Chronic alcohol intake elicits distinct multi-omic profiles in the liver versus skeletal muscle of mice

29 Alcohol-related liver disease and alcohol-related myopathy are widespread consequences of 30 chronic alcohol use. However, understanding of the associated molecular mechanisms and 31 effective treatments remains limited. To address this, we employed multi -omics to uncover 32 molecular blueprints of liver versus skeletal muscle responses to chronic alcohol exposure, 33 using a pre -clinical mouse model showing signs of alcohol -related liver dysregulation 34 (diminished liver phosphatidylcholine-to-phosphatidylethanolamine lipid ratio) and alcohol-35 related myopathy (reduced muscle mass and strength). We found that the liver was more 36 sensitive to chronic alcohol than muscle across the transcriptome, proteome and metabolome 37 levels, but both tissues were equally sensitive at the lipidome level. The liver displayed an 38 extensive and multi -layered metabolic molecular profile, while muscle was associated with 39 upregulated inflammatory and matrisome responses and impaired mitochondrial energetics. 40 Lipidome analyses also revealed a novel potential role for altered phospholipid remodelling in 41 the aetiology of alcohol-related myopathy. Finally, computational drug repurposing identified 42 several compounds for therapeutic targeting of alcohol -induced liver (e.g., saracatinib, 43 GSK126) and muscle (e.g., metformin, trichostatin A) pathophysiology, perhaps working partly 44 to counter metabolic dysregulation. Overall, our study provides a tractable list of therapeutic 45 targets and treatments to help expedite the understanding of and countermeasures against 46 alcohol-related liver disease and alcohol-related myopathy in humans. 47 48

49 alcohol; liver; skeletal muscle; multi-omics; transcriptomics; proteomics; metabolomics; 50 lipidomics. 51 52 53 54 55 56 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 3

57 Alcohol use remains widespread throughout society. In 2016, nearly a third of the world’s 58 population were current alcohol drinkers (i.e., any alcoholic drinks in the past 12 months), with 59 the average amount of alcohol consumed globally being ~1.2 standard drinks daily1. By 2030, 60 the global proportion of people drinking alcohol is expected to reach 50%2. Light-to-moderate 61 alcohol intake has been linked to reduced risk of diabetes and cardiovascular disease 3,4, 62 however the ‘true’ health benefits of small quantities of alcohol remain contentious and are 63 offset by an increased risk of other diseases (e.g., cancer) from modest alcohol consumption5. 64 It is also well established that chronic, heavy alcohol intake is associated with many negative 65 health outcomes. Over 200 health conditions , spanning multiple organ systems , are 66 attributable to alcohol6. Excessive alcohol consumption is a leading contributor to the global 67 burden of disease 7 and a major risk factor for mortality 8. In the US alone, excessive alcohol 68 use contributes to approximately 13% of deaths among 20 -to-64 year olds 9 and carries an 69 annual healthcare cost of $28 billion 10. An ongoing priority of the World Health Organization 70 is, therefore, to protect public health by preventing and reducing harmful use of alcohol11. 71 72 The liver is the main site for alcohol metabolism, making it one of the most afflicted organs by 73 chronic alcohol intake12. Excessive alcohol use can lead to alcohol-related liver disease (ALD), 74 which presents a wide clinical spectrum that includes alcohol-related fatty liver (steatosis), 75 alcohol-related hepatitis, liver cirrhosis and liver cancer13. Alcohol-related fatty liver occurs in 76 up to 90% of heavy alcohol drinkers and, while reversable with abstinence, can progress to 77 more severe liver disease (e.g., cirrhosis) and increase the risk of liver -related mortality14-16. 78 Liver cirrhosis is the 11th leading cause of death worldwide and develops in 10-20% of heavy 79 alcohol users, with 30 -50% of global cirrhosis -related deaths attribut able to alcohol 17,18. 80 However, despite ALD being the most prevalent type of chronic liver disease globally19 and 81 the leading cause of alcohol -specific death in some countries (e.g., the UK20), therapeutic 82 options are still lacking. Abstinence can be challenging for alcohol-dependent individuals and 83 cannot reverse advanced stages of ALD 21,22. Nevertheless, c urrently, there no approved 84 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 4 pharmacologic treatments for ALD18,23,24. 85 86 Another major organ impacted by alcohol is skeletal muscle . Excessive alcohol intake can 87 cause muscle atrophy and weakness, a condition known as alcohol-related myopathy25. 88 Skeletal muscle is the largest tissue in the human body 26 and is central to many key 89 physiological processes , including locomotion and metabolism 27. Alcohol-induced muscle 90 atrophy can therefore have profound negative implications on health and physical function , 91 ultimately reducing quality of life 28. Strength loss can exceed 30% in chronic , heavy alcohol 92 drinkers and is only partially restored with abstinence29. With alcohol-related myopathy 93 affecting 45-70% of heavy drinkers , it is one of the most widespread alcohol-related 94 conditions30. Nevertheless, the exact pathophysiologic mechanisms of alcohol -related 95 myopathy remain incompletely defined and likely multifactorial, with impaired muscle protein 96 synthesis, increased muscle protein degradation, mitochondrial dysfunction, perturbed 97 metabolism, inflammation, oxidative stress and diminished muscle regenerative capacity all 98 purported to contribute 28. Moreover, like ALD, therapeutic strategies for alcohol-related 99 myopathy are limited31. 100 101 The development and progression of ALD and alcohol-related myopathy are likely caused not 102 only by the direct effects of alcohol on liver (ALD) and muscle (alcohol-related myopathy), but 103 also via indirect effects across these two tissues. Indeed, ALD may contribute to alcohol-104 related myopathy, and vice versa, by influencing signalling between the liver and skeletal 105 muscle28,30,32-34. Certain alcohol-induced phenotypes can also be common between the liver 106 and muscle, such as tissue fibrosis13,35. Thus, ALD and alcohol-related myopathy encompass 107 distinct, shared, and interlinked pathophysiologic hallmarks. Understanding similarities and 108 differences in molecular responses of the liver and skeletal muscle to chronic alcohol exposure 109 could, therefore, expedite the development of more targeted therapeutic interventions for both 110 conditions. However, the molecular dynamics of excessive alcohol consumption in the context 111 of liver versus skeletal muscle remain largely unknown. 112 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 5 113 Molecular and translational medicine have been revolutionised by the paradigm shift of biology 114 into a ‘big data’ era. Indeed, modern ‘omics’ technologies now allow for the characterisation 115 of the molecular milieu at an immense resolution36, paving the way for unprecedented medical 116 advances from identifying individuals with undiagnosed disease to guiding personali sed 117 medicine37. The omics cascade includes several layers that bridge genotype to phenotype, 118 from genomics and transcriptomics through to proteomics and metabolomics/lipidomics 38. 119 Research efforts have often focussed on just a single omics layer at a time , overlooking the 120 fact that complex disease pathophysiology is underpinned by disturbances across multiple 121 layers of molecular biology and their interplay39. The integration and interrogation of different 122 omics data types, termed ‘multi-omics’, can improve biological insights by providing a more 123 holistic ‘systems-level’ view of disease aetiology40. In turn, multi-omics offers an unparalleled 124 platform for aiding the discovery and development of new disease diagnostics, prognostics, 125 treatments, and preventative strategies41. 126 127 Here, we harnessed the power of multi-omics to define and compare the molecular blueprints 128 of chronic alcohol use in the liver and skeletal muscle using an established pre-clinical mouse 129 model. Our findings provide valuable insights into the molecular mechanisms underlying 130 alcohol-induced liver and skeletal muscle pathophysiology, and reveal new therapeutic targets 131 and candidate drug options. Consequently, the results from this innovative multi-omics study 132 provide a strong foundation for accelerating the understanding of, and targeted treatments 133 against, ALD and alcohol-related myopathy in humans. 134 135

136 137 Mice that drink alcohol chronically show signs of alcohol-related myopathy and ALD 138 Transcriptomic, proteomic, metabolomic and lipidomic data were generated from the liver and 139 plantarflexor muscles ( gastrocnemius, plantaris, soleus) of adult C57BL/6 female mice that 140 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 6 had consumed either 100% water (Control mice, n = 9) or 80% water + 20% alcohol (ethanol) 141 (Alcohol mice, n = 14) for 34-40 weeks42,43. Mice that consumed alcohol for 34-40 weeks had 142 elevated blood alcohol levels (Control = 0 (3.1) mg.dL-1, Alcohol = 159.3 (120.6) mg.dL-1; P = 143 7.60E-05) and lower total body mass (Control = 28.1 (6.5) g, Alcohol = 25.8 (3.7) g; P = 5.55E-144 03), fat mass (Control = 7.0 (2.1) g, Alcohol = 5.4 (1.9) g; P = 1.73E-02) and lean mass (Control 145 = 18.5 (2.5) g, Alcohol = 17.0 (1.1) g; P = 1.34E-03) compared to controls. Plantarflexor muscle 146 mass (Figure 1A) and in vivo isometric torque (Figure 1B) were lower in alcohol -consuming 147 mice relative to control mice, consistent with alcohol -induced muscle atrophy and weakness 148 (i.e., alcohol -related myopathy). Alcohol-consuming mice also had an elevated ratio of 149 phosphatidylcholine (PC) to phosphatidylethanolamine (PE) lipid content in muscle (marker of 150 muscle insulin resistance 44) (Figure 1C), demonstrating that muscle metabolism may have 151 changed after alcohol intake25. While liver mass was similar between control mice and alcohol-152 consuming mice (Figure 1D), chronic alcohol drinking lowered hepatic PC:PE ratio values 153 (Figure 1E), which is indicative of diminished hepatic membrane potential and a strong 154 indicator of ALD in both pre-clinical models and human patients.45 155 156 Chronic alcohol consumption extensively disrupts liver and muscle transcriptomes 157 Chronic alcohol markedly perturbed the transcriptomes of both liver and muscle, with over 158 1,000 dysregulated genes in e ach case (Figure 2A). However, the liver transcriptome was 159 more sensitive to the effects of chronic alcohol, exhibiting more differentially expressed genes 160 and greater magnitudes of change compared to skeletal muscle (Figure 2A). An overlay of 161 dysregulated genes showed that liver and muscle experience largely unique transcriptomic 162 responses to chronic alcohol (Figure 2B). In the liver, uniquely upregulated genes were 163 enriched with MTA3 transcription factor (TF) targets and glycolysis genes, while uniquely 164 downregulated genes were associated with processes such as cholesterol homeostasis and 165 hypoxia (Figure 2C, Figure S1). In muscle, uniquely upregulated genes largely mapped to 166 inflammatory processes, apoptosis and extracellular matrix processes as well as CRELD1 and 167 LYL1 TF targets, while uniquely downregulated genes were linked to fatty acid metabolism, 168 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 7 oxidative phosphorylation and mitochondrial biogenesis, and included targets of the NR1H2 169 and PPARGC1 TFs (Figure 2C, Figure S1). Fewer genes were commonly dysregulated by 170 chronic alcohol across both tissues, with c ommonly upregulated gene involved in 171 carbohydrate metabolism processes, and commonly downregulated genes related to amino 172 acid metabolism, chromosome -related processes, and targets of the AFF4 and RXRG TFs 173 (Figure 2C, Figure S1). Tissue specificity was also evident among the top genes dysregulated 174 by alcohol, with virtually all top genes in liver (except Arsa, Sox12 and Paqr7) unaffected by 175 alcohol in muscle and, similarly, nearly all top genes in muscle (except Gstk1, Ivd, Pxmp2 and 176 Sertad3) unaffected by alcohol in the liver (Figure 2A). Top liver-specific genes included many 177 related to alcohol and/or cholesterol metabolism (Cyp2d9, Cyp3a41b, Cyp3a44, Cyp51, Idi1, 178 Msmo1, Nsdhl, Rdh11, Tkfc), a glutathione conjugator ( Gstm3; upregulated) and a leptin 179 receptor (Lepr; upregulated). Top muscle-specific genes included those linked to cytoskeletal 180 regulation (Actc1, Tmsb10, Vasp; all upregulated), metabolism of pro-inflammatory mediators 181 (Cyp4f18; upregulated) and mitochondrial energetics ( Coq10a, Mpc1; both downregulated) 182 (Figure 2A). 183 184 Dysregulation of the muscle proteome is minimal following chronic alcohol intake 185 Like the transcriptome, the liver proteome was more sensitive to chronic alcohol than the 186 muscle proteome (Figure 3A). While approximately 600 proteins were dysregulated by alcohol 187 in liver, only 8 proteins were dysregulated in muscle (Figure 3A). Overlaying these signatures 188 revealed a major liver -specific proteome response to chronic alcohol, accompanied by a 189 limited muscle-specific response and few commonly dysregulated proteins (Figure 3B). In 190 liver, uniquely upregulated proteins were involved in processes such as oxidative 191 phosphorylation, lipid metabolism, and mitochondrial translation , while u niquely 192 downregulated proteins were associated with cholesterol homeostasis, glycolysis and amino 193 acid metabolism (Figure 3C). In contrast, proteins uniquely upregulated in muscle were 194 enriched in processes such as coagulation, haemostasis, and the immune complement 195 system (including the direct terminal complement pathway inhibitor, clusterin46, and the 196 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 8 classical and alternative complement pathway modulator, antithrombonin -III47) (Figure 3C). 197 No proteins were uniquely downregulated in muscle ( Figure 3B), neither were there any 198 enriched terms for commonly dysregulated proteins nor enriched TF target sets in any case 199 (Figure 3C). Unsurprisingly, all top -dysregulated proteins in the liver were unaffected by 200 alcohol in muscle (Figure 3A), with most (except Alg3, Fcn1, Galt, Gnaq, Med,16, Psmb9) also 201 being uniquely dysregulated at the gene level in the liver. This included several metabolism-202 related proteins (Cbr3, Dcxr, Idi1, Lpcat3, Mvk, Pfkfb1, Retsat ) and two glutathione 203 conjugators (Gstm1, Gstm3; both upregulated) (Figure 3A). In contrast, top muscle -specific 204 proteins included stress -related chaperones (Bcap29, Clu; both upregulated), blood 205 coagulation regulators (Serpinc1; upregulated) and immune/inflammatory elements (Cd5l, 206 Mbl1; both upregulated) (Figure 3A). 207 208 Transcriptome and proteome responses to chronic alcohol use are more synergistic in 209 the liver compared to muscle 210 With our transcriptomic and proteomic data exhibiting some consistent outcome themes (e.g., 211 quantitative differential patterns and functions of dysregulated features), we further explored 212 the general concordance between gene and protein responses to chronic alcohol in each 213 tissue using ‘threshold-free’ approaches to maximise global biological signal48. We observed 214 a strong degree of agreement in the liver between the differential patterns of features present 215 at both the gene and protein levels (n = 3,303 genes/proteins) (Figure 4A). While a trend for 216 agreement between differential gene and protein patterns was also observed in muscle, th e 217 corresponding signal was much weaker compared to that in the liver (Figure 4A). Similarly, 218 the agreement for global pathway regulation between gene and protein levels was more 219 apparent in the liver than in muscle (Figure 4B). These analyses revealed numerous themes 220 in line with our individual transcriptome and proteome analyses. For example, in muscle, there 221 was a strong upregulation of inflammatory (e.g., ‘neutrophil degranulation’, ‘innate immune 222 system’, ‘platelet aggregation signalling and aggregation’, ‘complement cascade’ ), 223 extracellular matrix and apoptosis pathways. In the liver, there was a unique downregulation 224 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 9 of cholesterol homeostasis pathways. Both liver and muscle showed common downregulation 225 of amino acid metabolism pathways at the gene level, which extended only to the protein level 226 in the liver. Additionally, there were opposing mitochondrial and oxidative phosphorylation 227 responses in the liver compared to the muscle (Figure 4B). 228 229 Metabolome responses to chronic alcohol are less pronounced in muscle than the liver 230 As at the transcriptome and proteome levels, the liver metabolome was more sensitive to 231 chronic alcohol than the muscle metabolome, with six-fold more metabolites dysregulated in 232 the liver compared to muscle (Figure 5A). Overlaying differential metabolites revealed a strong 233 liver-specific response to alcohol , with fewer metabolites either uniquely dysregulated in 234 muscle or commonly dysregulated both tissues (Figure 5B). In the liver, uniquely upregulated 235 metabolites were strongly associated with the glycine, serine and threonine metabolism 236 pathway, while uniquely downregulated metabolites were linked to the carbohydrate 237 metabolism pathway and the organic acid class (Figure 5C). Metabolites downregulated by 238 alcohol in both tissues were predominantly organic nitrogen compounds and metabolites 239 involved in glycerophospholipid metabolism (Figure 5C). However, no enriched terms were 240 identified for metabolites either upregulated in both tissues or uniquely dysregulated in muscle 241 (Figure 5C). Interestingly, despite the general tissue specificity of upregulated metabolites 242 (Figure 5B), liver and muscle shared many top-upregulated metabolites, including 2-deoxy-d-243 glucose, 3-o-beta-d-galactosyl-sn-glycerol, azithromycin, fructose (all carbohydrate related) 244 and quinolinic acid (pyridine related) ( Figure 5A). Acetylcholine was the only top -245 downregulated metabolite in both tissues, with all other top-downregulated metabolites in the 246 liver being metabolites unaffected by alcohol in muscle ( Figure 5A). Conversely, only 4 top-247 downregulated metabolites in muscle were not dysregulated in the liver, namely hecogenin (a 248 terpenoid), cholfenethol (an acaricide), 4-tert-octylphenol monoethoxylate (a fatty alcohol) and 249 beta-asarone (a phenylpropanoid with anti-inflammatory and anti-apoptotic potential49) (Figure 250 5A). Top liver-specific metabolites included the upregulated compound (9cis)-retinal (a retinol-251 related compound), and downregulated compounds such as propionylcarnitine (a fatty acid 252 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 10 ester), bufotalin (a steroid), trans-3-indoleacrylic acid (an indole), hydroxyphenyllactic acid (a 253 phenylpropanoic acid ), beta-alanine and guanidinosuccinic acid (both amino acid -related 254 compounds) (Figure 5A). The identification confidence levels for all analysed metabolites are 255 detailed in Document S1 of the supplementary information. These range from Level 1 (highest 256 confidence), assigned to metabolites definitively identified by matching authentic standards, 257 to Level 4 (low confidence), representing unknown compounds or tentative assignments 258 based on database matches without confirmatory spectral or standard-based validation. 259 260 Liver and muscle lipidomes are equally sensitive to chronic alcohol consumption 261 Unlike all other omics layers studied, the lipidomes of liver and muscle were comparably 262 sensitive to chronic alcohol, with approximately 320 lipids differentially regulated in each case 263 (Figure 6A). Nevertheless, overlaying lipidome profiles revealed large tissue -specific lipid 264 responses to alcohol, although noticeable proportions of commonly dysregulated lipids were 265 also found (Figure 6B). Commonly upregulated lipids were enriched with P -ethanol 266 compounds, while commonly downregulated lipids were enriched with P-choline compounds 267 (Figure 6C). Lipids uniquely upregulated in the liver predominantly belonged to the P-inositol 268 class, whereas those uniquely downregulated in the liver were enriched with sphingolipids 269 (Figure 6C). Lipids uniquely upregulated in muscle strongly mapped to the P -ethanol Amine 270 class, although no enriched classes were identified for lipids uniquely downregulated in muscle 271 (Figure 6C). Among the top-downregulated lipids in muscle were P-choline, P-ethanol and P-272 ethanol Amine compounds that were either upregulated (Pc(36:6)(rep), Pet(16:0/20:3)) or 273 unperturbed (Pc(36:6), Pc(44:12), Dmepe(40:6p)) by alcohol in liver (Figure 6A). Similarly, half 274 of the top-upregulated lipids in muscle were compounds downregulated by alcohol in the liver 275 (Pc(16:0/18:2), Pc(17:0/18:2), Pc(33:2), Pc(36:3), Ps(39:3)) ( Figure 6A). In both liver and 276 muscle, the top -upregulated lipid was Pet(16:0/18:2), a P -ethanol compound and potential 277 alcohol biomarker in blood50 (Figure 6A). 278 279 Multi-omic networks discriminate chronic alcohol use in the liver and in muscle 280 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 11 We also performed discriminative multi-omic network analysis to model complex relationships 281 across the different omic s layers of molecular biology 51 and, in turn, establish further key 282 molecular drivers of alcohol-induced muscle/liver pathophysiology52. In the liver, a multi-omic 283 relevance network of 309 features was deduced, comprising a mix of genes (n = 187), proteins 284 (n = 48), metabolites (n = 19) and lipids (n = 55). The liver relevance network partitioned into 285 4 sub-networks of varying size (Liver C1-C4), each containing at least one feature from each 286 omics type (Figure 7A). Virtually all (n = 301) features of the liver relevance network showed 287 dysregulation by chronic alcohol in the liver (Figure 7A). Hub analyses further defined a list of 288 40 ‘priority’ features across the liver network , including 11 genes, 8 proteins, 6 metabolites, 289 15 lipids (Figure 7A). Many of these upregulated features were related to metabolism and 290 energy homeostasis, including : the genes Ephx1 (lipid metabolism), Pfkm (glycolysis) and 291 Zfp385a (adipogenesis) ( Liver C2 hubs) ; the proteins Ak2 (energy homeostasis), Hadha 292 (mitochondrial beta-oxidation), Htatip2 (redox sensor), Stbd1 (cargo receptor for glycogen) 293 (Liver C1 hubs) and Pc (glucose and lipid synthesis) (Liver C4 hub), and; the metabolites 3-o-294 beta-d-galactosyl-sn-glycerol (carbohydrate metabolism), fucose (carbohydrate class) (Liver 295 C1 hubs), 2 -deoxy-d-glucose (carbohydrate class) and azithromycin (carbohydrate class) 296 (Liver C4 hubs) (Figure 7A). Several upregulated features related to conjugation were also 297 identified among liver hubs, including Ugt1a9 (glucuronidation pathway) (Liver C2 hub) and 298 the reduced glutathione conjugators Gstp1 (Liver C2 hub) and Gstm1 (Liver C1 hub) (Figure 299 8A). Additionally, more than half of liver hub lipids were downregulated P-choline compounds, 300 including Lpc(15:1), Pc(31:0), Pc(32:1e) , Pc(39:7)(rep) (Liver C1 hubs), Pc(16:0p/22:5), 301 Pc(38:4e), Pc(38:6e)(rep) ( Liver C4 hubs), and the sole hub of Liver C3, Pc(36:4p) (Figure 302 7A). Pet(16:0/18:2) (P-ethanol compound), the top-upregulated lipid in liver (Figure 7A), was 303 also found among the hubs of Liver C4 (Figure 7A). 304 305 Compared to the liver, the muscle multi -omic relevance network was much smaller and 306 contained only genes (n = 8; all downregulated by alcohol in muscle) and lipids ( n = 2; both 307 upregulated by alcohol in muscle). The muscle relevance network partitioned into 2 sub -308 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 12 networks (Muscle C1 and C2), each comprising 1 lipid and 4 genes (Figure 7B). Muscle C1 309 contained upregulated genes related to transcription (Tcf19), endocytosis (Tbc1d2b), collagen 310 fibril assembly ( Efemp2) and actin cytoskeleton regulation ( Arhgap45), all centred around 311 Pe(18:0p/22:6)(rep), a downregulated P-ethanol Amide compound and the sole hub feature 312 within Muscle C1 (Figure 7B). Muscle C2 followed a similar topology, with upregulated genes 313 related to NF -Kappa-B inhibition ( Nfkbie), hexose phosphorylation ( Hk3) and actin 314 cytoskeleton dynamics (Arpc1b, Coro1a) all converging on one downregulated ‘hub’ P-ethanol 315 Amide compound, Pe(16:0p/22:6) (Figure 7B). 316 317 Chronic alcohol use associates with omic profiles that are targetable by drugs 318 Given the need to establish new therapeutics against ALD and alcohol-related myopathy, we 319 screened for potential drug treatments in humans based on the omic signatures derived in this 320 study. We specifically focused on transcriptomic and proteomic profiles due to the greater 321 availability of drug-target annotations compared to the metabolome and lipidome. 322 Consequently, a total of 745 compounds from the Drug Signature Database53 and Proteome 323 Drug Atlas54 were predicted to target features dysregulated by chronic alcohol in liver and/or 324 muscle. Consistent with the general tissue-specificity observed in transcriptome and proteome 325 responses to alcohol, virtually all (n = 742) compounds were predicted to target features 326 uniquely dysregulated by chronic alcohol in either liver or muscle. This trend remained evident 327 even when considering the top drug compound predictions based on significance (Figure 8). 328 Among the predicted compounds for liver -specific features, several were reproducibly 329 identified across transcriptome and proteome inputs . These included Saracatinib and 330 GSK126, which may potentially hold anti -liver fibrotic properties 55,56 (Figure 8). Predicted 331 compounds for features uniquely dysregulated by alcohol in muscle included metformin, 332 trichostatin A and retinoic acid, each of which may positively impact the phenotype observed 333 with alcoho l-related myopathy57-59 (Figure 8). Only three compounds were predicted for 334 features commonly upregulated by chronic alcohol in both liver and muscle, namely ketamine, 335 nicotinic acid and MERCK544, while no compounds were predicted for features commonly 336 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 13 downregulated by chronic alcohol in both tissues (Figure 8). However, several compounds, 337 such as estradiol and lycorine, were predicted to target unique responses in both tissues 338 (Figure 8), indicating that some compounds may impact both liver and muscle after alcohol 339 intake, but through distinct mechanisms . T he presence of carcinogenic factors among 340 predicted compounds (e.g., AFLATOXIN B1, benzo[a]pyrene) (Figure 8) further underscores 341 the toxic effects of chronic alcohol on both liver and skeletal muscle tissues. 342 343

344 Alcohol use is a leading cause of morbimortality worldwide7,8, such that minimising its harmful 345 effects is a major public health priority of the World Health Organization 11. Among the most 346 serious and prevalent consequences of excessive alcohol consumption are ALD and alcohol-347 related myopathy17,30. However, therapeutic options are currently lacking, underscoring the 348 urgency for continued research into the molecular blueprints of alcohol -induced liver and 349 muscle pathophysiology. Multi-omics has emerged as a cutting-edge frontier for accelerating 350 molecular understanding of and developing countermeasures against complex diseases40,41. 351 Therefore, we undertook the first multi-omic screen of chronic alcohol signatures in liver versus 352 skeletal muscle. Our findings reveal ed that liver and muscle are characteri sed by largely 353 unique molecular profiles in the context of chronic , excessive alcohol consumption. 354 Consequently, liver and muscle from mice that consumed alcohol were associated with mostly 355 divergent therapeutic targets and candidate pharmacologic interventions. These results have 356 important clinical implications for developing optimal strategies to ameliorate ALD and alcohol-357 related myopathy both individually and synergistically. 358 359 Mechanistic investigations of chronic alcohol drinking in people come with significant ethical 360 and technical challenges 60, highlighting the importance of pre -clinical studies that mimic 361 human responses to excessive alcohol consumption. In the current study, mice that drank 362 20% alcohol daily for 34-40 weeks had lower levels of muscle mass and contractile torque, 363 indicative of alcohol -induced muscle atrophy and weakness 42,43. Gold-standard clinical 364 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 14 markers of ALD include elevated serum levels of A ST and ALT61, which reflect their leakage 365 from hepatocytes upon alcohol-induced injury62. While serum AST and ALT levels were unable 366 to be quantified in the current cohort of mice due to lack of remaining sample, separate work 367 in another cohort of C57BL/6 (same strain as present study) mice that drank 20% alcohol for 368 ~60 weeks found elevated serum AST and ALT levels (unpublished observations), supporting 369 the likelihood that liver damage is already developing or progressing by the 34 -40 week 370 timepoint in our model. Moreover, in a recent large cohort study in humans, people with lower 371 hepatic levels of GOT1 (AST) and GPT (ALT) experienced more severe ALD 63, suggesting 372 that declines in AST and ALT levels in liver tissue itself may be indicative of ALD development. 373 Consistent with this, mice consuming alcohol in the present study had diminished levels of 374 liver Got1 and Gpt (see supplemental data), as well as a lower hepatic PC:PE ratio, which is 375 a hallmark of impaired hepatocyte membrane integrity that leads to progression of steatosis 376 into steatohepatitis64 and is also indicative of ALD in both patients and animal models45. Thus, 377 while we acknowledge that our current work would have benefited from liver histological 378 analysis to confirm the true presence and severity of ALD, the observations above offer some 379 support to the validity of our model for studying alcohol-induced dysregulation of both liver and 380 skeletal muscle at the molecular level. 381 382 The liver metabolises over 90% of absorbed alcohol, making it highly susceptible to the toxic 383 effects of chronic alcohol intake 12,13. Greater molecular disruption by chronic alcohol in the 384 liver compared to muscle is therefore logical. Consistent with this notion, we found that the 385 liver is more sensitive than muscle to chronic alcohol at the levels of the transcriptome, 386 proteome, and metabolome. Widespread changes across the liver transcriptome, proteome 387 and metabolome have also been observed in human ALD models 63,65, though omics work in 388 human alcohol myopathic muscle tissue remains lacking. Conversely, the lipidomes of liver 389 and muscle were found to be equally susceptible to chronic alcohol consumption. It is well 390 established that alcohol causes gross dysregulation of liver lipids45. Moreover, previous human 391 lipidomics research noted that wholesale hepatic lipidome changes associate with alcohol -392 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 15 related liver injury 66. We therefore postulate that liver and muscle lipidomes exhibit similar 393 sensitivity to chronic alcohol due to a relative increase in muscle lipidome sensitivity rather 394 than a decline in liver lipidome sensitivity. This could be due to several factors. First, it is 395 plausible that remodelling of the muscle lipidome after chronic alcohol intake resulted from 396 lipid “spillover” from other sources (e.g., adipose tissue ) that were then transported into 397 muscle. Second, it is possible that alcohol directly modified muscle lipids. Indeed, many 398 upregulated lipids in muscle (and liver) w ere phosphatidylethanols, which only form in the 399 presence of ethanol 50. Lastly, the lipidome provides a snapshot of processes across the 400 genome, transcriptome, and proteome 67, so cumulative changes across those layers could 401 predispose muscle to gross lipidome remodelling. Taken together, such data indicate that lipid 402 composition in skeletal muscle is highly influenced by chronic alcohol consumption. 403 404 Another interesting observation made in the present study was that there were few individual 405 protein abundance changes in muscle after chronic alcohol intake, even though a substantial 406 number of genes were found to be dysregulated. This lack of individual protein change s 407 relative to gene changes may be due to a ‘bulk’ impairment in translational efficiency, similar 408 to what occurs in other long -term models of muscle decline such as ageing68. Another 409 plausible explanation is that muscle exhibits a ‘biphasic’ adaptive temporal response to chronic 410 alcohol intake akin to other chronic stimuli like exercise training 69, such that the molecular 411 signatures captured at our end time point reflect the onset of ‘later’ molecular mechanisms of 412 alcohol-related myopathy which are initially prominent at the transcriptome level. Alternatively, 413 it may be that mRNA processing undergoes ‘fine -tuning’ in alcohol myopathic muscle once 414 protein levels converge on a ‘set point’ as rates of muscle decline subdue over time70. In this 415 scenario, larger changes in key mRNA could gradually couple to subtle protein-level changes 416 that still contribute to alcohol-related myopathy. Supporting this notion, we observed a strong 417 concordance between the transcriptome and proteome in whole-pathway changes related to 418 muscle maintenance/function after chronic alcohol intake (Figure 5). Thus, heavy alcohol 419 drinking is characterised by a more ‘global’ and persistent ‘steady -state’ mRNA -protein 420 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 16 response in the liver than in skeletal muscle. This observation may reflect temporal differences 421 in the onset and progression of ALD versus alcohol-related myopathy which might, in turn, 422 have implications for developing optimal therapeutics for each tissue. 423 424 Arguably the most pertinent finding in the current study was that the molecular profiles induced 425 in the liver and muscle by chronic alcohol are mostly unique to one another, implying that 426 largely distinct mechanisms characterise ALD and alcohol-related myopathy. The molecular 427 profile of the liver was mainly defined by an extensive and multi-layered metabolic remodelling 428 signature, encompassing changes related to lipid, carbohydrate and glucose/glycogen 429 metabolism that were not observed in muscle. Extensive metabolic reprogramming has also 430 been witnessed in human ALD liver tissue omics65, and our findings extend to suggest that 431 liver and muscle experience distinct metabolic reprogramming events due to chronic alcohol 432 drinking. Widespread metabolic dysregulation is a hallmark feature of alcohol-induced liver 433 damage. Indeed, the majority of absorbed alcohol is oxidised in the liver to acetate , during 434 which aberrant changes in hepatic metabolism across glycolytic, gluconeogenic and fatty acid 435 pathways occur, exacerbating the ALD phenotype71,72. Congruent with liver transcriptomics in 436 ALD patients73, cholesterol homeostasis was strongly downregulated in liver the current work. 437 Moreover, we found that Cyp3a family members involved in metabolism were among the 438 genes most significantly downregulated by alcohol in liver, marrying observations of 439 decreased hepatic CYP3A4 levels in human omics ALD research.73,74 440 441 Molecular profiles related to mitochondrial translation and oxidative phosphorylation were also 442 noted to be uniquely upregulated by alcohol in the liver herein, supporting the notion that 443 enhanced mitochondrial respiration in the liver due to alcohol consumption is associated with 444 greater injury and damage75. Transcriptomics data in ALD patients also points to elevated 445 oxidative phosphorylation gene expression in peripheral blood. 76 Elevated hepatic 446 mitochondrial respiration with chronic alcohol may reflect an adaptive response to enhance 447 liver alcohol metabolism or could signify chronically ‘overworked’ mitochondria. The latter 448 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 17 would lead to an increased mitochondrial reactive oxygen species (ROS) production77, which 449 is implicated as a key mechanism in ALD 78. Consistent with this , Glutathione S-transferase 450 (GST) enzymes (Gstm1, Gstm3, Gstp1) were among the top uniquely upregulated molecules 451 in the liver of mice that drank alcohol. GSTs help protect against cellular damage by 452 detoxifying ROS and other harmful products 79. Elevated hepatic GSTP1 abundance has 453 recently been uncovered as a strong candidate biomarker of ALD progression in patients,63 454 with meta-analyses also demonstrating that people with GSTM1 and GSTP1 allelic variations 455 hold increased susceptibility to ALD 80. Interestingly, we also noticed Fcn1 among the top 456 upregulated proteins by chronic alcohol specifically in liver , consistent with recent hepatic 457 proteomic observations in patients with severe ALD and possibly linked to inflammation 81. 458 Extending previous studies on liver damage82,83, our data also revealed MTA3 as a promising 459 new metabolism-related therapeutic target of ALD for future mechanistic investigation. 460 461 Compared to ALD, mechanistic investigations of alcohol -related myopathy are limited and 462 mainly restricted to conventional targets of muscle protein synthesis (MPS) and breakdown 463 (MPB)31. Here, in contrast to the liver, skeletal muscle was characterised by a molecular profile 464 consistent with impaired mitochondrial energetics. After chronic alcohol consumption, muscle 465 also displayed a unique molecular pattern primarily reflective of elevated inflammation and 466 cytoskeletal/ECM remodelling. Chronic inflammation and mitochondrial dysfunction can both 467 be deemed central tenets in the aetiology of muscle atrophy and weakness , potentially 468 contributing to alcohol-related myopathy by hindering MPS pathways and/or aggravating MPB 469 pathways84,85. Cytoskeletal/ECM alterations are also a major component of skeletal muscle 470 remodelling86. An upregulated matrisome profile in muscle after chronic alcohol intake has 471 been suggested to reflect a profibrotic phenotype28. Conversely, many matrisome features 472 upregulated by alcohol in the present work may, in fact, be upregulated to maintain or improve 473 muscle mass and function 87,88. Thus, whether these muscle matrisome changes truly 474 represent a profibrotic phenotype , or instead reflect a compensatory mechanism to avoid 475 advanced muscle wasting with prolonged alcohol use , warrants further determination. 476 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 18 Nonetheless, our analysis pinpoints LYL1 ( an inflammatory regulator), PPARGC1a ( a 477 regulator of mitochondrial biogenesis) and CRELD1 ( a matrisome member) transcription 478 factors as key mechanistic targets of alcohol-related myopathy. 479 480 It is also intriguing to rationale how the above findings parallel the gross lipidome remodelling 481 observed in skeletal muscle after chronic alcohol intake . On one hand, inflammation and 482 mitochondrial dysregulation are prime events that can influence muscle lipid dynamics 89,90. 483 Yet, the lipidome alterations we found appear underscored by wholesale changes in 484 phospholipids, particularly PC and PE species. Muscle phospholipids have been linked to 485 declines in muscle size and strength 91,92, albeit via unknown mechanisms , and PC:PE 486 dynamics are implicated in promoting mitochondrial dysfunction and inflammation 44,93. 487 Phospholipids are also integral components of the sarcolemma 94, suggesting that muscle 488 PC:PE ratio may be inherently linked to cytoskeletal/ECM remodelling. Thus, our current 489 findings support our recent postulation that remodelling of muscle phospholipid composition 490 may play a role in the aetiology of alcohol-related myopathy95 – potentially through some form 491 of mitochondria -inflammatory-matrisome regulatory circuit. Indeed, n etwork modelling 492 uncovered two mu lti-omic sub-networks in muscle after alcohol intake, both of which had a 493 central ‘hub’ phospholipid connected to cytoskeletal/ECM, inflammation and/or mitochondria-494 related transcript s. Coined the ‘lipotranscriptome’, lipid regulation of the transcriptome is 495 emerging as a promising target for the discovery and development of disease diagnostics and 496 therapeutics96. Alcohol-induced c hanges in muscle lipotranscriptome that centre on 497 phospholipid regulation of mitochondria, inflammation and/or the matrisome m ay, therefore, 498 be among the strongest candidates for understanding and treating chronic alcohol -related 499 myopathy. 500 501 Traditional drug discovery relying on lab -based compound screen s is cumbersome, costly, 502 labour-intensive and high-risk97. Omics circumvents many of these barriers to accelerate the 503 drug discovery process by enabling in silico drug repurposing on an unprecedented scale 98. 504 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 19 Exploiting this approach, we used human drug signature databases to computationally predict 505 a refined list of compounds that target molecular profiles of chronic alcohol-related liver and/or 506 muscle. While the interpretation of all predicted compounds is beyond the scope of this 507 discussion, this list provides a useful tool to expedite future hypothesis-driven work on ALD 508 and alcohol -related myopathy therapeutics. For example, among the most prevalent 509 compounds predicted to target the unique molecular profile of chronic alcohol -related liver 510 were saracatinib and GSK126. Saracatinib has been reported to attenuate liver fibrosis by 511 preventing activation of hepatic satellite cell activation56,99, while GSK126 also appears to have 512 anti-fibrotic liver properties55 and decreases liver fatty acid content100. These two compounds 513 may thus offer promising avenues for exploration in the context of treating ALD. We also 514 identified metformin and trichostatin A as strong candidate therapeutics for alcohol-related 515 myopathy. Indeed, trichostatin A has been shown to ameliorate muscle atrophy induced by 516 unloading58 and to reduce muscle fatty acid infiltration 101. While the muscle therapeutic 517 potential of metformin is somewhat equivocal, evidence indicates that it could help counter 518 muscle decline in pathological scenarios, as would be excessive chronic alcohol drinking, 519 potentially by normalising mitochondrial dysfunction or disturbed energy metabolism 59. 520 Further, compounds that target the molecular profiles of both alcohol-related muscle and liver 521 might represent the most viable concurrent treatments for ALD and alcohol-related myopathy. 522 Notably, MERCK544 may be a viable compound for dual -therapeutic targeting of ALD and 523 alcohol-related myopathy by mitigating 11β-HSD1-dependent metabolic dysregulation related 524 to both liver and muscle102-104. 525 526 In summary, we undertook the first multi -omic screening of liver versus skeletal muscle 527 responses to chronic alcohol use. Our results provide several new insights into therapeutic 528 targets of and pharmacologic interventions for ALD and alcohol -related myopathy. 529 Nonetheless, comprehensive follow -up investigations remain crucial to validate these 530 mechanistic lines of inquiry and to assess the true efficacy of our predicted drug compounds 531 for mitigating alcohol-induced liver and muscle maladaptations. Such work should include 532 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 20 tracking the time course of multi-omic responses to chronic alcohol in both sexes and 533 incorporate a wider range of physiological and morphological readouts (e.g., liver histological 534 analysis to assess stage of liver injury), to provide a more complete picture of the mechanisms 535 underpinning the onset and progression of ALD and alcohol -related myopathy. Thorough 536 examination of predicted drug compounds , including optimal dosing, delivery strategies and 537 safety is also warranted , at both the individual tissue and cross -tissue levels. Overall, our 538 current findings provide a strong benchmark for expediting the mechanistic understanding of 539 ALD and alcohol-related myopathy in humans and may help accelerate the development of 540 optimal personalised countermeasures. 541 542

543 544 Experimental overview 545 The mice and experimental procedures used are as previously reported 42,43. Briefly, female 546 C57BL/6 mice obtained from Jackson Laboratory were aged to 23 -28 weeks and then 547 randomly assigned access to either a 100% water bottle (Control mice; n = 9) or a bottle 548 containing 80% water + 20% alcohol (ethanol) (Alcohol mice; n = 14) for 34-40 weeks. Female 549 mice were used as it has been reported that females are more susceptible to alcohol -related 550 myopathy and ALD105-107. Alcohol-consuming mice were acclimatised to alcohol by increasing 551 alcohol concentration in 5% increments from 0% to 20% (w/v) across two weeks. All mice 552 were supplied standard rodent chow ad libitum throughout. At study completion, body 553 composition was assessed using the Bruker Minispec NMR Analyzer (LF50 Series, model mq 554 7.5), in vivo plantarflexor muscle isometric torque using a servomotor system (Model 300C -555 LR and 701C; Aurora Scientific, Aurora, Ontario, Canada), and blood alcohol concentration 556 using an AM1 alcohol analyser (Analox Instruments Ltd.). Mice were euthanised under 557 anaesthesia (2-to-3% isoflurane) by exsanguination followed by cervical dislocation, then 558 plantarflexor muscles and liver harvested, weighed, snap frozen in liquid nitrogen and stored 559 at -80C until further analyses. For the purposes of this work, a composite of blood alcohol 560 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 21 concentration, body composition, and muscle mass and torque data from two previous 561 publications has been included 42,43. Experimental procedures were approved by the Ohio 562 University Animal Care and Use Committee. All methods were performed in accordance with 563 the relevant guidelines and regulations. The study is reported in accordance with ARRIVE 564 guidelines. 565 566 Omics data generation 567 Harvested liver and muscle was shipped on dry ice to BGI Americas (San Jose), where tissue 568 samples were processed, and omics data generated. For transcriptomics, strand -specific 569 (second strand cDNA synthesis with dUTP) 100 bp paired -end reads were generated from 570 extracted RNA using the DNBseq platform. For proteomics, a label-free quantitative approach 571 was undertaken using nano flow HPLC (Ultimate 3000) followed by Orbitrap Eclipse Tribrid 572 Mass Spectrometer (Thermo Fisher Scientific, USA). A spectral library was constructed via 573 data-dependent acquisition using the MS2-based method, using a fractionated composite of 574 all digested samples. Data -independent acquisition was subsequently performed on each 575 sample via a high -resolution full mass spectrometry scan followed by two data -independent 576 acquisition segments. Raw mass spectrometer files were input into MaxQuant (v1.5.3.30; 577 https://www.maxquant.org/) for identification and quantification against the mouse database, 578 with identified peptides that satisfied a false discovery rate ≤ 1% used when constructing the 579 final spectral library. For metabolomics and lipidomics, features were extracted from tissue 580 samples using solvent-based precipitation, with feature separation and detection performed 581 using a UPLC I -Class Plus (Waters, USA) tandom Q Exactive high resolution mass 582 spectrometer (Thermo Fisher Scientific, USA), utilising AQUITY UPLC BEH C18 and Amide 583 columns for metabolites and a CSH C18 column for lipids (Waters, USA). In any case, QC 584 samples were produced by pooling equal volumes of prepared supernatant (10L) from each 585 sample. Off-line mass spectrometry data were subsequently input into Compound Discover 586 (v3.3; Thermo Fisher Scientific, USA) for metabolite peak extraction and identification (using 587 BGI metabolome, mzcloud and chemspider databases as reference) or LipidSearch (v4.1; 588 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 22 Thermo Fisher Scientific, USA) for lipid peak extraction and identification. 589 590 Omics data pre-processing 591 For transcriptomic data, reads were cleaned using SOAPnuke software (v2.2.1; 592 https://github.com/BGI-flexlab/SOAPnuke) to filter out adaptor sequences, contamination and 593 low-quality reads108. The clean reads were then aligned to the mouse reference transcriptome 594 using Kallisto (v0.46.1 , with bias correction ; https://pachterlab.github.io/kallisto/)109. The 595

transcriptome was compiled from the Ensembl Release 109 mouse reference 596 genome (primary assembly) and associated transcript annotations using GFFread (v0.12.7; 597 https://github.com/gpertea/gffread)110. Gene counts were inferred from transcript abundance 598 estimates scaled to library size using the tximport R package (v1.28.0; 599 https://doi.org/doi:10.18129/B9.bioc.tximport)111. Lowly expressed genes were then removed 600 using the limma R package (v3.56.2; https://doi.org/doi:10.18129/B9.bioc.limma) filterByExpr 601 function (with tissue as the grouping factor and ‘min.prop’ set to 1), and counts for remaining 602 genes (n = 14,017) normalised using the limma-voom approach as per developer guidelines 603 for mixed-design studies. For proteomic data, log2 transformation was applied, proteins with 604 more than 50% missing values omitted and the k-Nearest Neighbour Algorithm used to impute 605 missing values, resulting in normalised abundances for n = 3,512 proteins. For metabolomics 606 data, each UPLC column’s Compound Discover output was subject to probabilistic quotient 607 normalisation to obtain relative peak areas, quality control -based robust LOESS signal 608 correction to correct batch effect, removal of features with a coefficient of variation larger than 609 30% based on relative peak area in QC samples, filtering for features with a recognised KEGG 610 ID, and log2 transformation. Normalised metabolite abundances for each column were then 611 aggregated based on mean value to produce a singular metabolomic dataset ( n = 873 612 features). For lipidomic data, the LipidSearch output was subject to removal of lipids with > 613 50% missing values in QC samples and > 80% missing values in experimental samples, 614 imputation of remaining empty values via the k -Nearest Neighbour Algorithm, probabilistic 615 quotient normalisation to obtain relative peak areas, quality control -based robust LOESS 616 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 23 signal correction to correct batch effect, filtering to remove features with a coefficient of 617 variation larger than 30% based on relative peak area in QC samples, and log2 transformation, 618 leading to normalised abundances of n = 744 lipids. Prior to log2 transformation, percent 619 content of PC and PE lipid classes in each sample (sum of relative peak values for a given 620 class in the sample, divided by total of all relative peak values in the sample) were also 621 calculated and used to quantify muscle and liver PC:PE ratios. 622 623 Statistics 624 Statistical evaluation of end-point data: End-point variables (total body mass, lean mass, fat 625 mass, plantarflexor muscle mass, plantarflexor muscle torque, liver mass, liver-to-body mass 626 ratio, plantarflexor muscle PC:PE ratio, liver PC:PE ratio, blood alcohol concentrations) were 627 compared between alcohol -consuming mice and control mice on a per -tissue basis using 628 either the two -tailed Student’s independent t-test (when normal distribution and equal 629 variance), the two -tailed Welch’s independent t-test (when normal distribution but unequal 630 variance) or the Mann -Whitney U test (when non -normal distribution). Analyses were 631 conducted in R (v4.3.1; https://www.r-project.org/), with statistical significance accepted when 632 P ≤ 0.05. Unless otherwise stated, in-text descriptive statistics are presented as median (inter 633 quartile range). 634 635 Omics differential expression analysis: Differential analysis was undertaken at the level of 636 each individual omic strand using the limma R package (v3.56.2; 637 https://doi.org/doi:10.18129/B9.bioc.limma)112. Briefly, linear mixed effects models were fitted 638 with condition (Alcohol, Control) as a ‘fixed’ effect and mouse ID as a ‘random’ effect, utilising 639 the developers’ recommended duplicateCorrelation approach. The empirical Bayes method 640 was then used to calculate moderated t-scores113, and comparisons between Alcohol and 641 Control mice per tissue extracted. For each omics layer, features with a Benjamini-Hochberg 642 adjusted P-value ≤ 0.1 were defined as being differentially regulated by chronic alcohol. 643 Differentially regulated features were then overlaid to identify features commonly or uniquely 644 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 24 dysregulated by chronic alcohol in liver versus muscle. 645 646 Omics over-representation analysis of pathways, TF targets and molecular classes: Features 647 commonly or uniquely regulated by chronic alcohol in the liver versus muscle were subject to 648 over-representation analyses of pathways, TF targets and molecular class es using the 649 clusterProfiler R package (v4.8.3; https://doi.org/doi:10.18129/B9.bioc.clusterProfiler) enrichr 650 function114. For transcripts and proteins, analyses were performed against MSigDB 651 (v2023.1)115 mouse Molecular Hallmark, Reactome Pathway and TF target (GTRD) gene sets. 652 Default values for minGSSize and maxGSSize arguments were used, except for GTRD sets, 653 where maxGSSize was set unbounded. For metabolites, analyses were performed against 654 final class, super pathway and sub pathway sets as assigned during metabolite identification 655 (minGSSize = 3, maxGSSize unbounded). For lipids, analysis was performed against lipid 656 main class sets as assigned during lipid identification (minGSSize = 3, maxGSSize 657 unbounded). For each omic strand, the corresponding background list contained all annotated 658 features utilised during differential testing. Over-represented sets were selected as those with 659 a Benjamini -Hochberg corrected P-value ≤ 0.05 that were enriched for at least 2 given 660 features. 661 662 Rank-based analysis across the transcriptome and proteome: The rank-rank hypergeometric 663 overlap method 48 was employed (via the RRHO2 R package, v1.0; 664 https://github.com/RRHO2/RRHO2) to decipher the general degree of correspondence 665 between differential gene expression and protein abundance patterns in the liver and in 666 muscle. The algorithm was applied to unique features present at both the gene level and 667 protein level (n = 3,303 genes/proteins), with features ranked by t-score. Gene set enrichment 668 analysis was also employed as a rank -based method to elucidate global pathway regulation 669 at the gene level and protein level in the liver and in muscle. These analyses were performed 670 using the clusterProfiler R package (v4.8.3; 671 https://doi.org/doi:10.18129/B9.bioc.clusterProfiler) GSEA function against MSigSB Molecular 672 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 25 Hallmark and Reactome Pathway gene sets as above (default minGSSize and maxGSSize 673 argument values). In each case, the algorithm was applied to unique features present at both 674 the gene level and protein level ( n = 3,303 genes/proteins), with features ranked by t-score 675 and enrichment defined at the Benjamini-Hochberg corrected P-value ≤ 0.1 level. 676 677 Multi-omic relevance network analysis: Features from all omics layers were integrated using 678 multi-omic correlation network analysis with the Mixomics R package (v6.24.0; 679 https://doi.org/doi:10.18129/B9.bioc.mixOmics) DIABLO method52. Multiblock integration with 680 projection to latent structure models with discriminant analysis was performed, treating each 681 omic strand as a block and condition as the discriminator. A design matrix was used to 682 maximise the strength of relationships between blocks. Relevance networks were extracted 683 for strongly correlated features across omics layers (|correlation coefficient| > 0.85). 684 Component 1 was chosen for association estimates in the liver and component 2 for muscle, 685 based on inspection of individual sample plots and loading weights. Sub -networks of each 686 relevance network were determined via multi -level community analysis using the igraph R 687 package (v1.5.1; https://r.igraph.org/)116, with |correlation coefficient| used as edge weight. 688 Highly connected ‘hub’ features of each sub -network were defined with an eigenvector 689 centrality score > 0.7. Networks were visualised using Cytoscape (v3.10.0 ; 690 https://cytoscape.org/)117. 691 692 Omics-driven drug prediction analysis: Transcriptome and proteome features commonly or 693 uniquely dysregulated by chronic alcohol in the liver versus muscle were further subjected to 694 over-representation analyses against human Drug Signature Database53 and Proteome Drug 695 Atlas54 sets using the Enrichr online web tool (https://maayanlab.cloud/Enrichr/)118. For each 696 omic strand, the background list contained all annotated features utili sed during differential 697 testing. Ortholog conversions were handled implicitly as part of the built-in functionality of the 698 Enrichr tool . Over-represented sets were selected as those with a Benjamini -Hochberg 699 corrected P-value ≤ 0.05 that were enriched for at least 2 given features. 700 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 26 701 Data availability 702 Raw transcriptomic data from current study is deposited in the Sequence Read Archive 703 repository (https://www.ncbi.nlm.nih.gov/sra) with links to BioProject ID PRJNA1274880. Raw 704 proteomic data from current study is deposited in the EBI Proteomics Identification Database 705 repository ( https://www.ebi.ac.uk/pride/) under the dataset identifier PXD064997. Raw 706 metabolomic and lipidomic data are deposited in the EBI MetaboLights repository 707 (https://www.ebi.ac.uk/metabolights/) under the study identifie r MTBLS12616. Underlying 708 values of the data reported in the manuscript itself are provided as supplementary material in 709 Document S1. 710 711 Author contributions 712 Conceptualisation: C.R.G.W. & C.W.B.; Mouse experimental procedures: M.S.G., A.M.B., 713 S.E.M. & C.W.B.; Data analysis: C.R.G.W.; Figures and visualisation: C.R.G.W.; Manuscript 714 original draft: C.R.G.W.; Manuscript review, editing and approval: C.R.G.W., M.S.G., A.M.B., 715 S.E.M., N.J.S., B.C.C. & C.W.B.. 716 717

718 C.R.G.W. and C.W.B. acknowledge support from the UK Research and Innovation Doctoral 719 Career Development Fund scheme, which helped to fund this work. This work was also 720 funded, in part, through the University of Exeter’s Project ADA (Accelerating Data Science and 721 Artificial Intelligence). N.J.S. acknowledges the support of the Osteopathic Heritage 722 Foundation through funding for the Osteopathic Heritage Foundation Ralph S. Licklider, D.O., 723 Research Endowment in the Heritage College of Osteopathic Medicine. B.C.C. acknowledges 724 the support of the Osteopathic Heritage Foundation through funding for the Osteopathic 725 Heritage Foundation Harold E. Clybourne, D.O., Endowed Research Chair in the Heritage 726 College of Osteopathic Medicin e. C.W.B. acknowledges the support of the Osteopathic 727 Heritage Foundation through funding for the Osteopathic Heritage Foundation Ralph S. 728 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 27 Licklider, D.O., Endowed Faculty Fellowship in the Heritage College of Osteopathic Medicine. 729 This work was also supported, in part, by grants from the National Institutes of Health 730 (R01AG044424 & R01AG067758 to B.C.C.). The authors thank Thomas Krauss for assisting 731 with revisions. 732 733 Conflict of interests 734 B.C.C. reports that he has received grants in the past 5 -years related to muscle health from 735 the National Institutes of Health, Astellas Global Development Inc, RTI Solutions, Myolex, and 736 NMD Pharma, that he has received personal fees from Regeneron Pharmaceuticals and the 737 Gerson Lehrman Group, that he has received grants from OsteoDx Inc not related to the 738 submitted work, and that he also serves as co-founder and scientific director for OsteoDx Inc. 739 The other authors declare that they have no conflicts of interest. 740 741 Figure Legends 742 743 Figure 1: Indicators of alcohol -related myopathy and ALD following chronic alcohol 744 consumption. Boxplots illustrate plantarflexor muscle mass (Panel A), plantarflexor muscle 745 peak isometric contractile torque (Panel B), plantarflexor muscle PC:PE ratio (Panel C), liver 746 mass (Panel D) and liver PC:PE ratio (Panel E) in alcohol -consuming mice versus control 747 mice. Analyses via two -tailed Student’s/Welch’s independent t-test or Mann-Whitney U test, 748 as appropriate. *: P ≤ 0.05. 749 750 Figure 2: Liver versus muscle transcriptomic responses to chronic alcohol. Panel A: 751 Volcano plots for differential gene expression analysis (Alcohol versus Control) in the liver and 752 muscle. Red and blue shading denote significant upregulation and downregulation, 753 respectively (adjusted P ≤ 0.1). Annotated genes are those ranked in top 10 754 upregulated/downregulated based on t-score. Panel B: Venn diagrams illustrating degree of 755 overlap between genes upregulated/downregulated by chronic alcohol use in the liver and 756 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 28 muscle. Panel C: Bubble plot depicting results from over -representation analysis of MSigDB 757 mouse Molecular Hallmark and GTRD gene sets for genes commonly/uniquely dysregulated 758 by chronic alcohol use in the liver and muscle (as per panel B). Circle size is proportional to 759 the number of gene hits as a % of the total number of annotated genes for a given overlap. 760 Red and blue shading denote significant over -representation in upregulated and 761 downregulated genes, respectively (adjusted P ≤ 0.05, enriched for ≥ 2 genes). 762 763 Figure 3: Liver versus muscle proteomic responses to chronic alcohol. Panel A: Volcano 764 plots for differential protein abundance analysis (Alcohol versus Control) in the liver and 765 muscle. Red and blue shading denote significant upregulation and downregulation, 766 respectively (adjusted P ≤ 0.1). Annotated proteins are those significant proteins ranked in top 767 10 upregulated/downregulated based on t-score. Panel B: Venn diagrams showing degree of 768 overlap between proteins upregulated/downregulated by chronic alcohol use in the liver and 769 muscle. Panel C: Bubble plot depicting results from over -representation analysis of MSigDB 770 Molecular Hallmark and Reactome Pathways gene sets for proteins commonly/uniquely 771 dysregulated by chronic alcohol use in the liver and muscle (as per panel B). Circle size is 772 proportional to the number of hits as a % of the total number of annotated genes for a given 773 overlap. Red and blue shading denote significant over -representation in upregulated and 774 downregulated proteins, respectively (adjusted P ≤ 0.05, enriched for ≥ 2 features). 775 776 Figure 4: Global comparison between transcriptomic and proteomic responses to 777 chronic alcohol in the liver and muscle. Panel A: Rank -rank hypergeometric overlap 778 (RRHO) plots illustrating the degree of correspondence between gene level and protein level 779 responses to chronic alcohol use in the liver and in muscle. The lighter the colouring in the 780 lower-left quadrant, the stronger the concordance in upregulation between genes and 781 proteins. The lighter the colouring in the upper -right quadrant, the stronger the concordance 782 in downregulation between genes and proteins. For RRHO analysis, only unique features 783 present at both the gene level and protein level were included (n = 3303 gene/proteins), with 784 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 29 features ranked on t-score. Panel B: Heatmap comparing global pathway regulation by chronic 785 alcohol use in liver and muscle at gene and protein levels. Shown are those MSigDB mouse 786 Molecular Hallmarks and Reactome Pathways that are similarly regulated at gene and protein 787 levels in at least one of the two tissues. Results were derived via gene set enrichment analysis 788 applied to unique features present at both the gene level and protein level ( n = 3303 789 gene/proteins), with features ranked on t-score. Red and blue shading denote significant 790 upregulation (adjusted P ≤ 0.1, normalised enrichment score > 0) and downregulation 791 (adjusted P ≤ 0.1, normalised enrichment score < 0), respectively. G = gene, P = protein. 792 793 Figure 5: Liver versus muscle metabolomic responses to chronic alcohol. Panel A: 794 Volcano plots for differential metabolite analysis (Alcohol versus Control) in the liver and 795 muscle. Red and blue shading denote significant upregulation and downregulation, 796 respectively (adjusted P ≤ 0.1). Annotated metabolites are those significant metabolites 797 ranked in top 10 upregulated/downregulated based on t-score. Panel B: Venn diagrams 798 showing degree of overlap between metabolites upregulated/downregulated by chronic 799 alcohol use in the liver and muscle. Panel C: Bubble plot depicting results from over -800 representation analysis of metabolite final class, super pathway and sub pathway annotations 801 for metabolites commonly/uniquely dysregulated by chronic alcohol use in the liver and muscle 802 (as per panel B). Circle size is proportional to the number of metabolite hits as a % of the total 803 number of annotated metabolites for a given overlap. Red and blue shading denote significant 804 over-representation in upregulated and downregulated metabolites, respectively (adjusted P 805 ≤ 0.05, enriched for ≥ 2 metabolites). 806 807 Figure 6: Liver versus muscle lipidomic responses to chronic alcohol. Panel A: Volcano 808 plots for differential lipid analysis (Alcohol versus Control) in the liver and muscle. Red and 809 blue shading denote significant upregulation and downregulation, respectively (adjusted P ≤ 810 0.1). Annotated lipids are those significant lipids ranked in top 10 upregulated/downregulated 811 based on t-score. Panel B: Venn diagrams illustrating degree of overlap between lipids 812 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 30 upregulated/downregulated by chronic alcohol use in the liver and muscle. Panel C: Bubble 813 plot depicting results from over -representation analysis of main lipid classes for lipids 814 commonly/uniquely dysregulated by chronic alcohol use in the liver and muscle (as per panel 815 B). Circle size is proportional to the number of lipid hits as a % of the total number of annotated 816 lipids for a given overlap. Red and blue shading denote significant over -representation in 817 upregulated and downregulated lipids, respectively (adjusted P ≤ 0.05, enriched for ≥ 2 lipids). 818 819 Figure 7: Integrative multi-omic network modelling of the liver and muscle responses 820 to chronic alcohol. Communities of strongly connected molecular features in the liver (panel 821 A) and muscle (panel B) with chronic alcohol use, as derived via Mixomics DIABLO analyses 822 with downstream correlation network generation (|correlation coefficient| > 0.85, with 823 component 1 used for the liver and component 2 used for muscle). 824 825 Figure 8: Omics-driven repurposing of drug therapeutics for ALD and alcohol -related 826 myopathy. Heatmap includes results from over-representation analysis of human Drug 827 Signature Database (DSigDB) and Proteome Drug Atlas sets for genes/proteins 828 commonly/uniquely dysregulated by chronic alcohol use in the liver and muscle. Shown are 829 the top 10 most significant sets for each common/unique feature permutation. Red and blue 830 shading denote significant over -representation in upregulated and downregulated 831 genes/proteins, respectively (adjusted P ≤ 0.05, enriched for ≥ 2 features). G = enriched when 832 genes used as input, P = enriched when proteins used as input. 833 834 Supplementary Information Files 835 836 Figure S1: Reactome Pathways enriched with genes dysregulated by chronic alcohol 837 in the liver and/or muscle. Bubble plot depicts results from over-representation analysis of 838 MSigDB mouse Reactome Pathway gene sets for genes commonly/uniquely dysregulated by 839 chronic alcohol use in the liver and muscle. Circle size is proportional to the number of gene 840 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 31 hits as a % of the total number of annotated genes for a given overlap. Red and blue shading 841 denote significant over-representation in upregulated and downregulated genes, respectively 842 (adjusted P ≤ 0.05, enriched for ≥ 2 genes). 843 844 Document S1: A Single XLSX file containing underlying values of reported data. 845 846

847 848 1. Collaborators, G.B.D.A. (2018). Alcohol use and burden for 195 countries and 849 territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 850 2016. Lancet 392, 1015-1035. 10.1016/S0140-6736(18)31310-2. 851 2. Manthey, J., Shield, K.D., Rylett, M., Hasan, O.S.M., Probst, C., and Rehm, J. (2019). 852 Global alcohol exposure between 1990 and 2017 and forecasts until 2030: a modelling 853 study. Lancet 393, 2493-2502. 10.1016/S0140-6736(18)32744-2. 854 3. Polsky, S., and Akturk, H.K. (2017). Alcohol Consumption, Diabetes Risk, and 855 Cardiovascular Disease Within Diabetes. Curr Diab Rep 17, 136. 10.1007/s11892 -856 017-0950-8. 857 4. Ronksley, P.E., Brien, S.E., Turner, B.J., Mukamal, K.J., and Ghali, W.A. (2011). 858 Association of alcohol consumption with selected cardiovascular disease outcomes: a 859 systematic review and meta-analysis. BMJ 342, d671. 10.1136/bmj.d671. 860 5. Mitra, S., De, A., and Chowdhury, A. (2020). Epidemiology of non -alcoholic and 861 alcoholic fatty liver diseases. Transl Gastroenterol Hepatol 5, 16. 862 10.21037/tgh.2019.09.08. 863 6. Rehm, J., Baliunas, D., Borges, G.L., Graham, K., Irving, H., Kehoe, T., Parry, C.D., 864 Patra, J., Popova, S., Poznyak, V., et al. (2010). The relation between different 865 dimensions of alcohol consumption and burden of disease: an overview. Addiction 105, 866 817-843. 10.1111/j.1360-0443.2010.02899.x. 867 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 32 7. Rehm, J., Gmel, G.E., Sr., Gmel, G., Hasan, O.S.M., Imtiaz, S., Popova, S., Probst, 868 C., Roerecke, M., Room, R., Samokhvalov, A.V., et al. (2017). The relationship 869 between different dimensions of alcohol use and the burden of disease -an update. 870 Addiction 112, 968-1001. 10.1111/add.13757. 871 8. Jayasekara, H., English, D.R., Room, R., and MacInnis, R.J. (2014). Alcohol 872 consumption over time and risk of death: a systematic review and meta -analysis. Am 873 J Epidemiol 179, 1049-1059. 10.1093/aje/kwu028. 874 9. Esser, M.B., Leung, G., Sherk, A., Bohm, M.K., Liu, Y., Lu, H., and Naimi, T.S. (2022). 875 Estimated Deaths Attributable to Excessive Alcohol Use Among US Adults Aged 20 to 876 64 Years, 2015 to 2019. JAMA Netw Open 5, e2239485. 877 10.1001/jamanetworkopen.2022.39485. 878 10. Sacks, J.J., Gonzales, K.R., Bouchery, E.E., Tomedi, L.E., and Brewer, R.D. (2015). 879 2010 National and State Costs of Excessive Alcohol Consumption. Am J Prev Med 49, 880 e73-e79. 10.1016/j.amepre.2015.05.031. 881 11. Organization, W.H. (2023). Alcohol action plan 2022-2030. 882 12. Hyun, J., Han, J., Lee, C., Yoon, M., and Jung, Y. (2021). Pathophysiological Aspects 883 of Alcohol Metabolism in the Liver. Int J Mol Sci 22. 10.3390/ijms22115717. 884 13. Osna, N.A., Donohue, T.M., Jr., and Kharbanda, K.K. (2017). Alcoholic Liver Disease: 885 Pathogenesis and Current Management. Alcohol Res 38, 147-161. 886 14. Haflidadottir, S., Jonasson, J.G., Norland, H., Einarsdottir, S.O., Kleiner, D.E., Lund, 887 S.H., and Bjornsson, E.S. (2014). Long -term follow-up and liver-related death rate in 888 patients with non-alcoholic and alcoholic related fatty liver disease. BMC Gastroenterol 889 14, 166. 10.1186/1471-230X-14-166. 890 15. Toshikuni, N., Tsutsumi, M., and Arisawa, T. (2014). Clinical differences between 891 alcoholic liver disease and nonalcoholic fatty liver disease. World J Gastroenterol 20, 892 8393-8406. 10.3748/wjg.v20.i26.8393. 893 16. O'Shea, R.S., Dasarathy, S., and McCullough, A.J. (2010). Alcoholic liver disease. Am 894 J Gastroenterol 105, 14-32; quiz 33. 10.1038/ajg.2009.593. 895 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 33 17. Cheemerla, S., and Balakrishnan, M. (2021). Global Epidemiology of Chronic Liver 896 Disease. Clin Liver Dis (Hoboken) 17, 365-370. 10.1002/cld.1061. 897 18. Hernandez-Evole, H., Jimenez-Esquivel, N., Pose, E., and Bataller, R. (2024). Alcohol-898 associated liver disease: Epidemiology and management. Ann Hepatol 29, 101162. 899 10.1016/j.aohep.2023.101162. 900 19. Rehm, J., and Shield, K.D. (2019). Global Burden of Alcohol Use Disorders and Alcohol 901 Liver Disease. Biomedicines 7. 10.3390/biomedicines7040099. 902 20. The Lancet Gastroenterology, H. (2022). Digging deeper into alcohol -related deaths. 903 Lancet Gastroenterol Hepatol 7, 107. 10.1016/S2468-1253(21)00479-9. 904 21. Singal, A.K., and Anand, B.S. (2013). Recent trends in the epidemiology of alcoholic 905 liver disease. Clin Liver Dis (Hoboken) 2, 53-56. 10.1002/cld.168. 906 22. Becker, H.C. (2008). Alcohol dependence, withdrawal, and relapse. Alcohol Res Health 907 31, 348-361. 908 23. Yoon, E.L., and Kim, W. (2023). Current and future treatment for alcoholic-related liver 909 diseases. J Gastroenterol Hepatol 38, 1218-1226. 10.1111/jgh.16257. 910 24. Ciocan, D., and Cassard, A.M. (2022). In the quest for treating alcohol liver disease. 911 EBioMedicine 81, 104086. 10.1016/j.ebiom.2022.104086. 912 25. Urbano-Marquez, A., and Fernandez-Sola, J. (2004). Effects of alcohol on skeletal and 913 cardiac muscle. Muscle Nerve 30, 689-707. 10.1002/mus.20168. 914 26. Janssen, I., Heymsfield, S.B., Wang, Z.M., and Ross, R. (2000). Skeletal muscle mass 915 and distribution in 468 men and women aged 18 -88 yr. J Appl Physiol (1985) 89, 81-916 88. 10.1152/jappl.2000.89.1.81. 917 27. Frontera, W.R., and Ochala, J. (2015). Skeletal muscle: a brief review of structure and 918 function. Calcif Tissue Int 96, 183-195. 10.1007/s00223-014-9915-y. 919 28. Simon, L., Jolley, S.E., and Molina, P .E. (2017). Alcoholic Myopathy: Pathophysiologic 920 Mechanisms and Clinical Implications. Alcohol Res 38, 207-217. 921 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 34 29. Estruch, R., Sacanella, E., Fernandez-Sola, J., Nicolas, J.M., Rubin, E., and Urbano-922 Marquez, A. (1998). Natural history of alcoholic myopathy: a 5-year study. Alcohol Clin 923 Exp Res 22, 2023-2028. 924 30. Souza-Smith, F.M., Lang, C.H., Nagy, L.E., Bailey, S.M., Parsons, L.H., and Murray, 925 G.J. (2016). Physiological processes underlying organ injury in alcohol abuse. Am J 926 Physiol Endocrinol Metab 311, E605-619. 10.1152/ajpendo.00270.2016. 927 31. Simon, L., Bourgeois, B.L., and Molina, P .E. (2023). Alcohol and Skeletal Muscle in 928 Health and Disease. Alcohol Res 43, 04. 10.35946/arcr.v43.1.04. 929 32. Bourgeois, B.L., Levitt, D.E., Molina, P.E., and Simon, L. (2022). Chronic Alcohol and 930 Skeletal Muscle. In Handbook of Substance Misuse and Addictions: From Biology to 931 Public Health, (Springer), pp. 943-967. 932 33. Dasarathy, J., McCullough, A.J., and Dasarathy, S. (2017). Sarcopenia in Alcoholic 933 Liver Disease: Clinical and Molecular Advances. Alcohol Clin Exp Res 41, 1419-1431. 934 10.1111/acer.13425. 935 34. Henin, G., Lanthier, N., and Dahlqvist, G. (2022). Pathophysiological changes of the 936 liver-muscle axis in end-stage liver disease: what is the right target? Acta Gastroenterol 937 Belg 85, 611-624. 10.51821/85.4.10899. 938 35. Dekeyser, G.J., Clary, C.R., and Otis, J.S. (2013). Chronic alcohol ingestion delays 939 skeletal muscle regeneration following injury. Regen Med Res 1, 2. 10.1186/2050 -940 490X-1-2. 941 36. Dai, X., and Shen, L. (2022). Advances and Trends in Omics Technology Development. 942 Front Med (Lausanne) 9, 911861. 10.3389/fmed.2022.911861. 943 37. Karczewski, K.J., and Snyder, M.P . (2018). Integrative omics for health and disease. 944 Nat Rev Genet 19, 299-310. 10.1038/nrg.2018.4. 945 38. Misra, B.B., Langefeld, C.D., Olivier, M., and Cox, L.A. (2018). Integrated Omics: Tools, 946 Advances, and Future Approaches. J Mol Endocrinol. 10.1530/JME-18-0055. 947 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 35 39. Worheide, M.A., Krumsiek, J., Kastenmuller, G., and Arnold, M. (2021). Multi -omics 948 integration in biomedical research - A metabolomics-centric review. Anal Chim Acta 949 1141, 144-162. 10.1016/j.aca.2020.10.038. 950 40. Chen, C., Wang, J., Pan, D., Wang, X., Xu, Y., Yan, J., Wang, L., Yang, X., Yang, M., 951 and Liu, G.P. (2023). Applications of multi -omics analysis in human diseases. 952 MedComm (2020) 4, e315. 10.1002/mco2.315. 953 41. Subramanian, I., Verma, S., Kumar, S., Jere, A., and Anamika, K. (2020). Multi-omics 954 Data Integration, Interpretation, and Its Application. Bioinform Biol Insights 14, 955 1177932219899051. 10.1177/1177932219899051. 956 42. Moser, S.E., Brown, A.M., Ganjayi, M.S., Otis, J.S., and Baumann, C.W. (2023). 957 Excessive Ethanol Intake in Mice Does Not Impair Recovery of Torque after Repeated 958 Bouts of Eccentric Contractions. Med Sci Sports Exerc 55, 873 -883. 959 10.1249/MSS.0000000000003118. 960 43. Moser, S.E., Brown, A.M., Clark, B.C., Arnold, W.D., and Baumann, C.W. (2022). 961 Neuromuscular mechanisms of weakness in a mouse model of chronic alcoholic 962 myopathy. Alcohol Clin Exp Res 46, 1636-1647. 10.1111/acer.14907. 963 44. Lee, S., Norheim, F., Gulseth, H.L., Langleite, T.M., Aker, A., Gundersen, T.E., Holen, 964 T., Birkeland, K.I., and Drevon, C.A. (2018). Skeletal muscle phosphatidylcholine and 965 phosphatidylethanolamine respond to exercise and influence insulin sensitivity in men. 966 Sci Rep 8, 6531. 10.1038/s41598-018-24976-x. 967 45. Jeon, S., and Carr, R. (2020). Alcohol effects on hepatic lipid metabolism. J Lipid Res 968 61, 470-479. 10.1194/jlr.R119000547. 969 46. Tschopp, J., Chonn, A., Hertig, S., and French, L.E. (1993). Clusterin, the human 970 apolipoprotein and complement inhibitor, binds to complement C7, C8 beta, and the b 971 domain of C9. J Immunol 151, 2159-2165. 972 47. Weiler, J.M., and Linhardt, R.J. (1991). Antithrombin III regulates complement activity 973 in vitro. J Immunol 146, 3889-3894. 974 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 36 48. Cahill, K.M., Huo, Z., Tseng, G.C., Logan, R.W., and Seney, M.L. (2018). Improved 975 identification of concordant and discordant gene expression signatures using an 976 updated rank -rank hypergeometric overlap approach. Sci Rep 8, 9588. 977 10.1038/s41598-018-27903-2. 978 49. Balakrishnan, R., Cho, D.Y ., Kim, I.S., Seol, S.H., and Choi, D.K. (2022). Molecular 979 Mechanisms and Therapeutic Potential of alpha - and beta-Asarone in the Treatment 980 of Neurological Disorders. Antioxidants (Basel) 11. 10.3390/antiox11020281. 981 50. Aboutara, N., Jungen, H., Szewczyk, A., Muller, A., and Iwersen-Bergmann, S. (2023). 982 PEth 16:0/18:1 and 16:0/18:2 after consumption of low doses of alcohol-A contribution 983 to cutoff discussion. Drug Test Anal 15, 104-114. 10.1002/dta.3376. 984 51. Agamah, F.E., Bayjanov, J.R., Niehues, A., Njoku, K.F., Skelton, M., Mazandu, G.K., 985 Ederveen, T.H.A., Mulder, N., Chimusa, E.R., and t Hoen, P .A.C. (2022). 986 Computational approaches for network -based integrative multi-omics analysis. Front 987 Mol Biosci 9, 967205. 10.3389/fmolb.2022.967205. 988 52. Singh, A., Shannon, C.P ., Gautier, B., Rohart, F., Vacher, M., Tebbutt, S.J., and Le Cao, 989 K.A. (2019). DIABLO: an integrative approach for identifying key molecular drivers from 990 multi-omics assays. Bioinformatics 35, 3055-3062. 10.1093/bioinformatics/bty1054. 991 53. Yoo, M., Shin, J., Kim, J., Ryall, K.A., Lee, K., Lee, S., Jeon, M., Kang, J., and Tan, 992 A.C. (2015). DSigDB: drug signatures database for gene set analysis. Bioinformatics 993 31, 3069-3071. 10.1093/bioinformatics/btv313. 994 54. Mitchell, D.C., Kuljanin, M., Li, J., Van Vranken, J.G., Bulloch, N., Schweppe, D.K., 995 Huttlin, E.L., and Gygi, S.P . (2023). A proteome -wide atlas of drug mechanism of 996 action. Nat Biotechnol 41, 845-857. 10.1038/s41587-022-01539-0. 997 55. Zhang, Q., Jia, R., Chen, M., Wang, J., Huang, F., Shi, M., Sheng, H., and Xu, L. 998 (2023). Antagonizing EZH2 combined with vitamin D3 exerts a synergistic role in anti-999 fibrosis through bidirectional effects on hepatocytes and hepatic stellate cells. J 1000 Gastroenterol Hepatol 38, 441-450. 10.1111/jgh.16126. 1001 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 37 56. Seo, H.Y ., Lee, S.H., Lee, J.H., Kang, Y .N., Hwang, J.S., Park, K.G., Kim, M.K., and 1002 Jang, B.K. (2020). Src Inhibition Attenuates Liver Fibrosis by Preventing Hepatic 1003 Stellate Cell Activation and Decreasing Connetive Tissue Growth Factor. Cells 9. 1004 10.3390/cells9030558. 1005 57. Lamarche, E., Lala-Tabbert, N., Gunanayagam, A., St-Louis, C., and Wiper-Bergeron, 1006 N. (2015). Retinoic acid promotes myogenesis in myoblasts by antagonizing 1007 transforming growth factor -beta signaling via C/EBPbeta. Skelet Muscle 5, 8. 1008 10.1186/s13395-015-0032-z. 1009 58. Dupre-Aucouturier, S., Castells, J., Freyssenet, D., and Desplanches, D. (2015). 1010 Trichostatin A, a histone deacetylase inhibitor, modulates unloaded -induced skeletal 1011 muscle atrophy. J Appl Physiol (1985) 119, 342-351. 10.1152/japplphysiol.01031.2014. 1012 59. Shang, R., and Miao, J. (2023). Mechanisms and effects of metformin on skeletal 1013 muscle disorders. Front Neurol 14, 1275266. 10.3389/fneur.2023.1275266. 1014 60. D'Souza El -Guindy, N.B., Kovacs, E.J., De Witte, P., Spies, C., Littleton, J.M., de 1015 Villiers, W.J., Lott, A.J., Plackett, T.P., Lanzke, N., and Meadows, G.G. (2010). 1016 Laboratory models available to study alcohol -induced organ damage and immune 1017 variations: choosing the appropriate model. Alcohol Clin Exp Res 34, 1489 -1511. 1018 10.1111/j.1530-0277.2010.01234.x. 1019 61. Hall, P., and Cash, J. (2012). What is the real function of the liver 'function' tests? Ulster 1020 Med J 81, 30-36. 1021 62. Kew, M.C. (2000). Serum aminotransferase concentration as evidence of 1022 hepatocellular damage. Lancet 355, 591-592. 10.1016/S0140-6736(99)00219-6. 1023 63. Niu, L., Thiele, M., Geyer, P .E., Rasmussen, D.N., Webel, H.E., Santos, A., Gupta, R., 1024 Meier, F., Strauss, M., Kjaergaard, M., et al. (2022). Noninvasive proteomic biomarkers 1025 for alcohol-related liver disease. Nat Med 28, 1277-1287. 10.1038/s41591-022-01850-1026 y. 1027 64. Li, Z., Agellon, L.B., Allen, T.M., Umeda, M., Jewell, L., Mason, A., and Vance, D.E. 1028 (2006). The ratio of phosphatidylcholine to phosphatidylethanolamine influences 1029 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 38 membrane integrity and steatohepatitis. Cell Metab 3, 321 -331. 1030 10.1016/j.cmet.2006.03.007. 1031 65. Massey, V., Parrish, A., Argemi, J., Moreno, M., Mello, A., Garcia -Rocha, M., 1032 Altamirano, J., Odena, G., Dubuquoy, L., Louvet, A., et al. (2021). Integrated 1033 Multiomics Reveals Glucose Use Reprogramming and Identifies a Novel Hexokinase 1034 in Alcoholic Hepatitis. Gastroenterology 160, 1725 -1740 e1722. 1035 10.1053/j.gastro.2020.12.008. 1036 66. Thiele, M., Suvitaival, T., Trost, K., Kim, M., de Zawadzki, A., Kjaergaard, M., 1037 Rasmussen, D.N., Lindvig, K.P., Israelsen, M., Detlefsen, S., et al. (2023). 1038 Sphingolipids Are Depleted in Alcohol -Related Liver Fibrosis. Gastroenterology 164, 1039 1248-1260. 10.1053/j.gastro.2023.02.023. 1040 67. Belhaj, M.R., Lawler, N.G., and Hoffman, N.J. (2021). Metabolomics and Lipidomics: 1041 Expanding the Molecular Landscape of Exercise Biology. Metabolites 11. 1042 10.3390/metabo11030151. 1043 68. Roberts, M.D., Kerksick, C.M., Dalbo, V.J., Hassell, S.E., Tucker, P.S., and Brown, R. 1044 (2010). Molecular attributes of human skeletal muscle at rest and after unaccustomed 1045 exercise: an age comparison. J Strength Cond Res 24, 1161 -1168. 1046 10.1519/JSC.0b013e3181da786f. 1047 69. Brook, M.S., Wilkinson, D.J., Smith, K., and Atherton, P .J. (2016). The metabolic and 1048 temporal basis of muscle hypertrophy in response to resistance exercise. Eur J Sport 1049 Sci 16, 633-644. 10.1080/17461391.2015.1073362. 1050 70. Ganjayi, M.S., Brown, A.M., and Baumann, C.W. (2023). Longitudinal assessment of 1051 strength and body composition in a mouse model of chronic alcohol-related myopathy. 1052 Alcohol Clin Exp Res (Hoboken). 10.1111/acer.15149. 1053 71. Cunningham, C.C., and Van Horn, C.G. (2003). Energy availability and alcohol-related 1054 liver pathology. Alcohol Res Health 27, 291-299. 1055 72. Wilson, D.F., and Matschinsky, F.M. (2020). Ethanol metabolism: The good, the bad, 1056 and the ugly. Med Hypotheses 140, 109638. 10.1016/j.mehy.2020.109638. 1057 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 39 73. Wruck, W., and Adjaye, J. (2017). Meta -analysis reveals up-regulation of cholesterol 1058 processes in non-alcoholic and down-regulation in alcoholic fatty liver disease. World 1059 J Hepatol 9, 443-454. 10.4254/wjh.v9.i8.443. 1060 74. Prasad, B., Bhatt, D.K., Johnson, K., Chapa, R., Chu, X., Salphati, L., Xiao, G., Lee, 1061 C., Hop, C., Mathias, A., et al. (2018). Abundance of Phase 1 and 2 Drug-Metabolizing 1062 Enzymes in Alcoholic and Hepatitis C Cirrhotic Livers: A Quantitative Targeted 1063 Proteomics Study. Drug Metab Dispos 46, 943-952. 10.1124/dmd.118.080523. 1064 75. Han, D., Ybanez, M.D., Johnson, H.S., McDonald, J.N., Mesropyan, L., Sancheti, H., 1065 Martin, G., Martin, A., Lim, A.M., Dara, L., et al. (2012). Dynamic adaptation of liver 1066 mitochondria to chronic alcohol feeding in mice: biogenesis, remodeling, and functional 1067 alterations. J Biol Chem 287, 42165-42179. 10.1074/jbc.M112.377374. 1068 76. Yang, Z., Han, S., Zhang, T., Kusumanchi, P ., Huda, N., Tyler, K., Chandler, K., Skill, 1069 N.J., Tu, W., Shan, M., et al. (2022). Transcriptomic Analysis Reveals the Messenger 1070 RNAs Responsible for the Progression of Alcoholic Cirrhosis. Hepatol Commun 6, 1071 1361-1372. 10.1002/hep4.1903. 1072 77. Hoek, J.B., Cahill, A., and Pastorino, J.G. (2002). Alcohol and mitochondria: a 1073 dysfunctional relationship. Gastroenterology 122, 2049 -2063. 1074 10.1053/gast.2002.33613. 1075 78. Garcia-Ruiz, C., Kaplowitz, N., and Fernandez -Checa, J.C. (2013). Role of 1076 Mitochondria in Alcoholic Liver Disease. Curr Pathobiol Rep 1, 159 -168. 1077 10.1007/s40139-013-0021-z. 1078 79. Hayes, J.D., Flanagan, J.U., and Jowsey, I.R. (2005). Glutathione transferases. Annu 1079 Rev Pharmacol Toxicol 45, 51-88. 10.1146/annurev.pharmtox.45.120403.095857. 1080 80. Marcos, M., Pastor, I., Chamorro, A.J., Ciria -Abad, S., Gonzalez -Sarmiento, R., and 1081 Laso, F.J. (2011). Meta -analysis: glutathione -S-transferase allelic variants are 1082 associated with alcoholic liver disease. Aliment Pharmacol Ther 34, 1159 -1172. 1083 10.1111/j.1365-2036.2011.04862.x. 1084 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 40 81. Taiwo, M., Huang, E., Pathak, V., Bellar, A., Welch, N., Dasarathy, J., Streem, D., 1085 McClain, C.J., Mitchell, M.C., Barton, B.A., et al. (2024). Proteomics identifies 1086 complement protein signatures in patients with alcohol -associated hepatitis. JCI 1087 Insight 9. 10.1172/jci.insight.174127. 1088 82. Li, Q., Li, Z., Lin, Y., Che, H., Hu, Y ., Kang, X., Zhang, Y ., Wang, L., and Zhang, Y . 1089 (2019). High glucose promotes hepatic fibrosis via miR ‑32/MTA3‑mediated 1090 epithelial‑to‑mesenchymal transition. Mol Med Rep 19, 3190 -3200. 1091 10.3892/mmr.2019.9986. 1092 83. Wang, C., Li, G., Li, J., Li, J., Li, T., Yu, J., and Qin, C. (2017). Overexpression of the 1093 metastasis-associated gene MTA3 correlates with tumor progression and poor 1094 prognosis in hepatocellular carcinoma. J Gastroenterol Hepatol 32, 1525 -1529. 1095 10.1111/jgh.13680. 1096 84. Bonaldo, P., and Sandri, M. (2013). Cellular and molecular mechanisms of muscle 1097 atrophy. Dis Model Mech 6, 25-39. 10.1242/dmm.010389. 1098 85. Hyatt, H.W., and Powers, S.K. (2021). Mitochondrial Dysfunction Is a Common 1099 Denominator Linking Skeletal Muscle Wasting Due to Disease, Aging, and Prolonged 1100 Inactivity. Antioxidants (Basel) 10. 10.3390/antiox10040588. 1101 86. Goody, M.F., Sher, R.B., and Henry, C.A. (2015). Hanging on for the ride: adhesion to 1102 the extracellular matrix mediates cellular responses in skeletal muscle morphogenesis 1103 and disease. Dev Biol 401, 75-91. 10.1016/j.ydbio.2015.01.002. 1104 87. Deane, C.S., Willis, C.R.G., Phillips, B.E., Atherton, P .J., Harries, L.W., Ames, R.M., 1105 Szewczyk, N.J., and Etheridge, T. (2021). Transcriptomic meta -analysis of disuse 1106 muscle atrophy vs. resistance exercise -induced hypertrophy in young and older 1107 humans. J Cachexia Sarcopenia Muscle 12, 629-645. 10.1002/jcsm.12706. 1108 88. Nowak, K.J., Sewry, C.A., Navarro, C., Squier, W., Reina, C., Ricoy, J.R., Jayawant, 1109 S.S., Childs, A.M., Dobbie, J.A., Appleton, R.E., et al. (2007). Nemaline myopathy 1110 caused by absence of alpha -skeletal muscle actin. Ann Neurol 61, 175 -184. 1111 10.1002/ana.21035. 1112 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 41 89. Glass, C.K., and Olefsky, J.M. (2012). Inflammation and lipid signaling in the etiology 1113 of insulin resistance. Cell Metab 15, 635-645. 10.1016/j.cmet.2012.04.001. 1114 90. Montgomery, M.K., and Turner, N. (2015). Mitochondrial dysfunction and insulin 1115 resistance: an update. Endocr Connect 4, R1-R15. 10.1530/EC-14-0092. 1116 91. Selathurai, A., Kowalski, G.M., Mason, S.A., Callahan, D.L., Foletta, V.C., Della Gatta, 1117 P.A., Lindsay, A., Hamley, S., Kaur, G., Curtis, A.R., et al. (2019). Phosphatidylserine 1118 decarboxylase is critical for the maintenance of skeletal muscle mitochondrial integrity 1119 and muscle mass. Mol Metab 27, 33-46. 10.1016/j.molmet.2019.06.020. 1120 92. Hinkley, J.M., Cornnell, H.H., Standley, R.A., Chen, E.Y ., Narain, N.R., Greenwood, 1121 B.P ., Bussberg, V., Tolstikov, V.V., Kiebish, M.A., Yi, F., et al. (2020). Older adults with 1122 sarcopenia have distinct skeletal muscle phosphodiester, phosphocreatine, and 1123 phospholipid profiles. Aging Cell 19, e13135. 10.1111/acel.13135. 1124 93. Grapentine, S., and Bakovic, M. (2019). Significance of bilayer-forming phospholipids 1125 for skeletal muscle insulin sensitivity and mitochondrial function. J Biomed Res 34, 1-1126 13. 10.7555/JBR.33.20180104. 1127 94. Demonbreun, A.R., and McNally, E.M. (2017). Muscle cell communication in 1128 development and repair. Curr Opin Pharmacol 34, 7-14. 10.1016/j.coph.2017.03.008. 1129 95. Ganjayi, M.S., Krauss, T.A., Willis, C.R., and Baumann, C.W. Chronic Alcohol-Related 1130 Myopathy: A Closer Look at the Role of Lipids. Frontiers in Physiology 15, 1492405. 1131 96. Wang, X., Han, X., and Powell, C.A. (2022). Lipids and genes: Regulatory roles of 1132 lipids in RNA expression. Clin Transl Med 12, e863. 10.1002/ctm2.863. 1133 97. Matthews, H., Hanison, J., and Nirmalan, N. (2016). "Omics" -Informed Drug and 1134 Biomarker Discovery: Opportunities, Challenges and Future Perspectives. Proteomes 1135 4. 10.3390/proteomes4030028. 1136 98. Pulley, J.M., Rhoads, J.P., Jerome, R.N., Challa, A.P., Erreger, K.B., Joly, M.M., Lavieri, 1137 R.R., Perry, K.E., Zaleski, N.M., Shirey -Rice, J.K., and Aronoff, D.M. (2020). Using 1138 What We Already Have: Uncovering New Drug Repurposing Strategies in Existing 1139 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 42 Omics Data. Annu Rev Pharmacol Toxicol 60, 333-352. 10.1146/annurev-pharmtox-1140 010919-023537. 1141 99. Du, G., Wang, J., Zhang, T., Ding, Q., Jia, X., Zhao, X., Dong, J., Yang, X., Lu, S., 1142 Zhang, C., et al. (2020). Targeting Src family kinase member Fyn by Saracatinib 1143 attenuated liver fibrosis in vitro and in vivo. Cell Death Dis 11, 118. 10.1038/s41419-1144 020-2229-2. 1145 100. Wu, X., Wang, Y ., Wang, Y., Wang, X., Li, J., Chang, K., Sun, C., Jia, Z., Gao, S., Wei, 1146 J., et al. (2018). GSK126 alleviates the obesity phenotype by promoting the 1147 differentiation of thermogenic beige adipocytes in diet -induced obese mice. Biochem 1148 Biophys Res Commun 501, 9-15. 10.1016/j.bbrc.2018.04.073. 1149 101. Liu, X., Liu, M., Lee, L., Davies, M., Wang, Z., Kim, H., and Feeley, B.T. (2021). 1150 Trichostatin A regulates fibro/adipogenic progenitor adipogenesis epigenetically and 1151 reduces rotator cuff muscle fatty infiltration. J Orthop Res 39, 1452 -1462. 1152 10.1002/jor.24865. 1153 102. Penning, T.M. (2011). Human hydroxysteroid dehydrogenases and pre -receptor 1154 regulation: insights into inhibitor design and evaluation. J Steroid Biochem Mol Biol 1155 125, 46-56. 10.1016/j.jsbmb.2011.01.009. 1156 103. Loerz, C., and Maser, E. (2017). The cortisol-activating enzyme 11beta-hydroxysteroid 1157 dehydrogenase type 1 in skeletal muscle in the pathogenesis of the metabolic 1158 syndrome. J Steroid Biochem Mol Biol 174, 65-71. 10.1016/j.jsbmb.2017.07.030. 1159 104. Chapman, K., Holmes, M., and Seckl, J. (2013). 11beta -hydroxysteroid 1160 dehydrogenases: intracellular gate -keepers of tissue glucocorticoid action. Physiol 1161 Rev 93, 1139-1206. 10.1152/physrev.00020.2012. 1162 105. Shenkman, B., Zinovyeva, O., Belova, S., Samkhaeva, N., Shcheglova, N., Mirzoev, 1163 T., Vilchinskaya, N., Altaeva, E., Turtikova, O., and Nemirovskaya, T. (2016). The 1164 response of skeletal muscle to alcohol abuse: Gender differences. Biophysics 61, 793-1165 796. 1166 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 43 106. Becker, U., Deis, A., Sorensen, T.I., Gronbaek, M., Borch -Johnsen, K., Muller, C.F., 1167 Schnohr, P., and Jensen, G. (1996). Prediction of risk of liver disease by alcohol intake, 1168 sex, and age: a prospective population study. Hepatology 23, 1025 -1029. 1169 10.1002/hep.510230513. 1170 107. Urbano-Marquez, A., Estruch, R., Fernandez -Sola, J., Nicolas, J.M., Pare, J.C., and 1171 Rubin, E. (1995). The greater risk of alcoholic cardiomyopathy and myopathy in women 1172 compared with men. JAMA 274, 149-154. 10.1001/jama.1995.03530020067034. 1173 108. Chen, Y ., Chen, Y ., Shi, C., Huang, Z., Zhang, Y ., Li, S., Li, Y ., Ye, J., Yu, C., Li, Z., et 1174 al. (2018). SOAPnuke: a MapReduce acceleration -supported software for integrated 1175 quality control and preprocessing of high-throughput sequencing data. Gigascience 7, 1176 1-6. 10.1093/gigascience/gix120. 1177 109. Bray, N.L., Pimentel, H., Melsted, P., and Pachter, L. (2016). Near-optimal probabilistic 1178 RNA-seq quantification. Nat Biotechnol 34, 525-527. 10.1038/nbt.3519. 1179 110. Pertea, G., and Pertea, M. (2020). GFF Utilities: GffRead and GffCompare. F1000Res 1180 9. 10.12688/f1000research.23297.2. 1181 111. Soneson, C., Love, M.I., and Robinson, M.D. (2015). Differential analyses for RNA -1182 seq: transcript -level estimates improve gene -level inferences. F1000Res 4, 1521. 1183 10.12688/f1000research.7563.2. 1184 112. Ritchie, M.E., Phipson, B., Wu, D., Hu, Y ., Law, C.W., Shi, W., and Smyth, G.K. (2015). 1185 limma powers differential expression analyses for RNA -sequencing and microarray 1186 studies. Nucleic Acids Res 43, e47. 10.1093/nar/gkv007. 1187 113. Smyth, G.K. (2004). Linear models and empirical bayes methods for assessing 1188 differential expression in microarray experiments. Stat Appl Genet Mol Biol 3, Article3. 1189 10.2202/1544-6115.1027. 1190 114. Wu, T., Hu, E., Xu, S., Chen, M., Guo, P ., Dai, Z., Feng, T., Zhou, L., Tang, W., Zhan, 1191 L., et al. (2021). clusterProfiler 4.0: A universal enrichment tool for interpreting omics 1192 data. Innovation (Camb) 2, 100141. 10.1016/j.xinn.2021.100141. 1193 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 44 115. Castanza, A.S., Recla, J.M., Eby, D., Thorvaldsdottir, H., Bult, C.J., and Mesirov, J.P. 1194 (2023). Extending support for mouse data in the Molecular Signatures Database 1195 (MSigDB). Nat Methods. 10.1038/s41592-023-02014-7. 1196 116. Csardi, G.N.T. (2006). The igraph software package for complex network research. 1197 InterJournal Complex Systems, 1695. 1198 117. Shannon, P., Markiel, A., Ozier, O., Baliga, N.S., Wang, J.T., Ramage, D., Amin, N., 1199 Schwikowski, B., and Ideker, T. (2003). Cytoscape: a software environment for 1200 integrated models of biomolecular interaction networks. Genome Res 13, 2498-2504. 1201 10.1101/gr.1239303. 1202 118. Chen, E.Y ., Tan, C.M., Kou, Y., Duan, Q., Wang, Z., Meirelles, G.V., Clark, N.R., and 1203 Ma'ayan, A. (2013). Enrichr: interactive and collaborative HTML5 gene list enrichment 1204 analysis tool. BMC Bioinformatics 14, 128. 10.1186/1471-2105-14-128. 1205 1206 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint A B C * 125 150 175 200 Control Alcohol Plantarflexor muscle mass (mg) * 9 10 11 12 Control Alcohol Plantarflexor muscle torque (mN.m) * 2.5 3.0 3.5 4.0 Control Alcohol Plantarflexor muscle PC:PE D E * 4.0 4.5 5.0 Control Alcohol Liver PC:PE 1000 1250 1500 Control Alcohol Liver mass (mg) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint Cyp3a41b Cyp3a44 Lepr Gm6135 Cyp3a63−ps Idi1 Tkfc 2010003K11Rik Cyp51 Cyp2d9 Nsdhl Msmo1 Gstm3Rdh11Atp6v0d2 Paqr7 Lhx6 Sox12Arsa Gm35986 0 5 10 15 −8 −4 0 4 8 log2(fold−change) −log10(adjusted P−value) Liver Ivd ENSMUSG00000120512 Exoc3l4 Cyp4f18Mettl2 Vasp Coq10a Cdkn1cGm34455 Cth Fam174b Igfbp4 Sertad3 Actc1 Tmsb10 AnpepGstk1 Gsto1 Pxmp2Mpc1 0 5 10 15 −8 −4 0 4 8 log2(fold−change) −log10(adjusted P−value) Muscle Upregulated Downregulated GTRDHALLMARK Liver Both Muscle Liver Both Muscle ZFP449 TARGET GENES TBX20 TARGET GENES SPIB TARGET GENES RXRG TARGET GENES PPARGC1A TARGET GENES NR1H2 TARGET GENES MTA3 TARGET GENES L YL1 TARGET GENES CRELD1 TARGET GENES AFF4 TARGET GENES TNFA SIGNALING VIA NFKB OXIDATIVE PHOSPHORYLATION MYOGENESIS MTORC1 SIGNALING KRAS SIGNALING UP INTERFERON GAMMA RESPONSE INFLAMMATORY RESPONSE IL6 JAK STAT3 SIGNALING IL2 STAT5 SIGNALING HYPOXIA GL YCOL YSIS FATTY ACID METABOLISM EPITHELIAL MESENCHYMAL TRANSITION COMPLEMENT COAGULATION CHOLESTEROL HOMEOSTASIS BILE ACID METABOLISM APOPTOSIS ANGIOGENESIS ALLOGRAFT REJECTION ADIPOGENESIS MuscleLiver 1055 A CB Upregulated genes Muscle Liver 119 785 965 64 440 Downregulated genes Upregulated: n = 1183 Downregulated: n = 1043 Upregulated: n = 906 Downregulated: n = 514 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint Fcn1 Cbr3 Mvk Gstm3 Pfkfb1 Htatip2 Galt Arsa Med16 Ociad2 Gpcpd1 Gstm1 Ggct Retsat Alg3 Lpcat3 Idi1 Dcxr Psmb9 Gnaq 0 5 10 15 −10 −5 0 5 10 log2(fold−change) −log10(adjusted P−value) Liver Prps1 Clu Mbl1 Cd5l Itih4 Bcap29 Serpinc1 A1bg 0 5 10 15 −10 −5 0 5 10 log2(fold−change) −log10(adjusted P−value) Muscle Muscle Liver 338 A CB Upregulated proteins Muscle Liver 1 5 250 2 Downregulated proteins Upregulated: n = 339 Downregulated: n = 252 Upregulated: n = 6 Downregulated: n = 2 Upregulated Downregulated HALLMARKREACTOME Liver Both Muscle Liver Both Muscle OXIDATIVE PHOSPHORYLATION INTERFERON GAMMA RESPONSE GL YCOL YSIS COMPLEMENT COAGULATION CHOLESTEROL HOMEOSTASIS TRANSLATION PHOSPHOLIPID METABOLISM MITOCHONDRIAL TRANSLATION METABOLISM OF LIPIDS METABOLISM OF AMINO ACIDS AND DERIVATIVES HEMOSTASIS GL YCEROPHOSPHOLIPID BIOSYNTHESIS .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint Gene Protein −log10(P) 0 10 20 30 Liver Gene Protein −log10(P) 0 4 8 12 16 Muscle A B Liver Muscle HALLMARKREACTOME G P G P XENOBIOTIC METABOLISM OXIDATIVE PHOSPHORYLATION INTERFERON GAMMA RESPONSE COMPLEMENT COAGULATION CHOLESTEROL HOMEOSTASIS APOPTOSIS ADIPOGENESIS UCH PROTEINASES UBIQUITIN MEDIATED DEGRADATION OF PHOSPHORYLATED CDC25A UB SPECIFIC PROCESSING PROTEASES TRANSCRIPTIONAL REGULATION BY RUNX2 THE CITRIC ACID TCA CYCLE AND RESPIRATORY ELECTRON TRANSPORT TCR SIGNALING TCF DEPENDENT SIGNALING IN RESPONSE TO WNT SYNTHESIS OF PA STABILIZATION OF P53 SIGNALING BY THE B CELL RECEPTOR BCR SIGNALING BY HEDGEHOG RUNX1 REGULATES TRANSCRIPTION OF GENES INVOLVED IN DIFFERENT... RESPIRATORY ELECTRON TRANSPORT ATP SYNTHESIS BY CHEMIOSMOTIC... RESPIRATORY ELECTRON TRANSPORT REGULATION OF RAS BY GAPS PLATELET ACTIVATION SIGNALING AND AGGREGATION PHOSPHOLIPID METABOLISM NICOTINATE METABOLISM NEUTROPHIL DEGRANULATION NEURONAL SYSTEM METABOLISM OF CARBOHYDRATES METABOLISM OF AMINO ACIDS AND DERIVATIVES INTEGRIN SIGNALING INTEGRIN CELL SURFACE INTERACTIONS INNATE IMMUNE SYSTEM IMMUNE SYSTEM HEMOSTASIS HEDGEHOG ON STATE HEDGEHOG OFF STATE HEDGEHOG LIGAND BIOGENESIS GRB2 SOS PROVIDES LINKAGE TO MAPK SIGNALING FOR INTEGRINS GL YOXYLATE METABOLISM AND GL YCINE DEGRADATION GL YCEROPHOSPHOLIPID BIOSYNTHESIS GLI3 IS PROCESSED TO GLI3R BY THE PROTEASOME FCERI MEDIATED NF KB ACTIVATION EXTRACELLULAR MATRIX ORGANIZATION DOWNSTREAM TCR SIGNALING DOWNSTREAM SIGNALING EVENTS OF B CELL RECEPTOR BCR DEUBIQUITINATION DEGRADATION OF DVL DEGRADATION OF AXIN CROSS PRESENTATION OF SOLUBLE EXOGENOUS ANTIGENS ENDOSOMES COMPLEX I BIOGENESIS COMPLEMENT CASCADE CLASS I MHC MEDIATED ANTIGEN PROCESSING PRESENTATION ASYMMETRIC LOCALIZATION OF PCP PROTEINS ASSEMBL Y OF THE PRE REPLICATIVE COMPLEX APC C CDH1 MEDIATED DEGRADATION OF CDC20 AND OTHER APC C CDH... ANTIGEN PROCESSING UBIQUITINATION PROTEASOME DEGRADATION ANTIGEN PROCESSING CROSS PRESENTATION ACTIVATION OF NF KAPPAB IN B CELLS ACTIVATION OF APC C AND APC C CDC20 MEDIATED DEGRADATION OF ... Concordance Concordance .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint 2−deoxy−d−glucose 3−o−beta−d−galactosyl−sn−... Guanidinosuccinic acid Azithromycin FucoseBufotalin Pydanon Quinolinic acid 7−aminomethyl−7−deazaguan... Trans−3−indoleacrylic aci... 5−methyltetrahydrofolic a... N−acetyl−l−phenylalanine1−palmitoyl−2−oleoyl−sn−g... S−methylglutathione (9cis)−retinal Propionylcarnitine Acetylcholine Hydroxyphenyllactic acid .−alanine N−acetyl−.−d−glucosamine ... 0 10 20 30 −5 0 5 log2(fold−change) −log10(adjusted P−value) Liver 3−o−beta−d−galactosyl−sn−... 2−deoxy−d−glucose 4−hydroxybutyric acid (gh... Azithromycin FucoseQuinolinic acid N−acetyl−l−glutamine 1−[4−(9−benzyl−9h−fluoren... N−nitrosoproline 1−octadecanoyl−2−(7z,10z,... Acetylcholine 1,4−dihydro−4−imino−1−î²−... .−asarone Trimethylamine n−oxide P−dmea 11−deoxocucurbitacin i TrigonellineHecogenin Chlorfenethol4−tert−octylphenol monoet...0 10 20 30 −5 0 5 log2(fold−change) −log10(adjusted P−value) Muscle Muscle Liver 94 A CB Upregulated metabolites Muscle Liver 20 9 178 8 Downregulated metabolites Upregulated: n = 114 Downregulated: n = 186 Upregulated: n = 29 Downregulated: n = 20 12 Upregulated Downregulated Liver Both Muscle Liver Both Muscle Super Pathway: Carbohydrate metabolism Sub Pathway: Glycine, serine and threonine metabolism Sub Pathway: Glycerophospholipid metabolism Final Class: Organic nitrogen compounds Final Class: Organic acids .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint Pet(16:0/18:2) Cer(d34:1) Pet(16:0/20:5) Pet(16:0/20:4) Pet(16:0/20:3) Cl(82:11) Pet(16:0/22:6)Pc(38:6e)(rep) Pc(16:0p/22:5) Pc(36:4p) Pe(18:0/22:6)(rep) Pc(37:7)(rep) Pc(39:7)(rep) Pc(39:6p) Pe(18:1/22:6)(rep) Pc(38:4)(rep)(rep)(rep)(rep) Pc(40:8p) Pe(38:4)Pc(40:6p)(rep) Pc(34:2e) 0 10 20 30 −10 −5 0 5 10 log2(fold−change) −log10(adjusted P−value) Liver Pet(16:0/18:2) Pc(34:2)(rep)(rep)(rep) Pc(16:0/18:2) Dmepe(16:0/18:2) Pc(44:12) Ps(39:3) Pet(16:0/20:3) Pc(36:6) Pc(36:3) Pc(33:2) Pe(16:0/18:2) Pc(20:4/22:6) Pc(17:0/18:2) Dmepe(16:0/18:3) Pc(36:6)(rep) Pe(34:0)(rep)Pc(42:11) Dmepe(40:6p) Pe(18:0/16:0) Pe(41:6) 0 10 20 30 −10 −5 0 5 10 log2(fold−change) −log10(adjusted P−value) Muscle Upregulated Downregulated Liver Both Muscle Liver Both Muscle Sphingolipids P−Serine P−Inositol P−Glycerol P−Ethanol Amine P−Ethanol P−Choline Neutral Glycosphingolipids Neutral glycerolipid Glycosphingolipids Glycoglycerolipid Fatty Acid Coenzyme Cardiolipin MuscleLiver 79 A CB Upregulated lipids MuscleLiver 44 136 138 50 89 Downregulated lipids Upregulated: n = 123 Downregulated: n = 188 Upregulated: n = 180 Downregulated: n = 139 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint A B Liver C1 Liver C2 Liver C3 Liver C4 Muscle C1 Muscle C2 Node key: Shape: • Circle = gene • Diamond = protein • V shape = metabolite • Triangle = lipid Colour: • Red = significantly upregulated • Blue = significantly downregulated Size: • Larger nodes = eigenvector centrality > 0.7 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint Upregulated Downregulated DSigDBProteome Drug Atlas Liver Both Muscle Liver Both Muscle trichostatin A CTD 00000660 (G) thioridazine MCF7 UP (G) Tetradioxin CTD 00006848 (G) terfenadine MCF7 UP (G) tanespimycin MCF7 UP (G) Retinoic acid CTD 00006918 (G) pyrvinium PC3 UP (G) progesterone CTD 00006624 (G) pimozide MCF7 UP (G) perhexiline MCF7 UP (G) nicotinic acid BOSS (G) Metformin hydrochloride (G) mefloquine MCF7 UP (G) mebendazole HL60 UP (G) lycorine HL60 DOWN (G) ketamine BOSS (G) geldanamycin MCF7 UP (G) estradiol CTD 00005920 (G) Decitabine CTD 00000750 (G) cytarabine CTD 00005743 (G) Cube root extract CTD 00006707 (G) chlorprothixene MCF7 UP (G) Caspan CTD 00000180 (G) Benzo[k]fluoranthene CTD 00001069 (G) benzo[a]pyrene CTD 00005488 (G) anisomycin HL60 DOWN (G) alexidine PC3 UP (G) AFLATOXIN B1 CTD 00007128 (P) AFLATOXIN B1 CTD 00007128 (G) 8−Bromo−cAMP , Na CTD 00007044 (G) 5707885 PC3 UP (G) 22−Hydroxycholesterol CTD 00000121 (G) Zofenopril Down (G) ZINC8945 Up (G) YK−4−279 Down (P) Tubastatin A Up (G) TOFA Up (P) Tacrolimus Down (P) Saracatinib Up (P) Saracatinib Up (G) Ro 48−8071 Up (P) Ro 48−8071 Up (G) RGFP966 Up (G) Pyr1−Apelin−13 Down (P) PP 1 Up (G) PF−543 Up (P) PF−3758309 Up (G) PD 176252 Up (G) Obatoclax Up (P) Obatoclax Up (G) Mubritinib Up (G) MK2206 Up (G) MI−3 Up (G) MERCK544 Down (G) Mardepodect Up (P) L−741 Up (G) Iloprost Down (G) Guanfacine Down (P) Guanfacine Down (G) GSK2334470 Up (P) GSK2334470 Up (G) GSK126 Up (P) GSK126 Up (G) GBR 12909 Down (P) Gandotinib Up (G) G−1 Up (G) Ethoxzolamide Up (P) Eltrombopag Down (P) EIPA Down (P) Desipramine Down (P) Desipramine Down (G) Cyclopamine Up (G) Clozapine Down (P) CHEMBL200403 Up (G) CHEMBL200403 Down (P) CGP77675 Up (P) CGP77675 Up (G) Cediranib Up (P) Cediranib Up (G) CAY10650 Down (G) CAL−101 Down (G) BX 471 Up (G) BIX 01294 Up (G) AZ505 Down (P) AXL1717 Up (G) AS 1517499 Up (G) AGI−6780 Up (G) .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 10, 2025. ; https://doi.org/10.1101/2025.10.09.681415doi: bioRxiv preprint