Marijuana exerts its action by binding its main active component Δ(9-)tetrahydrocannabinol to cannabinoid (CB) receptors. They are seven-transmembrane-domain receptors coupled to G(i/o) protein and consist of two isoforms, CB1 and CB2. While the CB1 receptor is dominantly expressed in the brain, the CB2 receptor is expressed mainly in the immune system. Therefore, it is assumed that the CB1 receptor is responsible for most of the psychoactivity of cannabis. N-arachidonylethanolamide (anandamide) and 2-arachidonylglycerol (2-AG) have been identified as major endocannabinoids, that is, the endogenous ligands for CB receptors.

A major physiological role of endocannabinoids is to mediate retrograde signaling at synapses. Endocannabinoids are released from postsynaptic neurons, travel retrogradely across the synaptic cleft, activate CB1 receptors on presynaptic terminals, and suppress neurotransmitter release. Both short-term and long-term depression can be induced by endocannabinoids depending on the pattern and duration of CB1 receptor activation. The endocannabinoid-mediated retrograde signaling has been found in various brain regions and is now regarded as an important and fundamental mechanism of synaptic modulation.

Endocannabinoid production and release are induced by three distinct modes of activity in postsynaptic neurons: 1) strong depolarization of postsynaptic neurons and the resultant elevation of Ca(2+) concentration to micromolar range, 2) strong activation of G(q/11)-coupled receptors at basal intracellular Ca(2+) level and 3) combination of postsynaptic Ca(2+) elevation with G(q/11)-coupled receptor activation. Between the two major endocannabinoids, results from pharmacological studies suggest that 2-AG rather than anandamide mediates retrograde signaling. This is based on the results that endocannabinoid-mediated retrograde suppression was prolonged by inhibitors of monoacylglycerol lipase (MGL), a major 2-AG hydrolyzing enzyme, and that the suppression was abolished by inhibitors of diacylglycerol lipase (DGL), the enzyme that synthesize 2-AG. However the inhibitors used in previous studies are not specific for MGL or DGL. Moreover, there are two isoforms of DGL, DGLα and DGLβ, and inhibitors do not discriminate the two DGL isoforms.

In the first part of my thesis work, I examined whether 2-AG mediates retrograde signaling and, if so, relative contribution of the two DGL isoforms to 2-AG synthesis. I used DGLα and DGLβ knockout mice that were generated in collaboration with Prof. Sakimura's laboratory (Niigata University). Basal amounts of 2-AG in the cerebellum, hippocampus and striatum were markedly reduced in DGLα knockout mice, whereas those were normal in DGLβ knockout mice. These results indicate that DGLα is the major 2-AG synthesizing enzyme in the brain. Then I examined whether endocannabinoid-mediated retrograde suppression was affected in the cerebellum, hippocampus and striatum of the two strains of knockout mice. I made whole-cell recordings from Purkinje cells in cerebellar slices, from pyramidal cells in hippocampal slices, from neurons in dissociated hippocampal cultures and from medium spiny neurons in striatal slices. I found that transient suppression of excitatory synaptic currents (EPSCs) or inhibitory synaptic currents (IPSCs) induced by strong depolarization of postsynaptic neurons, by strong G(q/11)-coupled receptor activation, and by combined depolarization with G(q/11)-coupled receptor activation were all absent in the three brain areas of DGLα knockout mice. Long-term depression was also absent in the hippocampus of DGLα knockout mice. By marked contrast, transient suppression of EPSC or IPSC induced by the three modes of endocannabinoid release was normal in the cerebellum and hippocampus of DGLβ knockout mice. These results clearly indicate that 2-AG produced by DGLα, not by DGLβ, mediates retrograde signaling at central synapses.

In the second part of my thesis work, I examined how 2-AG is terminated after causing retrograde suppression of transmitter release. Most 2-AG is known to be hydrolyzed into arachidonic acid and glycerol by MGL that is enriched in presynaptic terminals. For this purpose, I focused on the cerebellar cortex that contains four distinct types of cannabinoid sensitive synapses. I used global MGL knockout mice and cerebellar granule cell (GC)-specific MGL knockout mice that were generated also in collaboration with Prof. Sakimura's laboratory (Niigata University). First, I examined localization of MGL in the cerebellar cortex of wild-type mice by immunofluorescence microscopic and immunogold electron microscopic analyses. MGL was expressed richly in terminals of parallel fibers (PFs), the axons of GCs, and weakly in Bergman glia, but was absent in climbing fiber (CF) terminals and GABAergic inhibitory terminals of basket cell (BC) and stellate cell (SC). Then I made whole-cell recordings from PCs in cerebellar slices and examined depolarization-induced retrograde suppression at the four types of synapses. Despite the highly selective MGL expression pattern, depolarization-induced 2-AG-mediated retrograde suppression was significantly prolonged at not only PF- PC synapse but also CF-PC synapse in GC-specific MGL knockout mice whose MGL expression in PF terminals was eliminated and MGL was present only in Bergmann glia. Moreover, 2-AG-mediated signaling at both PF-PC and CF-PC synapses was significantly shorter in GC-specific MGL-KO mice compared with global MGL-KO mice. These results indicate that MGL in PF terminals facilitates termination of 2-AG signaling not only "homosynaptically" at PFs but also "heterosynaptically" at CFs, and that MGL in Bergmann glia also contributes to termination of 2-AG signaling. To further examine the role of MGL in Bergmann glia, I prepared organotypic cerebellar slice cultures from global MGL-KO mice and introduced MGL into Bergmann glia by using a lentivirus vector. As expected, depolarization-induced retrograde suppression at PF-PC synapses was significantly shorter in slice cultures with MGL over-expression into Bergmann glia than in control cultures with GFP expression alone. Finally, I compared depolarization-induced retrograde suppression at SC-PC synapses and that at BC-PC synapses. Retrograde suppression was significantly prolonged at SC-PC synapses that are surrounded by PFs and located close to Bergmann glial processes, but not at BC-PC synapses that are remote from MGL-rich PFs. These results indicate taht 2-AG is degraded in a synapse type-independent manner by MGL present in PFs and Bergmann glia. The results of the present study strongly suggest that MGL regulates 2-AG signaling rather broadly within a certain range of neural tissue, although MGL expression is heterogeneous and limited to a subset of nerve terminals and astrocytes.

本研究は、(1)中枢神経系において逆行性シナプス伝達抑圧を担う内因性カンナビノイドを同定するために、内因性カンナビノイドの一つである2-アラキドノイルグリセロール(2-AG)の産生酵素、ジアシルグリセロールリパーゼ(DGL)の遺伝子改変マウス(DGLαノックアウトマウス、DGLβノックアウトマウス)の解析と、(2)小脳において、2-AG分解酵素であるモノアシルグリセロールリパーゼ(MGL)による逆行性シナプス伝達抑圧の制御機構を明らかにするために、MGL遺伝子改変マウスの解析を行ったものであり、下記の結果を得ている。

(1)

1、HPLC法によって小脳、海馬、線条体の2-AG量を計測した結果、DGLαノックアウトマウスでは、野生型に較べて2-AG量が減少しているのに対し、DGLβノックアウトマウスでは、野生型と差がなかった。この事から、2-AGの産生に関与する分子はDGLαであることが分かった。

2、電気生理学的解析によって、内因性カンナビノイドによるシナプス修飾について調べた。その結果、小脳、海馬、線条体において、DGLαノックアウトマウスでは、野生型で逆行性シナプス伝達抑圧が誘発される刺激を加えても、逆行性シナプス伝達抑圧は観察されなかった。

3、2と同様にDGLβノックアウトマウスについて逆行性シナプス伝達抑圧が観察されるかを調べた。その結果、DGLβノックアウトマウスでは、野生型と同様に逆行性シナプス伝達抑圧が誘発された。2、3の結果から、逆行性シナプス伝達抑圧を担う分子は、DGLαによって産生される2-AGであることが分かった。

(2)

1、免疫染色、免疫電子顕微鏡による解析により、小脳皮質において、MGLは平行線維終末に最も強く、またバーグマングリアに弱く発現しており、登上線維終末や抑制性終末には発現していないことが分かった。

2、顆粒細胞特異的MGLノックアウトマウスの電気生理学的解析により、顆粒細胞特異的MGLノックアウトマウスでは、登上線維シナプスにおける逆行性シナプス伝達抑圧が野生型よりも遷延していることが分かった。このことから、登上線維シナプスにおける逆行性シナプス伝達抑圧に、平行線維終末に局在するMGLが関与することが分かった。

3、小脳スライス培養の系を用いて、レンチウイルスによってMGLノックアウトマウスのバーグマングリアにMGLを強制発現させ、逆行性シナプス伝達抑圧の持続時間を調べた。その結果、バーグマングリアにMGLを強制発現させたスライスでは、逆行性シナプス伝達抑圧の持続時間が短くなっていることが分かった。このことから、逆行性シナプス伝達抑圧が起きているシナプス周囲に存在するMGLがシナプス非選択的に逆行性シナプス伝達抑圧を調節することが分かった。

4、抑制性シナプスには、小脳プルキンエ細胞の遠位樹状突起にシナプスを形成するもの(分子層でシナプスを形成する)と細胞体にシナプスを形成するもの(プルキンエ細胞層でシナプスを形成する)がある。MGLは平行線維終末に強く発現していることから、分子層に最も強くMGLは存在する。シナプスを形成している部位周囲のMGLの有無によって、逆行性シナプス伝達抑圧は影響を受けるかについて電気生理学的解析を用いて調べた。その結果、分子層に形成されるシナプスで起きる逆行性シナプス伝達抑圧は、MGLノックアウトマウスで野生型よりも遷延しているのに対し、細胞体に形成されるシナプスで起きる逆行性シナプス伝達抑圧は野生型とMGLノックアウトマウス間で差がなかった。

2、3、4の結果から、小脳において逆行性シナプス伝達抑圧はMGLによる2-AGのシナプス非選択的な分解によって調節されることが分かった。

以上、本研究によって、逆行性シナプス伝達抑圧を担う分子が2-AGであること、また逆行性シナプス伝達抑圧は、MGLによるシナプス非選択的な2-AGの分解によって調節されることが明らかになった。本研究は、内因性カンナビノイドによるシナプス修飾機構の解明に大きな貢献をなすものであり、学位の授与に値するものと考えられる。