Cover ArticleResearch Articles, Cellular/Molecular

APP Genetic Deficiency Alters Intracellular Ca2+ Homeostasis and Delays Axonal Degeneration in Dorsal Root Ganglion Sensory Neurons

Andrés de León, Julien Gibon and Philip A. Barker
Journal of Neuroscience 31 August 2022, 42 (35) 6680-6691; https://doi.org/10.1523/JNEUROSCI.0162-22.2022

Abstract

The activation of self-destructive cellular programs helps sculpt the nervous system during development, but the molecular mechanisms used are not fully understood. Prior studies have investigated the role of the APP in the developmental degeneration of sensory neurons with contradictory results. In this work, we sought to elucidate the impact of APP deletion in the development of the sensory nervous system in vivo and in vitro. Our in vivo data show an increase in the number of sciatic nerve axons in adult male and female APP-null mice, consistent with the hypothesis that APP plays a pro-degenerative role in the development of peripheral axons. In vitro, we show that genetic deletion of APP delays axonal degeneration triggered by nerve growth factor deprivation, indicating that APP does play a pro-degenerative role. Interestingly, APP depletion does not affect caspase-3 levels but significantly attenuates the rise of axoplasmic Ca2+ that occurs during degeneration. We examined intracellular Ca2+ mechanisms that could be involved and found that APP-null DRG neurons had increased Ca2+ levels within the endoplasmic reticulum and enhanced store-operated Ca2+ entry. We also observed that DRG axons lacking APP have more mitochondria than their WT counterparts, but these display a lower mitochondrial membrane potential. Finally, we present evidence that APP deficiency causes an increase in mitochondrial Ca2+ buffering capacity. Our results support the hypothesis that APP plays a pro-degenerative role in the developmental degeneration of DRG sensory neurons, and unveil the importance of APP in the regulation of calcium signaling in sensory neurons.

SIGNIFICANCE STATEMENT The nervous system goes through a phase of pruning and programmed neuronal cell death during development to reach maturity. In such context, the role played by the APP in the peripheral nervous system has been controversial, ranging from pro-survival to pro-degenerative. Here we present evidence in vivo and in vitro supporting the pro-degenerative role of APP, demonstrating the ability of APP to alter intracellular Ca2+ homeostasis and mitochondria, critical players of programmed cell death. This work provides a better understanding of the physiological function of APP and its implication in developmental neuronal death in the nervous system.

Introduction

Maturation of the nervous system requires the elimination of neuronal cells and connections generated in excess during early stages of embryonic development (Neukomm and Freeman, 2014). This physiological process requires activation of intrinsic self-destruction programs in response to embryonic stage-dependent signals that include guidance cues, neuronal activity patterns, and competition for limited target-derived trophic factors (Luo and O'Leary, 2005). The dysregulation or aberrant activation of mechanisms that resemble developmental neuronal cell death is observed in several neurodevelopmental and neurodegenerative disorders (Cashman and Höke, 2015; Thomas et al., 2016). In vitro models of cell death using embryonic sympathetic and DRG sensory neurons initially maintained and then deprived of NGF have identified several mechanisms that drive developmental neuronal degeneration (Unsain et al., 2013, 2014; Geden et al., 2019). Among these are the role of the Bcl-2 family members, in particular BAX, to actively induce mitochondrial permeabilization, the importance of intracellular Ca2+ overload, particularly through TRPV1 channels, and the involvement of the caspase family of cysteine proteases as executors of cellular destruction (Deckwerth et al., 1996; Putcha et al., 2001; Simon et al., 2012; Unsain et al., 2013; Johnstone et al., 2019). Despite these findings, the precise molecular mechanisms driving the process of neuronal elimination during development are still not fully understood.

The APP was reported to be actively involved in the elimination of neuronal connections during the development of PNS and CNS (Nikolaev et al., 2009; Olsen et al., 2014). APP was proposed to bind death receptor 6 (DR6) and thereby activate a degenerative cascade that resulted in the activation of the apoptotic machinery and the elimination of neuronal connections during development (Nikolaev et al., 2009; Olsen et al., 2014). However, the notion that APP contributes to developmental pruning of the PNS is controversial and remains unresolved (Nishimura et al., 2003; Nikolaev et al., 2009; Vohra et al., 2010; Olsen et al., 2014).

In the present study, we tested the hypothesis that APP mediates the degeneration of DRG sensory axons during development. We explored the PNS of adult male and female APP KO mice at different anatomic locations in search of alterations that reflect the role of APP at early developmental stages. We show that the sciatic nerve of adult APP-null mice contains significantly more axons than the sciatic nerve of WT mice, suggesting that APP may normally promote peripheral axon loss. To directly address the role of APP during developmental degeneration, we used NGF deprivation to induce axonal degeneration in DRG sensory neurons in vitro and definitively show that APP deletion reduces sensory neuron loss that normally occurs following NGF deprivation. DRG sensory neurons lacking APP display normal caspase-3 activation following NGF deprivation; however, the rise of axoplasmic Ca2+ that normally occurs before axonal disintegration is sharply reduced in APP-null neurons (Johnstone et al., 2018, 2019). This reduction in Ca2+ flux correlates with a higher endoplasmic reticulum (ER) Ca2+ content and a larger store-operated calcium entry (SOCE) response in DRG sensory neurons lacking APP. Further, APP-null DRG neurons show an increase in total mitochondrial mass and a decrease in mitochondrial potential. Together, these data indicate that APP plays a pro-degenerative role during PNS development that reflects changes in intracellular Ca2+ regulation.

Materials and Methods

Mouse strains

APP KO mice (Zheng et al., 1995) were maintained in a C57Bl6 strain background. Animal procedures and experiments were approved by the University of British Columbia animal care committee and the Canadian Council of Animal Care. Efforts were made to reduce animal handling and use.

Culture and NGF deprivation of DRG explants

DRGs were dissected from E13.5 mouse embryos and seeded in 12-well plastic (Grenier) or 4-well glass-bottom dishes (CellVis) sequentially coated with 1 mg/ml poly-D-lysine (Sigma-Aldrich), 10 µg/ml laminin-entactin complex (Corning), and 0.1 mg/ml PurCol bovine collagen (Advanced Biomatrix). Explants were grown in phenol-red Neurobasal media (Invitrogen) supplemented with 2% B27 serum-free supplement (Invitrogen), 1% L-glutamine (Wisent), 1% penicillin/streptomycin (Wisent), 10 µm 5-fluoro-2′-deoxyuridine (Sigma-Aldrich), and 12.5 ng/ml NGF (CedarLane) at 37°C, 5% CO2. Deprivation of neurotrophic support was accomplished using 2.0 µg/ml of function-blocking antibodies against NGF (rabbit polyclonal antibody raised against 2.5 s NGF) (Acheson et al., 1991) in fresh media without neurotrophic supplementation.

βIII-tubulin immunocytochemistry, imaging, and quantification of axon degeneration

DRG explants were fixed in 4% PFA solution in PBS for 15 min, washed once in PBS, and blocked in 5% milk in Tris-borate buffer and 0.3% Triton X-100 for 1 h at room temperature. Explants were incubated overnight at 4°C with mouse monoclonal antibody against βIII-tubulin (Millipore, MAB5564) diluted 1:10,000 in blocking solution. DRGs were washed twice in PBS and then incubated with goat anti-mouse conjugated to AlexaFluor-488 (Jackson ImmunoResearch Laboratories, 115-545-003) diluted 1:5000 in blocking solution for a minimum of 3 h at room temperature. Explants were imaged using a Zeiss Observer Z.1 inverted epifluorescence microscope with an automated motorized stage (5× magnification with tilling). From a stitched master image of the plate generated by Zen 2 software (Zeiss), quarter DRG fields were used to generate a set of images for analysis using the R script program Axoquant 2.0 (Johnstone et al., 2018). Final measurements were plotted as the mean axonal area of DRGs derived from at least three separate embryos. Increments of 500 µm were used for statistical analysis (normalized to same increments in control condition).

Tissue processing and histologic assessment of DRG, sciatic nerve, hind-paw skin innervation, and spinal cord motoneurons

Male and female WT and APP-null adult mice were anesthetized and perfused transcardially with 4% PFA in PBS. Dissected tissues were postfixed by immersion in 4% PFA in PBS ON at 4°C, profusely washed with PBS, and stored either in PB 20 mm, pH 7.4, with 0.1% sodium azide at 4°C or cryoprotected with 20% sucrose, blocked in OCT medium embedding compound (Sakura Finetek) on dry ice, and stored at −80°C.

Sciatic nerve processing, staining, and counting OCT-embedded sciatic nerves were transverse sectioned (15 µm), mounted in gelatin-coated slides, air-dried, rehydrated in PBS, permeabilized, blocked, and immunostained with mouse anti-βIII tubulin (Millipore, MAB5564) diluted 1:10,000 in blocking solution (0.3% Triton X-100, 5% normal goat serum in PBS) ON at 4°C. Slides were washed in PBS and incubated with goat anti-mouse conjugated to AlexaFluor-488 (Jackson ImmunoResearch Laboratories, 115-545-003), diluted 1:5000 in blocking solution at room temperature for 3 h. Slides were mounted with Fluoroshield mounting media (Sigma, F6182) and stored at 4°C. Sciatic nerve sections were tiled-imaged in Leica DMi8 confocal microscope and LAS X software with 488 nm laser in 1 µm z increments with 40× oil immersed objective. National Institutes of Health (NIH) ImageJ software Z stacks were converted into a maximum intensity projection image. Mean axonal number per animal was obtained by averaging blind counts of three sections using a manual particle trace and count in NIH ImageJ software.

DRG processing, staining, and counting OCT-embedded L4 DRGs were serially sectioned (30 µm), and every fifth section was collected onto separate gelatin-coated slides. Sections were air-dried ON, rehydrated in PBS, permeabilized, blocked, and immunostained with mouse anti-βIII-tubulin (Millipore, MAB5564), diluted 1:10,000 in blocking solution (0.3% Triton X-100, 5% normal goat serum in PBS) ON at 4°C. Slides were washed in PBS and incubated with goat anti-mouse conjugated to AlexaFluor-488 (Jackson ImmunoResearch Laboratories, 115-545-003), diluted 1:5000 in blocking solution ON at 4°C. Sections were counterstained with 1 µg/ml of DAPI and mounted with Fluoroshield mounting media (Sigma, F6182). DRG sections were tiled-imaged in Leica DMi8 confocal microscope and LAS X software with 488 nm laser in 1 µm z increments with 10× objective. The number of immunopositive DRG neurons was determined by counting neuronal nuclei (on every fifth section) as described previously (Nakamura et al., 2008).

Hind-paw skin processing, staining and free nerve ending (FNE) counting OCT-embedded hind-paw skin was transverse sectioned (20 µm), and every third section was collected onto separate gelatin-coated slides (0.5 mm total skin area sectioned per animal). Sections were air-dried ON, rehydrated in PBS, permeabilized, blocked, and immunostained with rabbit anti-CGRP (Sigma, C8198), diluted 1:3000 in blocking solution (0.3% Triton X-100, 5% normal goat serum in PBS) ON at 4°C. Slides were washed in PBS and incubated with goat anti-rabbit conjugated to AlexaFluor-546 (Invitrogen, A-11035) diluted 1:5000 in blocking solution ON at 4°C. Sections were counterstained with 1 µg/ml of DAPI and mounted with Fluoroshield mounting media (Sigma, F6182). Skin sections were imaged in Leica DMi8 confocal microscope and LAS X software with 546 nm laser in 1 µm z increments with 40× oil immersed objective. Using NIH ImageJ software, z stacks were converted into a maximum intensity projection image. The dermis–epidermis borderline within the image area was traced and its length measured. Immunolabeled intraepidermal CGRP-positive fibers at the glabrous skin were manually and blinded counted. At least three images per section and five sections per animals were analyzed. Counts were normalized to the measured epidermal length and displayed as number of FNEs per 500 µm (Lauria et al., 2005; Koivisto et al., 2012).

Lumbar spinal cord processing, motoneuron staining, and counting OCT-embedded L3-L5 lumbar spinal cords were trans-sectioned (20 µm), and every fifth section was collected onto separate gelatin-coated slides. Sections were air-dried ON, rehydrated in PBS, and subject to heat-induced epitope retrieval by heating to 95°C for 5 min in retrieval solution (10 mm Tris base, 1 mm EDTA solution, 0.05% Tween 20, pH 9.0). Slides were washed in PBS, permeabilized, blocked, and immunostained with goat anti-ChAT (Millipore, AB144P), diluted 1:300 in blocking solution (0.3% Triton X-100, 5% normal goat serum in PBS) ON at 4°C. Slides were washed in PBS and incubated with donkey anti-goat conjugated to Cy3 (Jackson ImmunoResearch Laboratories, 705-165-147), diluted 1:1000 in blocking solution ON at 4°C. Sections were counterstained with 1 µg/ml of DAPI and mounted with Fluoroshield mounting media (Sigma, F6182). Spinal cord sections were imaged in Leica DMi8 confocal microscope and LAS X software with 546 nm laser in 1 µm z increments with 40× oil immersed objective. ChAT-positive motoneurons in the spinal cord displaying a prominent nucleolus were counted in every fifth section, a minimum five sections per animal. Counts were displayed as number of motoneurons per 500 µm of lumbar spinal cord (Lalancette-Hebert et al., 2016).

Ca2+ imaging with fluo-4 and quantification

DRG explants were seeded on glass-bottom dishes (CellVis) and treated with 5 µm Fluo-4 AM (Invitrogen) in neurobasal media for 15 min at 37°C, washed with HBSS, and switched to clear HBSS-based complete media supplemented with HEPES (final concentration 20 mm) to maintain its physiological pH. Explants were tiled-imaged using a Zeiss Observer Z.1 inverted epifluorescence microscope with an automated motorized stage at 40× magnification. Using NIH ImageJ software, stitched master images of each explant were cropped to eliminated soma and Schwann-cell area. From there, a binary mask image of remaining axons was created to measure area and mean pixel intensity corrected by background signal. After calculating the intensity per unit of axonal area, DRG explants from the same embryo were pooled and averaged to generate the mean value per embryo. Measurements were normalized and expressed as fold change from NGF WT control.

Live ER-Ca2+ content and SOCE quantification

Dissociated DRG neurons were seeded on glass-bottom dishes (CellVis), grown for 24 h, and treated with 5 µm Fluo-4 AM (Invitrogen) in neurobasal media for 15 min at 37°C, washed with HBSS, and then allowed to equilibrate in fresh media for 15 min. Plates were then switched to clear HBSS-based media without Ca2+ supplemented with HEPES (final concentration 20 mm) to maintain its physiological pH. Fluo-4 fluorescence was recorded for ∼1 min to establish a baseline. For measurements of ER Ca2+ content, cells were treated with the sarcoplasmic and ER Ca2+ ATPase (SERCA) pump blocker thapsigargin (1 µm final, Sigma-Aldrich, T9033) while recording. A rise in Fluo-4 fluorescence corresponds to ER Ca2+ content release to the cytoplasm. Once fluorescence returns to baseline, Ca2+ influx through SOCE is measured by adding Ca2+ (2 mm final of CaCl2) to the media (Gibon et al., 2010). Fields with at least three DRGs were chosen for recording. Fluo-4 intensity was normalized by neuronal soma area and plotted relative to WT DRGs at an initial fluorescence baseline (without Ca2+ and thapsigargin).

Immunoblotting

For SDS-PAGE and Western blot analysis, a total of 25 DRG explants per well were seeded in 12-well plastic plates (Grenier). For protein harvesting, cultures were washed with PBS, and DRGs were scraped into 90 µl of sample buffer (4% SDS, 20% glycerol, 10% 2-mercaptoethanol, 0.004% bromophenol blue, and 0.125 m Tris-HCl, pH ∼6.8). Samples were boiled for 5 min, centrifuged, and stored at −80°C for later analysis. Antibodies used for immunoblotting were as follows: anti-βIII-tubulin (Millipore MAB5564, 1:10,000), anti-APP-Y188 (Abcam ab32136, 1:1000), anti-caspase-3 (NEB9662, 1:1000), and anti-TOMM20 (Abcam ab56783, 1:1000). Densitometric analysis was performed using Bio-Rad ImageLabTM software.

Mitochondria and axonal live staining with tetramethylrhodamine, ethyl ester (TMRE) and Calcein-AM

DRG explants were treated with 1 µg/ml Calcein-AM (AAT Bioquest) in neurobasal media for 1 h at 37°C. Twenty minutes before the end of the initial Calcein-AM incubation hour, explants were cotreated with 0.25 µm TMRE at 37°C. At the end of the incubation hour, DRGs were switched to clear HBSS-based complete media supplemented with HEPES to maintain physiological pH. Imaging was performed with a Leica DMi8 confocal microscope and LAS X software with 488 and 546 nm lasers in 0.5 µm z increments with 63× oil-immersed objective. The number of mitochondria and the pixel intensity of TMRE signal in the axons were measured using NIH ImageJ software, while the density of mitochondria was calculated per square micron of axon quantified by the area of Calcein-AM-stained axons over a specified threshold. Axons from the same DRG were pooled and averaged to generate the mean value for each DRG. Measurements of TMRE intensity were normalized relative to WT conditions.

Primary mice embryonic fibroblast (MEF) culture

E13.5 embryos from C57Bl6 mice were used to obtain primary WT and APP-null MEF cultures as previously described (Bellingham et al., 2004). For this, the head, liver, and gastrointestinal organs from individual embryos were removed and the remaining body was placed in a new dish with fresh HBSS buffer. The tissue was finely minced with sterile scissors, transferred to a 15 ml tube with 1 ml HBSS buffer, and trypsinized with 0.25% w/v trypsin at 37°C bath for 10 min. Then DNase 0.1 mg/ml was added and incubated at 37°C for 10 more minutes. Trypsinization was stopped using 10% FBS and solution replaced with high-glucose DMEM (Invitrogen) supplemented with 10% FBS. The tissue was then triturated with fire-polished Pasteur pipette, passed through a cell strainer, and pelleted by centrifugation. The cell pellet was resuspended in fresh DMEM + 10% FBS, seeded in 10 cm culture dishes and placed in a 37°C incubator with 5% CO2. The culture medium was replaced with fresh DMEM + 10% FBS after 24 h and then changed every 4 d. MEFs were maintained for at least 7 d before used in experiments. Experiments were performed on MEFs between passages 1-5.

Lipofectamine transfection and mitochondrial Ca2+ buffer capacity measurements

Liposomal transfection reagent Lipofectamine 2000 (Invitrogen) was used to introduce mito-GcAMP (2.5 µg, pCAG mito-GcAMP5G, Addgene #105009) and RFP (2.5 µg, pmRFP-C1) plasmids into dissociated embryonic DRG neurons or mice embryonic fibroblasts at 1:2 (DNA:lipofectamine) ratio according to the manufacturer's protocol. Medium was then removed after 24 h and replaced with fresh medium and incubated for a further 24 h. For mito-GcAMP and RFP fluorescence intensity measurements, time-series recordings at 488 and 546 nm were done using a Zeiss Observer Z.1 inverted epifluorescence microscope with an automated motorized stage (40× magnification). Mitochondrial Ca2+ buffer capacity in transfected cells was assessed after the release of Ca2+ from the ER using the SERCA pump blocker thapsigargin (1 µm final, Sigma-Aldrich, T9033), and once fluorescence returns to baseline, during Ca2+ influx through SOCE by adding back HBSS media with Ca2+ (2 mm final of CaCl2) (Gibon et al., 2010). Mitochondrial depolarization with carbonyl cyanide m-chlorophenylhydrazone 1 µm was done to assess the specificity of Ca2+ measurements within the mitochondria. At least two fields with one transfected cell were chosen for simultaneous recording per plate. Mito-GcAMP and RFP intensity was normalized by cell area and plotted relative to WT cells group at initial fluorescence baseline without Ca2+ and thapsigargin. Final measurements were plotted as the mean intensity per cell area.

Experimental design and statistical analysis

Data were plotted and analyzed using Prism 6 (GraphPad). All data are presented as mean ± SEM. The number of embryos n in each experiment or condition is described in each figure legend. Unpaired t test test (unpaired, two-tailed) was used for two-group experiments comparisons. Two-way ANOVA with Tukey's or Bonferroni's post hoc test was used to analyze differences in multiple groups.

Results

APP is actively involved in the elimination of neuronal connections during the development of the CNS (Olsen et al., 2014). Whether APP plays a similar role in the PNS remains uncertain. To address this knowledge gap, we first studied the sensory nervous system of adult APP-deficient mice. Analysis of sciatic nerve sections revealed that adult APP-deficient mice (7583 ± 157, SEM) have a large statistically significant increase in the number of axons compared with their WT counterparts (5359 ± 196 SEM), t(10) = −8.86, p < 0.01 (two-tailed), a novel finding not previously reported (Fig. 1A). Further stereometric analyses were performed to determine whether the increase in axon number correlated with changes in neuronal cell bodies or free nerve endings. The number of L4 DRG neurons showed no difference between WT and APP-null mice (Fig. 1B). The number of FNEs in the hind-paw skin (Fig. 1C; 31 ± 1.4 FNE per 500 µm of glabrous skin in WT against 33 ± 2.9 in APP-null, unpaired t test t(8) = 0.5, p = 0.588, two-tailed) and the number of lumbar motoneurons (Fig. 1D; 240 ± 10 ChAT-positive motoneurons per 0.5 mm of lumbar spinal cord in WT against 266 ± 12 in APP-null adult mice, unpaired t test t(4) = 1.70, p = 0.164, two-tailed) trended higher in APP-null adults compared with WT littermates, but these observations were not statistically significant.

Figure 1.
Figure 1.

PNS hypertrophy phenotype in adult APP-null mice. A, Representative transverse sections of adult WT and APP-null mice sciatic nerves stained with βIII-tubulin. Scale bar, 100 μm. Quantification shows a significant increase in the number of sciatic nerve axons in APP-null adult animals compared with their WT counterparts. Results show a mean of 5284 ± 292 axons in WT (n = 3) against 7482 ± 187 in APP-null sciatic nerves (n = 3), analyzed by unpaired t test t(4) = 6.23, p < 0.01, two-tailed. Values are mean ± SEM. B, Representative cross-sections of L4 DRG from adult WT and APP-null mice stained with βIII-tubulin. Scale bar, 250 μm. Quantification analysis shows no significant differences in the number of DRG neurons in APP-null adult animals compared with their WT counterparts. Results show a mean of 12,830 ± 682 neurons in WT (n = 3) against 13,035 ± 876 in APP-null DRG (n = 3), analyzed by unpaired t test t(4) = 0.184, p = 0.862, two-tailed. Values are mean ± SEM. C, Representative sections of adult WT and APP-null mice hind-paw skin with CGRP. Scale bar, 10 μm. There is not a significant difference in the number of FNEs in the hind-paw skin of adult APP-null mice compared with their WT counterparts. Results show a mean of 31 ± 1.4 FNE per 500 μm of glabrous skin in WT (n = 5) against 33 ± 2.9 in APP-null hind-paw skin (n = 5), analyzed by unpaired t test t(8) = 0.564, p = 0.588, two-tailed. Values are mean ± SEM. D, Representative sections of L3-L5 spinal cord from adult WT and APP-null mice stained with ChAT. Scale bar, 250 μm. There is not a significant difference in the number of ChAT-positive motoneurons in APP-null adult animals compared with their WT counterparts. Results show a mean of 240 ± 10 motoneurons per 0.5 mm of lumbar spinal cord in WT (n = 3) against 266 ± 12 in APP-null adult mice (n = 3), analyzed by unpaired t test t(4) = 1.70, p = 0.164, two-tailed. Values are mean ± SEM.

The role of APP during developmental remodeling in DRG sensory neurons in vitro remains uncertain. Nishimura et al. (2003) showed that APP protects DRG neurons deprived of NGF (Nishimura et al., 2003), whereas Marc Tessier-Lavigne's group indicated that it played a pro-degenerative role in the same experimental paradigm (Nikolaev et al., 2009; Olsen et al., 2014). To resolve whether APP exerts neuroprotective or neurodegenerative effects in DRG neurons during the period of developmental programmed cell death, WT and APP-KO DRGs were prepared from E13.5 mouse embryos, cultured for 60 h in the presence of NGF (to support survival and neurite outgrowth), and then deprived of NGF (Fig. 2A). Three independent experiments (analyzed alone or pooled) showed that APP-null DRG explants retained higher axonal density following NGF deprivation than their WT counterparts. Figure 2B compares the loss of axonal area between WT and APP-null DRG explants 21 and 24 h after NGF deprivation at two distances from the soma (500-1000 µm and 1000-1500 µm). At 21 h after NGF deprivation, DRG explants from APP-null have lost significantly fewer axons than WT DRG explants (Fig. 2Bi,Bii). This difference remains significant 24 h after NGF deprivation when quantified between 500 and 1000 µm from the soma (Fig. 2Biii); however, no differences between genotype were observed at further distances (Fig. 2Biv).

Figure 2.
Figure 2.

APP genetic deletion protects DRG sensory neurons from degeneration induced by NGF deprivation. A, WT and APP-null DRG explants cultured in the presence of NGF for 48 h and then either maintained with trophic support or deprived with a function blocking anti-NGF antibody (2 µg/ml) for the following 21 or 24 h, before fixation and immunostaining with βIII-tubulin. Scale bar, 250 µm. B, Quantification of axonal area as a function of the distance from the soma using Axoquant 2.0 (Johnstone et al., 2018). Bi–Biv, The relative axonal area lost for each genotype at 21 or 24 h of deprivation at 500-1000 µm or 1000-1500 µm from the soma. The difference of relative axonal area lost between WT and APP-null DRG at different times and distances from the soma was analyzed by unpaired t test and plotted with mean and SEM (n = 7 WT embryos, n = 8 APP-null embryos). *WT versus APP-null; not significant > 0.05. *p < 0.05. **p < 0.01. ***p < 0.001. C, WT and APP-null DRG explants cultured in the presence of NGF for 48 h and then either maintained with trophic support or deprived with a function blocking anti-NGF (2 µg/ml) for the following 24 h, before live-staining with Calcein-AM. Scale bar, 250 µm. D, Quantification of live axonal area as a function of the distance from the soma using Axoquant 2.0 (Johnstone et al., 2018), and plotted in 500 µm bins. Differences between the relative axonal area of WT and APP-null DRG at different time points were analyzed by two-factor ANOVA and Tukey's post hoc comparison and plotted with mean and SEM (n = 3 WT embryos, n = 3 APP-null embryos). *WT:NGF versus WT:dep. 24 h; ****p < 0.0001. # WT:dep. 24 h versus APP-null:dep. 24 h; #p < 0.05. ####p < 0.0001.

DRGs deprived of NGF were also assessed using a Calcein-AM method. Figure 2C, D shows that the drastic reduction of viable axons observed in WT DRGs deprived of NGF is sharply attenuated in APP-null DRGs subjected to NGF withdrawal. These results indicate that APP genetic deletion delays the degeneration of sensory axons induced by NGF deprivation.

Caspases are key executor proteases that are activated in DRG sensory axons deprived of NGF (Simon et al., 2012; Unsain et al., 2013). In our next experiments, we asked whether caspase activation levels differ between WT and APP-null DRG neurons deprived of NGF for 3, 6, 9, 12, and 15 h. Figure 3A–C shows that levels of cleaved caspase-3 within WT and APP-null DRG explants withdrawn from NGF for 3-15 h do not differ.

Figure 3.
Figure 3.

APP genetic deficiency reduces axoplasmic Ca2+ rise but not caspase-3 activation on NGF deprivation. A, Protein lysates from WT and APP-null E13.5 DRG explants cultured for 48 h in the presence of NGF (12.5 ng/ml) and then either maintained with trophic support or deprived of NGF and supplied with a function blocking anti-NGF (2 μg/ml) for 15 h were analyzed by Western blot against APP, caspase-3, and βIII-tubulin. B, Quantification by densitometry of the corresponding bands for cleaved caspase-3 normalized by βIII-tubulin levels and relative to WT NGF control. Cleaved caspase-3 levels significantly increase on NGF deprivation in WT and in APP-null DRG lysates; no significant difference was noted between the genotypes. Analysis by two-factor ANOVA followed by Bonferroni's post hoc comparison and plotted with mean, min/max, and 25%/75% for each panel (n = 4 WT embryos; n = 6 APP-null embryos). *p < 0.05. ***p < 0.001. C, Time course of cleaved caspase-3 levels from WT and APP-null DRG lysates deprived of NGF for 3, 6, 9, 12, and 15 h. Samples were analyzed by Western blot and quantified by densitometry of the corresponding bands for cleaved caspase-3 normalized by βIII-tubulin levels and relative to WT NGF control. No significant difference was noted between the genotypes. Analysis by two-factor ANOVA followed by Bonferroni's post hoc comparison and plotted with mean, min/max, and 25%/75% for each panel (n = 4 embryos per genotype). D, WT and APP-null DRG explants cultured in NGF were maintained in trophic media or withdrawn from trophic support for 15 h before staining with Fluo-4 and imaged by epifluorescence microscopy. Scale bar, 1000 μm. E, The rise in axonal Fluo-4 intensity observed in WT DRG explants on 15 h of NGF deprivation is significantly attenuated in deprived APP-null DRG axons. Box plots represent mean, min/max, and 25%/75% for each panel, analyzed by two-factor ANOVA followed by Bonferroni's post hoc comparison (n = 4 embryos per condition). *p < 0.05.

Another hallmark of sensory axon degeneration is the rise in axoplasmic Ca2+ that occurs several hours after NGF deprivation is initiated (Johnstone et al., 2018, 2019). We asked whether the genetic deletion of APP affects the rise in axoplasmic Ca2+ in DRG explants deprived of NGF. Figure 3D shows that WT sensory axons display a large increase in intra-axonal Ca2+ 15 h after NGF deprivation, whereas this effect is sharply reduced in APP-null axons. When quantified over several independent experiments, it was apparent that axons lacking APP display a statistically significant reduction in intracellular Ca2+ accumulation after NGF deprivation (Fig. 3D,E).

The ER and SOCE are key regulators of intracellular Ca2+ homeostasis. Therefore, in our next experiments, we addressed whether the lack of APP alters the ER Ca2+ content or SOCE. Dissociated DRG neurons from WT and APP-null E13.5 embryos were treated with thapsigargin, a SERCA pump inhibitor, in Ca2+-free media. Then, using the Ca2+ dye Fluo-4, we followed the intracellular rise of Ca2+ that occurs in response to thapsigargin, which will provide a read-out of ER Ca2+ content in these conditions. Figure 4 shows that ER Ca2+ content is higher in APP-null DRGs than WT counterparts (Fig. 4A–C). Further, the addition of Ca2+ to the extracellular media revealed that influx of Ca2+ through the SOCE system is higher in APP-null DRGs than their WT counterparts (Fig. 4D, Thapsigargin + Ca2+). These results indicate that DRG sensory neurons lacking APP have increased levels of Ca2+ within the ER and allow elevated SOCE. These results indicate that DRG sensory neurons lacking APP have defects in calcium homeostasis.

Figure 4.
Figure 4.

APP deletion attenuates ER Ca2+ content and SOCE in DRG neurons. Dissociated WT and APP-null embryonic DRG were cultured for 24 h in the presence of NGF (12.5 ng/ml), stained with Fluo-4 and live-imaged. The recording was initiated in Ca2+-free media to establish the baseline, followed by the treatment with the SERCA pump inhibitor thapsigargin (1 μm final) and ending in media with Ca2+ (2 mm CaCl2 final). A, Representative dissociated DRG soma stained with Fluo-4 at indicated stages during live recording. Scale bar, 20 μm. B, Plot represents the Fluo-4 intensity across time of WT and APP-null DRG somas relative to the WT baseline (first minute) and normalized by soma area. C, Bar graph represents the mean Fluo-4 intensity between WT and APP-null DRG somas at thapsigargin peak, reflecting ER Ca2+ content. Significantly increased Fluo-4 intensity was detected in APP-null DRG somas compared with WT DRGs, analyzed by unpaired t test t(773) = 2.288, p = 0.0224, two-tailed. Values are mean ± SEM (n = 400 WT and 375 APP-null DRG somas). *p < 0.05. D, Fluo-4 intensity after addition of Ca2+ is significantly increased in APP-null DRG somas compared with WT DRGs, reflecting increase SOCE-dependent Ca2+ influx, analyzed by unpaired t test t(773) = 3.196, p = 0.0015, two-tailed. Values are mean ± SEM (n = 400 WT and 375 APP-null DRG somas). **p < 0.01.

Mitochondria constitute an important decision node in axonal degeneration by driving destruction via the intrinsic apoptotic pathway. To probe whether mitochondria in APP-null DRG sensory axons have abnormal characteristics, we assessed mitochondrial proteins by immunoblot and examined mitochondrial potential using the live dye TMRE. Figure 5A shows that levels of TOM20, a subunit of the mitochondrial protein import complex TOM (often used as a surrogate measure of total mitochondrial mass), are significantly increased in APP-null DRGs compared with their WTs DRG counterparts, both in the presence and absence of NGF (Fig. 5A,B). However, staining with TMRE, which provides a readout of mitochondrial membrane potential, revealed a significant reduction in the intensity per axon area in APP-null DRG axons compared with WT neurons (Fig. 5C,D). Together, these data suggest that mitochondria within APP-null DRG axons are abnormal.

Figure 5.
Figure 5.

DRG sensory neurons lacking APP have reduced functional mitochondria and increased TOM20 levels. A, Protein lysates from WT and APP-null E13.5 DRG explants cultured for 48 h in the presence of NGF (12.5 ng/ml) and then either maintained with trophic support or deprived of NGF and supplied with a function blocking anti-NGF (2 μg/ml) for 15 h were analyzed by Western blot against APP, TOM20, and βIII-tubulin. B, Quantification by densitometry of the corresponding bands for TOM20 normalized by βIII-tubulin levels and relative to WT NGF control. Results show a significant increase of TOM20 levels in lysates from APP-null DRG in the presence of NGF compared with their WT counterparts, analyzed using an unpaired t test t(13) = 2.760, p = 0.0162, two-tailed. Values are the mean, min/max, and 25%/75% for each panel (n = 7 WT embryos; n = 8 APP-null embryos). *p < 0.05. C, Representative images of WT and APP-null DRG explants cultured for 24 h in the presence of NGF (12.5 ng/ml) and lived-stained with Calcein-AM and TMRE to identify axons and functional mitochondria, respectively. Scale bar, 20 μm. D, TMRE intensity normalized by axonal area and relative to WT. Quantification shows a significant decrease in relative TMRE intensity in APP-null DRG axons compared with their WT counterparts, analyzed by unpaired t test t(27) = 3.395, p = 0.0021, two-tailed. Values are mean ± SEM (n = 12 WT embryos and 16 APP-null embryos). **p < 0.01.

Mitochondria play a pivotal role in the regulation of intracellular Ca2+. In physiological conditions, the capacity of the mitochondria to buffer cytoplasmic Ca2+ prevents dangerously high concentrations of the ion from accumulating within the cell. Under stressful conditions, the flux of Ca2+ from the ER to the mitochondria, powered by the mitochondrial potential, is a key step in the apoptotic cascade. Given the Ca2+ homeostasis and mitochondrial defects observed in APP-null DRG neurons, we next asked whether mitochondria within DRG neurons deficient in APP have defects in Ca2+ buffering capacity. We transfected dissociated DRG neurons with the mitochondrial-targeted Ca2+ sensor mito-GcAMP and increased intracellular Ca2+ levels by thapsigargin stimulation. Unfortunately, few mito-GcAMP-labeled mitochondria accumulated in axons, and we were therefore only able to measure the mito-GcAMP intensity curve within the soma. The mito-GcAMP signal was slightly higher in APP-null DRGs than in their WT counterparts (Fig. 6A,B), yet the pool of four independent experiments did not show a significant difference in Ca2+ buffering between WT and APP-null DRG somatic mitochondria (Fig. 6C,D). The technical issues with our initial approach led us to consider whether the analysis of non-neuronal cell types may provide evidence for a systemic defect in mitochondrial Ca2+-buffer capacity of APP-nulls. To address this, we derived MEFs from E13.5 WT and APP-null E13.5 embryos, transfected them with mito-GcAMP, switched them to Ca2+-free-media, and then increased cytoplasmic Ca2+ by exposing them to thapsigargin. Interestingly, mitochondria from APP-null MEFs showed significantly higher mito-GcAMP intensity after thapsigargin stimulation (Fig. 6E–G; ER-Ca2+ release). Further, subsequent addition of extracellular Ca2+ revealed an increase in SOCE activity in these cells (Fig. 6H; SOCE). These results demonstrate that APP-null MEFs have an altered mitochondrial Ca2+-buffering capacity and suggest that APP deficiency may cause a systemic defect in this property in a variety of cell types.

Figure 6.
Figure 6.

APP deficiency alters mitochondrial Ca2+ buffer capacity. A, Representative images of dissociated WT and APP-null embryonic DRG cultured for 24 h in the presence of NGF (12.5 ng/ml), transfected with mitochondrial-Ca2+ reporter mito-GcAMP, and live-imaged 48 h later. Scale bar, 20 μm. The recording was initiated in Ca2+-free media to establish the baseline, followed by the treatment with thapsigargin (1 μm final) and ending in media with Ca2+ (2 mm CaCl2 final). B, Plot represents the mito-GcAMP intensity across time of WT and APP-null DRG somas relative to the WT baseline (first minute) and normalized by soma area. C, Bar graph represents the mean mito-GcAMP intensity between WT and APP-null DRG somas at thapsigargin peak, reflecting mitochondria capacity to buffer Ca2+ from ER. Nonsignificant increase in mito-GcAMP intensity was detected in APP-null DRG somas compared with WT DRGs. Analysis by unpaired t test t(90) = 1.423, p = 0.158, two-tailed. Values are mean ± SEM (n = 42 WT and 50 APP-null DRG somas); not significant, p > 0.05. D, Mito-GcAMP intensity at thapsigargin + Ca2+ peak reflecting mitochondrial capacity to buffer Ca2+ influx through SOCE. Mito-GcAMP intensity is not significantly different between genotypes. Analysis by unpaired t test t(90) = 1.165, p = 0.247, two-tailed. Values are mean ± SEM (n = 42 WT and 50 APP-null DRG somas). Not significant, p > 0.05. E, Representative images of WT and APP-null MEF cells cultured for 24 h after seeding, transfected with mitochondrial-Ca2+ reporter mito-GcAMP, and live-imaged 48 h later. Scale bar, 20 μm. The recording was initiated in Ca2+-free media to establish the baseline, followed by the treatment with thapsigargin (1 μm final) and ending in media with Ca2+ (2 mm CaCl2 final). F, Plot represents the mito-GcAMP intensity across time of WT and APP-null MEF cells relative to the WT baseline (first minute) and normalized by soma area. G, Bar graph represents the mean mito-GcAMP intensity between WT and APP-null MEF cells at thapsigargin peak, reflecting mitochondria capacity to buffer Ca2+ from ER. Significant increase in mito-GcAMP intensity was detected in APP-null DRG somas compared with WT DRGs. Analysis by unpaired t test t(149) = 2.518, p = 0.0129, two-tailed. Values are mean ± SEM (n = 83 WT and 68 APP-null MEF cells). *p < 0.05. H, Mito-GcAMP intensity at thapsigargin + Ca2+ peak reflecting mitochondrial capacity to buffer Ca2+ from SOCE. Mito-GcAMP intensity is significantly higher in APP-null MEF compared with WT counterparts. Analysis by unpaired t test t(149) = 2.499, p = 0.0136, two-tailed. Values are mean ± SEM (n = 83 WT and 68 APP-null MEF cells). *p < 0.05.

Discussion

Developmental neuronal cell death is essential for normal high-fidelity patterning of the nervous system, but its molecular mechanisms are still not fully understood. Several research avenues have opened during the last decade, one being the role played by APP, a protein intimately linked with Alzheimer's disease pathology. The available data support an essential role for APP in developmental neuronal remodeling, although its nature has been controversial, ranging from protective to pro-degenerative (Nishimura et al., 2003; Nikolaev et al., 2009; Olsen et al., 2014). Our work helps clarify the physiological role of APP during peripheral neuron development and provides new insights into the mechanisms by which APP contributes to developmental neuron elimination.

Evidence in vivo and in vitro supports a pro-degenerative role for APP during PNS development

A role for APP in neuronal remodeling during the development of the CNS has been established in pruning of RGC axons in the superior colliculus. In the adult brain, experience-dependent plasticity and axonal pruning also rely on APP (Olsen et al., 2014; Marik et al., 2016). Our study provides the first in vivo evidence that APP is required for neuronal remodeling during PNS development. We examined the PNS in adult mice deficient in APP and found a significant increase in the number of axons in the sciatic nerve of APP-null adult mice compared with WT counterparts. A recent study had observed a trend toward increased sciatic nerve axons, especially in the smaller nonmyelinated variety, in APP- and in APLP2-null adult mice; unlike in our study, this observation was not statistically significant (Truong et al., 2019). The reason for the increase in sciatic nerve axons in APP-null mice in our analyses is not certain; the number of the L4 DRG neurons and lumbar motoneurons animals in APP-null animals trended somewhat higher, but significant differences in these pools were not observed. We also noted a trend toward increased CGRP-positive nerve endings in the footpad skin of APP-deficient animals, but again, significant differences were not observed. Interestingly, analysis of APP and APLP2 double KO mice has revealed an excessive nerve terminal sprouting phenotype at the neuromuscular junction (Wang et al., 2005), and APP was recently identified as a novel receptor for the repulsive guidance cue Slit (Wang et al., 2017), which also plays a role in developmental axonal pruning (Vanderhaeghen and Cheng, 2010). Together, these data indicate that the effect of APP deletion on these individual measures is subtle, whereas the sciatic nerve axon counts represent contributions from a wide array of sensory and motoneuron pools that sum to produce a statistically significant result.

In vitro, the evidence supporting a role for APP in developmental neuronal remodeling of sensory neurons has been contradictory. In an early study, Nishimura et al. (2003) reported that degeneration of DRG neurons deprived of NGF is significantly more severe when APP expression is depleted (Nishimura et al., 2003). Olsen et al. (2014) later demonstrated that APP deficiency significantly reduces axonal loss in the same in vitro model (Olsen et al., 2014). In the present study, we report that APP has a pro-degenerative role in DRG neurons deprived of NGF. APP deficiency significantly rescues the loss of axons typically observed in WT DRG explants. These findings were observed using two distinct staining methods and assessed with Axoquant2.0, an unbiased semiautomatic quantification method that measures axonal area at different distances from cell soma (Johnstone et al., 2018).

APP deficiency reduces axoplasmic Ca2+ rise but does not attenuate caspase activation during DRG degeneration

Caspase activation and a rise in axoplasmic Ca2+ play critical roles during NGF-deprived DRG neuron degeneration. Inhibition of caspases through pharmacological or genetic means, or Ca2+ chelation with EGTA, rescues axonal degeneration induced by NGF withdrawal (Simon et al., 2012; Unsain et al., 2013; Johnstone et al., 2018, 2019). We quantified the levels of cleaved caspase-3 and axoplasmic Ca2+ in WT and APP-null DRG neurons 15 h after NGF deprivation. In accordance with the reduction in axon loss, NGF-deprived APP-null explants showed a significant decrease in axoplasmic Ca2+ levels compared with WT DRGs. However, the level of cleaved caspase-3 induced by NGF withdrawal in WT animals and APP nulls was not different. A recent study suggested that caspase activation lies upstream of the axoplasmic Ca2+ rise in the degenerative cascade (Yong et al., 2019), and it is possible that subtle differences in caspase activation, below our detection limits, may drive changes in axoplasmic Ca2+. Also, we cannot rule out differences in activity of other members of the caspase family. We expect that, if they exist, they are likely to be subtle and difficult to demonstrate.

Modulation of intracellular Ca2+ by APP: from the ER to the mitochondria

Dysregulation of ER Ca2+ homeostasis and the consequent induction of the ER stress response facilitate the intrinsic apoptotic pathway (Zhivotovsky and Orrenius, 2011; Logue et al., 2013). Although it is not certain that early ER Ca2+ dysregulation is a component of the degenerative process in NGF-deprived DRG neurons, several molecular markers of the ER-stress response increase in DRG neurons on NGF deprivation (Larhammar et al., 2017). ER Ca2+ release is followed by rapid Ca2+ replenishment to avoid ER stress. Ca2+ reuptake from the cytosol occurs via SERCA pumps, and this allows cytosolic Ca2+ levels to reach homeostasis, with Ca2+ entering the cell from the extracellular milieu via SOCE (Logue et al., 2013). Several studies have demonstrated that APP participates in ER stress-induced cell death in different cell types (Copanaki et al., 2007; O'Connor et al., 2008; Takahashi et al., 2009), and others have highlighted the capacity of APP to alter the basal ER Ca2+ levels and regulate SOCE (Hamid et al., 2007; Niu et al., 2009; Linde et al., 2011; Ma et al., 2012; Gazda et al., 2017). Our work demonstrates that, in the absence of APP, the ER Ca2+ content and SOCE are significantly increased in DRG neurons, suggesting that a delay in the induction of the ER stress response may contribute to the attenuated degenerative process in APP nulls. We recognize that germline APP deletion can have indirect effects in some of our outputs and that determining precisely how APP regulates ER Ca2+ content and SOCE will require further investigation.

Mitochondria act as decision hubs in multiple physiological and pathologic processes, including the modulation of intracellular Ca2+ stores and the control of cell death (Werth and Thayer, 1994; Jacobson and Duchen, 2004; Walsh et al., 2009; Santo-Domingo and Demaurex, 2010; Grimm, 2012; Williams et al., 2013). We found that APP-null DRG neurons have an increase in TOM20 levels, yet the axonal density of TMRE-stained active mitochondria is reduced in APP-null DRG axons. Recent studies have shown that the APP intracellular domain, generated by cleavage of APP by the γ-secretase, is able to act as a transcription factor to induce the expression of the phosphatase and tensin homolog-induced kinase 1 (Pink-1). Pink-1 controls mitochondrial dynamics and mitophagy by selectively enhancing mitochondrial fission and recruiting Parkin to mitochondria. In line with our results, these authors found that, when APP intracellular domain is not expressed or when γ-secretase is inhibited in the presence of Pink-1, TOM20 levels increase and TMRM fluorescence (mitochondrial membrane potential live dye) decreases, respectively (Goiran et al., 2018). Further research will be necessary to determine whether a similar mechanism explains mitochondrial physiology in APP-null DRG neurons.

A recent study performed on primary astrocytes demonstrated that mitochondrial Ca2+ sequestration is delayed in APP-null cells versus their WT counterparts, suggesting that the ability of the mitochondria to buffer Ca2+ in the absence of APP is altered (Montagna et al., 2019). We tested whether DRG from APP nulls also presented this phenotype. Our results showed a trend toward an increase in the capacity of APP-null DRG mitochondria to buffer Ca2+ derived from the ER or the extracellular milieu. For technical purposes, we had to use MEFs derived from WT and APP-null embryos to delve into this further; and using this system, we confirmed that mitochondria from APP-null cells buffer Ca2+ more than mitochondria in WT-derived MEF cells.

In conclusion, we observed that APP deletion increases mitochondrial mass, reduces TMRE signal, and increases the capacity of mitochondria to buffer calcium. Our data therefore suggest that APP deletion leads to more active mitochondria, with a higher buffering capacity and partially prevents axonal degeneration. Together, our in vivo and in vitro data strongly suggest a pro-degenerative role for APP during the development of the peripheral nervous system.

Footnotes

  • This work was supported by Canadian Institutes of Health Research 38942. We thank Svetlana Simtchouk for technical assistance and for managing the APP mouse colony.

  • The authors declare no competing financial interests.

  • Correspondence should be addressed to Philip A. Barker at philip.barker{at}ubc.ca

References

  1. ↵
    2. Acheson A,
    3. Barker PA,
    4. Alderson RF,
    5. Miller FD,
    6. Murphy RA
    (1991) Detection of brain-derived neurotrophic factor-like activity in fibroblasts and Schwann cells: inhibition by antibodies to NGF. Neuron 7:265–275. doi:10.1016/0896-6273(91)90265-2 pmid:1873030
    OpenUrlCrossRefPubMed
  2. ↵
    2. Bellingham SA,
    3. Ciccotosto GD,
    4. Needham BE,
    5. Fodero LR,
    6. White AR,
    7. Masters CL,
    8. Cappai R,
    9. Camakaris J
    (2004) Gene knockout of amyloid precursor protein and amyloid precursor-like protein-2 increases cellular copper levels in primary mouse cortical neurons and embryonic fibroblasts. J Neurochem 91:423–428. doi:10.1111/j.1471-4159.2004.02731.x pmid:15447675
    OpenUrlCrossRefPubMed
  3. ↵
    2. Cashman CR,
    3. Höke A
    (2015) Mechanisms of distal axonal degeneration in peripheral neuropathies. Neurosci Lett 596:33–50. doi:10.1016/j.neulet.2015.01.048 pmid:25617478
    OpenUrlCrossRefPubMed
  4. ↵
    2. Copanaki E,
    3. Schürmann T,
    4. Eckert A,
    5. Leuner K,
    6. Müller WE,
    7. Prehn JH,
    8. Kögel D
    (2007) The amyloid precursor protein potentiates CHOP induction and cell death in response to ER Ca2+ depletion. Biochim Biophys Acta 1773:157–165. doi:10.1016/j.bbamcr.2006.10.002 pmid:17113167
    OpenUrlCrossRefPubMed
  5. ↵
    2. Deckwerth TL,
    3. Elliott JL,
    4. Knudson CM,
    5. Johnson EM Jr.,
    6. Snider WD,
    7. Korsmeyer SJ
    (1996) BAX is required for neuronal death after trophic factor deprivation and during development. Neuron 17:401–411. doi:10.1016/S0896-6273(00)80173-7 pmid:8816704
    OpenUrlCrossRefPubMed
  6. ↵
    2. Gazda K,
    3. Kuznicki J,
    4. Wegierski T
    (2017) Knockdown of amyloid precursor protein increases calcium levels in the endoplasmic reticulum. Sci Rep 7:10. doi:10.1038/s41598-017-15166-2
    OpenUrlCrossRef
  7. ↵
    2. Geden MJ,
    3. Romero SE,
    4. Deshmukh M
    (2019) Apoptosis versus axon pruning: molecular intersection of two distinct pathways for axon degeneration. Neurosci Res 139:3–8. doi:10.1016/j.neures.2018.11.007 pmid:30452947
    OpenUrlCrossRefPubMed
  8. ↵
    2. Gibon J,
    3. Tu P,
    4. Bouron A
    (2010) Store-depletion and hyperforin activate distinct types of Ca2+-conducting channels in cortical neurons. Cell Calcium 47:538–543. doi:10.1016/j.ceca.2010.05.003 pmid:20621761
    OpenUrlCrossRefPubMed
  9. ↵
    2. Goiran T,
    3. Duplan E,
    4. Chami M,
    5. Bourgeois A,
    6. El Manaa W,
    7. Rouland L,
    8. Dunys J,
    9. Lauritzen I,
    10. You H,
    11. Stambolic V,
    12. Biféri MG,
    13. Barkats M,
    14. Pimplikar SW,
    15. Sergeant N,
    16. Colin M,
    17. Morais VA,
    18. Pardossi-Piquard R,
    19. Checler F,
    20. Alves da Costa C
    (2018) β-Amyloid precursor protein intracellular domain controls mitochondrial function by modulating phosphatase and tensin homolog–induced kinase 1 transcription in cells and in Alzheimer mice models. Biol Psychiatry 83:416–427. doi:10.1016/j.biopsych.2017.04.011 pmid:28587718
    OpenUrlCrossRefPubMed
  10. ↵
    2. Grimm S
    (2012) The ER–mitochondria interface: the social network of cell death. Biochim Biophys Acta 1823:327–334. doi:10.1016/j.bbamcr.2011.11.018 pmid:22182703
    OpenUrlCrossRefPubMed
  11. ↵
    2. Hamid R,
    3. Kilger E,
    4. Willem M,
    5. Vassallo N,
    6. Kostka M,
    7. Bornhövd C,
    8. Reichert AS,
    9. Kretzschmar HA,
    10. Haass C,
    11. Herms J
    (2007) Amyloid precursor protein intracellular domain modulates cellular calcium homeostasis and ATP content. J Neurochem 102:1264–1275. doi:10.1111/j.1471-4159.2007.04627.x pmid:17763532
    OpenUrlCrossRefPubMed
  12. ↵
    2. Jacobson J,
    3. Duchen MR
    (2004) Interplay between mitochondria and cellular calcium signaling. Mol Cell Biochem 256:209–257. doi:10.1023/B:MCBI.0000009869.29827.df pmid:14977182
    OpenUrlCrossRefPubMed
  13. ↵
    2. Johnstone AD,
    3. Hallett RM,
    4. de Léon A,
    5. Carturan B,
    6. Gibon J,
    7. Barker PA
    (2018) A novel method for quantifying axon degeneration. PLoS One 13:e0199570. doi:10.1371/journal.pone.0199570 pmid:30020957
    OpenUrlCrossRefPubMed
  14. ↵
    2. Johnstone AD,
    3. de Léon A,
    4. Unsain N,
    5. Gibon J,
    6. Barker PA
    (2019) Developmental axon degeneration requires TRPV1-dependent Ca2+ influx. eNeuro 6:ENEURO.0019-19.2019. doi:10.1523/ENEURO.0019-19.2019
    OpenUrlCrossRef
  15. ↵
    2. Koivisto A,
    3. Hukkanen M,
    4. Saarnilehto M,
    5. Chapman H,
    6. Kuokkanen K,
    7. Wei H,
    8. Viisanen H,
    9. Åkerman KE,
    10. Lindstedt K,
    11. Pertovaara A
    (2012) Inhibiting TRPA1 ion channel reduces loss of cutaneous nerve fiber function in diabetic animals: sustained activation of the TRPA1 channel contributes to the pathogenesis of peripheral diabetic neuropathy. Pharmacol Res 65:149–158. doi:10.1016/j.phrs.2011.10.006 pmid:22133672
    OpenUrlCrossRefPubMed
  16. ↵
    2. Lalancette-Hebert M,
    3. Sharma A,
    4. Lyashchenko AK,
    5. Shneider NA
    (2016) Gamma motor neurons survive and exacerbate alpha motor neuron degeneration in ALS. Proc Natl Acad Sci USA 113:E8316–E8325. doi:10.1073/pnas.1605210113 pmid:27930290
    OpenUrlAbstract/FREE Full Text
  17. ↵
    2. Larhammar M,
    3. Huntwork-Rodriguez S,
    4. Jiang Z,
    5. Solanoy H,
    6. Sengupta Ghosh A,
    7. Wang B,
    8. Kaminker JS,
    9. Huang K,
    10. Eastham-Anderson J,
    11. Siu M,
    12. Modrusan Z,
    13. Farley MM,
    14. Tessier-Lavigne M,
    15. Lewcock JW,
    16. Watkins TA
    (2017) Dual leucine zipper kinase-dependent PERK activation contributes to neuronal degeneration following insult. Elife 6:e20725. doi:10.7554/eLife.20725
    OpenUrlCrossRefPubMed
  18. ↵
    2. Lauria G,
    3. Cornblath DR,
    4. Johansson O,
    5. McArthur JC,
    6. Mellgren SI,
    7. Nolano M,
    8. Rosenberg N,
    9. Sommer C
    , European Federation of Neurological Societies (2005) EFNS guidelines on the use of skin biopsy in the diagnosis of peripheral neuropathy. Eur J Neurol 12:747–758. doi:10.1111/j.1468-1331.2005.01260.x pmid:16190912
    OpenUrlCrossRefPubMed
  19. ↵
    2. Linde CI,
    3. Baryshnikov SG,
    4. Mazzocco-Spezzia A,
    5. Golovina VA
    (2011) Dysregulation of Ca2+ signaling in astrocytes from mice lacking amyloid precursor protein. Am J Physiol Cell Physiol 300:C1502–C1512. doi:10.1152/ajpcell.00379.2010 pmid:21368296
    OpenUrlCrossRefPubMed
  20. ↵
    2. Logue SE,
    3. Cleary P,
    4. Saveljeva S,
    5. Samali A
    (2013) New directions in ER stress-induced cell death. Apoptosis 18:537–546. doi:10.1007/s10495-013-0818-6 pmid:23430059
    OpenUrlCrossRefPubMed
  21. ↵
    2. Luo L,
    3. O'Leary DD
    (2005) Axon retraction and degeneration in development and disease. Annu Rev Neurosci 28:127–156. doi:10.1146/annurev.neuro.28.061604.135632 pmid:16022592
    OpenUrlCrossRefPubMed
  22. ↵
    2. Ma T,
    3. Gong K,
    4. Yan Y,
    5. Song B,
    6. Zhang X,
    7. Gong Y
    (2012) Mitochondrial modulation of store-operated Ca2+ entry in model cells of Alzheimer's disease. Biochem Biophys Res Commun 426:196–202. doi:10.1016/j.bbrc.2012.08.062 pmid:22935417
    OpenUrlCrossRefPubMed
  23. ↵
    2. Marik SA,
    3. Olsen O,
    4. Tessier-Lavigne M,
    5. Gilbert CD
    (2016) Physiological role for amyloid precursor protein in adult experience-dependent plasticity. Proc Natl Acad Sci USA 113:7912–7917. doi:10.1073/pnas.1604299113 pmid:27354516
    OpenUrlAbstract/FREE Full Text
  24. ↵
    2. Montagna E,
    3. Crux S,
    4. Luckner M,
    5. Herber J,
    6. Colombo AV,
    7. Marinković P,
    8. Tahirovic S,
    9. Lichtenthaler SF,
    10. Wanner G,
    11. Müller UC,
    12. Sgobio C,
    13. Herms J
    (2019) In vivo Ca2+ imaging of astrocytic microdomains reveals a critical role of the amyloid precursor protein for mitochondria. Glia 67:985–998. doi:10.1002/glia.23584 pmid:30667091
    OpenUrlCrossRefPubMed
  25. ↵
    2. Nakamura S,
    3. Senzaki K,
    4. Yoshikawa M,
    5. Nishimura M,
    6. Inoue K,
    7. Ito Y,
    8. Ozaki S,
    9. Shiga T
    (2008) Dynamic regulation of the expression of neurotrophin receptors by Runx3. Development 135:1703–1711.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    2. Neukomm LJ,
    3. Freeman MR
    (2014) Diverse cellular and molecular modes of axon degeneration. Trends Cell Biol 24:515–523. doi:10.1016/j.tcb.2014.04.003 pmid:24780172
    OpenUrlCrossRefPubMed
  27. ↵
    2. Nikolaev A,
    3. McLaughlin T,
    4. O'Leary DD,
    5. Tessier-Lavigne M
    (2009) APP binds DR6 to trigger axon pruning and neuron death via distinct caspases. Nature 457:981–989. doi:10.1038/nature07767 pmid:19225519
    OpenUrlCrossRefPubMed
  28. ↵
    2. Nishimura I,
    3. Takazaki R,
    4. Kuwako K,
    5. Enokido Y,
    6. Yoshikawa K
    (2003) Upregulation and antiapoptotic role of endogenous Alzheimer amyloid precursor protein in dorsal root ganglion neurons. Exp Cell Res 286:241–251. doi:10.1016/s0014-4827(03)00066-1 pmid:12749853
    OpenUrlCrossRefPubMed
  29. ↵
    2. Niu Y,
    3. Su Z,
    4. Zhao C,
    5. Song B,
    6. Zhang X,
    7. Zhao N,
    8. Shen X,
    9. Gong Y
    (2009) Effect of amyloid β on capacitive calcium entry in neural 2a cells. Brain Res Bull 78:152–157. doi:10.1016/j.brainresbull.2008.10.003 pmid:19000747
    OpenUrlCrossRefPubMed
  30. ↵
    2. O'Connor T,
    3. Sadleir KR,
    4. Maus E,
    5. Velliquette RA,
    6. Zhao J,
    7. Cole SL,
    8. Eimer WA,
    9. Hitt B,
    10. Bembinster LA,
    11. Lammich S,
    12. Lichtenthaler SF,
    13. Hébert SS,
    14. De Strooper B,
    15. Haass C,
    16. Bennett DA,
    17. Vassar R
    (2008) Phosphorylation of the translation initiation factor eIF2α increases BACE1 levels and promotes amyloidogenesis. Neuron 60:988–1009. doi:10.1016/j.neuron.2008.10.047 pmid:19109907
    OpenUrlCrossRefPubMed
  31. ↵
    2. Olsen O,
    3. Kallop DY,
    4. McLaughlin T,
    5. Huntwork-Rodriguez S,
    6. Wu Z,
    7. Duggan CD,
    8. Simon DJ,
    9. Lu Y,
    10. Easley-Neal C,
    11. Takeda K,
    12. Hass PE,
    13. Jaworski A,
    14. O'Leary DD,
    15. Weimer RM,
    16. Tessier-Lavigne M
    (2014) Genetic analysis reveals that amyloid precursor protein and death receptor 6 function in the same pathway to control axonal pruning independent of β-secretase. J Neurosci 34:6438–6447. doi:10.1523/JNEUROSCI.3522-13.2014 pmid:24806670
    OpenUrlAbstract/FREE Full Text
  32. ↵
    2. Putcha GV,
    3. Moulder KL,
    4. Golden JP,
    5. Bouillet P,
    6. Adams JA,
    7. Strasser A,
    8. Johnson EM Jr.
    (2001) Induction of BIM, a proapoptotic BH3-only BCL-2 family member, is critical for neuronal apoptosis. Neuron 29:615–628. doi:10.1016/S0896-6273(01)00238-0 pmid:11301022
    OpenUrlCrossRefPubMed
  33. ↵
    2. Santo-Domingo J,
    3. Demaurex N
    (2010) Calcium uptake mechanisms of mitochondria. Biochim Biophys Acta 1797:907–912. doi:10.1016/j.bbabio.2010.01.005 pmid:20079335
    OpenUrlCrossRefPubMed
  34. ↵
    2. Simon DJ,
    3. Weimer RM,
    4. McLaughlin T,
    5. Kallop D,
    6. Stanger K,
    7. Yang J,
    8. O'Leary DD,
    9. Hannoush RN,
    10. Tessier-Lavigne M
    (2012) A caspase cascade regulating developmental axon degeneration. J Neurosci 32:17540–17553. doi:10.1523/JNEUROSCI.3012-12.2012 pmid:23223278
    OpenUrlAbstract/FREE Full Text
  35. ↵
    2. Takahashi K,
    3. Niidome T,
    4. Akaike A,
    5. Kihara T,
    6. Sugimoto H
    (2009) Amyloid precursor protein promotes endoplasmic reticulum stress-induced cell death via C/EBP homologous protein-mediated pathway. J Neurochem 109:1324–1337. doi:10.1111/j.1471-4159.2009.06067.x pmid:19476545
    OpenUrlCrossRefPubMed
  36. ↵
    2. Thomas MS,
    3. Davis R,
    4. Karmiloff-Smith A,
    5. Knowland VC,
    6. Charman T
    (2016) The over-pruning hypothesis of autism. Dev Sci 19:284–305. doi:10.1111/desc.12303 pmid:25845529
    OpenUrlCrossRefPubMed
  37. ↵
    2. Truong PH,
    3. Ciccotosto GD,
    4. Merson TD,
    5. Spoerri L,
    6. Chuei MJ,
    7. Ayers M,
    8. Xing YL,
    9. Emery B,
    10. Cappai R
    (2019) Amyloid precursor protein and amyloid precursor-like protein 2 have distinct roles in modulating myelination, demyelination, and remyelination of axons. Glia 67:525–538. doi:10.1002/glia.23561 pmid:30506868
    OpenUrlCrossRefPubMed
  38. ↵
    2. Unsain N,
    3. Higgins JM,
    4. Parker KN,
    5. Johnstone AD,
    6. Barker PA
    (2013) XIAP regulates caspase activity in degenerating axons. Cell Rep 4:751–763. doi:10.1016/j.celrep.2013.07.015 pmid:23954782
    OpenUrlCrossRefPubMed
  39. ↵
    2. Unsain N,
    3. Heard KN,
    4. Higgins JM,
    5. Barker PA
    (2014) Production and isolation of axons from sensory neurons for biochemical analysis using porous filters. J Vis Exp 89:51795. doi:10.3791/51795
    OpenUrlCrossRefPubMed
  40. ↵
    2. Vanderhaeghen P,
    3. Cheng HJ
    (2010) Guidance molecules in axon pruning and cell death. Cold Spring Harb Perspect Biol 2:a001859. doi:10.1101/cshperspect.a001859 pmid:20516131
    OpenUrlAbstract/FREE Full Text
  41. ↵
    2. Vohra BP,
    3. Sasaki Y,
    4. Miller BR,
    5. Chang J,
    6. DiAntonio A,
    7. Milbrandt J
    (2010) Amyloid precursor protein cleavage-dependent and-independent axonal degeneration programs share a common nicotinamide mononucleotide adenylyltransferase 1-sensitive pathway. J Neurosci 30:13729–13738. doi:10.1523/JNEUROSCI.2939-10.2010 pmid:20943913
    OpenUrlAbstract/FREE Full Text
  42. ↵
    2. Walsh C,
    3. Barrow S,
    4. Voronina S,
    5. Chvanov M,
    6. Petersen OH,
    7. Tepikin A
    (2009) Modulation of calcium signalling by mitochondria. Biochim Biophys Acta 1787:1374–1382. doi:10.1016/j.bbabio.2009.01.007 pmid:19344663
    OpenUrlCrossRefPubMed
  43. ↵
    2. Wang B,
    3. Li H,
    4. Mutlu SA,
    5. Bowser DA,
    6. Moore MJ,
    7. Wang MC,
    8. Zheng H
    (2017) The amyloid precursor protein is a conserved receptor for slit to mediate axon guidance. eNeuro 4:ENEURO.0185-17.2017. doi:10.1523/ENEURO.0185-17.2017
    OpenUrlAbstract/FREE Full Text
  44. ↵
    2. Wang P,
    3. Yang G,
    4. Mosier DR,
    5. Chang P,
    6. Zaidi T,
    7. Gong YD,
    8. Zhao NM,
    9. Dominguez B,
    10. Lee KF,
    11. Gan WB,
    12. Zheng H
    (2005) Defective neuromuscular synapses in mice lacking amyloid precursor protein (APP) and APP-like protein 2. J Neurosci 25:1219–1225. doi:10.1523/JNEUROSCI.4660-04.2005 pmid:15689559
    OpenUrlAbstract/FREE Full Text
  45. ↵
    2. Werth JL,
    3. Thayer SA
    (1994) Mitochondria buffer physiological calcium loads in cultured dorsal root ganglia neurons. J Neurosci 14:348–356. doi:10.1523/JNEUROSCI.14-01-00348.1994
    OpenUrlAbstract/FREE Full Text
  46. ↵
    2. Williams GS,
    3. Boyman L,
    4. Chikando AC,
    5. Khairallah RJ,
    6. Leder WJ
    (2013) Mitochondrial calcium uptake. Proc Natl Acad Sci USA 110:10479–10486. doi:10.1073/pnas.1300410110 pmid:23759742
    OpenUrlAbstract/FREE Full Text
  47. ↵
    2. Yong Y,
    3. Gamage K,
    4. Cheng I,
    5. Barford K,
    6. Spano A,
    7. Winckler B,
    8. Deppmann C
    (2019) p75NTR and DR6 regulate distinct phases of axon degeneration demarcated by spheroid rupture. J Neurosci 39:9503–9520. doi:10.1523/JNEUROSCI.1867-19.2019 pmid:31628183
    OpenUrlAbstract/FREE Full Text
  48. ↵
    2. Zheng H,
    3. Jiang M,
    4. Trumbauer ME,
    5. Sirinathsinghji DJ,
    6. Hopkins R,
    7. Smith DW,
    8. Heavens RP,
    9. Dawson GR,
    10. Boyce S,
    11. Conner MW,
    12. Stevens KA,
    13. Slunt HH,
    14. Sisoda SS,
    15. Chen HY,
    16. Van der Ploeg LH
    (1995) β-Amyloid precursor protein-deficient mice show reactive gliosis and decreased locomotor activity. Cell 81:525–531. doi:10.1016/0092-8674(95)90073-x pmid:7758106
    OpenUrlCrossRefPubMed
  49. ↵
    2. Zhivotovsky B,
    3. Orrenius S
    (2011) Calcium and cell death mechanisms: a perspective from the cell death community. Cell Calcium 50:211–221. doi:10.1016/j.ceca.2011.03.003 pmid:21459443
    OpenUrlCrossRefPubMed
Back to top

In this issue

Email
Print
View Full Page PDF
Citation Tools
Respond to this article
APP Genetic Deficiency Alters Intracellular Ca2+ Homeostasis and Delays Axonal Degeneration in Dorsal Root Ganglion Sensory Neurons
Andrés de León, Julien Gibon, Philip A. Barker
Journal of Neuroscience 31 August 2022, 42 (35) 6680-6691; DOI: 10.1523/JNEUROSCI.0162-22.2022
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords

Responses to this article

Respond to this article

Jump to comment:

No eLetters have been published for this article.