Figure 1.
Ataxin-3 interacts directly with hHR23A and VCP/p97.
(A) Confocal microscopy images of COS-7 cells fixed and immunostained for endogenous ataxin-3 (i-red) and hHR23A (ii-green). DNA was stained with Hoescht 33342 (blue) in the merged image (iii). (B) Endogenous ataxin-3 was immunoprecipitated from COS-7 cell extracts and co-immunoprecipitation of endogenous hHR23A was determined by western blotting. (C) Recombinant human ataxin-3 and recombinant human hHR23A were incubated under the same conditions applied in the in vitro protease assays, for 5 hours at 37°C. Recombinant ataxin-3 was precipitated using anti-ataxin-3 (1H9) antibody and hHR23A co-immunoprecipitation was analyzed by western blotting. (D) Visualization of endogenous ataxin-3 (i-red) and VCP/p97 (ii-green) through confocal microscopy in fixed COS-7 cells using specific antibodies. Nuclei were stained Hoescht 33342 dye (blue) for merged images. (E) Recombinant GST-VCP/p97 fusion protein was used to pull-down endogenous ataxin-3 from COS-7 total extracts. Endogenous ataxin-3 and recombinant VCP/p97 were detected by western blotting. (F) Recombinant human VCP/p97 and recombinant human (Q22) ataxin-3 were added to a buffered solution at 37°C for 5 hours. These samples were used to immunoprecipitate ataxin-3 and co-immunoprecipitation of VCP/p97 was assessed by western blotting.
Figure 2.
VCP/p97 enhances deubiquitinase activity of ataxin-3 in vitro.
K63-linked (A) and K48-linked (C) hexa-ubiquitin chains (250 nM) were incubated with recombinant human (Q22) ataxin-3 (100 nM) with or without recombinant human VCP/p97 (100 nM) for 20 hours at 37°C. Samples collected at 0, 2, 5 and 20 hours were submitted to SDS-PAGE analysis and reaction products were detected by immunoblotting with an anti-ubiquitin antibody. The graphs summarize the mean ± SEM of the increase in reaction products over time resulting from the hydrolysis of K63-linked (B) and K48-linked (D) hexa-ubiquitin chains by ataxin-3 in the absence (light grey squares) or presence (dark grey triangles) of VCP/p97 in 3–5 independent experiments. Statistical analysis: * p<0.05, compared to recombinant human wild-type ataxin-3 alone at 20 hours. (E) K63-linked hexa-ubiquitin chains (250 nM) were incubated with recombinant wild-type or catalytically inactive (C14A) ataxin-3 (100 nM) in the presence of VCP/p97 (100 nM) for 20 hours at 37°C. The 0, 2, 5 and 20 hours time-points were analysed by western blotting with an anti-ubiquitin antibody.
Figure 3.
VCP/p97 does not enhance expanded ataxin-3 activity in vitro.
(A) Recombinant human expanded (Q80) ataxin-3 (100 nM) was incubated with K63-linked hexa-ubiquitin chains (250 nM) in the absence or presence of recombinant human VCP/p97 (100 nM), for 20 hours at 37°C. Samples of 0, 2, 5 and 20 hours time-points were separated by SDS-PAGE and immunoblotted for ubiquitin. (B) The graph shows the mean ± SEM of the increase in reaction products over time resulting from the deubiquitinase activity of expanded ataxin-3 over K63-linked hexa-ubiquitin chains in the absence (light grey squares) or presence (dark grey triangles) of VCP/p97 in 4 independent experiments.
Figure 4.
VCP/p97 enhancement of wild-type ataxin-3 activity is mediated through a direct protein-protein interaction.
(A) Co-immunoprecipitation of recombinant human VCP/p97 (100 nM) with ataxin-3 (100 nM) with 22 or 80 glutamines in its wild-type or (282RKRR-HNHH) mutant form, after a pre-incubation of 5 hours at 37°C in a buffered solution. Ataxin-3 forms were precipitated with an anti-ataxin-3 polyclonal antibody (αMJD) and co-immunoprecipitation was assessed by western blotting using anti-VCP/p97 polyclonal antibody. (B) Ubiquitin protease assay for Q22 and Q80 (282RKRR-HNHH) ataxin-3 mutants (100 nM) in the absence or presence of recombinant human VCP/p97 (100 nM), using K63-linked hexa-ubiquitin chains (250 nM) as substrate. Samples were collected at times 0, 2, 5 and 20 hours and analyzed by western blotting using an anti-ubiquitin antibody. Graphs express the mean ± SEM of the increase in reaction products over time resulting from the hydrolysis of K63-linked hexa-ubiquitin chains by Q22 (C) and Q80 (D) (282RKRR-HNHH) ataxin-3 mutants in the absence (light grey squares) or presence (dark grey triangles) of VCP/p97 from 3 independent experiments.
Figure 5.
hHR23A has no effect over the ubiquitin hydrolase activity of ataxin-3.
Recombinant human (Q22) wild-type (A, C) or catalytic inactive (C14A) (E) ataxin-3 (100 nM) was incubated alone or together with recombinant hHR23A (100 nM) in a buffered solution containing hexa-ubiquitin chains (250 nM) with K63 (A, E) or K48 (C) linkages for 20 hours at 37°C. Samples were collected at 0, 2, 5 and 20 hours and analyzed by immunoblotting. The reaction products were detected using an antibody against ubiquitin (A, C, E). The appearance of reaction products resulting from the cleavage of K63-linked (B) or K48-linked (D) hexa-ubiquitin chains by ataxin-3 in the absence (light grey squares) or presence (dark grey triangles) of hHR23A was quantified; the mean ± SEM of 4 independent experiments was plotted in graphs.
Figure 6.
hHR23A blocks the VCP/p97 enhancement of ataxin-3 activity in vitro.
(A) Endogenous hHR23A was immunoprecipitated from COS-7 cell extracts and the co-immunoprecipitation of endogenous ataxin-3 and endogenous VCP/p97 was determined by western blotting. (B) In vitro deubiquitination assay of K63-linked hexa-ubiquitin chains (250 nM) by wild-type (Q22) ataxin-3 (100 nM) alone or in the presence of both hHR23A (100 nM) and VCP/p97 (100 nM). Western blotting was performed for samples collected after 0, 2, 5 and 20 hours of reaction and anti-ubiquitin reactivity was assessed. (C) Quantification of the increase in reaction products over time (time = 0 hours as control) resulting from the hydrolysis of K63-linked hexa-ubiquitin chains by wild-type (Q22) ataxin-3 alone (light grey squares) or in the presence (dark grey triangles) of hHR23A and VCP/p97 was plotted as the mean ± SEM of 4 independent experiments.
Figure 7.
Proposed model of sequential events during the interaction between ataxin-3, VCP/p97 and hHR23A.
(i) Ataxin-3 interacts with a monomer of the VCP/p97 hexamer through its 282RKRR region, assembling in a complex capable of recognizing the ubiquitinated substrate. (ii) Ataxin-3 acts on the branched ubiquitin chains attached to the ubiquitinated protein, originating a clear and perceptible ubiquitin signal. (iii) Due to its K63-linkage preference, ataxin-3 leaves an all K48-linkage ubiquitin chain on the surface of the protein. (iv) The ubiquitin signal released by ataxin-3 is recognized by hHR23A and interpreted as a degradation signal. (v) hHR23A delivers the ubiquitinated protein to the proteasome for degradation.