PM_2.5 was purchased from the National Institute of Standards and Technology (Gaithersburg, MD, USA). Human Wnt3a siRNA, negative control (NC) siRNA and antibodies against GRP94 (cat. no. sc-393402), CD63 (cat. no. sc-59284), Tsg101 (cat. no. sc-136111), Alix (cat. no. sc-53540), HSP70 (cat. no. sc-59570), Wnt3a (cat. no. sc-136163) and β-Actin (cat. no. sc-517582) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Human Wnt1, Wnt3a, Wnt4, Wnt7a, Wnt9a and Wnt10b primers were purchased from OriGene Technologies, Inc. (Rockville, MD, USA). An antibody against β-catenin (cat. no. ab6302) was purchased from Abcam (Cambridge, MA, USA). LF3, a specific inhibitor of Wnt/β-catenin signalling (24), was purchased from Selleckchem (Houston, TX, USA). Cell Counting Kit-8 (CCK-8) was purchased from Dojindo Molecular Technologies, Inc. (Tokyo, Japan). Matrigel matrix basement membrane was purchased from BD Biosciences (San Diego, CA, USA).

Female athymic nude mice (aged 6–8 weeks) were purchased from Joint Ventures Sipper BK Experimental Animal Co., Ltd. (Shanghai, China). The mice were maintained in specific pathogen-free facilities with temperature ranging from 22 to 24°C, humidity ranging from 50 to 60% and 12 h of light/dark cycle at Wenzhou Medical University (Wenzhou, China). Mice had free access to food and water, and all experiments using mice were approved by and performed according to the guidelines of the Animal Ethics Committee of Wenzhou Medical University.

The A549 cell line, a human lung adenocarcinoma cell line, was purchased from the American Type Culture Collection (ATTCC; Manassas, VA, USA), and cultured in RPMI-1640 media supplemented with 10% (v/v) fetal calf serum (FBS; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37°C in a 5% CO₂ incubator.

PM_2.5 (10 mg) was suspended in 1 ml of normal saline, sonicated at 20% power for 3 pulses of 10 sec each (waiting 5 sec between pulses), and then 4 ml of normal saline was added to a final concentration of 2 mg/ml. The PM_2.5 solution was aliquoted and stored at 4°C. Before use, the PM_2.5 solution was sonicated at 20% power for 3 pulses of 10 sec each (waiting 5 sec between pulses).

A549 cells (2.5×10⁵/ml) were seeded into 6-well plates in total volume of 2 ml/well. After 12 h, the supernatant was discarded, and 2 ml of fresh RPMI-1640 media was added with 100 µg/ml PM_2.5. After 24 h, the supernatant collected from all the wells (240 ml in total) was differentially centrifuged at 300 × g for 10 min, 1,200 × g for 20 min, and 10,000 × g for 30 min at 4°C. The supernatants from the final centrifugation were ultracentrifuged at 100,000 × g for 1 h at 4°C. After removing the supernatants, the exosomal pellets were washed in a large volume of ice-cold phosphate-buffered saline (PBS) and centrifuged at 100,000 × g for an additional 1 h at 4°C. The final pellets were resuspended in PBS. The amount of exosomal proteins was assessed by a BCA assay (Thermo Fisher Scientific, Inc.).

Nanoparticle tracking analysis of exosomes was assessed by NanoSight NS300 Particle Size Analyzer (Malvern Panalytical Ltd., Malvern, UK). To detect morphology of exosomes, exosomes were isolated and diluted in 100 µl of PBS, and 20 µl of the suspension was placed onto Formvar carbon-coated copper grids (Beijing XXBR Technology Co., Ltd., Beijing, China) at room temperature for 1 min. The excess suspension was removed using filter paper. Exosomes were stained with 2% phosphotungstic acid at room temperature for 5 min. The grids were then fixed with 2.5% glutaraldehyde at room temperature for 5 min followed by rinsing with PBS three times. Images were observed with a Philips Tecnai-10 transmission electron microscope operating at 80 kV (Phillips Electronic Instruments, Inc., Mahway, NJ, USA).

Total RNA was extracted with TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The following PCR conditions were used: 1 cycle at 95°C for 30 sec; and 40 cycles of 5 sec at 95 and 60°C at 34 sec. Real-time PCR was performed using an Applied Biosystems 7500 Real-Time PCR system (Applied Biosystems, Foster City, CA, USA). The following primers were used: Wnt1 forward, 5′-ctcttcggcaagatcgtcaacc-3′ and reverse, 5′-cga tgg aac ctt ctg agc agg a-3′; Wnt3a forward, 5′-atg aac cgc cac aac aac gag g-3′ and reverse, 5′-gtc ctt gag g aa gtc acc gat g-3′; Wnt4 forward, 5′-gct gga gaa gtg cgg ctg tga-3′ and reverse, 5′-cca caa acg act gtg aga agg c-3′; Wnt7a forward, 5′-agg aga agg ctc aca aat ggg c-3′ and reverse, 5′-cgg caa tga tgg cgt agg tga a-3′; Wnt9b forward, 5′-cct gct tga gtg cca gtt tca g-3′ and reverse, 5′-aca ccg cgt aca gga aag ctg t-3′; Wnt10b forward, 5′-ctc ggg att tct tgg att cca gg-3′ and reverse, 5′-gcc atg aca ctt gca ttt ccg c-3′.

For detection of β-catenin nuclear translocation, A549 cells were treated with 10 µg/ml exosomes for 0, 30 and 60 min. The cells were then fixed, permeabilized and incubated with rabbit polyclonal antibodies against β-catenin (cat. no. ab6302; Abcam) using a dilution of 1:1,000 at 4°C overnight. Subsequently, the cells were incubated with Alexa Fluor^® 647 conjugated goat anti-rabbit antibodies (cat. no. ab150079; Abcam) using a dilution of 1:200 for 1 h. Cells were counterstained with DAPI to indicate DNA. Stained cells were viewed using a confocal microscope (SP2; Leica, Solms, Germany).

A549 cells (1×10⁶/ml and 2.5×10⁵/ml) were treated with 10 µg/ml exosomes from A549 with mock treatment (EXO_Ctrl) or EXO_PM2.5 for 24 h at 37°C. Then, 2×10⁴ cells in 100 µl of serum-free media were seeded into the top chamber. The bottom chamber was filled with 800 µl of medium containing 20% serum. After being cultured for 18 h at 37°C, the cells were fixed with methanol for 20 min and washed with PBS three times. The fixed cells were stained with 10 µg/ml DAPI for 30 min and washed with PBS. The stained cells were examined using a fluorescence microscope.

After rehydration using a 6-fold volume of serum-free media, 50 µl of Matrigel was added on an 8-µm polycarbonate membrane in 24-well Transwell plates, and the Matrigel was solidified at 37°C. Then, 1×10⁶ A549 tumour cells were incubated with 10 µg/ml EXO_Ctrl or EXO_PM2.5 for 24 h at 37°C. Subsequently, 5×10⁴ A549 cells in 100 µl of serum-free media were seeded into the top chamber. The bottom chamber was filled with 800 µl of medium containing 20% FBS. After being cultured for 48 h at 37°C, the cells were fixed with methanol for 20 min and washed with PBS three times. The fixed cells were stained with 10 µg/ml DAPI for 30 min and washed with PBS. The stained cells were examined using a fluorescence microscope.

A549 cells (2×10⁴) were seeded into 96-well plates at 200 µl/well, and 2 µg of exosomes was then added for 24 h. To some of the wells, 10 µM LF3 was added. Four hours before the end of culture, 20 µl of CCK-8 was added. The optical density (OD) of each well was read at 450 nm using an automated microplate reader (Sunrise; Tecan Group, Ltd., Mannedorf, Switzerland).

Exosomes or crude proteins were extracted from cell lysates by RIPA lysis buffer (Beyotime Institute of Biotechnology, Shanghai, China) and then qualified by BCA Protein Assay kit (Beyotime Institute of Biotechnology). A total of 20 µg of proteins was separated by 10% SDS-PAGE and transferred onto a polyvinylidene difluoride polyvinylidene difluoride (PVDF) membrane (EMD Millipore, Billerica, MA, USA). The membrane was blocked with 5% BSA in TBST, and then incubated with corresponding primary antibodies overnight at 4°C. After incubating with HRP-coupled secondary antibodies for 1 h at room temperature, the membranes were scanned using a Tanon 4500 Gel Imaging System (Tanon Science and Technology Co., Ltd., Shanghai, China). The following primary antibodies and secondary antibodies were used: Rabbit monoclonal antibodies against GRP94 (dilution 1:1,000; cat. no. ab108606), rabbit monoclonal antibodies against Tsg101 (dilution 1:1,000; cat. no. ab125011), mouse monoclonal antibodies against Alix (dilution 1:500; cat. no. ab117600), mouse monoclonal antibodies against HSP70 (dilution 1:500; cat. no. ab47455), mouse monoclonal antibodies against Wnt3a (dilution 1:1,000; cat. no. ab81614), mouse monoclonal antibodies against CD63 (dilution 1:500; cat. no. ab193349), rabbit monoclonal antibodies against β-catenin (dilution 1:2,000; cat. no. ab32572), mouse monoclonal antibodies against β-actin (dilution 1:500; cat. no. ab8226;), HRP-coupled rabbit polyclonal antibodies against mouse (dilution 1:2,000; cat. no. ab6728) and HRP-coupled goat antibodies against rabbit (dilution 1:2,000; cat. no. ab6721; all were from Abcam).

For transient silencing of the Wnt3a gene, 40 nM siRNA was transfected into cells (2×10⁵/well) using 3 µl of INTERFERin siRNA transfection reagent (Polyplus-Transfection, New York, NY, USA) per well in a 24-well plate. The knockdown efficiency of Wnt3a was confirmed by western blotting.

To establish the tumour model, A549 cells (5×10⁶) were subcutaneously injected into nude mice. On day 10, the mice were randomly divided into three groups (each group with 5 mice/total 120 mice) and received intratumoural injections of 5 µg of exosomes every other day (total 14 injections). For the survival study, when the largest tumour volume reached 4,000 mm³ (60 mice), the mice were no longer free to move around. For humanitarian reasons, the mice were sacrificed by cervical dislocation after being intraperitoneally injected with 50 mg/kg pentobarbital sodium approved by Animal Ethics Committee of Wenzhou Medical University. The other 60 mice were also sacrificed by this way on day 36. The length and width of tumours were assessed every four days by vernier caliper and the tumour was calculated according to the following formula: Volume = (length × width²)/2.

Results were expressed as the mean ± SEM. Statistics were analysed by one-way or two-way analysis of variance (ANOVA) with Newman-Keuls post hoc test. The survival curves were calculated using the Kaplan-Meier method and the log-rank test was used for survival analysis. All statistics were analysed by GraphPad Prism 5.0 software (Graphpad Software, Inc., La Jolla, CA, USA). A P-value of <0.05 was considered to indicate a statistically significant difference.

To assess if PM_2.5 treatment upregulated expression of Wnt family members, the mRNA expression level of Wnt family members in PM_2.5-treated A549 cells was assessed. The Wnt3a mRNA level was greatly increased compared to control cells (Fig. 1A). To investigate if PM_2.5 exposure affected Wnt3a protein level in EXO_PM2.5, EXO_PM2.5 and exosomes from A549 cells with mock treatment (EXO_Ctrl) were isolated. Visualization using electron microscopy indicated that EXO_Ctrl and EXO_PM2.5 had similar morphology, and both had diameters ranging from 50 to 150 nm (Fig. 1B), indicating that PM_2.5 exposure did not affect exosomal morphology. Nanoparticle tracking analysis revealed that the size distribution of EXO_Ctrl and EXO_PM2.5 was 119±41.7 and 119±41.3 nm (mean ± SD), respectively (Fig. 1C). The protein components of EXO_Ctrl and EXO_PM2.5 were examined. Both exosomes were negative for the endoplasmic reticulum-residing protein, GRP94, and positive for HSP70 as well as the multivesicular body-related proteins, CD63, Tsg101 and Alix (Fig. 1D). According to a previous publication, CD63 was used as a loading control (25). Notably, Wnt3a was only detected in EXO_PM2.5 (Fig. 1D). These results indicated that EXO_PM2.5 has higher Wnt3a protein levels than EXO_Ctrl.

β-catenin is a key downstream effector in the Wnt signalling pathway (26). In the off-state, β-catenin is phosphorylated by CK1 and subsequently phosphorylated by GSK-3β (27,28), resulting in destabilization of β-catenin (29). In the on-state, a Wnt ligand binds to a Frizzled receptor and then prevents β-catenin phosphorylation by GSK-3β, leading to accumulation and increased nuclear import of β-catenin (30). Since Wnt3a was enriched in EXO_PM2.5, we investigated whether EXO_PM2.5 activated Wnt/β-catenin signalling. Treatment with EXO_PM2.5 markedly increased total β-catenin protein in A549 cells (Fig. 2A). In addition, enhanced nuclear translocation of β-catenin was observed in EXO_PM2.5-treated A549 cells (Fig. 2B). These results indicated that EXO_PM2.5 activated the Wnt/β-catenin pathway.

Since the Wnt/β-catenin pathway has been implicated in tumour migration and invasion (31,32), we investigated whether EXO_PM2.5 promoted A549 cell migration. As revealed in Fig. 3A, EXO_PM2.5 treatment had no effect on A549 cell migration (Fig. 3A and B). The role of EXO_PM2.5 on invasion of A549 cells was next examined. No difference of invasive ability was observed in A549 cells with or without EXO_PM2.5 treatment (Fig. 3C and D). These results indicated that EXO_PM2.5 does not alter the migration and invasion abilities of A549 cells.

β-catenin is implicated in tumourigenesis (33), and Wnt3a/β-catenin signalling has also been demonstrated to induce tumour cell proliferation (34). Thus, the effect of EXO_PM2.5 on A549 cell proliferation was investigated. EXO_PM2.5, but not EXO_Ctrl, significantly promoted A549 cell proliferation (Fig. 4A). To elucidate the role of Wnt3a in this effect, Wnt3a was knocked down in PM_2.5-treated A549 cells by Wnt3a siRNA. Exosomes with low amounts of Wnt3a protein were obtained from PM_2.5-treated A549 cells (Fig. 4B). Exosomes from PM_2.5-treated A549 cells transfected with NC siRNA (EXO_PM2.5/NC siRNA) promoted A549 cell proliferation (Fig. 4C). However, exosomes from PM_2.5-treated A549 cells transfected with Wnt3a siRNA (EXO_PM2.5/Wnt3a siRNA) had no effect on A549 cell proliferation (Fig. 4C). In the presence of LF3, a specific inhibitor of Wnt/β-catenin signalling, EXO_PM2.5 did not promote A549 cell proliferation (Fig. 4D). These results demonstrated that the effect of EXO_PM2.5 on the enhanced A549 cell proliferation was dependent on Wnt3a/β-catenin signalling.

Finally, the effect of EXO_PM2.5 on A549 cell progression in vivo was assessed. An A549 tumour model in nude mice was established, and intratumoural injection of EXO_PM2.5 was performed every other day. EXO_PM2.5 increased A549 cell growth (Fig. 5A) and reduced the survival rate of tumour-bearing mice (Fig. 5B). Immunohistochemical staining of Ki-67 revealed that EXO_PM2.5 significantly promoted tumour cell proliferation (Fig. 5C and D). To dissect the role of Wnt3a in EXO_PM2.5 in this process, intratumoural injection of EXO_PM2.5/NC siRNA or EXO_PM2.5/Wnt3a siRNA was performed every other day. EXO_PM2.5/NC siRNA, but not EXO_PM2.5/Wnt3a siRNA, increased tumour growth and reduced the survival rate of tumour-bearing mice (Fig. 5E and F). Immunohistochemical results also revealed that EXO_PM2.5/Wnt3a siRNA did not promote tumour cell proliferation (Fig. 5G and H). In the survival study, when the largest tumour volume reached 4,000 mm³, for humanitarian reasons, the mice were sacrificed. However, in Fig. 5B, compared with EXO_Ctrl or PBS-treated mice, the tumour volume of EXO_PM2.5-treated mice reached 4,000 mm³ much earlier. Similarly, in Fig. 5F, compared with EXO_PM2.5/Wnt3a siRNA or PBS-treated mice, the tumour volume of EXO_PM2.5/NC siRNA-treated mice reached 4,000 mm³ much earlier. Therefore, these results reflected the real tendency of survival time of tumour mice with different treatments. Altogether, these results indicated that EXO_PM2.5-induced tumour growth in vivo was dependent on Wnt3a.