JP4426549B2 - Diagnosis method of gastric cancer using methylation of novel gene ACMG1 as an index - Google Patents

Description

  The present invention relates to a novel gene named ACMG1, a diagnosis method and diagnostic kit for gastric cancer using the same, and a method for selecting candidate substances for gastric cancer therapeutic agents.

CpG islands are regions rich in CpG among genes, and about 60% of human genes have CpG islands in the 5 'region of genes. Under physiological conditions, gene methylation occurs during X-chromosome inactivation and genomic imprinting, but in other situations, CpG islands are not methylated. However, in cancer, it has become clear that CpG islands are often methylated, and that methylation suppresses gene expression. The present inventors have so far reported that expression of p16 ^INK4A , E-cadherin, hMLH1 and 14-3-3 sigma is suppressed by methylation of the promoter region in cancer. In cancer, it is estimated that several percent of genes present in the human genome are inactivated by abnormal methylation. Therefore, genetic modification information such as methylation can be a good molecular marker for identifying novel tumor suppressor genes. In addition, since gene methylation can be detected from serum and feces, it is expected to be used as a tumor marker in early diagnosis of cancer, prediction of recurrence, and postoperative monitoring.

The present inventors have previously developed a method for efficiently amplifying only methylated CpG islands, Methylated CpG island amplification (MCA) method (Toyota et al. Cancer Res, 59: 2307, 1999). By performing subtraction using amplicons derived from normal tissue and cancer tissue amplified by this method, it became possible to identify gene stumps that are abnormally methylated only in cancer. The present inventors have identified gene stumps that are abnormally methylated in human colorectal cancer using the MCA method. One of these gene fragments, MINT25, was located on human chromosome 22q11 and was found to be highly methylated in gastric cancer (Toyota et al., Cancer Res. 59, 5438, 1999).
Toyota et al., Cancer Res. 59: 2307, 1999 Toyota et al., Cancer Res. 59: 5438, 1999

  The present inventors have identified a novel gene ACMG1 in the vicinity of the abnormally methylated gene fragment MINT25, and abnormal methylation of the CpG island of this gene occurs frequently in gastric cancer cell lines, and the expression of the ACMG1 gene is It was found that it was suppressed, and the present invention was completed.

  That is, the present invention provides a method for detecting gastric cancer by measuring the frequency of ACMG1 gene methylation in a sample collected from a subject. The frequency of methylation means the number or ratio of methylated cytosines out of cytosines of dinucleotide CpG present in a given region of DNA to be measured. Alternatively, the methylation frequency means the proportion of CpG at a specific position in a collection of DNA prepared from a specimen. As a sample collected from a subject, a tissue sample, blood, serum, or plasma can be used. Preferably, in the method of the present invention, the frequency of methylation in a region containing the promoter of the ACMG1 gene, the 5 'untranslated region, the first exon and the first intron is measured. Particularly preferably, in the method of the present invention, the frequency of methylation in the CpG island in the above-mentioned region of the ACMG1 gene is measured.

  In another aspect, the present invention provides a method for detecting gastric cancer by measuring the expression level of ACMG1 gene in a specimen collected from a subject. The expression level of ACMG1 gene refers to the amount of mRNA transcribed from ACMG1 gene or the amount of ACMG1 protein.

  In another aspect, the present invention provides an oligonucleotide having 6 to 50 bases in the base sequence represented by SEQ ID NO: 3 (FIG. 10) or a base sequence complementary thereto. The present invention also provides an oligonucleotide having 6-50 bases in the base sequence shown in SEQ ID NO: 5 (FIG. 12) or a base sequence complementary thereto, and a base shown in SEQ ID NO: 6 (FIG. 13). An oligonucleotide having 6 to 50 bases in a sequence or a complementary base sequence is provided. These oligonucleotides can be used as primers and probes for measuring the frequency of methylation of the ACMG1 gene. In a particularly preferred embodiment, the present invention provides an oligonucleotide having 6 to 50 bases in the base sequence shown in SEQ ID NO: 7-26 (FIGS. 14-16) or a base sequence complementary thereto.

In another preferred embodiment, the present invention provides primers and probes having the following sequences:
Forward primer: 5'-AGYGAGAGAGYGYGAAYGAGTT-3 '(SEQ ID NO: 27)
Reverse primer: 5'-RCCRACTCCRCCRTAAC-3 '(SEQ ID NO: 28)
Probe: 5'-TGGTTCGGCGTTCGT-3 '(SEQ ID NO: 29)
(Where Y = C or T, R = A or G)
In another preferred embodiment, the present invention provides an oligonucleotide having the base sequence shown by any of SEQ ID NOs: 30-39 (Table 1).

  The present invention further provides a kit for detecting gastric cancer comprising these oligonucleotides of the present invention.

  In another aspect, the present invention provides a method for selecting candidate substances for therapeutic drugs for gastric cancer. In one embodiment, the method of the present invention comprises culturing mammalian cells in the presence and absence of a test substance, measuring the frequency of methylation of the ACMG1 gene in each cell, and suppressing methylation. Each step of selecting a test substance having an effect as a candidate for a gastric cancer therapeutic drug. In another embodiment, the selection method of the present invention comprises culturing mammalian cells in the presence and absence of a test substance, measuring the expression level of ACMG1 gene in each cell, and expressing ACMG1 gene. Each step of selecting a test substance that has an effect of promoting gastric cancer as a candidate substance for a therapeutic agent for gastric cancer is included.

  In yet another aspect, the present invention provides a novel DNA encoding the amino acid sequence shown in SEQ ID NO: 2 (FIG. 9). Preferably, the DNA of the present invention has the base sequence shown by SEQ ID NO: 1 (FIG. 8). In another aspect, the present invention provides a protein having the amino acid sequence represented by SEQ ID NO: 2 or a fragment thereof. ACMG1 gene and ACMG1 protein according to the present invention inhibits the growth of cancer cells because the growth of cells is suppressed when ACMG1 gene is introduced into and expressed in cells with ACMG1 gene methylation. It is thought that it can act as an agent.

  In still another embodiment, the present invention provides an oligonucleotide having 6-50 bases in the base sequence represented by SEQ ID NO: 1 or a base sequence complementary thereto. Such oligonucleotides are useful as probes and primers for amplifying and / or detecting the ACMG1 gene or its transcript.

  The present invention is useful for diagnosis of gastric cancer and screening of candidate substances for gastric cancer therapeutic agents.

The present inventors have identified a novel gene in the vicinity of the gene fragment MINT25 present in human chromosome 22q11. The gene from the share exons and 3 'ends CABIN1, this gene was named A lternative transcript of C ABIN1 M ethylated in G astric Cancer 1 (ACMG1). FIG. 10 (SEQ ID NO: 3) shows a sequence about 1.2 kb upstream from the transcription start site of the ACMG1 gene, 5 ′ untranslated region, first exon, and part of the first intron, and FIG. 11 shows the sequence of its complementary strand. Shown in (SEQ ID NO: 4). Further, FIG. 8 (SEQ ID NO: 1) shows the ACMG1 cDNA sequence, and FIG. 9 (SEQ ID NO: 2) shows the amino acid sequence of the polypeptide encoded by this sequence. CABIN1, which shares the gene C-terminus with ACMG1, has been identified as a protein that suppresses the activation of calcineurin, a protein phosphatase (Immunity, 8: 703, 1998). ACMG1 has a domain that can interact with calcineurin, but does not have a nuclear translocation signal, and is considered to be a gene with a physiological function different from CABIN1. To date, there has been no report of an isoform sharing the gene C-terminus with CABIN1.

  Although it has been reported that gene mutations such as APC, K-ras, and p53 are involved in the development of gastric cancer, the frequency is low and there are many unclear points regarding the mechanism of gastric cancer development. In the present invention, it was revealed that ACMG1 is inactivated by abnormal methylation in gastric cancer. Analysis of the gene function of ACMG1 may lead to the elucidation of the molecular mechanism of gastric cancer development. Aberrant methylation of the ACMG1 gene is specific for gastric cancer and is less likely to occur in normal tissues and other tumors, suggesting that it is useful as a tumor marker. It can be applied to genetic diagnosis using serum by using abnormal methylation of ACMG1 gene as an index, and is expected to be applied to early diagnosis of gastric cancer and monitoring of recurrence after surgery.

  The method of the present invention for detecting gastric cancer by measuring methylation of the ACMG1 gene can be performed as follows. First, a sample containing DNA is collected from a subject and DNA is prepared. As the specimen, tissue samples, blood, serum or plasma, urine, saliva, ascites, sputum, stool, pancreatic juice, bile fluid, and the like can be used. Tissue samples include tissue or part of it removed by surgery or endoscopy, biopsy specimens, organ washings, and the like. Preferably, the subject's serum or organ washing solution is used as the specimen. Methods for extracting DNA from a sample are well known in the art, and for example, extraction with phenol / chloroform can be used. Alternatively, a commercially available DNA extraction reagent may be used.

  In order to measure the degree and frequency of DNA methylation, COBRA (Combined bisulfite restriction analysis) method, bisulfite sequencing method or the like can be used. In these methods, when DNA is treated with an unmethylated cytosine modifying reagent such as bisulfite, unmethylated cytosine is converted to uracil, but methylated cytosine is not converted to uracil. based on. A technique for analyzing DNA methylation with high sensitivity and semi-quantitativeity is the Methylight method. In the Methylight method, a methylation is detected by a real-time PCR method using a TaqMan Probe in combination with a primer that recognizes a methylation-specific sequence.

  When the double-stranded DNA having the sequence of ACMG1 gene represented by SEQ ID NO: 3 is treated with bisulfite, the sequence shown in FIG. 12 (SEQ ID NO: 5) is obtained from the top strand, and the sequence shown in FIG. The sequence shown in 6) is obtained. In these sequences, the base represented by Y is C when CpG cytosine is methylated, and T when CpG cytosine is not methylated. When cytosine is treated with bisulfite, uracil is produced. In the present specification and drawings, this uracil is represented as T (thymine). This is because, in PCR amplification of DNA, the position of uracil is synthesized as thymine, and uracil can be treated the same as thymine as far as the explanation of complementarity of oligonucleotide sequences in PCR amplification and hybridization is concerned. It is.

  In the COBRA method, the DNA extracted as described above is treated with bisulfite. As the treatment conditions, for example, 2 μg of DNA can be treated with 2N sodium hydroxide at 37 ° C. for 10 minutes and then treated with 3M bisulfite at 50 ° C. for 16 hours. Thereafter, the bisulfite is removed using a DNA purification column or the like, ethanol precipitation is performed, the DNA is dissolved in purified water, and stored at −80 ° C. until use. Next, PCR is performed using a primer set designed to amplify the target region in the ACMG1 gene using DNA treated with bisulfite as a template.

  The target region is a region where methylation is to be detected. Preferably, the target region is a region containing the promoter of the ACMG1 gene, the 5 'untranslated region, the first exon and the first intron, and more preferably a CpG island in this region. In a particularly preferred embodiment of the present invention, the regions where methylation is to be detected are GM2 to GM6 shown in FIG. 1B. The base sequence of GM2 to GM6 and the base sequence obtained when double-stranded DNA having this sequence is treated with bisulfite are shown in FIGS.

  Primers used for PCR can be easily designed based on the sequence of ACMG1 gene shown in SEQ ID NO: 3-6 (FIGS. 10-13) and the sequence after bisulfite treatment. Primers may be designed not to contain potentially methylated cytosine, or for potentially methylated cytosine to hybridize to either cytosine or uracil. It may be designed as a mixture of suitable sequences. Examples of preferable primers include a set of primers having the sequences of SEQ ID NOs: 30-39 (Table 1). Oligonucleotides can be chemically synthesized using, for example, a general-purpose DNA synthesizer (for example, Model 394 manufactured by Applied Biosystems). Oligonucleotides may be synthesized using any other method well known in the art.

  Next, the PCR amplification product is cleaved with a restriction enzyme. The restriction enzyme is selected to cleave only when the target CpG cytosine is methylated (cytosine) or only when it is not methylated (uracil / thymine). Analysis of the cleavage product by electrophoresis can determine whether the target is methylated and / or the frequency of methylation.

  As an example, a procedure for detecting methylation in the GM2 region will be described. The base sequence of the GM2 region is shown as SEQ ID NO: 7 (top strand) and SEQ ID NO: 9 (bottom strand) in FIG. 14, and the position of CpG is underlined. When the DNA purified as described above is treated with bisulfite, the top strand is converted into the sequence represented by SEQ ID NO: 8, and the bottom strand is converted into the sequence represented by SEQ ID NO: 10, respectively. In these sequences, the base represented by Y is C when CpG cytosine is methylated, and T (U) when it is not methylated. In any of these cases, the GM2 region can be PCR amplified using primers designed to hybridize. For example, a primer set having the sequences shown in SEQ ID NOs: 30 and 31 in Table 1 can be used. Next, the PCR amplification product is cleaved with the restriction enzyme NruI. NruI recognizes the sequence 5′-TCGCGA-3 ′ and cleaves double-stranded DNA. Therefore, cleavage occurs when cytosine of CpG is methylated, and cleavage occurs when it is not methylated. Does not occur. Therefore, by analyzing the cleavage product by electrophoresis, it is possible to determine whether it is methylated or the ratio of methylation.

  In the bisulfite sequencing method, the region containing the target is PCR amplified in the same manner as the COBRA method described above, and then the base sequence of the PCR amplification product is directly sequenced and analyzed.

Alternatively, methylation can also be detected by combining bisulfite treatment and hybridization. That is, a methylation-specific probe and / or an unmethylation-specific probe designed to PCR-amplify a region containing a target in the same manner as in the COBRA method described above and then to be complementary to the base sequence in the target region And the PCR amplification product are hybridized. A methylation specific probe is a probe that hybridizes only when the specific CpG cytosine to be examined is methylated (cytosine), and an unmethylated specific probe is a target CpG cytosine. This probe hybridizes only when it is not methylated (uracil / thymine). Whether or not the CpG cytosine to be examined is methylated, or the frequency of methylation, by examining whether hybridization occurs between the PCR amplification products and these probes, or by determining the rate of hybridization. Can be determined. Preferred examples of primers and probes that can be used in such a method are as follows.
Forward primer: 5'-AGYGAGAGAGYGYGAAYGAGTT-3 '(SEQ ID NO: 27)
Reverse primer: 5'-RCCRACTCCRCCRTAAC-3 '(SEQ ID NO: 28)
Probe: 5'-TGGTTCGGCGTTCGT-3 '(SEQ ID NO: 29)
(Where Y = C or T, R = A or G)

  As another method for measuring the frequency of DNA methylation, methylation-specific PCR can be used. In this method, after the DNA extracted as described above is treated with bisulfite, PCR using a methylation specific primer and PCR using an unmethylation specific primer are performed separately using this DNA as a template. A methylation-specific primer is a primer designed to hybridize only when cytosine of the target CpG in the ACMG1 gene is methylated (cytosine), and an unmethylated-specific primer is Primer designed to hybridize only when the target CpG cytosine is not methylated (uracil / thymine). Such primers can be easily designed by appropriately selecting the base indicated by Y or the corresponding base in the complementary strand based on the sequences shown in SEQ ID NOs: 5 and 6. When PCR is performed using these primers, when the target CpG cytosine in the ACMG1 gene is methylated, a PCR amplification product is obtained from the methylation-specific primer, and when it is not methylated A PCR amplification product is obtained from unmethylated specific primers. By measuring and comparing the amounts of these amplification products by means such as gel electrophoresis, the frequency of DNA methylation in the target region in the ACMG1 gene can be determined.

  Using various methods as described above, measure the frequency of methylation of ACMG1 gene contained in the specimen, and then compare this with the frequency of methylation of ACMG1 gene contained in specimens derived from healthy individuals Thus, the presence of gastric cancer cells in the specimen can be detected.

  In another embodiment of the present invention, the method for detecting gastric cancer by measuring the expression level of the ACMG1 gene can be performed as follows. When measuring the amount of mRNA transcribed from the ACMG1 gene as the expression level of the ACMG1 gene, RNA is extracted from the specimen, and the ACMG1 mRNA is quantified using a method such as RT-PCR or Northern blotting. Methods for extracting RNA from a sample are well known in the art, and for example, guanidine-isothiocyanate and phenol / chloroform extraction can be used. When measuring the amount of ACMG1 protein as the expression level of ACMG1 gene, it can be measured by a method such as Western blotting, ELISA, immunoprecipitation using a specific antibody against ACMG1 protein. A specific antibody against the ACMG1 protein can be easily obtained by immunizing a mammal with a polypeptide having the amino acid sequence represented by SEQ ID NO: 2 or a partial sequence thereof by a method well known in the art. .

  In another aspect, the present invention provides a method for selecting candidate substances for therapeutic drugs for gastric cancer. As shown in the Examples described later, in the present invention, cell growth is suppressed by colony formation assay by introducing ACMG1 gene externally into a gastric cancer cell line in which ACMG1 expression is suppressed by methylation. It was found that ACMG1 gene expression was restored in the presence of methylation inhibitors and histone deacetylation inhibitors. In other words, ACMG1 was suggested to have a function as a tumor suppressor gene. Therefore, substances that can suppress the methylation of ACMG1 gene and promote its expression, or substances that can promote the expression of ACMG1 gene by any other method are expected to be useful for the prevention and treatment of gastric cancer. Is done.

  In one embodiment, the selection method of the present invention comprises culturing a mammalian cell in the presence and absence of a test substance, measuring the frequency of methylation of the ACMG1 gene in each cell, and measuring the methylation. Each step of selecting a test substance having an inhibitory effect as a candidate substance for a therapeutic agent for gastric cancer is included. In another embodiment, the selection method of the present invention comprises culturing mammalian cells in the presence and absence of a test substance, measuring the expression level of ACMG1 gene in each cell, and expressing ACMG1 gene. Each step of selecting a test substance that has an effect of promoting gastric cancer as a candidate substance for a therapeutic agent for gastric cancer is included. As a mammalian cell used in the method of the present invention, a cell isolated from a mammal-derived cancer tissue or a primary culture thereof may be used, or an established cell line may be used.

  ACMG1 gene and ACMG1 protein according to the present invention inhibits the growth of cancer cells because the growth of cells is suppressed when ACMG1 gene is introduced into and expressed in cells with ACMG1 gene methylation. It is considered useful as an agent. Accordingly, in still another aspect of the present invention, a novel DNA encoding the amino acid sequence represented by SEQ ID NO: 2 and a protein having the amino acid sequence represented by SEQ ID NO: 2 or a fragment thereof are provided. These DNAs, proteins or fragments thereof are considered useful for the prevention and treatment of gastric cancer.

  EXAMPLES The present invention will be described below in more detail with reference to examples, but the present invention is not limited to these examples.

Identification of novel gene ACMG1 The present inventors previously detected a DNA fragment abnormally methylated in colorectal cancer by the MCA method. One of them, MINT25, has been reported to be highly methylated in gastric cancer. CABIN1, which is an endogenous inhibitor of calcineurin in T lymphocytes, and one EST were identified by homology search using the BLAST program near MINT25. Using GENSCAN, we identified a new gene at this EST site. This gene than to share exons 3 'end and CABIN1, this gene was named A lternative transcript of C ABIN1 M ethylated in G astric Cancer 1 (ACMG1). FIG. 1A shows a partial structure of the ACMG1 gene. The gene sequence is shown in FIG. 10 (SEQ ID NO: 3). Vertical lines indicate exons. The upper row shows Cabin1, and the lower row shows ACMG1. Among these two genes, ACMG1 was the gene that had MINT25 at the 5 'end and could affect transcriptional activity. Therefore, it was suggested that the expression of ACMG1 is suppressed by methylation in gastric cancer. FIG. 1B shows a schematic diagram of a CpG island of ACMG1. MINT25 is located on Exon 1 of ACMG1. Vertical lines indicate CpG sites, and arrows indicate transcription start sites.

Detection of abnormal methylation of ACMG1 in various cell lines (Gastric cancer cell lines: SNU638, HSC39, HSC40, HSC41, HSC42, HSC43, HSC44, HSC45, MKN7, MKN28, MKN45, MKN74, JRST, KatoIII, NUGC3, colon DNA was extracted from cancer cell lines: Caco2, RKO, SW48, HCT116, DLD-1, LoVo, HT29, SW480), and the methylation of CpG islands existing upstream of ACMG1 was examined using the COBRA method. Six primers from GM2 to GM6 were set on the CpG island and examined. The position of each primer is shown in FIG. 1B, and each sequence is shown in the following table.

Bisulfite-PCR was used for methylation analysis. Briefly, DNA was treated with sodium bisulfite, PCR was performed, the PCR amplification product was digested with a restriction enzyme that cleaves only the methylated CpG site, and then electrophoresed on agarose. The ratio of methylated alleles was quantified with an image analyzer (Atto, Lane and Spot Analyzer).

  FIG. 2A shows the result of methylation detection. The upper row shows gastric cancer cell lines, and the lower row shows% methylation. M represents an allele that is methylated. The primers used are shown on the left. Abnormal methylation of ACMG1 was classified into two types: cell lines confined to GM2, GM4, and GM5 (14 out of 17 cases, 82%) and cell lines ranging from GM2 to GM6 (FIG. 2A).

  Next, in order to investigate the methylation status of individual CpG of ACMG1, the methylation of individual CpG sites was analyzed by bisulfite sequencing for gastric cancer cell lines MKN7 and MKN28 in which the GM2 region was methylated. did. A part of the result is shown in FIG. 2B. M represents a methylated CpG site. The region amplified by PCR contained 22 CpG sites. In the gastric cancer cell lines MKN7 and MKN28, which had been methylated by the COBRA method, high-density methylation was observed throughout the analyzed region. These results suggest that when GM2 is methylated, it is methylated over a wide range of CpG islands.

  In the colon cancer cell line and the hematopoietic tumor cell line, the methylation state in GM2 including the transcription start point was detected by the COBRA method. In 1 of 8 cases (13%) of colorectal cancer and 2 of 12 cases (17%) of hematopoietic tumors, methylation frequency was considerably lower than that of gastric cancer (Fig. 2C). This indicates that ACMG1 is frequently methylated in gastric cancer.

Measurement of ACMG1 expression
To investigate whether abnormal methylation of ACMG1 correlates with suppression of gene expression, RT-PCR was used to analyze ACMG1 expression in cell lines in which GM2 was not methylated and in methylated cell lines. RNA was extracted from tumor cell lines and gastric cancer tissues collected from patients, and cDNA was prepared using reverse transcriptase. Using the obtained cDNA, gene expression was analyzed by RT-PCR and real-time PCR. As a negative control, a PCR reaction was simultaneously performed with a sample (RT-) prepared without performing a reverse transcription reaction. The results are shown in FIG. 3A. The ACMG1 gene was expressed in cell lines (MKN74, NUGC4, HSC41, LoVo, HT29, SW480) in which GM2 methylation was not observed by the COBRA method. In cell lines in which methylation was observed in GM2 (JRST, HSC39, HSC40, HSC42, HSC43, HSC44, HSC45, MKN28), the expression of ACMG1 was suppressed. The cell line KatoIII, which shows relatively weak methylation, showed decreased expression. These results suggested that the expression of GM2 containing the transcription start site of ACMG1 is suppressed when methylated.

  For each gastric cancer cell line, the correlation between methylation pattern and ACMG1 gene expression was examined. Gastric cancer cell lines could be divided into two groups according to the methylation pattern in each region of GM2 to GM6 of ACMG1. One shows the methylation of GM3 and GM6 at the edge of the CpG island, but GM2, GM4 and GM5 in the center of the CpG island are not methylated (group1), and the other is that the whole region analyzed is It is methylated (group2) (FIG. 3B). The average value of% methylation of each primer setting site in each group is shown in a circle. Each primer setting site is shown in the upper row. The expression state of ACMG1 is shown on the right. Group1 showed ACMG1 expression, and group2 did not show ACMG1 expression. This suggests that ACMG1 expression involves GM2, GM4, GM5, or methylation at the center of the CpG island.

Reexpression of ACMG1 gene in tumor cells treated with DNMT inhibitor and HDAC inhibitor
To further investigate the role of DNA methylation in the suppression of ACMG1 gene expression, gastric cancer cell lines HSC44 and HSC45, which have lost expression due to methylation, were converted to 5-aza-, a DNA methylation (DNMT) inhibitor. Treatment with dC and TSA, a histone deacetylation (HDAC) inhibitor, examined whether reexpression of the gene was observed. The results are shown in FIG. The left shows HSC44 and the right shows HSC45. The vertical axis of FIG. 4B shows the value obtained by correcting the amount of acetylated histone by input (endogenous control). Reexpression and 5-SA-dC 0.2 μM showed almost no re-expression and TSA treatment, but when 5-aza-dC 0.2 μM and TSA were combined, strong re-expression was observed. In addition, strong re-expression was observed when treated with 5-aza-dC 2.0 μM. From the above, it was suggested that the expression of ACMG1 was suppressed by DNA methylation (FIG. 4).

Examination of methylation of ACMG1 in clinical cases DNA was prepared from clinical tissue samples of gastric cancer and other cancers, and bismuth-PCR was used to detect the GM2 region most likely to correlate with loss of ACMG1 expression. Methylation was examined. As a result, 52 out of 77 cases (68%) of gastric cancer beds, 5 out of 50 cases of colon cancer (10%), 1 out of 10 cases of acute lymphocytic leukemia (10%), 12 out of ovarian cancer cases Among them, 1 case (8%) and 2 out of 24 head and neck cancer cases (8%) had methylation, and ACMG1 methylation was observed in gastric cancer at a higher rate than other tumors. In addition, no methylation was observed in the normal gastric mucosa around the stomach cancer. An example of detection is shown in FIG. 5A. T indicates gastric cancer tissue, and N indicates normal gastric mucosa around the cancer. M represents an allele that is methylated.

  The cell lines were 15/19 (79%) for gastric cancer, 1/8 (13%) for colorectal cancer, 2/12 (17%) for hematopoietic hematopoietic tumor, and 1% for gastric cancer. A significant difference was observed. In primary tissue culture, 52/77 (68%) for gastric cancer, 5/50 (10%) for colorectal cancer, ALL 1/10 (10%), 1/12 (8%) for ovarian cancer, head and neck cancer In 2/24 (8%), the significance level was 1%. Thus, ACMG1 was methylated at a high rate only in gastric cancer in both cell lines and primary tissue cultures, suggesting its involvement with gastric cancer.

  Next, we analyzed the methylation of individual CpG sites around the transcription start site in clinical cases of gastric cancer by bisulfite sequencing. We examined methylation in clinical cases 6T, 8T and surrounding normal mucosa 6N, 8N in which GM2 region was methylated. The result is shown in FIG. 5B. There were 10 CpG site methylations in the sequenced region. M represents a methylated CpG site, and U represents an unmethylated CpG site. In other words, it was shown that methylation was observed at all CpG sites analyzed in cases where methylation was observed by the COBRA method. On the other hand, in the adjacent non-cancerous tissue, all CpG sites were not methylated.

  Based on the above results, methylation of ACMG1 is more frequent in gastric cancer than tumors derived from other tissues, and it is methylated specifically in cancer because no normal case is methylated in normal gastric mucosa around gastric cancer. It was suggested to be a gene. Thus, methylation of ACMG1 may be useful as a tumor marker.

Analysis of methylation in xenografts Human gastric cancer surgical material was transplanted into nude mice to prepare xenografts. Approximately 2 weeks after transplantation, the grafts were collected and extracted for DNA and RNA. In 15 cases of gastric cancer xenografts, methylation of ACMG1 in GM2 was examined using the COBRA method (FIG. 6A). M represents an allele that is methylated. The bottom line shows% methylation. Methylation was observed at a high rate of 7 out of 15 cases (47%). In addition, the expression level of ACMG1 was quantitatively analyzed using real-time PCR. The result is shown in FIG. 6B.

  ACMG1 was expressed in the sample without methylation, and ACMG1 was not expressed in the sample with methylation. In addition, those that were partially methylated showed weak expression. Therefore, the ratio of methylation and the expression level were considered to be related.

Inhibition of growth of gastric cancer cell lines by ACMG1 gene
In order to investigate cell growth inhibition by ACMG1 gene, ACMG1 gene was externally introduced into gastric cancer cell line and expressed, and the effect was examined. As an ACMG vector, pcDNA3.1-ACMG1 was used which was introduced into the pcDNA3.1 vector (invitrogen), which was regulated by the cytomegalovirus promoter and had a gene resistant to neomycin (G418). This vector was transformed into cell lines HSC44 and HSC45, in which ACMG1 is thought to be suppressed by methylation. As a control, a transformation of only pcDNA3.1 vector was used. Six hours after transformation, the cells were diluted 8-fold, and after 24 hours, they were cultured for 14 days in a medium containing 400 μg / ml G418. Thereafter, the cells were fixed with methanol, stained with 0.25% crystal violet, and the number of colonies was quantified with NIH Image software.

  When ACMG1 cDNA was introduced, the growth of colonies was suppressed as compared to the control transformed with pcDNA3.1 alone (FIG. 7A). When the control was 100, the relative number of colonies was 30% for HSC44 and 35% for HSC45, and the growth was significantly suppressed (FIG. 7B). From the above, it was suggested that ACMG1 has the ability to suppress cell growth.

Detection of abnormal methylation of ACMG1 gene using serum
Abnormal methylation of ACMG1 was detected using serum collected from patients with gastric cancer, because abnormalities of ACMG1 were frequent in gastric cancer and abnormal methylation was tumor-specific. Serum abnormal methylation of ACMG1 was observed in 15 of 20 cases of gastric cancer in which ACMG1 was abnormally methylated, indicating a high detection rate. On the other hand, 12 cases of gastric cancer in which ACMG1 was not abnormally methylated did not show any methylation and showed high specificity. From these results, it was shown that gastric cancer can be diagnosed using serum by measuring methylation of ACMG1.

Detection of abnormal methylation of ACMG1 gene using gastric lavage fluid
In order to detect abnormal methylation of ACMG1 gene from gastric lavage fluid, DNA was prepared from gastric lavage fluid, treated with bisulfite, and methylation was detected by methylight method. Abnormal methylation of ACMG1 was observed in 12 cases of gastric cancer tissues obtained from 16 cases of gastric cancer, of which 9 cases were also methylated in gastric lavage fluid (sensitivity 75%). In 4 cases in which no methylation was observed in gastric cancer tissues, no methylation was observed in gastric lavage fluid (specificity 100%). These results suggest that detection of methylation of ACMG1 from gastric lavage fluid is useful for diagnosis of residual gastric cancer after gastrectomy and prediction of recurrence.

  The present invention is useful for diagnosis of gastric cancer and screening of candidate substances for gastric cancer therapeutic agents.

FIG. 1 shows the structure of ACMG1 gene and CpG island. FIG. 2 shows detection of abnormal methylation of ACMG1 in various cell lines. FIG. 3 shows the measurement results of ACMG1 expression. FIG. 4 shows ACMG1 gene reexpression in tumor cells treated with DNMT inhibitor and HDAC inhibitor. FIG. 5 shows detection of methylation of ACMG1 in clinical cases. FIG. 6 shows the results of analysis of methylation in xenografts. FIG. 7 shows the growth suppression of gastric cancer cell lines by ACMG1 gene. FIG. 8 shows the base sequence of ACMG1 cDNA. FIG. 9 shows the amino acid sequence of the polypeptide encoded by ACMG1 cDNA. FIG. 10 shows a partial base sequence of the ACMG1 gene. FIG. 11 shows a sequence complementary to the base sequence shown in FIG. FIG. 12 shows the sequence of DNA obtained when the DNA having the sequence shown in FIG. 10 is treated with bisulfite. FIG. 13 shows the sequence of DNA obtained when the DNA having the sequence shown in FIG. 11 is treated with bisulfite. FIG. 14 shows the nucleotide sequences of GM2 and GM3, and the DNA sequences obtained when DNA having these sequences is treated with bisulfite. FIG. 15 shows the base sequences of GM4 and GM5, and the DNA sequence obtained when DNA having these sequences is treated with bisulfite. FIG. 16 shows the base sequence of GM6 and the DNA sequence obtained when DNA having these sequences is treated with bisulfite.

Claims (17)

A method of detecting gastric cancer by measuring the frequency of methylation of the ACMG1 gene encoding a protein having the amino acid sequence represented by SEQ ID NO: 2 in a sample collected from a subject. A method for detecting gastric cancer by measuring the expression level of an ACMG1 gene encoding a protein having the amino acid sequence represented by SEQ ID NO: 2 in a sample collected from a subject. The method according to claim 1 or 2, wherein the specimen collected from the subject is a tissue sample, blood, serum or plasma. An oligonucleotide composed of 20-50 bases in a base sequence represented by base 1-1551 in SEQ ID NO: 3 or a base sequence complementary thereto. An oligonucleotide composed of 20-50 bases in a base sequence represented by base 1-1551 in SEQ ID NO: 5 or a base sequence complementary thereto. An oligonucleotide composed of 20-50 bases in a base sequence represented by bases 1450-3000 in SEQ ID NO: 6 or a base sequence complementary thereto. An oligonucleotide composed of 20-50 bases in a base sequence represented by any of SEQ ID NOs: 7-18 or 23-26 or a base sequence complementary thereto. An oligonucleotide comprising the base sequence represented by any of SEQ ID NOs: 27-29. An oligonucleotide comprising the nucleotide sequence represented by any of SEQ ID NOs: 30-35, 38 or 39. A kit for detecting gastric cancer, comprising the oligonucleotide according to any one of claims 4-9. A method for selecting a candidate substance for a therapeutic agent for gastric cancer, comprising culturing mammalian cells in the presence and absence of a test substance and encoding a protein having the amino acid sequence shown in SEQ ID NO: 2 in each cell. A method comprising the steps of measuring the frequency of gene methylation and selecting a test substance having an effect of inhibiting methylation as a candidate substance for a therapeutic agent for gastric cancer. A method for selecting a candidate substance for a therapeutic agent for gastric cancer, comprising culturing mammalian cells in the presence and absence of a test substance and encoding a protein having the amino acid sequence shown in SEQ ID NO: 2 in each cell. A method comprising the steps of measuring the expression level of a gene and selecting a test substance having an effect of promoting the expression of the ACMG1 gene as a candidate substance for a therapeutic agent for gastric cancer. A DNA encoding the amino acid sequence represented by SEQ ID NO: 2. The DNA according to claim 13, which has the base sequence represented by SEQ ID NO: 1. A cancer cell growth inhibitor comprising the DNA according to claim 13 or 14. An oligonucleotide composed of 20-50 bases in a base sequence represented by SEQ ID NO: 1 or a base sequence complementary thereto. A protein having the amino acid sequence represented by SEQ ID NO: 2.