Methylation of the Oxytocin, Oxytocin Receptor, and Vasopressin Gene Promoters in Tobacco Use Disorder during Cessation

Tobacco use disorder (TUD) is one of the leading causes of preventable death [1]. In 2019, tobacco use attributed to approximately 7.69 million deaths and 200 million disability-adjusted life years [2]. While several pharmacological and psychological interventions have been investigated, successful treatment to induce smoking cessation remains challenging and complex [3]. As potential therapeutic targets, several neuropeptides have been studied, e.g., proopiomelanocortin [4], brain-derived neurotrophic factor [5, 6], leptin [7], and atrial natriuretic peptide [8].

In recent years, attention has shifted toward the nonapeptides arginine vasopressin (AVP) and oxytocin (OT) due to their therapeutic potential in treating substance use disorders (SUD). AVP and OT are both secreted in the neurohypophysis and structurally very similar, with only two differing amino acids at positions 3 and 8 [9, 10]. They share the ability to bind at both OT and vasopressin receptors [11, 12]. In humans, both proteins have similar influences on pair bonding, social behavior, and stress reactions [13] while also modulating the activation of the hypothalamus-pituitary-adrenal axis (HPA-axis) [14]. Hypothalamic neurons secreting OT and AVP connect to brain areas such as the amygdala, brainstem, and anterior pituitary [15]. Although both hormones share evolutionary [16], genetic, and functional similarities [17], their function and physiological effects also seem to differ across species and in the male and female sex [18‒21].

HPA-axis activation differs in people suffering from SUD (see, e.g. [22, 23]). From that perspective, it makes sense to investigate the OT and AVP system regarding a possible influence on craving, which is associated with the function of the HPA-axis [24, 25]. Several studies in animals have investigated both proteins as potential therapeutic targets. One recent study has shown that OT can attenuate methamphetamine seeking and demand through its effect on the nucleus accumbens (NuAcc) [26]. OT also reduced cocaine-induced increases in turnover of the neurotransmitter dopamine inside the NuAcc [27]. Interestingly, intracerebroventricular administration of OT could completely block ethanol-induced dopamine release within the NuAcc of rats [28].

In another study, the AVP1b receptor antagonist SSR149415 dose-dependently reduced excessive levels of ethanol self-administration in dependent animals [29]. In rats that underwent chronic intermittent escalating heroin administration, SSR149415 blunted the HPA activation that resulted from experimental stressors [30]. SSR149415 and naltrexone also synergistically decrease excessive alcohol drinking in mice [31]. Regarding smoking, chronic treatment of mice with SSR149415 prevents dysphoria associated with nicotine withdrawal [32].

However, conflicting evidence exists regarding their potential role in treating human SUDs. One phase 2, double-blind, placebo-controlled randomized trial on the efficacy of ABT-436 (V1b receptor antagonist) found no difference in measures of drinking, alcohol craving, or alcohol-related consequences [33]. Regarding OT, in one study, patients received either intranasal OT or a placebo, followed by a measurement of craving and cigarette demand. On average, the participants smoked fewer cigarettes after receiving OT compared with placebo, while there was no significant decrease in cigarette demand or craving [34].

In an RCT performed by the same research group 1 year later, the authors assessed the effect of OT on stress and tobacco craving. Interestingly, OT did not alter the response to stress, cigarette craving, anxiety, heart rate, blood pressure, and cortisol levels [35]. Another recent RCT also reported no effect of intranasal OT administration on cigarette craving [36]. This contrasts with a study that reported evidence of OT’s potentially attenuating effect on stress and drug craving [37].

Oxytocin-receptor (OXTR) promoter methylation could explain differences in the presented evidence. As its receptor mediates the effect of OT [38, 39], it is fair to assume that methylation of the OXTR promoter would influence the impact of endogenous and exogenous OT. Furthermore, OT and AVP both bind to OXTR, and vice versa. Thus, changes in OXTR promoter methylation also affect AVP’s effect [11].

It is well established that OT and AVP influence drug-seeking behavior (including craving) and have sex-specific effects on behavior in mammals, including mice, rats, and humans. Thus, we believe AVP and OT are regulated differently in males compared to females with TUD, which could have therapeutic implications regarding intranasal OT or AVP receptor antagonists. With this in mind, we hypothesize sex-specific changes in methylation levels of the AVP, OT, and OXTR genes during nicotine withdrawal. Furthermore, we hypothesize sex-specific differences in the mean methylation of AVP, OT, and OXTR in smokers compared with healthy controls. Since human and animal studies have gathered evidence that OT and vasopressin influence drug craving, we hypothesize that changes in mean methylation are associated with tobacco craving.

The present study is part of a longitudinal case-control study on the effect of nicotine withdrawal on appetite-regulating and volume-regulating peptides in nicotine dependency. It was conducted at the Department of Psychiatry, Social Psychiatry, and Psychotherapy of the Hannover Medical School in Germany. This study protocol was reviewed and approved by the Ethics Committee of Hannover Medical School (MHH), approval number 6695 and adhered to the Declaration of Helsinki. All participants gave written informed consent before entering the study. As exclusion criteria, we defined concurrent mental disorders, other substance use, cerebral ischemia or hemorrhage, epilepsy, cardiovascular or renal diseases, age under 18 years, pregnancy, and nicotine replacement therapy. Inclusion criteria were age between 18 and 65 years, tobacco dependence as defined by the International Classification of Diseases and Diagnostics (ICD-10) and Statistical Manual of Mental Disorders (DSM IV), and current smoker (defined as a minimum of seven cigarettes per week or one cigarette a day). For the control group, inclusion criteria were current nonsmoking and having smoked less than seven cigarettes during their lifetime. We assessed the severity of TUD using the Fagerström Test [40]. For craving assessment, we used the Questionnaire on Smoking Urges (QSU) [41].

After enrollment, all smokers underwent a detailed physical examination, routine laboratory testing, and drug screening for cotinine (urine). Fasting blood samples and urine samples were drawn from the TUD group on days 1, 7, and 14 of abstinence. On day one, the smokers were asked to smoke their last cigarette, blood was drawn, and urine was sampled immediately afterward. Following that, they completed psychometrics. From the controls, only a single time point was assessed. All samples were drawn in the morning between 08:00 and 10:00 am. Blood samples were anticoagulated with sodium EDTA. Separation of plasma was immediately done in a centrifuge (4,000 g). The aliquots were stored at −80°C after centrifugation until processing.

Extraction and clean-up of genomic DNA from blood were done using the NucleoMag^® Blood 200 µL DNA Kit (Macherey-Nagel, Düren, Germany). The bisulfite conversion of the acquired DNA samples was done using the EpiTect^® 96 Bisulfite Kit (Qiagen, Hilden, Germany) for OXTR and AVP, and EZ96 Magprep Bisulfite Kit (Zymo Research Europe, Freiburg, Germany) for OT following the manufacturer’s protocol.

All primers were manually designed to bisulfite-converted regions of the respective genes using the program Geneious (Biomatters, Auckland, New Zealand). All fragments covered the proximal promoter region of the respective gene. Before decision on the area of investigation for the promoter in question, we studied the literature as well as the biology of the promoter. As different areas play a role for individual genes, we carefully selected that which made most sense to us when considering the literature and previous studies. For our analysis of the OT gene, we designed and used the forward primer F1 TTT​GTT​TTA​TTT​TAG​TGG​TTT​AGG​TTA​T and reverse primer R1 CTC​AAC​TCC​TAA​AAT​TCT​CAA​A. The total fragment size was 512bp. For OXTR, we designed the forward primer F1 GGT​ATT​TTA​TTT​TTT​GTG​TTT​AGA​TTA​T and reverse primer ACT​CAC​TAC​AAA​CTC​TAC​CTC​C. The total fragment size was 379bp. For AVP, we designed the forward primer F1 GAA​AGT​TTA​GAG​ATG​GTT​TTT​AGG​T and the reverse primers R1 CAA​CCC​TAA​AAT​AAC​CCA​CAA​TA and R2 CAT​CCT​AAT​ACA​CAC​AAA​TAA​ACC. The total fragment size was 364bp.

We used the primer analysis tool Netprimer (http://www.premierbiosoft.com, accessed on August 01, 2022) to check for the presence of secondary structures (e.g., hairpins, self-primer, etc.). Melting temperatures were also retrieved from Netprimer analysis. The resulting primers were ordered from Metabion (Metabion, Steinkirchen, Germany).

All polymerase chain reactions (PCRs) were performed in a C1000™ Thermal Cycler (Bio-Rad, Hercules, CA, USA). The AVP, OT, and OXTR target sequences of the purified bisulfite-converted DNA were amplified following a standard PCR protocol. The reaction components used were as follows: 0.4 μL (20 μmol) forward primer (F1 for AVP, F1 for OXTR, F1 for OT), 0.4 μL (20 μmol) reverse primer (R1/R2 for AVP, R1 for OXTR, R1 for OT), 1 μL of DNA, 3.2 μL of H₂O, 5 μL HotStarTaq^® Master Mix Kit (Qiagen, Hilden, Germany). The bisulfite primer amplification temperature was set at 61°C (AVP)/53°C (OXTR)/57°C (OT) for the PCR. Amplification products were purified using the Agencourt^® AMPure^® XP magnetic beads (Beckman Coulter).

Sequencing PCR for the target fragment was performed using a BigDye^® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and an Applied Biosystems/HITACHI 3500xl Genetic Analyzer (Applied Biosystems) according to the manufacturer’s instructions. The bisulfite primers R2_AVP, F1_OXTR, and R1_OT were used in 5:00 pm concentrations for the PCR of the respective genes. The sequencing PCR products were purified using the Agencourt CleanSeq^® XP magnetic beads (Beckman Coulter) and then used for sequencing. Electropherograms and sequences detected by the genetic analyzer were analyzed using the specialized epigenetic sequencing methylation analysis software to determine the methylation rates for every CpG locus.

We used the Epigenetic Sequencing Methylation Software (ESME) software package to determine methylation rates. It aligns the generated sequence and the reference sequence for comparing methylation at each CpG site. CpG islands were then labeled in reference to the location from the first base pair of exon 1 (i.e., CpGp 134 would be 134bp in the 5′ direction from the exon 1 [p = plus], CpGm112 would be 112 bp in the 3′ direction from the exon 1 [m = minus]). We calculated quantitative methylation for each site per subject (proportion of cytosine and thymine normalized peak values) [42].

For all statistical analyses, we used the Statistical Package for Social Sciences 27 (SPSS, IBM, Armonk, NY, USA) and GraphPad Prism version 7 and 9 (San Diego, CA, USA), which we also used for data illustration. As a standard procedure in our laboratory, we performed quality control of our methylation analysis. We included only those CpGs with less than 5% missing values for each gene. Then, we excluded each sample with more than 5% missing values. After applying these criteria, the total number of CpGs for OT, OXTR, and AVP were 30, 29, and 10, respectively.

From visual inspection, normality was present in the methylation data. Therefore, we decided to perform parametric tests accordingly.

The association of psychometric variables with methylation values was done using Spearman’s rank correlation. For clarity, we decided to limit this analysis to mean methylation values and CpGs with significantly different methylation levels in smokers versus controls. To test for differences in QSU scores across the withdrawal period, we used the nonparametric Friedman’s Test for dependent samples, followed by Dunn’s test for pairwise comparison. The Bonferroni correction was applied to correct multiple testing.

To compare mean methylation and methylation at specific CpGs between subjects and controls, we used the t test for independent samples (comparing controls with time point 1), followed by a correction for multiple testing using Bonferroni’s method. The corrected alpha levels were 0.00166, 0.00172, and 0.005 for OT, OXTR, and AVP, respectively. To assess changes in gene methylation throughout the cessation period, we fitted a mixed linear model. In this model, mean methylation was set as the dependent variable, while CpG position, sex, and time point were selected as factors and QSU score as a covariate. The resulting estimated marginal means (EMMs) are used to compare methylation values corrected for the influence of the factors. We used post hoc tests to compare mean methylation at the different time points (Bonferroni’s method) and Bonferroni’s method to correct for multiple testing. To compare the effect of sex and time point on the EMMs, we fitted a two-way ANOVA with post hoc tests and correction for multiple testing (Tukey’s method).

The study population consists of 49 smokers (29 males, 20 females) with tobacco dependence and 51 healthy non-smokers as controls (25 males, 26 females). The study population is already described in detail elsewhere [6]. There was no change in craving throughout the study period (males: χ²(2) = 3.586, p = 0.166; females: χ²(2) = 2.026, p = 0.363). For a detailed description of the demographics, see Table 1. After quality control, 30 CpGs for the OT gene, 29 for OXTR, and 10 for the AVP were analyzed (Fig. 1).

In male smokers, two CpG sites of the OT gene had significantly higher methylation, and six of the AVP gene’s significantly lower methylation values than controls. Two CpGs of the OXTR gene were significantly different, with one having significantly higher and one having lower methylation levels. Furthermore, mean methylation of the AVP gene was significantly lower in smokers compared with controls. Regarding 5 CpGs of the AVP gene, there was a significant difference after Bonferroni correction (Table 2; Fig. 2).

In females, the methylation of two OXTR and one AVP CpG were significantly lower in smokers compared with controls. However, this was not the case after the Bonferroni correction (Table 2; Fig. 2). Of note, OXTR CpG p134 and AVP CpG m214 had (before correction) significantly lower methylation levels in both male and female smokers compared with controls. For a graphical overview of the promoter as well as CpG plots for the whole analyzed regions of all three loci, see online supplementary Figure S1 (for all online suppl. material, see https://doi.org/10.1159/000535663).

In our model, CpG position, time point, sex, QSU, and the interaction of time point and gender as well as time point, sex, and QSU, had a significant fixed effect on OT gene methylation. CpG position, time point, and QSU significantly affected AVP gene methylation (Table 3). There was no significant effect of QSU on OXTR mean methylation, in spite of significant effects of CpG position, time point, sex, QSU, and the interaction of time point and gender as well as time point, sex, and QSU on mean methylation (Table 3).

To compare the effect of sex and time point on mean methylation of the respective genes, we fitted a two-way ANOVA. Across all three genes, time point had a significant effect on mean methylation (OT: F(2, 138) = 3.429, p = 0.0352; OXTR: F(2, 139) = 6,527, p = 0.0020; AVP: F(2, 132) = 8.492, p = 0.0003).

Post hoc tests (Tukey) showed that in males, mean methylation (based on EMMs) of the OT gene was significantly higher at t14 compared to t0 and t7 (mean difference (MD) = −0.012, 95% CI: [−0.0226 to −0.001403], p = 0.0222; MD = −0.013, 95% CI: [−0.0236 to −0.002403], p = 0.0118, respectively). In females, mean methylation (based on EMM) of the OT gene was significantly lower at t14 compared to t0 and t7 (MD = 0.03, 95% CI: [0.01746–0.04254], p < 0.0001; MD = 0.024, 95% CI: [0.01146–0.03654], p < 0.0001, respectively). Regarding the OXTR gene, mean methylation was significantly higher at t14 and t0 compared to t7 in males (MD = 0.02, 95% CI: [0.006361–0.03364], p = 0.0020; MD = −0.016, 95% CI: [−0.02952 to −0.002481], p = 0.0158) and significantly lower at t14 compared with t0 in females (MD = 0.017, 95% CI: [0.0005083–0.03349], p = 0.0417). For the AVP gene, while there was no difference in females, in males we found significantly lower values at t0 compared with t7 and t14 (MD = −0.018, 95% CI: [−0.03168 to −0.00432], p = 0.0063; MD = −0.022, 95% CI: [−0.03568 to −0.00832], p = 0.0006).

In this study, we hypothesized that sex-specific changes in methylation levels of the AVP, OT, and OXTR genes would occur during nicotine withdrawal. Indeed, methylation levels of the AVP, OT, and OXTR genes changed significantly during nicotine withdrawal. We have shown sex-specific differences in overall methylation during smoking cessation, a significant decrease in mean methylation of the OT gene, and an increase in the mean methylation of the vasopressin gene across both sexes. Furthermore, we hypothesized sex-specific differences in the mean methylation of AVP, OT, and OXTR in smokers compared with healthy controls. Several CpGs and mean methylation of the AVP gene were methylated differently in smokers compared to controls. Since human and animal studies have gathered evidence that both OT and vasopressin influence drug craving we also hypothesized that changes in mean methylation are associated with tobacco craving. Indeed, although not changing significantly during withdrawal, QSU had a significant effect on OT and AVP methylation. Furthermore, to our knowledge, this is the first study on OT gene methylation in TUD.

At T0, right after they smoked the last cigarette, smokers had significantly higher OT and OXTR and lower AVP mean methylation levels. Our results from blood suggest that AVP expression decreases and OT expression increases throughout cessation. Also, we observe a lowered expression of OXTR receptors, suggesting a downregulation of the pathways targeted by OT that attenuate fear and craving. Although both hormones share similar functions, in animal studies, administered OT and AVP receptor antagonists were shown to reduce drug-seeking behavior [28, 29, 43].

When looking at different sexes, mean AVP promoter methylation was significantly lower in male smokers compared to controls, which was not the case in OXTR and OT methylation. Since decreased mean methylation and methylation of single CpGs can have an impact on gene expression we looked for potentially masked relevant findings and found alterations in the single position we reported. For all genes, sex significantly affected mean methylation during cessation. Furthermore, QSU was associated with OT and AVP methylation levels at single CpGs. Interestingly, the EMMs of OT gene methylation decreased in females and increased in males during cessation. Assuming increased promoter methylation leading to decreased protein expression, this change would translate to decreasing OT levels in females and increasing levels in males. This finding indicates sex-specific effects and differences of the hormone during nicotine withdrawal. For AVP, the EMMs increased during the cessation period in both sexes.

Our results imply that in TUD, AVP, and OT are regulated differently for men and women. This is in line with the evidence provided for sex-specific differences regarding the function of both hormones. Our study found differences between control group males and baseline male smokers in mean methylation of the AVP gene, with lower levels of AVP protein methylation in smokers. Interestingly, studies on alcohol use disorder (AUD) show different results. In one study during withdrawal therapy, we identified significantly elevated promoter-related DNA methylation while also being associated with craving [44]. In another study, we found no difference in mean methylation but at specific CpGs of the AVP gene, with higher and lower methylation values compared with controls [45].

Furthermore, this is in line with studies on reduced AVP levels in alcohol intoxication and during alcohol withdrawal [46, 47] assuming higher methylation levels equal lower protein expression. However, one possible explanation for differences in promoter methylation of the AVP gene in AUD compared with TUD is that ethanol directly affects AVP release due to increased plasma osmolality [48]. Ethanol also directly impacts AVP release in response to the intravenous hypertonic saline infusion [49]. Thus, because of ethanol’s direct effect on plasma levels, specific differences are present in gene methylation of both genes in TUD compared with AUD.

Since gene methylation modulates protein expression [50], this could translate into higher AVP levels in male smokers. Indeed, as a stress hormone, AVP is known to potentiate HPA-axis activation [51], and higher levels could therefore modulate HPA-axis activation and affect craving (see Introduction). In our study, craving, when considered a covariate, influenced the mean methylation of the AVP gene during smoking cessation.

Of note, craving did not change significantly during cessation at the assessed time points. This could be explained by the fact that each smoker filled out the questionnaire right after willingly smoking the last cigarette. The anticipation of never smoking again could therefore increase craving. Furthermore, evidence suggests that the changes in craving during nicotine withdrawal seem to mimic an exponential function [52]. Since we did not register relapse in our study and performed no follow-up, our craving scores are naturally different from those in other comparable studies. Nonetheless, craving still affected mean AVP and OT methylation, which changed during cessation.

For OT, methylation levels increased in female smokers during the cessation period, while mean methylation was partly predicted by craving. Supposing higher methylation leads to less gene transcription, our results suggest that the protein expression of OT increases during the cessation period in females. Due to its effect on brain areas involved in addictive behavior, this could directly affect substance craving. However, since it remains elusive whether peripheral methylation measurements reliably mirror central changes, it is unclear if those changes would coincide (s. limitations).

We did not perform case-control matching and did not include those patients in our study that relapsed during the study period. It would have been interesting to compare those to study abstainers concerning endpoints such as promoter methylation and craving. Furthermore, since we have only assessed mean methylation in healthy controls at one time point, it is possible that the changes observed in smokers could also have been observed in healthy controls. However, the variance of the mean methylation values in controls compared with smokers suggested otherwise. Nevertheless, we cannot exclude a random change in smokers that would also have occurred in healthy controls. Thus, in future studies it is important to assess natural changes in the methylation values of the investigated promoters with respect to possible influencing factors. In this study, we did not include longitudinal measurements in healthy controls since we wanted to focus on changes and sex-specific differences throughout cessation. Furthermore, since promoter methylation was assessed as a possible biomarker, we expected a certain variance in measurements depending on several factors.

Also, we used Sanger sequencing, as this technique allows the analysis of many samples while focusing on specific genes and regions of interest [53]. Modern sequencing methods such as target enrichment sequencing and Oxford Nanopore Technology Sequencing (ONT-S) have obvious methodic advantages compared with Sanger sequencing [54]. Sanger sequencing has some upsides, such as still being the gold standard of comparison as well as the effective investigation of specific genetic loci. Concerning the method’ s accuracy for measuring single promoter regions, our research group showed that Sanger sequencing achieves comparable results compared with (ONT-S) when analyzing promoter methylation [55].

As smoking substantially affects overall global methylation [56] and these influences may persist after cessation [57], lower levels of certain CpGs found in smokers could be due to the general effect of smoking. However, we found increasing and decreasing mean methylation values during the cessation process (e.g., mean methylation (as EMMs) of OT in our calculated model). Thus, our results suggest smoking-induced changes specific to the measured genes and sex rather than an overall change in smoking-related methylation levels.

Also, AVP, OXTR, and OT gene methylation were measured in peripheral blood, not allowing definite conclusions about release and changes in the central nervous system. However, in contrast to its receptor, OT is expressed only in a few peripheral organs and mainly distributed via the bloodstream, thereby creating the need for peripheral OXTR expression but not OT expression [17]. Furthermore, the authors review and summarize several studies that present evidence for different stimuli (e.g., mating, social interaction, stress, hyperosmotic stimulation) that cause an intracerebral release of OT parallel to OT secretion into the blood [17]. Indeed, Jurek and Neumann [17] conclude that glucocorticoids contribute to stress-induced alterations in OT release and OT neural activity. Nonetheless, OT is often released region-dependent and independently of peripheral OT secretion [58]. Therefore, our study’s design does not allow a conclusion about possible differences or similarities of OT and OXTR gene methylation in peripheral blood compared with brain regions. However, in TUD, HPA-axis activation is altered during smoking cessation [59, 60] since tobacco smoke modulates HPA-axis activity [61, 62], while HPA-axis activity is also associated with craving [63]. Thus, with smoking cessation being a potential stressor and having an apparent effect on physiological stress responses, peripheral changes in OT gene methylation during withdrawal could, to some extent, mirror changes in the central nervous system. Furthermore, we did not assess the promoter gene of the respective vasopressin receptors, which could have been interesting since both OT and vasopressin exert effects on those receptors.

As stress induces changes in the composition of leukocytes [64], changes in the composition of blood cells in the context of nicotine withdrawal might relate to the differences in methylation levels [65]. However, since we found sex-specific differences in methylation levels, it is unlikely they can be solely explained by stress-induced withdrawal.

Although several studies suggest an association between OXTR promoter methylation and OXTR expression in the brain [66, 67], Jurek and Neumann point out that it is elusive whether peripheral measurements reliably mirror neural processes. The authors point to one study on the effect of maternal care on OXTR methylation in regional brain tissue and blood cells in rats that found no correlation between striatal or hypothalamic OXTR methylation and OXTR methylation of blood cells and only a modest correlation with hippocampal cells [17, 68]. However, it is crucial to remember that these results cannot be easily extrapolated to our study.

Since we found sex-specific changes in methylation levels during the cessation period, future studies on the therapeutic utility of both targets need to analyze their effect in the context of sex, social factors, and gene regulation. Furthermore, our data imply that intranasal OT’s therapeutic effect and utility could also differ according to sex. Of three studies on the impact of OT on cigarette craving, two reported negative results, with only one comparing sex and finding no differences [34‒36]. Notably, McClure et al. [36] compared the effect in the context of the trier social stress test (TSST), which is known to induce a physiological stress response. Since in the context of the TSST, OT administration, together with social support, was shown to suppress cortisol and subjective responses to psychosocial stress [69], we believe that such factors could also mediate the effect of OT on craving. Exerting sex-specific differences in social pair bonding, we believe OT and AVP’s effect on craving strongly depends on contextual and epigenetic factors, the latter being observed in our study.

In summary, we present evidence for altered gene methylation patterns of the OT, OXTR, and AVP genes during nicotine withdrawal. Our findings align with studies on animals and humans regarding sex-specific differences in the function and regulation of both hormones. Furthermore, they underline the similarities in social behavior and substance use behavior. Our results have implications for future research on the utility of AVP receptor antagonists and intranasal OT or OT analogs on substance craving. Since we found sex-specific differences and changes in methylation levels during the cessation period, future studies on the therapeutic utility of both targets need to analyze their effect in the context of sex, social factors, and gene regulation.

This study protocol was reviewed and approved by the Ethics Committee of Hannover Medical School (MHH), approval number 6695 and adhered to the Declaration of Helsinki. All participants gave written informed consent before entering the study.

The authors have no conflicts of interest to declare.

This study received no funding.

P.J.P. planned the study, performed laboratory analyses, analyzed the data, and wrote the manuscript. S.B. participated in the planning of the study. V.B. performed laboratory analysis. MANM planned the study and recruited the participants. H.F. participated in planning the study and laboratory and statistical analysis. A.G. planned the study, recruited the participants, and analyzed the data. M.R. planned the study, performed laboratory analyses, analyzed the data, and wrote the manuscript. All authors read and approved the final manuscript.

Alexander Glahn and Mathias Rhein contributed equally to this study.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants. Further inquiries can be directed to the corresponding author.