In situ measurement of neuronal nitric oxide synthase activity in the spinal cord by NADPH-diaphorase histochemistry

Acute spatial spread of NO-mediated potentiation during hindpaw ischaemia in mice

Key points

Neuropathic pain spreads spatially beyond the injured sites, and the mechanism underlying the spread has been attributed to inflammation occurring in the spinal cord.

However, the spatial spread of spinal/cortical potentiation induced by conduction block of the peripheral nerves can be observed prior to inflammation.

In the present study, we found that spreading potentiation and hypersensitivity acutely induced by unilateral hindpaw ischaemia are nitric oxide (NO)‐dependent and that NO is produced by ischaemia and quickly diffuses within the spinal cord.

We also found that NO production induced by ischaemia is not observed in the presence of an antagonist for group II metabotropic glutamate receptors (mGluRs) and that neuronal NO synthase‐positive dorsal horn neurons express group II mGluRs.

These results suggest strongly that NO‐mediated spreading potentiation in the spinal cord is one of the trigger mechanisms for neuropathic pain.

Abstract

Cortical/spinal responses to hindpaw stimulation are bilaterally potentiated by unilateral hindpaw ischaemia in mice. We tested the hypothesis that hindpaw ischaemia produces nitric oxide (NO), which diffuses in the spinal cord to induce spatially spreading potentiation. Using flavoprotein fluorescence imaging, we confirmed that the spreading potentiation in hindpaw responses was induced during ischaemia in the non‐stimulated hindpaw. This spreading potentiation was blocked by spinal application of l‐NAME, an inhibitor of NO synthase (NOS). Furthermore, no spreading potentiation was observed in neural NOS (nNOS) knockout mice. Spinal application of an NO donor was enough to induce cortical potentiation and mechanical hypersensitivity. The spatial distribution of NO during unilateral hindpaw ischaemia was visualized using 4‐amino‐5‐methylamino‐2′,7′‐difluorofluorescein (DAF‐FM). An increase in fluorescence derived from the complex of DAF‐FM with NO was observed on the ischaemic side of the spinal cord. A similar but smaller increase was also observed on the contralateral side. Somatosensory potentiation after hindpaw ischaemia is known to be inhibited by spinal application of LY354740, an agonist of group II metabotropic glutamate receptors (mGluRs). We confirmed that the spinal DAF‐FM fluorescence increases during hindpaw ischaemia were not observed in the presence of LY354740. We also confirmed that approximately half of the nNOS‐positive neurons in the superficial laminae of the dorsal horn expressed mGluR2 mRNA. These results suggest that disinhibition of mGluR2 produces NO which in turn induces a spreading potentiation in a wide area of the spinal cord. Such spreading, along with the consequent non‐specific potentiation in the spinal cord, may trigger neuropathic pain.

Neuropathic pain spreads spatially beyond the injured sites, and the mechanism underlying the spread has been attributed to inflammation occurring in the spinal cord.

However, the spatial spread of spinal/cortical potentiation induced by conduction block of the peripheral nerves can be observed prior to inflammation.

In the present study, we found that spreading potentiation and hypersensitivity acutely induced by unilateral hindpaw ischaemia are nitric oxide (NO)‐dependent and that NO is produced by ischaemia and quickly diffuses within the spinal cord.

We also found that NO production induced by ischaemia is not observed in the presence of an antagonist for group II metabotropic glutamate receptors (mGluRs) and that neuronal NO synthase‐positive dorsal horn neurons express group II mGluRs.

These results suggest strongly that NO‐mediated spreading potentiation in the spinal cord is one of the trigger mechanisms for neuropathic pain.

Endogenous Nitric Oxide Inhibits, Whereas Awakening Stimuli Increase, the Activity of a Subset of Orexin Neurons

The lateral hypothalamic area contains neurons expressing neuronal nitric oxide synthase (nNOS), in addition to orexin neurons. Here we examined whether the activity of orexin neurons was regulated by endogenous nitric oxide (NO) in male C57BL/6 mice. Caffeine (30 mg/kg, intraperitoneally (i.p.)) increased the number of orexin neurons positive for c-Fos, a marker of neuronal activity, and also increased the number of NOS/c-Fos-positive cells as identified by reduced nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase histochemistry and c-Fos immunohistochemistry. Diphenhydramine hydrochloride (10 mg/kg. i.p.) decreased c-Fos-positive orexin neurons but had no significant effect on the number of c-Fos-positive NOS neurons. nNOS inhibitor 7-nitroindazole (25 mg/kg, i.p.) alone increased c-Fos-positive orexin neurons, and combined treatment with caffeine and 7-nitroindazole did not show additive effect in the number of c-Fos-positive orexin neurons. In contrast, 7-nitroindazole decreased c-Fos-positive NOS neurons and attenuated caffeine-induced increase in c-Fos-positive NOS neurons. Sleep deprivation increased c-Fos-positive cells in both orexin neurons and NOS neurons, and 7-nitroindazole did not show additive effect with sleep deprivation in the activation of orexin neurons. Together, these results suggest that endogenous NO negatively regulates the activity of a subset of orexin neurons, and this subset of orexin neurons overlaps with that activated by awakening stimuli.

Prenatal effects of retinoic acid on lumbar spinal cord development and liver antioxidants in rats

During embryonic and early postnatal development, retinoic acid (RA) regulates genes that control neuronal differentiation and neurite outgrowth from the neural tube. The effects of high levels of RA on the CNS can be detected via nitric oxide (NO), which plays a crucial role in neural transmission. The aim of the study was to investigate the prenatal influence of high levels of RA on postnatal development of nitrergic structures in lumbar spinal cord and antioxidant status. RA was administered orally at a dose of 10mg/kg body weight to pregnant female Wistar rats during days 8-10 of gestation. Neuronal nitric oxide synthase (nNOS) of lumbar spinal cord sections was processed for visualization via nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) histochemistry on postnatal day one, day twenty-one and in adults. The results suggest that prenatal administration of high levels of RA is not associated with postnatal morphological changes in nNOS-positive neurons in the rat lumbar spinal cord. An estimation of the activity of enzymes related to the storage of retinoid in the liver showed possible side effects. Suppression and deepening of superoxide dismutase activity persisted into adulthood, and a concurrent downregulation of glutathione reductase was noted. A decrease in reduced glutathione persisted until adulthood when other compensatory mechanisms were probably active to maintain an appropriate level.

Nitric oxide mediates selective degeneration of hypothalamic orexin neurons through dysfunction of protein disulfide isomerase

We addressed the role of nitric oxide (NO) in orexin neuron degeneration that has been observed under various pathological conditions. Administration of an NO donor NOC18 (50 nmol) into the third ventricle of mice resulted in a significant decrease of orexin-immunoreactive (-IR) neurons, in contrast to a modest change in melanin-concentrating hormone-IR neurons. In addition, NOC18 promoted formation of orexin-A-IR aggregates within orexin neurons. An endoplasmic reticulum stress inducer tunicamycin replicated the effect of NOC18 with regard to decrease of orexin-IR neurons and formation of aggregates. We also found that NOC18 caused an increase in S-nitrosation of protein disulfide isomerase (PDI) and a decrease in PDI activity in hypothalamic tissues. Moreover, PDI inhibitors, such as cystamine and securinine, caused a selective decrease of orexin neurons and promoted formation of orexin-A-IR aggregates. Aggregate formation in orexin-IR neurons was also induced by local injection of small interfering RNA targeting PDI. Interestingly, sleep deprivation for 7 consecutive days induced a selective decrease of orexin-IR neurons, which was preceded by aggregate formation in orexin-IR neurons and an increase in S-nitrosated PDI in the hypothalamus. Activity of neuronal NO synthase (nNOS)-positive neurons in the lateral hypothalamus as assessed by c-Fos expression was elevated in response to sleep deprivation. Finally, sleep deprivation-induced decrease of orexin-IR neurons, formation of aggregates, and S-nitrosation of PDI were not observed in nNOS knock-out mice. These results indicate that nNOS-derived NO may mediate specific pathological events in orexin neurons, including neuropeptide misfolding via S-nitrosation and inactivation of PDI.

Rapid S-nitrosylation of actin by NO-generating donors and in inflammatory pain model mice

Background

S-Nitrosylation, the reversible post-translational modification of reactive cysteine residues in proteins, has emerged as an important mechanism by which NO acts as a signaling molecule. We recently demonstrated that actin is a major S-nitrosylated protein in the spinal cord and suggested that NO directly attenuates dopamine release from PC12 cells by causing the breakdown of F-actin. However, the occurrence of S-nitrosylation of actin remained unclarified in animal pain model. Kinetic analysis of S-nitrosylation of actin in the present study was made by using NO-generating donors. The biotin-switch assay and purification on streptavidin-agarose were employed for identification of S-nitrosylated actin.

Results

Dopamine release from PC12 cells was markedly attenuated by NOR1 (t_1/2 = 1.8 min) and much less by NOR3 (t_1/2 = 30 min), but not by S-nitroso-glutathione, an endogenous NO donor. A membrane-permeable cGMP analogue could not substitute for NOR1 as a suppressor nor could inhibitors of soluble guanylate cyclase and cGMP-dependent protein kinase attenuate the suppression. S-Nitrosylated actin was detected by the biotin-switch assay at 5 min after the addition of NOR1. Consistent with the kinetic analysis, actin in the spinal cord was rapidly and maximally S-nitrosylated in an inflammatory pain model at 5 min after the injection of 2% formalin into the hind paws. In vivo patch-clamp recordings of the spinal dorsal horn, NOR3 showed an inhibitory action on inhibitory synaptic transmission in interneurons of the substantia gelatinosa.

Conclusions

The present study demonstrates that rapid S-nitrosylation of actin occurred in vitro in the presence of exogenous NO-generating donors and in vivo in inflammatory pain model mice. Our data suggest that, in addition to the well-known cGMP-dependent protein kinase pathway, S-nitrosylation is involved in pain transmission via disinhibition of inhibitory neurons.

Involvement of S-nitrosylation of actin in inhibition of neurotransmitter release by nitric oxide

BACKGROUND

The role of the diffusible messenger nitric oxide (NO) in the regulation of pain transmission is still a debate of matter, pro-nociceptive and/or anti-nociceptive. S-Nitrosylation, the reversible post-translational modification of selective cysteine residues in proteins, has emerged as an important mechanism by which NO acts as a signaling molecule. The occurrence of S-nitrosylation in the spinal cord and its targets that may modulate pain transmission remain unclarified. The "biotin-switch" method and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry were employed for identifying S-nitrosylated proteins.

RESULTS

Here we show that actin was a major protein S-nitrosylated in the spinal cord by the NO donor, S-nitroso-N-acetyl-DL-penicillamine (SNAP). Interestingly, actin was S-nitrosylated, more in the S2 fraction than in the P2 fraction of the spinal homogenate. Treatment of PC12 cells with SNAP caused rapid S-nitrosylation of actin and inhibited dopamine release from the cells. Just like cytochalasin B, which depolymerizes actin, SNAP decreased the amount of filamentous actin cytoskeleton just beneath the membrane. The inhibition of dopamine release was not attenuated by inhibitors of soluble guanylyl cyclase and cGMP-dependent protein kinase.

CONCLUSION

The present study demonstrates that actin is a major S-nitrosylated protein in the spinal cord and suggests that NO directly regulates neurotransmitter release by S-nitrosylation in addition to the well-known phosphorylation by cGMP-dependent protein kinase.

Nitric oxide (NO) serves as a retrograde messenger to activate neuronal NO synthase in the spinal cord via NMDA receptors

We have recently demonstrated that nitric oxide (NO) produced by neuronal NO synthase (nNOS) in the spinal cord is involved in the maintenance of neuropathic pain. To clarify whether NO itself affected nNOS activity in the spinal cord as a retrograde messenger, we examined the involvement of the NO/cGMP signaling pathway in the regulation of nNOS activity by NADPH-diaphorase histochemistry. NO-generating agents NOR3 (t(1/2)=30min) and SNAP (t(1/2)=5h), but not NOR1 (t(1/2)=1.8min), significantly enhanced NADPH-diaphorase staining in the spinal cord. 8-Br-cGMP also enhanced it similar to that by NOR3, and 8-Br-cAMP and forskolin, an activator of adenylate cyclase, enhanced it moderately. NOR1 and NOR3 markedly increased the cGMP level in the spinal cord. The enhancement of NADPH-diaphorase staining by NOR3 was significantly inhibited by CPTIO, an NO scavenger, ODQ, a soluble guanylate cyclase inhibitor, and KT5823, an inhibitor of cGMP-dependent protein kinase. Additionally, the NOR3-enhanced nNOS activity was completely inhibited by NMDA antagonists MK-801 and d-AP5, partially by the GluRepsilon2-selective antagonist CP-101,606, and was attenuated in GluRepsilon1(-/-) and GluRepsilon1(-/-)/epsilon4(-/-) mice. These results suggest that NO may regulate nNOS activity as a retrograde messenger in the spinal cord via activation of NMDA receptor containing GluRepsilon1 and GluRepsilon2 subunits.

Signal pathways coupled to activation of neuronal nitric oxide synthase in the spinal cord by nociceptin/orphanin FQ

Nociceptin/orphanin FQ (N/OFQ) was earlier shown to be involved in the maintenance of neuropathic pain by activating neuronal nitric oxide synthase (nNOS). We recently established an ex vivo system to elucidate biochemical and molecular mechanisms for nNOS activation by the use of a combination of isolated intact spinal cord preparations and NADPH-diaphorase histochemistry. Here we examined the N/OFQ signal pathways coupled to nNOS activation in the spinal cord by using this ex vivo system. N/OFQ enhanced nNOS activity in the superficial layer of the spinal cord, as assessed by NADPH-diaphorase histochemistry, in a time- and dose-dependent manner. The maximum effect was observed at 3-10 nM. The N/OFQ-stimulated nNOS activity was inhibited by NMDA receptor antagonists MK-801 and D-AP5, but not by the NR2B-selective antagonist CP-101,606; and the stimulated activity was observed in NR2D(-/-) mice, but not in NR2A(-/-) or NR2A(-/-)/NR2D(-/-) mice. N/OFQ receptor antagonists attenuated the nNOS activity stimulated by N/OFQ, but not that by NMDA. Furthermore, the potentiation of nNOS by N/OFQ was inhibited by calphostin C and Ro 31-8220, PP2, and KN-62, but not by H-89. These results suggest that N/OFQ stimulated nNOS activity by a biochemical cascade initiated by activation of NMDA receptors containing NR2A.

Peripheral antinociceptive effect of pertussis toxin: activation of the arginine/NO/cGMP/PKG/ ATP-sensitive K channel pathway

The aim of the present study was to determine the effect of pertussis toxin (PTX) on inflammatory hypernociception measured by the rat paw pressure test and to elucidate the mechanism involved in this effect. In this test, prostaglandin E(2) (PGE(2)) administered subcutaneously induces hypernociception via a mechanism associated with neuronal cAMP increase. Local intraplantar pre-treatment (30 min before), and post-treatment (5 min after) with PTX (600 ng/paw1, in 100 microL) reduced hypernociception induced by prostaglandin E(2) (100 ng/paw, in 100 microL, intraplantar). Furthermore, local intraplantar pre-treatment (30 min before) with PTX (600 ng/paw, in 100 microL) reduced hypernociception induced by DbcAMP, a stable analogue of cAMP (100 microg/paw, in 100 microL, intraplantar), which indicates that PTX may have an effect other than just G(i)/G(0) inhibition. PTX-induced analgesia was blocked by selective inhibitors of nitric oxide synthase (L-NMMA), guanylyl cyclase (ODQ), protein kinase G (KT5823) and ATP-sensitive K(+) channel (Kir6) blockers (glybenclamide and tolbutamide). In addition, PTX was shown to induce nitric oxide (NO) production in cultured neurons of the dorsal root ganglia. In conclusion, this study shows a peripheral antinociceptive effect of pertussis toxin, resulting from the activation of the arginine/NO/cGMP/PKG/ATP-sensitive K(+) channel pathway.

Ubiquitous presence of argininosuccinate at millimolar levels in the central nervous system of Aplysia californica

Endogenous nitric oxide (NO) is generated by nitric oxide synthases (NOSs), which convert arginine (Arg) and oxygen to citrulline (Cit) and NO. Cit can be enzymatically transformed back to Arg by argininosuccinate synthetase (ASS) and argininosuccinate lyase (ASL) via a pathway involving argininosuccinate (ArgSuc). Arg, Cit, and ArgSuc levels have been measured in single neurons, neuronal clusters, and neuropil from the nervous system of the common neurobiological model Aplysia californica. Using capillary electrophoresis with laser-induced fluorescence detection, ArgSuc was found to be present in the nervous system in millimolar concentrations at levels significantly exceeding Cit levels (p<0.01). ArgSuc levels are proportional to Arg concentrations in single neurons, whereas they have no clear correlation to the Cit or Arg/Cit ratio. NOS-expressing neurons often exhibit fixative-resistant nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) staining. Incubation of ganglia with Arg results in an increase in Cit and ArgSuc levels in the NADPH-d-positive neuropil with no effect on ArgSuc levels in NADPH-d-negative neurons, suggesting NOS activity in the neuropil. Similar incubation with Cit leads to decreased ArgSuc levels in NADPH-d-negative neurons. These results can be explained by localization of NOS and ASS in different neurons; therefore, the complete Arg-Cit-NO cycle may not be present in the same neuron. The surprisingly high intracellular ArgSuc concentration suggests alternative sources of ArgSuc and that at least a portion may be formed by the reverse reaction of ASL (catalyzing the conversion of Arg to ArgSuc), which can be inhibited by Cit.