Identification of inclusions and HTT molecular species in Q175 mice.

(A) IHC detects age-dependent increase in the number of mHTT inclusions. Brain sections from 6-mo- and 10-mo-old WT and Q175 mice were processed for IHC with antibody mEM48 (MAB5374) directed against mHTT (A). Images from sections without (A1-3) or with (A4) a cresyl violet counterstain are shown where the dark-brown puncta represent mHTT-positive inclusions. Arrowheads depict neuronal intranuclear inclusions (NIIs), determined with the assistance from the nuclear labeling by cresyl violet, while arrows indicate extranuclear inclusions, primarily the neuritic inclusion in the neuropil. Bars = 20 μm.

(B) mHTT inclusions are detected in nucleus, dendrites and axons of Q175 brains by IEM. Sagittal vibratome brain sections of 17-mo-old Q175 were cut and went through EM processing. Small blocks were obtained from the striatal areas for ultrathin sectioning. Tissue containing grids were processed for immunogold labeling procedure with antibody mEM48, using 10 nm gold followed by silver enhancement. Structures showing high level of silver-enhanced gold labeling were considered as mHTT-positive.

(C) Various forms of HTT molecules are detected with different antibodies by immunoblotting. Equal amounts of proteins from hemibrain homogenates of 17-mo-old WT, TRGL, Q175 and TRGL/Q175 (labeled as “Cross”) were subjected to SDS-PAGE and processed for WB with different antibodies directed against HTT/mHTT, including MAB1574 (C1, C2), mAb PHP2 (C1) and MAB5490 (C2). Images were collected by a digital gel imager (Syngene G:Box XX9). The arrowhead and arrow (C2) depict a 120 kDa and a 48 kDa fragment, respectively. (C3) Densitometry was performed with Image J for the blots shown in (C2) and the results were normalized by the immunoblot(s) of given loading control protein(s) (e.g., GAPDH). Values are the Mean +/- SEM for each group (n = 7 TRGL, 4 Q175 and 10 TRGL/Q175). Significant differences among the groups were analyzed by One-Way ANOVA followed by Sidak’s multiple comparisons test. * P < 0.05, ** P < 0.01.

Colocalization of mHTT with p62, Ub and CTSD.

(A) Triple labeling detects colocalization of mHTT with both autophagy adaptor proteins p62 and Ub. Brain sections from 6-mo- and 10-mo-old mice were immunostained with antibodies to HTT (antibody MW8; red), p62 (green) or pan-Ub (blue), followed by an additional DAPI (cyan) labeling, and confocal images from the striatum are shown. Boxed areas are enlarged and shown in the last column. Arrows and arrowheads depict the white areas representing triply labeled inclusions containing the signals of the three proteins, where arrows are for NIIs, determined with the assistance from the nuclear DAPI labeling, while the arrowheads are for extranuclear inclusions. Bar = 10 μm.

(B) mHTT is detected in vesicles of the ALP. (B1) Brain sections from 10-mo-old Q175 were double-immunolabeled with antibodies to HTT (antibody MW8; green) and CTSD (red), followed by an additional DAPI (blue) labeling, and a 3-color-merged confocal image from the striatum is shown. Large and small green arrows depict HTT nuclear and extranuclear inclusions, respectively, while yellow arrowheads depict yellow puncta showing HTT and CTSD signal colocalization. Bar = 10 μm. (B2) IEM with anti-HTT antibody (MAB1574) specifically detects HTT signal, represented by the silver-enhanced gold particles (red arrowheads), in AV/LY in cell bodies, dendrites and axons. (B3) To demonstrate the labeling specificity of the HTT antibody in this IEM study, the number of silver-enhanced gold particles in AV/LY existing in neuronal cell bodies and neurites was counted from 69 EM images from two 10-mo-old Q175 mice against the number of silver-enhanced gold particles in mitochondria on the same images, and the result is shown in the bar graph. Statistical significances between the two groups were analyzed by unpaired, two-tailed t-test. **** P < 0.0001.

Mild late-onset alterations in the ALP in the striatum of 17-mo-old Q175.

(A) Quantitation of AV/LY subtypes of striatal neurons detect increases in AL, pa-AL and a decrease in LY in 17-mo-old TRGL/Q175 vs. TRGL. (A1) Brain sections from TRGL and TRGL/Q175 (4 sections/mouse, 10 mice/genotype) were immunostained with an anti-CTSD antibody. Confocal images from the cranial-dorsal portion of the striatum (3 images at 120x/section) were collected and representative images for each eGFP-LC3 (green), mRFP- LC3 (red) and CTSD (blue) are shown. Arrowheads depict pa-AL. (A2) Hue angle-based analysis was performed for AV/LY subtype determination using the methods described in Lee et al., 2019. Data are presented as Vesicle #/Neuron (TRGL: n = 713 neurons; TRGL/Q175: n = 601 neurons). Statistical significances between the two groups for each vesicle type were analyzed by unpaired t-test. Two-tailed P value: *** P<0.001, **** P<0.0001.

(B) EM detects larger AL/lipofuscin granules in the Q175 striatum. Sagittal vibratome brain sections from 17-mo-old mice were cut and went through EM processing. Small blocks were obtained from the striatal area for ultrathin sectioning, followed by EM examinations of the grids. The red circle depicts NII, and the arrow depicts larger sized (> 1µm) lipofuscin granules, which were counted on randomly collected images of neurons from striatum of WT (420 neurons from n = 6 mice) or Q175 (586 neurons from n = 9 mice).

Decrease of DARPP-32-IR in striatal neurons of 17-mo-old TRGL/Q175 in the absence of a NeuN signal reduction.

Brain sections from TRGL and TRGL/Q175 (n = 10 mice/genotype, 4 sections/mouse) were double-immunostained with anti-DARPP-32 and -NeuN antibodies. Confocal images from the cranial-dorsal portion of the striatum (3 images at 120x/section) were collected and representative images are shown (A). Arrows depict examples of neurons showing strong NeuN signal while minimal DARPP-32-IR. (B) Images were quantified by Image J for Integrated Density of DARPP-32- (top) or NeuN-IR/Image field (middle), and for # of NeuN positive cells/Image field (bottom). Statistical significances between the two groups were analyzed by unpaired, two-tailed t-test. * P < 0.05.

mTORi INK exhibits target engagement and induces downstream responses in the ALP in 7-mo-old TRGL/Q175.

Equal amounts of proteins from brain homogenates of 7-mo-old TRGL/Q175 mice untreated (labeled as “Veh”) or mTORi INK treated [4 mg/kg (4-mpk), daily, 3 weeks; labeled as “INK”] were subjected to SDS-PAGE and processed for WB with antibodies directed against several marker proteins in the autophagy pathway, representing target engagement of INK or downstream responses. Representative blots are shown on the left (A) while quantitative results of the blots are shown on the right (B). The bottom LC3 blots represent a repeated immunoblotting experiment. Values are the Mean +/- SEM for each group (n = 6-7 mice per condition). Significant differences between the two groups were analyzed by unpaired, two- tailed t-test. * P<0.05, ** P<0.01, *** P<0.001, **** P<0.0001. tfLC3 = mRFP-eGFP-LC3; Endo-LC3 = endogenous LC3.

mTORi INK reverses the mild alteration of AV/LY subtypes in the striatum of 7- mo-old TRGL/Q175.

(A) Brain sections from untreated or INK (4-mpk, 3w)-treated TRGL/Q175 vs. TRGL (4 sections/mouse, 5-6 mice per condition) were immunostained with an anti-CTSD antibody. Confocal images from the cranial-dorsal portion of the striatum (3 images at 120x/section) were collected and representative images for each eGFP-LC3 (green), mRFP-LC3 (red) and CTSD (blue) are shown. (B) Hue angle-based analysis was performed for AV/LY subtype determination using the methods described in Lee et al., 2019. Data are presented as Vesicle #/Neuron (TRGL-Veh: n = 260 neurons; TRGL/Q175-Veh: n = 218 neurons; TRGL/Q175-INK: n = 287 neurons). Statistical significances among the groups were analyzed by One-Way ANOVA followed by Sidak’s multiple comparisons test. ** P<0.01, **** P<0.0001.

mTORi INK reduces HTT-, Ub- or p62-IR-covered areas parallelly in the striatum of 7-mo-old TRGL/Q175.

(A) Brain sections from INK (4-mpk, 3w)-treated or untreated TRGL/Q175 (4 sections/mouse, 6-7 mice/condition,) were immunostained with anti-HTT (MW8), -p62 or -pan-Ub antibodies. Confocal images from the cranial-dorsal portion of the striatum (3 images at 120x/section) were collected. Shown are single-channel images (i.e., without showing the eGFP and mRFP signals). (B1-B3) Areas covered by either the HTT-, p62- or Ub-IR on a per cell basis are quantified (for HTT-IR, TRGL/Q175-Veh: n = 196 neurons, TRGL/Q175-INK: n = 385 neurons; for p62-IR, TRGL/Q175-Veh: n = 311 neurons, TRGL/Q175-INK: n = 378 neurons; for Ub-IR, TRGL/Q175-Veh: n = 244 neurons, TRGL/Q175-INK: n = 347 neurons) and grouped as “Total Area” (B1), “AV-associated Form” (i.e., the IR which was associated with tfLC3 signals representing AVs)(B2) or “AV-unassociated Form” (i.e., the IR which was not associated with the tfLC3 signals)(B3). Statistical significances between the two groups were analyzed by unpaired t-test. Two-tailed P value: * P < 0.05, **** P < 0.0001.

INK treatment does not reverse the reduction of DARPP-32-IR in the striatum of 7-mo-old TRGL/Q175.

(A) Sagittal brain sections containing the striatum area were immunostained with an anti-DARPP-32 antibody, and 10x images taken from the cranial-dorsal portion of the striatum are shown (1st row). Boxed areas on the 1st row are enlarged and shown on the 2nd row for easy viewing of the immunostaining patterns for each condition. (B) For quantitation purpose, each whole 10x image (excluding the areas covered by the fiber bundles which usually exhibited minimal background staining – achieved by threshold setting) was quantified by Image J (1 image/section, 4 sections/mouse) and the results are expressed as the Integrated Density of DARPP-32-IR. n = 6-8 mice/condition. Statistical significances among the groups were analyzed by One-Way ANOVA followed by Sidak’s multiple comparisons test. ** P < 0.01 compared to TRGL-Veh.

Colocalization of HTT/Ub and Ub/p62 in IBs in the STR.

Brain sections from 10-mo-old mice were processed for double IF with antibodies against HTT (antibody MW8; red) and pan-Ub (green), or pan-Ub (red) and p62 (green), and confocal images from the striatum are shown. Arrowheads depict IBs showing colocalization signals.

Molecules involved in various stages of the ALP are largely unchanged in 17-mo-old TRGL/Q175.

Equal amounts of proteins from brain homogenates of 17-mo-old TRGL and TRGL/Q175 were subjected to SDS-PAGE and processed for WB with antibodies directed against a number of interested marker proteins in the ALP, grouped according to the functions of the proteins in the autophagy process: autophagy induction signaling (A), membrane nucleation/AP formation (B), autophagy adaptor proteins (C) and AL formation/substrate degradation (e.g., lysosomal hydrolases or structural components)(D). The blot of the loading control protein GAPDH is boxed and placed under each protein of interest. Images were collected by a digital gel imager (Syngene G:Box XX9). Densitometry was performed with Image J and the result for each protein of interest was normalized by its corresponding GAPDH blot and presented in the bar graph. Values are the Mean +/- SEM for each group (n = 7 TRGL and 10 TRGL/Q175). Significant differences between the 2 groups were analyzed by unpaired, two-tailed t-test. * P < 0.05, ** P < 0.01, *** P < 0.001.

Dosing tests demonstrate mTORi INK BBB penetration and target engagement even at low dosages.

(A) Dosing tests for INK. INK or the vehicle was administered by oral gavage. Two dosing tests were done in 6-mo-old WT mice, one with 10, 8 mg/kg (mpk) every other day or 4-mpk daily for 2 weeks and the other with 7.5, 5, 2.5-mpk daily for 2 weeks. Equal amounts of proteins from brain homogenates of mice receiving vehicle (labeled as “Veh”) or INK were subjected to SDS-PAGE and processed for WB with antibodies directed against marker proteins representing target engagement of INK, such as p-ULK1 (S757) and p-S6 – the drug effects are reflected by the changes within the boxed areas compared to the signals of Veh outside the boxed areas.

(B) INK was detected in the brains of mice receiving a single dose via oral gavage. After 6, 12 or 24 hrs post INK administration to 6-mo-old WT mice, INK was detected in cerebellum of mice treated with 1, 3 or 10-mpk INK. Total 60 mice: n = 6 mice/dose/time point, where the Vehicle control group was just tested once at the 6-hr time point.