Abstract
Background:
Bone is the most common metastatic site in prostate cancer (PCa) patients and serves as a key contributing factor to the poor prognosis observed in advanced-stage patients. Mammalian target of rapamycin (mTOR) inhibition has limited clinical efficacy, potentially due to pathway complexity. Prior to the colonization by tumor cells, primary PCa cells actively remodel the bone microenvironment through the secretion of mediators including extracellular vesicles (EVs). The objective of this research is to investigate the regulatory mechanisms of EV biogenesis and the effects of EVs on the bone pre-metastatic niche (PMN), offering a novel therapeutic strategy against bone metastasis.
Methods:
PCa cell lines were employed to detect mTOR and Ras-related protein Rab-1A (RAB1A) protein expression levels via Western blotting (WB). Functional assays (invasion and proliferation) were used to validate the impact of RAB1A expression on biological behavior. The biological characteristics of EVs were characterized using WB, nanoparticle tracking analysis, and transmission electron microscopy. Bone marrow cell subpopulation alterations were analyzed based on the GSE143791 single-cell dataset. Cells and animal models were treated with EVs to assess their effects on the bone marrow microenvironment, survival time, and bone metastatic burden. Finally, peripheral blood routine parameters were compared in patients with or without bone metastasis.
Results:
Utilizing PCa cell lines, we demonstrated that mTOR activation inhibits the ubiquitination activity of the oncogenic factor RAB1A, thereby stabilizing its expression. The EVs derived from tumor promoted bone immunosuppression via B-cell dysfunction and myeloid cell expansion, highlighting their role in PMN formation. In RAB1A-overexpressing PCa animal models, GW4869-mediated inhibition of EV secretion prolonged mice survival, ameliorated bone marrow abnormalities, enhanced B-cell activation capacity, and reduced regulatory B-cell proportions.
Conclusions:
Our findings elucidated the detailed mechanism by which mTOR/RAB1A regulates EV secretion, providing new insight into cellular changes involved in PMN formation and a theoretical basis for the inhibition of the PMN in the development of targeted therapies for PCa. RAB1A represents a therapeutic target to reverse tEV-mediated immunosuppression, while peripheral B-cell dynamics provide diagnostic biomarkers for early metastasis detection.
1 Introduction
Based on the global cancer statistics from the American Cancer Society journal CA: A Cancer Journal for Clinicians, prostate cancer (PCa) ranks as the leading malignancy of the male genitourinary system, accounting for 29% of new cancer cases among men worldwide (, ). The progression of PCa to castration-resistant prostate cancer (CRPC) is frequently driven by somatic alterations in the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway, indicating that therapies targeting this pathway could improve survival outcomes and therapeutic efficacy. However, mTOR blockade in CRPC results in limited effectiveness. In a phase 2 clinical trial (NCT00629525), CRPC patients treated with the mTOR inhibitor everolimus showed neither a decline in prostate-specific antigen levels nor objective clinical responses (). This limited efficacy may stem from the inherent complexity and compensatory activation mechanisms within the mTOR signaling pathway (). Thus, to explore precise therapeutic targets and enhance therapeutic strategies, we further analyzed the mTOR signaling pathway and its downstream proteins in PCa.
Bone metastasis is the most common site of metastasis in PCa patients (). Apart from prostatectomy and androgen deprivation treatment, immunotherapy has failed to achieve the anticipated therapeutic benefits for PCa patients with bone metastasis (). Mechanistically, prior to the occurrence of bone metastases, primary tumors can remotely distort bone marrow cell activity in response to pathological changes, reshaping the microenvironment of the pre-metastatic niche (PMN) to assist in the colonization and proliferation of circulating tumor cells (CTCs) (). Extracellular vesicles (EVs) are small, highly heterogeneous membranous particles derived from various cell types, while RAS-related protein (RAB) GTPases act as regulatory switches to control EV trafficking and secretion (). Numerous reports have described that tumor-derived EVs (tEVs) drive immunosuppressive PMN formation through multifaceted mechanisms, such as inactivation of the anti-tumor activities of natural killer (NK) cells and T cells, induction of myeloid-derived suppressor cell expansion and macrophage polarization (, ). As a common cell type in the bone marrow, B cells develop through defined stages (pre-pro-B, pro-B/pre-B, immature B) before migrating to secondary lymphoid organs for further maturation (). Recent studies indicate that tumors can reprogram B cells to blunt immune responses and induce metastasis-supporting regulatory B cells to promote metastasis (, ). However, the impact of tEVs on B cell development and the role of B cells in tumor immunosuppression remain less understood. Thus, the tEV-mediated induction in B-cell development need intensive research, particularly regarding their specific contributions to the regulation of immunosuppression by the PMN.
The mTOR signaling pathway critically regulates the EV biogenesis and cargo loading (, ). Hyperactivation of mTORC1 in cancer cells enhances EV secretion, which in turn delivers oncogenic miRNAs (e.g., miR-21) and phosphorylated signaling proteins to recipient cells, indicating the complexity of the mTOR signaling pathway (). According to the study by Thomas, Ras-related protein Rab-1A (RAB1A) overexpression positively correlates with mTOR activation and rapamycin sensitivity in colorectal cancer cell lines (). Thus, we hypothesized that these studies on the regulatory mechanisms of mTOR and RAB1A in PCa would address the dilemma of the apparently low clinical utility of a single mTOR inhibitor in PCa patients.
Herein, we discover that inhibiting the mTOR/RAB1A axis suppresses EV secretion, thereby inhibiting the formation of the pre-metastatic immunosuppressive microenvironment. Our results further support that tEVs disrupt normal bone marrow cell development and induce an imbalance between B lymphocytes and myeloid cells, indicating that RAB1A may serve as a novel therapeutic target for PCa treatment.
2 Material and methods
2.1 Animals
Wild-type male C57BL/6 mice and nude mice at 6–8 weeks of age were purchased from Vital River and kept in a specific pathogen-free facility at constant temperature and humidity for 1 week prior to tumor cell inoculation or tail vein injection of tEVs. All animal experiments in this study were conducted in accordance with animal welfare law and institutional guidelines with the approval of the Institutional Animal Care and Use Committees at the Fourth Hospital of Hebei Medical University (Approval number: 2023212). Bone marrow cells were flushed out of the femur and tibia using phosphate-buffered saline (PBS) containing 1% fetal bovine serum (FBS, VivaCell, C04001-050X10) and a 27G syringe needle. Each batch of isolated primary cells was pooled from three mice. CD19+ B cells were isolated with a Mouse CD19+ B Cell Isolation Kit (480002, BioLegend) and the purity of magnetic-activated cell sorting (MACS)-purified CD19+ cells was > 95% (Supplementary Figure S6A).
2.2 Cell culture
PCa cell lines (human DU145 and PC-3, and murine RM-1 (Pricella) were cultured in Ham’s F12K (BasalMedia, L450KJ), MEM (Gibco) and RPMI 1640 (Gibco), respectively, each supplemented with 10% FBS and 1% penicillin-streptomycin solution. Bone marrow cells and MACS-purified CD19+ B cells were cultured in 1640 supplemented with 10% FBS, 1% penicillin-streptomycin solution, 2-Mercaptoethanol (50 µM). The cell culture medium used for tEVs isolation was formulated with 10% exosome-free FBS and 1% penicillin-streptomycin solution. Plasmid construction and lentiviral packaging were mentioned as before (). shRNA-1 targeting RAB1A (human): 5-GGAAACCAGTGCTAAGAATGC-3, shRAB1A-2 targeting RAB1A (human): 5- CTTCTTAGGTTTGCAGATGAT -3. shRNA-1 targeting RAB1A (mus): 5- GGAGTCCTTCAATAACGTTAA-3, shRAB1A-2 targeting RAB1A (mus): 5-GCACAATTGGTGTGGATTTCA -3. The specific plasmid overexpressing RAB1A and luciferase lentivirus were purchased from Genechem (Shanghai). All cell lines were cultured in a humidified incubator with 5% CO2 at 37°C.
2.3 EV isolation
To mitigate the effects of apoptotic cells on the purity of EVs, it was advisable to collect the cell supernatant when the cells achieved approximately 80%. Cell culture medium containing EVs was pre-cleared first by a speed centrifugation step (3,000×g for 15 min) to remove debris and dead cells. Then, remaining larger particles were removed with a 0.45-µm filter. The culture medium was concentrated using 10-kDa MWCO Amicon Ultra-15 spin-filters (UFC901024, Millipore). The concentrated cell culture supernatant was utilized for the purification of EVs according to the exoEasy Maxi kit (76064, Qiagen) protocol.
2.4 RNA extraction and quantitative reverse transcription PCR (qRT-PCR)
Total RNA was extracted from the cultured cells using TriQuick Reagent (Solarbio, Beijing, China) and measured with a ND100 spectrophotometer (Nanodrop, Wilmington, DE, USA) according to a standard protocol. cDNA synthesis was conducted using PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Japan). The gene expression was quantified with qRT-PCR using SYBR Green PCR master mix (Yeasen, Shanghai, China). The relative quantitative analysis of genes were calculated by the 2-ΔΔCt method normalized against β-actin, an internal control gene. All primer sequences in RT-qPCR were provided in Supplementary Table 1.
2.5 Western blotting and co-immunoprecipitation
Cell samples were lysed in lysis buffer supplemented with protease inhibitors for 30 min. After centrifugation at 20,000×g for 10 min at 4 °C, the protein concentration of the supernatant was determined using a BCA Protein Assay Kit. Equal amounts of protein were separated by 8-12% SDS-PAGE and transferred onto PVDF membranes. The membranes were incubated overnight at 4°C with primary antibodies. The antibodies against β-actin (AC026), P70S6K1 (A2190), and phospho-P70S6K1 (AP0564) were from Abclonal; against mTOR (66888-1), RAB1A (11671-AP), GAPDH (60004-1-Ig), mTOR (66888-1), and RAB1A (11671-AP) were from Proteintech; against RAB27B (HA723666), RAB7 (ET1611-96), ALIX (ET-1705-74), TSG101 (ET1701-59), and calnexin (ER1803-42) were from Huabio; and against CD9 (380441) was from Zenbio. After washing, membranes were probed with species-matched secondary antibodies (goat anti-mouse: A0216, Beyotime; goat anti-rabbit: A0208, Beyotime). For Co-IP, antibodies against mTOR and RAB1A were mixed with magnetic beads were incubated with protein samples at 4°C. Subsequently, proteins were separated from the magnetic beads and detected by WB as previously described.
2.6 Single-cell data quality control
Single-cell RNA sequencing data were processed using R and standard pipelines. Raw counts were imported via Seurat’s Read10X function and stored as a dgCMatrix (). Samples were merged into a single object, and cell identifiers were standardized with RenameCells. Doublets were detected and removed using Scrublet (). Low-quality cells were filtered by excluding genes detected in fewer than 200 cells and removing cells containing <500 or >4,000 genes per cell. Gene expression was normalized via the LogNormalize method with a scale factor of 10,000. The FindVariableFeatures function was used to identify the top 2,000 variable genes for downstream analysis. Technical noise (UMI counts, mitochondrial content) was regressed out using ScaleData. Dimensionality reduction was performed via PCA (30 principal components), followed by batch correction using Harmony. Cells were visualized using UMAP. Clustering was conducted by constructing a shared nearest-neighbor graph with the Louvain algorithm (resolution tested: 0.1-1.0; optimal resolution =0.6, validated via clustree). Differential gene expression analysis was performed using FindAllMarkers, followed by marker-based annotation with the Cell Taxonomy database (https://ngdc.cncb.ac.cn/celltaxonomy/) and literature references to assign cell identities.
2.7 Pseudotime analysis
For pseudotime trajectory analysis, Seurat objects were converted to Monocle2 CellDataSet format using count matrices from the RNA assay. The conversion process involved creating AnnotatedDataFrame objects for both phenotypic and feature data, followed by establishing a CellDataSet with the negative binomial distribution family, appropriate for sparse count matrices. Size factors and dispersions were estimated using estimateSizeFactors() and estimateDispersions() functions, respectively, and expressed genes were detected with a minimum expression threshold of 3. Highly variable genes for trajectory inference were selected based on dispersion analysis using dispersionTable(), where genes with mean expression ≥ 0.1 and empirical dispersion ≥ fitted dispersion were retained as ordering genes via setOrderingFilter(). Dimensionality reduction was performed using the Discriminative Dimensionality Reduction via Learning a Tree (DDRTree) algorithm with reduceDimension() function, projecting cells into a two-dimensional space with maximum components set to 2. The DDRTree method constructs a tree-like trajectory structure that captures both linear and branching developmental paths while preserving the global structure of cell state transitions. Finally, cells were ordered along the inferred trajectory using orderCells() function, which assigns pseudotime values representing the progress of each cell along the developmental trajectory and identifies distinct cell states based on the tree structure. This approach enables the identification of transitional cell states located at branching points and the characterization of gene expression dynamics throughout the developmental process.
2.8 Characterization of tEVs
The expression of TSG101, CD9, calnexin, ALIX, and GAPDH was evaluated by WB. The size distribution and concentration of tEVs were determined using ZetaView (version 8.05.14 SP7). Each EV sample was injected and measured three times. Transmission electron microscopy (TEM) was employed to assess the size and morphology of EV particles using a HITACHI H-7650 instrument (Tokyo, Japan) following a previously described method ().
2.9 Heat inactivation of tEVs
As described by Doste R. Mamand, tEVs were subjected to heat inactivation. The heat inactivation process involved using a heating block to incubate the tEVs at either 100°C for 10min (in vivo studies) or 56°C (in vitro studies) ().
2.10 Cell uptake experiment
EVs derived from RM-1 were labeled using a 10 µM PKH26 staining solution (HY-D1451, MCE) according to the manufacturer guidelines, 2×109 PKH26-labelled tEVs were injected through tail vein for 4 hours. Bone marrow cell uptake was analyzed by flow cytometry.
2.11 Tumor induction and tEVs treatment in mice
For tumor induction, male C57BL/6 mice were either non-injected or injected with 1 × 105 RM-1 cells suspended in 100 µL PBS (n = 6) via the left cardiac ventricle (LCV). Bone metastases were monitored through serial body weight measurements, lameness assessment, and hematoxylin-eosin (HE) staining of femoral and tibial tissues (, ).
Male C57BL/6 mice received tail vein injections every 3 days for 2 weeks with PBS/1 × 109 heat-inactivated tEV particles (100°C, 10 min)/1 × 109 tEV particles (n=3).
Male C57BL/6 mice were randomly divided into three groups (n=6 per group): Control Group (100µL PBS administered via tail vein injection prior to intracardiac PBS injection), PCa Group (100µL PBS administered via tail vein injection prior to intracardiac RM-1 cell injection), tEVs-treated PCa Group (1 × 109 tEV particles administered via tail vein injection prior to intracardiac RM-1 cell injection). Mice were injected through tail vein every 3 days for 2 weeks. Bone metastases were visualized by using a small-animal in vivo optical imaging system (IVIS).
Male C57BL/6 mice injected with RM-1 were randomly divided into five groups (n=6 per group): PCa Group (received daily intraperitoneal (i.p.) injections of PBS); RAB1A Group (received an LCV injection of RM-1 cells overexpressing RAB1A (RM-1 RAB1A-OE) and received i.p. injections of PBS daily); GW4869 Group (i.p. injections of GW4869 daily, HY-19363, MedChemExpress) at a dose of 2.5 mg/kg); RAB1A+GW4869 Group (received an LCV injection of RM-1 RAB1A-OE cells and i.p. injections of GW4869 at a dose of 2.5 mg/kg daily); and tEVs Group (i.p. injections of tEVs at a concentration of 1 × 109 particles, every 3 days). A PBS-injected group of the same strain served as the Control Group.
2.12 Flow cytometric analysis
Bone marrow cells and peripheral blood (PB) cells were treated with red blood cell lysis buffer in the dark and on ice. Samples were stained with different combinations of anti-mouse flow cytometry antibodies according to manufacturer’s instructions. 7-aminoactinomycin D was used to immediately exclude dead cells. The samples were further evaluated utilizing flow cytometry and data were analyzed with FlowJo Software (Version 10.8.1). Detailed information of flow cytometry antibodies used in this study is listed in Supplementary Table S2.
2.13 Multiplex secretome analysis
Blood samples were collected into vacuum tubes containing EDTA. After thorough mixing for 10–20 mins, centrifuge at 1000 × g for 15 mins to collect the upper supernatant. Plasma samples were stored at -20°C and avoid repeated freeze-thaw cycles. The fluorescent coded microspheres (RK05203, ABclonal Technology) were combined and mixed with supernatants following the manufacturer’s instructions with the assistance of ABclonal Technology. Two technical replicates were designed per plasma sample.
2.14 RNA sequence
MACS-CD19+ cells were isolated from the bone marrow of wild type C57BL/6 mice and treated with PBS (Control group) or tEVs (Experimental group) for 24 hours. RNA sequencing was performed on samples from both groups using Illumina’s next-generation sequencing platform. Differentially expressed genes (DEGs) (Supplementary Table S3) were identified as statistically significant using thresholds of P-values ≤ 0.05 and |log2FC| ≥1.
2.15 Statistical analysis
All data were analyzed using Prism 8 (GraphPad Software). Results are expressed as mean ± standard error of the mean (SEM). Comparisons between two groups were performed using unpaired Student’s t-tests. One-way ANOVA was used to analyze three or more groups for statistical significance. The correlation between mTOR and RAB1A was examined using Spearman’s correlation analysis. P < 0.05 was considered statistically significant (*, P < 0.05; **, P < 0.01; ***, P <0.001) and ns indicates non-significant differences.
3 Results
3.1 Inactivation of the mTOR signaling pathway accelerates RAB1A degradation via the ubiquitin-proteasome system
We analyzed mTOR mRNA expression across metastatic CRPC subtypes using the GSE77930 dataset. Bone metastatic tumor tissues exhibited significantly elevated mTOR expression levels compared to other sites (Figure 1A). To further clarify the association between mTOR signaling pathway activity and bone metastasis, we revealed that tumor tissues from PCa patients with bone metastasis exhibited significantly elevated mTOR signaling pathway activity compared to localized-stage PCa using the GSE32269 dataset (Figure 1B). Notably, there was also a modest positive association between pathway activity and RAB1A expression (R = 0.4, P = 0.0033; Figure 1C). Given previous reports of RAB1A-mediated mTOR regulation in colorectal cancer (), we downregulated RAB1A expression of two human PCa cell lines with RAB1A-targeting shRNA lentiviruses (Supplementary Figure S1A). The results revealed that RAB1A knockdown did not alter S6K phosphorylation status, contrasting findings from Thomas’s study (Supplementary Figure S1B). However, mTOR inhibition via rapamycin treatment reduced RAB1A protein levels in PCa cells, while serum starvation (FBS removal) showed no effect on RAB1A expression (Supplementary Figure S1C). Next, consistent with public datasets, we confirmed elevated RAB1A expression in PC3 cells (derived from bone metastasis) compared to DU145 cells (originating from brain metastasis) (Supplementary Figure S1D).
Figure 1
Our previous study reported RAB1A protein modification in PCa cells (). Thus, we hypothesized that the mTOR signaling pathway regulated RAB1A at the protein level via the ubiquitin-proteasome system. Co-IP assays showed that mTOR indeed physically interacted with RAB1A (Supplementary Figures S1E, F). Co-treatment of PCa cell lines with rapamycin and the proteasome inhibitor MG132 restored RAB1A protein levels that were reduced by rapamycin alone. Notably, the mTOR inhibitor induced a further prominent increase in RAB1A ubiquitination (Figure 1D-E). To investigate RAB1A stability under rapamycin treatment, we performed half-life analyses using the protein synthesis inhibitor CHX. These experiments revealed that RAB1A protein was significantly more stable in the absence of rapamycin treatment in vitro (Figure 1F). Mechanistically, the mTOR signaling pathway was shown to regulate RAB1A post-translational modification in a ubiquitin-dependent manner.
3.2 RAB1A expression is positively correlated with PCa progression
We further investigated the impact of RAB1A on PCa by reducing its expression. Next, RAB1A knockdown significantly impaired the proliferation of both PCa cell lines (Figure 2A). Consistently, silencing of RAB1A led to significant reductions in the migration and invasion capacity of PCa cells (Figures 2B, C), and inhibited colony formation, as shown by colony formation assays (Figures 2D, E). Additionally, depleting RAB1A resulted in higher proportions of apoptotic cells using propidium iodide (PI)/annexin V staining (Figure 2F). Overall, our findings suggested that RAB1A knockdown attenuated tumor development, indicating that RAB1A might have potential as a therapeutic target to inhibit PCa development.
Figure 2
3.3 RAB1A expression positively influences EV production
To elucidate the potential downstream regulatory mechanisms, we analyzed differentially expressed genes (DEGs) between high- and low RAB1A expressing groups based on the median RAB1A mRNA using TCGA database data. The results showed that RAB1A was associated with a variety of genes involved in small GTPase-mediated signal transduction (Supplementary Figure S2A). It is well acknowledged that the Rab GTPase family is crucial in vesicular trafficking and EV secretion, with secretory vesicles linked to actin cytoskeletal dynamics (). To investigate the impact of RAB1A knockdown on EV biogenesis, we screened a panel of EV-associated factors (, , ). As shown in Supplementary Figure S2B, qRT-PCR analysis revealed significant downregulation of mRNA expression for selected factors (such as RAB27, RAB7, ALIX, RAB5, RAB11, SNAP23, RAB35). We further profiled selected factors (CD9, RAB7, ALIX, RAB27, TSG101) and validated their expression at the protein level (Figure 3A). WB analysis demonstrated reduced protein expression to varying degrees in shRAB1A PCa cells (Figure 3B, Supplementary Figure S2C). Consequently, these data establish RAB1A as a key regulator of EV biogenesis in PCa, likely acting through the coordinated suppression of a network of essential Rab GTPases (RAB27A, RAB7A, RAB5A, RAB11A, RAB35) and EV cargo-sorting/scaffolding factors (ALIX, TSG101, SNAP23, CD9).
Figure 3
We further validated that the typical markers for tEVs and EVs derived from shRAB1A PCa cells expressed lower levels of TSG101 and Alix at the same concentration (Figure 3C, Supplementary Figure S2D). TEM images showed invaginations and typical cup-shaped membrane vesicles (Figure 3D, Supplementary Figures S3A, E). Nanoparticle tracking analysis (NTA) showed a particle distribution within the 100–200 nm range, consistent with the diameter of functional EVs. shRAB1A PCa cells secreted fewer EVs with a smaller size compared to control cells (Figure 3E, Supplementary Figures S3C, D). These findings collectively demonstrate that RAB1A knockdown significantly disrupts EV biogenesis and secretion in PCa cells. The concomitant decrease in both EV yield and particle size further indicates the essential role of RAB1A in regulating EV quantity and size characteristics.
3.4 B lymphocyte-myeloid cell ratio imbalance in the pre-metastatic bone microenvironment
Primary tumors remotely reprogrammed an immunosuppressive microenvironment in the bone marrow niche by secreting many soluble molecules, providing the subsequent CTCs with a notable survival advantage (, ). Based on the GSE143791 single-cell dataset, we analyzed the sequencing data of bone marrow cells from patients undergoing hip replacement surgery (Benign Group) and different vertebral specimens distant from the tumor site (Distal Group). As shown in Figure 4A, quality control and normalization were performed on the two groups. The bone marrow cell subpopulation atlas revealed decreased B cell populations and increased macrophages (Figure 4B). The statistical results shown in Figure 4C indicated that the proportion of T cells in Distal Group showed no significant difference compared to the Benign Group, while the proportions of progenitor B cells, immature B cells, and mature B cells were all significantly reduced (P < 0.05). In contrast, the proportions of macrophages and NK cells were significantly increased (P < 0.05). Furthermore, the proportion of erythroid in the Distal Group was significantly increased (P < 0.05). In conclusion, prior to the formation of bone metastatic lesions, the bone marrow microenvironment had already participated in the PMN construction through various regulatory mechanisms, characterized by myeloid cell expansion and suppressed development of the B-lymphocyte lineage.
Figure 4
3.5 EVs derived from PCa dysregulate the proportions of lymphocytes and myeloid cells
Compared to orthotopic models with low spontaneous metastasis rates (), LCV injection directly targets skeletal dissemination, enabling effective modelling of the PCa bone metastasis immune microenvironment. Therefore, we established a mouse model of bone metastasis via LCV injection of RM-1 cells (Supplementary Figure S4A). Analysis of bone marrow cell subtypes in PCa mice with bone metastasis revealed a reduction in total B cells, an increase in CD11b+ cells, and no significant change in CD3+ cells, indicating severe disruption of bone marrow cell development (Figure 5A). Flow cytometry further characterized all major B-cell subsets in the bone marrow (Supplementary Figure S4B). To investigate the regulatory role of tEVs in bone marrow cell differentiation, we treated bone marrow cells with tEVs, heated-inactivated tEVs (heated at 56°C for 10 min to effectively denature the bioactive molecules), conditioned medium, or co-cultured them with RM-1 cells (Figure 5C). EVs-treated bone marrow cells demonstrated an increase in the frequency of pre-pro B cells (B220lo+CD24+BP-1-), alongside a decrease in pro-B/pre-B cells (B220+CD24+CD43-IgM-IgD-) both in vivo (Figure 5B) and in vitro (Figures 5C-E). These alterations correlated positively with tEV concentrations (Supplementary Figure S4C) and tEVs block the differentiation of pre-pro-B cells into pro-B/pre-B cells, which was attenuated by heat inactivation. Considering the aforementioned finding that RAB1A knockdown decreased EV production, we treated bone marrow cells with EVs derived from the equal number of RM-1 cells transduced with either control shRNA (shCtrl-EVs) or RAB1A-targeting shRNA (shRAB1A-EVs). Consistent with the concentration-dependence, shRAB1A-EVs (produced in lower quantities) induced less pronounced B-cell population shifts compared to shCtrl-EVs (Supplementary Figure S4D).
Figure 5
In vitro studies (Figure 5D) showed a significant decrease in B lymphocytes and an increase in CD11b+ myeloid cells, which was consistent with our in vivo observation (Figure 5A). Furthermore, there was a significant number of adherent cells after tEV treatment, most of which were CD11b+F4/80+ cells (Figure 5F, Supplementary Figure S4E). Analysis of hematopoietic transcription factors (TFs) (, ) revealed that tEVs reprogrammed the transcriptional profile of bone marrow cells towards a myeloid lineage, evidenced by decreased expression of PAX5 (a B-cell master regulator) and increased expression of myeloid-associated genes (Figure 5G). Collectively, tEVs induce a myeloid-biased signature in bone marrow cells, driven by both increased CD11b+ cell abundance and TF-mediated transcriptional reprogramming.
3.6 B cells transdifferentiate into myeloid cells under tEV-mediated induction
Multi-color IF staining for CD19 and CD68 revealed the presence of a small population of CD19+ cells within CD68+ myeloid cells in the bone metastatic sites of PCa models, suggesting a potential interaction between CD19+ B cells and CD68+ macrophages (Supplementary Figure S5A). Myeloid and lymphoid lineages arise from common multipotent progenitors and share a similar set of genes in their differentiation processes (). Under certain circumstances, the response to the ever-changing needs of the hematopoietic system can lead to transformation between the two lineages (). Single-cell trajectory analysis of sorted bone marrow cells demonstrated B-myeloid transdifferentiation within the bone PMN (Figure 6A). Subsequently, we employed pseudotemporal heatmap analysis to identify critical cell-state transition points and associated signaling pathways orchestrating stage-specific differentiation (Figure 6B).
Figure 6
To relatively mimic the condition for tEVs production at the PMN formation phase, we treated C57BL/6 mice with PBS, heat-inactivated tEVs, or tEVs through tail vein injection. First, we administered PKH26-labeled tEVs intravenously and analyzed bone marrow cells by flow cytometry after 4 hours (Supplementary Figure S5B). As shown in Figure 6C, bone marrow cells from tEVs-treated mice exhibited significant enrichment of myeloid cells and a sharp decrease in B lymphocytes. Furthermore, tEVs induced PD-L1+ Breg within bone marrow and significantly inhibited B cell activation (Figures 6D, E). Correspondingly, in the PB of the tEVs-treated Group, the total B cells were significantly decreased, while myeloid cells were increased, which mirrored the trends observed in the bone marrow (Supplementary Figure S5C). Simultaneously, the proportion of PD-L1+ Breg cells increased, and B cell activation status remained suppressed in bone marrow cells and PB cells (Supplementary Figures S5D, E). In addition, we found that the plasma cytokine profile in the tEVs group manifested pro-tumor characteristics, such as increased IL-6 and decreased IL-12p40, IL-12p70 (Supplementary Figure S5F) (). Treatment with tEVs led to slower growth in body weight, although the difference did not reach statistical significance (Supplementary Figure S5G). Systemic administration of active tEVs (but not heat-inactivated tEVs) in mice recapitulated PMN formation.
MACS-purified CD19+ B cells (> 95% purity; Supplementary Figure S6A) were cultured in a complete medium supplemented with tEVs. EV exposure induced significant morphological alterations in CD19+ B cells, manifested as increased cell size (quantified by FSC-A) (Supplementary Figure S6B). As shown in vivo (Figure 6C) and in vitro (Supplementary Figure S6C), B220+CD11b- cells were increased in both proportion and absolute cell numbers in tEVs-treated groups. Next, we performed transcriptome sequencing on CD19+ B cells treated with PBS or tEVs. In tEVs-treated CD19+ cells, we observed increased expression of genes related to M2 macrophages (Arginase-1, Myc, and CD163) and chemokines (Supplementary Figure S6D). GO analysis showed that tEVs inhibited the activity of immunoglobulin production, B cell activation, adaptive immune response, immune cell proliferation, among other processes (Supplementary Figure S6E). Taken together, these findings demonstrated that tEVs drove B-cell transdifferentiation into myeloid lineages, which contributed to the imbalance between B lymphocytes and myeloid cells in PMN formation.
3.7 PCa-derived EVs promote bone metastasis in vivo
Next, we investigated the impact of tEV treatment on survival outcomes in a PCa model mice (Figure 7A). Given that intracardiac PCa models inadequately recapitulate early metastatic events that occur in the primary tumor, we pretreated wild type mice with tEVs (). As expected, tEVs-pretreated PCa mice exhibited reduced survival and increased bone metastatic burden (Supplementary Figure S7A, Figure 7B). Quantitative analysis revealed a significant increase in CD11b+ myeloid cell frequencies accompanied by a concomitant decrease in B lymphocyte populations within the bone marrow of tEVs-treated PCa mice (Supplementary Figure S7B). Furthermore, tumor-bearing mice exhibited splenomegaly whereas tumor-free control mice did not (not shown). To elucidate the potential contribution of RAB1A to the immune modulation of cancer immunoediting, we used RAB1A overexpression plasmid and GW4869, an inhibitor of exosome biogenesis to establish the experimental model (). As shown in Figure 7E, GW4869 significantly restored the RAB1A-induced imbalance between myeloid cell and B cell. Simultaneously, we observed increased PD-L1+ Breg proportion (Figure 7G) and suppression of B-cell activation in bone marrow (Figure 7F). These results demonstrated that RAB1A promotes PCa progression by enhancing EV secretion, whereas inhibition of EV secretion restores the bone marrow microenvironment and suppresses tumor progression. Clinically, PB analysis revealed significantly reduced lymphocyte proportions (P = 0.042) but unchanged monocyte frequencies in PCa patients with bone metastasis compared to those without metastasis (Figure 7H). Collectively, these findings establish the RAB1A-EV secretion axis as a master regulator of bone metastatic niche formation, driving immunosuppression via myeloid expansion and B-cell dysfunction, while its therapeutic blockade offers a multifaceted strategy to disrupt the vicious cycle of PCa progression.
Figure 7
4 Discussion
At the early stage of bone metastasis development, tEVs are released into the blood circulation and transported to the bone, where they orchestrate an immunosuppressive and inflammatory PMN microenvironment through multifaceted molecular mechanisms (, ). Since that discovery, extensive research focused on characterizing the formation of the pre-metastatic microenvironment, primarily initiated by regulatory immune cells (, ). For instance, exosomes derived from melanoma-educated bone marrow progenitor cells promote a pro-metastatic phenotype (). Notably, our prior work revealed a novel mechanism by which EVs derived from esophageal cancer disrupted the equilibrium between circulating follicular helper T and circulating follicular regulatory T cells, thereby promoting immune escape and tumor progression (). Consequently, it was uncertain whether tEVs could indirectly modulate the function of other immune cells via B cells, which had remained a huge knowledge gap in understanding tEVs-B cell crosstalk during PMN formation ().
In our study, we confirmed that mTOR expression was elevated in PCa patients with bone metastasis compared to those with other metastases. Here, we found that RAB1A expression decreased in response to inactivation of the mTOR signaling pathway, which may elucidate the limited clinical efficacy of mTOR inhibitors in PCa treatment. Additionally, activation of the mTOR signaling pathway upregulated RAB1A expression and thus promoted EV release. Mice injected with tEVs exhibited significant reductions in B-lymphoid populations alongside expanded myeloid compartments, mirroring the immune cell pattern found in the bone microenvironment of the vertebral body distant from the tumor site in human patients (). These findings establish tEVs as direct mediators of bone marrow immunomodulation in metastatic progression.
In the bone marrow, B-cell precursors and immature IgM+ B cells retain plasticity and the potential for myeloid transdifferentiation (–). In addition, tEVs drive bone marrow reconstitution through myeloid differentiation and disruption of lymphoid-biased HSC development. Although our results demonstrated that tEVs could induce transdifferentiation of B cells into myeloid cells in vivo and in vitro, they also revealed variations in the heterogeneity and plasticity of cells at the early stage of bone metastasis. Nevertheless, the specific factors carried by the tEVs that account for the dysregulation of lymphopoiesis and myelopoiesis are unclear. It is reported that the enrichment of TGF-β and miR-21 in EVs provides candidate mediators for functional validation (, ). Further studies should employ EV-subtype fractionation and CRISPR-based cargo editing to establish causality. These findings establish tEVs as direct mediators of bone marrow immunomodulation in metastatic progression.
Collectively, our study delineates a novel mTOR/RAB1A-regulated EV secretion axis that orchestrates PCa bone metastasis. EVs drive immunosuppressive niche formation via B-lymphocyte suppression and myeloid skewing. The results of our in vivo experiments emphasized the significant role of tEVs in bone metastasis progression in PCa. In addition, RAB1A as a novel target with the potential to reverse immunosuppression and enhance immunotherapeutic responses in clinical practice. The limited efficacy of mTOR inhibitors in PCa may reflect compensatory EV-mediated crosstalk, suggesting that co-targeting RAB1A and PD-1/PD-L1 could overcome microenvironmental immunosuppression. Considering that primary tumors disrupt bone marrow cell activities by releasing factors, cell subpopulations in PB may serve as potential biomarkers for detecting bone metastasis and monitoring responses to antiresorptive therapies. We will further include prospective correlation analyses of bone marrow-peripheral B-cell dynamics to quantify proportional shifts and subtype alterations alongside clinical evidences of bone metastasis in longitudinal PCa cohorts.
Statements
Data availability statement
The data presented in the study are deposited in the SRA database, accession number PRJNA1321719.
Ethics statement
The studies involving humans were approved by The Ethics Committee of the Fourth Hospital of Hebei Medical University. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements. The animal study was approved by the Institutional Animal Care and Use Committees at the Fourth Hospital of Hebei Medical University. The study was conducted in accordance with the local legislation and institutional requirements.
Author contributions
TL: Data curation, Formal analysis, Methodology, Writing – review & editing, Writing – original draft. YG: Writing – review & editing, Methodology, Formal analysis, Data curation. YZ: Project administration, Resources, Investigation, Supervision, Writing – review & editing. JC: Writing – review & editing, Resources, Investigation, Supervision. XL: Conceptualization, Supervision, Writing – review & editing. DW: Methodology, Data curation, Writing – review & editing. XZ: Data curation, Visualization, Methodology, Writing – review & editing. DH: Methodology, Writing – review & editing, Resources. XG: Formal analysis, Resources, Writing – review & editing. CY: Resources, Writing – review & editing. ZW: Supervision, Methodology, Project administration, Writing – review & editing, Investigation, Visualization, Conceptualization, Resources, Validation, Funding acquisition.
Funding
The author(s) declare financial support was received for the research and/or publication of this article. This work was supported by National Natural Science Foundation of China under Grant [No. 81872101], Natural Science Foundation of Hebei Province [H2024206210], the Central Guidance on Local Science and Technology Development Fund of Hebei Province [226Z7717G] and Hebei Provincial Key Research Projects [21377704D].
Acknowledgments
We thank Michelle Kahmeyer-Gabbe, PhD, from Liwen Bianji (Edanz) (www.liwenbianji.cn) for editing the English text of a draft of this manuscript.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that no Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu.2025.1605494/full#supplementary-material
Supplementary Figure 1Correlation between mTOR and RAB1A in PCa. (A) Analysis of PCa cells infected with lentiviral RAB1A shRNAs. (B) Analysis of p-S6K1 (T389) and total S6K in PCa cells infected with lentiviral RAB1A shRNAs. (C) Detection of RAB1A protein levels in PCa cell lines treated with rapamycin (100 nM) or serum-free culture medium for 24 h. (D) WB analysis of RAB1A expression in PCa cells. (E, F) Co-IP analysis of mTOR and RAB1A in PCa cells. Normal IgG was used as a negative control. All experiments have been repeated at least twice. Data were analyzed using t-test (D) and one-way ANOVA with multiple comparisons test. P < 0.05 was considered statistically significant.
Supplementary Figure 2RAB1A knockdown affects EVs secretion in human PCa cell lines. (A) GO enrichment analysis of patients with high or low RAB1A expression from the TCGA prostate cancer dataset. (B) Expression levels of EV-related genes determined by PCR in shCtrl and shRAB1A PCa cells. (C) Protein expression of RAB27B and TSG101 in shCtrl and shRAB1A PCa cells. (D) CD9 protein expression levels in EVs derived from shCtrl and shRAB1A PCa cells. All experiments have been repeated at least three times. Data were analyzed using t-test (B) and one-way ANOVA with multiple comparisons test (C, D). P < 0.05 was considered statistically significant.
Supplementary Figure 3RAB1A affects EV secretion in RM-1 cells. (A) Knockdown efficiency of RAB1A in RM-1 cells. (B) Statistical analysis of RAB1A knockdown efficiency. (C) NTA results of EVs secreted by shCtrl and shRAB1A RM-1 cells. (D) Statistical analysis of NTA. (E) Representative TEM images of EVs (scale bar: 100 nm). All experiments have been repeated at least three times. Data were analyzed using t-test. P < 0.05 was considered statistically significant.
Supplementary Figure 4Both tEVs and murine PCa cells dysregulate the proportions of lymphocytes and myeloid cells. (A) Representative H&E staining of bone from C57BL/6 mice with bone metastasis (scale bar: 50 µm). (B) Representative flow cytometric analysis of B subtypes in bone marrow. (C) Dose-dependent effects of tEVs on B cell subpopulations in cultured bone marrow cells (1×, 5 × 108 particles/well; 2×, 1 × 109 particles/well). (D) Quantified percentages of B cell subtypes in bone marrow cells cultured with shCtrl-EVs or shRAB1A-EVs. (E) Representative imaging of bone marrow cells after tEV treatment (scale bar: 20 µm). All experiments have been repeated at least three times. Data were analyzed using one-way ANOVA with multiple comparisons test (C, D). P < 0.05 was considered statistically significant.
Supplementary Figure 5EVs derived from PCa influence immune microenvironment. (A) IF imaging of tumor sections stained for CD68 (red), CD19 (green), and DAPI (blue) in PCa model with bone metastasis (scale bar: 200 µm). (B) Analysis of bone marrow cells by flow cytometry 4 hours after intravenous injection of PKH26-labeled tEVs. (C) Frequencies of B220+ B cells and CD11b+ myeloid cells in PB of mice (n=3). (D) Frequency of PD-L1+ B cells among CD45+CD19+ B cells in PB (n=3). (E) Proportion of IgA−IgM−IgG+ B cells among CD45+CD19+ B cells in PB (n=3). (F) ELISA assay performed on mouse plasma samples (n=3). (G) Body weight of mice. All experiments have been repeated at least three times. Data were analyzed using t-test and one-way ANOVA with multiple comparisons test (A). P < 0.05 was considered statistically significant.
Supplementary Figure 6EVs impair the development of MACS-CD19+ B cells in vitro. (A) Purity of MACS-purified CD19+ B cells. (B) Median FSC−A of CD19+ cell population in different groups. (C) Total count of myeloid cells (B220-CD11b+) calculated by flow cytometry in 1×106 CD19+ cells treated with tEVs. (D) Volcano plot of DEGs. (E) GO enrichment analysis of DEGs downregulated in CD19+ cells treated with tEVs compared to control cells. All experiments have been repeated at least three times. Data were analyzed using one-way ANOVA with multiple comparisons test (B, C). P < 0.05 was considered statistically significant.
Supplementary Figure 7Pretreatment with tEVs shortens survival time and disrupts cell subtypes in mice. (A) Kaplan-Meier survival analysis of experimental mice. (B) Gating strategy for B cells and myeloid cells in bone marrow (top) and peripheral blood (bottom). (C) Quantification of B cells and myeloid cells in BM (left) and PB (right). Control Group, n=6; PCa Group, n=6; tEVs-treated PCa Group, n=4. All experiments have been repeated at least three times. Data were analyzed using one-way ANOVA with multiple comparisons test (B). P < 0.05 was considered statistically significant.
Supplementary Table 1Primers for mRNA expression analysis.
Supplementary Table 2Flow antibodies.
Supplementary Table 3DEGs between experimental groups and control groups.
References
1
SiegelRLGiaquintoANJemalA. Cancer statistics, 2024. CA: Cancer J Clin. (2024) 74:12–49. doi: 10.3322/caac.21820
2
WuYPengSChengBZhongHCenMFuJet al. FOXA1-dependent PUS1 regulates EIF3b stability in a non-enzymatic pathway mediating prostate cancer bone metastasis. Int J Biol Sci. (2024) 20:4566–84. doi: 10.7150/ijbs.100905
3
GeorgeDJHalabiSHealyPJonaschDAnandMRasmussenJet al. Phase 2 clinical trial of TORC1 inhibition with everolimus in men with metastatic castration-resistant prostate cancer. Urol Oncol. (2020) 38:79.e15–22. doi: 10.1016/j.urolonc.2019.08.015
4
StatzCMPattersonSEMockusSM. mTOR inhibitors in castration-resistant prostate cancer: A systematic review. Target Oncol. (2017) 12:47–59. doi: 10.1007/s11523-016-0453-6
5
YuGCornPGMakCSLLiangXZhangMTroncosoPet al. Prostate cancer-induced endothelial-cell-to-osteoblast transition drives immunosuppression in the bone-tumor microenvironment through Wnt pathway-induced M2 macrophage polarization. Proc Natl Acad Sci U States A. (2024) 121:e2402903121. doi: 10.1073/pnas.2402903121
6
FuresiGde Jesus DominguesAMAlexopoulouDDahlAHacklMSchmidtJRet al. Exosomal miRNAs from prostate cancer impair osteoblast function in mice. Int J Mol Sci. (2022) 23:1285. doi: 10.3390/ijms23031285
7
StenmarkH. Rab GTPases as coordinators of vesicle traffic. Nat Rev Mol Cell Biol. (2009) 10:513–25. doi: 10.1038/nrm2728
8
TaoSCGuoSC. Role of extracellular vesicles in tumour microenvironment. Cell Commun Signal: CCS. (2020) 18:163. doi: 10.1186/s12964-020-00643-5
9
KfouryYBaryawnoNSevereNMeiSGustafssonKHirzTet al. Human prostate cancer bone metastases have an actionable immunosuppressive microenvironment. Cancer Cell. (2021) 39:1464–78.e8. doi: 10.1016/j.ccell.2021.09.005
10
HaoXShenYLiuJAlexanderAWuLXuZet al. Solid tumour-induced systemic immunosuppression involves dichotomous myeloid-B cell interactions. Nat Cell Biol. (2024) 26:1971–83. doi: 10.1038/s41556-024-01508-6
11
GuYLiuYFuLZhaiLZhuJHanYet al. Tumor-educated B cells selectively promote breast cancer lymph node metastasis by HSPA4-targeting IgG. Nat Med. (2019) 25:312–22. doi: 10.1038/s41591-018-0309-y
12
RagonnaudEMoritohKBodogaiMGusevFGaraudSChenCet al. Tumor-derived thymic stromal lymphopoietin expands bone marrow B-cell precursors in circulation to support metastasis. Cancer Res. (2019) 79:5826–38. doi: 10.1158/0008-5472.CAN-19-1058
13
Kaeser-PebernardSVionnetCMariMSankarDSHuZRoubatyCet al. mTORC1 controls Golgi architecture and vesicle secretion by phosphorylation of SCYL1. Nat Commun. (2022) 13:4685. doi: 10.1038/s41467-022-32487-7
14
BarberiLPorcuCBocciaCCosentinoMNicolettiCPeruzziBet al. Circulating extracellular vesicles in alcoholic liver disease affect skeletal muscle homeostasis and differentiation. J Cachexia Sarcopenia Muscle. (2025) 16:e13675. doi: 10.1002/jcsm.13675
15
SchellSLBrickerKNFikeAJChodisettiSBDomeierPPChoiNMet al. Context-dependent miR-21 regulation of TLR7-mediated autoimmune and foreign antigen-driven antibody-forming cell and germinal center responses. J Immunol (Baltimore Md: 1950). (2021) 206:2803–18. doi: 10.4049/jimmunol.2001039
16
Sanchez-GurmachesJGuertinDA. mTORC1 gRABs the golgi. Cancer Cell. (2014) 26:601–3. doi: 10.1016/j.ccell.2014.10.011
17
LvTHeDZhangXGuoXLiZZhangAet al. SGOL2 promotes prostate cancer progression by inhibiting RAB1A ubiquitination. Aging. (2022) 14:10050–66. doi: 10.18632/aging.204443
18
HaoYHaoSAndersen-NissenEMauckWM3rdZhengSButlerAet al. Integrated analysis of multimodal single-cell data. Cell. (2021) 184:3573–87.e29. doi: 10.1016/j.cell.2021.04.048
19
WolockSLLopezRKleinAM. Scrublet: computational identification of cell doublets in single-cell transcriptomic data. Cell Syst. (2019) 8:281–91.e9. doi: 10.1016/j.cels.2018.11.005
20
LiZZhangYHaoHChenLLvTZhangXet al. Esophageal cancer cell-derived small extracellular vesicles decrease circulating Tfh/Tfr via sEV-PDL1 to promote immunosuppression. Cancer Immunol Immunother: CII. (2023) 72:4249–59. doi: 10.1007/s00262-023-03561-w
21
MamandDRBazazSMohammadDKLiangXPavlovaSMimCet al. Extracellular vesicles originating from melanoma cells promote dysregulation in haematopoiesis as a component of cancer immunoediting. J Extracell Vesicles. (2024) 13:e12471. doi: 10.1002/jev2.12471
22
ParkSHEberMRShiozawaY. Models of prostate cancer bone metastasis. Methods Mol Biol (Clifton NJ). (2019) 1914:295–308. doi: 10.1007/978-1-4939-8997-3_16
23
TanakaMDykesSSSiemannDW. Inhibition of the Axl pathway impairs breast and prostate cancer metastasis to the bones and bone remodeling. Clin Exp Metastasis. (2021) 38:321–35. doi: 10.1007/s10585-021-10093-z
24
ThomasJDZhangYJWeiYHChoJHMorrisLEWangHYet al. Rab1A is an mTORC1 activator and a colorectal oncogene. Cancer Cell. (2016) 30:181–2. doi: 10.1016/j.ccell.2016.06.014
25
HanQFLiWJHuKSGaoJZhaiWLYangJHet al. Exosome biogenesis: machinery, regulation, and therapeutic implications in cancer. Mol Cancer. (2022) 21:207. doi: 10.1186/s12943-022-01671-0
26
LariosJMercierVRouxAGruenbergJ. ALIX- and ESCRT-III-dependent sorting of tetraspanins to exosomes. J Cell Biol. (2020) 219(3). doi: 10.1083/jcb.201904113
27
HaoXShenYChenNZhangWValverdeEWuLet al. Osteoprogenitor-GMP crosstalk underpins solid tumor-induced systemic immunosuppression and persists after tumor removal. Cell Stem Cell. (2023) 30:648–64.e8. doi: 10.1016/j.stem.2023.04.005
28
ReyaTRudolfG. Transcriptional regulation of B-cell differentiation. Curr Opin Immunol. (1998) 10:158–65. doi: 10.1016/S0952-7915(98)80244-6
29
SchmidlinHDiehlSABlomB. New insights into the regulation of human B-cell differentiation. Trends Immunol. (2009) 30:277–85. doi: 10.1016/j.it.2009.03.008
30
LiggettLASankaranVG. Unraveling hematopoiesis through the lens of genomics. Cell. (2020) 182:1384–400. doi: 10.1016/j.cell.2020.08.030
31
JoplingCBoueSIzpisua BelmonteJC. Dedifferentiation, transdifferentiation and reprogramming: three routes to regeneration. Nat Rev Mol Cell Biol. (2011) 12:79–89. doi: 10.1038/nrm3043
32
KureshiCTDouganSK. Cytokines in cancer. Cancer Cell. (2025) 43:15–35. doi: 10.1016/j.ccell.2024.11.011
33
ChenLBrigstockDR. Integrins and heparan sulfate proteoglycans on hepatic stellate cells (HSC) are novel receptors for HSC-derived exosomes. FEBS Lett. (2016) 590:4263–74. doi: 10.1002/1873-3468.12448
34
WangMQinZWanJYanYDuanXYaoXet al. Tumor-derived exosomes drive pre-metastatic niche formation in lung via modulating CCL1(+) fibroblast and CCR8(+) Treg cell interactions. Cancer Immunol Immunother: CII. (2022) 71:2717–30. doi: 10.1007/s00262-022-03196-3
35
WangYJiaJWangFFangYYangYZhouQet al. Pre-metastatic niche: formation, characteristics and therapeutic implication. Signal Transduct Target Ther. (2024) 9:236. doi: 10.1038/s41392-024-01937-7
36
HaynesNMChadwickTBParkerBS. The complexity of immune evasion mechanisms throughout the metastatic cascade. Nat Immunol. (2024) 25(10):1793–808. doi: 10.1038/s41590-024-01960-4
37
PeinadoHAlečkovićMLavotshkinSMateiICosta-SilvaBMoreno-BuenoGet al. Melanoma exosomes educate bone marrow progenitor cells toward a pro-metastatic phenotype through MET. Nat Med. (2012) 18:883–91. doi: 10.1038/nm.2753
38
YuDAllmanDGoldschmidtMHAtchisonMLMonroeJGThomas-TikhonenkoA. Oscillation between B-lymphoid and myeloid lineages in Myc-induced hematopoietic tumors following spontaneous silencing/reactivation of the EBF/Pax5 pathway. Blood. (2003) 101:1950–5. doi: 10.1182/blood-2002-06-1797
39
ChenCParkBRagonnaudEBodogaiMWangXZongLet al. Cancer co-opts differentiation of B-cell precursors into macrophage-like cells. Nat Commun. (2022) 13:5376. doi: 10.1038/s41467-022-33117-y
40
YingWJunLPeterDBJi-YangW. B cell development and maturation. Adv Exp Med Biol. (2020) 1254:1–22. doi: 10.1007/978-981-15-3532-1_1
41
MiaoBPZhangRSLiMFuYTZhaoMLiuZGet al. Nasopharyngeal cancer-derived microRNA-21 promotes immune suppressive B cells. Cell Mol Immunol. (2015) 12:750–6. doi: 10.1038/cmi.2014.129
42
Flores-HermenegildoJMHernández-CázaresFJPérez-PérezDRomero-RamírezHRodríguez-AlbaJCLicona-LimonPet al. Lrba participates in the differentiation of IgA+ B lymphocytes through TGFβR signaling. Front Immunol. (2024) 15:1386260. doi: 10.3389/fimmu.2024.1386260
Summary
Keywords
B lymphocytes, extracellular vesicles, pre-metastatic niche, prostate cancer, Rab1A
Citation
Lv T, Guo Y, Zhang Y, Cao J, Li X, Wang D, Zhang X, He D, Guo X, Yang C and Wang Z (2025) Extracellular vesicles from prostate tumors reshape the pre-metastatic bone environment in an mTOR/RAB1A-dependent manner. Front. Immunol. 16:1605494. doi: 10.3389/fimmu.2025.1605494
Received
03 April 2025
Accepted
22 August 2025
Published
19 September 2025
Volume
16 - 2025
Edited by
Ti Wen, The First Affiliated Hospital of China Medical University, China
Reviewed by
Anil Kumar Kalvala, Texas Tech University Health Sciences Center, Abilene, United States
Dipendra Khadka, Purbanchal University, Nepal
Kinjal Bhadresha, National Institutes of Health (NIH), United States
Updates
Copyright
© 2025 Lv, Guo, Zhang, Cao, Li, Wang, Zhang, He, Guo, Yang and Wang.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Zhiyu Wang, drwangzhiyu@hebmu.edu.cn
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.