1 Introduction
Since the discovery of Δ9-tetrahydrocannabinol (THC) over 60 years ago, the main psychoactive component in cannabis, and subsequent identification of cannabinoid receptors and their endogenous agonists, the functional role of the endocannabinoid system (ECS) in animals has gained greater recognition (Rezende et al., 2023). Many fatty acid amides and related compounds, structurally similar to the main endocannabinoid, arachidonoyl ethanolamide (AEA), have been identified in the brain, yet with only a handful actually studied (Alharthi et al., 2018). Those that were studied showed diverse effects, including anti-inflammatory and neuroprotective activities (Donvito et al., 2018). Cannabinoid receptors and their endogenous agonists are found throughout the body and are involved in a range of physiological processes including inflammation (Eljaschewitsch et al., 2006), pain (Bruni et al., 2018), immune response (Rahaman and Ganguly, 2021), regulation of sleep (Kesner and Lovinger, 2020) and appetite (Aguilera Vasquez and Nielsen, 2022). The ECS is also important for learning and memory (Morena and Campolongo, 2014), as well as regulation of emotional, motivational, and cognitive functions (Viveros et al., 2007; Magen et al., 2009; Schechter et al., 2013; Morena and Campolongo, 2014; Manduca et al., 2015; Murlanova et al., 2022a). Modulation of the ECS has also been shown to be protective against neurodegenerative disorders (Kogan and Mechoulam, 2007). Further, altered levels of endocannabinnoids have been detected in neuropsychiatric disorders (Schaefer et al., 2014; Romero-Sanchiz et al., 2016; Wingenfeld et al., 2018). Some elements of the ECS have been proposed for use as biomarkers and even as potential targets for the treatment of anxiety and depression (Chadwick et al., 2020; Gallego-Landin et al., 2021; Rana et al., 2021; Bright and Akirav, 2022). Accumulating evidence also suggests a role for the ECS in borderline personality disorder, antisocial behavior, post-traumatic stress disorder, and manic-depressive disorder (Kolla and Mishra, 2018; Wingenfeld et al., 2018; Ferber et al., 2020; Ho and Kolla, 2022; Topuz et al., 2022; Spohrs et al., 2023).
The ECS is involved also in normal emotional processes (Viveros et al., 2007; Morena and Campolongo, 2014; Arndt and de Wit, 2017). AEA and 2-arachidonoylglycerol (2-AG) have been shown to play a role in regulation of social behavior (Wei et al., 2017). Anandamide levels are elevated in rat striatum after meeting an unfamiliar animal, compared with normal levels when a familiar animal was met (Marco et al., 2011). Knockout mutation of fatty acid amide hydrolase (FAAH), responsible for anandamide degradation, enchanced social interactions in mice (Cassano et al., 2011). Anandamide was found to reduce social anxiety (File and Seth, 2003). A “rough-and-tumble” social play in juvenile rats was found to be associated with increased levels of anandamide (Trezza et al., 2012) and 2-AG (Manduca et al., 2015). Mouse knockout mutation of the hydrolytic enzyme monoacylglycerol lipase produced impaired conditioned place preference to social stimuli; and prolonged social contact for 6 h was found to stimulate 2-AG mobilization (Wei et al., 2016). However, the role of other members of the ECS in social behavior remains unclear.
In this work, endocannabinoid brain profiles and responses to a selected endocannabinoid were evaluated in mouse models of social behavior developed through multiple generations of selective breeding based on social interaction food competition dominant-submissive relationship (DSR) paradigm (Feder et al., 2010). These animals exhibit strong and stable characteristics of dominance (Dom strain) or submissiveness (Sub strain), possess inherited stress resilience or vulnerability, respectively (Murlanova et al., 2022b), and differentially respond to psychotropic agents (Nesher et al., 2013).
4 Discussion
In this work for the first time, we explored the endocannabinoidome in the brains of mice with strong and stable characteristics of social dominance and social submissiveness. Our evaluation revealed a number of interesting patterns: i) long-chain polyunsaturated ethanolamides, specifically C20 (eicosa-) and C22 (docosa-), were found to be higher in the brains of the Sub group compared with their Dom counterparts; ii) fatty acid precursors, DA and EPA, of the related ethanolamides, DEA and EPEA, were higher in Sub mouse brains compared to Dom; iii) the transcription levels of genes encoding key enzymes involved in polyunsaturated long-chained fatty acid biosynthesis, FADS2 and ELOVL5, were significantly higher in Sub mice brains; iv) acute DEA administration dose-dependently reduced nociception and induced locomotion in Sub mice only; v) subchronic DEA administration markedly induced dominance behaviors in a phenotype-dependent manner.
The endocannabinoidome encompasses structurally-related lipid mediators (Di Marzo and Wang, 2014; Piscitelli, 2015) involved in diverse biological mechanisms, with broad modulatory effects extending beyond the traditional endocannabinoid signaling pathways mediated by anandamide and 2-AG (Pacher, 2006; Pacher and Kunos, 2013; Pacher et al., 2020). Given that endocannabinoids play a crucial role in brain development, neural plasticity and function (Viveros et al., 2007), it is expected that changes in their concentrations would be associated with specific behavioral manifestations. Currently, the endogenous function of only AEA and 2-AG has been extensively investigated, along with their impacts on behavior (Wei et al., 2017; Kolla and Mishra, 2018). In healthy individuals, for example, circulating levels of AEA have been found to be inversely associated with anxiety (Dlugos et al., 2012) and need for positive reward feedback (Redlich et al., 2021). Here, we observed significant differences in levels of DEA in both male and female mice between Dom and Sub groups. Although DEA was identified about 30 years ago (Hanus et al., 1993), it has yet to be adequately studied. DEA showed affinity to the CB1 receptor at concentrations close to structurally related compound AEA (IC50 190 nM vs. 160 nM, respectively) (Felder et al., 1993) and produced similar responses in a set of behavioral assays for cannabinoid-like effects (Barg et al., 1995). DEA has also been demonstrated to be an agonist of the vanilloid receptor (TRPV1) (Movahed et al., 2005) and, along with related endocannabinoids, is produced in the CNS by neurons, microglial cells, and astrocytes (Walter et al., 2002). The difference between DEA and AEA is in the fatty acid component of the molecule. Endocannabinoids mostly consist of fatty acids coupled to ethanolamide, glycerol, and amino acids. In the main 2 endocannabinoids AEA and 2-AG, the fatty acid is arachidonic acid (AA), while in DEA the fatty acid is docosatetraenoic acid (DA), both of them are polyunsaturated fatty acids (PUFA).
PUFA family (Spector, 1999) is typically classified based on the number and position of their double carbon bonds (carbons number:double bonds number). In addition to AA (20:4), the parent fatty acid of AEA and 2-AG, this family includes other fatty acids such as ɑ-linolenic acid (ALA; 18:3), EPA (20:5), DHA (22:6), linoleic acid (LA; 18:2), DA (22:4), and others. Dysregulation of PUFA homeostasis was found to worsen the course of mental illness and is associated with neurodegenerative processes (Petermann et al., 2022; Stachowicz, 2023). In a recent review of meta-analyses, PUFA supplementation was found to be partially favorable in treatment of anxiety, depression, attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), dementia, mild cognitive impairment, Huntington’s disease, and schizophrenia (Gao et al., 2022). Therefore, we may infer that distinct behavioral characteristics of Dom and Sub mice may partially stem from different brain levels of specific PUFA measured here. The concentrations of PUFA in the body are typically mirrored by the levels of their ethanolamide derivatives. For example, supplementing milk formulas with AA and DHA raised the levels of the corresponding NAEs, AEA, and DHEA, in certain brain regions in piglets (Banni and di Marzo, 2010). In addition, feeding with an AA-rich diet produced increased AEA whole-brain concentrations in mice (Banni and di Marzo, 2010). The differences in the levels of the parental fatty acids (DA and EPA) in brains of Dom and Sub mice hint that imbalance in the regulation of endocannabinoid biosynthesis in these behaviorally-distinct populations may occur upstream of the endpoint ethanolamide derivatives. Our assumption was further confirmed by differential mRNA expression of genes encoding the enzymes (FADS2 and ELOVL5) responsible for PUFA biosynthesis in the brains of Dom and Sub mice. An increased expression of enzymes involved in the desaturation and elongation of PUFA in Sub mice aligns with the results achieved by lipidomic profiling and further provides an explanation for the observed rise in DEA and related metabolites in these mice. It is noteworthy that Fads2 transcription, which is involved in fatty acid desaturation and arachidonic acid synthesis, was upregulated in high-fat diet treated mice exhibiting depressive-like behavior (Yu et al., 2021). ELOngation of Very Long-chain fatty acids (ELOVLs) are enzymes involved in the initial condensation reaction necessary to elongate fatty acids (Sassa and Kihara, 2014). ELOVL5 can condense a wide range of PUFA, including condensation of linoleic acid to produce arachidonic acid, α-linolenic acid to produce eicosapentaenoic acid, and stearidonic acid to produce docosahexaenoic acid (Shikama et al., 2015). Furthermore, ELOVL5 is highly expressed in the central nervous system and mutations in the gene were found to be the cause of spinocerebellar ataxia type 38 (SCA38), a rare autosomal neurological disease characterized by gait abnormality, dysarthria, dysphagia, hyposmia, and peripheral neuropathy (Balbo et al., 2021). While reduction in ELOVL5 expression has been linked to worsened prognosis in breast estrogen receptor-positive cancer patients (Kieu et al., 2022), its role in behavior regulation is still unclear. The observed increased expression of ELOVL5 and FADS2 in Sub mice, together with the elevation of the parental DA and EPA levels, as well as with DEA and related metabolites, suggest that the signaling pathways mediated by these molecules are of importance for social behavior and personality manifestations.
DEA was previously shown to produce classical cannabinoid-like effects at higher concentrations (15–60 mg/kg) (Barg et al., 1995). Indeed, we found that, among tested doses (5, 10, 15 mg/kg), significant antinociceptive effects were observed at a dose of 15 mg/kg, however only in Sub mice. While no anxiolytic-like effects were observed, 15 mg/kg induced locomotory activity in EPM assessments. Interesting, that this phenomenon was also examined in Sub, but not in Dom mice. However, dose as low as 5 mg/kg had no effect in either hot plate or EPM in Sub and Dom mice. Our previous observations demonstrated that Dom and Sub mice differentially respond to psychotropic agents (Nesher et al., 2013; Murlanova et al., 2021). Acute administration of the selective serotonin reuptake inhibitor paroxetine produced pronounced antidepressant-like effects in Sub, while paradoxical (frenetic) activity was observed in Dom (Nesher et al., 2013). Differential responses were also observed with mood stabilizers and addictive compounds (Nesher et al., 2013; Murlanova et al., 2021). We assume, that differential responses of Dom and Sub mice to DEA, observed in this study, may occur due to distinct brain-regions specific patterns of monoamines of these mice (Murlanova et al., 2021). In previous studies, an interaction was found between nociception and negative social interaction (defined as “social pain” (Sturgeon et al., 2020), moreover, they are processed by a shared neural circuitry (Zhang et al., 2019). Many therapeutics, such as cannabinoids, opioids, antidepressants, etc. also share the effect on both, affecting mutually pain perception and mood/behavior (Negri et al., 1996; Greenwald and Stitzer, 2000; Duman et al., 2004; Bomholt et al., 2005; Hache et al., 2011). Indeed, our results show that the treatment with DEA may exert modularity effect on social (hierarchy-based) personality. Research has shown that social hierarchy is linked to personality traits in both animals and humans (Colléter and Brown, 2011; David et al., 2011; Dasgupta et al., 2022; Vékony et al., 2022). For example, dominant individuals tend to be more assertive, confident, and proactive (Blanchard et al., 1988; Salonen et al., 2022), while submissive individuals are often more anxious, passive, and reactive (Catarino et al., 2014; Zaffar and Arshad, 2020). In addition, social hierarchy can shape individual behavior and physiology (Schmid Mast, 2010; Flota-Bañuelos et al., 2019) including stress responses (Knight and Mehta, 2017; Karamihalev et al., 2020), brain activity (Breton et al., 2014; Williamson et al., 2019a), and gene expression (Pohorecky et al., 2004; Horii et al., 2017; Lea et al., 2018; Williamson et al., 2019b).
As DEA is higher in the brain of Sub mice, it would be straightforward to assume that its administration may enhance submissiveness, but this assumption was not confirmed by pharmacological studies, where DEA administration was not able to reduce submissiveness of Sub animals. We may suggest that elevation of DEA in the brain of Sub mice acts as a compensatory mechanism to counterbalance submissiveness-associated features including enhanced stress vulnerability, anxiety, and social deficits (Gross and Pinhasov, 2016; Yanovich et al., 2018). Moreover, in support of this hypothesis, we clearly demonstrate here that DEA further induced dominance levels of selectively-bred dominant mice. This phenomenon should be further evaluated as it suggests that the ECS is important in regulation of social behavior. It is possible that DEA may invoke activation of a signal transduction pathway, distinct from canonical cannabinoid receptor-mediated neurotransmission. This pathway would be activated by DEA at levels that do not produce the classical CB1-mediated psychotropic and antinociceptive effects. This hypothesis is partially supported by gene transcription analysis of Sub and Dom mouse brain, which did not show any significant differences in the expression levels of cannabinoid system related genes Cnr1, Cnr2, Trpv1, Gpr55, and Gpr18.
The diverse endocannabinoid profiles observed in behaviorally-distinct populations of mice reported here, in addition to the phenotype-dependent effect of DEA on nociception, locomotion and modulation of dominance, highlight the role of the ECS in regulating social behavior. By unraveling the intricate relationship between endocannabinoids and social behavior, this research contributes to a deeper understanding of the neurobiological foundations of social behavior and holds promise for advancing our knowledge of psychiatric disorders characterized by social dysfunctions.