Variation and Selection in Axon Navigation Through Microtubule-Dependent Stepwise Growth Cone Advance

Abstract

24Myosin II (MII) activity is required for elongating mammalian sensory axons to change speed 25 and direction in response to Nerve Growth Factor (NGF) and laminin-1 (LN). NGF signaling 26 induces faster outgrowth on LN through regulation of actomyosin restraint of microtubule 27 advance into the growth cone periphery. It remains unclear whether growth cone turning on LN 28 works through the same mechanism and, if it does, how the mechanism produces directed 29 advance. Using a novel method for substrate patterning, we tested how directed advance 30 occurs on LN by creating a gap immediately in front of a growth cone advancing on a narrow 31 LN path. The growth cone stopped until an actin-rich protrusion extended over the gap, 32 adhered to LN, and became stabilized. Stepwise advance over the gap was triggered by 33 microtubule +tip entry up to the adhesion site of the protrusion and was independent of traction 34 force pulling. We found that the probability of microtubule entry is regulated at the level of the 35 individual protrusion and is sensitive to the rate of microtubule polymerization and the rate of 36 rearward actin flow as controlled by adhesion-cytoskeletal coupling and MII. We conclude that 37 growth cone navigation is an iterative process of variation and selection. Growth cones extend 38 leading edge actin-rich protrusions that adhere transiently (variation). Microtubule entry up to 39 an adhesion site stabilizes a protrusion (selection) leading to engorgement, consolidation, 40 protrusive activity distal to the adhesion site, and stepwise growth cone advance. The 41 orientation of the protrusion determines the direction of advance. 42 87 5 preference for LN primarily through traction force pulling supporting the prevailing view that 88 substrate coupling is instructive (necessary and sufficient) for advance during directed growth 89 and that the main function of MII is to develop pulling forces via the adhesion sites. However, if 90 MII inhibition does not block advance then the core mechanism may be regulation of 91 actomyosin restraint of microtubule advance as we recently proposed (Turney et al., 2016). 92Here we test between these possibilities. 93 94 RESULTS 95 96 Novel assay for analysis of stepwise growth cone advance 97 Mouse dorsal root ganglion (DRG) or superior cervical ganglion (SCG) neurons were cultured 98 in a microfluidic Campenot chamber that served to hinder migration of glia into the axon 99 compartment and to chemically isolate distal axon segments and growth cones from cell 100 bodies. Axons exiting the channels of the chamber (10 µm wide) grew onto narrow LN lanes 101 (9-12 µm wide). On many lanes we produced gaps (non-adhesive regions) over which axons 102 had to cross for the elongation to continue. Both the lanes and the gaps were created by a new 103 method of substrate patterning (Live Cell Substrate Patterning or LSCP; see Experimental 104 Procedures) which uses region of interest (ROI) scanning of intense multiphoton laser light to 105 remove PLO, FN and LN f...

Cytoskeletal Crosstalk in Cell Migration

“…The buckling of MTs and breaking of MTs due to the actin retrograde flow might also play a role in the polarized organization of the MT network (Figure 2B) [97,98]. For example, in neuronal growth cones, the entry of MTs into actin-rich protrusions highly depends upon actin retrograde flow rates and myosin II activity [99]. From these results, it appears that the polarized organization of actin contributes to the guidance and stabilization of MTs from their organizing centers towards the cell front and also influences the dynamics of MTs at the cell periphery.…”

Spatial confinement: A spur for axonal growth

“…Here GCs do not loose physical contact with their substrate but their impaired or modified adhesion trigger specific behaviors. GCs explore non-fouling or cell-repellant surfaces on a limited spatial range over maximum distances of the order of 50 µm [7,40,44,52,53,54]. Unexpectedly, the average velocity of dissociated dorsal root ganglion (DRG) neurons from chick embryos grown on array of LN stripes increases with the distance between stripes [52].…”

Probabilistic axon targeting dynamics lead to individualized brain wiring

“…This conserved growth cone behaviour may be induced by a change in substrate, when growth cones migrate from a dense neuropile to a sparser environment organized in "open channels" 32,33 . This radical change in morphology may also reflect a bet-hedging strategy as a growth cone crosses a non-adhesive environment, whereby the extension of multiple long parallel filopodia in the same direction would increase the probability of at least one fibre reaching the more adhesive target 34 . Similar to Drosophila DCNs, one fibre of the chick embryo spinal cord splitting axon is selected through microtubule stabilization, suggesting that microtubule-based axon amplification and stabilization is a conserved targeting mechanism.…”