Introduction

Cyclin D1 (CCND1; also named BCL1 alternatively, MIM 168461), located on chromosome 11q13, plays a crucial role in the transition from G1 to S phase of the cell cycle as a key cell-cycle regulator (Sherr 1995), while B-cell CLL\lymphoma 2 (BCL2, MIM 151430), located on chromosome 18q21.3, protects cells from apoptosis by binding to BCL2-associated X (BAX) protein (Leiter et al. 2000). CCND1 is expressed in response to mitogenic signals promoting transition through the restriction point in the G1 phase of the cell cycle, and its activity reaches a maximum by associating with cyclin-dependent kinase 4 (CDK4) and cyclin-dependent kinase 6 (CDK6) in mid to late G1 phase (Sherr 1996). Over-expression of CCND1, due to deregulation, leads to propagation and accumulation of cell damage or genetic errors, and consequently contributes to tumor development (Rimokh et al. 1994). Recent studies have shown that the polymorphisms of CCND1, especially 870G>A (Pro241Pro), are associated with the risk, early onset and the outcome of multiple malignancies, including colorectal cancer and squamous cell carcinoma of the head and neck (Zheng et al. 2001; Porter et al. 2002). CCND1 870G>A polymorphism, located on the splicing region of exon 4, modulates the production of two types of transcripts by affecting CCND1 splicing (Howe and Lynas 2001). BCL2 is localized to mitochondria and interferes with programmed cell death independent of promoting cell division (Hockenbery et al. 1990). When BCL2 is over-expressed, the cell killing induced by various stimuli can be prevented or markedly reduced (Reed 1994). The various anti-apoptotic functions resulting from the different expressions of the BCL2 protein in lymphocytes seem to be associated with development of autoimmune disease. The non-synonymous polymorphism of BCL2, Ala43Thr (G>A), may contribute to resistance to autoimmune disease (Komaki et al. 1998).

Recent studies have shown that CCND1 and BCL2 may influence on the body mass and the regulation of various metabolisms. CCND1 may affect the body size (Sicinski et al. 1995) and over-expression of CCND1 in breast cancer cells triggers autonomous growth by controlling cell-cycle exit (Zwijsen et al. 1996). BCL2 is well known as a mitochondrial membrane protein and might have an unknown role in the regulation of basal mitochondrial energy metabolism (Shimizu et al. 1998; Kowaltowski et al. 2000). Recent evidence suggests that BCL2 may also affect growth regulation. BCL2 delays the entry into the cell cycle of mitogen-stimulated B and T lymphocytes, and also delays the serum stimulation-induced entry into S phase of quiescent fibroblasts (O’Reilly et al. 1996). Furthermore, it was presented that the linkage between cholesterol esterification and the control of cell-cycle G1/S transition due to the function of CCND1 and BCL2 was plausible, although this relationship is still obscure (Batetta et al. 2003). Although several evidences have shown that metabolic phenotypes may be influenced by those two genes through cell proliferation and apoptosis, direct association of CCND1 and BCL2 with metabolic phenotypes has not been investigated.

In this study, we scrutinized the genetic polymorphisms in CCND1 and BCL2 by direct sequencing for the sake of the discovery of additional polymorphisms and the evaluation of the relation between the polymorphisms in two genes and metabolic phenotypes. Here, we present 22 and eight genetic variants found in CCND1 and BCL2, respectively. The association analyses with two metabolic phenotypes [body mass index (BMI) and total cholesterol] were also examined with the polymorphisms of CCND1 and BCL2 in a Korean population.

Materials and methods

Subjects

A total of 320 unrelated healthy subjects were randomly recruited from an unselected population at Hallym University Sacred Heart Hospital (in routine health check-ups). The study protocol was approved by the Institutional Review Board of the Clinical Research Institute at Hallym University Hospital. Informed consent was obtained from all subjects before drawing blood. All study subjects were examined in the morning after overnight fast. Height and weight were recorded and used for calculation of BMI. Blood samples were drawn for both biochemical measurements of total cholesterol and DNA extraction. The clinical characteristics of the subjects are shown in Table 1.

Table 1 Clinical profiles of the study subjects
Full size table

Sequencing analysis of human CCND1 and BCL2 genes

We sequenced the whole CCND1 gene and exon–intron boundaries of BCL2 gene, including each promoter region (∼1.5 kb), to investigate genetic variants in 24 Korean DNA samples using the ABI PRISM 3700 DNA analyzer (Applied Biosystems, Foster City, Calif., USA). Twenty-nine primer sets of CCND1 and 16 primer sets of BCL2 for the amplification and sequencing analysis were designed based on GenBank sequences (Ref. Genome seq. for CCND1; NT_078088.02 released in Jan. 2004, and Ref. Genome seq. for BCL2; NT_025028.13 released in January 2004), respectively. Information regarding primers is available on our website (http://www.snp-genetics.com/user/news_content.asp?board_idx=168). Sequence variants were verified by chromatograms. For comparison with other major ethnic groups of allele frequencies of CCND1 and BCL2 polymorphisms identified in the Korean population, we also genotyped 50 Caucasian and 50 African–American DNAs obtained from the Human Genetic Cell Repository (http://locus.umdnj.edu/nigms/).

Genotyping by single-base extension (SBE)

For genotyping of polymorphic sites, amplification and extension primers were designed for SBE (Vreeland et al. 2002). Primer extension reactions were performed with the SNaPshot ddNTP Primer Extension Kit (Applied Biosystems). Information regarding primers is available on our website (http://www.snp-genetics.com/user/news_content.asp?board_idx=168).

Statistics

We examined Lewontin’s D′ (|D′|) and LD coefficient r 2 between all pairs of biallelic loci (Hedrick and Kumar 2001; Hedrick 1987). Haplotypes of each individual were inferred using the algorithm developed by Stephens et al. (2001) (PHASE). Chi-square test was performed for measuring P-values for the difference of distribution of polymorphisms among three major ethnic groups, including Korean, African–American, and Caucasian. Means and standard deviations (SD) of phenotypes, as well as P-values were calculated by multiple regression analyses controlling age (continuous value) and sex (male=0 and female=1) as covariates.

Results and discussion

CCND1 (BCL1), a cell-cycle regulatory protein, is associated with cell proliferation, prognosis and recurrence of cancer and its over-expression is often found in many different types of tumors (Coco Martin et al. 1999). BCL2 is an anti-apoptotic protein and plays an important role in the regulation of the rate of apoptosis (Sugino et al. 2000). Despite of the important features of these two genes, only few polymorphisms, including 870G>A and 1722G>C in CCND1 (Betticher et al. 1995; Holley et al. 2001) and Ala43Thr (G>A) in BCL2 (Komaki et al. 1998), had been reported.

We identified 22 sequence variants within the 16 kb of whole CCND1 including the −1500 bp promoter region: one in intron 1, two in intron 2, seven in intron 3, one in exon 4, seven in intron 4, and four in the 3′ UTR region. We also found eight sequence variants within 7.5 kb exons and their boundaries of BCL2, including the −1500 bp promoter region: one in promoter, three in exon 1, and four in the 3′ UTR region (Fig. 1 a1,b1). CCND1 +6829G>A (Pro241Pro), CCND1 +10656G>C, and BCL2 +127G>A (Ala43Thr), in this study, correspond to CCND1 870G>A, CCND1 1722G>C, and BCL2 Ala43Thr (G>A) in previous reports (Betticher et al. 1995; Komaki et al. 1998; Holley et al. 2001), respectively.

Fig. 1
figure 1

Gene maps and haplotypes of CCND1 and BCL2. Coding exons are marked by black blocks and 5′ and 3′ UTR by white blocks. The first base of translational start site is denoted as nucleotide +1. Asterisks indicate polymorphisms genotyped in a larger Korean population (n=320). The frequencies of polymorphisms without larger scale genotyping are based on sequencing data (n=24). Bold face indicates polymorphisms that newly identified in this study. Number in parenthesis means distribution of rare allele of polymorphisms and change of amino acid are also indicated. r 2=1 means absolute LDs among polymorphisms. a1 Polymorphisms identified in CCND1 on chromosome 11q13 (Ref. Genome Seq. NT_078088.02). a2 Haplotypes of CCND1 and their frequencies among three ethnic groups. Haplotypes with frequencies ≥0.003 are presented. Others (1) contain rare haplotypes: CCGDelTCG, CCADelCCG, TCADelTGG, and TAADelCGG. b1 Polymorphisms identified in BCL2 on chromosome 18q21.3 (Ref. Genome Seq. NT_025028.13). b2 Haplotypes of BCL2 and their frequencies among three ethnic groups

Full size image

Seven polymorphisms in CCND1 and six in BCL2 were selected for a larger scale genotyping (n=320) by considering their allele frequencies, haplotype-tagging status, and LDs among polymorphisms. The frequencies of those 13 polymorphisms of CCND1 and BCL2 genes were shown in Table 2. The significant differences in frequencies of polymorphisms and haplotypes were observed among three ethnic groups (P=0.03 to <0.001) (Table 2; Fig. 1 a2,b2). Linkage disequilibrium coefficiencies (|D′|) and r 2, among polymorphisms were also calculated (data not shown; information about LDs among polymorphisms is available at http://www.snp-genetics.com/user/news_content.asp?board_idx=168).

Table 2 Single-nucleotide polymorphisms and allele frequencies of CCND1 and BCL2 in Korean (n=320), African–American (n=50) and Caucasian (n=50) subjects
Full size table

Positive associations of CCND1 870G>A polymorphism [+6829G>A (Pro241Pro), in this study], located on the splice donor region of exon 4, with a wide variety of cancers have been suggested in different populations including Chinese (Deng et al. 2002), Japanese (Wang et al. 2003), and Caucasian (Zheng et al. 2001; Porter et al. 2002). It has been also reported that CCND1 1722G>C (+10656G>C in this study) is associated with both poorly differentiated tumors and reduced disease-free interval independent of the influence of CCND1 870G>A (Holley et al. 2001). A positive association of BCL2 Ala43Thr (G>A) polymorphism [+127G>A (Ala43Thr) in this study] with autoimmune diseases such as type 1 diabetes mellitus (T1DM) has been reported in the Japanese population (Komaki et al. 1998), whereas there was no association with Caucasian (Heding et al. 2001).

The associations of CCND1 and BCL2 polymorphisms (seven polymorphisms and two major haplotypes in CCND1, and six polymorphisms and two major haplotypes in BCL2) with metabolic phenotypes (BMI and total cholesterol) were analyzed in this study. The results of statistical analyses are summarized in Table 3. In association analysis of cholesterol with CCND1 and BCL2 polymorphisms based on the evidence that cholesterol was related to the cell replication and apoptotic cell death mainly due to the functional distribution of CCND1 and BCL2 genes (Kockx et al. 1998), we could not find any association. On the other hand, among polymorphisms analyzed, BCL2 +21A>G (Thr7Thr) showed a weak association with BMI (P=0.03; Table 3). The highest BMI was found in +21AG genotypes, intermediate in +21AA genotypes, and the lowest in +21GG genotypes. However, this association of BCL2 +21A>G with BMI might be controversial because the P-value did not reach the significance level when multiple comparison tests were applied. Although it had been proposed that the modulation of the cell cycle and apoptosis of CCND1 and BCL2 was associated with increased BMI (Suga et al. 2001), polymorphisms of CCND1 and BCL2 including BCL2 +21A>G, seem to be not responsible for the association with cell cycle and apoptosis through change in levels of BMI. Additional studies would be needed to clarify the association with BMI.

Table 3 Regression analysis for age-adjusted and sex-adjusted phenotype with CCND1 and BCL2 polymorphisms in the Korean population (n=320). Genotype and haplotype distributions, means±SD of each value, and P-values of codominant models for regression analyses controlling for age (continuous value) and sex (male=0 and female=1) as covariates are shown
Full size table

In summary, we identified 22 and eight polymorphic sites in the human CCND1 and BCL2 genes, respectively. Among variants of two genes, seven variants in CCND1 and six in BCL2 were genotyped in larger scale subjects (n=320). Haplotypes, their frequencies and linkage disequilibrium coefficients (|D′| and r 2), were estimated. Using statistical analyses to evaluate the associations of CCND1 and BCL2 polymorphisms with two metabolic phenotypes, no reliable associations were found. This information about genetic polymorphisms in CCND1 and BCL2 might be useful for host genetic studies of other diseases, especially cancer and apoptosis.