The flavonoid chrysin (5,7-dihydroxyflavone) is found in high concentrations in propolis and honeybee, and particularly also in plants such as Passiflora coerulea, Passiflora incarnata and Matricaria chamomilla [1]. It exerts antioxidant, anticancer and anti-inflammatory activity, as well as important pharmacological effects in several disorders involving the central nervous system, such as epilepsy, Parkinson’s disease, multiple sclerosis, Alzheimer’s disease, depression and anxiety [1, 2]. Preclinical studies have demonstrated anxiolytic-like effects of chrysin in different anxiety models in mice and rats. For example, the injection of 1 mg/kg chrysin increases the time spent by male mice [3] and rats [4] on the open arms of the elevated plus maze (EPM). It also increases the time spent by male rats [5] and zebrafish [6] in the illuminated compartment in the light/dark test (LDT). Similarly, the administration of 2 mg/kg chrysin in ovariectomized [7] and cycling [8, 9] female rats produces an anxiolytic-like effect in the EPM and LDT tests. The effect is similar to that produced by anxiolytic drugs as diazepam [8] and some neuro-steroids such as 17$\beta$-estradiol [10, 11], pregnanolone [12] and allopregnanolone [13]. Additionally, intrahippocampal microinjection of 0.5 µg of chrysin or allopregnanolone produces anxiolytic-like effects in female rats during the diestrus phase [9]. These effects are blocked by prior administration of Gamma-Aminobutyric Acid sub-type A (GABA${}_{\text{A}}$) receptor antagonists such as flumazenil, bicuculline and picrotoxin, without affecting locomotor activity [9]. Overall, these findings suggest that chrysin shares the same mechanism of action with anxiolytic drugs and neuro-steroids, both of which involve the Gamma-Aminobutyric Acid (GABA)ergic system.

In experimental anxiety models, daily administration for 14–28 days of 5–20 mg/kg chrysin reduced the total time of immobility in forced swim and tail suspension tests [14, 15]. This was accompanied by higher consumption of sucrose, suggesting an antidepressant-like effect. The effects were also associated with an increased serotonin concentration in the prefrontal cortex, hippocampus and nucleus accumbens, which could be blocked by pretreatment with p-chlorophenylalanine, a selective inhibitor of the tryptophan hydroxylase enzyme involved in serotonin biosynthesis [14, 15]. At the clinical level, abnormalities in serotonin neurotransmission during depression can now be more reliably evaluated due to advances in measurement techniques. However, there is still much to learn about serotonin activity and its potential role in the causation of illness [16]. Novel research approaches go beyond the serotonin hypothesis of depression and have moved to molecular integration involving diverse receptors and biochemical systems, some of which could become novel targets for antidepressant drugs. For example, daily oral administration of 20 mg/kg chrysin or of 10 mg/kg fluoxetine for 28 days increased the levels of the neurotrophins Brain Derived Neurotrophic Factor (BDNF) and Nerve Growth Factor (NGF) compared to controls in the hippocampus and prefrontal cortex of both non-stressed and stressed mice [17]. This was associated with a decrease in depression-like behavior [18, 19], and suggests the potential for developing new drugs from their dipeptide mimetics [20]. Related to the above findings, the flavonoid chrysin, similar to the neuro-steroid allopregnanolone, exerts anxiolytic- and antidepressant-like effects through actions on GABAergic and serotonergic systems. These actions are likely to be dependent on the doses administered and the duration of treatment (21 to 28 days) [21, 22]. Therefore, we hypothesized that low and high doses of chrysin may produce antidepressant-like effects through different mechanisms of action. The goal of this study was to examine the effects of 1 mg/kg and 5 mg/kg chrysin after 0, 1, 14, and 28 days of treatment, and 48 h post-treatment (day 30) in the forced swim test. The results were compared with those of fluoxetine (a serotonergic drug) and allopregnanolone (a GABAergic drug) as pharmacological controls of antidepressant-like activity in adult male rats. In addition, the involvement of the GABA${}_{\text{A}}$ receptor in the long-term action of chrysin was also explored.

The present study compared the effects of low and high doses of chrysin on depression-like behavior in adult male rats with those of fluoxetine and allopregnanolone. In addition, we investigated the mechanism involving the GABA${}_{\text{A}}$ receptor in the antidepressant-like effects produced by chronic administration of chrysin. This was studied by antagonizing the GABA${}_{\text{A}}$ receptor with picrotoxin. We evaluated the anti-immobility effects of low and high doses of chrysin at different times during the chronic treatment. Picrotoxin attenuated the latency to first immobility at high dose chrysin, and the total time of immobility at low dose chrysin.

The evaluation of crossings with the LAT allows the detection of changes in general locomotor activity caused by treatment. It is well known that stimulants of the central nervous system reduce immobility times [37], but the LAT can discriminate motor effects from motivational effects. This can occur with a substance that has antidepressant activity, which reduces immobility in the FST without significant changes in locomotion [26, 28, 35]. For this reason, we and others run the LAT before the FST [45]. When longitudinal studies are carried out using LAT, reductions in crossings may occur even when anti-immobility effects are exerted by the substance with anti-despair properties [27]. Therefore, in the present study it is possible to disregard any motor influence on behavior detected by the FST. This demonstrates that chrysin, similar to allopregnanolone and fluoxetine, produces an antidepressant-like effect associated with motivation of the animals, rather than with a motor component.

Also in relation to the LAT, the time spent vertically exploring the environment (rearing) did not change with chrysin treatment, but differences were observed in grooming. Rearing and grooming are behavioral indicators of the emotional state of rats when exposed to novel environments [46]. Rearing is a measure of the active exploration carried out by the rat on the environment in which it is located. This can vary depending on the experimental situation. Some studies have reported that rearing increases after acute administration of drugs such as benzodiazepines, e.g., diazepam [32, 47, 48], or substances with anxiolytic potential such as neuro-steroids, e.g., progesterone or allopregnanolone, or flavonoids such as chrysin [7, 28]. However, other studies have reported no changes with the same treatments [4, 6, 9, 49]. There are no reports of increases in rearing after long-term administration of the abovementioned treatments. Indeed, no changes in this exploratory behavior are often reported [4, 27, 50], as found in the present study.

Grooming behavior is an indicator of animal motivation [51]. It may increase in mildly stressful situations, but is drastically reduced under conditions of severe stress [27, 52, 53]. Reduction of grooming time is prevented by anxiolytics, antidepressants and neuro-steroids, which re-establish this behavior to levels similar to those found in undisturbed animals [36, 53, 54, 55]. As expected, our results showed that grooming decreased during the course of treatment compared to the baseline value in the VEH group. However, all treatments maintained a value similar to baseline from day 1 of administration, except for fluoxetine. This treatment showed a “U-shaped” effect, with an increase from day 14, and similar values to the baseline at day 28. Interestingly, 48 hours after the last administration, only the C5 and FLX-treated groups maintained grooming behavior at levels similar to baseline. This may reflect increased motivation consistent with reduced total immobility time, and is possibly also associated with changes in brain plasticity produced by prolonged treatment with antidepressants [53]. It is worth noting this effect was not observed with low dose chrysin, thus supporting the hypothesis that low doses exert “anti-stress” actions through effects on the GABA${}_{\text{A}}$ receptor. This was supported by the results with picrotoxin, which canceled the pro-grooming effect observed with low dose chrysin, but not with high dose chrysin, allopregnanolone, or fluoxetine. This also supports the idea that chrysin exerts GABA-mediated effects in the short term, but that its effects in the long term involve other neurotransmission systems such as serotonin, as reported previously [28]. Therefore, the permanent effect of allopregnanolone and chrysin over picrotoxin during chronic administration may involve the actions of combined neurotransmission systems beyond GABA. Long-term treatment with allopregnanolone and chrysin may also produce significant changes in the BDNF, serotonin and dopamine neurotransmission systems [17, 56], since allopregnanolone maintained the reduced immobility 48 h after withdrawal. This may be due to plasticity after 28 days of administration.

In the FST, low and high doses of chrysin produced antidepressant-like effects similar to fluoxetine and allopregnanolone. Picrotoxin prevented the effects of low dose chrysin, but not of the higher dose, suggesting a differential mechanism for chrysin action depending of the dose. The mechanism for higher doses could involve the activation of other neurotransmitter systems (e.g., serotonin) and neurotrophic factors, as identified in previous studies with chronic administration of chrysin [17, 29, 56]. The FST has been validated as a behavioral model and reveals substances with antidepressant activity because they produce longer latency and shorter immobility time, similar to clinically effective antidepressant drugs [26, 37, 57, 58, 59]. However, the terms “depression-like” and “despair” may be incorrect and FST behavior in general may represent a copying behavior [60, 61]. Neural networks involved in the stress sessions of FST are heterogeneously complex and are linked by relevant factors to depression and to other conditions related to stress. The present work used naive rats to study the response to chrysin, and the findings with allopregnanolone and fluoxetine support the robust antidepressant-like effects observed with chrysin. In particular, 14 days of administration with high dose chrysin increased the latency to the first immobility, similar to the antidepressant drug fluoxetine and the neuro-steroid allopregnanolone. Moreover, the effect was maintained 48 h after ending the treatment, suggesting there was a long-term increase in motivation associated with changes in brain neuroplasticity, as occurs after treatment with antidepressant substances [29, 62].

As expected, the total immobility time decreased with both low and high doses of chrysin, but important differences were noted. Low dose chrysin produced a rapid (day 1) anti-immobility effect, but was no longer effective 48 h after the last administration. This finding suggests that 1 mg/kg chrysin may act through ionotropic receptors to exert a reduction in immobility that is unrelated to antidepressant-like effects. Instead, it may be an “anti-stress” effect that allows the rat to quickly deal with the urgent situation, as previously reported for allopregnanolone [30]. Low dose chrysin may act in a manner similar to the acute administration of neuro-steroids such as progesterone or allopregnanolone. These reduce immobility through actions on ionotropic receptors, exhibiting rapid (30 min) but short-lasting (6 h) activity after injection [53]. In contrast, chronic application of 5 mg/kg chrysin may instead involve neuronal plasticity changes that take time to establish, as reported in previous studies [63, 64]. This flavonoid continues to produce anti-immobility effects 48 h after the suspension of 28-day treatment. Such long-term effects are similar to those seen with antidepressant drugs such as fluoxetine, which require 2–3 weeks for their effects to occur [26], as seen here with the FLX group. This effect could be related to chrysin-induced changes in the level of metabotropic serotonergic receptors such as 5-hydroxytryptamine subtype 1A (5-HT${}_{\text{1A}}$) receptor [29], and in the production of serotonin. The establishment of long-term therapeutic effects could also involve changes in neuronal plasticity associated with the production of neurotrophic factors such as BDNF and NGF in structures involved with depression [65], such as the prefrontal cortex and the hippocampus [33, 66, 67]. Thus, it appears likely that 5 mg/kg chrysin induces neuronal plastic changes of long duration, as reported by other authors with fluoxetine [68, 69]. This allows animals to maintain reduced immobility, even after the suspension of treatment. Pretreatment with picrotoxin, a non-competitive antagonist that blocks the GABA${}_{\text{A}}$ receptor chloride channel, inhibits the reduction in immobility produced by 1 mg/kg chrysin and returns the values to control levels. This effect was not observed with high dose chrysin, allopregnanolone or fluoxetine, presumably because they involve the actions of combined neurotransmission systems beyond GABA. These treatments maintained the reduced immobility 48 h after withdrawal and may be involved in more permanent effects, such as those observed with the antidepressant phytomedicine Hiperikan® [70]. In summary, we propose that chrysin at higher doses may activate other neurotransmission systems (serotonergic, dopaminergic and norepinephrinergic) to increase the production of neurotrophic factors such as BDNF and NGF [14, 15, 29, 71]. Additionally, chrysin exerts antioxidant and anti-inflammatory effects on the central nervous system (CNS), which may also contribute antidepressant effects [33], as occurs with conventional antidepressant drugs. This may explain why the effects of chrysin were not blocked by the GABA${}_{\text{A}}$ receptor antagonist and highlights a crucial point in psychopharmacology. Substances such as many of the flavonoids, especially those found naturally and with pleiotropic effects, often do not have a single, clearly defined mechanism of action. Instead, their effects are the result of a complex interplay between multiple physiological systems. Therefore, single antagonism of the GABA${}_{\text{A}}$ receptor is not sufficient to block the effects of allopregnanolone and 5 mg/kg chrysin on immobility.

Many flavonoids also act in a biphasic manner by enhancing the actions of GABA at low concentrations, and inhibiting it at high concentrations [72]. In particular, chrysin has agonist actions on the GABA receptor, producing anxiolytic effects similar to diazepam, that disappear with the administration of GABAergic antagonists [3]. In the present study, the absence of an antidepressant-like effect with high dose chrysin on day 1 may be due to the overstimulation of GABA${}_{\text{A}}$ receptors. An excessive amount of drug or agonist substance may lead to overstimulation of the GABA${}_{\text{A}}$ receptor, resulting in decreased ability of the brain to regulate neural activity. A similar phenomenon has been reported for other substances, such as benzodiazepines [54, 73], neuro-steroids and flavonoids [72, 74]. Furthermore, most antidepressants including fluoxetine exhibit an initial delay of two or more weeks before establishing a therapeutic effect [75, 76]. At the preclinical level in rats, fluoxetine requires more time (14 to 21 days) to reduce immobility in the FST and to increase latency to the first immobility [26, 43]. This is similar to the observed delay in humans. The delay in therapeutic effects of antidepressant drugs has been related to synaptic and neuronal plasticity, including changes in brain-derived neurotrophic factor and in the number of dendrites or synaptic receptors [29, 62].

A limitation of the present study was that antagonism through a non-competitive mechanism using picrotoxin was only measured after 28 days of treatment. It would also be preferable to study the antagonism at days 1 and 14 of treatment. In addition, it would be preferable to antagonize the recognition site in the GABA${}_{\text{A}}$ receptor using the competitive antagonist bicuculline. With regard to the differential effects of low and high doses of chrysin, further work should investigate selective antagonism of the benzodiazepine recognition site in the GABA${}_{\text{A}}$ receptor using flumazenil. Futures studies should also investigate female rats in order to explore possible differences in the anti-immobility effects of low and high doses of chrysin according to the estrous phase. It was previously shown in ovariectomized rats that bicuculline blocked the anti-immobility effect of 1 mg/kg chrysin [26]. There is increasing interest in the use of chrysin as a potential therapeutic application for anxiety disorders, with many studies now being published on its anxiolytic effects. However, there has been less research on the antidepressant-like effects of chrysin and on its combined mechanism of action in the central nervous system. These areas clearly require further study.

The anti-immobility effects of low dose chrysin are time-dependent and different to those of high dose chrysin. Low dose chrysin produced a rapid onset, anti-immobility effect in the FST. In contrast, high dose chrysin produced a delayed but sustained anti-immobility effect during chronic treatment and 48 h after withdrawal, similar to the antidepressant fluoxetine. The mechanism underlying the antidepressant-like effects of low dose chrysin is GABAergic, while the antidepressant-like effects of high dose chrysin may be via other neurotransmission and neuromodulation systems likely related to serotonergic actions. Possible applications of chrysin for the treatment of comorbid anxiety-depression will first require the completion of relevant preclinical and clinical studies with this flavonoid.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

JFRL designed the research study. AKLV and OJOV performed the research. BBM provided help and advice on technical elements. GGR and JFRL analyzed the data. GGR, BBM, AKLV, OJOV and JFRL wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The experimental protocol received authorization from the Committee for the Use and Care of Laboratory Animals of the Biomedical Research Center from Universidad Veracruzana with the approval number CLCIB2023/2.

To SIREI project no. DGI: 26650202346 to J.F.R-L. and CONAHCYT (CVU 121508, 285449, 121609, 863438, 635392).

This research was funded by SEP-PROMEP (grant no. 103.5/05/1955, UVER-PTC-155), High Quality Postgraduate Academic Strengthening Program (I010/458/2013, C-703/2013), CONAHCYT (grant 1840).

The authors declare no conflict of interest.