Some anxiety disorders affect a higher percentage of women than men worldwide (Piggott et al., 2019), which have been associated with hormonal, biochemical, psychological and sociocultural factors (Velasco et al., 2019). Fluctuations in plasma and brain concentrations of oestradiol, progesterone, and some of their metabolites during the ovarian cycle are related to females’ higher vulnerability to stressors that may trigger anxiety (Rodríguez-Landa et al., 2015; Stanikova et al., 2019). At the preclinical level, anxiety-like behaviour is higher during metoestrus-dioestrus, characterised by low concentrations of steroid hormones than proestrus-oestrus which, in contrast, presents high hormonal concentrations (Marcondes et al., 2001). This naturally occurring anxiety-like behaviour prompted the evaluation of substances that relieve anxiety associated with metestrus-dioestrus (Ravenelle et al., 2018; García-Ríos et al., 2019; Rodríguez-Landa et al., 2021), which is a physiological state similar to the premenstrual period in women (Lovick, 2008). In this way, some flavonoids exert anxiolytic-like effects in rodents. The systemic administration of chrysin (5,7-dihydroxyflavone) produces anxiolytic-like effects in male rats (Salgueiro et al., 1997; Zanoli et al., 2000), ovariectomised and cycling female rats (Rodríguez-Landa et al., 2019, 2021). Such effects are blocked by gamma-aminobutyric acid-A (GABA_A) receptor complex antagonists, suggesting that chrysin shares similar mechanisms of action with anxiolytic drugs and neurosteroids (Crocetti and Guerrini, 2020). This flavonoid exerts neurochemical and neurotrophic effects in brain structures involved in the neurobiology of anxiety and depression, such as the prefrontal cortex and limbic cortex (Filho et al., 2015; Germán-Ponciano et al., 2021).

The GABA_A/benzodiazepine receptor complex is widely distributed in brain structures involved in emotional processing (Pirker et al., 2000; Engin and Treit, 2007b; Albrecht et al., 2021) such as the hippocampus, which is a target of anxiolytic drugs (Farajdokht et al., 2020). Intrahippocampal microinjections of GABAergic compounds produce anxiolytic- and antidepressant-like effects that are blocked by GABA_A receptor antagonists (Nin et al., 2008; Rezvanfard et al., 2009; Rodríguez-Landa et al., 2009; Farajdokht et al., 2020). However, it is unknown if the hippocampus modulates anxiolytic-like effects produced by the flavonoid chrysin in cycling female rats via GABAergic actions. Previously, we have demonstrated that specific doses of picrotoxin, bicuculline, and flumazenil (all injected i.p.) antagonise the anxiolytic-like effects of GABAergic compounds without exerting intrinsic activity (Rodríguez-Landa et al., 2013, 2019); which permits their use as a pharmacological tool to explore the potential mechanism of action of GABAergic compounds with anxiolytic effects. Therefore, we extend our knowledge of the participation of the hippocampus, where the Cornus Ammonis 1 (CA1) area plays a critical role in anxiety-like behaviour through ligands that act on the GABA_A receptor (Rezayat et al., 2005; Jimenez et al., 2018).

We hypothesised that a microinjection of chrysin in the CA1 area of the dorsal hippocampus reduces anxiety-like behaviour in dioestrus through the GABA_A/benzodiazepine receptor complex. GABA_A receptor antagonists such as picrotoxin, bicuculline, and flumazenil were used to explore the participation of chloride ion channels, the recognition site for GABA and the benzodiazepine binding site, respectively, in the effects of chrysin on anxiety-like behaviour. The effects of chrysin were compared with those of similar doses of the neurosteroid allopregnanolone, which was used as a reference anxiolytic GABAergic drug (Bitran et al., 1991).

The present study explored the effects of the dorsal hippocampus in chrysin and allopregnanolone on anxiety-like behaviour associated with low (dioestrus phase) and high (proestrus phase) concentrations of ovarian hormones and the involvement of the GABA_A/benzodiazepine receptor complex in such effects. Thus, this is the first report to show that chrysin, similar to allopregnanolone, regulates anxiety-like behaviour by targeting a brain structure involved in the physiopathology of anxiety (i.e., the hippocampus), which depends on the endocrine state associated with the ovarian cycle. This study contributes to the delineation of the neuroanatomical substrate involved in the anxiolytic-like effects of the flavonoid chrysin.

The EPM is a widely validated unconditioned model of anxiety that is used to study the potential anxiogenic and anxiolytic effects of drugs, their mechanisms of action and their brain targets. In this model, higher anxiety-like behaviour was observed when rats spent less time in the open arms, whereas lower anxiety-like behaviour was observed when rats spent more time in the open arms (Korte and de Boer, 2003). Interestingly, rats are treated with specific doses of anxiolytic drugs that activate the GABAergic system and spend more time in the open arms of the EPM without showing changes in their general locomotor activity (Rodgers et al., 1997). In the present study, 0.25 μg of chrysin and allopregnanolone did not modify the levels of anxiety that were observed in rats in dioestrus and oestrus. Increases in the number of entries into the open arms and time spent on the open arms were brought about by the microinjection of 0.5 μg of chrysin and allopregnanolone in the dorsal hippocampus in dioestrus rats. This behavioural change suggests an anxiolytic-like effect that is similar to that of anxiolytic drugs such as benzodiazepines and their derivatives (Rezvanfard et al., 2009; Ravenelle et al., 2018; Rombolà et al., 2019, 2020).

The anxiolytic effects of chrysin and allopregnanolone in dioestrus rats resemble anxiolysis, which naturally occurs in proestrus. In contrast, microinjections of 0.5 and 1 μg of chrysin and allopregnanolone in proestrus rats blocked the naturally occurring anxiolysis associated with this ovarian cycle phase. These results suggest a switch in the effects of chrysin and allopregnanolone when relatively high doses were microinjected into the dorsal hippocampus; however, they were not associated with sedative or stimulant effects on locomotor activity. One possible explanation for these findings in the EPM is the high plasticity of the GABA_A/benzodiazepine receptor complex in the presence of these ligands. In vitro studies have shown that high concentrations of GABAergic agonists may alter the GABA_A subunit configuration to elicit desensitization and even lead to the inactivation of this receptor subtype (Mathers, 1991; Krall et al., 2015). Both chrysin and allopregnanolone target GABA_A receptors, and steroid hormone concentrations are high in proestrus (MacKenzie and Maguire, 2014). Thus, the combination of high concentrations of endogenous and exogenous ligands (i.e., chrysin plus endogenous steroids or allopregnanolone plus endogenous steroids) may reduce GABA_A receptor sensitivity, interfere with naturally occurring proestrus-dependent anxiolysis and block the facilitation of GABA_A-mediated inhibition by exogenous ligands.

The effect of the highest dose of chrysin or allopregnanolone (1 μg) microinjected into the dorsal hippocampus of dioestrus rats supports the hypothesis of the lower sensitivity of the GABA_A receptor, showing an inverted-U effect and returning anxiety to a level that was observed in vehicle-treated rats in dioestrus. Interestingly, bilateral intrahippocampal microinjections of 0.5 μg of the GABA_A receptor agonist muscimol (Zhang et al., 2014) and 14 and 20 μg allopregnanolone (Engin and Treit, 2007a), produced anxiety-like behaviour in the EPM in male rats, whereas bilateral microinjections of the lower dose of 0.2 μg/side allopregnanolone in the dorsal intrahippocampus produced anxiolytic-like effects (Mòdol et al., 2011). Altogether, evidence suggests a dose-dependent modulation of anxiolytic-like effects of drugs targeting the GABA_A receptor. Another alternative explanation is that elevated levels of oestradiol are present during early proestrus, which does not occur during early dioestrus. Such differences in the endocrine milieu may be critical in preventing the synergistic effects of chrysin and allopregnanolone with sex hormones. Ovariectomy causes extremely low levels of sex hormones and elicits a progressive increase in [³H] flunitrazepam-specific binding that is linked to GABA_A receptors, and this increase is reduced by oestradiol treatment (Bossé and Di Paolo, 1996). In cycling female rats, progesterone levels, but not oestradiol levels, positively correlate with bicuculline-induced seizure thresholds (Wilson, 1992), supporting the protective role of progesterone and possibly its metabolites, on the behavioural effects elicited by drugs targeting the GABA_A receptor. This idea deserves to be evaluated in specific experiments in females treated with chrysin or allopregnanolone and coursing with different oestradiol and/or progesterone levels. The present results suggest that chrysin exerts actions that are similar to the endogenous neurosteroid allopregnanolone, modulating anxiety-like behaviour depending on the dose that is microinjected in the dorsal hippocampus and depending on the endocrine state of females. Higher doses of chrysin, which may induce adverse effects, remain unexplored. Further research should be conducted to evaluate the potential negative or toxic effects.

To avoid false positives or false negatives in anxiety models, possible non-specific motor effects of drugs or surgical manipulations need to be discarded. Closed-arm entries were introduced as a measure of motor activity in the EPM (Rodgers and Johnson, 1995). However, agreement as to the “pure” indicator of locomotor activity index of this variables in the EPM remains ambiguous because it is considered more as an index of protected exploration (Bourin, 2015). Recent studies have included the LAT as a complementary measure of a purer locomotor activity to either discard or identify motor changes associated with experimental manipulations (Rodríguez-Landa et al., 2009; Hernández-López et al., 2017). In the present study, none of the treatments affected the closed-arm entries in the EPM or the motor activity in the LAT (i.e., crossings, rearing, and grooming), thus demonstrating the specificity of anxiolytic-like effects in dioestrus and anxiogenic-like effects in proestrus when chrysin and allopregnanolone were microinjected into the hippocampus. These effects are consistent with those in previous studies that reported no significant effects of anxiolytic drugs, the ovarian cycle, or stereotactic surgery on locomotor activity (Rodríguez-Landa et al., 2009, 2013, 2019; Carro-Juárez et al., 2012; Hernández-López et al., 2017; García-Ríos et al., 2019). On the other hand, intraperitoneal injections of chrysin (2 mg/kg) in dioestrus rats increased grooming behaviour in the LAT (Rodríguez-Landa et al., 2021). This difference with the present study could be associated with the different routes of administration and specific experimental conditions, suggesting that the dorsal hippocampus does not directly participate in the regulation of grooming behaviour elicited by chrysin.

The present findings support the hypothesis that chrysin interacts with the GABAergic system in the dorsal hippocampus, which highly expresses GABA_A receptors and participates in the regulation of emotional processing, anxiety and neuropharmacological effects of anxiolytic drugs (Farajdokht et al., 2020). Studies have explored the participation of the GABA_A/benzodiazepine receptor complex in the actions of anxiolytic drugs using specific antagonists. Picrotoxin is a non-competitive GABA_A receptor antagonist that blocks the flux of chloride ions through the receptor channel, decreasing GABAergic function (Olsen, 2006). Through this mechanism, picrotoxin blocks the effects of drugs, such as diazepam, gabapentin and allopregnanolone (Kumar and Goyal, 2007; Kumar et al., 2009; Carro-Juárez et al., 2012). Bicuculline is a competitive antagonist of the gamma-aminobutyric acid recognition site in the GABA_A receptor. It blocks the activation of this receptor by flavonoids and neurosteroids (Cueto-Ecobedo et al., 2020). Flumazenil is a selective antagonist of the benzodiazepine recognition site of the GABA_A receptor and interferes with the anxiolytic effects of benzodiazepines and allopregnanolone (Fernández-Guasti and Picazo, 1995; Farajdokht et al., 2020). In the present study, the systemic administration of picrotoxin, bicuculline, and flumazenil completely prevented the anxiolytic-like effects of microinjections of 0.5 μg of chrysin and allopregnanolone in the dorsal hippocampus in dioestrus rats, underscoring the critical role of the GABA_A-benzodiazepine receptor complex in anxiolysis mediated by both drugs. Additionally, all GABAergic antagonists elicited a partial reversal of the anxiogenic-like effect of chrysin and allopregnanolone, which was observed in proestrus rats. Such effects may be explained by the saturation of the GABA_A receptor by its ligands, leading to changes in its functionality. The present results suggest that other endogenous substances that are associated with the endocrine milieu during proestrus may directly impact the effects of chrysin and allopregnanolone on GABA_A receptor function. Altogether, these findings indicate that the GABA_A/benzodiazepine receptor complex in the dorsal hippocampus participates in the neuropharmacological effects of chrysin and allopregnanolone on anxiety-like behaviour in female rats. The present findings should prompt further clinical research on chrysin, similar to the neurosteroid allopregnanolone, on its ability to ameliorate neuropsychiatric symptoms in women with low concentrations of steroid hormones (Paul et al., 2020; Pinna, 2020a,b).

The present study had several limitations. The groups in Experiment 2 were combined with groups that received microinjections of a vehicle and 0.5 μg of chrysin and allopregnanolone from Experiment 1. This decision was based on the 3R principles (Russell et al., 2005) to reduce the number of experimental animals. The statistical results validated this decision because between-group comparisons yielded significant differences (p < 0.05, p < 0.001), providing satisfactory statistical support for the neurobehavioural findings. Supporting this decision, we used antagonist doses previously reported in the literature by our laboratory group to avoid replicated groups. Such doses lack effects per se in the EPM and LAT (Rodríguez-Landa et al., 2013, 2019). Furthermore, future studies could be conducted to evaluate the microinjection of GABAergic antagonists in the hippocampus to identify possible local interactions among the GABA_A/benzodiazepine receptor complex, chrysin, or allopregnanolone. Finally, another potential limitation could be the use of unilateral vs. bilateral implants into the dorsal hippocampus, which was decided to avoid a potentially higher lesion in the brain, considering the dimensions of the implanted guide cannula. In support of our design, previous studies have reported anxiolytic- and antidepressant-like effects after the unilateral microinjection of substance P (Carvalho et al., 2008), progesterone (Estrada-Camarena et al., 2002), and allopregnanolone (Rodríguez-Landa et al., 2005, 2009), with no significant differences between the hippocampus side. In the present study, we microinjected chrysin and allopregnanolone into the left (dorsal hippocampus) hemisphere to observe clear anxiolytic-like effects.

The results of this study confirm that chrysin affects anxiety-like behaviour in females and that it directly affects the dorsal hippocampus, a structure with a key role in affective and emotional behaviour modulation. GABA_A receptors in the dorsal hippocampus seem to have different degrees of participation in the effects of chrysin on anxiety, which is modulated by the endocrine milieu in females, similar to allopregnanolone. Specifically, hormone conditions in dioestrus favor the anxiolysis elicited by chrysin, in interaction with the different binding sites of GABA_A receptors. In contrast, the proestrus endocrine milieu leads to anxiogenic effects of this flavonoid, which are partially mediated by the GABA_A/benzodiazepine receptor complex. Altogether, our findings contribute to the knowledge of the neuroanatomical and neuropharmacological bases of the effects of chrysin on anxiety-like behaviour that are associated with hormonal fluctuations during the ovarian cycle phases.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by the Internal Committee for the Care and Use of Laboratory Animals (CICUAL-UV).

JFR-L, DS, and FH-L: conceptualisation and methodology. FH-L, LM-M, and ER-D: methodology and formal analysis. JFR-L, BB-M, and DS: data curation. JFR-L, LM-M, and DS: writing manuscript. All authors reviewed, discussed, and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.