Activation of PKN mediates survival of cardiac myocytes in the heart during ischemia/reperfusion

Introduction

Although myocardial ischemia and reperfusion induce lethal injury in the heart, the heart and the cardiac myocytes therein have powerful endogenous mechanisms to protect themselves from energy deficiency, oxidative stress, protein aggregation and organelle malfunction, thereby minimizing myocardial injury ¹. Ischemic preconditioning is one of the strongest of these endogenous mechanisms of cardioprotection, and pharmacological intervention mimicking the ischemic preconditioning effect is considered to be a promising modality for the treatment of ischemic heart disease ¹. However, other endogenous mechanisms of cardioprotection activated during ischemia and reperfusion remain to be elucidated.

PKN, also known as protein kinase C-related kinase 1 (PRK1), is a serine (Ser)/threonine (Thr) kinase, with a molecular mass of 120KDa ². PKN consists of a catalytic domain highly homologous to that of protein kinase C (PKC) in the carboxyl-terminal region, and a regulatory domain in the amino-terminal region ². PKN localizes primarily in the cytosolic fraction and possesses a wide variety of functions, such as cytoskeletal regulation, cell adhesion, vesicle transport, and cell cycle regulation ³, and it has been shown to be involved in the pathogenesis of Alzheimer’s disease ⁴ and amyotrophic lateral sclerosis ⁵. Interestingly, PKN is cleaved by caspase 3 ⁶, and a truncated and kinase-active form of PKN has been found in ischemia/reperfusion (I/R) models of the rat retina ⁷ and brain ⁸, suggesting that PKN may modulate cell survival or death during I/R. Transgenic (Tg) mice overexpressing a constitutively activated (CA) form of PKN in mammary epithelium exhibit impaired tight junction sealing and increased apoptosis, resulting in precocious involution in the mammary gland ⁹.

PKN activates atrial natriuretic factor (ANF) gene expression in cardiac myocytes ¹⁰. However, the in vivo function of PKN is poorly understood in the heart. Thus, a major goal in this study was to elucidate the function of PKN in the heart in vivo. Our preliminary studies indicated that PKN is activated by stress, such as hypotonic stress and I/R. Therefore, we examined the role of PKN in regulating cell survival and death and myocardial injury in response to I/R, and the underlying molecular mechanisms mediating the action of PKN in cardiac myocytes. Our results suggest that activation of PKN is an endogenous mechanism of cardioprotection against I/R injury.

Materials and Methods

An expanded Methods section is available in the Online Data Supplement at http://circres.ahajournals.org. All animal protocols were approved by the review board of the Institutional Animal Care and Use Committee of the University of Medicine and Dentistry of New Jersey.

Results

Discussion

Our main findings in this work are 1) that PKN is activated by I/R in the heart in vivo and by oxidative stress in cardiac myocytes in vitro, 2) that activation of PKN causes cardiac hypertrophy without LV dysfunction, 3) that activation of PKN plays a cell protective role in cardiac myocytes during I/R in vivo and in response to oxidative stress in vitro, and 4) that PKN mediates its protective effects against I/R in part through activation of the alpha BC/proteasome pathway.

In our study, Thr774 phosphorylation of PKN, which is essential for its activation, was observed in hearts subjected to I/R in vivo. In yeast, Pkc1, whose amino-terminal regulatory region is highly homologous to that of PKN, is activated by cell wall stress, including hypotonic stress¹⁷. Pkc1 is also important for cell viability in the adaptive response to oxidative stress ^{18, 19} or nitrosative stress ¹⁹. I/R causes both oxidative and osmotic stress, both of which affect survival and death of cardiac myocytes in the heart ^{20, 21}. PKN is phosphorylated by H₂O₂ in cardiac myocytes in vitro, and PKN phosphorylation caused by I/R is inhibited in the presence of MPG, an antioxidant, in the heart in vivo (Online Fig. VIII), suggesting that oxidative stress plays an important role in mediating I/R-induced PKN activation. Furthermore, we found previously that hypotonic stress causes activation of PKN in vitro (manuscript in preparation by KK and JS). Thus, we speculate that osmotic stress may also mediate PKN activation in response to I/R. Interestingly, RhoA, a known activator of PKN ³, is activated by reactive oxygen species ²². Oxygen radicals can stimulate PKC directly by oxidative modification of its regulatory domain ²³. The role of RhoA and oxidative posttranslational modification of PKN in mediating activation of PKN during I/R remains to be elucidated.

Our findings suggest that activation of PKN in cardiac myocytes is necessary and sufficient for cardioprotection during I/R in vivo and in response to H₂O₂ in vitro. Although LV function in Tg-DNPKN mice was comparable to that in NTg mice at 3 months of age, nearly complete downregulation of PKN by Ad-sh-PKN induces cell death in cardiac myocytes in vitro (Online Fig. IX), suggesting that a low level of PKN is required for survival of cardiac myocytes. Thus, all lines of experimental evidence presented in this work support the notion that PKN promotes survival of cardiac myocytes. Since PKN reduces the number of TUNEL positive myocytes in the ischemic area after I/R, we speculate that the protective effect of PKN is mediated through suppression of apoptosis, but its effect upon other forms of cell death remains to be elucidated.

Since suppression of myocardial injury during I/R was exacerbated when activation of PKN was inhibited in Tg-DNPKN mice, activation of PKN during I/R protects the heart. However, since Tg-CAPKN mice have stronger and more persistent activation of PKN, the cardioprotective effect observed in Tg-CAPKN may also be mediated in part by preconditioning effects. Our preliminary results suggest that PKN is also activated by preconditioning (data not shown). Whether or not activation of endogenous PKN can achieve protection against prolonged ischemia or late preconditioning remains to be elucidated. Neither translocation of PKCε into the membrane fraction (Online Fig. X) nor Akt phosphorylation, common mediators of ischemic preconditioning, was induced by overexpression of PKN in cardiac myocytes in vitro (Online Fig. XI). Thus, it is likely that the cardioprotective effects of CAPKN are mediated by PKCε- and Akt-independent mechanisms. Development of small molecules that specifically stimulate PKN would be relevant clinically, considering the fact that stimulators of PKCε have thus far shown promising results in their preclinical studies ²⁴.

Tg-CAPKN mice showed mild LV hypertrophy. PKN is involved in actin reorganization in vascular smooth muscle cells ²⁵ and stimulates ANF gene expression in cardiac myocytes ¹⁰, integral features of cardiac hypertrophy. PKN has also been shown to interact directly with alpha-actinin ²⁶. Thus, activation of PKN by stress could stimulate cardiac hypertrophy in vivo. It should be noted that whether or not PKN is a physiological mediator of cardiac hypertrophy requires further evaluation with loss of function models of PKN. The presence of hypertrophy alone may secondarily affect the cardioprotective effect of PKN against I/R. However, since even short term activation or inactivation of PKN affects cell survival in a cell autonomous manner, the protective effect of PKN may be at least in part hypertrophy-independent.

Thus far, we have found that LV function in Tg-CAPKN mice is well maintained, even after over a year of follow up (data not shown). Together with the cardioprotective effect of PKN during I/R, we speculate that cardiac hypertrophy induced by PKN may be compensatory/physiological. However, this notion would require further testing by applying long-term hypertrophic stimulation, such as pressure overload, to Tg-CAPKN mice. In addition, the reduced heart rate could have resulted in a compensatory increase in ejection fraction, an index of LV contractility, in Tg-CAPKN mice. Thus, the effect of PKN upon the intrinsic contractility of cardiac myocytes should be evaluated.

As noted above, PKN physically interacts with RhoA ³, which positively regulates PKN. Transgenic mice with cardiac specific overexpression of constitutively active RhoA exhibit early lethality, and those with overexpression of wild type RhoA exhibit atrial enlargement, LV dilation and contractile failure, with a markedly reduced heart rate ²⁷. The phenotype of Tg-CAPKN mice appears to be quite different from either of the above, suggesting that PKN may not be a major effector of RhoA in the heart. It should be noted, however, that Tg-CAPKN mice do show mild bradycardia. Since RhoA regulates the function of ion channels, such as Kir2.1 ²⁸ and L type calcium channel ²⁹, PKN may participate in the regulation of ion channels by RhoA, thereby controlling heart rate.

Alpha BC, a member of the small heat shock protein and the molecular chaperone families, has protective effects against stresses ^{12, 30, 31}. Whereas some PKC isoforms phosphorylate alpha BC ¹⁵, PKN upregulates mRNA expression of alpha BC through heat shock factor-1 in HeLa S3 cells ¹³. Interestingly, although total expression of alpha BC did not differ between Tg-CAPKN and NTg hearts, both alpha BC phosphorylation at Ser59 and Ser45 and its expression in the cytoskeletal fraction were increased in Tg-CAPKN mice. Furthermore, I/R-induced increases in alpha BC phosphorylation at Ser59 and Ser45 were inhibited in Tg-DNPKN mice. These results are consistent with the notion that PKN plays an important role in mediating phosphorylation of alpha BC in response to I/R in the heart. Although phosphorylation of alpha BC at Ser59 can be mediated by activation of MAPKAPK-2 ³², MAPKAPK-2 is not activated in Tg-CAPKN mice (Online Fig. XII).

Although the specific role of PKN-induced Ser59 and Ser45 phosphorylation of alpha BC during I/R remains to be clarified, knockdown of alpha BC inhibited the protective effect of PKN in cardiac myocytes in response to H₂O₂, indicating that alpha BC mediates the cell protective effect of PKN. Phosphorylation of alpha BC at the Ser59 residue contributes to cytoprotection in the heart ^{12, 32}, primarily due to its association with cytoskeletal elements, where the chaperone stabilizes myofilament, thereby maintaining cellular integrity ^{33, 34}. Phosphorylation of alpha BC at either Ser59 or Ser45 stimulates its chaperone activity ¹⁶, which in turn plays an important role in mediating cellular protein quality control through modulation of the ubiquitin-proteasome system and autophagy. These mechanisms of protein degradation act as defense mechanisms against unfolded proteins, and are essential for cellular function and survival ³⁵. It has been shown that I/R decreases the proteasome activity but ischemic preconditioning improves the proteasomal activity after I/R, and that increased proteasome activity is involved in the cardioprotective effect of preconditioning ³⁶. Interestingly, Tg-CAPKN mice showed enhanced proteasome activity both at baseline and after I/R, and epoxomicin treatment partially reversed the cardioprotective effect of PKN against I/R injury. Furthermore, decreases in the proteasome activity during I/R were significantly enhanced in the Tg-DNPKN heart. Thus, these results suggest that the increased proteasome activity, in part, mediates the cardioprotective effect of PKN.

In conclusion, PKN is activated by I/R and activation of PKN plays a cell protective role in the heart in vivo. Furthermore, PKN mediates phosphorylation of alpha BC and stimulation of ubiquitin-proteasome activity, which in part mediates the protective effect of PKN in the heart (Online Fig. XIII). Thus, stimulation of PKN may represent a novel strategy for protecting the heart from I/R injury.

Supplementary Material