| 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 June 17, 2003 as doi:10.1096/fj.02-0700fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute of Clinical Pharmacology and Toxicology, Department of Clinical Pharmacology, Benjamin Franklin Medical Center, Freie Universität Berlin, Hindenburgdamm 30, 12200 Berlin, Germany
2Correspondence: Institute of Clinical Pharmacology and Toxicology, Benjamin Franklin Medical Center, Department of Clinical Pharmacology, Freie Universität Berlin, Hindenburgdamm 30, 12200 Berlin, Germany. E-mail: hans-dieter.orzechowski{at}medizin.fu-berlin.de
SPECIFIC AIMS
Genetic and phenotypic analysis of patients with congenital heart disease and mouse knockout models has revealed that the homeobox transcription factor Nkx2-5 and the zinc metalloprotease endothelin-converting enzyme-1 (ECE-1), expressed in at least three different isoforms (termed ECE-1a, -1b, -1c; see Fig. 3), are both essential for cardiac development. The aim of our study was to investigate whether a functional link between the homeobox transcription factor Nkx2-5 and the metalloprotease ECE-1 may exist, further supported by the identification of consensus sequences for Nkx2-5 binding in two of the three isoform-specific ECE-1 promoters previously cloned by our group.
PRINCIPAL FINDINGS
1. Nkx2-5 and ECE-1 isoforms are coexpressed in H9c2 cardiomyoblasts
Expression of Nkx2-5 in rat cardiomyoblast H9c2 cells was confirmed on mRNA and protein levels by RT/PCR and Western blot analysis, respectively. Isoform-specific ECE-1 mRNA expression was analyzed by RNase protection assay (RPA), demonstrating ECE-1c mRNA expression, and by RT/PCR, showing mRNA expression of ECE-1a, -1b, and -1c.
2. Nkx2-5 activates the isoform-specific ECE-1 promoters in H9c2 cardiomyoblasts
A promoter reporter construct containing 1206 bp of the ECE-1a-specific promoter was transfected in H9c2 cells. Cotransfection of an Nkx2-5 expression plasmid significantly increased luciferase activity by 1.8-fold compared with cotransfection of insertless expression plasmid. Point mutations were introduced in the Nkx2-5 consensus sequence of the ECE-1a promoter luciferase construct at position -421, which did not affect luciferase activity under basal conditions. Cotransfection of the mutated ECE-1a promoter and Nkx2-5 expression vector resulted in significantly increased luciferase activity comparable to the wild-type promoter. Analogous experiments were performed using a cloned 1282 bp fragment of the ECE-1b promoter demonstrating 5.0-fold increased luciferase activity by Nkx2-5 coexpression. Triple mutation of the Nkx2-5 consensus sequence at -1197 slightly reduced (P=0.11) the activity of the ECE-1b promoter. A similar effect was obtained after deletion of this consensus sequence. With both mutated constructs, we still found promoter activation by coexpression of Nkx2-5, but to a lesser extent (4.0- and 3.8-fold). Finally, cotransfection of a 969 bp ECE-1c promoter construct with the Nkx2-5 expression plasmid resulted in significantly increased ECE-1c promoter activity compared with cotransfection with empty control vector (2.6-fold).
3. Nkx2-5 binds to a consensus sequence in the ECE-1b promoter but not to a consensus sequence the ECE-1a promoter
To examine potential protein-DNA interactions, electromobility shift assay (EMSA) was performed using oligodeoxynucleotides (ODN) containing Nkx2-5 consensus sequences identified at positions -421 and -1197 in the ECE-1a and b promoters, respectively. Using nuclear extracts of H9c2 cells and the Nkx2-5 consensus ODN of the ECE-1b promoter EMSA resulted in three shifted bands with specificity demonstrated by competition experiments using unlabeled ODN. Analogous experiments using ECE-1a promoter Nkx2-5 consensus ODN did not yield clear band shifts, irrespective of competition conditions. Binding of recombinant Nkx2-5 to the ECE-1b, but not to the ECE-1a, consensus sequence was demonstrated by supershift analysis.
4. Nkx2-5 regulates ECE-1 in the chromatin context
To confirm our results in the context of nuclear chromatin, we generated stably transfected H9c2 cells overexpressing Nkx2-5 as shown by Western blot (Fig. 1
A). The relative expression ratio of ECE-1 isoforms was analyzed by RPA (Fig. 1B
). Only ECE-1c mRNA was detected, irrespective of the Nkx2-5 expression level, and the ECE-1c signal was apparently increased in Nkx2-5 overexpressing cells. Additional Northern blot analyses showed increases in total ECE-1 mRNA expression by 28% and 34% by stable Nkx2-5 overexpression (Fig. 1C
). Since ECE-1a and ECE-1b expression were below RPA detection level, we performed semiquantitative RT/PCR and real-time PCR analyses and found up-regulation of both isoforms in cells overexpressing Nkx2-5. ECE-1 isoform expression levels were concordant with corresponding promoter activities in stably transfected cells. Regardless of Nkx2-5 expression, the ECE-1c-specific promoter was 
eightfold more active than the ECE-1a and ECE-1b promoters. To analyze the regulatory mechanisms in a cell system without endogenous expression of Nkx2-5, we performed analogous experiments in endothelial EA.hy926 cells stably transfected with Nkx2-5 (Fig. 1D-F
). Using RPA, up-regulation of ECE-1b and ECE-1c mRNA was clearly demonstrated (Fig. 1E
). Northern blot analyses of two independently generated EA.hy926 cell lines revealed up-regulation of ECE-1 mRNA by 89% and 65% by Nkx2-5 overexpression (Fig. 1F
).
|
CONCLUSIONS AND SIGNIFICANCE
The homeobox transcription factor Nkx2-5 and the metalloprotease ECE-1 represent essential factors in cardiac development. Until now a functional link between these two genes has not been reported. We investigated a possible interaction in the cardiomyoblast cell line H9c2 that we have shown to constitutively express Nkx2-5 and ECE-1 isoforms. Our study demonstrates for the first time that Nkx2-5 transcriptionally regulates the expression of ECE-1 isoforms. Our results provide clear evidence that Nkx2-5 can transcriptionally activate the expression of all three ECE-1 isoforms (Fig. 2
).
|
Our data support an indirect mode of interaction in case of the ECE-1a and the ECE-1c promoter since a specific protein–DNA interaction was not detected (ECE-1a) or a consensus sequence for Nkx2-5 binding was not identified (ECE-1c).
We also demonstrated an indirect mechanism of ECE-1b promoter activation by Nkx2-5 coexpression in that a promoter mutant with a deletion of the complete consensus sequence at -1197 maintained strong inducibility by Nkx2-5 coexpression. A direct mechanism of ECE-1b promoter activation mediated by the -1197 cis-element has to be assumed for several reasons. First, promoter constructs carrying the deletion or a triple mutation of the Nkx2-5 consensus sequence showed only 80% inducibility of the wild-type promoter (3.8-/4.0-fold vs. 5.0-fold). Second, under basal conditions the triple mutation showed an 11% reduction compared with the wild-type promoter (P=0.11). Third, these functional promoter data are supported by our EMSA results. We demonstrated strong binding of H9c2 nuclear proteins and of recombinant Nkx2-5 to the -1197 consensus sequence in the ECE-1b promoter. The finding of increased ECE-1 expression (with regard to total and isoform-specific mRNA levels) in cardiomyoblast and endothelial cell lines overexpressing Nkx2-5 clearly indicates that the activating effects of Nkx2-5 are functional in the context of nuclear chromatin. In conjunction with these mRNA expression data, our functional promoter assays argue in favor of a transcriptional mechanism of ECE-1 isoform expression by Nkx2-5, in contrast to altered mRNA stability.
Several published reports show expression of either Nkx2-5 or ECE-1 in cardiomyocytes in situ, which, in conjunction with the functional data presented here, supports regulation of ECE-1 by Nkx2-5 in vivo. Transcriptional regulation of the ECE-1 gene by Nkx2-5 may be of developmental significance. Mouse knockout studies revealed that a single Nkx2-5 allele is necessary and sufficient for normal cardiac development (i.e., cardiac looping) at an early stage where a role for ECE-1 has not been identified. In contrast, the cardiac phenotype of patients with hemizygous Nkx2-5 gene mutations indicates that both functional alleles are required for normal cardiac development in a later stage. These patients can present with conotruncal malformations and/or abnormalities of the cardiac conduction system. Similar conotruncal malformations are observed in ECE-1-deficient mice, and an essential role for cardiac ECE-1 has been shown for Purkinje fiber differentiation. These observations support the notion that an interaction between Nkx2-5 and ECE-1 is important for the development of the cardiac outflow tract and the conducting system. Our data provide the molecular explanation of such an interaction.
In addition, the interaction between Nkx2-5 and ECE-1 may play a role in adult cardiac disease. Up-regulation of cardiac ECE-1 in human and experimental heart failure was reported by several groups. There is also evidence of Nkx2-5 up-regulation in congestive heart failure. Besides quantitative alterations in gene expression, ischemic heart failure can cause a shift in cellular localization of ECE-1 from nonmyocyte cells to cardiomyocytes, which is consistent with the exclusively myocytic expression of Nkx2-5 in the failing heart. The link between Nkx2-5 and ECE-1 presented here might therefore also be crucial with regard to congestive heart failure. The precise temporal and spatial pattern of Nkx2-5 and ECE-1 isoform expression during the development of heart failure remains to be investigated. Based on previous reports on the beneficial effects of endothelin blockade in chronic heart failure, a future anti-Nkx2-5 therapy may be a novel therapeutic option for this fatal disease.
Alternative promoters and differential splicing constitute two basic mechanisms to generate a transcriptome more complex than predicted from the number of genes. Alternative promoters are implicated in temporal (stimulus-dependent or developmental) and spatial (cell-specific) control of gene expression. These important regulatory mechanisms, in conjunction with different subcellular localizations of the expressed protein isoforms associated with alternative promoters, apply to ECE-1, which therefore represents a model gene in the so-called "postgenomic era."
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0700fje; doi: 10.1096/fj.02-0700fje 
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
