Abstract

TDP-43 was identified as the major component of ubiquitin and autophagosome-positive cytoplasmic inclusions in neurons in the large majority of amyotrophic lateral sclerosis (ALS) and frontotemporal lobar dementia (FTLD) patients. It has been shown that a loss of nuclear TDP-43 in combination with enhanced cytoplasmic mislocalization of TDP-43, which is associated with accumulation of TDP-43 aggregates in the cytosol, is an early and key event in TDP-43-mediated neurodegeneration. However, the mechanism underlying TDP-43 nucleocytoplasmic shuttling is still not clear. Here, we show that the tumor suppressor folliculin (FLCN) is a novel positive regulator of TDP-43 cytoplasmic translocation. FLCN directly interacts with TDP-43. The amino acids 202–299 of FLCN and RNA-recognition motif domains of TDP-43 are necessary for their interaction. In addition, both exogenous and endogenous FLCNs are required for TDP-43 cytoplasmic accumulation, protein aggregation and stress granule formation. Overall, our study suggests that FLCN may play an important role in the regulation of TDP-43 nucleocytoplasmic shuttling and TDP-43-mediated proteinopathy.

Introduction

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder characterized by the muscle weakness and degeneration of motor neurons (1). About 10% of ALS cases are familially inherited (fALS), and the remaining is sporadic ALS (sALS) that has no clear family history. SOD1 (Cu/Zn superoxide dismutase 1) was the first identified gene associated with the fALS (2), and most studies in ALS research have focused on SOD1 for a long time. Recently, the discovery of TDP-43, which is encoded by TARDBP, made a breakthrough in ALS research. TDP-43 is a RNA/DNA-binding protein that predominantly localizes in the nucleus, although it can be exported to the cytosol and transported between cytosol and the nucleus (3). The protein functions in the RNA processing and metabolism including RNA transcription, RNA splicing, RNA transport and stability (4). It contains two RNA-recognition motifs (RRM1 and RRM2) that mediate nucleic acids interactions (5), a C-terminal glycine-rich region that is capable of binding proteins (6), a nuclear localization signal (NLS) and a nuclear export signal (NES) that control nucleocytoplasmic shuttling (3,7).

TDP-43 was identified as a key component of the neuronal ubiquitin-positive inclusions in the central nervous system of the large majority of ALS and frontotemporal lobar dementia (FTLD) patients (8,9). Despite the fact that TDP-43 inclusions, the pathological hallmark of ALS and FTLD, are abnormally ubiquitinated in the cytoplasm of affected neurons (10) (1,8), those affected neurons show not only cytoplasmic TDP-43 aggregation but also a loss of nuclear TDP-43 (11). Since the nuclear localization of TDP-43 simply reflects the physiological function of this RNA-binding protein, a major unexplained question in TDP-43-related ALS pathogenesis is whether the disease is caused by a toxic gain of function (due to the abnormal accumulation of TDP-43 aggregates in the cytoplasm), a loss of function (due to a loss of TDP-43 in the nucleus) or a combination of both.

In response to cellular stress, such as heat shock and oxidative stress, mRNAs and RNA-binding proteins are sequestered to cytoplasmic bodies so called stress granules (12), which is regarded as a cellular self-protective process. Previous studies reported that TDP-43 could localize to stress granules as a specific component of stress granules under various types of stress stimuli (13–15). Moreover, increasing evidences showed that TDP-43 could bind to multiple stress granule component proteins, such as T-cell intracellular antigen 1 (TIA-1) and Ras GTPase-activating protein-binding protein 1 (G3BP1) (16), indicating a potential role of TDP-43 in stress granule formation associated with ALS pathology. The aberrant recruitment of TDP-43 to stress granules could subsequently promote TDP-43 aggregate formation or interaction with other proteins involved in stress granule formation (17,18), suggesting a potential relationship between stress granule formation and TDP-43 aggregate formation in ALS.

Although it is not yet clear whether the cytoplasmic aggregation of TDP-43, a common feature in neurodegenerative disorders such as ALS and FTLD, has a protective or toxic role, growing evidences suggest that an abnormal nucleocytoplasmic shuttling of TDP-43 leads to accumulation of TDP-43 aggregates in the cytosol, which is a key event TDP-43-mediated proteinopathy (1,18–21). A number of recent studies suggest that several cellular stressors and signaling pathways could control TDP-43 cytoplasmic localization (7,13,17,22–24); however, the mechanism by which TDP-43 is transported to and accumulated in the cytosol during stress is still not clear. Elucidating this process is critical to understand the early stage of cytoplasmic accumulation of TDP-43 in ALS and FTLD.

In this study, we identify that the tumor suppressor folliculin (FLCN), which is associated with Birt–Hogg–Dubé (BHD) syndrome, binds to TDP-43 and regulates nucleocytoplasmic shuttling of TDP-43. Overexpression of FLCN enhances TDP-43 cytoplasmic mislocalization, whereas knockdown of FLCN results in TDP-43 nuclear deposition, even with the treatment of stressors that usually induce TDP-43 cytoplasmic mislocalization. Moreover, the amino acids 202–299 of FLCN and RRM domains of TDP-43 are required for their interaction. Furthermore, the interaction leads to TDP-43 cytoplasmic accumulation, ubiquitin and autophagosome-positive aggregation, and stress granule formation.

Results

FLCN enhances translocation of TDP-43 to the cytoplasm

Previous studies have well demonstrated that cellular protein control system, such as autophagy, could play an important role in the regulation of TDP-43 (1,25–28). We wondered whether tumor suppressor FLCN, an evolutionarily conserved protein linking to autophagy (29,30), could regulate TDP-43. We co-transfected TDP-43 with FLCN and examined TDP-43 cellular localization in those transfected cells. Overexpressed TDP-43 showed three distinct distribution patterns in cells: nuclear, cytoplasmic or diffuse distribution (Fig. 1A). Strikingly, we found a dramatic increase of TDP-43 cytoplasmic mislocalization in FLCN-transfected cells (Fig. 1B). Moreover, overexpression of FLCN had a similar effect on cytoplasmic mislocalization of pathogenic TDP-43 G298S and A315 T mutants (data not shown). Consistent with immunofluorescent data, subcellular fractionation assays also showed that overexpression of FLCN resulted in a dramatic increase of cytoplasmic and a decrease of nuclear TDP-43 (Fig. 1C). Similar results were obtained when we detected endogenous TDP-43 in enhanced green fluorescent protein (EGFP)-FLCN-transfected cells under treatment of arsenite, a stressor that induces stress granule formation and TDP-43 cytoplasmic mislocalization (14–16,31,32) (Supplementary Material, Fig. S1A and B). However, knockdown or overexpression of TDP-43 had no effect on FLCN cellular localization (Supplementary Material, Fig. S2A and B).

Figure 1.

Overexpression FLCN recruits TDP-43 to the cytoplasm. (A) Three representative distribution patterns of TDP-43 in cells: nuclear, cytoplasmic or diffuse distribution. HEK 293 cells were transfected with HA-TDP-43. After 24 h transfection, the transfected cells were subjected to immunofluorescence assay using antibody against HA (red). 4′, 6-diamidino-2-phenylindole, dihydrochloride (DAPI) (blue) was used for nuclear staining. The stained cells were visualized using an inverted system microscope IX71 (Olympus). Scale bar, 5 µm. (B) HEK 293 cells were co-transfected with GFP or GFP-FLCN and HA-TDP-43. Twenty-four hours after transfection, the transfected cells were subjected to immunofluorescence assay and visualized using antibody against HA (red). DAPI (blue) was used for nuclear staining. Scale bar, 5 µm. The quantification of TDP-43 localization in (B) is shown on the lower side of (A). An average of 100 cells were analyzed per condition per experiment. The data from three independent experiments indicated the means ± S.E.M., **P < 0.01, one-way ANOVA. Notably, ** indicates the significance of ‘GFP-FLCN-nuclear versus GFP-nuclear’, ‘GFP-FLCN-cytoplasmic versus GFP-cytoplasmic’ and ‘GFP-FLCN-diffuse versus GFP-diffuse’. (C) HEK293 cells were similarly transfected as in (B). Twenty-four hours after transfection, the transfected cells were subjected to subcellular fractionation assay, and then the cell lysates were separated into cytoplasmic and nuclear fractions and subjected to western blotting analysis using anti-GFP, HA, GAPDH and Histone 2B antibodies. The relative densities of TDP-43 in cytoplasm and nucleus are shown on the lower panels. The data from three independent experiments are presented as means ± S.E.M., **P < 0.01, one-way ANOVA.

Figure 1.

Overexpression FLCN recruits TDP-43 to the cytoplasm. (A) Three representative distribution patterns of TDP-43 in cells: nuclear, cytoplasmic or diffuse distribution. HEK 293 cells were transfected with HA-TDP-43. After 24 h transfection, the transfected cells were subjected to immunofluorescence assay using antibody against HA (red). 4′, 6-diamidino-2-phenylindole, dihydrochloride (DAPI) (blue) was used for nuclear staining. The stained cells were visualized using an inverted system microscope IX71 (Olympus). Scale bar, 5 µm. (B) HEK 293 cells were co-transfected with GFP or GFP-FLCN and HA-TDP-43. Twenty-four hours after transfection, the transfected cells were subjected to immunofluorescence assay and visualized using antibody against HA (red). DAPI (blue) was used for nuclear staining. Scale bar, 5 µm. The quantification of TDP-43 localization in (B) is shown on the lower side of (A). An average of 100 cells were analyzed per condition per experiment. The data from three independent experiments indicated the means ± S.E.M., **P < 0.01, one-way ANOVA. Notably, ** indicates the significance of ‘GFP-FLCN-nuclear versus GFP-nuclear’, ‘GFP-FLCN-cytoplasmic versus GFP-cytoplasmic’ and ‘GFP-FLCN-diffuse versus GFP-diffuse’. (C) HEK293 cells were similarly transfected as in (B). Twenty-four hours after transfection, the transfected cells were subjected to subcellular fractionation assay, and then the cell lysates were separated into cytoplasmic and nuclear fractions and subjected to western blotting analysis using anti-GFP, HA, GAPDH and Histone 2B antibodies. The relative densities of TDP-43 in cytoplasm and nucleus are shown on the lower panels. The data from three independent experiments are presented as means ± S.E.M., **P < 0.01, one-way ANOVA.

FLCN is required for TDP-43 nuclear export, but not nuclear import

Based on the observation that FLCN enhanced the translocation of TDP-43 to the cytoplasm, we wondered whether the cytoplasmic distribution of TDP-43 in cells expressing FLCN was caused by an enhanced nuclear export or an impaired nuclear import of TDP-43. To address this issue, we examined the cellular localization of TDP-43 in FLCN-transfected cells treated with leptomycin B (LMB), a CRM1/exportin 1/Xpo1 inhibitor that inhibits protein nuclear export or cycloheximide (CHX) (an inhibitor of protein synthesis in the cytosol). We found that the LMB treatment did not prevent the translocation of TDP-43 to the cytoplasm induced by FLCN (Fig. 2), suggesting that FLCN regulated TDP-43 nucleocytoplasmic shuttling in a CRM1-independent manner. Meanwhile, CHX treatment did not block TDP-43 cytoplasmic localization driven by FLCN, despite that FLCN displayed a nuclear-enriched distribution in this case (Fig. 2), suggesting that the portion of TDP-43 in the cytosol was originated from the nuclear TDP-43 pool. More importantly, we found that TDP-43 displayed a similar distribution as FLCN upon the CHX treatment (Fig. 2), indicating that the distribution of TDP-43 might directly rely on FLCN. Taken together, our data suggest that FLCN is indispensable for non-CRM1-mediated TDP-43 nuclear export, but not nuclear import.

Figure 2.

FLCN is a regulator of TDP-43 localization. (A) HEK 293 cells were co-transfected with GFP or GFP-FLCN and HA-TDP-43 for 24 h. For CHX treatment, the cells were then incubated with CHX (100 µg/ml) for 12 h. For LMB treatment, the cells were then cultured for another 11 h, and then treated with LMB (20 ng/ml) for 1 h. Finally, the cells were subjected to immunofluorescence assay using antibody against HA (red). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm.

Figure 2.

FLCN is a regulator of TDP-43 localization. (A) HEK 293 cells were co-transfected with GFP or GFP-FLCN and HA-TDP-43 for 24 h. For CHX treatment, the cells were then incubated with CHX (100 µg/ml) for 12 h. For LMB treatment, the cells were then cultured for another 11 h, and then treated with LMB (20 ng/ml) for 1 h. Finally, the cells were subjected to immunofluorescence assay using antibody against HA (red). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm.

FLCN directly interacts with TDP-43

It was worth to note that, in part of (∼60%) the FLCN-transfected cells, the cytoplasmic TDP-43 displayed punctuate structures, and those punctuate structures of TDP-43 strongly co-localized with FLCN (Fig. 3A), suggesting that FLCN may target TDP-43 to the cytosol through a direct protein interaction with TDP-43. To explore the potential interaction between FLCN and TDP-43, we co-transfected HEK293 cells with FLCN and TDP-43 and performed immunoprecipitation assays. Our results showed that TDP-43 was co-precipitated with EGFP-FLCN, but not EGFP tag alone (Fig. 3B), and pathogenic TDP-43 mutants could interact with FLCN to a similar extent, compared with wild-type TDP-43 (data not shown). Moreover, the interaction between endogenous FLCN and TDP-43 in cells was also confirmed by immunoprecipitation assay (Fig. 3C). To further investigate whether there is a direct interaction between FLCN and TDP-43, we evaluated the binding of FLCN with TDP-43 using in vitro Glutathione S-transferase (GST) pull-down assay. Results showed that purified recombinant GST-TDP-43, but not GST tag alone, could interact with purified Histidine (His)-tagged FLCN in vitro (Fig. 3D). Taken together, our data suggest that FLCN directly interacts with TDP-43.

Figure 3.

FLCN interacts with TDP-43. (A) HEK 293 cells were co-transfected with GFP or GFP-FLCN and HA-TDP-43 for 24 h. Then the transfected cells were subjected to immunofluorescence assay using antibody against HA (red). DAPI (blue) was used for nuclear staining. The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. The quantification of cytoplasmic TDP-43 with puncta is shown on the lower panel. Data from three independent experiments represented as means ± S.E.M., **P < 0.01, one-way ANOVA. (B) HEK 293 cells were co-transfected with GFP or GFP-FLCN and Flag-TDP-43 for 36 h. Then the cells were subjected to immunoprecipitation analysis using anti-GFP antibody, and the inputs and immunoprecipitants were subjected to western blotting analysis with anti-GFP and Flag antibodies. (C) HEK 293 cells were subjected to immunoprecipitation analysis using rabbit IgG or anti-FLCN antibody. The inputs and immunoprecipitants were subjected to western blotting using anti-FLCN and TDP-43 antibodies. (D) The supernatant of E. coli crude extract containing His-FLCN protein was incubated with GST- or GST-TDP-43-bound glutathione sepharose 4B beads. The input and bound proteins were subjected to western blotting analysis with anti-GST and His antibodies.

Figure 3.

FLCN interacts with TDP-43. (A) HEK 293 cells were co-transfected with GFP or GFP-FLCN and HA-TDP-43 for 24 h. Then the transfected cells were subjected to immunofluorescence assay using antibody against HA (red). DAPI (blue) was used for nuclear staining. The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. The quantification of cytoplasmic TDP-43 with puncta is shown on the lower panel. Data from three independent experiments represented as means ± S.E.M., **P < 0.01, one-way ANOVA. (B) HEK 293 cells were co-transfected with GFP or GFP-FLCN and Flag-TDP-43 for 36 h. Then the cells were subjected to immunoprecipitation analysis using anti-GFP antibody, and the inputs and immunoprecipitants were subjected to western blotting analysis with anti-GFP and Flag antibodies. (C) HEK 293 cells were subjected to immunoprecipitation analysis using rabbit IgG or anti-FLCN antibody. The inputs and immunoprecipitants were subjected to western blotting using anti-FLCN and TDP-43 antibodies. (D) The supernatant of E. coli crude extract containing His-FLCN protein was incubated with GST- or GST-TDP-43-bound glutathione sepharose 4B beads. The input and bound proteins were subjected to western blotting analysis with anti-GST and His antibodies.

Functional domains involved in the interaction between FLCN and TDP-43

To further examine which domains of FLCN and TDP-43 are necessary for their interaction, we generated a series of deletion mutants of FLCN (Fig. 4A) and tested their interactions with TDP-43 using immunoprecipitation assays. We identified that the wild-type FLCN, FLCN (1–557), FLCN (1–344), FLCN (Δ301–449) and FLCN (Δ301–399) interacted with TDP-43, whereas FLCN (345–579), FLCN (Δ202–449), FLCN (Δ202–399) and FLCN (Δ202–299) did not (Fig. 4B). In addition, GST pull-down assay showed that wild-type FLCN and FLCN (1–344) could pull down TDP-43, whereas FLCN (345–579) and FLCN (Δ202–299) did not (Fig. 4C). To further identify which domain of TDP-43 was necessary for the interaction, we generated deletion mutants of TDP-43 (Fig. 4D) and analyzed their interactions with FLCN. Immunoprecipitation assays showed that wild-type TDP-43, TDP-43 (ΔGRD) and TDP-43 (35 kDa) could interact with FLCN, whereas TDP-43 (ΔRRM1 + ΔRRM2) did not. Meanwhile, TDP-43 (ΔRRM1) or TDP-43 (ΔRRM2) had relatively weaker association with FLCN (Fig. 4C and D). Note that TDP-43 (ΔRRM1) had a very faint interaction with FLCN, compared with TDP-43 (ΔRRM2). Consistent with these data, immunofluorescence assays also showed that FLCN mutants lacking amino acids 202–299 failed to target TDP-43 to the cytoplasm (Fig. 5A and Supplementary Material, Fig. S3). Meanwhile, FLCN failed to recruit TDP-43 (ΔRRM1 + ΔRRM2) and had a weak effect on cytoplasmic translocation of TDP-43 (ΔRRM1) or TDP-43 (ΔRRM2) (Fig. 5B). We found that the interaction between TDP-43 and FLCN was not affected by RNase treatment (data not shown), suggesting that their interaction was not RNA dependent. Taken together, our data indicate that amino acids 202–299 of FLCN and RRM1/RRM2 domains of TDP-43 are essential for their direct interaction and FLCN-mediated TDP-43 cytoplasmic translocation.

Figure 4.

Amino acids 202–299 of FLCN and RRM domains of TDP-43 are essential for the interactions. (A) Schematic diagram of FLCN deletion mutants. (B) HEK 293 cells were co-transfected with GFP, GFP-FLCN (WT), GFP-FLCN (1–557), GFP-FLCN (1–344), GFP-FLCN (345–579), GFP-FLCN (Δ202–449), GFP-FLCN (Δ202–399), GFP-FLCN (Δ301–449), GFP-FLCN (Δ301–399) or GFP-FLCN (Δ202–299) and Flag-TDP-43 for 36 h. The cells were subjected to immunoprecipitation analysis with anti-GFP antibody. The inputs and immunoprecipitants were subjected to western blotting analysis using anti-GFP and Flag antibodies. (C) The supernatant of E. coli crude extract containing His-FLCN (WT), His-FLCN (1–344), His-FLCN (345–579) or His-FLCN (Δ202–299) protein was incubated with GST- or GST-TDP-43-bound glutathione sepharose 4B beads. The input and bound proteins were subjected to western blotting analysis with anti-GST, His or FLCN antibodies. (D) Schematic diagram of TDP-43 deletion mutants. (E) HEK 293 cells were co-transfected with Flag-TDP-43 (WT), Flag-TDP-43 (ΔRRM1), Flag-TDP-43 (ΔRRM2) or Flag-TDP-43 (ΔRRM1 + ΔRRM2) and GFP-FLCN for 36 h. The cells were subjected to immunoprecipitation analysis with anti-Flag antibody. The inputs and immunoprecipitants were subjected to western blotting analysis using anti-Flag and GFP antibodies. (F) HEK 293 cells were co-transfected with Flag-TDP-43 (WT), Flag-TDP-43 (ΔGRD), Flag-TDP-43 (ΔRRM1 + ΔRRM2) or Flag-TDP-43 (35KD) and GFP-FLCN for 36 h. The cells were subjected to immunoprecipitation analysis with anti-Flag antibody. The inputs and immunoprecipitants were subjected to western blotting analysis using anti-Flag and GFP antibodies.

Figure 4.

Amino acids 202–299 of FLCN and RRM domains of TDP-43 are essential for the interactions. (A) Schematic diagram of FLCN deletion mutants. (B) HEK 293 cells were co-transfected with GFP, GFP-FLCN (WT), GFP-FLCN (1–557), GFP-FLCN (1–344), GFP-FLCN (345–579), GFP-FLCN (Δ202–449), GFP-FLCN (Δ202–399), GFP-FLCN (Δ301–449), GFP-FLCN (Δ301–399) or GFP-FLCN (Δ202–299) and Flag-TDP-43 for 36 h. The cells were subjected to immunoprecipitation analysis with anti-GFP antibody. The inputs and immunoprecipitants were subjected to western blotting analysis using anti-GFP and Flag antibodies. (C) The supernatant of E. coli crude extract containing His-FLCN (WT), His-FLCN (1–344), His-FLCN (345–579) or His-FLCN (Δ202–299) protein was incubated with GST- or GST-TDP-43-bound glutathione sepharose 4B beads. The input and bound proteins were subjected to western blotting analysis with anti-GST, His or FLCN antibodies. (D) Schematic diagram of TDP-43 deletion mutants. (E) HEK 293 cells were co-transfected with Flag-TDP-43 (WT), Flag-TDP-43 (ΔRRM1), Flag-TDP-43 (ΔRRM2) or Flag-TDP-43 (ΔRRM1 + ΔRRM2) and GFP-FLCN for 36 h. The cells were subjected to immunoprecipitation analysis with anti-Flag antibody. The inputs and immunoprecipitants were subjected to western blotting analysis using anti-Flag and GFP antibodies. (F) HEK 293 cells were co-transfected with Flag-TDP-43 (WT), Flag-TDP-43 (ΔGRD), Flag-TDP-43 (ΔRRM1 + ΔRRM2) or Flag-TDP-43 (35KD) and GFP-FLCN for 36 h. The cells were subjected to immunoprecipitation analysis with anti-Flag antibody. The inputs and immunoprecipitants were subjected to western blotting analysis using anti-Flag and GFP antibodies.

Figure 5.

Amino acids 202–299 of FLCN and RRM domains of TDP-43 are necessary for the recruitment. (A) HEK293 cells were co-transfected with GFP, GFP-FLCN or GFP-FLCN (Δ202–299) and HA-TDP-43 for 24 h. Then the transfected cells were subjected to immunofluorescence assay using antibody against HA (red). DAPI (blue) was used for nuclear staining. The stained cells were visualized using confocal microscopy. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. The quantification of TDP-43 localization is shown on the right side. The data from three independent experiments indicated the means ± S.E.M., ns, not significantly different; **P < 0.01, one-way ANOVA. (B) HEK293 cells were co-transfected with HA-TDP-43 (WT), HA-TDP-43 (ΔRRM1), HA-TDP-43 (ΔRRM2) or HA-TDP-43 (ΔRRM1 + ΔRRM2) and GFP-FLCN for 24 h. Then the transfected cells were subjected to immunofluorescence assay using anti-HA (red) antibody. The stained cells were visualized using a microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. The quantification of TDP-43 localization is shown on the right side. The data from three independent experiments indicated the means ± S.E.M., ns, not significantly different; *P < 0.05; **P < 0.01, one-way ANOVA. Similarly as in Figure 1, the significances indicate the comparison within the ‘nuclear’, ‘cytoplasmic’ and ‘diffuse’ groups, and all indicate the comparison with relative control (GFP in (A) and wild-type TDP-43 in (B)).

Figure 5.

Amino acids 202–299 of FLCN and RRM domains of TDP-43 are necessary for the recruitment. (A) HEK293 cells were co-transfected with GFP, GFP-FLCN or GFP-FLCN (Δ202–299) and HA-TDP-43 for 24 h. Then the transfected cells were subjected to immunofluorescence assay using antibody against HA (red). DAPI (blue) was used for nuclear staining. The stained cells were visualized using confocal microscopy. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. The quantification of TDP-43 localization is shown on the right side. The data from three independent experiments indicated the means ± S.E.M., ns, not significantly different; **P < 0.01, one-way ANOVA. (B) HEK293 cells were co-transfected with HA-TDP-43 (WT), HA-TDP-43 (ΔRRM1), HA-TDP-43 (ΔRRM2) or HA-TDP-43 (ΔRRM1 + ΔRRM2) and GFP-FLCN for 24 h. Then the transfected cells were subjected to immunofluorescence assay using anti-HA (red) antibody. The stained cells were visualized using a microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. The quantification of TDP-43 localization is shown on the right side. The data from three independent experiments indicated the means ± S.E.M., ns, not significantly different; *P < 0.05; **P < 0.01, one-way ANOVA. Similarly as in Figure 1, the significances indicate the comparison within the ‘nuclear’, ‘cytoplasmic’ and ‘diffuse’ groups, and all indicate the comparison with relative control (GFP in (A) and wild-type TDP-43 in (B)).

FLCN is required for TDP-43 cytoplasmic accumulation, aggregation and stress granule formation

As we have shown that FLCN could induce a dramatic increase of the cytoplasmic punctuate TDP-43, we wondered which subcellular organelle the punctuate TDP-43 located and what were those punctuate structures. We found that the punctuate TDP-43 in FLCN-transfected cells could localize to the lysosomes, but not mitochondria, endoplasmic reticulum (ER) or Golgi apparatus (Fig. 6A). In addition, we found that the punctuate TDP-43 could also co-localize with ubiquitin, p62 (an autophagic receptor and marker) and LC3 (autophagosomal marker), but not G3BP1 (Fig. 6B). Consistent with these data, immunoprecipitation assays also showed that endogenous TDP-43 interacted with p62 in EGFP-FLCN-transfected cells under the treatment of arsenite (Supplementary Material, Fig. S4). These results suggest that FLCN-mediated TDP-43 cytoplasmic accumulation may result in protein aggregation of TDP-43, and those TDP-43 aggregates were recognized by components in ubiquitin–proteasome system (UPS) and autophagy–lysosome pathway (ALP), as shown in previous studies (23,25,33).

Figure 6.

FLCN targets TDP-43 to stress granules and autophagosomes. (A) HEK293 cells were co-transfected with GFP-FLCN and HA-TDP-43. Twenty-four hours after transfection, the cells were subjected to immunofluorescence assay using antibodies against HA (blue), Tom20, Calnexin, GM130 and Lamp-2 or lysosome tracker (red). The stained cells were visualized using an inverted system microscope IX71 (Olympus). Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (B) HEK 293 cells were co-transfected with Flag-UB or RFP-LC3 and GFP-FLCN and HA-TDP-43 for 24 h. The cells were subjected to immunofluorescence assay using antibodies against HA (blue), Flag, P62 and G3BP1 (red). The stained cells were visualized using an inverted system microscope IX71 (Olympus). Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (C) HEK293 cells were co-transfected with GFP-LC3 and HA-TDP-43 for 24 h and incubated with arsenite (0.5 mm) for 15 min. The cells were subjected to immunofluorescence assay using antibodies against HA (red) and G3BP1 (blue). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (D) HEK 293 cells were co-transfected with Flag-UB and HA-TDP-43 for 24 h and incubated with arsenite (0.5 mm) for 15 min. The cells were subjected to immunofluorescence assay using antibodies against Flag (red) and HA (green). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm.

Figure 6.

FLCN targets TDP-43 to stress granules and autophagosomes. (A) HEK293 cells were co-transfected with GFP-FLCN and HA-TDP-43. Twenty-four hours after transfection, the cells were subjected to immunofluorescence assay using antibodies against HA (blue), Tom20, Calnexin, GM130 and Lamp-2 or lysosome tracker (red). The stained cells were visualized using an inverted system microscope IX71 (Olympus). Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (B) HEK 293 cells were co-transfected with Flag-UB or RFP-LC3 and GFP-FLCN and HA-TDP-43 for 24 h. The cells were subjected to immunofluorescence assay using antibodies against HA (blue), Flag, P62 and G3BP1 (red). The stained cells were visualized using an inverted system microscope IX71 (Olympus). Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (C) HEK293 cells were co-transfected with GFP-LC3 and HA-TDP-43 for 24 h and incubated with arsenite (0.5 mm) for 15 min. The cells were subjected to immunofluorescence assay using antibodies against HA (red) and G3BP1 (blue). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (D) HEK 293 cells were co-transfected with Flag-UB and HA-TDP-43 for 24 h and incubated with arsenite (0.5 mm) for 15 min. The cells were subjected to immunofluorescence assay using antibodies against Flag (red) and HA (green). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm.

Under arsenite treatment, TDP-43 was recruited onto the stress granules (G3BP1) that were also labeled by LC3 or ubiquitin (Fig. 6C and D), which was consistent with recent reports showing that stress granules have a close relationship with protein aggregates and UPS/ALP components (17,34–38). To test whether FLCN could regulate TDP-43 cytoplasmic accumulation, aggregation and stress granule formation at a physiological circumstance, we tested the effect of endogenous FLCN on the translocation of TDP-43 to the cytosol and stress granules. In cells treated with arsenite, TDP-43 showed an enhanced cytoplasmic localization and co-localized with stress granule marker G3BP1. However, in FLCN-depleted cells, TDP-43 dissociated with cytoplasmic stress granules and shuttled back into the nucleus after arsenite treatment (Fig. 7A and Supplementary Material, Fig. S5A). In cells transfected with FLCN alone, FLCN could not co-localize with G3BP1 with or without arsenite treatment (Supplementary Material, Fig. S6), and depletion of FLCN did not affect cytoplasmic aggregation of a TDP-43 mutant lacking a NLS (TDP-43 ΔNLS) upon MG132 treatment (Supplementary Material, Fig. S7), indicating that FLCN enhanced translocation of TDP-43 to the cytosol prior to TDP-43 aggregation and stress granule formation, and it was highly likely that accumulation of TDP-43 in the cytosol leads to TDP-43 aggregate and stress granules formation. In addition, in cells treated with MG132 (a proteasome inhibitor that induces protein aggregation), we found that the nuclear TDP-43 was also translocated to the cytosol and co-localized with p62. Similar to the results obtained from the arsenite treatment, the cytoplasmic translocation and aggregation of TDP-43 were also inhibited in FLCN-depleted cells (Fig. 7B and Supplementary Material, Fig. S5B). In addition, similar results were obtained when we detected endogenous TDP-43 in FLCN-depleted cells under the treatment of MG132 using biochemical fractionation assays (Supplementary Material, Fig. S8A and B). Taken together, these data suggest that FLCN is critical for TDP-43 cytoplasmic mislocalization, stress granule and aggregate formation.

Figure 7.

Knockdown of FLCN prevented TDP-43 nuclear export. (A) HEK 293 cells were transfected with siRNA against FLCN for 48 h, and then the cells were transfected with HA-TDP-43 for 24 h and incubated with arsenite (0.5 mm) for 15 min. The cells were fixed and subjected to immunofluorescence assay using antibodies against HA (green) and G3BP1 (red). DAPI (blue) was used for nuclear staining. The stained cells were visualized using confocal microscopy. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (B) HEK293 cells were similarly transfected as in (A). The cells were incubated with MG132 (10 µM) for 14 h. The cells were subjected to immunofluorescence assay using antibodies against HA (green) and P62 (red). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm.

Figure 7.

Knockdown of FLCN prevented TDP-43 nuclear export. (A) HEK 293 cells were transfected with siRNA against FLCN for 48 h, and then the cells were transfected with HA-TDP-43 for 24 h and incubated with arsenite (0.5 mm) for 15 min. The cells were fixed and subjected to immunofluorescence assay using antibodies against HA (green) and G3BP1 (red). DAPI (blue) was used for nuclear staining. The stained cells were visualized using confocal microscopy. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm. (B) HEK293 cells were similarly transfected as in (A). The cells were incubated with MG132 (10 µM) for 14 h. The cells were subjected to immunofluorescence assay using antibodies against HA (green) and P62 (red). The stained cells were visualized using microscope IX71. Regions within the dotted boxes are magnified in the insets. Scale bar, 5 µm.

Discussion

Although the identification of TDP-43 as the major aggregate-prone protein in most ALS patients was a breakthrough in ALS research (8,9,20), it is still far away to completely answer how exactly nuclear depletion and cytoplasmic aggregation of TDP-43 contribute to the disease pathogenesis (20). In attempt to clarify the role of loss of TDP-43 nuclear function in ALS and FTLD pathology, many TDP-43-targeting RNAs, such as Nefl, CFTR, HDAC6 and SMN, have been identified (39–44). Although it is generally believed that cytoplasmic translocation of TDP-43 is an early event and important step in ALS pathogenesis (1,19,21), the basic mechanism underlying TDP-43 nucleocytoplasmic shuttling is still not fully understood. Relevant to this, the identification of TDP-43 interacting proteins may help to provide crucial insight into this process. Among cellular stressors and genetic modifiers that were found to function in the regulation of TDP-43 cytoplasmic mislocalization (23,24,45–50), we identify FLCN as a novel strong modifier of TDP-43 nucleocytoplasmic shuttling and aggregation in the current study.

Previous studies have shown that TDP-43 could interact with several heterogeneous nuclear ribonucleoproteins (hnRNPs) such as hnRNPA2/B1, hnRNPA1, hnRNP A2 and hnRNP K (6,51). For example, amino acids 321–366 in the C-terminus of TDP-43 have been identified to interact with hnRNP A2 (51). In the current study, we found that the RRM domains of TDP-43 could associate with FLCN, leading to a recruitment of TDP-43 to the cytosol by FLCN under normal and stress conditions (Figs 1, 4 and 5). It is well known that TDP-43 could be transported to the nucleus through Importins that are involved in the nuclear import pathway (52), and it could also be transported to the cytosol through CRM1 (exportin 1/Xpo1) that is involved in the nuclear export pathway (21,23). Thus, given that FLCN may enhance TDP-43 cytoplasmic localization by either impairing TDP-43 nuclear import or promoting TDP-43 nuclear export, we carefully tested each possibility using CRM1 inhibitor LMB or a protein synthesis inhibitor CHX (Fig. 2). FLCN still targets TDP-43 to the cytosol under CHX treatment, indicating that the portion of cytoplasmic TDP-43 was originated from the nuclear pool. Thus, FLCN may enhance TDP-43 cytoplasmic transport rather than directly impair its nuclear transport. Interestingly and importantly, LMB treatment failed to affect FLCN-mediated TDP-43 nuclear export (Fig. 2), indicating that FLCN facilitates TDP-43 cytoplasmic translocation in a CRM1-independent manner. Besides the physical interactions that may retain TDP-43 in the cytosol by FLCN, as FLCN influences the aggregate and stress granule formation of TDP-43, we suppose that there might be several possibilities involved in FLCN-mediated TDP-43 nucleocytoplasmic shuttling: (1) FLCN interacts with TDP-43 to influence TDP-43 folding and conformation; (2) FLCN interacts with TDP-43 to affect the association between TDP-43 and its binding partners, which play a role in the regulation of TDP-43 nucleocytoplasmic shuttling; and (3) FLCN interacts with TDP-43 to affect TDP-43 posttranslational modifications, such as ubiquitination and phosphorylation of TDP-43, which are very important for TDP-43 function, cellular localization and aggregation (8,9,11,18,20).

Cytoplasmic accumulation of TDP-43 is usually accompanied by the translocation of TDP-43 to cytoplasmic aggregates and stress granules. Stress granules are the cytoplasmic compartments that are enriched with mRNA and RNA-binding proteins and have a strong relationship with many pathogenic proteins involved in ALS and FTLD (36–38), and it has been shown that TDP-43 protein aggregation is mediated, at least in part, through a molecular mechanism associated with stress granule formation (36–38). Interestingly, recent studies showed that TDP-43 could interact with TIA-1 and regulate G3BP1 expression, thereby regulating stress granule dynamics, size and assembly (15,16,32,53), suggesting that TDP-43 is a potential key component that is tightly associated with stress granule assembly. In FLCN-depleted cells, we found that TDP-43 dissociated with stress granules and shuttled back into the nucleus, and the size of stress granules was relatively small in those cells compared with control cells (Fig. 7B), suggesting that the FLCN-mediated cytoplasmic localization of TDP-43 is required for stress granule assembly.

Although FLCN can promote the abnormal cytoplasmic deposition of TDP-43, TDP-43 still localizes in the nucleus and soluble fractionation under normal condition (7,54). However, under pathological conditions and long-term stress, protein folding and posttranslational modifications of TDP-43 seem to be affected (1,18,21). Thus, FLCN may facilitate the nuclear loss and cytoplasmic accumulation of TDP-43 under these conditions, rather than direct impairment of TDP-43 nuclear import (Fig. 2) (21). Given that nuclear export and cytoplasmic accumulation of TDP-43 are the critical steps involved in depositing TDP-43 into the insoluble aggregates in the cytosol, our study offers an important conclusion that the pathological progress of neurodegeneration may require key player such as FLCN to assist TDP-43 cytoplasmic mislocalization and aggregation. Moreover, since FLCN is a multiple function protein involved in diverse aspects of biological events such as tumorigenesis, metabolism, development and autophagy (30,55–58), our results may also be of help to fully understand FLCN function. Taken together, the present study is not only helpful for understanding the basic mechanism underlying TDP-43 nucleocytoplasmic shuttling, but it also indicates a potential role of FLCN in ALS and FTLD pathogenesis.

Material and Methods

Plasmid constructs

Full-length FLCN cDNA was amplified from a human fetal brain cDNA library (Clonetech) with the primers 5′ CCCAAGCTTATGAATGCCATCGTGGCT 3′ and 5′ GCTCTAGAGTTCCGAGACTCCGAGGC 3′ and then inserted into 3 × FLAG at HindIII/XbaI sites. Deletion mutants of FLCN encoding amino acids 1–344 were created by subcloning polymerase chain reaction (PCR) products amplified with the primers 5′ CCCAAGCTTATGAATGCCATCGTGGCT 3′ and 5′ GCTCTAGATGGCAGCTTCCGGGGCTG 3′ into 3 × FLAG at HindIII/XbaI sites. The deletion mutants of 3 × FLAG-FLCN were generated by the site-directed mutagenesis using MutanBEST kit (Takara) with following primers: 5′ TCTAGAGGATCCGAACAAAAA 3′ and 5′ CAGGCCAGTCATCCAGAACTT 3′ for 3 × FLAG-FLCN (1–557) lacking amino acids 558–579, 5′ GTCTTCAAGTCCCTCCGG 3′ and 5′ AAGCTTGTCATCGTCATC 3′ for 3 × FLAG-FLCN (345–579) lacking 1–344 amino acids, and EGFP-N3-tagged FLCN was generated by excising full-length FLCN cDNA from 3 × FLAG-FLCN and inserting it into the pEGFP-N3 (Clontech) vector at HindIII/BamHI sites. EGFP-N3-FLCN (1–557), EGFP-N3-FLCN (1–344) and EGFP-N3-FLCN (345–579) were, respectively, generated by excising deletion mutant of FLCN cDNA from 3 × FLAG-FLCN (1–557), 3 × FLAG-FLCN (1–344) and 3 × FLAG-FLCN (345–579) and inserting them into the pEGFP-N3 vector at HindIII/BamHI sites. The deletion mutants of EGFP-N3-FLCN were generated by the site-directed mutagenesis using MutanBEST kit (Takara) with following primers: 5′ CACCCTGTGGGGTGTGAGGAT 3′ and 5′ CTGGAGCTCATCGATGAT 3′ for EGFP-N3-FLCN lacking 202–449 amino acids (Δ202–449); 5′ GTCCGCATCATCCCATACAGC 3′ and 5′ CTGGAGCTCATCGATGAT 3′ for EGFP-N3-FLCN lacking 202–399 amino acids (Δ202–399); 5′ CACCCTGTGGGGTGTGAG GAT 3′ and 5′ GTCCCAGCTTTCTGATTCCTC 3′ for EGFP-N3-FLCN lacking 301–449 amino acids (Δ301–449); 5′ GTCCGCATCATCCCATACAGC 3′ and 5′ GTCCCAGCTTTCTGATTCCTC 3′ for EGFP-N3-FLCN lacking 301–399 amino acids (Δ301–399); 5′ GACAACTCTGAGGCTGAAGAG 3′ and 5′ CTGGAGCTCATCGATGATT 3′ for EGFP-N3-FLCN lacking 202–299 amino acids (Δ202–299); full-length FLCN cDNA was created by subcloning PCR product, amplified using 3 × FLAG-FLCN as a template with the primers 5′ GGAATTCATGAATGCCATCGTGGCT 3′ and 5′ CCGCTCGAGGTTCCGAGACTCCGAGGC 3′, into pET-21a at its EcoRI/XhoI sites. Deletion mutants of FLCN encoding amino acids 345–579 were created by subcloning PCR products amplified with the primers 5′ GGAATTCACCATGGTCTTCAAG TCCCTC 3′ and 5′ CCGCTCGAGGTTCCGAGACTCCGAGGCTGT 3′ into pET-21a at its EcoRI/XhoI sites. The deletion mutants of pET-21a-FLCN were generated by the site-directed mutagenesis using MutanBEST kit (Takara) with following primers: 5′ CTCGAGCACCACCACCACCAC 3′ and 5′ TGGCAGCTTCCGGGGCTGCCA 3′ for pET-21a-FLCN lacking 345–579 amino acids (1–344), 5′ GACAACTCTGAGGCTGAAGAG 3′ and 5′ CTGGAGCTCATCGATGATT 3′ for pET-21a-FLCN lacking 202–299 amino acids (Δ202–299), and HA-TDP-43 was created by subcloning TDP-43 PCR product using pGEX-5x-1-TDP-43 as a template (25) into pKH3-HA vector at HindIII/XbaI sites. FLAG-tagged TDP-43 was generated by excising TDP-43 cDNA from HA-TDP-43 and inserting it into the 3 × FLAG at HindIII/XbaI sites. The deletion mutants of FLAG-TDP-43 were generated by using following primers: 5′ GGACGATGGTGTGACTGCAAA 3′ and 5′ GGATGTTTTCTGGACTGCTCT 3′ for FLAG-TDP-43 lacking 105–169 amino acids (ΔRRM1); 5′ TCCAATGCCGAACCTAAGCAC 3′ and 5′ CACTTTTCTGCTTCTCAAAGG 3′ for FLAG-TDP-43 lacking 194–257 amino acids (ΔRRM2); and 5′ GGACGATGGTGTGACTGCAAA 3′ and 5′ GGATGTTTTCTGGACTGCTCT 3′, 5′ TCCAATGCCGAACCTAAGCAC 3′ and 5′ CACTTTTCTGCTTCTCAAAGG 3′ for FLAG-TDP-43 lacking 105–169 and 194–257 amino acids (ΔRRM1 + ΔRRM2), respectively. HA-TDP-43 (ΔRRM1), HA-TDP-43 (ΔRRM2) and HA-TDP-43 (ΔRRM1 + ΔRRM2) were, respectively, generated by excising deletion mutant of TDP-43 cDNA from 3 × FLAG-TDP-43 (ΔRRM1), 3 × FLAG-TDP-43 (ΔRRM2) and 3 × FLAG-TDP-43 (ΔRRM1 + ΔRRM2) and inserting them into the pKH3-HA vector at HindIII/XbaI sites. The deletion mutants of HA-TDP-43 were generated by the site-directed mutagenesis using MutanBEST kit (Takara) for HA-TDP-43 lacking 1–85 amino acids (HA-TDP-43 35KDa). The point mutation of EGFP-N3-TDP-43 was generated by the site-directed mutagenesis using MutanBEST kit (Takara) with following primers: 5′ TGCGTTATCTTTTGGATAGTTGACAACA 3′ and 5′ GCAGCAATGGATGAGACAGATGCTTCAT 3′ for EGFP-N3-TDP-43 (ΔNLS) in which the lysine 82, arginine 83 and lysine 84 were converted to alanines; Flag-UB, EGFP-LC3 and RFP-LC3 constructs were described previously (59–61).

Cell culture, transfection and drug treatment

Human embryonic kidney 293 cells (HEK293) were cultured in Dulbecco's modified Eagle's medium (DMEM) (Gibco) containing 10% fetal bovine serum (FBS) (Gibco) with penicillin (10 000 Units/ml) and streptomycin (10 000 µg/ml). Cells were transfected with siRNAs against FLCN using the RNAiMAX (Invitrogen) transfection reagent upon splitting and transfected with plasmids using Lipofectamine 2000 transfection reagent (Invitrogen) according to the manufacturer's instructions. The oligonucleotides that target human FLCN were described in a previous study (58). Cells were treated with LMB (20 ng/ml) (Beyotime) for 1 h, or CHX (100 µg/ml) (Sigma) for 12 h or arsenite (Sigma) (0.5 mm) for 15 min, or MG132 (10 µM) (Calbiochem) for 14 h.

Western blotting

Cells were harvested and lysed in the cell lysis buffer [50 mm Tris–HCl, pH 7.6, 150 mm NaCl, 0.5% sodium deoxycholate, 1% Nonidet P-40 and protease inhibitor cocktail (Roche)]. Then the proteins were separated by 10, 12 or 13.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore).

Antibodies

The following primary antibodies were used: mouse monoclonal antibodies against Calnexin (BD), FLAG (Sigma), GAPDH (Chemicon), GFP (Santa Cruz), HA (Santa Cruz), GM130 (BD), His (Santa Cruz), GST (Santa Cruz), Tom20 (Santa Cruz), p62 (Enzo life science), rabbit polyclonal antibodies against FLCN (Cell Signaling Technology), G3BP1 (Proteintech), HA (Santa Cruz), Histone 2B (Abcam) and Lamp-2 (Abcam).

The following secondary antibodies were used: horseradish peroxidase-conjugated sheep anti-mouse and anti-rabbit antibodies (Amersham Pharmacia Biotech). The proteins were visualized with an ECL detection kit (Thermo).

The following fluorescent secondary antibodies were used: Alexa Fluor 350 goat anti-mouse IgG (Invitrogen), Alexa Fluor 350 goat anti-rabbit IgG (Invitrogen), Alexa Fluor 488 donkey anti-mouse IgG (Invitrogen), Alexa Fluor 488 donkey anti-rabbit IgG (Invitrogen), Alexa Fluor 594-conjugated Affinipure Donkey Anti-mouse IgG (Invitrogen) and Alexa Fluor 594-conjugated Affinipure Donkey Anti-rabbit IgG (Invitrogen).

Immunofluorescence

Cells were washed with 1× phosphate-buffered saline (PBS) (pH 7.4) (Sangon Biotech) once. Cells were fixed with 4% paraformaldehyde for 10 min and washed with 1× PBS three times. Subsequently, cells were treated with 0.25% Triton X-100 adding 0.2% FBS for 5 min and washed with 1× PBS three times. Then cells were incubated with the primary antibodies for 5 h. After washed with 1× PBS three times, the cells were treated with the fluorescent secondary antibodies for 1 h, and then were stained with DAPI (Sigma) for 8 min. The cells were visualized using a Zeiss LSM710 confocal microscope or an inverted system microscope IX71 (Olympus).

The live cells were staining with Lysotracker red DND-99 (Invitrogen) for 10 min, and then the cells were fixed for immunofluorescent assays.

Immunoprecipitation

Protein G-Sepharose (Roche) was incubated with anti-Flag or GFP antibodies and 0.01% bovine serum albumin (BSA) overnight at 4°C and washed with 1× PBS three times. Cells were harvested and lysed in cell lysis buffer, and then cell lysates were sonicated for 1 min on ice and centrifuged at 12 000 rpm for 30 min at 4°C to remove the cellular debris. The supernatants were added to the above incubated protein G with antibodies and allowed for shaking for 6 h on ice. After incubation, the beads were washed with cell lysis buffer six times. The proteins were eluted with 2 × SDS sample buffer, and then were subjected to western blotting analysis.

GST pull-down assay

GST or GST-TDP-43 protein was expressed by Escherichia coli strain DH5α, and His-FLCN protein was expressed by E. coli strain BL21. An aliquot containing 20 µg of GST or GST-TDP-43 protein from the supernatants of E. coli strain DH5α cell lysates was incubated with 20 µl of glutathione sepharose 4B beads (GE Healthcare) for 30 min on ice, and then the beads were washed with ice-cold 1× PBS three times. Subsequently, the beads were incubated with 50 µg of His-FLCN protein from the supernatants of E. coli strain BL21 for 3 h on ice, and then the beads were washed with ice-cold 1 × wash buffer (HNTG buffer) [10 mm sucrose, 1 mm CaCl2, 10 mm MgAc, 2.5 mm ethylenediaminetetraacetic acid (EDTA), 1 mm DL-Dithiothreitol (DTT), 1 mm Phenylmethyl sulfonyl fluoride (PMSF) and 0.5% NP-40] six times. The proteins were eluted with 2 × SDS sample buffer and subjected to western blotting analysis.

Biochemical fractionation

For cellular fractionation (cytoplasm and nucleus fractionations), cells were washed with 1× PBS (pH 7.4) once. Cells were harvested and lysed in sucrose buffer (10 mm sucrose, 1 mm CaCl2, 10 mm MgAc, 2.5 mm EDTA, 1 mm DTT, 1 mm PMSF and 0.5% NP-40), and then the cells were incubated on ice for 30 min and flipped every 5 min. Subsequently, cell lysates were centrifuged at 600g for 15 min at 4°C, and the cell lysates were separated into detergent-soluble (cytoplasm) and detergent-insoluble fractions (nucleus). The detergent-insoluble fractions were washed twice with sucrose buffer without NP-40, and then were resuspended and sonicated in the cell lysis buffer.

For aggregate-prone fractionation, cell lysates were sonicated in lysis buffer, and then the cell lysates were separated into supernatants and pellet by centrifugation at 12 000 rpm for 30 min at 4°C. The pellet was lysed in lysis buffer (10% SDS, 1% Nonidet P-40) as detergent-insoluble fraction containing protein aggregates.

Supplementary Material

Supplementary Material is available at HMG online.

Funding

This work was supported in part by the National High-tech Research and Development program of China 973-projects (2012CB947602), the National Natural Sciences Foundation of China (Nos 31330030, 81371393, 31200803 and 31371072), the Natural Science Foundation of Jiangsu Province (BK2012181), a project funded by the Jiangsu Key Laboratory of Translational Research and Therapy for Neuro-Psycho-Diseases (BM2013003) and a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.

Acknowledgements

We thank Dr Cheng Fu (University of Science & Technology of China) for helpful discussions and suggestions.

Conflict of Interest statement. None declared.

References

1
Ling
S.C.
,
Polymenidou
M.
,
Cleveland
D.W.
(
2013
)
Converging mechanisms in ALS and FTD: disrupted RNA and protein homeostasis
.
Neuron
 ,
79
,
416
438
.
2
Rosen
D.R.
,
Siddique
T.
,
Patterson
D.
,
Figlewicz
D.A.
,
Sapp
P.
,
Hentati
A.
,
Donaldson
D.
,
Goto
J.
,
O'Regan
J.P.
,
Deng
H.X.
et al
. (
1993
)
Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis
.
Nature
 ,
362
,
59
62
.
3
Winton
M.J.
,
Igaz
L.M.
,
Wong
M.M.
,
Kwong
L.K.
,
Trojanowski
J.Q.
,
Lee
V.M.
(
2008
)
Disturbance of nuclear and cytoplasmic TAR DNA-binding protein (TDP-43) induces disease-like redistribution, sequestration, and aggregate formation
.
J. Biol. Chem.
 ,
283
,
13302
13309
.
4
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
.
5
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
.
6
Buratti
E.
,
Brindisi
A.
,
Giombi
M.
,
Tisminetzky
S.
,
Ayala
Y.M.
,
Baralle
F.E.
(
2005
)
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.
 ,
280
,
37572
37584
.
7
Ayala
Y.M.
,
Zago
P.
,
D’Ambrogio
A.
,
Xu
Y.F.
,
Petrucelli
L.
,
Buratti
E.
,
Baralle
F.E.
(
2008
)
Structural determinants of the cellular localization and shuttling of TDP-43
.
J. Cell Sci.
 ,
121
,
3778
3785
.
8
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
.
9
Sreedharan
J.
,
Blair
I.P.
,
Tripathi
V.B.
,
Hu
X.
,
Vance
C.
,
Rogelj
B.
,
Ackerley
S.
,
Durnall
J.C.
,
Williams
K.L.
,
Buratti
E.
et al
. (
2008
)
TDP-43 mutations in familial and sporadic amyotrophic lateral sclerosis
.
Science
 ,
319
,
1668
1672
.
10
Neumann
M.
(
2009
)
Molecular neuropathology of TDP-43 proteinopathies
.
Int. J. Mol. Sci.
 ,
10
,
232
246
.
11
Lagier-Tourenne
C.
,
Cleveland
D.W.
(
2009
)
Rethinking ALS: the FUS about TDP-43
.
Cell
 ,
136
,
1001
1004
.
12
Anderson
P.
,
Kedersha
N.
(
2008
)
Stress granules: the Tao of RNA triage
.
Trends Biochem. Sci.
 ,
33
,
141
150
.
13
Colombrita
C.
,
Zennaro
E.
,
Fallini
C.
,
Weber
M.
,
Sommacal
A.
,
Buratti
E.
,
Silani
V.
,
Ratti
A.
(
2009
)
TDP-43 is recruited to stress granules in conditions of oxidative insult
.
J. Neurochem.
 ,
111
,
1051
1061
.
14
Dewey
C.M.
,
Cenik
B.
,
Sephton
C.F.
,
Dries
D.R.
,
Mayer
P.
III
,
Good
S.K.
,
Johnson
B.A.
,
Herz
J.
,
Yu
G.
(
2011
)
TDP-43 is directed to stress granules by sorbitol, a novel physiological osmotic and oxidative stressor
.
Mol. Cell. Biol.
 ,
31
,
1098
1108
.
15
McDonald
K.K.
,
Aulas
A.
,
Destroismaisons
L.
,
Pickles
S.
,
Beleac
E.
,
Camu
W.
,
Rouleau
G.A.
,
Vande Velde
C.
(
2011
)
TAR DNA-binding protein 43 (TDP-43) regulates stress granule dynamics via differential regulation of G3BP and TIA-1
.
Hum. Mol. Genet.
 ,
20
,
1400
1410
.
16
Liu-Yesucevitz
L.
,
Bilgutay
A.
,
Zhang
Y.J.
,
Vanderweyde
T.
,
Citro
A.
,
Mehta
T.
,
Zaarur
N.
,
McKee
A.
,
Bowser
R.
,
Sherman
M.
et al
. (
2010
)
Tar DNA binding protein-43 (TDP-43) associates with stress granules: analysis of cultured cells and pathological brain tissue
.
PLoS ONE
 ,
5
,
e13250
.
17
Parker
S.J.
,
Meyerowitz
J.
,
James
J.L.
,
Liddell
J.R.
,
Crouch
P.J.
,
Kanninen
K.M.
,
White
A.R.
(
2012
)
Endogenous TDP-43 localized to stress granules can subsequently form protein aggregates
.
Neurochem. Int.
 ,
60
,
415
424
.
18
Cohen
T.J.
,
Lee
V.M.
,
Trojanowski
J.Q.
(
2011
)
TDP-43 functions and pathogenic mechanisms implicated in TDP-43 proteinopathies
.
Trends Mol. Med.
 ,
17
,
659
667
.
19
Lagier-Tourenne
C.
,
Polymenidou
M.
,
Cleveland
D.W.
(
2010
)
TDP-43 and FUS/TLS: emerging roles in RNA processing and neurodegeneration
.
Hum. Mol. Genet.
 ,
19
,
R46
R64
.
20
Da Cruz
S.
,
Cleveland
D.W.
(
2011
)
Understanding the role of TDP-43 and FUS/TLS in ALS and beyond
.
Curr. Opin. Neurobiol.
 ,
21
,
904
919
.
21
Dormann
D.
,
Haass
C.
(
2011
)
TDP-43 and FUS: a nuclear affair
.
Trends Neurosci.
 ,
34
,
339
348
.
22
Ayala
V.
,
Granado-Serrano
A.B.
,
Cacabelos
D.
,
Naudi
A.
,
Ilieva
E.V.
,
Boada
J.
,
Caraballo-Miralles
V.
,
Llado
J.
,
Ferrer
I.
,
Pamplona
R.
et al
. (
2011
)
Cell stress induces TDP-43 pathological changes associated with ERK1/2 dysfunction: implications in ALS
.
Acta Neuropathol.
 ,
122
,
259
270
.
23
Kim
S.H.
,
Zhan
L.
,
Hanson
K.A.
,
Tibbetts
R.S.
(
2012
)
High-content RNAi screening identifies the type 1 inositol triphosphate receptor as a modifier of TDP-43 localization and neurotoxicity
.
Hum. Mol. Genet.
 ,
21
,
4845
4856
.
24
Liu
Y.J.
,
Ju
T.C.
,
Chen
H.M.
,
Jang
Y.S.
,
Lee
L.M.
,
Lai
H.L.
,
Tai
H.C.
,
Fang
J.M.
,
Lin
Y.L.
,
Tu
P.H.
et al
. (
2015
)
Activation of AMP-activated protein kinase alpha1 mediates mislocalization of TDP-43 in amyotrophic lateral sclerosis
.
Hum. Mol. Genet.
 ,
24
,
787
801
.
25
Wang
X.
,
Fan
H.
,
Ying
Z.
,
Li
B.
,
Wang
H.
,
Wang
G.
(
2010
)
Degradation of TDP-43 and its pathogenic form by autophagy and the ubiquitin-proteasome system
.
Neurosci. Lett.
 ,
469
,
112
116
.
26
Scotter
E.L.
,
Vance
C.
,
Nishimura
A.L.
,
Lee
Y.B.
,
Chen
H.J.
,
Urwin
H.
,
Sardone
V.
,
Mitchell
J.C.
,
Rogelj
B.
,
Rubinsztein
D.C.
et al
. (
2014
)
Differential roles of the ubiquitin proteasome system and autophagy in the clearance of soluble and aggregated TDP-43 species
.
J. Cell Sci.
 ,
127
,
1263
1278
.
27
Wang
I.F.
,
Tsai
K.J.
,
Shen
C.K.
(
2013
)
Autophagy activation ameliorates neuronal pathogenesis of FTLD-U mice: a new light for treatment of TARDBP/TDP-43 proteinopathies
.
Autophagy
 ,
9
,
239
240
.
28
Bose
J.K.
,
Huang
C.C.
,
Shen
C.K.
(
2011
)
Regulation of autophagy by neuropathological protein TDP-43
.
J. Biol. Chem.
 ,
286
,
44441
44448
.
29
Possik
E.
,
Jalali
Z.
,
Nouet
Y.
,
Yan
M.
,
Gingras
M.C.
,
Schmeisser
K.
,
Panaite
L.
,
Dupuy
F.
,
Kharitidi
D.
,
Chotard
L.
et al
. (
2014
)
Folliculin regulates AMPK-dependent autophagy and metabolic stress survival
.
PLoS Genet.
 ,
10
,
e1004273
.
30
Dunlop
E.A.
,
Seifan
S.
,
Claessens
T.
,
Behrends
C.
,
Kamps
M.A.
,
Rozycka
E.
,
Kemp
A.J.
,
Nookala
R.K.
,
Blenis
J.
,
Coull
B.J.
et al
. (
2014
)
FLCN, a novel autophagy component, interacts with GABARAP and is regulated by ULK1 phosphorylation
.
Autophagy
 ,
10
,
1749
1760
.
31
Tao
Z.
,
Wang
H.
,
Xia
Q.
,
Li
K.
,
Jiang
X.
,
Xu
G.
,
Wang
G.
,
Ying
Z.
(
2015
)
Nucleolar stress and impaired stress granule formation contribute to C9orf72 RAN translation-induced cytotoxicity
.
Hum. Mol. Genet.
 ,
24
,
2426
2441
.
32
Aulas
A.
,
Caron
G.
,
Gkogkas
C.G.
,
Mohamed
N.V.
,
Destroismaisons
L.
,
Sonenberg
N.
,
Leclerc
N.
,
Parker
J.A.
,
Vande Velde
C.
(
2015
)
G3BP1 promotes stress-induced RNA granule interactions to preserve polyadenylated mRNA
.
J. Cell Biol.
 ,
209
,
73
84
.
33
Kim
S.H.
,
Shi
Y.
,
Hanson
K.A.
,
Williams
L.M.
,
Sakasai
R.
,
Bowler
M.J.
,
Tibbetts
R.S.
(
2009
)
Potentiation of amyotrophic lateral sclerosis (ALS)-associated TDP-43 aggregation by the proteasome-targeting factor, ubiquilin 1
.
J. Biol. Chem.
 ,
284
,
8083
8092
.
34
Seguin
S.J.
,
Morelli
F.F.
,
Vinet
J.
,
Amore
D.
,
De Biasi
S.
,
Poletti
A.
,
Rubinsztein
D.C.
,
Carra
S.
(
2014
)
Inhibition of autophagy, lysosome and VCP function impairs stress granule assembly
.
Cell Death Differ.
 ,
21
,
1838
1851
.
35
Buchan
J.R.
,
Kolaitis
R.M.
,
Taylor
J.P.
,
Parker
R.
(
2013
)
Eukaryotic stress granules are cleared by autophagy and Cdc48/VCP function
.
Cell
 ,
153
,
1461
1474
.
36
Dewey
C.M.
,
Cenik
B.
,
Sephton
C.F.
,
Johnson
B.A.
,
Herz
J.
,
Yu
G.
(
2012
)
TDP-43 aggregation in neurodegeneration: are stress granules the key?
Brain Res.
 ,
1462
,
16
25
.
37
Ramaswami
M.
,
Taylor
J.P.
,
Parker
R.
(
2013
)
Altered ribostasis: RNA-protein granules in degenerative disorders
.
Cell
 ,
154
,
727
736
.
38
Li
Y.R.
,
King
O.D.
,
Shorter
J.
,
Gitler
A.D.
(
2013
)
Stress granules as crucibles of ALS pathogenesis
.
J. Cell Biol.
 ,
201
,
361
372
.
39
Buratti
E.
,
Baralle
F.E.
(
2010
)
The multiple roles of TDP-43 in pre-mRNA processing and gene expression regulation
.
RNA Biol.
 ,
7
,
420
429
.
40
Bose
J.K.
,
Wang
I.F.
,
Hung
L.
,
Tarn
W.Y.
,
Shen
C.K.
(
2008
)
TDP-43 overexpression enhances exon 7 inclusion during the survival of motor neuron pre-mRNA splicing
.
J. Biol. Chem.
 ,
283
,
28852
28859
.
41
Volkening
K.
,
Leystra-Lantz
C.
,
Yang
W.
,
Jaffee
H.
,
Strong
M.J.
(
2009
)
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.
 ,
1305
,
168
182
.
42
Fiesel
F.C.
,
Voigt
A.
,
Weber
S.S.
,
Van den Haute
C.
,
Waldenmaier
A.
,
Gorner
K.
,
Walter
M.
,
Anderson
M.L.
,
Kern
J.V.
,
Rasse
T.M.
et al
. (
2010
)
Knockdown of transactive response DNA-binding protein (TDP-43) downregulates histone deacetylase 6
.
EMBO J.
 ,
29
,
209
221
.
43
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
.
44
Alami
N.H.
,
Smith
R.B.
,
Carrasco
M.A.
,
Williams
L.A.
,
Winborn
C.S.
,
Han
S.S.
,
Kiskinis
E.
,
Winborn
B.
,
Freibaum
B.D.
,
Kanagaraj
A.
et al
. (
2014
)
Axonal transport of TDP-43 mRNA granules is impaired by ALS-causing mutations
.
Neuron
 ,
81
,
536
543
.
45
Nihei
Y.
,
Ito
D.
,
Suzuki
N.
(
2012
)
Roles of ataxin-2 in pathological cascades mediated by TAR DNA-binding protein 43 (TDP-43) and fused in sarcoma (FUS)
.
J. Biol. Chem.
 ,
287
,
41310
41323
.
46
Farg
M.A.
,
Soo
K.Y.
,
Warraich
S.T.
,
Sundaramoorthy
V.
,
Blair
I.P.
,
Atkin
J.D.
(
2013
)
Ataxin-2 interacts with FUS and intermediate-length polyglutamine expansions enhance FUS-related pathology in amyotrophic lateral sclerosis
.
Hum. Mol. Genet.
 ,
22
,
717
728
.
47
Moujalled
D.
,
James
J.L.
,
Yang
S.
,
Zhang
K.
,
Duncan
C.
,
Moujalled
D.M.
,
Parker
S.J.
,
Caragounis
A.
,
Lidgerwood
G.
,
Turner
B.J.
et al
. (
2015
)
Phosphorylation of hnRNP K by cyclin-dependent kinase 2 controls cytosolic accumulation of TDP-43
.
Hum. Mol. Genet.
 ,
24
,
1655
1669
.
48
Ritson
G.P.
,
Custer
S.K.
,
Freibaum
B.D.
,
Guinto
J.B.
,
Geffel
D.
,
Moore
J.
,
Tang
W.
,
Winton
M.J.
,
Neumann
M.
,
Trojanowski
J.Q.
et al
. (
2010
)
TDP-43 mediates degeneration in a novel Drosophila model of disease caused by mutations in VCP/p97
.
J. Neurosci.
 ,
30
,
7729
7739
.
49
Gitcho
M.A.
,
Strider
J.
,
Carter
D.
,
Taylor-Reinwald
L.
,
Forman
M.S.
,
Goate
A.M.
,
Cairns
N.J.
(
2009
)
VCP mutations causing frontotemporal lobar degeneration disrupt localization of TDP-43 and induce cell death
.
J. Biol. Chem.
 ,
284
,
12384
12398
.
50
Fung
G.
,
Shi
J.
,
Deng
H.
,
Hou
J.
,
Wang
C.
,
Hong
A.
,
Zhang
J.
,
Jia
W.
,
Luo
H.
(
2015
)
Cytoplasmic translocation, aggregation, and cleavage of TDP-43 by enteroviral proteases modulate viral pathogenesis
.
Cell Death Differ.
 , .
51
D'Ambrogio
A.
,
Buratti
E.
,
Stuani
C.
,
Guarnaccia
C.
,
Romano
M.
,
Ayala
Y.M.
,
Baralle
F.E.
(
2009
)
Functional mapping of the interaction between TDP-43 and hnRNP A2 in vivo
.
Nucleic Acids Res.
 ,
37
,
4116
4126
.
52
Nishimura
A.L.
,
Zupunski
V.
,
Troakes
C.
,
Kathe
C.
,
Fratta
P.
,
Howell
M.
,
Gallo
J.M.
,
Hortobagyi
T.
,
Shaw
C.E.
,
Rogelj
B.
(
2010
)
Nuclear import impairment causes cytoplasmic trans-activation response DNA-binding protein accumulation and is associated with frontotemporal lobar degeneration
.
Brain
 ,
133
,
1763
1771
.
53
Aulas
A.
,
Stabile
S.
,
Vande Velde
C.
(
2012
)
Endogenous TDP-43, but not FUS, contributes to stress granule assembly via G3BP
.
Mol. Neurodegener.
 ,
7
,
54
.
54
Xia
Q.
,
Wang
H.
,
Zhang
Y.
,
Ying
Z.
,
Wang
G.
(
2014
)
Loss of TDP-43 inhibits amyotrophic lateral sclerosis-linked mutant SOD1 aggresome formation in an HDAC6-dependent manner
.
J. Alzheimers Dis.
 ,
45
,
373
86
.
55
Betschinger
J.
,
Nichols
J.
,
Dietmann
S.
,
Corrin
P.D.
,
Paddison
P.J.
,
Smith
A.
(
2013
)
Exit from pluripotency is gated by intracellular redistribution of the bHLH transcription factor Tfe3
.
Cell
 ,
153
,
335
347
.
56
Yan
M.
,
Gingras
M.C.
,
Dunlop
E.A.
,
Nouet
Y.
,
Dupuy
F.
,
Jalali
Z.
,
Possik
E.
,
Coull
B.J.
,
Kharitidi
D.
,
Dydensborg
A.B.
et al
. (
2014
)
The tumor suppressor folliculin regulates AMPK-dependent metabolic transformation
.
J. Clin. Invest.
 ,
124
,
2640
2650
.
57
Tsun
Z.Y.
,
Bar-Peled
L.
,
Chantranupong
L.
,
Zoncu
R.
,
Wang
T.
,
Kim
C.
,
Spooner
E.
,
Sabatini
D.M.
(
2013
)
The folliculin tumor suppressor is a GAP for the RagC/D GTPases that signal amino acid levels to mTORC1
.
Mol. Cell
 ,
52
,
495
505
.
58
Petit
C.S.
,
Roczniak-Ferguson
A.
,
Ferguson
S.M.
(
2013
)
Recruitment of folliculin to lysosomes supports the amino acid-dependent activation of Rag GTPases
.
J. Cell Biol.
 ,
202
,
1107
1122
.
59
Liu
C.
,
Fei
E.
,
Jia
N.
,
Wang
H.
,
Tao
R.
,
Iwata
A.
,
Nukina
N.
,
Zhou
J.
,
Wang
G.
(
2007
)
Assembly of lysine 63-linked ubiquitin conjugates by phosphorylated alpha-synuclein implies Lewy body biogenesis
.
J. Biol. Chem.
 ,
282
,
14558
14566
.
60
Li
B.
,
Hu
Q.
,
Wang
H.
,
Man
N.
,
Ren
H.
,
Wen
L.
,
Nukina
N.
,
Fei
E.
,
Wang
G.
(
2010
)
Omi/HtrA2 is a positive regulator of autophagy that facilitates the degradation of mutant proteins involved in neurodegenerative diseases
.
Cell Death Differ.
 ,
17
,
1773
1784
.
61
Ren
H.
,
Fu
K.
,
Mu
C.
,
Li
B.
,
Wang
D.
,
Wang
G.
(
2010
)
DJ-1, a cancer and Parkinson's disease associated protein, regulates autophagy through JNK pathway in cancer cells
.
Cancer Lett.
 ,
297
,
101
108
.