Abstract

The aggregation and mislocalization of RNA-binding proteins leads to the aberrant regulation of RNA metabolism and is a key feature of many neurodegenerative diseases, including amyotrophic lateral sclerosis and frontotemporal dementia. However, the pathological consequences of abnormal deposition of TDP-43 and other RNA-binding proteins remain unclear, as the specific molecular events that drive neurodegeneration have been difficult to identify and continue to be elusive. Here, we provide novel insight into the complexity of the RNA-binding protein network by demonstrating that the inclusion of exon 17b in the SORT1 mRNA, a pathologically relevant splicing event known to be regulated by TDP-43, is also considerably affected by additional RNA-binding proteins, such as hnRNP L, PTB/nPTB and hnRNP A1/A2. Most importantly, the expression of hnRNP A1/A2 and PTB/nPTB is significantly altered in patients with frontotemporal dementia with TDP-43-positive inclusions (FTLD-TDP), indicating that perturbations in RNA metabolism and processing in FTLD-TDP are not exclusively driven by a loss of TDP-43 function. These results also suggest that a comprehensive assessment of the RNA-binding protein network will dramatically advance our current understanding of the role of TDP-43 in disease pathogenesis, as well as enhance both diagnostic and therapeutic capabilities.

Introduction

Following the discovery of TDP-43 within protein inclusions in patients with frontotemporal lobar degeneration (FTLD) and amyotrophic lateral sclerosis (ALS), the identification of pathogenic mechanisms triggered by TDP-43 aggregation has been a major research effort hindered by the massive number of RNA processing events regulated by TDP-43 (1,2). Similar to other RNA-binding proteins (RBPs) within the heterogeneous ribonuclear protein (hnRNP) family (3), most of the functional properties displayed by TDP-43 are mediated by binding to specific RNA target sequences (4–7). In addition, given the extensive TDP-43 interactome as indicated by several recent proteomic studies (8,9), including members of the hnRNP protein family, TDP-43 is able to regulate all steps of RNA maturation (transcription, processing, transport, stability and translation) (10). Therefore, the relative expression status of proteins within the TDP-43 interactome could significantly impact TDP-43 function, either through a direct interaction or by acting independently on the same cellular targets. This high degree of complexity may also explain the difficulty associated with pinpointing the specific misregulated events that can be directly linked to disease onset and progression in TDP-43 proteinopathies (11).

Recently, a specific mis-splicing event regulated by TDP-43 that involves mRNA processing of the neurotrophic receptor sortilin 1 (SORT1) within the brain of FTLD-TDP patients was described by us and others (4,12). SORT1 is primarily expressed in neurons and has been shown to regulate the intracellular trafficking of the neurotrophic factor progranulin (PGRN) (13,14). Given that mutations in the PGRN gene (GRN) lead to haploinsufficiency and are a genetic cause of FTLD-TDP, loss of PGRN is associated with disease (15–18), making the regulation of SORT1 expression essential to maintaining PGRN homeostasis. Under normal conditions, TDP-43 heavily represses the inclusion of a specific human SORT1 sequence, exon 17b, that carries a translational stop codon. Loss of TDP-43 promotes inclusion of this exon, leading to the production of a truncated and toxic SORT1 isoform that can bind but not internalize PGRN. Most importantly, the inclusion of this toxic isoform is elevated in FTLD-TDP patients, suggesting a possible link with disease (12). However, the fact that inclusion of exon 17b is not as heavily repressed in species that do not carry a translational stop codon, such as mouse or other rodents (12), indicates the evolution of additional mechanisms to prevent the production of this toxic SORT1 fragment in humans. As TDP-43 binds equally well to a conserved UG-rich region near the 3′ splice site (ss) of both human and mouse exon 17b sequences, the existence of other splicing regulatory elements must be present in the human exon 17b to ensure exclusion. In these studies, we identify additional hnRNP splicing factors that work in conjunction with TDP-43 to efficiently repress human exon 17b inclusion and demonstrate that expression of some of these factors is significantly altered in FTLD-TDP patients relative to controls. Overall, these findings indicate that expression of other hnRNP proteins strongly impacts the molecular consequences of decreased nuclear TDP-43 as a result of aggregation in disease.

Results

The splicing of exon 17b upon TDP-43 depletion is determined by the cell context

To study the alternative splicing regulation of human and mouse exon 17b sequences, we used two minigenes that contain either the human or mouse 17b exons flanked by 150 nucleotides of upstream and downstream introns (Fig. 1A). As previously described in human M17 and HeLa cells, TDP-43 binds to UG-repeat elements located near the 3′ss of exon 17b and represses inclusion (12). Here, we extended these analyses to evaluate exon 17b inclusion in a variety of human and mouse neuronal (SK-N-BE, SH-SY-5Y, NSC-34 and Neuro 2a) and non-neuronal cell types (Hep3B and N-muli). Following minigene transfection, the splicing pattern of exon 17b was analyzed by RT–PCR amplification, with specific primers generating either 246-bp and 345-bp bands for exon 17b exclusion and inclusion products, respectively (Fig. 1B and C). As shown in Figure 1, under normal conditions, low levels of exon 17b were detected in all cell lines tested, whether of human (Fig. 1B) or mouse origin (Fig. 1C). Of interest, the splicing pattern for mouse exon 17b was more variable than that of the human, with high levels of mouse exon 17b inclusion in the human neuronal cell lines (SK-N-BE, SH-SY-5Y), and very little inclusion in the human non-neuronal Hep3B cell line. These results are consistent with previous indications that inclusion of human exon 17b is more heavily repressed in neuronal cell lines than its mouse homolog.

Figure 1.

Variable effects of TDP-43 depletion in different cellular contexts. (A) Schematic presentation of the exon minigenes used in transfection experiments. The α-globin, fibronectin EDB and SORT1 exon17b (SORT1 Ex17b) are shown as black, white and gray boxes, respectively. The binding location of TDP-43 near the 3′ss of both human and mouse exons is also shown. Possible splicing events are indicated by single (exon inclusion) and dotted (exon skipping) lines. The sequences of human and mouse SORT1 Ex17b exons (bold uppercase letters) and their flanking introns (lowercase letters) are shown below. Nucleotide differences in the exonic regions between human and mouse are highlighted. (B and C) Basic splicing pattern and eventual effects of TDP-43 depletion on human or mouse SORT1 Ex17b inclusion in different human (B) and mouse (C) cell lines. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcripts (17b inclusion versus 17b skipping) is reported on the right of each agarose gel. Depletion of TDP-43 was determined by western blot assay, and tubulin was used as loading control.

Figure 1.

Variable effects of TDP-43 depletion in different cellular contexts. (A) Schematic presentation of the exon minigenes used in transfection experiments. The α-globin, fibronectin EDB and SORT1 exon17b (SORT1 Ex17b) are shown as black, white and gray boxes, respectively. The binding location of TDP-43 near the 3′ss of both human and mouse exons is also shown. Possible splicing events are indicated by single (exon inclusion) and dotted (exon skipping) lines. The sequences of human and mouse SORT1 Ex17b exons (bold uppercase letters) and their flanking introns (lowercase letters) are shown below. Nucleotide differences in the exonic regions between human and mouse are highlighted. (B and C) Basic splicing pattern and eventual effects of TDP-43 depletion on human or mouse SORT1 Ex17b inclusion in different human (B) and mouse (C) cell lines. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcripts (17b inclusion versus 17b skipping) is reported on the right of each agarose gel. Depletion of TDP-43 was determined by western blot assay, and tubulin was used as loading control.

To further define the role of TDP-43 in this critical splicing event (12), these analyses were carried out under TDP-43 knockdown conditions following siRNA treatment. Figure 1 shows that human exon 17b inclusion was only minimally affected by silencing of TDP-43 in most cell lines. The only exception was in Neuro2a cells (see Fig. 4 for quantification of this event), where a statistically significant increase in human exon 17b inclusion was observed. Conversely, TDP-43 silencing had a very positive effect on the inclusion levels of mouse exon 17b with just one exception, in the hepatocellular carcinoma-derived Hep3B cell line (Fig. 1B). Taken together, these data demonstrate that inclusion of the human exon 17b is much more heavily repressed than mouse exon 17b, a conclusion that is in agreement with our previous findings (12). However, the fact that mouse exon 17b inclusion varied considerably depending on the cell line tested (e.g. compare Hep3B and SH-SY-5Y cell lines, Fig. 1B) suggests that additional splicing regulatory factors can modulate this splicing event.

Identification of additional factors that differentially bind human and mouse exon 17b

In order to identify splicing factors that contribute to the regulatory action of TDP-43 on exon 17b splicing, we used a previously described affinity pulldown system that allows the identification of the binding location of such factors in the exon 17b context (12,19–21). The human and mouse exon 17b regions were first split into upstream intron, exon and downstream intron, and these sequences were transcribed separately (Fig. 2). Affinity RBP purification was performed using RNA-coated adipic acid dehydrazide beads, and the ability of the following hnRNP proteins to bind to these regions was tested by western blot: TDP-43, hnRNP H, hnRNP F, PTB (hnRNP I), hnRNP A1, hnRNP A2, DAZAP1, hnRNP L and hnRNP C1/C2. Many of these hnRNP proteins were previously identified as binding factors into the exon 17b context (TDP-43, hnRNP A1, hnRNP A2), as described in our previous publication (12). In addition, we also included the following hnRNPs: (1) hnRNP H/F, due to the fact that the high number of G-repeats in the upstream introns of mouse and human exon 17b represent well-known binding sites for this factor (22), (2) PTB (hnRNP I) since several known binding ‘cucc’ sequences are present in introns and in the 17b exon (23) and (3) hnRNP L, which may bind to ca-repeats located in the 17b exonic region (24). Although there are no well-known binding sequences for Dazap1, this protein was chosen because it preferentially binds downstream of 5′splice sites and has been described to act in concert with hnRNP A/Bs to regulate alternative splicing (25). Finally, hnRNP C was also tested because it is a very abundant hnRNP protein that binds u-rich sequences (26).

Figure 2.

Evaluation of binding efficiency of diverse hnRNPs to human and mouse exon 17b. (A and B) The binding efficiency of several hnRNP proteins was assessed by western blot after pulldown of in vitro-transcribed RNAs from human (A) and mouse (B) SORT1 Ex17b exons and their flanking introns.

Figure 2.

Evaluation of binding efficiency of diverse hnRNPs to human and mouse exon 17b. (A and B) The binding efficiency of several hnRNP proteins was assessed by western blot after pulldown of in vitro-transcribed RNAs from human (A) and mouse (B) SORT1 Ex17b exons and their flanking introns.

As shown in Figure 2, affinity purification showed that TDP-43 binds to the upstream intronic region of both human and mouse exon 17b sequences. This result is congruent with the presence of a UG-rich region in the intron upstream of exon 17b, as TDP-43 presents high binding affinity to such regions (7,27). Additionally, other hnRNP proteins displayed similar binding patterns to both human and mouse exon 17b contexts (hnRNP H, hnRNP F, hnRNP A1 and hnRNP L) or no binding (hnRNP C1/C2). This was expected based on the high sequence similarity between these human and mouse regions, especially within the exon (see Fig. 1A). However, hnRNP A2 and PTB/nPTB exhibited striking differences in the binding patterns between human and mouse exon 17b contexts. For example, hnRNP A2 showed binding within exon 17b in the human but not in the mouse context, while PTB showed binding to human upstream and downstream introns to a greater extent than the corresponding mouse intronic regions (Fig. 2).

Binding of hnRNP A2 to human 17b exon correlates with splicing repression

To understand the differential binding preference of splicing repressors for the human versus mouse contexts, we took advantage of a series of human-mouse hybrid minigenes generated in our previous study (12) in which the eight nucleotide differences existing between the human and mouse 17b exons were selectively mutated (Fig. 3A). Transfection of these minigenes in Neuro2a cells replicated the results previously observed in HeLa cells (12), with increased inclusion of exon 17b in mutants 1 h-to-m and 4 h-to-m (Fig. 3B). To determine whether these nucleotide changes would also affect the occupancy of hnRNP factors that exhibit differential binding to the mouse and human exon 17b contexts, we performed pulldown analyses on the human-mouse hybrid minigenes depicted in Figure 3A and looked for protein occupancy of the hnRNP factors hnRNP A2, PTB and hnRNP L. Interestingly, binding of hnRNP A2 was completely abolished in both the 1 h-to-m and 4 h-to-m mutants (Fig. 3C), which displayed the highest levels of exon inclusion (Fig. 3B). Conversely, hnRNP A2 was still able to bind the remaining 2 h-to-m and 3 h-to-m mutants, which showed similar levels of exon inclusion to the wild-type human sequence (Fig. 3B and C). These findings indicate that binding of hnRNP A2 is critical to repress the inclusion of human exon 17b. In addition, minor changes were observed in hnRNP L and PTB binding to the mutant versus wild-type sequences (Fig. 3C), as expected given that both hnRNP L and PTB proteins displayed similar binding to both the mouse and human exons (Fig. 2). While changes in PTB binding do not appear to correlate perfectly with splicing, it should be noted that in the h1-h4 human-mouse hybrid minigenes, PTB binding will still occur in the downstream human intron (Fig. 2). This may eventually affect the splicing by multimeric assembly of PTB complexes, as previously described (28), in ways that are not easy to predict.

Figure 3.

hnRNP A2 binding to the human 17b exon correlates with splicing repression. (A) Schematic diagram of the human to mouse (h to m) hybrid constructs based on the human Ex17b sequence (B) Transfection of the various h-to-m constructs in Neuro2a cells. The RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcripts (17b inclusion versus 17b skipping) is reported on the left of the agarose gel. The quantified percentage of SORT1 Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). *P < 0.05, **P < 0.01 (n = 3). (C) In vitro-transcribed RNA from wild-type and mutant exonic sequences were incubated with human HeLa nuclear protein extracts to assess their ability to bind the following proteins: hnRNP A2, hnRNP L and PTB.

Figure 3.

hnRNP A2 binding to the human 17b exon correlates with splicing repression. (A) Schematic diagram of the human to mouse (h to m) hybrid constructs based on the human Ex17b sequence (B) Transfection of the various h-to-m constructs in Neuro2a cells. The RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcripts (17b inclusion versus 17b skipping) is reported on the left of the agarose gel. The quantified percentage of SORT1 Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). *P < 0.05, **P < 0.01 (n = 3). (C) In vitro-transcribed RNA from wild-type and mutant exonic sequences were incubated with human HeLa nuclear protein extracts to assess their ability to bind the following proteins: hnRNP A2, hnRNP L and PTB.

hnRNP proteins work co-operatively to repress exon 17b inclusion

To dissect the individual and collaborative contributions of different hnRNP proteins in the exon 17b splicing event, we evaluated levels of exon 17b inclusion following single or double knock down of various hnRNP proteins. While TDP-43 knock down alone was capable of increasing exon 17 inclusion in both human and mouse minigenes (Fig. 4A), single knock down of hnRNP A1/A2, hnRNP L, PTB/nPTB, hnRNP H, hnRNP F and DAZAP1 (Supplementary Material, Fig. S1) had very minimal effect on basal levels of either human or mouse exon 17b inclusion. These results confirm that TDP-43 is indeed the principal repressor of both human and mouse 17b exons.

Figure 4.

Cooperation of TDP-43 and hnRNP L or hnRNP A2 to repress SORT1 Ex17b inclusion in the human and mouse minigene systems. (A) TDP-43 knockdown, alone or simultaneously with hnRNP A1/A2 (B), hnRNP L (C), PTB/nPTB (D), hnRNP H/F (E) or DAZAP1 (F) in Neuro 2a cell line. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of each agarose gel. Luciferase siRNA-treated cells have been used as a control and depletion efficiency was determined by western blot assay (extracts have been normalized using tubulin) (G and H). The quantified percentages of human (G) and mouse (H) Ex17b inclusion were determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey) *P < 0.05, **P < 0.01 (n = 3).

Figure 4.

Cooperation of TDP-43 and hnRNP L or hnRNP A2 to repress SORT1 Ex17b inclusion in the human and mouse minigene systems. (A) TDP-43 knockdown, alone or simultaneously with hnRNP A1/A2 (B), hnRNP L (C), PTB/nPTB (D), hnRNP H/F (E) or DAZAP1 (F) in Neuro 2a cell line. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of each agarose gel. Luciferase siRNA-treated cells have been used as a control and depletion efficiency was determined by western blot assay (extracts have been normalized using tubulin) (G and H). The quantified percentages of human (G) and mouse (H) Ex17b inclusion were determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey) *P < 0.05, **P < 0.01 (n = 3).

Next, to investigate the potential co-operative repressive role of hnRNP proteins and TDP-43, we performed a series of hnRNP knock down experiments in conjunction with TDP-43 knock down. Interestingly, knockdown of both hnRNP A1/A2 and TDP-43 significantly increased the level of human and mouse exon 17b inclusion (Fig. 4B). This result is consistent with the presence of a silencer element bound by hnRNP A2 in the human exon 17b that upon mutation abolishes the binding of this protein (Fig. 3). In addition, while no statistically significant increases were observed when TDP-43 was knocked down together with PTB/nPTB (Fig. 4D), hnRNP H/F (Fig. 4E) or Dazap1 (Fig. 4F), we detected a significant increase in exon 17b inclusion when both TDP-43 and hnRNP L were knocked down (Fig. 4C). Based on this finding, we then evaluated whether hnRNP L knock down would achieve similar results when a different hnRNP protein was concomitantly knocked down. We observed a modest increase of human exon 17b inclusion following double knock down of hnRNP L and hnRNP A1/A2 (Fig. 5A and D), but not when hnRNP L was knocked down with hnRNP PTB/nPTB (Fig. 5B and D). Conversely for mouse exon 17b inclusion, a statistically significant increase was found following hnRNP L and hnRNP PTB/nPTB knock down (Fig. 5B and E), while no effect was observed when hnRNP L was knocked down with hnRNP A1/A2 (Fig. 5A and E). However, for both mouse and human exon 17b inclusion, no effect was observed when hnRNP A1/A2 was knocked down together with hnRNP PTB/nPTB (Fig. 5C–E).

Figure 5.

Effects on human and mouse exon 17b recognition following knock down of hnRNPs in the presence of TDP-43. The following knockdowns were performed: hnRNP L together with hnRNP A1/A2 (A), PTB/nPTB (B) or PTB/nPTB with hnRNP A1/A2 in Neuro 2a cell line. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of each agarose gel. Luciferase siRNA-treated cells have been used as a control, and depletion efficiency was determined by western blot assay (extracts were normalized using tubulin). (D and E) The quantified percentage of human (D) and mouse (E). Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey) *P < 0.05 (n = 3).

Figure 5.

Effects on human and mouse exon 17b recognition following knock down of hnRNPs in the presence of TDP-43. The following knockdowns were performed: hnRNP L together with hnRNP A1/A2 (A), PTB/nPTB (B) or PTB/nPTB with hnRNP A1/A2 in Neuro 2a cell line. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of each agarose gel. Luciferase siRNA-treated cells have been used as a control, and depletion efficiency was determined by western blot assay (extracts were normalized using tubulin). (D and E) The quantified percentage of human (D) and mouse (E). Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey) *P < 0.05 (n = 3).

In order to verify that the effects observed when utilizing the minigene system resemble those happening endogenously, we also analyzed exon 17b inclusion of endogenous mouse SORT1 (Fig. 6). Similarly to the results obtained with the minigene system (Fig. 4), knock down of either TDP-43 alone, or in conjunction with hnRNP L or hnRNP A1/A2, was the most efficient combinations in promoting exon 17b inclusion. In conclusion, these findings demonstrate that while TDP-43 plays a central role in repressing human exon 17b inclusion, additional factors within the hnRNP network function with TDP-43 to regulate this critical splicing event.

Figure 6.

Effects of knocking down diverse hnRNPs on endogenous mouse Sort1 Ex17b in Neuro 2a cells cDNA prepared from the extracted RNA were subjected to PCR using primers in exons 17 and 19 of mouse Sort1. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of the agarose gel. The quantified percentage of Sort1 Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). **P < 0.01, ***P < 0.001 (n = 3).

Figure 6.

Effects of knocking down diverse hnRNPs on endogenous mouse Sort1 Ex17b in Neuro 2a cells cDNA prepared from the extracted RNA were subjected to PCR using primers in exons 17 and 19 of mouse Sort1. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of the agarose gel. The quantified percentage of Sort1 Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). **P < 0.01, ***P < 0.001 (n = 3).

Identification of a splicing enhancer of exon 17b in the mouse upstream intron

We previously identified two splicing enhancer factors, SRp40/SRSF5 and Tra2β, which selectively bound to the mouse, but not human, exon 17b context (12). To determine the exact binding location for both splicing factors, we performed the same pulldown analysis described in Figure 2. Western blot analyses using antibodies against SRp40/SRSF5 and Tra2β proteins demonstrate that both factors bind the intron upstream of mouse exon 17b with high efficiency (Fig. 7A). Note, no binding to either human or mouse exon 17b context was found for SRSF1, a negative control. To further pinpoint the region within the upstream intron of mouse exon 17b where these splicing enhancers bind, we created three human to mouse mutants in which we replaced the 150-bp human intron with selected mouse intronic regions: mutant H1 (1–43), mutant H2 (43–111) and mutant H3 (110–150) (Fig. 7B). Mutants were then transfected in Neuro 2a cells and TDP-43 was knocked down by siRNA treatment. As shown in Figure 7C, depletion of TDP-43 resulted in significantly increased exon 17b inclusion in the H2 mutant, suggesting that the 43–111 nucleotide replacement had introduced an enhancer element in the human context.

Figure 7.

Identification of a splicing enhancer of Ex17b in the mouse upstream intron. (A) Pulldown analysis performed for human and mouse Ex17b exons and their flanking introns (Up. and Down Int.). Bound proteins were then tested for binding of the following factors: SRp40/SRSF5, Tra2β and SRSF1/(ASF/SF2). (B) Schematic representation of hybrid human Ex17b minigenes constructed to assess the importance of mouse upstream intron in inclusion of Ex17b. (C) Hybrid constructs were expressed in Neuro 2a cells after TDP-43 knockdown. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of the agarose gel. Luciferase siRNA treated cells were used as a control and correct depletion was determined by western blot assay (extracts were normalized using tubulin). (D) The quantified percentage of SORT1 Ex17b inclusion for each construct was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). *P < 0.05, **P < 0.01 (n = 3).

Figure 7.

Identification of a splicing enhancer of Ex17b in the mouse upstream intron. (A) Pulldown analysis performed for human and mouse Ex17b exons and their flanking introns (Up. and Down Int.). Bound proteins were then tested for binding of the following factors: SRp40/SRSF5, Tra2β and SRSF1/(ASF/SF2). (B) Schematic representation of hybrid human Ex17b minigenes constructed to assess the importance of mouse upstream intron in inclusion of Ex17b. (C) Hybrid constructs were expressed in Neuro 2a cells after TDP-43 knockdown. RT–PCR samples are stained with ethidium bromide and run on a 1.5% agarose gel. A schematic diagram of the two minigene spliced transcript (17b inclusion versus 17b skipping) is reported on the right of the agarose gel. Luciferase siRNA treated cells were used as a control and correct depletion was determined by western blot assay (extracts were normalized using tubulin). (D) The quantified percentage of SORT1 Ex17b inclusion for each construct was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). *P < 0.05, **P < 0.01 (n = 3).

The presence of SR-binding splicing enhancer elements in intronic sequences is a rather rare but a previously described phenomenon for some genes, such as β-tropomyosin exon 6A (29), CT/CGRP gene exon 4 (30) and Trα2 (31). While in these genes the enhancer sequences are present in the downstream intron of the affected exonic region, in Sort1 the enhancer sequence is localized in the upstream intron. Therefore, in order to better define the properties of this upstream enhancer element, we overexpressed the SRp40/SRSF5 and Tra2β proteins in Neuro2a cells and determined their effect on the inclusion of the endogenous mouse exon 17b. This was performed both under normal conditions (Supplementary Material, Fig. S2A) and following TDP-43 siRNA knockout (Supplementary Material, Fig. S2B). Interestingly, SRp40/SRSF5 had the expected effect of promoting mouse exon 17b inclusion. However, this occurred only in the absence of TDP-43. In contrast, overexpression of Tra2β had no effects on inclusion in both conditions. Moreover, no effect could be observed following overexpression of other SR factors (SRp75/SRSF4 and SC-35/SRSF2), which do not bind within the exon 17b context (12) and were used as negative controls. Finally, to further clarify whether Tra2β can play a role in exon 17b splicing, we silenced this factor in Neuro2a cells. As shown, in Supplementary Material, Figure S3, silencing of Tra2α/β could successfully lower endogenous mouse exon 17b inclusion in the absence of TDP-43.

All these findings are summarized in the schematic diagram shown in Figure 8, in which the binding location and effect on exon 17b inclusion is depicted by font colour for the various hnRNP factors (no effect illustrated by black font, while positive or negative influence is depicted by either green or red font, respectively).

Figure 8.

Comparison of the splicing regulatory proteins binding to the human and mouse Ex17b exons. Schematic models of the human and mouse sortilin Ex17b exons and the splicing regulatory proteins (hnRNPs and SR proteins) binding to Ex17b exons and their flanking introns, as determined by our RNA-protein pulldown assays and functional analysis. Negative factors are marked in red font while positive factors are also marked in green font.

Figure 8.

Comparison of the splicing regulatory proteins binding to the human and mouse Ex17b exons. Schematic models of the human and mouse sortilin Ex17b exons and the splicing regulatory proteins (hnRNPs and SR proteins) binding to Ex17b exons and their flanking introns, as determined by our RNA-protein pulldown assays and functional analysis. Negative factors are marked in red font while positive factors are also marked in green font.

Altered expression of hnRNP proteins in the brain of FTLD-TDP patients

In order to determine whether a change in hnRNP protein expression might be associated with FTLD-TDP, we evaluated expression levels in the temporal cortex, a brain region in which we previously observed a significant increase in exon 17b inclusion in FTLD-TDP patients (12). Using quantitative RT–PCR (qRT-PCR), we detected a significant increase in the splicing repressors HNRNP A2B1 (hnRNP A1/A2) and PTBP1 (PTB) in FTLD-TDP patients relative to controls (Fig. 9). In addition, consistent with previous reports, no significant differences were detected in TARDBP expression levels between FTLD-TDP cases and controls. Taken together, these findings indicate that increased expression of the splicing repressors hnRNP A1/A2 and PTB may represent a compensatory response to the loss of nuclear TDP-43 function as a result of aggregation in disease.

Figure 9.

The RNA levels of several hnRNP proteins are altered in the brain of FTLD-TDP patients. Transcript levels of several hnRNPs measured by qRT-PCR in human FTLD-TDP cases (n = 20) and age-matched controls (n = 19) and normalized to GAPDH RNA levels. Statistical differences were assessed by Student's t-test (*P < 0.05, **P < 0.005).

Figure 9.

The RNA levels of several hnRNP proteins are altered in the brain of FTLD-TDP patients. Transcript levels of several hnRNPs measured by qRT-PCR in human FTLD-TDP cases (n = 20) and age-matched controls (n = 19) and normalized to GAPDH RNA levels. Statistical differences were assessed by Student's t-test (*P < 0.05, **P < 0.005).

Discussion

The eukaryotic genome encodes hundreds of RBPs that jointly coordinate and regulate all aspects of the life cycle of every RNA produced within cells, including transcription, capping, pre-mRNA splicing, editing, polyadenylation, transport, stability and translation (32). Among all these different events, the regulation of pre-mRNA splicing often involves an intricate network of RBPs that bind to particular sequences near or within the exon to be regulated (33,34). The concentration of these factors within different cells is one of the key elements that will determine whether a particular sequence is eventually included in the mature mRNA. Thus, given the importance of the RBP network in maintaining overall RNA processing and metabolism, it is intriguing that mutations or altered expression of RBPs are increasingly being shown to play an active role as causative agents of Mendelian disease in humans (35). In neurodegenerative diseases, insight into the role(s) played by RBPs in determining disease onset and progression has been provided by the discovery that alterations in genes involved in RNA/DNA metabolism are associated with FTD and ALS (36). In addition, alterations in RNA metabolism caused by misregulation of RBP expression are considered to be very damaging to neuronal function as described in recent reviews (37–41) and observational studies (42–45) on this subject. Therefore, the combined study of RBPs and RNA metabolism is expected to advance the field of neurodegenerative diseases in the near future (46).

Regarding this topic, one specific issue that needs to be addressed is how various hnRNP proteins may differentially contribute to neurodegeneration. In this work, we show that aberrant splicing of human exon 17b in the SORT1 gene is dependent on both TDP-43 as well as a number of other hnRNP proteins (hnRNP A2, hnRNP L, PTB), whose expression levels can vary considerably within FTLD-TDP and control individuals. These findings indicate that in order to produce this toxic exon 17b-containing isoform, not only TDP-43 needs to be removed from the nuclear environment through aggregation but also the relative concentration levels of these other repressive factors need to be decreased. Interestingly, the levels of both hnRNP A2 and PTB/nPTB are significantly upregulated in FTLD-TDP patients with respect to controls, suggesting the activation of in vivo compensatory mechanisms which lead to an increase in the expression of these splicing repressors to try and counteract the loss of TDP-43 function. This situation is in keeping with recent data demonstrating that overexpression of hnRNP U and hnRNP A1/A2 can inhibit TDP-43-induced neuronal cell death in NSC34 cells (41). Similarly in Drosophila, the ability of TDP-43 to suppress CGG RNA-associated phenotypes is dependent on its ability to interact with two fly homologs of hnRNP A2/B1, and overexpression of these factors could attenuate TDP43-induced cell death (41,47,48). Thus, our findings support this general view and provide a specific example through which overexpression of hnRNP proteins may represent a general defense mechanism following the loss through aggregation of RBPs, such as TDP-43. A key question that will need to be determined in future studies is the mechanism through which this increase in hnRNP A/B and PTB expression occurs. The fact that CLIP analyses have not identified these proteins as mRNA targets of TDP-43 (4,6), with the only exception possibly represented by the hnRNP A/B transcript according to RIP-seq (49), suggests that TDP-43 may not be able to directly repress the transcription of these factors. However, these results do not exclude the possibility that loss of TDP-43 may still lead to alterations in other factors that influence the transcription or stability of hnRNP A/B and PTB mRNAs. Another possibility stems from the fact that TDP-43 has been described to be part of the Drosha complex and negatively alter the expression of several miRNAs, recently reviewed by Volonte et al. (50), which might repress the expression of these other hnRNP factors.

The results presented in the current study could also explain differences in disease onset and progression among patients. In fact, it is quite conceivable that the presence of TDP-43 aggregation in FTLD-TDP patients will lead to misregulation of many events controlled by this protein. However, as shown by our exon 17b results, the extent to which these events will be misregulated also depends on changes in the relative expression levels of other hnRNPs. In this respect, it is also interesting to note that ALS patients have been described to present loss of hnRNP A1 expression in motor neurons that is concomitant with TDP-43 cytoplasmic inclusions, which may represent an additional disturbance in the control of RNA metabolism within these cells (51). As a result, different hnRNP expression levels, which may depend on individual or age-related differences, on top of similar TDP-43 pathologies could potentially account for a different disease course, with regards both to onset and progression. Therefore, a better knowledge of all these eventual connections is expected to have important implications in our understanding of TDP-43 proteinopathies and represents a very promising area of research for future studies given the therapeutic implications.

Materials and Methods

Hybrid minigene constructs

Human and mouse SORT1 exon 17b minigene constructs (hEx17b, mEx17b, 1 h to m, 2 h to m, 3 h to m, 4 h to m) have been previously described (12). Further modifications of human SORT1 Ex17b upstream intron (H1, H2, H3) were generated by PCR-directed mutagenesis with specific primers using human Ex17b plasmids as templates. The products were subcloned inside the NdeI restriction site of the pTB plasmid. Primer sequences for the making of each mutant can be provided upon request.

Cell culture; silencing and transfection

All human (SK-N-BE and SH-SY-5Y, and Hep3B) and mouse (NSC-34, Neuro 2a and N-muli) cell lines were cultured in DMEM–Glutamax-I (GIBCO) supplemented with 10% fetal bovine serum (GIBCO) and Antibiotic-Antimycotic-stabilized suspension (Sigma) in standard conditions. The siRNA sense sequences from Sigma used for silencing the different target proteins were the following: human TDP-43: gcaaagccaagaugagccu, mouse TDP-43: cgaugaacccauugaaaua, hnRNP-A1: cagcugaggaagcucuuca, hnRNP-A2: ggaacaguuccguaagcuc, PTB-P1: aacuuccaucauuccagagaa, PTB-N1: gagaggaucugacgaacua, hnRNP L: uauggcuuggaucaaucua, Tra2α: ggaucuucgugaaguauuu, Tra2β: ggaggauacagaucacguu, Luciferase (control): uaaggcuaugaagagauac, from Darmacon: Dazap1: gagacucugcgcagcuacu and hnRNP H: ggaaauagcugaaaagcu, and from Ambion: hnRNP F (the sequence is not provided). Each siRNA was transfected according to manufacturer's instructions. One day after silencing, cells were transfected by the minigenes mainly using Effectene (Qiagen) or by Lipofectamine 2000 (Invitrogen) in the case of mouse neuroblastoma cell lines (Neuro 2a and NSC-34) following the manufacturer's instructions. For overexpression experiments, the relevant plasmids [(SRp75/SRSF4, SC-35/SRSF2), Tra2β and SRp40/SRSF5] were transfected in Neuro 2a cells using Lipofectamine 2000 (Invitrogen) following the manufacturer's instructions. After 20 h post-transfection, cells were collected and divided into two parts: from the first part total-protein extracts were obtained by cell sonication in lysis buffer (1xPBS and 1× Protease inhibitor cocktail, Roche) in order to confirm the gene knockdown efficacy. Tubulin was used as total protein loading control. The other half part of the cells was used for RT–PCR analysis.

RNA extraction and RT–PCR analysis

RNA was obtained using Eurogold Trifast (Euroclone), as per the manufacturer's instructions. One microgram of total RNA was used in the retrotranscription reaction with random primers and Moloney murine leukemia virus (M-MLV) Reverse Transcriptase (Invitrogen). Spliced products from the transfected minigene were obtained using primers Bra2 5′ taggatccggtcaccaggaagttggttaaatca 3′ and α 2–3 5′ caacttcaagctcctaagccactgc 3′. PCR conditions were the following: 94°C for 2 min; 94°C for 30 s, 56°C for 1 min and 72°C for 45 s for 35 cycles and 72°C for 10 min for the final extension. PCRs were optimized to be in the exponential phase of amplification, and products were routinely fractionated in 1.5% (wt/vol) agarose gels. Moreover, the PCR cycling conditions described above and used to detect spliced products from the transfected minigene have been also used to detect endogenous mouse SORT1 product in Neuro 2a cell line using two other primers including: 5′ caaatgccaaggtgggatgaa 3′ and 5′ ttgaatccaaagcctctacgcc 3′ on exons 17 and 19, respectively.

In vitro transcription

RNA templates have been obtained in two different ways. For the experiments reported in Figures 2 and 6, the sequence of interest was cloned in pBluescript KS+ plasmid (Stratagene) that contains a T7 promoter. In this case, the 5′ end of the sequence to be transcribed is placed as close as possible to the end of the T7 promoter, and a restriction enzyme site for BamH1 is present at its 3′ end to linearize the plasmid. This is to minimize the length of plasmid-related RNA that will eventually be transcribed together with the sequence of interest. On the other hand, for Figure 3, the h-to-m mutants RNA was transcribed from PCR templates amplified from the respective plasmids. A T7 promoter sequence was added towards the 5′ end of the template using primers that carried a T7 sequence. For transcription, 5 µg of DNA template was used in a 60 µl T7 polymerase (Stratagene) transcription reaction for each sample. The synthesized RNA was then purified using standard Acid–Phenol purification method followed by Ethanol precipitation.

Pulldown experiments and western blot analysis

A total of 15 µg of transcribed RNA was oxidized in the dark for an hour with sodium m-periodate in a 400 µl reaction mixture (100 mm sodium acetate pH 5.2 and 5 mm sodium m-periodate). RNA was then ethanol precipitated and resuspended in 200 µl of 100 mm sodium acetate. Approximately 350 µl of prewashed equilibrated adipic acid dehydrazide-agarose beads (50% slurry; Sigma) were added to each oxidized RNA volume and placed in the rotor at 4°C for overnight incubation. This RNA-Bead covalent link formation was also performed in the dark.

The beads with the bound RNA were then pelleted 5′ 4000 rpm, washed two times with 1 ml of 2 m NaCl and equilibrated in washing buffer (20 mm HEPES pH = 7.5, 100 mm KCl, 0.2 mm EDTA, 0.5 mm DTT, 6% glycerol). They were then incubated on a rotator with ∼1 mg of HeLa cell nuclear extract (Cilbiotech) for 30 min at room temperature, in 400 µl final volume in a 1× binding buffer (200 mm HEPES pH = 7.5, 2 mm EDTA, 5 mm DTT, 60% glycerol, this is 10× binding buffer), plus 100 mm KCl, heparin was added to a final concentration of 5 mg/ml. The beads were then pelleted by centrifugation at 4000 rpm for 5 min at room temperature and washed four times with 1.5 ml of washing buffer, before addition of SDS sample buffer and loading on a 10% SDS–PAGE gel.

The gel was then electroblotted onto a Nitrocellulose blotting membrane according to standard protocols (Amersham Biosciences) and blocked with 5% skimmed milk (non-fat dry milk in 1× PBS and 0.1% Tween-20). Membranes targeted for SR protein recognition were blocked using western blocking reagent (Roche). Proteins were probed with different antibodies and detected with a chemiluminescence kit (ECL; Pierce Biotechnology).

Primary antibodies against TDP-43, hnRNP H, hnRNP F, hnRNP A1, hnRNP A2, hnRNP C1/C2, Dazap1, PTB and tubulin were already available in-house. The hnRNP L antibody was purchased from Abcam, SRp40/SRSF5 and SRSF2 antibodies were obtained from Invitrogen, and anti-Tra2β was a generous gift from G. Dreyfuss and I.C Eperon. Rabbit and mouse secondary antibodies were purchased from Dako.

Human sample collection and qRT-PCR analysis

RNA was available from our previous study (12) and was used to determine the mRNA levels of indicated RBPs by qRT-PCR, following previously described procedures (12). Briefly, cDNA (diluted 1:40) from temporal cortex of 20 FTLD-TDP and 19 age-matched controls (refer to (12) for detailed list of cases) was used for qRT-PCR in an HT7900 analyzer (Applied Biosystems). Each 5 µl reaction contained: 2 µl of cDNA diluted 1:40, 0.5 µl primer mix (1 µm for each primer) and 2.5 µl SYBR green (Invitrogen). The qRT-PCR program was as follows: 50°C 2 min, 95°C 2 min, 40 cycles of 95°C 15 s and 60°C 1 min. mRNA values were normalized to either GAPDH or RPLP0 values, endogenous transcript controls, resulting in similar values. Note mRNA levels of the different transcripts presented in Figure 9 were normalized to GAPDH. The sequences of the primer pairs used for qRT-PCR are as follows: hnRNP A2B1 5′ tttggggatggctataatgg 3′ and 5′ ccataaccggggctacct 3′; DAZAP1 5′ ccagacttcccctatggtca 3′ and 5′ gcctgaccataagactgtaaacaa 3′; hnRNP L 5′ agcggctcaagactgacaa 3′ and 5′ gtgcgggtcatcgtagttct 3′; PTBP1 5′ atgacaagagccgtgactacac 3′ and 5′ gagaggctgagattataccaggtg 3′, GAPDH 5′ gttcgacagtcagccgcatc 3′ and 5′ ggaatttgccatgggtgga 3′, RPLP0 5′ tctacaaccctgaagtgcttgat 3′ and 5′ caatctgcagacagacactgg 3′.

Statistical analysis

The quantified percentage of Sort1 Ex17b inclusion was determined by densitometric analysis using ImageJ software. Statistical differences were calculated by one-way ANOVA test (post-test Tukey). P<0.05 was considered significant (n = 3). Statistical differences in transcript levels between FTLD-TDP cases (n = 20) and controls (n = 19) were calculated by Student's t-tests. P < 0.05 was considered significant (*P < 0.05, **P < 0.005).

Supplementary Material

Supplementary Material is available at HMG online.

Funding

This work was supported by Thierry Latran Fondation (REHNPALS) and the EU Joint Programme-Neurodegenerative Diseases JPND (RiMod-FTD, Italy, Ministero della Sanita’) to E.B., National Institutes of Health/National Institute of Neurological Disorders and Stroke (R01 NS077402, P01 NS084974) and National Institute of Environmental Health Services (R01 ES020395) to L.P. and The ALS Association (771DOO) to M.P.

Conflict of Interest statement. None declared.

References

1
Neumann
M.
,
Sampathu
D.M.
,
Kwong
L.K.
,
Truax
A.C.
,
Micsenyi
M.C.
,
Chou
T.T.
,
Bruce
J.
,
Schuck
T.
,
Grossman
M.
,
Clark
C.M.
et al
. (
2006
)
Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis
.
Science
 ,
314
,
130
133
.
2
Janssens
J.
,
Van Broeckhoven
C.
(
2013
)
Pathological mechanisms underlying TDP-43 driven neurodegeneration in FTLD-ALS spectrum disorders
.
Hum. Mol. Genet.
 ,
22
,
R77
R87
.
3
Dreyfuss
G.
,
Kim
V.N.
,
Kataoka
N.
(
2002
)
Messenger-RNA-binding proteins and the messages they carry
.
Nat. Rev. Mol. Cell Biol.
 ,
3
,
195
205
.
4
Polymenidou
M.
,
Lagier-Tourenne
C.
,
Hutt
K.R.
,
Huelga
S.C.
,
Moran
J.
,
Liang
T.Y.
,
Ling
S.C.
,
Sun
E.
,
Wancewicz
E.
,
Mazur
C.
et al
. (
2011
)
Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability from loss of TDP-43
.
Nat. Neurosci.
 ,
14
,
459
468
.
5
Lagier-Tourenne
C.
,
Polymenidou
M.
,
Hutt
K.R.
,
Vu
A.Q.
,
Baughn
M.
,
Huelga
S.C.
,
Clutario
K.M.
,
Ling
S.C.
,
Liang
T.Y.
,
Mazur
C.
et al
. (
2012
)
Divergent roles of ALS-linked proteins FUS/TLS and TDP-43 intersect in processing long pre-mRNAs
.
Nat. Neurosci.
 ,
15
,
1488
1497
.
6
Tollervey
J.R.
,
Curk
T.
,
Rogelj
B.
,
Briese
M.
,
Cereda
M.
,
Kayikci
M.
,
Konig
J.
,
Hortobagyi
T.
,
Nishimura
A.L.
,
Zupunski
V.
et al
. (
2011
)
Characterizing the RNA targets and position-dependent splicing regulation by TDP-43
.
Nat. Neurosci.
 ,
14
,
452
458
.
7
Buratti
E.
,
Baralle
F.E.
(
2001
)
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.
 ,
276
,
36337
36343
.
8
Ling
S.C.
,
Albuquerque
C.P.
,
Han
J.S.
,
Lagier-Tourenne
C.
,
Tokunaga
S.
,
Zhou
H.
,
Cleveland
D.W.
(
2010
)
ALS-associated mutations in TDP-43 increase its stability and promote TDP-43 complexes with FUS/TLS
.
Proc. Natl Acad. Sci. USA
 ,
107
,
13318
13323
.
9
Freibaum
B.D.
,
Chitta
R.K.
,
High
A.A.
,
Taylor
J.P.
(
2010
)
Global analysis of TDP-43 interacting proteins reveals strong association with RNA splicing and translation machinery
.
J. Proteome Res.
 ,
9
,
1104
1120
.
10
Buratti
E.
,
Baralle
F.E.
(
2012
)
TDP-43: gumming up neurons through protein-protein and protein-RNA interactions
.
Trends Biochem Sci.
 ,
37
,
237
247
.
11
Buratti
E.
,
Romano
M.
,
Baralle
F.E.
(
2013
)
TDP-43 high throughput screening analyses in neurodegeneration: advantages and pitfalls
.
Mol. Cell Neurosci.
 ,
56C
,
465
474
.
12
Prudencio
M.
,
Jansen-West
K.R.
,
Lee
W.C.
,
Gendron
T.F.
,
Zhang
Y.J.
,
Xu
Y.F.
,
Gass
J.
,
Stuani
C.
,
Stetler
C.
,
Rademakers
R.
et al
. (
2012
)
Misregulation of human sortilin splicing leads to the generation of a nonfunctional progranulin receptor
.
Proc. Natl Acad. Sci. USA
 ,
109
,
21510
21515
.
13
Hu
F.
,
Padukkavidana
T.
,
Vaegter
C.B.
,
Brady
O.A.
,
Zheng
Y.
,
Mackenzie
I.R.
,
Feldman
H.H.
,
Nykjaer
A.
,
Strittmatter
S.M.
(
2010
)
Sortilin-mediated endocytosis determines levels of the frontotemporal dementia protein, progranulin
.
Neuron
 ,
68
,
654
667
.
14
Carrasquillo
M.M.
,
Nicholson
A.M.
,
Finch
N.
,
Gibbs
J.R.
,
Baker
M.
,
Rutherford
N.J.
,
Hunter
T.A.
,
DeJesus-Hernandez
M.
,
Bisceglio
G.D.
,
Mackenzie
I.R.
et al
. (
2010
)
Genome-wide screen identifies rs646776 near sortilin as a regulator of progranulin levels in human plasma
.
Am. J. Hum. Genet.
 ,
87
,
890
897
.
15
Pickering-Brown
S.M.
(
2007
)
Progranulin and frontotemporal lobar degeneration
.
Acta Neuropathol.
 ,
114
,
39
47
.
16
Gass
J.
,
Cannon
A.
,
Mackenzie
I.R.
,
Boeve
B.
,
Baker
M.
,
Adamson
J.
,
Crook
R.
,
Melquist
S.
,
Kuntz
K.
,
Petersen
R.
et al
. (
2006
)
Mutations in progranulin are a major cause of ubiquitin-positive frontotemporal lobar degeneration
.
Hum. Mol. Genet.
 ,
15
,
2988
3001
.
17
Cruts
M.
,
Gijselinck
I.
,
van der Zee
J.
,
Engelborghs
S.
,
Wils
H.
,
Pirici
D.
,
Rademakers
R.
,
Vandenberghe
R.
,
Dermaut
B.
,
Martin
J.J.
et al
. (
2006
)
Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to chromosome 17q21
.
Nature
 ,
442
,
920
924
.
18
Baker
M.
,
Mackenzie
I.R.
,
Pickering-Brown
S.M.
,
Gass
J.
,
Rademakers
R.
,
Lindholm
C.
,
Snowden
J.
,
Adamson
J.
,
Sadovnick
A.D.
,
Rollinson
S.
et al
. (
2006
)
Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17
.
Nature
 ,
442
,
916
919
.
19
Buratti
E.
,
Baralle
M.
,
De Conti
L.
,
Baralle
D.
,
Romano
M.
,
Ayala
Y.M.
,
Baralle
F.E.
(
2004
)
hnRNP H binding at the 5′ splice site correlates with the pathological effect of two intronic mutations in the NF-1 and TSHbeta genes
.
Nucl
 . Acids Res.,
32
,
4224
4236
.
20
Buratti
E.
,
Dork
T.
,
Zuccato
E.
,
Pagani
F.
,
Romano
M.
,
Baralle
F.E.
(
2001
)
Nuclear factor TDP-43 and SR proteins promote in vitro and in vivo CFTR exon 9 skipping
.
Embo J.
 ,
20
,
1774
1784
.
21
Haque
A.
,
Buratti
E.
,
Baralle
F.E.
(
2010
)
Functional properties and evolutionary splicing constraints on a composite exonic regulatory element of splicing in CFTR exon 12
.
Nucl
 . Acids Res.,
38
,
647
659
.
22
Caputi
M.
,
Zahler
A.M.
(
2001
)
Determination of the RNA binding specificity of the heterogeneous nuclear ribonucleoprotein (hnRNP) H/H'/F/2H9 family
.
J. Biol. Chem.
 ,
276
,
43850
43859
.
23
Singh
R.
,
Valcarcel
J.
,
Green
M.R.
(
1995
)
Distinct binding specificities and functions of higher eukaryotic polypyrimidine tract-binding proteins
.
Science
 ,
268
,
1173
1176
.
24
Hui
J.
,
Stangl
K.
,
Lane
W.S.
,
Bindereif
A.
(
2003
)
HnRNP L stimulates splicing of the eNOS gene by binding to variable-length CA repeats
.
Nat. Struct. Biol.
 ,
10
,
33
37
.
25
Pastor
T.
,
Pagani
F.
(
2011
)
Interaction of hnRNPA1/A2 and DAZAP1 with an Alu-derived intronic splicing enhancer regulates ATM aberrant splicing
.
PLoS One
 ,
6
,
e23349
.
26
Gorlach
M.
,
Burd
C.G.
,
Dreyfuss
G.
(
1994
)
The determinants of RNA-binding specificity of the heterogeneous nuclear ribonucleoprotein C proteins
.
J. Biol. Chem.
 ,
269
,
23074
23078
.
27
Lukavsky
P.J.
,
Daujotyte
D.
,
Tollervey
J.R.
,
Ule
J.
,
Stuani
C.
,
Buratti
E.
,
Baralle
F.E.
,
Damberger
F.F.
,
Allain
F.H.
(
2013
)
Molecular basis of UG-rich RNA recognition by the human splicing factor TDP-43
.
Nat. Struct. Mol. Biol.
 ,
20
,
1443
1449
.
28
Spellman
R.
,
Smith
C.W.
(
2006
)
Novel modes of splicing repression by PTB
.
Trends Biochem. Sci.
 ,
31
,
73
76
.
29
Gallego
M.E.
,
Gattoni
R.
,
Stevenin
J.
,
Marie
J.
,
Expert-Bezancon
A.
(
1997
)
The SR splicing factors ASF/SF2 and SC35 have antagonistic effects on intronic enhancer-dependent splicing of the beta-tropomyosin alternative exon 6A
.
Embo J.
 ,
16
,
1772
1784
.
30
Lou
H.
,
Neugebauer
K.M.
,
Gagel
R.F.
,
Berget
S.M.
(
1998
)
Regulation of alternative polyadenylation by U1 snRNPs and SRp20
.
Mol. Cell Biol.
 ,
18
,
4977
4985
.
31
Hastings
M.L.
,
Wilson
C.M.
,
Munroe
S.H.
(
2001
)
A purine-rich intronic element enhances alternative splicing of thyroid hormone receptor mRNA
.
RNA
 ,
7
,
859
874
.
32
Anko
M.L.
,
Neugebauer
K.M.
(
2012
)
RNA-protein interactions in vivo: global gets specific
.
Trends Biochem Sci.
 ,
37
,
255
262
.
33
Wang
Y.
,
Xiao
X.
,
Zhang
J.
,
Choudhury
R.
,
Robertson
A.
,
Li
K.
,
Ma
M.
,
Burge
C.B.
,
Wang
Z.
(
2013
)
A complex network of factors with overlapping affinities represses splicing through intronic elements
.
Nat. Struct. Mol. Biol.
 ,
20
,
36
45
.
34
Huelga
S.C.
,
Vu
A.Q.
,
Arnold
J.D.
,
Liang
T.Y.
,
Liu
P.P.
,
Yan
B.Y.
,
Donohue
J.P.
,
Shiue
L.
,
Hoon
S.
,
Brenner
S.
et al
. (
2012
)
Integrative genome-wide analysis reveals cooperative regulation of alternative splicing by hnRNP proteins
.
Cell Rep.
 ,
1
,
167
178
.
35
Castello
A.
,
Fischer
B.
,
Hentze
M.W.
,
Preiss
T.
(
2013
)
RNA-binding proteins in Mendelian disease
.
Trends Genet.
 ,
29
,
318
327
.
36
Hardy
J.
,
Rogaeva
E.
(
2014
)
Motor neuron disease and frontotemporal dementia: sometimes related, sometimes not
.
Exp. Neurol.
 ,
262
(Pt B)
,
75
83
.
37
Ling
S.C.
,
Polymenidou
M.
,
Cleveland
D.W.
(
2013
)
Converging Mechanisms in ALS and FTD: Disrupted RNA and Protein Homeostasis
.
Neuron
 ,
79
,
416
438
.
38
Renoux
A.J.
,
Todd
P.K.
(
2011
)
Neurodegeneration the RNA way
.
Prog. Neurobiol.
 ,
97
,
173
179
.
39
Ugras
S.E.
,
Shorter
J.
(
2012
)
RNA-binding proteins in amyotrophic lateral sclerosis and neurodegeneration
.
Neurol. Res. Int.
 ,
2012
,
432780
.
40
Strong
M.J.
(
2010
)
The evidence for altered RNA metabolism in amyotrophic lateral sclerosis (ALS)
.
J. Neurol. Sci.
 ,
288
,
1
12
.
41
Suzuki
H.
,
Shibagaki
Y.
,
Hattori
S.
,
Matsuoka
M.
(
2015
)
Nuclear TDP-43 causes neuronal toxicity by escaping from the inhibitory regulation by hnRNPs
.
Hum. Mol. Genet.
 ,
24
,
1513
1527
.
42
Prudencio
M.
,
Belzil
V.V.
,
Batra
R.
,
Ross
C.A.
,
Gendron
T.F.
,
Pregent
L.J.
,
Murray
M.E.
,
Overstreet
K.K.
,
Piazza-Johnston
A.E.
,
Desaro
P.
et al
. (
2015
)
Distinct brain transcriptome profiles in C9orf72-associated and sporadic ALS
.
Nat. Neurosci.
 ,
18
,
1175
1182
.
43
Berson
A.
,
Barbash
S.
,
Shaltiel
G.
,
Goll
Y.
,
Hanin
G.
,
Greenberg
D.S.
,
Ketzef
M.
,
Becker
A.J.
,
Friedman
A.
,
Soreq
H.
(
2012
)
Cholinergic-associated loss of hnRNP-A/B in Alzheimer's disease impairs cortical splicing and cognitive function in mice
.
EMBO Mol
 . Med.,
4
,
730
742
.
44
Pinkus
J.L.
,
Amato
A.A.
,
Taylor
J.P.
,
Greenberg
S.A.
(
2014
)
Abnormal distribution of heterogeneous nuclear ribonucleoproteins in sporadic inclusion body myositis
.
Neuromuscul. Disord.
 ,
24
,
611
616
.
45
Kim
H.J.
,
Kim
N.C.
,
Wang
Y.D.
,
Scarborough
E.A.
,
Moore
J.
,
Diaz
Z.
,
Maclea
K.S.
,
Freibaum
B.
,
Li
S.
,
Molliex
A.
et al
. (
2013
)
Mutations in prion-like domains in hnRNPA2B1 and hnRNPA1 cause multisystem proteinopathy and ALS
.
Nature
 ,
495
,
467
473
.
46
Nussbacher
J.K.
,
Batra
R.
,
Lagier-Tourenne
C.
,
Yeo
G.W.
(
2015
)
RNA-binding proteins in neurodegeneration: Seq and you shall receive
.
Trends Neurosci.
 ,
38
,
226
236
.
47
He
F.
,
Krans
A.
,
Freibaum
B.D.
,
Taylor
J.P.
,
Todd
P.K.
(
2014
)
TDP-43 suppresses CGG repeat-induced neurotoxicity through interactions with HnRNP A2/B1
.
Hum. Mol. Genet.
 ,
23
,
5036
5051
.
48
Romano
M.
,
Buratti
E.
,
Romano
G.
,
Klima
R.
,
Del Bel Belluz
L.
,
Stuani
C.
,
Baralle
F.
,
Feiguin
F.
(
2014
)
Evolutionarily conserved heterogeneous nuclear ribonucleoprotein (hnRNP) A/B proteins functionally interact with human and Drosophila TAR DNA-binding protein 43 (TDP-43)
.
J. Biol. Chem.
 ,
289
,
7121
7130
.
49
Sephton
C.F.
,
Cenik
C.
,
Kucukural
A.
,
Dammer
E.B.
,
Cenik
B.
,
Han
Y.
,
Dewey
C.M.
,
Roth
F.P.
,
Herz
J.
,
Peng
J.
et al
. (
2011
)
Identification of neuronal RNA targets of TDP-43-containing ribonucleoprotein complexes
.
J. Biol. Chem.
 ,
286
,
1204
1215
.
50
Volonte
C.
,
Apolloni
S.
,
Parisi
C.
(
2015
)
MicroRNAs: newcomers into the ALS picture
.
CNS Neurol. Disord Drug Targets
 ,
14
,
194
207
.
51
Honda
H.
,
Hamasaki
H.
,
Wakamiya
T.
,
Koyama
S.
,
Suzuki
S.O.
,
Fujii
N.
,
Iwaki
T.
(
2015
)
Loss of hnRNPA1 in ALS spinal cord motor neurons with TDP-43-positive inclusions
.
Neuropathology
 ,
35
,
37
43
.

Author notes

These authors contributed equally to this work.