Stimulation of “Stress-regulated” Mitogen-activated Protein Kinases (Stress-activated Protein Kinases/c-Jun N-terminal Kinases and p38-Mitogen-activated Protein Kinases) in Perfused Rat Hearts by Oxidative and Other Stresses*

“Stress-regulated” mitogen-activated protein kinases (SR-MAPKs) comprise the stress-activated protein kinases (SAPKs)/c-Jun N-terminal kinases (JNKs) and the p38-MAPKs. In the perfused heart, ischemia/reperfusion activates SR-MAPKs. Although the agent(s) directly responsible is unclear, reactive oxygen species are generated during ischemia/reperfusion. We have assessed the ability of oxidative stress (as exemplified by H 2 O 2 ) to activate SR-MAPKs in the perfused heart and compared it with the effect of ischemia/reperfusion. H 2 O 2 acti- vated both SAPKs/JNKs and p38-MAPK. Maximal activation by H 2 O 2 in both cases was observed at 0.5 m M . Whereas activation of p38-MAPK by H 2 O 2 was comparable to that of ischemia and ischemia/reperfusion, activation of the SAPKs/JNKs was less than that of ischemia/ reperfusion. As with ischemia/reperfusion, there was minimal activation of the ERK MAPK subfamily by H 2 O 2 . MAPK-activated protein kinase 2 (MAPKAPK2), a downstream substrate of p38-MAPKs, was activated by H 2 O 2 to a similar extent as with ischemia or

We and others have recently shown that p38-MAPK and MAPKAPK2 are strongly activated by ischemia in the perfused rat heart (32,33). On reperfusion, activation of these kinases is maintained, and in addition SAPKs/JNKs are activated (32)(33)(34). Consistent with this, SAPKs/JNKs are activated in neonatal cardiac myocytes subjected to hypoxia/reoxygenation but not hypoxia alone (35). The roles of these SR-MAPKs remain obscure. We are interested in signals that may potentially activate the SR-MAPKs in ischemic/reperfusion stress. During ischemia and on reperfusion of the ischemic myocardium, there is release of ROS as well as other factors (reviewed in Refs. 1, 2, and 4). Here, we have examined the potential of oxidative stress (as exemplified by perfusion with H 2 O 2 ) to activate SR-MAPKs and MAPKAPK2 in the isolated rat heart and have compared the effects of H 2 O 2 with those of ischemia and ischemia/reperfusion. Using antioxidants, we investigated the role of ROS in the activation of SR-MAPKs in ischemia/reperfusion. In addition, we have examined the effects of another pathophysiologically important stress (hypertensive stress) on activation of the SR-MAPKs.
Heart Perfusions-Adult male (250 -300 g) Sprague-Dawley rat hearts were perfused retrogradely at a pressure of 10 kilopascals (70 mm Hg) with Krebs-Henseleit bicarbonate-buffered saline (25 mM NaHCO 3 , 119 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl 2 , 1.2 mM KH 2 PO 4 (pH 7.6)) at 37°C supplemented with 10 mM glucose and equilibrated with 95% O 2 /5% CO 2 . The temperature of the perfusates and hearts was maintained at 37°C by the use of a water-jacketed apparatus. Coronary flows were determined at 10 min after cannulation and were also measured during and at the end of the experiments. All other times given refer to times following the appropriate equilibration period.
For experiments in which H 2 O 2 was the only agent added, hearts were perfused for a 15-min equilibration period. H 2 O 2 was added to the requisite concentration, and hearts were perfused for a further 5-30 min. For hearts subjected solely to simple global ischemia or to ischemia/reperfusion, the equilibration period was 15 min. The perfusion was then interrupted for 20 min by clamping the aortic perfusion line. Hearts ceased beating within 1 min of ischemia. Where indicated, ischemic hearts were reperfused for 10 min by reopening the aortic perfusion line. Hearts resumed beating within 1 min of reperfusion, and coronary flow returned to within 80% of control values (control, 13 ml/min/heart; reperfused, 11 ml/min/heart). Control hearts were perfused for up to 30 min after the pre-equilibration period without interruption to the perfusate flow. For hearts subjected to increased aortic pressure, the perfusion pressure was raised to 20 kilopascals (140 mm Hg) after the 15-min equilibration period, and the perfusions were continued for 20 min. When Me 2 SO, SB203580 (10 mM stock in Me 2 SO), or BPN (2.5 M stock in Me 2 SO) was used, the hearts were perfused for 15 min. Me 2 SO, SB203580/Me 2 SO, or BPN/Me 2 SO was then added, and the perfusions were continued for another 15 min before the addition of H 2 O 2 or the imposition of ischemia. At the end of all perfusions, hearts were "freeze-clamped" between aluminum tongs cooled in liquid N 2 and pulverized under liquid N 2 , and the powders were stored at Ϫ80°C.
Fast Protein Liquid Chromatography of ERKs and MAPKAPK2-Supernatants of heart powders homogenized with 3 volumes of buffer A were diluted 4-fold with buffer A and recentrifuged (10,000 ϫ g, 5 min, 4°C). Proteins in samples (0.5 ml) were separated by fast protein liquid chromatography (FPLC). ERKs were separated on a Mono Q HR5/5 column equilibrated with 50 mM Tris/HCl (pH 7.3), 2 mM EDTA, 2 mM EGTA, 0.1% (v/v) 2-mercaptoethanol, 5% (v/v) glycerol, 0.03% (v/v) Brij-35, 0.3 mM Na 3 VO 4 , 1 mM benzamidine, and 4 g/ml leupeptin. Following a 5-ml isocratic wash, ERKs were separated using a linear NaCl gradient (20 ml, 0 -0.33 M NaCl) at flow rate of 1 ml/min with collection of 0.5-ml fractions. They were assayed by the incorporation of 32 P from [␥-32 P]ATP into MBP by the direct method as described previously (43). Samples of fractions were also taken for in gel kinase assays and were boiled with 0.33 volume of SDS sample buffer.

Stimulation of "Stress-regulated" MAPKs in Heart
MAPKAPK2 was assayed by the incorporation of 32 P from [␥-32 P]ATP into KKLNRTLSVA peptide substrate (37) as described previously (32), except that the incubation time was decreased to 10 min. For the determination of total MAPKAPK2 activity, the areas under the FPLC peaks were integrated.

Activation of "Stress-regulated" MAPKs by H 2 O 2 -
In gel kinase assays of hearts perfused for 30 min showed maximal activation of both p46 and p54 SAPKs/JNKs with 0.5 mM H 2 O 2 (Fig. 1A). Activation was reduced at higher concentrations. This activation was never as great as that observed after ischemia (20 min) followed by reperfusion (10 min) (Fig. 1A). We also confirmed that ischemia alone (20 min) did not activate the SAPKs/JNKs (Fig. 1A). JNK1 antibodies immunoprecipitated all of the p46 SAPK/JNK activity (Fig. 1B, top panel) and protein (Fig. 1B, bottom panel), and approximately 25% of the p54 SAPK/JNK activity (Fig. 1B, top panel). It is not possible to determine the proportion of p54 SAPK/JNK protein immunoprecipitated by the JNK1 antibody (Fig. 1B, bottom panel) because of the interference by immunoglobulins in this region of the gel. These data confirm that the activities principally responsible for phosphorylation of c-Jun(1-135) in the in gel kinase assays (Fig. 1A) were SAPKs/JNKs.
Analogous experiments showed that p38-MAPK was also maximally activated by perfusion with 0.5 mM H 2 O 2 for 30 min ( Fig. 2A). In this case, activation was comparable with that induced by ischemia (20 min) alone or ischemia (20 min) followed by reperfusion (10 min) ( Fig. 2A). In gel phosphorylation of GST-MAPKAPK2(46 -400) was essentially completely inhibited by 10 M SB203580 confirming that the activity responsible was p38-MAPK (Fig. 2B). p38-MAPK antibodies immunoprecipitated only approximately 50% of the p38-MAPK activity (Fig. 2C, top panel) and p38-MAPK protein detected on immunoblots (Fig. 2C, bottom panel). This proportion was not altered by increasing the p38-MAPK antibody concentration (results not shown), demonstrating that the incomplete immunopre-cipitation did not result from insufficient antibody. Furthermore, the apparent molecular mass of the immunoprecipitated p38-MAPK activity (Fig. 2C, top panel) and protein (Fig. 2C, bottom panel) was slightly greater than that remaining in the supernatant, indicating that more than one isoform of p38-MAPK may be activated in perfused rat heart. It is apparent that although this antibody will detect more than one denatured form of p38-MAPK (Fig. 2C, bottom panel), not all native forms are recognized (Fig. 2C, top panel).
Activation of ERKs-Activation of MAPKs by 0.5 mM H 2 O 2 was mainly confined to the stress-regulated forms. In gel kinase assays showed minimal phosphorylation of the ERK substrate, MBP, by extracts of hearts perfused with H 2 O 2 for 30 min (results not shown). However, because ERKs are generally activated at earlier times than the SR-MAPKs in heart (29 -31), activation of ERKs by H 2 O 2 was also assessed in hearts perfused for 5 min using both in gel kinase assays and following FPLC on Mono Q columns. As a positive control, the effects of phorbol 12-myristate 13-acetate (PMA, which powerfully activates ERKs in hearts (29,43,44)) were also studied. p42 and p44 ERKs eluted from the Mono Q column at 0.22 M and 0.26 M NaCl, respectively, consistent with published data (29,43,44) and were activated in hearts perfused with PMA (1 M, 5 min) (Fig. 3, A and B). Relatively little activation of p42 and p44 ERKs (ERK2 and ERK1 respectively) was observed after perfusion with H 2 O 2 (Fig. 3, A and B). The identities of the kinases detected after FPLC were confirmed as p42 and p44 ERKs by in gel kinase assays with MBP as substrate (Fig. 3B).
Activation of MAPKAPK2-MAPKAPK2 is an established substrate of p38-MAPK (18). The activation of MAPKAPK2 was studied following FPLC on Mono S columns. Ischemia (Fig.  4A), ischemia/reperfusion (Fig. 4B), and H 2 O 2 (Fig. 4C) all induced activation of MAPKAPK2. The effects of ischemia and H 2 O 2 were comparable, but the effects of ischemia/reperfusion were consistently greater. SB203580 is a selective inhibitor for p38-MAPK and inclusion of 10 M SB203580 (in Me 2 SO, 0.1% (v/v), 14 mM final concentration) in the perfusion media abolished the activation of MAPKAPK2 by all of these interventions (Fig. 4, A-C). To ensure that these effects were independent of Me 2 SO, 14 mM Me 2 SO was included in all perfusions. Independently, we showed that inclusion of this concentration of Me 2 SO had no effect on the activation of p38-MAPK or MAPKAPK2 by ischemia or ischemia/reperfusion as compared with perfusions in the absence of Me 2 SO (results not shown).

Effects of Free Radical Scavengers on the Activation of SR-MAPKs and MAPKAPK2 by Ischemia and Ischemia/Reperfusion-
To demonstrate that ROS and other free radicals are involved in the activation of SR-MAPKs and MAPKAPK2 by ischemia and/or ischemia/reperfusion, hearts were perfused with the OH ⅐ scavenger, Me 2 SO (0.4% (v/v), 56 mM), or with the lipophilic spin trap radical scavenger, BPN (10 mM, added in Me 2 SO (56 mM final concentration)). Perfusion of hearts with Me 2 SO or BPN/Me 2 SO under control conditions (no interruption in coronary flow) did not affect the activity of either p38-MAPK or SAPKs/JNKs. The activation of p38-MAPK by ischemia was unaffected by the presence of Me 2 SO but was greatly reduced by BPN/Me 2 SO (Fig. 6A). Consistent with this, the activation of MAPKAPK2 by ischemia was not reduced (and may even be increased) in the presence of Me 2 SO, whereas BPN essentially abolished the activation of MAPKAPK2 by ischemia (Fig. 7). In contrast, the activation of SAPKs/JNKs (Fig. 6B, upper panel), p38-MAPK (Fig. 6B, lower panel) and MAPKAPK2 (Fig. 7) by ischemia/reperfusion was essentially completely inhibited by 56 mM Me 2 SO. DISCUSSION Global ischemia stimulates p38-MAPK and MAPKAPK2 activities in the perfused heart (32,33). On reperfusion of ischemic hearts, the stimulation of p38-MAPK and MAPKAPK2 are maintained or increased (32,33). In addition, activities of SAPKs/JNKs are stimulated by ischemia/reperfusion (32)(33)(34). A variety of potentially cytotoxic activators of the SR-MAPKs are produced by the heart in response to ischemia and reperfusion. However, it is not clear which agent(s) is responsible for the activation of these kinases. It was recognized some years ago that there was a significant release of ROS under these conditions (reviewed in Refs. 1, 2, and 4) . (48,49) have been detected in ischemia and in reperfusion following ischemia. These species then induce the production of other radicals (e.g. alkoxy and alkyl radicals) by reaction with membrane lipids (50). Here, we have assessed the potential of ROS, as exemplified by H 2 O 2 , to activate the SR-MAPKs in perfused heart and tested whether the production of ROS and free radicals may be responsible for the activation of these kinases by ischemia and ischemia/reperfusion. (Fig. 1, A and B) and p38-MAPK (Fig. 2, A and C). These results contrast with those of Knight and Buxton (34), who failed to detect activation of SAPKs/JNKs in hearts perfused with 0.5 mM H 2 O 2 (34). Activation of both SAPKs/JNKs and p38-MAPK appeared to be critically dependent on the concentration of H 2 O 2 because there was minimal activation at 0.1-0.2 mM, but with 0.5 mM H 2 O 2 activation was maximal (Figs. 1A and 2A). At higher concentrations (1 mM H 2 O 2 ), SAPK/JNK activity declined (Fig. 1A), but p38-MAPK activity was maintained. A similar activation pattern of the SAPKs/JNKs has been noted in astrocytes exposed to H 2 O 2 (51). The activation of p38-MAPK by 0.5-1 mM H 2 O 2 was comparable with the activation seen after ischemia and ischemia/reperfusion (Fig. 2A). These data suggest that increases in H 2 O 2 concentrations (or other ROS) in the heart during ischemia and ischemia/reperfusion (45) could play a role in activation of p38-MAPK. In contrast to the activation of p38-MAPK, the activation of SAPKs/JNKs by 0.5 mM H 2 O 2 was greater than that seen after ischemia but less than that after ischemia/reperfusion (Fig. 1A). This suggests that factors other than ROS may be involved in the activation of SAPKs/JNKs during ischemia/reperfusion.

Activation of SR-MAPKs and MAPKAPK2 by H 2 O 2 -Perfusion of rat hearts with H 2 O 2 stimulated SAPKs/JNKs
The species of ROS responsible for the activation of the SR-MAPKs by H 2 O 2 is not clear. Exposure of isolated hearts or ventricular myocytes to H 2 O 2 leads to an iron-dependent formation of OH ⅐ (52-54). Although OH ⅐ may be partly responsible for some of the cardiotoxic effects of H 2 O 2 , OH ⅐ -independent effects of H 2 O 2 have also been detected (55). At low concentrations (14 mM), the OH ⅐ scavenger Me 2 SO did not significantly inhibit the activation of p38-MAPK or MAPKAPK2 (an index of p38-MAPK activation) by 0.5 mM H 2 O 2 (results not shown). Further experiments examining the effects of higher concentrations of Me 2 SO on the activation of SR-MAPKs by H 2 O 2 are indicated.
Several isoforms of the SAPKs/JNKs have been identified by molecular cloning (56). At least three genes produce alternatively spliced transcripts encoding proteins of approximately 46 and 54 kDa (56). An antibody to human JNK1 immunoprecipitated all of the p46 SAPK/JNK activity and approximately 25% of the p54 SAPK/JNK activity stimulated by ischemia, ischemia/reperfusion, or H 2 O 2 in perfused heart (Fig. 1B, top panel). This suggests that there is activation of at least one other isoform of p54 SAPK/JNK. Although antibodies to JNK2 detect  a protein in neonatal rat ventricular myocytes, we found that these antibodies were not suitable for immunoprecipitation (results not shown).
SB203580 is a selective inhibitor of p38-MAPK (17,57). The 38-kDa MAPK activity detected using in gel assays with GST-MAPKAPK2(46 -400) as a substrate was completely inhibited by 10 M SB203580 (Fig. 2B), indicating that this activity is indeed attributable to p38-MAPK. However, an antibody to the C terminus of murine p38-MAPK immunoprecipitated only approximately 50% of the activity (Fig. 2C, top panel). The immunoprecipitated form of p38-MAPK migrated slightly more slowly than the residual activity in the supernatant (Fig. 2C,  top panel), suggesting that more than one isoform of p38-MAPK is activated by ischemia, ischemia/reperfusion, or H 2 O 2 in perfused heart. Using the same antibody for Western blots, bands of approximately 38 kDa were detected in both the immunoprecipitates and the residual supernatants (Fig. 2C,  bottom panel), although, consistent with the p38-MAPK activity data (Fig. 2C, top panel), the band in the immunoprecipitates migrated more slowly than the band in the supernatants (Fig. 2C, bottom panel). This suggests that although the antibody is more selective for a particular isoform(s) of p38-MAPK in the native form, it also detects other forms after denaturation. At least five isoforms of p38-MAPKs have been identified: p38-MAPK (15,16,18) of which there are two alternatively spliced isoforms (17), p38-MAPK␤ (58), p38-MAPK␥ (20,59), and p38-MAPK␦ (60,61). However, p38-MAPK␥ and p38-MAPK␦ transcripts are minimally expressed in heart and neither phosphorylates MAPKAPK2 effectively in an in gel kinase assay (59,61). Furthermore, p38-MAPK␥ and p38-MAPK␦ are resistant to inhibition by SB203580 (20,60). It is thus likely that ischemia and ischemia/reperfusion activate the alternatively spliced isoforms of p38-MAPK and/or p38-MAPK␤ in the heart.
One downstream substrate of p38-MAPK is MAPKAPK2 (18). Perfusion of hearts with 0.5 mM H 2 O 2 for 30 min powerfully activated MAPKAPK2 (Fig. 4C). The activation was similar to that seen after ischemia (Fig. 4A) but less than that after ischemia/reperfusion (Fig. 4B). In all cases, inclusion of 10 M SB203580 in the perfusion medium abolished MAPKAPK2 activation (Fig. 4, A-C). These data are consistent with activation of p38-MAPKs stimulating MAPKAPK2 activities.
Activation of ERKs by H 2 O 2 -H 2 O 2 activates the ERKs in HeLa, Rat1, NIH 3T3, and PC12 cell lines (62) and in primary cultures of rat astrocytes (51). There is disagreement about whether H 2 O 2 does (62) or does not (63) produce a significant activation of ERKs in primary cultures of vascular smooth muscle cells. In the study in which no ERK activation was detected (63), there was significant activation of ERKs by another ROS, namely O 2 . . We detected only minimal ERK1 and ERK2 activation in hearts perfused with 0.5 mM H 2 O 2 for 5 min, although there was significant activation in hearts perfused with 1 M PMA (Fig. 3, A and B). There was no detectable activation of ERKs after 30 min of perfusion with 0.5 mM H 2 O 2 (results not shown). These data contrast with a previous study that showed activation of ERKs rather than SAPKs/JNKs in hearts perfused with H 2 O 2 (34). However, in this study (34), ERKs were partially purified by batch elution from DEAE-Sephacel with 0.5 M NaCl and assayed with a tetrapeptide derived from MBP. Such an elution protocol would elute all MAPK species and is therefore less specific than separation on Mono Q FPLC using a NaCl gradient, which was used here.
Mechanisms of Activation of SR-MAPKs and MAPKAPK2 during Ischemia and Ischemia/Reperfusion-As discussed above, H 2 O 2 generates free radical ROS in the heart and activates SR-MAPKs (Figs. 1 and 2) and MAPKAPK2 (Fig. 4).
Equally, ROS are generated in the heart during ischemia and ischemia/reperfusion. We therefore sought evidence that ROS and other free radicals mediate the activation of SR-MAPKs in ischemia and ischemia/reperfusion. We used two different free radical scavengers, the OH ⅐ radical scavenger Me 2 SO and the lipophilic spin trap BPN. Both have been found to be cardioprotective under situations of increased oxidative stress (54,64). At a concentration of 10 mM BPN/56 mM Me 2 SO (solvent carry-over), activation of p38-MAPK and MAPKAPK2 by ischemia was inhibited (Figs. 6A and 7). In contrast, 56 mM Me 2 SO alone inhibited activation of SR-MAPKs and MAPKAPK2 after ischemia/reperfusion ( Fig. 6B and 7). These data implicate ROS and free radicals in the activation of SR-MAPKs during ischemia and ischemia/reperfusion but suggest that different radicals may be involved during the ischemic and reperfusion phases. It has recently been suggested that O 2 . and H 2 O 2 are formed during simulated ischemia in isolated cardiac myocytes, whereas OH ⅐ and further H 2 O 2 are generated during simulated reperfusion (65). This is entirely consistent with our data (Figs. 6 and 7). Activation of SR-MAPKs and MAPKAPK2 by High Pressure Perfusion-Using in gel kinase assays with MBP as a substrate, we have previously shown that short term perfusion (5 min) of hearts at high aortic pressures activates ERKs (44). Furthermore, the upstream activators of the ERKs, the MAPK (or ERK) kinase group of MAPK kinases, were also activated (44). This work was completed prior to the identification of the SR-MAPKs. Here, we show that perfusion of hearts at high aortic pressure for longer times (20 min) additionally stimulated the activities of SAPKs/JNKs, p38-MAPK, and MAP-KAPK2 (c.f. ischemia, ischemia/reperfusion, and H 2 O 2 ). Although the mechanism of activation of the SR-MAPKs by high pressure perfusion is unclear, it could represent ischemia/ reperfusion effects, because hypertension in vivo is known to induce subendomyocardial ischemia (reviewed in Ref. 3).
Significance of SR-MAPKs in the Heart-The biological consequences of activation of SR-MAPKs in the heart are poorly understood. Activation of the SAPKs/JNKs would be expected to result in phosphorylation of the c-Jun and ATF2 transcription factors, increasing their trans-activating activity (reviewed in Ref. 6). In this regard, we have shown that hyperosmotic stress activates SAPK/JNKs in ventricular myocytes (31), and c-Jun and ATF2 become phosphorylated (66). Transcription of c-jun is regulated by a number of cis-acting regulatory sequences in the c-jun promoter region, including two sites (jun1 and jun2) that bind c-Jun/ATF2 heterodimers. Thus activation of SAPKs/JNKs is potentially able to up-regulate c-jun expression. Increased expression of c-jun occurs during ischemia/reperfusion in isolated hearts (67) and during hypoxia in cultured myocytes (68). Furthermore, H 2 O 2 induces c-jun expression in NIH 3T3 cells (69). Thus SAPK/JNK-dependent activation of c-Jun might be expected to increase gene expression and anabolism. Indeed, SAPKs/JNKs have been proposed to be mediators of the ␣ 1 -adrenergic stimulation of hypertrophic growth in the ventricular myocyte (70). However, activation of SAPKs/JNKs and c-Jun (71) induces apoptosis in a number of cell lines, which may be particularly pertinent in the ischemic heart. The p38-MAPKs have also been implicated in ␣ 1 -adrenergic stimulation of hypertrophic growth in the ventricular myocyte (72). One of their substrates (MAPKAPK2) phosphorylates the small heat shock proteins Hsp25/27, which may be cytoprotective (26). However, activation of p38-MAPK may be apoptotic (73). The ultimate biological effects of activation of SR-MAPKs may depend on the duration and extent of their activation (74).