Pilot Projects 2008
Pilot Grant Awardee: Ettinger, AS
Project Title: Maternal Arsenic Exposure as a Risk Factor for Gestational Diabetes
Award Amount: $7,500
Description: The overall aim of this project is to determine if maternal arsenic exposure is associated with impaired glucose tolerance during pregnancy and if impaired maternal glucose metabolism is associated with adverse outcomes in offspring.
Funding from this pilot project was used to move three separate pilot projects forward:
1) Added glucose tolerance testing at 28 weeks gestation to ongoing birth cohort study of arsenic and reproductive outcomes in Bangladesh (P.I. Christiani). Data has been collected; waiting for data extraction.
2) Completed analysis of existing data from Children’s Center birth cohort study (P.I. Robert Wright) at the Tar Creek Superfund Site and published the first epidemiologic study to investigate arsenic exposure and impaired glucose tolerance during pregnancy. We studied 532 women living proximate to the Tar Creek Superfund Site to investigate whether arsenic exposure is associated with impaired glucose tolerance during pregnancy. This is the first published study to investigate arsenic exposure and risk of gestational diabetes (GD). After adjusting for age, Native-American race, prepregnancy body mass index, Medicaid use, and marital status, we found that women in the highest quartile of blood arsenic exposure had 2.8 higher odds of impaired GTT than women in the lowest quartile of exposure (95% confidence interval, 1.1–6.9) (p-trend = 0.008). Among this population of pregnant women, arsenic exposure was associated with increased risk of impaired GTT at 24–28 weeks gestation and therefore may be associated with increased risk of GD. These results were used as preliminary data for a new study designed to investigate health disparities in diabetes by enrolling pregnant women specifically to address this objective with more detailed and specific measures of exposure and outcomes.
3) Planned study to enroll pregnant women at a diabetes center in Pune, India.
1) Ettinger AS, Zota AR, Amarasiriwardena CJ, Hopkins MR, Schwartz J, Hu H, Wright RO. Maternal arsenic exposure and impaired glucose tolerance during pregnancy. Environ Health Perspect 2009 Jul; 117(7):1059-1064.
2) Ettinger AS. Maternal Arsenic Exposure in Relation to Maternal, Fetal and Child Adiposity and Diabetes
Risk Factors. Epidemiology 2009 Nov; 20(6):S234-S235 [abstract].
3) Ettinger AS, Hopkins MR, Amarasiriwardena CJ, Schwartz J, Hu H, Bellinger D, Wright RO. Hemochromatosis Gene Variants Modify Effect of Arsenic on Impaired Glucose Tolerance during Pregnancy. Epidemiology 2009 Nov; 20(6): S96 [abstract]
4) Ettinger AS, Baccarelli A, Tarantini L, Albetti B, Wright RO. Epigenetic Adaptation to Impaired Glucose Metabolism during Pregnancy. Accepted for presentation at: Harvard Program on Quantitative Genetics, November 2010.
Description: The goal of this pilot project is to develop preliminary data to demonstrate the feasibility of studying environmental PCB levels and PCB serum levels among occupants of PCB-contaminated buildings. An initial application to investigate the relationships between PCB exposure from indoor environments and biomarkers of thyroid function (1 R01 ES014077-01) was reviewed but not funded in 2005. The specific aims of that proposal were to identify and recruit occupants of buildings containing PCB-contaminated caulking, along with a comparison group without PCBs in their buildings. Comprehensive exposure assessments would be done, including serum collection for PCB analysis from the study subjects, and the relationships between serum PCBs with serum measures of thyroid status would explored. The reviewers’ enthusiasm for the proposed study was limited by the lack of pilot data available to demonstrate that certain parts of the study were feasible. The reviewers stated that the application would be greatly strengthened by the inclusion of pilot data demonstrating an increased potential exposure to PCBs in buildings with PCB-rich caulking materials, and preliminary data demonstrating the ability to recruit subjects.
The data we have collected in this pilot project will address the concern about the lack of pilot data on the potential for PCB exposure to occupants of PCB-containing buildings. On the concern about the lack of preliminary data demonstrating the ability to recruit subjects, the Massachusetts Teachers Association (MTA) collaborated in the identification of sites for study, individuals who may qualify for participation as subjects, and in the dissemination of the study results. The acquisition of preliminary data on serum PCB in a group of workers (MTA members) addresses the concern about our ability to recruit subjects and will greatly strengthen a revised proposal.
Description: The goals of this pilot study were 1) to investigate the association between maternal arsenic exposure with DNA methylation patterns in peripheral blood leukocytes collected in maternal blood at 28-weeks gestational age and umbilical cord blood; 2) To determine if global DNA methylation is associated with reproductive health outcomes in a prospective birth cohort of 160 maternal-infant pairs; and 3) To determine if genetic polymorphisms in the methionine cycle which regenerates SAM [MTHFR C677T, MTHFR A1298C, A66G MTRR, RFC-1] modifies the association of arsenic exposure with global DNA methylation in adults and infants.
Progress Report: We have completed the epigenetic analysis and have quantified the DNA methylation at LINE-1 (3 positions), Alu (3 positions), p16 (7 positions), and p53 (4 positions) in maternal peripheral blood and umbilical cord blood samples. Genotyping of the maternal and umbilical cord DNA is ongoing. The completed epigenetic data has been merged with the prenatal arsenic exposure data, birth outcome data, and additional collected covariates. Data analysis has been ongoing and we have observed the following in relationship to the aims stated above:
Using maternal drinking water arsenic exposure measured at the time the mother’s were recruited into the study, we categorized infants as either exposed (ie., mother’s had detectable levels of arsenic in their drinking water (1 ug/L or above) or unexposed (had less than 1 ug/L). This resulted in 52 unexposed infants (46.8%) and 59 exposed infants (53.2%) that had complete epigenetic data. Generalized linear models were used to evaluate the association between exposure status and DNA methylation. In models that were adjusted for maternal BMI, infant sex, delivery method (vaginal/cesarean), gestational age (preterm/full term), and maternal age we observed that infants that were exposed to arsenic in utero had significantly higher %methylation at LINE-1, Alu, and p16 (positions 2,4,5,6 and 7). There was no observed association between prenatal arsenic exposure and DNA methylation in p53. When looking at the same associations in maternal blood, those mother’s who were exposed to arsenic when they enrolled into the study had significantly higher %methylation at LINE-1 and p16 (positions 3 and 6). Therefore, it appears that an infant’s epigenome was modified by maternal exposure to arsenic early in their pregnancy.
We also evaluated the correlation of DNA methylation in maternal-infant pairs since the inheritance of DNA methylation patterns is a popular theory to explain the influence of parental genetic and environmental factors on the phenotype of their offspring and few studies have examined this relationship in humans. Using 120 paired maternal-umbilical cord blood samples randomly selected from a prospective birth cohort in Bangladesh, we quantified DNA methylation by pyrosequencing of seven CpG positions in the promoter region of p16, four CpG positions in the promoter region of p53, and LINE-1 and Alu. Positive correlations were observed between maternal and umbilical cord blood at p16, LINE-1, and Alu but not p53. Multiple linear regression models observed a significant association between maternal and umbilical cord blood at LINE-1 and Alu (LINE-1: β=0.63, p<0.0001; Alu: β=0.28, p=0.009). After adjusting for multiple comparisons, maternal methylation of p16 at position 4 significantly predicted methylation at the same position in umbilical cord blood (β=0.43, p=<0.0001). These models explained 48%, 5% and 16% of the observed variability in umbilical cord %5mC for LINE-1, Alu and p16 at position 4, respectively. These results suggest that DNA methylation in maternal blood was correlated with her offspring at LINE-1, Alu, and p16 but not p53. Additional studies are needed to confirm whether these observed associations were due to the inheritance of epigenetic events or the shared environment between mother and fetus. Future studies should also use a multi-generational family-based design that would quantify both maternal and paternal contributions to DNA methylation in offspring across more than one generation. These results were successfully published in PLOS One.
Aim 2: We have also begun preliminary analysis on the association between DNA methylation in umbilical cord blood, gestational age, and birthweight. Using linear regression models that were adjusted for gestational age, we have observed that DNA methylation in p16 is associated with a modest decrease in birthweight (β= -65.1 grams, p-value 0.08). In a separate model, there was a significant association between p16 %mC and a slight increase in gestational age (β= 0.52 weeks, p-value 0.004). These findings are novel and currently we are exploring different modeling techniques in order to try to isolate individual associations given the complex relationship between gestational age and birthweigth.
Resulting Grants: “Epigenetic Effects of Prenatal Arsenic Exposure and Fetal Growth” NIEHS Grant # 1 K01 ES017800-01, Start date: 01/19/2010 to 01/19/2015.
Harvard School of Public Health Dean’s Gene-Environment Pilot Grant ($195,000 over two year) to continue this research with additional samples and epigenetic markers.
Kile ML, Baccarelli A, Hoffman E, Wright RO, Christiani DC. (2008) Arsenic exposure and global DNA methylation. Epidemiology 19(6): S304-S305.
Kile ML, Baccarelli A, Tarantini L, Hoffman E, WrightRO, Christiani DC. (2010) DNA Methylation in maternal blood predicts global and gene-specific DNA methylation in offspring. PloS ONE, 5(10): e13730. doi:10.1371/journal.pone.0013730.
Description: Mutation of the HFE gene is the leading cause of hereditary hemochromatosis (HH) and is considered to be the most common genetic defect in the North American Caucasian population. Defects in the HFE gene promote increased intestinal absorption and progressive tissue deposition of iron. HH has also been associated with derangements in the metabolism of other divalent metals. We hypothesize that carriers of C282Y and/or H63D HFE alleles may also be more susceptible to manganese exposure through the olfactory pathway, and consequently may have impaired olfactory function as an early indicator of brain manganese neurotoxicity. Our model further suggests that Hfe-/- knockout mice would have increased 54Mn absorption through the olfactory system and that manganese exposures would impair olfaction in these animals. This pilot project will test the hypothesis that HFE acts as a genetic determinant of olfactory manganese absorption and toxicity. These planned studies will foster future research avenues for a new junior faculty member, broaden on-going collaborative efforts between laboratory scientists and epidemiologists, apply information gained from animal studies to human studies, focus on a central theme of genetic variability and gene-environment interactions, and bring new technologies to HSPH for assessing individual biological responses to exposure – olfactory testing and MR imaging – for both mice and humans.
Description: The association between short-term increases in ambient fine particulate matter (PM2.5) and cardiovascular morbidity and mortality is well documented. Specifically, acute elevations in particulate air pollution have been linked to the triggering of myocardial infarction, life-threatening cardiac arrhythmias, ischemic stroke, and decompensation of heart failure patients It is widely believed that subjects with pre-existing cardiovascular disease represent a group that may be more susceptible to the acute effects of ambient PM. This notion of individual susceptibility is supported by the observation that many chamber studies in healthy patients or animals have found only small or no effects of concentrated ambient particles (CAPs), while studies on diseased patients or animal models of cardiovascular disease often find stronger CAPs-related effects. With ischemia, even a brief episode, appears to change the response to air pollution indicative of a heightened response. This change in response appears to continue long after the ischemic event.
Our hypothesis is that even a brief ischemic event can produce epigenetic changes in the heart or vasculature which then influences the autonomic nervous system’s response to any stress, and specifically particulate exposure. The effect of myocardial ischemia with reperfusion on gene expression represents an area of intense research interest. However, whether myocardial ischemia/reperfusion results in epigenetic changes has not been explored. Ultimately, we propose to study DNA methylation patterns in lung tissue, brain tissue, cardiac tissues, isolated cardiac vessels, and autonomic ganglia as a means to test this hypothesis. The goal of this pilot project is to develop a model of ischemia in which there is a change in response to particle exposure comparable to the changes that we have observed in dogs and can be inferred from human studies. The proven model will then be used in subsequent studies to test our gene-silencing hypothesis and develop pilot data for a new R01 application in this area. The specific aims of the pilot project proposal are:
Specific Aim 1: To use a rat model of myocardial ischemia/reperfusion to assess whether a brief myocardial ischemic insult alters the cardiovascular response to CAPs, as assessed by heart rate and heart rate variability. Specific Aim 2: To generate pilot data to test the hypothesis that a brief period of myocardial ischemia results in epigenetic modifications in the heart.
Progress Report: Sprague-Dawley rats were purchased with implanted radio-telemeters already in place for the recording of ECG signals. Rats were exposed for 5 hours per day to either CAPs or filtered air using the Harvard Ambient Particle Concentrator (HAPC). Rats have been exposed to either CAPs or filtered air in a protocol such that each group received 4 CAPs exposures and 4 filtered air exposures prior to surgical intervention. Then, rats were randomly assigned to receive either a 5-min complete occlusion of the left coronary artery followed by reperfusion or a sham operation. The rats were again exposed to CAPs and filtered air in a protocol such that each received at least an additional 4 CAPs exposures and 4 filtered air exposures. In this protocol we are directly testing the hypothesis that brief myocardial ischemia can alter the cardiovascular response to CAPs. These post operative exposures were spread out over several months to determine the persistence of the effect. On completion of exposures, the animals were sacrificed and tissues collected for DNA analyses. Whole blood from which the buffy coat was separated was collected from each rat as well as 1-2 mm3 pieces of heart ventricle, sino-atrial nodal area tissue, superficial cardiac plexis area tissue, coronary arteries, cervical sympathetic ganglion, lung airway tissue, lung parenchymal tissue, and tissue from the area of the dorsal motor nucleus of the vagus and the medulla. All of these tissue pieces and the buffy coat cells were snap frozen and stored in liquid nitrogen for subsequent DNA extraction and assessment of methylation. Analyses of electrocardiogram wave forms are in progress. We are assessing heart rate, heart rate variability, ST- segment elevation/depression, and wave form intervals. These parameters are assessed before exposure during the initial baseline exposure periods, during the operation to produce ischemia, and during each of the post-ischemia periods. Respiratory breathing patterns are also being analyzed. Frozen samples for DNA methylation studies will be done if changes in the physiologic parameters under study have significant findings.
Pilot Grant Awardee: Weuve J
Project Title: Environmental determinants of cognitive impairment in Parkinson’s disease: a study to establish the feasibility and validity of telephone cognitive testing
Award Amount: $ 8,200
Description: Parkinson’s disease (PD) is characterized primarily by neuromotor dysfunction, but it is often accompanied by cognitive deficits as well. Although little is known about the causes of these deficits, exposures to environmental toxins such as lead are plausible candidates. This pilot project aimed to develop and validate a telephone-based cognitive assessment battery for persons with PD, with the goal of generating preliminary data for a grant to examine in more detail environmental determinants—and gene-environment determinants—of cognitive deficits in PD. Based on interviews with 100 PD cases and now 49 controls and 4 additional participants for a validation study, we have further established the feasibility of the telephone cognitive battery, generated preliminary data suggested an inverse association between higher concentrations of bone lead and worse performance on the tests, and generated preliminary data for formal validation testing of the telephone battery.
The risk of dementia among Parkinson’s disease (PD) patients has been estimated to be up to 6 times greater than in healthy controls  and is accompanied by increased risks for institutionalization, mortality, and caregiver distress [2-4]. Moreover, most PD patients without clinically evident intellectual deterioration exhibit relatively subtle deficits in specific areas of cognitive function, and these deficits, too, are associated with increased risk for disability and hospitalization. Nonetheless, research on risk factors for dementia or cognitive impairment in PD has been relatively limited [5, 6], and the possible involvement of environmental toxicants has not been addressed at all.
Exposure to environmental toxins such as lead may affect cognitive function in PD. As part of an NIH-funded grant, we found that higher levels of long-term lead exposure are associated with increased risk of PD, based on data from our case-control study of 340 cases recruited from 4 hospitals in the Boston area and 274 controls of a similar age and sex distribution (Weisskopf et al, in press). As part of this ongoing study we have lead exposure measures on all participants as well as extensive data related to pesticide exposures and other genetic and non-genetic characteristics. However, conducting follow-up in-person cognitive assessments of these participants may be challenging due to their mobility limitations, their wide dispersion across New England and potential reticence to travel into urban Boston.
Thus, we proposed to conduct a pilot study to develop and validate a telephone-based cognitive assessment. Preliminary data from this pilot study will be used to prepare an R01 proposal to complete cognitive testing on all study participants—which would open the possibility of looking at gene-environment interactions related to cognitive impairment in PD; more completely examine the role of other factors on which we have data; and extend the cognitive assessment longitudinally so that we can look at the progression of cognitive dysfunction in PD over time.
Telephone cognitive assessment. Our group developed a modified version of the telephone cognitive test battery developed and validated by Dr. Francine Grodstein, the PI of several large studies of cognitive aging. Based on further consultation with Dr. Daniel Press, consulting neuropsychologist specializing in cognitive impairments in PD, we revised the battery that was in the original proposal (details in the Progress Section).
Validation of the telephone cognitive assessment. Most elements of the telephone cognitive battery have been extensively validated against in-person testing among relatively healthy older adults. However, the validity of key novel tests in our battery and the validity of the telephone battery among PD patients have received less attention. Therefore, we are evaluating the validity of the telephone cognitive battery by administering the telephone battery and a similar in-person battery to up to 40 PD patients, including those with and without frank dementia. We will evaluate 20 persons without dementia and 20 PD patients with established dementia from the neuropsychology clinic of Dr. Press.
The specific aims we proposed to address are:
(1) Complete telephone cognitive testing of 100 PD patients, 50 selected from the lower quartile of lead exposure and 50 selected from the upper quartile.
(2) Validate the use of telephone assessments of cognitive function among 40 PD patients—at least 20 of whom will have been diagnosed as having PD with dementia—against in-person tests of cognitive function.
Pilot Grant Awardee: Fang, S
Project Title: A Toxicogenomic Approach to the Discovery of Early Response Biomarkers Associated with Exposure to Particulate Matter and Metals among Welders
Award Amount: $ 21,375
Description: We aim to discover specific pathway candidate biomarkers of early stage responses to metal-rich particulate exposures based on global gene expression profiles among welders. We will determine the association between the candidate biomarkers, measured in serum, and exposure to cumulative work-shift PM2.5 concentrations and specific metal components (manganese, nickel, and chromium).
Hypothesis: Analysis of global gene expression profile data with responses previously determined to be clustered into inflammatory and oxidative stress pathways will lead to the discovery of measurable candidate serum inflammatory and oxidative stress markers that will be associated with occupational PM2.5 exposure and its specific metal components (manganese, nickel, and chromium).
Progress Report: Complementary DNA (cDNA) microarray gene expression analysis techniques were used to analyze whole blood total RNA samples collected in 2003 and 2004 from 32 male boilermaker construction workers (mean age 39 years) before and after a work shift of welding or bookwork at a local union hall. Genes significantly associated (p<0.01) with personal work shift PM2.5 concentrations and “status” (pre-shift or post-shift) were identified from separate linear regression models adjusted for age, smoking status, and welding status. In addition, models adjusted for consistent expression signatures due to sources other than the variables of interest, calculated with the R package sva.
Subsequently, the identified genes were analyzed in GeneGo using an enrichment analysis with a threshold setting of p=0.05. A list of 498 genes from the PM2.5 lists was obtained and a list of 522 genes from the “status” list was obtained. The genes associated with secreted proteins were cross-referenced with available measurable analytes from Milliplex, a provider of multiplex detection assays for analysis with the Luminex System. A total of nine circulating cytokines/chemokines were identified: Il-1b, IL-2, IL-9, CD40L, TNF-a, VEGF, SCF, M-CSF, and lymphotactin.
Quantification of the selected proteins has been completed. A total of 109 serum and plasma samples were analyzed for the circulating proteins using the Luminex multiplex detection system. Analysis of the biomarker data in relation to exposure and covariate data is in progress.
Resulting Grants: Systemic inflammatory and oxidative stress responses to occupational particulate exposures, Funding Source: American Heart Association, NCRP Fall 09 Scientist Development Grant #10SDG2610159, 1/1/2010-12/31/2013, Annual Direct costs: $70,000.
Description: This study is a retrospective cohort study designed to assess the relationship of early life environmental lead exposure and risk factors for Alzheimer disease. This study is designed to test the specific hypotheses that higher blood lead levels at birth and early childhood are associated with 1) lower levels of intellectual function in adulthood and 2) higher serum levels of APP and beta-amyloid. In addition, this study is conducting an exploratory analysis to determine if gene-environment interactions modify the relationship between lead and APP / beta-amyloid levels.
Between August 1979 and April 1981, a cohort of 249 infants was established among babies born at the Brigham and Women’s Hospital in Boston. In January 2009, subjects were mailed an introductory letter explaining a new study regarding early-life lead exposure and health outcomes in adulthood. Names and last known addresses were available for only the 148 subjects who participated in 10-year follow-up. As of July 2010, of the 89 subjects located throughout the United States, 55 subjects enrolled in the study, which included completing a questionnaire and donating a blood sample. Forty-three subjects came to Children’s Hospital Boston (CHB) for additional neuropsychological testing.
Blood lead levels and neuropsychological testing
We examined the relationship between blood lead levels and intellectual function among the 43 subjects who participated in IQ testing (Wechsler Abbreviated Scale of Intelligence, or WASI). Participants were similar to subjects that did not come to CHB for assessment in terms of demographic factors, measures of socioeconomic status and IQ scores in early childhood. In our sample, Total IQ is related to blood lead at 6 months (p=0.033), blood lead at 4 years (p=0.062), blood lead at 10 years (p=0.020), average blood lead (p=0.033), and childhood blood lead (p=0.010). Although all these measures are related to Total IQ, childhood blood lead (defined as the average of 4 and 10 year lead levels) has the strongest relationship with Total IQ as measured by R2.
Blood lead levels and APP/beta-amyloid levels
In December 2009, we submitted 55 collected samples to the laboratory of Dennis Selkoe, M.D., at Brigham and Women’s Hospital for analysis of APP levels and beta-amyloid levels in serum. Dr. Selkoe’s laboratory developed an assay for measuring APP and beta-amyloid isomers. Extensive analyses showed no relationship between blood lead levels in early life and these proteins. There are many potential reasons for these negative results including 1) small sample size, 2) relatively young age of participants (29 years) and 3) no true relationship between lead and APP/beta-amyloid production.
Given the negative results of the lead and APP/beta-amyloid experiments, we decided to explore further the pathways that control APP/beta-amyloid production. Currently we are exploring whether childhood lead is associated with up-regulation of genes involved in APP/beta-amyloid production using RNA microarray analysis. If we find such an association, it would support the hypothesis that lead increases production of APP/beta-amyloid, but that our subjects, at age 29 years, did not yet have enough pathogenic protein to be detected.
Resulting Grants: Neurodevelopmental Effects of Early-Life Arsenic Exposure. NIH/NIEHS. 1 K23 ES 017437- 01 Mazumdar (PI). September 2009 – June 2014. This project is composed of three separate studies designed to epidemiologic associations of prenatal and early childhood arsenic exposure with measures of neurological development in children. These measures include 1) head circumference, 2) hearing impairment, 3) motor dysfunction and 4) formal neurocognitive testing. These studies will be conducted in Bangladesh in a cohort of 600 infants who are participating in a current study of reproductive health effects of arsenic.