Skip to main content
. 2016 Nov 28;5:e19103. doi: 10.7554/eLife.19103

Figure 3. Localization of the origins of thalamostriatal projections that converge with a corticostriatal projection in the striatum.

(a) Schematic sagittal view of the mouse brain, adapted from (Watson et al., 2012), indicating the location of M1/2. (b) Distribution of dense (dark yellow) and diffuse (light yellow) corticostriatal projections from M1/2. (c–d) Representative images of two coronal sections through the striatum of one example brain (left panels in c and d) showing the fluorescent thalamic axons in the striatum from injections described in panel e. Original images are on the left and segmented striatum and axon projection fields are on the right, with traveling axon bundles subtracted (black in right images). (e) Two views of a model thalamus (gray) showing the four thalamic viral injections that produced projections shown in panels c and d. Note that since thalamic projections do not cross the midline in mouse, a single injection spanning the midline was treated as two independent injections (injections 2 and 4). A darker center of each injection site represents the eroded ‘core’ of each injection defined previously for the thalamic injection dataset (Hunnicutt et al., 2014). (f) Projection distributions in the striatum for each of the injections shown in panel e (red and green) aligned and overlaid with the outlines of M1/2 projections in the striatum (yellow) delineated in panel b. (g) Injections were assigned to one or more of four categories based on quantification of the convergent volumes of thalamostriatal and corticostriatal projection fields (see Materials and methods). Inclusion in each category is used, as described in Figure 3—figure supplement 1, to localize the thalamic origins of convergence. (h–i) Fluorescent images of coronal sections through the thalamus, showing injection sites 1, 2, and 4. Insets show the segmented injection sites (solid white line) and the injection site core (dashed white line) (Hunnicutt et al., 2014). The dashed yellow line in panel h insert shows the brain midline. (j–k) Two example coronal sections, approximately corresponding to the position of panels h and i, respectively, of the thalamostriatal confidence maps for M1/2 convergence in panels h and i, respectively (top panels in j and k). The segmented injection sites are overlaid on their corresponding confidence maps (bottom panels in j and k). All scale bars, 1 mm.

DOI: http://dx.doi.org/10.7554/eLife.19103.012

Figure 3.

Figure 3—figure supplement 1. Illustration of the method used to generate thalamic confidence maps.

Figure 3—figure supplement 1.

(a) Illustration of eight hypothetical injection volumes labeled to indicate whether or not they individually satisfy each of three criteria for their projections in the striatum (right). Each injection indicates the borders of the full injection (1, 2, 3, and X) and the injection core (1c, 2c, 3c, and Xc), which is generated by eroding the full injection volume by 100 µm. (b) Injections shown in panel a, indicating the injections that fulfill each of the three criteria (green). (c) Binary injection masks representing the area covered by either the full injection (top) or injection cores (bottom) that satisfy each of the three criteria. (d) Binary injection masks representing the area covered by either the full injection (top) or injection cores (bottom) that do not satisfy each of the three criteria. Binary mask for the full injections that do not meet the hardest to satisfy criteria are not included. (e) Sum of binary masks generated in panel c. (f) Subtraction of the binary masks generated in panel d. (g) Example confidence map, generated by combining the summed binary masks from panel E and the subtracted binary masks from panel f. Values below zero set to zero.