Fig 2: HIPK3- and MAPK11- mediated HTT regulation is mHTT-dependent. (A) Representative western blots and quantification of the STHdhQ7/Q7 (WT) cells showing that expression of the mHTT (Q72) but not the wtHTT (Q25) N-terminal exon1 fragments enables the reduction of endogenous wtHtt levels by knocking down Mapk11. The wtHtt change by the siRNA targeting Mapk11 was calculated within each group of cDNA transfected samples and plotted as the percentage difference between each of the Mapk11 siRNA-transfected well and the average of the scramble siRNA-transfected control within the same group. The statistical analysis was performed by one-way ANOVA and Dunnett's post hoc tests. (B) Similar to (A), but showing the effects of knocking down Hipk3. (C) Representative western blots and quantifications (by ab1) of the HD (STHdhQ7/Q111) cells showing that specifically knocking down mHtt using the allele-specific siRNA (allele specific mHtt si) abolished the reduction of endogenous Htt levels by knocking down Mapk11 or Hipk3. The statistical analysis was performed by two-tailed unpaired t tests. (D) qPCR measurements of Hipk3 mRNA levels in the HD (Q7/Q111) versus the WT (Q7/Q7) STHdh cells (left), and in the HD (HdhQ140/Q7) versus the WT (HdhQ7/Q7) mouse striata of animals at indicated ages (middle and right). The levels were normalized to the average of WT samples. The statistical analysis was performed by two-tailed unpaired t tests. (E) qPCR measurements of Hipk3 mRNA levels in the HD (STHdhQ7/Q111) cells or the WT (STHdhQ7/Q7) cells transfected with the indicated siRNAs. B01, C01 and Hdh5 are siRNAs targeting Htt (see Supplementary information, Data S1). The levels were normalized to the average of scrambled siRNA-transfected control samples (Scramble), and the statistical analysis was performed by one-way ANOVA followed by post hoc Dunnett's tests. (F) Similar to (E), but in patient iPSC-derived neurons. HTT3 is the siRNA targeting human HTT. (G) Representative western blots and quantifications of phospho-MAPK11/MAPK11 ratio in STHdhQ7/Q7 WT cells transfected with indicated cDNA and treated with indicated compounds. The p-MAPK11/MAPK11 ratio change compared to the Mock control was quantified to assay the activation of MAPK11. (H) Western blots and quantification of Atf2 and phosphorylated Atf2 (p-Atf2) levels in striata lysates from 6.5-month-old mice with or without Mapk11 knockout in the wild-type background using specific antibodies targeting Atf2 versus phosphorylated Atf2, respectively. Atf2 phosphorylation was significantly decreased in Mapk11 knockout mice, confirming Atf2 as a substrate for Mapk11. Statistical analysis was performed by two-tailed unpaired t tests. (I) Similar to (H), but comparing HD (HdhQ7/Q140) versus WT mice (HdhQ7/Q7). Atf2 phosphorylation was significantly increased in HD mice, suggesting that Mapk11 activity may be increased. For all plots, error bars represent mean and SEM. *P < 0.05; **P < 0.01; ***P < 0.001; ns, P > 0.05.

Fig 3: A schematic model illustrating mHTT-dependent MAPK11-HuR mediated regulation of HTT mRNA stability.The presence of mHTT activates MAPK11 (ref. [19,20]), leading to translocation of HuR protein to the cytoplasm (ref. [21]) (Figs. 7 & 9) and its interaction with HTT mRNA in its exon 11 at the #105708 site (Figs. 4 & 5) to stabilize it in a mHTT-dependent manner (Figs. 2, 6 & 8B–D).

Fig 4: Mapk11 regulates HTT mRNA levels via regulation of HTT mRNA stability. (A) RT-qPCR results in STHdh cells showing that knockdown of Mapk11 lowers Htt mRNA levels in HD (left) but not wild-type cells (right). The X-axis indicates the Mapk11 mRNA expression level, whereas the Y-axis indicates the Htt mRNA level. As for B-D. Each single point exhibits the Htt versus Mapk11 mRNA levels of an independently transfected well, and the numbers are indicated (n). All data were normalized to the average of scrambled siRNA-transfected control samples. Red, scrambled siRNA-transfected control samples; Dark, cells transfected with either of the two Mapk11 siRNAs (Mapk11_si1&Mapk11_si2). The correlation coefficient R and P values on top of each plot were calculated by Spearman correlation analysis. “***” on the right side of each plot indicates that P< 0.001 calculated by unpaired two-tailed t tests for the Htt mRNA levels in Mapk11 siRNA-transfected sample versus the control. (B) Similar to (A), but in HD patient iPSC-derived neurons (Q47). Data plots and statistical analysis were performed in the same way as in (A). (C) Similar to (A), but the mRNAs were extracted from in vivo mouse striata (HdhQ7/Q140 with or without heterozygous knockout of Mapk11, 4-6.5-month-old), and the statistical analyses were performed the same. The “n” numbers indicate the number of mice. (D) The non-radioactive pulse-chase experiment showing that knockdown of Mapk11 does not influence Htt protein degradation. The siRNAs were transfected in STHdhQ7/Q111 cells. Three independently transfected wells for each point were tested. Data plotted as mean and SEM. The statistical analysis was performed by two-way ANOVA to determine if the transfection influences the protein degradation. “ns” on the right side of the plot indicates no statistically significant interaction between siRNA transfection and Htt protein degradation. (E) The Htt lowering by Mapk11 siRNA (Mapk11_si2) was measured in STHdh cells by 2B7/2166 HTRF. The proteasome inhibitor epoxomycin or the autophagy inhibitor bafilomycin A (BafA) did not reduce the lowering effect given by Mapk11 knockdown. Bars represent mean and SEM, and the numbers indicate the number of independently transfected wells. The statistical analysis was performed by one-way ANOVA and Dunnett's post hoc tests. **P< 0.01. (F) The Htt promoter activity was measured in STHdhQ7/Q111 by the luciferase reporter assay (see Materials and Methods). Knockdown of Mapk11 did not significantly influence the Htt promoter activity. Blue lines represent mean and SEM, and 10 independently transfected wells were tested for each condition. The statistical analysis was performed by unpaired two-tailed t tests. (G) Htt mRNA stability measurements in STHdhQ7/Q111 cells (left) and HD patient iPSC-derived neurons Q47 (right). The cells were transfected with indicated siRNAs for 36 h and then treated with actinomycin D to stop new mRNA synthesis. The Htt/HTT mRNA level was then measured at different time points using RT-qPCR, normalized to Hprt/HPRT mRNA levels and plotted as percentage of the initial time point. For HD patient iPSC-derived neurons, β-actin and 18S mRNA were used as controls, which were not influenced by MAPK11 knockdown. Since their degradation is close to HPRT and their levels were normalized to HPRT at each time point, the normalized levels were relatively flat across time points. Data were plotted as mean and SEM. The statistical analysis was performed by two-way ANOVA to determine if the transfection influences the mRNA degradation. ***P< 0.0001 for the effect of MAPK11 siRNA on Htt mRNA degradation.

Fig 5: Lowering Mapk11 rescues HD-relevant phenotypes in iPSC-derived HD neurons and HD knockin mouse model. (A) Left panels, representative phase-contrast (for cell confluence and morphology) and green fluorescence images (for caspase-3 activity using the caspase-3 detection dye) of the wild-type (WT, Q19) or the Huntington's disease (HD, Q47) iPSC-derived neurons transfected with the indicated siRNAs under BDNF deprived conditions for 48 h. Scale bar, 100 μm; Right panel, quantification of the number of caspase-3 signals (green particle counts × size × mean intensity) per field by IncuCyte software. The numbers on top indicate the number of independently transfected wells. 4 fields of each well were imaged for quantification by IncuCyte inside the cell culture incubator. All the signals were normalized to the average of the HD cells. Bars represent mean and SEM. The statistical analysis was performed by one-way ANOVA and Dunnett's post hoc tests. (B) Left panels, representative images of the Htt aggregates (by the antibody S830) alone or merged with DAPI and Darpp32. The scatter plot, quantification of the total aggregate signals (aggregate counts × size × mean intensity) per cell by particle analysis using ImageJ. Data were from 12 different slices from 3 mice of each genotype. Scale bar, 20 μm. The imaging and the analysis were performed blindly before revealing the genotypes. The statistical analysis was performed by one-way ANOVA and Dunnett's post hoc tests. (C) The total number of active behaviors (activities) in 5 min for each mouse of the indicated genotypes tested in the pen holder with mashed surface. Unpaired two-tailed t test was performed between the HD (HdhQ140/Q140) and WT (HdhQ7/Q7) mice without Mapk11 deletion (Mapk11+/+); one-way ANOVA and Dunnett's post hoc tests were performed within the HD and WT groups. (D) Left, the total distance traveled in 15 min for mice of the indicated genotypes tested in the open-field. Right, the open-field was divided into 4 × 4 virtue grids for analysis, and the number of times that each mouse crossed the borders of the middle 4 blocks was quantified. The statistical analysis was performed and indicated as in (C). (E) Representative footprints of mice of indicated genotypes. (F) The plot of the top principle component (Coupling PC1) of the gait coupling parameters (Supplementary information, Table S1). The statistical analysis was performed and indicated as in (C). (G) Summary of the number of rescued (green) versus exacerbated (orange) parameters by Mapk11 heterozygous or heterozygous knockout. The category was judged by the direction of changes. Heterozygous or homozygous knockout of MAPK11 in HD mice led to predominantly more rescues (green) than exacerbations (orange), suggesting rescue effects. See Supplementary information, Table S2 for the selected parameters. For all plots, error bars represent mean and SEM. The numbers on top of each plot indicate the number of mice. For all the behavioral experiments and aggregation immunofluorescence, mice were tested blindly and their genotypes were revealed after data recording and analysis. Immunofluorescence data were obtained from 11 month old mice and behavioral data were obtained from 7.5 month old mice.