A Coumarin‐ and Phthalide‐Rich Fraction From Baichuan Baile Formula Alleviates Reserpine‐Induced Depression‐Like Behaviors by Modulating the Monoaminergic System and Gut Microbiota

ABSTRACT

Baichuan Baile (BCBL) is a novel herbal formula composed of three herbs and a food additive with proven antidepressant‐like effects. However, its unclear bioactive constituents hinder quality control. This study evaluated BCBL‐CP, a fraction extracted via ethanol precipitation and macroporous resin chromatography, mainly comprising coumarins (5‐hydroxyxanthotoxin, oxypeucedanin hydrate, byakangelicin) and phthalides (senkyunolides H and I). We rapidly assessed BCBL‐CP's efficacy using behavioral despair and 5‐hydroxytryptophan‐induced head‐twitch tests. Subsequently, therapeutic effects on a reserpine‐induced depression‐like mouse model were evaluated through behavioral assessments. Its effects on the monoaminergic system, cerebral pathology, brain‐derived neurotrophic factor (BDNF), and gut microbiota were detected through enzyme‐linked immunosorbent assay, hematoxylin and eosin staining, immunofluorescence, and 16S rDNA sequencing. Results showed BCBL‐CP exhibited significant antidepressant‐like activity in both acute behavioral screening and chronic reserpine‐induced depression studies. Mechanistically, BCBL‐CP selectively increased prefrontal norepinephrine (p < 0.05), attenuated neuronal damage (p < 0.001), and showed a trend toward BDNF upregulation (p = 0.083). Uniquely, it restored gut microbiota α‐diversity (p < 0.01), normalized β‐diversity, and modulated eight key bacterial taxa (p < 0.01 or p < 0.05). These findings confirm that these five compounds constitute the major characteristic bioactive constituents of BCBL, providing quality‐control markers for developing this prospective antidepressant.

The coumarin‐ and phthalide‐rich fraction (BCBL‐CP) from Baichuan Baile formula (BCBL) effectively alleviates depression‐like behaviors in mice. This botanical agent acts through a multi‐target mechanism that selectively enhances prefrontal norepinephrine, prevents hippocampal neuronal damage, and restores gut microbiota homeostasis. These findings provide quality‐control markers for developing BCBL as a novel antidepressant.

1. Introduction

The World Health Organization forecasts that by 2030, depression will top the global disease‐burden list. It already afflicts more than 340 million people and generates staggering economic losses [1, 2]. Current first‐line therapeutics, including selective norepinephrine (NE) reuptake inhibitors, serotonin and NE reuptake inhibitors, monoamine oxidase inhibitors, and tricyclic antidepressants, primarily enhance monoaminergic signaling, yet approximately one‐third of patients fail to achieve remission and many experience adverse events that compromise adherence [3]. The delayed onset of action (weeks) and limited efficacy against anhedonia and cognitive deficits further underscore the need for novel therapeutic agents in clinical practice [4]. The development of functional foods or standardized botanical drugs from traditional Chinese medicinal (TCM) materials, particularly those with food‐drug homology or high safety profiles, is now regarded as a strategic pillar for depression prevention and therapy [2]. These products offer three key advantages. First, they are generally recognized as safe at culinary or nutraceutical doses, allowing for long‐term or adolescent use. Second, their multi‐component nature can simultaneously modulate multiple targets, thereby aligning with the multifactorial pathogenesis of major depressive disorder. Third, their long history of dietary use provides real‐world evidence that facilitates regulatory acceptance and enhances patient adherence [5, 6].

Major depressive disorder is a multifactorial neuropsychiatric disease with a complex and interconnected pathophysiological network, which extends far beyond the classical monoamine hypothesis. The core pathological mechanisms of depression include: (1) Monoaminergic dysregulation: Deficiency of monoamine neurotransmitters (serotonin, NE) in the prefrontal cortex and hippocampus is the most well‐established pathological basis of depression, which is also the core target of almost all first‐line antidepressants [7]. (2) Neuroinflammation and oxidative stress: Sustained stress‐induced microglial activation, excessive pro‐inflammatory cytokine release, and oxidative damage are key drivers of neuronal impairment in depression [8]. (3) Neurotrophic factor alterations: Downregulation of hippocampal brain‐derived neurotrophic factor (BDNF), a key mediator of neuronal survival and synaptic plasticity, leads to hippocampal atrophy and synaptic dysfunction, underlying the progression of core depressive symptoms [9]. (4) Microbiota–gut–brain axis disturbances: Gut dysbiosis can further aggravate central monoaminergic dysfunction, neuroinflammation, and neurotrophic deficiency via the microbiota–gut–brain axis; such dysbiosis has been recognized as a critical peripheral driver of depression [10, 11]. Therefore, evaluating the regulatory effects of candidate antidepressants across these multiple pathological dimensions is essential to comprehensively clarify their therapeutic mechanisms, serving as the core design of this study.

Baichuan Baile (BCBL) is a novel experimental formula based on TCM principles and modern nutritional assessment, comprising Angelica dahurica (Fisch. ex Hoffm.) Benth. et Hook. f. (A. dahurica), Ligusticum chuanxiong Hort. (L. chuanxiong), Schisandra chinensis (Turcz.) Baill. (S. chinensis), and l‐menthol. Our previous work demonstrated that BCBL exerted significant antidepressant‐like effects in multiple mouse models, including behavioral despair mice, chronic restraint stress and chronic noise exposure‐induced depression‐like mice, and chronic reserpine‐induced depression‐like mice. Drug–drug interaction tests, high‐performance liquid chromatography with electrochemical detection, and enzyme‐linked immunosorbent assay (ELISA) analyses validated that BCBL's antidepressant‐like effects were partially mediated through upregulation of the serotoninergic system [1]. However, the BCBL extract is a complex matrix containing thousands of chemical constituents, and the specific chemical scaffolds responsible for its antidepressant‐like activity remain incompletely characterized. This “material basis” gap not only hampers batch‐to‐batch quality control (a prerequisite for botanical drug approval) but also impedes mechanism‐oriented optimization and intellectual‐property positioning [12, 13, 14].

In the present study, an integrated phytochemical and pharmacological strategy was employed to demonstrate the chemical basis underlying the antidepressant‐like effects of BCBL. First, a fraction enriched in coumarins and phthalides, designated BCBL‐CP, was prepared from BCBL via ethanol precipitation and macroporous resin chromatography, and its chemical constituents were characterized by ultra‐performance liquid chromatography coupled with quadrupole‐orbitrap tandem mass spectrometry (UPLC‐Q‐Orbitrap‐MS/MS) and high‐performance liquid chromatography with photodiode array detection (HPLC‐PDA) [15]. Second, the tail suspension test (TST), forced swimming test (FST), and 5‐hydroxytryptophan (5‐HTP)‐induced head‐twitch test were conducted to rapidly assess the antidepressant‐like effects of acute pre‐treatment with BCBL‐CP in mice [5]. Acute models (TST, FST) are excellent for rapid screening, but they do not mimic the pathophysiology of clinical depression. Therefore, to rigorously build upon our acute screening results, we advanced to a chronic reserpine‐induced depression model, which was established based on prior studies to evaluate the antidepressant‐like effects of chronic BCBL‐CP co‐treatment [1]. ELISA, hematoxylin and eosin (HE) staining, immunofluorescence (IF) staining, and 16S rDNA (V3 + V4 regions) gene sequencing were conducted to elucidate the underlying mechanisms [16]. The overarching goal of this study was to identify quantitative markers for quality control, thereby guiding the development of BCBL as a novel antidepressant.

2. Results and Discussions

2.1. Extraction and Chemical Identification of BCBL‐CP

Following our previously established protocol [1], the BCBL extract was processed into BCBL‐CP in two streamlined steps. First, triple‐volume ethanol precipitation, followed by concentration, centrifugation, and drying, yielded the supernatant extract. Second, this extract was redissolved in distilled water, defatted with petroleum ether, and purified by AB‐8 macroporous resin chromatography to afford the final product after elution, concentration, and drying. The extraction yield of BCBL‐CP was 1.0% from raw herbal material. As illustrated in Figure 1, UPLC‐Q‐Orbitrap‐MS/MS and HPLC‐PDA identified five signature constituents of BCBL‐CP, including senkyunolide I (1), senkyunolide H (2), 5‐hydroxyxanthotoxin (3), oxypeucedanin hydrate (4), and byakangelicin (5). Of these, 5‐hydroxyxanthotoxin, oxypeucedanin hydrate, and byakangelicin are coumarins, whereas senkyunolides H and I are phthalides.

FIGURE 1.

Characteristic chromatogram of BCBL‐CP by HPLC‐PDA (254 nm). (1) Senkyunolide I. (2) Senkyunolide H. (3) 5‐Hydroxyxanthotoxin. (4) Oxypeucedanin hydrate. (5) Byakangelicin.

Coumarins and phthalides are two major classes of phenolic lactones widely distributed in Apiaceae, Rutaceae, and Leguminosae species. To date, over 1300 coumarins and 200 phthalides have been identified, many of which readily cross the blood–brain barrier (BBB) and demonstrate excellent preclinical safety profiles [17, 18]. Specifically, existing pharmacokinetic studies have demonstrated that senkyunolide I, byakangelicin, and related coumarin/phthalide components possess favorable BBB penetration capability and central nervous system bioavailability [19, 20, 21]. In addition to their classical antioxidant roles, emerging evidence highlights their neuromodulatory potential. Corylin, a structural analogue of 5‐hydroxyxanthotoxin from Psoralea corylifolia L., significantly alleviated oxidative damage and neuroinflammation by activating the Nrf2/HO‐1 and inhibiting the NF‐κB pathways, suggesting therapeutic potential for neurological disorders [22]. Similarly, senkyunolide H has been reported to attenuate lipopolysaccharide‐induced microglial activation, neuroinflammation, and oxidative stress in BV2 cells via modulation of ERK and NF‐κB pathways [23]. Consequently, supported by the fractionation process and preliminary activity evaluation, these convergent mechanisms position coumarins and phthalides as major bioactive constituents and candidate quality control markers for the antidepressant‐like activity of BCBL, thereby highlighting BCBL‐CP as a potential active fraction.

2.2. Antidepressant‐Like Effects of Acute Pre‐Treatment With BCBL‐CP in Mice

Figure 2A outlines the experimental design for evaluating rapid antidepressant‐like effects following acute pre‐treatment with BCBL‐CP in mice. l‐Menthol, a non‐toxic food additive that contributes significantly to the antidepressant‐like activity of the intact BCBL formula [1, 5], was included at dose‐matched levels to assess the contribution of the coumarin‐ and phthalide‐enriched fraction to the overall efficacy. As shown in Figure 2B,C, a single administration of duloxetine hydrochloride (DLX; 20 mg/kg) markedly shortened immobility time of behavioral despair mice relative to the control group (p < 0.05 and p < 0.01, respectively). Post‐hoc analyses revealed that low‐dose BCBL‐CP (15 mg/kg, plus 3.25 mg/kg l‐menthol) simultaneously reduced immobility in both the TST and FST (p < 0.01 and p < 0.05, respectively). The middle dose (30 mg/kg, plus 6.5 mg/kg l‐menthol) significantly decreased immobility in the TST (p < 0.05), whereas the high dose (60 mg/kg, plus 13 mg/kg l‐menthol) produced a clear downward trend in both tests (p = 0.061 and p = 0.072, respectively). As shown in Figure 2D, acute pre‐treatment with DLX (5 mg/kg) led to a significant increase in head twitches (p < 0.001). Both low‐ and high‐dose BCBL‐CP also significantly elevated the number of head twitches relative to the control group (p < 0.05 and p < 0.01, respectively).

FIGURE 2.

Effects of acute pre‐treatment with BCBL‐CP in behavioral despair mice and 5‐HTP‐induced head‐twitch mice. (A) Schematic illustration. (B and C) Immobility time in the TST and FST. (D) Number of head twitches in the 5‐HTP‐induced head‐twitch test. Measurement data were analyzed by Student's t‐test or one‐way analysis of variance (ANOVA) followed by the post‐hoc LSD test, and presented as mean ± standard error of the mean (SEM) (n = 10). *p < 0.05, **p < 0.01, and ***p < 0.001 in comparison to the control group.

The TST and FST are two well‐established paradigms of behavioral despair for rapid confirmation of antidepressant activity in rodents [24, 25, 26]. Consistent with our prior studies, acute pre‐treatment with DLX at a dose of 20 mg/kg reliably shortened immobility time in mice, corroborating the predictive validity of these models [1, 5]. Likewise, BCBL‐CP co‐administered with the formula‐relevant food additive l‐menthol reduced immobility time to some extent, with the low‐dose combination exhibiting superior efficacy in alleviating behavioral despair. The 5‐HTP‐induced head‐twitch test is a typical model to assess the effects of antidepressants on the serotoninergic system [5, 27, 28]. Our results showed that DLX (5 mg/kg) markedly increased twitch counts, consistent with its known mechanism of serotonin reuptake inhibition [29]. BCBL‐CP also produced a comparable potentiation of head twitches, implying that enhancement of the serotoninergic system contributes to its antidepressant‐like action, aligning with our previous findings [1, 5].

2.3. Antidepressant‐Like Effects of Chronic Co‐Treatment With BCBL‐CP in Mice

Figure 3A illustrates the establishment of a chronic reserpine‐induced depression model in mice to evaluate the sustained antidepressant‐like effects of BCBL‐CP [1, 30, 31]. For the initial 2 weeks, reserpine at a dose of 0.2 mg/kg was administered to induce depression‐like behaviors. For the subsequent 2 weeks, reserpine at a dose of 0.1 mg/kg was maintained to sustain the depression‐like phenotypes, during which DLX and BCBL‐CP were administered to assess therapeutic efficacy in this mouse model.

FIGURE 3.

Effects of chronic co‐treatment with BCBL‐CP in reserpine‐induced depression‐like mice. (A) Schematic illustration. (B) Rate of body weight change. The rate of body weight change was calculated using the body weight at the start of reserpine induction (Week 0) as the baseline, according to the following formula: Rate of body weight change (%) = [(weight at Week X—weight at Week 0)/weight at Week 0] × 100%. (C) Sucrose preference index at baseline, week 2, and week 4. (D and E) Total distance and percentage of distance in the center area at baseline, week 2, and week 4. (F) Rectal temperature at baseline, week 2, and week 4. (G) Immobility time in the TST at week 4. Measurement data were analyzed by Student's t‐test, one‐way ANOVA, or one‐way repeated‐measures ANOVA followed by the post‐hoc LSD test, and presented as mean ± SEM (n = 8). *p < 0.05, **p < 0.01, and ***p < 0.001 in comparison to the control group; ^# p < 0.05, ^## p < 0.01, and ^### p < 0.001 in comparison to the model group.

As illustrated in Figure 3B, chronic reserpine injection (0.2 mg/kg) for 2 weeks markedly suppressed body weight gain in model mice relative to the control group, and this suppression persisted through week 4 (p < 0.001). Administration of DLX or BCBL‐CP over the subsequent 2‐week period alleviated reserpine‐induced weight gain suppression. Both the DLX (20 mg/kg) and low‐BCBL‐CP (15 mg/kg, plus 3.25 mg/kg l‐menthol) groups showed significant differences compared with the model group (p < 0.01 and p < 0.05, respectively).

The sucrose preference test (SPT) was performed at baseline, week 2, and week 4 to assess anhedonia and evaluate the modulatory effects of BCBL‐CP on this phenotype. As shown in Figure 3C, no significant difference in sucrose preference was observed between groups at baseline. By week 2, chronic reserpine treatment significantly reduced sucrose preference compared with the control group (p < 0.001 or p < 0.05), indicating the successful induction of anhedonia. This reduction persisted through week 4, with model mice showing significantly lower sucrose preference than controls (p < 0.05). However, chronic co‐treatment with either DLX or low‐dose BCBL‐CP for 2 weeks significantly increased sucrose preference in model mice (p < 0.05).

As shown in Figure 3D,E, the open field test (OFT) was conducted to assess the effects of BCBL‐CP on locomotor activity in chronic reserpine‐treated mice. Prior to modeling, no significant difference in locomotor activity was observed. By week 2, chronic reserpine treatment led to significant reductions in both total distance traveled and percentage of distance in the central area (p < 0.001 or p < 0.05). By week 4, the total distance traveled and percentage of distance in the central area of model mice were also significantly lower than those of the control group (p < 0.001). Co‐treatment with DLX or BCBL‐CP improved abnormal activity and alleviated anxiety‐like symptoms in model mice to varying extents. Notably, both low‐ and high‐dose (60 mg/kg, plus 13 mg/kg l‐menthol) BCBL‐CP exhibited significant modulatory effects (p < 0.001 or p < 0.05).

Reserpine‐induced hypothermia is a well‐established parameter for evaluating the effects of antidepressants on the NEergic system in rodents. Therefore, the rectal temperature of each mouse was measured at baseline, week 2, and week 4. As shown in Figure 3F, no significant intergroup differences were observed at baseline. However, by week 2, chronic reserpine treatment significantly lowered body temperature compared with the control group (p < 0.001 or p < 0.05). By week 4, co‐treatment with DLX or BCBL‐CP effectively ameliorated hypothermia in model mice to varying degrees, with DLX as well as low‐ and middle‐dose (30 mg/kg, plus 6.5 mg/kg l‐menthol) BCBL‐CP showing superior efficacy (p < 0.001 or p < 0.05).

The TST was additionally performed to evaluate the effects of BCBL‐CP on immobility time in chronic reserpine‐treated mice. As shown in Figure 3G, chronic reserpine treatment significantly increased immobility time by week 4 compared with the control group (p < 0.001). Co‐treatment with DLX substantially reduced this parameter (p < 0.001). Furthermore, low‐dose BCBL‐CP also significantly decreased immobility time (p < 0.01). These findings further demonstrated the antidepressant‐like effects of BCBL‐CP in this established depression model.

Reserpine, a catecholamine‐depleting agent, has long served as a pharmacological tool for inducing depression‐like phenotypes in rodents by depleting monoamine neurotransmitters (serotonin, NE, and dopamine) [5, 32]. In the present study, chronic reserpine administration successfully induced a comprehensive depressive syndrome characterized by attenuated body weight gain, anhedonia, reduced locomotor activity, hypothermia, and increased behavioral despair. These behavioral alterations are consistent with previous reports, validating the reserpine‐induced model as a robust tool for screening antidepressant candidates [1, 30, 31]. Notably, BCBL‐CP co‐treatment demonstrated significant antidepressant‐like efficacy across multiple behavioral domains, mirroring the therapeutic effects of DLX, a dual serotonin‐NE reuptake inhibitor. The amelioration of reserpine‐induced weight loss, the restoration of sucrose preference, and the reduction of immobility time in the TST suggested that BCBL‐CP effectively reversed core depressive symptoms [33]. The improvement in locomotor activity and the reduction of anxiety‐like behaviors in the central zone of the OFT indicated that BCBL‐CP might possess broader psychotropic properties beyond mood regulation, potentially addressing the comorbid anxiety symptoms often observed in clinical depression [34]. Furthermore, the reversal of reserpine‐induced hypothermia by BCBL‐CP provided compelling evidence for its modulatory effects on the NEergic system, suggesting that the antidepressant‐like effects of BCBL‐CP may be partially mediated through this pathway [5].

2.4. Antidepressant‐Like Mechanisms of Chronic Co‐Treatment With BCBL‐CP in Mice

2.4.1. BCBL‐CP Modulates the Monoaminergic System in Reserpine‐Induced Depression‐Like Mice

As demonstrated by the 5‐HTP‐induced head‐twitch test and rectal temperature measurement in chronic reserpine‐treated mice, both serotonergic and NEergic systems appeared to mediate the antidepressant‐like effects of BCBL‐CP. To further evaluate these effects on the monoaminergic system, serotonin and NE levels in the prefrontal cortex (PFC) and hippocampus were measured by ELISA. As illustrated in Figure 4, chronic reserpine administration for 4 weeks significantly reduced serotonin and NE levels in both regions of model mice (p < 0.01 or p < 0.05), whereas co‐treatment with DLX or BCBL‐CP restored these levels to varying degrees. Notably, compared with model mice, the DLX group exhibited significant increases in prefrontal NE (p < 0.01) and hippocampal serotonin (p < 0.05) levels. Additionally, prefrontal NE levels were significantly increased in the low‐BCBL‐CP group (p < 0.05).

FIGURE 4.

Effects of chronic co‐treatment with BCBL‐CP on the prefrontal and hippocampal levels of serotonin and NE in reserpine‐induced depression‐like mice. (A and C) The standard curves and linear regression equations of serotonin and NE by ELISAcalc (four‐parameter). (B) The levels of serotonin and NE in the PFC. (D) The levels of serotonin and NE in the hippocampus. Measurement data were analyzed by Student's t‐test, and presented as mean ± SEM (n = 5). *p < 0.05 and **p < 0.01 relative to the control group; ^# p < 0.05 and ^## p < 0.01 relative to the model group.

The present findings demonstrated that BCBL‐CP exerted antidepressant‐like effects primarily through modulation of monoaminergic transmission in the PFC and hippocampus, two limbic regions critically implicated in depression pathophysiology [35, 36]. Specifically, the selective elevation of prefrontal NE by BCBL‐CP directly mediated the alleviation of reserpine‐induced hypothermia, while its modulation of monoamine levels in the PFC and hippocampus contributed to the reversal of anhedonia and behavioral despair observed in behavioral tests [35]. The successful establishment of chronic reserpine‐induced monoamine depletion validated the classical monoamine hypothesis underlying our model [7]. While DLX robustly restored both serotonin and NE levels as expected for a dual serotonin‐NE reuptake inhibitor [29], BCBL‐CP induced only partial restoration, as evidenced by the failure of low‐dose treatment to fully normalize prefrontal and hippocampal serotonin levels. This differential modulatory profile suggests that BCBL‐CP may possess distinct pharmacological properties or receptor selectivity compared with conventional antidepressants.

2.4.2. BCBL‐CP Ameliorates the Hippocampal Histomorphological Abnormality in Reserpine‐Induced Depression‐Like Mice

Additionally, hippocampal architecture was examined by HE staining. As illustrated in Figure 5, reserpine‐induced depression‐like mice exhibited marked nuclear condensation, with extensive neuronal necrosis and pyknotic nuclei evident in the dentate gyrus (DG) region (p < 0.001). Both DLX and low‐dose BCBL‐CP ameliorated these degenerative changes, restoring structural regularity, cellular integrity, and distinct nuclear morphology in the DG area (p < 0.001). In the cornu ammonis 1 (CA1) subfield, model mice showed a modest increase in degenerated neurons, while low‐dose BCBL‐CP tended to attenuate these alterations, albeit without reaching statistical significance (p = 0.063).

FIGURE 5.

Effects of chronic co‐treatment with BCBL‐CP on the hippocampal histomorphological abnormality in reserpine‐induced depression‐like mice. (A) Representative HE staining images of the hippocampus (scale bar = 400 or 100 µm). (B) Quantification of degenerated neurons in the DG and CA1 regions. Measurement data were analyzed by Student's t‐test, and represented as mean ± SEM (n = 3). p < 0.05, ***p < 0.001 relative to the control group; ^### p < 0.001 in contrast to the model group.

Our study provides histomorphological evidence that chronic reserpine exposure induces substantial neurodegenerative alterations in the hippocampus, particularly within the DG region, characterized by extensive neuronal necrosis and nuclear condensation. These findings align with the established understanding that depression is associated with hippocampal atrophy and neuronal vulnerability, underscoring the critical role of structural integrity in mood regulation [37]. Notably, chronic co‐treatment with low‐dose BCBL‐CP effectively reversed these pathological changes, restoring neuronal density and architectural regularity in the DG region. This reversal of DG neuronal damage directly contributes to the improvement of anxiety‐like behaviors and stress resilience observed in the OFT. This neuroprotective efficacy suggests that BCBL‐CP may confer resilience against hippocampal damage in mood disorders, warranting mechanistic elucidation in future investigations [38].

2.4.3. BCBL‐CP Attenuates the Hippocampal Neurotrophic Disturbance in Reserpine‐Induced Depression‐Like Mice

As a highly conserved neurotrophin, BDNF plays a pivotal role in neural development. To investigate how chronic BCBL‐CP co‐administration affected neurotrophic signaling, we measured hippocampal BDNF expression via IF staining in the reserpine‐induced depression‐like model. As illustrated in Figure 6, compared with controls, model animals exhibited markedly reduced BDNF levels in the hippocampus (p < 0.05). Furthermore, chronic co‐treatment with either DLX or low‐dose BCBL‐CP partially restored BDNF expression, albeit without reaching statistical significance (p = 0.098 and p = 0.083, respectively).

FIGURE 6.

Effects of chronic co‐treatment with BCBL‐CP on the hippocampal neurotrophic disturbance in reserpine‐induced depression‐like mice. (A) Representative IF staining images of the hippocampus (scale bar = 400 µm). (B) Quantification of BDNF expression in the hippocampus. Measurement data were analyzed by Student's t‐test, and represented as mean ± SEM (n = 3). *p < 0.05 relative to the control group.

BDNF, a critical mediator of synaptic plasticity and neuronal survival, plays an important role in the pathogenesis and progression of mood disorders. The observed downregulation of hippocampal BDNF in our study thus provides a molecular substrate for the structural and functional deficits characterizing this depression‐like phenotype, thereby supporting the well‐established neurotrophic hypothesis of depression [9, 39]. Disruption of BDNF‐mediated hippocampal synaptic plasticity is tightly linked to the progression of core depressive symptoms, including anhedonia and behavioral despair [40, 41]. Notably, chronic co‐administration of low‐dose BCBL‐CP partially restored BDNF levels, exhibiting a trend toward neurotrophic enhancement comparable to the reference antidepressant DLX. Although this restorative effect did not achieve statistical significance, the consistent directional changes across both treatment groups imply that BCBL‐CP may engage neurotrophic mechanisms similar to those of the established antidepressant DLX [42].

2.4.4. BCBL‐CP Corrects the Gut Microbiota Dysregulation in Reserpine‐Induced Depression‐Like Mice

Fecal 16S rDNA (V3 + V4 regions) sequencing was performed to assess BCBL‐CP‐mediated regulation of gut microbiota in reserpine‐induced mice. As shown in Figure 7A, Venn analysis identified 688, 468, and 843 unique amplicon sequence variants (ASVs) in the control, model, and low‐BCBL‐CP groups, respectively, with 219 shared across all groups. Model mice showed depleted unique ASVs versus controls, whereas low‐dose BCBL‐CP co‐treatment elevated ASV number. Notably, α diversity indices (Ace, Chao, Sobs) were reduced in model mice (Figure 7B–D), while low‐BCBL‐CP significantly restored Ace, Chao, and Sobs levels (p < 0.01). β diversity (Bray–Curtis PCoA) further confirmed community structure restoration, as samples from the low‐BCBL‐CP group clustered closely with controls (Figure 7F).

FIGURE 7.

Effects of chronic co‐treatment with BCBL‐CP on the richness and similarity of the gut microbiota among microbial communities in reserpine‐induced depression‐like mice. (A) Venn diagram on ASV level. (B–E) Comparison of Ace, Chao, Sobs, and Shannon indices based on ASV level. (F) PCoA of the fecal microbial communities based on order level. Measurement data were analyzed by Student's t‐test, and represented as mean ± SEM (n = 5). *p < 0.05 in comparison to the control group; ^## p < 0.01 in contrast to the model group.

Subsequently, we conducted analyses of the abundance and composition of the gut microbiota among different groups. As shown in Figure 8, order‐ and species‐level alterations within reserpine‐treated mice were observed in our study. On order level (Figure 8A,C,D), Bacteroidales and RF39 were depleted relative to the control group, with Bacteroidales reaching significance (p < 0.05). Low‐dose BCBL‐CP partially reversed the decrease in abundance of the two taxa, with RF39 reaching significance (p < 0.05). On species level (Figure 8B,E–N), chronic reserpine injection disrupted 13 taxa, significantly altering one species each within the families Muribaculaceae and Prevotellaceae (UCG‐001) and the genus Adlercreutzia (p < 0.01 or p < 0.05). BCBL‐CP could restore the abundance of these bacteria to some extent, with 8 reaching significance (p < 0.01 or p < 0.05), suggesting effective amelioration of reserpine‐induced dysbiosis.

FIGURE 8.

Effects of chronic co‐treatment with BCBL‐CP on the relative abundance of the gut microbiota on order and species levels in reserpine‐induced depression‐like mice. (A and B) Bacterial composition of the different communities on order and species levels. (C and D) The relative abundance of distinctive taxa on order level. (E–N) The relative abundance of distinctive strains on species level. Measurement data were represented as mean ± SEM (n = 5). *p < 0.05 and **p < 0.01 relative to the control group; ^# p < 0.05 and ^## p < 0.01 in comparison to the model group.

The above findings provide compelling evidence that chronic reserpine exposure induces substantial gut dysbiosis, characterized by reduced α diversity, altered community structure, and specific taxonomic perturbations on both order and species levels [43]. These results align with the emerging concept that depression is associated with compromised microbial richness and diversity, supporting the notion that the gut microbiota serves as a critical interface between environmental stressors and neuropsychiatric pathophysiology [10]. Gut microbiota dysbiosis can drive depressive‐like behaviors by modulating central monoaminergic function, neuroinflammation, and neurotrophic signaling through the microbiota–gut–brain axis [44]. The restoration of ASV numbers and α diversity indices (Ace, Chao, and Sobs) by low‐dose BCBL‐CP, together with the normalization of β diversity profiles, underscores its potential to restore microbial eubiosis. The regulation of gut microbiota homeostasis by BCBL‐CP is closely coupled with its central neuroprotective and monoaminergic modulatory effects, which collectively contribute to the reversal of core depressive‐like behaviors in this model. Additionally, the observed concordance between gut microbiota restoration and hippocampal BDNF upregulation in our study further supports the existence of a microbiota–gut–brain axis through which BCBL‐CP exerts its neurotrophic effects [45].

2.4.5. Comprehensive Comparative Analysis of BCBL‐CP and Duloxetine

Behaviorally, low‐dose BCBL‐CP (15 mg/kg, plus 3.25 mg/kg l‐menthol) and DLX (20 mg/kg) exerted comparable therapeutic effects. Both significantly alleviated core depressive phenotypes, including body weight loss, anhedonia, behavioral despair, and hypothermia (p < 0.01 or p < 0.05). BCBL‐CP also showed distinctive advantages: it markedly improved anxiety‐like behaviors and locomotor inhibition in the OFT (p < 0.01 or p < 0.05), whereas DLX only produced a mild and statistically non‐significant improvement. BCBL‐CP showed no strict linear dose‐dependent effect: the low dose exhibited the optimal efficacy, while middle and high doses showed weaker effects. This nonlinear response is common in multi‐component herbal medicines [46, 47].

Mechanistically, the two agents exhibited both similarities and key differences. Both significantly attenuated neuronal damage in the hippocampal DG region (p < 0.001) and tended to upregulate hippocampal BDNF expression, indicating shared neuroprotective mechanisms. However, their monoaminergic profiles differed considerably: DLX significantly restored prefrontal NE and hippocampal serotonin levels, whereas BCBL‐CP selectively elevated prefrontal NE with only mild modulation of serotonin. Notably, BCBL‐CP effectively restored depression‐related gut microbiota homeostasis, including α‐diversity, community structure, and the abundance of key depressive phenotype‐associated taxa, via the microbiota–gut–brain axis. This represents a major mechanistic advantage of BCBL‐CP over DLX, as existing studies have confirmed that DLX exerts antidepressant effects mainly through direct central monoamine modulation, with limited regulatory effects on gut dysbiosis [10, 45]. By contrast, normalization of the gut microbiota constitutes one of the core antidepressant mechanisms of BCBL‐CP.

3. Conclusions

The current study provided a coumarin‐ and phthalide‐enriched fraction (BCBL‐CP) from BCBL, which mainly contains 5‐hydroxyxanthotoxin, oxypeucedanin hydrate, byakangelicin, and senkyunolides H and I. Comprehensive pharmacological evaluation revealed that BCBL‐CP exerted significant antidepressant‐like effects in behavioral despair mice and chronic reserpine‐induced depression‐like mice, mediated through modulation of the monoaminergic system, amelioration of hippocampal histomorphological abnormalities, amelioration of neurotrophic disturbance, and attenuation of gut microbiota dysbiosis. These findings not only further validate the therapeutic potential of BCBL but also establish quantitative markers for quality control, thereby paving the way for its development as a novel antidepressant.

However, this study has several limitations. First, we did not perform independent pharmacological studies on the five identified quality markers. Second, IF staining, despite its high spatial resolution for mapping hippocampal BDNF distribution, is inherently a semi‐quantitative assay. Third, this study only included male mice, which is a limitation given the well‐documented sex disparities in major depressive disorder and the strong regulatory effect of sex hormones on gut microbiota. Future studies will clarify the antidepressant‐like effects and underlying mechanisms of the five quality markers, perform precise absolute quantification of BDNF via western blotting or ELISA, and include both male and female mice to explore sex‐related differences in the efficacy of BCBL, thereby providing a more robust scientific basis for the quality control and clinical application of this formula.

4. Experimental Section

4.1. Extraction and Chemical Identification of BCBL‐CP

4.1.1. Plant Material

Angelica dahurica roots (voucher no. 201116), L. chuanxiong rhizomes (voucher no. B8031511), S. chinensis fruits (voucher no. 21041002), and l‐menthol (voucher no. 210307025) were sourced from Beijing Tongrentang Co., Ltd. (Beijing, China) on March 4, 2022. All specimens were deposited in the Laboratory of Natural Products Chemistry, Academy of Military Medical Sciences.

4.1.2. Extraction

Angelica dahurica, L. chuanxiong, and S. chinensis (5:3:3, w/w) were soaked in 10‐fold distilled water (w/v) for 60 min, heated to boiling, and maintained at gentle boil for 60 min. After filtration, the residue was re‐extracted with 10‐fold distilled water for 60 min. Combined filtrates were centrifuged, concentrated, mixed with three volumes of 95% ethanol, and refrigerated at 4°C overnight. The supernatant was concentrated, dried, redissolved in distilled water, defatted with petroleum ether, and purified by AB‐8 macroporous resin chromatography to yield the final product.

4.1.3. Chemical Identification

UPLC‐Q‐Orbitrap‐MS/MS analysis of BCBL‐CP was performed on a Phenomenex Kinetex C18 column (2.1 × 100 mm, 2.6 µm) and a Waters ACQUITY BEH Amide column (2.1 × 100 mm, 1.7 µm) coupled to an Orbitrap Exploris 120. C18 conditions: (A) 0.01% aqueous acetic acid and (B) isopropanol–acetonitrile (1:1); flow rate 0.3 mL/min; injection volume 2 µL; 0–1 min (1% B), 1–8 min (1% → 99% B), 8–9 min (99% B), 9–9.1 min (99% → 1% B), 9.1–12 min (1% B). Amide conditions: (A) 25 mmol/L ammonium acetate and 25 mmol/L ammonia hydroxide in water (pH 9.75) and (B) acetonitrile; flow rate 0.5 mL/min; injection volume 2 µL; 0–0.5 min (95% B), 0.5–7 min (95% → 65% B), 7–8 min (65%→40% B), 8–9 min (40% B), 9–9.1 min (40% → 95% B), 9.1–12 min (95% B). MS parameters (DDA mode): m/z 70–1200; MS resolution 60 000; MS/MS resolution 15 000; stepped NCE 20/30/40; +3.8/−3.4 kV. Data processing: ProteoWizard (mzXML conversion), XCMS (peak processing), and BiotreeDB (annotation).

Subsequently, HPLC‐PDA coupled with comparison against reference standards was performed to assign the major constituents of BCBL‐CP. Analytical HPLC was conducted on a Shimadzu LC‐20AT system equipped with a PDA detector. Chromatographic separation was achieved on a Supersil AQ‐C18 column (4.6 mm × 250 mm, 5 µm) using a mobile phase consisting of 0.1% phosphoric acid in water (A) and acetonitrile (B) with gradient elution as follows: 0–10 min (15% → 18% B), 10–95 min (18% B), 95–100 min (18% → 15% B). The detection wavelength was set at 254 nm. The column temperature was maintained at 35°C, the flow rate was 0.8 mL/min, and the injection volume was 20 µL. The reference standard 5‐hydroxyxanthotoxin (PU0969) was purchased from Chengdu Pusi Biotechnology Co., Ltd. (Chengdu, China). Oxypeucedanin hydrate (B21189) and byakangelicin (B20027) were obtained from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China), while senkyunolide H (CFN99595) and senkyunolide I (CFN99596) were purchased from Wuhan Zhongbiao Technology Co., Ltd. (Wuhan, China).

4.2. Antidepressant‐Like Effects of Acute Pre‐Treatment With BCBL‐CP in Mice

4.2.1. Animals and Grouping

A total of 150 male ICR mice (weighing 20 ± 2 g, provided by SPF Biological Technology Co., Ltd., Beijing) were raised in an environment with a temperature of 24 ± 1°C and a 12 h light‐dark cycle, with free access to standard diet and water. After 7 days of adaptive feeding, the mice were randomly divided into five groups (n = 30 per group): the control group (vehicle; 20 mL/kg, i.g.), the DLX group (20 mg/kg for TST and FST, 5 mg/kg for 5‐HTP‐induced head‐twitch test, i.g.), the low‐BCBL‐CP group (15 mg/kg, plus 3.25 mg/kg L‐menthol, i.g.), the middle‐BCBL‐CP group (30 mg/kg, plus 6.5 mg/kg l‐menthol, i.g.), and the high‐BCBL‐CP group (60 mg/kg, plus 13 mg/kg l‐menthol, i.g.). All procedures were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee (approval IACUC‐DWZX‐2021‐618) and every effort was made to minimize suffering.

4.2.2. Drug Treatment

After 7 days of acclimatization, the mice were orally administered DLX (1304002; Shanghai Wandai Pharmaceutical Co., Ltd.) or various doses of BCBL‐CP, respectively. Mice in the control group were given an equivalent volume of vehicle.

4.2.3. Behavioral Testing

One hour after administration, 10 mice from each group were subjected to the FST, another 10 to the TST, and the remaining 10 to the 5‐HTP‐induced head‐twitch test, so as to assess the potential antidepressant‐like effects induced by acute pre‐treatment with BCBL‐CP. The detailed protocols for the TST, FST, and 5‐HTP‐induced head‐twitch test were performed in accordance with our previously published study [1].

4.3. Antidepressant‐Like Effects of Chronic Co‐Treatment With BCBL‐CP in Mice

4.3.1. Animals and Grouping

An additional 48 male ICR mice (20 ± 2 g, obtained from the same supplier) were randomly allocated into six groups (n = 8 per group): the control group (vehicle + vehicle; 20 mL/kg, i.g.), the model group (reserpine + vehicle; 20 mL/kg, i.g.), the DLX group (reserpine + DLX; 20 mg/kg, i.g.), the low‐BCBL‐CP group (reserpine + BCBL‐CP; 15 mg/kg, plus 3.25 mg/kg l‐menthol, i.g.), the middle‐BCBL‐CP group (reserpine + BCBL‐CP; 30 mg/kg, plus 6.5 mg/kg l‐menthol, i.g.), and the high‐BCBL‐CP group (reserpine + BCBL‐CP; 60 mg/kg, plus 13 mg/kg l‐menthol, i.g.). The housing conditions and ethical care of the mice were consistent with the detailed descriptions provided earlier.

4.3.2. Modeling and Drug Treatment

After a 7‐day acclimatization phase, mice received intraperitoneal injections of reserpine (0.2 mg/kg, dissolved in sterile endotoxin‐free 0.2% acetic acid solution) for 2 consecutive weeks to establish depression‐like behaviors. For the subsequent 2 weeks, the reserpine dosage was decreased to 0.1 mg/kg to sustain the depressive phenotype. During the period of low‐dose reserpine injection, DLX and varying doses of BCBL‐CP were concurrently administered orally as described above to evaluate the antidepressant‐like effects of chronic co‐treatment. The control group received an equivalent volume of vehicle throughout the experiment.

4.3.3. Behavioral Testing

The SPT, OFT, and rectal temperature measurement were performed at baseline, week 2, and week 4 to evaluate effects of BCBL‐CP on the anhedonia, locomotor dysfunction, and NEergic system, respectively. By week 4, the TST was also conducted to assess the effects of BCBL‐CP on behavioral despair in model mice. Detailed procedures can be found in our prior publication [16].

4.3.4. Sample Collection

Following the completion of behavioral tests, three mice per group were anesthetized with isoflurane, perfused with 4% paraformaldehyde, and prepared for HE and IF staining. The hippocampus and PFC of the remaining animals were collected and analyzed to determine the levels of serotonin and NE by ELISA. The commercial competitive ELISA kits used for neurotransmitter detection were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Meanwhile, colonic contents were collected from the remaining animals for gut microbiota analysis. Detailed procedures for HE and IF staining, ELISA, and gut microbiota analysis are available in our previous studies [1, 16].

4.3.5. Statistical Analysis

Behavioral and biochemical data presented as mean ± SEM were analyzed in SPSS 16.0 and plotted with GraphPad Prism 6.01. Two‐group comparisons used Student's t‐test; multi‐group comparisons used one‐way ANOVA or one‐way repeated‐measures ANOVA followed by LSD. p < 0.05 was considered significant and 0.05 ≤ p < 0.1 was noted as a trend. Gut–microbiota statistics (mean ± SEM) were run on the Majorbio cloud platform (Shanghai, China) with Student's t‐test for inter‐group comparisons. All plots were generated online.

Author Contributions

Chun‐Xue Gao: data curation, investigation. Shuai‐Ming Zhu: methodology, writing – original draft. Yue Sun: data curation, supervision. Fu‐Yao Luo: data curation, investigation. Ting Feng: data curation, validation. Yu Yang: supervision. Chang‐Wei Li: methodology, writing – original draft. Chun‐Yan Liu: supervision, writing – review and editing. You‐Zhi Zhang: project administration, writing – review and editing.

Supporting Information

Table S1 : Sequences for each sample on ASV level; Table S2 : α diversity estimators for each sample on ASV level; Table S3 : Relative abundance of gut microbiota among different groups on order level; Table S4 : Relative abundance of gut microbiota among different groups on species level.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: cbdv71322‐sup‐0001‐SuppMat.docx

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.