Breast Cancer Center of Excellence

Home | About | Alcohol Risk | Members | Resources | News |
Abstracts | Publications | Meeting

Meeting Abstracts

American Association for Cancer Research, 97th Annual Meeting, April 1 to 5, 2006, Washington, DC.

Poster Presentation: Malondialdehyde-DNA adducts in normal breast tissues

Ramona G. Dumitrescu, Armelle Munnia, Scott Spear, Jo L. Freudenheim, Peter G. Shields, Marco Peluso. Georgetown University Medical Ctr., Washington, DC, Tuscany Cancer Institute-CSPO, Florence, Italy, Bufallo University, Buffalo, NY.

Malondialdehyde (MDA) is a product of lipid-peroxidation that is also formed during prostaglandin biosynthesis via cyclooxygenase. MDA is a highly reactive electrophile that reacts with DNA, generating MDA-DNA adducts. DNA adduct levels reflect the biologically effective dose of oxygen radicals, and might be mechanistically related to breast carcinogenesis. Many breast cancer risk factors are known, but many patients do not have identifiable risk factors. Alcohol drinking is a known non-hormonal risk factor for breast cancer. Excessive alcohol consumption can increase oxidative stress and generates free radicals and lipid peroxidation.
Purpose
Determine the MDA-DNA adducts level in normal breast tissues from reduction mammoplasty subjects in order to examine associations with environmental factors which may induce oxidative stress.
Hypothesis
There is an interindividual variation in the MDA-DNA adducts level in these subjects, reflecting a different internal level of DNA damage as a result of the exposure to alcohol or other agents.
Methods
Breast tissue was collected from 55 normal women without history of breast cancer, undergoing reduction mammoplasty surgery. Questionnaire data were collected 48 hours before the surgery regarding alcohol consumption, smoking and other relevant exposures. DNA was isolated from frozen tissues. The production of MDA-DNA adducts was investigated using ³²P-postlabelling techniques. The association between the MDA-DNA adducts and alcohol consumption, smoking, age, race and family history of breast cancer was estimated using the Wilcoxon Rank Sum test.
Results
We observed more than 100-fold variation in MDA-adduct levels among these normal women. The mean MDA-DNA adducts were not significantly different by drinking status comparing regular drinkers to never drinkers (2.6 +/- 5.1 versus 2.3 +/- 3.0). There were differences for smoking status (smokers versus nonsmokers; 2.2 +/- 2.8 versus 1.0 +/- 0.9), race (African-American women versus Caucasian women; 3.7 +/- 6.1 versus 1.7 +/- 1.9) and for women who had history of breast cancer in the family versus those without (4.5 +/- 7.2 versus 1.7 +/- 2.0). However, using Wilcoxon Rank Sums test, the differences for these subgroups were not statistically significantly different.
Conclusions
Normal women undergoing reduction mammoplasty have variable level of MDA-DNA adducts in their breast tissue and this may reflect the internal level of DNA damage. There were higher adduct levels in the breast tissues of smokers, African-American women and women with a family history of breast cancer, but the differences were not statistically significantly different, likely owing to the small sample size. This study set will be expanded to improve statistical power.

10th International Workshop on Molecular and Cellular Biology of Plasminogen Activation, April 9 to 13, 2005, Washington, DC.

Hepatocyte growth factor activator inhibitor 1 (HAI-1): Role in matriptase function beyond a conventional inhibitor.

Oberst M, Chen LL, Kiyomiya K, Williams CA, Lee MS, Johnson MD, DicksonRB, and *Lin CY Department of Oncology, Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, DC 20057.

Matriptase and its inhibitor, hepatocyte growth factor activator inhibitor 1 (HAI-1), represent a cognate pair, a cell surface serine protease and a Kunitz-type serine protease inhibitor. Both are broadly expressed by epithelial elements in most epithelium-containing tissues and a variety of epithelium-derived human tumors. Matriptase has been implicated in cancer invasion and metastasis by serving as a cell surface activator for urokinase type plasminogen activator (uPA), HGF, and protease activated receptor 2 (PAR-2). In addition to serving as matriptase endogenous inhibitor, HAI-1 is paradoxically required for matriptase activation, as coexpression of wild-type matriptase with HAI-1 mutant in its LDL receptor domain abolishes matriptase activation. We further observed that the cellular ratio of HAI-1 to matriptase was maintained during matriptase translocation and activation at cell-cell junctions, a process induced by sphingosine 1-phosphate (S1P) in human mammary epithelial cells. However, when this ratio was changed by cell treatment with HAI-1 siRNA or anti-HAI-1 mAb M19, spontaneous activation of matriptase occurred in the absence of S1P-induced translocation; S1P-induced matriptase activation was also enhanced. These results support a role for HAI-1 in protection of cell from uncontrolled matriptase activation. We next expressed matriptase, either alone or with HAI-1 in breast cancer cells that do not endogenously express either protein. A defect in matriptase trafficking to the cell surface occurred if wild type matriptase was expressed in the absence of HAI-1; this defect appeared to result from matriptase toxicity to cells. Coexpression with matriptase of wild type HAI-1, but not HAI-1 mutants altered in its Kunitz domain 1, corrected the trafficking defect. In contrast, catalytically defective matriptase mutants were normal in their trafficking in the absence of HAI-1. These results are also consistent with a role for HAI-1 to prevent inappropriate matriptase proteolytic activity during its protein synthesis and trafficking. Taken together, these results support multiple roles for HAI-1 to regulate matriptase, including its proper expression, intracellular trafficking, activation, and inhibition.

American Association for Cancer Research, 96th Annual Meeting, April 16 to 20, 2005, Anaheim, CA. 

Glutathione-S-transferase M1 (GSTM1), total suspended particulate exposure and breast cancer.

Matthew R. Bonner, Jo L. Freudenheim, Daikwon Han, Jing Nie, Qing Lan, Peter Rogerson, John E. Vena, Stephen B. Edge, Peter Shields. Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, NIH, DHHS. Bethesda, MD 20892

Previously we found that postmenopausal women exposed to relatively high concentrations of total suspended particulates (TSP) were at increased risk for breast cancer. In Western New York, polycyclic aromatic hydrocarbons (PAHs) are important constituents of TSP. In addition, PAHs have been hypothesized to act as mammary carcinogens. GSTM1 detoxifies carcinogenic PAH metabolites and women who lack this enzyme (null homozygotes) may be at increased risk for breast cancer from TSP exposure compared with women who have this enzyme (carriers). We used the Western New York Exposures and Breast Cancer Study (WEB) to investigate GSTM1, TSP and the risk of breast cancer. Cases (n = 818) and controls (n = 1,525) were genotyped for GSTM1. TSP concentration for study participants’ domiciles at five time periods: 1) birth, 2) menarche, 3) the time of the study participant’s birth, 4) 20 years and 5) 10 years prior to diagnosis (cases) or interview (controls) were estimated with inverse distance squared spatial interpolation within a geographic information system (GIS). Cumulative exposure to TSP was calculated as the sum of domicile TSP concentrations at all five time periods. Unconditional logistic regression was used to calculate odds ratios (OR) and 95% confidence intervals (CI) adjusting for age, education, and parity. GSTM1 genotype did not modify the effect of cumulative exposure to TSP for either pre- or postmenopausal women. Among premenopausal women, there was no evidence of an association between TSP and breast cancer risk overall or within strata of GSTM1 (null homozygotes or carriers). Among postmenopausal women, however, cumulative exposure to TSP was associated with the risk of breast cancer, although GSTM1 genotype did not modify the effect of exposure. For postmenopausal women comparisons of those with high cumulative exposure to TSP to those with low exposure, ORs were 3.7 (95% CI = 1.1-12.9) and 3.8 (95% CI = 1.3-11.2) for GSTM1 null and carriers, respectively. Similarly, the effect of TSP exposure at the time of birth was not modified by GSTM1 genotype. Among postmenopausal with a GSTM1 null genotype, high TSP exposure at birth was associated a non-significant 2-fold (OR = 1.9; 95% CI = 0.5-8.0) increase in the risk of breast cancer, while high TSP exposure was not associated with breast cancer among GSTM1 carriers (OR = 1.0; 95% CI = 0.3-3.6); although the p for interaction was not significant (p= 0.4). In summary, GSTM1 genotype did not modify our previous observation of increased risk of postmenopausal breast cancer associated with cumulative exposure to TSP. For TSP exposure at the time of birth, however, the effect that GSTM1 genotype may have on the risk of breast cancer is less clear due to a small number of cases and controls in the low exposed (referent) category.

American Association for Cancer Research, 96th Annual Meeting, April 16 to 20, 2005, Anaheim, CA. 

Geographic differences in breast cancer cases and Controls by Genetic and Biological Characteristics: Explaining Clustering of Breast Cancer at Place of Birth.

Han, D. Freudenheim, JL. Nie, J. Bonner, MR. Muti, P. Trevisan, M. Vito, D. Edge, SB. Luyegu, K. Shields, P. Department of Social and Preventive Medicine, University at Buffalo, Buffalo, NY.

There is growing evidence that early environmental exposures may be related to risk of breast cancer. We had identified previously geographic clustering of residences in early life, especially at birth, in relation to breast cancer risk. Because women with glutathione-S-transferase M1 (GSTM1) null genotypes may be more sensitive to environmental insults, we report here on clustering by GST genotype. Further, we examined whether there were differences by estrogen receptor (ER) or progesterone receptor (PR) status. We conducted a population-based case control study of incident, primary, histologically-confirmed breast cancer with controls frequency matched to cases on age, race and county of residence (the WEB study). All participants provided lifetime residential histories; we report here on place of birth. ER and PR status were obtained from medical records, and GST genotype was determined by PCR and agarose gel resolution. Geographic differences in clustering of breast cancer cases and controls were examined by the k-function method, a test for general tendency of spatial clustering, in groups stratified by GST genotype and by ER and PR status. We found evidence that breast cancer cases with GSTM1 null genotype were more clustered than controls, while there was no such evidence among GSTM1 wild genotype. However, there was no indication that cases with GSTM1 null genotype were more clustered when compared to cases with the wild genotype. We also observed a tendency to clustering among those with ER+ tumors relative to controls, but not among other groups defined by either ER or PR status. Clustering of women with GSTM1 null genotype may indicate that breast cancer cases with GSTM1 null genotype are more likely to share common environmental exposures at place of birth, but not necessarily congregated in a specific geographic area, and that such genetic susceptibility to early environmental exposure may play a role in subsequent risk of breast cancer. The tendency for clustering of ER positive tumors may also indicate that there are effects of common exogenous exposures related to place of birth among ER positive tumors. These findings are provocative in providing an indication that exogenous exposures at the time of birth may affect breast cancer risk.

American Association for Cancer Research, 96th Annual Meeting, April 16 to 20, 2005, Anaheim, CA. 

Simultaneous activation and HAI-1-mediated inhibition of matriptase induced at activation foci in human mammary epithelial cells

Ming-Shyue Lee, Ken-ichi Kiyomiya, Christelle Benaud, Robert B. Dickson, and Chen-Yong Lin. Department of Oncology, Lombardi Cancer Center, Georgetown University Medical Center, Washington, DC 20057.

Activation of single chain, latent matriptase, a type II transmembrane serine protease, depends on the weak proteolytic activity of its own zymogen, as well as its cognate inhibitor, hepatocyte growth factor activator inhibitor 1 (HAI-1). Oligomerization of matriptase zymogens, HAI-1, and probably its interaction with other proteins has been proposed to occur during matriptase activation. In the current study, we examined the cellular events associated with matriptase activation, triggered either by the physiological inducer, sphingosine 1-phosphate (S1P) or by a chemical inducer, the polyanionic compound suramin. S1P-induced matriptase translocation to cell-cell contacts, where it is activated, is an F-actin polymerization-dependent process. Conversely, suramin-induced matriptase accumulation and activation at vesicle-like structures is an F-actin polymerization-independent process. While matriptase activation can occur at different subcellular locations, both S1P- and suramin-induced matriptase accumulation forms unique subcellular structures, termed activation foci, where oligomerization of matriptase zymogens and HAI-1 may occur, promoting matriptase activation. Furthermore, matriptase activation may be regulated by intracellular signaling, since Ro-31-8220, a bisindolylmaleimide protein kinase C inhibitor, inhibited both S1P- and suramin-induced activation. The requirement of HAI-1 for matriptase activation and the co-localization of HAI-1 and matriptase in activation foci apparently provide rapid access of HAI-1 for inhibition of matriptase, immediately following its activation. Indeed all activated matriptase was detected in complexes with HAI-1, only 5 min after suramin stimulation. The close temporal and spatial coupling of matriptase activation with its inhibition suggests that the proteolytic activity of this enzyme must be well controlled, and that the proteolysis of matriptase substrates may be tightly regulated by this mechanism.

American Association for Cancer Research, 96th Annual Meeting, April 16 to 20, 2005, Anaheim, CA. 

Environmental exposure to traffic polycyclic aromatic hydrocarbons (PAHs) and risk of breast cancer.

Jing Nie, Jan Beyea, Matthew Bonner, Daikwon Han, John Vena, Peter Rogerson, Dominica Vito, Paola Muti, Maurizio Trevisan, Jo Freudenheim. State Univ. of New York at Buffalo, Buffalo, NY, Consulting in the Public Interest, Lambertville, NJ, Occupational and Environmental Epidemiology Branch, National Cancer Institute, Bethesda, MD, Department of Epidemiology and Biostatistics, Arnold School of Public Health, University of South Carolina, Columbia, SC.

Polycyclic aromatic hydrocarbons (PAHs) are an important component of air pollution and potential human carcinogens. While they have been shown to cause mammary cancer in animal studies, the association between PAH exposure and breast cancer risk is not well understood. Traffic emissions are one of the major sources of PAH exposure in cities. Further, growing evidence suggests that there may be critical time periods of exposure in breast cancer initiation and development. In this study, we examined the association between breast cancer risk and exposure to PAHs from traffic emissions estimated for each woman at menarche, at the time when she had her first pregnancy and birth, and at 20 and 10 years prior to interview, using data collected from the Western New York Exposures and Breast Cancer (WEB) study, a population based case control study in western New York. All participants were women, aged 35-79, residents of Erie and Niagara Counties. Cases had incident, primary, histologically-confirmed breast cancer. Controls were randomly selected and frequency-matched to cases on age, race and county. In-person interviews were used to collect data on potential breast cancer risk factors including self-reported lifetime residential history. Traffic volumes on roads were obtained from historical records for the years from 1960-2002. Tailpipe emission data were based on previous reports, including measurements carried out in tunnels or on individual vehicles run in place on test beds. A geographic model, developed by Dr. Beyea and colleagues from the Long Island Breast Cancer project, was used to reconstruct historical traffic PAHs, using BaP as a surrogate for total PAH exposure. Cruise emissions, cold engine emissions and intersection emissions were used to estimate total traffic PAH emissions. Meteorological information was also utilized in the geographic dispersion model to assign PAH exposure at each residence. The model was validated using data collected from both Long Island and our study area. We found evidence that higher exposure to traffic PAH emissions at menarche was associated with increased risk of premenopausal breast cancer (OR 2.07, 95% CI 0.91-4.72, p for trend 0.03) and emissions at the time of a woman's first birth was associated with postmenopausal breast cancer (OR 2.58, 95% CI 1.15-5.83, p for trend 0.19). Both associations were limited to lifetime non-smokers. There was no association of traffic emissions with risk for any of the other time periods. These findings provide evidence for both the potential importance of early exposures and the potential importance of an environmental agent in risk of breast cancer.

American Association for Cancer Research, 96th Annual Meeting, April 16 to 20, 2005, Anaheim, CA. 

Methylenetetrahydrofolate reductase (MTHFR) and risk of breast cancer: The Western New York Exposures and Breast Cancer Study (WEB Study).

Mary E. Platek, Jo L. Freudenheim, Sylvia Quick, Jing Nie, Paola Muti, Susan E. McCann, Maurizio Trevisan, Peter Shields, Stephen B. Edge. University at Buffalo, Buffalo, NY, Roswell Park Cancer Institute, Buffalo, NY, Georgetown University, Washington, DC

Because there is evidence that the association of alcohol with breast cancer risk is limited to women with low folate status there is reason to believe that one-carbon metabolism is important to the carcinogenesis of breast cancer. MTHFR is a key enzyme involved in one-carbon metabolism and is responsible for the irreversible reduction of the 5,10 methylenetetrahydrofolate to 5-methyltetrahydrofolate which provides methyl groups for the vitamin B12 dependant remethylation of homocysteine to methionine required for the production of S-Adenosylmethionine (SAM), the universal methyl group donor. MTHFR is encoded by a gene that is highly polymorphic. Two variants have been shown to have functional impact, C677T and A1298C variants. We conducted a population based case control study in two counties of western New York. Cases were women with primary incident pathologically-confirmed breast cancer and controls were frequency matched to cases on age, race and county. Genotyping was done using PCR-RFLP and diet was assessed with a self-administered 104 item FFQ. Odds ratios (OR) and 95% confidence intervals (CI) were estimated by unconditional logistic regression adjusted for age, education, total energy and other nutrient and breast cancer risk factors. Associations within pre- and postmenopausal groups for high and low intakes of folate, vitamins B6 and B12, riboflavin, methionine and alcohol were examined for each variant. For the C677T variant, the risk for breast cancer was increased among postmenopausal women with the TT variant who had low folate intake (OR 1.47, CI 0.896-2.412). Increased risk was also noted among postmenopausal women with the CT variant who had low riboflavin and low vitamin B12 intakes. No significant associations were observed for premenopausal women for the C677T variant. Analysis for the A1298C variant of MTHFR showed decreased risk among postmenopausal women who had the CC variant and low methionine intake (OR 0.562, CI 0.318-0.994), but not for any other analysis. These findings suggest that MTHFR polymorphisms may impact postmenopausal breast cancer risk among women with lower one-carbon status.

Era of Hope 2005 Department of Defense Breast Cancer Research Program Meeting, Pennsylvania Convention Center, June 8 to 11, 2005, Philadelphia, PA.

Molecular epidemiology and mechanisms for breast carcinogenesis: alcohol drinking as a paradigm. A breast cancer challenge.

Peter G. Shields, M.D. Lombardi Comprehensive Cancer Center, Georgetown University, Washington DC. pgs2@georgetown.edu.

Alcohol drinking is an established risk factor for breast cancer, but biological mechanisms for this remain unexplained. In our funded Center, we will study 4 etiological mechanisms, namely: 1) perturbation of estrogen metabolism and response; 2) increased oxidative stress and damage to DNA and proteins; 3) mutagenesis by acetaldehyde and; 4) interactions with folate and one-carbon metabolism, methylation and mutations. Effect modification by genetic susceptibility and diet will be considered. The key investigators of this Center include Drs. Jo Freudenheim (SUNY: Buffalo), Celia Byrne (LCCC), Leena Hilakivi-Clarke (LCCC), BJ Song (NIAAA), and Matthew Freedman (LCCC). The Aims are: 1) to use a previously conducted breast cancer case-control study of 1131 cases and 2021 controls that will allow us to directly examine breast cancer risk; 2) to conduct a study of mammographic breast density in 1064 Caucasian and African American women considering breast density as an intermediate biomarker of breast cancer risk; 3) to conduct an intensive biomarker study of 250 Caucasian and African American cancer-free women undergoing reduction mammoplasty that will provide insights into early carcinogenic changes and; 4) to utilize several different experimental animal models that will provide mechanistic data that can be corroborated across studies. Across all the studies will be shared biomarkers relating to the above mechanisms, namely, gene hypermethylation, mitochondrial mutations, microarrays, p53 mutations, comparative genomic hybridization, protein analysis for adducts and metabolizing genes, markers of oxidative damage, estrogen receptor analyses, markers of proliferation, digital mammography, and MRI of mammary glands. In the human studies, we have identified increased risks with low folate, B6 and B12 in the diet with genetic polymorphisms and effects on hypermethylation frequency in breast cancers. We also have determined that 15-20% of healthy women already have evidence of hypermethylation in their breast tissue. Animal studies indicate the reversibility by folate on effects of ethanol-promoted and DMBA-induced mammary tumors. The attributable risk for alcohol drinking is about 17,000 breast cancers per year. Using alcohol as a paradigm for breast carcinogenesis, we can learn about other preventable causes of breast cancer as well (estrogen, diet, DNA repair, oxidation, etc.). Definitive studies on breast carcinogenesis would lead to improved public health recommendations, allow for women to make individual choices about lifestyle and risk, place alcohol drinking into a broader context of interactions with other choices such diet, hormone replacement therapy, etc., and lead to more rationale prevention strategies.

28th Annual Scientific Meeting of the Research Society on Alcoholism, June 25 to 29, 2005, Santa Barbara, CA.

Ethanol-mediated rapid degradation of Keap1 protein and activation of NRF2 transcription factor in human hepatoma and porcine kidney cell lines.

Aragon RA, Kim BJ, Brooks PJ, Shields PG, Song BJ. LMBB, National Institute on Alcohol Abuse and Alcoholism, Bethesda, MD 20892.

Acute or chronic ethanol exposure causes cellular damage in most cells. Such exposure, however, also results in the induction of certain proteins such as gamma-glutamyl cysteinyl synthase (GCS), which is a key enzyme in glutathione synthesis and is thus thought to have a protective or defensive role against oxidative stress or acute cellular damage. Recent experiments have established that acute ethanol exposure results in an increase in intracellular GCS levels. The latter protein was shown to be transcribed or otherwise regulated by the transcription factor, NRF2, which is retained in the cytosol by an actin-binding cytoskeletal protein (Keap1) but which translocates to the nucleus after exposure to alcohol or other toxic compounds such as acetaminophen. The exact mechanisms by which ethanol-mediated NRF2 activation and nuclear translocation occur, however, remain unclear. Therefore, we have investigated how NRF2 is activated after ethanol exposure in human HepG2 hepatoma cells transduced with ethanol-induced CYP2E1 (E47 cells) and in CYP2E1-containing porcine kidney LLC-PK1 cells. Ethanol caused a time-dependent degradation of Keap1 protein in both cell lines. We observed the translocation of NRF2 from the cytosol to the nucleus concurrent with Keap1 degradation. We are therefore investigating the potential mechanisms of Keap1 degradation as well as the up-regulation of NRF2 down-stream genes such as GCS and quinone reductase after alcohol exposure in two different cell lines and alcohol-exposed mouse liver.

DoD Breast Cancer Center of Excellence Resources: Posters

Annual Lombardi Comprehensive Cancer Center Research Fair, April 15, 2005, Washington, DC.

Biomarkers for susceptibility to primary lung cancer in women with breast cancer.

Tennis M¹, Krishnan S¹, Granath F², Hall P², Shields PG¹. ¹LCCC, GUH, Washington, DC. ²Karolinska Institutet, Stockholm, Sweden.

Click here to download pdf of poster.