Distinct in vivo roles of secreted APP ectodomain variants APP sα and APP sβ in regulation of spine density, synaptic plasticity, and cognition
2018; Springer Nature; Volume: 37; Issue: 11 Linguagem: Inglês
10.15252/embj.201798335
ISSN1460-2075
AutoresMax Richter, Susann Ludewig, Alex Winschel, Tobias Abel, Charlotte Bold, Leonie Salzburger, Susanne Klein, Kang Han, Sascha W. Weyer, Ann‐Kristina Fritz, Bodo Laube, David P Wolfer, Christian J. Buchholz, Martin Körte, Ulrike Müller,
Tópico(s)Neuroscience and Neuropharmacology Research
ResumoArticle16 April 2018free access Source DataTransparent process Distinct in vivo roles of secreted APP ectodomain variants APPsα and APPsβ in regulation of spine density, synaptic plasticity, and cognition Max C Richter Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Susann Ludewig Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany Search for more papers by this author Alex Winschel Department of Biology, Neurophysiology und Neurosensory Systems, TU Darmstadt, Darmstadt, Germany Search for more papers by this author Tobias Abel Paul-Ehrlich-Institut (PEI), Langen, Germany Search for more papers by this author Charlotte Bold Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Leonie R Salzburger Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany Search for more papers by this author Susanne Klein Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Kang Han Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Sascha W Weyer Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Ann-Kristina Fritz Institute of Anatomy, University of Zurich, Zurich, Switzerland Institute of Human Movements Sciences and Sport, ETH Zurich, Zurich, Switzerland Search for more papers by this author Bodo Laube Department of Biology, Neurophysiology und Neurosensory Systems, TU Darmstadt, Darmstadt, Germany Search for more papers by this author David P Wolfer Institute of Anatomy, University of Zurich, Zurich, Switzerland Institute of Human Movements Sciences and Sport, ETH Zurich, Zurich, Switzerland Search for more papers by this author Christian J Buchholz orcid.org/0000-0002-9837-7345 Paul-Ehrlich-Institut (PEI), Langen, Germany Search for more papers by this author Martin Korte orcid.org/0000-0001-6956-5913 Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany Helmholtz Centre for Infection Research, AG NIND, Braunschweig, Germany Search for more papers by this author Ulrike C Müller Corresponding Author [email protected] orcid.org/0000-0002-0587-2575 Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Max C Richter Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Susann Ludewig Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany Search for more papers by this author Alex Winschel Department of Biology, Neurophysiology und Neurosensory Systems, TU Darmstadt, Darmstadt, Germany Search for more papers by this author Tobias Abel Paul-Ehrlich-Institut (PEI), Langen, Germany Search for more papers by this author Charlotte Bold Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Leonie R Salzburger Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany Search for more papers by this author Susanne Klein Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Kang Han Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Sascha W Weyer Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Ann-Kristina Fritz Institute of Anatomy, University of Zurich, Zurich, Switzerland Institute of Human Movements Sciences and Sport, ETH Zurich, Zurich, Switzerland Search for more papers by this author Bodo Laube Department of Biology, Neurophysiology und Neurosensory Systems, TU Darmstadt, Darmstadt, Germany Search for more papers by this author David P Wolfer Institute of Anatomy, University of Zurich, Zurich, Switzerland Institute of Human Movements Sciences and Sport, ETH Zurich, Zurich, Switzerland Search for more papers by this author Christian J Buchholz orcid.org/0000-0002-9837-7345 Paul-Ehrlich-Institut (PEI), Langen, Germany Search for more papers by this author Martin Korte orcid.org/0000-0001-6956-5913 Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany Helmholtz Centre for Infection Research, AG NIND, Braunschweig, Germany Search for more papers by this author Ulrike C Müller Corresponding Author [email protected] orcid.org/0000-0002-0587-2575 Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany Search for more papers by this author Author Information Max C Richter1,‡, Susann Ludewig2,‡, Alex Winschel3, Tobias Abel4, Charlotte Bold1, Leonie R Salzburger2, Susanne Klein1, Kang Han1, Sascha W Weyer1, Ann-Kristina Fritz5,6, Bodo Laube3, David P Wolfer5,6, Christian J Buchholz4, Martin Korte2,7 and Ulrike C Müller *,1 1Institute of Pharmacy and Molecular Biotechnology (IPMB), Ruprecht-Karls University Heidelberg, Heidelberg, Germany 2Zoological Institute, Division of Cellular Neurobiology, TU Braunschweig, Braunschweig, Germany 3Department of Biology, Neurophysiology und Neurosensory Systems, TU Darmstadt, Darmstadt, Germany 4Paul-Ehrlich-Institut (PEI), Langen, Germany 5Institute of Anatomy, University of Zurich, Zurich, Switzerland 6Institute of Human Movements Sciences and Sport, ETH Zurich, Zurich, Switzerland 7Helmholtz Centre for Infection Research, AG NIND, Braunschweig, Germany ‡These authors contributed equally to this work *Corresponding author. Tel: +49 6221 546717; E-mail: [email protected] EMBO J (2018)37:e98335https://doi.org/10.15252/embj.201798335 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Increasing evidence suggests that synaptic functions of the amyloid precursor protein (APP), which is key to Alzheimer pathogenesis, may be carried out by its secreted ectodomain (APPs). The specific roles of APPsα and APPsβ fragments, generated by non-amyloidogenic or amyloidogenic APP processing, respectively, remain however unclear. Here, we expressed APPsα or APPsβ in the adult brain of conditional double knockout mice (cDKO) lacking APP and the related APLP2. APPsα efficiently rescued deficits in spine density, synaptic plasticity (LTP and PPF), and spatial reference memory of cDKO mice. In contrast, APPsβ failed to show any detectable effects on synaptic plasticity and spine density. The C-terminal 16 amino acids of APPsα (lacking in APPsβ) proved sufficient to facilitate LTP in a mechanism that depends on functional nicotinic α7-nAChRs. Further, APPsα showed high-affinity, allosteric potentiation of heterologously expressed α7-nAChRs in oocytes. Collectively, we identified α7-nAChRs as a crucial physiological receptor specific for APPsα and show distinct in vivo roles for APPsα versus APPsβ. This implies that reduced levels of APPsα that might occur during Alzheimer pathogenesis cannot be compensated by APPsβ. Synopsis Increasing evidence suggests that the synaptic functions of the amyloid precursor protein (APP), that is key to Alzheimer (AD) pathogenesis, may be carried out by its secreted ectodomain. Here, using AAV mediated intracranial expression, we studied the specific in vivo roles of APPs fragments generated by non-amyloidogenic or amyloidogenic APP processing. APPsα, but not APPsβ, efficiently rescued deficits in spine density, synaptic plasticity and spatial memory of mice lacking both APP and APLP2 (cDKO). The C-terminal 16 amino acids of APPsα (lacking in APPsβ) facilitate LTP to the same extent as APPsα. This activity of APPsα involves the α7-nAChR as a crucial physiological receptor specific for APPsα. APPsα potentiates α7-nAChRs expressed in oocytes and functions as an allosteric positive modulator Reduced levels of APPsα that may occur during Alzheimer pathogenesis cannot be compensated by APPsβ. Introduction Alzheimer's disease (AD) is characterized by the accumulation of β-amyloid peptides (Aβ) that are derived from the amyloid precursor protein (APP) by proteolytic cleavage (Selkoe & Hardy, 2016). Two principal physiological pathways either prevent or promote Aβ generation. Within the amyloidogenic pathway, APP processing is shifted towards the production of Aβ and secreted APPsβ, by consecutive β-secretase (BACE1) and γ-secretase cleavage (Vassar et al, 2014). In AD, BACE1 is upregulated, favoring amyloidogenic APP processing (Holsinger et al, 2002; Ahmed et al, 2010). In the alternative, non-amyloidogenic pathway cleavage of APP within the Aβ region by the major α-secretase ADAM10 (a disintegrin and metalloprotease) prevents Aβ generation and liberates APPsα that is secreted into the extracellular space (Saftig & Lichtenthaler, 2015). Shifting APP processing towards non-amyloidogenic processing has therefore been suggested as a therapeutic strategy for AD (Mockett et al, 2017). An important and still unresolved question is whether in addition to neurotoxic Aβ accumulation a concomitant reduction in APPsα level, or an altered APPsα/APPsβ ratio may contribute to AD symptoms and pathogenesis (reviewed by Mockett et al, 2017). In this regard, it will be crucial to know whether APPsα and APPsβ, that is only 16 amino acids shorter than APPsα, serve largely similar or distinct, possibly even opposite physiological functions. While our previous studies and work from others indicated that APPsα has neurotrophic and neuroprotective effects, including synaptogenic, LTP facilitating and memory enhancing properties (Meziane et al, 1998; Ring et al, 2007; Taylor et al, 2008; Milosch et al, 2014; Weyer et al, 2014; Hick et al, 2015; Hefter et al, 2016; Plummer et al, 2016), only few and mostly conflicting studies have as yet addressed the functions of APPsβ (Nikolaev et al, 2009; Li et al, 2010; Weyer et al, 2011; Chasseigneaux & Allinquant, 2012). So far, the molecular basis of any difference compared to APPsα remained unclear. Thus, a more detailed knowledge about the specific functions of the secreted APP ectodomains is essential to understand AD pathogenesis and evaluate risks of ongoing pharmacotherapy, as well as to elucidate APP physiology. There is a large body of evidence indicating that APP family proteins are multimodal proteins that can function as ligands via their secreted fragments or as cell surface proteins important for synaptic adhesion and signal transduction (Müller et al, 2017). Major insights into the physiological functions of APP and the related APLPs (APP like proteins) were obtained from knockout models (Müller et al, 2017). While most impairments of APP-KO mice emerged only in aged mice (Dawson et al, 1999; Seabrook et al, 1999; Ring et al, 2007; Lee et al, 2010; Tyan et al, 2012), combined APP/APLP2 double knockout mice die shortly after birth, likely due to severe deficits at the neuromuscular junction (von Koch et al, 1997; Heber et al, 2000; Wang et al, 2005; Klevanski et al, 2014). Recently generated forebrain-specific double knockout mice (termed NexCre cDKO), that lack APP from embryonic day 11.5 onwards in excitatory forebrain neurons on a global constitutive APLP2-KO background, showed a severe synaptic phenotype already at young age, including reduced spine density and impaired LTP in the hippocampus, as well as deficits in learning and memory (Hick et al, 2015). Interestingly, the LTP impairment of NexCre cDKO mice could be ameliorated by acute APPsα application onto brain slices in vitro, while the molecular mechanism and receptor(s) mediating its function remained unknown (Hick et al, 2015). More recently, we showed that AAV-mediated intracranial expression of APPsα can mitigate the Aβ-related synaptic deficits of APP/PS1 mice in vivo (Fol et al, 2016). Intracranial AAV-APPsα injections enhanced spine density, improved LTP deficits and memory, but at the same time also reduced soluble Aβ levels and plaque load, likely due to enhanced Aβ clearance (Fol et al, 2016). In addition, APPsα had been reported to lower Aβ by directly binding to and inhibiting BACE (Obregon et al, 2012). These results raised the question of whether the beneficial in vivo effects of APPsα are mainly due to its Aβ lowering properties. Here, we asked whether APPsα may also have synaptotrophic effects in an Aβ-independent pathology with synaptic impairments and used viral vectors to express APPsα intracranially in NexCre cDKO mice. Moreover, we set out to compare side by side the properties of APPsα and APPsβ in vivo. We demonstrate that APPsα is sufficient to fully rescue hippocampal spine density, to restore LTP and partially rescue spatial memory in adult NexCre cDKO mice. In sharp contrast, despite similar expression level, APPsβ failed to show any detectable effects on synaptic plasticity and spine density. Finally, we show that the CTα16 domain of APPsα (that is lacking in APPsβ) is able to facilitate LTP to the same extent as APPsα, in a mechanism that involves functional nicotinic α7 acetylcholine receptors (α7-nAChRs). Moreover, we show that nanomolar concentrations of APPsα (but not APPsβ) can directly potentiate α7-nAChRs-mediated currents upon heterologous expression in Xenopus oocytes and increases the apparent agonist affinity as a positive allosteric modulator. Collectively, our analysis identifies the α7-nAChR as a crucial physiological receptor for APPsα and reveals distinct in vivo roles of APPsα vs. APPsβ. Results AAV-APPsα injection mediates efficient and neuron-specific expression of APPsα in the hippocampus of cDKO mice To investigate whether APPsα is able to rescue the synaptic deficits of NexCre cDKO (further referred to as cDKO) animals in vivo, APPsα was expressed in the adult brain using stereotactic injection. We employed an AAV9-based bicistronic vector (AAV-APPsα; Fig 1A) coding for Venus and codon-optimized HA-tagged murine HA-APPsα that was inserted behind the APP signal peptide (SP). The HA-APPsα reading frame was fused to Venus by a T2A site to enable tracking of transduced cells. Expression was driven by the neuronal synapsin promoter. Monocistronic AAV-Venus vector served as a control (AAV-Venus; Fig 1A). Adult cDKO animals (aged 4–5 months) were either injected with AAV-APPsα or AAV-Venus control vector, whereas littermate control mice (LM control) received only AAV-Venus. Thus, comparison of AAV-Venus-injected cDKO mice with AAV-Venus-injected LM controls was expected to yield similar synaptic impairments as previously observed for uninjected cDKO mice (Hick et al, 2015). Vectors were bilaterally injected (dose: 1.0 × 109 gc/μl per injection spot) into the stratum lacunosum-moleculare region of the dorsal hippocampus and into the dentate gyrus (Fig 1B). To evaluate Venus and APPsα expression, animals were sacrificed 6 weeks post-injection and brain samples were analyzed by Western blot and immunohistochemistry using an HA-tag-specific antibody. Analysis of serial anteroposterior coronal brain sections (Bregma −1.10 to Bregma −2.70; see Fig EV1C) revealed widespread expression of APPsα in the hippocampus and, to a considerably lower extent, also in adjacent cortical areas (Figs 1C and EV1C). Along the longitudinal (dorsal-ventral) axis of the hippocampus, APPsα expression was much more pronounced in the dorsal region, whereas expression was not detectable in the ventral hippocampus (Fig EV1C). Figure 1. Expression of APPsα in the hippocampus of AAV-APPsα-injected cDKO animals A. Schematic representation of monocistronic and bicistronic AAV constructs enabling neuron-specific expression of (i) Venus and (ii) HA-tagged APPsα (+Venus). ITR: inverted terminal repeat, Synapsin: neuron-specific promotor, T2A: Thoseaasigna virus 2A site, SP: signal peptide, HA: influenza hemagglutinin tag. B. Scheme of the hippocampus with coordinates of the two injection sites (black stars). C. Overview of the hippocampus of an AAV-APPsα-injected cDKO mouse. HA-tag staining (red) reveals APPsα expression within the CA1, CA3, and DG of the hippocampus. Magnification shows the boxed CA1 region. Scale bars: 500 μm (left), 100 μm (right). DG: Dentate Gyrus, CA: Cornu Ammonis. D–G. Double immunostaining in CA1 region. APPsα (HA-tag, red) is exclusively expressed in neurons (NeuN, blue, D, E), but not in astrocytes (GFAP, blue, F) or microglia (Iba1, blue, G). Scale bars: 10 μm (D, F, G), 5 μm (E). H. Western blot analysis of APP expression in hippocampus of LM control (N = 5) and cDKO mice (N = 4) injected with AAV-Venus or AAV-APPsα (N = 5). Age of mice: 5–6 months. Source data are available online for this figure. Source Data for Figure 1H [embj201798335-sup-0003-SDataFig1.pdf] Download figure Download PowerPoint Click here to expand this figure. Figure EV1. AVV-Venus and AVV-APPs vectors mediate comparable neuron-specific expression throughout the dorsal hippocampus Double immunostaining of CA1 pyramidal cells. APPsβ (HA-tag, red) is exclusively expressed in neurons (NeuN, blue, left), but not in astrocytes (GFAP, blue, middle) or microglia (Iba1, blue, right). Scale bar: 20 μm. Expression was investigated by immunohistochemistry in different coronal sections from dorsal to ventral. The expression of Venus, APPsα, and APPsβ (HA-tag) is shown for different sections of the hippocampus. Expression is absent in ventral portions of the hippocampus. Scale bar: 1 mm. Download figure Download PowerPoint As Venus contained a membrane anchor, we observed prominent accumulation in dendritic regions, whereas HA-APPsα staining was intense in intracellular membrane compartments consistent with the transport of APPsα within the secretory pathway to the cell surface for secretion (see also Fig EV2D and E for co-localization with ER and Golgi markers). In the CA3 region, APPsα expression appeared slightly lower compared to the expression obtained in pyramidal cells of the subiculum, the CA1 (Fig 1C) and the CA2 regions. Consistent with previous studies (Jackson et al, 2016), we noted that APPsα was detectable in HEK cells transfected with synapsin promoter driven AAV constructs (see Fig EV2A), likely due to AAV plasmid overload. In vivo, however, AAV-APPsα expression was restricted to neurons, as shown by double immunofluorescence staining against the HA-tag and the neuronal marker NeuN (Fig 1D and E). Consistently, no overlapping expression pattern was detectable in astrocytes (GFAP; Fig 1F) and microglia (Iba1; Fig 1G). The AAV-Venus expression pattern obtained from injections of control vector was largely similar to that of AAV-APPsα (Fig EV1C). Western blot analysis of hippocampal homogenates, using the M3.2 antibody, confirmed efficient AAV-mediated APPsα expression in injected animals (Fig 1H). Note that antibody M3.2 is directed against an epitope located between the α- and β-secretase site and recognizes endogenous APP species (APP full length and APPsα) that are still expressed by interneurons and glia in cDKO mice (see Hick et al, 2015) and vector-derived AAV-APPsα. Collectively, our data demonstrate that AAV-APPsα injection results in efficient and neuron-specific expression of APPsα throughout the dorsal hippocampus. Click here to expand this figure. Figure EV2. APPsα and APPsβ are transported through the secretory pathway and secreted efficiently A. Western blot analysis of lysate (L) and supernatant (SN) of pAAV-APPsα and pAAV-APPsβ-transfected HEK293T cells using an HA-tag-specific antibody. Note the secretion of HA-APPsα and HA-APPsβ into the cell supernatant. Mature APPs is posttranslationally modified and runs at slightly higher MW. B. Western blot analysis of medium from cortical neurons transduced with AAV-APPsα or AAV-APPsβ. 105 cDKO neurons were transduced at DIV7 with 104 genome copies per cell of either AAV-APPsα or AAV-APPsβ. After 12 days, medium was exchanged and neuronal supernatants were harvested on the next day. Equal amounts (2 μl) of supernatant were subjected to capillary electrophoresis followed by immunodetection (WES system, ProteinSimple, USA). Note that similar amounts of HA-APPsα and HA-APPsβ are secreted into the medium of transduced neurons. C–E. Immunocytochemistry of neurons transduced with AAV-APPsα or AAV-APPsβ. Transduced neurons positive for the neuronal marker NeuN (red) were identified by Venus expression (green). (D, E) Confocal images of neurons stained with the ER marker Calreticulin (green), the Golgi marker Giantin (green), and α-HA to detect APPs (red). Note that both HA-APPsα and HA-APPsβ show co-localization (white arrows) with the ER (D) and Golgi markers (E) indicating successful transport through the secretory pathway. Scale bars: 100 μm (C), 10 μm (D, E). Download figure Download PowerPoint Impaired synaptic plasticity and reduced spine density of APP/APLP2 cDKO mice are rescued by APPsα expression in the adult brain Having demonstrated that HA-APPsα is efficiently expressed in the hippocampus of injected animals, we evaluated whether APPsα is sufficient to rescue impairments in functional network activity that were previously reported for cDKO mice (Hick et al, 2015). After 20 min of baseline recording, we induced long-term potentiation (LTP) at the Schaffer collateral to CA1 pathway by application of theta burst stimulation (TBS) in acute hippocampal slices of mice that had been injected with viral vectors 6 weeks earlier (at 4–5 months of age). Consistent with our previous results (Hick et al, 2015), AAV-Venus-injected cDKO mice exhibited significantly lower induction and maintenance of LTP (n = 25 slices), as compared to AAV-Venus-injected LM controls (n = 23; Fig 2A). AAV-Venus-injected LM control mice showed a potentiation of 156.69 ± 4.75% (t75–t80 after start of baseline recording) that was significantly reduced to only 128.12 ± 3.41% in AAV-Venus-injected cDKO mice (one-way ANOVA followed by Bonferroni's post hoc test, ###P < 0.001; Fig 2B). In contrast, the LTP curve recorded from acute slices of AAV-APPsα-injected cDKO mice (n = 24) closely overlapped with that of AAV-Venus-injected LM controls (Fig 2A). Quantification of the last 5 min of recording revealed that APPsα rescued LTP deficits of cDKO mice to a level statistically indistinguishable from that of Venus-injected LM controls (t75–80: 150.34 ± 3.55%, nsP > 0.05; Fig 2B). While basal synaptic transmission was comparable in all groups (see Fig EV3A and B), presynaptic function evaluated by paired pulse facilitation (PPF; Fig 2C) was significantly impaired in AAV-Venus-injected cDKO mice at the 10 ms (#P < 0.05) and 20 ms (#P < 0.05) inter-stimulus intervals (ISI) in comparison with AAV-Venus-injected LM controls. Strikingly, AAV-APPsα treatment leads to a highly significant rescue of short-term synaptic plasticity in cDKO mice as evidenced by PPF values statistically indistinguishable from LM controls (Fig 2C). Figure 2. AAV-APPsα rescues LTP, impaired short-term synaptic plasticity, and spine density of cDKO mice A, B. LTP was induced at hippocampal CA3-CA1 synapses after 20 min baseline recordings (arrowhead, TBS). cDKO mice expressing Venus (red) exhibited significant lower induction and maintenance of LTP (128.12 ± 3.41%) compared to Venus-injected LM controls (white, 156.69 ± 4.75%, ###P < 0.001). AAV-mediated expression of APPsα (green) restored potentiation after start of baseline recording and resulted in a LTP curve comparable to that of LM controls. The LTP induction rate is shown as percentage % of mean baseline slope. Data points were averaged over six time points. n = number of slices. N = number of animals. C. The deficit of PPF in Venus-injected cDKO mice at the 10 ms (#P < 0.05) and 20 ms (#P < 0.05) ISI was restored by expression of APPsα. n = number of slices. N = number of animals. D. Representative images of basal and midapical dendritic segments of CA1 neurons. Scale bar: 5 μm. E, F. The spine density deficit of Venus-injected cDKOs mice (11% in basal and 18% in midapical dendrites) is rescued by APPsα to LM control levels. Images are maximum projections of deconvolved z-stacks. Spine density was normalized to LM control levels. n = number of neurons (from five animals per condition). Data information: Age of mice at analysis: 5–6 months. Data represent mean ± SEM. Data were analyzed by one-way ANOVA followed by Bonferroni's post hoc test. # indicates significant differences between LM control and cDKO injected with AAV-Venus, * between cDKO injected with AAV-Venus or APPsα. nsP > 0.05, */#P < 0.05, **P < 0.01, ***/###P < 0.001. Download figure Download PowerPoint Click here to expand this figure. Figure EV3. Basal transmission in cDKO mice treated with APPsα or APPsβ is not altered compared to controls A–D. (A, C) Neuronal excitability is comparable at all stimulus intensities (25–250 μA) between all groups injected with the indicated AAV vectors. (B, D) No significant alterations can be detected when analyzing the input–output (IO) strength of the viral vector injected groups of mice at any fiber volley (FV) amplitude. n = number of recorded slices, N = number of animals. Data represent mean ± SEM. Download figure Download PowerPoint Next, we evaluated spine density as a correlate of excitatory synapses. To visualize neuronal morphology and spine density, we performed iontophoretic postfixation filling with a fluorescent dye of hippocampal CA1 pyramidal cells in brain slices from injected adult mice (aged 5–6 months). Consistent with our previous study (Hick et al, 2015), we confirmed a prominent reduction in spine density both in basal (100.00 ± 3.41 vs. 88.77 ± 3.10 spines/μm, #P < 0.05) and in midapical dendrites (100.00 ± 3.57 vs. 82.08 ± 4.73 spines/μm, #P < 0.05) of AAV-Venus-injected cDKO CA1 neurons as compared to AAV-Venus-injected LM controls (Fig 2D–F). In contrast, spine density of AAV-APPsα-injected cDKO mice did not significantly differ from that of AAV-Venus-injected LM control neurons, indicating that APPsα expression fully restored spine density in CA1 neurons. Collectively, these data demonstrate that acute expression of APPsα in adult mice is sufficient to rescue morphological and functional synaptic deficits in cDKO animals. APPsα ameliorates dendritic branching abnormalities of cDKO animals Prompted by the effect of APPsα on spine density, we further evaluated its influence on the overall morphology and complexity of hippocampal CA1 neurons (Fig 3). In view of their different morphology and connectivity, basal and apical dendrites of CA1 neurons were studied separately. Neurons of AAV-Venus-injected cDKO mice showed a significantly reduced total dendritic length and branching (as assessed by the number of nodes) in both basal and apical dendrites compared to AAV-Venus-injected LM controls (Fig 3A–F). Impairments were readily apparent when visually inspecting reconstructed images of individual neurons (Fig 3A). Figure 3. APPsα ameliorates dendritic branching abnormalities of cDKO animals A. Representative 3D-reconstructions of CA1 pyramidal neurons from AAV-Venus-injected LM controls (left), AAV-Venus-injected cDKOs (middle, red), and AAV-APPsα-injected cDKOs (right, green). B. Schematic representation of parameters assessed. C–F. Compared to LM controls, AAV-Venus-injected cDKO mice show a significantly reduced basal (##P < 0.01, C) and apical (##P < 0.01, E) dendritic length and reduced branching in basal (##P < 0.01, D) and apical (##P < 0.01, F) dendrites. AAV-APPsα injection did neither affect basal (nsP > 0.05, C) nor apical (nsP > 0.05, E) total dendritic length. However, the total number of basal nodes differed significantly compared to AAV-Venus-injected cDKO mice (*P < 0.05, D). G. Sholl analysis of basal dendritic length reveals a significant group effect (repeated measures ANOVA: genotype F(2, 29) = 5.038, P < 0.05) and a significant distance effect (repeated measures ANOVA: genotype F(5, 145) = 250.2, P < 0.0001). Due to a significant interaction effect (repeated measures ANOVA: genotype F(10, 145) = 4.466, P < 0.0001), a post hoc Sidak's multiple comparison test was performed to further evaluate effects between groups at distinct distances from soma. Compared to Venus-injected, AAV-APPsα injection of cDKO significantly increased basal dendritic length at 60 μm (*P < 0.05). H. Sholl analysis of apical dendritic length reveals an overall significant group effect (repeated measures ANOVA: genotype F(2, 37) = 4.776, P < 0.05) and a significant distance effect (repeated measures ANOVA: genotype F(18, 666) = 47.54, P < 0.0001). Due to a significant interaction effect (repeated measures ANOVA: genotype F(36, 666) = 1.640, P < 0.05), a post hoc Sidak's multiple comparison test was performed to further evaluate effects between groups at distinct distances from soma. AAV-APPsα expression in
Referência(s)