Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Dec 1;8(12):1421-1437.
doi: 10.15252/emmm.201606403. Print 2016 Dec.

Mitochondria-associated membrane collapse is a common pathomechanism in SIGMAR1- and SOD1-linked ALS

Seiji Watanabe  1 , Hristelina Ilieva  2   3 , Hiromi Tamada  4 , Hanae Nomura  1 , Okiru Komine  1 , Fumito Endo  1 , Shijie Jin  1 , Pedro Mancias  5 , Hiroshi Kiyama  4 , Koji Yamanaka  6
Affiliations

Mitochondria-associated membrane collapse is a common pathomechanism in SIGMAR1- and SOD1-linked ALS

Seiji Watanabe et al. EMBO Mol Med. .

Abstract

A homozygous mutation in the gene for sigma 1 receptor (Sig1R) is a cause of inherited juvenile amyotrophic lateral sclerosis (ALS16). Sig1R localizes to the mitochondria-associated membrane (MAM), which is an interface of mitochondria and endoplasmic reticulum. However, the role of the MAM in ALS is not fully elucidated. Here, we identified a homozygous p.L95fs mutation of Sig1R as a novel cause of ALS16. ALS-linked Sig1R variants were unstable and incapable of binding to inositol 1,4,5-triphosphate receptor type 3 (IP3R3). The onset of mutant Cu/Zn superoxide dismutase (SOD1)-mediated ALS disease in mice was accelerated when Sig1R was deficient. Moreover, either deficiency of Sig1R or accumulation of mutant SOD1 induced MAM disruption, resulting in mislocalization of IP3R3 from the MAM, calpain activation, and mitochondrial dysfunction. Our findings indicate that a loss of Sig1R function is causative for ALS16, and collapse of the MAM is a common pathomechanism in both Sig1R- and SOD1-linked ALS Furthermore, our discovery of the selective enrichment of IP3R3 in motor neurons suggests that integrity of the MAM is crucial for the selective vulnerability in ALS.

Keywords: amyotrophic lateral sclerosis; inositol 1,4,5‐triphosphate receptor type 3; mitochondria‐associated membrane; sigma 1 receptor.

PubMed Disclaimer

Figures

Figure 1
Figure 1. p.L95fs is a novel amyotrophic lateral sclerosis (ALS) causative mutation in SIGMAR1

  1. A family tree for the juvenile ALS patient with L95 fs mutation in sigma 1 receptor (Sig1R). The arrow indicates a proband.

  2. Schematic diagram of domain structure of Sig1R, coded by SIGMAR1. Amino acid sequence misplaced by L95fs mutation is shown by a red rectangle. Note that both ALS‐causative mutations (L95fs, E102Q; highlighted with red arrows) are located near the ligand‐binding motif. In addition, recently identified mutations that cause distal hereditary motor neuropathy (dHMN), E138Q, E150K, and a splice‐site mutation resulting in in‐frame deletion of 20 amino acids (Δ31–50) are indicated by green arrows.

Figure 2
Figure 2. Both E102Q and L95fs Sig1R variants are unstable and lose their abilities to control Ca2+ flux in Neuro2a cells
  1. A

    Immunoblotting analysis of Neuro2a (N2a) cells expressing Sig1R‐FLAG variants.

  2. B

    Cycloheximide chase analysis on N2a cells expressing Sig1R‐FLAG. Quantitative data of immunoblotting for the levels of Sig1R‐FLAG protein and its variants during the cycloheximide chase were plotted as mean ± SEM of three independent experiments (upper panel). Representative immunoblots for the levels of Sig1R‐FLAG proteins were shown (lower panel). **P < 0.01, *P = 0.0216 in E102Q versus wild type (WT); ## P < 0.01, # P < 0.05 in L95fs versus WT. A one‐way ANOVA with subsequent post hoc Tukey's test.

  3. C

    N2a cells transfected with Sig1R‐FLAG variants were incubated with MG‐132 (10 μM) or the combination of E64d and pepstatin A (5 μg/ml each) (E64d/PepA) for 8 h. The relative mean levels of Sig1R‐FLAG variants determined by immunoblotting from three independent experiments were plotted as mean ± SEM (upper panel). Representative immunoblots were shown (lower panel). *P < 0.05 versus no inhibitor control. A one‐way ANOVA with subsequent post hoc Tukey's test.

  4. D

    Schematic representation of subcellular fractionation used in this study. Details are described in the Materials and Methods section (Cyto, cytosol; P1, nucleus and debris; Mito, mitochondria; MAM, mitochondria‐associated membrane; P3, microsomal fraction).

  5. E, F

    Subcellular localization of inositol 1,4,5‐triphosphate receptor type 1 (IP3R1) and type 3 (IP3R3) in N2a cells stably expressing human IP3R3 (N2a‐IP3R3 cells, E) or HeLa cells (F). IP3R1 and IP3R3 in the isolated fractions were identified by immunoblotting with the specific antibodies as indicated. Proper fractionation of these samples was confirmed by the fraction‐specific markers as indicated.

  6. G

    Interaction of IP3R3 with wild‐type Sig1R or E102Q ALS‐linked Sig1R mutant. Sig1R‐FLAG variants were transfected in N2a‐IP3R3 cells, and IP3R3 was co‐immunoprecipitated using an anti‐FLAG antibody and identified by immunoblotting with the specific antibodies as indicated.

  7. H

    Suppression of Sig1R by siRNA. Lysates of N2a cells transfected with control siRNA (siCtrl) or siRNA against Sig1R (siSig1R) for 24 h were blotted. Similar results were obtained from three independent experiments. Paired t‐test was used for statistical analysis.

  8. I, J

    Cytoplasmic (I) or mitochondrial (J) calcium (Ca2+) flux in N2a or N2a cells stably expressing human IP3R3 (N2a‐IP3R3). siCtrl or siSig1R was transfected into N2a or N2a‐IP3R3 cells, then cytoplasmic and mitochondrial Ca2+ flux were determined with fluo‐4 and Case12‐mito, respectively. The fluorescence intensity was normalized to the intensity in resting state at 0 s and plotted as mean ± SD. **P < 0.01, *P < 0.05. Two‐way ANOVA with subsequent post hoc Tukey's test. n = 10 each.

  9. K, L

    Cytoplasmic (K) or mitochondrial (L) Ca2+ flux in N2a‐IP3R3 cells. siCtrl or siSig1R was transfected with or without the Sig1R‐FLAG variants. The data are obtained and plotted as described in (I and J). n = 10 each. **P < 0.01, *P < 0.05. Two‐way ANOVA with subsequent post hoc Tukey's test.

Source data are available online for this figure.
Figure 3
Figure 3. Accelerated disease onset in Sig1R‐deficient SOD1G85R mice
  1. A

    The levels of Sig1R in the lumbar spinal cord (LSC) or brain of Sig1R−/−, Sig1R+/−, or Sig1R+/+ mice at 5 months old. Representative immunoblots obtained from three independent experiments are shown. Asterisk denotes non‐specific bands.

  2. B, C

    Sig1R‐deficient SOD1G85R mice (SOD1G85R/Sig1R−/−) exhibited accelerated onset of the disease (B) and shortened survival time (C) compared to SOD1G85R mice with one or two copies of Sig1R (SOD1G85R/Sig1R+/−, SOD1G85R/Sig1R+/+). Open circles indicate Sig1R knockout mice (Sig1R−/−) that never developed motor neuron disease but that were sacrificed at 396 days of age (n = 8). Mean onset or survival times of mice with ± SD are shown on the top of Kaplan–Meyer curves. **P < 0.01, *P < 0.05; log‐rank test. n = 14 (SOD1G85R/Sig1R+/+), 29 (SOD1G85R/Sig1R+/−), 17 (SOD1G85R/Sig1R−/−).

  3. D, E

    Body weight was lost earlier both in male (D) and female (E) SOD1G85R/Sig1R−/− mice. Data are shown as mean ± SD. *P < 0.05 in SOD1G85R/Sig1R+/+ versus SOD1G85R/Sig1R−/−, † P < 0.05 in SOD1G85R/Sig1R+/− versus SOD1G85R/Sig1R−/−. Two‐way ANOVA with subsequent post hoc Tukey's test. n = 8 (SOD1G85R/Sig1R+/+), 13 (SOD1G85R/Sig1R+/−), 8 (SOD1G85R/Sig1R−/−) in male (D); and n = 6 (SOD1G85R/Sig1R+/+), 16 (SOD1G85R/Sig1R+/−), 9 (SOD1G85R/Sig1R−/−) in female (E). The animals were the same as those used in panels (B and C). Error bars denote SD.

  4. F

    Decreased performance on the rotarod test was observed in SOD1G85R/Sig1R−/− mice. The test was performed at 0–30 rpm with 0.1 rpm/s acceleration every week. Data are shown as mean ± SD. **P < 0.0001 in SOD1G85R/Sig1R+/+ versus SOD1G85R/Sig1R−/−, †† P < 0.0001 in SOD1G85R/Sig1R+/− versus SOD1G85R/Sig1R−/−. Two‐way ANOVA with subsequent post hoc Tukey's test. The animals were the same as those used in panels (B–E). Error bars denote SD.

  5. G, H

    Motor function of WT (Sig1R+/+) and Sig1R knockout (Sig1R−/−) mice at 5 or 12 months old was evaluated by the rotarod test (0–30 rpm, accelerated at 0.1 rpm/s). Average time on the rotating rod was plotted. Error bars denote SD. Unpaired t‐test.

Source data are available online for this figure.
Figure 4
Figure 4. Mutant SOD1 proteins accumulate at the MAM and mitochondria in mice and cells expressing mutant SOD1
  1. A, B

    N2a cells expressing SOD1WT, SOD1G85R or SOD1G93A (A), or spinal cords of WT or mutant SOD1 transgenic mice at end stage (B) were fractionated as described in the Materials and Methods section (Cyto, cytoplasm; P1, nuclei and debris; Mito, mitochondria; MAM, mitochondria‐associated membrane; P3, microsomal fraction). Proper fractionation was confirmed by the fraction‐specific markers as indicated and also shown in Fig EV1A–E. The asterisk denotes non‐specific bands.

  2. C

    Tissue‐ and cell type‐specific accumulation of mutant SOD1 at the MAM. Brain, liver, or primary astrocytes from SOD1G93A mice were fractionated and blotted as in (A and B) (upper panel). Quantification of SOD1 proteins at the MAM relative to ones in the cytoplasm was performed from immunoblotting data with an anti‐human SOD1 antibody in (B) and the top panel of (C), and the data are plotted as mean ± SEM from three independent experiments (lower panel). **P < 0.01 versus SOD1WT spinal cord; one‐way ANOVA with subsequent post hoc Tukey's test. Proper fractionation of these samples was confirmed in Fig EV1F–H.

  3. D

    Mutant SOD1 prevented association of ER with mitochondria. Isolated ER (P3: microsomal fraction) and mitochondria of N2a cells expressing wild‐type (WT) or G85R (85) SOD1 for 48 h were mixed and incubated. Mitochondrial pellets after centrifugation were analyzed by immunoblotting using anti‐PDI (ER marker) and anti‐VDAC (mitochondrial marker), respectively. All these results were confirmed by three independent experiments.

Source data are available online for this figure.
Figure 5
Figure 5. Accumulation of mutant SOD1 at the MAM is most prominent around the disease onset
  1. A, B

    Time course analyses of mutant SOD1 levels at the MAM in SOD1G93A mouse spinal cords (A) and brains (B). Immunoblots show levels of mutant SOD1 protein in the indicated fractions at various time points (left panels). Quantitative data at the right were plotted as mean ± SEM of three independent experiments. Arrows indicate a rapid reduction in SOD1 level at the MAM just following the onset of disease.

  2. C, D

    Time course analyses of mutant SOD1 accumulation at the MAM in SOD1G85R (C) and SOD1G37R (D) mouse spinal cords. Quantitative data from three independent experiments were plotted as mean ± SEM (bottom). Arrows indicate a rapid reduction in SOD1 level at the MAM just following the onset of disease.

  3. E

    Time course analyses of MAM‐specific proteins at MAM and in the whole lysates of SOD1G93A mouse spinal cords.

Data information: Fractions from a single mouse tissue at the indicated ages were loaded on each lane. Representative blots from at least three independent experiments were shown. GAPDH in the cytosolic fractions was used as a loading control.Source data are available online for this figure.
Figure EV1
Figure EV1. Confirmation of the subcellular fractionation in various samples
  1. A–H

    Representative immunoblotting images of fractionated samples from N2a cells transfected with mutant SOD1 (A and B), spinal cords (C–E), brain (F), primary cultured astrocytes (G), or liver (H) from SOD1 transgenic mice. Representative blots from at least three independent experiments are shown. The asterisks indicate non‐specific bands.

Source data are available online for this figure.
Figure 6
Figure 6. Loss of Sig1R or expression of mutant SOD1 compromises the MAM integrity in motor neurons
  1. A–I

    Immunofluorescence staining of spinal cords and brains from non‐transgenic (Non‐Tg), SOD1 transgenic, or Sig1R−/− mice. Transverse sections of mouse spinal cords (A, D–I) or sagittal sections of mouse brains (B, C) were stained using anti‐Sig1R (white), βIII‐tubulin (red), and IP3R3 (green) antibodies. Note that Sig1R and IP3R3 are co‐localized in the motor neurons of the anterior horn (A) and the hypoglossal nucleus (B), and IP3R3 was not expressed in hippocampal neurons (C). Mutant SOD1 induced aggregation of Sig1R and mislocalization of IP3R3 in anterior horn neurons (D–G), while their abnormalities were not observed in SOD1WT motor neurons (H). Mislocalization of IP3R3 was also observed in Sig1R−/− mouse spinal cords (I). Scale bars: 50 μm.

  2. J, K

    Representative electron micrographs of the MAM (J) (double‐headed arrows) in motor neurons of 12‐month‐old Non‐Tg, SOD1G85R, or Sig1R−/− mice. Note that ER–mitochondria contacted areas were reduced in both SOD1G85R or Sig1R−/− mice. Quantification of the mitochondria surface associated with ER was calculated in (K). For quantification, 13–19 motor neurons and 224–316 mitochondria with MAM in each animal (n = 2) were analyzed. Data are expressed as mean ± SEM. **P < 0.0001 versus Non‐Tg; one‐way ANOVA with subsequent post hoc Tukey's test. Scale bars: 300 nm.

  3. L

    The levels for IP3R3 and calreticulin were decreased in the MAM fractions of Sig1R−/− mouse brains. MAM fractions and whole tissue lysates of Sig1R+/+ or Sig1R−/− mouse brains were immunoblotted. Representative blots from three independent experiments are shown.

Source data are available online for this figure.
Figure EV2
Figure EV2. Selective expression of IP3R3 in the motor nuclei of mouse brain
  1. A–F

    Immunofluorescence staining of mouse brains from non‐transgenic mice. Sagittal sections of mouse brains were stained using anti‐Sig1R (white), βIII‐tubulin (red), and IP3R3 (green) antibodies. Sig1R was expressed in most of the neurons tested. However, IP3R3 was not expressed in cerebellum (A), hippocampus (B), or cortex (C), but specifically co‐localized with Sig1R in hypoglossal nucleus (D), facial motor nucleus (E), or motor nucleus of trigeminal (F), indicating selective expression in IP3R3 in the motor‐related nuclei. Scale bars: 50 ?m. The white boxed areas were magnified in the images at bottom.

Figure EV3
Figure EV3. Representative micrographs of ventral horn neurons in non‐transgenic, SOD1G85R, or Sig1R−/− mice

  1. Immunofluorescence staining of mouse ventral horn neurons from non‐transgenic or Sig1R−/− mice with anti‐PDI and NeuN antibodies. Note that general ER morphology was not affected by the Sig1R deficiency. Scale bars: 50 μm.

  2. Representative low‐magnification electron micrographs of the MAM (arrows) in motor neurons of 12‐month‐old non‐transgenic (Non‐Tg), SOD1G85R, or Sig1R−/− mice. Scale bars: 500 nm.

Figure 7
Figure 7. Sig1R inhibits mutant SOD1‐mediated, abnormal calpain activation and restores intracellular ATP levels
  1. A

    N2a cells were transfected with WT or mutant SOD1 in the presence or absence of IP3R3. Plotted viability of the cells measured by a neurotoxicity assay revealed that IP3R3 is involved in cell vulnerability against mutant SOD1. Data are expressed as mean ± SEM from three independent experiments, triplicated in each experiment.

  2. B–G

    Calpain activity (B–D) and intracellular ATP levels (E–G) were measured in N2a‐IP3R3 cells expressing Sig1R‐FLAG (B and E), SOD1 (C and F), or both (D and G). Mean ± SEM from three independent experiments is plotted.

  3. H, I

    Calpain activity in vivo was determined by the cleavage of spectrin α II. Calpain‐cleaved 150 kDa fragment of spectrin α II in mouse lumbar spinal cord or brain (H) was normalized to β‐actin (I). Quantitative data in immunoblotting analyses are plotted as mean ± SEM from three independent experiments.

Data information: **P < 0.01, *P < 0.05; one‐way ANOVA with subsequent post hoc Tukey's test (A–G, I).Source data are available online for this figure.
Figure 8
Figure 8. Pre‐symptomatic administration of PRE‐084, a Sig1R agonist, restores Sig1R functions at the MAM and prevents Sig1R aggregation in vivo
  1. A, B

    Cytoplasmic (A) or mitochondrial (B) Ca2+ flux was measured in N2a‐IP3R3 cells treated with PRE‐084 (5 μM) or NE‐100 (5 μM) for 1 h prior to fluorescent imaging. Cytoplasmic and mitochondrial Ca2+ flux were detected by fluo‐4 and Case12‐mito, respectively. The fluorescence intensity was normalized by the resting state at 0 s and plotted as mean ± SD. n = 10 each. **P < 0.01, *P < 0.05; two‐way ANOVA with subsequent post hoc Tukey's test.

  2. C, D

    Calpain activity and intracellular ATP levels were measured in N2a‐IP3R3 cells treated with PRE‐084 (5 μM) or NE‐100 (5 μM) for 24 h. Data are plotted as mean ± SEM from three independent experiments. *P < 0.05 versus mock control; one‐way ANOVA with subsequent post hoc Tukey's test.

  3. E, F

    SOD1G93A mice were intraperitoneally administered with saline or PRE‐084 (0.25 mg/kg, 3 times per week) from postnatal day 35 to 95. Lumbar spinal cord sections were stained by using anti‐Sig1R (white), IP3R3 (green), and βIII‐tubulin (red) at 95 days old. Arrowheads indicate the neurons with normal distribution of Sig1R and IP3R3. Scale bars: 50 μm.

See this image and copyright information in PMC

References

    1. Aggad D, Veriepe J, Tauffenberger A, Parker JA (2014) TDP‐43 toxicity proceeds via calcium dysregulation and necrosis in aging Caenorhabditis elegans motor neurons. J Neurosci 34: 12093–12103 - PMC - PubMed
    1. Al‐Saif A, Al‐Mohanna F, Bohlega S (2011) A mutation in sigma‐1 receptor causes juvenile amyotrophic lateral sclerosis. Ann Neurol 70: 913–919 - PubMed
    1. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, Mori H, Mann D, Tsuchiya K, Yoshida M, Hashizume Y et al (2006) TDP‐43 is a component of ubiquitin‐positive tau‐negative inclusions in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res Commun 351: 602–611 - PubMed
    1. Area‐Gomez E, de Groof AJ, Boldogh I, Bird TD, Gibson GE, Koehler CM, Yu WH, Duff KE, Yaffe MP, Pon LA et al (2009) Presenilins are enriched in endoplasmic reticulum membranes associated with mitochondria. Am J Pathol 175: 1810–1816 - PMC - PubMed
    1. Area‐Gomez E, Del Carmen Lara Castillo M, Tambini MD, Guardia‐Laguarta C, de Groof AJC, Madra M, Ikenouchi J, Umeda M, Bird TD, Sturley SL et al (2012) Upregulated function of mitochondria‐associated ER membranes in Alzheimer disease. EMBO J 31: 4106–4123 - PMC - PubMed

Publication types