Increased expression of the RNA-binding motif protein 47 predicts poor prognosis in non-small-cell lung cancer.

Lung cancer is the leading cause of cancer-associated mortality worldwide. In China, in particular, lung cancer mortality has markedly increased and is likely to continue to rise. RNA-binding proteins are pivotal to the development and progression of a variety of cancer types, including non-small cell lung cancer (NSCLC). RNA-binding motif protein 47 (RBM47) has been found to act as a tumor suppressor in breast cancer and NSCLC. However, to the best of our knowledge, RBM47 expression in NSCLC tissues has yet to be investigated. Analysis via the online database, Gene Expression Omnibus, revealed that RBM47 was upregulated in NSCLC and associated with pathological type, suggesting that RBM47 may play different roles in lung adenocarcinoma and lung squamous cell carcinoma. In the present study, the expression of RBM47 was examined by immunohistochemistry in 175 pairs of tumor and adjacent non-cancerous tissues resected from patients with NSCLC. The results indicated that the expression of RBM47 was significantly increased in NSCLC samples compared with that in the matched non-cancerous samples. Furthermore, RBM47 expression was higher in Xuanwei compared with that in non-Xuanwei NSCLC, suggesting that RBM47 is a more sensitive biomarker in Xuanwei NSCLC, and that it may serve as a candidate therapeutic target. In addition, RBM47 expression was associated with the pathological type, however not with the age, sex, lymph node metastasis, pT stage or pathological Tumor-Node-Metastasis stage of the patients. The increased expression level of RBM47 may indicate a worse overall survival rate for patients with NSCLC. In addition, multivariate survival analysis showed that the Xuanwei area is associated with poor prognosis for patients with NSCLC. In conclusion, the present study revealed that the upregulation of RBM47 accelerated the malignant progression of NSCLC, indicating that RBM47 may be a potential biomarker for NSCLC progression and a therapeutic target for NSCLC. Copyright: © Li et al.

Introduction

Lung cancer is one of the most prevalent malignancies worldwide (1). In 2014, ~781,500 new cases of lung cancer were reported in China, with an incidence of 5713/105, higher than the world standard rate of 3663/105 (2). A notable increase in mortality rates of lung cancer was observed in China from 5.46/105 in the 1970s, 17.54/105 in the 1990s, to 30.83/105 in the 2000s (3,4). Xuanwei has one of the highest mortality rates of lung cancer in China, and Xuanwei lung cancer (XWLC) is a term used to describe lung cancer cases recorded in Xuanwei City and Fuyuan County (5), which are known for extensive combustion of locally sourced bituminous (smoky) coal in Yunnan, China. XWLC is characterized by its high incidence and mortality rates, and a significant difference in incidence in terms of the sex of the patient (6), and a lower patient age at onset (7–9). The epidemiologic studies demonstrated that lung cancer mortality rates in Xuanwei were 4–5 times higher than the average in China (10). The mortality rate of lung cancer in Xuanwei in 2011–2013 was 82.53/105 and 62.62/105 for males and females, respectively. The age-standardized death rate (ASMR) of lung cancer among males in Xuanwei was three times of that in rural areas in China, and six times higher among females (6,11). A previous study demonstrated that indoor air pollution-releasing and cancer-causing substances, such as polycyclic aromatic hydrocarbons, also produce inorganic particulate matter that can lead to the damage of alveolar cells and activate NF-κB signaling pathways, ultimately leading to tumorigenesis (12). Non-small cell lung cancer (NSCLC) accounts for 80–85% of all lung cancer cases and small cell lung cancer (SCLC) for the remaining 15–20% (13). Lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) are the most common pathological types of NSCLC (14). In advanced NSCLC, due to its aggressiveness and resistance to treatment, the 5-year relative survival rate is ~17.4% (15), despite efforts to discover novel therapies, such as PD-1 inhibitors (16). In order to accurately diagnose and treat NSCLC, it is imperative that the molecular mechanisms of tumorigenesis and progression are accurately investigated, and novel therapeutic strategies are identified.

Previous investigations have indicated that RNA-binding proteins (RBPs) also serve a prominent role in tumorigenesis and cancer progression (17–19). Since they contain ≥1 RNA-binding domains, RBPs can bind to specific mRNA targets and participate in post-transcriptional RNA regulation, including splicing, polyadenylation and translation (20). For example, tumor spheres of lung cancer cells highly express ELAV-like RNA binding protein 1 (HuR), and lung cancer stem cells maintain stemness through the HuR/microRNA(miR)-873/cyclin-dependent kinase 3 (CDK3) and HuR/miR-125a-3p/CDK3 axes (21). Similarly, insulin-like growth factor 2 mRNA-binding protein 3 has been shown to promote lung tumorigenesis by attenuating p53 protein stability (22).

As a novel RBP, RNA-binding motif protein 47 (RBM47), located on chromosome 4p14, has a high affinity for poly-A, -C and -URNAs, and produces two transcripts due to alternative splicing during protein synthesis (23). In zebra fish embryogenesis, RBM47 serves a critical role in head formation and embryonic patterning through the Wnt8a signaling pathway (23). RBM47 also binds to Nanog transcript in mouse embryonic stem (ES) cells, regulating ES cell pluripotency (24). Regarding regulation, the basic machinery of C-to-U RNA editing comprised of RBM47 and apolipoprotein B mRNA editing enzyme catalytic subunit 1 (APOBEC1) to enable post-transcriptional modification (25).

Previous studies have demonstrated that RBM47 plays differing roles in breast cancer and lung cancer (26,27). It has been reported that RBM47 suppresses Wnt activity, inhibiting breast cancer progression and metastasis (26). It was also recently reported that RBM47 inhibits Nrf2 activity in LUAD, retarding tumor growth (27). However, to the best of our knowledge, the detection of RBM47 expression in lung cancer tissues has not been previously reported. The present study aimed to investigate the expression of RBM47 in tumor tissues, compared with matched adjacent non-neoplastic tissues from patients with NSCLC. The association between RBM47 expression and the clinicopathological characteristics of patients with NSCLC, including age, sex, pathological types, pT stage, lymph node metastasis, pathological Tumor-Node-Metastasis (pTNM) stage and overall survival rates were also analyzed.

Materials and methods

Gene expression omnibus (GEO) data from the Affymetrix platform

RBM47 expression data were retrieved from the GEO database (http://www.ncbi.nlm.ih.gov/geo/), using the Affymetrix platform (https://www.thermofisher.com/cn/zh/home/life-science/microarray-analysis/affymetrix.html) for the following six datasets; GSE27262 (25 NSCLC samples and 25 marginal samples collected for microarray analysis), GSE7670 (27 paired NSCLC samples and the corresponding adjacent non-cancerous samples as a control), GSE19804 (60 NSCLC tumor samples and 60 adjacent normal tissues), GSE10245 (40 LUAD and 18 LUSC samples), GSE14814 (71 LUAD and 52 LUSC samples) and GSE21933 (11 LUAD and 10 LUSC samples). Paired Student's t-test was employed to compare RBM47 expression levels between the different groups. P<0.05 was considered to indicate a statistically significant difference. Additional details regarding the data are presented in Table I.

Table I.

Detailed information from the Affymetrix Human Genome platform for the Gene Expression Omnibus datasets.

Series accession	Organism	Type	Affymetrix human genome platform	Reference
GSE7670	Homo sapiens	Expression profiling by array	GPL96 [HG-U133A] Affymetrix Human Genome U133A Array	(55)
GSE19804	Homo sapiens	Expression profiling by array	GPL570 [HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array	(56)
GSE27262	Homo sapiens	Expression profiling by array	GPL570 [HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array	(57)
GSE10245	Homo sapiens	Expression profiling by array	GPL570 [HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array	(58)
GSE14814	Homo sapiens	Expression profiling by array	GPL96 [HG-U133A] Affymetrix Human Genome U133A Array	(59)
GSE21933	Homo sapiens	Expression profiling by array	GPL6254 Phalanx Human OneArray	(60)

Clinical data collection

The clinical processes were approved by the Ethics Committee of the Third Affiliated Hospital of Kunming Medical University, and written informed consent was provided by all patients. Of the 175 cases, 107 were male and 68 were female, with a median age of 57 years (range, 27–78 years). XWLC fit the following criteria, three generations all residing in a coal region and >10 years of coal-burning history. Patients with NSCLC were divided into the XWLC group (72 patients) and non-XWLC group (103 patients). 175 pairs of samples were fixed in 10% formalin for 24 h at room temperature. Paraffin-embedded primary NSCLC tissues and paired non-cancerous lung tissues (5 cm away from the tumor edge) were obtained between December 2008 and October 2015 from the Third Affiliated Hospital of Kunming Medical University (Kunming, China). The stage of the tumor specimens was determined according to the 7th edition of the TNM Classification for Lung Cancer (28). American Joint Committee on Cancer (AJCC) staging (28) consists of clinical and pathological staging. Pre-operative CT examination was not performed on 20 patients due to economic reasons, resulting in an inaccurate reflection of the patient's clinical stage and therefore, the pathological stage was used. The patients included in the present study had been diagnosed with stage IA-IIIA NSCLC. Patients who had received preoperative radiation, chemotherapy or biotherapy were excluded from the study. None of the patients had been diagnosed with multiple primary cancers in other organs or tissues. The clinical data were derived from the patients' medical records. All patients had undergone complete surgical resection, including primary tumor plus mediastinal and bronchial lymph node dissection. All 175 patients with NSCLC had effective and accurate follow-up data (from 2–110 months). The follow-up data of patients was updated using medical records, interviews and via telephone. According to the National Comprehensive Cancer Network guidelines (29), adjuvant chemotherapy is not recommended for stage IA, but may be useful in a subset of patients at stage IB. For stage II and IIIA, which is a therapeutically challenging and controversial subset of lung cancer, four cycles of platinum-based adjuvant chemotherapy to address micro-metastatic disease is recommended.

Immunohistochemistry (IHC) and hematoxylin and eosin (H&E) staining

IHC was performed in order to determine RBM47 expression levels and distribution patterns of 175 pairs of NSCLC tissues. IHC was performed in accordance with previous studies (30,31). Tissue samples were cut into 4-µm-thick sections and incubated in Rabbit polyclonal RBM7 antibody (1:500; cat. no. HPA006347; Sigma-Aldrich; Merck KGaA) for 12 h at 4°C, in order to detect RBM47 protein expression. The sections were washed three times with PBS and incubated with avidin-biotin complex (Vector Laboratories, Inc.). 3–3′-diamino-benzidine was used as the chromogen for the color reaction. A total of 30 µl of the secondary antibody in the DAB kit (Dako; Agilent Technologies, Inc.), with original concentration, was added to the sections and incubated at 37°C for 30 min. H&E staining was used to evaluate histology (32). Paraffin-embedded tissues samples were cut into 4-µm-thick sections and incubated at 60°C for 1 h. The samples were subsequently deparaffinized in xylene at 60°C for 15 min and rehydrated prior to incubation with hematoxylin (original concentration; Fuzhou Maixin Biotech Co., Ltd.) at room temperature for 2 min, followed by incubation with eosin (original concentration; Guangzhou RiboBio Co., Ltd.) at room temperature for 1 min. Subsequently, the slices were dehydrated in a step-up ethanol series at room temperature and mounted in neutral gum prior to analysis using an optical microscope (Leica Microsystems GmbH; magnification, ×100 and ×200). The negative controls were obtained by replacing the primary antibody with non-immunized sheep serum (Fuzhou Maixin Biotech Co., Ltd.).

Evaluation of IHC

RBM47-positive cells displayed brownish yellow granules in the cytoplasm and nucleus. Two independent pathologists from The Third Affiliated Hospital of Kunming Medical University, who were blinded to the clinical parameters, determined the scores using a fluorescent inverted microscope (magnification, ×100 and ×200). The extent and intensity of immunoreactivity was scored in a semi-quantitative manner. The extent of immunoreactivity was scored as follows: 0, 0% immunoreactive cells; 1, <5% immunoreactive cells; 2, 5–50% immunoreactive cells and 3, >50% immunoreactive cells. The intensity of immunoreactivity was scored as follows: 0, negative; 1, weak; 2, intermediate and 3, strong. The samples were grouped according to the sum of the immunoreactivity score as follows: 0, negative; 1–2, weak; 3, moderate and 4–6, strong (31). Final immunoreactivity scores >0 were considered positive, and those at 0 were considered negative.

Statistical analysis

All data are expressed as the mean ± standard deviation. All experiments were performed in triplicate and statistical analyses were conducted using SPSS (version 18.0; SPSS, Inc.). The paired Student's t-test was used in order to determine the differences in RBM47 protein expression levels between different groups in the GEO database. The χ2 test was used in order to determine the differences in the RBM47 protein expression levels between NSCLC and non-cancerous lung tissues. The Kaplan-Meier method was used to plot survival curves, and Cox's proportional hazards regression model was used to analyze multivariate survival. P<0.05 was considered to indicate a statistically significant difference.

Results

Analysis of GEO data

In order to examine the expression patterns of RBM47 in NSCLC, the GSE27262 dataset was first analyzed, which included 25 NSCLC and 25 marginal samples collected for microarray analysis. As presented in Fig. 1A, the RBM47 expression level was significantly higher in the cancerous samples compared with the marginal samples in GSE27262 (P=0.0102). In order to verify the result, two additional datasets (GSE7670 and GSE19804; Fig. 1B and C), were utilized and data were consistent (all P<0.05). These data revealed that RBM47 was upregulated in NSCLC. Furthermore, the potential association between RBM47 expression and the pathological type of NSCLC was investigated using the GSE10245 dataset in Fig. 1D, which included 40 LUAD and 18 LUSC samples. The data analysis results indicated that the expression level of RBM47 was significantly different between LUAD and LUSC samples (P<0.001). Furthermore, two additional datasets (GSE14814 and GSE21933; Fig. 1E and F), demonstrated similar results (all P<0.05). Collectively, RBM47 expression exhibited an association with NSCLC and pathological type, suggesting that RBM47 may serve different roles in LUAD and LUSC.

Figure 1.

Analysis of RBM47 gene expression in NSCLC from GEO datasets. A total of six datasets, including (A) GSE27262, (B) GSE19804, (C) GSE7670, (D) GSE10245, (E) GSE14814 and (F) GSE21933 were acquired from the GEO repository and subjected to data analyses. (A-C) RBM47 expression was significantly increased in NSCLC compared with non-cancerous lung tissues. (D-F) GSE10245, GSE14814 and GSE21933 datasets were analyzed to confirm whether RBM47 protein exhibited significantly different expression levels between the LUAD and LUSC pathological types. *P<0.05, ***P<0.001 vs. NSCLC; ***P<0.001 vs. LUAD. RBM47, RNA-binding motif protein 47; NSCLC, non-small-cell lung cancer; GEO, Gene Expression Omnibus; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma.

Clinicopathological characteristics of patients with NSCLC

A total of 175 histologically confirmed, paraffin-embedded NSCLC tissues were investigated. The clinicopathological characteristics are presented in Table II. The majority of the samples were adenocarcinoma (131 cases, 74.9%) and the remaining were squamous cell carcinoma (44 cases, 25.1%). Tumor stage was determined according to the 7th edition of the AJCC cancer staging manual (19): 86 cases (49.1%) were stage I, 34 (19.4%) stage II and 55 (31.5%) stage III. XWLC fit the following conditions: i) Three generations all residing in a coal region; and ii) >10 years of coal-burning history. The patients were divided into 72 cases of XWLC (41.1%) and 103 cases of non-XWLC (58.9%). The median follow-up time was 39 months (range, 2–110). Of the 175 patients, 71 (40.6%) succumbed to NSCLC; the remaining 104 (59.4%) were still alive at the end of the follow-up. Detailed data are presented in Table II.

Table II.

Clinicopathological characteristics of patients with non-small cell lung cancer.

Characteristics	Patient, n (%)
Total	175
Sex
Male	107 (61.1)
Female	68 (38.9)
Age, years
<60	99 (56.6)
≥60	76 (43.4)
Area
Xuanwei	72 (41.1)
Non-Xuanwei	103 (58.9)
Pathological type
Adenocarcinoma	131 (74.9)
Squamous carcinoma	44 (25.1)
Tumor size (pT stage)
T1	94 (53.7)
T2	60 (34.3)
T3/T4	21 (12.0)
Lymph node metastasis (pN stage)
Negative	104 (59.4)
Positive	71 (40.6)
AJCC stage at diagnosis (pTNM)
I	86 (49.1)
II	34 (19.4)
IIIA	55 (31.5)

pT, pathological tumor size; pN, pathological lymph node stage; AJCC, American Joint Committee on Cancer; TNM, Tumor-Node-Metastasis.

RBM47 is upregulated in NSCLC tissues

The GEO dataset indicated increased expression of the RBM47 gene in NSCLC cases. The present study, to the best of our knowledge, was the first to measure RBM47 protein expression and subcellular localization by IHC staining, in an independent cohort of 175 patients with NSCLC with cancer tissues and their matched adjacent non-cancerous tissues. IHC staining revealed that RBM47 was localized to the nucleus and cytoplasm. In addition, it was observed that the staining density of RBM47 in the NSCLC tissues was more intense and had a broader distribution than that observed in non-cancerous tissues. In summary, RBM47 was upregulated in NSCLC. Representative IHC staining images of RBM47 protein expression in the tumor and non-cancerous lung tissues are shown in Fig. 2, and the difference in staining intensity between specimens is presented in Fig. 3. These results were consistent with the analysis of the GEO database.

Figure 2.

Representative immunohistochemistry images of RBM47 protein expression in NSCLC and non-cancerous lung tissues via hematoxylin & eosin staining. (A and C) RBM47 was positively expressed in NSCLC cells, but (B and D) negatively or weakly expressed in adjacent lung epithelial cells (all used magnification, ×100). RBM47, RNA-binding motif protein 47; NSCLC, non-small-cell lung cancer.

Figure 3.

Staining intensity of RBM47 differed between the specimens. Representative images demonstrated different staining intensities of RBM47 expression, as follows: (A and B) Negative staining, (C and D) weak staining, (E and F) moderate staining and (G and H) strong staining (A, C, E and F magnification, ×100; B, D, F and H magnification, ×200). RBM47, RNA-binding motif protein 47.

RBM47 expression is associated with the pathological type of NSCLC

To further determine the role of RBM47 expression in NSCLC, the association between RBM47 and the clinical characteristics of patients with NSCLC was analyzed. Table III shows the number and percentage of RBM47-positive samples in the two groups. It was also observed that RBM47 stained positive in 75.4% (132/175) of NSCLC tissues, but only 21.7% (38/175) in the matched adjacent non-cancerous tissues, which in turn was significantly lower compared with that in the NSCLC samples (P<0.001). The expression of RBM47 was associated with pathological type (P<0.001), as it was detected in 84% (110/131) of the samples collected from LUAD tissues, but only 50% (22/44) of LUSC tissues. No association was observed between RBM47 and the age, sex, tumor size, pT stage, lymph node metastasis and pTNM stage in patients with NSCLC (Table III). This result indicated that RBM47 expression was associated with the pathological type of NSCLC, which was consistent with the results of the GEO profiles.

Table III.

RBM47 expression in patients with NSCLC and its associations with different characteristics.

		RBM47 expression
Variables	Patient, n	Negative expression, n (%)	Positive expression, n (%)	P-value[a]
Tissues				<0.001
Normal lung tissues	175	137 (78.3)	38 (21.7)
Primary tumors of NSCLC	175	43 (24.6)	132 (75.4)
Sex				0.329
Male	107	29 (27.1)	78 (72.9)
Female	68	14 (20.6)	54 (79.4)
Age, years				0.125
<60	99	20 (20.2)	79 (79.8)
≥60	76	23 (30.3)	53 (69.7)
Pathological type				<0.001
Adenocarcinoma	131	21 (16.0)	110 (84.0)
Squamous carcinoma	44	22 (50.0)	22 (59.0)
Tumor size (pT stage)				0.784
T1	94	25 (26.6)	69 (73.4)
T2	60	13 (21.7)	47 (73.8)
T3/T4	21	5 (23.8)	16 (76.2)
Lymph node metastasis (pN stage)				0.578
Negative	104	24 (23.1)	80 (76.9)
Positive	71	19 (26.8)	52 (73.2)
AJCC stage at diagnosis				0.366
I	86	19 (22.1)	67 (77.9)
II	34	12 (35.3)	24 (64.7)
IIIA	55	12 (21.8)	43 (78.2)

According to the χ2 test. RBM47, RNA-binding motif protein 47; NSCLC, non-small cell lung cancer; pT, pathological tumor size; pN, pathological lymph node stage; AJCC, American Joint Committee on Cancer.

In order to compare RBM47 expression between XWLC and non-XWLC, patients with NSCLC were divided into the Xuanwei and non-Xuanwei groups. Patients in the Xuanwei group were all from Xuanwei city and Fuyuan County of the Yunnan Province. The association between the RBM47 expression and the clinicopathological characteristics of the two groups was subsequently investigated (Tables IV and V). The expression of RBM47 was only associated with the pathological type (P<0.001) in both the Xuanwei and non-Xuanwei groups. Samples of the Xuanwei group had a significantly higher RBM47 expression level than those in the non-Xuanwei group (P=0.031) (Table VI), suggesting that RBM47 was a more sensitive biomarker in Xuanwei NSCLC.

Table IV.

RBM47 expression in patients with NSCLC from the Xuanwei area.

		RBM47 expression
Variables	Patient, n	Negative expression, n (%)	Positive expression, n (%)	P-value[a]
Tissue				<0.001
Normal lung tissues	72	58 (80.6)	14 (19.4)
Primary tumors of NSCLC	72	12 (16.7)	60 (83.3)
Sex				0.329
Male	42	7 (16.7)	35 (83.3)
Female	30	5 (16.7)	25 (83.3)
Age, years				0.385
<60	55	8 (14.5)	47 (85.5)
≥60	17	4 (23.5)	13 (76.5)
Pathological type				0.007
Adenocarcinoma	73	8 (11.0)	65 (89.0)
Squamous carcinoma	9	4 (44.4)	5 (55.6)
Tumor size (pT stage)				0.296
T1	44	9 (20.5)	35 (79.5)
T2	19	1 (5.3)	18 (94.7)
T3/T4	9	2 (22.2)	7 (77.8)
Lymph node metastasis (pN stage)				0.814
Negative	52	9 (17.3)	43 (82.7)
Positive	20	3 (15.0)	17 (85.0)
AJCC stage at diagnosis				0.720
I	45	9 (20.0)	67 (80.0)
II	12	2 (16.7)	10 (83.3)
III	15	1 (6.7)	14 (93.3)

According to the χ2 test. RBM47, RNA-binding motif protein 47; NSCLC, non-small cell lung cancer; pT, pathological tumor size; pN, pathological lymph node stage; AJCC, American Joint Committee on Cancer.

Table V.

RBM47 expression in patients with NSCLC from the non-Xuanwei area.

		RBM47 expression
Variables	Patient, n	Negative expression, n (%)	Positive expression, n (%)	P-value[a]
Tissue				<0.001
Normal lung tissues	103	80 (77.7)	33 (22.3)
Primary tumors of NSCLC	103	32 (31.1)	71 (68.9)
Sex				0.278
Male	65	22 (33.8)	43 (66.2)
Female	38	9 (23.7)	29 (76.3)
Age, years				0.589
<60	44	12 (27.3)	32 (72.7)
≥60	59	19 (32.2)	40 (67.8)
Pathological type				0.001
Adenocarcinoma	68	13 (19.1)	55 (80.9)
Squamous carcinoma	35	18 (51.4)	17 (48.6)
Tumor size (pT stage)				0.605
T1	50	16 (32.0)	34 (68.0)
T2	41	12 (29.3)	18 (70.7)
T3/T4	12	3 (25.0)	9 (75.0)
Lymph node metastasis (pN stage)				0.623
Negative	52	15 (28.8)	37 (71.1)
Positive	51	17 (33.3)	34 (66.7)
AJCC stage at diagnosis				0.223
I	41	10 (24.4)	31 (75.6)
II	22	10 (45.5)	12 (54.5)
III	40	12 (30.0)	28 (70.0)

According to the χ2 test. RBM47, RNA-binding motif protein 47; NSCLC, non-small cell lung cancer; pT, pathological tumor size; pN, pathological lymph node stage; AJCC, American Joint Committee on Cancer.

Table VI.

Association between RBM47 expression and Xuanwei region in patients with NSCLC.

		RBM47 expression
Variables	Patient, n	Negative expression, n (%)	Positive expression, n (%)	P-value[a]
Xuanwei NSCLC	72	12 (16.7)	60 (83.3)	0.031
Non-Xuanwei NSCLC	103	32 (31.1)	71 (68.9)

According to the χ2 test. RBM47, RNA-binding motif protein 47; NSCLC, non-small cell lung cancer.

Association between RBM47 expression, and overall survival rates and prognosis in NSCLC

The association between RBM47 expression and overall survival in NSCLC was analyzed. The present data revealed that a high RBM47 expression level in patients with NSCLC was significantly associated with poor prognosis compared with a low RBM47 expression level (P=0.009; Fig. 4A). Furthermore, the potential differences in the prognostic roles of RBM47 between patients with Xuanwei and non-Xuanwei NSCLC were investigated (Fig. 4B and C). It was indicated that RBM47 did not serve as a prognostic factor for patients with Xuanwei NSCLC, since its quantity variance affected the statistical analysis. However, RBM47 was significantly associated with poor survival in patients with non-Xuanwei NSCLC (P=0.004).

Figure 4.

Analysis of overall survival for all patients by log-rank test according to RBM47 expression. (A) Survival curves of 175 patients with NSCLC. Negative RBM47 expression, n=43 and positive RBM47 expression, n=132. (B) Survival curves of 72 patients with Xuanwei NSCLC. Negative RBM47 expression, n=12 and positive RBM47 expression, n=60. (C) Survival curves for 103 patients with non-Xuanwei NSCLC. Negative RBM47 expression, n=32 and positive RBM47 expression, n=71. RBM47, RNA-binding motif protein 47; NSCLC, non-small-cell lung cancer.

The independent prognostic factors, including RBM47, location, age, sex, tumor size, pT stage, lymph node metastasis and pTNM stage, were analyzed in association with NSCLC patient survival using multivariate Cox's proportional hazards regression analysis. It was found that location, however not RBM47 expression, was an independent prognostic factor for NSCLC (P=0.001; Table VII). In conclusion, the present data demonstrated that the Xuanwei area is associated with poor prognosis in patients with NSCLC, and that RBM47 is a more sensitive prognostic biomarker in Xuanwei NSCLC.

Table VII.

Multivariate analysis of prognostic factors for patients with non-small cell lung cancer.

Prognostic factor	Hazard ratio	95% CI	P-value
Area (Xuanwei vs. Non-Xuanwei)	2.519	1.456–4.357	0.001
RBM47 expression (Positive vs. Negative)	0.721	0.408–1.275	0.261
Sex (Male vs. Female)	1.232	0.732–2.076	0.432
Age, years (≥60 vs. <60)	1.333	0.775–2.291	0.299
Pathological type (AC vs. SCC)	1.570	0.816–3.022	0.177
pT stage (T1-2 vs. T3-4)	1.220	0.62–2.401	0.564
Lymph node metastasis (Yes vs. No)	0.549	0.23–1.313	0.178
pTNM (I–II vs. IIIA)	1.480	0.574–3.82	0.417

CI, confidence interval; RBM47, RNA-binding motif protein 47; AC, adenocarcinoma; SCC, squamous cell carcinoma; pT, pathological tumor size; TNM, Tumor-Node-Metastasis.

Discussion

An increasing number of recent reports have indicated that RBPs serve an important role in cancer development (33–35). A number of studies have investigated RBPs using IHC staining, and demonstrated that they were abnormally expressed in cancer, compared with adjacent normal tissues, which was associated with patient prognosis (36–38). The RBP La-related protein 1 is a post-transcriptional regulator of ovarian cancer progression and chemotherapy resistance (39). Furthermore, the expression level of RBM4 is significantly lower in gastric cancer compared with adjacent non-cancerous tissues, and it's also associated with poor differentiation, lymph node status and distant metastasis (40). In addition, it was identified as a novel biomarker and an independent prognostic marker (40). In addition to protein expression levels, the biological functions of RBPs are varied, including RNA alternative splicing, stability, subcellular localization and translation (41,42). The roles of RBPs in cancer can therefore be diverse. RBM10 was reported to be consistently downregulated in LUAD, promoting NUMB mRNA exon 9 skipping to generate a NUMB isoform that blocks cell proliferation and inhibited Notch activity (43,44). Regarding mRNA stability, in multiple types of cancer (21,45,46), overexpressed Hu-Antigen R contributes to the accelerated proliferation of cancer cells by enhancing the stability of cell-cycle regulators containing cyclinA and cyclinB1 encoded by AU-rich element-containing mRNAs (47,48). Furthermore, RBM5, as a tumor suppressor in NSCLC, splices the pre-mRNA of multiple target genes, including the tumor suppressor protein p53, which participates in the induction of cell cycle arrest and apoptosis (49). Therefore, further examination on the functional role of RBPs may contribute to the diagnosis and treatment of human cancers.

In the present study, RBM47 expression was investigated in 175 paired tissue samples from patients with NSCLC. IHC data indicated that RBM47 was more highly expressed in cancer tissues compared with the matched adjacent tissues. In addition, the RBM47 expression levels were also elevated in the data from the three GEO datasets, suggesting that RBM47 may be involved in the development of NSCLC. The next investigation indicated that enhanced RBM47 expression was significantly associated with the pathological type of NSCLC, however not with the sex, age, tumor size, pT stage, lymph node metastasis or pTNM stage of patients with NSCLC. Subsequently, a higher expression level of RBM47 was detected in Xuanwei compared with non-Xuanwei NSCLC, which suggested that RBM47 was a more sensitive prognostic marker of Xuanwei NSCLC. As candidate oncogenes are generally associated with poor prognosis in patients with cancer (50), it was inferred that RBM47 may also be associated with the prognosis of those with NSCLC. Furthermore, the association between RBM47 expression and the overall survival of patients with NSCLC was analyzed. The data revealed that a high RBM47 expression was associated with a worse prognosis compared with a low RBM47 expression level in patients with NSCLC, including non-Xuanwei NSCLC. For patients with Xuanwei NSCLC, RBM47 did not serve as a prognostic factor, because its quantity variance affected the statistical analysis. However, a trend in prognostic differences was revealed by the survival curve. Unfortunately, due to the limited number of available specimens, the number of cases of XWLC could not be increased to improve the results of the statistical analyses. The results provided a greater understanding of the underlying role of RBPs in the development and progression of NSCLC. However, one limitation of the study was the lack of investigation into the role of RBM47 in lung cancer cell proliferation and other malignant processes, in vivo and in vitro. Further investigation is necessary to determine whether RBM47 promotes lung cancer and the exact underlying mechanisms.

Patients with breast cancer with high expression levels of RBM47 tend to achieve a relatively good clinical outcome, as demonstrated in a previous study (26). In experimental models, RBM47 regulated the target mRNA to inhibit the reinitiation and growth of breast cancer, rather than acting directly as a tumor suppressive gene (26). RBM47 bound to the 3′UTR of Dickkopf-related protein 1 mRNA to enhance its expression levels, which has been shown to inhibit breast cancer progression by antagonizing Wnt. Consistent with this observation, Sakurai et al (27) used a collection of published microarray data to show that the RBM47 expression may be associated with a longer survival time in patients with lung and gastric cancer. In addition to its tumor-suppressive role in lung cancer, RBM47 bound to the mRNA of kelch-like ECH-associated protein 1 and Cullin 3 to inhibit Nrf2 activity. Meanwhile, RBM47-knockdown accelerated tumor formation and metastasis in mouse xenograft models. The present study focused on the function of RBM47, without detecting the RBM47 expression levels in human lung cancer tissue samples. Taking the above into consideration, further research is required to clarify whether RBM47 indicates good or poor prognosis in lung cancer.

A recent study reported that miR-25 could directly target RBM47 to upregulate PI3K/Akt/mTOR signaling in melanoma (51). Interleukin (IL)-8, another target of RBM47, exhibited increased expression in buccal epithelial cells collected from healthy, non-smoking female residents of Xuanwei and Fuyuan, who burnt smoky and smokeless coal, as indicated in the genome-wide gene expression profiles. This suggested that the physiological response to smoky coal modulated pro-inflammatory events in smoky coal users (52). Moreover, RBM47 has been reported to serve an important role in promoting the regulatory functions of B cells by regulating IL-10 at the post-transcriptional level (53). Since RBM47 may serve various roles in different types of cancer, future studies should focus on the exact regulatory functions and the mechanism of RBM47 in the proliferation, migration and invasion of NSCLC cells. In addition to RBM47, other RBPs have been identified as crucial biomarkers or prognostic factors for NSCLC. For instance, the increased expression of hnRNPA2/B1 is associated with a poor prognosis in patients with NSCLC, and hnRNPA2/B1 activates prostaglandin G/H synthase 2 signaling in NSCLC cells to promote tumor cell growth (54).

In conclusion, the present study indicated that RBM47 is significantly upregulated in NSCLC tissues, as well as being a more sensitive biomarker in Xuanwei NSCLC, and that RBM47 expression is significantly associated with pathological type. Furthermore, the increased expression level of RBM47 may predict poor prognosis in patients with NSCLC. Taking the aforementioned into consideration it was therefore suggested that RBM47 promotes the malignant progression of NSCLC and may become a crucial biomarker and prognostic factor for patients with NSCLC.