Phosphorylation of mutant huntingtin at serine 116 modulates neuronal toxicity

Abstract

Phosphorylation has been shown to have a significant impact on expanded huntingtin-mediated cellular toxicity. Several phosphorylation sites have been identified on the huntingtin (Htt) protein. To find new potential therapeutic targets for Huntington's Disease (HD), we used mass spectrometry to identify novel phosphorylation sites on N-terminal Htt, expressed in HEK293 cells. Using site-directed mutagenesis we introduced alterations of phosphorylation sites in a N586 Htt construct containing 82 polyglutamine repeats. The effects of these alterations on expanded Htt toxicity were evaluated in primary neurons using a nuclear condensation assay and a direct time-lapse imaging of neuronal death. As a result of these studies, we identified several novel phosphorylation sites, validated several known sites, and discovered one phospho-null alteration, S116A, that had a protective effect against expanded polyglutamine-mediated cellular toxicity. The results suggest that S116 is a potential therapeutic target, and indicate that our screening method is useful for identifying candidate phosphorylation sites.

Figure 1. Isolation of Huntingtin N511 for mass spectrometry.

(A) Representation of the Htt N511 construct used for transfection. (B) HEK293 cells were transfected with the N511 construct for 24 hours. Protein extracts were immunoprecipitated using Flag antibody and treated with phosphatase inhibitors or phosphatase and separated on a gel. Coomassie Blue detection of proteins in gel shows N511 overexpression. The image shown is representative of 3 replicate experiments. (C) Peptide coverage in two mass spectrometry experiments obtained from trypsin digest of Htt N511. Underlined residues indicate detection, blue indicates previously known phosphorylatable residues, red indicates novel detected sites. (D–E) Examples of mass spectra of peptide showing isotopic distribution for the site at serine 116.

Figure 2. Generation and expression of N586-82Q constructs with phospho-site alterations.

(A) Schematic representation of the Htt N586 and the sites altered. Italics indicate previously described phosphorylation sites. Detected sites are in bold. (B) Expression of constructs in primary cortical neurons. (C) Quantification of expression of N586 constructs in primary neurons. Fluorescence was quantified using Volocity, and results are expressed as percent of N586-22Q level of expression. (D) Western blot of expression of N586 constructs with phospho-site alterations in primary cortical neurons. The image shown is representative of 4 replicate experiments.

Figure 3. Cell toxicity of N586-82Q constructs with alterations of phospho-sites in primary neurons.

Primary cortical neurons were transfected at DIV5. Automated quantification of nuclear condensation was performed 48(A) Toxicity of constructs with alterations of previously identified Htt phosphorylation sites. (B) Toxicity of constructs with alterations of putative Htt phosphorylation sites located between position 1 and 100. (C) Toxicity of constructs with alterations of putative Htt phosphorylation sites located between 100 and 200. (D) Toxicity of constructs with alterations of putative Htt phosphorylation sites located at the 457–464 cluster. (E) Toxicity of constructs with alterations of putative Htt phosphorylation sites located between 200 and 586. Results are expressed as mean ± sem. n = 5 independent experiments per condition. * p<0.05 compared to N586-82Q.

Figure 4. Time lapse imaging of toxicity of N586 constructs.

Primary cortical neurons were co-transfected at DIV5. Beginning 24 hours after transfection, GFP positive neurons were imaged every 10 mn for 10 hours. (A) Representative images of cells transfected with N586-22Q (top row), N586-82Q (middle row), or N586-82Q S116A (bottom row) at t = 0 (left column), t = 5 h (center column), and t = 10 h (right column). (B) Quantification of cell survival. For each time point cells were given the value of 100 if alive and 0 if dead. Results are expressed as mean ± sem. n = 200 cells analyzed in 5 independent experiments.