Tirzepatide attenuates dopamine reward signaling and suppresses alcohol drinking and relapse-like behaviors in rodents

GLP-1, GIP , alcohol, reward, dopamine 31 32 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 2

33 Alcohol use disorder (AUD) remains a major public health problem, with few effective 34 medications currently available. However, p eptides of the gut -brain axis appear to offer 35 promising therapeutic targets for AUD as they influence the mesolimbic reward circuitry. Here, 36 we examined the effects of tirzepatide, a long-acting dual glucagon-like peptide-1 receptor (GLP-37 1R) and glucose -dependent insulinotropic polypeptide receptor (GIPR) agonist approved for 38 diabetes and obesity, using behavioral assays, alcohol intake paradigms, and molecular 39 analyses in rodents. First, tirzepatide effectively attenuated the rewarding properties of alcohol, 40 measured through locomotor stimulation, conditioned place preference, and accumbal 41 dopamine release. Subsequently, this GLP -1R/GIPR agonist dose -dependently reduced 42 voluntary alcohol consumption, prevented binge and relapse -like drinking, and maintained 43 efficacy during repeated administration. Finally, tirzepatide induced sustained synaptic 44 depression in the lateral septum and further altered histone regulatory proteins in this region, 45 suggesting a potential neural substrate for its effects. Moreover, the GLP -1R/GIPR agonist 46 affected metabolic parameters including body weight, adipose tissue mass, hepatic triglycerides 47 and circulating pro -inflammatory cytokines. Together, our findings suggest tirzepatide 48 modulates alcohol-related behaviors through reward -related mechanisms while also affecting 49 physiological consequences associated with long -term alcohol use. Given tirzepatide's 50 established clinical use and the consistency of effects observed here, these results support 51 further investigation for treating AUD and associated complications. 52 53 54 55 56 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 3 SIGNIFICANCE STATEMENT 57 Existing treatments for alcohol use disorder show limited effectiveness, leaving patients without 58 viable therapeutic options. We demonstrate that tirzepatide, a long -acting gut peptide -based 59 drug already approved for diabetes and obesity, substantially reduces alcohol consumption and 60 prevents relapse-like behavior across multiple preclinical models. Tirzepatide appears to work 61 by influencing brain reward systems while simultaneously a ffecting metabolic complications 62 common in alcohol disorders. Given tirzepatide's clinical availability, these findings suggest 63 repurposing a recently approved drug to tackle one of medicine's more persistent treatment 64 challenges. 65 66 67 68 69 70 71 72 73 74 75 76 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 4

77 Alcohol use disorder (AUD) remains a major public health challenge, contributing to substantial 78 morbidity and mortality worldwide (1, 2) . Despite available treatments, current medications 79 show modest efficacy and are under-prescribed (3), highlighting the need for additional effective 80 therapeutic approaches with alternative mechanisms of action. The neurobiological 81 mechanisms underlying AUD involve the mesolimbic dopamine system, where alcohol-induced 82 dopamine release in the nucleus accumbens (NAc) reinforces consummatory behaviors and 83 contributes to the risk of developing AUD later in life (4-8). Long-term alcohol exposure leads to 84 persistent neuroadaptations that disrupt mesolimbic system function, contributing to craving 85 and relapse vulnerability (7-13). Research indicates that altered neuroplasticity and epigenetic 86 mechanisms, including histone modifications, maintain these neuroadaptations (9-13). This 87 complexity, combined with the limited success of existing therapies, suggests that current 88 treatment approaches may be insufficient. Effective interventions might require strategies that 89 address multiple interconnected systems influencing reward processing and addiction 90 vulnerability. 91 In this context, gut –brain axis peptides have emerged as promising therapeutic 92 candidates for AUD (14-16), given not only their apparent capacity to reduce alcohol intake (17) 93 but also their wide -ranging physiological effects (18). The incretin hormones glucagon -like 94 peptide-1 (GLP -1) and glucose -dependent insulinotropic polypeptide (GIP), traditionally 95 recognized for their metabolic functions, also appear to influence central reward processing (14-96 16, 19, 20) . Preclinical studies demonstrate that GLP -1 receptor (GLP -1R) agonists reduce 97 alcohol consumption, likely by attenuating alcohol's rewarding effects (21-28). Early clinical data 98 from randomized trials and observational studies further demonstrates that GLP-1R agonists can 99 reduce alcohol intake in humans (29-32). Building on these findings, clinical trials are now 100 investigating these therapeutic applications more systematically, including studies examining 101 incretin agonists for both alcohol consumption and alcohol-related disorders (ClinicalTrials.gov 102 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 5 identifiers: NCT06546384, NCT06409130, NCT07046819, NCT05891587, NCT05895643, 103 NCT06015893, NCT05892432, NCT06939088, NCT06727331, and NCT06994338). 104 Recent advances have further introduced multi -receptor incretin agonists for diabetes 105 and obesity treatment, with GLP -1R agonism serving as a central component (20). Given their 106 apparent multiple modes of action, these compounds may address several aspects of AUD's 107 complex pathophysiology. Among these, tirzepatide, a long -acting dual GLP -1R/GIPR agonist 108 approved for diabetes and obesity, shows enhanced therapeutic outcomes on cardiometabolic 109 diseases compared to selective GLP -1R agonists (20, 33, 34) . Tirzepatide's clinical availability 110 presents an opportunity to explore whether dual incretin agonists might offer advantages for AUD 111 treatment. However, whether tirzepatide even affects alcohol consumption, and if so through 112 what mechanisms, remains unexplored. 113 To address this knowledge gap, we conducted a systematic investigation of tirzepatide's 114 effects across multiple aspects of AUD using preclinical models. Our approach examined 115 tirzepatide's impact on alcohol -related reward processing, voluntary consumption, and binge 116 and relapse-like drinking in both sexes. We also assessed tirzepatide's influence on metabolic 117 and inflammatory parameters, which are often dysregulated in AUD (35-37). To identify potential 118 neural substrates underlying tirzepatide's effects, we employed electrophysiological recordings 119 across reward-related brain regions and conducted proteomic analysis of tissue samples from 120 alcohol-consuming rats. Together, these studies allowed us to evaluate tirzepatide's therapeutic 121 potential while beginning to characterize what biological mechanisms might account for any 122 observed effects. 123 124 125 126 127 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 6

128 Tirzepatide disrupts alcohol -induced behaviors and mesolimbic dopamine release 129 in male mice 130 Alcohol consumption activates the mesolimbic dopamine system (6), producing locomotor 131 stimulation, conditioned place preference (CPP), and dopamine release in the NAc. These 132 behavioral and neurochemical changes underlie alcohol's rewarding properties and contribute 133 to the risk of developing AUD (5-8). Given tirzepatide's potential to influence reward processing, 134 we first tested whether acute tirzepatide administration (0.144 mg/kg, subcutaneously; SC) 135 could alter these alcohol -induced (1.75 g/kg, intraperitoneally; IP) reward responses in male 136 mice. 137 We first examined alcohol -induced locomotor stimulation . Tirzepatide alone had no 138 effect on baseline locomotion compared to vehicle (P>0.999), while alcohol produced the 139 expected locomotor activation (P<0.001) (F 3,32=11.40, P<0.001; Fig. 1A ). Tirzepatide 140 pretreatment significantly blunted this alcohol-induced stimulation (P0.999). We next examined tirzepatide's influence on alcohol CPP. A 143 control experiment verified that tirzepatide itself did not affect place conditioning when vehicle 144 was paired with both compartments (t 14=0.49, P=0.631; Fig. 1B). When alcohol was paired with 145 one side, vehicle -treated mice developed clear preference for the alcohol -associated 146 environment. Tirzepatide treatment markedly reduced this preference (t 18=5.23, P<0.001; Fig. 147 1B). This guided us to examine another clinically relevant question. Recent evidence has shown 148 that GLP-1R agonist exenatide reduces alcohol cue reactivity in NAc and septal regions of AUD 149 patients (30), leading us to test whether tirzepatide might affect cue -induced place preference 150 after prolonged abstinence. For this experiment we incorporated both environmental and 151 olfactory cues in the CPP paradigm along with a 14-day abstinence period. While vehicle-treated 152 mice retained strong preference for alcohol-associated contexts and cues on day 20, tirzepatide 153 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 7 treatment substantially reduced this preference on the last test day (t18=4.13, P<0.001; Fig. 1C). 154 These behavioral findings across locomotion, CPP , and cue -induced preference 155 suggested that tirzepatide affects central reward processing mechanisms. We therefore 156 employed microdialysis to investigate the neurochemical basis underlying these effects, 157 examining alcohol-induced dopamine release in the NAc across two experimental paradigms. 158 Systemic alcohol administration ( Fig. 1D ) produced pronounced elevation in NAc dopamine 159 release compared to vehicle (treatment F3,28=57.47, P<0.001, interaction F39,364=17.42, P<0.001). 160 Tirzepatide pretreatment substantially reduced this alcohol -induced dopamine response. Area 161 under the curve analysis confirmed that tirzepatide alone did not affect baseline dopamine 162 release (P>0.999), with no significant differences between vehicle controls and tirzepatide -163 alcohol treated mice (P=0.162). This effect moreover extended to alcohol -induced increases in 164 dopamine metabolites 3,4-dihydroxyphenylacetic acid (DOPAC), 3-methoxytyramine (3-MT), and 165 homovanillic acid (HVA), which tirzepatide similarly reduced ( Fig. S1A -C). We also detected 166 alterations in noradrenergic and serotonergic transmission ( Fig. S1D-G), though these changes 167 appeared less pronounced than the dopaminergic effects ( Fig. S1H). To confirm these findings 168 reflected tirzepatide's influence on NAc dopamine responses specifically, rather than indirect 169 systemic effects, we perfused alcohol locally through the probe ( Fig. 1E ). Local alcohol 170 application evoked robust dopamine increases that systemic tirzepatide administration 171 significantly attenuated (treatment F 1,14=61.52, P<0.001, interaction F 13,182=25.12, P<0.001), 172 suggesting tirzepatide can modulate alcohol's dopaminergic effects within the reward circuitry 173 itself. 174 Acute tirzepatide treatment dose-dependently reduce s alcohol consumption in 175 male and female rats 176 To further assess tirzepatide's effectiveness across different alcohol drinking phenotypes and 177 potential sex-specific effects, we conducted complementary experiments in male and female 178 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 8 rodents. These experiments evaluated tirzepatide's effects on voluntary alcohol consumption 179 across multiple established paradigms: the intermittent access two -bottle choice model (38), 180 the drinking in the dark (DID) model (39), and the alcohol -deprivation effect (ADE) model (40). 181 We first assessed acute tirzepatide effects using two doses (0.048 and 0.144 mg/kg, SC) 182 versus vehicle control on alcohol intake in the intermittent access paradigm to examine potential 183 dose-dependent effects in both sexes. In male rats, the lower dose (0.048 mg/kg) showed 184 minimal impact on alcohol, water, or food intake at 4 hours ( Fig. S2A-C). By 24 hours, this dose 185 significantly reduced alcohol intake (t11=7.81, P<0.001), food consumption (t 11=6.07, P<0.001), 186 and body weight (t 11=9.60, P<0.001), while water intake remained unchanged ( Fig. S2D-G). The 187 higher dose (0.144 mg/kg) produced more pronounced effects, decreasing alcohol intake 188 (t11=4.47, P=0.001) and food consumption (t 11=4.75, P<0.001) while water intake showed an 189 upward trend (t 11=2.04, P=0.066) at the 4 -hour timepoint ( Fig. S2H -J). At 24 hours, this dose 190 significantly reduced alcohol consumption (t 11=6.14, P<0.001; Fig. 2A), decreased food intake 191 (t11=18.90, P<0.001) and body weight (t11=15.30, P<0.001), while elevating water intake (t11=3.24, 192 P=0.008; Fig. S2K-M). Direct comparison indicated that 0.144 mg/kg produced a significantly 193 greater reduction in alcohol consumption ( -51.7±6.3%) compared to the 0.048 mg/kg dose ( -194 30.9±3.3%; t22=2.93, P=0.008), confirming dose-dependent effects (Fig. S2N). 195 Parallel studies in female rats revealed similar dose -dependent responses. The lower 196 tirzepatide dose (0.048 mg/kg) produced minimal changes in alcohol, water, or food intake at 4 197 hours ( Fig. S3A -C). At 24 hours, we found a trend toward reduced alcohol intake (t 11=1.91, 198 P=0.083), with significant decreases in food consumption (t 11=3.78, P=0.003) and body weight 199 (t11=2.92, P=0.014), while water intake remained unaffected (Fig. S3D-G). The higher dose (0.144 200 mg/kg) in females produced a trend toward lowered alcohol intake (t 11=1.65, P=0.128), caused 201 no change in food intake, and increased water intake (t 11=2.71, P=0.020) at 4 hours ( Fig. S3H-J). 202 At 24 hours, this dose significantly reduced alcohol consumption (t 11=6.96, P<0.001; Fig. 2B), 203 diminished food intake (t11=10.24, P<0.001) and body weight (t11=7.12, P<0.001), while increasing 204 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 9 water intake (t 11=4.39, P=0.001; Fig. S3K -M). Direct dose comparison in females showed 205 significantly greater alcohol intake suppression with the 0.144 mg/kg dose ( -41.3±4.3%) 206 compared to the 0.048 mg/kg dose (-11.7±5.7%; t₂₂=4.15, P<0.001), suggesting dose-dependent 207 effects (Fig. S3N). We found no significant sex differences with the 0.144 mg/kg dose on alcohol 208 intake between males ( -51.7±6.3%) and females ( -41.3±4.3%; t 22=1.37, P=0.183; Fig. 2C ). 209 Alcohol consumption also returned to baseline levels 48 hours post-treatment in both sexes (Fig. 210 S4A-B), indicating that the suppressive effect of an acute tirzepatide injection on alcohol intake 211 did not extend beyond this timeframe. 212 Tirzepatide attenuat es binge-like drinking in male and female mice with dose -213 response effects similar to semaglutide 214 Given that binge-like alcohol consumption is a clinically relevant concern (41), we tested whether 215 tirzepatide affects this drinking phenotype in male and female mice using the DID paradigm. Mice 216 received either vehicle or tirzepatide (0.144 mg/kg, IP) one hour before dark onset, followed by 217 four-hour alcohol access during the dark phase. Both male (t 18=3.66, P=0.002; Fig. 2D ) and 218 female mice (t18=6.41, P<0.001; Fig. 2E) showed significant reductions in binge-like drinking, with 219 no significant sex differences in efficacy: males (64.5±16.9%) versus females (77.2±14.9%) 220 relative to vehicle (t18=0.56, P=0.581; Fig. 2F). 221 To further explore tirzepatide's dose-response characteristics and compare them with 222 semaglutide, given the latter's emerging clinical evidence for alcohol reduction (29), we 223 conducted additional studies using lower doses (0.001 -0.072 mg/kg, IP). Both tirzepatide and 224 semaglutide produced dose -dependent reductions in binge -like drinking across male 225 (tirzepatide: F 6,63=34.83, P<0.001; semaglutide: F 6,60=21.36, P<0.001; Fig. S5A -B) and female 226 mice (tirzepatide: F6,62=24.76, P<0.001; semaglutide: F6,61=50.45, P<0.001; Fig. S5C-D). 227 228 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 10 Tirzepatide prevents relapse-like drinking in male and female rats 229 As preventing relapse remains one of the most persistent challenges in AUD treatment (9, 42, 43), 230 we next tested whether tirzepatide could also affect relapse-like drinking behavior. Using the ADE 231 model, which captures the temporary increase in alcohol consumption following forced 232 abstinence (40), we assessed tirzepatide's (0.144 mg/kg, SC) acute impact on this relapse -like 233 response. 234 Following 10 days of alcohol deprivation, vehicle -treated male rats developed the 235 expected ADE, with consumption rising significantly above baseline levels (F 3,36=9.30, P<0.001; 236 Fig. 2G ). Tirzepatide treatment appeared to block this response entirely. Rather than showing 237 increased drinking, treated rats exhibited a trend toward reduced alcohol intake compared to 238 baseline (P=0.073). Direct comparison revealed substantial between -group differences: 239 tirzepatide-treated males showed 48.3±4.4% reduction from baseline while vehicle -treated 240 counterparts increased consumption by 63.5±8.3% (t 18=11.90, P<0.001; Fig. 2H ). Additionally, 241 tirzepatide increased water intake while reducing both food consumption and body weight ( Fig. 242 S6A-C). 243 Female rats showed comparable responses . Vehicle-treated females developed robust 244 ADE (F 3,36=35.03, P<0.001; Fig. 2I), whereas tirzepatide treatment again prevented the rebound 245 response and actually decreased alcohol consumption below baseline (P<0.001). Tirzepatide 246 produced substantial prevention in females as well, with 56.5±4.6% reductions compared to 247 55.7±4.9% increases in vehicle controls (t18=16.80, P<0.001; Fig. 2J). As in males, tirzepatide also 248 increased water intake while decreasing food consumption and body weight (Fig. S6D-F). 249 250 Sustained eff ect of tirzepatide in reducing alcohol consumption with repeated 251 administration in male and female rats 252 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 11 To address whether tirzepatide maintains effectiveness during repeated administration, a n 253 important consideration for any potential addiction therapy (44), we examined the effects of 254 repeated tirzepatide (0.144 mg/kg, SC) or vehicle administration across six alcohol drinking days 255 spanning two weeks following the baseline period. 256 Male rats receiving repeated tirzepatide showed sustained reductions in alcohol 257 consumption (treatment F 1,18=20.66, P<0.001, interaction F 5,90=1.39, P=0.234; Fig. 3A ), with 258 consistent decreases compared to vehicle throughout the study period. Water intake remained 259 unaffected by treatment ( Fig. S7A ), while tirzepatide significantly reduced both food 260 consumption (treatment F 1,18=78.47, P<0.001, interaction F 5,90=12.67, P<0.001; Fig. S7B ) and 261 body weight (treatment F1,18=93.75, P<0.001, interaction F5,90=5.68, P<0.001; Fig. S7C) across the 262 experimental timeline. 263 Female rats demonstrated similar sustained responses during repeated treatment. 264 Tirzepatide produced comparable reductions in alcohol intake (treatment F 1,18=84.59, P<0.001, 265 interaction F 5,90=1.24, P=0.296; Fig. 3B ) that persisted throughout the two -week period. As 266 observed in males, water consumption remained stable ( Fig. S7D), while both food intake and 267 body weight decreased significantly under tirzepatide administration (Fig. S7E-F). 268 Direct comparison revealed no significant sex differences in treatment response (sex 269 F1,18=1.60, P=0.222, interaction F5,90=0.63, P=0.679), with females showing 63.4±4.1% reductions 270 compared to 54.3±7.1% in males (Fig. 3C). 271 272 Tirzepatide improves metabolic and inflammatory markers in alcohol -consuming 273 rats across both sexes 274 Long-term alcohol use can lead to fatty liver disease, metabolic syndrome, and systemic 275 inflammation, conditions that complicate treatment (35-37). GLP -1R agonists and tirzepatide 276 have shown effects on metabolic liver disease and inflammatory processes (45-48). Given the 277 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 12 notable body weight changes we observed during repeated treatment, we additionally explored 278 whether tirzepatide might also affect these metabolic and inflammatory parameters in alcohol -279 consuming animals. Preliminary insights on these effects might inform whether tirzepatide could 280 serve dual therapeutic roles, treating both alcohol use behaviors and the complications that 281 often accompany AUD. 282 In male rats, body weight changes diverged significantly between treatment groups 283 (treatment F1,18=6.39, P=0.021, interaction F 11,198=95.70, P<0.001; Fig. 3D), with vehicle -treated 284 rats gaining weight while tirzepatide -treated rats lost weight. Percentage body weight analysis 285 showed more pronounced between -group differences (F 1,18=181.80, P<0.001, interaction 286 F11,198=97.35, P<0.001; Fig. 3E ), with the tirzepatide group showing an average decrease of 287 8.5±0.6% and vehicle group an average increase of 3.0±0.2%. Tissue -specific analysis revealed 288 selective effects of this weight reduction. Tirzepatide treatment did not significantly alter 289 gastrocnemius muscle (t 18=1.32, P=0.204; Fig. 3F) or intrascapular brown adipose tissue (iBAT) 290 mass (t18=1.22, P=0.238) compared to vehicle. However, tirzepatide significantly reduced white 291 adipose tissue (WAT) deposits across multiple locations: subcutaneous inguinal (sWAT; t18=3.10, 292 P=0.006), gonadal (gWAT; t 18=2.12, P=0.048), and retroperitoneal and perirenal (rpWAT; 293 t18=1.619, P=0.005). Tirzepatide also decreased liver weight in treated males (t 18=2.15, P=0.045; 294 Fig. 3F ), with accompanying reductions in hepatic triglyceride content (t 18=3.70, P=0.002; Fig. 295 3G). Tirzepatide significantly reduced serum concentrations of pro -inflammatory cytokines 296 interleukin-6 (IL-6; t18=2.49, P=0.023) and tumor necrosis factor alpha (TNFα; t 18=2.17, P=0.043; 297 Fig. 3H) in male rats, while levels of IL-1β, IL-10, and monocyte chemoattractant protein-1 (MCP-298 1) remained unaffected. 299 Female rats exhibited comparable metabolic and inflammatory responses to tirzepatide 300 treatment. Body weight changes showed similar divergence between treatment groups 301 (F1,18=12.93, P=0.002, interaction F11,198=39.48, P<0.001; Fig. 3I), with tirzepatide inducing weight 302 reduction while vehicle -treated females gained weight. Percentage body weight changes 303 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 13 revealed an 8.8±0.6% decrease in tirzepatide-treated females compared to a 2.3±0.5% increase 304 in vehicle-treated counterparts (F1,18=136.70, P<0.001, interaction F11,198=40.17, P<0.001; Fig. 3J). 305 As observed in males, tirzepatide's effects on body composition in females were tissue-specific. 306 Tirzepatide preserved gastrocnemius muscle (t 18=2.02, P=0.058; Fig. 3K ) and iBAT mass 307 (t18=0.59, P=0.564), while significantly reducing WAT deposits across all measured depots: sWAT 308 (t18=4.49, P<0.001), gWAT (t 18=6.38, P<0.001), and rpWAT (t 18=3.84, P=0.001). Tirzepatide 309 significantly reduced liver weight in treated females (t 18=2.58, P=0.019; Fig. 3K ), with 310 concomitant decreases in hepatic triglyceride content (t 18=2.99, P=0.008; Fig. 3L ). The 311 inflammatory profile mirrored that observed in males, with significant reductions in serum IL -6 312 (t18=2.62, P=0.017) and TNFα (t18=2.16, P=0.044; Fig. 3M) levels following tirzepatide treatment. 313 314 Ex vivo electrophysiological recordings identified lateral septum as a po ssible 315 target for tirzepatide's neural effects 316 Before conducting a proteomic analysis of brain tissue from the repeated tirzepatide and long-317 term alcohol drinking experiment, where rats were sacrificed 24 hours after their final treatment, 318 we performed electrophysiological screening to identify which reward -related circuits (9, 49) 319 show detectable responses to tirzepatide treatment. Using alcohol -naïve male mice, we 320 examined neural activity 24 hours after acute tirzepatide administration (0.144 mg/kg, SC) across 321 several regions including NAc core and shell, medial prefrontal cortex (mPFC), dorsolateral and 322 dorsomedial striatum (DLS/DMS), and lateral septum (LS). 323 Analysis of the LS recordings revealed that tirzepatide exposure 24 hours earlier 324 produced a sustained suppression of evoked field potentials (F1,56=4.74, P=0.034; Fig. 4A), which 325 was concomitant with a significant increase in paired -pulse ratio (t 40=2.32, P=0.026; Fig. 4B), 326 indicative of a decreased probability of neurotransmitter release. This suggests tirzepatide 327 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 14 induced lasting presynaptic modifications within the LS region. In contrast, we observed no 328 sustained differences when comparing tirzepatide -exposed mice with vehicle -treated mice in 329 other reward -related circuits, including the mPFC (F 1,19=0.01, P=0.924), DMS (F 1,34=0.01, 330 P=0.907), DLS (F 1,58=1.06, P=0.308), NAc core (F ₁,₁₉=0.26, P=0.617), or NAc shell (F 1,58=0.48, 331 P=0.490; Fig. 4C-G). 332 333 Proteomic analysis of the LS reveals chromatin regulatory proteins as potential 334 targets of tirzepatide's effects 335 Based on electrophysiology results identifying the LS as a region showing sustained responses 336 to tirzepatide, we conducted a global quantitative proteomic analysis of the LS using tissue from 337 male alcohol -drinking rats that received repeated tirzepatide treatment. We identified 4,359 338 distinct proteins using TMT mass spectrometry, with statistical analysis showing 51 proteins 339 differentially expressed between tirzepatide and vehicle groups ( Fig. 4H, Table S1). Tirzepatide 340 upregulated 35 proteins and downregulated 16 proteins compared to vehicle. 341 Gene Ontology analysis identified several functional categories among the differentially 342 expressed proteins (DEPs). We found proteins previously linked to alcohol consumption, 343 including peroxisomal trans -2-enoyl-CoA reductase (PECR), midkine (MDK), acetyl -CoA 344 acyltransferase 2 (ACAA2), ATP-binding cassette subfamily G member 2 (ABCG2), and reticulon 345 1 (RTN1). Tirzepatide also affected proteins involved in neurotransmission, such as solute carrier 346 family 6 member 6 (SLC6A6), reticulon 3 (RTN3), proline-rich transmembrane protein 2 (PRRT2), 347 alpha-aminoadipic semialdehyde synthase (AASS), and microtubule -associated protein 1A 348 (MAP1A). We also detected changes in neuroinflammation -related proteins, including PECR, 349 signal regulatory protein alpha (SIRPA), and leucine-rich repeat containing 14 (LRRC14). 350 The DEPs also included several histone proteins: H1-0, H1-4, H2A-1A, and H3-3B. Given 351 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 15 that histone modifications and chromatin remodeling represent important epigenetic 352 mechanisms involved in addiction pathophysiology (10, 11), we focused detailed analysis on GO 353 terms related to histone and chromatin processes. Cross -referencing with the UniProtKB 354 database (50) and existing literature revealed a total of 11 proteins associated with histone or 355 chromatin regulatory functions, visualized as a heat map (51) in Fig. 4I (with references listed in 356 Table S1 ). Statistical analysis showed significant differences between tirzepatide and vehicle 357 groups in these 11 proteins: ataxin -3 (ATXN3; t 15.9=2.88, P=0.011), histone H1 -0 (t 10.7=2.93, 358 P=0.014), histone H1-4 (t15.7=2.50, P=0.024), POU domain class 2 transcription factor 1 (POU2F1; 359 t15.7=2.14, P=0.049), dual specificity protein phosphatase 12 (DUSP12; t 15.5=3.24, P=0.005), 360 histone H2A-1A (t14.2=3.17, P=0.007), chromobox protein homolog 7 (CBX7; t 14.2=3.12, P=0.007), 361 histone H3-3B (t12.5=3.43, P=0.005), high mobility group nucleosome-binding domain-containing 362 protein 2 (HMGN2; t 14.2=3.41, P=0.004), PIH1 domain -containing protein 1 (PIH1D1; t 10.5=3.19, 363 P=0.009), and aprataxin (APTX; t14.0=2.22, P=0.043) (Fig. 4J). 364 365 366 367 368 369 370 371 372 373 374 375 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 16

376 Here we demonstrate that tirzepatide, a dual GLP -1R/GIPR agonist, affects alcohol intake and 377 alcohol-related responses across both sexes in rodents. Our findings reveal that tirzepatide 378 attenuates alcohol-induced dopamine signaling, reduces alcohol consumption, and suppresses 379 relapse-like behaviors. We also observed changes in metabolic and inflammatory parameters 380 linked to alcohol use. Through electrophysiological and proteomic approaches, we identified the 381 LS as what may be an important neuroanatomical substrate, with proteomic data suggesting 382 potential epigenetic involvement in tirzepatide's effects. These findings indicate that tirzepatide 383 may represent a promising therapeutic candidate for AUD, potentially addressing both drinking 384 behavior and the physiological consequences of alcohol intake that likely contribute to poor 385 treatment outcomes and relapse vulnerability. 386 387 Our initial investigation examined tirzepatide's effects on reward -related responses and 388 revealed significant suppression of alcohol -induced dopamine processing. Tirzepatide 389 consistently reduced alcohol-induced locomotor stimulation, place preference, and accumbal 390 dopamine release in male mice across these paradigms. This modulation is noteworthy given 391 that alcohol -induced dopamine release in the NAc appears to contribute to consummatory 392 behaviors and likely represents a risk factor for AUD development (5-8). Perhaps most 393 compelling was our observation that tirzepatide suppressed alcohol-induced dopamine release 394 regardless of whether alcohol was administered systemically or perfused locally within the NAc 395 itself. This suggests tirzepatide may directly influence the reward circuitry, though the precise 396 mechanisms warrant further investigation. These findings align with previous work showing that 397 GLP-1R agonists affect dopaminergic reward pathways across different substances of abuse 398 (23-25, 27, 52-58). 399 The observed effects on reward processing appeared to translate into substantial 400 reductions in alcohol -drinking behavior. Tirzepatide consistently decreased voluntary alcohol 401 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 17 intake across both sexes in multiple experimental paradigms. Single administration produced 402 dose-dependent reductions in alcohol consumption in rats while also suppressing binge -like 403 drinking in mice, suggesting cross -species effectiveness. These observations extend what we 404 and others have seen with GLP -1R agonists, where similar alcohol intake -suppressing effects 405 have emerged across various preclinical models (21, 22, 25, 26, 28, 59-62). 406 Our comparison with semaglutide using similar dose ranges revealed similar acute 407 effects on binge -like drinking for both compounds. This might suggest comparable therapeutic 408 potential for both drugs when administered acutely, though it should be recognized that optimal 409 dose ranges will likely differ between compounds in clinical settings, making direct dose 410 comparisons challenging to interpret. A m ore intriguing observation was that the r epeated 411 tirzepatide administration appeared to maintain its suppressive effects more consistently than 412 what we saw in our previous work with semaglutide (25), though we acknowledge this 413 comparison spans different studies with inherent limitations. The clinical landscape looks 414 increasingly promising. Early studies show that GLP-1R agonists reduce alcohol consumption in 415 humans (29-32), suggesting our preclinical models might capture clinically relevant 416 mechanisms. Cross -species translation always demands careful interpretation, though. Even 417 more encouraging, a recent case -control study found reduced alcohol consumption in obese 418 patients receiving tirzepatide (63), providing additional support for potential therapeutic 419 applications. While these observations emerge from metabolic treatment contexts rather than 420 addiction-focused studies, they provide what appears to be real -world validation of our 421 experimental findings. Since tirzepatide and other GLP -1R agonists already have clinical 422 approval for type 2 diabetes and obesity, a solid foundation exists for further AUD investigation. 423 This practical advantage appears to have facilitated current clinical initiatives, with multiple 424 studies now exploring incretin agonists potential as addiction treatment. 425 Our findings on relapse -like behaviors add further relevance to these clinical 426 applications. Relapse remains a clinical challenge in AUD and effective treatment options 427 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 18 remain scarce (9, 42, 43). We found that tirzepatide blocks ADE, a model of relapse-like drinking 428 (40), which aligns with our earlier work with liraglutide and semaglutide (25, 59). Tirzepatide also 429 attenuated cue -induced place preference following forced abstinence, suggesting possible 430 effects on drug memory and cue reactivity processes (42, 43). Existing literature indicates that 431 GLP-1R agonists also suppress reinstatement for cocaine (56), nicotine (57), and opioids (58), 432 further supporting a potential effect on relapse -like behaviors. Furthermore, our findings seem 433 translationally promising, given recent reports that semaglutide reduces alcohol cravings in 434 humans (29) and that exenatide diminishes cue reactivity in AUD patients (30). Both anecdotal 435 and research -based observations also support this interpretation, with reports of reduced 436 cravings for food and substances during GLP-1R agonist treatment (63-65). As both cravings and 437 environmental cues often precipitate relapse (42, 43), our findings further support the emerging 438 evidence that incretin agonists could help mitigate relapse risk. 439 The repeated administration study also revealed effects on metabolic and inflammatory 440 parameters that warrant further investigation. Given planned clinical trials exploring tirzepatide 441 and semaglutide for alcohol liver disease and AUD patients with metabolic comorbidities 442 (ClinicalTrials.gov identifiers: NCT06546384, NCT06409130, and NCT07046819), these findings 443 could take on added significance. While incretin agonists demonstrate established benefits in 444 metabolic and inflammatory conditions (45-48), their effects in long -term alcohol -consuming 445 populations remain less well characterized. The question is whether they maintain these effects 446 when alcohol consumption is involved. Our findings suggest they might. Repeated tirzepatide 447 treatment reduced body weight, adipose tissue mass, liver weight, and hepatic triglyceride 448 content, and also decreased pro -inflammatory cytokines (IL -6, TNFα) in both male and female 449 alcohol-consuming rats. These observations suggest tirzepatide may offer therapeutic potential 450 for alcohol-related complications such as fatty liver disease, a finding with clear implications for 451 patients managing concurrent AUD and metabolic disorders. However, clinical studies 452 specifically in alcohol -using populations will be necessary to validate these applications. 453 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 19 Moreover, our alcohol consumption studies demonstrated generally consistent effects across 454 sexes, though future work should examine whether subtle sex -specific mechanisms exist at 455 molecular and circuit levels. 456 457 Given the consistent effects across sexes, our subsequent investigations focused on 458 male rodents to identify potential brain regions mediating tirzepatide's effects on alcohol-related 459 responses. We started with electrophysiological recordings across several reward -related 460 circuits. This approach led us to the LS, where we observed the most pronounced changes in 461 neural activity, including increased paired -pulse ratios suggesting modified presynaptic 462 neurotransmitter release probability. These LS findings seemed particularly significant given the 463 region's established role in reward processing and drug -seeking behaviors (49, 66, 67) . Several 464 factors made this discovery especially compelling. The LS sits strategically near the subfornical 465 organ, a circumventricular structure that might serve as an access point for peripherally 466 administered incretin agonists (68). It also maintains direct connections to the NAc and ventral 467 tegmental area (49, 66, 67), positioning it to potentially influence mesolimbic dopamine signaling 468 (69). Both GLP-1R (70, 71) and GIPR (72, 73) are expressed in the LS, and peripherally 469 administered GLP -1R agonists can reach this region (21, 74, 75) . A recent clinical stud y have 470 further shown that these drugs can reduce septal cue reactivity to alcohol in AUD patients (30). 471 Previous preclinical work has also demonstrated LS involvement in GLP -1R-mediated reward 472 processing for both alcohol (21, 24) and cocaine-related behaviors (52, 76), as well as regulation 473 of alcohol -induced dopamine release in the NAc (24). This convergent evidence led us to 474 hypothesize that the LS might serve as a neuroanatomical substrate mediating tirzepatide's 475 effects on mesolimbic dopamine signaling and alcohol-related responses. 476 To further explore potential LS-related molecular mechanisms, we conducted a 477 proteomic analysis of LS tissue from alcohol-consuming male rats from the repeated tirzepatide 478 treatment experiment. This revealed differential expression of several histone and chromatin 479 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 20 regulatory proteins, including specific histone proteins (H1 -0, H1 -4, H2A -1A, H3 -3B). These 480 findings seemed noteworthy because histones regulate gene expression and serve as essential 481 components of chromatin architecture (10, 13). Given that alcohol exposure can cause histone 482 modifications, such as acetylations and methylations , which may contribute to addiction 483 pathophysiology (10, 11, 77, 78), our results suggest tirzepatide might influence epigenetic 484 mechanisms involved in alcohol drinking behaviors. Recent clinical research with semaglutide 485 in obesity has identified similar proteomic changes in blood samples, including proteins linked 486 to substance use disorders , such as histones (79). These findings hint that GLP -1R-based 487 therapeutics may engage epigenetic mechanisms, tentatively explaining their broad efficacy 488 across multiple disorders. However, future studies will need to investigate these potential 489 mechanisms more thoroughly. 490 491 Our experimental approach provides solid evidence for tirzepatide's therapeutic 492 potential, though several important considerations deserve attention. We were careful to 493 exclude potential confounding factors like sedation, anhedonia, and malaise since tirzepatide 494 did not affect baseline locomotor activity, dopamine levels per se, or kaolin consumption. Even 495 so, we did not specifically assess anxiety-like behaviors or taste aversion, which might influence 496 alcohol-related responses. Our proteomic analysis focused on just one brain region, limiting our 497 ability to understand broader molecular changes across reward circuits. An important 498 mechanistic question also remains unanswered: did tirzepatide directly cause the molecular 499 changes we observed, or did they simply result from reduced alcohol intake? This distinction 500 seems crucial for understanding how tirzepatide actually works long -term. Targeted 501 experimental designs could help address this to better understand the molecular effects we are 502 seeing in the LS. Future studies should also explore the brain -circuit connectivity between LS 503 GLP-1R/GIPR pathways and other regions since this could clarify which specific neural circuits 504 contribute to tirzepatide's effects on reward processing and alcohol consumption. It would also 505 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 21 prove valuable to directly compare tirzepatide versus semaglutide within the same study to 506 understand their different mechanisms. Another limitation worth noting: several experiments 507 used only males, which probably limits our insights into sex -specific mechanisms. Even with 508 these considerations, our findings suggest that tirzepatide influences the reward circuitry and 509 may serve as a therapeutic candidate for AUD and related alcohol-related conditions. 510 511 In summary, our findings indicate that tirzepatide influences alcohol -related responses 512 in ways that appear to have clinical potential. Tirzepatide consistently reduced alcohol intake 513 across different drinking paradigms and both sexes without signs of tolerance development. 514 Perhaps more significantly, tirzepatide's effects on relapse behaviors suggest it might help 515 decrease relapse vulnerability, a finding that could prove important for therapeutic applications. 516 The mesolimbic reward effects offer insights into tirzepatide's possible mode of action. Our data 517 suggest tirzepatide influences dopaminergic reward processes to suppress alcohol -drinking 518 behaviors. The LS findings add another piece to this puzzle, offering some initial clues about a 519 neural substrate where dual incretin receptor signaling might exert these effects. These results, 520 combined with tirzepatide's existing clinical approval, position this dual incretin agonist as a 521 promising therapeutic candidate for AUD and alcohol -related diseases, that warrants clinical 522 investigation. 523 524 525 526 527 528 529 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 22

AND MATERIALS 530 531 Animals 532 Adult male NMRI mice (8-10 weeks old, 25-30 g; Charles River, Sulzfeld, Germany) were used for 533 locomotor activity, CPP , microdialysis, food intake, and electrophysiological studies. Binge-like 534 drinking paradigms employed adult male and female C57BL/6J mice (8 -10 weeks old, 25 -30 g; 535 Jackson Laboratories, Bar Harbor, ME, USA). Intermittent access two -bottle choice alcohol 536 studies utilized adult male and female Rcc/Han Wistar rats (8 -9 weeks old, 180 -250 g; Envigo, 537 Horst, Netherlands), with tissue collected post-mortem for molecular analyses. The three rodent 538 strains were selected based on their established responsivity to alcohol and gut -brain peptides 539 (22-24, 80). Animals were group-housed upon arrival and acclimated for at least one week under 540 standardized conditions (12/12-hour light/dark cycle, 20°C, 50% humidity) with ad libitum access 541 to standard chow (Harland Teklad Rodent Diet #2916 & 2918, Madison, WI, USA) and water . 542 Animals used for microdialysis and alcohol intake studies were subsequently single-housed after 543 surgery or at the start of the alcohol baseline period to prevent implant damage and allow for 544 individual consumption measurements. Behavioral and microdialysis experiments were 545 conducted during the light phase when stimulation effects are more pronounced, with 60-minute 546 habituation to the testing environment. Alcohol intake studies were performed during dark and 547 light phases for rats and exclusively during the dark phase for mice, when drinking behavior is 548 heightened. All experiments received approval from the Ethics Committee for Animal Research 549 in Gothenburg, Sweden (ethical permits: 4685/23, 3348/20, 3276/20) or the Institutional Animal 550 Care and Use Committee at the Medical University of South Carolina, USA. Studies adhered to 551 the NIH Guide for the Care and Use of Laboratory Animals, ARRIVE guidelines, and 3Rs principle. 552 553 554 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 23 Drugs 555 For behavioral and neurochemical experiments, alcohol (95% Ethanol, Solveco; Stockholm, 556 Sweden or Warner Graham Co., Cockeysville, MD, USA) was diluted with vehicle (0.9% NaCl) to 557 a 15% (w/v) solution and administered IP at 1.75 g/kg, 5 minutes prior to testing. Microdialysis 558 experiments required local alcohol administration, achieved by diluting alcohol in modified 559 Ringer's solution (140 mM NaCl, 1.2 mM CaCl₂, 3.0 mM KCl, and 1.0 mM MgCl₂, Sigma -Aldrich, 560 Darmstadt, Germany) to 300 mM, corresponding to approximately 50 -60 mM outside the probe 561 in the NAc (81). These alcohol doses and concentrations were selected for their established 562 ability to stimulate the mesolimbic dopamine system with reproducible effects (22-24). Alcohol 563 drinking studies employed a 20% (v/v) solution prepared with tap water. 564 Tirzepatide (LY3298176 HCl, MedChemExpress, Sollentuna, Sweden) was dissolved in vehicle 565 (40 mM Tris-HCl, pH 8.0) and administered SC 30 minutes before behavioral testing or alcohol 566 exposure. Dose selection was guided by initial dose -response studies (0.048, 0.144, 0.240, and 567 0.336 mg/kg) that evaluated effects on locomotor activity and food/kaolin intake. While no dose 568 altered baseline two -hour locomotor activity or gross behavior ( Fig. S8A -D), dose -dependent 569 decreases in food intake and body weight appeared at 24 hours without affecting kaolin or water 570 intake ( Fig. S9A -D). Based on these findings, 0.144 mg/kg was selected for subsequent 571 experiments to achieve consistent effects 30 minutes post-administration. The lower dose (0.048 572 mg/kg) was additionally tested in acute alcohol drinking studies to evaluate dose -dependent 573 effects. Binge-like alcohol drinking experiments compared lower tirzepatide doses (0.001, 0.003, 574 0.009, 0.018, 0.036, and 0.072 mg/kg) with corresponding doses of semaglutide 575 (MedChemExpress). Both compounds were administered IP one hour before dark onset in these 576 studies. To validate this methodological variation, we compared SC versus IP administration 577 effects on food intake for both tirzepatide ( Fig. S9A -D) and semaglutide at a dose previously 578 shown to attenuate alcohol -related behaviors (25) (Fig. S9E -H). Both administration routes 579 produced comparable effects (Fig. S9A-H). 580 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 24 581 Locomotor activity 582 Horizontal and vertical activity were recorded in six sound-attenuated, ventilated, and dimly lit (3 583 lx) locomotor boxes (42×42×20 cm; Open Field Activity System, Med Associates Inc., Georgia, VT, 584 USA). Movement was detected via a two -layered infrared photobeam grid and recorded using 585 Activity Monitor software (Version 7, Med Associates Inc.) (22-24). Male mice underwent 60 -586 minute habituation in the test arena before first receiving tirzepatide treatment and then 30 587 minutes later an alcohol injection. Activity recording began 5 minutes after the final injection and 588 continued for 60 minutes. 589 590 Conditioned place preference 591 CPP experiments employed four two -chambered arenas (50×24×24 cm, custom -made, 592 University of Gothenburg, Gothenburg, Sweden) under dim lighting (3 lx), with chambers 593 distinguished by distinct tactile and visual cues (22-24). The protocol was conducted in male 594 mice, beginning with a 20 -minute pre -test (day 1) to assess initial place preference following 595 vehicle injection. Conditioning sessions (days 2 -5, 20 minutes each) followed a biased design, 596 pairing alcohol with the least preferred chamber and vehicle with the preferred chamber. Daily 597 sessions included one alcohol injection and one vehicle injection in a balanced design, 598 alternating between morning and afternoon. On test day (day 6), mice received tirzepatide or 599 vehicle before place preference monitoring for 20 minutes. A control experiment was conducted 600 to assess tirzepatide's effect on CPP independently of alcohol, following identical procedures but 601 employing vehicle injections in both chambers during conditioning. A third experiment 602 investigated tirzepatide's influence on cue -induced place preference following forced 603 abstinence, using the same biased alcohol paradigm with added neutral-valence olfactory cues 604 (82, 83) . Caraway odor (S -carvone, Sigma -Aldrich) was paired with the alcohol chamber and 605 mineral oil (Sigma-Aldrich) with the vehicle chamber. Odorants (one drop) were applied to filter 606 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 25 paper in perforated plastic tubes positioned at chamber tops. This protocol included a first test 607 day (day 6) where mice received vehicle before a 20 -minute test with olfactory cues present. 608 Following two weeks of forced home cage abstinence, mice were divided into equal groups based 609 on first test day results and received tirzepatide or vehicle before a second 20 -minute test (day 610 20) with olfactory cues present. All experiments were analyzed using Observer XT software 611 (Version 15, Noldus, Wagenegen, Netherlands). CPP expression was calculated as the difference 612 in percentage of total time spent in the drug -paired compartment between pre -test and test 613 sessions. 614 615 Microdialysis 616 An I-shaped microdialysis probe (20 kDa cut-off membrane with 1 mm exposed length, HOSPAL, 617 Gambro, Sweden) was surgically implanted in the NAc shell four days before experiments as 618 previously described (22-24). Mice were anesthetized with isoflurane (Baxter, Apoteket AB, 619 Gothenburg, Sweden), placed in a stereotaxic frame, and maintained on a heating pad. Local 620 anesthesia (Xylocaine with adrenaline, 10 mg/ml, 5 μg/ml; Pfizer Inc, Apoteket AB, Gothenburg, 621 Sweden) was applied at the incision site. Carprofen (Rimadyl® , 5 mg/kg, AstraZeneca, Apoteket 622 AB, Gothenburg, Sweden), 0.9% NaCl, and Viscotears were administered for pain management, 623 rehydration, and eye protection. After exposing the skull, holes were drilled for the probe and 624 anchoring screws. The probe was secured with dental cement (DENTALON® Plus, Agntho's AB, 625 Lidingö, Sweden). On experiment days, the probe was connected to a pump and perfused with 626 Ringer's solution at 1.6 μl/min. After a two-hour equilibration period, samples were collected at 627 20-minute intervals throughout the experiment. Following baseline measurements (minutes -40 628 to 0), tirzepatide or vehicle was administered at minute 10. Thirty minutes later (minute 40), 629 alcohol was either injected systemically (Experiment 1) or perfused through the probe for the 630 remainder of the experiment (40 -220 minutes, Experiment 2). Nine additional samples were 631 collected following alcohol exposure. Probe placement was verified histologically using a brain 632 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 26 atlas (84), and only data from correctly placed probes without hemorrhage were included in 633 analyses (Fig. S10A-B). Microdialysate samples were analyzed using HPLC with electrochemical 634 detection, as described before (24). Changes in dopamine and other monoamines and their 635 metabolites were calculated as percentages of the mean of three baseline values before 636 tirzepatide/vehicle treatment. The area under the curve following alcohol exposure (40 -220 637 minutes) was additionally calculated for further analysis. 638 639 Intermittent access two-bottle choice and the alcohol deprivation effect 640 This paradigm provided rats with alcohol and water access during three 24-hour sessions weekly 641 (Monday, Wednesday, Friday), with water-only access on intervening days (24, 38). Bottles were 642 switched at dark -phase onset, while food and water remained continuously available. Rats 643 underwent an 8-week baseline period before experimental interventions, during which alcohol, 644 water, and food intake were measured daily and body weight recorded weekly. Following the 645 baseline period, rats were allocated to treatment groups with matched baseline alcohol intake 646 levels. During experiments , consumption measurements occurred at 4 and 24 hours post -647 treatment, with corresponding 24 -hour body weight changes documented. To assess 648 tirzepatide's impact on relapse-like behavior, we utilized the ADE model (25, 40) . This protocol 649 involved an 8-week baseline alcohol consumption period followed by 10-day alcohol deprivation 650 within the intermittent access paradigm. Rats then received single tirzepatide or vehicle 651 administration before alcohol reintroduction, with relapse-like drinking quantified as percentage 652 change from baseline intake. A separate study examined repeated tirzepatide or vehicle 653 administration effects spanning six alcohol drinking days across two weeks following the 654 baseline period. This design allowed assessment of treatment efficacy over extended an 655 timeframe. Upon completing the repeated administration study, rats were euthanized 24 hours 656 after final treatment following a full day of alcohol access. Brains were rapidly removed, flash -657 frozen, and stored at -80°C for subsequent analysis. We additionally dissected and weighed 658 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 27 several metabolically relevant tissues: gastrocnemius muscle, iBAT, sWAT, gWAT, rpWAT, and 659 liver. The liver's middle lobe was flash-frozen and stored at -80°C. Trunk blood was also collected 660 using serum tubes (Z-gel tubes with clotting activator, Sarstedt, Germany) and stored at -80°C for 661 later analysis. 662 663 Drinking in the dark 664 A standardized DID protocol in adult male and female mice were used to examine binge -like 665 alcohol drinking (39). The experimental design utilized repeated four-day alcohol access cycles, 666 where days 1-3 involved two-hour sessions and day 4 extended to four hours as the primary test 667 day. Three days without alcohol separated each cycle. Sessions started three hours after dark -668 phase onset, with food remaining available throughout while water was temporarily removed 669 during alcohol access periods. The protocol timing required modification based on 670 pharmacological considerations. Our initial intermittent access studies suggested that lower 671 tirzepatide doses might not produce measurable effects within the standard four -hour 672 assessment window, prompting this temporal adjustment to better capture potential treatment 673 effects. 674 675 Measurements of liver triglycerides and serum cytokines 676 Liver tissue samples from the repeated alcohol drinking experiment were processed for 677 triglyceride analysis using standard lipid extraction techniques. Tissue was lysed in 2:1 678 chloroform:methanol and washed with 0.9 M NaCl to achieve phase separation, as previously 679 described (85). The triglyceride-containing lower phase was collected and evaporated overnight. 680 Dried triglyceride pellets were resuspended in isopropanol (Sigma Aldrich) and quantified using 681 a commercial triglyceride kit (Randox Laboratories Ltd, Crumlin, UK) according to manufacturer's 682 instructions. Absorbance was measured at 500 nm with 546 nm correction using a Spectramax 683 i3x multiplate reader (Molecular Devices, San Jose, CA, USA). Serum cytokine levels were 684 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 28 measured using a custom -ordered Bio -Plex Pro™ rat Cytokine Assay kit (10014905, Bio -Rad, 685 Hercules, CA, USA) to quantify five cytokines, based on the literature (86, 87): IL-1β, IL-6, IL-10, 686 TNFα, and MCP -1 using the Bio -Plex 200 system (Bio -Rad) according to manufacturer's 687 instructions. 688 689 Electrophysiological recordings 690 Coronal brain slices (300 μm) were prepared 24 hours after tirzepatide or vehicle administration, 691 as previously described (24, 88) . Field potentials were evoked using a stimulation electrode 692 positioned near (0.2-0.3 mm) the recording electrode in each brain region (NAc core/shell, mPFC, 693 DLS, DMS, LS) . Population spike (PS) amplitudes were evoked using seven -step increasing 694 stimulation protocols. Paired-pulse stimulation (50 ms interpulse interval, 0.1 Hz) was used to 695 calculate paired-pulse ratio (PPR, PS2/PS1), to estimate changes in the probability of transmitter 696 release. Data were acquired using Clampfit 10.2 software (Molecular Devices , Foster City, CA, 697 USA). 698 699 Proteomics - Global relative quantification 700 The LS was microdissected using a brain-slicing matrix on dry ice, weighed, and stored at -80°C 701 until analysis as previously described (24). Protein extraction employed lysis buffer (2% sodium 702 dodecyl sulfate, 100 mM triethylammonium bicarbonate) with a Covaris ML230 ultrasonicator. 703 Protein concentrations were determined using the Pierce BCA Protein Assay Kit (Thermo 704 Scientific, Gothenburg, Sweden). Sample processing followed a modified SP3 method. Samples 705 and references (40 μg) underwent reduction (100 mM DTT), alkylation (20 mM iodoacetamide), 706 and precipitation on Sera -Mag™ SpeedBeads (Cytiva, Uppsala, Sweden) using ethanol. After 707 washing and drying, beads were resuspended in 100 mM TEAB for protein digestion with 708 Trypsin/Lys-C mix (1:25) for two hours, followed by trypsin (1:50) overnight. Following bead 709 removal, peptide concentrations were determined using Pierce ™ Quantitative Fluorometric 710 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 29 Peptide Assay (Thermo Scientific). Peptide samples (20 μg) were labeled using TMT pro 18 -plex 711 isobaric mass tagging reagents (Thermo Fisher Scientific), pooled into one TMT-set, and purified 712 using HiPPR Detergent Removal Resin and Pierce™ Peptide Desalting Spin Columns. The TMT-set 713 underwent basic reversed-phase chromatography (bRP-LC, pH10) fractionation into 36 fractions 714 over 70 minutes. Mass spectrometry analysis employed an Orbitrap Eclipse Tribrid mass 715 spectrometer with FAIMS Pro ion mobility system interfaced with an nLC 1200 liquid 716 chromatography system. Peptides were separated on a C18 35 cm column over 90 minutes, with 717 data acquired using SPS MS3 methodology. Raw files were processed using Proteome Discoverer 718 (Ver 3.0, Thermo Scientific) against UniProt Swiss -Prot Rattus norvegicus database using 719 Sequest search engine. Only unique peptides were used for relative quantification, with proteins 720 required to pass a 5% false discovery rate threshold. Proteins showing significant expression 721 changes were cross -referenced with existing literature and the UniProtKB database (50) to 722 identify those linked to histone and chromatin processes (complete reference list in Table S1). 723 724 Statistics 725 Statistical analyses were performed using GraphPad Prism (version 10.4.1, GraphPad Software 726 Inc., Boston, MA, USA). The statistical approach was as described previously (24). In brief, normal 727 distribution was assessed using the Shapiro-Wilk test. All subsequent tests were two-tailed with 728 significance threshold at p<0.05. For comparisons between two groups in behavioral, intake, or 729 electrophysiology experiments, paired or unpaired Student's t-tests were applied as appropriate. 730 Comparisons among three or more groups employed one -way ANOVA with Bonferroni post-hoc 731 tests. For experiments with repeated measures (microdialysis, repeated alcohol drinking, and 732 electrophysiology), repeated -measures two -way ANOVA with Bonferroni post -hoc tests were 733 utilized. Welch's t-test was used on log2 -transformed data to identify DEPs . Proteins with a p -734 value<0.05 and fold -change ≥10% (log2 fold -change ≤ -0.137 or ≥0.137) were considered as 735 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 30 differentially expressed. All data are presented as mean ± standard error of the mean (SEM) with 736 individual values shown when appropriate. 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 31

769 The authors gratefully acknowledge the technical assistance and expertise of Ebba Frövenholt, 770 Erika Lucente, Manisha Bergum Samad and Anna -Lena Leverin. Proteomic analysis was 771 performed at the Proteomics Core Facility, Sahlgrenska academy, Gothenburg University, with 772 financial support from SciLifeLab and BioMS. Animations were created using BioRender. 773 774 Funding: The study is supported by grants from the Swedish Research Council (2023-2600, 2020-775 00559, 2020-01463, 2024-03054), LUA/ALF (grant no. 723941 & 1005347 ) from the Sahlgrenska 776 University Hospital, Alcohol Research Council of the Swedish Alcohol Retailing Monopoly 777 (FO2024-0048), Herbert & Karin Jacobssons Foundation (2024 -Forskning-225), Adlerbertska 778 Research Foundation (2024 -791), Wilhelm & Martina Lundgren’s Research Foundation (2024 -779 SA-4698) and Mary von Sydow Foundation (2024 -36). Thaynnam A Emous held an international 780 internship scholarship from the São Paulo Research Foundation (FAPESP), Process Number 781 #2023/18470-5, while conducting research at the University of Gothenburg. 782 783 Data availability: All data sets generated or analyzed during this study are available in the Source 784 Data file. The MS proteomics data have been deposited to the ProteomeXchange Consortium 785 (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository with the data set 786 identifier PXD063324. Any additional information required to reanalyze the data reported in this 787 paper is available upon request from the corresponding author. 788 789 790 791 792 793 794 795 796 797 798 799 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 32

800 1. B. F . Grant et al., Epidemiology of DSM-5 Alcohol Use Disorder: Results From the 801 National Epidemiologic Survey on Alcohol and Related Conditions III. JAMA 802 Psychiatry 72, 757-766 (2015). 803 2. M. Heilig et al., Addiction as a brain disease revised: why it still matters, and the 804 need for consilience. Neuropsychopharmacology 46, 1715-1723 (2021). 805 3. W. Yaseen, J. Mong, J. Zipursky, Sobering Perspectives on the Treatment of Alcohol 806 Use Disorder. JAMA Netw Open 7, e243340 (2024). 807 4. K. C. Berridge, T. E. Robinson, Liking, wanting, and the incentive -sensitization 808 theory of addiction. Am Psychol 71, 670-679 (2016). 809 5. H. de Wit, T. J. Phillips, Do initial responses to drugs predict future use or abuse? 810 Neurosci Biobehav Rev 36, 1565-1576 (2012). 811 6. G. Di Chiara, Alcohol and dopamine. Alcohol Health Res World 21, 108 -114 812 (1997). 813 7. A. King et al. , Subjective Responses to Alcohol in the Development and 814 Maintenance of Alcohol Use Disorder. American Journal of Psychiatry 178, 560-815 571 (2021). 816 8. A. C. King et al. , Real -Time Assessment of Alcohol Reward, Stimulation, and 817 Negative Affect in Individuals With and Without Alcohol Use Disorder and 818 Depressive Disorders. Am J Psychiatry 182, 187-197 (2025). 819 9. G. F . Koob, N. D. Volkow, Neurobiology of addiction: a neurocircuitry analysis. 820 Lancet Psychiatry 3, 760-773 (2016). 821 10. E. J. Nestler, C. Luscher, The Molecular Basis of Drug Addiction: Linking Epigenetic 822 to Synaptic and Circuit Mechanisms. Neuron 102, 48-59 (2019). 823 11. D. Ron, S. Barak, Molecular mechanisms underlying alcohol-drinking behaviours. 824 Nat Rev Neurosci 17, 576-591 (2016). 825 12. J. A. Kauer, R. C. Malenka, Synaptic plasticity and addiction. Nat Rev Neurosci 8, 826 844-858 (2007). 827 13. E. J. Nestler, The biology of addiction. Sci Signal 18, eadq0031 (2025). 828 14. N. Bruns Vi, E. H. Tressler, L. F . Vendruscolo, L. Leggio, M. Farokhnia, IUPHAR 829 review - Glucagon-like peptide -1 (GLP -1) and substance use disorders: An 830 emerging pharmacotherapeutic target. Pharmacol Res 207, 107312 (2024). 831 15. E. Jerlhag, Gut-brain axis and addictive disorders: A review with focus on alcohol 832 and drugs of abuse. Pharmacol Ther 196, 1-14 (2019). 833 16. M. Kruse Klausen, M. Thomsen, G. Wortwein, A. Fink-Jensen, The role of glucagon-834 like peptide 1 (GLP-1) in addictive disorders. Br J Pharmacol 179, 625-641 (2022). 835 17. G. N. Petrie, L. M. Mayo, GLP-1 receptor agonists for the treatment of alcohol use 836 disorder. J Clin Invest 135 (2025). 837 18. A. K. Singal, L. Leggio, GLP -1 receptor agonists in alcohol use disorder and 838 alcohol-associated liver disease. Lancet Gastroenterol Hepatol 10, 707 -709 839 (2025). 840 19. J. J. Holst, Incretin hormones and the satiation signal. Int J Obes (Lond) 37, 1161-841 1168 (2013). 842 20. C. M. Kusminski et al., Transforming obesity: The advancement of multi-receptor 843 drugs. Cell 187, 3829-3853 (2024). 844 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 33 21. M. L. Allingbjerg, S. N. Hansen, A. Secher, M. Thomsen, Glucagon -like peptide-1 845 receptors in nucleus accumbens, ventral hippocampus, and lateral septum 846 reduce alcohol reinforcement in mice. Exp Clin Psychopharmacol 31, 612 -620 847 (2023). 848 22. E. Egecioglu et al. , The glucagon -like peptide 1 analogue Exendin -4 attenuates 849 alcohol mediated behaviors in rodents. Psychoneuroendocrinology 38, 1259 -850 1270 (2013). 851 23. D. Vallof, A. L. Kalafateli, E. Jerlhag, Brain region specific glucagon -like peptide-1 852 receptors regulate alcohol -induced behaviors in rodents. 853 Psychoneuroendocrinology 103, 284-295 (2019). 854 24. C. E. Edvardsson, D. Cadeddu, M. Ericson, L. Adermark, E. Jerlhag, An inhibitory 855 GLP-1 circuit in the lateral septum modulates reward processing and alcohol 856 intake in rodents. EBioMedicine 115, 105684 (2025). 857 25. C. Aranas et al., Semaglutide reduces alcohol intake and relapse -like drinking in 858 male and female rats. EBioMedicine 93, 104642 (2023). 859 26. V . Chuong et al. , The glucagon -like peptide -1 (GLP -1) analogue semaglutide 860 reduces alcohol drinking and modulates central GABA neurotransmission. JCI 861 Insight 8 (2023). 862 27. D. Vallof, J. Vestlund, E. Jerlhag, Glucagon -like peptide -1 receptors within the 863 nucleus of the solitary tract regulate alcohol -mediated behaviors in rodents. 864 Neuropharmacology 149, 124-132 (2019). 865 28. A. Fink-Jensen et al., Effect of the glucagon-like peptide-1 (GLP-1) receptor agonist 866 semaglutide on alcohol consumption in alcohol-preferring male vervet monkeys. 867 Psychopharmacology (Berl) 10.1007/s00213-024-06637-2 (2024). 868 29. C. S. Hendershot et al. , Once -Weekly Semaglutide in Adults With Alcohol Use 869 Disorder: A Randomized Clinical Trial. JAMA Psychiatry 870 10.1001/jamapsychiatry.2024.4789 (2025). 871 30. M. K. Klausen et al., Exenatide once weekly for alcohol use disorder investigated 872 in a randomized, placebo-controlled clinical trial. JCI Insight 7 (2022). 873 31. W. Wang et al. , Associations of semaglutide with incidence and recurrence of 874 alcohol use disorder in real-world population. Nat Commun 15, 4548 (2024). 875 32. M. O'Farrell, F . I. Almohaileb, C. W. le Roux, Glucagon -like peptide-1 analogues 876 reduce alcohol intake. Diabetes Obes Metab 27, 1601-1604 (2025). 877 33. J. P . Frias et al., Tirzepatide versus Semaglutide Once Weekly in Patients with Type 878 2 Diabetes. N Engl J Med 385, 503-515 (2021). 879 34. A. M. Jastreboff et al. , Tirzepatide Once Weekly for the Treatment of Obesity. N 880 Engl J Med 387, 205-216 (2022). 881 35. D. Martinez-Urbistondo, N. Perez-Diaz-Del-Campo, M. F . Landecho, J. A. Martinez, 882 Alcohol Drinking Impacts on Adiposity and Steatotic Liver Disease: Concurrent 883 Effects on Metabolic Pathways and Cardiovascular Risks. Curr Obes Rep 13, 461-884 474 (2024). 885 36. F . Aberg, C. D. Byrne, C. J. Pirola, V . Mannisto, S. Sookoian, Alcohol consumption 886 and metabolic syndrome: Clinical and epidemiological impact on liver disease. J 887 Hepatol 78, 191-206 (2023). 888 37. H. J. Wang, S. Zakhari, M. K. Jung, Alcohol, inflammation, and gut -liver-brain 889 interactions in tissue damage and disease development. World J Gastroenterol 890 16, 1304-1313 (2010). 891 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 34 38. J. A. Simms et al. , Intermittent Access to 20% Ethanol Induces High Ethanol 892 Consumption in Long –Evans and Wistar Rats. Alcoholism: Clinical and 893 Experimental Research 32, 1816-1823 (2008). 894 39. T. E. Thiele, J. C. Crabbe, S. L. Boehm, 2nd, "Drinking in the Dark" (DID): a simple 895 mouse model of binge-like alcohol intake. Curr Protoc Neurosci 68, 9 49 41-49 49 896 12 (2014). 897 40. V . Vengeliene, A. Bilbao, R. Spanagel, The alcohol deprivation effect model for 898 studying relapse behavior: a comparison between rats and mice. Alcohol 48, 313-899 320 (2014). 900 41. R. W. Hingson, W. Zha, A. M. White, Drinking Beyond the Binge Threshold: 901 Predictors, Consequences, and Changes in the U.S. Am J Prev Med 52, 717-727 902 (2017). 903 42. N. Vafaie, H. Kober, Association of Drug Cues and Craving With Drug Use and 904 Relapse: A Systematic Review and Meta -analysis. JAMA Psychiatry 79, 641-650 905 (2022). 906 43. J. Witteman et al., Cue reactivity and its relation to craving and relapse in alcohol 907 dependence: a combined laboratory and field study. Psychopharmacology (Berl) 908 232, 3685-3696 (2015). 909 44. A. T. McLellan, D. C. Lewis, C. P . O'Brien, H. D. Kleber, Drug dependence, a chronic 910 medical illness: implications for treatment, insurance, and outcomes evaluation. 911 JAMA 284, 1689-1695 (2000). 912 45. R. Loomba et al. , Tirzepatide for Metabolic Dysfunction -Associated 913 Steatohepatitis with Liver Fibrosis. N Engl J Med 391, 299-310 (2024). 914 46. A. J. Sanyal et al. , Phase 3 Trial of Semaglutide in Metabolic Dysfunction -915 Associated Steatohepatitis. N Engl J Med 392, 2089-2099 (2025). 916 47. M. J. Gonzalez-Rellan, D. J. Drucker, The expanding benefits of GLP -1 medicines. 917 Cell Rep Med 6, 102214 (2025). 918 48. C. K. Wong et al., Central glucagon-like peptide 1 receptor activation inhibits Toll-919 like receptor agonist-induced inflammation. Cell Metab 36, 130-143 e135 (2024). 920 49. T. P . Sheehan, R. A. Chambers, D. S. Russell, Regulation of affect by the lateral 921 septum: implications for neuropsychiatry. Brain Res Brain Res Rev 46, 71 -117 922 (2004). 923 50. C. UniProt, UniProt: the Universal Protein Knowledgebase in 2025. Nucleic Acids 924 Res 53, D609-D617 (2025). 925 51. D. Tang et al., SRplot: A free online platform for data visualization and graphing. 926 PLoS One 18, e0294236 (2023). 927 52. A. E. Harasta et al. , Septal Glucagon -Like Peptide 1 Receptor Expression 928 Determines Suppression of Cocaine -Induced Behavior. 929 Neuropsychopharmacology 40, 1969-1978 (2015). 930 53. E. Egecioglu, J. A. Engel, E. Jerlhag, The glucagon-like peptide 1 analogue, exendin-931 4, attenuates the rewarding properties of psychostimulant drugs in mice. PLoS 932 One 8, e69010 (2013). 933 54. E. Egecioglu, J. A. Engel, E. Jerlhag, The glucagon-like peptide 1 analogue Exendin-934 4 attenuates the nicotine -induced locomotor stimulation, accumbal dopamine 935 release, conditioned place preference as well as the expression of locomotor 936 sensitization in mice. PLoS One 8, e77284 (2013). 937 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 35 55. N. S. Hernandez et al. , GLP -1 receptor signaling in the laterodorsal tegmental 938 nucleus attenuates cocaine seeking by activating GABAergic circuits that project 939 to the VTA. Mol Psychiatry 26, 4394-4408 (2021). 940 56. R. Merkel et al. , An endogenous GLP -1 circuit engages VTA GABA neurons to 941 regulate mesolimbic dopamine neurons and attenuate cocaine seeking. bioRxiv 942 10.1101/2024.06.20.599574 (2024). 943 57. R. J. Herman, M. R. Hayes, J. Audrain -McGovern, R. L. Ashare, H. D. Schmidt, 944 Liraglutide attenuates nicotine self-administration as well as nicotine seeking and 945 hyperphagia during withdrawal in male and female rats. Psychopharmacology 946 (Berl) 240, 1373-1386 (2023). 947 58. Y . Zhang et al. , Activation of GLP -1 receptors attenuates oxycodone taking and 948 seeking without compromising the antinociceptive effects of oxycodone in rats. 949 Neuropsychopharmacology 45, 451-461 (2020). 950 59. D. Vallof et al., The glucagon-like peptide 1 receptor agonist liraglutide attenuates 951 the reinforcing properties of alcohol in rodents. Addict Biol 21, 422-437 (2016). 952 60. M. Thomsen et al. , The glucagon -like peptide 1 receptor agonist Exendin -4 953 decreases relapse -like drinking in socially housed mice. Pharmacol Biochem 954 Behav 160, 14-20 (2017). 955 61. G. Sorensen, S. B. Caine, M. Thomsen, Effects of the GLP-1 Agonist Exendin-4 on 956 Intravenous Ethanol Self-Administration in Mice. Alcohol Clin Exp Res 40, 2247 -957 2252 (2016). 958 62. V . N. Marty et al. , Long -Acting Glucagon -Like Peptide -1 Receptor Agonists 959 Suppress Voluntary Alcohol Intake in Male Wistar Rats. Front Neurosci 14, 599646 960 (2020). 961 63. F . Quddos et al. , Semaglutide and Tirzepatide reduce alcohol consumption in 962 individuals with obesity. Sci Rep 13, 20998 (2023). 963 64. J. Blundell et al., Effects of once -weekly semaglutide on appetite, energy intake, 964 control of eating, food preference and body weight in subjects with obesity. 965 Diabetes Obes Metab 19, 1242-1251 (2017). 966 65. C. K. Martin et al., Tirzepatide on ingestive behavior in adults with overweight or 967 obesity: a randomized 6-week phase 1 trial. Nat Med 10.1038/s41591-025-03774-968 9 (2025). 969 66. A. Besnard, F . Leroy, Top -down regulation of motivated behaviors via lateral 970 septum sub-circuits. Mol Psychiatry 10.1038/s41380-022-01599-3 (2022). 971 67. H. S. Wirtshafter, M. A. Wilson, Lateral septum as a nexus for mood, motivation, 972 and movement. Neurosci Biobehav Rev 126, 544-559 (2021). 973 68. S. Buller, C. Blouet, Brain access of incretins and incretin receptor agonists to their 974 central targets relevant for appetite suppression and weight loss. Am J Physiol 975 Endocrinol Metab 326, E472-E480 (2024). 976 69. S. Jonsson, J. Morud, R. Stomberg, M. Ericson, B. Soderpalm, Involvement of 977 lateral septum in alcohol's dopamine-elevating effect in the rat. Addict Biol 22, 93-978 102 (2017). 979 70. S. C. Cork et al. , Distribution and characterisation of Glucagon -like peptide -1 980 receptor expressing cells in the mouse brain. Mol Metab 4, 718-731 (2015). 981 71. I. Merchenthaler, M. Lane, P . Shughrue, Distribution of pre -pro-glucagon and 982 glucagon-like peptide -1 receptor messenger RNAs in the rat central nervous 983 system. J Comp Neurol 403, 261-280 (1999). 984 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 36 72. J. Nyberg, C. Jacobsson, M. F . Anderson, P . S. Eriksson, Immunohistochemical 985 distribution of glucose -dependent insulinotropic polypeptide in the adult rat 986 brain. J Neurosci Res 85, 2099-2119 (2007). 987 73. A. M. Kaplan, S. R. Vigna, Gastric inhibitory polypeptide (GIP) binding sites in rat 988 brain. Peptides 15, 297-302 (1994). 989 74. S. Gabery et al., Semaglutide lowers body weight in rodents via distributed neural 990 pathways. JCI Insight 5 (2020). 991 75. G. Skovbjerg et al., Uncovering CNS access of lipidated exendin -4 analogues by 992 quantitative whole -brain 3D light sheet imaging. Neuropharmacology 238, 993 109637 (2023). 994 76. I. A. Reddy et al., Glucagon-like peptide 1 receptor activation regulates cocaine 995 actions and dopamine homeostasis in the lateral septum by decreasing 996 arachidonic acid levels. Transl Psychiatry 6, e809 (2016). 997 77. T. M. Cruise, K. Kotlo, E. Malovic, S. C. Pandey, Advances in DNA, histone, and RNA 998 methylation mechanisms in the pathophysiology of alcohol use disorder. Adv 999 Drug Alcohol Res 3, 10871 (2023). 1000 78. S. C. Pandey, E. J. Kyzar, H. Zhang, Epigenetic basis of the dark side of alcohol 1001 addiction. Neuropharmacology 122, 74-84 (2017). 1002 79. L. Maretty et al. , Proteomic changes upon treatment with semaglutide in 1003 individuals with obesity. Nat Med 31, 267-277 (2025). 1004 80. H. C. Becker, M. F . Lopez, Animal Models of Excessive Alcohol Consumption in 1005 Rodents. Curr Top Behav Neurosci 10.1007/7854_2024_461 (2024). 1006 81. A. Loften et al. , Alcohol -induced accumbal dopamine - and taurine release in 1007 female and male Wistar rats, an in vivo microdialysis study. J Neural Transm 1008 (Vienna) 10.1007/s00702-025-02928-w (2025). 1009 82. D. J. Ottenheimer, M. M. Hjort, A. J. Bowen, N. A. Steinmetz, G. D. Stuber, A stable, 1010 distributed code for cue value in mouse cortex during reward learning. Elife 12 1011 (2023). 1012 83. L. R. Saraiva et al., Combinatorial effects of odorants on mouse behavior. Proc 1013 Natl Acad Sci U S A 113, E3300-3306 (2016). 1014 84. G. Paxinos, Paxinos and Franklin's the Mouse Brain in Stereotaxic Coordinates 1015 (Academic Press, ed. Fifth Edition, 2019). 1016 85. M. Vujicic et al., A macrophage -collagen fragment axis mediates subcutaneous 1017 adipose tissue remodeling in mice. Proc Natl Acad Sci U S A 121, e2313185121 1018 (2024). 1019 86. S. F . Mehdi et al. , Glucagon -like peptide -1: a multi -faceted anti -inflammatory 1020 agent. Front Immunol 14, 1148209 (2023). 1021 87. J. J. H. Bray et al., Glucagon-like peptide-1 receptor agonists improve biomarkers 1022 of inflammation and oxidative stress: A systematic review and meta -analysis of 1023 randomised controlled trials. Diabetes Obes Metab 23, 1806-1822 (2021). 1024 88. L. Adermark et al. , Temporal Rewiring of Striatal Circuits Initiated by Nicotine. 1025 Neuropsychopharmacology 41, 3051-3059 (2016). 1026 1027 1028 1029 1030 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 37 FIGURES 1031 Fig. 1 1032 1033 1034 1035 1036 1037 1038 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 38 Fig. 1. Impact of tirzepatide on alcohol -related reward behaviors and dopamine release in 1039 male mice. 1040 A. Tirzepatide (Tirz; 0.144 mg/kg) attenuates alcohol (Alc; 1.75 g/kg, IP) -induced locomotor 1041 stimulation (n=9/group, one -way ANOVA). B. Tirzepatide (0.144 mg/kg) treatment reduces the 1042 expression of alcohol (1.75 g/kg, IP) -induced conditioned place preference (CPP), without 1043 affecting CPP itself (n=8 -10/group, unpaired t -test). C. On day 20 (D20) following a period of 1044 forced abstinence from day 6 (D6), tirzepatide (0.144 mg/kg) attenuates the expression of alcohol 1045 (1.75 g/kg, IP) and cue -induced CPP that persists in vehicle -treated mice (Veh), with the cue 1046 present on both testing days (n=10/group, unpaired t -test). D. Tirzepatide (0.144 mg/kg) 1047 significantly mitigates alcohol (1.75 g/kg, IP) -induced dopamine release in the nucleus 1048 accumbens (NAc) following systemic alcohol injection (n=8/group, repeated measures two-way 1049 ANOVA). E. Tirzepatide similarly blocks dopamine release when we perfused alcohol (300 mM) 1050 locally in the NAc (n=8/group, repeated measures two -way ANOVA). Data show mean ± SEM. 1051 *P<0.05, **P<0.01, ***P<0.001, #P<0.05, ##P<0.01, ###P<0.001. 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 39 Fig. 2 1064 1065 1066 1067 1068 1069 1070 1071 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 40 Fig. 2. Effects of single administration of tirzepatide on alcohol intake, binge -like drinking 1072 and relapse-like drinking in male and female rodents. 1073 A. Single tirzepatide administration (0.144 mg/kg) significantly reduces 24-hour alcohol intake in 1074 male rats compared to vehicle (n=12, paired t-test). B. Female rats show similar reduction in 24-1075 hour alcohol consumption following single tirzepatide treatment (0.144 mg/kg; n=12, paired t -1076 test). C. Percent reduction in alcohol intake relative to vehicle demonstrates comparable 1077 tirzepatide efficacy (0.144 mg/kg) across both sexes with no statistically significant difference 1078 between males and females (n=12/group, unpaired t -test). D. Single tirzepatide administration 1079 (0.144 mg/kg) significantly attenuates binge -like drinking in male mice (n=10/group, unpaired t -1080 test). E. Female mice also exhibit significant reduction in binge -like alcohol consumption 1081 following single tirzepatide treatment (0.144 mg/kg; n=10/group, unpaired t -test). F. Percent 1082 comparison against vehicle demonstrates comparable tirzepatide effectiveness (0.144 mg/kg) 1083 on binge-like drinking between sexes with no statistically significant difference between males 1084 and females (n=10/group, unpaired t -test). G. Vehicle-treated male rats exhibit significantly 1085 elevated alcohol intake during post -deprivation sessions compared to baseline consumption 1086 levels, whereas tirzepatide administration (0.144 mg/kg) effectively blocks this relapse -like 1087 drinking behavior (n=10/group, one -way ANOVA). H. Percent change from baseline in males 1088 demonstrates significant attenuation of relapse -like drinking behavior by tirzepatide (0.144 1089 mg/kg; n=10/group, unpaired t -test). I. Vehicle-treated females demonstrate significantly 1090 elevated alcohol intake in post-deprivation sessions compared to baseline, whereas single-dose 1091 tirzepatide administration (0.144 mg/kg) inhibits this relapse -like alcohol drinking effect and 1092 further reduces alcohol intake compared to baseline levels (n=10/group, one -way ANOVA). J. 1093 Percent change from baseline in females shows tirzepatide efficacy (0.144 mg/kg) in preventing 1094 relapse-like drinking (n=10/group, unpaired t-test). Data show mean ± SEM. *P<0.05, **P<0.01, 1095 ***P<0.001. 1096 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 41 Fig. 3 1097 1098 1099 1100 1101 1102 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 42 Fig. 3. Effects of repeated tirzepatide administration on alcohol consumption, body 1103 composition and inflammation parameters in male and female rate. 1104 All experiments used n=10/group with repeated tirzepatide (0.144 mg/kg) or vehicle treatment. A. 1105 Tirzepatide significantly attenuates alcohol intake in male rats on all alcohol days compared to 1106 vehicle group, which shows similar baseline (BL) alcohol intake levels (repeated measures two -1107 way ANOVA). B. Female rats show comparable reduced alcohol intake following tirzepatide 1108 administration, with consistent efficacy throughout all alcohol days (repeated measures two-way 1109 ANOVA). C. Alcohol intake relative to vehicle demonstrates comparable reducing effects across 1110 sexes with no statistically significant sex differences (repeated measures two-way ANOVA). D-E. 1111 Tirzepatide reduces body weight and percent body weight change in male rats compared to 1112 vehicle (repeated measures two -way ANOVA). F. Post-mortem tissue analysis in males shows 1113 significant reductions in subcutaneous inguinal (sWAT), gonadal (gWAT), and retroperitoneal 1114 (rpWAT) white adipose tissues and liver weight, while muscle (gastrocnemius) and intrascapular 1115 brown adipose tissue (iBAT) remain unaffected (unpaired t -tests). G. Tirzepatide significantly 1116 reduces hepatic triglyceride content in alcohol -drinking males (unpaired t -test). H. Treatment 1117 decreases interleukin (IL)-6 and tumor necrosis factor alpha (TNFα) serum levels, while IL-1 beta 1118 (β), IL-10 and monocyte chemoattractant protein-1 (MCP-1) remain unaffected (unpaired t-test). 1119 I-J. Female rats exhibit decreased body weight and percent body weight change following 1120 tirzepatide treatment (repeated measures two-way ANOVA). K. Tirzepatide significantly reduces 1121 white adipose tissue depots in females (unpaired t -tests). L. Tirzepatide significantly reduces 1122 hepatic triglyceride content in treated females (unpaired t -test). M. Treatment reduces IL-6 and 1123 TNFα serum levels in females, while levels of IL-1β, IL-10 and MCP-1 remain unaffected (unpaired 1124 t-test). Data show mean ± SEM with individual data points. *P<0.05, **P<0.01, ***P<0.001. 1125 1126 1127 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 43 Fig. 4 1128 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint 44 Fig. 4. Electrophysiological effects of tirzepatide and proteomic alterations following 1129 tirzepatide administration in alcohol-exposed male rodents. 1130 A. Input-output curves derived from ex vivo brain slice field recordings from male mice 1131 demonstrate tirzepatide-induced reduction in population spike (PS) amplitude within the lateral 1132 septum (LS) (n=28 -30/group, repeated measures two -way ANOVA). B. Paired-pulse ratio (PPR, 1133 PS2/PS1) measurements in the LS reveal significant effects in tirzepatide-treated mice compared 1134 to the vehicle group (n=20 -22/group, unpaired t -test). C-G. Input-output curves in the medial 1135 prefrontal cortex (mPFC), dorsomedial striatum (DMS), dorsolateral striatum (DLS), nucleus 1136 accumbens (NAc) core, and NAc shell demonstrate region -specific electrophysiological 1137 responses, with no statistically significant differences observed in these regions following 1138 tirzepatide treatment (n=9 -30/group, repeated measures two -way ANOVA). H. Volcano plot 1139 visualizes the proteomics data in the LS of alcohol -consuming male rats following repeated 1140 tirzepatide (0.144 mg/kg) administration, showing the differential protein expression profile. I. 1141 Circular heatmap visualizes differentially expressed proteins associated with histone and 1142 chromatin processes following repeated tirzepatide treatment in alcohol -drinking male rats. J. 1143 Quantification shows fold changes in specific proteins associated with histone and chromatin 1144 processes comparing tirzepatide treatment against vehicle, with significant differences observed 1145 in multiple proteins (n=9/group, Welch's t-test, *p<0.05, **p<0.01). 1146 1147 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.26.672374doi: bioRxiv preprint