Functional proteomic analysis of a three-tier PKC -Akt-eNOS signaling module in cardiac protection

Final accepted version H-00756-2004, Zhang et al.

Functional proteomic analysis of a three tier PKCε-Akt-eNOS signaling

module in cardiac protection

Jun Zhang1, Christopher P. Baines2, Nobel C. Zong1, Ernest M. Cardwell1, Guangwu Wang1, Thomas M. Vondriska1, Peipei Ping1.

1Departments of Physiology and Medicine, Division of Cardiology, University of California at Los Angeles, Los Angeles, CA; 2Division of Molecular Cardiovascular

Biology, Children’s Hospital Medical Center, Cincinnati, OH Running Title: Myocardial PKCε-Akt-eNOS signaling modules §To whom correspondence should be addressed: Peipei Ping, PhD Cardiovascular Research Laboratories Departments of Physiology and Medicine Division of Cardiology David Geffen School of Medicine at UCLA Room 1619 MRL Building Los Angeles, CA 90095 Tel: 310.267.5624 Fax: 310.267.5623 E-mail: [email protected]

Articles in PresS. Am J Physiol Heart Circ Physiol (November 4, 2004). doi:10.1152/ajpheart.00756.2004

Copyright © 2004 by the American Physiological Society.


2

ABSTRACT

Cardiac protective signaling networks have been shown to involve protein kinase C

epsilon (PKCε). However, the molecular mechanisms by which PKCε interacts with other

members of these networks to form task-specific modules remain unknown. Among 93

different PKCε-associated proteins identified to date, Akt and endothelial nitric oxide

synthase (eNOS) are of importance due to their independent abilities to promote cell

survival and prevent cell death. However, the simultaneous association of PKCε, Akt and

eNOS has never been examined, and in particular, the formation of a module containing

these three proteins, and the role of such a module in the regulation of nitric oxide (NO)

production and cardiac protection, is unknown. Accordingly, the present study was

undertaken to determine whether these molecules form a signaling module and thereby

play a collective role in cardiac signaling. Using both recombinant proteins in vitro and

PKCε transgenic mouse hearts, we demonstrate that: (i) PKCε, Akt and eNOS interact

and form signaling modules both in vitro and in the mouse heart; activation of either

PKCε or Akt enhances the formation of PKCε-Akt-eNOS signaling modules; (ii) PKCε

directly phosphorylates and enhances activation of Akt in vitro, and PKCε activation

increases both phosphorylation and activation of Akt in PKCε transgenic mouse hearts;

(iii) PKCε directly phosphorylates eNOS in vitro, and this phosphorylation enhances

eNOS activity; activation of PKCε in vivo increased phosphorylation of eNOS at Ser1177,

indicating eNOS activation. This study characterizes, for the first time, the physical, as

well as functional, coupling of PKCε, Akt, and eNOS in the heart, and implicates these

PKCε-Akt-eNOS signaling modules as critical signaling elements during PKCε-induced

cardiac protection.


3

Key words: signaling network, protein-protein interactions, phosphorylation,

cardioprotection, preconditioning, proteomics


4

Introduction

Intracellular signaling networks are composed of protein modules specified to execute

distinct tasks (5,17,35). In the heart, complex cellular phenotypes, such as resistance to

ischemic cell death, involve activation of various signaling molecules. Despite this

information, the manner in which the actions of these individual proteins are integrated

into functional modules has only begun to be understood. Numerous studies have

demonstrated modules composed of two molecules, however, the next level of

hierarchical structure, that is a three tier module, remains scarcely defined. While “three

tier” modules have been described in certain canonical signaling responses (e.g.

MAPKs), there is a paucity of investigations demonstrating simultaneous association of

three molecules as a generalized organizational pattern to recruit proteins with disparate

functions together for shared signaling actions. The present study was designed to

address this limitation of our understanding.

Protein kinase C epsilon (PKCε) has been well documented to play an important

role in the genesis of cardioprotection (11, 22). In particular, previous studies from our

laboratory have shown that activation of PKCε in the heart is sufficient to significantly

reduce myocardial infarction due to coronary artery occlusion (26,27). Moreover,

Akt/protein kinase B (14,15,32) and endothelial nitric oxide synthase (eNOS; ref. 31)

have both been independently implicated as protective molecules in the setting of

oxidative stress and ischemic injury to the myocardium. As described above, however,

the information gained about these molecules through previous investigations was

acquired either in isolation (e.g. transgenic activation of PKCε is sufficient to reduce

infarct size [26]) or in a binary sense (e.g. regulation of NO production through eNOS by

Akt [3,15,34]). To our knowledge, no previous investigations have examined the

possibility that these three molecules—PKCε, Akt and eNOS—together constitute a

module, assembly of which is a critical mechanism of protective signaling in the heart. In


5

the present study, we took a combined proteomic and biochemical approach to

characterize native protein complexes containing PKCε, Akt and eNOS. We examine

PKCε-Akt-eNOS signaling modules in vitro and in the mouse heart in terms of molecular

architecture (i.e. protein-protein interactions) and signal transduction (i.e.

posttranslational modification and alteration of enzymatic activity). Our findings indicate

formation of a three tier module in the heart and suggest that such signaling units, like

that formed by PKCε, Akt and eNOS, may represent a mechanism utilized to manipulate

multipurpose signaling proteins to carry out distinct tasks in the myocyte.


6

Materials and Methods

All procedures were performed in accordance with the Animal Research Committee

guidelines at UCLA and the Guide for the Care and Use of Laboratory Animals,

published by the National Institutes of Health.

Materials—Recombinant active Akt1, unactive Akt1, pleckstrin homology domain (PH)

domain of Akt1, PH domain deleted Akt1 fusion proteins corresponding to human

Akt1/PKB alpha were purchased from Upstate Biotechnology (Lake Placid, NY).

Recombinant human PKCε and bovine eNOS were purchased from Biomol (Plymouth

Meeting, PA) and Calbiochem (San Diego, CA), respectively. Anti-Akt1 polyclonal

antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-PKCε

and -eNOS monoclonal antibodies were purchased from BD Pharmingen (San Diego,

CA). Anti-phospho-Akt (Thr308 and Ser473), anti-phospho-eNOS (Ser1177), and anti-

phospho-glycogen synthase kinase (GSK)-3β (Ser21/9) monoclonal antibodies and Akt

kinase assay kit were from Cell Signaling Technology Inc. (Beverly, MA). All other

chemicals were purchased from Sigma-Aldrich (St Louis, MO).

PKCε Transgenic Mice—The transgenic mice with cardiac-specific activation of PKCε

used in this study exhibit ~6.2-fold overexpression of PKCε and have been previously

described (26). Transgenic mice and their non-transgenic littermates were used at 9-12

weeks of age.

Immunoprecipitation—Immunoprecipitation was performed as previously described

(2,12,26-28,33,36). Briefly, covalently cross-linked PKCε monoclonal antibodies were

incubated with protein samples overnight at 4°C. Immunocomplexes were then washed

three times with buffer containing 150 mM NaCl, 20 mM Tris-HCl (pH 7.4), 10 mM

EDTA, 1 % (v/v) Nonidet P40 (NP-40), 1 mM Na3VO4, and a protease inhibitor cocktail


7

(Roche, Indianapolis, IN). After the final wash, the protein sample was eluted from the

beads by re-suspension in Laemmli buffer, boiled, and then subjected to SDS-PAGE.

Immunoblotting—Standard protocols were applied for immunoblotting (2,26-28,36).

Briefly, following SDS-PAGE separation, proteins were transferred to nitrocellulose

membranes and blotted in either 5% milk or 5% Bovine serum albumin (BSA) for

phospho-specific antibodies) in Tris-buffered saline supplemented with 0.5% Tween 20

(TBS-T; 10mM Tris-HCl, pH 7.5, 100mM NaCl, 0.5% Tween 20). The membranes were

then immunoblotted using the ECL detection system (Amersham, Piscataway, NJ).

Gel filtration chromatography—Gel filtration chromatography was carried out as

described (8) with a few modifications. Two mouse hearts were homogenized in Buffer A

containing: 20mM HEPES, pH 7.9, 1.5mM MgCl2, 150mM NaCl, 0.2mM EDTA, 0.5%

(v/v) NP-40, and a cocktail of protease inhibitors. The samples were then dialyzed

overnight at 4oC in Buffer A without NP-40, and clarified prior to chromatography.

Samples were then loaded onto a pre-calibrated Sephacryl S-400 column (XK 26/70,

26mm diameter, 70cm length, Amersham) using Buffer A without NP-40 as running

buffer. Ten milliliter fractions were collected. Proteins were then subjected to PKCε

immunoprecipitation followed by Akt and eNOS immunoblottings. Thyroglobulin

(669kDa), ferritin (440kDa), catalase (232kDa) and aldolase (158kDa) (Amersham,

Piscataway, NJ) were used as molecular standards.

Affinity Pull-down Assay—Glutathione-S-transferase (GST) affinity pull-down assays

were performed as previously reported (26,28,36). Briefly, functionally viable GST-PKCε

recombinant protein was generated using the baculovirus system (BD Pharmingen) and

incubated with either recombinant Akt or eNOS proteins in binding buffer containing

0.5% (v/v) Triton X-100, 20mM Tris-HCl, pH7.4, 150mM NaCl, 1mM EDTA, 1mM EGTA,


8

and a cocktail of protease inhibitors overnight at 4oC. The beads were washed three

times and the proteins eluted with glutathione elution buffer (BD Pharmingen).

Phosphorylation of Akt or eNOS by PKCε—Recombinant Akt or eNOS was incubated

with recombinant PKCε in reaction buffer containing 0.03mg/ml Lα-phosphatidyl-L-serine

(PS), 2.5µg/ml phorbol 12-myristate 13-acetate (PMA), 3.5mM dithiothreitol, 100µM

ATP, 6.5mM magnesium chloride, 50mM Tris-HCl, pH7.5 and 0.2µCi adenosine-5’-[γ-

32P] ATP (for in vitro phosphorylation assay) at 30oC for 30min (33). The reaction was

terminated by addition of Laemmli buffer and boiling for 5 min. The proteins were then

separated by SDS-PAGE and phosphorylation signal detected by autoradiography.

Akt Activity Assay—Akt kinase activity was measured using a kit from Cell Signaling

Technology Inc. according to the manufacturer’s instructions. Briefly, Akt

immunocomplexes were incubated with 1µg GSK-3β fusion protein in the presence of

ATP in buffer containing 25mM Tris·HCl (pH 7.5), 5mM β-glycerolphosphate, 2mM

dithiothreitol, 0.1mM Na3VO4, 10mM MgCl2 at 30oC for 30 min. Phosphorylation of GSK-

3β is measured by Western blotting using a phospho-GSK-3β (Ser21/9) antibody.

eNOS Activity Assay—eNOS activity was measured as NO2/NO3 production using a

colorimetric assay kit from Calbiochem. Briefly, recombinant eNOS, in the presence and

absence of PKCε, was incubated in buffer containing 1mM NADPH, 10mM arginine,

1mM CaCl2, 3µM tetrahydrobiopterin, 1µM flavin adenine dinucleotide, 1µM flavin

adenine mononucleotide, and 25mM Tris-HCl (pH7.4) for 30 min at 30oC. The nitrate

produced was then converted to nitrite with nitrate reductase treatment. Total nitrite was

measured by using the Griess method and quantified by a Wallace 1420 multilabel

counter (16).


9

Statistical Analysis—Data are reported as mean±SEM. Differences among the

experimental groups were analyzed using one-way ANOVA. If the ANOVA showed an

overall significance, post hoc contrasts were performed with Student t test (27,37). P

value less than 0.05 was considered significant.


10

Results

Formation of PKCε-Akt-eNOS modules in vitro—The formation of this three tier module

was initially examined using recombinant proteins in vitro.

We first determined the ability of these three proteins to directly interact in a

binary fashion. GST-PKCε fusion proteins were incubated with recombinant Akt or

eNOS, GST pull-down performed, and the products separated by SDS-PAGE and

Western blotted for Akt or eNOS. Figure 1 demonstrates direct interaction between

PKCε and Akt in vitro. Two further observations regarding this interaction warrant

remark. First, GST-PKCε preferentially interacted with recombinant active Akt,

(130.0±2.3% of unactive Akt) versus unactive Akt with no treatment (Figure 1A). Second,

using recombinant PH-deleted Akt or Akt PH domain proteins, the interaction between

PKCε and Akt was found to occur via the PH domain on Akt (Figure 1B). Interaction of

Akt and eNOS is well-established and was also observed in this study (data not shown).

Next, we investigated the formation of this module when all three components

were present. Akt and eNOS were incubated in the presence or absence of PKCε

followed by eNOS immunoprecipitation and Akt immunoblotting. The data indicate that

addition of PKCε increases the interaction between eNOS and Akt (Figure 2A). To

determine the role of PKCε activation on its ability to interact with Akt and eNOS, these

respective proteins were incubated with PKCε in the presence or absence of potent PKC

activators phorbol ester (PMA) and phosphatidylserine (PS) followed by

immunoprecipitation of PKCε and blotting for either Akt or eNOS. Figure 2B shows that

activation of PKCε with PMA/PS enhances its interaction with both Akt and eNOS.

Similarly, to examine the effect of Akt’s activation on its interaction of PKCε and eNOS,

these respective proteins were incubated with active or unactive Akt recombinant protein

followed by immunoprecipitation and immunoblotting. Interestingly, Akt activation not


11

only enhanced its interaction with PKCε (Figure 2C, bottom panel), but moreover,

activation of Akt enhanced the interaction between PKCε and eNOS (Figure 2C, top

panel). These data suggest that activation of Akt is a key event to facilitate assembly of

PKCε-Akt-eNOS modules.

Formation of PKCε-Akt-eNOS modules in the mouse heart—Having established

formation of functional PKCε-Akt-eNOS modules in vitro, we next wanted to explore the

assembly of, and signal transduction by, these modules in the heart. Previous studies

have shown that Akt and eNOS are present in cardiac PKCε signaling complexes (28).

Despite this information, nothing was known regarding the nature of interactions

between PKCε and Akt or eNOS (i.e. whether they modulate each other’s activity or

post-translational modification state), the architecture of the signaling complexes formed

by these molecules, and whether the three localize in the same signaling unit in the

heart.

To assess native, that is, intact complexes that have not been disrupted by harsh

detergents or heating and thus maintain their endogenous interactions, mouse hearts

were homogenized and separated via gel filtration chromatography. With this method,

intact myocardial protein complexes are separated on the basis of their physical size.

Individual gel filtration fractions were immunoprecipitated with PKCε antibodies to isolate

from the given elution fraction (representative of a given molecular size) the native

complexes containing PKCε. This last step is essential because while some native

complexes of a given size may contain PKCε, other complexes of identical or similar

sizes most certainly exist that do not contain PKCε but that may co-elute from the

column nonetheless. The isolated PKCε immunocomplexes were then separated by

denaturing SDS-PAGE and immunoblotted for PKCε, Akt and eNOS. Figure 3A shows

the western immunoblot analysis of the chromatographic elution fractions following


12

immunoprecipitation for PKCε in the wild type non-transgenic mouse heart. The data

indicate that Akt and eNOS associate with PKCε in multiple fractions, suggesting that

these three molecules interact with each other in a variety of different-sized multiprotein

complexes. Next, the identical analysis was performed using hearts from PKCε

transgenic mice (Figure 3B), which are inherently resistant to ischemic injury (26). When

compared with the profiles obtained from non-transgenic mouse hearts (Figure 3A), the

expression patterns of PKCε-Akt-eNOS signaling modules in PKCε transgenic mice are

shifted towards a higher molecular weight (i.e. a lower elution fraction; Figure 3B),

indicating that during protection, PKCε-Akt-eNOS signaling modules are assembled

within native complexes of greater molecular size.

Signal transduction through a three tier PKCε-Akt-eNOS module—To investigate signal

transduction by PKCε-Akt-eNOS modules, the effect of these proteins to post-

translationally modify each other and to influence each other’s enzymatic activity was

determined.

Recombinant Akt or eNOS was incubated with recombinant PKCε in the

presence of the PKC activators PMA and PS and P32-γ-ATP. PKCε was found to directly

phosphorylate both Akt and eNOS (Figure 4A). Next, the effect of these PKCε-induced

modifications on the activation status of Akt and eNOS was examined. As seen in Figure

4B, Akt phosphorylation activity directed at the well known substrate GSK-3β was

significantly enhanced in the presence of PKCε, suggesting that the PKCε-dependent

phosphorylation of Akt (Figure 4A) leads to increased Akt kinase activity. Similarly,

addition of PKCε significantly enhanced eNOS activity, as measured by nitrate/nitrite

production, (178.8±11%, P<0.05 vs eNOS alone) when compared with eNOS alone

(Figure 4B). These data suggest that the PKCε-dependent phosphorylation of eNOS

(Figure 4A) triggers the increase in eNOS activity.


13

Akt is known to be activated by phosphorylation of Thr308 within its activation

loop, a modification that stabilizes the active conformation of the molecule.

Subsequently, phosphorylation of Ser473 at the C-terminus is essential for full activation

of Akt. While it is well known that PDK1 is responsible for Thr308 phosphorylation, the

upstream kinase(s) that catalyze the phosphorylation of Ser473 have not been defined. In

this study, we tested whether PKCε could be a putative “candidate PDK2” that targets

Ser473 for phosphorylation and completes the activation process of Akt. Indeed, addition

of PKCε in vitro resulted in a 58.3±2.8% increase in Akt phosphorylation at the Ser473

residue above that seen in the absence of PKCε (using a site specific antibody to

phosphor-Ser473; Figure 5), suggesting that PKCε may be a kinase responsible for

PDK2-activity directed at Akt. This possibility was further supported by the data

described below from PKCε transgenic mice.

Role of PKCε-Akt-eNOS modules in PKCε-induced cardiac protection—To confirm the

functional importance of these post-translational modifications observed in vitro, Akt and

eNOS phosphorylation was also examined in the mouse heart. First, PKCε cardiac

protected transgenic mice hearts were used to test the effect of activation of PKCε on

Akt by examining the two conserved phosphorylation sites of Akt. Using the site-specific

phospho-antibodies of Akt, we found that phosphorylation of Akt on both activation sites

Thr308 (336.1±15.1% vs NTG) and Ser473 (181.3±18.2% vs NTG), was significantly

increased in PKCε transgenic mice when compared with non-transgenic animals (Figure

6A). These alterations occurred in the absence of any change in Akt protein level (Figure

6A). Importantly, in agreement with the foregoing in vitro data, PKCε activation was

concomitant not only with post-translational modification of Akt, but also with enhanced

total Akt kinase activity (Figure 6B, top panel). Moreover, when only PKCε-associated

Akt activity was examined (i.e. by immunoprecipitating PKCε and performing an Akt


14

kinase activity assay; Figure 6B, lower panel), there was an even greater increase in Akt

activity in the protected mice as compared with the increase in total Akt activity (i.e.

PKCε-associated and non-PKCε-associated Akt activity) in the protected mice. These

data strongly support that association of Akt with this module in the mouse heart

significantly potentiates its activity.

It is well documented that active Akt can phosphorylate eNOS at serine 1177

(Ser1177) resulting in activation of eNOS and the production of nitric oxide. Therefore, the

phosphorylation of eNOS in PKCε transgenic hearts was examined to determine

whether functional coupling of PKCε to Akt resulted in downstream activation of eNOS,

as suggested by the in vitro findings. Phosphorylation of eNOS on Ser1177 was drastically

enhanced in the PKCε transgenic mice (407.8±43.7% vs NTG, p<0.05; Figure 7),

indicating that activation of PKCε increased eNOS activity in vivo. Meanwhile, total

eNOS protein expression was also significantly increased in PKCε transgenic mice

(140.9±7.2% vs NTG, p<0.05).


15

Discussion

Many subcellular functions, such as signal transduction, are accomplished by

multiprotein complexes. These complexes are differentially assembled within cells in

response to given stimuli, and they coordinate modules of proteins (two or more

interacting molecules) that are targeted to carry out specific tasks. The functions of these

modules, therefore, are engendered by the properties of the molecules within them and

the interactions among these individual components. Several recent studies have

emphasized the role of modules containing two proteins as effectors of signaling tasks in

the setting of cardiac protection. However, the next level of hierarchical structure, that is

integration of a third molecule into these modules, is poorly understood. On the basis of

findings from other laboratories suggesting the potential for a three tier module

containing PKCε, Akt and eNOS, we designed the present study to functionally

characterize this signaling unit containing two kinases and nitric oxide-generating

enzyme.

The molecular architecture of, and signal transduction by, PKCε-Akt-eNOS

signaling modules was first examined in the in vitro setting. The role of Akt to activate

eNOS has been established (3,15,25,38) and several investigations have suggested that

PKC signaling can directly influence Akt activity (12,20,21,23). However, the role of the ε

isoform of PKC to modulate Akt is completely unknown, and the formation of a module

containing PKCε, Akt and eNOS has never been studied. In the present study, all three

molecules were found to form protein-protein interactions with each other in a pair-wise

fashion, but importantly, the interaction was significantly potentiated when all three

molecules were present (Figure 2). We found that PKCε preferentially interacts with the

PH domain of Akt, congruent with previous studies in non-cardiac cells indicating this

domain of Akt as critical for its activation (4). Moreover, we found that the PH domain of


16

Akt appears to not be necessary for binding of eNOS to Akt in vitro (data not shown),

suggesting that the interactions of PKCε and eNOS with Akt are non-competing, i.e. they

occur through different domains. These data suggest that the module concept, whereby

two or more molecules are recruited into close apposition for signal transduction, is

tenable, because a linear, or “pathway” model may not involve simultaneous interaction

of more than two molecules.

To determine whether these interactions lead to signal transduction, we

examined the ability of PKCε to post-translationally modify Akt and eNOS and to

regulate their activity. The findings display that PKCε phosphorylates Akt and leads to

increased Akt activity (as evidenced by GSK3β phosphorylation; Figure 4). Likewise,

PKCε phosphorylates and activates eNOS leading to increased nitrate/nitrite production.

Thus, PKCε is physically and functionally coupled to Akt and eNOS in vitro, providing the

impetus to examine this module in vivo.

Accordingly, we next examined PKCε-Akt-eNOS modules in a line of PKCε

transgenic mice with cardiac-specific over-expression of an active mutant of PKCε that

display a powerful inherent resistance to ischemia/reperfusion injury (26). We wished to

examine the presence and characteristics of PKCε-Akt-eNOS modules in these PKCε

mice, with the goal to understand what role these modules may play in a cardiac

protective phenotype.

Indeed, our in vitro findings were substantiated by the studies in murine hearts:

activation of PKCε (PKCε transgenic mice) was associated with increased formation of

functional PKCε-Akt-eNOS modules. PKCε mice demonstrated enhanced post-

translational modification of Akt and eNOS, as well as enhanced activation of these

molecules (analogous to the in vitro experiments in which PKCε was activated with

PMA/PS). The phosphorylation and inactivation of GSK3β observed in this study is in

agreement with that documented during cardiac protection by other investigators (34). It


17

should be noted that recombinant PKCε alone could increase the phosphorylation of

GSK3β (Figure 4B). However, this was considerably less than when Akt was also

present, suggesting that the preferred mechanism of activation of GSK3β by PKCε in

this module is indirect, that is, through Akt. Therefore, although we cannot rule out that

the increase in phosphorylation of GSK3β in the PKCε TG mice is at least partly due to a

direct action by PKCε, we believe that this more likely achieved through Akt activation.

Additional investigations will be necessary in the future to elucidate the functional

importance of the direct effects of PKCε on the GSK3β phosphorylation during ischemic

injury and protection.

Using non-denaturing gel filtration liquid chromatography (to separate protein

complexes) followed by immunoprecipitation for PKCε (to isolate PKCε-containing native

complexes) we were able to examine intact protein complexes containing PKCε. To

determine which of these complexes contained Akt and eNOS, we separated the

complexes immunoprecipitated from the gel filtration fractions by SDS-PAGE and

western blotted for Akt and eNOS (Figure 3). The data indicate that PKCε, Akt and

eNOS co-reside in multiple native complexes in the mouse heart and that these

complexes display a large molecular weight range. Furthermore, PKCε-induced cardiac

protection (i.e. PKCε transgenic mice) was associated with increased localization of

these PKCε-Akt-eNOS modules to higher molecular weight complexes. To our

knowledge, these findings are the first to demonstrate formation of native complexes in

the myocardium that are modulated with regard to their molecular components (here,

PKCε-Akt-eNOS modules) concomitant with changes in the phenotype of the organ.

These findings are of salient interest because they suggest that native protein

complexes, such as PKCε complexes, containing modules, such as the PKCε-Akt-eNOS

module described herein, are mechanisms of signal transduction and not solely artifacts

of biochemical analyses. Taken with the striking data described above regarding post-


18

translational modifications and enhanced enzymatic activity engendered by assembly of

this signaling unit, these findings support PKCε-Akt-eNOS modules as critical

components of PKCε protective signal transduction. Since each component within the

complex contributes to the complex formation, it is very likely that expression changes in

either Akt and/or eNOS may also influence assembly of this module. Consequently,

future studies will be required to investigate Akt- and eNOS-induced changes in PKCε-

Akt-eNOS module formation, potentially using transgenic mouse models with cardiac-

specific overexpression of either Akt or eNOS (6, 9).

In contrast to our findings, other investigators have reported that in A549 and

HEK293 cells, the phosphatidylinositol 3-kinase/Akt signaling pathway was regulated by

PKC in a negative manner (10,39). PMA-induced apoptosis was accompanied by an

inhibition of Akt activity (29). Similarly, PKC activation inhibited eNOS activity by

attenuating eNOS phosphorylation on Ser1177 and increasing phosphorylation of Thr495 in

cardiovascular endothelial cells (24,25). The present findings suggest that cardiac PKCε

serves as a positive regulator of the formation and regulation of PKCε-Akt-eNOS

signaling modules in mouse hearts. Possible reasons for these differences include

isoform-specific functions of PKC and/or distinct cell types. These findings highlight the

versatility of the enzymes to participate in a host of distinct cellular processes, and re-

emphasize the importance of studying modules of proteins (and correlating these

findings with a phenotype) that are regulated in a cell type and stimulus-specific manner.

One caveat with the present studies is the use of the PKCε transgenic mouse

model. This is a chronic genetic model in which the protein has been overexpressed for

most of the animal’s post-natal life. This activation can in turn induce long-standing

genetic and proteomic changes in the heart. Moreover, overexpression of a protein can

lead to its mislocalization to different subcellular compartments. Consequently, caution

needs to be exercised when extrapolating the present data to more acute models of


19

cardioprotection such as ischemic preconditioning or adenosine administration, or

physiological alterations in PKCε signaling. Future studies will examine changes in

module formation under these conditions. A second caveat is that PKCε may not be the

predominant “protective” isoform in other animal models. Studies from our lab have

indicated that PKCε is the major mediator of protection in mice and rabbits (27,28).

However, PKCα appears to be more important in dogs and pigs (19,30), as does PKCδ

in the setting of opioid-induced protection (13). Whether PKCα also forms a similar

signaling module with Akt and eNOS in these species remains to be determined.

Akt has two conserved phosphorylation sites, Thr308, and Ser473, which must be

modified to induce full activation. The upstream kinase that phosphorylates Thr308 is

PDK-1. However, the protein kinase responsible for Ser473 phosphorylation is unknown,

and has been tentatively named “PDK-2” (7). Several candidates have been proposed,

including atypical PKC (40), lipid raft-associated activity (18), or simply the

autophosphorylation processes that are triggered by PDK-1 activity (1). Due to the data

in the present study indicating a link between PKCε and Akt, we examined the ability of

PKCε to phosphorylate Akt at Ser473, thereby behaving as the aforementioned PDK-2.

Interestingly, we found that PKCε can in fact phosphorylate Akt at Ser473 in vitro (Figure

5). Taken with the data demonstrating increased Ser473 phosphorylation of Akt in PKCε

transgenic mice (Figure 6), these findings strongly suggest that PKCε is a candidate for

the in vivo “PDK2-activity” associated with the cardiac protected phenotype displayed by

these PKCε mice. Moreover, these findings have important implications for regulation of

the PKCε-Akt-eNOS module. Specifically, once Akt has been primed by PDK-1,

integration into the PKCε-Akt-eNOS module would theoretically allow for full activation of

Akt on the basis of the PDK-2-activity ascribed herein to PKCε. Further experimentation

will be required to fully establish or refute this concept.


20

In summary, we have demonstrated assembly of PKCε-Akt-eNOS signaling

modules in vitro and in the mouse heart. These modules appear to be involved in

cardiac protection, and provide a mechanistic link between these three proteins

previously associated with such protection. The concept of modules of proteins is not

new, indeed the field of MAPK signaling, for instance, has long recognized the

importance of conserved mechanisms of activation. A departure from previous

investigations that is offered by these investigations is the principle that multipurpose

signaling proteins, themselves not constrained to mandatory activation pathways, can be

manipulated by the cell to form distinct signaling modules. In other words, module

formation may be an organizational tool of the intracellular signaling network to elicit

specialized responses.


21

Acknowledgements

This study was supported in part by the National Institutes of Health Grants HL-69301

and HL-65431 (Peipei Ping), American Heart Association Grants 0120412B (Christopher

P. Baines) and 0110053B (Thomas M. Vondriska) and the Laubisch Endowment at

UCLA.


22

References

1. Adams JA. Kinetic and catalytic mechanisms of protein kinases. Chem Rev.

2001;101:2271-2290.

2. Baines CP, Zhang J, Wang GW, Zheng YT, Xiu JX, Cardwell EM, Bolli R, Ping

P. Mitochondrial PKCε and MAPK form signaling modules in the murine heart:

enhanced mitochondrial PKCε-MAPK interactions and differential MAPK activation

in PKCε-induced cardioprotection. Circ Res. 2002;90:390-397.

3. Bell RM, Yellon DM. Bradykinin limits infarction when administered as an adjunct

to reperfusion in mouse heart: the role of PI3K, Akt and eNOS. J Mol Cell Cardiol.

2003;35:185-193.

4. Bellacosa A, Chan TO, Ahmed NN, Datta K, Malstrom S, Stokoe D,

McCormick F, Feng J, Tsichlis P. Akt activation by growth factors is a multiple-

step process: the role of the PH domain. Oncogene. 1998;17:313-25.

5. Bhalla US, Iyengar R. Functional modules in biological signaling networks.

Novartis Found Symp. 2001;239:4-13.

6. Brunner F, Andrew P, Wolkart G, Zechner R, Mayer B. Myocardial contractile

function and heart rate in mice with myocyte-specific overexpression of endothelial

nitric oxide synthase. Circulation. 2001;104:3097-3102.

7. Chan TO, Tsichlis PN. PDK2: a complex tail in one Akt. Science STKE.

2001;23:PE1.

8. Chen J, Rappsilber J, Chiang YC, Russell P, Mann M, Denis CL. Purification

and characterization of the 1.0 Mda CCR4-NOT complex identifies two novel

components of the complex. J Mol Biol. 2001;314:683-694.

9. Condorelli G, Drusco A, Stassi G, Bellacosa A, Roncarati R, Iaccarino G,

Russo MA, Gu Y, Dalton N, Chung C, Latronico MV, Napoli C, Sadoshima J,


23

Croce CM, Ross J Jr. Akt induces enhanced myocardial contractility and cell size

in vivo in transgenic mice. Proc Natl Acad Sci U S A. 2002;99:12333-12338.

10. Doornbos RP, Theelen M, van der Hoeven PC, van Blitterswijk WJ, Verkleij

AJ, van Bergen en Henegouwen PM. Protein kinase C zeta is a negative

regulator of protein kinase B activity. J Biol Chem. 1999;274:8589-96.

11. Dorn 2nd GW, Souroujon MC, Liron T, Chen C-H, Gray MO, Zhou HZ, Csukai

M, Wu G, Lorenz JN, Mochly-Rosen D. Sustained in vivo cardiac protection by a

rationally designed peptide that causes ε PKC translocation. Proc Natl Acad Sci

USA. 1999;96:12798-12803.

12. Edmondson RD, Vondriska TM, Biederman K, Zhang J, Jones R, Zheng YT,

Allen DL, Xiu JX, Cardwell EM, Pisano MR, Ping P. PKCε complexes contain

metabolism and transcription/translation related proteins: complementary

separation techniques with LC/MS/MS. Mol Cell Proteomics. 2002;6:421-433.

13. Fryer RM, Wang Y, Hsu AK, Gross GJ. Essential activation of PKCδ in opioid-

initiated cardioprotection. Am J Physiol. 2001;280:H1346-1353.

14. Fujio Y, Nguyen T, Wencker D, Kitsis RN, Walsh K. Akt promotes survival of

cardiomyocytes in vitro and protects against ischemia-reperfusion injury in mouse

heart. Circulation. 2001;101:660-667.

15. Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K, Franke TF,

Papapetropoulos A, Sessa WC. Regulation of endothelium-derived nitric oxide

production by the protein kinase Akt. Nature. 1999;399:597-601.

16. Gilliam MB, Sherman MP, Griscavage JM, Ignarro LJ. A spectrophotometric

assay for nitrate using NADPH oxidation by Aspergillus nitrate reductase. Anal

Biochem. 1993;212:359-365.

17. Hartwell LH, Hopfield JJ, Leibler S, Murry AW. From molecular to modular cell

biology. Nature. 1999;402:C47-C52.


24

18. Hill MM, Feng J, Hemmings BA. Identification of a plasma membrane Raft-

associated PKB Ser473 kinase activity that is distinct from LK and PDK1. Curr Biol.

2002;12:1251-1255.

19. Kitakaze M, Funaya H, Minamino T, Node K, Sato H, Ueda Y, Okuyama Y,

Kuzuya T, Hori M, Yoshida K. Role of protein kinase C-α in activation of ecto-5’-

nucleotidase in the preconditioned canine myocardium. Biochem Biophys Res

Commun. 1997;239:171-175.

20. Konoshi H, Kuroda S, Tanaka M, Matsuzaki H, Ono Y, Kameyama K, Haga T,

Kikkawa U. Molecular cloning and characterization of a new member of the RAC

protein kinase family: association of the pleckstrin homology domain of three types

of RAC protein kinase with protein kinase C subspecies and beta gamma subunits

of G proteins. Biochem Biophys Res Commun. 1995;216:526-534

21. Li W, Zhang J, Flechner L, Hyun T, Yam A, Franke TF, Pierce JH. Protein

kinase C alpha overexpression stimulates Akt activity and suppresses apoptosis

induced by interleukin 3 withdrawal. Oncogene. 1999;18:6564-72.

22. Liu GS, Cohen MV, Mochly-Rosen D, Downey JM. PKCε is responsible for the

protection of preconditioning in rabbit cardiomyocytes. J Mol Cell Cardiol.

1999;31:1937-1948.

23. Mao M, Fang X, Lu Y, Lapushin R, Bast RC Jr, Mills GB. Inhibition of growth-

factor-induced phosphorylation and activation of protein kinase B/Akt by atypical

protein kinase C in breast cancer cells. Biochem J. 2000;352:475-8

24. Michell BJ, Chen ZP, Tiganis T, Stapleton D, Katsis F, Power DA, Sim AT,

Kemp BE. Coordinated control of endothelial nitric oxide synthase phosphorylation

by protein kinase C and the cAMP-dependent protein kinase. J Biol Chem.

2001;276:17625-17628.


25

25. Michell BJ, Griffiths JE, Mitchelhill KI, Rodriguez-Crespo I, Tiganis T,

Bozinovski S, de Montellano PR, Kemp BE, Pearson RB. The Akt kinase

signals directly to endothelial nitric oxide synthase. Curr. Biol. 1999;9:845-848.

26. Ping P, Song CX, Zhang J, Guo Y, Cao X, Li RC, Wu W, Vondriska TM, Pass

JM, Tang XL, Pierce WM, Bolli R. Formation of PKCε-Lck modules confers

cardioprotection. J Clin Invest. 2002;109:499-507.

27. Ping P, Takano H, Zhang J, Tang XL, Qiu Y, Li RC-X, Banerjee S, Dawn B,

Balafanova Z, Bolli R. Isoform-selective activation of PKC by nitric oxide in the

heart of conscious rabbits: a signaling mechanism for both nitric oxide-induced and

ischemia-induced preconditioning. Circ Res. 1999;84:587-604.

28. Ping P, Zhang J, Pierce WM Jr, Bolli R. Functional proteomics analysis of PKCε

signaling complexes in the normal heart and during cardioprotection. Circ Res.

2001;88:59-62.

29. Ragolia L, Palaia T, Paric E, Maesaka JK. Elevated L-PGDS activity contribute to

PMA-induced apoptosis concomitant with downregulation of PI3-K, Am J Physiol.

2003;284:C119-26.

30. Schulz R, Gres P, Skyschally A, Duschin A, Belosjorow S, Konietzka I,

Heusch G. Ischemic preconditioning preserves connexin 43 phosphorylation

during sustained ischemia in pig hearts in vivo. FASEB J. 2003,17:1355-7.

31. Shi Y, Baker JE, Zhang C, Tweddell JS, Su J, Pritchard KA Jr. Chronic hypoxia

increases endothelial nitric oxide synthase generation of nitric oxide by increasing

heat shock protein 90 association and serine phosphorylation. Circ Res.

2002;91:300-306.

32. Shiraishi I, Melendez J, Ahn Y, Skavdahl M, Murphy E, Welch S, Schaefer E,

Walsh K, Rosenzweig A, Torella D, Nurzynska D, Kajstura J, Leri A, Anversa

P, Sussman MA. Nuclear targeting of Akt enhances kinase activity and survival of

cardiomyocytes. Circ. Res. 2004;94:884-891.


26

33. Song CX, Vondriska TM, Wang GW, Klein JB, Cao X, Zhang J, Kang YJ,

D’Souza S, Ping P. Molecular conformation dictates module formation: example of

PKCε and Src. Am J Physiol. 2002;282:H1166-H1171.

34. Tong H, Imahashi K, Steenbergen C, Murphy E. Phosphorylation of glycogen

synthase kinase-3β during preconditioning through a phosphatidylinositol-3-kinase-

dependent pathway is cardioprotective. Circ Res. 2002;90:377-379.

35. Vondriska TM, Klein JB, Ping P. Functional proteomics to investigate PKCε-

mediated cardioprotection: the signaling module hypothesis. Am J Physiol.

2001;280:H1434-1441.

36. Vondriska TM, Zhang J, Song C, Tang XL, Cao X, Baines CP, Pass JM, Wang

S, Bolli R, Ping P. PKCε-Src modules direct signal transduction in nitric oxide-

induced cardioprotection: complex formation as a means for signal transduction.

Circ Res. 2001;88:1306-1313.

37. Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in

circulation research. Circ Res. 1980;47:1-9.

38. Webster KA. Aktion in the nucleus. Circ Res. 2004;94:865.

39. Wen HC, Huang WC, Ali A, Woodgett JR, Lin WW. Negative regulation of

phosphatidylinositol 3-kinase and Akt signaling by PKC. Cell Signal. 2003;15:37-

45.

40. Ziegler WH, Parekh DB, Le Good JA, Whelan RD, Kelly JJ, Frech M,

Hemmings BA, Parker PJ. Rapamycin-sensitive phosphorylation of PKC on a

carboxy-terminal site by an atypical PKC complex. Curr Biol. 1999;9:522-529.


27

Figure Legends

Figure 1. PKCε directly interacts with Akt and eNOS in vitro GST-PKCε was

incubated with recombinant Akt or eNOS, followed by SDS-PAGE and immunoblotting

for Akt or eNOS. A, GST-PKCε favors interaction with active (lanes 1 and 2), as

compared to unactive (lanes 3 and 4), Akt. GST-null (GST without insert proteins)

exhibited minimal interactions with active or unactive Akt (lanes 6 and 7 respectively). B,

GST-PKCε favors interaction with the Akt PH domain (lanes 3 and 4) as compared to PH

domain-deleted Akt (Akt ∆PH; lanes 1 and 2). Lanes 6 and 7 are positive controls (direct

loading of recombinant proteins) for Akt ∆PH and Akt PH domain, respectively. C, GST-

PKCε directly interacts with eNOS (lanes 1 and 2). GST-null exhibited minimal

interactions with eNOS (lane 4). All results are representative of three or more

independent experiments. “+” indicates presence of a protein and “-“ indicates absence.

Figure 2. Activation of PKCε or Akt enhances PKCε-Akt-eNOS module formation in

vitro. A, Left panel, Recombinant Akt and eNOS were incubated in the absence (lane 1)

or presence (lane 2) of PKCε followed by immunoprecipitation (IP) for eNOS and

immunoblotting (IB) for PKCε (top blot) or Akt (bottom blot). PKCε is sufficient to

increase interactions between eNOS and Akt. B, recombinant PKCε was incubated with

Akt and eNOS in the absence (lane 1) or presence (lane 2) of the PKC activators

phorbol 12-myristate 13-acetate (PMA) and phosphatidyl-L-serine (PS) followed by IP for

PKCε and IB for eNOS (top blot) or Akt (bottom blot). Activation of PKCε with PMA/PS

increases its affinity for Akt and eNOS. C, Active (lane 2) or unactive (lane 3) Akt

recombinant proteins were incubated with PKCε and eNOS followed by IP for PKCε and

IB for Akt (top blot) or eNOS (bottom blot). Activation of Akt increases its interaction with


28

PKCε, and in addition, increases the interaction of PKCε with eNOS. Results are

representative of three or more independent experiments and IgG and beads-alone

controls all demonstrated negligible signal (data not shown). “+” indicates presence of a

protein or the lipids and “-” indicates absence.

Figure 3. Formation of PKCε-Akt-eNOS signaling modules in vivo. Non-transgenic

(top panel) or PKCε transgenic (bottom panel) mouse hearts were homogenized and

subjected to gel filtration chromatography using a Sephacryl S400 column followed by

immunoprecipitation (IP) for PKCε and immunoblotting (IB) for PKCε (top blots, both

panels), Akt (center blots, both panels) or eNOS (bottom blots, both panels). Fraction

numbers are indicated on top of the blots: larger numbers are indicative of later elution

fractions and therefore smaller molecular masses. Areas of salient correlation in the

protein content of PKCε, Akt, and eNOS are indicated with the dashed lines. Cardiac

protection is associated with formation of larger complexes containing PKCε-Akt-eNOS

modules, as demonstrated by the leftward shift of IB signal for the three proteins toward

lower elution fractions. Results are representative of three or more independent

experiments.

Figure 4. PKCε phosphorylates, and enhances the activity of, Akt and eNOS. A,

Recombinant Akt (top panel) or eNOS (bottom panel) was incubated with recombinant

PKCε (lane 2) in the presence of PKC activators PMA and PS and P32-γ-ATP. Following

SDS-PAGE separation, phosphorylation was visualized by autoradiography. Lanes 1

and 3 demonstrate minimal background phosphorylation signal. B, After incubation with

PKCε, Akt (blot; lanes 2 and 3) and eNOS (graph) activities were assessed by the ability

of Akt to phosphorylate GSK-3β or eNOS to produce nitrate/nitrite. In the absence of

PKCε, unactive Akt does not phosphorylate GSK-3β (lane 1). Results are representative


29

of three or more independent experiments. “+” indicates presence of a protein and “-“

indicates absence.

Figure 5. PKCε phosphorylate Akt on Ser473 in vitro. PKCε was incubated with

unactive Akt (lanes 3 and 4) and Ser473 phosphorylation detected by site-specific

antibody-based immunoblotting (lanes 1 and 2, unactive Akt negative control). In the

absence of PKCε (lanes 5 and 6) Akt is not phosphorylated at Ser473. Results are

representative of three or more independent experiments. “+” indicates presence of a

protein and “-“ indicates absence.

Figure 6. Activation of PKCε leads to phosphorylation and activation of Akt in

vivo. A, Non-transgenic (NTG; left lanes) and PKCε transgenic (PKCε TG; right lanes)

mouse heart lysates were immunoblotted for Akt, PKCε and site-specific phospho-Akt

(Ser473 and Thr308). Blots (left panel) and graph (right panel) demonstrate that activation

of PKCε in vivo leads to increased total Akt protein and increased phosphorylation at

both residues critical for its activation. No change in total Akt protein level was observed

in the PKCε TG mice (bottom blot, left). B, Total Akt activity (top blot) and PKCε-

associated (determined following PKCε IP; bottom blot) Akt activity were measured by

the ability of Akt to phosphorylate its well-characterized substrate, GSK-3β. PKCε

activation in vivo (PKCε TG mice, right lanes) leads to increased enzymatic activity of

Akt, as compared to NTG control (left lanes). All results are representative of three or

more independent experiments.


30

Figure 7. PKCε activates eNOS in vivo. Non-transgenic (NTG; left lanes) and PKCε

transgenic (PKC TG; right lanes) mouse heart lysates were immunoblotted for Ser1177-

phosphorylated eNOS (top blot) and for total eNOS (bottom blot). PKCε activation

promotes a modest increase in total eNOS protein and a dramatic increase in Ser1177

phosphorylation of eNOS (quantitative data are reported in graph; open bars indicate

NTG, striped bars PKCε TG hearts). Results are representative of three or more

independent experiments.

1 2 3 4 5 6 7A.

GST-PKCεGST-null

Active AktUnactive Akt

+ + + +

+ ++ + +

++++

_ _ _

___

_____

____

Akt

B.

+ + + + _ _ _

+ +

_ _

_ _

_ _ + + +_ _

_ __ + _

1 2 3 4 5 6 7Akt ∆PHAkt PH domain

GST-PKCεGST-nullAkt ∆PH

+ + +Akt PH Domain

C.

GST-PKCεGST-null

eNOS

+

+ +

+

+++__

_ _

_

1 2 3 4

eNOS

Figure 1. 31

A. B.

1 2 1 2

eNOS IP : PKCε IB

_ +

PKCε IP : eNOS IB

+++

_

PKCε IP : Akt IBeNOS IP : Akt IBPKCεeNOS

AktPMA + PS

Figure 2.

PKCε IP : eNOS IB

PKCε IP : Akt IB__

_

+++

++

+ _+

1 2 3

Unactive Akt

eNOSPKCε

Active Akt

PKCε++

eNOSAkt +

+

C.

++

++

+

32

A. NTG Mouse Hearts

Gel Filtration Elution Fraction #

Hela15 16 17 18 19 201 2 3 4 5 6 7 8 9 10 11 12 13 14

PKCεPKCε IP : PKCε IB

AktPKCε IP : Akt IB

eNOSPKCε IP : eNOS IB

High MW Low MW 669kDa 440kDa 232kDa

B. PKCε TG Mouse Hearts Gel Filtration Elution Fraction #

Hela15 16 18 19 201 2 3 4 5 6 7 8 9 10 11 12 13 14 17

PKCε IP : PKCε IB PKCε

AktPKCε IP : Akt IB

eNOSPKCε IP : eNOS IB

669kDa 440kDa 232kDaLow MW High MW

Figure 3. 33

Phosphorylated Akt

+ + __ + +

1 2 3

A.Akt

PKCε

1 2 3

+ + __ + +

Phosphorylated eNOS

eNOSPKCε

+ + + __ + + +

1 2 3 4

Phosphorylated GSK3βB.

PKCεUnactive Akt

** P<0.05 vs eNOS alonen=6/group

200

Figure 4.

eNOS

80

120

160

eNOS Activity(% of eNOS alone)

40

0eNOS + PKCε

34

Phospho-Akt Ser4731 2 3 4 5 6

PKCε

Akt

_

PKCε

Unactive Akt

+

+

+

+

+ +

_ + +_ _

Figure 5. 35

A.NTG PKCε TG

PKCε

0

100

200

300

400

Phospho-Akt Thr308

Phospho-Akt Ser473

Total Akt protein* P<0.05 vs NTG (n=5)

Inte

nsity

(% o

f NTG

)

*

*Phospho-Akt Thr308

Phospho-Akt Ser473

Akt

NTG PKCε TG

B.

Phospho-GSK3β

PKCε TGNTG

Total lysates

Phospho-GSK3βPKCε IP

Figure 6. 36

NTG PKCε TG

Figure 7.

Phospho-eNOS Ser1177

eNOS

*

*500

0

100

200

300

400

Phospho-eNOS Ser1177 Total eNOS protein

PKCε TGNTG

* P<0.05 vs NTG (n=5)

Inte

nsity

(% o

f NTG

)

37

Functional proteomic analysis of a three-tier PKC -Akt-eNOS signaling module in cardiac protection

Documents

Transcript of Functional proteomic analysis of a three-tier PKC -Akt-eNOS signaling module in cardiac protection