Introduction

Inflammatory bowel disease (IBD) commonly refers to ulcerative colitis (UC) and Crohn's disease (CD), which is a chronic inflammatory disease of the gastrointestinal tract of unknown etiology (Blumberg et al., 1999). IBDs are complex and multifactorial involving genetic, environmental and microbial factors (Fiocchi, 1998; Podolsky, 2002). The balance between pro- and anti-inflammatory cytokines secreted by T cells is responsible for both initiation and perpetuation of IBD. Cytokine production in lamina propria CD4+ T lymphocytes differs between CD and UC. Whereas CD is associated with increased production of T helper 1 cell (Th1) type cytokines, such as IFN-γ and TNF-α, UC is associated with T cells that produce large amounts of the Th2 type cytokine IL-5, however, IFN-γ production is not affected (Targan et al., 1995; Fuss et al., 1996; Plevy et al., 1997).

Data from genome-wide linkage analysis has implicated multiple IBD susceptibility loci in IBD affected sibling or relative-pair families (Ahmad et al., 2004). Results from a number of studies have suggested the existence of at least nine IBD loci (referred to as IBD 1-9) in the human genome. Whereas some loci appear to be specific to UC (IBD2) (Parkes et al., 2000), others seem to confer susceptibility to both CD and UC (IBD3) (Satsangi et al., 1996; Hampe et al., 1999; Van Heel et al., 2004).

Interferon induced transmembrane protein 3 (IFITM3, also known as 1-8U) is a member of the IFITM family, which mediates cellular processes, including the homotypic cell adhesion functions of interferons (IFNs). Expression levels of IFITM genes have been found to be up-regulated in gastric cancer cell and colorectal tumors (Yang et al., 2005; Andreu, 2006). IFITM3 was initially cloned from a human lymphoid cell cDNA library (Lewin et al., 1991), and is located on chromosome 11p15.5 (Lange et al., 2003). Hisamatsu and colleagues reported that the IFITM3 gene demonstrated strong expression in UC-associated cancers, describing it as a preferential marker of colitis-associated colon cancer in UC (Hisamatsu et al., 1999). Although IFITM3 is one of the crucial candidate genes for IBD pathogenesis, a correlation between single nucleotide polymorphisms (SNPs) of IFITM3 and individual susceptibility to various diseases, including UC, has not been reported. In an attempt to better understand the genetic influences of IFITM3 on UC, we have identified possible variation sites and SNPs through the two exons of IFITM3 and their boundary intron sequences, including the ~2.1 kb promoter regions. To determine whether or not these IFITM3 SNPs are associated with susceptibility to UC, genotype and allele frequencies of IFITM3 polymorphisms were analyzed on genomic DNAs isolated from patients with UC and healthy controls. We also investigated haplotype frequencies constructed by these SNPs in both groups.

Results

We examined expression patterns of IFITM3 mRNA in 15 normal human tissues and peripheral leukocytes (Figure 1). Our results showed that the expression level of IFITM3 mRNA was highest in peripheral leukocytes, while the expression level of IFITM3 mRNA in brain and thymus was barely detected (Figure 1A). We also examined expression patterns of IFITM3 mRNA in the various tissues of the human digestive system (Figure 1). The expression level of IFITM3 mRNA was significantly higher in tissues of the ileum and cecum than those in other tissues of the digestive system (Figure 1B). These results suggest that IFITM3 might modulate immune cell function and the intestinal immune system.

Figure 1
figure 1

Expression patterns of the IFITM3 mRNA. The expression levels of IFITM3 mRNA in various human tissues: 1, lung; 2, brain; 3, muscle; 4, pancreas; 5, kindey; 6, placenta; 7, heart; 8, liver; 9, spleen; 10, thymus; 11, prostate; 12, testis; 13, ovary; 14, intestine; 15, colon; 16, leukocyte (A), and the tissues of human digestive system: 1, liver; 2, stomach; 3, colon (desending); 4, colon (transverse); 5, colon (ascending); 6, ileocecum; 7, duodenum; 8, rectum; 9, jejunum; 10, ileum; 11, cecum; 12, esophagus (B).

To determine variation sites in the IFITM3 gene, including the promoter region (~2.0 kb), we scanned genomic DNAs isolated from 48 unrelated controls using direct sequence analysis. We identified a total of seven SNPs, g.-1129G>A (novel SNP) in the promoter region, g.-204T>G (rs3888188), g.-188T>C (rs6598045), g.-181T>C (rs7478728), g.-178A>C (novel SNP), g.-175T>C (rs7479267) in the 5'-UTR region, and g.42C>T (rs12252) in the exon 1 (Figure 2). The g.42C>T polymorphism located in the coding region was synonymous SNP (Ser14Ser). To identify multiple variation sites in the promoter region, PCR and sequencing analysis were performed using IFITM3-SF2 and IFITM3-SR2 primers (Figure 2). There exist at least five types with the difference length by repeated sequence (data not shown). Linkage disequilibrium coefficients (|D'|) between all SNP pairs were calculated, and found the absolute LD (|D'| = 1 and r2 = 1) between g.-204T>G and g.-178A>C (data not shown). Among the polymorphisms identified in this study, three SNPs (g.-204T>G, g.-175T>C and g. 42C>T) were selected for large samples genotype analysis. The D' values between g.-204T>G and g.-175T>C (or g. 42C>T) were 0.617 and 0.943, respectively, and between g.-175T>C and g. 42C>T was -0.873 (data not shown). We analyzed genotype and allele frequencies of patients with UC and healthy controls (Table 1, co-dominant analysis). In patients with UC, genotype frequency of the g.-204T>G polymorphism was significantly different from the frequency of the healthy control group (P = 0.0146).

Figure 2
figure 2

Locations of each single nucleotide polymorphisms (SNPs) and variation sites in IFITM3. Coding exons are marked by black blocks and 5'- and 3'-UTR by white blocks. The positions of SNPs were calculated from the translation start site. Putative transcription factor sites were searched at www.cbrc.jp/research/db/TFSEARCH.html. The reference sequence for IFITM3 was based on the sequence of human chromosome 11, clone RP13317D12.

Table 1 Genotype and allele analyses of the polymorphisms of IFITM3 gene in UC patients and healthy controls. *Calculated from the translation start site. †Logistic regression analyses were used for calculating OR (95% CI; confidence interval). ‡Value was determined by Fisher's exact test or χ2 test from 2 × 2 contingency table

We also evaluated haplotype frequencies among g.-204T>G, g.-175T>C and g. 42C>T polymorphisms in both healthy controls and patients with UC (Table 2). While four major (more than 5%) haplotypes explaining more than 95.1% of distributions were identified in patients with UC, three major haplotypes (95.3%) out of eight possible haplotypes were identified in the healthy control group. The distribution of haplotype TCT and GTT were significantly different between healthy controls and patients with UC (P = 0.001 and P = 0.015, respectively). These results suggested that IFITM3 polymorphisms might be one of the most important genetic factors in IBD susceptibility.

Table 2 The haplotype frequencies in both IBD patients and controls in IFITM3 polymorphisms. *Values were constructed by EM algorithm with genotyped SNPs. †Values were analyzed by Chi-square

Discussion

During the developmental stage, the activated Th cells differentiate into two phenotypically and functionally distinct cell types, Th1 and Th2 (Mosmann and Coffman, 1989; Abbas et al., 1996). Th1 cells produce the cytokines IFN-γ, IL-2 and lymphotoxin, which are commonly associated with cell-mediated immune response to intracellular pathogens and induction of organ-specific autoimmune disease (Kuchroo et al., 1995; Abbas et al., 1996). On the other hand, the Th2 cell related cytokines IL-4, IL-5, and IL-10, have a known association with atopic and allergic diseases. Th1 and Th2 cells cross-regulate one another in their differentiation. In IBD, UC is associated with T cells production of large amounts of Th2-type cytokines and eosinophils (Plevy et al., 1997; Farrell, 2002). There are many differences regarding disease extension, localization, behavior, and occurrence of extra- intestinal manifestations within UC (Forbes, 2003; Travis and Jewell, 2003).

Multiple IBD susceptibility loci (IBD 1-9) have been implicated. Among these loci, IBD1 and IBD3 are the most extensively studied genetic regions in IBD. The human leucocyte antigen (HLA) complex is located on the IBD3 locus (Williams et al., 2002; van Heel et al., 2004). HLA class II variants are associated with the susceptibility and phenotype in both CD and UC (Satsangi et al., 1996; Hampe et al., 1999; Stokkers et al., 1999; Yang et al., 1999; Dechairo et al., 2001). We previously reported on the exon 4 variation of the Tim-1 gene, and the polymorphisms of eotaxin-2 and eotaxin-3 are associated with UC in the Korean population (Song et al., 2004; Park et al., 2005).

The human IFITM3 gene, a member of the IFITM family, consists of two exons and one intron. IFITM proteins were first discovered in T98G neuroblastoma cells that express the proteins in response to interferon stimulation (Friedman et al. 1984). These proteins play distinct roles in mouse primordial germ cell homing and repulsion (Tanaka et al. 2005). In the aspect of gastrointestinal tract, IFITM3 gene was first isolated from colitis-associated cancer tissues and severely inflamed mucosa in the colon of UC patients using differential display PCR (Hisamatsu et al. 1999). Although the IFITM3 gene may be an important candidate gene for UC, few results for this gene have been published; therefore, the precise function of the IFITM3 gene is unknown, there has been no research on genetic variation. The IFITM3 gene was recently identified, and several reports have suggested that IFITM3 is one of the crucial candidate genes for IBD (Wu et al. 2007). In this study, we showed that expression level of IFITM3 mRNA was significantly higher in the ileum and cecum tissues of the human digestive system. We identified a total of seven SNPs and multiple variation regions in the IFITM3 gene. The genotype frequency of g.-204T>G polymorphisms in patients with UC was significantly different from that of the healthy control group (Table 2). This result strongly suggests that SNPs of IFITM3 may be associated with susceptibility to UC. Polymorphisms within the binding site of the promoter region may influence expression level, by suppression of binding between the specific transcriptional binding site, and the transcription factor. Accordingly, it is interesting to know that the IFITM3 polymorphism (g.-204T>G) may have some influence on susceptibility to UC. Thus, our results will be important in future studies, to determine whether or not this polymorphism affects the levels of IFITM3 gene expression and function. In summary, our results strongly suggest that polymorphisms of the IFITM3 gene may be associated with susceptibility to UC, as well as the TCT haplotype, by g.-204T>G, g.-175T>C, and g. 42C>T polymorphisms in the IFITM3 gene, and may be an important genetic marker for IBD.

Methods

Patients and DNA samples

Blood samples were obtained from 183 UC patients (100 males and 83 females) and 566 healthy controls (351 males and 215 females). Mean ages of IBD patients and controls were 41.1 yr and 40.5 yr, respectively. Genomic DNA was extracted from leukocyte in peripheral blood by use of a standard phenol-chloroform method or by use of a Genomic DNA Extraction kit (iNtRON Biotechnology, Korea) according to the manufacturer's directions. IBD patients were recruited from the outpatient clinic at Wonkwang University Hospital. Patients were classified into the IBD group according to clinical features, endoscopic findings, and histopathologic examinations. Healthy controls were recruited from the general population, and had received comprehensive medical testing at the Wonkwang University Hospital. All subjects in this study were Korean.

PCR and sequence analysis

The entire coding regions of the IFITM3 gene, including the ~2.0 kb promoter regions, were partially amplified by PCR using the two primer pairs (Supplemental Data Table S1). PCR reactions were prepared by previously described procedures (Chae et al., 2004). Amplification was carried out in a GeneAmp PCR system 9700 thermocycler (PE Applied Biosystem) at 95℃ for 5 min in order to pre-denature the template DNA, followed by 30 cycles of denaturation at 98℃ for 10 s, annealing at 68℃ for 30 s and extension at 72℃ for 2.0 or 2.5 min. The final extension was completed at 72℃ for 7 min. PCR products purified by use of a PCR purification kit (Millipore) were used template DNA for sequencing analysis. Purified PCR products were sequenced using the ABI Prism BigDye Terminator cycle sequencing system (PE Applied Biosystems) on the ABI 3100 automatic sequencer (PE Applied Biosystem). Both sense and antisense strands of PCR products were directly sequenced using the same primers used for the PCR amplification, and five primers were additionally used to sequence the promoter and intron 1 region (Supplemental Data Table S1). SNPs and variation sites of the IFITM3 gene were detected by direct sequence analysis. The reference sequence for the IFITM3 gene was based on the sequence of human chromosome 11, clone RP13- 317D12.

IFITM3 mRNA expression level

The expression level of IFITM3 mRNA in various tissues was determined using MTC multiple tissue cDNA panels (Clontech, CA), and IFITM3-MF1 and IFITM3-MR1 primers (Supplemental Data Table S1).

PCR-RFLP

Genotyping for g.42T>C in the IFITM3 gene was performed by PCR-RFLP. A region of the IFITM3 gene containing the g.42T>C polymorphism was partially amplified by PCR using the IFITM3-PF2 and IFITM3-SR1 primer pairs (Supplemental Data Table S1). PCR products were digested with 1 units of Bal I (Takara Co., Ltd., Japan) for 3 h at 37℃, and then separated on 1.5% agarose gel and visualized under UV with ethidium bromide. Following restriction enzyme digestion, the PCR products for g.42T>C (1,357 bp) took the form of two fragments, of 776 bp and 581 bp (Supplemental Data Figure S1).

Single-base extension (SBE)

Genotype analysis for g.-175T>G and g.-204T>G in the IFITM3 gene was performed by way of single-base extension (SBE), using the ABI Prism® SNaPshot™ Multiplex kit (Applied Biosystems). PCR products purified with a PCR purification kit (Millipore) were used as template DNA for two SBE primers (Supplemental Data Table S1). The SBE reaction mix was prepared according to the manufacturer's directions. The primer extension reaction was performed at 96℃ for 1 min, followed by 25 cycles at 96℃ for 10 s, 55℃ for 40 s, and 60℃ for 30 s. To clean up the primer extension reaction, 1.5 units of CIP (Promega) was added to the reaction mixture; the mixture was incubated at 37℃ for 90 min, followed by 15 min at 72℃, for the purposes of enzyme inactivation. Purified extension products were added to Hi-Di formamide (Applied Biosystems) and incubated at 95℃ for 5 min, followed by 5 min on ice; electrophoresis was then performed using the ABI Prism 3100 Genetic Analyzer. Results were analyzed using the ABI Prism GeneScan and Genotyper software (Applied Biosystems).

Statistic analysis

IBD patients and control groups were compared using case-control association analysis. The χ2 test was used to estimate Hardy-Weinberg equilibrium (HWE). Allele frequency was defined as the percentage of individuals carrying the allele among the total number of individuals. Logistic regression analyses were used to calculate odds ratios (95% confidence interval) for SNP sites. Linkage Disequilibrium (LD) analyses by pair-wise comparison of biallelic loci and haplotype frequencies of the IFITM3 gene for multiple loci were estimated using the expectation maximization (EM) algorithm with SNPAlyze software (DYNACOM, Japan). A P-value of less than 0.05 was considered an indication of statistical significance.