fetus. Toxic chemicals in the environment act on reproduction directly by affecting the conceptus, or by affecting maternal organs that can result in altered hormone secretion and hence altered reproduction (6). Estimating the incidence of these effects is difficult and determining the causes is even more complex because of the many confounding factors that must be considered. These factors include but are not limited to: incomplete information concerning dose, timing, and duration of exposure; unknown interactions between causes; difficulty in obtaining specimens; inadequacy of analytical techniques; the large number of potential causes of abortion; and variations in individual susceptibility due to differences in genotype (7). To help clarify these effects on reproduction, this review will address the physiology of pregnancy and fetal development, the etiology and mechanisms of reproductive failure and teratogenesis, and assay systems to measure these effects.
Physiology of pregnancy Gestation is the period of fetal development from the time of conception to the time of birth. The average length of gestation in humans is 266 days or approximately 9 months (8). Gestation is commonly divided into three periods of three months each referred to as the first, second, and third trimesters. Fertilization of the ovum by the sperm, or conception, occurs in the fallopian tube, which connects the ovary to the uterus. It takes the fertilized ovum three days to reach the uterus and another four to five days before implantation in the wall of the uterus occurs. The blastocyst, a hollow sphere of cells, is formed during the first week of gestation. At this stage, the cells comprising the conceptus are not differentiated. Therefore, injury to a few of these cells does not result in a specific developmental defect but in an overall delay in the development of the fetus. However, if a sufficiently high dosage of toxicant reaches the conceptus, death can occur (9). By the end of the second week, blood flow through the embryo is established (10) via the placental connection between the mother and embryo. Embryonic differentiation begins during the third week of gestation when the cells of the blastocyst are segregated into the three embryonic germ layers, referred to as endoderm, mesoderm, and ectoderm. During the embryonic period (Weeks 4-8), each of the three germ layers gives rise to specific groups of cells known as primordia. During this process, known as organogenesis, the differentiated cells have more specialized metabolic requirements and are 1188
Environ Sa.Technol., Vol. 23. No. 10, 1989
therefore more vulnerable to damage by adverse influences. Damage to a particular primordium at this time may result in specific structural defects in the fetus. Upon completion of organ formation at the end of the 12th week, the induction of major structural defects is no longer a factor of concern (9). The period from the beginning of the third month to birth is known as the fetal period. It is characterized by rapid growth of the fetus and continued tissue differentiation or histogenesis (11). The progress of histogenesis is closely correlated with the development of the functional activity of the fetal organs. Adverse influences during this stage result in microscopic structural defects and possible functional abnormalities (9). Because structural and functional maturation continue after birth in many organ systems such as the immunologic system, nervous system, liver, kidneys, and other endocrine organs, there is a growing concern about the possible adverse effects of environmental factors during infancy and childhood (9).
Reproductive failure, teratogenesis Interference with the normal development of a fetus caused by exposure to various compounds may have one of four possible outcomes-death, malformation, growth retardation, or organ dysfunction of the fetus (9). Death of the embryo during the early stages of gestation (Weeks 1-4) results in resorption of the conceptus by the maternal system. Death of the embryo from Week 4 through Week 8 results in heavy bleeding, which is frequently undetected as a fetal death by the mother. Many pregnancies are not diagnosed until the woman seeks medical attention, 8-10 weeks after conception. Death of the fetus during Weeks 8 through 36 results in expulsion of the uterine contents, referred to as a spontaneous abortion. The rate of spontaneous abortions (percentage) in a population is determined by multiplying the number of spontaneous abortions by a factor of 100 and dividing by the sum of the number of spontaneous abortions and the number of births (12). The estimated rate of abortion reported in the literature is variable. One source estimates that 75-78% of all conceptions are resorbed or aborted (13); another estimates the rate as being 30-50% (14). Another study indicates the incidence of spontaneous abortion of recognized pregnancies to be 15-20 % (3). The reason for these discrepancies is the inclusion or exclusion of the embryonic period in the estimation of spontaneous abortion. Until recently, it has been impossible to detect pregnancy in its early stages
and thus the incidence of pregnancy wastage has been frequently underestimated because of inadequacies in the methods of analysis of pregnancy loss (14). With the development of new and innovative immunoassays able to detect the presence of urinary /3 chorionic gonadotrophin-a hormone produced by the placenta-nine days after conception (1.51, it should be possible in the future to more accurately determine the rate of spontaneous abortion in the first two months of pregnancy. The outcome of pregnancy following exposure to a chemical substance depends upon the length of exposure, the stage of fetal development at the time of exposure, the magnitude of exposure, and the nature of the chemical substance. Not all exposures result in fetal death and spontaneous abortion. Many exposures result in teratogenesis, defined as the production of any significant change in structure or function of the fetus that can be detected in the postnatal period (1). The possible consequences of teratogenesis are death of the fetus (spontaneous abortion), congenital malformation, growth retardation, and functional disorder of an organ system (16). The final manifestation of fetal exposure to an adverse influence will depend primarily on the susceptibility of the stage of development at the time of the exposure and the magnitude of the exposure (9). The total dose of chemical that reaches the embryo or fetus is dependent on several factors: the magnitude of the dose, the physical form of the agent, the route of exposure, the rate of absorption by the maternal system, and the effectiveness of maternal homeostatic devices that function to protect the fetus. The primary function of the maternal homeostatic devices is to reduce the blood concentration of the toxicant so that the number of molecules free to cross the placenta is minimized. This is accomplished by detoxification, excretion, or storage of the toxicant by the maternal system. The embryo or fetus is thought to have a threshold dose for most toxic substances. Below this dose, no effect occurs; above this dose, permanent changes in the fetus may be induced (9). Although the embryo or fetus may receive only a small fraction of the dose the mother was exposed to, it is frequently sufficient to produce an embryotoxic or teratogenic response. Because of the extraordinary sensitivity of the conceptus, it is possible for embryotoxicity to occur in the absence of maternal toxicity following exposure to a teratogen (I 7). Susceptibility to teratogens depends on two major factors: the genotype of the fetus and the stage of development
at the time of exposure. Genotype is the term applied to the entire, unique genetic makeup of an individual. Variations in genotype among individuals result in differences in their biochemical and morphological makeup. It is these differences that acwunt for the variability in response of different individuals to a particular toxicant (9). Such differences could affect the absorption, transformation, and e l i i a tion of the toxicant by the m a t e d system; the rate of absorption of the toxicant across the placental membrane; and the type of interaction b e tween the toxicant and the conceptus tissues (9).
Efiefts of environmentsl chemicals Reproductive failure can occur as a result of the exposure of either male or female to toxic substances, as shown in Figure 1. Adverse reproductive outcomes in males result from direct chemical effects on the testes or though interference with the steroidigenic function of the testes. A number of chemical substances capable of damaging the anatomical structure and function of the testes have been identified. nKo such compounds are dibromochloroprcpne (DBCP), a potent alkylating agent, and lead, which produces malformed sperm (teratospermia) in the ejaculate (18). F‘renatal exposure of males to diethylstilbestrol (DES), a drug used after World War II to prevent miscarriage, resulted in abnormalities in the male reproductive tract manifested as epididymal cysts, seminal vesicle enlargement, prostatic inflammation, undescended testes, and the production of abnormal sperm (19). Use of this drug clearly demonstrates the vulnerability of the developing reproductive tract to toxicants. By interfering with the male hormones necessary for development and differentiation of the male reproductive tract, prenatal DES exposure is responsible for morphological abnormalities that are expressed in later life. F’esticides such as chlordecone and 09DDT possess significant estrogenic activity, whereas chlordane decreases androgenic activity by increasing the metabolism rate of the active intermediate (IS). Adverse reproductive outcomes in females result from the direct effect of a chemical substance on the reproductive hact itself or through i n t e r f m c e with hormonal balance, which is normally maintained by extragonadal tissues such as the pituitary, hypothalamus, or brain. In the female, the development of ova, which are released during adult life, OCCUIS during fetal development; no ova are produced after birth. Therefore the female fetus is extremely vul-
nerable to prenatal exposure to toxicants. Any alteration in the development or formation of ova by toxicants during the prenatal period may result in decreased reproductive capacity as an adult. These effects may be manifested by reduced fertility or by carcinogenic effects on the fetal organs that may be manifested in later life (18). For example, prenatal exposure of the female fetus to DES has resulted in the development of a d e n m i n o m a , a malignant cancer of the reproductive tract, i i early adult life. AII& studies have shown that fetal ovaries are the target of certain environmental pollutants, particularly polycyclic aromatic hydrocarbons, which deplete the number of the cells that will develop into ova (IS). CLarses of reproductive toxicants Metals. Metals, including mercury, lead, arsenic, cadmium, aluminum, and lithium, have been shown to pose a risk to the developing embryo. Mercury is the metal for which the greatest documentation of teratology has been recorded. During the period 19541960, the population of Minimata Bay in Japan was exposed to methylmercury via the food chain. A chlor-alkali plant, using mercury as a catalyst, discharged its wastes into the bay. Subsequent methylation of the mercury by bacteria and uptake of the lipphylic methylmercury by fish resulted in bioconcentration of the compound into fat and muscle. Because fish is the staple source of protein for the inhabitants of the area, a number of people were exposed to the compound (20.21). Infants exposed in utero via the consump tion of contaminated fish by their moth-
ers demonstrated varying degrees of neurological symptoms resembling cerebral palsy (20,21). Similar effects of mercury on development also have been documented in the United States (22) and Iraq (23) following accidental inclusion of mercurous fungicides into grains. For nearly a century it has been known that lead has an adverse effect on the reproduction of women exposed, at work and on their offspring. Extensive studies in Sweden by Nordsaiim and w-workers have reported adverse reproductive effects in women living close to or working in a smelter discharging arsenic and lead. An increased spontaneous abortion rate in women living close to the smelter has been documented (24-26). Decreased birth weight of offspring not associated with cigarette smoking has been observed (26),as have deleterious effects among men working in the plant. An increased frequency of chromosomal aberrations of cultured lymphocytes of male workers exposed to arsenic from the smelter was reported. Lead was implicated in the abnormal development of sperm resulting in impaired morphology, motility, and an increased incidence of chromosomal aberrations (27).As these workers were exposed to arsenic, cadmium; and other smelter fumes concomitantly with lead, these effects may not be due to a particular metal alone but may result from an interaction between these toxic metals. Although exposure to inorganic lead has been found to result in teratospermia, a relationship between adverse reproductive outcomes and abnormally shaped sperm has not been established (28).
Environ. Sci.Technol.,Vol. 23,No. 10,1989
1188
Epidemiological studies relating aluminum and lithium to adverse effects have also been reported. In a study in south Wales, a significant positive correlation between central nervous system malformations and aluminum was noted (29). Lithium in the form of lithium carbonate, which is used therapeutically to treat psychiatric disorders, has been associated with teratogenicity in infants born to mothers exposed during the first trimester of pregnancy (30. 31). hugs. Exposure of nurses during the first trimester to antineoplastic drugs including cyclophosphamide, doxorubicin, and vincristine has resulted in an increased incidence of fetal loss (32). The independent effect of each drug could not be proven, as more than one drug was handled by each nurse during the exposure period. Drugs that have been established as embryotoxic and teratogenic in humans include thalidomide; androgens; aminop terin (2); warfarin and coumarin derivatives; trimethadione, an anticonvulsant; and isotretinoin, used in acne treatment (33). Gases. A number of epidemiological studies have shown that exposure of o p erating-room nurses and female anesthesiologists (34. 35) to anesthetic gases has resulted in an increased rate of spontaneous abortion as compared to unexposed control women. These gases have also been implicated as an etiological factor in birth defects in children born to mothers exposed during pregnancy (36, 37). In Finland, an increase in spontaneous abortion was also observed in women involved in the sterilization of hospital instruments using ethylene oxide ( 3 8 ) . Exposure of personnel ranged from 5-10 ppm with occasional acute concentrations of 250 ppm. Neither teratogenic effects nor decreased birth weights were observed. Although ethylene oxide ranks 26th in the volume of organic chemicals produced in the United States, less than 1 % is used in hospitals (5). Organic chemicals. More than 2000 people were exposed to cooking oil contaminated with polychlorinated biphenyls (PCB) and dibenzofurans (DBF) during a six-month period in Taiwan. Malformed children were still being born to mothers exposed more than six years previously to the cooking oil (39J.This phenomenon can possibly be related to bioaccumulation of ingested PCB, which has an estimated half-life in humans of seven years, for extended periods of time in fatty tissue. Abnormalities of the gingiva, skin. nails. teeth, and lungs were observed in a statistically larger percentage of offspring of exposed mothers. Neurologically, the exposed infants did not differ 1190 Envlron. ScI. Technol.. Vol. 23. No. 10, 1989
from the unexposed controls, but a delay in the performance of certain tasks requiring motor coordination was observed. The exposed children scored lower on three I.Q. tests that were evaluated. An earlier study in Japan with a smaller cohort of children whose mothers had been exposed to thermally degraded cooking oil showed similar abnormalities (40).At seven years of age, 70% of these children were apathetic and had IQs in the 70s. The effects of organic solvents and gases do not Seem to pose a great risk to reproduction in humans. No definitive association between vinyl chloride and spontaneous abortion or teratology has been reported, although an increase in the frequency of chromosomal aberrations in sperm has been noted in a num-
ber of studies (4/-43). An increased incidence of spontaneous abortions (44) and congenital malformations (45) have been reported in laboratory workers exposed to various solvents. In Sweden, a statistically higher rate of miscarriage was found in women working in a petrochemical plant but no differences were Observed in women living close to the plant (46). A similar nonsignificant increase in reproductive outcomes was noted in a study of women living close to a phenoxy herbicide plant (47). A number of studies of laboratory animals have demonstrated reproductive hazards associated with many organic solvents. Common solvents that have been found to be teratogenic or embryotoxic include benzene (48). xylene (49). cyclohexanone (50). propylene glycol (51). alkane sulfones (52). glycol ethers (53). acetamides, and formamides (54). Reduced sperm counts, testicular atrophy, and decreased fertility were documented in male employees in a plant "IfaCturing DBCP (55, 56). Similar reproductive effects were noted in occupationally exposed workers applying the nematocide (57).Subchronic exposure of laboratory animals to DBCP produced similar changes to those observed in humans (58). Pesticides. Adverse effects of pesticides on reproduction have been reported in occupationally exposed men and women. Impotence, miscarriage, chromosomal aberrations, and infertility are the major manifestations that have been recorded (59-6/), A significant increase in abortions among c o u p les in India who were occupationally exposed to nine pesticides including DDT, dieldrin, dichlorovos, parathion, and copper sulfate has been reported. An increase in lymphocyte chromosomal aberrations was also documented in these workers (62). Conflicting reports on the teratogenic potential of 2.4.5-T and its major contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) has appeared in the literature (5).An extensive study by the U S . Department of Defense on 0.5 million exposed Vietnamese reported no increased incidence of either miscarriages or teratogenic outcomes (63). but an American Association for the Advancement of Science report showed an increase in cleft palate and spina bifida in children in Viemam from 1964 to 1%8 (64).A study by LaPorte (65) substantiated the results of the AAAS, although a study by Kundstadter found no increased incidence reported in hospital records (66). In general, insecticides appear to have low embryotoxic potential in mammals, but a number of teratogenic
coumarin, doxorubicin, hycanthone, ketoconazole, methotrexate, and trypan blue demonstrated that the number of somites after exposure was as reliable an index of teratogenicity as were other morphological changes. The concentrations that elicited a teratogenic response ranged from 3 n g / d (actinomycin D) In vitro short-term tests to 150 pglml (trypan blue). From this Whole embryo systems. Mamma- study, the author was able to determine linn embryos. Early (preimplantation) that each compound elicited a specific mammalian embryos can be collected response resulting in a defect at a parthrough the blastocyst stage and cul- ticular embryonic site(+ The results of this study (86)were tured in vitro in serum-amended media (79,So). Exposure of the embryo can similar to the results obtained in in vivo be undertaken during culturing or by studies with actinomycin D @7), hyexposure of the mother before the em- canthone (88),and methotrexate (89). bryo is flushed from the reproductive Although coumarin was shown to cause tract. Morphological alteration or central nervous systems malformations growth retardation are the end points in this assay (86)and has been implimeasured. To date only a few com- cated in congenital defects of the cen91), pounds have been evaluated for repro- tral nervous system in humans (W, ductive toxicity using this assay, and it was negative in in vivo studies in rabmore studies will be necessary to vali- bits (92)and mice (93).In addition to morphological characteristics, various date this method. The original method involving post- biochemical parameters including proimplantation embryos was developed tein content, DNA content, uptake of by New and co-workers (81) and in- 3H thymidine and '*C- amino acids into volves the removal of 9- or 10-day-old DNA and proteins, respectively, can be Evaluation of compounds (equivalent to 4-12 somites) mice or rat measured. Three other vertebrate embryo asReproductive toxicity assays can be embryos from the uterus followed by in divided into two groups: those that de- vitro culturing in rat or human seflltn says that have not been as well docutect reprodudve toxicants and those for 1-48 hours (82-84).During the in- mented have also been described. that detect malformations occurring cubation period, test chemicals are Chick (94),fish (93,and frog (%) emduring development of the fetus. These added to the media in various concen- bryo cultures have been used to assess tests can be undertaken on intact mam- trations. Embryos are observed after the teratogenicity of a number of commals, on nonmammalian animals, or by the exposure period for heartbeat, yolk pounds that were positive and negative sac circulation, crown rump length, in the mammalian assays. In general in vitro assays. The following three levels of screen- somite numbers, closed or open neural ahout a 7&80% agreement with maming are used in the evaluation of a com- tubes, otic and optic vesicles, and de- malian assays was obtained. The use of in vitro assays has many pound for its effects on the developing velopment of limbs. A morphological scoring system for advantages as well as a number of infetus: Prescreening employs simple, inex- teratogenic changes in rats has been de- herent problems. Advantages are the pensive, short-term testing and is veloped by Brown and Fabro (85). A relatively low costs, the simplicity and most commonly performed in vitro study by Schmid (86)using actinomy- the rapidity with which results are obcin D, azathioprine, colchicine, tained, the many variations that can be (78); devised for the scientific protocol, the precise control over exposure to a specific developmenral stage of the embryo, the composition of the culture medium, the concentration of the test compound, and the duration of exposure. The major disadvantages include the isolation of the embryo from the mother-thus eliminating the modulation of the metabolic system of the mother-and the difficulty of assessing the long-term consequences of brief embryonic exposures used in embryo culture systems. Additional problems include the lack of standardiition of the methods and quantitation of normal and abnormal rat embryo development. The influence of chronic exposure cannot be analyzed by any of these methodologies, and the extrapolation of the results found in laboratory animals to humans is difficult. However, this technique does have potential for the future, as shown by the good correlation be-
effects have been recorded. Organochlorine insecticides including aldrin, dieldrin, and endrin have been shown to be teratogenic to the mouse and hamster (67).A few organophosphates are known to be teratogenic-cabaryl in the guinea pig (68)and dog (69),demeton (70)and parathion (71)in the mouse, and dichlorovos in the rabbit (72).Evidence that the fungicides manoseb, dinocap, and nitrofen are potent teratogens has been presented (73).DDE, a metabolite of DDT, is known to have detrimental effects on reproduction in birds by interfering with the hormonal activity of estrogen, which results in decreased deoosition of calcium in the egg shell. Cigarette smoke. Smokine durine preghncy has been associateiiwith rel d u d birth weight of infants (74)and an increased risk of spontaneous abortions (75,76). This effect, which has been coupled to growth retardation, has been found to more pronounced in blacks than in whites (77). Table 1 summarizes the effects of these classes of reproductive toxicants.
Animal testing involves the testing of chemical agents for teratogenicity in small laboratory animals; and Epidemiological studies in humans help determine an association between a chemical and its reproductive effect (I).
Environ. Scl. Technol., Val. 23, NO. 10, 1989 1191
tween the in vitro results and in vivo studies. Invertebrate embryos. The test system using the freshwater coelenterate Hydra attenwta was developed by Johnson and co-workers (97-99)and has been applied in the study of glycol ethers (100). Adult hydra polyps are exposed to log concentrations of the test substance (1 x 10-3 to I x iw ,,m), and the minimum concentration required to produce a toxic response is determined. The dose of a test compound that results in toxicity in the adult (A) is then compared to that which causes developmental toxicity (D).The lowest whole log concenmtion of the chemical causing disruption in the development of the embryo is determined by disassociating 700 to loo0 hydra into their component cells, forming an "artificial embryo." These preparations consist of two cell types. The major component consists of groups of cells that give rise to a new population of adult polyps in about 90 h. The second group consists of a small number of undifferentiated interstitial cells capable of rapid proliferation and subsequent differentiation into a typical adult hydra. These cells are then exposed to the test chemical to determine the developmental toxicty dose. An A:D ratio can then be calculated. More than 30 compounds that have been tested in this assay have a similar A:D ratio in in vivo studies (96, 97). Thus this assay seems extremely promising as a means of assessing reproductive toxicants. A system devised by Best and Morita (101) using the freshwater planarian Lhgesia dortocephala has been used to assess the effects of toxicants on reproduction and teratology. Planaria exhibit bilateral symmetry, are hermaphroditic, can reproduce sexually or asexually, and have a hue brain indicative of complex behavior. Planaria reproduce by fission in the transverse plane with regeneration of the anterior or posterior section. Regeneration involves the migration of undifferentiated stem cells or neoblasts. Because this process simulates embryogenesis, it is thought to be sensitive to teratogenic agents. W o a p proaches have been developed. Surgical fragments of planaria can be exposed to the test material during regeneration to assess the effect of the compound on the rapid multiplication of cells, or intact planaria can be exposed to the test material and examined for morphological or behavioral abnormalities. This assay has been shown to be less sensitive to three hown teratogens than the hydra assay (96). Organ culture. In the mouse embryo limb bud cell culture assay, 11- to 12day-old mice embryos are removed 1192 Environ. Sci. Technol.. Vol. 23,No. 10, 1989
from the dams and the limb bud dissected from the embryo under a stereomicroscope (102). The buds are incubated for several days in a nutrient medium in the presence or absence of the test chemical. The cultured limbs can then be scored qualitatively for the presence or absence of malformations and quantitatively for the amount of cartilage formed in the l i b by using toluidine-blue stain. A quantitative a p proach to the measurement of the formation of cartilage is the determination of the uptake of 'H-proline or "5s-SUIfate into condroitin sulfate, the major component of cartilage. A modification of this method developed by Guntakatta (103) depends on the disassociation of the limb bud into individual cells, which are then exposed to 'H-thymidine and 3sS-sulfate in culture. Tissue culture. Cell lines have been established from mouse ovarian tumor cells (104), human embryonic palatal mesenchyme cells (IOS), and human neuroblastoma cells (106). Recent studies have indicated that these cell lies may be promising for screening of p tentially teratogenic compounds (107). Cell attachment, growth inhibition, and inhibition or stimulation of differentiation are the parameters that are measured following exposure to nontoxic doses of the test chemical. Table 2 illustrates the end points for short-term bioassays.
In vivo Imm ' "assays Mammalian three-generation studies. In these reproductive toxicity tests, three generations of animals are exposed to the toxic agent in question at various dose levels. The test substance is administered to both pa" throughout the study with the highest dose level beimg toxic to the test animal and the lowest being innocuous (10s).Following exposure of the animals during gestation, various developmental param-
eters are measured in the pups. offspring of the mating are then continuously exposed to the compound, weaned, allowed to mature, and mated amongst themselves. This procedure is followed for three succeeding generations. The advantage of this technique is the determination of the effect during in utero exposure and subsequent reproductive performance. A complication that does arise is the selection of the animals for mating for the succeeding generations. Fetal survival, litter sue, and weight of the pups may affect the selection process. At the end of each reproductive cycle, all the pups are examined for physical abnormalities. The number of viable, stillborn, and dead pups in each litter is recorded, and the number of survivors on Days 1, 4, 7, 14, and 21 after birth are recorded. Body weight of the pups is recorded at weaning. W o indexes, the gestation index and the viability index, can be calculated. The gestation index-a measure of the percentage of pregnancies resulting in live litters-is not a meaningfd measure as it does not take into account stillborn pups unless the whole litter succumbs. The viability index W.1.) refers to the percentage of pups that SUIvives for a specific period. This index is extremely important because early survival of the pups may be dependent on the excretion of a toxic compound into the milk of the lactating dam and subsequent ingestion of the compound by the pups. In these three-generation studies, a number of parameters cannot be measured. For example, the actual litter size is seldom known because the mothers are not sacrificed after parturition. No information is collected on sperm motility or viabfity. Groups of untreated control animals must be included in these studies so that comparisons with the exposed group can be
made. The reproductive organs of only those pups not being used for breeding can be evaluated hy weighing or by examination for histological changes. In most studies, only the male reproductive organ (the testicle) is studied,but in some situations, the ovary may be examined. Mammalian single-generation studies. In single-generation reprcductive studies, both male and female animals receive the compound for 60 days, prior to conception and then through gestation. The effect on libido, estrus cycle, and reproductive capability; toxicity of the compound to both mother and fetus; postnatal pup development; and adequacy of lactation in the mother can be evaluated. In these conventional reprcduction studies, the animals are mated once or twice following exposure to the test chemical. A new protocol, the Fertility Assessment by Continuous Breeding, which has been developed by the National Toxiw logical Program, allows multiple matings for each pair of animals during the 14 weeks of cohabitation and exposure to the toxicant. In these studies, females generally deliver four to five litters during the exposure period, thus, a meaningful fertility index can be calculated (109). Another approach and one that is used more frequently is the administration of the toxicant to the mother following mating. Several dose levels of the chemical are administered to different groups of females. The highest dose is based on the maximum tolerated dose (MTD) and frequently results in maternal toxicity. The lower doses are set at specific fractions of the MTD. To prevent cannibalism of the pups, the fetuses are usually delivered by cesarean section prior to their expected date of birth. The newborn pups are examined for external and internal abnormalities. A major problem associated with this approach is that teratogenic changes that occur late in life will not be de-
tected. Extrapolation from animal studies In extrapolating from animal studies to humans, a number of factors must be taken into 'consideration. The appropriate laboratory animal must be used, the number of animals on each treatment must allow for statistical evaluation, and the dose levels that are used should be relevant to human exposure. Far more animal than human teratogens have been identified because of the ways we are able to test animals. Experimental animals have a short reproductive cycle and multiple offspring. Therefore, many fetuses can be examined in a short time and at a relatively low cost. An equivalent human study
would cost millions of dollars and require a number of years. The dosage administered to animals can be adjusted so that it is many times greater than would be expected in a normal human exposure situation. The experiment may also be designed to correlate e x p sure of the embryo or fetus with its most sensitive time in development (1). All of these factors improve the possibility of identifying a chemical compound as a teratogen in animals. Animal Studies are, however, not without fault. As discussed earlier, the variation in genotype among different individuals influences their response to a teratogen and hence their susceptibility to a particular toxicant. Even greater variations in genotype exist be tween different species. Therefore, a chemical proven to be nonteratogenic in a certain laboratory species may well be teratogenic in a different laboratory species or even in humans. A classic example of such a situation is thalide mide. The laboratory animals most widely used in teratogenicity testing (mouse, rat, and rabbit) are 5-500 times less sensitive than humans to this sedative (5), which is one of the most potent human teratogens !mown. Another drawback of animal studies is that chemicals may combine and permutate after they enter the environment. Even if a chemical is found to be nontoxic in animal studies, the safety of the chemical cannot be assured. The possible potentiation of a chemical's effect because of its interaction with other chemical agents or other toxic factors once it enters the environment cannot be dismissed.
Epidemiolngid studies in humans Epidemiological studies of reproductive loss are usually undertaken using retrospective approaches including both personal interviews and medical records. If only personal interview data are used, many confounding factors can intefiere with an accurate assessment of a spontaneous abortion. Patients are frequently confused about delayed menstruation and have difficulty recalling events. The confirmation of a mis-
etic transmission
camage through medical documentation is thus necessary to obtain relevant data (108,110). Selection bias of exposed and nonexposed women must also be avoided in these studies. A number of factors not involving exposure have also been associated with spontaneous abortion. These include maternal age, previous spontaneous abortions, cervical incompetence, maternal fever (69, 75, ill), diet, health status, and weight (112). In an epidemiological study, timing and duration of exposure is crucial. Exposure of the father for four months prior to conception must be monitored because this is the interval required for a complete cycle of spermatogenesis. Possible fetal exposure during the first trimester is the most critical although exposure during the second and third trimester can result in adverse effects. Quantification of exposure is frequently the most difficult parameter to measure. An innovative approach that has recently been applied in assessing exposure is the use of biological markersnoninvasive measures of dose-that can then be related to potential adverse reproductive outcomes (113). Measurement of mutagens, thioethers, of Dglucaric acid in the urine are frequently used. Use of these markers requires information regarding the normal values of the marker, inter- and intravariability among individuals, and possible confounding effects. In order to assess the effect of exposure on reprcductive outcomes, two sequential epidemiological techniques, descriptive and analytical, have been used (114, 115). In descriptive studies, information regarding the distribution and frequency of an outcome is o b tained. The impetus for these studies is usually based on an increased number of case reports on spontaneous abortion or teratology in a particular geogmphical area.The second stage is an analytical study that is designed to test a hypothesis or generate a new hypothesis regarding an association between exposure and reproductive outcomes. Either case control or cohort studies are used. In the case control study a retrospective assessment of the exposure factors of the cases and controls is performed. In a cohort study, both individuals exposed to a particular factor and nonexposed individuals are followed over time, using either retrospective or prospective techniques to observe the outcome of interest. A risk for the particular outcome of both groups can then be quantified. Conclusion It is evident that there are many possible adverse health effects that may result from exposure to the numerous Environ. Sci. Technoi.. Vol. 23. No. 10, 1989 1193
chemicals present in the environment. Exposure of both male and female organisms to chemicals during the reproductive cycle can have deleterious effects on the development of the fetus. These effects are manifested as fetal death, malformation, retarded growth, and organ dysfunction. Our knowledge of the adverse health effects that are produced as the result of a complex network of factors is limited. This is the key point-there is much that is unknown with regard to the chemicals to which humans are exposed and how these chemicals interact with biological systems. To aid in assessing whether a compound is a teratogen, a number of short-term in vitro assays have been developed. At present scientists are validating these assays. There is much progress to be made in defining the incidence of such effects, the magnitude of dose in specific exposure situations, and the possibility of interactions between causes. The progress that has been made so far has indicated the importance of limiting exposure to chemicals as much as is practical and feasible. Future research efforts should be directed toward establishing the mechanisms of toxicological effects and developing more accurate means of identifying hazardous compounds.
Acknowledgments The author wishes to acknowledge Dr. June Sutherlin, D.V.M., for her contribution and suggestions, and Mrs. Ann Soike for typing the manuscript. The support of the Geismar Area Industrial Technical Group is gratefully acknowledged.
References (1) Shepard, T. H. Ann. N . Y Acad. of Sci. 1986, 477, 105-15. (2) Wilson, J. G. In Handbook of Teratology; Wilson, J. G . ; Fraser, E C., Eds.; Plenum: New York, 1977; Vol. I, pp. 357-85. (3) Byrn. F. W. ; Gibson, M. Clin. Obstet. Gvnecol. 1986.29. 925-40. (4) Sfellman, J. M. Ann. N. I: Acad. of Sci. 1986, 477, 116-21. ( 5 ) Schardein, J. L., Keller, K. A . CRC Crit. Rev. Toxicol. 1989,19, 251-339. (6) Butcher. R. L.: Page, R. D. Environ. Health Perspect. 19f1, 38, 35-37 (7) Pernoll, M. L. Clin. Obster. Gynecol. 1986, 29, 953-58. (8) Dorland S Nlustrated Medical Dictionary, 25th ed.; Friel, J. P., Ed.; W. B. Saunders: Philadelphia, PA, 1974. (9) Wilson, J. G. In Handbook of Teratology; Wilson, J . G . ; Fraser, E C . , Eds.: Plenum: New York, 1977; Vol. I, pp. 4974. (10) Guyton, A. C . Textbook ofMedica1 Physiology, 6th ed.; W. B. Saunders: Philadelphia, PA, 1981; pp. 1022-37. (1 1) Langman, J. Medical Embryology, 2nd ed.: Williams and Wilkins: Baltimore, MD, 1969. (12) Taskinen, H.; Lindbohm, M. L . ; Hemminki. K. B,: J . lnd. Med. 1986, 43, 199-205. (13) Johnston, S . D.; Raksil, S . Veterinary Clinics of North America: Small Animal Pracrice; 1987, 17, 535-54. 1194
Environ. Sci. Technol., Vol. 23,No. 10, 1989
Carr, D. H. In Handbook of Teratology; Wilson, J. G . ; Fraser, E C . , Eds.; Plenum: New York, 1977: Vol. 111, pp. 200-2 13. Sweeney, A. M. et al. Am. J . Epidemiol. 1988, 127, 843-50. Beck, E In Developmental Toxicology; Kimmel, C . A , ; Buelke-Sano, J . , Eds.; Raven: New York, 1981; pp. 35-54. Wilson, J. G . Fed. Proc. 1977,36, 16981703. McLachlan, J.A. et al. In Developmental To.xicology; Kimmel, C . A , ; Buelke-Sam, I . Eds.; Raven: New York, 1981; pp, 213-32. Gill, W. B . ; Schumacher, G. F. B . ; Bibbo, M. J. Reproduct. Med. 1976, 16, 147-53. Matsumoto, H. G . ; Goyo, K.; Takevchi, T. J. Neuropathol. Exp. Neurol. 1965, 24, 563-74. Harada, Y. In Minimata Disease; Kumamota University study group of Minimata disease: Kumamota. Japan, 1968; pp, 73-91. Snyder, R. D. N . Engl. J . Med. 1971, 284, 1014-16. Amin-Zaki, L. et al. Pediatrics 1974, 54, 587-95. Nordstrom, S . ; Beckman, L.; Nordenson! I. Hereditas 1978, 88, 43-46. Nordstrom, S.; Beckman, L.; Nordenson, I. Hereditas 1978,88, 51-54. Nordstrom, S . ; Beckman, L.; Nordenson, I . Hereditas 1979, 90, 291-96. Nordenson, I . et al. Hereditas 1978, 88, 263-67. Lancranjan, I. et al. Arch. Environ. Health 1975,30, 396-401. Morton, M. S . ; Elwood, P. C. Teratology 1974,10, 318. Frankenburg, E R.; Lipinski, J. E N . Ena. J . Med. 1983.309. 311-12. Wehstein, M. M.:’Goldfield, M. Am. J . Psychiatry 1975, 132, 529-3 1 . Selevan, -S. G. et al. N. Eng. J . Med. 1985, 313, 1173-78. Khera. K. S. CRC Crit. Rev. Toxicol. 1987, 17, 345-75. Cohen, E. N . ; Bellville, J. W.; Brown, B. W. Anesthesiology 1971, 35, 343-47. Knill-Jones, R. P.; Newman, B. J . ; Spence, A . A. Lancet 1975,2, 807-9. Corbett. T. H. et al. Anesthesiology 1974, 41, 341-44. Pharoah, P.O.D. et al. Lancet 1977, I , 34-36. Hemminki, K. et al. B,: Med. J . 1982, 285, 1461-63. Rogan, W. J . et al. Science 1986, 241, 334-36. Rogan. W. J. Teratology 1982, 26, 25961. Purchase, I. E et al. Mutat. Res. 1978, 57. 325-34. Heath, C . W.; Dumont, C. R.; Gamble, J. Environ. Res. 1977, 14, 68-72. Fleig, I.; Thiess, A. M. J. Occup. Med. 1978,20, 557-61. Hemminki. K.: Franffila. E.; Vainio, H. Int. Arch. Occup. Environ. Health 1980, 45, 123-26. Holmberg, €? C. Lancet 1979,2, 177-79. Axelsson, G . ; Molin, I . Int. J . Epidemiol. 1988, 17, 363-69. Kallen, B.; Thorbert, G. Environ. Res. 1985,37, 313-19. Murray, E J. et al. Am. lnd. Hyg. Assoc. J . 1979, 40, 993-98. Kuchera, J . J. Pediatrics 1968, 72, 85759.
Weller, E . M.; Griggs, J. H. Teratology 1973, 7, 1-30. Gebhardt, D. 0. E. Teratology -. 1968, I , 153-62, Hemsworth, B. N. J . Reproduct. Fertil. 1968, 17, 325-34. Schuler, R. L. et al. Environ. Health Perspect. 1984,57, 141-46. Thiersch, J. B. J . Reproduct. Fertil. 1962, 4, 219-20.
(55) Whorton, M. D. et al. Lancet 1977, 2, 1259-61 ---_ (56) Biava, C. G.; Smukler, E. A , ; Whorton, M. D. Exp. Mol. Pathol. 1978, 29, 44858. (57) Lipschultz, L. I. et al. J . Urol. 1980, 12, 464-68. (58) Whorton, M. D.; Foliart, D. E. Mutat. Res. 1983,123, 13-30. (59) Whorton, M. D.; Watson, M.; Benson, W. W. Mutat. Res. 1973,21, 335-40. (60) Espir, M.L.E. et al. Bs Med. J . 1970, I , 423-25, (61) Kiraly, J.; Szentesi, I . ; Ruzicska, M. Arch. Environ. Contam. Toxicol. 1979,8, 309-19. (62) Rita, I?; Reddy, P. P.; Reddy, S . V. Environ. Res. 1987, 44, 1-5. (63) Cutting, R. T.; Phuoc, T. H.; Ballo, J.; Benenson, M. W.; Evans, C. H. Congenital Malformations, Hydatiform Moles and Stillbirths and the Republic of Vietnam 1960-1969; U.S. Government Printing Office: Washington, DC, 1970; No. 903-223. (64) Meselsohn M. S. et al. 1970 Annual Meeting of AAAS Herbicide Assessment Commission of the American Association for the Advancement of Science; background material, revised January 14, 1971. (65) LaPorte, J. R. Lancet 1977, 1, 1049-50. (66) Kundstadter, P. A Study of Herbicides and Birth Defects in the Republic of Vietnam: An Analysis of Hospital Records; National Academy of Sciences: Washington. DC, 1982. (67) Ottolenghi, A. D.; Haseman, J . K.: Suggs, E Teratology 1974, 9, 11-16. (68) Robens, J. E Toxicol. Appl. Pharmacol. 1969, 15, 152-63. (69) Smalley, H. E . ; Curtis, J. M.; Earl, E L. Toxicol. Appl. Pharm. 1968, 13, 392403. (70) Budreau, C . H.; Singh, R. €? Toxicol. Appl. Pharm. 1973,24, 324-32. (71) Tanimura, T.; Katsuya, T.; Nishimura, H . Arch. Environ. Health 1967, 15, 609-13. (72) Kimbrough, R. D.; Gaines, T. B. Arch. Environ. Health 1968, 16, 805-8. (73) Wang, G. M. Teratog. Carcinog. Mutagen. 1988, 8, 117-26. (74) Butler, N. R.; Goldstein, H.; Ross, E. M. BE Med. J . 1972,2, 127-30. (75) Himmelberger, D. V.; Brown, B. W.; Cohen, E. N. Am. J . Epidemiol. 1978, 108, 670-79. (76) Harlap, S . ; Shioni, P. Lancet 1980, 2, 173-76. (77) Lubs, M. L. E. Am. J . Obstet. Gynecol. 1973, 115, 66-76. (78) Faustman, E. M. Mutat. Res. 1988, 205, 355-84. (79) Spielmann, H.; Kruger, C . ; Vogel, R. Conceuts Toxicol. 1985.3. 22-28. (80) SpielGann, H. et al. Drug Res. 1986, 36, 219-23. (81) New, D.A.T. Biof. Rev. 1978, 53, 81122. (82) Fantel, A. G. Bratog. Carcznog. Mutagen. 1982,2, 231-42. (83) Kochar, D. M. Teratog. Carcinog. Mutagen. 1980, l , 63-74. (84) Van Maele-Fabry, G.; Picard, J . J. Teratology 1987, 36, 95-106. (85) Brown, N. A , ; Fabro, S. Teratology 1981,24, 65-78. (86) Schmid, B. E! In Concepts in Toxicology; Homburger, F; Goldberg, A. M., Eds.; S. Karger: Basel, Switzerland, 1985; Vol. 3 , pp. 46-57. (87) Tuchmann-Duplessis, H.; Mercier-Parot, L. In Ciba Foundation Symposium on Congenital Malformation; Wolstenholme C . , Ed.; Little Brown: Boston, 1958; pp. 115-28. (88) Moore, J. A. Nature 1972,239, 107-09. (89) Wilson, J. G. et al. Teratology 1979, 19, 71-80. (90) Warkany, J . Teratology 1976, 14, 20509.
The Basics of Technical
communicating
Aauatic Humic Influence on Fate and T " m t of Pollutantc
U
ntil now lnde i n f o " has been available on the study of humic sub stances. This d i m new valume offersa cohesiw compilabon of the latest re search in thls arebbringlng together theuretKal and experimental a p p m k as well as practical applicafions on this unique aspea of water purification. Offering an indepth look at a wide range of important tcpia. this valuable title men: e tlumwbatlcm md e n r i " 6 n W lmpxt of aquatic humk rubstaruer e their nxtianr in nrmd wmn md adhnnts 0 their 1M"mta on wmr t"fnt
Wfih 45 dewriptie chapters. this book details m r a g e of hazardous waste chemicals water solubility enhancement sorption. metal p i a Mn. and photochemlmy. In addition. speck types of treatment pmcesses are examinedd s o w n . oxldation. enzymatic d e t o M o n . a mgulabon. ion exchange, and membrane processes. This volume daws together important re Search and offers many compelling. prr " s for detailed studies. It will appeal to a variety of readers interested in ensuring environmental p r o t d o n in a technological Mclety.
I.H. Suffet. Editw. EnvlmmentalStudii Institute. Drexel Uniwrsty Patrick MacCarthy. €dltor. Color& %hod of Mines M o p e d hrm a m u m spaaaed LytkM sim of EmimmfntiChfmbhy ofme Chemral Seem ACS in Chamistry series ~ 0 . 2 1 9 844 pages 11988) Clothbound ISBN 0-8412.1428.X LC 88.58029 US 8 Canada $109.95 Exwrt $131.95
ACIWCSS
S. Shone is (01 ~,r\ro~iirreprofessoor of roxicolonv i n rhr ln.srirute for Environmemo/ Sriiies ur Louiriunu irate Universify in Baron Rouxe. She rweiwd her B.S. and Ph.D. degree.s from rhe Universify of Wirwarersrand in South Africa. Her major research inreresrr center oround rhe developmenr and use of shorr-rerm r e m IO assess rhe risk of exposure of humans ro environmenral mixrures. She is currenrly srudying rhe effecr of air toxics on human health in Louisiana.
American Chemical kclety Dlnnbunon me.Dept. 14 11SSSuteenthSt.NW Washington. Dc uy136
WWrmLREE
ommunications sMlls-they're a must for successful xienti-. Now you can increax your effectiveness and improve your communication sIuIIs with this new. easyto-use reference book. Wth i 8 chapters. this handy guide starts with the basics. such as eliminating wordiness and jargon in technical communications.using correct punctuation. and seleaing appropriate verbs From there. you'll learn h w to awmble your papers and gather data for bath m e n and oral reports. You'll learn c o w documenting prccedures. including footnoting and wndng a bibliqraphy. You'll cover the use of wsual aids and graphcs. abstract preparation. the use of computers and proofreading. A wide variety of practical applications is MY. ered in this volume. including labaratory and business repom.journa1 publications. grants and proposals, business correspondence. resumk. and memos The &sics of Technkal Communicating is a convenient. essential reference-one that you'll use time and time again! It complements The Al3Sfyie Guide and Wnting the laboratory Notebwx to give scientists a complete package for professional development in technical communication slcllls. by 8. Edward Cain. Rmhester lnstrlote of Technolcgy 2M) pages (1 988) ISBN 0-8412-1451-4 U: 88-3325 Clothbound US 8 Canada $29.95 Eqmrt $35.95 ISBN 0-8412-1452-2 K 88-3325 Paperbound US 8 Canada $19.95. Exprt SW.9S Order from: A" Cmrlul SaMy
C
Datnbudan Gffm Dept. 90 1155 Weenth St. NW Washington. D.C. 2w36
MCALLTOUREE
800-227-5558 800-227-5558 (in Washinqmn. 0 C 872 43631 and UP P u r credt rad
and use your credt cardl
Iron. SCI.Technol.. VOl. 23, No. 10, 1989 1195
