Extended Data Figure 3. Evaluation of CAV-Cre spread by using the OB→APC projection.
a, Four representative 60-μm coronal sections of the CAV-Cre injection site in the anterior piriform cortex (APC) of four Ai14 Cre reporter mice. Red, tdTomato; green, retrobeads; blue, DAPI. The location of the injection site was readily visualized by concentrated retrobeads. L, lateral; M, medial; D, dorsal; V, ventral. In each mouse, we determined the minimal distance (D) between the injection site and layer 1a of the piriform cortex, where mitral cell axons terminate, or lateral olfactory tract, where mitral cell axon bundles are present. Dashed lines represent the boundary between layer 1a and layer 1b. For each sample, we counted the number of tdTomato-labeled mitral cells (numbers below each image) from serial OB sections. b, An example 60-μm coronal section of the OB. Both tdTomato and retrobead signals were found to be mostly restricted to the mitral cell layer (M) of the main olfactory bulb (MOB) and accessory olfactory bulb (AOB) with minor labeling in the granule cell layer (Gra). As AOB mitral cells do not form synapses in the APC, this observation indicates that CAV-Cre can infect axons-in-passage. c, Distribution of D among 26 injections (x-axis) and relationship between D and the numbers of labeled cells in the MOB (y-axis). d, Histogram based on (c). Dense labeling (over 1000) was obtained only when D < 100 μm. CAV-Cre injections with D > 800 μm rarely labeled the OB (2.8±1.9 cells/bulb, n=4). e, Cumulative distribution plot of MOB cell counts. A sample of the 9th smallest D (D = 200μm) reached 90% of the labeling (indicated by vertical dotted line) detected in all 26 samples, suggesting that given our sample distribution, ~90% of axonal transduction occurred within 200 μm from the CAV-Cre injection site. Scale, 100 μm. Error bars, s.e.m.