Facilitation of axon outgrowth via a Wnt5a-CaMKK-CaMKIα pathway during neuronal polarization

Abstract

Background: Wnt5a, originally identified as a guidance cue for commissural axons, activates a non-canonical pathway critical for cortical axonal morphogenesis. The molecular signaling cascade underlying this event remains obscure.

Results: Through Ca(2+) imaging in acute embryonic cortical slices, we tested if radially migrating cortical excitatory neurons that already bore primitive axons were sensitive to Wnt5a. While Wnt5a only evoked brief Ca(2+) transients in immature neurons present in the intermediate zone (IZ), Wnt5a-induced Ca(2+) oscillations were sustained in neurons that migrated out to the cortical plate (CP). We wondered whether this early Wnt5a-Ca(2+) signaling during neuronal polarization has a morphogenetic consequence. During transition from round to polarized shape, Wnt5a administration to immature cultured cortical neurons specifically promoted axonal, but not dendritic, outgrowth. Pharmacological and genetic inhibition of the CaMKK-CaMKIα pathway abolished Wnt5a-induced axonal elongation, and rescue of CaMKIα in CaMKIα-knockdown neurons restored Wnt5a-mediated axon outgrowth.

Conclusions: This study suggests that Wnt5a activates Ca(2+) signaling during a neuronal morphogenetic time window when axon outgrowth is critically facilitated. Furthermore, the CaMKK-CaMKIα cascade is required for the axonal growth effect of Wnt5a during neuronal polarization.

Fig. 1

Activation of Wnt5a-Ca²⁺ signaling in radially migrating cortical neurons. a Protocol for Fluo-4 Ca²⁺calcium imaging in immature migrating cortical neurons. b Top, image acquired from a Fluo-4-labeled acute slice, in which immature migrating neurons expressing TagRFP where identified in CP and IZ. Bottom, magnified image of an immature cortical neuron in CP from the boxed region in the upper panels. Scale bars: 50 μm (top) and 10 μm (bottom). c Left, Fluo-4 Ca²⁺ imaging showing immature migrating cortical neurons experiencing spontaneous Ca²⁺ activity. The neurons are magnified from the boxed areas in (b). Right, Ca²⁺ transient traces of immature cortical neurons, with vertical lines corresponding to the time points of the left images. Scale bars: 10 μm. d Wnt5a administration evoked Ca²⁺ transients in immature migrating neurons, both in CP (#1–4 cells) and IZ (#5, #6 cells). Several of the immature neurons in CP (e.g. #1–3 cells) showed sustained Ca²⁺ oscillations. Scale bar: 50 μm

Fig. 2

Axon-specific morphogenetic effects of Wnt5a in immature cultured cortical neurons. a Left, cortical neurons were stimulated with Wnt5a from 6 h until 48 h after plating. Wnt5a administration induced elongation of axons (arrowheads) but not dendrites (arrows). Right, a scatter plot of data points for both axonal and dendritic lengths obtained from individual neurons. Number of neurons: n = 15 for both Vehicle and Wnt5a groups; Scale bar: 50 μm. b Quantification of morphometric parameters in Wnt5a-stimulated neurons. For quantification, the total length, the branch tip number, and the longest processes were calculated over the axonal or the dendritic arborization for all branches exceeding 7 μm in length. Top: quantification of Wnt5a-stimulated neurons. The total length of axons and the length of the longest axons were increased. Bottom: none of the 3 morphometric parameters were significantly altered in dendrites. Number of neurons: n = 15 for both the vehicle and Wnt5a groups. **p < 0.01, n.s., not significant (p > 0.05) (t-test)

Fig. 3

Wnt5a promotes axonal elongation via a CaMKK-CaMK cascade in immature cortical neurons. a Cortical neurons were stimulated with Wnt5a in the presence or absence of KN-93, a general CaMK inhibitor. Wnt5a treatment facilitated axonal elongation and this effect was potently inhibited in neurons treated with KN-93. Number of neurons: n = 15 for all groups. b Treatment with STO-609, a selective inhibitor of CaMKKα/β, the upstream kinases of CaMKI/IV, severely impaired Wnt5a-dependent facilitation of axonal elongation. Number of neurons: n = 15 for all groups. c Left, Wnt5a-dependent facilitation of axonal outgrowth was abolished in CaMKKα/β-DKO neurons. Note that the basal growth of axons was strongly impaired in CaMKKα/β-DKO neurons, in the absence of Wnt5a. Right, similar to axonal elongation, basal dendritic growth was inhibited in CaMKKα/β-DKO neurons in the absence of Wnt5a. However, dendritic growth was not facilitated by Wnt5a administration, either in the WT or DKO. Number of neurons: n = 15 for all groups. Insets in both graphs show no significant changes in axonal or dendritic growths in the cultured cortical neurons from either WT or CaMKKα/β double heterozygous (DHT) mice. ***p < 0.001, **p < 0.01, *p < 0.05, n.s., not significant (p > 0.05) (two-way ANOVA with Turkey’s test)

Fig. 4

CaMKIα mediates Wnt5a-facilitated axon outgrowth. a Left, Specific effects of Wnt5a treatment and CaMKIα knockdown on total axonal lengths during neuronal polarization. Wnt5a treatment facilitated axon outgrowth and CaMKIα KD abolished this facilitation, in keeping with results from CaMKK blockade. Note that CaMKIα KD alone inhibited basal axonal elongation even in the absence of Wnt5a. Right, Neither Wnt5a treatment nor CaMKIα KD had any detectable effect on total dendritic lengths. Number of neurons: n = 15 for all groups. ***p < 0.001, **p < 0.01, n.s., not significant (p > 0.05) (two-way ANOVA with Turkey’s test). b Suppression of Wnt5a-stimulated axonal elongation by CaMKIα KD is rescued by co-expression of a shRNA-resistant wild-type CaMKIα (WTres). Number of neurons: n = 15 for all groups. ***p < 0.001 (one-way ANOVA with Turkey’s test)