TDP-43: gumming up neurons through protein–protein and protein–RNA interactions

Since the discovery that 43 kDa TAR DNA binding protein (TDP-43) is involved in neurodegeneration, studies of this protein have focused on the global effects of TDP-43 expression modulation on cell metabolism and survival. The major difficulty with these global searches, which can yield hundreds to thousands of variations in gene expression level and/or mRNA isoforms, is our limited ability to separate specific TDP-43 effects from secondary dysregulations occurring at the gene expression and various mRNA processing steps. In this review, we focus on two biochemical properties of TDP-43: its ability to bind RNA and its protein–protein interactions. In particular, we overview how these two properties may affect potentially very important processes for the pathology, from the autoregulation of TDP-43 to aggregation in the cytoplasmic/nuclear compartments. Since the discovery that 43 kDa TAR DNA binding protein (TDP-43) is involved in neurodegeneration, studies of this protein have focused on the global effects of TDP-43 expression modulation on cell metabolism and survival. The major difficulty with these global searches, which can yield hundreds to thousands of variations in gene expression level and/or mRNA isoforms, is our limited ability to separate specific TDP-43 effects from secondary dysregulations occurring at the gene expression and various mRNA processing steps. In this review, we focus on two biochemical properties of TDP-43: its ability to bind RNA and its protein–protein interactions. In particular, we overview how these two properties may affect potentially very important processes for the pathology, from the autoregulation of TDP-43 to aggregation in the cytoplasmic/nuclear compartments. In this review, we focus on the intrinsic biochemical properties of TDP-43, such as its interactions with RNA, protein–protein interactions, and its propensity for aggregation. A better understanding of TDP-43 biochemistry (RNA binding kinetics and specificity, molecular determinants of protein–protein interactions, and others) might give clues into the cellular processes leading to disease, and provide us with a better ability to develop novel therapeutic options. Since TDP-43 involvement in amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD) was initially described in 2006 [1Neumann M. et al.Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis.Science. 2006; 314: 130-133Crossref PubMed Scopus (4288) Google Scholar], the number of laboratories focusing on this protein has increased dramatically. The main reason for this interest is that aberrant TDP-43 aggregation is at the center of an extensive network of neuronal diseases that are collectively been referred to as the TDP-43 proteinopathies [2Geser F. et al.Clinical and pathological continuum of multisystem TDP-43 proteinopathies.Arch. Neurol. 2009; 66: 180-189Crossref PubMed Scopus (201) Google Scholar]. TDP-43 is the prototype of this type of pathological process, but there are other RNA binding proteins such as fused in sarcoma/translocated in liposarcomas (FUS/TLS) and other heterogeneous nuclear ribonucleoproteins (hnRNPs) that are also mutated and/or aggregated in diseased brains. Like TDP-43, they are involved in the general control of mRNA processing steps. Several extensive reviews have already been written on the TDP-43 proteinopathies, so they are not discussed here [3Mackenzie I.R. et al.TDP-43 and FUS in amyotrophic lateral sclerosis and frontotemporal dementia.Lancet Neurol. 2010; 9: 995-1007Abstract Full Text Full Text PDF PubMed Scopus (639) Google Scholar]. The discovery that TDP-43 is involved in neurodegeneration, followed closely by similar findings for FUS/TLS and C9orf72, have opened up the entire field of RNA binding proteins (RBPs) and RNA metabolism as a new and promising area of research in neuroscience [4Lee E.B. et al.Gains or losses: molecular mechanisms of TDP43-mediated neurodegeneration.Nat. Rev. Neurosci. 2012; 13: 38-50Google Scholar]. Establishing the direct links between TDP-43 and disease has not been easy and might not get easier in the near future. The principal difficulty comes from the huge number of processes that can be aberrantly affected by TDP-43 aggregations in neurons and glia, and by nuclear depletion of TDP-43. TDP-43 was initially described as a transcription factor [5Ou S.H. et al.Cloning and characterization of a novel cellular protein, TDP-43, that binds to human immunodeficiency virus type 1 TAR DNA sequence motifs.J. Virol. 1995; 69: 3584-3596Crossref Google Scholar], however, the roles played by TDP-43 have rapidly expanded to include regulation of splicing [6Buratti E. et al.Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo CFTR exon 9 skipping.EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar], mRNA stability (including its own) [7Ayala Y.M. et al.TDP-43 regulates its mRNA levels through a negative feedback loop.EMBO J. 2011; 30: 277-288Crossref PubMed Scopus (370) Google Scholar, 8Volkening K. et al.Tar DNA binding protein of 43 kDa (TDP-43), 14-3-3 proteins and copper/zinc superoxide dismutase (SOD1) interact to modulate NFL mRNA stability. Implications for altered RNA processing in amyotrophic lateral sclerosis (ALS).Brain Res. 2009; 1305: 168-182Crossref Scopus (155) Google Scholar, 9Fiesel F.C. et al.Knockdown of transactive response DNA-binding protein (TDP-43) downregulates histone deacetylase 6.EMBO J. 2010; 29: 209-221Crossref PubMed Scopus (178) Google Scholar], microRNA processing [10Buratti E. et al.Nuclear factor TDP-43 can affect selected microRNA levels.FEBS J. 2010; 277: 2268-2281Crossref PubMed Scopus (171) Google Scholar, 11Kawahara Y. Mieda-Sato A. TDP-43 promotes microRNA biogenesis as a component of the Drosha and Dicer complexes.Proc. Natl. Acad. Sci. U.S.A. 2012; 109: 3347-3352Crossref PubMed Scopus (292) Google Scholar], mRNA transport and translation (Figure 1) [12Wang I.F. et al.TDP-43, the signature protein of FTLD-U, is a neuronal activity-responsive factor.J. Neurochem. 2008; 105: 797-806Crossref PubMed Scopus (283) Google Scholar, 13Fiesel F.C. et al.TDP-43 regulates global translational yield by splicing of exon junction complex component SKAR.Nucleic Acids Res. 2011; (doi:10.1093.nar.gkr1082)Google Scholar, 14Godena V.K. et al.TDP-43 regulates Drosophila neuromuscular junctions growth by modulating Futsch/MAP1B levels and synaptic microtubules organization.PLoS ONE. 2011; 6: e17808Crossref PubMed Scopus (93) Google Scholar]. Dysregulation of many of these processes might precede or follow the aberrant aggregation and modification of TDP-43 in affected cells. While evaluating all of these potential pathways and mechanisms, however, the basic characteristics of TDP-43 should be considered because they will ultimately control disease pathogenesis. Structurally, TDP-43 belongs to the very large family of nuclear factors known as hnRNPs (Box 1). Like most members of this family, the main distinguishing feature of TDP-43 is its ability to bind RNA in a single-stranded and sequence-specific manner. This is achieved by the presence of two 60 residue-long motifs that fold in a conserved three dimensional conformation, which are known as RNA recognition motifs (RRMs) (Figure 2a) . In TDP-43, these regions are highly evolutionarily conserved, such that its Drosophila melanogaster and Caenorhabditis elegans homologs can functionally substitute for the human protein (and vice versa) in a variety of experimental systems [15Ayala Y.M. et al.Human, Drosophila, and C. elegans TDP-43: nucleic acid binding properties and splicing regulatory function.J. Mol. Biol. 2005; 348: 575-588Crossref Scopus (263) Google Scholar, 16Ash P.E. et al.Neurotoxic effects of TDP-43 overexpression in C. elegans.Hum. Mol. Genet. 2010; 19: 3206-3218Crossref Scopus (165) Google Scholar, 17Feiguin F. et al.Depletion of TDP-43 affects Drosophila motoneurons terminal synapsis and locomotive behavior.FEBS Lett. 2009; 583: 1586-1592Abstract Full Text Full Text PDF PubMed Scopus (237) Google Scholar]. Currently, there is scant detailed structural analysis of TDP-43 RRMs bound to RNA sequences, owing to the great propensity of this protein to aggregate. The only exceptions are the crystallographic structure of RRM2 bound to a small DNA molecule [18Kuo P.H. et al.Structural insights into TDP-43 in nucleic-acid binding and domain interactions.Nucleic Acids Res. 2009; 37: 1799-1808Crossref PubMed Scopus (208) Google Scholar] and the functionally validated computer model of the RRM1 structure [15Ayala Y.M. et al.Human, Drosophila, and C. elegans TDP-43: nucleic acid binding properties and splicing regulatory function.J. Mol. Biol. 2005; 348: 575-588Crossref Scopus (263) Google Scholar]. Nonetheless, it is already clear from the initial characterization of the RNA binding preferences of TDP-43 that it prefers to bind UG-repeated motifs, for which RRM1 plays a crucial role [19Buratti E. Baralle F.E. Characterization and functional implications of the RNA binding properties of nuclear factor TDP-43, a novel splicing regulator of CFTR exon 9.J. Biol. Chem. 2001; 276: 36337-36343Crossref PubMed Scopus (475) Google Scholar]. The RNA binding ability of TDP-43 is essential for several RNA processing steps (Figure 1), in particular alternative splicing. When localized near the 3′ splice sites or 5′ splice sites of several exons, TDP-43 binding to UG-repeats can silence splicing, as in the case of cystic fibrosis transmembrane regulator (CFTR) exon 9 [6Buratti E. et al.Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo CFTR exon 9 skipping.EMBO J. 2001; 20: 1774-1784Crossref PubMed Scopus (487) Google Scholar], apolipoprotein AII (ApoAII) exon 3 [20Mercado P.A. et al.Depletion of TDP-43 overrides the need for exonic and intronic splicing enhancers in the human apoA-II gene.Nucleic Acids Res. 2005; 33: 6000-6010Crossref Scopus (191) Google Scholar], eukaryotic translation termination factor 1 (ETF1), and retinoid X receptor gamma (RXRG) [21Passoni M. et al.UG repeats/TDP-43 interactions near 5′ splice sites exert unpredictable effects on splicing modulation.J. Mol. Biol. 2011; 415: 46-60Crossref Scopus (22) Google Scholar]; or, it can enhance splicing, as in the case of a breast cancer 1 (BRCA1)-mutated substrate [21Passoni M. et al.UG repeats/TDP-43 interactions near 5′ splice sites exert unpredictable effects on splicing modulation.J. Mol. Biol. 2011; 415: 46-60Crossref Scopus (22) Google Scholar] and polymerase delta interacting protein/S6 kinase 1 Aly/REF-like target (POLDIP3/SKAR) (Figure 2b) [13Fiesel F.C. et al.TDP-43 regulates global translational yield by splicing of exon junction complex component SKAR.Nucleic Acids Res. 2011; (doi:10.1093.nar.gkr1082)Google Scholar].Box 1What are hnRNP proteins?hnRNP proteins were initially identified as the major proteins directly associated with heterogeneous nuclear RNA. They possess the ability to bind RNA in a sequence-specific manner (although many of them also have a nonspecific affinity for RNA) and often contain glycine-rich regions that are important in mediating protein–protein interactions. Because of this loose definition, it is difficult to ascribe a specific function to individual hnRNP proteins in the absence of further characterization. This is because individual members can play a role in all aspects of RNA metabolism, from transcription and splicing to miRNA maturation, mRNA transport and degradation, and even protein translation. In addition, very few hnRNPs have just one well-defined function. Rather, the most common situation is that the same hnRNP can play different roles in different processes depending on its subcellular localization, relative abundance, binding context within a particular RNA, and the interactions with other cellular components. However, despite this variability the hnRNPs are some of the most important negative regulators of alternative splicing, and function antagonistically to the well-known class of splicing enhancers, the serine–arginine family of splicing factors (SR proteins). hnRNP proteins were initially identified as the major proteins directly associated with heterogeneous nuclear RNA. They possess the ability to bind RNA in a sequence-specific manner (although many of them also have a nonspecific affinity for RNA) and often contain glycine-rich regions that are important in mediating protein–protein interactions. Because of this loose definition, it is difficult to ascribe a specific function to individual hnRNP proteins in the absence of further characterization. This is because individual members can play a role in all aspects of RNA metabolism, from transcription and splicing to miRNA maturation, mRNA transport and degradation, and even protein translation. In addition, very few hnRNPs have just one well-defined function. Rather, the most common situation is that the same hnRNP can play different roles in different processes depending on its subcellular localization, relative abundance, binding context within a particular RNA, and the interactions with other cellular components. However, despite this variability the hnRNPs are some of the most important negative regulators of alternative splicing, and function antagonistically to the well-known class of splicing enhancers, the serine–arginine family of splicing factors (SR proteins). Recently, the strong preference of TDP-43 for UG-repeats has been confirmed in vivo by several high throughput studies using techniques known as in vivo crosslinking and immunoprecipitation-sequencing (CLIP-seq) [22Tollervey J.R. et al.Characterizing the RNA targets and position-dependent splicing regulation by TDP-43.Nat. Neurosci. 2011; 14: 452-458Crossref PubMed Scopus (718) Google Scholar, 23Polymenidou M. et al.Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability from loss of TDP-43.Nat. Neurosci. 2011; 14: 459-468Crossref PubMed Scopus (803) Google Scholar, 24Xiao S. et al.RNA targets of TDP-43 identified by UV-CLIP are deregulated in ALS.Mol. Cell. Neurosci. 2011; 47: 167-180Crossref PubMed Scopus (122) Google Scholar] and RNA immunoprecipitation-sequencing (RIP-seq) [25Sephton C.F. et al.Identification of neuronal RNA targets of TDP-43-containing ribonucleoprotein complexes.J. Biol. Chem. 2011; 286: 1204-1215Crossref PubMed Scopus (305) Google Scholar]. Several of these studies have also highlighted the ability of TDP-43 to bind other types of loosely conserved sequence motifs (Figure 2c). At the moment, the functional significance of many of these binding sites is mostly unknown. TDP-43 binds UG-rich sequences throughout the genome. UG sequences are one of the more common repeats in the human genome, and are found more often in long introns and 3′ untranslated regions (UTRs) than in exons. In view of the distribution of its target sequences, it has been inferred that TDP-43 might help to control long intron splicing and mRNA stability [23Polymenidou M. et al.Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability from loss of TDP-43.Nat. Neurosci. 2011; 14: 459-468Crossref PubMed Scopus (803) Google Scholar]. In support of the latter, TDP-43 has been found to control neurofilament (hNFL) mRNA stability by acting together with other proteins implicated in ALS pathology [8Volkening K. et al.Tar DNA binding protein of 43 kDa (TDP-43), 14-3-3 proteins and copper/zinc superoxide dismutase (SOD1) interact to modulate NFL mRNA stability. Implications for altered RNA processing in amyotrophic lateral sclerosis (ALS).Brain Res. 2009; 1305: 168-182Crossref Scopus (155) Google Scholar]. A novel and important function of human TDP-43, which is dependent on RNA binding, was recently observed [7Ayala Y.M. et al.TDP-43 regulates its mRNA levels through a negative feedback loop.EMBO J. 2011; 30: 277-288Crossref PubMed Scopus (370) Google Scholar]. Interestingly, an extended binding region for TDP-43 (called TDPBR) was identified in the 3′ UTR of TDP-43 mRNA. The TDPBR contains several non-UG sequences, which are essential for autoregulation of TDP-43 mRNA levels. These results, which were obtained in human cells, were confirmed soon afterwards by similar observations in transgenic mouse systems [23Polymenidou M. et al.Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability from loss of TDP-43.Nat. Neurosci. 2011; 14: 459-468Crossref PubMed Scopus (803) Google Scholar, 26Xu Y.F. et al.Wild-type human TDP-43 expression causes TDP-43 phosphorylation, mitochondrial aggregation, motor deficits, and early mortality in transgenic mice.J. Neurosci. 2010; 30: 10851-10859Crossref PubMed Scopus (368) Google Scholar, 27Igaz L.M. et al.Dysregulation of the ALS-associated gene TDP-43 leads to neuronal death and degeneration in mice.J. Clin. Invest. 2011; 121: 726-738Crossref PubMed Scopus (281) Google Scholar]. A model of the autoregulation process is shown in Figure 2c. Briefly, low nuclear TDP-43 levels cause the most efficient polyA1 site to be used instead of the other possible choices (pA2 to pA4). Conversely, an abundance of nuclear TDP-43 results in the use of the suboptimal splice sites, followed by rapid degradation of the mRNA. By modulating this feedback loop, the cell is capable of keeping TDP-43 concentrations constant. Although the mechanism and possible consequences to cellular metabolism have been recently reviewed elsewhere [28Buratti E. Baralle F.E. TDP-43: new aspects of autoregulation mechanisms in RNA binding proteins and their connection with human disease.FEBS J. 2011; 278: 3530-3538Crossref Scopus (39) Google Scholar, 29Budini M. Buratti E. TDP-43 autoregulation: implications for disease.J. Mol. Neurosci. 2011; 45: 473-479Crossref Scopus (36) Google Scholar], it is important to note that disruption of this regulation could play an important role in TDP-43 aggregation. Indeed, formation of TDP-43 aggregates within the cell nucleus or cytoplasm will probably result in reduced free nuclear TDP-43, therefore the 3′ UTR TDPBR sensor will detect a drop in protein levels and respond with increased TDP-43 production. Such a situation would result in a vicious circle, which could lead to cell stress and death, even in the absence of directly toxic effects from TDP-43 aggregates. Depletion of TDP-43 from the nucleus can also have some very serious consequences on the expression levels and RNA processing of many other transcripts (Figure 1, Figure 2) [28Buratti E. Baralle F.E. TDP-43: new aspects of autoregulation mechanisms in RNA binding proteins and their connection with human disease.FEBS J. 2011; 278: 3530-3538Crossref Scopus (39) Google Scholar, 29Budini M. Buratti E. TDP-43 autoregulation: implications for disease.J. Mol. Neurosci. 2011; 45: 473-479Crossref Scopus (36) Google Scholar]. Like RNA-binding, most protein–protein interactions are determined by particular domains or sequences within the architecture of a protein. An important characteristic of TDP-43, therefore, is its ability to bind several proteins that modulate its RNA processing abilities and possibly its aggregation properties. Among the most important binding partners are several members of the hnRNP family, such as hnRNP A1 and A2, which are necessary for many, but not all, of the splicing inhibitory properties of TDP-43 [30Buratti E. et al.TDP-43 binds heterogeneous nuclear ribonucleoprotein A/B through its C-terminal tail: an important region for the inhibition of cystic fibrosis transmembrane conductance regulator exon 9 splicing.J. Biol. Chem. 2005; 280: 37572-37584Crossref PubMed Scopus (349) Google Scholar, 31D'Ambrogio A. et al.Functional mapping of the interaction between TDP-43 and hnRNP A2 in vivo.Nucleic Acids Res. 2009; 37: 4116-4126Crossref Scopus (158) Google Scholar]. Equally important is FUS/TLS, which, together with TDP-43, regulates histone deacetylase (HDAC)6 expression levels. However, unlike the hnRNP–TDP-43 interaction, the FUS/TLS–TDP-43 interaction seems to be restricted to only a minority (10%) of cells, and it is greatly enhanced by mutant TDP-43 [32Kim S.H. et al.Amyotrophic lateral sclerosis-associated proteins TDP-43 and FUS/TLS function in a common biochemical complex to co-regulate HDAC6 mRNA.J. Biol. Chem. 2010; 285: 34097-34105Crossref PubMed Scopus (167) Google Scholar, 33Ling S.C. et al.ALS-associated mutations in TDP-43 increase its stability and promote TDP-43 complexes with FUS/TLS.Proc. Natl. Acad. Sci. U.S.A. 2010; 107: 13318-13323Crossref Scopus (305) Google Scholar]. Other important TDP-43-binding hRNPs are those that are implicated in stress granule formation, such as T cells restricted intracellular antigen-1 (TIA-1) [34Liu-Yesucevitz L. et al.Tar DNA binding protein-43 (TDP-43) associates with stress granules: analysis of cultured cells and pathological brain tissue.PLoS ONE. 2010; 5: e13250Crossref PubMed Scopus (419) Google Scholar]. Importantly, TDP-43 also forms homo-dimers and -multimers (a property common to many hnRNPs). Indeed, two recent reports have suggested that TDP-43 is able to dimerize in normal conditions [18Kuo P.H. et al.Structural insights into TDP-43 in nucleic-acid binding and domain interactions.Nucleic Acids Res. 2009; 37: 1799-1808Crossref PubMed Scopus (208) Google Scholar, 35Shiina Y. et al.TDP-43 dimerizes in human cells in culture.Cell. Mol. Neurobiol. 2010; 30: 641-652Crossref PubMed Scopus (63) Google Scholar]. In these studies, the domains involved in self-interaction were suggested to include RRM2 [18Kuo P.H. et al.Structural insights into TDP-43 in nucleic-acid binding and domain interactions.Nucleic Acids Res. 2009; 37: 1799-1808Crossref PubMed Scopus (208) Google Scholar] and the N-terminal sequence from residues 3 to 183 [35Shiina Y. et al.TDP-43 dimerizes in human cells in culture.Cell. Mol. Neurobiol. 2010; 30: 641-652Crossref PubMed Scopus (63) Google Scholar]. It was also initially noted that a Gln/Asn-rich region in the TDP-43 C-terminal tail could disrupt endogenous TDP-43-hnRNP A2 complexes, and possibly form high molecular weight complexes in vitro [31D'Ambrogio A. et al.Functional mapping of the interaction between TDP-43 and hnRNP A2 in vivo.Nucleic Acids Res. 2009; 37: 4116-4126Crossref Scopus (158) Google Scholar]. More recently it has become evident that the core sequence of this Gln/Asn-rich region, corresponding to residues 345 to 366, is essential for TDP-43 self-interaction [36Budini M. et al.Cellular model of TAR DNA binding protein 43 (TDP-43) aggregation based on its C-terminal Gln/Asn-rich region.J. Biol. Chem. 2012; 287: 7512-7525Crossref PubMed Scopus (85) Google Scholar]. This region appears to be particularly important for the aggregation properties of TDP-43 because introducing repeated units of this sequence within cell lines or primary neuron cultures induces aggregate formation that recapitulates many, but not all, of the characteristics of aggregates in patient cells [36Budini M. et al.Cellular model of TAR DNA binding protein 43 (TDP-43) aggregation based on its C-terminal Gln/Asn-rich region.J. Biol. Chem. 2012; 287: 7512-7525Crossref PubMed Scopus (85) Google Scholar]. It should also be noted, however, that the protein–protein interactions of TDP-43 are not limited to hnRNPs and even within this group we have not been exhaustive. In fact, proteomic studies performed on TDP-43 in several cell lines have detected a huge number of potential TDP-43 interacting partners. Some of these putative interactions have been at least somewhat studied, such as the Drosha complex, but many remain to be validated, especially at the functional level [33Ling S.C. et al.ALS-associated mutations in TDP-43 increase its stability and promote TDP-43 complexes with FUS/TLS.Proc. Natl. Acad. Sci. U.S.A. 2010; 107: 13318-13323Crossref Scopus (305) Google Scholar, 37Freibaum B.D. et al.Global analysis of TDP-43 interacting proteins reveals strong association with RNA splicing and translation machinery.J. Proteome Res. 2010; 9: 1104-1120Crossref PubMed Scopus (315) Google Scholar]. Taken together, these data suggest that TDP-43 is a very flexible protein in terms of its interactions with RNA and proteins. In the future, one of the critical issues will be to identify the key interactions that are primarily responsible for the phenomena observed in the brains of patients with neurodegenerative diseases. This will be extremely important for both the search for therapeutic options that target single splicing events or specific protein–protein interactions, and also the search for biomarkers to monitor disease progression. As mentioned, one of the fundamental characteristics of TDP-43 is its intrinsic propensity to aggregate [38Johnson B.S. et al.TDP-43 is intrinsically aggregation-prone, and amyotrophic lateral sclerosis-linked mutations accelerate aggregation and increase toxicity.J. Biol. Chem. 2009; 284: 20329-20339Crossref PubMed Scopus (483) Google Scholar, 39Buratti E. Baralle F.E. Multiple roles of TDP-43 in gene expression, splicing regulation, and human disease.Front. Biosci. 2008; 13: 867-878Crossref PubMed Scopus (365) Google Scholar]. It is now widely accepted that its C-terminal tail is responsible for most of its tendency to aggregate, even in the absence of particular cofactors or modifications, as recently demonstrated by physicochemical studies using synthetic peptides [40Saini A. Chauhan V.S. Delineation of the core aggregation sequences of TDP-43 C-terminal fragment.Chembiochem. 2011; 12: 2495-2501Crossref Scopus (62) Google Scholar, 41Chen A.K. et al.Induction of amyloid fibrils by the C-terminal fragments of TDP-43 in amyotrophic lateral sclerosis.J. Am. Chem. Soc. 2010; 132: 1186-1187Crossref Scopus (101) Google Scholar]. An intriguing hypothesis that has recently been proposed is that this region contains a potentially infectious 'prion domain' from residues 277 to 414 [42Cushman M. et al.Prion-like disorders: blurring the divide between transmissibility and infectivity.J. Cell Sci. 2010; 123: 1191-1201Crossref PubMed Scopus (229) Google Scholar]. This hypothesis has received some support from experimental evidence that in vitro prepared TDP-43 fibrils can be taken up by cultured cells and function to trigger intracellular aggregates [43Furukawa Y. et al.A seeding reaction recapitulates intracellular formation of sarkosyl-insoluble TAR DNA binding protein-43 inclusions.J. Biol. Chem. 2011; 286: 18664-18772Crossref PubMed Scopus (181) Google Scholar]. These studies suggest that propagation of these fibrils may act as a seeding reaction to allow cell-to-cell spreading. However, this observation needs further experimental confirmation. In recent years, several factors that affect the TDP-43 aggregation process have been described. As expected, they are either changes in the protein architecture itself (through amino acid substitutions or post-translational modifications) or the surrounding protein environment (Figure 3). Regarding TDP-43 modifications, most attention has focused on the potential role of the C-terminal fragments (CTFs) that are characteristically produced in the neurons of ALS and FTLD patients. Expression of CTFs of TDP-43 was shown to promote aggregation of TDP-43 in a considerable number of cell models [44Yang C. et al.The C-terminal TDP-43 fragments have a high aggregation propensity and harm neurons by a dominant-negative mechanism.PLoS ONE. 2010; 5: e15878Crossref PubMed Scopus (126) Google Scholar, 45Igaz L.M. et al.Expression of TDP-43 C-terminal fragments in vitro recapitulates pathological features of TDP-43 proteinopathies.J. Biol. Chem. 2009; 284: 8516-8524Crossref PubMed Scopus (269) Google Scholar, 46Zhang Y.J. et al.Aberrant cleavage of TDP-43 enhances aggregation and cellular toxicity.Proc. Natl. Acad. Sci. U.S.A. 2009; 106: 7607-7612Crossref PubMed Scopus (428) Google Scholar]. Most recently, de novo intranuclear cleavage of TDP-43, producing CTFs, has been shown to cause aggregation only in the presence of a 'second hit', such as impairment of dynein-mediated microtubule transport or loss of direct and indirect RNA interaction [47Pesiridis G.S. et al.A "two-hit" hypothesis for inclusion formation by carboxyl-terminal fragments of TDP-43 protein linked to RNA depletion and impaired microtubule-dependent transport.J. Biol. Chem. 2011; 286: 18845-18855Crossref PubMed Scopus (87) Google Scholar]. If confirmed, this finding would tend to downplay the physiological relevance of CTFs because they may not be a self-sufficient source of aggregation. In contrast with these aggregation-promoting effects, hyperphosphorylation of TDP-43 might be protective [48Li H.Y. et al.Hyperphosphorylation as a defense mechanism to reduce TDP-43 aggregation.PLoS ONE. 2011; 6: e23075Crossref Scopus (72) Google Scholar]. However, S409/410 phosphorylation has been shown to be required for toxicity in a C. elegans model [49Liachko N.F. et al.Phosphorylation promotes neurotoxicity in a Caenorhabditis elegans model of TDP-43 proteinopathy.J. Neurosci. 2010; 30: 16208-16219Crossref PubMed Scopus (147) Google Scholar] and phosphorylated fragments are degraded less efficiently [50Zhang Y.J. et al.Phosphorylation regulates proteasomal-mediated degradation and solubility of TAR DNA binding protein-43 C-terminal fragments.Mol. Neurodegener. 2010; 5: 33Crossref Scopus (53) Google Scholar], suggesting that further work is needed to clarify whether phosphorylation of TDP-43 is beneficial or detrimental to neurons. As recently discussed by Lee et al. [4Lee E.B. et al.Gains or losses: molecular mechanisms of TDP43-mediated neurodegeneration.Nat. Rev. Neurosci. 2012; 13: 38-50Google Scholar], the role of TDP-43 phosphorylation in disease is still very much an open question that remains to be elucidated. In addition, although aggregation of TDP-43 usually oc