Association Between Long Noncoding RNA MEG3 Polymorphisms and Lung Cancer Susceptibility in Chinese Northeast Population

Abstract

Long noncoding RNA (lncRNA) maternal-expressed gene 3 (MEG3) is associated with proliferation of various tumor cells and has decreased expression in many types of cancers. In this study, we aimed at demonstrating the association between MEG3 polymorphisms and the risk of lung cancer in northeast China. There were 526 lung cancer patients and 526 healthy controls included in this hospital-based case-control study. The genotyping of two polymorphisms, rs7158663 G > A and rs4081134 G>A, was performed by the Taqman allelic discrimination method. We found that MEG3 rs4081134-AA may be associated with the risk of lung cancer (AA vs. GG: adjusted odds ratio [OR] = 0.487, confidence interval [95% CI] = 0.257–0.897, p = 0.030; AA vs. AG+GG: adjusted OR = 0.522, 95% CI = 0.274–0.992, p = 0.047). Similar associations in several subgroups were found in subsequent stratified analysis. Further, there were no statistically significant interactions of rs4081134 polymorphism and smoking to lung cancer susceptibility. In addition, the associations between the MEG3 rs7158663 polymorphism and lung cancer susceptibility were not found. These results indicate that the MEG3 rs4081134 polymorphism was significantly associated with lung cancer susceptibility in the Chinese population.

Introduction

Lung cancer is the most frequent cancer in men and the second most common cancer in women. According to Cancer Incidence in Five Continents (CI5) Volume XI, published by the International Agency for Research on Cancer (IARC) and the International Association of Cancer Registries (IACR), the age-standardized incidence rate of lung cancer in Shenyang, China, was 52.4 (52.7 cases/100,000 men) for men and 28.4 (28.4 cases/100,000 women) for women from 2008 to 2012 (International Agency for Research on Cancer, 2017). This is similar to the overall incidence of lung cancer in China (Chen et al., 2016). Smoking is considered one of the major environmental risk factors for human lung cancer. Compared with nonsmokers, long-term smokers are estimated to be 10 to 30 times more likely to develop lung cancer (Samet, 1991). Nevertheless, 25% of lung cancer patients are nonsmokers in the world (Ferlay et al., 2010). The results suggest that the occurrence of lung cancer is not only related to environmental exposure but also related to the population's genetic susceptibility.

With no protein-coding capability, long noncoding RNAs (lncRNAs), longer than 200 nucleotides, participate in many biological processes in cells. Although more than 3000 lncRNAs have been identified, only a small part of lncRNAs were described in detail (Khalil et al., 2009; Ponting et al., 2009). The biological function of lncRNAs remains unclear. It has been reported that lncRNAs are involved in a wide range of biological processes through different mechanisms, such as chromosome dose compensation, epigenetic regulation, post-transcriptional regulation of mRNA splicing, nuclear and cytoplasmic transport, regulation of apoptosis and invasion, reprogramming, and serving as a marker of embryonic stem cell fate (Louro et al., 2009; Mercer et al., 2009; Wilusz et al., 2009; Loewer et al., 2010; Spitale et al., 2011; Huang et al., 2012; Wierzbicki, 2012; Kopp and Mendell, 2018). LncRNAs are emerging as an important component of carcinogenesis transcription. Therefore, we can reasonably predict that lncRNAs may have an effect on the progression of lung cancer.

The lncRNA maternal-expressed gene 3 (MEG3), discovered in recent years, is a tumor suppressor gene located in the chromosome 14q32 (Benetatos et al., 2011; Zhou et al., 2012). MEG3 is expressed in many normal tissues, but it is inhibited in primary human tumors, including glioma, hepatocellular carcinoma, meningioma, and bladder cancer (Anwar et al., 2012; Wang et al., 2012; Balik et al., 2013; Ying et al., 2013). More importantly, as a tumor suppressor gene, MEG3 is downregulated in lung cancer tissue, and the forced expression of MEG3 in lung cancer will lead to the decline of tumor cell proliferation and the increase of apoptosis rate (Liu et al., 2015; Xia et al., 2015; Yan-Hua et al., 2015; Kruer et al., 2016). These findings make us very interested in investigating whether MEG3 polymorphisms may also be involved in the risk of lung cancer.

It has been reported that single-nucleotide polymorphisms (SNPs) may play important roles in carcinogenesis by affecting gene expression and function (Yin et al., 2016, 2017; Gao and Wei, 2017). In addition to SNPs in protein-coding genes, polymorphisms in functional lncRNAs may also contribute to the development of cancer (Gao et al., 2018; Lv et al., 2018). Moreover, polymorphisms are widely distributed in lncRNAs and some polymorphisms may affect the expression and secondary structure of lncRNAs (Yin et al., 2009). There was one report that polymorphisms in the lncRNA MEG3 were associated with an increased risk of colorectal cancer (Cao et al., 2016). Another study showed that MEG3 genetic polymorphisms were associated with platinum-based chemotherapy toxicity in Chinese patients with lung cancer (Gong et al., 2017). In addition, a latest study suggested that MEG3 rs4081134 was associated with the risk of neuroblastoma in Chinese children (Zhuo et al., 2018). However, there were few studies on the association between MEG3 polymorphisms and the risk of lung cancer.

Considering the important roles of MEG3 in the development of cancer as well as the unclear effects of MEG3 polymorphisms in lung cancer risk, we performed a case-control study to analyze the association between MEG3 rs7158663 and rs4081134 and lung cancer susceptibility in northeast China. Both SNPs are tagging SNPs (tagSNPs) in MEG3 and may be located at transcription factor binding sites.

Materials and Methods

Study subjects

This study was a hospital-based case-control study conducted in the Shenyang city, which is situated in northeast China. According to the inclusion and exclusion criteria, there were 526 lung cancer patients and 526 healthy controls included in our study. Those 526 cases were recruited by the following criteria: (1) they were newly diagnosed with primary lung cancer, (2) they were confirmed by experienced pathologists, and (3) they were able to accept a 1.5-h interview. After the interview, 10 mL of venous blood was drawn from each subject. Individuals who smoked ≥100 cigarettes during their lifetime were defined as smokers, otherwise they were classified as nonsmokers. The controls were healthy participants included in the medical examination centers of the corresponding hospital. All of the investigations performed in studies involving human participants were in accordance with the ethical criteria of the Ethics Committee of China Medical University, and informed consent was signed by each participant.

SNP selection and genotyping

Using the data of Han Chinese from the 1000 Genome Projects, we selected the tagSNPs of MEG3 by the pairwise option of the Haploview 4.2 software (setting r ² ≥ 0.8, minor allele frequency >0.05). Then, we conducted a potential functional prediction of these tagSNPs. Finally, rs7158663 G > A and rs4081134 G > A were selected for analysis. Both the SNPs are located in transcription factor binding sites. Genomic DNA was isolated from venous blood by using the phenol-chloroform method. All of the genotyping was performed by the Taqman^® allelic discrimination (Applied Biosystems, Foster City, CA) with a commercially available primer probe set. PCR and fluorescence signal reading were performed on an Applied Biosystems 7500 FAST Real-Time PCR System. Genotyping results was conducted by using SDS software. Quality controls were carried out by including negative controls in each run and selecting 10% samples randomly. The same genotyping was performed twice on these selected samples by different investigators, and there is reasonable concordance between the two sets of results.

Statistical analysis

The differences in the distributions of selected variables between the lung cancer cases and controls were performed by the χ ² test and Student's t-test. The goodness-of-fit χ ² test was used to test Hardy–Weinberg equilibrium of the genotypes. Unconditional logistic regression analysis was performed to assess the association between two SNPs and lung cancer risk, adjusted by gender, age, and smoking status. The crude odds ratio (ORs), adjusted ORs, and their confidence interval (95% CIs) were also calculated. The Cochran–Armitage test was used for trend analysis. Statistical power was calculated by using “Quanto” software version 1.2.4 (University of Southern California, Los Angeles, CA). Statistical analysis was performed with SPSS 20.0 software (SPSS, Inc. Chicago, IL). All the tests were two-sided, and statistical significance was defined as a p-value less than 0.05.

Results

As shown in Table 1, a total of 1052 participants were included in the study, which contains 526 cases and 526 controls. In the case group, there were 424 non-small cell lung cancer patients, 74 small cell lung cancer patients, and 28 other types. Among the non-small cell lung cancer patients, 250 were adenocarcinoma and 157 were squamous cell carcinoma. As expected, the smoking exposure rate was 44.1% in the patients, whereas it was 29.3% in the controls, suggesting that the smoking exposure was a risk factor for lung cancer (p < 0.001). There was no significant difference in the distribution of gender between the two groups (p = 0.711). However, the t-test showed that the age distribution between the two groups was statistically significant, with the mean age being 59.88 ± 10.54 and 54.69 ± 15.97 in the cases and controls, respectively (p < 0.001). Therefore, we adjusted for age and sex and smoking status in the statistical analysis. The genotype frequencies of rs7158663 and rs4081134 in the control group were under the Hardy–Weinberg equilibrium (χ ² = 0.142, p = 0.706 for rs7158663; χ ² = 0.284, p = 0.594 for rs4081134).

Table 1.

Demographics of Cases with Lung Cancer and Controls

Variable	Case (n = 526) (%)	Control (n = 526) (%)	p
Age, year (mean ± SD)	59.88 ± 10.54	54.69 ± 15.97	<0.001
Age, year
≤57	193 (36.7)	267 (50.8)	<0.001
>57	333 (63.3)	259 (49.2)
Gender
Male	275 (52.3)	281 (53.4)	0.711
Female	251 (44.7)	245 (46.6)
Smoking status
Never	294 (55.9)	372 (70.7)	<0.001
Ever	232 (44.1)	154 (29.3)
Pathological type
Non-small cell lung cancer	424 (80.6)
Adenocarcinoma	250 (47.5)
Squamous cell carcinoma	157 (29.8)
Small cell lung cancer	74 (14.1)
Other	28 (5.3)

The associations between the two polymorphisms and risks of lung cancer and non-small cell lung cancer are shown in Table 2. MEG3 rs4081134 logistic regression analysis showed that carriers of AA genotype significantly reduced the risk of lung cancer and non-small cell lung cancer compared with homozygous GG (adjusted OR = 0.487, 95% CI = 0.257–0.897, p = 0.030 for lung cancer group; adjusted OR = 0.428, 95% CI = 0.206–0.889, p = 0.023 for non-small cell lung cancer group). What is more, the associations between rs4081134 polymorphism and reduced risk of lung cancer and non-small cell lung cancer were also observed in the recessive model (AA vs. AG+GG: adjusted OR = 0.522, 95% CI = 0.274–0.992, p = 0.047 for lung cancer group; adjusted OR = 0.450, 95% CI = 0.218–0.929, p = 0.031 for non-small cell lung cancer group). For the distribution of rs4081134, the χ ² test showed that the A allele could reduce the risk of lung cancer and non-small cell lung cancer compared with the G allele (the adjusted OR = 0.752, 95% CI = 0.608–0.930, p = 0.008 for lung cancer group; adjusted OR = 0.760, 95% CI = 0.607–0.952, p = 0.017). The difference was statistically significant. However, statistical analysis found that the MEG3 rs7158663 polymorphism was not associated with the risk of lung cancer. Table 3 described the results of subgroup analysis. In the lung adenocarcinoma group, rs4081134 showed significant results. In comparison with the GG genotype, the risk of lung adenocarcinoma was significantly decreased in AA genotype carriers (adjusted OR = 0.325, 95% CI = 0.121–0.869, p = 0.025). In the recessive model, the results of rs4081134 were also statistically significant (AA vs. AG+GG: adjusted OR = 0.344, 95% CI = 0.129–0.915, p = 0.033). The results of stratification by smoking exposure are presented in Table 4. In the nonexposure smoking group of rs4081134, the patients with the AG genotype had a lower risk of lung cancer, compared with the homozygous GG (adjusted OR = 0.708, 95% CI = 0.505–0.994, p = 0.046). Using the GG genotype as reference, there were also significant results in the dominant model (AG+AA vs. GG: adjusted OR = 0.679, 95% CI = 0.490–0.942, p = 0.020). Stratified by average age, the associations between two polymorphisms and lung cancer risk are presented in Table 5. There were no statistical correlations in all of the age subgroups. However, it was noteworthy that we observed a borderline significance in the dominant model of the rs4081134 subgroups of age >57 (p = 0.052).

Table 2.

Association Between the Two Single-Nucleotide Polymorphisms and Risks of Lung Cancer and Non-small Cell Lung Cancer

		Lung cancer			Non-small cell lung cancer
SNPs	Control (%) (n = 526)	Case (%) (n = 526)	OR ^a (95% CI)	p^a	Case (%) (n = 424)	OR ^a (95% CI)	p^a
Rs7158663
GG (ref)	289 (54.9)	268 (51.0)	1.00 (ref)		215 (50.7)	1.00 (ref)
AG	204 (38.8)	219 (41.6)	1.149 (0.885–1.493)	0.298	178 (42.0)	1.152 (0.874–1.519)	0.316
AA	33 (6.3)	39 (7.4)	1.314 (0.788–2.192)	0.296	31 (7.3)	1.286 (0.749–2.209)	0.362
AG+AA vs. GG			1.172 (0.912–1.505)	0.216		1.170 (0.898–1.525)	0.245
AA vs. AG+GG			1.238 (0.751–2.039)	0.402		1.210 (0.714–2.051)	0.479
G	782	755			608
A	270	297			240
Rs4081134
GG (ref)	314 (59.7)	349 (66.4)	1.00 (ref)		278 (65.6)	1.00 (ref)
AG	182 (34.6)	161 (30.6)	0.819 (0.625–1.073)	0.148	135 (31.8)	0.864 (0.650–1.148)	0.312
AA	30 (5.7)	16 (3.0)	0.487 (0.255–0.933)	0.030	11 (2.6)	0.428 (0.206–0.889)	0.023
AG+AA vs. GG			0.773 (0.596–1.002)	0.051		0.803 (0.610–1.056)	0.116
AA vs. AG+GG			0.522 (0.274–0.992)	0.047		0.450 (0.218–0.929)	0.031
G	810	859	1.00 (ref)		691	1.00 (ref)
A	242	193	0.752 (0.608–0.930)	0.008	157	0.760 (0.607–0.952)	0.017

Adjusted for age, gender, and smoking status.

CI, confidence interval; OR, odds ratio; SNPs, single-nucleotide polymorphisms.

Table 3.

Association Between the Two Single-Nucleotide Polymorphisms and Risks of Lung Adenocarcinoma and Squamous Cell Lung Cancer

		Lung adenocarcinoma			Squamous cell carcinoma
SNPs	Control (%) (n = 526)	Case (%) (n = 250)	OR ^a (95% CI)	p^a	Case (%) (n = 157)	OR ^a (95% CI)	p^a
Rs7158663
GG (ref)	289 (54.9)	130 (52.0)	1.00 (ref)		80 (51.0)	1.00 (ref)
AG	204 (38.8)	97 (38.8)	1.041 (0.751–1.442)	0.811	70 (44.6)	1.225 (0.825–1.820)	0.315
AA	33 (6.3)	23 (9.2)	1.498 (0.834–2.690)	0.176	7 (4.5)	0.898 (0.356–2.262)	0.819
AG+AA vs. GG			1.105 (0.811–1.507)	0.525		1.186 (0.807–1.742)	0.386
AA vs. AG+GG			1.473 (0.834–2.603)	0.182		0.821 (0.332–2.033)	0.670
G	782	357			230
A	270	143			84
Rs4081134
GG (ref)	314 (59.7)	165 (66.0)			106 (67.5)
AG	182 (34.6)	80 (32.0)	0.845 (0.606–1.178)	0.320	46 (29.3)	0.840 (0.552–1.278)	0.415
AA	30 (5.7)	5 (2.0)	0.325 (0.121–0.869)	0.025	5 (3.2)	0.499 (0.179–1.395)	0.185
AG+AA vs. GG			0.773 (0.560–1.066)	0.117		0.788 (0.526–1.181)	0.248
AA vs. AG+GG			0.344 (0.129–0.915)	0.033		0.528 (0.191–1.465)	0.220
G	810	410	1.00 (ref)		258	1.00 (ref)
A	242	90	0.735 (0.561–0.962)	0.025	56	0.727 (0.526–1.003)	0.052

Adjusted for age, gender, and smoking status.

Table 4.

Associations Between the Two Single-Nucleotide Polymorphisms and Lung Cancer Risk, Stratified by Smoking Exposure

SNPs	Smoking	Genotype	Control (%)	Case (%)	OR ^a (95% CI)	p^a
Rs7158663	Exposure	GG (ref)	85 (55.2)	120 (51.7)	1.00 (ref)
		AG	64 (41.6)	96 (41.4)	0.987 (0.635–1.534)	0.955
		AA	5 (3.2)	16 (6.9)	1.878 (0.638–5.533)	0.253
		AG+AA vs. GG			1.054 (0.686–1.617)	0.812
		AA vs. AG+GG			1.889 (0.653–5.465)	0.240
	Nonexposure	GG (ref)	204 (54.8)	148 (50.3)	1.00 (ref)
		AG	140 (37.6)	123 (41.8)	1.227 (0.882–1.706)	0.225
		AA	28 (7.5)	23 (7.8)	1.155 (0.632–2.113)	0.640
		AG+AA vs. GG			1.215 (0.887–1.663)	0.225
		AA vs. AG+GG			1.058 (0.588–1.902)	0.851
Rs4081134	Exposure	GG (ref)	97 (63.0)	151 (65.1)	1.00 (ref)
		AG	46 (29.9)	73 (31.5)	1.169 (0.729–1.872)	0.517
		AA	11 (7.1)	8 (3.4)	0.600 (0.213–1.690)	0.333
		AG+AA vs. GG			1.073 (0.685–1.680)	0.760
		AA vs. AG+GG			0.569 (0.205–1.582)	0.280
	Nonexposure	GG (ref)	217 (58.3)	198 (67.3)	1.00 (ref)
		AG	136 (36.6)	88 (29.9)	0.708 (0.505–0.994)	0.046
		AA	19 (5.1)	8 (2.7)	0.468 (0.196–1.114)	0.086
		AG+AA vs. GG			0.679 (0.490–0.942)	0.020
		AA vs. AG+GG			0.527 (0.223–1.245)	0.144

Adjusted for age and gender.

Table 5.

Associations Between the Two Single-Nucleotide Polymorphisms and Lung Cancer Risk, Stratified by Average Age

SNPs	Age	Genotype	Control (%)	Case (%)	OR ^a (95% CI)	p^a
Rs7158663	≤57	GG (ref)	148 (55.4)	104 (53.9)	1.00 (ref)
		AG	101 (37.8)	77 (39.9)	1.108 (0.740–1.658)	0.618
		AA	18 (6.7)	12 (6.2)	1.045 (0.465–2.347)	0.915
		AG+AA vs. GG			1.099 (0.746–1.619)	0.632
		AA vs. AG+GG			1.001 (0.454–2.206)	0.999
	>57	GG (ref)	141 (54.4)	164 (49.2)	1.00 (ref)
		AG	103 (39.8)	142 (42.6)	1.155 (0.818–1.630)	0.414
		AA	15 (5.8)	27 (8.1)	1.516 (0.767–2.996)	0.232
		AG+AA vs. GG			1.200 (0.862–1.672)	0.280
		AA vs. AG+GG			1.422 (0.732–2.764)	0.299
Rs4081134	≤57	GG (ref)	162 (60.7)	125 (64.8)	1.00 (ref)
		AG	87 (32.6)	61 (31.6)	0.870 (0.573–1.322)	0.514
		AA	18 (6.7)	7 (3.4)	0.449 (0.177–1.143)	0.093
		AG+AA vs. GG			0.796 (0.534–1.186)	0.261
		AA vs. AG+GG			0.471 (0.187–1.186)	0.110
	>57	GG (ref)	152 (58.7)	224 (67.3)	1.00 (ref)
		AG	95 (36.7)	100 (30.0)	0.736 (0.516–1.050)	0.091
		AA	12 (4.6)	9 (2.7)	0.515 (0.209–1.272)	0.150
		AG+AA vs. GG			0.711 (0.504–1.002)	0.052
		AA vs. AG+GG			0.573 (0.234–1.403)	0.223

Adjusted for gender and smoking status.

We then assessed the combined effect of the risk alleles (rs4081134-G and rs7158663-A) on the risk of lung cancer. With the subjects carrying “0” risk alleles as a reference, subjects with “2” risk alleles and subjects carrying “3–4” risk alleles had an increased risk of lung cancer and non-small cell lung cancer. The trend test showed that the risk of lung cancer and non-small cell lung cancer gradually increased with the increase of the number of risk alleles in study subjects, and the differences were statistically significant. The data are summarized in Table 6.

Table 6.

Cumulative Effects of rs4081134-G and rs7158663-A on Lung Cancer and Non-small Cell Lung Cancer Susceptibility

		Lung cancer					Non-small cell lung cancer
Variant	Control (%) (n = 526)	Case (%) (n = 526)	OR (95% CI)	p	OR ^a (95% CI)	p^a	Case (%) (n = 424)	OR (95% CI)	p	OR ^a (95% CI)	p^a
0 (ref)	30	15	1.00 (ref)		1.00 (ref)		10	1.00 (ref)		1.00 (ref)
1	115	102	1.774 (0.903–3.483)	0.096	1.757 (0.873–3.533)	0.144	85	2.217 (1.028–4.782)	0.042	2.172 (0.983–4.799)	0.055
2	211	212	2.009 (1.051–3.843)	0.035	2.051 (1.047–4.018)	0.036	172	2.445 (1.163–5.142)	0.018	2.478 (1.151–5.334)	0.020
3–4	170	197	2.318 (1.206–4.452)	0.012	2.258 (1.147–4.442)	0.018	157	2.771 (1.312–5.853)	0.008	2.644 (1.223–5.717)	0.013
Trend			P ^b = 0.011					P ^b = 0.017
0	30	15	1.00 (ref)		1.00 (ref)		10	1.00 (ref)		1.00 (ref)
1–4	496	511	2.060 (1.095–3.876)	0.025	2.054 (1.066–3.955)	0.031	414	2.504 (1.210–5.183)	0.013	2.465 (1.164–5.219)	0.018

Adjusted for age, gender, and smoking status.

p-Value of linear-by-linear association.

In addition, we evaluated the interaction between the rs4081134 genotype and smoking exposure. There was no significant interaction between rs4081134 and smoking on lung cancer and non-small cell lung cancer risk. Tables 7 and 8 showed the results of crossover analysis.

Table 7.

Interaction Between the rs4081134 Polymorphisms and Smoking Exposure on Lung Cancer and Non-small Cell Lung Cancer Risk

			Lung cancer			Non-small cell lung cancer
Genotype	Smoking	Controls (%)	Case (%)	OR ^a (95% CI)	p^a	Case (%)	OR ^a (95% CI)	p^a
Rs4081134
AA	Nonexposure	19 (3.6)	8 (1.5)	1 (ref)		5 (1.2)	1 (ref)
AG+GG	Nonexposure	353 (67.1)	286 (54.4)	1.889 (0.797–4.478)	0.149	242 (57.1)	2.619 (0.941–7.290)	0.065
AA	Exposure	11 (2.1)	8 (1.5)	2.537 (0.707–9.108)	0.153	6 (1.4)	3.425 (0.801–14.642)	0.097
AG+GG	Exposure	143 (27.2)	224 (42.6)	4.953 (2.013–12.188)	0.000	171 (40.3)	6.347 (2.196–18.344)	0.001

Adjusted for age, gender, and smoking status.

Table 8.

Crossover Analysis of Interaction Between rs4081134 Polymorphisms and Smoking on Lung Cancer Susceptibility

		Lung cancer		Non-small cell lung cancer
SNP	Measure	Estimate	95% CI	Estimate	95% CI
Rs4081134	RERI	1.069	−0.714 to 2.851	0.866	−1.470 to 3.205
	AP	0.287	−0.220 to 0.794	0.191	−0.352 to 0.733
	S	1.647	0.482–5.624	1.323	0.510–3.434

AP, attributable proportion due to interaction; RERI, relative excess risk due to interaction; S, synergy index.

Discussion

It is estimated that about 1.6 million people worldwide died of lung cancer in 2012, and lung cancer has become a major public health problem that threatens human health (Torre et al., 2015). We performed a hospital-based case-control study to assess the association between LncRNA MEG3 polymorphisms and lung cancer susceptibility in 526 cases and 526 controls. Our results suggested that the rs4081134-AA genotype was associated with a decreased risk of lung cancer and non-small cell lung cancer. In stratification by smoking status, the rs4081134-AG genotype and dominant model (AG+AA) in the nonsmoking group could reduce the risk of lung cancer, compared with GG. To date, this is the first study to evaluate the correlation between MEG3 rs7158663 and rs4081134 polymorphisms and the susceptibility of lung cancer in northeast China.

To evaluate the positive findings, we calculated the statistical power and false-positive report probability (FPRP) (Wacholder et al., 2004; He et al., 2013). The statistical power result and FPRP was 0.998 and 0.162 for the lung cancer group, 0.999 and 0.118 for the non-small cell lung cancer group. To control age-related confounding, we only observed borderline significance in the rs4081134 dominant model, which was stratified by average age. The possible reason may be the small sample size and the insufficient power of the study after stratification (power = 0.503, FPRP = 0.479). It also suggests that larger sample sizes are required to conduct further exploration.

With the exploration of lncRNAs function, massive studies have demonstrated that lncRNAs may have effect on the development, progression, and susceptibility of lung cancer. For example, Li et al. (2016) revealed the association between LncRNA ZNRD1-AS1 polymorphisms and lung cancer susceptibility. Moreover, compared with normal tissues, ZNRD1-AS1 expression was increased in lung cancer tissues. A case-control study conducted by Gong et al. (2016) found that the polymorphisms of HOTTIP, CCAT2, and H19 were significantly associated with the risk of lung cancer. To explore the role of MEG3 in the development and progression of cancer, researchers in related fields have made great efforts in recent years. It is believed that the low expression of MEG3 plays an important role in tumor suppressor, which is closely related to tumor occurrence, development, metastasis, and prognosis (Qin et al., 2013; Ying et al., 2013; Sun et al., 2014). Through in vivo and in vitro tests, Zheng et al. (2018) demonstrated that MEG3 affects the risk of liver cancer by modulating the activity of the PKM2 and β-catenin pathways. Further, the expression of MEG3 was significantly lower in gastric cancer than in adjacent normal tissues, and meanwhile it was also significantly correlated with TNM staging, depth of invasion, and tumor size (Sun et al., 2014).

It has been reported that the decrease of MEG3 expression in non-small cell lung cancer (NSCLC) may be affected by DNA methylation, which partly regulates the proliferation and apoptosis of NSCLC cells through the activation of p53 (Lu et al., 2013). It was also found that the expression of MEG3 in NSCLC was at a lower level than that in normal tissues, which also correlated with advanced pathological stages and tumor size. Besides, MEG3 affects JARID2 and EZH2 histone H3 methylation, which plays a key role in epigenetic regulation of epithelial interstitial transformation in lung cancer cell lines (Terashima et al., 2017). MEG3 was also reported to regulate the translation of Skp2 mRNA in NSCLC cells and to inhibit its growth (Su et al., 2016).

It has been shown that several polymorphisms can regulate gene expression by causing variations in transcription factor binding sites. For instance, gene mutation can lead to the disappearance of an existing transcription factor binding site or the production of a new transcription factor binding site (Yee et al., 2009; Su et al., 2015). Based on SNP function prediction, we found that both MEG3 rs7158663 and rs4081134 are located in transcription factor binding sites (Xu and Taylor, 2009) (https://snpinfo.niehs.nih.gov/cgibin/snpinfo/tfbs.cgi?2_rs7158663) (https://snpinfo.niehs.nih.gov/cgibin/snpinfo/tfbs.cgi?2_rs4081134). Therefore, we speculate that the polymorphisms of rs7158663 and rs4081134 may change the gene expression of LncRNA MEG3 and affect the risk of lung cancer. However, according to the bioinformatics method used by Huang et al. (2014), we further analyzed these possible transcription factor binding sites. Regrettably, we failed to observe the transcription factor binding sites involving lung cancer development. By using the RNAfold web server, we predicted that rs4081134 polymorphisms might alter the centroid secondary structure and minimal free energy (MFE), and also change the folding and function of MEG3. The MFE of the MEG3 centroid secondary structure containing the rs4081134 G allele and the A allele were −62.90 and −54.98 kcal/mol, respectively. However, our speculation requires further confirmation.

A hospital-based case-control study of colorectal cancer risk in a Chinese population found that individuals with the MEG3 rs7158663 AA genotype were at higher risk than those with the GG genotype (Cao et al., 2016). In addition, studies have shown that MEG3 rs116907618 was not associated with the susceptibility of lung cancer, but a significant association could be observed in lung cancer patients with platinum-based chemotherapy toxicity (Gong et al., 2016, 2017). However, another case-control study by Zhou et al. (2018) demonstrated that subjects carrying the rs4081134 AG/AA genotype have a significantly increased risk of developing neuroblastoma among children with age >18 months. These previous studies and our studies suggest that MEG3 polymorphisms may have an effect on susceptibility, drug resistance, and prognosis of malignant tumors. However, the exact mechanisms of MEG3 polymorphisms in lung cancer remain undetermined, and further studies will be needed in the future.

In this study, we have a series of strict inclusion and exclusion criteria in selecting newly diagnosed patients, so it can effectively avoid Neyman bias. In the statistical analysis, age, sex, and smoking status were used as adjustment factors for unconditional logistic regression analysis to reduce confounding bias. Meanwhile, there are still some limitations that should not be ignored. First, although we selected cases and controls from multiple hospitals, it is possible to have Berkson bias in this study. Besides, individuals in the control group were selected from the medical examination center of the hospital, and its representativeness may not be very well.

In conclusion, we indicate that MEG3 rs4081134 polymorphisms are associated with decreased risk of lung cancer in northeast China. Our findings suggest that MEG3 polymorphisms might serve as potential predictive markers of lung cancer. In the future, more large-scale studies and functional experiments on lung cancer in other ethnic groups are needed to confirm our findings.

Conclusion

MEG3 polymorphisms might be associated with the decreased risk of lung cancer in northeast China.

Footnotes

Acknowledgments

The authors would like to thank all patients and hospital investigators for their participation. This study was funded by the National Natural Science Foundation of China (No. 81673261) and the Science Foundation of School of Public Health, China Medical University.

Disclosure Statement

No competing financial interests exist.

References

Anwar

S.L.

, Krech

, Hasemeier

, Schipper

, Schweitzer

, Vogel

, et al. (2012). Loss of imprinting and allelic switching at the DLK1-MEG3 locus in human hepatocellular carcinoma. PLoS One, 7, e49462.

Balik

, Srovnal

, Sulla

, Kalita

, Foltanova

, Vaverka

, et al. (2013). MEG3: a novel long noncoding potentially tumour-suppressing RNA in meningiomas. J Neurooncol, 112, 1–8.

Benetatos

, Vartholomatos

, and Hatzimichael

(2011). MEG3 imprinted gene contribution in tumorigenesis. Int J Cancer, 129, 773–779.

Cao

, Zhuang

, Hu

, Xi

, Deng

, Sheng

, et al. (2016). Associations between polymorphisms of long non-coding RNA MEG3 and risk of colorectal cancer in Chinese. Oncotarget, 7, 19054–19059.

Chen

, Zheng

, Baade

P.D.

, Zhang

, Zeng

, Bray

, et al. (2016). Cancer statistics in China, 2015. CA Cancer J Clin, 66, 115–132.

Ferlay

, Shin

H.R.

, Bray

, Forman

, Mathers

, and Parkin

D.M.

(2010). Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer, 127, 2893–2917.

Gao

, Li

, Lv

, Zhou

, and Yin

(2018). Association between four polymorphisms in lncRNA and risk of lung cancer in a Chinese never-smoking female population. DNA Cell Biol. 37, 651–658.

Gao

, and Wei

G.H.

(2017). Genomic insight into the role of lncRNA in cancer susceptibility. Int J Mol Sci, 18, 1239.

Gong

W.J.

, Peng

J.B.

, Yin

J.Y.

, Li

X.P.

, Zheng

, Xiao

, et al. (2017). Association between well-characterized lung cancer lncRNA polymorphisms and platinum-based chemotherapy toxicity in Chinese patients with lung cancer. Acta pharmacol Sin, 38, 581–590.

10.

Gong

W.J.

, Yin

J.Y.

, Li

X.P.

, Fang

, Xiao

, Zhang

, et al. (2016). Association of well-characterized lung cancer lncRNA polymorphisms with lung cancer susceptibility and platinum-based chemotherapy response. Tumour Biol, 37, 8349–8358.

11.

, Wang

M.Y.

, Qiu

L.X.

, Zhu

M.L.

, Shi

T.Y.

, Zhou

X.Y.

, et al. (2013). Genetic variations of mTORC1 genes and risk of gastric cancer in an Eastern Chinese population. Mol Carcinog, 52 Suppl 1, E70–E79.

12.

Huang

, Whitington

, Gao

, Lindberg

J.F.

, Yang

, Sun

, et al. (2014). A prostate cancer susceptibility allele at 6q22 increases RFX6 expression by modulating HOXB13 chromatin binding. Nat Genet, 46, 126–135.

13.

Huang

, Liu

, Wang

J.P.

, Wang

Y.Q.

, Yu

X.L.

, Wang

Z.B.

, et al. (2012). Regulatory long non-coding RNA and its functions. J Physiol Biochem, 68, 611–618.

14.

International Agency for Research on Cancer. (2017). Cancer Incidence in Five Continents. Available at http://ci5.iarc.fr IARC (Lyon). (Last accessed January 5, 2018 .)

15.

Khalil

A.M.

, Guttman

, Huarte

, Garber

, Raj

, Rivea Morales

, et al. (2009). Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression. Proc Natl Acad Sci U S A, 106, 11667–11672.

16.

Kopp

, and Mendell

J.T.

(2018). Functional classification and experimental dissection of long noncoding RNAs. Cell, 172, 393–407.

17.

Kruer

T.L.

, Dougherty

S.M.

, Reynolds

, Long

, de Silva

, Lockwood

W.W.

, et al. (2016). Expression of the lncRNA maternally expressed gene 3 (MEG3) contributes to the control of lung cancer cell proliferation by the Rb pathway. PLoS One, 11, e0166363.

18.

, Song

, Wen

, Li

, Jie

, Wang

, et al. (2016). Strong evidence for LncRNA ZNRD1-AS1, and its functional Cis- eQTL locus contributing more to the susceptibility of lung cancer. Oncotarget, 7, 35813–35817.

19.

Liu

, Wan

, Lu

, Sun

, Pan

, Zhang

, et al. (2015). The Long Noncoding RNA MEG3 contributes to cisplatin resistance of human lung adenocarcinoma. PLoS One, 10, e0114586.

20.

Loewer

, Cabili

M.N.

, Guttman

, Loh

Y.H.

, Thomas

, Park

I.H.

, et al. (2010). Large intergenic non-coding RNA-RoR modulates reprogramming of human induced pluripotent stem cells. Nat Genet, 42, 1113–1117.

21.

Louro

, Smirnova

A.S.

, and Verjovski-Almeida

(2009). Long intronic noncoding RNA transcription: expression noise or expression choice?. Genomics, 93, 291–298.

22.

K.H.

, Li

, Liu

X.H.

, Sun

, Zhang

M.L.

, Wu

W.Q.

, et al. (2013). Long non-coding RNA MEG3 inhibits NSCLC cells proliferation and induces apoptosis by affecting p53 expression. BMC Cancer, 13, 461.

23.

, Cui

, Li

, Yang

, Bi

, et al. (2018). Polymorphism in lncRNA AC008392.1 and its interaction with smoking on the risk of lung cancer in a Chinese population. Cancer Manag Res, 10, 1377–1387.

24.

Mercer

T.R.

, Dinger

M.E.

, and Mattick

J.S.

(2009). Long non-coding RNAs: insights into functions. Nat Rev Genet, 10, 155–159.

25.

Ponting

C.P.

, Oliver

P.L.

, and Reik

(2009). Evolution and functions of long noncoding RNAs. Cell, 136, 629–641.

26.

Qin

, Chen

, Ding

, Hao

, Hu

, and Guo

(2013). Long non-coding RNA MEG3 inhibits the proliferation of cervical carcinoma cells through the induction of cell cycle arrest and apoptosis. Neoplasma, 60, 486–492.

27.

Samet

J.M.

(1991). Health benefits of smoking cessation. Clin Chest Med, 12, 669–679.

28.

Spitale

R.C.

, Tsai

M.C.

, and Chang

H.Y.

(2011). RNA templating the epigenome: long noncoding RNAs as molecular scaffolds. Epigenetics, 6, 539–543.

29.

, Su

, Shang

, Wan

, Chen

, and Rao

(2015). SNP detection of TLR8 gene, association study with susceptibility/resistance to GCRV and regulation on mRNA expression in grass carp, Ctenopharyngodon idella . Fish Shellfish Immunol, 43, 1–12.

30.

, Han

, Wu

, and Huo

(2016). Skp2 regulates non-small cell lung cancer cell growth by MEG3 and miR-3163. Tumour Biol, 37, 3925–3931.

31.

Sun

, Xia

, Jin

, Xu

, Liu

, De

, et al. (2014). Downregulated long noncoding RNA MEG3 is associated with poor prognosis and promotes cell proliferation in gastric cancer. Tumour Biol, 35, 1065–1073.

32.

Terashima

, Tange

, Ishimura

, and Suzuki

(2017). MEG3 long noncoding RNA contributes to the epigenetic regulation of epithelial-mesenchymal transition in lung cancer cell lines. J Biol Chem, 292, 82–99.

33.

Torre

L.A.

, Bray

, Siegel

R.L.

, Ferlay

, Lortet-Tieulent

, and Jemal

(2015). Global cancer statistics, 2012. CA Cancer J clin, 65, 87–108.

34.

Wacholder

, Chanock

, Garcia-Closas

, El Ghormli

, and Rothman

(2004). Assessing the probability that a positive report is false: an approach for molecular epidemiology studies. J Natl Cancer Inst, 96, 434–442.

35.

Wang

, Ren

, and Sun

(2012). Overexpression of the long non-coding RNA MEG3 impairs in vitro glioma cell proliferation. J Cell Biochem, 113, 1868–1874.

36.

Wierzbicki

A.T.

(2012). The role of long non-coding RNA in transcriptional gene silencing. Curr Opin Plant Biol, 15, 517–522.

37.

Wilusz

J.E.

, Sunwoo

, and Spector

D.L.

(2009). Long noncoding RNAs: functional surprises from the RNA world. Genes Dev, 23, 1494–1504.

38.

Xia

, He

, Liu

, Wang

, and Chen

(2015). Downregulation of MEG3 enhances cisplatin resistance of lung cancer cells through activation of the WNT/beta-catenin signaling pathway. Mol Med Rep, 12, 4530–4537.

39.

, and Taylor

J.A.

(2009). SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies. Nucleic Acids Res, 37, W600–W605.

40.

Yan-Hua

, Xiang-Lei

, Hong

, Jian-Jun

(2015). Long noncoding ribonucleic acids maternally expressed gene 3 inhibits lung cancer tumor progression through downregulation of MYC. Indian J Cancer, 52 Suppl 3, E190–E193.

41.

Yee

S.W.

, Shima

J.E.

, Hesselson

, Nguyen

, De Val

, Lafond

R.J.

, et al. (2009). Identification and characterization of proximal promoter polymorphisms in the human concentrative nucleoside transporter 2 (SLC28A2). J Pharmacol Exp Ther, 328, 699–707.

42.

Yin

J.Y.

, Huang

, Yang

, Zhang

J.T.

, Zhong

M.Z.

, Zhou

H.H.

, et al. (2009). Characterization and analyses of multidrug resistance-associated protein 1 (MRP1/ABCC1) polymorphisms in Chinese population. Pharmacogenet Genomics, 19, 206–216.

43.

Yin

, Cui

, Ren

, Xia

, Li

, and Zhou

(2017). MiR-196a2 and lung cancer in Chinese non-smoking females: a genetic association study and expression analysis. Oncotarget, 8, 70890–70898.

44.

Yin

, Cui

, Ren

, Xia

, Wang

, Zhang

, et al. (2016). Association between polymorphisms in pre-miRNA genes and risk of lung cancer in a Chinese non-smoking female population. Lung Cancer, 94, 15–21.

45.

Ying

, Huang

, Chen

, Wang

, Xia

, Chen

, et al. (2013). Downregulated MEG3 activates autophagy and increases cell proliferation in bladder cancer. Mol BioSyst, 9, 407–411.

46.

Zheng

, Lin

, Xu

, Lu

, Meng

, Wang

, et al. (2018). Long noncoding RNA MEG3 suppresses liver cancer cells growth through inhibiting beta-catenin by activating PKM2 and inactivating PTEN. Cell Death Dis, 9, 253.

47.

Zhou

, Zhang

, and Klibanski

(2012). MEG3 noncoding RNA: a tumor suppressor. Journal of Mol Endocrinol, 48, R45–R53.

48.

Zhuo

Z.J.

, Zhang

, Zhu

, Yang

, Zou

, et al. (2018). Associations between lncRNA MEG3 polymorphisms and neuroblastoma risk in Chinese children. Aging, 10, 481–491.