Hypoxic Tumor-Derived Exosomes Induce M2 Macrophage Polarization via AMPK/p38 To Promote Lung Cancer Progression

Hypoxia is a major regulator of tumor aggressiveness and metastasis in cancer progression. Exosomes (exos) play an important role in the communication between lung cancer and hypoxic microenvironment. However, the underlying mechanisms are largely undened. Exos isolated from A549 cells under hypoxia conditions. Transmission electron microscopy and nanoparticle tracking analysis were carried out to characterize exos. CCK-8 assay, ow cytometry, Western blot, wound healing and transwell assays were performed to assess the proliferation, apoptosis, migration, and invasion of A549 cells, respectively. The M2 polarization of macrophages was evaluated by RT-qPCR and Western blot analysis. In vivo nude mice model was established to determine the regulatory effect of hypoxia/exos on the progression of lung cancer. hypoxia/exos induced M2 macrophage polarization via AMPK/p38 pathway and thus promotes lung tumor growth and metastasis. This study manifests hypoxia/exos as the possible targets for ghting against cancer progression and metastasis.


Introduction
Lung cancer is the most common malignancy worldwide and its mortality ranks rst (1). Despite the advance in the current therapies, the 5-year survival of petients with lung cancer remains unsatisfactory (2). Therefore, identi cation of novel diagnostic/therapeutic targets is urgently needed.
Intertumoral hypoxia is a dynamic and heterogeneous feature of most solid tumors, which drives multiple cancer processes, such as cell proliferation, epithelial-mesenchymal transition (EMT), invasion, and metastasis (3). Hypoxia regulates the cellular communication and crosstalk between tumor cells and their microenvironment via diverse secretory factors, including exosomes (exos) (4). Exos are vesicles of endocytic origin ranging 30-150nm in diameter and carry packages of molecular information, including proteins, nucleic acids, lipids, and metabolites wrapped in their lipid bilayer (5). Accumulated evidence demonstrates that exos derived from hypoxic tumor cells contribute to cancer progression by promoting tumor cell proliferation, migration, and invasion and conferring chemo-resistance (6-8).
Macrophages are one of the most abundant cells with noteworthy plasticity in tumor microenvironment (9). Macrophages could be polarized to classically activated M1 and alternatively activated M2 macrophages, depending on the microenvironmental signals (10). M1 macrophages exert proin ammatory phenotype and tumor cytotoxicity, while M2 macrophages suppress the in ammation, contributing to angiogenesis and favoring tumor progression (11,12). However, the association and mechanism of macrophages polarization involved in lung cancer progression remain to be elucidated.
In this study, we investigated the effect of hypoxic A549 cells-derived exos on the biological behavior of A549 cells and found that hypoxia/exos could promote the proliferation and migration of lung cancer cells both in vitro and in vivo. Moreover, we found that hypoxic exosomal PKM2 induced M2 polarization of macrophages through AMPK/p38 pathway, thus contributing to the promotion of tumor progression.

Methods
Cell culture and hypoxia treatment A549 and RAW264.7 cells were obtained from the American Type Culture Collection (Manassas, VA, USA) and cultured in DMEM medium with 10% fetal calf serum (FBS) and 1% penicillin/streptomycin at 37˚C in an atmosphere containing 5% CO 2 . Hypoxia environment was established with a hypoxia cell incubator of 1% O 2.
Exos isolation and identi cation A549 cells were cultured under normoxic (21% CO 2 ) or hypoxic (1% O 2 ) conditions for 3 days. The medium was replaced with 10% exos-depleted FBS at 80-90% con uence. Then the medium was collected and centrifuged at 300g for 10 min, 3000g for 20 min, 10,000g for 30 min, and ultra-centrifuged at 100,000 × g for 2 h at 4 °C to precipitate exos pellets. After the pellets were resuspended in PBS and stained with the phosphotungstic acid for 10 min, transmission electron microscopy experiment was performed for exos identi cation (13).

Page 4/16
Cells were harvested and ultra-centrifuged at 100,000 × g for 30 min at 4°C to collect proteins. The concentration of proteins was determined with the BCA protein assay kit (Thermo Fisher; Waltham, MA, USA). 20 μg protein was loaded on an SDS-PAGE, separated by electrophoresis, and transferred to PVDF membrane. The membranes were blocked with 5% skimmed milk for 1 h and incubated with speci c antibodies overnight at 4°C subsequently. Monoclonal antibodies (Abcam; Cambridge, UK) used in this study were as follows: Wound-healing assay A549 cells were seeded into 6-well plates. Then cells were scratched with a 200-µl sterile pipette tip at the con uence of 90%. The wounded monolayers were washed with PBS to remove cell debris and A549 cells were cultured with fresh serum-free medium with normoxia/exos or hypoxia/exos. The distance between the two edges of the wound was calculated at 0 h, 24 h, 48 h to evaluate the migration capacity.

Cell apoptosis analysis
Cell apoptosis was examined with the Annexin V-FITC apoptosis detection kit (BD Biosciences; New Jersey, USA). After treatment with normoxia/exos or hypoxia/exos, A549 cells were collected and resuspended in 1×binding buffer, followed by staining with Annexin V-FITC and PI in the dark at room temperature. Then the percentage of apoptotic cells was immediately analyzed by a ow cytometer (BD, Biosciences; New Jersey, USA).

Exo uptake assay
After being labeled with the PKH67 Green Fluorescent Cell Linker Kit (Sigma-Aldrich; St. Louis, MO, USA), exos were added to RAW264.7 cells. Then, RAW264.7 cells were xed with 4% paraformaldehyde and stained with DAPI prior to its observation by a uorescence microscopy.

Real-Time quantitative PCR (RT-qPCR)
Total RNA was isolated with the TRIzol reagent (Invitrogen; Carlsbad, CA, USA) according to the manufacturer's instructions. Reverse transcription reaction was performed by using reverse transcriptase kit (Transgen Biotech; Beijing, China). Subsequently, PCR was performed with a standard SYBR Green PCR kit (Transgen Biotech). GAPDH expression was regarded as the normalization control.

Migration and invasion assays
Migration and invasion assays were performed using transwell chambers (24 wells) (Corning; NY, USA) coated without or with Matrigel (BD Biosciences), respectively. 5 × 10 4 A549 cells were re-suspended in serum-free medium and placed in the top chamber. The lower chamber was lled with 1x10 4 macrophages induced by hypoxic/exos and normoxia/exos in 600 μl medium containing 10% FBS. After incubation for 24 h, the migrating or invading cells were xed with methyl alcohol and subsequently stained with 0.1% crystal violet. The number of migrating or invading cells was counted using an Inverted microscope.

Hematoxylin-eosin (HE) staining
Suspicious lung metastasis tissues were assessed by HE staining. The tissues were placed in 10% formalin overnight, and then dehydrated and embedded in para n. Then the tissue samples were sliced to 4 μm thick sections, xed on a glass slide, dried and stained. The sections were successively immersed in xylene, ethanol with gradient concentrations, and hematoxylin. After being mounted with resin, the sections were dried naturally, and observed and photographed with a light microscope.

Animal experiments
All animal experiments were approved by the Guangzhou Medical University. Nude mice (male, 4-6 weeks of age, 18-20 g) were purchased from Guangdong Animal Experiment Center. Luciferase-labelled A549 cells were subcutaneously injected with into the oxter of nude mice. After the tumor grew to 50-100 mm 3 , mice were randomly divided into 3 groups: control, normoxia/exos, and hypoxia/exos (n=5). Exos derived from A549 cells under normoxia or hypoxia conditions were injected into mice via the tail vein for the treatment. In vivo uorescence imaging was performed to visualize the tumor growth. Tumors size was measured using a caliper weekly. Four weeks after injection, mice were sacri ced under general anesthesia. For metastasis, A549 cells were injected into nude mice via the tail vein. 24 h after injection, mice were treated with exos via tail vein injection. Suspicious lung metastasis tissues were analyzed by HE staining.

Statistical analysis
All in vitro results were derived from at least three independent experiments. Statistical analyses, histograms drawing, and scatter plots were performed with SPSS 22.0 software (SPSS Inc. USA) and GraphPad Prism 6.0 software (GraphPad Software, Inc.) The signi cance of differences between two groups was determined using Student's t-test. Comparations among more than two groups were performed with one-way analysis of variance (ANOVA). P<0.05 was considered to indicate a statistically signi cant difference.

Results
Characterization of exos from A549 cells under normoxia and hypoxia A549 cells were placed under normoxia and hypoxia for 48 h and exos were isolated from the conditioned media. The exo morphology was identi ed by transmission electron microscope assay. The exos were presented as cup-shaped double-layer membrane structure ranging from 30 to150 nm in diameter (Fig.   1A). Nanoparticle tracking analysis showed that the nanoparticle relative concentration of exos derived from hypoxic A549 cells was signi cantly higher than that from normoxic condition (Fig. 1B). Hypoxia exos exhibited the elevated expression of exo speci c markers (CD9, CD81, TSG-101, HSP70 and ALIX), as characterized by Western blot (Fig 1C).
Hypoxia/exos promote the proliferation and migration, and inhibit apoptosis in A549 cells Next, we explored the effect of hypoxia/exos on A549 cells. As shown in Fig. 2A, A549 cells cultured with hypoxia/exos exhibited higher proliferation capacity than those cultured with normoxia/exos. Wound healing assay demonstrated that hypoxia/exos enhanced the migration ability of A549 cells (Fig. 2B).
Moreover, ow cytometry analysis showed that administration of hypoxia/exos inhibited the apoptosis of A549 cells, compared with normoxia/exos (Fig. 2C). In line with these observations, the expression of cleaved-caspase3 and Bax was lower in the hypoxia/exo group than the normoxia/exo group (Fig. 2D), while the level of Bcl2 was increased in the hypoxia/exo group compared with the normoxia/exo group. These data indicated that hypoxia/exos could facilitated the proliferation and migration, and antagonized the apoptosis in A549 cells.
Hypoxia/exos induced the polarization of M2 macrophages M2 macrophages are bene cial for cancer progression. We then investigated whether hypoxia/exos had an effect on macrophage polarization. The uptake assay showed that PKH67-labeled exos could be internalized by RAW264.7 cells (Fig. 3A). M1 macrophages are characterized by the expression of iNOS and IL-1β, while M2 macrophages are characterized by the expression of CD206, CD163, TGF-β, IL-10, and Arginase-1. The results of RT-qPCR showed that the mRNA expression of CD206, CD163, TGF-β, IL-10, Arginase-1 was signi cantly higher in macrophages incubated with hypoxia/exos than those incubated with normoxia/exos (Fig. 3B). Consistently, the protein levels of M2 markers (CD206 and Arginase-1) were increased following the treatment with hypoxia/exos (Fig. 3C). Together, these results indicated that hypoxia/exos could induce macrophage towards to the M2 phenotype.
Glucose metabolism is important change in cancer cells under a hypoxic environment. PKM2 has been reported to play a critical role in glycolysis (15,16). Interestingly, we found that the expression of PKM2 in hypoxia/exo was signi cantly increased compared with normoxia/exo (Fig. 3D). Western bolt results manifested higher levels of p-AMPK, and p-P38 in macrophages following stimulation with hypoxia/exos (Fig. 3E). Thus, these results implied that hypoxia/exos induces M2 polarization is associated with exosomal PKM2-mediated AMPK/p38 activation.

Hypoxia/exos-induced M2 macrophages favor the migration, invasion, and EMT in A549 cells
To determine the effects of M2 macrophages on the migration and invasion ability of A549 cells, we performed transwell assay. The results revealed that A549 cells co-incubated with M2 macrophages treated with hypoxia/exos exhibited superior migration and invasion abilities compared with the control cells ( Fig. 5A and 5B). It is known that re-activation of the EMT program promotes cancer progression and enhances the metastatic phenotype (17). Western blot analysis presented that the expression of Ecadherin, a major epithelial marker, was decreased, while the expression levels of mesenchymal cell markers N-cadherin and vimentin were enhanced in A549 co-cultured with hypoxia/exo-treated macrophages (Fig. 5C). All together, these results con rmed a more superior role of M2 macrophages induced by hypoxia/exos in promoting migration, invasion, and EMT in A549 cells.

Hypoxia/exos facilitate tumor growth and metastasis of A549 cells in vivo
Given the important role of hypoxa/exos in the proliferation and migration of A549 cells in vitro, we further veri ed the function of hypoxia/exos using the xenograft model via injection with luciferaselabeled A549 cells. Higher uorescence intensity and increased tumor volume and tumor weight were observed in the hypoxia/exos group compared with the normoxia/exos group (Fig. 6A-C), indicating that hypoxia/exos exacerbated progressive growth in vivo. In addition, more lung metastatic nodules were observed in hypoxia/exos-treated mice compared with the control mice ( Fig. 6D and 6E). Collectively, these data suggest that hypoxia/exos have a promoting effect on lung cancer progression.

Discussion
Hypoxia is a common phenomenon in the tumor microenvironment and is associated with tumor progression and drug resistance (3). Accumulating studies have been indicated that hypoxia triggers exos release from cancer cells, thus facilitating tumor progression (18). Herein, we isolated the exos from A549 cells under hypoxia conditions and investigated their function and mechanism in lung cancer progression. Our results showed that hypoxia/exos have an oncogenic activity in lung cancer, which is associated with exosomal PKM2-mediated induction of M2 macrophages via AMPK/p38 signaling.
Exos secreted by multiple cells have been reported to involve in tumorous biological function and exosomal concentration/cargo can be altered by hypoxia (4). Previous studies have been demonstrated that the acceleration of tumor progression was closely associated with the exos released from hypoxia tumor cells. For example, it has been reported that ovarian cancer cells exposed to hypoxia signi cantly increase their exos release and treatment with hypoxia-induced exos lead to a more aggressive and chemoresistant ovarian cancer phenotype (6). In this work, we also found that culture under hypoxic conditions favored the release of exos. Hypoxia/exos favored the proliferation and migration of A549 cells in vitro. Also, treatment with hypoxia/exos signi cantly promoted the tumor growth and metastasis of A549 cells in vivo, indicating an oncogenic activity of hypoxia/exos in lung cancer.
It is well acknowledged that macrophages act as scavengers, modulating the immune response and maintaining tissue homeostasis (19). Macrophages can be reprogramed by metabolites, cytokines, and other signaling mediators, which makes them the key participants in tumor microenvironment and associated with enhanced tumor progression (20). It has been reported that M2 macrophage polarization could be regulated by exos. Bardi et al. have showed that exos released from melanoma cells induced a mixed M1 and M2 pro-tumor macrophages (21). Exos from adipose-derived stem cells polarize M2 macrophages and thus attenuate adipose in ammation and obesity (22). In the present study, the results showed that hypoxia/exos induced M2 macrophage polarization and therefore facilitated the migration and invasion of A549 cells.
Glucose metabolism change is a hallmark of cancer under hypoxia conditions. PKM2 is a key enzyme involved in glycolysis regulation and has an important regulatory role in cancer progression/resistance under hypoxia conditions (23)(24)(25)(26). It has been reported that exosomal PKM2 secreted by hypoxic cisplatin-resistance cells transmitted cisplatin-resistance to sensitive non-small cell lung cancer (NSCLC) cells both in vitro and in vivo (27). In present study, we found that hypoxia/exos derived from A549 cells could signi cantly elevated PKM2 expression. AMPK has been reported to play an essential role M2 macrophage polarization (28, 29). Consistently, our data showed treatment with hypoxia/exos signi cantly increased the phosphorylation of AMPK and its downstream p38 in macrophages. Collectively, our results implied that hypoxic exosmal PKM2 induces M2 macrophages, which was related to activation of AMPK/p38 pathway.