TDP-43 neurotoxicity and protein aggregation modulated by heat shock factor and insulin/IGF-1 signaling

Abstract

TAR DNA-binding protein 43 (TDP-43) plays a key role in the neurodegenerative diseases including amyotrophic lateral sclerosis and frontotemporal lobar degeneration. The nature of the TDP-43-mediated neurotoxicity associated with these diseases is not yet understood. Here, we have established transgenic Caenorhabditis elegans models that express human TDP-43 variants in the nervous system, including the full-length wild-type (WT) and mutant proteins and a pathologic C-terminal fragment. The C. elegans models developed severe locomotor defects associated with the aggregation of TDP-43 in neurons. In comparison to parallel Cu/Zn superoxide dismutase worm models, transgenic full-length TDP-43, including the WT protein, was highly neurotoxic. In addition, TDP-43 demonstrated an unusually high tendency to aggregate, a property intrinsic to the WT protein. The C-terminal 25 kDa fragment of TDP-43 was unstable but remarkably aggregation-prone. Distinct disulfide-linked TDP-43 dimers and oligomers were detected. In C. elegans, the neurotoxicity and the protein aggregation of TDP-43 were regulated by environmental temperature and heat shock transcriptional factor 1, indicating that a deficiency in protein quality control is a risk factor for TDP-43 proteinopathy. Furthermore, the neurotoxicity and the protein aggregation of TDP-43 can be significantly attenuated by a deficiency in the insulin/insulin-like growth factor 1 (IGF-1) signaling in C. elegans and mammalian cells. These results suggest that protein misfolding underlies the aging-dependent neurodegeneration associated with TDP-43 and that the insulin/IGF-1 signaling may be a target for therapies.

Figure 1.

Transgenic C. elegans expressing neuronal TDP-43 develops pronounced movement defects associated with protein aggregation. (A) Relative motility, as measured by the thrashing rate in liquid medium, was compared among 1-day adult C. elegans expressing variants of human SOD1-YFP or TDP-43-YFP under the control of the snb-1 promoter. n = 32, error bars represent the standard error of means (SEM). The TDP-43 strains had significantly worse locomotion than the mutant SOD1 strain (*P < 0.05). The TDP-C25 strain had better movement than the other TDP-43 strains (**P < 0.05). The expression levels of total SOD1 or TDP-43 proteins were shown by immunoblotting against YFP. (B) The insolubility assay as a measure of the aggregation of SOD1-YFP or TDP-43-YFP variants. From detergent-extracted C. elegans homogenates, 10 µg (∼1/20) of soluble supernatant protein and 5 µg (∼1/4) of insoluble pellet protein were analyzed by immunoblotting. (C) Nuclear localization of full-length TDP-43-YFP and cytoplasmic localization of the fragment TDP-C25-YFP. Representative images of motor neurons in the ventral cord stained with DAPI are shown. Note the well-demarked aggregates of TDP-C25-YFP of various sizes in the cytoplasm. (D) Schematic drawing of the ventral and dorsal cords in the C. elegans nervous system. The neuronal cell bodies (arrow) are located in the ventral cord, and the neuronal processes (arrowheads) are projected circumferentially and located in both ventral and dorsal cords. (E) In addition to the large aggregates of TDP-C25-YFP in the neuronal cell bodies, smaller aggregates were observed in the neuronal processes of both ventral and dorsal cords. Scale bars: 5 µm.

Figure 2.

FRAP analysis of TDP-43 aggregates in C. elegans neurons. (A) Representative images from the FRAP analysis of neurons expressing YFP only, WT TDP-43-YFP localized to the nucleus and TDP-C25-YFP, which form aggregates in the cytoplasm. The circles mark the photobleached region. Colors indicate high (H) or low (L) intensity of fluorescence. (B) The ratio of the intensities in bleached and adjacent unbleached regions is used to assess the diffusion rate of the fluorescent proteins. n = 5, error bars represent the SEM.

Figure 3.

High aggregation potential of TDP-43 proteins in human cells. (A) N-terminal Myc-tagged TDP-43 and SOD1 were transiently overexpressed in HEK293T cells, and the lysates were extracted with detergents. Two micrograms (∼1/500) of soluble supernatant protein and 8 µg (∼1/50) of insoluble pellet protein were analyzed by denaturing and reducing SDS–PAGE followed by immunoblotting with anti-Myc antibodies. (B) The relative aggregation potential, as measured by the ratio of insoluble pellet proteins to soluble supernatant proteins, is compared among variants of SOD1 and TDP-43. TDP-43 is more aggregation-prone than mutant G85R SOD1 (*P < 0.05). A nominal aggregation potential for TDP-C25 is given because of high degree of instability of the soluble protein. n = 3, error bars represent the SEM. (C) Cycloheximide chase experiments for TDP-43 proteins. Twenty-four hours after the transfection of the Myc-tagged proteins, HeLa cells were treated with 200 µg/ml of cycloheximide. The cells were harvested at various time points during the subsequent 24 h. The cells were lysed with the extraction buffer and centrifuged, and 8 µg of soluble supernatant protein was analyzed by immunoblotting with the anti-Myc antibody (long exposure for the TDP-C25 immunoblot). The TDP-C25 protein was estimated to have a half-life of 3.3 h, when compared with 11.3 h for the WT TDP-43 protein. (D) Endogenous TDP-43 in HEK293T cells is relatively enriched in insoluble pellets, when compared with the endogenous SOD1. (E) Nuclear localization of full-length Myc-TDP-43 and cytoplasmic localization of aggregated Myc-TDP-C25 are indicated by immunofluorescence staining in transfected HEK293T cells. Scale bar: 5 µm.

Figure 4.

TDP-43 protein dimers and oligomers linked by disulfide bonds are enriched in aggregates in vivo. (A) N-terminal Myc-tagged TDP-43 and SOD1 variants were transfected into HEK293T cells, and the lysates were detergent extracted in the presence of the thiol blocker iodoacetamide. The insoluble pellet and the soluble supernatant fractions were both analyzed with denaturing and non-reducing SDS–PAGE without β-mercaptoethanol. The top panel shows the pellet fraction in which the aggregated protein was enriched. In contrast to mutant SOD1 G85R, which is ‘glued’ together by disulfide bonds into large aggregates, distinct TDP-43 dimers (arrows) and oligomers (arrowhead) dependent on β-mercaptoethanol are observed. The bottom panel shows the supernatant fraction, in which the majority of soluble TDP-43 or SOD1 proteins are not disulfide cross-linked. The doublet bands of WT SOD1 are due to its intramolecular disulfide bond. A fraction of soluble WT SOD1 protein was disulfide-reduced and subsequently modified by iodoacetamide, resulting in a slower migration, as described previously (68). The soluble SOD1 G85R protein has a characteristic faster migration than WT SOD1 and is presumably all in a reduced state. (B) Mutating the only cysteine in TDP-C25 (C243A) blocked the formation of the insoluble dimers (arrow), but the aggregation of the majority of the protein was not significantly affected.

Figure 5.

Elevated environmental temperature increases protein aggregation and neurotoxicity in transgenic TDP-43 C. elegans. Both WT TDP-43-YFP (A) and TDP-C25-YFP (B) proteins show increased accumulation in soluble and insoluble fractions at 25°C when compared with those at 20°C (*P < 0.05; n = 3). Ten micrograms (∼1/20) of soluble supernatant protein and 5 µg (∼1/4) of insoluble pellet protein were analyzed by immunoblotting. (C) Elevated temperature increases the accumulation of WT TDP-43-YFP in C. elegans neurons. Anterior (A) and posterior (P) regions of heads of L4 animals are shown. (D) Aggregates of TDP-C25-YFP are markedly increased in number and size in response to the elevated temperature. Head neurons of L1 animals are shown. (E) Quantitation of the effects of temperature elevation on the locomotor behavior of 1-day adult TDP-C25-YFP animals (*P < 0.05). n = 32, error bars represent the SEM. Scale bars: 5 µm.

Figure 6.

HSF-1 robustly modulates protein aggregation and neurotoxicity in transgenic TDP-43 C. elegans. (A and B) Reduction in hsf-1 by RNAi increases the accumulation of WT TDP-43-YFP and TDP-C25-YFP in C. elegans neurons. Integrated TDP-43 transgenic C. elegans lines were sensitized for RNAi treatment by crossing into a background of eri-1(mg366);lin-15B(n744) and fed with bacteria expressing double-stranded RNA targeting hsf-1. The hsf-1 RNAi increases the accumulation of TDP-43-YFP in the neuronal nuclei and the aggregates of TDP-C25-YFP in the cytoplasm. Anterior (A) and posterior (P) regions of heads of L4 animals are shown. (C) Since homozygous TDP-C25-YFP transgenic animals fail to thrive in the presence of hsf-1(sy441), quantitation of the effects on the locomotor behavior was carried out on TDP-C25-YFP hemizygous transgenic animals (TDP-C25-YFP +/−; 1-day adult) (*P < 0.05). n = 32, error bars represent the SEM. (D) In the presence of loss-of-function mutant hsf-1(sy441), aggregates of TDP-C25-YFP are markedly increased in number and size in response to the elevated temperature. Head neurons of L1 animals are shown. Scale bars: 5 µm.

Figure 7.

Loss-of-function mutant daf-2(e1370) suppresses neurotoxicity and aggregation of WT full-length TDP-43-YFP in C. elegans. (A) The neurotoxicity of WT TDP-43-YFP in the C. elegans models, as measured by the thrashing rate in liquid medium, is significantly suppressed by a loss-of-function allele of the insulin/IGF1R daf-2, e1370. n = 32, error bars represent the SEM. (B) Loss of function of daf-2 decreases the soluble and insoluble aggregated TDP-43 proteins in C. elegans. The C. elegans homogenates were detergent extracted, and 10 µg (∼1/20) of soluble supernatant protein and 5 µg (∼1/4) of insoluble pellet protein were analyzed by immunoblotting. (C) Quantification of the TDP-43 proteins levels (*P < 0.05, n = 3). (D) Loss of function of daf-2 decreases the accumulation of full-length TDP-43-YFP in the nuclei of C. elegans neurons. Representative 1-day adult images are shown. Scale bars: 5 µm.

Figure 8.

daf-2(e1370) and daf-16(mu86) exert opposing effects on neurotoxicity and aggregation of TDP-C25-YFP in C. elegans. (A) The neurotoxicity of TDP-C25-YFP in the C. elegans models, as measured by the thrashing rate in liquid medium, is significantly suppressed by the loss-of-function mutant daf-2(e1370) but increased by the loss-of-function allele daf-16(mu86). n = 32, error bars represent the SEM. (B) daf-2(e1370) and daf-16(mu86) show opposing effects on the aggregation of TDP-C25-YFP. Ten micrograms (∼1/20) of soluble supernatant protein and 5 µg (∼1/4) of insoluble pellet protein were analyzed by immunoblotting. (C) Quantification of the TDP-C25 proteins levels (*P < 0.05, n = 3). (D) Loss of function of daf-2 decreases the number and size of neuronal aggregates of TDP-C25. Representative L2 worm images are shown. Scale bar: 15 µm.

Figure 9.

Reduction in the function of insulin/IGF1R suppresses the aggregation of TDP-43 in human cells. (A) Knockdown of IGF1R by shRNA in HEK293T cells decreases the accumulation of soluble and insoluble aggregated Myc-TDP-43 Q331K. Twelve micrograms (∼1/50) of soluble supernatant protein and 15 µg (∼1/10) of insoluble pellet protein were analyzed by immunoblotting with the anti-Myc antibody. (B) Quantitation of the knockdown of the IGF1R protein expression is shown. (C) Quantitation of soluble and insoluble Myc-TDP-43 Q331K protein levels after the knockdown of IGF1R. *P < 0.05, n = 3, error bars represent the SEM.

Figure 10.

A model of TDP-43 misfolding and aggregation as a modulator of its toxicity. A schematic representation of the mechanisms through which the HSF-1 and IGF-1 pathways may regulate the folding and aggregation of TDP-43 proteins. The normal DNA/RNA-binding functions of TDP-43 could be disrupted by the misfolding and aggregation of the protein. The nuclear and cytoplasmic aggregates formed by misfolded TDP-43 and its fragments, such as TDP-C25, may further perturb cellular functions by burdening the protein cellular control. The HSF-1 and IGF-1 signaling may regulate the misfolding and toxicity of TDP-43 through modulation of the protein quality-control system, such as the molecular chaperone networks. Oval and hexagon shapes represent native and misfolded TDP-43, respectively. HSP, heat shock protein; pA, polyadenylation.