Intrathecally administered cGMP-dependent protein kinase Ialpha inhibitor significantly reduced the threshold for isoflurane anesthesia: implication for a novel role of cGMP-dependent protein kinase Ialpha

Sensitivity to Sevoflurane anesthesia is decreased in mice with a congenital deletion of Guanylyl Cyclase-1 alpha

BACKGROUND

Volatile anesthetics increase levels of the neurotransmitter nitric oxide (NO) and the secondary messenger molecule cyclic guanosine monophosphate (cGMP) in the brain. NO activates the enzyme guanylyl cyclase (GC) to produce cGMP. We hypothesized that the NO-GC-cGMP pathway contributes to anesthesia-induced unconsciousness.

METHODS

Sevoflurane-induced loss and return of righting reflex (LORR and RORR, respectively) were studied in wild-type mice (WT) and in mice congenitally deficient in the GC-1α subunit (GC-1^-/- mice). Spatial distributions of GC-1α and the GC-2α subunit in the brain were visualized by in situ hybridization. Brain cGMP levels were measured in WT and GC-1^-/- mice after inhaling oxygen with or without 1.2% sevoflurane for 20 min.

RESULTS

Higher concentrations of sevoflurane were required to induce LORR in GC-1^-/- mice than in WT mice (1.5 ± 0.1 vs. 1.1 ± 0.2%, respectively, n = 14 and 14, P < 0.0001). Similarly, RORR occurred at higher concentrations of sevoflurane in GC-1^-/- mice than in WT mice (1.0 ± 0.1 vs. 0.8 ± 0.1%, respectively, n = 14 and 14, P < 0.0001). Abundant GC-1α and GC-2α mRNA expression was detected in the cerebral cortex, medial habenula, hippocampus, and cerebellum. Inhaling 1.2% sevoflurane for 20 min increased cGMP levels in the brains of WT mice from 2.6 ± 2.0 to 5.5 ± 3.7 pmol/mg protein (n = 13 and 10, respectively, P = 0.0355) but not in GC-1^-/- mice.

CONCLUSION

Congenital deficiency of GC-1α abolished the ability of sevoflurane anesthesia to increase cGMP levels in the whole brain, and increased the concentration of sevoflurane required to induce LORR. Impaired NO-cGMP signaling raises the threshold for producing sevoflurane-induced unconsciousness in mice.

Oxidant-induced activation of cGMP-dependent protein kinase Iα mediates neuropathic pain after peripheral nerve injury

AIMS

Emerging lines of evidence indicate that oxidants such as hydrogen peroxide exert specific signaling functions during the processing of chronic pain. However, the mechanisms by which oxidants regulate pain processing in vivo remain poorly understood. Here, we investigated whether cyclic guanosine monophosphate (cGMP)-dependent protein kinase Iα (cGKIα), which can be activated by oxidants independently of cGMP, serves as a primary redox target during pain processing.

RESULTS

After peripheral nerve injury, oxidant-induced cGKIα activation is increased in dorsal root ganglia of mice. Knock-in (KI) mice in which cGKIα cannot transduce oxidant signals demonstrated reduced neuropathic pain behaviors after peripheral nerve injury, and reduced pain behaviors after intrathecal delivery of oxidants. In contrast, acute nociceptive, inflammatory, and cGMP-induced pain behaviors were not impaired in these mice.

INNOVATION

Studying cGKIα KI mice, we provide the first evidence that oxidants activate cGKIα in sensory neurons after peripheral nerve injury in vivo.

CONCLUSION

Our results suggest that oxidant-induced activation of cGKIα specifically contributes to neuropathic pain processing, and that prevention of cGKIα redox activation could be a potential novel strategy to manage neuropathic pain.

Putative role of nitric oxide synthase isoforms in the changes of nitric oxide concentration in rat brain cortex and cerebellum following sevoflurane and isoflurane anaesthesia

We have previously observed an increase in nitric oxide (NO) content in rat brain cortex following halothane, sevoflurane or isoflurane anaesthesia. This study was undertaken in order to determine whether isoform-specific nitric oxide synthase (NOS) inhibitors and inducers could modify these increases in NO contents. Rats were subjected to isoflurane and sevoflurane anaesthesia with concomitant administration of neuronal nitric oxide synthase (nNOS) inhibitor 7-Nitro-indazole (7-NI), inducible nitric oxide synthase (iNOS) inhibitor 2-amino-5,6-dihydro-6-methyl-4H-1,3-thiazine (AMT) or lipopolysaccharide. NO concentration in different organs was measured by electron paramagnetic resonance (EPR) spectroscopy. 7-NI significantly decreased NO concentration in cerebellum but not in brain cortex, whereas AMT decreased NO in all the organs studied. Anaesthesia significantly increased NO concentration in brain cortex and decreased that in cerebellum. AMT abolished the NO increase in brain cortex. Anaesthesia enhanced the drastic increase in NO concentration in brain cortex after intraventricular lipopolysaccharide administration. Isoflurane was found to inhibit recombinant nNOS and iNOS activities at high concentrations (EC50=20 mM). Our data suggest a putative role for iNOS in the increase in NO levels produced by isoflurane and sevoflurane, whereas nNOS activity is probably inhibited during anaesthesia.

Phosphorylation of septin 3 on Ser-91 by cGMP-dependent protein kinase-I in nerve terminals

The septins are a family of GTPase enzymes required for cytokinesis and play a role in exocytosis. Among the ten vertebrate septins, Sept5 (CDCrel-1) and Sept3 (G-septin) are primarily concentrated in the brain, wherein Sept3 is a substrate for PKG-I (cGMP-dependent protein kinase-I) in nerve terminals. There are two motifs for potential PKG-I phosphorylation in Sept3, Thr-55 and Ser-91, but phosphoamino acid analysis revealed that the primary site is a serine. Derivatization of phosphoserine to S-propylcysteine followed by N-terminal sequence analysis revealed Ser-91 as a major phosphorylation site. Tandem MS revealed a single phosphorylation site at Ser-91. Substitution of Ser-91 with Ala in a synthetic peptide abolished phosphorylation. Mutation of Ser-91 to Ala in recombinant Sept3 also abolished PKG phosphorylation, confirming that Ser-91 is the major site in vitro. Antibodies raised against a peptide containing phospho-Ser-91 detected phospho-Sept3 only in the cytosol of nerve terminals, whereas Sept3 was located in a peripheral membrane extract. Therefore Sept3 is phosphorylated on Ser-91 in nerve terminals and its phosphorylation may contribute to the regulation of its subcellular localization in neurons.

Effect of knock down of spinal cord PSD-93/chapsin-110 on persistent pain induced by complete Freund's adjuvant and peripheral nerve injury

PSD-93/chapsin-110 is a neuronal PDZ domain-containing protein that binds to and clusters the N-methyl-D-aspartate receptor (NMDAR) at synapses in the central nervous system. It also assembles a specific set of signaling proteins around the NMDAR and mediates downstream signaling by the NMDAR. Thus, PSD-93/chapsin-110 might be involved in many physiological and pathophysiological actions triggered via the activation of the NMDAR. In the current study, we report that abundant PSD-93/chapsin-110 protein was detected in rat spinal cord, particularly in the superficial dorsal horn. The rats injected intrathecally with PSD-93/chapsin-110 antisense oligodeoxynucleotide every 24 h for 4 days displayed not only a remarkable decrease in spinal cord PSD-93/chapsin-110 expression but also a significant reduction in the paw withdrawal responses to thermal and mechanical stimuli during complete Freund's adjuvant-induced inflammatory pain and peripheral nerve injury-induced neuropathic pain. In contrast, the rats injected intrathecally with PSD-93/chapsin-110 missense oligodeoxynucleotide did not exhibit these changes. We also found that pretreatment with PSD-93/chapsin-110 antisense oligodeoxynucleotide did not change the locomotor activity or the responses to acute noxious thermal and mechanical stimuli in intact rats. The present results indicate that the deficiency of spinal cord PSD-93/chapsin-110 protein significantly attenuates thermal and mechanical hyperalgesia in complete Freund's adjuvant- or peripheral nerve injury-induced chronic pain. This suggests that spinal cord PSD-93/chapsin-110 might be involved in the central mechanism of chronic pain. Our work might provide a new target for the therapy of chronic pain.

Inhibition of cyclic guanosine 5'-monophosphate-dependent protein kinase I (PKG-I) in lumbar spinal cord reduces formalin-induced hyperalgesia and PKG upregulation

Nitric oxide-mediated nociception has been suggested to involve formation of cyclic guanosine 5'-monophosphate (cGMP) and activation of cGMP-dependent protein kinase (PKG). To further evaluate this pathway we assessed the effects of the PKG-inhibiting cGMP analog Rp-8-Br-cGMPS in the rat formalin assay and analyzed the regulation of PKG expression in rat lumbar spinal cord. Spinally delivered Rp-8-Br-cGMPS (0.1-0.5 micro mol i.t.) reduced the nociceptive behavior in a dose-dependent manner. Similar effects were achieved with Rp-8-Br-PET-cGMPS (0.5 micro mol i.t.), another PKG-inhibitory cGMP analog. In contrast, Rp-8-Br-cAMPS (0.5 micro mol i.t.), an inhibitor of protein kinase A, had no effect in this model. Formalin treatment resulted in a rapid (within 1h), long-lasting (up to 96h) upregulation of PKG-I protein expression. This increase was prevented in animals pretreated with Rp-8-Br-cGMPS (0.5 micro mol i.t.) or morphine (2.5-5mg/kg i.p.) 10min prior to formalin injection. Spinal delivery of 8-Br-cGMP, a PKG-activating cGMP analog, without subsequent formalin treatment also caused an increase of PKG-I protein expression. Hence, the upregulation of PKG-I might possibly be mediated by cGMP itself. Our data suggest that PKG-I activation is involved in the synaptic transmission of nociceptive stimuli in the spinal cord and that PKG-I inhibitors might be interesting novel drugs for pain treatment.

Increased synthesis of nitric oxide in rat brain cortex due to halogenated volatile anesthetics confirmed by EPR spectroscopy

BACKGROUND

Halogenated volatile anesthetics (HVAs) are considered to be inhibitors of nitric oxide synthase (NOS). On other hand, NO mediates the vasodilation produced by HVAs. Thus, both increase and decrease of NO concentration in brain tissues are possible during anesthesia. Previously, we have observed an increase of NO content in rat brain cortex under halothane anesthesia. The goal of this study was to determine whether the observed phenomenon was general for this anesthetic group, if it was specific for brain cortex, and if the NO increase was due changes in NOS activity.

METHODS

NO scavengers were injected to adult rats 30 min prior to anesthesia. Rats were anesthetized by inhalation of an O2 mixture with volatile anesthetics (1.5% for halothane; 1% for isoflurane, 2% for sevoflurane). After 30 min of anesthesia, rats were decapitated and brain cortex, cerebellum, liver, heart, kidneys and testes were dissected, frozen in liquid nitrogen and subjected to EPR spectroscopy. Nitric oxide content was determined quantitatively based on the intensity of the NO-Fe-DETC complex spectrum and its comparison with the calibration curve.

RESULTS

In rats anesthetized with HVAs, we observed a greater than twofold increase of NO content in brain cortex as compared to the nonanesthetized animals. No significant changes were detected in other organs. The NOS inhibitor N(omega)-nitro-L-arginine abolished the increase of NO content in brain produced by volatile anesthetics.

CONCLUSION

The action of volatile anesthetics is coupled with an increase of NO content in the cortex dependent on NOS activity.

Spinal cord nitric oxide synthase and heme oxygenase limit morphine induced analgesia

Spinal cord tissue contains two enzyme systems capable of producing monoxide gases which in turn are linked to the stimulation of soluble guanylate cyclase, nitric oxide synthase (NOS) which produces NO and heme oxygenase (HO) which produces CO. Reports from several laboratories link these two enzyme systems to pain of inflammatory and neuropathic etiologies. Additional studies have demonstrated that the activation of the NOS system by morphine limits the spinal analgesic action of this drug. In this study we first employed the hot plate model of pain to demonstrate that the NOS inhibitor L-NAME and the HO inhibitor Sn-P potentiate the analgesic actions of intrathecally administered morphine while having no intrinsic analgesic action at the doses used. We then determined that L-NAME loses its ability to potentiate morphine in nNOS null-mutant mice, while Sn-P no longer potentiates morphine in mice lacking a functional HO-2 gene. The intrathecal injection of the cGMP analog 8-Br cGMP caused hyperalgesia in the hot plate assay. Focusing on the possible involvement of cGMP metabolism, we documented that morphine stimulates cGMP production in a spinal cord slice model in a concentration dependent and naloxone reversible manner. Both L-NAME and Sn-P were potent inhibitors of morphine-stimulated cGMP production. Buffer containing either CO or the NO donor compound SNAP stimulated cGMP production as well. In spinal cord slices from either nNOS or HO-2 null-mutant animals morphine did not stimulate cGMP production. Taken together our data suggest that spinal monoxide generation modifies the acute analgesic actions of morphine.

Release of glutamate, nitric oxide and prostaglandin E2 and metabolic activity in the spinal cord of rats following peripheral nociceptive stimulation

Peripheral tissue injury and inflammation may result in a facilitated spinal nociceptive transmission and central sensitization. Particularly, nitric oxide (NO) and prostaglandins (PGs) have been shown to be key mediators involved in the induction and maintenance of this state. By means of spinal cord microdialysis we have determined interstitial glutamate, NO (NO2-/NO3-), PGE2, glycerol, glucose and lactate concentrations in the dorsal horns of the spinal cord following peripheral nociceptive stimulation to gain further insight into the link between excitatory neurotransmitters and metabolic functions in the spinal cord during nociception. Formalin and zymosan injection into one hind paw evoked a biphasic release of glutamate and NO with the glutamate peaks preceding those of NO. Moreover, zymosan induced a biphasic increase of interstitial glycerol concentrations accompanied by an increase of interstitial lactate indicating metabolic disturbances. In contrast, formalin injection led to an elevation of dialysate glucose concentrations which may be interpreted as an indication of enhanced metabolic activity. The sequential release of glutamate and NO in the dorsal horns of the spinal cord in response to peripheral nociceptive stimulation supports the theory that NO may act as a retrograde transmitter. The metabolic changes observed after formalin and zymosan injection suggest that an intense peripheral nociceptive stimulation may not only activate but also disturb metabolic activity and possibly membrane integrity in the spinal cord.