Fig 2: Effects of different intrabody fragments on mHTT. a) Schematics of the intrabody fragments SM1, SM2, and SM3 (24, 29, and 23 amino acids), and N‐terminal mutant HTT fragment (HTT‐N171‐150Q). The intrabody peptides are overlapping fragments of variable region heavy (VH) chain of the scFv‐mEM48, which is tagged with the HA epitope and linked to a LAMP1 signal. N‐terminal mHTT fragment (HTT1–171aa) contains 150 glutamine repeats (150Q). b) Transfection of SM1, SM2, or SM3 with N171‐150Q in HEK 293T cells showed that SM3 can reduce aggregates more efficiently. Scale bar: 20 µm. c) Western blotting analysis of SM1 or SM3 and mutant HTT transfected HEK293T cells using mEM48 antibody. d) Quantification of the aggregates in co‐transfected HEK 293T cells (n = 8 images per group). Data were analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and are presented as mean ± SEM. **P = 0.0085 (150Q+HA vs 150Q+SM1); **P = 0.0020(150Q+HA vs 150Q+SM2); ****P < 0.0001 (150Q+HA vs 150Q+SM3). e) Quantification of the ratios of mHTT to vinculin on the western blots. The data were obtained from four independent experiments (n = 4). Data were analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. **P = 0.0025; ***P = 0.0003; ****P < 0.0001. f) Quantification of HEK293 cell viability using cell counting kit‐8 (CCK8). The data were obtained from four independent experiments (n = 4). Data were analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. *P = 0.031; ***P = 0.0001.

Fig 3: Western blotting of SM3 and HTT‐N171‐150Q transfected 293 cells. a) Western blotting analysis of 293T cells transfected with HTT‐N171‐150Q or HTT‐171‐150Q cot‐transfected with SM3. b,c) Quantification of the ratios of mHTT recognized by mEM48 or 1C2 to vinculin on the western blots. The data were obtained from four independent experiments (n = 4). Data were analyzed by two‐tailed Student's t test and presented as mean ± SEM. *P = 0.0143; ***P = 0.0002. d) Quantification of human HTT using ELISA. n = 4 per group. Data were analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. *P = 0.0155; ***P = 0.0005(150Q+HA vs 150Q+SM3); ***P = 0.0008(150Q+HA vs 150Q+SM3+DMSO); ****P <0.0001; ns = 0.2693. e) Western blotting analysis of the transfected 293T cells, probed with LAMP1, LAMP2, P62, Beclin 1, and LC3 antibodies to detect whether the lysosomes‐autophagy pathway was activated. Vinculin served as a loading control. f) Quantification of the ratios of LAMP1, LAMP2, P62, and Beclin 1 to vinculin or LC3II to LC3I on the western blots. The data were obtained from four independent experiments (n = 4). Data are analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. Beclin 1: *P = 0.0113; **P = 0.0029. LC3: **P = 0.0018; ***P = 0.0008; ****P <0.0001. WT293 = Untransfected 293 cells.

Fig 4: Lysosomal fractionation of SM3 and mHTT‐transfected 293T cells. a) Western blotting analysis of transfected mHTT and SM3 with mEM48 antibody to detect HTT‐N171‐150Q. The blots were also probed with LAMP1 and Vinculin antibodies. b) Quantification of the ratios of mHTT to vinculin (total protein) or LAMP1 (lysosome) on the western blots. The data were obtained from four independent experiments (n = 4). Data were analyzed by two‐tailed Student's t test and presented as mean ± SEM. ***P = 0.0002; ****P <0.0001. c) Western blotting analysis of mHTT, which in transfected 293T cells with SM3 and added proteasome inhibitor MG132 or lysosome inhibitor bafA1. HA transfection was used as control for SM3. d) Quantification of the ratios of mHTT to vinculin on the western blots. The data were obtained from four independent experiments (n = 4). Data were analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. ***P = 0.0001; ****P <0.0001. WT293 = un‐transfected 293 cells.

Fig 5: Analysis of HD KI‐140Q mice after stereotaxic injection with SM3. a) Schematic diagram of AAV‐PHP.eB‐HA‐SM3 viral vector for the brain injection, which expresses HA‐tagged SM3 under the control of the UBC promoter. b) Western blotting analysis of the striatum of HD KI‐140Q and wild type (WT) mice one month after stereotaxic injection with GFP (control) or SM3. Western blots were probed with 1C2 or mEM48 antibody to detect mHTT and its aggregates. Representative western blots show samples from two animals in each group. mHTT aggregates are in the stacking gel, the full‐length mutant HTT is indicated by an arrowhead. c) Quantification of the ratios of mHTT to vinculin on the western blots. The data were obtained from four independent experiments (n = 4). Data were analyzed by two‐tailed Student's t test and are presented as mean ± SEM. **P = 0.001 (1C2); **P = 0.002. d) Immunofluorescent images of brain sections stained with antibodies mEM48, GFP, and HA (SM3). DAPI is used for nuclear staining. Scale bars: 20 µm. e) Representative immunofluorescent fluorescent images of the striatum from GFP or SM3 injected HD KI‐140Q mice. Antibodies for NeuN, GFAP, and Iba1 were used. Scale bars: 40 µm. f) ELISA assay of human HTT. n = 4 mice per group. Data were analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. **P = 0.0011; ****P < 0.0001. g) Quantification of mHTT aggregates in SM3‐injected mice. Cells expressing mHTT aggregates were counted per 0.1mm2, n = 8 mice per group. Data are analyzed by two‐tailed Student's t test and presented as mean ± SEM. **P = 0.0056. h) Quantification of the NeuN, GFAP, Iba1 immunostaining. Data are analyzed by one‐way ANOVA with Dunnett's multiple comparisons test and presented as mean ± SEM. n = 8 mice per group; GFAP: *P = 0.0262 (WT GFP verses HD KI‐140Q+SM3); **P = 0.0056 (WT GFP verses HD KI‐140Q). Iba1: *P = 0.0448 (WT GFP verses HD KI‐140Q +SM3); **P = 0.0071 (WT GFP verses HD KI‐140Q). Note that the numbers of NeuN‐positive cells were not changed in all scenarios but injection of SM3 decreased the GFAP or Iba1 positive cells.