Background
Lung cancer is the leading cause of cancer-related deaths among humans worldwide, and non-small cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer cases [
1]. Even when NSCLC is diagnosed at an operable stage and treatment with chemo- or radio-therapies is applied, the incidence of recurrence and metastasis remains high with a median survival of less than 10–12 months. At present, gemcitabine-based chemotherapy is an established multimodal therapy for NSCLC treatment. However, its clinical efficacy remains limited by the development of acquired resistance following tumor metastasis and relapse. Increasing evidence indicates that micro RNAs (miRNAs) are involved in gemcitabine resistance and tumor progression in several malignant cancers. In breast cancer, miR-21 was found to induce epithelial–mesenchymal transition (EMT) and gemcitabine resistance through activation of the PTEN/AKT pathway [
2]. In addition, a recent study found that downregulation of miR-130a-3p can induce the EMT and malignancy of hepatocellular carcinoma cells via inhibition of Smad4 [
3]. Also, miR-101-3p reverses gemcitabine resistance by inhibiting ribonucleotide reductase M1 in pancreatic cancer [
4].
miR-222 was initially revealed to induce polarization of tumor-associated macrophages in epithelial ovarian cancer [
5]. Further studies indicated that elevated levels of miR-222-3p can promote cell proliferation and tumor metastasis via targeting ERα in endometrial carcinoma [
6], whereas down-regulation of miR-222 restrains cell proliferation and migration by activating SIRT1 in prostate cancer [
7]. Moreover, the miR-221 and miR-222 cluster can regulate TRAIL resistance and enhance tumorigenicity through PTEN and TIMP3 down-regulation [
8]. All these findings suggest that miR-222 plays an important role in tumorigenesis and tumor metastasis. However, the precise role of miR-222-3p in modulating malignant progression and chemo-resistance in lung cancer is largely unknown.
As post-transcriptional regulators, miRNAs negatively regulate gene expression by binding directly to the 3’untranslated regions (UTRs) of target mRNAs, thereby inducing mRNA degradation or repression of protein translation. The suppressor of cytokine signaling 3 (SOCS3), a primary member of the SOCS family of proteins, is a negative feedback regulator of the JAK/STAT signaling pathway [
9]. Recent studies have revealed SOCS3 overexpression can inhibit cell proliferation and anchorage-independent growth in breast cancer cells [
10], whereas down-regulation of SOCS3 reflects a poor prognosis in gastric cancer and hepatocellular carcinoma [
11,
12]. However, whether SOCS3 acts as the target of miR-222-3p, thereby contributing to gemcitabine resistance in NSCLC, remains unclear.
Exosomes are nano-sized microvesicles (30–100 nm in diameter) that are formed by the inward budding of late endosomes released into the extracellular environment upon fusion with the plasma membrane [
13,
14]. Exosomes can be produced by many cell types, including T cells, B cells, dentritic cells, epithelial cells, and tumor cells [
15]. Emerging studies have demonstrated that tumor cells can secrete exosomes into the extracellular space, and the exosomes then migrate far away from their initial position, transferring various types of functional effectors (RNAs, miRNAs, and proteins) to recipient cells [
16,
17]. Moreover, studies have shown exosomes play pleiotropic roles in regulating tumor progression, metastasis, chemoresistance and immune dysregulation [
18‐
21]. Several studies have indicated that abundant cell-free miRNAs from exosomes within biological fluids can induce a miRNA-mediated repression of target genes, and these exogenous miRNAs can function effectively in various physiological and pathological processes as either oncogenes or tumor suppressors in recipient normal or tumor cells [
22,
23]. Examples include the ability of exosomes derived from hypoxic oral squamous cell carcinoma cells to deliver miR-21 to normoxic cells and elicit a pro-metastatic phenotype as well as the ability of exosomic miR-29a to bind Toll-like receptors in nearby tumor-associated macrophages and trigger a protumoral inflammatory reaction in lung cancer [
24]. However, the mechanisms underlying the association of exosomes with gemcitabine resistance during tumor progression and relapse in NSCLC remain poorly understood. Multiple studies have suggested that tumor-derived exosomes can, at least in part, mediate the malignancy of either normal or tumor cells through transported miRNAs, and several exosomal miRNAs now serve as cancer biomarkers and therapeutic targets [
25‐
27]. Given that chemo-resistance can not be induced in all tumor cells simultaneously, we hypothesized that miRNAs might be transported from gemcitabine-resistant (GR) cells to non-resistant cells through exosomes to spread gemcitabine resistance and accelerate tumor progression in NSCLC.
In this study, we evaluated the content of exosomes shed by A549-GR cells and found they are rich in miR-222-3p, which readily entered recipient cells via both caveolin- and lipid raft-dependent endocytosis. We then found that this exosomic transfer of miR-222-3p promoted cell growth and metastasis by down-regulating SOCS3 expression. In addition, our results indicated that the level of exosomic miR-222-3p from patient sera was inversely associated with prognosis and tumor metastasis following gemcitabine therapy. Therefore, our findings suggest a novel molecular mechanism and a potential prognostic biomarker of tumor resistance to gemcitabine in human NSCLC.
Methods
Patients and tissue samples
Fifty patients diagnosed with NSCLC were enrolled in the Department of Thoracic Surgery of the First Affiliated Hospital of Jilin University (Changchun, Jilin, China) from January 2012 to December 2014. The clinical and pathologic characteristics of the patients are summarized in Table
1. All patients received routine gemcitabine-platinum chemotherapy (no surgery and completed 2–4 cycles). Informed consent was obtained from each patient, and the protocol was approved by the Ethics Committee of the First Affiliated Hospital of Jilin University.
Table 1
Clinicopathologic parameters and circulating exosomal miR-222-3p level in 50 NSCLC patients
Overall | 50 | 25 | 25 | |
Age |
≥ 60 | 24 | 14 | 10 | 0.396 |
< 60 | 26 | 11 | 15 |
Gender |
Male | 40 | 21 | 19 | 0.725 |
Female | 10 | 4 | 6 |
Histology |
Squamous | 22 | 9 | 13 | 0.393 |
Adenocarcinoma | 28 | 16 | 12 |
Differentiation |
Well | 3 | 1 | 2 | 0.820 |
Moderate | 18 | 10 | 8 |
Poor | 29 | 14 | 15 |
cTNM stage |
II | 6 | 4 | 2 | 0.489 |
III | 22 | 12 | 10 |
IV | 22 | 9 | 13 |
cT stage |
cT1–2 | 27 | 16 | 11 | 0.256 |
cT3–4 | 23 | 9 | 14 |
cN stage |
cN0 | 12 | 9 | 3 | 0.095 |
cN+ | 38 | 16 | 22 |
cM stage |
cM0 | 28 | 20 | 8 | 0.001** |
cM1 | 22 | 5 | 17 |
Response after Gemb
|
PR | 27 | 19 | 8 | 0.006** |
SD | 7 | 3 | 4 |
PD | 16 | 3 | 13 |
Cell culture and transfection
All human NSCLC cell lines were purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco’s Modified Eagle Medium (DMEM)/F-12 medium with 10% FBS (Gibco) at 37 °C in 5% CO
2. The gemcitabine-resistant cell line (A549-GR) was established as rapamycin-resistant cells as described previously [
28]. Briefly, gemcitabine-sensitive A549 parental cells (A549-P) were exposed to gradually increasing concentrations of gemcitabine (LC laboratories) from 10 nM initially up to 10 μM over a 6-month period. Transfection of miR-222-3p mimics/inhibitor (Ambion, Carlsbad, CA), or GV-144-SOCS3 plasmid (GeneChem, Shanghai, China) was carried out using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. Stable A549-P-
luc2, A549-P-KD-
luc2, A549-GR-
luc2, and A549-GR-KD-
luc2 cell lines were established using lentivirus pGC-FU-LUC-IRES-puromycin carrying negative control or oligonucleotides against miR-222-3p (GeneChem), and cells were continually incubated with puromycin (2.5 μg/ml, Sigma) to allow for acquired resistance.
Exosome isolation
Exosomes were isolated by differential centrifugation of conditioned media collected from A549-P/GR cells. Cells were grown in medium containing 10% exosome-depleted fetal bovine serum (FBS, SBI System Biosciences, Palo Alto, CA, USA). After 3 days’ incubation, the conditioned medium was initially cleared of cellular debris, and the dead cells were removed with two sequential centrifugation steps at 2500 g for 10 min at 4 °C. The supernatants were then spun at 110,000×g for 70 min at 4 °C. The pellets were washed with phosphate-buffered saline (PBS) and the ultracentrifugation protocol was repeated. The final exosome pellet was resuspended in PBS. To isolate exosomes from human peripheral blood, samples were centrifuged twice at 2000×g for 10 min to separate the plasma from red blood cells, and exosomes were isolated via ultracentrifugation as described above. Protein amounts in exosomes were quantified using the bicinchoninic acid assay.
Transmission electron microscopy (TEM)
First 20 μg of exosomes was loaded onto parafilm, and then a 300 mesh copper grid (Agar Scientific Ltd., Stansted, UK) was placed over the drop for 2 min. After the excess liquid was removed by blotting with filter paper, the grid was negatively stained with 2% phosphotungstic acid (PTA) for 2 min and examined at 80 kV with a JEM-1200 EXII TEM (JEOL, Ltd., Tokyo, Japan).
Immunofluorescence assay
Purified exosomes were labeled with green fluorescent linker PKH-67 (Sigma) according to the manufacture’s protocol. Cells were seeded in 8-well chamber slides (8000 cells/well) and pre-treated with pharmacological inhibitors for 2 h. Then 5 μl of PKH67-dyed exosomes were added before a 4-h incubation to allow internalization. Finally, slides were washed twice with PBS, fixed with 4% paraformaldehyde (PFA), and mounted with DAPI-containing mounting media (Vector Labs). Images were taken using a Zeiss LSM 780 (Zeiss, Jena, Germany) confocal microscope.
Western blot analysis
Total proteins were extracted using an extraction buffer with a protease inhibitor cocktail (Thermo Scientific), and equal amounts of protein (50 μg) were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore). Membranes were blocked and probed with primary antibodies overnight including those against Alix, TSG101, CD81, SOCS3, JAK2(T/P), Stat3(T/P), Bcl-2, Bax and Bcl-XL (Cell Signaling Technology and Abcam). After incubation with secondary antibodies, the membranes were developed for chemiluminescence measurement.
Quantitative real-time polymerase chain reaction (PCR)
Total RNA was isolated from cells or exosomes using RNeasy Kit (Qiagen). cDNA was synthesized from 1 to 10 μg RNA using the TaqMan® MicroRNA Reverse Transcription Kit (Applied Biosystems). Aliquots of the reaction mixture were used for PCR with TaqMan® 2× Universal PCR Master Mix. All PCR experiments were performed in triplicate. The U6 RNA level was used as an internal control for data normalization.
Cell proliferation assay
Cells were seeded in 96-well plates at 8000 cells/well and cultured overnight. After treatment with exosomes or drugs for 48 h, MTS was added at 20 μl/100 μl medium. The absorbance was measured with a spectrophotometer (Bio-Rad Inc) at 490 nm, and cell growth inhibition was calculated using the equation: cell viability (%) = (At/Ac) × 100%, where At and Ac represent the absorbance in the treated and control cultures, respectively [
12].
Cells were trypsinized (single-cell suspension) and seeded into 6-well plates (800 cells/well) for culture in medium containing gemcitabine (1 μM), which was refreshed every 3 days. After 10 days of treatment, the medium was discarded, the cell colonies were stained with crystal violet (0.1% in 20% methanol) and imaged using a digital camera to record the results.
Luciferase reporter assay
The dual-luciferase vectors psiCHECK-SOCS3–3’UTR-WT and psiCHECK-rcmiR −222-WT were constructed by synthesizing the candidate seed sequences in the 3′-UTR of SOCS3 or the reverse complementary sequence of miR-222 and inserting them into the psiCHECK-2 vector. For mutant vectors, 3–4-bp mutations were introduced into the seed sequences. All plasmids were confirmed by DNA sequencing. For reporter assays, HEK-293 T cells were seeded in 24-well plates and transfected with 0.8 μg recombinant vectors alone or plus 30 nM mimics or inhibitors. Firefly and Renilla luciferase activities in cell lysates were measured 48 h later using the dual-Glo reporter assay system (Promega).
Cell migration assay
A wound-healing assay was performed to assess any influence of treatment on cell migration. Cells were seeded in 6-well plates to create a confluent monolayer, and then a scratch wound was made with a sterile pipette tip in a straight line. After rinsing with PBS, cells were incubated with medium containing exosomes (10 μg/well, 0.5% FBS). Images were captured under a microscope at 12 h or 24 h post-wounding.
Cell invasion assay
An invasion assay was performed using 6.5-mm Transwell chambers with 8-μm pores (Costar) according to the manufacturer’s instructions. Briefly, 1 × 105 cells in serum-free medium were seeded in the upper insert precoated with matrigel (1:4, BD Biosciences), and then 600 μl complete medium containing 10 μg exosomes was added to the bottom chamber as a chemoattractant. After incubation for 24 h, the upper surface of each membrane was cleaned, and cells adhered to the insert surface were fixed and stained with 0.5% crystal violet.
Anchorage-independent soft agar assay
Cells were suspended in 0.35% agarose and medium supplemented with 10% FBS, and the mixture was seeded in 6-well plates containing a basal layer of 0.6% agarose at 5000 cells/well. Then medium was replaced twice per week. After 3–4 weeks of routine culture, colonies were stained with 0.005% crystal violet for longer than 1 h, and images were captured using an SZX12 microscope (Olympus, Japan). Viable colonies larger than 0.1 mm in diameter were counted.
Animal experiments
Animal experiments were performed according to protocols approved by the Institutional Animal Care and Use Committees of Jilin University. Female severe combined immunodeficient (SCID) mice at ~6 weeks old (Vital River, Beijing, China) were injected 1 × 106 viable tumor cells via the tail vein. The mice were treated with PBS or exosomes (n = 8 each). Successful injections were confirmed by immediate luciferase imaging. For this, mice were anesthetized and injected intraperitoneally with luciferin (25 mg/ml in 100 μl PBS), and images were collected beyond 15 min post-injection using an IVIS-Lumina system (Caliper, USA). Light emission from animal tissue (photons/s) was measured using software provided by the vendor (Xenogen, Corp., Alameda, CA). Luciferase imaging was performed once a week for a total of 6 weeks unless significant morbidity and earlier euthanasia were required. Finally, mice were sacrificed, and tumor tissues were frozen in liquid nitrogen.
Statistical analysis
All statistical analyses were performed using GraphPad Prism software (San Diego, CA). Data are presented as mean value ± standard error, and the clinicopathological parameters were compared using the Fisher’s exact test. The statistical significance of differences between two groups was analyzed using two-tailed unpaired Student’s t-tests, and P < 0.05 was considered to be statistically significant.
Discussion
Gemcitabine is one of the most important chemotherapeutic drugs for human NSCLC treatment, and it improves survival through its abilities to inhibit DNA chain termination, cell cycle progression, apoptosis and metastasis [
32,
33]. Especially for elderly patients with advanced NSCLC, a mono-chemotherapy treatment with gemcitabine or docetaxel is the recommended option, because it offers survival and quality of life benefits with acceptable toxicity [
34]. However, side effects and acquired resistance still limit the clinical efficacy of gemcitabine therapy. Previous studies indicated that several resistance-related proteins, such as P-glycoprotein, topoisomerases, and thymidylate synthase, are involved in this process [
35]. Herein, we aimed to clarify the effects of exosomes from GR cancer cells on other sensitive cells as well as their effects on cancer cell proliferation, tumor metastasis, and chemoresistance in NSCLC treatment.
It is well recognized that several miRNAs are involved in tumor resistance to gemcitabine-based chemotherapy. Dhayat et al. demonstrated miR-138, miR-147b and miR-99a are upregulated, and miR-31 and miR-422a are downregulated in PANC-1-GR cells when compared with parental/sensitive cells [
36]. Furthermore, emerging evidence indicates that tumor-derived exosomes as vesicular carriers frequently transfer miRNAs into subcellular sites in recipient cells to induce repression of the expression of certain target genes [
23,
37]. However, fewer studies have focused on the roles of GR cell-derived exosomes in onco-miRNA transfer to sensitive cells, and their subsequent roles in the malignancy of receipt cells during NSCLC progression.
In this study, we first purified nano-sized exosomes that were abundantly present in tumor cell-conditioned media of both GR and parental NSCLC cells. Our results indicated that the internalization of A549-GR–Exo was mainly mediated via caveolin- and lipid raft-dependent endocytosis mechanisms, rather than micropinocytosis- or clathrin-dependent endocytosis (Fig.
1). Intriguingly, this is not completely consistent with its effects seen in other cells lines. Pancreatic cancer-derived exosomes can enter paraneoplastic β-cells through caveolin- mediated endocytosis or micropinocytosis [
38], and internalization of glioblastoma cell- derived exosomes was shown to be mediated by lipid raft-dependent endocytosis, suggesting that exosome internalization might be closely associated with cell phenotypes, which are mediated by different protein distributions or signaling activations on the cell membrane [
39].
In the present study, to elucidate the functions of exosomic miRNAs in gemcitabine resistance and tumor malignancy, a microRNA array was performed to identify 116 miRNAs that were differentially expressed between A549-GR and A549-P exosomes, including 45 upregulated and 70 downregulated miRNAs. Among these differentially regulated miRNAs, 23 miRNAs were found to be significantly up- or down- regulated by at least 5-fold between these two cell lines (
P < 0.01). Given the conflicting reports regarding whether exosomal miRNA profiles resemble those of parental cells [
40,
41], we compared the most remarkably upregulated miRNAs. miR-222-3p was upregulated in A549-GR exosomes to a level even higher than in donor A549-GR cells and could be internalized by receipt cells. We also found the level of miR-222-3p was elevated from A549-P to A549-GR cells, indicating that miR-222-3p might play an important role in tumorigenesis and progression (Fig.
2). Additionally, we observed some other miRNAs were altered in A549-GR-Exo or recipient cells, such as miR-135b, miR-10a, and miR-221-3p were upregulated in both exosomes and recipient cells, whereas miR-181a was specially increased in exosomes, miR-224-5p was upregulated in GR cells and exosomes, but failed to transfer to recipient cells effectively. These findings suggest the exosomic miRNA profile does not resemble that of donor or recipient cells completely. We expect that some miRNAs may not be suitable for packaging into exosomes, and some miRNAs may be less stable in the cytoplasm than in exosomes and tend to degrade.
Although miR-222 has been demonstrated to be overexpressed in several types of cancer and to promote tumor progression and drug resistance via genetic or epigenetic mechanisms [
8,
42,
43], the functions of exosomic miR-222-3p and the molecular mechanisms by which miR-222-3p modulates drug resistance and tumor progression have yet to be revealed. Our data demonstrate that miR-222-3p KD in recipient cells could inhibit gemcitabine resistance, colony formation, tumor invasion, and migration in vivo and in vitro (Figs.
2 and
5). Further investigation revealed that miR-222-3p block the expression of SOCS3 by directly binding with the 3’-UTR of SOCS3 to upregulate the expression levels of Stat3 and Bcl-2 (Fig.
3). Constitutive activation of Stat3 had been confirmed to inhibit apoptosis and promote migration in various tumors, including lung cancer [
44]. Moreover, we also clarified that the exogenous expression of SOCS3 neutralized the activity of SOCS3 and alleviated GR-Exo–induced malignancy, subsequently (Fig.
4). Bcl-2 families are key regulators of cell anti-apoptosis, the increased ratio of Bcl-2/Bax better elucidates Bcl-2, acts as an downstream effector of SOCS3, contributes to miR-222-3p induced cell anti-apoptotic (Fig.
3e). Additionally, we observed Bcl-XL (anti-apoptotic protein) has a slight decrease, our explanation is that some other mechanisms might be involved in regulating the Bcl-XL expression, especially, when cells received so many exosomic-miRNAs simultaneously, and this data suggest that Bcl-2, but not Bcl-XL plays key roles in inducing cellular anti-apoptosis in cells which involving increased miR-222-3p. Regretfully, our current study could not explain why gemcitabine induced the upregulation of miR-222-3p in NSCLC, although some studies reported that nuclear factor (NF)-κB and c-Jun can induce the expression of miR-221&222 in prostate carcinoma and glioblastoma cells. In addition, over-activated TEAD1 upregulates miR-222 expression via physically binding to its promoter in gastric cancer cells, and a DNA damaging agent can cause dysregulation of miRNA expression at the transcriptional level [
45,
46].
Additionally, our results demonstrated that A549-P–derived exosomes failed to promote malignant phenotypes of A549-P cells, and GR-Exo had no strong influence on the proliferation of receipt A549-P cells, which is not consistent with the previous finding that tumor-secreted exosomes derived from BT-474 cells can accelerate the proliferation of BT-474 cells [
47,
48]. With increasing evidence that tumor-derived exosomes can confer either anti-tumorigenic or pro-tumorigenic effects [
49], one probable explanation for these seemingly conflicting effects may relate to the complex interactions between exosomes, recipient cells, and intercellular environmental factors [
50]. On the other hand, the original proliferation rate of A549-P cells is very fast, so it would be difficult to obtain the further promotion. Meanwhile, our data reveals even A549-GR cells involving high level of miR-222-3p, the proliferation rate only increases less than 20% when compared with parental cells (Fig.
2e), which is helpful to explain why A549-GR-derived exosomes show only marginal survival benefit. Importantly, this data is not contradicted with our conclusion that GR-Exo indeed deliver the functional miR-222-3p into recipient cells, and certainly, the role of miR-222-3p in enhancing the drug-resistance and motility is more potent than that in promoting cell proliferation in NSCLC.
Considering gemcitabine mono-therapy is usually administered to elderly patients, to exclude the limitation of age, we recruited NSCLC patients who received routine gemcitabine-platinum chemotherapy (4–6 courses). In parallel, our results demonstrated that platinum has no influence on miR-222-3p expression with or without gemcitabine treatment in advance, which excludes the potential disturbance of platinum, indicating that NSCLC patients with a higher level of circulating exosomic miR-222-3p might experience worse tumor metastasis (P = 0.001). Moreover, a higher level of circulating exosomic miR-222-3p correlated with a failed response to gemcitabine therapy (PD), whereas a lower level correlated with an acceptable response (PR).
In summary, our study provides evidence that miR-222-3p was upregulated in both A549-GR–derived exosomes and their donor cells. Upon internalization, these miR-222-3p-rich exosomes induce a more malignant phenotype in recipient sensitive cells by activating the SOCS3/Stat3 signaling pathway. Moreover, circulating exosomic miR-222-3p may function as a biomarker for predicting the response to gemcitabine of patients with advanced NSCLC.