Over the past decade, evidence has accumulated that environmental contaminants, often at low levels, are causing infertility, declines in sperm count, miscarriages, birth defects, and cancers. Yet, a majority of health professionals, policy makers, and public health advocates have little knowledge of these findings.
To help remedy this, the University of California, San Francisco, and the multinational advocacy group known as the Collaborative on Health & the Environment (CHE) convened a cross-disciplinary conference at the end of January. UCSF's department of obstetrics, gynecology, and reproductive sciences and CHE brought together basic and clinical researchers, physicians, public health professionals, policy makers, and environmental activists with the goal of bridging the knowledge gaps among these groups.
At the conference, participants discussed public policies that would improve reproductive health and how some research results might be used to influence clinical care. They also came up with new research strategies that would help illuminate the relationship between chemical contaminants and reproductive outcomes.
The presentations showed that, in regard to the impact of exposures to certain environmental chemicals, such as the industrial compounds bisphenol A (BPA) and phthalates, a convergence is emerging between research on animals and on humans. That is, more and more, researchers are finding that the effects seen in animals correlate with the effects seen in humans. Also, the presentations indicated that exposures over a lifetime, or sometimes even before birth, influence the reproductive success of the adult. The research is complicated, however, and in most cases it has not yet led to consistent or definitive conclusions.
"There are signs that some aspects of human male reproductive health are deteriorating in some parts of the world," said Niels E. Skakkebaek, physician at Copenhagen University Hospital. The incidence of testicular cancer, primarily a disease of young men, has increased in all industrialized countries, he said. "In some European nations, there has been a three- to fourfold rise in the past 50 years, but in many African and Asian countries, the incidence remains very low."
In addition, in certain European populations, semen quality has deteriorated. Twenty percent of 18- to 20-year-old men in Denmark and Norway now have sperm counts below 20 million spermatozoa per mL, and 40% have counts below 40 million per mL, Skakkebaek said. According to recent publications, 40 million spermatozoa per mL may be the "threshold below which fecundity declines," he explained. Also, among 18- to 20-year-old men in Denmark, 93% of sperm are abnormal by modern criteria for sperm morphology, Skakkebaek said. At the same time, he noted, the rates of undescended testes (cryptorchidism) and a condition in which the urethra opens at the side rather than the end of the penis (hypospadias) have increased.
"There is evidence that all of these conditions have their basis in a common origin: specific errors during development of fetal testes," Skakkebaek said. More than 95% of all testicular tumors seem to be derived from primordial germ cells that failed to differentiate in the womb, he said. This same process also may be responsible for hypospadias and undescended testes and may cause low rates of spermatogenesis in adulthood, he conjectured.
It is interesting to consider the differences between Denmark and Finland, Skakkebaek said. Finnish men have lower risks of testicular cancer and better semen quality than do Danish men. Newborn Finnish boys also have a lower incidence of undescended testes and hypospadias than Danish babies, he observed. Some differences in lifestyle or environmental exposures must be responsible for these disparities in reproductive health, he said. "It would be unwise to comfort ourselves thinking that the widespread poor semen quality as we see in Denmark would have no effect on the fecundity of our societies."
"It is very difficult to test the hypothesis that exposures to environmental chemicals in the womb cause the constellation of male reproductive effects Skakkebaek described," cautioned Richard M. Sharpe, a researcher at the Human Reproductive Sciences Unit of the University of Edinburgh, in Scotland. However, he outlined research that he used to study these disorders in animals. Because the sequence of developmental events is very similar in humans and other mammals, rodents provide a good research model, he explained. All mammals depend upon a large and intricate array of chemical signaling systems that control biological development.
Sharpe exposed pregnant rats to a ubiquitous environmental chemical found in many polymer formulations and personal care products—n-dibutyl phthalate (DBP)—during the last week of gestation. That exposure resulted in dose-dependent induction of cryptorchidism, hypospadias, and impaired spermatogenesis in the male pups, he said. It also reduced the anogenital distance—the distance between the anus and the penis. That measure is considered a sensitive indicator of male sexual developmental problems in the offspring. DBP exposure did not cause cancer, but it impaired germ cell differentiation in the offspring, which is probably how testicular cancer arises in humans, Sharpe noted. "When the DBP-exposed rat pups reached early puberty, their germ cell development was retarded by 80% compared with controls, and many were infertile," he said at the meeting.
Shanna H. Swan, professor of obstetrics and gynecology at the University of Rochester School of Medicine & Dentistry, also cited data that show a decline in men's reproductive health. "Earlier studies found a 1% per year decrease in sperm count in North America and Western Europe over the past 40 to 50 years," she said. New studies show that testosterone levels declined at similar rates in Boston and Copenhagen over that same period, she said. "The two trends taken together increase the plausibility of a significant decrease in male reproductive function."
One type of chemical exposure that could be responsible for this decline in men's reproductive health involves phthalates, Swan said. Several studies show that prenatal exposure to phthalates leads to a decrease in anogenital distance in male rodents. In a new human study Swan conducted, "higher levels of five phthalate metabolites in a pregnant mother's urine predicted a shorter anogenital distance in her male offspring," she said. Also, by looking at levels of phthalates in babies' urine samples, "we have shown that all babies, male and female, are exposed to these ubiquitous chemicals and that the more baby care products a mother applies, the higher the levels of phthalates in her baby's urine." Phthalates are ingredients in nearly all personal care and cleaning products that contain fragrances and are added to polyvinyl chloride to make it flexible.
"Although it hasn't been shown conclusively that postnatal exposure to phthalates can alter male reproductive capacity, recent data suggest the possibility," Swan said. "Large epidemiological studies that prospectively assess later reproductive performance after pre- and postnatal exposures to phthalates will be necessary to establish clear links."
Laura Fenster, epidemiologist in the Occupational Health Branch of the California Department of Health Services, said that exposure to environmental and workplace chemicals may in some cases lead to preterm delivery, unusually low birth weights, and stillbirths. Preterm delivery, defined as birth at a gestational age of less than 37 weeks, is a particularly acute and increasing problem in the U.S., Fenster said. Twelve percent of babies are delivered preterm, a 30% increase since 1980.
Preterm delivery is the number one cause of death in the first month of life and leads to long-term health problems that cost about $26 billion in 2005, she explained. There are several nonenvironmental risk factors for preterm delivery, including advanced maternal age, race, vaginal infections, and premature rupture of membranes, she observed, "but half of preterm births occur in mothers with no known risk factors."
There is clear evidence that smoking and secondhand tobacco smoke cause preterm delivery, Fenster said. "There is less certain, but good, evidence that exposure to DDT, carbon monoxide, benzene, ethylene oxide, lead, diethylstilbestrol (DES), and particulate air pollution can lead to preterm delivery," she said. And, she added, there is "limited" evidence that carbon disulfide, phenoxyacetic herbicides, and phthalates—especially di(2-ethylhexyl) phthalate (DEHP)—cause preterm birth.
Fourteen studies of possible associations between prenatal pesticide exposure and birth weight have produced conflicting results, Fenster said. Ten indicated that pesticide exposure reduced birth weight, and four showed no effects. "Another study of births in upper Manhattan concluded that residential exposure to the insecticides chlorpyrifos and diazinon reduced birth weight," she said. After the Environmental Protection Agency banned these products for indoor use in 2000, infant birth weights and birth lengths improved immediately, the research found.
Fenster went on to chronicle an Italian study, which indicated that higher levels of DEHP and its metabolite MEHP in umbilical cord blood were associated with shorter gestational times. Babies with DEHP exposure were born, on average, one week earlier than unexposed babies, she said.
Furthermore, according to a case-control study of women from 10 California counties, occupational and home exposures to pesticides during early pregnancy are linked to stillbirths, she said. She recommended that future studies on the effects of chemicals on pregnancy outcomes "should use a prospective design, assess exposure with biomarkers where available, and consider genetic susceptibility."
Patricia A. Hunt, a geneticist at Washington State University, described research that suggests that the weak estrogen BPA could perturb meiosis, that is, cell division, in developing human eggs. She and her colleagues exposed pregnant mice to a low dose of BPA—20 µg per kg of body weight per day—for one week and found that the process of meiosis was perturbed in the eggs of the female fetuses. Forty percent of the eggs were chromosomally abnormal. After these mouse pups reached adulthood and their eggs were fertilized, these perturbations led to an increase in chromosomally abnormal embryos, she said.
"Our studies also suggest that BPA acts on the fetal ovary, not by mimicking the actions of estrogen but by interfering with the function of one of the known estrogen receptors," Hunt explained. "These results raise the possibility that BPA and other substances—both man-made and naturally occurring—that mimic or interfere with the action of estrogen may affect early egg development in humans," she warned. If BPA perturbs egg development, it would affect the grandchildren of the exposed mother, increasing the chances they will have abnormal chromosomes, she said. One in 10 human fetuses is chromosomally abnormal, she said. Many of them are lost due to miscarriage or abortion after tests indicate a problem, and those that survive are generally profoundly disabled with disorders such as Down's syndrome, Hunt observed.
"The levels of BPA the female mice were exposed to were only slightly higher than what people commonly encounter," Hunt said. A common source of exposure to BPA is polycarbonate plastic, which is made from BPA, in items such as clear plastic baby bottles and many types of food and drink containers.
One mechanism by which BPA and other environmental contaminants could affect development of the female reproductive tract is by altering a particular pattern of gene expression, said Hugh S. Taylor, a researcher at Yale University School of Medicine. The ensemble of genes involved are known as the HOX genes, which function in patterning the body axis, thereby determining where limbs and other body parts will grow in a developing fetus. For example, it is well-known that DES affects the development of the uterus in the human fetus, he said. It does this, at least in part, through its ability to alter the normal pattern of HOX gene expression.
"We tested the ability of BPA to alter expression of HOXA10, the HOX gene necessary for uterine development in the fetus," Taylor said. To determine if in utero BPA exposure results in a lasting alteration of uterine HOXA10 expression, pregnant mice were treated with 0.5 to 5.0 mg per kg of BPA on gestational days nine to 16, he said. A dose-response increase was seen in stromal cell HOXA10 expression in adult mice exposed in utero. Because BPA, as well as several other synthetic estrogens, alters HOX gene expression, it appears that HOX genes are a common target of endocrine disruption, he said. "In utero exposure to synthetic estrogens produces reproductive tract alterations by imprinting [silencing] essential developmental regulatory genes."
Louis J. Guillette Jr., a zoologist at the University of Florida, suggested another molecular biology context in which environmental chemicals interact with genes and affect reproductive health. "Recent research on animals has begun to uncover mechanisms, such as [DNA] methylation, by which environmental factors can lead to altered gene expression patterns and consequent abnormalities of the reproductive and immune systems," he explained. For example, when cytosine nucleotides in DNA are methylated to 5-methylcytosine, this affects gene expression, he said, and the number and pattern of methylated cytosines affect the expression of specific genes.
"When a gene is not methylated it can be turned on, but if it is methylated it can't be," Guillette said. "Environmental contaminants can cause some genes to be methylated at times during development when methylation would not normally occur." Low levels of methylation are associated with genes that potentially can be turned on, and high levels of methylation correspond with low gene activity.
At this time in the 21st century, "we have more knowledge of the genetic basis for health and disease than at any time in history," Guillette said. "A growing body of scientific literature indicates that it is the interaction of an organism's genetic makeup with various environmental factors that primarily leads to health or disease. Yet we continue to use genetic information as if disease were a direct result of a heritable gene mutation alone."
Some general principles should be considered when conducting or evaluating research on the effects of environmental exposures, said Linda S. Birnbaum, director of EPA's Experimental Toxicology Division. "Environmental chemicals can cause a broad spectrum of effects, which depend not only on the route of exposure and dose but on the susceptibility of the exposed individual," she said. For example, age, gender, and genotype can determine whether an exposure causes a problem.
Even though the research is complicated and much is not known, some results are quite clear, Birnbaum said. "By using biomarkers to measure exposure and effect, as well as studies of some unfortunate poisoning episodes, we do know the effects of some environmental exposures," she observed. "Just as in pharmaceutical research, animal studies give us much of our knowledge. When a particular environmental toxicant causes multiple effects in multiple species, people are highly likely to be susceptible."
It is clear from the talks and data presented at the meeting that exposure to environmental chemicals is harming reproductive health and fertility. But not enough research has been done to tease out the relative importance of the various contaminants. Also, as several speakers pointed out, the interaction of many different factors needs to be considered when conducting and evaluating this research.
"We should regularly and systematically look at the convergence of multiple factors when trying to understand the incidence, prevalence, and distribution of diseases, disabilities, and other conditions," said Ted Schettler, director of the Science & Environmental Health Network, a consortium guided by a better-safe-than-sorry mindset. "Multiple factors—genetics, chemical exposures, infectious agents, nutritional status—interacting in complex ways determine reproductive health status in individuals or populations."