Soluble β-amyloid Precursor Protein Alpha Binds to p75 Neurotrophin Receptor to Promote Neurite Outgrowth

Background The cleavage of β-amyloid precursor protein (APP) generates multiple proteins: Soluble β-amyloid Precursor Protein Alpha (sAPPα), sAPPβ, and amyloid β (Aβ). Previous studies have shown that sAPPα and sAPPβ possess neurotrophic properties, whereas Aβ is neurotoxic. However, the underlying mechanism of the opposing effects of APP fragments remains poorly understood. In this study, we have investigated the mechanism of sAPPα-mediated neurotrophic effects. sAPPα and sAPPβ interact with p75 neurotrophin receptor (p75NTR), and sAPPα promotes neurite outgrowth. Methods and Findings First, we investigated whether APP fragments interact with p75NTR, because full-length APP and Aβ have been shown to interact with p75NTR in vitro. Both sAPPα and sAPPβ were co-immunoprecipitated with p75NTR and co-localized with p75NTR on COS-7 cells. The binding affinity of sAPPα and sAPPβ for p75NTR was confirmed by enzyme-linked immunosorbent assay (ELISA). Next, we investigated the effect of sAPPα on neurite outgrowth in mouse cortical neurons. Neurite outgrowth was promoted by sAPPα, but sAPPα was uneffective in a knockdown of p75NTR. Conclusion We conclude that p75NTR is the receptor for sAPPα to mediate neurotrophic effects.


Introduction
APP, a single transmembrane protein with a long N-terminal extracellular domain and a short cytoplasmic domain, can be processed by two distinct pathways to generate multiple cleaved products [1]. In the primary pathway, a-secretase catalyzes the cleavage of APP to generate a soluble peptide, sAPPa, which includes Ab sequence, thereby preventing Ab generation. In the alternative pathway, b-secretase cleaves APP to generate an alternate soluble peptide, sAPPb, followed by c-secretase to generate Ab.
The start of APP expression occurs when neurons initiate differentiation at embryonic day (E) 9.5 in the mouse brain [2]. In addition, APP cleavage occurs at the embryonic stage [3][4][5][6] as well as injured brain tissue [7][8][9]. These observations suggest that APP fragments may have multiple roles in normal brain development and CNS injury. Indeed, it has been shown that sAPPa possesses neurotrophic effects; for example, it promotes neurite outgrowth in vitro [10] and protects neural tissue after brain injury [8,[11][12][13]. However, the underlying mechanism of its neurotrophic effect remains largely unknown.
p75 NTR mediates a diverse set of functions, including axonal elongation, neuronal survival, and modulation of synaptic transmission [14]. Furthermore, p75 NTR can transmit both positive and negative signals for neuronal action. For example, p75 NTR mediates axonal elongation through binding to neurotrophins, whereas it is also involved in axon growth inhibition through its interactions with the Nogo receptor (NgR) and LINGO co-receptors [14,15]. Regarding APP, p75 NTR has been reported to associate with both full-length APP and Ab [16][17][18]. Indeed, Ab induces cell death via p75 NTR in various types of cells, including neurons [19]. This neurotoxic effect occurs through c-Jun kinase (JNK) and c-Jun [20][21][22]. A recent report further demonstrated that the N-terminal fragment of APP (N-APP) interacts with p75 NTR [18].
In this study, we assessed whether sAPPa and sAPPb will also associate with p75 NTR . We show that sAPPa and sAPPb bind to p75 NTR , and that sAPPa binding stimulates neurite outgrowth. These results indicate that p75 NTR is the receptor for sAPPa to mediate neurotrophic effects.

Mice
All experiments were conducted in accordance with the Osaka University Medical School Guide for the Care and Use of Laboratory Animals, and were approved by the institutional committee of Osaka University (Permit Number: 24-067-005). C57BL/6J mice were purchased from Kiwa Animal Farm (Wakayama, Japan).

Plasmid constructs and small interfering RNA (siRNA)
Mouse sAPPa cDNA was generated by polymerase chain reaction (PCR) using primers constructed from APP valiant 2 (accession No. NM_007471) from a postnatal day (P) 4 mouse spinal cord cDNA library. The cDNA of sAPPa was inserted into a pMD20-T vector (TaKara, Shiga, Japan), and then subcloned into a pcDNA5/FRT vector (Invitrogen, Carlsbad, CA, USA). Amino-terminally Hemagglutinin (HA)-tagged full-length human p75 NTR was subcloned into the pcDNA3 vector (Invitrogen) [23]. Mouse p75 NTR siRNA was designed as described previously [24]. Scrambled siRNA was used as a negative control.

Neurite outgrowth assay
Primary dissociated cultures of cortical neurons were prepared from E16 C57BL/6J mice by using a previously described protocol [25]. Briefly, cortices were dissected and removed, minced into small pieces on ice, and then collected in ice-cold PBS. The cells were then incubated with 0.25% trypsin (Gibco/ Invitrogen, Paisley, UK) and 500 mg/mL DNase1

Nucleofection
Cortical neurons were washed and resuspended in Mouse Neuron Nucleofector Solution (Lonza, Basel, Switzerland) at a final concentration of 5610 6 neurons per 100 mL. The cellnucleofector solution complex (100 mL) and the p75 NTR siRNA or control scrambled siRNA (500 pmol) were then gently mixed and transferred into a cuvette, followed by nucleofection using the nucleofector program O-05. Immediately after electroporation, the cells were mixed with 500 mL of pre-warmed DMEM/F12 containing 10% FBS, followed by transference of the cell suspension into 3.5-cm dishes coated with PLL. After 2 hincubation, the medium was changed to DMEM/F12 containing B27 supplement and penicillin/streptomycin. After 3 days when the expression of p75 NTR was reduced by siRNA, neurons were replated on 3.5-cm dishes coated with PLL at a density of 0.5610 5 neurons/dish in DMEM/F12 containing 10% FBS. After another 2-h incubation, the medium was changed to DMEM/F12 containing B27 supplement, penicillin/streptomycin and 1.22 nM sAPPa or PBS control. The neurons were incubated for 24 h, fixed in 4% PFA and immunostained with polyclonal anti-TuJ1 antibody (1:1000). The lengths of the longest neurites were measured by the ImageJ software.
Co-culture of cortical neurons with Chinese hamster ovary (CHO) cells CHO cells were plated on 3.5-cm dishes coated with PLL at a density of 3610 5 cells/dish in DMEM/F12 containing 10% FBS 24 h before transfection. pcDNA5/FRT vector or sAPPa inserted pcDNA5/FRT vector were transfected into CHO cells. The expression of sAPPa protein was confirmed as described below. At 12 h after transfection, the medium was changed to new DMEM/ F12 containing 10% FBS. At 15 h after transfection, cortical neurons (0.5610 5 cells/dish) were plated on CHO cells. After another 2 h the medium was changed to DMEM/F12 containing B27 supplement and penicillin/streptomycin. At 40 h after coculture, CHO cells and neurons were fixed and immunostained with polyclonal anti Tuj1-antibody (1:1000) and monoclonal anti sAPPa-antibody (3:1000). The lengths of the longest neurites were measured by using the ImageJ software. On the other hand expression of sAPPa was examined by western blotting. At 40 h after co-culture, CHO cells were lysed with lysis buffer (50 mM Tris-HCl, pH 7.8, 150 mM NaCl, 1% NP-40, 2 mM Na 3 VO 4 , 1 mM EDTA). The lysates and the medium of the CHO cells were centrifuged at 13,0006g for 5 min and the supernatants were collected. The supernatant of the CHO cell culture medium was collected and concentrated using centrifugal filter units (Amicon Ultra-0.5 mL 30 K MWCO, Millipore). The supernatant of the lysates and the concentrated supernatant of the medium were boiled in sample buffer for 5 min and subjected to SDS-PAGE. The proteins were transferred onto polyvinylidene difluoride (PVDF) membranes and blocked for 1 h in 5% skim milk. Membranes were blotted overnight with monoclonal anti-sAPPa antibody (3:1000), followed by incubation with HRP-linked secondary antibody. For detection, an ECL chemiluminescence system (GE Healthcare, Little Chalfont, UK) was used.

Statistical analysis
All values are expressed as mean 6 SEM. Tukey-Kramer test followed by Bonferroni/Dunn test was used in growth assay by sAPPa addition. Student's t test was applied in neurite growth assay by the co-culture method. Scheffe's F test was used in neurite growth assay followed by p75 NTR neucleofecton. Kolmogrov-Smirnov test was applied for analysis of distribution of neurite length. P,0.05 was considered statistically significant.

p75 NTR interacts with sAPPa
To assess the possible involvement of p75 NTR in the APP fragments ( Figure 1A) signal transduction pathway, we first examined whether sAPPa interacted with p75 NTR by a pull-down assay. His-tagged sAPPa was incubated with Ni-agarose beads to precipitate any bound protein, and then p75-Fc or IgG-Fc as a control was added. p75 NTR , but not control IgG protein, was detected in sAPPa precipitates ( Figure 1B). Comparable experiments using sAPPb revealed that p75 NTR protein was also detected in sAPPb, and the C-sAPPa had precipitated p75 NTR ( Figure 1C and D). C-sAPPa is the carboxyl-terminal region of sAPPa, corresponding to aa 314-612 of sAPPa (aa 1-612)   Figure 1A). These results indicate that APP fragments interact with p75 NTR .
To examine whether APP fragments bind to p75 NTR on cell surfaces, we performed cell-based binding assays. COS-7 cells were transfected with either empty control vector or HA-tagged p75 NTR inserted vector. After 40 h, cells were fixed and incubated with recombinant protein of His-tagged sAPPa recombinant protein. Bound ligand was immunostained with anti-sAPPa antibody. sAPPa bound to cells expressing p75 NTR but not to cells transfected with control vector ( Figure 1E). We also found that sAPPb and C-sAPPa bound to p75 NTR -expressing cells ( Figure 1F and G). These results suggest that APP fragments bind to p75 NTR on cell surfaces.

Affinity of the sAPPa-p75 NTR interaction
Next, we examined the affinity of the each APP fragments-p75 NTR interactions by ELISA. The recombinant p75 NTR ECD-Fc or IgG-Fc was added to plastic wells coated with one of the APP fragments (sAPPa, sAPPb, or C-sAPPa). The binding was detected by HRP-conjugated anti-human Fc antibody. The interaction between p75 NTR ECD-Fc and sAPPa was higher than that between IgG-Fc and sAPPa (Figure 2A), indicating specific binding between sAPPa and p75 NTR ECD. sAPPb and C-sAPPa also bound to p75 NTR ECD-Fc ( Figure 2C and E). The sigmoid dose-response formulas were used to calculate the EC 50 . sAPPa, sAPPb, and C-sAPPa bound to p75 NTR ECD-Fc, and EC 50 were 90, 120, and 150 nM, respectively ( Figure 2B, D, F). Taken together, our observations indicate that p75 NTR ECD binds to APP peptides, thereby suggesting that p75 NTR is the receptor for APP fragments.

sAPPa promotes neurite outgrowth
It has been reported that sAPPa exerts neuroprotective effects in the traumatic brain injury model [8,[11][12][13]. Therefore, we focused on cortical neurons to examine the effect of sAPPa on neurite outgrowth. For this purpose, we compared the neurite length of sAPPa-treated neurons and control ones. Cortical neurons from E16 mice were treated with IgG-Fc as control or sAPPa at concentrations of 1.22 nM, 2.44 nM, or 4.88 nM, and cultured for 24 h. Neurite outgrowth was enhanced by sAPPa treatment (Figure 3A, B, and S1A).
We further examined neurite length by the co-culture method. In this method, CHO cells were transfected with either empty vector or His-tagged sAPPa inserted vector. We observed that sAPPa protein expression was only detected in sAPPa-transfected CHO cells ( Figure 3C and D). Neurite outgrowth was promoted when the neurons were cultured on sAPPa-expressing CHO cells, compared to those on control CHO cells ( Figure 3E, F, and S1B). These results demonstrate that sAPPa promotes neurite outgrowth in embryonic cortical neurons.

p75 NTR is required for sAPPa-induced neurite outgrowth
The aforementioned results suggest that p75 NTR interacted with APP fragments (Figure 1 and 2). To address whether p75 NTR is a functional receptor for sAPPa, we performed a series of loss-offunction experiments using siRNA for p75 NTR [24]. We first confirmed the knockdown efficacy of p75 NTR siRNA in cortical neurons endogenously expressing p75 NTR . Efficient downregulation of p75 NTR protein was specifically observed in p75 NTR siRNA-transfected cells ( Figure 4A), indicating successful siRNAmediated knockdown of p75 NTR protein. We next examined whether p75 NTR mediated neurite elongation by sAPPa. sAPPa promoted neurite outgrowth of E16 cortical neurons up to 18.7% of control levels. Transfection of p75 NTR siRNA reversed the effect of sAPPa on neurite outgrowth to control levels ( Figure 4B, C, and S1C). These results demonstrate that p75 NTR mediates the promotion of neurite outgrowth by sAPPa.

PKA mediates sAPPa-induced neurite outgrowth
Our previous study demonstrated that neurotrophin binding to p75 NTR promoted neurite outgrowth through cyclic adenosine monophosphate-protein kinase A (cAMP-PKA) [26]. We examined the hypothesis that cAMP-PKA is located downstream of p75 NTR in the signaling pathway mediated by sAPPa. We confirmed that treatment of sAPPa to the culture of cortical neurons significantly enhanced neurite outgrowth in the presense of DMSO, which was used as a solvent control for PKA inhibitor KT5720. By contrast, treatment with KT5720 suppressed the effect of sAPPa on neurite outgrowth ( Figure 5A, B, and S1D). These results demonstrate that PKA activation is essential for sAPPa-induced neurite outgrowth.

Discussion
In this study, we demonstrated that sAPPa binds to p75 NTR and promotes neurite outgrowth. Furthermore, sAPPb also binds to p75 NTR . These results implicate p75 NTR as the receptor for sAPPa in promoting neurite outgrowth.
Although the effect was modest, sAPPa significantly enhanced the neurite outgrowth ( Figure 3A and B). We observed that the treatment of sAPPa increased the number of cells, which had neurites longer than 180 mm ( Figure S1A). These results demonstrated modest but significant effects of sAPPa on neurite elongation. In some cases, excessive neurite elongation may burden on the cells. To promote axon outgrowth, neurons undergo expansion of the plasma membrane [27]. Therefore, rapid neurite outgrowth may result in exhausting cellular biosynthesis. It is possible that sAPPa promotes neurite outgrowth with a lower stress on neurons.
In addition, both fragments possess the region involved in the promotion of neurite outgrowth [28][29][30][31]. These observations suggest that both N-and C-terminal regions of sAPPa contribute to interaction with p75 NTR and the regulation of neurite outgrowth.
We observed that the EC 50 of sAPPa-p75 NTR interaction was lower than that of sAPPb-p75 NTR , indicating that, while sAPPb also binds to p75 NTR , sAPPa binds to p75 NTR with greater affinity. It was reported that sAPPa is more efficient in protecting hippocampal neurons and promoting neurite outgrowth compared to sAPPb [32,33]. These findings suggest that the greater binding affinity of sAPPa-p75 NTR might affect the neuroprotective and neurotrophic function of sAPPa.
In contrast, p75 NTR also functions as a signal transducer of neurite outgrowth inhibition. When myelin-derived proteins bind to the NgR, which lacks an intracellular domain, p75 NTR interacts with NgR to transduce the inhibitory signals intracellularly [39]. Next, p75 NTR facilitates the release of RhoA from Rho-GDPdissociation inhibitor (Rho-GDI), resulting in RhoA activation. The activation of RhoA has a critical role in inducing the inhibition of neurite outgrowth [40]. In this study, we showed that the PKA inhibitor KT5720 inhibited sAPPa-induced neurite outgrowth. These observations lead to our hypothesis that sAPPa also suppresses RhoA activation through p75 NTR . Further studies are required to assess the validity of this hypothesis.
Additionally, APP cleavage occurs during embryogenesis [3][4][5], suggesting that APP fragments are required for embryonic development. In addition, the axons of p75 mutant embryos are disturbed [41]. Based on these findings, sAPPa-p75 NTR signaling may be involved in normal brain development. Furthermore, APP is expressed and cleaved dramatically in CNS injuries, such as spinal cord or traumatic brain injuries [7][8][9]. Therefore, APP cleaved products and the p75 NTR signal may affect the recovery process of neural tissues. Understanding the molecular pathway may assist in the elucidation of novel therapeutic targets for CNS diseases.
In conclusion, we revealed that both sAPPa and sAPPb interact with p75 NTR on COS cells. Knockdown of p75 NTR suppressed the effect of sAPPa. These results support the hypothesis that p75 NTR is the receptor for sAPPa in neurite outgrowth.