Wnt-5a Promotes Neural Development and Differentiation by Regulating CDK5 via Ca²⁺/Calpain Pathway

The roles of Wnt signaling in physiological and pathological processes have been comprehensively studied, including its contributions to cell development, differentiation, proliferation, and cancer [1-5]. Most Wnt signaling can be classified into canonical and non-canonical pathways [6]. A majority of Wnt ligands regulate canonical signaling via Wnt/β-catenin pathways, including those mediated by the ligands Wnt-1 and Wnt-3a [7, 8]. Other ligands mediate the non-canonical Wnt signaling-independent Wnt/β-catenin pathway, including Wnt-4, Wnt-5a, and Wnt-11 [9-11]. Most Wnt ligands have critical functions that contribute to neuronal differentiation of embryonal carcinoma and embryonic stem cells [12]. Wnt also affects several other signaling pathways, including Notch and PI3K/ AKT, whose activities affect neuronal development and differentiation [13]. Canonical Wnt signaling regulates axon genesis and the differentiation of cerebellar granule neuron(CGN) progenitors. Non-canonical Wnt signaling regulates vertebrate development and neurite outgrowth through the planar cell polarity (PCP) and Wnt/Ca²⁺ pathways [14, 15]; however, canonical Wnt can activate non-canonical pathways, and vice versa, by regulating the same downstream components, and human neural stem cell differentiation can be mediated via switching from canonical to non-canonical Wnt signaling. The molecular mechanisms underlying the effects of non-canonical Wnt signaling in brain development and differentiation are currently unclear.

Cyclin-dependent kinase 5 (CDK5) is a proline-directed serine/threonine kinase, which is essential for several neuronal functions [16, 17]. CDK5 activation requires a coactivator, such as p35 or p39 [18]. Intracellular calcium influx can activate calpain to promote conversion of p35 to p25 [19] and a CDK5/p25 complex phosphorylates substrates to regulate neuronal death, development, and differentiation [20-22]. CDK5 regulatory subunit-associated protein 1-like1 (CDKAL1) regulates adipocyte differentiation through the activation of the Wnt pathway [23]. A novel CDK5 splicing variant (CDK5-SV) may be a negative regulator of Wnt/β-catenin signaling [24]. In addition, CDK5 phosphorylates liprinα1 to mediate the neuronal activity-dependent synapse development [25].

Recently, CDK5 and Wnt signaling have been shown to be correlated in different circumstances. Wnt signaling and the regulation of CDK5 are critical for the development and differentiation of the nervous system; however, their correlation is seldom reported in this context. In this study, we determined that interaction between Wnt signaling and CDK5/p25 can mediate neuronal development and differentiation, as well as exploring the underlying mechanisms. We showed that Wnt-5a upregulates p25 expression, which elevated CDK5 activity during neuronal development and differentiation of SH-SY5Y cells. Consequently, Wnt-5a promoted activation of CDK5/p25-mediated neuronal development and differentiation of SH-SY5Y cells via regulation of the Ca²⁺/calpain signaling. In addition, we demonstrate that Wnt-5a activates CDK5, establishing a link between Wnt signaling and CDK5 kinase activity.

Recombinant human/Rat Wnt-5a was acquired from Promega (Madison, WI, USA). EGTA-AM, BAPTA-AM, and the calcium ionophore A23187 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Antibodies used for immunostaining or immunoblotting included the following: CDK5 (Santa Cruz, CA, USA), MAP2 (Cell Signaling Technology, MA, USA), p35 (Santa Cruz, CA, USA), and cleaved α-fodrin (Cell Signaling Technology, MA, USA). Wnt-5a neutralizing antibody and anti β-actin were obtained from Sigma-Aldrich (St. Louis, MO, USA). Alexa Fluor 488- and 594-conjugated secondary antibodies were purchased from Invitrogen (Carlsbad, CA, USA).

Primary cultures of rat cortical neurons were established from an E18 rat. After 3 days, neurons were treated with Wnt-5a or DMSO. SH-SY5Y human neuroblastoma cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS) in 5% CO₂ at 37°C. DMEM and FBS were obtained from Gibco (Rockville, MD, USA). SH-SY5Y cells were differentiated by incubation with retinoic acid and brain-derived growth factor (RA-BDNF) for 6 days.

Cells were lysed in RIPA lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 2 mM EDTA, 10 mM NaF, 2 mM Na3VO4) containing 0.01% protease inhibitor cocktail. The protein concentrations of lysates were determined using a BCA kit (Beyotime, Nanjing, China). Total lysate aliquots containing an equivalent of 20 µg protein per sample were separated by SDS-PAGE and transferred to PVDF membranes. Membranes were blocked using 5% skim milk or 5% BSA in TBST. Subsequently, levels of immunodetected protein were analyzed using ECL kits (Amersham, Buckinghamshire, UK).

Cell culture media were transferred to centrifuge tubes and centrifuged to remove debris. Concentrations of various Wnt ligands were detected using their corresponding ELISA kits, according to the manufacturer’s instructions (Cusabio, Wuhan, China). The absorbance of samples was measured at 450 nm using a Multiskan Spectrum (Multiskan, Thermo, USA).

Cultured cells were fixed for 1 h in 4% paraformaldehyde in phosphate-buffered saline (PBS), then analyzed using a polyclonal rabbit anti-MAP2 primary antibody, and anti-Rat Alexa Fluor 488 and anti-rabbit Alexa Fluor 594 secondary antibodies. Cells were examined using a Nikon microscope (Nikon, Japan). Subsequently, images were analyzed using Image J Lab software (Bio-Rad, Hercules, CA, USA).

For quantitative analysis of gene expression involved in Wnt signaling, total RNA was isolated from cortical neuron or SH-SY5Y cells using TRIzol reagent (Invitrogen) and equivalent amounts of cDNA synthesized using the Superscript RT enzyme(Vazyme,china). qPCR was performed using SYBR Green Mix (Takara, Japan), and gene expression level values were normalized to those of the internal control, GAPDH. Primer sequences were as follows: Wnt-1, forward 5’-GACGATCTTGCCGAAGAG-3’ and reverse 5’-CTGTGCGAGAGTGCAA-3’ (human), forward 5’-CAAGATCGTCAACCGAGGCT-3’ and reverse 5’- GTAGTCGCAGGTGCAGGATT-3’ (rat); Wnt-3a, forward 5’- GCCATCGGTGACTTCCTCAA-3’ and reverse 5’- TTGAAGTAGGTGTAGCGCGG-3’ (human), forward 5’-AGGTGCTAACACAGCAGCTTA-3’ and reverse 5’- CAGTGAGGAGTACTGGGGTC-3’ (rat); Wnt-4, forward 5’- GGCTTCCAGTGGTCAGGATG-3’ and reverse 5’- TTGTGGAGGTTCATGAGGGC-3’ (human), forward 5’- CTGTTCCACACTGGACTCCC-3’ and reverse 5’- ACACCTGCTGAAGAGATGGC-3’ (rat); Wnt-5a, forward 5’- AGAAGAAACTGTGCCACTTGTATCAG-3’ and reverse 5’- CCTTCGATGTCGGAATTGATACT -3’ (human), forward 5’- GAAGGCGAGCTGTCTACCTG-3’ and reverse 5’- ACACCTGCTGAAGAGATGGC-3’ (rat); Wnt-11, forward 5’- GGATATCCGGCCTGTGAAGG-3’ and reverse 5’- GTTGCACTGCCTGTCTTGTG -3’ (human), forward 5’- AGACAGGCAGTGCAACAAGA-3’ and reverse 5’- GTCCGTGTAGGGGTTGTAGC-3’ (rat); GAPDH, forward 5’- ACCCAGAAGACTGTGGATGG-3’ and reverse 5’- TTCAGCTCAGGGATGACCTT-3’ (human), forward 5’-GACATGCCGCCTGGAGAAAC-3’ and reverse 5’- AGCCCAGGATGCCCTTTAGT-3’ (rat).

Purified Histone H1 protein (1.5 μg) was incubated for 1 h at 30°C with p25/CDK5 in kinase buffer (20 mM MOPS, pH 7.0, 10 mM MgCl2, 1 mM DTT, and 20 μM sodium orthovanadate) containing 100 μM non-radioactive ATP and 5 μCi [γ-32P] ATP. Reaction mixtures were separated by SDS-PAGE and measured by autoradiography [17].

All data were presented as means ± standard error of the mean(SEM). Comparisons between two groups were conducted using two-tailed Student’s t-tests. Anova was used here for comparing more than two groups. P < 0.05 was considered statistically significant. All the experiments were repeated three times excluding certain chance factors.

To determine the role of canonical and non-canonical Wnt pathways in cortical development, we first evaluated the expression patterns of Wnt family genes throughout Rat brain development from embryonic day 11.5 to stage P0 (E11.5–P0). Next, we analyzed the expression levels of the canonical (Wnt-1 and Wnt-3a) and non-canonical (Wnt-4, Wnt-5a, and Wnt-11) Wnt ligands by qPCR. Wnt-1 mRNA levels did not alter significantly, while those of Wnt-3a were reduced; however, the mRNA levels of Wnt-4, Wnt-5a, and Wnt-11 increased with progressing embryonic development (Fig. 1a, 1b).

In the RA/BDNF-stimulated SH-SY5Y cell differentiation model, non-canonical Wnt signaling was upregulated at the mRNA and protein levels in differentiated SH-SY5Y cells compared with levels prior to differentiation (Fig. 1c, 1d). Together, these results indicate that non-canonical Wnt signaling is critical for cortical neuronal development and differentiation.

Next, we conducted experiments to investigate the roles of CDKs/ cyclins. To understand the function of CDKs during neuronal development, we assessed the mRNA levels of CDK2, CDK4, and CDK6 by qPCR analysis and found no significant differences in their levels during cortical neuronal development (Fig. 2a). Therefore, we investigated the cyclin partners of CDK2, CDK4, and CDK6, cyclin D and cyclin E, in the same context and found that expression levels of cyclin D was downregulated and cyclin E was unaltered (Fig. 2b), indicating that CDKs/cyclins do not have significant roles in cortical cell development.

CDK5 is a unique member of the CDK family, with diverse functions related to nervous system regulation. Therefore, we investigated whether CDK5 can regulate cortical neuronal development and found high levels of p25 expression during neuronal development (Fig. 2c). Moreover, western blotting revealed that CDK5 kinase activity increased with progressing embryonic development and differentiation (Fig. 2d). These data suggested a potential role for CDK5 in cortical neuronal development.

To further evaluate the role of CDK5/p25 in neuronal development, we used the RA/ BDNF-induced SH-SY5Y cell differentiation model. The mRNA levels of CDK2, CDK4, CDK6, and Cyclin E did not differ significantly in these cells before and after differentiation; however, levels of Cyclin D declined dramatically in differentiated SH-SY5Y cells (Fig. 3a, 3b). CDK5 mRNA and protein levels did not differ significantly between the two groups; however, those of p25 increased markedly. In addition, p25, a degradation product of p35, was detected in differentiated SH-SY5Y cells (Fig. 3c, 3d), and CDK5 kinase activity was upregulated (Fig. 3d).

Our previous results showed that non-canonical Wnt signaling was upregulated during neuronal development and differentiation. Next, we focused on studying the mechanism underlying non-canonical Wnt signaling in this context. We stimulated cortical neurons with different concentrations of various recombinant non-canonical pathway Wnt ligands. The results indicated that CDK5 kinase activity was upregulated in response to stimulation with Wnt-5a; however, no significant changes in CDK5 kinase activity were observed in response to Wnt-3a, Wnt-4, or Wnt-11 (Fig. 4a). Similarly, stimulation with Wnt-5a at various time points demonstrated that the CDK5 kinase activity was upregulated in cortical neurons (Fig. 4c). mRNA and protein levels of p35 were also upregulated in response to Wnt-5a (Fig. 4b, 4c). In addition, cleaved α-fodrin, a substrate of calpain, appeared on Wnt-5a stimulation (Fig. 4c); therefore, we deduced that calpain participates in Wnt-5a/CDK5 signaling in this context.

In the present study, enhanced CDK5 kinase activity in neurons was mediated by Wnt-5a; however, the mechanisms triggering CDK5 kinase activity remain unknown. Therefore, we sought to elucidate how Wnt-5a induces CDK5 kinase activity. Recent reports have highlighted that Ca²⁺/calpain signaling activity is a significant downstream target of Wnt-5a stimulation. Moreover, CDK5 signaling is regulated by Ca²⁺/calpain activity, which degrades p35 into p25, a highly active molecular chaperone involved in CDK5 activation; therefore, we hypothesized that Wnt-5a activates CDK5 through the Ca²⁺/calpain pathway. CDK5 kinase activity was upregulated by exposure of cultured cortical neurons to a calcium ionophore at different time points (0–8 h) (Fig. 5a). In addition, Wnt-5a-mediated upregulation of CDK5 kinase activity was reduced by treatment with different calcium chelators (EGTA-AM and BAPTA-AM) (Fig. 5b). Consistent with these results, we found that AK295, a calpain inhibitor, and Roscovitine (Ros) attenuated CDK5 kinase activity in differentiated SH-SY5Y cells (Fig. 5c). Moreover, numbers of differentiated SH-SY5Y cells were markedly reduced by AK295 or Ros in the RA/BDNF model (Fig. 5d).

Our previous results demonstrated that Wnt-5a could activate CDK5 kinase activity. Next, we examined whether Wnt-5a regulated neuronal differentiation through its activation of CDK5. We firstly tested that Wnt-5a downregulation using a Wnt neutralizing antibody could inhibit CDK5 kinase activity (Fig. 6a). The antibody could partially block Wnt-5a binding to Wnt and suppress SH-SY5Y cell differentiation, which was also inhibited by Box5 (a Wnt-5a antagonist), demonstrating a critical role for Wnt-5a in BDNF-induced SH-SY5Y cell differentiation (Fig. 6b, 6c). We also constructed an siRNA that specifically silenced Wnt-5a expression in SH-SY5Y cells and confirmed its knockdown efficiency in vivo (Fig. 6d). Our results verified that activation of CDK5 could be inhibited and the proportion of differentiated SH-SY5Y cells reduced by treatment with Wnt-5a-specific siRNA (Fig. 6e, 6f). Hence, Wnt-5a appears to be a critical factor in SH-SY5Y cell maturation. Collectively, these results indicate that Wnt-5a supports SH-SY5Y cell differentiation.

Previous studies have demonstrated a significant role for Wnt signaling in neuronal development and differentiation; however, little is known about β-catenin-independent Wnt signaling in this context [26, 27]. Among the effects of Wnt proteins on the nervous system, the Wnt/β-catenin pathway has a crucial role in cell proliferation and differentiation, while non-Wnt/β-catenin signaling is key for regulation of cell polarity and motility [28]. Wnt/β-catenin signaling also controls postnatal Rat epididymal development, while it has no influence on epithelial cell differentiation [29]. Low-affinity nerve growth factor receptor (LNGFR) targets the Wnt/β-catenin pathway and promotes osteogenic differentiation in rat ecto-mesenchymal stem cells [30]. Celecoxib-induced inhibition of neurogenesis in the fetal frontal cortex is attenuated by curcumin via the Wnt/β-catenin pathway; however, the expression patterns and functions of Wnt proteins in neuronal development and differentiation have yet to be elucidated. Recently, the non-Wnt/β-catenin pathway was identified as involved in embryonic development [31, 32]. The data presented in this report demonstrate the role of Wnt-5a signaling in the regulation of neuronal development and differentiation for the first time.

Individual Wnt proteins can initiate multiple downstream signaling events [33]. Several Wnt ligands, including Wnt-4a, Wnt-5a, Wnt-7a, Wnt-11, and Wnt-16, can activate the non-canonical Wnt/β-catenin pathway. Wnt-5a has been extensively studied [34]; it is required for notochord cell intercalation through the Wnt/PCP pathway in the ascidian Halocynthia roretzi [35] and can also activate the Wnt/β-catenin pathway under certain conditions [36, 37]. The Wnt/PCP pathway activates JUK/Rho, while the Wnt/Ca²⁺ pathway increases intracellular calcium concentrations and activates protein kinase C (PKC). Wnt-5a can both activate and repress Wnt/β-catenin signaling during Rat embryonic development [37]. It also regulates the expression of MESP1 and cardiomyocyte differentiation via non-Wnt/β-catenin signaling [38]. Wnt-5a is expressed at an early of in vitro osteogenesis differentiation [39]. Together, these experiments show that Wnt-5a can activate multiple receptors to influence various cellular processes during embryonic development. Therefore, we hypothesized that Wnt-5a may be important for neuronal development and differentiation. Finally, we demonstrated a relationship between Wnt-5a and CDK5 activation, where Wnt-5a regulated CDK5 activation through modulation of the Ca²⁺/calpain pathway.

Previous studies have demonstrated that regulation of CDK5 is essential for early neuronal development and differentiation. CDK5 activation, by conversion of p35 to p25, is triggered by calpain activation, which is mediated by intracellular Ca²⁺ levels [40]. Here, we studied the effect of CDK5-regulated Wnt signaling. In the Wnt/PKC-Ca²⁺ pathway, PKC triggers Ca²⁺ release and activates the transcription factor NFAT (nuclear factor of activated T-cells) [41]. Wnt/PKC-Ca²⁺ is activated via Wnt-5a; therefore, we hypothesize that Wnt-5a increases intracellular Ca²⁺ levels to activate calpain in SH-SY5Y cells stimulated by RA-BDNF. Furthermore, we provide evidence for the involvement of Wnt-5a-activated CDK5 in the regulation of neuronal development and differentiation.

Calpain, a Ca²⁺-mediated protease, is regulated by a rarely reported Wnt signaling pathway. In SW480 cells, calpain is negatively regulated by β-catenin signaling, which is downregulated by Wnt-5a. Thus, it would be of interest to verify the correlation between Wnt-5a and β-catenin. Here, we demonstrate that the upregulation of Wnt-5a is a key factor in activation of CDK5 kinase through regulation of Ca²⁺-mediated calpain activity. These results suggest that CDK5 is a key player downstream of Wnt-5a. Blocking calpain activity reduces Wnt-5a-mediated CDK5 kinase activity; therefore, our investigation of Wnt-5a in neuronal development and differentiation reveal critical mechanisms and enrich our understanding of CDK5/p25 functions. Several recent studies have shown that Wnt-5a and ROR2 (an orphan tyrosine kinase receptor) can regulate Ca²⁺/calpain activity in melanoma cells. Our future investigations will focus on the Wnt-5a/ROR2 pathway.

In the current study, we present data revealing the effects of Wnt-5a on the central nervous system, demonstrating that Wnt-5a promotes neuronal development and SH-SY5Y cell differentiation. Importantly, we also identified the cellular mechanism underlying Wnt-5a/Ca²⁺-calpain/CDK5-mediated regulation of neuronal development and differentiation.

This work was supported financially by the National Natural Science Foundation of China (No. 81601043 to Y. S. and No. 31371384 and 31571044 to B. T.), the Natural Science Foundation of Jiangsu Province (No. BK 20160548 to Y. S.), the Major Social Development Project of Zhenjiang city (No. SH2016032 to Y. S.), and Doctor startup Fund Program (No. JDFYRC-2015005 to Y. S.).

All the authors declare no competing interests.

Y. Shu and M. Xiang contributed equally to this work