Transcript Novel Functions of PXR: A Bioinformatic Approach

Novel Functions of PXR: A Bioinformatic Approach
Callum Allison, Hannah Sian-McGuiness, Dr Ian Bailey
Contact: [email protected]
Introduction:
The pregnane x receptor (PXR, NR1I2) is well established as being chiefly responsible for the coordinate regulation of xenobiotic metabolism. The promiscuous and prolific
nature of the ligand binding domain is such that a wide variety of steroidal molecules will bind to, and activate the receptor. In recent years some interesting and novel roles
for PXR have come to the forefront, with expression being confirmed in the vascular endothelium whereby it acts to mediate xenobiotic disposition through this vital tissue,
and mediate oxidative stress (Swales, 2012). It is also apparent that PXR mediates expression of novel and interesting genes, such as CDKN1A, CDKN1B, RBL2, GAS1,
SERPINE1, PLAUR, SKP2 AND FBXW7 indicating a role in cell cycle and proliferation (Shizu, 2013).
In taking a bioinformatic approach to identifying novel roles of PXR, it is necessary to move away from the more traditional agonists, such as rifampicin, as the literature basis
for these compounds focuses very heavily on their role in drug metabolism. Instead it was decided to use the isoquinoline alkaloid berberine as a basis for this search, as
while this is a confirmed agonist of PXR the research surrounding the compound focusses on a wide range of cellular effects.
Computational analysis of berberine:
Berberine is an established agonist for PXR, with evidence of its modulation of CYP3A genes in a PXR dependent manner. Using the Agilent literature plugin for Cytoscape, a
gene interaction network was generated for berberine. This network was then analysed for the nature of the interaction and colour coded according to the expression of
genes associated with the interactions between PXR and berberine (Figure 1). The genes which are regulated were then analysed for a transcriptional involvement of PXR,
some of which are highlighted in figures 2a through f.
Experimental:
A cell viability (MTT) assay was performed on HuH7 cells over 24, 48 and 72 hours (Figure 3) to evaluate the cytotoxicity of berberine. Results were analysed using GraphPad
Prism. Concentrations which did not elicit an overt cytotoxicity were selected for further analysis (1 and 5 µM).
Further to this, HuH7 cells were treated with berberine chloride at 1 and 5 µM, total RNA was extracted (Machery-Nagel) and converted into cDNA using random primers and
M-MLV reverse transcriptase (Promega). cDNA was quantitated (NanoDrop) and a stock concentration of 400 ng/µl was created. Gene specific primers were used to analyse
for CYP3A4, CYP2B6, SULT1A1, VEGFA, MMP2, CDK2, CDKN1A and GAPDH (sequences may be found in Table 1). Results were analysed by agarose gel electrophoresis for
qualitative assessment.
Gene
CYP3A4
CYP2B6
SULT1A1
VEGFA
MMP2
CDK2
CDKN1A
Forward (Upstream) Primer
5’-GCACCACCCACCTATGATACT-3’
5’-GGAGATTGAACAGGTGATTGG-3’
5’-CCGAAAAGGGAGATTCAAAAG-3’
5’-TGGTGAAGTTCATGGATGTCTA-3’
5’-GTCCACTGTTGGTGGGAAC-3’
5’-GCTCCAGGGCCTAGCTTTCT-3’
5’-GAGCGATGGAACTTCGACTT -3’
Reverse (Downstream) Primer
5’-CTTTCAGGGAGGAACTTCTCAG-3’
5’-GGATGATGTACCCTCGGAAG-3’
5’-ATGAAGGGGGAGATGCTGT-3’
5’-CTGCATGGTGATGTTGGACT-3’
5’-CTTGGTCAGGGCAGAAGC-3’
5’-CCGGAAGAGCTGGTCAATCTCAGA-3’
5’- AGGTCCACATGGTCTTCCTCT -3’
Amplicon Length (bp)
314
283
308
218
474
123
380
Table 1 – Oligonucleotide primer sequences.
Figure 4 – Qualitative gene expression analysis:
Huh7 cells were treated for 48 hours with concentrations of
1 or 5 µM berberine chloride. RNA was extracted and cDNA
synthesised. PCR was performed on cDNA samples
standardised to 400 ng/µl, using gene specific primers (Table
1) Results were analysed using the GelRed nucleic acid stain
and 1% agarose gel electrophoresis. GAPDH is used as the
loading control to demonstrate consistency of amplification.
Figure 3 – Cell viability in response to berberine chloride:
Huh7 cells were treated with concentrations of berberine chloride
between 0 and 20 µM. Cell viability was assessed by MTT assay using
the DMSO vehicle as a control. Results were processed by one way
analysis of variance with Bonferroni multiple comparisons post hoc
testing using GraphPad Prism, significant results are indicated with
*=P≤0.05 ** P≤0.01.
Discussion
Figure 1: Cytoscape interaction map generated from berberine. Red genes are downregulated,
green genes are upregulated and yellow/blue indicates possible association and more research is
needed.
A
B
C
In silico analysis of the berberine array network indicates some interesting potential
PXR interactions.
• CDK2, 4 and 6 are downregulated by berberine, and are known to suppress the
expression of PXR-regulated genes.
• CDKN1A and CDKN1B are upregulated in response to berberine; this is known to
be regulated by PXR directly.
• MMP2 has previously been shown to be inhibited in response to UDCA, an FXR
agonist and possible PXR agonist.
• The vascular endothelial growth factor, VEGFA has been shown to be regulated by
PXR in response to rifampicin,
The MTT results indicate that berberine is significantly cytotoxic at doses over 5 µM
at 48 hours. Therefore doses of 1 and 5 µM were chosen to examine any gene
expression changes.
E
D
Figure 2a-e: Sections of the Cytoscape network
isolated to highlight areas of interest. A) Cell growth,
B) Growth Factors and Cytokines, C) Xenobiotic
Metabolism, D) Cell Signalling and E) Apoptosis.
As before, red genes are downregulated, green genes
are upregulated and yellow/blue indicates possible
association and more research is needed.
The expected changes seen in expression of CYP3A4 were not observed in this
system, likewise neither were CYP2B6 and SULT1A1. These genes are classical
markers of PXR activity and their lack of observable change may indicate the PXR is
not responsible for the gene changes observed in response to berberine.
Future Work
Using Rifampicin as a control. Using HepG2 and HepaRG cells. qPCR for gene
expression. Reporter gene assays for a number of receptors. CHiPSeq to identify
targeted promoter regions.