Systemic overexpression of SQSTM1/p62 accelerates disease onset in a SOD1H46R-expressing ALS mouse model

Systemic overexpression of SQSTM1/p62 accelerates disease onset in a SOD1H46R-expressing ALS... Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease characterized by a selective loss of upper and lower motor neurons. Recent studies have shown that mutations in SQSTM1 are linked to ALS. SQSTM1 encodes SQSTM1/p62 that regulates not only autophagy via the association with MAP1LC3/LC3 and ubiquitinated proteins but also the KEAP1-NFE2L2/Nrf2 anti-oxidative stress pathway by interacting with KEAP1. Previously, we H46R have demonstrated that loss of SQSTM1 exacerbates disease phenotypes in a SOD1 -expressing ALS mouse H46R model. To clarify the effects of SQSTM1 overexpression in this model, we generated SQSTM1 and SOD1 double- H46R H46R transgenic (SQSTM1;SOD1 ) mice. SQSTM1;SOD1 mice exhibited earlier disease onset and shorter lifespan than H46R did SOD1 mice. Conversely, disease progression after the onset rather slightly but significantly slowed in H46R SQSTM1;SOD1 mice. However, there were observable differences neither in the number of Nissl positive neurons H46R nor in the distribution of ubiquitin-positive and/or SQSTM1-positive aggregates between SOD1 and SQSTM1; H46R SOD1 mice. It was noted that these protein aggregates were mainly observed in neuropil, and partly localized to astrocytes and/or microglia, but not to MAP2-positive neuronal cell bodies and dendrites at the end-stage of disease. Nonetheless, the biochemically-detectable insoluble SQSTM1 and poly-ubiquitinated proteins were H46R H46R significantly and progressively increased in the spinal cord of SQSTM1;SOD1 mice compared to SOD1 mice. H46R These results suggest that overexpression of SQSTM1 in SOD1 mice accelerates disease onset by compromising the protein degradation pathways. Keywords: Amyotrophic lateral sclerosis, SOD1, SQSTM1/p62, Ubiquitin-positive aggregates Introduction different genes that are linked to ALS have been identified, Amyotrophic lateral sclerosis (ALS) is a progressive neuro- molecular mechanisms by which motor neurons are select- degenerative disease characterized by a selective loss of ively and progressively degenerated are still unknown [3]. upper and lower motor neurons. ALS patients exhibit en- Recently, mutations in the SQSTM1 gene have been hanced tendon reflex caused by loss of upper motor neurons identified in patients with ALS and ALS/frontotem- and muscle atrophy, impaired motor control caused by loss poral dementia (FTD) [4–8]. SQSTM1 mutations have of lower motor neurons without sensory disturbance, and originally been identified in Paget disease of bone [9]. ultimately die within 3 to 5 years due to respiratory failure The SQSTM1 gene product, sequestosome1 (SQSTM1/ [1]. An approximately 10% of ALS cases is familial and the p62), is a multi-functional adapter protein. SQSTM1 remaining 90% are sporadic [2]. Although more than 50 regulates not only autophagy via the association with microtubule-associated protein 1 light chain 3 (MAP1LC3/LC3) and poly-ubiquitinated proteins [10, * Correspondence: shinji@is.icc.u-tokai.ac.jp 11] but also the Kelch-like ECH-associated protein 1 Department of Molecular Life Sciences, Tokai University School of Medicine, (KEAP1)-nuclear factor related erythroid 2-related fac- Isehara, Kanagawa 259-1193, Japan The Institute of Medical Sciences, Tokai University, Isehara, Kanagawa tor 2 (NFE2L2/Nrf2) anti-oxidative stress pathway by 259-1193, Japan interacting with KEAP1 [12]. Indeed, ALS-linked Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mitsui et al. Molecular Brain (2018) 11:30 Page 2 of 16 SQSTM1 missense mutant (Leu341Val) [7]showeda compromising the protein degradation pathways in a H46R decreased affinity to LC3, resulting in impaired au- SOD1 -expressing ALS mouse model. tophagy [13]. It has also been shown that SQSTM1 is multiply phosphorylated. Especially, the phosphoryl- Methods ation of SQSTM1 at serine (Ser) 403 (corresponding to Animals H46R mouse Ser405) increases the binding affinity to SOD1 -tg mice express human SOD1 carrying the poly-ubiquitin molecules and enhances the autophagic deg- H46R mutation under control of the human SOD1 pro- radation of poly-ubiquitinated proteins [14, 15]. In addition, motor [22]. CAG-SQSTM1-HA (SQSTM1)-tg mice ex- it has been reported that the phosphorylation at Ser349 press human SQSTM1 with hemagglutinin (HA) tag at (corresponding to mouse Ser351) enhances the interaction the C-terminus under control of the CAG [(C) the cyto- between SQSTM1 and KEAP1, thereby activating the megalovirus (CMV) early enhancer element, (A) the pro- KEAP1-NFE2L2/Nrf2 pathway [16]. With respect to the re- moter, the first exon and the first intron of chicken lationship between SQSTM1 and neurodegenerative dis- β-actin gene, and (G) the splice acceptor of the rabbit H46R eases, it has been demonstrated that SQSTM1-positive β-globin gene] promoter [24]. Both SOD1 -tg and aggregates are accumulated in the spinal cord and brain of SQSTM1-tg mice were backcrossed to 57BL/6 N (B6) ALS and ALS/FTD patients [17–19]. mice more than 10 generations, and were maintained as On the other hand, the SOD1 gene has been identified as B6 congenic lines. We generated SQSTM1-overexpressing H46R H46R a causative gene for an autosomal dominant form of famil- SOD1 mice (SQSTM1;SOD1 )bycrossing male H46R ial ALS in 1993. Thus far, more than 180 different SOD1 SOD1 mice with female SQSTM1 mice. The offsprings mutations have been identified (http://alsod.iop.kcl.ac.uk/ were genotyped by PCR using genomic DNA extracted H46R als/). Several transgenic (tg) ALS mouse models expressing from ear tissues. Primers used were as follows; SOD1 : different SOD1 mutants have been generated [20]. Among hSOD1_ex2: 5′-TCAGAAACTCTCTCCAACTTTGC-3′ H46R them, SOD1 mutation has originally been identified in and 5′-CAAGTATGGGTCACCAGCAC-3′,hSOD1_ex4: Japanese kindred, showing a slow progression of symptoms 5′-GGCATCAGCCCTAATCCATC-3′ and 5′-CCGC H46R [21]. SOD1 -tg mouse recapitulates many neurological GACTAACAATCAAAGTG-3′, SQSTM1-HA:5′-AGCT and pathological features observed not in only familial but GCCTTGTACCCACATC-3′ and 5′-AGCGTAATC also in sporadic ALS patients, which include progressive TGGAACATCGT-3′.Micewerehousedat22 °Cwitha motor neuron degeneration, impaired motor control, and 12 h light-dark cycle. Food and water were fed ad libitum. accumulation of SQSTM1 and ubiquitin-positive aggregates in the spinal cord [19, 20, 22, 23]. Antibodies It has been shown that, in other motor neuron disease Primary antibodies used for immunohistochemistry were (MND) mouse model, i.e., spinal and bulbar muscular atro- listed as follows; anti-human Ubiquitin (MBL MK-11-3; phy (SBMA) mice, loss of SQSTM1 causes earlier disease mouse monoclonal, 1:300), anti-human Ubiquitin (Dako onset, whereas overexpression of it ameliorates disease Z0458; rabbit polyclonal, 1:500), anti-human SQSTM1 symptoms [24]. We have previously demonstrated that loss (Thermo PA5–20839; rabbit polyclonal, 1:300), anti-human of SQSTM1 exacerbates disease phenotypes, including a p62-C (PROGEN GP62-C; guinea-pig polyclonal, 1:500), shorter lifespan, accelerated body weight decline and motor anti-rat MAP2 (SIGMA M9942; mouse monoclonal, H46R dysfunction, in a mutant SOD1 -tg ALS mouse model 1:500), anti-GFAP (NICHIREI 422251; rabbit polyclonal, [25]. Thesefindings suggest aneuroprotective role of 1:2), anti-mouse Iba1 (Abcam ab178896; rabbit monoclo- SQSTM1 in vivo. However, the effect of SQSTM1 overex- nal, 1:500), and anti-HA-tag (Cell Signaling 3724; rabbit pression on the disease onset and progression in mutant monoclonal, 1:500) antibodies. Secondary antibodies in- SOD1-expressing mice remains unknown. cluded Alexa 594 conjugated anti-mouse IgG (Molecular In this study, to clarify the effect of systemic overexpres- Probes, 1:500), Alexa 594 conjugated anti-Rabbit IgG sion of SQSTM1 on the disease onset and progression as (Molecular Probes, 1:500), Alexa 594 conjugated well as the accumulation of aggregated proteins in anti-guinea-pig IgG (Molecular Probes, 1:500), Alexa H46R SOD1 mice, we generated SQSTM1-overexpressing 488 conjugated anti-mouse IgG (Molecular Probes, H46R H46R SOD1 double-tg (SQSTM1;SOD1 ) mice, and ana- 1:500), and Alexa 488 conjugated anti-Rabbit IgG lyzed lifespan, body weight, the number of motor neurons, (Molecular Probes, 1:500) antibodies. and the distribution of ubiquitin-positive aggregates Primary antibodies used for western blotting included in the spinal cord. In addition, we conducted a series anti-human SOD1 (SANTA CRUZ cs-11,407; rabbit poly- of western blot analysis to quantify the amount of clonal, 1:15000), anti-human misfolded SOD1 (C4F6; expressed protein levels including phosphorylated mouse monoclonal, 1:3000) [26], anti-bovine ubiquitin forms of SQSTM1. We here revealed that overexpres- (SANTA CRUZ cs-8017; mouse monoclonal, 1:3000), sion of SQSTM1 accelerated disease onset by anti-human SQSTM1 (MBL PM045; rabbit polyclonal, Mitsui et al. Molecular Brain (2018) 11:30 Page 3 of 16 1:2000), anti-human phosphorylated SQSTM1 (S403) overnight. Lumbar cord was further immersed in metha- (MBL D343–3; rat monoclonal, 1:3000), anti-mouse phos- nol:chloroform = 1:1 solution for 30 min, and in 100% phorylated SQSTM1 (S351) (MBL PM074; rabbit poly- chloroform for 2 h followed by the treatment with fresh clonal, 1:2000), anti-human LC3 (SIGMA L8918; rabbit chloroform overnight. The resulting lumbar cord sam- polyclonal, 1:5000), anti-human NQO1 (Abcam ab34173; ples were embedded in paraffin. Paraffin-embedded rabbit polyclonal, 1:3000), anti-rabbit GAPDH (MBL transverse sections at a thickness of 6 μm were prepared M171–3; mouse monoclonal, 1:5000), anti-Actin (SIGMA by microtome (LEICA RM2165) for Nissl staining and A5060; rabbit polyclonal, 1:1000), and anti-HA-tag immunohistochemistry. (SANTA CRUZ sc-805; rabbit polyclonal, 1:3000) anti- bodies. Secondary antibodies included horseradish peroxid- Nissl staining ase (HRP)-conjugated anti-mouse IgG (Jackson, 1:5000), Lumbar cord paraffin sections from mice with four dif- HRP-conjugated anti-rabbit IgG (GE Healthcare NA934, ferent genotypes at 16 and 22 weeks of age (WT, H46R H46R 1:5000), and HRP-conjugated anti-rat IgG (SANTA CRUZ SQSTM1, SOD1 , and SQSTM1;SOD1 ), and at H46R H46R sc-2006, 1:5000) antibodies. end-stage (SOD1 and SQSTM1;SOD1 )or 28 weeks of age (WT and SQSTM1), were treated with Lifespan analysis xylene (5 min × 3 times) and ethanol (5 min × 3 times), Lifespan (endpoint) of tg mice with each genotype and hydrated. Slices were permeabilized in 1% Triton H46R H46R (SOD1 and SQSTM1;SOD1 ) were determined by X-100/phosphate buffered saline (PBS) for 15 min and the observations that mice were unable to crawl by their stained with 50 μl of NeuroTrace green fluorescent Nissl forelimb and to eat/drink by themselves. stains (ThermoFisher N21480) (1:500)/PBS for 25 min. Slices were washed with PBS for 2 h and mounted by VEC- Growth curve analysis TASHIELD Mounting Medium with DAPI for nuclear Body weight of mice with each genotype [wild-type staining. A total of 5 representative Nissl staining images of H46R H46R (WT), SQSTM1, SOD1 , and SQSTM1;SOD1 ] every tenth serial sections from each sample was captured was weekly monitored from 5 weeks to a maximum of by fluorescence microscope (KEYENCE BZ-9000). 32 weeks of age. Counting the number of large Nissl-positive neurons Disease onset and progression analyses Large Nissl-positive neurons, representing motor neu- Onset of the disease of mice with each genotype rons, were identified by using Dynamic cell count (KEY- H46R H46R (SOD1 and SQSTM1;SOD1 ) was defined as the ENCE). In brief, tones of the captured images were point (date) at which body weight was successively de- binarized, and the area covering 600 pixels (px) × 400 px creased from its peak-value in each animal without any corresponding to anterior horn of the spinal cord was increases thereafter. Post-onset survival was defined as selected. Automatic cell separation procedures (setting the duration from the onset of disease to the endpoint. resolution; 14, threshold; 45 with excluding cells whose cross-sectional areas were less than 240 px) were exe- Preparation of paraffin sections cuted. Neurons identified were defined as large We obtained tissue samples of the spinal cord from Nissl-positive neurons and counted. The numbers of H46R H46R SOD1 and SQSTM1;SOD1 mice at 16 and large Nissl-positive neurons from 5 representative im- 22 weeks of age, and at end-stage. We also obtained the ages from each sample were summed. samples from WT and SQSTM1 mice at 16, 22, and 28 weeks of age. Samples from WT and SQSTM1 mice Immunohistochemistry at 28 weeks of age were used to compare to those from Spinal cord sections were deparafinized in xylene tg mice at their end-stage. Mice were anesthetized with (5 min × 3 times) and ethanol (5 min × 3 times), and 4% isoflurane by inhalation and perfused with physio- were hydrated. Sections were heated by microwave oven logical saline containing 100 U/ml heparin, followed by in 300 ml of 0.1 M citric acid (pH 5.0) for 15 min. After 4% paraformaldehyde (PFA)/0.1 M phosphate buffer cooling at room temperature (RT), sections were washed (PB) (pH 7.2). Spinal cord together with the spinal col- with PBS and blocked with 50 μl of 0.1% Triton X-100/ umn was removed and fixed with 4% PFA/PB overnight 5% normal goat serum (NGS)/PBS for 30 min. Sections at 4 °C. Lumbar cord (L4-L5) was removed from the were incubated with primary antibody in 50 μl of the spinal column, cut by an approximately 2 mm length antibody diluted solution (0.05% Triton X-100, 2% NGS/ (rostral to caudal direction), and post-fixed with 4% PBS) overnight at 4 °C. Sections were washed with PBS PFA/PB overnight at 4 °C. After washing with PB over- for 2 h and incubated with secondary antibody in 50 μl night twice, lumbar cord was immersed in 100% metha- of the antibody diluted solution for 2 h. Sections were nol for 1 h followed by soaking to fresh methanol washed with PBS for 1 h and mounted using Mitsui et al. Molecular Brain (2018) 11:30 Page 4 of 16 VECTASHIELD Mounting Medium with DAPI. Signals (BIO-RAD 1610374) was used as molecular weight were analyzed by fluorescence microscope (KEYENCE markers. Proteins were electro-transferred onto poly- BZ-710). vinylidene fluoride (PVDF) membrane (Merckmilli- pore Immobilon-P) in the transfer buffer [25 mM Preparation of tissue samples for western blot analysis Tris-HCl, 192 mM glycine, 20% (w/v) methanol]. The To analyze the expression and distribution of transgene membranes were blocked with 0.5% skimmed milk/ product; i.e., SQSTM1-HA, we obtained tissue samples TBST [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, from WT and SQSTM1 mice at 16 weeks of age. Tissues 0.1% Tween20] for 1 h at RT. When using including the liver, skeletal muscle, olfactory bulb, cere- anti-phosphorylation SQSTM1 antibody, membranes bral cortex, hippocampus, cerebellum, and whole spinal were blocked with 50% blocking one (Wako)/TBST cord were removed, immediately frozen on dry ice, and for 30 min at RT. Membranes were incubated with stored at − 80 °C until use. To analyze the expression of the primary antibody in TBST overnight at 4 °C. particular proteins of interest in the spinal cord, we ob- After washing with TBST (10 min, 6 times), mem- tained the samples from mice with four different geno- branes were incubated with HRP-conjugated second- H46R H46R types (WT, SQSTM1, SOD1 ,and SQSTM1;SOD1 ) ary antibody in TBST for 2 h at RT. After washing H46R at 16 and 22 weeks of age, and at end-stage (SOD1 with TBST, signals were visualized by Immobilon H46R and SQSTM1;SOD1 ) or 28 weeks of age (WT and (Merck Millipore) and detected by Ez-Capture MG SQSTM1). The whole spinal cord was removed from each (ATTO). mouse, immediately frozen on dry ice, and stored at − 80 ° C until use. Quantification of the signal intensities on western blotting Preparation of protein samples We conducted a quantitative analysis of the signal intensities Spinal cord tissues were weighted and homogenized in 2 on western blotting to determine the expression levels of weight-volume (mg/μl) of PBS by sonication (Output 2, proteins of interest in 1% Triton X-100 soluble and insoluble 10–20 times, BRANSON Sonifier 450). One-hundred factions from mice with four different genotypes (WT, H46R H46R microliters of homogenates and 700 μl of Lysis buffer A’ SQSTM1, SOD1 ,and SQSTM1;SOD1 )at16and H46R [25 mM Tris-HCl (pH 7.5), 50 mM NaCl, 1% Triton 22 weeks of age, and at end-stage (SOD1 and H46R X-100, Complete Protease Inhibitor Cocktail (Roche), SQSTM1;SOD1 ) or 28 weeks of age (WT and SQSTM1). Phosphatase inhibiter Cocktail I (SIGMA)] were The signal intensities were quantified by using CS Analyzer mixed, followed by centrifugation at 23,000×g for Ver3.0 (ATTO) and were normalized by the levels of glycer- 20 min at 4 °C. Supernatant was collected as 1% Tri- aldehyde 3-phosphate dehydrogenase (GAPDH) for soluble ton X-100 soluble fraction. The insoluble pellet was fractions or β-actin for insoluble fractions. washed by mixing with 600 μl of Lysis buffer A’ and centrifuged again, and resulting pellet was suspended Statistical analysis in 400 μl of Lysis buffer B′ [25 mM Tris-HCl All statistical analyses were conducted using Prism 5 (Graph (pH 7.5), 50 mM NaCl, 5% sodium dodecyl sulfate pad). Data for lifespan, onset of disease, and post-onset sur- H46R H46R (SDS)], sonicated (Output 2, 5–10 times), and used as vival between SOD1 and SQSTM1;SOD1 mice were 1% Triton X-100 insoluble fraction. Protein concen- compared using Kaplan-Meier survival analysis with tration was determined by Micro BCA (Thermo, Log-rank test. Statistical significances for body weight, the MicroBCA Protein Assay kit #23235). Each protein number of spinal motor neurons, and quantitative western fraction was diluted by the corresponding lysis buffer blotting data between groups were evaluated by One-way and adjusted its concentration at 0.75 μg/μl. or Two-way ANOVA with Bonferroni post hoc tests. A One-hundred microliters of diluted samples were p-value < 0.05 was considered as reaching statistical mixed with 50 μl of 3xSDS buffer [187 mM Tris-HCl significance. (pH 6.8), 6% SDS, 30% glycerol, 0.015% bromophenol blue (BPB), 15% 2-melcaptoethanol] and heated for Results 5minat 95°C. Expression and tissue distribution of SQSTM1-HA in SQSTM1 transgenic mice Western blot analysis SQSTM1-tg mouse carries the expression-cassette that Equal amount of proteins was subjected to can express the C-terminally HA-tagged human SDS-polyacrylamide gel electrophoresis (PAGE) (Wako SQSTM1 (SQSTM1-HA) protein under control of the Supersep™Ace, 5–20% 17well) in SDS-PAGE buffer CAG promoter (Additional file 1: Figure S1a) [24]. To [25mM Tris-HCl, 192mMglycine,0.1%(w/v)SDS]. determine the distribution and expression levels of Precision Plus Protein standards dual color standard SQSTM1-HA in SQSTM1 mice, we prepared protein Mitsui et al. Molecular Brain (2018) 11:30 Page 5 of 16 samples from the liver, skeletal muscle, olfactory bulb, (23.8 ± 2.1 weeks) (Fig. 1d and Table 1; Log-rank test; cerebral cortex, hippocampus, cerebellum, and spinal p = 0.0007). To investigate the difference in disease H46R H46R cord from WT and SQSTM1 mice, and performed west- progression between SOD1 and SQSTM1;SOD1 ern blotting using anti-SQSTM1 and anti-HA antibodies. mice, we analyzed the survival interval after disease onset Mouse endogenous SQSTM1 was detected in all tissues (post-onset survival). Contrary to the disease onset, examined with highest in the cerebral cortex and hippo- survival interval after the onset of disease in H46R campus (Additional file 1: Figure S1b, see WT). The SQSTM1;SOD1 mice (3.9 ± 1.3 weeks) was slightly CAG promoter-driven human SQSTM1-HA was highly but significantly extended when compared to H46R expressed in skeletal muscle, and also detected in the cen- SOD1 mice (3.0 ± 1.1 weeks) (Fig. 1e and Table 1; tral nervous system including the spinal cord, while it was Log-rank test; p = 0.0314). undetectable in the liver (Additional file 1: Figure S1b). Progressive decreases in the number of motor neurons in H46R Overexpression of SQSTM1 accelerates body weight the lumbar spinal cord of SOD1 -expressing mice H46R declines in SOD1 mice To investigate whether earlier disease onset observed in H46R We performed growth curve analysis of WT, SQSTM1, SQSTM1;SOD1 mice was associated with motor H46R H46R SOD1 , and SQSTM1;SOD1 mice. Both WT and neuron loss in the spinal cord, we performed Nissl stain- SQSTM1 mice were viable for at least 32 weeks of age. ing and counted the number of large neurons in the an- Throughout the experimental periods, body weights of terior horn of lumbar spinal cord. In this analysis, mice H46R SQSTM1 mice tended to be lower than those of WT with four different genotypes; WT, SQSTM1, SOD1 , H46R mice in female (Fig. 1a). Body weights of female and SQSTM1;SOD1 mice at 16 and 22 weeks of age, H46R H46R H46R SQSTM1;SOD1 mice were significantly lower than and at end-stage (SOD1 and SQSTM1;SOD1 )or H46R those of SOD1 mice at 5, 8–13, 16–25, and 28 weeks of age (WT and SQSTM1), were used. The num- 28 weeks of age (Fig. 1a). In male, body weights of bers of large Nissl-positive neurons were progressively H46R H46R H46R SQSTM1;SOD1 mice were significantly lower than decreased both in SOD1 and SQSTM1;SOD1 H46R those of SOD1 mice at 25 and 26 weeks of age mice compared to WT and SQSTM1 mice (Fig. 2 and (Fig. 1b). These results indicate that overexpression of Additional file 2:Fig.S2a). However, therewereno H46R SQSTM1 in SOD1 mice accelerates weight loss. significant differences in the numbers of Nissl positive large neurons between groups both at 22 weeks of Overexpression of SQSTM1 shortens lifespans in ageand theend-stage,suggestingthatdecrease in the H46R SOD1 mice number of motor neurons did not explain the reasons In thepreviousstudy,wehaveshown that loss of SQSTM1 why overexpression of SQSTM1 not only accelerates H46R significantly shortens lifespans in SOD1 mice [25]. Fur- the onset of disease but also slows survival after the H46R ther, it has been reported that loss of SQSTM1 shortens onset in SOD1 mice. lifespan while overexpression of it prolongs lifespan in SBMA mice [24]. To clarify the effect of overexpression of Accumulation of ubiquitin/SQSTM1-positive aggregates H46R SQSTM1 on the onset and progression in an ALS mouse and the activation of glial cells in SOD1 -expressing H46R model (SOD1 mice), we performed lifespan analysis mice H46R H46R using SOD1 and SQSTM1;SOD1 mice. Since there SQSTM1 acts as an adaptor protein that directly binds were no gender differences of lifespans in B6 congenic to poly-ubiquitinated proteins destined for degradation H46R SOD1 mice [27], we used the gender-combined data. in autophagy [10]. It has been reported that accumula- Kaplan-Meier survival analysis revealed that the lifespan of tion of ubiquitin- and SQSTM1-positive aggregates is H46R SQSTM1;SOD1 mice (179.3 ± 2.6 days) was signifi- observed not only in ALS patients but also in a H46R H46R cantly shorter than that of SOD1 mice (189.3 ± 2.2 days) SOD1 ALS mouse model [22], and that proliferation (Fig. 1c;Log-ranktest; p = 0.0013). of astrocytes and the activation of microglia also occur H46R in SOD1 mice as disease progresses. To clarify the Overexpression of SQSTM1 accelerates onset of disease effect of SQSTM1 overexpression on the localization H46R but slightly extends post-onset survival in SOD1 of ubiquitin-/SQSTM1-positive aggregates and the ac- mice tivation of glial cells, we conducted immunohisto- Previously, it has been reported that onset of disease chemical analysis of ubiquitin and SQSTM1 in the H46R can be estimated by the turning-point of the body lumbar cord (L4–5) from WT, SQSTM1, SOD1 , H46R weight from gain to loss in mutant SOD1-tg mice and SQSTM1;SOD1 mice at 16 and 22 weeks of H46R H46R [27–29]. Growth curve analysis revealed that onset of age, and at end-stage (SOD1 and SQSTM1;SOD1 ) H46R disease in SQSTM1;SOD1 mice (21.5 ± 2.1 weeks) or 28 weeks of age (WT and SQSTM1)(Fig. 3 and H46R was significantly earlier than those of SOD1 mice Additional file 2: Figure S2b). We also conducted Mitsui et al. Molecular Brain (2018) 11:30 Page 6 of 16 H46R Fig. 1 Analysis of lifespan, disease onset, and progression in SQSTM1;SOD1 mice. a Growth curve in female mice with four different genotypes H46R H46R [wild-type (WT; n =8–12), SQSTM1 (SQSTM1; n =9–12), SOD1 (H46R; n =2–12), and SQSTM1;SOD1 (SQSTM1;H46R; n =1–12)]. b Growth curve in male mice with four different genotypes [WT; n =4–12, SQSTM1; n =11–12, H46R; n =2–12, and SQSTM1;H46R; n =4–12]. a, b Body weight of H46R H46R SQSTM1;SOD1 mice was significant lower than that of SOD1 mice (*p <0.05, **p < 0.01, ***p <0.001, ****p < 0.0001). There are no significantly differences in body weight between WT and SQSTM1 mice. Values were mean ± s.d.. Statistical significance was evaluated by one-way ANOVA with H46R H46R Bonferroni’s post hoc test. c Lifespan analysis of SOD1 mice (H46R; n = 31) and SQSTM1;SOD1 mice (SQSTM1;H46R; n = 28). Kaplan-Meier analysis H46R H46R demonstrated that lifespan in SQSTM1;SOD1 mice (mean ± s.d.; 179.3 ± 2.6 days) was significantly shorter than that in SOD1 mice (189.3 ± H46R H46R 2.2 days) (Log-rank test; p = 0.0013). d Onset of the disease in SOD1 (H46R; n =24) and SQSTM1;SOD1 (SQSTM1;H46R; n = 24) mice. Onset of H46R disease is defined as the turning-point of body weight successive reduction. Onset in SQSTM1;SOD1 mice (mean ± s.d.; 21.5 ± 2.1 weeks) was H46R significantly earlier than that in SOD1 mice (23.8 ± 2.1 weeks) (Log-rank test; p =0.0007). e Survival after the onset of disease (post-onset survival H46R H46R interval) in SOD1 (H46R; n =24) and SQSTM1;SOD1 (SQSTM1;H46R; n = 24) mice. Survival after the onset is defined as the period between the H46R onset of disease and end-point. Post-onset survival interval in SQSTM1;SOD1 mice (mean ± s.d.; 3.9 ± 1.3 weeks) was significantly longer than that in H46R SOD1 mice (3.0 ± 1.1 weeks) (Log-rank test; p =0.0314) Mitsui et al. Molecular Brain (2018) 11:30 Page 7 of 16 H46R H46R Table 1 Summary of survival and onset point analysis in SOD1 and SQSTM1;SOD1 mice Genotype Number End point (weeks) s.d. Onset (weeks) s.d. Post-onset survival interval (weeks) s.d. H46R SOD1 24 26.7 1.5 23.8 2.1 3.0 1.1 H46R SQSTM1;SOD1 24 25.4 1.5 21.5 2.1 3.9 1.3 Number represents the number of mice used in survival and onset analyses. Values represent mean ± s.d. (weeks) H46R H46R immunohistochemistry for microtubule-associated protein cord of SOD1 -expressing mice (SOD1 and H46R 2 (MAP2), glial fibrillary acidic protein (GFAP), and ion- SQSTM1;SOD1 ) as disease progressed (Fig. 3 and Add- ized calcium binding adaptor molecule 1 (Iba1) using mice itional file 2: Figure S2b). It was also noted that a majority of H46R with four different genotypes at end-stage (SOD1 and the ubiquitin-positive signals was colocalized with H46R SQSTM1;SOD1 ) or 28 weeks of age (WT and SQSTM1 (Figs. 3 and 5a). Further, SQSTM1-HA ag- SQSTM1). MAP2, GFAP, and Iba1 were used as cellular gregates were immunohistochemically detected in H46R markers for neuronal soma and dendrite, astrocytes, and SQSTM1;SOD1 mice at 22 weeks and end-stage microglia, respectively (Figs. 4 and 5, and Additional file 2: by using the anti-HA-tag antibody (Additional file 3: Figure S2c). Figure S3), indicating that overexpressing SQSTM1 Double immunostaining of ubiquitin and SQSTM1 re- was also progressively accumulated in the spinal cord vealed that a progressive accumulation of ubiquitin- and as was endogenous SQSTM1. However, there were no SQSTM1-positive aggregates was observed in the spinal apparent differences in the amount and/or number of Fig. 2 Quantitative analysis of motor neurons in the anterior horn of lumbar spinal cord. a Representative images for fluorescence Nissl staining in the anterior horn. Nissl positive neurons were detected in the anterior horn of lumbar spinal cord (L4–5) from the mice with four different H46R H46R genotypes; wild-type (WT), SQSTM1, SOD1 (H46R), and SQSTM1;SOD1 (SQSTM1;H46R) at 16 and 22 weeks of age (wk), and at end-stage (H46R and SQSTM1;H46R) or at 28 wk. (WT and SQSTM1). Cells covering the area of over 240 square pixels (px), corresponding to large motor neurons, are surrounded by red circles, and counted. Scale bars = 50 μm. b Number of large Nissl-positive neurons in the anterior horn of lumbar spinal cord. Vertical axis represents the cumulative number of motor neurons identified in 5 representative Nissl staining images of every tenth H46R H46R serial sections from each sample (n =3–4, SOD1 and SQSTM1;SOD1 mice at 22 wk. and end-stage; n =4, others; n =3). Values are mean ± s.e.m.. Statistical significance was evaluated by two-way ANOVA with Bonferroni’spost hoc test (*p < 0.05, **p < 0.01) Mitsui et al. Molecular Brain (2018) 11:30 Page 8 of 16 Fig. 4 Extensive accumulation of ubiquitin-positive aggregates in the H46R Fig. 3 Progressive accumulation of ubiquitin/SQSTM1-positive spinal cord from SOD1 -expressing mice at end-stage. a-c aggregates in the anterior horn of lumbar spinal cord. a-c Representative images of double immunostaining with MAP2 (green; Representative images of double immunostaining with Ubiquitin neuron marker) and Ubiquitin (red) (a), Ubiquitin (green) and GFAP (red; (green) and SQSTM1 (red) in the lumbar spinal cord (L4–5) from wild- astrocyte marker) (b), Ubiquitin (green) and Iba1 (red; microglia marker) H46R H46R type (WT), SQSTM1 (SQSTM1), SOD1 (H46R) and SQSTM1;SOD1 (c) in the lumbar spinal cord (L4–5) from wild-type (WT) and SQSTM1 H46R H46R (SQSTM1;H46R) mice at 16 weeks of age (wk) (a), 22 wk. (b), and end- (SQSTM1) at 28 weeks of age, SOD1 (H46R) and SQSTM1;SOD1 stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1) (c). The (SQSTM1;H46R) mice at end-stage. The nuclei were counterstained with nuclei were counterstained with DAPI (blue). Scale bars = 50 μm. DAPI (blue). Scale bars = 50 μm. b Arrows indicate colocalization of Ubiquitin-positive aggregates and SQSTM1 aggregates were observed ubiquitin-positive aggregates with GFAP-positive astrocytes. c Arrows H46R H46R in the anterior horn of SOD1 (H46R) and SQSTM1;SOD1 indicate colocalization of ubiquitin-positive aggregates with (SQSTM1;H46R) mice at 22 wk. and end-stage. Arrows indicate Iba1-positive microglia ubiquitin-positive aggregates that are colocalized with SQSTM1 aggregates were localized to astrocytes (Figs. 4b and 5c; H46R ubiquitin-/SQSTM1-positive aggregates between SOD1 arrows). Activation of Iba1-positive microglia was also H46R H46R and SQSTM1;SOD1 mice. observed in SOD1 -expressing mice (Fig. 4c and Double immunostaining of ubiquitin and MAP2 re- Additional file 2: Figure S2c). Ubiquitin-positive aggre- vealed that ubiquitin-positive aggregates were not gates that were surrounded by microglia were occasionally present in MAP2-positive neuronal cell bodies or den- observed (Fig. 5d; arrows). However, there were no differ- H46R drites in the lumbar cord of SOD1 -expressing mice ences in the levels of glial cell activation between H46R H46R (Figs. 4a and 5b and Additional file 2: Figure S2c). On SOD1 and SQSTM1;SOD1 mice. the other hand, double immunostaining of ubiquitin Together, these results indicate that ubiquitin-/ and GFAP demonstrated that the number of SQSTM1-positiveaggregatesweremainlylocalized to GFAP-positive astrocytes was significantly increased in neuropil and occasionally to astrocytes and/or micro- H46R SOD1 -expressing mice (Fig. 4b and Additional file 2: glia but not to the cell body of neurons at the H46R Figure S2c), and that small numbers of ubiquitin-positive end-stage of SOD1 -expressing mice. However, Mitsui et al. Molecular Brain (2018) 11:30 Page 9 of 16 soluble and insoluble proteins from the spinal cord of mice with four different genotypes; WT, SQSTM1, H46R H46R SOD1 , and SQSTM1;SOD1 mice at 16 and H46R 22 weeks of age, and at end-stage (SOD1 and H46R SQSTM1;SOD1 ) or at 28 weeks of age (WT and SQSTM1) and performed western blot analysis, followed by the quantification of signal intensities. Although a progressive accumulation of insoluble high-molecular weight SOD1 (SOD1_HMW) was ob- H46R H46R served both in SOD1 and SQSTM1;SOD1 mice as disease progressed (Fig. 6a and Additional file 4: Figure S4c), quantitative analyses showed no signifi- cant differences of the amount of insoluble SOD1_HMW H46R and monomer SOD1 (SOD1_mono) between SOD1 H46R and SQSTM1;SOD1 mice (Fig. 6c). Further, we de- tected misfolded SOD1 molecules using anti-human misfolded SOD1 antibody (C4F6) [26]. Misfolded H46R monomeric SOD1 was observed both in SOD1 H46R and SQSTM1;SOD1 mice, but not in WT and SQSTM1 mice (Fig. 6b). Quantitative analyses showed that both soluble and insoluble misfolded SOD1 in H46R SQSTM1;SOD1 mice at 16 weeks of age were signifi- H46R cantly higher than those in SOD1 mice, while there were no differences between two groups at the later stages of disease (22 weeks of age and end-stage) (Fig. 6c and Additional file 5:FigureS5). Both soluble and insoluble SQSTM1 were significantly H46R and progressively increased in SOD1 and H46R SQSTM1;SOD1 mice as disease progressed (Fig. 7a, b and Additional file 4: Figure S4d and S4e). Compared to H46R SOD1 mice, the accumulated levels of insoluble H46R SQSTM1 in SQSTM1;SOD1 mice were much higher at 16 and 22 weeks of age, and at end-stage (Fig. 7b). Fig. 5 High-magnification images for the distribution of ubiquitin- To distinguish between mouse endogenous and H46R positive aggregates in the spinal cord from SOD1 -expressing transgene-derived SQSTM1, we performed western blot mice. a-d Representative images of double immunostaining with analysis using anti-HA antibody, and revealed that Ubiquitin (green) and SQSTM1 (red) (a), MAP2 (green; neuron transgene-encoded SQSTM1-HA was also progressively marker) and Ubiquitin (red) (b), Ubiquitin (green) and GFAP (red; H46R accumulated as insoluble forms in SQSTM1;SOD1 astrocyte marker) (c), Ubiquitin (green) and Iba1 (red; microglia H46R marker) (d) in the lumbar cord (L4–5) from SOD1 (H46R) and mice (Fig. 7a). H46R SQSTM1;SOD1 (SQSTM1;H46R) mice at end-stage. The nuclei were Accumulation of insoluble poly-ubiquitinated proteins counterstained with DAPI (blue). Scale bars = 10 μm. c Arrows that were represented by smear signals was observed in indicate that ubiquitin-positive aggregates are present in GFAP- H46R SOD1 -expressing mice (Fig. 7a and Additional file 4: positive astrocytes. d Arrows indicate that ubiquitin-positive Figure S4 g). Although poly-ubiquitinated proteins de- aggregates are surrounded by microglia H46R tected in the spinal cord of SQSTM1;SOD1 mice H46R were comparable to those in SOD1 mice at 22 weeks H46R there were no discernible differences in the distribu- of age, those in SQSTM1;SOD1 mice at end-stage H46R tion and the degree of colocalization of ubiquitin-positive were significantly higher than that in SOD1 mice H46R aggregates to glial cells between SQSTM1;SOD1 and (Fig. 7a and c). H46R SOD1 mice. These results suggest that overexpression of SQSTM1 accelerates the accumulation not only of Overexpression of SQSTM1 increases insoluble poly- misfolded SOD1 at the earlier stage, but also of in- H46R ubiquitinated proteins in the spinal cord of SOD1 mice soluble poly-ubiquitinated proteins at the later stage H46R To biochemically investigate the accumulation of SOD1, of disease in the spinal cord of SOD1 -expressing SQSTM1, and ubiquitin, we extracted Triton X-100 mice. Mitsui et al. Molecular Brain (2018) 11:30 Page 10 of 16 H46R Fig. 6 Overexpression of SQSTM1 increases misfolded SOD1 in the spinal cord of SOD1 -expressing mice. a Western blot analysis of SOD1. b Western blot H46R H46R analysis of misfolded SOD1. a, b The spinal cord samples were obtained from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1). Two fractions; 1% Triton X-100 soluble and 1% Triton X-100 insoluble/5%SDS soluble fractions were analyzed. SOD1_mono and SOD1_HMW represent monomeric and high molecular- weight forms of SOD1, respectively. SOD1(C4F6) represents misfolded SOD1. GAPDH and β-actin (Actin) were used for a loading control in Triton X-100- soluble and -insoluble fractions, respectively. c Quantification of 1% Triton X-100 soluble SOD1, insoluble SOD1_mono, insoluble SOD1_HMW, soluble misfolded SOD1, and insoluble misfolded SOD1. Values are mean ± s.e.m. (n = 4) in an arbitrary unit relative to H46R mice at 16 weeks of age. Signal intensities were normalized by the levels of GAPDH (soluble fractions) and Actin (insoluble fractions). Statistical significance was evaluated by two-way ANOVA with Bonferroni’s post hoc test (comparisons between different genotypes in the same age; *p < 0.05, **p < 0.01) Overexpression of SQSTM1 increases insoluble of SQSTM1 overexpression on autophagy and H46R H46R phosphorylated SQSTM1 in the spinal cord of SOD1 anti-oxidative stress responses in SOD1 mice, we per- mice formed western blot analysis using antibodies to specific- Previous studies have shown that phosphorylation of ally detect Ser403(human)/Ser405(mouse)-phosphorylated SQSTM1 at Ser403 regulates degradation of SQSTM1, Ser349(human)/Ser351(mouse)-phosphorylated poly-ubiquitinated proteins by autophagy [14], and that SQSTM1, MAP1LC3/LC3, and NAD(P)H dehydrogenase protein aggregates containing Ser403-phosphorylated [quinone] 1 (NQO1) in the spinal cord of WT, SQSTM1, H46R H46R SQSTM1 are enriched in the brain of ALS patients [30]. SOD1 ,and SQSTM1;SOD1 mice at 16 and H46R Further, phosphorylation of SQSTM1 at Ser351 increases 22 weeks of age, and at end-stage (SOD1 and H46R the binding of KEAP1 to SQSTM1, thereby activating the SQSTM1;SOD1 ) or at 28 weeks of age (WT and NFE2L2/Nrf2 mediated anti-oxidative stress pathway [16]. SQSTM1). These phosphorylation sites are conserved between hu- The levels of insoluble Ser403/Ser405-phosphorylated H46R man and mouse and can be detected by specific SQSTM1 were increased both in SOD1 and H46R anti-phospho-SQSTM1 antibodies. To elucidate the effect SQSTM1;SOD1 mice at 22 weeks of age and Mitsui et al. Molecular Brain (2018) 11:30 Page 11 of 16 H46R Fig. 7 Overexpression of SQSTM1 increases the insoluble poly-ubqiuitinated proteins in the spinal cord of SOD1 -expressing mice. a Western blot H46R analysis of SQSTM1, SQSTM1-HA (HA) and Ubiquitin (Ub) in the spinal cord from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and H46R SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1). Two fractions; 1% Triton X-100 soluble and 1% Triton X-100 insoluble/5%SDS soluble fractions were analyzed. SOD1_mono and SOD1_HMW represent monomeric and high molecular-weight forms of SOD1, respectively. Ub_mono and Ub_HMW represent monomeric ubiquitin and poly-ubiquitinated proteins, respectively. GAPDH and β-actin (Actin) were used for a loading control in Triton X-100-soluble and -insoluble fractions, respectively. Since the blots used in these western analyses were same as those in Fig. 6a, same images for GAPDH and Actin (asterisks) in Fig. 6a were used again as references. b Quantification of soluble and insoluble SQSTM1. c Quantification of soluble and insoluble poly-ubiquitinated proteins (Ub_HMW). b, c Values are mean ± s.e.m. (n = 4) in an arbitrary unit relative to WT mice at 16 weeks of age. Signal intensities were normalized by the levels of GAPDH (soluble fractions) and Actin (insoluble fractions). Statistical significance was evaluated by two-way ANOVA with Bonferroni’spost hoc test (comparisons between different genotypes in the same age; *p < 0.05, **p < 0.01, ***p <0.001, ****p <0.0001) H46R H46R end-stage compared to WT and SQSTM1 mice. How- SQSTM1;SOD1 , but not in SOD1 , mice as dis- ever, there were no differences in the amount of Ser403/ ease progressed (Fig. 8a, c and Additional file 6:Fig.S6d). H46R Ser405-phosphorylated SQSTM1 between SOD1 The levels of LC3-II in insoluble fraction were pro- H46R H46R and SQSTM1;SOD1 mice, despite that the total gressively increased in SOD1 -expressing mice amount of SQSTM1 was progressively increased at the (Fig. 8a and Additional file 6:FigureS6f). Thequan- end-stage (Figs. 7a, b and 8a, b, and Additional file 6: titative analysis of the LC3-II to LC3-I ratio (LC3-II/ Figure S6b). Remarkably, Ser349/Ser351-phosphorylated LC3-I), a marker of the autophagosome formation, SQSTM1 was specifically detected in insoluble but not demonstrated that increase in the levels of LC3-II/ H46R in soluble fractions. Further, insoluble Ser349/Ser351-- LC-I was more prominent in SOD1 mice com- H46R phosphorylated SQSTM1 was significantly increased in pared to SQSTM1;SOD1 mice (Fig. 8a and d). Mitsui et al. Molecular Brain (2018) 11:30 Page 12 of 16 H46R Fig. 8 Accumulation of phosphorylated SQSTM1 in the spinal cord of SOD1 -expressing mice. a Western blot analysis of Ser403(human)/ Ser405(mouse)-phosphorylated SQSTM1 (p-S403/405), Ser349(human)/Ser351(mouse)-phosphorylated SQSTM1 (p-S349/351), LC3, and NQO1 in H46R H46R the spinal cord from mice with four different genotypes; wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1). Two fractions; 1% Triton X-100 soluble and 1% Triton X-100 insoluble/5%SDS soluble fractions were analyzed. GAPDH and β-actin (Actin) were used for a loading control in Triton X-100-soluble and -insoluble fractions, respectively. To properly compare the intensities of signals between phosphorylated SQSTM1 and total SQSTM1, the results from same blots for SQSTM1 in Fig. 7a, GAPDH, and Actin (asterisks) in Fig. 6a are used again as references. b Quantification of 1% Triton X-100 soluble and insoluble S403/S405-phosphorylated SQSTM1. c Quantification of insoluble S349/S351- phosphorylated SQSTM1. d Quantification of the LC3-II to LC3-I ratio (LC3-II/LC3-I) in insoluble fraction. e Quantification of soluble NQO1. b-e Values are mean ± s.e.m. (n = 4) in an arbitrary unit relative to WT mice at 16 weeks of age. Signal intensities were normalized by the levels of GAPDH (soluble fractions) and Actin (insoluble fractions). Statistical significance was evaluated by two-way ANOVA with Bonferroni’s post hoc test (comparisons between different genotypes in the same age; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001) On the other hand, the levels of NQO1, which was conditions in which the levels of Ser349/Ser351-phosph- one of the NFE2L2/Nrf2 target proteins, were af- lylated SQSTM1 is upregulated. It is thus reasonable to fected neither by overexpression of mutant SOD1, assume that phosphorylated SQSTM1 molecules present SQSTM1, nor both of them (Fig. 8a and e,and Add- in the insoluble fraction might be biologically inactive itional file 6:FigureS6g). and are present as inert insoluble aggregates in the H46R These results suggest that overexpression of SQSTM1 spinal cord of SOD1 mice. affects the autophagy activation and/or the impairment of autophagic degradation of LC3-II positive autophago- Discussion somes, whereas it has no impact on the NFE2L2/Nrf2 We have previously shown that loss of SQSTM1 exacer- H46R mediated anti-oxidative stress pathway even in bates disease phenotypes in SOD1 mice [25]. Further, Mitsui et al. Molecular Brain (2018) 11:30 Page 13 of 16 in a SBMA mouse model, loss of SQSTM1 also acceler- proteins [10]. Recently, it has been demonstrated that ates the onset of disease, while overexpression of it ame- the crossbreeding of conditional Atg7 knockout mice G93A liorates disease phenotypes [24]. These findings suggest a with SOD1 mice remarkably extends the lifespan, protective role of SQSTM1 in motor neuron degeneration despite that the onset of hindlimb tremor is acceler- in vivo. In this study, we investigated the effect of systemic ated [33]. Further, the pharmacological induction of overexpression of SQSTM1 on the disease onset and pro- mTOR-independent autophagy has failed to slow the H46R G93A gression in SOD1 mice. Contrary to our expectation, disease progression of SOD1 mice [34]. More im- we revealed that systemic overexpression of SQSTM1 portantly, motor neuron-specific disruption of autoph- resulted in earlier onset and shorter lifespan in a agy displays no phenotypic changes in ALS mice, H46R SOD1 -expressing ALS mouse model. while proteasome inhibition leads to ALS-associated It has been reported that phosphorylation at Ser349 pathology and paralysis [35]. We here showed that H46R (corresponding to mouse Ser351) of SQSTM1 enhances misfolded SOD1 in SQSTM1;SOD1 mice at the binding of SQSTM1 to KEAP1 which in turn acti- 16 weeks of age were significantly higher than those H46R vates the NFE2L2/Nrf2 pathway and NFE2L2/Nrf2 me- in SOD1 mice (Fig. 6c). We also demonstrated diated anti-oxidative stress responses in mice [16]. On that the extensive accumulation of insoluble SQSTM1 H46R the other hand, the previous study on NFE2L2/Nrf2 de- occurred in SQSTM1;SOD1 mice even prior to H46R ficient SOD1 mice has revealed that loss of NFE2L2/ the onset of disease; i.e., at 16 weeks of age (Fig. 7a Nrf2 affects neither the onset of disease nor the progres- and b). Although phosphorylation of human SQSTM1 sion of disease [25]. In this study, despite that the accu- at Ser403 was known to induce autophagic degradation mulation of Ser349/Ser351-phosphorylated SQSTM1 of ubiquitinated proteins [14], the levels of Ser403 (corre- H46R was increased in SQSTM1;SOD1 mice at 22 weeks sponding to mouse Ser405)-phosphorylated SQSTM1 be- H46R H46R of age and end-stage (Fig. 8a and c), there were no dif- tween SQSTM1;SOD1 and SOD1 mice were most ferences in the expression levels of NQO1, which was comparable (Fig. 8a and b). Further, the ratio of LC3-II/ H46R one of the NFE2L2/Nrf2 downstream targets, between LC3-I in SQSTM1;SOD1 mice was lower than that of H46R H46R H46R SOD1 and SQSTM1;SOD1 mice. Further, im- SOD1 mice (Fig. 8a and d) at onset stage; i.e., 22 week munohistochemical analysis revealed that ubiquitin/ of age. These results combined suggest that overexpres- SQSTM1-positive aggregates were not localized in the sion of SQSTM1 does not activate autophagy, but rather cell body and dendrite of neurons but mainly to accelerates the accumulation of SQSTM1 by possibly hin- H46R neuropil (Figs. 4 and 5), indicative of dysfunctional dering the UPS and autophagic degradation in SOD1 SQSTM1 molecules in insoluble aggregates. Taken to- mice. Since the accumulation of autophagosome-like gether, we speculate that progressively-increased structures and SQSTM1-positive aggregates are among Ser349/Ser351-phosphorylated SQSTM1 is physiolo- the cardinal pathological features in patients with ALS as gically inactive, and thus the NFE2L2/Nrf2 anti-oxidative well as in murine ALS models [18, 23], it is reasonable to stress pathway plays a minor role both in the disease onset assume that overexpression of SQSTM1 enhances mutant and progression. SOD1-associated neurotoxicity by further compromising We suppose three possible reasons as to why overex- proteostasis, thereby leading to an earlier onset of disease. pression of SQSTM1 accelerates the onset of disease in On the other hand, SBMA-linked toxic entities; e.g., H46R SOD1 mice, unlike in the case of a SBMA mouse polyglutamine-mediated transcriptional dysregulation model [24]. One possibility is that protein homeostasis [24], may play an indirect role in the regulation of the H46R (proteostasis) in the spinal cord of SOD1 mice is UPS and/or autophagy, and thus the SQSTM1-associated more severely compromised than those in SBMA mice beneficial effect becomes more tangible in SBMA mice. under conditions in which SQSTM1 is overexpressed. A Second possibility is that SQSTM1 expressing in cells large number of studies has revealed that the accumula- and tissues other than neurons affects the onset of dis- H46R tion of aggregated proteins is one of the pathological ease in SOD1 mice. Evidences showing that, in skel- G93A features of neurodegenerative diseases. Misfolded ubi- etal muscle of a SOD1 -expressing ALS mouse quitinated proteins are preferentially degraded by the model, the UPS is activated prior to the onset of disease, ubiquitin-proteasome system (UPS), but undegradable while the autophagic degradation pathway is activated misfolded proteins by the UPS form aggregates, which after the onset of disease till at end-stage [36], sug- in turn are degraded by autophagy [31]. SQSTM1 acts gest a possible pathogenic interaction between pro- as an autophagy adaptor for the degradation of ubiquiti- teostasis in skeletal muscles and neurodegeneration. nated proteins [11]. Thus, in addition to the UPS, autoph- Since the expression level of transgene-coded agy is the important intracellular degradation pathway SQSTM1 in skeletal muscle is higher than those in to maintain proteostasis [32], and its impairment re- other tissues (Additional file 1: Figure S1b), it is possible H46R sults in the accumulation of SQSTM1 and ubiquitinated that the earlier onset of disease in SQSTM1;SOD1 Mitsui et al. Molecular Brain (2018) 11:30 Page 14 of 16 mice is linked to dysregulation of autophagy by SQSTM1 assume that such glial contribution may not play a overexpression in skeletal muscles. However, even if that majorrole intheextensionofsurvivalafter theonset. would be the case, the underlying mechanism has yet to On theother hand,wehavepreviouslyreportedthat H46R be investigated. loss of SQSTM1 in SOD1 mice decreases the Third possibility is that body weight at pre-symptomatic biochemically-detectable insoluble ubiquitinated pro- H46R healthy stages affects the onset of disease in SOD1 teins in the spinal cord in parallel with the accumula- mice. It has been shown that a high fat food feeding tion of ubiquitin-positive inclusions in somal area of in a mutant TAR DNA-binding protein 43-kD motor neurons [25]. Since loss of SQSTM1 exacerbates H46R (TARDBP)-expressing ALS mouse model extends their disease symptoms in SOD1 mice, we consider that the lifespans [37], and that the activation of AMP-activated accumulation of biochemically-detectable insoluble ubi- protein kinase (AMPK), which functions as an energy sen- quitinated proteins is beneficial and/or protective rather sor, results in the extension of lifespans in a mutant than harmful to motor neurons, while the intrasomal SOD1-expressing ALS mouse model [38]. These studies ubiquitin-positive inclusions are poisonous [25]. Indeed, on ALS and energy metabolism suggest that nutritional we showed that biochemically-detectable insoluble conditions affect the disease phenotypes in ALS animal ubiquitinated proteins were increased in the spinal H46R models. Interestingly, we revealed that body weight of cord of SQSTM1;SOD1 mice at end-stage (Fig. 7a SQSTM1 mice tended to be lower than that of WT mice and c). Further, immunohistochemistry revealed that (Fig. 1a and b), and that body weight of female ubiquitin-positive inclusions were exclusively observed H46R SQSTM1;SOD1 mice was significantly lower than that in neuropil but not in somal area of motor neurons H46R H46R of SOD1 mice throughout their pre-symptomatic of SQSTM1;SOD1 mice (Figs. 4 and 5). Thus, the period (Fig. 1a). Taken together, it is possible that the re- effects of loss and overexpression of SQSTM1 on duction of body weight that is associated with SQSTM1 these factors are obviously opposite to each other. overexpression causes the earlier onset of disease in Taken together, it is still possible that the insolubiliza- H46R SOD1 mice. Since loss of SQSTM1 in mice causes hy- tion of poly-ubiquitinated proteins and the exclusion perphagia that results in obese phenotypes [39], a systemic of ubquitinated protein aggregates from neuronal cell overexpression of SQSTM1 may conversely induce hypo- body by SQSTM1 are implicated in some neuropro- phagia thereby decreasing the body weight in mice. In any tective roles, particularly, at the stage of disease pro- cases, pathogenic mechanisms by which overexpression of gression, thereby slowing disease progression by H46R H46R SQSTM1 accelerates the onset in a SOD1 -expressing SQSTM1 overexpression in SOD1 mice. However, ALS mouse model remain unknown, and thus further in order to conclude as such, further studies that in- studies are required. cludes other methods to determine the onset of dis- One of the interesting findings in this study is that ease, such as grip power, the hind-limb reflex, rotarod survival after the onset is slightly extended in retention time, and balance-beam test will be H46R SQSTM1;SOD1 mice, despite that these mice show required. earlier onset of disease (Fig. 1). However, we are cur- rently considering that, although the extension of Conclusions post-onset survival reaches to a statistically significant In conclusions, we demonstrated that, unlike in SBMA level, its reliability may still not be high enough to con- mice, SQSTM1 served as an onset accelerating factor for H46R clude that overexpression of SQSTM1 definitely extends SOD1-linked toxicities, at least, in a SOD1 -expres- survival, because the prediction of disease onset is solely sing ALS mouse model, implying that SQSTM1 upregu- determined by a single measure; i.e., body weight lation might not always to be beneficial to the treatment changes, in this study. Previously, it has been shown that of ALS. mutant SOD1-mediated toxicities in neurons determine the disease onset, whereas those in astrocytes or micro- Additional files glia regulate the disease progression in ALS mouse models [29, 40]. Based on these findings, it could be hy- Additional file 1: Figure S1. Schema of the transgene construct for SQSTM1-tg mouse and the distribution of SQSTM1 in tissues. a Schematic pothesized that mutant SOD1-mediated toxicities in glial diagram of the transgene cassette encoding the C-terminally cells relative to neurons are decreased by systemic overex- hemagglutinin (HA)-tagged human SQSTM1 cDNA. Transgene construct H46R pression of SQSTM1 in SOD1 mice. Although quanti- consists of cytomegalovirus enhancer (E), chicken β-actin promoter (Pro), rabbit β-globin splice acceptor (S), full length human SQSTM1 cDNA tative evaluation of GFAP and Iba1 levels by western (SQSTM1), hemagglutinin tag (HA), and rabbit β-globin poly A (Poly A). b blotting remained to be performed, immunohistochemical Expression and distribution of SQSTM1 in the liver, skeletal muscle, olfactory analysis revealed that the glial cells activation in the spinal bulb, cerebral cortex, hippocampus, cerebellum, and spinal cord in wild-type H46R H46R (WT) and SQSTM1-tg (SQSTM1) mice. Expression of mouse endogenous cord between SOD1 and SQSTM1;SOD1 mice SQSTM1 and human SQSTM1 (Endogenous + SQSTM1-HA) was simultaneously were almost comparable (Figs. 4 and 5). Thus, it is fair to Mitsui et al. Molecular Brain (2018) 11:30 Page 15 of 16 PBS: Phosphate-buffered saline; PFA: Paraformaldehyde; PVDF: Polyvinylidene detected by western blotting using anti-SQSTM1 antibody. Values shown under difluoride; RT: Room temperature; SBMA: Spinal and bulbar muscular atrophy; each lane were the signal intensities in an arbitrary unit relative to that of the SDS: Sodium dodecyl sulfate; SQSTM1: Sequestosome 1/p62; TARDBP: TAR liver in WT mouse. Human SQSTM1 (hSQSTM1-HA) was specifically detected by DNA-binding protein 43-kD; tg: Transgenic; UPS: Ubiquitin-proteasome anti-HA antibody. GAPDH was used as a loading control. (PDF 795 kb) system; WT: Wild-type Additional file 2: Figure S2. Whole transverse sectional images of the lumbar spinal cord in Figs. 2, 3 and 4. a Whole transverse sectional images of Acknowledgments fluorescence Nissl staining in the lumbar spinal cords in Fig. 2. b Whole We thank Dr. Makoto Urushitani at Shiga University of Medical Science for transverse sectional images of double immunostaining with Ubiquitin (green) the generous gift of anti-misfolded SOD1 antibody (C4F6), and all the mem- andSQSTM1(red)in thelumbarspinalcordinFig. 3. c Whole transverse bers of Support Center for Medical Research and Education at Tokai Univer- sectional images of double immunostaining with MAP2 (green) and Ubiquitin sity for their technical help. (red), Ubiquitin (green) and GFAP (red), and Ubiquitin (green) and Iba1 (red) in the lumbar spinal cord in Fig. 4. a-c Scale bars = 300 μm. b, c The nuclei were Funding counterstained with DAPI (blue). (PDF 7238 kb) This study was supported by Japanese Society for Promotion of Science Additional file 3: Figure S3. SQSTM1-HA positive aggregates in the (JSPS) [grant numbers 26290018 and 24650189 to SH], and the National anterior horn of lumbar spinal cord. a-c Representative images of double Natural Science Foundation of China (NSFC) and JSPS Bilateral Joint Research immunostaining with Ubiquitin (green) and HA (SQSTM1-HA; red) in the Project (to FHS and SH). H46R lumbar spinal cord (L4–5) from wild-type (WT), SQSTM1 (SQSTM1), SOD1 H46R (H46R) and SQSTM1;SOD1 (SQSTM1;H46R)miceat16weeks of age (wk) Availability of data and materials (a), 22 wk. (b), and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and The datasets supporting the conclusions of this article is included within the SQSTM1) (c). The nuclei were counterstained with DAPI (blue). Scale article and its Additional files. bars = 50 μm. Ubiquitin-positive aggregates and SQSTM1-HA aggregates were H46R observed in the anterior horn of SQSTM1;SOD1 (SQSTM1;H46R) mice at 22 Authors’ contributions wk. and end-stage. Arrows indicate that ubiquitin-positive aggregates coloca- lizing with SQSTM1-HA. (PDF 1718 kb) Conceived and designed the experiments: SM, AO, FHS and SH. Performed the experiments: SM, MN, SO, KS, RS and AO. Analyzed the data: SM. Additional file 4: Figure S4. Immunoblot-images used for the quantitative Contributed reagents/materials/analysis tools: HA, MA, GS and SH. Wrote the analysis in Figs. 6a and Fig. 7. The immunoblots of (a)soluble SOD1,(b) paper: SM and SH. All authors reviewed and approved the manuscript. insoluble SOD1 monomer (Mono), (c) insoluble high-molecular weight (HMW) SOD1, (d) soluble SQSTM1, (e) insoluble SQSTM1, (f) soluble poly-ubiquitinated Ethics approval proteins, (g) insoluble poly-ubiquitinated proteins, (h)soluble GAPDH, and(i) All animal experimental procedures were carried out in accord with the insoluble actin were analyzed for the quantitative analysis (Figs. 6a and 7). The H46R Fundamental Guidelines for Proper Conduct of Animal Experiment and spinal cord from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and H46R Related Activities in Academic Research Institutions under the jurisdiction of SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and the Ministry of Education, Culture, Sports, Science and Technology (MEXT), end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1) were Japan, and reviewed and approved by The Institutional Animal Care and Use used. C (control sample) used as internal control indicates soluble and H46R Committee at Tokai University. insoluble fractions from 22 week-old SQSTM1;SOD1 mouse. Asterisk represents the measured area of poly-ubiquitinated proteins. (PDF 1973 kb) Competing interests Additional file 5: Figure S5. Immunoblot-images used for the quantita- The authors declare that they have no competing interests. tive analysis in Fig. 6b. The immunoblots of (a) soluble misfolded SOD1 and GAPDH, (b) insoluble misfolded SOD1 and actin were analyzed for H46R the quantitative analysis (Fig. 6b). The spinal cord from SOD1 (H46R), H46R Publisher’sNote and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 Springer Nature remains neutral with regard to jurisdictional claims in wk., and end-stage were used. (PDF 444 kb) published maps and institutional affiliations. Additional file 6: Figure S6. Immunoblot-images used for the quanti- tative analysis in Fig. 8. The immunoblots of (a)soluble and(b)insoluble Author details Ser403(human)/Ser405(mouse)-phosphorylated SQSTM1 (p-S403/405), (c) Department of Molecular Life Sciences, Tokai University School of Medicine, soluble and (d) insoluble Ser349(human)/Ser351(mouse)-phosphorylated Isehara, Kanagawa 259-1193, Japan. The Institute of Medical Sciences, Tokai SQSTM1 (p-S349/351), (e) soluble LC3, (f) insoluble LC3, (g) soluble NQO1, University, Isehara, Kanagawa 259-1193, Japan. Micro/Nano Technology and (h) insoluble NQO1 were analyzed for the quantitative analysis (Fig. 7). Center, Tokai University, Hiratsuka, Kanagawa 259-1292, Japan. Department H46R The spinal cord from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), of Anesthesiology, Tokai University School of Medicine, Isehara, Kanagawa H46R and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., 259-1193, Japan. Department of Neurology, University of Occupational and and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1) were Environmental Health School of Medicine, Kitakyushu, Fukuoka 807-0804, used. C (control sample) used as internal control indicates soluble and Japan. Department of Neurology, Tohoku University Graduate School of H46R insoluble fractions from 22 week-old SQSTM1;SOD1 mouse. (PDF 607 kb) Medicine, Sendai, Miyagi 980-8575, Japan. Department of Neurology, Nagoya University Graduate School of Medicine, Nagoya, Aichi 466-8550, Japan. Department of Neurology, West China Hospital, Sichuan University, Abbreviations Chengdu 610041, Sichuan, China. Research Center for Brain and Nervous ALS: Amyotrophic lateral sclerosis; AMPK: Adenosine mono-phosphate acti- Diseases, Tokai University Graduate School of Medicine, Kanagawa, Isehara vated protein kinase; ANOVA: Analysis of variance; BPB: Bromophenol blue; 259-1193, Japan. CAG: (C) the cytomegalovirus (CMV) early enhancer element, (A) the promoter, the first exon and the first intron of chicken beta-actin gene, and Received: 12 March 2018 Accepted: 20 May 2018 (G) the splice acceptor of the rabbit beta-globin gene; DAPI: 4′,6-diamidino- 2-phenylindole dihydrochloride; FTD: Frontotemporal dementia; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; GFAP: Glial fibrillary acidic protein; HA: Hemagglutinin; HRP: Horseradish peroxidase; Iba1: Ionized References calcium binding adaptor molecule 1; KEAP1: Kelch-like ECH-associated pro- 1. Robberecht W, Philips T. The changing scene of amyotrophic lateral tein 1; MAP1LC3/LC3: Microtubule-associated protein 1 light chain 3; sclerosis. Nat Rev Neurosci. 2013;14:248–64. MAP2: Microtubule-associated protein 2; MND: Motor neuron disease; 2. Rowland LP, Shneider NA. Amyotrophic lateral sclerosis. N Engl J Med. 2001; NFE2L2/Nrf2: Nuclear factor related erythroid 2-related factor 2; NGS: Normal 344:1688–700. goat serum; NQO1: NAD(P)H dehydrogenase [quinone] 1; 3. Renton AE, Chio A, Traynor BJ. State of play in amyotrophic lateral sclerosis PAGE: Polyacrylamide gel electrophoresis; PB: Phosphate buffer; genetics. Nat Neurosci. 2014;17:17–23. Mitsui et al. Molecular Brain (2018) 11:30 Page 16 of 16 4. Fecto F, Yan J, Vemula SP, Liu E, Yang Y, Chen W, Zheng JG, Shi Y, Siddique 25. Hadano S, Mitsui S, Pan L, Otomo A, Kubo M, Sato K, Ono S, Onodera W, Abe N, Arrat H, et al. SQSTM1 mutations in familial and sporadic amyotrophic K, Chen X, et al. Functional links between SQSTM1 and ALS2 in the lateral sclerosis. Arch Neurol. 2011;68:1440–6. pathogenesis of ALS: cumulative impact on the protection against mutant 5. Hirano M, Nakamura Y, Saigoh K, Sakamoto H, Ueno S, Isono C, Miyamoto K, SOD1-mediated motor dysfunction in mice. Hum Mol Genet. 2016;25:3321–40. Akamatsu M, Mitsui Y, Kusunoki S. Mutations in the gene encoding p62 in 26. Urushitani M, Ezzi SA, Julien JP. Therapeutic effects of immunization with Japanese patients with amyotrophic lateral sclerosis. Neurology. 2013;80: mutant superoxide dismutase in mice models of amyotrophic lateral 458–63. sclerosis. Proc Natl Acad Sci U S A. 2007;104:2495–500. 27. Pan L, Yoshii Y, Otomo A, Ogawa H, Iwasaki Y, Shang HF, Hadano S. 6. Shimizu H, Toyoshima Y, Shiga A, Yokoseki A, Arakawa K, Sekine Y, Different human copper-zinc superoxide dismutase mutants, SOD1G93A Shimohata T, Ikeuchi T, Nishizawa M, Kakita A, et al. Sporadic ALS with and SOD1H46R, exert distinct harmful effects on gross phenotype in mice. compound heterozygous mutations in the SQSTM1 gene. Acta PLoS One. 2012;7:e33409. Neuropathol. 2013;126:453–9. 28. Boillee S, Yamanaka K, Lobsiger CS, Copeland NG, Jenkins NA, Kassiotis G, 7. Chen Y, Zheng ZZ, Chen X, Huang R, Yang Y, Yuan L, Pan L, Hadano S, Kollias G, Cleveland DW. Onset and progression in inherited ALS Shang HF. SQSTM1 mutations in Han Chinese populations with sporadic determined by motor neurons and microglia. Science. 2006;312:1389–92. amyotrophic lateral sclerosis. Neurobiol Aging. 2014;35:726. e7–9 29. Yamanaka K, Chun SJ, Boillee S, Fujimori-Tonou N, Yamashita H, Gutmann 8. Yang Y, Tang L, Zhang N, Pan L, Hadano S, Fan D. Six SQSTM1 mutations in DH, Takahashi R, Misawa H, Cleveland DW. Astrocytes as determinants of a Chinese amyotrophic lateral sclerosis cohort. Amyotroph Lateral Scler disease progression in inherited amyotrophic lateral sclerosis. Nat Neurosci. Frontotemporal Degener. 2015;16:378–84. 2008;11:251–3. 9. Laurin N, Brown JP, Morissette J, Raymond V. Recurrent mutation of the 30. Kurosawa M, Matsumoto G, Sumikura H, Hatsuta H, Murayama S, Sakurai T, gene encoding sequestosome 1 (SQSTM1/p62) in Paget disease of bone. Shimogori T, Hattori N, Nukina N. Serine 403-phosphorylated p62/SQSTM1 Am J Hum Genet. 2002;70:1582–8. immunoreactivity in inclusions of neurodegenerative diseases. Neurosci Res. 10. Komatsu M, Waguri S, Koike M, Sou YS, Ueno T, Hara T, Mizushima N, Iwata J, 2016;103:64–70. Ezaki J, Murata S, et al. Homeostatic levels of p62 control cytoplasmic inclusion 31. Lystad AH, Simonsen A. Assays to monitor aggrephagy. Methods. 2015;75:112–9. body formation in autophagy-deficient mice. Cell. 2007;131:1149–63. 32. Mizushima N, Komatsu M. Autophagy: renovation of cells and tissues. Cell. 11. Pankiv S, Clausen TH, Lamark T, Brech A, Bruun JA, Outzen H, Overvatn A, 2011;147:728–41. Bjorkoy G, Johansen T. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate 33. Rudnick ND, Griffey CJ, Guarnieri P, Gerbino V, Wang X, Piersaint JA, Tapia degradation of ubiquitinated protein aggregates by autophagy. J Biol JC, Rich MM, Maniatis T. Distinct roles for motor neuron autophagy early Chem. 2007;282:24131–45. and late in the SOD1(G93A) mouse model of ALS. Proc Natl Acad Sci U S A. 12. Komatsu M, Kurokawa H, Waguri S, Taguchi K, Kobayashi A, Ichimura Y, Sou 2017;114:E8294–E303. YS, Ueno I, Sakamoto A, Tong KI, et al. The selective autophagy substrate 34. Perera ND, Sheean RK, Lau CL, Shin YS, Beart PM, Horne MK, Turner BJ. p62 activates the stress responsive transcription factor Nrf2 through Rilmenidine promotes MTOR-independent autophagy in the mutant SOD1 inactivation of Keap1. Nat Cell Biol. 2010;12:213–23. mouse model of amyotrophic lateral sclerosis without slowing disease 13. Goode A, Butler K, Long J, Cavey J, Scott D, Shaw B, Sollenberger J, Gell C, progression. Autophagy. 2018;14:534–51. Johansen T, Oldham NJ, et al. Defective recognition of LC3B by mutant 35. Tashiro Y, Urushitani M, Inoue H, Koike M, Uchiyama Y, Komatsu M, Tanaka SQSTM1/p62 implicates impairment of autophagy as a pathogenic K, Yamazaki M, Abe M, Misawa H, et al. Motor neuron-specific disruption of mechanism in ALS-FTLD. Autophagy. 2016;12:1094–104. proteasomes, but not autophagy, replicates amyotrophic lateral sclerosis. J 14. Matsumoto G, Wada K, Okuno M, Kurosawa M, Nukina N. Serine 403 Biol Chem. 2012;287:42984–94. phosphorylation of p62/SQSTM1 regulates selective autophagic clearance of 36. Olivan S, Calvo AC, Gasco S, Munoz MJ, Zaragoza P, Osta R. Time-point ubiquitinated proteins. Mol Cell. 2011;44:279–89. dependent activation of autophagy and the UPS in SOD1G93A mice 15. Matsumoto G, Shimogori T, Hattori N, Nukina N. TBK1 controls skeletal muscle. PLoS One. 2015;10:e0134830. autophagosomal engulfment of polyubiquitinated mitochondria through 37. Coughlan KS, Halang L, Woods I, Prehn JH. A high-fat jelly diet restores p62/SQSTM1 phosphorylation. Hum Mol Genet. 2015;24:4429–42. bioenergetic balance and extends lifespan in the presence of motor 16. Ichimura Y, Waguri S, Sou YS, Kageyama S, Hasegawa J, Ishimura R, Saito T, dysfunction and lumbar spinal cord motor neuron loss in TDP-43A315T Yang Y, Kouno T, Fukutomi T, et al. Phosphorylation of p62 activates the mutant C57BL6/J mice. Dis Model Mech. 2016;9:1029–37. Keap1-Nrf2 pathway during selective autophagy. Mol Cell. 2013;51:618–31. 38. Coughlan KS, Mitchem MR, Hogg MC, Prehn JH. “Preconditioning” with 17. Mizuno Y, Amari M, Takatama M, Aizawa H, Mihara B, Okamoto K. latrepirdine, an adenosine 5′-monophosphate-activated protein kinase Immunoreactivities of p62, an ubiqutin-binding protein, in the spinal activator, delays amyotrophic lateral sclerosis progression in SOD1(G93A) anterior horn cells of patients with amyotrophic lateral sclerosis. J Neurol mice. Neurobiol Aging. 2015;36:1140–50. Sci. 2006;249:13–8. 39. Harada H, Warabi E, Matsuki T, Yanagawa T, Okada K, Uwayama J, Ikeda A, 18. Sasaki S. Autophagy in spinal cord motor neurons in sporadic amyotrophic Nakaso K, Kirii K, Noguchi N, et al. Deficiency of p62/Sequestosome 1 causes lateral sclerosis. J Neuropathol Exp Neurol. 2011;70:349–59. hyperphagia due to leptin resistance in the brain. J Neurosci. 2013;33: 19. Al-Sarraj S, King A, Troakes C, Smith B, Maekawa S, Bodi I, Rogelj B, Al- 14767–77. Chalabi A, Hortobagyi T, Shaw CE. p62 positive, TDP-43 negative, neuronal 40. Yamanaka K, Boillee S, Roberts EA, Garcia ML, McAlonis-Downes M, Mikse cytoplasmic and intranuclear inclusions in the cerebellum and OR, Cleveland DW, Goldstein LS. Mutant SOD1 in cell types other than hippocampus define the pathology of C9orf72-linked FTLD and MND/ALS. motor neurons and oligodendrocytes accelerates onset of disease in ALS Acta Neuropathol. 2011;122:691–702. mice. Proc Natl Acad Sci U S A. 2008;105:7594–9. 20. Kato S. Amyotrophic lateral sclerosis models and human neuropathology: similarities and differences. Acta Neuropathol. 2008;115:97–114. 21. Aoki M, Ogasawara M, Matsubara Y, Narisawa K, Nakamura S, Itoyama Y, Abe K. Mild ALS in Japan associated with novel SOD mutation. Nat Genet. 1993;5:323–4. 22. Sasaki S, Nagai M, Aoki M, Komori T, Itoyama Y, Iwata M. Motor neuron disease in transgenic mice with an H46R mutant SOD1 gene. J Neuropathol Exp Neurol. 2007;66:517–24. 23. Hadano S, Otomo A, Kunita R, Suzuki-Utsunomiya K, Akatsuka A, Koike M, Aoki M, Uchiyama Y, Itoyama Y, Ikeda JE. Loss of ALS2/Alsin exacerbates motor dysfunction in a SOD1-expressing mouse ALS model by disturbing endolysosomal trafficking. PLoS One. 2010;5:e9805. 24. Doi H, Adachi H, Katsuno M, Minamiyama M, Matsumoto S, Kondo N, Miyazaki Y, Iida M, Tohnai G, Qiang Q, et al. p62/SQSTM1 differentially removes the toxic mutant androgen receptor via autophagy and inclusion formation in a spinal and bulbar muscular atrophy mouse model. J Neurosci. 2013;33:7710–27. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Molecular Brain Springer Journals

Systemic overexpression of SQSTM1/p62 accelerates disease onset in a SOD1H46R-expressing ALS mouse model

Free
16 pages

Loading next page...
 
/lp/springer_journal/systemic-overexpression-of-sqstm1-p62-accelerates-disease-onset-in-a-5swHSmI0t1
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Neurology; Psychopharmacology
eISSN
1756-6606
D.O.I.
10.1186/s13041-018-0373-8
Publisher site
See Article on Publisher Site

Abstract

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease characterized by a selective loss of upper and lower motor neurons. Recent studies have shown that mutations in SQSTM1 are linked to ALS. SQSTM1 encodes SQSTM1/p62 that regulates not only autophagy via the association with MAP1LC3/LC3 and ubiquitinated proteins but also the KEAP1-NFE2L2/Nrf2 anti-oxidative stress pathway by interacting with KEAP1. Previously, we H46R have demonstrated that loss of SQSTM1 exacerbates disease phenotypes in a SOD1 -expressing ALS mouse H46R model. To clarify the effects of SQSTM1 overexpression in this model, we generated SQSTM1 and SOD1 double- H46R H46R transgenic (SQSTM1;SOD1 ) mice. SQSTM1;SOD1 mice exhibited earlier disease onset and shorter lifespan than H46R did SOD1 mice. Conversely, disease progression after the onset rather slightly but significantly slowed in H46R SQSTM1;SOD1 mice. However, there were observable differences neither in the number of Nissl positive neurons H46R nor in the distribution of ubiquitin-positive and/or SQSTM1-positive aggregates between SOD1 and SQSTM1; H46R SOD1 mice. It was noted that these protein aggregates were mainly observed in neuropil, and partly localized to astrocytes and/or microglia, but not to MAP2-positive neuronal cell bodies and dendrites at the end-stage of disease. Nonetheless, the biochemically-detectable insoluble SQSTM1 and poly-ubiquitinated proteins were H46R H46R significantly and progressively increased in the spinal cord of SQSTM1;SOD1 mice compared to SOD1 mice. H46R These results suggest that overexpression of SQSTM1 in SOD1 mice accelerates disease onset by compromising the protein degradation pathways. Keywords: Amyotrophic lateral sclerosis, SOD1, SQSTM1/p62, Ubiquitin-positive aggregates Introduction different genes that are linked to ALS have been identified, Amyotrophic lateral sclerosis (ALS) is a progressive neuro- molecular mechanisms by which motor neurons are select- degenerative disease characterized by a selective loss of ively and progressively degenerated are still unknown [3]. upper and lower motor neurons. ALS patients exhibit en- Recently, mutations in the SQSTM1 gene have been hanced tendon reflex caused by loss of upper motor neurons identified in patients with ALS and ALS/frontotem- and muscle atrophy, impaired motor control caused by loss poral dementia (FTD) [4–8]. SQSTM1 mutations have of lower motor neurons without sensory disturbance, and originally been identified in Paget disease of bone [9]. ultimately die within 3 to 5 years due to respiratory failure The SQSTM1 gene product, sequestosome1 (SQSTM1/ [1]. An approximately 10% of ALS cases is familial and the p62), is a multi-functional adapter protein. SQSTM1 remaining 90% are sporadic [2]. Although more than 50 regulates not only autophagy via the association with microtubule-associated protein 1 light chain 3 (MAP1LC3/LC3) and poly-ubiquitinated proteins [10, * Correspondence: shinji@is.icc.u-tokai.ac.jp 11] but also the Kelch-like ECH-associated protein 1 Department of Molecular Life Sciences, Tokai University School of Medicine, (KEAP1)-nuclear factor related erythroid 2-related fac- Isehara, Kanagawa 259-1193, Japan The Institute of Medical Sciences, Tokai University, Isehara, Kanagawa tor 2 (NFE2L2/Nrf2) anti-oxidative stress pathway by 259-1193, Japan interacting with KEAP1 [12]. Indeed, ALS-linked Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mitsui et al. Molecular Brain (2018) 11:30 Page 2 of 16 SQSTM1 missense mutant (Leu341Val) [7]showeda compromising the protein degradation pathways in a H46R decreased affinity to LC3, resulting in impaired au- SOD1 -expressing ALS mouse model. tophagy [13]. It has also been shown that SQSTM1 is multiply phosphorylated. Especially, the phosphoryl- Methods ation of SQSTM1 at serine (Ser) 403 (corresponding to Animals H46R mouse Ser405) increases the binding affinity to SOD1 -tg mice express human SOD1 carrying the poly-ubiquitin molecules and enhances the autophagic deg- H46R mutation under control of the human SOD1 pro- radation of poly-ubiquitinated proteins [14, 15]. In addition, motor [22]. CAG-SQSTM1-HA (SQSTM1)-tg mice ex- it has been reported that the phosphorylation at Ser349 press human SQSTM1 with hemagglutinin (HA) tag at (corresponding to mouse Ser351) enhances the interaction the C-terminus under control of the CAG [(C) the cyto- between SQSTM1 and KEAP1, thereby activating the megalovirus (CMV) early enhancer element, (A) the pro- KEAP1-NFE2L2/Nrf2 pathway [16]. With respect to the re- moter, the first exon and the first intron of chicken lationship between SQSTM1 and neurodegenerative dis- β-actin gene, and (G) the splice acceptor of the rabbit H46R eases, it has been demonstrated that SQSTM1-positive β-globin gene] promoter [24]. Both SOD1 -tg and aggregates are accumulated in the spinal cord and brain of SQSTM1-tg mice were backcrossed to 57BL/6 N (B6) ALS and ALS/FTD patients [17–19]. mice more than 10 generations, and were maintained as On the other hand, the SOD1 gene has been identified as B6 congenic lines. We generated SQSTM1-overexpressing H46R H46R a causative gene for an autosomal dominant form of famil- SOD1 mice (SQSTM1;SOD1 )bycrossing male H46R ial ALS in 1993. Thus far, more than 180 different SOD1 SOD1 mice with female SQSTM1 mice. The offsprings mutations have been identified (http://alsod.iop.kcl.ac.uk/ were genotyped by PCR using genomic DNA extracted H46R als/). Several transgenic (tg) ALS mouse models expressing from ear tissues. Primers used were as follows; SOD1 : different SOD1 mutants have been generated [20]. Among hSOD1_ex2: 5′-TCAGAAACTCTCTCCAACTTTGC-3′ H46R them, SOD1 mutation has originally been identified in and 5′-CAAGTATGGGTCACCAGCAC-3′,hSOD1_ex4: Japanese kindred, showing a slow progression of symptoms 5′-GGCATCAGCCCTAATCCATC-3′ and 5′-CCGC H46R [21]. SOD1 -tg mouse recapitulates many neurological GACTAACAATCAAAGTG-3′, SQSTM1-HA:5′-AGCT and pathological features observed not in only familial but GCCTTGTACCCACATC-3′ and 5′-AGCGTAATC also in sporadic ALS patients, which include progressive TGGAACATCGT-3′.Micewerehousedat22 °Cwitha motor neuron degeneration, impaired motor control, and 12 h light-dark cycle. Food and water were fed ad libitum. accumulation of SQSTM1 and ubiquitin-positive aggregates in the spinal cord [19, 20, 22, 23]. Antibodies It has been shown that, in other motor neuron disease Primary antibodies used for immunohistochemistry were (MND) mouse model, i.e., spinal and bulbar muscular atro- listed as follows; anti-human Ubiquitin (MBL MK-11-3; phy (SBMA) mice, loss of SQSTM1 causes earlier disease mouse monoclonal, 1:300), anti-human Ubiquitin (Dako onset, whereas overexpression of it ameliorates disease Z0458; rabbit polyclonal, 1:500), anti-human SQSTM1 symptoms [24]. We have previously demonstrated that loss (Thermo PA5–20839; rabbit polyclonal, 1:300), anti-human of SQSTM1 exacerbates disease phenotypes, including a p62-C (PROGEN GP62-C; guinea-pig polyclonal, 1:500), shorter lifespan, accelerated body weight decline and motor anti-rat MAP2 (SIGMA M9942; mouse monoclonal, H46R dysfunction, in a mutant SOD1 -tg ALS mouse model 1:500), anti-GFAP (NICHIREI 422251; rabbit polyclonal, [25]. Thesefindings suggest aneuroprotective role of 1:2), anti-mouse Iba1 (Abcam ab178896; rabbit monoclo- SQSTM1 in vivo. However, the effect of SQSTM1 overex- nal, 1:500), and anti-HA-tag (Cell Signaling 3724; rabbit pression on the disease onset and progression in mutant monoclonal, 1:500) antibodies. Secondary antibodies in- SOD1-expressing mice remains unknown. cluded Alexa 594 conjugated anti-mouse IgG (Molecular In this study, to clarify the effect of systemic overexpres- Probes, 1:500), Alexa 594 conjugated anti-Rabbit IgG sion of SQSTM1 on the disease onset and progression as (Molecular Probes, 1:500), Alexa 594 conjugated well as the accumulation of aggregated proteins in anti-guinea-pig IgG (Molecular Probes, 1:500), Alexa H46R SOD1 mice, we generated SQSTM1-overexpressing 488 conjugated anti-mouse IgG (Molecular Probes, H46R H46R SOD1 double-tg (SQSTM1;SOD1 ) mice, and ana- 1:500), and Alexa 488 conjugated anti-Rabbit IgG lyzed lifespan, body weight, the number of motor neurons, (Molecular Probes, 1:500) antibodies. and the distribution of ubiquitin-positive aggregates Primary antibodies used for western blotting included in the spinal cord. In addition, we conducted a series anti-human SOD1 (SANTA CRUZ cs-11,407; rabbit poly- of western blot analysis to quantify the amount of clonal, 1:15000), anti-human misfolded SOD1 (C4F6; expressed protein levels including phosphorylated mouse monoclonal, 1:3000) [26], anti-bovine ubiquitin forms of SQSTM1. We here revealed that overexpres- (SANTA CRUZ cs-8017; mouse monoclonal, 1:3000), sion of SQSTM1 accelerated disease onset by anti-human SQSTM1 (MBL PM045; rabbit polyclonal, Mitsui et al. Molecular Brain (2018) 11:30 Page 3 of 16 1:2000), anti-human phosphorylated SQSTM1 (S403) overnight. Lumbar cord was further immersed in metha- (MBL D343–3; rat monoclonal, 1:3000), anti-mouse phos- nol:chloroform = 1:1 solution for 30 min, and in 100% phorylated SQSTM1 (S351) (MBL PM074; rabbit poly- chloroform for 2 h followed by the treatment with fresh clonal, 1:2000), anti-human LC3 (SIGMA L8918; rabbit chloroform overnight. The resulting lumbar cord sam- polyclonal, 1:5000), anti-human NQO1 (Abcam ab34173; ples were embedded in paraffin. Paraffin-embedded rabbit polyclonal, 1:3000), anti-rabbit GAPDH (MBL transverse sections at a thickness of 6 μm were prepared M171–3; mouse monoclonal, 1:5000), anti-Actin (SIGMA by microtome (LEICA RM2165) for Nissl staining and A5060; rabbit polyclonal, 1:1000), and anti-HA-tag immunohistochemistry. (SANTA CRUZ sc-805; rabbit polyclonal, 1:3000) anti- bodies. Secondary antibodies included horseradish peroxid- Nissl staining ase (HRP)-conjugated anti-mouse IgG (Jackson, 1:5000), Lumbar cord paraffin sections from mice with four dif- HRP-conjugated anti-rabbit IgG (GE Healthcare NA934, ferent genotypes at 16 and 22 weeks of age (WT, H46R H46R 1:5000), and HRP-conjugated anti-rat IgG (SANTA CRUZ SQSTM1, SOD1 , and SQSTM1;SOD1 ), and at H46R H46R sc-2006, 1:5000) antibodies. end-stage (SOD1 and SQSTM1;SOD1 )or 28 weeks of age (WT and SQSTM1), were treated with Lifespan analysis xylene (5 min × 3 times) and ethanol (5 min × 3 times), Lifespan (endpoint) of tg mice with each genotype and hydrated. Slices were permeabilized in 1% Triton H46R H46R (SOD1 and SQSTM1;SOD1 ) were determined by X-100/phosphate buffered saline (PBS) for 15 min and the observations that mice were unable to crawl by their stained with 50 μl of NeuroTrace green fluorescent Nissl forelimb and to eat/drink by themselves. stains (ThermoFisher N21480) (1:500)/PBS for 25 min. Slices were washed with PBS for 2 h and mounted by VEC- Growth curve analysis TASHIELD Mounting Medium with DAPI for nuclear Body weight of mice with each genotype [wild-type staining. A total of 5 representative Nissl staining images of H46R H46R (WT), SQSTM1, SOD1 , and SQSTM1;SOD1 ] every tenth serial sections from each sample was captured was weekly monitored from 5 weeks to a maximum of by fluorescence microscope (KEYENCE BZ-9000). 32 weeks of age. Counting the number of large Nissl-positive neurons Disease onset and progression analyses Large Nissl-positive neurons, representing motor neu- Onset of the disease of mice with each genotype rons, were identified by using Dynamic cell count (KEY- H46R H46R (SOD1 and SQSTM1;SOD1 ) was defined as the ENCE). In brief, tones of the captured images were point (date) at which body weight was successively de- binarized, and the area covering 600 pixels (px) × 400 px creased from its peak-value in each animal without any corresponding to anterior horn of the spinal cord was increases thereafter. Post-onset survival was defined as selected. Automatic cell separation procedures (setting the duration from the onset of disease to the endpoint. resolution; 14, threshold; 45 with excluding cells whose cross-sectional areas were less than 240 px) were exe- Preparation of paraffin sections cuted. Neurons identified were defined as large We obtained tissue samples of the spinal cord from Nissl-positive neurons and counted. The numbers of H46R H46R SOD1 and SQSTM1;SOD1 mice at 16 and large Nissl-positive neurons from 5 representative im- 22 weeks of age, and at end-stage. We also obtained the ages from each sample were summed. samples from WT and SQSTM1 mice at 16, 22, and 28 weeks of age. Samples from WT and SQSTM1 mice Immunohistochemistry at 28 weeks of age were used to compare to those from Spinal cord sections were deparafinized in xylene tg mice at their end-stage. Mice were anesthetized with (5 min × 3 times) and ethanol (5 min × 3 times), and 4% isoflurane by inhalation and perfused with physio- were hydrated. Sections were heated by microwave oven logical saline containing 100 U/ml heparin, followed by in 300 ml of 0.1 M citric acid (pH 5.0) for 15 min. After 4% paraformaldehyde (PFA)/0.1 M phosphate buffer cooling at room temperature (RT), sections were washed (PB) (pH 7.2). Spinal cord together with the spinal col- with PBS and blocked with 50 μl of 0.1% Triton X-100/ umn was removed and fixed with 4% PFA/PB overnight 5% normal goat serum (NGS)/PBS for 30 min. Sections at 4 °C. Lumbar cord (L4-L5) was removed from the were incubated with primary antibody in 50 μl of the spinal column, cut by an approximately 2 mm length antibody diluted solution (0.05% Triton X-100, 2% NGS/ (rostral to caudal direction), and post-fixed with 4% PBS) overnight at 4 °C. Sections were washed with PBS PFA/PB overnight at 4 °C. After washing with PB over- for 2 h and incubated with secondary antibody in 50 μl night twice, lumbar cord was immersed in 100% metha- of the antibody diluted solution for 2 h. Sections were nol for 1 h followed by soaking to fresh methanol washed with PBS for 1 h and mounted using Mitsui et al. Molecular Brain (2018) 11:30 Page 4 of 16 VECTASHIELD Mounting Medium with DAPI. Signals (BIO-RAD 1610374) was used as molecular weight were analyzed by fluorescence microscope (KEYENCE markers. Proteins were electro-transferred onto poly- BZ-710). vinylidene fluoride (PVDF) membrane (Merckmilli- pore Immobilon-P) in the transfer buffer [25 mM Preparation of tissue samples for western blot analysis Tris-HCl, 192 mM glycine, 20% (w/v) methanol]. The To analyze the expression and distribution of transgene membranes were blocked with 0.5% skimmed milk/ product; i.e., SQSTM1-HA, we obtained tissue samples TBST [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, from WT and SQSTM1 mice at 16 weeks of age. Tissues 0.1% Tween20] for 1 h at RT. When using including the liver, skeletal muscle, olfactory bulb, cere- anti-phosphorylation SQSTM1 antibody, membranes bral cortex, hippocampus, cerebellum, and whole spinal were blocked with 50% blocking one (Wako)/TBST cord were removed, immediately frozen on dry ice, and for 30 min at RT. Membranes were incubated with stored at − 80 °C until use. To analyze the expression of the primary antibody in TBST overnight at 4 °C. particular proteins of interest in the spinal cord, we ob- After washing with TBST (10 min, 6 times), mem- tained the samples from mice with four different geno- branes were incubated with HRP-conjugated second- H46R H46R types (WT, SQSTM1, SOD1 ,and SQSTM1;SOD1 ) ary antibody in TBST for 2 h at RT. After washing H46R at 16 and 22 weeks of age, and at end-stage (SOD1 with TBST, signals were visualized by Immobilon H46R and SQSTM1;SOD1 ) or 28 weeks of age (WT and (Merck Millipore) and detected by Ez-Capture MG SQSTM1). The whole spinal cord was removed from each (ATTO). mouse, immediately frozen on dry ice, and stored at − 80 ° C until use. Quantification of the signal intensities on western blotting Preparation of protein samples We conducted a quantitative analysis of the signal intensities Spinal cord tissues were weighted and homogenized in 2 on western blotting to determine the expression levels of weight-volume (mg/μl) of PBS by sonication (Output 2, proteins of interest in 1% Triton X-100 soluble and insoluble 10–20 times, BRANSON Sonifier 450). One-hundred factions from mice with four different genotypes (WT, H46R H46R microliters of homogenates and 700 μl of Lysis buffer A’ SQSTM1, SOD1 ,and SQSTM1;SOD1 )at16and H46R [25 mM Tris-HCl (pH 7.5), 50 mM NaCl, 1% Triton 22 weeks of age, and at end-stage (SOD1 and H46R X-100, Complete Protease Inhibitor Cocktail (Roche), SQSTM1;SOD1 ) or 28 weeks of age (WT and SQSTM1). Phosphatase inhibiter Cocktail I (SIGMA)] were The signal intensities were quantified by using CS Analyzer mixed, followed by centrifugation at 23,000×g for Ver3.0 (ATTO) and were normalized by the levels of glycer- 20 min at 4 °C. Supernatant was collected as 1% Tri- aldehyde 3-phosphate dehydrogenase (GAPDH) for soluble ton X-100 soluble fraction. The insoluble pellet was fractions or β-actin for insoluble fractions. washed by mixing with 600 μl of Lysis buffer A’ and centrifuged again, and resulting pellet was suspended Statistical analysis in 400 μl of Lysis buffer B′ [25 mM Tris-HCl All statistical analyses were conducted using Prism 5 (Graph (pH 7.5), 50 mM NaCl, 5% sodium dodecyl sulfate pad). Data for lifespan, onset of disease, and post-onset sur- H46R H46R (SDS)], sonicated (Output 2, 5–10 times), and used as vival between SOD1 and SQSTM1;SOD1 mice were 1% Triton X-100 insoluble fraction. Protein concen- compared using Kaplan-Meier survival analysis with tration was determined by Micro BCA (Thermo, Log-rank test. Statistical significances for body weight, the MicroBCA Protein Assay kit #23235). Each protein number of spinal motor neurons, and quantitative western fraction was diluted by the corresponding lysis buffer blotting data between groups were evaluated by One-way and adjusted its concentration at 0.75 μg/μl. or Two-way ANOVA with Bonferroni post hoc tests. A One-hundred microliters of diluted samples were p-value < 0.05 was considered as reaching statistical mixed with 50 μl of 3xSDS buffer [187 mM Tris-HCl significance. (pH 6.8), 6% SDS, 30% glycerol, 0.015% bromophenol blue (BPB), 15% 2-melcaptoethanol] and heated for Results 5minat 95°C. Expression and tissue distribution of SQSTM1-HA in SQSTM1 transgenic mice Western blot analysis SQSTM1-tg mouse carries the expression-cassette that Equal amount of proteins was subjected to can express the C-terminally HA-tagged human SDS-polyacrylamide gel electrophoresis (PAGE) (Wako SQSTM1 (SQSTM1-HA) protein under control of the Supersep™Ace, 5–20% 17well) in SDS-PAGE buffer CAG promoter (Additional file 1: Figure S1a) [24]. To [25mM Tris-HCl, 192mMglycine,0.1%(w/v)SDS]. determine the distribution and expression levels of Precision Plus Protein standards dual color standard SQSTM1-HA in SQSTM1 mice, we prepared protein Mitsui et al. Molecular Brain (2018) 11:30 Page 5 of 16 samples from the liver, skeletal muscle, olfactory bulb, (23.8 ± 2.1 weeks) (Fig. 1d and Table 1; Log-rank test; cerebral cortex, hippocampus, cerebellum, and spinal p = 0.0007). To investigate the difference in disease H46R H46R cord from WT and SQSTM1 mice, and performed west- progression between SOD1 and SQSTM1;SOD1 ern blotting using anti-SQSTM1 and anti-HA antibodies. mice, we analyzed the survival interval after disease onset Mouse endogenous SQSTM1 was detected in all tissues (post-onset survival). Contrary to the disease onset, examined with highest in the cerebral cortex and hippo- survival interval after the onset of disease in H46R campus (Additional file 1: Figure S1b, see WT). The SQSTM1;SOD1 mice (3.9 ± 1.3 weeks) was slightly CAG promoter-driven human SQSTM1-HA was highly but significantly extended when compared to H46R expressed in skeletal muscle, and also detected in the cen- SOD1 mice (3.0 ± 1.1 weeks) (Fig. 1e and Table 1; tral nervous system including the spinal cord, while it was Log-rank test; p = 0.0314). undetectable in the liver (Additional file 1: Figure S1b). Progressive decreases in the number of motor neurons in H46R Overexpression of SQSTM1 accelerates body weight the lumbar spinal cord of SOD1 -expressing mice H46R declines in SOD1 mice To investigate whether earlier disease onset observed in H46R We performed growth curve analysis of WT, SQSTM1, SQSTM1;SOD1 mice was associated with motor H46R H46R SOD1 , and SQSTM1;SOD1 mice. Both WT and neuron loss in the spinal cord, we performed Nissl stain- SQSTM1 mice were viable for at least 32 weeks of age. ing and counted the number of large neurons in the an- Throughout the experimental periods, body weights of terior horn of lumbar spinal cord. In this analysis, mice H46R SQSTM1 mice tended to be lower than those of WT with four different genotypes; WT, SQSTM1, SOD1 , H46R mice in female (Fig. 1a). Body weights of female and SQSTM1;SOD1 mice at 16 and 22 weeks of age, H46R H46R H46R SQSTM1;SOD1 mice were significantly lower than and at end-stage (SOD1 and SQSTM1;SOD1 )or H46R those of SOD1 mice at 5, 8–13, 16–25, and 28 weeks of age (WT and SQSTM1), were used. The num- 28 weeks of age (Fig. 1a). In male, body weights of bers of large Nissl-positive neurons were progressively H46R H46R H46R SQSTM1;SOD1 mice were significantly lower than decreased both in SOD1 and SQSTM1;SOD1 H46R those of SOD1 mice at 25 and 26 weeks of age mice compared to WT and SQSTM1 mice (Fig. 2 and (Fig. 1b). These results indicate that overexpression of Additional file 2:Fig.S2a). However, therewereno H46R SQSTM1 in SOD1 mice accelerates weight loss. significant differences in the numbers of Nissl positive large neurons between groups both at 22 weeks of Overexpression of SQSTM1 shortens lifespans in ageand theend-stage,suggestingthatdecrease in the H46R SOD1 mice number of motor neurons did not explain the reasons In thepreviousstudy,wehaveshown that loss of SQSTM1 why overexpression of SQSTM1 not only accelerates H46R significantly shortens lifespans in SOD1 mice [25]. Fur- the onset of disease but also slows survival after the H46R ther, it has been reported that loss of SQSTM1 shortens onset in SOD1 mice. lifespan while overexpression of it prolongs lifespan in SBMA mice [24]. To clarify the effect of overexpression of Accumulation of ubiquitin/SQSTM1-positive aggregates H46R SQSTM1 on the onset and progression in an ALS mouse and the activation of glial cells in SOD1 -expressing H46R model (SOD1 mice), we performed lifespan analysis mice H46R H46R using SOD1 and SQSTM1;SOD1 mice. Since there SQSTM1 acts as an adaptor protein that directly binds were no gender differences of lifespans in B6 congenic to poly-ubiquitinated proteins destined for degradation H46R SOD1 mice [27], we used the gender-combined data. in autophagy [10]. It has been reported that accumula- Kaplan-Meier survival analysis revealed that the lifespan of tion of ubiquitin- and SQSTM1-positive aggregates is H46R SQSTM1;SOD1 mice (179.3 ± 2.6 days) was signifi- observed not only in ALS patients but also in a H46R H46R cantly shorter than that of SOD1 mice (189.3 ± 2.2 days) SOD1 ALS mouse model [22], and that proliferation (Fig. 1c;Log-ranktest; p = 0.0013). of astrocytes and the activation of microglia also occur H46R in SOD1 mice as disease progresses. To clarify the Overexpression of SQSTM1 accelerates onset of disease effect of SQSTM1 overexpression on the localization H46R but slightly extends post-onset survival in SOD1 of ubiquitin-/SQSTM1-positive aggregates and the ac- mice tivation of glial cells, we conducted immunohisto- Previously, it has been reported that onset of disease chemical analysis of ubiquitin and SQSTM1 in the H46R can be estimated by the turning-point of the body lumbar cord (L4–5) from WT, SQSTM1, SOD1 , H46R weight from gain to loss in mutant SOD1-tg mice and SQSTM1;SOD1 mice at 16 and 22 weeks of H46R H46R [27–29]. Growth curve analysis revealed that onset of age, and at end-stage (SOD1 and SQSTM1;SOD1 ) H46R disease in SQSTM1;SOD1 mice (21.5 ± 2.1 weeks) or 28 weeks of age (WT and SQSTM1)(Fig. 3 and H46R was significantly earlier than those of SOD1 mice Additional file 2: Figure S2b). We also conducted Mitsui et al. Molecular Brain (2018) 11:30 Page 6 of 16 H46R Fig. 1 Analysis of lifespan, disease onset, and progression in SQSTM1;SOD1 mice. a Growth curve in female mice with four different genotypes H46R H46R [wild-type (WT; n =8–12), SQSTM1 (SQSTM1; n =9–12), SOD1 (H46R; n =2–12), and SQSTM1;SOD1 (SQSTM1;H46R; n =1–12)]. b Growth curve in male mice with four different genotypes [WT; n =4–12, SQSTM1; n =11–12, H46R; n =2–12, and SQSTM1;H46R; n =4–12]. a, b Body weight of H46R H46R SQSTM1;SOD1 mice was significant lower than that of SOD1 mice (*p <0.05, **p < 0.01, ***p <0.001, ****p < 0.0001). There are no significantly differences in body weight between WT and SQSTM1 mice. Values were mean ± s.d.. Statistical significance was evaluated by one-way ANOVA with H46R H46R Bonferroni’s post hoc test. c Lifespan analysis of SOD1 mice (H46R; n = 31) and SQSTM1;SOD1 mice (SQSTM1;H46R; n = 28). Kaplan-Meier analysis H46R H46R demonstrated that lifespan in SQSTM1;SOD1 mice (mean ± s.d.; 179.3 ± 2.6 days) was significantly shorter than that in SOD1 mice (189.3 ± H46R H46R 2.2 days) (Log-rank test; p = 0.0013). d Onset of the disease in SOD1 (H46R; n =24) and SQSTM1;SOD1 (SQSTM1;H46R; n = 24) mice. Onset of H46R disease is defined as the turning-point of body weight successive reduction. Onset in SQSTM1;SOD1 mice (mean ± s.d.; 21.5 ± 2.1 weeks) was H46R significantly earlier than that in SOD1 mice (23.8 ± 2.1 weeks) (Log-rank test; p =0.0007). e Survival after the onset of disease (post-onset survival H46R H46R interval) in SOD1 (H46R; n =24) and SQSTM1;SOD1 (SQSTM1;H46R; n = 24) mice. Survival after the onset is defined as the period between the H46R onset of disease and end-point. Post-onset survival interval in SQSTM1;SOD1 mice (mean ± s.d.; 3.9 ± 1.3 weeks) was significantly longer than that in H46R SOD1 mice (3.0 ± 1.1 weeks) (Log-rank test; p =0.0314) Mitsui et al. Molecular Brain (2018) 11:30 Page 7 of 16 H46R H46R Table 1 Summary of survival and onset point analysis in SOD1 and SQSTM1;SOD1 mice Genotype Number End point (weeks) s.d. Onset (weeks) s.d. Post-onset survival interval (weeks) s.d. H46R SOD1 24 26.7 1.5 23.8 2.1 3.0 1.1 H46R SQSTM1;SOD1 24 25.4 1.5 21.5 2.1 3.9 1.3 Number represents the number of mice used in survival and onset analyses. Values represent mean ± s.d. (weeks) H46R H46R immunohistochemistry for microtubule-associated protein cord of SOD1 -expressing mice (SOD1 and H46R 2 (MAP2), glial fibrillary acidic protein (GFAP), and ion- SQSTM1;SOD1 ) as disease progressed (Fig. 3 and Add- ized calcium binding adaptor molecule 1 (Iba1) using mice itional file 2: Figure S2b). It was also noted that a majority of H46R with four different genotypes at end-stage (SOD1 and the ubiquitin-positive signals was colocalized with H46R SQSTM1;SOD1 ) or 28 weeks of age (WT and SQSTM1 (Figs. 3 and 5a). Further, SQSTM1-HA ag- SQSTM1). MAP2, GFAP, and Iba1 were used as cellular gregates were immunohistochemically detected in H46R markers for neuronal soma and dendrite, astrocytes, and SQSTM1;SOD1 mice at 22 weeks and end-stage microglia, respectively (Figs. 4 and 5, and Additional file 2: by using the anti-HA-tag antibody (Additional file 3: Figure S2c). Figure S3), indicating that overexpressing SQSTM1 Double immunostaining of ubiquitin and SQSTM1 re- was also progressively accumulated in the spinal cord vealed that a progressive accumulation of ubiquitin- and as was endogenous SQSTM1. However, there were no SQSTM1-positive aggregates was observed in the spinal apparent differences in the amount and/or number of Fig. 2 Quantitative analysis of motor neurons in the anterior horn of lumbar spinal cord. a Representative images for fluorescence Nissl staining in the anterior horn. Nissl positive neurons were detected in the anterior horn of lumbar spinal cord (L4–5) from the mice with four different H46R H46R genotypes; wild-type (WT), SQSTM1, SOD1 (H46R), and SQSTM1;SOD1 (SQSTM1;H46R) at 16 and 22 weeks of age (wk), and at end-stage (H46R and SQSTM1;H46R) or at 28 wk. (WT and SQSTM1). Cells covering the area of over 240 square pixels (px), corresponding to large motor neurons, are surrounded by red circles, and counted. Scale bars = 50 μm. b Number of large Nissl-positive neurons in the anterior horn of lumbar spinal cord. Vertical axis represents the cumulative number of motor neurons identified in 5 representative Nissl staining images of every tenth H46R H46R serial sections from each sample (n =3–4, SOD1 and SQSTM1;SOD1 mice at 22 wk. and end-stage; n =4, others; n =3). Values are mean ± s.e.m.. Statistical significance was evaluated by two-way ANOVA with Bonferroni’spost hoc test (*p < 0.05, **p < 0.01) Mitsui et al. Molecular Brain (2018) 11:30 Page 8 of 16 Fig. 4 Extensive accumulation of ubiquitin-positive aggregates in the H46R Fig. 3 Progressive accumulation of ubiquitin/SQSTM1-positive spinal cord from SOD1 -expressing mice at end-stage. a-c aggregates in the anterior horn of lumbar spinal cord. a-c Representative images of double immunostaining with MAP2 (green; Representative images of double immunostaining with Ubiquitin neuron marker) and Ubiquitin (red) (a), Ubiquitin (green) and GFAP (red; (green) and SQSTM1 (red) in the lumbar spinal cord (L4–5) from wild- astrocyte marker) (b), Ubiquitin (green) and Iba1 (red; microglia marker) H46R H46R type (WT), SQSTM1 (SQSTM1), SOD1 (H46R) and SQSTM1;SOD1 (c) in the lumbar spinal cord (L4–5) from wild-type (WT) and SQSTM1 H46R H46R (SQSTM1;H46R) mice at 16 weeks of age (wk) (a), 22 wk. (b), and end- (SQSTM1) at 28 weeks of age, SOD1 (H46R) and SQSTM1;SOD1 stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1) (c). The (SQSTM1;H46R) mice at end-stage. The nuclei were counterstained with nuclei were counterstained with DAPI (blue). Scale bars = 50 μm. DAPI (blue). Scale bars = 50 μm. b Arrows indicate colocalization of Ubiquitin-positive aggregates and SQSTM1 aggregates were observed ubiquitin-positive aggregates with GFAP-positive astrocytes. c Arrows H46R H46R in the anterior horn of SOD1 (H46R) and SQSTM1;SOD1 indicate colocalization of ubiquitin-positive aggregates with (SQSTM1;H46R) mice at 22 wk. and end-stage. Arrows indicate Iba1-positive microglia ubiquitin-positive aggregates that are colocalized with SQSTM1 aggregates were localized to astrocytes (Figs. 4b and 5c; H46R ubiquitin-/SQSTM1-positive aggregates between SOD1 arrows). Activation of Iba1-positive microglia was also H46R H46R and SQSTM1;SOD1 mice. observed in SOD1 -expressing mice (Fig. 4c and Double immunostaining of ubiquitin and MAP2 re- Additional file 2: Figure S2c). Ubiquitin-positive aggre- vealed that ubiquitin-positive aggregates were not gates that were surrounded by microglia were occasionally present in MAP2-positive neuronal cell bodies or den- observed (Fig. 5d; arrows). However, there were no differ- H46R drites in the lumbar cord of SOD1 -expressing mice ences in the levels of glial cell activation between H46R H46R (Figs. 4a and 5b and Additional file 2: Figure S2c). On SOD1 and SQSTM1;SOD1 mice. the other hand, double immunostaining of ubiquitin Together, these results indicate that ubiquitin-/ and GFAP demonstrated that the number of SQSTM1-positiveaggregatesweremainlylocalized to GFAP-positive astrocytes was significantly increased in neuropil and occasionally to astrocytes and/or micro- H46R SOD1 -expressing mice (Fig. 4b and Additional file 2: glia but not to the cell body of neurons at the H46R Figure S2c), and that small numbers of ubiquitin-positive end-stage of SOD1 -expressing mice. However, Mitsui et al. Molecular Brain (2018) 11:30 Page 9 of 16 soluble and insoluble proteins from the spinal cord of mice with four different genotypes; WT, SQSTM1, H46R H46R SOD1 , and SQSTM1;SOD1 mice at 16 and H46R 22 weeks of age, and at end-stage (SOD1 and H46R SQSTM1;SOD1 ) or at 28 weeks of age (WT and SQSTM1) and performed western blot analysis, followed by the quantification of signal intensities. Although a progressive accumulation of insoluble high-molecular weight SOD1 (SOD1_HMW) was ob- H46R H46R served both in SOD1 and SQSTM1;SOD1 mice as disease progressed (Fig. 6a and Additional file 4: Figure S4c), quantitative analyses showed no signifi- cant differences of the amount of insoluble SOD1_HMW H46R and monomer SOD1 (SOD1_mono) between SOD1 H46R and SQSTM1;SOD1 mice (Fig. 6c). Further, we de- tected misfolded SOD1 molecules using anti-human misfolded SOD1 antibody (C4F6) [26]. Misfolded H46R monomeric SOD1 was observed both in SOD1 H46R and SQSTM1;SOD1 mice, but not in WT and SQSTM1 mice (Fig. 6b). Quantitative analyses showed that both soluble and insoluble misfolded SOD1 in H46R SQSTM1;SOD1 mice at 16 weeks of age were signifi- H46R cantly higher than those in SOD1 mice, while there were no differences between two groups at the later stages of disease (22 weeks of age and end-stage) (Fig. 6c and Additional file 5:FigureS5). Both soluble and insoluble SQSTM1 were significantly H46R and progressively increased in SOD1 and H46R SQSTM1;SOD1 mice as disease progressed (Fig. 7a, b and Additional file 4: Figure S4d and S4e). Compared to H46R SOD1 mice, the accumulated levels of insoluble H46R SQSTM1 in SQSTM1;SOD1 mice were much higher at 16 and 22 weeks of age, and at end-stage (Fig. 7b). Fig. 5 High-magnification images for the distribution of ubiquitin- To distinguish between mouse endogenous and H46R positive aggregates in the spinal cord from SOD1 -expressing transgene-derived SQSTM1, we performed western blot mice. a-d Representative images of double immunostaining with analysis using anti-HA antibody, and revealed that Ubiquitin (green) and SQSTM1 (red) (a), MAP2 (green; neuron transgene-encoded SQSTM1-HA was also progressively marker) and Ubiquitin (red) (b), Ubiquitin (green) and GFAP (red; H46R accumulated as insoluble forms in SQSTM1;SOD1 astrocyte marker) (c), Ubiquitin (green) and Iba1 (red; microglia H46R marker) (d) in the lumbar cord (L4–5) from SOD1 (H46R) and mice (Fig. 7a). H46R SQSTM1;SOD1 (SQSTM1;H46R) mice at end-stage. The nuclei were Accumulation of insoluble poly-ubiquitinated proteins counterstained with DAPI (blue). Scale bars = 10 μm. c Arrows that were represented by smear signals was observed in indicate that ubiquitin-positive aggregates are present in GFAP- H46R SOD1 -expressing mice (Fig. 7a and Additional file 4: positive astrocytes. d Arrows indicate that ubiquitin-positive Figure S4 g). Although poly-ubiquitinated proteins de- aggregates are surrounded by microglia H46R tected in the spinal cord of SQSTM1;SOD1 mice H46R were comparable to those in SOD1 mice at 22 weeks H46R there were no discernible differences in the distribu- of age, those in SQSTM1;SOD1 mice at end-stage H46R tion and the degree of colocalization of ubiquitin-positive were significantly higher than that in SOD1 mice H46R aggregates to glial cells between SQSTM1;SOD1 and (Fig. 7a and c). H46R SOD1 mice. These results suggest that overexpression of SQSTM1 accelerates the accumulation not only of Overexpression of SQSTM1 increases insoluble poly- misfolded SOD1 at the earlier stage, but also of in- H46R ubiquitinated proteins in the spinal cord of SOD1 mice soluble poly-ubiquitinated proteins at the later stage H46R To biochemically investigate the accumulation of SOD1, of disease in the spinal cord of SOD1 -expressing SQSTM1, and ubiquitin, we extracted Triton X-100 mice. Mitsui et al. Molecular Brain (2018) 11:30 Page 10 of 16 H46R Fig. 6 Overexpression of SQSTM1 increases misfolded SOD1 in the spinal cord of SOD1 -expressing mice. a Western blot analysis of SOD1. b Western blot H46R H46R analysis of misfolded SOD1. a, b The spinal cord samples were obtained from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1). Two fractions; 1% Triton X-100 soluble and 1% Triton X-100 insoluble/5%SDS soluble fractions were analyzed. SOD1_mono and SOD1_HMW represent monomeric and high molecular- weight forms of SOD1, respectively. SOD1(C4F6) represents misfolded SOD1. GAPDH and β-actin (Actin) were used for a loading control in Triton X-100- soluble and -insoluble fractions, respectively. c Quantification of 1% Triton X-100 soluble SOD1, insoluble SOD1_mono, insoluble SOD1_HMW, soluble misfolded SOD1, and insoluble misfolded SOD1. Values are mean ± s.e.m. (n = 4) in an arbitrary unit relative to H46R mice at 16 weeks of age. Signal intensities were normalized by the levels of GAPDH (soluble fractions) and Actin (insoluble fractions). Statistical significance was evaluated by two-way ANOVA with Bonferroni’s post hoc test (comparisons between different genotypes in the same age; *p < 0.05, **p < 0.01) Overexpression of SQSTM1 increases insoluble of SQSTM1 overexpression on autophagy and H46R H46R phosphorylated SQSTM1 in the spinal cord of SOD1 anti-oxidative stress responses in SOD1 mice, we per- mice formed western blot analysis using antibodies to specific- Previous studies have shown that phosphorylation of ally detect Ser403(human)/Ser405(mouse)-phosphorylated SQSTM1 at Ser403 regulates degradation of SQSTM1, Ser349(human)/Ser351(mouse)-phosphorylated poly-ubiquitinated proteins by autophagy [14], and that SQSTM1, MAP1LC3/LC3, and NAD(P)H dehydrogenase protein aggregates containing Ser403-phosphorylated [quinone] 1 (NQO1) in the spinal cord of WT, SQSTM1, H46R H46R SQSTM1 are enriched in the brain of ALS patients [30]. SOD1 ,and SQSTM1;SOD1 mice at 16 and H46R Further, phosphorylation of SQSTM1 at Ser351 increases 22 weeks of age, and at end-stage (SOD1 and H46R the binding of KEAP1 to SQSTM1, thereby activating the SQSTM1;SOD1 ) or at 28 weeks of age (WT and NFE2L2/Nrf2 mediated anti-oxidative stress pathway [16]. SQSTM1). These phosphorylation sites are conserved between hu- The levels of insoluble Ser403/Ser405-phosphorylated H46R man and mouse and can be detected by specific SQSTM1 were increased both in SOD1 and H46R anti-phospho-SQSTM1 antibodies. To elucidate the effect SQSTM1;SOD1 mice at 22 weeks of age and Mitsui et al. Molecular Brain (2018) 11:30 Page 11 of 16 H46R Fig. 7 Overexpression of SQSTM1 increases the insoluble poly-ubqiuitinated proteins in the spinal cord of SOD1 -expressing mice. a Western blot H46R analysis of SQSTM1, SQSTM1-HA (HA) and Ubiquitin (Ub) in the spinal cord from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and H46R SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1). Two fractions; 1% Triton X-100 soluble and 1% Triton X-100 insoluble/5%SDS soluble fractions were analyzed. SOD1_mono and SOD1_HMW represent monomeric and high molecular-weight forms of SOD1, respectively. Ub_mono and Ub_HMW represent monomeric ubiquitin and poly-ubiquitinated proteins, respectively. GAPDH and β-actin (Actin) were used for a loading control in Triton X-100-soluble and -insoluble fractions, respectively. Since the blots used in these western analyses were same as those in Fig. 6a, same images for GAPDH and Actin (asterisks) in Fig. 6a were used again as references. b Quantification of soluble and insoluble SQSTM1. c Quantification of soluble and insoluble poly-ubiquitinated proteins (Ub_HMW). b, c Values are mean ± s.e.m. (n = 4) in an arbitrary unit relative to WT mice at 16 weeks of age. Signal intensities were normalized by the levels of GAPDH (soluble fractions) and Actin (insoluble fractions). Statistical significance was evaluated by two-way ANOVA with Bonferroni’spost hoc test (comparisons between different genotypes in the same age; *p < 0.05, **p < 0.01, ***p <0.001, ****p <0.0001) H46R H46R end-stage compared to WT and SQSTM1 mice. How- SQSTM1;SOD1 , but not in SOD1 , mice as dis- ever, there were no differences in the amount of Ser403/ ease progressed (Fig. 8a, c and Additional file 6:Fig.S6d). H46R Ser405-phosphorylated SQSTM1 between SOD1 The levels of LC3-II in insoluble fraction were pro- H46R H46R and SQSTM1;SOD1 mice, despite that the total gressively increased in SOD1 -expressing mice amount of SQSTM1 was progressively increased at the (Fig. 8a and Additional file 6:FigureS6f). Thequan- end-stage (Figs. 7a, b and 8a, b, and Additional file 6: titative analysis of the LC3-II to LC3-I ratio (LC3-II/ Figure S6b). Remarkably, Ser349/Ser351-phosphorylated LC3-I), a marker of the autophagosome formation, SQSTM1 was specifically detected in insoluble but not demonstrated that increase in the levels of LC3-II/ H46R in soluble fractions. Further, insoluble Ser349/Ser351-- LC-I was more prominent in SOD1 mice com- H46R phosphorylated SQSTM1 was significantly increased in pared to SQSTM1;SOD1 mice (Fig. 8a and d). Mitsui et al. Molecular Brain (2018) 11:30 Page 12 of 16 H46R Fig. 8 Accumulation of phosphorylated SQSTM1 in the spinal cord of SOD1 -expressing mice. a Western blot analysis of Ser403(human)/ Ser405(mouse)-phosphorylated SQSTM1 (p-S403/405), Ser349(human)/Ser351(mouse)-phosphorylated SQSTM1 (p-S349/351), LC3, and NQO1 in H46R H46R the spinal cord from mice with four different genotypes; wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1). Two fractions; 1% Triton X-100 soluble and 1% Triton X-100 insoluble/5%SDS soluble fractions were analyzed. GAPDH and β-actin (Actin) were used for a loading control in Triton X-100-soluble and -insoluble fractions, respectively. To properly compare the intensities of signals between phosphorylated SQSTM1 and total SQSTM1, the results from same blots for SQSTM1 in Fig. 7a, GAPDH, and Actin (asterisks) in Fig. 6a are used again as references. b Quantification of 1% Triton X-100 soluble and insoluble S403/S405-phosphorylated SQSTM1. c Quantification of insoluble S349/S351- phosphorylated SQSTM1. d Quantification of the LC3-II to LC3-I ratio (LC3-II/LC3-I) in insoluble fraction. e Quantification of soluble NQO1. b-e Values are mean ± s.e.m. (n = 4) in an arbitrary unit relative to WT mice at 16 weeks of age. Signal intensities were normalized by the levels of GAPDH (soluble fractions) and Actin (insoluble fractions). Statistical significance was evaluated by two-way ANOVA with Bonferroni’s post hoc test (comparisons between different genotypes in the same age; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001) On the other hand, the levels of NQO1, which was conditions in which the levels of Ser349/Ser351-phosph- one of the NFE2L2/Nrf2 target proteins, were af- lylated SQSTM1 is upregulated. It is thus reasonable to fected neither by overexpression of mutant SOD1, assume that phosphorylated SQSTM1 molecules present SQSTM1, nor both of them (Fig. 8a and e,and Add- in the insoluble fraction might be biologically inactive itional file 6:FigureS6g). and are present as inert insoluble aggregates in the H46R These results suggest that overexpression of SQSTM1 spinal cord of SOD1 mice. affects the autophagy activation and/or the impairment of autophagic degradation of LC3-II positive autophago- Discussion somes, whereas it has no impact on the NFE2L2/Nrf2 We have previously shown that loss of SQSTM1 exacer- H46R mediated anti-oxidative stress pathway even in bates disease phenotypes in SOD1 mice [25]. Further, Mitsui et al. Molecular Brain (2018) 11:30 Page 13 of 16 in a SBMA mouse model, loss of SQSTM1 also acceler- proteins [10]. Recently, it has been demonstrated that ates the onset of disease, while overexpression of it ame- the crossbreeding of conditional Atg7 knockout mice G93A liorates disease phenotypes [24]. These findings suggest a with SOD1 mice remarkably extends the lifespan, protective role of SQSTM1 in motor neuron degeneration despite that the onset of hindlimb tremor is acceler- in vivo. In this study, we investigated the effect of systemic ated [33]. Further, the pharmacological induction of overexpression of SQSTM1 on the disease onset and pro- mTOR-independent autophagy has failed to slow the H46R G93A gression in SOD1 mice. Contrary to our expectation, disease progression of SOD1 mice [34]. More im- we revealed that systemic overexpression of SQSTM1 portantly, motor neuron-specific disruption of autoph- resulted in earlier onset and shorter lifespan in a agy displays no phenotypic changes in ALS mice, H46R SOD1 -expressing ALS mouse model. while proteasome inhibition leads to ALS-associated It has been reported that phosphorylation at Ser349 pathology and paralysis [35]. We here showed that H46R (corresponding to mouse Ser351) of SQSTM1 enhances misfolded SOD1 in SQSTM1;SOD1 mice at the binding of SQSTM1 to KEAP1 which in turn acti- 16 weeks of age were significantly higher than those H46R vates the NFE2L2/Nrf2 pathway and NFE2L2/Nrf2 me- in SOD1 mice (Fig. 6c). We also demonstrated diated anti-oxidative stress responses in mice [16]. On that the extensive accumulation of insoluble SQSTM1 H46R the other hand, the previous study on NFE2L2/Nrf2 de- occurred in SQSTM1;SOD1 mice even prior to H46R ficient SOD1 mice has revealed that loss of NFE2L2/ the onset of disease; i.e., at 16 weeks of age (Fig. 7a Nrf2 affects neither the onset of disease nor the progres- and b). Although phosphorylation of human SQSTM1 sion of disease [25]. In this study, despite that the accu- at Ser403 was known to induce autophagic degradation mulation of Ser349/Ser351-phosphorylated SQSTM1 of ubiquitinated proteins [14], the levels of Ser403 (corre- H46R was increased in SQSTM1;SOD1 mice at 22 weeks sponding to mouse Ser405)-phosphorylated SQSTM1 be- H46R H46R of age and end-stage (Fig. 8a and c), there were no dif- tween SQSTM1;SOD1 and SOD1 mice were most ferences in the expression levels of NQO1, which was comparable (Fig. 8a and b). Further, the ratio of LC3-II/ H46R one of the NFE2L2/Nrf2 downstream targets, between LC3-I in SQSTM1;SOD1 mice was lower than that of H46R H46R H46R SOD1 and SQSTM1;SOD1 mice. Further, im- SOD1 mice (Fig. 8a and d) at onset stage; i.e., 22 week munohistochemical analysis revealed that ubiquitin/ of age. These results combined suggest that overexpres- SQSTM1-positive aggregates were not localized in the sion of SQSTM1 does not activate autophagy, but rather cell body and dendrite of neurons but mainly to accelerates the accumulation of SQSTM1 by possibly hin- H46R neuropil (Figs. 4 and 5), indicative of dysfunctional dering the UPS and autophagic degradation in SOD1 SQSTM1 molecules in insoluble aggregates. Taken to- mice. Since the accumulation of autophagosome-like gether, we speculate that progressively-increased structures and SQSTM1-positive aggregates are among Ser349/Ser351-phosphorylated SQSTM1 is physiolo- the cardinal pathological features in patients with ALS as gically inactive, and thus the NFE2L2/Nrf2 anti-oxidative well as in murine ALS models [18, 23], it is reasonable to stress pathway plays a minor role both in the disease onset assume that overexpression of SQSTM1 enhances mutant and progression. SOD1-associated neurotoxicity by further compromising We suppose three possible reasons as to why overex- proteostasis, thereby leading to an earlier onset of disease. pression of SQSTM1 accelerates the onset of disease in On the other hand, SBMA-linked toxic entities; e.g., H46R SOD1 mice, unlike in the case of a SBMA mouse polyglutamine-mediated transcriptional dysregulation model [24]. One possibility is that protein homeostasis [24], may play an indirect role in the regulation of the H46R (proteostasis) in the spinal cord of SOD1 mice is UPS and/or autophagy, and thus the SQSTM1-associated more severely compromised than those in SBMA mice beneficial effect becomes more tangible in SBMA mice. under conditions in which SQSTM1 is overexpressed. A Second possibility is that SQSTM1 expressing in cells large number of studies has revealed that the accumula- and tissues other than neurons affects the onset of dis- H46R tion of aggregated proteins is one of the pathological ease in SOD1 mice. Evidences showing that, in skel- G93A features of neurodegenerative diseases. Misfolded ubi- etal muscle of a SOD1 -expressing ALS mouse quitinated proteins are preferentially degraded by the model, the UPS is activated prior to the onset of disease, ubiquitin-proteasome system (UPS), but undegradable while the autophagic degradation pathway is activated misfolded proteins by the UPS form aggregates, which after the onset of disease till at end-stage [36], sug- in turn are degraded by autophagy [31]. SQSTM1 acts gest a possible pathogenic interaction between pro- as an autophagy adaptor for the degradation of ubiquiti- teostasis in skeletal muscles and neurodegeneration. nated proteins [11]. Thus, in addition to the UPS, autoph- Since the expression level of transgene-coded agy is the important intracellular degradation pathway SQSTM1 in skeletal muscle is higher than those in to maintain proteostasis [32], and its impairment re- other tissues (Additional file 1: Figure S1b), it is possible H46R sults in the accumulation of SQSTM1 and ubiquitinated that the earlier onset of disease in SQSTM1;SOD1 Mitsui et al. Molecular Brain (2018) 11:30 Page 14 of 16 mice is linked to dysregulation of autophagy by SQSTM1 assume that such glial contribution may not play a overexpression in skeletal muscles. However, even if that majorrole intheextensionofsurvivalafter theonset. would be the case, the underlying mechanism has yet to On theother hand,wehavepreviouslyreportedthat H46R be investigated. loss of SQSTM1 in SOD1 mice decreases the Third possibility is that body weight at pre-symptomatic biochemically-detectable insoluble ubiquitinated pro- H46R healthy stages affects the onset of disease in SOD1 teins in the spinal cord in parallel with the accumula- mice. It has been shown that a high fat food feeding tion of ubiquitin-positive inclusions in somal area of in a mutant TAR DNA-binding protein 43-kD motor neurons [25]. Since loss of SQSTM1 exacerbates H46R (TARDBP)-expressing ALS mouse model extends their disease symptoms in SOD1 mice, we consider that the lifespans [37], and that the activation of AMP-activated accumulation of biochemically-detectable insoluble ubi- protein kinase (AMPK), which functions as an energy sen- quitinated proteins is beneficial and/or protective rather sor, results in the extension of lifespans in a mutant than harmful to motor neurons, while the intrasomal SOD1-expressing ALS mouse model [38]. These studies ubiquitin-positive inclusions are poisonous [25]. Indeed, on ALS and energy metabolism suggest that nutritional we showed that biochemically-detectable insoluble conditions affect the disease phenotypes in ALS animal ubiquitinated proteins were increased in the spinal H46R models. Interestingly, we revealed that body weight of cord of SQSTM1;SOD1 mice at end-stage (Fig. 7a SQSTM1 mice tended to be lower than that of WT mice and c). Further, immunohistochemistry revealed that (Fig. 1a and b), and that body weight of female ubiquitin-positive inclusions were exclusively observed H46R SQSTM1;SOD1 mice was significantly lower than that in neuropil but not in somal area of motor neurons H46R H46R of SOD1 mice throughout their pre-symptomatic of SQSTM1;SOD1 mice (Figs. 4 and 5). Thus, the period (Fig. 1a). Taken together, it is possible that the re- effects of loss and overexpression of SQSTM1 on duction of body weight that is associated with SQSTM1 these factors are obviously opposite to each other. overexpression causes the earlier onset of disease in Taken together, it is still possible that the insolubiliza- H46R SOD1 mice. Since loss of SQSTM1 in mice causes hy- tion of poly-ubiquitinated proteins and the exclusion perphagia that results in obese phenotypes [39], a systemic of ubquitinated protein aggregates from neuronal cell overexpression of SQSTM1 may conversely induce hypo- body by SQSTM1 are implicated in some neuropro- phagia thereby decreasing the body weight in mice. In any tective roles, particularly, at the stage of disease pro- cases, pathogenic mechanisms by which overexpression of gression, thereby slowing disease progression by H46R H46R SQSTM1 accelerates the onset in a SOD1 -expressing SQSTM1 overexpression in SOD1 mice. However, ALS mouse model remain unknown, and thus further in order to conclude as such, further studies that in- studies are required. cludes other methods to determine the onset of dis- One of the interesting findings in this study is that ease, such as grip power, the hind-limb reflex, rotarod survival after the onset is slightly extended in retention time, and balance-beam test will be H46R SQSTM1;SOD1 mice, despite that these mice show required. earlier onset of disease (Fig. 1). However, we are cur- rently considering that, although the extension of Conclusions post-onset survival reaches to a statistically significant In conclusions, we demonstrated that, unlike in SBMA level, its reliability may still not be high enough to con- mice, SQSTM1 served as an onset accelerating factor for H46R clude that overexpression of SQSTM1 definitely extends SOD1-linked toxicities, at least, in a SOD1 -expres- survival, because the prediction of disease onset is solely sing ALS mouse model, implying that SQSTM1 upregu- determined by a single measure; i.e., body weight lation might not always to be beneficial to the treatment changes, in this study. Previously, it has been shown that of ALS. mutant SOD1-mediated toxicities in neurons determine the disease onset, whereas those in astrocytes or micro- Additional files glia regulate the disease progression in ALS mouse models [29, 40]. Based on these findings, it could be hy- Additional file 1: Figure S1. Schema of the transgene construct for SQSTM1-tg mouse and the distribution of SQSTM1 in tissues. a Schematic pothesized that mutant SOD1-mediated toxicities in glial diagram of the transgene cassette encoding the C-terminally cells relative to neurons are decreased by systemic overex- hemagglutinin (HA)-tagged human SQSTM1 cDNA. Transgene construct H46R pression of SQSTM1 in SOD1 mice. Although quanti- consists of cytomegalovirus enhancer (E), chicken β-actin promoter (Pro), rabbit β-globin splice acceptor (S), full length human SQSTM1 cDNA tative evaluation of GFAP and Iba1 levels by western (SQSTM1), hemagglutinin tag (HA), and rabbit β-globin poly A (Poly A). b blotting remained to be performed, immunohistochemical Expression and distribution of SQSTM1 in the liver, skeletal muscle, olfactory analysis revealed that the glial cells activation in the spinal bulb, cerebral cortex, hippocampus, cerebellum, and spinal cord in wild-type H46R H46R (WT) and SQSTM1-tg (SQSTM1) mice. Expression of mouse endogenous cord between SOD1 and SQSTM1;SOD1 mice SQSTM1 and human SQSTM1 (Endogenous + SQSTM1-HA) was simultaneously were almost comparable (Figs. 4 and 5). Thus, it is fair to Mitsui et al. Molecular Brain (2018) 11:30 Page 15 of 16 PBS: Phosphate-buffered saline; PFA: Paraformaldehyde; PVDF: Polyvinylidene detected by western blotting using anti-SQSTM1 antibody. Values shown under difluoride; RT: Room temperature; SBMA: Spinal and bulbar muscular atrophy; each lane were the signal intensities in an arbitrary unit relative to that of the SDS: Sodium dodecyl sulfate; SQSTM1: Sequestosome 1/p62; TARDBP: TAR liver in WT mouse. Human SQSTM1 (hSQSTM1-HA) was specifically detected by DNA-binding protein 43-kD; tg: Transgenic; UPS: Ubiquitin-proteasome anti-HA antibody. GAPDH was used as a loading control. (PDF 795 kb) system; WT: Wild-type Additional file 2: Figure S2. Whole transverse sectional images of the lumbar spinal cord in Figs. 2, 3 and 4. a Whole transverse sectional images of Acknowledgments fluorescence Nissl staining in the lumbar spinal cords in Fig. 2. b Whole We thank Dr. Makoto Urushitani at Shiga University of Medical Science for transverse sectional images of double immunostaining with Ubiquitin (green) the generous gift of anti-misfolded SOD1 antibody (C4F6), and all the mem- andSQSTM1(red)in thelumbarspinalcordinFig. 3. c Whole transverse bers of Support Center for Medical Research and Education at Tokai Univer- sectional images of double immunostaining with MAP2 (green) and Ubiquitin sity for their technical help. (red), Ubiquitin (green) and GFAP (red), and Ubiquitin (green) and Iba1 (red) in the lumbar spinal cord in Fig. 4. a-c Scale bars = 300 μm. b, c The nuclei were Funding counterstained with DAPI (blue). (PDF 7238 kb) This study was supported by Japanese Society for Promotion of Science Additional file 3: Figure S3. SQSTM1-HA positive aggregates in the (JSPS) [grant numbers 26290018 and 24650189 to SH], and the National anterior horn of lumbar spinal cord. a-c Representative images of double Natural Science Foundation of China (NSFC) and JSPS Bilateral Joint Research immunostaining with Ubiquitin (green) and HA (SQSTM1-HA; red) in the Project (to FHS and SH). H46R lumbar spinal cord (L4–5) from wild-type (WT), SQSTM1 (SQSTM1), SOD1 H46R (H46R) and SQSTM1;SOD1 (SQSTM1;H46R)miceat16weeks of age (wk) Availability of data and materials (a), 22 wk. (b), and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and The datasets supporting the conclusions of this article is included within the SQSTM1) (c). The nuclei were counterstained with DAPI (blue). Scale article and its Additional files. bars = 50 μm. Ubiquitin-positive aggregates and SQSTM1-HA aggregates were H46R observed in the anterior horn of SQSTM1;SOD1 (SQSTM1;H46R) mice at 22 Authors’ contributions wk. and end-stage. Arrows indicate that ubiquitin-positive aggregates coloca- lizing with SQSTM1-HA. (PDF 1718 kb) Conceived and designed the experiments: SM, AO, FHS and SH. Performed the experiments: SM, MN, SO, KS, RS and AO. Analyzed the data: SM. Additional file 4: Figure S4. Immunoblot-images used for the quantitative Contributed reagents/materials/analysis tools: HA, MA, GS and SH. Wrote the analysis in Figs. 6a and Fig. 7. The immunoblots of (a)soluble SOD1,(b) paper: SM and SH. All authors reviewed and approved the manuscript. insoluble SOD1 monomer (Mono), (c) insoluble high-molecular weight (HMW) SOD1, (d) soluble SQSTM1, (e) insoluble SQSTM1, (f) soluble poly-ubiquitinated Ethics approval proteins, (g) insoluble poly-ubiquitinated proteins, (h)soluble GAPDH, and(i) All animal experimental procedures were carried out in accord with the insoluble actin were analyzed for the quantitative analysis (Figs. 6a and 7). The H46R Fundamental Guidelines for Proper Conduct of Animal Experiment and spinal cord from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), and H46R Related Activities in Academic Research Institutions under the jurisdiction of SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., and the Ministry of Education, Culture, Sports, Science and Technology (MEXT), end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1) were Japan, and reviewed and approved by The Institutional Animal Care and Use used. C (control sample) used as internal control indicates soluble and H46R Committee at Tokai University. insoluble fractions from 22 week-old SQSTM1;SOD1 mouse. Asterisk represents the measured area of poly-ubiquitinated proteins. (PDF 1973 kb) Competing interests Additional file 5: Figure S5. Immunoblot-images used for the quantita- The authors declare that they have no competing interests. tive analysis in Fig. 6b. The immunoblots of (a) soluble misfolded SOD1 and GAPDH, (b) insoluble misfolded SOD1 and actin were analyzed for H46R the quantitative analysis (Fig. 6b). The spinal cord from SOD1 (H46R), H46R Publisher’sNote and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 Springer Nature remains neutral with regard to jurisdictional claims in wk., and end-stage were used. (PDF 444 kb) published maps and institutional affiliations. Additional file 6: Figure S6. Immunoblot-images used for the quanti- tative analysis in Fig. 8. The immunoblots of (a)soluble and(b)insoluble Author details Ser403(human)/Ser405(mouse)-phosphorylated SQSTM1 (p-S403/405), (c) Department of Molecular Life Sciences, Tokai University School of Medicine, soluble and (d) insoluble Ser349(human)/Ser351(mouse)-phosphorylated Isehara, Kanagawa 259-1193, Japan. The Institute of Medical Sciences, Tokai SQSTM1 (p-S349/351), (e) soluble LC3, (f) insoluble LC3, (g) soluble NQO1, University, Isehara, Kanagawa 259-1193, Japan. Micro/Nano Technology and (h) insoluble NQO1 were analyzed for the quantitative analysis (Fig. 7). Center, Tokai University, Hiratsuka, Kanagawa 259-1292, Japan. Department H46R The spinal cord from wild-type (WT), SQSTM1 (SQSTM1), SOD1 (H46R), of Anesthesiology, Tokai University School of Medicine, Isehara, Kanagawa H46R and SQSTM1;SOD1 (SQSTM1;H46R) mice at 16 weeks of age (wk), 22 wk., 259-1193, Japan. Department of Neurology, University of Occupational and and end-stage (H46R and SQSTM1;H46R) or 28 wk. (WT and SQSTM1) were Environmental Health School of Medicine, Kitakyushu, Fukuoka 807-0804, used. C (control sample) used as internal control indicates soluble and Japan. Department of Neurology, Tohoku University Graduate School of H46R insoluble fractions from 22 week-old SQSTM1;SOD1 mouse. (PDF 607 kb) Medicine, Sendai, Miyagi 980-8575, Japan. Department of Neurology, Nagoya University Graduate School of Medicine, Nagoya, Aichi 466-8550, Japan. Department of Neurology, West China Hospital, Sichuan University, Abbreviations Chengdu 610041, Sichuan, China. Research Center for Brain and Nervous ALS: Amyotrophic lateral sclerosis; AMPK: Adenosine mono-phosphate acti- Diseases, Tokai University Graduate School of Medicine, Kanagawa, Isehara vated protein kinase; ANOVA: Analysis of variance; BPB: Bromophenol blue; 259-1193, Japan. CAG: (C) the cytomegalovirus (CMV) early enhancer element, (A) the promoter, the first exon and the first intron of chicken beta-actin gene, and Received: 12 March 2018 Accepted: 20 May 2018 (G) the splice acceptor of the rabbit beta-globin gene; DAPI: 4′,6-diamidino- 2-phenylindole dihydrochloride; FTD: Frontotemporal dementia; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; GFAP: Glial fibrillary acidic protein; HA: Hemagglutinin; HRP: Horseradish peroxidase; Iba1: Ionized References calcium binding adaptor molecule 1; KEAP1: Kelch-like ECH-associated pro- 1. Robberecht W, Philips T. The changing scene of amyotrophic lateral tein 1; MAP1LC3/LC3: Microtubule-associated protein 1 light chain 3; sclerosis. Nat Rev Neurosci. 2013;14:248–64. MAP2: Microtubule-associated protein 2; MND: Motor neuron disease; 2. Rowland LP, Shneider NA. Amyotrophic lateral sclerosis. N Engl J Med. 2001; NFE2L2/Nrf2: Nuclear factor related erythroid 2-related factor 2; NGS: Normal 344:1688–700. goat serum; NQO1: NAD(P)H dehydrogenase [quinone] 1; 3. Renton AE, Chio A, Traynor BJ. State of play in amyotrophic lateral sclerosis PAGE: Polyacrylamide gel electrophoresis; PB: Phosphate buffer; genetics. Nat Neurosci. 2014;17:17–23. Mitsui et al. Molecular Brain (2018) 11:30 Page 16 of 16 4. Fecto F, Yan J, Vemula SP, Liu E, Yang Y, Chen W, Zheng JG, Shi Y, Siddique 25. Hadano S, Mitsui S, Pan L, Otomo A, Kubo M, Sato K, Ono S, Onodera W, Abe N, Arrat H, et al. SQSTM1 mutations in familial and sporadic amyotrophic K, Chen X, et al. Functional links between SQSTM1 and ALS2 in the lateral sclerosis. Arch Neurol. 2011;68:1440–6. pathogenesis of ALS: cumulative impact on the protection against mutant 5. Hirano M, Nakamura Y, Saigoh K, Sakamoto H, Ueno S, Isono C, Miyamoto K, SOD1-mediated motor dysfunction in mice. Hum Mol Genet. 2016;25:3321–40. Akamatsu M, Mitsui Y, Kusunoki S. Mutations in the gene encoding p62 in 26. Urushitani M, Ezzi SA, Julien JP. Therapeutic effects of immunization with Japanese patients with amyotrophic lateral sclerosis. Neurology. 2013;80: mutant superoxide dismutase in mice models of amyotrophic lateral 458–63. sclerosis. Proc Natl Acad Sci U S A. 2007;104:2495–500. 27. Pan L, Yoshii Y, Otomo A, Ogawa H, Iwasaki Y, Shang HF, Hadano S. 6. Shimizu H, Toyoshima Y, Shiga A, Yokoseki A, Arakawa K, Sekine Y, Different human copper-zinc superoxide dismutase mutants, SOD1G93A Shimohata T, Ikeuchi T, Nishizawa M, Kakita A, et al. Sporadic ALS with and SOD1H46R, exert distinct harmful effects on gross phenotype in mice. compound heterozygous mutations in the SQSTM1 gene. Acta PLoS One. 2012;7:e33409. Neuropathol. 2013;126:453–9. 28. Boillee S, Yamanaka K, Lobsiger CS, Copeland NG, Jenkins NA, Kassiotis G, 7. Chen Y, Zheng ZZ, Chen X, Huang R, Yang Y, Yuan L, Pan L, Hadano S, Kollias G, Cleveland DW. Onset and progression in inherited ALS Shang HF. SQSTM1 mutations in Han Chinese populations with sporadic determined by motor neurons and microglia. Science. 2006;312:1389–92. amyotrophic lateral sclerosis. Neurobiol Aging. 2014;35:726. e7–9 29. Yamanaka K, Chun SJ, Boillee S, Fujimori-Tonou N, Yamashita H, Gutmann 8. Yang Y, Tang L, Zhang N, Pan L, Hadano S, Fan D. Six SQSTM1 mutations in DH, Takahashi R, Misawa H, Cleveland DW. Astrocytes as determinants of a Chinese amyotrophic lateral sclerosis cohort. Amyotroph Lateral Scler disease progression in inherited amyotrophic lateral sclerosis. Nat Neurosci. Frontotemporal Degener. 2015;16:378–84. 2008;11:251–3. 9. Laurin N, Brown JP, Morissette J, Raymond V. Recurrent mutation of the 30. Kurosawa M, Matsumoto G, Sumikura H, Hatsuta H, Murayama S, Sakurai T, gene encoding sequestosome 1 (SQSTM1/p62) in Paget disease of bone. Shimogori T, Hattori N, Nukina N. Serine 403-phosphorylated p62/SQSTM1 Am J Hum Genet. 2002;70:1582–8. immunoreactivity in inclusions of neurodegenerative diseases. Neurosci Res. 10. Komatsu M, Waguri S, Koike M, Sou YS, Ueno T, Hara T, Mizushima N, Iwata J, 2016;103:64–70. Ezaki J, Murata S, et al. Homeostatic levels of p62 control cytoplasmic inclusion 31. Lystad AH, Simonsen A. Assays to monitor aggrephagy. Methods. 2015;75:112–9. body formation in autophagy-deficient mice. Cell. 2007;131:1149–63. 32. Mizushima N, Komatsu M. Autophagy: renovation of cells and tissues. Cell. 11. Pankiv S, Clausen TH, Lamark T, Brech A, Bruun JA, Outzen H, Overvatn A, 2011;147:728–41. Bjorkoy G, Johansen T. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate 33. Rudnick ND, Griffey CJ, Guarnieri P, Gerbino V, Wang X, Piersaint JA, Tapia degradation of ubiquitinated protein aggregates by autophagy. J Biol JC, Rich MM, Maniatis T. Distinct roles for motor neuron autophagy early Chem. 2007;282:24131–45. and late in the SOD1(G93A) mouse model of ALS. Proc Natl Acad Sci U S A. 12. Komatsu M, Kurokawa H, Waguri S, Taguchi K, Kobayashi A, Ichimura Y, Sou 2017;114:E8294–E303. YS, Ueno I, Sakamoto A, Tong KI, et al. The selective autophagy substrate 34. Perera ND, Sheean RK, Lau CL, Shin YS, Beart PM, Horne MK, Turner BJ. p62 activates the stress responsive transcription factor Nrf2 through Rilmenidine promotes MTOR-independent autophagy in the mutant SOD1 inactivation of Keap1. Nat Cell Biol. 2010;12:213–23. mouse model of amyotrophic lateral sclerosis without slowing disease 13. Goode A, Butler K, Long J, Cavey J, Scott D, Shaw B, Sollenberger J, Gell C, progression. Autophagy. 2018;14:534–51. Johansen T, Oldham NJ, et al. Defective recognition of LC3B by mutant 35. Tashiro Y, Urushitani M, Inoue H, Koike M, Uchiyama Y, Komatsu M, Tanaka SQSTM1/p62 implicates impairment of autophagy as a pathogenic K, Yamazaki M, Abe M, Misawa H, et al. Motor neuron-specific disruption of mechanism in ALS-FTLD. Autophagy. 2016;12:1094–104. proteasomes, but not autophagy, replicates amyotrophic lateral sclerosis. J 14. Matsumoto G, Wada K, Okuno M, Kurosawa M, Nukina N. Serine 403 Biol Chem. 2012;287:42984–94. phosphorylation of p62/SQSTM1 regulates selective autophagic clearance of 36. Olivan S, Calvo AC, Gasco S, Munoz MJ, Zaragoza P, Osta R. Time-point ubiquitinated proteins. Mol Cell. 2011;44:279–89. dependent activation of autophagy and the UPS in SOD1G93A mice 15. Matsumoto G, Shimogori T, Hattori N, Nukina N. TBK1 controls skeletal muscle. PLoS One. 2015;10:e0134830. autophagosomal engulfment of polyubiquitinated mitochondria through 37. Coughlan KS, Halang L, Woods I, Prehn JH. A high-fat jelly diet restores p62/SQSTM1 phosphorylation. Hum Mol Genet. 2015;24:4429–42. bioenergetic balance and extends lifespan in the presence of motor 16. Ichimura Y, Waguri S, Sou YS, Kageyama S, Hasegawa J, Ishimura R, Saito T, dysfunction and lumbar spinal cord motor neuron loss in TDP-43A315T Yang Y, Kouno T, Fukutomi T, et al. Phosphorylation of p62 activates the mutant C57BL6/J mice. Dis Model Mech. 2016;9:1029–37. Keap1-Nrf2 pathway during selective autophagy. Mol Cell. 2013;51:618–31. 38. Coughlan KS, Mitchem MR, Hogg MC, Prehn JH. “Preconditioning” with 17. Mizuno Y, Amari M, Takatama M, Aizawa H, Mihara B, Okamoto K. latrepirdine, an adenosine 5′-monophosphate-activated protein kinase Immunoreactivities of p62, an ubiqutin-binding protein, in the spinal activator, delays amyotrophic lateral sclerosis progression in SOD1(G93A) anterior horn cells of patients with amyotrophic lateral sclerosis. J Neurol mice. Neurobiol Aging. 2015;36:1140–50. Sci. 2006;249:13–8. 39. Harada H, Warabi E, Matsuki T, Yanagawa T, Okada K, Uwayama J, Ikeda A, 18. Sasaki S. Autophagy in spinal cord motor neurons in sporadic amyotrophic Nakaso K, Kirii K, Noguchi N, et al. Deficiency of p62/Sequestosome 1 causes lateral sclerosis. J Neuropathol Exp Neurol. 2011;70:349–59. hyperphagia due to leptin resistance in the brain. J Neurosci. 2013;33: 19. Al-Sarraj S, King A, Troakes C, Smith B, Maekawa S, Bodi I, Rogelj B, Al- 14767–77. Chalabi A, Hortobagyi T, Shaw CE. p62 positive, TDP-43 negative, neuronal 40. Yamanaka K, Boillee S, Roberts EA, Garcia ML, McAlonis-Downes M, Mikse cytoplasmic and intranuclear inclusions in the cerebellum and OR, Cleveland DW, Goldstein LS. Mutant SOD1 in cell types other than hippocampus define the pathology of C9orf72-linked FTLD and MND/ALS. motor neurons and oligodendrocytes accelerates onset of disease in ALS Acta Neuropathol. 2011;122:691–702. mice. Proc Natl Acad Sci U S A. 2008;105:7594–9. 20. Kato S. Amyotrophic lateral sclerosis models and human neuropathology: similarities and differences. Acta Neuropathol. 2008;115:97–114. 21. Aoki M, Ogasawara M, Matsubara Y, Narisawa K, Nakamura S, Itoyama Y, Abe K. Mild ALS in Japan associated with novel SOD mutation. Nat Genet. 1993;5:323–4. 22. Sasaki S, Nagai M, Aoki M, Komori T, Itoyama Y, Iwata M. Motor neuron disease in transgenic mice with an H46R mutant SOD1 gene. J Neuropathol Exp Neurol. 2007;66:517–24. 23. Hadano S, Otomo A, Kunita R, Suzuki-Utsunomiya K, Akatsuka A, Koike M, Aoki M, Uchiyama Y, Itoyama Y, Ikeda JE. Loss of ALS2/Alsin exacerbates motor dysfunction in a SOD1-expressing mouse ALS model by disturbing endolysosomal trafficking. PLoS One. 2010;5:e9805. 24. Doi H, Adachi H, Katsuno M, Minamiyama M, Matsumoto S, Kondo N, Miyazaki Y, Iida M, Tohnai G, Qiang Q, et al. p62/SQSTM1 differentially removes the toxic mutant androgen receptor via autophagy and inclusion formation in a spinal and bulbar muscular atrophy mouse model. J Neurosci. 2013;33:7710–27.

Journal

Molecular BrainSpringer Journals

Published: May 29, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off