Sprouty1 regulates neuritogenesis and survival of cortical neurons

Abstract

In multicellular organisms, receptor tyrosine kinases (RTKs) control a variety of cellular processes, including cell proliferation, differentiation, migration, and survival. Sprouty (SPRY) proteins represent an important class of ligand‐inducible inhibitors of RTK‐dependent signaling pathways. Here, we investigated the role of SPRY1 in cells of the central nervous system (CNS). Expression of SPRY1 was substantially higher in neural stem cells than in cortical neurons and was increased during neuronal differentiation of cortical neurons. We found that SPRY1 was a direct target gene of the CNS‐specific microRNA, miR‐124 and miR‐132. In primary cultures of cortical neurons, the neurotrophic factors brain‐derived neurotrophic factor (BDNF) and Basic fibroblast growth factor (FGF2) downregulated SPRY1 expression to positively regulate their own functions. In immature cortical neurons and mouse N₂A cells, we found that overexpression of SPRY1 inhibited neurite development, whereas knockdown of SPRY1 expression promoted neurite development. In mature neurons, overexpression of SPRY1 inhibited the prosurvival effects of both BDNF and FGF2 on glutamate‐mediated neuronal cell death. SPRY1 was also upregulated upon glutamate treatment in mature neurons and partially contributed to the cytotoxic effect of glutamate. Together, our results indicate that SPRY1 contributes to the regulation of CNS functions by influencing both neuronal differentiation under normal physiological processes and neuronal survival under pathological conditions.

We found sprouty1 (Spry1) was a direct target gene of the central nervous system (CNS)‐specific microRNAs, miR‐124 and miR‐132. Our research implies that Spry1 participates in the regulation of CNS functions by influencing both neuronal differentiation under normal physiological processes and neuronal survival under pathological conditions.

Abbreviations

BDNF

brain‐derived neurotrophic factor

CNS

central nervous system

CREB

cAMP response element‐binding protein

DIV

days in vitro

ERK

extracellular signal‐regulated kinase

ESC

embryonic stem cell

FGF2

fibroblast growth Factor2

MAPK

mitogen‐activated protein kinase

NSC

neural stem cell

PI3K

phosphoinositide 3‐kinase

PLCγ

phospholipase C‐γ

RTKs

receptor tyrosine kinases

SPRY

sprouty

NGN1

neurogenin 1

shRNA

short hairpin RNA

EGF

epidermal growth factor

mESCs

mouse embryonic stem cells

EGFP

enhanced green fluorescent protein

1. INTRODUCTION

Understanding the mechanisms underlying neuronal development, neurodegeneration, and identification of novel regulatory molecules during neuronal development and prosurvival signaling cascades is crucial for the development of therapies for neurological disorders. Receptor tyrosine kinases (RTKs) are important mediators of signal transduction pathways in response to an extracellular stimulus that influences cell proliferation, differentiation, and survival (Schlessinger, 2014). Precise spatial and temporal regulation of RTK‐mediated signaling events ensures a physiologically appropriate biological outcome. Indeed, a large number of disease states result from dysregulation of RTK signaling (Neben, Lo, Jura, & Klein, 2017; Rosenzweig, 2012).

Over the last three decades, growth factors have emerged as critical players in the development, remodeling, and long‐term survival of brain structure and function (Castren & Antila, 2017; Oliveira et al., 2013). In addition, they appear to perform other, more acute modulatory functions that remain to be fully understood or explored. Recent studies have underscored the importance of feedback control of RTK function as a mechanism for ensuring signaling patterns that are compatible with the induction of a particular cellular response.

Sprouty (SPRY) proteins were first discovered in Drosophila (Hacohen, Kramer, Sutherland, Hiromi, & Krasnow, 1998). Subsequent research revealed that Drosophila Spry is a critical inhibitor of RTK‐mediated Ras signaling. The vertebrate orthologs of SPRY share similar sequences to Drosophila Spry in the C‐terminus but are different from each other, and markedly so from the Drosophila Spry protein, in the N‐terminus (de Maximy et al., 1999).

Early elegant studies demonstrated that SPRY proteins specifically inhibit RAS/extracellular signal‐regulated kinase (ERK)/mitogen‐activated protein kinase (MAPK) signaling pathway by RTKs, without interrupting the phosphoinositide 3‐kinase (PI3K)‐AKT, the phospholipase C‐γ (PLCγ) pathway, and other MAPK pathways (Gross, Bassit, Benezra, & Licht, 2001; Yusoff et al., 2002). The point at which SPRY regulate the ERK/MAPK signaling remains controversial, and considerable evidence to date suggests that multiple mechanisms exist, which depend on the various cellular context and/or the type of the RTK. SPRY proteins interact with several critical components of RTK signaling pathways, but how these interactions modulate ERK signaling is still unclear (Hanafusa, Torii, Yasunaga, & Nishida, 2002; Sasaki et al., 2003; Tefft et al., 2002). As a negative regulator of RTKs, SPRY themselves are subject to extensive regulation at multiple levels, including through differential cellular localization (Impagnatiello et al., 2001), posttranslational modification (X. Li, Brunton, Burgar, Wheldon, & Heath, 2004), and protein degradation (Hall et al., 2003).

Although amply expressed in the developing brain (Zhang, Lin, Itaranta, Yagi, & Vainio, 2001), the role of Spry genes in neuronal differentiation or survival has not yet been explored extensively. Given the essential roles of growth factors and neurotrophins in the regulation of neuronal differentiation and survival through their activation from RTK signaling, we investigated the potential role of Spry1, an important member of the Spry family, in neurons of the central nervous system.

2. MATERIALS AND METHODS

2.1. Ethics statement

This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of Southern Medical University (Permit number: SYXK2011‐0074).

2.2. Cell culture and transfection

Primary cortical neurons and NSCs were obtained from E16 and E10.5 murine embryos (C57BL/6 J mice, RRID: IMSR JAX: 000664; B6), respectively, as described previously. Briefly, dorsal cerebral cortex was microdissected and stripped of meninges and tissues were enzymatically treated with 0.05% trypsin‐ethylenediaminetetraacetic acid for 15 min at 37℃ in a humidified incubator (5% CO₂/95% air) and then mechanically dissociated into single‐cell suspensions. Neuron cells were cultured in poly‐l‐lysine‐coated 24‐well plates at a density of 100,000 cells per well in neurobasal medium (Cat #21103049; Life Technologies, Gaithersburg, MD) supplemented with 2% B27 (Cat #0080085SA; Life Technologies), 0.5 mM glutamine (Cat #25030081; Life Technologies), and penicillin–streptomycin (Cat #15070063; Life Technologies) and were grown at 37℃ in a humidified incubator. NSCs were cultured in suspension in Dulbecco's modified Eagle's medium (DMEM)/F‐12 (Cat #11320082; Life Technologies) supplemented with 1 mM l‐glutamine, 1% N2 supplement (Cat #17502048; Life Technologies), 0.5% B27 supplement (Life Technologies), 100 μg/ml penicillin–streptomycin (Life Technologies), 20 ng/ml epidermal growth factor (EGF; Cat #315‐09; PeproTech, London, UK), and 20 ng/ml fibroblast growth factor (bFGF/FGF2; Cat #100‐18B; PeproTech). N₂A cells were obtained from the American Type Culture Collection (Cat #CCL‐131) and maintained in DMEM (Cat #41965062; Life Technologies) supplemented with 10% fetal bovine serum (FBS; Cat #10270‐106; Life Technologies) and antibiotics. Cells were transfected with Lipofectamine 2000 (Cat #11668019; Life Technologies), according to the manufacturer's instructions.

2.3. Plasmids

The miR‐124 and miR‐9 expression plasmids (U6‐miR‐124‐CMV‐EGFP and U6‐miR‐132‐CMV‐EGFP) were constructed using the pGenesil‐1 vector (Cat #VRG0358; YRGENE, China). miR‐124 and miR‐9 precursor sequences and dozens of flanking nucleotides were amplified from mouse genomic DNA with the following primers: miR‐124, forward 5′‐GAGAATTCGCACGCGTCGCCAGCTTTTTC‐3′ and reverse 5′‐TCTCTAGATGCAGCTGC AGCGCTGAGATC‐3′; miR‐132, forward 5′‐GTCTGGATCCGTGCTGACGTCAGCCTGC‐3′ and reverse 5′‐GTCTAAGCTTCCAAAAAACTCCTCTTGCTCTGTATC‐3′ and were inserted into pGenesil‐1. The miR‐124 and miR‐132 mutant plasmids were constructed using the QuikChange Site‐Directed Mutagenesis Kit (Cat #200518; Stratagene, La Jolla, CA) with the sequences “AAGG” in miR‐124 changed to “TTCC” (underlined in the primer) and the sequences “ACAG” in miR‐132 changed to “TTTT” (underlined in the primer). The following primers were used to generate these mutations: miR‐124 , forward 5′‐AATGTCCATACAATTTTCCCACGCGGTGAATGCC‐3′ and reverse 5′‐GGCATTCACCGCGTGGGAAAATTGTATGGACATT‐3′; miR‐132, forward 5′‐GTGGGAACCGGAGGTATTTTTCTACAGCCATGGTCGC‐3′ and reverse 5′‐GCGACCATGGCTGTAGAAAAATACCTCCGGTTCCCAC‐3′. The shRNA plasmids against SPRY1 were constructed using the pGenesil‐1 vector and target sequences used were: 5′‐GAATACACAGAGGGACCTTCG‐3′ for Spry1‐kd1 and 5′‐GCAGGTGTAGAAACTCCAACA‐3′ for Spry1‐kd2. The 3′‐UTR sequence of the Spry1 mRNA, which contained the miR‐124 and miR‐132 binding sites, was amplified from mouse brain complementary DNA (cDNA) using the following primers: 5′‐GTCTGAATTCTTAAGCAAACTGTCATGACTTC‐3′ and 5′‐GTCTGGATCCTTCTTCGAGTCACCTTGC‐3′. The amplified sequence was inserted into the 3′‐UTR of the EGFP gene of the pEGFP‐C1 vector (Cat #6084‐1; Clontech Laboratories, Inc., Mountainview, CA) with a stop codon (TAA, underlined in the primer) inserted in frame with EGFP gene to acquire the sensor plasmid EGFP‐Spry1‐3′‐UTR. The plasmid EGFP‐Spry1‐3′‐UTRMut carries the following mutations in the “seeds” of the conserved miR‐124‐binding sites: “TGCCTTA” is mutated to “TAAAATA” and “seeds” of the conserved miR‐132‐binding sites: “GACTGTT” is mutated to “GAAAAAA.” For the SPRY1 and NGN1 expression constructs, the full‐length coding region of Spry1 or Ngn1 was amplified from mouse brain cDNA and inserted into the pLVX‐IRES‐tdTomato vector (Cat #63628; Clontech). The primers used are: forward 5′‐GTCTGAATTCATCACTCCACATGGATTC‐3′ and reverse 5′‐GTCTGGATCCGAAGTCATGACAGTTTGC‐3′ for Spry1; 5′‐GTCTGAATTCCGTCTGTCGGTCCTGCAC‐3′ and reverse 5′‐GTAATCTAGACAAAGGCCTAGTGGTATG‐3′ for Ngn1. The nucleotides of miR‐124 mimics and miR‐132 mimics were purchased from GenePharma (Shanghai, China).

2.4. Luciferase assay

Neurons seeded in 24‐well plates were transfected with 0.8 μg of the corresponding expression plasmids, 0.1 μg of the cAMP response element‐binding protein (CREB) firefly luciferase reporter plasmid p‐CREB‐Luc (Invitrogen) and 0.08 μg of the Renilla luciferase control plasmid pRL‐TK (Promega, Madison, WI). Transfections were performed using Lipofectamine 2000 (Life Technologies). Each treatment was performed in triplicate in three independent experiments. The activities of firefly and Renilla luciferase were measured using the dual‐luciferase assay system (Cat #E1910; Promega).

2.5. Western blot analysis

Cells cultured in 24‐well plates were lysed in 80 µl radioimmunoprecipitation assay buffer (50 mM Tris; pH 8.0; 150 mM NaCl, 1% Nonidet P40, and 0.1% Sodium dodecyl sulfate [SDS]) buffer. Extracts were sonicated and boiled for 10 min followed by centrifugation at 12,000g for 5 min. Protein extracts (40 µg per well) were separated with 12% SDS‐polyacrylamide gel electrophoresis, electrotransferred to nitrocellulose membranes, blocked with fat‐free milk (5% in Tris‐buffered saline and polysorbate 20 [TBST]), and incubated with primary antibodies (rabbit anti‐SPRY1, Cat #13013; 1/2,000; rabbit anti‐MAP2, Cat #4542; 1/2,000; rabbit anti‐synapsin I, Cat #D12G5; 1/1,000; rabbit anti‐nestin 1/1,000; rabbit anti‐ERK, Cat #4695; 1/1,000; rabbit anti‐p‐ERK, Cat #4370; 1/1,000; rabbit anti‐CREB, Cat #9197; 1/1,000 [Cell Signaling, Beverly, MA]; mouse anti‐TUJ1, Cat #AT809; 1/1,000; rabbit anti‐green fluorescent protein (GFP), Cat #AF1483; 1/1,000 [Beyotime Institute of Biotechnology, Nantong, Jiangsu, China]; rabbit anti‐p‐CREB, Cat #ab32096; 1/1,000 [Abcam Inc., Cambridge, MA]) and the relevant secondary horseradish peroxidase–conjugated antibody (Cat #ZB2301, #ZB5305; 1/5,000; Zhongshan Jinqiao Biotech, Beijing, China). Membranes were developed by enhanced chemiluminescence detection.

2.6. Immunocytochemistry, Hoechst staining, and fluorescence microscopy

Neurons were fixed with 4% paraformaldehyde in phosphate‐buffered saline (PBS) for 15 min at room temperature. Subsequently, cells were permeabilized and blocked with 0.3% Triton X‐100 and 5% bovine serum albumin in PBS for 30 min. Primary antibodies were added in the same buffer without Triton X‐100 at a concentration of 1/500 for 16 hr. After three PBS washes, Alexa Fluor 595 or 488 secondary antibodies (Cat #A11008, #35510; 1/2,000; Life Technologies) were added for 2 hr. Cells were then washed three times with PBS and incubated for 10 min with the nuclear Hoechst 33258 dye (Cat #B1155; 1 μg/ml; Sigma‐Aldrich, St Louis, MO). Fixed cells were observed at ×10, ×20, or ×40 with an Olympus IX71 microscope and images were acquired with an Olympus DP72 camera (Olympus, Tokyo, Japan) linked to image acquisition software (Olympus, Rungis, France) or the ZOE Fluorescent Cell Imager (Bio‐Rad Laboratories Inc., Hercules, CA). Cell morphology or fluorescence was examined under the same instruments, and neurite length was manually measured using the ImageJ software (National Institutes of Health, Bethesda, MD).

2.7. Real‐time quantitative polymerase chain reaction (qPCR)

The total RNA was extracted from neuron cells using TRIzol (Cat #15596026; Life Technologies), according to the manufacturer's instructions. For real‐time (RT)‐qPCR of protein‐coding genes, the first‐strand cDNA was reverse‐transcribed from 1 μg of total RNA using ReverTra Ace qPCR RT Master Mix (Cat #FSQ‐101; Toyobo, Tokyo, Japan). RT‐qPCR was performed with a Stratagene Mx3005P qPCR system (Agilent Technologies, Palo Alto, CA) using SYBR Green (Cat #QPS‐101; Toyobo). The housekeeping gene Gapdh was used as a loading control. The following primers were used: Gapdh, forward 5′‐AGAAGGCTGGGGCTCATTTG‐3′ and reverse 5′‐AGGGGCCATCCACAGTCTTC‐3′; Spry1, forward 5′‐CAGGTCATAGGTCAGATC‐3′ and reverse 5′‐TCTCCACATTTGCACTTG‐3′. miRNA levels were measured with RT‐qPCR using miDETECT A Track™ miRNA qRT‐PCR (Cat #R10048.3; RiboBio, Guangzhou, China). Primers for miR‐124, miR‐132, and U6 small nuclear RNA were obtained from RiboBio. The RNA expression levels were normalized to the expression of the internal control using the $2^{-\mathrm{\Delta }\mathrm{\Delta }{C}_{\mathrm{t}}}$ method.

2.8. Statistical analysis

All experiments were repeated at least three times. All statistical analyses were performed using one‐way analysis of variance or Student's t test. The data were expressed as the mean ± SD. p < 0.05 was considered statistically significant.

3. RESULTS

3.1. SPRY1 is expressed in cortical neurons and is upregulated during neuronal differentiation

To study the role of SPRY1 in the CNS, we used the broadly utilized mouse primary cortical neurons as a cellular model. We first examined SPRY1 expression during cortical neuron differentiation. Cortical neurons were isolated from E16–E17 mouse cortex and cultured in the appropriate neuronal growth media. At Day 1, plated cells were round, without neurites. From Days 3 to 11, these immature neurons differentiated, grew neurites and gradually formed complex neuronal networks (Figure 1a). Immunocytofluorescence with anti‐SPRY1 and anti‐TUJ1 antibodies at Day 7 confirmed coexpression of SPRY1 and neuron‐specific TUJ1 in the same cells (Figure 1b). Neurons were collected at different developmental time‐points, as indicated in Figure 1c, to examine protein expression of SPRY1. The differentiation status of the neurons was determined by examining the expression of the neuronal markers MAP2 and synapsin1. At Day 1, the expression level of SPRY1 was relatively low, became stronger at Day 3 and further increased as neurons matured.

Figure 1.

Expression of SPRY1 in cortical neurons. (a) Morphology of cortical neurons from Days 1 to 11 in culture. (b) Detection of SPRY1 and TUJ1 in differentiated cortical neurons at Day 7 with immunocytofluorescence (scale bar, 50 μm). (c) Western blot showing SPRY1 MAP2, synapsin1, and α‐tubulin expression in primary cortical neurons at the indicated time‐points of culture. (d) Western blot showing SPRY1, TUJ1, nestin, and α‐tubulin expression in neural stem cells and primary cortical neurons at Day 9 of culture. (e) Real‐time quantitative polymerase chain reaction (qPCR) showing Spry1 mRNA expression in neural stem cells and primary cortical neurons at the indicated day of culture; the values are in means ± SD; *p < 0.05 and **p < 0.01, obtained in a typical experiment performed in triplicate. mRNA: messenger RNA; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

We also compared expression of the SPRY1 protein in Day 9 neurons and neural stem cells (NSCs). Neuroepithelial cells were isolated from E10.5 cortex and cultured in suspension for 6 days in vitro (DIV) in the presence of the EGF and FGF2 and then plated onto poly‐l‐lysine‐coated dishes for another 2 days before lysis. The NSC marker nestin and the neuron marker TUJ1 were used to illustrate the purity of cells. The Day 9 neurons showed a substantially higher level of TUJ1 and lower level of nestin than NSCs (Figure 1d). The expression level of SPRY1 was dramatically higher in NSCs than in Day 9 neurons (Figure 1d). We also examined Spry1 messenger RNA (mRNA) expression in NSCs and neurons. Consistent with the expression pattern of SPRY1 protein, the mRNA level of Spry1 exhibited higher expression in NSCs than in neurons (Figure 1e) and expression increased gradually during neuronal differentiation (Figure 1e).

3.2. MicroRNAs and ERK/CREB signaling pathway regulate SPRY1 expression in cortical neurons

Next, we sought to explore the regulatory mechanisms underlying SPRY1 expression during neuronal differentiation. In recent years, microRNAs (miRNAs) have attracted a lot of attention as regulators of gene expression (Lewis, Burge, & Bartel, 2005) and are essential for cellular differentiation and development (Huang et al., 2011). We hypothesized that miRNAs may play a role in regulating expression of SPRY1 during neuronal differentiation. Analysis with the widely used PicTar (http://pictar.mdc‐berlin.de) (Krek et al., 2005) and TargetScan (http://www.targetscan.org) (H. Liu, Yue, Chen, Gao, & Huang, 2010) software programs, revealed miR‐124 and miR‐132, two CNS‐enriched miRNAs, as potential regulators of SPRY1. The predicted canonical binding sites in the Spry1 locus are conserved among mammals listed in the programs (Figure 2a).

Figure 2.

SPRY1 expression is regulated by miR‐124 and miR‐132. (a) Sequence alignment of the 3′‐UTR of the Spry1 locus from various species. (b) Schematic diagram of the sensor construct. (c) Immunoblotting for enhanced green fluorescent protein (EGFP) expression showed the importance of the miR‐124‐ and miR‐132‐binding sites for SPRY1 expression. Immunoblotting for α‐tubulin served as a loading control. (d) Representative immunoblot of endogenous expression of SPRY1 in cells transfected with the indicated plasmids. (e) A scheme for experimental design. (f) EGFP expression in the Day 7 cortical neurons for the indicated plasmids is shown (scale bar, 200 μm). (g) Immunoblotting with anti‐GFP antibody showing expression of the GFP. (h) qPCR showing miR‐124 and miR‐132 expression in NSCs and primary cortical neurons at the indicated day of culture; the values are in means ± SD; *p < 0.05, **p < 0.01, and ***p < 0.001, obtained in a typical experiment performed in triplicate. (i) Western blot showing expression of active ERK (p‐ERK) and active CREB (p‐CREB) in primary cortical neurons at the indicated day of culture. (j) Day 9 cortical neurons were grown in the presence of the indicated inhibitors (U0126, 20 μM; SGC‐CBP30, 2.5 μM) for 36 hr before lysis and immunoblotting were performed with the indicated antibodies. CREB: cAMP response element‐binding protein; ERK: extracellular signal‐regulated kinase; GFP: green fluorescent protein; miR: microRNA; NSC: neural stem cell; qPCR: quantitative polymerase chain reaction [Color figure can be viewed at wileyonlinelibrary.com]

To experimentally verify that Spry1 is a direct target gene of miR‐124 and miR‐132, we constructed a sensor plasmid inserting the 3′‐UTR sequences of the Spry1 gene, including the two miRNA‐binding sites, downstream of an EGFP expression cassette (Figure 2b, herein referred to as EGFP‐3′‐UTR). We transfected EGFP‐3′‐UTR into N₂A cells together with miR‐124, or miR‐132 or control plasmids and the expression of EGFP was analyzed with immunoblotting. miR‐124 or miR‐132 overexpression caused a dramatic reduction in the expression of EGFP‐3′‐UTR (Figure 2c, cf., lanes 2 and 3 to lane 5). Moreover, miR‐124 and miR‐132 coexpression suppressed EGFP expression to a greater extent than miR‐124 or miR‐132 plasmids alone (Figure 2c, lane 1). However, neither miR‐124 nor miR‐132 suppressed EGFP expression when their binding sites on the sensor plasmid were mutated (Figure 2c, lane 6). Furthermore, endogenous SPRY1 protein expression is reduced in N₂A cells transfected with miR‐124 or miR‐132 mimics (Figure 2d). Transfection of both miR‐124 and miR‐132 mimics decreased SPRY1 levels to a greater extent (Figure 2d), consistent with our previous results. Taken together, our results suggest an additive effect of these two miRNAs on SPRY1 expression.

To explore whether endogenously expressed miR‐124 and miR‐132 were able to reduce SPRY1 expression in primary cortical neurons, we transfected cells with equal amounts of either EGFP‐3′‐UTR or EGFP‐3′‐UTRmut and observed the fluorescence of EGFP 24 hr later (Figure 2e). An obvious reduction of EGFP fluorescence was observed in EGFP‐3′‐UTR‐transfected cells, whereas rescue of EGFP expression was observed in neurons expressing EGFP‐3′‐UTRmut (Figure 2f). We also examined EGFP expression with immunoblot and found the same change tendency (Figure 2g). These results indicated that endogenous miR‐124 and miR‐132 regulate SPRY1 expression, depending on the miRNA‐binding sites located in the 3′‐UTR of the gene.

We next measured the expression of miR‐124 and miR‐132 in NSCs and developing neurons. The levels of miR‐124 and miR‐132 were considerably higher in Day 1 postmitotic neurons than in NSCs. Their expression is inversely related to SPRY1 expression, which is consistent with our conclusion that miR‐124 and miR‐132 inhibit SPRY1 expression. Paradoxically, however, during development and maturation of neuron cells, as the expression levels of miR‐124 and miR‐132 gradually increased, SPRY1 expression exhibited a positive correlation with the two miRNAs (Figure 2h and Figure 1c). As the ERK/CREB pathway stimulates SPRY gene expression in various cellular contexts, we next examined ERK activity during neuron cell development. We found that, as neurons mature, the level of active ERK (p‐ERK) and the active form of its downstream substrate CREB (p‐CREB) gradually increased (Figure 2i). To examine whether the elevated ERK and CREB activity contributed to the increased level of SPRY1, we inhibited ERK or CREB activity with specific inhibitors on Day 9 developing neurons and determined SPRY1 expression 36 hr later. As shown in Figure 2j, either the selective inhibitor of ERK1/2, U0126, or the selective inhibitor of CREB, SGC‐CBP30 dramatically reduced expression of SPRY1, which suggested that ERK/CREB pathway promotes SPRY1 expression during neuronal differentiation. These results suggest that SPRY1 is bidirectionally regulated during neuronal development by miRNAs and the ERK/MAPK signaling pathway.

3.3. Treatment with neurotrophic factors represses SPRY1 expression

As SPRY proteins are well‐studied regulators of RTK signaling and are usually upregulated by RTK signaling (Impagnatiello et al., 2001; Ozaki et al., 2001; Sasaki, Taketomi, Wakioka, Kato, & Yoshimura, 2001), we next explored whether this occurs in neurons. The Day 7 neurons were treated with 25 ng of brain‐derived neurotrophic factor (BDNF) or FGF2 and the SPRY1 expression was examined with immunoblotting at the indicated time‐points. Surprisingly, we did not observe an increase in SPRY1 protein expression; in contrast, SPRY1 expression initially decreased after neurotrophic factor treatment, and subsequently gradually increased to the initial level (Figure 3a,b). We further examined the miR‐124 and miR‐132 levels during this process and found that miR‐132 expression was dramatically increased upon BDNF but not FGF2 treatment; neither treatment changed the expression of miR‐124 (Figure 3c,d). These results suggested that miR‐132 may mediate the regulation of SPRY1 after BDNF treatment.

Figure 3.

Regulation of SPRY1 expression by BDNF and FGF2. Influence of BDNF (a) or FGF2 (b) on SPRY1 protein expression in cortical neurons. The Day 7 cortical neurons were grown in the presence of 25 ng/ml of BDNF or FGF2 for the indicated time before lysis. Cell extracts were analyzed by immunoblotting with an antibody against SPRY1. qPCR showing expression of miR‐124 and miR‐132 in primary cortical cells at the indicated time‐points of culture upon BDNF (c) or FGF2 (d) treatment; the values are in means ± SD; *p < 0.05 and **p < 0.01, obtained in a typical experiment performed in triplicate. Western blot showing p‐ERK expression in differentiated N₂A cells transfected with the indicated plasmids in the presence of BDNF (e) or FGF2 (f). (g) Immunocytofluorescence showing expression of p‐ERK in differentiated cortical neurons transfected with the indicated plasmids after BDNF or FGF2 treatment of 2 hr at Day 8 (scale bar, 100 μm). BDNF: brain‐derived neurotrophic factor; ERK: extracellular signal‐regulated kinase; FGF2: fibroblast growth factor 2; SPRY1: sprouty1; qPCR: quantitative polymerase chain reaction [Color figure can be viewed at wileyonlinelibrary.com]

To explore the effect of SPRY1 on ERK signaling in neuronal differentiation, we transfected N₂A cells with SPRY1 overexpression or silencing plasmids together with a puromycin expression vector. The transfected cells were selected with puromycin for 24 hr. Then the cell culture medium was changed with DMEM plus 1% FBS containing 10 μM retinoic acid to induce neuronal differentiation. Three days after neuronal differentiation, BDNF or FGF2 were added to the culture for 2 hr before immunoblotting. SPRY1 overexpression repressed, whereas SPRY1 knockdown enhanced p‐ERK level compared with control groups (Figure 3e,f). Further, we transfected the Day 7 cortical neurons with the SPRY1 overexpression plasmid for 24 hr followed by neurotrophic factor treatment of 2 hr before immunocytochemistry analysis. As shown in Figure 3g, SPRY1 overexpression reduced the fluorescence intensity of p‐ERK. These results suggested that SPRY1 negatively regulates neurotrophic factor‐stimulated RTK signaling in differentiated neurons.

3.4. Knockdown of SPRY1 expression promotes neurite outgrowth

As both miR‐124 and miR‐132 are important for neurite development, we next examined whether SPRY1 is involved in this process. To explore whether inhibition of SPRY1 activity may influence neurite development, two short hairpin RNA (shRNA) plasmids SPRY1‐kd1 and SPRY1‐kd2 were generated to knockdown SPRY1 expression. To determine the knockdown efficiency, we cotransfected N₂A cells with SPRY1 overexpression and the silencing plasmids as indicated in Figure 4a together with the puromycin expression vector, followed by puromycin selection and immunoblot 2 days later. Our results indicated that transfection with the overexpression construct dramatically increased SPRY1 protein expression and both Spry1‐kd1 and Spry1‐kd2 can effectively reduce the amount of SPRY1. We then used these validated plasmids to investigate the function of SPRY1 in neuronal neurite development. N₂A cells were transfected with the two silencing plasmids or the control vector for 48 hr, then the cell culture medium was replaced with the DMEM plus 1% fetal bovine serum and 10 μM retinoic acid to induce neuronal differentiation. Cell morphology was observed 48 hr later under fluorescence microscope by monitoring GFP fluorescence. We found that knockdown of endogenous SPRY1 expression by the respective shRNA constructs resulted in increased neurite outgrowth (Figure 4b,c). Next, we cotransfected the silencing plasmids together with a bright red fluorescent protein (RFP) expressing vector (pmCherry‐N1) into cultured Day 2 cortical neurons. Cell morphology was observed under a fluorescence microscope by monitoring RFP fluorescence 48 hr later. The knocked‐down SPRY1 did not influence the neurite outgrowth without neurotrophic factors. However, when the neurotrophic factors BDNF or FGF2 were added to the culture medium, Spry1‐kd1 significantly stimulated neurite elongation (Figure 4d,e).

Figure 4.

Downregulation of SPRY1 favors neurite elongation. (a) Immunoblotting for SPRY1 expression showed that Spry1‐kd1 and Spry1‐kd2 reduced the amount of overexpressed SPRY1 in N₂A cells. (b) The morphology of N₂A cells, transfected with the indicated plasmids, was analyzed by monitoring GFP fluorescence (scale bar, 50 μm). (c) Histogram showing the stimulation of neurite outgrowth in N₂A cells after silencing of SPRY1. The values are in means ± SD (n = 30 cells); ***p < 0.001. (d) The morphology of neurons, transfected with the indicated plasmids, was analyzed by monitoring RFP fluorescence (scale bar, 200 μm). (e) Histogram showing the stimulation of neurite outgrowth in cortical neurons after silencing of SPRY1. The values are in means ± SD (n = 20 neurons); ***p < 0.001. GFP: green fluorescent protein; RFP: red fluorescent protein; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

3.5. Overexpression of SPRY1 inhibits neurite outgrowth

To explore whether upregulation of SPRY1 expression may influence neurite development, N₂A cells were cotransfected with plasmids expressing SPRY1 and the proneural factor, neurogenin 1 (NGN1). After 24 hr of transfection, the cell culture medium was changed with DMEM plus 1% FBS and 10 μM retinoic acid to induce neuronal differentiation. Cell morphology was observed 24 hr later under a fluorescence microscope by monitoring RFP fluorescence. As shown in Figure 5a,b, overexpression of NGN1 strongly promoted neurite outgrowth of differentiated N₂A cells, and SPRY1 coexpression dramatically inhibited this effect. We also overexpressed SPRY1 in developing cortical neurons. Cortical neurons were transfected with the corresponding vector on Day 2 and the neurite length was measured 48 hr later. Overexpression of SPRY1 did not influence neurite outgrowth without neurotrophic factors; however, when the neurotrophic factors BDNF or FGF2 were added to the culture medium, SPRY1 overexpression significantly reduced the total neurite length (Figure 5c,d). Together, these results indicated that SPRY1 negatively regulates neurite elongation in response to neurotrophic factors.

Figure 5.

Overexpression of SPRY1 inhibits neuritogenesis. (a) The morphology of N₂A cells, transfected with the indicated plasmids, was analyzed by monitoring RFP fluorescence (scale bar, 50 μm). (b) Histogram showing the inhibition of neurite outgrowth in N₂A cells after overexpression of SPRY1. The values are in means ± SD (n = 30 cells);***p < 0.001. (c) The morphology of neurons, transfected with the indicated plasmids, was analyzed by monitoring RFP fluorescence (scale bar, 200 μm). (d) Histogram showing the inhibition of neurite outgrowth in cortical neurons after overexpression of SPRY1. The values are in means ± SD (n = 20 neurons); ***p < 0.001. RFP: red fluorescent protein; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

3.6. SPRY1 inhibits miR‐124‐ and miR‐132‐mediated neurite outgrowth

To determine whether SPRY1 may influence neurite growth mediated by miR‐124 and miR‐132, N₂A cells or cortical neurons were cotransfected with miR‐124, miR‐132, and SPRY1 expression constructs and then differentiated as previously. As shown in Figure 6a–d, expression of miR‐124 and miR‐132 greatly promoted neurite elongation of both differentiated N₂A cells and primary cortical neurons, whereas these effects were attenuated when SPRY1 was coexpressed in N₂A cells and in neurons in the presence of BDNF or FGF2, thus suggesting that SPRY1 inhibited neurite growth stimulated by miR‐124 and miR‐132.

Figure 6.

SPRY1 expression inhibits the neurite stimulating effect of miR‐124 and miR‐132. (a) The morphology of N₂A cells, transfected with the indicated plasmids, was analyzed by monitoring RFP fluorescence (scale bar, 50 μm). (b) Histogram showing the inhibition of miR‐124‐ and miR‐132‐mediated neurite outgrowth in N₂A cells after overexpression of SPRY1. The values are in means ± SD (n = 30 cells); ***p < 0.001. (c) The morphology of neurons, transfected with the indicated plasmids, was analyzed by monitoring RFP fluorescence (scale bar, 200 μm). (d) Histogram showing the inhibition of miR‐124‐ and miR‐132‐mediated neurite outgrowth in cortical neurons after overexpression of SPRY1. The values are in means ± SD; (n = 20 neurons); ***p < 0.001. miR: microRNA; RFP: red fluorescent protein; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

3.7. SPRY1 antagonizes the neuronal protective effect of neurotrophic factors and exacerbates glutamate‐induced neurotoxicity

FGF2 and BDNF have been involved in promoting the survival of CNS neurons (B. Cheng & Mattson, 1991; Ghosh, Carnahan, & Greenberg, 1994). Therefore, we hypothesized that SPRY1 may influence neuronal survival through the regulation of the RAS/ERK/MAPK signaling pathway or other pathways induced by these neurotrophic factors. To study whether SPRY1 was involved in the regulation of neurotrophic factor‐mediated neuronal protection, Day 12 cortical neurons were transfected with SPRY1‐expressing or control plasmids. BDNF or FGF2 was added to culture medium at a concentration of 25 ng/ml at Day 14 followed by glutamate treatment 2 hr later. RFP‐positive neurons were observed by fluorescence microscopy. Cells with condensed or fragmented nuclei were considered as dead or dying (apoptotic; Figure 7a, lower panel). In the control group, approximately 15–30% of transfected neurons were apoptotic. Overexpression of SPRY1 leads to a dramatic increase in apoptosis, reaching approximately 80% in the BDNF group and 70% in the FGF2 group (Figure 7b,c). These results indicated that SPRY1 counteracts neurotrophic factor‐mediated neuronal protection.

Figure 7.

SPRY1 inhibits the neuronal protective effect of neurotrophic factors in glutamate‐induced neurotoxicity stress. Mature cortical neurons were transfected for 48 hr with an expression plasmid for SPRY1 or control plasmid and were treated with 25 ng/ml BDNF or FGF2 for 2 hr before addition of glutamate (200 μM) for another 4 hr. After fixation and Hoechst 33258 dye staining, RFP‐positive cells (a, yellow arrows) were identified by fluorescence microscopy and nucleus morphology was examined (a, white arrows): the arrow in the upper shows a representative healthy transfected cell; the two arrows in the lower panel denote dying cells (scale bar, 50 μm). Percentages of apoptotic neurons among RFP‐positive cells in the BDNF treatment group (b) or in the FGF2 treatment group (c) were quantified and presented as means ± SD; ***p < 0.001, obtained in a typical experiment performed in triplicate. BDNF: brain‐derived neurotrophic factor; FGF2: fibroblast growth factor 2; RFP: red fluorescent protein; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

3.8. SPRY1 regulates neurotrophic factor‐mediated neuronal survival through inhibition of the ERK/CREB signaling pathway

As CREB is an important downstream molecule of the RAS/ERK/MAPK pathway in the regulation of cell survival (Jhala et al., 2003; Walton & Dragunow, 2000), we next investigated whether SPRY1 regulates neuronal survival under glutamate stress by affecting the activity of CREB. Neurons were cotransfected with the SPRY1‐expressing SPRY1 knockdown or control vectors together with a CRE‐luciferase reporter plasmid after 7 DIV and stimulated 24 hr later with FGF2 or BDNF for 6 hr. Expression of SPRY1 inhibited luciferase activity, whereas SPRY1 knockdown enhanced luciferase activity upon BDNF or FGF2 treatment (Figure 8a,b). We further verified the critical role of CREB activity on neuronal survival with ERK/CREB pathway agonists or antagonists. As shown in Figure 8c,d, BDNF, FGF2, or db‐cAMP (CREB agonists) treatment significantly decreased neuronal death caused by cytotoxic glutamate stimulation, which can be partially prevented by the ERK inhibitor U0126 or the CREB inhibitor SGC‐CBP30. Together, these results suggest that neurotrophic factors regulated SPRY1 expression promotes CREB activity which may least partially contributed to the prosurvival effect of these neurotrophic factors.

Figure 8.

SPRY1 attenuates neurotrophic factor‐mediated cell survival via a CREB‐dependent mechanism. Neurons were cotransfected with p‐CREB‐luc and the indicated plasmids, and were stimulated 48 hr later with BDNF (a) or FGF2 (b) (25 μg/ml, 6 hr). The ratio of firefly to Renilla luciferase luminescence was recorded and expressed as relative luciferase units. Data are means ± SD, n = 3. (c) Mature cortical neurons were treated with the indicated molecules. After fixation and Hoechst 33258 dye staining, their nucleus morphology was examined (scale bar, 50 μm). (d) The percentage of transfected neurons undergoing apoptosis is shown. Percentages of apoptotic neurons were quantified and presented as means ± SD, (n = 15); ***p < 0.001. BDNF: brain‐derived neurotrophic factor; CREB: cAMP response element‐binding protein: FGF2: fibroblast growth factor 2 [Color figure can be viewed at wileyonlinelibrary.com]

3.9. Glutamate‐mediated cytotoxicity differentially regulates SPRY1 expression at different developmental stages

To examine whether glutamate treatment influences expression of SPRY1, primary cultures of Days 7 and 14 cortical neurons were exposed to a neurotoxic dosage of glutamate and expression of SPRY1, miR‐124 and miR‐132 were examined at different time‐points as indicated in Figure 9a–d. Glutamate treatment results in a gradual downregulation of SPRY1 in Day 7 developing neurons. However, in mature neurons, SPRY1 expression level increased at 0.5 hr, and is maintained during the following hours. Expression of miR‐124 and miR‐132 initially decreased and then increased; miR‐132 expression returned back to the initial levels in the Day 7 neurons, whereas miR‐124 and miR‐132 expression declined and maintained at relatively low levels in mature neurons upon glutamate treatment (Figure 9b,d). Furthermore, we also examined the change of ERK and CREB activity during these conditions by immunoblotting. We observed a sharp decrease in the levels of both phosphorylated ERK and phosphorylated CREB in the Day 14 neurons but not in the Day 7 neurons (Figure 9a,c). We next compared the effect of glutamate on neuronal death in the Days 7 and 14 neurons. Consistent with previous studies, we found that immature Day 7 neurons were significantly more resistant to glutamate cytotoxicity than Day 14 neurons (Figure 9e). When SPRY1 was knocked‐down in mature neurons, the apoptosis rate dramatically decreased (Figure 9f). Finally, we examined whether SPRY1 influences CREB activity upon glutamate stimulation. Day 12 neurons were transfected with Spry1‐kd or control plasmids and treated with glutamate on Day 14. As phosphorylated CREB decreased rapidly upon glutamate stimulation in mature neurons, we examined p‐CREB by cytoimmunofluorescence 15 min after glutamate stimulation to preclude cell apoptosis. As shown in Figure 9g, SPRY1 knockdown did not influence the p‐CREB immunofluorescence levels, which indicated that signaling pathways other than ERK/CREB participated in SPRY1 regulation of neuronal survival during excitatory neurotoxicity caused by glutamate.

Figure 9.

Glutamate stress differently regulates SPRY1 expression at different developmental stages. (a) Day 7 cortical neurons were treated with 50 or 200 μM glutamate. Western blot showing expression of various proteins in primary cortical neurons in the indicated time‐points of culture. (b) qPCR showing miR‐124 and miR‐132 expression in primary cortical neurons in the indicated time‐points of culture. (c) Day 14 cortical neurons were treated with 50 or 200 μM glutamate. Western blot showing expression of various proteins in primary cortical neurons in the indicated time‐points of culture. (d) qPCR showing miR‐124 and miR‐132 expression in primary cortical neurons in the indicated time‐points of culture. (e) The percentage of neurons undergoing apoptosis after stimulation of Days 7 or 14 neurons with 200 μM glutamate for 4 hr is shown. Percentages of apoptotic neurons were quantified and presented as means ± SD; ***p < 0.001, obtained in a typical experiment performed in triplicate. (f) Mature cortical neurons were transfected for 48 hr with the indicated plasmids, stimulated with 200 μM glutamate for 4 hr. The percentage of transfected neurons undergoing apoptosis is shown and presented as means ± SD; ***p < 0.001. (g) Immunocytofluorescence showing expression of p‐CREB in mature cortical neurons transfected with the indicated plasmids after glutamate treatment for 15 min. Arrows in the pictures in the upper and lower panels indicate the same cell (scale bar, 200 μm). CREB: cAMP response element‐binding protein; miR: microRNA; qPCR: quantitative polymerase chain reaction; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

4. DISCUSSION

Neurotrophic factors are generally regarded as important mediators of neuronal survival and differentiation (Henderson, 1996). These effects are primarily mediated by binding of neurotrophic factors to their high‐affinity tyrosine kinase receptors, which initiate some parallel signaling cascades, including activation of PLCγ, ERK, and PI3K pathways. In this study, we provide evidence that SPRY1, a neurotrophic factor‐regulated gene, is progressively increased during differentiation of primary cortical neurons, is negatively regulated by the neurotrophins BDNF and FGF2, is repressed by the CNS‐specific miRNAs, miR‐124 and miR‐132, and is positively regulated by ERK/MAPK signaling. Glutamate neurotoxicity also influences SPRY1 expression, which varies at different developmental stages, and inhibition of SPRY1 expression reduced glutamate neurotoxicity in mature neurons. We also demonstrated that SPRY1 expression influences neuronal physiology by establishing a synergistic feedback loop with BDNF or FGF2, whose excessive expression reduces neuritogenesis and survival (Figure 10).

Figure 10.

Model showing the function of SPRY1 in cortical neurons. Expression of SPRY1 is regulated by the BDNF/ERK/CREB pathway and neuron‐enriched miRNAs, miR‐124 and miR‐132 in cortical neurons during terminal differentiation. SPRY1 inhibits BDNF‐ and FGF2‐induced signaling cascades and participates in the control of neurite elongation and neuronal survival. Therefore, SPRY1 is involved in a positive‐feedback loop initiated by BDNF or FGF2, thus controlling neurite growth and neuronal survival mediated by these neurotrophic factors. BDNF: brain‐derived neurotrophic factor; CREB: cAMP response element‐binding protein; ERP: extracellular signal‐regulated kinase; FGF2: fibroblast growth factor 2; miR: microRNA; SPRY1: sprouty1 [Color figure can be viewed at wileyonlinelibrary.com]

4.1. SPRY1 forms a positive‐feedback loop that regulates BNDF and FGF2 activity in cortical neurons

In the mouse embryo, SPRY1 is expressed in the neocortex, cranial flexure, and primordium of the cerebellum around embryonic days (Zhang et al., 2001). Expression of SPRY1 in primary cultures of differentiated cortical neurons was confirmed with coexpression with the neuronal marker TUJ1. Although SPRY1 expression increased during neuronal differentiation of cortical neurons, expression was much lower than in NSCs. This result is similar to that of a previous study, which examined SPRY1 expression in undifferentiated mouse embryonic stem cells (mESCs) and also showed that SPRY1 was dispensable for the maintenance of mESC self‐renewal; expression of SPRY1 was decreased during neural induction of mESCs and suppression of SPRY1 expression in mESC promoted neural differentiation (Jung et al., 2012).

miRNAs are small noncoding RNAs that act as important posttranscriptional gene expression regulators and have been implicated in the regulation of developmental and disease processes (Bartel, 2004; Stefani & Slack, 2008). miRNAs mainly act as repressors of gene expression either by inducing cleavage of their target mRNAs or by inhibiting translation of target mRNAs (Cai, Yu, Hu, & Yu, 2009). miRNAs are also important for adult neurogenesis in vivo in the subventricular zone stem cell niche (L. C. Cheng, Pastrana, Tavazoie, & Doetsch, 2009). In the current study, we found that expression of SPRY1 was repressed by the neuron‐enriched miRNAs, miR‐124 and miR‐132. Consistent with previous studies, expression of miR‐124 and miR‐132 in NSCs was substantially lower than in neurons (L. C. Cheng et al., 2009; Luikart et al., 2011). However, expression of SPRY1 in neurons was not completely repressed but gradually increased in parallel with miR‐124 and miR‐132 as neurons matured. Our results revealed that this upregulation was caused by the activation of the ERK/CREB pathway, which promoted SPRY1 expression during neuronal maturation. This discovery is not surprising as previous studies have revealed that miRNAs not only reduce expression of target genes but also limited too high expression of their target genes under certain conditions: for example, expression of the neuron‐enriched protein MeCP2 is repressed by miR‐132 in neurons, thus maintaining MeCP2 protein levels in a high but narrow range, which is essential for CNS normal development (Klein et al., 2007).

Previous findings suggested that Sprouty gene expression is usually positively regulated by the ERK pathway downstream of RTK (Minowada et al., 1999; Ozaki et al., 2001; Sasaki et al., 2001) and forms a negative‐feedback loop with RTK signaling. In our study, we found a downregulation of SPRY1 protein upon neurotrophic factor stimulation, which resulted in augmented activation of their downstream ERK/CREB signaling pathway in cortical neurons. Therefore, our study supports the notion that growth factors are able to positively or negatively control RTK signaling by regulating the expression of their own antagonists, depending on the growth factor involved and the cellular context. Mechanistically, BDNF treatment stimulated miR‐132 expression, which may contribute to the downregulation of SPRY1. However, we did not observe an upregulation of either miR‐132 or miR‐124 when cortical neurons were treated with FGF2. Indeed, c‐Cbl‐mediated proteolytic degradation of hSPRY2 was observed in response to FGF2 and EGF stimulation (Hall et al., 2003) and SPRY1 was also reported to interact with c‐Cbl (Lee et al., 2009); hence, we speculate that, in cortical neurons, FGF2 may repress SPRY1 expression through the ubiquitin‐dependent proteasome pathway via c‐Cbl.

4.2. SPRY1 negatively regulates neurite development

SPRY1 has been reported to control organ development and fundamental biological processes including cell proliferation, differentiation, and survival (Edwin, Anderson, Ying, & Patel, 2009), but so far only a few studies have examined the function of SPRY1 in neuronal development. In the mouse embryo, fine‐tuning of the ERK pathway by SPRY1 and SPRY2 is essential for proper brain morphogenesis (Faedo, Borello, & Rubenstein, 2010). The SPRY1 expression is high in undifferentiated mESCs, where it controls ERK1/2 activity. SPRY1 was dispensable for the maintenance of self‐renewal of mESC and suppression of SPRY1 expression promoted neural differentiation and inhibited endothelial differentiation of mESCs (Jung et al., 2012). Herein, we showed that SPRY1 also regulated neuritogenesis of cortical neurons in the presence of neurotrophic factors: overexpression of SPRY1 inhibited neurite elongation of differentiated immature cortical neurons, whereas knockdown of SPRY1 expression favored neuritogenesis. This observation is consistent with previous studies showing that overexpression of the SPRY protein family member, SPRY4, blocked neurite development and branching induced by nerve growth factor in PC12 cell neuronal differentiation (Alsina et al., 2012) and overexpression of SPRY2 inhibited neurite outgrowth of mouse cerebellar granule neurons (Gross et al., 2007).

Furthermore, we confirmed that SPRY1 was a direct target gene of brain‐enriched miR‐124 and miR‐132. In fact, our recent studies and others demonstrated the relevance of miR‐124 in the process of neuron generation: miR‐124 promoted neurite development of CAD cells derived from CNS catecholaminergic neurons (Makeyev, Zhang, Carrasco, & Maniatis, 2007). In mouse P19 embryonal carcinoma cells, miR‐124 promoted neurite outgrowth during neuronal differentiation (Yu, Chung, Deo, Thompson, & Turner, 2008). Our previous study revealed that miR‐124 promoted neurite elongation of both differentiated BE(2)‐M17 human neuroblastoma cells and mouse primary cortical neurons through repressing ROCK1 expression (Gu et al., 2014). Several studies have also revealed that miR‐132 is essential for neuron cell morphology. In adult hippocampus, miR‐132 stimulated dendritic growth and arborization of newborn neurons (Magill et al., 2010). In mouse dorsal root ganglion cells, miR‐132 promoted axonal extension through locally regulating gene expression within the axon (Hancock, Preitner, Quan, & Flanagan, 2014). Every cellular process, including processes involved in neuronal growth, is the result of a synergistic action of multiple molecules and joint activity of multiple signaling pathways. We found that SPRY1 was regulated by both miRNAs and ERK/MAPK signaling cascades during neuronal differentiation: miR‐124 and miR‐132 repressed SPRY1 expression, whereas ERK stimulated SPRY1 expression. As miR‐124, miR‐132 and ERK signals positively regulate neuronal development (Perron & Bixby, 1999), the increased expression of SPRY1 during neuronal differentiation contributes to the formation of a negative‐feedback loop, which may fine‐tune the right dose of specific signaling required for appropriate neurite growth and branching at different developmental stages.

4.3. SPRY1 regulates neuronal survival under glutamate neurotoxic stress

Cell death induced by excess glutamate release is broadly thought to contribute to neuronal loss associated with both acute and chronic neurodegenerative conditions (Arundine & Tymianski, 2004). Thus, a thorough understanding of glutamate‐dependent signal transduction may have great clinical significance for the treatment of neurodegenerative diseases. Herein, we demonstrated that SPRY1 inhibited neuronal protection mediated by neurotrophic factors under glutamate‐induced neurotoxicity: overexpression of SPRY1 antagonized the neuronal protective function of both BDNF and FGF2. Paradoxically, most studies have shown that SPRY1 can promote cell viability and reduced expression of SPRY1 induced apoptosis. For example, the prosurvival function of serum in adrenal cortex adenocarcinoma (SW13) cells was blocked after silencing of SPRY2 expression (Edwin & Patel, 2008); in human ovarian cancer cells, SPRY1 acts as a negative regulator of cell vitality and survival by inducing expression of a number of proapoptotic proteins (Masoumi‐Moghaddam, Amini, Ehteda, Wei, & Morris, 2014). One explanation for these contradictory results is that SPRY1 may function in a cell context‐dependent manner. As those apoptotic events caused by silencing of SPRY proteins are mostly observed in tumor cells, which rely on EGF to survive, and SPRY proteins in most cases promoted but not antagonized EGF‐mediated RTK cascades. In contrast, we found that in cortical neurons, SPRY1 antagonized BDNF‐ and FGF2‐mediated RTK signaling, which may account for these opposite cytophysiological effects.

The fact that glutamate stress causes the opposite SPRY1 expression changes in different developmental stages is interesting. Immature cultured neurons are more resistant to N‐methyl‐d‐aspartate (NMDA) neurotoxicity (Choi, Maulucci‐Gedde, & Kriegstein, 1987; Marks, Friedman, & Haddad, 1996; Wahl, Schousboe, Honore, & Drejer, 1989). Mizuta, Katayama, Watanabe, Mishina, and Ishii (1998) demonstrated that cortical neurons on Day 11 exhibited marked glutamate sensitivity, but were almost unaffected by glutamate treatment on culture from Days 7–9. Our observations that SPRY1 is downregulated in cultured Day 7 neurons but upregulated in cultured Day 14 neurons upon glutamate exposure and that SPRY1 has a negative role in the survival of glutamate‐treated mature neurons may partially explain the susceptibility of mature neurons to glutamate cytotoxicity. As miR‐124 and miR‐132 expression are inversely related to SPRY1 upon glutamate exposure, it is possible that expression of SPRY1 is under the control of these two miRNAs. miR‐124 and miR‐132 were downregulated in various neurodegenerative conditions (Hwang, Kaneko, Noh, Pontarelli, & Zukin, 2014; Kanagaraj, Beiping, Dheen, & Tay, 2014; Salta, Sierksma, Vanden Eynden, & De Strooper, 2016; Sun et al., 2013) and increased expression of these miRNAs promoted neuronal survival. Therefore, our study indicates that SPRY1 is, at least partially, implicated in the protective effect of miR‐124 and miR‐132 on neurons under pathological conditions.

To date, it is widely accepted that glutamate‐mediated excitotoxicity is the main factor that contributes to several brain pathologies such as ischemia, traumatic brain injury, and Alzheimer's disease. Clinically, it is now clear that blocking of glutamate receptors is not an efficient approach. In fact, glutamate receptor inhibition has recently been reported to influence normal brain function and produced severe adverse side effects (De Keyser, Sulter, & Luiten, 1999). NR2B subunits of NMDA glutamate receptor is the major focus of studies in the treatment of neurological diseases due to their unique characteristics (Hardingham, Fukunaga, & Bading, 2002; Y. Liu et al., 2007). First, the NMDA receptor NR2B subunits are localized predominantly at extrasynaptic sites (Brickley, Misra, Mok, Mishina, & Cull‐Candy, 2003), which enables the receptor to detect glutamate that spill over from synapses during various neuropathological conditions (Gass et al., 2018; S. Li et al., 2011; Rossi, Oshima, & Attwell, 2000). Second, it is reported that the extrasynaptic NMDA receptors are associated with specific signals that cause neurons to die (Hardingham et al., 2002; Y. Liu et al., 2007). However, it is currently believed that the NR2B subunits may also bind to NR1/NR2A subunits and form NR1/NR2A/NR2B complex receptors at synapses (Thomas, Miller, & Westbrook, 2006). Therefore, the selective inhibition of extrasynaptic NMDA receptors by NR2B antagonists remains problematic. Therefore, current studies are focusing on the control of intracellular signaling pathways and their regulatory pathways triggered by glutamate exposure (Pose‐Utrilla et al., 2018; Tu et al., 2010) and our research may provide a possible pharmacological target for this endeavor.

CONFLICTS OF INTEREST

The authors declare that there are no conflicts of interest.