Polyglutamine expansion affects huntingtin conformation in multiple Huntington's disease models

Abstract

Conformational changes in disease-associated or mutant proteins represent a key pathological aspect of Huntington's disease (HD) and other protein misfolding diseases. Using immunoassays and biophysical approaches, we and others have recently reported that polyglutamine expansion in purified or recombinantly expressed huntingtin (HTT) proteins affects their conformational properties in a manner dependent on both polyglutamine repeat length and temperature but independent of HTT protein fragment length. These findings are consistent with the HD mutation affecting structural aspects of the amino-terminal region of the protein, and support the concept that modulating mutant HTT conformation might provide novel therapeutic and diagnostic opportunities. We now report that the same conformational TR-FRET based immunoassay detects polyglutamine- and temperature-dependent changes on the endogenously expressed HTT protein in peripheral tissues and post-mortem HD brain tissue, as well as in tissues from HD animal models. We also find that these temperature- and polyglutamine-dependent conformational changes are sensitive to bona-fide phosphorylation on S13 and S16 within the N17 domain of HTT. These findings provide key clinical and preclinical relevance to the conformational immunoassay, and provide supportive evidence for its application in the development of therapeutics aimed at correcting the conformation of polyglutamine-expanded proteins as well as the pharmacodynamics readouts to monitor their efficacy in preclinical models and in HD patients.

Figure 1

A schematic model illustrating the 2B7/MW1 conformational TR-FRET immunoassay for detection of HTT amino-terminal conformation (elaborated from ref. 11). The principal domains of HTT exon 1 are indicated (N17, polyQ and polyproline), together with the identity of the corresponding monoclonal antibodies used for TR-FRET detection (2B7, MW1 and 4C9, respectively). (A) The TR-FRET antibody pair (2B7/MW1) including one antibody targeting the polyQ domain (MW1) engages polyQ epitopes on wild type HTT differently depending on incubation temperature, as the wild type polyQ repeat is flexible and can adopt different conformations (a and b). This results in a higher ratio (≫1) between TR-FRET signals obtained from the same sample at the different temperatures (c). (B) The mutant polyQ repeat in mHTT is less flexible and availability/position of the polyQ epitopes for detection by the same TR-FRET antibody pair is relatively invariant at different temperatures (a and b), resulting in ratio between TR-FRET signals obtained from the same sample at the different temperatures close to 1(c). (C) Instead, a TR-FRET antibody pair (2B7/4C9) which does not include one antibody targeting the polyQ domain does not interrogate changes in polyQ conformation, does not significantly discern between wild type and mutant HTT and produces a ratio between TR-FRET signals obtained from the same sample at the different temperatures close to 1 (c).

Figure 2

The 2B7/MW1 conformational immunoassay preferentially detects mHTT conformation in cocktails of isolated HTT proteins with wild type or mutant polyQ. (A) The 2B7/MW1 TR-FRET assay performed on wild type HTT, mHTT or a cocktail (of equivalent final protein concentration) of the two proteins at 20 °C (a) or measured after shifting the same samples to 4 °C (b). For simplicity, only results for a representative experiment performed on the N573 HTT protein fragments are shown. (B) Summary of data (ratio TR-FRET signals at 4 °C/20 °C) obtained on individual isolated HTT proteins (3 lengths, namely exon 1, N573 and full length) or cocktails of equivalent final protein concentration (Wt + Mut). (C) Same as A, obtained on the same proteins using a (control) TR-FRET immunoassay (2B7/4C9) which does not interrogate the polyQ region and does not detect a temperature- and polyQ dependent conformational change^, , demonstrating efficient detection of all proteins. (D) Same as B (ratio TR-FRET signals at 4 °C/20 °C) obtained on the same proteins using a (control) TR-FRET immunoassay (2B7/4C9) which does not interrogate the polyQ region and does not detect a temperature- and polyQ dependent conformational change. In (B and D) values represent means and standard deviations of the means of three independent experiments (two-way ANOVA with Bonferroni’s post-test, degrees of significance are indicated).

Figure 3

The 2B7/MW1 conformational immunoassay preferentially detects mHTT conformation in lysates of HEK293T cells transfected with plasmids encoding wild type and mutant HTT proteins. (A) The 2B7/MW1 TR-FRET assay performed on cell lysates expressing wild type HTT, mutant HTT or a cocktail of the two proteins (produced by co-transfecting equivalent concentrations of plasmid constructs encoding wild type and mutant HTT proteins) at 20 °C (a) or measured after shifting the same samples to 4 °C (b). For simplicity, only results for a representative experiment performed on lysates transfected with constructs encoding N548 HTT proteins are shown. (B) Summary of data (ratio TR-FRET signals at 4 °C/20 °C) obtained on lysates of HEK293T cells transfected with plasmids encoding wild type and mutant HTT proteins (3 lengths, namely exon 1, N548 and full length) or cocktails of plasmids encoding wild type and mutant HTT of equivalent concentration (Wt + Mut). (C) Same as A, obtained on the same lysates using a (control) TR-FRET immunoassay (2B7/4C9) which does not interrogate the polyQ region and does not detect a temperature- and polyQ dependent conformational change^, , demonstrating efficient detection of all proteins. (D) Same as B (ratio TR-FRET signals at 4 °C/20 °C) obtained on the same lysates using a (control) TR-FRET immunoassay (2B7/4C9) which does not interrogate the polyQ region and does not detect a temperature- and polyQ dependent conformational change. In (B and D) values represent means and standard deviations of the means of three independent experiments (two-way ANOVA with Bonferroni’s post-test, degrees of significance are indicated).

Figure 4

Representative Western blotting experiments of HEK293T cell lysates obtained by transient transfection with plasmids encoding HTT proteins, probed with antibody 4C9 (detecting HTT). (A) Western blot of HEK293T cell lysates transfected with either individual full length HTT proteins, wild type or mutant, or a cocktail of plasmids encoding both (of equivalent DNA concentration); samples were loaded on 6% Gel in order to better resolve the two full lengths proteins. (B) Western blot of HEK293T cell lysates transfected with either individual N548 HTT proteins, wild type or mutant, or a cocktail of plasmids encoding both (of equivalent DNA concentration). (C) Western blot of HEK293T cell lysates transfected with either individual exon 1 HTT proteins, wild type or mutant, or a cocktail of plasmids encoding both (of equivalent DNA concentration). In B and C samples were loaded on 4–12% gradient gel. The WB membranes in B and C were probed with anti-GAPDH antibody as loading control. GAPDH images are reported in supplementary information: Figs S2 and 3. It was not possible to test GAPDH as control in A because, the gel was run in order to resolve high molecular weight proteins, for distinguishing the wild type form from the mutant form of HTT. Indeed, the molecular weights included into the gel did not comprise proteins lower than 100 kDa (as GAPDH). All images were cropped from the original acquired file. Full-length blots are reported in Supplementary Information: Figures S1, S2 and S3.

Figure 5

The 2B7/MW1 conformational immunoassay detects mHTT conformation in lysates of immortalized control and HD human fibroblasts (obtained from heterozygous HD donors). (A) Table illustrating the salient features of human control and HD immortalized fibroblasts used in this study. (B) Summary of 2B7/MW1 conformational immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on lysates of control and HD fibroblasts. (C) Summary of 2B7/4C9 control immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on the same lysates of control and HD fibroblasts. (D) Same as A, plotted against CAG (polyQ) repeat length rather than against control/HD genotype. In B and C values represent means and standard deviations of the means of three independent experiments (two-way ANOVA with Bonferroni’s post-test, degrees of significance are indicated).

Figure 6

The 2B7/MW1 conformational immunoassay detects mHTT conformation in a Drosophila transgenic HD model. (A) The 2B7/MW1 TR-FRET assay performed on tissue homogenates of flies expressing a human wild type HTT N469 fragment or a mHTT N469 fragment at 20 °C (a) or measured after shifting the same samples to 4 °C (b). (B) Summary of data (ratio TR-FRET signals at 4 °C/20 °C) from three independent experiments. (C) Same as A, obtained on the same samples using a (control) TR-FRET immunoassay (2B7/4C9) which does not interrogate the polyQ region and does not detect a temperature- and polyQ dependent conformational change^, , demonstrating efficient detection of all proteins. (D) Same as B (ratio TR-FRET signals at 4 °C/20 °C) obtained on the same samples using a (control) TR-FRET immunoassay (2B7/4C9) which does not interrogate the polyQ region and does not detect a temperature- and polyQ dependent conformational change. In B and D values represent means and standard deviations of the means of three independent experiments (Datasets were tested for normality and Student’s t-test, degrees of significance are indicated).

Figure 7

The 2B7/MW1 conformational immunoassay detects mHTT conformation in lysates of control and HD human PBMCs. (A) Table illustrating the salient features of human control and HD donors of samples for this study. (B) Summary of 2B7/MW1 conformational immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on lysates of control and HD PBMCs. (C) Summary of 2B7/4C9 control immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on the same lysates of control and HD fibroblasts. (D) Same as A, plotted against CAG (polyQ) repeat length rather than against control/HD genotype. In B and C values represent means and standard deviations of the means of three independent experiments (Datasets were tested for normality and Student’s t-test, degrees of significance are indicated).

Figure 8

The 2B7/MW1 conformational immunoassay detects mHTT conformation in homogenates of control and HD post-mortem human brains (cortex). (A) Table illustrating the salient features of human control and HD donors of samples for this study. (B) Summary of 2B7/MW1 conformational immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on homogenates of control and HD brain samples. (C) Summary of 2B7/4C9 control immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on the same homogenates of control and HD brain samples. (D) Same as A, plotted against CAG (polyQ) repeat length rather than against control/HD genotype. In B and C values represent means and standard deviations of the means of three independent experiments (Datasets were tested for normality and Student’s t-test, degrees of significance are indicated). Signals obtained from TR-FRET immunoassays, preparatory for the conformational analysis, are reported in Supplementary Information: Figure S6.

Figure 9

S13/S16 modifications in the N17 domain of HTT alter mHTT conformation as detected by the 2B7/MW1 conformational immunoassay. (A) Western blot of semisynthetic HTT exon 1 proteins (100 ng/lane; Q42/43 and Q23/22) used in this study (detection with mAb 4C9). WB images were cropped from the acquired original file. Full-length blots are reported in Supplementary Information: Figures S4 and S5. (B) Upper panels: 2B7/MW1 TR-FRET assay signal obtained on semisynthetic unmodified (Q23) or pS13/pS16 HTT exon 1 Q22 at the two temperatures. Lower panels: 2B7/MW1 TR-FRET assay signal obtained on semisynthetic unmodified (Q43) or pS13/pS16 HTT exon 1 Q42 at the two temperatures. (C) Summary of 2B7/MW1 conformational immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on all wild type (Q23/22) semisynthetic proteins from three independent experiments. (D) Summary of 2B7/MW1 conformational immunoassay data (ratio TR-FRET signals at 4 °C/20 °C) obtained on all mutant (Q43/42) semisynthetic proteins from three independent experiments, each with three technical replicates (two-way ANOVA with Bonferroni’s post-test, degrees of significance are indicated).

Figure 10

The 2B7/4C9 control immunoassay in semisynthetic HTT exon 1 proteins. (A) Upper panels: 2B7/4C9 TR-FRET assay signal obtained on semisynthetic unmodified(Q23) or pS13/pS16 HTT exon 1 Q22 at the two temperatures. Lower panels: 2B7/4C9 TR-FRET assay signal obtained on semisynthetic unmodified (Q43) or pS13/pS16 HTT exon 1 Q42 at the two temperatures. (B) Summary of 2B7/4C9 (control) TR-FRET immunoassay (ratio TR-FRET signals at 4 °C/20 °C), which does not detect the polyQ-dependent conformational change, obtained on all wild type (Q23/22) and all mutant (Q43/42) semisynthetic proteins from three independent experiments, each with three technical replicates (two-way ANOVA with Bonferroni’s post-test, degrees of significance are indicated).