Targeting the mevalonate pathway sensitizes multiple myeloma cells to bortezomib through induction of mitochondrial metabolic disruption

Abstract

Background

Multiple myeloma (MM) remains an incurable hematologic malignancy with bortezomib resistance representing a major therapeutic challenge. The mevalonate (MVA) pathway, a crucial metabolic cascade involved in cholesterol synthesis, has emerged as a potential target for cancer therapy.

Methods

In this study, we first investigated the correlations between the MVA pathway and MM progression. Leveraging public databases, transcriptome sequencing, mass cytometry, as well as in vitro and in vivo experimental models, we further systematically evaluated the therapeutic efficacy and underlying mechanisms of statins, a class of clinically available MVA pathway inhibitors, for the treatment of MM.

Results

Using microarray and sequencing datasets, we found that the MVA pathway is dysregulated in MM and correlated with poor prognosis, as core genes in this pathway were significantly dysregulated in MM patients. Statins directly inhibit MM cell viability, induce apoptosis, and up-regulate apoptosis-associated proteins. These effects of statins are associated with the activation of the ERK pathway. They also induce dysregulation of metabolism, oxidative stress, and autophagy in MM cells. Moreover, the combination of statins and bortezomib exerts a synergistic anti-MM effect both in vitro and in vivo.

Conclusions

Our preclinical data suggest that the MVA pathway can serve as a therapeutic target for MM and the combination of statins, widely used and safe drugs in the clinic, with bortezomib offers a promising strategy to overcome bortezomib resistance and improve therapeutic outcomes for MM patients.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12964-026-03076-8.

Background

Despite remarkable therapeutic advancements, multiple myeloma (MM) persists as an incurable malignancy of plasma cells. The advent of proteasome inhibitors (PIs), such as bortezomib, and immunomodulatory drugs has undeniably enhanced patient survival. However, the twin challenges of drug resistance and relapse continue to loom large in clinical practice, casting a shadow over treatment efficacy [1, 2]. Current standard treatment regimens typically involve the combination of PIs with other agents. Regrettably, the emergence of refractory disease often curtails their effectiveness, leaving patients with limited options [3, 4]. As a result, there is an urgent and pressing need for the development of novel therapeutic strategies [5]. Recent research has shed light on the metabolic vulnerabilities of MM cells, opening up new avenues for therapeutic intervention. For example, the mitochondrial fission factor (MFF) has been identified as a key driver of mitochondrial fragmentation in aggressive MM subtypes. Inhibiting MFF has been shown to effectively reduce tumor growth, highlighting the potential of targeting metabolic pathways to surmount treatment resistance [6]. This growing body of knowledge provides a strong and persuasive rationale for exploring the mevalonate (MVA) pathway-a critical metabolic cascade-as a potential therapeutic target in MM. By delving into this pathway, we may uncover new treatment approaches that could revolutionize the management of this challenging disease.

The MVA pathway stands as a pivotal metabolic route, playing a fundamental role in cholesterol biosynthesis and the generation of essential isoprenoid intermediates [7]. This intricate metabolic cascade gives rise to key intermediates, namely farnesyl pyrophosphate and geranylgeranyl pyrophosphate. These intermediates are indispensable as they facilitate the post-translational modification of small GTPases, such as Ras and Rho [8]. These modifications are of utmost significance, serving as critical regulators of cellular proliferation, survival, and migration. Beyond its well-established function in maintaining cholesterol homeostasis, the MVA pathway has emerged as a central player in cancer biology [7]. It actively supports multiple oncogenic processes by enabling the prenylation of proteins that are integral to growth and survival signaling. By precisely regulating the synthesis of isoprenoids, the MVA pathway is crucial for the post - translational modification of small GTPases involved in cell growth and survival. Consequently, it has become an appealing target for cancer therapy [9, 10].

The MVA pathway has emerged as a captivating area of research in MM. Although direct evidence regarding its role in MM is still in the process of being established, findings from other cancer types offer compelling reasons to explore it further. MM cells presumably rely on this pathway for cholesterol homeostasis and protein prenylation, which are crucial for growth and survival signaling within the bone marrow microenvironment [11]. Statins, which act by inhibiting HMG-CoA reductase, the rate-limiting enzyme of the MVA pathway, have been extensively investigated for their capacity to reduce cholesterol synthesis and modulate various cellular processes, including proliferation, apoptosis, and differentiation [12]. In the field of cancer biology, statins have captured attention as potential adjunctive therapies because they can target key oncogenic signaling pathways and disrupt the essential molecular machinery required for tumor growth [13, 14]. However, despite these encouraging results, the exact role of the MVA pathway in MM and whether inhibiting it in combination with proteasome inhibition can present a feasible therapeutic strategy remains ambiguous.

Against this backdrop, the current study aims to assess the potential of combining statins with bortezomib to improve the treatment efficacy in MM. We particularly focus on unraveling the molecular mechanisms that underlie the synergistic effects of this combination therapy. We hypothesize that simultaneously inhibiting the MVA pathway and proteasomal activity will further trigger apoptosis in MM cells, thereby overcoming bortezomib resistance and enhancing the therapeutic outcomes for MM patients.

Methods

Cell culture

Human MM cell lines, including RPMI-8226, MM1S, and U266, were purchased from the China Center for Type Culture Collection (Wuhan, China) and maintained in RPMI 1640 medium (Invitrogen, Carlsbad, CA, USA), supplemented with 10% fetal bovine serum (FBS, Sigma-Aldrich, St. Louis, MO, USA), 2 mM L-glutamine (VivaCell, Shanghai, China), and 100 U/mL penicillin/streptomycin (VivaCell). LP-1 cells were purchased from Cellcook (Guangzhou, China) and cultured in IMDM (Invitrogen), supplemented with 10% FBS, 2 mM L-glutamine, and 100 U/mL penicillin/streptomycin. Bone marrow (BM) samples were collected from newly diagnosed MM patients after obtaining informed consent, and the studies were conducted in accordance with the Declaration of Helsinki. All human sample-related studies were approved by the Ethics Committee of Sun Yat-sen University Cancer Center and the Medical Ethics Committee of Guangzhou Medical University. BM mononuclear cells (BMMCs) were isolated using Lymphoprep (Stemcell Technologies, Inc., Vancouver, BC, Canada) and cultured in RPMI 1640 medium supplemented with 10% FBS, 2 mM L-glutamine, and 100 U/mL penicillin/streptomycin.

Reagents and antibodies

Bortezomib, lovastatin, simvastatin, rosuvastatin, atorvastatin, BAPTA-AM, N-Acetylcysteine, SCH772984, and temuterkib were purchased from Selleck Chemicals (Houston, TX, USA). Zombie Violet Fixable Viability Kit was purchased from Biolegend (San Diego, CA, USA). Mitochondrial Calcium Assay Kit with Rhod-2 AM (Rhod-2 AM), and Mitochondrial Superoxide Assay Kit with MitoSO Red were purchased from Beyotime (Shanghai, China). CellROX Deep Red was purchased from Invitrogen. Antibodies against glyceraldehyde-3‐phosphate dehydrogenase (GAPDH, 5174 S, 1:1000), β-actin(3700 S, 1:1000), Caspase-3 (9668T, 1:1000), poly ADP-ribose polymerase (PARP, 9532 S, 1:1000), p-ERK1/2 (4370 S, 1:1000), ERK1/2 (4695 S, 1:1000), and LC3‐I/II (12741 S, 1:1000), P62 (8025 S, 1:1000) were purchased from Cell Signaling Technology (Danvers, MA, USA). IRDye 800CW or 680RD goat anti‐rabbit/mouse IgG secondary antibodies (1:10000) were purchased from LI‐COR Biosciences (Lincoln, NE, USA).

Cell viability assay

MM cells (5 × 10⁵ cells/mL) were treated with a series of concentration gradients of lovastatin, simvastatin, rosuvastatin, or atorvastatin, in the absence or presence of bortezomib at the indicated concentration for 48 h and the cell viability was identified by the Cell Titer glo Luminescent Viability assay (Promega, Madison, WI, USA). All viability assays were determined in triplicate and replicated in at least three independent experiments.

Apoptosis analysis

MM cells treated with the indicated drugs under the given conditions were harvested and labeled with Annexin V-FITC/APC (Biolegend) and 7‐Aminoactinomycin D (7‐AAD, Biolegend). For detecting primary cell apoptosis, BMMCs were treated with lovastatin for indicated hours and stained with anti-human CD38-pacific blue, anti-human CD45-PE, and Annexin V-FITC. The apoptotic cells in cell lines or primary MM (CD45⁻CD38^high) cells were determined using a CytoFLEX flow cytometry (Beckman Coulter, Brea, CA, USA) and analyzed with FlowJo software (TreeStar, Ashland, OR, USA).

Measurement of the mitochondrial calcium and oxidative stress

MM cells(1 × 10⁶/mL) were treated with the indicated statins at the indicated concentrations for 48 h and stained with Zombie Violet dye. After washing with PBS for 3 times, these cells were stained with Rhod-2 AM, MitoSO Red, or CellROX Deep Red according to the manufacturer’s instruction. The mitochondrial calcium and oxidative stress of MM cells were detected using flow cytometry on a CytoFLEX and analyzed with FlowJo software.

RNA sequencing and analysis

Total RNA was extracted from RPMI-8226 cells treated with simvastatin or lovastatin. The RNA sequencing and analysis were conducted by IGE Biotechnology Co., Ltd. (Guangzhou, China). Heatmaps, the GSEA, GO and KEGG analyses were conducted by IGE Biotechnology Co., Ltd. Figures for GO and KEGG were generated on chiplot.online.

Cell staining for mass cytometry

Mass cytometry and data analysis were performed as described previously [12]. Briefly, MM cells were treated with lovastatin or simvastatin for 48 h and incubated with IdU (50 µM) for another 30 min. Immediately after the incubation, the cells were fixed using Fix I buffer (Standard BioTools) and barcoded separately using the Cell-ID 20-Plex Pd Barcoding Kit (Standard BioTools) according to the manufacturer’s instructions. After barcoding, all the cells were washed with cell staining buffer and combined into a single pool. The pooled cells were first stained with metal isotope-tagged antibodies targeting surface proteins and then with antibodies against intracellular proteins (Table S1) after permeabilization. Once the antibody staining was completed, the cells were resuspended in Fix & Perm Buffer (Standard BioTools) supplemented with 125 nM Cell-ID Intercalator-Ir (Standard BioTools). The cells were then stored at 4 °C overnight for further processing.

Mass cytometry data acquisition and analyses

The stained samples were washed with ultrapure water for three times and subsequently resuspended in ultrapure water supplemented with 15% EQ Four Element Calibration Beads (Standard BioTools). All samples were then introduced into a Helios mass cytometer (Standard BioTools) for data acquisition. CyTOF software 7.0 (Standard BioTools) was employed to perform normalization and debarcoding processes on the acquired data. These data were uploaded to cytobank.cn for Uniform Manifold Approximation and Projection (UMAP) analysis after excluding cell debris and doublets.

Western blot analysis

MM cell lysates were prepared using RIPA buffer (Beyotime) supplemented with a proteinase inhibitor cocktail (Beyotime). The protein concentration in the lysate was determined using the Pierce BCA Protein Assay Kit (Thermo Scientific, Waltham, MA, USA). The lysate, mixed with loading buffer, was then loaded onto SDS-PAGE gels and subjected to electrophoresis. Afterward, the proteins were transferred from the gel to PVDF membranes for immunoblotting, and the membranes were blocked using QuickBlock Blocking Buffer (Beyotime). Following this, the membranes were incubated with primary antibodies, followed by fluorescent dye-conjugated secondary antibodies. The resulting protein bands were visualized and captured using an Odyssey CLx imaging system (LI-COR Biosciences).

In vivo study

Female NOD-Prkdcscid IL2rgtm1/Bcgen (B‐NDG) mice at six‐to eight‐week‐old were purchased from Biocytogen (Beijing, China) and were housed and treated following conditions approved by the Institutional Animal Care and Use Committee of Guangzhou Medical University. A total of 5 × 10⁶ RPMI-8226 cells expressing luciferase (RPMI-8226‐Luc) were intravenously injected into B‐NDG mice to generate a human myeloma xenograft mouse model. Bortezomib and/or simvastatin were administered by intraperitoneal injection at the indicated times. For bioluminescence imaging, mice were anesthetized with 1–3% isoflurane via inhalation and intraperitoneally injected with 150 mg/kg of D-luciferin (PerkinElmer, Waltham, MA, USA) and bioluminescence signal was captured using a IVIS Lumina III in vivo imaging System (PerkinElmer). Bioluminescence intensity was quantified using a Living Images software (PerkinElmer). Hindlimb paralysis was used as the endpoint criterion. Animals meeting these criteria were humanely euthanized.

Statistical analysis

The Shapiro-Wilk test was applied to assess the normality of the data distribution. If the data followed normal distributions, a parametric test (Student’s t-test) was used to evaluate the statistical significance between two groups. In cases where the data were not normally distributed, a nonparametric test (Mann-Whitney U test) was employed. For comparisons involving multiple groups, one-way ANOVA followed by Tukey’s post-hoc test was conducted. All analyses were performed using GraphPad Prism software (Version 10, GraphPad Software, La Jolla, CA, USA). The error bars represent the mean ± standard deviation (SD) or standard error of the mean (SEM), as specified. Statistical significance was defined as p < 0.05.

Results

The MVA pathway is dysregulated in MM and correlated with poor prognosis

To identify critical therapeutic targets for MM, we investigated the clinical relevance of 12 core genes, including ACAT1, ACAT2, FDPS, GGPS1, HMGCR, HMGCS1, HMGCS2, IDI1, IDI2, PMVK, MVK, and MVD, in the MVA pathway using publicly available microarray datasets GSE5900 (Fig. 1A). Among them, the levels of ACAT1, FDPS, GGPS1, HMGCR, HMGCS1, HMGCS2, IDI1, PMVK, and MVK were significantly dysregulated in CD138⁺ plasma cells obtained from MM or monoclonal gammopathy of undetermined significance (MGUS) patients compared to those from healthy donors (HD) (Figs. 1B and S1A), suggesting continuous activation of the MVA pathway during myeloma progression. Additionally, Kaplan-Meier survival curves showed that high expression of 8 core genes (ACAT2, FDPS, GGPS1, HMGCR, HMGCS1, PMVK, IDI1, IDI2, and MVK) was significantly associated with poor prognosis in MM patients (Figs. 1C and S1B). Among these, FDPS and HMGCS1 had the strongest association with adverse outcomes, with hazard ratios (HR) of 2.161 and 2.251, respectively (both P < 0.0001), confirming them as robust poor prognostic markers. High expression of ACAT2 (HR = 1.704, P = 0.0009), HMGCR (HR = 1.656, P = 0.0017), and PMVK (HR = 1.807, P = 0.0002) was also linked to significantly shortened survival. These results provide the direct clinical evidence that the MVA pathway is abnormally activated in myeloma progression and that high expression of its core genes is closely associated with poor prognosis.

Fig. 1.

Expression and prognostic significance of key genes in the MVA pathway in MM progression. A A heatmap representing the expression of 12 key genes in the MVA pathway in CD138⁺ plasma cells from HD, patients with MGUS, and MM patients. B Bar plots showing the expression of the indicated genes in CD138⁺ plasma cells from HD (n = 22), patients with MGUS (n = 44), and MM patients (n = 12). C Overall survival according to high and low expression of the indicated genes in plasma cells was determined in MM patients (MMRF CoMMpass, n = 785) using the Log-rank (Mantel-Cox) test (cut-off point used is the median value). Error bar, mean ± SD. *, p < 0.05; **, p < 0.01; ****, p < 0.0001

Statins directly inhibit MM cell viability and induce apoptosis

Statins are inhibitors of the MVA pathway, functioning by blocking HMG-CoA reductase to reduce cholesterol synthesis. Since the MVA pathway is activated in MM cells, we further validated the in vitro anti-myeloma effects of statins. Both lovastatin and simvastatin exhibited concentration-dependent inhibitory effects on the viability of RPMI-8226, LP-1, and U266 cells (Figs. 2 A, 2B, S2A, and S2B). Additionally, flow cytometry analysis confirmed that statins significantly induced apoptosis in different myeloma cells (Figs. 2 C-F, S2C and S2D). Other statins, such as atorvastatin and rosuvastatin also inhibited the MM cell viability and induced cell apoptosis (Figures S2E-H). Furthermore, the apoptosis of primary MM cells obtained from patients was also induced by statins (Figure S2I). These results support the inhibition of the MVA pathways by statins as an effective anti-MM strategy.

Fig. 2.

Inhibitory effects of statins on MM cell viability and survival. A-B RPMI-8226 and LP-1 cells were cultured with or without lovastatin or simvastatin at the indicated concentration for 48 h and the cell viability was measured (n = 3). C-F RPMI-8226 and LP-1 cells were cultured with or without lovastatin or simvastatin at the indicated concentration for 48 h. The apoptotic cells were determined using flow cytometry (n = 3). LOV, lovastatin; SIM, simvastatin; Apop, apoptotic. Error bar, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Statins upregulate apoptosis-associated proteins and activate the ERK pathway in MM Cells

To clarify the molecular mechanism by which statins induce apoptosis, we assessed the expression of proteins involved in the apoptotic and ERK signaling pathways. Treatment with lovastatin and simvastatin resulted in a concentration-dependent increase in the cleavage of caspase-3 and PARP, indicating that statins induce apoptosis through the activation of the caspase-dependent intrinsic apoptotic pathway (Figs. 3 A–D). Additionally, statin treatment significantly enhanced ERK1/2 phosphorylation in MM cells (Figs. 3E–H). Notably, pharmacological inhibition of ERK1/2 using two distinct inhibitors, SCH772984 and temuterkib, attenuated statin-induced apoptosis in MM cells (Figures S3A and S3B). These findings collectively indicate that activation of the ERK1/2 signaling pathway constitutes a partial mechanistic basis for statin-mediated apoptosis in MM cells.

Fig. 3.

The regulatory effects of statins on apoptosis pathways and the ERK signaling pathway in MM cells. RPMI-8226 and LP-1 cells were cultured with or without lovastatin or simvastatin at the indicated concentration for 48 h (n = 3–4). A-D The protein levels of cleaved caspase-3, caspase-3, cleaved PARP and PARP were detected using western blot. β-actin was included as a loading control. E-H The protein levels of p-ERK and ERK were detected using western blot. GAPDH was included as a loading control. Cl, Cleaved; LOV, lovastatin; SIM, simvastatin. Error bar, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Differentially expressed genes regulated by statins are enriched in lipid metabolism and apoptosis pathways

To systematically decipher the molecular regulatory network of statins, we performed differentially expressed gene (DEG) screening and functional enrichment analysis via transcriptome sequencing. Lovastatin significantly upregulated 533 genes and downregulated 255 genes in MM cells (Fig. 4A). Lovastatin mainly regulated genes involved in the important GO or KEGG pathways, inducing the defense response to viruses, cholesterol biosynthetic process, and steroid biosynthesis (Figs. 4B-D). Similarly, simvastatin significantly induced up-regulation of 414 genes and down-regulation of 163 genes in MM cells (Fig. 4E). Simvastatin also mainly regulated genes involved in the cholesterol biosynthetic process and steroid biosynthesis, etc. (Figs. 4F-H).

Fig. 4.

Screening of differentially expressed genes and functional enrichment analysis in MM cells treated with statins. A Heatmap of differentially expressed genes (DEGs, |LogFC| > 1 and P < 0.05) in the vehicle vs. lovastatin group. B GO functional enrichment analysis of DEGs in the vehicle vs. lovastatin group. Different colors indicate BP, MF, and CC categories, with core enriched terms and the number of genes labeled. C-D Bubble plots of GO (C) and KEGG (D) pathway enrichment for DEGs in the vehicle vs. lovastatin group. E-H Heatmap (E), GO enrichment analysis (F, G), and KEGG pathway enrichment analysis (H) of DEGs in the vehicle vs. simvastatin group

Various MsigDB pathways, such as genes up-regulated in response to IFN-γ and genes up-regulated in response to IFN-α proteins, were significantly enriched in and positively correlated with both lovastatin and simvastatin treatment. We further focused on differentially expressed genes co-regulated by lovastatin and simvastatin to identify their core synergistic mechanisms. The GO/KEGG enrichment analysis of co-differentially expressed genes showed that co-upregulated genes were enriched in the “sterol biosynthetic process”, “defense response to virus”, “steroid biosynthesis”, and “terpenoid backbone biosynthesis” pathways (Fig. 5C and E), further confirming that lipid metabolism inhibition and antiviral immune activation are the primary modes of action. In contrast, co-downregulated genes were enriched in “amide transport”, “intracellular lipid transport”, “cornified envelope formation”, and “ABC transporters” (Fig. 5D and F), indicating that statins simultaneously inhibit both cholesterol synthesis and transport, thus dual-blocking lipid metabolism.

Fig. 5.

Functional enrichment and pathway analysis of differentially expressed genes co-regulated by statins. A-B Gene Set Enrichment Analysis (GSEA) of DEGs in the vehicle vs. lovastatin (A) and vehicle vs. simvastatin (B) groups. C-D Bubble plots of GO functional enrichment for co-upregulated (C) and co-downregulated (D) DEGs by lovastatin (LOV) and simvastatin (SIM). E-F Bubble plots of KEGG pathway enrichment for co-upregulated (E) and co-downregulated (F) DEGs by LOV and SIM

Statins induce metabolism dysregulation in MM cells

Based on the sequencing data, we sought to determine whether statins affect the metabolism in MM cells. Mass cytometry was used to measure the levels of key proteins that regulate signaling pathways and glucose and energy metabolism in statin-treated MM cells at the single-cell level. Using UMAP algorithm analysis, a heterogeneity of MM cell lines was clearly observed (Fig. 6A). According to the levels of cleaved caspase-3, Ki-67, and IdU in the RPMI-8226 cells, four clusters were identified (Figure S4A). Statins treatment significantly induced an increase of cleaved caspase-3⁺ cells (Fig. 6C), indicating the enhancement of apoptosis. Meanwhile, statins significantly reduced the proportion of IdU⁺ cells (Fig. 6C), suggesting the inhibition of cell proliferation. The proportion of IdU⁻Ki-67⁺ cells that mainly represent proliferation preparation cells was also decreased by statins (Fig. 6C). Additionally, statins also significantly regulated the levels of phosphorylated proteins of key signaling pathways, as well as the levels of key enzymes of metabolism in these four clusters (Fig. 6D and E, S4B, and S4C). Similarly, statins increased the apoptosis, reduced the proliferation, and dysregulated the signaling pathways and metabolism in LP-1 cells (Figs. 6F-I and S4D-G). All these single-cell data suggested that statins also significantly trigger dysregulation of signaling pathways and metabolic pathways in MM cells in addition to inducing cell apoptosis.

Fig. 6.

Statins regulate metabolism in MM cells. A-E Multiple metabolism-related proteins in RPMI-8226 cells treated with lovastatin (LOV, 50 µM), simvastatin (SIM, 50 µM), or vehicle for 48 h were analyzed using mass cytometry (n = 4). A UMAP maps colored by the normalized expression of the indicated markers in RPMI-8226 cells. B UMAP maps colored by four cell populations after clustering. C Bar plots showing the frequencies of the indicated clusters in RPMI-8226 cells. D and E Bar plots showing the mean transformed ratio of the indicated proteins in the indicated RPMI-8226 cell clusters. F-I Multiple metabolism-related proteins in LP-1 cells treated with LOV (100 µM), SIM (50 µM), or vehicle for 48 h were analyzed using mass cytometry. F UMAP maps colored by four cell populations after clustering in LP-1 cells. G Bar plots showing the frequencies of the indicated clusters in LP-1 cells. H and I Bar plots showing the mean transformed ratio of the indicated proteins in the indicated LP-1 cell clusters. Cl, Cleaved; LOV, lovastatin; SIM, simvastatin. Error bar, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Statins induce oxidative stress and autophagy dysregulation in MM Cells

Guided by bioinformatics analysis highlighting the “response to reactive oxygen species”(Figure S5A), we validated key indicators related to oxidative stress. After treatment with lovastatin or simvastatin, we observed a significant increase in both intracellular total reactive oxygen species (ROS) and mitochondrial ROS in all and living MM cells (Figs. 7A-D, S5B, and S5C). The intracellular Ca²⁺ concentration, measured by Rhod2-AM staining, also increased by 1.5-2 folds in statin-treated MM cells (Fig. 7E and F, and S5D). Furthermore, treatment with the ROS scavenger N-acetylcysteine (NAC) and the intracellular calcium chelator BAPTA-AM significantly attenuated statin-induced apoptosis in MM cells (Figures S5E and S5F). These findings suggest that oxidative stress and dysregulated calcium homeostasis contribute at least partially to statin-mediated cytotoxicity in MM cells. Moreover, the gene set enrichment analysis (GSEA) demonstrated that autophagy was also regulated by statins (Figure S5G). Both lovastatin and simvastatin increased the LC3-II/LC3-I ratio and increased the p62 level (Figs. 7G-J). Through transcriptomic sequencing analysis, we found that the mRNA level of p62 was up-regulated (Figure S5H), and the degree of this up-regulation was consistent with the increase in p62 protein level. This result indicates that the accumulation of p62 is primarily driven by enhanced synthesis. When combined with the elevated LC3-II/LC3-I ratio, our results provide compelling evidence that statins promote autophagosome formation while disrupting autophagic flux. These findings confirm the involvement of oxidative stress and autophagy dysregulation in statin-induced mitochondrial dysfunction in MM cells.

Fig. 7.

The regulatory effects of statins on oxidative stress and autophagy in MM cells.A-F RPMI-8226 and LP-1 cells were treated with LOV, SIM, or vehicle for 48 h (n = 3). A-B After CellROX Deep Red staining, intracellular reactive oxygen species (ROS) were determined using flow cytometry.C-D After MitoSOX Red staining, mitochondrial ROS were determined using flow cytometry. E-F The mitochondrial calcium influx was determined using flow cytometry after Rhod-2 AM staining. G-J RPMI-8226 and LP-1 cells were treated with LOV, SIM, or vehicle for 48 h. G-H The protein levels of LC3-Ⅰ and LC3-Ⅱ were detected using western blot (n = 4). GAPDH was included as a loading control. I-J The protein level of p62 was detected using western blot (n = 6 or 8). GAPDH was included as a loading control. LOV, lovastatin; SIM, simvastatin. Error bar, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Combination of statins and bortezomib exerts a synergistic anti-MM effect

To optimize clinical treatment strategies, we investigated the combined effect of statins and bortezomib. Co-treatment with low concentrations of lovastatin (12.5 µM) or simvastatin (6.25 µM) and bortezomib resulted in a significantly higher cell viability inhibition rate compared to bortezomib alone, indicating a synergistic effect (Fig. 8A and B). Additionally, the proportion of late apoptotic and dead cells was significantly increased in the combination group in RPMI-8226 and LP-1 cells (Figs. 8C-E). Furthermore, the combination treatment led to significantly higher expression levels of cleaved-caspase-3 and cleaved-PARP compared to the bortezomib-only group, suggesting an enhanced activation of the apoptotic pathway (Figs. 8F-H). Additionally, statins also enhanced the phosphorylation of ERK1/2 in the absence of bortezomib (Figures S6A and S6B). These results suggest statins as potent adjuvants to enhance the anti-MM efficacy of bortezomib.

Fig. 8.

The enhanced anti-MM effect of statins combined with bortezomib on MM cells. RPMI-8226 and LP-1 cells were treated with lovastatin (LOV, 12.5 µM), simvastatin (SIM, 6.25 µM) or vehicle in the absence or presence of bortezomib (BTZ) at the indicated concentration for 48 h. A-B The cell viability was measured (n = 3). C-E The apoptotic cells were determined using flow cytometry (n = 3). F-H The protein levels of cleaved caspase-3, caspase-3, cleaved PARP and PARP were detected using western blot (n = 4–6). β-actin was included as a loading control. LOV, lovastatin; SIM, simvastatin; BTZ, bortezomib; Apop, apoptotic. Error bar, mean ± SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Statins enhance bortezomib-inhibited growth of MM xenografts in vivo

To confirm the anti-MM activity of statins in vivo, we established an orthotopic MM model by injecting RPMI-8226-Luc cells, which express luciferase, into B-NDG mice via tail vein injection. Mice were treated with vehicle or 20 mg/kg simvastatin (SIM) daily via intraperitoneal injection on days 20 after MM implantation (Figs. 9A). MM progression was monitored using bioluminescence imaging at multiple time points (Fig. 9A and B). The results indicate that mice treated with SIM did not exhibit a significant reduction in MM growth compared to the vehicle group (Fig. 9B). However, combination therapy with simvastatin and bortezomib showed a more pronounced inhibitory effect on MM growth compared to the bortezomib-only group (Fig. 9C). Furthermore, at 78 days post-inoculation, the total flux in the combination group was significantly lower than that in the bortezomib-only group, confirming that combination therapy effectively reduces in vivo MM burden (Fig. 9D). These findings suggest that the combination of statins and bortezomib significantly enhances the therapeutic efficacy in vivo.

Fig. 9.

In vivo anti-MM effects of statins in combination with bortezomib. A and B B-NDG mice were intravenously inoculated with RPMI-8226-Luc cells. Twenty days after inoculation, these mice were randomly divided into two groups and intraperitoneally injected with 20 mg/kg simvastatin (SIM, n = 5) or vehicle (n = 5) once daily. A The distribution of MM cells in these mice was measured at the indicated days after inoculation using a living imaging system. B The total flux in each mouse at the indicated days after inoculation was determined. C-D B-NDG mice were intravenously inoculated with RPMI-8226-Luc cells. Three weeks after inoculation, these mice were intraperitoneally injected with 20 mg/kg SIM (n = 4) or vehicle (n = 3) once daily. Two weeks later, these mice were intraperitoneally injected with 0.5 mg/kg bortezomib three times a week. C The distribution of MM cells in these mice was measured using a living imaging system. D The total flux in each mouse at the indicated days after inoculation was determined. E Schematic diagram showing how targeting the MVA pathway by statins facilitates MM treatment through regulating autophagy and redox homeostasis. The figure was created with BioRender.com/vom1js8. Error bar, mean ± SEM. *, p < 0.05; **, p < 0.01; ****, p < 0.0001

Discussion

MM is a genetically heterogeneous hematological malignancy characterized by clonal plasma cell proliferation, and despite advances in proteasome inhibitors, immunomodulatory drugs, and CAR-T cell therapy, most patients eventually develop refractory disease or relapse [15, 16]. Metabolic reprogramming, particularly dysregulated lipid metabolism, has emerged as a key hallmark of MM progression and drug resistance, making it a promising therapeutic target [17, 18]. In this study, we systematically dissected the role of the cholesterol synthesis pathway in MM and validated statins as repurposable anti-myeloma agents (Fig. 9E), with findings that bridge basic science and clinical translation.

A pivotal discovery of our work is the identification of the MVA pathway as a driver of MM progression and a robust prognostic signature. We demonstrated that 10 core genes in this pathway are progressively upregulated from healthy donors to MGUS and MM, which aligns with recent studies showing that the MVA pathway supports cancer cell proliferation by providing membrane components, signaling molecules, and energy substrates [19, 20]. A previous study reported that lipid metabolism rewiring promotes MM cell survival under nutrient stress [21], but our work extends this by defining specific cholesterol synthesis genes as actionable prognostic biomarkers-filling the gap in current MM stratification strategies that lack metabolism-related markers.

Our mechanistic studies revealed that statins exert anti-MM effects through a multi-pronged network, beyond their classic cholesterol-lowering role. Statins inhibit proliferation and induce caspase-dependent apoptosis by activating the ERK signaling pathway, which is consistent with findings that sustained ERK activation drives apoptosis [22]. One plausible mechanism by which statins activate ERK is the inhibition of the prenylation of small GTPases, including Ras and Rho [23, 24]. Since these small GTPases rely on prenylation for proper membrane localization and function, statin-induced prenylation inhibition might lead to dysregulated signaling that activates ERK. However, alternative pathways could also be involved, and further research is needed to clarify the exact mechanism. Transcriptome analysis uncovered that statins co-regulate lipid metabolism (steroid biosynthesis, terpenoid backbone synthesis) and immune-related pathways (interferon-γ signaling, NOD-like receptor pathway)-a novel crosstalk not previously reported in MM. This aligns with emerging evidence that metabolic inhibitors can modulate anti-tumor immunity [25–27], which suggests that statins may simultaneously target cancer cell metabolism and the immunosuppressive tumor microenvironment. We have validated that statins induce oxidative stress and Ca²⁺ dyshomeostasis, which can trigger autophagosome formation. Recent work showed that excessive ROS can shift autophagy from cytoprotection to cytotoxicity in MM [28], and our data support this “ROS-driven autophagic apoptosis” model-providing a unifying mechanism for statin’s anti-MM activity. Statins elevate the LC3-II/LC3-I ratios, which is a clear indication of the enhanced formation of autophagosomes. Concurrently, the augmented synthesis of p62 has the potential to disrupt the normal autophagic process. As more p62 is generated but fails to be efficiently degraded, it results in the blockage of autophagy. These findings contribute to a better understanding of the mechanism by which statins affect autophagy in MM cells.

Additionally, we demonstrated that statins synergize with bortezomib, a first-line MM drug. Low concentrations of statins enhanced bortezomib’s anti-proliferative and pro-apoptotic effects, which is clinically relevant given that bortezomib resistance remains a major challenge. Mechanistically, this synergy stems from enhanced apoptotic pathway activation rather than further ERK activation, indicating statins complement bortezomib’s proteasome inhibition by targeting lipid metabolism-a non-overlapping mechanism that reduces the likelihood of cross-resistance [29–31]. In our in vivo study, simvastatin monotherapy failed to significantly reduce the growth of MM, while the combination with bortezomib was effective. This could be attributed to pharmacokinetic limitations. For instance, simvastatin may have difficulty penetrating the bone marrow, resulting in insufficient drug concentration at the tumor site to exert its anti-MM effect. Moreover, the tumor microenvironment might also influence the efficacy of simvastatin. Cytokines and signaling pathways in the MM microenvironment may also antagonize the action of simvastatin, preventing it from inhibiting MM growth on its own.

Statins are well-tolerated, oral, and cost-effective drugs with decades of clinical use for cardiovascular diseases [32], eliminating the need for lengthy safety profiling. Our in vivo data showed the combination of simvastatin with bortezomib achieves an additional MM reduction, supporting their use as adjuvant therapy. The cholesterol synthesis gene signature (e.g., FDPS^high/HMGCS1^high) can be used to stratify MM patients who are most likely to benefit from statin treatment. This addresses the unmet need for personalized MM therapy, as current treatments are often administered without taking into account metabolic heterogeneity. A clinical study reported that MM patients with high cholesterol levels have poorer responses to bortezomib [33], and our preclinical data suggest statins can reverse this by disrupting cholesterol metabolism, which provides a rationale for testing statins in bortezomib-refractory patients. Statins are well-known to have both cholesterol-dependent and cholesterol-independent (pleiotropic) mechanisms [34]. The pleiotropic effects, such as anti-inflammatory and anti-angiogenic properties, might play a crucial role in the anti-MM activity, regardless of the initial cholesterol status. Further research is needed to disentangle the contributions of these two types of mechanisms. Notably, our findings also support the concept of “old drugs, new uses” in oncology, which accelerates clinical translation. Several phase II trials are currently evaluating statins in solid tumors [35–37], but our work provides the first comprehensive preclinical evidence for their use in MM-laying the groundwork for future clinical trials. For example, a biomarker-driven trial could enroll FDPS^high/HMGCS1^high MM patients to receive statins plus standard-of-care therapy, with progression-free survival as the primary endpoint. Numerous retrospective studies have consistently demonstrated an inverse association between statin use and both the incidence and mortality of MM [38–41], further supporting the use of statins in MM treatment.

Despite these strengths, our study has limitations that warrant further investigation. Future work should include primary MM cells and patient-derived xenograft (PDX) models, which better recapitulate the genetic heterogeneity of human MM [42]. While we identified key pathways regulated by statins, the molecular crosstalk between cholesterol synthesis inhibition, interferon signaling, and autophagy remains unclear. Future research directions should also explore combination strategies beyond bortezomib. For instance, statins could be combined with lenalidomide (an immunomodulatory drug) or daratumumab (a CD38 antibody), as these agents target distinct pathways (immune modulation, cell surface antigens) that may synergize with statins’ metabolic effects. Additionally, investigating the role of statins in MM bone disease-another major clinical burden-would be valuable, given that cholesterol metabolism has been linked to osteoclast activation [43–45]. In pre-clinical in vitro studies, the use of relatively high concentrations of lovastatin and simvastatin is a common practice. However, these concentrations are much higher than those achievable in clinical settings. This disparity may limit the direct translation of our in vitro findings to the clinical scenario. Although our in vitro and in vivo findings substantiate the synergistic efficacy of statins and bortezomib, when translating these results to clinical use, potential pharmacokinetic interactions between the two drugs must be taken into account. Specifically, they may compete for metabolic enzymes, which could alter drug concentrations in the body. For MM patients with high expression of MVA pathway genes, statins have the potential to resensitize bortezomib-resistant cells. Therefore, this patient subgroup is likely to benefit from the combination therapy.

Our study establishes the MVA pathway as a key oncogenic driver and prognostic marker in MM, and validates statins as multi-mechanistic anti-MM agents. These findings not only advance our understanding of metabolic reprogramming in MM but also provide a feasible, cost-effective strategy for clinical translation. By stratifying patients based on cholesterol synthesis gene expression and repurposing statins as adjuvant therapy, we can move toward personalized MM treatment that improves outcomes while minimizing toxicity. Future preclinical and clinical studies focusing on pathway crosstalk, PDX validation, and biomarker-driven trials will further unlock the potential of statins in MM management.

Abbreviations

DEG

Differentially expressed gene

HD

Healthy donors

HR

Hazard ratios

MFF

Mitochondrial fission factor

MGUS

Monoclonal gammopathy of undetermined significance

MM

Multiple myeloma

MVA

Mevalonate

PDX

Patient-derived xenograft

PIs

Proteasome inhibitors

ROS

Reactive oxygen species

Authors’ contributions

J.W. and H.W. designed the experiments. S.L., N.H., X.L., and Y.F. performed the experiments. Z.G. and Y.R. provided necessary assistance in the experiment. S.L., N.H., and J.W. analyzed the data and wrote the paper. J.W. and H.W. supervised the whole study and edited the manuscript.

Funding

This work received support from the National Natural Science Foundation of China (32570827 to J.W.), the Guangdong Basic and Applied Basic Research Foundation (2024A1515013250 to J.W.), the Guangdong Provincial Key Areas Special Projects for Regular Higher Education Institutions (2024ZDZX2054 to J.W) and the Young Top-notch Talent of Guangdong Special Support Plan (2025TQ09A778).

Data availability

RNA-seq data generated in this study are publicly available in the Sequence Read Archive database under accession numbers PRJNA1478343. All other data reported in this paper will be shared by J.W. upon request. All materials generated in this paper will be available from J.W. with a completed Materials Transfer Agreement.

Declarations

Ethics approval and consent to participate

All animal procedures were approved by the Institutional Animal Care and Use Committee of Guangzhou Medical University.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.