Central Thyroid Hormone Receptor-Beta: Sensitivity to Alcohol and A Role in Regulating Alcohol Drinking

Abstract

Clinical and preclinical evidence indicate that both peripheral and central elements of the hypothalamic-pituitary-thyroid (HPT) axis are dysregulated in alcohol use disorder, and that thyroid hormone system dysregulation is associated alcohol craving and co-morbid mood and depression-related disorders. Yet, no study has investigated if central nervous system (CNS) thyroid hormone receptors, primary targets of thyroid hormone and major regulators of the HPT axis are involved in alcohol consumption. We utilized a 24-hour access two-bottle choice (2BC) voluntary ethanol (EtOH) drinking paradigm to assess if the expression of CNS thyroid hormone receptors is sensitive to voluntary alcohol consumption in C57BL/6J mice. We found that thyroid hormone receptor-beta (Thrb/THRβ) mRNA expression was significantly reduced in the paraventricular nucleus of the hypothalamus of EtOH drinking mice compared to water controls. In addition, EtOH drinking mice exhibited peripheral elevation of serum triiodothyronine. Next, we utilized the CNS selective THRβ agonist, Sob-AM2, to determine if central activation of THRβ would influence voluntary alcohol drinking in mice in the same EtOH 2BC drinking paradigm. We found that repeated treatment with Sob-AM2 significantly reduced daily EtOH intake and preference, while in conjunction increasing water intake. In summary, we found that hypothalamic Thrb expression is sensitive to voluntary alcohol drinking, and that CNS THRβ activity regulates alcohol consumption in mice. Taken together, our results highlight an important role of central thyroid hormone receptor signalling in alcohol drinking and indicate therapeutic potential of CNS specific thyromimetics in treatment of alcohol use disorder.

Introduction:

Alcohol use disorder (AUD) is a widespread public health problem with few reliable pharmacotherapeutics despite an economic burden of ~$250 billion per year and alcohol use being involved in ~3–6% of annual deaths worldwide (Bardach et al. 2019; Di Castelnuovo et al. 2019; GBD 2016 Alcohol Collaborators 2018; Sacks et al. 2015; CDC 2023). Alcohol consumption is deleterious to many organ systems including neuroendocrine axes which maintain homeostasis through central feedback onto the hypothalamus and influence behavior through hormonal actions in the brain (Rachdaoui and Sarkar 2017; Emanuele and Emanuele 1997; Clarke et al. 2008; Kenna et al. 2012). Dysregulation of these systems is thought to contribute to the maintenance and continuation of maladaptive behaviors related to AUD (Kiefer and Wiedemann 2004). Thus, neuroendocrine systems provide intriguing targets for pharmacotherapeutics for AUD (Rachdaoui and Sarkar 2017; Leggio 2009). Interactions of alcohol and the hypothalamic-pituitary-adrenal axis, hypothalamic-pituitary-gonadal axis, and hypothalamic-posterior-pituitary axis have received much attention preclinically. However, there is also evidence that alcohol use is associated with dysregulation of the hypothalamic-pituitary-thyroid (HPT) axis (Reviewed in: Hermann et al. 2002; Balhara and Deb 2013).

The HPT axis regulates metabolism through the release of triiodothyronine (T3) and thyroxine (T4) from the thyroid gland in the periphery. Thyrotropin releasing hormone (TRH) is secreted from the paraventricular nucleus (PVN) of the hypothalamus and acts at the pituitary gland to drive the release of thyroid stimulating hormone (TSH). TSH then signals to the thyroid gland to produce and release T3 and T4 into circulation. Once in circulation T3 and T4 regulate target gene expression by binding to one of their nuclear hormone receptors, thyroid hormone receptors (THRα/β), in cells throughout the body (Lazar 1993; Wu and Koenig 2000). Importantly, in recent decades understanding of thyroid hormone’s role has expanded beyond regulation of metabolism. Not only do THRs serve as key transcriptional regulators in the central nervous system of adults (Schroeder and Privalsky 2014; Hochbaum et al. 2024; Bernal 2007), they are also necessary for development and function of GABAergic (Richard et al. 2020) and monoaminergic systems (Bauer et al. 2002; Ookubo et al. 2015; Bennett et al. 2024), suggesting that thyroid hormone and its receptors are critical for reward-related behaviors. Indeed, thyroid hormone dysregulation is associated with mood- and reward-related disorders such as bipolar disorder and major depression, which are often co-morbid with AUD (Liappas et al. 2006; McHugh and Weiss 2019; Li et al. 2004; Farren et al. 2012; Chakrabarti 2011).

With regard to AUD and ethanol (EtOH) exposure, clinical and preclinical evidence suggests that EtOH can dysregulate the thyroid hormone system, as indicated by a blunting of TSH release (Pienaar et al. 1995; Zoeller et al. 1996; Välimäki et al. 1984; Anderson et al. 1992; Sellman and Joyce 1992), and decreased T3 and T4 levels (Mason et al. 1988; Singh et al. 1979; Rasmussen 2003; Gruppen et al. 2020; Reviewed in: Hermann et al. 2002; Balhara and Deb 2013) in humans and in models of forced EtOH exposure in rats. Further clinical evidence suggests that these findings have behavioral relevance as studies report that serum T3 is positively associated with self-reported measures of craving (Aoun et al. 2015; Leggio et al. 2008), and that HPT dysregulation may contribute to withdrawal symptoms and relapse (Liappas et al. 2006; Heinz et al. 1996; Pienaar et al. 1995). Meanwhile, inducing hypothyroidism in rats results in increased binge-like EtOH intake, confirming experimentally the behavioral relevance of thyroid hormone signalling for EtOH consumption (Echeverry-Alzate et al. 2019). Such findings provide ample evidence that EtOH exposure and withdrawal are associated with dysregulation of the peripheral HPT axis, and that these changes are relevant for behavior. However, the molecular mechanisms underlying HPT dysregulation within the CNS are difficult to assess in clinical settings, necessitating further studies in preclinical models.

Several preclinical studies indicate EtOH-related dysregulation in thyroid hormone signalling in the CNS. In rats, EtOH injections increase thyrotropin releasing hormone mRNA in the PVN of the hypothalamus and reduce HPT axis response to central stimulation (Zoeller et al. 1996; Zoeller and Rudeen 1992). Additionally, activities of type 2 and type 3 deiodinases (Dio2 and Dio3), responsible for intracellular processing of thyroid hormones, are dysregulated in the prefrontal cortex (PFC) and amygdala of rats following EtOH vapor exposure (Baumgartner et al. 1997). In mice, expression of transthyretin (Ttr) mRNA, encoding for a protein important to central thyroid hormone transport, is upregulated in the PFC during withdrawal from EtOH (Hashimoto et al. 2011) and following chronic EtOH exposure in multiple brain regions (Ferguson et al. 2022). Such findings indicate a dysregulation of the thyroid hormone system by forced EtOH exposure that extends to central elements of the HPT axis. Yet, no study has investigated if central thyroid hormone receptors, primary targets of thyroid hormone, major drivers of transcriptional remodeling in the brain, and critical CNS regulators of the HPT axis, are affected by or play a role in voluntary EtOH drinking.

Parsing the role of central thyroid hormone signalling in EtOH consumption comes with difficulties, particularly because treatment with T3 or T4 results in activation of both central and peripheral THRs. In the last several years, efforts to more selectively target central THR activity have led to the development of CNS selective pro-drug agonists, such as Sob-AM2. Sob-AM2 is a pro-drug agonist for THRβ which, when administered systemically, remains masked from THRs in the periphery, but is converted to its active form when transported into the CNS where it selectively activates central THRs (Meinig et al. 2017). The ability to specifically target THRs in the CNS has already seen successful application in various preclinical models of neurological disorders (Bohlen et al. 2023; Chaudhary et al. 2021; Hartley et al., 2020), and CNS selective thyromimetics are currently in clinical trials for mood disorders (NCT06633016, NCT06869187). Thus, CNS selective thyromimetics afford a unique opportunity to investigate the role of central thyroid hormone signalling in EtOH consumption and have direct translational potential for AUD.

While previous studies have outlined that peripheral and central thyroid hormone signaling can be dysregulated by forced EtOH exposure, it remains unclear if the HPT axis in rodents is dysregulated by voluntary EtOH drinking. Likewise, no study has investigated if central thyroid hormone receptors play a role in EtOH drinking, despite their key role in the regulation of the HPT axis and behavior. To address these gaps in knowledge we utilized a C57BL/6J mouse model of 24-hour access two-bottle choice (2BC) voluntary EtOH drinking to 1) assess the effects of voluntary alcohol consumption on hypothalamic thyroid hormone related gene expression and serum thyroid hormone levels, and 2) investigate if activation of central THRβ, by Sob-AM2, influences voluntary EtOH consumption.

Methods:

Subjects:

All experiments were approved by the Institutional Animal Care and Use Committee at OHSU, Portland, OR, United States and were conducted in accordance with the National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals. Adult C57BL/6 J mice (N = 72; 36 males and 36 females; sex defined gonadally) arrived from the Jackson Laboratory (Sacramento, CA) at 8 weeks of age. Upon arrival, mice were housed in sex-matched groups of four in standard shoebox cages (18.4 × 29.2 × 12.7 cm) with food and water (H₂O) available ad libitum during all experiments. All cages contained a cotton nestlet and paper nesting materials. After 48h, male and female mice were isolated and transferred to experimental housing 2 BC drinking experiments. Experimental rooms were kept on a 12:12 light/dark cycle (lights off at 11:00).

Two-bottle Choice:

Following 48h of habituation to the new vivarium environment mice were transferred to isolated standard shoebox cages. All mice underwent H₂O habituation for 7 days, where they were given 24-hour access to two bottles containing autoclaved H₂O in their home cage environment (2BC exposure started at 11:00). Immediately following H₂O habituation, individually housed mice either underwent a 24-hour access 2BC EtOH procedure or had access to two bottles of autoclaved H₂O for approximately three weeks. For EtOH 2BC mice, a bottle of EtOH (escalating dosage of 3, 6, and 10%, volume/volume in autoclaved H₂O; Fig. 1; Experiment 1) and a bottle of autoclaved H₂O were continuously available for drinking. For the first two days of the experiment 3% (volume/volume) EtOH was available, followed by 6% EtOH for two days, and finally 10% EtOH for the remainder of the experiment (19 days; Fig. 1; Experiment 1). This is a standard method in alcohol research that results in pharmacologically relevant blood EtOH concentrations (Fulenwider et al. 2021; Dole and Gentry 1984) and signs of hyperalgesia upon EtOH withdrawal, which is a sign of developing EtOH dependence (Smith et al. 2016; 2017). H₂O control mice had access to two bottles of autoclaved H₂O for the same time-period. Fluid consumption was measured daily. Mice were weighed weekly, and their EtOH and H₂O intake were normalized to body weight. Immediately following all 2BC experiments, mice were euthanized via live decapitation 3-hours into the dark cycle for analysis of gene expression in the PVN and hormonal levels in trunk blood.

Figure 1.

Experimental Timelines. (A) Timeline for EtOH 2BC Drinking. (B) Timeline for Sob-AM2 effects on EtOH 2BC drinking.

Sob-AM2

Amide pro-drug THRβ agonist, Sob-AM2, was synthesized at Oregon Health & Science University as described previously (Meinig et al. 2017). Aliquots of Sob-AM2 dissolved in DMSO at a concentration of 10mg/mL and were frozen at −20°C. Each day prior to treatment, frozen aliquots were thawed and diluted with 0.9% sodium chloride to working concentrations. For the low-dose, aliquots were diluted to a concentration of 0.06 mg/mL and administered at a dose of 0.3 mg/kg bodyweight. For the high-dose, aliquots were diluted to a concentration 0.1mg/mL and administered at a dose of 1.0 mg/kg bodyweight. Vehicle treatments were 10% DMSO in 0.9% sodium chloride. All injections were performed subcutaneously (sc).

Effects of Sob-AM2 on EtOH 2BC drinking:

For Experiment 2A, mice underwent ~ 3-week EtOH 2BC and had the 10% portion of the experiment extended for Sob-AM2 or vehicle treatment (Fig. 1; Experiment 2A). Once EtOH drinking stabilized (Supplementary Fig. 1A), mice were given a habituation injection of vehicle (10% DMSO in 0.9% sodium chloride) on the first day of this extension and subsequently received a daily subcutaneous injection of either a low-dose of Sob-AM2 (0.3 mg/kg) or vehicle for 6 days. Finally, for the remaining 5 days of treatment mice received daily subcutaneous injections of a high-dose (1 mg/kg) of Sob-AM2. Injections were performed approximately 1 hour prior to the onset of the dark cycle. After the final treatment mice were euthanized 3-hours into the dark cycle as in Experiment 1. For Experiment 2B, mice underwent the exact same 2BC and injection schedule as in Experiment 2A, however, only water was available in both bottles for the entirety of the experiment (Supplementary Figs. 3-5).

Tissue Processing and qPCR:

Immediately following euthanasia, brains were sectioned on ice using 1 mm metal brain block and 14-gauge punches of PVN were dissected on ice and stored at −80ºC. RNA was isolated from frozen tissue punches using a modified RNA extraction protocol (Walker et al. 2022; Invitrogen, #12183020). RNA concentrations were determined on a Nanodrop spectrophotometer and RNA quality was assessed using a Bioanalyzer 2100 and only samples with RINs > 8 were included. 300ng of RNA were converted to cDNA using a high capacity cDNA Reverse Transcription Kit (Invitrogen, #4368814). qPCR was performed using generated cDNA, gene target primers (Supplementary Table 1), and PowerUp Sybr Green Mastermix (Applied Biosystems, #A25742) on the ABI quant studio 5 under the following parameters: 50°C for 2 min; 95°C for 10 min and 45 cycles (95°C for 15 sec; 60°C for 15 sec). All samples were run in triplicates. The median of the three Cts for each gene/sample were analyzed via the comparative Ct method (Livak and Schmittgen 2001).

Serum Thyroid Hormones:

Immediately following euthanasia fresh trunk blood was collected from mice in all experiments. Trunk blood was allowed to coagulate and was spun at 11,292G for 15 minutes at a temperature of 4 ºC, serum supernatant was then collected and stored at −80ºC. Serum thyroid hormones assays for T3, T4 and TSH were performed using the Rat Thyroid Panel RTHYMAG-30K (Millipore Sigma, #12161503; inter-assay cv: <15%, intra-assay cv: <10%) according to manufacturer’s guidelines, by the Endocrinology Core facility at OHSU. The limit of detection for this kit is between 96–24,000 pg/mL and any sample that was outside the limit of detection was replaced by the value of the respective upper or lower limit (i.e.: sample < 96 pg/mL = 96 pg/mL; sample > 24,000 pg/mL = 24,000 pg/mL). Due to the limited amount of serum collected from mice, one sample reading was run per animal.

Statistical Analysis:

GraphPad Prism was used for data analysis and graph generation. All data are presented as means ± standard error of the mean. Grubbs test for outliers was performed on the raw data of each output measure prior to analysis. No more than one data point per group per analysis were removed. Partial eta squared (η²_p = SS_Effect / SS_Effect + SS_Error) was calculated to determine the effect size of significant ANOVA effects. Meanwhile Cohen’s d (d = x̄₁ – x̄ ₂ / s_pooled) was used to calculate the effect size of significant t-tests and Tukey’s post-hoc tests.

Experiment 1:

Two-way repeated measures ANOVAs (EtOH by sex by day) were performed for EtOH intake and EtOH preference on the last 5 days of 2BC EtOH drinking. For gene expression analysis, two-way ANOVAs (EtOH by sex) were performed on relative gene target ΔCt values normalized to the geometric mean of Cts from housekeeping genes (18s, Hprt1). There were no significant differences in house-keeping gene Ct values across any groups (See Supplementary Table 1). Final bodyweights and serum thyroid hormone levels were each analyzed using two-way ANOVAs (EtOH by sex). Tukey’s post-hoc analyses were performed if interactions were observed between factors.

Experiment 2:

Three-way repeated measures ANOVAs (treatment by sex by day) were performed for EtOH intake and EtOH preference on the last 5 days of 2BC EtOH drinking prior to treatment with Sob-AM2 or vehicle (Supplementary Fig. 1). Mice were counterbalanced prior to treatment such that EtOH intake did not significantly differ between treatment groups during the last 5 days prior to treatment (Supplementary Fig. 1A). The average of these last 5 days of EtOH intake and preference were taken as baseline values for analysis. For EtOH intake, EtOH preference, H₂O intake, and total fluid intake, three-way repeated measures ANOVAs (treatment by sex by time) were performed on baseline values, habituation injection day, days of low-dose treatment (6 days; 0.3 mg/kg), and days of high-dose treatment (4 days; 1.0 mg/kg). If interactions of treatment by day were observed, comparisons between baseline, the last day of low-dose, and the last day of high-dose were performed with three-way repeated measures ANOVAs to identify the cumulative effects of multiple days of dosing with Sob-AM2 at either dose. Since this comparison resulted in a significant interaction of treatment by time by sex in EtOH intake (Fig. 4B; F _{(2, 40)} = 4.686, p < 0.05), all related comparisons (EtOH preference, total fluid intake, and H₂O intake) across these three timepoints were performed as two-way ANOVAs (treatment by time) within each sex for internal consistency of comparisons. Significant interactions within these ANOVAs were followed up with Tukey’s post-hoc comparisons. Percentage change from baseline to the last day of high-dose values was calculated for all output measures and two-way ANOVAs (treatment by sex) were performed. Sob-AM2 effects on bodyweight were analyzed using a three-way repeated measures ANOVA (treatment by sex by time) and serum thyroid hormone levels were analyzed using two-way ANOVAs (EtOH by sex). For gene expression analysis in Sob-AM2 experiments, 12 samples were either lost during RNA isolation or displayed low RNA integrity values (RIN < 8) and were not included in qPCR experiments. Lost samples were evenly distributed across groups (EtOH VEH: 1M, 2F; EtOH Sob-AM2: 2M, 1F; H₂O VEH: 1M, 2F; H₂O Sob-AM2: 2M, 1F). Two-way ANOVAs (Sob-AM2 by sex) were performed on relative gene target ΔCt values normalized to the geometric mean of Cts from housekeeping genes (18s, Hprt1). As in Experiment 1, there were no significant differences in house-keeping gene Ct values across any groups (See Supplementary Table 1). Statistics were performed separately, but in the same manner for EtOH-naïve mice in Experiment 2b.

Figure 4.

Effects of Sob-AM2 on EtOH 2BC (A) Daily EtOH intake (g/kg) during Sob-AM2 Treatment. (B) EtOH intake at baseline, final day of low dose treatment (0.3 mg/kg), and final day of high-dose treatment (1 mg/kg). * (p < 0.01), **(p < 0.01), and *** (p< 0.001) indicate significant post-hoc Tukey’s tests. (C) Percentage change in EtOH intake (g/kg) from baseline to the final day of treatment. * indicates significant effect of Sob-AM2 treatment (p < 0.05). (D) Daily EtOH preference during Sob-Am2 Treatment. (E) EtOH preference at baseline, end of low dose treatment (0.3 mg/kg), and end of high-dose treatment (1 mg/kg). * (p < 0.01), **(p < 0.01), and *** (p< 0.001) indicate significant post-hoc Tukey’s tests. (F) Percentage change in EtOH preference from baseline to the final day of treatment. ** indicates significant effect of Sob-AM2 treatment (p < 0.01).

Results:

Experiment 1:

Experiment 1. EtOH 2BC Drinking

A two-way repeated measures ANOVA revealed that female mice consumed significantly more EtOH than males during the last 5 days of 10% drinking (F _{(1, 10)} = 13.63, p < 0.01, η²_p = 0.73; Fig. 2A), however, preference did not significantly differ (Fig. 2B). Furthermore, when assessing average daily EtOH intake over the course of the entire experiment female mice consumed significantly more EtOH than male mice (t ₍₁₀₎ = 3.409, p < 0.01, d = 1.97; Fig. 2C). Next, to assess if EtOH consumption had a meaningful impact on metabolism due to caloric intake we compared body weights on the final day of 2BC experiments using a two-way ANOVA with factors of sex by EtOH. While there was a significant and expected difference between sexes (F _{(1, 20)} = 115.6, p < 0.0001, η²_p = 0.85; Fig. 2D), we observed no significant differences in bodyweight between animals in the H₂O and EtOH conditions (F _{(1, 20)} = 0.043, p = 0.84; Fig. 2D).

Figure 2.

Serum thyroid hormones in EtOH 2BC mice and H₂O drinking controls (A) Last 5 days of EtOH intake. (B) Last 5 days of EtOH preference. (C) Average daily EtOH intake over the course of the study. ## Indicates significant effect of sex (p < 0.01) (D) Final bodyweights of EtOH 2BC. #### Indicates significant effect of sex (p < 0.0001). (E) Serum TSH levels in EtOH 2BC mice and H₂O drinking controls. (F) Serum T3 levels in EtOH 2BC mice and H₂O drinking controls. * Indicates significant effect of EtOH (p < 0.05). (G) Serum T4 levels in EtOH 2BC mice and H₂O drinking controls. # Indicates significant effect of sex (p < 0.05). (H) Serum T3/T4 ratio in EtOH 2BC mice and H₂O drinking controls. # Indicates significant effect of sex (p < 0.05). * Indicates significant effect of EtOH (p < 0.05).

Effects of EtOH Drinking on Serum Thyroid Hormones

To assess the effects of voluntary EtOH consumption on peripheral thyroid hormone levels we compared T3, T4, T3/T4 ratios, and TSH levels in H₂O drinking and 2BC EtOH drinking mice using two-way ANOVAs of sex by EtOH. We observed no significant differences in TSH levels in any conditions (Fig. 2E). However, voluntary EtOH drinking produced a significant increase in peripheral T3 levels regardless of sex (F _{(1, 20)} = 4.359, p < 0.05, η²_p = 0.18, Fig. 2F), while T4 levels were higher in female than in male mice regardless of condition (F _{(1, 20)} = 15.74, p < 0.001, η²_p = 0.44, Fig. 2G). Furthermore, T3/T4 ratios were significantly higher in EtOH drinking mice than H₂O drinking mice (F _{(1, 20)} = 5.579, p < 0.05, η²_p = 0.22) and males exhibited higher T3/T4 ratios than females (F _{(1, 20)} = 15.26, p < 0.001, η²_p = 0.43).

Effects of EtOH Drinking on Gene Expression in the PVN

To test if voluntary EtOH consumption resulted in changes to central thyroid hormone signaling we used RT-qPCR to compare mRNA expression of thyroid hormone related genes (See Supplementary Table 1) in the PVN, the primary central hub of HPT axis (Kondo et al. 2022), in mice from the above experiment. A two-way ANOVA of EtOH by sex revealed that EtOH drinking mice significantly decreased Thrb mRNA expression regardless of sex compared to H₂O drinking controls (F _{(1, 20)} = 8.157, p < 0.01, η²_p = 0.29; Fig. 3C). Furthermore, we observed that androgen receptor (Ar) mRNA expression was also decreased in EtOH drinking mice compared to H₂O drinking controls (F _{(1, 20)} = 9.061, p < 0.01, η²_p = 0.31), an effect that has been observed previously in protein levels of the hypothalamus of rats following chronic EtOH exposure (Chung 1989). We found no other significant differences in thyroid hormone related gene expression following the 2BC drinking paradigm. (See Supplementary Table 1).

Figure 3.

mRNA expression in the PVN of EtOH 2BC mice and H₂O drinking controls normalized to the geometric mean of Hprt1 and 18s. (A) PVN thyrotropin releasing hormone expression. (B) PVN thyroid hormone receptor-alpha expression. (C) PVN thyroid hormone receptor-beta expression. ** Indicates significant effect of EtOH (p < 0.01).

Experiment 2:

Experiment 2A. Effects of Sob-AM2 on EtOH Drinking and Preference in EtOH Drinking Mice

In Experiment 1 we determined that 2BC EtOH drinking significantly decreases Thrb mRNA expression in the PVN of male and female mice compared to H₂O drinking controls. Therefore, we then tested whether increasing central THR activity would modulate alcohol drinking. To investigate this, we treated male and female mice with increasing doses of the CNS specific THRβ agonist Sob-AM2 and assessed EtOH intake and preference. Mice underwent EtOH 2BC procedure as described above (Fig. 1; Experiment 2A). Once EtOH drinking stabilized (Supplementary Fig. 1A), mice were given one subcutaneous habituation injection of vehicle (10% DMSO in Saline), then treated with daily subcutaneous injections of either Sob-AM2 at 0.3 mg/kg for 6 days, and then 4 days of 1.0 mg/kg, or vehicle for the same period. Using a three-way repeated measures ANOVA with factors of treatment, sex, and time, we evaluated EtOH intake across baseline (baseline = the average intake across the last 5 days of EtOH 2BC, Supplementary Fig. 1A), the habituation injection day, and 10 days of treatment with Sob-AM2 (6 days at 0.3 mg/kg & 4 at 1.0 mg/kg; Fig. 4A). ANOVAs revealed a significant interaction of treatment by day for both EtOH intake (F _{(11, 220)} = 2.378, p < 0.01, η²_p = 0.11; Fig. 4A) and EtOH preference (F _{(1, 20)} = 6.386, p < 0.0001, η²_p = 0.18; Fig. 4D). Sob-AM2 is a thyromimetic, and primarily affects gene expression via actions of central THRs, thus, effects are expected to occur after several days of dosing (Ferrara and Scanlan 2020). Additional studies confirm that molecular and behavioral changes induced by Sob-AM2 take several days of dosing to arise, presumably because of the slower actions of nuclear hormone receptors as transcription factors (Bohlen et al. 2023; Bárez-López et al. 2018; Chaudhary et al. 2021; Valcárcel-Hernández et al. 2023). Therefore, we compared the effects of Sob-AM2 on EtOH intake and preference at baseline (BL), the last day of low-dose treatment (6), and the last day of high-dose treatment (10) using three-way repeated measures ANOVAs of treatment, sex, and time to assess the cumulative effects of multiple days of treatment at either dose. Analysis of EtOH intake revealed a significant interaction of sex by treatment by time on EtOH intake (F _{(2, 40)} = 4.686, p < 0.05, η²_p = 0.19; Fig. 4B). To further parse this interaction, data were separated by sex, and two-way ANOVAs were performed of treatment by time on EtOH intake within both sexes respectively. In males, a significant interaction of treatment by time was observed (F _{(2, 20)} = 3.676, p < 0.05, η²_p = 0.27; Fig. 4B). Tukey’s post-hoc comparisons in males revealed that on the final day of high-dose (10), Sob-AM2 treated males had significantly lower EtOH intake compared to vehicle males (p < 0.001, d = 3.02) and compared to Sob-AM2 male baseline drinking levels (p < 0.05, d = 2.33). Additionally, Sob-AM2 males had significantly decreased EtOH intake (p < 0.05, d = 1.50) by the final day of low-dose treatment, compared to baseline, while vehicle treated significantly increased their EtOH intake from the final day of low-dose treatment to the final day of high-dose treatment (p < 0.01, d = 0.58). In female mice, a two-way ANOVA also revealed a significant interaction of treatment by time (F _{(2, 20)} = 10.05, p < 0.001, η²_p = 0.50; Fig. 4B). Tukey’s post-hoc comparisons revealed that on the final day of high-dose, Sob-AM2 treated females had significantly lower EtOH intake compared to vehicle females (p < 0.05, d = 1.50), and had significantly reduced their intake compared to low-dose treatment (p < 0.05, d = 1.96). Meanwhile, vehicle treated females, akin to males, increased EtOH intake from the final day of low-dose treatment to the final day of high-dose treatment (p < 0.01, d = 1.02). After observing significant decreases in EtOH intake in both males and females on the final day of high-dose treatment we further probed how individual mice changed their EtOH intake from baseline to the final day of high-dose treatment by calculating a percentage change from baseline in vehicle and Sob-AM2 treated mice. A two-way ANOVA of the percentage change in EtOH intake from baseline to the final day of high-dose treatment revealed a significant effect of treatment (F _{(1, 20)} = 6.107, p < 0.05, η²_p = 0.23; Fig. 4C) indicating a greater decrease from baseline in Sob-AM2 mice than in vehicle treated control mice that displayed no change from baseline.

To confirm if the observed decreases in EtOH intake were selective to EtOH, we assessed the effects of Sob-AM2 on EtOH preference across the same three timepoints (BL, 6, and 10). Due to a significant interaction effect of sex on EtOH intake, and to maintain internal consistency between comparisons we also separated our analysis of EtOH preference by sex and performed two-way ANOVAs within each sex across the same timepoints. Analysis of EtOH preference in males revealed a significant interaction of treatment by time (F _{(2, 20)} = 4.624, p < 0.05, η²_p = 0.32). Tukey’s post-hoc comparisons indicated that Sob-AM2 male mice had significantly reduced EtOH preference compared to vehicle treated male mice at the final day of low-dose (p < 0.01, d = 2.08) and the final day of high-dose (p < 0.001, d = 3.38). Meanwhile Sob-AM2 male mice had significantly reduced EtOH preference on the final day of high-dose compared to the final day of low-dose (p < 0.05, d = 1.61) and at baseline (p < 0.05, d = 2.26). In female mice, analysis revealed a significant interaction of treatment by time (F _{(2, 20)} = 11.89, p < 0.001, η²_p = 0.54). Tukey’s post-hoc comparisons indicated that Sob-AM2 treated female mice had significantly reduced EtOH preference compared to vehicle treated female mice at the final day of high-dose (p < 0.05, d = 1.87), and that Sob-AM2 treated female mice had reduced EtOH preference on the final day of high-dose compared to the final day of low-dose (p < 0.05, d = 2.12). In contrast, vehicle treated female mice increased EtOH preference across the same timepoints (p < 0.05, d = 0.90). Finally, to further assess changes in EtOH preference within each mouse treated with Sob-AM2 or vehicle from baseline to the final day of high-dose we assessed the percentage change from baseline between these two timepoints. A two-way ANOVA of sex by treatment revealed a significant effect of treatment (F _{(1, 20)} = 12.27, p < 0.01, η²_p = 0.38) indicating a greater decrease in EtOH preference from baseline in Sob-AM2 mice than in vehicle treated mice who displayed no change from baseline. Taken together these results indicate that Sob-AM2 selectively reduces EtOH intake and not H₂O intake in both male and female mice.

Effects of Sob-AM2 on H₂O Drinking and total fluid intake in EtOH Drinking Mice

In addition to analyzing effects of Sob-AM2 on EtOH intake and preference, we also assessed its effects on H₂O intake (g/kg) in the same mice and across the same timepoints as EtOH intake (Fig. 4). Using a three-way repeated measures ANOVA with factors of treatment, sex, and time, we first evaluated H₂O intake (g/kg) across baseline, the habituation injection day, and 10 days of treatment with Sob-AM2 (6 days at 0.3 mg/kg & 4 at 1.0 mg/kg; Supplementary Fig. 2A). These analyses revealed significant interactions of treatment by time on H₂O intake (F _{(11, 220)} = 10.29, p < 0.0001, η²_p = 0.24; Supplementary Fig. 2A). To further probe this interaction, we compared H₂O intake at the same timepoints as previously for EtOH intake and preference (BL, 6, 10) using two-way ANOVAs of treatment by time for each sex. Analyses of H₂O intake in male mice indicated a significant interaction of treatment by time (F _{(10, 20)} = 3.143, p < 0.01, η²_p = 0.42; Supplementary Fig. 2B). Tukey’s post-hoc comparisons revealed that Sob-AM2 males had significantly higher H₂O intake on the last day of low-dose (p < 0.05, d = 1.60) and the last day of high-dose (p < 0.05, d = 2.03) compared to vehicle males. Meanwhile, Sob-AM2 males had significantly higher H₂O intake at the last day of high-dose than at the last day of low-dose (p < 0.01, d = 0.81) and at baseline (p < 0.05, d = 1.98). In females, analyses also revealed a significant effect of treatment by time (F _{(2, 20)} = 14.35, p < 0.001, η²_p = 0.60; Supplementary Fig. 2B). Tukey’s post-hoc comparisons revealed that Sob-AM2 females had significantly higher H₂O intake than vehicle treated females at the last day of high-dose (p < 0.01, d = 2.12). Meanwhile, Sob-AM2 females had significantly higher H₂O intake on the last day of high-dose compared to the last day of low-dose (p < 0.05, d = 1.75) and compared to baseline (p < 0.05, d = 2.33). Finally, to further assess the change in H₂O intake within individual mice we compared the percentage change from baseline to the last-day of high-dose treatment between Sob-AM2 and vehicle treated mice using a two-way ANOVA with factors of sex and treatment. Analyses revealed a significant effect of treatment (F _{(1, 20)} = 14.46, p < 0.001, η²_p = 0.42; Supplementary Fig. 2F). With these findings, we confirm that H₂O intake was increased in Sob-AM2 treated mice in conjunction with the previously observed decreases in EtOH intake and preference compared to vehicle control mice.

Next, we assessed if Sob-AM2 would result in a significant change in total fluid consumption regardless of fluid type (combined mL of EtOH and H₂O). As with previous comparisons, we first evaluated total fluid intake across baseline, the habituation injection day, and 10 days of treatment with Sob-AM2 (6 days at 0.3 mg/kg & 4 at 1.0 mg/kg; Supplementary Fig. 2A). A three-way repeated measure ANOVA revealed a significant interaction of treatment by time. total fluid intake (F _{(11, 220)} = 6.493, p < 0.0001, η²_p = 0.25; Supplementary Fig. 2D) We again compared the data at the same three timepoints (BL, 6, 10) using two-way ANOVAs of treatment by time for each sex. Analysis of male total fluid intake indicated a significant interaction of treatment by time (F _{(2, 20)} = 3.982, p < 0.05, η²_p = 0.29; Supplementary Fig. 2E). Tukey’s post-hoc indicated that Sob-AM2 treated males had significantly higher total fluid consumption than vehicle treated males at the final day of low-dose (p < 0.05, d = 1.48) and at the final day of high-dose (p < 0.05, d = 1.79). Meanwhile, Sob-AM2 treated males had significantly higher total fluid intake on the final day of high-dose than on the final day of low-dose (p < 0.01, d = 0.65). In female mice, analyses also revealed a significant interaction of treatment by time (F _{(10, 20)} = 1.375, p < 0.01, η²_p = 0.40; Supplementary Fig. 2E). Tukey’s post-hoc interactions indicate that at baseline prior to treatment total fluid intake was significantly lower in Sob-AM2 females than in the vehicle group (p < 0.05, d = 0.55), however, there were no significant differences at other timepoints. Meanwhile, Sob-AM2 treated mice did have significantly higher total fluid intake compared to baseline at the final day of low-dose (p < 0.05, d = 1.02) and at the final day of high-dose (p < 0.05, d = 2.33) indicating an elevation of total fluid intake within these mice as treatment progressed. Finally, to further evaluate the change in total fluid intake within individual mice we assessed the percentage change from baseline to the end of high-dose treatment using a two-way ANOVA with factors of treatment and sex. Analysis revealed a significant effect of treatment (F _{(1, 20)} = 11.40, p < 0.01, η²_p = 0.36; Supplementary Fig. 2F) indicating that Sob-AM2 treated mice had a greater increase from baseline in total fluid intake than did vehicle treated controls who observed relatively no change from baseline. These results indicate that while EtOH intake is decreased, and H₂O intake is increased, such that total fluid consumption is overall increased in mice treated with Sob-AM2 compared to vehicle controls.

Effects of Sob-AM2 on Serum Thyroid Hormones and Bodyweight in EtOH Drinking Mice

Treatment with Sob-AM2 has been shown previously to alter some elements of the HPT axis (Ferrara et al. 2018). Thus, we aimed to compare both bodyweight and serum thyroid hormone measures in EtOH 2BC mice treated with either Sob-AM2 or vehicle. First, we compared bodyweight of Sob-AM2 and vehicle treated mice prior to treatment and following treatment. A three-way ANOVA with factors of sex, treatment, and time revealed a significant effect of sex (F _{(1, 20)} = 154.7, p < 0.0001, η²_p = 0.99; Fig. 5A) indicating that male mice had higher bodyweight than females, and of time (F _{(1, 20)} = 84.37, p < 0.0001, η²_p = 0.81; Fig. 5A) indicating that all groups increased bodyweight post-treatment, regardless of treatment.

Figure 5.

Effects of Sob-AM2 treatment on serum thyroid hormones in EtOH drinking mice. (A) Bodyweight pre and post treatment. #### indicates a significant effect of sex (p < 0.0001). (B) Serum TSH levels. (C) Serum T3 levels. (D) Serum T4 levels. **** indicates a significant effect of Sob-AM2. (E) Serum T3/T4 ratios. **** indicates a significant main effect of Sob-AM2.

To assess the effects of Sob-AM2 treatment on peripheral thyroid hormones in EtOH drinking mice T3, T4, T3/T4 ratios, and TSH levels in serum were compared between Sob-AM2 treated mice and vehicle controls using two-way ANOVAs with factors of sex and treatment. We found that neither serum T3 nor TSH in Sob-AM2 treated mice differed from vehicle treated controls. Meanwhile, serum T4 was significantly decreased (F _{(1, 20)} = 82.87, p < 0.0001, η²_p = 0.81; Fig. 5D) and, in conjunction, serum T3/T4 ratios were elevated (F _{(1, 20)} = 21.17, p < 0.001, η²_p = 0.51; Fig. 5E) in Sob-AM2 mice compared to vehicle controls. Despite these changes in serum T4 levels, bodyweight was not significantly altered by Sob-AM2 treated mice compared to vehicle treated controls regardless of sex. In conjunction with unchanged peripheral serum T3 levels these findings suggest that Sob-AM2 did not result in observable changes to general metabolism in EtOH drinking mice.

Effects of Sob-AM2 on Gene Expression in the PVN of EtOH Drinking Mice

Due to the centrally acting nature of Sob-AM2 we tested if treatment with Sob-AM2 in these same EtOH drinking mice altered gene targets of thyroid hormone within the PVN. We found that Sob-AM2 treatment decreased Trh mRNA expression regardless of sex (F _{(1, 14)} = 4.956, p < 0.05). The Trh gene is canonically regulated by THRβ at several elements in the gene promoter (Guissouma et al. 2002; Sugrue et al. 2010), and a reduction in Trh mRNA as a consequence of Sob-AM2 has been reported previously in male mice (Ferrara et al. 2018). We also observed a significant decrease in Oxt expression in Sob-AM2 mice compared to vehicle controls (F _{(1, 14)} = 4.769, p < 0.05; See Supplementary Table 1). No other effects on targets of the thyroid hormone system were observed (See Supplementary Table 1).

Discussion:

Thyroid hormones regulate a myriad of behaviors through actions at their nuclear hormone receptors in the brain: including feeding, affective states, and reward related behaviors (Bernal 2007; Bauer and Whybrow 2002). Likewise, clinical studies have linked HPT axis dysregulation to alcohol use disorder (Hermann et al. 2002). Here, we hypothesized that CNS thyroid hormone receptors would have a functional relationship with voluntary EtOH consumption. In turn, we show that voluntary alcohol consumption in mice results in both central and peripheral dysregulation of thyroid hormone signaling. In addition, we show for the first time that subcutaneous treatment with a CNS selective thyromimetic reduces alcohol drinking and preference and suggest potential mechanisms for these effects.

Voluntary EtOH consumption attenuates PVN Thrb expression and dysregulates peripheral thyroid hormones

To our knowledge, we provide the first evidence that thyroid hormone receptor, specifically Thrb, mRNA expression is disrupted by voluntary alcohol drinking in mice. THRβ is an important transcriptional regulator (Jones et al. 2003; Ayers et al. 2014) that is abundantly expressed within the PVN of the hypothalamus across species (Lechan et al. 1994; Alkemade et al. 2005). It is well known that THRβ plays an important role in regulating the HPT axis through negative regulation of Trh gene expression in TRH neurons of the PVN (Guissouma et al. 2002; Nillni 2010; Sugrue et al. 2010). Interestingly, others have reported an elevation of Trh mRNA in the hypothalamus of rats following chronic EtOH injections, which was accompanied by decreased stimulation of thyroid hormone release in a cold-test (Zoeller et al. 1996). Thus, attenuated expression of Thrb due to EtOH observed in our study could reduce inhibition of Trh gene and explain the elevation of TRH due to EtOH reported in several studies (Zoeller et al. 1996; Zoeller and Rudeen 1992). However, we found no indication of EtOH induced dysregulation of Trh mRNA in our study (Fig. 3). This discrepancy could be explained by difference in the route of exposure, length of exposure or perhaps differences in endogenous ligand binding (T3) that compensate for any Trh effects induced by the loss of Thrb expression. Regardless, our data add to the growing body of literature suggesting that chronic EtOH dysregulates thyroid hormone signaling within the hypothalamus. Future studies should further examine these interactions and establish if Thrb related dysregulation by EtOH may underly blunting of the HPT axis response reported clinically in alcohol drinking patients.

In addition to central attenuation of Thrb expression, we also observed that peripheral serum T3 levels were elevated in EtOH drinking mice regardless of sex (Fig. 2F) without expected decreases in serum TSH. This finding suggests that negative feedback of the HPT axis may be dysregulated by daily alcohol consumption. If T3 levels are consistently elevated by EtOH consumption, then this could lead to attenuated Thrb expression in the PVN (Fig. 3C) through allostatic compensation. Curiously however, elevations in T3 are in contrast to previous preclinical studies suggesting that EtOH can decrease peripheral T3, T4, and TSH measures (Singh et al. 1979; Mason et al. 1988; Rasmussen 2003). While this may appear as an inconsistency, more recent clinical literature further supports a potential bidirectional relationship between thyroid hormone signaling and alcohol consumption that may depend on the dose of EtOH exposure. Alcohol craving and compulsive drinking scores positively correlate with T3 measures in some studies (Aoun et al. 2015; Leggio et al. 2008), while others report a negative relationship between T3 measures and alcohol consumption specifically in heavy drinkers (Gruppen et al. 2020). Accordingly, the effects reported in the current study are observed in mice that had lower EtOH intake (~11g/kg daily; Figure 2B) than what is generally published from C57BL/6J mice in a voluntary 2BC procedure (over 12 g/kg daily). Although BECs were not measured in this study, previous work in our lab has shown that mice attain a range of pharmacologically relevant BEC’s (5.97mg/dL-77.58mg/dL) in 2BC with lower EtOH intakes than those observed in our current study (Fulenwider et al. 2021). Though it should be mentioned that while such BECs are achieved, they are not maintained throughout the entirety of the dark cycle (Dole and Gentry 1984). Regardless, the doses of EtOH that mice are exposed to in our voluntary 2BC procedure are likely lower than those in previous preclinical studies utilizing chronic forced EtOH exposure procedures, which report reductions in serum thyroid hormone levels and are also presumably more stressful. Additionally, it is possible that circadian dysregulation plays a role in observed differences in serum thyroid hormones as one study found that EtOH exposure increased serum T3 just before lights out, similar to our study, but had the opposite effect early in the light cycle (Rasmussen 2003). We do not suspect that elevations in T3 are simply due to caloric value of EtOH as bodyweight and other HPT measures (T4, TSH, Trh) remain unchanged from H₂O drinking controls.

Sob-AM2 decreases voluntary EtOH intake and preference while increasing H₂O and total fluid intake

We show for the first time that treatment with CNS selective thyromimetic Sob-AM2 significantly reduces EtOH intake and preference in mice in a 2BC drinking paradigm and thus decreases EtOH consumption selectively. More specifically, we show that 24-hour EtOH intake and preference are decreased by 6 days of treatment with the 0.3 mg/kg dose in male mice, and with 4 days of treatment with the 1.0 mg/kg dose in both sexes (Fig. 4). The reasons underlying the lack of effects in females at the 0.3 mg/kg dose remain unclear. However, THRs compete for binding with estrogen receptors at the DNA to modulate gene expression due to shared elements of their DNA binding motifs (Dellovade et al. 1996; Vasudevan et al. 2002). Thus, it is possible that circulating ovarian hormones such as estrogen may delay or dampen the effects of Sob-AM2 on EtOH consumption at the 0.3 mg/kg dose in females, and such interactions are over-powered with longer dosing or at higher doses of Sob-AM2. Furthermore, Sob-AM2 has been shown to take multiple days to produce effects on behavior and on gene expression (Bohlen et al. 2023; Bárez-López et al. 2018; Chaudhary et al. 2021; Valcárcel-Hernández et al. 2023). The design of our experimental paradigm does not afford us the ability to distinguish if decreases in EtOH intake and preference were due to increasing of dose, or cumulative effects of dosing over time. It is possible that the 0.3 mg/kg dose of Sob-AM2 could achieve comparable decreases in EtOH consumption with dosing over a longer period. Future studies should conduct between subject dose-response experiments to elucidate the effects of the low-dose further. Nonetheless, effects of Sob-AM2 treatment on EtOH consumption are promising, as in recent years thyroid hormone treatment has gained traction in its potential to treat mood disorders such as bipolar disorder and major depression (Bauer and Whybrow 2021; Bauer et al. 2016; Singh and Sundaresh 2022) and these disorders are often co-morbid with AUDs (McHugh and Weiss 2019; Farren et al. 2012). Likewise, a CNS selective thyromimetic is currently in clinical trials for mood disorders (NCT06633016, NCT06869187). Future experiments should also validate the effects of Sob-AM2 in other pre-clinical models of AUD, such as the “drinking-in-the-dark” model of binge EtOH consumption, EtOH dependence-enhanced drinking models, and operant EtOH self-administration.

Importantly, we observed that Sob-AM2 treatment concomitantly increased H₂O intake and total fluid intake in EtOH consuming mice that reduced their EtOH intake and preference (Supplementary Fig. 2; Fig. 4). This observation indicates selectivity of Sob-AM2s effects in decreasing EtOH consumption rather than unidirectionally modulating all fluid intake. Increases in H₂O were not dependent on the presence of EtOH as increases in H₂O consumption from baseline were also observed in EtOH naïve mice (Supplementary Fig. 3C). Such findings are not entirely surprising, as it is well established that thyroid hormone can modulate consummatory behaviors through hypothalamic nuclei (Hameed et al. 2017; Alva-Sánchez et al. 2012). For example, T4 treatment in adult C57BL/6J mice, which resulted in elevations of T4 concentrations in the brain and periphery, increased daily H₂O and food consumption (Niedowicz et al. 2021). Thus, it would be apt to suspect that Sob-AM2 produces its behavioral effects via promoting consummatory behavior centrally in a similar manner to T4 treatment and thus increases polydipsia. However, such an explanation would fail to indicate why, in EtOH 2BC, only H₂O intake is selectively increased while EtOH intake and preference are conversely decreased in Sob-AM2 treated mice. Such a distinction suggests a specific and more complex relationship between Sob-AM2 treatment and central thyroid hormone signalling on EtOH consumption that is not observed in H₂O intake alone. It could be theorized that Sob-AM2 increases taste aversion to EtOH, as thyroidal status can influence taste preferences in both humans (McConnell et al. 1975; Deniz et al. 2016; Baskoy et al. 2016) and rodents (Gordon et al. 1992; Brosvic et al. 1992). However, we believe that it is unlikely to be the primary factor in our effects as EtOH naïve mice also elevated their H₂O consumption in the absence of EtOH (Supplementary Fig. 3C). Nonetheless, it would also be apt to test Sob-AM2’s effects on quinine related taste aversion in future studies, and likewise investigate the effects of Sob-AM2 treatment on other naturally rewarding fluids in 2BC drinking models to confirm specificity of such an effect to EtOH.

Potential mechanisms of Sob-AM2’s effects on EtOH and H₂O consumption

Activity of prodrug agonist Sob-AM2 is designed for and has shown highly selective transport to the CNS via cleavage of its amide group by central FAAH receptors, while it remains inert in peripheral tissues (Ferrara and Scanlan 2020; Meinig et al. 2017). Thus, we suspect that Sob-AM2s effects on EtOH drinking are produced by activation of central THRs and resultant changes in gene transcription. The CNS selectivity of Sob-AM2 does not exclude possible effects of centrally mediated dysregulation of peripheral HPT measures, which in turn may impact EtOH consumption as treatment with Sob-AM2 has been shown to deplete peripheral T3 and T4 levels, although it does so without resulting in a hypothyroid phenotype (Ferrara et al. 2018). It is possible that depleted T4 levels and elevated T3/T4 ratios induced by Sob-AM2 (Fig. 5D-E) contribute to the observed decreases in EtOH intake and preference in our study. However, this explanation would run contrary to other work indicating that depleting peripheral thyroid hormone levels pharmacologically results in increased EtOH consumption in rats (Echeverry-Alzate et al. 2019). Furthermore, T3 levels in the periphery and bodyweight of Sob-AM2 treated EtOH drinking mice remained normal compared to vehicle treated controls in our study (Fig. 5A, E) suggesting that baseline metabolism had not been significantly altered by treatment with Sob-AM2 in these mice. Such evidence suggests a specific role for central thyroid hormone signalling in EtOH consumption that is largely independent of peripheral thyroid hormone receptor activity.

As Sob-AM2 is a central modulator of the HPT axis, it is tempting to suspect that effects of Sob-AM2 on EtOH intake and preference are mediated via activation of hypothalamic THRβ (Taylor et al. 1990; Kondo et al. 2022). To this point, data presented here indicate a dysregulation of Thrb mRNA in the PVN by EtOH drinking, and previous studies have highlighted thyroid hormone-related dysregulation of the hypothalamus by EtOH exposure in rats (Zoeller et al. 1996; Zoeller and Rudeen 1992). Sob-AM2 preferentially activates central THRβ, and to a lesser degree THRα (Meinig et al. 2017; Ferrara and Scanlan 2020), which act as transcription factors modulating transcription at various sites throughout the genome (Bernal 2007; Jones et al. 2003). Thus, it is possible that Sob-AM2 directly alters the expression of genes known to regulate EtOH drinking and fluid consumption in the hypothalamus. For example, hypothalamic vasopressin and oxytocin expression and release, which are involved in both alcohol drinking behaviors and regulating fluid homeostasis (Scarboro et al. 2024; Antunes-Rodrigues et al. 2004; Ryabinin and Fulenwider 2021), are sensitive to fluctuations in thyroid hormone signalling (Stohn et al. 2018; Adan et al. 1992; Ciosek and Drobnik, 2004; Ciosek and Izdebska 2009) and could pose intriguing transcriptional targets for mediating the divergent effects of Sob-AM2 on EtOH and H₂O intake observed in our study. In support of such a hypothesis, we did observe decreased Oxt mRNA expression in EtOH drinking mice treated with Sob-AM2, but not in EtOH naïve mice treated with Sob-AM2 (Supplementary Table 1). However, THRβ is expressed across the brain, and further experiments would be needed to confirm the neuroanatomical specificity of Sob-AM2s behavioral effects to the hypothalamus. Alternatively, thyroid hormone signalling has been shown to influence reward related behaviors by modulating transcription in the PFC (Hochbaum et al. 2024), the ventral tegmental area (Bennett et al. 2024), and the medial amygdala (Walker et al. 2022). Likewise, central thyroid hormone signalling has considerable influence on gene expression in GABAergic (Richard et al. 2020) and monoaminergic systems (Ookubo et al. 2015; Bauer et al. 2002), which play critical roles in reward related behaviors. Thus, it is reasonable to surmise that behavioral effects of Sob-AM2 treatment on EtOH consumption are likely mediated via THRβ agonism in the CNS through transcriptional reprogramming of multiple brain regions. The specific brain regions and transcriptional targets involved in this behavioral phenomenon remain speculative and should be the topic of future studies.

Conclusion:

Taken together, these findings reveal a role for central thyroid hormone receptor expression and activity in alcohol consumption and preference in mice. We establish for the first time that Thrb mRNA is significantly decreased in the PVN of the hypothalamus of EtOH drinking mice, and that a CNS selective THRβ agonist, via Sob-AM2, significantly decreases both EtOH intake and EtOH preference in mice, while increasing H₂O intake. These results further suggest that THRβ agonism in the CNS represents a potentially viable clinical target for treatment of AUD.

Figure 6.

Effects of Sob-AM2 treatment PVN mRNA expression in EtOH drinking mice normalized to the geometric mean of Hprt1 and 18s. (A) PVN thyrotropin releasing hormone expression. * Indicates significant effect of EtOH (p < 0.05). (B) PVN thyroid hormone receptor-alpha expression. (C) PVN thyroid hormone receptor-beta expression.

Highlights:

Thrb mRNA expression is reduced in the paraventricular nucleus of the hypothalamus by voluntary alcohol drinking in C57BL/6J mice.

Treatment with CNS selective THRβ agonist, Sob-AM2, reduces voluntary alcohol consumption and preference C57BL/6J mice.

Findings indicate therapeutic potential in targeting central thyroid hormone receptors for treatment of alcohol use disorder.