Multi-omics Identify Serotonin Transporter as a Promising Therapeutic Target for Essential Tremor

18 Essential tremor (ET) stands as one of the most prevalent cerebellar movement 19 disorders. However, effective treatment remains elusive, largely due to a limited 20 understanding of its molecular pathology. Harmaline-induced tremor in mouse 21 is a well-established animal model for ET, while with enigmatic mechanism. The 22 aim of this study was to get insight into the molecular intricacies underlying 23 cerebellar dysfunction in harmaline-induced tremor. Combining LC-MS/MS and 24 RNA-Seq analysis, we delved into the variation of the cerebellum between 25 harmaline-induced tremor and the control ones. This comprehensive 26 investigation revealed a profile of this mouse model from mRNA and protein 27 level, highlighting 5194 correlated coding molecules, with 19 proving to be 28 significant. Further KEGG enrichment analysis identified cerebellar serotonin 29 transporter (SERT) as the key molecule in harmaline-induced tremor. The 30 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint implications of this transcriptomic and proteomic exploration underscore the 31 potential therapeutic value of targeting SERT as a novel treatment approach for 32 ET. In general, our study unveils crucial insights that could pave the way for 33 molecular target identification and effective therapeutic interventions for ET. 34 35 Key Words: 36 Essential tremor, Transcriptomic sequencing, Proteomic sequencing, Serotonin 37 transporter, Therapeutic target 38 39

40 Essential tremor (ET) is one of the most common movement disorders 1,2, 41 characterized by the rhythmic oscillation of agonist and antagonist muscle 42 groups, typically occurring at a frequency of 8 to 12 Hz3,4. The incidence of ET 43 has been reported to be around 0.9%, with a significant increase in prevalence 44 among individuals over 65 years old, reaching 4.6% 5. Drugs with established 45 efficacy include propranolol, a β-adrenergic blocker, and primidone, an 46 anticonvulsant. However, these medications are associated with side effects, 47 and their efficacy is limited, resulting in an average tremor reduction of only 48 around 50%. Consequently, there is pressing need for targeted therapies that 49 explore new biological pathways. Nevertheless, the underlying mechanisms of 50 ET remain elusive. Harmaline -induced tremor is a recognized model used to 51 screen new therapies for ET6. In laboratory animals, a tremor at a frequency of 52 8-12 Hz can be generated after intraperitoneal injection of harmaline at doses 53 ranging from 10 to 50 mg/kg, with a time latency of 3-10 minutes7. Furthermore, 54 Louis et al ., discovered elevated concentrations of harmane, another 55 compound of harmala alkaloids, in ET patients compared to controls in the 56 Faroe island 8. This suggests that β-carboline alkaloid may contribute to ET 57 development in both patients and laboratory animals. However, the exact 58 pathology underlying harmaline -induced tremor has not been conclusively 59 elucidated9. 60 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint The prevailing viewpoint suggests that harmaline can instigate abnormal 61 activation of climbing fibers (CF) within the inferior olivary nucleus (ION), 62 leading to the formation of aberrant synapses with Purkinje cells (PC) in the 63 cerebellar cortex 10. While recent findings indicate that harmaline induces a 64 burst pattern of activity in Purkinje cells, and the absence of Purkinje cell 65 neurotransmission can attenuate harmaline -induced tremor 11. This 66 underscores the crucial role of the cerebellar cortex itself, particularly the 67 Purkinje cell s, in harmaline -induced tremor. Understanding the mechanism 68 underlying harmaline -induced tremor in the cerebellum may help reveal the 69 molecular basis of ET. 70 Transcriptomic analysis has been widely recognized as an efficient method for 71 unveiling tissue -specific alterations, including gene splicing, structural 72 variations and transcription al changes12. This technique is capable of 73 uncovering variations in gene expression13. However, the significance of post-74 transcriptional modifications and protein turnover in determining protein 75 function should not be underestimated14. The unique advantage of combined 76 analyses of transcriptomics and proteomics lies in their integration, which helps 77 mitigate systematic errors associated with each method individually. In this 78 study, we employed a comprehensive approach integrating both transcriptomic 79 and proteomic analysis to thoroughly disclose the alteration in the cerebellum 80 induced by harmaline treatment. To gain deeper insights into the mechanism 81 contributing to harmaline -induced tremor in the cerebellum, we conducted a 82 comparative analysis of the cerebellums of mice treated with harmaline and 83 their control counterpart. Through the integration of transcriptomic results and 84 proteomic traits, numerous candidate genes were identified. Subsequent 85 extended biological experiments were conducted to validate the role of the 86 candidate molecule serotonin transporter ( SERT) in cerebellar PC activity 87 during harmaline-induced tremor. Thus, this study revealed a novel molecular 88 mechanism for harmaline -induced ET and screened potential therapeutic 89 targets. 90 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint 91

92 Tremor detection in freely moving mice 93 Tremors were recorded using a tremor detector (Medusa, bio -signal), by pre -94 implanting an electrode slice on the skull of the mice. During tremor 95 measurement, the electrode slice was connected to a transformer, enabling the 96 recording of behavioral data and transforming vibratory signals into digital 97 signals for subsequent analysis. The collected data were analyzed using a 98 preset program that employed the power spectrum density function and further 99 transformed it into frequency domains. Normalization of the spectrum data was 100 achieved through logarithmic (lg) calculation. Adult mice were administered 30 101 mg/kg harmaline (Topscience, T2792) via intraperitoneal injection (i .p.). For 102 specific treatments, mice were administered DSP -1053 (10 mg/kg), harmaline 103 (30 mg/kg) or an equal amount of saline as a control. DSP-1053 was applied to 104 mice (i.p.) 1 hour before harmaline, ensuring that the blood concentration of 105 serotonin would almost reach its peak when treated with harmaline15. The mice 106 were then sacrificed 30 min after harmaline injection. 107 108 Animals 109 The study was conducted in accordance with the approval of the Animal Ethics 110 Committee of the (Shanghai, China). Analyses were conducted on 6 -week-old 111 C57BL/6 male mice, which were subjected with harmaline and DSP -1053 112 injections, along with their control wild -type littermates (WT). The mice were 113 housed under controlled conditions of lighting (12-hour light, 12-hour dark cycle) 114 and temperature (22 ± 2°C), with unrestricted access to food and water. 30 115 minutes after injection, animals were anesthetized with isoflurane and 116 subsequently euthanized by decapitation. Cerebellar cortexes samples were 117 promptly dissected and either flash -frozen in liquid nitrogen for proteomic 118 analysis or preserved in RNALater (Beyotime, R0118) for transcriptomic 119 sequencing afterwards. 120 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint 121 Experimental design 122 In the proteomic analyses, three mice were included in each group (harmaline-123 induced tremor vs WT). Cerebellar cortex samples were collected, and proteins 124 were extracted, reduced and alkylated. Subsequently, each fraction underwent 125 trypsin digestion to generate peptides. LC -MS/MS spectrometry analysis was 126 performed to identify and quantity proteins , and the resulting data were 127 subjected to the final bioinformatics analysis. For transcriptomic analyses, an 128 equal number of 3 mice were utilized in each group. Following RNA extraction 129 and verification, cDNA libraries were constructed and sequenced to obtain data 130 for the final bioinformatics analysis. 131 132 Protein extraction and digestion and protein-protein interaction analysis 133 The protein samples were dissociated by SDT buffer (4% SDS, 100 mM Tris-134 HCl!1 mM DTT!pH 7.6), and the quantity of which were determined by BCA 135 (Bio-Rad, USA). Filter-aided sample preparation (FASP) procedure was used 136 to digest protein with trypsin 16,17, then the samples were desalted by C18 137 Cartridges (Empore™ SPE Cartridges C18 , Sigma), concentrated and 138 reconstituted by formic acid. 139 The IntAct molecular interaction database ( http://www.ebi.ac.uk/intact/) and 140 STRING software (http://string-db.org/) was used to analyze the protein–protein 141 interaction (PPI) information. Cytoscape software (http://www.cytoscape.org/, 142 version 3.2.1) was then utilized for the PPI visualization. 143 144 RNA extraction and transcriptomic analysis 145 Total RNA was extracted from cerebellum cortex samples preserved in 146 RNALater using TRIzol reagent. The RNA concentration was quantified at 260 147 nm using the RNA 6000 Nanodrop on the Agilent 4150 Bioanalyzer (Agilent). 148 Afterwards, mRNAs were isolated using beads with Oligo (dT) and subjected to 149 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint random fragmentation buffer. Complementary DNAs (cDNAs) were then 150 synthesized based on the extracted mRNA and purified by AMPure XP beads. 151 The enriched cDNAs were further amplified via PCR. 152 153 Liquid chromatograph triple quadrupole mass spectrometer analysis 154 Q Exactive mass spectrometer (Thermo Scientific) was used to analyze Liquid 155 chromatograph triple quadrupole mass spectrometer (LC-MS/MS) coupled with 156 Easy nLC (Proxeon Biosystems, now Thermo Fisher Scientific) . Peptides was 157 dissolved in buffer A ( 0.1% Formic acid) loading into a a reverse phase trap 158 column (Thermo Scientific Acclaim ), the column was connected to a C18-159 reversed phase analytical column (Thermo Scientific Easy Column ). Buffer B 160 (84% acetonitrile and 0.1% Formic acid) was used to separate the peptides with 161 the rate of 300 nL/min. 162 The mass spectrometer was operated under positive ion mode , the most 163 abundant precursor ions were acquired by HCD fragmentation scanning. The 164 dynamic exclusion duration was set at 40.0 s, survey scans were acquired at a 165 resolution of 70,000 at m/z 200, and the resolution for HCD spectra was set at 166 17,500 at m/z 200, the isolation width was 2 m/z. 167 168 Bioinformatic analysis 169 GO annotation 170 NCBI BLAST+ client software were utilized to search th e differentially 171 expressed proteins sequences, after the homologous sequences were 172 identified with InterProScan, the protein sequences were mapped with gene 173 ontology (GO) terms , annotated by Blast2GO software , and subsequently 174 plotted by R scripts. 175 176 KEGG annotation 177 After completing the annotation , Kyoto Encyclopedia of Genes and Genomes 178 (KEGG) orthology was identified by KEGG database (http://geneontology.org/) 179 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint with a BLAST search, and then mapped to pathways. 180 181 Western blot 182 Fresh cerebellar cortex samples from mice were solubilized in lysis buffer 183 containing proteinase inhibitors (Thermo Fisher, A32965) and phosphatase 184 inhibitors (Thermo Fisher, 78420), After sonication and centrifugation, the 185 supernatant was combined with loading buffer (Thermo Fisher, AM8547). 186 Protein concentration was determined using the Pierce BCA Protein Assay Kit 187 (Thermo Fisher, 23225). Following sample preparation, the proteins were 188 loaded onto a 10% SDS -PAGE gel and transferred onto a PVDF membrane 189 (Millipore). The membrane was blocked with 3% BSA (BioFroxx, 4240GR100), 190 and then incubated with primary antibodies: SERT (1:1000, Abcam, ab102048) 191 and GAPDH (Thermo Fisher, A300-639A-T). After an overnight incubation with 192 the primary antibodies, the respective secondary antibodies (1:2000) were 193 applied. Signals were detected using Tanon-5200 system (Tanon). 194 195 ELISA assay 196 An ELISA kit (MM -0443M1) was used to determine the concentration of 197 serotonin in the cerebellum of harmaline-treated mice and their corresponding 198 control counterpart. The cerebellum samples were washed, sonicated , and 199 centrifuged with PBS ( pH=7.4, Sangon biotech, B548117 -0500), and the 200 resulting liquid supernatant was collected for further examination. The results 201 were measured using a multi-mode microplate reader. 202 203 Primary culture of cerebellar cortical neuron 204 At embryonic day 18 (E18), mice were anesthetized with diethyl ether. 205 Following sterilization with 75% ethyl alcohol, a midline incision was made in 206 the abdomen to expose and separate the uterus form the pregnant mice. Next, 207 fetal mouse heads were extracted and placed in a Petri plate containing 208 dissection buffer (DMEM + 5% penicillin/ streptomycin). The scalps of the fetal 209 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint mice were then cut open using sterilized scissors, and the brain tissues were 210 carefully extracted. Under an optical microscope, the cerebellum was separated 211 from the whole brain tissue, the meninges were peeled off from the surface of 212 the brain. The isolated cerebellum was placed in a 7 mL centrifuge tube with 213 digestion buffer (2 mL 0.25% trypsin + 2 mL dissection buffer). The centrifuge 214 tube was transferred to a constant -temperature and humidity incubator for 15 215 minutes, with gentle shaking every 5 minutes during incubation. After incubation, 216 excess digestion buffer was removed and 4 mL of complete culture medium 217 (DMEM + 10% FBS + 1% GlutaMAX) was added. The mix ture was then 218 homogenized with the digested brain tissue, filtered through a filter net, and the 219 resulting cell suspension was centrifuged at 1000 rpm for 5 minutes. After 220 centrifugation, the cell pellet was resuspended in 2 mL of complete culture 221 medium. After cell counting, the cell suspension was added to cell -culture 222 dishes preloaded with 2 mL of culture medium for cerebellar neurons (50% 223 complete culture medium + 50% neural selective medium (Neuralbasal + 2% 224 B27 + 1% GlutaMAX) + 2 uL T3 (20 mg/ml)). Cerebellar neurons were cultured 225 in a constant-temperature and humidity incubator. 226 227 Cell electrophysiology 228 The electrophysiological characteristics of cultured cerebellar cortical neuron s 229 were recorded using voltage clamp techniques in the whole-cell mode of patch 230 clamp at room temperature. Data acquisition and analysis were performed 231 using the patch clamp amplifier system (MultiClamp 700B) and digital analog 232 converter (Digidata 1550B and pClamp10). For the recording, glass electrodes 233 filled with filtered electrode fluid (10 mM NaCl, 5 mM KCl, 1 mM MgCl 2, 2 mM 234 CaCl2, 10 mM HEPES, 10 mM Glucose, pH 7.4, 310-320 mOsm/L) were utilized. 235 The resistance of the glass electrode, when immersed in the extracellular fluid 236 (150 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 210 mM glucose), ranged 237 between 2 -5 M Ω. Following baseline normaliz ation, the glass electrode was 238 connected to the cultured cerebellar cortical neuron under negative pressure. 239 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Then, the cell membrane was aspirated until it broke, achieving high -pressure 240 sealing and establishing the whole-cell mode of patch clamp. 241 242 Virus injection and optical recording 243 The mouse was anesthetized with isoflurane. Then, the mouse’s head was 244 restrained in a stereotaxic instrument, and the scalp above the cerebellar region 245 was incised following fur removal. Adequate sterilization was applied, and a 246 hole was drilled in the skull. A Hamilton syringe containing the virus was inserted 247 into the mouse cerebellum with the coordinates ( relative to Bregma: AP -6.75 248 mm, ML 1.8 mm, DV -2.5 mm). A total of 300 nL of the virus, administered at a 249 rate of 0.1 uL per minute, was injected into the cortex of the mouse’s cerebellum. 250 Ten minutes after completing the injection, the syringe was gradually removed 251 at a rate of 0.05 mm per minute. Subsequently, an optogenetic fiber (Thinker 252 Tech Nanjing Biotech Co., Ltd) was implanted above the virus -injection area 253 (relative to Bregma: AP -6.75 mm, ML 1.8 mm, DV: -2 mm). Following these 254 procedures, the scalp was sutured, and the mice received appropriate post -255 operative care. Upon full recovery and viral transgene expression, the optical 256 signal of serotonin was detected using Signal-channel Fiber Photometry 257 (Thinker Tech Nanjing Biotech Co., Ltd). 258 259 Statistical analysis 260 Tissue level of serotonin system, tremor detection of mice and 261 electrophysiological parameters were evaluated using GraphPad Prism9 262 (GraphPad Software, La Jolla, CA). A two -tailed Student’s t-test was used for 263 the comparison of two groups, while one -way analysis of variance (ANOVA) 264 was used for comparisons involving more than two groups. The homogeneity 265 of variance was assessed using Brown -Forsythe and Bar tlett’s tests. All data 266 were presented as mean ± SEM (standard error of the mean), with ‘n’ 267 representing the sample number (i.e., the number of independent experiments 268 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint or cell numbers). Significant differences were denoted as *P < 0.05!**P < 0.01!269 ***P < 0.001 and ****P < 0.0001. 270 271

272 Rhythmic activity detection of mice induced by harmaline. 273 We established a harmaline -induced tremor model in mice following 274 established protocols from previous studies6,18,19 (Fig. 1A). Upon injection with 275 harmaline (dissolved in DMSO and diluted with saline, in vivo 30 mg/kg), mice 276 exhibited pronounced tremor s manifesting across the head, trunk , and limbs. 277 Compare to their control littermates injected with the vehicle (DMSO diluted with 278 saline), the tremor s presented in harmaline -injected mice exhibited a distinct 279 frequency range of 8 -20 Hz, initiating approximately 3 minutes after 280 intraperitoneal injection ( Fig. 1B-G). Additionally, mice subjected to harmaline 281 treatment displayed significantly higher intensity in the 8 -20 Hz range in 282 contrast with the control group ( Fig. 1H). The tremors persisted for 283 approximately 2 hours, reaching their peak at 30 minutes post -injection. 284 Subsequently, we sacrificed the model mice 30 minutes after harmaline 285 administration and separated cerebellar cortex for protein and RNA extraction. 286 287 Transcriptomic and proteomic workflow and overall characterization 288 following harmaline-induced ET model. 289 To date, no study has documented transcriptomic or proteomic changes in 290 cerebellar cortex of harmaline -administrated mice for ET model. As illustrated 291 in the schematic workflow ( Fig. 2A), we performed both transcriptomic and a 292 proteomic analysis respectively, comparing 3 mice treated with harmaline to 3 293 vehicle-treated littermates so as to unravel the transcriptomic and proteomic 294 profile changes associate with harmaline administration. For each cerebellar 295 cortex lysate sample, we performed LS-MS/MS mass spectrometry analysis in 296 triplicate. The reliability of our analysis can be pledged by the high technical 297 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint reproducibility observed in our experiments and the minimal variation in our 298 cerebellar cortex samples. Notably, for transcriptomic analysis, RNA samples 299 met all the criteria for cDNA library construction standard (all OD 260/280=2.1, 300 RINs (RNA Integrity Number) vary from 8.9 to 9.2). As for proteomic analysis, 301 all electrophoretic bands were clear and the volumes of each sample were 302 sufficient (1431.6 μg ~ 2555.6 μg). 303 304 In total, we identified 35125 genes from RNA-seq and 5661 proteins from LC-305 MS/MS. Through gene and protein difference analysis ( Fig. 2B-C), it is 306 noteworthy that in the transcriptomic analysis, 614 gene s were up-regulated, 307 18 genes were down-regulated; in comparison, 469 proteins were up-regulated, 308 271 proteins were down -regulated for proteomic analysis. However, the 309 correlation of the gene expression level between these up-regulated and down-310 regulated genes and proteins in harmaline -treated mice remain unclear at this 311 point. 312 313 Co-occurring alterations in the cerebellum of harmaline -induced ET 314 identified through integrated analysis of transcriptomics and proteomics. 315 We integrated transcriptomic and proteomic information derived from the same 316 treatment, considering genes that exhibited co -directional changes in 317 transcription and translation as the correlated ones. In this context, 5194 318 correlated genes were identified, among which 19 were deemed significant, all 319 showing upregulation ( Fig.3A). Subsequently, a clustering analysis of these 320 significant correlated genes revealed a consistent pattern of upregulation 321 across all 19 (Fig. 3B). 322 To gain insights into the functional implicat ions of those correlated genes, we 323 conducted enrichment analys is. GO enrichment analysis identified over 30 324 significant GO terms that exhibiting concordance in both omics data (Fig. 3C). 325 Notably, among these terms, proteins and genes located at the cell membrane 326 are considered to undergo most significant changes (Fig . 3D). In KEGG 327 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint enrichment, only the serotonin synapse and GABA synapse pathways stood 328 out as highly significant in both transcriptomic and proteomic analyses (Fig. 3E). 329 Specifically, two genes: serotonin transporter (SERT) and GABA transporter 330 (GAT) were identified as the most significant ones (Fig. 3F). This suggests that 331 SERT and GAT may play a significant role in harmaline-induced tremor. 332 333 The pivotal role of serotonin transporter of cerebellar cortex in harmaline-334 induced tremor. 335 To verify the potential function of SERT in harmaline -induced tremor , we 336 conducted Western Blot analysis. Using the stable mouse model of harmaline-337 induced ET, we observed an increase in SERT protein levels in the mouse 338 cerebellar cortex compared to the vehicle control group, which aligns with our 339 integrated analysis of transcriptomics and proteomics (Fig . 4A-B). However, 340 despite an upward trend, the quantification of GABA transporter (GAT) showed 341 no significant difference between the two groups (Fig . 4A-C), indicating that 342 SERT may play a more crucial role in harmaline-induced tremor. 343 To further investigate the relationship between SERT and serotonin (5-HT), we 344 utilized Enzyme-Linked Immunosorbent Assay (ELISA) and found a significant 345 down-regulation of 5-HT in the cerebellum during harmaline -induced tremor 346 (Fig. 4D). We then locally expressed genetic sensor of 5-HT in the cerebellar 347 cortex by AAV to continuously monitor the fluctuation of 5-HT with high temporal 348 and spatial resolution (Fig. 4E). The results revealed that within 30 minutes of 349 harmaline administration (intraperitoneal), the content of 5-HT declined with 350 fluctuations (Fig. 4E, G), confirming the ELISA findings. This suggests that the 351 down-regulation of 5-HT may be one of the consequences of the increase in 352 SERT. 353 354 Targeted inhibition of SERT reverses the excitability and tremor induced 355 by harmaline. 356 To functionally elucidate the role of SERT in harmaline -induced tremor, we 357 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint conducted behavioral experiments in mice by inhibiting SERT. The mice were 358 pre-injected with DSP -1053, an inhibitor of SERT, one hour before harmaline 359 administration (Fig . 5A). Remarkably, w ith DSP -1053 pre -treatment, the 360 harmaline-induced tremors were significantly attenuated in mice ( Fig. 5B-D). 361 These results emphasize the critical role of SERT in harmaline-induced tremor. 362 To further investigate the potential mechanism of harmaline-induced tremor and 363 targeted inhibition of SERT , we conducted electrophysiology experiments on 364 cultured primary cerebellar cortical neurons. Using the whole-cell patch clamp 365 technique, we observed that harmaline perfusion enhance d the activity of 366 Purkinje cells (PCs) by increasing spike firing, both in amplitude and frequency 367 (Fig. 5H-L), confirming the impact of harmaline on PC activity. Importantly, pre-368 perfusion with DSP -1053 significantly reverse d the effect of harmaline, 369 highlighting the involvement of SERT in harmaline-enhanced cerebellar activity 370 (Fig. 5H-L). These results demonstrated that SERT plays a pivotal role in 371 harmaline-induced tremor by affecting PC excitability. 372 373

374 The mounting evidence implicates alterations in cerebellum in ET. Postmortem 375 examinations of ET patients have revealed characteristic neuropathological 376 changes, including torpedo-shaped swelling of neuronal branches 20 and 377 decreased density in Purkinje cells21 , as well as changes in hairy baskets for 378 basket cells22, among others. In laboratory animals, harmaline-induced tremor 379 serves as a recognized model for studying ET, providing a valuable tool for 380 qualitatively assessing the effectiveness of pharmaceutical interventions. 381 However, the potential mechanisms underlying harmaline -induced tremor 382 remain largely unknown. 383 Our approach, which combines proteomic and transcriptomic sequencing 384 technologies, introduces a novel dimension to elucidating gene signatures by 385 reducing the false discovery rates inherent in individual technologies and 386 utilizing a more precise bioinformatics infrastructure23,24. Within our dataset, we 387 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint observed significant up -regulation of 19 genes along with their associated 388 proteins. Through KEGG pathway analysis , we identified two specific genes, 389 GAT and SERT. However, our Western blot results revealed a disparity between 390 the two. Unlike SERT, the quantity of GAT did not show a significant increase 391 after harmaline treatment. This discrepancy may be attributed to the lower 392 precision of WB, as it only reflects proteins with marked variations at the protein 393 level. 394 SERT, a component of the serotonin system , is known for its involvement in 395 various neuropsychiatric disorders, including depression, bipolar disorder, 396 anxiety, and neurodegenerative conditions25. The structural similarity between 397 β-carboline alkaloid and serotonin has been reported 26. However, the role of 398 serotonin system in harmaline -induced tremor, particularly in olivocerebellar 399 function, remains a highly debated topic27-30 , with limited research highlighting 400 its role in ET. Our research group made a pioneering discovery by revealing a 401 conspicuous up-regulation of SERT in harmaline-induced tremor, suggesting its 402 potential as a key regulatory factor in this context. 403 To strengthen the validity of our findings, we employed multiple experimental 404

including WB, ELISA, and optical fiber recording. The collective results 405 consistently confirmed that harmaline induces the up -regulation of SERT, 406 leading to an increase in serotonin uptake and subsequently causing a 407 decrease in cerebellar serotonin levels. Furthermore, we conducted 408 electrophysiological assessments on cultured primary cerebellar cortical 409 neurons and observed tremors in mice. These experiments demonstrated that 410 the excitability of neurons and harmaline -induced tremors can be significantly 411 suppressed. This further validated the critical role of SERT in harmaline -412 induced tremor. To provide additional support for this conclusion, we utilized 413 DSP-1053, an inhibitor of SERT, and observed its effect on tremor suppression. 414 The application of DSP -1053 further confirmed the involvement of SERT in 415 harmaline-induced tremor. 416 Our findings align closely with previous studies that have suggested the 417 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint potential effectiveness of serotonergic agonist, such as trazodone , in 418 ameliorating ET31-34. While significant improvements were observed in only two 419 small clinical trials (involving 2 patients32 in one study and 5 out of 6 patients33 420 in another), our results make a contribution to the increasing body of evidence 421 supporting the involvement of the serotonin system in ET. These collective 422 findings emphasize the potential relevance of targeting the serotonin system as 423 a therapeutic approach for ET. 424 Nonetheless, some published articles have presented contrasting views. For 425 instance, citalopram, a selective serotonin reuptake inhibitor, has been reported 426 to augment harmaline-induced tremor35. Additionally, after harmaline injection, 427 elevated level of 5 -HT level was observed in striatum, cortex, hypothalamus, 428 hippocampus 36,37 and brainstem35. Furthermore, sertraline escitalopram, one 429 of the SSRI, was implicated in inducing movement disorders, including dystonia, 430 akathisia, parkinsonian symptoms in a 38-year-old male patient38. 431 Several factors may contribute to the discrepancies noted above. Firstly, 432 previous studies did not specifically focus on serotonin level in the cerebellar 433 cortex, where our findings revealed a decline in 5 -HT. Moreover, the observed 434 augmentation in climbing fiber reuptake of 5 -HT matches with the elevation of 435 5-HT in brainstem. Additionally, citalopram exihibits low affinity for various 436 receptors35, including SERT, dopamine receptors and monoamine oxidase 437 inhibitor receptors. As the citalopram dosage rise, the inhibition of these 438 receptors could contribute to the augmentation of tremors. Conversely, it is 439 theorized that SSRI may lead to a further reduction of serotonergic activity in 440 brainstem, enhancing glucose metabolism and adenosine, eventually bringing 441 about tremors35. 442 443

and perspective 444 The current study employed transcriptomic and proteomic analysis to elucidate 445 alterations within cerebellar cortex following harmaline treatment in mice. Our 446 findings confirm the up -regulation of SERT, a component of the serotonin 447 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint system, in harmaline-induced tremor. This highlights a potential role for SERT 448 in the pathogenesis of ET, offering a novel research strategy and identifying a 449 potential molecular target for further investigation. Moving forward, several 450 avenues for further research are suggested. Firstly, it is crucial to demonstrate 451 the molecular pathway of SERT in the pathophysiology of ET. Secondly, 452 exploring the potential of 5 -HT in peripheral blood as an auxiliary indicator for 453 the diagnosis of ET could enhance diagnostic approach. In addition, 454 considering the widespread use of 5 -HT reuptake inhibitors in depression, a 455 similar therapeutic targeting SERT could be investigated for its applicability in 456 ET treatment. 457 458 Conflict of interest 459 The authors claim that there are no conflicts of interest. 460 461

462 We thank Dr. Jun Zhou (Ruijin Hospital) for advice and help in English writing. 463 This work was supported by grants from 1. the National Natural Science 464 Foundation of China (82171239, 82371248); 2. Shanghai Jiao Tong University 465 School of Medicine physician-scientist program. 466 467

468 1. Louis ED. Essential tremor. Lancet Neurology 2005; 4(2): 100-10. 469 2. Louis ED. Essential tremor. New England Journal of Medicine 2001; 345(12): 887-91. 470 3. Zesiewicz TA, Elble R, Louis ED, et al. Practice parameter: Therapies for essential tremor - 471 Report of the quality standards subcommittee of the American Academy of Neurology. Neurology 472 2005; 64(12): 2008-20. 473 4. Elias WJ, Huss D, Voss T, et al. A Pilot Study of Focused Ultrasound Thalamotomy for Essential 474 Tremor. New England Journal of Medicine 2013; 369(7): 640-8. 475 5. Louis ED, Ferreira JJ. How Common Is the Most Common Adult Movement Disorder? Update 476 on the Worldwide Prevalence of Essential Tremor. Movement Disorders 2010; 25(5): 534-41. 477 6. Miwa H. Rodent models of tremor. Cerebellum 2007; 6(1): 66-72. 478 7. Handforth A. Harmaline tremor: underlying mechanisms in a potential animal model of 479 essential tremor. Tremor Other Hyperkinet Mov (N Y) 2012; 2. 480 8. Louis ED, Eliasen EH, Ferrer M, et al. Blood Harmane (1-Methyl-9H-Pyrido 3,4-b indole) and 481 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Mercury in Essential Tremor: A Population-Based, Environmental Epidemiology Study in the Faroe 482 Islands. Neuroepidemiology 2020; 54(3): 272-80. 483 9. Kosmowska B, Wardas J. The Pathophysiology and Treatment of Essential Tremor: The Role 484 of Adenosine and Dopamine Receptors in Animal Models. Biomolecules 2021; 11(12). 485 10. Brown AM, White JJ, van der Heijden ME, Zhou J, Lin T, Sillitoe RV. Purkinje cell misfiring 486 generates high-amplitude action tremors that are corrected by cerebellar deep brain stimulation. 487 eLife 2020; 9: e51928. 488 11. Brown AM, White JJ, van der Heijden ME, Zhou J, Lin T, Sillitoe RV. Purkinje cell misfiring 489 generates high-amplitude action tremors that are corrected by cerebellar deep brain stimulation. 490 Elife 2020; 9. 491 12. Trapnell C, Williams BA, Pertea G, et al. Transcript assembly and quantification by RNA -Seq 492 reveals unannotated transcripts and isoform switching during cell differentiation. Nature 493 Biotechnology 2010; 28(5): 511-U174. 494 13. Su AI, Wiltshire T, Batalov S, et al. A gene atlas of the mouse and human protein -encoding 495 transcriptomes. Proceedings of the National Academy of Sciences of the United States of America 496 2004; 101(16): 6062-7. 497 14. Vogel C, Marcotte EM. Insights into the regulation of protein abundance from proteomic and 498 transcriptomic analyses. Nature Reviews Genetics 2012; 13(4): 227-32. 499 15. Yoshinaga H, Nishida T, Sasaki I, et al. Discovery of DSP -1053, a novel benzylpiperidine 500 derivative with potent serotonin transporter inhibitory activity and partial 5 -HT1A receptor 501 agonistic activity. Bioorganic & Medicinal Chemistry 2018; 26(8): 1614-27. 502 16. Wisniewski JR. Filter Aided Sample Preparation - A tutorial. Analytica Chimica Acta 2019; 1090: 503 23-30. 504 17. Wisniewski JR, Zougman A, Nagaraj N, Mann M. Universal sample preparation method for 505 proteome analysis. Nature Methods 2009; 6(5): 359-U60. 506 18. Wilms H, Sievers J, Deuschl G. Animal models of tremor. Movement Disorders 1999; 14(4): 507 557-71. 508 19. Martin FC, Le AT, Handforth A. Harmaline-induced tremor as a potential preclinical screening 509

for essential tremor medications. Movement Disorders 2005; 20(3): 298-305. 510 20. Louis ED, Faust PL, Vonsattel JPG, et al. Neuropathological changes in essential tremor: 33 511 cases compared with 21 controls. Brain 2007; 130: 3297-307. 512 21. Axelrad JE, Louis ED, Honig LS, et al. Reduced Purkinje cell number in essential tremor. 513 Archives of Neurology 2008; 65(1): 101-7. 514 22. Louis ED. Essential tremor: evolving clinicopathological concepts in an era of intensive post-515 mortem enquiry. Lancet Neurology 2010; 9(6): 613-22. 516 23. Qie J, Liu Y, Wang Y, et al. Integrated proteomic and transcriptomic landscape of 517 macrophages in mouse tissues. Nature communications 2022; 13(1): 7389. 518 24. Nesvizhskii AI. Proteogenomics: concepts, applications and computational strategies. Nature 519

2014; 11(11): 1114-25. 520 25. Heils A, Teufel A, Petri S, et al. Allelic variation of human serotonin transporter gene 521 expression. Journal of Neurochemistry 1996; 66(6): 2621-4. 522 26. Hamid HA, Ramli ANM, Yusoff MM. Indole Alkaloids from Plants as Potential Leads for 523 Antidepressant Drugs: A Mini Review. Frontiers in Pharmacology 2017; 8. 524 27. Shen HW, Jiang XL, Winter JC, Yu AM. Psychedelic 5 -Methoxy-N,N-Dimethyltryptamine: 525 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Metabolism, Pharmacokinetics, Drug Interactions, and Pharmacological Actions. Current Drug 526 Metabolism 2010; 11(8): 659-66. 527 28. McIsaac WM, Khairallah PA, Page IH. 10 -METHOXYHARMALAN, A POTENT SEROTONIN 528 ANTAGONIST WHICH AFFECTS CONDITIONED BEHAVIOR. Science 1961; 134(348): 674-&. 529 29. Beaszarate C, Moralesvillagran A, Tapiaarizmendi G, Feriavelasco A. EFFECT OF 3 -530 ACETYLPYRIDINE ON SEROTONIN UPTAKE AND RELEASE FROM RAT CEREBELLAR SLICES. 531 European Journal of Pharmacology 1991; 198(1): 7-14. 532 30. Sugihara I, Lang EJ, Llinas R. SEROTONIN MODULATION OF INFERIOR OLIVARY 533 OSCILLATIONS AND SYNCHRONICITY - A MULTIPLE -ELECTRODE STUDY IN THE RAT 534 CEREBELLUM. European Journal of Neuroscience 1995; 7(4): 521-34. 535 31. Marin-Lahoz J, Gironell A. Linking Essential Tremor to the Cerebellum: Neurochemical 536 Evidence. Cerebellum 2016; 15(3): 243-52. 537 32. McLeod NA, White LE, Jr. Trazodone in essential tremor. Jama 1986; 256(19): 2675-6. 538 33. Sanson F, Schergna E, Semenzato D, et al. Therapeutic effects of trazodone in the treatment 539 of tremor. Multicentric double-blind study. Rivista di neurologia 1986; 56(6): 358-64. 540 34. Koller WC. Tradozone in essential tremor. Probe of serotoninergic mechanisms. Clinical 541 neuropharmacology 1989; 12(2): 134-7. 542 35. Arshaduddin M, Al Kadasah S, Biary N, Al Deeb S, Al Moutaery K, Tariq M. Citalopram, a 543 selective serotonin reuptake inhibitor augments harmaline -induced tremor in rats. Behavioural 544 Brain Research 2004; 153(1): 15-20. 545 36. Abdel-Fattah AF, Matsumoto K, Murakami Y, Adel -Khalek Gammaz H, Mohamed MF, 546 Watanabe H. Central serotonin level -dependent changes in body temperature following 547 administration of tryptophan to pargyline- and harmaline-pretreated rats. Gen Pharmacol 1997; 548 28(3): 405-9. 549 37. Abdel-Fattah AF, Matsumoto K, Murakami Y, El -Hady KA, Mohamed MF, Watanabe H. 550 Facilitatory and inhibitory effects of harmaline on the tryptophan -induced 5-hydroxytryptamine 551 syndrome and body temperature changes in pargyline -pretreated rats. Jpn J Pharmacol 1996; 552 72(1): 39-47. 553 38. Ozalp E, Soygur H, Cankurtaran ES, et al. Sertraline, escitalopram and tianeptine related 554 abnormal movements but not with bupropion: A case report. Progress in Neuro -555 Psychopharmacology & Biological Psychiatry 2006; 30(7): 1337-9. 556 557 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Fig 1. Establishment and phenotypic characterization of the Harmaline - induced mouse tremor model. (A) Schematic paradigm for detection of harmaline-induced essential tremor. ( B-C) Representative s pectrograph calculated from 30 -minute tremor measurements based on head vibration respectively in control group ( B) and harmaline -exposure group ( C). ( D-E) Representative raw data of weight measurement using tremor detector based on head vibration in control group (D) and harmaline-exposure group (E). (F-G) Averaged power spectrum from data in (B) and (C). n = 8 for control group (F) and n = 3 for harmaline-exposure group (G). (H) Comparison of averaged peak power from 8 to 20 Hz between control (n = 8) and harmaline-exposure (n = 3) group. Data presented as mean ± SEM , Ctrl, -59.34 ± 0.3005; Harmaline, - 45.43 ± 0.6715. ****P < 0.0001, two-tailed Students’ t-test. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Fig 2. Transcriptomic and proteomic workflow and preliminary screening of candidate genes involved in harmaline -ET. (A) Schematic workflow of RNA-seq and LC-MS/MS and joint bioinformatic analysis based on mouse cerebellar cortex. After vehicle or harmaline treatment, mouse cerebellar cortexes were dissected, followed by independent mass spectrometry and bulk RNA-Seq and combined analysis. ( B) 5661 proteins were identified from proteomic analysis, including 469 up -regulated proteins and 271 down - regulated proteins. (C) 35125 genes were identified in transcriptomic analysis, including 614 up-regulated genes and 18 down-regulated genes. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Fig 3. Integrated informatic analysis based on transcriptomic and proteomic data. (A) Integrated analysis revealed 5194 correlated genes with 19 significant upregulation. (B) The profile and expression richness of 19 genes based on clustering analysis . (C) GO analysis of e nrichment correlation between proteome and transcriptome. (D) Combined GO analysis of enrichment correlation between proteome and transcriptome. ( E) Combined KEGG analysis of enrichment correlation between proteome and transcriptome. (F) KEGG analysis of enrichment correlation between proteome and transcriptome. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Fig 4. Regulation of SERT expression and 5 -HT content in harmaline - induced tremor. (A-C) Representative blots (A) and quantification of SERT (B) and GAT (C) from cerebellar cortex in mice following vehicle or harmaline treatment as indicated. GAPDH was used as a cytoplasmic protein control. The protein level presented as mean ± SEM. For GAT, P = 0.1039, no significance between two groups (ctrl, 0.4331±0.0473; harmaline, 0.5471±0.268). For SERT, **P < 0.01 ( ctrl, 0.7003±0.0192; harmaline, 1.008±0.426). 3 independent experiments. (D) ELISA quantification of 5 -HT content in mouse cerebellar cortex 30 minutes after vehicle/harmaline treatment. Based on the standard curve, the 5-HT concentration: ctrl, 0.3378±0.0047; harmaline, 0.3152±0.0016. **P < 0.01. 3 independent experiments. (E) Schematic diagram for 5-HT sensor transfection and fiber photometry. ( F) Quantification analysis of 5 -HT sensor fluorescent fluctuation following harmaline treatment. n = 4 mice. (G) Representative heatmap of time-lapse recording of 5-HT sensor fluorescence. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint Fig 5. The cellular and molecular mechanisms underlying harmaline - induce tremor. (A) Schematic paradigm for behavioral experiments targeting SERT inhibition. (B) Representative raw data of head vibration using tremor detector in vehicle, harmaline with or without DSP-1053 treatment group. (C-D) Averaged power spectrum from data in ( B). n = 8 for control group , n = 3 for harmaline-exposure group and n =3 for harmaline with DSP -1053 group. (D) Comparison of averaged peak power from 8 to 20 Hz among control (n = 8) , harmaline (n = 3) and harmaline with DSP -1053 group. Data presented as mean ± SEM, Ctrl, -48 ± 0.4252; Harmaline, -39.82 ± 0.6153; Harmaline+DSP- 1053, -58.6±0.2877. ****P < 0.0001, one-way ANOVA. (E-G) Representative spectrograph calculated from 30-minute tremor measurements based on head vibration respectively in control group , harmaline group and harmaline+DSP- (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint 1053 group . (H) Representative traces of spontaneous postsynaptic current (sPSC) on cultured primary cerebellar neurons at DIV10. ( I-J) Amplitude quantification of sPSC. (K-L) Frequency quantification of sPSC. n = 4 cells. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted March 20, 2024. ; https://doi.org/10.1101/2024.03.18.585649doi: bioRxiv preprint