Inhibition of cAMP-Dependent PKA Activates β2-Adrenergic Receptor Stimulation of Cytosolic Phospholipase A2 via Raf-1/MEK/ERK and IP3-Dependent Ca2+ Signaling in Atrial Myocytes

We previously reported in atrial myocytes that inhibition of cAMP-dependent protein kinase (PKA) by laminin (LMN)-integrin signaling activates β2-adrenergic receptor (β2-AR) stimulation of cytosolic phospholipase A2 (cPLA2). The present study sought to determine the signaling mechanisms by which inhibition of PKA activates β2-AR stimulation of cPLA2. We therefore determined the effects of zinterol (0.1 μM; zint-β2-AR) to stimulate ICa,L in atrial myocytes in the absence (+PKA) and presence (-PKA) of the PKA inhibitor (1 μM) KT5720 and compared these results with atrial myocytes attached to laminin (+LMN). Inhibition of Raf-1 (10 μM GW5074), phospholipase C (PLC; 0.5 μM edelfosine), PKC (4 μM chelerythrine) or IP3 receptor (IP3R) signaling (2 μM 2-APB) significantly inhibited zint-β2-AR stimulation of ICa,L in–PKA but not +PKA myocytes. Western blots showed that zint-β2-AR stimulation increased ERK1/2 phosphorylation in–PKA compared to +PKA myocytes. Adenoviral (Adv) expression of dominant negative (dn) -PKCα, dn-Raf-1 or an IP3 affinity trap, each inhibited zint-β2-AR stimulation of ICa,L in + LMN myocytes compared to control +LMN myocytes infected with Adv-βgal. In +LMN myocytes, zint-β2-AR stimulation of ICa,L was enhanced by adenoviral overexpression of wild-type cPLA2 and inhibited by double dn-cPLA2S505A/S515A mutant compared to control +LMN myocytes infected with Adv-βgal. In–PKA myocytes depletion of intracellular Ca2+ stores by 5 μM thapsigargin failed to inhibit zint-β2-AR stimulation of ICa,L via cPLA2. However, disruption of caveolae formation by 10 mM methyl-β-cyclodextrin inhibited zint-β2-AR stimulation of ICa,L in–PKA myocytes significantly more than in +PKA myocytes. We conclude that inhibition of PKA removes inhibition of Raf-1 and thereby allows β2-AR stimulation to act via PKCα/Raf-1/MEK/ERK1/2 and IP3-mediated Ca2+ signaling to stimulate cPLA2 signaling within caveolae. These findings may be relevant to the remodeling of β-AR signaling in failing and/or aging heart, both of which exhibit decreases in adenylate cyclase activity.


Introduction
We previously reported that attachment of atrial myocytes to the extracellular matrix protein laminin (LMN) acts via β 1 integrin receptors to decrease β 1 -adrenergic receptor (AR) and increase β 2 -AR stimulation of L-type Ca 2+ current (I Ca,L ) [1]. Cell attachment to LMN decreases β 1 -AR signaling by inhibiting adenylate cyclase activity and diminishing cAMP levels via integrin-dependent activation of focal adhesion kinase (FAK)/phosphatidyinositol-3' kinase (PI-3K)/protein kinase B (Akt) signaling [2]. We also reported that atrial cell attachment to LMN enhances β 2 -AR signaling by activating Gi/ERK/cytosolic phospholipase A 2 (cPLA 2 )/arachidonic acid (AA) stimulation of I Ca,L [3]. β 2 -AR activation of cPLA 2 signaling is dependent on concomitant LMN-mediated inhibition of adenylate cyclase/cAMP-dependent kinase (PKA) [3]. In other words, cell attachment to LMN acts via inhibition of adenylate cyclase/PKA to both inhibit β 1 -AR signaling and enhance β 2 -AR signaling through activation of cPLA 2 . In embryonic chick ventricular myocytes [4] and rat ventricular myocytes [5] β 2 -AR stimulation also activates cPLA 2 /AA signaling. Moreover, these authors proposed that activation of β 2 -AR/cPLA 2 signaling may compensate for depressed cAMP signaling [4]. Interestingly, in both of these studies by Pavoine et al. (1999) and Ait-Mamar et al., (2005) cardiomyocytes were cultured on LMN, supporting our findings that cell attachment to LMN may be responsible for inhibition of PKA and activation of β 2 -AR/cPLA 2 signaling. However, the mechanism by which PKA inhibition activates β 2 -AR/cPLA 2 signaling is not clear.

Ethics Statement
The animal and experimental protocols used in this study were approved by the Institutional Animal Care and Use Committee (IACUC) of Loyola University Medical Center, Maywood, IL. IACUC prescribed the rules for the animal care and supervised their enforcement. Animals were obtained from a licensed vendor (R & R Research, Howard City, MI., USA), and housed and fed in our AAALAC approved Comparative Medicine Department. Adult cats of either sex (n = 32 cats) were anesthetized with sodium pentobarbital (50 mg/kg, IP).

Isolation of atrial myocytes
Once fully anesthetized, a bilateral thoracotomy was performed, and the heart was rapidly excised and mounted on a Langendorff perfusion apparatus. After enzyme (collagenase; type II, Worthington Biochemical) digestion, atrial myocytes were isolated as previously reported [11].

Perforated patch clamp experiments
Electrophysiological recordings from atrial myocytes were performed in the perforated (nystatin) patch whole-cell configuration at room temperature, as previously described [11]. L-type Ca 2+ current (I Ca,L ) was activated by depolarizing pulses from a holding potential of -40 mV to 0 mV for 200 ms every 5 s and measured in relation to steady-state current. β 2 -AR stimulation was achieved by 0.1 μM zinterol (zint-β 2 -AR), a specific β 2 -AR agonist [12]. Agonist was applied for approximately 4 min and the effects on peak I Ca,L amplitude were recorded at the steady-state response.

Plating of atrial myocytes on LMN coated glass coverslips
Generally, we compared freshly isolated atrial myocytes obtained from the same hearts, as previously described [3]; atrial myocytes on uncoated glass cover-slips in the absence of PKA inhibitor (+PKA) and atrial myocytes on uncoated glass coverslips exposed to the specific PKA inhibitor (1 μM) KT5720 (-PKA). Because pharmacological inhibition of PKA elicits signaling mechanisms that are similar to those elicited by LMN-integrin signaling, we performed some experiments on atrial myocytes attached to glass cover-slips coated with laminin (+LMN; 40 ug/ml) for at least 2 hrs, as previously described [11]. Inhibition of PKA by KT5720 typically decreases basal I Ca,L amplitude by 15-20% [3], consistent with the relatively high endogenous PKA activity in cat atrial myocytes [11]. In addition, a variety of experimental results indicate that atrial cell attachment to LMN is not restoring LMN-mediated signaling somehow lost during the cell isolation procedure. For example, control experiments have shown that atrial myocytes plated on poly-L-lysine, a non-specific substrate for cell attachment, fail to exhibit changes in β-AR signaling similar to cells attached to LMN [1]. Moreover, freshly isolated cardiomyocytes not plated on LMN exhibit responses to β-AR stimulation which are similar to multicellular cardiac preparations, i.e. exhibit predominantly β 1 -AR over β 2 -AR signaling [1]. However, cell attachment to LMN decreases the β 1 -/β 2 -AR signaling ratio resulting in predominantly β 2 -AR over β 1 -AR signaling [1]. Moreover, pharmacological inhibition of PKA in myocytes not attached to LMN mimics the effects of cell attachment to LMN [3].

Adenoviral infection of atrial myocytes
In some experiments, atrial myocytes were attached to laminin (2h) and then infected (100 moi, 24h) with replication-defective adenovirus (Adv) prior to electrophysiological recording. PKCα was inhibited by infection with an Adv expressing kinase-inactive mouse PKCα [13], kindly provided by Dr. Trevor Biden, Garvan Institute of Medical Research, St. Vincent's Hospital, Sydney Australia. An Adv expressing a dominant-negative (dn) mutant of rabbit PKCε [14] was kindly provided by Dr. Peipei Ping, University of California-Los Angeles. Adv expressing wild type, and a non-phosphorylatable mutant of human cPLA 2 (S515A/S505A double mutant) [15] were generously provided by Dr. K.U. Malik, University of Tennessee Health Science Center, Memphis, TN. An Adv expressing a dn mutant of human Raf-1 [9] was kindly provided by Dr. Dan Kuppuswamy, Medical University of South Carolina, Charleston, SC. A control Adv expressing nuclear-encoded β-galactosidase (Adv-βgal) was used to control for nonspecific effects of adenoviral infection [16]. Adenoviruses were amplified and purified using HEK293 cells, and the multiplicity of infection (moi) for each virus was determined by dilution assay in HEK293 cells grown in 96 well clusters, as previously described [16]. Preliminary experiments using 5-bromo-4-chloro-3-indolyl-beta-D-galactopyranoside (X-gal) staining of Adv-βgal infected cells determined that a concentration of 100 moi infected 93±3% (n = 3 expts, 400-700 cells/expt) of cultured myocytes. The IP 3 affinity trap consists of the ligand binding domain of the rat type 1 IP 3 R. The construction of this vector and subsequent production of adenovirus was previously described in detail [17]. Freshly isolated atrial myocytes were infected with Adv-IP 3 affinity trap or the Adv-βgal (control) for 1 h, followed by 18 h short-term culture at 37˚C.

Western blot
Atrial myocytes were plated on poly-L-lysine, a biologically inactive substrate and therefore represent +PKA myocytes. Atrial myocytes were either untreated or treated with Zinterol, KT5720 alone (PKA inhibitor) and KT combined with Zinterol. A positive control showing ERK1/2 and p38MAPK phosphorylation in A7r5 cells is used after stimulation with 1 μM angiotensin II (5 min). Briefly atrial homogenates were centrifuged at 10,000 g for 2 min and the supernatant was collected. The protein concentrations of the samples were determined using the Bradford protein assay (Bio-Rad). Aliquots of the samples (40 μg) were dissolved in a Laemmli Sample buffer containing: 60 mM Tris-HCl, 2% SDS, 20% Glycerol, 5% β-mercaptoethanol, 0.01% bromophenol blue (pH 6.8) and the proteins were separated on a 4-15% Mini-PROTEAN TGX Gel (Bio-Rad). After transfer, the membrane was blocked with TBS-T Buffer containing 20 mM Tris pH 7.5, 150 mM NaCl, 0.05% Tween, and a 5% blocking powder (Bio Rad) at 4˚C for 1 hr. The blot was probed with primary and secondary antibodies for phorpho-ERK1/2 and phosphor-p38 MAPK that were purchased from cell signaling technologies (Danvers, MA). The blot was subsequently probed with GAPDH primary antibody for 2 hrs, followed by HRP-conjugated goat anti-mouse secondary antibody (1:10,000, Santa Cruz Biotechnology) at 4˚C for 1 hr. Specific binding was visualized by chemiluminescence (Immun-Star Western C Kit, Bio-Rad) using a ChemiDoc XRS imager (Bio-Rad Life Science Research, Hercules, CA). The intensities of the bands corresponding to each protein were quantified using ImageLab software (Bio-Rad). The relative intensity for each band was normalized to the intensity of the GAPDH staining.

Statistics
Data are mean ± standard error (SE) of the mean. Measurements were analyzed using either paired or unpaired Student's t test for significance at P<0.05. Multiple comparisons were performed by ANOVA followed by a Student-Newman-Keuls test with significance at P<0.05.

Results
Role of Raf-1 in +PKA and -PKA atrial myocytes

Comparison of zint-β 2 -AR induced I Ca,L currents in the Adv-βgal and dn-Raf-1 mutant infected atrial myocytes
To further establish the role of Raf-1, we infected atrial myocytes with an Adv that expresses a dn-Raf-1 mutant (generously provided by Dr. Kuppuswamy [9,18]. Control cells were infected with Adv-βgal. Infected cells were cultured on LMN overnight and therefore represent +LMN myocytes. As shown in Fig 2A, in control +LMN myocytes expressing βgal, zint-β 2 -AR stimulation elicited a typically enhanced increase in I Ca,L (202±19%). In +LMN myocytes expressing the dn-Raf-1 mutant, zint-β 2 -AR stimulation of I Ca,L was significantly inhibited (105±14%; P<0.02) compared to control. Together, these findings indicate that Raf-1 signaling plays no role in β 2 -AR stimulation of I Ca,L in freshly isolated atrial myocytes not attached to LMN (Fig 1A). However, when PKA is inhibited by either PKA inhibitor or cell attachment to LMN, β 2 -AR stimulation acts via Raf-1 to activate cPLA 2 and stimulate I Ca,L . . This raises the question of how β 2 -AR stimulation activates Raf-1. In a variety of cell systems [10], including cardiac muscle [9], protein kinase C (PKC) activates Raf-1/MEK/ERK1/2. Receptor-mediated activation of phospholipase C (PLC) and the hydrolysis of phospholipids results in the production of diacylglycerol and stimulation of PKC isoenzymes. Therefore, as shown in  [3]. In contrast to +PKA myocytes, edelfosine now significantly inhibited zint-β 2 -AR stimulation of I Ca,L (78±11%; P<0.0005). Additional experiments in +LMN myocytes showed results similar to those found in-PKA myocytes, i.e. zint, 204±2% vs edelfosine, 84±8%; P<0.03 (see S2 Fig for data). In other words, inhibition of PKA allows β 2 -AR stimulation to act via PLC to stimulate cPLA 2 signaling.
Role of cPLA2 in the enhanced zint-β 2 -AR induced I Ca,L currents in LMN plated atrial myocytes Activation of cPLA 2 requires phosphorylation at serine sites S505 and S515 [21][22][23]. We therefore determined the effect of β 2 -AR stimulation in +LMN myocytes infected with Adv that either expressed a non-phosphorylatable dominant-negative cPLA 2 (dn-cPLA 2 S505A/S515A ) mutant, wild-type cPLA 2 (wt-cPLA 2 ) or βgal as control. Infected atrial myocytes were cultured on LMN overnight and therefore represent +LMN myocytes. . These results confirm that cell attachment to LMN activates β 2 -AR/cPLA 2 signaling and further indicates that β 2 -AR stimulation of cPLA 2 requires serine phosphorylation at one or both sites. Role of IP3 receptor in the enhanced zint-β 2 -AR induced I Ca,L currents in -PKA atrial myocytes Our previous work showed that β 2 -AR stimulation of cPLA 2 signaling is dependent on intracellular Ca 2+ [3]. The fact that β 2 -AR stimulation acts via PLC to activate cPLA 2 (Fig 3) suggests the potential involvement of IP 3 -mediated Ca 2+ signaling. We therefore determined the effects of 2 μM 2-APB, a putative IP 3 receptor (IP 3 R) blocking agent [24], on β 2 -AR stimulation of I Ca,L in +PKA and-PKA myocytes. As shown in Fig 7A, in +PKA myocytes 2-APB had   S3 Fig online data).
In another approach, we infected atrial myocytes with an adenovirus that expresses an IP 3 affinity trap which binds to IP 3 in the cytosol and thereby inhibits IP 3 -mediated Ca 2+ signaling by preventing IP 3 from reaching and activating the IP 3 receptor (IP 3 R) [17]. Cells were cultured overnight on LMN and therefore represent +LMN myocytes. As shown in Fig 8, in control +LMN myocytes expressing βgal, zint-β 2 -AR stimulation elicited a typical increase in I Ca,L (182±4%). However, in +LMN myocytes expressing the IP 3 trap zint-β 2 -AR stimulation of I Ca,L (111±17%) was significantly inhibited compared to controls (P<0.002). Together, these findings suggest that IP 3 R signaling plays no role in zint-β 2 -AR stimulation of I Ca,L in freshly isolated atrial myocytes. However, when PKA is inhibited by either PKA inhibitor (Fig 7) or cell attachment to LMN (Fig 8), β 2 -AR stimulation of cPLA 2 is dependent on IP 3 -mediated Ca 2+ signaling.
Role of IP3 receptors located on SR or nuclear envelope in the enhanced zint-β 2 -AR induced I Ca,L currents in -PKA atrial myocytes IP 3 Rs are thought to be located primarily on the sarcoplasmic reticulum (SR) and nuclear envelope membranes. Because the SR and nuclear envelope membranes are highly interconnected, inhibition of SR Ca 2+ uptake by thapsigargin depletes intracellular Ca 2+ stores from both sites [25]. Therefore, in Fig 9 we determined whether depletion of SR and nuclear Ca 2+ by 5 μM thapsigargin (10 min; Thaps) inhibits β 2 -AR stimulation of I Ca,L in +PKA (A) and-PKA (B) myocytes. In +PKA myocytes (A) thapsigargin had no effect on basal I Ca,L amplitude (open bars). Compared to control responses (107±8%), thapsigargin slightly enhanced zint-β 2 -AR stimulation of I Ca,L (143±4%), although the change was not statistically significant. This modest increase in β 2 -AR stimulation of I Ca,L is consistent with the effects of thapsigargin to inhibit SR Ca 2+ release and thereby inhibit Ca 2+ -mediated inactivation of I Ca,L . In fact, separate experiments showed that thapsigargin abolished SR Ca 2+ transients (data not shown). In-PKA myocytes (B), zint-β 2 -AR stimulation of I Ca,L (207±17%) was typically enhanced compared to control responses (107±8%) obtained in +PKA myocytes (A). Interestingly, in-PKA myocytes treated with thapsigargin β 2 -AR stimulation of I Ca,L (238±12%) was still enhanced. In other words, depletion of Ca 2+ from SR and nuclear membranes failed to prevent PKA inhibition from enhancing β 2 -AR signaling. Moreover, the last column in Fig 9B shows that in-PKA myocytes treated with thapsigargin, AACOCF 3 (cPLA2 inhibitor) significantly inhibited zintβ 2 -AR stimulation of I Ca,L (62±12%), indicating that the enhanced response to zint-β 2 -AR stimulation was in fact due to activation of cPLA 2 signaling and that depletion of SR and nuclear Ca 2+ stores failed to prevent β 2 -AR/cPLA 2 signaling. Similar results were obtained when Ca 2+ stores were depleted by treatment (10 min) with 10 μM ryanodine (data not shown). These results indicate that IP 3 -dependent Ca 2+ signaling is not mediated via IP 3 Rs located on SR or nuclear envelope membranes.
Role of IP3 receptors located in caveolae in the enhanced zint-β 2 -AR induced I Ca,L currents in -PKA atrial myocytes Alternatively, IP 3 R protein is present in caveolae [26,27], which are abundant in atrial myocytes [28]. We therefore determined the effects of 10 mM methyl-β-cyclodextrin (MCD; 30 min), an agent that disrupts caveolae formation [29], on zint-β 2 -AR stimulation of I Ca,L in +PKA and-PKA myocytes. As shown in Fig 10A, in +PKA myocytes MCD caused a modest but significant inhibition of zint-β 2 -AR stimulation of I Ca,L (control,122±5% vs MCD, 93±7%; P<0.05), representing a 24% decrease. This is consistent with the idea that β 2 -ARs are normally localized to caveolae [30]. However, in-PKA myocytes (B) MCD elicited a significantly larger inhibition of zint-β 2 -AR stimulation of I Ca,L (control, 235±7% vs MCD, 54±12%; P<0.008), representing a 77% decrease. The fact that MCD elicited a significantly larger inhibition of β 2 -AR signaling in-PKA compared to +PKA myocytes supports the idea that the IP 3 Rs that are essential for β 2 -AR/ cPLA 2 signaling are localized to the caveolae.
Our previous findings showed that strong chelation of intracellular Ca 2+ by BAPTA prevented β 2 -AR stimulation via cPLA 2 . This is consistent with the fact that several of the signaling molecules involved in the proposed β 2 -AR/cPLA 2 signaling cascade are Ca 2+ -dependent, including PLC, PKCα and cPLA 2 . In the present study, we investigated more specifically which source of intracellular Ca 2+ is required for β 2 -AR/cPLA 2 signaling. We found that 2-APB, an agent that inhibits IP 3 -mediated Ca 2+ release in cat atrial myocytes [24] significantly inhibited β 2 -AR stimulation of I Ca,L in both-PKA and +LMN myocytes but not in +PKA myocytes. Moreover, adenoviral expression of an IP 3 affinity trap which inhibits IP 3 -mediated Ca 2+ signaling [17] resulted in a similar inhibition of β 2 -AR signaling in +LMN myocytes. The fact that 2-APB had no effect on β 2 -AR signaling in control +PKA myocytes and that it exerted effects similar to those of the Adv-IP 3 affinity trap suggests that 2-APB acted specifically to inhibit IP 3mediated Ca 2+ signaling. Together, these results indicate that β 2 -AR stimulation of cPLA 2 is dependent on IP 3 -mediated Ca 2+ signaling. Although IP 3 Rs are typically located on SR and nuclear membranes, the present results showed that thapsigargin, an agent that depletes both SR and nuclear envelope Ca 2+ stores [25] failed to prevent β 2 -AR stimulation of I Ca,L via cPLA 2 . However, disruption of caveolae formation by methyl-β-cyclodextrin elicited a significantly larger inhibition of β 2 -AR stimulation in-PKA myocytes than control +PKA myocytes. We therefore conclude that IP 3 Rs on the SR and nuclear envelope membranes are not involved in the Ca 2+ signaling required for β 2 -AR stimulation of cPLA 2 . Alternatively, β 2 -AR stimulation of I Ca,L via cPLA 2 is dependent on IP 3 -mediated Ca 2+ signaling in caveolae. In endothelial and smooth muscle cells plasmalemmal caveolae contain IP 3 R protein that is speculated to mediate Ca 2+ influx through the caveolar membrane [26,27]. Moreover, in rat ventricular myocytes (cultured on laminin) β 2 -ARs stimulate translocation of cPLA 2 to the low density caveolin-3 enriched membrane fraction suggesting that cPLA 2 translocates to caveolae [5]. In fact, lipid rafts i.e. caveolae contain a wide variety of signaling components including β 2 -ARs, G i , PKC, Raf-1, ERK1/2, IP 3 Rs [36], involved in β 2 -AR/cPLA 2 signaling.

Conclusions
These findings may have important implications with respect to the aging and/or failing heart, both of which exhibit decreases in adenylate cyclase activity. In both animal models [37] and in the human right atrium [38], increasing age is associated with a decrease in β 1 -AR function that results from a decrease in adenylate cyclase activity. Likewise, adenylate cyclase activity is depressed in the failing human heart [39] and canine pacing-induced heart failure [40, 41] and yet β 2 -AR signaling is preserved. Previous studies by Nalli et al suggest that PKA and PKG phosphorylates PLCβ3 in gastric smooth muscle cells and decreases PI hydrolysis [42]. The decrease in PI hydrolysis has been suggested to diminish IP 3 -mediated Ca 2+ release decreasing muscle contraction. Along these lines, the present studies also suggest that feline atrial cardiomyocytes exhibit a similar regulation as that seen in gastric muscle. Our research also suggests that increases in extracellular matrix proteins, i.e. fibrosis, may contribute to decreases in adenylate cyclase/PKA activity in the aging and/or failing heart, which in turn switches β 2 -AR signaling from cAMP/PKA to cPLA 2 /AA. This mechanism may be responsible for the preservation of β 2 -AR signaling while β 1 -AR signaling is depressed. Thus, the diminished PKA activity in cardiac fibrosis may decrease phosphorylation of PLC resulting in an increase in IP hydrolysis, Ca 2+ release and greater contraction. Indeed, our present results indicate that inhibition of PLC significantly inhibited β 2 -AR stimulation of I Ca,L in-PKA or +LMN myocytes but not in +PKA myocytes. In addition, our results suggest that 2-APB, an agent that inhibits IP 3 -mediated Ca 2+ release in cat atrial myocytes [24] significantly inhibited β 2 -AR stimulation of I Ca,L in both-PKA and +LMN myocytes but not in +PKA myocytes. Together, these results indicate that β 2 -AR stimulation of cPLA 2 in both-PKA and +LMN myocytes is dependent on PLC/IP 3 -mediated Ca 2+ signaling. Given that cPLA 2 /AA signaling is a potentially pro- Both basal and stimulated PKA activity inhibits Raf-1 signaling, thereby preventing zint-β 2 -AR stimulation of I Ca,L via cPLA 2 . B; in cells not attached to LMN, inhibition of PKA by KT5720 removes inhibition of Raf-1. C; cell attachment to LMN acts via β 1 integrins and FAK/PI-(3)K/Akt signaling to inhibit adenylate cyclase (AC)/ cAMP-dependent kinase (PKA) activity, thereby removing inhibition of Raf-1. β 2 -AR stimulation acts via G i to stimulate PLC leading to activation of PKCα and IP 3 R-mediated Ca 2+ signaling within caveolae. With PKA inhibited, PKCα stimulates Raf-1/MEK/ERK1/2 signaling. Together, ERK1/2 and IP 3 -mediated Ca 2+ signaling activate cPLA/AA, resulting in stimulation of I Ca,L . inflammatory mediator, the present findings suggest that fibrosis in the aging and/or failing heart may predispose inflammation and atrial dysfunction. In fact, AA is reported to slow atrial conduction and has been implicated in the development of postoperative atrial fibrillation [43]. On the other hand, activation of the Raf/MEK/ERK pathway may be cardioprotective [44]. In fact, inhibition of adenylate cyclase/PKA and subsequent activation of Raf-1/ MEK/ERK1/2 signaling enhances cardiac resistance to oxidative stress, increases cell survival, and extends lifespan [45]. Therefore, inhibition of adenylate cyclase/PKA activity and the remodeling of β 2 -AR signaling to activate Raf/MEK/ERK1/2 may help ameliorate the deterioration of function that occurs in the aging and/or failing atrium. Previous studies have indicated that, beside inhibition of COX, aspirin has COX-independent mechanisms that plays a role in tumor suppression [46,47]. Aspirin reportedly inhibits PI3K/Akt kinase activity in epithelial ovarian cancer cells [48]. Aspirin down regulates the expression of PI3K, Akt and ERK in rodent model of acute pulmonary embolism [49] and in turn inhibit the release of inflammatory cytokines [50]. On the other hand, our studies indicate that laminin acts via FAK/PI-(3)K/Akt signaling to inhibit adenylate cyclase-mediated stimulation of I Ca,L . Although, there is no direct evidence suggesting that aspirin or NSAIDs play a role in cardiomyocyte aging, we hypothesize that aspirin or NSAIDs may influence cPLA2/AA pathway and this requires further investigation.