Title: APP and APLP2 are essential at PNS and CNS synapses for transmission, spatial learning and LTP
Abstract: Article26 April 2011free access APP and APLP2 are essential at PNS and CNS synapses for transmission, spatial learning and LTP Sascha W Weyer Sascha W Weyer Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Maja Klevanski Maja Klevanski Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Andrea Delekate Andrea Delekate Cellular Neurobiology, Zoological Institute, TU Braunschweig, Braunschweig, Germany Search for more papers by this author Vootele Voikar Vootele Voikar Institute of Anatomy and Zurich Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland Search for more papers by this author Dorothee Aydin Dorothee Aydin Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Meike Hick Meike Hick Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Mikhail Filippov Mikhail Filippov Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Natalia Drost Natalia Drost Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Kristin L Schaller Kristin L Schaller Department of Cell and Developmental Biology, Institute of Physiology and Biophysics, University of Colorado School of Medicine, Aurora, CO, USA Search for more papers by this author Martina Saar Martina Saar Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Miriam A Vogt Miriam A Vogt Department for Psychiatry and Psychotherapy, Central Institute of Mental Health, Mannheim, Germany Search for more papers by this author Peter Gass Peter Gass Department of Biology, Institute of Human Movement Sciences, ETH Zurich, Zurich, Switzerland Search for more papers by this author Ayan Samanta Ayan Samanta Department of Pharmaceutical Chemistry, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Andres Jäschke Andres Jäschke Department of Pharmaceutical Chemistry, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Martin Korte Martin Korte Cellular Neurobiology, Zoological Institute, TU Braunschweig, Braunschweig, Germany Search for more papers by this author David P Wolfer David P Wolfer Institute of Anatomy and Zurich Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland Department of Biology, Institute of Human Movement Sciences, ETH Zurich, Zurich, Switzerland Search for more papers by this author John H Caldwell John H Caldwell Department of Cell and Developmental Biology, Institute of Physiology and Biophysics, University of Colorado School of Medicine, Aurora, CO, USA Search for more papers by this author Ulrike C Müller Corresponding Author Ulrike C Müller Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Sascha W Weyer Sascha W Weyer Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Maja Klevanski Maja Klevanski Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Andrea Delekate Andrea Delekate Cellular Neurobiology, Zoological Institute, TU Braunschweig, Braunschweig, Germany Search for more papers by this author Vootele Voikar Vootele Voikar Institute of Anatomy and Zurich Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland Search for more papers by this author Dorothee Aydin Dorothee Aydin Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Meike Hick Meike Hick Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Mikhail Filippov Mikhail Filippov Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Natalia Drost Natalia Drost Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Kristin L Schaller Kristin L Schaller Department of Cell and Developmental Biology, Institute of Physiology and Biophysics, University of Colorado School of Medicine, Aurora, CO, USA Search for more papers by this author Martina Saar Martina Saar Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Miriam A Vogt Miriam A Vogt Department for Psychiatry and Psychotherapy, Central Institute of Mental Health, Mannheim, Germany Search for more papers by this author Peter Gass Peter Gass Department of Biology, Institute of Human Movement Sciences, ETH Zurich, Zurich, Switzerland Search for more papers by this author Ayan Samanta Ayan Samanta Department of Pharmaceutical Chemistry, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Andres Jäschke Andres Jäschke Department of Pharmaceutical Chemistry, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Martin Korte Martin Korte Cellular Neurobiology, Zoological Institute, TU Braunschweig, Braunschweig, Germany Search for more papers by this author David P Wolfer David P Wolfer Institute of Anatomy and Zurich Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland Department of Biology, Institute of Human Movement Sciences, ETH Zurich, Zurich, Switzerland Search for more papers by this author John H Caldwell John H Caldwell Department of Cell and Developmental Biology, Institute of Physiology and Biophysics, University of Colorado School of Medicine, Aurora, CO, USA Search for more papers by this author Ulrike C Müller Corresponding Author Ulrike C Müller Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany Search for more papers by this author Author Information Sascha W Weyer1,‡, Maja Klevanski1,‡, Andrea Delekate2,‡, Vootele Voikar3, Dorothee Aydin1, Meike Hick1, Mikhail Filippov1, Natalia Drost1, Kristin L Schaller4, Martina Saar1, Miriam A Vogt5, Peter Gass6, Ayan Samanta7, Andres Jäschke7, Martin Korte2, David P Wolfer3,6, John H Caldwell4 and Ulrike C Müller 1 1Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany 2Cellular Neurobiology, Zoological Institute, TU Braunschweig, Braunschweig, Germany 3Institute of Anatomy and Zurich Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland 4Department of Cell and Developmental Biology, Institute of Physiology and Biophysics, University of Colorado School of Medicine, Aurora, CO, USA 5Department for Psychiatry and Psychotherapy, Central Institute of Mental Health, Mannheim, Germany 6Department of Biology, Institute of Human Movement Sciences, ETH Zurich, Zurich, Switzerland 7Department of Pharmaceutical Chemistry, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Heidelberg, Germany ‡These authors contributed equally to this work *Corresponding author. Department of Bioinformatics and Functional Genomics, Institute of Pharmacy and Molecular Biotechnology, Heidelberg University, Im Neuenheimer Feld 364, Heidelberg 69120, Germany. Tel.: +49 622 154 6717; Fax: +49 662 154 5830; E-mail: [email protected] The EMBO Journal (2011)30:2266-2280https://doi.org/10.1038/emboj.2011.119 Correction(s) for this article APP and APLP2 are essential at PNS and CNS synapses for transmission, spatial learning and LTP17 May 2011 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 Despite its key role in Alzheimer pathogenesis, the physiological function(s) of the amyloid precursor protein (APP) and its proteolytic fragments are still poorly understood. Previously, we generated APPsα knock-in (KI) mice expressing solely the secreted ectodomain APPsα. Here, we generated double mutants (APPsα-DM) by crossing APPsα-KI mice onto an APLP2-deficient background and show that APPsα rescues the postnatal lethality of the majority of APP/APLP2 double knockout mice. Surviving APPsα-DM mice exhibited impaired neuromuscular transmission, with reductions in quantal content, readily releasable pool, and ability to sustain vesicle release that resulted in muscular weakness. We show that these defects may be due to loss of an APP/Mint2/Munc18 complex. Moreover, APPsα-DM muscle showed fragmented post-synaptic specializations, suggesting impaired postnatal synaptic maturation and/or maintenance. Despite normal CNS morphology and unaltered basal synaptic transmission, young APPsα-DM mice already showed pronounced hippocampal dysfunction, impaired spatial learning and a deficit in LTP that could be rescued by GABAA receptor inhibition. Collectively, our data show that APLP2 and APP are synergistically required to mediate neuromuscular transmission, spatial learning and synaptic plasticity. Introduction Synaptic dysfunction, cognitive decline, and deposition of the β-amyloid peptide Aβ, derived by proteolytic processing from the amyloid precursor protein (APP), are hallmark features of Alzheimer's disease (AD). Despite biochemical and genetic evidence that put Aβ as a central trigger for AD pathogenesis, the physiological role of APP and the question of whether a loss of its functions contributes to AD are still unclear. Interestingly, the neuroprotective secreted ectodomain APPsα is reduced in both familial and sporadic AD (e.g. Lannfelt et al, 1995; and references in Ring et al (2007)) and loss of APPsα-mediated functions may thus contribute to AD pathogenesis. Thus, it is essential to elucidate the in vivo function(s) of APP and its various proteolytic fragments, including their roles for synapse formation and function, as well as learning and memory. APP processing is initiated by α-secretase cleavage within the Aβ region or by β-secretase (BACE) cleavage at the N-terminus of Aβ, leading to the secretion of large soluble ectodomains, termed APPsα and APPsβ, respectively. Subsequent processing by γ-secretase generates Aβ and the APP intracellular domain (AICD). APP belongs to a gene family including the amyloid precursor-like proteins, APLP1 and APLP2, in mammals (Anliker and Müller, 2006). Although APLP1 and APLP2 lack the Aβ region, they are similarly processed by α-, β-, and γ-secretases (Walsh et al, 2007). APP and APLP2 are highly expressed in neurons and peripheral tissue including skeletal muscle, are axonally transported (Szodorai et al, 2009), and have been localized to peripheral neuromuscular junction (NMJ) and central synapses. Our analysis of APP knockout (KO) mice revealed deficits in body and brain weight, a deficit in grip strength and in aged mice also impairments in learning and memory associated with a deficit in LTP (Ring et al, 2007). Combined knockouts indicated functional complementation within the APP gene family and revealed a key physiological role of APLP2. Indeed, APP−/−APLP1−/− mice are viable, whereas APP−/−APLP2−/− and APLP1−/−APLP2−/− mice die within 24 h after birth, likely due to impaired neuromuscular transmission (Heber et al, 2000; Herms et al, 2004; Wang et al, 2005). Whereas brain morphology of newborn double mutants was normal, triple KO mice showed cortical neuronal ectopias, indicating a role of the APP family for neuronal positioning (Herms et al, 2004). At the NMJ, APP and APLP2 have a redundant and essential role for synapse formation and function, as APP/APLP2 double knockout (DKO) mice exhibit widening of the endplate band and severely impaired neurotransmission (Wang et al, 2005). More recently, a study with selective gene knockout either in nerve or in muscle suggested a requirement of APP/APLP2 at both pre- and post-synaptic sites (Wang et al, 2009). Nonetheless, a detailed understanding of the specific function(s) of APP family members at the synapse and the role of the various proteolytic fragments remain unclear. Knockout of the C. elegans orthologue APL-1 disrupts molting and morphogenesis and results in larval lethality. Interestingly, this lethality could be rescued by neuronal expression of the secreted extracellular domain (Hornsten et al, 2007). To dissect domains important for APP function in mice, we previously generated APPsα knock-in (KI) mice expressing only secreted APPsα. When compared with completely APP-deficient animals, APPsα-KI mice displayed a wild-type-like phenotype suggesting that APPsα (similar to secreted APL-1) is sufficient to rescue all abnormalities previously observed in adult APP-KO animals (Ring et al, 2007). Surprisingly, however, recently generated APPsβ-KI mice proved unable to rescue the perinatal lethality of APP/APLP2-DKO mice (Li et al, 2010), raising the question whether different secreted APP fragments may perform distinct functions. Until now, the perinatal lethality of APP/APLP2-DKO precluded the analysis of APP/APLP2-mediated functions in the adult nervous system. To test whether APPsα, as opposed to APPsβ, may rescue the lethality and neuromuscular deficits of APP/APLP2-DKO mice, we crossed APPsα-KI mice onto an APLP2-deficient background. The majority of these APPsα-DM mice survive into adulthood revealing a complex phenotype with deficits both in synaptic transmission in the adult PNS and CNS and in learning and memory. Results Viable APPsα-DM mice show muscular weakness and deficits in challenging neuromotor tasks In APPsα-KI mice, we inserted a stop codon behind the α-secretase cleavage site of the endogenous APP locus. Thus, APPsα-KI mice express only secreted APPsα (Supplementary Figure S1A). To test whether the apparent absence of phenotypic abnormalities in APPsα mice is due to functional complementation by APLP2, we crossed APPsα-KI mice with APLP2-deficient mice. Resulting APPsα-DM double mutant mice were born at normal Mendelian frequencies (χ2 F(2)=1.4056 ns). Remarkably, about half of these combined APPsα-DM mutants survived into adulthood (Supplementary Figure S1B) showing a reduction in body weight (Supplementary Figure S1C). Thus, expression of APPsα was sufficient for rescuing to a large extent the lethality of APP/APLP2-DKO mice that die shortly after birth. This viability allowed us to assess postnatal functions mediated by APP and APLP2. We focused our analysis on the NMJ, hippocampal formation, and behavioural tests of learning and memory. Previously, we had demonstrated a deficit in grip strength in adult APP-KO mice (Ring et al, 2007), whereas both APPsα-KI mice and APLP2-KO mice were normal. On the accelerating rotarod, that assesses motor coordination and muscle fatigue, APPsα-DM mice performed poorly, falling off much earlier than APLP2-KO controls (Figure 1A showing female mice and Supplementary Figure S8A for animals of both sexes). In addition, APPsα-DM mice exhibited pronounced muscular weakness in the grip strength test (Figure 1B; Supplementary Figure S8B). Whereas all control mice were able to hang for at least 60 s from an inverted cage lid, APPsα-DM mice fell off immediately or after a few seconds (Figure 1C; Supplementary Figure S8C). As a baseline for subsequent cognitive tests, we also assessed activity in the home cage and the open field (Supplementary Figure S2A–C). Basal activity in the home cage was increased (Supplementary Figure S2A). In the novel open field arena overall activity, velocity and acceleration were not reduced in APPsα-DM mice (Supplementary Figure S2B and C). These data indicate that in the absence of APLP2, APPsα is not sufficient to mediate motor function needed for sustained high contraction forces, while basal locomotion is not impaired. Figure 1.APPsα-DM mice show muscular weakness. (A) Rotarod testing. Time to fall during trials 1–5 of the accelerating rotarod test, max. time 300 s. APPsα-DM mice showed variable performance and were overall impaired (genotype F(1,20)=17.0 P<0.0005, trial F(4,80)=3.6 P<0.0092, trial × genotype F(4,80)=0.5 ns). APPsα-DM n=10, APLP2-KO n=14, all female. (B) Grip test. Average grip force during two 5-trial testing sessions. APPsα-DM mice were strongly impaired (genotype F(1,20)=168.6 P<0.0001, session F(1,20)=8.5 P<0.0086, session × genotype F(1,20)=2.2 ns). APPsα-DM n=10, APLP2-KO n=14, all female. (C) Cage lid hang test. APPsα-DM mice were strongly impaired (Mann–Whitney genotype U=140 P<0.0001). APPsα-DM n=10, APLP2-KO n=14, all female. Download figure Download PowerPoint Impaired neuromuscular transmission: reduction in quantal content, readily releasable pool, and ability to sustain vesicle release To examine whether the muscle weakness of APPsα-DM mice is a consequence of impaired NMJ function, we first studied spontaneous synaptic transmission. Miniature endplate potential (MEPP) frequency in both APLP2-KO and APPsα-DM muscle fibres was in the normal range of 0.5–1/s (Figure 2A). However, the distribution of frequencies was sharply altered in the APPsα-DM fibres with a high proportion of fibres showing very low frequency responses and some fibres with higher than normal MEPP frequencies (3–5/s; Figure 2B). Overall, this resulted in a significant decrease of mean MEPP frequency in double mutants (Figure 2A). Mean MEPP amplitudes, however, were increased by about 29% in APPsα-DM mice (Figure 2C), possibly due to post-synaptic changes that were not investigated in more detail. Evoked responses were recorded to study a potential defect in transmitter release. Single action potentials in the pre-synaptic axon produce the coordinated exocytosis of 50–100 synaptic vesicles (termed quantal content). We observed a dramatic reduction (by 45%) in quantal content in APPsα-DM muscles (29±1.7) compared with APLP2-KO littermate control muscles (52.9±3.3; Figure 2D; Supplementary Figure S3C). Short-term synaptic plasticity, assessed by paired-pulse-facilitation (PPF), was not significantly different (Supplementary Figure S3A and B). Figure 2.APPsα-DM mice show reduced quantal content, RRP and ability to sustain vesicle release. Spontaneous (A–C) and evoked (D–H) vesicle release recorded with intracellular recording in mouse diaphragm muscle fibres. (A) MEPP frequencies in APLP2-KO (black bar; freq=0.78±0.05/s, mean±s.e.m.; 103 fibres, 9 muscles) and APPsα-DM (red bar; freq=0.55±0.07/s, s.e.m.; 155 fibres, 15 muscles) were significantly different (**P=0.01, t-test). (B) Distribution of MEPP frequency from fibres in panel (A). Note the large increase in fibres with low spontaneous release of vesicles in APPsα-DM. (C) MEPP amplitudes were significantly larger in APPsα-DM (1.39±0.04 mV, s.e.m.) than in APLP2-KO fibres (1.08±0.04 mV, s.e.m.; ***P<10−6, t-test). Same fibres and muscles as panel (A). (D) Quantal content (QC) during evoked release was over 80% larger in APLP2-KO fibres (QC=52.9±3.3, s.e.m.; 12 fibres, 3 muscles) than in APPsα-DM fibres (QC=29±1.7, s.e.m.; 45 fibres, 3 muscles; ***P<10−7). (E) Readily releasable pool (RRP) size in APPsα-DM was less than a third compared with APLP2-KO fibres. Mean RRP values for APLP2-KO (12 fibres, 3 muscles; RRP=1004±109, s.e.m.) and APPsα-DM (24 fibres, 3 muscles; RRP=298±31, s.e.m.; ***P<10−8). (F) Probability of release in APPsα-DM was double that of APLP2-KO fibres. Same synapses as panel (E). Calculated as the ratio of the quantal content of the first response of a train divided by the RRP of that synapse. APLP2-KO: 0.05±0.005; APPsα-DM: 0.1±0.007; ***P<10−4. (G, H) Decrease in evoked post-synaptic response during a 10 s, 20 Hz train of action potentials in the phrenic nerve. Responses during the first and last second are shown (stimulation continued during the 8 s gap). Arrowhead: start of stimulus, filled circle: end of stimulus. (H) Endplate potentials were measured at 1, 2, 4, 6, 8, and 10 s by averaging groups of five responses occurring during a 200-ms window. Amplitudes were normalized for APLP2-KO (n=12 fibres, 3 muscles) and APPsα-DM (n=25 fibres, 3 muscles). Error bars (s.e.m.) are smaller than the symbol for some data points. Differences at each time point (1–10 s) were statistically significant (P<10−3). Download figure Download PowerPoint The reduced quantal content could be due to a smaller readily releasable pool (RRP) and/or a smaller probability of release. The APLP2-KO muscle had a greater initial quantal content (Figure 2D) and released ∼2.5-fold more vesicles during the 2 s stimulation (1610 versus 652; arrowheads in Supplementary Figure S3C). The RRP (determined according to Elmqvist and Quastel (1965)) in APPsα-DM muscles (298±31) was about 30% of that of the APLP2-KO muscles (1004±109; Figure 2E). Surprisingly, the probability of release from motoneurons (calculated as quantal content of the first response divided by the RRP) in the APPsα-DM muscles was twice as large as in the APLP2-KO muscles (0.1 versus 0.05; Figure 2F). Thus, the smaller quantal content in the APPsα-DM muscles was solely due to the reduction in RRP. Since APPsα-DM mice are unable to maintain their grip upside down on a wire mesh (Figure 1C) for longer than 5–10 s, we stimulated the phrenic nerve for 10 s with either 20 or 40 Hz trains of action potentials (Figure 2G and H; Supplementary Figure S3D). Within the first second of stimulation, EPP size (normalized to the first response) in APPsα-DM muscles was ∼20% smaller than in APLP2-KO muscles. A large deficit (15%) in APPsα-DM muscles was also observed for 40 Hz stimulation (Supplementary Figure S3D and E). Overall, these data suggest that neuromuscular weakness of APPsα-DM mice is likely due to both a reduction in quantal content, as well as impaired ability to maintain sustained transmitter release. Of note, this pre-synaptic defect in transmitter release seems to be specific for peripheral cholinergic neuromuscular synapses since no alterations in basal synaptic transmission or evoked pre-synaptic stimulation were detectable for glutamatergic CA3/CA1 excitatory synapses within the hippocampus (see Supplementary Figure S11B and C). Mint2 binding to APP provides a link to Munc18-1 Our data suggest that holo-APP/APLP2 may be required for organizing the molecular complex for exocytosis and/or in positioning the release ready vesicles at the NMJ (Neher and Sakaba, 2008). In this regard, it is striking that Mint/X11 family members are known to bind the cytoplasmic domain of APP. Neuronal Mint1, 2 (X11α, β) both bind Munc18-1, which in turn binds the SNARE complex (Rogelj et al, 2006). Thus, we wondered whether members of the Mint/X11 family might link APP and Munc18-1. To test this hypothesis, we co-transfected HEK293 cells with APP, Mint2 and Munc18-1 and performed immunoprecipitations followed by western blotting (Figure 3A). Indeed, both Mint2 and APP could be co-immunoprecipitated with Munc18-1. Figure 3.Interactions of APP with Mint2 and Munc18. (A) Immunoprecipitation and immunoblotting of HEK293 cells co-transfected with APP, Mint2, and HA-tagged Munc18-1. Cell lysates were immunoprecipitated with an anti-HA antibody (IP: Munc18, anti-HA) and probed with anti-APP (C1/6.1), anti-Mint or anti-Munc18 antibodies. IP with rabbit IgG served as a negative control. Lysate of non-transfected cells reveals endogenous APP. (B) Model of tripartite complex of APP-VC, Mint2 and Munc18-1-YN. (C) Visualization of the APP/Mint2/Munc18-1 complex by BiFC. Cos7 cells were co-transfected with YN-HA–Munc18 (all panels), mycAPP-VC (upper and middle panel) or mycAPPΔCT-VC (lower panel). In addition, co-transfections contained either Flag–Mint2 (upper and lower panel) or were mock transfected with pcDNA (middle panel). Expression was visualized by immunocytochemistry using anti-myc (for APP), anti-HA (for Munc18-1), and anti-mint antibodies. BiFC signal was imaged 24 h after transfection (scale bar: 10 μm). Download figure Download PowerPoint To explore the interaction of APP/Mint2/Munc18-1 in living cells, we employed the BiFC assay (Kerppola, 2008) by introducing APP-VC and Munc18-YN fusion proteins either alone or together with Mint2 (Figure 3B and C). As expected, expression of APP-VC together with Munc18-YN did not lead to an appreciable fluorescent BiFC signal in co-transfected cells. In contrast, significant fluorescence was detectable in triple-transfected cells expressing Mint2 in addition to APP-VC and Munc18-YN, suggesting that Mint2 serves as a scaffold that brings APP and Munc18-1 in close proximity. Importantly, formation of this tripartite complex critically depends on the APP C-terminus, as APPΔCT-VC constructs lacking the YENPTY interaction motive for Mint2, failed to yield fluorescence, despite readily detectable expression (as shown by immunocytochemistry) of transfected proteins (Figure 3C, middle). APPsα-DM mice exhibit a widened endplate band To assess whether deficits in synaptic transmission are associated with abnormal NMJ morphology, as seen in newborn DKO mutants (Wang et al, 2005), we studied the diaphragm from young adults (4 and 8 weeks of age). In APPsα-DM mice endplates visualized by bungarotoxin (BTX)-rhodamine staining appeared scattered and distributed over a much wider muscle territory, which was reflected by an increase in endplate band area of about 50% compared with APLP2-KO (Figure 4A–C). For quantification, the lateral distance and density of individual endplates from the medial tendon insertion was measured. When fitted with a Gaussian, the half-maximal width of the distribution in APPsα-DM muscle (450±20 μm) was significantly increased by about 1.3-fold compared with muscles of APLP2-KO controls (346±27 μm, P<0.05, t-test; Figure 4C). Widening of the endplate band was paralleled by a pronounced increase in secondary nerve branching in APPsα-DM mice (Figure 4D). Although the precise branching pattern varies between individual mice, abnormalities were obvious in all mutants (data not shown). Figure 4.Widening of the endplate band in APPsα-DM mice. (A) Whole mount BTX staining of diaphragm. (B) Muscle area covered by AChR-rich synapses (height of microscopic field=2485 μm). Values represent mean±s.e.m.; mean number of synapses per mouse: APLP2-KO n=575; APPsα-DM n=752, n=3 mice/genotype *P<0.05, t-test (age: 2 months). (C) % Synapses within seven stripes covering the synapse band. When fitted with a Gaussian the full width at half maximum (FWHM) of the synapse distribution is significantly increased in APPsα-DM (P<0.05; t-test). (D) Neurofilament staining indicates increased branching of the phrenic nerve in APPsα-DM (age: 1 month). Scale bars: (A)=250 μm and (D)=150 μm. Download figure Download PowerPoint APP/APLPs are necessary for NMJ synapse maturation and maintenance Closer examination of synapses revealed several striking abnormalities in APPsα-DM mutants: (1) reduced size of pre- and post-synaptic specializations, (2) defective apposition of pre- and post-synaptic elements, and (3) abnormal topology of post-synaptic structures. In APPsα-DM mice, both the area covered by AChRs and the area of synaptophysin-immunoreactive pre-synaptic specializations were considerably reduced (Figure 5A–C). In addition, APPsα-DM revealed reduced coverage of the pre-synaptic marker synaptophysin with post-synaptic AChRs (Figure 5D). Interestingly, whereas all post-synaptic sites were innervated, we found in APPsα-DM animals several isolated synaptophysin-positive patches indicating pre-synaptic specializations that lacked any BTX staining (Figure 5E), suggesting either postnatal nerve terminal sprouting or more likely, persistence of pre-synaptic structures at sites at which post-synaptic specializations had been lost. Figure 5.APPsα-DM mice show aberrant synaptic morphology and abnormal synaptic maturation. (A) In APPsα-DM synapses frequently show a fragmented structure with multiple discrete non-connected structures, APLP2-KO synapses were pretzel shaped. AChRs labelled by BTX (red), pre-synaptic terminals by anti-synaptop