Activities of 7-Nitroindazole and 1-(2-(Trifluoromethylphenyl)-imidazole Independent of Neuronal Nitric-Oxide Synthase Inhibition

A large number of published studies have used 7-nitroindazole (NI), a neuronal NO synthase (nNOS) inhibitor, to examine the role of nNOS in both the central and the peripheral nervous system. Most studies using 7-NI to show the involvement of nNOS have obtained results consistent with those in nNOS-deficient mice: renin release (Ren et al., 2001), glutamate neurotoxicity (Ayata et al., 1997), cocaine-induced conditional place preference (Itzhak et al., 1998b), prolactin mRNA synthesis (Chen et al., 2004), ischemic damage (Goyagi et al., 2001), kainic acid-induced seizures (Noh et al., 2006), methamphetamine toxicity (Itzhak et al., 1998a), aggressive behavior (Demas et al., 1999), and a vasoconstriction deficit in septic mice (McKinnon et al., 2006).

However, other studies have raised problems with interpretation. For example, nNOS-deficient mice remain sensitive to 7-NI (Engelhardt et al., 2006). Although 7-NI inhibited carrageenan-induced thermal hyperalgesia in the early phase of inflammation in wild-type mice, the hyperalgesia persisted in the early phase in nNOS-deficient mice (Tao et al., 2004). Purinergic neurotransmission in the rat vas deferens was inhibited by 7-NI, whereas no NOS activity was detected in the same tissue (Allawi et al., 1994) and a very high concentration (1 mM) of l-NAME had no effect on purinergic neurotransmission (Ventura and Burnstock, 1997). 7-NI at a dose that altered neither NOS activity nor the NO concentration inhibited picrotoxin-induced convulsions (Paul and Ekambaram, 2003). There are several possible explanations for these inconsistencies. One would be nonspecific effects unrelated to nNOS inhibition.

In this study, we examined the effects of nNOS inhibitors including 7-NI and TRIM, deletion of the nNOS gene, and other indazole-containing compounds devoid of nNOS inhibitory effects on 2-amino-4-methylphosphinobutyric acid (glufosinate)-induced convulsions, as a model system. Our findings suggest 7-NI and TRIM exert nNOS-unrelated effects.

Materials and Methods

Chemicals and Materials. Glufosinate was purchased from Riedel-de Haen (Seeize, Germany) and PhytoTechnology Laboratories (Shawnee Mission, KS), and N⁵-(1-Iminoethyl)-l-ornithine (l-NIO) was from Cayman Chemical (Ann Arbor, MI). NADPH, l-NAME, TRIM, 7-NI, 6-NI, indazole, imidazole, monoclonal anti-β-actin (clone AC-74), and Dowex AG50WX8-200 H⁺ were from Sigma-Aldrich (St. Louis, MO). l-[³H]Arginine (60 Ci/mmol) was obtained from GE Healthcare (Chalfont St. Giles, UK). A monoclonal anti-nNOS antibody (clone 16) was purchased from BD Biosciences Pharmingen (San Diego, CA).

Animals. Specific pathogen-free 20- to 25-g male ddY mice were purchased from Sankyo Laboratory (Tokyo, Japan). Eight mutant mice (5–6 weeks old) with targeted disruption of the nNOSα gene (B6; 129S4-nos1^tm1Plh/J) and eight wild-type mice (5–6 weeks old; B6129 SF 2/J) were obtained from The Jackson Laboratory (Bar Harbor, ME). The mice were placed in 35-× 25-× 17-cm cages. The cages were kept in a temperature-controlled room under a 12-h light/dark schedule. The experiments conformed to the standards put forth by the Animal Ethics Committee of Teikyo University School of Medicine (Tokyo, Japan) as well as to the Guide for the Care and Use of Laboratory Animals, as adopted and promulgated by the National Institutes of Health (Bethesda, MD).

Glufosinate-Induced Convulsions. Glufosinate (80 mg/kg) was administered (i.p.) to ddY mice. Eighty-two percent of tested mice displayed generalized convulsions within 7 h after the glufosinate challenge (Fig. 3). These generalized convulsions were characterized as tonic and/or clonic in nature, and they gradually increased in severity. The generalized convulsions were monitored for 8 h after glufosinate injection. To evaluate the generalized convulsions, we used the system for scoring of kainate-induced seizures (stage 6) (Morrison et al., 1996).

Drug Treatment. To examine the roles of NOS in glufosinate-induced convulsions, mice received injections with NOS inhibitors, and then they were monitored for 480 min after the glufosinate injection. l-NAME (100 mg/kg; 5 ml/kg i.p.) was administered once at 100 min after glufosinate injection. Five injections of 7-NI (25 mg/kg; 0.153 mmol/kg; 2 ml/kg i.p.) were performed at 100, 160, 220, 280, and 340 min after glufosinate injection. This repeated injection protocol was chosen because of the rapid metabolism of 7-NI in vivo (MacKenzie et al., 1994), the delayed occurrence of glufosinate-induced convulsions, and its successful use by other researchers demonstrating the inhibitory effects of 7-NI in vivo (Lapouble et al., 2002). For a dose-response effect of 7-NI, 7.5 and 75 mg/kg were also used. Other indazole derivatives such as 6-NI and indazole (25 and 18.1 mg/kg, respectively; 0.153 mmol/kg; 2 ml/kg i.p.) were injected by the same protocol as that used for 7-NI. l-NIO (30 mg/kg; 2 ml/kg i.p.) was also administered using this protocol except for the dose, which was chosen in accordance with those in other reports on eNOS inhibition in vivo (Salter et al., 1995). Four injections of TRIM (50 mg/kg; 0.236 mmol/kg; 5 ml/kg i.p.) were performed at 0, 90, 180, and 270 min after glufosinate injection. Imidazole (16 mg/kg; 0.236 mmol/kg; 5 ml/kg i.p.) was injected by the same protocol as that used for TRIM. Glufosinate, l-NAME, and imidazole were dissolved in saline, and TRIM was suspended in saline. 7-NI, 6-NI, and indazole were suspended or dissolved in peanut oil, and control mice received injections with saline or peanut oil.

Tissue Extraction, SDS-PAGE, and Western Blot Analysis. Mice were decapitated at 0, 50, 150, 250, or 350 min after glufosinate injection. The cerebra were removed, they were frozen with a freezing-clamp precooled in liquid nitrogen, and then they were stored at –80°C until SDS-PAGE. Each cerebrum (approximately 0.3 g) was homogenized in 1 ml of 20 mM Tris-HCl, pH 7.6, 1 mM EDTA, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, 10 mM Na₃VO₄, 1 μg/ml pepstatin A, 0.5 μg/ml leupeptin, 0.2 μg/ml aprotinin, and 0.1 μg/ml chymostatin, and then they were centrifuged for 1 h (15,000g) at 4°C. Protein concentrations in the supernatants were determined using a DC Protein Assay kit (Bio-Rad, Hercules, CA). Equal amounts of protein were loaded per lane, and then they were separated by 7.5% SDS-PAGE with transfer to a polyvinylidene difluoride membrane. Blots were probed with anti-nNOS mouse monoclonal antibody (1:1000, clone16; BD Biosciences Pharmingen) or anti-β-actin mouse monoclonal antibody (1:1000, clone AC-74; Sigma-Aldrich) in 5% skim milk overnight at 4°C. Bands were visualized with the Vistra ECF System (GE Healthcare) or ECL Plus Western Blotting Detection System (GE Healthcare), respectively, and quantified using ImageQuant software (GE Healthcare).

NOS Activity Assay in Vivo. NOS activity in vivo was measured by the method of Dwyer et al. (1991). In brief, mice that been treated with anticonvulsive compounds were decapitated 260 or 360 min after the first drug injection, and then their cerebella were removed, frozen rapidly in liquid nitrogen, and stored at –80°C until use. Each cerebellum (approximately 56 mg) was homogenized in 10 volumes (v/w) of 20 mM Tris-HCl buffer, pH 7.4, containing 2 mM EDTA using a Polytron homogenizer (Kinematica, Littau-Lucerne, Switzerland). Homogenates were centrifuged at 10,000g for 15 min at 4°C. The supernatants were used for the NOS activity assay described below. The protein concentration of the supernatant was determined using the DC Protein Assay kit (Bio-Rad). NOS activity was measured by monitoring the conversion of l-[³H]arginine to l-[³H]citrulline. The test tubes contained 25 μl of crude supernatant, 0.5 μCi l-[³H]arginine, 0.75 mM CaCl₂, and 0.5 mM NADPH in a total volume of 100 μl. After a 15-min incubation at 37°C, the reactions were terminated with 3 ml of ice-cold 20 mM HEPES, pH 5.5, with 2 mM EDTA. The reaction mixture was applied to 0.5-ml columns of Dowex AG50WX8-200 (Na⁺ form), and it was eluted with 0.5 ml of water. The Dowex AG50WX8-200 H⁺ form was converted into the Na⁺ form by stirring with a magnetic mixer for 2 h in 2 M sodium hydroxide. Control incubations were carried out using boiled supernatant. l-[³H]Citrulline was quantified by liquid scintillation spectroscopy of flow-through. NOS activity was expressed either as picomoles of citrulline formed per milligram of protein for 15 min or as the percentage of inhibition of enzyme activity compared with animals that received vehicle injections.

Statistical Analysis. The convulsion onset time was plotted according to the Kaplan-Meier method. Statistical differences between the Kaplan-Meier curves for vehicle control and test compounds were analyzed by the generalized Wilcoxon test. Statistical analysis of cerebellar NOS activity was performed by Student's t test. The relationship between NOS inhibition and anticonvulsant activity was analyzed using the Pearson correlation test. A P value of 0.05 was considered statistically significant. Results are expressed as means ± S.E.M. The binomial test was used to evaluate the anticonvulsant effect of various doses of 7-NI.

Results

Brain nNOS Level. Up-regulation of nNOS expression is associated with the onset of some seizures (Chavko et al., 2001; Bagetta et al., 2002; Itoh et al., 2004). To investigate nNOS up-regulation during glufosinate-induced convulsions, the time course of nNOS protein levels was measured after glufosinate (80 mg/kg i.p.) injection (Fig. 1A). nNOS protein levels did not change significantly during the observation period. Although 60 to 80% of mice exhibited generalized convulsions at 350 min after glufosinate injection, nNOS protein was not up-regulated.

Glufosinate-Induced Convulsions in nNOS-Deficient Mice. Glufosinate elicited convulsions in nNOS-deficient mice, as in wild-type mice, and there was no statistically significant difference between wild-type and nNOS-deficient mice (Fig. 1B).

Effects of l-NAME and l-NIO on Glufosinate-Induced Convulsions. To further explore the possible role of nNOS in glufosinate-induced convulsions, other NOS inhibitors, such as the nonselective NOS inhibitor l-NAME, were tested. A single administration of l-NAME (100 mg/kg i.p.) at 100 min after glufosinate (80 mg/kg i.p.) injection did not prevent generalized convulsions (P = 0.421) (Fig. 2A), whereas cerebellar NOS activity in vivo was inhibited by more than 90% at 360 min after l-NAME treatment (Table 2).

Because eNOS has been identified in neurons of some brain areas (Dinerman et al., 1994; O'Dell et al., 1994), we investigated the effect of l-NIO, an eNOS inhibitor (McCall et al., 1991). Because of its rapid metabolism in vivo, l-NIO (30 mg/kg i.p.) was administered five times, with a 60-min-interval starting at 100 min after glufosinate injection. l-NIO did not prevent glufosinate-induced generalized convulsions (P = 0.828) (Fig. 2B), whereas cerebellar NOS activity was inhibited by 80% (Table 1), suggesting that l-NIO reached the brain at concentrations sufficient to inhibit eNOS and/or nNOS. Because eNOS activity in the cerebellum accounts for approximately 30% of total NOS activity (Price and Hanson, 1998), the l-NIO dosage under these experimental conditions inhibited both eNOS and nNOS activities. These results suggest that neither nNOS nor eNOS plays a role in glufosinate-induced generalized convulsions.

Anticonvulsant Effects of Indazole Derivatives. A dose-dependent anticonvulsant effect of 7-NI was observed, with the first injection at 100 min after glufosinate injection (Table 1). Therefore, the dose of 25 mg/kg (0.153 mmol/kg) was considered to be suitable for comparison between the effect of 7-NI and that of other indazole analogs on glufosinate-induced generalized convulsions. We examined the structure-activity relationship of indazole derivatives in NOS inhibition and anticonvulsant activity. The drugs were injected using the same protocol (0.153 mmol/kg i.p.) as for 7-NI. 6-NI, 7-NI, and indazole significantly reduced glufosinate-induced generalized convulsions compared with the vehicle group (Fig. 3). At 360 min after glufosinate injection, mice treated with 7-NI, 6-NI, or indazole exhibited a significantly lower incidence of convulsions (26.6, 20.0, and 33.3%, respectively) compared with vehicle-treated mice (58.8%). At this time point, normalized anticonvulsant activities (% inhibition, with vehicle taken as 100%) of 7-NI, 6-NI, and indazole were 55, 66, and 43%, respectively (Table 2).

Effect of TRIM on Glufosinate-Induced Convulsions. TRIM, which acts on nNOS and iNOS in vitro (Handy et al., 1996), was tested as an anticonvulsant. When administered four times at a dose of 50 mg/kg, TRIM completely prevented convulsions (P < 0.05) (Table 2). Because TRIM is an imidazole derivative, the imidazole nucleus itself was tested for its capacity to inhibit glufosinate-induced convulsions. Imidazole failed to inhibit these convulsions (data not shown).

Relation between NOS Inhibition and Anticonvulsive Effects. To investigate whether nNOS inhibition correlates with anticonvulsant activity, the NOS inhibitory activities of indazole analogs, TRIM, and arginine analog NOS inhibitors were evaluated in vivo using an ex vivo cerebellum NOS assay. Owing to the high activity of nNOS in the cerebellum, this brain area was used to study the effects of these compounds on NOS activity. As shown in Fig. 4, there was no correlation between NOS inhibition and anticonvulsant activity (r = 0.675, P = 0.835, Pearson correlation coefficient).

In Table 2, results are grouped according to chemical structures: indazole, phenyl imidazole, and arginine analogs. The anticonvulsant effects of indazole analogs showed no correlation with NOS inhibitory activity. It is surprising to note that none of the arginine derivatives tested was effective against convulsions, despite their NOS inhibitory activities.

Discussion

Glufosinate ammonium, a broad-spectrum herbicide, causes generalized seizures characterized by tonic-clonic convulsions in rodents and humans (Hack et al., 1994). This poisoning is characterized by various neurological symptoms such as disturbances of consciousness and convulsions, which occur after an asymptomatic interval of several hours. As yet, no specific detoxifying drug for this poisoning has been developed.

nNOS expression is up-regulated in pentylenetetrazole-kindling (Itoh et al., 2004), tacrine-induced limbic seizures (Bagetta et al., 2002), and hyperbaric oxygen seizures (Chavko et al., 2001). We did not observe nNOS up-regulation during glufosinate convulsions, showing that these convulsions are not caused by elevated nNOS expression and that the underlying mechanisms thus differ from those of some other types of convulsions (Chavko et al., 2001; Bagetta et al., 2002; Itoh et al., 2004).

We previously showed glufosinate-induced generalized convulsions to be inhibited by N-methyl-d-aspartate receptor antagonists such as MK-801 and LY 235959 (Matsumura et al., 2001). Moreover, an electroencephalographic study by Lapouble et al. (2002) confirmed our result and also showed glufosinate-induced convulsions to be inhibited by 7-NI, suggesting nNOS involvement in these convulsions. In this study, we confirmed the inhibitory action of 7-NI on glufosinate-induced convulsions.

We showed that 7-NI, an indazole derivative and selective inhibitor of nNOS, inhibited both NOS activity in vivo and glufosinate-induced convulsions. Another nNOS inhibitor, TRIM, yielded results similar to those obtained with 7-NI. l-NIO, a selective inhibitor of eNOS, significantly inhibited NOS activity in vivo, while failing to inhibit glufosinate-induced convulsions. These results suggest the involvement of nNOS in glufosinate-induced convulsions.

However, this interpretation raises several problems. First, l-NAME, a nonselective NOS inhibitor, inhibited NOS activity by 93% in vivo, but it had no effect on the convulsions. Second, in nNOSα-deficient mice, glufosinate elicited convulsions at the same dose as that used in wild-type mice. Third, the order of potency, according to IC₅₀ concentrations for NOS inhibition in a previous report (Babbedge et al., 1993), is 7-NI (0.9 μM) > 6-NI (31.6 μM) > indazole (177.8 μM). In the present study, the order of NOS inhibition magnitude in vivo was similar. However, these compounds significantly prevented and/or inhibited glufosinate-induced convulsions, to an extent similar to that obtained with 7-NI.

The nNOSβ isoform is reported to be slightly up-regulated in some brain regions of nNOSα-deficient mice (Huang et al., 1993), and the residual isoform corresponds to 5% of NOS activity in the brain. Therefore, it is possible that a small amount of the nNOSβ isoform is involved in glufosinate-induced convulsions. However, in view of the lack of an l-NAME effect on the convulsions observed in this study, we may be able to rule out this possibility, because up-regulated nNOSβ should have a greater sensitivity to l-NAME (Hurt et al., 2006). Therefore, we favor the interpretation that nNOS activity is not involved in glufosinate-induced convulsions.

TRIM, an imidazole derivative, showed the strongest anti-convulsant activity, whereas TRIM had much less effect on nNOS activity than 7-NI, in agreement with other reports of in vivo results (Handy et al., 1995, 1996). Moreover, TRIM exhibited more potent antinociceptive activity and less potent nNOS inhibition than l-NAME and 7-NI in vitro (Moore et al., 1993; Handy et al., 1996). 7-NI but not l-NAME inhibits pilocarpine-induced seizures and electroconvulsions (Przegaliñski et al., 1996; Baran et al., 1997). These previously reported profiles of NOS inhibitors are similar to that of glufosinate-induced convulsions. These results suggest that the anticonvulsant activities of 7-NI, 6-NI, indazole, and TRIM are attributable to an unknown mechanism that does not involve NOS inhibition.

Administration of imidazole has an antiwrithing analgesic effect, with an ED₅₀ of 23 mg/kg p.o., indicating that the dose (16 mg/kg i.p., four times) used in the current study was sufficient for imidazole to reach the brain. However, imidazole did not inhibit glufosinate-induced convulsions (unpublished data). Therefore, the active structural moiety of TRIM may not be the imidazole nucleus. It is noteworthy that these compounds, which are effective against glufosinate-induced convulsions, have chemical structures common to indazim (Tyacke et al., 2003) and norharmane (Anderson et al., 2003), both of which are imidazoline receptor ligands. Considering the anticonvulsive property and monoamine oxidase inhibition common to 7-NI (Desvignes et al., 1999) and harmane (Anderson et al., 2003), it is tempting to speculate that 7-NI may bind with imidazoline receptors.

We conclude that TRIM and indazole analogs, including 7-NI, 6-NI, and indazole itself, have anticonvulsant activities independent of nNOS inhibition. In addition, we propose that the effects of both l-NAME as a positive control and indazole as a negative control should be tested in a given system to appropriately interpret the pharmacological data obtained with 7-NI and TRIM.