Figure 1.
Expression of Robo receptors in CAD cells.
(A) Representative images of a differentiated CAD cell and a dorsal root gangion (DRG) neuron immunostained for tubulin and actin cytoskeleton. Note that in general, the cytoskeletal organization of a neurite-like process in a differentiated CAD cell is similar to that of a primary neuron. In both cells, microtubules labeled the neurite or the neurite-like process along the length, whereas the actin cytoskeleton was enriched at the distal end. Bar, 25 µm. (B) Undifferentiated and differentiated CAD cells were tested for expression of the robo gene using RT-PCR. The PCR products of robo1 and robo2 were sequenced to confirm their identity. (C) CAD cells were transiently transfected with an HA-tagged Robo1 construct. Immunostaining with anti-HA antibodies revealed that ectopically expressed Robo1 was enriched at the growth cone (arrows). The cells were co-transfected with green fluorescent protein (GFP) to illustrate cell morphology. Bar, 25 µm.
Figure 2.
Recombinant mouse Slit2 (mSlit2) inhibits the extension of neurite-like processes in CAD cells.
(A, B) Differentiated CAD cells were plated at low density and cultured for 72 hr in the absence or presence of recombinant mSlit2 (25 nM) and/or recombinant rRobo1 (56 nM), as indicated. CAD cells were fixed and immunostained for tubulin. Representative images are shown in (A), and quantification of the lengths of neurite-like processes is shown in (B). Bar, 200 µm. *** p<0.001.
Figure 3.
Robo receptors mediate the mSlit2-induced inhibition of process outgrowth.
(A, B) CAD cells were subjected to real-time quantitative RT-PCR to confirm the down-regulation of robo1 (A) or robo2 (B) after transfection of siRNAs against robo1 (siRobo1) or robo2 (siRobo2). ** p<0.01; * p<0.05. (C, D) CAD cells transfected with either td-Tomato alone or together with siRobo1 and/or siRobo2 were cultured under serum-deprived conditions in the presence or absence of mSlit2, as indicated. Cells were fixed at 48 hr after incubation and immunostained for βIII-tubulin (TuJ1) to measure the lengths of neurite-like processes. Representative images (C) and quantification of process length (D) are shown. Process length was normalized against vehicle control to compare the effect of mSlit2. Bar, 100 µm. *** p<0.001.
Figure 4.
mSlit2 inhibits the extension of neurite-like processes in a concentration-dependent manner.
(A, B) Differentiated CAD cells were plated at low density and cultured for 72 hr in the absence or presence of different concentrations of recombinant mSlit2, as indicated. Representative images are shown in (A), and quantification of process length is shown in (B). Bar, 200 µm. *** p<0.001.
Figure 5.
Time-course of process extension in the presence or absence of mSlit2.
(A, B) Differentiated CAD cells were plated at low density and cultured for the indicated period of time in the absence or presence of recombinant mSlit2 (25 nM). Cells were fixed and stained for tubulin at three different time points (24 hr, 48 hr, 72 hr). Representative images are shown in (A), and quantification of the lengths of neurite-like processes is shown in (B). Bar, 200 µm.
Figure 6.
mSlit2 induces phosphorylation and inactivation of GSK3β.
(A, B) Undifferentiated and differentiated CAD cell lysates were subjected to Western blot analysis to compare the level of phospho-GSK3β. Lysates for differentiated CAD cells were collected at 24 hr after the induction of differentiation. For quantification in (B), the level of phospho-GSK3β was normalized against GAPDH loading control. Presented are representative immunoblots (A) and quantification of Western blot analysis (B). (C–G) Differentiated CAD cells were stimulated with two different concentrations of recombinant mSlit2 (5 nM or 25 nM). Protein lysates were prepared and subjected to Western blot analysis. Expression and/or activation/inactivation of GSK3 as well as other kinases, such as AKT and ERK1/2, were emxamined. The level of phosphorylated CRMP2, a substrate of GSK3 was also monitored. Representative immunoblots are shown in (C) and quantification of Western blot analysis is shown in (D–G). * p<0.05; ** p<0.01; n.s. not statistically significant.
Figure 7.
Inactivation of GSK3β is required for mSlit2-induced inhibition of process extension.
(A, B) CAD cells transfected with GSK3β-S9A mutant or EGFP, as a control, were cultured under serum-deprived conditions in the presence or absence of mSlit2, as indicated. Cells were fixed at 72 hr after incubation and immunostained for tubulin to measure the lengths of neurite-like processes. Representative images (A) and quantification of process length (B) are shown. Bar, 200 µm. *** p<0.001. (C) CAD cells were transfected with shGSK3 or treated with 6-bromoindirubin-3′-acetoxime (300 nM), an inhibitor of GSK3 (GSK3i), and cultured for 72 hr under serum-deprived condition. Cells were then fixed and stained for tubulin to measure the lengths of neurite-like processes. *** p<0.001 compared to control.
Figure 8.
mSlit2 inhibits GSK3β and neurite outgrowth in DRG neurons.
(A, B) DRG neurons from conditioning lesioned mice were treated with recombinant mSlit2 (25 nM) or vehicle control. Protein lysates were subjected to Western blot analysis and the level of phospho-GSK3β was examined. For quantification in (B), the level of phospho-GSK3β was normalized against GAPDH loading control. Presented are representative immunoblots (A) and quantification of Western blot analysis (B). * p<0.05. (C, D) DRG neurons from conditioning lesioned mice were cultured overnight in the presence or absence of mSlit2 (25 nM). Neurons were then fixed and stained for βIII-tubulin to measure neurite length. Presented are representative images (C) and quantification of neurite length (D). Bar, 200 µm. *** p<0.001.