Multiplex SILAC analysis of a cellular TDP-43 proteinopathy model reveals protein inclusions associated with SUMOylation and diverse polyubiquitin chains

Abstract

Transactive response (TAR) DNA-binding protein 43 (TDP-43) is a major protein component within ubiquitin-positive inclusions of frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Although TDP-43 is a nuclear DNA/RNA-binding protein, in pathological conditions, TDP-43 has been reported to redistribute to the cytoplasm where it is cleaved and forms insoluble, ubiquitinated, and phosphorylated inclusions. Here we present a cellular model in which full-length human TDP-43 or a splicing isoform (TDP-S6) that lacks the C terminus is overexpressed in a human cell line and mouse primary neurons. Whereas recombinant and endogenous TDP-43 was primarily localized in the nucleus, the shorter TDP-S6 formed highly insoluble cytoplasmic and nuclear inclusions reminiscent of disease-specific pathology. Western blot analysis of detergent-insoluble extracts showed an increase in high molecular weight immunoreactive species for TDP-S6 compared with TDP-43, consistent with ubiquitination or ubiquitin-like modifications. We used a multiplex stable isotope labeling with amino acids in cell culture approach to compare the detergent-insoluble proteome from mock-, TDP-43-, and TDP-S6-transfected cells. TDP-S6 overexpression caused a concomitant increase in both ubiquitin (Ub) and the small Ub-like modifier-2/3 (SUMO-2/3) within the insoluble proteome. Similarly, full-length TDP-43 overexpression also resulted in the elevation of SUMO-2/3. Immunofluorescence showed strong co-localization of endogenous Ub with both cytoplasmic and nuclear TDP-S6 inclusions, whereas SUMO-2/3 was co-localized mainly with the nuclear inclusions. Quantitative mass spectrometry further revealed that mixed Lys-48 and Lys-63 polyUb linkages were associated with the TDP insoluble fractions. Together our data indicate that expression of a TDP-43 splice variant lacking a C terminus recapitulates many of the cellular and biochemical features associated with disease pathology and that the interplay of ubiquitination and SUMOylation may have an important role in TDP-43 regulation.

Fig. 1.

Expression of TDP-43 and shorter alternative splicing isoform, TDP-S6, in HEK-293 cells. A, TDP-43 consists of six exons that encode a protein of 414 amino acids. RRM1 spans exons 3 and 4, RRM2 spans exons 5 and 6, and the glycine-rich C-terminal domain lies within exon 6. TDP-S6 is generated by an additional splicing event resulting in a reading frameshift after amino acid 277. Thus, TDP-S6 contains 18 unique amino acids (278–295) on its C terminus (grey) corresponding to residual intron RNA included during an alternative splicing event. B, Western blotting (WB) analysis of transfected control (mock plasmid), HA-tagged TDP-43, or TDP-S6. C, loss of TDP-43 phosphorylation during phosphatase treatment. The HA-TDP-43-transfected total cell lysate was incubated for 2.5 h with the addition of an increasing concentration of alkaline phosphatase followed by Western analysis. Phosphorylated TDP-43 isoforms are indicated by asterisks.

Fig. 2.

TDP-S6 translocation and aggregation in HEK-293 cells and primary hippocampal neurons. A, HEK-293 cells were transfected with HA-TDP-43 or HA-TDP-S6 and stained with TDP-43 antibody for both recombinant and endogenous proteins (green), HA antibody for recombinant TDP-43 exclusively (red), and Hoechst stain for the nucleus (blue). The white bars represent a distance of 10 μm. B, neurons were transfected and stained with the neuron-specific marker enolase (green), HA antibody (red), and Hoechst stain (blue). Two different sections of the same neuron, corresponding to separate focal planes, are shown in the middle and bottom panels of TDP-S6 (Soma), indicating cytoplasm and nucleus, respectively. C, TDP-S6 was also aggregated in neuronal processes, shown under higher magnification.

Fig. 3.

Characterization of Sarkosyl-insoluble fraction by Western blotting and mass spectrometry. A, Western blot (WB) analysis of sarkosyl-soluble and -insoluble HA-TDP-43, native TDP-43, and HA-TDP-S6 using an anti-TDP-43 antibody. The asterisk (*) indicates phosphorylated forms. B, densitometry analysis of multiple TDP-43 forms. C, large scale profiling of Sarkosyl-soluble and -insoluble fractions by LC-MS/MS. Spectral counts of identified proteins were used as a semiquantitative index to derive probability values. snRNP, small nuclear ribonucleoprotein.

Fig. 4.

TDP-43 and TDP-S6 display post-translational modifications reminiscent of biochemical signature observed in neurodegeneration. The cells were transfected with mock, HA-TDP-43, or HA-TDP-S6 plasmids and extracted by a Sarkosyl buffer (A), and the pellet was solubilized with urea buffer (B). The sequentially solubilized proteins were analyzed by Western blotting (WB). In detergent-insoluble fractions, modified TDP-43 phosphorylated species (marked by asterisks), proteolytic fragments (marked by arrowheads), and high molecular mass TDP immunoreactivity are observed. An increase in global protein ubiquitination with TDP-43 and more dramatically with TDP-S6 overexpression is observed using a Ub-specific antibody (B, right panel).

Fig. 5.

Quantitative proteomics of Sarkosyl-insoluble fraction by multiplex SILAC. A, diagram for biological replicates of SILAC analysis (experiments 1 and 2). The amino acid labeling order was swapped in the repeated analysis. B, silver-stained SDS gel of the isolated Sarkosyl-insoluble fractions. The gel lanes were excised as indicated. C, a representative fitting of a pairwise comparison in the first experiment, shown as a histogram fitted to a Gaussian curve. D, the experimental variations were similar in multiple comparisons, indicated by the values of S.D. L, light; M, medium; H, heavy; Ctl, control. GeLC-MS/MS is the method of one dimensional SDS gel and LC-MS/MS.

Fig. 6.

SILAC analysis of TDP insoluble proteome and characterization of SUMOylation. Shown are MS spectra of detected peptide ions for MOV10 (a negative control), TDP proteins, Ub, and SUMO-2/3 in one dimensional SDS gel and LC-MS/MS from the top gel band (>180 kDa) in experiments 1 (A) and 2 (B). C, Western blotting of SUMO-2/3 and HA in Sarkosyl-insoluble fractions from cells transfected with mock, TDP-43, and TDP-S6 plasmids. D, both Sarkosyl-soluble (10 μg) and -insoluble (1 μg) fractions were incubated with SENP2 followed by immunoblotting with antibodies against SUMO-2/3, TDP-43, and HA. The three bands indicated by arrows correspond to TDP-43 modified by one, two, or three SUMO-2/3 molecules, respectively. WB, Western blot.

Fig. 7.

Localization of SUMO-2/3, polyubiquitin (Lys-48 and Lys-63), and TDP proteins in HEK-293 cells. Cells were transfected with plasmids expressing HA-TDP-43 or HA-TDP-S6 and then co-stained with HA antibodies (red), and other antibodies (green) against SUMO-2/3 (A) and Lys-48 (B) and Lys-63 (C) polyUb linkages. The nuclei were stained by Hoechst (blue). Nuclear SUMO-2/3 and TDP-43 co-localization is indicated by white arrowheads (A).

Fig. 8.

Multiple polyUb linkages were detected in TDP-S6 and TDP-43 insoluble proteome. Transfected cells were harvested and mixed with metabolically labeled cells as internal standards followed by differential extraction. The Sarkosyl-insoluble samples were used for SRM-based analysis to quantify all polyUb linkages and two other proteins (E1 and Rpn2 as loading control). A, the monitored product ion pair of Lys-48 linkage-specific peptide. B, the co-elution of Lys-48 linkage peptide pair during an LC/SRM run. C, the comparison of measured linkages and proteins in three samples: mock-, HA-TDP-43- or HA-TDP-S6-transfected cells. The data were normalized to values in the mock sample. Error bars indicate S.E., and significance (*) was determined by performing an unpaired two-tailed t test (p < 0.01). WT, wild type (Full length TDP-43).