Inhibition of Glycogen Synthase Kinase 3β During Heart Failure Is Protective
Glycogen synthase kinase (GSK)-3, a negative regulator of cardiac hypertrophy, is inactivated in failing hearts. To examine the histopathological and functional consequence of the persistent inhibition of GSK-3β in the heart in vivo, we generated transgenic mice with cardiac-specific overexpression of dominant negative GSK-3β (Tg-GSK-3β-DN) and tetracycline-regulatable wild-type GSK-3β. GSK-3β-DN significantly reduced the kinase activity of endogenous GSK-3β, inhibited phosphorylation of eukaryotic translation initiation factor 2Bε, and induced accumulation of β-catenin and myeloid cell leukemia-1, confirming that GSK-3β-DN acts as a dominant negative in vivo. Tg-GSK-3β-DN exhibited concentric hypertrophy at baseline, accompanied by upregulation of the α-myosin heavy chain gene and increases in cardiac function, as evidenced by a significantly greater Emax after dobutamine infusion and percentage of contraction in isolated cardiac myocytes, indicating that inhibition of GSK-3β induces well-compensated hypertrophy. Although transverse aortic constriction induced a similar increase in hypertrophy in both Tg-GSK-3β-DN and nontransgenic mice, Tg-GSK-3β-DN exhibited better left ventricular function and less fibrosis and apoptosis than nontransgenic mice. Induction of the GSK-3β transgene in tetracycline-regulatable wild-type GSK-3β mice induced left ventricular dysfunction and premature death, accompanied by increases in apoptosis and fibrosis. Overexpression of GSK-3β-DN in cardiac myocytes inhibited tumor necrosis factor-α–induced apoptosis, and the antiapoptotic effect of GSK-3β-DN was abrogated in the absence of myeloid cell leukemia-1. These results suggest that persistent inhibition of GSK-3β induces compensatory hypertrophy, inhibits apoptosis and fibrosis, and increases cardiac contractility and that the antiapoptotic effect of GSK-3β inhibition is mediated by myeloid cell leukemia-1. Thus, downregulation of GSK-3β during heart failure could be compensatory.
GSK-3β is a ubiquitously expressed serine/threonine kinase that has versatile biological functions in cells, including regulation of metabolism, cell growth/death, and protein translation and transcription.1,2 Unlike most protein kinases, GSK-3β remains active in the resting state and is inactivated when cells are stimulated by mitogens, by other protein kinases, such as Akt, or by the Wnt pathway. In cardiac myocytes, GSK-3β phosphorylates β-catenin,3 eukaryotic translation initiation factor (eIF)2Bε,4 NFAT,5 GATA4,6 myocardin,7 and other proteins, thereby negatively regulating protein synthesis and gene expression. GSK-3β downregulates SERCA2a8 and enhances mitochondrial permeability transition,9 thereby leading to an inability to normalize cytosolic Ca2+ in diastole and reduced cell survival, respectively.
GSK-3β is an important negative regulator of cardiac hypertrophy.10 GSK-3β negatively regulates β-adrenergic and endothelin-induced cardiac hypertrophy in cultured neonatal myocytes in vitro.6,11,12 Cardiac specific expression of constitutively active GSK-3β (GSK-3β[S9A]) in transgenic mice inhibits cardiac hypertrophy in response to pressure overload and stimulation of the β-adrenergic receptors.13 Conditional expression of GSK-3β(S9A) in transgenic mice reversed pressure overload–induced cardiac hypertrophy.14 Many hypertrophic stimuli inhibit GSK-3β, thereby removing its negative constraints on hypertrophy, which represents a unique mechanism of cardiac hypertrophy.
GSK-3β is phosphorylated at Ser9 and inactivated in heart failure patients.15 Considering the fact that GSK-3β is an important regulator of growth, death, and metabolism in other cell types, changes in the activity of GSK-3β should cause significant biological effects during heart failure. Constitutively active GSK-3β prevents pathological hypertrophy in transgenic mice and thus seems to be salutary.13,14 On the other hand, overexpression of wild-type GSK-3β prevents physiological growth of the heart and causes cardiac dysfunction.8 Expression of constitutively active GSK-3β in a genetic background of hypertrophic cardiomyopathy inhibits the development of hypertrophy but induces heart failure.16 Pharmacological (short-term) inhibition of GSK-3β mimics the protective effects of ischemic preconditioning.17 Inhibition of GSK-3β prevents mitochondrial permeability transition in cardiac myocytes.9 These studies appear to indicate that activation of GSK-3β is detrimental. Thus, whether GSK-3β is protective or detrimental for myocytes seems controversial.18 What previous studies have not addressed is whether or not long-term suppression of GSK-3β activity is protective or detrimental for the heart in vivo.
The goal of this investigation was to elucidate the effect of persistent inactivation of GSK-3β on cardiac phenotype at baseline and in response to pressure overload, using transgenic mice expressing a kinase inactive (dominant negative) form of GSK-3β.
Materials and Methods
For an expanded Materials and Methods section, please refer to the online data supplement at http://circres.ahajournals.org. All experiments involving animals were approved by the Institutional Animal Care and Use Committee at New Jersey Medical School. The data are expressed as means±SEM. Statistical analyses were performed using ANOVA and the Tukey post test procedure. Values of P<0.05 were considered significant.
Generation of Tg-GSK-3β-DN
We generated transgenic mice with cardiac-specific expression of dominant negative GSK-3β (Tg-GSK-3β-DN) using the α myosin heavy chain (αMHC) promoter. We used a double-lysine mutant of GSK-3β, known to act as a dominant negative.19 Tg-GSK-3β-DN line no. 19 mice had 4.7±0.2-fold overexpression of GSK-3β (Figure 1A). Although GSK-3β immunoprecipitated from nontransgenic (NTg) mouse hearts phosphorylated Ser199 of tau, GSK-3β from Tg-GSK-3β-DN hearts induced little phosphorylation of tau (Figure 1B). GSK-3α immunoprecipitated from Tg-GSK-3β-DN hearts exhibited ≈40% less phosphorylation of Ser199 tau than that from NTg (not shown). Thus, GSK-3β-DN almost fully suppressed the activity of GSK-3β and modestly reduced the activity of GSK-3α, thereby acting as a dominant negative primarily against GSK-3β. Consistently, basal phosphorylation of eIF2Bε, a substrate of GSK-3β,4 was significantly reduced in Tg-GSK-3β-DN compared with NTg mice (Figure 1C). Furthermore, both β-catenin and myeloid cell leukemia (MCL)-1, the expressions of which are negatively regulated by GSK-3β,20 were significantly upregulated in Tg-GSK-3β-DN (Figure 1D). Collectively, the activity of GSK-3β is downregulated in Tg-GSK-3β-DN hearts.
Tg-GSK-3β-DN Develop Physiological Hypertrophy
Tg-GSK-3β-DN (line no. 19) showed a significantly greater left ventricular (LV) weight/body weight (LVW/BW) than NTg at 2.5 months and thereafter (Figure 2A). Tg-GSK-3β-DN (line no. 11, homozygous), with 3.2±0.9-fold overexpression, also exhibited a slightly but significantly greater heart weight (HW)/BW than NTg (4.35±0.05 versus 4.05±0.06, P<0.05) at 6 months. Thus, inhibition of GSK-3β enhanced cardiac growth. We primarily describe the phenotype of line no. 19. Individual myocytes isolated from Tg-GSK-3β-DN hearts at 2.5 to 3 months of age have significantly greater length and width, as well as electrophysiologically determined whole cell capacitance, proportional to the cell surface area, than those from NTg hearts (Figure 2B).
At 3 months, mRNA expression of atrial natriuretic factor was modestly upregulated in Tg-GSK-3β-DN hearts. mRNA expression of α-skeletal-actin in Tg-GSK-3β-DN hearts was not significantly different from that in NTg, whereas mRNA expression of βMHC was significantly downregulated in Tg-GSK-3β-DN hearts (Figure 2C). Interestingly, mRNA expression of αMHC was significantly upregulated in Tg-GSK-3β-DN hearts (Figure 2C). Thus, the expression pattern of hypertrophy-associated genes in Tg-GSK-3β-DN hearts mimics that of physiological hypertrophy.
Histological analyses conducted at 2.5 to 3 and 4 months of age indicated that the myocyte cross-sectional area was significantly greater in Tg-GSK-3β-DN hearts than in NTg hearts (Figure I in the online data supplement). No fibrosis was observed in Tg-GSK-3β-DN hearts (data not shown).
Tg-GSK-3β-DN Mice Have Elevated Maximum Cardiac Contractility
We did not observe significant increases in premature death in Tg-GSK-3β-DN up to 24 months of age. Echocardiographic measurement at baseline showed no significant differences in the chamber size. Although LV fractional shortening percentage (%FS) was slightly greater in Tg-GSK-3β-DN than in NTg (Table), the difference was not statistically significant.
Hemodynamic measurements indicated that Tg-GSK-3β-DN showed a slightly higher Emax and dP/dts than NTg at baseline. The difference in Emax and dP/dts between Tg-GSK-3β-DN and NTg became prominent after dobutamine injection (Figure 3A through 3C). Increases in cardiac output induced by dobutamine were significantly greater in Tg-GSK-3β-DN than in NTg (Figure 3D). The heart rate was not different between NTg and Tg-GSK-3β-DN after dobutamine administration (supplemental Figure II). These results suggest that the maximum contractility of Tg-GSK-3β-DN hearts is greater than that of NTg hearts.
The contractile function was also evaluated using isolated ventricular myocytes (3 to 4 months of age). Indices of isolated myocyte contraction, including percentage of contraction and +dL/dt, were greater in Tg-GSK-3β-DN than in NTg (supplemental Figure IIIA and IIIB). To examine the cellular mechanism for the enhanced myocyte shortening in Tg-GSK-3β-DN, we measured L-type Ca2+ channel currents (ICa). Myocytes from Tg-GSK-3β-DN exhibited a significantly greater ICa than those from NTg. Peak ICa (pA/pF), plotted as a function of voltage, demonstrated no significant difference in the current–voltage relationships but showed a significantly greater peak ICa density in myocytes from Tg-GSK-3β-DN than in those from NTg (supplemental Figure IIIC). Protein expression of L-type Ca2+ channel was significantly elevated, whereas that of phospholamban and SERCA2a was not, in Tg-GSK-3β-DN hearts (supplemental Figure IV). These results suggest that increases in ICa may contribute to the increased myocyte contractility in Tg-GSK-3β-DN.
The Effect of Transverse Aortic Constriction on LV Hypertrophy and Function in Tg-GSK-3β-DN
To examine whether inhibition of GSK-3β affects the development of heart failure under pathological stresses, we applied transverse aortic constriction (TAC) for 4, 6, and 8 weeks to Tg-GSK-3β-DN and NTg. To induce transition into decompensated hypertrophy, TAC was conducted using a 28-gauge needle. A comparable pressure gradient was applied to each group after 4 and 6 weeks, whereas the pressure gradient was 20% greater in Tg-GSK-3β-DN than in NTg at 8 weeks, possibly because of better cardiac function in Tg-GSK-3β-DN (supplemental Figure V). Kaplan–Meier survival analysis showed that there was no statistical difference in mortality after TAC between Tg-GSK-3β-DN and NTg. After TAC, HW/tibial length (TL) was greater in Tg-GSK-3β-DN than in NTg at all time points (Figure 4A). However, when the level of cardiac hypertrophy after TAC was normalized to that after sham operation, the TAC-induced fold increase in hypertrophy was similar between Tg-GSK-3β-DN and NTg (supplemental Figure VI), suggesting that GSK-3β-DN primarily affects basal hypertrophy, but not TAC-induced hypertrophy.
Changes in LV systolic function during TAC were evaluated using echocardiography. LV %FS in NTg was significantly reduced at 6 and 8 weeks, whereas that in Tg-GSK-3β-DN was maintained up to 8 weeks and was significantly greater than that in NTg at 6 and 8 weeks (Figure 4B). When the relationship between HW/BW, representing the extent of LV hypertrophy, and lung weight (LW)/BW, representing the extent of LV decompensation, is plotted on x and y axes, respectively, the slope may indicate the extent of cardiac decompensation for a given extent of cardiac hypertrophy. Although the slope in Tg-GSK-3β-DN was significantly steeper than that in NTg at 4 weeks, the slope became less steep at 6 weeks and significantly less steep at 8 weeks than that in NTg (Figure 4C). These results suggest that Tg-GSK-3β-DN mice are more resistant to decompensation at a given extent of cardiac hypertrophy than NTg mice after 8 weeks of TAC.
Histological analyses of LV sections indicated that Tg-GSK-3β-DN mice after TAC have significantly fewer TUNEL-positive myocytes than NTg subjected to TAC (Figure 4D and supplemental Figure VII). Masson’s trichrome staining showed that less fibrosis was observed in Tg-GSK-3β-DN than in NTg mice after 8 weeks of TAC (Figure 4D). Taken together, long-term inactivation of GSK-3β decreases apoptosis/fibrosis and enhances functional compensation in response to pressure overload.
Inhibition of GSK-3β Failed to Exert an Additive Effect on Exercise-Induced Cardiac Hypertrophy
Because Tg-GSK-3β-DN developed well-compensated hypertrophy, we examined whether baseline cardiac hypertrophy in Tg-GSK-3β-DN and exercise-induced cardiac hypertrophy develop through a common mechanism. To this end, Tg-GSK-3β-DN and NTg mice were subjected to a 6-week swimming protocol.21 Although swimming significantly increased HW/BW in both groups, expression of GSK-3β-DN failed to exert an additive effect on the swimming-induced hypertrophy observed in NTg (supplemental Figure VIII). Together with recent evidence that GSK-3β is phosphorylated in response to exercise,22 these results suggest that GSK-3β-DN– and exercise-induced (physiological) hypertrophy are mediated in part through a common signaling mechanism.
Overexpression of Wild-Type GSK-3β Induces Heart Failure
To further examine the cardiac effects of GSK-3β, we generated cardiac-specific wild-type GSK-3β overexpression mice using the tetracycline (tet)-OFF system. Mice carrying the bidirectional tet-responsive promoter followed by a myc-GSK-3β cDNA in 1 direction and a DNA encoding β-galactosidase in the other direction (Tg-tetGSK)23 were crossed with mice carrying αMHC-tTA (Tg-tTA). In Tg-tetGSK-tTA mice, transgene expression in the heart was tightly regulated by doxycycline (Dox). Namely, a 2.4-fold overexpression of GSK-3β was induced in the heart in the absence of Dox, whereas expression of the transgene was abolished in the presence of Dox (Figure 5A). Transgene expression was never observed in Tg-tetGSK mice without tTA. Immune complex kinase assays, using tau as a substrate, showed that the total activity of GSK-3β was 5 times greater in Tg-tetGSK-tTA (without Dox) than in Tg-tetGSK (Figure 5A). The heart-specific expression of the transgene was also confirmed by comparison of β-galactosidase staining in Tg-tetGSK-tTA without Dox and in Tg-tetGSK (Figure 5B).
Neither an obvious baseline cardiac phenotype nor premature death was observed in Tg-tetGSK. In contrast, Tg-tetGSK-tTA without Dox started to die at ≈10 weeks, and ≈90% of Tg-tetGSK-tTA died by 30 weeks (Figure 5C). Echocardiographic measurement showed that Tg-tetGSK-tTA without Dox exhibited increased LV end-diastolic dimension and reduced %FS at 30 weeks (supplemental Table I). LW/TL was significantly greater in Tg-tetGSK-tTA without Dox than in Tg-tetGSK (supplemental Figure IXA and Table II), suggesting that the mice developed heart failure. Interestingly, HW/TL was also greater in Tg-tetGSK-tTA without Dox than in Tg-tetGSK, suggesting that cardiac hypertrophy developed even in the presence of GSK-3β overexpression (supplemental Figure IXB). Histological analyses of the LV sections showed that Tg-tetGSK-tTA without Dox developed prominent fibrosis, whereas fibrosis was not observed in Tg-tetGSK (Figure 5D). The number of TUNEL-positive cells was significantly higher in Tg-tetGSK-tTA without Dox than in Tg-tetGSK (Figure 5D). These results suggest that increased expression of GSK-3β in the myocardium stimulates apoptosis.
The progressive increase in premature death of Tg-tetGSK-tTA was also observed when Dox treatment was withdrawn at 10 weeks after birth (supplemental Figure XA), which eliminated contributions of transgene expression during fetal and neonatal development to the premature death of the mice. After Dox withdrawal, Tg-tetGSK-tTA exhibited progressive increases in LV end-diastolic dimension and decreases in %FS, as determined by echocardiographic measurement (supplemental Figure XB and Table III), suggesting that the postnatal conditional increase in GSK-3β expression induces cardiac dysfunction.
GSK-3β is inactivated during cardiac hypertrophy and heart failure.12 To examine the role of GSK-3β inactivation during heart failure, TAC was applied to Tg-tetGSK-tTA in the presence of Dox, and, after 4 weeks, Dox was withdrawn and the mice were followed for another 4 weeks (Figure 6A). Eight weeks of TAC significantly reduced the kinase activity of GSK-3β in Tg-tetGSK-tTA mice in the presence of Dox. However, Dox withdrawal induced transgene expression of GSK-3β, and the GSK-3β activity was no longer inhibited after 8 weeks of TAC (Figure 6B). Although the activity of GSK-3β after Dox withdrawal was significantly greater than in control mice (≈1.7-fold), it was not as high as in Dox-untreated Tg-tetGSK-tTA mice without TAC (≈5-fold shown in Figure 6A). In these experiments, all male Tg-tetGSK-tTA subjected to TAC (n=5) died after Dox withdrawal, but all female Tg-tetGSK-tTA mice survived. %FS was significantly reduced, whereas LV end-diastolic dimension was significantly increased in female Tg-tetGSK-tTA after Dox withdrawal compared with those treated with Dox (Figure 6C). Dox withdrawal significantly reduced septal and posterior wall thickness in Tg-tetGSK-tTA in the presence of TAC (supplemental Figure XI). At the time of euthanasia (8 weeks after TAC), Tg-tetGSK-tTA mice that underwent Dox withdrawal exhibited significantly greater LW/TL and HW/TL than Tg-tetGSK-tTA that were kept on Dox (Figure 6D). These results suggest that decreases in GSK-3β activity during heart failure are protective for the heart.
Mechanisms of Apoptosis
We examined the molecular mechanism by which inhibition of GSK-3β protects cardiac myocytes from apoptosis. We used adenovirus vector harboring GSK-3β-DN (Ad-GSK-3β-DN) to achieve inhibition of GSK-3β in a specific manner. Treatment of control virus (Ad-LacZ)–transduced cardiac myocytes with tumor necrosis factor (TNF)α, a pro-inflammatory cytokine known to induce apoptosis in cardiac myocytes, in the presence of cycloheximide24 significantly increased apoptosis, as determined by cytoplasmic accumulation of mono- and oligonucleosomes. Expression of GSK-3β-DN significantly reduced TNFα-induced cardiac myocyte apoptosis (Figure 7A). Expression of GSK-3β-DN induced significant accumulation of MCL-1,25 an antiapoptotic molecule (see Figure 1C), but not Bcl-2 nor Bcl-xL (supplemental Figure XII). To examine the role of MCL-1 in mediating the antiapoptotic effect of GSK-3β-DN, we conducted knock down of MCL-1 using adenovirus harboring short hairpin (sh)RNA-MCL-1. shRNA-MCL-1, but not control shRNA, downregulated GSK-3β-DN–induced expression of MCL-1 in cardiac myocytes (Figure 7B). Downregulation of MCL-1 by shRNA-MCL-1, but not control shRNA, abolished the protective effects of GSK-3β-DN (Figure 7A). These results suggest that accumulation of MCL-1 plays an important role in mediating the antiapoptotic effect of GSK-3β inhibition.
Our results suggest that persistent inhibition of GSK-3β induces cardiac hypertrophy in mouse hearts. Cardiac hypertrophy observed in Tg-GSK-3β-DN was well compensated. Hypertrophy induced by inhibition of GSK-3β appears to have properties of physiological hypertrophy for a number of reasons. First, hypertrophy in Tg-GSK-3β-DN mice is not accompanied by LV dilation. Second, the maximum contractility of the heart after dobutamine infusion and contractility of individual myocytes isolated from Tg-GSK-3β-DN mice are enhanced. Third, αMHC expression was significantly higher in Tg-GSK-3β-DN. Increases in αMHC are frequently observed in physiological hypertrophy.26 Fourth, neither fibrosis nor apoptosis was increased in conjunction with cardiac hypertrophy in Tg-GSK-3β-DN. Fifth, GSK-3β-DN and swimming exercise failed to show completely additive effects on cardiac hypertrophy, suggesting that GSK-3β-DN and exercise induce hypertrophy partially through a common mechanism. These results collectively suggest that inhibition of GSK-3β mediates physiological growth of the heart. These observations are relevant to the mechanism of heart failure because physiological hypertrophy could negatively regulate the development of pathological hypertrophy.22
GSK-3β phosphorylates GATA4 and eIF2Bε, thereby inhibiting cardiac hypertrophy through inhibition of either gene expression or protein translation.4,6 Because GATA4 protects the heart from load-induced heart failure,27 the cardioprotective effects of GSK-3β inhibition may be mediated in part through GATA4. β-Catenin may also regulate physiological growth of cardiac myocytes through regulation of lymphocyte enhancer factor–dependent signaling mechanisms.28 The downstream signaling mechanism mediating physiological hypertrophy in Tg-GSK-3β-DN remains to be elucidated.
The activity of GSK-3β is gradually downregulated by pressure overload in humans and experimental animals.15 When the extent of hypertrophy after TAC was expressed as relative to that of sham-operated mice, there was no significant enhancement in the TAC-induced hypertrophy in Tg-GSK-3β-DN compared with NTg. This indicates either that further inhibition of GSK-3β does not have an additive effect on TAC-induced hypertrophy or that other signaling mechanisms are sufficient to induce TAC-induced hypertrophy.
Although chronic inhibition of GSK-3β failed to enhance TAC-induced hypertrophy, it maintains LV function against pressure overload and is therefore protective. Several lines of evidence support this notion. First, persistent inhibition of GSK-3β inhibits LV systolic dysfunction during pressure overload, as evidenced by preserved %FS at 6 and 8 weeks. Second, the level of lung congestion, as determined by LW/BW, is less in Tg-GSK-3β-DN than in NTg at a given level of cardiac hypertrophy, which indicates that Tg-GSK-3β-DN are more resistant to cardiac decompensation. Third, both apoptosis and fibrosis are less severe in Tg-GSK-3β-DN than in NTg in response to pressure overload. Collectively, these data indicate that inhibition of GSK-3β activates cardioprotective effects, thereby inhibiting apoptosis and fibrosis during cardiac stresses. Because increases in baseline contractility and L-type Ca2+ channel current were observed in single cardiac myocytes isolated from Tg-GSK-3β-DN, the direct effect of GSK-3β inhibition on cardiac contractility may also contribute to the preserved cardiac function during pressure overload in Tg-GSK-3β-DN. Other unknown mechanisms may also contribute to the protective effects of GSK-3β inhibition. Transgenic expression of GSK-3β-DN enhanced cardioprotective effects under TAC despite the fact that endogenous GSK-3β is partially inhibited during hypertrophy. Either nearly complete inhibition of GSK-3β activity or inhibition of GSK-3β before its inhibition by TAC in NTg may contribute to the enhancement of cardioprotection in Tg-GSK-3β-DN mice.
To further examine the role of GSK-3β in regulating heart failure, we generated a gain-of-function model using an inducible system. Our results suggest that stimulation of GSK-3β is detrimental in the adult heart, consistent with the previous report by Michael et al.8 The inducible system, however, allowed us to eliminate the influence of transgene expression during the fetal period. Furthermore, the inducible system allowed us to restore the activity of GSK-3β after the animal established hypertrophy in response to pressure overload. Our results clearly indicate that inhibition of GSK-3β is protective during cardiac hypertrophy because increasing the activity of GSK-3β after the animals develop hypertrophy caused cardiac dysfunction and increased mortality.
Previous reports suggested that GSK-3β(S9A) protects the heart from pressure overload13,14 but not from hypertrophic cardiomyopathy.16 Because GSK-3β(S9A) cannot be phosphorylated by upstream kinases, such as Akt, it is a constitutively active GSK-3β, phosphorylating both C-terminally phosphorylated (primed) and unphosphorylated (unprimed) substrates. On the other hand, Ser9 of wild-type GSK-3β may be partially phosphorylated on hypertrophic stimulation. Serine 9 phosphorylated GSK-3β cannot phosphorylate primed substrates, although it may still phosphorylate unprimed substrates.1 Thus, overexpression of GSK-3β(S9A) and GSK-3β (wild type) could differentially affect phosphorylation of downstream targets, thereby causing distinct phenotypes. Importantly, we confirmed that conditional overexpression of wild-type GSK-3β restored the kinase activity of GSK-3β during pressure overload.
Our results suggest that stimulation of GSK-3β induces apoptosis in the heart in vivo, whereas inhibition of GSK-3β inhibits apoptosis. Although antiapoptotic effects of GSK-3β inhibition during ischemia/reperfusion have been suggested using SB 216763, a chemical inhibitor,17 the long term effect of GSK-3β inhibition had not been previously demonstrated in vivo. Furthermore, our result demonstrated for the first time that stimulation of GSK-3β is able to induce apoptosis in the heart in vivo. Although GSK-3β is active even under unstimulated conditions, it is further activated when Ser9-phosphorylating kinases are inactivated10 or Tyr216/279 kinases, including Pyk2, are activated.29 For example, GSK-3β is activated in diabetic tissues, and GSK-3β inhibitors have been developed to treat several human disorders, including Alzheimer’s disease, bipolar disorder, cancer, and diabetes.30 GSK-3β is transiently activated by isoproterenol in adult cardiac myocytes in vitro.31 It will be important to correctly identify the conditions under which GSK-3β is activated in the heart and test the involvement of GSK-3β activation in cardiac myocyte apoptosis.
Our results suggest that accumulation of MCL-1 may play an important role in mediating the antiapoptotic effects of GSK-3β inhibition. GSK-3β induces proteasomal degradation of MCL-1 through direct phosphorylation.20 Inhibition of GSK-3β induced significant accumulation of MCL-1, and downregulation of MCL-1 abolished the antiapoptotic effects of GSK-3β inhibition on TNFα-induced apoptosis. Although GSK-3β regulates apoptosis of cardiac myocytes through interaction with the mitochondrial permeability transition pore complex,9 such interaction may be regulated through MCL-1. Inhibiting GSK-3β–mediated phosphorylation of MCL-1 could be a novel strategy for protecting cardiac myocytes from apoptosis during heart failure.
Another interesting observation in this report is that postnatal GSK-3β overexpression allows the heart to develop hypertrophy in the presence of heart failure. This may contradict the notion that GSK-3β is a negative regulator of cardiac hypertrophy. However, cardiac hypertrophy induced by postnatal overexpression of GSK-3β was accompanied by cardiac dilation, reduced contractility, and increases in apoptosis. Thus, GSK-3β may not inhibit pathological hypertrophy, which is presumably mediated through GSK-3β–insensitive mechanisms. Inhibition of physiological hypertrophy by GSK-3β could indirectly facilitate pathological hypertrophy.
In summary, inhibition of GSK-3β at baseline induces well-compensated cardiac hypertrophy similar to physiological hypertrophy and stimulates cardiac function. Inhibition of GSK-3β during pressure overload plays a protective role, possibly by inhibiting apoptosis and fibrosis, thereby preventing cardiac decompensation. GSK-3β inhibitors may be beneficial for heart failure treatment.
We thank Daniela Zablocki for critical reading of the manuscript.
Sources of Funding
Supported by United States Public Health Service grants HL 59139, HL67724, HL69020, AG23039, and AG28787; and American Heart Association Grant 0340123N. S.H. and H.T. were supported by Postdoctoral Fellowships from the American Heart Association Heritage Affiliate.
↵*The first two authors contributed equally to this work.
Original received April 9, 2007; resubmission received July 25, 2007; revised resubmission received September 12, 2007; accepted September 13, 2007.
Doble BW, Woodgett JR. GSK-3: tricks of the trade for a multi-tasking kinase. J Cell Sci. 2003; 116: 1175–1186.
Haq S, Michael A, Andreucci M, Bhattacharya K, Dotto P, Walters B, Woodgett J, Kilter H, Force T. Stabilization of beta-catenin by a Wnt-independent mechanism regulates cardiomyocyte growth. Proc Natl Acad Sci U S A. 2003; 100: 4610–4615.
Hardt SE, Tomita H, Katus HA, Sadoshima J. Phosphorylation of eukaryotic translation initiation factor 2Bepsilon by glycogen synthase kinase-3beta regulates beta-adrenergic cardiac myocyte hypertrophy. Circ Res. 2004; 94: 926–935.
Beals CR, Clipstone NA, Ho SN, Crabtree GR. Nuclear localization of NF-ATc by a calcineurin-dependent, cyclosporin-sensitive intramolecular interaction. Genes Dev. 1997; 11: 824–834.
Morisco C, Seta K, Hardt SE, Lee Y, Vatner SF, Sadoshima J. Glycogen synthase kinase 3beta regulates GATA4 in cardiac myocytes. J Biol Chem. 2001; 276: 28586–28597.
Badorff C, Seeger FH, Zeiher AM, Dimmeler S. Glycogen synthase kinase 3beta inhibits myocardin-dependent transcription and hypertrophy induction through site-specific phosphorylation. Circ Res. 2005; 97: 645–654.
Michael A, Haq S, Chen X, Hsich E, Cui L, Walters B, Shao Z, Bhattacharya K, Kilter H, Huggins G, Andreucci M, Periasamy M, Solomon RN, Liao R, Patten R, Molkentin JD, Force T. Glycogen synthase kinase-3beta regulates growth, calcium homeostasis, and diastolic function in the heart. J Biol Chem. 2004; 279: 21383–21393.
Juhaszova M, Zorov DB, Kim SH, Pepe S, Fu Q, Fishbein KW, Ziman BD, Wang S, Ytrehus K, Antos CL, Olson EN, Sollott SJ. Glycogen synthase kinase-3beta mediates convergence of protection signaling to inhibit the mitochondrial permeability transition pore. J Clin Invest. 2004; 113: 1535–1549.
Hardt SE, Sadoshima J. Glycogen synthase kinase-3beta: a novel regulator of cardiac hypertrophy and development. Circ Res. 2002; 90: 1055–1063.
Morisco C, Zebrowski D, Condorelli G, Tsichlis P, Vatner SF, Sadoshima J. The Akt-glycogen synthase kinase 3beta pathway regulates transcription of atrial natriuretic factor induced by beta-adrenergic receptor stimulation in cardiac myocytes. J Biol Chem. 2000; 275: 14466–14475.
Haq S, Choukroun G, Kang ZB, Ranu H, Matsui T, Rosenzweig A, Molkentin JD, Alessandrini A, Woodgett J, Hajjar R, Michael A, Force T. Glycogen synthase kinase-3beta is a negative regulator of cardiomyocyte hypertrophy. J Cell Biol. 2000; 151: 117–130.
Antos CL, McKinsey TA, Frey N, Kutschke W, McAnally J, Shelton JM, Richardson JA, Hill JA, Olson EN. Activated glycogen synthase-3 beta suppresses cardiac hypertrophy in vivo. Proc Natl Acad Sci U S A. 2002; 99: 907–912.
Sanbe A, Gulick J, Hanks MC, Liang Q, Osinska H, Robbins J. Reengineering inducible cardiac-specific transgenesis with an attenuated myosin heavy chain promoter. Circ Res. 2003; 92: 609–616.
Haq S, Choukroun G, Lim H, Tymitz KM, del Monte FF, Gwathmey J, Grazette L, Michael A, Hajjar R, Force T, Molkentin JD. Differential activation of signal transduction pathways in human hearts with hypertrophy versus advanced heart failure. Circulation. 2001; 103: 670–677.
Luckey SW, Mansoori J, Fair K, Antos CL, Olson E, Leinwand LA. Blocking cardiac growth in hypertrophic cardiomyopathy induces cardiac dysfunction and decreased survival only in males. Am J Physiol Heart Circ Physiol. 2007; 292: H838–H845.
Tong H, Imahashi K, Steenbergen C, Murphy E. Phosphorylation of glycogen synthase kinase-3beta during preconditioning through a phosphatidylinositol-3-kinase-dependent pathway is cardioprotective. Circ Res. 2002; 90: 377–379.
McMullen JR, Shioi T, Zhang L, Tarnavski O, Sherwood MC, Kang PM, Izumo S. Phosphoinositide 3-kinase(p110alpha) plays a critical role for the induction of physiological, but not pathological, cardiac hypertrophy. Proc Natl Acad Sci U S A. 2003; 100: 12355–12360.
Konhilas JP, Watson PA, Maass A, Boucek DM, Horn T, Stauffer BL, Luckey SW, Rosenberg P, Leinwand LA. Exercise can prevent and reverse the severity of hypertrophic cardiomyopathy. Circ Res. 2006; 98: 540–548.
Lucas JJ, Hernandez F, Gomez-Ramos P, Moran MA, Hen R, Avila J. Decreased nuclear beta-catenin, tau hyperphosphorylation and neurodegeneration in GSK-3beta conditional transgenic mice. EMBO J. 2001; 20: 27–39.
de Moissac D, Mustapha S, Greenberg AH, Kirshenbaum LA. Bcl-2 activates the transcription factor NFkappaB through the degradation of the cytoplasmic inhibitor IkappaBalpha. J Biol Chem. 1998; 273: 23946–23951.
Chang KC, Figueredo VM, Schreur JH, Kariya K, Weiner MW, Simpson PC, Camacho SA. Thyroid hormone improves function and Ca2+ handling in pressure overload hypertrophy. Association with increased sarcoplasmic reticulum Ca2+-ATPase and alpha-myosin heavy chain in rat hearts. J Clin Invest. 1997; 100: 1742–1749.
Bisping E, Ikeda S, Kong SW, Tarnavski O, Bodyak N, McMullen JR, Rajagopal S, Son JK, Ma Q, Springer Z, Kang PM, Izumo S, Pu WT. Gata4 is required for maintenance of postnatal cardiac function and protection from pressure overload-induced heart failure. Proc Natl Acad Sci U S A. 2006; 103: 14471–14476.
Chen X, Shevtsov SP, Hsich E, Cui L, Haq S, Aronovitz M, Kerkela R, Molkentin JD, Liao R, Salomon RN, Patten R, Force T. The beta-catenin/T-cell factor/lymphocyte enhancer factor signaling pathway is required for normal and stress-induced cardiac hypertrophy. Mol Cell Biol. 2006; 26: 4462–4473.
Sayas CL, Ariaens A, Ponsioen B, Moolenaar WH. GSK-3 is activated by the tyrosine kinase Pyk2 during LPA1-mediated neurite retraction. Mol Biol Cell. 2006; 17: 1834–1844.