T-type channel blocking properties and antiabsence activity of two imidazo[1,2-b]pyridazine derivatives structurally related to indomethacin

Abstract

It is presently unclear whether the antiseizure effects exerted by NSAIDs are totally dependent on COX inhibition or not. To clarify this point we investigated whether 7-methyl-2-phenylimidazo[1,2-b]pyridazine-3-carboxylic acid (DM1) and 6-methoxy-2-phenylimidazo[1,2-b]pyridazine-3-carboxylic acid (DM2), two imidazo[1,2-b]pyridazines structurally related to indomethacin (IDM) but ineffective in blocking COXs, retain IDM antiabsence activity. When administered by intraperitoneal injection in WAG/Rij rats, a rat strain which spontaneously develops SWDs, both DM1 and DM2 dose-dependently suppressed the occurrence of these seizures. Importantly, these compounds were both more potent in suppressing SWD occurrence than IDM. As T-type channel blockade is considered a mechanism of action common to many antiabsence drugs we explored by whole cell patch clamp electrophysiology in stably transfected HEK-293 the effect of DM1 and DM2 on Ca_V3.1 channels, the T-type channel subtype preferentially expressed in ventrobasal thalamic nuclei. Both these compounds dose-dependently suppressed the currents elicited by membrane depolarization in these cells. A similar T-type blocking effect was also observed when the cells were exposed to IDM. In conclusion, DM1 and DM2 whilst inactive on COXs, are potent antiabsence drugs. This suggests that compounds with structural features typical of NSAIDs may exert antiepileptic activity independently from COX inhibition and possibly by a direct interaction with T-type voltage-dependent Ca²⁺ channels.

Introduction

There is an increasing awareness that the pharmacological effects of non-steroidal anti-inflammatory drugs (NSAIDs) are in part independent of the inhibition of cyclooxygenases (COXs) (Tegeder et al., 2001a). For instance, NSAIDs affect the activity of nitric oxide synthase (O'Kane et al., 2003), have free radical scavenger properties (De Gaetano et al., 1985, Mehta et al., 1989) and inhibit the kinase activity IKKβ (Yin et al., 1998) thereby repressing NF-κB activity (Kopp and Ghosh, 1994) and, consequently, the differentiation of immune cells (Matasic et al., 2000) and the expression of adhesion molecules (Pierce et al., 1996, Sakurada et al., 1996) and proinflammatory cytokines (Tegeder et al., 2001b). COX-independent NSAID effects have been observed also on a number of other transductional mediators including MAP kinases, STAT/Jak, p90 rsk and AP-1 (reviewed by Tegeder et al., 2001a). In addition, the antineoplastic effects of COX-2 inhibitors are largely independent from COX inhibition and may be explained by the ability of these drugs to induce the expression of NSAID-activated gene 1 (NAG-1), a member of the TGFβ family which induced apoptosis in cancer cells, and to suppress the expression of the antiapoptotic protein survivine 1 and of carbonic anhydrase (as reviewed by Rigas and Kashfi, 2005). Consistently, COX-2 antineoplastic activity can be observed also in tumor cells not expressing COX; furthermore, NSAID derivatives ineffective on COXs, such as sulindac sulfone, may inhibit cell proliferation (Hanif et al., 1996). Finally, members of the NSAID superfamily do affect the activity of ion channels by a direct interaction with channel subunits, as suggested by the evidence that acid sensitive channels (Voilley et al., 2001) are directly blocked by aspirin, diclofenac, and flurbiprofen, high voltage activated Ca²⁺ channels (Zhang et al., 2007) and voltage gated Na⁺ channel (Park et al., 2007) activity is increased by celecoxib, and chloride channels (Liantonio et al., 2007), KCNQ channels (Peretz et al., 2005) and hERG channels (Malykhina et al., 2002) are activated by fenamates.

However, while the role of COX-independent effects in determining the effectiveness of NSAIDs in specific conditions such as pain and cancer is currently under extensive investigation, it remains totally unexplored in other conditions like epilepsy. While, indeed, a number of studies addressed the issue of the role played by arachidonic acid (AA) and prostaglandins (PGs) in various experimental models and found that COX inhibitors decrease seizure frequency and severity in pentylenetetrazole (PTZ)- and kainic acid-induced epilepsy (Dhir and Kulkarni, 2006, Dhir et al., 2006; Tandon et al., 2003), the possible implications of COX-independent mechanisms in determining the efficacy of these drugs has not been explored yet. Structural analogues of classical NSAIDs ineffective on COXs would be powerful instruments to address similar experimental questions by evaluating whether specific pharmacological effects of their parental compounds are retained despite the loss of activity on PG-synthase.

In this perspective, in the present paper we examined the antiepileptic effects of two imidazo[1,2-b]pyridazine derivatives, namely 7-methyl-2-phenylimidazo[1,2-b]pyridazine-3-carboxylic acid (DM1) and 6-methoxy-2-phenylimidazo[1,2-b]pyridazine-3-carboxylic acid (DM2). These molecules belong to a series of heterocyclic compounds, including imidazo[1,2-a]pyridines, imidazo[1,2-a] and [1,2-c]pyrimidines, and imidazo[1,2-a]pyrazines (Abignente, 1991), which were synthesized as derivatives of the common general structure shown in Fig. 1a, designed according to the model proposed by Gund and Shen (1977) for the active site of COX.

Such compounds related to structure a) should have the minimal structural features required to act as a COX inhibitor and indeed most of these derivatives showed more or less significant anti-inflammatory and analgesic activity in vivo (Abignente, 1991). However, some imidazo[1,2-b]pyridazine derivatives (Fig. 1b) including DM1 and DM2 showed peculiar pharmacological properties in vivo, namely potent analgesic activity together with low or negligible anti-inflammatory and gastric ulcerogenic actions (Abignente et al., 1990a, Abignente et al., 1990b, Luraschi et al., 1997). These results suggested that the inhibition of PG biosynthesis played at the most a secondary role in the mode of action of imidazo[1,2-b]pyridazine derivatives.

Here we show that DM1 and DM2 are completely devoid of COX-1 and COX-2 inhibitory activity, but are effective in suppressing spike and wave discharges (SWDs) in WAG/Rij rats, a genetic rodent model of absence epilepsy, similar to what was recently described for their structural congener indomethacin (IDM), which significantly decreased SWDs in these rats (Kovacs et al., 2006). As T-type channel blockade has been considered as an electrophysiological feature common to antiabsence drugs, we also investigated whether DM1 and DM2 COX-independent antiseizure effect could depend on T-type channel blockade and we found that these compounds are indeed powerful T-type channel blockers and that this property is displayed by IDM as well.