Inhibition of TDP-43 accumulation by bis(thiosemicarbazonato)-copper complexes

Abstract

Amyotrophic lateral sclerosis (ALS) is a progressive, fatal, motor neuron disease with no effective long-term treatment options. Recently, TDP-43 has been identified as a key protein in the pathogenesis of some cases of ALS. Although the role of TDP-43 in motor neuron degeneration is not yet known, TDP-43 has been shown to accumulate in RNA stress granules (SGs) in cell models and in spinal cord tissue from ALS patients. The SG association may be an early pathological change to TDP-43 metabolism and as such a potential target for therapeutic intervention. Accumulation of TDP-43 in SGs induced by inhibition of mitochondrial activity can be inhibited by modulation of cellular kinase activity. We have also found that treatment of cells and animal models of neurodegeneration, including an ALS model, with bioavailable bis(thiosemicarbazonato)copper(II) complexes (Cu(II)(btsc)s) can modulate kinase activity and induce neuroprotective effects. In this study we examined the effect of diacetylbis(-methylthiosemicarbazonato)copper(II) (Cu(II)(atsm)) and glyoxalbis(-methylthiosemicarbazonato)copper(II) (Cu(II)(gtsm)) on TDP-43-positive SGs induced in SH-SY5Y cells in culture. We found that the Cu(II)(btsc)s blocked formation of TDP-43-and human antigen R (HuR)-positive SGs induced by paraquat. The Cu(II)(btsc)s protected neurons from paraquat-mediated cell death. These effects were associated with inhibition of ERK phosphorylation. Co-treatment of cultures with either Cu(II)(atsm) or an ERK inhibitor, PD98059 both prevented ERK activation and blocked formation of TDP-43-and HuR-positive SGs. Cu(II)(atsm) treatment or ERK inhibition also prevented abnormal ubiquitin accumulation in paraquat-treated cells suggesting a link between prolonged ERK activation and abnormal ubiquitin metabolism in paraquat stress and inhibition by Cu. Moreover, Cu(II)(atsm) reduced accumulation of C-terminal (219-414) TDP-43 in transfected SH-SY5Y cells. These results demonstrate that Cu(II)(btsc) complexes could potentially be developed as a neuroprotective agent to modulate neuronal kinase function and inhibit TDP-43 aggregation. Further studies in TDP-43 animal models are warranted.

Figure 1. Effect of Cu^II(atsm) and Cu^II(gtsm) on TDP-43 and HuR localization in SY5Y cells.

Cells were exposed overnight with 1 mM paraquat in the presence or absence of 1 µM Cu^II(atsm) or 50 nM Cu^II(gtsm) and TDP-43 and HuR localization was examined by immunofluorescence. A–D: untreated, E–H: paraquat, I–L: paraquat and Cu^II(atsm), M–P: paraquat and Cu^II(gtsm). Green  =  TDP-43, red  =  HuR, blue  =  DAPI. Bottom panels  =  merged images of TDP-43 and HuR above. Arrows indicate TDP-43 and HuR SGs. Bar  = 10 µm.

Figure 2. Effect of Cu^II(btsc)s and metals on TDP-43 and HuR SGs in SY5Y cells.

Cells were treated overnight with 1 mM paraquat in the presence or absence of 1 µM Cu^II(atsm), 1 and 100 µM CuCl₂ and 100 µM FeCl₂. A–B: untreated, C–D: paraquat, E–F: paraquat and Cu^II(atsm), G–H: paraquat and 1 µM CuCl₂, I–J: paraquat and 100 µM CuCl₂, K–L: paraquat and 100 µM FeCl₂. Green  =  TDP-43, blue  =  DAPI. Arrows indicate SGs. Bar  = 10 µm. M: SGs (TDP-43 or HuR-positive) per cell after overnight treatment with 1 mM paraquat and co-treatment with Cu^II(atsm), Cu^II(gtsm), CuCl₂ (1 and 100 µM) or FeCl₂. **p<0.01 compared to paraquat treatment.

Figure 3. Effect of Cu^II(atsm) on SGs in HeLa cells induced by sodium arsenite.

Cells were treated for 1 hr with 500 µM sodium arsenite in the presence or absence of 1 µM Cu^II(atsm) (1 hr) and TDP-43 and HuR were examined by immunofluoresence. A–C: untreated, D–F: sodium arsenite treated, G–I: sodium arsenite and Cu^II(atsm). Green  =  TDP-43, red  =  HuR, blue  =  DAPI. Arrows indicate TDP-43 and HuR SGs. Bar  = 10 µm. J: SGs (TDP-43 or HuR-positive) per cell after 1 hr treatment with 500 µM sodium arsenite and co-treatment with 1 µM Cu^II(atsm).

Figure 4. Effect of Cu^II(atsm) on TDP-43, HuR and kinase phosphorylation.

Graphs show densitometric analysis of altered protein expression or phosphorylation from two-three separate experiments relative to loading controls. A: Cells were treated with 1 mM paraquat overnight in the presence or absence of 20 µM SP600125, 1 µM Cu^II(atsm) or 1 µM CuCl₂ and immunoblotted for TDP-43 or HuR. B–C: Cells were treated for 2 hr with paraquat in the presence or absence of 1 µM Cu^II(atsm), 1 µM CuCl₂ or 10 µM PD98059 (ERK inhibitor). Cells were immunoblotted for phospho-ERK (p-ERK), total ERK, phospho-JNK (p-JNK) and total JNK. D–E: Cells were treated overnight with paraquat in the presence or absence of 1 µM Cu^II(atsm), 1 µM CuCl₂ or 10 µM PD98059 (ERK inhibitor). Cells were immunoblotted for phospho-ERK (p-ERK), total ERK, phospho-JNK (p-JNK) and total JNK. F: Effect of CuCl₂ on ERK and JNK phosphorylation. Cells were treated overnight with 1 mM paraquat or paraquat plus 100 µM CuCl₂ and cells were examined for expression ERK and JNK phosphorylation. G: HeLa cells were treated for 1 hr with 500 µM sodium arsenite in presence or absence of 1 µM Cu^II(atsm). Cells were immunoblotted for phospho-ERK (p-ERK), total ERK, phospho-JNK (p-JNK) and total JNK. *p<0.05 compared to untreated control. **p<0.05 compared to treatment alone.

Figure 5. Effect of Cu^II(atsm) and kinase inhibitors on TDP-43 and HuR-positive SG formation.

Cells were treated overnight with 1 mM paraquat in the presence or absence of 1 µM Cu^II(atsm), 10 µM PD98059 or 20 µM SP600125 and examined for TDP-43 and HuR by immunofluorescence. A–C: untreated, D–F: paraquat, G–I: paraquat and Cu^II(atsm), J–L: paraquat and PD98059, M–O: paraquat and SP600125. Green  =  TDP-43, red  =  HuR, blue  =  DAPI. Arrows indicate TDP-43 and HuR SGs. Bar  = 10 µm. P: SGs (TDP-43 or HuR-positive) per cell after overnight treatment with 1 mM paraquat and co-treatment with Cu^II(atsm) (1 µM), PD98095 or SP600125. **p<0.01 compared to paraquat treatment.

Figure 6. Effect of Cu^II(atsm) on paraquat-mediated changes to ubiquitin.

Cells were treated with 1 mM paraquat overnight in the presence or absence of 1 µM Cu^II(atsm), 10 µM PD98059 or 100 µM CuCl₂. Cells were also exposed overnight to 75 µM rotenone. Cells were examined for TDP-43 and ubiquitin by immunofluorescence. A–C: untreated, D–F: paraquat, G–I: rotenone, J–L: paraquat and Cu^II(atsm), M–O: paraquat and PD98059 (ERK inhibitor), P–R: paraquat and CuCl₂. Green  =  TDP-43, red  =  ubiquitin, blue  =  DAPI. Arrowheads indicate ubiquitin aggregation. Bar  = 10 µm. S: Percentage of cells containing ubiquitin aggregates after overnight treatment with 1 mM paraquat, 75 µM rotenone, paraquat plus 1 µM Cu^II(atsm), paraquat plus PD98059 or paraquat plus 100 µM CuCl₂. **p<0.01 compared to paraquat treatment.

Figure 7. Effect of Cu^II(btsc)s on paraquat toxicity.

A: Cells were treated overnight with 1 or 10 mM paraquat in the presence or absence of 1 µM Cu^II(atsm), 50 nM Cu^II(gtsm) or 100 µM CuCl₂ and cell death was measured with the LDH assay. *p<0.01 compared to untreated cells. **p<0.01 compared to 10 mM paraquat alone. B: In-gel SOD activity assay. Cells were treated with 1 mM paraquat overnight alone or in the presence of 1 µM Cu^II(atsm), 50 nM Cu^II(gtsm) or 1 µM CuCl₂. Cu^II(atsm), Cu^II(gtsm) and CuCl₂ were also added at the same concentration to cultures without paraquat. Cell lysates were analyzed for SOD activity using a SOD zymography gel assay. The band representing SOD activity was analyzed by densitometry and revealed no significant change compared to untreated control.

Figure 8. Effect of Cu^II(atsm) on aggregation of CTF-TDP-43 219–414.

Cells were transfected with WT TDP-43-GFP (full length) or CTF-TDP-43 219–414-GFP and formation of TDP-43 aggregates was examined after 48 hr. A–C: Aggregates of TDP-43 were not observed in cells transfected with WT-TDP-43. D–F: Upon transfection with CTF-TDP-43 219–414, cells revealed widespread formation of cytosolic aggregates after 48 hr. G–I: Addition of Cu^II(atsm) (1 µM) to cells at 24 hr resulted in a reduction in frequency of cells expressing CTF-TDP-43 219–414 aggregates at 48 hr. J–L: Addition of CuCl₂ (1 µM) to cells at 24 hr resulted in no significant change to the frequency of cells expressing CTF-TDP-43 219–414 aggregates at 48 hr. C, F, I and L represent merged images of TDP and DAPI. Arrows indicate TDP-43 aggregates. Green  =  TDP-43, blue  =  DAPI. Bar  = 10 µm.