Regulation of Interferon-β-Modified Human Umbilical Cord Mesenchymal Stem Cell-Derived Exosomes in Proliferation and Apoptosis of Prostate Cancer Cells

ABSTRACT

Prostate cancer (PCa) poses a serious burden to men. Interferon-β (IFN-β) is implicated in cancer cell growth. This study hence explored the regulation of IFN-β-modified human umbilical cord mesenchymal stem cell-derived exosomes (hUCMSC-Exos) in PCa cells. In vitro-cultured hUCMSCs were transfected with pcDNA3.1-IFN-β plasmid or IFN-β siRNA. hUCMSC-Exos were extracted by ultracentrifugation and identified. PCa cells (PC3 and LNCap) were treated with Exos. Cellular internalization of Exos by cells was detected by uptake assay. Cell proliferation, cycle, and apoptosis were evaluated by CCK-8, EdU staining, and flow cytometry. Levels of cell cycle-related proteins (cyclin D/cyclin E) were determined by Western blot. The effect of IFN-β-modified hUCMSC-Exos in vivo was analyzed. IFN-β-modified hUCMSC-Exos (Exo^oe-IFN−β or Exo^si-IFN−β) were successfully isolated. IFN-β was encapsulated in Exos, and PCa cells could uptake Exos. After treating with Exo^oe-IFN−β, PCa cell proliferation was impeded, the percentage of cells in the G0/G1 phase, cyclin D/cyclin E levels, and cell apoptotic rate were elevated, while cells treated with Exo^oe-IFN−β exhibited contrary trends. IFN-β-modified hUCMSC-Exos reduced PCa tumor size and weight in vivo. Conjointly, IFN-β-modified hUCMSC-Exos suppress PCa cell proliferation and facilitate apoptosis.

GRAPHICAL ABSTRACT

KEYWORDS:

• Apoptosis
• cell cycle
• human umbilical cord mesenchymal stem cell-derived exosomes
• interferon-β
• PC3 cells
• proliferation
• prostate cancer

Introduction

Prostate cancer (PCa) ranks as the second most prevailing malignancy and the fifth dominant cause of cancer-associated death among men globally, contributing to approximately 1.4 million new cases and 375,000 deaths in 2020.¹ The burden of PCa is expected to increase to nearly 2.3 million newly-diagnosed patients and 740,000 deaths by 2040 partly due to population growth and aging.² Death from PCa is mainly attributed to the spread of cancer cells to other parts of the body, such as the brain, bone, rectum, bladder, pelvic, retroperitoneal lymph nodes, and spinal cord.³ Compared to other tumors, treatment of advanced PCa is particularly challenging owing to the limited access to metastatic tissues given the predominance of bone metastases as well as the failure to capture intraindividual tumor heterogeneity by analyzing one tumor region.⁴ The occurrence and development of PCa exhibit a close correlation with androgen.⁵ Currently, the therapies available for PCa including hormonal treatment, radiotherapy, and chemotherapy, are beneficial for only a minority of patients and present adverse side effects that ultimately affect patients’ quality of life.⁶ Thereby, there is an urgent need to find and identify novel optimal therapeutic modalities for PCa management.

Mesenchymal stem cells (MSCs) are widely distributed in various tissues, such as bone marrow, placenta, umbilical cord, adipose, fetal tissue, muscle, peripheral blood, and amniotic fluid, and the MSC-based therapies are considered an attractive option for cancer treatment.^7,8 In particular, human umbilical cord MSCs (hucMSCs) have attracted extensive attention because of the characteristics of large cell content, convenient isolation, low cost, minimal invasiveness, low immunogenicity, and high gene transfection efficiency, which indeed exert therapeutic effects primarily through extracellular vesicles (EVs) secreted by paracrine actions.⁹ As one of the extensively studied EVs, exosomes (Exos) naturally deliver an array of bioactive molecules such as proteins, RNAs, lipids, and metabolites that can modulate diverse signaling pathways in recipient cells and repress different stages of the carcinogenic process, showing therapeutic potential for various cancers, including PCa.^10,11 A previous study has illuminated that MSC-Exos suppress the proliferation of both PC3 (androgen-independent) and LNCap (androgen-dependent) PCa cells and induce their apoptosis.¹² Hence, exploring the application of hucMSC-Exos in the field of PCa is worthwhile.

Interferon-β (IFN-β), one member of Type I IFNs, is a cytokine with anti-viral, anti-proliferative, anti-tumor, and immunomodulatory properties.¹³ In PCa cells (PC3, Du145, and LNCap), miR-139, functioning as an immune agonist of RIG-1, possesses the ability to up-regulate IFN-β expression.¹⁴ The hormone sensitivity and androgen receptor of the human PCa cell line (PC3) are subjected to IFN-β regulation.¹⁵ IFN-β-induced cell cycle arrest is associated with increased G0/G1 cell arrest.¹⁶ It is interesting to note that the application of MSCs in cancer therapy depends on direct delivery of the IFN-β gene, and the treatment with IFN-β-MSCs not only suppresses PCa cell growth but also hinders tumor progression and prolongs the survival of PCa-bearing mice.¹⁷ Nevertheless, research focusing on the action of IFN-β-modified hUCMSC-Exos in PCa cell (PC3 and LNCap) growth is limited. Therefore, the study investigated the precise function of IFN-β-modified hUCMSC-Exos in PCa cell proliferation and apoptosis, expecting to provide a new theoretical basis and valuable ideas for gene therapy of PCa.

Materials and methods

Ethics statement

All animal experiments were ratified by the ethics committee of Peking University Shenzhen Hospital(approval no. 2022–715). We made considerable efforts to minimize the animal quantity and their suffering.

Culture and grouping of hUcmscs

hUCMSCs (Cyagen Biosciences, Guangzhou, Guangdong, China) were identified by the short tandem repeat (STR) method. hUCMSCs were subcultured in Dulbecco’s modified Eagle medium/F12 (Gibco, Grand Island, NY, USA) comprising 10% Exo-free fetal bovine serum (FBS; Gibco) and 1% penicillin-streptomycin solution (Zeye Biotechnology, Shanghai, China) in a 37°C incubator with 5% CO₂. The cells at passage 3 that reached approximately 80% confluence were selected for subsequent analyses.

hUCMSCs were allocated into 5 groups: blank group, oe-IFN-β group (transfected with pcDNA3.1-IFN-β plasmids), oe-negative control (NC) group (transfected with pcDNA3.1-NC plasmid), si-IFN-β [transfected with IFN-β siRNA¹⁸], and si-NC (transfected with Scramble siRNA). hUCMSCs were transfected with 10 nmol/L pcDNA3.1-IFN-β, pcDNA3.1-NC, IFN-β siRNA or Scramble siRNA at a final concentration of 3.75 µL/mL using Lipofectamine RNAiMAX Transfection reagent (Cat# 13778150, Invitrogen, Carlsbad, CA, USA).¹⁹ pcDNA3.1-IFN-β, pcDNA3.1-NC, IFN-β siRNA, and Scramble siRNA were provided by GenePharma (Shanghai, China).

Isolation and identification of hUCMSC-Exos

When hUCMSCs reached a confluence of 80%, the supernatant of hUCMSCs was collected, centrifuged at 2000×g for 30 min, and then added with total Exo isolation reagent (Thermo Fisher Scientific). The concentrated supernatant was incubated overnight at 2–8°C, and next centrifuged at 10,000×g for 1 h at 4°C. The precipitates were resuspended in phosphate-buffered saline (PBS), centrifuged for another 1 h, resuspended in PBS, and next stored in a −80°C freezer.

The isolated hUCMSC-Exos was identified by the following methods: (a) hUCMSC-Exos were fixed with 4% paraformaldehyde (Beyotime, Shanghai, China) in PBS for 20 min at room temperature, adsorbed onto a copper gate, and fixed again with 4% paraformaldehyde for 30 s. The morphology of Exos was observed under a transmission electron microscope (TEM, Thermo Fisher Scientific); (b) The particle size distribution of Exos was detected by nanoparticle tracking assay (NTA) using a NanoSight apparatus (Malvern Instruments, Malvern, UK); (c) The expression patterns of positive markers CD63, CD9, and CD81 and negative marker calnexin on the surface of Exos were determined by Western blot.

Exos were grouped as follows: 1) Exo: directly isolated from hUCMSCs; 2) GW: isolated from hUCMSCs treated with 20 μM GW4869 (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) for 2 h;²⁰ 3) Exo^oe-NC: extracted from hUCMSCs transfected with pcDNA3.1-NC plasmids for 48 h; 4) Exo^oe-IFN−β: extracted from hUCMSCs delivered with pcDNA3.1-IFN-β plasmids for 48 h; 5) Exo^oe-IFN−β + PK: Exo^oe-IFN−β added with proteinase K (20 mg/mL, Beyotime); 6) Exo^oe-IFN−β + PK + T: Exo^oe-IFN−β treated with 20 mg/mL proteinase K and 1% Triton X-100 (Sigma-Aldrich);¹⁹ 7) Exo^si-IFN−β: extracted from hUCMSCs transfected with IFN-β siRNA for 48 h; 7) Exo^si-NC: extracted from hUCMSCs delivered with Scramble siRNA for 48 h.

Culture and treatment of PCa cells

Human PCa cells (PC3 and LNCap) were provided by Tongpai Biotechnology (Shanghai, China) and identified using STR method. PCa cells were then cultured in RPMI-1640 medium (Thermo Fisher Scientific) comprising 10% FBS along with 100 IU/mL penicillin and 0.1 mg/mL streptomycin. Cell cultures were maintained at 37°C in a humidified atmosphere with 95% air and 5% CO₂. The 3^rd passage cells were collected when reaching 80% confluence. In this study, 5 μg/mL hUCMSC-Exos were used to pretreat PC3 and LNCap cells in accordance with the reference²¹ and the pre-experiment (Supplementary figure S1). Alternatively, 10 nmol/L of pcDNA3.1-IFN-β was transfected at a final concentration of 3.75 µL/mL,¹⁹ and the next experiment was carried out 48 h later.

Uptake assay

Exos were stained and labeled with PHK26 (MedChemExpress, Monmouth Junction, NJ, USA) and then co-cultured with PCa cells for 24 h. Following fixing with 4% paraformaldehyde (Beyotime) for 20 min, cell slides were rinsed thrice with PBS and stained with 4–6-diamino-2-phenylindole (Beyotime) for 1 h. Subsequently, the uptake of Exos by PCa cells was observed under a confocal fluorescence microscope (Carl Zeiss, Oberkochen, Germany).

Western blot

Exos or PC3 cells were lysed with protease inhibitor-contained radio-immunoprecipitation assay buffer (Beyotime) and the supernatant was collected after a 20-min centrifugation at 12,000×g at 4°C, followed by determination using bicinchoninic acid kits (Sigma-Aldrich). Following separation by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Thermo Fisher Scientific), proteins were transferred onto polyvinylidene fluoride membranes. Membranes were next incubated overnight at 4°C with primary antibodies: anti-CD63 (1:1000, ab134045, Abcam, Cambridge, UK), anti-CD9 (1:1000, ab236630, Abcam), anti-CD81 (1:1000, ab219209, Abcam), anti-calnexin (1:500, ab227310, Abcam), anti-cyclin D (1:1000, ABE52, Merck Millipore, Billerica, MA, USA), anti-cyclin E (1:1000, PA5–16237, Invitrogen). Subsequently, membranes were rinsed and probed for 1 h with horseradish peroxidase-labeled goat anti-rabbit secondary antibody (IgG, 1:2000, ab6721, Abcam) under room temperature, followed by detection with enhanced chemiluminescence. The gray analysis of bands was conducted with Image J software (NIH, Bethesda, MD, USA), with β-actin (1:200, ab115777) as an internal reference.

Cell counting kit-8 (CCK-8)

Cell suspension (100 μL/well) was seeded in 96-well plates and next pre-incubated at 37°C with 5% CO₂. Cell proliferation was evaluated utilizing CCK-8 kits (GM-040101, Genomeditech, Shanghai, China). Each well was supplemented with 10 μL CCK-8 solution and next incubated for 1 h at 37°C. The absorbance at 450 nm was measured on a microplate reader (Thermo Fisher Scientific).

5’-ethynyl-2’-deoxyuridine (EdU) staining

The proliferation of PCa cells was evaluated utilizing Cell-Light EdU DNA cell proliferation kits (RiboBio, Guangzhou, Guangdong, China). After 2-h culture in the medium added with the thymidine analog EdU at 37°C, PC3 cells were fixed for 30 min with 4% (v/v) paraformaldehyde and then permeabilized with 0.5% (v/v) Triton X-100. Thereafter, cells were incubated with 1 × Apollo reaction cocktail for 30 min. The incorporation of EdU into genomic DNA was visualized under the Axio Imager A2 fluorescence microscope (Carl Zeiss, Germany). Five fields were arbitrarily selected for determination of the percentage of EdU-positive cells.

Flow cytometry for cell apoptosis

The cell apoptosis was assessed with fluorescein isothiocyanate-Annexin V Apoptosis Detection Kits (BD Biosciences, San Diego, CA, USA). After detachment with trypsin, cells were cultured with AnnexinV/propidium iodide (PI) for 25 min at 20–25°C. Apoptosis was analyzed using a flow cytometer (FACScan, BD Biosciences) equipped with Cell Quest 3.0 software.

Flow cytometry for cell cycle

Cells (1 × 10⁶/mL) in each group were fixed overnight at 4°C with 70% ethanol and next added with PI dye (50 µg/mL, Gudao Biotechnology, Shanghai, China), followed by heating at 37°C for 30 min. The number of cells at different cell cycle phases was analyzed using a FACSCaliber flow cytometer (BD Biosciences) and the corresponding percentages were calculated.

In vivo xenograft tumor growth

Thirty BALB/c nude mice (aged 4 weeks, weighting 16 ± 2 g) were purchased from Kaixue Biotechnology (Shanghai, China). Mice were reared at 25 ± 2°C under controlled conditions of 12-h light/dark cycles and 60 ± 5% relative humidity, with free access to food and water.

BALB/c nude mice were assigned to 5 groups (6 mice per group): 1) NM + PC3 group: injected with normal PC3 cells; 2) NM + PC3 (Exo^oe-IFN−β): injected with PC3 cells that were co-cultured with Exo^oe-IFN−β; 3) NM + PC3 (Exo^oe-NC): injected with PC3 cells after co-culture with Exo^oe-NC; 4) NM + PC3 (Exo^si-IFN−β): injected with PC3 cells after co-culture with Exo^si-IFN−β; 5) NM + PC3 (Exo^si-NC): injected with PC3 cells after co-culture with Exo^si-NC. During the experimentation, 1 × 10⁷ PC3 cells²² were subcutaneously injected into the right dorsonuchal area of mice.

Four weeks later, mice were euthanized by an injection of 100 mg/kg pentobarbital sodium, and the tumor volume size and weight as well as IFN-β mRNA and protein levels in tumor tissues were measured. Tumor volume was calculated as follows: (a × b²)/2, where a and b indicated the longest longitudinal and transverse diameters, respectively.²²

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

The extraction of total RNA from cells, Exos, or tumor tissues was implemented by TRIzol reagent (Invitrogen), and cDNA was synthesized by reverse transcription with PrimeScriptTM RT reagent kits (TaKaRa, Kyoto, Japan). TaqMan primers and probes were provided by TaKaRa. qPCR assay was performed using an ABI 7500 system (ABI, Carlsbad, CA, USA) under specific reaction conditions: pre-denaturation at 95°C for 10 min, and next 40 cycles of denaturation at 95°C for 10 s, annealing at 6,0°C for 20 s, and extension at 72°C for 34 s. Three replicates were carried out for each sample. β-actin served as an internal control and data were analyzed using the 2^−ΔΔCt method. Primer sequences are exhibited in Table 1.

Table 1. Primer sequences.

Gene	Forward 5’-3’	Reverse 5’-3’
IFN-β	CGTTCCTGCTGTGCTTCTC	TGTAACTCTTCTCCATCTGTGAC
β-actin	CTGTCCCTGTATGCCTCTG	ATGTCACGCACGATTTCC

Enzyme-linked immunosorbent assay (ELISA)

The expression levels of IFN-β in cells, Exos, or tumor tissues were determined in compliance with the provided instructions of human IFN-β ELISA kits (Solarbio, Beijing, China).

Statistical analysis

All data were statistically analyzed and graphed using GraphPad Prism 8.01 (GraphPad Software Inc., San Diego, CA, USA). Data were depicted as mean ± standard deviation (SD). Comparisons between multiple groups were performed using one-way analysis of variance (ANOVA) and Tukey’s multiple comparisons test was conducted for post hoc analysis. The p value < .05 was regarded as statistically significant.

Results

Extraction and characterization of hUCMSC-Exos

Exos were extracted from hUCMSCs by ultracentrifugation. TEM revealed that Exos had a diameter of 40–160 nm and presented elliptical, and the membranous structures were present in the periphery of Exos (Figure 1a). Subsequently, NTA showed that Exos were mainly distributed around 100 nm at a concentration of 2.3 × 10⁶ cells/mL (Figure 1b). Western blot illustrated that compared with the GW group, the expression patterns of CD63, CD9, and CD81 on the surface of Exos were positive, but calnexin was not significantly expressed (Figure 1c). The above results evinced the successful extraction of hUCMSC-Exos.

Figure 1. Extraction and characterization of hUCMSC-Exos. (a): morphology of Exos observed by TEM; (b): size of Exos determined by NTA; (c): expression patterns of positive markers CD63, CD9, and CD81 as well as negative marker calnexin on the surface of Exos detected by Western blot. GW4869 was an inhibitor released by Exos.

IFN-β-modified hUCMSC-Exos hampered proliferation and accelerated apoptosis of PCa cells

The purchased hUCMSCs were firstly subcultured in vitro and later the cells at passage 3 were transfected with pcDNA3.1-IFN-β or pcDNA3.1-NC plasmids when reaching 80% confluence. The transfection efficiency was verified by RT-qPCR and ELISA kits, which revealed that IFN-β mRNA and protein levels were prominently increased in hUCMSCs after transfection with pcDNA3.1-IFN-β (all p < .001) (Figure 2a,b). Additionally, Exo^oe-IFN−β had higher IFN-β levels than Exo^oe-NC (all p  < .001) (Figure 2c,d). Exo^oe-IFN−β was further treated with proteinase K and Triton X-100, and it was noted that IFN-β levels in Exo^oe-IFN−β were not changed significantly after treatment with proteinase K (p  > .05), but prominently decreased after co-treatment with proteinase K and Triton X-100 (all p  < .001) (Figure 2c,d). The aforementioned results revealed that IFN-β-modified hUCMSC-Exos were successfully isolated and IFN-β was encapsulated within Exos.

Figure 2. IFN-β-modified hUCMSC-Exos hampered proliferation and accelerated apoptosis of PCa cells. (a): IFN-β mRNA level in hUcmscs measured by RT-qPCR; (b): IFN-β expression in hUcmscs measured by ELISA kits; (c): IFN-β mRNA levels in Exos measured by RT-qPCR; (d): IFN-β expression in Exos measured by ELISA kits; (e): the uptake of Exos by PCa cells; (f): IFN-β mRNA level in PC3 cells measured by RT-qPCR; (g): IFN-β expression in PC3 cells measured by ELISA; (h): cell proliferation assessed by CCK-8; (i): cell proliferation assessed by EdU staining; (j): cell apoptosis assessed by flow cytometry. Proteinase K could be used to degrade the envelope protein; triton X-100 could destroy the phospholipid bilayer structure of Exos. Data were presented as mean ± SD, and one-way ANOVA was implemented for comparisons among multiple groups, followed by Tukey’s test. **p < .01, ***p < .001.

To investigate the exact effects of IFN-β-modified hUCMSC-Exos on proliferation and apoptosis of PCa cells, PC3 and LNCap cells were cultured in vitro and next treated with IFN-β-modified hUCMSC-Exos or pcDNA3.1-IFN-β transfection. Firstly, the uptake assay unveiled that both Exo^oe-NC and Exo^oe-IFN−β could be taken up by PCa cells (Figure 2e). As detected by RT-qPCR and ELISA, the levels of IFN-β mRNA and protein in PCa cells treated with Exo^oe-IFN−β were obviously increased (all p < .001), whereas the levels were higher in PCa cells transfected with pcDNA3.1-IFN-β than in PCa cells treated with Exo^oe-IFN−β (all p > .05). (Figure 2f,g). CCK-8 revealed that the proliferative ability of PCa cells was prominently weakened after treatment with Exo^oe-IFN−β (p  < .001), while the proliferative ability of PCa cells transfected with pcDNA3.1-IFN-β was lower than that of PCa cells treated with Exo^oe-IFN−β (p > .05). (Figure 2h). EdU staining indicated that the proportion of PCa positive cells was markedly diminished after treatment with Exo^oe-IFN−β (all p  < .01), and PCa positive cells transfected with pcDNA3.1-IFN-β was lower than PCa cells treated with Exo^oe-IFN−β(p > .05) (Figure 2i). In addition, flow cytometry unraveled that the apoptotic rate of PCa cells was raised after treatment with Exo^oe-IFN−β (all p  < 0.01), while PCa cells transfected with pcDNA3.1-IFN-β exhibited a higher apoptotic rate than PCa cells treated with Exo^oe-IFN−β (p > .05). (Figure 2j). Taken together, IFN-β-modified hUCMSC-Exos restrained PCa (PC3 and LNCap) cell proliferation and facilitated apoptosis.

Inhibition of IFN-β partially counteracted the effect of IFN-β-modified hUCMSC-Exos on PC3 cell growth

To further validate whether IFN-β inhibition partially counteracted the effect of IFN-β-modified hUCMSC-Exos on PCa cell (PC3 and LNCap) proliferation and apoptosis, we isolated Exos after hUCMSCs were delivered with IFN-β siRNA or Scramble siRNA plasmids. The transfection efficiency was detected by RT-qPCR and ELISA kits, which showed diminished IFN-β mRNA and protein levels in hUCMSCs after transfection with IFN-β siRNA plasmid (all p  < .05) (Figure 3a,b). Additionally, Exo^si-IFN−β presented lower IFN-β levels than Exo^si-NC (all p  < .001) (Figure 3c,d). IFN-β mRNA and protein expression levels were strikingly reduced in PCa cells treated with Exo^si-IFN−β (all p  < .05) (Figure 3e,f). Moreover, the proliferative and apoptotic abilities of PCa cells were evaluated, and the results unveiled that after inhibiting IFN-β expression in Exos, the proliferation of PCa cells was enhanced (p  < .001) (Figure 3g), the proportion of PCa positive cells was elevated (all p  < .05) (Figure 3h), and the apoptosis was decreased (all p  < .05) (Figure 3i). All in all, IFN-β silencing partly reversed the impact of IFN-β-modified hUCMSC-Exos on suppressing proliferation and promoting apoptosis of PC3 cells.

Figure 3. Inhibition of IFN-β partially counteracted the effect of IFN-β-modified hUCMSC-Exos on PC3 cell growth. (a): RT-qPCR determined IFN-β mRNA level in hUcmscs; (b): ELISA kits determined IFN-β expression in hUcmscs; (c): RT-qPCR determined IFN-β mRNA level in Exos; (d): ELISA kits determined IFN-β expression in Exos; (e): RT-qPCR determined IFN-β mRNA level in PCa cells; (f): ELISA determined IFN-β expression in PCa cells; (g): CCK-8 evaluated cell proliferation; (h): EdU staining evaluated cell proliferation; I: flow cytometry assessed cell apoptosis. Data were displayed as mean ± SD, and one-way ANOVA was conducted for comparisons among multiple groups, followed by Tukey’s test. *p  < .05, **p  < .01, ***p  < .001.

IFN-β-modified hUCMSC-Exos induced G0/G1 arrest of PCa cells

IFN-β has been documented to induce apoptosis and block the cell cycle.²³ Therefore, we examined the precise role of Exos in PCa cell cycle by flow cytometry. After PCa cells were treated with Exo^oe-IFN−β, the percentage of cells arresting in G0/G1 phase was increased (all p  < .05) (Figure 4a) and the percentage of cells in S phase was reduced (all p  < .05) (Figure 4a); while after PCa cells were treated with Exo^si-IFN−β, fewer cells were arrested in G0/G1 phase (all p  < .05) (Figure 4a) and more cells were arrested in S phase (all p  < .05) (Figure 4a). In addition, Western blot revealed that the levels of cyclin D and cyclin E were upregulated in PCa cells after treatment with Exo^oe-IFN−β (all p  < .05) (Figure 4b), but downregulated upon Exo^si-IFN−β treatment (all p  < .05) (Figure 4b). Collectively, IFN-β-modified hUCMSC-Exos promoted G0/G1 arrest of PCa cells.

Figure 4. IFN-β-modified hUCMSC-Exos induced PCa cell cycle arrest by regulating cycle-related proteins. (a): flow cytometry detected cell cycle (the percentage of cells arrested in G0/G1, S, and G2/M phases); (b): Western blot determined expression levels of cell cycle-related proteins (cyclin D and cyclin E). One-way ANOVA was used for comparisons among multiple groups, followed by Tukey’s test. *p  < .05, **p < .01, ***p < .001.

IFN-β-modified hUCMSC-Exos repressed PCa progression in vivo

To investigate the specific effect of IFN-β-modified hUCMSC-Exos on the progression of PCa in vivo, the mouse xenograft models were established. The differently-treated Exos were co-cultured with PC3 cells for 24 h and next the PC3 cells were subcutaneously injected into nude mice. Four weeks later, nude mice were euthanized and their tumor volume and weight were measured. Compared with the NM + PC3 (Exo^oe-NC) group, tumor volume and weight were decreased in the NM + PC3 (Exo^oe-IFN−β) group (all p  < .001) (Figure 5a-c), but increased in the NM + PC3 (Exo^si-IFN−β) group compared to the NM + PC3 (Exo^si-NC) (all p  < .001) (Figure 5a-c). IFN-β mRNA and protein levels in tumor tissues were also measured by RT-qPCR and ELISA kits, which demonstrated that the NM + PC3 (Exo^oe-IFN−β) group had higher IFN-β levels (all p  < .001) (Figure 5d) and the NM + PC3 (Exo^si-IFN−β) group had lower IFN-β levels than the NM + PC3 (Exo^si-NC) group (all p  < .05) (Figure 5e). Briefly, IFN-β-modified hUCMSC-Exos conferred an inhibitory impact on PCa tumor progression in vivo.

Figure 5. IFN-β-modified hUCMSC-Exos repressed PCa progression in vivo. (a,b): volume of the xenograft tumor, n = 6; (c): weight of the xenograft tumor, n = 6; D: IFN-β mRNA level in tumor tissues determined by RT-qPCR, n = 6; E: IFN-β expression in tumor tissues determined by ELISA kit, n = 6. One-way ANOVA was performed for comparisons among multiple groups, followed by Tukey’s test. *p  < .05, **p < .01, ***p < .001.

Discussion

As a multifocal disease, PCa is primarily characterized by a high propensity for bone metastasis and great intratumoral heterogeneity.²⁴ At present, hormonal and surgical treatments are effective in treating localized and androgen-dependent PCa, but cancer easily progresses to an androgen-independent stage where the tumor continues to grow despite hormone depletion, thus leading to poor prognosis and treatment failure.²⁵ Recently, Exos emerge as novel and potent biological drug carriers in anti-cancer therapy owning to low immunogenicity and toxicity, high stability and bioavailability, and the stimulation of anti-tumor immune responses.²⁶ As vital immune response factors, type I IFNs (IFN-α and IFN-β) can potently enhance the vitality of natural killer cells, thereby facilitating secretion of cytokines and exerting broad-spectrum anti-viral and anti-tumor roles.²⁷ In our studies, we selected the androgen-independent PCa cell line PC3 and androgen-dependent PCa cell line LNCap as the study subjects to explore the function of Exos carrying IFN-β in cell growth and the findings elucidated that IFN-β-modified hUCMSC-Exos suppressed malignant behaviors of PCa cells.

IFN-β, a well-known cytokine secreted by fibroblasts and a vital element of the host defense system, exhibits unique functions in repressing tumor proliferation and triggering apoptosis in studies of lung cancer, breast cancer, ovarian cancer, glioblastoma, and melanoma.²⁸ However, the clinical application of IFN-β in cancer treatment is limited due to its short half-life and toxicity.²⁹ Exos are emerging as natural and efficient carriers of drug delivery, which can enhance the pharmacokinetic effects (such as drug solubility, bioavailability, and stability), weaken the damage to normal cells or tissues, and strengthen the affinity to target organs or cells, finally reducing toxicity and side effects.³⁰ It is noteworthy that MSCs are well-recognized as a powerful producer of Exos³¹ and can release anti-cancer molecules (such as IFN-β and tumor necrosis factor) and non-coding RNAs through Exos that functionally act as tumor suppressors or pro-apoptotic factors.³² As reported, Exos are capable of inducing apoptosis of PC3 and LNCap cells and precluding cancer cell proliferation.¹² In our study, hUCMSCs were firstly transfected with pcDNA3.1-IFN-β or IFN-β siRNA and next hUCMSC-Exos overexpressing IFN-β or silencing IFN-β were successfully extracted. Our findings elicited that the effect of pcDNA3.1-IFN-β direct transfection on PCa cell proliferation and apoptosis was weaker than that of the Exo^oe-IFN−β treatment group. We postulated that the transmission effectiveness of Exos or the impact of other compounds included in Exos could be contributing factors. Subsequently, our future investigations will delve deeper into the effects of other substances in Exos on the proliferation and apoptosis of PCa cells.

Our further experiments revealed that IFN-β-modified hUCMSC-Exos restrained PC3 and LNCap cell proliferation and strengthened apoptosis, but IFN-β silencing in Exos unleashed contrary roles in PC3 and LNCap cell behaviors. Consistently, EVs derived from genetically modified MSCs overexpressing IFN-β1 not only effectively activate human immune cells but also induce apoptosis in various carcinomas in vitro.³³ Prior research has revealed that the downregulation of IFN-inducible proteins in prostate epithelial cells shows a close association with PCa development and progression, and IFN-β treatment in PCa cells leads to reduced proliferative ability.¹⁴ The HVJ-E-induced IFN-β production contributes to the activation of PARP, caspase-3, and caspase-8 in PC3 cells, ultimately causing apoptosis.³⁴ Importantly, compelling evidence suggests the suppression of IFN-β-producing MSCs in PC3 cell growth.¹⁷ Besides, IFN-β genes are upregulated in PC3, Du145, and LNCaP PCa cells by miR-139, which contributes to autophagy and inhibition of cell proliferation.¹⁴ Also, IFN-β genes can dramatically repress the viability of LNCaP cells when they are exposed to the prodrugs.³⁵ Although the inhibitory action of IFN-β in PCa cell growth is extensively studied, the utility of hUCMSC-Exos as carriers of IFN-β in PCa cells has not been elucidated yet. Our study for the first time elicited that IFN-β-modified hUCMSC-Exos repressed PCa cell malignant behaviors.

It is noteworthy that numerous anticancer molecules induce cell death mainly by triggering tumor cell apoptosis and blocking cell cycle progression, or by the merged effects of both.^36,37 Major proteins such as cyclin A, cyclin B, cyclin D, and cyclin E are implicated in regulating cell cycle process, and dysfunction of these proteins results in cell cycle arrest.³⁸ IFN-β is capable of affecting all phases of the mitotic cell cycle commonly through an arrest in G1 phase or occasionally by prolonging all cell cycle phases, and more importantly, adipose tissue-derived MSC-IFN-β interferes with tumor cell proliferation by altering cell cycle progression.³⁹ Accordingly, our results demonstrated that IFN-β-modified hUCMSC-Exos induced more PCa cells arrested in G0/G1 phase and fewer cells in S phase, as well as an increase in cyclin D and cyclin E levels, in contrast to the effect of IFN-β-silenced Exos. Of note, the anti-neoplastic role of IFN-β1a is extensively associated with cell cycle perturbation in DU-145 and PC3 cells.⁴⁰ It was also reported that Exos can boost G1 arrest of PCa cells (PC3 and LNCap).¹² Likewise, the anti-cancer action of IFN-β-modified hUCMSC-Exos is partly attributed to the induction of G0/G1 arrest in PCa cells.

There is evidence to suggest that IFN-γ-modified exosomal vaccine is capable of restricting tumor progression and prolonging the survival period of PCa mice.⁴¹ The intravenous injection of IFN-β-MSCs can inhibit PC-3 ×enograft growth.¹⁷ Finally, we conducted animal experiments to further verify in vitro results and noted that tumor volume and weight in PCa xenograft mouse models were reduced upon treatment with IFN-β-overexpressed Exos, but raised by IFN-β-silenced Exos. The aforementioned evidence confirmed the in vivo anti-cancer activity of IFN-β-modified hUCMSC-Exos in PCa.

In conclusion, by culturing PCa cells in vitro and treating the cells with IFN-β-modified hUCMSC-Exos, the current study underlined that IFN-β-modified hUCMSC-Exos suppressed proliferation and potentiated apoptosis of PCa cells. Meanwhile, PCa xenograft mouse models were established for in vivo validation. Exos can effectively impede the proliferation and facilitate the apoptosis of androgen-dependent and androgen-independent PCa cells, which is consistent with the literature report.¹² Hence, this study offers a theoretical foundation and serves as a valuable reference for the therapy and clinical investigation of PCa. This article has several important limitations to consider. Firstly, the function mechanism of IFN-β-modified hUCMSC-Exos in regulating PCa cell behaviors is complicated, involving multiple pathways, but our studies only focus on the cell cycle. In the future, more comprehensive and in-depth studies should be designed to investigate the precise mechanism of IFN-β-modified hUCMSC-Exos in mediating PCa cell proliferation and apoptosis. Secondly, Exos are the extracellular vesicles actively secreted by cells, having a lipid bilayer membrane structure (roughly 40–150 nm) and containing various biological active substances, like lipids, proteins, and miRNAs, and are considered to be an important substance for the information exchange or interactions between the cells and the surrounding target cells.^42,43 However, we only focused on the effects of IFN-β-modified hUCMSC-Exos on the proliferation and apoptosis of the human PCa cell line, and did not conduct in-depth studies on the effects of other substances in Exos, and this drawback should be overcome in the future. Thirdly, hUCMSCs has been used based on several literature reports,^44–46 this study centers its investigation on hUCMSCs due to resource constraints, including experimental funds and time limitations. Nevertheless, future research endeavors will explore the impact on other types of umbilical cord stem cells.

Author contributions

YW and ZPZ is the guarantor of integrity of the entire study; YW and ZPZ contributed to the study concepts and study design; YW contributed to the definition of intellectual content; HCY and SML contributed to the literature research; HCY and SML contributed to the manuscript preparation and YW and ZPZ contributed to the manuscript editing and review; SML contributed to the animal studies; HCY contributed to the experimental studies and data acquisition; HCY and SML contributed to the data analysis and statistical analysis. All authors read and approved the final manuscript.

Data availability of statement

All the data generated or analyzed during this study are included in this published article.

Ethics statement

All animal experiments were ratified by the ethics committee of Peking University Shenzhen Hospital(approval no. 2022–715). We made considerable efforts to minimize the animal quantity and their suffering.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15476278.2023.2285836

Additional information

Funding

This work was supported by Research Foundation of Peking University Shenzhen Hospital (Grant no.JCYJ2021013) and The Health and Family Planning Commision of Shenzhen (Grant no. SZBC2017021).