Loss of heterozygosity and mRNA expression at deleted in colorectal cancer gene locus in gastric cancer

Abstract

AIM: To assess the effects of the deleted in colorectal cancer (DCC) gene changes on the development and progression of gastric cancer.

METHODS: The loss of heterozygosity (LOH) and mRNA expression DCC gastric cancer using a PCR-based detection method.

RESULTS: LOH was found in 35.3% (18/51) of the specimens, and the LOH was more frequently detected in tumors from patients with stage III or IV cancer (50.5%) than those in stages I or II (14.3%) (P < 0.05). The occurrence of LOH was not found to correlate with the histological type, tumor size, invasion depth and lymph node metastasis of gastric cancer. The mRNA expression of the DCC gene was studied in 26 of the 51 cases, of which LOE was found in 30.8% (8/26). LOE was not significantly correlated to LOH or other clinicopathological parameters.

CONCLUSION: LOH and LOE of DCC gene are frequently encountered in gastric cancer, and the LOH of DCC gene is a late event associated with progression of gastric cancer.

Key Words: Stomach neoplasms; DCC gene; Gene expression; mRNA; Heterozygosity loss

INTRODUCTION

Tumor suppressor genes play an important role in regulating normal cellular proliferation[1,2]. Conversely, inactivation of tumor suppressor genes at both alleles may allow a cell to lose normal growth controls and acquire a malignant phenotype. This inactivation may occur through a variety of mechanisms including deletion, rearrangement, point mutation, gene conversion, and binding of suppressor gene products with viral or cellular inactivating proteins[1,3]. The deleted in colorectal cancer (DCC) gene was first cloned based on frequent deletions affecting the 18q21 region in colon cancer[4]. Subsequently, loss of heterozygosity (LOH) or loss of expression (LOE) of DCC has been reported in several other tumor types, including breast[5,6], pancreatic[7], prostate[8] and testicular[9]carcinomas, glioblastomas[10] and hematological malignancies[11].

In a study of human gastric cancer, chromosome 18q was frequently affected by the loss of heterozygosity detectable in more than 60% of cases[12]. However, there have been no studies reported on LOE of DCC gene in gastric cancer. In order to investigate the effects of the DCC gene abnormality on the development and progression of gastric cancer, LOH and LOE of DCC gene were examined using a PCR based detection method.

MATERIALS AND METHODS

Tissue specimens

Tumor and corresponding noncancerous tissues were obtained from 51 patients who underwent surgical resection for gastric carcinoma between January 1993 and October 1996 at the Southwest Hospital. None of the patients had received any radiotherapy or chemotherapy preoperatively. Each pair of tumor and corresponding non-tumor tissues was stored at -80 °C immediately after the resection for experimental use. A 6 μm section was cut from each tissue and stained with hemotoxylin/eosin for pathological diagnosis. After diagnostic confirmation, a visual assessment was made of the approximate proportion of tumor cells vs normal cells in the tumor. Only the specimens in which tumor cells represented ≥ 60% of the tumor tissue were accepted for LOH and LOE analysis.

Total RNA isolation and DNA extraction

Total RNA was prepared from tumor and noncancerous tissues using the acid guanidinium thiocyanate method[13] and high molecular weight DNA was extracted using proteinase K digestion and phenol chloroform isoamyl alcohol extraction as previously described[14].

RT-PCR assay of DCC gene expression

RT-PCR was performed as described previously with some modifications[15]. DCC complementary DNA was amplified at 94 °C for 40 s; 49 °C for 40 s, and 72 °C for 1 min in a Perkin Elmer Thermocycler 2400 for 35 cycles. DCC primers were located on exons O and P, amplifying a 233 base pair fragment from the human mRNA[16]. A fragment of this size cannot be amplified from genomic DNA, for the primers were designed to frame sequences that cross an intron on the DCC gene. RT-PCR without RNA or without reverse transcriptase were included in each experiment as negative controls. Primers used were 5′ TTCCGCCATGGTTTTTAAATCA 3′ (DCC sense), and 5′ AGCCTCATTTTCAGCCACACA 3′ (DCC antisense).

PCR-LOH analysis

Fifty to 500 ng of genomic DNA were placed at 95 °C for 5 min in 20 μL buffer containing 10 mmol/L Tris (pH 8.3), 5 mmol/L KCl, 2.5 mmol/L MgCl₂, 0.1 μg/μL bovine serum albumin, sense and antisense primers at 1 μm concentration. Then 2.5 units of Ampli Taq DNA polymerase was added and PCR was run at 94 °C for 40s; 56 °C for 40 s, 72 °C for 1 min, for 35 cycles. For M2 and M3 polymorphism[17], PCR products were digested with MspI and analyzed on 25% agarose gels. For VNTR polymorphism[18,19], PCR products were directly separated on 2.5 gels. The gel was then stained with ethidium bromide and photographed under UV light. The primers were: 5′TGCACCATGCTGAAGATTGT 3′ (M2 sense), 5′AGTACAACACAAGGTATGTG 3′ (M2 antisense); 5′ CGACTCGATCCTACAAAATC 3′ (M3 sense), 5′ TCTACCCAGGTCTCAGAG 3′ (M3 antisense); 5′ GATGACATTTTCCCTCTAG 3′ (VNTR sense), and 5′GTGGTTATTGCCTTGAAAAG 3′ (VNTR antisense). Negative controls without genomic DNA were performed for each set of PCR reaction.

Data analysis

Photographs of thidium-stained gels were read by two observers independently. LOH and LOE was defined by a visible change in that allele: allele ratio in tumor compared to the matching normal tissues. A reduction of allelic intensity over 50% in tumor compared to the matching tissues was taken to be indicative of LOH or LOE, (Figure 1, Figure 2, Figure 3).

Figure 1 Heterozygosity loss of deleted in colorectal cancer (DCC) gene (M2) of gastric cancer. N: Normal tissue DNA; T: Tumor tissue DNA; 1: Heterozygote; 2-4: loss of heterozygosity (LOH).

Figure 2 Heterozygosity loss of deleted in colorectal cancer (DCC) gene (M3) of gastric cancer. N: Normal tissue DNA; T: Tumor tissue DNA; 1: Heterozygote; 4: Homozygote; 2 and 3: loss of heterozygosity (LOH).

Figure 3 Heterozygosity loss of deleted in colorectal cancer (DCC) gene (VNTR) of gastric cancer. N: Normal tissue; T: Tumor tissue DNA; 1 and 3: Heterozygote; 2 and 4: loss of heterozygosity (LOH).

Statistical analyses

Associations between variables were made with the Chi square test, a p value of less than 0.05 was considered to be statistically significant.

RESULTS

LOH of the DCC gene was determined by PCR-LOH in 51 specimens of gastric cancer. In order to raise the assay sensitivity, three different sites, i.e., M2, M3 and VNTR were used in this study. LOH of DCC was observed in 9 of 47 (19.0%) at M2, 7 of 50 (14.0%) at M3 and 3 of 26 (11.5%) at VNTR sites, respectively. If a positive allelic deletion of DCC was judged by LOH at one or any combination of these three sites, the incidence of LOH at the DCC locus was 35.3% (18/51). LOH was detected in 37.5% (6/16) of intestinal type of gastric cancer and 36.4% (12/33) of gastric type. Of the 51 cases of gastric cancer, 26 underwent the examination of expression of DCC mRNA, and LOE was observed in 30.8% (8/26) (Figure 4). The incidence of LOE was 44.4% (4/9) in the intestinal type and 23.5% (4/17) in gastric type. χ² test revealed no significant difference of the LOH and LOE between these two types of cancer (p > 0.05).

Figure 4 Loss of deleted in colorectal cancer (DCC) gene expression of gastric cancer. N: Normal tissue T: Tumor tissue; M: Marker of molecular weight, PBR 322/Hae III; 1 and 3: Normal expression; 2: loss of heterozygosity (LOH).

Correlation between LOH and LOE of DCC and clinicopathological data of gastric cancer are illustrated in Table 1. χ² test demonstrated that LOH of DCC was significantly higher in stages III and IV gastric cancer than that in stage I or II (P < 0.05). Correlation between LOH and LOE is shown in Table 2. The paired data were analyzed with χ² test and no significant correlation was found between LOH and LOE (p > 0.05).

Table 1 Relationship of loss of heterozygosity (LOH) and mRNA expression (LOE) of deleted in colorectal cancer (DCC) with clinicopathological parameters.

Clinicopathologic parameters	LOH/informative (%)	LOE/No. examined (%)
Differentiation
Well/moderate	4/12 (33.3)	1/6 (16.7)
Poor	12/28 (42.9)	5/14 (35.7)
Mucinous carcinoma	2/11 (18.2)	2/6 (33.3)
Tumor size
< 5 cm	4/20 (20.0)	4/12 (33.3)
> 5 cm	14/31 (45.2)	4/14 (28.6)
Serosal invasion
Absent	4/18 (22.2)	3/12 (25.0)
Present	12/33 (36.4)	5/14 (35.7)
Lymph node metastasis
Absent	5/24 (20.8)	2/13 (15.4)
Present	13/27 (48.1)	6/13 (46.2)
Clinical staging
Stages I-II	3/21 (14.3)	2/11 (18.2)
Stages III-IV	15/30 (50.0)a	6/15 (40.0)

^aP < 0.05 as compared with that of stage I and II.

Table 2 Relationship between loss of heterozygosity (LOH) and mRNA expression (LOE) of deleted in colorectal cancer (DCC) gene.

Groups	LOH	LOE	No. of cases (%)
1	+	+	4 (15.4)
2	-	-	14 (53.8)
3	-	+	4 (15.4)
4	+	-	4 (15.4)

+: Positive; and -: Negative.

DISCUSSION

The DCC gene is located on the human chromosome 18q^21.3. It was reported that the inactivation of this gene is closely related to the pathogenesis of colorectal, esophageal and pancreatic carcinoma, and this gene is considered a susceptible gene in gastrointestinal carcinomas. Uchino et al[12] and Ranzani et al[19] reported that the rate of LOH of DCC in gastric cancer was 61% and 42.9%, respectively; in our study, this rate was 35.5%. Together, these results suggest that DCC gene takes part in the pathogenesis of gastric cancer through its LOH.

To our knowledge, this is the first report on the expression of mRNA of DCC gene in gastric cancer, which confirmed the LOE of DCC mRNA in the gastric cancers with RT-PCR. This suggests that the inactivation of DCC in gastric cancer occurs in various patterns, and further confirms that the DCC gene is the susceptible gene of gastric cancer, playing an important role in the pathogenesis of gastric cancer.

The histological progression associated with the intestinal type of gastric cancer has been well documented, with apparent evolution through a sequence of superficial gastritis, intestinal metaplasia and dysplasia[20,21]. Lesions indicating an adenoma carcinoma sequence similar to that in the colorectum were also observed in the stomach[22]. In the present study we have found that the rate of LOE of DCC was as high as 44.4% in intestinal type of gastric cancer, which approached that of the rate seen in colorectal carcinoma[23]. However, the rate of LOE was rather low in gastric type of gastric cancer. Though there was no statistical difference between the two types of gastric cancer, similarities in genetic alterations between colorectal carcinoma and intestinal type of carcinoma may reflect a carcinogenetic pathway common to colorectal carcinoma and gastric carcinoma. It was also found in our study that the rate of LOH of intestinal type of DCC was similar between the two types of gastric cancer. Whether the effects of LOH and LOE of DCC gene are different between the two types of gastric cancer needs further studies.

The DCC gene encodes a molecule which shares high homology with the neural cell adhesion molecule[16]. Cell adhesion molecules are cell surface receptors that play critical roles in a number of different processes, including embryogenesis, thrombosis, wound healing, cell homing, and immunoreactivity, as well as tumor progression and metastasis. The inactivation of DCC gene may result in malignant degeneration of the cells and aid in the invasion and metastasis of a tumor. Kato et al[24] found that the incidence of LOH at DCC locus in colorectal carcinoma was significantly greater for patients with liver metastasis than for patients with no liver metastasis. Iino et al[23] and Yanoshita et al[13] observed that the expression of DCC gene in mRNA was greatly reduced or not detectable in invasive colorectal carcinoma in comparison with carcinoma in adenoma and intramucosal carcinoma. They indicated that the inactivation of the DCC gene was associated with the progression of early stage carcinoma to advanced stage. Itoh et al[25] revealed that the expression level of DCC mRNA was lower in liver matastasis than in primary carcinoma. These findings imply that the inactivation of the DCC gene occurs in the late stage of colorectal carcinoma, and was of prognostic significance. There were similar conclusions in the study of esophageal and pancreatic carcinoma[7,26]. In our study, the rate of LOH of DCC rose along with the increase of tumor size and the depth of invasion and the metastasis to lymph nodes, and LOE of DCC frequently occurred in gastric cancer of stages III and IV with lymph node metastasis. Though these data did not find a statistical significance, they may suggest that DCC gene plays a definite role in the proliferation, invasion and metastasis of gastric cancer. The LOH rate of DCC in our study was significantly higher in the stages III and IV than that in stages I or II, which indicates that LOH of DCC occurs in the late stage and is related to the advances of the malignancy. Ranzani et al[19] had similar findings as ours. Thus, it is expected that LOH and LOE of DCC may be potentially used as a prognostic factor for gastric cancer. However, to accurately reveal the correlation of LOH and LOE of DCC with clinicopathologic factors and prognosis, a larger number of cases should be examined.

The interrelation of LOH and LOE of DCC gene was preliminarily studied and it was found that LOH does not seem to be necessary for LOE of DCC mRNA, which was similar to the finding of others[14,27]. There might be some other causes, such as alterations in sequences controlling transcriptional regulation, point mutation or insertions within the DCC gene, or alterations in other gene controlling DCC gene expression. Further studies are required to examine these possibilities.