| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 2, 2004 as doi:10.1096/fj.03-1118fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institut National de la Santé et de la Recherche Médicale U589, Hormones, Facteurs de Croissance et Physiopathologie Vasculaire, Institut Louis Bugnard IFR31, Hôpital Rangueil, Toulouse, France;
Department of Stress Science, Institute of Health Biosciences, University of Tokushima Graduate School, Tokushima, Japan; and
# MilleGen, Prologue Biotech, Labège, France
1Correspondence: Institut National de la Santé et de la Recherche Médicale U589, Hormones, Facteurs de Croissance et Physiopathologie Vasculaire, Institut Louis Bugnard, IFR31, CHU Rangueil, 31059 Toulouse Cedex 9, France. E-mail: anne-catherine.prats{at}toulouse.inserm.fr
SPECIFIC AIMS
Internal ribosome entry site (IRES) elements allow a subtle regulation of gene expression at the translational level particularly in response to stress. The specific aim of our study was to explore translation control in diabetic conditions (hyperglycemia). We focused on the FGF-2 gene, a major angiogenic growth factor implicated in diabetes-associated vascular complications of atherosclerosis and whose mRNA contains an IRES and different initiation codons of translation giving rise to several FGF-2 isoforms.
PRINCIPAL FINDINGS
1. Hyperglycemia generates a specific activation of FGF-2 IRES-dependent translation in transgenic mice aorta
IRESs are RNA structural elements present in 5' untranslated regions of a small number of mRNAs, principally coding for control proteins (growth factors and transcription factors), which mediate a nonclassical cap-independent translation initiation mechanism. It has been shown in transfected cells that IRESs can mediate translational activation of gene expression in response to stress (e.g., heat shock, amino acid deprivation, and apoptosis) when cap-dependent translation is blocked.
To study in vivo FGF-2 IRES regulation, we developed transgenic mice that express bicistronic mRNA containing two open reading frames (ORFs) encoding two distinct luciferases, Renilla luciferase (LucR) and firefly luciferase (LucF) (Fig. 1
A). In these conditions, translation of LucF is driven by FGF-2 IRES. We addressed pathophysiological regulation of FGF-2 IRES in response to hyperglycemia by using a streptozotocin (STZ)-induced diabetic model. Four organs were tested for FGF-2 expression: eye, heart, and aorta, which are typical target organs of hyperglycemia, and brain in which FGF-2 IRES activity is very high and where FGF-2 plays a critical role.
|
As shown in Fig. 1B
, STZ-induced hyperglycemia affected LucR reporter activity in the aorta, heart, and eye. LucR activities were correlated with levels of reporter mRNA in all organs tested (Fig. 1D
), indicating that LucR activity reflects its mRNA levels. mRNA quantification using LucR primers was always equal to that using LucF primers, indicating the bicistronic status of mRNA in all conditions. In contrast to the results of LucR activity, FGF-2 IRES-dependent LucF activity was independent on the level of its mRNA (Fig. 1C, D
). In the aorta, reporter mRNA decreased significantly during hyperglycemia in a time-dependent manner. However, LucF activity increased until wk 3 (243% of control mice) and decreased to basal level (Fig. 1C
). In contrast, LucF activity decreased selectively in the brain and was not affected significantly in the eye during hyperglycemia (Fig. 1C
). Only the heart showed a close relationship between levels of reporter mRNA and reporter activities (Fig. 1B-D
). These data indicate that hyperglycemia affects FGF-2 IRES activity in a tissue-specific manner.
2. Hyperglycemia induces aorta-specific translational activation of endogenous FGF-2 expression
To further examine relationships between hyperglycemia-associated translational activity and endogenous FGF-2 protein expression, we analyzed levels of FGF-2 mRNA and protein (Fig. 2
). In the aorta (Fig. 2A-C
), hyperglycemia significantly increased FGF-2 protein levels, with kinetics similar to that observed for IRES activity in Fig. 1
. Both endogenous FGF-2 protein expression and IRES activation disappeared at wk 7, and this feature was specific to aorta. In contrast, FGF-2 mRNA expression was slightly affected by hyperglycemia, indicating that hyperglycemia-induced FGF-2 expression in the aorta could be regulated mainly by IRES-dependent translation.
|
In the heart, FGF-2 protein levels slightly decreased in association with a decrease in its mRNA during hyperglycemia (Fig. 2D-F
). In the eye levels of FGF-2 protein and mRNA both increased during hyperglycemia (Fig. 2G-I
), suggesting that FGF-2 expression in these organs might be regulated transcriptionally.
In the brain, FGF-2 protein levels showed no change (Fig. 2J, K
), although mRNA significantly increased during hyperglycemia (Fig. 2L
). Absence of correlation between FGF-2 mRNA and protein levels suggests that FGF-2 protein expression might be inhibited at the translational level.
In conclusion, translational changes of FGF-2 expression induced by hyperglycemia were observed in the aorta and the brain. Translational activation of endogenous FGF-2 in aorta fully correlated with activation of FGF-2 IRES.
3. Glucose stimulates biosynthesis of FGF-2 protein in an IRES-dependent mechanism
To further characterize hyperglycemia-mediated translational control of FGF-2 expression in aortic tissue, we isolated aortic smooth muscle cells (SMC) from nondiabetic transgenic mice. SMC were subjected to normal (5.5 mM) and high (25 mM) glucose for 6 d. FGF-2 IRES-dependent LucF activity increased in 180% of control by high glucose, whereas LucR activity decreased to 58% of control. High glucose reduced overall protein synthesis activity to 57% of control cells, whereas biosynthesis and total levels of endogenous FGF-2 protein increased to 220% and 280% of control by high glucose. These results indicated that high glucose-induced FGF-2 protein accumulation could be regulated translationally and that FGF-2 IRES allowed efficient translation under high glucose conditions. We further investigated whether this property could be shared by other cellular IRESs: we compared the effect of glucose on IRES activities from FGF-2, PDGF and c-myc. LucF activity from FGF-2 IRES was significantly increased by high glucose, and activity from other IRESs maintained the level seen in normal glucose conditions. In contrast, LucR activity was significantly reduced to around 70% of control. Maintenance of translational activity may be a general property of cellular IRES contained in glucose-induced genes whereas activation seems specific to FGF-2 IRES.
CONCLUSIONS AND SIGNIFICANCE
We have shown that FGF-2 expression is translationally activated by an IRES-dependent mechanism in response to hyperglycemia in mice aorta in vivo and in SMC ex vivo. To the best of our knowledge, this is the first in vivo example of an IRES-dependent translational activation of gene expression in pathophysiological conditions. The present report shows that hyperglycemia-induced FGF-2 is regulated translationally in a tissue-specific manner and that FGF-2 expression in aorta and brain is dependent on FGF-2 IRES activity, but not on its mRNA level (Figs. 1
, 2)
.
Our data allows us to propose a schema of the translational response to diabetes occurring in aorta (Fig. 3
). Hyperglycemia may generate suppression of cap-dependent, global translation that leads to reduction of overall gene expression (Fig. 3
, left panel). This process concerns at least vascular SMC, which is the main cell type in this tissue, but we cannot rule out that endothelial cells or, eventually, macrophages may be concerned by such a regulation. Hyperglycemia may induce pathological activation of FGF-2 IRES through an induction and/or activation of FGF-2 IRES trans-acting factors (ITAFs) in the aorta but not in other organs tested (Fig. 3
, right panel). Whereas ITAFs described up to now for other IRESs are mainly activators, we have previously shown that FGF-2 IRES is destabilized and inactivated by p53. Nevertheless, induction of activating ITAFs or removal of inhibitory ITAFs effect could generate an increase of FGF-2 IRES-dependent translation, leading to overproduction of FGF-2.
|
Several lines of evidence show an important in vivo implication of FGF-2 in the pathogenesis of vascular dysfunction induced by hyperglycemia, including vascular hyperpermeability and hemodynamic change, and increased aortic smooth muscle growth response to vascular injury. We propose that pathological IRES activation, generating FGF-2 overproduction, might be an important parameter in diabetes-associated vascular complications such as atherosclerosis and neovascularization. These and previous data enable us to affirm that translational regulation in vessels is more general and may be a key phenomenon in the link between diabetes and atherosclerosis. A previous report has shown that another gene, CD36 scavenger receptor, contributing to accelerated atherosclerosis in response to glucose, is translationally activated in vascular wall macrophages. Glucose sensitive translation is mediated by CD36 mRNA 5'UTR and involves translation of small upstream open reading frames (uORF). uORF can be coupled with IRES-dependent translation: activation of arginine/lysine transporter cat-1 mRNA IRES in response to amino-acid deficiency is mediated by translation of a uORF responsible for mRNA leader remodeling. Maintenance of translation efficiency under high glucose conditions and glucose up-regulated translation of specific gene suggest that the mechanism of IRES-mediated translational activation in response to glucose may be able to regulate several genes involved in the pathophysiology of diabetes.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-1118fje;
This article has been cited by other articles:
|
|
 |
|
  S. Bornes, L. Prado-Lourenco, A. Bastide, C. Zanibellato, J. S. Iacovoni, E. Lacazette, A.-C. Prats, C. Touriol, and H. Prats Translational Induction of VEGF Internal Ribosome Entry Site Elements During the Early Response to Ischemic Stress Circ. Res., February 16, 2007; 100(3): 305 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Zhang, H. Cao, and G. N. Rao Fibroblast Growth Factor-2 Is a Downstream Mediator of Phosphatidylinositol 3-Kinase-Akt Signaling in 14,15-Epoxyeicosatrienoic Acid-induced Angiogenesis J. Biol. Chem., January 13, 2006; 281(2): 905 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. HOLCIK, T. GRABER, S. M. LEWIS, C. A. LEFEBVRE, E. LACASSE, and S. BAIRD Spurious splicing within the XIAP 5' UTR occurs in the Rluc/Fluc but not the {beta}gal/CAT bicistronic reporter system RNA, November 1, 2005; 11(11): 1605 - 1609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Miura, J. Thompson, J. V. Chakkalakal, M. Holcik, and B. J. Jasmin The Utrophin A 5'-Untranslated Region Confers Internal Ribosome Entry Site-mediated Translational Control during Regeneration of Skeletal Muscle Fibers J. Biol. Chem., September 23, 2005; 280(38): 32997 - 33005. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. Komar and M. Hatzoglou Internal Ribosome Entry Sites in Cellular mRNAs: Mystery of Their Existence J. Biol. Chem., June 24, 2005; 280(25): 23425 - 23428. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
