The Ziyin Buyang Formula (ZYBYF) comprises two distinct formulae: the Ziyin Formula (ZYF) and the Buyang Formula (BYF), with detailed compositions provided in Table 1. Bioactive components of ZYBYF constituents were retrieved from the Traditional Chinese Medicine Systems Pharmacology Database (TCMSP, https://ibts.hkbu.edu.hk/LSP/tcmsp.php) and supplementary literature (Li et al., 2021). Active compounds were selected based on pharmacokinetic criteria: oral bioavailability (OB) ≥30% and drug-likeness (DL) ≥0.18. Potential targets were predicted using TCMSP algorithms, with gene symbols standardized via the UniProtKB database (https://www.uniprot.org).

PCOS-associated targets were systematically collated from four databases: GeneCards (https://www.genecards.org/; keyword “Polycystic ovary syndrome”, relevance score >20), DrugBank (https://go.drugbank.com/), DisGeNET (https://www.disgenet.org/), and PharmGKB (https://www.pharmgkb.org/). Duplicate entries were removed to generate a consolidated PCOS target repository.

Differentially expressed genes (DEGs) from ovarian tissues of circadian-disrupted rats versus controls were identified through transcriptomic analysis. DEGs were defined by an adjusted p-value <0.05 and an absolute log2-fold change >1.0. Orthologous human gene mapping was performed using the NCBI HomoloGene database (https://www.ncbi.nlm.nih.gov/homologene) to translate rat gene symbols into their human equivalents. The mapping was followed by manual verification to resolve ambiguities, ensuring functional conservation by cross-referencing gene annotations.

Venn diagram analysis (BioVenn web tool) identified overlapping targets among three datasets: ZYBYF component targets (1,258 genes), PCOS-associated targets (892 genes), and circadian disruption-related DEGs (327 genes). The intersection (68 genes) was retained for subsequent analyses.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were conducted using the R package clusterProfiler (v4.0). Statistical significance was determined at adjusted p < 0.05 and q-value <0.05. Protein-protein interaction networks were constructed using STRING (v11.5; confidence score >0.7) and visualized in Cytoscape (v3.9.1).

Thirty specific pathogen-free (SPF) female Sprague-Dawley rats (7 weeks old, 180–220 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (SCXK 2021-0011). Animals were housed under controlled conditions (22°C ± 1 °C, 50%–60% humidity) with ad libitum access to food and water. Following 1-week acclimatization under 12:12 light-dark (LD) cycles, rats were randomized into three groups (n = 10/group):

• Control (D/L): Standard LD cycles

Circadian disruption was induced via 24-h light exposure (500 ± 20 lux intensity, LED panels) for 10 weeks (Ma et al., 2018). Drug administration commenced post-induction: ZYF (5.88 mg/kg) during proestrus/estrus phases and BYF (6.195 mg/kg) during metestrus/diestrus phases via oral gavage. All procedures were approved by the Nanjing University of Chinese Medicine Animal Ethics Committee (202201A002).

Ovarian tissues were fixed in 4% paraformaldehyde (24 h), paraffin-embedded, and sectioned (5 μm thickness). Hematoxylin and eosin (H&E) staining was performed using standard protocols. Follicular morphology and cystic changes were assessed by two blinded pathologists using an Olympus BX53 microscope.

Serum anti-Müllerian hormone (AMH, #JM-01626R1), estradiol (E2, #JM-01981R1), luteinizing hormone (LH, #JM-02207R1), follicle-stimulating hormone (FSH, #JM-01972R1), testosterone (T, #JM-01983R1), and adrenocorticotropic hormone (ACTH, #JM-01971R1) levels were measured using ELISA kits (Jingmei Biotech) according to manufacturer protocols. Absorbance was quantified at 450 nm using a microplate reader (BioTek Synergy H1).

Total protein was extracted using RIPA lysis buffer supplemented with protease inhibitors. Proteins (20 μg/lane) were separated on 10% SDS-PAGE gels and transferred to PVDF membranes. After blocking with 5% BSA, membranes were incubated overnight at 4 °C with primary antibodies: p-p38 (Abclonal AP1311, 1:2,000), p-ERK1/2 (Zen-Bio 310289, 1:1,000), p-JNK (Abclonal AP0631, 1:2,000), BAX (Proteintech 50599-2-Ig, 1:5,000), Bcl-2 (Proteintech 68103-1-Ig, 1:5,000), CLOCK (Abcam, ab229495, 1:1,000), β-actin (Zen-Bio 200068-8F10, 1:5,000). HRP-conjugated secondary antibodies (1:3,000) were detected using ECL Prime (Amersham). Band intensities were quantified via ImageJ (v1.53k).

Total RNA was isolated using TRIzol reagent (Invitrogen) and reverse-transcribed with PrimeScript RT Master Mix (Takara). SYBR Green-based qPCR was performed on a StepOnePlus system (Applied Biosystems) using the following cycling parameters: 95 °C for 1 min; 40 cycles of 95 °C for 20 s, 60 °C for 20 s, and 72 °C for 30 s. Primer sequences: lncRNA TCONS_00265853: F: 5′-CCT​TCC​CTC​CTC​AAG​TTG​CCT-3′, R: 5′-GTC​CTC​ACA​TAG​ACA​ATG​CCA​AA-3'; miR-421-5p: F: 5′-CAC​ACA​GAA​GGC​CAC​AAA​AA-3′, R: 5′-TAT​GGT​TTT​GAC​GAC​TGT​GTG​AT-3'; Clock: F: 5′-TCT​CTT​CCA​AAC​CAG​ACG​CC-3′, R: 5′-TGC​GGC​ATA​CTG​GAT​GGA​AT-3'; β-actin (reference): F: 5′-AGG​GTG​TGA​TGG​TGG​GTA​TG-3′, R: 5′-AGG​ATG​CCT​CTC​TTG​CTC​TG-3'. Relative expression was calculated using the 2^−ΔΔCT method with normalization to β-actin.

Apoptosis was assessed using a TUNEL assay kit (Solarbio T2130). Deparaffinized sections were incubated with proteinase K (20 μg/mL), equilibrated, and labeled with TdT reaction mix (37 °C, 1 h). Nuclei were counterstained with DAPI. Images were captured using a Nikon Eclipse CI fluorescence microscope (20× objective). Apoptotic indices were calculated as TUNEL-positive cells/total cells ×100% in five random fields per section.

Data are expressed as mean ± SEM. Normality of data was assessed using the Kolmogorov-Smirnov test. For normally distributed data, one-way ANOVA with Newman-Keuls post hoc test was used. Non-parametric data were analyzed using the Kruskal-Wallis test with Dunn’s correction. For bioinformatics enrichment analysis, p-values were corrected for multiple comparisons using the Benjamini-Hochberg method. Statistical significance was defined as p < 0.05. All analyses were performed in SPSS 23.0 (IBM).

Through the identification of active ingredients, 40 compounds from ZYF and 66 from BYF were recognized. This led to the determination of 226 action targets for ZYF and 239 for BYF. After eliminating duplicates, a total of 261 unique ZYBYF action targets were identified (Figure 1A). Additionally, 5434 PCOS-associated targets were compiled from various databases: 4977 from GeneCards, 166 from PharmGKB, 988 from DisGeNET, 31 from DrugBank, and 182 from OMIM (Figure 1B). Moreover, 1,415 differential ovarian genes were identified in rats exposed to continuous light, with 1,150 human homologs obtained from the HomoloGene database. Analysis revealed 25 intersection genes among the 261 potential ZYBYF targets, the 5,434 PCOS targets, and the 1,150 circadian rhythm disruption targets, which represent the target genes of ZYBYF in treating PCOS with circadian rhythm disruption (Figure 1C).

The 25 intersection genes underwent analysis using the R package, yielding 450 biological processes in GO functional enrichment and 38 pathways in KEGG pathway analysis. The 20 most significant terms are presented in Figure 2. The MAPK signaling pathway emerged as a pivotal pathway.

MAPK signaling is crucial in regulating circadian rhythm and is closely linked to PCOS (Dehghani et al., 2020). Enrichment analysis indicated that the MAPK signaling pathway is pivotal in PCOS with circadian rhythm disruption. STRING data predicted an interaction between CLOCK and intersection genes enriched in the MAPK signaling pathway (PRKCA, IL1B, ERBB2, KDR, VEGFA, and TGFB1), suggesting that CLOCK may target PRKCA to regulate the MAPK pathway (Figure 3).

Continuous light exposure for 10 weeks disrupted the estrous cycle in rats (Figure 4A), resulting in reduced body weights (Figures 4B,C), increased ovary weight and ovarian index (Figures 4D,E), and the appearance of polycystic ovarian morphology with increased cystic follicles and reduced granulosa cell layers (Figure 4F). Serum levels of AMH, LH, T, and ACTH increased, while E2 and FSH levels decreased (Figure 4G), confirming that circadian rhythm disruption induces PCOS in rats. Following ZYBYF treatment, the estrous cycle, body weights, ovarian morphology, and hormonal profiles were significantly restored (Figure 4).

In rats exposed to continuous light, the expression of lncRNA TCONS_00265853 and CLOCK were downregulated, while miR-421-5p expression was upregulated (Figure 5). Post-ZYBYF treatment, the expressions of lncRNA TCONS_00265853 and CLOCK increased, and miR-421-5p expression decreased, suggesting ZYBYF treatment helps normalize this regulatory axis.

In ovarian tissues of rats exposed to continuous light, Prkca mRNA expression decreased, while Il1b, Vegfa, Kdr, Erbb2, and Tgfb1 expression increased (Figure 6A). At the protein level, the expression of p-p38, p-ERK1/2, and p-JNK increased, indicating MAPK pathway activation (Figure 6B). After ZYBYF treatment, Prkca expression was upregulated, while the expression of the other five genes was downregulated (Figure 6A). Concurrently, the phosphorylation levels of p-p38, p-ERK1/2, and p-JNK decreased (Figure 6B).

In rats subjected to continuous light exposure, BAX protein expression increased, Bcl-2 expression decreased, and the apoptosis rate of ovarian cells increased significantly (Figure 7). Post-ZYBYF treatment, BAX expression decreased, Bcl-2 expression increased, and the apoptosis rate of ovarian cells decreased, indicating that ZYBYF has a protective effect against apoptosis in the context of CRD-induced PCOS.