Voltage-gated Cav2.1 (P/Q-type) and Cav2.2 (N-type) channels are two closely related calcium channels that play indispensable roles in signal transduction pathways by regulating neurotransmitter release. Despite having highly conserved amino acid sequences, they are differentially modulated by calmodulin, which mediate two important feedback mechanisms known as Ca2+-dependent inactivation (CDI) and Ca2+-dependent facilitation (CDF). These dual regulatory mechanisms contribute to synaptic plasticity, but only CDI is observed in Cav2.2 channel, while both CDI and CDF are present in Cav2.1 channel. Previously, it was hypothesized that the lack of CDF in Cav2.2 channel is due to the pre-IQ-IQ domain of the channel’s lower binding affinity for calmodulin compared to that of Cav2.1 channel. Now that the EF-hand domain of calcium channels is identified as one of the two minimally required molecular determinants that are responsible for supporting CDF in Cav2.1 channel and preventing CDF in Cav2.2 channel, it was necessary to determine the role of EF-hand domain in calmodulin binding of Cav2.1 and Cav2.2 channels. Using pull-down binding assays, this study finds that the EF-hand domain enhances calmodulin binding to the proximal C-terminal domain of Cav2.2 channel, which suggests that the lack of CDF in Cav2.2 does not result from the channel’s weak interaction with CaM, but from the EF-pre-IQ-IQ domain of the channel’s inability to allow calmodulin from fully exerting its effects.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第22124号 / 工博第4654号 / 新制||工||1726(附属図書館) / 京都大学大学院工学研究科合成・生物化学専攻 / (主査)教授 森 泰生, 教授 跡見 晴幸, 教授 浜地 格 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DGAM

The Rem, Rem2, Rad, and Gem/Kir GTPases, comprise a novel subfamily of the small Ras-related GTP-binding proteins known as the RGK GTPases, and have been shown to function as potent negative regulators of high voltage-activated (HVA) Ca2+ channels upon overexpression. HVA Ca2+ channels modulate Ca2+ influx in response to membrane depolarization to regulate a wide variety of cellular functions and they minimally consist of a pore-forming α1 subunit, an intracellular β subunit, and a transmembrane complex α2/δ subunit. While the mechanisms underlying RGK-mediated Ca2+ channel regulation remain poorly defined, it appears that both membrane localization and the binding of accessory Ca2+ channel β subunits (CaVβ) are required for suppression of Ca2+ channel currents. We identified a direct interaction between Rem and the L-type Cavα1 C-terminus (CCT), but not the CCT from CaV3.2 T-type channels. Deletion mapping studies suggest that the conserved CB-IQ domain is required for Rem:CCT association, a region known to contribute to both Ca2+-dependent channel inactivation and facilitation through interactions of Ca2+-bound calmodulin (CaM) with the proximal CCT. Furthermore, both Rem2 and Rad GTPases display similar patterns of CCT binding, suggesting that CCT represents a common binding partner for all RGK proteins. While previous studies have found that association of the Rem C-terminus with the plasma membrane is required for channel inhibition, it is not required for CaVβ- subunit binding. However, Rem:CCT association is well correlated with the plasma membrane localization of Rem and more importantly, Rem-mediated channel inhibition upon overexpression. Moreover, co-expression of the proximal CB-IQ containing region of CCT (residues 1507-1669) in HIT-T15 cells partially relieves Rem blockade of ionic current. Interestingly, Ca2+/CaM disrupts Rem:CCT association in vitro. Moreover, CaM overexpression partially relieves Rem-mediated L-type Ca2+ channel inhibition and Rem overexpression alters the kinetics of calcium-dependent inactivation. Together, these data suggest that the association of Rem with the CCT represents a crucial molecular determinant for Rem-mediated L-type Ca2+ channel regulation and provides new insights into this novel channel regulatory process. These studies also suggest that instead of acting as complete Ca2+ channel blockers, RGK proteins may function as endogenous regulators for the channel inactivation machinery.

No description available.

Calmodulin (CaM) is a ubiquitous calcium-binding protein responsible for the binding and activation of a vast number of enzymes and signaling pathways. It contains two lobes that bind two calcium ions each, separated by a flexible central linker. This structural flexibility allows CaM to bind and regulate a large number of diverse protein targets within the cell in response to Ca2+ gradients. Voltage gated calcium channels (CaVs), as main sources of extracellular Ca2+, are crucial for a number of physiological processes, from muscle contraction to neurotransmission and endocrine function. These large transmembrane proteins open in response to membrane depolarization and allow gated entry of Ca2+ ions into the cytoplasm. Their regulation is currently the subject of intense investigation due to its pharmacological and scientific importance. CaM has been previously shown to pre-associate and act as a potent inhibitor of one class of high-voltage activated (HVA) channels called L-type channels via its interaction with their C-terminal cytoplasmic region. This interaction is primarily mediated by a conserved CaM-binding motif called the ‘IQ’ motif (for conserved isoleucine and glutamine residues), although the exact molecular details of its involvement in inactivation are currently unclear. Elucidation of these details was the primary objective of this dissertation. Recently, a novel sequence motif within this channel called ‘NSCaTE’ (N-terminal spatial calcium transforming element) has been described as an important contributor to calcium-dependent inactivation (CDI) of L-type channels. It was presumed to be unique to vertebrates, but we also show its conservation in a distantly related L-type channel homolog of Lymnaea stagnalis (pond snail). The interaction of CaM with a number of peptides representing the different regulatory motifs (IQ and NSCaTE) for both mammalian and snail isoforms was characterized in an attempt to better understand their role in CDI. Biophysical work with peptides as well as electrophysiology recordings with an N-terminal truncation mutant of Lymnaea CaV1 homolog were performed to expand our understanding of how the interplay between these channel elements might occur. In brief, the most striking feature of the interaction concerns the strong evidence for a CaM-mediated bridge between the N- and C-terminal elements of L-type channels. Further investigation of the CaM interaction with both IQ and NSCaTE peptides using Ca2+-deficient CaM mutants reveals a preference of both peptides for the Ca2+-C-lobe of CaM, and a much higher affinity of CaM for the IQ peptide, suggesting that the N-lobe of CaM is the main interaction responsible for the physiological effects of NSCaTE. These results are consistent with our electrophysiology findings that reveal a distinct buffer-sensitive CDI in wild type LCaV1 that can be abolished by the N-terminal truncation spanning the NSCaTE region. In addition to L-type channels, CaM has also been shown to have an indirect role in the regulation of low-voltage activated (LVA) or T-type channels (CaV3.x), via their phosphorylation by CaM-dependent protein kinase II (CaMKII). Using a primary sequence scanning algorithm, a CaM-binding site was predicted in a cytoplasmic region of these channels that was also previously shown to be important in channel gating. Biophysical experiments with synthetic peptides spanning this gating brake region from the three human and the single Lymnaea isoform strongly suggest that there is a novel, bona fide CaM interaction in this channel region, and also hint that this interaction may be a Ca2+-dependent switch of some sort. The results confirm a possible new role for CaM in the direct regulation of these channels, although the exact mechanism remains to be elucidated.