Tumor suppressive role of a 2.4 Mb 9q33–q34 critical region and DEC1 in esophageal squamous cell carcinoma

Article metrics

Abstract

The key genes involved in the development of esophageal squamous cell carcinoma (ESCC) remain to be elucidated. Previous studies indicate extensive genomic alterations occur on chromosome 9 in ESCC. Using a monochromosome transfer approach, this study provides functional evidence and narrows down the critical region (CR) responsible for chromosome 9 tumor suppressing activity to a 2.4 Mb region mapping to 9q33–q34 between markers D9S1798 and D9S61. Interestingly, a high prevalence of allelic loss in this CR is also observed in primary ESCC tumors by microsatellite typing. Allelic loss is found in 30/34 (88%) tumors and the loss of heterozygosity (LOH) frequency ranges from 67 to 86%. Absent to low expression of a 9q32 candidate tumor suppressor gene (TSG), DEC1 (deleted in esophageal cancer 1), is detected in four Asian ESCC cell lines. Stably expressing DEC1 transfectants provide functional evidence for inhibition of tumor growth in nude mice and DEC1 expression is decreased in tumor segregants arising after long-term selection in vivo. There is 74% LOH in the DEC1 region of ESCC primary tumors. This study provides the first functional evidence for the presence of critical tumor suppressive regions on 9q33–q34. DEC1 is a candidate TSG that may be involved in ESCC development.

Introduction

Esophageal squamous cell carcinoma (ESCC) is a very deadly disease and key genes involved in its development still remain to be elucidated. Previous studies show frequent allele losses on chromosomes 9 and 18 by both loss of heterozygosity (LOH) (Aoki et al., 1994; Mori et al., 1994; Shibagaki et al., 1994; Wang et al., 1996; Muzeau et al., 1997; Naidoo et al., 1999; Karkera et al., 2000; Yang et al., 2004) and comparative genomic hybridization (CGH) studies (Pack et al., 1999; Tada et al., 2000; Yen et al., 2001, 2003). These results suggest that tumor suppressor genes (TSGs) associated with esophageal tumorigenesis may be located on these chromosomes.

DEC1, deleted in esophageal cancer 1, is a candidate TSG characterized by a high LOH frequency and reduced expression in tumors and the ability to suppress cancer cell growth in vitro (Nishiwaki et al., 2000). There is some controversy with respect to the chromosomal location of DEC1: the gene maps to 9q33.1, according to the Ensembl and UCSC databases, and 9q32, according to the NCBI database. No genetic alterations in the gene have been detected and its function is still unknown.

Somatic cell genetics offers a functional approach to map chromosome regions associated with tumor suppressive activities. The microcell-mediated chromosome transfer (MMCT) method allows transfer of exogenous genetic materials in a physiological state closely resembling its endogenous expression (Saxon et al., 1985; Murakami, 2002). To date, no MMCT studies with ESCC cell lines have been published. The objective of this present study is to identify whether tumor suppressive activities of chromosomes 9 and 18 contribute to ESCC and to investigate a candidate TSG DEC1 in an ESCC cell line by in vivo studies.

Results

Transfer of human chromosomes 9 and 18 into ESCC cell lines

The Neo. 9 donor cell line contains a human chromosome 9 with a microdeletion at 9p21 and is null for the p16INK4a gene. Another donor cell line, microcell hybrid (MCH)912.1, contains an intact human chromosome 18. Both donor chromosomes were introduced into the ESCC cell line SLMT-1S1, which is highly tumorigenic in nude mice. G418-resistant MCHs for chromosome 9, MCH9.1, MCH9.3, and MCH9.6, and for chromosome 18, MCH18.3, MCH18.8, and MCH18.10, were obtained. Slot blot hybridization of MCHs detected no mouse DNA contamination (data not shown).

Microsatellite typing and fluorescence in situ hybridization (FISH) analysis

Polymerase chain reaction (PCR) microsatellite typing with 39 chromosome 9 and 10 chromosome 18 specific highly polymorphic microsatellite markers verifies the successful transfer of the donor chromosomes in the MCHs. For chromosome 9, eight markers (D9S324, D9S168, D9S269, D9S254, D9S157, D9S171, D9S126, and D9S970) show a microdeletion in the donor cells, which spans a 19.6 Mb region around 9p21, where a well-known TSG p16INK4a maps. For chromosome 18, all markers are present in the donor cells. Information for chromosome 9q markers is summarized in Table 1 .

Table 1 Chromosome 9q-specific microsatellite markers

At the cytogenetic level, whole chromosomes 9 and 18 painting was used for verification of the extra copy of chromosomes 9 and 18 transferred into the MCHs. The SLMT-1S1 recipient cell line contains eight copies of chromosome 9 and three copies of chromosome 18. An additional chromosome 9 is observed in 94, 95, and 96% of the MCH9.1, MCH9.3, and MCH9.6 hybrid metaphase spreads, respectively. However, for MCH9.3 fewer than 20% of the metaphases show transfer of the complete donor chromosome 9, while the remaining metaphases show transfer of only a partial donor chromosome. This result is also verified by molecular allelotyping. An additional chromosome 18 is observed in 95, 100, and 100% of the MCH18.3, MCH18.8, and MCH18.10 hybrid metaphase spreads, respectively (data not shown).

Tumorigenicity assay of MCH cell lines

Figure 1 shows the kinetics of tumor formation in athymic nude mice for the recipient ESCC cell line SLMT-1S1 and the MCHs. Significant differences in the tumor growth kinetics are observed for chromosome 9 MCH cell lines, with reduced tumor sizes and much longer latency periods for MCH9.1 and MCH9.6, compared with the recipient SLMT-1S1 (P=0.029) (Figure 1a). In contrast, there is no obvious tumor suppression observed with the three chromosome 18 hybrids (Figure 1b). Table 2 summarizes this data.

Figure 1
figure1

(a) Tumor growth kinetics of chromosome 9 MCHs; (b) tumor growth kinetics of chromosome 18 MCHs. SLMT-1-S1 and MCH9.1, MCH9.3, MCH9.6, MCH18.3, MCH18.8, and MCH18.10 cell lines were tested in nude mice. The curves represent an average tumor volume of all sites inoculated for each cell line

Table 2 Tumorigenicity assay of recipient SLMT-1S1, chromosome 9 MCHs, chromosome 18 MCHs, pcDNA3.1 (+) vector-alone transfectants, and DEC1 stable transfectants

Chromosome 9 MCH tumor segregant (TS) cell line analysis

Four TS cell lines were established from each chromosome 9 hybrid and are designated MCH9.1-1-4TS, MCH9.3-1-4TS, and MCH9.6-1-4TS. Detailed microsatellite typing of MCHs and their TSs was performed with 20 microsatellite markers (Figure 2). Both MCH9.1 and MCH9.6 TSs were excised from mice after 3 months growth in vivo. The MCH9.1 tumors grow more aggressively and are on average over 2000 mm3 upon excision, while the MCH9.6 tumors are on average 1275 mm3. The MCH9.3 tumors grow the most aggressively and their TSs were excised from mice after 2 months growth, with an average size of 1075 mm3. All 12 TSs suffer loss of the D9S112 marker, identifying a 2.4 Mb critical region (CR) between markers D9S1798 and D9S61. Representative microsatellite typing results of the D9S112 marker mapping in the CR for the three MCH cell lines and their corresponding TSs are shown in Figure 3a.

Figure 2
figure2

Microsatellite analyses of hybrid cell lines (MCH9.1, 9.3, 9.6) and their TSs. The presence () and absence (•) of the marker, and absence from hybrid cell (X) in TSs are as indicated. Nonrandom losses are found at one 2.4 Mb CR around D9S112. DEC1 maps to the region between KM9.1 and D9S177. The LOH frequencies of six microsatellite markers in the 9q32–q34 region in 34 primary ESCC tissues are indicated on the right

Figure 3
figure3

(a) Representative results of PCR microsatellite typing for hybrid cell lines and TSs. The recipient, donor, hybrids, and TSs are analysed with the D9S112 marker. The combined peak patterns of the MCHs show successful transfer of chromosome 9. The comparative intensity of the donor allele in MCH9.3 is lower than that present in MCH9.1 and MCH9.6. Nonrandom loss of the donor allele from the hybrid cell lines is observed in their corresponding TSs. The arrows designate donor allele transfer () and loss (); (b) representative LOH analysis of marker D9S112 with patient sample 31. The arrow () indicates allelic deletion in the tumor tissue. N: normal; T: tumor

Deletion of 9q32–q34 region in ESCC primary tumors

DNAs from 34 ESCC tumors and the corresponding normal tissue were analysed using five 9q32–q34 microsatellite markers localized to the 4.7 Mb interval between D9S778 and D9S752; 30 tumors (88%) show allelic loss. In total, 19 tumors (56%) lost an allele for all of the informative markers, indicating possible extensive deletion of the entire 9q32–q34 region flanked by these five markers. Figure 2 shows the frequency of allelic loss of these markers. Representative results of the microsatellite typing for one pair of tissues are shown in Figure 3b.

DEC1 gene expression status of ESCC cell lines

DEC1 expression was examined by reverse transcription–PCR (RT–PCR) in four Asian ESCC cell lines, HKESC-2, T.Tn6, SLMT-1S1, and 81T. Decreased expression level of a 156 bp DEC1 PCR fragment is observed compared to a normal esophageal tissue (Figure 4a).

Figure 4
figure4

(a) RT–PCR analysis of DEC1 expression in ESCC cell lines. Normal esophageal tissue (N1) and DEC1 plasmid are used as the positive controls. All four ESCC cell lines show reduced expression of DEC1; (b) RT–PCR analysis of DEC1 stable transfectants and TSs. SLMT-1S1, V1 vector-alone stable transfectant, and C4, C8, C9 DEC1 stable transfectants, and two TS cell lines C8-2TS and C8-4TS, were analysed; (c) tumor growth kinetics of V1, V2, V6 vector-alone stable transfectants, and C4, C8, C9 DEC1 stable transfectants in nude mice. The curves represent an average tumor volume of all sites inoculated for each cell line

DEC1 cDNA transfection and colony formation assay

DEC1 cDNA was transfected into three DEC1 downregulated cell lines, SLMT-1S1, HKESC-2, and T.Tn6. In comparison to the vector-alone transfected groups, no obvious reduction in colony numbers is observed in DEC1-transfected groups (data not shown).

Tumorigenicity assay of DEC1 stable transfectants

The DEC1 gene expression of SLMT-1S1, the pcDNA3.1 (+) vector-alone stable transfectant, and the DEC1 stable transfectants is shown in Figure 4b. The DEC1 stable transfectants express much greater quantities of the DEC1 gene compared to the SLMT-1S1 recipient cell line. All three DEC1 stable transfectants display a delayed latency period in forming tumors (average latency period of 32–91 days), and tumors are of reduced size compared with the vector-alone control (average latency period of 8–28 days) (Figure 4c). These differences are highly significant (P<0.0001). These results are summarized in Table 2.

Status of DEC1 in TSs, ESCC primary tumors and chromosome 9 MCHs

Tumors arising in nude mice injected with the more aggressive C8 DEC1 stable transfectant were excised for further analysis. Two TS cell lines, C8-2TS and C8-4TS, were analysed by RT–PCR and show a decreased DEC1 gene expression compared to the original C8 DEC1 stable transfectant (Figure 4b).

Tissues from 34 ESCC patients were analysed using the KM9.1 microsatellite marker mapping to 9q32–q33 and flanking the position of the DEC1 gene (Miura et al., 1996). An LOH frequency of 74% (20/27) is observed (Figure 2).

DEC1 maps to the region marked by KM9.1, which is around 11 Mb away from the CR identified by the MMCT approach (Figure 2). DEC1 is present in donor chromosome 9 cell line Neo.9, as observed by both RT–PCR and KM9.1 microsatellite typing. To check the expression status of DEC1 following MMCT, RT–PCR analysis was performed, and it showed no appreciable increase in DEC1 expression in each MCH cell line compared to the SLMT-1S1 recipient (data not shown).

Discussion

Candidate ESCC TSGs mapped to chromosome 9 include p16INK4a and DEC1. In a previous study, we found that transfected wild-type p16INK4a can suppress cell growth of ESCC cell lines in vitro (Kwong et al., 2004). DEC1 is also a candidate TSG, based on its high LOH frequency, reduced expression and ability to suppress cell growth in vitro (Nishiwaki et al., 2000). However, no in vivo evidence of its tumor suppressive effect is available. Tumorigenicity in nude mice is recognized as the most reliable indicator of malignancy (Fogh et al., 1977). Since malignant and transformed phenotypes may be under separate genetic control (Stanbridge et al., 1978), the results from in vitro colony formation assays may not always be consistent with in vivo tumorigenicity assays. In vivo studies are required to further investigate DEC1 tumor suppressive activities.

In this study, an LOH frequency of 74% for the DEC1 gene is observed in 34 ESCC primary tumors. DEC1 was not able to suppress cell growth in a Hong Kong ESCC cell line, SLMT-1S1, in contrast to the findings in Japanese ESCC cell lines (Nishiwaki et al., 2000). This difference may be attributed to patterns of suppression of malignancy varying depending on not only the genes to be transfected but also on the recipient cells utilized (Murakami, 2002). On the other hand, tumor growth of stable transfectants with DEC1 overexpression was markedly suppressed in nude mice. Further monitoring of DEC1 expression in TS cell lines demonstrated a decreased DEC1 expression, supporting its importance in ESCC tumorigenicity.

Numerous studies in chondrosarcomas (Jagasia et al., 1996), skin cancers (Holmberg et al., 1996), lung carcinomas (Wiest et al., 1997), neuroblastomas (Marshall et al., 1997), breast carcinomas (Gorgoulis et al., 1998), sporadic basal cell carcinomas (He et al., 1999), cutaneous malignant melanomas (Parris et al., 1999), bladder cancers (Wu et al., 1996), renal cell carcinomas (Grady et al., 2001), and ESCC (Nishiwaki et al., 2000) have suggested that chromosome 9 may harbor several other TSGs besides p16INK4a. Tumor suppression using MMCT helps narrow down the search for key genes contributing to tumorigenesis. Our previous studies using an MMCT approach in another cancer, nasopharyngeal carcinoma (NPC), successfully identified a CR for tumor suppression mapping to chromosome 3p21.3 (Cheng et al., 1998). More recent studies have validated the usefulness of this approach and at least two NPC-associated candidate TSGs, RASSF1A and BLU, have now been mapped to this region (Chow et al., 2004; Qiu et al., 2004).

To date, only four MMCT studies of chromosome 9 have been reported. These include studies on uterine endometrial carcinoma (Yamada et al., 1990), bladder carcinoma (Wu et al., 1996), cutaneous malignant melanoma (Parris et al., 1999), and NPC (Cheng et al., 2000). Two of these studies (Wu et al., 1996; Cheng et al., 2000) show that the 9p21 p16INK4a gene is important for tumor formation, while TSGs other than p16INK4a are implicated in melanoma and uterine carcinomas.

The current study shows that a chromosome 9 null for p16INK4a gene expression can still functionally suppress tumor formation by ESCC cells, resulting in tumors of reduced size and longer latency periods in the MCH9.1 and MCH9.6 cell lines. The higher tumorigenic potential of MCH9.3, compared to MCH9.1 and MCH9.6, may be attributed to the finding that less than 20% of its cells show transfer of the whole chromosome 9 by karyotyping, in contrast to results observed by FISH for the other two MCHs. The corresponding microsatellite typing analysis shows that for a number of markers, the intensity of the donor allele in MCH9.3 is lower than that present in the other two MCHs. Figure 3a illustrates this for the marker D9S112. The MCH cell lines have nine chromosome 9-staining copies by FISH analysis. The subsequent loss of the one donor allele in the TS is expected to show only a reduction of 1/9 (11%) in peak signal intensities. However, the cutoff value used in this study for LOH for the TSs was a 20% difference in peak heights. Although the extent of loss observed with the MCH9.1.2TS is not as obvious as that for the other two TSs in Figure 3a, its loss is significant. Genetic analysis of these TSs identifies a CR of common loss, presumably associated with recurrence of the tumorigenic phenotype. In the current study, detailed genotyping of MCHs and their corresponding TSs showed one 2.4 Mb CR commonly associated with tumor suppression mapping to 9q33–q34.

In an earlier study (Yang et al., 2004), we observed extensive and multiply lost regions on chromosome 9q in 34 ESCC tissues. Since the 10 chromosome 9q markers used in this earlier study did not contain several markers mapped to the CR observed in this MMCT functional study, we utilized these additional five markers including and neighboring the D9S112 marker, which appeared to be critically eliminated in all 12 of the MCH TSs analysed, to examine the 34 ESCC tissues for allelic loss. LOH frequencies of 67–86% were observed for these markers. This high loss is consistent with the evidence from MMCT that critical TSGs mapping to this region may be essential for ESCC tumorigenesis. It is also obvious that the extent of the losses detected by this type of molecular approach is so great that no single gene can be isolated and that other functional approaches are required to narrow down the regions to which candidate TSGs are likely to map.

The current study identifies an important gene-rich region at 9q32–q34, to which several human candidate cancer genes map. The region includes two bladder cancer-associated genes, DBC1 mapped at 9q32–q33 (Habuchi et al., 1998; Nishiyama et al., 1999) and TSC1 mapped at 9q34 (Sampson et al., 1995). Other genes in this vicinity include PBX3, mapped at 9q33–q34 (Qin et al., 2004), ENG, which is a component of the TGF-β receptor complex and can regulate the biological effect of TGF-β in tumor angiogenesis (Parker et al., 2003) mapped at 9q33–q34.1, and the myeloid leukemia-associated gene, SET, which is a regulator of G2/M transition (Li et al., 1996) mapped at 9q34. Indeed, the search in the 9q32–q34 region remains a high-priority initiative, since both LOH studies in primary tumor tissues and functional MMCT studies implicate this region as being involved in ESCC tumorigenesis.

The expression level of DEC1 is downregulated in the SMLT-1S1 recipient. Transfer of the 9q-intact chromosome suppressed tumorigenicity, but failed to restore DEC1 expression to normal levels as monitored by RT–PCR. One possibility, that remains to be tested, is there may be important upstream elements involved in the DEC1 expression regulatory pathway, which are defective in these MCHs, which fails to restore DEC1 gene expression after chromosome 9 transfer into recipient cells. These observations suggest that other unidentified TSGs mapping at 9q33–q34 are involved in ESCC development, genes, which are neither the p16INK4a nor the DEC1 gene.

Although frequent deletions have also been detected on chromosome 18 in ESCC by LOH and CGH studies, the present current MMCT study indicates that chromosome 18 transfer was unable to suppress tumorigenic behavior in animal studies. Thus, the change in the suppression phenotype observed in chromosome 9 MMCT studies is neither due to the process of MMCT nor to the possible clonal variation of the recipient, but rather may be attributed to the specific chromosome introduced. Similar results were observed in our previous studies in which chromosome 17 did not suppress tumorigenicity in an animal model, while chromosome 11 did suppress tumor growth of a NPC cell line HONE1 (Cheng et al., 2000).

In conclusion, we show that chromosome 9 deletions occur at a significant frequency in ESCC tumors, provide functional evidence for the presence of other TSGs on chromosome 9 important for ESCC besides p16INK4a, and narrow down the candidate region for TSGs to a 2.4 Mb region. Another candidate TSG DEC1 is also implicated in this cancer, but does not appear to be the suppressor transferred by chromosome 9 in these studies. Work towards verification and identification of the chromosome 9 TSGs essential for ESCC tumorigenesis is continuing.

Materials and methods

Cell lines, culture conditions, and tissue specimens

The SLMT-1S1 cell line was established from a Hong Kong Chinese ESCC patient (Tang et al., 2001) and used as a recipient for the MMCT study. ESCC cell lines, SLMT-1S1, HKESC-2 (Hu et al., 2002), and T.Tn6 (Takahashi et al., 1990) were used as recipients for gene transfection. The 81T cell line was established from a Taiwanese patient (Hu et al., 1984). These cell lines were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% fetal bovine serum (FBS), 5% newborn calf serum (NCS) and 1% penicillin/streptomycin (Invitrogen, Grand Island, NY, USA). Neo.9 is an A9 mouse cell line containing a human chromosome 9 microdeleted at 9p21 (Yamada et al., 1990; Cheng et al., 2000). MCH912.1 is a mouse A9 cell containing an intact human chromosome 18 (Goyette et al., 1992). Neo.9 and MCH912.1 were used as the donor cell lines for MMCT. The cells were grown in DMEM/5% FBS/5% NCS/G418 (Geneticin) (Invitrogen, San Diego, CA, USA) at 800 μg/ml. All SLMT-1S1 MCHs, DEC1 transfectants and pcDNA3.1 (+) vector alone transfectants were selected in growth medium containing 400–500 μg/ml G418. All TSs were grown under the same condition as the recipient cell SLMT-1S1. The cells were regularly monitored for mycoplasma presence and were consistently negative.

DNAs from 34 pairs of tumor and corresponding normal tissues were obtained from Chinese ESCC patients diagnosed at Tuen Mun Hospital in Hong Kong from 1993 to 1999, as described previously (Yang et al., 2004).

Normal esophageal tissues were extracted from Hong Kong gastric cancer patients, who underwent surgery in Queen Mary Hospital. The esophageal tissues were verified as normal under the microscope by a pathologist.

Microcell-mediated human chromosomes 9 and 18 transfer

MMCT of chromosomes 9 and 18 is essentially performed as described previously (Goyette et al., 1992). After fusion, the MCHs were selected in medium containing G418 at 500 μg/ml and hypoxanthine/aminopterin/thymidine (HAT) (Sigma, St Louis, MO, USA).

DNA slot blot assay

To detect any mouse DNA contamination of MCHs, a DNA slot blot was performed as described previously (Cheng et al., 1998).

Fluorescence in situ hybridization

Whole chromosomes 9 and 18 FISH probes were purchased from Vysis (Downers Grove, IL, USA). Hybridization protocols are as described previously (Cheng et al., 2000). The probe signals were processed by use of SPOT software (Sterling Heights, MI, USA) on an Olympus BX51 microscope (Olympus, Tokyo, Japan). A minimum of 20 metaphase spreads were analysed.

Microsatellite markers

In all, 39 chromosome 9 and 10 chromosome 18 specific microsatellite markers were utilized. Details about the cytogenetic localization of the chromosome 9q microsatellite markers used are seen in Table 1. The other chromosome 9p markers utilized in this study include D9S1779, D9S1810, D9S324, D9S168, D9S269, D9S254, D9S157, D9S171, D9S126, D9S259, D9S1853, D9S1874, D9S55, and D9S970. The chromosome 18 markers include D18S1132, D18S1150, D18S1107, D18S468, D18S34, D18S450, D18S46, DCC, D18S465, and D18S469. Primer information was from the Genome Data Base (http://www.gdb.org) and NCBI genome database (http://www.ncbi.nlm.nih.gov/). Information on KM9.1, KM9.2, and DCC was as reported previously (Huang et al., 1992; Miura et al., 1996).

PCR microsatellite assay and LOH studies

Microsatellite analysis of recipient, donor, MCH, and TS cell lines was performed using fluorescent PCR-based analysis on an Applied Biosystems Prism 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA), essentially as described previously (Cheng et al., 1998). Reactions were carried out in a volume of 10 μl containing 10 × PCR buffer II, 200 μ M dNTP, 4 μ M F-dNTP (for nonfluorescent labeled markers), 1.5 mM MgCl2, 5 pmol each primer, 0.5 U AmpliTaq Gold DNA polymerase, and sterile deionized water. Between 20 and 100 ng DNA was amplified for 30 cycles. The annealing time is 30 s for all markers, except 15 s for D9S778 and 45 s for D9S61. Since there are nine copies of chromosome 9 in the MCH cell lines, allelic loss in the TSs was considered significant, if allelic ratios are less than or equal to 0.8 or greater than or equal to 1.2. For LOH studies, PCR products were similarly analysed using Genescan and Genotyper 2.1 software (Yang et al., 2004). Each allele was scored by comparing the ratio of the signal intensities between tumor and corresponding normal DNA in heterozygous individuals. Based on the formula T1:T2/N1:N2, where T stands for tumor and N stands for normal alleles, LOH is defined as having a ratio of less than or equal to 0.67 or greater than or equal to 1.5. Homozygous specimens are considered uninformative. All samples showing LOH result were confirmed at least twice.

Tumorigenicity assay and TS analyses

Chromosomes 9 and 18 MCHs and DEC1 stable transfectants were inoculated into nude mice to compare the tumor growth ability with the recipient cell line and pcDNA3.1 (+) vector-alone stable transfected cell lines. A total of 5 × 106 cells were inoculated subcutaneously into 6- to 8-week-old female athymic Balb/c Nu/Nu mice. For each cell line, six sites were injected into three animals, except for MCH9.3, for which only five sites were injected. Tumor growth in the nude mice was measured weekly with calipers. Tumors arising after chromosome 9 MCH and C8 DEC1 stable transfectant injections were reconstituted into cell culture for subsequent molecular analyses. The statistical significance of the difference in times for appearance of tumors at 3 weeks after injection was determined by the independent t-test using the SPSS version 10.0.1.

DNA transfection and colony formation assay

The pcDNA3.1 (+) expression plasmid constructs contain a selectable neomycin-resistance gene. The 525 bp cDNA clone of the wild-type DEC1 gene, mapping from 213 to 737 bp on the DEC1 cDNA sequence (Nishiwaki et al., 2000) and control pcDNA3.1 (+) vector (5.4 kb) were transfected into SLMT-1S1, T.Tn6, and HKESC-2 cells using lipofectamine and Plus™ reagent (Invitrogen, San Diego, CA, USA). The method was according to the manufacturer's protocols. DNA (1 μg) is thoroughly mixed with 4–6 μl Plus™ reagent and 5 μl lipofectamine for 15 min at room temperature separately before addition to the cells for 6–8 h. The cells are subsequently split into three 100 mm dishes 24 h later. Three separate experiments were performed. After 18–21 days of selection in growth medium containing 400–500 μg/ml G418, colonies were fixed and stained with giemsa (Sigma, St Louis, MO, USA). Stable clones from the vector-alone group and DEC1-transfected group were subcloned and analysed.

RT–PCR analysis

Total RNA was extracted from the cell lines using an RNeasy mini kit (QIAGEN, Chatsworth, CA, USA). Total RNA (1 μg) was reverse transcribed using the M-MLV Reverse Transcriptase (USB, Cleveland, OH, USA) for first-strand cDNA synthesis. cDNA (2 μl) is used for PCR in a 20 μl reaction mixture with 5 pmol of each primer, 5 mM dNTP, 1 U Taq polymerase (Boehringer Mannheim GmbH, Germany), 1 × PCR buffer with MgCl2, and sterile deionized water. PCR for DEC1 transcript was performed under the following conditions: 5 min at 94°C; 30 cycles of 30 s at 94°C, 30 s at 60°C, 1 min at 72°C; followed by final 5 min at 72°C. As an internal control, the GAPDH gene was amplified from the same cDNA samples for 25 cycles. Normal esophageal tissues and the pcDNA3.1 (+)/DEC1 plasmid serve as positive controls. For amplification of the DEC1 gene and the GAPDH gene, two sets of sense and antisense oligonucleotides were used. These are 5′-IndexTermAAGTGGAAGAGCATTGTGCC-3′ and 5′-IndexTermCCTGGCAAGAGATACAGCAA-3′, with nucleotide positions 550–569 and 686–705 for DEC1, and 5′-IndexTermGAGTCAACGGATTTGGTCGT-3′ and 5′-IndexTermATCCACAGTCTTCTGGGTGG-3′, with nucleotide positions 123–143 and 551–670 for GAPDH. A 156 bp fragment of the DEC1 and 548 bp fragment of the GAPDH transcript was then amplified by PCR.

References

  1. Aoki T, Mori T, Du X, Nishihira T, Matsubara T and Nakamura Y . (1994). Genes Chromosome Cancer, 10, 177–182.

  2. Cheng Y, Poulos NE, Lung ML, Hampton G, Ou B, Lerman MI and Stanbridge EJ . (1998). Proc. Natl. Acad. Sci. USA, 95, 3042–3047.

  3. Cheng Y, Stanbridge EJ, Kong H, Bengtsson U, Lerman MI and Lung ML . (2000). Genes Chromosome Cancer, 28, 82–91.

  4. Chow LS, Lo KW, Kwong J, To KF, Tsang KS, Lam CW, Dammann R and Huang DP . (2004). Int. J. Cancer, 109, 839–847.

  5. Fogh J, Fogh JM and Orfeo T . (1977). J. Natl. Cancer Inst., 59, 221–226.

  6. Gorgoulis VG, Koutroumbi EN, Kotsinas A, Zacharatos P, Markopoulos C, Giannikos L, Kyriakou V, Voulgaris Z, Gogas I and Kittas C . (1998). Mol. Med., 4, 807–822.

  7. Goyette MC, Cho K, Fasching CL, Levy DB, Kinzler KW, Paraskeva C, Vogelstein B and Stanbridge EJ . (1992). Mol. Cell. Biol., 12, 1387–1395.

  8. Grady B, Goharderakhshan R, Chang J, Ribeiro-Filho LA, Perinchery G, Franks J, Presti J, Carroll P and Dahiya R . (2001). J. Urol., 166, 1088–1092.

  9. Habuchi T, Luscombe M, Elder PA and Knowles MA . (1998). Genomics, 48, 277–288.

  10. He C, Zhang X, Wang Y, Sun K and Chen HD . (1999). Chin. Med. J., 112, 497–500.

  11. Holmberg E, Rozell BL and Toftgard R . (1996). Br. J. Cancer, 74, 246–250.

  12. Hu CP, Hsieh HG, Chien KY, Wang PY, Wang CI, Chen CY, Lo SJ, Wu KD and Chang CM . (1984). J. Natl. Cancer Inst., 72, 577–583.

  13. Hu YC, Lam KY, Law SY, Wan TS, Ma ES, Kwong YL, Chan LC, Wong J and Srivastava G . (2002). Cancer Genet. Cytogenet., 135, 120–127.

  14. Huang Y, Boynton RF, Blount PL, Silverstein RJ, Yin J, Tong Y, McDaniel TK, Newkirk C, Resau JH, Sridhara R, Reid BJ and Meltzer SJ . (1992). Cancer Res., 52, 6525–6530.

  15. Jagasia AA, Block JA, Qureshi A, Diaz MO, Nobori T, Gitelis S and Iyer AP . (1996). Cancer Lett., 105, 91–103.

  16. Karkera JD, Ayache S, Ransome Jr RJ, Jackson MA, Elsayem AF, Sridhar R, Detera-Wadleigh SD and Wadleigh RG . (2000). Clin. Cancer Res., 6, 3565–3569.

  17. Kwong FM, Tang JC, Srivastava G and Lung ML . (2004). Cancer Lett., 208, 207–213.

  18. Li M, Makkinje A and Damuni Z . (1996). J. Biol. Chem., 271, 11059–11162.

  19. Marshall B, Isidro G, Martins AG and Boavida MG . (1997). Cancer Genet. Cytogenet., 96, 134–139.

  20. Miura K, Suzuki K, Tokino T, Isomura M, Inazawa J, Matsuno S and Nakamura Y . (1996). Cancer Res., 56, 1629–1634.

  21. Mori T, Yanagisawa A, Kato Y, Miura K, Nishihira T, Mori S and Nakamura Y . (1994). Hum. Mol. Genet., 3, 1969–1971.

  22. Murakami Y . (2002). Oncogene, 21, 6936–6948.

  23. Muzeau F, Flejou JF, Thomas G and Hamelin R . (1997). Int. J. Cancer, 72, 27–30.

  24. Naidoo R, Tarin M, Reddi A and Chetty RD . (1999). Diagn. Mol. Pathol., 8, 131–137.

  25. Nishiwaki T, Daigo Y, Kawasoe T and Nakamura Y . (2000). Genes Chromosome Cancer, 27, 169–176.

  26. Nishiyama H, Takahashi T, Kakehi Y, Habuchi T and Knowles MA . (1999). Genes Chromosome Cancer, 26, 171–175.

  27. Pack SD, Karkera JD, Zhuang Z, Pak ED, Balan KV, Hwu P, Park WS, Pham T, Ault DO, Glaser M, Liotta L, Detera-Wadleigh SD and Wadleigh RG . (1999). Genes Chromosome Cancer, 25, 160–168.

  28. Parker WL, Goldring MB and Philip A . (2003). J. Bone Miner. Res., 18, 289–302.

  29. Parris CN, Harris JD, Griffin DK, Cuthbert AP, Silver AJR and Newbold RF . (1999). Cancer Res., 59, 516–520.

  30. Qin P, Haberbusch JM, Soprano KJ and Soprano DR . (2004). J. Cell. Biochem., 92, 147–163.

  31. Qiu GH, Tan LK, Loh KS, Lim CY, Srivastava G, Tsai ST, Tsao SW and Tao Q . (2004). Oncogene, 23, 4793–4806.

  32. Sampson JR, Patel A and Mee AD . (1995). J. Med. Genet., 32, 848–850.

  33. Saxon PJ, Srivatsan ES, Leipzig GV, Sameshima JH and Stanbridge EJ . (1985). Mol. Cell. Biol., 5, 140–146.

  34. Shibagaki, Shimada Y, Wagata T, Ikenaga M, Imamura M and Ishizaki K . (1994). Cancer Res., 54, 2996–3000.

  35. Stanbridge EJ and Wilkinson J . (1978). Natl. Acad. Sci. USA, 75, 1466–1469.

  36. Tada K, Oka M, Hayashi H, Tangoku A, Oga A and Sasaki K . (2000). Cancer Genet. Cytogenet., 117, 108–112.

  37. Takahashi K, Kanazama H, Chan CT, Hosono T, Takahara M and Sato K . (1990). Jpn. J. Oral Maxillofac. Surg., 36, 307–316.

  38. Tang JCO, Wan TSK, Wong N, Pang E, Lam KY, Law SY, Chow LMC, Ma ESK, Chan LC, Wong J and Srivastava G . (2001). Cancer Genet. Cytogenet., 124, 36–41.

  39. Wang L, Li W, Wang X, Zhang C, Zhang T, Mao X and Wu M . (1996). Oncogene, 12, 699–703.

  40. Wiest JS, Franklin WA, Otstot JT, Forbey K, Varella-Garcia M, Rao K, Drabkin H, Gemmill R, Ahrent S, Sidransky D, Saccomanno G, Fountain JW and Anderson MW . (1997). Cancer Res., 57, 1–6.

  41. Wu Q, Possati L, Montesi M, Gualandi F, Rimessi P, Morelli C, Trabanelli C and Barbanti-Brodano G . (1996). Int. J. Cancer, 65, 840–846.

  42. Yamada H, Wake N, Fujimoto S, Barrett JC and Oshimura M . (1990). Oncogene, 5, 1141–1147.

  43. Yang LC, Tang CMC, Lau KW and Lung ML . (2004). Cancer Lett., 203, 71–77.

  44. Yen CC, Chen YJ, Chen JT, Hsia JY, Chen PM, Liu JH, Fan FS, Chiou TJ, Wang WS and Lin CH . (2001). Cancer, 92, 2769–2777.

  45. Yen CC, Chen YJ, Lu KH, Hsia JY, Chen JT, Hu CP, Chen PM, Liu JH, Chiou TJ, Wang WS, Yang MH, Chao TC and Lin CH . (2003). Int. J. Oncol., 23, 871–881.

Download references

Acknowledgements

We thank Y Cheng and HL Lung for helpful discussions during manuscript preparation. We would like to acknowledge the financial support of the Research Grants Council of the Hong Kong Special Administration Region, China (HKUST) to MLL, for HKUST 6106/00M grant.

Author information

Correspondence to Maria L Lung.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • chromosome 9
  • deleted in esophageal cancer 1
  • tumor suppressor gene
  • esophageal carcinoma
  • microcell-mediated chromosome transfer

Further reading