|
Effects of low levels of exposure: why do we care? Much of the world’s current safe drinking water standards for arsenic are based on risk estimates using data on people exposed to very high levels of arsenic through their occupations or through drinking water in areas such as Bangladesh, Taiwan and parts of South America. In recent years, exposure to drinking water arsenic has also been identified as a potential health concern in regions of the United States where bedrock contains unusually high levels of arsenic, such as areas of New Hampshire, Maine, Michigan and regions in the Southwest and Rockies. In these regions, elevated levels of arsenic has been detected from private wells which are not routinely tested for drinking water contaminants. These levels —typically below 200 parts per billion — are much lower than those of Bangladesh but even low levels may pose a health risk according to a 2001 report issued by the National Academies' National Research Council. For this reason, the US Environmental Protection Agency lowered the maximum containment level for arsenic from 50 to 10 parts per billion, or 0.010 milligrams per liter. All public water systems must comply with the new standard beginning January 23, 2006. A study led by Dartmouth epidemiologist Margaret Karagas is designed to determine the health effects that these lower doses of arsenic, consumed over long periods of time, pose to people in the United States. The project focuses on New Hampshire, where a large proportion of the population (about 40%) gets its drinking water from private drinking water supplies, and one in ten or more of these contains arsenic in excess of the revised U.S. standard. The research team, including epidemiologists, biostatisticians, pathologists, chemists and geologists, has evaluated the sources of arsenic in the state and is now exploring the association between cancer and levels of arsenic exposure in a random sample of residents. At high levels of exposure, arsenic has been specifically linked to skin and bladder cancer. Throughout New England, bladder cancer mortality rates are higher than other parts of the country for reasons that are not understood. Karagas and colleagues hope to provide important information for evaluating whether arsenic poses a risk of cancer at the low-to-moderate exposures found commonly in the United States. The researchers are also studying links between arsenic and skin cancer. Though cancer registries do not usually record skin cancers, the Dartmouth group enlisted the collaboration of dermatologists and pathology laboratories throughout New Hampshire and bordering regions, forming one of the few U.S. skin cancer registries. Though skin cancers are strongly tied to arsenic exposure in Taiwan and similar regions, there’s no direct evidence of risk associated with the low to moderate levels of arsenic exposure commonly experienced in the U.S. People may vary in their sensitivity to arsenic depending on their genetic background, diet, health care and other environmental factors, making comparisons with the U.S. population uncertain. The Dartmouth group is working to tease out which individuals may be at greater risk of cancer from low-dose exposures to arsenic. To date, the researchers have studied more than 6,000 New Hampshire residents. Establishing a dose-response relationship for a U.S. population that is, the point at which an exposure is clearly related to the incidence of disease, would be the ideal scientific evidence to guide public policy. This study is among the first to provide information regarding individual exposure to arsenic using biologic measurements and risk of cancer for a geographically defined U.S. population. In addition, Karagas and her team are currently investigating whether certain subgroups of the U.S. population are more vulnerable to arsenic-induced disease, either because of nutrition, genetic factors or because they are exposed to additional cancer-causing agents.
Though laboratory studies have suggested that even trace amounts of arsenic can harm human cells, reliable techniques to measure arsenic in the environment at very low levels has only recently become available. The Dartmouth epidemiology group collaborated with the Dartmouth Trace Analysis Laboratory to develop ways to accurately detect as little as a few parts per trillion of arsenic in water or biological tissue. These amounts are incredibly small; an analogy for a concentration of one part per trillion is one grain of salt in an Olympic sized swimming pool. To trace the source of the arsenic in New Hampshire's drinking water, Karagas and colleagues tested well water samples from thousands of New Hampshire households. The group also drew upon well data compiled by New Hampshire's Department of Environmental Services, which maintains a comprehensive database of results from well water testing. Of the wells tested in the study, over one quarter found to contain more than two parts per billion, (0.002 milligrams per liter) of arsenic; about one in ten measured above 10 ppb (0. 010 mg/l); and one in 40 were over 50 ppb (0.050 mg/l). The Dartmouth researchers estimate that 35 to 40 percent of the households in New Hampshire use private wells as their water source. Federal regulations addressing the allowable levels of arsenic in drinking water do not apply to domestic wells, defined as those serving fewer than 15 households or 25 individuals. Most of these wells have not been tested for arsenic.
An important aspect of the Dartmouth study is the use of human tissue samples to assess concentrations of arsenic as a biomarker — direct biological evidence of an individual's exposure to arsenic. Most of what is known about the link between drinking-water arsenic and disease is based on ecological studies that look for an increased incidence of disease in geographical regions where the average concentration of arsenic in drinking water is known to be high. These studies, including the small number that have examined U.S. populations, do not account for the wide variation in arsenic levels from well to well, nor do they establish that the individuals who became ill where the ones who drank water from tainted wells. Also, exposure to arsenic can occur both at home and outside the home. Calculations of exposure based solely on drinking water supplies especially at low levels could be subject to error. A biomarker of exposure helps to measure how much actually arsenic gets into people's bodies. Arsenic binds tightly to the sulfur in certain proteins that make up hair and nails. For that reason, measurements of arsenic in hair and nails provide biological evidence that a person has been exposed to arsenic. In an earlier study, Karagas and colleagues found that toenail arsenic concentrations were correlated with drinking water concentrations, particularly among those with water containing one part per billion or more of arsenic and that toenail tissue is not subject to external contamination. The Dartmouth group, in collaboration with investigators at the University of Arizona, is also assessing arsenic concentrations in urine to individuals’ exposure. Though measurements of arsenic in urine have generally been used to determine acute exposure, the utility of urine measurements for assessing low to moderate arsenic exposures is unknown. One reason is that the arsenic in seafood, a less toxic, inorganic form of arsenic, can show up in urine and must be accounted for when assessing the more toxic drinking-water form of arsenic. Finding specific chemical forms of arsenic in urine is thought to reflect variations in the way individuals process arsenic in their bodies, and is also believed to be a possible marker for increased susceptibility to arsenic's toxic effects.
Tobacco smoking, certain occupational exposures, and exposure to arsenic in drinking water have been associated with the occurrence of bladder cancer. However, the mechanisms by which these toxins contribute to disease is not well understood. Animal and in-vitro studies suggest that arsenic and other exposures may act through epigenetic modifications; that is, changes in DNA and its associated proteins that alter gene expression without altering the DNA sequence itself. Epigenetic modifications have been linked to a number of diseases that that result in inappropriate gene silencing. Most cancers, for example, for example, involve the epigenetic silencing of genes that control normal cell function. One way epigenetic modifications can occur in DNA is through the addition of a chemical group (methyl) to DNA. Methylation can switch off gene expression by interfering with the process through which the gene encodes protein. Dr. Alan Schned, an expert urologic pathologist on the project, has systematically re-reviewed the tumor tissue samples of bladder cancers in New England. Dr. Karl Kelsey and colleagues at the Harvard Superfund Program are applying a combination of techniques to identify genes that are silenced in this way in the bladder tumors collected by the Dartmouth study. In a published study, the investigators examined the relationship between the silencing of three tumor suppressor genes — p16, RASSF1A, and PRSS3 — and exposure to both tobacco and arsenic. Arsenic exposure, measured in toenail samples collected from patients, was associated with methylation of two of the three tumor suppressor genes (RASSF1A and PRSS3), and cigarette smoking was associated with an increased risk of methylation in the p16 gene only. These results, from human bladder tumors, add to the body of animal and in-vitro evidence that suggests epigenetic alterations play a role in bladder carcinogenesis. Another goal of the studies is to establish how much arsenic is needed to get an increased risk of cancer. Most of the population studies linking arsenic to bladder cancer have been conducted in Bangladesh, southwest Taiwan and other areas of the world where well water contains levels of arsenic as high as 1200 ppb (1.20 mg/l), more than ten times higher than elevated levels typically found in New Hampshire. It’s not clear whether arsenic levels commonly found in U.S. can cause bladder cancer. Finding links between a disease such as cancer and exposure to very low doses of an environmental toxin is complicated by other factors as well. For example, individuals in the same community and even in the same household may have very different exposures to arsenic-contaminated well water; some individuals might be exposed to arsenic from other sources than the water at home, such as arsenic in the workplace. Using questionnaires to gain information on individual exposure to arsenic over their lifetime, combined with biologic tissue levels of arsenic, Dr. Tor Tosteson and his colleagues are working to improve arsenic exposure risk models.
Dartmouth researchers also are trying to understand other ways in which arsenic increases the risk of certain kinds of cancer. Unlike many other known chemical carcinogens, arsenic at low levels does not appear to directly damage DNA or cause mutations in genes. Instead, it seems to indirectly modify the way cells behave in ways that increase their probability of becoming cancer cells. One way this may occur is through inhibition of DNA repair mechanisms, leading indirectly to increased mutations from other DNA damaging agents, such as cigarette smoke or UV irradiation from sunlight. By inhibiting an individual’s DNA repair capacity, arsenic may serve as a potent co-carcinogen. Dr. Angeline Andrew and her colleagues at Dartmouth tested this
hypothesis by analyzing the expression of repair genes and arsenic
exposure among individuals enrolled in the epidemiologic study
investigating arsenic exposure and cancer risk in New Hampshire.
Arsenic levels were correlated with the expression of several
critical DNA repair genes, the findings showed a particularly
strong association between biomarkers of arsenic exposure and
expression of DNA repair genes. The researchers are now pursuing
studies with larger groups using assays that test DNA repair
function to confirm the hypothesis that inhibition of DNA repair
capacity is a potential mechanism for the co-carcinogenic activity
of arsenic. top |
