Abstract

Oxidized DNA base lesions, such as thymine glycol (Tg) and 8-hydroxyguanine, are often toxic and mutagenic and have been implicated in carcinogenesis. To clarify whether NEIL1 protein, which exhibits excision repair activity towards such base lesions, is involved in gastric carcinogenesis, we examined 71 primary gastric cancers from Japanese patients and four gastric cancer cell lines for mutations and genetic polymorphisms of the NEIL1 gene. We also examined 20 blood samples from Chinese patients for NEIL1 genetic polymorphisms. Three mutations (c.82_84delGAG:p.Glu28del, c.936G > A and c.1000A > G:p.Arg334Gly) and two genetic polymorphisms were identified. When the excision repair activity towards double-stranded oligonucleotide containing a Tg:A base pair was compared among six types of recombinant NEIL1 proteins, p.Glu28del-type NEIL1, found in a primary case, was found to exhibit an extremely low activity level. Moreover, c.936G > A, located in the last nucleotide of exon 10 and detected in the KATO-III cell line, was shown to be associated with a splicing abnormality using an in vivo splicing assay. An immunofluorescence analysis showed that the wild-type NEIL1 protein, but not the truncated protein encoded by the abnormal transcript arising from the c.936G > A mutation, was localized in the nucleus, suggesting that the truncated protein is unlikely to be capable of repairing nuclear DNA. An expression analysis revealed that NEIL1 mRNA expression was reduced in six of 13 (46%) primary gastric cancer specimens that were examined. These results suggest that low NEIL1 activities arising from mutations and reduced expression may be involved in the pathogenesis in a subset of gastric cancers.

Introduction

Oxidative DNA damage is widely accepted to be a causative factor of cancer; this type of damage includes pre-mutagenic and blocking lesions that may lead to mutation and cell death, respectively ( 1 , 2 ). 8-Hydroxyguanine (oh8G) and thymine glycol (Tg) are typical examples of the former and latter lesions, respectively, and are used as biomarkers of oxidative stress ( 13 ). The base excision repair pathway is a universal mechanism for such base lesions in DNA and is initiated by DNA glycosylases that hydrolyze the N -glycosidic bond, releasing the damaged base from double-stranded DNA. Human DNA glycosylases encoded by the OGG1 (MIM# 601982) and NTH1 (MIM# 602656) genes have the activity to remove oh8G and Tg from DNA, respectively ( 412 ). So far, OGG1 somatic mutations have been reported in some types of cancers, and the impairment of DNA repair activity in these mutant proteins has been proven experimentally ( 1316 ). In these papers, the impaired DNA repair activity of the OGG1 protein was speculated to lead to an accumulation of oxidative DNA damage and an increase in the mutation rate of cells, which may be related to cancer. Moreover, an association between the OGG1 genetic polymorphism and cancer has also been reported ( 17 , 18 ). The NEIL1 gene (MIM# undecided) has been isolated recently and revealed to encode DNA glycosylase for oh8G and Tg ( 1926 ). Judging from the similarity of the substrates, an investigation of the somatic mutations and genetic polymorphisms of the NEIL1 gene would seem to be worthwhile. Stomach tissue, in particular, is exposed to oxidative stresses, including inflammation induced by sodium chloride, Helicobacter pylori infection and smoking ( 2729 ). The huge amounts of oxidative DNA damage may play an important role in carcinogenesis and a predisposition to gastric cancer. Therefore, in this study, we examined primary gastric cancers for somatic mutations and genetic polymorphisms of the NEIL1 gene and assessed the functional status of the detected NEIL1 variants. We also assessed NEIL1 expression in tumor tissue. This is the first report to describe a genetic alteration of the NEIL1 gene in human cancer tissue.

Materials and methods

Samples

Gastric cancers from a total of 71 sporadic cases and corresponding normal mucosae were obtained from Hamamatsu University Hospital, Shizuoka, Japan. The four gastric cancer cell lines used in this study were MKN45, MKN74, TMK-1 and KATO-III ( 30 ). Blood samples from a total of 20 Chinese primary gastric cancer patients were obtained from Huaian City Municipal Hospital, China ( 31 ). The protocol was approved by the Human Institutional Review Board of Hamamatsu University School of Medicine. DNA was extracted by standard SDS–proteinase K digestion followed by phenol–chloroform extraction and ethanol precipitation.

NEIL1 variant screening

NEIL1 variants were screened with the WAVE DNA Fragment Analysis System (Transgenomic, Omaha, NE). The method is based on ion-pair, reverse-phase high-performance liquid chromatography and temperature-modulated heteroduplex analysis. Nine coding exons of the NEIL1 gene were amplified by genomic polymerase chain reaction (PCR) using 10 sets of primers: 5′-AGC CGC TAC CTC ACA AAG TC-3′ and 5′-GCT GAA AAG AGC CGG ACA TG-3′ for exon 4a, 5′-CTA CCG CAT CTC AGC TTC AG-3′ and 5′-AGA AGG CAC TAA GAG AGT CAG-3′ for exon 4b, 5′-ACG CAC CCA GAC CGT GTT C-3′ and 5′-GGC AAG CCA TGC TAG GCA G-3′ for exon 5, 5′-AGA GAT CCT GTA CCG GTC AG-3′ and 5′-CCC CTG TTG TTG AGC TAA GC-3′ for exon 6, 5′-AGC TCA GAG AGA AGG TCA GC-3′ and 5′-TGC TGA AGC AGA GGT TGC TG-3′ for exon 7, 5′-TCA GTC CAT CAG CTT GTG TAG-3′ and 5′-CTA GAC AGA CCC TGA GCA TC-3′ for exon 8, 5′-CCT GCC TCT CCA AGG AAT AC-3′ and 5′-TGT AGC CCT TGC AGG CGT G-3′ for exon 9, 5′-CCA ACT CCA ACA CCA GTG TC-3′ and 5′-AGG GTA AGG CAG CTG CCT G-3′ for exon 10, 5′-ACC TCT GAA CTG CTT TCT GAG-3′ and 5′-CTT TGT ACA CGC GCT CCT AG-3′ for exon 11, and 5′-CAG CCC CTG GAG TCT TAG C-3′ and 5′-ACC TTC AGA TAT TGC CTG CTC-3′ for exon 12. PCR was performed in 25 µl reaction mixtures containing AmpliTaq Gold (Applied Biosystems, Tokyo, Japan) under the following conditions: 30 s at 94°C, 30 s at 56°C and 60 s at 72°C for 35 cycles. PCR products were analyzed using WAVE and WAVEMaker software 4.0 (Transgenomic). PCR products exhibiting different peaks in the WAVE analysis were directly sequenced using a BigDye Terminator Cycle Sequencing Reaction Kit and the ABI 3100 Genetic Analyzer (Applied Biosystems).

Preparation of the recombinant NEIL1 proteins

NEIL1 protein tagged with His 6 at its C-terminus was prepared according to the pET System protocol (Novagen, Darmstadt, Germany). Briefly, NEIL1 cDNA-containing pET25b(+) vector was constructed by inserting a PCR product of NEIL1 cDNA into the Nde I and Xho I sites of a pET25b(+) vector (Novagen). Escherichia coli BL21-CodonPlus (DE3)-RP competent cells (Stratagene, La Jolla, CA) were transformed with the NEIL1 cDNA-containing pET25b vectors and cultured at 37°C until A 600 reached 0.5. After incubation with 0.5 mM isopropyl-1-thio- β - d -galactopyranoside (IPTG) at 25°C for 12 h, NEIL1-His 6 protein was purified using TALON metal affinity resins (Clontech, Palo Alto, CA) under native conditions according to the manufacturer's instructions. The protein was then dialyzed against the buffer containing 20 mM sodium phosphate (pH 7.5), 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 1 mM phenylmethylsulfonylfluoride and 20% glycerol. The quality and concentration of the protein were determined using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). For the construction of the vector for the NEIL1variant forms, a QuikChange Site-Directed Mutagenesis kit (Stratagene) was used.

DNA cleavage activity assay

A 30mer oligonucleotide containing or not containing a single 5 R ,6 S -Tg (5′-CTG GTG GCC TGA C[Tg or T]C ATT CCC CAA CTA GTG-3′) was chemically synthesized and purified by polyacrylamide gel electrophoresis (PAGE) (Japan Bio Services, Saitama, Japan). The oligonucleotide was 32 P-labeled at the 5′ terminus with a MEGALABEL kit (Takara, Osaka, Japan) and [γ- 32 P]ATP (Amersham Biosciences, Piscataway, NJ) and annealed to a complementary strand containing an adenine opposite the Tg or T. The reaction was performed with 20 µl of a mixture containing 25 nM NEIL1-His 6 protein, 20 mM sodium phosphate (pH 7.5), 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 2.5 nM labeled DNA and 50 µg/ml bovine serum albumin at 37°C. After the reaction, denaturing formamide dye was added to the mixture, heated at 95°C for 3 min, and subjected to 20% PAGE. A 13mer oligonucleotide (5′-CTGGTGGCCTGAC P -3′) was 32 P-labeled at the 5′ terminus and used as a size marker for the cleaved products. The radioactivities of intact and cleaved oligonucleotides were quantified using a bioimaging analyzer (BAS1000, Fuji Photo Film, Tokyo) and ImageGauge software (Fuji Photo Film).

RNA isolation and reverse transcription (RT)–PCR

Total RNA was extracted with an RNeasy Mini Kit (Qiagen, Valencia, CA) and converted to first-strand cDNA with a SuperScript First-Strand Synthesis System for RT–PCR (Invitrogen, Carlsbad, CA) according to supplier's protocol. PCR amplification was performed using the following sets of primers: 5′-GAA ATC CAA GGC CAC ACA GC-3′ and 5′-TTG CAG TCC TCT TAG GAA GGT C-3′ for NEIL1 transcripts and 5′-CCA AGG TCA TCC ATG ACA AC-3′ and 5′-CAC CCT GTT GCT GTA GCC A-3′ for GAPDH transcripts. PCR products were fractionated by electrophoresis on an agarose gel and stained with ethidium bromide. The band detected by electrophoresis was excised from the gel and directly sequenced.

In vivo splicing assay

NEIL1-936G and -936A type vectors were constructed by inserting a PCR product of the NEIL1 gene into the Eco RI and Mlu I sites of a pALTER-MAX (Promega, Madison, WI) mammalian expression vector, as indicated in Figure 2C . The expression vectors were transfected into a lung cancer cell line, NCI-H1299, using the LipofectAMINE 2000 reagent (Invitrogen) according to the supplier's recommendations. At 24 h post-transfection, the cells were harvested, and total RNA was extracted and converted to cDNA. RT–PCR was performed using a set of primers, pAM-F (5′-CAG CTC TTA AGG CTA GAG TAC-3′) and pAM-R (5′-CTT ATC ATG TCT GCT CGA AGC-3′), and PCR products were fractionated by agarose gel electrophoresis.

Immunofluorescence analysis

Wild-type and truncated-type NEIL1 expression vectors containing a FLAG tag were constructed by inserting each cDNA with the FLAG sequence at the C-terminus into the Xba I and Eco RI sites of a pcDNA3.1(+) plasmid vector (Invitrogen). The p.Arg334Gly type NEIL1 expression vector was prepared using a QuikChange Site-Directed Mutagenesis kit (Stratagene) and a wild-type NEIL1-FLAG expression vector. The expression vector was then transfected into NCI-H1299 cells using LipofectAMINE 2000 reagent (Invitrogen), according to the supplier's recommendations. After 24 h, the cells were fixed with 4% paraformaldehyde. After microwave treatment and permeabilization, the cells were incubated with 40 μg/ml of anti-FLAG M2 antibody (Sigma, St Louis, MO) at room temperature for 1 h. Indirect immunofluorescence labeling was performed using a fluorescein 5-iso-thiocyanate (FITC)-conjugated anti-mouse IgG secondary antibody (Cappel, Aurora, OH) at room temperature for 30 min, and the nuclei were stained with 4′-6-diamidino-2-phenylindole (DAPI) (Vysis, Downers Grove, IL). The slides were examined under a fluorescence microscope (Olympus BX-50-FL; Olympus) equipped with epifluorescence filters and a photometric CCD camera (Quantix 1400; Roper Scientific, Tucson, AZ). The images captured were digitized and stored in the image analysis program (IPlab Spectrum; Scanalytics, Fairfax, VA).

Quantitative real-time (QRT)–PCR

Expression of the NEIL1 mRNA transcript was measured by QRT–PCR using a LightCycler instrument (Roche, Palo Alto, CA). The specimens available for analysis of its expression represented only a small fraction (18.3%: 13 of 71 cases) of the surgical specimens, because (i) most of the surgical specimens were small; (ii) nothing was left after most of the surgical specimens had been used for DNA extraction; and (iii) RNA samples of clearly low quality were rejected. The mirror section of the tissue sample for RNA extraction was confirmed to have the proper components using hematoxylin and eosin staining. PCR amplification of the NEIL1 transcript and the transcript of a control housekeeping gene, porphobilinogen deaminase ( PBGD ), was performed using a QuantiTect SYBR Green PCR kit (Qiagen). The following PCR primers were used: 5′-AGA AGA TAA GGA CCA AGC TGC-3′ and 5′-GAT CCC CCT GGA ACC AGA TG-3′ for the NEIL1 transcript and 5′-GTC TGG TAA CGG CAA TGC GG-3′ and 5′-TCC CCT GTG GTG GAC ATA GC-3′ for the PBGD transcript. The relative amounts of NEIL1 transcript were standardized according to those of the PBGD transcript. The relative expression of each sample was calculated by comparing the expression of non-cancerous gastric tissue in a primary gastric cancer patient (case no. 1). The T/N ratio was calculated in each case.

Statistics

A Wilcoxon signed-ranks test was performed using StatView software (Abacus Concepts, Berkeley, CA).

Results

Identification of NEIL1 mutations in gastric cancer

We examined 71 primary gastric cancers from Japanese patients and four gastric cancer cell lines for mutations and genetic polymorphisms of the NEIL1 gene. We also examined blood samples from 20 Chinese primary gastric cancer patients for genetic polymorphisms of the NEIL1 gene. A survey for all the coding sequences of the NEIL1 gene with intronic or non-coding sequence-based primers disclosed three mutations (c.82_84delGAG:p.Glu28del, c.936G > A and c.1000A > G:p.Arg334Gly) and two genetic polymorphisms ( Table I ). None of the three mutations were observed in any of the non-cancerous tissues examined in this study, and all of the mutations were observed in heterozygosity.

Table I.

Somatic mutations and genetic polymorphisms of the NEIL1 gene

Exon/intron a  Nucleotide change b  Predicted effect (polypeptide length c )  Sample  Allele frequency d
 
 

 

 

 

 
Japanese
 
Chinese
 
Somatic mutation      
    Exon 4 c.82_84delGAG p.Glu28del (389 aa) GC-16 (surgical specimen) 0.000 0.000 
    Exon 10 c.936G > A splicing abnormality (337 aa) KATO-III (cell line) 0.000 0.000 
    Exon 11 c.1000A > G p.Arg334Gly (390 aa) GC-71 (surgical specimen) 0.000 0.000 
Genetic polymorphism      
    Intron 7 [c.719–80delT + c.719–82T > G] Unknown JS-159 (blood sample) 0.000 0.025 
    Exon 8 c.733G > A p.Gly245Arg (390 aa) JS-158 (blood sample) 0.000 0.025 
Exon/intron a  Nucleotide change b  Predicted effect (polypeptide length c )  Sample  Allele frequency d
 
 

 

 

 

 
Japanese
 
Chinese
 
Somatic mutation      
    Exon 4 c.82_84delGAG p.Glu28del (389 aa) GC-16 (surgical specimen) 0.000 0.000 
    Exon 10 c.936G > A splicing abnormality (337 aa) KATO-III (cell line) 0.000 0.000 
    Exon 11 c.1000A > G p.Arg334Gly (390 aa) GC-71 (surgical specimen) 0.000 0.000 
Genetic polymorphism      
    Intron 7 [c.719–80delT + c.719–82T > G] Unknown JS-159 (blood sample) 0.000 0.025 
    Exon 8 c.733G > A p.Gly245Arg (390 aa) JS-158 (blood sample) 0.000 0.025 
a

According to LocusLink homepage (Locus ID: 79661) of the NCBI web site.

b

Nucleotide +1 is the A of the ATG-translation initiation codon.

c

Length of wild-type NEIL1 protein: 390 aa.

d

The allele frequencies were determined using analyses of non-cancerous tissues from Japanese and Chinese patients with gastric cancer.

Impairment in Tg DNA glycosylase activity of mutant-type NEIL1 protein

To assess the functional status of the proteins coded by the variant-type NEIL1 genes, each type of recombinant NEIL1 protein tagged with His 6 at its C-terminus was prepared by site-directed mutagenesis in a wild-type NEIL1 cDNA-containing pET25b(+) expression vector. Since the N-terminal Pro2 is an active site ( 19 ), we also prepared a p.Pro2del construct as a negative control. Although a p.Lys242Arg variant was not detected in the present series, this variant and p.Lys245Arg variant described here is available on the LocusLink homepage of the NCBI Web Site. Thus, we also prepared a p.Lys242Arg construct. In total, six recombinant NEIL1-His 6 proteins, including a wild-type protein, were expressed in E.coli and purified to apparent homogeneity ( Figure 1A ). Their molecular size of ∼47 kDa was determined by SDS–PAGE and corresponded with that calculated from the cDNA sequences. The Tg DNA glycosylase activity of the NEIL1 protein was tested by determining its capacity to cleave a double-stranded oligonucleotide containing a Tg:A mispair ( Figure 1B and C ). The enzymatic activities of the p.Lys242Arg, p.Gly245Arg and p.Arg334Gly type proteins were similar to that of the wild-type protein. However, the enzymatic activity of the p.Glu28del type protein was similar to that of the p.Pro2del type protein, indicating a strong impairment in the DNA repair activity of the p.Glu28del type protein.

Fig. 1.

Comparison of the cleavage activity toward double-stranded DNA containing a Tg:A mispair among NEIL1 variant proteins. ( A ) Purification of recombinant NEIL1 proteins resolved by PAGE and stained with Coomassie Brilliant Blue. NEIL1 protein tagged with His 6 at its C-terminus was prepared according to the pET System protocol. Lysates of E.coli culture with/without expression of wild-type (wt) NEIL1 protein (lanes 2–5) and purified various NEIL1 proteins (lanes 6–11) are shown. Pa. (lanes 2 and 3) and wt (lanes 4 and 5) indicate the transformation of the host cells with parental pET25b(+) and NEIL1 cDNA-containing pET25b(+) vector, respectively. − and + mean the absence and presence of IPTG induction, respectively. The NEIL1-His 6 protein band is indicated. ( B ) Cleavage activity of NEIL1 variant proteins toward double-stranded DNA containing a Tg:A mispair. Various types of NEIL1 proteins and a 32 P-labeled double-stranded oligonucleotide containing or not containing a single 5 R ,6 S -Tg:A mispair were incubated and subjected to 20% PAGE. A 32 P-labeled 13mer oligonucleotide was used as a size marker for the cleaved products. A representative result is shown in the panel. (−) at the top of the panel indicates no addition of NEIL1 protein. The intact 30mer and cleaved 13mer oligonucleotides are indicated by the arrows. The band with an asterisk in the lane of the Tg substrate is an oligonucleotide impurity. ( C ) The percentage of cleaved products per total under conditions of 37°C for 15 min was calculated as % incision and listed.

Fig. 1.

Comparison of the cleavage activity toward double-stranded DNA containing a Tg:A mispair among NEIL1 variant proteins. ( A ) Purification of recombinant NEIL1 proteins resolved by PAGE and stained with Coomassie Brilliant Blue. NEIL1 protein tagged with His 6 at its C-terminus was prepared according to the pET System protocol. Lysates of E.coli culture with/without expression of wild-type (wt) NEIL1 protein (lanes 2–5) and purified various NEIL1 proteins (lanes 6–11) are shown. Pa. (lanes 2 and 3) and wt (lanes 4 and 5) indicate the transformation of the host cells with parental pET25b(+) and NEIL1 cDNA-containing pET25b(+) vector, respectively. − and + mean the absence and presence of IPTG induction, respectively. The NEIL1-His 6 protein band is indicated. ( B ) Cleavage activity of NEIL1 variant proteins toward double-stranded DNA containing a Tg:A mispair. Various types of NEIL1 proteins and a 32 P-labeled double-stranded oligonucleotide containing or not containing a single 5 R ,6 S -Tg:A mispair were incubated and subjected to 20% PAGE. A 32 P-labeled 13mer oligonucleotide was used as a size marker for the cleaved products. A representative result is shown in the panel. (−) at the top of the panel indicates no addition of NEIL1 protein. The intact 30mer and cleaved 13mer oligonucleotides are indicated by the arrows. The band with an asterisk in the lane of the Tg substrate is an oligonucleotide impurity. ( C ) The percentage of cleaved products per total under conditions of 37°C for 15 min was calculated as % incision and listed.

Effect of the NEIL1 c.936G > A mutation on pre-mRNA splicing and subcellular localization

c.936G > A was detected in one gastric cancer cell line, KATO-III. As this nucleotide substitution has not been found in any other samples, it is probably a somatic mutation. However, as DNA from corresponding normal tissue for the KATO-III cell line was not available, we could not exclude the possibility that this substitution might be a rare genetic polymorphism. c.936G > A was located in the last nucleotide of exon 10. Guanine at the last nucleotide of the exon resides in the consensus sequence at the 5′ splice site, which is recognized and bound by U1 small nuclear RNP during 5′ splice site selection ( 32 ). Moreover, the G > A change at the last nucleotide of the exon in the OGG1 gene was associated with the retention of the following intron ( 13 ). Therefore, the c.936G > A change in the NEIL1 gene may weaken the splicing signal. RT–PCR and sequencing analyses revealed that a NEIL1 mRNA transcript derived from KATO-III cells, but not from TMK-1 cells, contained a 160-bp intron 10 sequence ( Figure 2A , band II). The transcript derived from KATO-III cells encoded the truncated-type NEIL1 protein ( Figure 2B ). To confirm the contribution of c.936G > A to this splicing abnormality, a pALTER-MAX mammalian expression vector was inserted by a PCR product containing c.936G or c.936A and surrounding the NEIL1 sequence, including intron 10, as illustrated in Figure 2C . These vectors and parental pALTER-MAX vector were then transfected into NCI-H1299 cells, and the total RNA was extracted from these cells. An RT–PCR analysis using a set of primers for the pALTER-MAX sequence and a sequencing analysis of the bands on the gel revealed that a transcript derived from the cells transfected with the NEIL1-c.936A type vector, but not from those with the NEIL1-c.936G type vector, contained the 160-bp intron 10 sequence ( Figure 2D ). These results indicate that the c.936G > A change caused the splicing abnormality. This abnormal splicing pattern leads to the replacement of the 78 aa C-terminus with 25 unrelated amino acids, resulting in the deletion of part of the nuclear localization signal (NLS). As the NEIL1 protein has been reported to be localized in the cell nucleus ( 20 , 21 ), we investigated whether the structural change resulting from the c.936G > A mutation actually affected subcellular localization. Wild-type and truncated-type NEIL1 cDNA fused with the FLAG sequence at the C-terminus were inserted into pcDNA3.1 mammalian expression vectors, and then transiently transfected into the NCI-H1299 cell line. The presence of exogenous NEIL1-FLAG proteins in the transfected cells was then examined using fluorescence microscopy ( Figure 3 ). Consistent with previous results ( 20 , 21 ), the wild-type NEIL1-FLAG protein was localized in the nucleus; however, the truncated-type NEIL1-FLAG protein was localized in the cytoplasm. We also evaluated the subcellular localization of p.Arg334Gly-type protein in the same experimental design, however, the protein was localized in the nucleus ( Figure 3 ). These results indicate that the c.936G > A-type truncated NEIL1 protein, lacking an NLS, is localized in the cytoplasm, unlike the nuclear localization of the wild-type NEIL1 protein, suggesting that the excisional repair ability of the truncated-type NEIL1 protein is impaired, preventing oxidatively damaged bases in nuclear DNA from being corrected.

Fig. 2.

Detection of the NEIL1 splicing abnormality due to the c.936G > A change by in vivo splicing assay. ( A ) Detection of the abnormally spliced product from KATO-III cell line by RT–PCR analysis. The genotypes of TMK-1 and KATO-III cells are homozygous for the c.936G allele and heterozygous for the c.936G and c.936A alleles, respectively. Expression of the NEIL1 mRNA transcript was examined using the primers located in exons 10 and 11. GAPDH expression was also examined. The band detected by electrophoresis was excised from the gel and directly sequenced. Bands indicated by arrowheads I and II are PCR products retaining or not retaining intron 10, respectively. ( B ) Diagram showing the polypeptide sequence translated from the read-through transcript of intron 10. ( C ) Schema of a part of pALTER-MAX expression vectors containing exon 10, intron 10 and exon 11 of the NEIL1 gene. ( D ) In vivo splicing assay. The NEIL1-c.936G and -c.936A type expression vectors were transiently transfected into the NCI-H1299 cell line. At 24 h post-transfection, the cells were harvested, and RT–PCR was performed using the set of primers pAM-F/pAM-R, as indicated in (C), or a set of primers for GAPDH. PCR products were fractionated by agarose gel electrophoresis, and each band was directly sequenced. The 119-, 260- and 420-bp products correspond to a vector sequence without NEIL1 , a NEIL1 sequence without intron 10, and a NEIL1 sequence with intron 10, respectively.

Fig. 2.

Detection of the NEIL1 splicing abnormality due to the c.936G > A change by in vivo splicing assay. ( A ) Detection of the abnormally spliced product from KATO-III cell line by RT–PCR analysis. The genotypes of TMK-1 and KATO-III cells are homozygous for the c.936G allele and heterozygous for the c.936G and c.936A alleles, respectively. Expression of the NEIL1 mRNA transcript was examined using the primers located in exons 10 and 11. GAPDH expression was also examined. The band detected by electrophoresis was excised from the gel and directly sequenced. Bands indicated by arrowheads I and II are PCR products retaining or not retaining intron 10, respectively. ( B ) Diagram showing the polypeptide sequence translated from the read-through transcript of intron 10. ( C ) Schema of a part of pALTER-MAX expression vectors containing exon 10, intron 10 and exon 11 of the NEIL1 gene. ( D ) In vivo splicing assay. The NEIL1-c.936G and -c.936A type expression vectors were transiently transfected into the NCI-H1299 cell line. At 24 h post-transfection, the cells were harvested, and RT–PCR was performed using the set of primers pAM-F/pAM-R, as indicated in (C), or a set of primers for GAPDH. PCR products were fractionated by agarose gel electrophoresis, and each band was directly sequenced. The 119-, 260- and 420-bp products correspond to a vector sequence without NEIL1 , a NEIL1 sequence without intron 10, and a NEIL1 sequence with intron 10, respectively.

Fig. 3.

The distinct subcellular localization of the wild- and mutant-type of NEIL1 protein. The wild-type and mutant-type NEIL1 with the FLAG epitope tag at the C-terminus were expressed in NCI-H1299 cells and stained with anti-FLAG M2 as the first antibody and FITC-conjugated anti-mouse IgG as the second antibody. The immunofluorescence microscopic image of FITC (green)-stained cells showed the localization of exogenous NEIL1 protein. The nuclei were counterstained with DAPI (blue).

Fig. 3.

The distinct subcellular localization of the wild- and mutant-type of NEIL1 protein. The wild-type and mutant-type NEIL1 with the FLAG epitope tag at the C-terminus were expressed in NCI-H1299 cells and stained with anti-FLAG M2 as the first antibody and FITC-conjugated anti-mouse IgG as the second antibody. The immunofluorescence microscopic image of FITC (green)-stained cells showed the localization of exogenous NEIL1 protein. The nuclei were counterstained with DAPI (blue).

Reduction in NEIL1 expression in gastric cancer

We next assessed the levels of NEIL1 mRNA expression in 13 primary gastric cancers and corresponding non-cancerous mucosae by QRT–PCR using the PBGD housekeeping gene as a control. The ratio of the NEIL1 mRNA expression level in the tumor compared with the corresponding non-cancerous tissue (T/N ratio) was calculated in each case ( Figure 4 ). Six out of 13 (46%) of the primary gastric cancers exhibited a reduction in NEIL1 expression (0.5 > T/N ratio). A Wilcoxon signed-ranks test also showed that the NEIL1 mRNA level in the cancer tissue was significantly lower than that in the non-cancerous tissue ( P < 0.01). A significant reduction in NEIL1 expression in the cancer tissue compared with the non-cancerous tissue was also observed when GAPD was used as the control housekeeping gene (data not shown). The level of NEIL1 expression in KATO-III cells carrying the c.936G > A mutation was less than one-tenth lower than that in the non-cancerous gastric tissue sample from case no. 1. These results indicate that NEIL1 expression may be reduced in some primary gastric cancers, although the underlying mechanism of this phenomenon is presently unknown.

Fig. 4.

NEIL1 expression in gastric cancer and corresponding non-cancerous tissue. Expression of the NEIL1 mRNA transcript was measured by QRT–PCR. The relative amounts of NEIL1 transcript were standardized according to those of the PBGD transcript. The relative expression of each sample was calculated by comparing the expression of non-cancerous gastric tissue in a primary gastric cancer patient (case no. 1). The T/N ratio is shown below the case number.

Fig. 4.

NEIL1 expression in gastric cancer and corresponding non-cancerous tissue. Expression of the NEIL1 mRNA transcript was measured by QRT–PCR. The relative amounts of NEIL1 transcript were standardized according to those of the PBGD transcript. The relative expression of each sample was calculated by comparing the expression of non-cancerous gastric tissue in a primary gastric cancer patient (case no. 1). The T/N ratio is shown below the case number.

Discussion

Three mutations and two genetic polymorphisms in the NEIL1 gene locus were identified in the present study. When the Tg DNA glycosylase activity was compared among six types of recombinant NEIL1 proteins, p.Glu28del-type NEIL1, first identified in a primary gastric cancer, was found to exhibit an extremely low activity level. Moreover, the c.936G > A truncated NEIL1 protein, first detected in the KATO-III cell line, was shown to be associated with a splicing abnormality using an in vivo splicing assay. In contrast to the results for the wild-type NEIL1 protein, an immunofluorescence analysis showed that the c.936G > A-type truncated NEIL1 protein was localized in the cytoplasm, not the nucleus, suggesting that the truncated protein's ability to correct nuclear DNA damage via its excisional repair ability was likely to be impaired. An expression analysis revealed that NEIL1 mRNA expression was reduced in six of 13 (46%) of the primary gastric cancer specimens that were examined. These results suggest that low NEIL1 activities arising from mutations and reductions in expression may be involved in the pathogenesis of a subset of gastric cancers.

In this study, the NEIL1-p.Glu28del-type protein resulting from the c.82_84delGAG mutation was shown to exhibit a greatly reduced ability to cleave a double-stranded oilognucleotide containing a single 5 R ,6 S -Tg:A mispair. As mammalian NEIL1 protein has been shown to efficiently excise the substrate ( 23 , 26 ), the assay system used in this study was expected to be adequate to determine the level of the NEIL1 protein's repair ability. As expected, the repair ability of the p.Glu28del-type protein as well as the p.Pro2del-type protein, which is known to be an inactive form, was lower than that of the wild-type protein. However, the reason for the reduction in the repair ability was unclear, because the Glu28 residue is not located in a domain that is known to be necessary for the protein's repair function. The Glu28 residue may play an important role in the recognition of the DNA substrate, or some other related function. In the future, a crystal structure analysis of the NEIL1 protein alone and covalently complexed with DNA, in conjunction with the present finding that the p.Glu28del-type protein exhibited an extremely low activity level, would contribute to establishing further correlations between the structure and repair function of the NEIL1 protein.

The c.936G > A mutation caused a splicing abnormality that led to the production of an aberrant mRNA transcript encoding a truncated NEIL1 protein. Unlike the wild-type protein, the truncated NEIL1 protein was not localized in the nucleus, suggesting that the c.936G > A-type truncated protein could not repair nuclear DNA. KATO-III cells carrying this c.936G > A mutation also showed a reduction in NEIL1 mRNA expression, agreeing with the observation that NEIL1 mRNA expression was reduced in some primary gastric cancers in the present study. Thus, both genetic alterations may contribute to the low NEIL1 activity in the KATO-III cell line.

Regarding the prevalence of somatic mutations in human gastrointestinal cancer, Wang et al . ( 33 ) reported that only three non-synonymous somatic mutations, each in a different gene, were identified in ∼3.2 Mb of coding colon cancer DNA, containing 1811 exons from 470 genes. Considering this prevalence, the observation of three NEIL1 mutations in this study implies a relatively high prevalence of NEIL1 somatic mutations in gastric cancer. Gastric cells with a NEIL1 somatic mutation may have been selected during tumorigenesis. However, as we did not compare the prevalence of the NEIL1 mutation in a variety of human cancers, it is unknown whether its prevalence in gastric cancer is significantly higher than in other cancers. We are planning to investigate this issue in the near future.

A reduction in the DNA repair function of the NEIL1 protein in human cells, particularly, where severe oxidative stress is present, could lead to a high level of accumulated oxidative DNA damage. This situation might lead to an increase in the incidence of gene mutation in the cells. Moreover, a 20–35% reduction in the repair capacity of cells has been associated with a 4–6-fold higher risk of developing cancer ( 34 ). Thus, it is suggested that low NEIL1 activities caused by mutations and reduced mRNA expression may contribute to the alteration of genes involved in gastric carcinogenesis in a subset of gastric cancers.

We thank Dr M.Takao of the Tohoku University, Sendai, Japan, for valuable advice on the expression of the NEIL1 protein, Dr Y.Gao of the Hamamatsu University School of Medicine, Japan, for instruction on the use of the LightCycler instrument, and Ms C.Tatebayashi of Hamamatsu University School of Medicine for her technical assistance. This work was supported in part by a Grant-in-Aid from the Ministry of Health, Labour and Welfare for the 2nd-term Comprehensive 10-Year Strategy for Cancer Control and from the Ministry of Education, Culture, Sports, Science and Technology of Japan for Scientific Research on Priority Area and the 21st century COE program ‘Medical Photonics’.

References

1.
Beckman,K.B. and Ames,B.N. (
1997
) Oxidative decay of DNA.
J. Biol. Chem.
  ,
272
,
19633
–19636.
2.
Wallace,S.S. (
2002
) Biological consequences of free radical-damaged DNA bases.
Free Radic. Biol. Med.
  ,
33
,
1
–14.
3.
Kasai,H. and Nishimura,S. (
1991
) Formation of 8-hydroxyguanosine in DNA by oxygen radicals and its biological significance. In Sies,H. (ed.) Oxidative Stress: Oxidants and Antioxidants. Academic Press, London, pp. 99–116.
4.
Aburatani,H., Hippo,Y., Ishida,T. et al . (
1997
) Cloning and characterization of mammalian 8-hydroxyguanine-specific DNA glycosylase/apurinic, apyrimidinic lyase, a functional mutM homologue.
Cancer Res.
  ,
57
,
2151
–2156.
5.
Aspinwall,R., Rothwell,D.G., Roldan-Arjona,T. et al . (
1997
) Cloning and characterization of a functional human homolog of Escherichia coli endonuclease III.
Proc. Natl Acad. Sci. USA
  ,
94
,
109
–114.
6.
Lu,R., Nash,H.M. and Verdine,G.L. (
1997
) A mammalian DNA repair enzyme that excises oxidatively damaged guanines maps to a locus frequently lost in lung cancer.
Curr. Biol.
  ,
7
,
397
–407.
7.
Radicella,J.P., Dherin,C., Desmaze,C., Fox,M.S. and Boiteux,S. (
1997
) Cloning and characterization of hOGG1 , a human homolog of the OGG1 gene of Saccharomyces cerevisiae .
Proc. Natl Acad. Sci. USA
  ,
94
,
8010
–8015.
8.
Bjoras,M., Luna,L., Johnsen,B., Hoff,E., Haug,T., Rognes,T. and Seeberg,E. (
1997
) Opposite base-dependent reactions of a human base excision repair enzyme on DNA containing 7,8-dihydro-8-oxoguanine and abasic sites.
EMBO J.
  ,
16
,
6314
–6322.
9.
Roldan-Arjona,T., Wei,Y., Carter,K.C., Klungland,A., Anselmino,C., Wang,R., Augustus,M. and Lindahl,T. (
1997
) Molecular cloning and functional expression of a human cDNA encoding the antimutator enzyme 8-hydroxyguanine-DNA glycosylase.
Proc. Natl Acad. Sci. USA
  ,
94
,
8016
–8020.
10.
Shinmura,K., Kasai,H., Sasaki,A., Sugimura,H. and Yokota,J. (
1997
) 8-Hydroxyguanine (7,8-dihydro-8-oxoguanine) DNA glycosylase and AP lyase activities of hOGG1 protein and their substrate specificity.
Mutat. Res.
  ,
385
,
75
–82.
11.
Ikeda,S., Biswas,T., Roy,R., Izumi,T., Boldogh,I., Kurosky,A., Sarker,A.H., Seki,S. and Mitra,S. (
1998
) Purification and characterization of human NTH1, a homolog of Escherichia coli endonuclease III. Direct identification of Lys-212 as the active nucleophilic residue.
J. Biol. Chem.
  ,
273
,
21585
–21593.
12.
Shinmura,K. and Yokota,J. (
2001
) The OGG1 gene encodes a repair enzyme for oxidatively damaged DNA and is involved in human carcinogenesis.
Antioxid. Redox Signal.
  ,
3
,
597
–609.
13.
Kohno,T., Shinmura,K., Tosaka,M., Tani,M., Kim,S.R., Sugimura,H., Nohmi,T., Kasai,H. and Yokota,J. (
1998
) Genetic polymorphisms and alternative splicing of the hOGG1 gene, that is involved in the repair of 8-hydroxyguanine in damaged DNA.
Oncogene
  ,
16
,
3219
–3225.
14.
Audebert,M., Chevillard,S., Levalois,C. et al . (
2000
) Alterations of the DNA repair gene OGG1 in human clear cell carcinomas of the kidney.
Cancer Res.
  ,
60
,
4740
–4744.
15.
Hyun,J.W., Choi,J.Y., Zeng,H.H., Lee,Y.S., Kim,H.S., Yoon,S.H. and Chung,M.H. (
2000
) Leukemic cell line, KG-1 has a functional loss of hOGG1 enzyme due to a point mutation and 8-hydroxydeoxyguanosine can kill KG-1.
Oncogene
  ,
19
,
4476
–4479.
16.
Bruner,S.D., Norman,D.P. and Verdine,G.L. (
2000
) Structural basis for recognition and repair of the endogenous mutagen 8-oxoguanine in DNA.
Nature
  ,
403
,
859
–866.
17.
Sugimura,H., Kohno,T., Wakai,K. et al . (
1999
) hOGG1 Ser326Cys polymorphism and lung cancer susceptibility.
Cancer Epidemiol. Biomarkers Prev.
  ,
8
,
669
–674.
18.
Goode,E.L., Ulrich,C.M. and Potter,J.D. (
2002
) Polymorphisms in DNA repair genes and associations with cancer risk.
Cancer Epidemiol. Biomarkers Prev.
  ,
11
,
1513
–1530.
19.
Hazra,T.K., Izumi,T., Boldogh,I., Imhoff,B., Kow,Y.W., Jaruga,P., Dizdaroglu,M. and Mitra,S. (
2002
) Identification and characterization of a novel human DNA glycosylase for repair of cytosine-derived lesions.
Proc. Natl Acad. Sci. USA
  ,
99
,
3523
–3528.
20.
Takao,M., Kanno,S., Kobayashi,K., Zhang,Q.M., Yonei,S., van der Horst,G.T. and Yasui,A. (
2002
) A back-up glycosylase in Nth1 knock-out mice is a functional Nei (endonuclease VIII) homologue.
J. Biol. Chem.
  ,
277
,
42205
–42213.
21.
Morland,I., Rolseth,V., Luna,L., Rognes,T., Bjoras,M. and Seeberg,E. (
2002
) Human DNA glycosylases of the bacterial Fpg/MutM superfamily: an alternative pathway for the repair of 8-oxoguanine and other oxidation products in DNA.
Nucleic Acids Res.
  ,
30
,
4926
–4936.
22.
Bandaru,V., Sunkara,S., Wallace,S.S. and Bond,J.P. (
2002
) A novel human DNA glycosylase that removes oxidative DNA damage and is homologous to Escherichia coli endonuclease VIII.
DNA Rep.
  ,
1
,
517
–529.
23.
Rosenquist,T.A., Zaika,E., Fernandes,A.S., Zharkov,D.O., Miller,H. and Grollman,A.P. (
2003
) The novel DNA glycosylase, NEIL1, protects mammalian cells from radiation-mediated cell death.
DNA Rep.
  ,
2
,
581
–591.
24.
Kroeger,K.M., Hashimoto,M., Kow,Y.W. and Greenberg,M.M. (
2003
) Cross-linking of 2-deoxyribonolactone and its beta-elimination product by base excision repair enzymes.
Biochemistry
  ,
42
,
2449
–2455.
25.
Dou,H., Mitra,S. and Hazra,T.K. (
2003
) Repair of oxidized bases in DNA bubble structures by human DNA glycosylases NEIL1 and NEIL2.
J. Biol. Chem.
  ,
278
,
49679
–49684.
26.
Miller,H., Fernandes,A.S., Zaika,E., McTigue,M.M., Torres,M.C., Wente,M., Iden,C.R. and Grollman,A.P. (
2004
) Stereoselective excision of thymine glycol from oxidatively damaged DNA.
Nucleic Acids Res.
  ,
32
,
338
–345.
27.
Baik,S.C., Youn,H.S., Chung,M.H., Lee,W.K., Cho,M.J., Ko,G.H., Park,C.K., Kasai,H. and Rhee,K.H. (
1996
) Increased oxidative DNA damage in Helicobacter pylori -infected human gastric mucosa.
Cancer Res.
  ,
56
,
1279
–1282.
28.
Farinati,F., Cardin,R., Degan,P., Rugge,M., Mario,F.D., Bonvicini,P. and Naccarato,R. (
1998
) Oxidative DNA damage accumulation in gastric carcinogenesis.
Gut
  ,
42
,
351
–356.
29.
Tredaniel,J., Boffetta,P., Buiatti,E., Saracci,R. and Hirsch,A. (
1997
) Tobacco smoking and gastric cancer: review and meta-analysis.
Int. J. Cancer
  ,
72
,
565
–573.
30.
Kiyokawa,E., Takai,S., Tanaka,M., Iwase,T., Suzuki,M., Xiang,Y.Y., Naito,Y., Yamada,K., Sugimura,H. and Kino,I. (
1994
) Overexpression of ERK, an EPH family receptor protein tyrosine kinase, in various human tumors.
Cancer Res.
  ,
54
,
3645
–3650.
31.
Takezaki,T., Gao,C.M., Wu,J.Z. et al . (
2002
) hOGG1 Ser(326)Cys polymorphism and modification by environmental factors of stomach cancer risk in Chinese.
Int. J. Cancer
  ,
99
,
624
–627.
32.
Chabot,B. (
1996
) Directing alternative splicing: cast and scenarios.
Trends Genet.
  ,
12
,
472
–478.
33.
Wang,T.L., Rago,C., Silliman,N., Ptak,J., Markowitz,S., Willson,J.K., Parmigiani,G., Kinzler,K.W., Vogelstein,B. and Velculescu,V.E. (
2002
) Prevalence of somatic alterations in the colorectal cancer cell genome.
Proc. Natl Acad. Sci. USA
  ,
99
,
3076
–3080.
34.
Mohrenweiser,H.M. and Jones,I.M. (
1999
) Variation in DNA repair is a factor in cancer susceptibility: a paradigm for the promises and perils of individual and population risk estimation?
Mutat. Res.
  ,
400
,
15
–24.

Author notes

1First Department of Pathology and 2Department of Laboratory Medicine, Hamamatsu University School of Medicine, 1-20-1 Handayama, Hamamatsu, Shizuoka 431-3192, Japan and 3Department of International Island and Community Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima 890-8544, Japan