Background
Lung cancer is the highly malignant tumor with leading cancer-related mortality worldwide [
1]. The most common type of lung cancer is non-small cell lung cancer (NSCLC), which can be further categorized into three subtypes: large cell lung carcinoma, lung squamous cell carcinoma and lung adenocarcinoma (LUAD). The latter is the most common type and accounts for 40 % of all lung cancers [
2]. More than 50 % of LUAD patients are initially diagnosed with advanced metastatic disease, and the 5-year survival rate of those patients is only 4 % [
3]. Accumulating evidence demonstrates that tumor microenvironment (TME) extremely involves in regulating tumor angiogenesis, cell proliferation, epithelial-mesenchymal transition (EMT), and even metastasis [
4,
5]. Hypoxia, defined as a condition of insufficient oxygen, is a common feature of the TME and serves to accelerate tumor progression and metastasis by activating hypoxic signaling pathways [
6,
7]. Increasing evidence suggests that the aberrant expression of hypoxia-related biomarkers is strongly associated with a worse clinical outcome [
8]. However, it remains unclear how tumor cells communicate with their microenvironment in response to hypoxic conditions, and subsequently exhibit a more aggressive phenotype.
Exosomes comprise a class of membrane-bound extracellular vesicles with a size range of 50–150 nm. Exosomes could be released extracellularly by almost all types of cells following the fusion of multivesicular bodies or mature endosomes with the cellular membrane [
9‐
11]. An early hypothesis supported the view that exosomes function as garage bags that eliminate nonfunctional cellular constituents. Subsequently, it was found that exosomes carry various bioactive molecules, such as proteins, lipids, and non-coding RNAs that can be transmitted to recipient cells and produce physiological or pathological effects on those cells [
12]. Tumor-derived exosomes have been found to regulate cancer progression by inducing autocrine/paracrine oncogenesis, angiogenesis, TME remodeling, modulating the immune system and pre-metastatic niche formation [
13,
14]. MicroRNAs (miRNAs), small class of endogenous non-coding RNAs, are major bioactive components carried by tumor-derived exosomes. MiRNAs encapsulated by exosomes regulate gene expression in recipient cells, and can thereby contribute to tumor aggressiveness and invasiveness [
15].
Hypoxia has been shown to stimulate exosome release and lead to significant changes in cellular contents and functions, indicating critical effects of exosomes as regulators in distant intercellular communication [
7]. Li et al [
16] reported that hypoxia significantly increased the levels of exosomal miR-21 and significantly enhanced oral squamous cell carcinoma (OSCC) cell migration and invasion in HIF-1α- and HIF-2α–dependent manners. Additionally, Tadokoro et al [
17] showed that hypoxic leukemia cells were responsible for the exosome-mediated transferring of miR-210 to facilitate human umbilical vein endothelial tube formation. These findings indicate that exosomes derived from hypoxic cells can alter the miRNA profiles of their target cells, and induce changes in cell phenotype.
In present study, we found that hypoxia increased the release of exosomes by LUAD cells. Hypoxic LUAD cell-derived exosomes (HExo) could enhance cell motility and metastasis. Furthermore, we examined the miRNA profiles of HExo and normoxic LUAD cell-derived exosomes (NExo). Exosomal miR-31-5p was found to be remarkably upregulated in HExo, and promotes tumor progression by targeting Special AT-Rich Sequence-Binding Protein 2 (SATB2) and activating MEK/ERK pathway. We also demonstrated that the expression of plasma-derived exosomal miR-31-5p were significantly increased in LUAD patients when compared with healthy control subjects. These data suggest that exosomal miR-31-5p is involved in tumor development and might serve as a diagnostic marker for LUAD.
Materials and methods
Cell culture and hypoxic exposure
The human LUAD cell lines A549, H1299, H292 and H1975 were obtained from the Cell Bank of the Chinese Academy of Sciences. All LUAD cell lines were cultured at 37℃ in a humidified incubator with a 5 % CO
2 and 20 % O
2 atmosphere in RPMI-1640 medium (Gibco, Waltham, MA, USA) supplemented with 10 % fetal bovine serum (FBS) and 1 % penicillin-streptomycin. For hypoxic treatment, the cells were cultured in a H35 HEPA hypoxystation (Whitley, UK) flushed with a mixture of 1 % O
2, 94 % N
2, and 5 % CO
2 for 24–48 h. Prior to exosome isolation, the cells were incubated in medium supplemented with 10 % exosome-depleted FBS obtained by overnight ultracentrifugation at 100,000 g to avoid the effects of bovine serum exosomes [
18].
Clinical samples
Samples of human blood plasma from 82 LUAD patients and 39 healthy control subjects, as well as 50surgically resected LUAD tissues were collected at Fujian Medical University Union Hospital (Fuzhou, Fujian, P.R. China). All the subjects provided their written informed consent, and the study protocol was approved by the Ethics Committee of Fujian Medical University Union Hospital.
Exosome isolation, purification. and characterization
Exosomes were isolated from LUAD cell-derived supernatants as previously described [
19]. The isolation method included a penultimate ultracentrifugation step (100,000 g for 70 min at 4 °C) needed to obtain the exosomes pellets, which were then washed with phosphate-buffered saline (PBS) and subjected to a second ultracentrifugation at 100,000 g for 70 min at 4 °C. When isolating exosomes from the plasma of LUAD patients, 1 mL of 0.8 μm-filtered blood plasma was purified by using Exosupur columns (Echo biotech, China) and a combination of size-exclusion chromatography (SEC) and ultrafiltration [
20]. The final pellets were resuspended in PBS for use in subsequent experiments. The quantities of exosomes isolated for use both
in vivo and
in vitro experiments were determined using a BCA Protein Assay Kit (Beyotime, Shanghai, China). For transmission electron microscope (TEM) studies, purified exosomes pellets were fixed with 4 % paraformaldehyde (PFA) and placed onto copper mesh Formvar-coated grids. The grids were then stained with 2 % phosphotungstic acid for 5 min. The samples were observed using a JEOL TEM (JEM1230, JEOL, Tokyo, Japan). The size distribution and concentration of exosomes were detected by Nanoparticle Tracking Analysis (NTA). The data were analyzed with a ZetaView PMX 110 Nanoparticle Analyzer (Particle Metrix, Meerbusch, Germany) and the corresponding software ZetaView 8.04.02.
Scratch assays
Cells were seeded on each side of an Ibidi culture insert (Ibidi, Munich, Germany) with a separation between each side of the well, and allowed to create a confluent monolayer. The cells were then treated with exosomes (10 µg/mL) for 2 h following removal of the insert, and were cultured with 1 % exosome-depleted FBS [
21]. The first image (0 h) was obtained and then cultured in a 37℃ incubator for 24 h prior to acquisition of the second image. Percent of wound closure was calculated as: migrated cell surface area/total surface area × 100.
Cell invasion and migration assays
Cell invasion assays were performed in triplicate using a 24-well Transwell plate (Corning, NY, USA) coated with Matrigel (Corning). The cell migration assays were constructed using a similar protocol, but without the Matrigel coating. Cells were treated with 10 µg of exosomes for 2 h and then seeded in the upper chamber of a Transwell plate containing serum-free RPMI-1640 medium. The bottom chamber was filled with 10 % FBS medium. After 24 h of incubation, migrated or invaded cells to the bottom of the insert membrane were fixed with 4 % paraformaldehyde and stained with 1 % crystal violet.
RNA extraction and real-time quantitative polymerase chain reaction (RT-qPCR) analysis
The total RNA of cells and total miRNA of exosomes were extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and a miRNeasy Mini Kit (Qiagen, Germany), respectively. Prior to the isolation of exosomal RNA, 25 fmol of
C. elegans cel-miR-39 standard RNA (RiboBio, Guangzhou, China) was added to each sample as a spike-in control [
22]. cDNA synthesis was performed using a PrimeScript™ RT reagent kit (Takara, Shiga, Japan) according to the manufacturer’s instructions. The miRNAs were reverse transcribed using a Mir-X miRNA First-Strand Synthesis Kit (Takara, Shiga, Japan). RT-qPCR was performed using SYBR Green GoTaq Master Mix (Promega, Madison, WI, USA) and the products were analyzed on an ABI 7500 thermocycler (Applied Biosystems, Foster City, CA, USA). The relative levels of mRNA and miRNA in cells were normalized to those of β-actin and U6, respectively, and the levels of exosomal miR-31-5p was normalized to those of cel-miR-39.
GW4869 and RNase A treatment
To block exosomes release, LUAD cells exposed to hypoxic condition were pretreated with GW4869 (an inhibitor of exosome formation and release, Sigma-Aldrich, St. Louis, MO USA) at a concentration of 10 µM for 24 h. For RNase A treatment, exosomes were incubated with RNase A (Takara, Shiga, Japan, final concentration of 10 µg/mL) alone or combined with 0.1 % Triton X-100 for 20 min.
Library preparation and sequencing
A 100 ng sample of total RNA was ligated with sequencing adapters using a TruseqTM Small RNA sample prep Kit (Illumina, San Diego, CA, USA), and then reversed transcribed to cDNA and PCR amplified. sRNA sequencing libraries were prepared using a TruSeq Small RNA Sample Preparation Guide (Illumina, San Diego, CA, USA) at Shanghai Majorbio Bio-pharm Technology Co. Ltd, and sequenced on an Illumina HiSeq platform. For quality control, low-quality bases (Sanger base quality of < 20) of the 3’ end were trimmed using in-house perl scripts. All identical sequences of sizes ranging from 18 to 32 nt were counted and eliminated from the initial data set. Through a BLAST search of the miRbase and the Rfam database, perfectly matched sequences were used to count, and then remove non-miRNA sequences (rRNA, tRNA, snoRNA, etc).
Western blot analysis
Total proteins were isolated with RIPA lysis buffer (Beyotime) and quantified with a BCA kit. An equal amount of protein lysate from each sample was separated on SDS-polyacrylamide gels, and the protein bands were transferred onto polyvinylidene difluoride membranes that were subsequently blocked with 5 % skimmed dry milk. Next, the membranes were incubated with the following antibodies: anti-HIF-1α (Hypoxia Inducible Factor 1 Subunit Alpha, Abcam, Cambridge, MA, USA), anti-CD63 (Abcam), anti-TSG101 (Tumor Susceptibility 101, Abcam), anti-GAPDH (glyceraldehyde 3-phosphate dehydrogenase)anti-Flotillin (Abcam), anti-SATB2 (Abcam); anti-phosphorylated (p)-MEK (Abcam), anti-MEK (Abcam), anti-phosphorylated (p)-ERK (Affinity Biosciences, Cincinnati, OH, USA), anti-ERK (Affinity); anti-E-cadherin (Cell Signaling Technology, Danvers, MA, USA), anti-N-cadherin (Cell Signaling Technology), anti-Vimentin (Cell Signaling Technology) at 4℃ overnight and then incubated with a secondary antibody for 1 h at room temperature. β-actin (Abcam) and Calnexin (Boster, Wuhan, China) were served as a loading control and a negative control for exosomal markers, respectively. The chemiluminescent signals were detected and the immunostaining intensity of each band was quantified using a Bio-Rad ChemiDoc XRS system (Bio-Rad, Hercules, CA, USA).
Animal studies
Athymic nude mice (male, 5 weeks of age) were purchased from the SLUAD Laboratory Animal, Shanghai, China, and housed in a specific pathogen-free environment. Twenty mice were randomly assigned to four groups (5 mice per group). A549 cells were intravenously injected into the tail vein of each mouse (2 × 106 cells/200 µL/per mouse) to establish lung metastatic model. Next, equal amounts (10 µg) of purified exosomes were also intravenously injected into the tail vein of each mouse twice per week for 4 weeks. All the mice were sacrificed after 8 weeks, and their lungs were collected and subjected to hematoxylin and eosin (HE) staining. The numbers of metastatic nodules in the lung tissues were also counted. The protocols for all animal studies were approved by the Ethics Committee of the Animal Center of Fujian Medical University.
Luciferase activity assay
The 3’UTR segments of SATB2 genes were amplified by the PCR and inserted into SATB2 vectors. Co-transfection with the vectors and miR-31-5p mimics was performed using Lipofectamine 3000 (Invitrogen). Luciferase activity was measured after 36 h of incubation using a Dual-Luciferase Reporter Assay Kit (TransGen Biotech, Beijing, China) as described in the manufacturer’s instructions.
Cell transfection and internalization of exosomes
SATB2 siRNA, miR-31-5p mimics and inhibitors were obtained from RiboBio Company (Guangzhou, China). Cells were seeded into 6-well plates for 24 h before transfection. When the cells reached 60 % confluence, they were transfected with SATB2 siRNA, miRNA mimics or inhibitors using Lipofectamine 3000. To determine whether the exosomes could be internalized by LUAD cells, the exosomes were labeled with PKH67 (Sigma) as the manufacturer described. Briefly, purified exosomes were suspended in a mixture consisting of 500 µL of Diluent C and 4 µL of PKH67, and then incubated in the dark for 4 min at room temperature. Next, 2 mL of 0.5 % bovine serum albumin (BSA) was added to stop staining, and the mixture was subsequently centrifuged at 100,000 g for 1 h. The labeled exosomes were suspended with PBS and then used immediately or stored at 20℃. LUAD cells were treated with labeled exosomes for 8 h at 37℃, and the nuclei were stained by Dapi (Sigma) for 30 min. After staining, the samples were washed twice with PBS and observed under a fluorescence microscope.
Immunohistochemistry
Samples of metastatic lung tissue from each mouse were fixed in 10 % formalin and paraffin embedded. Section (4 μm thickness) were placed onto glass slides. The slides were incubated at 37℃ with mouse anti-E-cadherin and mouse anti-Vimentin for 2 h; Slides then were incubated with HRP-conjugated anti-mouse antibody for 30 min and stained with DAB.
Statistical analysis
All the statistical analyses were performed using IBM SPSS Statistics for Windows, Version 22.0 software (IBM Corp, Armonk, NY, USA). Graphpad Prism 8.0 software (Graphpad Prism Software, Inc, San Diego, CA, USA) was used to create graphs. Results are presented as a mean value ± standard deviation. One-way analysis of variance or the unpaired Student’s t-test was used for statistical analysis. Relationships between exosomal miR-31-5p and other factors were analyzed using the χ2 test. Receiver operating characteristic (ROC) curves were generated to assess the diagnostic value of exosomal miR-31-5p. The association between SATB2 and miR-31-5p was explored by Pearson’s correlation. A P-value<0.05 was considered to be statistically significant.
Discussion
Hypoxia has been universally considered as a key hallmark of solid tumors [
6]. Metastasis, a multifaceted process in LUAD, requires tumor cells to migrate toward regions of sufficient oxygen to strengthen their development, as based on necessary structural changes that need to occur, including EMT [
28]. Emerging evidence suggests that hypoxia stress induces the release of exosomes that carry proteins and miRNAs into TME, and thereby influences exosome-mediated intercellular communication [
29]. In the present study, we isolated exosomes from the supernatants of hypoxic and normoxic LUAD cells. Combined validation with TEM, NTA and western blotting revealed that hypoxia-induced stress increased the production of exosomes, which is consistent with findings in studies conducted with other malignant tumors [
7,
30,
31]. The transfer of exosomes can change the phenotypes of the recipient cells. We observed that exosomes derived from hypoxic cells induced EMT process and enhanced the migration and invasion capabilities of recipient cells. A previous study showed that exosomal miR-23a derived from hypoxic lung cancer cells promoted angiogenesis and vascular permeability [
14]. However, the phenotype change that contributed to tumor progression was different from that observed in our study, which suggests that there are other informative biomolecules that greatly influence LUAD.
An increasing body of evidence indicates that miRNAs are major bioactive factors abundantly enriched in exosomes [
23,
24]. We compared the miRNA profiles of hypoxic and normoxic exosomes derived from LUAD cells. Initial screening and subsequent verification studies conducted with different types of LUAD cell lines indicated that exosomal miR-31-5p might be a critical factor that influences LUAD development. Exosomal non-coding RNAs can stably exist in body fluids and be protected from degradation by ribonuclease because they are protected by the exosomal membrane [
32]. Therefore, we observed that miR-31-5p expression was decreased by treatment with RNase and Triton X-100. Similarity, administration of GW4869 also decreased the levels of miR-31-5p. These data confirmed that miR-31-5p was encapsulated in HExo.
miR-31-5p is ectopically upregulated in various malignant tumors. Previous studies showed that miR-31-5p expression was correlated with lung tumorigenesis and the Warburg effect [
33,
34]. However, the effects of miR-31-5p in different space distributions, and especially in exosomes, have not been completely elucidated in LUAD. We discovered that exosomal miR-31-5p could significantly accelerate the metastasis of LUAD cells
in vitro and
in vivo, and the effects could be reversed by administration of an miR-31-5p inhibitor. Those findings suggest that miR-31-5p is the bioactive factor in hypoxic exosomes, which induce development of aggressive LUAD phenotypes.
SATB2, an AT-rich binding transcription factor, plays a vital role in tumorigenesis and regulation of gene expression by influencing chromatin structure remodeling [
35]. It is well established that SATB2 acts as a negative regulator of EMT in colorectal cancer and non-small-cell lung carcinoma [
36]. MiRNAs commonly bind to the 3’UTR of their downstream target mRNA molecules to repress gene expression. We predicted the potential target of exosomal miR-31-5p by conducting a bioinformatics analysis. We subsequently confirmed that SATB2 mRNA was the target of exosomal miR-31-5p by conducting luciferase assays and western blot studies. Notably, knockdown of SATB2 suppressed the migration and invasion of LUAD cells, and that effect was increased by transfection with an miR-31-5p inhibitor. Moreover, studies using the xenograft mouse model further validated our hypothesis. Unfortunately, ROC analysis showed that the synergistic effect of miR-31-5p and SATB2 in LUAD clinics is limited, which indicated that there may exist other axis in controlling LUAD metastasis.
MEK/ERK signaling regulates SATB2-reversed osteogenic differentiation in bone mesenchymal stem cells [
37]. To explore the underlying mechanism by which exosomal miR-31-5p mediates tumor progression, we focused on factors which activate MEK/ERK pathway. Surprisingly, our findings showed that upregulation of exosomal miR-31-5p decreased the levels of SATB2 expression and remarkably accelerated EMT process and phosphorylation of MEK and ERK, which aberrantly activates MEK/ERK pathway by addition of miR-31-5p in HExo. Therefore, we believed that hypoxia leads to abundance of miR-31-5p in HExo, which activates MEK/ERK signaling in normoxic cells to promote malignant transformation.
Exosomes released by cancer cells can enter the circulatory system. Recently, informative noncoding RNAs encased in plasma exosomes were isolated and detected [
38]. Exosomal miRNAs are strongly correlated with clinicopathological features [
27]. Here, our results revealed that the levels of exosomal miR-31-5p were much higher in metastatic patients than in non-metastatic patients, which further validates the metastatic effect of exosomal miR-31-5p on LUAD cells. Moreover, we found that an upregulation of exosomal miR-31-5p was significantly associated with N and TNM stages in LUAD patients, and could be used to discriminate between LUAD patients and healthy control subjects. These findings suggest that exosomal miR-31-5p could serve as a diagnostic biomarker for LUAD. Unfortunately, we were unable to evaluate the prognostic value of exosomal miR-31-5p due to a lack of sufficient follow-up time.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.