TDP-43 real-time quaking induced conversion reaction optimization and detection of seeding activity in CSF of amyotrophic lateral sclerosis and frontotemporal dementia patients

Abstract

The pathological deposition of the transactive response DNA-binding protein of 43 kDa occurs in the majority (∼97%) of amyotrophic lateral sclerosis and in around 45% of frontotemporal lobar degeneration cases. Amyotrophic lateral sclerosis and frontotemporal lobar degeneration clinically overlap, presenting a continuum of phenotypes. Both amyotrophic lateral sclerosis and frontotemporal lobar degeneration lack treatments capable of interfering with the underlying pathological process and early detection of transactive response DNA-binding protein of 43 kDa pathology would facilitate the development of disease-modifying drugs. The real-time quaking-induced conversion reaction showed the ability to detect prions in several peripheral tissues of patients with different forms of prion and prion-like diseases. Despite transactive response DNA-binding protein of 43 kDa displays prion-like properties, to date the real-time quaking-induced conversion reaction technology has not yet been adapted to this protein. The aim of this study was to adapt the real-time quaking-induced conversion reaction technique for the transactive response DNA-binding protein of 43 kDa substrate and to exploit the intrinsic ability of this technology to amplify minute amount of mis-folded proteins for the detection of pathological transactive response DNA-binding protein of 43 kDa species in the cerebrospinal fluid of amyotrophic lateral sclerosis and frontotemporal lobar degeneration patients. We first optimized the technique with synthetic transactive response DNA-binding protein of 43 kDa-pre-formed aggregates and with autopsy-verified brain homogenate samples and subsequently analysed CSF samples from amyotrophic lateral sclerosis and frontotemporal lobar degeneration patients and controls. Transactive response DNA-binding protein of 43 kDa real-time quaking-induced conversion reaction was able to detect as little as 15 pg of transactive response DNA-binding protein of 43 kDa aggregates, discriminating between a cohort of patients affected by amyotrophic lateral sclerosis and frontotemporal lobar degeneration and age-matched controls with a total sensitivity of 94% and a specificity of 85%. Our data give a proof-of-concept that transactive response DNA-binding protein of 43 kDa is a suitable substrate for the real-time quaking-induced conversion reaction. Transactive response DNA-binding protein of 43 kDa real-time quaking-induced conversion reaction could be an innovative and useful tool for diagnosis and drug development in amyotrophic lateral sclerosis and frontotemporal lobar degeneration. The cerebrospinal fluid detection of transactive response DNA-binding protein of 43 kDa pathological aggregates may be exploited as a disease biomarker for amyotrophic lateral sclerosis and frontotemporal lobar degeneration patients.

Keywords: ALS; FTD; RT-QuIC; TDP-43; biomarker.

Graphical Abstract

Figure 1

Aggregation kinetics of recombinant TDP-43 and AFM seeds characterization. Purified seed-free recombinant HuTDP-43 (2 mg/mL) (A) and HuTDP-43(263-414) (0.2 mg/mL) (B) were induced to aggregate by alternating cycles of 60 s (s) of shaking and 60 s of incubation at 40°C (black lines). The final product of their aggregation was collected and efficiently employed as synthetic seed (green lines). ThT fluorescence is plotted against time. Each graph has a specific range of ThT fluorescence values and duration of a given reaction. The experiment was performed in triplicate and each replica was performed three times. Curves represent means for all experimental replicates and bars indicate the standard deviations (SD). (C, D) AFM analysis of TDP-43 aggregates. HuTDP-43 aggregates showed features of amorphous aggregates (C), whereas HuTDP-43(263-414) (D) acquired a fibrillary structure after aggregation.

Figure 2

RT-QuIC analysis of BH samples. In brief, 20 μL of sonicated and diluted (10⁻³) BHs collected from three FTLD-TDP patients (BH*, purple lines) and one CTRL patient (CTRL BH, black line) were added to purified seed-free recombinant HuTDP-43 (1 mg/mL) (A) and HuTDP-43(263-414) (0.2 mg/mL) (B) and analysed by means of RT-QuIC. The reaction was exposed to 60 s of shaking and 60 s of rest. All positive BHs efficiently seeded the reaction, whereas the negative control BH at the same dilution (10⁻³) did not affect the aggregation kinetics of the reaction. (C) Optimized protocol for BH analysis using purified seed-free HuTDP-43(263-414) (0.05 mg/mL) as substrate with a different reaction buffer (no NaCl, Gdn-HCl added) and a modified protocol (shaking at 100 rpm for 15 s every 30 min at 40°C). ThT fluorescence intensity was plotted against time. Each graph has a specific range of ThT fluorescence values and duration of a given reaction. The experiment was performed in triplicate and each replica was performed three times. Curves represent means for all experimental replicates and bars indicate the standard deviations (SD).

Figure 3

RT-QuIC analysis of ALS and FTLD-TDP CSF samples. In brief, 6 μL of undiluted CSF collected from patients with a known pathological TDP-43-related mutation (C9orf72, green lines, TARDBP, magenta lines, GRN, blue lines) and from age-matched controls (CTRLs, orange lines) were added to purified seed-free recombinant HuTDP-43(263-414) (0.05 mg/mL) and analysed by means of RT-QuIC with the optimized protocol (shaking at 100 rpm for 15 s every 30 min at 40°C). ThT fluorescence intensity is plotted against time. Dotted black lines represent the ThT fluorescence and lag-phase thresholds. In the first four panels (A–D) the curves for each group are shown separately, whereas in (E) samples are shown all together. The experiment was performed in triplicate and each replica was performed three times. Each curve shows the median aggregation curve for each sample.

Figure 4

RT-QuIC analysis of immune-depleted CSF samples. One true-positive (TP) and two false-positive (FP) CSF samples were immune-depleted of TDP-43 aggregates using magnetic beads coated with a cocktail of two anti-TDP-43 antibodies directed against two different epitopes of the protein. Samples before and after immune-depletion were added to purified seed-free recombinant HuTDP-43(263-414) (0.05 mg/mL) and analysed by means of RT-QuIC (shaking at 100 rpm for 15 s every 30 min at 40°C). ThT fluorescence intensity is plotted against time. Dotted black lines represent the ThT fluorescence and lag-phase thresholds. (A) After depletion, the true-positive sample tested negative (blue dashed line). After depletion, one of the false-positive samples became negative (red dashed line), whereas the other one remained positive, even if presenting a diminished intensity of ThT fluorescence emission (yellow dashed line). The experiment was performed in triplicates and each replica was performed three times. The image represents the median aggregation curve for each sample.

Figure 5

Performance of TDP-43 RT-QuIC with tau and α-synuclein aggregates. A previously screened negative CSF sample was spiked with 1.5 ng of tau K18, α-syn or TDP-43–pre-formed seeds. The CSF samples spiked with tau K18 (blue line) or α-syn (yellow line) fibrils presented no seeding activity in the optimized TDP-43 RT-QuIC. TDP-43 preformed aggregates diluted in the same negative CSF efficiently seeded the reaction (green line). Protocol details: 0.05 mg/mL of purified seed-free recombinant HuTDP-43(263-414) substrate; shaking at 100 rpm for 15 s every 30 min at 40°C. ThT fluorescence intensity is plotted against time. The experiment was performed in triplicates and each replica was performed three times. The image represents the median aggregation curve for each sample.

Figure 6

Performance of TDP-43 RT-QuIC with TDP-43 CSF. Different dilutions of in vitro-obtained HuTDP-43(263-414) aggregates were added to a previously screened negative CSF sample. TDP-43 RT-QuIC was able to detect as little as 15 pg of TDP-43 pathological aggregates. Protocol details: 0.05 mg/mL of purified seed-free recombinant HuTDP-43(263-414) substrate; shaking at 100 rpm for 15 s every 30 min at 40°C. ThT fluorescence intensity is plotted against time. Dotted black lines represent the ThT fluorescence and lag-phase thresholds. The experiment was performed in triplicate and each replica was performed three times. The image represents the median aggregation curve for each sample.