|Home | About | Journals | Submit | Contact Us | Français|
TAR DNA-binding protein-43 (TDP-43) is a highly conserved, ubiquitously expressed nuclear protein that was recently identified as the disease protein in frontotemporal lobar degeneration with ubiquitin positive inclusions (FTLD-U) and amyotrophic lateral sclerosis (ALS). Pathogenic TDP-43 gene (TARDBP) mutations have been identified in familial ALS kindreds, and here we report a TARDBP variant (A90V) in a FTLD/ALS patient with a family history of dementia. Significantly, A90V is located between the bipartite nuclear localization signal sequence of TDP-43 and the in vitro expression of TDP-43-A90V led to its sequestration with endogenous TDP-43 as insoluble cytoplasmic aggregates. Thus, A90V may be a genetic risk factor for FTLD/ALS because it predisposes nuclear TDP-43 to redistribute to the cytoplasm and form pathological aggregates.
Frontotemporal lobar degeneration (FTLD) refers to a clinically, genetically and pathologically heterogeneous group of neurodegenerative diseases that account for up to 20% of presenile dementia cases. Clinically, FTLD is characterized by behavioral and/or language dysfunction [1–3], but it also can co-occur with movement abnormalities such as parkinsonism and motor neuron disease, including the most common type, amyotrophic lateral sclerosis (ALS) [4,5]. Although neurodegenerative tauopathies account for many familial and sporadic cases of FTLD, those cases with ubiquitin-positive tau- and α-synuclein-negative inclusions (UBIs) make up the most common neuropathological subtype of FTLD, i.e. FTLD-U [6,7]. Recently, the transactive response (TAR) DNA binding protein 43 (TDP-43) was identified as the major disease protein in UBIs that accumulate in the central nervous system (CNS) of patients with FTLD-U as well as in patients with sporadic and familial ALS , but not in the majority of patients with familial ALS (FALS) due to SOD-1 gene mutations [8–12]. These data provided compelling evidence that FTLD-U and ALS represent a clinicopathological spectrum of the same neurodegenerative disorder, i.e. TDP-43 proteinopathy, and this view is supported by the recent detection of several pathogenic TARDBP mutations in a number of FALS kindreds [13–16]. TDP-43, encoded by the TARDBP gene on chromosome 1, is a highly conserved, ubiquitously expressed nuclear protein implicated in repression of gene transcription, inhibition of exon splicing and interactions with splicing factors and nuclear bodies [17,18]. Under physiological conditions TDP-43 is predominately localized to the nucleus; however, pathological TDP-43 forms inclusions in neuronal perikarya and neurites, suggesting that the redistribution and sequestration of TDP-43 in the cytoplasm is a pathogenic mechanism . Moreover, we have identified a specific bipartite nuclear localization signal (NLS) sequence in the amino terminal domain of TDP-43 that is required for nuclear targeting . Over-expression of TDP-43 with mutated NLS sequence (ΔNLS mutants) not only re-directed TDP-43 to the cytoplasm, but also reduced its solubility, resulting in the accumulation of cytoplasmic aggregates, high Mr smears and C-terminal fragments of TDP-43 similar to those seen in FLTD-U/ALS cases . Thus, these data imply that perturbation of nuclear and cytoplasmic trafficking of TDP-43 leads to the formation of cytoplasmic aggregates with morphological similarities to authentic TDP-43 pathology in FTLD-U/ALS . Therefore, to demonstrate physiological relevance of increased cytoplasmic TDP-43 as a potential pathogenic mechanism, we investigated whether genetic variant(s) with defective TARDBP NLS sequences are found in FTLD patients with or without ALS and if such variant(s) perturbed TDP-43 distribution in the nucleus and cytoplasm.
Genomic DNA was extracted from blood of living patients or from postmortem brains using standard methods (Qiagen Inc., Valencia, CA). The TARDBP gene was screened for mutations in a cohort of patients from University of California at San Francisco (UCSF). This included 134 patients with clinical FTLD or FTLD/ALS, as well as in autopsy cases with confirmed TDP-43 pathology and neuropathological diagnoses of ALS (n=2) and FTLD-U or FTLD plus MND (n=12). Autopsy cases were conducted at the Center for Neurodegenerative Disease Research at the University of Pennsylvania (UPenn). Cases with mutations in the progranulin gene were excluded. The coding region of TARDBP, consisting of exons 2–5 and the first 531 nucleotides (nt) of exon 6, as well as at least 100 nt of the flanking intronic regions of each exon were fully sequenced in both directions. Primers used to amplify part or all of each exon were selected using Primer3 software and are available upon request.
Amplification reactions (50 μl) were performed with 200 ng DNA using AmpliTaq Gold DNA polymerase (Applied Biosystems, Foster City, CA) and 0.8 μM (final concentration) each primer. Touchdown amplification protocol used consisted of 95°C (10 min), followed by 14 cycles of 95°C (30 s), 60°C (exons 4–6) or 58°C (exons 2, 3) with a reduction of 0.5°C per cycle (30 s), 72°C (1 min), and 20 cycles of 95°C (30 s), 53°C (30 s), 72°C (1 min), with a final 72°C extension (5 min). Amplification products were purified using AMPure (Agencourt Bioscience, Beverly, MA) followed by single pass bidirectional sequencing performed by Agencourt Bioscience or CEQ8000 (Beckman Coulter, Fullerton, CA) with Beckman Coulter sequencing reagents. Results were analyzed using Mutation Surveyor software (SoftGenetics, PA).
Genotyping of control samples for TARDBP c.269C>T (p.A90V) was performed using a TaqMan chemistry-based allelic discrimination assay with “Assay by Design” (Applied Biosystems, Foster City, CA) probes on an Applied Biosystems 7900 followed by analysis with Sequence Detection System 2.2.1 software (Applied Biosystems). A total of 1385 control samples were obtained from the following sources: 276 controls from the Coriell Institute (Neurologically Normal Caucasian control panels, Camden, NJ; mean age 70 y), 48 clinical controls (mean age 76 y) from the Alzheimer Disease Core Center at UPenn, 42 normal postmortem brain samples (mean age 69 y) from the UPenn brain bank, 638 normal controls (mean age 63 y) from the Mayo Clinic Scottsdale and 381 normal controls (mean age 76 y) from the University of Washington. The DNA sample with the variant identified by DNA sequencing served as a positive control for each assay.
cDNA encoding human TDP-43 (Accession No. NM 007375) in the plasmid pENTR-221 was obtained from Invitrogen (Carlsbad, CA). The addition of a myc epitope tag to the 5′-end of TDP-43 was achieved by PCR, using the primers; 5′-AAGCTT GATGGAACAAAAACTCATCTCGGAAGAGGATCTGTCTGAATATATTCGGGTA ACC-3′ and 5′-TCTAGAGCTACATTCCCCAGCCAGAAGACTTAGA-3′. The PCR product was cloned into the pGEM-T vector (Promega, Madison, WI). Following sequence analysis the PCR product was then subcloned into the plasmid pcDNA 3.1 (Invitrogen) using restriction sites HindIII and XbaI, creating pcDNA 3.1/myc-TDP-43-WT. Site-directed mutations (see below) were generated using the plasmid pcDNA 3.1/myc-TDP-43-WT as a template.
Site-directed mutagenesis (Quikchange kit; Strategene, La Jolla, CA) was used to create ΔNLS1  and A90V missense mutation in the longest human TDP-43 isoform. Sequence of the mutagenized oligonucleotides are as follows: A90V: 5′-G GAT GAG ACA GAT GTT TCA TCA GCA GTG AAA GTG-3′. All constructs were subjected to sequence analysis.
QBI-293 cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin and 1% L-glutamate. QBI-293 cells were transfected using the Amaxa Nucleofector system (Amaxa Inc., Gaithersburg, MD) or Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions.
Antibodies used here included: rabbit anti-TDP-43 antibody raised to recombinant TDP-43 (Protein Tech Group, Chicago, IL); anti-myc mAb (9E10, Santa Cruz Biotechnology, Santa Cruz, CA) and an anti-α-tubulin mAb (Sigma, St. Louis, MO).
Cells were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS), permeabilized with 0.02% or 0.2% Triton X-100 (Sigma) in PBS for 10 min, blocked with 5% powdered-milk in PBS for 2-hrs and incubated overnight with primary antibody at 4ºC. Primary antibodies were visualized with secondary antibodies conjugated with either Alexa Fluor 488 or Alexa Fluro 594 (Vector Laboratories, Burlingame, CA) and nuclei were detected using DAPI. All cells were analyzed using a Nikon TE-2000-E (Nikon, Tokyo, Japan) and images were captured using a CoolSnap-HQ camera (Photometrics, Tuscon, AZ). All micrographs show individual cells representative of the total cell population. For all experiments, at least four separate experiments were performed in duplicate.
To examine the solubility profile of TDP-43, sequential extractions were performed. Cells were washed twice with PBS, lysed in cold RIPA buffer, and sonicated. Cell lysates were cleared by centrifugation at 40,000 rpm for 30 min at 4ºC to generate the RIPA soluble samples. To prevent carry over the resulting pellets were washed twice (i.e. re-sonicated and re-centrifuged). Only the supernatants from the first centrifugation were analyzed and all supernatants produced from the second and third wash step were discarded. RIPA insoluble pellets were then extracted with urea buffer (7M urea, 2M thiourea, 4% 3-[(3-Cholamidopropyl) dimethylammonio]-1-propanesulfonate (CHAPS), 30mM Tris, pH 8.5), sonicated and centrifuged at 40,000 rpm for 30 min at 22ºC. Protease inhibitors were added to all buffers prior to use (1 mM PMSF and a cocktail of protease inhibitors). Protein concentration was determined by bicinchoninic acid method (Pierce, Rockford, IL) and all proteins were resolved by 10% SDS-PAGE and transferred to nitrocellulose membranes. Following transfer, nitrocellulose membranes were blocked in 5% powdered milk and incubated in the primary antibody overnight at 4ºC. Primary antibodies were detected with horseradish peroxidase-conjugated anti-IgG antibody (Jackson ImmunoResearch, Wegate, PA) and blots were developed with Renaissance Enhanced Luminal Reagents (NEN Life Science Products, Boston, MA). Digital images were acquired using a Fuji Film Intelligent Darkbox II (Fuji Systems, Stamford, CT).
Genetic evaluation of TARDBP was completed on 148 patients with a clinical and/or autopsy proven diagnosis of FTLD or FTLD/ALS and identified a single living patient with a c.269C>T (A90V) substitution in exon 3. This 72 year-old Caucasian with a family history of dementia (Supplementary Fig. 1) presented clinically with symptoms consisting of slowly progressive FTLD/ALS (age of onset 57 y). Additional family members (deceased, affected or living unaffected) were unavailable for testing and no autopsies have been performed in the proband’s family to study the neuropathological changes linked to the A90V TARDBP variant. Therefore, association of this variant with disease in the family could not be confirmed at this time. Testing of 1385 age-matched controls for the c.269C>T substitution by SNP genotyping identified the variant in a single control case (1 out of 1385 cases; 0.07%), 76 years old at inclusion of the study indicating that this amino acid substitution may be a polymorphism in the TARDBP gene. However, based on the functional data presented below we speculate that A90V may increase the risk of FTLD/ALS.
As the A90V variant is located between the bipartite NLS sequence of TDP-43, we examined its functional consequences in vitro. Myc-TDP-43-A90V was transiently expressed in QBI-293 cells and two-color immunofluorescence was performed on cells 72-hrs post-transfection. Both myc-TDP-43-WT and myc-TDP-43-ΔNLS1 were used as controls and as reported previously , myc-TDP-43-WT was localized exclusively to the nucleus (Fig. 1a-1c) and myc-TDP-43-ΔNLS1 was detected primarily in the cytoplasm (Fig. 1d–f). However, myc-TDP-43-A90V was detected both in the nucleus and in the cytoplasm of a subset of cells, suggesting only a partial disruption of TDP-43 nuclear localization by the alanine to valine substitution (Fig. 1g-i). Quantitative analysis of transfected cells showed that 78 % and 22 % of myc-TDP-43-A90V expressing cells showed nuclear and cytoplasmic localization of TDP-43, respectively (Fig. 1j). Further, since the A90V variant is located in between the bipartite NLS sequence, and not directly in the functional region, we hypothesize that this variant may only cause a mild disruption in the nuclear localization of TDP-43 and as a result the mislocalization of TDP-43 only occurs in a sub-set of transfected cells.
To determine if myc-TDP-43-A90V accumulated as an insoluble pool and sequestered endogenous TDP-43, RIPA and urea fractions of transfected cells were analyzed by immunoblot. Since myc-tagged TDP-43 protein migrated slower than endogenous TDP-43 in SDS-PAGE, each could be analyzed separately. At all time points examined, both endogenous and myc-TDP-43-WT were recovered exclusively in the RIPA fractions (Fig. 2a and 2b). Consistent with the expected ΔNLS sequence disruption described previously , myc-TDP-43-A90V was detected in the urea fraction as well as in the RIPA fraction 24 hrs post-transfection and by 72 hrs both endogenous and the myc-TDP-43-A90V proteins were recovered in the insoluble urea fraction (compare Fig. 2a with 2b). Thus, the myc-TDP-43-A90V genetic variant showed partial NLS sequence disruption and caused a similar but milder phenotype like that seen in the myc-TDP-43-ΔNLS mutants  including sequestration of endogenous TDP-43 by myc-TDP-43-A90V in the urea fraction (Fig. 2a and 2b).
The identification of a human genetic variant in the region between the bipartite NLS sequence, in concert with results from in vitro cell biology experiments gives credence to the notion that functional disruption of the highly conserved NLS sequence could increase the risk for FTLD/ALS. Unfortunately, no biological samples were available from other members of the proband’s family for genetic analysis; however, there is a very strong documented family history of dementia. Furthermore, the clinical presentation of both FTLD and ALS in the proband is consistent with the current hypothesis that FTLD and ALS represent two extremes of a clinicopathological spectrum of disease, TDP-43 proteinopathies .
We have reported previously that over-expression of TDP-43 ΔNLS mutants in vitro recapitulated the biochemical phenotype found in human neurodegenerative TDP-43 proteinopathies . In concert with this finding a similar, but milder, phenotype was also observed with A90V over-expression. However, as our manuscript was in preparation, the A90V TDP-43 variant was identified in a 53 year-old control (1/372 cases screened)  as well as an unspecified control (1/360) . Thus, as of today, this variant has been found in 1 affected and 3 control subjects. Therefore we cannot exclude the possibilities that A90V represents a rare benign polymorphism. However, our in vitro data and the strong family history of dementia suggest that the A90V variant maybe a potential genetic risk factor for FTLD. Similar to other known disease risk factor genes, characteristics and lifestyle of the person, and environmental and additional genetic factors can contribute to the development of the disease. Reduced penetrance and variation in expressivity may also play a role in the disease-associated risk of the A90V variant. This is consistent with our in vitro data as the mislocalization of TDP-43 was observed in only a subset of transfected cells. Thus, the presence of the A90V variant may increase the likelihood of developing FTLD/ALS
The recent discovery of multiple pathogenic TARDBP mutations in several FALS kindreds [13–16] further supports the importance of TDP-43 in the etiology and pathologenesis of ALS; however, unlike the A90V TDP-43 substitution, these recently identified TARDBP mutations are localized to the C-terminus of TDP-43 and thus, are likely to cause the disease through different mechanisms. As the current literature suggests that the presence of disease segregating mutations in TDP-43 are rare, and to date, minimal neuropathological data from affected family members has been published. Therefore, further genetic analysis needs to be completed in more familial cases of FTLD/ALS before we can conclusively comment on the genetic and mechanistic consequences of TDP-43 mutations, as well as the TDP-43 A90V variant.
Furthermore, although TDP-43 inclusions are found in the nucleus and cytoplasm of diseased cells in association with nuclear clearing of TDP-43, it remains uncertain how TARDBP mutations or TDP-43 pathology causes neurodegeneration in FTLD-U and ALS. Rapid progress is being made regarding the normal and pathological biology of TDP-43 and this will accelerate efforts to develop new and more effective therapies for neurodegenerative TDP-43 proteinopathies.
Supplementary Figure 1. Pedigree of the family in which the A90V variant was identified. Additional affected individuals were not available to determine segregation of the mutation with disease. NOS: not otherwise specified, i.e. dementia of uncertain type. Squares have been used to identify male individuals, circles have been used to identify female individuals and diamonds have been used to conceal gender for privacy. The proband is III: 4 (indicated with the arrowhead).
This work was funded by grants from the National Institute of Health/National Institute on Aging (AG10124, AG17586, AG05142, AG10129, AG16574, MH57899) and the Pacific Alzheimer Research Foundation. VM-YL is the John H. Ware III Professor of Alzheimer’s Research and JQT is the William Maul Measey-Truman G. Schnabel, Jr., M.D. Professor of Geriatric Medicine and Gerontology. The authors would like to thank the families of patients who made this research possible.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
This PDF receipt will only be used as the basis for generating PubMed Central (PMC) documents. PMC documents will be made available for review after conversion (approx. 2–3 weeks time). Any corrections that need to be made will be done at that time. No materials will be released to PMC without the approval of an author. Only the PMC documents will appear on PubMed Central -- this PDF Receipt will not appear on PubMed Central.