Loss of Tmem106b is unable to ameliorate frontotemporal dementia-like phenotypes in an AAV mouse model of C9ORF72-repeat induced toxicity

Loss of Tmem106b is unable to ameliorate frontotemporal dementia-like phenotypes in an AAV mouse... Loss-of-function mutations in progranulin (GRN) and a non-coding (GGGGCC) hexanucleotide repeat expansions in C9ORF72 are the two most common genetic causes of frontotemporal lobar degeneration with aggregates of TAR DNA binding protein 43 (FTLD-TDP). TMEM106B encodes a type II transmembrane protein with unknown function. Genetic variants in TMEM106B associated with reduced TMEM106B levels have been identified as disease modifiers in individuals with GRN mutations and C9ORF72 expansions. Recently, loss of Tmem106b has been reported to protect the FTLD-like phenotypes in Grn−/− mice. Here, we generated Tmem106b−/− mice and examined whether loss of Tmem106b could rescue FTLD-like phenotypes in an AAV mouse model of C9ORF72-repeat induced toxicity. Our results showed that neither partial nor complete loss of Tmem106b was able to rescue behavioral deficits induced by the expression of (GGGGCC) repeats (66R). Loss of Tmem106b also failed to ameliorate 66R-induced RNA foci, dipeptide repeat protein formation and pTDP-43 pathological burden. We further found that complete loss of Tmem106b increased astrogliosis, even in the absence of 66R, and failed to rescue 66R-induced neuronal cell loss, whereas partial loss of Tmem106b significantly rescued the neuronal cell loss but not neuroinflammation induced by 66R. Finally, we showed that overexpression of 66R did not alter expression of Tmem106b and other lysosomal genes in vivo, and subsequent analyses in vitro found that transiently knocking down C9ORF72, but not overexpression of 66R, significantly increased TMEM106B and other lysosomal proteins. In summary, reducing Tmem106b levels failed to rescue FTLD-like phenotypes in a mouse model mimicking the toxic gain-of-functions associated with overexpression of 66R. Combined with the observation that loss of C9ORF72 and not 66R overexpression was associated with increased levels of TMEM106B, this work suggests that the protective TMEM106B haplotype may exert its effect in expansion carriers by counteracting lysosomal dysfunction resulting from a loss of C9ORF72. * Correspondence: rademakers.rosa@mayo.edu Alexandra M. Nicholson and Xiaolai Zhou contributed equally to this work. Department of Neuroscience, Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA 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. Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 2 of 14 Introduction aimed to examine whether loss of Tmem106b expression Frontotemporal dementia (FTD) is a devastating neuro- was able to rescue FTD-like behavioral and pathological degenerative disorder with initial symptoms occurring in features observed in an adeno-associated virus (AAV)-- the fifth or sixth decade of life. While most cases of FTD based mouse model mimicking the toxic gain-of-functions develop sporadically, 30–50% of FTD cases report a associated with overexpression of (GGGGCC) repeats. family history [23, 43, 44, 47, 61], in support of a strong genetic component to the disease. Two of the most Methods common gene mutations found to cause FTD reside in Tmem106b knockout mice the progranulin (GRN) and chromosome 9 open reading Tmem106b knockout mice were generated at the Knockout frame 72 (C9ORF72) genes [6, 14, 17, 45]. Causative Mouse Project (KOMP) Repository at the University of GRN mutations leading to FTD include heterozygous California, Davis using the PGS00041_A_C06 targeting vec- missense, nonsense, or frameshift changes that most tor and blastocyst injection of the targeted embryonic stem often lead to nonsense-mediated decay of the mutant cell clone EPD0047_1_E02 generated from C57BL/6 N mRNA and an associated loss of progranulin protein mice. This knock-in first strategy results in the insertion of (PGRN). Individuals with GRN mutations invariably a lacZ gene trap between the first two coding exons (exons present with aggregates of the TAR DNA binding protein 3 and 4) of the mouse Tmem106b gene. Cryopreserved 43 (TDP-43) in affected brain regions, and are thus patho- sperm were purchased and used to inseminate oocytes ob- logically classified as FTLD-TDP [4, 36]. In C9ORF72,a tained from 3-week-old C57BL/6N female mice (Harlan, non-coding (GGGGCC) hexanucleotide repeat expansion Indianapolis, IN). Zygotes that reached the 2-cell-stage 24 h is responsible for up to 25% of familial and 5% of sporadic post insemination were surgically transferred into foster FTD patients [17, 45]. Extensive research has shown that dams (Harlan). DNA obtained from subsequent pups was the presence of these expanded repeats leads to multiple screened by multiplex polymerase chain reaction (PCR) for pathogenic mechanisms, including a loss of C9ORF72 thepresenceofthe NEOcassette beforebreedingasacol- mRNA expression and toxic gain-of-functions resulting ony founder (CSD-Tmem106b-F: 5’-TTCTCTCCATGTGC from nuclear RNA aggregates and dipeptide repeats pro- TGCATTATGAGC-3′; CSD-Neo-F: 5’-GGGATCTCATG teins [5, 17, 33, 34, 45]. FTD patients with C9ORF72 CTGGAGTTCTTCG-3′; CDS-Tmem106b-ttR: 5’-ACGTG expansions also present with FTLD-TDP at autopsy, CTTCTCTCATCTACAGTTTTCC-3′). A Tmem106b+/− suggesting a potentially convergent disease mechanism x Tmem106+/− breeding scheme was used to generate between GRN- and C9ORF72-induced pathogenesis. Tmem106b +/+, +/−,and −/− mice for the experiments. In 2010, a genome-wide association study (GWAS) Both male and female mice of each Tmem106b genotype identified genetic variants at the transmembrane protein were used for all the experiments. All animal studies were 106 B (TMEM106B) gene locus as the first genetic mod- approved by the Mayo Clinic Institutional Animal Care and ifiers of FTLD-TDP [59]. TMEM106B variants were Use Committee. found to be a modifier of disease risk in FTLD-TDP pa- tients of unknown cause, and a modifier of disease pene- Genotyping trance and presentation in GRN mutation and C9ORF72 Genomic DNA (gDNA) was extracted and PCR-amplified expansion carriers [13, 19, 21, 30, 37, 58–60]. Specific- using the Phire Tissue Direct Master Mix kit (Thermo ally, in C9ORF72 carriers, we showed that individuals Scientific Inc., Waltham, MA) per the manufacturer’s who were also homozygous for the minor alleles at the instructions. Briefly, mouse hair follicles were digested associated TMEM106B variants were significantly pro- in Dilution Buffer supplemented with DNARelease tected from developing FTD but not amyotrophic lateral Additive for 2 min at room temperature, followed by sclerosis (ALS) symptoms [18, 58], another common 2 min incubation at 95 °C. Samples were briefly phenotypic presentation in C9ORF72 expansion carriers. centrifuged and 1 μl of supernatant containing the The TMEM106B protein resides in lysosomal com- gDNA was used for each PCR reaction. Tmem106b partments where it might be involved in lysosomal func- gene products were amplified using a multiplex PCR tion and/or trafficking [7, 11, 29, 50, 53]. Overexpression approach containing 0.8 μMofeachforward primer of TMEM106B results in abnormal lysosomal size, num- (CSD-Tmem106b-F and CSD-Neo-F), 0.8 μMof reverse ber, and acidification [7, 11]. Interestingly, recent studies primer (CDS-Tmem106b-ttR), and Phire Tissue Direct determined that the protective TMEM106B variants are PCR Master Mix (Thermo Scientific). associated with reduced levels of TMEM106B [20, 37, 59], suggesting that lowering TMEM106B might be thera- Viral production and injections peutic in the context of FTD. In fact, lysosomal deficits Viruses were generated as previously described [12, 22, 54]. observed in Grn knockout mice were recently rescued by Briefly, (GGGGCC) (2R) or (GGGGCC) (66R) C9ORF72 2 66 loss of Tmem106b expression [26]. In this study, we repeats were cloned into the pAM/CBA-pl-WPRE-BGH Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 3 of 14 vector containing inverted repeats of serotype 2. AAV vec- cages and transferred to a different room with reduced tors containing the repeats were packaged into the serotype light for at least 1 h. Contextual cues were changed by 9 type capsid by co-transfection with helper plasmids into altering the environment, shape, and smell of the testing HEK293T cells. The cells were harvested and lysed 2 days chamber, as well as by covering the chamber floor with post transfection in the presence of 0.5% sodium deoxycho- opaque plastic. Each mouse was placed back into the test late and 50 U/ml Benzonase (Sigma Aldrich, St. Louis, chamber and the auditory stimulus was presented. Freez- MO) by freeze thawing. The virus was isolated using a ing was recorded for 3 min (cued test). For both the con- discontinuous iodixanol gradient, and qPCR was used to text and cued tests, baseline freezing time was subtracted determine the genomic titer of each virus. 2R and 66R from the freezing time obtained during each test. AAV were diluted to 1 13 genomes/ml in sterile phosphate-buffered saline (PBS) before injection. Mouse Social interaction test pups underwent intracerebroventricular injections with Each mouse was placed into a rectangular box subdi- virus at postnatal day 0 (P0) [10, 12, 25]. Pups were cryoa- vided into three chambers. Two larger chambers nesthetized on ice and their heads were wiped with a sterile measured 17 × 40 cm with a smaller chamber of 5 × 40 alcohol pad. Two microliters of virus were manually cm in the middle. The three chambers were connected injected into each cerebral ventricle using a 32 gauge needle by an 8 × 5 cm opening to allow the mouse free access attached to a 10 μl syringe (Hamilton Company, to all chambers. Two empty, inverted wire-mesh cylin- Reno,NV).After injection,pups were warmed on a ders were placed in opposite corners of each large cham- heating pad and placed back with the dam. All litters ber. In the first trial, mice were placed in the box and were injected within an 8-day timeframe and mice allowed to explore the apparatus freely for 4 min before were aged to 12 months before assessing behavior being placed into a temporary holding cage. Next, a and pathological manifestations. Small subsets of mice probe mouse (matched for sex/strain) was placed in one were harvested at 3 months of age to study of the cylinders for 3 min prior to reintroduction of the Tmem106b expression and validate the model. test mouse. An overhead camera and Anymaze software (Stoelting Co.) were used to monitor mouse interactions Open field test for 10 min. The time each test mouse spent in the area Mice were placed in a square, Perspex box (40x40x30cm, containing the cylinder with the probe mouse was used LxWxH) containing side-mounted photobeams placed to determine sociability. 7.6 cm above the bottom of the box. Mice were allowed to move freely for 15 min, during which locomotor activity Tissue harvests and anxiety measurements were taken. The Perspex box Mice were subjected to carbon dioxide narcosis and was illuminated by a light suspended over the chamber, body weight was obtained (SCALTEC SBA 53 scale; and an overhead camera and AnyMaze software (Wood Denver Instrument, Bohemia, NY) before decapitation. Dale, IL) were used to monitor mouse movement, such as Mouse blood was collected in tubes containing 1.6 mg/ time mobile, total distance traveled, and distance traveled ml EDTA and placed on ice. Blood samples were centri- in the outer and center zones. Mouse rearing was fuged at 4 °C for 10 min at 5000 rpm, after which the recorded by breaking of the photobeams. resulting plasma supernatant was transferred to a new tube for storage at -80 °C until use. The brain was re- Conditional fear testing moved and its weight recorded (SCALTEC SBC 32 scale; Each mouse was placed in a sound-reducing chamber Denver Instrument) before separating the hemispheres. containing a grid floor capable of inducing an electric Whole mouse brains from uninjected mice were either shock. An overhead camera and FreezeFrame software immediately dehydrated and flash frozen in a beaker of (Actimetrics, Wilmette, IL) were used to measure freez- isopentane on dry ice or fixed for 24 h at 4 °C in 4% para- ing. The mice were left undisturbed for the first 2 min formaldehyde (PFA) prepared in PBS. For all injected of the test and baseline freezing was recorded. An 80-dB mice, the left hemisphere was fixed in PFA at 4 °C for white noise was then administered for 30 s (conditioned 48 h. The right hemisphere was immediately dehydrated stimulus; CS). During the last 2 s of the CS, a 0.5 mA and flash frozen in a beaker of isopentane on dry ice. Fol- foot shock was administered to the mouse (uncondi- lowing PFA fixation, brain tissues were washed and stored tioned stimulus; US). After 1 min, a second CS-US pair in PBS at 4 °C until being embedded in paraffin wax. was given to the mouse and the mouse was removed from the chamber and placed in his/her home cage 30 s Cell culture and transfection later. Each mouse was returned to the testing chamber HeLa and U251 cells (ATCC, Manassas, VA) were cul- 24 h later and freezing behavior was recorded for 5 min tured and maintained in Eagle’s Minimum Essential (context test). All mice were returned to their home Medium (EMEM) supplemented with 10% FBS and 1% Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 4 of 14 penicillin/streptavidin. All cell lines were maintained at Mm00479862_g1), Gfap (Mm01253033_m1), and Gapdh 37 °C, 5% CO . For overexpression studies, cells were (Mm99999915_g1). transiently transfected with pAAV C9ORF72 2R or pAAV C9ORF72 66R using Lipofectamine 2000 (Invitrogen, Western blotting Carlsbad, CA) by mixing DNAs with the transfection Protein samples were mixed with an equivalent volume reagent in OptiMEM (Life Technologies, Carlsbad, CA, of 2X Novex sample buffer (Life Technologies) supple- USA) according to the manufacturer’sprotocol. Forthe mented to 5% β-mercaptoethanol. Proteins were dena- siRNA knockdown experiments, HeLa and U251 cells tured by incubating at room temperature for 30 min or were transfected with 20 nM of either negative control by heating at 95 °C for 1–5 min before loading into siRNA or siRNAs against human C9ORF72 using SDS-polyacrylamide gels (Life Technologies). Proteins Lipofectamine RNAiMAX Reagent (Life Technologies) were transferred to Immobilon membranes (Millipore, according the manufacturer’s protocol. The control siRNA Billerica, MA) and immunoblotted with the primary (5’-UGGUUUACAUGUCGACUAA-3′, D-001210-05) and antibody at 4 °C overnight. The next day, blots were human C9ORF72 siRNA (5’-CAUAGAGUGUGUGUUG incubated with an HRP-conjugated secondary antibody AUA-3′, J-013341-11) were purchased from Dharmacon (Promega, Madison, WI) and bands were detected by (Lafayette, CO). Cells were harvested for protein extraction enhanced chemiluminescence using Western Lightning 48 and 72 h post transfection for overexpression and Plus-ECL reagents (Perkin Elmer, Waltham, MA). Primary siRNA experiments, respectively. antibodies included: rabbit anti-Tmem106b from Bethyl Laboratories (A303-439A), rabbit anti-Tmem106b gener- RNA and protein extraction ously shared and derived in the lab of Dr. Fenghua Hu, Frozen brain tissue was homogenized by sonication in sheep anti-mouse progranulin (AF2557; R&D sys- tris-buffered saline (TBS) containing 2X protease and tems, Minneapolis, MN), goat anti-human progranu- phosphatase inhibitors (Thermo Scientific). RNA was lin (AF2420; R&D systems, Minneapolis, MN), mouse isolated from 75 μl of brain homogenate using the anti-Gapdh (H86504M; Meridian Life Sciences, Cincinnati, RNeasy Plus Mini Kit (Qiagen, #74136) according to the OH), mouse anti-Lamp1 (sc-20,011; Santa Cruz Biotech- manufacturer’s instructions. Briefly, brain tissue was nology, Dallas, TX), rabbit anti-C9ORF72 (ABN1645; lysed using Buffer RLT containing β-ME, and then Millipore), mouse anti-HA (clone12CA5; #11583816001; passed through the gDNA Eliminator column to remove Roche, Indianapolis, IN), and goat anti-Cathepsin-D (clone DNA. The RNA containing flow-through was precipi- C-20; sc-6486; Santa Cruz Biotechnology, Dallas, TX). tated by 70% ethanol and passed through an RNeasy Bands of Western blots were quantified using Image J pink spin column. RNA was eluted from the column (NIH, Bethesda, MD). with RNase-free water. Protein was also isolated from 75 μl of brain homogenate by adding and equivalent vol- Poly(GP) immunoassay ume of 2X Radioimmunoprecipitation Assay (RIPA) Poly(GP) protein levels were measured in 10 μgof buffer (Boston BioProducts, Ashland, MA). For cell protein in duplicate from mouse brain lysates using a culture experiments, media was removed and RIPA buf- sandwich immunoassay utilizing MesoScale Discovery fer was added directly to PBS-rinsed cell culture wells. (MSD) technology as previously described [12, 22]. Serial All RIPA samples were incubated on ice and centrifuged dilutions of recombinant (GP) were used as a standard at 4 °C for 5 min at 6000 rpm to clear debris. Protein curve. Response values were measured using the MSD content in brain samples was measured in the super- QUICKPLEX SQ120 and are defined as the intensity of natant using the bicinchoninic acid (BCA) assay emitted light upon electrochemical stimulation. Each (Thermo Fisher Scientific). sample’s response value was corrected for background response detected in 2R-injected mouse samples prior to Quantitative PCR interpolation of poly(GP) levels using the standard Brain RNA was reverse transcribed using the Superscript curve. III complimentary DNA (cDNA) synthesis kit, random hexamers (Life Technologies), and an equal ratio of random Immunohistochemistry and digital analysis hexamers and Oligo dT primers (Thermo Fisher Scientific). Paraffin-embedded mouse brain tissues were cut on a Real-time quantitative PCRs (qPCRs) were conducted using sagittal plane at a thickness of 5 μm, deparaffinized with TaqMan gene expression assays and the QuantStudio 7 xylene, and rehydrated in a series of ethanol washes. For Flex Real-Time PCR System (Applied Biosystems, Foster slides stained with mouse anti-NeuN (ABN78; Millipore) City, CA). All probes were purchased from Life Technolo- or rabbit anti-pTdp-43 (TIP-PTD-P01; pSer409/410; gies: Tmem106a (Mm01246747_m1), Tmem106b (Mm005 Cosmo Bio USA, Carlsbad, CA), antigen retrieval was 10952_m1), Tmem106c (Mm01303550_m1), Iba1 (Aif1; performed by steaming slides for 30 min with distilled Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 5 of 14 water before blocking in 0.03% hydrogen peroxide. Immu- under 63× magnification and the number of cells con- nostaining of sections was done using a Dako Autostainer taining RNA foci were quantified. and Envision + HRP system (Dako, Carpintaria, CA) per the manufacturer’s instructions. For slides stained with Statistical analyses rabbit anti-pTdp-43 (pSer409/410; gift from Dr. Leonard For experiments in which only two groups were com- Petrucelli), antigen retrieval was performed by steaming pared, significance was measured using a Student’s slides for 30 min in sodium citrate buffer (10 mM sodium t-test. For analyses involving more than two groups, citrate, pH 6.0 with 0.05% Tween-20) and immunostaining GraphPad Prism 5.04 (GraphPad Software) was utilized was performed using the VectaStain Elite ABC HRP kit to perform a one-way ANOVA or two-way ANOVA (Vector Laboratories, Burlingame, CA) per the manufac- followed by the Fisher’s LSD post hoc test. turer’s instructions. All slides were counterstained with hematoxylin, washed in a series of alcohols, and dehy- Results drated in xylene. Glass coverslips were mounted using Generation and validation of Tmem106b knockout mice Cytoseal XYL (Thermo Scientific) and were left to set at Tmem106b knockout alleles were generated by insertion room temperature for 48 h before scanning with an of a lacZ gene trap in the intronic region between the Aperio ScanScope AT2 Slide Scanner (Leica Biosystems, first two coding exons, exons 3 and 4. The inserted gene Buffalo Grove, IL). ImageScope software (v12.1.0.5029; is transcribed by the endogenous Tmem106b promoter Leica) was used to annotate the cerebral cortex and along with the upstream exons and leads to a premature motor cortex of NeuN- and pTdp-43-stained slides. A termination of transcription (Fig. 1a). Inheritance of this custom-designed algorithm was applied to detect the targeted gene disruption was confirmed by PCR amplifi- number of pTdp-43-positive nuclei per area (mm ) cation of the genomic DNA isolated from Tmem106b [12] when labeled with the Cosmo antibody. The total +/+, +/−, and −/− mice (Fig. 1b). qPCR analysis in 3, 8 number of pTdp-43-positive cells were counted and 15 months old mice further confirmed the loss of manually for slides stained with the pTdp-43 antibody Tmem106b mRNA transcripts with mice heterozygous provided by Dr. Petrucelli. for the knockout allele showing approximately 50% loss as compared to age matched wild-type mice and a near RNA fluorescence in situ hybridization (FISH) complete loss of Tmem106b mRNA in Tmem106b −/− RNA FISH was performed in fixed mouse brain tissue as mice (Fig. 1c, Additional file 1: Figure S1). Loss of done previously [12, 28]. Briefly, paraffin embedded Tmem106b transcripts did not alter the expression of brain sections were deparaffinized in xylene and rehy- other Tmem106 family members, Tmem106a and drated in a series of ethanol solutions. Sections were Tmem106c (Additional file 1:FigureS2). Western permeabilized with ice-cold 2% acetone in PBS prepared blotting of wild-type mouse brain lysates with a in DEPC-treated water for 5 min. Sections were then Tmem106b antibody (Bethyl Laboratories) revealed a washed twice with DEPC-treated water and dehydrated robust Tmem106b-immunoreactive band at the pre- in a series of ethanol solutions before incubating 30 min dicted 43 kDa molecular weight. This band’sintensity at 66 °C in pre-hybridization buffer [50% formamide/2X was reduced approximately 50% in samples from SSC (MIDSCI, Valley Park, MO), 10% dextran sulfate Tmem106b +/− mice and was undetectable in Tmem106b (Millipore), 2× saline-sodium citrate buffer, 50 mM so- −/− brain tissue (Fig. 1d, e). Of note, using an in-house dium phosphate buffer, pH 7.0]. A fluorescently labeled developed Tmem106b antibody against the intracellular locked nucleic acid (LNA) probe (5’-TYE563/CCCC domain of TMEM106B (residues 1–96) [7], small molecu- GGCCCCGGCCCC-3′, Exiqon, Inc.; batch number lar weight bands could also be detected upon long expos- 612968) was diluted to 40 nM in in hybridization buffer ure in Tmem106b +/− and Tmem106b −/− mice. These (10% dextran sulfate, 50% formamide, 20 ng/μl BSA, additional bands likely correspond to Tmem106b-lacZ 25 mM tRNA, 25 nM EDTA, 2X SSC, 25 mM sodium truncated fragments which are expected in mouse phosphate buffer) and denatured at 80 °C for 5 min models created by the lacZ gene trap approach [9] before hybridizing to the tissue for 24 h at 66 °C in a (Additional file 1:FigureS3). dark, humid chamber. Sections were subsequently washed once with 2X SSC (0.1% Tween-20) at room Tmem106b loss does not reverse abnormal behaviors in temperature for 5 min and washed twice with mice expressing (GGGGCC) repeat pre-warmed 0.2X SSC at 55 °C for 10 min in the dark. To determine whether reduced Tmem106b levels might Slides were mounted with Vectashield mounting media be protective against FTD-like behavioral phenotypes in containing DAPI (Vector Laboratories). Cortical images the AAV-(GGGGCC) mouse model, the newborn of RNA foci were obtained using a Zeiss Axio Imager Z1 Tmem106b +/+, +/−, and −/− mice were injected with fluorescent microscope (Zeiss, Oberkochen, Germany) either AAV- 66R or AAV-2R (control) and aged to Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 6 of 14 Fig. 1 Generation of mice with targeted Tmem106b gene disruption. a Genomic structure of the wild-type mouse Tmem106b allele and the gene trap vector used to target the mouse Tmem106b gene. Rectangular boxes represent exons that are shaded in gray or white to denote coding and non-coding exons, respectively. White arrowheads signify Flp recombination target (FRT) sites and black arrowheads represent loci of crossover in P1 (LoxP) sites. Genotyping primer binding sites are labeled and denoted with black arrows above each genomic structure. b PCR analysis of DNA obtained from the hair follicles of Tmem106b wild-type (+/+), heterozygous (+/−), or knockout (−/−) mice using the primers depicted in panel a. c Quantitative PCR analysis measuring Tmem106b mRNA levels in Tmem106b +/+, +/−, and −/− mouse brain (n = 4 per genotype) at 3 months of age. The graph represents the mean ± S.E.M.; ****p < 0.0001 by one-way ANOVA followed by a Fisher’s LSD post-hoc test. d Western blot depicting Tmem106b protein levels (black arrowhead) in 3-month-old Tmem106b +/+, +/−, and −/− mouse brain tissue. Gapdh was used as a loading control. e Quantification of Tmem106b protein levels in Tmem106b +/+, +/-, and -/- mouse brain at 3 months of age (n =3 per genotype) 12 months of age. 66R virus injection did not affect the all 66R-injected mouse brains, which showed no sig- overall activity (data not shown) or body weight of the nificant difference in 66R mRNA expression among mice; except for a reduced body weight in female Tmem106b +/+, +/−,and −/− mice (Fig. 2f). Tmem106b +/+ mice injected with 66R as compared to 2R which could be contributed to the small number of (GGGGCC) repeat-induced neuropathology is not mice in this sub-group (Additional file 1: Figure S4). In rescued by lowering Tmem106b expression line with the original study describing this model [12], We next determined whether lowering Tmem106b injection of the 66R virus into wild-type mice induced expression was able to ameliorate key hallmarks of significant behavioral deficits, including anxiety and re- neurodegeneration, such as neuroinflammation and duced sociability as compared to 2R-injected wild-type neuronal loss, previously reported in the AAV-66R mice (p=0.019 by Student’s t-test for all wild-type 2R vs. mouse model [12]. At 12 months of age, wild-type mice 66R analyses; Fig. 2a-e). However, neither partial nor injected with 66R virus showed significantly increased complete reduction of Tmem106b altered the anxiety levels of Iba1 and Gfap mRNA transcripts as compared phenotype observed in 66R-injected wild-type mice as to 2R-injected animals, indicating pronounced neuroin- determined by the open field assay (Fig. 2a, b) or fear flammation (p = 0.024 by Student’s t-test for both ana- conditioning tests (Fig. 2c, d). Also, the reduced mouse lyses; Fig. 3a, b). However, neither partial nor complete sociability observed in 66R-injected mice could not be reduction of Tmem106b levels was able to rescue these rescued in mice with partial or complete loss of changes; in fact, 66R-injected Tmem106b −/− mice had Tmem106b (Fig. 2e). Since the amount of 66R viral significantly higher Gfap mRNA levels than 66R-injected expression could affect the presence and/or severity wild-type mice (Fig. 3b). Follow-up analysis in unin- of these phenotypes, we performed qPCR analysis on jected Tmem106b +/+, +/−, and −/− mice further Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 7 of 14 levels in Tmem106b −/− as compared to Tmem106 +/+ mice (Additional file 1: Figure S5). Gfap mRNA levels in Tmem106b +/− mice were not significantly different from Tmem106b +/+ mice. We next assessed the effect of Tmem106b reduction on 66R-induced neuronal loss. As expected, 66R-injected Tmem106b +/+ mice portrayed a significantly reduced number of cells immunoreactive for the neuronal marker, NeuN, as compared to 2R-injected mice (p =0.016 by Student’s t-test; Fig. 3c). Complete loss of Tmem106b did not modify this phenotype; however, partial reduction of Tmem106b significantly lessened the extent of neuronal loss associated with 66R injection. In fact, 66R-injected Tmem106b +/− cortical NeuN counts were not sig- nificantly different from that of 2R-injected animals (p = 0.0512 by Student’st-test; Fig. 3c, Additional file 1: Figure S6). This was not due to the presence of more NeuN-positive cells present in Tmem106b heterozygous mice given that uninjected Tmem106b +/+, +/−,and −/− mice have comparable NeuN counts (Additional file 1:FigureS5). We next studied two unique pathologies induced by overexpression of the 66R repeat: RNA foci and dipep- tide repeat proteins. Parallel to what was previously re- ported, RNA foci and poly-glycine/proline dipeptides [poly(GP)] were detected at significant levels in 66R-injected wild-type mice and not 2R-injected mice (Fig. 3d-f). We quantified the percentage of cells con- taining one or more RNA foci in the cortex of 66R-injected Tmem106b +/+, +/−, and −/− mice. As depicted in Fig. 3e, partial or complete reduction of Tmem106b levels did not significantly change the number of RNA foci-containing cells. Similarly, poly(GP) levels in 66R-injected Tmem106b +/− and −/− mice were similar to that of 66R-injected wild-type mice (Fig. 3f). Fig. 2 (GGGGCC) repeat-induced behavioral deficits are Finally, we examined the effect of Tmem106b unchanged by Tmem106b reduction. a Quantification of the average reduction on the formation of intracellular inclusions distance traveled obtained during the open field assay for 66R-injected of phosphorylated Tdp-43 (pTdp-43), a key patho- Tmem106b +/+, +/−, −/− mice at 12 months of age. Wild-type mice of logical feature observed both in patients and in mice the same age that had been injected with 2R C9ORF72 were used as a injected with 66R [12, 17, 45]. We detected intense control. b Quantification of the average speed traveled during the open field assay for mice as described in (a). c-d Quantification of the pTdp-43-positive inclusions in 66R- but not 2R-injected time each mouse spent freezing during the contextual (c) and cued Tmem106b +/+ brains using two different antibodies that (d) fear conditioning tests. e Quantification of the time each mouse recognize Tdp-43 in its phosphorylated form (Fig. 4a, spent exploring the mouse-containing cup during the social Additional file 1: Figure S7). However, neither partial interaction test (f qPCR quantification measuring the amount of 66R nor complete loss of Tmem106b significantly changed mRNA obtained from the brains of 12-month-old 66R-injected Tmem106b +/+, +/−,and −/− mice. Graphs represent the mean ± the number of pTdp-43-positive cells in the cortex or S.E.M. Data was analyzed by one-way ANOVA followed by Fisher’sLSD hippocampus (Fig. 4b; Additional file 1:FigureS7). post-hoc, and results are shown for 66R-injected Tmem106b +/− and −/− mice as compared to 66R-injected wild-type mice (n ≥ 12 for all C9ORF72 loss increases TMEM106B levels unlike (GGGG groups). NS, not significant CC) overexpression To further study the connection between TMEM106B showed that full loss of Tmem106b (in the absence of and C9ORF72-related disease mechanisms, we deter- repeat overexpression), is sufficient to induce astrogliosis mined Tmem106b protein levels in wild-type mice that as determined by a significant increase in Gfap mRNA had been injected with 66R or 2R virus. Tmem106b Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 8 of 14 Fig. 3 Assessment of (GGGGCC) -mediated neuropathology in response to changes in Tmem106b levels (a-b) Quantification of Iba1 (a)or Gfap (b) mRNA levels in the cortex of 2R-injected mice, or in 66R-injected Tmem106b +/+, +/−,or −/− mice. c Quantitative analysis of the number of cells immunoreactive for NeuN in the cortex of 2R-injected mice, or in 66R-injected Tmem106b +/+, +/−,or −/− mice. d Image depicting the presence of RNA foci (arrowheads) in the nuclei of cortical cells in wild-type 66R- versus 2R-injected mice. e Quantification of the number of cells with RNA foci in the cortex of wild-type 2R-injected mice, or in 66R-injected Tmem106b +/+, +/−,or −/− mice. f Quantitative assessment of the presence of poly(GP) peptides detected in the brains of mice injected as described in panel e. Graphs represent the mean ± S.E.M. Data was analyzed by one-way ANOVA followed by Fisher’s LSD post-hoc, and results are shown for 66R-injected Tmem106b +/− and −/− mice as compared to 66R-injected wild-type mice (n = 12 per group). NS, not significant.; *p < 0.05; **p < 0.001 Fig. 4 Tmem106b levels do not affect (GGGGCC) repeat induced pTdp-43 inclusion body formation. a Representative images of pTdp-43 (pS409/410 from Cosmo Bio) staining of motor cortex region of the mouse brains from indicated Tmem106b genotypes 12 months after 2R and 66R AAV injection. Arrow heads indicate pTdp-43-positive inclusion bodies. b and c Quantification of pTdp-43 inclusion body positive cells in different brain regions: cortex (b) or hippocampus (c) from Tmem106b +/+, +/−, and −/− mice as compared to WT 2R injected mice (n = 4). Graphs represent the mean ± S.E.M. Data was analyzed by one-way ANOVA followed by Fisher’s LSD post-hoc, and results are shown for 66R-injected Tmem106b +/− and −/− mice as compared to 66R-injected wild-type mice (n = 12 per group). NS, not significant Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 9 of 14 levels in 66R injected brains were almost identical to was chosen as the model system in this study due to the that of 2R control brains (Fig. 5a, b). The 66R-injected demonstration of various phenotypes associated with mice also had normal levels of other lysosomal proteins, C9ORF72 repeat expansions as early as 6 months of age such as Lamp1, cathepsin D (both pro- and mature [12]. At the time this study commenced, only two other forms), and progranulin (Fig. 5a, quantifications not characterized C9ORF72-repeat expansion mouse models shown). Furthermore, whereas human patients with had been generated. Both of these models were created C9ORF72-related FTD have reduced C9ORF72 levels in using a bacterial artificial chromosome (BAC) for the addition to the GGGGCC-repeat associated toxicities, expression of the complete [39] or partial [41] C9ORF72 endogenous C9ORF72 expression levels were not chan- coding region with incorporated (GGGGCC) repeats of ged in our 66R overexpression model (Additional file 1: various lengths. These models successfully demonstrated Figure S8). This prompted us to compare the effects of pathologies directly related to the repeat expansion as 66R overexpression and C9ORF72 loss on Tmem106b early as 4–6 months of age, including RNA foci and the levels in cell culture (techniques validated in Additional generation of dipeptide repeat proteins. However, in file 1: Figure S9). Consistent with the in vivo data, over- contrast to the (GGGGCC) -AAV model, many of the expression of 66R in both HeLa and U251 cells failed to key features of C9ORF72-mediated FTD, especially change the expression of various lysosomal proteins, behavioral deficits, neuroinflammation, neuronal loss, including TMEM106B (Fig. 6a-e and Additional file 1: and pTdp-43 pathology, were not observed even in aged Figure S10). However, knockdown of C9ORF72 signifi- mice in these models [39, 41]. Since TDP-43 pathology cantly increased protein levels of TMEM106B together is a common denominator of the human disease popula- with other lysosomal resident proteins (Fig. 6f-j and tions associated with TMEM106B (FTLD-TDP, GRN-car- Additional file 1: Figure S10). riers, C9ORF72-carriers and AD patients with TDP-43 pathology [38]) the selection of a mouse model with Discussion pTdp-43 pathology was considered essential. Importantly Genetic variants in TMEM106B associated with reduced however, neither the BAC mice nor our (GGGG TMEM106B expression have been shown to significantly CC) -AAV mice recapitulate the loss of C9ORF72 expres- protect individuals with either GRN mutations or sion consistently observed in human C9ORF72 expansion C9ORF72 repeat expansions from the development of carriers. FTD symptoms [13, 19, 21, 30, 37, 58–60]. Moreover, Neuronal loss and neuroinflammation are common depletion of Tmem106b was recently shown to rescue features among neurodegenerative disorders. Protective several disease-relevant phenotypes observed in Grn −/− TMEM106B variants were found to associate with mice [26]. This study is the first to determine the impact increased neuronal gene expression, reduced expression of Tmem106b reduction on C9ORF72-related disease. of genes involved in inflammation, and better cognitive Contrary to the relative success in the Grn −/− mouse performanceinhealthyagedindividuals [46]. However, we model, we demonstrate that neither partial nor complete did not see a significant improvement in measures of loss of Tmem106b is sufficient to rescue the behavioral neuroinflammation or behavioral deficits induced by changes or neuropathological phenotypes that manifest the overexpression of the (GGGGCC) repeat when in an AAV-based mouse model of C9ORF72-associated Tmem106b levels were reduced. In fact, (GGGGCC) -in- (GGGGCC) expansions. The (GGGGCC) -AAV model jected Tmem106b knockout mice showed more severe n 66 Fig. 5 (GGGGCC) repeat expansion overexpression does not alter Tmem106b protein levels in mouse brain. (a) Western blot of brain tissue obtained from 2R- or 66R-injected wild-type mice using antibodies against various lysosomal proteins. Gapdh was used as a loading control. (b) Quantification of Tmem106b protein levels by Western blot as depicted in panel a. The graph represents the mean ± S.E.M. by Student’s t- test (n = 8 for 2R; n = 12 for 66R ); NS, not significant Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 10 of 14 Fig. 6 The effect of (GGGGCC) overexpression or C9ORF72 knockdown on TMEM106B protein levels in HeLa cells. a Western blot of HeLa cells transfected with either 2R or 66R pAAV. b-e Protein quantification of TMEM106B (b), LAMP1 (c), CTSD (pro-form, mature form has similar results as the pro-form) (d), and PGRN (e) in cells transfected as described in panel (a). f Western blot of HeLa cells transfected with either control siRNA or siRNA against C9ORF72 (g-j), Protein quantification of TMEM106B (g), LAMP1 (h), CTSD (i), and PGRN (j) in cells transfected as described in panel f. GAPDH was used as a loading control. Graphs represent the mean ± S.E.M. by Student’st-test (n = 6 for all groups). NS, not significant; *p <0.05, **p <0.01 astrogliosis, evidenced by increased Gfap expression, as that did not find an association between TMEM106B compared to Tmem106b wild-type or heterozygous mice variants and dipeptide repeat pathology in C9ORF72 injected with the expanded repeat. We showed that mutation carriers [16]. Given the strong genetic associ- this was due to increased astrogliosis from Tmem106b ation of both GRN and C9ORF72 carriers with loss alone, excluding the possibility that loss of TMEM106B variants, the lack of association in our Tmem106b renders mice more sensitive to a C9ORF72 Tmem106b model with pathological features that are repeat-mediated inflammatory response. Our observa- unique to C9ORF72GGGGCC repeat expansion carriers tion of astrogliosis resulting from loss of Tmem106b may not be surprising. Nevertheless, both C9ORF72 and alone suggests that Tmem106b plays a unique and po- GRN mutation carriers present with TDP-43 pathology tentially necessary role in astrocytes. Tmem106b is a at autopsy. Indeed, TMEM106B variants were first lysosomal resident protein, and loss of Tmem106b has discovered as disease modifiers in an FTD cohort been recently shown to cause lysosomal dysfunctions comprised of individuals with TDP-43 brain pathology including lysosomal acidification and trafficking prob- regardless of underlying cause [59]. TMEM106B variants lems [26, 49]. Importantly, dysfunction of lysosomes were additionally found to associate with the presence of in multiple diseases, for instance, lysosomal storage TDP-43 pathology in other diseases, such as Alzheimer’s disorders has been tightly linked to astroglial activa- disease and hippocampal sclerosis [3, 35, 48]. In fact, the tion [42]. Interestingly, activation of the Tmem106b protective TMEM106B variants appeared to corres- paralog, Tmem106a, was shown to be immunostimula- pond with lessened TDP-43 aggregate burden in a tory in mouse macrophages [15]. As such, our data preliminary study of eight C9ORF72 mutation carriers suggests that TMEM106B might play a novel, recipro- [58]. Collectively, these findings suggest that TMEM106B cal role in inflammatory modulation. In relation to the protective variants may reduce one’s risk of developing neuronal loss, we did observe that partial, but not TDP-43 proteinopathies; however, despite careful analysis complete loss of Tmem106b significantly lessened the with two independent antibodies, we found that reducing extent of neuronal loss in the AAV-(GGGGCC) Tmem106b levels in (GGGGCC) -injected mice did not injected mice, suggesting that, if pursued, partial lessen the development of phosphorylated pTdp-43 aggre- TMEM106B reduction may be a more viable avenue gates in the cortex or hippocampus at 12 months of age. for future TMEM106B-related therapeutic approaches The inability to ameliorate neuropathological and in FTD. behavioral deficits through the reduction of Tmem106b Tmem106b levels also had no effect on two other key in the AAV-(GGGGCC) model requires a careful pathological features observed in the (GGGGCC) -AAV examination of the approaches employed in this study. model: RNA foci and the generation of dipeptide repeat First, we modelled the protective TMEM106B human proteins. These data are in line with a previous report haplotype by reducing Tmem106b by approximately 50 Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 11 of 14 or 100%. This seems appropriate since multiple studies lysosomal function [2, 51, 55, 62]. Specifically, loss of have observed lower TMEM106B mRNA levels or a fas- C9ORF72, as seen in human patients carrying C9ORF72 ter TMEM106B protein degradation associated with the repeat expansions, causes impaired mTORC1 signaling TMEM106B protective haplotype [20, 37, 59]. We can- and abnormal lysosome morphology indicative of dysfunc- not, however, exclude that the associated variants affect tion [2, 57]. Thus, it is conceivable that the effect of the TMEM106B in an unknown fashion, for example as a protective TMEM106B haplotype in C9ORF72 expansion result of the p.Thr185Ser variant (rs3173615) which al- carriers is to counteract lysosomal dysfunction that results ters the protein coding sequence of TMEM106B and is from the loss of C9ORF72 expression. In support of this inherited as part of the protective haplotype [59]. It is hypothesis, we showed that reducing C9ORF72 levels in also possible that the protective TMEM106B haplotype human cell lines significantly increased levels of various induces a more subtle decrease in TMEM106B than lysosomal proteins, including TMEM106B, which is what we modeled in this study or that the N-terminal thought to signify lysosomal dysfunction. These types of TMEM106B fragments, which we detected at low lysosomal changes were not observed by overexpression levels in knockout animals, may have retained some of (GGGGCC) in cell culture, nor were these changes partial TMEM106B function. Second, we modelled observed in our AAV-(GGGGCC) mouse model. the C9ORF72-associated repeat expansion through the overexpression of (GGGGCC) by AAV. This ap- Conclusions proach importantly results in pTdp-43-positive inclu- In summary, we show that reducing the levels of sions as early as 6 months of age [12]; however, this Tmem106b in a mouse model mimicking the toxic model may have been too aggressive to reverse the gain-of-functions associated with the C9ORF72 (GGGG neurodegenerative, neuropathological, and behavioral CC) repeat expansions is unable to ameliorate key phenotypes in these mice at 12 months of age. More pathological features seen in FTD patients, including importantly, our approach only recapitulated the RNA pTdp-43 pathology. We further provide support for the and protein toxic gain-of-function mechanisms associated growing body of evidence linking the loss of C9ORF72 with the repeat sequence and failed to model the reduction expression to the pathobiology of C9ORF72, in this case in C9ORF72 transcripts which is now considered and inte- through the induction of lysosomal dysfunction. As gral part of C9ORF72 disease pathogenesis [1, 17, 45, 56]. such, it will be critically important that further examin- Indeed, endogenous C9orf72 levels were unchanged in our ation of the effects of Tmem106b reduction on 66R mice at 12 months of age. C9ORF72 pathobiology be studied in models that in- TMEM106B is a type II lysosomal membrane protein clude reduction of C9ORF72. with currently unknown function. Increases in TMEM106B levels have been found to be cytotoxic and are associated Additional file with increases in lysosomal size and reduced lysosomal Additional file 1: Figures S1 through S10. Figure S1. Transcript acidification, leading to the disruption of endolysosomal- expression of Tmem106b in Tmem106b deficiency mice at different ages. and autophagic-lysosomal degradation [38]. Recent Figure S2. Tmem106b reduction does not alter the expression of its work undeniably links PGRN to lysosomal biology family members. Figure S3. Tmem106b immunoreactivity in mice with Tmem106b gene interruption using an additional antibody. Figure S4. [24, 31, 64, 65] and, as such, it may not have been The body weight of 2R and 66R injected mouse. Figure S5. Tmem106b surprising that Tmem106b loss reversed some of the reduction alone induces astrogliosis. Figure S6. Heterozygous loss of Grn knockout-mediated lysosomal enzyme dysregula- Tmem106b partially rescues 66R injection-induced neuronal loss. Figure S7. pTdp-43 immunoreactivity in 2R and 66R injected mouse brain. tion [26]. This prompts the question of whether Figure S8. Endogenous C9orf72 protein levels in 2R- and 66R-injected Tmem106b reduction might only confer protection in mouse brain. Figure S9. Validation of (GGGGCC) repeat overexpression disease models that portray abnormal lysosomal biol- and C9ORF72 knockdown. Figure S10. The effect of (GGGGCC) overexpression or C9ORF72 knockdown on TMEM106B protein levels in ogy. While lysosomal dysfunction has been implicated U251 cells. (DOCX 26231 kb) in C9ORF72-related pathogenesis, much of the evi- dence to support this comes from studies investigating Acknowledgements the function of the C9ORF72 protein itself and not the This work is funded in part by a Mayo Clinic Edward C. Kendall Research (GGGGCC) repeat expansions [2, 8, 27, 40, 52, 55]. The n Fellowship (AN) and a research fellowship from The Bluefield Project to Cure FTD (XZ). Additional funding was obtained by NIH/NINDS grants C9ORF72 protein sequence contains DENN-like domains, R35NS097261 (RR), UG3NS0103870 (RR) and P01NS084974 (LP, RR, DWD) and making it a part of the DENN protein superfamily which The Bluefield Project to Cure FTD (RR). is known to be involved in regulating membrane traffick- ing and autophagy [32, 63]. Genetic and cell biology Authors’ contributions RR, AMN and XZ designed the study, oversaw the experiments and drafted studies have shown that C9ORF72 interacts with the manuscript. JC and AMN injected the AAV virus into Tmem106b mice. other DENN domain-containing proteins linked to RBP, TP, MB, NF, and BM harvested mouse brains and assisted in mTORC1 signaling, whose activity is closely tied to biochemical, gene expression and genetic analysis of mouse tissues. NF and Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 12 of 14 TP performed and analyzed the cell culture studies. AK, MB and JF 12. Chew J, Gendron TF, Prudencio M, Sasaguri H, Zhang YJ, Castanedes-Casey performed the mouse behavioral analyses. MC-C, LR, VP, MEM and DD M, Lee CW, Jansen-West K, Kurti A, Murray ME et al (2015) performed the immunohistochemical analyses of mouse brains. MD-H Neurodegeneration. C9ORF72 repeat expansions in mice cause TDP-43 performed FISH analysis to quantify RNA foci burden and LD and TFG pathology, neuronal loss, and behavioral deficits. Science 348:1151–1154. performed quantitative analysis of GP levels. LP, KRJ-W and EP generated the https://doi.org/10.1126/science.aaa9344 AAV virus and FH provided Tmem106b antibody. All authors read and a 13. Cruchaga C, Graff C, Chiang HH, Wang J, Hinrichs AL, Spiegel N, Bertelsen S, pproved the final manuscript. Mayo K, Norton JB, Morris JC et al (2011) Association of TMEM106B gene polymorphism with age at onset in Granulin mutation carriers and plasma Ethics approval Granulin protein levels. Arch Neurol. https://doi.org/10.1001/archneurol.2010.350 All animal studies were approved by the Mayo Clinic Institutional Animal 14. Cruts M, Gijselinck I, van der Zee J, Engelborghs S, Wils H, Pirici D, Rademakers Care and Use Committee. R, Vandenberghe R, Dermaut B, Martin JJ et al (2006) Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to Competing interests chromosome 17q21. Nature 442:920–924. https://doi.org/10.1038/nature05017 The authors declare that they have no competing interests. 15. Dai H, Xu D, Su J, Jang J, Chen Y (2015) Transmembrane protein 106a activates mouse peritoneal macrophages via the MAPK and NF-kappaB signaling pathways. Sci Rep 5:12461. https://doi.org/10.1038/srep12461 Publisher’sNote 16. Davidson YS, Barker H, Robinson AC, Thompson JC, Harris J, Troakes C, Springer Nature remains neutral with regard to jurisdictional claims in Smith B, Al-Saraj S, Shaw C, Rollinson S et al (2014) Brain distribution of published maps and institutional affiliations. dipeptide repeat proteins in frontotemporal lobar degeneration and motor neurone disease associated with expansions in C9ORF72. Acta Neuropathol Author details Commun 2:70. https://doi.org/10.1186/2051-5960-2-70 Department of Neuroscience, Mayo Clinic Jacksonville, 4500 San Pablo Road, 17. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, Jacksonville, FL 32224, USA. Department of Molecular Biology and Genetics, Rutherford NJ, Nicholson AM, Finch NA, Flynn H, Adamson J et al (2011) Weill Institute for Cell and Molecular Biology, Cornell University, 345 Weill Expanded GGGGCC hexanucleotide repeat in noncoding region of Hall, Ithaca, NY 14853, USA. C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron 72:245–256. https://doi.org/10.1016/j.neuron.2011.09.011 Received: 10 May 2018 Accepted: 11 May 2018 18. Deming Y, Cruchaga C (2014) TMEM106B: a strong FTLD disease modifier. Acta Neuropathol 127:419–422. https://doi.org/10.1007/s00401-014-1249-3 19. Finch N, Carrasquillo MM, Baker M, Rutherford NJ, Coppola G, Dejesus- References Hernandez M, Crook R, Hunter T, Ghidoni R, Benussi L et al (2011) 1. Amick J, Ferguson SM (2017) C9orf72: at the intersection of lysosome cell TMEM106B regulates progranulin levels and the penetrance of FTLD in GRN biology and neurodegenerative disease. Traffic 18:267–276. https://doi.org/ mutation carriers. Neurology 76:467–474. https://doi.org/10.1212/WNL. 10.1111/tra.12477 0b013e31820a0e3b 2. Amick J, Roczniak-Ferguson A, Ferguson SM (2016) C9orf72 binds SMCR8, 20. Gallagher MD, Posavi M, Huang P, Unger TL, Berlyand Y, Gruenewald AL, localizes to lysosomes, and regulates mTORC1 signaling. Mol Biol Cell 27: Chesi A, Manduchi E, Wells AD, Grant SF et al (2017a) A dementia- 3040–3051. https://doi.org/10.1091/mbc.E16-01-0003 associated risk variant near TMEM106B alters chromatin architecture and 3. Aoki N, Murray ME, Ogaki K, Fujioka S, Rutherford NJ, Rademakers R, Ross gene expression. Am J Hum Genet 101:643–663. https://doi.org/10.1016/j. OA, Dickson DW (2015) Hippocampal sclerosis in Lewy body disease is a ajhg.2017.09.004 TDP-43 proteinopathy similar to FTLD-TDP type a. Acta Neuropathol 129: 21. Gallagher MD, Suh E, Grossman M, Elman L, McCluskey L, Van Swieten JC, 53–64. https://doi.org/10.1007/s00401-014-1358-z Al-Sarraj S, Neumann M, Gelpi E, Ghetti B et al (2014) TMEM106B is a 4. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, Mori H, Mann D, genetic modifier of frontotemporal lobar degeneration with C9orf72 Tsuchiya K, Yoshida M, Hashizume Y et al (2006) TDP-43 is a component of hexanucleotide repeat expansions. Acta Neuropathol 127:407–418. https:// ubiquitin-positive tau-negative inclusions in frontotemporal lobar doi.org/10.1007/s00401-013-1239-x degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res 22. Gendron TF, Bieniek KF, Zhang YJ, Jansen-West K, Ash PE, Caulfield T, Commun 351:602–611. https://doi.org/10.1016/j.bbrc.2006.10.093 Daughrity L, Dunmore JH, Castanedes-Casey M, Chew J et al (2013) Antisense 5. Ash PE, Bieniek KF, Gendron TF, Caulfield T, Lin WL, Dejesus-Hernandez M, transcripts of the expanded C9ORF72 hexanucleotide repeat form nuclear RNA van Blitterswijk MM, Jansen-West K, Paul JW 3rd, Rademakers R et al (2013) foci and undergo repeat-associated non-ATG translation in c9FTD/ALS. Acta Unconventional translation of C9ORF72 GGGGCC expansion generates Neuropathol 126:829–844. https://doi.org/10.1007/s00401-013-1192-8 insoluble polypeptides specific to c9FTD/ALS. Neuron 77:639–646. https:// 23. Goldman JS, Farmer JM, Wood EM, Johnson JK, Boxer A, Neuhaus J, Lomen- doi.org/10.1016/j.neuron.2013.02.004 Hoerth C, Wilhelmsen KC, Lee VM, Grossman M et al (2005) Comparison of 6. Baker M, Mackenzie IR, Pickering-Brown SM, Gass J, Rademakers R, Lindholm family histories in FTLD subtypes and related tauopathies. Neurology 65: C, Snowden J, Adamson J, Sadovnick AD, Rollinson S et al (2006) Mutations 1817–1819. https://doi.org/10.1212/01.wnl.0000187068.92184.63 in progranulin cause tau-negative frontotemporal dementia linked to 24. Holler CJ, Taylor G, Deng Q, Kukar T (2017) Intracellular proteolysis of chromosome 17. Nature 442:916–919. https://doi.org/10.1038/nature05016 Progranulin generates stable, lysosomal Granulins that are Haploinsufficient 7. Brady OA, Zheng Y, Murphy K, Huang M, Hu F (2012) The frontotemporal in patients with frontotemporal dementia caused by GRN mutations. lobar degeneration risk factor, TMEM106B, regulates lysosomal morphology eNeuro 4. https://doi.org/10.1523/ENEURO.0100-17.2017 and function. Hum Mol Genet. https://doi.org/10.1093/hmg/dds475 25. Kim JY, Grunke SD, Levites Y, Golde TE, Jankowsky JL (2014) 8. Burberry A, Suzuki N, Wang JY, Moccia R, Mordes DA, Stewart MH, Suzuki- Intracerebroventricular viral injection of the neonatal mouse brain for Uematsu S, Ghosh S, Singh A, Merkle FT et al (2016) Loss-of-function persistent and widespread neuronal transduction. J Vis Exp 51863. https:// mutations in the C9ORF72 mouse ortholog cause fatal autoimmune disease. doi.org/10.3791/51863 Sci Transl Med 8: 347ra393 Doi https://doi.org/10.1126/scitranslmed.aaf6038 26. Klein ZA, Takahashi H, Ma M, Stagi M, Zhou M, Lam TT, Strittmatter SM 9. Cecconi F, Meyer BI (2000) Gene trap: a way to identify novel genes and (2017) Loss of TMEM106B ameliorates lysosomal and frontotemporal unravel their biological function. FEBS Lett 480:63–71 dementia-related phenotypes in Progranulin-deficient mice. Neuron 95(281– 10. Chakrabarty P, Rosario A, Cruz P, Siemienski Z, Ceballos-Diaz C, Crosby K, 296):e286. https://doi.org/10.1016/j.neuron.2017.06.026 Jansen K, Borchelt DR, Kim JY, Jankowsky JL et al (2013) Capsid serotype and timing of injection determines AAV transduction in the neonatal mice 27. Koppers M, Blokhuis AM, Westeneng HJ, Terpstra ML, Zundel CA, Vieira de brain. PLoS One 8:e67680. https://doi.org/10.1371/journal.pone.0067680 Sa R, Schellevis RD, Waite AJ, Blake DJ, Veldink JH et al (2015) C9orf72 11. Chen-Plotkin AS, Unger TL, Gallagher MD, Bill E, Kwong LK, Volpicelli-Daley ablation in mice does not cause motor neuron degeneration or motor L, Busch JI, Akle S, Grossman M, Van Deerlin V et al (2012) TMEM106B, the deficits. Ann Neurol 78:426–438. https://doi.org/10.1002/ana.24453 risk gene for frontotemporal dementia, is regulated by the microRNA-132/ 28. Lagier-Tourenne C, Baughn M, Rigo F, Sun S, Liu P, Li HR, Jiang J, Watt AT, 212 cluster and affects progranulin pathways. J Neurosci 32:11213–11227. Chun S, Katz M et al (2013) Targeted degradation of sense and antisense https://doi.org/10.1523/JNEUROSCI.0521-12.2012 C9orf72 RNA foci as therapy for ALS and frontotemporal degeneration. Proc Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 13 of 14 Natl Acad Sci U S A 110:E4530–E4539. https://doi.org/10.1073/pnas. 46. Rhinn H, Abeliovich A (2017) Differential aging analysis in human cerebral cortex 1318835110 identifies variants in TMEM106B and GRN that regulate aging phenotypes. Cell 29. Lang CM, Fellerer K, Schwenk BM, Kuhn PH, Kremmer E, Edbauer D, Capell Syst 4(404–415):e405. https://doi.org/10.1016/j.cels.2017.02.009 A, Haass C (2012) Membrane orientation and subcellular localization of 47. Rosso SM, Donker Kaat L, Baks T, Joosse M, de Koning I, Pijnenburg Y, de Jong transmembrane protein 106B (TMEM106B), a major risk factor for D, Dooijes D, Kamphorst W, Ravid R et al (2003) Frontotemporal dementia in the frontotemporal lobar degeneration. J Biol Chem 287:19355–19365. https:// Netherlands: patient characteristics and prevalence estimates from a population- doi.org/10.1074/jbc.M112.365098 based study. Brain 126:2016–2022. https://doi.org/10.1093/brain/awg204 30. Lattante S, Le Ber I, Galimberti D, Serpente M, Rivaud-Pechoux S, Camuzat 48. Rutherford NJ, Carrasquillo MM, Li M, Bisceglio G, Menke J, Josephs KA, A, Clot F, Fenoglio C, Scarpini E, Brice A et al (2014) Defining the association Parisi JE, Petersen RC, Graff-Radford NR, Younkin SG et al (2012) of TMEM106B variants among frontotemporal lobar degeneration patients TMEM106B risk variant is implicated in the pathologic presentation of with GRN mutations and C9orf72 repeat expansions. Neurobiol Aging Alzheimer disease. Neurology 79:717–718. https://doi.org/10.1212/WNL. 35(2658):e2651–e2655. https://doi.org/10.1016/j.neurobiolaging.2014.06.023 0b013e318264e3ac 31. Lee CW, Stankowski JN, Chew J, Cook CN, Lam YW, Almeida S, Carlomagno 49. Schwenk BM, Lang CM, Hogl S, Tahirovic S, Orozco D, Rentzsch K, Y, Lau KF, Prudencio M, Gao FB et al (2017) The lysosomal protein cathepsin Lichtenthaler SF, Hoogenraad CC, Capell A, Haass C et al (2014a) The FTLD L is a progranulin protease. Mol Neurodegener 12:55. https://doi.org/10. risk factor TMEM106B and MAP6 control dendritic trafficking of lysosomes. 1186/s13024-017-0196-6 EMBO J 33:450–467. https://doi.org/10.1002/embj.201385857 32. Levine TP, Daniels RD, Gatta AT, Wong LH, Hayes MJ (2013) The product of 50. Schwenk BM, Lang CM, Hogl S, Tahirovic S, Orozco D, Rentzsch K, C9orf72, a gene strongly implicated in neurodegeneration, is structurally Lichtenthaler SF, Hoogenraad CC, Capell A, Haass C et al (2014b) The FTLD related to DENN Rab-GEFs. Bioinformatics 29:499–503. https://doi.org/10. risk factor TMEM106B and MAP6 control dendritic trafficking of lysosomes. 1093/bioinformatics/bts725 Embo J. https://doi.org/10.1002/embj.201385857 33. Mori K, Arzberger T, Grasser FA, Gijselinck I, May S, Rentzsch K, Weng SM, 51. Sellier C, Campanari ML, Julie Corbier C, Gaucherot A, Kolb-Cheynel I, Oulad- Schludi MH, van der Zee J, Cruts M et al (2013a) Bidirectional transcripts of Abdelghani M, Ruffenach F, Page A, Ciura S, Kabashi E et al (2016) Loss of the expanded C9orf72 hexanucleotide repeat are translated into C9ORF72 impairs autophagy and synergizes with polyQ Ataxin-2 to induce aggregating dipeptide repeat proteins. Acta Neuropathol 126:881–893. motor neuron dysfunction and cell death. EMBO J 35:1276–1297. https:// https://doi.org/10.1007/s00401-013-1189-3 doi.org/10.15252/embj.201593350 34. Mori K, Weng SM, Arzberger T, May S, Rentzsch K, Kremmer E, Schmid B, 52. Shi Y, Lin S, Staats KA, Li Y, Chang WH, Hung ST, Hendricks E, Linares GR, Kretzschmar HA, Cruts M, Van Broeckhoven C et al (2013b) The C9orf72 Wang Y, Son EY et al (2018) Haploinsufficiency leads to neurodegeneration GGGGCC repeat is translated into aggregating dipeptide-repeat proteins in in C9ORF72 ALS/FTD human induced motor neurons. Nat Med 24:313–325. FTLD/ALS. Science 339:1335–1338. https://doi.org/10.1126/science.1232927 https://doi.org/10.1038/nm.4490 35. Murray ME, Cannon A, Graff-Radford NR, Liesinger AM, Rutherford NJ, Ross 53. Stagi M, Klein ZA, Gould TJ, Bewersdorf J, Strittmatter SM (2014) Lysosome OA, Duara R, Carrasquillo MM, Rademakers R, Dickson DW (2014) Differential size, motility and stress response regulated by fronto-temporal dementia clinicopathologic and genetic features of late-onset amnestic dementias. modifier TMEM106B. Mol Cell Neurosci 61:226–240. https://doi.org/10.1016/j. Acta Neuropathol 128:411–421. https://doi.org/10.1007/s00401-014-1302-2 mcn.2014.07.006 36. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT, 54. Su Z, Zhang Y, Gendron TF, Bauer PO, Chew J, Yang WY, Fostvedt E, Jansen- Bruce J, Schuck T, Grossman M, Clark CM et al (2006) Ubiquitinated TDP-43 West K, Belzil VV, Desaro P et al (2014) Discovery of a biomarker and lead in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. small molecules to target r(GGGGCC)-associated defects in c9FTD/ALS. Science 314:130–133 Neuron 83:1043–1050. https://doi.org/10.1016/j.neuron.2014.07.041 37. Nicholson AM, Finch NA, Wojtas A, Baker MC, Perkerson RB 3rd, 55. Sullivan PM, Zhou X, Robins AM, Paushter DH, Kim D, Smolka MB, Hu F Castanedes-Casey M, Rousseau L, Benussi L, Binetti G, Ghidoni R et al (2016) The ALS/FTLD associated protein C9orf72 associates with SMCR8 and (2013) TMEM106B p.T185S regulates TMEM106B protein levels: WDR41 to regulate the autophagy-lysosome pathway. Acta Neuropathol implications for frontotemporal dementia. J Neurochem 126:781–791. Commun 4:51. https://doi.org/10.1186/s40478-016-0324-5 https://doi.org/10.1111/jnc.12329 56. Taylor JP, Brown RH Jr, Cleveland DW (2016) Decoding ALS: from genes to 38. Nicholson AM, Rademakers R (2016) What we know about TMEM106B in mechanism. Nature 539:197–206. https://doi.org/10.1038/nature20413 neurodegeneration. Acta Neuropathol 132:639–651. https://doi.org/10.1007/ 57. Ugolino J, Ji YJ, Conchina K, Chu J, Nirujogi RS, Pandey A, Brady NR, s00401-016-1610-9 Hamacher-Brady A, Wang J (2016) Loss of C9orf72 enhances Autophagic 39. O'Rourke JG, Bogdanik L, Muhammad A, Gendron TF, Kim KJ, Austin A, Cady activity via deregulated mTOR and TFEB signaling. PLoS Genet 12:e1006443. J, Liu EY, Zarrow J, Grant S et al (2015) C9orf72 BAC transgenic mice display https://doi.org/10.1371/journal.pgen.1006443 typical pathologic features of ALS/FTD. Neuron 88:892–901. https://doi.org/ 58. van Blitterswijk M, Mullen B, Nicholson AM, Bieniek KF, Heckman MG, Baker 10.1016/j.neuron.2015.10.027 MC, Dejesus-Hernandez M, Finch NA, Brown PH, Murray ME et al (2014) 40. O'Rourke JG, Bogdanik L, Yanez A, Lall D, Wolf AJ, Muhammad AK, Ho R, TMEM106B protects C9ORF72 expansion carriers against frontotemporal Carmona S, Vit JP, Zarrow J et al (2016) C9orf72 is required for proper dementia. Acta Neuropathol. https://doi.org/10.1007/s00401-013-1240-4 macrophage and microglial function in mice. Science 351:1324–1329. 59. Van Deerlin VM, Sleiman PM, Martinez-Lage M, Chen-Plotkin A, Wang LS, https://doi.org/10.1126/science.aaf1064 Graff-Radford NR, Dickson DW, Rademakers R, Boeve BF, Grossman M et al 41. Peters OM, Cabrera GT, Tran H, Gendron TF, McKeon JE, Metterville J, Weiss (2010) Common variants at 7p21 are associated with frontotemporal lobar A, Wightman N, Salameh J, Kim J et al (2015) Human C9ORF72 degeneration with TDP-43 inclusions. Nat Genet 42:234–239. https://doi.org/ Hexanucleotide expansion reproduces RNA foci and dipeptide repeat 10.1038/ng.536 proteins but not neurodegeneration in BAC transgenic mice. Neuron 88: 60. van der Zee J, Van Langenhove T, Kleinberger G, Sleegers K, Engelborghs S, 902–909. https://doi.org/10.1016/j.neuron.2015.11.018 Vandenberghe R, Santens P, Van den Broeck M, Joris G, Brys J et al (2011) 42. Platt FM, Boland B, van der Spoel AC (2012) The cell biology of disease: TMEM106B is associated with frontotemporal lobar degeneration in a lysosomal storage disorders: the cellular impact of lysosomal dysfunction. J clinically diagnosed patient cohort. Brain 134:808–815. https://doi.org/10. Cell Biol 199:723–734. https://doi.org/10.1083/jcb.201208152 1093/brain/awr007 43. Rascovsky K, Hodges JR, Knopman D, Mendez MF, Kramer JH, Neuhaus J, 61. Wood EM, Falcone D, Suh E, Irwin DJ, Chen-Plotkin AS, Lee EB, Xie SX, Van van Swieten JC, Seelaar H, Dopper EG, Onyike CU et al (2011) Sensitivity of Deerlin VM, Grossman M (2013) Development and validation of pedigree revised diagnostic criteria for the behavioural variant of frontotemporal classification criteria for frontotemporal lobar degeneration. JAMA Neurol dementia. Brain 134:2456–2477. https://doi.org/10.1093/brain/awr179 70:1411–1417. https://doi.org/10.1001/jamaneurol.2013.3956 44. Ratnavalli E, Brayne C, Dawson K, Hodges JR (2002) The prevalence of 62. Yang M, Liang C, Swaminathan K, Herrlinger S, Lai F, Shiekhattar R, frontotemporal dementia. Neurology 58:1615–1621 Chen JF (2016) A C9ORF72/SMCR8-containing complex regulates ULK1 45. Renton AE, Majounie E, Waite A, Simon-Sanchez J, Rollinson S, Gibbs JR, and plays a dual role in autophagy. Sci Adv 2:e1601167. https://doi.org/ Schymick JC, Laaksovirta H, van Swieten JC, Myllykangas L et al (2011) A 10.1126/sciadv.1601167 hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21- 63. Zhang D, Iyer LM, He F, Aravind L (2012) Discovery of novel DENN proteins: linked ALS-FTD. Neuron 72:257–268. https://doi.org/10.1016/j.neuron.2011.09.010 implications for the evolution of eukaryotic intracellular membrane Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 14 of 14 structures and human disease. Front Genet 3:283. https://doi.org/10.3389/ fgene.2012.00283 64. Zhou X, Paushter DH, Feng T, Pardon CM, Mendoza CS, Hu F (2017a) Regulation of cathepsin D activity by the FTLD protein progranulin. Acta Neuropathol 134:151–153. https://doi.org/10.1007/s00401-017-1719-5 65. Zhou X, Paushter DH, Feng T, Sun L, Reinheckel T, Hu F (2017b) Lysosomal processing of progranulin. Mol Neurodegener 12:62. https://doi.org/10.1186/ s13024-017-0205-9 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals

Loading next page...
 
/lp/springer_journal/loss-of-tmem106b-is-unable-to-ameliorate-frontotemporal-dementia-like-Dl3ts2Ujpi
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Pathology; Neurology
eISSN
2051-5960
D.O.I.
10.1186/s40478-018-0545-x
Publisher site
See Article on Publisher Site

Abstract

Loss-of-function mutations in progranulin (GRN) and a non-coding (GGGGCC) hexanucleotide repeat expansions in C9ORF72 are the two most common genetic causes of frontotemporal lobar degeneration with aggregates of TAR DNA binding protein 43 (FTLD-TDP). TMEM106B encodes a type II transmembrane protein with unknown function. Genetic variants in TMEM106B associated with reduced TMEM106B levels have been identified as disease modifiers in individuals with GRN mutations and C9ORF72 expansions. Recently, loss of Tmem106b has been reported to protect the FTLD-like phenotypes in Grn−/− mice. Here, we generated Tmem106b−/− mice and examined whether loss of Tmem106b could rescue FTLD-like phenotypes in an AAV mouse model of C9ORF72-repeat induced toxicity. Our results showed that neither partial nor complete loss of Tmem106b was able to rescue behavioral deficits induced by the expression of (GGGGCC) repeats (66R). Loss of Tmem106b also failed to ameliorate 66R-induced RNA foci, dipeptide repeat protein formation and pTDP-43 pathological burden. We further found that complete loss of Tmem106b increased astrogliosis, even in the absence of 66R, and failed to rescue 66R-induced neuronal cell loss, whereas partial loss of Tmem106b significantly rescued the neuronal cell loss but not neuroinflammation induced by 66R. Finally, we showed that overexpression of 66R did not alter expression of Tmem106b and other lysosomal genes in vivo, and subsequent analyses in vitro found that transiently knocking down C9ORF72, but not overexpression of 66R, significantly increased TMEM106B and other lysosomal proteins. In summary, reducing Tmem106b levels failed to rescue FTLD-like phenotypes in a mouse model mimicking the toxic gain-of-functions associated with overexpression of 66R. Combined with the observation that loss of C9ORF72 and not 66R overexpression was associated with increased levels of TMEM106B, this work suggests that the protective TMEM106B haplotype may exert its effect in expansion carriers by counteracting lysosomal dysfunction resulting from a loss of C9ORF72. * Correspondence: rademakers.rosa@mayo.edu Alexandra M. Nicholson and Xiaolai Zhou contributed equally to this work. Department of Neuroscience, Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA 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. Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 2 of 14 Introduction aimed to examine whether loss of Tmem106b expression Frontotemporal dementia (FTD) is a devastating neuro- was able to rescue FTD-like behavioral and pathological degenerative disorder with initial symptoms occurring in features observed in an adeno-associated virus (AAV)-- the fifth or sixth decade of life. While most cases of FTD based mouse model mimicking the toxic gain-of-functions develop sporadically, 30–50% of FTD cases report a associated with overexpression of (GGGGCC) repeats. family history [23, 43, 44, 47, 61], in support of a strong genetic component to the disease. Two of the most Methods common gene mutations found to cause FTD reside in Tmem106b knockout mice the progranulin (GRN) and chromosome 9 open reading Tmem106b knockout mice were generated at the Knockout frame 72 (C9ORF72) genes [6, 14, 17, 45]. Causative Mouse Project (KOMP) Repository at the University of GRN mutations leading to FTD include heterozygous California, Davis using the PGS00041_A_C06 targeting vec- missense, nonsense, or frameshift changes that most tor and blastocyst injection of the targeted embryonic stem often lead to nonsense-mediated decay of the mutant cell clone EPD0047_1_E02 generated from C57BL/6 N mRNA and an associated loss of progranulin protein mice. This knock-in first strategy results in the insertion of (PGRN). Individuals with GRN mutations invariably a lacZ gene trap between the first two coding exons (exons present with aggregates of the TAR DNA binding protein 3 and 4) of the mouse Tmem106b gene. Cryopreserved 43 (TDP-43) in affected brain regions, and are thus patho- sperm were purchased and used to inseminate oocytes ob- logically classified as FTLD-TDP [4, 36]. In C9ORF72,a tained from 3-week-old C57BL/6N female mice (Harlan, non-coding (GGGGCC) hexanucleotide repeat expansion Indianapolis, IN). Zygotes that reached the 2-cell-stage 24 h is responsible for up to 25% of familial and 5% of sporadic post insemination were surgically transferred into foster FTD patients [17, 45]. Extensive research has shown that dams (Harlan). DNA obtained from subsequent pups was the presence of these expanded repeats leads to multiple screened by multiplex polymerase chain reaction (PCR) for pathogenic mechanisms, including a loss of C9ORF72 thepresenceofthe NEOcassette beforebreedingasacol- mRNA expression and toxic gain-of-functions resulting ony founder (CSD-Tmem106b-F: 5’-TTCTCTCCATGTGC from nuclear RNA aggregates and dipeptide repeats pro- TGCATTATGAGC-3′; CSD-Neo-F: 5’-GGGATCTCATG teins [5, 17, 33, 34, 45]. FTD patients with C9ORF72 CTGGAGTTCTTCG-3′; CDS-Tmem106b-ttR: 5’-ACGTG expansions also present with FTLD-TDP at autopsy, CTTCTCTCATCTACAGTTTTCC-3′). A Tmem106b+/− suggesting a potentially convergent disease mechanism x Tmem106+/− breeding scheme was used to generate between GRN- and C9ORF72-induced pathogenesis. Tmem106b +/+, +/−,and −/− mice for the experiments. In 2010, a genome-wide association study (GWAS) Both male and female mice of each Tmem106b genotype identified genetic variants at the transmembrane protein were used for all the experiments. All animal studies were 106 B (TMEM106B) gene locus as the first genetic mod- approved by the Mayo Clinic Institutional Animal Care and ifiers of FTLD-TDP [59]. TMEM106B variants were Use Committee. found to be a modifier of disease risk in FTLD-TDP pa- tients of unknown cause, and a modifier of disease pene- Genotyping trance and presentation in GRN mutation and C9ORF72 Genomic DNA (gDNA) was extracted and PCR-amplified expansion carriers [13, 19, 21, 30, 37, 58–60]. Specific- using the Phire Tissue Direct Master Mix kit (Thermo ally, in C9ORF72 carriers, we showed that individuals Scientific Inc., Waltham, MA) per the manufacturer’s who were also homozygous for the minor alleles at the instructions. Briefly, mouse hair follicles were digested associated TMEM106B variants were significantly pro- in Dilution Buffer supplemented with DNARelease tected from developing FTD but not amyotrophic lateral Additive for 2 min at room temperature, followed by sclerosis (ALS) symptoms [18, 58], another common 2 min incubation at 95 °C. Samples were briefly phenotypic presentation in C9ORF72 expansion carriers. centrifuged and 1 μl of supernatant containing the The TMEM106B protein resides in lysosomal com- gDNA was used for each PCR reaction. Tmem106b partments where it might be involved in lysosomal func- gene products were amplified using a multiplex PCR tion and/or trafficking [7, 11, 29, 50, 53]. Overexpression approach containing 0.8 μMofeachforward primer of TMEM106B results in abnormal lysosomal size, num- (CSD-Tmem106b-F and CSD-Neo-F), 0.8 μMof reverse ber, and acidification [7, 11]. Interestingly, recent studies primer (CDS-Tmem106b-ttR), and Phire Tissue Direct determined that the protective TMEM106B variants are PCR Master Mix (Thermo Scientific). associated with reduced levels of TMEM106B [20, 37, 59], suggesting that lowering TMEM106B might be thera- Viral production and injections peutic in the context of FTD. In fact, lysosomal deficits Viruses were generated as previously described [12, 22, 54]. observed in Grn knockout mice were recently rescued by Briefly, (GGGGCC) (2R) or (GGGGCC) (66R) C9ORF72 2 66 loss of Tmem106b expression [26]. In this study, we repeats were cloned into the pAM/CBA-pl-WPRE-BGH Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 3 of 14 vector containing inverted repeats of serotype 2. AAV vec- cages and transferred to a different room with reduced tors containing the repeats were packaged into the serotype light for at least 1 h. Contextual cues were changed by 9 type capsid by co-transfection with helper plasmids into altering the environment, shape, and smell of the testing HEK293T cells. The cells were harvested and lysed 2 days chamber, as well as by covering the chamber floor with post transfection in the presence of 0.5% sodium deoxycho- opaque plastic. Each mouse was placed back into the test late and 50 U/ml Benzonase (Sigma Aldrich, St. Louis, chamber and the auditory stimulus was presented. Freez- MO) by freeze thawing. The virus was isolated using a ing was recorded for 3 min (cued test). For both the con- discontinuous iodixanol gradient, and qPCR was used to text and cued tests, baseline freezing time was subtracted determine the genomic titer of each virus. 2R and 66R from the freezing time obtained during each test. AAV were diluted to 1 13 genomes/ml in sterile phosphate-buffered saline (PBS) before injection. Mouse Social interaction test pups underwent intracerebroventricular injections with Each mouse was placed into a rectangular box subdi- virus at postnatal day 0 (P0) [10, 12, 25]. Pups were cryoa- vided into three chambers. Two larger chambers nesthetized on ice and their heads were wiped with a sterile measured 17 × 40 cm with a smaller chamber of 5 × 40 alcohol pad. Two microliters of virus were manually cm in the middle. The three chambers were connected injected into each cerebral ventricle using a 32 gauge needle by an 8 × 5 cm opening to allow the mouse free access attached to a 10 μl syringe (Hamilton Company, to all chambers. Two empty, inverted wire-mesh cylin- Reno,NV).After injection,pups were warmed on a ders were placed in opposite corners of each large cham- heating pad and placed back with the dam. All litters ber. In the first trial, mice were placed in the box and were injected within an 8-day timeframe and mice allowed to explore the apparatus freely for 4 min before were aged to 12 months before assessing behavior being placed into a temporary holding cage. Next, a and pathological manifestations. Small subsets of mice probe mouse (matched for sex/strain) was placed in one were harvested at 3 months of age to study of the cylinders for 3 min prior to reintroduction of the Tmem106b expression and validate the model. test mouse. An overhead camera and Anymaze software (Stoelting Co.) were used to monitor mouse interactions Open field test for 10 min. The time each test mouse spent in the area Mice were placed in a square, Perspex box (40x40x30cm, containing the cylinder with the probe mouse was used LxWxH) containing side-mounted photobeams placed to determine sociability. 7.6 cm above the bottom of the box. Mice were allowed to move freely for 15 min, during which locomotor activity Tissue harvests and anxiety measurements were taken. The Perspex box Mice were subjected to carbon dioxide narcosis and was illuminated by a light suspended over the chamber, body weight was obtained (SCALTEC SBA 53 scale; and an overhead camera and AnyMaze software (Wood Denver Instrument, Bohemia, NY) before decapitation. Dale, IL) were used to monitor mouse movement, such as Mouse blood was collected in tubes containing 1.6 mg/ time mobile, total distance traveled, and distance traveled ml EDTA and placed on ice. Blood samples were centri- in the outer and center zones. Mouse rearing was fuged at 4 °C for 10 min at 5000 rpm, after which the recorded by breaking of the photobeams. resulting plasma supernatant was transferred to a new tube for storage at -80 °C until use. The brain was re- Conditional fear testing moved and its weight recorded (SCALTEC SBC 32 scale; Each mouse was placed in a sound-reducing chamber Denver Instrument) before separating the hemispheres. containing a grid floor capable of inducing an electric Whole mouse brains from uninjected mice were either shock. An overhead camera and FreezeFrame software immediately dehydrated and flash frozen in a beaker of (Actimetrics, Wilmette, IL) were used to measure freez- isopentane on dry ice or fixed for 24 h at 4 °C in 4% para- ing. The mice were left undisturbed for the first 2 min formaldehyde (PFA) prepared in PBS. For all injected of the test and baseline freezing was recorded. An 80-dB mice, the left hemisphere was fixed in PFA at 4 °C for white noise was then administered for 30 s (conditioned 48 h. The right hemisphere was immediately dehydrated stimulus; CS). During the last 2 s of the CS, a 0.5 mA and flash frozen in a beaker of isopentane on dry ice. Fol- foot shock was administered to the mouse (uncondi- lowing PFA fixation, brain tissues were washed and stored tioned stimulus; US). After 1 min, a second CS-US pair in PBS at 4 °C until being embedded in paraffin wax. was given to the mouse and the mouse was removed from the chamber and placed in his/her home cage 30 s Cell culture and transfection later. Each mouse was returned to the testing chamber HeLa and U251 cells (ATCC, Manassas, VA) were cul- 24 h later and freezing behavior was recorded for 5 min tured and maintained in Eagle’s Minimum Essential (context test). All mice were returned to their home Medium (EMEM) supplemented with 10% FBS and 1% Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 4 of 14 penicillin/streptavidin. All cell lines were maintained at Mm00479862_g1), Gfap (Mm01253033_m1), and Gapdh 37 °C, 5% CO . For overexpression studies, cells were (Mm99999915_g1). transiently transfected with pAAV C9ORF72 2R or pAAV C9ORF72 66R using Lipofectamine 2000 (Invitrogen, Western blotting Carlsbad, CA) by mixing DNAs with the transfection Protein samples were mixed with an equivalent volume reagent in OptiMEM (Life Technologies, Carlsbad, CA, of 2X Novex sample buffer (Life Technologies) supple- USA) according to the manufacturer’sprotocol. Forthe mented to 5% β-mercaptoethanol. Proteins were dena- siRNA knockdown experiments, HeLa and U251 cells tured by incubating at room temperature for 30 min or were transfected with 20 nM of either negative control by heating at 95 °C for 1–5 min before loading into siRNA or siRNAs against human C9ORF72 using SDS-polyacrylamide gels (Life Technologies). Proteins Lipofectamine RNAiMAX Reagent (Life Technologies) were transferred to Immobilon membranes (Millipore, according the manufacturer’s protocol. The control siRNA Billerica, MA) and immunoblotted with the primary (5’-UGGUUUACAUGUCGACUAA-3′, D-001210-05) and antibody at 4 °C overnight. The next day, blots were human C9ORF72 siRNA (5’-CAUAGAGUGUGUGUUG incubated with an HRP-conjugated secondary antibody AUA-3′, J-013341-11) were purchased from Dharmacon (Promega, Madison, WI) and bands were detected by (Lafayette, CO). Cells were harvested for protein extraction enhanced chemiluminescence using Western Lightning 48 and 72 h post transfection for overexpression and Plus-ECL reagents (Perkin Elmer, Waltham, MA). Primary siRNA experiments, respectively. antibodies included: rabbit anti-Tmem106b from Bethyl Laboratories (A303-439A), rabbit anti-Tmem106b gener- RNA and protein extraction ously shared and derived in the lab of Dr. Fenghua Hu, Frozen brain tissue was homogenized by sonication in sheep anti-mouse progranulin (AF2557; R&D sys- tris-buffered saline (TBS) containing 2X protease and tems, Minneapolis, MN), goat anti-human progranu- phosphatase inhibitors (Thermo Scientific). RNA was lin (AF2420; R&D systems, Minneapolis, MN), mouse isolated from 75 μl of brain homogenate using the anti-Gapdh (H86504M; Meridian Life Sciences, Cincinnati, RNeasy Plus Mini Kit (Qiagen, #74136) according to the OH), mouse anti-Lamp1 (sc-20,011; Santa Cruz Biotech- manufacturer’s instructions. Briefly, brain tissue was nology, Dallas, TX), rabbit anti-C9ORF72 (ABN1645; lysed using Buffer RLT containing β-ME, and then Millipore), mouse anti-HA (clone12CA5; #11583816001; passed through the gDNA Eliminator column to remove Roche, Indianapolis, IN), and goat anti-Cathepsin-D (clone DNA. The RNA containing flow-through was precipi- C-20; sc-6486; Santa Cruz Biotechnology, Dallas, TX). tated by 70% ethanol and passed through an RNeasy Bands of Western blots were quantified using Image J pink spin column. RNA was eluted from the column (NIH, Bethesda, MD). with RNase-free water. Protein was also isolated from 75 μl of brain homogenate by adding and equivalent vol- Poly(GP) immunoassay ume of 2X Radioimmunoprecipitation Assay (RIPA) Poly(GP) protein levels were measured in 10 μgof buffer (Boston BioProducts, Ashland, MA). For cell protein in duplicate from mouse brain lysates using a culture experiments, media was removed and RIPA buf- sandwich immunoassay utilizing MesoScale Discovery fer was added directly to PBS-rinsed cell culture wells. (MSD) technology as previously described [12, 22]. Serial All RIPA samples were incubated on ice and centrifuged dilutions of recombinant (GP) were used as a standard at 4 °C for 5 min at 6000 rpm to clear debris. Protein curve. Response values were measured using the MSD content in brain samples was measured in the super- QUICKPLEX SQ120 and are defined as the intensity of natant using the bicinchoninic acid (BCA) assay emitted light upon electrochemical stimulation. Each (Thermo Fisher Scientific). sample’s response value was corrected for background response detected in 2R-injected mouse samples prior to Quantitative PCR interpolation of poly(GP) levels using the standard Brain RNA was reverse transcribed using the Superscript curve. III complimentary DNA (cDNA) synthesis kit, random hexamers (Life Technologies), and an equal ratio of random Immunohistochemistry and digital analysis hexamers and Oligo dT primers (Thermo Fisher Scientific). Paraffin-embedded mouse brain tissues were cut on a Real-time quantitative PCRs (qPCRs) were conducted using sagittal plane at a thickness of 5 μm, deparaffinized with TaqMan gene expression assays and the QuantStudio 7 xylene, and rehydrated in a series of ethanol washes. For Flex Real-Time PCR System (Applied Biosystems, Foster slides stained with mouse anti-NeuN (ABN78; Millipore) City, CA). All probes were purchased from Life Technolo- or rabbit anti-pTdp-43 (TIP-PTD-P01; pSer409/410; gies: Tmem106a (Mm01246747_m1), Tmem106b (Mm005 Cosmo Bio USA, Carlsbad, CA), antigen retrieval was 10952_m1), Tmem106c (Mm01303550_m1), Iba1 (Aif1; performed by steaming slides for 30 min with distilled Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 5 of 14 water before blocking in 0.03% hydrogen peroxide. Immu- under 63× magnification and the number of cells con- nostaining of sections was done using a Dako Autostainer taining RNA foci were quantified. and Envision + HRP system (Dako, Carpintaria, CA) per the manufacturer’s instructions. For slides stained with Statistical analyses rabbit anti-pTdp-43 (pSer409/410; gift from Dr. Leonard For experiments in which only two groups were com- Petrucelli), antigen retrieval was performed by steaming pared, significance was measured using a Student’s slides for 30 min in sodium citrate buffer (10 mM sodium t-test. For analyses involving more than two groups, citrate, pH 6.0 with 0.05% Tween-20) and immunostaining GraphPad Prism 5.04 (GraphPad Software) was utilized was performed using the VectaStain Elite ABC HRP kit to perform a one-way ANOVA or two-way ANOVA (Vector Laboratories, Burlingame, CA) per the manufac- followed by the Fisher’s LSD post hoc test. turer’s instructions. All slides were counterstained with hematoxylin, washed in a series of alcohols, and dehy- Results drated in xylene. Glass coverslips were mounted using Generation and validation of Tmem106b knockout mice Cytoseal XYL (Thermo Scientific) and were left to set at Tmem106b knockout alleles were generated by insertion room temperature for 48 h before scanning with an of a lacZ gene trap in the intronic region between the Aperio ScanScope AT2 Slide Scanner (Leica Biosystems, first two coding exons, exons 3 and 4. The inserted gene Buffalo Grove, IL). ImageScope software (v12.1.0.5029; is transcribed by the endogenous Tmem106b promoter Leica) was used to annotate the cerebral cortex and along with the upstream exons and leads to a premature motor cortex of NeuN- and pTdp-43-stained slides. A termination of transcription (Fig. 1a). Inheritance of this custom-designed algorithm was applied to detect the targeted gene disruption was confirmed by PCR amplifi- number of pTdp-43-positive nuclei per area (mm ) cation of the genomic DNA isolated from Tmem106b [12] when labeled with the Cosmo antibody. The total +/+, +/−, and −/− mice (Fig. 1b). qPCR analysis in 3, 8 number of pTdp-43-positive cells were counted and 15 months old mice further confirmed the loss of manually for slides stained with the pTdp-43 antibody Tmem106b mRNA transcripts with mice heterozygous provided by Dr. Petrucelli. for the knockout allele showing approximately 50% loss as compared to age matched wild-type mice and a near RNA fluorescence in situ hybridization (FISH) complete loss of Tmem106b mRNA in Tmem106b −/− RNA FISH was performed in fixed mouse brain tissue as mice (Fig. 1c, Additional file 1: Figure S1). Loss of done previously [12, 28]. Briefly, paraffin embedded Tmem106b transcripts did not alter the expression of brain sections were deparaffinized in xylene and rehy- other Tmem106 family members, Tmem106a and drated in a series of ethanol solutions. Sections were Tmem106c (Additional file 1:FigureS2). Western permeabilized with ice-cold 2% acetone in PBS prepared blotting of wild-type mouse brain lysates with a in DEPC-treated water for 5 min. Sections were then Tmem106b antibody (Bethyl Laboratories) revealed a washed twice with DEPC-treated water and dehydrated robust Tmem106b-immunoreactive band at the pre- in a series of ethanol solutions before incubating 30 min dicted 43 kDa molecular weight. This band’sintensity at 66 °C in pre-hybridization buffer [50% formamide/2X was reduced approximately 50% in samples from SSC (MIDSCI, Valley Park, MO), 10% dextran sulfate Tmem106b +/− mice and was undetectable in Tmem106b (Millipore), 2× saline-sodium citrate buffer, 50 mM so- −/− brain tissue (Fig. 1d, e). Of note, using an in-house dium phosphate buffer, pH 7.0]. A fluorescently labeled developed Tmem106b antibody against the intracellular locked nucleic acid (LNA) probe (5’-TYE563/CCCC domain of TMEM106B (residues 1–96) [7], small molecu- GGCCCCGGCCCC-3′, Exiqon, Inc.; batch number lar weight bands could also be detected upon long expos- 612968) was diluted to 40 nM in in hybridization buffer ure in Tmem106b +/− and Tmem106b −/− mice. These (10% dextran sulfate, 50% formamide, 20 ng/μl BSA, additional bands likely correspond to Tmem106b-lacZ 25 mM tRNA, 25 nM EDTA, 2X SSC, 25 mM sodium truncated fragments which are expected in mouse phosphate buffer) and denatured at 80 °C for 5 min models created by the lacZ gene trap approach [9] before hybridizing to the tissue for 24 h at 66 °C in a (Additional file 1:FigureS3). dark, humid chamber. Sections were subsequently washed once with 2X SSC (0.1% Tween-20) at room Tmem106b loss does not reverse abnormal behaviors in temperature for 5 min and washed twice with mice expressing (GGGGCC) repeat pre-warmed 0.2X SSC at 55 °C for 10 min in the dark. To determine whether reduced Tmem106b levels might Slides were mounted with Vectashield mounting media be protective against FTD-like behavioral phenotypes in containing DAPI (Vector Laboratories). Cortical images the AAV-(GGGGCC) mouse model, the newborn of RNA foci were obtained using a Zeiss Axio Imager Z1 Tmem106b +/+, +/−, and −/− mice were injected with fluorescent microscope (Zeiss, Oberkochen, Germany) either AAV- 66R or AAV-2R (control) and aged to Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 6 of 14 Fig. 1 Generation of mice with targeted Tmem106b gene disruption. a Genomic structure of the wild-type mouse Tmem106b allele and the gene trap vector used to target the mouse Tmem106b gene. Rectangular boxes represent exons that are shaded in gray or white to denote coding and non-coding exons, respectively. White arrowheads signify Flp recombination target (FRT) sites and black arrowheads represent loci of crossover in P1 (LoxP) sites. Genotyping primer binding sites are labeled and denoted with black arrows above each genomic structure. b PCR analysis of DNA obtained from the hair follicles of Tmem106b wild-type (+/+), heterozygous (+/−), or knockout (−/−) mice using the primers depicted in panel a. c Quantitative PCR analysis measuring Tmem106b mRNA levels in Tmem106b +/+, +/−, and −/− mouse brain (n = 4 per genotype) at 3 months of age. The graph represents the mean ± S.E.M.; ****p < 0.0001 by one-way ANOVA followed by a Fisher’s LSD post-hoc test. d Western blot depicting Tmem106b protein levels (black arrowhead) in 3-month-old Tmem106b +/+, +/−, and −/− mouse brain tissue. Gapdh was used as a loading control. e Quantification of Tmem106b protein levels in Tmem106b +/+, +/-, and -/- mouse brain at 3 months of age (n =3 per genotype) 12 months of age. 66R virus injection did not affect the all 66R-injected mouse brains, which showed no sig- overall activity (data not shown) or body weight of the nificant difference in 66R mRNA expression among mice; except for a reduced body weight in female Tmem106b +/+, +/−,and −/− mice (Fig. 2f). Tmem106b +/+ mice injected with 66R as compared to 2R which could be contributed to the small number of (GGGGCC) repeat-induced neuropathology is not mice in this sub-group (Additional file 1: Figure S4). In rescued by lowering Tmem106b expression line with the original study describing this model [12], We next determined whether lowering Tmem106b injection of the 66R virus into wild-type mice induced expression was able to ameliorate key hallmarks of significant behavioral deficits, including anxiety and re- neurodegeneration, such as neuroinflammation and duced sociability as compared to 2R-injected wild-type neuronal loss, previously reported in the AAV-66R mice (p=0.019 by Student’s t-test for all wild-type 2R vs. mouse model [12]. At 12 months of age, wild-type mice 66R analyses; Fig. 2a-e). However, neither partial nor injected with 66R virus showed significantly increased complete reduction of Tmem106b altered the anxiety levels of Iba1 and Gfap mRNA transcripts as compared phenotype observed in 66R-injected wild-type mice as to 2R-injected animals, indicating pronounced neuroin- determined by the open field assay (Fig. 2a, b) or fear flammation (p = 0.024 by Student’s t-test for both ana- conditioning tests (Fig. 2c, d). Also, the reduced mouse lyses; Fig. 3a, b). However, neither partial nor complete sociability observed in 66R-injected mice could not be reduction of Tmem106b levels was able to rescue these rescued in mice with partial or complete loss of changes; in fact, 66R-injected Tmem106b −/− mice had Tmem106b (Fig. 2e). Since the amount of 66R viral significantly higher Gfap mRNA levels than 66R-injected expression could affect the presence and/or severity wild-type mice (Fig. 3b). Follow-up analysis in unin- of these phenotypes, we performed qPCR analysis on jected Tmem106b +/+, +/−, and −/− mice further Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 7 of 14 levels in Tmem106b −/− as compared to Tmem106 +/+ mice (Additional file 1: Figure S5). Gfap mRNA levels in Tmem106b +/− mice were not significantly different from Tmem106b +/+ mice. We next assessed the effect of Tmem106b reduction on 66R-induced neuronal loss. As expected, 66R-injected Tmem106b +/+ mice portrayed a significantly reduced number of cells immunoreactive for the neuronal marker, NeuN, as compared to 2R-injected mice (p =0.016 by Student’s t-test; Fig. 3c). Complete loss of Tmem106b did not modify this phenotype; however, partial reduction of Tmem106b significantly lessened the extent of neuronal loss associated with 66R injection. In fact, 66R-injected Tmem106b +/− cortical NeuN counts were not sig- nificantly different from that of 2R-injected animals (p = 0.0512 by Student’st-test; Fig. 3c, Additional file 1: Figure S6). This was not due to the presence of more NeuN-positive cells present in Tmem106b heterozygous mice given that uninjected Tmem106b +/+, +/−,and −/− mice have comparable NeuN counts (Additional file 1:FigureS5). We next studied two unique pathologies induced by overexpression of the 66R repeat: RNA foci and dipep- tide repeat proteins. Parallel to what was previously re- ported, RNA foci and poly-glycine/proline dipeptides [poly(GP)] were detected at significant levels in 66R-injected wild-type mice and not 2R-injected mice (Fig. 3d-f). We quantified the percentage of cells con- taining one or more RNA foci in the cortex of 66R-injected Tmem106b +/+, +/−, and −/− mice. As depicted in Fig. 3e, partial or complete reduction of Tmem106b levels did not significantly change the number of RNA foci-containing cells. Similarly, poly(GP) levels in 66R-injected Tmem106b +/− and −/− mice were similar to that of 66R-injected wild-type mice (Fig. 3f). Fig. 2 (GGGGCC) repeat-induced behavioral deficits are Finally, we examined the effect of Tmem106b unchanged by Tmem106b reduction. a Quantification of the average reduction on the formation of intracellular inclusions distance traveled obtained during the open field assay for 66R-injected of phosphorylated Tdp-43 (pTdp-43), a key patho- Tmem106b +/+, +/−, −/− mice at 12 months of age. Wild-type mice of logical feature observed both in patients and in mice the same age that had been injected with 2R C9ORF72 were used as a injected with 66R [12, 17, 45]. We detected intense control. b Quantification of the average speed traveled during the open field assay for mice as described in (a). c-d Quantification of the pTdp-43-positive inclusions in 66R- but not 2R-injected time each mouse spent freezing during the contextual (c) and cued Tmem106b +/+ brains using two different antibodies that (d) fear conditioning tests. e Quantification of the time each mouse recognize Tdp-43 in its phosphorylated form (Fig. 4a, spent exploring the mouse-containing cup during the social Additional file 1: Figure S7). However, neither partial interaction test (f qPCR quantification measuring the amount of 66R nor complete loss of Tmem106b significantly changed mRNA obtained from the brains of 12-month-old 66R-injected Tmem106b +/+, +/−,and −/− mice. Graphs represent the mean ± the number of pTdp-43-positive cells in the cortex or S.E.M. Data was analyzed by one-way ANOVA followed by Fisher’sLSD hippocampus (Fig. 4b; Additional file 1:FigureS7). post-hoc, and results are shown for 66R-injected Tmem106b +/− and −/− mice as compared to 66R-injected wild-type mice (n ≥ 12 for all C9ORF72 loss increases TMEM106B levels unlike (GGGG groups). NS, not significant CC) overexpression To further study the connection between TMEM106B showed that full loss of Tmem106b (in the absence of and C9ORF72-related disease mechanisms, we deter- repeat overexpression), is sufficient to induce astrogliosis mined Tmem106b protein levels in wild-type mice that as determined by a significant increase in Gfap mRNA had been injected with 66R or 2R virus. Tmem106b Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 8 of 14 Fig. 3 Assessment of (GGGGCC) -mediated neuropathology in response to changes in Tmem106b levels (a-b) Quantification of Iba1 (a)or Gfap (b) mRNA levels in the cortex of 2R-injected mice, or in 66R-injected Tmem106b +/+, +/−,or −/− mice. c Quantitative analysis of the number of cells immunoreactive for NeuN in the cortex of 2R-injected mice, or in 66R-injected Tmem106b +/+, +/−,or −/− mice. d Image depicting the presence of RNA foci (arrowheads) in the nuclei of cortical cells in wild-type 66R- versus 2R-injected mice. e Quantification of the number of cells with RNA foci in the cortex of wild-type 2R-injected mice, or in 66R-injected Tmem106b +/+, +/−,or −/− mice. f Quantitative assessment of the presence of poly(GP) peptides detected in the brains of mice injected as described in panel e. Graphs represent the mean ± S.E.M. Data was analyzed by one-way ANOVA followed by Fisher’s LSD post-hoc, and results are shown for 66R-injected Tmem106b +/− and −/− mice as compared to 66R-injected wild-type mice (n = 12 per group). NS, not significant.; *p < 0.05; **p < 0.001 Fig. 4 Tmem106b levels do not affect (GGGGCC) repeat induced pTdp-43 inclusion body formation. a Representative images of pTdp-43 (pS409/410 from Cosmo Bio) staining of motor cortex region of the mouse brains from indicated Tmem106b genotypes 12 months after 2R and 66R AAV injection. Arrow heads indicate pTdp-43-positive inclusion bodies. b and c Quantification of pTdp-43 inclusion body positive cells in different brain regions: cortex (b) or hippocampus (c) from Tmem106b +/+, +/−, and −/− mice as compared to WT 2R injected mice (n = 4). Graphs represent the mean ± S.E.M. Data was analyzed by one-way ANOVA followed by Fisher’s LSD post-hoc, and results are shown for 66R-injected Tmem106b +/− and −/− mice as compared to 66R-injected wild-type mice (n = 12 per group). NS, not significant Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 9 of 14 levels in 66R injected brains were almost identical to was chosen as the model system in this study due to the that of 2R control brains (Fig. 5a, b). The 66R-injected demonstration of various phenotypes associated with mice also had normal levels of other lysosomal proteins, C9ORF72 repeat expansions as early as 6 months of age such as Lamp1, cathepsin D (both pro- and mature [12]. At the time this study commenced, only two other forms), and progranulin (Fig. 5a, quantifications not characterized C9ORF72-repeat expansion mouse models shown). Furthermore, whereas human patients with had been generated. Both of these models were created C9ORF72-related FTD have reduced C9ORF72 levels in using a bacterial artificial chromosome (BAC) for the addition to the GGGGCC-repeat associated toxicities, expression of the complete [39] or partial [41] C9ORF72 endogenous C9ORF72 expression levels were not chan- coding region with incorporated (GGGGCC) repeats of ged in our 66R overexpression model (Additional file 1: various lengths. These models successfully demonstrated Figure S8). This prompted us to compare the effects of pathologies directly related to the repeat expansion as 66R overexpression and C9ORF72 loss on Tmem106b early as 4–6 months of age, including RNA foci and the levels in cell culture (techniques validated in Additional generation of dipeptide repeat proteins. However, in file 1: Figure S9). Consistent with the in vivo data, over- contrast to the (GGGGCC) -AAV model, many of the expression of 66R in both HeLa and U251 cells failed to key features of C9ORF72-mediated FTD, especially change the expression of various lysosomal proteins, behavioral deficits, neuroinflammation, neuronal loss, including TMEM106B (Fig. 6a-e and Additional file 1: and pTdp-43 pathology, were not observed even in aged Figure S10). However, knockdown of C9ORF72 signifi- mice in these models [39, 41]. Since TDP-43 pathology cantly increased protein levels of TMEM106B together is a common denominator of the human disease popula- with other lysosomal resident proteins (Fig. 6f-j and tions associated with TMEM106B (FTLD-TDP, GRN-car- Additional file 1: Figure S10). riers, C9ORF72-carriers and AD patients with TDP-43 pathology [38]) the selection of a mouse model with Discussion pTdp-43 pathology was considered essential. Importantly Genetic variants in TMEM106B associated with reduced however, neither the BAC mice nor our (GGGG TMEM106B expression have been shown to significantly CC) -AAV mice recapitulate the loss of C9ORF72 expres- protect individuals with either GRN mutations or sion consistently observed in human C9ORF72 expansion C9ORF72 repeat expansions from the development of carriers. FTD symptoms [13, 19, 21, 30, 37, 58–60]. Moreover, Neuronal loss and neuroinflammation are common depletion of Tmem106b was recently shown to rescue features among neurodegenerative disorders. Protective several disease-relevant phenotypes observed in Grn −/− TMEM106B variants were found to associate with mice [26]. This study is the first to determine the impact increased neuronal gene expression, reduced expression of Tmem106b reduction on C9ORF72-related disease. of genes involved in inflammation, and better cognitive Contrary to the relative success in the Grn −/− mouse performanceinhealthyagedindividuals [46]. However, we model, we demonstrate that neither partial nor complete did not see a significant improvement in measures of loss of Tmem106b is sufficient to rescue the behavioral neuroinflammation or behavioral deficits induced by changes or neuropathological phenotypes that manifest the overexpression of the (GGGGCC) repeat when in an AAV-based mouse model of C9ORF72-associated Tmem106b levels were reduced. In fact, (GGGGCC) -in- (GGGGCC) expansions. The (GGGGCC) -AAV model jected Tmem106b knockout mice showed more severe n 66 Fig. 5 (GGGGCC) repeat expansion overexpression does not alter Tmem106b protein levels in mouse brain. (a) Western blot of brain tissue obtained from 2R- or 66R-injected wild-type mice using antibodies against various lysosomal proteins. Gapdh was used as a loading control. (b) Quantification of Tmem106b protein levels by Western blot as depicted in panel a. The graph represents the mean ± S.E.M. by Student’s t- test (n = 8 for 2R; n = 12 for 66R ); NS, not significant Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 10 of 14 Fig. 6 The effect of (GGGGCC) overexpression or C9ORF72 knockdown on TMEM106B protein levels in HeLa cells. a Western blot of HeLa cells transfected with either 2R or 66R pAAV. b-e Protein quantification of TMEM106B (b), LAMP1 (c), CTSD (pro-form, mature form has similar results as the pro-form) (d), and PGRN (e) in cells transfected as described in panel (a). f Western blot of HeLa cells transfected with either control siRNA or siRNA against C9ORF72 (g-j), Protein quantification of TMEM106B (g), LAMP1 (h), CTSD (i), and PGRN (j) in cells transfected as described in panel f. GAPDH was used as a loading control. Graphs represent the mean ± S.E.M. by Student’st-test (n = 6 for all groups). NS, not significant; *p <0.05, **p <0.01 astrogliosis, evidenced by increased Gfap expression, as that did not find an association between TMEM106B compared to Tmem106b wild-type or heterozygous mice variants and dipeptide repeat pathology in C9ORF72 injected with the expanded repeat. We showed that mutation carriers [16]. Given the strong genetic associ- this was due to increased astrogliosis from Tmem106b ation of both GRN and C9ORF72 carriers with loss alone, excluding the possibility that loss of TMEM106B variants, the lack of association in our Tmem106b renders mice more sensitive to a C9ORF72 Tmem106b model with pathological features that are repeat-mediated inflammatory response. Our observa- unique to C9ORF72GGGGCC repeat expansion carriers tion of astrogliosis resulting from loss of Tmem106b may not be surprising. Nevertheless, both C9ORF72 and alone suggests that Tmem106b plays a unique and po- GRN mutation carriers present with TDP-43 pathology tentially necessary role in astrocytes. Tmem106b is a at autopsy. Indeed, TMEM106B variants were first lysosomal resident protein, and loss of Tmem106b has discovered as disease modifiers in an FTD cohort been recently shown to cause lysosomal dysfunctions comprised of individuals with TDP-43 brain pathology including lysosomal acidification and trafficking prob- regardless of underlying cause [59]. TMEM106B variants lems [26, 49]. Importantly, dysfunction of lysosomes were additionally found to associate with the presence of in multiple diseases, for instance, lysosomal storage TDP-43 pathology in other diseases, such as Alzheimer’s disorders has been tightly linked to astroglial activa- disease and hippocampal sclerosis [3, 35, 48]. In fact, the tion [42]. Interestingly, activation of the Tmem106b protective TMEM106B variants appeared to corres- paralog, Tmem106a, was shown to be immunostimula- pond with lessened TDP-43 aggregate burden in a tory in mouse macrophages [15]. As such, our data preliminary study of eight C9ORF72 mutation carriers suggests that TMEM106B might play a novel, recipro- [58]. Collectively, these findings suggest that TMEM106B cal role in inflammatory modulation. In relation to the protective variants may reduce one’s risk of developing neuronal loss, we did observe that partial, but not TDP-43 proteinopathies; however, despite careful analysis complete loss of Tmem106b significantly lessened the with two independent antibodies, we found that reducing extent of neuronal loss in the AAV-(GGGGCC) Tmem106b levels in (GGGGCC) -injected mice did not injected mice, suggesting that, if pursued, partial lessen the development of phosphorylated pTdp-43 aggre- TMEM106B reduction may be a more viable avenue gates in the cortex or hippocampus at 12 months of age. for future TMEM106B-related therapeutic approaches The inability to ameliorate neuropathological and in FTD. behavioral deficits through the reduction of Tmem106b Tmem106b levels also had no effect on two other key in the AAV-(GGGGCC) model requires a careful pathological features observed in the (GGGGCC) -AAV examination of the approaches employed in this study. model: RNA foci and the generation of dipeptide repeat First, we modelled the protective TMEM106B human proteins. These data are in line with a previous report haplotype by reducing Tmem106b by approximately 50 Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 11 of 14 or 100%. This seems appropriate since multiple studies lysosomal function [2, 51, 55, 62]. Specifically, loss of have observed lower TMEM106B mRNA levels or a fas- C9ORF72, as seen in human patients carrying C9ORF72 ter TMEM106B protein degradation associated with the repeat expansions, causes impaired mTORC1 signaling TMEM106B protective haplotype [20, 37, 59]. We can- and abnormal lysosome morphology indicative of dysfunc- not, however, exclude that the associated variants affect tion [2, 57]. Thus, it is conceivable that the effect of the TMEM106B in an unknown fashion, for example as a protective TMEM106B haplotype in C9ORF72 expansion result of the p.Thr185Ser variant (rs3173615) which al- carriers is to counteract lysosomal dysfunction that results ters the protein coding sequence of TMEM106B and is from the loss of C9ORF72 expression. In support of this inherited as part of the protective haplotype [59]. It is hypothesis, we showed that reducing C9ORF72 levels in also possible that the protective TMEM106B haplotype human cell lines significantly increased levels of various induces a more subtle decrease in TMEM106B than lysosomal proteins, including TMEM106B, which is what we modeled in this study or that the N-terminal thought to signify lysosomal dysfunction. These types of TMEM106B fragments, which we detected at low lysosomal changes were not observed by overexpression levels in knockout animals, may have retained some of (GGGGCC) in cell culture, nor were these changes partial TMEM106B function. Second, we modelled observed in our AAV-(GGGGCC) mouse model. the C9ORF72-associated repeat expansion through the overexpression of (GGGGCC) by AAV. This ap- Conclusions proach importantly results in pTdp-43-positive inclu- In summary, we show that reducing the levels of sions as early as 6 months of age [12]; however, this Tmem106b in a mouse model mimicking the toxic model may have been too aggressive to reverse the gain-of-functions associated with the C9ORF72 (GGGG neurodegenerative, neuropathological, and behavioral CC) repeat expansions is unable to ameliorate key phenotypes in these mice at 12 months of age. More pathological features seen in FTD patients, including importantly, our approach only recapitulated the RNA pTdp-43 pathology. We further provide support for the and protein toxic gain-of-function mechanisms associated growing body of evidence linking the loss of C9ORF72 with the repeat sequence and failed to model the reduction expression to the pathobiology of C9ORF72, in this case in C9ORF72 transcripts which is now considered and inte- through the induction of lysosomal dysfunction. As gral part of C9ORF72 disease pathogenesis [1, 17, 45, 56]. such, it will be critically important that further examin- Indeed, endogenous C9orf72 levels were unchanged in our ation of the effects of Tmem106b reduction on 66R mice at 12 months of age. C9ORF72 pathobiology be studied in models that in- TMEM106B is a type II lysosomal membrane protein clude reduction of C9ORF72. with currently unknown function. Increases in TMEM106B levels have been found to be cytotoxic and are associated Additional file with increases in lysosomal size and reduced lysosomal Additional file 1: Figures S1 through S10. Figure S1. Transcript acidification, leading to the disruption of endolysosomal- expression of Tmem106b in Tmem106b deficiency mice at different ages. and autophagic-lysosomal degradation [38]. Recent Figure S2. Tmem106b reduction does not alter the expression of its work undeniably links PGRN to lysosomal biology family members. Figure S3. Tmem106b immunoreactivity in mice with Tmem106b gene interruption using an additional antibody. Figure S4. [24, 31, 64, 65] and, as such, it may not have been The body weight of 2R and 66R injected mouse. Figure S5. Tmem106b surprising that Tmem106b loss reversed some of the reduction alone induces astrogliosis. Figure S6. Heterozygous loss of Grn knockout-mediated lysosomal enzyme dysregula- Tmem106b partially rescues 66R injection-induced neuronal loss. Figure S7. pTdp-43 immunoreactivity in 2R and 66R injected mouse brain. tion [26]. This prompts the question of whether Figure S8. Endogenous C9orf72 protein levels in 2R- and 66R-injected Tmem106b reduction might only confer protection in mouse brain. Figure S9. Validation of (GGGGCC) repeat overexpression disease models that portray abnormal lysosomal biol- and C9ORF72 knockdown. Figure S10. The effect of (GGGGCC) overexpression or C9ORF72 knockdown on TMEM106B protein levels in ogy. While lysosomal dysfunction has been implicated U251 cells. (DOCX 26231 kb) in C9ORF72-related pathogenesis, much of the evi- dence to support this comes from studies investigating Acknowledgements the function of the C9ORF72 protein itself and not the This work is funded in part by a Mayo Clinic Edward C. Kendall Research (GGGGCC) repeat expansions [2, 8, 27, 40, 52, 55]. The n Fellowship (AN) and a research fellowship from The Bluefield Project to Cure FTD (XZ). Additional funding was obtained by NIH/NINDS grants C9ORF72 protein sequence contains DENN-like domains, R35NS097261 (RR), UG3NS0103870 (RR) and P01NS084974 (LP, RR, DWD) and making it a part of the DENN protein superfamily which The Bluefield Project to Cure FTD (RR). is known to be involved in regulating membrane traffick- ing and autophagy [32, 63]. Genetic and cell biology Authors’ contributions RR, AMN and XZ designed the study, oversaw the experiments and drafted studies have shown that C9ORF72 interacts with the manuscript. JC and AMN injected the AAV virus into Tmem106b mice. other DENN domain-containing proteins linked to RBP, TP, MB, NF, and BM harvested mouse brains and assisted in mTORC1 signaling, whose activity is closely tied to biochemical, gene expression and genetic analysis of mouse tissues. NF and Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 12 of 14 TP performed and analyzed the cell culture studies. AK, MB and JF 12. Chew J, Gendron TF, Prudencio M, Sasaguri H, Zhang YJ, Castanedes-Casey performed the mouse behavioral analyses. MC-C, LR, VP, MEM and DD M, Lee CW, Jansen-West K, Kurti A, Murray ME et al (2015) performed the immunohistochemical analyses of mouse brains. MD-H Neurodegeneration. C9ORF72 repeat expansions in mice cause TDP-43 performed FISH analysis to quantify RNA foci burden and LD and TFG pathology, neuronal loss, and behavioral deficits. Science 348:1151–1154. performed quantitative analysis of GP levels. LP, KRJ-W and EP generated the https://doi.org/10.1126/science.aaa9344 AAV virus and FH provided Tmem106b antibody. All authors read and a 13. Cruchaga C, Graff C, Chiang HH, Wang J, Hinrichs AL, Spiegel N, Bertelsen S, pproved the final manuscript. Mayo K, Norton JB, Morris JC et al (2011) Association of TMEM106B gene polymorphism with age at onset in Granulin mutation carriers and plasma Ethics approval Granulin protein levels. Arch Neurol. https://doi.org/10.1001/archneurol.2010.350 All animal studies were approved by the Mayo Clinic Institutional Animal 14. Cruts M, Gijselinck I, van der Zee J, Engelborghs S, Wils H, Pirici D, Rademakers Care and Use Committee. R, Vandenberghe R, Dermaut B, Martin JJ et al (2006) Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to Competing interests chromosome 17q21. Nature 442:920–924. https://doi.org/10.1038/nature05017 The authors declare that they have no competing interests. 15. Dai H, Xu D, Su J, Jang J, Chen Y (2015) Transmembrane protein 106a activates mouse peritoneal macrophages via the MAPK and NF-kappaB signaling pathways. Sci Rep 5:12461. https://doi.org/10.1038/srep12461 Publisher’sNote 16. Davidson YS, Barker H, Robinson AC, Thompson JC, Harris J, Troakes C, Springer Nature remains neutral with regard to jurisdictional claims in Smith B, Al-Saraj S, Shaw C, Rollinson S et al (2014) Brain distribution of published maps and institutional affiliations. dipeptide repeat proteins in frontotemporal lobar degeneration and motor neurone disease associated with expansions in C9ORF72. Acta Neuropathol Author details Commun 2:70. https://doi.org/10.1186/2051-5960-2-70 Department of Neuroscience, Mayo Clinic Jacksonville, 4500 San Pablo Road, 17. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, Jacksonville, FL 32224, USA. Department of Molecular Biology and Genetics, Rutherford NJ, Nicholson AM, Finch NA, Flynn H, Adamson J et al (2011) Weill Institute for Cell and Molecular Biology, Cornell University, 345 Weill Expanded GGGGCC hexanucleotide repeat in noncoding region of Hall, Ithaca, NY 14853, USA. C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron 72:245–256. https://doi.org/10.1016/j.neuron.2011.09.011 Received: 10 May 2018 Accepted: 11 May 2018 18. Deming Y, Cruchaga C (2014) TMEM106B: a strong FTLD disease modifier. Acta Neuropathol 127:419–422. https://doi.org/10.1007/s00401-014-1249-3 19. Finch N, Carrasquillo MM, Baker M, Rutherford NJ, Coppola G, Dejesus- References Hernandez M, Crook R, Hunter T, Ghidoni R, Benussi L et al (2011) 1. Amick J, Ferguson SM (2017) C9orf72: at the intersection of lysosome cell TMEM106B regulates progranulin levels and the penetrance of FTLD in GRN biology and neurodegenerative disease. Traffic 18:267–276. https://doi.org/ mutation carriers. Neurology 76:467–474. https://doi.org/10.1212/WNL. 10.1111/tra.12477 0b013e31820a0e3b 2. Amick J, Roczniak-Ferguson A, Ferguson SM (2016) C9orf72 binds SMCR8, 20. Gallagher MD, Posavi M, Huang P, Unger TL, Berlyand Y, Gruenewald AL, localizes to lysosomes, and regulates mTORC1 signaling. Mol Biol Cell 27: Chesi A, Manduchi E, Wells AD, Grant SF et al (2017a) A dementia- 3040–3051. https://doi.org/10.1091/mbc.E16-01-0003 associated risk variant near TMEM106B alters chromatin architecture and 3. Aoki N, Murray ME, Ogaki K, Fujioka S, Rutherford NJ, Rademakers R, Ross gene expression. Am J Hum Genet 101:643–663. https://doi.org/10.1016/j. OA, Dickson DW (2015) Hippocampal sclerosis in Lewy body disease is a ajhg.2017.09.004 TDP-43 proteinopathy similar to FTLD-TDP type a. Acta Neuropathol 129: 21. Gallagher MD, Suh E, Grossman M, Elman L, McCluskey L, Van Swieten JC, 53–64. https://doi.org/10.1007/s00401-014-1358-z Al-Sarraj S, Neumann M, Gelpi E, Ghetti B et al (2014) TMEM106B is a 4. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, Mori H, Mann D, genetic modifier of frontotemporal lobar degeneration with C9orf72 Tsuchiya K, Yoshida M, Hashizume Y et al (2006) TDP-43 is a component of hexanucleotide repeat expansions. Acta Neuropathol 127:407–418. https:// ubiquitin-positive tau-negative inclusions in frontotemporal lobar doi.org/10.1007/s00401-013-1239-x degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res 22. Gendron TF, Bieniek KF, Zhang YJ, Jansen-West K, Ash PE, Caulfield T, Commun 351:602–611. https://doi.org/10.1016/j.bbrc.2006.10.093 Daughrity L, Dunmore JH, Castanedes-Casey M, Chew J et al (2013) Antisense 5. Ash PE, Bieniek KF, Gendron TF, Caulfield T, Lin WL, Dejesus-Hernandez M, transcripts of the expanded C9ORF72 hexanucleotide repeat form nuclear RNA van Blitterswijk MM, Jansen-West K, Paul JW 3rd, Rademakers R et al (2013) foci and undergo repeat-associated non-ATG translation in c9FTD/ALS. Acta Unconventional translation of C9ORF72 GGGGCC expansion generates Neuropathol 126:829–844. https://doi.org/10.1007/s00401-013-1192-8 insoluble polypeptides specific to c9FTD/ALS. Neuron 77:639–646. https:// 23. Goldman JS, Farmer JM, Wood EM, Johnson JK, Boxer A, Neuhaus J, Lomen- doi.org/10.1016/j.neuron.2013.02.004 Hoerth C, Wilhelmsen KC, Lee VM, Grossman M et al (2005) Comparison of 6. Baker M, Mackenzie IR, Pickering-Brown SM, Gass J, Rademakers R, Lindholm family histories in FTLD subtypes and related tauopathies. Neurology 65: C, Snowden J, Adamson J, Sadovnick AD, Rollinson S et al (2006) Mutations 1817–1819. https://doi.org/10.1212/01.wnl.0000187068.92184.63 in progranulin cause tau-negative frontotemporal dementia linked to 24. Holler CJ, Taylor G, Deng Q, Kukar T (2017) Intracellular proteolysis of chromosome 17. Nature 442:916–919. https://doi.org/10.1038/nature05016 Progranulin generates stable, lysosomal Granulins that are Haploinsufficient 7. Brady OA, Zheng Y, Murphy K, Huang M, Hu F (2012) The frontotemporal in patients with frontotemporal dementia caused by GRN mutations. lobar degeneration risk factor, TMEM106B, regulates lysosomal morphology eNeuro 4. https://doi.org/10.1523/ENEURO.0100-17.2017 and function. Hum Mol Genet. https://doi.org/10.1093/hmg/dds475 25. Kim JY, Grunke SD, Levites Y, Golde TE, Jankowsky JL (2014) 8. Burberry A, Suzuki N, Wang JY, Moccia R, Mordes DA, Stewart MH, Suzuki- Intracerebroventricular viral injection of the neonatal mouse brain for Uematsu S, Ghosh S, Singh A, Merkle FT et al (2016) Loss-of-function persistent and widespread neuronal transduction. J Vis Exp 51863. https:// mutations in the C9ORF72 mouse ortholog cause fatal autoimmune disease. doi.org/10.3791/51863 Sci Transl Med 8: 347ra393 Doi https://doi.org/10.1126/scitranslmed.aaf6038 26. Klein ZA, Takahashi H, Ma M, Stagi M, Zhou M, Lam TT, Strittmatter SM 9. Cecconi F, Meyer BI (2000) Gene trap: a way to identify novel genes and (2017) Loss of TMEM106B ameliorates lysosomal and frontotemporal unravel their biological function. FEBS Lett 480:63–71 dementia-related phenotypes in Progranulin-deficient mice. Neuron 95(281– 10. Chakrabarty P, Rosario A, Cruz P, Siemienski Z, Ceballos-Diaz C, Crosby K, 296):e286. https://doi.org/10.1016/j.neuron.2017.06.026 Jansen K, Borchelt DR, Kim JY, Jankowsky JL et al (2013) Capsid serotype and timing of injection determines AAV transduction in the neonatal mice 27. Koppers M, Blokhuis AM, Westeneng HJ, Terpstra ML, Zundel CA, Vieira de brain. PLoS One 8:e67680. https://doi.org/10.1371/journal.pone.0067680 Sa R, Schellevis RD, Waite AJ, Blake DJ, Veldink JH et al (2015) C9orf72 11. Chen-Plotkin AS, Unger TL, Gallagher MD, Bill E, Kwong LK, Volpicelli-Daley ablation in mice does not cause motor neuron degeneration or motor L, Busch JI, Akle S, Grossman M, Van Deerlin V et al (2012) TMEM106B, the deficits. Ann Neurol 78:426–438. https://doi.org/10.1002/ana.24453 risk gene for frontotemporal dementia, is regulated by the microRNA-132/ 28. Lagier-Tourenne C, Baughn M, Rigo F, Sun S, Liu P, Li HR, Jiang J, Watt AT, 212 cluster and affects progranulin pathways. J Neurosci 32:11213–11227. Chun S, Katz M et al (2013) Targeted degradation of sense and antisense https://doi.org/10.1523/JNEUROSCI.0521-12.2012 C9orf72 RNA foci as therapy for ALS and frontotemporal degeneration. Proc Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 13 of 14 Natl Acad Sci U S A 110:E4530–E4539. https://doi.org/10.1073/pnas. 46. Rhinn H, Abeliovich A (2017) Differential aging analysis in human cerebral cortex 1318835110 identifies variants in TMEM106B and GRN that regulate aging phenotypes. Cell 29. Lang CM, Fellerer K, Schwenk BM, Kuhn PH, Kremmer E, Edbauer D, Capell Syst 4(404–415):e405. https://doi.org/10.1016/j.cels.2017.02.009 A, Haass C (2012) Membrane orientation and subcellular localization of 47. Rosso SM, Donker Kaat L, Baks T, Joosse M, de Koning I, Pijnenburg Y, de Jong transmembrane protein 106B (TMEM106B), a major risk factor for D, Dooijes D, Kamphorst W, Ravid R et al (2003) Frontotemporal dementia in the frontotemporal lobar degeneration. J Biol Chem 287:19355–19365. https:// Netherlands: patient characteristics and prevalence estimates from a population- doi.org/10.1074/jbc.M112.365098 based study. Brain 126:2016–2022. https://doi.org/10.1093/brain/awg204 30. Lattante S, Le Ber I, Galimberti D, Serpente M, Rivaud-Pechoux S, Camuzat 48. Rutherford NJ, Carrasquillo MM, Li M, Bisceglio G, Menke J, Josephs KA, A, Clot F, Fenoglio C, Scarpini E, Brice A et al (2014) Defining the association Parisi JE, Petersen RC, Graff-Radford NR, Younkin SG et al (2012) of TMEM106B variants among frontotemporal lobar degeneration patients TMEM106B risk variant is implicated in the pathologic presentation of with GRN mutations and C9orf72 repeat expansions. Neurobiol Aging Alzheimer disease. Neurology 79:717–718. https://doi.org/10.1212/WNL. 35(2658):e2651–e2655. https://doi.org/10.1016/j.neurobiolaging.2014.06.023 0b013e318264e3ac 31. Lee CW, Stankowski JN, Chew J, Cook CN, Lam YW, Almeida S, Carlomagno 49. Schwenk BM, Lang CM, Hogl S, Tahirovic S, Orozco D, Rentzsch K, Y, Lau KF, Prudencio M, Gao FB et al (2017) The lysosomal protein cathepsin Lichtenthaler SF, Hoogenraad CC, Capell A, Haass C et al (2014a) The FTLD L is a progranulin protease. Mol Neurodegener 12:55. https://doi.org/10. risk factor TMEM106B and MAP6 control dendritic trafficking of lysosomes. 1186/s13024-017-0196-6 EMBO J 33:450–467. https://doi.org/10.1002/embj.201385857 32. Levine TP, Daniels RD, Gatta AT, Wong LH, Hayes MJ (2013) The product of 50. Schwenk BM, Lang CM, Hogl S, Tahirovic S, Orozco D, Rentzsch K, C9orf72, a gene strongly implicated in neurodegeneration, is structurally Lichtenthaler SF, Hoogenraad CC, Capell A, Haass C et al (2014b) The FTLD related to DENN Rab-GEFs. Bioinformatics 29:499–503. https://doi.org/10. risk factor TMEM106B and MAP6 control dendritic trafficking of lysosomes. 1093/bioinformatics/bts725 Embo J. https://doi.org/10.1002/embj.201385857 33. Mori K, Arzberger T, Grasser FA, Gijselinck I, May S, Rentzsch K, Weng SM, 51. Sellier C, Campanari ML, Julie Corbier C, Gaucherot A, Kolb-Cheynel I, Oulad- Schludi MH, van der Zee J, Cruts M et al (2013a) Bidirectional transcripts of Abdelghani M, Ruffenach F, Page A, Ciura S, Kabashi E et al (2016) Loss of the expanded C9orf72 hexanucleotide repeat are translated into C9ORF72 impairs autophagy and synergizes with polyQ Ataxin-2 to induce aggregating dipeptide repeat proteins. Acta Neuropathol 126:881–893. motor neuron dysfunction and cell death. EMBO J 35:1276–1297. https:// https://doi.org/10.1007/s00401-013-1189-3 doi.org/10.15252/embj.201593350 34. Mori K, Weng SM, Arzberger T, May S, Rentzsch K, Kremmer E, Schmid B, 52. Shi Y, Lin S, Staats KA, Li Y, Chang WH, Hung ST, Hendricks E, Linares GR, Kretzschmar HA, Cruts M, Van Broeckhoven C et al (2013b) The C9orf72 Wang Y, Son EY et al (2018) Haploinsufficiency leads to neurodegeneration GGGGCC repeat is translated into aggregating dipeptide-repeat proteins in in C9ORF72 ALS/FTD human induced motor neurons. Nat Med 24:313–325. FTLD/ALS. Science 339:1335–1338. https://doi.org/10.1126/science.1232927 https://doi.org/10.1038/nm.4490 35. Murray ME, Cannon A, Graff-Radford NR, Liesinger AM, Rutherford NJ, Ross 53. Stagi M, Klein ZA, Gould TJ, Bewersdorf J, Strittmatter SM (2014) Lysosome OA, Duara R, Carrasquillo MM, Rademakers R, Dickson DW (2014) Differential size, motility and stress response regulated by fronto-temporal dementia clinicopathologic and genetic features of late-onset amnestic dementias. modifier TMEM106B. Mol Cell Neurosci 61:226–240. https://doi.org/10.1016/j. Acta Neuropathol 128:411–421. https://doi.org/10.1007/s00401-014-1302-2 mcn.2014.07.006 36. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT, 54. Su Z, Zhang Y, Gendron TF, Bauer PO, Chew J, Yang WY, Fostvedt E, Jansen- Bruce J, Schuck T, Grossman M, Clark CM et al (2006) Ubiquitinated TDP-43 West K, Belzil VV, Desaro P et al (2014) Discovery of a biomarker and lead in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. small molecules to target r(GGGGCC)-associated defects in c9FTD/ALS. Science 314:130–133 Neuron 83:1043–1050. https://doi.org/10.1016/j.neuron.2014.07.041 37. Nicholson AM, Finch NA, Wojtas A, Baker MC, Perkerson RB 3rd, 55. Sullivan PM, Zhou X, Robins AM, Paushter DH, Kim D, Smolka MB, Hu F Castanedes-Casey M, Rousseau L, Benussi L, Binetti G, Ghidoni R et al (2016) The ALS/FTLD associated protein C9orf72 associates with SMCR8 and (2013) TMEM106B p.T185S regulates TMEM106B protein levels: WDR41 to regulate the autophagy-lysosome pathway. Acta Neuropathol implications for frontotemporal dementia. J Neurochem 126:781–791. Commun 4:51. https://doi.org/10.1186/s40478-016-0324-5 https://doi.org/10.1111/jnc.12329 56. Taylor JP, Brown RH Jr, Cleveland DW (2016) Decoding ALS: from genes to 38. Nicholson AM, Rademakers R (2016) What we know about TMEM106B in mechanism. Nature 539:197–206. https://doi.org/10.1038/nature20413 neurodegeneration. Acta Neuropathol 132:639–651. https://doi.org/10.1007/ 57. Ugolino J, Ji YJ, Conchina K, Chu J, Nirujogi RS, Pandey A, Brady NR, s00401-016-1610-9 Hamacher-Brady A, Wang J (2016) Loss of C9orf72 enhances Autophagic 39. O'Rourke JG, Bogdanik L, Muhammad A, Gendron TF, Kim KJ, Austin A, Cady activity via deregulated mTOR and TFEB signaling. PLoS Genet 12:e1006443. J, Liu EY, Zarrow J, Grant S et al (2015) C9orf72 BAC transgenic mice display https://doi.org/10.1371/journal.pgen.1006443 typical pathologic features of ALS/FTD. Neuron 88:892–901. https://doi.org/ 58. van Blitterswijk M, Mullen B, Nicholson AM, Bieniek KF, Heckman MG, Baker 10.1016/j.neuron.2015.10.027 MC, Dejesus-Hernandez M, Finch NA, Brown PH, Murray ME et al (2014) 40. O'Rourke JG, Bogdanik L, Yanez A, Lall D, Wolf AJ, Muhammad AK, Ho R, TMEM106B protects C9ORF72 expansion carriers against frontotemporal Carmona S, Vit JP, Zarrow J et al (2016) C9orf72 is required for proper dementia. Acta Neuropathol. https://doi.org/10.1007/s00401-013-1240-4 macrophage and microglial function in mice. Science 351:1324–1329. 59. Van Deerlin VM, Sleiman PM, Martinez-Lage M, Chen-Plotkin A, Wang LS, https://doi.org/10.1126/science.aaf1064 Graff-Radford NR, Dickson DW, Rademakers R, Boeve BF, Grossman M et al 41. Peters OM, Cabrera GT, Tran H, Gendron TF, McKeon JE, Metterville J, Weiss (2010) Common variants at 7p21 are associated with frontotemporal lobar A, Wightman N, Salameh J, Kim J et al (2015) Human C9ORF72 degeneration with TDP-43 inclusions. Nat Genet 42:234–239. https://doi.org/ Hexanucleotide expansion reproduces RNA foci and dipeptide repeat 10.1038/ng.536 proteins but not neurodegeneration in BAC transgenic mice. Neuron 88: 60. van der Zee J, Van Langenhove T, Kleinberger G, Sleegers K, Engelborghs S, 902–909. https://doi.org/10.1016/j.neuron.2015.11.018 Vandenberghe R, Santens P, Van den Broeck M, Joris G, Brys J et al (2011) 42. Platt FM, Boland B, van der Spoel AC (2012) The cell biology of disease: TMEM106B is associated with frontotemporal lobar degeneration in a lysosomal storage disorders: the cellular impact of lysosomal dysfunction. J clinically diagnosed patient cohort. Brain 134:808–815. https://doi.org/10. Cell Biol 199:723–734. https://doi.org/10.1083/jcb.201208152 1093/brain/awr007 43. Rascovsky K, Hodges JR, Knopman D, Mendez MF, Kramer JH, Neuhaus J, 61. Wood EM, Falcone D, Suh E, Irwin DJ, Chen-Plotkin AS, Lee EB, Xie SX, Van van Swieten JC, Seelaar H, Dopper EG, Onyike CU et al (2011) Sensitivity of Deerlin VM, Grossman M (2013) Development and validation of pedigree revised diagnostic criteria for the behavioural variant of frontotemporal classification criteria for frontotemporal lobar degeneration. JAMA Neurol dementia. Brain 134:2456–2477. https://doi.org/10.1093/brain/awr179 70:1411–1417. https://doi.org/10.1001/jamaneurol.2013.3956 44. Ratnavalli E, Brayne C, Dawson K, Hodges JR (2002) The prevalence of 62. Yang M, Liang C, Swaminathan K, Herrlinger S, Lai F, Shiekhattar R, frontotemporal dementia. Neurology 58:1615–1621 Chen JF (2016) A C9ORF72/SMCR8-containing complex regulates ULK1 45. Renton AE, Majounie E, Waite A, Simon-Sanchez J, Rollinson S, Gibbs JR, and plays a dual role in autophagy. Sci Adv 2:e1601167. https://doi.org/ Schymick JC, Laaksovirta H, van Swieten JC, Myllykangas L et al (2011) A 10.1126/sciadv.1601167 hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21- 63. Zhang D, Iyer LM, He F, Aravind L (2012) Discovery of novel DENN proteins: linked ALS-FTD. Neuron 72:257–268. https://doi.org/10.1016/j.neuron.2011.09.010 implications for the evolution of eukaryotic intracellular membrane Nicholson et al. Acta Neuropathologica Communications (2018) 6:42 Page 14 of 14 structures and human disease. Front Genet 3:283. https://doi.org/10.3389/ fgene.2012.00283 64. Zhou X, Paushter DH, Feng T, Pardon CM, Mendoza CS, Hu F (2017a) Regulation of cathepsin D activity by the FTLD protein progranulin. Acta Neuropathol 134:151–153. https://doi.org/10.1007/s00401-017-1719-5 65. Zhou X, Paushter DH, Feng T, Sun L, Reinheckel T, Hu F (2017b) Lysosomal processing of progranulin. Mol Neurodegener 12:62. https://doi.org/10.1186/ s13024-017-0205-9

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: May 31, 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