ml DEX media
11 10µl ethanol 10ml Na-butyrate
12 10µl 5-AZA 10ml DEX media
13 10µl trichostatin A 10ml DEX media
14 200µl DMSO 10ml DEX media
Figure 4. microscope image at x200 magnification
Figure 3. table showing treatments added to wild type HPAC culture flasks
over 48 hours. Flask 1 was the control and reference. 5-AZA (5-Azacytidine,
25mM in DMSO), DMSO (dimethyl sulphoxide), RU486 (Mifepristone, 10mM
in DMSO), SGK1 (100µM in DMSO), Trichostatin A, (25µM in DMSO). Na –
butyrate, (2mM directly to media). All concentrations are stock.

After lower than expected DNA concentrations were recorded it
was decided that overnight incubation at 55°C (after proteinase k
treatment) was necessary, instead of 2 hours; and this was
particularly the case for the 12 to 48hr samples, which were less
confluent so had less DNA to isolate. After the DNA isolation of
the 0hr to 14 day samples, the first DNA methylation assay was
conducted as per the protocol and absorbance readings recorded
for each. The Imprint® Methylated DNA quantification kit’s
protocol suggested using a single point method to calculate the
percent methylation of the samples relative to methylated control
DNA supplied with the kit. Given that at any point 70% of all CpG
dinucleotides in the mammalian genome are methylated (24), a
global variation in this methylation across the genome can change
gene expression (25). These shifts in DNA methylation have been
observed in a range of diseases from cancers to autoimmune
illnesses (26, 27). As a result, determining the degree of
methylation in DNA can shed some light on the epigenetic
changes taking place.
Initially a positive methylated DNA control was used as advised
but the absorbance readings for the control were consistently
lower than expected. At 450nm the absorbance was regularly
similar to the blank (DNA buffer only) which had no DNA and
therefore minimal if any methylation. The positive control should
have had one of the highest absorbance readings so as to
represent 100% methylation in the single point method. This
meant that percentage methylation could not be identified as
figures over one hundred percent would be commonplace and not
Figure 6. 0 days (0hrs) to 14 days (336 hours)
0
0.5
1
1.5
2
2.5
3
3.5
4
0 31 60 91 121 152 182 213 244 274 305 335
Foldmethylationcomparedto0hr(nounits)
Time (hours)
DEX treatment
ethanol control
Figure 5. microscope image at x200 magnification showing ethanol control treatment A-E and dexamethasone treatment F-J.

comparable. Instead the 0 hour HPAC wild type was used as a
reference and any methylation could be seen as a fold increase or
decrease from this value. This meant that methylation changes
could still be compared to these B-13 cell data in figure 1.
The 0hr to 14 day assay (figure 6) showed a jump in global DNA
methylation in the 24 hour DEX treated culture which was 2.36
fold higher than the wild type methylation. Methylation then
dropped to 2.04 fold increase from wild type and then down to
1.2 fold at 7 days (162 hours), before a further decrease to 0.87
fold. Ethanol control did not reach as high a methylation as the
DEX treatments. At 24 hours vc it was just 0.76 fold difference to
the wild type. There was an increase to 1.08 fold and 1.58 fold at
48 hour and 7 day respectively and then a drop to 0.85 fold by 14
day (336 hours) vc.
There was quite a large degree of standard deviation for some
samples, for example it was as high as 2.36± 1.36 in 24 hour DEX
and 2.04± 1.3 in 48 hour DEX. This was due to some anomalous
results when calculating averages. In the case of average
absorbance readings for 24 hour DEX treatment, of the three
readings 0.272, 1.584 and 1.597, the first was clearly an anomaly.
In this methylation there were three other such anomalies where
a reading varied by greater than 50 percent of the average.
Henceforth another graph (figure 7) interpreted the experiment
with anomalies removed. There was always at least two of the
three samples at each time point used to determine the average
fold methylation and for this graph 29 of the original 33 samples
were used.
The manipulated data shows a very defined peak at 24 hours DEX
of 4.25 fold increase from the wild type. This is much more in the
region of what was expected from the B-13 research. At 48 hours
DEX it has dropped way down to 1.37 fold contrary to figure 6
data. From 48 hours through to 14 day, vc and DEX treatments
have a very similar minimal effect on methylation. Most
importantly the standard deviation was vastly smaller without the
anomalies, ranging from 4.25± 0.01 to 1.44± 0.22. The
manipulated graph is very similar to the pattern shown in B-13 cell
methylation in figure 1. The same peak can be seen at 24 hours
DEX HPACs and 24 hours DEX B-13, then the same drop in
methylation and plateauing off.
Another repeat of the 0 to 7 day DEX and ethanol treatment was
done to try to get some conclusive results and these can be seen
in figure 8. The same spike in DNA methylation at 24 hours DEX
can be seen again and the same decrease at 48 hours DEX.
However this time there is a large increase at 7 days. The 7 day
average results are still similar as in the case of the previous
graphs but they are 4 fold higher than what was expected. Of
course the exact methylation response of HPACs to DEX treatment
is not confirmed, so this may be the case but such an increase is
unlikely from current research. The higher than expected result
could be due to contamination of 7 day DNA and therefore a
higher amount of DNA than the universal 100ng/µl. The similar
increase suggests the same mistake was made to both DEX
treated 7 day and 7 day ethanol cultures. Furthermore, there was
no anomaly in six samples of 7 day cultures, suggesting a
consistent error or a valid result.
On the other hand, there was a 3.1 fold increase at 24 hours DEX
which was to be expected and then a drop to 1.7 fold, not
dissimilar from the first experiment. Again there was a large
standard deviation for some results. 7 day vc was 4.8 fold± 0.89
and 48 hour DEX was 1.7 fold± 1.78.
However, there was still no overlap between standard deviation
for 24 hour DEX and vc so the results for 24 hour cultures were
statistically different. This would suggest that there is a difference
in methylation in the DEX treated cultures and ethanol cultures at
the 24 hour mark.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
0 31 60 91 121 152 182 213 244 274 305 335
Time (hours)
DEX treatment
ethanol control
Figure 7. data minus anomalies. N.B. very small standard deviation at 24 hours DEX treatment (too small to see on graph).

The data was once again manipulated by removing two anomalous results. Both anomalies were in the DEX treatment. 24 hour DEX treatment
had three samples, 0.371, 0.446 and 0.242 absorbance readings and the last sample was taken to be anomalous. The 48 hour DEX treatment
had a large anomaly of 0.447 compared to 0.145 and 0.100. A graph depicting these manipulated data can be seen in figure 9. Standard
deviation has been reduced in both treatments for 24 hour and 48 hour time points. However the standard deviation of the wild type control
of 1± 0.91 questions the reliability of these data, as this was the value that the rest of the data was compared with.
Figure 8. 0-7 days treatment
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7
Time (days)
DEX treatment
ethanol control
Figure 9. 0-7 day treatment minus anomalies.
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7
Foldmethylationcomparedtoohr(nounits)
Time (days)
DEX treatment
ethanol control

Effect of Inhibitors
The results for inhibitors are summarised in figure 10. Ethanol
media plus 10µl DMSO (dimethyl sulphoxide) was used as the
control for referencing the fold methylation increase or decrease,
as inhibitors were dissolved in DMSO. The control, for some
reason, produced the largest absorbance reading and was much
higher than the DEX media plus 10µl DMSO treatment which was
hard to understand. 200µl DMSO treatments were much lower
than the control despite DMSO to be a known inducer of DNA
hydroxymethylation in embryonic stem cells (28). As such, the
results of the inhibitor experiment could not be analysed and
compared logically to each other.
According to what is already know about inhibitors, the only two
things we might definitely expect to inhibit DEX induced
methylation are RU486 and 5-AZA. This is because, as mentioned
previously, transdifferentiation of B-13 to B-13/H cells is
associated with an induction of SGK1 gene expression. This
induction is crucial for the cross-talk between two essential
pathways of the transdifferentiation process, the glucocorticoid
and WNT signalling pathways (22). Consequently, an equal
decrease in fold methylation was expected in DEX and vehicle
media when these inhibitors were added. There was a decrease
but not by the same amount.
Na- butyrate is a known HDAC inhibitor and experiments have
shown how it can cause histone modification in HeLa cells, which
are a human cell line (29). However it can also induce a 20-30%
hypomethylation in mammalian cell DNA (30). There was a 26%
methylation fold decrease in cells treated with DEX and Na-
butyrate. This would have been expected from the ethanol control
which was in actual fact significantly smaller, more like a 2 fold
(100%) hypomethylation. Looking back on results for 24 hour DEX,
Na-butyrate could have potentially shown what has a greater
effect
on methylation: the glucocorticoid which causes DNA
hypermethylation, or Na-butyrate which causes hypomethylation.
Of course, neither have ever been tested together on HPACs so
the real outcome is very much unknown.
DMSO increases expression of genes involved in DNA
hydroxymethylation and decreases global DNA methylation in the
MC3T3-E1 cell line, which are pre-osteoblastic stem cells (28).
Cytosine hydroxymethylation is an important epigenetic
modification on mammalian DNA. Hydroxymethylation replaces a
cytosine at the C5 position with a hydrogen atom by using a
hydroxymethyl group (31). Hydroxymethylation level has been
demonstrated to be involved in gene regulation (32) and also
found to be associated with pluripotency of stem cells (33). Prior
to the experiments of this study the influence of DMSO on the
HPACs was previously unexplored but it could be fair to assume
that an increase in DMSO could cause a decrease in fold
methylation.
Mifepristone (RU486) is a glucocorticoid receptor antagonist and
has already been shown to block transdifferentiation in B-13 cells.
It prevents transdifferentiation by inhibiting the glucocorticoid
receptor in such a way that it can no longer interact with parts of
the WNT-signalling pathway (22). This is as much as we
understand about its role in transdifferentiation, but the effects of
RU486 have been examined in other studies where it had an
antagonistic effect on methylprednisolone protection (another
synthetic glucocorticoid) (34).
Trichostatin A is a Streptomyces product and another histone
deacetylase inhibitor similar to Na-butyrate. It causes inhibition of
the rat cell cycle at G1 and G2 phases at ‘extremely’ low
concentrations (35). Trichostatin A has already been used on B-13
cells to inhibit transdifferentiation to B-13/H in unpublished data
Media Inhibitor Absorbance 450nM Fold compared to control
Dex SGK1 0.506 0.750
VC SGK1 0.326 0.411
Dex 5-AZA 0.457 0.657
VC 5-AZA 0.504 0.746
Dex Na-butyrate 0.498 0.734
VC Na-butyrate 0.085 -0.043
Dex Trich-A 0.203 0.179
VC Trich-A 0.083 -0.047
Dex RU486 0.262 0.290
VC RU486 0.172 0.121
Dex 10µl DMSO 0.115 0.013
VC 10µl DMSO 0.639 1.000
Dex 200µl DMSO 0.063 -0.085
VC 200µl DMSO 0.152 0.083
Figure 10. Table showing the results of inhibitors on HPACS.

by the Wright group (Newcastle University). It is thought to act as
an HDAC, like Na-butyrate.
5- Azacytidine is a nucleoside analog and is a methylation
inhibitor when present in DNA. As such, it has been used
therapeutically in cancer treatment to silence critical regulatory
genes (36). Again it’s been used with B-13 cells in unpublished
experiments to inhibit DNA methylation and transdifferentiation
to B-13/H cells.
SGK1 results
Rat SGK1 constructs 1 to 6 on figure 11 can be seen along with
the uncut plasmid. The uncut plasmid had no bands below about
4kb, instead there was just supercoiled DNA, nicked DNA and the
same uncut 4kb band present in all 6 constructs. ECoRI was used
to cut the plasmids at different places to produce varying sizes of
linear, sticky ended DNA. The predicted 241bp band suspects can
be seen in the lane with 1 ECoRI cut and the lane with 2 ECoRI
cuts. Therefore, these two constructs were sent for sequencing
and the results of the sequence can be seen below in figure 13.
The red sequence represents the SGK1 promoter database
predicted sequence which matches the orange rat sequence for
the most part.
B-13 SGK1 constructs can be seen in figure 12. as ECoRI digests of
clones 1-12. The same bands of approximately 241bp can be seen
in lanes 9 and 10 along with the smaller, lighter coloured, sub
100bp bands below them; which were also in the rat constructs.
Therefore, constructs from lane 9 and 10 were sent for
sequencing as well. Figure 13. shows the returned B13 sequence
in green and it proved to match the red database sequence
exactly, thus identifying the SGK1 gene in B13 cells and showing
no mutations.
Figure 13. Sequences of Rat and B-13 constructs containing the SGK1 sequence predicted in red. The rat sequence doesn’t match
the database sequence completely however B-13 cell does
Figure 11. Rat SGK1 clones on a 1.5% agarose gel.
US primer binding site
AAGCCAGTCTCAACAACTTGATTCCCTCTCTGTTAAATGATAAGGTTAGCAGACCATCCGTCTCCAATCGCTGACCATTCCGGGCACGGAGAGG
|| | ||||||||||||||||||||||||||||||||||||||||||||||||
CCCGTATGATCGTAGCAGACCATCCGTCTCCAATCGCTGACCATTCCGGGCACGGAGAGG
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
AAGCGGCACATCCAAATCCGTCCCTTTCTGTTCCAACCACAAATATATGGTTGAGCTTTCTCTTTATTGCACTTGGAGGCCTCTGACTTGAATT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DS primer binding site
CCGTGTTCCTTTTGGAATGCATGCCAGCTTCCCTGGGAGACCAGACCCTGAGCA
||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||||||||||||||||||||||||||||||||||||||||||||||||||
||||||
CCGTGT
KEY
Red = database sequence
Orange = sequence from rat DNA
Green = sequence from B-13 cells
Figure 12. ECoRI digests of clones 1-12 of B-13 SGK1 on
1.5% agarose gel.

5. Discussion
It is clear from previous research that glucocorticoid treatment
causes B13 cell transdifferentiation to B13/H over the course of
about 6 days (6, 10, 13, 14). It is also widely believed that the
mechanisms underlying this process are more than likely due to
epigenetic modifications, with the hypothesis that these changes
are due to the action of dexamethasone at the glucocorticoid
receptor, which acetylates β-catenin and switches-off WNT
signalling. SGK1c is likely to be involved in this mechanism as its
inhibition by knockdown siRNA stops differentiation taking place
(22). Unpublished data has also shown SGK1c to be inhibited by a
specific SGK1 inhibitor in B-13 cells, the same SGK1 inhibitor used
in this study.
Given HPACs’ morphological similarity to the pancreatic B-13 cell,
it was expected that there would be similarities; the first of which
was the differentiation to hepatocyte-like cells. The same
flattening and becoming more epithelial-like was observed in
HPACs, albeit over a slightly longer time period. As a result, this
study can conclusively say that dexamethasone treatment does
lead to the transdifferentiation of HPACs to hepatocyte-like cells.
This effect can be clearly seen in the phenotypic change captured
in figures 4 and 5.
DNA methylation showed consistent peaks in fold methylation at
24 hours DEX treatment which would suggest that it is at this
point an increase in global DNA methylation begins the
differentiation process. Given that the B-13 cell percentage
methylation has a very large peak at 24 hours DEX, it could be fair
to assume that the same epigenetic modifications are happening
in the two cells prior to differentiation. However another
experiment to calculate percentage methylation would have to be
done for HPACs, as conclusive evidence cannot be based on fold
methylation and percentage methylation comparisons. There
were also various anomalous absorbance readings in the data and
large standard deviations proving some data insignificant.
These anomalous data could be explained by mistakes being
made in the lengthy and multi-step protocol. For example, there
could have been mistakes in making dilutions as every DNA
sample required a different volume of DNA buffer to make them
all 100ng/µl each as per the protocol. There were upwards of 70
dilutions made and it would be difficult to mix these all completely
accurately. Contamination of DNA with foreign DNA could have
also affected absorbance readings, as the foreign DNA would also
have methylation which would result in overestimated readings.
The inhibitor results weren’t able to be analysed because of the
higher than expected control. So in the future this experiment
would have to be repeated, with repeats of samples for average
absorbance readings as in the 0 to 14 day experiment. This way
anomalies can be determined. The experiment would also need to
be slightly altered by treating cells with inhibitors 24 hours before
DEX and ethanol medium were added and then treating with
inhibitors again as the medium is added. This is because from the
results of this study DNA methylation in HPACs likely peaks at 24
hours DEX treatment; so the DNA needs to be isolated 24 hours
after DEX is added or 48 hours into the experiment. In this study it
was wrongly isolated 48 hours after DEX treatment, after the
methylation pulse.
SGK1c gene sequencing concluded that the same SGK1 promoter
gene that is in B-13 is likely found in the rat genome. The gel
electrophoresis of both B-13 and rat constructs verified the same
241bp SGK1 promoter as a linear band of DNA. The presence of
different size bands in the SGK1c PCR experiment could be due to
non-specific amplification of other sequences, or the presence of
insertions in some of the promoters of the SGK1c gene (there are
at least 4 in the B-13 cell because it is 4n). These are being
sequenced to determine whether this is the case. Future
experiments could look into the HPAC and human form of SGK1
promoter, to look for further similarities between the cell lines.
Given more time in the laboratory, further experimentation could
have investigated DEX/ethanol treatments over 3, 4, 5 and 6 day
time points as there is a gap in data for 48 hour to 7 day
methylation. Immunocytochemistry for phenotypic markers such
as albumins could also be performed for cell identification, which
has been used to show the difference between B-13 and B-13/H
cells (10). Investigation of histone translational modification could
also be used to identify their role in the differentiation
progression since DNA and histone lysine methylation rely
mechanistically on each other (15, 37).
Acknowledgments: Many Thanks go to the Wright group (ICM),
Newcastle University and in particular Prof. Wright and Dr Fairhall
without whom this study would not be possible.
References
1. Marek, C. J., Cameron, G. A., Elrick, L. J., Hawksworth, G.
M., and Wright, M. C. (2003) Generation of hepatocytes
expressing functional cytochromes P450 from a
pancreatic progenitor cell line in vitro. Biochem J 370,
763-769
2. Friedman, S. L. (2000) Molecular regulation of hepatic
fibrosis, an integrated cellular response to tissue injury. J
Biol Chem 275, 2247-2250
3. Parker, R., Armstrong, M. J., Bruns, T., Hodson, J., Rowe,
I. A., Corbett, C. D., Reuken, P. A., Gunson, B. K.,
Houlihan, D. D., Stephenson, B., Malessa, C., Lester, W.,
and Ferguson, J. W. (2014) Reticulocyte count and
hemoglobin concentration predict survival in candidates
for liver transplantation. Transplantation 97, 463-469
4. Sauer, V., Roy-Chowdhury, N., Guha, C., and Roy-
Chowdhury, J. (2014) Induced pluripotent stem cells as a
source of hepatocytes. Curr Pathobiol Rep 2, 11-20
5. Simbulan, R. K., Di Santo, M., Liu, X., Lin, W., Donjacour,
A., Maltepe, E., Shenoy, A., Borini, A., and Rinaudo, P.
(2015) Embryonic Stem Cells Derived from In Vivo or In
Vitro-Generated Murine Blastocysts Display Similar
Transcriptome and Differentiation Potential. PLoS One
10, e0117422
6. Fairhall, E. A., Charles, M. A., Wallace, K., Schwab, C. J.,
Harrison, C. J., Richter, M., Hoffmann, S. A., Charlton, K.
A., Zeilinger, K., and Wright, M. C. (2013) The B-13
hepatocyte progenitor cell resists pluripotency induction
and differentiation to non-hepatocyte cells. Toxicology
Research 2, 308-320
7. Mann, D. A. (2015) Human induced pluripotent stem
cell-derived hepatocytes for toxicology testing. Expert
Opin Drug Metab Toxicol 11, 1-5
8. Wallace, K., Marek, C. J., Hoppler, S., and Wright, M. C.
(2010) Glucocorticoid-dependent transdifferentiation of
pancreatic progenitor cells into hepatocytes is
dependent on transient suppression of WNT signalling. J
Cell Sci 123, 2103-2110

9. Wallace, K., Flecknell, P. A., Burt, A. D., and Wright, M. C.
(2010) Disrupted pancreatic exocrine differentiation and
malabsorption in response to chronic elevated systemic
glucocorticoid. Am J Pathol 177, 1225-1232
10. Shen, C. N., Slack, J. M., and Tosh, D. (2000) Molecular
basis of transdifferentiation of pancreas to liver. Nat Cell
Biol 2, 879-887
11. Longnecker, D. S., Lilja, H. S., French, J., Kuhlmann, E.,
and Noll, W. (1979) Transplantation of azaserine-
induced carcinomas of pancreas in rats. Cancer Lett 7,
197-202
12. Wallace, K., Marek, C. J., Currie, R. A., and Wright, M. C.
(2009) Exocrine pancreas trans-differentiation to
hepatocytes--a physiological response to elevated
glucocorticoid in vivo. J Steroid Biochem Mol Biol 116,
76-85
13. Probert, P. M. E., Chung, G. W., Cockell, S. J., Agius, L.,
Mosesso, P., White, S. A., Oakley, F., Brown, C. D. A., and
Wright, M. C. (2014) Utility of B-13 Progenitor-Derived
Hepatocytes in Hepatotoxicity and Genotoxicity Studies.
Toxicological Sciences 137, 350-370
14. Fairhall, E. A., Wallace, K., White, S. A., Huang, G. C.,
Shaw, J. A., Wright, S. C., Charlton, K. A., Burt, A. D., and
Wright, M. C. (2013) Adult human exocrine pancreas
differentiation to hepatocytes - potential source of a
human hepatocyte progenitor for use in toxicology
research. Toxicology Research 2, 80-87
15. Rose, N. R., and Klose, R. J. (2014) Understanding the
relationship between DNA methylation and histone
lysine methylation. Biochim Biophys Acta 1839, 1362-
1372
16. Berger, J., and Bird, A. (2005) Role of MBD2 in gene
regulation and tumorigenesis. Biochem Soc Trans 33,
1537-1540
17. Luger, K. (2003) Structure and dynamic behavior of
nucleosomes. Curr Opin Genet Dev 13, 127-135
18. Marks, P. A., Miller, T., and Richon, V. M. (2003) Histone
deacetylases. Curr Opin Pharmacol 3, 344-351
19. Tan, X., Behari, J., Cieply, B., Michalopoulos, G. K., and
Monga, S. P. (2006) Conditional deletion of beta-catenin
reveals its role in liver growth and regeneration.
Gastroenterology 131, 1561-1572
20. Thompson, M. D., and Monga, S. P. (2007) WNT/beta-
catenin signaling in liver health and disease. Hepatology
45, 1298-1305
21. Hoppler, S., and Kavanagh, C. L. (2007) Wnt signalling:
variety at the core. J Cell Sci 120, 385-393
22. Wallace, K., Long, Q., Fairhall, E. A., Charlton, K. A., and
Wright, M. C. (2011) Serine/threonine protein kinase
SGK1 in glucocorticoid-dependent transdifferentiation of
pancreatic acinar cells to hepatocytes. J Cell Sci 124,
405-413
23. Lang, F., Bohmer, C., Palmada, M., Seebohm, G., Strutz-
Seebohm, N., and Vallon, V. (2006) (Patho)physiological
significance of the serum- and glucocorticoid-inducible
kinase isoforms. Physiol Rev 86, 1151-1178
24. Shen, L., Guo, Y., Chen, X., Ahmed, S., and Issa, J. P.
(2007) Optimizing annealing temperature overcomes
bias in bisulfite PCR methylation analysis. Biotechniques
42, 48, 50, 52 passim
25. Rodenhiser, D., and Mann, M. (2006) Epigenetics and
human disease: translating basic biology into clinical
applications. Cmaj 174, 341-348
26. Leone, G., Teofili, L., Voso, M. T., and Lubbert, M. (2002)
DNA methylation and demethylating drugs in
myelodysplastic syndromes and secondary leukemias.
Haematologica 87, 1324-1341
27. Beier, V., Mund, C., and Hoheisel, J. D. (2007)
Monitoring methylation changes in cancer. Adv Biochem
Eng Biotechnol 104, 1-11
28. Thaler, R., Spitzer, S., Karlic, H., Klaushofer, K., and
Varga, F. (2012) DMSO is a strong inducer of DNA
hydroxymethylation in pre-osteoblastic MC3T3-E1 cells.
Epigenetics 7, 635-651
29. Riggs, M. G., Whittaker, R. G., Neumann, J. R., and
Ingram, V. M. (1977) n-Butyrate causes histone
modification in HeLa and Friend erythroleukaemia cells.
Nature 268, 462-464
30. Niehrs, C. (2009) Active DNA demethylation and DNA
repair. Differentiation 77, 1-11
31. Severin, P. M., Zou, X., Schulten, K., and Gaub, H. E.
(2013) Effects of cytosine hydroxymethylation on DNA
strand separation. Biophys J 104, 208-215
32. Munzel, M., Globisch, D., and Carell, T. (2011) 5-
Hydroxymethylcytosine, the sixth base of the genome.
Angew Chem Int Ed Engl 50, 6460-6468
33. Tahiliani, M., Koh, K. P., Shen, Y., Pastor, W. A.,
Bandukwala, H., Brudno, Y., Agarwal, S., Iyer, L. M., Liu,
D. R., Aravind, L., and Rao, A. (2009) Conversion of 5-
Methylcytosine to 5-Hydroxymethylcytosine in
Mammalian DNA by MLL Partner TET1. Science 324, 930-
935
34. Hirose, Y., Tabuchi, K., Oikawa, K., Murashita, H., Sakai,
S., and Hara, A. (2007) The effects of the glucocorticoid
receptor antagonist RU486 and phospholipase A2
inhibitor quinacrine on acoustic injury of the mouse
cochlea. Neuroscience Letters 413, 63-67
35. Yoshida, M., Kijima, M., Akita, M., and Beppu, T. (1990)
Potent and specific inhibition of mammalian histone
deacetylase both in vivo and in vitro by trichostatin A. J
Biol Chem 265, 17174-17179
36. Christman, J. K. (2002) 5-Azacytidine and 5-aza-2'-
deoxycytidine as inhibitors of DNA methylation:
mechanistic studies and their implications for cancer
therapy. Oncogene 21, 5483-5495
37. Zhou, J., and Cidlowski, J. A. (2005) The human
glucocorticoid receptor: one gene, multiple proteins and
diverse responses. Steroids 70, 407-417

Dissertation final complete1

More Related Content

What's hot (20)

Viewers also liked (10)

Similar to Dissertation final complete1 (16)

Dissertation final complete1