G Model

ARTICLE IN PRESS

BC 4555 1–10

The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel

PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling

1

2

Q1

3 4 5

Janice Hiu Chor Plani-Lam a , Tai Cheong Chow a , Kam Leung Siu a , Wing Hin Chau a , Ming-Him James Ng a,b , Suying Bao a , Cheung Toa Ng a , Pak Sham c,d , Daisy Kwok Yan Shum a , Evan Ingley e , Dong Yan Jin a , You-Qiang Song a,∗ a

6

Department of Biochemistry, University of Hong Kong, 21 Sassoon Road, Hong Kong, China Poison Treatment Centre, Department of Medicine and Therapeutics, Prince of Wales Hospital, Shatin, Hong Kong, China c Department of Psychiatry, University of Hong Kong, 21 Sassoon Road, Hong Kong, China d Centre for Genomic Sciences, University of Hong Kong, 5 Sassoon Road, Hong Kong, China e Harry Perkins Institute of Medical Research, QEII Medical Centre, Nedlands and Centre for Medical Research, The University of Western Australia, Crawley, Western Australia, 6009, Australia b

7 8 9 10 11 12

13 28

a r t i c l e

i n f o

a b s t r a c t

14 15 16 17 18 19

Article history: Received 11 October 2014 Received in revised form 27 January 2015 Accepted 3 February 2015 Available online xxx

20

Keywords: PTPN21 NRG3 23 ErbB4 24 Elk-1 25 Pro-neuronal survival 26 27 Q2 Neuritic elongation 21 22

29

Although expression quantitative trait locus, eQTL, serves as an explicit indicator of gene–gene associations, challenges remain to disentangle the mechanisms by which genetic variations alter gene expression. Here we combined eQTL and molecular analyses to identify an association between two seemingly non-associated genes in brain expression data from BXD inbred mice, namely Ptpn21 and Nrg3. Using biotinylated receptor tracking and immunoprecipitation analyses, we determined that PTPN21 de-phosphorylates the upstream receptor tyrosine kinase ErbB4 leading to the up-regulation of its downstream signaling. Conversely, kinase-dead ErbB4 (K751R) or phosphatase-dead PTPN21 (C1108S) mutants impede PTPN21-dependent signaling. Furthermore, PTPN21 also induced Elk-1 activation in embryonic cortical neurons and a novel Elk-1 binding motif was identified in a region located 1919 bp upstream of the NRG3 initiation codon. This enables PTPN21 to promote NRG3 expression through Elk-1, which provides a biochemical mechanism for the PTPN21–NRG3 association identified by eQTL. Biologically, PTPN21 positively influences cortical neuronal survival and, similar to Elk-1, it also enhances neuritic length. Our combined approaches show for the first time, a link between NRG3 and PTPN21 within a signaling cascade. This may explain why these two seemingly unrelated genes have previously been identified as risk genes for schizophrenia. © 2015 Published by Elsevier Ltd.

1. Introduction

Q3

The neurotrophic factor Neuregulin 3 (NRG3), has been iden31Q4 tified as a risk gene for Schizophrenia in multiple genome-wide 30

Abbreviations: NRG3, Neuregulin 3; eQTL, expression quantitative trait locus; SNPs, single nucleotide polymorphisms; PTPN21, protein tyrosine phosphatase nonreceptor 21; GWAS, genome-wide association study; RTK, receptor tyrosine kinase; FAK, focal adhesion kinase; KIF1C, tyrosine-phosphorylated kinesin-like Protein; Etk, Tec tyrosine kinase; Elk-1, ETS domain-containing protein; ErbB4, V-erb-b2 avian erythroblastic leukemia viral oncogene homolog 4; IP, immunoprecipitation; p-Tyr, tyrosine phosphatase; RT-qPCR, quantitative reverse transcriptase-PCR. ∗ Corresponding author at: Room L3-63, Department of Biochemistry, University of Hong Kong, 21 Sassoon Road, Hong Kong, China. Tel.: +852 2819 9245; fax: +852 2855 1254. E-mail address: [email protected] (Y.-Q. Song).

association studies in Ashkenazi Jewish (Chen et al., 2009), Australian (Morar et al., 2011), German (Meier et al., 2013) and Han Chinese (Wang et al., 2008) populations. The presence of other important genetic components is indicated by numerous other genomic studies, including a recent GWAS meta-analyses that identified protein tyrosine phosphatase non-receptor 21 (PTPN21) as a potential risk gene for schizophrenia (Chen et al., 2011). This genetic study identified two missense non-synonymous single nucleotide polymorphisms (SNPs) in PTPN21, rs2401751 and rs2274736, which change the size and alter the hydrophobicity of residues that may impact enzyme activity. Intriguingly, an independent quantitative trait locus (eQTL), of brain in BXD recombinant inbred mice also identified that the genetic variants of PTPN21 mediate Nrg3 expression in mouse neurons. This eQTLs mapping method revealed association between genetic polymorphisms variation and mRNA expression variation (Michaelson et al.,

http://dx.doi.org/10.1016/j.biocel.2015.02.003 1357-2725/© 2015 Published by Elsevier Ltd.

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

G Model BC 4555 1–10

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

2

2009). Since the results of both eQTL and multiple schizophrenia genome-wide association study (GWAS) studies are significantly in support of Ptpn21–Nrg3 association, we hypothesized that the 50 expression of PTPN21 would facilitate the transcription of neu51 rotrophic factor NRG3 in a targeted manner. 52 PTPN21 was first discovered in human bone by Moller et al. 53 (1994). While the function of PTPN21 is not fully understood, 54 it has been suggested that it to directly dephosphorylates an 55 56Q5 unidentified receptor tyrosine kinase (RTK) (Carlucci et al., 2008). This twenty-first member of the PTP family is known to inter57 act with Focal Adhesion kinase (FAK) (Carlucci et al., 2008) and 58 tyrosine-phosphorylated kinesin-like protein (KIF1C) (Dorner et al., 59 1998) which drives constitutive Src signal activation (Cardone et al., 60 2004; Livigni et al., 2006; Moller et al., 1994) as well as promot61 ing the activity of ETS domain-containing protein (Elk-1) (Cardone 62 et al., 2004) and Tec tyrosine kinase (Etk) (Jui et al., 2000) through 63 the regulation of the MEK-Elk-1 signaling pathway. Yet the ques64 tion remains as to how a phosphatase, such as PTPN21, normally 65 associated with switching off signal transduction cascades, pro66 motes Src, Elk-1 and Etk pathways. Consequently, there is a need 67 to identify the RTK that is dephosphorylated by PTPN21, in order to 68 improve our understanding of this phosphatase as a mediator for 69 positive Src/Elk-1/Etk signaling. 70 Here we first specifically investigate whether a genetic variant 71 of Ptpn21, identified in eQTL brain expression data, could regulate 72 Nrg3 expression, in mouse cortical neurons and a human cell line. 73 Subsequently, we elucidated the mechanism linking PTPN21–NRG3 74 by identifying the unknown RTK that PTPN21 interacts with, as 75 V-erb-b2 avian erythroblastic leukemia viral oncogene homolog 4 76 (ErbB4). Biotinylated receptor tracking assays and immunoprecip77 itation analyses revealed that PTPN21 binds ErbB4 at the plasma 78 membrane, resulting in reduced phospho-ErbB4 levels, which leads 79 to ErbB4 ubiquitination resistance and its accumulation. In addi80 tion, our results revealed that PTPN21 shares similar biological roles 81 as Elk-1 in promoting neuronal survival and neuritic elongation 82 (Lavaur et al., 2007). When used in conjunction, the bioinformat83 ics and molecular biology approaches made it possible for us to 84 demonstrate the importance of PTPN21 in the regulation and main85 tenance of neurotrophic factor NRG3 levels in primary embryonic 86 mouse cortical neurons. This, therefore, provides the biochemical 87 link that underlies the bioinformatic association between NRG3 and 88 PTPN21 as risk genes for schizophrenia. 89 48 49

90

2. Materials and methods

91

2.1. Ethics statement

92 93 94 95

96 97

98 99 100 101 102 103 104 105 106 107

All experiments were conducted in accordance with the Code of Practice for the Care and Use of Animals were approved by the CULATR Ethics Committee for animal experimentation (CULATR 1745-08 and 2910-12). 2.2. BXD mice and expression quantitative trait locus database (eQTL) model Transcription data were collected from brain expression data for 48 BXD recombinant inbred mature mice strains (Peirce et al., 2004), DBA/2J, C57BL/6J and F1 hybrids. Data were processed by The University of Tennessee Health Science Center, and the St. Jude Children’s Research Hospital. Brain samples were hybridized to Affymetrix M430A and M430B arrays in triplicate. Reduced Major Axis (RMA) and Position-Dependent Nearest Neighbor (PDNN) protocols served as the platforms to generate association data. RMA and PDNN values of each array were adjusted according to the database developer. The abscissa axis represents

chromosome location (in mega base basis) and the ordinate axis represents the value of likelihood ratio statistics (LSR, LOD × 4.61). The bars indicate the “best” QTLs from 2000 bootstrap resampling tests. The protocol was as stated on WebQTL database resource (www.genenetwork.org) and the Mouse Genome Informatics (MGI) database resource (www.informatics.jax.org) with GeneNetwork accession number GN113.

2.3. Murine embryonic cortical neuron extraction and cell culture Cortical neurons were isolated from the cortex of embryonic 14.5 d.p.c C57BL/6J mice and cultured with neurobasal medium as previously described (Chishti et al., 2001; Dudal et al., 2004). In general, a single 14.5 d.p.c. mouse cortex preparation produced a yield of primary cortical neurons sufficient for approximately four samples of 2 × 106 cells. The HEK 293 human embryonic kidney cell line was cultured in DMEM/10%FBS (GIBCO) (Chow et al., 1999).

2.4. Transfection, immunoblotting, immunofluorescent microscopy, immunoprecipitation, and biotinylated membrane receptor pull-down assays C57BL/6J cortical neurons were transfected using Lipofectamine2000 (Invitrogen), while HEK 293 cells were transfected with GeneJuice (Novagen) according to the manufacturer’s protocol. The pcDNA PTPN21 plasmid was provided by Dr. Antonio Feliciello (Università di Napoli Federico II, Napoli, Italy) (Carlucci et al., 2008, 2010). The pcDNA PTPN21 myc plasmid was constructed by insertion of the full-length PTPN21 sequence in-frame into pcDNATM 4/myc vector. The kinase-dead plasmids were purchased from Addgene and the phosphatase-dead plasmids were constructed by site directed mutagenesis (Millipore) according to the manufacturer’s protocol (oligonucleotide sequences available upon request). HEK 293 cells were serum-starved for 24 h and then stimulated or treated with EGF/HB-EGF [50 ng/ml] for 10 min to 3 h, leupeptin [200 ␮g/ml] for an hour and MG132 [20 ␮M] for 30 min, and then lysed in buffer (50 mM Tris–HCl, 150 mM NaCl, 1% NP40, 0.5% sodium deoxycholate, 1% Triton 100, 50 mN NaF, 5 mM NaVo3 , 100 mM PMSF and Roche Complete Cocktail). Immunoprecipitation or biotinylated pull-down assays were conducted with 1–2 mg protein lysate with antibodies as indicated and proteinG-agarose (Roche) or streptavidin (Pierce). Immunoprecipitations, pull-downs and total cell lysates were separated by SDS-PAGE and then transferred to Hybond ECL nitrocellulose membranes (Amersham Biosciences). Visualization was by enhanced chemiluminescence (Amersham Biosciences). For immunofluorescent microscopy, C57BL/6J cortical neurons were fixed in 4% paraformaldehyde and then probed with anti-ErbB4 and anti-PTPN21 or anti-phospho-Elk and anti-NRG3 antibodies, and then nuclei stained with Hoechst (Life Technologies) as previously described (Carlucci et al., 2008). Antibodies used to detect the protein levels and/or immunoprecipitation and/or immunofluorescents were biotin (ab1227, abcam), ␤-actin (A5316, Sigma), ␤-casine (ab112595, abcam), c-myc (sc40, Santa Cruz), Cyclin D1 (sc753, Santa Cruz), phospho1056-ErbB4 (sc33040, Santa Cruz), ErbB4 (sc8050, Santa Cruz), ErbB4 (sc283, Santa Cruz), PTPN21 (ab12250, abcam), Elk-1 (Santa Cruz) and Phospho 383-Elk-1 (Abcam), Phospho-Tyrosine-100 (9411, Cell Signaling), PTPN21 immunoreactivity employed anti-PTPN21 and/or anti-myc antibodies. Slides were mounted and examined by fluorescence microscopy. Images were taken with 40× or 60× objectives using an LSM 710 (Carl Zeiss) or BX50WI (Olympus) microscope, respectively.

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

108 109 110 111 112 113 114

115

116 117 118 119 120 121 122

123 124 125

126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167

G Model

ARTICLE IN PRESS

BC 4555 1–10

J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx 168

169 170 171 172 173 174 175 176 177 178 179 180 181 182 183

184 185

186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209

210

2.5. mRNA extraction, quantitative RT-PCR and quantitative PCR

2.9. Statistical analyses

Messenger RNA (mRNA) was extracted using 1 ml Trizol reagent (Invitrogen) for approximately 1 × 107 cells, according to the manufacturer’s protocol. RNA levels were analyzed by real time reverse transcriptase-PCR using the FastStart Universal SYBR Green Master Mix (Roche) or semi-quantitative PCR analyses preformed using StepOneTM /StepOnePlusTM systems (ABI) or by ethidium bromide staining of 2% agarose gel. Primer sequences used to detect mRNA levels are: GAPDH sense primer 5 CCTCCCGCTTCGCTCTCT 3 and GAPDH antisense primer 5 CTGGCGACGCAAAAGAAGAT 3 ; PTPN21 sense primer 5 GCTACACGGTGTCCAGCAAGA 3 and PTPN21 antisense primer 5 TCTCCACGGACAGGGTGAAC 3 ; NRG3 sense primer 5 TCATCCACAAGGGCCAGTTC 3 and NRG3 antisense primer 5 GAAATCCCTGGCATTTGCA 3 ; ErbB4 sense primer 5 ACCGGGACCTCTCCTTCCT 3 and ErbB4 antisense primer 5 TAGTGGCTCTTAATCAGTTTCGTTACC 3 .

Statistical analyses were performed with Graphpad Prism (GraphPad Software Inc.) and Excel software (Microsoft). The data are illustrated as the mean ± s.e.m., a Student’s t-test was used to compare two groups. Differences in protein expression level variation among subgroups were analyzed with Image J (National Institutes of Health, USA). For all tests, differences were consider significant at p-value < 0.05 (*), p-value < 0.01 (**) and pvalue < 0.001 (***). Data are represented as the means of duplicate or triplicate independent experiments.

2.6. Dual-Elk-1 activity, Elk-1 binding assay and chromatin immunoprecipitation Dual-Elk-1 activity PathDetect assay (Stratagene) was performed according to Ng et al. (2010) using 4–5 × 104 cells. To confirm the binding site, four nucleotide substitutions were modified at sites of consensus sequence (5 -GGAC3 ) of Elk-1 binding motif using QuikChange Site-Directed Mutagenesis kit (Stratagene), primers were Sense primer 5 GCACCAAACCCCAGGGGCTTgagaTTTCAATGCATCCTCTGCTC-3 , Antisense primer 5 -GAGCAGAGGATGCATTGAAAtctcAAGCCCCTGGGGTTTGGTGC-3 and then subjected to sequencing for confirmation. Dual Elk-1-dependent luciferase and binding activity assays were performed according to the manufacturer’s protocol (Promega). Primers were c-fos sense primer 5 GCGAGCAGTTCCCGTCAAT 3 and c-fos antisense primer 5 GAAAGGCCGTGGAAACCTG 3 ; GAPDH sense primer 5 TACTAGCGGTTTTACGGGCG 3 and GAPDH antisense primer 5 TCGAACAGGAGGAGCAGAGAGCGA 3 ; NRG3 (+1878) promoter region sense primer 5 TCTCAATGTTATAGCTGAAAGAGGTG 3 and NRG3 (+1878) promoter region antisense primer 5 CACTGAAGAGACTGCAGAGAGC 3 ; NRG3 (+1833) promoter region sense primer 5 ATAATTTATTAACCTTGTCTGTGCAAA 3 and NRG3 (+1833) promoter region antisense primer 5 TGAAGTTTTCAAGGTATCCATTTTATT 3 . Chromatin-immunoprecipitation used 1 × 107 cell and 1.0–2.5 ␮g of antibodies as indicated and was preformed according to EZ-ChIPTM manufacturer’s protocol (Millipore). 2.7. Apoptosis and neuritic length analyses

218

After 72 h of C57BL/6J embryonic E14.5 extracted cortical neuron culture, embryonic cortical neurons were transfected with CFP or PTPN21 CFP then subjected to trophic factor-deprivation according to indicated period. Neuritic length was measured with NIH Image J software (NIH) and cells were immunostained with annexin V-FITC and propidium iodide (PI), an early apoptotic marker and a necrotic death marker, respectively. Numbers of annexin V- and/or PI-positive cells were counted using a BD LSR Fortessa Analyzer.

219

2.8. Phosphatase activity assay

211 212 213 214 215 216 217

Purified PTPN21 myc, with or without ErbB4 was dissolved in 40 ␮l of buffer (50 mM Tris, pH 6.8, 2 mM dithiothreitol) and incu221 bated at room temperature in a 96-well microtiter plate in 10 mM 222 pNPP (New England Biolab). The plates were read at absorbance 223 405 nm. The stop buffer (1 N NaOH) were added into each 224 225Q6 individual wells in a time interval manner as 0, 2, 5, 10, 20, 30, 40 min (Lorenz, 2011). 226 220

3

3. Results 3.1. Identification of PTPN21 as an NRG3 mediator candidate by eQTL analyses of brains of recombinant inbred mice The association between Ptpn21 and Nrg3 was originally identified by a robust open-access eQTL data set, available on the WebQTL database resource (www.genenetwork.org) with GeneNetwork accession number GN113. A putative associated (PA) region, which contributed to the mRNA expression variation of Nrg3 was found to be located on chromosome 12 near 99.89 Mb. Several genes along with Ptpn21 are within the identified PA region, which contributed to the mRNA expression variation of Nrg3 with a likelihood-ratio score (LRS) of 13.56 (Fig. 1A). Steps were taken to systematically prioritize the candidate genes. The genes where the SNPs were absent between two parenteral C57BL/6J and DBA/2J strains according to information from the WebQTL database were first eliminated, since the two parenteral strains exhibited enormous genetic variation hence a gene that lacks genetic variation cannot be responsible for the expression variation of Nrg3. The sequences of the remaining genes (Galc, Kcnk10 and Ptpn21) were downloaded from the Mouse Genome Informatics (MGI) website and the National Center for Biotechnology Information (NCBI). Within this candidate region, there is a SNP in the 3 UTR of both the Ptpn21 and the Galc genes (Fig. S1A). As the 3 UTR may be responsible for mRNA stability and the regulation of translation efficiency, Ptpn21 and Galc may act as regulators of Nrg3. To prioritize the two candidate genes, correlation analyses were performed and the expression of Ptpn21 was shown to be significantly linked with the expression of Nrg3 in neuronal cells (correlation 0.33, p-value 0.02), while Galc was not significantly correlated with Nrg3 in brain expression data (correlation 0.05, p-value 0.73). Therefore, Ptpn21 was selected for further molecular analyses. Supplementary Fig. S1 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biocel.2015.02.003. 3.2. Enhancement of PTPN21 levels in HEK293 and embryonic cortical neurons induces NRG3 expression The eQTL suggested that Ptpn21 mediates Nrg3 mRNA expression in mouse brain tissue; consequently we used molecular and cellular studies to determine whether PTPN21 expression variation would account for the differences in NRG3 levels. Quantitative reverse transcriptase-PCR (RT-qPCR) was carried out and an 8-fold increase in PTPN21 mRNA expression caused a time-dependent elevation in NRG3 mRNA, 48 h = 0.74 ± 0.57-fold, 72 h = 1.25 ± 0.20-fold, and 96 h = 2.80 ± 0.21-fold (Fig. S1B). To determine the long-term effects of increased PTPN21 levels, six stable PTPN21-overexpressing HEK 293 clones were established. As anticipated, PTPN21 expression levels were found to positively correlate with NRG3 levels in several clones (Fig. S1C). Furthermore, this positive correlation was found to be dose-dependent (Fig. S1D). In addition, overexpression of PTPN21 significantly

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

227

228 229 230 231 232 233 234 235 236

237

238 239

240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269

270 271

272 273 274 275 276 277 278 279 280 281 282 283 284 285

G Model BC 4555 1–10 4

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Fig. 1. Ptpn21 mediates Nrg3 expression in murine neuronal tissues. (A) BXD recombinant inbred strains based on the genetic variation within the putative association region of chromosome 12 near 99.89 Mb causes mRNA expression elevation of the Nrg3 gene. Association identified by eQTL Nrg3 linkage mapping confirmed that Ptpn21 positively modulates Nrg3 mRNA levels. Genome-wide scan analyses identified the main loci (bars) and LRS values across the considered location (dashed line). (B) PTPN21 overexpression modulates NRG3 protein expression levels in embryonic C57BL/6J cortical neurons in contrast to neurons transiently transfected with vector (pcDNA). Samples were treated with EGF for 1–2 h before cell lysate collection.

286 287 288 289 290 291 292

293 294

295 296 297

elevated NRG3 expression by 5.39 ± 0.51-fold in embryonic C57BL/6J cortical neurons (Fig. 1B). In contrast, when PTPN21 expression was knocked down (by 0.49-fold, ±0.04, p < 0.005) with small interfering RNA (siPTPN21), NRG3 expression was also mildly reduced to 0.54-fold (±0.10, p < 0.03) (Fig. S1E). These results demonstrate that PTPN21 is significantly and positively correlated with NRG3 expression levels. 3.3. PTPN21 up-regulates ErbB4 expression without triggering ErbB4 transcription To identify the molecular pathways trigged by PTPN21, we first monitored the transcriptional and translational rates of a known NRG3 binding RTK receptor, ErbB4 and found PTPN21 does not

affect the ErbB4 mRNA levels (Fig. S2A). However, PTPN21 was positively correlated with total ErbB4 protein levels (Fig. S2B). Moreover, PTPN21 translocated from the nucleus/cytosol to the plasma membrane fraction 30 min after stimulation with known ErbB4 ligands, EGF/HB-EGF (Fig. S2C). These results suggest a possible interaction between PTPN21 and ErbB4 after induction with EGF/HB-EGF. To confirm whether PTPN21 interacts with ErbB4, we used an immunoprecipitation assay and found interactions between endogenous PTPN21 and ErbB4 after 30 min of EGF stimulation (Fig. 2A). Furthermore we detected PTPN21 (using anti-PTPN21 and anti-Myc antibodies) from ErbB4 plasma membrane fractions of ErbB4 immunoprecipitation complexes after 30 min stimulation with EGF in HEK 293 cells (data not shown). We showed ErbB4 recruited PTPN21 to the plasma membrane

Fig. 2. PTPN21 dephosphorylates ErbB4 at Tyr1162 at cell membrane. All samples treated with EGF for the time 30–90 min before following assay were carried out. (A) The endogenous PTPN21 and ErbB4 shown interaction in HEK 293 cell line after EGF simulation for 30 min using anti-ErbB4(n) and anti-ErbB4(c) antibodies for immunoprecipitation assay. (B) Pull-down (PD) assay using anti-biotin antibody confirmed that PTPN21 and ErbB4 form an immunoprecipitated complex at the cell membrane after treated with 200 ␮g/ml of leupeptin for an hour and 20 ␮M of MG132 for 30 min. (C) Full length PTPN21 Myc or Phosphatase domain delation Myc mutant (PD; 1–667 amino acid) or partial FERM deletion Myc mutant (ED; 1–325 amino acid) are co-introduced by transfection with full length ErbB4. PTPN21 with FERM domain successfully forming a complex with ErbB4 revealed by IP. (D) The tyrosine mutation was induced using site directed mutagenesis and sample are collected after EGF induction for the time indicated in the figure (C = pcDNA, T = PTPN21, D = phosphatase dead PTPN21). The total phospho-ErbB4 level reveals by anti-phospho-Tyr 100 antibody, PTPN21-dependent dephosphorylation of ErbB4 is abolished by 1162 Tyrosine-dead ErbB4. ErbB4(n) represent anti-ErbB4 antibody at an epitope between amino acids 1280–1308 and ErbB4(c) represent anti-ErbB4 at the C-terminal.

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

298 299 300 301 302 303 304 305 306 307 308 309 310 311

G Model BC 4555 1–10

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

324

after 30 min of EGF stimulation, which plateaued off at 60 min (Fig. 2B; input and S2D IP: ErbB4 (c)). Furthermore, a biotinylated pull-down (PD) subcellular fractionation assay tracked the biotinylated-membrane bound ErbB4 movement back to the membrane after 30–60 min of EGF induction (Fig. 2B; PD: biotin). This binding association between ErbB4 and PTPN21 is novel and suggests that PTPN21 might inhibit ErbB4 degradation. Taken together, these data place PTPN21 not only in close proximity to ErbB4, but also show that PTPN21 forms a complex with the cytosolic face of ErbB4 at the membrane, promotes ErbB4 accumulation and its recycling back to the plasma membrane. Supplementary Fig. S2 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biocel.2015.02.003.

325

3.4. Characterizing the PTPN21–ErbB4 interaction

312 313 314 315 316 317 318 319 320 321 322 323

326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361

362 363

364 365 366 367 368 369 370 371 372 373

We have demonstrated for the first time that full length PTPN21 binds to ErbB4. We then wanted to analyze which binding domain of PTPN21 is required for ErbB4 binding. PTPN21 protein tyrosine phosphatase is composed of an ezrin-band 4.1-radixin-merlin (FERM) motif at its N-terminus and a catalytic phosphatase domain at its C-terminus separated by a histone promoter control 2 domain. Deletion of the phosphatase-domain and partial deletion of the FERM-domain illustrated that an intact FERM domain is necessary for ErbB4 interaction in immunoprecipitation assays (Fig. 2C). We then investigated which tyrosines on ErbB4 that PTPN21 could dephosphorylate, since PTPN21 is a phosphatase, by systematically mutating each tyrosine (Y) residue on ErbB4 to phenylalanine (F) using site-mutagenesis. If the tyrosine motif is essential, then the mutation should abolish the PTPN21-dependent regulation of total phospho-ErbB4 levels. In this study, the Y1162F mutation of ErbB4 stopped PTPN21 from dephosphorylating ErbB4 (reducing ErbB4 pY levels to that of control as well as the phosphatase-dead PTPN21 mutant) after 30–45 min of EGF induction (Fig. 2D). A PTPN21-dependent enhancement of the total ErbB4 levels, but not of phospho-ErbB4, was detected in samples without EGF stimulation (Fig. 2D), which was consistent with previous findings (Fig. S2B). Suggesting Y1162 site on ErbB4 is required for PTPN21-mediated dephosphorylation for ErbB4. To determine whether phsopho-Y1162 acts as a scaffold for PTPN21, mutant Y1162A, Y1162F and wild type ErbB4 were co-transfected with PTPN21, revealed Y1162A and Y1162F ErbB4 mutants fail to bind with the PTPN21 (Fig. S2E). To determine whether PTPN21 is a direct substrate for ErbB4, in vitro p-nitrophenyl phosphate dephosphorylation (pNPP) dephosphorylation assay were performed and revealed PTPN21 dephosphorylation is significantly higher in the present of ErbB4 in a time-dependent manner (Fig. S3). Taking together, following ErbB4 activation, PTPN21 promotes ErbB4 accumulation dependent on Erbb4 Y1162 site. One mechanism for this accumulation could be the inhibition of ErbB4 degradation. Supplementary Fig. S3 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biocel.2015.02.003. 3.5. PTPN21 prevents ErbB4 from undergoing ubiquitination through promoting ErbB4 dephosphorylation Based on our results we hypothesized that PTPN21 would reduce phospho-ErbB4 levels and in turn inhibit ErbB4 degradation. Immunoprecipitation (IP) with anti-ErbB4 using an N-terminal antibody was employed to determine the dynamics of ErbB4 ubiquitination and phosphorylation. Normalized against total ErbB4, the magnitude of ubiquitin-conjugation and tyrosine phosphorylation (p-Tyr) of ErbB4 was shown to have an inverse correlation with the amount of functional PTPN21 (Fig. 3). Importantly, a phosphatase-dead mutant of PTPN21 failed to significantly change the levels of ubiquitin-conjugated ErbB4 and p-Tyr ErbB4 (Fig. 3).

5

Fig. 3. PTPN21 binds to ErbB4 and controls its phospho-ErbB4 and ubiquitinconjugated ErbB4 levels. PTPN21 reduces phospho-ErbB4 and ubiquitin-conjugated ErbB4 levels as shown by immunoprecipitation with anti-ErbB4 at the N-terminal followed by immunoblotting. HEK 293 cells were co-transfected with functional ErbB4 and PTPN21. All samples were treated with 200 ␮g/ml of leupeptin for an hour and 20 ␮M of MG132 for 30 min, samples were then treated with serum free medium (–) or EGF (50 ng/ml) for 10–15 min. The effect of PTPN21 on reducing phospho-ErbB4 and ubiquitin-conjugated ErbB4 (UbHA ) levels is consistently shown in a concentration-dependent manner. In contrast, samples co-transfected with phosphatase-dead PTPN21 (PD) failed to influence phospho-ErbB4 and ubiquitinconjugated ErbB4 levels.

This novel finding indicates that PTPN21 promotes the stability of ErbB4 and dephosphorylates ErbB4. 3.6. Kinase-dead ErbB4 mutant or phosphatase-dead PTPN21 fail to confer ErbB4 accumulation To provide further evidence that PTPN21 acts as a phosphatase in the ErbB4 pathway, we investigated whether a functional ErbB4 is needed for PTPN21 to promote ErbB4 accumulation. To eliminate any secondary feedback of EGF induction, a shorter EGF stimulation period of 10–15 min was used, which sowed the kinase-dead mutant of ErbB4 (K751R) (Sundvall et al., 2007) could reverse PTPN21-induced ErbB4 accumulation (Fig. 4A). Furthermore, a phosphatase-dead mutant of PTPN21 (C1108S) (Carlucci et al., 2008) inhibited PTPN21-induced ErbB4 accumulation (Fig. 4B). These data support the notion that activation of ErbB4 is a prerequisite for PTPN21 to take part in the ErbB4 signaling cascade, resulting in the stabilization of ErbB4 and indirectly promoting ErbB4 pathway activation. 3.7. PTPN21 causes MEK dependent phospho-Elk-1 elevation in C57BL/6J embryonic cortical neurons We then determined whether PTPN21 regulates downstream ErbB4 signaling through MEK/Elk-1, by investigating Elk-1 activity. PTPN21 positively mediates Elk-1 signaling in the HEK 293 cell line upon EGF induction in a time-dependent manner (Fig. S4A). In

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

374 375

376 377

378 379 380 381 382 383 384 385 386 387 388 389 390

391 392

393 394 395 396

G Model BC 4555 1–10 6

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Fig. 4. The kinase activity of ErbB4 is critical for PTPN21 mediated ErbB4 accumulation and resulted in elevation of phospho-ErbB4 due to increase of total ErbB4 levels. All samples were transfected for 36–48 h and were serum starved overnight, then treated with ErbBs agonist EGF (E)/ErbB4 agonist HB-EGF (H) (50 ng/ml) for 10–15 min, or for an hour with ErbBs inhibitor AG1478 (A) (5–10 ␮M). (A) HEK 293 cells were co-transfected with functional ErbB4 and PTPN21. The promoting effect of PTPN21 on total ErbB4 levels and consequently the phospho-ErbB4 (p-ErbB4) level was consistently shown in comparison to samples co-transfected with backbone vector and ErbB4. (B) The presence of kinase-dead (KD) ErbB4 abolishes the PTPN21-dependent ErbB4 accumulation. While, phosphatase-dead PTPN21 significantly reduces ErbB4 levels.

Fig. 5. PTPN21 regulates phospho-Elk-1 levels and NRG3 promoter binding activity. All samples were transfected for 36–48 h and were serum starved overnight, then treated with EGF (E) for 10–15 min. (A) Chromatin-immunoprecipitation (Ch-IP) shows significant binding activity enhancement in samples transfected with PTPN21 and treated with EGF, while U0126 halts NRG3 promoter region binding activity by Elk-1 at the baseline level. (B) Identification of the Elk-1 binding site within the NRG3 promoter region. NRG3 promoter binding activity assay illustrated enhanced binding levels in samples co-transfected with PTPN21 and the wild type NRG3 promoter region, while samples co-transfected with the Sub1919 mutant abolished Elk-1 binding.

404

addition, PTPN21 induces Elk-1 signaling, while the MEK inhibitor U0126 abrogates PTPN21-induced Elk-1 signaling in embryonic C57BL/6J cortical neurons (Fig. S4B). This indicates that PTPN21 acts through MEK to activated Elk-1, which corroborates previous reports using bladder carcinoma cells and HEK 293 cells (Carlucci et al., 2010). Supplementary Fig. S4 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biocel.2015.02.003.

405

3.8. Elk-1 is crucial for transcription of the NRG3 gene

397 398 399 400 401 402 403

406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422

We next wanted to determine how PTPN21 regulates the expression of NRG3. We have shown that PTPN21 can confer activation of both Elk-1 and NRG3 in cortical neurons. These results were strengthened by the identification of a consensus Elk-1 (5 -GGAC-3 ) sequence at +1919 as an Elk-1 binding site within the NRG3 promoter region using TFsearch software. Using chromatin immunoprecipitation assays of the nuclear fractions of HEK293 cells with anti-Elk-1 antibodies, we successfully precipitated the NRG3 promoter region, monitored by NRG3 promoter binding activity assay using specific c-fos primers as a positive control and two sets of primers that can detect the binding activity of the consensus Elk-1 binding sites within NRG3. An increase in Elk-1 binding activity was detected in both sets of binding activity primers, set one = 30.59 ± 5.15% and set two = 60.89 ± 2.66% in the HEK293 cell line, in contrast to the untreated samples where binding activity remained at baseline levels (Fig. 5A). To confirm the binding motif of Elk-1 within the NRG3 promoter region we mutated the predicted

Elk-1 consensus-binding motif from 5 -GGAC-3 to 5 -TCTC-3 . This mutated site reduced Elk-1-dependent binding activity by 3.35-fold (±0.16) (Fig. 5B). Consistent with the above findings, transient co-expression of Elk-1and PTPN21, increased Elk-1 binding to the NRG3 promoter by 6.49-fold (±0.88) after EGF stimulation (10 min); and a 4.27-fold (±0.35) increase without EGF treatment. Importantly, the kinase-dead Elk-1 mutant (mutS383A) showed no elevated binding activity (Fig. S5A). Immunoblotting confirmed the importance of functional Elk-1 (using wild type Elk-1 and dominant-active Elk-1), which resulted in elevated NRG3 protein levels after 30 min of EGF treatment (Fig. 6A). Furthermore, mutS383A, or samples treated with the MEK inhibitor U0126 failed to show elevation of NRG3 (Fig. 6A). Immunofluorescent microscopy provided further confirmation that both PTPN21 and Elk-1 promote NRG3 expression in C57BL/6J embryonic cortical neurons upon EGF treatment (at 1–2 h) in contrast to samples transfected with an empty plasmid (Figs. 1B, 6B and Fig. S4). Supplementary Fig. S5 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biocel.2015.02.003. 3.9. PTPN21 positively influences neuronal survival in embryonic cortical neurons under hypo-neurotrophic factor stress As Elk-1 and NRG3 promote neuron and oligodendrocyte Q7 survival, respectively [1,22], we speculated that PTPN21 might also be required for neuronal survival. To test this hypothesis, wild-type PTPN21 and phosphatase-dead PTPN21

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442

443 444

445 446 447 448

G Model BC 4555 1–10

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

7

Fig. 6. Elk-1 and PTPN21 regulate NRG3 expression levels in HEK 293 cells and murine cortical neurons. (A) In HEK 293 cells wild type (WT) Elk-1 or dominant active Elk-1 (mutDA) induced alterations in the levels of NRG3 and phospho-Elk-1, in samples treated with EGF in a time-dependent manner. Conversely, the Elk-1 catalytic dead (mut383) mutant or samples co-treated with U0126 did not show up-regulated levels of NRG3 and phospho-Elk-1. (B) Immunofluorescent microscopy reveals p-Elk-1 and NRG3 expression levels change in embryonic C57BL/6J cortical neurons transfected with PTPN21 and EGF stimulation. (a and d) vector (PV), (b and e) PTPN21 and (c and f) Elk-1, (d–f) stimulated with EGF and (a–c) non-stimulated samples. Pictures were taken at 480× magnification and scale bar represents 10 ␮m.

449 450 451 452 453 454 455 456 457 458 459 460 461

(G3323C) were expressed in cortical neurons and the cells subjected to trophic factor deprivation. Annexin V and propidium iodide were used to assess the pro-neuronal survival role of PTPN21. Time-dependent neuronal apoptosis assays illustrated the trend of PTPN21 positively enhanced neuronal survival under trophic factor B27 withdrawal, while phosphatase dead PTPN21 mutant (DN PTPTN21) failed to reduce neuronal apoptosis (Fig. S5B). Furthermore, the phosphatase-dead PTPN21 (DN PTPN21) mutant failed to reduce neuronal apoptosis after trophic factor withdrawal (Fig. 7A). Additionally, PTPN21 influenced neuritic length, in both normal medium and trophic factor withdrawal medium. PTPN21 transfected cortical neurons displayed on average, longer neuritic extensions

in both media, in contrast to the vehicle transfected cortical neurons (Fig. 7B). Collectively, these data show that PTPN21 promotes neuritic survival and neuritic length under trophic factor withdrawal conditions in 14.5 d.p.c. C57BL/6J embryonic mouse cortical neurons. 4. Discussion PTPN21 protein tyrosine phosphatase is composed of a FERM motif, a histone promoter control 2 domain and a catalytic phosphatase domain. Despite the well-defined PTPN21 domains, the function of PTPN21 remains largely unknown. To date, bioinformatics methods, such as eQTL analyses have led to the identification

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

462 463 464 465 466

467

468 469 470 471 472

G Model BC 4555 1–10 8

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Fig. 7. PTPN21 positively promotes cortical neuron survival and neuritic elongation in murine cortical neurons under trophic factor deprivation. (A) (a and b) After 90 h of B27 trophic factor withdrawal, light field pictures captured 14.5 d.p.c. mouse cortical neurons transfected with PTPN21 and pCFP promotes healthy cortical neurons (black arrow), while deformed cortical neurons (white arrow) are found in the cortical neurons transfected with vehicle only. (c) 5000–10,000 cells were counted, the abscissa axis is the time line and the ordinate axis cells in the Q3 region (Healthy cell) in terms of Annex V and propidium iodide signal. Cortical neurons were transfected with PTPN21, pcDNA or catalytic dead PTPN21 (DN). (B) (a–f) Pictures taken of embryonic cortical neurons after 0–120 h of trophic factor deprivation. Samples were co-transfected with, (a–c) pCFP (Vehicle) with or without PTPN21 DN; (e–g) PTPN21 and Vehicle. (i–k) The neuritic length was measured using Image J software at the 0 h, 72 h and 120 h time points. Neuritic length was significantly longer in PTPN21 transfected cells (black arrow) compared to pCFP transfected cells (white arrow). Scale bar represents 10 ␮m.

473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498

of promising trans-regulator signal associations in specific tissues. Based on an eQTL analyses using murine neuronal tissues, we found that a genetic variation of Ptpn21 was associated with mRNA expression level variation of Nrg3. The genetic variation identified within Ptpn21 is located in its 3 UTR (Fig. S1A); a region attributed to translational efficiency and mRNA stability, leading us to speculate that Ptpn21 expression variation alters Nrg3 mRNA expression levels. Our results provide deeper insight into the mechanism of PTPN21 regulation of Nrg3. Based on the reported link between the MEK/ELK-1 signaling and PTPN21, we used the MEK inhibitor U0126 to investigate this link in mouse embryonic cortical neurons (Fig. S3). We also investigated which receptor tyrosine kinase (RTK) was involved in PTPN21 signalling, and we observed that enhanced expression of PTPN21 robustly elevated ErbB4/Elk-1signal transduction and stabilized ErbB4 levels, in a manner that was independent of any role in ErbB4 transcription regulation (Fig. S2A). Our novel findings are consistent with previous studies, showing that PTPN21 stabilizes ErbB1 receptor levels (Carlucci et al., 2010). However, as PTPN21 is a phosphatase and not a kinase, downstream signal enhancement appears contradictory. Consequently we hypothesized that PTPN21 might be exerting its effect by reducing ErbB4 receptor ubiquitination, since non-phosphorylated ErbB receptors are known to have a slower rate of ubiquitination (Galcheva-Gargova et al., 1995; Grøvdal et al., 2004; Thien

and Langdon, 2001; Wang et al., 1996). In our ubiquitination studies, PTPN21 reduced ErbB4 phosphorylation in a dose-dependent manner and ubiquitin-conjugated ErbB4 was detected after treatment with the lysosome and proteasome inhibitors leupeptin and MG132, respectively (Fig. 3). In addition, subcellular fractionation pull-down and immunoprecipitation assays confirmed that ErbB4 bound to PTPN21 at the plasma membrane (Fig. 2A and S2D) after 30–90 min of EGF stimulation. Furthermore, the subcellular fractionation biotinylated pull-down assay also showed that PTPN21, along with biotinylated-membrane bound ErbB4 recycled back to the plasma membrane. This suggests that PTPN21 bound to both ErbB4 and a recycling complex after EGF stimulation. Further, the FERM domain of PTPN21 is essential for ErbB4 binding and the Y1162 motif of ErbB4 is required for PTPN21 to dephosphorylate ErbB4 (Fig. 2B and C). These findings, in conjunction with those by Carlucci et al. (2010) who identified associations of PTPN12 with ErbB1, another family member of ErbB4, in colon carcinoma cell model, strongly implicate PTPN21 in the molecular mechanisms of the ErbB signaling cascade. In order to show the PTPN21-dependent ErbB4 signal transduction after ligand stimulation, we stimulated HEK 293 cells expressing PTPN21 with known ErbB4 ligands, EGF or HB-EGF, which resulted in a higher magnitude of ErbB4 signaling (Figs. 3 and 4). In contrast, the kinase-dead ErbB4 and phosphatase-dead PTPN21 mutants prevented elevation of ErbB4 levels and its activity (Fig. 4B). This supports previous findings in which loss of

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524

G Model BC 4555 1–10

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589

ubiquitin-conjugated ErbB receptors resulting from loss of E3 ligase Cbl is to known to indirectly enhance ErbB receptor signaling (Lai et al., 2012). Consequently, PTPN21 is able to induce ErbB4 accumulation and indirectly promotes Elk-1 phosphorylation in both HEK 293 and embryonic cortical neurons (Fig. S4A and B, respectively), supporting previous reports on PTPN21/ErbB1/Elk-1 signaling (Carlucci et al., 2008). It is important to note that Sacco and colleagues demonstrated that PTPN21 interacts with Grb2, a downstream target of ErbB4 and an upstream target of Elk-1 signaling (Sacco et al., 2014). Linking PTPN21 to both upstream and downstream elements, ErbB4 and Grb2, respectively, results in regulation of the Elk-1 signaling cascade. Our studies comprised extensive mechanical characterization of this signaling pathway and demonstrated that although PTPN21 enhances phospo-Elk-1 shortly after EGF induction (50 ng/ml; 5–10 min), there are no significant effects on prolonged EGF stimulation (50 ng/ml; >10–15 min). Our investigation reveals a moderate dephosphorylation capability of PTPN21 toward ErbB4. Through pre-treatment with leupeptin and MG132; the effects of PTPN21 on ErbB4 accumulation through reduction of ErbB4 degradation were also assessed (Figs. 2A and 3). In order to evaluate the biological significance of PTPN21 activity, we relied on previous findings regarding the function of Elk-1, such as regulation of c-Fos (Dalton and Treisman, 1992; Deng, 1994), and promotion of neuronal survival (Demir et al., 2011; Repici et al., 2009). Given our novel findings that NRG3 translation is dependent on Elk-1 activation (Figs. 5 and 6), and NRG3s role in promoting oligodendrocyte survival (Carteron et al., 2006), we hypothesize that, similar to both Elk-1 and NRG3, PTPN21 possesses neuronal survival properties. Indeed, our results show that PTPN21 enhances embryonic e14.5 C57BL/6J cortical neuronal survival under hyponeurotrophic stress (Fig. 7A). Another intriguing biological function of PTPN21 derived from the present study is that PTPN21 increases neuritic length (Fig. 7B), which is vital for maintaining normal neuronal function. This is in line with previous reports that dominant negative Elk-1 halts dendritic elongation (Sharma et al., 2010). Intriguingly, dysfunctional neurite elongation regulators, such as GABAergic circuits, are often associated with the pathophysiology of Schizophrenia (Benes and Berretta, 2001; Lewis et al., 2005). Morphological studies also revealed that abnormal neurite elongation during neurogenesis correlated with depression (Hashimoto et al., 2006), a common symptom in people with Schizophrenia. Combining GWAS and biological functional studies suggests a potentially important role for PTPN21 in schizophrenia, although its precise involvement in the pathological development of this disease remains to be elucidated. Importantly, the present study is the first to report the association between PTPN21 and NRG3. PTPN21 promotes neuronal survival and neuritic elongation, and it consistently positively correlates with NRG3 expression in embryonic cortical neurons. However, other variables in the pathway, such as MEK/Elk-1, do serve as multiple conditional limiting factors. Consequently, we do not anticipate that PTPN21 initiates a positive feedback loop. Additionally, the de-phosphorylation activity of PTPN21 remains to be elucidated. Our findings have shown a direct link between PTPN21 and two other known Schizophrenia risk genes, ErbB4 and NRG3 that are also supported by GWAS meta-analyses, which identified PTPN21 as being associated with schizophrenia (Chen et al., 2011). Since the pathophysiology of schizophrenia has been linked to neurotrophic factors (Durany and Thome, 2004; Insel, 2005), the underlying molecular relationships between schizophrenia risk genes must be further investigated to fully understand the pathways controlling this disease, and help develop an effective treatment for those who suffer from schizophrenia.

9

Acknowledgements The PTPN21 vector was kindly provided by Prof. Antonio Feliciello, Dipartimento di Neuroscienze, Università “Federico II,” Italy and the ErbB4 vector was kindly provided by Dr. Paul Rammer, Danish Cancer Society Research Center, Denmark. We thank Prof. Sookja K. Chung, Prof. Xin-yuan Guan, Dr. Abel Chun, Dr. Raven Kok, Dr. Graham Shea, Dr. Emily H.M. Wong, Dr. Stacey S. Cherny, Michael Su Wang and Jian Gina for their helpful suggestions and/or technical support. This work was funded by grants from NSFC Q8 grant (No. 81271226 to YQS) and the Research Grants Council of Hong Kong (No. 17119114 to YQS). A/Prof. Evan Ingley is supported by Sock-it-to-Sarcoma and Hollywood Private Hospital Medical Research Foundation.

References Benes FM, Berretta S. GABAergic interneurons: implications for understanding schizophrenia and bipolar disorder. Neuropsychopharmacology 2001;25:1–27. Cardone L, Carlucci A, Affaitati A, Livigni A, DeCristofaro T, Garbi C, et al. Mitochondrial AKAP121 binds and targets protein tyrosine phosphatase D1, a novel positive regulator of src signaling. Mol Cell Biol 2004;24:4613–26. Carlucci A, Gedressi C, Lignitto L, Nezi L, Villa-Moruzzi E, Avvedimento EV, et al. Protein-tyrosine phosphatase PTPD1 regulates focal adhesion kinase autophosphorylation and cell migration. J Biol Chem 2008;283:10919–29. Carlucci A, Porpora M, Garbi C, Galgani M, Santoriello M, Mascolo M, et al. PTPD1 supports receptor stability and mitogenic signaling in bladder cancer cells. J Biol Chem 2010;285:39260–70. Carteron C, Ferrer-Montiel A, Cabedo H. Characterization of a neural-specific splicing form of the human neuregulin 3 gene involved in oligodendrocyte survival. J Cell Sci 2006;119:898–909. Chen J, Lee G, Fanous AH, Zhao Z, Jia P, O’Neill A, et al. Two non-synonymous markers in PTPN21, identified by genome-wide association study data-mining and replication, are associated with schizophrenia. Schizophr Res 2011;131:43–51. Chen PL, Avramopoulos D, Lasseter VK, McGrath JA, Fallin MD, Liang KY, et al. Fine mapping on chromosome 10q22-q23 implicates Neuregulin 3 in schizophrenia. Am J Hum Genet 2009;84:21–34. Chishti MA, Yang DS, Janus C, Phinney AL, Horne P, Pearson J, et al. Early-onset amyloid deposition and cognitive deficits in transgenic mice expressing a double mutant form of amyloid precursor protein. J Biol Chem 2001;276:21562–70. Chow JC, Young DW, Golenbock DT, Christ WJ, Gusovsky F. Toll-like receptor-4 mediates lipopolysaccharide-induced signal transduction. J Biol Chem 1999;274:10689–92. Dalton S, Treisman R. Characterization of SAP-1, a protein recruited by serum response factor to the c-fos serum response element. Cell 1992;68:597–612. Demir O, Aysit N, Onder Z, Turkel N, Ozturk G, Sharrocks AD, et al. ETS-domain transcription factor Elk-1 mediates neuronal survival: SMN as a potential target. Biochim Biophys Acta 2011;1812:652–62. Deng T, Karin M. c-Fos transcriptional activity stimulated by H-Ras-activated protein kinase distinct from JNK and ERK. Nature 1994;371:171–5. Dorner C, Ciossek T, Müller S, Møller PH, Ullrich A, Lammers R. Characterization of KIF1C, a new kinesin-like protein involved in vesicle transport from the Golgi apparatus to the endoplasmic reticulum. J Biol Chem 1998;273:20267–75. Dudal S, Krzywkowski P, Paquette J, Morissette C, Lacombe D, Tremblay P, et al. Inflammation occurs early during the A␤ deposition process in TgCRND8 mice. Neurobiol Aging 2004;25:861–71. Durany N, Thome J. Neurotrophic factors and the pathophysiology of schizophrenic psychoses. Eur Psychiatry 2004;19:326–37. Galcheva-Gargova Z, Theroux SJ, Davis RJ. The epidermal growth factor receptor is covalently linked to ubiquitin. Oncogene 1995;11:2649–55. Grøvdal LM, Stang E, Sorkin A, Madshus IH. Direct interaction of Cbl with pTyr 1045 of the EGF receptor (EGFR) is required to sort the EGFR to lysosomes for degradation. Exp Cell Res 2004;300:388–95. Hashimoto R, Numakawa T, Ohnishi T, Kumamaru E, Yagasaki Y, Ishimoto T, et al. Impact of DISC1 Ser704Cys polymorphism on risk for major depression, brain morphology and ERK signaling. Hum Mol Genet 2006;15:3024–33. Insel TR. Rethinking schizophrenia. Nature 2005;468:187–93. Jui HY, Tseng RJ, Wen X, Fang HI, Huang LM, Chen KY, et al. Protein-tyrosine phosphatase D1, a potential regulator and effector for Tec family kinases. J Biol Chem 2000;29:41124–32. Lai AZ, Durrant M, Zuo D, Ratcliffe CD, Park M. Met kinase-dependent loss of the E3 ligase Cbl in gastric cancer. J Biol Chem 2012;287:8048–59. Lavaur J, Bernard F, Trifilieff P, Pascoli V, Kappes V, Pages C, et al. A TAT-DEF-Elk-1 peptide regulates the cytonuclear trafficking of Elk-1 and controls cytoskeleton dynamics. J Neurosci 2007;27:14448–58. Lewis DA, Hashimoto T, Volk DW. Cortical inhibitory neurons and schizophrenia. Nat Rev Neurosci 2005;6:312–24. Livigni A, Scorziello A, Agnese S, Adornetto A, Carlucci A, Garbi C, et al. Mitochondrial AKAP121 links cAMP and src signaling to oxidative metabolism. Mol Biol Cell 2006;17:263–71.

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

590

591 592 593 594 595 596 597 598 599 600 601 602

603

604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666

G Model BC 4555 1–10 10 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685

ARTICLE IN PRESS J.H.C. Plani-Lam et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Meier S, Strohmaier J, Breuer R, Mattheisen M, Degenhardt F, Mühleisen TW, et al. Neuregulin 3 is associated with attention deficits in schizophrenia and bipolar disorder. Int J Neuropharmacol 2013;16:549–56. Michaelson JJ, Loguercio S, Beyer A. Detection and interpretation of expression quantitative trait loci (eQTL). Methods 2009;48:265–76. Moller NPH, Moller KB, Lammers R, Kharitonenkov A, Sures I, Ullrich A. Src kinase associates with a member of a distinct subfamily of protein-tyrosine phosphatases containing an ezrin-like domain. Proc Nat Acad Sci 1994;91:7477–81. Morar B, Dragovic´ M, Waters FA, Chandler D, Kalaydjieva L, Jablensky A. Neuregulin 3 (NRG3) as a susceptibility gene in a schizophrenia subtype with florid delusions and relatively spared cognition. Mol Psychiatry 2011;16:860–6. Ng MH, Ng RK, Kong CT, Jin DY, Chan LC. Activation of Ras-dependent Elk-1 activity by MLL-AF4 family fusion oncoproteins. Exp Hematol 2010;38:481–8. Peirce JL, Lu L, Gu J, Silver LM, Williams RW. A new set of BXD recombinant inbred lines from advanced intercross populations in mice. BMC Genet 2004;5:7. Repici M, Mare L, Colombo A, Ploia C, Sclip A, Bonny C, et al. c-Jun N-terminal kinase binding domain-dependent phosphorylation of mitogen-activated protein kinase kinase 4 and mitogen-activated protein kinase kinase 7 and balancing cross-talk between c-Jun N-terminal kinase and extracellular signal-regulated kinase pathways in cortical neurons. Neuroscience 2009;159:94–103.

Sacco F, Boldt K, Calderone A, Panni S, Paoluzi S, Castagnoli L, et al. Combining affinity proteomics and network context to identify new phosphatase substrates and adapters in growth pathways. Front Genet 2014:115. Sharma A, Callahan LM, Sul JY, Kim TK, Barrett L, Kim M, et al. A neurotoxic phosphoform of Elk-1 associates with inclusions from multiple neurodegenerative diseases. PLoS ONE 2010;5:e9002. Sundvall M, Peri L, Määttä JA, Tvorogov D, Paatero I, Savisalo M, et al. Differential nuclear localization and kinase activity of alternative ErbB4 intracellular domains. Oncogene 2007;26:6905–14. Thien CB, Langdon WY. Cbl: many adaptations to regulate protein tyrosine kinases. Nat Rev Mol Cell Biol 2001;2:294–307. Lorenz U. Protein tyrosine phosphatase assays. Curr Protoc Immunol 2011. Q9 Wang Y, Yeung YG, Langdon WY, Stanley ER. c-Cbl is transiently tyrosine-phosphorylated, ubiquitinated, and membrane-targeted following CSF-1 stimulation of macrophages. J Biol Chem 1996;271:17–20. Wang YC, Chen JY, Chen ML, Chen CH, Lai IC, Chen TT, et al. Neuregulin 3 genetic variations and susceptibility to schizophrenia in a Chinese population. Biol Psychiatry 2008;64:1093–6.

Please cite this article in press as: Plani-Lam JHC, et al. PTPN21 exerts pro-neuronal survival and neuritic elongation via ErbB4/NRG3 signaling. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.02.003

686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703