Introduction

In general, gastrointestinal cancer (GI) includes esophageal cancer, gastric cancer, colon cancer, rectal cancer, pancreatic cancer, liver cancer, gallbladder cancer, bowel cancer, appendicitis, and anal cancer, which is noted as a major cause of cancer-related death (Bray et al. 2018). Mucins are high molecular weight epithelial glycoprotein contained in mucus and is a viscous secretion that covers the epithelial surface of the digestive or respiratory system. These mucins form the first line of physical defense and act as chemical and immunological barriers (Byrd and Bresalier 2004). Abnormal mucin expression and glycosylation are related to chronic inflammation and gastrointestinal cancer.

Twenty human mucin genes are divided into functionally secreted gel-forming mucins and transmembrane mucins, and some MUC gene products do not fit well in either class or have both properties (Byrd and Bresalier 2004). Among them, the MUC2 gene encodes gel-forming mucin, located between H-Ras and IGF2 on the 11p15.5 chromosome (Byrd and Bresalier 2004). It is also confirmed that the central domain of mucin has variable number of tandem repeats (VNTR, minisatellites), consisting of threonine-, serine- and proline-rich repeated peptides (Byrd and Bresalier 2004). While the effect of minisatellites on gene function is unclear, several genes (MUC genes, H-Ras, hTERT, and Boris) have shown susceptibility to certain diseases depending on the length of the minisatellites (Carvalho et al. 1997; Kwon et al. 2019; Kyo et al. 1999; Weitzel et al. 2000; Yoon et al. 2010a, b, 2016).

In previous work, we detected eight tandem repeats regions within MUC2, of which the short rare alleles of MUC2-MS6 were associated to gastric cancer (Jeong et al. 2007). In particular, some minisatellites regions were reported to affect gene expression (Yoon et al. 2016), and they also showed loss-of-heterozygosity (LOH) due to genomic instability during cancer development (Jeong et al. 2007; Kwon et al. 2019).

In this study, we examined the multi-allelic properties of MUC2-MS8, and then we identified whether there was an association between certain specific alleles of MUC2–MS8 and susceptibility for gastrointestinal cancer. Next, we compared the length difference of MUC2-MS8 in genomic DNAs from matched blood and cancer tissue obtained from patients with gastric cancer, colon cancer, or rectal cancer. Here, we suggest that MUC2-MS8 alleles may be associated to genomic instability and related to susceptibility to rectal cancer.

Materials and methods

Preparation of genomic DNA from peripheral blood and cancer tissues

We conducted a case–control study with genomic DNA obtained from 1229 cancer-free controls, 455 gastric cancer, 192 colon cancer, and 278 rectal cancer (Table 1). In interviews, people without a history of cancer and currently without cancer were recruited as control groups. We also isolated genomic DNAs from 107 stomach cancer patients, 125 colon cancer patients, and 85 rectal cancer patients by matching each patient's blood and cancer tissue. After obtaining approval from the bioethics committees of Dong-A University Hospital (#IRB-07-10-7; Busan, Korea), Inje University Paik Hospital (#IRB11-011; Busan, Korea) and Chungbuk National University Hospital (#IRB-2006-1; Cheongju, Korea), cancer cases and control groups were recruited.

Table 1 Age and sex distribution of cases and controls
Full size table

PCR analysis of minisatellite polymorphism in MUC2-MS8

We analyzed MUC2–MS8 polymorphisms by PCR using primers designed in our previous study (Jeong et al. 2007); MS8, F-GTAGGCCCCACCGTGTTT & R-AGAAGCTCTGACATGACATCTTGGCC. After the PCR reaction was carried out as in the previous study (Jeong et al. 2007), PCR products were separated by gel electrophoresis using 2% SeaKem LE agarose (Cambrex, ME) (Figs. 1, 2).

Fig. 1
figure 1

Haplotyic patterns of MUC2-MS8 minisatellites in cancer-free controls and cases with gastrointestinal cancers. A The consensus sequence of MUC2-MS8 minisatellites repeat units (Jeong et al. 2007). Comparison of haplotype patterns of MUC2-MS8 between controls (B), (Jeong et al. 2007) and patients with gastric (C), and rectal cancers (D). Rare alleles are indicated by white arrows in (B, C) and (D). Size markers (M) are given in kb. E The repeat number, size of the PCR products and allelic frequency in controls and gastrointestinal cases are indicated

Full size image
Fig. 2
figure 2

Effect of the MUC2-MS8 alleles in luciferase reporter vector. Two different cell lines [293 T (A) and LoVo (B)] were transfected with eight different vectors. As a control, the pGL3 promoter vector was used, and the other TR vectors (6TR to 15TR) are constructs inserted into the pGL3 promoter vector with varying copy numbers of the repeat unit of the MUC2-MS8 region. Results are showed as the ratio of activity of Firefly to Renilla luciferase. The average mean ± SE values from triplicate transfections are shown, representative of four independent experiments

Full size image

Construction of the pGL3-promoter vector with MUC2-MS8 tandem repeats

Three common (6TR, 8TR and 10TR) allele and four rare alleles (7TR, 9TR, 13TR and 15TR) were amplified from genomic DNA from controls and cases with gastrointestinal cancers (Fig. 2). Seven different length fragments of the MUC2-MS8 were amplified and inserted into BamHI/SalI sites (in the enhancer region) of the pGL3-Promoter vector (Promega, WI, USA), respectively. All seven different constructs were confirmed by DNA sequencing.

Cells and luciferase assay

The effect of MUC2-MS8 alleles on luciferase expression were investigated using 293 T [human embryonic kidney cell line, Korea Cell Line Bank (KCLB), Korea] and LoVo (colorectal cancer cell line, KCLB, Korea) cells. For the luciferase assay, cells were transfected with each construct of the seven different MUC2–MS8 (0.5 µg per well) as described in the previous study (Kim et al. 2021).

Statistical analyses

Heterozygosity was confirmed by calculating the degree of polymorphism in the MUC2-MS8 region in a range between 0 and 1 (Nei and Roychoudhury 1974). Odds ratio (OR) and 95% confidence interval (CI) determination and statistical significance validation were performed as described in previous studies (Jeong et al. 2007).

Results

Polymorphic analysis of MUC2-MS8 alleles

In a previous study, we identified nine MUC2 minisatellites through analysis of the MUC2 genome sequence (Jeong et al. 2007). For MUC2-MS8 (repeat unit 46 bp, Fig. 1A), located in Intron 40, polymorphism was identified using a cancer-free control group (Jeong et al. 2007). All identified MUC2-MS8 alleles can be classified as rare alleles if their frequency is less than 1%, and the rest can be classified as common alleles. In the control group, we found 4 alleles with repeat counts of 6 to 10 (6TR, 8TR, 9TR, and 10TR), with 10 repeats present as the most common allele (57%) (Jeong et al. 2007) (Fig. 1) and 9TR was identified as a rare allele (Fig. 1B) (Jeong et al. 2007).

We analyzed unrelated cancer-free individuals (1229) and cancer cases [cases with gastric cancer (486), colon cancer (220), and rectal cancer (278)] to assess the degree of polymorphism in MUC2-MS8 (Table 1). The heterozygosity of the MUC2-MS8 region ranged from 0.583 to 0.635 in the control group and the three types of cancer patient groups (controls, 0.585; gastric cancer, 0.583; colon cancer, 0.635; rectal cancer, 0.588), and there was no significant difference.

We also compared the distribution and frequency of the polymorphic MUC2-MS8 alleles between controls (Fig. 1B) and cancer patients with gastric (Fig. 1C), rectal cancers (Figs. 1D, 2A). The identified MUC2-MS8 appeared as 7 different length alleles with 6–15 repeat copies and 833 bp-1247 bp in length, with the most common allele (56.37%) containing 10 repeats in total controls and cases (Fig. 1E). In this study, additional three rare alleles (cancer specific rare alleles) were confirmed that were not identified in the control group (Jeong et al. 2007): two in the gastric cancer group (13TR, 15TR; Fig. 1C, E) and one in the rectal group (7TR; Fig. 1D, E).

Genetic susceptibility of MUC2-MS8 rare alleles to cancer

MUC2, called intestinal mucin, is the main gel-forming intestinal mucus and a major structural component of the mucus gel (Corfield et al. 2000). This study analyzed the distribution and frequency of polymorphic MUC2-MS8 in gastrointestinal cancer patients to determine whether this variable minisatellite exhibits gastrointestinal cancer sensitivity. Table 2 summarizes the frequency of each MUC2-MS8 allele identified in cancer patients and control groups. In rectal cancer cases, the frequency of rare MUC2-MS8 alleles were 0.54%, compared to 0.08% in cancer-free controls. The analysis revealed a statistically significant association between the rare alleles and odds of rectal cancer (OR, 6.66; 95% CI, 1.11–39.96; P = 0.0165). In particular, there was a significant association between the short rare alleles (< 10TR) and rectal cancer. In addition, the frequency of incidence of C/R genotypes with rare alleles also showed statistically significant values (P = 0.0164) when comparing the rectal cancer group and the control group (Table 3). These results indicate that the rare MUC2-MS8 alleles is associated with the risk of rectal cancer. In other cancers, the association between rare alleles and cancer was not statistically significant but showed an increasing trend (Tables 2, 3).

Table 2 Frequency of MUC2–MS8 rare alleles and risk of cancer
Full size table
Table 3 Univariate ORs and 95% CIs for cancer associated with MUC2-MS8 rare alleles
Full size table

Table 4 summarizes the frequency of rare MUC2-MS8 alleles according to age at diagnosis. In controls, no difference in the frequency of rare alleles was found between younger (< 55 years) and older individuals (≥ 55 years) (P = 0.8925). In comparison to older patients (≥ 55 years), however, we found that younger individuals (< 55 years) with cancer had an increased OR (8.30, CI: 0.74–93.19: P = 0.0416) of association between rare MUC2-MS8 alleles and rectal cancer (Table 4). In particular, a comparison of young groups in controls and rectal cancer cases showed a significant difference in the association between rectal cancer and rare MUC2-MS8 allele in young patients: OR 24.93 (CI: 2.22–279.3, P = 0.0001). These results suggest that rare MUC2-MS8 alleles may be genetically associated with rectal cancer.

Table 4 Frequency of rare alleles at MUC2-MS8 loci associated with age
Full size table

Possible function of MUC2-MS8 alleles for the expression of MUC2

To examine the possible role of MUC2-MS8 in intron regions, we constructed reporter vectors that included the seven different alleles (6TR, 7TR, 8TR, 9TR, 10TR, 13TR, and 15TR) of MUC2-MS8 (Fig. 2) inserted into the pGL3-Promoter vector. Then, two cell lines (293 T/HEK293T and LoVo) used to determine the effect of MUC2-MS8 alleles on MUC2 expression. After transfection into 293 T and LoVo cells using these seven types of vectors, luciferase activity was investigated (Fig. 2). In 293 T cells, the activity of luciferase was significantly reduced when transfecting vectors containing TR alleles of MUC2-MS8 compared to when transfected with pGL3-Promoter control vector (Fig. 2). These results suggest that the repeat sequence of MUC2-MS8 suppresses the expression of MUC2. Even when the colorectal cancer cell line LoVo was used, significant expression inhibition by the repeat sequence was observed (Fig. 2).

Analysis of MUC2-MS8 instability in gastrointestinal cancer tissues

Loss of heterozygosity (LOH) at chromosome has been found in a variety of human carcinoma, suggesting that chromosomal arrangements important to tumorigenesis (Lazar et al. 1998). Since the MUC2 gene contains a high-density minisatellite, we investigated whether this region is involved in chromosomal instability in cancer tissues. This idea was examined by comparing the polymorphic MUC2-MS8 alleles in the blood and cancer tissues from the same patient with gastrointestinal cancer (107 gastric cancer, 125 colon cancer and 85 rectal cancer; Fig. 3). In genomic DNA obtained from blood and cancer tissue from the same patient with gastrointestinal cancer, there were eight cases of LOH in MUC2-MS8 in DNA obtained from cancer tissues (Fig. 3). When 107 gastric cancer patients were examined, the frequency of rearrangement was 2.8% (2/107) (Fig. 3A). In addition, the LOH frequency of MUC2-MS8 was 0.8% (1/125) in colon cancer tissues (Fig. 3B) and 5.9% (5/85) in rectal cancer tissues (Fig. 3C). The frequency of these LOH shows a trend very similar to the results of the previous cases-controls study that confirmed the association with cancer.

Fig. 3
figure 3

LOH at MUC2-MS8 region in gastrointestinal cancer tissues. Comparison MUC2-MS8 alleles between blood and cancer tissues obtained from the same patient with gastric (A), colon (B), or rectal cancer (C). LOH in cancer tissues are indicated by white arrows in electrophoretic patterns. Cancer tissue samples are indicated by asterisks and M indicates the size marker

Full size image

Discussion

Four secretion mucin genes, MUC2, MUC5AC, MUC5B, and MUC6, are clustered on chromosome 11, of which MUC2 has high intestinal expression and MUC5AC, MUC5B and MUC6 have high bronchial and upper gastrointestinal expression (Sylvester et al. 2001; Van Klinken et al. 1995). The main characteristic of apomucins is that they have a variable number tandem repeat (VNTR, minisatellites) sequence (Gendler and Spicer 1995; Van Klinken et al. 1995).

Minisatellites are tandem repeating DNA sequences distributed throughout the human genome, especially in the terminal regions of chromosomes (Jeffreys et al. 1985). Although the characteristics of these minisatellites are not clear, it has been reported that certain minisatellites are involved in chromosomal rearrangement or susceptibility to disease (Jeong et al. 2007; Kwon et al. 2019; Kyo et al. 1999). In a previous study, nine minisatellites were identified by analyzing the entire MUC2 sequence, eight of which were identified by PCR (Jeong et al. 2007). In MUC2-MS8 (introns 40), we found several canonical CACGT-binding sites for the MYC family of oncogenic transcription factors (Jeong et al. 2007). These transcription factors located in MUC2-MS8 region may be related to a carcinogenesis by the regulation of MUC2 expression. We examined whether the MUC2-MS8 region effects on the gene expression; we constructed vectors containing each minisatellite in enhancer regions of a reporter plasmid. Coincidence with luciferase assay, this analysis suggests that MUC2-MS8 is a potential region for regulation of gene expression even though it located in intron, a non-coding region (Fig. 2).

Rare alleles of minisatellites are associated high risk for various types of cancer (Jeong et al. 2007; Kwon et al. 2019; Kyo et al. 1999; Yoon et al. 2010b, 2016). Among the minisatellites of the MUC2 gene, a short rare gene of MUC2-MS6 has been identified as associated with stomach cancer, and this study has identified a link between rare MUC2-MS8 allergies and rectal cancer. In addition, the present study indicated that the incidence of the rare MUC2-MS8 allele was significantly higher in younger rectal cancer patients than in younger cancer-free controls when age was considered. These results suggest that the rare MUC2-MS8 allele may be genetically linked.

Numerous studies have reported that genetic instability appears in human neoplasia, and that persistent genetic instability is important in the pathogenesis of all colorectal cancers (Lengauer et al. 1997, 1998). MUC2 is located at the end of chromosome 11 and contains dense minisatellites involved in chromosome instability, and LOH mutations and microsatellite instability have also been reported in colorectal cancer (Jeong et al. 2007; Lengauer et al. 1998). We analyzed the MUC2-MS8 alleles in gDNA from blood and cancer tissue derived from patients with gastrointestinal cancers, and detected nine cases with LOH in MUC2-MS8 (Fig. 3).

Therefore, our observations show that MUC2-MS8 minisatellites can function as cancer risk indicators for rectal cancer and may be related to chromosomal instability. This study will provide useful references for understanding the functionality and complex genetic characteristics of MUC2.