UBC Faculty Research and Publications

Reducing TDP-43 aggregation does not prevent its cytotoxicity Liu, Rui; Yang, Guang; Nonaka, Takashi; Arai, Tetsuaki; Jia, William; Cynader, Max S Aug 9, 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-40478_2013_Article_65.pdf [ 985.1kB ]
JSON: 52383-1.0221602.json
JSON-LD: 52383-1.0221602-ld.json
RDF/XML (Pretty): 52383-1.0221602-rdf.xml
RDF/JSON: 52383-1.0221602-rdf.json
Turtle: 52383-1.0221602-turtle.txt
N-Triples: 52383-1.0221602-rdf-ntriples.txt
Original Record: 52383-1.0221602-source.json
Full Text

Full Text

RESEARCH Open AccessReducing TDP-43 aggregation does not preventits cytotoxicityRui Liu1, Guang Yang1, Takashi Nonaka2, Tetsuaki Arai3, William Jia1 and Max S Cynader1*AbstractBackground: TAR DNA-binding protein 43 (TDP-43) is a protein that is involved in the pathology of AmyotrophicLateral Sclerosis (ALS) and Frontotemporal Lobar Degeneration (FTLD). In patients with these neurodegenerativediseases, TDP-43 does not remain in its normal nuclear location, but instead forms insoluble aggregates in both thenucleus and cytoplasm of affected neurons.Results: We used high density peptide array analysis to identify regions in TDP-43 that are bound by TDP-43 itselfand designed candidate peptides that might be able to reduce TDP-43 aggregation. We found that two of thesynthetic peptides identified with this approach could effectively inhibit the formation of TDP-43 proteinaggregates in a concentration-dependent manner in HeLa cells in which a mutated human TDP-43 gene wasoverexpressed. However, despite reducing aggregation, these peptides did not reduce or prevent cell death. Similarresults were observed in HeLa cells treated with arsenite. Again we found reduced aggregation, in this case of wildtype TDP-43, but no difference in cell death.Conclusions: Our results suggest that TDP-43 aggregation is associated with the cell death process rather thanbeing a direct cause.Keywords: TDP-43, Aggregation, Peptides, Cell deathBackgroundRecent evidence links TDP-43 pathology to at least twoforms of neurodegeneration that had heretofore beenthought to be quite separate. FTLD is the second mostcommon type of early-onset neurodegenerative dementiaafter Alzheimer’s disease, and ALS is the most commonadult-onset progressive motor neuron disease (MND).The TAR DNA binding protein 43 (TDP-43) has beenfound to be the major protein constituent of the intracel-lular aggregated inclusions in both FTLD with ubiquitin-positive inclusions (FTLD-U) and ALS [1,2].TDP-43 is a 414 amino acid protein encoded by theTARDBP gene on chromosome 1. It was originally iden-tified as a transcriptional repressor of the human im-munodeficiency virus type 1 (HIV-1) gene [3,4] and themammalian gene SP-10 [5]. TDP-43 normally is foundin the nucleus where it regulates RNA splicing, mRNAstability and microRNA processing [6-9], but TDP-43 inpathological inclusions is generally hyperphosphorylated,ubiquitinated [10], and cleaved to 35 and 25 kDa species[11]. The pathological aggregates are frequently found inthe cytoplasm rather than at TDP-43’s normal nuclearlocation [1,2,12].The mechanism through which TDP-43 is involved inneuronal death and degeneration remains unclear. Oneof the best characterized pathological features of TDP-43 proteinopathies is the cytoplasmic inclusions of TDP-43 aggregates. As with other protein misfolding diseases,TDP-43 mediated toxicity may result from a toxic gainof function associated with its aggregation. Johnson andcolleagues have established a yeast model involving over-expressed full-length human TDP-43 or various TDP-43truncation products [13]. They found that express-ing a truncated form of TDP-43 containing both theC-terminal and RRM2 promoted aggregation. Only theaggregated form of TDP-43 induced toxicity to yeastcells. This suggested that TDP-43 misfolding and aggre-gation might be an important cause of cell death in neu-rodegenerative diseases. Another group supported thisconclusion in human cell models [14]. The 25 kDa* Correspondence: cynader@brain.ubc.ca1Brain Research Center, University of British Columbia, 2211 Wesbrook Mall,Vancouver, BC V6T2B5, CanadaFull list of author information is available at the end of the article© 2013 Liu et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.Liu et al. Acta Neuropathologica Communications 2013, 1:49http://www.actaneurocomms.org/content/1/1/49C-terminal fragment of TDP-43, which is likely to be acaspase-3 cleavage product [15], was overexpressed inHEK293 and differentiated M17 neuroblastoma cells. The25 kDa fragment of TDP-43 formed cytoplasmic inclu-sions, induced cell toxicity, but did not disturb endo-genous TDP-43 functions. Moreover, it was found thataggregation of mitochondria seems to be a common fea-ture when overexpressing TDP-43 in transgenic mice[16,17]. All these data point to the toxicity of TDP-43 ag-gregates, or of the processes by which they form.To further elucidate the links between TDP-43 aggre-gation and disease mechanisms, we used two differentmodels. The first was an aggregation model pioneeredby Nonaka and colleagues [18]. They found that a mu-tant TDP-43 lacking residues 187–192 formed intra-nuclear dot-like inclusions when expressed in SY5Ycells. A second model relied in recent findings that ar-senite treatment can induce aggregation of wild-typeTDP-43. When HeLa cells are treated with arsenite, anestablished stress granule (SG) inducer [19,20], TDP-43aggregation occurs in the cytoplasm. Arsenite reactswith oxygen, induces oxidative stress and then activatesHRI (Heme-regulated inhibitor), whose expression is re-quired for SG formation [21]. SGs are cytoplasmic sites ofstalled mRNA pre-initiation complexes induced by mul-tiple stressors. Recent data have shown that SGs partici-pate in the process of TDP-43 accumulation [22,23]. Withthese two models in hand, we asked whether selectivelyblocking aggregation would affect cellular toxicity. Inhib-ition of TDP-43 aggregation is considered to be a majorpotential therapeutic avenue for ALS and FTLD-U. Simi-lar to the situation for other neurodegenerative diseases,potential tools include antibodies, molecular chaperones,chemical compounds and synthetic peptides. The strategywe used to design aggregation blockers was to identifyand synthesize TDP-43 fragments that bind with fulllength TDP-43 protein. Our findings indicate that the syn-thetic peptides reduced the formation of TDP-43 aggre-gates in both models used, but did not reduce or preventcell death.ResultsTDP-43 protein binds with selected TDP-43 derivedpeptides on the membraneThe protein array described in the Methods section wasused to identify candidate binding regions at which TDP-43 might interact with itself. Given the nature of theoverlapping sequences associated with each neighboringpeptide (frame shift of 2 aa), only positive spots in a serieswere considered as possible binding sites. Five separateand robust binding regions were found on the TDP-43peptide array after incubating the membrane with recom-binant TDP-43 protein followed by visualization of anantibody against TDP-43 (Figure 1). This result suggeststhat the five regions may potentially be involved in selfbinding or interaction of TDP-43. Based on these pos-sible interaction domains, five peptide candidates, calledPeptide A-E, were designed and synthesized. The pep-tides were derived from the sequence of full lengthTDP-43. The overlapping amino acids are the possiblebinding domains and are indicated in Figure 1. The dis-tribution of five peptides is shown on the schematic ofTDP-43 structure.The synthetic peptides were able to block the interactionbetween recombinant TDP-43 protein and its peptidearray membraneTo verify whether the five regions of TDP-43 identifiedthrough the array assay are actually involved in selfbinding of the protein, the five peptides indicated inFigure 1 were synthesized and tested for their ability toinhibit the binding between full length TDP-43 and theTDP-43 peptide array (Figure 2). A scrambled peptideTC was designed as a control. Peptide TC has the samelength, net charge, and number of hydrophilic andhydrophobic amino acids as Peptide C. It has a randomorder and substitutions with the structure disruptingamino acids of original peptide. The results showed thata mixture of the five peptides blocked the interactionbetween TDP-43 and its peptide array (decrease by76.05% ± 3.46% compared to control) (Figure 2A, B).However, the control peptide TC was not able to blockthe interaction (Figure 2C).Synthetic peptides inhibit aggregation of overexpressedmutated TDP-43 in HeLa cellsTo examine whether the peptides blocking interaction ofrecombinant TDP-43 protein with the peptide array couldalso inhibit TDP-43 aggregation in cells, a construct ex-pressing a mutant TDP-43 lacking residues 187–192 [18]was transfected into HeLa cells. A construct expressingwild-type TDP-43 was used as a control. Immunohisto-chemical analysis was performed 48 hours after trans-fection. As shown in Figure 3, while all cells expresswild-type TDP-43, only cells transfected with mutantTDP-43 showed TDP-43 aggregates (18.40% ± 0.62%),demonstrated by punctate staining in the nuclei.To investigate whether the synthetic peptides could in-hibit the aggregation of mutant TDP-43, N-terminallyTAT conjugated peptides A, B, C, D, E and/or TC wereadded to the cell culture medium after transfection. Thepeptides were added either individually or combined to-gether in various ways (peptide A-E mixture, or peptideB and C mixture). The TAT peptide has been used todeliver large molecules and small particles across boththe plasma membrane and the nuclear membrane [24].The number of mutant TDP-43 inclusions decreased incells transduced with the peptides (Figures 3 and 4).Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 2 of 11http://www.actaneurocomms.org/content/1/1/49Peptides B and C were found to be much more effect-ive at reducing aggregation than the other syntheticpeptides derived from TDP-43 or than the control TCpeptide. Compared to untreated cells, cells with TDP-43 aggregates were reduced by 52% ± 8.85% (p < 0.01),72% ± 2.33% (p < 0.01), and 91% ± 3.56% (p < 0.01) aftertreatment with Peptide B (20 μM), Peptide C (20 μM)and a Peptide B and C mixture (20 μM), respectively(Figure 4A). In addition, the peptides blocked aggrega-tion in a concentration-dependent manner (Figure 4B).Peptides A, D, E or the scrambled control TC did notshow statistically significant effects (Figure 4A). Pep-tides A-E applied to the untransfected cells had no ef-fect on cell viability.Synthetic peptides inhibit aggregation of endogenousTDP-43 induced by arsenite stress in HeLa cellsTo investigate whether our peptides could inhibit aggre-gation of wild-type TDP-43 under stress conditions,HeLa cells were treated with arsenite with or withoutaddition of synthetic peptides. After exposure to 0.5 μMarsenite with or without 10 μM of the synthetic peptidemixture (ABCDE) for 4 hours, the cell lysates weresequentially extracted and separated into soluble andinsoluble proteins. The relative ratio of insoluble tosoluble TDP-43 was used as a measure of TDP-43 aggre-gation propensity. After the treatment with arsenite,soluble endogenous TDP-43 was transformed into pre-dominantly insoluble protein and showed a significantincreased aggregation propensity (243.23 ± 22.97%) com-pared with control cells without treatment (41.02 ±24.02%) (Figure 5). When the synthetic peptides (PeptideA-E mixture) were coincubated along with the arsenite,the ratio of insoluble to soluble TDP-43 decreased to109.13 ± 25.00%. Treatment with the scrambled peptidedid not lead to any reduction compared to non-treatedarsenite-stressed cells (221.57% ± 46.76%) (Figure 5).These results show that our synthetic peptides can re-duce the formation of insoluble wild-type TDP-43 understress conditions.Figure 1 Identification of peptide candidates that may block TDP-43 binding to itself using the high density peptide array. TDP-43protein associates with TDP-43 membrane at specific regions. Control, binding of the antibody only to the TDP-43 membrane; TDP-43, TheTDP-43 membrane was incubated with TDP-43 protein (three repeats). The schematic representation shows the five peptide candidates labeledon the membranes. The peptides were designed based on the identified regions of the membrane to which purified TDP-43 bound.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 3 of 11http://www.actaneurocomms.org/content/1/1/49The inhibition of aggregation cannot prevent cell deathTo ask whether TDP-43 aggregation is the cause of celldeath, we used MTT and TUNEL assays to examinecell viability in cells expressing the mutant TDP-43.Overexpression of the mutant TDP-43 did indeed inducesignificant cell death (64.60 ± 5.44% cell viability, p <0.01) compared to untreated cells. Treatment with Pep-tide B and C, which significantly reduced the aggrega-tion, led to similar levels of cell viability (71.37% ± 3.30%,p < 0.01) as was found in untreated and scrambled pep-tide treated cells (68.90% ± 2.96%, p < 0.01) (Figure 6).Similar results were observed in the cells in whichwild-type TDP-43 aggregation was reduced after treat-ment with arsenite and synthetic peptides. Treatmentwith arsenite induced significant HeLa cell death(53.32 ± 8.13% cell viability, p < 0.01) compared to thebasal condition. Our synthetic peptides decreased in-soluble TDP-43 but did not prevent cell death (55.10 ±9.38% cell viability, p < 0.01) (Figure 7). These resultsindicate that TDP-43 aggregation was associated withcell death, but inhibition of TDP-43 aggregation couldnot rescue or prevent cell death.To further elucidate the relationship between the celldeath and TDP-43 aggregation, the cells were triple-labeled with TDP-43, GFP and the cell death marker,TUNEL. The numbers of transfected cells, aggregation-positive cells, or TUNEL-positive cells were counted bya blinded independent investigator. We measured fourfields and 1000–1200 cells per microscope slide. Wefound that almost all the cells with TDP-43 aggregationthat were untreated, Peptide B and C treated, or scram-bled peptide (TC) treated showed TUNEL positivesignals. In addition, even though Peptides B and C com-bined reduced TDP-43 aggregation, a much higher per-centage of dead cells were observed in Peptides B and Ctreated cells without aggregation (12.22% ± 0.75%) thanin scrambled peptide treated cells (5.95% ± 0.95%).Finally, we also measured cell death in cells withoutmutant TDP-43 overexpression but treated with thepeptides. Cell death percentage in the untransfectedcells of wild-type TDP-43, mutant TDP-43, mutantTDP-43 plus peptides BC and mutant TDP-43 plusscrambled peptide groups were 3.14% ± 0.56%, 2.28% ±0.36%, 3.22% ± 0.35%, and 3.57% ± 0.39%, respectively(Figure 8). There was no difference among all of thesegroups, indicating that the peptides themselves werenoncytotoxic.DiscussionHere we report the development of synthetic peptides thatcan reduce the aggregation of mutant TDP-43 and canprevent the formation of stress induced wild type TDP-43inclusions in HeLa cell models. We were concerned about“off-target” interactions of our peptides. We found noother proteins sharing the same sequences of TDP-43 towhich peptides bind after searching the NCBI molecularbiology database. We therefore think that off-target inter-actions may be less likely.Our findings confirm an earlier report by Nonaka andcolleagues in SY5Y cells by showing that mutated TDP-43(Δ187-192) can induce TDP-43 aggregation and cell deathin HeLa cells [18]. However, although many TDP-43 mu-tations have been found, the majority of ALS and FTLD-Ucases have no known TDP-43 mutations and instead, it iswild-type TDP-43 that aggregates in these conditions [25].Accordingly, we tested our peptides on a second model inHeLa cells in which the formation of insoluble endogen-ous TDP-43 aggregates was induced by oxidative stressorarsenite [26]. Increased TDP-43 aggregation was observedand quantified using immunocytochemical and biochem-ical methods (separation of soluble and insoluble protein).Importantly, while the peptides reduced both mutantTDP-43 and endogenous TDP-43 aggregation induced bystress in a concentration-dependent manner, they did notFigure 2 Validated peptide candidates that may block TDP-43binding determined using a high density peptide array. Thepeptide mixture inhibits the binding between TDP-43 protein andthe membrane. Membranes were incubated with A, TDP-43 protein;B, TDP-43 protein and a mixture of the five candidate peptides;C, TDP-43 protein and the scrambled peptide TC. The data showthat the five candidate peptides can effectively inhibit the bindingof TDP-43 to the TDP-43 membrane. The experiment was repeatedthree times with similar results.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 4 of 11http://www.actaneurocomms.org/content/1/1/49prevent or reduce the cytotoxicity caused by overexpressionof the mutant TDP-43. These data suggest that while TDP-43 aggregation is associated with the process of cytotox-icity, it is not the cause of cell death in our model.Our results indicating that aggregation may not be thecausative factor for cytotoxicity in cells expressing mu-tant TDP-43, appear consistent with other emerging evi-dence that aggregation is not necessarily the cause ofcellular toxicity. Previous work from our lab [27] showedthat knockdown of progranulin in mouse cortical neu-rons induced TDP-43 translocation from nucleus to thecytoplasm and resulted in enhanced vulnerability to sev-eral stressors, including H2O2 and NMDA. Despite thisincreased neuronal vulnerability, we did not observe ei-ther nuclear or cytoplasmic TDP-43 aggregation. Theseresults suggesting that reduced levels of TDP-43 in thenucleus may be critical for cytoxicity are consistent withevidence reviewed by Z.S. Xu in a recent paper [28].These results are also consistent with those of Barmadaet al. who recently established a TDP-43 proteinopathymodel by transfecting a mutant form of TDP-43 in ratprimary cortical neurons. They observed that their mu-tant TDP-43 is translocated from nucleus to cytoplasmand observed increased cell death. However, they foundno aggregation or inclusions in their neurons [29]. Re-cent evidence from transgenic mice has also beenreported and appears consistent with this idea. Expres-sion of a mutant TDP-43 plasmid induced neuron deathand degeneration, but did not lead to cytoplasmic inclu-sions [30]. Moreover, some studies with other proteinshave shown that aggregation may even serve a protect-ive role. In Huntington’s disease, for instance, mutantHuntington protein (mHtt) is toxic when it is soluble[31,32]. Overexpression of mHtt in HEK293 cells causedaggregation but had no effect on cell survival. However,when mHtt was overexpressed with the small guaninenucleotide-binding protein Rhes, it showed decreasedaggregation but increased cell death [33]. Togetherthose results are leading to the view that protein aggre-gation is not an obligatory factor leading to cytotoxicity.Protein aggregation has been observed in a number ofneurodegenerative diseases. While aggregation is gener-ally to be associated with cytotoxicity, it can also be aphenomenon associated with excessive ER stress, whichis caused by an unusually large amount of misfoldedproteins in the cell [34-38]. In fact, it is now clear thatoverexpression of wild-type TDP-43 can also cause ag-gregation and cytotoxicity as observed by us (Figure 6)and others [16,39]. Expression of large amounts of wild-type TDP-43 might cause an ER stress response andthen initiate cell death. By reducing the protein-proteininteraction that leads to aggregation, the present study,Figure 3 Expression of mutated TDP-43 (deletion of 187–192 AA) results in the formation of inclusions. Immunofluorescent detection ofTDP-43 in HeLa cells with anti-TDP-43 antibody (left panel, green), nuclear staining by Hoechst (right panel, blue). A: untransfected HeLa cells;B: HeLa cells examined 48 h post transfection with mutated TDP-43; C and D: HeLa cells examined 48 h post transfection with mutated TDP-43,followed by addition of the mixture of five peptides (10 μM) (C) or the scrambled peptide TC (10 μM) (D) to the culture medium. The peptideseffectively inhibited aggregation of TDP-43 in HeLa cells. The experiment was repeated three times with similar results.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 5 of 11http://www.actaneurocomms.org/content/1/1/49Figure 4 Synthetic peptides can inhibit TDP-43 aggregation. A: Peptide B, C and B + C combined were able to inhibit aggregation moreefficiently than the other three peptide candidates. B: Peptide B, C and the B + C mixture inhibit aggregation in a concentration-dependentmanner (*, p < 0.05; **, p < 0.01, one-way ANOVA). The data shown are from an experiment with three replicates.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 6 of 11http://www.actaneurocomms.org/content/1/1/49to our best knowledge, is the first attempt to separatethe phenotype of TDP-43 aggregation from that ofcytotoxicity. It also provides insights into the mechan-ism of TDP-43 toxicity in FTD and ALS. Our study suf-fers from the limitation that our experiments wereperformed in a HeLa cell model. FTLD and ALS arediseases of neurons, and moreover of specific popula-tions of neurons. Our experiments need to be extendedand replicated in selected populations of cultured neu-rons. In addition, further studies in transgenic mice willbe needed to determine whether our peptides can alsoreduce TDP-43 aggregation in vivo. Furthermore, thesame approach as the one taken here may be used tostudy the role of aggregation of other proteins, e.g.Huntingtin, α-synuclein, β-amyloid, in other neurode-generative diseases.Normalized cell viability (%)Figure 6 Assessment of cell viability using the MTT assay. Cells were transfected with either wild type or mutated TDP-43 plasmids. PeptidesB and C were unable to prevent cell death. Cell viability was normalized to the basal condition (*, p < 0.05; **, p < 0.01, one-way ANOVA). Thedata shown are from an experiment with three replicates.Ratio of insoluble to soluble TDP-43(Insoluble TDP-43/ Soluble TDP-43) *100%Figure 5 Synthetic peptides decreased arsenite induced TDP-43 aggregation. Insoluble and soluble TDP-43 levels were quantified underdifferent conditions (basal, arsenite, arsenite + Peptide ABCDE, arsenite + Scrambled peptide) (*, p < 0.05; **, p < 0.01, one-way ANOVA). The datashown are from an experiment with three replicates.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 7 of 11http://www.actaneurocomms.org/content/1/1/49ConclusionIn this study, we are the first group to identify theregions in TDP-43 protein that involve in its self-aggregation. Peptides derived from those regions caneffectively reduce the aggregation in cells transfectedwith an aggregation prone mutant TDP-43 or treatedwith arsenite. We then showed that reducing TDP-43aggregation did not protect cells from cell deathcaused by either the expression of mutant TDP-43 orarsentie treatment. Thus, our study is the first toseparate aggregation of TDP-43 from other factorsinvolved in TDP-43 proteinopathy and demonstratethat TPD-43 aggregation may not be the cause ofcytotoxicity.Percentage of viable cellsFigure 7 Synthetic peptides did not rescue arsenite induced cell toxicity. Cell viability was assessed using the MTT assay and normalized tothe basal condition (*, p < 0.05; **, p < 0.01, one-way ANOVA). The data shown are from an experiment with three replicates.Figure 8 Schematic flowchart of the cell death studies. Cell death was identified using the TUNEL assay in HeLa cells. Cells were transfectedwith GFP and the wild type TDP-43 or mutated TDP-43 plasmid. About 1000 cells of each sample were counted. After treatment with Peptide Band C, the amount of cells with aggregation decreased. However, despite the reduction in aggregation, increased cell death was observed in thenon-aggregated cells. This suggests that the peptides can inhibit aggregation but not cell death. In addition, the very low percentage of celldeath in untransfected cells suggests the peptides are not toxic to cells.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 8 of 11http://www.actaneurocomms.org/content/1/1/49MethodsProtein arrayThe peptides were synthesized on derivatized cellulose-based membranes (Intavis AG, Köln, Germany) by theUBC Peptide Synthesis facility using a previously de-scribed protocol [40]. The peptide scans were performedby synthesizing overlapping 12 or 14-mer peptides span-ning the entire 414 amino acid sequence of TDP-43 witha frame shift of 2 amino acids per spot. The TDP-43 pro-tein was purified from E. coil overexpressing the humanTDP-43 gene. Membranes were blocked with 5% sucroseand 4% nonfat dry milk in Tris-buffered Saline Tween-20(TBST) for 4 h and then incubated with TDP-43 protein(3–10 μg/ml) or peptides overnight at 4°C. Then mem-branes were incubated with rabbit polyclonal TDP-43antibody (1:5000; Protein Tech Group, Chicago, IL) over-night at 4°C. After washing three times for 15 min, mem-branes were incubated with donkey anti-goat conjugatedwith donkey anti-rabbit IgG conjugated with horseradishperoxidase (1: 5000; R&D Systems, Minneapolis, MN) for3 h at room temperature. Membranes were then washedthree times for 15 min and protein interaction was visual-ized with an enhanced chemiluminescence reaction assay(PerkinElmer Life Sciences).Cell-penetrating peptide synthesisCell-penetrating peptides consisting of the truncatedTAT domain at the N-terminal were synthesized by GLbiochem Ltd. (Shanghai) and purified by HPLC. TAT,the trans-acting activator of transcription of the humanimmunodeficiency virus (HIV-1), has been used as an ef-ficient way of delivering proteins or peptides into cells[41]. The resulting peptides were more than 90% pureand verified by mass spectrometry. All the peptides werefirst dissolved in DMSO and further diluted in water be-fore usage.Blocking assay of TDP-43 interaction with the membraneMembranes were incubated with the mixture of TDP-43protein (5 μg/ml) and/or single synthetic peptides(100 μg/ml) overnight at 4°C. Then the membranes wereincubated sequentially with primary antibody and sec-ondary antibody and washed with TBST. The proteininteraction was visualized using an enhanced chemilu-minescence reaction assay (PerkinElmer Life Sciences).Optical densities of each peptide array were measuredusing NIH ImageJ software.Cell cultures, transfection and treatmentHeLa cells were obtained from the American Type Cul-ture Collection (ATCC) and grown in Dulbecco’s modifiedEagle’s medium (DMEM; Sigma Chemical Co., St. Louis,MO) supplemented with 10% fetal bovine serum (Gibco-BRL, Grand Island, NY) and 1% antibiotics (Gibco-BRL,Grand Island, NY). Cultures were maintained at 37°C in ahumidified incubator (NuAir, Plymouth, MN) with 5%CO2. Wild type TDP-43 plasmid and mutant TDP-43plasmid lacking residues 187–192 were obtained fromDr. Nonaka and Dr. Arai. The plasmids were transfectedinto the HeLa cells using the Lipofectamine 2000 system(Invitrogen, Carlsbad, CA) according to the manufac-turer’s instructions. After incubating the cells with thetransfection mixture for 5 hours, the medium was re-placed with fresh DMEM (10% fetal bovine serum) with2.5–20 μM synthetic peptides. After 24 hours, the mediumwas replaced with fresh peptides. The cells were returnedto the incubator for an additional 24 hours before study.For studies using arsenite, arsenite (Fisher scientific CO.,Pittsburg, PA) was added to the cell culture medium to afinal concentration of 0.5 μM with or without the presenceof synthetic peptides. The cells were then returned to theincubator for 4 hours before harvesting and analysis.Immunofluorescence studiesCells were fixed with pre-warmed 4% Paraformaldehyde(PFA; Sigma, Saint Louis, MO) containing 4% sucrose(Sigma, Saint Louis, MO) in PBS for 10 minutes at roomtemperature and permeabilized with 0.1% Trition X-100(Sigma, Saint Louis, MO) in PBS for 2–3 minutes atroom temperature. Then the reaction was blocked with10% Bovine serum albumin (BSA; Invitrogen, Carlsbad,CA) in PBS for 1 hour at room temperature and incu-bated with a primary antibody against TDP-43 (1:150;ProteinTech, Chicago, IL) overnight at 4°C. Reactionswere visualized with anti-rabbit antibody conjugatedwith Alexa 488 (Invitrogen, Carlsbad, CA). Cell nucleiwere labeled with DAPI (1:10000; Invitrogen, Carlsbad,CA) for 2–3 minutes at room temperature. Images wereobtained with an Olympus Fluoview FV1000 Confocalscanning microscope.MTT assay for cell viabilityThe viability of cultured cells was assessed usingthe MTT assay. MTT (3-(4,5-dimethylthiazol-yl)-2,5-diphenyltetrazolium bromide) (Sigma, Saint Louis,MO) was added to each well (20 μl, 5 mg/ml). After4 h incubation, cells were lysed overnight. The opticaldensity of test and control samples was measured at590 nm absorbance with a reference filter of 750 nmusing a “uQuant” microplate spectrophotometer (Bio-Tek Instruments, USA). The absorbance was translatedinto cell viability ratios for comparison: cell viabilityratio = (test sample absorbance/control sample absorb-ance) × 100%.TUNEL assay for cell viabilityDeoxynucleotidyl-transferase-mediated dUTP nick end-labeling (TUNEL) staining was performed using an inLiu et al. Acta Neuropathologica Communications 2013, 1:49 Page 9 of 11http://www.actaneurocomms.org/content/1/1/49situ cell death detection kit (Roche Applied Science, IN).Cells were seeded on poly-D-lysine coated cover slips ata density of 2.5 × 104 cells per well in 24 well plate. Aftertreatment, cells were fixed and stained with anti-TDP-43(1:150; ProteinTech, Chicago, IL), anti-GFP (1:1000;Invitrogen, Carlsbad, CA), Alexa 647 anti-rabbit (1:1000;Invitrogen, Carlsbad, CA) and Alexa 488 anti-chicken(1:1000; Invitrogen, Carlsbad, CA). The enzyme solution(terminal transferase) and label solution were mixed in avolume ratio of 1:9 to obtain the TUNEL reaction mix-ture. The cells were incubated with the 50 μl TUNEL re-action mixture at 37°C for 1 hour and then washed threetimes with PBS. Then the cover slips were mounted onglass slides with antifade reagent (Invitrogen, Carlsbad,CA). Images were obtained with an Olympus FluoviewFV1000 Confocal scanning microscope.Biochemical fractionationTo examine the effect of arsenite on solubility of endogen-ous TDP-43, preparation of soluble/insoluble protein wereperformed. HeLa cells were washed twice with cold PBS,lysed in cold lysis buffer (PBS with 1% Triton X-100,10 μg/ml aprotinin, 0.5 mM PMSF). Cell lysates were ro-tated for 15 min and then centrifuged at 55,000 rpm for15 min at 4°C. The supernatants were collected as solubleproteins. To prevent contamination caused by carryingover, the pellets were re-centrifuged at 55,000 rpm for15 min at 4°C and washed once with lysis buffer. Then thepellets were lysed in PBS with 1% SDS and sonicated. Thesupernatants were collected as insoluble protein. Solubleand insoluble proteins were analyzed by immunoblotting.ImmunoblottingProtein concentrations of cell lysate and medium weremeasured using a standard BCA assay (Bio-Rad) as perthe manufacturer’s instructions. Proteins were separatedin SDS–PAGE (10%) and transferred to nitrocellulosemembranes (Bio-Rad) at 4°C. Membranes were blockedwith 5% nonfat dry milk in TBST for 1 hour, then incu-bated in 5% nonfat dry milk/TBST overnight at 4°C withantibodies for TDP-43 or β-actin (rabbit IgG, 1:1000;Cell Signaling, Danvers, MA). After washing with TBST,membranes were incubated with anti-rabbit secondaryantibody conjugated with horseradish peroxidase for1 hour at room temperature. Blots were developed usingan enhanced chemiluminescence reaction assay.Statistical analysisAll data were expressed as mean ± SD and analyzed byone-way ANOVA. A value of p < 0.05 was consideredstatistically significant.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsRL, WJ and MSC conceived and designed the experiments. RL performed theexperiments and analyzed the data. GY provided assistance in application ofimmunofluorescence methods. TN and TA were involved in plasmidconstruction. RL, WJ and MSC drafted and edited the manuscript. All Authorsread and approved the final manuscript.AcknowledgmentsThis work was supported by the Pacific Alzheimer Research Foundation(PARF).Author details1Brain Research Center, University of British Columbia, 2211 Wesbrook Mall,Vancouver, BC V6T2B5, Canada. 2Department of Neuropathology and CellBiology, Tokyo Metropolitan Institute of Medical Science, Setagaya-ku, Tokyo156-8506, Japan. 3Department of Psychiatry, Graduate School ofComprehensive Human Sciences, University of Tsukuba, 1-1-1 Tennoudai,Tsukuba, Ibaraki 305-8575, Japan.Received: 31 July 2013 Accepted: 2 August 2013Published: 9 August 2013References1. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, et al: TDP-43 is acomponent of ubiquitin-positive tau-negative inclusions infrontotemporal lobar degeneration and amyotrophic lateral sclerosis.Biochem Biophys Res Commun 2006, 351:602–611.2. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, et al:Ubiquitinated TDP-43 in frontotemporal lobar degeneration andamyotrophic lateral sclerosis. Science 2006, 314:130–133.3. Ou SH, Wu F, Harrich D, Garcia-Martinez LF, Gaynor RB: Cloning andcharacterization of a novel cellular protein, TDP-43, that binds to HumanImmunodeficiency Virus Type 1 TAR DNA sequence motifs. J Virol 1995,69:3584–3596.4. Buratti E, Brindisi A, Giombi M, Tisminetzky S, Ayala YM, et al: TDP-43 bindsheterogeneous nuclear ribonucleoprotein A/B through its C-terminal tail:an important region for the inhibition of cystic fibrosis transmembraneconductance regulator exon 9 splicing. J Biol Chem 2005, 280:37572–37584.5. Acharya KK, Govind CK, Shore AN, Stoler MH, Reddi PP: cis-Requirement forthe maintenance of round spermatid-specific transcription. Dev Biol 2006,295:781–790.6. Wang HY, Wang IF, Bose J, Shen CK: Structural diversity and functionalimplications of the eukaryotic TDP gene family. Genomics 2004,83:130–139.7. Mercado PA, Ayala YM, Romano M, Buratti E, Baralle FE: Depletion of TDP43 overrides the need for exonic and intronic splicing enhancers in thehuman apoA-II gene. Nucleic Acids Res 2005, 33:6000–6010.8. Buratti E, Dork T, Zuccato E, Pagani F, Romano M, et al: Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo CFTR exon 9 skipping.EMBO J 2001, 20:1774–1784.9. Strong MJ, Volkening K, Hammond R, Yang W, Strong W, et al: TDP43 is ahuman low molecular weight neurofilament (hNFL) mRNA-bindingprotein. Mol Cell Neurosci 2007, 35:320–327.10. Kim SH, Shanware NP, Bowler MJ, Tibbetts RS: Amyotrophic lateralsclerosis-associated proteins TDP-43 and FUS/TLS function in a commonbiochemical complex to co-regulate HDAC6 mRNA. J Biol Chem 2010,285:34097–34105.11. Rutherford NJ, Zhang YJ, Baker M, Gass JM, Finch NA, et al: Novelmutations in TARDBP (TDP-43) in patients with familial amyotrophiclateral sclerosis. PLoS Genet 2008, 4:e1000193.12. Davidson Y, Kelley T, Mackenzie IR, Pickering-Brown S, Du Plessis D, et al:Ubiquitinated pathological lesions in frontotemporal lobar degenerationcontain the TAR DNA-binding protein, TDP-43. Acta Neuropathol 2007,113:521–533.13. Johnson BS, McCaffery JM, Lindquist S, Gitler AD: A yeast TDP-43proteinopathy model: exploring the molecular determinants of TDP-43aggregation and cellular toxicity. Proc Natl Acad Sci U S A 2008,105:6439–6444.14. Zhang YJ, Xu YF, Cook C, Gendron TF, Roettges P, Link CD, et al: Aberrantcleavage of TDP-43 enhances aggregation and cellular toxicity. Proc NatlAcad Sci U S A 2009, 21:7607–7612.Liu et al. Acta Neuropathologica Communications 2013, 1:49 Page 10 of 11http://www.actaneurocomms.org/content/1/1/4915. Zhang YJ, Xu YF, Dickey CA, Buratti E, Baralle F, et al: Progranulin mediatescaspase-dependent cleavage of TAR DNA binding protein-43. J Neurosci2007, 27:10530–10534.16. Xu YF, Gendron TF, Zhang YJ, Lin WL, D’Alton S, et al: Wild-type humanTDP-43 expression causes TDP-43 phosphorylation, mitochondrialaggregation, motor deficits, and early mortality in transgenic mice.J Neurosci 2010, 30:10851–10859.17. Shan X, Chiang PM, Price DL, Wong PC: Altered distributions of Gemini ofcoiled bodies and mitochondria in motor neurons of TDP-43 transgenicmice. Proc Natl Acad Sci U S A 2010, 107:16325–16330.18. Nonaka T, Arai T, Buratti E, Baralle FE, Akiyama H, et al: Phosphorylated andubiquitinated TDP-43 pathological inclusions in ALS and FTLD-U arerecapitulated in SH-SY5Y cells. FEBS Lett 2009, 583:394–400.19. Kedersha N, Anderson P: Mammalian stress granules and processingbodies. Methods Enzymol 2007, 431:61–81.20. Kedersha N, Anderson P: Stress granules: sites of mRNA triage thatregulate mRNA stability and translatability. Biochem Soc Trans 2002,30:963–969.21. McEwen E, Kedersha N, Song B, Scheuner D, Gilks N, et al: Heme-regulatedinhibitor kinase-mediated phosphorylation of eukaryotic translationinitiation factor 2 inhibits translation, induces stress granule formation,and mediates survival upon arsenite exposure. J Biol Chem 2005,280:16925–16933.22. McDonald KK, Aulas A, Destroismaisons L, Pickles S, Beleac E, et al: TARDNA-binding protein 43 (TDP-43) regulates stress granule dynamics viadifferential regulation of G3BP and TIA-1. Hum Mol Genet 2011,20:1400–1410.23. Meyerowitz J, Parker SJ, Vella LJ, Ng D, Price KA, et al: C-Jun N-terminalkinase controls TDP-43 accumulation in stress granules induced byoxidative stress. Mol Neurodegeneration 2011, 6:57.24. Nagahara H, Vocero-Akbani AM, Snyder EL, Ho A, Latham DG, et al:Transduction of full-length TAT fusion proteins into mammalian cells:TAT-p27Kip1 induces cell migration. Nat Med 1998, 4:1449–1452.25. Barber SC, Shaw PJ: Oxidative stress in ALS: key role in motor neuroninjury and therapeutic target. Free Radic Biol Med 2010, 48:629–641.26. Parker SJ, Meyerowitz J, James JL, Liddell JR, Crouch PJ, et al: EndogenousTDP-43 localized to stress granules can subsequently form proteinaggregates. Neurochem Int 2012, 60:415–424.27. Guo A, Tapia L, Bamji SX, Cynader MS, Jia W: Progranulin deficiency leadsto enhanced cell vulnerability and TDP-43 translocation in primaryneuronal cultures. Brain Res 2010, 1366:1–8.28. Xu ZS: Does a loss of TDP-43 function cause neurodegeneration?Mol Neurodegeneration 2012, 7:27.29. Barmada SJ, Skibinski G, Korb E, Rao EJ, Wu JY, et al: Cytoplasmicmislocalization of TDP-43 is toxic to neurons and enhanced by amutation associated with familial amyotrophic lateral sclerosis.J Neurosci 2010, 30:639–649.30. Igaz LM, Kwong LK, Lee EB, Chen-Plotkin A, Swanson E, Unger T, Malunda J,et al: Dysregulation of the ALS-associated gene TDP-43 leads to neuronaldeath and degeneration in mice. J Clin Invest 2011, 121:726–738.31. Saudou F, Finkbeiner S, Devys D, Greenberg ME: Huntingtin acts in thenucleus to induce apoptosis but death does not correlate with theformation of intranuclear inclusions. Cell 1998, 95:55–66.32. Arrasate M, Mitra S, Schweitzer ES, Segal MR, Finkbeiner S: Inclusion bodyformation reduces levels of mutant huntingtin and the risk of neuronaldeath. Nature 2004, 431:805–810.33. Subramaniam S, Sixt KM, Barrow R, Snyder SH: Rhes, a striatal specificprotein, mediates mutant-huntingtin cytotoxicity. Science 2009,324:1327–1330.34. Rao RV, Bredesen DE: Misfolded proteins, endoplasmic reticulum stressand neurodegeneration. Curr Opin Cell Biol 2004, 16:653–662.35. Kopito RR: Aggresomes, inclusion bodies and protein aggregation.Trends Cell Biol 2000, 10:524–530.36. Harding HP, Calfon M, Urano F, Novoa I, Ron D: Transcriptional andtranslational control in the mammalian unfolded protein response.Annu Rev Cell Dev Biol 2002, 18:575–599.37. Rutkowski DT, Kaufman RJ: A trip to the ER: coping with stress. Trends CellBiol 2004, 14:20–28.38. Paschen W, Frandsen A: Endoplasmic reticulum dysfunction—a commondenominator for cell injury in acute and degenerative diseases of thebrain? J Neurochem 2001, 79:719–725.39. Ash PE, Zhang YJ, Roberts CM, Saldi T, Hutter H, et al: Neurotoxic effects ofTDP-43 overexpression in C. elegans. Hum Mol Genet 2010, 19:3206–3218.40. Hilpert K, Winkler DF, Hancock RE: Peptide arrays on cellulose support:SPOT synthesis, a time and cost efficient method for synthesis of largenumbers of peptides in a parallel and addressable fashion. Nat Protoc2007, 2:1333–1349.41. Schwarze SR, Ho A, Vocero-Akbani A, Dowdy SF: In vivo proteintransduction: delivery of a biologically active protein into the mouse.Science 1999, 285:1569–1572.doi:10.1186/2051-5960-1-49Cite this article as: Liu et al.: Reducing TDP-43 aggregation does notprevent its cytotoxicity. Acta Neuropathologica Communications 2013 1:49.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitLiu et al. Acta Neuropathologica Communications 2013, 1:49 Page 11 of 11http://www.actaneurocomms.org/content/1/1/49


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items