Characterization of a series of 4-aminoquinolines that stimulate caspase-7 mediated cleavage of TDP-43 and inhibit its function

Abstract

Dysfunction of the heterogeneous ribonucleoprotein TAR DNA binding protein 43 (TDP-43) is associated with neurodegeneration in diseases such as amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD). Here we examine the effects of a series of 4-aminoquinolines with affinity for TDP-43 upon caspase-7-induced cleavage of TDP-43 and TDP-43 cellular function. These compounds were mixed inhibitors of biotinylated TG6 binding to TDP-43, binding to both free and occupied TDP-43. Incubation of TDP-43 and caspase-7 in the presence of these compounds stimulated caspase-7 mediated cleavage of TDP-43. This effect was antagonized by the oligonucleotide TG12, prevented by denaturing TDP-43, and exhibited a similar relation of structure to function as for the displacement of bt-TG6 binding to TDP-43. In addition, the compounds did not affect caspase-7 enzyme activity. In human neuroglioma H4 cells, these compounds lowered levels of TDP-43 and increased TDP-43 C-terminal fragments via a caspase-dependent mechanism. Subsequent experiments demonstrated that this was due to induction of caspases 3 and 7 leading to increased PARP cleavage in H4 cells with similar rank order of the potency among the compounds tests for displacement of bt-TG6 binding. Exposure to these compounds also reduced HDAC-6, ATG-7, and increased LC3B, consistent with the effects of TDP-43 siRNA described by other investigators. These data suggest that such compounds may be useful biochemical probes to further understand both the normal and pathological functions of TDP-43, and its cleavage and metabolism promoted by caspases.

Figure 1

Structures of 4-aminoquinoline 1 and 2 with affinity for TDP-43.

Figure 2. 4-aminoquinoline 1 exhibits mixed inhibition of bt-TG6 binding to TDP-43 using AlphaScreen^® technology

TDP-43 (0.1 nM) and 10 μg/mL of both streptavidin donor beads and anti-GST acceptor beads were incubated with various concentrations of biotinylated-TG6 (bt-TG6) and 1 for 3 hr at room temperature as described in section 2.2. (A) Saturation binding isotherms of bt-TG6 binding to TDP-43 in the absence (○), and presence of 1.0 μM (□), 3.2 μM (△), or 10 μM (◇) of 1. Data fit best to the mixed model inhibition equation compared to competitive inhibition equation (P < 0.01) in Graph Pad Prism^® with a K_i value of 0.61 μM and α value of 7.0. (B) Dose-response curves for 1 in the presence of 0.32 nM (◇), 1.0 nM (△), 3.2 nM (□), 10 nM (○) bt-TG6. (C) Linear regression of pIC₅₀ values from (B) vs log [bt-TG6] yields a slope of 0.51. Data represent mean ± S.E.M. of 4 determinations.

Figure 3. Stimulation of caspase-7 mediated cleavage of TDP-43 by 4-aminoquinolines is due to a specific interaction with TDP-43

(A) Time dependence of 2 stimulation of caspase-7 mediated cleavage of TDP-43. TDP-43 (2 μg/mL) was preincubated with 10 μM of 2 for 1 hr followed by the addition of caspase-7 (5 U/mL). Reactions were incubated at 37 °C and terminated at the times indicated with SDS sample buffer followed by immunoblotting (section 2.3). (B) TG12 mediated antagonism of 2 stimulation of caspase-7 mediated cleavage of TDP-43. TDP-43 (2 μg/mL) was preincubated for 1 hr with indicated concentrations of 2 in the presence and absence of 1 μM TG12 followed by the addition of caspase-7. Reactions were terminated after 1 hr at 37 °C with SDS sample buffer. (C) Stimulation of caspase-7 cleavage of TDP-43 requires correctly folded TDP-43. TDP-43 (0.2 mg/mL) was incubated in the absence and presence of 6 M guanidine HCl for 1 hr, followed by dialysis for 3 hr at room temperature. Dialyzed TDP-43 was then diluted to 4 μg/mL and preincubated in the presence and absence of 2 for 1 hr followed by addition of caspase-7 and subsequent 1 hr incubation at 37 °C. Blots are examples of single experiments that were replicated with similar results. (□) TDP-43, (◇) TDP-43 CTF35, (○) nonspecific band. (D). 4-aminoquinoline 2 does not affect caspase-7 mediated cleavage of Ac-DEVD-AMC. Caspase-7 (4 U/mL) was incubated with 1 μM Ac-DEVD-AMC in the presence and absence of indicated concentrations of 2 and change in fluorescence was monitored as described in section 2.4. Data represent the mean ± S.E.M of four determinations.

Figure 4. Treatment with 4-aminoquinoline probes results in a caspase-dependent reduction in TDP-43 levels and increase in CTFs in human H4 cells

(A) 4-aminoquinolines reduce TDP-43 levels in H4 cells with SAR similar to binding and stimulation of caspase-7 cleavage of TDP-43. H4 cells were plated onto 24-well plates the day before and then treated with test compound for 24 hrs, followed by lysis in SDS sample buffer and immunoblotting (section 2.3). (B) 4-aminoquinoline 1 and 2 reduce full length TDP-43 and increase TDP-43 CTFs in the triton X100 insoluble fraction, and to a lesser extent in the soluble fraction. H4 cells were plated in 6 well plates and treated with test compound as before and subject to fractionation of triton X100 soluble and insoluble fractions as described in section 2.5. (C) The effects of 1 on TDP-43 levels are attenuated by the caspase inhibitors Q-VD-OPh and Ac-DEVD-CHO. H4 cells were plated in 24-well plates and treated with 1 in the presence and absence of 10 μM Q-VD-OPh or Ac-DEVD-CHO for 24 hrs, followed by immunoblotting as before. Blots are examples of single experiments that were replicated with similar results.

Figure 5. 4-aminoquinoline probes activate caspase-3 and caspase-7 with subsequent PARP cleavage

(A) H4 cells were plated onto 24 well plates the day before and treated with test compounds for overnight, followed by lysis in SDS sample buffer with subsequent immunoblotting (section 2.3). Blots represent single experiments that were replicated with similar results. (B) 4-aminoquinolines induce PARP cleavage with similar SAR to TDP-43 binding. Human H4 cells were treated with compounds 1 (□), 2 (○), 3 (●), 4 (■), 5 (△), and 6 (▲) at concentrations ranging from 0.33 to 32 μM overnight. Levels of cleaved PARP were measured using solid state immunoassay as described in section 2.6.

Figure 6. 4-aminoquinoline probes reduce levels of HDAC6 and inhibit autophagy in H4 neuroglioma cells

H4 cells were plated onto 24 well plates the day before and treated with test compounds for overnight, followed by lysis in SDS sample buffer with subsequent immunoblotting (section 2.3). Blots represent single experiments that were replicated with similar results.