This article is mentioned in:

Introduction

Circadian rhythms are daily oscillations in physiological processes driven by endogenous clocks, that exist in the brain and peripheral tissues and which are synchronized to external environmental cycles such as day and night (1). They regulate numerous biological functions in the human body, including sleep and wakefulness, body temperature, blood pressure, hormone production, digestive secretion and immune activity (2).

The molecular mechanisms of 24 h timekeeping and circadian rhythm generation in the central clock of the body are based on interactive positive and negative transcriptional-translational feedback loops generated by circadian clock genes (3). Up to 10% of the genome of cells is expressed, at the RNA level, in a circadian-coordinated manner, with each tissue exhibiting unique tissue-specific expression profiles (4). The circadian clock controls cellular processes, including proliferation, apoptosis, DNA repair, metabolism, detoxification and the DNA damage response (5). Alterations in clock genes may disrupt this rhythmic control, resulting in abnormalities in cell proliferation, apoptosis, DNA damage response and metabolism, which may subsequently contribute to coronary heart attacks, depression and tumor promotion (4,6). Therefore, circadian clock regulation is a critical component of disease pathogenesis.

Cancer may be a circadian associated disorder, as the extensive evidence gathered in mice and human studies, has suggested. Earlier investigations have revealed that the cycle of cell division is under circadian control, that the rate of tumor growth demonstrates a daily circadian rhythm and shift work or jet lag may be contributing factors to the increased incidence of cancer and mortality rate in humans (7,8).

At the molecular level, the circadian clock is composed of the products of at least eight core genes (Clock, Ck1ɛ, Cry1, Cry2, Per1, Per2, Per3 and Bmal1), which are organized in a transcriptional-translational regulatory network (9,10). Molecular-genetic analysis studies in vertebrates have provided evidence that consistent alterations in gene expression critically controls the expression patterns of further output genes (clock-controlled genes), such as Nocturnin (CCRN4L/NOC), albumin D-box binding protein (DBP), RORα and REV-ERBα (11,12).

Lung cancer is the leading cause of cancer-related mortality worldwide due to its high incidence, malignant behavior and an lack of effective treatments that are able to improve disease prognosis and survival rate (13). It is, therefore, a necessity to unravel and better understand the molecular mechanisms associated with cancer development and progression (14).

In the present study, we analyzed polymorphisms in CLOCK, period 3 (PER3) and CCRN4L genes and investigated, in a case-control study, their association with non-small-cell-lung cancer (NSCLC) in Brazilian patients.

Materials and methods

Subjects

The study’s cohort consisted of 78 patients with primary diagnosis of NSCLC, all eligible for surgery and with no previous history of chemotherapy or radiotherapy. The age range was 25–82 years (mean, 60.1±12.6; median, 62 years). Patients were recruited from a reference center of thoracic surgery from the Hospital Julia Kubitscheck (Minas Gerais, Brazil). Local Ethics Committee approval was obtained and all participants signed a written informed consent form. The study protocol was also approved by the Ethics Committee of the Universidade Federal de Minas Gerais (Minas Gerais, Brazil; ETIC 473-05). As a control group for CCRN4L, CLOCK and PER3 genotyping, we studied 74 healthy individuals with no cancer history and all >55 years old.

Polymorphism genotyping

Genomic DNA was isolated from lung tissue samples according to a proteinase K-based protocol (15). Peripheral blood samples from the control group were collected in vacuum tubes and genomic DNA was isolated using the high salt method (16). HapMap database (www.hapmap.org) as well as previous published literature were used for selecting single-nucleotide polymorphisms (SNPs) covering CCRN4L, CLOCK and PER3 genes; 20 ng of DNA of each patient was used for TaqMan SNP genotyping assays (rs938836, rs17050680, rs3805213, rs6855837, rs228727, rs228644, rs228729, rs707467, rs10462020) according to the manufacturer’s instructions (Applied Biosystems, Foster City, CA, USA; Table I). Genotyping was performed by Real-Time PCR (RT-PCR) using allelic discrimination in the 7500 RT-PCR System (Applied Biosystems). PCR parameters involved an initial denaturation at 95°C for 10 min followed by 50 cycles at 95°C for 15 sec and 60°C for 1 min. Each reaction contained 5.0 μl of mix, 0.1 μl of probe, 3.9 μl of deionized water and 60 ng of DNA. We retyped at least 10% of the samples for quality control.

Table 1 Position of all SNPs used in this study.

Table 1

Position of all SNPs used in this study.

SNP	Chromosome	Chromosome position	Gene	Gene location	Ancestral allele
rs938836a	4	139939653	CCRN4L	Intron 1–2	G
rs17050680a	4	139946648	CCRN4L	Intron 1–2	A
rs3805213a	4	139965724	CCRN4L	Intron 2–3	G
rs6855837	4	56319244	CLOCK	Exon 14	G
rs228727	1	7847836	PER3	Intron 3–4	A
rs228644	1	7866083	PER3	Intron 9–10	G
rs228729a	1	7845695	PER3	Intron 2–3	G
rs707467	1	7861684	PER3	Intron 7–8	T
rs10462020	1	7880683	PER3	Exon 15	T

a tag SNP.

{ label (or @symbol) needed for fn[@id='tfn2-mmr-10-01-0435'] } SNP, single nucleotide polymorphism; PER3, period 3.

Statistical analysis

Allele and genotype frequencies were compared between case and control groups with the χ2 test using the UNPHASED software program (v.3.0.13, https://sites.google.com/site/fdudbridge/software/unphased-3-1). HAPLOVIEW 4.1 software (Broad Institute, Cambridge, MA, USA) was used to evaluate pairwise linkage disequilibrium (LD) matrices between each tag SNP to examine the LD block structure and Hardy-Weinberg equilibrium (HWE; significance cutoff, 0.05). FAMHAP software (famhap.19c, http://famhap.meb.uni-bonn.de/; Germany) was used for haplotype association analysis. The odds ratios (ORs) and 1,000 permutations were performed using UNPHASED. We performed 1,000 permutations in each test to estimate the global significance of the results and to validate the expectation-maximization values. P<0.05 was considered to indicate a statistically significant result.

Results

Patient characteristics (Table II)

There were 30 female and 48 male participants with a median age of 60.1 years. Of the 78 tumors analyzed in this study, 25 (32%) were diagnosed as squamous cell carcinoma (SCC), 39 (50%) as adenocarcinoma (ADC) and 14 (18%) as other subtypes of NSCLC. Sixty-two patients were former or current-smokers and sixteen were never-smokers. The control group had 74 individuals, 40 female and 34 male with a median age of 75.4±7.96 years (range, 65–96 years).

Allele and genotype frequency of all SNPs (Table III)

All samples were in HWE for eight SNPs, except for rs938836 (CCRN4L) and rs6855837 (CLOCK), 2.096×10−13 and 0.007, respectively. One CCRN4L marker (tag SNP rs3805213) showed a significant association with NSCLC to allele and genotype frequency (P=0.005 and P=0.004, respectively). For these tag SNPs, the T/C heterozygosity showed a significant association with increased risk for logarithm of the odds (LOD; 47.5%) correlated to comparison group (24.3%; P=0.004, OR =1.5 (0.45–4.94), χ2=10.9, df=2; Table III). Allelic analysis of tag SNP rs3805213 demonstrated a risk higher between allele T and LOD (33.7%) related to comparison group (19.8%; P=0.005, χ2=7.81, df=1). Following the permutation tests, the values remained significant (Table III).

Another tag SNP tested from the PER3 gene (rs228729) also showed a significant association with the disease to allele and genotype frequency (P=0.004 and P=0.02, respectively; Table III). In the control group, analyses of rs228729 in T/T homozygous individuals showed a higher frequency (21%) compared with patients (11%; P=0.02, OR=1, χ2=7.77, df=2; Table III). Allelic analysis for this PER SNP showed a higher risk between allele T and LOD (45%) related to the comparison group (29.9%; P=0.004, χ2=8.06, df=1). Following the permutation tests, the values remained significant (Table III). No association was detected for the other markers of PER3, CLOCK and CCRN4L (rs938836, rs17050680, rs6855837, rs228727, rs228644, rs707467 and rs10462020).

Haplotype analysis

The analysis revealed a strong association with rs228729 (PER3) allele T and the increased risk to LOD. The T-T-A-G-T haplotype (rs228729, rs228727, rs707467, rs228644, rs10462020, respectively) occurred more in patients (62%) than in the comparison group (3%; P=0.015; Table IV). These results reinforce the presence of the allele T (rs228729) increased risk for LOD. To avoid genotyping errors (quality control) at least 10% of the samples were retyped.

Table IV PER3 haplotype frequency in lung cancer patients and controls.

Table IV

PER3 haplotype frequency in lung cancer patients and controls.

Haplotype ID	rs228729	rs228727	rs707467	rs228644	rs10462020	Case freq	Control freq	Freq	P-valuea
A	C	C	A	A	T	0.075	0.092	0.083	-
B	C	C	A	G	G	0.073	0.156	0.114	-
C	C	C	A	G	T	0.022	0.128	0.074	-
D	C	T	A	G	T	0.168	0.142	0.151	-
E	C	T	C	G	T	0.189	0.173	0.181	-
F	T	C	A	A	T	0.380	0.268	0.328	-
G	T	T	A	G	T	0.062	0.030	0.046	0.015 (0.020)a

a Adjusted P value.

{ label (or @symbol) needed for fn[@id='tfn7-mmr-10-01-0435'] } Freq, frequency; PER3, period 3.

Pairwise LD (Fig. 1)

The analysis of CCRN4L demonstrated a significant LD between rs17050680 and rs3805213, and also among rs938836 and rs3805213 (D′=1.0). Similarly, there was a moderate LD in CCRN4L rs9302648 and rs17050680 polymorphisms (D′=0.61). The Fig. 1B shows one LD block comprising only of three SNPs of the PER3 gene tested (rs228727, rs707467 and rs228644; D′=1.0). The rs228729 and rs10462020 were not included in the block.

Discussion

The present study was based on case-control association evaluating the clock-controlled gene CCRN4L and clock genes, including CLOCK and PER3 with NSCLC. We analyzed nine SNPs (rs938836, rs17050680, rs3805213, rs6855837, rs228727, rs228644, rs228729, rs707467, rs10462020) in these genes in 78 Brazilian patients with NSCLC.

PER3 is a circadian regulation protein that affects the cell cycle, growth and differentiation. Previous studies have also demonstrated an association between the PER3 gene and tumorigenesis. Wang et al determined the association between PER3 expression and colon cancer incidence/progression, as compared with normal tissue. As a 2.8-fold decrease in PER3 mRNA levels in colon cancerous tissue was observed (17). Climent et al (18) demonstrated that mice deficient in PER3 had increased susceptibility to breast cancer induced by carcinogen treatment or by overexpression of Erbb2. Zienolddiny et al (19) revealed that SNPs in the PER3 gene were associated with decreased breast cancer risk and an association with prostate and colorectal cancer risk has also been identified (20,21). In our analysis, we identified a statistically significant association for SNPs rs228729 [P=0.02, OR=0.51 (0.32–0.81), χ2=7.8, df=2] in the PER3 gene with NSCLC suggesting that PER3 may be correlated with the development of lung cancer. Furthermore, one LD block comprising of three SNPs of the PER3 gene (rs228727, rs707467 and rs228644) was observed and rs228729 demonstrated a significant association with the disease, however was not located in this block.

Zieker et al (22) conducted a comparative microarray analysis to investigate the relative mRNA expression of clock-controlled genes (including Noc, hPER2, hCRY1, SMAD5) throughout a 24 h period in cell samples obtained from oral mucosa. The detected circadian expression profile of CCRN4L constituted a protective feature during daytime upregulation of the gene. The authors also suggested that this gene may be involved in specific tumor suppressor functions in human oral mucosa. In the present study, we identified an association between SNP rs3805213 in CCRN4L and NSCLC [P=0.004, OR=1.5 (0.45–4.94), χ2=10.9, df=2], suggesting a possible correlation between this protein and NSCLC. The rs3805213 of CCRN4L gene did not form LD block with others SNPs despite demonstrating an association with disease.

Our results did not reveal a correlation between the rs6855837 in CLOCK gene and NSCLC. Huang et al (23) suggested that CLOCK gene expression seems to be altered in certain human cancer types compared with healthy corresponding tissues, possibly as a result of other events, such as promoter methylation. Therefore, further studies are required to evaluate the involvement of CLOCK genes in the tumorigenesis of NSCLC.

Our study provides preliminary evidence that polymorphisms in the CCRN4L and PER3 genes may represent a risk factor in the occurrence of NSCLC in Brazilian patients. Further large scale genetic studies are needed to confirm these observations, since the sample size in the present study was relatively small. Furthermore, functional studies may clarify the role of other variants and haplotypes in patient susceptibility to NSCLC.

Acknowledgements

This study was supported by grants from CNPq, FAPEMIG and CAPES (Brazil).

References