Transmission-selective muscle pathology induced by active propagation of mutant huntingtin across the human neuromuscular synapse

11 A potential explanation for the spatiotemporal accumulation of pathological lesions in the brain of 12 patients with neurodegenerative protein misfolding diseases (PMDs) is cell-to-cell transmission 13 of aggregation-prone, misfolded proteins. Little is known about central to peripheral transmission 14 and its contribution to pathology. We show that transmission of Huntington’s disease(HD-) 15 associated mutant HTT exon 1 (mHTTEx1) occurs across the neuromuscular junctions in human 16 iPSC cultures and in vivo in wild-type mice. We found that transmission is an active and dynamic 17 process, that happens prior to aggregate formation and is regulated by synaptic activity. 18 Furthermore, we find that transmitted mHTTEx1 causes HD-relevant pathology at a molecular 19 and functional level in human muscle cells, even in the presence of ubiquitous expression 20 mHTTEx1. With this work we uncover a casual-link between mHTTEx1 synaptic transmission and 21 pathology, highlighting the therapeutic potential in blocking toxic protein transmission in PMDs. 22 23 Introduction 24 Neurodegenerative protein misfolding diseases (PMDs) are a group of unrelated illnesses, 25 including Alzheimer’s(AD), Parkinson’s-(PD), Huntington’s disease (HD), Amyotrophic lateral 26 sclerosis (ALS) and frontotemporal lobar dementia (FTLD). They are all characterized by 27 misfolding and aggregation of a disease-specific protein, cell-type specific vulnerability to 28 degeneration and progressive loss of structure and function of the nervous system. The disease 29 process is already active for years, prior to revealing itself mostly around mid-age with initially 30 discrete neurobehavioral and neuropsychiatric symptoms, which progressively worsen into 31 cognitive impairment. Currently no therapies are available to cure or at least slow down the 32 progression of these devastating illnesses. 33 It has been suggested that intra brain transmission of the toxic misfolded protein species 34 might be a potential explanation for the spatiotemporal propagation of the pathological lesions 35 through the brain. A casual-link between transcellular spreading of misfolded prion proteins (PrP 36 scrapie or PrPSc) and pathology has been demonstrated in prion diseases. For the other 37 neurodegenerative PMDs it has been by now firmly demonstrated that tau (AD), α-synuclein (PD), 38 mutant huntingtin (HD) and tdp-43 (ALS, FTLD) are transmitted between cells and functional 39 connected brain regions (for review see). This transmission is accompanied by the appearance 40 of protein aggregates in the acceptor cells. Furthermore, a decline of cognitive and motor behavior 41 has been associated with regionor cell-type specific misfolded protein-expression . 42


24
Neurodegenerative protein misfolding diseases (PMDs) are a group of unrelated illnesses, 25 including Alzheimer's-(AD), Parkinson's-(PD), Huntington's disease (HD), Amyotrophic lateral 26 sclerosis (ALS) and frontotemporal lobar dementia (FTLD). They are all characterized by 27 misfolding and aggregation of a disease-specific protein, cell-type specific vulnerability to 28 degeneration and progressive loss of structure and function of the nervous system. The disease 29 process is already active for years, prior to revealing itself mostly around mid-age with initially 30 discrete neurobehavioral and neuropsychiatric symptoms, which progressively worsen into 31 cognitive impairment 1 . Currently no therapies are available to cure or at least slow down the 32 progression of these devastating illnesses. 33 It has been suggested that intra brain transmission of the toxic misfolded protein species 34 might be a potential explanation for the spatiotemporal propagation of the pathological lesions 35 through the brain 2 . A casual-link between transcellular spreading of misfolded prion proteins (PrP 36 scrapie or PrPSc) and pathology has been demonstrated in prion diseases 3,4 . For the other 37 neurodegenerative PMDs it has been by now firmly demonstrated that tau (AD), α-synuclein (PD), 38 mutant huntingtin (HD) and tdp-43 (ALS, FTLD) are transmitted between cells and functional 39 connected brain regions (for review see 5-7 ). This transmission is accompanied by the appearance 40 of protein aggregates in the acceptor cells. Furthermore, a decline of cognitive and motor behavior 41 has been associated with region-or cell-type specific misfolded protein-expression 8-12 . 42 Various cellular mechanism responsible for cell-to-cell transmission of misfolded proteins 43 have been proposed 13 . It has been shown that transneuronal transmission of A-β and tau is 44 enhanced by neuronal activity and synaptic connectivity and that preventing synaptic vesicle 45 release reduces the transmission of mutant huntingtin (mHTT) [14][15][16][17][18] . Together, with the 46 observations that tau, α-synuclein (α-syn), mHTT and tdp-43 are transmitted between functional 47 connected brain regions in vivo in mice and drosophila, this strongly suggests a transsynaptic 48 transmission pathway of misfolded proteins [8][9][10][11]14,17,[19][20][21] . Synaptic connections are not only present 49 in the central nervous system CNS, but also allow transcellular communication between the CNS 50 and the periphery, as for example the neuromuscular junction (NMJ) between spinal motor 51 neurons and skeletal muscles. mHTT expressed in either the skeletal muscle or brain in 52 Caenorhabditis (C.) elegans has been shown to travel between the CNS and skeletal muscles 22 . 53 Thus, transmission of misfolded proteins could represent a systemic disease pathway affecting 54 not only the CNS, but also contributing to a progressive deterioration of peripheral systems. 55 Patients with HD suffer from a decline in skeletal muscle function, which progressively 56 worsen with disease course 23,24 . HD is an autosomal dominant disorder that develops with 57 hundred percent penetrance when the number of CAG triplets in the HTT gene exceeds 35 58 repeats. This repeat is translated into a pathogenic polyglutamine stretch in the exon1 of the HTT 59 protein 25 . Incomplete mRNA splicing of the mHTT results in a toxic exon 1 fragment of the protein, 60 which is highly prone to aggregation and aberrantly translocates to the nucleus where it interferes 61 with transcription [26][27][28][29] . Human neuronal cell lines overexpressing the HTT exon 1 (HTTEx1) 62 develop intra-nuclear inclusions and mitochondrial dysfunction 30 . These pathologies are also 63 observed in skeletal muscles of HD patients and animal models, together with skeletal muscle 64 wasting and fatigue 24,31-36 . 65 Using an isogenic human induced pluripotent stem cell-(hiPSC-) neuromuscular (NM) 66 model combined with high-throughput live-cell imaging, functional analysis and microfluidic 67 systems, we addressed whether mHTTEx1 cell-to-cell transmission can occur across the NMJ 68 and how synaptic density and activity could influence this process. We also examined whether 69 mHTT NM transmission can contribute to skeletal muscle pathology including conditions 70 resembling ubiquitous expression of mHTTEx1. We show that mHTTEx1 is transmitted from 71 neurons to myotubes across the NMJ and that transmission is elevated by increased NMJ density 72 and modulated by neuronal activity. Moreover, our data reveal that transmission occurs 73 independent of mHTTEx1 aggregation, already during NMJs assembly and is enhanced during 74 their functional maturation. Furthermore, our data discloses that mHTTEx1 transmission results 75 in fragmented mitochondria, increased intra-nuclear aggregates and a functional decline of 76 myotube contractibility. Importantly, these pathologies are enhanced or specifically induced by 77 transmission in the presence of cell autonomous mHTTEx1 in myotubes. Finally, we show that 78 mHTTEx1 expressed specifically in the pyramidal neurons in the M1 motor cortex in vivo in mice 79 is transmitted to the spinal motor neurons and triceps muscles. Our findings therefore suggest 80 that mHTTEx1 cell-to-cell transmission occurs between the central nervous system and the 81 periphery and might contribute to pathological alterations of the NM system already at early, 82 preclinical stages of the disease. More broadly, our findings also support the notion that cell-type 83 specific vulnerability might be determined by the level of functional synaptic connectivity in 84 combination with trans-synaptic transmission of the misfolded proteins.

90
Characterization of an in vitro hiPSC-derived NM co-culture to study transmission of 91 pathogenic HTT. 92 To assess whether mHTT transmission can contribute to skeletal muscle pathology in HD 93 patients, we designed an in vitro, isogenic hiPSC-derived NM co-culture system, using two 94 transgenic cell lines, one bearing a doxycycline-(dox)-inducible pro-neuronal transcription factor, 95 neurogenin 2 (Ngn2) transgene 37 and one bearing a Dox-inducible pro-skeletal muscle 96 transcription factor, myoblast determination protein 1 (MyoD) transgene. We generated four 97 isogenic hiPSC Ngn2 and one hiPSC iMyoD line, isogenic to the Ngn2 lines, to establish two NM-98 co-culture systems: 1) a Ngn2 line expressing the exon1 of the HTT gene, with 72 (pathogenic) 99 triplets encoding for glutamine, fused to a Cre sequence without the additional nuclear localization 100 signal (iNgn2;HTTEx1Q72-cre) and a MyoD hiPSC line with a LoxP-GFP construct (iMyoD;LoxP-101 GFP; Supplemental Fig. 1a). This system can be used for high-throughput, low-resolution, live-102 cell imaging to follow transmission over weeks (Fig. 1a); 2) a HTTEx1Q72 fused to a mCherry 103 (iNgn2;HTTEx1Q72-mCherry; Supplemental Fig. 1a). The mCherry-tag allows to follow the 104 HTTEx1Q72 transmission to myotubes quantitatively over time. Once a LoxP-GFP myotube turns 105 green, further transmission of HTTEx1Q72 cannot be visualized. In contrast, a change in mCherry 106 labelling reveals the dynamics of this process and allows to correlate the amount of transmitted 107 protein with the pathology. Importantly, using HTTEx1Q72 fused to two different tags we verify 108 that transmission is independent of the tag. 109 We assessed transgene expression with western blot (WB) analysis using anti-HTT exon 110 1 antibody, which revealed that the HTTEx1Q72-cre line expressed the lowest and that we had 111 iPSC clones with different levels of HTTEx1Q72-mCherry expression (Supplemental Fig. 1b). To 112 assess whether the different protein expression levels result in distinct propensity to aggregation 113 we differentiated the hiPSC lines into neurons. At day 1 of differentiation we observed loss of 114 pluripotency with decrease in Oct4 expression (Supplemental Fig. 1c). After 7 and 21 days of 115 differentiation we assessed mHTT aggregates with Em48 antibody which has high affinity for the 116 aggregated form. Aggregation was lowest in HTTEx1Q72-cre line and increased with increasing 117 expression level of the HTTEx1Q72 protein in HTTEx1Q72-mCherry clones (Supplemental Fig.  118 1d). 119 To test the cre-lox system, we electroporated iMyoD;LoxP-GFP hiPSCs with the HTTEx1-120 cre construct and the iNgn2;HTTEx1Q72-cre hiPCSs with a FloxP-mCherry plasmid. This resulted 121 in GFP and mCherry expressing cells, resp. In the absence of Cre we never observed GFP 122 expression in the iMyoD;LoxP-GFP hiPSCs (n = 3; Supplemental Fig. 1e). Altogether, these 123 analyses demonstrate a successful generation of four hiPSC-lines, which can be used to study 124 cell-to-cell transmission of mHTTEx1.

128
In the next step we established NM co-cultures using a two-step differentiation protocol 129 (Supplemental Fig. 2a). A molecular maturation of the myotubes and neurons in co-cultures was 130 assessed by WB at DCC 1, 7, 14, 21 and 28 using myotube-specific antibodies (myosin heavy 131 chain embryonic and postnatal isoform (MHC3 and MHC8, resp.) and neuronal (doublecortin (DC; 132 4 neuronal precursor marker) and motor neuron (Islet 1 and choline acetyl transferase (ChaT)), 133 specific antibodies. With increasing co-culture time, we found a decrease in the precursor markers 134 and an increase in the postnatal markers (Fig. 1b). This demonstrates a molecular maturation of 135 the two cell types, hereafter referred to as Neu HTTEx1Q72-cre for the neurons and Myo LoxP-136 GFP for the myotubes. Immunofluorescence antibody staining (IF) further revealed that NMJs are 137 formed between Neu HTTEx1Q72-cre and Myo LoxP-GFP. We observed close appositions of the 138 neuronal presynaptic active zone marker Bassoon (BSN) and the acetylcholine receptor (AChR) 139 marker α-bungarotoxin (α-BgTx) on myotubes (which represent the postsynaptic structure of the 140 NMJ; Fig. 1c). Patch-clamp recordings from Neu HTTEx1Q72-cre revealed functional maturation 141 of a current-induced action potential firing pattern from a mixed phasic / adaptive at DCC7 to a 142 mainly tonic pattern at DCC 21 (Fig. 1d, Supplemental Fig. 2b). We further demonstrated that 143 myotube contractions disappeared upon addition of the NMJ-activity blocker α-BgTx at DCC21 144 (Supplemental Fig 2c). These data demonstrate the establishment of functional NMJs between 145 Neu HTTEx1Q72-cre and Myo LoxP-GFP. To gain better insight into the temporal development 146 of these NMJs we followed myotube contractions for 29 days in the same wells of either co-147 cultures or monocultures of Myo LoxP-GFP. This revealed a temporal increase in both myotube 148 activity and contracting area only in co-cultures (Fig. 1e). In addition, we analyzed the variability 149 of these parameters within one culture well. The variability significantly decreased from DCC 15 150 onwards in co-cultures, but stayed high in monocultures (Supplemental Fig. 2d), indicative of 151 triggered neuron-induced contractions in co-cultures. Concordantly, DCC15 is also a time of steep 152 increase in myotube contracting area and of neuronal maturation based on a more active AP firing 153 pattern (DCC 14 to 21, tonic firing from 14% to 80%) (Fig. 1d, e). These data together demonstrate 154 the establishment of functional NMJs between Neu HTTEx1Q72-cre and Myo LoxP-GFP in this 155 NM-co-culture system and validates it for addressing the question whether HTTEx1Q72-cre can 156 be transmitted from motor neurons to myotubes across functional NMJs.

158
Neuromuscular transmission of HTTEx1Q72-cre occurs with time and in the absence of 159 aggregates in the neurons. 160 To assess HTTEx1Q72-cre transmission, we performed high-throughput live-cell fluorescent 161 imaging from the same wells from DCC 4 to 28. At DCC 4 first GFP+ myotubes appeared and 162 their number increased with co-culture time until day 21, after that the number stayed stable ( Fig.  163 1f, g). This timing correlated with establishment of functional NMJs (Fig. 1d, e). Based on Em48 164 staining at DCC 7 and 21 we did not observe aggregated form of HTT in HTTEx1Q72-cre neurons 165 (Supplemental Fig. 1d). When we stained the co-culture at DDC 28 we detected presence of few 166 Em48 positive aggregates selectively in GFP+ myotubes (Fig. 1h). Transmission thus likely 167 occurs in a non-aggregated form and aggregation takes place in the myotubes. 168 To prove that HTTEx1Q72-cre NM transmission requires direct cell-cell contact and is not 169 transferred via the culture media, we placed a two-chamber cell culture insert w/o bottom in one 170 dish to allow physical separation of Neu HTTEx1Q72-cre from Myo LoxP-GFP, while the medium 171 was shared. After attachment of the cells, the inserts were removed (DCC1). The surface between 172 the inserts was not coated to prevent the movement of the cells and extension of the axons to the 173 myotubes. In these co-cultures we never observed GFP+ myotubes (Supplemental Fig. 3a, b).

174
Taken together, the NM co-culture system allows to follow pathogenic HTTEx1 cell-to-cell 175 transmission over weeks with high-throughput low-resolution live-cell imaging. Furthermore, with 176 the expression of GFP in Myo LoxP-GFP we demonstrated that HTTEx1Q72-cre is transmitted 177 from neurons to muscles and can enter the cytosol and the nucleus of the myotubes. 178 179 NM-co-cultures in microfluidic devices reveal HTTEx1Q72 transmission across the NMJs. 180 To demonstrate that HTTEx1Q72 is transmitted across the NMJ and assess whether these 181 structures play a role in determining the efficiency of pathogenic HTT transmission we established 182 NM co-cultures in microfluidic devices (MFDs). These devices allow to co-culture two cell 183 populations in two isolated compartments, connected with microgrooves through which axons can 184 grow and reach the other compartment, allowing them to form connections with myotubes ( Fig.  185 2a, upper panel). To quantify the HTTEx1Q72 NM transmission we used here the co-cultures of 186 Neu HTTEx1Q72-mCherry clone#75 with Myo LoxP-GFP (we kept using this myotube line, but to 187 prevent confusion we will refer to it as Myo when we used it in co-culture with Neu HTTEx1Q72-188 mCherry). The mCherry labeling of the HTTEx1Q72-mCherry expressing neurons showed that 189 these projected their axons from the presynaptic neuronal compartment to the postsynaptic 190 myotube compartment (Fig. 2a, lower panel). In the myotube compartment NMJs were 191 established, as visualized with IF staining's of BSN and AChR appositions at DCC21 (Fig. 2b).

192
AChR clusters on the surface of myotubes, can be classified based on their shapes 38 . We 193 performed a detailed shape analysis of these clusters at DCC21 and classified them into four 194 categories: small&elongated, small&round, big&elongated, big&round (Supplemetal Fig. 4a-c).

195
When we compared the clusters of myotubes in mono versus co-cultures we found that the 196 clusters of small&round type were the majority in both cultures but, there was a significant 197 increase in the density of both types of big clusters in the co-cultures (Supplemental Fig. 4d, e).

198
The big-cluster types are thus likely those constituting the NMJs. Supporting this notion, we found 199 that big cluster types were over-represented among those associated with the presynaptic marker 200 BSN, compared to the clusters w/o BSN (Fig. 2c). With these analyses we demonstrate the 201 presence of structural NMJs in the co-cultures grown in MFDs. To judge whether Neu 202 HTTEx1Q72-mCherry clone#75 neurons are able to potentially trigger myotube contractions we 203 performed patch-clamp recordings and found that these cells develop from DCC 7 to 21 a more 204 active firing pattern of current-induced action potential from mainly phasic at DCC 7 to mainly 205 adaptive at DCC 21 ( Fig. 2d). 206 Next, using IF labeling, we revealed that HTTEx1Q72-mCherry protein is present in the 207 myotubes in the postsynaptic compartment, indicating transmission from the neurons (Fig. 2e). 208 Similar as for the HTTEx1Q72-cre, we observed a continuous transmission of HTTEx1Q72-209 mCherry to Myo, increasing from DCC 7 to DCC 21 quantified as number of HTTEx1Q72-mCherry 210 puncta in the myotubes (Fig. 2f). We analyzed the volume of HTTEx1Q72-mCherry puncta over 211 time and found that there is a dynamic change of a size distribution (Fig. 2g). At DCC 14 we saw 212 appearance of larger aggregates with sizes above 15 µm 3 and loss of major contribution of smaller 213 assemblies comparing to DCC7. At DDC 21 there was again a major contribution of small 214 assemblies with sizes below 1 µm 3 , together with a persistent presence of bigger aggregates. 215 This suggests that new molecules arrive into the muscle as small assemblies and that aggregation 216 occurs over time. Together, with these experiments we provide evidence that HTTEx1Q72 is 217 transmitted across the NMJ, most probably in a form of small protein complexes. 218 219 220 6 The load of HTTEx1Q72-mCherry correlates with increasing NM connections. 221 One of the presymptomatic pathologies in HD patients is a loss of functional neuronal connectivity, 222 that first arises in the cortico-striatal pathway and then progresses to cortical and other subcortical 223 brain regions [39][40][41][42][43] . Interestingly, the most vulnerable brain regions form a selective network with 224 higher connectivity than other brain regions 44 -a so called 'rich club'. To assess whether a higher 225 density of NMJ connections leads to more HTTEx1Q72-mCherry puncta in myotubes, we divided 226 the postsynaptic compartment into 3 bins, starting with bin 1 closest to the microgrooves. The 227 area of neuronal processes was highest in bin 1 and decreased towards bin 3 ( Fig. 3a, b, 228 Supplemental Fig. 5a). Similarly, we found that the number of HTTEx1Q72-mCherry puncta in 229 myotubes at DCC21 was highest in bin 1 and steeply decreased towards bin 3 (Fig. 3a, c). 230 Interestingly, when we compared the distribution in bins at all time points we observed, a delayed 231 increase in the number of puncta in bin 2 compared to bin 1 (Supplemental Fig 5b). Axons will 232 arrive slightly later in bin 2 than bin 1, followed by delayed formation of NMJs, explaining the 233 temporal delay in HTTEx1Q72-mCherry accumulation in myotubes. Further validating our 234 assumption that HTTEx1Q72-mCherry proteins reach the myotubes via the NMJs, we found a 235 positive correlation between the density of NMJs (BSN-BgTx complexes) and the number of 236 HTTEx1Q72-mCherry puncta in myotubes (Fig. 3d, e).

237
Previously we showed that mHTT is transmitted from mouse cells in HD-derived mouse 238 organotypic brain slices (OTBS) to human stem cell-derived neurons (h-neurons). During time of 239 transmission mHTT co-localized with the presynaptic marker synaptophysin and post-synaptic 240 density protein-95 (PSD-95) in neurons 17 . In the NM-co-cultures we also observed around 20% 241 of HTTEx1Q72-mCherry puncta to be associated with the postsynaptic AChRs (Supplemental 242 Fig. 4b). Interestingly, around 60% of the AChR clusters associated with HTTEx1Q72-mCherry 243 puncta were of the big-type, while among those w/o HTTEx1Q72-mCherry only around 10 % were 244 big (Fig. 3f). The big clusters are likely representing those incorporated in the NMJs, since this 245 type increased in the presence of neurons and also in association with the presynaptic marker 246 BSN (Supplemental Fig. 4e, Fig. 2c). Summarizing, we observe a positive correlation of NM-247 connectivity with HTTEx1Q72-mCherry load in myotubes and preferential association of 248 HTTEx1Q72-mCherry with NMJ-forming AChR clusters.

250
Modulating neuronal activity alters NM transmission of mHTTEx1 251 Neuron-to-neuron transmission of mHTT in the OTBS -h-neuron co-cultures and also in vivo in 252 drosophila has been shown to be vastly blocked by preventing SNARE-dependent fusion of 253 synaptic vesicles to the presynaptic membrane and subsequent release of its content 14,17 . 254 Therefore, we applied the SNARE-cleaving Tetanus neurotoxin (TeNT) 45 at DCC 10 in 255 HTTEx1Q72-mCherry clone#75 Myo co-cultures. Similar to neuron-to-neuron transmission we 256 observed a significant decrease in HTTEx1Q72-mCherry NM transmission, measured by number 257 of mCherry foci within the myotubes at DCC 21 ( Fig. 4a,b). Interestingly, the proportions of AChR 258 cluster types were not affected (Fig.4c), confirming that the observed effect is due to the blocking 259 of the presynaptic neuronal terminals and not due to a re-organization of postsynaptic structures. 260 We hypothesized that synaptic transmission of mHTTEx1 can be regarded as a clearance 261 mechanism by which cells get rid of toxic protein species. We therefore addressed if blocking of 262 this rescue process by the TeNT treatment increased pathological consequences in neurons.

263
Previously, we showed that mHTTEx1 transmitted from mouse cells to human stem cell-derived 264 neurons, first appeared as cytoplasmic aggregates and with time aggregates appeared in the 265 nucleus. Nuclear aggregation correlated with the time when pathological changes occurred in the 266 human neurons 17 . Indeed, in our co-culture system TeNT treatment resulted in increase of the 267 number and size of neuronal intra-nuclear assemblies (Fig. 4d,e). 268 As inhibition of synaptic vesicle release reduced HTTEx1Q72-mCherry transmission, we 269 asked if opposite effect can be obtained by depolarization of neurons, triggering increased AP 270 firing and synaptic vesicle release 46 . We exposed the HTTEx1Q72-mCherry clone#72 Myo co-271 culture at DCC 21 for 10 minutes to 10mM KCl, followed by 2 hours in 2.5mM KCl. We chose to 272 use the clone#72 for this experiment, because of its lower expression of HTTEX1Q72-mCherry, 273 which provides a larger range for increase in transmission without a risk of system saturation.

274
Upon exposure to 10mM KCl we observed more HTTEx1Q72-mCherry puncta in myotubes 275 compared to control co-cultures exposed to 2.5mM KCl (Fig.4f). This increase was not associated 276 with a re-organization of the postsynaptic structures, as AChR proportions did not change after 277 KCl treatment (Fig 4g). To further exclude influence of muscle depolarization on this effect we 278 repeated the experiment in MFDs and applied the 10mM KCl only to the pre-synaptic side ( Fig.4h  279 upper panel). We again observed increase in HTTEx1Q72-mCherry puncta in myotubes upon 280 treatment ( Fig.4h lower panel, i). When we compared the volume of these HTTEx1 Q72-mCherry 281 puncta, we observed larger contribution of small assemblies in KCl treated compared to non-282 treated co-cultures, supporting transmission in a form of small protein complexes (Fig. 4j). 283 The above results of two opposite synaptic manipulations demonstrate that synaptic 284 activity regulates HTTEx1Q72-mCherry transmission and that the pathway of transmission is 285 coupled with synaptic vesicle release. 286 287 HTTEx1Q72-mCherry aggregates accumulate at the myosin surface.

288
Our results so far demonstrate a trans-NM pathway of HTTEx1Q72 transmission. To reveal a 289 potential pathology triggered by transmitted HTTEx1Q72, we assessed the intracellular 290 localization of this protein in the myotubes from DCC 7 to 21. In particular, we analyzed the 291 localization of the HTTEx1Q72-mCherry puncta at the cellular surface. We used the surface 292 function of the Imaris software (Oxford Instruments) to define the surface of myotubes based on 293 MHC1 staining. By visual inspection of images, we observed a striking localization of the puncta 294 to and partially passing through the MHC1+ myotube surface (Fig. 5a, b). Based on a 295 quantification, at DCC 7 half of the puncta localized at the myotube surface (at the surface: 0 -296 0.05 µm to myotube surface) and half inside the myotube (inside: > 0.05 µm to myotube surface) 297 and were mostly small (majority below 4 µm 3 ) (Fig. 5c, d). At DCC 15 even more -79% of 298 HTTEx1Q72-mCherry puncta accumulated at the surface, at DCC 21 this shifted back to 52% 299 (Fig. 5c, d). Furthermore, as we found before, we observed growing number of HTTEx1Q72-300 mCherry puncta with larger volume (Fig. 5c). It has been shown that mHTT has high affinity for 301 bioengineered lipid membranes and that insertion of these proteins into these membranes triggers 302 their aggregation 47 . Our analysis also revealed that the largest HTTEx1Q72-mCherry puncta 303 (volume > 5 µm 3 ) were nearly all (90-100%) localized at the myotube surface at DCC 7-21 ( Fig.  304 5c, e). 305 306 HTTEx1Q72-mCherry transmission induces and aggravates pathological alterations in 307 myotubes. 308 8 An important open question in the field of misfolded proteins is whether the transmission can 309 trigger or aggravate the pathology caused by the cell-autonomous presence of the toxic protein.

310
Addressing this critical question will reveal whether toxic protein transmission is a novel disease 311 pathway in neurodegenerative PMDs. Therefore, we assessed HD-specific pathological 312 alterations in myotubes in the following Neu Myo co-culture combinations: 1) Neu control (ctr)/Myo 313 ctr (no expression of the pathogenic HTTEx1Q72), 2) Neu ctr/Myo HTTEx1Q72-mCherry (cell-314 autonomous), 3) Neu HTTEx1Q72-mCherry/Myo ctr (transmission) and 4) Neu HTTEx1Q72-315 mCherry/Myo HTTEx1Q72-mCherry (transmission + cell autonomous). 316 Mitochondrial dysfunction is a characteristic observed in skeletal muscle obtained from 317 HD patients and animal models 24 . Typically, a disbalance in the fission and fusion events occur 318 which lead to more fragmented structures and a reduced filamentous network 34,48 . To assess the 319 effect of cell-autonomous and transmitted HTTEx1Q72 on mitochondria fission/fusion we 320 compared mitochondrial length, area weighted form factor and form factor in cell-autonomous and 321 transmission co-cultures and compared these to control. We used MFDs to avoid contamination 322 with neuronal mitochondria. We did not analyze cell autonomous + transmission co-cultures, 323 where we cannot discriminate myotubes which received HTTEx1Q72-mCherry from neurons from 324 those which did not. We observed a significant reduction in all three parameters when 325 HTTEx1Q72-mCherry was expressed in myotubes and when the myotubes received 326 HTTEx1Q72-mCherry from the neurons (Fig. 6a). A fragmentation of the mitochondrial 327 filamentous network is likely to impair mitochondrial function. For example, in HD patient skeletal 328 muscles, a reduction in ATP production has been observed and patients suffer from exercise-329 induced muscle fatigue already at preclinical stages of the disease 33,35 . To assess whether any 330 functional change occurred in myotubes, we measured the myotube contractions. Strikingly, we 331 observed a nearly complete loss of myotube contractions both measured by activity and 332 contraction area, selectively in transmission or in transmission + cell autonomous co-cultures, 333 despite the fact that these neurons displayed AP firing upon current injections (Fig. 6b, 2d).

334
Interestingly, we also observed a smaller in magnitude but significant increase in activity in 335 myotubes with cell autonomous expression that was independent of neuronal transmission. (Fig.  336 6b). 337 To further assess pathological consequences of transmitted HTTEx1Q72 for myotubes 338 we analyzed the extend of nuclear accumulation of HTTEx1Q72-mCherry aggregates in mixed 339 genotype co-cultures. Nuclear aggregates in skeletal muscle of R6/2 mouse models of HD have 340 been detected and their increase correlated with the worsening of disease pathology 36 . We 341 observed the lowest number of nuclear aggregates in transmission co-cultures, this number was 342 slightly higher when the protein was expressed cell autonomously in myotubes and significantly 343 increased in the concurrent presence of cell autonomous expression and transmission from 344 neurons (Fig. 6c).

346
Mutant HTTEx1 is transmitted from the motor cortex to skeletal muscle in vivo in mice 347 With the in vitro experiments we so far demonstrated that mutant HTTEx1 is transmitted across 348 the NMJ from neurons to muscle cells and that this induces pathological changes in the receiving 349 myotubes. To understand whether this is also likely to occur in patients we studied the 350 transmission of mutant HTTEx1 from the motor cortex to the skeletal muscles in vivo in mice. To 351 this end we designed adeno-associated viruses carrying a floxed HTTEx1Q138-v5 plasmid 352 (AAV_LoxP-Q138-v5). We chose a longer (138) CAG repeat, since mice are more resistant to 353 CAG repeat expansion than humans. We used a 9 amino acid long v5 reporter tag to further 354 exclude that the longer mCherry and Cre tags that we used in in-vitro were the driving force of 355 the mutant HTTEx1 transmission. The AAV_LoxP-Q138-v5 was injected in the right hemisphere 356 of the primary motor cortex (PM1) of mice expressing Cre selectively in the layer 5 pyramidal 357 neurons (nex-cre mice). We observed HTTEx1Q138-v5 expression in the motor cortex (Fig. 7a). 358 After 6 Months we analyzed the brachial spinal cord and the Triceps forelimb muscle and found 359 HTTEx1Q138-v5 positive aggregates (Fig. 7b, c). We observed higher number of aggregates in 360 contralateral side (left), confirming transmission following neuronal connectivity (Fig. 7d). 361 362 363 364

366
When cell-to-cell transmission of mHTTEx1 occurs and whether it is regulated by functional 367 synaptic connectivity and can contribute to disease in an environment of ubiquitous expression of 368 the mutant protein is to date not well understood. To advance our current understanding of these 369 processes we established two in-vitro hiPSC derived neuro-muscular co-culture systems to study 370 the role of neuromuscular connections in the development of HD-related skeletal muscle 371 pathology. We provide evidence that neuromuscular transmission of mHTTEx1 can occur across 372 the human neuromuscular synapse, likely already at early preclinical stages of disease, and 373 contributes to skeletal muscle pathology. Furthermore, our findings suggest that mHTTEx1 374 transmission is more efficient when synaptic activity and density are increased. Finally, we show 375 that mHTTEx1 is transmitted along the corticospinal pathway to skeletal muscles in mice in vivo.

376
With the newly established cre-lox co-culture system we could follow HTTEx1Q72-cre 377 transmission in the same culture, over weeks with fluorescent live-cell imaging. This revealed that 378 transmission happened over time in the absence of detectable Em48+ aggregates in the neurons 379 and that aggregation occurred first in the myotubes. Thus, mHTTEx1 is transmitted in the form of 380 smaller protein structures, potentially as oligomers. Oligomeric structures are more soluble then 381 larger aggregates, such as fibrils or inclusion bodies and therefore can diffuse more easily 49 . 382 Furthermore, mHTTEx1 transmission happened already during the time of NMJ assembly in vitro. 383 This thus may represent a very early pathological process in HD, possibly regulated by the 384 immediate activity of the synaptic vesicle fusion machinery upon the contact of the growth cone 385 with the muscle 50 . 386 Further, using Neu HTTEx1Q72-mCherry/Myo co-cultures we confirmed that transmission 387 resulted in an increase of predominantly small protein-assemblies over time in the myotubes. 388 HTTEx1Q72-mCherry puncta were associated with NMJ-forming AChR clusters types and 389 transmission was positively correlated with NMJ density. Previously, it has been shown that 390 synaptic density elevates neuron-to-neuron transmission of Tau 15 . These data together suggest 391 that the density of synaptic connectivity between cells might be an important factor affecting toxic 392 protein levels in postsynaptic cells. Additionally, our work encloses that mHTTEx1 secretion can 393 be both regulated and constitutive. We found that HTTEx1Q72-mCherry transmission is elevated 394 by neuronal depolarization, while preventing neuronal presynaptic release results in decreased 395 transmission. This, together with previously published data on Aβ, Tau and mHTTEx1, strongly 396 suggest that transmission occurs across functional synapses and is regulated by at least 397 presynaptic activity 14-17,51 . In addition, a constitutive secretion is supported by our observation that 398 hiPSC clones with higher HTTEx1Q72 expression levels showed more transmission to muscle 399 (~15-fold increase in HTTEx1Q72-mCherry puncta between myotubes cultured with clones #72 400 and #75). A positive correlation between mHTTEx1 concentration and aggregation has been 401 previously shown 52 . Similar, for α-syn and Aβ a positive correlation between the intracellular 402 levels and the amount that is released has been reported 53,54 . Thus, intracellular presence of 403 misfolded proteins might trigger a highly sensitive stress response resulting in active transmission 404 of the toxic species. 405 The pathobiological relevance of the misfolded protein transmission in an environment of 406 ubiquitous expression of the toxic protein, as it is the case in HD, has not been assessed so far. 407 It has been shown that transmission alone induces non-cell autonomous pathology in hiPSC- in the membrane is likely to cause a disruption of the lipid bilayer, with potentially a distorted 427 localization of membrane receptors, including those required for normal transsynaptic signaling 62 . 428 Finally, we show that chronic inhibition of neurotransmitter release by exposing NM co-429 cultures to TeNT not only reduced release, but also resulted in increased nuclear aggregate 430 pathology in HTTEx1Q72-mCherry expressing neurons. Clearance of misfolded proteins by the 431 ubiquitin-proteasome system and autophagy is crucial to prevent protein accumulation to avoid 432 aggregation 63,64 . Our finding suggest that toxic protein release might resemble a so far undefined 433 pathway of misfolded protein clearance. A similar observation has been made for Aβ 53 . In this 434 light, the new drug discovery strategies should promote release and prevent uptake of mHTT. 435 The current antibody-based therapy designed to prevent Aβ, tau and α-syn accumulation in tau-436 and synucleinopathies, would also be a valuable strategy to test in HD 65,66 . 437 Taken together, the positive correlation that we observe between NMJ density and the 438 HTTEx1-mCherry puncta, with transmission-triggered pathology suggests that the high number 439 of synaptic connections in the CNS and between the spinal motor neurons and skeletal muscle 440 makes these structures particular vulnerable to HD 35 . Given the peripheral phenotype, these 441 findings also provide novel opportunities for biomarker development to assess the presence and 442 contribution of this pathway in HD patients. 443 In a broader context, transsynaptic transmission of misfolded proteins is likely a common 444 mechanism in PMDs, by which these toxic species spread through the brain and the periphery, 445 contributing to a temporal decline of patient's functional abilities. 446 447 448

548
Cell culture inserts 549 Culture-Insert 2 Well from IBIDI (81176) were used to seed iND3 and iMD3 in spatially separated 550 areas of the well. Cell densities were adapted to this format. iND3 were seeded on Poly-L-lysine 551 (P1524, Sigma) and laminin-521, iMD3 were seeded on laminin-521. Cells were monitored with 552 EVOS M7000 microscope at day 1, 4, 8, 15 after the removal of the culture insert to check for the 553 presence of GFP cells.

555
Contractility assay in on-top co-culture 556 The primary readout for the amount of contraction in any on-top co-culture was captured as the 557 total amount of motion within any given field of view over time. The Yokogawa CV7000 558 microscope with a 10x objective (NA 0.3) was adopted for this assay. The raw 559 images were acquired as a series of 2560x2180 16-bit grayscale brightfield images with a 560 frequency of 2 Hz., for a total amount of 60 images per field with 4 fields of view in each well. This 561 assay was performed in live cell-imaging conditions (37°C, 5%CO2). At least 3 wells per 562 experimental condition were acquired and analyzed. 563 For each consecutive pair of image frames a motion field was computed which provides, for 564 each pixel location, a direction and magnitude of projected spatial motion. Thus, for N image 565 frames we obtained N-1 motion frames. A numerical threshold on the magnitude of the motion 566 vectors was applied to eliminate possible noise and vibration artefacts and to obtain 567 a reliable binary image map of region-of-contraction. The union of all such pixels over all motion 568 frames in the time series was computed and used as the final region-of-contraction map for 569 comparative analysis between cell lines or treatments. These values were used to describe the 570 on-top co-culture functionality as follows: 571 • Total contracting area normalized to well area (%): sum of moving pixels normalized to 572 the acquired fields area (namely, sum of the pixels occupied by the 4 fields of view) 573 washed twice with ice-cold PBS, and subsequently lysed in RIPA buffer supplemented with 603 complete EDTA-free protease inhibitor mixture (11873580001, Roche). Lysates were incubated 604 on ice for 15 min and cleared via centrifugation (10,000 × g) for 10 min at 4 °C. Supernatants were 605 collected, and the protein concentration was determined using a BCA assay kit (Thermo Scientific 606 Pierce, 23227). Lysates were resolved using standard SDS-PAGE gels and after blocking, blots 607 were incubated with primary antibodies overnight at 4 °C. After washing, blots were incubated 608 with secondary antibodies and visualized using SuperSignal Femto chemiluminescent detection 609 kit (Thermo Scientific) in Odyssey Infrared Imager (LiCor, 9120). The image in Fig. 1b is  610 representative from 3 independent experiments. 611 612 Immunofluorescence on coverglass culture 613 Cells on glass coverslips (in format 24 well plate with 3 × 10 5 myotubes and 3 × 10 5 neurons 614 density) were fixed for 5 min in 4% PFA/4% sucrose at RT, permeabilized with PBS+/+ (D8662, 615 Sigma, supplemented with 1 mM MgCl2 and 0.1 mM CaCl2)/Triton-0.1%, blocked with 5% BSA 616 in PBS+/+ and labeled with primary antibodies in PBS+/+ (D8662, Sigma) and 5% BSA overnight 617 at 4°C and secondary antibodies for 1h RT. PBS+/+ washes were performed after each antibody 618 incubation. Coverslips were mounted on glass slides in Prolong (P36930, Invitrogen). 619 620 Microfluidic devices (MFD) culture. 621 We used XonaChips XC450 devices from Xona Microfluidics. iMD3 cells were growth in 5%KSR 622 medium (Alpha-MEM (12571-063, Gibco); 5% KSR (10828028, Gibco); 1% Pen/Strep (15140-623 122, Gibco); 100µM β-Mercaptoethanol (21985-023, Gibco), supplemented with 1µg/ml 624 doxycycline (D1822, Sigma) and 20ng/ml FGF (300-112P, Gemini Bio). At DIV 3 the cells were 625 seed in final format for differentiation: In myocytes side, 3 × 10 5 cells were seed and in the 626 neuronal side, 1.5 × 10 5 cells in 5 µl medium, were seeded to give support to the motor neurons. 627 The myotubes growth for 7 days in "Differentiation medium" (DMEM subtracted (rolling ball radius = 50 pixels) and uneven labeling of mitochondria was improved 643 through local contrast enhancement using contrast-limited adaptive histogram equalization 644 ("CLAHE"). To segment mitochondria, the "Tubeness" filter was applied. After setting an 645 automated threshold, the "Analyze Particles" plugin was used to determine the area and perimeter 646 of individual mitochondria and the "Skeletonize" function was used to measure mitochondrial 647 length.

648
Three parameters were assessed: 649 -Mitochondrial length: the length reports the mitochondrial length or elongation in pixel, after the 650 mitochondria are reduced to a single-pixel-wide shape ("Skeletonize" function on ImageJ).

651
-Form factor (FF): The FF value describes the particle's shape complexity of the mitochondria, 652 as the inverse of the circularity.

653
-Area-weighted form factor (AWFF): a variant of FF with a bias towards larger mitochondria or 654 mitochondrial networks. AWFF provides more realistic results in cases where highly elongated 655 mitochondria are overlapping 656 657 Animal husbandry 658 Adult NEX-Cre were kindly provided by Dr. Sandra Goebbels (Max-Planck-Institute of 659 Experimental Medicine, Goettingen, Germany). All mice were housed in temperature (22°C) and 660 light-controlled environment on a 12-light dark cycle and had access to food and water ad libitum. 661 All experimental procedures were carried out according to Basel University animal care and use 662 guidelines. They were approved by the Veterinary Office of the Canton of Basel-Stadt, 663 Switzerland.

665
Delivery of viral vectors 666 Four-week old female NexCre mice were anaesthetized by the administration of 4% isoflurane, 667 were maintained under isoflurane anesthesia (1-2%) and kept warm with a heating pad (53800, 668 Stoeling). The head was fixed to a stereotaxic frame (Kopf Instruments) with ear bars and the skin 669 was disinfected with 70% ethanol and polyvidone iodine. The skin was cut with surgical scissors 670 to expose the skull, allowing the identifications of bregma and lambda. Using a borosilicate glass 671 pipette and a pressure ejection system (Eppendorf) 250 nl of the self-complementary AAV-9/2-672 DIO-mHTTExon1Q138-V5 (VVF, Zürich) were injected in the layer V of the primary motor cortex, 673 using the following coordinates AP (anterior-posterior): + 1.18, ML (medial-lateral): + 2.00, DV 674 (dorsal-ventral): + 2.00, according to the Paxinos and Franklin mouse brain atlas (Paxinos and 675 Franklin, 2019, eBook ISBN: 9780128161609). The mice were placed in a recovery cage to 676 awaken before returning to their home cage. 6-month-old female mice were anesthetized by the 677 administration of 4% isoflurane and were fast decapitated.

679
Immunohistochemistry for spinal cord and muscle samples. 680 The spinal cord, biceps and triceps were dissected on ice and embedded in low-melting agarose 681 (16520050, ThermoFisher Scientific). Samples were sliced in 100-150 µm thick sections using a 682 vibratome (VT1200, Leica). Then, fixed in 4% paraformaldehyde (PFA) at room temperature for 683 10 min, followed by 3 DPBS (14190, Sigma-Aldrich) washings, 10 minutes each. DPBS 684 supplemented with 0.1% Triton X-100 (for permeabilization) and 1% BSA (blocking) was used for 685 primary antibodies labelling, for at least 2 days at 4°C. After three washing steps with DPBS, cells 686 were incubated for 3h with secondary antibodies (diluted 1:800, Invitrogen). Afterwards, cells were 687 washed in DPBS for 3 times. Sections were then mounted on glass slides using ProLong Gold 688 (P10144, ThermoFisher Scientific) and acquired with LSM800 with 40x objective (ZEISS, EC 689 Plan-NEOFLUAR 40X/1,3 Oil) 690 691 Image acquisition and analysis 692 Fluorescence signals in "on top" culture for iPSC-derived co-culture were imaged with Zeiss LSM-693 700 system with a Plan-Apochromat 40 × /NA 1.30 oil DIC, using Zen 2010 software. For bin 694 analysis in MFD, section of 0-160, 160-320 and 320-480 µm were taken in the using Zeiss LSM-695 800II inverted system with a Plan-Apochromat 40 × /NA 1.30 oil DIC, using Zen blue 2.6 software. 696 Whole-cell, 16-bit stacks images with 0.33-μm step size were acquired (15-30 planes). Immersion 697 oil with 1.518 refractive index at room temperature was applied to the lens. Coverslips were 698 mounted with ProLong Gold anti-fade reagent (P36930, Thermofisher) with a refractive index of 699 1.46. All images were acquired with identical microscope settings within individual experiments. 700 Brightness and contrast were adjusted equally for all images, and cropped insets were generated 701 in the same manner among all the experiments to facilitate visualization of representative cells.

702
Saturation was avoided by using image acquisition software to monitor intensity values. For any 703 image adjustment, identical settings were always applied to all cells, irrespective of genotype. 704 Cells that were clumped or overlapping were excluded from quantification. For quantification, 705 values were averaged over multiple cells from at least three independent culture preparation. 706 Quantification of number and volume HTTEx1Q72-mCherry puncta was done using Imaris 707 Software (v.9.6.0; Oxford Instruments) function and measurement based on mCherry 708 fluorescence staining. Aggregates with volume above 0.02 and below 30 µm 3 and localized within 709 the surface generated based on MHC1 staining were analyzed. 710 Intracellular localization of aggregates was analyzed using distances between surfaces generated 711 based on mCherry staining for aggregates and MHC1 staining for muscle by Imaris software. 712 Localization at the surface was defined as distance between 0 and 0.05 µm. 713 Quantification of nuclei containing HTTEx1Q72-mCherry were done using Image J 714 software. Images were background subtracted and after setting an automated threshold a mask 715 for DAPI positive nuclei (in MHC1 positive myotubes or MAP2 positive neurons) was applied and 716 the "Analyze Particles" plugin was used to determine the number of puncta per nucleus. 717 Quantification of HTTEx1Q138-V5 puncta in triceps were done using Image J software. 718 Images were background subtracted and after setting an automated threshold, the "Analyze 719 Particles" plugin was used to determine the number of puncta in MHC1 positive staining. 720 Integrated density of EM48/MAP2+NF staining was done using Image J software. Images were 721 background subtracted and after setting an automated threshold, integrated density was 722 measured in the full image, to consider EM48 staining in the soma and neurites of the neurons.

723
The values were normalized to the neuronal market MAP2 and NF. 724 AChR clusters were analyzed using Imaris Software (v.9.6.0; Oxford Instruments) based 725 on immunofluorescent images acquired by Confocal microscope. 3D reconstruction of AChR and 726 BSN structures were done using Imaris surface function. Automatically generated values for 727 volume and sphericity were used to characterize the clusters. Only structures with volume above 728 0.02 and below 20 µm 3 were analyzed for BSN and only structures above 0.024 µm 3 for AChR.

729
Distances between the surfaces provided by Imaris software were used to identify AChR cluster 730 and BSN association. Association was defined as distance below 0.05 µm. 731 Multiple images were analyzed using Imaris Batch function. The data on volume, 732 sphericity and distances between the surfaces were exported and further analyzed using R 733 (