Human MM.1R, ARD, and RPMI-8226 MM cell lines were from the American Type Culture Collection (ATCC). Short tandem repeat (STR) profiling confirmed cell identity, and cultures tested negative for mycoplasma. Cells were maintained in RPMI-1640 medium (01-106-1A, Biological Industries Inc., Beit Haemek, Israel) with 10% fetal bovine serum (A5256701, Gibco, Shanghai, China) at 37 °C in a humidified 5% CO₂ incubator.

MM.1R cells were seeded into 96-well plates and exposed to epirubicin hydrochloride (56390-09-1, Selleck, Shanghai, China) for 48 h. CCK-8 reagent (10 µL/well; C0038, Beyotime Biotechnology, Shanghai, China) was added 4 h before endpoint measurement. Optical density was recorded at 450 nm via microplate reader (Varioskan™ ALF Multimode, Thermo Fisher Scientific Inc., shanghai, China). The inhibitory concentration (IC)50 was subsequently used for microarray analysis, Western blotting, and cell-cycle assays.

MM.1R cells (2 $\times$ 10⁵ per well) were cultured in 6-well plates and treated with epirubicin or vehicle control for 48 h. Three biological replicates per condition were collected. Microarray processing and analysis were performed by Shanghai YBR Biotechnology Co., Ltd.

Total RNA was extracted using TRIzol (15596-018, Solarbio, Shanghai, China), and cDNA synthesis was performed using the Vazyme RT Reagent Kit (P612, Vazyme, Shanghai, China). Reactions were run in triplicate using AceQ SYBR Green Master Mix (Q111-02, Vazyme, Shanghai, China) on a VIIA-7 PCR system (Thermo Fisher Scientific Inc., shanghai, China). Relative expression was calculated by the 2^{-Δ⁢Δ⁢CT} method, normalized to GAPDH. Primer sequences were: CDC20 (Sangon Biotech, China) 5^′-AATGGAGCAGCCTGGGGAATA-3^′ (sense) and 5^′-CGGGCAGAGTGACTGGTCATAT-3^′ (antisense). GAPDH (SangonBiotech, Shanghai, China) 5^′-TGACTTCAACAGCGACACCCA-3^′ (sense) and 5^′-CACCCTGTTGCTGTAGCCAAA-3^′ (antisense). Experiments were repeated $\ge$3 times with technical triplicates.

Full-length CDC20 sequences were cloned into LV-002 and LV-003 vectors (BGI, Shanghai, China). Control vectors were purchased from YBR (Shanghai, China). HEK-293T cells were transfected using Lipofectamine™ 2000 (Invitrogen, Shanghai, China) along with packaging plasmids pMD2.G and psPAX2 (pMD2.G #12259, psPAX2 #12260, Addgene, MA, USA). Viral supernatants were harvested at 48 h and used to transduce MM.1R cells in the presence of 0.1% polybrene (H-8968, Sigma-Aldrich, Shanghai, China).

Immediately following cell harvest, total proteins were isolated using RIPA lysis buffer supplemented with protease inhibitors (R0010, Solarbio, Shanghai, China). Lysates were clarified by centrifugation at 12,000 $\times$g for 15 min at 4 °C. Protein concentrations were quantified using a BCA detection kit (E112-01, Vazyme Biotech, Shanghai, China). Equal amounts of protein (25 µg) were combined with SDS loading buffer and denatured at 95 °C for 5 min, then resolved on 4–12% SDS-polyacrylamide gels (ET12412, ACE, Shanghai, China). Proteins were transferred to PVDF membranes by wet-transfer and subsequently blocked in 5% skim milk. Membranes were incubated overnight at 4 °C with primary antibody against CdC20 (dilution ratio 1:100, SC-13162, Santa Cruz, Shanghai, China), followed by incubation with HRP-conjugated secondary antibodies (dilution ratio 1:500, A0208, Beyotime, Shanghai, China). Bands were visualized by enhanced chemiluminescence using an ECL kit (p10300, New Seme, Shenzhen, China).

After collection, cells were fixed in pre-chilled 70% ethanol at –20 °C overnight. Fixed cells were washed with PBS and stained using a cell-cycle detection kit (KGA9101-50, KeyGen Biotech, Shanghai, China) according to the manufacturer’s protocol. DNA content and distribution across cell-cycle phases were assessed by flow cytometry (FACSCanto™ II, BD Biosciences, USA), and the proportion of cells in each phase was quantified.

Six-week-old female nude mice were obtained from Hangzhouziyuan Experimental Mobility Co., Ltd. Animals received subcutaneous and intravenous injections of 2 $\times$ 10⁶ MM.1R cells expressing control vectors or CDC20-overexpression constructs. Beginning on day 14 after tumor cell implantation, epirubicin was administered intraperitoneally at 3.5 mg/kg. Tumor dimensions were recorded every 2 days. On day 31, mice were euthanized and tumors were excised for measurement of volume and weight. The study followed the guidelines of Zhengzhou University’s Animal Experiment Committee, the Safety and Environmental Management Department, and the Laboratory Animal Welfare and Ethics Committee (Approval No. DW2023078). Experimental protocols were approved by the Research Committee of Zhengzhou University Affiliated Tumor Hospital. All methods adhered to applicable guidelines and regulations, and the research was reported in compliance with the ARRIVE guidelines.

Euthanasia protocol: Based on American Veterinary Medical Association (AVMA) standards, mice were euthanized via CO₂ exposure to ensure humane termination. The surgery performed on the mice requires anesthesia. The anesthetic used is tribromophenol, with a concentration of 1.25%, and the dosage for each mouse is 200 µL. Place the animal in the box and slowly inject CO₂ into the box at a rate of 10–30% of the box’s volume per minute (for this system, a 20% volume flow rate is 5.8 L/min. Adjust the flow rate by turning the left knob). Ensure that the animal is motionless, not breathing, and its pupils are dilated. Turn off the CO₂ cylinder switch and observe for another 2–3 minutes to confirm the animal’s death.

Quality control procedures included assessment of intensity distributions, normalization, sample correlation analysis, and principal-component analysis. Probes with mean signal values below 0.005 were removed. Differentially expressed genes (DEGs) were identified using an empirical Bayes-based linear-model approach to determine p-values, minimizing false-positive rates. Significant genes were defined using $|$fold change$|$ $\ge$1.3 and p 0.05. Functional enrichment and visualization of DEG clusters were performed using Metascape (http://metascape.org/) [7].

Four publicly available datasets were utilized for this study, including GSE24080, GSE4204, GSE57317, and GSE9782. The details of these datasets were as follows: GSE24080: Contributed by the Myeloma Institute for Research and Therapy at the University of Arkansas for Medical Sciences, comprising 340 training and 214 validation samples; GSE4204: Consisted of pre-treatment bone-marrow aspirates from MM patients; GSE57317: Included gene-expression profiles from 55 previously treated MM patients; GSE9782: Contained data from a bortezomib clinical trial cohort (528 samples), representing proteasome-inhibitor-treated MM.

SPSS software (SPSS Statistics 27, IBM, NY, USA) or Microsoft Excel 2010 (Microsoft, WA, USA) were used for all analyses. Results are expressed as mean $\pm$ SD. Student’s t-tests or one-way ANOVAs were applied when appropriate. Significance thresholds were: *p 0.05, **p 0.01, and ***p 0.001.

Gene-expression profiling was performed to compare MM cells before and after epirubicin hydrochloride treatment. A total of 115 transcripts were significantly upregulated, whereas 25 genes were significantly downregulated (Fig. 1A,B). Functional enrichment analysis of upregulated genes (log FC $>$1) using Metascape indicated enrichment in pathways associated with p53-mediated signaling, DNA-damage responses, cytokine-receptor interactions, regulation of MAPK cascades, TP53-related networks, vitamin-D-receptor signaling, and immune-activation processes, as well as malignant pleural mesothelioma-related pathways (Fig. 1C). Conversely, downregulated genes (log FC –1) were predominantly linked to mitotic progression, nuclear division, regulation of cytokinesis and small-GTPase signaling, kinetochore assembly, microtubule-related transport, phosphorylation of Emi1, cellular senescence, synaptic-plasticity regulation, positive regulation of transferase activity, and modulation of cell-growth programs (Fig. 1D).

Fig. 1.

Differentially expressed gene analysis. (A) Volcano plot showing differentially expressed genes. (B) Heatmap of differentially expressed genes. (C) Enrichment analysis of upregulated DEGs. (D) Enrichment analysis of downregulated DEGs. (E) Enrichment analysis of significantly down-regulated genes (log FC –1.5). DEGs, differentially expressed genes.

To elucidate the molecular pathways and biological functions modulated by epirubicin in MM cells, genes exhibiting marked expression shifts were subjected to enrichment analysis using Metascape. Functional clustering focused on transcripts that showed a pronounced increase (log FC $>$2). This analysis demonstrated that 12 genes were strongly upregulated, predominantly associated with the enhancement of apoptotic signaling. Representative genes included ITGAM (log FC = 3.746), GADD45G (log FC = 2.743), ATF3 (log FC = 2.563), PHLDA3 (log FC = 2.151), PLK2 (log FC = 2.038), and PMAIP1 (log FC = 2.012). Conversely, six transcripts exhibited substantial downregulation (log FC –1.5), primarily participating in cell cycle regulation and the PID PLK1 cascade (Fig. 1E). Notably, among these cell cycle‒associated genes were LMNB1 (log FC = –1.589), KIF20A (log FC = –1.693), CCNB1 (log FC = –1.787), and CDC20 (log FC = –2.409), indicating an epirubicin-driven suppression of mitotic processes

CDC20 has previously been identified as one of the most highly upregulated chromosomal instability-related regulators in MM, promoting proliferation and contributing to therapeutic resistance [8]. High-risk MM cohorts demonstrate elevated CDC20 expression, and enhanced CDC20 activity correlates with worse clinical outcomes [9]. Functional suppression of CDC20 has been reported to induce metaphase arrest and apoptosis in MM models, supporting the concept that the APC/C-CDC20 axis represents a promising therapeutic target, particularly in aggressive disease subsets [9]. Consistent with these observations, our microarray profiling identified CDC20 as the most significantly suppressed transcript following epirubicin treatment, raising the possibility that individuals with inherently high CDC20 expression could derive enhanced benefit from epirubicin-based therapy.

Although previous work has linked elevated CDC20 levels to inferior survival in MM, heterogeneity across molecular subgroups remains poorly explored. Identifying which patient groups derive the greatest benefit from therapeutic modulation of CDC20 is a major priority in precision oncology. To refine this association, we examined survival outcomes in a cohort of 1416 MM patients integrated through the KMplot platform, incorporating four datasets: GSE24080 [10], GSE4204 [11], GSE57317 [12], and GSE9782 [13, 14]. Across the full cohort, increased CDC20 expression was significantly associated with reduced overall survival (OS), event-free survival (EFS), and post-progression survival (PPS). Similar trends were observed among male patients, whereas in female patients, only the PPS correlation remained but did not reach statistical significance (Fig. 2). When stratified by immunoglobulin subtype, elevated CDC20 expression corresponded to shortened PPS in the serum free light chain (FLC) group and was strongly linked to inferior OS, EFS, and PPS in immunoglobulin A (IgA) and immunoglobulin G (IgG) subtypes. Interestingly, in the light chain subtype subset, high CDC20 expression was paradoxically associated with improved OS and PPS outcomes (Fig. 3).

Fig. 2.

OS, EFS, and PPS for patients stratified based on sex and CDC20 expression. OS, overall survival; EFS, event-free survival; PPS, post-progression survival.

Fig. 3.

OS, EFS, and PPS for patients stratified according to CDC20 expression and myeloma subtype. OS, overall survival; EFS, event-free survival; PPS, post-progression survival; FLC, serum free light chain; IgA, immunoglobulin A; IgG, immunoglobulin G.

Given CDC20’s critical role in coordinating mitotic progression, we next analyzed cell cycle distribution prior to and following epirubicin exposure. As illustrated in Fig. 4A, epirubicin treatment caused a marked accumulation of cells in the G2/M phase, with MM.1R cells increasing from 20.1% to 46.4% and ARD cells rising from 43.6% to 57.9%. Parallel assessments using qPCR and western blotting confirmed a substantial reduction in CDC20 expression at both mRNA and protein levels following treatment (Fig. 4B,C). These findings collectively indicate that epirubicin-induced G2/M arrest occurs concurrently with transcriptional and translational downregulation of CDC20.

Fig. 4.

Epirubicin induces G2/M phase arrest in MM cells. (A) Cell cycle profiles before and after drug exposure. (B) Changes in CDC20 mRNA expression following treatment. (C) Alterations in CDC20 protein levels post-treatment. Data are presented as means $\pm$ SD; *p 0.05, ***p 0.001, ****p 0.0001, nd, no difference.

To evaluate whether MM cells with heightened baseline CDC20 levels exhibit increased vulnerability to epirubicin, we established CDC20-overexpressing cell lines and assessed drug sensitivity using both in vitro and in vivo platforms. Colony formation assays demonstrated a pronounced reduction in clonogenic survival in the CDC20-overexpression group relative to controls when treated with epirubicin (Fig. 5A). Consistent with this, CCK-8 analysis revealed a lower IC50 in CDC20-overexpressing cells compared with controls, indicating enhanced drug sensitivity (Fig. 5B). EdU labeling assays further supported this trend, showing significantly decreased proliferation rates under epirubicin challenge (Fig. 5C). In vivo xenograft studies corroborated these findings, with tumors derived from CDC20-high cells exhibiting reduced volume and weight after epirubicin administration relative to control tumors (Fig. 5D–F).

Fig. 5.

High levels of CDC20 expression make cells more sensitive to epirubicin. (A) Colony-forming capacity under drug exposure. (B) IC50 comparison between control and CDC20-overexpressing groups after 48 hours. (C) Proliferation rates determined after 48 hours of treatment. Scale bar = 50 μm. (D,E) Tumor volume changes in wild-type and overexpression groups treated with epirubicin. (F) Tumor weight comparisons between wild-type and overexpression groups following epirubicin treatment. Data are means $\pm$ SD; *p 0.05, **p 0.01, nd, no difference.