Fig. 4.

Compartmentalization of RA synthesis is essential for exosome mediate RA transfer in neurite outgrowth. (a) Neurite growth in neuron-NG2 + cell cultures treated for 72 h with vehicle and retinal, alone or with the addition of BMS, GW4869 or DEAB. Scale bar is 50 μm. (b) Quantification of the longest neurite lengths in the cultures showed a significant reduction in the retinal induced neurite outgrowth when: either activation of RARs is inhibited (BMS) or RA synthesis (DEAB) or exosome transfer are prevented (GW4869). Data represent mean ± SEM. **p ≤ 0.01***p ≤ 0.001, One-way ANOVA followed by Tukey's test. (c) Diagram illustrating LV shRNA Raldh2 injection site at the DREZ, performed at the same time of injury in C5-T1 of rats that were subsequently treated with the RARβ agonist for 4 weeks. (d) Expression of GFP, Raldh2 and NG2 in RARβ treated rats that received LV shRNA Raldh2, shows effective transduction. Scale bar is 100 μm. (e) Loss of Raldh2 in the NG2 + cells at the DREZ in the same section shown in (d) compared to the expression of Raldh2 in the NG2 + cells of a RARβ agonist treated group that had not been transfected. Scale bar is 100 μm. Insets (I) and (II) show a higher magnification image of SC areas the differences in Raldh2 in NG2 + cells can be clearly seen between the two groups. Scale bar is 40 μm. (f) In the LV shRNA Raldh2 transduced rats, the axons fail to enter the SC and numerous axonal stumps remain at the DREZ (indicated by the white arrows in the lower insets, I and II). In contrast, in the non-transduced RARβ agonist treated animals, bundles of axons are seen entering the DREZ surrounded by NG2 + cells (seen clearly in the lower inset III). Scale bar is 100 μm for upper panel and 40 μm for lower insets (I, II and III). (g) Quantification of the axonal stumps at the DREZ. Data represent mean ± SEM. ***p ≤ 0.001, Student's t-test.