Abstract |
The aim of the present PhD thesis was to investigate the molecular mechanisms
that control the expression of genes that are involved in lipoprotein metabolism and to
identify new molecular approaches that increase HDL levels in plasma.
Lipoprotein metabolism involves the transport of lipids, particularly cholesterol
and triglycerides, from intestine and liver to peripheral tissues and from periphery
back to the liver (reverse cholesterol transport) and is facilitated by numerous proteins
including apolipoproteins, membrane transporters, plasma enzymes and lipoprotein
receptors. Lipoprotein lipase (LPL) plays a critical role in lipoprotein remodeling by
catalyzing the hydrolysis of triglycerides (TGs) present in TG-rich lipoprotein
particles such as very low density lipoprotein (VLDL) and chylomicrons (CMs) to
free fatty acids for the subsequent storage in adipose tissue or utilizations for the
production of energy by various tissues. LPL is expressed primarily in the adipose
tissue but it is also expressed at lower levels in other tissues including the liver. The
role of LPL in the adult liver has been controversial due its low levels of expression
but recent studies in mouse models of liver LPL overexpression or deficiency have
revealed important new roles of the enzyme in glucose and lipid metabolism.
In Part I we characterized the mechanism that controls the expression of human
LPL in hepatic cells at the level of transcription. First, we cloned the human LPL
promoter and performed a deletion analysis using transactivation assays in human
hepatoblastoma HepG2 cells. We revealed that the proximal region -109/-28 is
important for basal hepatic LPL promoter activity. An in silico analysis of this region
showed that it harbors a putative binding site, at position -47/-40, for the hepatic
transcription factor forkhead box A2 (FOXA2) or Hepatocyte Nuclear factor 3β
(HNF-3β), shown previously to play important roles in lipid and glucose homeostasis.
Silencing of endogenous FOXA2 expression in HepG2 cells (using a specific siRNA)
reduced the LPL mRNA and protein levels. Direct binding of FOXA2 to the novel
binding site was established using chromatin immunoprecipitation assays, ex vivo and
DNA affinity precipitation assays in vitro. This element was further characterized by
site directed mutagenesis and it was found that five nucleotide substitutions in the
FOXA2 site abolished the binding of FOXA2 and reduced the basal activity of the
LPL promoter and the FOXA2-mediated transactivation. Next, we studied the effect of insulin on the hepatic regulation of the LPL gene in HepG2 cells and we showed
that insulin induces the phosphorylation of AKT and the nuclear export of FOXA2
causing a reduction in the LPL mRNA levels and promoter activity. Based on these
findings, we proposed a novel role of FOXA2 in the regulation of the human LPL
gene in hepatic cells by insulin.
In Part II we elucidated the mechanism of regulation of the human LPL gene by
Liver X Receptors (LXR) in hepatic cells. Previous studies in mice had shown that the
expression of LPL gene in the liver is strongly induced by high fat diets and synthetic
agonists that activate the nuclear receptors LXR and RXR (Retinoid X Receptor). In
agreement with these findings, we showed that treatment of HepG2 cells or primary
mouse hepatocytes with the LXR synthetic agonist T0901317 upregulated the
expression of LPL gene in mRNA and protein levels. Moreover, the nuclear receptors
LXRα and RXRα transactivated strongly the human LPL promoter in response to
their ligands in HepG2 cells and deletion analysis of the human LPL promoter
established that the minimal region required for LXR/RXR transactivation was the -
109/-28, which involves the FOXA2 binding site. Interestingly, we demonstrated very
weak binding of nuclear receptors LXRα and RXRα to the proximal human LPL
promoter, using chromatin immunoprecipitation assays, suggesting that additional
factors are required for LXR action. Silencing of the endogenous FOXA2 gene using
a specific siRNA in HepG2 cells and in mouse primary hepatocytes caused an
inhibition of the oxysterol-inducible expression of LPL gene at both the mRNA and
protein levels. Importantly, insulin, which inactivates FOXA2 via its nuclear
exclusion, reduced the oxysterol-inducible expression of LPL gene, indicating the
importance of FOXA2 in the lipid homeostasis in the liver. Next, we found that
FOXA2 and ligand-activated LXRα/RXRα transactivated the human LPL promoter in
a synergistic fashion. The mutations in the FOXA2 binding site (-47/-40) inhibited the
synergistic transactivation of the LPL promoter by FOXA2 and LXRα/RXRα. Finally,
we performed co-immunoprecipitation and GST pull down assays and established
physical interactions between FOXA2, LXRα and RXRα in vitro and in vivo. An
extended DNA binding domain (DBD) of LXRα is required for physical interactions
with the at least one of the two transactivation domains of FOXA2. In conclusion, the
findings of Part I and II suggest that the newly identified FOXA2 binding site in the
LPL promoter serves as a novel LXRE that facilitates the induction of the LPL gene
by oxysterols via FOXA2. Through an insulin-AKT-FOXA2-LPL signaling pathway the overexpression of LPL is prevented in the liver under conditions of cholesterol
overload protecting this tissue from the toxic effects of LPL.
In Part III we focused on the regulation of genes that are involved in HDL
biogenesis and the remodeling by transcription factor FOXA2 and LXRs in hepatic
cells. Previous studies had shown that mice heterozygous for FOXA2 have reduced
levels of HDL in the plasma. Using in silico analysis, we identified putative binding
sites for the FOXA2 factor in proximity with characterized LXR responsive elements
(LXREs) in the promoters of various human HDL genes encoding the lipid
transporters ABCG1, the hepatic lipase (LIPC) and the cholesteryl ester transfer
protein (CETP). In agreement with previous studies, we showed that treatment of
HepG2 cells and primary mouse hepatocytes with the synthetic LXR ligand,
T0901317, caused a strong induction of mRNA levels of ABCG1, ABCG5, ABCG8
and CETP genes. Inactivation of the FOXA2 factor by siRNA silencing or insulin in
primary mouse hepatocytes and HepG2 cells reduced the basal mRNA levels and also
the induction of ABCG5 and ABCG8 genes, by T0901317, indicating that FOXA2 is
critical for the upregulation of these lipid transporters by the LXR ligands. With
transactivation assays, we established that FOXA2 or LXRα/RXRα overexpression in
the presence of their ligands in HepG2 cells increased the activity of the promoters of
the ABCG5 and ABCG8 genes. In agreement with these findings, silencing the
expression of LXRα or LXRβ in HepG2 cells, via lentivirus expressing shRNAs
specific for each LXR isoform, reduced the mRNA levels of ABCG8 and ABCG5
genes. Furthermore, FOXA2 and ligand-activated LXRα/RXRα transactivated in a
synergistic manner the promoter of ABCG8 gene, but not the ABCG5 promoter.
Unexpectedly, FOXA2 inhibited the induction of the ABCG5 promoter by LXRs and
oxysterols in HepG2 cells, suggesting that more distal transcription factor binding
sites far from the coding regions of the genes may be involved and via DNA looping
regulate coordinately the expression of ABCG5 and ABCG8 genes. These findings
are in line with the synergistic transactivation of LPL promoter by nuclear receptors
LXRα/RXRα and the transcription factor FOXA2, as described above.
Understanding in depth the mechanisms by which lipid and glucose metabolic
pathways are interconnected in the liver by factors such as FOXA2 may open the way
to novel therapeutic strategies to increase HDL levels and protect patients with
metabolic diseases such as coronary heart disease, dyslipidemia, diabetes and the
metabolic syndrome.
|