4
Preterm Birth—Gene–Environment Interactions*

The environment in which we live, play, and work has long been suspected to play a major role in determining the health of individuals and animals. Using a broader definition of environmental health, workshop participants addressed some of the research that provides a direct linkage between environmental toxicants and preterm birth. One observation that has led to many studies of the role of the fetus in initiating labor and an early clue to disruption of normal pregnancy is the ingestion of Veratrum californicum, a flowering plant found commonly in Rocky Mountain meadows. It is known to contain a toxic compound that produces malformations in the sheep fetus when ingested during pregnancy. According to Nathanielsz, the fetus is severely deformed, with a large central cyclopian eye, and macroglossia (the tongue protrudes from the mouth because it is too large), and the base of the brain is disorganized. Further the fetus continues to grow and, at term, onset of labor of labor does not occur (see Figure 4.1). Observations such as this have led researchers to ask how the environment may interact with the normal processes of pregnancy to interfere with labor and delivery. This chapter summarizes some of the research and research programs that may shed light on environmental causes of preterm birth.

*  

This chapter was prepared from the transcript of the meeting by a rapporteur. The discussions were edited and organized around major themes to provide a more readable summary and to eliminate duplication of topics.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary 4 Preterm Birth—Gene–Environment Interactions* The environment in which we live, play, and work has long been suspected to play a major role in determining the health of individuals and animals. Using a broader definition of environmental health, workshop participants addressed some of the research that provides a direct linkage between environmental toxicants and preterm birth. One observation that has led to many studies of the role of the fetus in initiating labor and an early clue to disruption of normal pregnancy is the ingestion of Veratrum californicum, a flowering plant found commonly in Rocky Mountain meadows. It is known to contain a toxic compound that produces malformations in the sheep fetus when ingested during pregnancy. According to Nathanielsz, the fetus is severely deformed, with a large central cyclopian eye, and macroglossia (the tongue protrudes from the mouth because it is too large), and the base of the brain is disorganized. Further the fetus continues to grow and, at term, onset of labor of labor does not occur (see Figure 4.1). Observations such as this have led researchers to ask how the environment may interact with the normal processes of pregnancy to interfere with labor and delivery. This chapter summarizes some of the research and research programs that may shed light on environmental causes of preterm birth. *   This chapter was prepared from the transcript of the meeting by a rapporteur. The discussions were edited and organized around major themes to provide a more readable summary and to eliminate duplication of topics.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary ENVIRONMENTAL FACTORS ASSOCIATED WITH PRETERM BIRTH Adverse effects on the developing organism, which may be detected at any point in the life span of the organism, may result from exposure, prior to conception, of either parent, during prenatal development, or exposure postnatally up to the time of sexual maturation. Researchers are only beginning to identify those exposures and levels of exposures that affect reproductive outcome. According to Matthew Longnecker, National Institute of Environmental Health Sciences (NIEHS), the environmental risk factors for other reproductive outcomes can differ from those for preterm birth. The potential risk factors studied can be divided into a number of broad categories, including: occupation or occupational exposures, air pollutants, exposure to POPs (persistent organic pollutants), exposure to DDE (1,1-dichloro-2, 2-bis(chlorophenyl) ethylene) a metabolite of DDT (1,1,1-trichloro-2,2-bis (p-chlorophenyl) ethane (DDT), and PCBs (polychlorinated biphenyls), metals (e.g., lead, arsenic), water disinfection byproducts, and video display terminals. FIGURE 4.1 Sheep that ingest the flowering plant Veratrum californicum during pregnancy prolong the length of gestation. A pair of twin sheep delivered at 230 days of gestation by cesarean section (left), approximately 80 days longer than the normal pregnancy (right). This would be equivalent to a woman being pregnant for 15 months and not showing any indication of entering the stages of labor. SOURCE: Binns et al., 1964. © 1964 Annals of the New York Academy of Sciences. Reprinted with permission.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary Among the exposures linked to occupation and occupational exposures, factors that were included in the discussions at the workshop were for those exposures in which the relative risk of preterm birth was 1.5 or greater or the finding was statistically significant, and the finding was reported in two or more studies. Longnecker noted a difference of relative risk between women and men. Women who were employed in metal working, electrical occupations, janitorial work, food service, and textiles had a higher relative risk. In contrast, men who were working in food service or textiles had some association. This sparked some discussion among participants, who wondered why food service as an occupation had a higher relative risk. Some participants noted that occupations and titles often are not clear and additional work would be needed to answer the questions surrounding these observations. There was a further difference between females and males regarding occupational exposures. Longnecker noted that solvents were the most common risk factor for maternal occupational exposures while the most common risk factors for males were pesticide exposures. The second category of exposure that has been studied in relation to premature birth is air pollutants. These can be divided into two types: tobacco-related (e.g., passive smoking, environmental tobacco smoke) and traditional air pollutants (e.g., particulates, sulfur). The evidence on direct tobacco use is discussed in Chapter 5; however, Longnecker reported that studies showed that a moderately elevated risk of premature birth was associated with direct smoking. It was interesting that the relative risks for environmental tobacco smoke tended to be higher than those for direct smoking. Researchers are beginning to study the relationship between traditional air pollutants and premature birth. Longnecker noted that of the three studies that had ecologic time-series designs and some degree of individual-level data, in two there was an increase in preterm delivery associated with exposure to sulfur dioxide, which comes primarily from coal burning. Even at lower levels of particulates, the relative risk associated with particles less than 10 microns in diameter was high as measured in one study in Los Angeles. Two other studies also supported an adverse effect of total suspended particles. The time of exposure during pregnancy may be important for causing preterm birth. In one study, researchers found that if exposures occurred in the first trimester, the relative risks were greater. Longnecker noted that many questions remain. For example, data on sulfur dioxide in the United States are lacking. They don’t burn coal in Los Angeles, but they do in many other places in the United States. Collecting better data would be a challenge because individual-level measurements are very expensive. The next category of risk factors involves DDE, the metabolite of the pesticide DDT. Although the use of DDT has been banned in the United States since the 1970s, it is still used or approved for malaria control in 25 countries. The levels of the metabolites in humans are at now 5 percent of the levels found in those areas where DDT spraying is ongoing. In a recent study, Longnecker looked at the stored serum of U.S. women from the 1960s—a time when DDT

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary TABLE 4.1 Odds Ratio for Preterm Birth DDE (µg/L) Cases (n=361) Controls (n=2019) Odds Ratio for Preterm Birth* 95% Confidence Interval I <15 34 375 1   15–29 153 944 1.5 1.0–2.3 30–44 80 404 1.6 1.0–2.6 45–59 50 176 2.5 1.5–4.2 ≥ 60 44 120 3.1 1.8–5.4 *A Preterm birth is defined as less than 37 weeks gestational age for this experiment. NOTE: Trend p < 0.0001. SOURCE: Modified from Table 2 in Longnecker et al., 2001. Reprinted with permission. spraying was occurring. As shown in Table 4.1, there was an increased risk of preterm birth with increasing levels of DDE in serum (Longnecker, 2001). In the 2000s, most individuals have a very low (less than 15 micrograms per liter) DDE level. He noted that upon further examination there was a flat dose-response relation for exposures of less than 15 micrograms per liter. This suggests that DDE exposure as a risk factor for preterm birth may be more important in those countries where DDT is used for malaria control but not in the United States. Longnecker reviewed a number of other potential risk factors for pre-term birth—many of which had few supportive data, conflicting results, or not enough information to make a conclusion. Chemical exposures such PCBs have shown an association in some studies, but further work is needed. Metals such as arsenic, cadmium, and lead have not been fully investigated. In Bangladesh, where arsenic levels are twice as high as in the United States, there is a suggestion from one study that there may be an association. Alcohol Use According to some workshop participants, a study done at the University of Washington suggests that women and their partners who used alcohol and drugs reported significantly higher levels of stress, weaker social support, and poorer levels of self-esteem (Lindenberg et al., 1999). Whether this accounts for an increase in preterm labor has yet to be investigated. The effects of alcohol on preterm birth are complex and can induce different pregnancy outcomes depending on the differential pattern and timing of exposure. In the rat and the rabbit, research has shown that parturition is prolonged if alcohol is given chronically (Cook and Randall, 1998). However, alcohol consumption on day 17 of gestation has been shown to induce preterm birth in the mouse. Kimmel reported that an ongoing study in humans suggests that moderate drinking tends to be associated with preterm labor, while social drinking has no effect.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary Tobacco Use Cigarette smoking by women during pregnancy continues to be a substantial contributor to poor perinatal outcomes in the United States. As discussed by many workshop participants, smoking during pregnancy is associated with low birth weight, premature birth, and reduced neonatal lung function. In addition to these effects on the fetus and infant, pregnant women who smoke are more likely to have problems such as miscarriage, premature rupture of membranes, and placenta previa (Surgeon General, 1989, 1994). A clear dose-response relationship exists between the number of cigarettes smoked during pregnancy and the birth weight deficit. Compared with nonsmokers, light and heavy smokers have a 54 and 130 percent increase, respectively, in the prevalence of newborns weighing less than 2,500 grams (Surgeon General, 1989). The reduction in birth weight associated with maternal tobacco use seems to be a direct effect of smoking on fetal growth (Surgeon General, 1989). Mothers who smoke also have increased rates of premature delivery. The newborns are also smaller at every gestational age. The infants display generally symmetrical fetal growth retardation with deficits in measurements of crown-heel length, chest and head circumferences, and birth weight (Surgeon General, 1989). Low birth weight infants have a greater risk of neurological and developmental handicaps and congenital anomalies. In addition, they may be susceptible to respiratory infections as well as other disorders. Preterm low birth weight was associated with a higher risk of poor health and asthma among children when all the other selected risk factors were controlled. Poor maternal health and maternal smoking were important risk factors for poor child health (Chen and Millar, 1999). Tobacco and Gene–Environment Interactions Wang and colleagues investigated maternal cigarette smoking, metabolic gene polymorphism, and reduced birth weight in a U.S. population in order to examine whether maternal genotypes can modify the association between maternal cigarette smoking and infant birth weight. Two metabolic genes that are involved in benzene detoxification and excretion, CYP1A1 and GSTT, were used to assess a metabolic gene–smoking interaction. Consistent with previous data, the study found that maternal smoking during pregnancy was associated with reduced infant birth weight and gestational age. The study further demonstrated that the adverse effects of maternal smoking on infant birth weight and gestational age were modified by maternal genotypes, suggesting gene–smoking interactions.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary CONVERGENCE OF CELLULAR PATHWAYS The work of Rita Loch-Caruso, University of Michigan, provides insight into the cellular mechanisms by which some environmental chemicals such as DDT could initiate parturition. According to Loch-Caruso, it is often difficult to study toxicant effects that might influence parturition using the standard small laboratory animal models because of the diversity among species in initiating parturition. Her work is based on the knowledge that there is an apparent convergence of mechanisms for parturition at the level of the myometrium. At the level of the uterine smooth muscle, there are similarities across species in the myometrial responses that are necessary for parturition. By recording contractions from the myometrium of pregnant rats in a standard organ bath, her laboratory has found that some toxicants, such as PCBs and DDT, can act directly on the myometrium to alter contractions. For example, DDT, which was described above as having a potential role in premature birth, was shown to increase the oscillations (contraction frequency) of the myometrium. The increased response persisted even after rinsing the strips with buffer (see Figure 4.2). In addition, PCBs, also candidates for reproductive effects, were tested in the organ bath. The commercial PCB mixture, Aroclor 1242, showed a dose-dependent effect on myometrial contraction. At 10 µM Aroclor, there was little effect; however 50 and 100 µM Aroclor increased the frequency of contractions. FIGURE 4.2 (a) Polygraph tracing showing the effect of 100 µM o,p′-DDT (1,1,1-trichloro-2,2-bis(p-chlorophenyl) ethane) on rat uterine contraction frequency during the post-treatment period. The top two tracings are controls and bottom two represent DDT-treated strips, all of the same animal. The duration of the period shown is 10 minutes. (b) Effect of o,p′-DDT on isometric contraction frequency in rat uterine strips treated in vitro. SOURCE: Juberg et al., 1991. © 1991 Society of Toxicology. Reprinted with permission.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary By understanding some of the basic cellular mechanisms of myometrial activation, Loch-Caruso and her colleagues have begun to improve our understanding of how chemical toxicants might recruit these normal processes. Aroclor 1242 (100 µM) was shown in her laboratory to increase the intracellular concentration of calcium by approximately 778 percent. This increase was dependent on the influx of calcium through voltage-dependent calcium channels in the cellular membrane. Removal of calcium from the buffer (extracellular space), or blocking the L-type calcium channel with nifedipine, blocked the increase. The cellular mechanism by which this occurs is still under investigation; however, Loch-Caruso recently proposed that Aroclor 1242 increases phospholipase A2 to release arachidonic acid, which works through a direct or indirect mechanism on the voltage-dependent calcium channels to increase intracellular calcium. There is some evidence for this pathway. Using reverse-phase high-performance liquid chromatography (HPLC) to analyze released arachidonic acid, 100 µM Aroclor 1242 increased the concentration of arachidonic acid (over controls) that was not converted to its metabolites. Further by using uterine strips in the organ bath, she was able to show that pretreatment with phospholipase A2 inhibitors was able to block the stimulatory action of Aroclor 1242. Loch-Caruso concluded that some environmental toxicants can act directly through cellular mechanisms to alter uterine contractions. In vitro approaches using laboratory animal myometrial tissue and cells may prove useful for assessing the potential risks of some environmental chemicals for pregnant women. She cautioned however, that these procedures will not detect chemicals that act through other mechanisms, for example, a toxicant that acts at the level of fetal and maternal signaling. GENE–ENVIRONMENT INTERACTIONS AND PRETERM BIRTH To date, most studies of preterm delivery have focused on social, environmental, or clinical variables. The role of genetic susceptibility and gene–environment interactions has been largely unexplored, in part because the study of complex gene–environment interactions involves dealing with a number of methodological challenges, including determining appropriate study design and the appropriate statistical analysis for testing the gene–environment interaction in relation to preterm births. Empirically, gene–environment interactions can be assessed in epidemiological model in which markers of genetic susceptibility can be incorporated to test these interactions. These regression models incorporate the genetic effect, environmental effect, and gene–environment interactions. In studying gene–environment interactions, it is critical to accurately measure environmental exposures and to determine the genotypes. However, in reality, misclassification of environmental exposures can occur because of poor recall of previous exposures, complex patterns of long-term exposures, and lack of

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary good biological indicators of exposure levels. In addition, the timing of exposures, given the importance of the development of the fetus, is unknown, and there can be misclassification of genotypes. Study findings also could be biased because of confounding variables, from either environmental or genetic sources (e.g., linkage disequilibrium). Other unique challenges faced by those conducting research on preterm birth, compared to those conducting research involving other complex human diseases, are determining the role of maternal versus fetal genes in the metabolic pathways involved in the pathogenesis of preterm birth and identifying appropriate study designs and analytical methods for addressing this issue. Inspired by previous work on the intergenerational influence on low birth weight, Xiaobin Wang of Boston University School of Medicine began to study the aggregation of low birth weight among Caucasians and African Americans in the United States by assessing the association between the birth weight of the mother and the index child and the risk of low birth weight among the siblings of that index child. The study consisted of a cohort of the 1988 U.S. National Maternal and Child Health Survey, and the analysis included 1,691 Caucasian mothers and 1,461 African-American mothers who had two or more live-born singleton children. The four study groups were defined as follows: (1) neither the mother nor the index child had a low birth weight; (2) only the mother had a low birth weight; (3) only the index child had a low birth weight; and (4) both the mother and the index child had low birth weights. As illustrated in Figure 4.3, the lowest odds ratio is found in the first group and the highest in the last. The pattern is similar for both Caucasian and African-American mothers. In fact, Wang et al. (1995) found that the risk of low birth weight and preterm birth was greatly increased if both the mother and the index child were of low birth weight. The findings persist in both African-American and Caucasian populations even after adjustment for maternal and infant characteristics. Wang suggested that the strong familial aggregation of low birth weight might be attributable to environmental factors, genetic factors, or both. To begin to address the questions of gene–environment interaction, Wang began a second large-scale, epidemiological study of preterm births in a low-risk, homogeneous Chinese population as a function of benzene exposure. Wang’s study sought to examine genetic differences in benzene metabolism and its relationship to the length of gestation. Benzene was of interest because it is a known reproductive toxicant and is a ubiquitous solvent. According to the National Institute of Occupational Safety and Health, an estimated 10 million U.S. workers are exposed to benzene. Further, most Americans are exposed to minor amounts of benzene through various sources, including the gastrointestinal system. Metabolism is essential for benzene detoxification, and it is speculated that an individual’s reproductive risk associated with benzene exposure may be modified by genetic variation in metabolic detoxification activity. A two-phase pro-

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary FIGURE 4.3 The risk of low birth weight may be predicted by whether the mother was low birth weight or whether she had a prior low birth weight infant (index child). The risk is lowest when neither the mother nor the sibling was low birth weight (group 1), only the mother had low birth weight (group 2), or only the index child had low birth weight (group 3). The highest risks were noted when both the mother and the index child were low birth weight (group 4). SOURCE: Wang et al., 1995. © 1995 Massachusetts Medical Society. Reprinted with permission. cess is necessary to detoxify and excrete benzene. The cytochrome P450 family, including CYP1A1, serves as the major enzyme system involved in phase one. Phase two relies on glutathione S-transferases such as GSTT1 to make the end product a stable hydrophilic compound that can be excreted easily. The study population consisted of 542 postpartum Chinese female workers, of whom 302 had known low-level benzene exposure and 240 had no known exposure during pregnancy. As illustrated in Figure 4.4, benzene was shown to shorten the gestational age, which was dependent on maternal genotype. When the mother had the AA allele, there was a decrease in the gestational age at birth. The largest association occurred when the mother had AA allele and GSTT1 was absent. This study demonstrates that benzene exposure, even at a very low level, was associated with shortened gestation, thus suggesting a gene–environment interaction. IMMUNE-DEFICIENT KNOCKOUT MICE AS A MODEL TO ELUCIDATE ENVIRONMENTAL TOXICANTS Participants considered other animal models to begin to understand the effect of environmental toxicants on preterm birth and other reproductive abnor-

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary FIGURE 4.4 Gestational age by maternal benzene exposure status and CYP1A1 and GSTT1 genotypes, with adjustment for maternal age, education, parity, passive smoking, stress, prepregancy weight and height, and infant sex, Beijing, China, June 1995 to June 1997. SOURCE: Wang et al., 2000. © 2000 Oxford University Press. Reprinted with permission. malities. During her presentation, Anne Croy of the University of Guelph suggested that if mice lacking lymphocytes are housed under strict microbiological barriers, they are excellent models for reproductive studies because they are immune deficient. The immune system is composed of highly mobile cells that normally circulate in the blood and have the ability to move in and out of tissues. Immune-competent cells move in and out of tissues by recognizing receptors on cells, particularly on endothelial cells, and by responding to gradients that can either attract or repel the lymphocytes within the tissue. The mammalian uterus is endowed with all of the known lineages of immune-competent cells and as such is a highly active immune organ.

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary In women, pregnancy is accompanied by the transient appearance in the uterus of a population of granulated lymphocytes. These specialized lymphocytes become the dominant immune cells in the first half of gestation and have been estimated to represent more than 70 percent of the maternal bone marrow-derived cells at implantation sites. This population, which has been defined in women and in mice as a subset within the natural killer (NK) lymphocyte lineage, has a strong association with uterine stroma. The granulated cells first appear and proliferate as uterine stromal cells transform from fibroblasts into decidual cells. Uterine natural killer (uNK) cells decline in the later part of pregnancy and are absent from the postpartum uterus. Croy has found that mice that are genetically deficient in lymphocyte lineages have quantifiable implantation site defects. Specific genes are key to regulating and controlling the health of the site of uterine implantation. In one model, interferon-gamma (IFN-g), a key product of NK cells, appears to be a central regulator of the expression of these genes and could thus serve as a potential target molecule for monitoring the effects of environmental contaminants on the uterine immune system. Establishing and quantifying consistent phenotypes in the uterine implantation sites of specific immune-deficient mice can provide powerful animal research tools in which reconstitution of implantation site structure can be assessed. These approaches involving reconstitution can be exploited in many ways in investigations of environmental toxicants, said Croy. She described a number of models (see “Models That Study the Action of Uterine Lymphocytes During Pregnancy” in the abstracts) that focus on uNK cells as indicators of environmental interference in the implantation process and, thus have some effect on the subset of preterm births that may be due to problems of implantation. FEDERAL GOVERNMENT TOXICOLOGY PROGRAMS In addition to the work at academic institutions, the federal government has a number of research and testing programs that are poised to test reproductive and development effects of chemicals. Three such programs are housed at the National Institutes of Health (NIH), the Environmental Protection Agency (EPA), and the Food and Drug Administration (FDA). The EPA has regulatory requirements for industry to conduct testing on pesticides and industrial chemicals, and the FDA has requirements for the testing of pharmaceuticals, food additives, and contaminants. The NIEHS National Toxicology Program conducts testing on a number of chemicals. EPA’s Risk Assessment Guidelines In her role at the Environmental Protection Agency, Carole Kimmel has been actively involved in the development of a number of risk assessment guide-

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary lines, three of which are most relevant for a discussion of preterm birth and environmental chemicals: developmental toxicity, reproductive toxicity, and immunotoxicity. Current protocols use rodents as a model, although developmental toxicity testing also uses rabbits. Extrapolating the results from animals to humans may be problematic because of the differences in physiology. According to Kimmel, some of the differences are known, while others are not clearly understood. It is not clear from the literature how important some of these differences are for the human situation. Developmental toxicity research looks at the adverse effects on the developing organism that result from exposure prior to conception (an exposure of either parent prior to the pregnancy), exposure during prenatal development; or exposure postnatally, to the time of sexual maturation. With the recognition that exposures during development may have effects that can be detected at any point in the future, there is a burgeoning area of research today on fetal origins of adult disease. The possible types of adverse developmental outcomes include death through early spontaneous abortions, perinatal death, or postnatal deaths; birth defects; altered or reduced growth; and functional deficits, e.g., mental retardation. Related to the work on developmental toxicity is the study of reproductive toxicity that includes adverse effects on the reproductive system, such as changes in gestation, parturition, lactation, onset of puberty, gamete production and transport, and reproductive senescence. The EPA, along with other regulatory agencies, have developed a set of harmonized testing guidelines that are used by industrialized countries as guidance for testing of pesticides and industrial chemicals, according to Kimmel. These include the prenatal developmental toxicity study, which is typically done on rats or rabbits, although other species are used if warranted. Exposure occurs from day 6 of gestation, which correlates with the time of implantation in the rabbit, to approximately days 29–30 in rabbits or day 20 in rats—just prior to term, which allows for systematic evaluation of structural defects. Because of the testing procedure, it is not possible to study the issue of preterm delivery. From other research, it has been shown that rats shows prolongation of gestation rather a preterm delivery. Rabbits, however, do have preterm delivery, as early as days 17–18, but may also deliver as late as days 28–29. Thus, the rabbit may be a better model for understanding the effects of environmental exposures on preterm delivery. Another study for which a harmonized testing guideline exists is the developmental neurotoxicity study, which was developed primarily to provide a better evaluation of developmental effects on the nervous system. It is used typically after a chemical has shown other indications of possible neurotoxic effects, in either adult or prenatal studies, and usually in rats. The original protocol was developed to start exposures around implantation and finish by postnatal day 10. However, the exposure period has recently been extended out to weaning (postnatal day 21), in order to cover even more of the developing nervous system. In

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary this type of study, it is possible to evaluate the timing of parturition and gestational length, and to measure both maternal and pup toxicity. The third testing protocol that is most relevant to the discussion of preterm birth is the two generation reproduction study, which is a standard study that is used for all pesticides. It is done mainly on rats, although the mouse may be used in some testing. In this protocol, the exposure begins before mating and continues until lactation in the F1 generation. Selected offspring are continued with the exposure through mating and lactation of the F2 offspring. This allows researchers to evaluate a long-term chronic exposure through two generations. The study is designed to look primarily at developmental effects as well as reproductive development and function. It follows development of the reproductive system over two generations, including gestation length. However, as indicated before, rats typically do not abort or deliver prematurely, even in the face of severe toxicity. Drug Testing and Preterm Birth The Food and Drug Administration has also developed testing guidelines and approaches for reproductive and developmental toxicity. These cover a number of drugs, chemicals, additives, and contaminants, and the testing guidelines are similar to those that EPA has developed for pesticides, according to Kimmel. For pharmaceutical agents, the approach to testing varies some, using what is called a three-segment approach. Segment I covers exposure from mating until the time of implantation. It follows the animals until they have offspring. Segment II is similar to the prenatal developmental toxicity study. The exposure is from the period of implantation to around the time of closure of the hard palate. Segment III is from the time of closure of the hard palate through weaning. Animal Data on Preterm Birth Through a review of the literature, Kimmel described some of the studies that have been reported on premature delivery from exposure. Using protocols similar to those in the testing guidelines, researchers found that parturition in the rat and rabbit is delayed if alcohol is given chronically. However, if alcohol is given acutely on approximately day 17 of gestation in the mouse, the animals experience preterm labor. According to Kimmel, there seems to be a differential effect of alcohol depending on the timing and pattern of exposure. A number of agents such as diethylenetriamine, tioconazole, and methyl methacrylate in rats, and prulifloxacin, idoxifene, and ostidine in rabbits, have been shown to cause premature delivery. In concluding, Kimmel noted that the current testing guidelines do include some evaluation of gestation length, but the question remains whether researchers have the appropriate animal models to ask the appropriate questions regarding preterm birth. She suggested that current study designs may be able to be

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary modified to look at parturition in more detail. Currently, when effects are reported, it is generally at higher doses where there are other signs of maternal and developmental toxicity. NIEHS’s National Toxicology Program The National Institute of Environmental Health Sciences’ mission is to reduce the burden of human illness and dysfunction from environmental causes through a better understanding of environment, individual susceptibility and age, and how they interrelate. To meet part of this goal, the National Toxicology Program (NTP) was created to coordinate the toxicological testing programs within the Department of Health and Human Services and to strengthen science-based knowledge. This is an arduous task because there are approximately 80,000 chemicals approved for use in the United States, and approximately 2,000 additional chemicals are introduced annually. A major emphasis of the NTP is to safeguard public health by identifying and characterizing toxic effects of environmental chemicals, including the effects on reproduction, said Jack Bishop of NIEHS. Research designs used by the NTP to assess reproductive toxicology include (1) the total reproductive capacity test (TRCT); (2) the dominant lethal test; (3) the reproductive assessment by continuous breeding (RACB) test; (4) the reproductive, development, and general toxicity (RDGT) test; and (5) the teratology test. He suggested that there may be opportunities for adjustment of these test methods to ask questions that are relevant to the issues of premature birth. The total reproduction capacity test is primarily conducted in mice and generally involves single acute exposures. Approximately 30 females are treated per chemical dose group and concurrent control. After treatment, each female is caged with a male and mated for her reproductive life span. The end points measured are average litter size per mating interval, the proportion of females with young per mating interval, and the mean total number of offspring per female across all mating intervals. The TRCT was originally designed to identify environmental agents that might induce genetic damage in female germ cells or otherwise cause premature cessation of ovulation. The TRCT, as conducted for these purposes, does not capture length of gestation information needed to assess premature birth. One of the classic toxicology study paradigms is the dominant lethal assay, according to Bishop. Male or female animals are given either a single or multiple exposures to a chemical. The females are checked for the presence of a vaginal copulatory plug, so the time of mating is known in this test. With this paradigm researchers measure the number of implantation sites, the number of viable and dead fetuses and the number of resorptions. A minimum of 300 implantations per dose group need to be analyzed for an adequate test. The conduct of a female dominant lethal test is more complex than a male dominant lethal test because

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary true genetic damage to the oocyte can be easily confounded by maternal toxicities. The disadvantage of this paradigm for the study of preterm birth is that the females are sacrificed prior to term, which means that information regarding gestational length cannot be obtained. The third paradigm is reproductive assessment by continuous breeding. Exposure occurs during the F0 and F1 generations from prior to conception through weaning. Similar to the EPA two-generation study, some F1 offspring are exposed and mated. Generally during this experiment they do not check for vaginal copulatory plug, so data on gestational length is not captured. “Average Gestation Length” has been determined in select studies where plug checks have been performed and another statistic, “Cumulative Days to Litter”, has been calculated in the absence of plug check data (see Table 4.2). The final assay is the reproductive, developmental, and general toxicity protocol. This assay or study paradigm has many versions that can look at different outcomes (gestational length, average time to litter, cumulative time per litter, numbers of pups per litter, decreased pup weight, altered pup sex ratio, etc.), and may be applicable for studying preterm birth. Of the chemicals reviewed to date using either the RACB or RDGT paradigms, only three exhibited a significant exposure-related change in gestation length (see Table 4.2): ethoxyquin increased the time to litter, and sodium selenate increased the gestation length. Only nitrofurazone decreases the average time to litter. The current animal models that test for chemical toxicity may be inadequately designed to capture information on chemically induced alteration of gestation length, according to Bishop. We don’t report many effects, which may suggest that we need to look at the experimental design. Further, the rodent model has a number of limitations. The hormonal milieu of the rat and mouse are very different from that of humans. Another critical issue, in terms of gestational length, is that in the rodent we must measure “hours of gestation,” which current protocols are not designed to measure. He further suggested that we need to consider developing alternative models to better test for potential environmental effects on preterm birth. The current animal models that test for chemical toxicity may be inadequately designed to capture information on chemically induced alteration of gestation length. Jack Bishop

OCR for page 46
The Role of Environmental Hazards in Premature Birth: Workshop Summary TABLE 4.2 Chemicals Reviewed in RACB 1/N RDGT Studies Average Gestation LengthST Cumulative/Average Days to Litter Bromoacetonitrile Methylene Blue trihydrateT2/4 Bromochloroacetic acid Nitrofurazone ↓ Bromodichloromethane Propylthiouracil F0/1 Chlorodibromomethane Sodium BromateF0/1 Dibromoacetonitrile TetrachlorobenzeneF0/1 Hexachloroacetone ThiophenolT2/4 Sodium bromate AZTF0F0/1 Sodium selenate ↑ CD1 (22 vs 23 days)* BTCAF0/1 Tribromoacetic Acid CaffeineT2/4 Sodium Bromate DibutylpthalateF0/1 ElmironT2/4 DicyclopentiadineT2/4 MethacrylonitrileT2/4 DiethylhexylpthalateF0/1 Tamoxifen citrateF1 EthoxyquinT2/4 ↑ Potassium dichromateT2/4 BalbC 19.5 HexachlorobenzendF0/1   Indium trichloride TetrahydrateT2/4 IsoeugenolF0/1 Of the 30 RACB-RDGT studies reviewed, only three chemicals altered either time to litter or gestational length. The remaining chemicals listed in this table have been tested but not found to have an effect on either time to litter or gestational length. Ethoxyquin increased the time to litter and sodium selenate increased the gestational length. Only nitrofurazone decreased the average time to litter. * Sodium selenate increased the time of gestation from 22 to 23 days. SOURCE: NIEHS/NTP Reproductive Toxicology Database. Reprinted with permission.