SCPNs and CPNs exhibit similar strength of electrical coupli

SCPNs and CPNs exhibit similar strength of electrical coupling (Maruoka et al., 2017) and had similar membrane properties (Experimental Procedures), and showed no significant difference in amplification and synchronization (Figs. 2E and 3C). However, a subset of CPNs exhibit very strong electrical coupling (Maruoka et al., 2017). Similarly, amplification and synchronization were very strong for a subset of CPNs (Figs. 2E and 3C). Therefore, the strength of electrical coupling and its functional effects may be more varied among CPNs than among SCPNs. The coupling-mediated synchronization observed in the present study was much slower than that observed in cortical inhibitory neurons or sister pyramidal neurons (Gibson et al., 1999, Yu et al., 2012) and in many other types of neurons (Landisman et al., 2002, Long et al., 2002, van Welie et al., 2016). During the first postnatal week, spontaneous activity in the retina drives wave-like correlated activities in the visual pitavastatin (Ackman et al., 2012). These waves sweep the cortex at a speed of approximately 100 μm/s or ~100 ms/10 μm. Because coupled neurons mostly align radially with a tangential distance of <15 μm (Maruoka et al., 2017) and the time constant of the electrical transmission is ~65 ms, spatial propagation of the electrical transmission occurs at a speed similar to that of traveling waves. Therefore, the slow electrical transmission may facilitate propagation of wave-like activities in the visual cortex. Electrical coupling may also synchronize the activities of radially aligned neurons with retina-driven waves and may coordinate the development of visual circuits, including those that organize visual responses within microcolumns (Maruoka et al., 2017). Our simulation study suggested electrical coupling at distal dendritic positions, and electrical coupling was correlated to dendritic bundling. These results are consistent with the hypothesis that (1) a significant fraction of gap junctions is on apical dendritic bundles and that (2) this dendritic coupling causes the slow transmission. To confirm apical dendritic coupling, it is critical to determine the exact location of gap junctions using methods such as immunoelectron microscopy and protein tagging. In particular, because electrical coupling was also present for neuron pairs without apical dendritic bundling (Fig. 5D and F), some gap junctions are also likely present on other sites, such as the soma or basal dendrites. Therefore, the exact distribution of gap junctions must be determined to obtain a comprehensive picture of the network. In addition to the position of coupling, the type of connexins may also affect the time-course of electrical transmission because different connexins have different kinetics (Bukauskas and Verselis, 2004). Therefore, the types of connexins expressed in SCPNs and CPNs and their contribution to slow transmission must be clarified. Coupling on apical dendrites may be part of the mechanism for the preferential coupling of radially aligned neurons (Maruoka et al., 2017). Bundles of the apical dendrites of layer 5 neurons have been described in many cortical areas (Innocenti and Vercelli, 2010). Although bundles are implicated in functional clustering (Kondo et al., 2016), the physiological functions of bundling are not well understood (Rockland and Ichinohe, 2004, Krieger et al., 2007). Our findings suggest that a function of dendritic bundling is to provide gap junctional coupling during development. Gap junctions on dendritic bundles may also structurally related to "neuronal domains", radial clusters of neonatal neurons that are suggested to be electrically coupled (Yuste et al., 1992, Peinado et al., 1993, Yuste et al., 1995). On the other hand, the cell-adhesive function of gap junctions (Elias et al., 2007) may contribute to the formation and/or maintenance of dendritic bundles.
Acknowledgements We thank Naomi Matsumoto for technical assistance. This work was supported by research funds from RIKEN to T.H. and a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan to T.H. (Innovative Areas “Mesoscopic Neurocircuitry”; 22115004) and N.N. (16K14565). N.N. performed all experiments and analyses. T.H. conducted the study and performed the modeling analysis. N.N and T.H. wrote the manuscript.