Gabapentin脊髓腔止痛作用機制之研究

Abstract

Gabapentin為一個新一代抗癲癇藥物，其化學結構式為模仿抑制性神經傳導物質gamma-aminobutyric acid (GABA)。Gabapentin除了具抗癲癇的作用，在動物及臨床實驗中也被發現具有止痛作用。雖然不少理論被提出，但gabapentin的作用機轉迄今尚未有定論。在早期的agonist-receptor binding study中，gabapentin並無法被證實為GABAA或GABAB受體的agonist。最近Ng等人指出gabapentin為一GABAB1a 與GABAB2 二種次單元組成的GABAB受體agonist。另外，NMDA受體與KATP鉀離子通道的活化也可能與gabapentin的作用有關。在中樞神經系統，gabapentin被發現與鈣離子通道的alpha2delta subunit有特定結合，且此結合可被magnesium chloride、ruthenium red與spermine影響。在脊髓中存在有L, P/Q, N及T型鈣離子通道，這些鈣離子通道皆具有alpha2delta subunit，其中T型鈣離子通道與疼痛傳導路徑的突觸可塑性(synaptic plasticity)有關，此突觸可塑性可發生於發炎性疼痛引起的中樞過敏化(central sensitization)。 我們在老鼠術後疼痛模式中發現在脊髓腔中給予gabapentin可以產生止痛作用，利用此一模式，我們測試 一) Gabapentin是否為一GABAB 受體的agonist；二) NMDA受體及KATP 鉀離子通道的活化是否與gabapentin的止痛作用有關； 三) magnesium chloride、ruthenium red或spermin是否會影響gabapentin的止痛作用；四)老鼠腰椎的alpha2delta-1 subunit蛋白質表現量是否有改變；五) 各種鈣離子通道阻斷劑的止痛作用。最後，我們測試脊髓腔T型鈣離子通道阻斷劑在福馬林疼痛模式的止痛作用。 在術後疼痛模式，於isoflurane麻醉下，我們在帶有脊髓腔導管的大白鼠(250-300 公克)右腳掌製造一公分傷口，以von Frey filament分別在術前，術後兩小時及脊髓腔給藥後15, 30, 45, 60, 90及120分鐘測試傷口旁縮腿閾值(Withdrawal threshold)。藥物的止痛效果以最大可能效果百分比(Percent of maximal possible effect: %MPE)表示。相關受體及離子通道(GABAA, GABAB and NMDA 受體及 KATP 離子通道) 的抑制劑在給gabapentin前5或10分鐘經脊髓腔給予. 在福馬林疼痛模式，帶有脊髓腔導管的雄性大白鼠(250-300 公克)接受右腳掌皮下打入50 microL 5% 福馬林。在打入福馬林10分鐘前，分別經脊髓腔導管給予生理食鹽水或T型鈣離子通道阻斷劑。在打入福馬林後0分鐘到9分鐘(第一期)及10分鐘到60分鐘(第二期)，老鼠會產生抖腳或縮腳的動作。我們在打入福馬林後1分鐘，5分鐘及每隔5分鐘(共計60分鐘)，計算1分鐘内老鼠的抖腳或縮腳次數。另外，在打入福馬林後，老鼠會產生舔腳或咬腳的動作。在打入福馬林後，我們同時計算在打入福馬林後0–5分鐘(前期)及20–40分鐘(後期)，老鼠的舔腳或咬腳的累計時間。 在術後疼痛模式中，gabapentin (30-200 microg)具止痛作用。GABAB 受體的拮抗劑 CGP 35348 (60 microg) 與CGP 55845 (3 microg) 無法反轉gabapentin (100 microg) 的止痛作用，但可反轉GABAB 受體agonist baclofen的止痛作用。NMDA 受體拮抗劑 APV (6 microg) 或 MK-801 (14 microg)，GABAA 受體拮抗劑 bicuculline (0.3 microg)與 KATP 離子通道拮抗劑 glibenclamide (300 microg)並不影響gabapentin的止痛作用。GABAA 受體的agonist isoguvacine (20 microg)，KATP 離子通道的opener pinacidil (300 microg)與diazoxide (1200 microg)本身並不具止痛作用。 在脊髓腔中同時給予magnesium chloride (5-20 microg) 或ruthenium red (0.2-20 ng) 可以減緩gabapentin (30-200 microg) 的止痛作用，但spermine在不引起運動障礙的劑量下並不影響gabapentin的止痛作用。Magnesium chloride及ruthenium red並不影響morphine的止痛作用。另外，我們也發現在此一術後疼痛模式中，老鼠腰椎的alpha2delta-1 subunit蛋白質表現量在劃刀受傷後有增加的趨勢。 在脊髓腔給予N型鈣離子通道阻斷劑(omega-conotoxin GIVA 0.1-3 microg)具止痛作用。然而，P/Q 型鈣離子通道阻斷劑 (omega-agatoxin IVA 0.1-0.3 microg)，L型鈣離子通道阻斷劑(verapamil 300 microg, diltiazem 500 microg, nimodipine 500 microg) 與T 型鈣離子通道阻斷劑(mibefradil 600 microg)皆無止痛作用。另一方面，脊髓腔内注射T型鈣離子通道阻斷劑mibefradil (50-500 microg)或nickel (1-10 microg)，在福馬林疼痛模式中的第一期與第二期皆具有止痛作用﹔然而ethosuximide (100-1200 microg) 並不具止痛效果。 這些結果支持在老鼠術後疼痛模式中，gabapentin的脊髓腔止痛作用與N型鈣離子通道alpha2delta subunit的結合有關，而GABAB 受體、NMDA 受體與KATP離子通道可能不參與其中。另外，T型鈣離子通道的活化可能與福馬林引起的疼痛行為有關。

Gabapentin, a gamma-aminobutyric acid (GABA) analogue anticonvulsant, possesses antinociceptive effects in several animal pain models and clinical trials. Up to now, the action mechanism(s) of gabapentin remains unknown and a variety of hypotheses have been proposed. Gabapentin was initially shown to have little affinity at GABAA or GABAB receptors but was recently reported to be a selective agonist at the GABAB receptor consisting of GABAB1a-GABAB2 heterodimers. In addition, activation of NMDA receptors and ATP-sensitive K+ (KATP) channels were also reported to be involved in the cellular actions of gabapentin. A specific binding site of [3H]gabapentin in the brain has been reported to be the alpha2delta subunit of voltage-dependent Ca2+ channels, at which the binding of gabapentin was modulated by magnesium chloride, ruthenium red and spermine. In this study, we validated if these proposed cellular mechanisms contribute to the spinal analgesic effect of gabapentin using a rat model of postoperative pain. The alpha2delta subunit is present in several types (L-, P/Q-, N-, T-type) of Ca2+ channels, which are all distributed in the spinal cord. Among all types of Ca2+ channels, T-type Ca2+ channels have been implicated in the synaptic plasticity at synapses from nociceptive nerve fibers, which might be involved in the central sensitization of inflammatory pain. We have examined the following in the postoperative pain model: 1) if gabapentin acts as an GABAB receptor agonist to achieve it antiallodynic effect; 2) if activation of NMDA receptors or KATP channels are involved in the antiallodynic effect of intrathecal gabapentin; 3) if magnisium chloride, ruthenium red or spermine could affect the antiallodynic effect of intrathecal gabapentin; 4) if the expression of spinal alpha2delta-1 subunit of Ca2+ channels was altered; 5) which type of Ca2+ channel blockers would mimic the antiallodynic effects of intrathecal gabapentin. Finally, the possible antinociceptive effects of intrathecal T-type Ca2+ channel blockers were examined in the rat formalin test. The postoperative pain model was conducted in male Sprague-Dawley rats (250-300 gm). Under isoflurane anesthesia, the rats with chronic intrathecal catheters received an incision over plantar surface of right hindpaw to produce punctate mechanical allodynia. Withdrawal thresholds to von Frey filament stimulation near the incision site were measured before, 2 h after incision and 15, 30, 45, 60, 90 and 120 min after intrathecal drug administrations. Receptor or channel (GABAA, GABAB and NMDA receptors and KATP channel) ligands were intrathecally pretreated 5 or 10 min before gabapentin was administered. The formalin test was conducted by giving subcutaneous injection of 5% formalin 50 microL into rat (Sprague-Dawley, male, 250-300 gm) right hindpaw. T-type Ca2+ channel blockers were given intrathecally 10 min before formalin injection. Biphasic flinching or shaking of the injected paw were observed after formalin injection. Phase 1 and 2 were defined as 0–9 and 10–60 min after formalin injection, respectively. The number of flinches was counted for 1-min periods at 1 and 5 min and at 5-min intervals from 10 to 60 min. The biting and licking time was also monitored during 0–5 and 20–40 min after formalin injection. Intrathecal injection of gabapentin (30-200 microg) dose-dependently reduced incision-induced allodynia. GABAB receptor antagonists, CGP 35348 and CGP 55845, did not affect gabapentin-induced antiallodynic effect at doses effective in antagonizing the antiallodynic effect of baclofen, a GABAB receptor agonist. Intrathecal pretreatment with NMDA receptor antagonists, APV (6 microg) and MK-801 (14 microg), GABAA receptor antagonist, bicuculline (0.3 microg), and KATP channel blocker, glibenclamide (300 microg), did not attenuate the antiallodynic effect of gabapentin. The GABAA receptor agonist, isoguvacine (20 microg), and KATP channel openers, pinacidil (100-300 microg) and diazoxide (600-1200 microg), per se, had little effect on the postincision allodynic response. The antiallodynic effect of gabapentin was inhibited by the alpha2delta subunit binding modulators, MgCl2 (5-20 microg) and ruthenium red (0.2-20 ng), non-competitively, but not by spermine at doses without inducing motor weakness. On the other hand, the antiallodynic effect of intrathecal morphine was unaffected by either alpha2delta subunit binding modulators. In addition, an upregulation of the alpha2delta-1 subunit of Ca2+ channels in the L4-6 dorsal spinal cord was observed in this postoperative pain model, similar to the findings in other gabapentin-sensitive animal pain models. Intrathecal omega-conotoxin GIVA (0.1-3 microg), an N-type Ca2+ channel blocker, like gabapentin, attenuated the allodynic response induced by paw incision. However, P/Q-type (omega-agatoxin IVA 0.1-0.3 microg), L-type (verapamil 300 microg, diltiazem 500 microg and nimodipine 500 microg) and T-type (mibefradil 600 microg) Ca2+ channel blockers had little effect on the postincisional allodynia. On the other hand, intrathecal injection of T-type Ca2+ channel blocker, mibefradil (50-500 microg) or nickel (1-10 microg), but not ethosuximide (100-1200 microg) dose-dependently inhibited the nociceptive behaviors in phase 1 and 2 of the formalin test. In conclusion, our results provide behavioral evidences to support that the alpha2delta subunit of N-type Ca2+ channels, but not GABAB receptors, NMDA receptors and KATP channels, may be involved in the antiallodynic action of intrathecal gabapentin in the postoperative pain model. In addition, activation of spinal T-type Ca2+ channels might be involved in formalin-induced nociceptive behaviors.