Spine-neck geometry determines NMDA receptor-dependent Ca2+ signaling in dendrites

NMDA receptors mediate increases in the intracellular free Ca2+ concentration （［Ca2+］i） of neurons that lead to bidirectional regulation of synaptic plasticity, which supports self-organization of neuronal networks in the central nervous system. Synaptic functions and plasticities have been proposed to be dependent on synaptic structure. With the use of two-photon photolysis of caged glutamate, Matsuzaki et al. previously showed that spine-head volume is an important determinant of the expression of AMPA receptors. The same approach was recently applied to induce NMDA receptor （NMDAR）-dependent long-term plasticity at visually selected single spines； this plasticity was dependent on spine-head volume and was induced more efficiently in smaller spines than in larger ones. However, the dependence of NMDAR-mediated Ca2+ signaling on spine structure has not been clarified, given that selective stimulation of single presynaptic fibers that innervate identified spines has been difficult to achieve electrically because of the spread of the electric current. The role of diverse spine-neck structures in the regulation of neuronal signaling is also unclear. In this study, we combined two-photon photolysis of a caged glutamate with two-photon Ca2+ imaging and whole-cell patch-clamp recording, to monitor the activation of NMDARs in neurons with a high spatial resolution. In acute hippocampal slices, we imaged CA1 pyramidal neurons clamped in the whole-cell mode with a solution containing the low-affinity Ca2+ indicator OregonGreen- BAPTA-5N （OGB-5N, 500μM） and the Ca2+ -insensitive dye Alexa Fluor 594 （40μM）. To monitor ［Ca2+]i in the spine head and at the base of the spine simultaneously, we performed line scanning along the axis of the spine after stimulation. Since OGB-5N has a low affinity for Ca2+ （dissociation constant, 32μM） and its Ca2+ binding ratio （κ） is 2+ buffers of neurons only minimally, and therefore would be expected to have the least effect on the spatiotemporal pattern of Ca2+ signaling. The NMDAR blocker APV eliminated both the increase in ［Ca2+]i （Δ［Ca2+]i） and the whole-cell current induced by photolysis of MNI-glutamate, indicating that the response was mediated by NMDARs. The full-width-at-half-maximal （FWHM） resolution of NMDAR-mediated current （INMDA） estimated was ~1.4μm, consistent with the prediction made from the established slow gating of NMDARs. Increases in ［Ca2+]i were often undetectable （0.5 ± 0.5μM, mean ± SD） when uncaging was effected at arbitrary points along the dendritic shaft at a distance of >2μm from a neighboring spine, indicating that NMDARs were present in relatively small numbers on the dendritic shaft compared with the spine head. We analyzed the data from 213 spines （17 dendrites from different animals）, and found that the amplitude of INMDA positively correlated with spine-head volume （VH）. A substantial INMDA was detected even in small spines, in contrast to AMPAR-mediated currents. This observation was confirmed by the fact that a marked Δ［Ca2+]i was detected in the head of most small spines. This indicates that small spines nearly correspond to post-synaptic "silent synapses," which express significant NMDARs but little AMPARs. Moreover, we found that Δ［Ca2+］i of spine head tended to be largest in spines with a small VH, with the mean value （〓） of correlation coefficients being -0.50 for the four dendrites. We found that neck Ca2+ conductance （gN） was highly related to VH： gN was small （3 s-1） when the spine-head volume was 3, but it increased markedly in spines with head volumes of >0.1μm3. A double logarithmic plot revealed that gN was roughly proportional to the second power of VH. The head-neck relation was confirmed by values of gN （gN*） which estimated from morphology of the spine necks. We found that the values of gN were well correlated with those of gN*, especially for spines on the same dendrite. We found that gN* also depended on the second power of VH, and that this relation was due to the correlation of VH with both neck diameter and neck length. We think that gN （spine neck geometry） has much effect on NMDAR-mediated Ca2+ signaling, because we found that another Ca2+ clearance mechanisms of spine heads, Ca2+ pumps （gH）, played minor role in determining the VH dependence of Δ［Ca2+］i. Thus, Δ［Ca2+]i was larger and more confined within a head of smaller spines, mainly because they tend to have thin and/or long necks （smaller gN） which prevent outflow of Ca2+ into dendritic shafts. In contrast, Δ［Ca2+]i was smaller but spread into dendritic shafts from larger spines because of their thick and/or short necks （larger gN）. We conclude that spine-neck geometry is key determinant of spine Ca2+ signaling, allowing small spines to be the preferential sites for isolated induction of long-term potentiation