CED-10/Rac1 mediates axon guidance by regulating the asymmetric distribution of MIG-10/lamellipodin

Abstract

Axon migrations are guided by extracellular cues that induce asymmetric outgrowth activity in the growth cone. Several intracellular signaling proteins have been implicated in the guidance response. However, how these proteins interact to generate asymmetric outgrowth activity is unknown. Here, we present evidence that in C. elegans, the CED-10/Rac1 GTPase binds to and causes asymmetric localization of MIG-10/lamellipodin, a protein that regulates actin polymerization and has outgrowth-promoting activity in neurons. Genetic analysis indicates that mig-10 and ced-10 function together to orient axon outgrowth. The RAPH domain of MIG-10 binds to activated CED-10/Rac1, and ced-10 function is required for the asymmetric MIG-10 localization that occurs in response to the UNC-6/netrin guidance cue. We also show that asymmetric localization of MIG-10 in growth cones is associated with asymmetric concentrations of f-actin and microtubules. These results suggest that CED-10/Rac1 is asymmetrically activated in response to the UNC-6/netrin signal and thereby causes asymmetric recruitment of MIG-10/lamellipodin. We propose that the interaction between activated CED-10/Rac1 and MIG-10/lamellipodin triggers local cytoskeletal assembly and polarizes outgrowth activity in response to UNC-6/netrin.

Figure 1

CED-10/Rac1 binds to MIG-10/lamellipodin and is required for the asymmetric localization of MIG-10/lamellipodin. (A), GST fusion proteins encoding activated mutants of the Rho family GTPases (RhoL63, RacL61, and Cdc42L61) were incubated with the RAPH domain of MIG-10 fused to GFP (MIG-10 RAPH::GFP). MIG-10 RAPH::GFP bound to RacL61 but not Rho L63 or Cdc42L61. (B), MIG-10 RAPH::GFP was incubated with an activated Rac mutant (RacL61) or an inactive Rac mutant (RacN17). MIG-10 RAPH::GFP bound to RacL61 but not RacN17. (C), The UNC-6/netrin guidance cue is expressed from cells located ventral to the HSN and is thought to form a chemotactic gradient that polarizes the HSN [6]. (D), In the wild type background, MIG-10::YFP is localized to the ventral edge of the HSN. Previous work has indicated that this ventral localization is netrin-dependant [6]. (E), In the ced-10(n3246) mutant background, MIG-10::YFP fails to localize to the ventral edge of the HSN neuron. (F), Average ventral enrichment of MIG-10::YFP in wildtype (n=12) and ced-10(n3246) (n=15) backgrounds. Asterisk indicates p<0.0001 (student’s t-test). Myristolated::GFP (MYR::GFP) was not enriched in the wildtype or ced-10(n3246) backgrounds. Error bars represent standard error of the mean. Images are of collapsed stacks of optical sections. Ventral is down and anterior is to the left. Scale bars are 5μm. The molecular weight markers represent 150, 100, 75 and 50 Kilodaltons.

Figure 2

CED-10 functions with MIG-10 to orient HSN axon growth. (A) Schematic of normal HSN axon trajectory. The HSN axon makes a ventral migration and then makes an anterior turn to migrate along the ventral nerve cord. (B) Example of HSN axon in the wildtype background. (C and D) Examples of HSN axon in mig-10(ct41) background. In nearly all cases the axon eventually reached the ventral nerve cord. In defective axons, the axon reached the nerve cord after making a lateral migration in the anterior (example shown in c) or posterior (example shown in D) direction. (E and F) Similar defects were observed in the ced-10(n3246) mutant background. (G) Percentage of HSN axons that exhibited guidance defects. The ced-10;mig-10 double mutant was not enhanced relative to either single mutant. For all genotypes n=150. Error bars represent standard error of the proportion. Images represent collapsed stacks of optical sections. Ventral is down and anterior is to the left. Scale bars are 5 μm.

Figure 3

MAX-2/PAK-1 functions in parallel to MIG-10/lamellipodin to guide the HSN axon. (A) Example of HSN axon in the max-2(cy2) null mutant background. Nearly all axons had a normal ventral migration. (B) Example of HSN axon in the mig-10(ct41) null mutant background. Many of the axons made an abnormal lateral migration prior to making a ventral migraion. (C) Example of max-2(cy2); mig-10(ct41) double null mutant. Many axons failed to make a ventral migration. (D) Example of HSN axon in the unc-6(ev400) null mutant background. (E) Percentage of HSN axons that exhibited mild defects (where the HSN eventually reached the VNC) and percentage of HSN axons that exhibited severe defects (where the HSN never reached the VNC). For all genotypes n=150. Standard error of the proportion is shown. Images represent collapsed stacks of optical sections. Ventral is down and anterior is to the left. Scale bars are 5 μm.

Figure 4

CED-10 is required for the polarization of the MIG-10 outgrowth-promoting activity. (A) Example of AVM neuron overexpressing MIG-10A in the wildtype background. (B) Example of AVM neuron overexpressing MIG-10A in the ced-10(n3246) mutant background. (C) Example of neuron overexpressing MIG-10A in the ced-10(n3246); slt-1(eh15) double mutant background. (D) Percentage of AVM neurons overexpressing GFP or MIG-10A with more than one process growing out of the cell body. The urEx305 transgene[5], encoding mec-4::mig-10a, was used to overexpress MIG-10A in all experiments. For all genotypes n=150. Error bars represent standard error of the proportion. Images represent collapsed stacks of optical sections. Ventral is down and anterior is to the left. Scale bars are 5 μm.

Figure 5

Asymmetric MIG-10 is associated with asymmetric concentrations of f-actin and microtubules in cultured growth cones. (A-D) Representative example of a growth cone exhibiting asymmetric concentration of MIG-10::GFP double stained for f-actin and microtubules. (E) Analysis of actin and microtubule distribution in growth cones that exhibit asymmetric concentration of MIG-10::GFP (n=6) and growth cones that do not exhibit asymmetric concentration of MIG-10::GFP (n=44). Each growth cone was bisected with a line in parallel to the neurite shaft and the ratio of actin or microtubule fluorescence was calculated. The numerator corresponded to the side of the growth cone with the highest MIG-10 fluorescence. The denominator corresponded to the side of the growth cone with the lowest MIG-10 fluorescence. Asterisk indicates significant difference relative to growth cones without asymmetric MIG-10 (p<0.001, student’s t-test). Error bars represent standard error of the mean. Scale bar is 5 μm.