Fig. 4. Mutant sensory axons fail to respond to Sema3A chemorepulsion. (A-C) Co-cultures of Sema3A-transfected Cos7 cell aggregates (S3A) with E13.5 DRG from wild-type (A), L1-null (B) or TAG-1-null (C) embryos. (D) Wild-type DRG co-cultured with Sema3A-transfected Cos7 cells in the presence of soluble NRP1-AP. (E) P/D ratios of axons numbers in wild type (WT; mean=0.096, s.e.m.=0.03, n=25), L1 null (L1-; mean=0.88, s.e.m.=0.047, n=20), TAG-1 null (TAG-; mean=0.71, s.e.m.=0.05, n=15) and wild type plus soluble NRP1-AP (WT+NRP1; mean=1.0, s.e.m.=0.04, n=10). (F,G) Wild-type DRG and VSC cultured together with soluble NRP1-AP (F) and plotted with data from Fig. 3E for comparison (G). There was no significant difference (P>0.05) between L1- and WT+NRP1 in either E or G. Scale bars: 100 µm. (H) NGF-dependent growth cones from wild type (WT) or TAG-1 null (TAG-) with (+Sema3A) or without Sema3A. Growth cones are immunolabelled with anti-L1. Scale bar: 20 µm. (I) Percentage of collapsed growth cones from wild-type or TAG-1-null DRG with (S3A) or without (cont.; supernatant from mock-transfected Cos7 cells) Sema3A. Over 600 growth cones were assessed for each DRG; each datapoint is the mean of data from n=18 (WT+S3A), n=17 (WT cont.), n=7 (TAG- +S3A) and n=5 (TAG- cont.) DRG. ^***P<0.001, ^**P<0.01; unpaired t-test.