Activation of 5′-AMP-activated Kinase Is Mediated through c-Src and Phosphoinositide 3-Kinase Activity during Hypoxia-Reoxygenation of Bovine Aortic Endothelial Cells

AMP-activated kinase (AMPK) is a fuel-sensing enzyme present in most mammalian tissue. In response to a decrease in the energy state of a cell AMPK is phosphorylated and activated by still poorly characterized upstream events. Exposure of bovine aortic endothelial cells (BAEC) to chemically synthesized ONOO– acutely and significantly increased phosphorylation of c-Src, PDK1, AMPK, and its downstream target, acetyl-CoA carboxylase (ACC), without affecting cellular AMP. This novel pathway for AMPK activation was confirmed by the use of pharmacological inhibitors and dominant-negative mutants. Exposure of BAEC to hypoxia-reoxygenation (H/R) caused a biphasic increase in AMPK and ACC phosphorylation, which was prevented by adenoviral overexpression of superoxide dismutase (SOD) or inhibition of nitric-oxide synthase (NOS) implicating a role of ONOO– formed during H/R. Furthermore, dominant-negative mutants of c-Src or kinase-defective PDK1 also blocked H/R-induced AMPK activation indicating that, as with addition of exogenous ONOO–, both c-Src and PI 3-kinase are upstream of AMPK. Moreover, H/R, like ONOO–, significantly increased co-immunoprecipitation of AMPK with c-Src, suggesting that ONOO– favors physical association of AMPK with upstream kinases. Taken together, our results indicate a novel pathway by which H/R via ONOO– activates AMPK in a c-Src-mediated, PI 3-kinase-dependent manner, and suggest that ONOO–-induced activation of AMPK might thereby regulate metabolic enzymes, such as ACC.

rapidly increased AMPK phosphorylation as well as its activity on downstream targets (22). Because vascular endothelial cells are able to generate O 2 . and NO simultaneously under pathological conditions such as hypoxia-reoxygenation (H/R) (24 -25), we further investigated whether or not H/R via ONOO Ϫ could regulate AMPK activity and if so, the mechanism of activation. Here we report that upstream of AMPK, ONOO Ϫ activates c-Src and PI 3-kinase without a change in cellular AMP or ATP content. Furthermore, this novel activation scheme may be implicated during H/R where we found that AMPK activation depends on ONOO Ϫ formation, as well as c-Src and PI 3-kinase. As evidenced by co-immunoprecipitation, both ONOO Ϫ and H/R lead to physical association of c-Src and AMPK suggesting that this enables this signaling pathway for activation of AMPK downstream targets including ACC.
This article has been withdrawn by the authors. Analysis performed by the Journal determined the following. The c-Src immunoblot in

Methods
Peroxynitrite Synthesis-ONOO Ϫ was either obtained from Calibiochem (San Diego, CA) or synthesized using a quenched-flow reaction as previously described (19). Briefly, an aqueous solution of 0.6 M sodium nitrite was mixed rapidly with an equal volume of 0.7 M H 2 O 2 containing 0.6 M HCl and immediately quenched with the same volume of 1.5 M NaOH. Residual H 2 O 2 was removed by treatment with granulate manganese dioxide. All solutions were kept on ice. The concentrations of ONOO Ϫ were determined spectrophotometrically in 0.1 M NaOH (⑀ 302 ϭ 1670 M Ϫ1 s Ϫ1 ).
Treatment of BAEC Cells with Peroxynitrite-To confluent BAEC in 6-well plates was added 950 l of 100 mmol/liter HEPES buffer, pH 7.4 after being rinsed twice with phosphate-buffered saline buffer, pH 7.4. 50 l of concentrated ONOO Ϫ in 0.1 mol/liter NaOH was evenly but quickly added to the plates while being rapidly rotated on orbital shakers at room temperature. There was no pH shift during treatment with ONOO Ϫ . The same volumes of 0.1 mol/liter NaOH or decomposed ONOO Ϫ were used as controls (ONOO Ϫ was first decomposed in 1 mol/liter Tris buffer, pH 7.4, and kept for 5 min or overnight).
Adenoviral Transfection-BAEC were transfected with adenovirus expressing green fluorescence protein (GFP) as a control, superoxide dismutase (SOD1 or SOD2), catalase, or dominant-negative mutants for c-Src (c-Src-DN) (26 -27), a kinase-defective mutant PDK1 (PDK1-KD, mutation of Lys 114 to glycine) (28 -30). BAEC were infected in medium with 2% fetal calf serum overnight. The cells were then washed and incubated in fresh EGM medium without fetal calf serum for an additional 12 h prior to experimentation. Using these conditions, infection efficiency was typically Ͼ80% as determined by GFP expression.
Measurement of Cellular ATP, ADP, and AMP Content-After treatment with ONOO Ϫ or vehicle, BAEC (100-cm 2 dishes) were immediately covered with 6 ml of ice-cold 1% trichloroacetic acid and kept on ice for 5 min. The cells were then scraped and centrifuged at 4°C (5 min, 14,000 rpm). After centrifugation, the supernatants were neutralized by ether extraction, freeze-dried (SpeedVac) and then stored at Ϫ80°C until they were re-dissolved in 0.5 ml water for assay. The contents of ATP, ADP, and AMP were assayed by bioluminescent methods as described previously (31).
Hypoxia Reoxygenation of BAEC-BAEC were cultured in 6-well plates. The cells were first transfected with adenoviral vectors for 2 days if required. The cells were placed in a water bath (37°C, total volume of 1 liter) filled with 1 prewarmed Krebs-Ringer's buffer, gassed with 95% O 2 , 5% CO 2 . After 30 min incubation, the oxygen tension was reduced abruptly from 95% O 2 , 5% CO 2 to 95% N 2 , 5% CO 2 and was maintained for the time indicated. After this phase of hypoxia, 95% O 2 /5% CO 2 was resumed (reoxygenation) for 30 min. After that the cells were washed with phosphate-buffered saline buffer twice and collected for Western blot and immunoprecipitation assays. Control BAEC were gassed only with 95% O 2 , 5% CO 2 for equivalent periods.
Immunoprecipitation and Western Blots-Immunoprecipitation and Western blots were performed as described previously (21)(22). To determine the interaction of c-Src and AMPK, c-Src was immunoprecipitated and stained for AMPK, or vice versa.
Quantification of Western Blot-The intensity (area ϫ density) of the individual bands on Western blots was quantitated by densitometry (Model GS-700, Imaging Densitometer, Bio-Rad). The background was subtracted from the calculated area. The results were calculated as percentage change compared with the corresponding control band.
Assay of AMP Kinase Activity-Confluent BAEC were exposed to ONOO Ϫ or subjected to H/R as described above. After treatment, cells were immediately washed with 2 ml of ice-cold phosphate-buffered saline buffer and scraped with a rubber spatula in lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 1 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride). The cell lysates were then sonicated twice for 10 s in an Ultrasonic dismemberator with output 10% (Model 500, Fisher Scientific) and then centrifuged at 14,000 ϫ g for 20 min at 4°C. The pellets were discarded and supernatants were assayed for protein concentration. Duplicate tubes with 200 g protein from each sample were prepared and were mixed with 500 l of IP buffer (lysis buffer plus 1 M NaCl and 1 mM dithiothreitol). AMPK was then immunoprecipitated by adding 10 g of polyclonal antibody against AMPK (Cell Signaling) and 25 l of protein A-G agarose (Santa Cruz Biotechnology) and incubated at 4°C. After centrifugation (14,000 ϫ g, 1 min), the beads were washed with IP buffer and then twice with 10ϫ reaction buffer (400 mM HEPES, pH 7.4, 800 mM NaCl, 50 mM MgCl 2 , 1 mM dithiothreitol). The AMPK activity was assayed by adding 50 l of reaction mixtures, consisting of 5 l of reaction buffer, 10 l of SAMS peptide (1 mg/ml), 10 l of ATP working stock consisting of 0.1 l of 100 mM ATP, 1 l of [ 32 P]ATP, and 8.9 l of H 2 0, 25 l H 2 0, or 25 l of 400 M AMP and incubated at 37°C for 10 min. The beads were quickly pelleted and 25 l of supernatant was spotted onto P81 Whatman paper. The filter papers were then washed 4 -5 times with 1% phosphoric acid. After the final wash, the filters were quickly dried and counted in a scintillation counter. The difference between the presence and absence of AMP is calculated as the AMPK activity.

Activation of AMPK by ONOO Ϫ Is Independent of Cellular
AMP/ATP Ratio-Because the cellular AMP/ATP ratio is the best recognized factor controlling AMPK activity (1-6), we first determined if ONOO Ϫ increased the AMP/ATP ratio. As shown in Fig. 1a, treating cells with ONOO Ϫ (100 mol/L) neither decreased the content of ATP or ADP nor increased that of AMP in BAEC (Fig. 1a). Because the concentrations of AMP were low, we next determined if ONOO Ϫ affected the ADP to ATP ratios, an index that reflect the ratio of AMP to ATP. As shown in Fig. 1b, ONOO Ϫ did not affect the ratios of ADP to ATP, which was in line with unaffected AMP content. These results suggest that ONOO Ϫ activated AMPK via a mechanism other than an increased AMP.
ONOO Ϫ -mediated AMPK Activation Is c-Src-dependent-The Src family of protein tyrosine kinases is important in the regulation of cell growth and differentiation and has been reported to play an important role in cell stress responses. Phosphorylation of Tyr 416 of c-Src, which is located in the activation loop of the kinase domain, up-regulates the activity of the kinase (32). We next investigated if exogenous ONOO Ϫ activates AMPK via c-Src in BAEC. As shown in Fig. 1c, addition of chemically synthesized ONOO Ϫ (1-100 mol/liter), but not decomposed ONOO Ϫ , dose-dependently increased phosphorylation of c-Src (Tyr 416 ), indicating activation of c-Src. We next addressed whether or not c-Src activation contributed to ONOO Ϫ -induced AMPK activation. Phosphorylation of Thr 172 in the active site of AMPK is essential for AMPK activity (9 -10), and therefore reflects AMPK activity. Using a highly specific antibody against phosphorylated Thr 172 of AMPK, we found that PP2 (10 mol/liter), which selectively inhibits both c-Src activity and c-Src phosphorylation (33)(34), not only attenuated ONOO Ϫ -induced AMPK (Thr 172 ) phosphorylation, but also prevented phosphorylation of ACC, indicating inhibition of AMPK activity (Fig. 1d). Moreover, overexpression of a dominant-negative c-Src mutant (c-Src-DN) attenuated the increased phosphorylation of AMPK as well as that of ACC (Fig. 1d) indicating that ONOO Ϫ activates AMPK at least in part via c-Src. The inhibition of c-Src by both PP2 and c-Src-DN overexpression was further confirmed by decreasing phospho-rylation of SAMS peptide using [ 32 P]ATP assays (Fig. 1e). As shown in Fig. 1e, both PP2 and c-Src-DN strongly inhibited ONOO Ϫ up-regulated phosphorylation of SAMS peptides.
In search of intermediates involved in the activation of AMPK by c-Src, we next addressed whether or not activation of c-Src by ONOO Ϫ leads to PI 3-kinase activation. Phosphoinositide-dependent kinase-1 (PDK1), which is activated by PI 3-kinase lipid products serves as a link between PI 3-kinase and its downstream targets, such as p70S6 kinase (33,34). As shown in Fig. 1f, ONOO Ϫ increased the phosphorylation of PDK1 and p70S6 kinase. Inhibition of c-Src by PP2 or by overexpressing the c-Src-DN mutant blunted the effect of ONOO Ϫ on both PDK1 and p70S6 kinase, suggesting that c-Src activation by ONOO Ϫ is linked to activation of PI 3-kinase.
ONOO Ϫ Activates AMPK via PI 3-Kinase-To further explore the role of PI 3-kinase, we investigated if growth factors that are recognized to activate PI 3-kinase stimulated AMPK. As expected, VEGF (50 ng/ml) or insulin (0.01 Unit/ml, 5 nM) added for 10 min did not affect AMPK or ACC phosphorylation. However, in the presence of each growth factor, ONOO Ϫ increased both AMPK and ACC phosphorylation to a similar extent as in the absence of the growth factor (Fig. 2a). This suggests that PI 3-kinase activation is not normally linked to AMPK activation.
It was therefore interesting to investigate whether or not inhibition of PI 3-kinase prevented the ONOO Ϫ -enhanced phosphorylation of AMPK as well as its downstream targets. Either wortmannin (100 nmol/liter) or LY294002 (50 mol/ liter) prevented ONOO Ϫ -enhanced phosphorylation of PDK1 and p70S6 kinase (Fig. 2b). In addition, although neither of the PI 3-kinase inhibitors affected the basal level of AMPK Thr 172 phosphorylation (data not shown), both inhibitors prevented ONOO Ϫ -enhanced AMPK phosphorylation (Fig. 2b). In parallel, either wortmannin or LY294002 also attenuated ACC phosphorylation that was enhanced by ONOO Ϫ (Fig. 2b) and activity of AMPK as indicated by phosphorylation of the SAMS peptide (Fig. 2c). In contrast, although PP2 prevented c-Src phosphorylation following ONOO Ϫ , neither wortmannin nor LY294002 affected the enhanced c-Src phosphorylation caused by ONOO Ϫ , consistent with c-Src being upstream of PI 3-kinase (Fig. 2d). There was no effect of wortmannin or LY294002 on AMPK or ACC phosphorylation under basal conditions in the absence of ONOO Ϫ (data not shown), again indicating that PI 3-kinase is not sufficient for AMPK phosphorylation. Fig. 2, e and f, overexpression of a PDK1-KD mutant prevented ONOO Ϫ -enhanced phosphorylation of both PDK1 and p70S6 kinase. We also found that the PDK-1-KD decreased ONOO Ϫ -enhanced phosphorylation of AMPK and ACC (Fig. 2, e and f) indicating that ONOO Ϫ activates AMPK via PI 3-kinase and PDK1. Importantly, the PDK-1-KD had no effect on AMPK or ACC phosphorylation under basal conditions in cells not exposed to ONOO Ϫ (data not shown). Thus, these data indicate that ONOO Ϫ activates a signaling pathway involving c-Src, PI 3-kinase, and PDK1 that leads to activation of AMPK and ACC.

PI 3-Kinase Activates AMPK via Its Downstream Kinase, PDK1-As shown in
Coordinated Activation of c-Src, AMPK, and ACC by Hypoxia-Reoxygenation-Because AMPK is activated by cellular stresses such as ischemia (1-6, 12-13) and because ischemia is associated with the generation of ONOO Ϫ in endothelial cells (23)(24)(25), we determined if H/R activated a similar signaling sequence as does synthesized ONOO Ϫ . As shown in Fig. 3a, short term exposure to hypoxia (5 or 15 min) increased phosphorylation of c-Src, AMPK, and ACC. Following the initial increased activation, a significant decline in phosphorylated c-Src, AMPK and ACC occurred after 30 min of hypoxia (Fig. 3,  a and b). Reoxygenation of hypoxic BAEC caused another increase in phosphorylated c-Src, AMPK, and ACC with the maximal effect seen after 5 min. Prolonged reoxygenation did not further increase, but rather diminished the phosphorylation of AMPK and ACC at 30 min (Fig. 3, a and b). These data indicate that phosphorylation of c-Src, AMPK, and ACC are temporally coordinated during H/R, and that activation of AMPK after 30 min of hypoxia is less at a time when ATP depletion would be expected to be greatest.
ONOO Ϫ Mediates H/R Activation of c-Src and AMPK-Because our studies indicate that ONOO Ϫ can activate c-Src and AMPK, we next determined if endogenous generation of ONOO Ϫ was involved in c-Src and AMPK activation caused by H/R. Production by BAEC of ONOO Ϫ was inhibited either by overexpressing SOD to scavenge O 2 . , or treating the cells with L-NAME (1 mmol/liter) to prevent formation of NO. As shown in Fig. 4, a and b, overexpression of SOD1 or SOD2, or treatment of the cells with L-NAME, attenuated H/R-enhanced c-Src, AMPK, and ACC phosphorylation, as well as AMPK activity as assayed by SAMS peptide phosphorylation. Infection of cells with adenovirus encoding GFP alone had no effect on H/R-induced phosphorylation, excluding a nonspecific effect of viral infection. Because we previously showed that neither NO nor O 2 . alone activated AMPK in BAEC (22), the attenuation of c-Src and AMPK phosphorylation by either L-NAME or overexpression of SOD1 and SOD2 is likely explained by the formation of ONOO Ϫ during H/R, which then activates both c-Src and AMPK kinase in the coordinated fashion observed. c-Src-mediated and PI 3-Kinase-dependent AMPK Activation in Hypoxia-Reoxygenated BAEC-We next determined whether or not c-Src and PI 3-kinase are involved in AMPK activation in H/R-treated cells due to the generation of ONOO Ϫ . Overexpression of the c-Src-DN mutant significantly blunted the H/R-enhanced AMPK phosphorylation (Fig. 4c). Furthermore, the c-Src-DN also blocked the increased phosphorylation of ACC during H/R (Fig. 4c). In parallel, overexpression of the PDK-1-KD mutant, also prevented H/R-enhanced phosphorylation of both AMPK and ACC (Fig. 4c). While it was not feasible to treat BAEC exposed to H/R with wortmannin or LY294002 because of the large volume of buffer bathing the cells, these studies suggest that like chemically synthesized ONOO Ϫ , H/R activates AMPK and ACC via a c-Src and PI 3-kinase-dependent mechanism. Our previous studies showed that ONOO Ϫ increased phosphorylation of another target of AMPK, eNOS-Ser 1179 (22). In other studies, we also found that H/R increases eNOS-Ser 1179 phosphorylation, and that this is prevented by the c-Src-DN and PDK-1-KD mutants (data not shown).
We previously reported that ONOO Ϫ increases the association of AMPK with its downstream target, eNOS. To better understand how ONOO Ϫ and H/R activate AMPK, we determined if c-Src was also physically associated with AMPK after ONOO Ϫ , or after H/R treatment of BAEC. As shown in Fig. 4d, addition of ONOO Ϫ to BAEC increased the association of c- Src   FIG. 4. ONOO ؊ -dependent activation of c-Src, AMPK, and ACC in BAEC exposed to hypoxia-reoxygenation. a, phosphorylation of c-Src, AMPK, and ACC in BAEC treated with hypoxia for 30 min followed by reoxygenation for 30 min was inhibited by the NOS inhibitor, L-NAME, or overexpression of SOD1 or SOD2 (Control versus H/R, GFP plus H/R, or catalase plus H/R, n ϭ 5, #, p Ͻ 0.01; H/R versus H/R plus L-NAME, SOD-1, SOD-2, n ϭ 5, *, p Ͻ 0.01). b, decreased formation of ONOO Ϫ by overexpression of SOD1 or SOD2 or adding the NOS inhibitor, L-NAME) blocked H/R-enhanced phosphorylation of SAMS peptide in BAEC (Control versus H/R or GFP plus H/R, n ϭ 5, #, p Ͻ 0.01; H/R versus H/R plus SOD or L-NAME, n ϭ 5, *, p Ͻ 0.01). c, ONOO Ϫ increased the physical association of AMPK and c-Src. d, overexpression of c-Src-DN and PDK-1-KS attenuated H/R-enhanced phosphorylations of AMPK and ACC (Control versus GFP plus H/R, n ϭ 6, #, p Ͻ 0.01; GFP plus H/R versus H/R plus PDK-1-KD or c-Src-DN, n ϭ 6, *, p Ͻ 0.01). e, to determine the interaction of c-Src and AMPK caused by ONOO Ϫ , either c-Src or AMPK were immunoprecipitated, and blotted for both c-Src and AMPK, or vice versa. The blots are representative of 3-5 independent experiments. f, H/R also increased the association of c-Src with AMPK in BAEC, which was prevented by over-expression of SOD1 or SOD2, or by L-NAME (1 mmol/liter). The blots are representative of three independent experiments. with AMPK-␣1/␣2 subunits. To further clarify the AMPK subunits interacting with c-Src, we first imunoprecipitated c-Src and Western blotted with the antibodies against AMPK␣ (␣1 or ␣2) subunits. C-Src was found to mainly cross-react with AMPK␣1 subunit, but a weak staining was also seen with AMPK␣2 (data not shown). H/R, like ONOO Ϫ , also significantly increased the association of c-Src and AMPK-␣1 subunit (Fig. 4e).
To address whether or not ONOO Ϫ during H/R is responsible for the increased association of c-Src with AMPK, the effects of blocking ONOO Ϫ production during H/R on the association of c-Src with AMPK was determined. Adenoviral overexpression of SOD1 and SOD2, but not GFP attenuated H/R-induced association of AMPK and c-Src (Fig. 4e). L-NAME also blunted the effects of H/R on c-Src and AMPK association. However, overexpression of catalase had no effect, excluding a role of hydrogen peroxide (Fig. 4e). These data suggest that H/R increases the physical association of c-Src and AMPK via the formation of ONOO Ϫ . DISCUSSION H/R was one of the first identified physiological activators of AMPK (1)(2)(3)(4)(5)(6)(12)(13)(14)(15). It is generally believed that H/R activates AMPK by increasing the AMP/ATP ratio by inhibiting mitochondrial ATP generation (1)(2)(3)(4)(5)(6). Although H/R activation of AMPK has been demonstrated in various tissues, and a mechanism other than AMP/ATP has been suggested, none has yet been identified. The present study has for the first time demonstrated that H/R via the generation of oxidants, likely ONOO Ϫ , activates AMPK via a c-Src-mediated, PI 3-kinase-dependent pathway. Although an increased AMP/ATP ratio obviously contributes to hypoxia induced AMPK activation in many tissues, ONOO Ϫ generation in BAEC during H/R provides a novel pathway to regulate AMPK independently of altered cellular AMP/ATP ratio.
The evidence that activation of AMPK in the early phase of H/R is likely to be explained by the increased formation of ONOO Ϫ rather than, or in addition to, altered AMP/ATP ratio is severalfold. First, phosphorylated AMPK and ACC were detected as early as the first 5-15 min of hypoxia. In addition, after 30 min of hypoxia there was a decreased level of phosphorylated AMPK and ACC compared with those at 5 and 15 min. Because prolonged periods of hypoxia would be expected to further increase the AMP/ATP ratio, these data suggest a mechanism other than AMP/ATP ratio that can temporally activate AMPK. Reoxygenation of hypoxic BAEC, which is associated with the generation of ONOO Ϫ in endothelial cells (23)(24)(25)(26), also activated AMPK and ACC for 5 to 30 min, suggesting that the AMP/ATP ratio is not the sole factor that determines AMPK activity.
Second, inhibition of ONOO Ϫ formation by overexpression of SOD (to scavenge O 2 . ) or NOS inhibition with L-NAME (to prevent the formation of NO) attenuated H/R-enhanced phosphorylation of both AMPK and ACC. Because NO or O 2 . alone have no effect on AMPK activation (22), and addition of authentic ONOO Ϫ did not affect the AMP/ATP ratio, the attenuation of AMPK and ACC phosphorylation by L-NAME and SOD in cells exposed to H/R suggests the involvement of ONOO Ϫ and can not be explained solely by the AMP/ATP ratio. Third, in parallel with increased AMPK phosphorylation, H/R also increased c-Src phosphorylation. Inhibition of c-Src activity with a dominant-negative mutant also blocked both AMPK and ACC phosphorylation. Because H/R-enhanced c-Src phosphorylation increases in parallel with that of AMPK, and there is no evidence that activation of c-Src is dependent upon the AMP/ATP ratio, these data also suggest that H/R, via c-Src activation, leads to AMPK activation in H/R-treated BAEC.
Fourth, inhibition of PI 3-kinase signaling by overexpression of a PDK-1-KD mutant effectively blocked the phosphorylation of AMPK and ACC, indicating that the downstream effector of PI 3-kinase signaling, PDK-1, is involved in H/R-induced AMPK activation. There is also no evidence to suggest that PI3-kinase or PDK-1 is activated by altered AMP/ATP ratio.
Finally, the postulate that ONOO Ϫ generated during H/R can activate AMPK independently of AMP/ATP ratio is substantiated by the fact that synthetic ONOO Ϫ was shown to do so without a measurable change in AMP/ATP. Thus, our results suggest a signaling pathway initiated by ONOO Ϫ capable of activating AMPK independently of changes in AMP/ATP. This novel activation scheme (Fig. 5) for ONOO Ϫ generated during H/R was anticipated from studies of the effect of synthetic ONOO Ϫ . Relatively low concentrations of ONOO Ϫ dosedependently increased the phosphorylation of c-Src and AMPK. Most importantly, inhibition of c-Src activity with a selective c-Src inhibitor, PP2, or overexpression of a c-Src mutant attenuated ONOO Ϫ -stimulated AMPK and ACC phosphorylation. ONOO Ϫ increased phosphorylation of PDK1 and P70S6 kinase, two downstream targets of PI 3-kinase, confirming that ONOO Ϫ activates PI 3-kinase. Furthermore, either wortmannin or LY294002 abolished ONOO Ϫ -induced AMPK phosphorylation as well as AMPK activity, as evidenced both by phosphorylation of the SAMS peptide and phosphorylation of ACC. Moreover, overexpression of the inactive PDK1 mutant likewise blocked ONOO Ϫ -activated AMPK phosphorylation and activity, strongly indicating that as a consequence of ONOO Ϫ , AMPK is a downstream target of PI 3-kinase and PDK1.
Precisely how PI 3-kinase regulates AMPK remains a subject for further study. Because inhibition of PI 3-kinase either with wortmannin or LY294002, or by overexpressing PDK-1-KS, did not affect basal AMPK phosphorylation in quiescent cells, nor did activation of PI 3-kinase by growth factors such as insulin and VEGF increase phosphorylation of AMPK in the absence of ONOO Ϫ , it is apparent that activation of PI 3-kinase alone is FIG. 5. Scheme illustrating proposed activation of AMPK by ONOO ؊ . ONOO Ϫ formed during hypoxia-reoxygenation or added exogenously activates c-Src which leads to PI 3-kinase and PDK1 activation. AMPK thus activated can phosphorylate and inhibit acetyl-CoA carboxylase and then increases oxidation of free fatty acids (41)(42). not sufficient to activate AMPK under normal conditions. However, the fact that the PDK1-KD or pharmacological inhibitors of PI 3-kinase prevented ONOO Ϫ -induced activation of AMPK, strongly suggests that PI 3-kinase plays a key role in cells challenged with the reactive nitrogen species.
Another novel finding of the present study is that oxidants such as ONOO Ϫ might be important in regulating metabolic enzymes such as ACC. Phosphorylation and inactivation of ACC (35)(36) and/or phosphorylation and activation of malonyl-CoA decarboxylase (37) by ONOO Ϫ -induced activation of AMPK will lead to a decrease in the concentration of malonyl-CoA (25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37). Because malonyl-CoA is a potent allosteric inhibitor for carnitine palmitoyltransferase I (CPT 1) (1-6), a key enzyme involved in fatty acid transport into the mitochondrial matrix, a decrease in malonyl-CoA will lead to an increase in fatty acid oxidation. AMPK and ACC have been reported to play a key role in the physiological regulation of fatty acid oxidation in human umbilical vein endothelial cells (38), and activation of AMPK in hypoxic heart results in phosphorylation and inactivation of ACC. Therefore, increased formation of reactive oxygen and nitrogen species reported to occur in H/R (18 -23) or physiological exercise (38 -40) might contribute to accelerated fatty acid oxidation aiding metabolism in the ischemic heart or exercising muscle. Activation of AMPK by ONOO Ϫ via c-Src, PI 3-kinase, and PDK1 might therefore be implicated in regulating cellular energy status particularly for cells responding to such stresses as H/R or exercise. Our results for the first time demonstrate that oxidants, which have been considered as toxic and damaging to cells in high concentrations, may be involved at lower concentrations as physiological regulators of metabolic and catabolic pathways. In other studies, we also have shown that exaggerated formation of ONOO Ϫ can contribute to metabolic disorders such as diabetes mellitus (21,22).
In summary, we have demonstrated that H/R activates AMPK via temporally coordinated activation of c-Src and PI 3-kinase via the endogenous generation of ONOO Ϫ . Activation by ONOO Ϫ of AMPK via this scheme can proceed independently of altered AMP/ATP ratio, as demonstrated explicitly with authentic ONOO Ϫ . Oxidants such as ONOO Ϫ might thereby regulate metabolic enzymes in many physiological conditions, such as physical exercise and pathophysiological conditions such as hypoxia-reperfusion.