Supplementary Figure 1: Expression of GCaMP6s in LGN and their axons in V1.

(a, b) Coronal section of LGN and V1 expressing GCaMP6s. a Coronal slice (bregma -2.3 mm) including thalamus confirmed that GCaMP6s expression was restricted within LGN. b The section of V1 was stained with fluorescent Nissl (red) and layers were assigned according to the size and density of cells (Anderson et al., Brain Res. 2009) (top). The same section from the top panel with GCaMP6s signal (green, bottom) is shown with assigned layers. (c) A part of V1 area (a bottom) was expanded. Intensity of GCaMP6s signal (LGN axons, green) was summed horizontally and plotted as normalized intensity on the right along the cortical depth. LGN axons distributing in layer 4 were particullary numerous and much less amount in layers 1 and 2/3. Small numbers of axons can be also seen in layers 5 and 6. We repeated the experiments three times. Previous studies showed the LGN projection pattern to V1 in higher mammals. In cat, X and Y cells mainly target layers 4 and 6, while W cells mainly target layer 1 and some amount to layers 3, 4a, 5a (Boyd & Matsubara, J. Comp. Neurol. 1996). In macaque monkey, similar to cat, P and M cells target mainly layer 4 and less amount layer 6, while K cells target mainly layers 1 and 2/3 (Lund, Annu. Rev. Neurosci. 1988). Therefore, projection patterns from LGN to V1 in mouse is similar to cat and macaque in principle. (d) Three-dimensionally reconstructed image of V1 from Scnn1a-cre/Ai14 mouse in vivo. Most of the cell bodies (red) can be seen at the depth of ~400μm from the pia mater. (e) Merged image of LGN axons (green) and layer 4 neurons (red) obtained from Scnn1a-cre/Ai14 mouse. LGN axons (green) can be seen among layer 4 neurons (red). Some imagings from layer 4 were obtained by using Scnn1a-cre/Ai14 mice (11 mice). (f, g, h) An example of cranial window for the in vivo imaging (f). Intrinsic optical imaging was performed and retinotopic map was created to determine the V1/V2 border (11 mice, g). tdTomato (red) was co-expressed with GCaMP6s (green) in LGN axons. Because GCaMP6s signal is low at the base level, we used the co-expressed tdTomato signal to find the area to image. Two-photon calcium imagings of LGN axons were typically obtained several ROIs (eg. white rectangles) within V1 (h). The dotted line indicates V1/V2 border determined from the retinotopic map (f-h).