Accumulation of mutant proteins is a major cause of many diseases (collectively called proteopathies), and lowering the level of these proteins can be useful for treatment of these diseases. We hypothesized that compounds that interact with both the autophagosome protein microtubule-associated protein 1A/1B light chain 3 (LC3)1 and the disease-causing protein may target the latter for autophagic clearance. Mutant huntingtin protein (mHTT) contains an expanded polyglutamine (polyQ) tract and causes Huntington’s disease, an incurable neurodegenerative disorder2. Here, using small-molecule-microarray-based screening, we identified four compounds that interact with both LC3 and mHTT, but not with the wild-type HTT protein. Some of these compounds targeted mHTT to autophagosomes, reduced mHTT levels in an allele-selective manner, and rescued disease-relevant phenotypes in cells and in vivo in fly and mouse models of Huntington’s disease. We further show that these compounds interact with the expanded polyQ stretch and could lower the level of mutant ataxin-3 (ATXN3), another disease-causing protein with an expanded polyQ tract3. This study presents candidate compounds for lowering mHTT and potentially other disease-causing proteins with polyQ expansions, demonstrating the concept of lowering levels of disease-causing proteins using autophagosome-tethering compounds.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Discovery of an autophagy inducer J3 to lower mutant huntingtin and alleviate Huntington’s disease-related phenotype
Cell & Bioscience Open Access 08 October 2022
Molecular Cancer Open Access 11 April 2022
Journal of Hematology & Oncology Open Access 05 February 2022
Subscribe to Nature+
Get immediate online access to Nature and 55 other Nature journal
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
The protein structure data has been uploaded to the Protein Data Bank with accession number 6J04. Source data for all figure plots are provided with the paper. The full gel blots and the proteomics data sets have been provided in the Supplementary Information. The data that support the findings of this study are available from the corresponding authors upon reasonable request.
Kabeya, Y. et al. LC3, a mammalian homologue of yeast Apg8p, is localized in autophagosome membranes after processing. EMBO J. 19, 5720–5728 (2000).
Scherzinger, E. et al. Huntingtin-encoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo. Cell 90, 549–558 (1997).
Warrick, J. M. et al. Expanded polyglutamine protein forms nuclear inclusions and causes neural degeneration in Drosophila. Cell 93, 939–949 (1998).
Fire, A. et al. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806–811 (1998).
Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–826 (2013).
Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819–823 (2013).
Winter, G. E. et al. Phthalimide conjugation as a strategy for in vivo target protein degradation. Science 348, 1376–1381 (2015).
Lu, K., den Brave, F. & Jentsch, S. Pathway choice between proteasomal and autophagic degradation. Autophagy 13, 1799–1800 (2017).
Mizushima, N., Levine, B., Cuervo, A. M. & Klionsky, D. J. Autophagy fights disease through cellular self-digestion. Nature 451, 1069–1075 (2008).
Zhu, C. et al. Developing an efficient and general strategy for immobilization of small molecules onto microarrays using isocyanate chemistry. Sensors 16, E378 (2016).
Fei, Y. et al. Screening small-molecule compound microarrays for protein ligands without fluorescence labeling with a high-throughput scanning microscope. J. Biomed. Opt. 15, 016018 (2010).
Mangiarini, L. et al. Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87, 493–506 (1996).
Liu, H. et al. Nuclear cGAS suppresses DNA repair and promotes tumorigenesis. Nature 563, 131–136 (2018).
Landry, J. P. et al. Discovering small molecule ligands of vascular endothelial growth factor that block VEGF–KDR binding using label-free microarray-based assays. Assay Drug Dev. Technol. 11, 326–332 (2013).
Fei, Y. et al. Characterization of receptor binding profiles of influenza A viruses using an ellipsometry-based label-free glycan microarray assay platform. Biomolecules 5, 1480–1498 (2015).
Zhu, X. et al. Oblique-incidence reflectivity difference microscope for label-free high-throughput detection of biochemical reactions in a microarray format. Appl. Opt. 46, 1890–1895 (2007).
Landry, J. P., Zhu, X. D. & Gregg, J. P. Label-free detection of microarrays of biomolecules by oblique-incidence reflectivity difference microscopy. Opt. Lett. 29, 581–583 (2004).
Zhu, C. et al. Fast focal point correction in prism-coupled total internal reflection scanning imager using an electronically tunable lens. Sensors 18, E524 (2018).
Menalled, L. B., Sison, J. D., Dragatsis, I., Zeitlin, S. & Chesselet, M. F. Time course of early motor and neuropathological anomalies in a knock-in mouse model of Huntington’s disease with 140 CAG repeats. J. Comp. Neurol. 465, 11–26 (2003).
Bondeson, D. P. et al. Catalytic in vivo protein knockdown by small-molecule PROTACs. Nat. Chem. Biol. 11, 611–617 (2015).
Miller, J. et al. Identifying polyglutamine protein species in situ that best predict neurodegeneration. Nat. Chem. Biol. 7, 925–934 (2011).
Fu, Y. et al. A toxic mutant huntingtin species is resistant to selective autophagy. Nat. Chem. Biol. 13, 1152–1154 (2017).
Baldo, B. et al. TR-FRET-based duplex immunoassay reveals an inverse correlation of soluble and aggregated mutant huntingtin in Huntington’s disease. Chem. Biol. 19, 264–275 (2012).
Weiss, A. et al. Single-step detection of mutant huntingtin in animal and human tissues: a bioassay for Huntington’s disease. Anal. Biochem. 395, 8–15 (2009).
Lu, B. et al. Identification of NUB1 as a suppressor of mutant Huntington toxicity via enhanced protein clearance. Nat. Neurosci. 16, 562–570 (2013).
Mizushima, N. et al. Dissection of autophagosome formation using Apg5-deficient mouse embryonic stem cells. J. Cell Biol. 152, 657–668 (2001).
Lackey, K. et al. The discovery of potent cRaf1 kinase inhibitors. Bioorg. Med. Chem. Lett. 10, 223–226 (2000).
Reddy, K. & D’Orazio, A. Highlights from the international conference on molecular targets and cancer therapeutics: discovery, biology, and clinical applications, Philadelphia, PA. ECCO 13–The European Cancer Conference, Paris, France, October 30-November 3, 2005. Clin. Genitourin. Cancer 4, 156–159 (2005).
Johnson, G. L. & Lapadat, R. Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science 298, 1911–1912 (2002).
Tao, W. et al. An inhibitor of the kinesin spindle protein activates the intrinsic apoptotic pathway independently of p53 and de novo protein synthesis. Mol. Cell. Biol. 27, 689–698 (2007).
Trettel, F. et al. Dominant phenotypes produced by the HD mutation in STHdh Q111 striatal cells. Hum. Mol. Genet. 9, 2799–2809 (2000).
Kimura, S., Noda, T. & Yoshimori, T. Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy 3, 452–460 (2007).
Zhang, J., Wang, J., Ng, S., Lin, Q. & Shen, H. M. Development of a novel method for quantification of autophagic protein degradation by AHA labeling. Autophagy 10, 901–912 (2014).
Ni, H. M. et al. Dissecting the dynamic turnover of GFP–LC3 in the autolysosome. Autophagy 7, 188–204 (2011).
Feng, X., Luo, S. & Lu, B. Conformation polymorphism of polyglutamine proteins. Trends Biochem. Sci. 43, 424–435 (2018).
Wang, G., Liu, X., Gaertig, M. A., Li, S. & Li, X. J. Ablation of huntingtin in adult neurons is nondeleterious but its depletion in young mice causes acute pancreatitis. Proc. Natl Acad. Sci. USA 113, 3359–3364 (2016).
Vijayvargia, R. et al. Huntingtin’s spherical solenoid structure enables polyglutamine tract-dependent modulation of its structure and function. eLife 5, e11184 (2016).
Hancock, M. K., Hermanson, S. B. & Dolman, N. J. A quantitative TR-FRET plate reader immunoassay for measuring autophagy. Autophagy 8, 1227–1244 (2012).
Sapp, E. et al. Native mutant huntingtin in human brain: evidence for prevalence of full-length monomer. J. Biol. Chem. 287, 13487–13499 (2012).
Ko, J., Ou, S. & Patterson, P. H. New anti-huntingtin monoclonal antibodies: implications for huntingtin conformation and its binding proteins. Brain Res. Bull. 56, 319–329 (2001).
We thank J. Lu, M. Jiang, L. Liu and Q. Huang for their technical support with mouse behavioural experiments, Y. Xu for technical support with protein purification and H. Saiyin for help with obtaining human patient fibroblasts. We thank the following for funding support: National Key Research and Development Program of China (2016YFC0905100), National Natural Science Foundation of China (8192500069, 81870990, 31961130379, 91649105, 31470764, 91527305 and 61505032), Science and Technology Commission of Shanghai Municipality (18410722100), Natural Science Foundation of Shanghai (19ZR1405200), Shanghai Municipal Science and Technology Major Project (No.2018SHZDZX01), ZJLab and Hereditary Disease Foundation.
B.L., Y.F., Y. Ding and Y. Dang have filed two patents together on the basis of this study to the State Intellectual Property Office of China (201910180674.7 and 201910180717.1).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Peer review information Nature thanks David Rubinsztein, Huda Yahya Zoghbi and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
a, SDS–PAGE and linear mode MALDI–TOF mass spectrometry analysis of the expression and purification of recombinant LC3B protein. Left, SDS–PAGE: lane 1, the whole cell lysate before induction; lane 2, the whole cell lysate after induction; lane 3, the supernatants of induced cells; lane 4, the flow through fraction of Ni-NTA chromatography; lane 5, the eluates of Ni-NTA chromatography (GST–His8–LC3); lane 6, LC3B eluate after removal of GST–His8 tag by TEV protease; lane 7, the eluates of size-exclusion chromatography; lane 8, molecular weight marker. Right, m/z peak of recombinant LC3B is 14,660.811, consistent with theoretical calculations. b, Structural alignment of purified recombinant LC3B(ΔG120) (PDB ID: 6J04, yellow) with published LC3B structure (PDB ID: 1UGM, cyan) by PyMOL. c, d, SDS–PAGE and linear mode MALDI–TOF mass spectrometry analysis of the HTTexon1 proteins. c, Left, SDS–PAGE for HTTexon1(Q72)–MBP: lane 1, the supernatants of induced cells; lane 2, the insoluble fraction of induced cells; lane 3, the flow through fraction of Ni-NTA chromatography; lane 4, the eluates of Ni-NTA chromatography; lane 5, the eluates of size-exclusion chromatography; lane 6, molecular weight marker. d, Left, SDS–PAGE for HTTexon1(Q25)–MBP: lane 1, molecular weight marker; lane 2, the induced cell lysate; lane 3, the supernatant fraction of induced cells; lane 4, the flow-through fraction of Ni-NTA chromatography; lanes 5 and 6, the eluates of Ni-NTA chromatography; lanes 7 and 8, the eluates of size-exclusion chromatography. The m/z peaks of HTTexon1(Q72)–MBP (c, right) and HTTexon1(Q25)–MBP (d, right) are 64,225.946 and 58,228.893, consistent with theoretical calculations. e, Left and middle, size-exclusion chromatography of the recombinant full-length HTT(Q73) (flHTT–Q73) and HTT(Q23) (flHTT-Q23) proteins using Superose 6 5/150 GL. The major peak fractions were collected pooled together for the SDS–PAGE analysis (right). f, SDS–PAGE analysis of purified MBP–His8 (MBP), sfGFP (GFP) and Rpn10 proteins.
Extended Data Fig. 2 Negative controls for OI-RD measurements and validation of the compounds’ interaction with HTT and LC3 by MST.
a, Similar to Fig. 1c–e, but for negative control proteins MBP–His8 (MBP), sfGFP and Rpn10 (Rpn10). Association–dissociation curves of surface immobilized compounds 8F20 and 10O5 with these proteins were measured by OI-RD, and no compound–protein interactions were detected. For all association–dissociation curves, vertical dashed lines mark the starts of association and dissociation phases of the binding event. b, Binding of 10O5 and 8F20 to full-length HTT(Q73) (flHTT(Q73), black dots) or LC3B (red dots) in standard treated capillaries measured by MST. The compound-bound protein fractions (bound/total) were calculated from the MST signals (Fnorm) at each compound concentration, as well as the bound (Fnorm_bound, set as 100%) and the unbound (Fnorm_unbound, set as 0%) MST signals: bound/total = (Fnorm – Fnorm_unbound)/(Fnorm_bound – Fnorm_unbound) × 100%. The fitted curves and Kd values calculated by Nanotemper analysis software (v.1.5.41) for flHTT(Q73) and LC3B are indicated in each panel. Consistent with the OI-RD measurements (Fig. 1e), no binding was observed for the flHTT(Q23) protein (blue dots). The MST experiments were repeated more than three times and showed consistent results. c, Similar to b, except using the compounds indicated on the x axis. MST measurements of the binding of indicated compounds to full-length HTT(Q73) (flHTT-Q73), full-length HTT(Q23) (flHTT-Q23) and LC3B in standard treated capillaries. The proteins tested are indicated in the legends. d, Similar to Fig. 1c–e, but plotting the association–dissociation curves of surface immobilized compound AN2 with full-length HTT(Q73) (Q73), or full-length HTT(Q23) (Q23), LC3B or the negative-control proteins MBP–His8 (MBP), sfGFP and Rpn10. For all association–dissociation curves, vertical dashed lines mark the starts of association and dissociation phases of the binding event. The red dashed lines are global fits to a Langmuir reaction model with the global fitting parameters listed at the bottom of each plot. No binding signals were observed for full-length HTT(Q23) proteins, and thus the parameters were not presented. e, Cell viability measurement of cultured HD neurons measured by the CellTiter-glo assay. No toxicity was observed within the concentration range presented in Fig. 2, although the compound 8F20 became toxic to the cells when the concentration reached 300 nM.
Extended Data Fig. 3 mHTT-lowering effects by mHTT-linker compounds could be detected by multiple antibodies and were dependent on autophagy.
a, Representative western blots (HTT detected by the 2166 antibody) and quantifications of compound-treated cultured cortical neurons from HdhQ7/Q140 HD-knock-in mice. The neurons were treated with the indicated compounds (100 nM for 10O5, 8F20 and AN1; 50 nM for AN2) with or without the autophagy inhibitor NH4Cl (top) or chloroquine (bottom left), or the autophagy activator rapamycin (bottom right). The same amount of culture medium was added in the controls (top). The statistical analysis was performed by one-way ANOVA with post hoc Dunnett’s tests, and the F, degree of freedom and post hoc P values are indicated in each bar plot. b, Western blots using indicated HTT or polyQ antibodies for samples from cultured cortical neurons treated with the indicated compounds: 10O5 (100 nM), 8F20 (100 nM) or AN2 (50 nM). The HTT gel blots presented in Fig. 2d (right) were cropped from first four blots. The low molecular weight bands were run out in these blots so that the wtHTT and mHTT could be better separated. Note that the weak bands just above 250 kDa in the first two blots were leftover signals from the spectrin blotting. The spectrin signals were too strong to be stripped completely. c, Western blots using the antibody MW1 or 3B5H10, which detects mHTT specifically. We ensured that the relatively low-molecular-weight proteins did not run out of the gels. No increase of potential polyQ-containing mHTT N-terminal fragments was observed. d, iPS-cell-derived striatal neurons from a patient with HD (Q47) were treated with the indicated compounds (100 nM, with 0.1% DMSO) in presence of an additional 0.1% DMSO or 10 mM NH4Cl, and the mHTT levels were measured by HTRF using the 2B7/MW1 antibody pair. All signals were normalized to the averaged signals from the DMSO control group. The statistical analysis was performed by one-way ANOVA with post hoc Dunnett’s tests, and F, degree of freedom and post hoc P values are indicated in each bar plot. ****P < 0.0001. The post hoc analysis was not performed if the ANOVA tests did not show significance (P > 0.05). e, Immortalized fibroblasts from a patient with HD (Q47) were transfected with the non-targeting control siRNA (Neg siRNA) or the ATG5 siRNA (target sequence, GCCUGUAUGUACUGCUUUA; ATG5 mRNA was knocked down to 17.7 ± 3.0%, n = 3, as tested by reverse transcription with quantitative PCR), and then treated after 24 h with the indicated compounds (100 nM) for a further 48 h. mHTT levels were then measured by HTRF using the 2B7/MW1 antibody pair. All signals were normalized to the averaged signals from the DMSO control group. The statistical analysis was performed by one-way ANOVA with post hoc Dunnett’s tests, and F, degree of freedom and post hoc P values are indicated in each bar plot. ****P < 0.0001. The post hoc analysis was not performed if the ANOVA tests did not show significance (P > 0.05). The western blot of LC3 confirmed the partial inhibition of autophagy in the ATG5-knockdown cells. f, Similar to e, but in wild-type (WT) or Atg5-knockout (Atg5 KO) mouse embryonic fibroblast lines (MEF) transfected with full-length mHTT (flHTT-Q73). The western blot of LC3 confirmed the inhibition of autophagy in the Atg5-KO cells. For all panels, n indicates the number of independently plated wells, and bars represent mean and s.e.m. Full-blots of cropped gels are shown in Supplementary Fig. 1.
Extended Data Fig. 4 Potential influence on c-Raf and KSP pathways following treatment with the mHTT–LC3 linker compounds.
a, Representative results (from three biological repeats) of the in vitro c-Raf kinase assay (see Methods) showing that only 10O5 inhibits c-Raf activity within the concentration range tested. b, Representative western blots and quantifications of phospho-MEK and phospho-ERK as indicators of Raf activity (left) and phospho-BUBR1 as an indicator of KSP inhibition (right) in cultured cortical neurons treated with indicated compounds (100 nM for 10O5, 8F20, AN1, and 50 nM for AN2) or the DMSO control. c, Similar to b, but in immortalized fibroblasts from a patient with HD (Q47). Note that phospho-BUBR1 is essentially absent and too weak to quantify, indicating that KSP was not inhibited by any of the compounds at the concentration tested. Data are mean ± s.e.m. In b, c, all data were corrected by the loading control (β-tubulin) and normalized to the averaged signal of the DMSO control group. The statistical analysis was performed by one-way ANOVA and F, degree of freedom and post hoc P values are indicated in each bar plot. The n number indicates the number of independently plated and treated wells.
a, Overlay between LC3B and predicted Atg8 structure showing high structural similarities. b, Transgenic flies expressing full-length HTT(Q128) driven by elav-GAL4 were fed with indicated compounds at 10 μM for 6 days, and protein lysates were extracted from the heads. mHTT was then measured by HTRF using the 2B7/MW1 antibody pair. Each dot represents the HTRF signal from each individual sample extracted from five fly heads. All the data were normalized to the average of the DMSO-fed control samples. The statistical analysis was performed by one-way ANOVA and Dunnett’s post hoc tests. F(4, 31) = 15.67; ****P < 0.0001. c, 10O5 (top) and AN2 (bottom) concentrations in heart plasma and brain tissues were measured by mass spectrometry at the indicated time points for compound-injected mice (0.5 mg kg−1). In the brain tissue, the 10O5 concentrations were ~20 to ~200 nM, and the AN2 concentrations were ~20 to ~40 nM, close to the effective doses that were capable of lowering mHTT in cultured neurons. Data are mean ± s.e.m.
a, Western blots (4 mice (3 months old) for each group) and quantifications of mHTT and wtHTT in the cortices from HdhQ7/Q140-knock-in mice with intracerebroventricular injection of the indicated compounds (2 μl at 25 μM for each mouse) for 10 days at one dose per day. HTT was detected by western blot using the 2166 antibody, and the statistical analysis was performed by one-way ANOVA and post hoc Dunnett’s tests. F, degree of freedom and post hoc P values are indicated below each bar plot. b, Similar to a, except that the compounds were delivered to 5-month-old HdhQ7/Q140 mice by intraperitoneal injection (0.5 mg kg−1) for 14 days at one dose per day. c, Similar to b, but from striata of intraperitoneally injected mice. The mice were injected at 10 months old for 14 days at one dose per day. d, Left, representative dot blot results (from two technical replicates) of the protein lysates from b using the 4C9 antibody, which preferentially detects mHTT aggregates23. Middle, quantification of the dot blots based on the averaged signals from two technical replicates. Right, measurement of mHTT aggregates by the 4C9–4C9 HTRF assay23. In all panels, n indicates the number of mice tested, and bars represent mean and s.e.m. For quantification, two to three technical replicates were averaged for each mouse. Statistical analysis was performed by one-way ANOVA with post hoc Dunnett’s tests, and F, degree of freedom and post hoc P values are indicated in each bar plot.
a, HeLa cells stably expressing GFP–LC3B were treated with 2 μl vehicle (0.1% DMSO), 10O5 or AN2 for the indicated concentration for 24 h; chloroquine (CQ, 20 μM) treatment was used as a control. After 24 h, cells were fixed and images were acquired by confocal microscopy. The number and size of GFP vesicles per cell was determined using ImageJ software (n indicated on top of each plot). For each treatment, more than 20,000 puncta were quantified (~100 puncta per cell from 226 cells). Scale bar, 10 μm. b, Representative images and quantifications of the numbers of autophagosomes (GFP+ puncta) and autolysosomes (RFP+GFP− puncta) in HeLa cells stably expressing mRFP–GFP–LC3B. Scale bar, 10 μm. Autophagosome numbers or sizes were not influenced by 10O5 and AN2 at the indicated concentrations after 24 h treatment (or 4 h treatment, not shown). The autophagsome fusion was also unaffected as indicated by the autolysosome number. Note that the autophagosome and autolysosome numbers and sizes were based on image analysis of the puncta, some of which may represent multiple vesicles. Green vesicles are considered to be autophagosomes (GFP+ puncta) and red vesicles are considered to be both autophagosomes and autolysosomes. The number of autolysosomes (RFP+GFP− puncta) was calculated by subtracting the number of green vesicles from that of the red vesicles. More than 10,000 puncta from 194 cells were analysed. c, Representative western blots and quantifications of HeLa cells stably expressing GFP–LC3B. The ‘free GFP’ was generated by lysosomal cleavage, and thus the free GFP/GFP–LC3B ratio was used as an index for autophagy flux, which was unaffected by 10O5 or AN2, but decreased by the autophagy flux inhibitor chloroquine. d, Representative western blots and quantifications of the chase signal of long-lived proteins in HeLa cells as an indicator of autophagy flux (see Methods). Consistent with previous reports33, starvation reduced the long-lived protein chase signal, whereas rapamycin treatment had a milder effect. The mHTT–LC3 linker compounds 10O5 and AN2 had no influence in this assay. e, Representative western blots and quantifications of LC3 in cultured cortical neurons treated with the indicated compounds. Normalized LC3-II/LC3-I was used as the indicator of autophagy. Right blot: 10O5, 100 nM; AN2, 50 nM. f, SQSTM1 (p62) levels were determined by western blot for the cortical tissues from mice injected with the indicated compounds or DMSO control. Bars indicate mean and s.e.m.; n indicated in each bar shows the number of cells (a, b), the number of independently plated wells (c–e) or the number of mice (f). Data are mean ± s.e.m. The statistical analysis was performed by one-way ANOVA with post hoc Dunnett’s tests (a–e) or two-tailed unpaired t-tests (f). Note that the post hoc tests were not performed if the ANOVA tests failed to show significance. ****P < 0.0001 (post hoc test).
Extended Data Fig. 8 Investigation on the specificity of mHTT-lowering effects of mHTT–LC3 linker compounds.
a, Representative western blots and quantifications of cultured cortical neurons treated with the indicated compounds. None of the proteins tested showed a clear effect (>10%). b, Volcano plots of the proteomics analysis of cortices from intraperitoneally injected HD mice (10 month old, 4 mice per group, injected for 14 days). Mice were injected with 0.5 mg kg−1 protein with 110 μg kg−1 DMSO, and equal amount of vehicle containing DMSO was injected in the control mice. Only proteins detected in both groups of samples used for comparisons were calculated and plotted. Red arrows indicate HTT. See Supplementary Table 2 for complete datasets. The bar plots indicate the total HTT levels normalized to the DMSO control. The actual mHTT reduction is anticipated to be higher, because the compounds reduced mHTT in an allele-selective manner. c, Similar to b, but in cultured cortical neurons (from postnatal day 0 pups, three wells per group). See Supplementary Table 3 for complete datasets. In all panels, data are mean ± s.e.m.
Extended Data Fig. 9 mHTT–LC3 linker compounds lowered the mutant ATXN3 protein with polyQ expansion in an allele-selective manner.
a, Representative western blots and quantifications of ATXN3 levels in a fibroblast line from a patient with SCA3 treated with the indicated compounds. The lowering of mutant (Q74) but not wild-type (Q27) ATXN3 was observed by treatment of linker compounds tested. b, Quantification of the GFP intensity as an indicator of polyQ–sfGFP (25Q–GFP, 38Q–GFP, 46Q–GFP and 72Q–GFP) protein levels in transfected HEK293T cells treated with the indicated compounds using Incucyte. Reduction of 72Q–GFP, 46Q–GFP and 38Q–GFP but not 25Q-GFP was observed. In a and b, the compound concentrations were 100 nM for 10O5 and AN1, and 50 nM for AN2. Bar plots present mean ± s.e.m., and n indicates the number of independently plated wells. c, SDS–PAGE analysis of polyQ–sfGFP proteins (25Q, 38Q, 46Q, 53Q and 72Q) purified from HEK293T cells. The protein purification methods were similar to those for HTT proteins. d, Binding of 10O5, AN1 and AN2 to sfGFP (GFP) or different polyQ–sfGFP (25Q–GFP, 38Q–GFP and 72Q-GFP) proteins in standard treated capillaries measured by MST, performed and analysed similarly as in Extended Data Fig. 2b. All these compounds interact with 38Q–GFP and 72Q–GFP but not with 25Q–GFP or GFP. e, Association–dissociation curves of surface-immobilized compounds 10O5, AN1 and AN2 with polyQ–sfGFP (72Q, 53Q, 46Q, 38Q and 25Q) proteins. For all association–dissociation curves, vertical dashed lines mark the starts of association and dissociation phases of the binding event. The red dashed curves are fits to a Langmuir reaction model with the fitting parameters listed at the bottom of each plot. No binding signals were observed for 25Q–sfGFP (25Q). f–h, Results of mouse behavioural test performed similarly to those in Fig. 5d–f, except that the mice were injected with saline (0.9% NaCl) with DMSO (110 μg kg−1) or without DMSO. The statistical analysis was performed by two-way ANOVA with post hoc Bonferroni’s tests, and F, P values and degrees of freedom are indicated in the table below each plot. In all panels, data are mean ± s.e.m.
Full blot images of all cropped bands presented in data figures, unless shown elsewhere already. Note that all the loading controls were on the same gel as target protein, except the middle panel of Fig. 2d, which utilized a sample processing control.
Supplementary Table 1. Data collection and refinement statistics (molecular replacement). Data collection and refinement statistics of the full length human LC3BΔG120 mutant crystal. The values in parentheses are for the high resolution shells.
Supplementary Table 2. Proteomics analysis of tissue samples. An excel file of the proteomics data showing the quantification of the abundance (FOT: fraction of total) of each protein in mouse cortices. Note that only proteins detected in all samples utilized for comparisons were listed. The data of proteins that were significantly changed (ratio change > 0.2 and P value < 0.01) were copied into a separate sheet as indicated in the title of each sheet. The GO analysis (http://geneontology.org/) for pathways (http://pantherdb.org/) was performed and the results were listed in those sheets.
Supplementary Table 3. Proteomics analysis of cultured neuron samples. Similar to Supplementary Table 2, but for the proteomics analysis of cultured cortical neurons.
About this article
Cite this article
Li, Z., Wang, C., Wang, Z. et al. Allele-selective lowering of mutant HTT protein by HTT–LC3 linker compounds. Nature 575, 203–209 (2019). https://doi.org/10.1038/s41586-019-1722-1
This article is cited by
Journal of Neurology (2023)
Cellular and Molecular Neurobiology (2023)
Molecular Cancer (2022)
Journal of Hematology & Oncology (2022)
Discovery of an autophagy inducer J3 to lower mutant huntingtin and alleviate Huntington’s disease-related phenotype
Cell & Bioscience (2022)