Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium AACR Conference on Molecular Diagnostics - 2008
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, S.-Y.
Right arrow Articles by Chen, Y.-M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, S.-Y.
Right arrow Articles by Chen, Y.-M. A.
[Cancer Research 64, 3617-3623, May 15, 2004]
© 2004 American Association for Cancer Research

Regular Articles

Glycine N-Methyltransferase Tumor Susceptibility Gene in the Benzo(a)pyrene-Detoxification Pathway

Shih-Yin Chen1, Jane-Ru Vivan Lin1, Ramalakshmi Darbha2, Pinpin Lin3, Tsung-Yun Liu4 and Yi-Ming Arthur Chen1

1 Division of Preventive Medicine, Institute of Public Health, and AIDS Prevention and Research Centre, National Yang-Ming University, Taipei, Taiwan, Republic of China; 2 Biomolecular Structure Section, Macromolecular Crystallography Laboratory, National Cancer Institute-Frederick, Frederick, Maryland; 3 Institute of Toxicology, Chung-Shan Medical University, Taichung, Taiwan, Republic of China; and 4 Department of Medical Research, Taipei Veterans General Hospital, Taiwan, Republic of China


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Glycine N-methyltransferase (GNMT) affects genetic stability by (a) regulating the ratio of S-adenosylmethionine to S-adenosylhomocystine and (b) binding to folate. Based on the identification of GNMT as a 4 S polyaromatic hydrocarbon-binding protein, we used liver cancer cell lines that expressed GNMT either transiently or stably in cDNA transfections to analyze the role of GNMT in the benzo(a)pyrene (BaP) detoxification pathway. Results from an indirect immunofluorescent antibody assay showed that GNMT was expressed in cell cytoplasm before BaP treatment and translocated to cell nuclei after BaP treatment. Compared with cells transfected with the vector plasmid, the number of BaP-7,8-diol 9,10-epoxide-DNA adducts that formed in GNMT-expressing cells was significantly reduced. Furthermore, the dose-dependent inhibition of BaP-7,8-diol 9,10-epoxide-DNA adduct formation by GNMT was observed in HepG2 cells infected with different multiplicities of infection of recombinant adenoviruses carrying GNMT cDNA. According to an aryl hydrocarbon hydroxylase enzyme activity assay, GNMT inhibited BaP-induced cytochrome P450 1A1 enzyme activity. Automated BaP docking using a Lamarckian genetic algorithm with GNMT X-ray crystallography revealed a BaP preference for the S-adenosylmethionine-binding domain of the dimeric form of GNMT, a novel finding of a cellular defense against potentially damaging exposures. In addition to GNMT, results from docking experiments showed that BaP binds readily with other DNA methyltransferases, including HhaI, HaeIII, PvuII methyltransferases and human DNA methyltransferase 2. We therefore hypothesized that BaP-DNA methyltransferase and BaP-GNMT interactions may contribute to carcinogenesis.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Benzo(a)pyrene (BaP) is a carcinogen produced by organic material combustion. Workers in gas generation and steel plants and individuals engaged in aluminum reduction and roofing have higher cancer risks associated with long-term exposure to various polycyclic aromatic hydrocarbons including BaP (1) . After diffusing into a cell, BaP binds with an aryl hydrocarbon receptor (AhR), translocates into the cell’s nuclei, and transactivates the cytochrome P450 1A1 (CYP1A1) gene (2, 3, 4) . A metabolic BaP product known as BaP-7,8-diol 9,10-epoxide (BPDE) is capable of forming DNA adducts and triggering mutagenesis (5) .

Glycine N-methyltransferase (GNMT; EC2.1.1.20), a protein with multiple functions, affects genetic stability by (a) regulating the ratio of S-adenosylmethionine (SAM) to S-adenosylhomocystine and (b) binding to folate (6 , 7) . We have reported previously (8 , 9) on diminished GNMT expression levels in both human hepatocellular carcinoma cell lines and tumor tissues. In previous projects, we localized the human GNMT gene to the 6p12 chromosomal region and characterized its polymorphism (10 , 11) . Genotypic analyses of several human GNMT gene polymorphisms showed a loss of heterozygosity in 36–47% of the genetic markers in hepatocellular carcinoma tissues (11) . In this study, we evaluated the effects of GNMT on liver cells treated with BaP in a transient transfection system or with stably expressed clones, based on the identification of GNMT as a 4 S polycyclic aromatic hydrocarbon-binding protein (12) . We also used automated docking with a Lamarckian genetic algorithm (LGA) to elucidate GNMT-BaP interaction. In light of results showing a BaP binding preference for the SAM-binding domain of GNMT, we expanded our BaP docking experiments to include other SAM-dependent methyltransferases; our results show that BaP interacts readily with DNA methyltransferases that use cytosine as a target atom.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Lines and Culture.
Two hepatocellular carcinoma cell lines [Huh7 (13) and HA22T/VGH (14) ] and one human hepatoblastoma cell line [HepG2 (15) ] were used in this study. Cells were cultured in DMEM (GIBCO BRL, Grand Island, NY) with 10% heat-inactivated fetal bovine serum (HyClone, Logan, UT), penicillin (100 IU/ml), streptomycin (100 µg/ml), nonessential amino acids (0.1 mM), fungizone (2.5 mg/ml), and L-glutamine (2 mM) in a humidified incubator with 5% CO2.

Construction of pGNMT, pGNMT-antisense, and pGNMT-His-short Plasmids.
To construct plasmid pGNMT containing the cytomegalovirus (CMV) promoter and GNMT cDNA fragment, we used plasmid-pFLAG-CMV-5 (Kodak, Rochester, NY) as a vector and the pBluescript-GNMT-9-1-2 phagemid (8) as the PCR template for generating the insert. A 0.9-kb DNA fragment containing the GNMT cDNA sequence and restriction enzyme sites on both ends was amplified. All PCR conditions were as recommended by the manufacturer (Perkin-Elmer, Norwalk, CT), with two exceptions (2 mM MgCl2 and 150 nM primer). Twenty amplification cycles were performed using Perkin-Elmer Amplitaq Gold Taq DNA polymerase with DNA Thermal Cycler. Each PCR cycle entailed a primer annealing step at 60°C for 30 s and an extension step at 72°C for 30 s. The upstream primer (5'-gcggaattcATGGTGGACAGCGTGTAC-3') included a 3-bp "clamp" (gcg) at the 5' end followed by a single restriction enzyme site (EcoRI) and the GNMT cDNA sequence. The downstream primer (5'-gcggaattcGTCTGTCCTCTTGAGCAC-3') contained a similar structural motif as the upstream primer; however, it consisted of a negative strand sequence from the terminal region of the GNMT cDNA. Immediately after amplification, SDS (0.1%) and EDTA (5 mM) were added to the PCR reaction; DNA was precipitated with 2.5 M ammonium acetate and 70% ethanol. After digestion with EcoRI, the DNA fragment was isolated by elution from the agarose gel and ligated to EcoRI-digested pFLAG-CMV-5.

We used two primers (F1, 5'-gcggaattcATGGTGGACAGCGTGTAC-3; R1, 5'-gcggaattcTGTACTCGGCGGTGCGGC-3) to construct an antisense-GNMT plasmid (pGNMT-antisense) for amplifying a 136-bp DNA fragment from phagemid pBluescript-GNMT-9-1-2 (8) . The fragment contained an antisense sequence spanning the GNMT translational starting site and two restriction enzyme sites (EcoRI and BamHI) at its terminals. Cloning procedures were similar to those described for pGNMT.

To express the GNMT recombinant protein (RP) in Escherichia coli, we constructed plasmid pGNMT-His-short. The large S-tag DNA fragment was excised from the pGNMT-His (9) using EcoRI and NdeI restriction enzymes (Stratagene, La Jolla, CA); the resulting plasmid DNA was religated after a Klenow reaction. Plasmid DNA sequences were confirmed with a DNA sequencer equipped with a dye terminator cycle sequencing core kit (Applied Biosystems Model 373A, Version 1.0.2; Applied Biosystems, Foster City, CA).

GNMT RP Expression and Purification.
pGNMT-His-short was used to transform the E. coli BL21 bacteria used for isopropyl-1-thio-ß-D-galactopyranoside induction (induction time, 3 h; bacterial culture absorbance, 0.6–0.7). GNMT RP purification was performed using a Ni2+-charged histidine-binding resin column according to the manufacturer’s guidelines (Novagen, Madison, WI). RP concentration was measured with a BCA protein assay (Pierce, Rockford, IL); purity was tested by running samples on a 12.5% SDS-polyacrylamide mini-gel (Bio-Rad Laboratories, Richmond, CA).

Transfection.
All plasmid DNA samples were prepared using Qiagen mega kits (Hilden, Germany). Standard calcium phosphate coprecipitation methodology (16) was used to transfect cultured cells from various liver cancer cell lines with plasmid DNA. Forty-eight h after transfection, cells were treated with different concentrations (1–10 µM) of BaP (Sigma-Aldrich, Steinheim, Germany) dissolved in DMSO (Nacalaitesque, Osaka, Japan) for 16 h. Treated cells were subjected to either indirect immunofluorescent antibody assay or 32P postlabeling. To produce a negative control, 0.1% DMSO was added to the cell culture.

Establishing Stable Clones Expressing GNMT.
Using calcium phosphate methodology, HepG2 cells were cotransfected with pGNMT and pTK-Hyg (Clontech, Palo Alto, CA) plasmid DNAs. Cells were placed in a selection medium containing 300 µg/ml hygromycin (17) . More than 12 clones were selected, and GNMT expression was analyzed with a Western blot (WB) assay using cell lysate collected from each clone. Among them, SCG2--1-1 and SCG2-1-11 were chosen for further study based on their expression level of GNMT. SCG2-neg, a stable clone selected from HepG2 cells cotransfected with pFLAG-CMV-5 and pTK-Hyg plasmids, was also used as a control in this study.

Indirect Immunofluorescent Antibody Assay.
Cultured HA22T/VGH or Huh7 cells were placed on cover slides, treated with 10 µM BaP or 0.1% DMSO, fixed with solution I (4% paraformaldehyde and 400 mM sucrose in PBS) at 37°C for 30 min, fixed with solution II (fixing solution I plus 0.5% Triton X-100) at room temperature for 15 min, and fixed with blocking buffer (0.5% BSA in PBS) at room temperature for 1 h. After washing with PBS, the slides were allowed to react with various primary antibodies at 4°C overnight. The two antibodies were anti-FLAG monoclonal antibody (1:500 dilution; Kodak) and rabbit anti-GNMT antiserum-R4 (1:200 dilution; Ref. 12 ). FITC-conjugated antimouse IgG and tetramethylrhodamine isothiocyanate-conjugated antirabbit IgG (Sigma-Aldrich) were used as secondary antibodies. After four washes with PBS, slides were mounted and observed using a confocal fluorescence microscope (TCS-NT, Hilden, Germany). DNA was stained with Hoechst H33258 (Sigma-Aldrich) to localize cell nuclei.

Generating Adenovirus Carrying GNMT cDNA (Ad-GNMT).
To construct a GMNT recombinant adenovirus controlled by a CMV promoter, pGEX-GNMT (9) was digested with XhoI (filled-in) and BamHI before insertion into the XbaI (filled-in) and BamHI sites of pBluescript SK(–) (Stratagene). GNMT cDNA was also cloned into the HindIII and NotI sites of pAdE1CMV/pA (18) , a shuttle vector containing the left arm of a virus genome, to generate pXCMV-GNMT. A recombinant adenovirus appeared within 7–12 days after the cotransfection of pXCMV-GNMT and pJM17 (18) into 293 cells. Individual virus clones were isolated and identified using PCR with primer sets specific to the adenoviral sequence (18) , the insertion flanking regions (18) , and the GNMT cDNA (8) . Virus titer was determined via the plaque assay method described above (18) .

32P Postlabeling and Five-Dimensional TLC for Quantifying BPDE-DNA Adducts.
We used SCG2 cells and hepatocellular carcinoma cell lines transiently transfected with pGNMT plasmid DNA for 48 h for our experiments. DNA was extracted from cells treated with 10 µM BaP or 0.1% DMSO (control) for 16 h (19) and digested with micrococcal endonuclease and spleen phosphodiesterase in succinate buffer (20 mM sodium succinate and 10 mM CaCl2) for 3 h at 37°C. The resulting 3' nucleotides were further extracted twice with butanol solution and labeled with [{gamma}-32P]ATP with T4 kinase in labeling buffer at 38°C for 30 min. Five-dimensional TLC was used to elucidate labeled DNA adducts (20) . Relative adduct level was calculated as cpm in adducted nucleotides/(cpm in total nucleotides x dilution).

Aryl Hydrocarbon Hydroxylase Assay.
To measure CYP1A1 enzyme activity, approximately 100 µg of cellular homogenates were incubated with reactive solution (100 mM HEPES, 0.4 mM NADPH, 1 mM MgCl2, and 20 µM BaP) at 37°C for 10 min. Supernatant protein concentrations were determined using a Bio-Rad protein assay kit (Hercules, CA). Reactions were stopped by the addition of acetone; extraction was performed with hexane and 1N NaOH. NaOH fractions were read on a spectrofluorometer (Hitahi Instrument F4500) with excitation and emission wavelengths of 396 nm and 522 nm, respectively. Reaction product (3-hydroxy-BaP) concentrations were calculated by comparison with a standard; procedural details are given in Ref. 21 .

WB Assay.
WB was used to detect GNMT in transfected cells or SCG2 clones. Anti-GNMT monoclonal antibody 14-1 was used to detect GNMT (9) . A detailed description of WB procedures is presented in Ref. 22 .

LGA Dockings.
LGA was used to elucidate interaction sites between BaP and various forms of GNMT. Autodock 3.0 software was used to identify the most favorable ligand binding interactions. We empirically determined van der Waals hydrogen bonding, hydrophobic desolvations, and electrostatic and torsional free energy to reproduce ligand-protein binding free energies (23) . We used X-ray crystallography data from rat GNMT for docking due to its 91% amino acid sequence homology with human GNMT (24 , 25) . We analyzed interactions between BaP and methyltransferase-1VID (26) , 1HMY (27) , 2ADM (28) , 1DCT (29) , 1BOO (30) , 2DPM (31) , 1EG2 (32) , and 1G55 (33) . Parameters were as follows: 10 runs; a population size of 50; and a run-termination criterion of a maximum of 27,000 generations or 2.5 x 105 energy evaluations, whichever came first. A root mean square deviation conformational clustering tolerance of 0.5 Å was calculated from the ligand’s crystallographic coordinates. Procedural details are available in Ref. 34 .

GNMT Enzyme Activity Assay.
GNMT RP purified from a Ni2+-charged histidine-binding resin column was used for an enzyme activity assay. GNMT RP (10 mg) was mixed with 10, 50, or 100 µM BaP or DMSO solvent (control) at room temperature for 60 min before treatment with 100 µl of 100 mM Tris buffer (pH 7.4) containing 50 mM glycine, 0.23 mM SAM, and 2.16 µM S-adenosyl-L-[methyl-3H]methionine (76.4 Ci/mmol). After incubation at 37°C for 30 min, individual reactions were terminated by the addition of a 50-µl mixture of 10% trichloroacetic acid and 5% activated charcoal. Each reaction was performed in triplicate. This procedure has been described in detail by Cook and Wagner (35) .


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
GNMT Nuclear Translocation Was Induced by BaP in Both HA22T/VGH and Huh7 Cells.
GNMT was expressed in the cytoplasm of HA22T/VGH cells 48 h after transfection with pGNMT DNA (double indirect immunofluorescent antibody assay with both rabbit anti-GNMT antiserum and mouse anti-FLAG monoclonal antibody; Fig. 1, A and BCitation ). Similar results were noted in control Huh7 cells treated with DMSO solvent (Fig. 1, C and D)Citation . In contrast, GNMT proteins were only partly translocated into the nuclei of Huh7 cells treated with 10 µM BaP for 16 h (Fig. 1, E and F)Citation . DNA was stained with Hoechst H33258 to localize cell nuclei (Fig. 1, D and F)Citation .



View larger version (55K):
[in this window]
[in a new window]
  		Fig. 1. Nuclear translocation of glycine N-methyltransferase (GNMT) after cell treatment with benzo(a)pyrene. A and B, a double indirect immunofluorescent antibody assay was performed on HA22T/VGH cells transfected with pGNMT. Antisera: A, rabbit anti-GNMT antibody; B, mouse anti-FLAG antiserum. C–F, indirect immunofluorescent antibody assay on Huh7 cells transfected with pGNMT and treated with either DMSO solvent (C and D) or benzo(a)pyrene (E and F) before being fixed and reacted with mouse anti-FLAG antiserum. Immunofluorescent staining was performed with rhodamine-conjugated goat antirabbit antibodies (A) or FITC-conjugated rabbit antimouse antibodies (B–F). Nuclei were stained with Hoechst H33258.

 
Inhibitory Effect of GNMT on BPDE-DNA Adduct Formation.
32P postlabeling and five-dimensional TLC were used to quantify BPDE-DNA adduct formation. After treatment with 10 µM BaP for 16 h, BPDE-DNA adduct formation in HepG2, Huh7, and HA22T/VGH cells transfected with pGNMT decreased 52.8%, 13.5%, and 20.7%, respectively, compared with cells transfected with the vector plasmid (Table 1)Citation . Because the inhibitory effect of GNMT on BPDE-DNA adduct formation was strongest in the HepG2 cells, we used that cell line as the target in subsequent experiments. HepG2 cell DNA transfection efficiency was approximately 30%. In addition to pGNMT, a plasmid containing an antisense GNMT sequence was constructed for the purpose of verifying the specificity of the GNMT effect. After BaP treatment, a 34.1% decrease was noted in BPDE adducts formed in pGNMT-transfected cells compared with cells transfected with the vector control plasmid (Fig. 2ACitation , 3Citation and 4Citation ). In contrast, a 29.2% increase in BPDE adducts was noted in HepG2 cells transfected with pGNMT-antisense (Fig. 2ACitation , 5Citation ). Quantities of BPDE-DNA adducts formed in cells transfected with equal amounts (20 µg) of pGNMT and pGNMT-antisense were approximately equal to those formed in the vector control cells (Fig. 2ACitation , 6Citation ). GNMT expression in different transfection experiments and the effects of antisense GNMT cDNA plasmid construct (pGNMT-antisense) were verified by WB assays with mouse anti-GNMT monoclonal antibody. As shown in Lane 4 of Fig. 2BCitation , GNMT was not detected in the lysates of cells transfected with equal amounts of pGNMT and pGNMT-antisense.


View this table:
[in this window]
[in a new window]
  		Table 1 Effects of GNMTa expression on BPDE-DNA adduct formation in HCC cell lines

 


View larger version (39K):
[in this window]
[in a new window]
  		Fig. 2. Effects of glycine N-methyltransferase (GNMT) on benzo(a)pyrene-7,8-diol 9,10-epoxide (BPDE)-DNA adduct formation. A, amount (relative adduct level) of BPDE-DNA adducts using a combination of 32P postlabeling and five-dimensional TLC. 1, DMSO solvent control; 2, mock transfection; 3, cells transfected with 40 µg of control (pFLAG-CMV-5) vector; 4, cells transfected with 40 µg of pGNMT; 5, cells transfected with 40 µg of pGNMT-antisense; 6, cells cotransfected with 20 µg of pGNMT and 20 µg of pGNMT-antisense. DNA adduct quantities/108 nucleotides (relative adduct level) were as follows: 1, 0; 2, 1031.7; 3, 1092.4; 4, 719.8; 5, 1411.3; 6, 1079.7. B, Western blot analysis of GNMT expression in HepG2 cells transfected with the control (pFLAG-CMV-5) vector (Lane 1), pGNMT (Lane 2), pGNMT-antisense (Lane 3), or pGNMT/pGNMT-antisense (Lane 4). Bottom row shows ß-actin expression levels for the four experiments. C, amounts of of BPDE-DNA adducts in HepG2, SCG2-1-1, and SCG2-1-11 cells treated with 1 or 10 µM benzo(a)pyrene (BaP). 1 and 4, HepG2 cells treated with 1 or 10 µM BaP; 2 and 5, SCG2-1-1 cells treated with 1 or 10 µM BaP; 3 and 6, SCG2-1-11 cells treated with 1 or 10 µM BaP. DNA adduct quantities/108 nucleotides (relative adduct level) were as follows: 1, 161.9; 2, 26.4; 3, 55.2; 4, 682.1; 5, 354.9; 6, 506.5. D, Western blot analysis of GNMT expression in HepG2 (Lane 1), SCG2-1-1 (Lane 2), and SCG2-1-11 (Lane 3) cells. Twenty µg of cell lysates from each cell line were used for the PAGE. Bottom row shows ß-actin expression levels for the four experiments.

 


View larger version (48K):
[in this window]
[in a new window]
  		Fig. 3. Effects of glycine N-methyltransferase (GNMT) expression on benzo(a)pyrene (BaP)-7,8-diol 9,10-epoxide-DNA adduct formation in HepG2 cells infected with Ad-GFP or various MOIs of Ad-GNMT. A: 1, cells infected with Ad-GFP and treated with DMSO solvent; 2, cells infected with Ad-GFP and treated with BaP; 3, cells infected with MOI = 100 of Ad-GNMT and treated with BaP; 4, cells infected with MOI = 250 of Ad-GNMT and treated with BaP; 5, cells infected with MOI = 1000 of Ad-GNMT and treated with BaP. DNA adduct quantities/108 nucleotides (relative adduct level) were as follows: 1, 0; 2, 638.9; 3, 514.2; 4, 405.3; 5, 224.3. B, Western blot analysis of GNMT expression in the same experiment. Lane 1, Ad-GFP control; Lane 2, Ad-GNMT (MOI = 100); Lane 3, Ad-GNMT (MOI = 250); Lane 4, Ad-GNMT (MOI = 1000).

 


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 4. Cytochrome P450 1A1 (CYP1A1) enzyme activity induced by benzo(a)pyrene (BaP) in SCG2-neg and SCG2-1-1 cells as measured by an aryl hydrocarbon hydroxylase (AHH) assay. 1–4, CYP1A1 activity in SCG2-neg cells; 5–8, CYP1A1 activity in SCG2-1-1 cells. Treatments were as follows: 1 and 5, DMSO solvent; 2 and 6, 3 µM BaP; 3 and 7, 6 µM BaP; 4 and 8, 9 µM BaP. The CYP1A1 enzyme activity [means (in pmol/mg/min) and SDs (in parentheses)] was as follows: 1, 14.5 (0.27); 2, 24.47 (0.14); 3, 41.5 (1.42); 4, 71.3 (1.75); 5, 16.2 (3.6); 6, 20.1 (1.5); 7, 27.7 (1.2); 8, 36.2 (1.7).

 


View larger version (43K):
[in this window]
[in a new window]
  		Fig. 5. Model of benzo(a)pyrene (BaP) docking with dimeric and tetrameric forms of GNMT using the Lamarckian genetic algorithm. A, BaP (red) docked with S-adenosylhomocystine (white)-bound tetrameric form of rat glycine N-methyltransferase [GNMT (cyan; Protein Data Bank code 1D2H)]. B, BaP (red) docked with the dimeric form of rat GNMT (yellow; Protein Data Bank code 1D2C). C, dimeric form of GNMT (yellow) superimposed on tetrameric form of GNMT (cyan). GNMT amino acid residues (Ile34, Thr37, Gly137, His142, and Leu240 of one dimeric subunit and Glu15 of another) in close proximity to several BaP carbon atoms are indicated based on Protein Data Bank code 1D2C and BaP docking model.

 


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 6. Inhibition of glycine N-methyltransferase enzyme activity by benzo(a)pyrene (BaP). Glycine N-methyltransferase enzyme activity was measured as 2810.8 ± 73.7 nmol/h/µg for treatment with DMSO solvent, 1563.3 ± 127.4 nmol/h/µg for treatment with 10 µM BaP, 1069.5 ± 124.2 nmol/h/µg for treatment with 50 µM BaP, and 1083.3 ± 175.9 nmol/h/µg for treatment with 100 µM BaP. Each reaction set was performed in triplicate, as were individual experiments.

 
We used two stable clones (SCG2-1-1 and SCG2-1-11) from HepG2 cells transfected with pGNMT in the same experiments described above. Results from a Northern blot assay indicate that copy numbers (per cell) of GNMT cDNA present in SCG2-1-1 and SCG2-1-11 cells were 3 and 1, respectively (data not shown). Results from a WB assay showed that the GNMT expression level in the SCG2-1-1 cells was nearly three times that in the SCG2-1-11 cells (Fig. 2DCitation , Lanes 2 and 3). After treating the SCG2-1-1 and SCG2-1-11 cells with 1 or 10 µM BaP, BPDE-DNA adduct formation inhibition was proportional to GNMT-expression levels under both treatment conditions (Fig. 2C)Citation .

We performed the same experiment using an adenovirus carrying GNMT cDNA (Ad-GNMT). A positive linear relationship was noted between the multiplicities of infection (MOIs) of the Ad-GNMT and BPDE-DNA-adduct formation inhibition (Fig. 3)Citation . Compared with Ad-GFP-control-infected cells, the Ad-GNMT MOIs increased from 100 to 250 to 1000, and BPDE-DNA adduct formation decreased 19.5%, 36.6%, and 61.8%, respectively (Fig. 3A)Citation . GNMT expression levels in HepG2 cells infected with MOI = 100 of Ad-GFP control and MOI = 100, 250, and 1000 of Ad-GNMT were analyzed by WB; results are shown in Fig. 3BCitation , Lanes 1–4.

GNMT Effect on CYP1A1 Enzyme Activity Induced by BaP.
SCG2-1-1 and SCG2-neg cells were treated with varying concentrations of BaP for 16 h before using aryl hydrocarbon hydroxylase assay to measure their cellular CYP1A1 enzyme activity. CYP1A1 activity in cells treated with 3, 6, and 9 µM BaP was 24.5, 41.5 and 71.3 pmol/mg/min for SCG2-neg cells, respectively, and 20.1, 27.7, and 36.2 pmol/mg/min for SCG2-1-1 cells, respectively (Fig. 4)Citation . For cells treated with 9 µM BaP, this represents a 45% reduction in CYP1A1 enzyme activity in GNMT-expressing cells (i.e., SCG2-1-1) compared with SCG2-neg cells.

Modeling GNMT-BaP Interaction.
We used a LGA to predict physical GNMT-BaP interaction. Again, due to its 91% homology with human GNMT proteins, rat GNMT X-ray crystallography was used for the BaP docking experiments. As shown in Fig. 5, A and BCitation , we found that BaP binds with both dimeric (yellow) and tetrameric (cyan) forms of GNMT but that it prefers binding with the dimeric form (Protein Data Bank code 1D2C). This cluster is located at the intersection of the SAM- and S-adenosylhomocystine-binding sites (Table 2Citation ; Fig. 5BCitation ). The low (–9.10 kcal/mol) binding energy between the dimeric form of GNMT and BaP suggests that BaP may displace the SAM position; the high (254.9 kcal/mol) binding energy of BaP with a GNMT dimer already bound with SAM (Protein Data Bank code 1XVA) suggests that BaP and SAM are in competition for binding with GNMT (Table 2)Citation . Accordingly, several GNMT amino acid residues (including Thr37, Gly137, and His142 of one dimer subunit and Glu15 of another subunit) are in close proximity to BaP (Fig. 5C)Citation .


View this table:
[in this window]
[in a new window]
  		Table 2 Lamarckian genetic algorithm dockings of GNMTa protein and BaP molecules

 
BaP-Induced Inhibition of GNMT Enzyme Activity.
Based on the inference that BaP can bind with GNMT, we studied the potential effects of BaP on GNMT enzyme activity by constructing plasmid pGNMT-His-Short to express a His-tag-GNMT RP in E. coli. GNMT RP purified from a Ni2+-charged histidine-binding resin column was used for our analysis. As shown in Fig. 6Citation , GNMT enzyme activity from reactions containing 10 and 50 µM BaP decreased 44% and 62%, respectively, compared with the DMSO control.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We used an indirect immunofluorescent antibody assay to demonstrate the power of BaP to induce the nuclear translocation of GNMT. Our results show that GNMT not only inhibits BPDE-DNA adduct formation but also down-regulates CYP1A1 enzyme activity; conversely, BaP also inhibits GNMT enzyme activity. Finally, we used a docking experiment to show the exact location of BaP-GNMT interaction. These results represent a novel finding of a cellular defense mechanism against potentially damaging forms of exposure. We confirmed the inhibition of BPDE-DNA adduct formation by GNMT via transient transfection, stable clone selection, and adenovirus infection systems, with consistent results throughout. An antisense construct for GNMT cDNA was used to demonstrate interaction specificity (Fig. 2A)Citation , and WB assays were used to monitor GNMT expression levels in various gene transfer experiment sets. The dose-dependent inhibitory effect of GNMT on BPDE-DNA adduct formation was further elaborated with HepG2 stable clones and a recombinant adenovirus carrying GNMT cDNA (Figs. 2CCitation and 3A)Citation .

Many polycyclic aromatic hydrocarbons induce cytochrome P450 expression through an AhR-dependent pathway (36) . After diffusing into a cell, BaP binds with AhR and translocates into the nuclei, where BaP-AhR heterodimers form complexes with Ah receptor nuclear translocator (Arnt) proteins (2) . The BaP-AhR-Arnt complexes then transactivate the CYP1A1 gene via interaction with its xenobiotic responsive element in the promoter region (37) . In addition to the inhibition of BPDE-DNA adduct formation, our results show that GNMT is capable of reducing CYP1A1 enzyme activity induced by BaP (Fig. 4)Citation . Foussat et al. (38) used AhR-deficient transgenic mice to demonstrate that GNMT is not a transcriptional activator of the CYP1A1 gene (38) . Preliminary data from our real-time PCR analysis showed that after BaP treatment, CYP1A1 gene expression was reduced by approximately 20% in SCG2-1-1 cells compared with HepG2 cells.5

Previous research has shown that the tetrameric form of rat GNMT acts as an enzyme and that the dimeric form of rat GNMT is capable of binding with polycyclic aromatic hydrocarbons (39) . In this study, we used LGA and a scoring function for estimating binding-related free energy change to locate possible sites for interactions between BaP and various forms of GNMT; we used X-ray crystallography data for rat GNMT for this purpose. Our results indicate that (a) the BaP-binding domain is located at the substrate (SAM)-binding site of GNMT and (b) BaP prefers binding with the dimeric form of GNMT. The R175K mutant form of the GNMT tetramer (Protein Data Bank code 1D2G) was used to demonstrate that although the R/K residue is near the binding site (~5 Å from the SAM position), it exerts practically no effect on GNMT-BaP cluster formation (Table 2)Citation . In comparison, the presence of an acetate ion favors the formation of the second preferred cluster in GNMT-SAM binding in the 1XVA crystal structure (Table 2Citation , final entry). It has been demonstrated that of various search systems, the LGA method is the most likely to locate crystallographic structures (23) . Heavily populated clusters usually correspond to crystallographically determined positions that show 0.2–0.8 Å root mean square differences from the crystal structures. For most ligands, our docking simulation predicted single binding modes that matched crystallographic binding modes within 1.0 Å root mean square deviation (23) . We therefore suggest that the LGA is a reliable method for predicting the bound conformation of a ligand to its macromolecular target. BaP-GNMT interaction was also confirmed by a functional assay showing that GNMT enzyme activity was reduced nearly 50% in the presence of BaP (Fig. 6)Citation .

Because BaP prefers binding with the SAM-binding domain of GNMT, we used the LGA to study interactions between BaP and eight other SAM-dependent methyltransferases: catechol O-methyltransferase; HhaI DNA methyltransferase; TaqI DNA methyltransferase; HaeIII DNA methyltransferase; PvuII DNA methyltransferase; DpnII DNA methyltransferase; RsrI DNA methyltransferase; and DNA methyltransferase 2. Our results show that BaP was capable of binding with the HhaI, HaeIII, and PvuII DNA methyltransferases and DNA methyltransferase 2, but not with catechol O-methyltransferase or TaqI, DpnII, and RsrI DNA methyltransferases (Table 3)Citation . It is interesting to note that the target atom of all of the BaP-preferred DNA methyltransferases is cytosine and not adenine (40) . To our knowledge, this is the first evidence suggesting that an environmental carcinogen such as BaP has the potential to interact with different DNA methyltransferases. In light of evidence showing that the induction of GNMT enzyme activity by all-trans-retinoic acid causes DNA hypomethylation in rat hepatocytes (22) , we suggest that BaP may affect DNA methylation via interactions with DNA methyltransferase and GNMT and thus contribute to a carcinogenic pathway.


View this table:
[in this window]
[in a new window]
  		Table 3 Lamarckian genetic algorithm dockings of some SAMa-dependent methyltransferases and BaP moleculesb

 


   ACKNOWLEDGMENTS
 
We thank Prof. Chi-Hong Lin (National Yang-Ming University) for help with confocal microscopy, Drs. Jer-Tsong Hsieh (University of Texas Southwestern Medical Center) and Hong-Cheng Hsiao (Kaohsiung Veterans General Hospital) for help with Ad-GNMT generation, Dr. Wey-Jing Lin (National Yang-Ming University) for help with the GNMT enzyme activity assay, Jon Lindemann for help in editing the manuscript, and members of the Division of Preventive Medicine of the Institute of Public Health at National Yang-Ming University for helpful feedback and technical support. Ramalakshmi Darbha thanks her supervisor, Dr. Xinhua Ji, for permission to use the facilities at the Macromolecular Crystallography Laboratory, National Cancer Institute-Frederick, (Frederick, MD).


   FOOTNOTES
 
Grant support: Grant NSC 92-3112-B-010-001 from the National Research Program for Genomic Medicine of the National Science Council, Republic of China.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Yi-Ming Arthur Chen, Institute of Public Health, National Yang-Ming University, Shih-Pai District, Taipei, Taiwan 112, Republic of China. Phone: 886-2-28267193; Fax: 886-2-28270576; E-mail: arthur{at}ym.edu.tw

5 C. M. Lee, S. Y. Chen, C. Y. Huang, Y. C. G. Lee, Y. M. A. Chen, manuscript in preparation. Back

Received 11/30/03. Revised 3/ 3/04. Accepted 3/ 5/04.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Fischman ML, Cadman EC, Desmond S. Occupational cancer LaDou J eds. . Occupational medicine, 182-208, Appleton and Lange Norwalk, CT 1990.
  2. Whitlock JP, Jr., Okino ST, Dong L, et al Cytochromes P450 5: induction of cytochrome P4501A1: a model for analyzing mammalian gene transcription. FASEB J, 10: 809-18, 1996.[Abstract]
  3. Foldes RL, Hines RN, Ho KL, et al 3-Methylchlanthrene-induced expression of the cytochrome P-450c gene. Arch Biochem Biophys, 239: 137-46, 1985.[CrossRef][Medline]
  4. Raval P, Iversen PL, Bresnick E. Induction of cytochromes P450IA1 and P450IA2 as determined by solution hybridization. Biochem Pharmacol, 41: 1719-23, 1991.[CrossRef][Medline]
  5. Wijnhoven SW, Kool HJ, van Oostrom CT, et al The relationship between benzo[a]pyrene-induced mutagenesis and carcinogenesis in repair-deficient Cockayne syndrome group B mice. Cancer Res, 60: 5681-7, 2000.[Abstract/Free Full Text]
  6. Kerr SJ. Competing methyltransferase system. J Biol Chem, 247: 4248-52, 1972.[Abstract/Free Full Text]
  7. Yeo EJ, Wagner C. Tissue distribution of glycine N-methyltransferase, a major folate-binding protein of liver. Proc Natl Acad Sci USA, 91: 210-4, 1994.[Abstract/Free Full Text]
  8. Chen YMA, Shiu JY, Tzeng SJ, et al Characterization of glycine-N-methyltransferase-gene expression in human hepatocellular carcinoma. Int J Cancer, 75: 787-93, 1998.[CrossRef][Medline]
  9. Liu HH, Chen KH, Lui WY, Wong FW, Chen YMA. Characterization of reduced expression of glycine N-methyltransferase in the cancerous hepatic tissues using two newly developed monoclonal antibodies. J Biomed Sci, 10: 87-97, 2003.[CrossRef][Medline]
  10. Chen YMA, Chen LY, Wong FH, et al Genomic structure, expression and chromosomal localization of the human glycine N-methyltransferase gene. Genomics, 66: 43-7, 2000.[CrossRef][Medline]
  11. Tseng TL, Shih YP, Huang YC, et al Genotypic and phenotypic characterization of a putative tumor susceptibility gene, GNMT, in liver cancer. Cancer Res, 63: 647-54, 2003.[Abstract/Free Full Text]
  12. Raha A, Wagner C, MacDonald RG, Bresnick E. Rat liver cytosolic 4 S polycyclic aromatic hydrocarbon-binding protein is glycine N-methyltransferase. J Biol Chem, 269: 5750-6, 1994.[Abstract/Free Full Text]
  13. Nakabayashi H, Taketa K, Miyano K, Yamane T, Sato J. Growth of human hepatoma cell lines with differentiated functions in chemically defined medium. Cancer Res, 42: 3858-63, 1982.[Abstract/Free Full Text]
  14. Fogh J, Trempe G, Loveless JD. New human tumor cell lines Fogh J eds. . Human tumor cells in vitro, 115-9, Plenum Press New York 1977.
  15. Aden DP, Fogel H, Plotkin S, Damjanov I, Knowles BB. Controlled synthesis of HbsAg in a differentiated human liver-carcinoma-derived cell line. Nature (Lond), 282: 615-6, 1979.[CrossRef][Medline]
  16. Clark SJ, Harrison J, Paul CL, Frommer M. High sensitivity mapping of methylated cytosines. Nucleic Acids Res, 22: 2990-7, 1994.[Abstract/Free Full Text]
  17. Gurtu V, Yan G, Zhang G. IRES bicistronic expression vectors for efficient creation of stable mammalian cell lines. Biochem Biophys Res Commun, 229: 295-8, 1996.[CrossRef][Medline]
  18. Kleinerman D, Zhang WW, von Eschenbach AC, Lin S-H, Hsieh JT. Application of a tumor suppressor gene, C-CAM1, in androgen-independent prostate cancer therapy: a preclinical study. Cancer Res, 55: 2831-6, 1995.[Abstract/Free Full Text]
  19. Vesselinovitch SD, Koka M, Mihailovich N, Rao KVN. Carcinogenicity of diethyl-nitrosamine in newborn, infant and adult mice. J Cancer Res Clin Oncol, 108: 60-5, 1984.[CrossRef][Medline]
  20. Roggeband R, Wolterbeek AP, Rutten AA, Baan RA. Comparative 32P-postlabeling analysis of benzo[a]pyrene-DNA adducts formed in vitro upon activation of benzo[a]pyrene by human, rabbit and rodent liver microsomes. Carcinogenesis (Lond), 14: 1945-50, 1993.[Abstract/Free Full Text]
  21. Chang KW, Lee H, Wang HJ, Chen SY, Lin P. Differential response to benzo[a]pyrene in human lung adenocarcinoma cell lines: the absence of aryl hydrocarbon receptor activation. Life Sci, 65: 1339-49, 1999.[CrossRef][Medline]
  22. Rowling MJ, McMullen MH, Schalinske KL. Vitamin A and its derivatives induce hepatic glycine N-methyltransferase and hypomethylation of DNA in rats. J Nutr, 132: 365-9, 2002.[Abstract/Free Full Text]
  23. Rosenfeld RJ, Goodsell DS, Musah RA, et al Automated docking of ligands to an artificial active site: augmenting crystallographic analysis with computer modeling. J Comput Aid Mol Des, 17: 525-36, 2003.
  24. Fu Z, Hu Y, Konishi K, et al Crystal structure of glycine N-methyltransferase from rat liver. Biochemistry, 35: 11985-93, 1996.[CrossRef][Medline]
  25. Huang Y, Komoto J, Konishi K, et al Mechanisms for auto-inhibition and forced product release in glycine N-methyltransferase: crystal structures of wild-type, mutant R175K and S-adenosylhomocysteine-bound R175K enzymes. J Mol Biol, 298: 149-62, 2000.[CrossRef][Medline]
  26. Vidgren J, Svensson LA, Liljas A. Crystal structure of catechol O-methyltransferase. Nature (Lond), 368: 354-8, 1994.[CrossRef][Medline]
  27. Cheng X, Kumar S, Posfai J, Pflugrath JW, Roberts RJ. Crystal structure of the HhaI DNA methyltransferase complexed with S-adenosyl-L-methionine. Cell, 74: 299-307, 1993.[CrossRef][Medline]
  28. Schluckebier G, Kozak M, Bleimling N, Weinhold E, Saenger W. Differential binding of S-adenosylmethionine, S-adenosylhomocysteine and Sinefungin to the adenine-specific DNA methyltransferase M.TaqI. J Mol Biol, 265: 56-67, 1997.[CrossRef][Medline]
  29. Reinisch KM, Chen L, Verdine GL, Lipscomb WN. The crystal structure of HaeIII methyltransferase convalently complexed to DNA: an extrahelical cytosine and rearranged base pairing. Cell, 82: 143-53, 1995.[CrossRef][Medline]
  30. Gong W, O’Gara M, Blumenthal RM, Cheng X. Structure of pvu II DNA-(cytosine N4) methyltransferase, an example of domain permutation and protein fold assignment. Nucleic Acids Res, 25: 2702-15, 1997.[Abstract/Free Full Text]
  31. Tran PH, Korszun ZR, Cerritelli S, Springhorn SS, Lacks SA. Crystal structure of the Dpnm DNA adenine methyltransferase from the DpnII restriction system of Streptococcus pneumoniae bound to S-adenosylmethionine. Structure, 6: 1563-75, 1998.[Medline]
  32. Scavetta RD, Thomas CB, Walsh M, et al Structure of RsrI methyltransferase, a member of the N6-adenine B class of DNA methyltransferases. Nucleic Acids Res, 28: 3950-61, 2000.[Abstract/Free Full Text]
  33. Dong A, Yoder JA, Zhang X, Zhou L, Bestor TH, Cheng X. Structure of human Dnmt2, an enigmatic DNA methyltransferase homolog that displays denaturant-resistant binding to DNA. Nucleic Acids Res, 29: 439-48, 2001.[Abstract/Free Full Text]
  34. Morris GM, Goodsell DS, Halliday RS, et al Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function. J Comput Chem, 19: 1639-62, 1998.[CrossRef]
  35. Cook RJ, Wagner C. Glycine N-methyltransferase is a folate binding protein of rat liver cytosol. Proc Natl Acad Sci USA, 81: 3631-4, 1984.[Abstract/Free Full Text]
  36. Okey AB, Bendy GP, Mason ME, et al Temperature-dependent cytosol-to-nucleus translocation of the Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin in continuous cell culture lines. J Biol Chem, 255: 11415-22, 1980.[Free Full Text]
  37. Hapgood J, Cuthill S, Denis M, Poellinger L, Gustafsson JA. Specific protein-DNA interactions at a xenobiotic-responsive element: copurification of dioxin receptor and DNA-binding activity. Proc Natl Acad Sci USA, 86: 60-4, 1989.[Abstract/Free Full Text]
  38. Foussat J, Costet P, Galtier P, Pineau T, Lesca P. The 4S benzo(a)pyrene-binding protein is not a transcriptional activator of Cyp1a1 gene in Ah receptor-deficient (AHR–/–) transgenic mice. Arch Biochem Biophys, 349: 349-55, 1998.[CrossRef][Medline]
  39. Bhat R, Wagner C, Bresnick E. The homodimic form of glycine N-methyl transferase acts as a polycyclic aromatic hydrocarbon-binding receptor. Biochem J, 36: 9906-10, 1997.
  40. Chang X, Roberts RJ. AdoMet-dependent methylation, DNA methyltransferases and base flipping. Nucleic Acids Res, 29: 3784-95, 2001.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Nutr.Home page
E. O. Uthus and S. A. Ross
Dietary Selenium Affects Homocysteine Metabolism Differently in Fisher-344 Rats and CD-1 Mice
J. Nutr., May 1, 2007; 137(5): 1132 - 1136.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
Y.-C. Huang, C.-M. Lee, M. Chen, M.-Y. Chung, Y.-H. Chang, W. J.-S. Huang, D. M.-T. Ho, C.-C. Pan, T. T. Wu, S. Yang, et al.
Haplotypes, Loss of Heterozygosity, and Expression Levels of Glycine N-Methyltransferase in Prostate Cancer
Clin. Cancer Res., March 1, 2007; 13(5): 1412 - 1420.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
Z. Luka, S. Pakhomova, L. V. Loukachevitch, M. Egli, M. E. Newcomer, and C. Wagner
5-Methyltetrahydrofolate Is Bound in Intersubunit Areas of Rat Liver Folate-binding Protein Glycine N-Methyltransferase
J. Biol. Chem., February 9, 2007; 282(6): 4069 - 4075.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
Y.-K. Leung, K.-M. Lau, J. Mobley, Z. Jiang, and S.-M. Ho
Overexpression of Cytochrome P450 1A1 and Its Novel Spliced Variant in Ovarian Cancer Cells: Alternative Subcellular Enzyme Compartmentation May Contribute to Carcinogenesis
Cancer Res., May 1, 2005; 65(9): 3726 - 3734.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, S.-Y.
Right arrow Articles by Chen, Y.-M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, S.-Y.
Right arrow Articles by Chen, Y.-M. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online