Stress-induced cell-cycle activation in Tip60 haploinsufficient adult cardiomyocytes

Abstract

Background: Tat-interactive protein 60 (Tip60) is a member of the MYST family of histone acetyltransferases. Studies using cultured cells have shown that Tip60 has various functions including DNA repair, apoptosis and cell-cycle regulation. We globally ablated the Tip60 gene (Htatip), observing that Tip60-null embryos die at the blastocyst stage (Hu et al. Dev.Dyn.238:2912;2009). Although adult heterozygous (Tip60(+/-)) mice reproduce normally without a haploinsufficient phenotype, stress caused by Myc over-expression induced B-cell lymphoma in Tip60(+/-) adults, suggesting that Tip60 is a tumor suppressor (Gorrini et al. Nature 448:1063;2007). These findings prompted assessment of whether Tip60, alternative splicing of which generates two predominant isoforms termed Tip60α and Tip60β, functions to suppress the cell-cycle in adult cardiomyocytes.

Methodology/principal findings: Western blotting revealed that Tip60α is the predominant Tip60 isoprotein in the embryonic heart, transitioning at neonatal stages to Tip60β, which is the only isoprotein in the adult heart wherein it is highly enriched. Over-expression of Tip60β, but not Tip60α, inhibited cell proliferation in NIH3T3 cells; and, Tip60-haploinsufficient cultured neonatal cardiomyocytes exhibited increased cell-cycle activity. To address whether Tip60β suppresses the cardiomyocyte cell-cycle in the adult heart, hypertrophic stress was induced in Tip60(+/+) and Tip(+/-) littermates via two methods, Myc over-expression and aortic banding. Based on immunostaining cell-cycle markers and western blotting cyclin D, stress increased cardiomyocyte cell-cycle mobilization in Tip60(+/-) hearts, in comparison with Tip60(+/+) littermates. Aortic-banded Tip60(+/-) hearts also exhibited significantly decreased apoptosis.

Conclusions/significance: These findings provide evidence that Tip60 may function in a tumor suppressor pathway(s) to maintain adult cardiomyocytes in replicative senescence.

Figure 1. Tip60 Expression during Heart Development & Maturation.

Total RNA and protein was purified at the indicated stages of development and subjected to semi-quantitative (A) RT/PCR and (B) western blotting analysis. A, the 184 bp band in the upper panel is amplified from all known isoforms of Tip60, whereas the 402 and 558 bp bands in the lower panel are respectively amplified from the Tip60 β and α isoforms. B, the upper panel is a western blot sequentially probed with antibodies recognizing Tip60 (α & β isoproteins were detected with the same antibody) and GAPDH; protein from individual hearts was evaluated at each postnatal day, whereas pools of three hearts each were evaluated at each embryonic day. The lower panel shows densitometric analysis in which each point indicates the mean of three (N = 3) independent western blot determinations. Error bars indicate ±SEM.

Figure 2. Tip60 Expression in Adult Organs.

Tissues from 14 week-old adult Tip60^+/+ and Tip60^+/− mice were processed for (A) semi-quantitative RT/PCR to assess transcript levels (of bulk Tip60 isoforms) in WT and Het tissues and (B) Western blotting to determine Tip60 protein levels. The primer pair used for RT/PCR (A) detected transcripts encoding both Tip60α and Tip60β; GAPDH was assessed as loading control. The bar graph in the lower part of B shows densitometric analysis of protein bands assessed from duplicate animals, normalized to GAPDH and expressed as a percentage of the most abundant band (in WT heart); vertical bars  =  range. +/+  =  WT; +/−  =  Het.

Figure 3. Over-Expression of Tip60β, but not Tip60α, Inhibits Cell Proliferation in NIH/3T3 Cells.

Cultured NIH/3T3 cells were transfected with plasmid p3xFLAG-CMV-7.1 (empty vector), or with the same vector containing cDNA encoding either Tip60α or Tip60β. A shows the average cell number in each well of a 12-well plate, five days after transfecting the cells with plasmid. Error bars  =  ±SEM; p-values were calculated using Student's t-test (two-tailed, unpaired). B shows western blots confirming exogenous expression of Tip60α and Tip60β isoproteins.

Figure 4. Increased Cell-Cycle Activity in Tip60^+/− Neonatal Cardiomyocytes.

Cardiomyocytes were isolated from 2 day-old neonatal hearts and cultured on gelatin-coated dishes. When the cells were ∼60% confluent (72 hrs later) they were double-immunostained for phosphorylated histone-H3 (H3P) to detect M-phase cells (arrows in A & C) and for sarcomeric α-actin (not shown) to verify cardiomyocyte identity. Nuclei were stained with DAPI (B,D); H3P-labeled nuclei are encircled because DAPI is obscured DAB-stained nuclei. E shows percentages of H3P-positive neonatal cardiomyocytes, based on enumerating 1,000-2,000 cells in each dish; error bars  =  +/−SEM. Scale bars in A–D = 10 µm.

Figure 5. Tip60-Haploinsufficiency Augments 4-OHT-Induced Induction of Cell-Cycle Activity in MycER Transgenic Hearts.

Eight week old MycER transgenic Tip60 wild-type (WT) and heterozygous (Het) mice were induced with 4-OHT for seven days and assessed for cell-cycle activity by immunostaining phosphorylated histone H3 (H3P; arrows in A,B). This was verified by immunostaining BrdU-incorporated nuclei (arrows in D,E). Percentages of labeled cells were determined by evaluating at least 5,000 (C) or 2,500 (F) hematoxylin-stained nuclei for H3P or BrdU antigen, respectively. (N)  =  number of hearts; vertical bars  =  ±SEM; p-values were calculated using Student's t-Test (two-tailed, unpaired). The scale bar in all images  =  10 µm. (Note: A control utilizing WT-MycER mice demonstrated that 4-OHT had no effect on these parameters [26]).

Figure 6. Most H3P-Positive Nuclei are in Cardiomyocytes.

Sections from hearts processed in Figure 5 were double-immunostained for phosphorylated H3P and Nkx2.5 to determine cardiomyocyte identity. A & D show Nkx2.5 staining (nucleus-specific brown DAB reaction product) in cardiomyocyte nuclei, as distinct from smaller non-myocyte nuclei in which only (blue) hematoxylin counter-stain is seen. B & E show H3P fluorescent green signal in the same nuclei. C and F are merged images of A–B and D–E, respectively. In C & F, dark arrows denote nuclei double-stained for Nkx2.5 and H3P; white arrows denote nuclei expressing only Nkx2.5. G summarizes results from enumerating a minimum of 150 H3P-positive nuclei per heart for co-localization of Nkx2.5. Error bars  =  ±SEM; statistical significance was determined by Student's t-Test (two-tailed, unpaired). Scale bars  = 20 µM.

Figure 7. Tip60-Haploinsufficiency Increases 4-OHT-Induced Expression of Cyclin D2.

Eight week old MycER transgenic Tip60 wild-type (WT) or heterozygous (Het) mice were induced with 4-OHT for seven days and assessed for cyclin D2 by western blotting. A is a western blot showing cyclin D2 and GAPDH levels in non-stressed (no 4-OHT) and stressed (+4-OHT) WT and Het myocardium. Each lane contained 10 µg total protein from separate hearts. B shows densitometry of the bands in A, with cyclin D2 normalized to GAPDH. Numbers (N) in parentheses indicate numbers of hearts evaluated. Vertical bars  =  ±SEM; p-values were calculated using Student's t-Test (two-tailed, unpaired).

Figure 8. Induction of Cell-Cycle Activity and Reduced Apoptosis in Aortic-Banded Mice are Enhanced by Tip60-Haploinsufficiency.

Mice were subjected to trans-aortic constriction (TAC) for two weeks. Panels A-B show Ki-67 (A) and phosphorylated histone H3 (H3P; B), which are significantly increased in Tip60-haploinsufficient myocardium. Panel C (TUNEL) shows a trend toward, and D (caspase-3 staining) demonstrates a significant reduction in, apoptosis in heterozygous myocardium. In C, cells were scored as TUNEL-positive if at least 50% of the nucleus contained signal. The microscopic images were selected as representative of the staining quality used to obtain data in the bar graphs. At least 1,500 DAPI-stained nuclei were enumerated in each heart for cell-cycle or apoptotic markers. (N)  =  number of hearts evaluated; vertical bars  =  ±SEM; statistical significance was via Student's t-Test (two-tailed, unpaired). Arrows in A–C denote labeled nuclei; arrows in D denote peri-nuclear loci of caspase-3. Bars in A = 25 µm; B = 15 µm.; C–D = 40 µm.