Cdc42 is an antihypertrophic molecular switch in the mouse heart

To improve contractile function, the myocardium undergoes hypertrophic growth without myocyte proliferation in response to both pathologic and physiologic stimulation. Various membrane-bound receptors and intermediate signal transduction pathways regulate the induction of cardiac hypertrophy, but the cardioprotective regulatory pathways or effectors that antagonize cardiac hypertrophy remain poorly understood. Here we identify the small GTPase Cdc42 as a signaling intermediate that restrained the cardiac growth response to physiologic and pathologic stimuli. Cdc42 was specifically activated in the heart after pressure overload and in cultured cardiomyocytes by multiple agonists. Mice with a heart-specific deletion of Cdc42 developed greater cardiac hypertrophy at 2 and 8 weeks of stimulation and transitioned more quickly into heart failure than did wild-type controls. These mice also displayed greater cardiac hypertrophy in response to neuroendocrine agonist infusion for 2 weeks and, more remarkably, enhanced exercise-induced hypertrophy and sudden death. These pathologies were associated with an inability to activate JNK following stimulation through a MEKK1/MKK4/MKK7 pathway, resulting in greater cardiac nuclear factor of activated T cells (NFAT) activity. Restoration of cardiac JNK signaling with an Mkk7 heart-specific transgene reversed the enhanced growth effect. These results identify what we believe to be a novel antihypertrophic and protective cardiac signaling pathway, whereby Cdc42-dependent JNK activation antagonizes calcineurin-NFAT activity to reduce hypertrophy and prevent transition to heart failure.

The adult heart hypertrophies in response to environmental stress as a means of augmenting pump function and reducing wall stress (1). The heart undergoes pathologic cardiac hypertrophy in response to hypertension, genetic mutations in genes encoding sarcomeric proteins, or valvular disease or following myocardial infarction, all of which predispose to arrhythmia and sudden death and the development of heart failure (2, 3). The transition between pathological hypertrophy and heart failure is associated with a decrease of cardiac contractility, ventricular remodeling and fibrosis, as well as myocyte loss. The hypertrophic program is initiated at the cell membrane by receptors that bind various neurohormonal factors to activate specific intracellular signaling pathways. One of the most highly characterized hypertrophic signaling cascades involves the Ca²⁺-dependent serine/threonine protein phosphatase calcineurin (4). Elevated intracellular Ca²⁺ levels activate calcineurin, which then dephosphorylates the transcription factor nuclear factor of activated T cells (NFAT) in the cytoplasm, resulting in its translocation to the nucleus and the subsequent activation of hypertrophy-associated genes (5). MAPKs, such as ERK, JNK, and p38, are also activated by membrane-bound receptors on cardiomyocytes, where they then regulate aspects of the cardiac hypertrophic response (6, 7).

Surprisingly, the link between neuroendocrine factor–sensing receptors and calcineurin or MAPK signaling pathways is not well understood, although the small G proteins may be involved (8, 9). Cdc42 is a ubiquitously expressed member of the Rho family of small GTPases that, together with Rac1 and RhoA, have been implicated as signaling effectors in the heart (8, 10, 11). Cdc42 cycles between a GDP-bound inactive state and a GTP-bound active state through the action of GTPase-activating proteins (GAPs), guanine nucleotide exchange factors (GEFs), and guanine nucleotide dissociation inhibitors (11–14). Cdc42 is dynamically regulated at the plasma membrane by GTP loading through cell type–specific GEFs and GAPs that respond to unique signals. Outside the heart, Cdc42 has multiple cellular functions, including the regulation of the actin and microtubule cytoskeleton formation that secondarily influences cellular morphology, polarity, migration, cell cycle progression, and cellular growth and death (11, 15). Analysis of gene-targeted mice has further verified many of these diverse functions for Cdc42, such as in maintaining the proper shape of keratinocytes and neuroepithelial cells as well as regulating hematopoietic stem cell maturation (16–20). By comparison, Cdc42GAP-deficient mice, which have enhanced Cdc42 activity, showed growth retardation, reduced cellular proliferation, and advanced aging (21, 22).

Very little is known about the function of Cdc42 in cardiomyocytes, although overexpression of Cdc42 was reported to enhance myocyte growth in culture, while overexpression of a dominant-negative Cdc42 mutant reduced leukemia inhibitory factor–induced (LIF-induced) myocyte growth (23). Here we investigate the role of Cdc42 in the adult mouse heart by using a conditional Cdc42-targeted allele in conjunction with a cardiac-specific cre transgenic line. In contrast to the single study in cultured myocytes discussed above (23), we observed that Cdc42 actually functions as an antihypertrophic molecular switch in the heart. Indeed, heart-specific deletion of Cdc42 enhances the cardiac growth response to both pathologic and physiological stimuli. Mechanistically, loss of Cdc42 renders JNK1/2 inactive to applied stimuli, and transgene-mediated restoration in cardiac JNK activity reduced the enhanced growth effect associated with Cdc42 deletion. The exacerbated hypertrophic response associated with Cdc42 deletion increased NFAT transcriptional activity, and since JNKs normally antagonize NFAT activity, these results suggest what we believe is a novel antihypertrophic signaling network in the heart.

Generation and characterization of Cdc^αMHC-cre mice. Conventional gene targeting of Cdc42 results in early embryonic lethality, precluding an analysis of Cdc42 function in adult disease-related processes (24). To circumvent this limitation, the Cdc42 gene was targeted with LoxP sites to permit tissue-specific deletion in a conditional manner through cre-mediated excision (17, 19, 20). We investigated the function of Cdc42 in the adult mouse heart by crossing the Cdc42-LoxP–targeted allele with transgenic mice expressing cre under the control of the cardiac-specific α-myosin heavy chain (α-MHC) promoter (Cdc^αMHC-cre mice). Cdc^αMHC-cre mice were born at the expected Mendelian ratios and were overtly normal. Hearts from 2-month-old Cdc^αMHC-cre mice showed an 88% deletion of Cdc42 protein compared with WT^αMHC-cre mice (Figure 1, A and B). Earlier and more efficient deletion of Cdc42 using the Nkx2.5-cre knockin allele resulted in early embryonic lethality, indicating that complete loss of Cdc42 from the heart at this early time point was not tolerated, necessitating our use of the α-MHC–cre transgene (data not shown). Deletion of Cdc42 in myocytes of the adult heart did not influence protein expression of the related small GTPase family members RhoA or Rac1 or the activated form of Rac1 bound to GTP (Figure 1C). Cardiac phenotyping in Cdc^αMHC-cre mice at 1 year of age revealed no difference in heart weight normalized to body weight (HW/BW), ventricular fractional shortening (FS) measured by echocardiography, or ventricular and cellular morphologies measured by histological methods (Figure 1, D–F, and Table 1). We also investigated whether deletion of Cdc42 from the heart might alter myocyte number. Hearts were harvested from 1-week-old Cdc^αMHC-cre and WT^αMHC-cre mice, and heart weights were measured and myocyte surface areas quantified from histological sections (Figure 1, G and H). No differences were noted in heart weights or cellular surface areas of Cdc^αMHC-cre mice compared with controls at 1 week of age, suggesting that total myocyte content was unchanged. Thus, deletion Cdc42 with the α-MHC–cre transgene did not appreciably alter the baseline phenotype of the heart.

Figure 1

Generation of cardiac-specific Cdc^αMHC-cre mice. (A) Western blot analysis of Cdc42 protein levels in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre hearts. GAPDH was used as a loading control. (B) Quantitation of the Western blot results shown in A. (C) Western blot analysis of total protein levels of RhoA and Rac1 in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre hearts. Rac baseline GTP loading for activity was also assayed. (D) Quantification of HW/BW at 1 year of age in WT^αMHC-cre and Cdc^αMHC-cre animals. (E) Echocardiographic analysis of left ventricular FS in 1-year-old WT^αMHC-cre and Cdc^αMHC-cre animals (n = 4 WT^αMHC-cre and n = 3 Cdc^αMHC-cre). (F) H&E staining of longitudinal sections from hearts of WT^αMHC-cre and Cdc^αMHC-cre mice at 1 year of age. (G) HW/BW at 1 week of age in WT^αMHC-cre (n = 4) and Cdc^αMHC-cre (n = 6) mice. (H) Myocyte cell surface areas in heart histological sections at 1 week of age in WT^αMHC-cre and Cdc^αMHC-cre mice. (I) Cdc42-GTP pull-down activity assay in neonatal rat cardiomyocytes stimulated for 10 minutes with the indicated hypertrophic agonists at 1 μM. A representative Western blot of 3 independent experiments is shown. (J) Rac1-GTP and Cdc42-GTP pull-down activity assays from hearts of sham- or TAC-operated WT mice subjected to TAC for 4, 7, or 14 days.

Table 1

Echocardiographic assessment of cardiac structure and function in 1-year-old WT^αMHC-cre and Cdc^αMHC-cre mice

We were also interested in determining whether endogenous Cdc42 might be activated in response to hypertrophic stimulation to the heart or myocytes in culture. Indeed, stimulation of neonatal rat cardiomyocytes in culture for 10 minutes with LIF, angiotensin II (Ag), phenylephrine (PE), or isoproterenol (Iso) each increased the GTP-bound, activated state of Cdc42 (Figure 1I). Cdc42 activation (GTP bound) was also increased in hearts of WT mice subjected to transverse aortic constriction (TAC) for 4, 7, and 14 days compared with hearts from sham-operated mice (Figure 1J). By comparison, Rac1 activation was not increased and in fact appeared to be decreased in the hearts of mice subjected to TAC (Figure 1J).

Deletion of Cdc42 from the adult heart exacerbates pressure overload hypertrophy. Given the results presented above, we hypothesized that Cdc42 might be more specialized as a signaling adaptor that connects neuroendocrine and stress-related membrane receptor activation to downstream pathways that program hypertrophy and/or dilated cardiomyopathy. Pressure overload by TAC was therefore performed in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice. Mice were sacrificed 2 weeks later and showed significant increases in the HW/BW ratio in both groups compared with sham-operated controls (Figure 2A). Interestingly, the amount of cardiac hypertrophy in Cdc^αMHC-cre mice was significantly greater than in WT^αMHC-cre mice (Figure 2A). This greater hypertrophy response was not due to a difference in the TAC procedure, as the TAC pressure gradients were similar between the 2 groups (Table 2). Analysis of myocyte areas from histological sections after 2 weeks of TAC also revealed greater cellular hypertrophy in Cdc^αMHC-cre myocytes compared with WT^αMHC-cre (Figure 2B). The exacerbated hypertrophy observed in Cdc^αMHC-cre hearts was not initially associated with a reduction in FS after 2 weeks of TAC (Table 2). However, histological analysis of fibrosis by Masson’s trichrome staining showed a 6-fold increase in collagen deposition in the hearts of Cdc^αMHC-cre mice after TAC compared with WT^αMHC-cre (Figure 2, C and D). Deletion of Cdc42 with the α-MHC–cre transgene also enhanced levels of apoptosis and necrosis after TAC stimulation compared with controls. Apoptosis was directly measured by TUNEL from histological sections (Figure 2E), while necrosis was indirectly evaluated by assessing Evan’s blue dye (EBD) uptake in vivo, followed by histological analysis (Figure 2F). A mouse model of muscular dystrophy with known membrane leakiness, mdx, was used as a control for EBD uptake. Both Cdc^αMHC-cre and mdx mice showed baseline EBD uptake that was significantly enhanced by TAC (Figure 2F). Thus, deletion of Cdc42 from the heart promoted greater cardiac hypertrophy following 2 weeks of TAC stimulation, and while function was not compromised at this time point, other indices of disease were enhanced.

Figure 2

Exacerbated hypertrophy in Cdc^αMHC-cre hearts. (A) Quantification of HW/BW in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice subjected to TAC for 2 weeks or a sham surgical procedure. The number of mice analyzed in each group is shown in each bar. *P < 0.05 versus sham; ^#P < 0.05 versus TAC WT^αMHC-cre. (B) Quantification of myocyte surface area from cardiac histological sections of sham- or TAC-operated WT^αMHC-cre and Cdc^αMHC-cre mice (n ≥ 500 cells from at least 3 different animals per group). *P < 0.05 versus sham; ^#P < 0.05 versus TAC WT^αMHC-cre. (C) Histological view of cardiac fibrosis stained in blue by Masson’s trichrome in WT^αMHC-cre and Cdc^αMHC-cre mice subjected to TAC for 2 weeks (original magnification, ×200). (D) Quantitation of fibrosis induced by 2 weeks of TAC in hearts of WT^αMHC-cre and Cdc^αMHC-cre mice (n = 3). ^#P < 0.05 versus WT^αMHC-cre. (E) Immunohistochemical assessment of TUNEL in the hearts of WT^αMHC-cre and Cdc^αMHC-cre mice subjected to 1 week of TAC (at least 100,000 nuclei were counted; n = 3 mice). (F) Quantification of EBD uptake in WT^αMHC-cre and Cdc^αMHC-cre hearts subjected to TAC for 1 week or a sham procedure. Dystrophin-deficient mdx mice were used as a positive control (at least 100,000 cells from 3 different mice were analyzed). *P < 0.05 versus sham; ^#P < 0.05 versus TAC WT^αMHC-cre.

Table 2

Echocardiographic assessment of cardiac structure and function before and after 2 weeks of TAC

Cardiac-specific Cdc^αMHC-cre mice are more susceptible to heart failure. To more carefully extend the results presented above after 2 weeks of TAC, we subjected 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice to long-term (8-week) pressure overload stimulation. Eight weeks of TAC also produced significantly greater increases in HW/BW ratios and myocyte cross-sectional areas in Cdc^αMHC-cre mice compared with WT^αMHC-cre controls (Figure 3, A and B). After 8 weeks of TAC, WT^αMHC-cre mice showed a significant reduction in FS, yet this decrease was significantly greater in Cdc^αMHC-cre mice, suggesting that deletion of Cdc42 in the heart predisposed to failure (Figure 3C). Measurement of TAC pressure gradients was performed after 2 weeks of TAC in these mice, revealing no differences (data not shown). Lung weight normalized to body weight (LW/BW) was significantly elevated in Cdc^αMHC-cre mice after 8 weeks of TAC, indicating pulmonary congestion due to poor cardiac function (Figure 3D). Consistent with this increased cardiac disease profile, cardiac fibrosis was greater in Cdc^αMHC-cre mice compared with WT^αMHC-cre mice after 8 weeks of TAC (Figure 3E). Analysis of hypertrophic/stress responsive gene expression in the heart by RT-PCR analysis also showed a greater response in Cdc^αMHC-cre mice after 8 weeks of TAC compared with WT^αMHC-cre mice (Figure 3F). Collectively, these results indicate that deletion of Cdc42 from the heart with the α-MHC–cre transgene predisposed to a greater hypertrophic response that transitioned more readily to heart failure, suggesting that Cdc42 is normally protective and antihypertrophic.

Figure 3

Cdc^αMHC-cre mice are more susceptible than WT^αMHC-cre mice to heart failure. (A) Quantification of HW/BW in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice subjected to TAC for 8 weeks compared with sham operation. The number of mice analyzed in each group is shown inside the bars. (B) Myocyte surface areas from histological sections of mice described in A (n ≥ 500 cells from at least 3 mice per group). (C) Echocardiographic assessment of FS after a sham procedure or 8 weeks of TAC for the mice described in A and B. (D) Quantification of LW/BW after 8 weeks of TAC for the mice described in A and B. (E) Quantitation of fibrosis induced by 8 weeks of TAC in hearts of WT^αMHC-cre and Cdc^αMHC-cre mice (n = 3). *P < 0.05 versus sham; ^#P < 0.05 versus TAC WT^αMHC-cre. (F) RT-PCR of mRNA from the hearts of mice described in A for ANP, BNP, β-MHC, α-Ska, and the L7 control RNA.

Cardiac-specific Cdc^αMHC-cre mice have exacerbated pathologic and physiologic hypertrophy. In addition to pressure overload stimulation, we also investigated the response of Cdc^αMHC-cre mice to neuroendocrine infusion and physiologic hypertrophy stimulation by exercise. First, chronic Ag and PE co-infusion was performed for 2 weeks with Alzet minipumps, revealing a significantly greater hypertrophic response in Cdc^αMHC-cre mice compared with WT^αMHC-cre control mice (Figure 4, A and B). Cdc^αMHC-cre and control mice were also subjected to forced swimming exercise daily for 3 weeks. Remarkably, two-thirds of the Cdc^αMHC-cre mice died suddenly during the swimming protocol, compared with no death in WT^αMHC-cre controls or rested mice (Figure 4C). While we are uncertain exactly why Cdc^αMHC-cre mice perished during exercise, the remaining mice in this cohort showed significantly greater increases in HW/BW ratios and myocyte surface areas from histological sections compared with the WT^αMHC-cre control group (Figure 4, D and E). Despite this greater increase in hypertrophy following swimming, cardiac FS was not reduced in Cdc^αMHC-cre mice, nor was there any evidence of histopathology such as fibrosis or TUNEL (Figure 4, F–H). Taken together, these results further suggest that Cdc42 is an antihypertrophic mediator that also functions during physiologic exercise stimulation to partially limit heart growth.

Figure 4

Cdc^αMHC-cre mice show greater hypertrophy to agonist and exercise stimulation. (A) Quantitation of HW/BW in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice stimulated for 2 weeks with Ag (432 μg/kg/d) and PE (100 mg/kg/d) or vehicle (Veh) released by Alzet osmotic mini-pumps. The number of mice analyzed in each group is shown inside each bar. *P < 0.05 versus vehicle; ^#P < 0.05 versus Ag/PE WT^αMHC-cre. (B) Quantitation of myocyte surface areas from the mice shown in A. At least 500 myocytes were quantified from 3 different mice in each group. *P < 0.05 versus vehicle; ^#P < 0.05 versus Ag/PE WT^αMHC-cre. (C) Survival in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre subjected to 3 weeks of daily swimming (initially, n = 7 WT^αMHC-cre, n = 12 Cdc^αMHC-cre; P < 0.001). (D) Quantitation of HW/BW in resting WT^αMHC-cre and Cdc^αMHC-cre mice and in the remaining exercised animals. *P < 0.05 versus rest; ^#P < 0.05 versus swim WT^αMHC-cre. (E) Quantitation of myocyte surface areas from the mice described in D. *P < 0.05 versus rest; ^#P < 0.05 versus swim WT^αMHC-cre. (F) Echocardiographic assessment of FS in WT^αMHC-cre and Cdc^αMHC-cre mice over 11 days of swimming. (G and H) Analysis of cardiac fibrosis and TUNEL after swimming in heart histological sections from mice in the indicated groups.

Deletion of Cdc42 from the heart alters JNK signaling. To investigate the potential mechanism whereby loss of Cdc42 promoted greater hypertrophy and predisposition to heart failure, we surveyed downstream signaling pathways. Most notably, Cdc42 was previously shown to affect MAPK signaling pathways, which are known to affect cellular growth (11, 25). Here we observed a prominent defect in JNK phosphorylation in the hearts of Cdc^αMHC-cre mice after 2 and 7 days of TAC stimulation compared with WT^αMHC-cre controls, while phosphorylation of p38, ERK1/2, Akt, and glycogen synthase kinase 3α/β (GSK3α/β) were unaltered (Figure 5A). We repeated the assay for phosphorylated JNK a second time from different extracts and again observed a defect in activation after TAC in Cdc^αMHC-cre mice (Figure 5A). Moreover, direct examination of JNK activity by measuring phosphorylation of GST-cJun in vitro showed a significant reduction in the hearts from Cdc^αMHC-cre mice after TAC (Figure 5B). We also instituted an in vitro approach in cultured neonatal rat cardiomyocytes using a lentivirus-encoding shRNA construct against Cdc42 to reduce the expression/activity of this protein. The shRNA reduced Cdc42 protein levels by approximately 74%, which correlated with a reduction in JNK phosphorylation after LIF and Iso stimulation (Figure 5C). We used these samples to examine the activity of upstream signaling components that might directly respond to Cdc42. Interestingly, the direct Cdc42 target, MEKK1, also showed reduced phosphorylation after agonist stimulation in the presence of the Cdc42 shRNA (Figure 5C). Downstream of MEKK1 but directly upstream of JNK1/2, MKK7 phosphorylation was similarly reduced with Cdc42 shRNA lentiviral infection after agonist stimulation (Figure 5C). We also observed a similar reduction in MKK4 activation after LIF stimulation (data not shown). Taken together, these data suggest that Cdc42 signals directly to MEKK1 in cardiomyocytes, which in turn alters MKK4/7 activity, leading to the observed alteration in JNK activity. Indeed, Sadoshima and colleagues showed that MEKK1^–/– mice had significantly reduced JNK activation in the heart after TAC stimulation (26). Finally, no changes in p21-activated kinases (PAK) or focal adhesion kinase (FAK) were observed in this assay (Figure 5C).

Figure 5

Inhibited JNK activation following TAC in Cdc^αMHC-cre mice. (A) Western blotting from heart protein extracts for phosphorylated and total JNK, p38, ERK1/2, Akt, and GSK3 in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice subjected to TAC for 2 days (T2) or 7 days (T7) versus sham-operated mice. A representative Western blot of 3 independent experiments is shown. A second Western blot for phospho-JNK1/2 was generated from different mice. (B) Quantitation of JNK1/2 activity determined by radioactive detection of GST-cJun phosphorylation from protein extracts after 1 week of TAC in WT^αMHC-cre or Cdc^αMHC-cre hearts. *P < 0.05 versus WT^αMHC-cre. (C) Western analysis of total Cdc42 protein levels and various kinases in protein extracts generated from neonatal rat cardiomyocytes stimulated with Iso (1 μM) or LIF (1 μM) or infected with a lentivirus containing a shRNA against Cdc42.

JNK is a critical effector of Cdc42 in regulating cardiac hypertrophy. JNK was selected for more thorough investigation because we previously showed that inhibition of this kinase promoted greater cardiac hypertrophy in both dominant-negative transgenic mice and Jnk1/2 gene–targeted mice (27). We used α-MHC–MKK7 transgenic mice, which are known to have a 4-fold increase in JNK activity (28), to enhance JNK signaling in the Cdc^αMHC-cre background. At baseline, MKK7 transgenic mice did not present with appreciable cardiac hypertrophy when crossed into the Cdc42-loxP–targeted background, either with or without the α-MHC–cre transgene (Figure 6, A and C). However, the enhanced growth response observed in Cdc^αMHC-cre mice after 2 weeks of TAC was reduced to WT^αMHC-cre TAC levels with the MKK7 transgene (Figure 6, A and C). All cohorts subjected to TAC showed similar TAC pressure gradients, indicating that the observed differences in organ and cellular hypertrophy was not caused by disparity in the pressure overload stimulation (Figure 6B). Similarly, the enhancement in cardiac fibrosis in Cdc^αMHC-cre mice after TAC was reduced to baseline levels by the MKK7 transgene (Figure 6E). Taken together, these results suggest that reestablishing JNK signaling downstream of Cdc42 deletion normalizes the cardiac hypertrophic response in Cdc^αMHC-cre hearts.

Figure 6

Restoring JNK activity in Cdc^αMHC-cre mice with an MKK7 transgene normalizes the cardiac hypertrophic response. (A) Quantitation of HW/BW in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice carrying either 1 or 2 transgenes (α-MHC–cre and/or α-MHC–MKK7) subjected to 2 weeks of TAC or a sham procedure. The number of mice analyzed in each group is shown inside the bars. *P < 0.05 versus sham; ^†P < 0.05 versus TAC WT^αMHC-cre; ^#P < 0.05 versus TAC Cdc^αMHC-cre. (B) Quantitation of TAC pressure gradients in the different experimental groups described in A. (C) Quantitation of myocyte surface areas from histological sections of the different groups subjected to sham or TAC for 2 weeks (n ≥ 500 myocytes from at least 3 mice in each group) *P < 0.05 versus sham; ^†P < 0.05 versus TAC WT^αMHC-cre; ^#P < 0.05 versus TAC Cdc^αMHC-cre. (D) Quantitation of luciferase activity in 2-month-old WT^αMHC-cre and Cdc^αMHC-cre mice carrying the NFAT luciferase reporter transgene subjected to 2 weeks of TAC or sham. *P < 0.05 versus sham; ^#P < 0.05 versus TAC WT^αMHC-cre. (E) Quantitation of fibrosis in Masson’s trichrome–stained histological heart sections from sham- and TAC-operated mice shown in A (n = 3 mice in each group). *P < 0.05 versus sham; ^†P < 0.05 versus TAC WT^αMHC-cre; ^#P < 0.05 versus TAC Cdc^αMHC-cre. (F) Quantitation of HW/BW in sham-operated double-transgenic WT^αMHC-cre and Cdc^αMHC-cre mice subjected to sham operation or TAC for 2 weeks. *P < 0.05 versus sham; ^#P < 0.05 versus TAC WT^αMHC-cre.

One potential mechanism whereby decreased JNK activity might promote greater cardiac hypertrophy is by permitting greater NFAT activity following stress stimulation. Indeed, JNK is known to directly phosphorylate NFAT transcription factors to antagonize their nuclear translocation and induction of hypertrophic genes (27, 29, 30). We previously showed that JNK-inhibited mice had greater NFAT activity in the heart (27). Here we introduced the NFAT-luciferase reporter transgene in the WT^αMHC-cre and Cdc^αMHC-cre backgrounds. As predicted, NFAT-luciferase activity was significantly higher in Cdc^αMHC-cre hearts compared with WT^αMHC-cre after 2 weeks of TAC stimulation (Figure 6D). Importantly, the NFAT-luciferase transgene was neutral and did not influence the hypertrophic response of WT^αMHC-cre and Cdc^αMHC-cre mice following 2 weeks of TAC, except that Cdc^αMHC-cre mice continued to show greater hypertrophy (Figure 6F). In conclusion, targeted deletion of the small GTPase Cdc42 from the adult myocardium allows greater cardiac hypertrophy following stress stimulation, in part through defective JNK signaling, which renders NFAT more active.

Here we show that the loss of Cdc42 renders the mouse heart more capable of hypertrophic growth following pressure overload, neuroendocrine agonist infusion, and exercise, suggesting that it normally functions to antagonize cardiac growth when activated. This proposed function for Cdc42 is unique among the Rho GTPases, as Ras, Rac1, and RhoA have all been shown to facilitate cardiac hypertrophy or myopathy in genetically modified mice. For example, overexpression of activated Ras in the hearts of transgenic mice generated cardiac hypertrophy that was pathologic and associated with reduced sarcoplasmic reticulum calcium handling (31). Similarly, Rac1 overexpression in transgenic mice induced early cardiac hypertrophy or dilated heart failure (32). In support of these results, in one study, cardiac-specific deletion of Rac1 in mice resulted in reduced cardiac growth at baseline associated with reduced NADPH oxidase activity (33). In contrast, another study found that RhoA overexpression in the hearts of transgenic mice did not produce left ventricular hypertrophy, although it did mediate atrial enlargement, alterations in the cardiac conduction system, and ventricular dilation with aging (34). By comparison, in the present study, Cdc42 appeared to be functionally distinct from the other Rho GTPases, where it served as an antihypertrophic and protective mediator during stress stimulation, and it had no reported linkage to NADPH oxidase activity. However, the observed protective and antihypertrophic function for Cdc42 in the heart is inconsistent with studies conducted in cultured cardiomyocytes with adenoviral gene transfer of dominant-negative and constitutively active Cdc42 mutants. Those studies suggested that Cdc42 regulates myocyte growth by the addition of sarcomeres in series, although the disparity with our study may simply reflect the different experimental systems employed (23).

While a large number of pro-hypertrophic signaling pathways have been proposed and validated in genetically modified mouse models, surprisingly few antihypertrophic signaling pathways have been described. Perhaps the only well-described antihypertrophic regulatory pathway to date is that involving cGMP. Increased levels of cGMP in cardiomyocytes through stimulation of the atrial natriuretic peptide (ANP) receptor, enhanced activity of cytosolic guanylyl cyclases through nitric oxide, or inhibition of select phosphodiesterases (PDEs) renders myocytes less able to hypertrophy to stress or neuroendocrine agonist stimulation. Increased intracellular cGMP results in cGMP-dependent protein kinase type I (PKG) activation and inhibition of cardiac hypertrophy, in part, by inhibition of calcineurin-NFAT signaling (35). Indeed, inhibition of PDE5A with sildenafil, which increases total cGMP and leads to PKG activation, renders the heart protected from pressure overload stimulation and subsequent hypertrophy (36).

In addition to cGMP/PKG signaling, terminal activation of p38 or JNK may have an antihypertrophic effect on the heart under select conditions. For example, expression of dominant-negative p38 or JNK in the hearts of transgenic mice, or even targeted deletion of Jnk1/2 or p38α permitted greater cardiac hypertrophy following pressure overload stimulation, exercise, or with neuroendocrine-like agonists (27, 37, 38). This increase in cardiac hypertrophy associated with p38 or JNK inhibition was reversed by deletion of calcineurin Aβ, showing that calcineurin-NFAT signaling was a mechanism whereby these kinases mediated their antihypertrophic effect. That JNK and p38 can have an antihypertrophic effect is also somewhat consistent with the phenotype of mice with MKK3/6 or MKK7 overexpression in the heart, which showed severe cardiac dilation and failure without prominent hypertrophy (39, 40). We also previously showed that Jnk1^+/–Jnk2^–/– mice developed spontaneous cardiac hypertrophy with aging and that young mice subjected to TAC had a greater hypertrophic response (27). The latter result is especially important, as it is obviously specific to JNK signaling and is not confounded by divergent signaling that characterizes other effectors used to examine this pathway. For example, overexpression of dominant-negative MKK4 diminished cardiac hypertrophy in a rat model of pressure overload, although this effector is not completely JNK specific (41). Our results are consistent with data from MEKK1^–/– mice, which, like Cdc^αMHC-cre mice, showed less JNK activation after TAC stimulation and greater cardiac hypertrophy at the 2-week time point (26). However, Schneider and colleagues showed that hypertrophy and heart disease observed in Gαq-overexpressing transgenic mice was rescued by crossing into the MEKK1^–/– background (42). These seemingly discordant results are obviously confusing, but perhaps JNK signaling plays both roles (pro- and antihypertrophic), depending on the activity of parallel regulatory pathways, the duration and/or strength of JNK signaling, or the intracellular location of activated JNK associated with different upstream effector pathways. It is also formally possible that other signaling effectors are more critical in regulating cardiac hypertrophy downstream of Cdc42.

Our results suggest a straightforward pathway in which Cdc42 receives signals from membrane-bound receptors to initiate the JNK cascade through MEKK1. Indeed, MEKK1 is known to associate with Cdc42 and to be directly activated by it (43), and others have shown a prominent link between Cdc42 and JNK activity (44, 45). Once activated by Cdc42, MEKK1 can directly phosphorylate and activate MKK4 and MKK7, which in turn directly phosphorylate JNK1/2. Previous work in mouse embryonic fibroblasts showed that loss of Cdc42 attenuates JNK phosphorylation, whereas activation of Cdc42 in Cdc42GAP^–/– mouse embryonic fibroblasts promotes JNK phosphorylation (22). Taken together with our observations, we can conclude that Cdc42 likely regulates JNK signaling directly through MEKK1 in cardiomyocytes, although we cannot rule out an additional effect from MEKK4 or other upstream kinases that might bind Cdc42.

Given that standard deletion of Cdc42 is embryonic lethal, the generation of LoxP-targeted Cdc42 mice was critical for determining its functional role in vivo. Previous analysis of these mice, or of mice with deletion of Cdc42GAP, revealed altered JNK signaling and effects on cell and organ size or cellular proliferation (21, 22). Cdc42 typically regulates cellular morphology, migration, and cell cycle progression by directly controlling actin and microtubule cytoskeleton formation and remodeling (11, 15). However, cardiomyocytes are unique in that they are structurally restrained and not typically capable of migration and cell division in adulthood. Not surprisingly, we failed to observe any alteration in myocyte number in adult hearts or hearts analyzed at 7 days after birth using the α-MHC–cre approach, suggesting that the lack of Cdc42 in cardiomyocytes did not influence cell number. However, myocytes can change their morphology in response to stress stimulation, and Cdc42 might help regulate the hypertrophy of myocytes by controlling non-contractile actin and microtubules at the cellular periphery, where sarcomeres are added or subtracted. Indeed, Cdc42 can respond to signals from integrins, such as integrin α2/β1 in kidney cells (46). However, while we believe that loss of JNK responsiveness in Cdc^αMHC-cre mice is a primary mechanism responsible for the increased hypertrophic response, we cannot rule out other potential signaling effects downstream of Cdc42. Moreover, the rescue of enhanced hypertrophy in Cdc^αMHC-cre mice with MKK7 overexpression does not prove that JNK is a direct pathway recipient of Cdc42 signaling in controlling cardiac hypertrophy, as these effects could be parallel.

The observation that most Cdc^αMHC-cre mice died during exercise stimulation was unexpected, as was the greater hypertrophic response. Even though other indices of pathologic hypertrophy were not observed in Cdc^αMHC-cre mice after exercise, such as fibrosis and a reduction in cardiac performance, the increase in mortality suggests some sort of maladaptation (although we were not able to determine the exact defect). While physiological hypertrophy is certainly nonpathological in humans, too much of a physiological stimulus might be maladaptive (47). Indeed, overexpression of myristoylated or phosphomimetic forms of AKT1, a major transducer of physiological hypertrophy, induces pathological features in the heart such as interstitial fibrosis and impaired cardiac function (48, 49). In this manner, even physiologic stimulation of hypertrophy could have a maladaptive component that is normally counteracted by protective pathways. Cdc42 signaling is potentially one of those protective pathways, such that rigorous exercise training without this effector leads to either arrhythmia and sudden death or greater hypertrophy. Thus, Cdc42 likely plays a unique signaling role in the heart, whereby its activation in response to both pathologic and physiologic stimuli is beneficial and antihypertrophic. This suggests that pharmacologic strategies designed to increase Cdc42 activity might be of therapeutic value in humans.

This work was supported by the NIH (grants to J.D. Molkentin and Y. Zheng). This work was also supported by an international grant in heart failure research from the Fondation Leducq (to J.D. Molkentin).

Address correspondence to: Jeffery D. Molkentin, Cincinnati Children’s Hospital Medical Center, Molecular Cardiovascular Biology, Howard Hughes Medical Institute, 240 Albert Sabin Way, MLC 7020, Cincinnati, Ohio 45229, USA. Phone: (513) 636-3557; Fax (513) 636-5958; E-mail: jeff.molkentin@cchmc.org.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J. Clin. Invest.119:3079–3088 (2009). doi:10.1172/JCI37694.

1. Claycomb, W.C. 1983. Cardiac muscle cell proliferation and cell differentiation in vivo and in vitro. Adv. Exp. Med. Biol. 161:249-265.
   View this article via: PubMed Google Scholar
2. Levy, D., Garrison, R.J., Savage, D.D., Kannel, W.B., Castelli, W.P. 1990. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N. Engl. J. Med. 322:1561-1566.
   View this article via: PubMed Google Scholar
3. Ho, K.K., Pinsky, J.L., Kannel, W.B., Levy, D. 1993. The epidemiology of heart failure: the Framingham Study. J. Am. Coll. Cardiol. 22:6A-13A.
   View this article via: PubMed Google Scholar
4. Molkentin, J.D., et al. 1998. A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell. 93:215-228.
   View this article via: CrossRef PubMed Google Scholar
5. Heineke, J., Molkentin, J.D. 2006. Regulation of cardiac hypertrophy by intracellular signalling pathways. Nat. Rev. Mol. Cell Biol. 7:589-600.
   View this article via: PubMed Google Scholar
6. Liang, Q., Molkentin, J.D. 2003. Redefining the roles of p38 and JNK signaling in cardiac hypertrophy: dichotomy between cultured myocytes and animal models. J. Mol. Cell. Cardiol. 35:1385-1394.
   View this article via: CrossRef PubMed Google Scholar
7. Bueno, O.F., Molkentin, J.D. 2002. Involvement of extracellular signal-regulated kinases 1/2 in cardiac hypertrophy and cell death. Circ. Res. 91:776-781.
   View this article via: CrossRef PubMed Google Scholar
8. Brown, J.H., Del Re, D.P., Sussman, M.A. 2006. The Rac and Rho hall of fame: a decade of hypertrophic signaling hits. Circ. Res. 98:730-742.
   View this article via: CrossRef PubMed Google Scholar
9. Lezoualc’h, F., Metrich, M., Hmitou, I., Duquesnes, N., Morel, E. 2008. Small GTP-binding proteins and their regulators in cardiac hypertrophy. J. Mol. Cell. Cardiol. 44:623-632.
   View this article via: CrossRef PubMed Google Scholar
10. Adams, A.E., Johnson, D.I., Longnecker, R.M., Sloat, B.F., Pringle, J.R. 1990. CDC42 and CDC43, two additional genes involved in budding and the establishment of cell polarity in the yeast Saccharomyces cerevisiae. J. Cell Biol. 111:131-142.
    View this article via: CrossRef PubMed Google Scholar
11. Jaffe, A.B., Hall, A. 2005. Rho GTPases: biochemistry and biology. Annu. Rev. Cell Dev. Biol. 21:247-269.
    View this article via: CrossRef PubMed Google Scholar
12. Olofsson, B. 1999. Rho guanine dissociation inhibitors: pivotal molecules in cellular signalling. Cell Signal. 11:545-554.
    View this article via: CrossRef PubMed Google Scholar
13. Schmidt, A., Hall, A. 2002. Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev. 16:1587-1609.
    View this article via: CrossRef PubMed Google Scholar
14. Bernards, A. 2003. GAPs galore! A survey of putative Ras superfamily GTPase activating proteins in man and Drosophila. Biochim. Biophys. Acta. 1603:47-82.
    View this article via: PubMed Google Scholar
15. Etienne-Manneville, S., Hall, A. 2002. Rho GTPases in cell biology. Nature. 420:629-635.
    View this article via: CrossRef PubMed Google Scholar
16. Wu, X., et al. 2006. Cdc42 controls progenitor cell differentiation and beta-catenin turnover in skin. Genes Dev. 20:571-585.
    View this article via: CrossRef PubMed Google Scholar
17. Chen, L., et al. 2006. Cdc42 deficiency causes Sonic hedgehog-independent holoprosencephaly. Proc. Natl. Acad. Sci. U. S. A. 103:16520-16525.
    View this article via: CrossRef PubMed Google Scholar
18. Cappello, S., et al. 2006. The Rho-GTPase cdc42 regulates neural progenitor fate at the apical surface. Nat. Neurosci. 9:1099-1107.
    View this article via: CrossRef PubMed Google Scholar
19. Yang, L., et al. 2007. Rho GTPase Cdc42 coordinates hematopoietic stem cell quiescence and niche interaction in the bone marrow. Proc. Natl. Acad. Sci. U. S. A. 104:5091-5096.
    View this article via: CrossRef PubMed Google Scholar
20. Yang, L., et al. 2007. Cdc42 critically regulates the balance between myelopoiesis and erythropoiesis. Blood. 110:3853-3861.
    View this article via: CrossRef PubMed Google Scholar
21. Wang, L., Yang, L., Debidda, M., Witte, D., Zheng, Y. 2007. Cdc42 GTPase-activating protein deficiency promotes genomic instability and premature aging-like phenotypes. Proc. Natl. Acad. Sci. U. S. A. 104:1248-1253.
    View this article via: CrossRef PubMed Google Scholar
22. Yang, L., Wang, L., Zheng, Y. 2006. Gene targeting of Cdc42 and Cdc42GAP affirms the critical involvement of Cdc42 in filopodia induction, directed migration, and proliferation in primary mouse embryonic fibroblasts. Mol. Biol. Cell. 17:4675-4685.
    View this article via: CrossRef PubMed Google Scholar
23. Nagai, T., et al. 2003. Cdc42 plays a critical role in assembly of sarcomere units in series of cardiac myocytes. Biochem. Biophys. Res. Commun. 305:806-810.
    View this article via: CrossRef PubMed Google Scholar
24. Chen, F., et al. 2000. Cdc42 is required for PIP(2)-induced actin polymerization and early development but not for cell viability. Curr. Biol. 10:758-765.
    View this article via: CrossRef PubMed Google Scholar
25. Bishop, A.L., Hall, A. 2000. Rho GTPases and their effector proteins. Biochem. J. 348:241-255.
    View this article via: CrossRef PubMed Google Scholar
26. Sadoshima, J., et al. 2002. The MEKK1-JNK pathway plays a protective role in pressure overload but does not mediate cardiac hypertrophy. J. Clin. Invest. 110:271-279.
    View this article via: JCI PubMed Google Scholar
27. Liang, Q., et al. 2003. c-Jun N-terminal kinases (JNK) antagonize cardiac growth through cross-talk with calcineurin-NFAT signaling. EMBO J. 22:5079-5089.
    View this article via: CrossRef PubMed Google Scholar
28. Kaiser, R.A., et al. 2005. Genetic inhibition or activation of JNK1/2 protects the myocardium from ischemia-reperfusion-induced cell death in vivo. J. Biol. Chem. 280:32602-32608.
    View this article via: CrossRef PubMed Google Scholar
29. Chow, C.W., Rincon, M., Cavanagh, J., Dickens, M., Davis, R.J. 1997. Nuclear accumulation of NFAT4 opposed by the JNK signal transduction pathway. Science. 278:1638-1641.
    View this article via: CrossRef PubMed Google Scholar
30. Porter, C.M., Havens, M.A., Clipstone, N.A. 2000. Identification of amino acid residues and protein kinases involved in the regulation of NFATc subcellular localization. J. Biol. Chem. 275:3543-3551.
    View this article via: CrossRef PubMed Google Scholar
31. Zheng, M., et al. 2004. Sarcoplasmic reticulum calcium defect in Ras-induced hypertrophic cardiomyopathy heart. Am. J. Physiol. Heart Circ. Physiol. 286:H424-H433.
    View this article via: PubMed Google Scholar
32. Sussman, M.A., et al. 2000. Altered focal adhesion regulation correlates with cardiomyopathy in mice expressing constitutively active rac1. J. Clin. Invest. 105:875-886.
    View this article via: JCI CrossRef PubMed Google Scholar
33. Satoh, M., et al. 2006. Requirement of Rac1 in the development of cardiac hypertrophy. Proc. Natl. Acad. Sci. U. S. A. 103:7432-7437.
    View this article via: CrossRef PubMed Google Scholar
34. Sah, V.P., et al. 1999. Cardiac-specific overexpression of RhoA results in sinus and atrioventricular nodal dysfunction and contractile failure. J. Clin. Invest. 103:1627-1634.
    View this article via: JCI CrossRef PubMed Google Scholar
35. Fiedler, B., et al. 2002. Inhibition of calcineurin-NFAT hypertrophy signaling by cGMP-dependent protein kinase type I in cardiac myocytes. Proc. Natl. Acad. Sci. U. S. A. 99:11363-11368.
    View this article via: CrossRef PubMed Google Scholar
36. Kass, D.A., Champion, H.C., Beavo, J.A. 2007. Phosphodiesterase type 5: expanding roles in cardiovascular regulation. Circ. Res. 101:1084-1095.
    View this article via: CrossRef PubMed Google Scholar
37. Braz, J.C., et al. 2003. Targeted inhibition of p38 MAPK promotes hypertrophic cardiomyopathy through upregulation of calcineurin-NFAT signaling. J. Clin. Invest. 111:1475-1486.
    View this article via: JCI PubMed Google Scholar
38. Taniike, M., et al. 2008. Apoptosis signal-regulating kinase 1/p38 signaling pathway negatively regulates physiological hypertrophy. Circulation. 117:545-552.
    View this article via: CrossRef PubMed Google Scholar
39. Liao, P., et al. 2001. The in vivo role of p38 MAP kinases in cardiac remodeling and restrictive cardiomyopathy. Proc. Natl. Acad. Sci. U. S. A. 98:12283-12288.
    View this article via: CrossRef PubMed Google Scholar
40. Petrich, B.G., et al. 2004. Targeted activation of c-Jun N-terminal kinase in vivo induces restrictive cardiomyopathy and conduction defects. J. Biol. Chem. 279:15330-15338.
    View this article via: CrossRef PubMed Google Scholar
41. Choukroun, G., et al. 1999. Regulation of cardiac hypertrophy in vivo by the stress-activated protein kinases/c-Jun NH(2)-terminal kinases. J. Clin. Invest. 104:391-398.
    View this article via: JCI CrossRef PubMed Google Scholar
42. Minamino, T., et al. 2002. MEKK1 is essential for cardiac hypertrophy and dysfunction induced by Gq. Proc. Natl. Acad. Sci. U. S. A. 99:3866-3871.
    View this article via: CrossRef PubMed Google Scholar
43. Fanger, G.R., Johnson, N.L., Johnson, G.L. 1997. MEK kinases are regulated by EGF and selectively interact with Rac/Cdc42. EMBO J. 16:4961-4972.
    View this article via: CrossRef PubMed Google Scholar
44. Coso, O.A., et al. 1995. The small GTP-binding proteins Rac1 and Cdc42 regulate the activity of the JNK/SAPK signaling pathway. Cell. 81:1137-1146.
    View this article via: CrossRef PubMed Google Scholar
45. Minden, A., Lin, A., Claret, F.X., Abo, A., Karin, M. 1995. Selective activation of the JNK signaling cascade and c-Jun transcriptional activity by the small GTPases Rac and Cdc42Hs. Cell. 81:1147-1157.
    View this article via: CrossRef PubMed Google Scholar
46. Yeh, Y.C., Wang, C.Z., Tang, M.J. 2009. Discoidin domain receptor 1 activation suppresses alpha(2)beta(1) integrin-dependent cell spreading through inhibition of Cdc42 activity. J. Cell. Physiol. 218:146-156.
47. Shiojima, I., Walsh, K. 2006. Regulation of cardiac growth and coronary angiogenesis by the Akt/PKB signaling pathway. Genes Dev. 20:3347-3365.
    View this article via: CrossRef PubMed Google Scholar
48. Matsui, T., et al. 2002. Phenotypic spectrum caused by transgenic overexpression of activated Akt in the heart. J. Biol. Chem. 277:22896-22901.
    View this article via: CrossRef PubMed Google Scholar
49. Shioi, T., et al. 2002. Akt/protein kinase B promotes organ growth in transgenic mice. Mol. Cell. Biol. 22:2799-2809.
    View this article via: CrossRef PubMed Google Scholar
50. Oka, T., et al. 2006. Cardiac-specific deletion of Gata4 reveals its requirement for hypertrophy, compensation, and myocyte viability. Circ. Res. 98:837-845.
    View this article via: CrossRef PubMed Google Scholar
51. Wilkins, B.J., et al. 2004. Calcineurin/NFAT coupling participates in pathological, but not physiological, cardiac hypertrophy. Circ. Res. 94:110-118.
    View this article via: CrossRef PubMed Google Scholar
52. Wilkins, B.J., et al. 2002. Targeted disruption of NFATc3, but not NFATc4, reveals an intrinsic defect in calcineurin-mediated cardiac hypertrophic growth. Mol. Cell. Biol. 22:7603-7613.
    View this article via: CrossRef PubMed Google Scholar
53. Shai, S.Y., et al. 2002. Cardiac myocyte-specific excision of the beta1 integrin gene results in myocardial fibrosis and cardiac failure. Circ. Res. 90:458-464.
    View this article via: CrossRef PubMed Google Scholar
54. Parsons, S.A., et al. 2004. Genetic loss of calcineurin blocks mechanical overload-induced skeletal muscle fiber type switching but not hypertrophy. J. Biol. Chem. 279:26192-26200.
    View this article via: CrossRef PubMed Google Scholar
55. De Windt, L.J., Lim, H.W., Haq, S., Force, T., Molkentin, J.D. 2000. Calcineurin promotes protein kinase C and c-Jun NH2-terminal kinase activation in the heart. Cross-talk between cardiac hypertrophic signaling pathways. J. Biol. Chem. 275:13571-13579.
    View this article via: CrossRef PubMed Google Scholar
56. Maillet, M., et al. 2003. Crosstalk between Rap1 and Rac regulates secretion of sAPPalpha. Nat. Cell Biol. 5:633-639.
    View this article via: PubMed Google Scholar