Glycogen synthase kinase 3ß functions as a positive effector in the WNK signaling pathway

PLOS ONE, Mar 2018

The with no lysine (WNK) protein kinase family is conserved among many species. Some mutations in human WNK gene are associated with pseudohypoaldosteronism type II, a form of hypertension, and hereditary sensory and autonomic neuropathy type 2A. In kidney, WNK regulates the activity of STE20/SPS1-related, proline alanine-rich kinase and/or oxidative-stress responsive 1, which in turn regulate ion co-transporters. The misregulation of this pathway is involved in the pathogenesis of pseudohypoaldosteronism type II. In the neural system, WNK is involved in the specification of the cholinergic neuron, but the pathogenesis of hereditary sensory and autonomic neuropathy type 2A is still unknown. To better understand the WNK pathway, we isolated WNK-associated genes using Drosophila. We identified Glycogen synthase kinase 3ß (GSK3ß)/Shaggy (Sgg) as a candidate gene that was shown to interact with the WNK signaling pathway in both Drosophila and mammalian cells. Furthermore, GSK3ß was involved in neural specification downstream of WNK. These results suggest that GSK3ß/Sgg functions as a positive effector in the WNK signaling pathway.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0193204&type=printable

Glycogen synthase kinase 3ß functions as a positive effector in the WNK signaling pathway

March Glycogen synthase kinase 3û functions as a positive effector in the WNK signaling pathway Atsushi Sato 0 1 2 Hiroshi Shibuya 0 1 2 0 Department of Molecular Cell Biology and Joint Usage/Research Center for Intractable Diseases, Medical Research Institute, Tokyo Medical and Dental University (TMDU) , Bunkyo-ku, Tokyo , Japan 1 Aid for Scientific Research from the Ministry of Education , Culture, Sports , Science and Technology of Japan , 26460386, to Dr. Atsushi Sato; and Nanken-Kyoten, TMDU, 17M03, to Prof. Hiroshi Shibuya 2 Editor: Renping Zhou, Rutgers University , UNITED STATES The with no lysine (WNK) protein kinase family is conserved among many species. Some mutations in human WNK gene are associated with pseudohypoaldosteronism type II, a form of hypertension, and hereditary sensory and autonomic neuropathy type 2A. In kidney, WNK regulates the activity of STE20/SPS1-related, proline alanine-rich kinase and/or oxidative-stress responsive 1, which in turn regulate ion co-transporters. The misregulation of this pathway is involved in the pathogenesis of pseudohypoaldosteronism type II. In the neural system, WNK is involved in the specification of the cholinergic neuron, but the pathogenesis of hereditary sensory and autonomic neuropathy type 2A is still unknown. To better understand the WNK pathway, we isolated WNK-associated genes using Drosophila. We identified Glycogen synthase kinase 3û (GSK3û)/Shaggy (Sgg) as a candidate gene that was shown to interact with the WNK signaling pathway in both Drosophila and mammalian cells. Furthermore, GSK3û was involved in neural specification downstream of WNK. These results suggest that GSK3û/Sgg functions as a positive effector in the WNK signaling pathway. Data Availability Statement; All relevant data are within the paper Introduction The with no lysine (WNK) protein kinases are atypical members of the serine/threonine kinase family, and are conserved among many species [1±3]. The mammalian WNK family has four members: WNK1±4. WNK1 and WNK4 have been identified as causative genes of pseudohypoaldosteronism type II (PHAII) [ 4 ], and WNK1 is also a causative gene of hereditary sensory and autonomic neuropathy type 2A (HSAN2A) [ 5 ]. Several groups including ours have attempted to identify the functions of the WNK family. In the kidney, WNK1 and WNK4 phosphorylate and activate STE20/SPS1-related, proline alanine-rich kinase (SPAK) and oxidative-stress responsive 1 (OSR1) kinases, which in turn regulate various ion co-transporters [6±9]. Because knock-in mice of Wnk4D561A (the mutation found in PHAII patients) display similar phenotypes to PHAII, dysregulation of this WNK signaling pathway was thought to cause hypertension in PHAII patients [ 10 ]. In the neural system, a neural-specific alternatively spliced isoform of WNK1 is expressed, which includes the neural-specific exon HSN2. In HSAN2A patients, mutations were found in this HSN2 exon [ 5, 11 ], but HSN2 knock-out mice have no discernable morphological phenotype [ 11, 12 ]. Furthermore, in other familial HSAN2A patients, mutations were found outside the HSN2 exon in WNK1 [ 13 ]. Thus, the pathogenesis of HSAN2A remains unclear. WNK kinases are required for EGF-mediated ERK5 activation, and WNK family members are also involved in proliferation, migration, and differentiation [14±16]. Recently, we found that WNK1 and WNK4 induced Lhx8 expression and were important for neural specification [ 17 ]. Moreover, WNK was identified as a positive regulator of the Wnt signaling pathway; however, the detailed mechanisms of this are unknown [ 18 ]. Although WNK has a range of functions during many developmental processes, little is known about the components of the WNK signaling pathway, except for the main molecules WNK1/4±SPAK/OSR1. In the kidney, ASK3 inhibits WNK1 [ 19 ], and the PI3K/AKT signaling pathway activates the WNK±SPAK/ OSR1±NCC pathway [ 20 ]. Other upstream or downstream component(s) are still unknown. Glycogen synthase kinase 3û (GSK3û) is a ubiquitously expressed serine/threonine kinase that was originally identified as the regulatory kinase of glycogen synthase. Since then, GSK3û has been shown to be involved in many biological processes [ 21 ]. GSK3û plays important roles in several signaling pathways, especially PI3K/AKT and Wnt signaling pathways. In the PI3K/ AKT signaling pathway, AKT phosphorylates Ser9 of GSK3û which inhibits its activity, thus phosphorylating cyclin D1 and regulating the cell cycle [ 22 ]. In the Wnt signaling pathway, GSK3û is a major component of the destruction complex that phosphorylates û-catenin, which in turn is degraded by proteasomes [ 22 ]. In this study, we attempted to identify a new component of the WNK signaling pathway using Drosophila, and identified the shaggy gene (sgg) as a possible candidate. Sgg is a Drosophila homolog of mammalian GSK3û. We found that GSK3û worked as a positive effector downstream of WNK in both mammalian and Drosophila cells. Materials and methods Ethics statement All the animal experiments were performed under the ethical guidelines of Tokyo Medical and Dental University, and animal protocols were reviewed and approved by the animal welfare committee of the Tokyo Medical and Dental University. Fly stocks and genetics Fly strains used in this study were; Canton-S, y w, EY10165 (UAS-Wnk; Bloomington Stock Center #16970), UAS-frayS347D [ 17 ], WnkEY18 FRT2A [ 17 ], Akt104226 (Bloomington Stock Center #11627), sgg1 (Bloomington Stock Center #9095), sggM11 (Bloomington Stock Center #31308), UAS-sgg (Bloomington Stock Center #5435), hh-Gal4 (Bloomington Stock Center #67046), arm-Gal4 (Bloomington Stock Center #1560), hsGFP hsCD2(y+) M(3)i55 Tub>Gal80 FRT2A (provided by G. Struhl). We made hh-Gal4 EY10165 recombinant flies for screening. We crossed these flies with the fray mutant and confirmed the suppression as described previously [ 17 ]. For initial screening, we crossed several mutants and isolated candidate suppressor genes (data not shown), including sgg. Histology and staining All wings were mounted in GMM [ 23 ]. Images were obtained using SteREO Discovery and Axioscope microscopes (Carl Zeiss), and were processed using Axiovision with extended focus (Carl Zeiss) and Photoshop (Adobe). 2 / 11 Cultured cell lines Neuro2A cells [ 17 ] were grown in DMEM with 20% FBS. Polyethylenimine (Polysciences) was used to transfect plasmids, and the TransIT-X2 Dynamic Delivery System (Mirus Bio) was used to transfect small interfering (si)RNA and co-transfect siRNA and plasmids. The plasmids we used are: pRK5-Flag-hWNK1, pRK5-Flag-hWNK1D368A, pRK5-Flag-OSR1, pRK5T7-OSR1, pRK5-Flag-OSR1K46M, pRK5-T7-OSR1K46M, pRK5-Flag-OSR1S325D, pcDNA-FlagGSK3û, pRK5-Flag-GSK3û, pRK5-Flag-GSK3ûK85M. siRNA target sequences were described previously [ 17 ], and were as follows: mouse Osr1, 50-GAUAUUCGAUUUGAAUUUA-30; and mouse GSK3ß, 50-GAAAUGAACCCAAAUUAUA -30. To differentiate Neuro2A cells, they were induced in serum-free DMEM with 10 μM retinoic acid for 24 h. Immunoprecipitation Neruo2A cells were transfected with indicated expression vectors. Then, lysates were prepared from transfected cells, and immunoprecipitated with indicated antibodies and Protein A/G PLUS-agarose (Santa Cruz). Immunoprecipitates were subjected to SDS-PAGE and western blotting, and bands were detected by the LAS-4000 mini (GE) image analyzer. For sequential immunoprecipitation, the anti-Flag antibody (M2, Sigma) was used and eluted with Flag peptides (Sigma). Eluates were divided and immunoprecipitated with anti-T7 antibodies and control mouse normal IgG. RT-PCR analysis Total RNA was isolated by TRIzol (Invitrogen). cDNA synthesis was carried out using Moloney murine leukemia virus reverse transcriptase (Invitrogen). GAPDH was used for the normalization of cDNA samples. Primer pairs were described previously [ 17 ], and were as follows: mouse GSK3ß, 50-GCAGCAAGGTAACCACAGTAGTGGC-30 and 50-TGGTGCCCT GTAGTACCGAGAACAG-30. In vitro kinase assay Neuro2A cells were transfected with Flag-WNK1, Flag-WNK1D368A, Flag-OSR1, FlagOSR1K46M, or Flag-OSR1S325D expression plasmids. Lysates were prepared from transfected cells and immunoprecipitated with an anti-Flag M2 antibody (Sigma) and Protein A/G PLUSagarose (Santa Cruz). Immunoprecipitates were incubated with bacterially-expressed GST fusion proteins (GST-GSK3û K85M) in kinase buffer containing 10 mM HEPES (pH 7.4), 1 mM DTT, 5 mM MgCl2, and 5 μCi [γ-32P]-ATP at 30ÊC. Phosphorylated substrates were subjected to SDS-PAGE, and bands were detected by the FLA3000 image analyzer (Fujifilm). Antibodies Antibodies used in this report were: mouse anti-Flag M2 (Sigma; 1:400 for immunoprecipitation), rabbit anti-Flag (Sigma; 1:1000 for western blotting), rat anti-HA (Roche; 1:1000 for western blotting), mouse anti-T7 (Merck; 1:2000 for immunoprecipitation), rabbit anti-T7 (MBL; 1:1000 for western blotting), anti-rabbit HRP-conjugated (GE; 1:10000 for western blotting), and anti-rat HRP-conjugated (GE; 1:10000 for western blotting). Quantification and statistical analysis Quantitative PCR was performed with an Applied Biosystems 7300 Real-Time PCR Cycler (ABI) using THUNDERBIRD SYBR qPCR Mix (TOYOBO). Primer sequences for Lhx8, ChAT, Gad1, and GAPDH were described previously [ 17 ]. GAPDH was used for the normalization of 3 / 11 cDNA samples. Neurite lengths were measured using ImageJ software (NIH). Data were computed using Microsoft Excel (Microsoft) and StatPlus (AnalystSoft). Values and error bars represent the means and SDs, and are representative of at least three independent experiments. Results Shaggy is a novel candidate effector of the WNK signaling pathway Overexpression of Drosophila WNK using hh-Gal4 driver resulted in an ectopic vein around vein 5 in the adult wing (Fig 1B compared with Fig 1A; [ 17 ]). As we showed previously, a heterozygous mutation of fray, which encodes a Drosophila homolog of SPAK/OSR1 that is a downstream effector of WNK, suppressed this phenotype [ 17 ]. Therefore, we performed screening to identify a new effector of the WNK signaling pathway using this system. We selected several mutants known to be components of other signaling pathways, such as the Wnt pathway, Notch pathway, TGFû pathway, and the EGF pathway. We obtained several candidate suppressor genes, including sgg, which encodes the Drosophila homolog of mammalian GSK3û whereas some genes, including Akt1, did not show any interaction (Fig 1C and data not shown). Two independent sgg mutants (sgg1 and sggM11) suppressed the wing phenotypes by the overexpression of Wnk (Fig 1D and 1E), suggesting that sgg is a suppressor of the WNK signaling pathway and that this suppression is not an effect of the genetic background. We further tested the interaction between sgg and fray. Ectopic expression of frayS347D (the constitutively active form of fray) resulted in a similar phenotype to that seen following the ectopic expression of Wnk (Fig 1F; [ 17 ]). The sgg1 mutant also repressed these phenotypes (Fig 1G) suggesting that sgg interacts with the WNK signaling pathway in Drosophila. We next confirmed this genetic interaction between Wnk and sgg. Because WnkEY18 mutant clones led to abdominal developmental defects (Fig 1H; [ 17 ]), we attempted to rescue this phenotype by the overexpression of sgg. Using a combination of the FLP/FRT mosaic system and Gal80 suppression, we induced the local expression of sgg in WnkEY18 minute clones. As shown in Fig 1I, sgg overexpression partially rescued the abdominal phenotype of the WnkEY18 minute clones. These results suggest that sgg is a novel effector of the WNK signaling pathway, not only in wing development but also in abdominal development. GSK3û functions as a positive effector downstream of WNK Because GSK3ß and the WNK pathway are highly conserved among many species [ 2, 3, 22 ], we next examined whether the interaction between WNK and GSK3û was also conserved in mammalian cells. In Neuro2A cells, WNK1 expression induced the expression of Lhx8 ([ 17 ]; see also Fig 2B lane 2). As shown in Fig 2, the expression of GSK3û also induced the expression of Lhx8 in Neuro2A cells (Fig 2A lane 2). However, the kinase dead form of GSK3û (GSK3û K85M) could not activate the expression of Lhx8, suggesting that GSK3û kinase activity is required for its expression (Fig 2A lane 3). Next, we examined the epistatic interaction between WNK1 and GSK3û. The induction of Lhx8 was suppressed by the knockdown of GSK3ß (Fig 2B lane 4). However, Lhx8 induction by GSK3û was not suppressed by the knockdown of both Wnk1 and Wnk4 (Fig 2C lanes 5 and 6), even though this knockdown did suppress Lhx8 induction by retinoic acid (RA) stimulation (Fig 2C lanes 3 and 4). The expression of OSR1, a downstream molecule of WNK, and its constitutively active form (OSR1S325D) induced the expression of Lhx8 (Fig 2D lanes 3 and 5; [ 17 ]). This activation was also suppressed by the knockdown of GSK3ß (Fig 2D lanes 4 and 6). In contrast, the induction of Lhx8 by GSK3û was not suppressed by the knockdown of Osr1 (Fig 2E lanes 5 and 6), although the induction of Lhx8 by WNK1 was suppressed by the knockdown of Osr1 (Fig 2E 4 / 11 Fig 1. sgg is downstream of Wnk in the Drosophila wing vein and abdominal patterning. (A) Wild-type wing. (B) Wing from EY10165 (UAS-Wnk) fly driven by hh-Gal4. Additional veins around vein 5 (arrowhead) were observed. (C) Wing from fly overexpressing Wnk driven by hh-Gal4 with the Akt104226 heterozygous mutant. (D) Wing from fly overexpressing Wnk driven by hh-Gal4 with the sgg1 heterozygous mutant. (E) Wing from fly overexpressing Wnk driven by hh-Gal4 with the sggM11 heterozygous mutant. (F) Wing from UAS-frayS347D fly driven by hh-Gal4. Additional veins around vein 5 (arrowhead) were observed. (G) Wing from fly overexpressing frayS347D driven by hhGal4 with the sgg1 heterozygous mutant. (H) Abdomen from adult fly with DWNKEY18 minute clones. Thin black lines indicate the clone border. (I) Abdomen from adult fly with DWNKEY18 minute clones and sgg overexpression. sgg was expressed only in DWNKEY18 minute clones using the Gal80 suppression technique. Thin black lines indicate the clone border (also the sgg expression area). Black arrowheads show rescued abdominal bristles. The detailed genotype is y w hsflp; arm-Gal4 / UAS-sgg; WnkEY18 FRT2A / hsGFP hsCD2(y+) M(3)i55 Tub>Gal80 FRT2A. The numbers of wings or abdomina showing the phenotypes and of total observed wings or abdomina are indicated. lanes 3 and 4). These data suggest that the WNK±OSR1±GSK3û pathway is conserved not only in flies but also in mammals, and that GSK3û functions as a positive effector downstream of the WNK signaling pathway. 5 / 11 Fig 2. GSK3û is a positive effector downstream of WNK-OSR1. (A) Gene expression determined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells overexpressing GSK3û or GSK3û K85M. The value obtained from each sample was normalized to that of GAPDH. The value of Lhx8 from overexpressing GSK3û (lane 2) was set to 100. (B) Gene expression detemined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells overexpressing hWNK1 under GSK3ß knockdown using siRNA. The value obtained from each sample was normalized to that GAPDH. The value of Lhx8 from overexpressing hWNK1 (lane 2) was set to 100. (C) Gene expression determined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells stimulated by retinoic acid (RA) or overexpressing GSK3û under both Wnk1 and Wnk4 knockdown using siRNA. The value obtained from each sample was normalized to that of GAPDH. The value of Lhx8 from overexpressing GSK3û (lane 3) was set to 100. (D) Gene expression determined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells overexpressing OSR1 or OSR1S324D (constitutively active form of OSR1) under GSK3ß knockdown using siRNA. The value obtained from each sample was normalized to that of GAPDH. The value of Lhx8 from overexpressing OSR1 (lane 3) was set to 100. (E) Gene expression determined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells overexpressing hWNK1 or GSK3û under Osr1 knockdown using siRNA. The value obtained from each sample was normalized to that of GAPDH. The value of Lhx8 from overexpressing hWNK1 (lane 3) was set to 100. WNK1 and OSR1 form a complex with GSK3û We next investigated the biochemical interaction between WNK1 and GSK3û. We transiently expressed Flag-tagged GSK3û together with HA-tagged WNK1 or T7-tagged OSR1. When cell extracts were subjected to immunoprecipitation with anti-Flag, anti-HA (for WNK1), or anti6 / 11 Fig 3. GSK3û forms a complex with WNK1, but is not directly phosphorylated. (A±B) Interactions between WNK1 (A) or OSR1 (B), and GSK3û were examined in Neuro2A cells by co-immunoprecipitation. Immunoprecipitates (IP) were subjected to western blotting (WB) with the indicated antibodies. +, present; -, absent. (C) Interaction among WNK1, OSR1 and GSK3û were examined in Neuro2A cells by sequential immunoprecipitation. An anti-FLAG antibody was used for the first immunoprecipitation, and immunoprecipitates were subjected to a second immunoprecipitation with IgG or T7 antibodies. (D) Phosphorylation of GSK3û by WNK1 or OSR1. Upper panel shows the result of an in vitro kinase assay. Black arrow in the left upper panel (32P-Flag-WNK1/DA) represents the auto-phosphorylation of WNK1. Black arrow in the right upper panel (32P-Flag-OSR1/KM/SD) represents the auto-phosphorylation of OSR1. Grey arrows in both left and right upper panels indicated the size of GSK3û (32P-GST-GSK3û K85M). Lower panels show the total protein. Arrowheads in both left and right lower panels represent the indicated proteins. T7 (for OSR1) antibodies followed by immunoblotting, we found that GSK3û interacted with WNK1 (Fig 3A lane 3), but not with OSR1 (Fig 3B lane 3). Because WNK1 bound to OSR1 [ 6, 7 ], we investigated whether the WNK1±OSR1 complex interacted with GSK3û. As shown in Fig 3C, GSK3û was immunoprecipitated after sequential IP by anti-Flag and anti-T7 antibodies, suggesting that GSK3û forms a complex with WNK1 and OSR1. GSK3û is positively and negatively regulated by phosphorylation [ 24 ]. Because WNK1 and OSR1 are both Ser/Thr kinases [ 3, 25, 26 ], and WNK1±OSR1±GSK3û forms the complex shown above, we examined whether WNK1 or OSR1 directly phosphorylated and regulated GSK3û. To perform the in vitro kinase assay, we purified Flag-tagged WNK1 or OSR1 from cultured cell extracts, and produced a GST-fusion protein of the kinase dead form of GSK3û (GST-GSK3û K85M) in bacteria. We did not observe phosphorylation of GSK3û by WNK1 or OSR1 (Fig 3D lanes 1±5). These results suggest that GSK3û forms a complex with WNK1 and OSR1, but that the regulation of GSK3û by the WNK signaling pathway does not depend on direct phosphorylation. GSK3û is involved in neural specification As we showed previously [ 17 ], WNK plays an important role in neural specification through the regulation of Lhx8 expression. We examined whether GSK3û was also involved in neural 7 / 11 specification downstream of WNK. Knockdown of GSK3ß caused the shortening of neurites after RA stimulation (Fig 4B compared with Fig 4A, quantified in Fig 4C). Knockdown of GSK3ß also decreased the expression of Lhx8 and the choline acetyltransferase gene (ChAT; a marker for cholinergic neuron) (Fig 4D lane 4). However, the gene expression of glutamic acid decarboxylase 1 (Gad1; a marker for GABAergic neurons) increased (Fig 4D lane 4). These Fig 4. The WNK-OSR-GSK3û pathway is involved in the neural development. (A±B) siRNA-treated differentiated Neuro2A cells induced by RA for 24 h; (A) Control siRNA, (B) siGSK3ß. (C) The average length of neurites in siRNAtreated differentiated Neuro2A cells induced by RA for 24 h, shown in A and B (Control siRNA (n = 93), siGSK3ß (n = 91)). p<0.0005 calculated by the Student's t-test. (D) Gene expression determined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells. Cells treated with siRNA against GSK3ß (siGSK3ß); (lanes 1 and 2) undifferentiated cells, (lanes 3 and 4) cells differentiated by RA for 24 h. The value obtained from each sample was normalized to that of GAPDH. The value of Lhx8, ChAT or Gad1 from differentiated cells under control siRNA treatment (lane 3) was set to 100. (E±H) Differentiated Neuro2A cells were transfected with various combinations of siRNAs and expression plasmids. (E) Control siRNA and control vector. (F) Control siRNA and GSK3û. (G) siWnk1 and siWnk4, and control vector. (H) siWnk1 and siWnk4, and GSK3û. (I) The average length of neurites in siRNAtreated differentiated Neuro2A cells induced by RA for 24 h, shown in E-H (Control siRNA and Control vector (n = 81), Control siRNA and GSK3û (n = 87), siWnk1 and siWnk4, and Control vector (n = 103), siWnk1 and siWnk4, and GSK3û (n = 77)). p<0.0005 calculated by the Bonferroni correction. ns indicated non-significance. (J) Gene expression determined by RT-PCR or quantitative RT-PCR analysis was examined in Neuro2A cells. Cells were treated with various combinations of siRNAs and expression plasmids; (lanes 1±4) undifferentiated cells, (lanes 5±8) cells differentiated by RA for 24 h, (lanes 1±2 and 5±6) control siRNA, (lanes 3±4 and 7±8) siWnk1 and siWnk4, (lanes 1, 3, 5 and 7) control vector, (lanes 2, 4, 6 and 8) GSK3û. The value obtained from each sample was normalized to that of GAPDH. The value of Lhx8, ChAT or Gad1 from differentiated cells under the treatment of control siRNA (lane 5) was set to 100. 8 / 11 results suggest that GSK3û is involved in neural specification, similar to that induced by the knockdown of both Wnk1 and Wnk4 as shown previously ([ 17 ]; see also Fig 4G and 4J). We next examined whether the expression of GSK3û suppressed the neural specification phenotypes caused by the knockdown of Wnk. While the expression of GSK3ß did not affect the elongation of neurites (Fig 4F compared with Fig 4E, quantified in Fig 4I), the expression of GSK3ß induced Lhx8 and ChAT expression, and reduced Gad1 expression after RA stimulation in Neuro2A cells (Fig 4J lanes 5 and 6). Under conditions of both Wnk1 and Wnk4 knockdown, the expression of GSK3ß partially rescued the elongation of neurites (Fig 4H compared with Fig 4G, summarized in Fig 4I), and Lhx8 expression (Fig 4J lanes 7 and 8); this in turn increased ChAT expression and decreased Gad1 expression (Fig 4J lanes 7 and 8). These results suggest that GSK3û is involved in neural development and functions downstream of the WNK signaling pathway. Discussion The WNK signaling pathway is involved in many biological processes, but the details of its components are unclear, except for in the kidney. Here, we screened candidate genes that genetically interact with the WNK signaling pathway in Drosophila. Among these, we identified shaggy, which encodes the Drosophila homolog of mammalian GSK3û (Fig 1). We showed that GSK3û activated Lhx8 expression and that GSK3û functions downstream of the WNK± OSR1 pathway by epistasis analysis (Fig 2). We also showed that GSK3û might form a tertiary complex with WNK1 and OSR1 (Fig 3). Furthermore, GSK3û was found to be involved in neural specification and neurite elongation (Fig 4), and GSK3û rescued the neural phenotypes induced by the knockdown of both Wnk1 and Wnk4 (Fig 4). However, we did not observe direct phosphorylation of GSK3û by WNK1 or OSR1 (Fig 3). This suggests that GSK3û functions as a positive downstream effector in the WNK signaling pathway, although the regulation of GSK3û activity by the signaling pathway remains unclear and requires further study to elucidate how WNK±OSR1 transduces the signal to GSK3û. GSK3û plays many roles in various signaling pathways. In the PI3K/AKT signaling pathway, AKT phosphorylates Ser-9 of GSK3û and inhibits GSK3û activity for cell proliferation [ 22 ]. Previous research has shown that the PI3K/AKT signaling pathway activates the WNK± OSR1±NCC pathway to regulate blood pressure [ 20 ]. However, GSK3 was reported to be a negative regulator for the destruction complex in the Wnt signaling pathway [ 22 ]. A recent study showed that WNK is a positive regulator of the Wnt signaling pathway [ 18 ]. Here, we found that GSK3û is a positive regulator downstream of the WNK±SPAK/OSR1 signaling pathway. These contradictions with regard to the regulation and role of GSK3û clearly indicate that the WNK±SPAK/OSR1±GSK3û signaling pathway for neural development is independent of PI3K/AKT or Wnt signaling pathways. The exact interaction between WNK and other signaling pathways remains to be determined and will require further analyses. GSK3û is also known to be involved in the Notch signaling pathway in which the intercellular domain (ICD) of Notch directly regulates the transcription of target genes with several cofactors [ 27 ]. GSK3û was previously shown to bind and phosphorylate the Notch ICD which increased the transcriptional activity of the Notch ICD complex [ 28 ]. However, the mechanisms of GSK3û activation in the Notch signaling pathway are still unclear. Since our initial screening of Drosophila Wnk-related genes showed that Wnk has a weak genetic interaction with the Notch signaling pathway (data not shown), we hypothesize that the WNK pathway positively regulates the Notch signaling pathway through GSK3û in neural development. Further study will be required to prove this hypothesis, which is likely to be important for understanding the pathogenesis of HSAN2A. 9 / 11 Acknowledgments We thank Gary Struhl and the Bloomington stock center for fly stocks, Toshiyasu Goto for helpful discussion, and Yoko Mitsutomo for technical assistance. We thank Sarah Williams, PhD, from Edanz Group (www.edanzediting.com) for editing a draft of this manuscript. Author Contributions Conceptualization: Hiroshi Shibuya. Data curation: Atsushi Sato, Hiroshi Shibuya. Formal analysis: Atsushi Sato. Supervision: Hiroshi Shibuya. Validation: Atsushi Sato. Visualization: Atsushi Sato. Writing ± original draft: Atsushi Sato. Writing ± review & editing: Hiroshi Shibuya. 10 / 11 1. Moniz S , Jordan P . Emerging roles for WNK kinases in cancer . Cell Mol Life Sci . 2010 ; 67 ( 8 ): 1265 ± 76 . https://doi.org/10.1007/s00018-010 -0261-6 PMID: 20094755 2. VerÂõssimo F , Jordan P. WNK kinases, a novel protein kinase subfamily in multi-cellular organisms . Oncogene . 2001 ; 20 ( 39 ): 5562 ±9. https://doi.org/10.1038/sj.onc. 1204726 PMID: 11571656 3. Xu B , English JM , Wilsbacher JL , Stippec S , Goldsmith EJ , Cobb MH . WNK1, a novel mammalian serine/threonine protein kinase lacking the catalytic lysine in subdomain II . J Biol Chem . 2000 ; 275 ( 22 ): 16795 ± 801 . PMID: 10828064 4. Wilson FH , Disse-Nicodème S , Choate KA , Ishikawa K , Nelson-Williams C , Desitter I , et al. Human hypertension caused by mutations in WNK kinases . Science . 2001 ; 293 ( 5532 ): 1107 ± 12 . https://doi.org/ 10.1126/science.1062844 PMID: 11498583 5. Shekarabi M , Girard N , Rivière JB , Dion P , Houle M , Toulouse A , et al. Mutations in the nervous system Ðspecific HSN2 exon of WNK1 cause hereditary sensory neuropathy type II . J Clin Invest . 2008 ; 118 ( 7 ): 2496 ± 505 . https://doi.org/10.1172/JCI34088 PMID: 18521183 6. Moriguchi T , Urushiyama S , Hisamoto N , Iemura S , Uchida S , Natsume T , et al. WNK1 regulates phosphorylation of cation-chloride-coupled cotransporters via the STE20-related kinases, SPAK and OSR1 . J Biol Chem . 2005 ; 280 ( 52 ): 42685 ± 93 . https://doi.org/10.1074/jbc.M510042200 PMID: 16263722 7. Vitari AC , Deak M , Morrice NA , Alessi DR . The WNK1 and WNK4 protein kinases that are mutated in Gordon's hypertension syndrome phosphorylate and activate SPAK and OSR1 protein kinases . Biochem J . 2005 ; 391 (Pt 1): 17 ± 24 . https://doi.org/10.1042/BJ20051180 PMID: 16083423 8. Gagnon KB , England R , Delpire E . Volume sensitivity of cation-Cl- cotransporters is modulated by the interaction of two kinases: Ste20-related proline-alanine-rich kinase and WNK4 . Am J Physiol Cell Physiol . 2006 ; 290 ( 1 ):C134± 42 . https://doi.org/10.1152/ajpcell.00037. 2005 PMID: 15930150 9. Piechotta K , Lu J , Delpire E. Cation chloride cotransporters interact with the stress-related kinases Ste20-related proline-alanine-rich kinase (SPAK) and oxidative stress response 1 (OSR1) . J Biol Chem . 2002 ; 277 ( 52 ): 50812 ±9. https://doi.org/10.1074/jbc.M208108200 PMID: 12386165 10. Yang SS , Morimoto T , Rai T , Chiga M , Sohara E , Ohno M , et al. Molecular pathogenesis of pseudohypoaldosteronism type II: generation and analysis of a Wnk4(D561A/+) knockin mouse model . Cell Metab . 2007 ; 5 ( 5 ): 331 ± 44 . https://doi.org/10.1016/j.cmet. 2007 . 03 .009 PMID: 17488636 11. Rivière JB , Verlaan DJ , Shekarabi M , Lafrenière RG , BeÂnard M , Der Kaloustian VM , et al. A mutation in the HSN2 gene causes sensory neuropathy type II in a Lebanese family . Ann Neurol . 2004 ; 56 ( 4 ): 572 ± 5. https://doi.org/10.1002/ana.20237 PMID: 15455397 12. Kahle KT , Schmouth JF , Lavastre V , Latremoliere A , Zhang J , Andrews N , et al. Inhibition of the kinase WNK1/HSN2 ameliorates neuropathic pain by restoring GABA inhibition . Sci Signal . 2016 ; 9 ( 421 ):ra32. https://doi.org/10.1126/scisignal.aad0163 PMID: 27025876 13. Rotthier A , Baets J , De Vriendt E , Jacobs A , Auer-Grumbach M , LeÂvy N , et al. Genes for hereditary sensory and autonomic neuropathies: a genotype-phenotype correlation . Brain . 2009 ; 132 (Pt 10): 2699 ± 711 . https://doi.org/10.1093/brain/awp198 PMID: 19651702 14. Sun X , Gao L , Yu RK , Zeng G . Down-regulation of WNK1 protein kinase in neural progenitor cells suppresses cell proliferation and migration . J Neurochem . 2006 ; 99 ( 4 ): 1114 ± 21 . https://doi.org/10.1111/j. 1471- 4159 . 2006 . 04159 . x PMID : 17018027 15. Tu SW , Bugde A , Luby-Phelps K , Cobb MH . WNK1 is required for mitosis and abscission . Proc Natl Acad Sci U S A . 2011 ; 108 ( 4 ): 1385 ± 90 . https://doi.org/10.1073/pnas.1018567108 PMID: 21220314 16. BjoÈrklund M , Taipale M , Varjosalo M , Saharinen J , LahdenperaÈ J , Taipale J . Identification of pathways regulating cell size and cell-cycle progression by RNAi . Nature . 2006 ; 439 ( 7079 ): 1009 ± 13 . https://doi. org/10.1038/nature04469 PMID: 16496002 17. Sato A , Shibuya H. WNK Signaling Is Involved in Neural Development via Lhx8/Awh Expression . PLoS One . 2013 ; 8 ( 1 ):e55301. https://doi.org/10.1371/journal.pone. 0055301 PMID: 23383144 18. Serysheva E , Berhane H , Grumolato L , Demir K , Balmer S , Bodak M , et al. Wnk kinases are positive regulators of canonical Wnt/β-catenin signalling . EMBO Rep . 2013 ; 14 ( 8 ): 718 ± 25 . https://doi.org/10. 1038/embor. 2013 .88 PMID: 23797875 19. Naguro I , Umeda T , Kobayashi Y , Maruyama J , Hattori K , Shimizu Y , et al. ASK3 responds to osmotic stress and regulates blood pressure by suppressing WNK1-SPAK/OSR1 signaling in the kidney . Nat Commun . 2012 ; 3 : 1285 . https://doi.org/10.1038/ncomms2283 PMID: 23250415 20. Nishida H , Sohara E , Nomura N , Chiga M , Alessi DR , Rai T , et al. Phosphatidylinositol 3-kinase/Akt signaling pathway activates the WNK-OSR1/SPAK-NCC phosphorylation cascade in hyperinsulinemic db/ db mice . Hypertension . 2012 ; 60 ( 4 ): 981 ± 90 . https://doi.org/10.1161/HYPERTENSIONAHA.112. 201509 PMID: 22949526 21. McCubrey JA , Steelman LS , Bertrand FE , Davis NM , Sokolosky M , Abrams SL , et al. GSK-3 as potential target for therapeutic intervention in cancer . Oncotarget . 2014 ; 5 ( 10 ): 2881 ± 911 . https://doi.org/10. 18632/oncotarget.2037 PMID: 24931005 22. Woodgett JR . Judging a protein by more than its name: GSK-3 . Sci STKE . 2001 ; 2001 (100):re12. https://doi.org/10.1126/stke. 2001 . 100 .re12 PMID: 11579232 23. Lawrence PA , and Johnston P. Methods of marking cells . In: Roberts DB, editor. Drosophila: A Practical Approach . Oxford: IRL Press; 1986 . p. 229 ± 42 . 24. Jope RS , Yuskaitis CJ , Beurel E. Glycogen synthase kinase-3 (GSK3): inflammation, diseases, and therapeutics . Neurochem Res . 2007 ; 32 ( 4 ±5): 577 ± 95 . https://doi.org/10.1007/s11064-006 -9128-5 PMID: 16944320 25. Johnston AM , Naselli G , Gonez LJ , Martin RM , Harrison LC , DeAizpurua HJ . SPAK, a STE20/SPS1- related kinase that activates the p38 pathway . Oncogene . 2000 ; 19 ( 37 ): 4290 ±7. https://doi.org/10. 1038/sj.onc. 1203784 PMID: 10980603 26. Tamari M , Daigo Y , Nakamura Y. Isolation and characterization of a novel serine threonine kinase gene on chromosome 3p22-21.3 . J Hum Genet . 1999 ; 44 ( 2 ): 116 ± 20 . https://doi.org/10.1007/ s100380050121 PMID: 10083736 27. Artavanis-Tsakonas S , Rand MD , Lake RJ . Notch signaling: cell fate control and signal integration in development . Science . 1999 ; 284 ( 5415 ): 770 ± 6 . PMID: 10221902 28. Foltz DR , Santiago MC , Berechid BE , Nye JS . Glycogen synthase kinase-3beta modulates notch signaling and stability . Curr Biol . 2002 ; 12 ( 12 ): 1006 ± 11 . PMID: 12123574


This is a preview of a remote PDF: http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0193204&type=printable

Atsushi Sato, Hiroshi Shibuya. Glycogen synthase kinase 3ß functions as a positive effector in the WNK signaling pathway, PLOS ONE, 2018, DOI: 10.1371/journal.pone.0193204