Background
Lung cancer is the leading cause of cancer-related deaths worldwide [
1], and the current pathologic staging is inadequate to predict outcome for patient treatment. The development of molecular target therapy has improved the management of patients with lung cancer who are at high risk of relapse following surgery. One of the most promising classes of molecular markers in tumor prognosis is the small noncoding RNAs, or microRNAs (miRNAs) [
2‐
4]. MiRNAs, acting as oncogenes or tumor suppressors, have been shown to regulate the expression of hundreds of targeted genes at the posttranscriptional level and are implicated in the pathogenesis and therefore prognosis of human cancers [
5‐
9]. For non-small cell lung cancer (NSCLC) in particular, previous reports have indicated that miRNA expression patterns could be potential biomarkers used for diagnosis, prognosis, and personalized therapy [
4,
8‐
11]. In addition, some studies found that human serum or plasma contained large numbers of stable miRNAs and that the expression profiles of some specific circulating miRNAs could be useful in the diagnosis and prognosis of cancer [
12‐
14].
In our study, we performed a miRNA expression array by quantitative real-time reverse transcription polymerase chain reaction (RT-qPCR) and identified the expression of members of the miR-183 family (miR-96, miR-182, and miR-183) when comparing primary NSCLC tumor with adjacent normal lung tissues. The miR-183 family is located on human chromosome 7 and members of this family have been identified as potential oncogenes in several tumor types, including medulloblastomas [
15], breast cancer [
16], prostate cancer [
17], hepatocellular tumors [
18], and colon cancer [
19,
20], as well as lung cancer [
21,
22]. However, the role that these miRNAs play in the diagnosis and prognosis of lung cancer patients remains unknown.
We compared the expression levels of the miR-183 family in lung cancer cell lines with normal lung cells, and in primary tumor tissues and sera from NSCLC patients with normal volunteers. Our results indicate for the first time that members of the miR-183 family expressed in tumors and sera may be potential biomarkers in the diagnosis and prognosis of human lung cancer.
Methods
Sample population
NSCLC and matched adjacent noncancerous tissues were collected from patients undergoing lung resection surgery from January 2008 to May 2008 at Zhoushan Hospital, Zhejiang Province, China. The details of patients were shown in Table
1. Eligible samples were obtained from primary lung cancer that had not received any preoperative radiotherapy or chemotherapy, and in which there was no co-existing disease. Moreover, we excluded tissue blocks of mixed histology, or specimens with insufficient tumor material. The subtypes included 36 squamous-cell carcinoma and 34 adenocarcinoma. All human materials were obtained with patients' informed consent and the Ethical Review Committee of Zhoushan Municipal Government of China approved this study.
Table 1
Clinico-pathological Characteristics of 70 Patients with NSCLC
Mean age | |
< 60 | 34 |
≥60 | 36 |
Sex
| |
Male
|
56
|
Female
|
14
|
Smoking | |
Nonsmokers | 24 |
Current smokers | 46 |
Tumor size
| |
0-3 cm
|
18
|
> 3 cm
|
52
|
Histological classification | |
Adenocarcinoma | 34 |
SCC | 36 |
Invasion to lung membrane
| |
Yes
|
17
|
No
|
53
|
differentiation | |
Mod-well | 46 |
Poor | 24 |
Lymph node
| |
Negative
|
38
|
Positive
|
32
|
Stage classification | |
StageI | 36 |
StageII,III and IV | 34 |
Upon removal, the surgical specimens were immediately transported to the clinical pathology laboratory, where each sample was placed in a cryovial and flash-frozen in liquid nitrogen within 30 minutes and then stored at -80°C until analyzed. Sera from all NSCLC patients and healthy volunteers were also collected. There was no significant difference in gender or age between NSCLC patients and healthy volunteers. All cases were reviewed by two pathologists and diagnoses were confirmed according to the criteria recently established by the National Comprehensive Cancer Network (NCCN).
Cell lines and cell culture conditions
The following cell lines were cultured individually in RPMI-1640 medium (Gibco): A549, H1299, and SPC-A1 human lung adenocarcinoma; 95C and 95D human giant-cell lung carcinoma; NCI-H466 human small cell lung carcinoma; NCI-H460 human large-cell lung carcinoma; and human bronchial epithelia. In addition, SK-MES1 human squamous-cell lung carcinoma cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco). Both media were supplemented with 10% fetal bovine serum (Gibco), 2 mM L-glutamine, 100 IU/mL penicillin, and 100 mg/mL streptomycin. Cells were incubated in 5% CO2 at 37°C.
RNA isolation
Total RNA isolated from 6 NSCLC and 6 matched adjacent noncancerous tissues by the Trizol (Invitrogen) method were prepared for miRNA microarray according to the manufacturer's instructions. miRNA isolated from cells, tissues, and sera were obtained using the miRNA Isolation Kit and mirVana PARIS Kit (Applied Biosystems, Foster City, CA, USA), according to the manufacturer's protocol. The specific sources of miRNAs were: 106 to 107 cells (described above), 100 mg of tissue from each of 70 NSCLC patients and matched adjacent noncancerous tissues, and circulating miRNAs from 600 μL of serum from each of 70 NSCLC patients and 44 normal volunteers. The RNA concentration was measured by a NanoDrop ND-1000 spectrophotometer (nm readings: A260/280 > 2.0, A260/230 > 1.8; NanoDrop Technologies).
MicroRNA microarray and data analysis
Three to five micrograms of total RNA were labeled by ligating the fluorescent RNA-linker 5'-cytidine bisphosphate-cyanine 3 (pCp-Cy3) to the 3' end of miRNAs. Slides were then incubated with the labeled RNA and washed. Five hundred and nine miRNAs passed the initial screening criteria of the normalized median fluorescence signal. The images were analyzed using SpotReader software (Niles Scientific, Portola Valley, CA). The significance of differences in expression levels was assessed by a two-sided paired t-test within the significance analysis of microarrays (SAM). P < 0.05 indicated a significant difference.
RT-qPCR of miRNA derived from lung cancer cell lines and lung primary tissues
The amounts of the miRNAs in lung tumors, analyzed by microarray, were quantified by RT-qPCR using TaqMan MicroRNA Assay Kits (Applied Biosystems, USA). Briefly, the reverse transcription (RT) reaction was carried out with a TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems) according to the instruction of the protocol. One to ten nanograms of total RNA per 15 μL RT reaction were processed at 16°C for 30 min, 42°C for 30 min, and 85°C for 5 min. Following the RT, quantitative real-time PCR was performed in an ABI 7500 Real-Time PCR system (Applied Biosystems) at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for one minute.
The cycle threshold (Ct) values were calculated with SDS 2.0.1 software (Applied Biosystems). The average expression levels of miRNAs in tissues and sera were normalized with U6 small nuclear RNA (snRNA) and U48 snRNA using the 2
-ΔΔCt method [
23]. The mean Ct value in three candidate miRNAs was calculated, excluding outliers (i.e., replicates with a Ct differing by > one cycle from the median). If Ct
U6, ave and Ct
U48, ave were not within 20 and 32 cycles, the assay was repeated. Samples with low U6 or U48 snRNA levels were not included in this study.
Statistical analyses
The statistical analyses were performed with Graphpad Prism 5.0 statistical software. The data were examined according to the degree of homogeneity. The paired t-test, unpaired t-test, or Mann-Whitney U-test was used to analyze the correlation between the miRNA expression levels and clinical-pathological features of the patients. The paired sample t-test was used to compare the differences in miRNA expression between lung tissues and sera. All data were expressed as the mean ± standard error of the mean (SEM). Binary logistic regression was used to assess associations between the miR-183 family and the clinical-pathological features of the patients. Survival analysis was performed with the Kaplan-Meier method, and the log-rank test was used to compare survival times between groups and was adjusted according to age and tumor stages. The Cox hazard regression model was used to analyze the miRNAs as risk factors for lung cancer. A probability (P)-value < 0.05 was considered statistically significant.
Discussion
Lung carcinoma is one of the most common malignancies in the world, and is the leading cause of cancer death in men and women in the United States with a dismal five-year survival rate (< 15%) [
1]. Although recent advances have been made in improving diagnosis and treatment strategies, the prognosis of lung cancer patients remains unchanged and early detection is critical in improving survival duration. Studies have shown that miRNAs play an important role in tumorigenesis and they have been considered potential biomarkers for early diagnosis and prognosis in a wide variety of human cancers [
6,
14,
25,
26]. More importantly, circulating miRNAs might act as noninvasive blood-based biomarkers for cancer diagnosis [
14,
27]. One recent report demonstrated that miRNAs in tissues and plasma could have a critical role as molecular predictors of lung cancer development and management outcome [
28].
The upregulation of members of the miR-183 family (miR-96, miR-182, miR-183) has been reported in several types of cancers [
15‐
20]. Consistent with our results, members of the miR-183 family were elevated in lung cancer, revealed by novel rank-based statistical methods using microarrays [
21,
22,
29]. miR-96, a known onco-miRNA, targets the tumor suppressor gene forkhead box O3 (
FOXO3; a transcription factor that plays important roles in tumorigenesis) by binding to the 3
'-UTR of
FOXO3 mRNA and thereby mediating protection against apoptosis and promoting cell survival [
16,
30]. Studies have also reported that miR-27a, miR-96, miR-182, and miR-183 coordinately regulated the expression of forkhead box O1 (
FOXO1) by directly targeting the mRNA
FOXO1 3'-UTR [
31,
32]. In addition, miR-183 functioned as an onco-miRNA, shown to target the tumor suppressor transcription factor early growth response 1 (
EGR1), and phosphatase and tensin homolog (
PTEN) to promote tumor cell migration [
19].
Consistent with other reports [
6,
19,
32], we showed that compared to adjacent non-cancerous lung tissues and normal volunteers, members of the miR-183 family were highly expressed in lung cancer primary tissues and sera. Moreover, we found that the expression levels of miR-96, miR-182, and miR-183 were higher in A549 and 95D cells than in H1299 and 95C cells, respectively. This implies that the miR-183 family might participate in tumor metastasis [
19,
30,
33,
34]. Consistent with this, other studies demonstrated that overexpressed miR-96 and miR-182 in melanoma cell lines and breast cancer cell lines resulted in enhanced oncogenic properties as well as invasion and metastasis
in vitro [
30,
33]. Sarver
et al. [
19] showed that a reduction in miR-183 led to a decrease in migration in colon cancer and synovial sarcoma cell lines. Abraham
et al. [
34] reported that overexpression of miR-183 in medullary thyroid cancer predicted lymph node metastasis.
We note that opposite results have also been reported. According to one study, expression of miR-96 decreased cell invasion and migration in pancreatic cancer [
35]. Another report demonstrated that miR-182 suppressed cell proliferation of lung cancer cell line A549
in vitro and tumor growth
in vivo through its interference with the target gene cortactin (
CTTN) by epigenetic modification [
36]. Others found that overexpression of miR-183 inhibited migration and invasion of lung cancer and breast cancer cells, by targeting the protein ezrin [
22,
37]. These discrepancies might be due to different cancer cell lines used, and suggest that miRNA might have distinct functions depending upon cell type [
38].
Our results demonstrate an association between the expressions of miR-183 and miR-182 and metastasis, invasion, advanced clinical stage, and tumor sizes of NSCLC. Together with other reports, they imply that miR-182 and miR-183 might play a role in lung cancer invasion and metastasis, and progression of NSCLC [
28,
30]. Moreover, in our study patients with high tumor or serum levels of miR-96, miR-182, and miR-183 had shorter overall post-operative survival times. The hazard ratios adjusted for gender, age and tumor stage indicated that these miRNAs could be used independently for evaluating the prognoses of NSCLC patients.
In addition, our results demonstrated that serum miRNAs might be used as potential biomarkers for diagnosis and prognosis of lung cancers. Under lung cancer condition, miRNAs entered the circulation mainly from tumor cells [
27]. One study showed that miRNAs in tissue specimens were not detected by Solexa sequencing in sera [
39]. By comparing the expressions of miR-96, miR-182 and miR-183 in tumors and sera, our results suggest that the levels of serum and tumor miR-96 were correlated, while those of miR-182 and miR-183 were not. These results are consistent with reports from Mitchell [
27] and Hu [
39]. The actual role and functions of these miRNAs need to be further investigated.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
WYZ, XGL, and YKZ were involved in the conception and design of the study. WYZ, JYH, DDC, and YYH were involved in the provision of study material and patients. WYZ and XGL performed the data analysis and interpretation. WYZ wrote the manuscript. YKZ approved the final version. All authors read and approved the final manuscript.