Inhibition of neuronal nitric oxide synthase results in neurodegenerative changes in the axotomised dorsal root ganglion neurons: evidence for a neuroprotective role of nitric oxide in vivo

Neuroscience Research 40 (2001) 37 – 44 www.elsevier.com/locate/neures

Inhibition of neuronal nitric oxide synthase results in neurodegenerative changes in the axotomised dorsal root ganglion neurons: evidence for a neuroprotective role of nitric oxide in vivo T. Thippeswamy a,*, R.K. Jain b, Nazli Mumtaz c, R. Morris a a

Department of Veterinary Preclinical Sciences, Uni6ersity of Li6erpool, Veterinary Science Building, Crown Street, Li6erpool, L69 7ZJ, UK b Department of Anatomy and Histology, Haryana Agricultural Uni6ersity, College of Veterinary Science, Hisar, Haryana 125004, India c Department of Veterinary Anatomy, Uni6ersity of Kafkas, Kars, Turkey Received 11 October 2000; accepted 18 January 2001

Abstract In axotomised adult rat dorsal root ganglion (DRG), many neurons show a marked increase in expression of neuronal nitric oxide synthase (nNOS). It has been established that NO functions as a neuron-glial signalling molecule by generating cGMP in glia cells that surround the neuron in DRG. Furthermore, in cultures of dissociated DRG deprived of nerve growth factor, many neurons expressed nNOS and cGMP and subsequently died if either enzyme’s activity was inhibited suggesting that NO–cGMP pathway could be neuroprotective in stressed DRG neurons. This has now been tested in vivo. It was found, 10 days after sciatic axotomy that nNOS was expressed in 36% of DRG neurons in the L5 and L6 ganglia giving rise to the damaged nerve, compared with 6% in contralateral ganglia. Almost all nNOS neurons and 24% of non-nNOS neurons expressed c-Jun in their nuclei. Ten days following axotomy, treatment with the relatively selective nNOS-blocker, 1-(2-trifluoromethylphenyl) imidazole (TRIM), caused morphology changes in approximately 50% of neurons that consisted of vacuolations, blebbing and highly irregular cell boundaries. Sham operated, TRIM treated, nerve-sectioned, vehicle treated, and controls did not show these changes. These observations further support the view that NO could be neuroprotective in some injured/stressed primary sensory neurons. © 2001 Elsevier Science Ireland Ltd and the Japan Neuroscience Society. All rights reserved. Keywords: Nitric oxide synthase; DRG; Neuroprotection; Neurodegeneration; In vivo; NOS inhibition

1. Introduction NO is becoming increasingly important as a biological messenger molecule in recent years. The role of NO in mediation of cytoprotective or toxic process still remains controversial. NO is synthesised from one of the essential amino acids, L-arginine by the enzyme NO synthase (NOS). There are three isoforms; neuronal (nNOS) expressed in neurons, inducible (iNOS) in immune cells and endothelial (eNOS) in endothelium of blood vessels. Although there are no conventional receptors for NO, soluble isoform of guanylate cyclase (sGC) is considered to be its principal target, the binding of which generates cGMP (Garthwaite and Boul* Corresponding author. Tel: + 44-151-7944242; fax: + 44-1517944243. E-mail address: [email protected] (T. Thippeswamy).

ton, 1995). In acutely isolated dorsal root ganglion (DRG), application of NO-donor compounds causes a marked elevation of cGMP in the satellite glia cells surrounding DRG neurons (Morris et al., 1992). This elevation of cGMP in glia cells is also seen in vivo following peripheral nerve section (Shi et al., 1998). The nNOS mRNA and protein expression increased in nearly two third of the DRG neurons following section of their peripheral axons (Verge et al., 1992; FiallosEstrada et al., 1993; Zhang et al., 1993). In dissociated DRG cultures, absence of exogenous nerve growth factor (NGF) in the medium and serum-free medium causes increased nNOS expression in many neurons (Thippeswamy and Morris, 1997a). Subsequent inhibition of nNOS or sGC activity leads to death of both nNOS containing and non-nNOS neurons (Thippeswamy and Morris, 1997b). Very recently, Andoh et al. (2000) have also shown the induction of

0168-0102/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd and the Japan Neuroscience Society. All rights reserved. PII: S 0 1 6 8 - 0 1 0 2 ( 0 1 ) 0 0 2 0 5 - X

38

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44

nNOS in serum deprived human brain-derived SHSY5Y neurotrophic cells and the inhibition of nNOS or sGC caused apoptosis. These observations suggest that NO could be neuroprotective in stressed DRG neurons by signalling glia via cGMP. A more widespread neuroprotective role for NO in nervous system is beginning to emerge in recent years. NO serves as an important signal for establishing an ischemic-tolerant state in brain (Gidday et al., 1999; Gonzalez-Zulueta et al., 2000) and it modulates survival of neuronal PC12 cells by activating Ras proteins (Teng et al., 1999). Pantazis et al. (1998) have demonstrated the involvement of NO – cGMP pathway in promoting cerebellar granule cell survival in culture and protecting them against ethanol toxicity. However, high concentrations of NO may induce both necrotic and apoptotic cell death (Ho et al., 1999; Melino et al., 1999). Hence, the block of NOS has been considered as a therapeutic strategy in many neurodegenerative diseases. In view of this, we considered it important to establish whether the neuroprotective effect of NO, we have identified in vitro is of any importance in vivo. To examine this issue, the effects of a relatively selective nNOS blocker, 1-(2-trifluoromethylphenyl) imidazole (TRIM) (Handy et al., 1996) that acts systemically (Dzoljic et al., 1997; Handy and Moore, 1998; Towler et al., 1998) on the axotomised DRG neurons has been evaluated. The proto-oncogene, c-Jun, an inducible transcription factor has been implicated in control of neuronal responses to injury (Gu et al., 2000) and in axonal growth or regenerative processes (Kenney and Kocsis, 1998). We have used antibodies to c-Jun as a marker for axotomised/stressed neurons in our experiments.

2. Materials and methods

2.1. Surgery Twenty-two adult Wister albino rats of either sex weighing between 200 and 250 g were studied. All procedures were carried out in accordance with UK Home Office statutory regulations. All surgical procedures followed by animal recovery were performed under aseptic conditions and general anaesthesia. Anaesthesia was induced with nitrous oxide (60%) and halothane (5%), and maintained with Hypnorm (0.3 ml/kg i.m.) and diazepam (2.5 mg/kg i.p.). Post-operative analgesia employed using Buprenorphine (0.5 mg/ kg i.m.). Sixteen rats were subjected to unilateral sciatic nerve section on the left side. A segment of the trunk of the sciatic nerve (4– 5 mm) was removed at the level of mid-thigh. Another six rats were sham operated, in which all procedures except the nerve section were

carried out. All rats included in the study had no abnormal swelling of the hind limb or showed any autotomy.

2.2. Drug treatment On day 10, six axotomised and three sham-operated rats were injected intraperitonially (i.p.) with a specific inhibitor of nNOS, 1-(2-trifluoromethylphenyl) imidazole (TRIM; RBI, Cat No. T-175; dissolved in sterile distilled water 5 mg/ml; 50 mg/kg). Injections were repeated three times at 2 h intervals (0, 2 and 4 h). Two more axotomised rats were given two injections at 2 h interval (0 and 2 h) and another two axotomised rats were given single injection for 2 h. The effective drug concentration was chosen based on other researchers findings (Handy et al., 1996; Dzoljic et al., 1997; Handy and Moore, 1998; Towler et al., 1998). Another six axotomised and three sham-operated rats were injected with identical volumes of vehicle (sterile distilled water). To minimise distress during injections of these relatively large volumes (2–2.5 ml) the animals were briefly anaesthetised with Halothane (2%). The behaviour of the rats was monitored throughout the period of drug treatment. Two hours after the last injection animals were terminally anaesthetised with pentobarbitone (80 mg/kg i.p.), and fixed by vascular perfusion with 4% paraformaldehyde 0.1 M phosphate buffered saline (PBS). Two animals each at 2 and 4 h drug treatment were also perfused to check the effect of the drug on axotomised DRG at shorter durations. Six hours treatment regime was chosen based on our earlier results in vitro, where in the neurons showed abnormal morphology, 6 h following NOS inhibition (Thippeswamy and Morris, 1997b). Though a systematic motor activity assessment was not the primary objective of this experiment, animals were constantly observed for their gross behaviour. There was no abnormal behaviour or motor activity in TRIM treated rats except the expected abnormal gait during locomotion (foot not raised and placed properly but dragged) as a result of sciatic nerve section in both vehicle and TRIM treated animals.

2.3. Immunocytochemistry DRG from the fifth and sixth lumbar segment (L5 and L6), ipsilateral and contralateral to the nerve sectioned or sham operated were dissected and post-fixed in the same fixative for 4 h at 4°C. DRG were cryo-protected overnight with 25% sucrose in PBS. Sections were cut on a cryostat (10 mm thick) and thaw mounted onto a set of ten slides such that each slide consisted of every tenth section. They were then washed with PBS and incubated overnight with antibodies to nNOS raised in sheep (1:2000 supplied by Dr P.C. Emson) and c-Jun raised in rabbit (1:40; Santa Cruz). Earlier exper-

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44

39

Fig. 1. Photomicrographs of cryostat sections of L5 DRG: A–D are double immunostained for nNOS and c-Jun, E and F are cresyl violet stained. Red labelled neurons in A–D are nNOS positive and green/orange-green labelled nuclei are c-Jun positive. Scale bar: A – D, 20 mm; E and F, 50 mm. (A) Ipsilateral (to sciatic nerve section) L5 DRG from TRIM treated rat for 6 h (50 mg/kg, i.p; three doses at 2 h interval). There is more number of nNOS neurons and c-Jun positive neurons in this section compared with the sham operated (ipsilateral C, contralateral D). Note the abnormal morphology of both nNOS containing and non-nNOS neurons. Some of the nNOS containing neurons show irregular cytoplasmic boundaries, and blebbing (examples are indicated by arrows). (B) Ipsilateral (to sciatic nerve section) L5 DRG from vehicle treated animal. This is control for A. Though there are many nNOS and c-Jun positive neurons in this section compared with controls (C, D), the morphology of neurons is not disrupted like in part A. (C) and (D) Ipsilateral and contralateral (L5 DRG) to Sham operation in TRIM treated animal. Note that there are not many nNOS and c-Jun positive neurons compared with A and B although there is small induction of nNOS and c-Jun in C compared with D (D is control for C; C is control for A and B). Note that the morphology of neurons is not affected following TRIM treatment. (E) and (F) Cresyl violet stained sections of L5 DRG (ipsilateral to nerve section) from TRIM (E) and vehicle treated for 6 h (F). Arrows in E indicate the neurons with cytoplasmic vacuolations and irregular cell boundaries where as in F, neurons appear normal (except the one indicated by an arrow showing cytoplasmic vacuole) with smooth cell boundaries and clear eccentric nucleus with nucleolus, characteristic of axotomy.

40

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44

iments have shown that the nNOS antibodies are highly specific for this isoform of the enzyme (Thippeswamy and Morris, 1997a). However, primary antibody omission was routinely carried out as control. In addition, bacterial lipolysaccharide or cytokine-stimulated lymphocytes (which express iNOS) were used to check the specificity of nNOS antisera. After thorough washing, sections were treated with the appropriate anti-species antibodies (anti-sheep Cy3, 1:100; biotinylated anti-rabbit, 1:200) (both from Jackson) for 1 h at room temperature followed by streptavidin-FITC (1:80, Vector). Sections were washed several times in between each step with PBS and finally cover slipped using Vectashield mounting medium (Vector). Representative sections from drug treated and control were stained with cresyl violet for morphological analysis. To reveal whether cells undergo apoptosis following treatment with TRIM for 2, 4, and 6 h, the TUNEL method (Gavrieli et al., 1992; with little modifications) was employed. Sections from rat colon with and without DNase I treatment were used as positive control. Each slide represented a random sample of the whole ganglion at 100 mm intervals. The number of immunofluorescent positive and negative neurons was counted (L5), only those sectioned through the nucleolus to avoid double counting the same cell and were made in three to five contiguous sections from each ganglion. Statistical analysis was done using Sigma stat software. Student’s t-test was used to compare paired data and P-values obtained.

(3.009 1.97%) (Figs. 1 and 2). These counts were from cresyl violet stained sections (Fig. 1E and F). The morphology changes were clear at 6 h drug treatment compared with 4 and 2 h treatment. The morphology changes consisted of highly irregular neuronal boundaries, cytoplasmic vacuolations (Fig. 1E) and blebbings (Fig. 1A). Some of the non-nNOS neurons (close to nNOS neurons), ipsilateral to the nerve section in TRIM treated animals, also showed distorted boundaries and their nuclei sometimes appeared fragmented (Fig. 1A). Cells showing these changes can be discriminated from control DRG neurons, which have smooth round or oval profiles (Fig. 1B, D and F). On contralateral side, no neurons showing these morphology changes were observed. In the vehicle treated animals on the nerve-sectioned side very few neurons showed abnormal morphology (Fig. 1F; Fig. 2, Table 1). Sham operated animals did not show either a significant increase in number of nNOS neurons nor abnormal morphology following TRIM treatment (Fig. 1C and D, Fig. 2). We had known from dissociated DRG culture studies, the neurons undergo apoptosis following NOS block (6 h) as a result of the induction of pro-apoptotic bax and caspases (Thippeswamy and Morris, 1999). Hence, it was of interest to know whether neurons undergo apoptosis in vivo following nNOS inhibition.

3. Results There was a statistically highly significant increase in nNOS containing neurons in DRG (36.59 2.8%, P\ 0.0001) ipsilateral to the sciatic nerve section when compared with the contralateral ganglia (6.792.6%) (Table 1). This increase in nNOS expression was predominantly in small and medium sized neurons (Fig. 1). Axons were also intensely stained for nNOS. A similar increase in the number of neurons expressing c-Jun occurred on the nerve sectioned side (60.09 3.6%, P\ 0.0001, ipsilateral; 5.99 2.6% contralateral). The c-Jun immunoreactivity was confined to the nucleus of neurons in both ipsi- and contra-lateral DRG. In ipsilateral to nerve sectioned ganglia almost all nNOS containing neurons (  36%) expressed c-Jun and the other 24% c-Jun immunoreactivity was seen in neurons that did not express nNOS (Fig. 1, Table 1). The most significant finding in this study was the effect of the nNOS inhibitor, TRIM on cell morphology of DRG neurons ipsilateral to the nerve section. Nearly 50% (50.309 4.37%, P \ 0.0001) neurons showed morphology changes in 6 h TRIM treated compared with vehicle treated ipsilateral DRG neurons

Fig. 2. Histogram showing the effect of nNOS blocker, TRIM on neuronal cells morphology following sciatic nerve section in ipsilateral L5 DRG. The abnormal nNOS neurons are plotted as the percentage of total nNOS positive neurons for vehicle and TRIM treated axotomised animals (‘Vehicle’ and ‘TRIM’ labelled columns, respectively) and sham operated TRIM treated animals (column labelled ‘Sham’). The ipsilateral L5 DRG from axotomised TRIM treated animal contained large numbers of abnormal nNOS neurons when compared with vehicle controls (P \0.0001; Student’s t-test; n =4).

Ipsilateral Contralateral Ipsilateral Contralateral

TRIM treated

a

36.59 2.8 6.7 9 2.6 38.89 2.7 5.19 2.6

% nNOS positive neurons

% nNOS and c-Jun co-localisation 36.5 92.8 3.4 9 1.4 38.8 92.7 5.1 9 2.1

% c-Jun positive neurons 60.0 93.6 5.992.6 62.6 93.7 10.3 9 2.7

26.4 9 8.7 2.5 91.7 23.9 93.4 3.6 9 3.1

% c-Jun expression in non-nNOS neurons

39.69 4.3 94.19 1.7 37.39 4.6 89.7 91.6

% negative neurones to both c-Jun and nNOS

50.394.4 0 3.0 91.9 0

% nNOS showing morphology changes

Note that all other parameters except the percentage of abnormal morphology of nNOS neurons differ significantly in TRIM treated ipsilateral ganglia (also see Fig. 2).

Vehicle treated

Side of ganglia in relation to nerve section

Treatment

Table 1 Counts showing the baseline expression of nNOS and c-Jun in nerve sectioned animals and effects of TRIM treatmenta

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44 41

42

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44

The TUNEL method did not reveal any neurons undergoing this type of cell death at any time point tested (2, 4 and 6 h) whilst, positive nuclei were seen in some connective tissue cells and leucocytes in blood vessels. Control rat colon sections used in parallel showed numerous apoptotic positive nuclei. However, it appears that 6 h treatment in vivo is not sufficient to cause cell death. 4. Discussion The significant finding of this study is that the prevention of NO synthesis by inhibiting its enzyme activity for 6 h causes a rapid change in cellular morphology in axotomised DRG neurons. This confirms our earlier studies from NOS blocking experiments in NGF deprived DRG cultures (Thippeswamy and Morris, 1997b) and further reinforces the view that the increase in nNOS expression in stressed or injured neurons can contribute to neuronal survival. This has an important implication for any therapeutic strategy that may employ NOS blocking drugs. The well documented increase in nNOS expression in DRG neurons following axotomy (Verge et al., 1992; Zhang et al., 1993) was confirmed. Most nNOS neurons on the nerve sectioned side expressed c-Jun, confirming earlier findings (Fiallos-Estrada et al., 1993) and this transcription factor also expressed in other neurons those did not express nNOS. The induction of c-Jun as a result of growth factor deprivation or injury to the axons of the neurons is an established fact. However, mere presence of c-Jun is not important but its physiological state of activity determines the cell fate. Either diphosphorylation (Gu et al., 2000) or S-nitrosylation (Lipton, 1999; Teng et al., 1999) of c-Jun terminal protein kinases activates c-Jun. It has been proposed that, similar to phosphorylation, protein nitrosylation by NO is a sequence-specific, reversible and covalent modification event (Liu and Stamler, 1999). The role of NO in activation or inhibition of transcription factormediated gene expression has been the subject of a range of different studies. Some of these have suggested that NO does not activate c-Jun. For example, in one study NO enhanced the binding of AP-1 (Jun/Fos) to DNA, which was due to the stimulation of jun B and c-fos mRNA but not c-jun (Pilz et al., 1995). In PC12 cells, which are often used as model neurons, NO stimulated production of cGMP lead to an increase in c-fos and jun B expression but not c-jun or jun D (Haby et al., 1994). However, in another study, S-nitrosylation by NO of c-jun terminal kinase has been proposed as a mechanism whereby NO can prevent apoptosis (So et al., 1998). Very recently, it is shown that preconditioning ischemia prevents c-Jun N-terminal protein kinases activation that in turn protect cells against lethal ischemia (Gu et al., 2000).

It is interesting to note that the cultures deprived of serum (Andoh et al., 2000) and or NGF (Thippeswamy and Morris, 1997a) many neurons expressed nNOS that were co-localised with c-Jun positive nucleus (unpublished data). Although adult DRG neurons do not require NGF for their survival, axotomised adult neurons at least in some aspects resemble the serum and or NGF deprived cultured neurons. In the NGF responsive cells, binding of NGF to its receptor tyrosine kinase A, recruits several signal-transducing molecules that include components of the Ras-MAP kinase cascade and phosphotidylinositol-3-kinase (PI3K) (Greene and Kaplan, 1995; Segal and Greenberg, 1996). In the absence of NGF, the up-regulated nNOS synthesise NO that directly interact and activates Ras proteins via S-nitrosylation on a single cysteine (Cys-118) residue (Lander et al., 1996). Teng et al. (1999) have reported a similar mechanism of activation of c-Ha-Ras by NO in modulating survival-promoting effect in neuronal PC12 cells. Hence, it appears that following peripheral nerve injury, neurons are deprived of target-derived NGF and start to express nNOS. In the absence of NGF, the downstream effector pathway of NO itself or via cGMP (cGMP-dependent protein kinase-fos pathway) (Gudi et al., 1999) could cross talk with NGF downstream MAPK or JNK/JAK or upstream Ras-PI3K signal transduction pathway. The involvement of NO in endothelial protection against apoptosis by down regulating MKP-3 (MAPK phosphatase-3) mRNA has been reported (Rossig et al., 2000). It is likely that NO activates Ras-PI3K pathway (Teng et al., 1999), which in turn might prevent c-Jun kinases activation (Gu et al., 2000) to promote neuronal survival. In another study, Park et al. (2000) have demonstrated that endogenously produced NO can suppress c-Jun N-terminal kinase activation in intact cells. However, further studies are needed to reveal whether the elevation of nNOS and c-Jun that occurs in axotomised DRG neurons are functionally related. Studies from primary dissociated DRG cultures revealed that the neuronal cell death that was precipitated by inhibition of nNOS and sGC for 6 h involved the activation of the synthesis of Bax and that could be partially reversed with caspase blockers (Thippeswamy and Morris, 1999) suggesting that the cells were dying by apoptosis. However, in the present study the TUNEL staining method did not show any neurons undergoing nuclear fragmentation at any time interval (2, 4 and 6 h) tested, although some cells other than neurons had TUNEL positive nuclei revealing that some non-neuronal cells were undergoing apoptosis. Control rat colon sections with or without DNase I treatment, processed in parallel with DRG sections for TUNEL staining showed clear apoptotic nuclei suggesting that the method had worked. It is possible that the time point after drug treatment, 6 h, was not long

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44

enough such that these cells were entering apoptosis but had not started to undergo DNA disruption. However, it is not conclusive that neurons do not undergo apoptosis but it is equally possible that they are starting to undergo necrotic changes. The type of cell death, either apoptosis or necrosis depends on the intracellular ATP concentrations. It has been proposed that NO may impair ATP concentrations and the establishment of amplifying or inhibiting feedback loops that normally accelerate or prevent apoptotic cell death (Leist et al., 1999) depends on the redox state of the cell. The other possible explanation is the role of cGMP. The cGMP mediated anti-apoptotic activity is mediated via the induction of anti-apoptotic Ref-1, bcl2 and the inhibition of caspase 3 (Edwards et al., 1998; Grosch and Kaina 1999). Studies from human neuroblastoma cell lines also demonstrated the cGMP-dependent cytoprotection via up-regulation of anti-apoptotic protein bcl2 and down-regulation of p66shc (cytoplasmic signal transducer leading to oxidative damage and apoptosis) (Andoh et al., 2000). Cultured DRG neurons express cGMP (Thippeswamy and Morris, 2000) but neurons in vivo do not. Hence, it is probable that neurons undergo apoptosis in vitro due to inhibition of cGMP activity in addition to NO inhibition. The cGMP expression is restricted to glia cells in vivo and following TRIM treatment (see later), glia cells start expressing proapoptotic bax but not the neurons (unpublished data). The induction of apoptosis in Schwann cells via p75 (low affinity NGF receptor with death domains) as a result of peripheral axotomy has been reported (Syroid et al., 2000). A selective and time dependent glial and neuronal apoptosis after sciatic nerve section in neonatal rats (Whiteside et al., 1998) suggest a possible pattern of cell death as a result of interference with neuron-glial intercommunication. However, 6 h drug treatment is too early to say whether the neurons with morphological changes ultimately die. In vitro studies revealed that overnight to 24 h treatment with NOS blockers caused death of neurons (Thippeswamy and Morris, 1997b). However, studies of a more chronic nature with TRIM or sGC blocker at different time intervals from day 1 nerve section is required to establish the longer term fate of these neurons and whether the morphological changes seen are irreversible leading to cell death. Unfortunately, the current license does not permit for such chronic systemic injection of drugs in vivo. The morphological changes were specific to ganglion neurons on the nerve-sectioned side and it would be reasonable to assume that the neurons showing these changes were the axotomised ones. This is important as it suggests that TRIM is not just a non-specific neurotoxin. Furthermore, the fact that these changes were not confined to nNOS containing neurons suggests that NO is having a neuroprotective effect, which involves

43

inter-cellular communication. This coupled with the observation that exogenous NO can cause an increase in cGMP expression in DRG satellite glia cells (Morris et al., 1992) leads to the suggestion that the increase in nNOS seen after nerve injury could be protective not only to the neurons expressing this enzyme but to the surrounding glia cells as well. This role is further supported by the in vivo increase in cGMP in DRG satellite cells seen after axotomy (Shi et al., 1998). We have recently found that following TRIM treatment, the cGMP expression in satellite glia cells was abolished in axotomised DRG (Thippeswamy and Morris, 2000) suggesting the neuronal source of NO. In conclusion, the present data support the view that NO has a role in neuroprotection in many axotomised sensory neurons. This could be of considerable importance in understanding the processes that permit adult sensory neurons to withstand damage to their peripheral axons and regenerate. The neuroprotective role of NO is probably of importance for other regions of nervous system, for example, in protecting central neurons against ischemic induced oxidative stress (Andoh et al., 2000). This also suggests that the indiscriminate use of NOS-blockers to inhibit neurodegeneration (Chabrier et al., 1999) could cause undesirable effects.

Acknowledgements We are thankful to Dr P. Emson for providing nNOS antibodies. This research was supported by grants from the Wellcome Trust and The University of Liverpool. The Commonwealth Fellowship Commission supported Dr Jain and Dr Nazli by travel funds provided by SmithKline Beecham.

References Andoh, T., Lee, S.Y., Chiueh, C.C., 2000. Preconditioning regulation of bcl-2 and p66shc by human NOS1 enhances tolerance to oxidative stress. The FASEB journal express article 10.1096/fj.000151fje, published online. Chabrier, P.E., Demerle-Pallardy, C., Auguet, M., 1999. Nitric oxide synthases: targets for therapeutic strategies in neurological diseases. Cell Mol. Life Sci. 55, 1029– 1035. Dzoljic, E., De Vries, R., Dzoljic, M.R., 1997. Anticonvulsant activity of new and potent inhibitors of nitric oxide synthase. Brain Res. Bull. 43, 191– 195. Edwards, M., Rassin, D.K., Izumi, T., Mitra, S., Perez-Polo, J.R., 1998. APE/Ref-1 responses to oxidative stress in aged rats. J. Neurosci. Res. 54, 635– 638. Fiallos-Estrada, C.E., Kummer, W., Mayer, B., Bravo, R., Zimmermann, M., Herdegen, T., 1993. Long-lasting increase of nitric oxide synthase immunoreactivity, NADPH-diaphorase reaction and c-JUN co-expression in rat dorsal root ganglion neurons following sciatic nerve transection. Neurosci. Lett. 150, 169–173. Garthwaite, J., Boulton, C.L., 1995. Nitric oxide signaling in the central nervous system. Annu. Rev. Physiol. 57, 683– 706.

44

T. Thippeswamy et al. / Neuroscience Research 40 (2001) 37–44

Gavrieli, Y., Sherman, Y., Ben-Sasson, S.A., 1992. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119, 493–501. Gidday, J.M., Shah, A.R., Maceren, R.G., Wang, Q., Pelligrino, D.A., Holtzman, D.M., Park, T.S., 1999. Nitric oxide mediates cerebral ischemia tolerance in neonatal rat model of hypoxic preconditioning. J. Cereb. Blood Flow Metab. 19, 331–340. Gonzalez-Zulueta, M., Feldman, A.B., Klesse, L.J., Kalb, R.G., Dillman, J.F., Parada, L.F., Dawson, T.M., Dawson, V.L., 2000. Requirement for nitric oxide activation of p21ras/extracellular regulated kinase in neuronal ischemic preconditioning. Proc. Natl. Acad. Sci. 97, 436– 441. Greene, L.A., Kaplan, D.R., 1995. Early events in neurotrophin signalling via Trk and p75 receptors. Curr. Opin. Neurobiol. 5, 579– 587. Grosch, S., Kaina, B., 1999. Transcriptional activation of apurinic/ apyrimidinic endonuclease (Ape, Ref-1) by oxidative stress requires CREB. Biochem. Biophys. Res. Commun. 261, 859– 863. Gudi, T., Hong, G.K.P., Vaandrager, A.B., Lohmann, S.M., Pilz, R.B., 1999. Nitric oxide and cGMP regulate gene expression in neuronal and glial cells by activating type II cGMP-dependent protein kinase. FASEB J. 13, 2143–2152. Gu, Z., Jiang, Q., Zhang, G., Cui, Z., Zhu, Z., 2000. Diphosphorylation of extracellular signal regulated kinases and c-Jun N-terminal protein kinases in brain ischemic tolerance in rat. Brain Res. 860, 157– 160. Haby, C., Lisovoski, F., Aunis, D., Zwiller, J., 1994. Stimulation of the cyclic GMP pathway by NO induces expression of the immediate early genes c-fos and junB in PC12 cells. J. Neurochem. 62, 496– 501. Handy, R.L., Harb, H.L., Wallace, P., Gaffen, Z., Whitehead, K.J., Moore, P.K., 1996. Inhibition of nitric oxide synthase by 1-(2trifluoromethylphenyl) imidazole (TRIM) in vitro: antinociceptive and cardiovascular effects. Br. J. Pharmacol. 119, 423–431. Handy, R.L., Moore, P.K., 1998. Effects of selective inhibitors of neuronal nitric oxide synthase on carrageenan-induced mechanical and thermal hyperalgesia. Neuropharmacology 37, 37– 43. Ho, Y.S., Liu, H.L., Duh, J.D., Chen, R.J., Ho, W.L., Jeng, J.H., Wang, Y.J., Lin, J.K., 1999. Induction of apoptosis by S-Nitrosoglutathione, Cu2 + or Ni2 + ion through modulation of bax, bad and bcl2 proteins in human colon adenocarcinoma cells. Mol. Carcinog. 26, 201– 211. Kenney, A.M., Kocsis, J.D., 1998. Peripheral axotomy induces longterm c-Jun amino-terminal kinase-1 activation and activator protein-1 binding activity by c-Jun and junD in adult rat dorsal root ganglia in vivo. J. Neurosci. 18, 1318–1328. Lander, H.M., Milbank, A.J., Tauras, J.M., Hajjar, D.P., Hempstead, B.L., Schwartz, G.D., Kraemer, R.T., Mirza, U.A., Chait, B.T., Burk, S.C., Quilliam, L.A., 1996. Redox regulation of cell signaling. Nature 381, 380–381. Leist, M., Single, B., Naumann, H., Fava, E., Simon, B., Kuhnle, S., Nicotera, P., 1999. Inhibition of mitochondrial ATP generation by nitric oxide switches apoptosis to necrosis. Exp. Cell Res. 249, 396– 403. Lipton, S.A., 1999. Neuronal protection and destruction by NO. Cell Death Differ. 6, 943–951. Liu, L., Stamler, J.S., 1999. NO: prevents cell death. Cell Death Differ. 6, 937– 942. Melino, G., Bernassola, F., Knight, R.A., Corasaniti, M.T., Nistico, G., Finazi-Agro, A., 1999. S-nitrosylation regulates apoptosis. Nature 388, 432– 433. Morris, R., Southam, E., Braid, D.J., Garthwaite, J., 1992. Nitric oxide may act as a messenger between dorsal root ganglion neurons and their satellite cells. Neurosci. Lett. 137, 29–32. .

Pantazis, N.J., West, J.R., Dai, D., 1998. The NO-cGMP pathway plays an essential role in both promoting cell survival of cerebellar granule cells in culture and protecting the cells against ethanol neurotoxicity. J. Neurochem. 70, 1826– 1838. Park, H.S., Huh, S.H., Kim, M.S., Lee, S.H., Choi, E.J., 2000. Nitric oxide negatively regulates c-Jun N-terminal kinase/stress-activated protein kinase by means of S-nitrosylation. Proc. Natl. Acad. Sci. 97, 14382– 14387. Pilz, R.B., Suhasini, M., Idriss, S., Meinkoth, J.L., Boss, G.R., 1995. Nitric oxide and cGMP analogs activate transcription from AP-1responsive promoters in mammalian cells. FASEB J. 9, 552–558. Rossig, L., Haendeler, J., Hermann, C., Malchow, P., Urbich, C., Zeiher, A.M., Dimmeler, S., 2000. Nitric oxide down-regulates MKP-3 mRNA levels: involvement in endothelial cell protection from apoptosis. J. Biol. Chem. 275, 25502– 25507. Segal, R.A., Greenberg, M.E., 1996. Intracellular signaling pathways activated by neurotrophic factors. Annu. Rev. Neurosci. 19, 463– 489. Shi, T.J., Holmberg, K., Xu, Z.Q., Steinbusch, H., de Vente, J., Hokfelt, T., 1998. Effect of peripheral nerve injury on cGMP and nitric oxide synthase levels in rat dorsal root ganglia: time course and coexistence. Pain 78, 171– 180. So, H.S., Park, R.K., Kim, M.S., Lee, S.R., Jung, B.H., Chung, S.Y., Jun, C.D., Chung, H.T., 1998. Nitric oxide inhibits c-Jun N-terminal kinase 2 (JNK2) via S-nitrosylation. Biochem. Biophys. Res. Commun. 247, 809– 813. Syroid, D.E., Maycox, P.J., Soilu-Hanninen, M., Petratos, S., Bucci, T., Burrola, P., Murray, S., Cheema, S., Lee, K.F., Lemke, G., Kilpatrick, T.J., 2000. Induction of postnatal Schwann cell death by the low-affinity neurotrophin receptor in vitro and after axotomy. J. Neurosci. 20, 5741– 5747. Teng, K.K., Esposito, D.K., Schwartz, G.D., Lander, H.M., Hempstead, B.L., 1999. Activation of c-Ha-Ras by nitric oxide modulates survival responsiveness in neuronal PC12 cells. J. Biol. Chem. 274, 37315– 37320. Thippeswamy, T., Morris, R., 1997a. Nerve growth factor inhibits the expression of nitric oxide synthase in neurons in dissociated cultures of rat dorsal root ganglia. Neurosci. Lett. 230, 9–12. Thippeswamy, T., Morris, R., 1997b. Cyclic guanosine 3%,5%monophosphate-mediated neuroprotection by nitric oxide in dissociated cultures of rat dorsal root ganglion neurons. Brain Res. 774, 116– 122. Thippeswamy, T., Morris, R., 1999. How does nitric oxide protects dorsal root ganglion neurons? 9th World Congress on Pain, IASP abstract vol. 106, pp. 34. Thippeswamy, T., Morris, R., 2000. Evidence that nitric oxide induced synthesis of cGMP occurs in a paracrine but not autocrine fashion and that the site of its release can be regulated: studies in dorsal root ganglia in vivo and in vitro. In: Nitric Oxide: Biology and Biochemistry, in press. Towler, P.K., Bennett, G.S., Moore, K., Brain, S.D., 1998. Neurogenic oedema and vasodilataion: effect of a selective neuronal NO inhibitor. Neuroreport 11, 1513– 1518. Verge, V.M., Xu, Z., Xu, X.J., Wiesenfeld, H.Z., Hokfelt, T., 1992. Marked increase in nitric oxide synthase mRNA in rat dorsal root ganglia after peripheral axotomy: in situ hybridization and functional studies. Proc. Natl. Acad. Sci. USA 89, 11617– 11621. Whiteside, G., Doyle, C.A., Hunt, S.P., Munglani, R., 1998. Differential time course of neuronal and glial apoptosis in neonatal rat dorsal root ganglia after sciatic nerve axotomy. Eur. J. Neurosci. 10, 3400– 3408. Zhang, X., Verge, V.M., Wiesenfeld, H.Z., Ju, G., Bredt, D., Synder, S.H., Hokfelt, T., 1993. Nitric oxide synthase-like immunoreactivity in lumbar dorsal root ganglia and spinal cord of rat and monkey and effect of peripheral axotomy. J. Comp. Neurol. 335, 563– 575.