Skip Navigation


Carcinogenesis Advance Access originally published online on April 29, 2007
Carcinogenesis 2007 28(10):2184-2192; doi:10.1093/carcin/bgm100
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/10/2184    most recent
bgm100v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Bradley, C.
Right arrow Articles by Parl, F. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bradley, C.
Right arrow Articles by Parl, F. F.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Carcinogen-induced histone alteration in normal human mammary epithelial cells

Chastity Bradley, Riet van der Meer, Nady Roodi, Heping Yan1, Mahesh B. Chandrasekharan1, Zu-Wen Sun1, Ray L. Mernaugh1 and Fritz F. Parl*

Department of Pathology
1 Department of Biochemistry, Vanderbilt University Medical Center, 4918 TVC, 22nd Avenue South Pierce, Nashville, TN 37232

* To whom correspondence should be addressed. Tel: +615 343 9117; Fax: +615 343 9563; Email: fritz.parl{at}vanderbilt.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Little is known about early carcinogen-induced protein alterations in mammary epithelium. Detection of early alterations would enhance our understanding of early-stage carcinogenesis. Here, normal human mammary epithelial cells (HMECs) were exposed to dietary and environmental carcinogens [2-amino-1-methyl-6-phenylimidazo[4,5b]pyridine (PhIP), 4-aminobiphenyl (ABP), benzo[a]pyrene, 2,3,7,8-tetrachlorodibenzo-p-dioxin] individually or in combination. A phage display library of single-chain variable fragment antibodies was used to screen protein targets altered by the treatment. In combination with matrix-assisted laser desorption time of flight, we identified histone H3 as a target antigen. Although histone H3 total protein remained unchanged in control and treated HMEC, the methylation of lysine 4 was altered. A reduction in mono-methyl histone H3 (Lys 4) was observed in treated HMEC compared with control HMEC. This alteration was shown to be dependent on carcinogen concentration and specific for PhIP and ABP. To characterize potential histone demethylation mechanisms, localization and protein expression patterns of lysine-specific demethylase 1 (LSD1) were analyzed. In control HMEC, LSD1 was present at the nuclear periphery. However, following 72 h carcinogen treatment, LSD1 localized within the nucleus. Within 48 h after treatment, mono-methyl histone H3 (Lys 4) was restored and LSD1 localization was reversed. Protein expression levels of LSD1 were also increased in treated HMEC compared with control HMEC. Our data suggest that the induction of a single enzyme, LSD1, represents an early response to carcinogen exposure, which leads to the demethylation of histone H3 (Lys 4), which, in turn, may influence the expression of multiple genes critical in early-stage mammary carcinogenesis.

Abbreviations: ABP, aminobiphenyl; B[a]P, benzo[a]pyrene; DAPI, 4',6-diamidino-2-phenylindole; FMAT, fluorimetric microvolume assay technology; HMEC, human mammary epithelial cells; LSD1, lysine-specific demethylase 1; MALDI-TOF, matrix-assisted laser desorption time of flight; PBS, phosphate-buffered saline; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5b]pyridine; ScFV, single-chain variable fragment; TCDD2, 3,7,8-tetrachlorodibenzo-p-dioxin


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Chemical carcinogens are commonly present in the environment, diet and cigarette smoke. Prolonged exposure to these toxins can lead to the formation of DNA adducts and eventually cancer (1–4). To date, little is known about early carcinogen-induced protein alterations in mammary epithelium (5,6). The identification of early biochemical alterations in carcinogen-exposed mammary epithelium would be of great theoretical interest to enhance our understanding of early-stage carcinogenesis and of practical importance for the potential diagnosis of precursor lesions.

A major challenge in studying early carcinogen-induced protein alterations is access to reliable detection and cell model systems, which may account for the small number of reported protein alterations (7,8). Previous studies have determined gene expression signatures in normal and breast cancer cells (9,10). However, most of these studies have examined cDNA levels that may not be indicative of protein expression. Furthermore, mRNA levels are not informative of proteins that are post-translationally modified. Screening phage display libraries offers the unbiased evaluation of all proteins whose expression levels change in response to carcinogen exposure (11). While phage display-based screening of the complete array of cellular proteins does not guarantee detection of causality, this unbiased screening approach, in general, is a significant step forward from the traditional arbitrary selection of a few proteins that may or may not be causally related to tumorigenesis.

Since breast cancers arise from mammary epithelial cells, we analyzed carcinogen-exposed normal human mammary epithelial cells (HMECs) to address early biochemical alterations. In this study, we exposed HMEC to four compounds that are representative of the main classes of chemical carcinogens. The heterocyclic amine 2-amino-1-methyl-6-phenylimidazo[4,5b]pyridine (PhIP) and the aromatic amine 4-aminobiphenyl (ABP) have been shown to induce mammary tumors in animal models (12,13). The polycyclic aromatic hydrocarbon benzo[a]pyrene (B[a]P) and the halogenated hydrocarbon 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) are also known mammary carcinogens (14–17).

We were able to identify single-chain variable fragment (ScFv) antibodies from a phage-displayed recombinant antibody library that distinguished carcinogen-treated from untreated HMEC. One target antigen was histone H3 whose methylation status at lysine residue 4 changed upon carcinogen treatment, possibly mediated by the demethylating enzyme, lysine-specific demethylase 1 (LSD1). These data suggest that the methylation of histone H3 is altered in carcinogen-treated HMEC and may be directly involved in regulating the expression of genes critical in early-stage mammary carcinogenesis.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Chemicals
PhIP was obtained from Toronto Research Chemicals (North York, ON) and 4-ABP and B[a]P were purchased from Sigma Chemical Company (St Louis, MO). All other chemicals of molecular biology grade were obtained from Sigma.

HMEC culture and carcinogen treatment
HMEC were obtained at passage 7 from Cambrex (Walkersville, MD) and grown in Medium 171 (Cascade Biologics, Portland, OR) according to the manufacturer's specifications. Cells were cultured in T-175 flasks to a total of 4 x 108 and all experiments were carried out in passage 12. To detect early carcinogen-induced protein alterations in HMEC, we selected low doses and short exposure times from treatment regimens reported in previous studies (14,18–20). Initially, cells were treated with 10 nM TCDD for 72 h and 5 µM PhIP, B[a]P and 4-ABP for 48 h. For individual concentration-dependent studies, HMEC were exposed to 5, 20 or 40 µM of PhIP, ABP or B[a]P for 72 h. For combined treatment studies, HMEC were treated with 10 µM PhIP, ABP and B[a]P for 72 h. Control cells were exposed to vehicle (0.001% dimethyl sulphoxide). Final cell confluency did not exceed 80%. Cells were harvested using Trypsin–ethylenediaminetetraacetic acid (Cambrex) and collected for protein extraction and immunocytochemistry.

Protein extraction and biotinylation
A modified version of a previously published protocol was for biotinylation (21). Protein concentrations of control and treated HMEC were determined using Bio-Rad protein assay kit (Hercules, CA). Cells were lysed by adding an equal volume of 2x sample buffer [8 M urea, 50 mM sodium phosphate, 4% 3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate, 130 µM dithiothreitol in phosphate-buffered saline (PBS)] and heated to 100°C for 10 min. The average molecular weight of proteins in cell lysates were arbitrarily assigned a value of 50, 000 per protein molecule for biotinylation purposes. The pH of cell lysates were adjusted to ~9.4 using carbonate/phosphate buffer. Pierce EZ-Link NHS-LC-LC-Biotin (Rockford, IL) was added to each sample to achieve a ratio of five biotins per molecule. No attempt was made to remove any unincorporated biotin from the reaction mix. The biotinylated proteins were separated from unbiotinylated proteins with streptavidin-coated magnetic beads and used for antibody selection.

Phage-displayed ScFv library
A modified version of a previously published protocol was used to develop a large phage-displayed ScFv recombinant antibody library (22). Spleens from outbred, newborn and 3- to 3-week-old mice and rats were used as a source of antibody genetic starting material to produce a library that contained ~2.9 x 109 members. Excluding dyes, all reagents, materials, vectors, etc. used for studies involving ScFv antibodies were obtained from G.E. Healthcare (Piscataway, NJ). All ScFv stemming from the recombinant antibody library had been cloned into Escherichia coli TG1 cells using the pCANTAB5E phagemid vector. The vector contains a lac promoter to control for ScFv and gene-3 protein expression and an ampicillin-resistance gene to select for bacterial clones that contain the ScFv-encoding phagemid. In addition, this vector also contains an amber stop codon located between the DNA encoding the tag on the ScFv C-terminus and gene-3 protein. Therefore, the bacteria can express either phage-displayed ScFv (if helper phage rescued) or soluble ScFv if grown in the presence of isopropyl ß-D-1-thiogalactopyranoside. All ScFv were expressed either as phage gene-3 fusion protein (for phage display and selection purposes) or as an E-tagged soluble ScFv for all assays performed throughout the study. Phage-displayed ScFv were not used for any assays. Expressed ScFv are recognized by the anti-E tag and horseradish peroxidase/anti-E tag monoclonal antibodies. The anti-E tag antibody can be used to detect ScFv bound to antigens in assays.

Phage display ScFv antibody selection
A modified version of ScFv antibody selection was performed as described previously (22). For phage antibody selections, 200 µg of biotinylated control and treated lysates were added to 200 µl of ProMega Streptavidin MagneSphere Paramagnetic Particles (vendor to Promega, Madison, WI) and incubated for 5–15 min at room temperature. Biotinylated proteins not bound to beads were removed by washing twice with 0.1% PBS–Tween 20. To reduce the number of ScFv antibodies recognizing epitopes present on control HMEC, 1 ml of phage antibody library (~1013 phage/ml) in 0.1% Tween 20 was added to biotinylated control lysates conjugated to streptavidin-coated beads and incubated for 1 h. Phage-displayed ScFv antibodies bound to biotinylated control lysates were separated from unbound phage-displayed ScFv antibodies using a magnet. The supernatant containing the unbound antibodies were incubated with biotinylated treated lysates to select for phage-displayed ScFv antibodies. After a 3-h incubation, bound phage-displaying ScFv were eluted with 100 mM triethanolamine for 10 min at room temperature and neutralized with 1 M Tris, pH 7.4. Retrieved phage-displaying ScFv were used to infect E.coli TG1 cells. Phage-infected bacteria were plated on 2xYTAG agar plates (17 g bacto-tryptone, 10 g bacto-yeast extract, 5 g NaCl, 20 g glucose, 100 mg ampicillin, 15 g agar/l, final concentration) and grown overnight at 30°C. TG1 cells were phage rescued with M13KO7 helper virus to facilitate ScFv antibody expression on the phage surface. Ten milliliters of phage-displayed ScFv were precipitated using polyethylene glycol, the pellet was resuspended in 1 ml of 0.15 M NaCl containing 0.1% Tween 20 and a second round of selection was performed.

Soluble E-tagged ScFv production
Briefly, diluted bacteria transformed with the output of the second round selection were plated on 2xYTAG agar plates and grown overnight at 30°C. Individual bacterial ScFv colonies were picked using an automated colony picker QPix (Genetix, Boson, MA), transferred into a sterile Nunc 384-well microtiter plate containing 100 µl/well of 2xYTAI (17 g bacto-tryptone, 10 g bacto-yeast extract, 5 g NaCl, 100 mg ampicillin, 1 mM isopropyl ß-D-1-thiogalactopyranoside final concentration) and grown overnight at 30°C. A 384-pin tool was used to transfer bacteria from the 2xYTAI 384 well (master plate) to a 384-well 2xYTAG replica plate. Master plates were centrifuged to pellet E.coli TG1 cells, and the supernatant was removed. The cell pellets from the master plates were resuspended in 40 µl of TES (0.2 M Tris, pH 8.0, 0.5 mM ethylenediaminetetraacetic acid, 0.5 M sucrose) and 60 µl of 1/5x TES and incubated on ice for 1 h at room temperature to obtain soluble E-tagged ScFv in periplasmic extract for use in assays.

Differential ScFv Screen
The fluorimetric microvolume assay technology (FMAT) ScFv cell-staining assay has been previously described (21). Briefly, control and treated HMEC were plated into separate 384-well Costar 3712 microtiter plates at a density of 3000 cells/well. Cells were allowed to attach, fixed with methanol and blocked with 5% bovine serum albumin in PBS. FMAT Blue Monofunctional Dye (Applied Biosystems, Foster City, CA excitation 633 nm, emission 670 nm) was conjugated to anti-E monoclonal antibody at a ratio of 2.25 dye molecules per antibody using FMAT Blue Monofunctional Dye Kit (Applied Biosystems) according to manufacturer's directions. Twenty-five microliters of conjugated anti-E tag monoclonal antibody diluted to 1 µg/ml in 0.1% PBS–Tween 20 was added to each well of the 384-well microtiter plates. Twenty-five microliters of soluble E-tagged ScFv in periplasmic extract was transferred from 384-well culture plates to corresponding wells containing control and treated HMEC. The plates were incubated for 1 h at room temperature in the dark. The fluorescence intensity in each well was measured by a FMAT 8100 Analyzer (Applied Biosystems) that scans the bottom of the microtiter plates at a depth of focus of 100 µm. Only fluorescence that is bound to the cells will be detected which eliminates the need for wash steps. In addition, all reagents are added simultaneously which greatly reduces incubation time for each reagent and significantly increases productivity. For any given ScFv antibody, the difference in fluorescence intensity between control and treated cells reflected relative expression levels of the respective target antigens. Escherichia coli TG1 cells containing ScFv of interest could be transferred, grown in separate culture containers and used to prepare frozen glycerol stocks. Glycerol stocks containing E.coli TG1 cells with ScFv in pCANTAB5E were grown on a large scale and used to produce milligram amounts of soluble E-tagged ScFv. Soluble E-tagged ScFv were purified using an RPAS purification kit according to the manufacturer's instructions (G.E. Healthcare).

Soluble E-tagged ScFv immunoprecipitation
A modified version of a previously published protocol was used to perform immunoprecipitation using soluble ScFv antibodies (23). Fifty micrograms of HMEC lysates were incubated with soluble ScFv overnight at 4°C with end-over-end mixing. Monoclonal anti-E antibody was conjugated to N-hydroxy succinimide-activated SepharoseTM 4 Fast Flow (G.E. Healthcare) at a ratio of 1 mg of anti-E antibody to 1 ml of sepharose beads using a NapTM 10 column (G.E. Healthcare) according to the manufacturer's directions. Following overnight ScFv incubation, 50 µg conjugated anti-E sepharose beads were added to HMEC lysates for 1 h. Beads were centrifuged and washed three times with 1 ml of buffer containing 50 mM Tris, 150 mM NaCl, pH 8.0, 1 mM ethylenediaminetetraacetic acid, 0.1% Triton X-100 and once with cold PBS. Beads were heated at 100°C in 20 µl of 2x Laemmli buffer (Sigma) and 20 µl was loaded onto a 4–20% Tris/Glycine gel for sodium dodecylsulphate–polyacrylamide gel electrophoresis.

Matrix-assisted laser desorption
Proteins were separated by sodium dodecylsulphate–polyacrylamide gel electrophoresis and the gel was stained with colloidal coomassie (Invitrogen, Carlsbad, CA). Individual protein bands were identified by the Vanderbilt Proteomics Laboratory, Mass Spectrometry Research Center, as described previously (24). Briefly, protein bands were excised, equilibrated in 50 mM NH4HCO3, reduced with dithiothreitol (3 mM in 100 mM NH4HCO3, 37°C for 15 min) and alkylated with iodoacetamide (6 mM in 50 mM NH4HCO3for 15 min). The gel slice was then dehydrated with acetonitrile and rehydrated with 15 ml 12.5 mM NH4HCO3 containing 0.01 mg/ml modified trypsin (Promega, Madison, WI), and trypsin digestion was carried out for >2 h at 37°C. Peptides were extracted with 60% acetonitrile, 0.1% trifluoroacetic acid and dried by vacuum centrifugation. Peptides were reconstituted in 5 ml 60% acetonitrile and 0.1% trifluoroacetic acid, and 0.4 ml of the reconstituted peptides were applied to a target plate and overlayed with 0.4 ml {alpha}-cyano-4-hydroxycinnamic acid matrix (5 mg/ml in 60% acetonitrile, 0.1% trifluoroacetic acid) supplemented with ammonium citrate (1 mg/ml). The high abundance of the target proteins allowed for mass spectrometry on the reconstituted peptide mixtures without additional concentration or cleanup steps. Peptide mixtures were analyzed by matrix-assisted laser desorption ionization time of flight (MALDI-TOF) and TOF/TOF tandem mass spectrometry using a Voyager 4700 mass spectrometer (Applied Biosystems). Mass spectral data, in the form of peptide mass maps of the intact molecular peptide ions (M+H), as well as fragmentation data derived from individual peptide ions were used to interrogate the Swiss-Prot and NCBInr protein databases for statistically significant protein matches using GPS Explorer software (Applied Biosystems) running the MASCOT search engine (Matrix Science, Boston, MA). Trypsin autolytic peptide ions (m/z = 842.51, 2211.10) were used for internal calibration of the peptide mass maps, allowing for searches to be performed with mass accuracy of <20 p.p.m. Searches also allowed for one missed cleavage, complete carbamidomethylation of cysteine sulfhydryls and partial oxidation of methionine residues.

Histone extraction
Histones were extracted from control and treated HMEC as described by Upstate (Charlottesville, VA). Cells were lysed with lysis buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol, 1.5 mM phenylmethylsulfonyl fluoride). HCL (6 N) was added to achieve a 0.2 N final concentration for each sample. Cells were centrifuged at 12 000 r.p.m. for 10 min at 4°C. The supernatant was loaded into a 3500 MWCO Dialysis Cassette (Pierce) and dialyzed against 0.1 N acetic acid twice for 1 h each and H2O for 1 h, 3 h and overnight. Dialysate was collected and centrifuged at 14 000 r.p.m. for 1 h. The pelleted histones were dissolved in 2x Laemmli buffer.

Western immunoblot
A modified version of the previously published protocol was used to determine histone expression levels (25). Control and carcinogen-treated HMEC lysates or nuclear extracts were separated on either a 15 or 4–20% Tris/Glycine gel, transferred onto polyvinylidene difluoride and incubated with the following anti-human IgG antibodies:sheep anti-histone H3 1:1000 (Novus Biologicals, Littleton, CO), rabbit anti-mono-methyl-histone H3 (Lys 4) 1:250, rabbit anti-di-methyl histone H3 (Lys 4) 1:250, rabbit anti-tri-methyl histone H3 (Lys 4) 1:250 (Upstate), mouse anti-phospho-histone H3 (Ser 10) 1:1000 (Abcam, Cambridge, MA), rabbit anti-histone H4 1:250 and rabbit anti-LSD1 1:250 (Abgent, San Diego, CA). Corresponding horseradish peroxidase-conjugated secondary IgG antibodies [histone H3 and phospho-histone H3 1:4000, mono-methyl histone H3 (Lys 4), di-methyl histone H3 (Lys 4), tri-methyl histone H3 (Lys 4), H4 and LSD1 1:1000] were applied. Enhanced chemiluminescent detection system was used according to manufacturer's directions (Pierce).

Immunofluorescence microscopy
A modified version of a previously published protocol was used for immunofluorescent cell staining (26). Sections (7 µm) of paraffin-embedded control and carcinogen-treated HMEC were deparaffinized with xylene, rehydrated in decreasing ethanol gradient and incubated with Target Retrieval Solution (DakoCytomation, Carpinteria, CA) in a pressure cooker to assist in antigen retrieval according to manufacturer's directions. Endogenous peroxidase activity was quenched by incubating sections with 0.5% hydrogen peroxide/30%methanol/PBS for 30 min. Sections were washed with PBS, blocked with 5% bovine serum albumin for 30 min and incubated with Image-iT FX Signal Enhancer (Molecular Probes, Eugene, OR) according to manufacturer's directions. Primary antibodies (histone H3 1:400, mono-methyl histone H3 (Lys 4) 1:50, LSD1 1:75) were applied to sections and incubated in a humidified chamber for 1 h. Unbound antibody was removed by washing with PBS. Fluorophore-conjugated secondary IgG antibodies [histone H3–Alexa-Fluor 594 (Molecular Probes) 1:200; mono-methyl histone H3 (Lys 4)–fluorescein isothiocyanate (Sigma) 1:80; LSD1–fluorescein isothiocyanate 1:200] were applied for 45 min followed by washing. The 4',6-diamidino-2-phenylindole (DAPI) dihydrochloride (Invitrogen) diluted in PBS to a final concentration of 1 µM was applied to each section and incubated for 3 min. Sections were washed and ProLong Gold anti-fade reagent was applied according to manufacturer's directions (Invitrogen). Coverslips were affixed and photographs were taken with an Olympus BX41 microscope equipped with a digital MicroFire camera.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Detection of early carcinogen-induced protein alterations using phage display
HMEC were exposed to 10 nM TCDD for 72 h and 5 µM PhIP, ABP and B[a]P for 48 h. As a result of carcinogen exposure, the cell number was reduced ~25% in treated HMEC compared with control HMEC. However, the majority of treated HMEC was viable and revealed little variation in cellular morphology. To identify carcinogen-induced protein alterations, we used a phage display antibody library to select prospective antibodies capable of distinguishing target antigens in control and treated HMEC (Figure 1). To further enrich the ScFv antibody pool, a fluorescent-based high-throughput screening assay as demonstrated in Figure 2A was performed following the phage antibody selection. Figure 2B illustrates a representative analysis using the FMAT, which quantitatively compared fluorescence intensity of antibody–antigen complexes in control and treated HMEC. This application allowed us to assess the interaction of multiple ScFv antibodies with target antigens (as indicated by individual bars) present in control and treated HMEC. We pursued ScFv antibodies with signal intensities differing at least 4-fold between control and treated HMEC.


Figure 1
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Phage antibody selection. To select ScFv antibodies from the phage display antibody library, biotinylated control HMEC lysates conjugated to streptavidin-coated magnetic beads are used to cross-absorb phage-displayed ScFv antibodies. Biotinylated treated HMEC lysates conjugated to streptavidin-coated magnetic beads are then incubated with depleted antibody library and ScFv antibodies are eluted. Eluted phage-displayed ScFv antibodies are used to infect bacteria. Infected bacteria are subsequently infected with a helper virus, M13KO7. Phage-displayed ScFv antibodies are precipitated and used to perform a second round selection.

 


Figure 2
View larger version (75K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. FMAT for high-throughput screening of soluble ScFv recombinant antibodies. (A) FMAT is a fluorescence-based high-throughput screening assay designed to simultaneously screen multiple soluble recombinant ScFv antibodies to determine relative target antigen expression levels in control and treated HMEC. In step 1, control and treated HMEC are plated at a density of 3000 cells/well in separate 384-well microtiter plates. Each well contained either control or treated HMEC, a single ScFv antibody, and FMAT Blue-conjugated anti-E antibody. In step 2, each microtiter plate is scanned by the FMAT 8100 analzyer at a depth of focus of 100 µm. Only fluorescence bound to cells will be detected by FMAT. For any given soluble ScFv, differences in fluorescence intensity in corresponding control and treated HMEC wells reflect relative expression levels of the respective antigens. In step 3, the average fluorescence intensity for each well is converted into graphical form. Prospective soluble ScFv clones can be picked, purified and used in future diagnostic assays. (B) Illustration of FMAT results. Generally, the overall fluorescence intensity in treated HMEC was greater than in the control counterpart (arrow, right panel). However, certain protein targets showed higher expression levels in control than in treated HMEC (arrow, left panel). ScFv antibodies with signal intensities differing at least 4-fold between control and treated HMEC were picked, grown in Escherichia coli TG1 cells and purified for diagnostic assays.

 
Histone H3 is an early carcinogen-induced protein target
Soluble ScFv candidates were used for immunoprecipitation using control and treated HMEC. We identified several differentially expressed antigens ranging in size from 14 to 97 kDa. Some showed higher expression levels in treated cells whereas others were more strongly expressed in control HMEC. To identity two target antigens of interest, MALDI-TOF analysis was performed. The mass spectra generated from our data identified antigens, uracil DNA glycosylase and histone H3 as the respective target antigens (Figure 3). These two antigens were altered as a result of carcinogen treatment, i.e. uracil DNA glycosylase was upregulated and histone H3 appeared downregulated. We decided to focus our attention on histone H3 whose identification was surprising because histones are considered housekeeping proteins present in fixed concentrations in all cells (27). At the same time, histones are known to be modified by methylation, acetylation and phosphorylation. Therefore, we speculated that the carcinogen-induced alteration in histone H3 might reflect a change in post-translational modification.


Figure 3
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. MALDI/TOF analysis. Control and carcinogen-treated HMEC lysates were immunoprecipitated using a soluble recombinant ScFv antibody selected from the phage display antibody library. Proteins were separated using sodium dodecylsulphate–polyacrylamide gel electrophoresis and stained with colloidal coomassie. Bands of interest were excised and MALDI/TOF analysis was performed. A complete description of the procedure can be found in the Materials and methods.

 
To examine post-translational modifications of histone H3 in control and treated HMEC, we used a panel of antibodies directed against histone H3, mono-, di- and tri-methyl histone H3 (Lys 4), phospho-histone H3 (Ser 10) and histone H4 (Figure 4A). Of the antibodies tested on control and treated HMEC lysates, only mono-methyl histone H3 (Lys 4) revealed a differential methylation pattern, i.e. histone H3 methylation at lysine 4 residue was observed in control HMEC and reduced in treated HMEC (Figure 4A). No expression of di- or tri-methyl histone H3 (Lys 4) was detected in either control or treated HMEC. Equal levels of histone H3 (Ser 10) phosphorylation in control and treated HMEC suggest that the carcinogen treatment is modification specific. In addition, control and treated HMEC contained equal levels of histone H4, acetyl histone H4 (Lys 8) and pentacetyl histone H4 (results not shown), suggesting that the carcinogen-induced alteration is mono-methyl histone H3 (Lys 4) specific.


Figure 4
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Histone analysis in control HMEC and in HMEC treated with 5 µM PhIP, B[a]P, ABP and 10 nM TCDD for 72 h. (A) Western immunoblots. A panel of anti-human histone H3 and histone H4 antibodies, including mono-methyl histone H3 (Lys 4), di-methyl histone H3 (Lys 4), tri-methyl histone H3 (Lys 4), phospho-histone H3 (Ser 10), and histone H4 was used to determine the expression levels of histones in control and treated HMEC. The results show a striking reduction in mono-methyl histone H3 (Lys 4) levels in treated compared with control HMEC. (B) Histone H3 Immunocytochemistry of control and treated HMEC. Paraffin-embedded control and treated HMEC were stained with anti-histone H3 and anti-mono-methyl histone H3 (Lys 4). Secondary fluorophores (Alexa Red 594 and fluorescein isothiocyanate), respectively, were used to detect cellular localization. Row 1: hematoxylin and eosin stains revealed little variation in control and treated HMEC morphology. Row 2: anti-histone H3 stained the nuclei of both control and treated HMEC. Row 3: mono-methyl histone H3 (Lys 4) levels in treated HMEC were greatly reduced compared with control HMEC. Row 4: Overlay of rows 2 and 3. Insets: single cells stained with either anti-histone H3 or anti-mono-methyl histone H3 (Lys 4) and counterstained with DAPI to confirm nuclear localization of proteins. Magnification x400; insets x600.

 
We also evaluated the cellular localization and changes in relative expression levels of histone H3 and mono-methyl histone H3 (Lys 4). As expected, immunocytochemical studies localized histone H3 to the nuclei of both control and treated cells (Figure 4B). Histone H3 nuclear localization was confirmed by counterstaining nuclei with DAPI (Figure 4B, inset). Mono-methyl histone H3 (Lys 4) also localized to the nuclei but compared with control HMEC, levels were significantly reduced in carcinogen-treated HMEC (Figure 4B). The insets show single cells stained with anti-mono-methyl histone H3 (Lys 4) and counterstained with DAPI. In control HMEC, mono-methyl histone H3 (Lys 4) co-localizes with nuclear counterstain (blue-green) whereas in treated HMEC mono-methyl histone H3 (Lys 4) is absent; therefore, the nucleus is stained only with DAPI (blue). At this point, we focused on characterizing mono-methyl histone H3 (Lys 4) in control and carcinogen-treated HMEC. Histone H3 was used as an internal control for all assays performed throughout the study.

Mono-methyl histone H3 (Lys 4) methylation pattern is carcinogen concentration dependent and reversible
Since mono-methyl histone H3 (Lys 4) was reduced in HMEC treated with a combination (PhIP, ABP, B[a]P, TCDD) of dietary and environmental carcinogens, we next wanted to determine individual carcinogen contributions to this protein alteration. HMEC were exposed individually to 5 µM PhIP, B[a]P or ABP for 48 h. TCDD is no longer present in the environment and, therefore, was excluded from the study at this point. A comparison of control and treated HMEC showed no reduction in mono-methyl histone H3 (Lys 4) in the latter (results not shown). This data suggest that HMEC treated individually with 5 µM PhIP, B[a]P or ABP does not alter the methylation status of mono-methyl histone H3 (Lys 4). Therefore, we hypothesized that histone H3 methylation may be regulated in a concentration- and/or time-dependent manner.

To test whether alteration of mono-methyl histone H3 (Lys 4) in carcinogen-treated HMEC was concentration dependent, HMEC were exposed to each carcinogen individually at 5, 20 or 40 µM for 72 h. As shown in Figure 5, HMEC exposed to increasing concentrations of either PhIP or ABP resulted in mono-methyl histone H3 (Lys 4) being reduced in a carcinogen concentration-dependent manner. In contrast, HMEC exposure to increasing concentrations of B[a]P did not affect the methylation of mono-methyl histone H3 (Lys 4), suggesting that the modification of histone H3 is carcinogen specific. Although exposure to 5 µM for 48 h did not cause any change in histone H3 mono-methylation, we wondered if longer treatment would induce changes. However, even exposure for 72 h did not elicit any change in mono-methylation status.


Figure 5
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Histone H3 and mono-methyl histone H3 Western analysis of HMEC treated with varying concentrations of individual carcinogens. HMEC were treated with PhIP, B[a]P or ABP at 5, 20 or 40 µM for 72 h. Histone methylation was not affected by carcinogen treatment at low concentrations (5 µM). However, HMEC exposed to either PhIP or ABP at higher concentrations (20, 40 µM) significantly increased histone H3 demethylation. Methylation of histone H3 was not altered by B[a]P treatment.

 
To determine whether the altered methylation pattern can be reversed, HMEC were exposed to a combination of carcinogens (PhIP, ABP, B[a]P) at 10 µM for 72 h, medium was then removed and fresh medium added for 48 h. Immunofluorescence analysis revealed a nuclear staining pattern for histone H3 in control HMEC as well as treatment groups (Figure 6A). Localization of mono-methyl histone H3 (Lys 4) was observed in the nuclei of control HMEC as confirmed by nuclear counterstaining with DAPI (Figure 6A, inset). However, after 72 h carcinogen treatment, there was a significant reduction in the levels of mono-methyl histone H3 (Lys 4) as expected. Reduced levels of mono-methyl histone H3 (Lys 4) present in the nucleus resulted in only the nucleus being counterstained (Figure 6A, inset). Within 48 h after treatment, the levels of mono-methyl histone H3 (Lys 4) were restored and co-localization was detected with nuclear counterstain (inset).


Figure 6
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. LSD1 localization pattern in control and carcinogen-treated HMEC. (A) Immunocytochemistry. HMEC were exposed to a combination (ABP, B[a]P, PhIP) of carcinogens each at 10 µM for 72 h. Medium was removed after 72 h and replaced with fresh medium for 48 h. Paraffin-embedded control and carcinogen-treated HMEC were stained with anti-histone H3 (row 1), anti-LSD1 (row 2) and anti-mono-methyl histone H3 (Lys 4) (row 3). Insets: single cells stained with the antibody indicated and counterstained with DAPI. Magnification x400; insets x600. (B) Western immunoblot. Nuclear extracts of control and PhIP-treated HMEC were separated on a 10% Tris/Glycine gel. Proteins were transferred onto polyvinylidene difluoride and blotted with anti-LSD1 antibody. Nuclear LSD1 protein levels increased in carcinogen-treated HMEC compared with control HMEC.

 
LSD1 and its potential role in demethylating mono-methyl histone H3 (Lys 4) in carcinogen-treated HMEC
LSD1 was recently identified as a demethylating enzyme specific for the demethylation of histone H3 (Lys 4) mono- and di-methyl (28,29). Based on these findings, we hypothesized that carcinogen exposure might induce LSD1 expression. To determine whether LSD1 is involved in demethylating mono-methyl histone H3 (Lys 4) in carcinogen-treated HMEC, we examined the localization patterns and protein expression levels of LSD1 in control and carcinogen-treated HMEC. Immunofluorescence studies were performed using control and carcinogen-treated HMEC (PHIP, B[a]P, ABP 10 µM 72 h treatment and PhIP, B[a]P, ABP 10 µM 72 h treatment + 48 h fresh medium) to address the LSD1 localization pattern (Figure 6A). Interestingly, selective cells that expressed LSD1 demonstrated a unique staining and localization pattern. In control HMEC, LSD1 was localized predominantly to the nuclear periphery in a punctate manner. LSD1 localization to the nuclear periphery was confirmed by counterstaining the nuclei with DAPI (Figure 6A, inset). After 72 h treatment, LSD1 localization was altered and stained primarily within the nucleus in a homogeneous manner (Figure 6A). Co-localization of LSD1 and nuclear counterstain was detected (inset). The localization of LSD1 reversed back to its original location of the nuclear periphery within 48 h after treatment comparable with LSD1 localization observed in control HMEC. Utilization of another anti-LSD1 antibody also revealed only selective staining (results not shown). The carcinogen-induced LSD1 localization pattern correlated with demethylation of mono-methyl histone H3 (Lys 4), which was also shown to be reversible. HMEC treated with increasing concentrations of PhIP were used to determine LSD1 protein expression levels. Western analysis of nuclear extracts revealed little LSD1 in control HMEC and increased expression in carcinogen-treated HMEC (Figure 6B). Therefore, our localization and protein expression studies of LSD1 strongly support our hypothesis that carcinogen treatment induces LSD1 expression resulting in mono-methyl histone H3 (Lys 4) demethylation in treated HMEC.

An increase in mono-methyl histone H3 (Lys 4) demethylation is specific to carcinogen treatment
To address the concern whether other types of cell stress are capable of inducing similar protein alterations in mammary epithelium, HMEC were either heat shocked at 42°C for 2 h (30) or treated with 0.2 mM H2O2 for 24 h (31). Western analysis confirmed that mono-methyl histone H3 (Lys 4) methylation was not affected by non-specific cell stress, such as heat shock or hydrogen peroxide treatment (Figure 7). HMEC cultured at 37°C or exposed to 0.001% dimethyl sulphoxide served as controls, respectively. Mono-methyl histone H3 (Lys 4) methylation was detectable in both control samples, as expected. These data provide evidence that the protein alterations we observed in mammary epithelial cells are specific to carcinogen treatment.


Figure 7
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7. Altered mono-methyl histone H3 methylation pattern in treated HMEC is specific to carcinogen treatment. HMEC were either heat shocked at 42°C for 2 h or treated with 0.2 mM H2O2 for 24 h. Histone H3 and mono-methyl histone H3 (Lys 4) methylation was not affected by non-specific cell stress, such as heat shock or hydrogen peroxide treatment, indicating that the identified protein alteration is specific to carcinogen treatment.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
All women are exposed to dietary and environmental carcinogens and due to the lipophilicity of these agents they may accumulate in mammary tissue leading to carcinogen–DNA interactions. This was demonstrated by the detection of PhIP, ABP and B[a]P adducts in DNA isolated from exfoliated ductal epithelial cells in milk from normal healthy, non-smoking women (32). We selected these compounds to examine the effects of carcinogen exposure on early protein alterations in normal mammary epithelium. To identify target proteins that may be altered following carcinogen exposure, ScFv antibodies were screened and selected from a phage display antibody library.

Phage display is becoming a widely used technique in a number of applications some of which include affinity screening of peptide libraries to identify ligands for peptide receptors, to assign epitopes for monoclonal antibodies, to select enzymes substrates and to screen cloned antibody repertoires (33–35). This technique offers several advantages over traditional screening methods such as the vast diversity of variant proteins that can be represented (36), the selections can either be performed in vivo or in vitro (37–39), and selection conditions and stringencies can be altered (34,40). Overall, phage display is relatively inexpensive, simple to perform and allows high-throughput detection. Microarray technology offers similar advantages but is limited by available arrays and lack of standardized data analysis (41). Moreover, mRNA levels are not informative of proteins that are post-translationally modified, such as histones.

In this study, the phage display screen in combination with MALDI-TOF identified a nuclear protein, histone H3, as a target antigen. Histones are proteins that are conserved among species from yeast to human (42). These small proteins, ranging from 13–17 kDa, are highly basic with numerous lysine and arginine residues located on the N-terminal tail and the histone core (42). Due to their basic charge, they associate strongly with acidic DNA in order to compactly condense chromatin into the nucleus (42). More importantly, post-translational modifications (methylation, acetylation, phosphorylation) that occur on histone lysine and arginine residues lead to the recruitment of other binding proteins and can, ultimately, regulate chromatin structure and gene transcription (43–46). Histone methylation, in particular, can either be mono-, di- or tri-methylated with each level of modification serving as an additional regulatory control to chromatin accessibility. Since histone H3 total protein was not altered as a result of carcinogen exposure, we focused our attention on characterizing alterations of histone H3, specifically mono-methyl histone H3 (Lys 4), in carcinogen-treated HMEC.

As shown in Figure 4, the exposure to four carcinogens resulted in a virtual elimination of mono-methyl histone H3 (Lys 4) in treated HMEC. This means that even after cross-absorption of ScFv during the phage selection process, the relative excess of mono-methyl histone H3 (Lys 4) in control HMEC lysate was sufficient to be distinguished by FMAT. Although the ScFv recognized and immunoprecipitated mono-methyl histone H3 (Lys 4) in the lysates from control HMEC, the standard MALDI-TOF did not detect this post-translational modification because it was calibrated to search for changes across the entire protein spectrum. At that time, we were unaware that a post-translational modification may be involved and that more specialized MALDI techniques capable of identifying mono-methyl histone H3 (Lys 4) would be needed. Therefore, the MALDI-TOF analysis employed in this study simply identified histone H3. Full characterization of the specific post-translational modification that was altered as a result of carcinogen treatment was based on the use of modification-specific histone antibodies.

The decrease in mono-methyl histone H3 (Lys 4) observed in treated HMEC was concentration dependent (Figure 5). Although PhIP, ABP, B[a]P have each been shown to induce mammary tumors in animal models (12,13,15–17), only two, i.e. PhIP and ABP, induced mono-methyl histone H3 (Lys 4) alterations. While we cannot explain this differential effect, it is interesting to note that the heterocyclic and aromatic amines (PhIP and ABP) have an amine group in common. Histone H3 protein alterations did not occur in response to random cell stress, such as heat shock or hydrogen peroxide treatment, but were in fact specific to the carcinogen treatment (Figure 7).

For years, many believed that methylation was a permanent histone mark unlike other types of modifications, such as phosphorylation and acetylation, which are rapidly regulated by specialized enzymes (47). However, recently, LSD1 was identified which catalyzes the removal of methyl groups via an amine oxidase reaction (48). LSD1, a flavin-containing amine oxidase, specifically targets methylated amine groups on mono- and di-methylated histone H3 (Lys 4) substrates (28,29,48,49). As mentioned above, increased expression of LSD1 and the associated demethylation of histone H3 (Lys 4) were only evident in HMEC treated individually with carcinogens containing an amine group, PhIP and ABP. Our study revealed the dynamics of LSD1 distribution from the nuclear periphery prior to carcinogen treatment, movement into the nucleus during treatment and relocalization of the enzyme to the nuclear periphery after carcinogen treatment (Figure 6A). Moreover, our data suggest that carcinogen-exposed HMEC upregulate LSD1. The combined upregulation and nuclear localization are mechanisms responsible for altering histone H3 methylation at lysine residue 4. This finding supports recent evidence that the turnover of histone lysine methylation is faster than previously assumed (46). In control, HMEC LSD1 was predominately localized to the nuclear periphery, which may explain why the Western analysis of nuclear extract revealed little LSD1 (Figure 6B). Our observation is in accordance with previous reports that demonstrate localization of LSD1 to the nuclei of HeLa cervical carcinoma cells and LnCaP prostate tumor cells (28,29). More importantly, our results suggest that LSD1 is a carcinogen-induced protein alteration, which may have implications in early-stage carcinogenesis. It is unclear, however, why LSD1 is expressed in only a few control HMEC and in an even smaller number of treated HMEC. It may be possible that additional post-translational modifications, interacting protein-binding partners or the activation status of LSD1 influences this selective expression pattern.

The emerging field of modified histones and their potential role in cancer biology is becoming a growing area of interest. Carcinogen exposure leads to DNA adduct formation (1–4) and, as demonstrated in our study, can also induce histone H3 protein alterations. While the DNA adduct formation is likely to be dispersed indiscriminately throughout the genome, the alteration of H3 is limited to one specific residue, i.e. lysine 4. Recent studies suggest that the positioning of nucleosomes is linked to DNA sequence (50). Alteration of either the DNA or histone by carcinogens would therefore be expected to regulate chromatin structure and gene transcription (43,44,46,50). Recent studies provide evidence that histone H3 K4 methylation affects gene transcription and plays a potential role in carcinogenesis. RNAi inhibition of LSD1 caused an increase in H3 K4 methylation and concomitant derepression of target genes, suggesting that LSD1 represses transcription via histone demethylation (28). One of the target genes negatively regulated by LSD1 was p57Kip2, a cyclin-dependent kinase inhibitor implicated in tumorigenesis. An H3 K4 methylase, SMYD3, has been shown to be upregulated in colorectal and hepatocarcinoma cells (51). Overproduction of SMYD3 increased cell proliferation dependent on the histone methylase activity. It is noteworthy that histone modification by histone deacetylase inhibitors leads to differentiation, growth arrest and apoptosis (52,53). Although transformed cells are 10-fold more sensitive to histone deacetylase inhibitors than normal cells, the exact anticancer mechanism is still unknown (54–57). Interestingly, a recent study demonstrated that histone deacetylase and demethylase enzymatic activities are functionally linked, suggesting a potential mechanism of action for histone deacetylase inhibitors (58). Furthermore, recent findings indicate that global levels of histone modifications are predictive of prostate cancer clinical outcome in individuals with low-grade prostate cancer (59). This evidence further strengthens the significance of our findings in an effort to better understand the implications of altered histone H3 methylation patterns in cancer research.

In summary, we were able to detect a carcinogen-induced protein alteration in HMEC using a phage display antibody library. In combination with a high-throughput selection technique, FMAT, a ScFv antibody was isolated. Immunoassays using this ScFv antibody demonstrated a differential expression pattern of a target antigen in control and treated HMEC. Interestingly, MALDI-TOF analysis identified the target antigen as histone H3. Although it was surprising for a housekeeping protein, such as histone H3, to be identified, characterization of histone H3 and its common modifications revealed that mono-methyl histone H3 (Lys 4) was specifically altered by carcinogen treatment. Decreased mono-methyl histone H3 (Lys 4) levels in treated HMEC were concentration dependent in PhIP- and ABP-treated HMEC and carcinogen specific. In addition, we were able to show that carcinogen exposure altered the localization of a mono-methyl histone H3 (Lys 4)-specific demethylase, LSD1, and upregulated its protein expression compared with control HMEC. Furthermore, removal of carcinogens from the cell culture medium reversed LSD1 localization and restored mono-methyl histone H3 (Lys 4) levels. In conclusion, our data provide direct evidence linking carcinogen exposure to upregulation of a single enzyme, LSD1, which leads to the demethylation of mono-methyl histone H3 (Lys 4). Exact mechanisms responsible for inducing LSD1 protein expression and localization changes are yet to be determined. Nevertheless, detection of early carcinogen-induced protein alterations, such as mono-methyl histone H3 (Lys 4), could serve as a common ‘signature’ for mammary epithelium that has been exposed to environmental and dietary carcinogens and ultimately lead to improved diagnostic, prognostic and therapeutic strategies.


   Acknowledgments
 
We thank Dr David Friedman of the Vanderbilt Proteomics Laboratory, Mass Spectrometry Research Center for performing the MALDI-TOF analysis and Violeta Sanchez for assisting with antigen retrieval processing. This work was supported in part by National Institutes of Health grants 1RO1CA83752 (C.B., R.v.d.M., N.R. and F.F.P.), T32 HL007751-12 (C.B.), RO1CA109355 (M.B.C. and Z-W.S.) Vnaderbilt-Ingram Cancer Center and The Robert J. Kleberg, Jr. and Helen C. Kleberg Foundation (M.B.C. and Z-W.S.) and American Cancer Society grant IRG-58-009-46 (M.B.C. and Z-W.S.).

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Snyderwine EG, et al. Proliferation, development and DNA adduct levels in the mammary gland of rats given 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine and a high fat diet. Carcinogenesis (1998) 19:1209–1215.[Abstract/Free Full Text]
  2. Dingley KH, et al. DNA and protein adduct formation in the colon and blood of humans after exposure to a dietary-relevant dose of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine. Cancer Epidemiol. Biomarkers Prev. (1999) 8:507–512.[Abstract/Free Full Text]
  3. Fang M, et al. Urinary N2-(2'-deoxyguanosin-8-yl)PhIP as a biomarker for PhIP exposure. Carcinogenesis (2004) 25:1053–1062.[Abstract/Free Full Text]
  4. Fields WR, et al. Differential c-myc expression profiles in normal human bronchial epithelial cells following treatment with benzo[a]pyrene, benzo[a]pyrene-4,5 epoxide, and benzo[a]pyrene-7,8-9,10 diol epoxide. Mol. Carcinog. (2004) 40:79–89.[CrossRef][Web of Science][Medline]
  5. Ellsworth DL, et al. Genomic instability in histologically normal breast tissues: implications for carcinogenesis. Lancet Oncol. (2004) 5:753–758.[CrossRef][Web of Science][Medline]
  6. Kendall SD, et al. A network of genetic events sufficient to convert normal human cells to a tumorigenic state. Cancer Res. (2005) 65:9824–9828.[Abstract/Free Full Text]
  7. Emberley ED, et al. Psoriasin interacts with Jab1 and influences breast cancer progression. Cancer Res. (2003) 63:1954–1961.[Abstract/Free Full Text]
  8. Gauthier ML, et al. p38 regulates cyclooxygenase-2 in human mammary epithelial cells and is activated in premalignant tissue. Cancer Res. (2005) 65:1792–1799.[Abstract/Free Full Text]
  9. Nagaraja GM, et al. Gene expression signatures and biomarkers of noninvasive and invasive breast cancer cells: comprehensive profiles by representational difference analysis, microarrays and proteomics. Oncogene (2006) 25:2328–2338.[CrossRef][Web of Science][Medline]
  10. Yao J, et al. Combined cDNA array comparative genomic hybridization and serial analysis of gene expression analysis of breast tumor progression. Cancer Res. (2006) 66:4065–4078.[Abstract/Free Full Text]
  11. Stausbol-Gron B, et al. De novo identification of cell-type specific antibody-antigen pairs by phage display subtraction. Isolation of a human single chain antibody fragment against human keratin 14. Eur. J. Biochem. (2001) 268:3099–3107.[Web of Science][Medline]
  12. Ito N, et al. A new colon and mammary carcinogen in cooked food, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). Carcinogenesis (1991) 12:1503–1506.[Abstract/Free Full Text]
  13. Snyderwine EG. Some perspectives on the nutritional aspects of breast cancer research. Food-derived heterocyclic amines as etiologic agents in human mammary cancer. Cancer (1994) 74:1070–1077.[CrossRef][Web of Science][Medline]
  14. Sutter TR, et al. Targets for dioxin: genes for plasminogen activator inhibitor-2 and interleukin-1 beta. Science (1991) 254:415–418.[Abstract/Free Full Text]
  15. Perera FP, et al. Carcinogen-DNA adducts in human breast tissue. Cancer Epidemiol. Biomarkers Prev. (1995) 4:233–238.[Abstract]
  16. Li D, et al. Aromatic DNA adducts in adjacent tissues of breast cancer patients: clues to breast cancer etiology. Cancer Res. (1996) 56:287–293.[Abstract/Free Full Text]
  17. Phillips DH. Polycyclic aromatic hydrocarbons in the diet. Mutat. Res. (1999) 443:139–147.[Web of Science][Medline]
  18. Gould MN, et al. Human breast cell-mediated mutagenesis of mammalian cells by polycyclic aromatic hydrocarbons. Cancer Res. (1982) 42:4619–4624.[Abstract/Free Full Text]
  19. Debiec-Rychter M, et al. DNA adduct formation in human and rat mammary epithelium by N-hydroxy derivatives of 2-aminofluorene and 4-aminobiphenyl. Teratog. Carcinog. Mutagen. (1998) 18:35–39.[CrossRef][Web of Science][Medline]
  20. Jeffy BD, et al. Epigenetics of breast cancer: polycyclic aromatic hydrocarbons as risk factors. Environ. Mol. Mutagen. (2002) 39:235–244.[CrossRef][Web of Science][Medline]
  21. Zhang Y, et al. Selection of active ScFv to G-protein-coupled receptor CCR5 using surface antigen-mimicking peptides. Biochemistry (2004) 43:12575–12584.[CrossRef][Medline]
  22. Pope T, et al. Construction and use of antibody gene repertories. In: Antibody Engineering: A practical approach—McCafferty J, Chiswell D, Hoogenboom H, eds. (1996) IRL. 1–40.
  23. Martinez-Torrecuadrada J, et al. Targeting the extracellular domain of fibroblast growth factor receptor 3 with human single-chain Fv antibodies inhibits bladder carcinoma cell line proliferation. Clin. Cancer Res. (2005) 11:6280–6290.[Abstract/Free Full Text]
  24. Friedman DB, et al. Proteome analysis of human colon cancer by two-dimensional difference gel electrophoresis and mass spectrometry. Proteomics (2004) 4:793–811.[CrossRef][Web of Science][Medline]
  25. Gallagher S, et al. Immunoblotting and immunodection. In: Current protocols in Molecular Biology—Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, StruhlK, eds. (2004) John Wiley & Sons, Inc. 10.8.1–10.8.24.
  26. Hendzel MJ, et al. Fixation-dependent organization of core histones following DNA fluorescent in situ hybridization. Chromosoma (1997) 106:114–123.[CrossRef][Web of Science][Medline]
  27. Choi HS, et al. Phosphorylation of Ser28 in histone H3 mediated by mixed lineage kinase-like mitogen-activated protein triple kinase alpha. J. Biol. Chem. (2005) 280:13545–13553.[Abstract/Free Full Text]
  28. Shi Y, et al. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell (2004) 119:941–953.[CrossRef][Web of Science][Medline]
  29. Metzger E, et al. LSD1 demethylates repressive histone marks to promote androgen-receptor-dependent transcription. Nature (2005) 437:436–439.[Medline]
  30. Sheffield LG. Epidermal growth factor as an autocrine modulator of stress response in mammary epithelial cells. J. Endocrinol. (1998) 159:111–116.[Abstract]
  31. Mori K, et al. Invasive potential induced under long-term oxidative stress in mammary epithelial cells. Cancer Res. (2004) 64:7464–7472.[Abstract/Free Full Text]
  32. Gorlewska-Roberts K, et al. Carcinogen-DNA adducts in human breast epithelial cells. Environ. Mol. Mutagen. (2002) 39:184–192.[CrossRef][Web of Science][Medline]
  33. Peletskaya EN, et al. Characterization of peptides that bind the tumor-associated Thomsen-Friedenreich antigen selected from bacteriophage display libraries. J. Mol. Biol. (1997) 270:374–384.[CrossRef][Web of Science][Medline]
  34. Smith GP, et al. Phage display. Chem. Rev. (1997) 97:391–410.[CrossRef][Web of Science][Medline]
  35. Koivunen E, et al. Tumor targeting with a selective gelatinase inhibitor. Nat. Biotechnol. (1999) 17:768–774.[CrossRef][Web of Science][Medline]
  36. Willats WG. Phage display: practicalities and prospects. Plant Mol. Biol. (2002) 50:837–854.[CrossRef][Web of Science][Medline]
  37. Pasqualini R, et al. Organ targeting in vivo using phage display peptide libraries. Nature (1996) 380:364–366.[CrossRef][Medline]
  38. Arap W, et al. Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model. Science (1998) 279:377–380.[Abstract/Free Full Text]
  39. Newton JR, et al. In vivo selection of phage for the optical imaging of PC-3 human prostate carcinoma in mice. Neoplasia (2006) 8:772–780.[CrossRef][Web of Science][Medline]
  40. Zou J, et al. Biodistribution of filamentous phage peptide libraries in mice. Mol. Biol. Rep. (2004) 31:121–129.[CrossRef][Web of Science][Medline]
  41. Russo G, et al. Advantages and limitations of microarray technology in human cancer. Oncogene (2003) 22:6497–6507.[CrossRef][Web of Science][Medline]
  42. Bode AM, et al. Inducible covalent posttranslational modification of histone H3. Sci. STKE (2005) 2005:re4.[Abstract/Free Full Text]
  43. Jenuwein T, et al. Translating the histone code. Science (2001) 293:1074–1080.[Abstract/Free Full Text]
  44. Kubicek S, et al. A crack in histone lysine methylation. Cell (2004) 119:903–906.[CrossRef][Web of Science][Medline]
  45. Cheung P, et al. Epigenetic regulation by histone methylation and histone variants. Mol. Endocrinol. (2005) 19:563–573.[Abstract/Free Full Text]
  46. Trojer P, et al. Histone lysine demethylases and their impact on epigenetics. Cell (2006) 125:213–217.[CrossRef][Web of Science][Medline]
  47. Byvoet P, et al. The distribution and turnover of labeled methyl groups in histone fractions of cultured mammalian cells. Arch. Biochem. Biophys. (1972) 148:558–567.[CrossRef][Web of Science][Medline]
  48. Bannister AJ, et al. Reversing histone methylation. Nature (2005) 436:1103–1106.[CrossRef][Medline]
  49. Forneris F, et al. Human histone demethylase LSD1 reads the histone code. J. Biol. Chem. (2005) 280:41360–41365.[Abstract/Free Full Text]
  50. Segal E, et al. A genomic code for nucleosome positioning. Nature (2006) 442:772–778.[CrossRef][Medline]
  51. Hamamoto R, et al. SMYD3 encodes a histone methyltransferase involved in the proliferation of cancer cells. Nat. Cell Biol. (2004) 6:731–740.[CrossRef][Web of Science][Medline]
  52. Marks P, et al. Histone deacetylases and cancer: causes and therapies. Nat. Rev. Cancer (2001) 1:194–202.[CrossRef][Medline]
  53. Secrist JP, et al. HDAC inhibitors for the treatment of cancer. Curr. Opin. Investig. Drugs (2003) 4:1422–1427.[Medline]
  54. Richon VM, et al. A class of hybrid polar inducers of transformed cell differentiation inhibits histone deacetylases. Proc. Natl Acad. Sci. USA (1998) 95:3003–3007.[Abstract/Free Full Text]
  55. Scott GK, et al. Transcriptional repression of ErbB2 by histone deacetylase inhibitors detected by a genomically integrated ErbB2 promoter-reporting cell screen. Mol. Cancer Ther. (2002) 1:385–392.[Abstract/Free Full Text]
  56. Kelly WK, et al. Phase I clinical trial of histone deacetylase inhibitor: suberoylanilide hydroxamic acid administered intravenously. Clin. Cancer Res. (2003) 9:3578–3588.[Abstract/Free Full Text]
  57. Marks PA, et al. Histone deacetylases. Curr. Opin. Pharmacol. (2003) 3:344–351.[CrossRef][Web of Science][Medline]
  58. Lee MG, et al. Functional interplay between histone demethylase and deacetylase enzymes. Mol. Cell. Biol. (2006) 26:6395–6402.[Abstract/Free Full Text]
  59. Seligson DB. Global histone modification patterns predict risk of prostate cancer recurrence. Nature (2005) 435:1262–1266.[CrossRef][Medline]
Received January 17, 2007; revised April 11, 2007; accepted April 17, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/10/2184    most recent
bgm100v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Bradley, C.
Right arrow Articles by Parl, F. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bradley, C.
Right arrow Articles by Parl, F. F.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?