A calcineurin-dependent transcriptional pathway for cardiac hypertrophy

Abstract

In response to numerous pathologic stimuli, the myocardium undergoes a hypertrophic response characterized by increased myocardial cell size and activation of fetal cardiac genes. We show that cardiac hypertrophy is induced by the calcium-dependent phosphatase calcineurin, which dephosphorylates the transcription factor NF-AT3, enabling it to translocate to the nucleus. NF-AT3 interacts with the cardiac zinc finger transcription factor GATA4, resulting in synergistic activation of cardiac transcription. Transgenic mice that express activated forms of calcineurin or NF-AT3 in the heart develop cardiac hypertrophy and heart failure that mimic human heart disease. Pharmacologic inhibition of calcineurin activity blocks hypertrophy in vivo and in vitro. These results define a novel hypertrophic signaling pathway and suggest pharmacologic approaches to prevent cardiac hypertrophy and heart failure.

Figures

Figure 1. Interactions between GATA4 and NF-AT3 in the Two-Hybrid System

(A) Schematic diagrams of GATA4 and NF-AT3 proteins. The portion of GATA4 used as bait in the two-hybrid system encompassed amino acids 130–409 and is shown beneath the full-length protein. The portion of NF-AT3 recovered in the yeast two-hybrid screen spanned amino acids 522–902. The Rel homology domain (RHD) extends from amino acids 404–694 and the conserved phosphorylation domain from 145–275. (B) Amino acids 522–902 of NF-AT3 were fused in-frame to the GAL4 DNA-binding domain (DBD) and used as bait in a two-hybrid assay in transfected 10T1/2 cells.

Figure 2. Coimmunoprecipitation of GATA4 and NF-AT3

(A) NF-AT3 with a Flag epitope tag and GATA4, as indicated, were translated in a rabbit reticulocyte lysate in the presence of 35S-methionine. Anti-Flag antibody was then used for coimmunoprecipitation assays. Proteins were resolved by SDS-PAGE. Lanes 1 and 2 show the in vitro translation products. Lanes 3–5 show the coimmunoprecipitations. The anti-Flag antibody selectively immunoprecipitates NF-AT3 (lane 3) but does not recognize GATA4 (lane 4). However, when NF-AT3 is mixed with GATA4, GATA4 is coimmunoprecipitated (asterisk in lane 5). (B) The domain of GATA4 that interacts with NF-AT3 was mapped by translating NF-AT3-Flag with a series of GATA4 deletion mutants (shown in [C]), followed by coimmunoprecipitation. Even-numbered lanes show the GATA4 or GATA6 (lane 14) deletion mutants loaded directly on the gel. The other lanes contain the indicated GATA4 or GATA6 (lane 15) deletion mutants plus NF-AT3 and were immunoprecipitated with anti-Flag antibody. All GATA4 deletion mutants except 80–441/d265–294, which lacks the second zinc finger, were coimmunoprecipitated. A GATA6 deletion mutant containing the two zinc fingers was also coimmunoprecipitated with NF-AT3. (C) Summary of coimmunoprecipitation results. F1 and F2 denote the two zinc fingers and NLS designates the nuclear localization signal. (D) The C-terminal region of NF-AT3, encompassing the Rel homology domain (RHD), was translated separately or together with GATA4 deletion mutant 80–328, as indicated. Lanes 1 and 2 show the individual in vitro translation products loaded directly on the gel. Lanes 3–5 show the results of immunoprecipitation with anti-NFAT antibody that recognizes the NF-AT3 RHD. This region is sufficient for interaction with GATA4 (indicated by an asterisk in lane 5).

Figure 3. Regulation of the BNP Promoter by NF-AT3 in Primary Cardiomyocytes

(A) Primary rat cardiomyocytes were transiently transfected with a CAT reporter gene linked to the BNP 5′-flanking region and expression vectors encoding NF-AT3, activated calcineurin, or GATA4, as indicated. Forty-eight hours later, cells were harvested and CAT activity was determined. In the lane labeled −927 site mutant, a BNP-CAT reporter gene, in which the NF-AT3 site at −927 was mutated, was used. (B) 32P-labeled oligonucleotide probes corresponding to the potential NF-AT binding sites from the IL-2 promoter and from −927, −327, and −27 bp 5′ of the BNP gene were used in gel mobility shift assays with unprogrammed reticulocyte lysate (lanes 1–4) or in vitro translated NF-AT3 (lanes 5–8). Only the IL-2 and BNP-927 probes yielded a specific DNA–protein complex. (C) A 32P-labeled oligonucleotide probe corresponding to the −927 BNP site was used in gel mobility shift assays with protein extracts from neonatal rat cardiomyocytes. Multiple specific complexes were generated (lane 1), which were competed by unlabeled cognate site (lane 2), but not by an unrelated E-box oligonucleotide (lane 3). Addition of NF-AT3 antibody eliminated the major DNA–protein complex and created a minor ternary complex (asterisk in lane 5), whereas nonimmune rabbit serum had no effect (lane 4). Free probe was electrophoresed off the bottom of the gel in both experiments.

Figure 4. Inhibition of AngII- and PE-Dependent Hypertrophy of Primary Cardiocytes by CsA and FK506

(A–F) Primary rat cardiocytes in serum-free medium were stimulated with AngII (10 nM) or PE (10 μM) for 72 hr. Cells were then fixed and stained with anti-α-actinin antibody to reveal sarcomeres and Hoechst stain to reveal nuclei. CsA (500 ng/ml) was added to one set of cultures at the time of AngII addition. (G) Total RNA was isolated from primary cardiocyte cultures treated with AngII in the presence or absence of CsA as in (A) and analyzed for expression of GAPDH and ANF transcripts by dot blot. (H) Primary rat cardiomyocytes were transiently transfected with an NF-AT-dependent luciferase reporter gene. Cells were then treated with AngII or PE in the presence or absence of CsA, as described above. Forty-eight hours later, cells were harvested and luciferase activity was determined.

Figure 5. Hypertrophy of α-MHC-Calcineurin Transgenic Hearts

(A) Control and α-MHC-calcineurin transgenic littermates were sacrificed at 18 days of age and hearts were removed and photographed. (B and C) Hearts shown in (A) were sectioned longitudinally. The left ventricular wall and septum are extensively hypertrophied in the calcineurin transgenic (C). (D and E) High magnification views of left ventricular walls of hearts shown in (B) and (C), respectively. (F and G) Transverse sections through hearts of control and calcineurin transgenics at 9 weeks of age stained with H and E. The calcineurin transgenic died suddenly and shows hypertrophy and dilatation. (H) Sections from the calcineurin transgenic heart shown in (G) were stained with Masson trichrome to reveal collagen. Note the extensive interstitial collagen deposits surrounding degenerated cardiomyocytes. Ca, coronary artery; la, left atrium; lv, left ventricle; ra, right atrium; rv, right ventricle. Bars in (B), (C), (F), (G), and (H) = 1 μm. Bars in (D) and (E) = 50 μm.

Figure 6. Changes in Cardiac Gene Expression in α-MHC-Calcineurin Transgenic Mice

Total RNA was isolated from hearts of control and α-MHC-calcineurin transgenic mice at 6 weeks of age. The indicated transcripts were detected by dot blot analysis, and their levels in transgenic hearts relative to controls are shown.

Figure 7. Hypertrophy of α-MHC-NF-AT3 Transgenic Hearts

(A) Structure of NF-AT3 and NF-AT3Δ317 mutant. RHD, Rel homology domain; Reg, regulatory domain. Amino acid positions are indicated. (B and C) Primary cardiomyocytes were transfected with full-length NF-AT3 and NF-AT3Δ317 mutant cDNAs, respectively, and cells were stained with anti-NF-AT antibody. NF-AT3 is localized to the cytoplasm whereas NF-AT3Δ317 is localized to the nucleus. (D and E) Hearts from nontransgenic and NF-AT3Δ317 littermates at 4 weeks of age were sectioned longitudinally and stained with H and E. La, left atrium; lv, left ventricle; ra, right atrium; rv, right ventricle. Bars in (D) and (E) = 2 μm.

Figure 8. Prevention of Calcineurin-Dependent Hypertrophy by CsA

(A) The regimen for CsA treatment is shown. (B) α-MHC-calcineurin transgenic (TG) and nontransgenic mice were treated with or without CsA (25 mg/kg body weight), as indicated. Heart-to-body weight ratios are expressed ± standard deviations. Transgenic littermates obtained from male calcineurin transgenic #37 (see Table 1) were injected subcutaneously twice daily with CsA or vehicle alone beginning at 9 days of age. At 25 days of age, animals were sacrificed and hearts were removed and sectioned longitudinally. (C) H and E sections of hearts from nontransgenic (control) and transgenic mice treated with vehicle (center panel) or CsA (right panel). La, left atrium; lv, left ventricle; ra, right atrium; rv, right ventricle. The right atrium was removed from the control, and both atria were removed from the transgenic treated with CsA. Bar = 2 μm.

Figure 9. A Model for the Calcineurin-Dependent Transcriptional Pathway in Cardiac Hypertrophy

AngII, PE, and possibly other hypertrophic stimuli acting at the cell membrane lead to elevation of intracellular Ca2+ and activation of calcineurin in the cytoplasm. Calcineurin dephosphorylates NF-AT3, resulting in its translocation to the nucleus, where it interacts with GATA4 to synergistically activate transcription. Whether all actions of NF-AT3 are mediated by its interaction with GATA4 or whether there are GATA4-independent pathways for activation of certain hypertrophic responses remains to be determined. Solid arrows denote pathways that are known. Dotted lines denote possible pathways that have not been demonstrated.