Pathobiology of the autophagy-lysosomal pathway in the Huntington’s disease brain

23 Accumulated levels of mutant huntingtin protein (mHTT) and its fragments are considered 24 contributors to the pathogenesis of Huntington’s disease (HD ). Although lowering mHTT by 25 stimulating autophagy has been considered a possible therapeutic strategy , the role and 26 competence of autophagy-lysosomal pathway (ALP) during HD progression in the human disease 27 remains largely unknown. Here, we used multiplex confocal and ultrastructural 28 immunocytochemical analyses of ALP functional markers in relation to mHTT aggresome 29 pathology in striatum and the less affected cortex of HD brains staged from HD2 to HD4 by 30 Vonsattel neuropathological criteria compared to controls . Immuno labeling revealed the 31 localization of HTT/mHTT in ALP vesicular compartments labeled by autophagy-related adaptor 32 proteins p62/SQSTM1 and ubiquitin, and cathepsin D (CTSD) as well as HTT-positive inclusions. 33 Although comparatively normal at HD2, neurons at later HD stages exhibited progressive 34 enlargement and clustering of CTSD-immunoreactive autolysosomes/lysosomes and , 35 ultrastructurally, autophagic vacuole/lipofuscin granule s accumulated progressively, more 36 prominently in striatum than cortex. These changes were accompanied by rises in levels of 37 HTT/mHTT and p62/SQSTM1, particularly their fragments, in striatum but not in the cortex, and 38 by increases of LAMP1 and LAMP2 RNA and LAMP1 protein. Importantly, no blockage in 39 autophagosome formation and autophagosome -lysosome fusion was detected, thus pinpointing 40 autophagy substrate clearance deficits as a basis for autophagic flux declines . The findings 41 collectively suggest that upregulated lysosomal biogenesis and preserved proteolysis maintain 42 autophagic clearance in early-stage HD, but failure at advanced stages contributes to progressive 43 HTT build-up and potential neurotoxicity. These findings support the prospect that ALP 44 stimulation applied at early disease stages, when clearance machinery is fully competent, may have 45 therapeutic benefits in HD patients. 46 47 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 3

48 Huntington’s disease (HD) is caused by a mutation in the gene encoding the huntingtin protein 49 (HTT) resulting in expansion of the polyglutamine (polyQ) stretch on its amino -terminus (The-50 Huntington's-Disease-Collaborative-Research-Group 1993, Bates, Dorsey et al. 2015, Jiang, 51 Handley et al. 2023, Franklin, Teive et al. 2024) . In the neostriatum, the brain region most 52 vulnerable and devastated by pathology in HD, the spatiotemporal advance of atrophy dorso -53 ventrally, caudo-rostrally, and medio -laterally has been used to stage disease pathology severity 54 as Grade s 0-4 (HD0 -HD4) (Vonsattel, Myers et al. 1985) . Striatal GABA-containing medium 55 spiny projection neurons are most susceptible to cell death, even in premanifest HD0, while other 56 striatal neuronal populations, such as aspiny interneurons, seem more resistant to toxicity 57 (Vonsattel 2008, Hedreen, Berretta et al. 2024) . In the more advanced HD3 and HD4, pyramidal 58 cells in Layers III, V and VI of the cerebral cortex also show cell loss (Sotrel, Paskevich et al. 59 1991). More recent studies have revealed that loss of Layer Va pyramidal neurons, identified as 60 corticostriatal cells, can occur in the early stage of HD (Pressl, Mätlik et al. 2024) 61 Neuronal intranuclear inclusions (NII) and neuropil inclusions are present in HD brains 62 (DiFiglia, Sapp et al. 1997) and are positive for mutant huntingtin (mHTT) and ubiquitin (Ub) 63 (DiFiglia, Sapp et al. 1997, Becher, Kotzuk et al. 1998, Gutekunst, Li et al. 1999), suggesting that 64 there may be a deficiency in the proteolytic machinery responsible for normally clearing these 65 proteins, resulting in their accumulation to form the inclusions . Autophagy is generally the 66 principal mechanism by which cells clear organelles, long-lived proteins and damaged, misfolded, 67 or aggregated proteins that are poor substrates for the ubiquitin-proteasome system (UPS). In cell 68 or mouse models of HD, HTT accumulates in autophagosomes (AP) and autolysosomes (AL) 69 along with lysosomal enzyme cathepsin D (CTSD) in proportion to HTT polyQ length. This has 70 suggested that the autophagic-lysosomal pathway (ALP) may be a major, if not the major, path for 71 HTT proteolysis and degradation which mobilizes to clear an overload of exogenously expressed 72 protein, particularly forms that misfold and potentially aggregate (Kegel, Kim et al. 2000, 73 Ravikumar, Duden et al. 2002, Qin, Wang et al. 2003, Yamamoto, Cremona et al. 2006, Heng, 74 Detloff et al. 2010) . Persistent accumulation of mutant proteins/aggregates in the ALP could 75 possibly reflect impaired autolysosomal proteolytic clearance and/or inadequate autophagy 76 induction and flux of substrates through the ALP. 77 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 4 Studies in the past decade , conduct ed primarily with cell and/or animal models, have 78 demonstrated multiple roles of HTT and mHTT in autophagy in relation to HD (Martin, Ladha et 79 al. 2015, Croce and Yamamoto 2019, Klionsky, Petroni et al. 2021) . Wild type HTT participates 80 in normal autophagy by (1) releasing ULK1 from the inhibition of mTOR and (2) serving as a 81 scaffold to facilitate cargo sequestration through improving the interaction of p62/SQSTM1 (p62) 82 with ubiquitinated cargos and with LC3 (Ochaba, Lukacsovich et al. 2014, Rui, Xu et al. 2015) . 83 mHTT influences autophagy in HD settings in multiple ways . Although mHTT may activate 84 autophagy through sequestering mTOR and therefore reduce mTOR activity (Ravikumar, Vacher 85 et al. 2004) , most reported mHTT effects on the ALP appear to be inhibitory for autophagy, 86 impairing earlier stages of autophagy, including initiation signaling, phagophore nucleation and 87 cargo recognition/AP formation. The related mechanisms involve: binding to Rheb and promoting 88 mTOR signaling (Pryor, Biagioli et al. 2014) ; interfering with ULK1 activities leading to 89 impairment of the Beclin1-PIK3C3/VPS34 and ATG14 complex (Wold, Lim et al. 2016) ; 90 impairing autophagosomal cargo recognition (Martinez-Vicente, Talloczy et al. 2010, Rui, Xu et 91 al. 2015); and interfering with the interaction between Ataxin 3 and Beclin-1, resulting in Beclin-92 1 degradation by the UPS (Ashkenazi, Bento et al. 2017). Additional effects of mHTT on the endo-93 lysosomal system include inducing extensive endosomal tubulation (Kegel, Kim et al. 2000) , 94 reducing exocytosis and promoting AL accumulation (Zhou, Peskett et al. 2021), and decreasing 95 transport of late autophagic structures from the neurites to the soma (Pircs, Drouin-Ouellet et al. 96 2022). 97 It should be noted , however, that most of the above findings have been obtained from cell 98 and/or mouse models of HD, including very recent studies using induced neurons through 99 reprogramming human fibroblasts (Oh, Lee et al. 2022, Pircs, Drouin -Ouellet et al. 2022) . 100 Remarkably, information about potential alterations of the ALP in the human HD brain is very 101 limited in the literature . For example, e arly studies show that activities of several lysosomal 102 enzymes (β -glucuronidase, α -glucosidase, dipeptidyl aminopeptidase II and cathepsin H) are 103 altered in brains of patients with HD (Cross, Crow et al. 1986, Mantle, Falkous et al. 1995) . 104 Fragmentary information on early neuropathological characterization of HD brain has suggested 105 an association of HTT with endo -lysosomal compartments such as multivesicular bodies (Sapp, 106 Schwarz et al. 1997) , and an increased frequency of dystrophic neurites (Jackson, Gentleman et 107 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 5 al. 1995, Maat -Schieman, Dorsman et al. 1999) although the extent and the specific nature and 108 composition of dystrophic neurites in HD brain are little known. 109 Thus, in the HD brain, it is yet unknown whether autophagy induction is stimulated or 110 suppressed in neurons at any stage of the disease. It is also not known whether autophagy clearance 111 steps, such as AP -lysosome (LY) fusion and autolysosomal proteolytic function are competent. 112 Such information, particularly defining the site(s) at which autophagy may be disrupted, is crucial 113 to design ing possible interventions for HD based on autophagy modulation. Therefore, in th is 114 study, we aimed to characterize the status of the ALP in relation to disease progression in human 115 brain samples, employing ultrastructural, immunohistochemical, and molecular analyses for the 116 caudate nucleus of the striatum (STR) and the prefrontal cortex (CTX) from control cases and 117 HD2-HD4 patients. Importantly, for these analyses, we accessed a postmortem HD brain with 118 exceptionally well-preserved ultrastructural preservation enabling detailed neuropathological and 119 immunochemical analyses of autophagy -related alterations. To our knowledge, this study is 120 unique in terms of the relatively large number of HD cases used (Table 1 ), the level of 121 ultrastructural analysis, the range of autophagy related processes analyzed, and the value of human 122 brain-derived information regarding this disease. 123 124 125

126 HTT inclusions: types and close relationships with autophagy adaptor proteins Ub and p62 127 Consistent with previous studies reporting NIIs in human HD brains (DiFiglia, Sapp et al. 128 1997), our u ltrastructural analysis of affected neurons in the STR and CTX revealed nuclei 129 containing single discrete ovoid or irregular shaped NIIs, 1 -4 µm in diameter, composed of 130 relatively uniform meshwork of granular or short fibrous elements (Fig. 1A1, arrowheads), which 131 were not detected in control human brains (Fig. S1a). At the light microscope level, NIIs were 132 readily detected by antibodies directed against HTT (mEM48) or autophagy adaptor proteins 133 including Ub and p62 (Fig. 1A2). The extent to which nuclei were filled by mHTT, Ub or p62 134 immunoreactivity (IR) varied, ranging from only a small punctate, or the rimming of the outer 135 surface of the nuclear envelope, to progressive labeling of the entire nucleus (Fig. S1b – using Ub 136 as an example). Double immunofluorescence labeling detected a high degree of colocalization of 137 nuclear p62-IR and Ub-IR (Fig. 1A3). 138 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 6 Another major type of inclusion was neuritic inclusions randomly distributing in the neuropil. 139 They were also readily detected by anti -mHTT, -Ub and -p62 antibodies in the STR (Fig. 1B1, 140 B2) and the CTX (Fig. S1c). The majority of neuritic inclusions exhibited spherical or oval shapes, 141 with hugely varied sizes (0.5-8 µm) (Fig. 1B1, B2; Fig. S1c). High levels of colocalization were 142 observed between mHTT-IR and Ub-IR (Fig. 1B2a) or p62-IR and Ub-IR (Fig. 1B2b), implying 143 colocalization of the three proteins within the inclusions, which was seen in our study in the brain 144 of a knock-in HD mouse model Q175 (Stavrides, Goulbourne et al. 2024 bioRxiv). The extent of 145 overlapped area of the two colors from p62-IR and Ub -IR appeared to increase with disease 146 progression (Fig. 1B2b, HD2 vs HD4), suggesting that p62 and Ub proteins, and mHTT as well, 147 each can aggregate independently earlier and then further develop to form composite structures . 148 Besides these spherical/ovoid neuritic inclusions, there were elongated types of modestly enlarged 149 neurites (Fig. 1B5a) with much lower frequency – for example, they were hardly found even in 150 images at low magnification (Fig. 1B1; Fig. S1c). 151 The neuritic inclusions in the neuropil appeared to fall into 3 forms ultrastructurally. (1) Most 152 common were 1-8 µm spherical inclusions that contained no evident limiting membrane and often 153 occupied the entire cross -sectional area of a given neurite, leaving only a narrow surround of 154 cytoplasm between the inclusion and the plasma membrane of the neurite (Fig. 1B3). The internal 155 structure of these inclusions, like inclusions in the nucleus, consisted of short, fine fibrous elements 156 (Fig. 1B3, Inset). Autophagic vacuoles (AVs) or other small membranous vesicles were sometimes 157 trapped within these structures (Fig. 1B3, arrows) . (2) The 2nd type of inclusion in the neuropil, 158 which were l ess common than th e above inclusions, but still frequently seen, were 0.5-2.0 µm 159 membrane-bound spherical structures containing multiple smaller fingerprint profiles , most of 160 which appeared to be composed of well-arranged bundles of filaments or microtubules (Fig 1B4). 161 These two forms of inclusions may correspond to the spherical /ovoid type found at the light 162 microscopic level (Fig. 1B1, B2). (3) The least common inclusions in the neuropil were elongated 163 or comet -shaped structures (Fig. 1B5b) , most likely corresponding to the aforementioned 164 elongated type under light microscope (Fig. 1B5a), which were part ially or fully occupied by 165 microtubule-like elements (Fi g. 1B5b , left ) or fibrillar bundles which were positive for Ub as 166 detected by Immuno-Gold EM (IEM) (Fig. 1B5b, right). 167 In addition to the NIIs and the neuritic inclusions, there were cytoplasmic inclusions , which, 168 ultrastructurally, could be identified as 3 forms. (1) one or several large (~3 µm) ovoid or spherical 169 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 7 structures devoid of limiting membrane and consisting of mainly a meshwork of short, thin fibrous 170 elements intermixed with small numbers of membranous elements (Fig. 1C1; Fig. S1d). These 171 inclusions resembled the similar size fibrous structures seen in the neuropil (Fig. 1B3). (2) 2-5 µm 172 membrane-bound profiles containing collections of very small clear vesicles (<30 nm diameter) 173 (Fig. 1C2; Fig. S1d). (3) Fiber bundles similar to those found in the neuropil (e.g., Fig. 1B5b) were 174 occasionally observed perinuclearly within the cytoplasm (Fig. 1A1, STR, arrow). 175 176 Immunoblotting analyses of HTT proteins and adaptor proteins p62, TRAF6 and Ub 177 The above morphological observations suggest close spatial relationships among mHTT and 178 the adaptor proteins Ub and p62. Both proteins are involved in UPS and ALP proteostasis under 179 physiological conditions. In the pathological situation of abundant mHTT aggregates , however, 180 ALP may become more influential as these lesions are poor substrates for the UPS. We further 181 investigated such relationships biochemically by immunoblot analyses of these proteins in the STR 182 and the CTX from HD patients (Vonsattel neuropathological stages HD2-HD4) and corresponding 183 controls (Fig. 2). It should be noted that a large number of striatal and cortical sam ples were 184 derived from the same patients , thus lending a great deal of confidence in comparing and 185 contrasting these cohorts (see Table 1, Demographics). That said, we notably observed regionally 186 specific patterns in the levels of HTT aggregates, intact HTT protein and HTT fragments. 187 First, to evaluate mHTT aggregation, we performed immunoblotting with MW8, an antibody 188 targeting the C -terminus of exon 1 protein located in the N -terminus of HTT and capable of 189 detecting the aggregated form of HTT and the exon 1 protein (Ko, Ou et al. 2001, Baldo, Paganetti 190 et al. 2012), which showed HTT immunoreactivity rimming sample wells within the stacking gel 191 region that, in HD striatal samples, increased and peaked in HD3 (Fig. 2A1, Row 1; Fig. 2B1 a, 192 left) )(see Fig. S 2 for the uncropped full -size blots). Only trace levels were detected in cortical 193 samples; however, levels were greatest in HD3 as well. (Fig. 2A2 , Row 1; Fig. 2B1 a, right ). 194 Overall, the levels were roughly 20 -fold higher in the STR versus the CTX, suggesting that 195 aggregation occurs in a regionally specific manner. 196 Second, the levels of intact HTT (i.e., full-length HTT) were greatly reduced in the STR of HD 197 cases (as early as HD2) compared to controls when detected by another N -terminal antibody, 198 mEM48 (Gutekunst, Li et al. 1999) (Fig. 2A1, Row 3, fl (full-length)*; Fig. 2B1b left), while the 199 difference in the levels of this HTT species in the CTX between the control and HD cases was 200 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 8 only marginal (Fig. 2A2, Row 3, fl*; Fig. 2B1b left). The decline of the intact HTT in the STR of 201 HD stages was accompanied by a marked accumulation of 45 -48 kDa N -terminal fragments of 202 HTT (Fig. 2A1, Row 3; Fig. 2B1b right), while such fragments were absent in the CTX (Fig. 2A2 203 Row 3; Fig. 2B1b right) (see Discussion). 204 Immunoblotting analyses of p62 and TRAF6, two adaptor proteins known to interact with each 205 other, revealed that these two proteins also generated proteolytic fragments (32-48 kDa) which, 206 interestingly, increased also selectively in the HD STR at a level of nearly 4-fold over the controls 207 (Fig. 2A1, Rows 5, 6; Fig. 2B2), similar to the selective accumulation of mHTT fragments in the 208 HD STR described above. By contrast, there were no differences in the levels of such fragments 209 between HD and control cases in the CTX (Fig. 2A2, Rows 5, 6; Fig. 2B2). In addition, analysis 210 of ubiquitination revealed a significant increase in K48 and K63 ubiquitination and a strong trend 211 of elevation in Total ubiquitination (P = 0.06) in the STR and a significant increase only in K48 212 ubiquitination in the CTX (Fig. 2A1 and 2A2 , Rows 8-10; Fig. 2B3) . Together, the 213 generation/accumulation of proteolytic fragments of mHTT, p62 and TRAF6, and the increase in 214 ubiquitination, in the STR appear to be disease-related and brain region-selective. 215 216 ALP pathology develops at the late stages of the disease as revealed by CTSD 217 immunolabeling 218 We subsequently assessed the status of the ALP in HD brains, compared to control brains, with 219 immunohistochemistry (IHC) of CTSD, which captures the status of both LY and AL collectively. 220 AL can be distinguished from LY by residual presence of LC3 or another autophagy -specific 221 substrate. Anti-CTSD antibody labeled many neurons in multiple layers of the CTX in control 222 cases with small positive puncta (visible in some neurons even at a low magnification) (Fig. 3A). 223 Brain sections at HD2 stage exhibited a similar CTSD labeling pattern (Fig. 3B) to that seen in the 224 Controls, suggesting an absence of lysosomal pathology at this earlier stage of the disease 225 progression. However, a mild abnormal staining pattern was observed in sections from HD4, 226 exhibiting as stronger, clump ing intraneuronal IR particularly at the basal poles, and an overall 227 patchy staining pattern likely due to neuronal loss and increased staining in the neuropil (Fig. 3C). 228 Similarly, in the STR, an apparently normal CTSD staining pattern was seen in sections from 229 control cases even though there was some small clumping staining (Fig. 3D), and again, abnormal 230 staining pattern was evident at HD4 stage reflected by strongly stained punct a and clumps (of 231 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 9 puncta) within neuronal bodies or in the neuropil (Fig. 3E), which was more severe than that seen 232 in the CTX (Fig. 3C). At a higher magnification (Fig. 3F-I), small and moderately CTSD-positive 233 puncta, representing normal AL/LY morphology, were seen in neurons of control and HD2 cases 234 (Fig. 3F, G). By contrast, strongly stained, grossly enlarged CTSD granules in neurons and 235 neuropil became predominant at HD4 (Fig. 3I), indicating obvious lysosomal pathology. As 236 expected, the STR at HD3 stage exhibited a CTSD staining pattern in between what seen in HD2 237 and HD4, i.e., some neurons were still normal appearing (Fig. 3H), while others did become 238 abnormal (Fig. 3H, inset). T ogether, lysosomal pathology, as revealed by CTSD IHC, develops 239 severely in the STR and mildly in the CTX of HD brains late in the disease progression, particularly 240 at HD3-HD4 stages. 241 Immunolabeling with LC3 , another marker for the ALP, revealed LC3 puncta in neurons in 242 the HD STR which, even at HD4, were not evidently increased in number beyond the range seen 243 in control brains (Fig. 3J). Isolated neurons undergoing degeneration often labeled strongly but 244 diffusely with LC3 antibodies suggesting upregulation of LC3 -I but not necessarily increased 245 formation of LC3 -II puncta (Fig. 3J, inset). In general, the numbers of LC3 puncta and size 246 distribution did not reflect any abnormal proliferation or enlargement due to impairment in a 247 particular step of autophagy (e.g., AP formation or AP-LY fusion). 248 249 Association of mHTT signal with CTSD IR during disease progression 250 Considering the occurrence of lysosomal pathology in HD brains appearing late in the 251 pathology, we next evaluated possible relationship s of the lysosomal abnormality with the 252 potential substrates (e.g., HTT) reaching LYs through the autopha gic process using 253 immunofluorescence. Double immunolabeling for CTSD and HTT ( mEM48) in striatal neurons 254 from control cases revealed moderately stained CTSD puncta (Fig. 4A). HD neurons revealed 255 mHTT immunolabeling with mEM48 that was surrounded by CTSD (Fig. 4B, C, red boxes, and 256 4D, arrowheads). Also, at HD4, a splotchy CTSD staining pattern was evident by the fluorescent 257 labeling (Fig. 4C), similar to that revealed by DAB staining shown above (Fig. 3C, E), and CTSD 258 and/or HTT positive vesicles in many cells had become grossly enlarged and clustered (Fig. 4D). 259 The relatively weak intraluminal HTT signal at HD2 (Fig. 4B) vs. the strong signal at HD4 (Fig. 260 C, D) may imply a relatively competent lysosomal degradative function in clearing HTT in the 261 early stage and a progressive delay in the clearance of this substrate in overtly degenerating 262 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 10 neurons during disease progression . At HD4, HTT-positive inclusions in the neuropil were 263 numerous and often larger (Fig. 4C , blue boxes ), colocalized with (therefore seen as yellow 264 fluorescence), or surrounded by, CTSD IR puncta, suggesting that the HTT inclusions were too 265 large to be contained within individual ALs. These structures in the neuropil may correspond to 266 the neuritic inclusion shown in Fig. 1B3 where AVs exist inside or surrounding a large inclusion. 267 The lysosomal pathology in affected neurons in late-stage HD brains (Figs. 3, 4) involved the 268 accumulation of C TSD positive vesicles , which were decorated by HTT (Fig. 4B-D) and 269 concentrating in one pole of the neuron (Fig. 3C, 3E; Fig. 4D ), a distribution pattern associated 270 with lipofuscin granules. This was consistent with the abundance of lipofuscin granules and 271 clusters seen ultrastructurally (see below) and was further verified by IEM where lipofuscin 272 granules were positively labeled by 4 markers, CTSD, p62, Ub and HTT (Fig. 4E). 273 274 Further ultrastructural analyses of cytoplasmic vesicular organelles 275 The above single and double immunostaining demonstrate that a dominant vesicular pathology 276 in affected neurons of late-stage HD involves the accumulation and enlargement of CTSD positive 277 vesicles which correspond to AL/LY and lipofuscin granules. We then extended our EM analysis 278 to verify the incidence of these structures and to further assess autophagy related or unrelated 279 vesicular structures in brain samples from a HD4 case , in addition to the previously described 280 ultrastructures of cytoplasmic inclusions shown in Fig. 1C. 281 First, we verified that the cytoplasm of affected neurons contained a range of AVs (AP, AL, 282 LY) (<500 nm) and abundan t lipofuscin granules (which could also be referred to as pigmented 283 ALs)(Sulzer, Mosharov et al. 2008) with either typical bipartite lipid/protein morphology (double 284 arrows in Fig. 5A1, top inset; also see Fig. 1A1, CTX, top) or clusters of more amorphous granules 285 with heterogeneous content of varying electron density corresponding to early forms of lipofuscin 286 granules (Fig. 5A1, arrows; also see Fig. 1C2, right ), all reflecting incomplete degradation of 287 lysosomal substrates within . Some of these small lipofuscin granules or granule clusters were 288 surrounded by a double membrane suggesting attempts by the cell to digest these structures by 289 lysophagy, a process of AL/ LY autophagy (Fig. 5A1, bottom inset) . Second, of particular note 290 were the collections of single membrane-limited 300-500 nm vesicles in some affected neurons , 291 exhibiting a relatively uniform content of mainly granular and fibrous material (Fig. 5A2, single 292 arrowheads) resembling in part the components of large fibrous cytoplasmic inclusions described 293 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 11 above (Fig. 1C1). This population of vesicles, presumably ALs, coexisted with double membrane-294 limited structures, presumably APs (Fig. 5A2, double arrowheads) in the perikarya, implying no 295 evident block in A P-LY fusion or generation of AL s. The large number of these AVs in some 296 neurons support the concept that either the generation of APs/ALs has increased or more consistent 297 with our other data, that clearance of ALs has decreased. Third, in dystrophic neurites (Fig. 5B1, 298 B2), which were not frequent in the HD brain, AVs were readily identified which were present as 299 double-membrane bound vacuoles (i.e. APs) or multilamellar bodies (MLB) containing low 300 electron-density intralumenal content , consistent with successful sequestration and delivery of 301 autophagy cargos along the ALP. 302 Notably, there were clusters of clear vesicles of varying sizes characterized by their minimal 303 intralumenal content (i.e. “empty”), most of which were single membrane-limited (Fig. 5C1, C2, 304 single arrowheads; also see Fig. 5B2), while some exhibited a double membrane appearance (Fig. 305 5C2, double arrowheads), raising the possibility that the latter type represents the “empty APs” 306 due to defective cargo sequestration (Martinez-Vicente, Talloczy et al. 2010). On the other hand, 307 however, mitochondria were seen in double membrane vesicles (Fig. 5C1, double arrowheads), 308 implying that at least sequestration of this cargo, is unimpeded. 309 310 Immunoblotting and qPCR of lysosomal markers suggest upregulation of lysosomal 311 biogenesis in the STR 312 To survey ALP-related biochemical changes in the HD brain using immunoblotting , we 313 analyzed protein markers of the different phases of the ALP (i.e., autophagy induction signaling, 314 AV formation and lysosomal substrate clearance) on staged brains from HD2 - HD4 and control 315 cases compared in two brain regions – the STR and the CTX. Notably, the majority of marker 316 proteins for the early phases of the ALP , including p-p70S6 (as an indicator for mTOR activity), 317 BECN1, VPS34, ATG5, ATG7 and LC3, showed no changes apparent with disease progression 318 in both the STR and the CTX (not shown). 319 Next, for markers relating to lysosomal degradative functions at the later phase of the ALP, we 320 did not observe significant alterations in the protein levels (Fig. 6A, C), and enzymatic activities 321 (Fig. 6B, D), of lysosomal enzymes CTSD and CTSB involved in substrate clearance in the STR 322 (Fig. 6A, B) and the CTX (Fig. 6C, D). However, an intriguing observation was a striking elevation 323 of the levels of LAMP1 in the STR in the HD brains, appearing as early as HD2 (Fig. 6A1, A2), 324 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 12 which was otherwise not observed in the CTX (Fig. 6B1, B2), representing a major different 325 observation between the two brain regions and consistent with aforementioned CTSD IHC results 326 showing severely -altered vs. mildly -altered CTSD staining patterns in the STR vs the CTX, 327 respectively. 328 We next examined transcripts of genes in the ALP that revealed a trend of upregulation of 329 TFEB mRNA expression in the STR and CTX of HD and significant upregulation of LAMP1 and 330 LAMP2 mRNAs in the STR of HD compared to control cases (Fig. 7), which may be the basis for 331 the increased levels of LAMP1 protein in the STR of HD described above. No significant 332 alterations were found in other transcripts including LC3, p62, CTSD, CTSB and HEXA in the 333 STR and CTX of HD compared to control cases (Fig. 7). 334 335 336

337 Most of the knowledge about the relationship of the ALP with HD including the roles of 338 HTT/mHTT in autophagy has been obtained from studies using cell and animal models. However, 339 here, we analyzed a relatively large number of human brain samples from controls and HD2-HD4 340 patients, which has provided human brain -derived information that can greatly expand our 341 knowledge about the status of ALP in HD brain, its relationship to the development of mHTT 342 aggresome/inclusion pathology, and the relevance of animal models as surrogates to characterize 343 HD neuropathology and pathobiology. Moreover, the findings we have described have 344 implications as support for potential therapeutic value of specific strategies of autophagy 345 modulation in HD. 346 347 The status of ALP in the HD brain: STR vs CTX 348 At a global level, we have demonstrated in the human brain that HTT/mHTT are substrates of 349 ALP. Moreover, in HD brain, our collective data suggest that ALP in affected neurons is relatively 350 competent to maintain autophagy flux clearance capacity at an early disease stage, but that at later 351 stages, autophagy flux decline s and ALs laden with substrates including HTT accumulate . The 352 later stage pattern is a pathological state associated with neurodegeneration in more than several 353 major adult onset neurodegenerative diseases (Nixon and Rubinsztein 2024) . A gradual 354 bidirectional pathological relationship (a “vicious cycle”) is suggested between mHTT build-up 355 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 13 within AL and the decline in autolysosomal clearance efficacy that is likely multifactorial (Nixon 356 2020, Nixon and Rubinsztein 2024). The relative competence of autophagy flux at an early disease 357 stage of symptomatic HD contrasts with the temporal pattern in AD, where impaired flux emerges 358 very early in pre-symptomatic disease as a result of direct impact of disease-related genetic factors 359 on AL/LY proteolytic efficacy (Bordi, Berg et al. 2016, Nixon 2020, Im, Jiang et al. 2023, Nixon 360 and Rubinsztein 2024) and autophagy failure progresses to an unusually extreme degree as disease 361 advances (Lee, Yang et al. 2022) (Lee et al., 2024 abstract). Preservation of autophagy early in 362 symptomatic HD suggest s the potential opportunity to intervene therapeutically at this stage to 363 stimulate autophagy flux with less concern than applies to AD where overburdening failing AL/LY 364 and exacerbating build-up of toxic metabolites within the pathway may be counter-productive. 365 At the early HD2 stage, there is limited evidence suggesting a major alteration in autophagy 366 induction and upstream steps of autophagy, such as AP formation or AP-LY fusion. We cannot, 367 however, exclude a possible impairment in the engagement of certain autophagy cargoes by 368 adaptor proteins which might impede sequestration and reduce ALP flux. We observed varying 369 size clusters of double membrane limited vesicular profiles containing minimal intralumenal 370 content (Fig . 5C1, C2), raising the possibility that they represent the “empty APs” described 371 previously in mouse, cell, and patient fibroblast models of HD (Martinez-Vicente, Talloczy et al. 372 2010). Because these patterns were also often seen in fresh postmortem brains from individuals 373 with AD clinical diagnoses in our other studies (Fig. S3A,B,D,E), but not seen in our previously 374 published analyses of biopsied brain from AD patients (Nixon, Wegiel et al. 2005), we are inclined 375 to attribute much of this pattern to postmortem and fixation artifact, mainly resulting in a swelling 376 and vacuolization of endoplasmic reticulum. This is supported by the common observation in the 377 HD brain of mitochondria within AP (mitophagy) and by our further immunogold labeling by an 378 anti-calnexin antibody in HD brain which showed a significant decoration of these vesicular 379 membranes (not shown). Artifactual vacuolization of lipid granules in lipofuscin may also give 380 rise to a somewhat similar membrane pattern (Fig. S3C). Thus, our studies so far neither give 381 significant support to the presence of empty APs nor refute the notion that they may be present in 382 HD cells and cell models (Yang, Chen et al. 2021). 383 Interestingly, we detected spatial differences in the early ALP responses between the STR and 384 the CTX. In the STR, but not the CTX, increased mRNA levels of LAMP1 and LAMP2 and protein 385 levels of LAMP1 that imply lysosomal membrane expansion suggest a possible compensatory 386 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 14 upregulation in lysosomal biogenesis. The absence of changes in levels of upstream ALP 387 components ( e.g., BECN1, ATG5, ATG7 and LC3 ) supports absence of deficiency in AP 388 formation or compartment size although a caveat is the potential for glial cell autophagy systems 389 to mask changes in neuronal levels in immunoblot analyses at the brain tissue level. 390 The less vulnerable but still affected CTX displayed no significant alterations in all the ALP 391 markers examined biochemically, including enzymatic activities of lysosomal proteases and 392 mRNA levels, suggesting that an apparently normal ALP machinery is maintained in this brain 393 region. This absence of biochemical changes, together with the mild morphological changes [incl. 394 AL clustering revealed by CTSD IHC (Fig. 3, top), inclusion formation revealed by mHTT IHC 395 with mEM48 (Fig. S1c)], lower level s of aggregated HTT detected by immunoblotting with 396 antibody MW8 (Fig. S2) and minimal HTT fragment generation (Fig. 2A2), highlight the regional 397 difference of the CTX from the STR in the severity of pathological changes. 398 In neurons in both brain regions at late stages of disease, HTT/mHTT exists in CTSD-positive 399 ALs which are abnormally enlarged and clustered. The continuing accumulation of these protein 400 aggregates in the HD brain, especially in the STR, may be explained as a result of continuing 401 overload of the aggregation -prone proteins on to the neurons which is beyond the degradative 402 capacities of both macroautophagy (this study) and chaperone -mediated autophagy, which is 403 known to play a role in HTT clearance and be upregulated in experimental HD models (Koga, 404 Martinez-Vicente et al. 2011). Consistent with this, larger mHTT inclusions were observed in the 405 neuropil without being completely contained within ALs (Fig. 4C), suggesting that increasing 406 amounts of mHTT accumulation at the late stage of disease progression led to the formation of 407 aggregates outside of vesicular structures that are too large to be processed by autophagy. 408 However, we cannot exclude a possible mechanism that impairments in the interactions of adaptor 409 proteins with autophagy cargoes could lead to slower rates of clearance of substrates including 410 mHTT (Ochaba, Lukacsovich et al. 2014, Rui, Xu et al. 2015). 411 412 The classification of inclusion bodies 413 We observed nuclear, neuritic and cytoplasmic inclusions and various subtypes in each 414 category, particularly their ultrastructures . Although certain of these are described previously in 415 literature (Tellez-Nagel, Johnson et al. 1974, Roizin 1979, DiFiglia, Sapp et al. 1997, Gutekunst, 416 Li et al. 1999, Rudnicki, Pletnikova et al. 2008) , they have not been presented collectively or 417 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 15 described systematically and therefore what we present here may represent a relatively 418 comprehensive collection of the inclusion types in the human HD brain. Among the NIIs, the main 419 subtype is the pale-staining, spherical/ovoid fine granular and/or fibrous inclusion (Fig. 1A1, left). 420 This may represent the most commonly reported inclusion type in brains of both human HD 421 (Roizin 1979, DiFiglia, Sapp et al. 1997, Rudnicki, Pletnikova et al. 2008) and HD mouse models 422 including R6/2, YAC128, HdhQ92 (Davies, Turmaine et al. 1997, Bayram -Weston, Jones et al. 423 2012, Bayram-Weston, Jones et al. 2012) and Q175 (our own study)(Stavrides, Goulbourne et al. 424 2024). Among the neuritic inclusion subtypes we identified (Fig. 1B), the major subtype (Fig. 1B3) 425 also exhibited similar aggregate ultrastructure to the aforementioned main NII subtype (i.e., 426 spherical/ovoid fine granular and/or fibrous inclusion) , along with additional AVs and 427 mitochondria inside and surrounding the inclusion. This subtype was also observed in other studies 428 in the brain of human HD by HTT IEM (DiFiglia, Sapp et al. 1997, Gutekunst, Li et al. 1999) and 429 HD mouse models (Bates, Mangiarini et al. 1998, Davies, Turmaine et al. 1999, Stavrides, 430 Goulbourne et al. 2024). Again, similar aggregate ultrastructure of this type was also seen in some 431 cytoplasmic inclusions (Fig. 1C1). Together, this fine granular and/or fibrous structure of 432 aggregates appears to represent a major common ultrastructure , implying that the source of the 433 aggregate material may be the same. 434 Another major ultrastructural feature of aggregates is a fiber-bundle subtype with variable 435 shapes: rod- or comet-shaped or just parallelly arranged , which is more observed in the neurit ic 436 inclusions (Fig. 1B4, B5), but can be occasionally seen in the nucleus (Fig. 1A1, right, arrowhead) 437 and the cytoplasm (Fig. 1A1, left, arrow). The fingerprint-like features of neuritic inclusion (Fig. 438 1B4) were considered as fibrillary fascicles from abnormal mitochondria in a previous study 439 (Tellez-Nagel, Johnson et al. 1974). However, we tend to interpret them as HTT -derived fibrillar 440 bundle aggregates as revealed from our high -resolution EM images (Fig. 1B4, enlarged images), 441 and therefore include them within this fiber-bundle category. 442 These two ultrastructural aggregate types (compact fine granular and/or fine fibrous aggregate 443 vs. fiber-bundle aggregate) apparently represent different ultrastructures, whereas they are very 444 likely derived from and/or primarily composed of the same aggregating material, namely mHTT, 445 particularly mHTT N-terminal fragments especially the Exon 1 fragments, which is supported by 446 the following considerations. First, the direct evidence is that both types were positively labeled 447 by anti-HTT antibodies by IEM in both human and mouse brains, and both fine granular/fibrous 448 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 16 and bundle structures can be detected even in the same single inclusion (DiFiglia, Sapp et al. 1997, 449 Davies, Turmaine et al. 1999, Gutekunst, Li et al. 1999) . Second, studies using recombinant or 450 overexpressed proteins of mHTT N -terminal fragments have demonstrated that in addition to 451 oligomers and protofibrils, there were two main mature aggregation forms generated from the 452 protein fragments, i.e., short fibrils and more aggregated bundles, which are similar to the two 453 types of ultrastructures in the HD brain described above. Also, these studies found that the various 454 patterns of aggregation existing during the aggregation process in vitro were interconvertible 455 (Scherzinger, Lurz et al. 1997, Sahl, Weiss et al. 2012, Ko, Isas et al. 2018, Mario Isas, Pandey et 456 al. 2021). However, it should also be noted that the aggregation formation process and the final 457 ultrastructural morphologies of aggregates in vitro or in vivo can be affected by many factors such 458 as experimental conditions, peptide sequence length, posttranslational modifications, and lipids, 459 proteins and cellular membranes existing in an in vivo environment (Wang and Lashuel 2013, 460 Kolla, Gopinath et al. 2021, Riguet, Mahul-Mellier et al. 2021). 461 462 The significance of HTT and p62 fragmentation in HD 463 mHTT fragments are believed to be critical for the pathogenesis of HD and numerous studies 464 have reported the presence of N -terminal and C -terminal fragments in human STR samples 465 (Mende-Mueller, Toneff et al. 2001). Fragments may be generated by aberrant splicing of HTT or 466 proteolytic cleavage of HTT (Sathasivam, Neueder et al. 2013) . Particularly, more studies have 467 focused on N -terminal fragments and revealed their pathological significance including their 468 contribution to the formation of inclusions (Mangiarini, Sathasivam et al. 1996, Davies, Turmaine 469 et al. 1997, Lunkes, Lindenberg et al. 2002, Schilling, Klevytska et al. 2007) . Multiple sites for 470 cleavages by proteases like caspases, calpains and for posttranslational modifications have been 471 identified in the N-terminus (Kim, Yi et al. 2001, Gafni and Ellerby 2002, Wellington, Ellerby et 472 al. 2002, Saudou and Humbert 2016, Sap and Reits 2020, Chiki, Zhang et al. 2021) . In HTT-473 Knock-In mice, the majority of N -terminal fragments are most likely proteolytic products while 474 the smallest fragment, i.e., the exon 1 protein, may be a product of incomplete splicing (Landles, 475 Sathasivam et al. 2010). 476 An early study had reported the presence of HTT fragments of about 40 kDa in CTX samples 477 from juvenile HD patients (65->70 CAG) but not in those from controls, and these fragments are 478 the predominant species in the nuclear fraction (DiFiglia, Sapp et al. 1997), implying that they are 479 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 17 the major species for NII formation. In our immunoblotting studies of the two brain regions, 480 fragments of HTT (45-48 kDa) were not detected in the CTX but were readily detected in the STR 481 where they were present at much higher levels in samples of HD patients than those of control 482 cases. Together with the data from DiFiglia et al. (1997), our data suggest region-specific, disease 483 severity-dependent, and/or CAG length related generation of the fragments. These 45 -48 kDa 484 species can be considered as N -terminal fragments as they were detected by the N -terminal 485 antibody mEM48 but not by the Ab D7F7 which targets residues surrounding Pro1220 (not 486 shown). Further, based on their size of 45 -48 kDa, it is possible that they are generated from the 487 cleavage at one of the calpain proteolytic sites, 437 (Gafni and Ellerby 2002), but not from other 488 known calpain and caspase cleavages (i.e., calpain at 469 and 536; caspase at 513, 552 and 586) 489 (Gafni and Ellerby 2002, Saudou and Humbert 2016) which would have generated larger 490 fragments than 45 -48 kDa. However, a possibility that these 45 -48 kDa fragments can be the 491 sequential proteolytic products from initial fragments generated by these proteases cannot be 492 excluded (Kim, Yi et al. 2001, Gafni and Ellerby 2002) . Moreover, suspected contribution of 493 aspartic endopeptidases (Lunkes, Lindenberg et al. 2002) and/or matrix metalloproteinase (at aa 494 402) (Miller, Holcomb et al. 2010) to the generation of these fragments may also be considered. 495 On the other hand, such interpretations may not be accurate or necessary given that many 496 posttranslational modifications may occurs at the N-terminus (Saudou and Humbert 2016) and that 497 the gel migration of the HTT fragments are retarded by the expanded polyQ tract as mentioned 498 previously (Sathasivam, Neueder et al. 2013, Landles, Milton et al. 2020). All of these possibilities 499 for fragment speciation make it difficult to establish the actual aa sequence size and the responsible 500 protease(s). However, no matter how these fragments are generated, their specific increase in the 501 HD STR (vs the controls) starting at HD2 may suggest their involvement in inclusion formation 502 and proclivity to form inclusions rather than indicate only a deficit in their clearance by the 503 proteolytic systems. 504 In a number of HD mouse models, exon 1 protein generated as a consequence of aberrant 505 splicing can be detected by MW8, an antibody targeting the C -terminus of exon 1 protein and 506 capable of detecting the aggregated form of the exon 1 HTT protein, where the size of the exon 1 507 protein on the gel varies depending on the size of the polyQ (Sathasivam, Neueder et al. 2013, 508 Landles, Milton et al. 2020) . However, this exon 1 protein was not detectable with MW8 in 509 cerebellum (CBM) homogenates from HD patients (Neueder, Landles et al. 2017). In the current 510 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 18 study, we were interested in examining if exon 1 protein was present in STR and CTX samples 511 with MW8. Although we could detect aggregated HTT in the stacking gel in samples from STR 512 (Fig. 2A1, Row 1; Fig. S2) and barely in a few cases of CTX samples (Fig. 2A2, Row 1; Fig. S2), 513 we failed to detect any bands below 250 kDa as shown on the full-size blots (Fig. S2). This result 514 is not surprising given that the levels of exon 1 mRNA in CTX, CBM and hippocampus are only 515 significantly higher in juvenile HD cases but are very low in adult -onset HD patients, which are 516 not significantly different from those of controls (Neueder, Landles et al. 2017). It is notable that 517 more recent studies in human brains using a highly sensitive approach detected fair normalized 518 HTT1a (to HTTex2) expression levels in samples with CAG sizes in the juvenile -onset range. 519 However, the two samples with repeats in the adult -onset range only show very low 520 HTT1a/HTTex2 ratio, similar to the levels in the controls (Hoschek, Natan et al. 2024). Similarly, 521 aberrant spliced HTT exon1 could be detected in the brains of HD pigs, but it was expressed at a 522 much lower level than the normally spliced HTT exon products (Tong, Yang et al. 2023). Together, 523 even though it cannot be ruled out at this point that our immunoblotting approach is not sensitive 524 enough to detect the very levels of exon 1 protein, if any, the absence of MW8 band(s) below 250 525 kDa in our study , including the 45 -48 kDa range, suggests that the 45 -48 kDa fragments in our 526 STR samples (Fig. 2A) are not likely to be the exon 1 protein, but are proteolytic products 527 discussed above. Further studies may be required to verify if exon 1 protein is actually produced 528 in human brain samples, like in the brain of mouse models. A much higher level of aggregated 529 HTT in the stacking gel in HD STR samples compared to HD CTX samples is striking, suggesting 530 many more aggregates (in the form of inclusions, and maybe in other forms as well) exist in the 531 STR which would be apparently consistent with a general view that the STR is a more affected 532 region than the CTX. These differences in the levels of the aggregated HTT (by MW8) and the 45-533 48 kDa HTT fragments (by mEM48) between the STR and the CTX may be a n underlying 534 mechanism for the different degrees of lysosomal membrane expansion/AL accumulation between 535 these brain regions discussed above. 536 We also observed increased levels of p62 fragments in the STR samples of HD compared to 537 controls. Previous in vitro studies have found that caspases cleave p62 at several putative sites 538 (D256, D329, D337 and D347), generating fragments of various sizes (30, 35, 37, 40, 46 kDa) 539 (Norman, Cohen et al. 2010, Jamilloux, Lagrange et al. 2018, Sanchez -Garrido, Sancho-Shimizu 540 et al. 2018, Valionyte, Yang et al. 2022). Among these cleavages, cleavage at D329 by caspase-1 541 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 19 generates a 30 kDa fragment and diminishes p62-LC3 interaction (Jamilloux, Lagrange et al. 542 2018), or by caspase-8 generat es a 40 kDa fragment and promot es mTOR activity (Sanchez-543 Garrido, Sancho-Shimizu et al. 2018) ; and cleavage at D256 by caspase -6 produces a 30 kDa 544 fragment and attenuates p62-droplet dependent AP formation (Valionyte, Yang et al. 2022) , all 545 indicating negative effects on autophagy. In our present in vivo study, while we detected fragments 546 with similar sizes (ranging 32-48 kDa), we are not sure if any of the fragments correspond to any 547 found in vitro. Particularly, the anti-p62 antibody we used is directed against the C-terminal 20 aa, 548 and therefore we would have seen C-terminal fragments of ~ 20 kDa or smaller if caspase -549 medicated cleavages did occur (i.e., the largest C-terminal fragment released from the 440-aa full-550 length p62 would be ~184 aa if cleaved at D256 by caspase -6). Thus, the fragment size range of 551 32-48 kDa implies that other proteolytic events rather than/in addition to the caspase cleavages 552 may be involved. A potential contribution of calpain in the generation of the observed fragments 553 remains to be studied even if an in vitro study did not detect fragments after treat ing p62 with 554 calpain 1 which suggested that calpain 1 is capable of completely digesting p62 in vitro, without 555 leaving detectable fragments (Norman, Cohen et al. 2010) . Nevertheless, in terms of negative 556 effects on autophagy under a caspase-cleavage situation, our data, which indicate a moderate 557 compensatory upregulation in lysosomal biogenesis without signs of blockage in autophagy steps 558 (e.g., AP formation or AP-LY fusion) as mentioned above, do not support an existing autophagy 559 inhibition as an explanation for the p62 fragments. That a high level of fragments is present at 560 HD2 when the LY function is intact does not support a likelihood that increased levels of p62 561 fragments are due to an impaired clearance of these fragments in the ALP. 562 563 In summary, our main observations of ALP alterations in the STR from this study include: 564 moderately upregulated lysosomal biogenesis reflected by increased LAMP1 and LAMP2 marker 565 levels, engagement of mHTT as an ALP substrate, absent upstream autophagy pathway deficits, 566 and emergence of AL enlargement/clustering at late disease stages (mainly in HD4 ). These 567 findings together suggest that the autophagy machinery and the rates of autophagy flux in HD 568 brain may not significantly decline until late when mHTT degradation become s delayed. A 569 striatum specific finding is the high levels of fragments of HTT, p62 and TRAF6 , three cross -570 interacting proteins (Zucchelli, Marcuzzi et al. 2011, Linares, Duran et al. 2013), appearing at HD2 571 and persisting to HD4, which may primarily reflect enzymatic cleavages promoting formation of 572 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 20 inclusions rather than defective lysosomal clearance. The observations in HD revealing mild AV 573 accumulation and AL enlargement even at HD4 along with preserved lysosomal enzymatic 574 activities contrast with an AD-like early and progressive lysosomal proteolytic decline and 575 massive autophagic substrate build-up. These contrasting situations suggest that, unlike concerns 576 for AD therapy (Nixon and Rubinsztein 2024) , pharmacologic stimulation of autophagy in HD 577 beginning in early symptomatic stage , as proposed by others (Ravikumar, Vacher et al. 2004, 578 Sarkar and Rubinsztein 2008) , may well show greater efficacy in eliminating autophagic 579 substrates including mHTT in HD while avoiding ALP flux backup due to lysosomal degradation 580 disruption. This idea has been successfully verified in our study in the Q175 mouse model 581 (Stavrides, Goulbourne et al. 2024) where administration of a mTOR inhibitor to 6-mo-old Q175 582 normalized LY number, ameliorated aggresome pathology while reducing HTT-, p62- and Ub-IR, 583 suggesting beneficial potential of autophagy modulation at early stages of disease progression. 584 585 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 21

586 This work was supported by the CHDI Foundation (R.A.N.) and the National Institute of Aging 587 (P01 AG017617 to R.A.N.). We are very grateful to all brain banks for providing the valuable 588 human brain samples, including Harvard Brain Tissue Resource Center, Emory Center for 589 Neurodegenerative Disease, New York Brain Bank at Columbia and Mount Sinai Neuropathology 590 Brain Bank & Research Core. 591 592 CONFLICT OF INTEREST STATEMENT 593 All authors declare no conflict of interests. 594 595 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 22 FIGURE LEGENDS 596 Fig. 1. HTT inclusions: types, distributions and close relationship with autophagy adaptors 597 Ub and p62. 598 (A) Representative micrographs taken from the STR and /or CTX (as depicted on the individual 599 panels) of HD brains for demonstrating NIIs. (A1) EM images depicting the NII (arrowhead). Bar 600 = 500 nm. (A2) LM images showing that NIIs (arrowheads) are detected by antibodies to mHTT 601 (mEM48), pan-Ub or p62. Bar = 10 m. (A3) Confocal images from brain sections double-labeled 602 with anti-p62 and -Ub antibodies depicting a high degree of colocalization of the two markers 603 within NIIs. Bar = 10 m. (B) Representative micrographs taken from the STR and /or CTX (as 604 depicted on the individual panel s) of HD brains for demonstrating n euritic inclusions. (B1) LM 605 images showing neuritic inclusions (arrowheads) in the neuropil detected by antibodies to mHTT 606 (mEM48), pan-Ub or p62. Bar = 20 m. (B2) Confocal images from brain sections double-labeled 607 with anti -mHTT (mEM48) and -pan-Ub (B2a), or anti -p62 and -pan-Ub (B2b) antibodies 608 demonstrating labeled neuritic inclusions (arrows) in the neuropil. Yellow arrows depict double -609 labeled inclusions indicating colocalization, while red and green arrows point to singly labeled 610 inclusions without colocalization. Bar = 50 m (B2a) and = 20 m (B2b). (B3) A representative 611 EM image for the most common type (see text) of neuritic inclusions (double arrowheads depict 612 the boundary of the neurite), usually 1-8 m in diameter, filled with short, fine fibrous elements 613 (Inset). Arrows indicate AVs within the inclusions. Bar = 500 nm. (B4) Representative EM images 614 for the less common type (see text) of neuritic inclusions (circled by double arrowheads), usually 615 0.5-2 m in diameter, characterized by a fingerprint feature, appearing to be composed of bundles 616 of filaments/microtubules. Bar = 500 nm. (B5) Representative LM (B5a) and EM images (B5b) 617 for the least common type (see text) of neuritic inclusions (surrounded by double arrowheads) . 618 (B5a) shows the elongated feature of the inclusions revealed by mHTT (mEM48) or pan-Ub 619 antibodies. Bars = 1 0 m. (B5b) depicts the filaments/microtubules revealed by either 620 conventional EM or IEM with an anti-pan-Ub antibody, indicating specific immunogold labeling 621 on the filaments. Bars = 500 nm. (C) Representative EM micrographs taken from HD brains for 622 demonstrating cytoplasmic inclusions, which exhibit as either accumulation of fibrous filaments 623 devoid of a limiting membrane (C1), or accumulation of very small (< 30 nm in diameter) clear 624 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 23 vesicles which are positive for mHTT (mEM48) as shown by labeling of immunogolds (red 625 arrows) (C2). High resolution images for (C) are presented as Fig. S1d. 626 627 Fig. 2. Protein levels of HTT and autophagy adaptor Ub, p62 and TRAF6 in the STR and 628 CTX. 629 (A) Western blots of samples from the STR (A1) and CTX (A2) for assessing the levels of HTT 630 aggregates (with antibody MW8), HTT intact and/or fragmented species (with antibody mEM48), 631 as well as UPS/autophagy adaptor proteins including p62, TRAF6 and Ub. All tissue samples were 632 run at the same time in wide-format gels, electroblotted onto same membrane for immunodetection 633 by ECL ( Representative Ponceau S Red -stained blots or ACTB blots detected by colorimetric 634 assay using DAB are shown). Color bars above the blots denote HD staging (Blue = 635 Control/Ctl; Red = HD2; Green = HD3; Purple = HD4) . (B) Bar graphs showing the 636 quantitative results for HTT aggregates (B1a), intact and fragmented HTT species (B1b), adaptor 637 protein/aggresomal markers (B2), and adaptor protein/ubiquitination species (B3). Except B1a 638 where data from each HD stage are shown to better demonstrate the difference of signals between 639 the STR and CTX, each bar in other graphs (i.e. B1b, B2, B3) represents a single result of HD data 640 pooled from all HD2, HD3 and HD4 samples and expressed as % of Ctl (set as 100% depicted by 641 the dashed line) ± SEM. n-d: not detectable. Significant differences between the Ctl and the HD 642 case of each brain region , or between the STR and the CTX of HD case were analyzed by two-643 tailed Student’s t-test. * signs: comparisons with the Ctl; # signs: comparisons between STR and 644 CTX. * or # P<0.05, ** or ## P<0.01, *** or ### P<0.001, **** or #### P<0.0001. For STR, n = 645 7 control and 21 HD cases, for CTX, n = 10 control and 18 HD cases. Please note that the images 646 of blots for ubiquitination species were compressed in the vertical direction to be 15% relative to 647 the original height, but quantitation was done on the original blots. PSRed = Ponceau S Red 648 staining. 649 650 Fig. 3. HD brains develop ALP pathology in the later stages of the disease progression, as 651 revealed by immunostaining of CTSD. 652 (A-I) Brain sections from control and HD cases were immunostained with an anti-CTSD antibody. 653 Low magnification images were taken from the CTX of Control (Ctl) (A), HD2 (B) and HD4 (C) 654 stages and from the STR of Ctl (D) and HD4 (E) depicting the normal CTSD staining pattern in 655 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 24 the Ctl brain while abnormal pattern in the HD4 brain represented by strong and clumping IR at 656 one pol e of the neurons (arrows) and strong and increased staining in swollen neurites in the 657 neuropil (arrowheads). (F -I) High magnification images from the STR showing small punctate 658 CTSD granules (i.e., AL/ LY) in neurons of Ctl, HD2 and HD3 (F -H), while grossly enlarged 659 positive granules in HD3 (H, Inset) and dominantly in HD4 (I and Inset). (J) Representative 660 confocal images from Ctl and HD4 cases labeled for LC3, showing a lack of obvious alteration in 661 LC3 staining pattern even at the late disease stage, except that some degenerating neurons 662 exhibited enhanced diffuse staining in the cytoplasm (Inset). 663 664 Fig. 4. Association of mHTT with CTSD IR during disease progression. 665 (A-D) Representative confocal images taken from the STR from control (A), HD2 (B) and HD4 666 (C, D) cases double labeled for mHTT (mEM48) and CTSD, depicting association/colocalization 667 of mHTT IR in CTSD positive vesicles (AL/LY), where CTSD signal strongly decorates the rim 668 of the lumen (arrowheads). CTSD signal also decorates mHTT-positive neuritic inclusions in the 669 neuropil (blue boxes), in the forms of either discrete puncta (C, upper Inset) or continuous ring (C, 670 middle Inset). (D) depicts grossly enlarged (note that D has the same magnification as A -C) and 671 clustered vesicles at HD4 stage, positive for CTSD or mHTT or both. Bars = 20 m. ( E) 672 Representative IEM images of HD brain samples labeled with antibodies against either CTSD, 673 mHTT (mEM48) , pan-Ub and p62, particularly demonstrating the labeling of gold particles on 674 lipofuscin granules. Bars = 500 nm. 675 676 Fig. 5. Membranous/vesicular pathologies in the later stages of HD – Accumulation of 677 vesicles with various types of intraluminal features. 678 Representative EM images from the STR and the CTX (as depicted on the individual panels) of a 679 HD4 case demonstrating cytoplasmic membranous/vesicular pathologies. (A) Representative EM 680 images depicting recognizable cytoplasmic vesicles of the ALP, including mature lipofuscin 681 granules with bipartite protein/lipid morphology (A1, upper Inset, double arrows), early forms of 682 lipofuscin granules (A1, single arrows), double (double arrowheads) or single (single arrowheads) 683 membrane-limited vesicles (A 2), presumably corresponding to APs and ALs respectively. The 684 lower Inset in A1 depicts a compounded dense AL/lipofuscin granule within a double membrane 685 sac, implying lysophagy. Bars = 500 nm. (B) Representative EM images of dystrophic neurites 686 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 25 depicting recognizable AVs including APs with double limiting membrane (double arrowheads) 687 and multilamellar bodies (MLB, triple arrowheads). Clear vesicles with apparent single limiting 688 membrane within the same dystrophic neurite (B2) are indicated by single arrowheads. Bars = 500 689 nm. (C) Representative EM images for other vesicular structures of unidentified origins, including 690 single membrane-limited clear vesicles of varying sizes with minimal intralumenal contents (C1, 691 C2, single arrowheads), similar clear vesicles exhibiting apparent double membrane ( C2, double 692 arrowheads. Note that the inner membrane is faint, questioning a fixation artifact). The double 693 arrowheads in C1 point to a double membrane vacuole containing a mitochondrion. Bars = 500 694 nm. 695 696 Fig. 6. Levels of proteins involved in the lysosomal clearance phase of the ALP and enzymatic 697 activity assays of cathepsins in the STR and the CTX. 698 (A1 and C1) Representative western blots of striatal (A1) or cortical (C1) lysates demonstrating 699 the levels of proteins involved in the lysosomal clearance stages of the ALP in control and HD 700 cases. Color bars above the blots denote HD staging (Blue = Ctl; Red = HD2; Green = HD3; Purple 701 = HD4). The western blots are accompanied with a representative colorimetric panel of ACTB as 702 a loading control as well as a representative Ponceau S Red staining to highlight total protein s in 703 samples shown (20 g/lane). (A2 and C2) Bar graphs for quantitation results for the blots shown 704 in (A1) and (C1), respectively. Each bar represents the result of either HD2, HD3, HD4 or the 705 pooled data from all HD samples and is expressed as % of Ctl (set as 100% depicted by the dashed 706 line) ± SEM. (B and D) Quantitation of CTSD and CTSB/L enzymatic activity in striatal (B) or 707 cortical (D) lysates of controls or HD cases. Each bar represents the result of either HD2, HD3, 708 HD4 or the pooled data from all HD samples and is expressed as % of Ctl (set as 100% depicted 709 by the dashed line) ± SEM. Significant differences were analyzed by One-way ANOVA followed 710 by post hoc Tukey’s multiple comparisons test. * signs: comparisons of each bar/group with the 711 Ctl; # signs (if shown): comparisons among the groups. * P<0.05, ** P<0.01. For STR, n = 7 712 control and 21 HD cases; for CTX, n = 10 control and 18 HD cases. 713 714 Fig. 7. qPCR of selected ALP-related targets in HD brains. 715 qPCR for analysis of mRNA transcripts from the STR and the CTX of control and HD cases (HD2 716 – HD4 stages) for assessing transcripts of genes involved in the ALP. Results are expressed as 717 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 26 ΔΔCt relative to % of control ACTB levels ± SEM, with all HD staged levels pooled to yield a 718 single result. Significant differences between the control and the HD case of each brain region, or 719 between the STR and the CTX of HD case were analyzed by two-tailed Student’s t-test. * signs: 720 comparisons with the Ctl; # signs (if shown): comparisons between STR and CTX. * or # P<0.05, 721 ** or ## P<0.01, *** or ### P<0.001, **** or #### P<0.0001. For STR, n = 8 control and 20 HD 722 cases; for CTX, n = 9 control and 21 HD cases. 723 724 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 27 SUPPLEMENTAL FIGURE LEGENDS 725 Fig. S1 (related to Fig. 1). 726 a. A representative EM image from a control human brain demonstrating a neuron (center) and 727 two glial cells (top), as a reference for normal ultrastructure of neuronal perikarya. 728 b. A series of LM images from the STR at the HD3 stage depicting the varying sizes of the Ub -729 positive NIIs. 730 c. LM images showing neuritic inclusions positive for either mHTT (mEM48), pan-Ub or p62 731 randomly distributing in the neurpil in the CTX, similar to the pattern in the STR shown in main 732 Fig. 1. 733 d. These are the regular size (i.e., uncompressed) EM images shown in main Fig. 1C, facilitating 734 visualization of detailed ultrastructures. 735 736 Fig. S2 (related to Fig. 2). mHTT signal is only detected in the stacking gel but not below by 737 antibody MW8. 738 These are the full -size blots of those MW8 strip s shown in Fig. 2 ( A1, A2, Rows 1 ). Brain 739 homogenates from STR and CTX of HD patients and controls were probed with mAb MW8 (Refer 740 to the above Fig. 2 Legend for additional info). Shown are overlayed images of chemiluminescence 741 signal representing the i mmunostaining and visible light ( for easy viewing Mr Markers). Red 742 arrows depict mHTT aggregates detected in the stacking gels which have been shown in Fig. 2A1, 743 A2, Rows 1. No additional immunoreactive species are detected below the 250 kDa marker. 744 745 Fig. S3 (related to Fig. 5C). Membrane pathology observed in the AD brain. 746 EM images collected from different brain regions of individuals with AD clinical diagnoses , 747 demonstrating single or double membrane-limited clear /empty vesicles of varying sizes with 748 minimal intralumenal contents ( A, B, D, E, red single or double arrowheads, respectively ), or 749 artifactual vacuolization of lipid granules within lipofuscin granules (C, black arrowheads). Bars 750 = 500 nm. 751 752 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 28

753 Brain tissue: Brain samples were obtained from the following brain banks: Harvard Brain Tissue 754 Resource Center (HBTRC), Emory Center for Neurodegenerative Disease (ECND) and New York 755 Brain Bank at Columbia (NYBBC). These banks use Vonsatt el’s grading system of 756 neuropathological severity to stage brains from individuals diagnosed clinically as having HD as 757 Grade 0 to Grade 4 (HD0 – HD4) (Vonsattel, Myers et al. 1985). Two brain regions were used in 758 this study (STR = caudate nucleus of the striatum, CTX = prefrontal cortex) as indicated in Table 759 1 which provides detailed demographic information. 760 761 Antibodies for immunohistochemistry (IHC), western blotting (W B): The following primary 762 antibodies were used in this study. (1) from Cell Signaling Technology: tHTT rabbit mAb ( clone 763 D7F7, #5656, targeting residues surrounding Pro1220 of human HTT and detecting total HTTs), 764 p70S6K pAb (#9202), p-p70S6K (T389) pAb (#9205), ULK1 pAb (#4773), p-ULK1 (S757) pAb 765 (#6888, #14202; detecting S757 or S758 of mouse or human ULK1, respectively ), ATG5 rabbit 766 mAb (#12994), ATG7 pAb (#2631), ATG13 rabbit mAb (#13273), p-ATG13 (S355) rabbit mAb 767 (#26839), VPS34 rabbit mAb (#81453), TRAF6 rabbit mAb (#8028). (2) from Millipore -Sigma: 768 ntHTT mAb (N -Terminus-specific, mEM48, #MAB5374 , preferentially recognizing aggregated 769 HTT)(Gutekunst, Li et al. 1999) , ATG5 pAb (#ABC14), K48 - or K63-specific ubiquitin mAb 770 (#05-1307, #05-1308, respectively), III-tubulin mAb (#SAB4700544), −actin mAb (#A1978). 771 (3) from other vendors: BECN1 mAb (BD Biosciences, #612113); LC3 pAb (Novus Biologics, 772 #NB100-2220), ATG9 (Novus Biologics, #B -110-56893); p62 mAb (BD Biosciences, #610832) 773 or C-term-specific p62 Guinea Pig pAb (Progen Biotechnik #C-1620); total ubiquitin pAb (Dako 774 Agilent, #Z0458), LAMP1 or LAMP2 rat mAb (Developmental Studies Hybridoma Bank, 775 University of Iowa, #H4A3 or #H4B4, respectively); CTSD sheet pAb (D -2-3, in-house made) 776 (Cataldo, Thayer et al. 1990) ; CTSD pAb (Scripps Laboratories, #RC245), CTSD mAb (CD1.1 , 777 in-house made) (Lie, Yang et al. 2021) ; CTSB pAb (Cortex Biochemicals, # CR6009RP), CTSB 778 goat pAb (Neuromics, #GT15047). 779 The following secondary antibodies and reagents for immunoperoxidase labeling were 780 purchased from Vector Laboratories (Burlingame, CA): biotinylated goat anti -rabbit or -mouse 781 IgG/IgM, Vectastain ABC kit (PK -4000), and DAB Peroxidase Substrate Kit (SK -4100). The 782 following secondary antibodies for immunofluorescence were purchased from Thermo Fisher 783 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 29 Scientific (Waltham, MA ): Alexa Fluor 568 -conjugated goat anti -mouse IgG (A11031), Alexa 784 Fluor 488-conjugated goat anti-rabbit IgG (A11034), and Alexa Fluor 568 -conjugated goat anti-785 rabbit IgG (A11036). 786 787 Immunolabeling of brain sections: Formalin-fixed, tissue blocks of human brain (Table 1) were 788 sectioned at 40 m on a vibratome , or paraffin embedded and sectioned at 7 m. Sections were 789 deparaffinized as necessary. Antigen-retrieval was performed by boiling sections in sodium citrate 790 buffer at 95C for 30 minutes. Sections were blocked and incubated in primary antibody O/N (up 791 to 3 days in some cases) at 4 C. Alexa -Fluor conjugated secondary antibodies were used for 792 immunofluorescence and ABC detection method was used for immunoperoxidase labeling with 793 DAB. Autofluorescence was quenched with 1% Sudan black (Sigma -Aldrich; St. Louis, MO) in 794 70% ethanol for 20 minutes. DAB labeling was inspected on a Zeiss AxioSkop II equipped with a 795 HrM digital camera (Carl Zeiss, Germany). Immunofluoescent images were collected on a Zeiss 796 LSM510 Metal confocal microscope. 797 798 Ultrastructural analyses: For EM, vibratome sections of human brain (Table 1) were post-fixed 799 in 1% osmium tetroxide. Following alcohol dehydration, sections were embedded in Epon (EMS, 800 Hatfield, PA). One -micron-thick sections were stained with toluidine blue for light microscopic 801 examination and ultrathin sections prepared and stained with uranyl acetate and lead citrate. 802

was viewed with a Philips CM 10 electron microscope equipped with a digital camera 803 (Hamamatsu, model C4742 -95) aided by AMT Image Capture Engine software (version 804 5.42.443a). 805 Post-embedding IEM with gold -conjugated secondary antibody was performed to detect 806 mHTT (antibody mEM48), pan-Ub, p62 and CTSD signal in neuronal cell bodies and the neuropil 807 using a previously described protocol (88). Ultrathin sections were placed on nickel grids, air -808 dried, and etched briefly with 1% sodium metaperiodate in PBS followed by washing in filtered 809 double-distilled water and incubated with 1% BSA for 2 hours. Sections then were incubated 810 overnight in the anti -CTSD antibody (RU2, 1:1000) in a humidified chamber overnight at 4°C, 811 washed in PBS, and incubated in a secondary antibody conjugated with 10 -nm gold particles 812 (Amersham, Buckinghamshire, UK) for 2 hours at room temperature. Grids were washed and 813 briefly stained with uranyl acetate and lead citrate before examination. 814 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 30 Please note that the majority of EM images were from one HD4 case which had shorter PMI 815 and exhibited an exceptionally high level of preservation of ultrastructure compared to over 20 816 postmortem HD brains surveyed. However, we did examine additional HD4 and HD3 cases 817 (Table 1) with sufficient preservation of features to provide similar info to what we found from 818 the above HD4 case, although their ultrastructure were suboptimal for performing quantitative EM 819 analyses. 820 821 Preparation of tissue extracts: Grey matter (0.5 g) was dissected from the STR and the CTX 822 (Brodmann’s area 9/10 ) of human brains (Table 1) and homogenized in RIPA buffer (50 mM 823 Tris/HCl pH 7.4; 0.15 M NaCl; 5 mM EDTA; 1 mM EGTA; 0.5% Sodium deoxycholate; 1% 824 NP40, 0.1% SDS with protease inhibitors 1 mM AEBSF (Gold Biotechnology, St. Louis, MO) 825 and 20 g/ml of leupeptin and pepstatin (US Biochemicals, Cleveland, OH), and phosphatase 826 inhibitor microcystin LR (1 ng/ml, Enzo) . Lysates were frozen and thawed 3 times followed by 827 centrifugation at 10,000 g for 30 min to yield a total tissue lysate supernatant. Protein content was 828 determined by the BCA method (Smith, Krohn et al. 1985) . Brain lysates were examined by 829 western blotting for various marker proteins for the ALP. 830 831 SDS-PAGE and western blotting: Lysate extracts (10-40 g total protein) were separated on 4 -832 20% or 10% Tris-glycine SDS-PAGE gels and transferred to nitrocellulose (Pall, Pensacola, FL) 833 for probing with antisera as noted along with appropriate external controls. Blots were blocked 834 for 1 hr at 37oC in 1x TBST containing 5% blotting grade dry milk (W/V), incubated in 1 o Ab in 835 block solution O/N at 4oC, washed 3x 10 min in TBST at RT followed by incubation with 2 o Ab 836 conjugated to horseradish peroxidase diluted in block solution for 90 min at RT. Blots were 837 washed 3x 10 min and immunoreactive bands were visualized with ECL reagent ( RPN2209, 838 Millipore Sigma) and the bands were quantified using MultiGauge V. 3.0 (Fuji Film) software . 839 Target proteins were normalized against -actin, unless otherwise noted. 840 Exposures were approximated for multiple conditions based on levels found for external 841 controls as noted in Results. In some cases, especially in cases of assaying housekeeping or other 842 highly expressed antigens that exposure times are difficult to control and maintain signal linearity 843 on film, blots were developed using DAB enhanced with nickel. Blots were stripped in 0.2 M 844 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 31 HCl, pH 2.2 containing 0.1% SDS and 1% Triton X-100 for 1.5 hr at 25oC and reprobed using the 845 above protocol. 846 847 RNA preparation: For routine qPCR analyses 100 mg of grey matter from the STR and the CTX 848 (Brodmann’s area 9/10) of human brains (Table 1) was dissected and extracted in 1.5 ml Trizol 849 reagent (ThermoFisher/Life Technologies, 15596026) using a hand -held homogenizer (6 x15 s 850 bursts)(Pro-Scientific, Oxford, CT) followed by mixing with 300 L of chloroform. Samples were 851 centrifuged at 12000 g for 15 min at 4 °C. The aqueous phase was collected and 750 L 852 isopropanol was added; samples were spun again at 12000 g for 10 min at 4 oC. Supernatant was 853 removed and pellet washed 2 times with 75% ice-cold ethanol and centrifuged 7500 g for 5 min at 854 4oC. The pellet was redissolved in 100 L of RNAase -free distilled water. Total RNA quality 855 (RNA Integrity Number, RIN) was assessed on an Agilent Bioanalyzer using an RNA Nanochip 856 (2100; Agilent Technologies; Santa Clara, CA). RNA quantity was interpolated from the Agilent 857 chip by using an RNA ladder with a known concentration of 150 ng. 858 859 Preparation of cDNA and qPCR: Starting concentrations of total RNA were normalized for 860 samples whose concentrations were estimated on different Agilent chips. cDNA was prepared 861 using TaqMan Reverse Transciption Reagent kit (Applied Biosystems; Branchburg, NJ) according 862 to manufacturer’s instructions. Following reverse transcription, sample cDNA was loaded in 863 triplicate into wells of a 96 -well optical reaction plate containing appropriate target gene primer 864 (Applied Biosystems, Branchburg, NJ). GAPDH (glyceraldehyde 3 -phosphate dehydrogenase), 865 ACTB (actin ), and HPRT1 (hypoxanthine phosphoribosyltransferase 1) were run as 866 housekeeping genes, also in triplicate, for each sample and on the same plate, as endogenous 867 controls. Total reaction volume per well was 20 L. qPCR was performed in the ABI Prism 868 7900HT Sequence Detection System (Applied Biosystems Branchburg, NJ) as described 869 previously (Alldred, 2015; Alldred, 2015). 870 871 Calculation of qPCR results: Following qPCR, the target genes were normalized against the three 872 housekeeping genes (GAPDH, ACTB, and HPRT). Results were calculated using the Ct 873

(Applied Biosystems, Branchburg, NJ Bulletin #2). Control values were averaged as a 874 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 32 geometric mean, and sample values were recalculated and expressed as percent control. Outliers 875 were recognized as values falling beyond two standard deviation s of mean, and were discarded 876 from the analyses. 877 878 Enzymatic assays in brain homogenates: Cathepsins B and L were assayed by measuring the 879 release of 7-amino-4-methylcoumarin (amc) from Z-Phe-Arg-amc at pH 5.5 (substrate recognized 880 by both enzymes (Enzo, Plymouth Reading, PA) modified from the method of Barrett and 881 Kirschke (Barrett, 1981 ) to utilize microplate procedures. Typically, assays were performed in 882 black microplates in a volume of 100 l mixture (1-5 l of enzyme in 50 mM NA-Acetate, pH 6.5 883 containing 1 mM EDTA and 10 M Z-Phe-Arg-amc). Fluorescence of amc released was read at 884 different time points in a Wallac Victor -2 spectrofluorimetric plate reader with a filter set 885 optimized for detection of 4-methyl-7-aminocoumarin (-amc) standard solution with excitation at 886 365 nm and emission at 440 nm. The reaction was linear up to 2 hours. Enzyme activity was 887 expressed as the amount of amc released per hour per mg protein. 888 Cathepsin D was assayed at 37oC at pH 4.0 by measuring the release of amc containing peptide, 889 7-methoxycoumarin-4-acetyl-Gly-Lys-Pro-Ile-Leu-Phe from 7 -methoxycoumarin-4-acetyl-Gly- 890 Lys-Pro-Ile-Leu-Phe-Phe-Arg-Leu-Lys (Dnp) -D-Arg-NH2 (BioMol-Enzo, Plymouth Reading, 891 PA), according to the method of Yasuda et.al. (Yasuda, 1999). Assays were performed in black 892 microplates in a total volume of 100 l (0.1 M sodium acetate buffer pH 4.0 containing 20 M 893 substrate with and without 3 g of pepstatin) for one hour. Fluorescence released was read in a 894 Wallac Victor -2 Spectroflurimetric plate reader with a filter optimized for detection of amc 895 standard solution with excitation at 365 nm and emission at 440 nm. However instead of using 896 amc standard, a quenched standard 7 -methoxycoumarin-4-acetyl-Pro-Leu-OH was used for 897 expressing enzyme activity to account for the release of peptide containing amc instead of free 898 amc. Enzyme activity was expressed as the relative amount of quenched standard released per 899 hour per mg protein. The specific activity of cathepsins were calculated by calculating the ratio of 900 enzyme activity to the densitometric data obtained from western blots for each enzyme. 901 902 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 33

903 904 Ashkenazi, A., C. F. Bento, T. Ricketts, M. Vicinanza, F. Siddiqi, M. Pavel, F. Squitieri, M. C. Hardenberg, S. 905 Imarisio, F. M. Menzies and D. C. Rubinsztein (2017). "Polyglutamine tracts regulate beclin 1 -906 dependent autophagy." Nature 545(7652): 108-111. 907 Baldo, B., P. Paganetti, S. Grueninger, D. Marcellin, L. S. Kaltenbach, D. C. Lo, M. Semmelroth, A. Zivanovic, 908 D. Abramowski, D. Smith, G. P. Lotz, G. P. Bates and A. Weiss (2012). "TR -FRET-based duplex 909 immunoassay reveals an inverse correlation of soluble and aggregated mutant huntingtin in 910 huntington's disease." Chem Biol 19(2): 264-275. 911 Bates, G. P., R. Dorsey, J. F. Gusella, M. R. Hayden, C. Kay, B. R. Leavitt, M. Nance, C. A. Ross, R. I. Scahill, 912 R. Wetzel, E. J. Wild and S. J. Tabrizi (2015). "Huntington disease." Nat Rev Dis Primers 1: 15005. 913 Bates, G. P., L. Mangiarini and S. W. Davies (1998). "Transgenic mice in the study of polyglutamine repeat 914 expansion diseases." Brain Pathol 8(4): 699-714. 915 Bayram-Weston, Z., L. Jones, S. B. Dunnett and S. P. Brooks (2012). "Light and electron microscopic 916 characterization of the evolution of cellular pathology in HdhQ92 Huntington's disease knock -in 917 mice." Brain Res Bull 88(2-3): 171-181. 918 Bayram-Weston, Z., L. Jones, S. B. Dunnett and S. P. Brooks (2012). "Light and electron microscopic 919 characterization of the evolution of cellular pathology in YAC128 Huntington's disease transgenic 920 mice." Brain Res Bull 88(2-3): 137-147. 921 Becher, M. W., J. A. Kotzuk, A. H. Sharp, S. W. Davies, G. P. Bates, D. L. Price and C. A. Ross (1998). 922 "Intranuclear neuronal inclusions in Huntington's disease and dentatorubral and pallidoluysian 923 atrophy: correlation between the density of inclusions and IT15 CAG triplet repeat length." 924 Neurobiol Dis 4(6): 387-397. 925 Bordi, M., M. J. Berg, P. S. Mohan, C. M. Peterhoff, M. J. Alldred, S. Che, S. D. Ginsberg and R. A. Nixon 926 (2016). "Autophagy flux in CA1 neurons of Alzheimer hippocampus: Increased induction 927 overburdens failing lysosomes to propel neuritic dystrophy." Autophagy 12(12): 2467-2483. 928 Cataldo, A. M., C. Y. Thayer, E. D. Bird, T. R. Wheelock and R. A. Nixon (1990). "Lysosomal proteinase 929 antigens are prominently localized within senile plaques of Alzheimer's disease: evidence for a 930 neuronal origin." Brain Res 513(2): 181-192. 931 Chiki, A., Z. Zhang, K. Rajasekhar, L. A. Abriata, I. Rostami, L. F. Krapp, D. Boudeffa, M. Dal Peraro and H. 932 A. Lashuel (2021). "Investigating Crosstalk Among PTMs Provides Novel Insight Into the Structural 933 Basis Underlying the Differential Effects of Nt17 PTMs on Mutant Httex1 Aggregation." Front Mol 934 Biosci 8: 686086. 935 Croce, K. R. and A. Yamamoto (2019). "A role for autophagy in Huntington's disease." Neurobiol Dis 122: 936 16-22. 937 Cross, A. J., T. J. Crow, J. M. Dawson, I. N. Ferrier, J. A. Johnson, T. J. Peters and G. P. Reynolds (1986). 938 "Subcellular pathology of human neurodegenerative disorders: Alzheimer -type dementia and 939 Huntington's disease." J Neurochem 47(3): 882-889. 940 Davies, S. W., M. Turmaine, B. A. Cozens, M. DiFiglia, A. H. Sharp, C. A. Ross, E. Scherzinger, E. E. Wanker, 941 L. Mangiarini and G. P. Bates (1997). "Formation of neuronal intranuclear inclusions underlies the 942 neurological dysfunction in mice transgenic for the HD mutation." Cell 90(3): 537-548. 943 Davies, S. W., M. Turmaine, B. A. Cozens, A. S. Raza, A. Mahal, L. Mangiarini and G. P. Bates (1999). "From 944 neuronal inclusions to neurodegeneration: neuropathological investigation of a transgenic mouse 945 model of Huntington's disease." Philos Trans R Soc Lond B Biol Sci 354(1386): 981-989. 946 DiFiglia, M., E. Sapp, K. O. Chase, S. W. Davies, G. P. Bates, J. P. Vonsattel and N. Aronin (1997). 947 "Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain." 948 Science 277(5334): 1990-1993. 949 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 34 Franklin, G. L., H. A. G. Teive, F. S. Tensini, C. H. F. Camargo, N. S. C. de Lima, D. C. de Dos Santos, A. T. 950 Meira and S. J. Tabrizi (2024). "The Huntington's Disease Gene Discovery." Mov Disord 39(2): 227-951 234. 952 Gafni, J. and L. M. Ellerby (2002). "Calpain activation in Huntington's disease." J Neurosci 22(12): 4842-953 4849. 954 Gutekunst, C. A., S. H. Li, H. Yi, J. S. Mulroy, S. Kuemmerle, R. Jones, D. Rye, R. J. Ferrante, S. M. Hersch 955 and X. J. Li (1999). "Nuclear and neuropil aggregates in Huntington's disease: relationship to 956 neuropathology." J Neurosci 19(7): 2522-2534. 957 Hedreen, J. C., S. Berretta and C. L. White Iii (2024). "Postmortem neuropathology in early Huntington 958 disease." J Neuropathol Exp Neurol 83(5): 294-306. 959 Heng, M. Y., P. J. Detloff, H. L. Paulson and R. L. Albin (2010). "Early alterations of autophagy in Huntington 960 disease-like mice." Autophagy 6(8): 1206-1208. 961 Hoschek, F., J. Natan, M. Wagner, K. Sathasivam, A. Abdelmoez, B. von Einem, G. P. Bates, G. B. 962 Landwehrmeyer and A. Neueder (2024). "Huntingtin HTT1a is generated in a CAG repeat -length-963 dependent manner in human tissues." Mol Med 30(1): 36. 964 Im, E., Y. Jiang, P. H. Stavrides, S. Darji, H. Erdjument-Bromage, T. A. Neubert, J. Y. Choi, J. Wegiel, J. H. Lee 965 and R. A. Nixon (2023). "Lysosomal dysfunction in Down syndrome and Alzheimer mouse models is 966 caused by v-ATPase inhibition by Tyr(682)-phosphorylated APP betaCTF." Sci Adv 9(30): eadg1925. 967 Jackson, M., S. Gentleman, G. Lennox, L. Ward, T. Gray, K. Randall, K. Morrell and J. Lowe (1995). "The 968 cortical neuritic pathology of Huntington's disease." Neuropathol Appl Neurobiol 21(1): 18-26. 969 Jamilloux, Y., B. Lagrange, A. Di Micco, E. Bourdonnay, A. Provost, R. Tallant, T. Henry and F. Martinon 970 (2018). "A proximity -dependent biotinylation (BioID) approach flags the p62/sequestosome -1 971 protein as a caspase-1 substrate." J Biol Chem 293(32): 12563-12575. 972 Jiang, A., R. R. Handley, K. Lehnert and R. G. Snell (2023). "From Pathogenesis to Therapeutics: A Review 973 of 150 Years of Huntington's Disease Research." Int J Mol Sci 24(16). 974 Kegel, K. B., M. Kim, E. Sapp, C. McIntyre, J. G. Castano, N. Aronin and M. DiFiglia (2000). "Huntingtin 975 expression stimulates endosomal -lysosomal activity, endosome tubulation, and autophagy." J 976 Neurosci 20(19): 7268-7278. 977 Kim, Y. J., Y. Yi, E. Sapp, Y. Wang, B. Cuiffo, K. B. Kegel, Z. H. Qin, N. Aronin and M. DiFiglia (2001). "Caspase 978 3-cleaved N -terminal fragments of wild -type and mutant huntingtin are present in normal and 979 Huntington's disease brains, associate with membranes, and undergo calpain -dependent 980 proteolysis." Proc Natl Acad Sci U S A 98(22): 12784-12789. 981 Klionsky, D. J., G. Petroni, R. K. Amaravadi, E. H. Baehrecke, A. Ballabio, P. Boya, J. M. Bravo-San Pedro, K. 982 Cadwell, F. Cecconi, A. M. K. Choi, M. E. Choi, C. T. Chu, P. Codogno, M. I. Colombo, A. M. Cuervo, 983 V. Deretic, I. Dikic, Z. Elazar, E. L. Eskelinen, G. M. Fimia, D. A. Gewirtz, D. R. Green, M. Hansen, M. 984 Jaattela, T. Johansen, G. Juhasz, V. Karantza, C. Kraft, G. Kroemer, N. T. Ktistakis, S. Kumar, C. Lopez-985 Otin, K. F. Macleod, F. Madeo, J. Martinez, A. Melendez, N. Mizushima, C. Munz, J. M. Penninger, R. 986 M. Perera, M. Piacentini, F. Reggiori, D. C. Rubinsztein, K. M. Ryan, J. Sadoshima, L. Santambrogio, 987 L. Scorrano, H. U. Simon, A. K. Simon, A. Simonsen, A. Stolz, N. Tavernarakis, S. A. Tooze, T. 988 Yoshimori, J. Yuan, Z. Yue, Q. Zhong, L. Galluzzi and F. Pietrocola (2021). "Autophagy in major human 989 diseases." EMBO J 40(19): e108863. 990 Ko, J., J. M. Isas, A. Sabbaugh, J. H. Yoo, N. K. Pandey, A. Chongtham, M. Ladinsky, W. L. Wu, H. Rohweder, 991 A. Weiss, D. Macdonald, I. Munoz -Sanjuan, R. Langen, P. H. Patterson and A. Khoshnan (2018). 992 "Identification of distinct conformations associated with monomers and fibril assemblies of mutant 993 huntingtin." Hum Mol Genet 27(13): 2330-2343. 994 Ko, J., S. Ou and P. H. Patterson (2001). "New anti -huntingtin monoclonal antibodies: implications for 995 huntingtin conformation and its binding proteins." Brain Res Bull 56(3-4): 319-329. 996 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 35 Koga, H., M. Martinez -Vicente, E. Arias, S. Kaushik, D. Sulzer and A. M. Cuervo (2011). "Constitutive 997 upregulation of chaperone-mediated autophagy in Huntington's disease." J Neurosci 31(50): 18492-998 18505. 999 Kolla, R., P. Gopinath, J. Ricci, A. Reif, I. Rostami and H. A. Lashuel (2021). "A New Chemoenzymatic 1000 Semisynthetic Approach Provides Insight into the Role of Phosphorylation beyond Exon1 of 1001 Huntingtin and Reveals N -Terminal Fragment Length -Dependent Distinct Mechanisms of 1002 Aggregation." J Am Chem Soc 143(26): 9798-9812. 1003 Landles, C., R. E. Milton, N. Ali, R. Flomen, M. Flower, F. Schindler, C. Gomez-Paredes, M. K. Bondulich, G. 1004 F. Osborne, D. Goodwin, G. Salsbury, C. L. Benn, K. Sathasivam, E. J. Smith, S. J. Tabrizi, E. E. Wanker 1005 and G. P. Bates (2020). "Subcellular Localization And Formation Of Huntingtin Aggregates Correlates 1006 With Symptom Onset And Progression In A Huntington'S Disease Model." Brain Commun 2(2): 1007 fcaa066. 1008 Landles, C., K. Sathasivam, A. Weiss, B. Woodman, H. Moffitt, S. Finkbeiner, B. Sun, J. Gafni, L. M. Ellerby, 1009 Y. Trottier, W. G. Richards, A. Osmand, P. Paganetti and G. P. Bates (2010). "Proteolysis of mutant 1010 huntingtin produces an exon 1 fragment that accumulates as an aggregated protein in neuronal 1011 nuclei in Huntington disease." J Biol Chem 285(12): 8808-8823. 1012 Lee, J. H., D. S. Yang, C. N. Goulbourne, E. Im, P. Stavrides, A. Pensalfini, H. Chan, C. Bouchet -Marquis, C. 1013 Bleiwas, M. J. Berg, C. Huo, J. Peddy, M. Pawlik, E. Levy, M. Rao, M. Staufenbiel and R. A. Nixon 1014 (2022). "Faulty autolysosome acidification in Alzheimer's disease mouse models induces autophagic 1015 build-up of Abeta in neurons, yielding senile plaques." Nat Neurosci 25(6): 688-701. 1016 Lie, P. P. Y., D. S. Yang, P. Stavrides, C. N. Goulbourne, P. Zheng, P. S. Mohan, A. M. Cataldo and R. A. Nixon 1017 (2021). "Post -Golgi carriers, not lysosomes, confer lysosomal properties to pre -degradative 1018 organelles in normal and dystrophic axons." Cell Rep 35(4): 109034. 1019 Linares, J. F., A. Duran, T. Yajima, M. Pasparakis, J. Moscat and M. T. Diaz -Meco (2013). "K63 1020 polyubiquitination and activation of mTOR by the p62 -TRAF6 complex in nutrient-activated cells." 1021 Mol Cell 51(3): 283-296. 1022 Lunkes, A., K. S. Lindenberg, L. Ben-Haiem, C. Weber, D. Devys, G. B. Landwehrmeyer, J. L. Mandel and Y. 1023 Trottier (2002). "Proteases acting on mutant huntingtin generate cleaved products that 1024 differentially build up cytoplasmic and nuclear inclusions." Mol Cell 10(2): 259-269. 1025 Maat-Schieman, M. L., J. C. Dorsman, M. A. Smoor, S. Siesling, S. G. Van Duinen, J. J. Verschuuren, J. T. den 1026 Dunnen, G. J. Van Ommen and R. A. Roos (1999). "Distribution of inclusions in neuronal nuclei and 1027 dystrophic neurites in Huntington disease brain." J Neuropathol Exp Neurol 58(2): 129-137. 1028 Mangiarini, L., K. Sathasivam, M. Seller, B. Cozens, A. Harper, C. Hetherington, M. Lawton, Y. Trottier, H. 1029 Lehrach, S. W. Davies and G. P. Bates (1996). "Exon 1 of the HD gene with an expanded CAG repeat 1030 is sufficient to cause a progressive neurological phenotype in transgenic mice." Cell 87(3): 493-506. 1031 Mantle, D., G. Falkous, S. Ishiura, R. H. Perry and E. K. Perry (1995). "Comparison of cathepsin protease 1032 activities in brain tissue from normal cases and cases with Alzheimer's disease, Lewy body 1033 dementia, Parkinson's disease and Huntington's disease." J Neurol Sci 131(1): 65-70. 1034 Mario Isas, J., N. K. Pandey, H. Xu, K. Teranishi, A. K. Okada, E. K. Fultz, A. Rawat, A. Applebaum, F. Meier, 1035 J. Chen, R. Langen and A. B. Siemer (2021). "Huntingtin fibrils with different toxicity, structure, and 1036 seeding potential can be interconverted." Nat Commun 12(1): 4272. 1037 Martin, D. D., S. Ladha, D. E. Ehrnhoefer and M. R. Hayden (2015). "Autophagy in Huntington disease and 1038 huntingtin in autophagy." Trends Neurosci 38(1): 26-35. 1039 Martinez-Vicente, M., Z. Talloczy, E. Wong, G. Tang, H. Koga, S. Kaushik, R. de Vries, E. Arias, S. Harris, D. 1040 Sulzer and A. M. Cuervo (2010). "Cargo recognition failure is responsible for inefficient autophagy 1041 in Huntington's disease." Nat Neurosci 13(5): 567-576. 1042 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 36 Mende-Mueller, L. M., T. Toneff, S. R. Hwang, M. F. Chesselet and V. Y. Hook (2001). "Tissue -specific 1043 proteolysis of Huntingtin (htt) in human brain: evidence of enhanced levels of N- and C-terminal htt 1044 fragments in Huntington's disease striatum." J Neurosci 21(6): 1830-1837. 1045 Miller, J. P., J. Holcomb, I. Al -Ramahi, M. de Haro, J. Gafni, N. Zhang, E. Kim, M. Sanhueza, C. Torcassi, S. 1046 Kwak, J. Botas, R. E. Hughes and L. M. Ellerby (2010). "Matrix metalloproteinases are modifiers of 1047 huntingtin proteolysis and toxicity in Huntington's disease." Neuron 67(2): 199-212. 1048 Neueder, A., C. Landles, R. Ghosh, D. Howland, R. H. Myers, R. L. M. Faull, S. J. Tabrizi and G. P. Bates 1049 (2017). "The pathogenic exon 1 HTT protein is produced by incomplete splicing in Huntington's 1050 disease patients." Sci Rep 7(1): 1307. 1051 Nixon, R. A. (2020). "The aging lysosome: An essential catalyst for late-onset neurodegenerative diseases." 1052 Biochim Biophys Acta Proteins Proteom 1868(9): 140443. 1053 Nixon, R. A. and D. C. Rubinsztein (2024). "Mechanisms of Autophagy -Lysosome Dysfunction in 1054 Neurodegenerative Diseases. ." Nat Rev Mol Cell Biol In press. 1055 Nixon, R. A., J. Wegiel, A. Kumar, W. H. Yu, C. Peterhoff, A. Cataldo and A. M. Cuervo (2005). "Extensive 1056 involvement of autophagy in Alzheimer disease: an immuno -electron microscopy study." J 1057 Neuropathol Exp Neurol 64(2): 113-122. 1058 Norman, J. M., G. M. Cohen and E. T. Bampton (2010). "The in vitro cleavage of the hAtg proteins by cell 1059 death proteases." Autophagy 6(8): 1042-1056. 1060 Ochaba, J., T. Lukacsovich, G. Csikos, S. Zheng, J. Margulis, L. Salazar, K. Mao, A. L. Lau, S. Y. Yeung, S. 1061 Humbert, F. Saudou, D. J. Klionsky, S. Finkbeiner, S. O. Zeitlin, J. L. Marsh, D. E. Housman, L. M. 1062 Thompson and J. S. Steffan (2014). "Potential function for the Huntingtin protein as a scaffold for 1063 selective autophagy." Proc Natl Acad Sci U S A 111(47): 16889-16894. 1064 Oh, Y. M., S. W. Lee, W. K. Kim, S. Chen, V. A. Church, K. Cates, T. Li, B. Zhang, R. E. Dolle, S. Dahiya, S. C. 1065 Pak, G. A. Silverman, D. H. Perlmutter and A. S. Yoo (2022). "Age -related Huntington's disease 1066 progression modeled in directly reprogrammed patient-derived striatal neurons highlights impaired 1067 autophagy." Nat Neurosci 25(11): 1420-1433. 1068 Pircs, K., J. Drouin -Ouellet, V. Horvath, J. Gil, M. Rezeli, R. Garza, D. A. Grassi, Y. Sharma, I. St -Amour, K. 1069 Harris, M. E. Jonsson, P. A. Johansson, R. Vuono, S. V. Fazal, T. Stoker, B. A. Hersbach, K. Sharma, J. 1070 Lagerwall, S. Lagerstrom, P. Storm, S. S. Hebert, G. Marko -Varga, M. Parmar, R. A. Barker and J. 1071 Jakobsson (2022). "Distinct subcellular autophagy impairments in induced neurons from patients 1072 with Huntington's disease." Brain 145(9): 3035-3057. 1073 Pressl, C., K. Mätlik, L. Kus, P. Darnell, J. D. Luo, M. R. Paul, A. R. Weiss, W. Liguore, T. S. Carroll, D. A. Davis, 1074 J. McBride and N. Heintz (2024). "Selective vulnerability of layer 5a corticostriatal neurons in 1075 Huntington's disease." Neuron. 1076 Pryor, W. M., M. Biagioli, N. Shahani, S. Swarnkar, W. C. Huang, D. T. Page, M. E. MacDonald and S. 1077 Subramaniam (2014). "Huntingtin promotes mTORC1 signaling in the pathogenesis of Huntington's 1078 disease." Sci Signal 7(349): ra103. 1079 Qin, Z. H., Y. Wang, K. B. Kegel, A. Kazantsev, B. L. Apostol, L. M. Thompson, J. Yoder, N. Aronin and M. 1080 DiFiglia (2003). "Autophagy regulates the processing of amino terminal huntingtin fragments." Hum 1081 Mol Genet 12(24): 3231-3244. 1082 Ravikumar, B., R. Duden and D. C. Rubinsztein (2002). "Aggregate-prone proteins with polyglutamine and 1083 polyalanine expansions are degraded by autophagy." Hum Mol Genet 11(9): 1107-1117. 1084 Ravikumar, B., C. Vacher, Z. Berger, J. E. Davies, S. Luo, L. G. Oroz, F. Scaravilli, D. F. Easton, R. Duden, C. J. 1085 O'Kane and D. C. Rubinsztein (2004). "Inhibition of mTOR induces autophagy and reduces toxicity 1086 of polyglutamine expansions in fly and mouse models of Huntington disease." Nat Genet 36(6): 585-1087 595. 1088 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 37 Riguet, N., A. L. Mahul -Mellier, N. Maharjan, J. Burtscher, M. Croisier, G. Knott, J. Hastings, A. Patin, V. 1089 Reiterer, H. Farhan, S. Nasarov and H. A. Lashuel (2021). "Nuclear and cytoplasmic huntingtin 1090 inclusions exhibit distinct biochemical composition, interactome and ultrastructural properties." 1091 Nat Commun 12(1): 6579. 1092 Roizin, L. S., S.; Liu, J.C. (1979). Neuronal nuclear -cytoplasmic changes in Huntington's chorea: Electron 1093 microscope investigation. T. N. Chase. New York, Raven Press. Advances in Neurology 23: 95-122. 1094 Rudnicki, D. D., O. Pletnikova, J. P. Vonsattel, C. A. Ross and R. L. Margolis (2008). "A comparison of 1095 huntington disease and huntington disease-like 2 neuropathology." J Neuropathol Exp Neurol 67(4): 1096 366-374. 1097 Rui, Y. N., Z. Xu, B. Patel, Z. Chen, D. Chen, A. Tito, G. David, Y. Sun, E. F. Stimming, H. J. Bellen, A. M. 1098 Cuervo and S. Zhang (2015). "Huntingtin functions as a scaffold for selective macroautophagy." Nat 1099 Cell Biol 17(3): 262-275. 1100 Sahl, S. J., L. E. Weiss, W. C. Duim, J. Frydman and W. E. Moerner (2012). "Cellular inclusion bodies of 1101 mutant huntingtin exon 1 obscure small fibrillar aggregate species." Sci Rep 2: 895. 1102 Sanchez-Garrido, J., V. Sancho -Shimizu and A. R. Shenoy (2018). "Regulated proteolysis of p62/SQSTM1 1103 enables differential control of autophagy and nutrient sensing." Sci Signal 11(559). 1104 Sap, K. A. and E. A. Reits (2020). "Strategies to Investigate Ubiquitination in Huntington's Disease." Front 1105 Chem 8: 485. 1106 Sapp, E., C. Schwarz, K. Chase, P. G. Bhide, A. B. Young, J. Penney, J. P. Vonsattel, N. Aronin and M. DiFiglia 1107 (1997). "Huntingtin localization in brains of normal and Huntington's disease patients." Ann Neurol 1108 42(4): 604-612. 1109 Sarkar, S. and D. C. Rubinsztein (2008). "Huntington's disease: degradation of mutant huntingtin by 1110 autophagy." FEBS J 275(17): 4263-4270. 1111 Sathasivam, K., A. Neueder, T. A. Gipson, C. Landles, A. C. Benjamin, M. K. Bondulich, D. L. Smith, R. L. 1112 Faull, R. A. Roos, D. Howland, P. J. Detloff, D. E. Housman and G. P. Bates (2013). "Aberrant splicing 1113 of HTT generates the pathogenic exon 1 protein in Huntington disease." Proc Natl Acad Sci U S A 1114 110(6): 2366-2370. 1115 Saudou, F. and S. Humbert (2016). "The Biology of Huntingtin." Neuron 89(5): 910-926. 1116 Scherzinger, E., R. Lurz, M. Turmaine, L. Mangiarini, B. Hollenbach, R. Hasenbank, G. P. Bates, S. W. Davies, 1117 H. Lehrach and E. E. Wanker (1997). "Huntingtin-encoded polyglutamine expansions form amyloid-1118 like protein aggregates in vitro and in vivo." Cell 90(3): 549-558. 1119 Schilling, G., A. Klevytska, A. T. Tebbenkamp, K. Juenemann, J. Cooper, V. Gonzales, H. Slunt, M. Poirer, C. 1120 A. Ross and D. R. Borchelt (2007). "Characterization of huntingtin pathologic fragments in human 1121 Huntington disease, transgenic mice, and cell models." J Neuropathol Exp Neurol 66(4): 313-320. 1122 Smith, P. K., R. I. Krohn, G. T. Hermanson, A. K. Mallia, F. H. Gartner, M. D. Provenzano, E. K. Fujimoto, N. 1123 M. Goeke, B. J. Olson and D. C. Klenk (1985). "Measurement of protein using bicinchoninic acid." 1124 Anal Biochem 150(1): 76-85. 1125 Sotrel, A., P. A. Paskevich, D. K. Kiely, E. D. Bird, R. S. Williams and R. H. Myers (1991). "Morphometric 1126 analysis of the prefrontal cortex in Huntington's disease." Neurology 41(7): 1117-1123. 1127 Stavrides, P., C. N. Goulbourne, J. Peddy, C. Huo, M. V. Rao, V. Khetarpal, D. M. Marchionini, R. A. Nixon 1128 and D. S. Yang (2024). "Oral mTOR inhibition in Q175 Huntington’s disease model mice ameliorates 1129 neuronal autophagy and mutant huntingtin clearance." bioRxiv. 1130 Sulzer, D., E. Mosharov, Z. Talloczy, F. A. Zucca, J. D. Simon and L. Zecca (2008). "Neuronal pigmented 1131 autophagic vacuoles: lipofuscin, neuromelanin, and ceroid as macroautophagic responses during 1132 aging and disease." J Neurochem 106(1): 24-36. 1133 Tellez-Nagel, I., A. B. Johnson and R. D. Terry (1974). "Studies on brain biopsies of patients with 1134 Huntington's chorea." J Neuropathol Exp Neurol 33(2): 308-332. 1135 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 38 The-Huntington's-Disease-Collaborative-Research-Group (1993). "A novel gene containing a trinucleotide 1136 repeat that is expanded and unstable on Huntington's disease chromosomes. ." Cell 72(6): 971-983. 1137 Tong, H., T. Yang, L. Liu, C. Li, Y. Sun, Q. Jia, Y. Qin, L. Chen, X. Zhao, G. Zhou, S. Yan, X. J. Li and S. Li (2023). 1138 "Aberrant splicing of mutant huntingtin in Huntington's disease knock -in pigs." Neurobiol Dis 187: 1139 106291. 1140 Valionyte, E., Y. Yang, S. A. Griffiths, A. T. Bone, E. R. Barrow, V. Sharma, B. Lu and S. Luo (2022). "The 1141 caspase-6-p62 axis modulates p62 droplets based autophagy in a dominant-negative manner." Cell 1142 Death Differ 29(6): 1211-1227. 1143 Vonsattel, J. P. (2008). "Huntington disease models and human neuropathology: similarities and 1144 differences." Acta Neuropathol 115(1): 55-69. 1145 Vonsattel, J. P., R. H. Myers, T. J. Stevens, R. J. Ferrante, E. D. Bird and E. P. Richardson, Jr. (1985). 1146 "Neuropathological classification of Huntington's disease." J Neuropathol Exp Neurol 44(6): 559-1147 577. 1148 Wang, Z. M. and H. A. Lashuel (2013). "Discovery of a novel aggregation domain in the huntingtin protein: 1149 implications for the mechanisms of Htt aggregation and toxicity." Angew Chem Int Ed Engl 52(2): 1150 562-567. 1151 Wellington, C. L., L. M. Ellerby, C. A. Gutekunst, D. Rogers, S. Warby, R. K. Graham, O. Loubser, J. van 1152 Raamsdonk, R. Singaraja, Y. Z. Yang, J. Gafni, D. Bredesen, S. M. Hersch, B. R. Leavitt, S. Roy, D. W. 1153 Nicholson and M. R. Hayden (2002). "Caspase cleavage of mutant huntingtin precedes 1154 neurodegeneration in Huntington's disease." J Neurosci 22(18): 7862-7872. 1155 Wold, M. S., J. Lim, V. Lachance, Z. Deng and Z. Yue (2016). "ULK1 -mediated phosphorylation of ATG14 1156 promotes autophagy and is impaired in Huntington's disease models." Mol Neurodegener 11(1): 1157 76. 1158 Yamamoto, A., M. L. Cremona and J. E. Rothman (2006). "Autophagy -mediated clearance of huntingtin 1159 aggregates triggered by the insulin-signaling pathway." J. Cell Biol. 172(5): 719-731. 1160 Yang, J., X. Chen and H. Xu (2021). "SQSTM1/p62 droplet -mediated autophagosome formation:insights 1161 into Huntington disease." Autophagy 17(10): 3256-3259. 1162 Yu, W. H., A. M. Cuervo, A. Kumar, C. M. Peterhoff, S. D. Schmidt, J.-H. Lee, P. S. Mohan, M. Mercken, M. 1163 R. Farmery, L. O. Tjernberg, Y. Jiang, K. Duff, Y. Uchiyama, J. Naslund, P. M. Mathews, A. M. Cataldo 1164 and R. A. Nixon (2005). "Macroautophagy--a novel -amyloid peptide-generating pathway activated 1165 in Alzheimer's disease." J Cell Biol 171(1): 87-98. 1166 Zhou, Y., T. R. Peskett, C. Landles, J. B. Warner, K. Sathasivam, E. J. Smith, S. Chen, R. Wetzel, H. A. Lashuel, 1167 G. P. Bates and H. R. Saibil (2021). "Correlative light and electron microscopy suggests that mutant 1168 huntingtin dysregulates the endolysosomal pathway in presymptomatic Huntington's disease." 1169 Acta Neuropathol Commun 9(1): 70. 1170 Zucchelli, S., F. Marcuzzi, M. Codrich, E. Agostoni, S. Vilotti, M. Biagioli, M. Pinto, A. Carnemolla, C. Santoro, 1171 S. Gustincich and F. Persichetti (2011). "Tumor necrosis factor receptor-associated factor 6 (TRAF6) 1172 associates with huntingtin protein and promotes its atypical ubiquitination to enhance aggregate 1173 formation." J Biol Chem 286(28): 25108-25117. 1174 1175 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint B5a B5b 20 um p62mHTT Ub mHTT Ub 500 nmp62 Ub mHTT Ub CTXSTR 10 um Ub 10 ump62 p62 Fig. 1. HTT inclusions: types, distributions and close relationship with autophagy adaptors Ub and p62. A1 B1 B2a B3 B2b A2 A3 HD2 HD4 A: Neuronal intranuclear inclusions (NIIs) B: Neuritic inclusions C: Cytoplasmic inclusions C1 B4 STR STR CTX CTX CTX mHTT Ub 10 um CTX STR HD3 HD3 C2 500 nm HTT IEM STR HD4 HD4 HD4 HD4 HD4 HD4 CTX HD4 500 nm Ub IEM CTX HD4 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Fig. 2. Protein levels of HTT and autophagy adaptor ubiquitin, p62 and TRAF6 in the STR and the CTX. was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Ctl HD2 HD3 HD4 STR CTX Ctl HD2 HD4 Ctl HD4 Fig. 3. HD brains develop ALP pathology in the later stages of the disease progression, as revealed by immunostaining of CTSD. STR A B C D E F G H I HD4Ctrl J was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Fig. 4. Association of mHTT signal with CTSD IR during disease progression. Ctl HD2 HD4 CTSD/mHTT STR CTSD mHTT Ub 500 nm p62 HD4 A B C D E 20 m IEM 20 m was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Recognizable AVs/lipofuscin in cytoplasm Str Lysophagy A1 A2 Fig. 5. Membranous/vesicular pathologies in the later stages of HD – Accumulation of vesicles with various types of intraluminal features. STR STR “Empty” vacuoles and abnormal Mito STR CTX HD4 HD4 HD4 HD4 Recognizable AVs in dystrophic neurites B1 B2 500 nm M C1 C2 M M STR STR HD4 HD4 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint % Rel to Ctl Rel to Ctl (%) Fig. 6. Levels of proteins involved in the lysosomal clearance phase of the ALP and enzymatic activity assays of cathepsins in the STR and the CTX. A2A1 B STR Rel to Ctl (%) CTX C2C1 D was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Fig. 7 qPCR of selected ALP-related targets in HD brains. 0 100 200 300 mRNA Expression of Select Markers in the STR and the CTX of HD Rel Ct vs ACTB (% Rel to Ctl) TFEB LC3 LAMP1 LAMP2 CTSD CTSB HEXA p62/SQSTM1 *** P=0.07 P=0.08 STR CTX was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Fig. S1 (related to Fig. 1). Ub IHC HD3, STR b. a. Neuron and glia from a control human brain HD2 p62mHTT Ubc. was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 500 nm d. Fig. 1C1 Enlarged Fig. 1C2 Enlarged HTT IEM Fig. S1 (related to Fig. 1) (Cont’d). was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint 15 16 17 18 19 20 21 22 23 24 25 26 27 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 CTX STR kDa --250 --250 --100 -- 75 -- 50 -- 37 -- 25 -- 20 -- 15 -- 10 --Comb Well Fig. S2 (related to Fig. 2). mHTT signal is only detected in the stacking gel but not below by antibody MW8. was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Fig. S3 (related to Fig. 5C). Membrane pathology in AD brain. was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint Table 1. Demographics of human brain samples utilized for qPCR for mRNA analysis (qPCR), western blotting (WB), cathepsin enzymatic activity assay (Enzyme), immunohistochemistry (IHC) and electronic microscopy (EM). Case Number HD Stage PMI (hr) Age (y) Brain Banks Analysis Performed (regions used) qPCR WB Enzyme IHC EM STR CTX STR CTX STR CTX STR CTX STR CTX B7010 Control 18.9 64 HBTRC X X B7360 Control 38.1 51 HBTRC X X B7970 Control 25.8 70 HBTRC X X 6916 Control Na Na HBTRC X 6919 Control Na Na HBTRC X X 8341 Control 15.7 82 HBTRC X X X X X X 8176 Control 25.3 86 HBTRC X X X X X X 8108 Control 31.6 70 HBTRC X X X X X X AN01404 Control 18.6 74 HBTRC X X AN01614 Control 18.7 80 HBTRC X X AN03398 Control 12.1 75 HBTRC X X AN07810 Control 18.1 65 HBTRC X X AN11537 Control 12.5 60 HBTRC X X AN12699 Control 11.0 55 HBTRC X X E04-32 Control 70 ECND X E04-34 Control 17 57 ECND X X X X X E04-46 Control 31 40 ECND X X X X E05-74 Control 6 59 ECND X X X X X X E06-41 Control 10 57 ECND X X X X X X X E06-45 Control 6.5 46 ECND X X X E06-114 Control 6.5 53 ECND X X X X X X X E09-170 Control 14.5 88 ECND X E11-33 Control 15 43 ECND X X X X OS03-299 Control 6.0 69 ECND X OS03-380 Control 12 61 ECND X X X X OS03-390 Control 7 74 ECND X X T-272 Control Na Na NYBBC X T-346 Control 10 84 NYBBC X X Avg/Tot 16.2+8.8 65.3+13.4 E05-119 HD2 11 56 ECND X X X X X X X X E05-154 HD2 33.0 67 ECND X E10-05 HD2 8.5 62 ECND X X X X X X X X OS01-04 HD2 24 51 ECND X X X X X X X X X OS01-12 HD2 na 73 ECND X OS01-114 HD2 2.5 68 ECND X X X X X X X X T-197 HD2 2.9 78 NYBBC X T-264 HD2 7.5 69 NYBBC X T-272 HD2 17 83 NYBBC X X X T-289 HD2 24.5 54 NYBBC X T-309 HD2 4 80 NYBBC X T-550 HD2 82 NYBBC X X X X X X T1146 HD2 39 69 NYBBC X X X X X Avg/Tot 15.8+12.7 68.6+10.6 B8001 HD3 18.3 47 HBTRC X X X B8007 HD3 22.6 65 HBTRC X X X X X B7901 HD3 23.3 73 HBTRC X 7684 HD3 22.7 61 HBTRC X X 7939 HD3 22.6 70 HBTRC X X 7989 HD3 24 52 HBTRC X X X X 8000 HD3 15.6 59 HBTRC X X X X 8232 HD3 15.3 58 HBTRC X X X X X X 8234 HD3 20.8 68 HBTRC X X X X X X 8268 HD3 19.9 55 HBTRC X X X X X X E05-24 HD3 18.5 63 ECND X X X X X X X X X X E05-38 HD3 12.5 70 ECND X E07-188 HD3 12.5 63 ECND X X X X X X X X X X E09-06 HD3 55 ECND X X X X OS00-03 HD3 6.5 54 ECND X X X X X X OS00-09 HD3 96 64 ECND X OS01-88 HD3 5 43 ECND X X X OS99-17 HD3 8 83 ECND X X X X X X X OS99-19 HD3 6.5 67 ECND X T-4276 HD3 25 59 NYBBC X X X X X X T4584 HD3 Na Na NYBBC X T-4830 HD3 24 63 NYBBC X X Avg/Tot 21.0+18.8 61.5+9.1 was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint B7624 HD4 24.1 50 HBTRC X X X B7735 HD4 23.1 60 HBTRC X X X B7822 HD4 29.1 54 HBTRC X X X X X 7319 HD4 13.5 76 HBTRC X X 7681 HD4 26 74 HBTRC X X 7684 HD4 Na Na HBTRC X 7791 HD4 Na Na HBTRC X 7792 HD4 13.0 42 HBTRC X X X X 7822 HD4 29.1 54 HBTRC X X X 7939 HD4 24 52 HBTRC X 7989 HD4 Na Na HBTRC X 7991 HD4 21.8 71 HBTRC X X X X X X X X 8083 HD4 20.1 52 HBTRC X X X X X X 8093 HD4 24.2 51 HBTRC X X X X X X 8150 HD4 21.1 57 HBTRC X X X X X X 8207 HD4 29.1 80 HBTRC X X X X X X T-128 HD4 3.9 50 NYBBC X T-141 HD4 6.4 77 NYBBC X T-225 HD4 17.9 72 NYBBC X T-295 HD4 1.5 46 NYBBC X T-329 HD4 0.3 54 NYBBC X T-4584 HD4 Na Na NYBBC X T-4817 HD4 31.5 42 NYBBC X X X X T-5017 HD4 12.0 66 NYBBC X X OS01-03 HD4 5 58 ECND X X X X X X X Autopsy 49 HD4 Na Na MSNBBRC 1719 HD4 Na Na MSNBBRC X X Avg/Tot 17.9+9.9 58.6+12.0 Abbreviations: Harvard Brain Tissue Resource Center, HBTRC; Emory Center for Neurodegenerative Disease, ECND; New York Brain Bank at Columbia, NYBBC; Mount Sinai Neuropathology Brain Bank & Research CoRE, MSNBBRC; Na, Not Available; Str, Striatum; Ctx, Cortex, BA9/10; Cbm, Cerebellum. was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (whichthis version posted May 30, 2024. ; https://doi.org/10.1101/2024.05.29.596470doi: bioRxiv preprint