Transcriptional analysis of the mammalian heart with special reference to its endocrine function

Abstract

Background: Pharmacological and gene ablation studies have demonstrated the crucial role of the endocrine function of the heart as mediated by the polypeptide hormones ANF and BNP in the maintenance of cardiovascular homeostasis. The importance of these studies lies on the fact that hypertension and chronic congestive heart failure are clinical entities that may be regarded as states of relative deficiency of ANF and BNP. These hormones are produced by the atrial muscle cells (cardiocytes), which display a dual secretory/muscle phenotype. In contrast, ventricular cardiocytes display mainly a muscle phenotype. Comparatively little information is available regarding the genetic background for this important phenotypic difference with particular reference to the endocrine function of the heart. We postulated that comparison of gene expression profiles between atrial and ventricular muscles would help identify gene transcripts that underlie the phenotypic differences associated with the endocrine function of the heart.

Results: Comparison of gene expression profiles in the rat heart revealed a total of 1415 differentially expressed genes between the atria and ventricles based on a 1.8 fold cut-off. The identification of numerous chamber specific transcripts, such as ANF for the atria and Irx4 for the ventricles among several others, support the soundness of the GeneChip data and demonstrates that the differences in gene expression profiles observed between the atrial and ventricular tissues were not spurious in nature. Pathway analysis revealed unique expression profiles in the atria for G protein signaling that included Galphao1, Ggamma2 and Ggamma3, AGS1, RGS2, and RGS6 and the related K+ channels GIRK1 and GIRK4. Transcripts involved in vesicle trafficking, hormone secretion as well as mechanosensors (e.g. the potassium channel TREK-1) were identified in relationship to the synthesis, storage and secretion of hormones.

Conclusion: The data developed in this investigation describes for the first time data on gene expression particularly centred on the secretory function of the heart. This provides for a rational approach in the investigation of determinants of the endocrine of the heart in health and disease.

Figure 1

Comparison of microarray and real-time PCR results. Relative fold change of 14 candidate genes are shown. The results demonstrate agreement between the microarray and RT-PCR data, although the degree of fold change can vary.

Figure 2

Functional classification of differentially expressed genes found between the atria and the ventricles with known Gene Ontology annotations. Note that the number of differentially expressed genes are plotted according to muscle type, i.e. atrium or ventricle.

Figure 3

Normalized intensity values of specific transcripts in all atrial and ventricular microarray replicates. The data is displayed in four different panels for visual simplicity. Note that atrial and ventricular microarrays are labelled as LA(n) and LV(n), and the scale of the ordinate axis is different for the upper panel due to high signal intensity values for ANF and BNP. Also note that RGS5 and Kcnj3 are represented by two distinct probe sets.

Figure 4

Paraffin-embedded, paraformaldehyde-fixed atrial tissue section stained with SGNE1-FITC antibody showing punctuated fluorescence structures suggesting adrenergic varicosities (white arrows).

Figure 5

G protein signaling pathways generated by GenMapp. Genes labeled in red indicate a higher expression level in the atria as compared to the ventricles based on a 1.8 fold threshold cut-off. Genes labeled in blue indicate a higher expression level in the ventricles as compared to the atria based on a -1.8 fold threshold cut-off. Genes labeled in white were not found in the list of statistically significant differentially expressed genes that was uploaded in GenMapp.