| 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 22, 2005 as doi:10.1096/fj.04-3064fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
33
,2
* Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri, USA;
Department of Cardiology, Children’s Hospital, Boston, Massachusetts, USA;
Cardiology Division, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA; and
Department of Experimental Therapeutics, Translational Research Center, Kyoto University School of Medicine, Kyoto, Japan
1Correspondence: Washington University School of Medicine, Departments of Pediatrics and Genetics, 660 S. Euclid Ave., Box 8208, St. Louis, MO 63110, USA. E-mail: jay_p{at}kids.wustl.edu
SPECIFIC AIMS
Heterozygous mutations of transcription factors cause a number of known human diseases, including cancer, birth defects, and syndromes. The identification of genes that are "direct" transcriptional targets is necessary to elucidate the molecular pathogenesis. Two forms of evidence are commonly offered to prove that a transcription factor directly regulates a gene. First, the suspected target must be abnormally expressed in a cell or tissue depleted of the factor, as in a homozygous knockout mouse. Second, the promoter of the putative target gene must contain binding sites for the transcription factor and transactivate a reporter gene construct when coexpressed with the transcription factor in cell culture or in transgenic animals. These two experimental strategies have yielded important insights, but generalizations about mechanisms of disease may be flawed because the results are obtained outside the physiologic range of transcription factor concentration.
Nkx2-5 is an evolutionarily conserved transcription factor that regulates cardiac development from fly to man. Homozygous null mutations cause arrest of cardiac development at the looped heart tube stage and embryonic lethality in mouse models. Heterozygous mutations in mice and humans cause congenital cardiac malformations and atrioventricular conduction defects. Therefore, genes that Nkx2-5 regulates are of interest. Connexin 40, a gap junction protein, was suspected as a candidate for the conduction defect because its promoter contains Nkx2-5-binding sites that can transactivate a heterologous reporter gene construct and because Connexin 40 knockout mice have conduction defects also. However, direct quantification of Connexin 40 protein in the Purkinje cells of wild-type and Nkx2-5+/– mice revealed identical levels of expression. Rather than abnormal Connexin 40 expression, the conduction defect in Nkx2-5 haploinsufficiency likely results from a developmental defect leading to anatomic hypoplasia of the conduction system. The Connexin 40 measurements in Nkx2-5 haploinsufficient mice led to a broader question regarding the physiologic validity of direct gene targets identified outside the normal range of transcription factor activity.
Thus, the specific aim of this work was to measure the expression of four well-described targets of the cardiac transcription factor Nkx2-5 in haploin-sufficient animals during embryonic development and the healthy and failing adult heart. Abundant data define atrial natriuretic factor (ANF), brain natriuretic peptide (BNP), the RNA helicase Csm, and the homeodomain only protein (HOP) as direct targets of Nkx2-5, based on their absence of expression in null mutant hearts and promoter activity in transactivation assays. Roles for each of these genes have been shown in cardiac development, growth, or function. ANF and BNP inhibit cardiac hypertrophy and fibrosis, respectively, in the adult mouse. Csm potentiates the hypertrophic response of cultured cardiac myocytes to phenylephrine, and HOP regulates embryonic cardiac growth. Despite the evidence that Nkx2-5 directly regulates the expression of these genes, the quantitative measurements of wild-type and Nkx2-5 haploinsufficient hearts indicate that the relationship between transcription factor dosage and target gene expression in vivo is variable and depends upon the specific gene and physiologic context.
PRINCIPAL FINDINGS
1. Nkx2-5 protein levels are significantly reduced in Nkx2-5 haploinsufficiency
One could reasonably presume, based on the presence of heterozygous phenotypes, that Nkx2-5 protein is quantitatively reduced in genetic haploinsufficiency, but this has not been directly shown previously. Western blot analyses of nuclear protein fractions from whole heart tissue show that Nkx2-5 protein in Nkx2-5+/– animals is probably no more than half the wild-type level (Fig. 1
). Nkx2-5 is undetectable in the cardiac cytosolic fraction and in the liver, as expected. Sp1, a ubiquitous transcription factor, is equally expressed in Nkx2-5+/– and wild-type hearts. Endogenous Nkx2-5 protein was undetectable unless most of the cytosolic fraction was eliminated, as shown by GAPDH immunoblot analysis.
|
2. Target gene expression in the embryonic heart is not affected by Nkx2-5 haploinsufficiency
The expression of ANF, BNP, Csm, and HOP in the ventricular myocardium of homozygous Nkx2-5 null mutant embryos is markedly down-regulated or absent. In contrast, no significant difference was detected in the expression of any of these genes between wild-type and heterozygous null mutant embryos from E12.5 through 17.5; representative data from E13.5 are shown (Fig. 2
A). Calibration with a standard sample indicated that we could resolve at least a 25% difference in the four putative target genes and GAPDH. Any undetected difference resulting from Nkx2-5 haploinsufficiency would thus be small.
|
3. Nkx2-5 haploinsufficiency does not affect the expression of ANF or BNP in the healthy and failing adult heart
In the adult heart, ANF and BNP are normally expressed at very low levels and significantly up-regulated in heart failure. In a mouse model of myocardial infarction, both wild-type and Nkx2-5+/– animals developed moderate dilatation of the left ventricle and diminished systolic function one month after ligation of the left anterior descending coronary artery. ANF and BNP were undetectable or expressed at very low levels in sham-operated mice and up-regulated to the same extent in the noninfarcted, failing left ventricular myocardium of wild-type and Nkx2-5+/– animals (Fig. 2B
). ANF and BNP expression, therefore, exhibits no dependence on Nkx2-5 in a physiologic range of transcription factor activity in the embryo or adult.
4. Nkx2-5 haploinsufficiency causes half-normal expression of Csm and HOP in the healthy and failing adult heart
In contrast to ANF and BNP, Csm and HOP were down-regulated in heart failure, which has not been previously reported to our knowledge. Both Csm and HOP were expressed at about half of the wild-type levels in healthy (sham-operated) and failing (postinfarction) Nkx2-5+/– ventricular myocardium (Fig. 2B
). Down-regulation of Csm and HOP in the Nkx2-5+/– heart and further down-regulation in failing wild-type and Nkx2-5+/– hearts raise interesting questions about the cellular levels of the respective mRNA and protein species reminiscent of our past observations with Cx40 and about the functional implications, but they are beyond the scope of the present work.
CONCLUSIONS AND SIGNIFICANCE
A gene is usually considered a direct target of a transcription factor based on experiments that show differential expression between either none and normal or normal and supraphysiologic levels of the transcription factor. The conclusion is then commonly extrapolated to explain mutant phenotypes where the activity of the transcription factor is around half-normal. The results of this work, which examines Nkx2-5+/– and wild-type hearts, suggests that the common conception greatly oversimplifies any model of gene regulation in vivo. Genes that depend upon Nkx2-5 concentration in one circumstance may not in another. For example, the embryonic heart tube has an absolute requirement for Nkx2-5 to express ANF and BNP, but later stage embryonic hearts and adult Nkx2-5+/– hearts express normal amounts. Paradoxically, conditional deletion of Nkx2-5 in the adult ventricle has been associated with up-regulation of ANF and BNP, which further demonstrates the inconsistent correlation between the expression of these two genes and Nkx2-5. A linear, one-component model consisting of Nkx2-5 acting on a target gene cannot explain the regulation of the four putative target genes across a range of biological conditions (Fig. 3
).
|
Additional factors that are gene-specific and context-dependent must contribute to the regulation of target gene expression and can negate any dependence on the transcription factor in its physiologic range of activity. For example, the different dependences of Csm and HOP expression on Nkx2-5 in embryonic and adult hearts indicate that mechanisms associated with embryonic development circumvent the deficiency of Nkx2-5. One could envision a scenario in which a cofactor acting stoichiometrically with Nkx2-5 could regulate expression of Csm and HOP in the adult but not embryonic heart. Nkx2-5-independent pathways, nonlinear or stochastic Nkx2-5-dependent mechanisms, or both must lead to the variable nature of target gene expression in Nkx2-5+/– animals. When such ad hoc regulatory modules are necessary, the unqualified concept of a direct transcriptional target becomes too vague to explain the mechanism of mutant phenotypes that develop under specific conditions or stages of development.
The phenotypes of animals that have none or supranormal levels of a transcription factor are most commonly reported and molecular sequelae are probably studied because experimental results are most obvious in the extremes. Such experiments, however, do not accurately predict the expression patterns of ANF, BNP, Csm, and HOP in the physiologically relevant setting of Nkx2-5 haploinsufficiency in vivo. Thus, the main purpose of experiments that measure gene expression outside the physiologic range of transcription factor activity may be to suggest prototypical forms of gene regulation or candidate effector genes for mutant phenotypes. It is beyond the scope of this work to identify specific regulatory mechanisms for each of the four genes examined, but the existence of heterozygous transcription factor diseases and mutant phenotypes clearly implies that the expression of some genes must depend on transcription factor level. The unpredictable nature of putative target gene expression in Nkx2-5+/– hearts under diverse conditions underscores the need to study heterozygous mutants directly to understand the mechanism of transcription factor diseases in general.
FOOTNOTES
2 Current address: Novartis Institutes for Biomedical Research, 100 Technology Sq., Cambridge, MA 02139, USA. 
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/1096/fj.04-3064fje;
This article has been cited by other articles:
|
|
 |
|
  Q. Li, Y. Guo, W. Tan, Q. Ou, W.-J. Wu, D. Sturza, B. Dawn, G. Hunt, C. Cui, and R. Bolli Cardioprotection Afforded by Inducible Nitric Oxide Synthase Gene Therapy Is Mediated by Cyclooxygenase-2 via a Nuclear Factor-{kappa}B Dependent Pathway Circulation, October 2, 2007; 116(14): 1577 - 1584. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, down-regulated relative to wild type in the same physiologic condition. 
, up-regulated. •, low or undetectable expression. Change in parentheses denotes expression of failing wild-type ventricular myocardium relative to healthy, sham-operated control.