Secreted Amyloid Precursor Protein-Alpha Promotes Arc Protein Synthesis in Hippocampal Neurons

Secreted amyloid precursor protein-α (sAPPα) is a neuroprotective and memory-enhancing molecule, however, the mechanisms through which sAPPα promotes these effects are not well understood. Recently, we have shown that sAPPα enhances cell-surface expression of glutamate receptors. Activity-related cytoskeletal-associated protein Arc (Arg3.1) is an immediate early gene capable of modulating long-term potentiation, long-term depression and homeostatic plasticity through regulation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor localization. Accordingly, we hypothesized that sAPPα may enhance synaptic plasticity, in part, by the de novo synthesis of Arc. Using primary cortical and hippocampal neuronal cultures we found that sAPPα (1 nM, 2 h) enhances levels of Arc mRNA and protein. Arc protein levels were increased in both the neuronal somata and dendrites in a Ca2+/calmodulin-dependent protein kinase II-dependent manner. Additionally, dendritic Arc expression was dependent upon activation of mitogen-activated protein kinase and protein kinase G. The enhancement of dendritic Arc protein was significantly reduced by antagonism of N-methyl-D-aspartate (NMDA) and nicotinic acetylcholine (α7nACh) receptors, and fully eliminated by dual application of these antagonists. This effect was further corroborated in area CA1 of acute hippocampal slices. These data suggest sAPPα-regulated plasticity within hippocampal neurons is mediated by cooperation of NMDA and α7nACh receptors to engage a cascade of signal transduction molecules to enhance the transcription and translation of Arc.


INTRODUCTION
Secreted amyloid precursor protein-alpha (sAPPα) is a neuroprotective and neurotrophic protein, derived from the same parent protein as neurotoxic amyloid-ß. The levels of endogenous sAPPα are reduced in neurological disorders, including Alzheimer's disease (AD; Lannfelt et al., 1995;Kim et al., 2009). By contrast, enhancement of sAPPα levels is protective against AD-associated memory impairments (Fol et al., 2016;Tan et al., 2018) and attenuates excitotoxic injury in vivo and in vitro (Mucke et al., 1996;Ryan et al., 2013). Further, sAPPα is able to facilitate long-term potentiation (LTP; Taylor et al., 2008;Moreno et al., 2015), stimulate neurite outgrowth (Clarris et al., 1994), and regulate spine morphology (Hick et al., 2015). Recently, it has been shown that the molecular mechanisms underpinning these actions include enhancement of glutamate receptor trafficking, synaptodendritic protein synthesis and new gene transcription (Claasen et al., 2009;Chasseigneaux et al., 2011;Ryan et al., 2013;Mockett et al., 2019), yet these and other mechanisms have not been fully explored.
Notably, while the receptor/s mediating the actions of sAPPα have not been conclusively identified, candidate receptors and downstream pathways overlap with those identified in relation to the regulation of Arc. Somatodendritic translation of Arc mRNA is dependent on Ca 2+ signaling via ionotropic receptors, including the N-methyl-d-aspartate receptor (NMDA; Chen et al., 2017) and α7 nicotinic acetylcholine receptor (α7nAchR; Kristensen et al., 2007); these receptors are both candidates for mediating sAPPα's actions (Taylor et al., 2008;Richter et al., 2018;Mockett et al., 2019). Furthermore, downstream signaling molecules such as protein kinase G (PKG), mitogen activated protein kinase (MAPK) and CaMKII have not only been shown to enhance Arc mRNA or regulate Arc protein expression (Huang et al., 2007;Gakhar-Koppole et al., 2008;Ota et al., 2010;Chasseigneaux et al., 2011), but also mediates the neuroprotective, neurotrophic and plasticity-enhancing effects of sAPPα (Furukawa et al., 1996a;Claasen et al., 2009;Mockett et al., 2019).
Based on the commonality in pathways regulated by sAPPα and those which enchance Arc expression, we hypothesized that heightening sAPPα levels would upregulate Arc expression. Using primary neuronal cultures, we found that exogenously delivered recombinant sAPPα (1 nM, 2 h) enhanced both Arc mRNA and protein through activation of both NMDA and α7nACh receptors, and that this effect is dependent on the activity of CaMKII, MAPK and PKG.

RESULTS
In order to investigate the expression of the key plasticity protein Arc, we first sought to confirm that DIV24-27 primary neuronal cultures form mature synapses. Consistent with previous literature (Basarsky et al., 1994;Papa et al., 1995;Grabrucker et al., 2009), we found that our cultures coexpress the presynaptic marker synapsin-1 and the postsynaptic AMPA receptor subunit GluA1 on MAP2-positive neurons ( Figure 1A). Co-expression was evident in both somatic and dendritic compartments, as previously observed (Richmond et al., 1996). Additionally, our cultures show populations of GFAP-positive astrocytes closely associated with GluA1-positive neurons ( Figure 1B). This association has been shown to support the development of synapses in vitro (Jones et al., 2012). Further, ultrastructural analysis of our cultured neurons shows evidence of mature synapses ( Figure 1C; Robert et al., 2012).
sAPPα Mediated New Synthesis of Arc Protein: Detected Using FUNCAT-PLA Next, we directly visualized de novo Arc protein synthesis in response to sAPPα using fluorescence non-canonical amino acid tagging with proximity ligation assays (FUNCAT-PLA: Figure 3). Here, we found that sAPPα (1 nM, 2 h) induced a highly significant increase in newly synthesized Arc protein in both the dendrites (2.72 ± 0.41, p ≤ 0.0001; Figures 3C-E) and somata (1.69 ± 0.32, p = 0.03) of cultured hippocampal neurons. Co-incubation with protein synthesis inhibitor anisomycin eliminated labeling (somata, 0.44 ± 0.06, p ≤ 0.0001; dendrites, 0.004 ± 0.004, p ≤ 0.0001; Figures 3B,D,E), confirming the punctate signal was specific for de novo synthesized protein. The relative integrated intensity of (D) somatic and (E) dendritic signal from treatment groups is expressed as drug/average of control and presented as mean ± SEM (n = 25-34 cells from three independent experiments). Co-incubation of AHA with anisomycin inhibited Arc protein synthesis. Incubation with sAPPα significantly increased the expression of de novo synthesized Arc protein in the somata and dendrites. Images show whole cell (scale bar = 50 µm), and magnified somatic (inset, bottom right) and dendritic (50 µm; lower panels; scale bar = 10 µm). Outliers were removed from each experiment prior to amalgamation using Grubb's tests, and normality was detected by D'Agostino and Pearson omnibus normality tests. Significance was calculated on data expressed relative to control by use of one sample t-tests; hashes (#) indicate significance between control and sAPPα-treated, asterisks ( * ) indicate significance between control and anisomycin treated, # p = 0.038, * * * /### p ≤ 0.0001.
Interestingly, a few puncta were found located outside the MAP2 positive cells. These may reflect a low level of background stain or the synthesis of Arc within astorcytes (Rodriguez et al., 2008).

sAPPα Facilitates Arc Protein Expression in Primary Hippocampal Neurons
Using western blot analysis we showed that the anti-Arc primary antibody detected a signal band of the expected molecular weight (Supplementary Figure 1). Next, using this antibody in immunocytochemistry, we showed that application of sAPPα significantly increased Arc protein expression in a concentration dependent manner. Specifically, 0.1 nM sAPPα was sufficient to significantly increase dendritic Arc expression (1.35 ± 0.05, p = 0.0002), although it did not affect somatic Arc protein expression (1.15 ± 0.39, p = 0.24; Figures 4A,B,F,G). Conversely, 1 nM sAPPα significantly increased both somatic (1.35 ± 0.46, p ≤ 0.0001) and dendritic (1.92 ± 0.81, p = 0.0003) Arc protein expression relative to controls (Figures 4A,C,F,G; see Supplementary  Figure 2 for Arc-only gray scale image). Further, using confocal microscopy we were able to show Arc expression to be punctate in dendrites, typical of synapse-associated proteins (Supplementary Figure 3). sAPPβ differs from sAPPα by 16 C-terminal amino acids. sAPPα is derived following cleavage of amyloid precursor protein by a-secretase activity, whereas sAPPβ is liberated by β-secretase 1. sAPPβ has been described as 100-fold less effective in ameliorating excitotoxicity and attenuating glucose deprivation compared to sAPPα (Furukawa et al., 1996b;Turner et al., 2007), but appears to remain unaltered in AD (Sennvik et al., 2000). Accordingly, we next examined the effect of sAPPβ on Arc expression. We found that sAPPβ did not affect Arc protein in dendrites (0.1 nM: 0.97 ± 0.49; 1 nM: 0.92 ± 0.45, p ≥ 0.99, respectively; Figures 4D-G). Interestingly, 0.1 nM sAPPβ did not affect Arc protein expression in the somata (0.91 ± 0.34, p ≥ 0.99), while 1 nM sAPPβ resulted in a small but significant decrease in somatic Arc expression (0.82 ± 0.38, p = 0.03).

sAPPα-Induced Arc Protein Expression Is Dependent on CaMKII, MAPK, and PKG Signaling
Previous research has identified CaMKII, MAPK, and PKG as key downstream kinases regulating sAPPα-mediated protein synthesis in synaptoneurosomes (Claasen et al., 2009). Here we sought to determine whether the sAPPα-induced expression of Arc protein utilizes these same signaling kinases. We observed that the sAPPα-mediated increase in Arc protein in the somata (Figures 7A,B  and PKG is necessary to facilitate sAPPα-induced Arc expression, but that different regulatory mechanisms may exist in somatic and dendritic compartments in primary hippocampal neurons.

sAPPα-Induced Dendritic Arc Protein Expression Is Dependent on Activation of NMDA-and α7nACh Receptors
Despite many studies, the cell surface receptor(s) which transduce the sAPPα signal are yet to be conclusively identified, though specific candidates have emerged (Rice et al., 2017(Rice et al., , 2019Richter et al., 2018;Mockett et al., 2019). Here, we pharmacologically inhibited likely candidates mediating sAPPα's plasticity-promoting effects and observed the effect on sAPPα-induced Arc protein levels (Figure 8). We found that application of antagonists targeting GABA B (CPG55845; 50 µM), TrkB (ANA-12; 100 µM), or mGluRI/II receptors (MCPG; 500 µM) had no significant effect on dendritic Arc protein expression (Figures 8A,B Figures 8E,I). These results suggest a novel mechanism whereby synergistic action between NMDA and α7nAch receptors governs an enhancement in sAPPα-mediated Arc protein expression.

sAPPα Increases CREB Phosphorylation and Arc Protein Levels in Area CA1 of Acute Hippocampal Slices
We next examined whether sAPPα could modulate transcription and Arc expression in acute hippocampal slices. Focusing on area CA1 (Figure 9A), we found that sAPPα (1 nM, 15 min) significantly increased phosphorylated cAMP element binding protein at serine 133 (pCREB) levels, a marker of transcriptional regulation as well as NMDAR signaling (Xia et al., 1996)   followed a modified protocol based on Banker and Goslin (1998) and Kaech and Banker (2006). Cortex and hippocampi from Sprague-Dawley rat pups were dissociated using papain (Sigma) and plated at a low density on glass-bottomed culture dishes (40,000 cells/cm 2 ; Mattek) for immunolabelling, or at 300,000 cells/well of a 6-well plate for lysis and subsequent protein or RNA extraction and RT-qPCR. Cells were cultured in Neurobasal A medium (Life Technologies #10888-022), supplemented with B27 (Life Technologies, #17504-001) and Glutamax (Life Technologies, #35050-061) at 37 • C/5% CO 2 for 24-27 days in vitro (DIV). Control samples were collected at the same time as treated cells in matched sets of culture dishes (immunocytochemistry, FUNCAT-PLA) or wells (RT-qPCR; western blot). All experiments conducted on acute tissue were prepared from young adult male Sprague-Dawley rats (42-56 days), as described previously (Mockett et al., 2019). sAPPα and sAPPβ production and purification was carried out according to the protocols developed by the Tate lab (Turner et al., 2007).
and amplification with Duolink detection reagent Texas Red according to the manufacturer's instructions. Neuronal somata, dendrites and nuclei were visualized by addition of anti-Guinea Pig Alexa Fluor 488 and DAPI, respectively.

Sample Preparation for Transmission Electron Microscopy
Primary hippocampal neurons were grown on 3 mm carboncoated sapphire disks until DIV 25, then fixed using 2% PFA with 2% glutaraldehyde in 0.1 M cacodylate buffer.
Cells were postfixed with 1% osmium tetroxide, followed by 1% uranyl acetate. The samples were then dehydrated using an ethanol gradient, prior to infiltration with EmBed 812 epoxy resin (Electron Microscopy Sciences) using BDMA as an accelerator. Resin-filled BEEM capsules were attached to the cell surface to enable separation of the cell layer from the disk, and the resin was cured at 60 • C for 36 h. Ultrathin sections (90 nm) were then cut by diamond knife using a Leica UC6 Ultramicrotome, and mounted on formvar-coated copper slot grids for viewing the ultrastructure.

Microscopy
FUNCAT-PLA images were captured with a LSM780 confocal microscope (Zeiss), using a 40x/1.4-N.A oil objective (Plan Apochromat DIC M27), a pinhole setting of 90 µm and all lasers at 2% power. The images were acquired as 8-bit mode z stacks, with the 1,024 × 1,024 pixel xy resolution through the thickness of the cell. The optical slice thickness was set to two-times oversampling, and pixel dwell times were set to 0.39-0.80 µs. For experiments examining the expression of protein using immunofluorescence, images were acquired using an Olympus IX71 inverted light microscope using a 20x/0.45-N.A objective (LUCPFLN) or 4x/0.13-N.A objective (UPFLN). The images were captured using a Hamamatsu Orca-AG camera (C4742-80-12AG) in 1024 × 1024 pixel 8-bit mode. Images were saved as.tif files. Electron micrographs were captured using a Philips CM100 BioTWIN transmission electron microscope with a LaB6 emitter (Philips/FEI Corporation), and images were captured using a MegaView III digital camera (Olympus).

Image Analysis
To quantify the FUNCAT-PLA signal a custom-made ImageJ script created by Maximilian Heumüller (Max Planck Institute for Brain Research, Frankfurt) was used (tom Dieck et al., 2015), and punctate PLA signal was subsequently dilated using ImageJ for ease of viewing. To quantify immunofluorescence, neurons were outlined using ImageJ. An 'integrated intensity/neuronal area' value was generated for each cell and somatic compartment, including all dendrites up until intersection with neighboring dendrites. This value was corrected for average background fluorescence by subtracting average background fluorescence. Dendritic fluorescence was determined by subtracting corrected somatic values from whole cell values. Primary and secondary dendrites were analyzed by sampling gray values at 0.5 µm increments, and binning the average of each 50-and 25 µm segments, respectively. Statistical analysis was achieved by averaging each dendritic segment per cell, for a total cell average. For experiments examining pCREB expression in acute hippocampal slices, DAPI was used to define a 'mask' around the nuclear layer and pCREB was measured as 'integrated intensity/DAPI area.' For experiments examining Arc expression in acute hippocampal slices, area CA1 was defined by a square area adjacent to the hippocampal fissure (encompassing both the pyramidal cell layer and stratum radiatum) and an integrated intensity/area' value was generated for each slice.

RT-qPCR
As the data exhibited a normal distribution (Shapiro-Wilk normality test), significance was assessed using Students t-tests where P < 0.05 was accepted as significantly different. Data were normalized to the control gene HPRT and expressed relative to no-drug control.

Imunocytochemistry
Statistics for all immunocytochemistry experiments were performed using Kruskal-Wallis one-way ANOVA with Dunn's multiple comparisons test. Data were not normally distributed (D'Agostino and Pearson omnibus normality test). Outliers within the raw data sets were detected using the Grubb's test.

FUNCAT-PLA
Prior to data amalgamation, outliers were removed from each experiment using Grubb's tests, and normality was assessed (D'Agostino and Pearson omnibus normality tests). Significance was calculated on data expressed relative to control by use of one sample t-tests.

Acute Hippocampal Slices
Data for all slice work exhibited a normal distribution (Shapiro-Wilk normality test). Experiments examining pCREB expression were analyzed by use of students t-tests, and experiments analyzing Arc protein expression were analyzed using one-way ANOVA followed by Šidák's multiple comparisons test. Data are expressed as fold change relative to control values.

DISCUSSION
Here, we show that sAPPα promotes the expression of the key plasticity protein Arc in primary neuronal cultures. Notably, Arc expression was dependent on CaMKII, MAPK, and PKG activity, which is consistent with previous reports of these kinases being essential for sAPPα's facilitation of local protein synthesis in hippocampal synaptoneurosomes and glutamate receptor trafficking in acute hippocampal slices (Claasen et al., 2009;Mockett et al., 2019). Additionally, we have identified a critical interaction between NMDA and α7nAch receptors in triggering this effect.
Our research provides evidence of a coordinated set of signaling mechanisms through which sAPPα regulates the expression of Arc, and by association, synaptic plasticity. Similar to the well-studied neurotrophin brain-derived neurotrophic factor (BDNF), nanomolar amounts of sAPPα promotes a gradual increase in Arc and likewise an increase in Zif268 mRNA. However, the relative change induced by sAPPα appears to be more modest than that of BDNF (El-Sayed et al., 2011), suggesting a neuromodulatory role for sAPPα. Further, we described a concentration-dependent effect in response to sAPPα, but not sAPPβ. While previous reports indicate that sAPPα is able to rescue morphological and plasticity-related deficits observed in APP and APP-like protein 2 (APPL2) knockout mice (Ring et al., 2007;Fol et al., 2016) and conditional APP/APPL2 NexCre knockdown mice Hick et al., 2015), sAPPβ is unable to ameliorate these deficits. This difference may underlie a divergence in signaling cascades between the two sAPP metabolites, further elucidating their distinct biological functions.
While previous work has described a multitude of signaling pathways mediating sAPPα-induced neuroprotection, and plasticity, of particular interest is sAPPα's ability to increase cGMP and MAPK activity (Furukawa et al., 1996a;Gakhar-Koppole et al., 2008). Strikingly, in previous studies using isolated synapses, we have shown that sAPPα enhances de novo protein synthesis in a manner partially dependent on CaMKII and MAPK, and fully depending on PKG (Claasen et al., 2009). Likewise, trafficking of GluA1-containing AMPA receptors following sAPPα-enhanced LTP requires CaMKII and PKG, as well as protein synthesis (Mockett et al., 2019). Interestingly, a recently published study (Martinsson et al., 2019) has shown that knockdown or knockout of endogenous APP leads to increased levels of GluA1. Given that Arc is known to internalize GluA1-and/or GluA2-containing AMPA receptors Rial Verde et al., 2006), this suggests that endogenous APP or sAPPα may contribute to regulation of Arc levels (and therefore surface expression of GluA1). Building on this, and as primary hippocampal neurons express GluA2-lacking AMPA receptors (Jaafari et al., 2012), we propose that the signaling mechanisms harnessed by sAPPα not only promote increased AMPA receptor cell surface expression, but also initiate a concomitant homeostatic response whereby Arc promotes delayed AMPA receptor internalization. Future experiments should aim to explore whether this occurs globally, or specifically at activated synapses.
In the present experiments, inhibition of CaMKII significantly impaired both somatic and dendritic Arc expression, while inhibition of MAPK and PKG significantly reduced Arc expression in the dendrites alone. Thus CaMKII, MAPK, and PKG may mediate distinct aspects of sAPPα-induced functions in each neuronal compartment. Indeed, evidence suggests that PKG may act to facilitate trafficking of Rab11-positive vesicles. Rab11 is a protein primarily associated with recycling endosomes, and can mediate anterograde trafficking from the trans-Golgi network and perinuclear endosome (Chen et al., 1998;Ang et al., 2004;Lock and Stow, 2005;Takahashi et al., 2012). This trafficking is achieved through close association with the Ca 2+ -sensitive motor protein myosin Vb (Wang et al., 2008). Interestingly, PKG has been linked to the nitric oxide-dependent simulation of anterograde trafficking of Rab11A-positive recycling endosomes (Zhai et al., 2017) and has been shown to directly bind Rab11B (Reger et al., 2014). Importantly, Arc protein and mRNA colocalize with Rab11 (Wu et al., 2011) and inhibition of Rab11 activity impairs postsynaptic expression of Arc protein in Drosophila motor neurons (Ashley et al., 2018). Thus, PKG may play an important role in the activity-dependent transport of Arc-containing vesicles throughout the dendrites. Once present at the synapse, PKG, in concert with MAPK and CaMKII, may then regulate synthesis of Arc protein (Browning et al., 2000;Kleppisch et al., 2003;Atkins et al., 2004;Claasen et al., 2009;Michel et al., 2011;Mockett et al., 2011). CaMKII may also contribute to the localization of Arc protein within the somata and dendrites (Husi et al., 2000;Donai et al., 2003;Okuno et al., 2012), however, our studies do not distinguish between CaMKII isoforms. Further, it has been hypothesized that there may exist separate pools of translating and nontranslating Arc mRNA (Steward et al., 2014) such that a pool of Arc mRNA may be translated within the somata, while a separate, translationally repressed pool is trafficked throughout the dendrites primed and waiting to be translated locally (Link et al., 1995). A third pool may govern rapid transcriptionindependent translation of Arc mRNA already associated with polyribosomes throughout the dendrites (Bagni et al., 2000;Na et al., 2016). Therefore, our data may reflect the presence of distinct pools of Arc mRNA, regulated differentially by CaMKII, MAPK, and PKG. Additionally, recent work has identified a novel role for Arc protein through shared properties with retroviral Gag proteins: Arc forms virus-like capsids capable of encapsulating and transporting functional RNA and protein between cells in endosomal-derived extracellular vesicles (Pastuzyn et al., 2018), therefore suggesting Arc can modulate synapse-, cell-, and network-wide plasticity.
A crucial, yet outstanding question in the literature is the identity of the 'sAPPα receptor.' Previous evidence has described a role of GABA B (Rice et al., 2017(Rice et al., , 2019, Na/K ATP'ase (Dorard et al., 2018), α7nAch (Richter et al., 2018) and NMDA receptors (Gakhar-Koppole et al., 2008;Mockett et al., 2019) in mediating sAPPα's plasticity enhancing effects. To understand the receptors contributing to the facilitated Arc expression, we investigated the possible contributions of these and other candidate receptors (mGluRI/II and TrkB) due to their similar neuromodulatory or plasticity-promoting properties (Raymond et al., 2000;Mockett et al., 2011). Our data suggest a synergistic effect between activation of NMDARs and α7nAChRs. Sole inhibition of either receptor alone led to a partial impairment in the sAPPα-mediated expression of Arc protein, whereas simultaneous inhibition of both receptors fully eliminated the enhancement of Arc protein expression in the dendrites. Further, we have extended our studies to the more complex biological system of acute hippocampal slices. Here, our findings closely reflected that of the primary hippocampal culture work, as sAPPα significantly enhanced Arc protein expression in area CA1 of the hippocampus in a manner dependent on activation of NMDA and α7nACh receptors. This finding was corroborated by an increase in pCREB, a downstream NMDAR signaling event (Xia et al., 1996), in area CA1. Importantly, many of the reports describing sAPPα's ability to enhance LTP has been shown in CA1, including the dependence on NMDAR (Mockett et al., 2019) and α7nACh receptors (Richter et al., 2018). The mechanism governing this effect may involve proteinprotein interactions between α7nAChR-NMDAR complexes (Li et al., 2012(Li et al., , 2013, or direct binding of sAPPα to NMDA or α7nAChR (Cousins et al., 2009;Innocent et al., 2012;Forest et al., 2018). Likewise, trafficking of NMDA receptors has been observed in response to sAPPα (Mockett et al., 2019) and may occur through α7nAChR activation (Marchi et al., 2015). This therefore may be a mechanism which provides additive Ca 2+ or activation of downstream signaling pathways (Aramakis and Metherate, 1998;Brunzell et al., 2003;Gould and Lewis, 2005). A similar synergistic effect has been found between the activation of NMDARs and Gs-coupled dopamine and β-adrenergic receptors in the PKA-dependent increase in Arc protein (Bloomer et al., 2008). PKG-dependent NMDAR activation has been shown to enhance the expression of IEG Arc, c-Fos and Zif268 in the lateral amygdala in vivo in response to both fear conditioning (Ota et al., 2010) and LTP-inducing stimulation (Ping and Schafe, 2010) and inhibition of neuronal nitric oxide (nNOS), soluble guanylyl cyclase (sGC), or PKG attenuated the bicuculline-induced expression of Arc, c-Fos and Zif268 in neuronal cultures (Gallo and Iadecola, 2011). Further, both NMDAR and α7nAch can facilitate CaMKII and MAPK activation (Cammarota et al., 2000;Thalhammer et al., 2006;Gubbins et al., 2010). Therefore, concerted activity between NMDARs and α7nAChRs may act to promote sAPPα-mediated Arc expression through synergistic activation of downstream cascades involving CaMKII, PKG and MAPK. sAPPα shows hallmark neuroprotective and neurotrophic properties and is capable of stimulating neurite outgrowth, regulating spine morphology (Clarris et al., 1994;Hick et al., 2015), enhancing synaptodendritic protein synthesis, and facilitating protein synthesis-dependent LTP (Taylor et al., 2008;Claasen et al., 2009;Moreno et al., 2015;Mockett et al., 2019), as well as protecting against excitotoxic injury in vitro and in vivo (Mucke et al., 1996;Ryan et al., 2013). Likewise, the synthesis and function of Arc protein has been previously linked to many of these characteristics, such as stimulation of CaMKIIdependent neurite extension (Donai et al., 2003), the homeostatic intergration of synapse downscaling following seizure activity and chronic excitation (Lyford et al., 1995;Shepherd et al., 2006;Gao et al., 2010) with increased morphological plasticity and spine size (Messaoudi et al., 2007;Peebles et al., 2010), and its role in facilitating LTP (Guzowski et al., 2000;Plath et al., 2006;Messaoudi et al., 2007;Nakayama et al., 2016). Our studies provide insight into how sAPPα promotes the expression of these mechanisms through the utilization of Arc protein. It is important to note that we have titrated sAPPa concentrations and found that 1 nM is sufficient to initiate a variety of molecular events (Claasen et al., 2009;Ryan et al., 2013;Mockett et al., 2019). There is, however, only limited knowledge of the physiological levels of sAPPα and to what extent they are dynamically regulated in vivo.

CONCLUSION
This work extends the current understanding of possible mechanisms through which sAPPα enhances synaptic plasticity.
Here we have described an enhancement of Arc protein synthesis driven by activation of NMDA and α7nAch receptors, and downstream activation of CaMKII, MAPK, and PKG. sAPPα levels are reduced in many disease states and neurological disorders, many of which involve impaired or altered plasticity (Lannfelt et al., 1995;Kim et al., 2009;Selnes et al., 2010;Jakobsson et al., 2013;Miyajima et al., 2013;Rolstad et al., 2015). Therefore, furthering our understanding of the underlying processes governing sAPPα's regulation of key plasticity proteins, including Arc, is a fundamental step in understanding dysfunction which arises in many cognitive diseases.

DATA AVAILABILITY
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

AUTHOR CONTRIBUTIONS
RL was the major contributor to the experimental aspects of the study, contributed to the design of the study, prepared the primary neuronal cultures, isolated RNA, carried out the RT-qPCR and immunochemistry experiments, analyzed and interpreted the corresponding data, and drafted the manuscript. ME carried out the FUNCAT-PLA study and the corresponding data analysis and critically revised the manuscript. MB and CW carried out immunochemistry experiments, analyzed and interpreted the corresponding data. KP produced recombinant sAPPα, analyzed and interpreted the corresponding data. WT designed the protocols for recombinant sAPPα production, analyzed and interpreted the corresponding data, critically reviewed the manuscript. WA critically revised the manuscript. JW conceived and participated in the design and co-ordination of the study, undertook data analysis and interpretation, and critically assessed the manuscript.

FUNDING
Research reported in this article was supported by the Health Research Council of New Zealand, the University of Otago (Department of Anatomy, and postgraduate scholarships to RL and ME) and a German Academic Exchange Service scholarship to ME. We acknowledge the support and mentorship of Prof. E. Schuman (Max Planck Institute for Brain Research).