Moderately Different NADPH-Diaphorase Positivity in the Selected Peripheral Nerves after Ischemia/ Reperfusion Injury of the Spinal Cord in Rabbit

Abstract

1. To vicariously investigate the nitric oxide synthase (NOS) production after spinal cord injury, NADPH-d histochemistry was performed on the selected peripheral nerves of adult rabbits 7 days after ischemia. The effect of transient spinal cord ischemia (15 min) on possible degenerative changes in the motor and mixed peripheral nerves of Chinchilla rabbits was evaluated.

2. The NADPH-diaphorase histochemistry was used to determine NADPH-diaphorase activity after ischemia/reperfusion injury in radial nerve and mediane nerve isolated from the fore-limb and femoral nerve, saphenous nerve and sciatic nerve separated from the hind-limb of rabbits. The qualitative analysis of the optical density of NADPH-diaphorase in selected peripheral nerves demonstrated different frequency of staining intensity (attained by UTHSCSA Image Tool 2 analysis for each determined nerve).

3. On the seventh postsurgery day, the ischemic spinal cord injury resulted in an extensive increase of NADPH-d positivity in isolated nerves. The transient ischemia caused neurological disorders related to the neurological injury—a partial paraplegia. The sciatic, femoral, and saphenous nerves of paraplegic animals presented the noticeable increase of NADPH-d activity. The mean of NADPH-diaphorase intensity staining per unit area ranged from 134.87 (±32.81) pixels to 141.65 (±35.06) pixels (using a 256-unit gray scale where 0 denotes black, 256 denotes white) depending on the determined nerve as the consequence of spinal cord ischemia. The obtained data were compared to the mean values of staining intensity in the same nerves in the limbs of control animals (163.69 (±25.66) pixels/unit area in the femoral nerve, 173.00 (±32.93) pixels/unit area in saphenous nerve, 186.01 (±29.65) pixels/unit area in sciatic nerve). Based on the statistical analysis of the data (two-way unpaired Mann–Whitney test), a significant increase (p≤0.05) of NADPH-d activity in femoral and saphenous nerve, and also in sciatic nerve (p≤0.001) has been found. On the other hand, there was no significant difference between the histochemically stained nerves of fore-limbs after ischemia/reperfusion injury and the same histochemically stained nerves of fore-limbs in control animals.

4. The neurodegenerative changes of the hind-limbs, characterized by damage of their motor function exhibiting a partial paraplegia after 15 min spinal cord ischemia and subsequent 7 days of reperfusions resulted in the different sensitivity of peripheral nerves to transient ischemia. Finally, we suppose that activation of NOS indirectly demonstrable through the NADPH-d study may contribute to the explanation of neurodegenerative processes and the production of nitric oxide could be involved in the pathophysiology of spinal cord injury by transient ischemia.

INTRODUCTION

The biology of nitric oxide (NO) received an enormous interest from scientists of different scientific disciplines for last 10 years (Lukacova et al., 1999, 2003; Keilhoff et al., 2002). This interest resulted from a unique role of NO in various physiological systems (Del-Bel et al., 2005). The nature and the possible participation of this molecule on neuromodulation have been discussed by Rand and Li (1995).

This nitrergic transmitter with diverse functions in signal transduction (Snyder and Bredt, 1991; Sanders and Ward, 1992) and with an extreme diversity of functions in physiology and pathology is generated by a family of enzymes known as NO-synthases (Prast and Philippu, 2001a). Three different forms of this enzyme are known, the endothelial (eNOS) responsible for cardiovascular actions, the inducible (iNOS) found originally in macrophages and involved mainly in immunological processes, and the neuronal one (nNOS). Each of the isoforms also possesses a nicotinamide adenine dinucleotide phosphate (NADPH) reduction capacity (the so-called NADPH-diaphorase activity) which is used as a histochemical detection method for NO-producing structures (Bredt et al., 1991). A considerable number of studies demonstrated that NADPH-d activity found in neuronal structures is congruent with the presence of nNOS (Bredt et al., 1991; Dawson et al., 1991; Garthwaite, 1991; Decker and Reuss, 1994).

The production of NO is a calmodulin-dependent process; therefore, it must be preceded by elevation of intracellular Ca²⁺ concentration (Kiss, 2000). Ca²⁺ influx is induced by activation of glutamate receptors, preferentially NMDA receptors (Prast and Philippu, 2001b).

The biomolecule of NO has one unpaired electron, making it a free radical that avidly reacts with other molecules (Cullota and Koshland, 1992). Nitric oxide synthase, recently shown to be a 150 kDa, NADPH-dependent enzyme in brain, is responsible for the calcium/calmodulin-dependent synthesis of the guanylyl cyclase activator nitric oxide from L-arginine (Hope et al., 1991). The role of enhanced levels of NO implicated by the upregulation of NOS has not been clearly defined.

The NOS activity and NADPH-diaphorase copurify to homogeneity and both these activities could be immunoprecipitated with an antibody recognizing neuronal NADPH-diaphorase. Furthermore, NO synthase can be competitively inhibited by the NADPH-diaphorase substrate, nitro blue tetrazolium. The neuronal NADPH-diaphorase is a nitric oxide synthase, and NADPH-diaphorase histochemistry, therefore, provides a specific histochemical marker for neurons producing nitric oxide (Hope et al., 1991).

NOS-associated NADPH-d activity can be found in highly selective locations in the central and peripheral nervous system (Marsala et al., 2003). NO appears to have a variety of interesting local actions after nerve injury. An increase in the NO production under some neuropathological conditions may result, due to increased activity of NOS which is histochemically detectable, in increased expression of a reduced form of nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase (Ando et al., 1996; Kucharova et al., 2001; Keilhoff et al., 2002; Lukacova et al., 2003).

The nervous system damage caused by a transient ischemia-induced injury is often seen, and as documented, its consequences are among most serious complications in the central and peripheral nervous system (Marsala et al., 1997; Kucharova et al., 2004).

During ischemia and reperfusion multiple independently fatal terminal pathways are activated involving loss of membrane integrity, progressive proteolysis, and inability of control these processes. The changes started by hypoxia lead to NOS activation, resulting in nitric oxide production (White et al., 2000).

In the present study, the NADPH-d histochemical reaction was applied to longitudinal sections of motor and mixed peripheral nerves, so as to demonstrate the NADPH-d reactivity in the nervous system of rabbit. In addition, physiological demonstration of transient ischemia forms the basis of this paper.

MATERIALS AND METHODS

Chinchilla rabbits of both sexes were supplied from the Animal Farm, Kosice, Slovak Republic. The animals used in the experiment had free access to water and food during the experiment. Experimental protocols were approved by the Institute of Neurobiology Animal Care Committee with the aim of minimizing both the suffering and the number of animals used. All experiments conformed to established international guidelines on the ethical use of animals. All efforts were made to minimize animal suffering and to reduce the number of animals used. The experiments were performed on 11 adult rabbits, male and female, weighing 3000–3500 g divided into two groups: (1) control animals (n=5) without ischemic injury and (2) animals (n=6) with ischemia/reperfusion injury.

Ischemia/Reperfusion Injury

The animals were housed under standard conditions for at least 2 weeks before starting the experiments. The rabbits were anesthetized i.m. with 5% Narkamon (ketaminum 50 mg/mL) and 2% Rometar (Xylazinum 20 mg/mL) in 2:1 ratio. Animals were brought to a total anesthesia using inhalation mask with 1.0–2.0% halothane. Spinal cord ischemia was induced by a Fogarty catheter 4F (Arterial Embolectomy Catheter) occlusion of abdominal aorta, through the femoral artery, below the left renal artery (de Haan et al., 1998). The balloon in the end of catheter was quickly inflated for the period of 15 min using syringe with 1.5 mL of air. The balloon was inflated up, thus achieving the loss of distal aorta pressure. The surgical procedure was performed on the thermal electric pad. After the surgery, the animals were treated with 1 mL Novalgin (analgesic) and 1 mL Amoclen (antibiotic). The reperfusion period after ischemic injury persisted for 7 days. The animal was considered ischemic when, pulled by the tail, it failed to fully extend of the injured limbs and turned contralaterally (Vannucchi et al., 2005).

Transcardial Perfusion

As soon as the 7-day postsurgery period passed, the animals were deeply anesthetized with intramuscular injection of 5% Narkamon (ketaminum 50 mg/mL) and 2% Rometar (Xylazinum 20 mg/mL) in 2:1 ratio. Transcardial perfusion was performed with saline followed by freshly prepared 4% paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.4. Both solutions were preheated to 37°C and infused at a pressure of 130 mmHg.

Histological Processing—NADPH-d Histochemistry

After perfusion, the animals were utilized for histochemical analysis. The peripheral nerves were carefully dissected out from right and left fore-limb and hind-limb of rabbits. The samples were stored in the same fixative 4% paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.4 for 48 h at 4°C and then transferred to 30% sucrose in PBS at 4°C for 1 h. The use of sucrose as a cryoprotectant may enhance the staining intensity of NADPH-d expressing pyramidal cells in the hippocampal formation. This idea is supported by the fact that preprocessing sucrose incubation results in an enhancement of pyramidal cell eNOS, but not nNOS, immunoreactivity (Vaid et al., 1996). The material was sectioned longitudinally at 35 μm on LEICA cryostat and processed for NADPH-d staining.

The sections were incubated for 3 h at 37°C in a solution containing 0.25 mg/mL of β-NADPH (Tetrasodium Salt M.W.=833,4 [2646-71], EEC N. 220-163-3, Sigma Chemical Co., St. Louis, MO 63178, USA), 0.5 mg/mL of nitroblue tetrazolium (2,2′-Di-p-nitrophenyl-5,5′-diphenyl-3,3′-[3,3′-dimethoxy-4,4′-diphenylene]ditetrazolium chloride), [298-83-9] NBT M.W.=817.6, Sigma Aldrich Chemie GmBh, Steinheim, Germany), 6 μL/mL of N,N-dimethylformamidum (C₃H₇NO, M.W.=73.09; Sigma M1125), 0.8 mg/mL monosodium malate (C₄H₄O₄, M.W.=116.07; Sigma, M-1125), and 0.8% Triton X-100 dissolved in 0.1 M phosphate buffer. The section were then rinsed in 0.1 M phosphate buffer, mounted onto alcohol coated glass slides and air-dried overnight at room temperature. The same sections were after NADPH-d staining additionally counterstained for neutral red.

Neutral Red Staining

The glass slides with the section of NADPH-d stained nerves were submerged to the descending sequence of alcohol solutions (95% alcohol, 70% alcohol, 50% alcohol) in the intervals of 10 s. The sections were then rinsed in distilled water and incubated in Neutral red (Sigma [553-24-2], Sigma Chemical, St. Louis, MO, USA, 3-Amino-7-dimethylamino-2-methylphenazine hydrochloride, F.W.=288.8; pH range red 6.8) for 1 min. Incubation was followed by rinsing of slides with sections in distilled water repeatedly. Before Entellan fixation (Entellan new [1.07961.0500], Merck KgaA, Darmstadt, Germany) the sections were submerged to the ascending sequence of alcohol solutions (50% alcohol, 70% alcohol, 95% alcohol) in the intervals of 10 s and rinsed twice in xylene.

Qualitative Analysis

The data were collected with Nikkon-microscope, a digital camera DP50 and Olympus DP-soft, version 3.0 and the density of nerves was analyzed. The criteria for the computer evaluation were established by a light microscopic observation. The 35 μm thick sections were transformed by a digital camera. The magnification (10×) and the resolution (2776×2074) made it possible to distinguish the changes with the distance of 3 μm in the computer image. Qualitative image analysis was performed on a minimum of five sections of peripheral nerve from each animal using the UTHSCSA Image Tool 2 program. Threshold intensity was determined for each image, above which most background noise was suppressed. By using a 256-unit gray scale (0: black, 256: white) the optical density of NADPH-d was determined. The calculated percentage change values for each section were then averaged and a mean and standard deviation were determined per selected nerve.

Data Analysis

Mann-Whitney unpaired tests were used in statistical analysis of the data, which are presented as means±SEM, and p values less than 0.05 were considered significant.

This approach was used for studying the localization of NADPH-d in the peripheral nerves.

RESULTS

Using a transient spinal cord ischemia model in rabbit we demonstrated that changes in NOS reactivity correlate with the changes in NADPH-d positivity of peripheral nerves. Ischemic rabbits of both sexes experienced a significantly higher threshold for a NADPH-d density in affected peripheral nerves.

We focused on comparison of NADPH-d staining intensity in peripheral nerves isolated from rabbit's fore-limb (radial nerve and mediane nerve) and the same hind-limb (femoral nerve, sciatic nerve, and saphenous nerve). When NADPH-d positivity was divided into two groups of darkly (higher positivity) and lightly (lower positivity) stained nerves, the significant difference was observed between the individual nerves.

The qualitative evaluation of the optical density of NADPH-diaphorase in selected peripheral nerves of rabbit displayed the different frequency of staining intensity attained by Image Tool analysis for each determined nerve. The mean of NADPH-diaphorase intensity staining per unit area reached 163.69 (±25.66) pixels in femoral nerve, 173.00 (±32.93) in saphenous nerve, 186.01 (±29.65) pixels in sciatic nerve, 179.83 (±31.02) pixels in radial nerve and 171.93 (±30.11) pixels in mediane nerve. To sum up, they ranged from 163.69 (±25.66) to 186.01 (±29.65) pixels depending on the determined nerve (Table I). The mean relative intensity of NADPH-d staining reached 63% of minimum possible intensity (the strongest positivity) in femoral nerve, 67% in mediane nerve, and 68% in saphenous nerve. The farthest from the maximum NADPH-d positivity were the mean values obtained by the evaluation of NADPH-d activity in sciatic nerve (72% of minimum possible intensity) and in radial nerve (69% of minimum possible intensity). Based on the statistical analysis of the data (two-way unpaired Mann–Whitney test), it has been presented a significant increase (p≤0.05) of NADPH-d activity in femoral and saphenous nerve, so as in sciatic nerve (p≤0.001) after spinal cord ischemia. There was no significant difference between the histochemically stained nerves of fore-limbs after ischemia/reperfusion injury and the same histochemically stained nerves of fore-limbs in control animals.

Table I.

The Effect of Ischemia/Reperfusion Injury on Selected Peripheral Nerves

	Control animals (n=5)		15 min ischemia (n=6)
	Pixels	%	Pixels	%	Δ of NADPH-d intensity (%)
Saphenous nerve	173.00±32.93	67	141.65±35.06	55	12*
Sciatic nerve	186.01±29.65	72	139.87±19.28	54	18**
Femoral nerve	163.69±25.66	63	134.87±32.81	52	11*
Radial nerve	179.83±31.02	69	154.84±21.25	60	9
Mediane nerve	171.93±30.11	67	154.00±22.49	60	7

Note. The analysis of the data accomplished by using a 256-unit gray scale (0: black, 256: white). The change of optical density (control and 15-min ischemia) manifested by the increase of NADPH-d intensity staining in injured peripheral nerves. Mann–Whitney unpaired tests were used in statistical analysis of the data, which are presented as means±SEM, p-values less then 0.05 were considered significant. NADPH-d: nicotinamide adenine dinucleotidphosphate diaphorase.^* p≤0.05.^** p≤0.001.

Short-termed ischemia resulted in a large increase of NADPH-d positivity in the identified nerves. After a suitable ischemic intervention, the sensitivity of the peripheral nerves resulted in a high optical density of NADPH-d in motor peripheral nerve. The mean relative intensity in femoral nerve reached 52% of minimum possible intensity after ischemic/reperfusion injury. It means that stronger positivity was visible there than in case without intervention. This significant divergence presents the most noticeable NADPH-d reactivity in presented peripheral nerves. The high NADPH-d positivity is not negligible in mixed nerves isolated from hind-limbs (sciatic nerve—the 18% increase of positivity and saphenous nerve—the 12% increase of positivity after injury). The different sensitivity of the peripheral nerves to transient ischemia has been reflected in the greater spread of medians on densitograms in ischemic/reperfusion group (Fig. 1). The NADPH-d positivity in the control group reached the mean value 175.88 (±30.57) pixels/unit area in mediane nerve and radial nerve, and the mean value 174.23 (±29.41) pixels/unit area in femoral nerve, saphenous nerve, and ischiadicus nerve. The mean value of NADPH-d positivity after 15 min spinal cord ischemia in the nerves isolated from the fore-limb exceeded 154.42 (±21.87) pixels and for the femoral, saphenous, and sciatic nerve separated from the hind-limb reached the mean value 138.80 (±29.05) pixels/unit area. On the basis of acquired data, we concluded that the difference of NADPH-d positivity between nerves of fore-limb and nerves of hind-limb seems to be greater in the ischemic group than in control group.

Fig. 1.

The qualitative evaluation of the optical density of NADPH-diaphorase in selected peripheral nerves. Densitograms display the different frequency of staining intensity (in pixels/cryostat section) attained by qualitative Image Tool analysis. The ischemic/reperfusion injury merged into the increase of optical density visually manifested by the increase of intensity NADPH-d staining. Left: control group. Right: ischemic/reperfusion group closely fits to a Gaussian distribution.

Short-lasting ischemia caused variously graded damage in ischemia injured rabbits and, as a rule, led to a mild or intense paraplegia. Generally, the involvement of the peripheral nerves innervating hind-limbs was expressed more strongly than in those innervating fore-limbs. This difference appeared significant when compared with control animals and also in those cases when the intensity of NADPH-d staining responding to ischemia (it means in hind-limbs) was compared with the intensity detected in the nerves innervating the fore-limbs (Fig. 2).

Fig. 2.

The effect of transient ischemia on NADPH-d reactivity of the peripheral nerves. Microphotographs of 35 μm thick sections with the NADPH-d positive nerve fibers running longitudinally. Left: moderate and high levels of NADPH-d staining 7 days after ischemic injury. Right: control group. A1, A2: saphenous nerve. B1, B2: sciatic nerve. C1, C2: femoral nerve. D1, D2: mediane nerve. E1, E2: radial nerve. Scale bar=200 μm.

To additionally identify the areas subject to ischemic damage, we applied counterstaining—the neutral red staining, for assessing the outcome of spinal cord ischemia in the peripheral nerves. The NADPH-d staining highlighted the nerve fibers, whereas the epineurium of singled peripheral nerves was picked out with neutral red stain (Fig. 3).

Fig. 3.

The effect of transient ischemia on NADPH-d reactivity of the peripheral nerves. Microphotographs of 35 μm thick sections with the NADPH-d positive nerve fibers running longitudinally. The samples were counterstained with Neutral Red. Right: moderate and high levels of NADPH-d staining 7 days after ischemic injury. (A) Control group—neutral red staining; (B) Control group—NADPH-d staining counterstained with Neutral Red; (C) Ischemia/reperfusion group—NADPH-d staining counterstained with Neutral Red. 1: mediane nerve; 2: radial nerve; 3: femoral nerve; 4: sciatic nerve; 5: saphenous nerve. Scale bar=200 μm.

In conclusion, transient ischemia/reperfusion led to an upregulation of NOS reactivity through NADPH-diaphorase activity accompanied by a strongly enhanced NADPH-d staining in the individual peripheral nerves.

DISCUSSION

The main aim of this study points to the importance of NADPH-diaphorase and in the same way nitric oxide in the peripheral nervous system. Our goal was also to evaluate the significance of presence of this neuronal transmitter in the peripheral nerves through the NADPH-d activity and to consider the possible involvement of peripheral nerves in further investigation of neurodegeneration.

The statistical data obtained in the present study revealed the NADPH-d positivity in all five demonstrated nerves (radial nerve, mediane nerve, femoral nerve, saphenous nerve, and sciatic nerve). The transient ischemia induced a discernible increase of NADPH-d positivity in femoral nerve, saphenous nerve, and sciatic nerve and evoked a partial paraplegia exemplified by a partial immobility of extremities.

Described results partially confirm those reported by Ando et al. (1996) indicating that the nerve fibers of lumbosacral intermediolateral neurons found mainly in IML region, areas adjacent to the central canal, and lamina I, lamina II of the dorsal horn, were intensively stained for NADPH-diaphorase.

The NADPH-d staining by using transverse, horizontal, and sagittal sections from cervical, thoracic, and lumbosacral segments of rabbits has shown a morphologically heterogeneous appearance of NADPH-d exhibiting neurons located in the superficial and deep dorsal horn, in the pericentral region all along the rostrocaudal axis of the spinal cord, and in the dorsal gray commissure in segments S1–S3. A homogeneous population of NADPH-d labeled neurons was located in the IML and sacral parasympathetic nucleus, with a small population of interneurons in the sacral parasympathetic nucleus. In the pericentral region (lamina X), a close association of NADPH-d exhibiting somata and fibers and mostly longitudinally oriented blood vessels was detected (Marsala et al., 1999).

Very dense NADPH-diaphorase-positive nerve terminal fields were seen in the olfactory tubercle, cortex, caudate nucleus, putamen, dentate gyrus, and interpeduncular nucleus. Intensely stained NADPH-diaphorase-positive nerve fibers were found in the stria terminalis, marginal region of the central tegmental field, dorsal tegmental nucleus, and spinal trigeminal tract (Mizukawa et al., 1989). The extensively represented NADPH-diaphorase positivity in the central nervous system seems to be the origin for the NADPH-d positivity in the peripheral nerves. This consideration supports a massive occurrence of axonal NADPH-d positivity detected in the juxtagriseal layer of the ventral column along the rostrocaudal axis of the spinal cord. The prominent NADPH-d-exhibiting bundles containing thick, smooth, nonvaricose axons were identified in the mediobasal and central portion of the ventral column (Marsala et al., 2002).

Based on the comparison of results related with NADPH-d positivity separately in male and female rabbits, we concluded that there were no significant differences between different sexes (unpublished data). On that account, the sex of experimental animals was not explicitly specified.

The positivity of NADPH-d staining of peripheral nerves was enhanced due to ischemic insult. Based on the study of Young et al. (1997), the presence of NADPH-diaphorase staining compared with the immunohistochemical localization of four NADPH-dependent enzymes—neuronal, inducible, and endothelial nitric oxide synthase and cytochrome P450 reductase—confirmed the fact that while tissues that demonstrate immunoreactivity for neuronal and endothelial NOS also stain positively for NADPH-diaphorase activity, the presence of NADPH-diaphorase staining does not reliably or specifically indicate the presence of one or more NOS isoforms. The NADPH-d method is also unspecific, as it does not allow the differentiation between the NOS isoforms.

The NADPH-d staining might be sensitive to prolonged periods of fixation. The duration of postfixation longer than 12 h may inactivate the enzyme resulting in diminished or even abolished staining (Reuss and Riemann, 2000). A modification of NADPH-d staining by performing the incubation in the presence of formaldehyde allows for a more specific detection of the histochemical nNOS activity than the conventional method (Grozdanovic et al., 1995).

The tissue incubation in sucrose solution prior to immunocytochemistry, enhances immunoreactivity of the endothelial isoform of NOS whilst having little effect on neuronal NOS reactivity (Vaid et al., 1996). A comparable enhancement of eNOS, but not nNOS, neuronal immunoreactivity was also seen in the section after brain incubation, suggesting that the incubation procedure may preserve the viability of this membrane-associated enzyme thereby enabling the NADPH-d histochemical reaction to proceed. The NADPH-d procedure maintains the viability of the active form of eNOS in pyramidal cells, which is predominantly membrane-associated enzyme. This may be a consequence of the osmotic gradient arising from sucrose preincubation, which results in alterations of membrane structures in a way that allows for enhanced histochemical/immunohistochemical staining, possibly by increasing the availability of membrane-associated eNOS for detection. This active form of eNOS would then be able to transfer electrons to the tetrazolium salt, nitroblue tetrazolium, and the NADPH-d histochemical reaction would be able to occur. The differences in the degree of reactivity of NADPH-d expressing pyramidal cells in the hippocampal formation may be due to a number of factors. First, differences in intensity may reflect possible differences in the degree of NOS activity, either eNOS, or nNOS, or both. Alternatively, the densitometric gradients may reflect differences in the inherent properties of the cells, i.e. the neurotransmitter/neuropeptide content. In relation to nNOS neuronal staining, there was no apparent difference in neuronal immunoreactivity from sections obtained from sucrose-incubated brains (Vaid et al., 1996).

Neutral red staining seems to be a simple method for assessing the outcome of focal cerebral ischemia, in an appropriate model of cerebral ischemia (Uno et al., 1995). Therefore, we decided to apply this additional counterstaining on selected peripheral nerves in the model of spinal cord ischemia.

In contrast to our results dealing with the transient ischemia, seems to be the data published by Qi et al. (2001). The results of these authors showed that, after ischemia (2 h), both nNOS and eNOS protein expressions decreased. After ischemia/reperfusion (2 h of ischemia followed by 3 h of reperfusion), expression of both nNOS and eNOS mRNA and protein decreased further. In contrast, iNOS mRNA significantly increased after ischemia and was further upregulated (14-fold) after I/R, while iNOS protein was not detected. The results reveal the dynamic expression of individual NOS isoforms during the course of ischemia/reperfusion injury. An understanding of this modulation on a cellular and molecular level may lead to understanding the mechanisms of ischemic/reperfusion injury.

The duration of spinal cord ischemia used in the current study was short-termed (15 min) but it seems to be strong enough to induce NO-mediated neurotoxicity.

Consistently with the description of an intense NADPH-d staining of thick myelinated afferents of dorsal horn, a noticeable number of identified axons located in the nerve system demonstrated the high level of NADPH-d staining (Marsala et al., 2002).

Other studies report that NADPH-d activity matches the presence of NOS. Thus, it appears that the staining techniques are complementary, the detection of NADPH-d reactivity primarily gives information about the enzymatic activity of NO-producing neurons, while the presence of nNOS immunopositivity gives a measure of the occurrence of the enzyme but not necessarily its activity (Hope and Vincent, 1989; Hope et al., 1991).

Because nNOS is distributed throughout the cytoplasm of neurons, NO might be produced in axons, terminals, dendrites, and somata. Electrochemical measurements of NO in the molecular layer of the cerebellum (Shibuki and Kimura, 1997), the substance gelatinosa of the spinal cord (Kimura et al., 1999), and dialysis measurements in the thalamus (Williams et al., 1997), which is innervated by axons of NOS-containing mesopontine cholinergic neurons, suggest that activity-dependent NO production in these structures arises from axon terminals. However, given the high density of NOS-containing somata and dendrites within the laterodorsal tegmentum, it is more likely that the electrically evoked changes in NO measured here were generated at these somatic and dendritic sites (Leonard et al., 2001). In some neurons, nNOS appears to be selectively regulated by NMDA receptor-evoked Ca²⁺ entry (Kiedrowski et al., 1992). This may be mediated by the association of nNOS with NMDA receptors and by their mutual targeting to postsynaptic densities (Christopherson et al., 1999) where Ca²⁺ influx through NMDA receptors mediates large increases in local Ca²⁺ (Yuste et al., 1999). Elevation of cytoplasmatic Ca²⁺ by several different pathways might activate nNOS. The Ca²⁺ imaging studies from NOS-containing laterodorsal tegmental neurons indicate than even weak activation of NMDA receptors stimulates large somatodendritic Ca²⁺ transients mediated by action potential-activated voltage-gated Ca²⁺ channels (Leonard et al., 2000) rather than by Ca²⁺ influx through NMDA receptors. Thus it seems unlikely that the observed NMDA-evoked changes in NO could be mediated by a restricted influx of Ca²⁺ through NMDA receptors. It is more likely that Ca²⁺ influx through voltage-gated Ca²⁺ channels played a common role in both the NMDA- and electrically evoked NO production (Leonard et al., 2001).

The NADPH-d histochemistry has been shown to stain the nerve fibers of peripheral nerves which contain NO synthase responsible for the biosynthesis of the freely diffusable NO. The variations of NADPH-d staining in neurons and their fibers may reflect different roles of NO in different subpopulation of neurons. It will be necessary to examine whether ischemia induces de novo synthesis of NOS and/or whether a low basal level of NOS becomes manifest after ischemia as a result of upregulated expression and increased protein synthesis. The imbalance in regulation of nitric oxide could disturb the physiological function of peripheral nerve system.

To determine if reactive nitrogen species or NOS-induced changes in ischemia/reperfusion are primary causes of specific neurological diseases or merely secondary effects, remains to be clarified in several areas from inflammatory processes to neurological diseases. With respect to the functional role of lesion-induced NOS activity, it has been suggested that the increase in NOS activity in injured neurons and their axons may be involved in neuronal processes such as regeneration leading to survival and degeneration resulting in cell death (Clowry, 1993; Hokfelt et al., 1994). The level of accumulation or production of NOS may be a crucial factor in determining whether injured neurons might regenerate or degenerate (Wu, 1993; Yu, 1994).

It must be pointed out that the results of the present experiments cannot confirm directly the localization of NOS. Moreover, the question remains unanswered of how NOS released by selected peripheral nerves may affect the excitability of each individual nerve. It should be admitted that NOS can act via a cooperative effect with glutamate or by phospholipase C activity (Lukacova et al., 2003).

It is important to note that our results do not imply that only NOS in large-diameter nerve fiber is involved in the process of neurological changes. In the near future, the NO and further free radicals research can surely clarify many of our still unanswered questions.