Fig 2: Putative Arrangements of Robo1 on the Cell Surface(A) Compact inactive Robo1 dimers on the cell surface can (B) undergo a conformational change but remain intact upon stimulation with Slit2-N.

Fig 3: Structure of Human Robo1 Ig1-4 and Ig5(A) The domain composition of Robo1 and Slit2. The Robo1 Ig and Fn domains are colored green and blue, respectively. The Slit2 LRR, EGF, Lamin, and CTCK domains are colored orange, yellow, red, and brown, respectively. The cleave site of Slit2 is indicated by a dashed line.(B) Robo1 Ig1-4 domains adopt an extended structure. The N-glycosylation at N160 is shown in stick representation.(C) The major crystallographic contacts are mediated by Ig1, Ig3, and Ig4. Interface 1 is symmetric and mediated by Ig4 (blue); interface two is mediated by Ig2-3 (yellow) and Ig4 (orange); and interface three is mediated by Ig3 (red) and Ig4 (salmon); interface four is mediated by Ig1 (light and dark green) and overlaps the Slit2 D2 binding site, illustrated as a ribbon representation (N- and C-terminal caps colored in magenta and cyan, respectively, and LRR colored in orange).(D) The Robo1 Ig5 domain structure showing a canonical I-set fold.(E) A superposition of Robo1 Ig domains with Ig1, Ig2, Ig3, Ig4, and Ig5 colored in red, gray, cyan, blue and green, respectively. One potential conformation of K137 and R136 (disordered) is shown in stick representation to highlight the Robo1 Ig1 heparin binding region (βE-βF loop).

Fig 4: N-cadherin is required for the trafficking of Slit2/Slit3 to the cell surface.(a–d) Representative confocal images of control or N-cadherin (N-Cad) knockdown Schwann cells (SCs). Cells were labelled with antibodies to (a) Slit2 or (c) Slit3 (green) with quantification of Slit2 and Slit3 levels in the cell protrusions indicated by the boxes (n = 3, mean ± SEM for both conditions). Scale bars = 15 μm. p-values were calculated using an unpaired t-test. (e) Representative confocal images of rescue experiments in which SCs depleted of N-Cad were transfected with the N-Cad ECD tagged with tomato, or tomato control vector and immunolabelled to detect Slit2 (green). Scale bar = 15 μm. (f) Quantification of (e) (n = 3, mean ± SEM). p-values were calculated using a one-way ANOVA followed by Tukey’s multiple comparisons tests. (g) As (e) but stained for Slit3 (green). (h) Quantification of (g) (n = 3, mean ± SEM). Scale bar = 15 μm. p-values were calculated using a one-way ANOVA followed by Tukey’s multiple comparisons tests. *p<0.05, **p<0.01. Figure 5—source data 1.Excel spreadsheet containing data used to generate graphs in Figure 5.

Fig 5: The Slit-repulsive signal is required for the efficient collective migration of Schwann cells (SCs) during nerve regeneration.(a) Quantification of the collective migration of control compared to Slit2/Slit3 knockdown SCs at 6 hr using a chamber assay (Figure 6—video 1). Data is normalised to control and presented as mean area ± SEM of n = 3 independent experiments. p-values were calculated using an unpaired t-test with Welch’s correction compared to control. (b) Representative confocal images of SCs treated with recombinant-Slit2 (rSlit2) or PBS immunolabelled to detect N-Cadherin (N-Cad) (green) and co-stained with phalloidin (red) to detect F-actin and Hoechst to detect nuclei (blue). (n = 3). Scale bar = 50 μM. (c) Quantification of SC clusters from (b) (n = 3, mean ± SEM) (Figure 6—video 2). p-values were calculated using a two-way ANOVA followed by Sidak’s test for multiple comparisons. (d) Graph shows the collective migration of SCs in a chamber assay treated with rSlit2 or PBS control (n = 3, mean area ± SEM). p-values were calculated using an unpaired t-test with Welch’s correction compared to PBS controls. (e) Graph shows the Euclidean distance (shortest distance travelled) for cells in (d) at 24 hr (PBS n = 141, rSlit2 n = 138 from n = 3 independent experiments). The red line denotes the mean. p-values were calculated using an unpaired two-tailed t-test. (f) Graph shows tracks of individual cells in the collective migration assay quantified in (d) n = 3. (g) Representative confocal images from three independent experiments of Sox2-induced SC clusters treated with Shield for 24 hr and PBS or rSlit2 immunolabelled to detect N-Cad (green) and co-stained with phalloidin (red) to detect F-actin and Hoechst to detect nuclei (blue) (Figure 6—video 3). Scale bar = 50 μM. (h) Quantification of cell roundness of Sox2-induced SC clusters treated with PBS or rSlit2. (n = 3, mean ± SEM). Data was normalised to Sox2 controls and p-values calculated using an unpaired t-test. (i) Polar histograms showing alignment of nuclei within each cluster from PBS (n = 168) or rSlit2 (n = 166) treated Sox2 SCs as an indicator of polarisation. Angles closer to 0 represent more aligned nuclei in the PBS (blue) whereas they are more randomly distributed in the rSlit2-treated Sox2 SCs (orange). Data is representative of n = 3. (j) Representative confocal images of Sox2 SC clusters treated with control, or Slit2/Slit3 siRNAs and immunolabelled to detect N-Cad (green) and co-stained with phalloidin (red) to detect F-actin and Hoechst to detect nuclei (blue). Scale bar = 50 μm. (k) Quantification of the cell roundness of individual cells in Sox2-induced SC clusters treated with control or Slit2/Slit3 siRNA1 or 2 (n = 3, mean ± SEM). Data was normalised to Sox2 controls and p-values calculated using an unpaired two-tailed t-test. (l) Schematic illustrating that in control conditions SC exhibit CIL but upon KD of N-cadherin, CIL is lost and SCs become overlapping. Knockdown or inhibition of Slit2/Slit3 inhibits CIL but the persistence of N-cadherin expression results in the formation of non-migratory clusters. In contrast, SC cords in which Sox2 is activated maintain CIL signals which drive their collective migration. *p<0.05, **p<0.01, ***p<0.001. Figure 6—source data 1.Excel spreadsheet containing data used to generate graphs in Figure 6.