Cardiac expression of Brn-3a and Brn-3b POU transcription factors and regulation of Hsp27 gene expression

Abstract

The Brn-3 family of transcription factors play a critical role in regulating expression of genes that control cell fate, including the small heat shock protein Hsp27. The aim of this study was to investigate the relationship between Brn-3a and Brn-3b and Hsp27 expression in the developing rodent heart. Brn-3a and Brn-3b were detected from embryonic days 9.5-10.5 (E9.5-E10.5) in the mouse heart, with significant increases seen later during development. Two isoforms (long and short) of each protein were detected during embryogenesis and postnatally. Brn-3a messenger RNA (mRNA) and protein were localized by E13.0 to the atrio-ventricular (AV) valve cushions and leaflets, outflow tract (OFT), epicardium and cardiac ganglia. By E14.5, Brn-3a was also localised to the septa and compact ventricular myocardium. An increase in expression of the long Brn-3a(l) isoform between E17 and adult coincided with a decrease in expression of Brn-3b(l) and a marked increase in expression of Hsp27. Hearts from Brn-3a-/- mice displayed a partially penetrant phenotype marked by thickening of the endocardial cushions and AV valve leaflets and hypoplastic ventricular myocardium. Loss of Brn-3a was correlated with a compensatory increase in Brn-3b and GATA3 mRNA but no change in Hsp27 mRNA. Reporter assays in isolated cardiomyocytes demonstrated that both Brn-3a and Brn-3b activate the hsp27 promoter via a consensus Brn-3-binding site. Therefore, Brn-3 POU factors may play an important role in the development and maintenance of critical cell types and structures within the heart, in part via developmental regulation of myocardial Hsp27 expression. Furthermore, Brn-3a may be necessary for correct valve and myocardial remodelling and maturation.

Fig. 1

Analysis of Brn-3a protein expression in heart and isolated cardiomyocytes. a immunoblot of Brn-3a protein in cultured neonatal rat ventricular cardiomyocytes (NRVM) and neonatal rat heart (NRH). Lane 1 Purified protein from Jurkat cells (lane 2), rat brain tissue, NRVM (lane 3) and NRH (lane 4) at stage P1 were blotted and probed with a monoclonal anti-Brn-3a antibody. The samples were also probed with a polyclonal anti-actin antibody to assess variation in protein loading. b NRVM cultured for 2 days, and Brn-3a(l) and Brn-3b expression assessed by Western blotting. NRVM at day 0 (lane 1) was compared to NRVM after 2 days in culture (lane 2), NRH at P1, adult rat heart (ARH) and isolated adult rat ventricular cadiomyocytes (ARVM). Blots were also probed for pan-myosin light chain (MLC) and troponin I (TnI) for comparison. C Immunohistochemical analysis of Brn-3a (1) and Brn-3b (2) in sections of adult rat heart

Fig. 2

Levels of total Brn-3a and Brn-3b protein and mRNA in the heart during development. Immunoblots of both the long and short isoforms of Brn-3a and Brn-3b in embryonic (1) and neonatal (2) heart tissue samples are shown in a and b, respectively. Protein levels were quantified using laser densitometry and normalised to that of actin (c and d). Positive controls were brain (Brn) and expression of the long form of Brn-3a expressed in cells from a transfected cDNA (+ve). Values were obtained from three experiments carried out using independent heart samples from separate litters, and data is presented as mean ± SD. Brn-3a(l) mRNA levels were quantified by qPCR using primers specific for exon 1 and normalised to that of GAPDH (en = 5 hearts per time point). Statistical analysis was carried out using the students t test, and significance is expressed as asterisks where **p < 0.005 and where *p < 0.05

Fig. 3

Expression of Brn-3a protein and mRNA in different cardiac structures at E14.5. Immunohistochemistry was carried out using a monoclonal Brn-3a antibody on transverse sections (8 μm) of E14.5 embryos to show sites of expression. a Nuclear Brn-3a expression is visible in isolated cells of the ventricular myocardium (black arrowheads) and epicardium. The lower panel shows negative control with secondary antibody alone. Expression of Brn-3a protein in the endocardium surrounding the mitral and tricuspid valves (arrowheads, b) and in a layer of cells beneath the endocardial layer. The lower panels show negative control. Brn-3a expression is also noted in the semilunar valve leaflets (arrowheads, e). Ao aorta. Brn-3a expression in the cardiac innervation was compared with the distribution of Neurofilament protein in the same region to confirm Brn-3a expression in the cardiac ganglia. f Prominent Brn-3a protein expression is seen in the cardiac ganglia (black arrowheads), as seen with the mRNA expression; the DRG’s exhibit strong Brn-3a protein expression. The lower panel shows negative control carried out with secondary antibody alone. Strong expression of neurofilament protein is apparent in the neuronal processes of the cardiac innervation and in the DRG (arrowheads) and neurons in the spinal cord. Br bronchus, Oe oesophagus. Slide in situ hybridisation using Brn-3a-specific DIG-labelled riboprobes on transverse sections (8 μm) of E12.5–E14.5 embryos. Expression of Brn-3a mRNA was observed in the ventricular myocardium of the apex at E14.5, the muscular interventricular septum and myocytes adjacent to the epicardium (c), the endocardium surrounding the mitral valve (left), tricuspid valve (right) and a layer of cells beneath the endocardial layer (arrowheads, d). Negative (sense) controls for the corresponding sections of the heart are shown. LA left atrium, MV mitral valve, TV tricuspid valve

Fig. 4

Expression of Brn-3a mRNA in different cardiac structures. Slide in situ hybridisation using Brn-3a-specific DIG-labelled riboprobes on transverse sections (8 μm) of E12.5–E14.5 embryos. Expression of Brn-3a mRNA was observed in the developing endocardial cushions (arrowheads) at E12.5 (a), the epicardium and the semilunar valve leaflets (arrowheads, b) at E13.0, the aorta and pulmonary trunk and pulmonary valve leaflets (arrowheads, c) at E14.5, the individual preganglionic nerves and ganglia (arrowheads, d) at E14.5. Negative (sense) controls for the corresponding sections of the heart are shown. Ao aorta, dAo descending aorta, LA left atrium, RA right atrium, CCV common cardinal vein, PT pulmonary trunk, LL left lung, RL right lung, IVS interventricular septum, LV left ventricle, RV right ventricle, VC vena cava, Br and asterisk bronchioles, T trachea, MV mitral valve, TV tricuspid valve

Fig. 5

Analysis of Hsp27 protein levels in hearts during development. a Representative mmunoblots of Hsp27 and actin protein in heart samples during development. b Band densities were obtained by laser densitometry from three independent sets of protein samples and the Hsp27 values normalized to actin. Hsp27/actin ratios are expressed as mean ± SD, and statistical analysis was conducted using the t test and is expressed as asterisks where **p < 0.005 versus E14

Fig. 6

Comparison of the hearts of Brn-3a null and wild-type mice at E14.5 (abnormal phenotype). Brn-3a null and wild-type E14.5 embryos were sectioned (8 μm) transversely and subjected to H & E staining to examine the morphology of their hearts. Wild-type (+/+) mice are shown in the left panel and Brn-3a null (−/−) mice are shown in the right panel. The position of the endocardial cushions are indicated by arrows. In this example, the Brn-3a null mouse exhibits thickened endocardial cushions with increased cell density

Fig. 7

Comparison of Brn-3b mRNA levels in Brn-3a KO and wild-type hearts. RT-PCR (a) and quantitative real-time PCR (qRT-PCR; b–d) of hearts isolated from wild-type or Brn-3a null mice at E14.5 and E15. Levels of cyclophilin (a) or GAPDH (b–d) mRNA were measured to control for variations in cDNA levels between samples. b A typical Brn-3b mRNA standard curve used to quantify the Brn-3b mRNA. c Typical cycle curves, with the Brn-3b mRNA levels being measured during the linear phase of the reaction at the point marked on the graph. d The mean values in femtograms (fg) for each sample normalized to GAPDH. Values were obtained in triplicate for each heart with an n = 4 hearts for E14.5 and n = 2 hearts for E15.5. Data is expressed as mean ± SD, and statistical analysis was conducted using the Student’s t test and is expressed as **p < 0.005

Fig. 8

Mutation of the putative Brn3-binding site results in a loss of activation of the Hsp27 promoter in cardiomyocytes. a The 200-bp proximal region of the Hsp27 promoter showing the major transcriptional regulatory elements including the TATA, HSE and ERE, and the putative Brn-3-binding site. b Comparison of the Brn-3-binding site found in the Hsp27 promoter with published Brn-3 consensus binding sites. c Activation of the wild-type Hsp27-luciferase reporter by Brn-3a(l), Brn-3a(s) and Brn-3b(l) (open bars) and the loss of activation following mutation of the putative Brn-3-binding site (filled bars). The Hsp27 promoter alone and empty pLTR vectors were used as controls. Luciferase activation was expressed as percent of pLTR, which was taken as 100%. Luminescence was determined from three independent experiments carried out in triplicate, and results are expressed as mean ± SD. Statistical analysis was carried out using the one-way ANOVA and is expressed as **p < 0.005 and as *p < 0.05 versus pLTR, and as ††p < 0.005 and as †p < 0.05 versus the WT promoter

Fig. 9

Levels of Hsp27 and GATA3 mRNA in wild-type and Brn-3a−/− hearts at P0. Hsp27 (a) and GATA3 (b) mRNA levels were quantified by qPCR and normalised to that of GAPDH (en = 5 hearts for each genotype)