Blocking a vicious cycle nNOS/peroxynitrite/AMPK by S-nitrosoglutathione: implication for stroke therapy

Abstract

Background: Stroke immediately sets into motion sustained excitotoxicity and calcium dysregulation, causing aberrant activity in neuronal nitric oxide synthase (nNOS) and an imbalance in the levels of nitric oxide (NO). Drugs targeting nNOS-originated toxicity may therefore reduce stroke-induced damage. Recently, we observed that a redox-modulating agent of the NO metabolome, S-nitrosoglutathione (GSNO), confers neurovascular protection by reducing the levels of peroxynitrite, a product of aberrant NOS activity. We therefore investigated whether GSNO-mediated neuroprotection and improved neurological functions depend on blocking nNOS/peroxynitrite-associated injurious mechanisms using a rat model of cerebral ischemia reperfusion (IR).

Results: IR increased the activity of nNOS, the levels of neuronal peroxynitrite and phosphorylation at Ser(1412) of nNOS. GSNO treatment of IR animals decreased IR-activated nNOS activity and neuronal peroxynitrite levels by reducing nNOS phosphorylation at Ser(1412). The Ser(1412) phosphorylation is associated with increased nNOS activity. Supporting the notion that nNOS activity and peroxynitrite are deleterious following IR, inhibition of nNOS by its inhibitor 7-nitroindazole or reducing peroxynitrite by its scavenger FeTPPS decreased IR injury. GSNO also decreased the activation of AMP Kinase (AMPK) and its upstream kinase LKB1, both of which were activated in IR brain. AMPK has been implicated in nNOS activation via Ser(1412) phosphorylation. To determine whether AMPK activation is deleterious in the acute phase of IR, we treated animals after IR with AICAR (an AMPK activator) and compound c (an AMPK inhibitor). While AICAR potentiated, compound c reduced the IR injury.

Conclusions: Taken together, these results indicate an injurious nNOS/peroxynitrite/AMPK cycle following stroke, and GSNO treatment of IR inhibits this vicious cycle, resulting in neuroprotection and improved neurological function. GSNO is a natural component of the human body, and its exogenous administration to humans is not associated with any known side effects. Currently, the FDA-approved thrombolytic therapy suffers from a lack of neuronal protective activity. Because GSNO provides neuroprotection by ameliorating stroke's initial and causative injuries, it is a candidate of translational value for stroke therapy.

Figure 1

Effect of GSNO on the activity of nNOS and the levels of NO and peroxynitrite (3-NT) in ipsilateral (penumbra) area of the IR brain. nNOS activity at 1 h (a), NO levels at 1 h (b) and peroxynitrite levels (3-NT, IHC [c] and IHC colocalization of 3-NT and neuronal marker NeuN [d]) at 1 h after reperfusion were determined. GSNO were administered 0 h after reperfusion began. NO levels in Sham animals were 39.8 ± 4.2 nmol/mg protein. Data are presented as mean ± SD (n = 5). ***p < 0.001 vs. Sham, ^###p < 0.001, ^#p < 0.05 vs. IR, ^$$p < 0.01, ^$p < 0.05 vs. Sham.

Figure 2

Effect of GSNO on neuronal degeneration and axonal loss in ipsilateral (penumbra) area of IR brain at 4 h after reperfusion. Nissl staining a shows significant number of degenerating neurons (black arrow) in IR compared with GSNO and sham groups (intact neurons are shown by blue arrows). Bielshowsky silver and b H&E stainings c show loss of axonal and tissue integrity (red arrow), respectively, in IR compared with GSNO and sham. IR also shows pyknosis of neurons (black arrow). Photomicrographs are representative of n = 3 in each group.

Figure 3

Effect of GSNO on nNOS phosphorylation status in ipsilateral (penumbra) area of IR brain. Phosphorylation at Ser^{1417 [equivalent to1412 in human]} of nNOS and its expression (a) and their densitometry (b) were evaluated at 1 h of reperfusion. Data are presented as mean ± SD (n = 5). ***p < 0.001 vs. Sham and GSNO.

Figure 4

Effect of nNOS inhibitor 7-NI and GSNO on brain infarctions and neurological score. Representative TTC stained sections (#3 and 4 out of six consecutive sections from cranial to caudate region) (a), infarct area (b), infarct volume (c) and neurological score (d) at 24 h of the reperfusion after 60 min middle cerebral artery occlusion. Both 7-NI and GSNO were administered 1 h after reperfusion began. Data are presented as mean ± SD (n = 5). ***p < 0.001, **p < 0.01 vs. IR, ⁺⁺⁺p < 0.001 vs. 7-NI.

Figure 5

Effect of GSNO on LKB1, ACC and AMPK phosphorylation status in ipsilateral (penumbra) area of IR brain. Western blot at 1 h of reperfusion showing phosphorylation of pLKB1 (a, a′), pACC and pAMPK/pACC (b, b′). GSNO was administered at 0 h after reperfusion began. Data are presented as mean ± SD (n = 5). ***p < 0.001, **p < 0.01 vs. Sham and GSNO.

Figure 6

Effect of AMPK inhibitor Comp c, peroxynitrite scavenger FeTPPS and AMPK activator AICAR on infarct volume and neurological score. Infarct volume (a calculated from TTC stained sections) and neurological score (b) at 24 h of IR. While Comp c, FeTPPS and AICAR were administered 1 h, GSNO was administered 0 h after reperfusion began. Data are presented as mean ± SD (n = 5). ***p < 0.001 vs. IR, ^$$$p < 0.001 vs. AICAR, ⁺⁺⁺p < 0.001 vs. GSNO.

Figure 7

Schematic showing hypothesized nNOS/peroxynitrite-mediated deleterious events and GSNO-mediated targets. IR-induced excitotoxicity and calcium dysregulation cause aberrant nNOS activation, peroxynitrite formation and LKB1/AMPK activation. GSNO treatment of IR blocks the vicious cycle by inhibiting the aberrant activity of nNOS and LKB1/AMPK, leading to neuroprotection and improved neurobehavioral function.