National Academies Press: OpenBook

Drinking Water and Health,: Volume 6 (1986)

Chapter: 3. Reproductive Toxicology

« Previous: 2. Developmental Effects of Chemical Contaminants
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 35
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 36
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 37
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 38
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 39
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 40
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 41
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 42
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 43
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 44
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 45
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 46
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 47
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 48
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 49
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 50
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 51
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 52
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 53
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 54
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 55
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 56
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 57
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 58
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 59
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 60
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 61
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 62
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 63
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 64
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 65
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 66
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 67
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 68
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 69
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 70
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 71
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 72
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 73
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 74
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 75
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 76
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 77
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 78
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 79
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 80
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 81
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 82
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 83
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 84
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 85
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 86
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 87
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 88
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 89
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 90
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 91
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 92
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 93
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 94
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 95
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 96
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 97
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 98
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 99
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 100
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 101
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 102
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 103
Suggested Citation:"3. Reproductive Toxicology." National Research Council. 1986. Drinking Water and Health,: Volume 6. Washington, DC: The National Academies Press. doi: 10.17226/921.
×
Page 104

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Reproductive Toxicology Reproductive dysfunction is broadly defined in this chapter to include all effects resulting from paternal or maternal exposure that interfere with the conception, development, birth, and normal growth of offspring. Chap- ter 3 is a broad discussion of the end points included under the heading of reproductive toxicology with the exception of embryo (fetal) death, growth retardation, and malformations, which are covered in Chapter 2. The relationship between exposure and reproductive dysfunction is highly complex because exposure of the mother, the father, or both may influence reproductive outcome. In addition, these exposures may have occurred at some time in the past, immediately before conception, or during gestation. For some specific dysfunctions, the relevant period of exposure is limited; for others, it is not. For example, chromosome abnormalities detected in the embryo can arise from lesions in the germ cells of either parent before conception or at fertilization, or from direct exposure of embryonic tissues during gestation. Major malformations, however, usually occur when ex- posure occurs during a discrete period of pregnancy, extending from the third to the eighth week of human development. Many cases of infertility can probably be attributed to postfertilization reproductive failure, i.e., repeated early spontaneous abortion. Such peri- implantation embryonic mortality may not be clinically apparent, since abortion could occur before the expected time of menstruation. Approx- imately 15% of clinically recognized pregnancies terminate in spontaneous abortion. Embryonic death rates in humans may be substantially higher. In recent studies subclinical spontaneous abortion rates were found to be 21% (Chartier et al., 1979) and 34% (Overstreet, 19841. Nonetheless, 35

36 DRINKING WATER AND HEALTH methods to assess early, subclinical spontaneous abortions are currently inadequate. Although there are extensive data on reproductive performance in human populations, most have been collected for routine surveillance not for environmental monitoring. Even though the effects of specific agents on reproductive function cannot be discerned from such data, useful infor- mation on trends and patterns in the frequency of various reproductive outcomes can be derived. For example, an estimated 11 million married couples in the United States are infertile (i.e., not capable of having children); 3 million of these couples have at least one partner who is noncontraceptively sterile (Mosher, 19851. Although early spontaneous abortions often go unreported, especially among pregnancies of less than 20 weeks duration, they are estimated to result in the termination of 15% of all pregnancies (Warburton and Fraser, 19641. This is generally regarded as an underestimate of the true rate insofar as most spontaneous abortions occur early in gestation, often before the mother is clinically recognized as pregnant. Approximately 7% of all babies are born prematurely (before the 37th week of gestation). Of those born at full term, an estimated 7% have low birth weights (2.5 kg or less) (Niswander and Gordon, 19721. Of the approximately 4 million infants born alive in the United States each year, 10.5 per 1,000 die within the first year (NCHS, 1985), and 2% to 3% of the 4 million infants have major congenital malformations that are recognized within that year (Edmonds et al., 19811. When defects that become apparent later in life are included, the frequency of major and minor malformations increases to about 16% (Chung and Myriantho- poulos, 19751. In very few cases has it been possible to separate a specific chemical exposure's impact on human reproduction from the background rate of spontaneous genetic defects or from other causes, such as radiation, in- fection, nutritional deficiencies, or maternal metabolic imbalance. There are also ethical limitations to conducting human studies, especially those concerning reproductive function. Consequently, the bulb of the infor- mation on specific exposures reported to affect reproductive function is derived from animal studies. The standard toxicological testing procedures for acute, subacute, and chronic exposures are not appropriate for detecting reproductive effects either in humans or in animals. Therefore, a separate series of tests has been developed to monitor reproductive function. These tests can provide both qualitative and quantitative analyses of reproductive toxicants in animals. This chapter contains brief descriptions of the biological development and function of the male and female reproductive systems. Stages partic- ularly susceptible to chemical insult are emphasized. This material is

Reproductive Toxicology 37 presented to provide a context in which animal data can be applied to humans in estimating the risk of reproductive toxicity. Toxicity to the embryo, fetus, or placenta, resulting in spontaneous abortion, teratogenicity, or other reproductive anomalies, has long been of concern. Other areas not as extensively studied are toxicities affecting the male and female reproductive systems, resulting in sexual dysfunction and infertility. Direct damage to germ cells, neuroendocrine imbalances, and alterations in accessory reproductive organs, which can be involved in these toxicological processes, are described in the following sections as they have been elucidated in animal studies. Consideration is also given to the use of these data to predict reproductive risk to humans. SUSCEPTIBILITY OF THE NONPREGNANT FEMALE TO REPRODUCTIVE IMPAIRMENT Maturation of the Female Reproductive System The development of the female genital tract and subsequent attainment of fertility are processes susceptible to disruption by chemical agents. Reduced fertility in offspring is one of the most sensitive indicators of prenatal exposure to reproductive toxicants. The female fetus is particu- larly vulnerable to germ cell toxicity, since the development of the oocyte occurs prenatally and the maximum number of oocytes available for sub- sequent ovulation is present at the time of birth. Damage to oocytes during the perinatal period may result in decreased reproductive capacity that will not be evident until sexual maturity is reached. Early in embryonic development, the progenitors of the germ cells, called primordial germ cells, are segregated from somatic cells. At 3 weeks of human development, these germ cells are first detectable in the yoLk sac. Thereafter, they undergo mitotic divisions and migrate to the uro- genital ridge where they populate the so-called indifferent gonad. Pri- mordial germ cells then differentiate into oogonia. The oogonial stage is characterized by active mitotic divisions; the daughter cells do not separate, but remain attached to each other by interconnecting cytoplasmic bridges. In the human fetal ovary, approximately 1,700 germ cells migrate to the gonads. By 2 months of gestation, the number of germ cells increases to about 6 x 105. Mitotic activity peaks by the fifth month at approximately 7 x 106 cells. Oogonia first begin to enter meiosis at the third month, and by the end of the fifth month, all the oogonia have entered early prophase I of meiosis and are called primary oocytes (Gondos, 19781. The timing of gonadal sex differentiation and of ovarian germ cell development in various mammalian species is presented in Table 3-1.

38 DRINKING WATER AND H"LTH TABLE 3-1 Ontogeny of Ovanan Germ Cell Development in Mammalsa Events in Germ Cell Development in Days of Gestation (or Postnatal Age) Length Completion Test of Gonadal Sex Initiation of Arrest of Animal Gestation Differentiation of Meiosis Oogenesisb Meiosis Mouse 19 12 13 16 (5) Rat 21 13-14 17 19 (5) Hamster 16 11-12 (1) (5) (9) Rabbit 31 15-16 (1) (10) (21) Rhesus monkey 165 38 56 165 Newborn Human 270 40-42 84 150 Newborn aAdapted from Gondos, 1978. bCompletion of oogenesis refers to the time when all oogonia have been transformed to primary oocytes. Replicative DNA synthesis occurs during the final interphase before the oogonia enter meiosis. The primary oocyte in prophase I of meiosis thus contains two sets of chromosomes; that is, it is diploid (2N) but contains four strands of DNA. The process of meiosis consists of two cell divisions. First, the number of chromosomes is halved, resulting in the formation of secondary oocytes, each containing one chromosome (IN), i.e., haploid, and two strands of DNA. In the second meiotic division, the chromosome number remains the same but the amount of DNA is halved. Thus, the ovum contains one chromosome and one strand of DNA. Figure 3-1 depicts the process of oocyte maturation in the fetus and adult. Each meiotic division has four stages: prophase, metaphase, anaphase, and telophase. The first meiotic division is initiated late in fetal life and progresses into early prophase during the fetal or neonatal period. By 8 weeks after birth, human oocytes have entered a resting phase of oocyte maturation, where meiosis remains blocked until the beginning of puberty (Biggers, 19801. Given the long duration of prophase I, this stage has been subdivided into five subphases: leptotene, zygotene, pachytene, dip- lotene, and dictyate (resting phase). Each substage is characterized by cytogenetic criteria of chromosome configuration. Extensive physiological degeneration of germ cells occurs during the oogonial and primary oocyte stages of development. In humans, an es- timated 70% of the germ cells present in a 5-month-old fetus are lost before birth (Biggers, 19801. Three distinct waves of degeneration occur in the human ovary, affecting oogonia in mitosis or in the. final interphase, oocytes in the pachytene stage, and oocytes in the diplotene stage of prophase. Oogonia connected by cytoplasmic bridges undergo atresia in

FIRST MEIOTIC DIVISION PROPHASE RESUMPTION OF MEIOSIS 1 LE PTOTE N E 7 DIAKINESIS 2 ZYGOTENE I ~ ~ UJ 8 METAPHASE 3 PACHYTENE z 4 DIPLOTENE 5 D I CTYATE STAGE 6 9 ANAPHASE 1 0 TE LOPHASE Reproductive Toxicology 39 SECOND ME IOTIC D IV ISION 1 1 METAPHASE OVULATION 1 2 ANAPHASE < o C] 1 3 TELOPHASE 14 PRONUCLEAR EGG . , z o - ~r N J - LO IL FIGURE 3-1 Oocyte maturation. Prophase of the first meiotic division (1-4) occurs during fetal life. In the zygotene stage, homologous chromosomes pair; in the pachytene stage, they form bivalent chromosomes. Genetic material is interchanged by a crossover process. At the diplotene stage, the chromosomes remain united at the points of interchange, the chiasmata. The meiotic process is arrested at the dictyate stage. When meiosis is resumed, the first division is completed (7-11). Ovulation occurs at metaphase of the second division (11), and maturation of the oocyte occurs in the oviduct (12-14) following sperm penetration. Adapted from Tsafriri, 1978. synchrony, which accounts for the majority of germ cell loss. After com- pletion of the meiotic prophase, groups of oocytes no longer appear to undergo atresia simultaneously, but individual oocytes may degenerate at all stages of development. It is not understood why some oocytes degen- erate while others mature. At some point during oogonial proliferation, all the oocytes within a syncytial mass will recruit granulosa cells from the surrounding ovarian

40 DRINKING WATER AND HEATH stroma and enter meiosis. The mechanism of this process, termed follic- ulogenesis or the formation of follicle complexes, is unknown. Once meiosis is initiated, germ cells lost to physiological atresia cannot be replaced. The maximum number of germ cells potentially available for ovulation in the offspring is fixed in the fetal period when oogonia mature into primary oocytes; the number continues to decrease due to physiolog- ical atresia and ovulation (Hertig and Barton, 1973~. During the prepubertal and reproductive periods, the majority of germ cells remain as primary oocytes enclosed within unilamellar follicles. These resting follicles comprise the pool from which a select number of oocytes are recruited for further maturation to preovulatory or graaf- ian follicles. In those follicles selected for maturation, a zone pellucida forms and separates the oocyte from the follicle cells. Thereafter, the follicular cell layer increases in size and the oocyte undergoes tremen- dous growth. Once a follicle embarks on this maturation process, it either reaches a preovulatory stage or it undergoes atresia (Tsafriri, 1978). At puberty, release of gonadotropins, particularly luteinizing hormone (LH) and to a lesser extent follicle-stimulating hormone (FSH), initiates the resumption of meiosis (see Figure 3-21. Following the rise in gonad- otropin levels, the primary oocytes in preovulatory follicles progress Trough the rest of the first meiotic division and form secondary oocytes that are blocked in metaphase of the second-division. The first polar body is extruded; this body contains half the chromosomes (IN) present in the primary oocytes (2N). As the time of ovulation nears, the follicle becomes more vascular and swells out from the ovarian surface. It is macroscop- ically visible as a blisterlike protuberance. The secondary oocyte is re- leased at metaphase of the second meiotic division, and it stays in this stage pending fertilization. At fertilization, the second meiotic division is completed, the second polar body is extruded, and the female pronucleus is formed. The male and female pronuclei combine to reestablish the diploid state (Espey, 19781. In the absence of fertilization, the secondary oocyte degenerates. When the menstrual cycle is established in humans, ovulation occurs on the average of every 28 days, during which time those follicles recruited from the resting follicle pool (graafian follicles) are stimulated to ovulate by elevation in gonadotropin levels. This process continues throughout the reproductive life until the population of primordial follicles is depleted or menopause occurs. Toxic effects on oocytes and effects on oogenesis are discussed in the following section. Germ cell mutagenesis is the subject of a separate section toward the end of this chapter.

Reproductive Toxicology 41 Pituitary Gonadotropins, Mi U/ml 5 LU an: cn Gonadal ~ Steroid > Estrogen, pg/ml 200 c) 100 400 o 99 o Basa1 Body Temperature 98 97 60 _ j — FSH 20 15 _ DEVELOPING - |~-- CORPORA- - | FOLLICLE is, LUTEA 10 Progesterone, ng/ml 5 O 1,, 1 1 1 1 1 1 1 1 1 1 ~ 1 1, 1 1 1 ~ 1 1 1 1 1 1 14 CYCLE DAYS ~FOLLICULAR —|- LUTEAL PHASE PHASE OVU LATION FIGURE 3-2 Endocrinology of the menstrual cycle in humans. From Haney, 1985, with per- . . mlsslon. Oocyte Toxicity The ovary, as a repository of oocytes and as a source of steroid hormones that control the functional development of reproductive organs, plays a major role in fertility and initiation of pregnancy. As indicated in the preceding section, when folliculogenesis is complete in the female during the perinatal period, oogonial cells no longer persist. The ovary cannot replace oocytes destroyed by toxicants. Complete destruction of oocytes prepubertally will result in primary amenorrhea and failure of pubertal onset. Complete oocyte destruction after puberty will produce premature menopause (Mattison, 19831. Follicle growth can be separated into two phases: gonadotropin inde- pendent and gonadotropin dependent. Recruitment of follicles from the resting pool and the initial phase of follicle growth to the preantral stage is gonadotropin independent and may be controlled by an intraovarian

42 DRINKING WATER AND HEALTH regulatory mechanism. Further growth and development to the preovu- latory stage requires the support of gonadotropins. Follicle growth is initiated at all ages, but in the absence of gonadotropins, growing follicles undergo atresia (Mattison, 1985~. If the dominant follicle is destroyed, fertility will be immediately in- terrupted. Follicles in the growing pool can repopulate the preovulatory pool, followed by a resumption in fertility. If a toxicant destroys growing, gonadotropin-independent follicles, but spares preovulatory follicles, the delay in the onset of infertility will be proportional to the time required for follicles to reach the preovulatory stage. Destruction of resting follicles has the greatest delayed effect on fertility, and the results will not be evident until the end of the reproductive life. Partial destruction of the resting follicle pool is manifested as premature onset of menopause. Men- opause generally occurs between 45 and 55 years of age; when it occurs before age 35, it is usually regarded as premature (Mattison, 19851. In a mathematical model of functional ovarian life span, Mattison ( 1985) has estimated that menopause occurs when there are fewer than 3,500 oocytes per ovary. Calculations based on this model indicate that the age of menopause is weakly dependent on the number of oocytes at birth. When 75%, 50%, or 25% of the normal complement of oocytes are present at birth, menopause is estimated to occur at 47, 44, or 37 years, respec- tively. Mattison reported, however, that varying the normal rate of atresia, or oocyte half-life (9.2 years), had a strong influence on age at menopause. When oocyte half-life was 75%, 50%, or 25% of the normal rate, the age at menopause was 38, 25, or 12 years, respectively. The results of this model are consistent with data on humans suggesting that most forms of premature ovarian failure, both genetically and xe- nobiotically determined, are due to an increased rate of atresia. Surgical procedures such as unilateral oophorectomy or bilateral wedge resection that decrease resting oocyte number without altering the rate of atresia do not appear to influence the age of menopause (Mattison, 19851. Effects of Radiation on Oogenesis In rodent species, female germ cells are extraordinarily sensitive to killing by exposure to ionizing radiation, especially during neonatal life. Primordial, or resting, follicles in juvenile mice have an LDso of only 6 reds (Dobson and Felton, 1983), whereas typical LDsoS for most other cell types range from 100 to 300 reds. The entire primordial follicle pool in female squirrel monkeys is destroyed by prenatal exposure to only 0.7 red/day throughout pregnancy. Histopathological examination of other tissues failed to yield evidence of cytotoxic effects at any other site (Dob- son and Felton, 19831. High oocyte radiosensitivity has been demonstrated

Reproductive Toxicology 43 in relatively few species, most notably in the neonatal mouse (Dobson and Cooper, 1974), the prenatal pig (Erickson, 1978), and the prenatal squirrel monkey (Dobson et al., 1978~. In Swiss-Webster mice, oocyte radiosensitivity appears shortly after birth, increases rapidly to peak sen- sitivity from days 5 to 17 of life, and decreases moderately to adult levels (Dobson and Felton, 19831. The rat displays a similar pattern but is considerably less radiosensitive (Mandl and Beaumont, 19641. In contrast, maturing oocytes in the guinea pig are more radiosensitive than primordial oocytes (Oakberg and Clark, 19641. The magnitude of prenatal germ cell loss in squirrel monkeys has led to examinations of loss in other nonhuman primates. In one such study, exposure of rhesus and bonnet monkeys to radiation during pregnancy failed to yield evidence of oocyte radiosensitivity (Dobson and Felton, 19831. In another, Baker (1978) found that oocytes in humans are resistant to radioactivity, reporting that LDsoS from x-ray exposure have reached 400 reds. X-ray exposures of the human ovary have most often been examined at prepubertal and adult stages, however, and a critical period during late fetal to early neonatal life would most likely have been missed. In adult human females, the growing follicles appear to be most sensitive to ionizing radiation, partly because of the rapid rate of granu-losa cell proliferation. The effects of ionizing radiation on the ovaries of women of reproductive age have been tabulated by Ash (19801. Exposure to less than 60 reds had no deleterious effects at any age. At 150 reds, women over 40 were at risk of becoming sterile. From 250 to see reds, women under 40 had temporary amenorrhea and 60% of the women became permanently sterile. All women over 40 had become permanently sterile at this level of radiation. With the onset of preovulatory oocyte maturation and resumption of meiosis after the dictyate stage (see Figure 3-1), susceptibility to the lethal effects of radiation decreases but sensitivity to heritable genetic damage increases. Preovulatory oocytes in multilayered follicles are relatively resistant to radiation-induced death, but they are sensitive to induction of both recessive and dominant mutations. In irradiated preovulatory oocytes, the incidence of dominant lethal mutations is highest at the first metaphase, slightly less at the second, and low at other stages (Baker, 19781. Effects of Xenobiotic Compounds on Oogenesis POLYCYCLIC AROMATIC HYDROCARBONS (PAHS) These compounds have been demonstrated to cause ovarian tumors, chromosome aberrations during oocyte meiosis, and decreased fertility in laboratory animals (see Mattison et al., 1983, for a review). Several

44 DRINKING WATER AND H"LTH investigators have also demonstrated that PAHs destroy oocytes in resting follicles in mice and rats at a rate depending on strain, species, age, dose, and metabolism (Feiton et al., 1978; Mattison and Thorgeirsson, 1978, 1979; Mattison et al., 19831. The following PAHs are known to have this effect: benzofa~pyrene, 3-hydroxybenzoLa~pyrene, 4,5-dihydroepoxy- benzoLa~pyrene, cis-4,5-dihydrodiolbenzofa~pyrene, trans-4,5-dihydro- diolbenzoka~pyrene, 7,8-dihydrodiolbenzofa~pyrene, 7,12-dime- thylbenzota~pyrene, 7,12-dimethylbenzanthracene, and 3-methylcholan- threne (Chapman, 1983; Dobson and Felton, 1983; EPA-ORNL, 1982; Haney, 1985~. The oocytes are actually destroyed by reactive intermediates formed from the parent compound in the ovary by enzyme action. Although this metabolic process is necessary for oocyte destruction, inducibility at the Ah locus is not as highly correlated with this effect as is the rate of metabolism along the pathway leading to formation of the dihydrodiol epoxide (Felton et al., 1978; Mattison and Thorgeirsson, 1978, 1979; Mattison et al., 19831. There are indications that cigarette smoking causes a toxic ovarian response in humans, resulting in premature onset of menopause. The incidence of infertility, defined as a woman never being pregnant through- out her reproductive life, was approximately 12% among white non- smokers compared with 18% among white smokers from a total of 1,728 women in the study (Tokuhata, 19684. There are more than 3,000 iden- tifiable compounds in cigarette smoke, and the specific agents responsible for this effect are not known, although PAHs and nicotine have been implicated (Surgeon General, 1981~. ANTINEOPLASTIC AGENTS A variety of antineoplastic agents have also been associated with ovu- latory dysfunction and destruction of oocytes (Haney, 19851. Included among these are adriamycin, 5-fluorouracil, ~-asparaginase, 6-mercap- topurine, bleomycin, methotrexate, busulfan, nitrogen mustard (mechlor- ethamine), chlorambucil, prednisone, corticosteroids, procarbazine, cyclophosphamide, vinblastine, p-cytosine arabinoside, and vincristine (Chapman, 1983; Dobson and Felton, 1983; EPA-ORNL, 1982; Haney, 19851. These agents destroy rapidly dividing granulosa cells in growing fol- licles as well as in the resting primordial follicle. When prepubertal girls are treated with antineoplastic drugs, complete loss of germ cells is un- likely. In young, postpubertal women, however, fertility may be impaired despite the onset of normal menstrual cycles. As a general rule, the greater the number of chemotherapeutic agents used, and the older the woman,

Reproductive Toxicology 45 the higher the likelihood of gonadal 1985). Dobson and Felton (1983) reviewed data on the oocyte toxicity of 77 chemicals in 11 chemical classes. Of the 77 chemicals tested, 21 caused destruction in resting primordial follicles in mice. Positive compounds were found in 7 of the 11 classes, notably among the PAHs, alkylating agents, esters, epoxides and carbamates, fungal toxins and antibiotics, and nitrosamines. The four negative classes were the aromatic amines, aryl halides, metals, and steroids. injury and permanent sterility (Haney, Alterations in Reproductive Endocrinology In addition to direct effects on the survival of oocytes, exposure to xenobiotic substances can impair female fertility through alterations in the function of the hypothalamic-pituitary-uterine-ovarian axis. The central nervous system (CNS) component of the female reproductive system func- tions in a permissive, integrating role. Hypothalamic neurons synthesize and secrete gonadotropin-releasing hormone (GnRH). These hypothalamic neurons adjoin a portal vascular system that transports GnRH, which is secreted in a pulsatile pattern to the anterior pituitary gland. GnRH func- tions at this site in a permissive capacity, allowing the release of FSH and LH. The pattern with which these gonadotropins are released is con- trolled by the circulating levels of sex hormones (Knobil, 19801. FSH and LH stimulate follicular maturation from the preantral to the preovulatory stage. In addition, they influence the synthesis and secretion of estrogen by thecal and granulosa cells in the follicle. Estrogen is critical to the viability of follicles because it is mitogenic to granulosa cells. During the surge in gonadotropins at midcycle, a series of events is set into motion that culminates in ovulation. These events include intrafollicular prosta- glandin synthesis, terminal oocyte maturation, a shift in steroidogenesis from estrogen to progesterone production by granulosa cells, morphologic luteinization, and, f~nally, rupture of the follicle and release of the oocyte (Takizawa and Mattison, 1983~. Without sustaining factors secreted by a conceptus, the corpus luteum undergoes regression. Peripheral progester- one levels begin to rise with the initiation of the LH surge and continue to increase until the midpoint of the luteal phase, when they begin a gradual decline that results in menses. In humans, human chorionic go- nadotropin appears responsible for maintenance of the corpus luteum dur- ing early pregnancy. There are a number of endocrine processes in which xenobiotic com- pounds can interfere with ovarian function aside from any direct injury of the oocyte. It is difficult, however, to separate direct injury to the follicle from alterations in hypothalamic-pituitary-gonadal function. In-

46 DRINKING WATER AND HEATH terference with specific endocrine functions critical to follicular devel- opment produces the same end points as direct oocyte toxicity—ovulatory dysfunction and reproductive failure. ENDOCRINE ALTERATIONS IN THE PERINATAL PERIOD Steroid hormones have been the most thoroughly studied agents in female reproductive toxicology and have served as model agents for the effects of xenobiotic substances. It has been well established that exposure of female rodents to pharmacological doses of androgens or estrogens during fetal or neonatal life results in disruption of the mechanisms that control cyclic secretion of gonadotropins (Kraulis et al., 1978~. In addition, several contaminants in drinking water, such as Kepone, exhibit estrogenic activity and disrupt control mechanisms (Hudson et al., 19841. The critical period for exposure to steroid hormones extends from day 18 of pregnancy to days ~ to 10 of neonatal life, in the rat. During this time, the hypothalamic centers believed to be involved in control of cyclic hormone secretion undergo neuronal maturation. Disruption or alteration of hypothalamic maturation has a permanent effect that results in an acyclic (tonic) pattern of hormone release like that found in males, and a persistent estrous syndrome characterized by inferdlity-disrupted periodicity (MacLusky and Naftolin, 1981~. In the rat and other rodent species, the persistent estrous syndrome appears to result from the action of estrogens on hypothalamic development during the neonatal period. The effect of androgens is also believed to stem from their aromatization to estrogens in the CNS. Neonatal exposure to estrogenic substances stimulates uterine growth and early vaginal open- ing. These two responses are good indicators of estrogenic action, and when they occur during the neonatal period they can be predictive of persistent estrous and reproductive tract anomalies in the adult (Sheehan et al., 19801. Physiological estrogens (such as estradiol, estriol, and es- trone), nonphysiological estrogens [such as diethylstilbestrol (DES), Ke- pone, dichlorodiphenyltrichloroethane (DDT), and methoxychlorl, and triphenylethylene drugs (such as nafoxidine, tamoxifen, and clomiphene) are known to cause these effects in the rat (Clark, 19821. In addition to effects on the reproductive cycle, exposure of rodents to steroid hormones during neonatal life also causes abnormalities in the reproductive system. Neonatal or chronic exposure of adult hamsters to estrogens results in preneoplastic and neoplastic changes in the vagina, uterus, and pituitary gland (Leavitt et al., 1982; Lin et al., 1982~. Neonatal treatment of mice with androgens causes persistent estrous as well as squamous metaplasia and alterations of stromal collagen in the uterus (Takasugi, 19761. These may result from conversion of the androgens to

Reproductive Toxicology 47 estrogens during the neonatal period or from tonic release of gonadotro- pins, which would result in continuous exposure to ovarian estrogens during adult life. Nonphysiological compounds with estrogenic activity such as DES and clomiphene also cause reproductive tract abnormalities in rodents (Leavitt et al., 19821. Endogenous physiological estrogens are prevented from exerting these toxicities because of their extensive binding to serum proteins. Conse- quently, the level of free hormone is low, leaving relatively little to bind to cellular estrogen receptors. This scenario has been well established in the rat, which has substantial quantities of a-fetoprotein (AFP) in the blood during fetal and neonatal development. AFP binds estradiol with high affinity and thus reduces the level of free hormone in the blood of rodents. DES, a nonphysiological estrogen, is weakly bound by AFP, however, permitting more interaction of this estrogenic substance with cellular receptors. Those estrogens not extensively bound to AFP tend to be potent estrogenic agents capable of disrupting normal reproduction in the rat (McEwen, 19811. Although these concepts have been validated in rodents, there is con- troversy about their applicability to humans. AFP does not bind physio- logical estrogens well in humans, and it cannot be equated with AFP in rodents. Other proteins, such as steroid hormone-binding globulin, along with the high levels of progesterone during human pregnancy may protect against estrogen action (Clark, 19821. The mechanisms that control the sexual development of the reproductive system in primates, including humans, also appear to act differently or to be less sensitive to toxic hormonal influences than they are in rodents. In the rat, it is generally accepted that androgens secreted by the testes during development are converted to estrogens in the hypothalamus (McEwen, 19811. These estrogens act to defeminize the hypothalamus and to produce an acyclic, male pattern of gonadotropin secretion. However, these mech- anisms do not appear to operate in primates; instead, testosterone is con- verted to dihydrotestosterone, which is the active agent. Insofar as aromatization of androgens to estrogens is not involved in masculinization of the primate hypothalamus, exposure to estrogenic agents is not likely to lead to abnormal patterns of gonadotropin release or male infertility. The primate hypothalamus appears to be insensitive to androgens; pharm- acological exposure of female fetuses to androgens can masculinize the external genitalia without influencing the periodicity of the adult menstrual cycle (Clark, 1982). Even though the evidence suggests that hormonal insult from steroids during human development will not influence hypothalamic maturation or cyclic gonadotropin release, it is clear that masculinized behavior patterns are produced in female primates exposed to androgens during pregnancy.

48 DRINKING WATER AND H"LTH Also, maternal exposure to DES during pregnancy has been associated with menstrual irregularity and subfertility in the daughters (Rosenfeld and Bronson, 1980), although the evidence for these effects is not as strong as for the structural and preneoplastic lesions in the reproductive tract. We can conclude, therefore, that exposure to steroid hormones during the perinatal and adult periods is associated with a number of reproductive tract abnormalities in females. Moreover, although there may be differ- ences in the critical period and in the mechanism of action between rodent and primate species, pharmacological exposure to sex steroids early in life predisposes the adults of all mammalian species to subsequent effects on their reproductive system (Clark, 19821. CNS-MEDIATED ENDOCRINE ALTERATIONS IN THE ADULT Certain exposures can cause reversible disruption of hypothalamic pi- tuitary function and gonadotropin release in the adult. In laboratory ani- mals, these effects are seen as temporary suppression of the estrous cycle, ovulation, and fertility in females or as inhibition of androgen production and suppression of spermatogenesis in males (Smith and Gilbeau, 1985~. The evidence indicates that the major pathways involved in hypotha- lamic control of gonadotropins are adrenergic and dopaminergic (Smith and Gilbeau, 1985~. There is profuse catecholaminergic innervation in the hypothalamus, and catecholamines play an important role in gonadotropin release. The surge type of gonadotropin release associated with preovu- latory release of LH and FSH is under noradrenergic control and is stim- ulated by dopamines, norepinephrine, and epinephrine. These catecholamines stimulate the release of GnRH, which in turn controls the release of LH and, to a lesser extent, FSH. Other experiments have demonstrated that (x- but not ,8-adrenergic blocking drugs can suppress GnRH release (McCann et al., 19821. There is growing evidence that the endogenous opioid peptides may also be involved in GnRH release by a reduction in endogenous inhibitory tone at the time of the preovulatory surge in LH and FSH. Injections of morphine or opioid peptides inhibit LH secretion, and injections of na- loxone given to block opioid receptors augment the magnitude of the preovulatory LH surge (McCann, 19821. The precise mechanism by which opioids modulate neuroendocrine function is unknown. Preliminary ob- servations indicate that opioids may affect secretions of biogenic amines; i.e., they may decrease dopamine turnover and norepinephrine concen- trations. A variety of pharmacological agents can modify catecholamine levels by altering synthesis, release, receptor activation, and uptake. Drugs and

Reproductive Toxicology 49 environmental contaminants that produce actions of this type are neuro- pharmacological agents that either inhibit CNS activities (e. g., anesthetics, analgesics, pesticides, sedatives, solvents, and tranquilizers) or that stim- ulate them (e.g., the antidepressants, hallucinogens, natural products, and stimulants). In addition, drugs of abuse are increasingly implicated in disruption of the hypothalamic-pituitary system, leading to reproductive dysfunction (Smith and Gilbeau, 19851. Marijuana and its principal psychoactive ingredient, l\-9-tetrahydro- cannabinol (THC), inhibit secretion of FSH, LH, and prolactin in rodent and nonhuman primate models. In primates, acute administration of THC results in 50% to 80% reductions in serum FSH and LH for up to 24 hours (Smith et al., 19791. Prolactin levels are maximally suppressed between 30 and 90 minutes after drug treatment (Smith et al., 19801. At blood THC levels comparable to those found in regular human marijuana users, nonhuman primates experienced disruption of the menstrual cycle and inhibition of ovulation (Asch et al., 1981~. An 18-day exposure to THC resulted in disruption of the menstrual cycles that persisted until 6 months after treatment (Asch et al., 1979~. In both rodent and primate models, the antifertility effects of THC could be reversed by treatment with GnRH, suggesting that the primary lesion occurred at the hypothalamic level. Narcotic drugs have been found to cause reproductive dysfunction in human addicts. Clinical manifestations of decreased sexual desire and performance, menstrual irregularities, and infertility have all been attrib- uted to altered hypothalamic-pituitary function (Gaulden et al., 1964; Hollister, 1973; Mintz et al., 19741. Acute doses of morphine inhibit ovulation in rats and rabbits (Barraclough and Sawyer, 1955; Sawyer, 19631. Chronic doses of morphine or heroin disrupt estrous cyclicity in rodents and the menstrual cycle in women (Gaulden et al., 1964; Packman and Rothchild, 19761. Evidence for a primary hypothalamic involvement has come from studies in male nonhuman primates, in which GnRH treatment prevented or reversed opioid-induced decrease in plasma tes- tosterone levels (Scher et al., 1983~. Barbiturates are sedative-hypnotic agents that have been used as an- esthetics in laboratory animals for many years. Their general effect is an inhibition of both LH and FSH release and a subsequent depression in steroid hormone levels. Nansel et al. (1979) found that phenobarbital inhibits gonadotropin secretion and thus blocks the rise in serum gonad- otropin levels that would normally follow castration (Nansel et al., 19791. They also found that LH secretion and ovulation can be restored in these animals by treatment with GnRH, indicating a hypothalamic site of action (Nansel et al., 1979; Wedig and Gay, 1973~. Phencyclidine hydrochloride was developed as a tranquilizer for ani- mals, but it has been used by humans as a drug of abuse. Many areas of

50 DRINKING WATER AND HEATH the CNS are affected by phencyclidine hydrochloride, which alters several neurotransmitter systems. Acute administration of phencyclidine hydro- chloride to male rats at dosage levels equivalent to those used by human drug abusers produced slight depressions in serum testosterone and LH levels. Marked depression occurred after nine daily treatments. After treat- ment, LH and testosterone levels were significantly elevated over controls, and they did not return to normal levels until 60 days after withdrawal of the drug (Harclerode et al., 1984~. Juvenile male rats receiving an identical treatment regimen during sexual maturation had severalfold higher ele- vations in hormone levels after withdrawal than did adult male rats, and the period of elevation persisted for 80 days. Other tranquilizers known to alter the hypothalamic GnRH level through effects on endogenous catecholamines are reserpine, chlorpromazine, and perphenazine (Mc- Lachlan et al., 19811. In both human males and laboratory animal males, acute exposure to alcohol primarily affects testicular synthesis and secretion of testosterone (Mendelson and Mello, 19841. Both ethanol and acetaldehyde inhibit enzymes involved in gonadal testosterone synthesis (Johnston et al., 19811. Multiple endocrine abnormalities, including hypogonadism and gyneco- mastia, sometimes occur along with alcoholic cirrhosis (Gordon et al., 19781. Levels of estrogenic steroids increase as a result of altered hepatic metabolism and clearance of androgens. Women and nonhuman primate females do not appear to be as sensitive to the direct gonadal effects of alcohol and may be less vulnerable to antifertility effects with chronic alcohol abuse (Van Thiel et al., 1977, 1978~. Most neuroactive drugs produce transient effects on CNS pathways necessary for normal gonadotropin secretion. The disruptive effects of these drugs on sexual and reproductive function are likely to be transient and completely reversible. Adults with compromised reproductive function and prepubertal adolescents may be at greater risk to long-term impairment due to lack of hypothalamic-pituitary-gonadal homeostasis. Test Systems for Detecting Female Infertility Female fertility can be disrupted at a number of points during life and at a number of levels in the hypothalamic-pituitary-uterine-ovarian axis. Laboratory tests used in safety studies, however, tend to measure apical end points of estrous cyclicity and the ability to conceive and bear off- spring. Entry-level tests can include exposure of males, females, or both sexes. Two basic designs have been developed for reproductive toxicity testing, one involving exposure during one generation and the. other based on exposure across several generations. (For details of the tests discussed in this section, see the appendix to this chapter.) The reproductive toxicity

Reproductive Toxicology 51 Fo Parental Animals 1st /1-2 weeks \ mati ng / - ' after weaning \ F'A Autopsy at weaning Initiate dosing at 30-40 days, perform first mating at 100-120 days of age. 2nd mating (continue mating for F. C F. B TWean Select F1 Parental Animals F2A F2 B T T Autopsy at weaning or F1 D) | Wean Select F2 Parental Animals / ~ F3A F3B T T Autopsy at Autopsy at weaning, weaning histopathology FIGURE 3-3 Multigeneration study encompassing Tree generations. From Manson et al., 1982, with penn~ssion. Of therapeutic drugs is usually evaluated in single-generation studies, on the premise that most are taken for relatively short periods and have comparatively short half-lives in the body. Multigeneration studies are used for compounds likely to concentrate in the body as a result of long- term exposure, such as exposures to pesticides and food additives. In these studies, animals are continuously exposed to the test compound, usually in the food or drinking water, for three generations. A new test procedure entitled Fertility Assessment by Continuous Breeding (FACB), under de- velopment in the National Toxicology Program, may provide a more accurate assessment of long-term, multigeneration effects on fertility. The basic designs of the multigeneration and FACB tests are shown in Figures 3-3 and 3-4. These tests are designed to give an overview of the reproductive process. As such, they cast a broad net over a spectrum of reproductive end points to determine if any are impaired. Impairment often takes the form of an inability to conceive and to bear viable litters. If parental weight gain is approximately within a 10% range of control values, such impairments are indicative of a specific effect on the repro-

52 DRINKING WATER AND H"LTH Task 1 Dose Finding (14 days) Task 2 Continuous Breeding (7+98+21 days) ~ ' A Asses s m e no ( N 0 ) <A Lit I (Optional ) Task 3 ____ Determination of Affected Sex I (Optional) Task 5 (Optional) _ Measurement of Hormone Patterns FIGURE 3-4 Flow chart for National Toxicology Program Fertility As- sessment by Continuous Breeding protocol. From Lamb et al., 1984, with . . permlsslon. ductive system. As time-consuming and laborious as these tests are, they rarely provide any insight into the mechanism of action or critical period of exposure, nor should they be expected to do so. In the context of safety testing for persistent environmental agents, a multigeneration-type test is used at initial stages to determine if an agent has any overall effect on reproduction. Consequently, these tests give a qualitative, yes-or-no signal about reproductive toxicity. If well performed, they also provide infor- mation of use in the selection of doses for single-generation studies. In Segment I studies, which are entry-level fertility tests for drugs, the effects of the test agent on gonadal function, estrous cyclicity, mating behavior, conception rates, and development are assessed. In addition, an overall view of the reproductive process within one generation is obtained.

Reproductive Toxicology 53 20 ~ 60 days \ Gestation Lactation bow /! 1 14 days Mate 1 d+2Q Birth 22 days Weaning 21 days FIGURE 3-5 Segment I (general fertility and reproductive performance) test. From Manson et al., 1982, with permission. The basic design of such tests is illustrated in Figure 3-5. If adequately performed, Segment I studies can yield information on the critical period of exposure, but they very seldom give information on the site or mech- anism of action. Although lowest-observed-effect levels (LOELs) and no-observed- effect levels (NOELs) can be identified from these studies, it should be understood that mating performance in rodents is not a sensitive indicator of the integrity of the reproductive system. Extensive germ cell loss, particularly of resting follicles, can occur in female rodents before fertility is measurably disrupted. When an agent tests positive in safety studies (i.e., in multigeneration or Segment I studies), additional studies to pursue the site and mechanism of action should be conducted. Information on some of these tests (e.g., protocols for identifying estrogenic activity and primordial oocyte toxicity) is given in the appendix to this chapter. These tests are more likely to yield the most accurate and quantitative data on reproductive toxicity. When available, results from these tests should be used in identifying LOELs and NOELs. SUSCEPTIBILITY OF THE MALE TO REPRODUCTIVE IMPAIRMENT The proliferation of human and laboratory studies conducted over the past 5 years to determine the impact of environmental exposures on male reproductive function reflects a growing concern about such effects. This section describes the susceptible stages in the development of the male reproductive system and the types of reproductive toxicities that are as- sociated with exposure to xenobiotic compounds. Maturation of the Male Reproductive System In the early fetal period, prespermatogonial germ cells undergo mitotic proliferation and migration to the gonad, as described earlier in this chapter

54 DRINKING WATER AND HEALTH for the female. When the gonad differentiates into a testis, Sertoli and Leydig cells form, androgen secretion begins, and the basic pattern of the male reproductive tract is established. Sertoli cells are epithelial cells lining the seminiferous tubules of the testis; they provide the cellular matrix within which germ cells develop. Leydig cells appear in the interstitial regions between Sertoli cells and are the site of de novo androgen synthesis. As a direct consequence of androgen synthesis in the fetal testis, the male reproductive tract grows and differentiates, and the accessory glands and external genitalia are formed (Gondos, 19801. A major difference in male and female germ cell development is that male germ cells do not enter meiosis until puberty. Rather, in the late fetal, early neonatal period, when female oogonia enter prophase I of meiosis, the male prespermatogonial cells enter a period of mitotic arrest. Mitotic activity of prespermatogonial cells is reinitiated at about 10 years of age, and active spermatogenesis begins some 3 years later (Hafez, 19771. Sertoli cells undergo extensive changes during the postnatal period under the influence of FSH. Cell division and growth continue until the onset of puberty, at which time the Sertoli cells become nonproliferating. FSH stimulates the production of androgen-binding protein by Sertoli cells, resulting in a local accumulation of androgen necessary for the initiation of spermatogenesis. Intercellular bridges formed between Sertoli cells under the influence of FSH constitute the blood-testis barrier. This divides the seminiferous tubules into basal and adluminal compartments and iso- lates the spermatid cell stages from blood-borne chemicals during its development from the spermatocyte stage. Leydig cells undergo a regres- sion in late fetal, early neonatal life, and testosterone production declines. In the postnatal period, Leydig cells redifferentiate, and testosterone and androstenedione synthesis increases, reaching a maximum shortly before puberty (Gondos, 19801. At puberty, prespermatogonial cells are converted to spermatogonial cells, and meiosis is initiated. In this process, stem cells, the type A spermatogonia, pass through a complex series of transformations to give rise to spermatozoa. During the first phase, type A spermatogonia undergo six mitotic divisions to form type B spermatogonia. After the last mitotic division, preleptotene primary spermatocytes form and pass through the blood-testis barrier into the more protected environment of the adluminal compartment of the seminiferous tubule. These spermatocytes replicate their DNA, enter meiosis, and proceed to the leptotene stage. During the subsequent zygotene stage, the pairing of homologous chromosomes oc- curs. Cells with completely paired chromosomes are termed pachytene primary spermatocytes. The progression from early pachytene through the middle and late pachytene stage to form secondary spermatocytes is the longest process in meiotic prophase and is also the point at which cells

Reproductive Toxicology 55 are especially susceptible to damage. The secondary spermatocytes have a short life span, and without replication of their DNA, they enter the second meiotic division and form haploid spermatids. The mitotic divisions of spermatogonia into primary spermatocytes are not hormone dependent, but the meiotic divisions of primary spermato- cytes into spermatids are believed to be testosterone dependent. The second stage of spermatogenesis, known as spermiogenesis, is characterized by morphological and functional maturation of spermatids to form sperma- tozoa. Acrosome formation, nuclear condensation, and tail formation take place, accompanied by the loss of cytoplasm. Even after completion of spermatogenesis, significant changes occur in the sperm of mammals during epididymal transport. Mammalian sperm are incapable of fertilization when they enter the epididymis, but acquire this capacity as they pass through the epididymal duct. They undergo changes in metabolism and exhibit alterations in motility, acrosome shape, degree of cross-linking of nuclear chromatin, and surface charge. Sperm retain their fertilizing capacity in the epididymis for 20 to 35 days in most mammalian species (Bedford, 19664. They attain their full capacity for fertilization after they enter the female reproductive tract in the process of capacitation, when the pattern of energy metabolism changes and mo- tility increases. Capacitation is believed to involve the removal of seminal plasma factors from the surface of the sperm (Aonuma et al., 1973~. Differences between species in various aspects of sperrnatogenesis are given in Table 3-2. Sperm Toxicity An estimated cell loss of 35% occurs from the early spermatocyte to the spermatid stage due to physiological atresia (Salisbury et al;, 19771. Unlike germ cells in the female, however, cells lost to atresia in the male can be replaced by continual production of spermatocyte cells from di- vision of stem cells. Thus, physiological or drug-induced destruction of germ cells does not necessarily result in a shortening of the fertility span in the male, although it may cause temporary reduction in sperm count. This is true as long as some stem cells are spared; if the insult completely destroys the stem cell population, then permanent infertility will occur. Testes function is controlled by the gonadotropins LH and FSH. LH stimulates Leydig cells to synthesize androgens, which control the func- tional activity of accessory sex organs and the development of secondary sexual characteristics. Prolactin, in the presence of other pituitary hor- mones, has a stimulatory effect on steroidogenesis in Leydig cells. Ele- vated prolactin levels have the opposite effect and can lead to impotence and infertility. FSH acts on Sertoli cells and is important in the initiation

56 DRINKING WATER AND H"LTH TABLE 3-2 Species Differences in Spennatogenic Parametersa Charactenstic Mouse Rat Rabbit Dog Monkey (Beagle) (Rhesus) Man Duration of seminiferous epithelium cycle (days) 8.9 12.9 10.7 13.6 9.5 16.0 Life span (days) of: Type B spermatogonia 1.5 2.0 1.3 4.0 2.9 6.3 Leptotene spermatocytes 2.0 1.7 2.2 3.8 2.1 3.8 Pachytene sperrnatocytes 8.0 11.9 10.7 12.4 9.5 12.6 Golgi spermatids 1.7 2.9 2.1 6.9 1.8 7.9 Cap spermatids 3.6 5.0 5.2 3.0 3.7 1.6 Fraction of life span as: Type B spe~matogonia 0.11 0.10 0.08 0.19 0.19 0.25 Primary spermatocyte 1.00 1.00 1.00 1.00 1.00 1.00 Round spermatocyte Epididymal transit time (days) Testes wt (g) Daily sperm production Per gram testis (106/g) Per male (106) Sperm reserves in cauda at sexual rest (106) 0.41 0.2 28 s 0.40 0.43 0.48 0.35 0.38 8.1 12.7 10.5 12.0 3.7 6.4 12.0 49 34 24 25 86 160 20 23 300 1,100 4.4 125 49 440 1,600 5,700 420 aAdapted from Amann et al., 1976, and Amann, 1982. Ot spermatogenesis at puberty and in the maintenance of optimal testicular function in the adult. The binding of FSH to Sertoli cells results in a general increase in protein synthesis through a cyclic adenosine mono- phosphate-mediated process. One protein synthesized is inhibin, which, when secreted, acts as a negative feedback control on the release of FSH. Another is androgen-binding protein, which maintains the high testoster- one levels in Sertoli cells needed for maintenance of spermatogenesis. FSH also influences the biosynthesis and interconversions of steroid hor- mones (Phillips et al., 19851. Effects of Radiation on Spermatogenesis The effects of radiation on spermatogenesis have been studied in a variety of species with different model systems and indices of measure-

Reproductive Toxicology 57 meet. Although the majority of information has been collected in mice and humans, some investigators have studied the effects of irradiation on the fruit fly, Drosophila melanogaster (Eeken and Sobels, 1985; Miya- moto, 1983; Sinclair and Grigliatti, 1985), the Amazon molly, Poecilia formosa (Woodhead et al., 1984), rats (Dedov, 1980), and male dogs (Fedorova et al., 19851. Spermatogenesis in the mouse has been proposed as an in viva test system with utility as a biological dosimeter (Hacker et al., 1980, 19811. Manikowska-Czerska et al. (1985) demonstrated that radiation interfered with spermatogenesis in the mouse, with chemical protection against these effects having low efficacy (Pomerantseva and Ramaija, 19841. Not all stages of spermatogenesis are affected equally in this species, however. Furthermore, cross-species and even cross-strain comparisons of the var- ious stages of spermatogenesis and their susceptibility to radiation should be carefully examined before general conclusions can be drawn (Leenhouts and Chadwick, 19811. Thus, for example, the repair capacity of stem cells and spermatogonia in mice is remarkably high (Hacker-Klom, 1985), in sharp contrast to that observed in man (Greiner, 19821. Similarly, significant strain-specific differences in DNA repair of ultraviolet-induced lesions have been demonstrated in early mouse embryos (Bennett and Pedersen, 19841. Several investigators have reported that spermatogenesis is considerably more sensitive to the effects of radiation in humans than in mice (Clifton and Bremner, 19831. This difference can be observed in the effects still present 2 to 9 months after exposure (Meistrich and Samuels, 19851. Martin et al. (1985) noted a progression to azoospermia and poor egg penetration in humans following exposure to radiation. All patients re- covered from the azoospermia 36 to 48 months after exposure. Thus, while man may be more sensitive than mice to the effects of radiation on spermatogenesis, in terms of duration of observable effect, it appears the species are comparable in their sensitivity to longer-term, irreversible damage (Meistrich and Samuels, 19851. Test Systems for Evaluating Mate Reproductive Toxicity in Laboratory Animals ENTRY-LEVEL TESTING The study of male reproductive toxicity encompasses scientific ap- proaches ranging from general toxicity testing to examination of the mech- anism of action. The first step in assessing the potential action of an agent on male reproductive function should be to conduct tests in the intact animal so that the spectrum of physiological processes involved can be examined at one time, regardless of the target organ, cell, or molecule.

58 DRINKING WATER AND H"LTH A comprehensive test system would include the varying susceptibility of the stages of spermatogenesis, the endocrine control of reproduction, the effects on libido, the ability of mature sperm to reach and fertilize the ovum, and, finally, the ability of the conceptus to complete development successfully. Current approaches to evaluating reproductive function in laboratory animals are reviewed in the appendix to this chapter. The decision whether to conduct a multigeneration or a single-generation (Segment I) study depends on the length of human exposures and the type of agent. In single- generation studies, male rodents are exposed to the test agent for 60 to 80 days and then mated with control females to determine their mating ability (i.e., the number of females inseminated compared with the number of females that become pregnant). After mating, the males are killed and their reproductive organs are weighed and histologically examined. Mated females are killed at midpregnancy or at term, and their uterine contents are examined for preimplantation and postimplantation death. When mated females are killed at term, the fetuses can be weighed and examined for malformations. Although this type of test is comprehensive, there is growing concern that rodent mating trials may be too insensitive to assess even dramatic effects on spermatogenesis and epididymal functioning. The insensitivity of rodent mating trials is due to the vast excess of sperm ejaculated over the amount required for normal fertility. Males of conventional laboratory animal species produce and ejaculate from 10 to 100 times more sperm than are necessary for normal fertility and litter size (Aafjes et al., 1980; Amann, 19821. Thus, the number of sperm available for ejaculation in rodent species can be reduced by as much as 90% before sterility occurs (Meistrich et al., 1978~. Evaluations of human semen and testicular function have led to the conclusion that "in comparison to laboratory animals, human testes are often functioning at the threshold of pathology" (Amann, 19821. For many men, the number of sperm ejaculated is only two to three times higher than the level at which fertility might be expected to decline (MacLeod and Wang, 1979), whereas in animals the number is many times higher (Amann, 19821. Median counts from four studies of fertile men ranged from 38 million to 68 million sperm per milliliter of ejaculate, while the potential for male infertility is believed to increase when counts fall below 20 million/ml (reviewed in MacLeod and Wang, 19791. Furthermore, the percentages of progressively motile sperm and of morphologically normal sperm in human semen are lower than those typical of males in other species. Consequently, measures of fertility and fecundity in animal mod- els may be insensitive criteria for monitoring the integrity of the male reproductive system, and for identifying agents with the potential for causing reproductive toxicity in men (Amann, 19821.

Reproductive Toxicology 59 A major difference between humans and laboratory animals is in the daily sperm production rate per gram of testis (see Table 3-21. The species of choice for reproductive toxicity testing are the rat and the rabbit. Rats are preferable to other rodents because of their well-characterized repro- ductive processes and general use in toxicological studies. As in all ro- dents, however, semen quality in rats cannot be evaluated in longitudinal studies. Rabbits are good in such studies because of the large amount of background information on their reproduction and because their semen can be quantitatively collected in longitudinal studies (Amann, 19821. Current recommendations for entry-level testing are to expose rats to an agent for the complete period of spermatogenesis, which extends over approximately 77 days and includes 6 cycles of the seminiferous epithe- lium. After 77 days, treated males are mated with control females to evaluate fertility. After mating, males are given 6 to 8 days of sexual rest to allow restoration of sperm reserves in the cauda epididymis. If combined with a 90-day subchronic toxicity study, fertility testing would thus be postponed until days 78 to 84. At the time of sacrifice, reproductive organs are weighed and placed in fixative for histological evaluation. One testis and epididymis are saved to evaluate testicular sperm production rate and epididymal sperm numbers, morphology, transit time, and motility (Amann, 1982; EPA-ORNL, 1982). This testing procedure is comprehensive and sensitive. Disruption of any of the reproductive processes in the male could be detected by de- creased testicular or epididymal characteristics or by lowered fertility or fecundity. On the basis of results from this entry-level test, decisions could be made to conduct secondary-level testing to pursue positive results. If an elevation in epididymal sperm with morphological abnormalities was identified, for example, secondary testing could be conducted to identify germ cell mutations, which is described later in this chapter. Similarly, agents that did not affect sperm production but that caused failures in mating could be evaluated for effects on the physiology of ejaculation. Human Test Systems According to Overstreet (1984), more than 15% of all recently married couples will have major difficulties in conceiving a child. Approximately one-third of these cases of infertility result from a pathological condition in the male, one-third are attributed to the female, and one-third are due to a combination of factors in both partners. More than 75% of female infertility problems can be diagnosed and treated, but the biological bases of male infertility are not well understood. Few if any therapies can be offered to most infertile men. This clinical inadequacy can be attributed to a lack of basic research on male infertility, as the brief review in this

60 DRINKING WATER AND HEALTH section indicates, and to the inadequacy of current diagnostic procedures (Overstreet, 19841. SPERM MORPHOLOGY Extensive studies have been performed on the induction of abnormal sperm shape by physical and chemical agents in laboratory animals (Wy- robek, 19771. The visual assessment of sperm morphology is very sub- jective and critically dependent upon the classification scheme used. It is relatively easy to detect subtle deviations in the shape of the highly angular sperm heads in laboratory rodents. The more ovoid shape of the typical human sperm head, however, makes subtle differences in shape difficult to detect. There are also large interlaboratory differences in sperm mor- phology criteria (Wyrobek et al., 19841. Recent work in mice suggests that generalized toxicity in the whole animal can result in an increased incidence of abnormal sperm forms (Komatsu et al., 1982), indicating that sperm abnormalities may not always be the direct effect of chemically induced gonadal damage. In the human, the link between semen quality and adverse pregnancy outcome is am- biguous. In some studies (Furuhjelm et al., 1962), poor sperm quality (count and morphology) was linked to the frequency of early pregnancy loss. Fathers of 201 spontaneous abortions had significantly more sperm shape abnormalities and lower sperm counts than did fathers of 116 normal pregnancies. Although several studies support a link between sperm de- fects and abnormal reproductive outcome, others found no correlation (see Wyrobek et al., 1984, for a review). More studies in humans are needed to compare exposures of male parents with resulting sperm defects and reproductive outcomes. The morphology of sperm in the semen of fertile humans is stable. Approximately 60~o of the sperm heads have the normal oval shape, and 40% have some type of abnormal form. The percentage of abnormally shaped sperm in an ejaculate remains relatively constant for one person but varies considerably between men. To detect relatively small changes in the percentage of abnormally shaped sperm following an exposure, a preexposure baseline should be established for each male. A procedure entailing repeated sampling and analysis of the ejaculate is the most ac- curate way to detect increases in abnormally shaped sperm. Sperm morphology has gained popularity as a human surveillance tech- nique because, unlike sperm number and motility, it is unaffected by frequency of ejaculation. In a study of six repeated measurements of semen samples from 100 males, large variations in volume, sperm number, and motility were found, but sperm morphology was identified as the most predictive and stable parameter for diagnosing infertility (Sherins et al., 1977~. As indicated in Table 3-3, a relatively small population (at least

Reproductive Toxicology 61 TABLE 3-3 Relative Sensitivities of Three Assays of Human Sperma Type of Assay (and Population Size)b Sensitivity Counts Measure (N=214) Morphology, % (N=26) 2F Bodies, % (N=41) Mean valuesC 132 x 106/ml 41.9 0.8 Standard deviation 160 x 106/ml 12.4 0.7 aAdapted from Edmonds et al., 1981, pp. 47 and 69, with permission of the copyright holder, the March of Dimes Birth Defects Foundation. bSample size for 5% level test with 80% power. CMean sperm counts, percentage of morphologically abnormal sperm, and percentage of spend with 2F bodies in semen samples from fertile men. 26 men) is needed in each group to conduct statistically valid studies of changes in sperm morphology. Figure 3-6 contains drawings of morphologically normal and abnormal human sperm cells. The abnormal shapes are relatively easy to identify from stained preparations of semen smears. Other abnormal shapes, es- pecially those involving sperm with small heads, are difficult to detect, and classification is usually subjective. Recent attempts to use computer- assisted technology for morphometric analysis of head length, maximum head width, head area, head circumference, and maximum midpiece width should greatly reduce the subjectivity of sperm shape assessment (Over- street, 19841. Since some morphologically abnormal sperm lack motility, the extent to which malformed sperm participate in fertilization is unclear. In the mouse, the proportion of sperm with abnormal head shape remains un- changed during transit from the testis, through the epididymis and vas deferens, and into the uterus. Around the egg, however, the proportion of abnormally shaped sperm is substantially lower (Krzanowska, 19741. Other studies have suggested that sperm with abnormally shaped heads but with normal motility and intact acrosomes can still fertilize (Clavert et al., 19751. In a review of this topic, Salisbury et al. (1977) contend that sperm take part in fertilization in numbers proportional to those reach- ing the egg. Selective pressures against sperm with grossly imbalanced chromosome complements and abnormal morphology are exerted during spermatogenesis and in the female reproductive tract. If genetically dam- aged sperm do fertilize an egg, a major selective pressure is embryo death; relatively few if any embryos carrying the genetic damage survive to term. Thus, although abnormal sperm shape is clearly indicative of a disrup- tion in spermatogenesis, it is not necessarily predictive of adverse preg- nancy outcome. A more definitive conclusion about the role of these

62 DRINKING WATER AND H"LTH \ / Abnorma I ~ \ Normal FIGURE 3-6 Normal and abnormal morphologies of human spend cells. _ Acrosome Nucleus ~ Centrosome \. Mitochondria Soiral Ring Centriole abnormalities in determining the nature of the offspring will be important, because semen analysis is particularly adaptable to human reproductive risk assessment. As this method gains wide acceptance, it will be necessary to determine how predictive it is of male reproductive insult versus the potential for adverse pregnancy outcome. SPERM MOTILITY The motility of sperm in human semen has been used for many years to predict male infertility (MacLeod and Gold, 1951~. It is usually rated in two ways: by percentage of motile cells and by the quality of sperm movement (e.g., how fast and how straight the sperm swim). Sperm movement has traditionally been rated on a scale of 0 to 4 +, 2 + being regarded as normal. The normal percentage of motile cells in human semen is at least 50% to 60% (Eliasson, 19759. The subjectivity of this assessment has greatly limited its usefulness. More recent, less subjective methods are based on photomicrographic techniques to obtain objective measurements of human sperm movement. Time-exposure photomicroscopy of moving sperm produces tracks on the negative that can be measured to calculate swimming speed. A mean swimming speed of 25 Am per second is considered adequate (Overstreet,

Reproductive Toxicology 63 19841. In more recent procedures, videotapes of semen have been used to measure sperm swimming speeds. Videomicrography is more conve- nient and easier to interpret than time-exposure photography, since the sperm themselves rather than their tracks can be visualized (Katz and Overstreet, 19811. SEMEN ANALYSIS At present, semen analysis is the primary method used to assess male fertility in humans. Induction of pregnancy is the only other test of this potential, but this end point is subject to highly confounding and variable factors such as female infertility, frequency of sexual contact, and con- traceptive use. The strong differences in opinion on what constitutes a normal ejaculate are based on variations in methods for collecting semen as well as on the wide range of normal variability in the human population. Methods for analyzing semen from humans and laboratory animals have undergone extensive development in the past 5 years. The evolution of basic research in mammalian sperm physiology has resulted in several new techniques for assessing human sperm functions. The following pa- rameters can be measured from semen samples: · gross parameters, including sperm concentration, sperm motility, and sperm morphology; · functional parameters, including sperm-cervical mucus interaction, sperm-oocyte interaction, and seminal fluid content; · genetic parameters, such as sperm chromosome complement (as mea- sured with the YFF, or 2F bodies test, as described below). SPERM CONCENTRATION Sperm concentrations in the ejaculates of fertile and infertile men have been studied with the goal of identifying a range of sperm concentrations that would be predictive of fertility. Traditionally, subfertility was believed to occur when the sperm concentration was less than 20 million/ml of ejaculate (MacLeod and Gold, 1951~. More recent studies have indicated that this level may be too high for an accurate assessment of fertility (Sherins et al., 1977; Smith and Steinberger, 19771. The division between fertility and subfertility has more recently been set at 10 million/ml, or 50 million sperm/ejaculate, a figure supported by conception rates as well as by plasma gonadotropin levels. Significant elevation in plasma FSH, indicative of germinal epithelial damage, occurs in men with a sperm concentration less than 10 million/ml and with total sperm counts below 25 million/ejaculate (Smith and Steinberger, 19779. It is important that

64 DRINKING WATER AND H"LTH semen donors have 2 to 3 days of abstinence from ejaculation before providing a sample for sperm count determination. Although sperm concentration is one of the most frequently examined semen parameters, it is not a sensitive indicator of early spermatogenic insult. Since sperm counts vary considerably, even among fertile and presumably healthy men, many cooperative subjects are needed to estab- lish differences between control and exposed groups in studies where each male is sampled only once. As indicated in Table 3-3, at least 214 men are needed in each group (control and exposed) to detect a 25% change in mean sperm count values. With mean sperm count values of 132 million/ ml, men in the infertility range of less than 10 million/ml have most likely already experienced a considerable degree of spermatogenic insult. Thus, sperm count measurements alone can be used to detect severe cases of potential subfertility. When coupled with other semen parameters, a more sensitive indication of early or marginal insult can be obtained. SPERM CHROMOSOME COMPLEMENT The Y chromosome in human sperm can be differentially stained with quinacrine dye so that it becomes a visible fluorescent spot in individual sperm cells. This procedure, called the YFF or 2F bodies test, is based on scoring the frequency of sperm with two fluorescent (F) bodies and thus presumably two Y chromosomes in an ejaculate (Kapp and Jacobson, 19801. Although 50% of the sperm would be expected to have one spot, in practice, frequencies range from 30% to 40%. Approximately 1% to lo of the sperm in ejaculates from fertile men contain two Y chromo- somes, or two F bodies. Sperm with two Y chromosomes are believed to be formed as a result of meiotic nondisjunction, when homologous chro- mosomes are not equally partitioned to secondary spermatocytes during metaphase of the first meiotic division. An elevated frequency of sperm with two Y chromosomes has been interpreted as indicating nondisjunction of the Y and other chromosomes, and thus of chromosomal abnormality of the sperm (Kapp and Jacobson, 19801. Given the tight packing of nuclear material in sperm heads, it is not possible to visualize any chro- mosomes other than the Y for cytogenetic analysis. Although Kapp and Jacobson suggested that sperm with two spots contain two Y chromo- somes, major uncertainties remain in this interpretation. Validation of this test has been hampered by the lack of fluorescent Y chromosomes in the sperm of common laboratory or domestic animals. This quality of fluorescence seems to be unique to the Y chromosomes of humans and the higher apes. Nonetheless, this system may provide a useful model of chemically induced nondisjunction in male germ cells. Relatively small test populations (at least 41 men in each treatment group)

Reproductive Toxicology 65 are needed for statistically valid application of the YFF test (see Table 3-31. Measurement of chromosome aberrations in Recondensed human sperm DNA may also be useful as an assay for human germ cell mutagens (Preston, 19821. But interindividual variation in background aberration rates must be thoroughly understood before this test can be meaningfully applied. The chemical induction of a positive response in an in vitro genetic toxicity assay cannot be considered an indication of the chemical's po- tential as a germ cell mutagen in vivo. Numerous compounds that exhibit genetic toxicity in vitro and many compounds that are carcinogens in viva in animals fail to produce measurable genotoxicity in germ cells in vivo. Dimethylnitrosam~ne, for example, is a potent genotoxic carcinogen (IARC, 19781; however, it does not produce detectable aLkylation of mouse sperm heads in vivo (Stott and Watanabe, 1980), it does not induce DNA repair in spermatocytes in vivo (Doolittle et al., 1984; Working and Butterworth, 1984), it does not alter sperm number or the incidence of morphologically abnormal sperm in hamsters (Wyrobek et al., 1978), and it is uniformly negative in mammalian germ cell mutation assays (Epstein et al., 19721. A variety of pharmacokinetic and adaptational factors in germ cells are probably the cause of this apparent mutagenic selectivity. Pharmacokinetic parameters that can modify the mutagenic potential of xenobiotic compounds in the gonads include, but are not limited to, those that control the absorption and detoxification of xenobiotic compounds within the exposed animal as well as the biotransformation capability of the gonads. Evidence indicates that the gonads contain relatively low levels of biotransformation activity with respect to monooxygenation reactions and, in contrast, that detoxifying enzymatic activities predominate in both ovaries and testes (see review by Heinrichs and Juchau, 19801. Epoxide hydratase, glutathione-S-transferase, and aryl hydrocarbon hydroxylase activities have been measured in the testes and ovaries and in some cases are differentially distributed between germ cell and interstitial cell com- partments (Mukhtar et al., 1978; Oesch et al., 1977~. Gonadal biotrans- formation is an important factor in the chemical induction of germ cell mutagenesis, since highly reactive electrophilic metabolites have ex- tremely short biological half-lives and, thus, are not likely to be transferred from one organ to another (Nelson et al., 1977~. Hence local, i.e., ex- trahepatic, bioactivation of promutagens plays an important role in the mediation of chemically induced genotoxicity within the gonads. Accurate information about the mutagenic hazards of chemical exposure has been difficult to obtain and is therefore limited. The few assays suitable for detecting germ cell mutagens in vivo generally fall into one of two broad categories: (1) those that detect effects presumed to be related to

66 DRINKING WATER AND H"LTH DNA alterations and that may be useful as predictors of mutagenic po- tential, and (2) those that directly measure heritable mutagenic events. Assays in the former category measure germ cell effects but not necessarily heritable mutation. Assays for germ cell effects include the measurement of DNA damage (Skare and Schrotel, 1984) or repair (Pedersen and Bran- driff, 1980; Sega, 1982; Working and Butterworth, 1984), the assessment of sperm morphology (Wyrobek et al., 1982), the measurement of mu- tagen-induced chemical modification of DNA (Stott and Watanabe, 1980), the detection of chemically induced chromosome aberrations in the zygotes (Adler and Brewen, 1982), and the dominant lethal test (Ehling, 1977) which scores all genetic effects in early embryos that cause death of offspring. Positive results in any of these assays should not be considered as final evidence that a chemical is a germ cell mutagen but, instead, that it is an indicator of possible mutagenicity. A positive response can, however, be taken as presumptive evidence that a suspected mutagen or its metabolites did reach the germ cells. Assays of this type detect actual genomic alter- ation, measured as the effects of gene mutation or gross alteration or damage of chromosomes. The heritable translocation test (Generoso et al., 1980) detects primary reciprocal translocations; the specific locus test (Russell et al., 1981) measures intragenic lesions at marked loci, which are recovered in Fit offspring. Many factors must be weighed when extrapolating data from these in vivo animal test systems to humans. Differences in background mutation rates have been observed between different strains of mice (Generoso et al., 1983), between different animal species (Lyon and Smith, 1971), and even between humans from urban and nonurban populations (Neel and Rothman, 19811. Nevertheless, an indication of mutagenic activity in the germ cells of test animals must be taken as a warning of potential mutagenic activity in humans. The most relevant animal models are those in which germ cell mutations can be measured directly by the heritable translocation test and the specific locus test. Assays that measure germ cell DNA effects, including dominant lethals, DNA damage and repair, chemical alterations in germ cell DNA, and alterations of chromosome structure in zygotes, may be less relevant but can provide information relating to the accessibility of the gonads to the putative mutagen. SPERM-CERVICAE MUCUS INTERACTION Sperm functions related to transport in the female can be assessed by observations of sperm-cervical mucus interaction. Sperm deposited in the vagina must migrate through cervical mucus en route to the site of fer-

Reproductive Toxicology 67 tilization. A postcoital test to examine sperm in the mucus is an important part of any infertility evaluation (Overstreet, 19841. Sperm penetration of midcycle conical mucus in humans significantly correlates with fertility (Kunitake and Davajan, 1970; Ulstein, 19731. In one study, the relative fertility of sperm donors in artificial insemination programs correlated more strongly with cervical mucus penetration than with any other semen parameter (Ulstein7 1973~. Katz and Overstreet (1980) developed a new quantitative technique for evaluating this inter- action through the use of flat capillary tubes. In this assay, the expected number of collisions between sperm cells and mucus is determined and the proportion of those collisions that result in successful penetration is calculated. In addition, the vitality of sperm that enter the mucus is mea- sured by comparing their swimming speeds there with their speeds in the semen. The difficulty of obtaining midcycle human cervical mucus for this test, however, limits its immediate application to human reproductive risk assessment. SPERM-OOCYTE INTERACTION Yanagimachi et al. (1976) have developed a test of human sperm fer- tilizing capacity that involves the incubation of human sperm with hamster ova. In this system, the zone pellucida of the hamster ova, which con- stitutes a barrier to cross-species fertilization, is enzymatically removed. This permits penetration of human sperm into the hamster ooplasm. Both capacitation and the acrosome reaction are required for this interspecies fertilization, and the hamster egg system is a suitable assay for these sperm functions. The penetration of the treated hamster ova by sperm from infertile men was significantly lower and was not correlated with sperm concentration, motility, or morphology. Although tests of sperm-oocyte interactions appear to be sensitive in detection of the early stages of functional infertility (i.e., not associated with abnormal counts, mor- phology, or motility), they are technically difficult to perform. At present, sperm-oocyte interaction tests are most appropriately applied in settings such as artificial insemination clinics rather than in surveillance studies of human populations. EXAMINATION OF SEMINAL FLUID The average ejaculate from a fertile man comprises approximately 3 ml of semen containing 50 million to 120 million sperm per milliliter of seminal plasma. Approximately 60% of the volume of the seminal plasma is contributed by seminal vesicle secretions, 30% by the prostate gland, and the remainder by the urethral and bulbourethral glands, ampullae,

68 DRINKING WATER AND H "LTH epididymis, and testes (Lewin, 19771. Sperm cells, along with accumu- lated rete testis fluid, pass into the caput epididymis, which absorbs most of the rete testis fluid and secretes certain components of it principally carnitine. Mature sperm move from the epididymis into the vas deferens and then to the ampulla. Upon ejaculation, sperm are mixed with secretions from the seminal vesicle and prostate, which together provide the bulk of seminal fluid volume. The biochemical composition of the seminal plasma is complex and differs in many respects from that of blood plasma and other body fluids. The measurement of components in seminal fluid specific to the secretions of each accessory gland in the male reproductive system can aid the clinician in evaluating the functional status of those glands. To evaluate prostate function, the levels of acid phosphatase, citric acid, inositol, zinc, and magnesium can be monitored. Fructose levels are used as a marker for seminal vesicle function, and carnitine levels for the epididymis (Elias- son, 19821. Alterations in the levels of these seminal plasma constituents can indicate functional alterations in the accessory organs that are related to male fertility. The excretion of drugs into semen is a phenomenon that is not well understood. In addition to influencing sperm motility and capacitation, drugs in semen have been reported to cause allergic reactions in women after intercourse. Cases of vaginitis have been reported in women whose sexual partners were taking vinblastine therapeutically (Paladine et al., 1975), pre- sumably due to excretion of this drug in the semen. Various drugs excreted into the semen of rats have caused pharmacological responses in female rats after mating (Ericsson and Baker, 19661. A number of chemical agents have been shown to be rapidly absorbed through the vaginal mucosa and to enter the systemic circulation of the female (Hardnan, 19591. Lutwak-Mann et al. (1967) found thalidomide or its metabolites in rabbit semen after giving them i4C-thalidomide by oral intake. When added to semen in vitro, thalidomide bound to rabbit sperm. Progeny of thalido- mide-treated male rabbits exhibited low birth weights, elevated incidences of birth defects, and poor neonatal survival. Similar findings have been obtained with methadone. Male rats given methadone sired offspring with low birth weights and depressed neonatal survival. High levels of methadone are excreted in the semen, giving a semen:blood ratio of 0.82 to 4.72 (Gerber and Lynn, 1976~. A number of antibiotics have also been measured in human semen in studies to develop improved treatments for venereal disease and chronic prostatitis (Malmborg et al., 19761. Thus, a spectrum of drugs are excreted in the semen and have the potential of being spermatotoxic and of entering the systemic circulation of the female after intercourse. The impact of this route of exposure on sperm and on the female has not been fully explored.

Reproductive Toxicology 69 TABLE 3-4 Chromosome Macrolesions in Germ Cells Outcome Stage of Genn Cell Embryonic Late Fetal Postnatal Type of Lesion Induction Death Death Effect Survival Chromosome break Gonial + Postgonial + Balanced chromosome rearrangement Gonial + + + a Postgonial + a Missegregation Premeiotic + b + c + ~ + e Meiotic + c + ~ + e aRoughly 50% of survivors will carry the balanced rearrangement. Some of these may show organismic disorders. All will have some chromosomally unbalanced offspring. bPossibly some monosom~es. CMonosom~es. Most autosomal trisom~es die. eSex chromosome numerical anomalies and a few autosomal trisom~es. GERM CELL MUTAGENESIS IN MALES AND FEMALES Genetic toxicants must be regarded as potential germ cell mutagens and, thus, as threatening irreversible damage to the human gene pool. Although there is as yet no documented case of chemically induced her- itable genetic disease in humans (ICPEMC, 1983b; Mohrenweiser and Neel, 1982), epidemiological studies suggest that the association found in animal studies may also apply to humans (Strobino et al., 19781. This potential association is of particular concern when chemical exposure occurs through contaminated drinking water, since exposure may be chronic throughout the reproductive years. Test Systems for Germ Cell Mutations Two types of mutagenic damage can be examined in germ cells: ma- crolesions and microlesions. A macrolesion is a microscopically visible change in either chromosome number or structure. The types of macrole- sions that have been observed, their transmissibility, and their effects on reproductivity are described in Table 3-4. Macrolesions may be detectable in the germ cells of the parental animal or human in any cell of any offspring to which they have been transmitted. Certain macrolesions ob- served in specific germ cell stages of the parent are likely to lead to germ cell death, and these will not be transmitted to offspring. For example, chromosome breakage in spermatogonia can produce a dicentric chro- mosome and an acentric fragment during spermatogonial replications,

70 DRINKING WATER AND H"LTH leading to loss of the affected chromosome and probable cell death. On the other hand, chromosome breakage in a meiotic or postmeiotic germ cell does not lead to chromosome loss until after the resulting gamete has participated in fertilization. This produces early embryonic death, i.e., an effect transmitted from the exposed parent to his or her offspring. Chromosome arrangements such as translocations or inversions are not lethal to cells, but segregation during meiosis produces chromosomally unbalanced as well as balanced gametes, all of them capable of partici- pating in fertilization. Conceptuses resulting from the former type of gamete usually die during prenatal life. Those resulting from balanced gametes survive postnatally and have roughly a 50-50 chance of being chromosomally normal, or carrying the balanced rearrangement and in turn transmitting it to some of their offspring. (They also transmit un- balanced chromosome complements to others of their offspring, which will die as early embryos). Although in the past it has been assumed that carriers of balanced rearrangements are themselves normal, there is in- creasing evidence that some balanced translocations can be associated with certain organismic disorders (Russell and Shelby, 1985; Rutledge et al., 1986). Heritable numerical chromosome anomalies result from missegregation of chromosomes during mitotic or meiotic divisions of the germ cells. Offspring can be trisomic or monosomic for the chromosome that has missegregated. If the affected chromosome is an X or Y. the offspring often survive to adulthood. Monosomies for any of the other chromosomes lead to early embryonic death, but trisomies do not kill until late fetal stages and are associated with organismic defects. In humans, trisomies for chromosomes 21 (Down's syndrome), 18, and 13 survive postnatally. Microlesions, such as frameshifts and base-pair substitutions, are chem- ical alterations in DNA that are transmissible without cell death as dom- inant or recessive mutations. Dominant mutations are expressed in the Fit generation and can be transmitted to all subsequent generations in up to 50% of the offspring. They are often recognizable, however, so that the possibility of transmitting this effect further could be reduced by knowl- edgeable genetic counseling. Recessive mutations probably represent the greatest long-term hazard in that they can be accumulated in the population in the heterozygous state (Brusick, 1978~. Mutations can and probably do occur both in coding and noncoding regions of the genome. Many of these will be unnoticed as silent mutations. Their silence is often a reflection of the accuracy and sensitivity of ana- lytical techniques, which range from morphologic analysis of offspring to a search for alterations in protein structure or function and in DNA sequence. Any change induced at a single genetic locus is a monogenic or Mendelian mutation that can be inherited as a recessive or dominant

Reproductive Toxicology 71 trait. A recessive mutation is not expressed in an individual when the normal allele is present at the same locus in the homologous chromosome. However, its presence can often be detected by molecular techniques. Other abnormalities are transmitted by polygenic inheritance in which more than one genetic locus is responsible for their expression. Examples of these are seen in some adult diseases (diabetes, coronary artery disease, hypertension), psychiatric diseases (schizophrenia, depression), and con- genital defects (cleft lip, club foot, neural tube defects, congenital heart defects) (Dean, 19831. A number of studies have been conducted to determine if heritable . .. . mutations occur in the human germ line as a result of exposure to Ionizing radiation or various chemicals. To date, studies of the offspring of large populations of atomic bomb survivors in Japan have shown no increase in the frequency of heritable genetic diseases that can be attributed to the radiation exposure of the parents, in whom significant somatic cell effects were observed. The genetic end points monitored in the progeny have included a variety of easily observable phenotypic changes. Smaller-scale studies of patients who received medical radiation for testicular cancer or lymphomas provide further evidence that genetic effects, if any, are not easily detectable in human populations, which are very heterogeneous genetically. A number of occupational and environmental exposures to various chemicals have been associated with heritable genetic defects induced in the male germ line. These results were obtained from highly controversial studies, however, and have not been uniformly replicable (reviewed by Dean, 19831. Studies on exposures to agents resulting in adult male reproductive dysfunction are summarized in Tables 3-5 and 3-6. In general, agents have been examined for their effects on fertility, testicular histology, and neuroendocrine balance. Information on the mechanism of action whereby these agents exert their effects on the reproductive system is scarce; it is inferred mostly from studies of mechanisms in other organ systems. Most of the agents listed cause reversible or short-term effects; permanent ster- ility rarely occurs under the exposure conditions described. The paucity of observations in humans (Table 3-5) is not a result of negative results but a reflection of the absence of studies. The concordance in response between humans and laboratory animals to those agents studied in both species is apparent, implying that much more experimentation in humans is needed. The findings from studies in laboratory animals do, therefore, show that germ cell mutations induced by chemical exposure can result in pas- sage of genetic diseases to the offspring. The dominant lethal, heritable translocation, and specific locus tests are the methods most commonly used in such studies to assess the mutagenic potential of chemicals in

72 so Cd As: 2 ._ to Cal ._ Cal us 2 m Cal o Cal Cd Cal Us ._ Cal an ._ V) Can ~ o ~ Cal ~ _ U. o ,4: - 1 ~ _ Ct m of, EM O Cal V, o U. ~ O O V ~ ~ ~0 O O ~ ~ ~ I: EM i Cal ~ C) ~ ~ C) ~ Dig 1i-~^~Eo · I En ~ ~ ~ ~ ~ ~ ~ ~ .' c .~ c.~: c c "c .= .~-c .' =: ¢) c ~ ca S .> ~ ~ ~ ~ ~ ;;- ;~ ~ ;^ ~ O TIC) ~ ~ CD ~ ~ C) ~ C) ~ C) ~ C) at ~ ~ ~ C) ~ ~ ¢) ~ C) C) ~ C) Ce E ~ E s:u E ~ C C E S ~ E ~ ~ ~ ~ ~ ~ ~ fi =' ~ '~ ~ ~ Cal Us Us Cal ~ Cal ~ ~ ~ ~ A= ~ ~ ~ ~ ~O 43) O3 4~ 30 4' 30 a~ O3 ~/ o ~ ~ 30 ~ 3 ~ 3 3 3 ce~ 3 :~: c<5 cx5 ce ct~ _ ~ ~ _ _ (t _ _ _ ~ ~ ;: c: ~ ~ ~ ~ ~ c: s: ~ O _ _ O _ ~ _ O ~ O O _ ~ =, p" ~ =, - 0 =, cL, - 0 ~ =, =, c~ c~ c~ ~ c~ ~ c~ c~ ~ c~ c~ c~ c~ c~ ~ c~ ~ ~ c~ ~ c~ O O O O 0 m0 0 m0 0 0 c~ ~ c~ O -2 ~ S ~ e ~ ~ e~ ° ~ U. :: O C~ ~: Ct ~S

Reproductive Toxicology 73 mammalian germ cells (Russell and Shelby, 1985~. They have been ex- tensively used in studies of radiation exposure and, to a lesser extent, chemical mutagens. The genetic risk evaluated in these tests is based on an increase in embryonic death and inherited alterations in the Fit gener- ation. Since analogous studies cannot be conducted in humans, results from these tests are extrapolated to determine the potential risk to humans. DOMINANT LETHAL TEST The dominant lethal test is the conventional entry-level test for assessing clastogenic damage to the germ cell. It can be used to measure germ cell macrolesions in both male and female animals, although typically it is conducted in males, since embryonic death following exposure of females might be the result of adverse effects on the maternal environment, rather than of genetic damage in the conceptus. Male animals are treated for 1 to 5 days and then sequentially mated with groups of control females for the duration of spermatogenesis (Green et al., 1985~. The sensitivities of different stages of spermatogenesis to mutation vary widely, in some cases by as much as 100- to 1,000-fold. Therefore, male rats are conventionally bred on a weekly basis for 10 consecutive weeks after exposure. This timing permits sampling of germ cells exposed in the mature sperm, spermatid, spermatocyte, spermatogonia, and stem cell stages of maturation. The mating index (the number of females inseminated divided by the number of females caged with males) and the fertility index (the number of females pregnant divided by the number of females in- seminated) are calculated. Females are killed at midpregnancy and ex- amined for the number of living or dead embryos, implants in the uterine horns, and corpora lutea in the ovary. The rates of preimplantation death (the number of corpora lutea divided by the number of implantation sites) and postimplantation death (the number of resorption sites divided by the number of implantation sites) are calculated. Mutation of sperm is mea- sured as increased mortality among embryos conceived with treated sperm (Green et al., 1985~. The dominant lethal assay has been the test most frequently used to measure germ cell mutations. One disadvantage is that a high spontaneous background of embryonic mortality, on the order of 7% to 10%, occurs in most rodent species, which reduces the sensitivity of the test. Only dominant mutations that are lethal to the embryo are measured, and the genetic origin of these macrolesions cannot be fully determined because only one generation the Fit is examined. Advantages of the assay are that it measures the most prevalent reproductive outcomes of germ cell damage, i.e., infertility and embryo death, and thus requires fewer animals than other assays that may be more sensitive but measure rarer outcomes.

74 o Cal o x no Cq Cal Cal s°- - Ct Cal is: Cal - .= ._ Cal ._ o ~ Cal 4_ Cal _ 1 ~ ,~ no m EM Cal 4) - V) o C) ca C) ·C~ as Cal ~0 ~ ~ ~ _ ~ _ ~ ~ ~ ~ 3 ~ ta o ~ C ~ ~ _ C ~ C [_ ~ , C >, ~0 C) ~ ~ C ~ . ~ , ~ ~ ~ ~ ~ ~ ,, ~ E hi, ~ ~ ~ ~ , 5 E ~ ~ C) Cal Cal ~ , C) E A.= o ,,, c _ o C ~ E ' o C) ~ Ct O Ce O Ce CtS ~ ~ Ce P~ , ~ ~ c o~ C ~ .~ m=C; ;n ' V ~ ~ 5 ~ ~ ~ ~ ~ C ·= o 1 0 ~ SO a c ~ U ,y ~ ~ .8 ~ ~o =,, ~ ~o Y a

75 ~ ~ ~ ~ ~ ON ~ ~ _ X X X ¢,, "' o Y V) ~ ~ C ~ e, ~ ^, ~ ~ ~ c O D U ~ ~ _ ~ D ~ - C' ~ C'2 ·~ ~ A C.) C) ~ ~ ~ ·— ~ ~ it= ~ ~ ~ O ~ ~ O ·— ~ ~ ~ ~ ~ —~ ~ ~ ~ ~ ~ ~ ~ ~ fi ~ o o ~ ~ ~ ~ ~ s ~ ~ ~ ~ 1 1 . . ~ ~ s -5 . s ~ Ce Cl ~ Cal ~ .= c — c ° ~ 3 0 ~ 0 ~ 0 ~ ~ ~ ~ 0 ~o ~ ~ ~ ~ ~ m ~ ~ ~ ~ C) C) ~ _ _ ~ _ ~ ~ C) Co · ~ C-) ° ;> ~ ;> ~ ;~ ~ O O O O ~ C) C) 1 1 o~ ~ ~ ~ ~ ~ ~ ~ ;. ~ I ~o 0 Ce Ct ^ ^ ~ 0 ) —° ~ =~) ~ ~ ~ ~ ·- ~ 0 o =^ E E ~ ~ o V, E _ ~ C~ _ C~ ~ ~ C~ _ ~ ~ ° ~ o o~ = == ~ ~ ~ 0 ~ O

76 C) C) C) C) Cal C) C40 LO o C) . C) C) V) s s m U. ~ =0 U. ·~ O —v X O Cal lo: y ~ _ 2 ~ ~ go e a ~ ~ 2 >4 ~ ~ ~ ~ I Cal ~ C.) ~ , i, u — ~ ' 2 ' D ' " B ' ' 3 ~ ~~ ~ a 5: ~ ~ ~ ~ ~ 1 1 ~ ~ Ct . . V 0C Ce ~ ~ O ° a., (u ° . Cal Ce . C) Cal Ct Ct X ~ ~ ~ ~ X ~ C) ~ 0= ° - ~ 0= ~ 0" ~ 0= 0 0 C,3 . O .s O ~ O O O Tic a ~ a ~ ~ ~ ~ et ~ 8 ~ ~ ~ ~o X ~ ~ ~ I =, o t o _ ~ · ~ _ o ~ _ o — ~ V ^ o o V o oc ~ ~ ~ ~ ~ ° o ~ o ~ ~ E ° o ~ ~ =- ~ E v ~ ~ ~ ~ :;: ~ ~ o ~ _ ~ o ·E :s ~ o o o 13 ~ E ,Z ° ~: a O Z O

77 _ ~ ~ _ _ ~ ~ ~ _ Y — Y ~ ~ . ~ C of ~ y oN ~ ~ ° C ~ b ~ ' i, ~ ', ~ ~ ~ -, ~ ~ . ~ ~ j c ' - ' ~ °, Cal ;~l, · .^ C) Us _ a,~ C~ te . _ ~ -_ ~ ~ C) ;^ _ _ C > ~ o _ _ Ce ~ ~ ;~ Ce _ _ . _ Ce Ce ce .O ·O ·o ~ ~ ~ .O .O .O ~ ,= ~ ~ o o ~ ·C~ ~ o o o ·— (L) ~ ~ O O ~ ~ ~ ~ ~ .= o ~ _. sin D _ ~ ID ~ ~ O ~ ~ ~ ~ C<! Cal Cal i_ Ce Ct Cal Ce ~ - 8 ~ =~t ~ ~ O ~ ~ 3 =~ C~ ~ ~ O U~ C~ S: . o E == ~ = ' =~ °~ ~ ~ j ~ y ~ ~ Y C) o C~ o C~ ~ ._ _ ~ I:: o ._ C~ . C) . ~ ~ _ ~o ~ ~ .5: ~ =.= _I ·_ ~ a~ ~ ~ ~ . c: ~ ~ Co~ ~ . _ o ^ ~ 0 = = o o ~ = C ~ U, ._ ~s ~ . . C) ~ C~ __ ., =. .

78 DRINKING WATER AND H"LTH TABLE 3-7 Cell Divisions in the Formation of Germ Cellsa Females No. of Cell Divisions Males No. of Cell Divisions Before Maximum Before To Division No. of Germ Cell Dunng No. of Germ Cell Mature Cycle Stem Species Formation Oogenesis Oocytes Formation Spenn (days) Cells Humans Unknown 21 3.4 x 106 Unknown 380-540b 16 5 x 108 Mice 10-13 13 1x104 10-13 40-80 ~ 1 x106 aAdapted from Lyon et al., 1979. bBetween 28 and 35 years of age. In addition, cell stage specificity can be detected through the sequential mating design. Overall, the sensitivity of the dominant lethal assay is such that germ cell mutagens can be identified with certainty if large numbers of animals are used (50 rodents per group) and if the compound is a potent mutagen (Green et al., 19851. The increased sensitivity of male germ cells to dominant lethal or other types of mutations is based on the fact that many mutations occur pre- dominantly in replicating cells. As indicated in Table 3-7, the stem cell spermatogonia of male mammals continue to replicate throughout the breeding life, which confers susceptibility to replication-dependent mu- tations. In female germ cells, as noted earlier, mitosis ceases at the fetal stage, and the oocyte remains in a resting stage of meiotic prophase from birth until puberty. Thus, since the oocyte spends a greater proportion of time in a nonreplicating state, the incidence of replication-dependent mu- tations will be lower in females than in males (Lyon et al., 19791. A dominant lethal assay in females would be based on a similar design, except that exposed females would be mated with control males at weekly intervals up to 7 weeks after exposure to sample the different stages of oocyte maturation. Dominant lethal tests are not routinely performed in females, however, because it is difficult to discriminate between nonspe- cific systemic toxicity (uterine toxicity resulting in failure to implant) and true genetic alterations in the germ cells (Badr and Badr, 1974~. The sex chromosome loss test can be used to detect clastogenic damage in female germ cells. This test utilizes genetic markers on the X chromosome, so that it can be identified by a marker phenotype rather than by a reduced number of survivors as in the dominant lethal test (Russell and Shelby, 1985~. The presumed XO-females detected by the marker phenotype can be verified cytogenetically by the presence of only 39 chromosomes. The end point observed in the dominant lethal test is embryonic death, presumably due to chromosome aberrations in sperm. Thus, a tremendous

Reproductive Toxicology 79 inherent selectivity is involved in the identification of mutagenic events to be monitored; i.e., there is a focus only on aberrant germ cells that survive the meiotic sieve, that are competitive with normal germ cells, and that result in early embryonic death. Somatic cell assessments (meta- phase analysis) can detect a wider spectrum of chromosome-damaging events, including the type that could lead to a dominant lethal event. In a study comparing the two types of assays for 76 compounds, the total incidence of concordance (positive-positive or negative-negative) between somatic and germ cell assays was 75%. Eighteen compounds gave a positive response in one or more somatic cell assays but were negative in germ cell assessment; no compounds were positive in germ cell assays but negative in somatic cells (Holder, 19821. In a comparison of available published data, the International Commission for Protection against En- vironmental Mutagens and Carcinogens (ICPEMC, 1983) found 88% con- cordance between germ cell assays (mouse specific locus, heritable translocation, and rodent dominant lethal tests) and cytogenetic in vitro tests, 83% concordance between germ cell assays and sister chromatic exchange tests, and 64% concordance between germ cell assays and the Ames test. This leaves open the question of whether germ cells are equipped with more effective mechanisms for removing mutagenic lesions. These mech- anisms may take the form of DNA packaging, DNA repair capability, or the meiotic sieve (Holder, 1982), which confer selective disadvantage to genetically abnormal cells. Each of these would be expected to reduce the mutagenic response of germ cells relative to that of somatic cells. The important application of these assays is to determine whether an agent that causes genetic toxicity in somatic cells also causes heritable mutations in germ cells, or if testicular lesions or reductions in male fertility identified in reproductive toxicology studies are due to germ cell mutations. HERITABLE TRANSLOCATION TEST The heritable translocation test is also used frequently to assess germ cell mutations. Male mice are exposed and mated to control females, which are allowed to deliver their litters. The Fit male progeny are reared to reproductive maturity, mated to control females, and then examined to estimate their sterility or semisterility. Criteria for Fit male sterility include an absence of embryos in females with vaginal plugs; for a classification of semisterility, there must be a decrease in the number of live embryos per litter relative to controls. The germ cells of males exhibiting varying degrees of sterility are analyzed cytogenetically at meiotic metaphase for translocations (Brusick, 1978; Generoso et al., 19801.

80 DRINKING WATER AND HEALTH TABLE 3-8 Relative Inducibility of Dominant Lethal Mutations and Heritable Translocations in Male Gerrn Cellsa Mutagen (and Dose) Germ Cell Stage Percentage of Dominant Lethal Frequency Percentage of Hentable Translocation Frequency X ray Spermatozoa (700 reds) and spennatids 67 27.0 Ethyl methane sulfonate Spermatozoa 69 32.0 (200 mg/kgb) Tnethylenemelam~ne Spennatids 75 29.0 (0.2 mg/kg) Isopropyl methane Spennatozoa 82 2.7 sulfonate (75 mg/kg) Spermatids 58 0.4 Benzo[a]pyrene Spermatozoa 27-50 0.18 (500 mg/kg) aAdapted from Generoso, 1982. bPer kilogram of body weight in each case. Carriers of translocations have reduced fertility a criterion used to select Fit progeny for cytogenetic analysis. Complete sterility is caused by certain types of translocations (X-autosome and c/t type). Transloca- tions are cytogenetically observed in meiotic cells at metaphase I of males and in either Fit males or male offspring of Fit females (Russell and Shelby, 1985~. The background rate of heritable translocations is relatively low: max- imum of one carrier per 1,000 control animals (Generoso, 1973~. In the heritable translocation test, only one type of chromosome aberration is measured a balanced translocation, which can occur only through ge- netic alteration of the germ line but not through nonspecific cytotoxicity. More animals are needed for this test than for the dominant lethal assay because the control rate is very low. At least 500 male progeny should be tested per dose group (ICPEMC, 19831. The levels of dominant lethal and heritable translocation mutations induced in male germ cells by x rays, by several direct-acting alkylating agents' and by benzotalpyrene are listed in Table 3-8. For a number of mutagens, including ethyl methane sulfonate, triethylenemelamine, and x rays, there is a positive correlation between the induction of dominant lethal mutations and the occurrence of heritable translocations. The fre- quency of heritable translocations induced by the maximum tolerated dose of any of these three agents is about 30% among live progeny in each case. This figure is reached when the frequency of dominant lethal mu- tation is approximately 60~o or more. Studies of isopropyl methane sul-

Reproductive Toxicology B! fonate and benzoLa~pyrene have indicated, however, that these compounds differ markedly in that few or no heritable translocations were induced, even at doses that caused high levels of dominant lethal mutations. Thus, some compounds can induce both types of genetic changes, whereas others induce only one (Generoso, 19821. SPECIFIC LOCUS TEST As discussed above, the dominant lethal and heritable translocation assays measure dominant mutations that are macrolesions in the chro- mosomes of germ cells. In the specific locus test, recessive point muta- tions, or microlesions, can be measured, and mutations to recessive alleles at a small number of specified loci can be detected in the first generation. The test method consists of mating treated, wild-type male mice to a test stock of untreated females that are homozygous for recessive markers at several loci. When there is no mutation at these loci in the treated male germ cells, none of the recessive markers contributed by the mother would be expressed in the Fit generation because they would be in the hetero- zygous state. In the presence of mutation, alterations in coat color, eye color, and other visibly apparent morphological features occur (W. L. Russell, 19511. To measure rare events occurring at only a few targets, one must ex- amine a large population in order to observe a significant change in mu- tation frequency. Since only seven loci are available for monitoring in the specific locus tests with mice, many animals must be used. Thus, this test can be used to measure frequencies of recessive mutations but not to evaluate large numbers of chemicals. Several investigators have developed methods to detect electrophoretic variants of proteins as measures of point mutations (Johnson and Lewis, 1981; Valcovic and Malting, 1975~. These methods increase the total number of loci assessed in the specific locus test and thus require obser- vation of fewer Fit animals, although the time required for examination of each Fit animal is considerably longer. Combining the specific locus test with the scoring for cataracts in Fit progeny quadruples the number of loci that can be scored, since dominant mutations at approximately 20 loci can produce cataracts (Ehling, 1980~. Examination of Fat skeletons can detect dominant mutations at over 100 loci (Selby, 19831. The specific locus test can also be applied to females to study recessive mutations in oocytes. In this case, the treated, wild-type parent would be female, and the untreated parent would be a homozygous recessive male. Although relatively few studies have been conducted in females, it appears that specific locus mutations, like macrolesions, are predominantly in- duced in the first 7 weeks after treatment. Experiments with the mutagens

82 DRINKING WATER AND H"LTH procarbazine (Ehling, 1980) and ethylnitrosourea (Russell et al., 1979) indicate that oocytes may be less sensitive than spermatogonia for induc- tion by chemicals of specific locus mutations. RISK ASSESSMENT IN REPRODUCTIVE TOXICOLOGY In light of the 15% infertility rate among married couples, it is surprising that risk assessment is infrequently based on reproductive toxicity data and that safety testing programs do not always include measurements of adverse effects on reproduction. For example, the Toxic Substances Con- trol Act, P.L. 94-469, identifies four progressive levels of testing, de- pending on the extent, frequency, and nature of chemical use, but information on reproductive effects is not required until level IV, when the product is already on the market. At level II, the most comprehensive test is a subchronic study designed to predict effects on humans exposed during production or industrial use. As demonstrated by Koeter (1983), there is concern that the acceptable daily exposure level estimated from the subchronic study may be too high with respect to effects on reproductive function. Koeter evaluated toxicity data on 37 compounds tested both in subchronic studies and in one or more reproductive toxicity studies to determine the impact of the latter in identifying the NOEL and the LOEL. For the LOEL, reproductive toxicity studies were more sensitive than the subchronic studies for 35% (13) of the compounds. They were equally sensitive for another 35% (13~. For 65% (17) of these 26 compounds, effects related to reproduction and development were factors in the determination of the LOEL. For 30% (11) of the total number of compounds tested, the reproductive toxicity studies were less sensitive than the subchronic studies. For 8 of the 37 compounds tested (>20%), fertility or reproduction end points were the most sensitive, and this solely determined the LOEL. Koeter (1983) con- cluded that reproductive function is highly sensitive to impairment and should be examined at earlier stages of safety testing. Even when data from reproductive toxicity studies are available for use in risk assessment for a particular compound, there is much confusion about how to apply results from animal studies to humans. The confusion is due partly to the fact that considerably less is understood about the underlying events leading to reproductive toxicity than is known about corresponding processes in mutagenesis and carcinogenesis. There are currently no agreed-upon standard quantitative methods for cross-species extrapolation. Some of the conditions under which reproductive toxicity data from animal studies can be applied to predict human risk and the methods currently used are examined in Chapter 8.

Reproductive Toxicology 83 APPENDIX: DETAILS OF TEST PROTOCOLS FOR REPRODUCTIVE TOXICOLOGY TESTING MULTIGENERATION TEST The classic multigeneration study encompasses three generations. Fig- ure 3-3 provides the basic outline of the procedure but does not encompass all possible variations. First, weanling animals (usually rats or mice, 30 to 40 days of age) are randomly assigned to control or test groups, and the test compound is administered continuously in the diet or drinking water or via inhalation for the duration of the study. After at least 60 days of exposure, or when these parental, Fo generation animals are 100 to 120 days old, males and females within a treatment group are mated to produce the MA generation. Within 24 hours after birth, each pup is weighed, sexed, and examined for gross abnormalities. On day 4, litters of more than 10 pups are culled to 10 or some smaller constant number by random selection. Separate weights for male and female offspring and their survival are recorded at birth and on days 4, 7, 14, and 21 of life. After weaning (21 days of age), MA offspring are sacrificed and autopsied to detect internal abnormalities. After weaning of MA pups, 1 to 2 weeks are allowed to pass before the second mating of the parental generation takes place for production of the FIB offspring. The same procedure is followed for observation and data collection, except that a sufficient number of animals are selected at weaning to serve as the parents of the F2 generation. Males and females are randomly paired within a treatment group, but brother-sister matings are avoided. FIB weanlings remaining after selection of parents are sac- rificed and autopsied. The F2A and F2B litters are generated in the same manner as the MA and FIB litters, and the same observations are made from birth to weaning. The F3 generation litters are produced from parents selected from the F2B litters. The pups are weighed, sexed, and weaned in the same manner, but at sacrifice, tissues from at least 10 animals are preserved for histological examination. Complete necropsies are per- formed on all animals that die spontaneously during the study. At routine necropsy, observations are made on all major organ systems, and weights are obtained. Although the protocol described in Figure 3-3 still serves as a model for multigeneration studies, several modifications are often conventionally applied (Collins, 1978a,b). A teratology study can be added by mating MA offspring instead of sacrificing them at weaning. Likewise, the F3B offspring can be mated and their progeny used for teratology studies. After resting, the parental animals may be mated to produce an F~C or an FED

84 DRINKING WATER AND H"LTH generation to study the effects of exposure on reproduction in aging an- imals. The F2A generation can be maintained for chronic studies to ex- amine the carcinogenic effect of the exposure. According to current recommendations, three dose levels and one con- trol group are used. The highest dose level may be a multiple of the human exposure level or it may be one-tenth of the test species' LDso. The low dose is determined from subchronic or chronic toxicity studies and is expected to be a no-observable-effect level. The intermediate dose level is logarithmically spaced between the high and low levels. When rodents are used, enough females should be started at each dose level so that at least 20 pregnant females per dose level for each generation are obtained. All doses are reported as mg/kg body weight per day, which necessitates measurement of food and water intake. Rodent species are used most frequently, so that a three-generation study can be conducted within 20 months. Although other species are acceptable, rats and mice are the species of choice because they are inexpensive and there is a large amount of historical data on testing in these animals. The following indices are calculated from data derived from multigen- eration studies: · The fertility index: (the number of pregnancies/the number of matings) x 100. This index represents the percentage of matings that result in pregnancies. · The gestation index represents the number of live fetuses per litter or the number of live-born per total born. · The sex ratio is calculated at birth and at 4, 7, 14, and 21 days of age. · The viability index: (the number of pups alive at 4 days/the number of pups born alive) x 100. The number of pups remaining after culling (10 or 8) is used in the denominator on days 7, 14, and 21 of age. · The weaning index: (the number of pups alive at 21 days/the number of pups maintained at 4 days) x 100. · The growth index represents the average weights of male and female offspring at birth and at 4, 7, 14, and 21 days of age. To obtain results that are statistically sound, it is critical that at the beginning of the study, parental animals be randomly assigned to control and treatment groups. This procedure should also be followed in culling litters to a constant size and in selecting parental animals from each generation for mating. A number of variants of this procedure, based on regulatory and inter- national guidelines, have been described by Palmer (19811.. For further

Reproductive Toxicology 85 elaboration on multigeneration tests, consult the review prepared by Col- lins (1978a,b). FERTILITY ASSESSMENT BY CONTINUOUS BREEDING The National Toxicology Program is validating its new reproductive toxicity test, Fertility Assessment by Continuous Breeding (FACB) (Lamb et al., 19841. This test provides an alternative to multigeneration studies, producing similar comprehensive reproductive toxicity data at a lower cost and within a shorter time. In addition, it provides for multiple, sequential breeding (up to five mating cycles) of one parental generation, which permits identification of latent toxic effects on immature germ cells (resting follicles, spermatogonia). During the first task, a vehicle control and five dose groups (eight/sex/ group) of 8-week-old CD-1 mice are repeatedly dosed for 14 days, during which time dose levels for the second task are identified. The high dose, or maximal tolerated dose, may produce some significant signs of toxicity, but it should not suppress body weight gain to levels more than 10% below that of the controls and it should allow 90% or greater survival. The intermediate dose should produce minimal or no toxic effects, and the low dose should ideally be a no-effect level. An LD~o calculation is made, clinical signs are observed, body weight gain during the 14-day period is measured, and either food or water consumption is measured, depending on the route of exposure. The continuous breeding phase occurs in task 2, which consists of a vehicle control group (40/sex) and Tree dose groups (20/sex/group). Eleven- week-old male and female CD- 1 mice are exposed to the test agent during a 7-day premating period, during which time clinical signs are observed and measurements are made of weight changes and water consumption, if water is the route of exposure. The mice are then randomly paired (1:1) within each group and housed together for 98 days (14 weeks). Newborn litters are evaluated and immediately sacrificed. Exposure of the parents is continuous during the 98-day cohabitation period and the subsequent 21-day period of segregation. Observations made throughout these two periods include number of litters produced per breeding pair; number and percentage of fertile pairs; number and percentage of live newborns per litter; mean body weight of newborns; length of time between litters; parental body weights at weeks 2, 5, 9, 13, and 18; and water consumption at weeks 1, 2, 5, 9, 13, and 18. At the end of the 98-day cohabitation period, data are evaluated to determine whether to proceed to task 3 or task 4. The final litters born to the control and high-dose pairs during the segregation period may be weaned, reared to maturity, and evaluated for

86 DRINKING WATER AND H"LTH reproductive performance in task 4. Chemical exposure of the offspring is continued. If significant effects on reproductive performance (fertility or litter size) are found in task 2, then the affected sex is identified in task 3. Task 3 is a continuation of the 119-day mating trial. On day 120, the following cross-matings (20/sex) are performed: high-dose male with control fe- males, control males with high-dose females, and control females with control males. All controls are assigned new cagemates. Treatment is discontinued during the 7-day cohabitation period; after mating has oc- curred, parental animals are placed back on their original treatments for 21 days. To determine reproductive performance, the following observations are made and evaluated: vaginal cytology in females not mating, percentage of fertile pairs, mean litter size, number and percentage of live newborns, and mean litter weight. If males are identified as the affected sex, they are killed and bled by cardiac puncture. The following observations are then made and evaluated: body weight, liver weight, fixed brain weight, fixed pituitary weight, right testis weight, ventral prostate weight, seminal vesicle weight, right epi- didymal weight, number of cauda epididymal sperm per milligram of tissue, and sperm morphology and motility. If females are the affected sex, they are also killed and bled by cardiac puncture. The following observations are then made and evaluated: body weight, liver weight, fixed brain weight, fixed pituitary weight, and weight of the reproductive tract (ovaries, oviducts, uterus, and upper half of vagina). Task 4 is a continuation of the 119-day mating trial. If no adverse effects on fertility are found during continuous breeding, offspring ob- tained between days 98 and 119 from high-dose and control groups are reared to reproductive maturity. At 70 + 10 days of age, 20 male offspring and 20 female offspring from the same treatment groups are paired. Chem- ical exposures continue throughout task 4, until litters are born. If possible, all high-dose and control task 2 litters are sampled and sibling matings are avoided. The pairs are housed together for 7 days, or less if a copulatory plug is observed, and treatment is continued during cohabitation. Reproductive performance is determined by evaluating the litters. If it is found to be adversely affected in the high-dose pairs, then task 3 can be carried out with these same mice to determine whether one or both sexes are affected. In this instance, the high-dose animals are bred to control animals of the opposite sex. If treatment has an adverse effect on fertility or reproductive perfor- mance in parental mice (task 3) or in their offspring (task 4), radioim- munoassays for reproductive hormones can be performed on plasma collected at necropsy. Hormones evaluated in the male are testosterone, FSH, and

Reproductive Toxicology 87 LH; in females, they are estradiol-17 (or total estrogens), prolactin, FSH, and LH. The following end points are evaluated in task 4: number and percentage of fertile pairs, litter size, litter weight, and number and percentage of live-born animals. SEGMENT ~ STUDIES: GENERAL FERTILITY AND REPRODUCTIVE PERFORMANCE Segment I studies can be performed with exposures of males alone, females alone, or both, depending on the background information avail- able. According to Food and Drug Administration guidelines, 20 male rodents are exposed for 70 days to cover the duration of spermatogenesis. A minimum of 20 females are exposed for 14 days to cover three estrous cycles. Treated animals are mated, and exposure of inseminated females is continued throughout pregnancy and lactation. Either at midpregnancy or just before delivery, half the females per test group (at least 10) are killed, and their uterine contents are examined for preimplantation and postimplantation death. The remainder (a minimum of 10) deliver spon- taneously and wean their offspring. Weanlings are killed and autopsied for gross observations and examination of viscera. Common variations in Segment I testing are to mate treated males and females with untreated counterparts to determine sex-specific effects on fertility and pregnancy outcome. Also, the midpregnancy sacrifice can be delayed to day 20 so that fetuses can be examined for malformations. Recommendations have been made that weanlings be spared and raised to maturity to detect latent effects on behavior, physiological development, and reproductive capacity (Collins, 1978b). The following observations are made in Segment I studies: · Preimplantation death: number of corpora lutea in ovaries/number of implantation sites in uterine horns. Most control dams have a slight excess of corpora lutea, which appear as highly vascularized bulges on the surface of the ovary. · Postimplantation death: (the number of resorption sites in the uterus/ number of implantation sites) x 100. One to two resorption sites are conventionally found in control litters. Statistical- procedures based on the assumption of a normal distribution are not preferable for Segment I studies in that the occurrence of embryo death in litters more closely follows a Poisson distribution. Variance anal- ysis with nonparametric tests, such as the Mann-Whitney U-test for com- parison of two groups and the Kruskal-Wallis test for comparison of more than two groups, are appropriate (Gaylor, 19781. The fertility of the parental animals and viability of the offspring can be expressed in the

88 DRINKING WATER AND H"LTH manner used for findings of the multigeneration test, i.e., fertility index, gestation index, sex ratio, viability index, weaning index, and growth index. Experience with this test procedure has led some investigators (Palmer, 1978) to conclude that the most common outcome observed with positive agents is infertility. Thus, mating success is a critical factor that should be carefully assessed. When nonpregnancy is encountered, additional stud- ies that include examination of paternal contribution at lower exposure levels should be performed to ensure that infertility does not mask effects such as delayed implantation or later developmental toxicities. With 10 animals per treatment group for each observation (10 for midpregnancy examination, 10 litters for weanling autopsies), only highly active repro- ductive toxicants can be detected. A 40% to 50% difference between control and treatment groups must be obtained to achieve statistical sig- nificance at the 0.05 level by the Fischer exact test. Thus, compounds with borderline toxicity are not likely to be detected by this test unless large numbers of animals are used (Palmer, 1978~. For Segment I studies, dosages should be extrapolated from subchronic studies in adult animals. The choice of dose should be made with an understanding of the complex relationship between maternal toxicity and embryonic toxicity. Adverse effects on the embryo may occur secondarily and nonspecifically from adverse effects on the mother. The highest dose used should not cause frank maternal effects (e.g., sedation, hyperactivity, and respiratory distress) or death. Treatment should be regulated so that no more than a 10% reduction in maternal weight gains compared with controls occurs in the highest dose group. Dosages causing minimal signs of maternal toxicity, such as increased liver weight, can be selected from subchronic studies in adult animals. The choice of low and intermediate doses should be made only after a suitable high dose has been selected. The lowest dose should cause therapeutic or physiological effects similar to those intended for humans or, in the case of environmental agents, should lead to measurable tissue levels or enzyme induction without toxic effects. The intermediate dose should lie logarithmically between the low- est and highest level. TIME OF VAGINAL OPENING IN THE RAT PUP This test can be used to identify estrogenic activity of test agents such as the estrogen agonists DES, clomiphene, and tamoxifen, which cause premature vaginal opening in the rat. First, newborn litters are culled to a constant size at birth (not less than six pups). The neonates are injected on days 1, 3, and 5 of postnatal life, and the time of vaginal opening is

Reproductive Toxicology 139 noted. Vaginal opening should occur by day 7 in controls (Clark and McCormack, 1980~. UTERINE EPITHELIAL HYPERTROPHY Rat pups are treated as described above, and uteri are removed, weighed, and processed for histological evaluation on days 7 to 10 after birth. Compounds such as Kepone, DES, DDT, and zearlenone have been shown to stimulate uterine growth and cause epithelial hyperplasia (Clark and Peck, 19791. PRIMORDIAL OOCYTE DESTRUCTION IN JUVENILE MICE Female mice are injected intraperitoneally at 7 to 21 days of age with six graded doses of the test agent. There are five animals per treatment group. The highest dose should be at the adult LDso for the agent. The day after the last dose, ovaries are removed, fixed, serially sectioned (5 ~m), and stained. Oocytes and follicles are classified according to the procedure of Pedersen and Peters (1968) by histological examination of every twentieth section. The percentage of primordial follicles surviving relative to controls is plotted against dose, resulting in sigmoidal dose- response curves for positive agents. Quantitative comparisons between agents can be made in terms of the oocyte toxicity index. This is the ratio of the LDso for the juvenile mouse to the oocyte LDso (Dobson and Felton, 1983~. SPERM PRODUCTION RATES IN RATS One testis (freed of the epididymis and spermatic cord) from each male is weighed. Absolute testis weight should be reported rather than weight per gram of body weight. In normal adult males, testis weight and body weight are independent variables (r = 0.24 for 125-day-old rats; Robb et al., 19781. Relative changes in sperm production rates can be established by direct comparison of testicular spermatid reserves, i.e., the number of spermatids per testis, per pair of testes, or per unit weight of testicular parenchyma. The technique for determining testicular spermatid reserves involves ho- mogenization of testicular tissue and subsequent enumeration of elongated spermatids by hemacytometry. Because the time required for these cells to be transformed into sperm is relatively constant for members of the same species, relative rates of sperm production can be determined and the rates in control and experimental animals can be compared. These procedures are much simpler and less tedious than histological methods,

90 DRINKING WATER AND H"LTH and a larger sample of the testis can be analyzed. Since elongated sper- matids are the last cells to be formed in the testis, alterations in sper- matogenesis at any stage should ultimately be reflected in changes in spermatid reserves (Berndtson, 19771. Also, ejaculation frequency does not influence daily testicular sperm production (Amann, 19821. Rat testicular parenchyma is homogenized in 20 to 30 rnl of a cold 0.9% sodium chloride solution containing 0.01% Triton X-100. The tissue is homogenized at low speed in a laboratory blender for 2 minutes, and spermatids are concentrated by low-speed centrifugation. This procedure destroys the nuclei of somatic cells and all germinal cells except elongated spermatid nuclei with shapes characteristic of step 17 to 19 spermatids (Clermont, 19721. Resistant spermatid nuclei are counted in a hemacy- tometer (at least six chambers per sample), and counts are expressed per testis and per milligram of parenchyma. These values are divided by 6.10 days (the time required for spermatids to form spermatozoa in the rat) to obtain the daily sperm production rate. In an alternative procedure, homogenization-resistant spermatids are counted with a Coulter counter. This requires sonication of the sample after homogenization and filtration through a screen for removal of residual debris prior to counting (Mien et al., 19771. Likewise, linear gradients of 22% to 36% Percoll can be used to obtain highly purified preparations of rat spermatids (Meistrich et al., 19791. ASSESSMENT OF EPIDIDYMAL SPERM NUMBERS AND TRANSIT T! M ES Epididymal sperm numbers can be determined from fresh or frozen samples. The epididymis is weighed and divided into cauda and corpus plus caput sections. Each section is minced and homogenized separately in 20 ml of sodium chloride-Triton X-100 solution, and the number of spermatozoa in each section is enumerated in a hemacytometer, as de- scribed above. Epididymal transit times are calculated by dividing the total number of spermatozoa per whole epididymis by the sperm production rate of the associated testis on a per animal basis (Robb et al., 19781. ASSESSMENT OF SPERM MOT! LlTY The cauda epididymis is excised immediately after sacrifice and placed in a petri dish containing 10 ml of Dulbecco's calcium- and magnesium- free phosphate-buffered saline and a 10-mg/ml concentration of bovine serum albumin. The solution is kept at 37°C. The tissue is finely minced with scalpels for approximately 1 minute. Then the dish is placed in a 37°C incubator for 15 minutes before assessing sperm movement. After

Reproductive Toxicology 91 incubation, a 15-~1 aliquot of the sperm suspension is placed on a mi- croscope slide and a cover slip is added. At least nine different microscopic fields are then videotaped for 10 to 20 seconds using phase-contrast mi- croscopy and a 16x objective. The videotapes are later analyzed for percentage of motile cells (50 cells per animal) and straight-line swimming speeds (25 cells per animal). The video system and methods for analyzing the tapes have been described by Katz and Overstreet (19811. The sperm specimen must be held at 37°C during taping in an air-curtain incubator. HISTOLOGICAL EVALUATION OF THE TESTIS The choice between histological or homogenization techniques for quan- tifying relative changes in sperm production depends upon the nature and severity of anticipated effects. One limitation of homogenization tech- niques is that they are not useful in establishing the relative severity of effects on earlier stages of spermatogenesis. Also, the recovery of sper- matogonia, spermatocytes, and early spermatid stages cannot be evaluated in the absence of elongated spermatids. Neither histological nor homog- enization techniques are clearly superior in all instances. Consequently, it is advisable to process testicular tissue so that both procedures can be utilized. A number of excellent reviews have been published on procedures for preparing slides and carrying out histological analysis of the testis. Lamb and Chapin (1985) have reviewed in detail the different methods of fixation and embedment for rat testicular tissue. The most common procedures, i.e., fixation in 10% neutral buffered formalin followed by paraffin embed- ding, produces highly unsatisfactory results that preclude even qualitative evaluation of the testis. A major improvement occurs with fixation in Bouin's, Zenker's, or Helly's fixatives followed by paraffin embedment. When tissues are embedded in glycol methacrylate, formalin becomes a suitable fixative for the testis (although glutaraldehyde is better) and su- perior resolution of cellular detail is obtained. Moreover, the entire cross section of the rat testis can be obtained in one block, permitting exami- nation of up to 400 tubules. Coupled with the range of usable stains and histochemical techniques, the water-based plastic embedding media are preferable for most light microscopic studies. In histological procedures, cross sections of seminiferous tubules are considered to be representative of the testis as a whole. This is valid provided all seminiferous tubules are affected uniformly and the histo- logical sections are separated by a reasonable distance. The seminiferous tubules of most mammals are highly convoluted, and cross sections of the same tubule may appear in a section of testicular tissue from a single location. Consequently, at least five tissue sections separated from each .

92 DRINKING WATER AND H"LTH other by a minimum distance of 1,000 ,um should be used for histological evaluations (Amann, 19821. Initial qualitative appraisal of the testis should take into account the appearance of Leydig cells, the occurrence of lymphoid cell or macrophage infiltration, the presence of each stage of germ cell (sperrnatogonia, sper- matocytes, spermatids, sperm) in seminiferous tubules, and the presence of large numbers of degenerating, multinucleate, or abnormal cells (EPA- ORNL, 19824. In addition, the percentage of tubular cross sections with no evidence of sperrnatogenesis should be scored during brief examination of 250 cross sections per testis magnified 100 or 400 times. The percentage of tubules with spermatids lining the lumen should be approximately 30% for normal rats, and the percentage of those with no evidence of sper- matogenesis should be less than 5% (Amann, 19821. Such examination can be performed 2 to 7 days after acute treatment or six cycles after initiation of chronic treatment. Most likely, these qualitative assessments will be highly correlated with measurements of daily sperm production rates. Different types of information are obtained from observations of long- term and short-term exposures to (or sacnfice after administration of) an agent. In short-term studies, several hours or a few days are usually necessary to observe results after single or multiple exposures to an agent. The initial morphological changes are the primary indicators of a defect, and they provide the best clues to the mechanism of damage. Long-term administration of agents results in end-stage effects or in maturation- depletion effects, in that no further deterioration of the testis occurs with time. Expenments designed to show end-stage effects best reveal the extent to which maturation of germ cells has been arrested. Detailed descriptions of methods for precise staging of the seminiferous epithelium of the rat have been described by Leblond and Clermont (1952a,b), Berndtson (1977), and Russell (19831. REFERENCES Aafjes, J. H., J. M. Vets, and E. Schenck. 1980. Fertility of rats with artificial oligo- zoospermia. J. Reprod. Fertil. 58:345-351. Adler, I.-D., and J. G. Brewen. 1982. Effects of chemicals on chromosome-aberration production in male and female germ cells. Pp. 1-35 in F. J. de Serres and A. Hollaender, eds. Chemical Mutagens. Principles and Methods for Their Detection. Vol. 7. Plenum, New York. Albert, P. S., R. G. Salerno, S. N. Kapoor, and J. E. Davis. 1975. The nitrofurans as sperm-immobilizing agents, their tissue toxicity, and their clinical application in vasec- tomy. Fertil. Steril. 26:485-491. Amann, R. P. 1982. Use of animal models for detecting specific alterations in reproduction. Fund. Appl. Toxicol. 2:13-26.

Reproductive Toxicology 93 Amann, R. P., L. Johnson, D. L. Thompson, Jr., and B. W. Pickett. 1976. Daily sper- matozoal production, epididymal spermatozoa! reserves and transit time of spermatozoa through the epididymis of the rhesus monkey. Biol. Reprod. 15:586-592. Aonuma, S., T. Mayumi, K. Suzuki, T. Noguchi, M. Iwai, and M. Okabe. 1973. Studies on sperm capacitation. I. The relationship between a guinea-pig sperm-coating antigen and a sperm capacitation phenomenon. J. Reprod. Fertil. 35:425-432. Artamonova, V. G., and Z. N. Klishova. 1972. Pathogenesis of chronic carbon bisulphide poisoning. Gig. Tr. Prof. Zabol. 16(10):22-25. Asch, R. H., C. G. Smith, T. M. Siler-Khodr, and C. J. Pauerstein. 1979. Effects of ~9- tetrahydrocannabinol administration on gonadal steroidogenic activity in viva. Fertil. Steril. 32:576-582. Asch, R. H., C. G. Smith, T. M. Siler-Khodr, and C. J. Pauerstein. 1981. Effects of ~9- tetrahydrocannabinol during the follicular phase of the rhesus monkey (Macaca mulatta). J. Clin. Endocrinol. Metab. 52:50-55. Ash, P. 1980. The influence of radiation on fertility in man. Br. J. Radiol. 53:271-278. Badr, F. M., and R. S. Badr. 1974. Studies on the mutagenic effect of contraceptive drugs. I. Induction of dominant lethal mutations in female mice. Mutat. Res. 26:529-534. Baker, T. G. 1978. Effects of ionizing radiations on mammalian oogenesis: A model for chemical effects. Environ. Health Perspect. 24:31-37. Balentine, J. D. 1966. Pathologic effects of exposure to high oxygen tensions. N. Engl. J. Med. 275:1038-1040. Barraclough, C. A., and C. H. Sawyer. 1955. Inhibition of the release of pituitary ovulatory hormones in the rat by morphine. Endocrinology 57:329-337. Bedford, J. M. 1966. Development of the fertilizing ability of spermatozoa in the epididymis of the rabbit. J. Exp. Zool. 163:319-329. BenneK, J., and R. A. Pedersen. 1984. Early mouse embryos exhibit strain variation in radiation-induced sister-chromatic exchange: Relationship with DNA repair. Mutat. Res. 126: 153-157. Berndtson, W. E. 1977. Methods of quantifying mammalian spermatogenesis: A review. J. Anim. Sci. 44:818-833. Biggers, J. D. 1980. Oogenesis. Pp. 820-827 inJ. J. Gold and J. B. Josimovich, eds. Gynecologic Endocrinology, 3rd ed. Harper & Row, New York. Blijleven, W. G. H. 1977. Mutagenicity of four hair dyes in Drosophila melanogaster. Mutat. Res. 48:181-186. Brusick, D. J. 1978. Alterations of germ cells leading to mutagenesis and their detection. Environ. Health Perspect. 24:105-112. Cater, B. R., M. W. Cook, S. D. Gangolli, and P. Grasso. 1977. Studies on dibutyl phthalate-induced testicular atrophy in the rat: Effects on zinc metabolism. Toxicol. Appl. Pharmacol. 41 :609-618. Cattanach, B. M., I. Murray, and J. M. Tracey. 1977. Translocation yield from the immature mouse testis and the nature of spermatogonial stem cell heterogeneity. Mutat. Res.44:105-117. Chapin, R. E., S. L. Dutton, M. D. Ross, and J. C. Lamb IV. 1985. Effects of ethylene glycol monomethyl ether (EGME) on mating performance and epididymal sperm param- eters in F344 rats. Fund. Appl. Toxicol. 5:182-189. Chapman, R. M. 1983. Gonadal injury resulting from chemotherapy. Am. J. Ind. Med. 4: 149-161. Chartier, M., M. Roger, J. Barrat, and B. Michelon. 1979. Measurement of plasma human chorionic gonadotropin (hCG) and ,B-hCG activities in the late luteal phase: Evidence of

94 DRINKING WATER AND H"LTH the occurrence of spontaneous menstrual abortions in infertile women. Fertil. Steril. 31:134-137. Chung, C. S., and N. C. Myrianthopoulos. 1975. Factors Affecting Risks of Congenital Malformations. The National Foundation March of Dimes. Birth Defects: Original Article Series, Vol. XI, No. 10. Symposia Specialists, Miami, Fla. 38 pp. Clark, J. H. 1982. Sex steroids and maturation in the female. Pp. 315-328 in V. R. Hunt, M. K. Smith, and D. Worth, eds. Environmental Factors in Human Growth and De- velopment. Banbury Report 11. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Clark, J. H., and S. A. McCormack. 1980. The effect of clomid and other triphenylethylene derivatives during pregnancy and the neonatal period. J. Steroid Biochem. 12:47-53. Clark, J. H., and E. J. Peck, Jr. 1979. Female Sex Steroids: Receptors and Function. Springer-Verlag, New York. 245 pp. Clavert, A., B. Brun, and G. Bollecker. 1975. Teratospermie et migration des sperma- tozo~des in vitro et in viva. C. R. Seances Soc. Biol. Ses Fill 169:1281-1284. (English summary) Clermont, Y. 1972. Kinetics of spermatogenesis in mammals: Seminiferous epithelium cycle and spermatogonial renewal. Physiol. Rev. 52:198-236. Clifton, D. K., and W. J. Bremner. 1983. The effect of testicular x-irradiation on sper- matogenesis in man: A comparison with the mouse. J. Androl. 4:387-392. Collins, T. F. X. 1972. Dominant lethal assay. I. Captan. Food Cosmet. Toxicol. 10:353- 361. Collins, T. F. X. 1978a. Multigeneration reproduction studies. Pp. 191-214 in J. G. Wilson and F. C. Fraser, eds. Handbook of Teratology. Vol. 4. Research Procedures and Data Analysis. Plenum, New York. Collins, T. F. X. 1978b. Reproduction and teratology guidelines: Review of deliberations by the National Toxicology Advisory Committee's Reproduction Panel. J. Environ. Pathol. Toxicol. 2:141-147. Dean, J. 1983. Preimplantation development: Biology, genetics, and mutagenesis. Am. J. Ind. Med. 4:31-49. Dedov, V. I. 1980. The ultrastructure of the Sertoli and Leydig cells in the intact rats and under conditions of a long-term inner irradiation. Tsitologiia 22:1153-1156. (In Russian; English summary) Der, R., Z. Fahim, M. Yousef, and M. Fahim. 1976. Environmental interaction of lead and cadmium on reproduction and metabolism of male rats. Res. Commun. Chem. Pathol. Pharmacol. 14:689-713. Dobson, R. L., and M. F. Cooper. 1974. Tritium toxicity: Effect of low-level 3HOH exposure on developing female germ cells in the mouse. Radiat. Res. 58:91-100. Dobson, R. L., and J. S. Felton. 1983. Female germ cell loss from radiation and chemical exposures. Am. J. Ind. Med. 4:175-190. Dobson, R. L., C. G. Koehler, J. S. Felton, T. C. Kwan, B. J. Wuebbles, and D. C. L. Jones. 1978. Vulnerability of female germ cells in developing mice and monkeys to tritium, gamma rays, and polycyclic aromatic hydrocarbons. Pp. 1-14 in D. D. Mahlum, M. R. Sikov, P. L. Hackett, and F. D. Andrew, eds. Developmental Toxicology of Energy-Related Pollutants. Proceedings of the Seventeenth Annual Hanford Biology Symposium at Richland, Washington, October 17-19, 1977. Technical Information Cen- ter, U.S. Department of Energy, Oak Ridge, Tenn. (Available from the National Tech- nical Information Service, Springfield, Va., as Publication No. CONF-771017.) Donnelly, P. A., and D. E. Monty, Jr. 1977. Toxicological effects of cadmium chlonde on the canine testis following various routes of administration. Toxicol. Lett. 1:53-58.

Reproductive Toxicology 95 Doolittle, D. J., E. Bermudez, P. K. Working, and B. E. Butterworth. 1984. Measurement of genotoxic activity in multiple tissue following inhalation exposure to dimethylnitro- samine. Mutat. Res. 141:123-127. Edmonds, L., M. Hatch, L. Holmes, J. Kline, G. Letz, B. Levin, R. Miller, P. Shrout, Z. Stein, D. Warburton, M. Weinstock, R. D. Whorton, and A. Wyrobek. 1981. Report of Panel II: Guidelines for reproductive studies in exposed human populations. Pp. 37- 110 in A. D. Bloom, ed. Guidelines for Studies of Human Populations Exposed to Mutagenic and Reproductive Hazards. Proceedings of conference held January 26-27, 1981, in Washington, D.C. March of Dimes Birth Defects Foundation, White Plains, N.Y. Eeken, J. C. J., and F. H. Sobels. 1985. Studies on mutagen-sensitive strains of Drosophila melanogaster. VI. The effect of DNA-repair deficiencies in spermatids, spermatocytes and spermatogonia irradiated in N2 or O2. Mutat. Res. 149:409-414. Ehling, U. H. 1971. Comparison of radiation- and chemically-induced dominant lethal mutations in male mice. Mutat. Res. 11:35-44. Ehling, U. H. 1977. Dominant lethal mutations in male mice. Arch. Toxicol. 38:1-11. Ehling, U. H. 1980. Induction of gene mutations in germ cells of the mouse. Arch. Toxicol. 46:123-138. Eliasson, R. 1975. Analysis of semen. Pp. 691-713 in S. J. Behrman and R. W. Kistner, eds. Progress in Infertility, 2nd ed. Little, Brown, Boston. Eliasson, R. 1982. Biochemical analysis of human semen. Int. J. Androl. (Suppl. 5):109- 119. Embree, J. W., J. P. Lyon, and C. H. Hine. 1977. The mutagenic potential of ethylene oxide using the dominant-lethal assay in rats. Toxicol. Appl. Pharmacol. 40:261-267. EPA-ORNL (U.S. Environmental Protection Agency-Oak Ridge National Laboratory). 1982. Assessment of Risks to Human Reproduction and to Development of the Human Conceptus from Exposure to Environmental Substances. Report No. EPA-600/9-82-001. U.S. Environmental Protection Agency, Washington, D.C. [169 pp.] Epstein, S. S., E. Arnold, J. Andrea, W. Bass, and Y. Bishop. 1972. Detection of chemical mutagens by the dominant lethal assay in the mouse. Toxicol. Appl. Pharmacol. 23:288- 325. Erickson, B. H. 1978. Interspecific comparison of the effects of continuous ionizing ra- diation on the primitive mammalian stem germ cell. Pp. 57-67 in D. D. Mahlum, M. R. Sikov, P. L. Hackett, and F. D. Andrew, eds. Developmental Toxicology of Energy- Related Pollutants. Proceedings of the Seventeenth Annual Hanford Biology Symposium at Richland, Washington, October 17-19, 1977. Technical Information Center, U.S. Department of Energy, Oak Ridge, Tenn. (Available from the National Technical In- formation Service, Springfield, Va., as Publication No. CONF-771017.) Ericsson, R. J., and V. F. Baker. 1966. Transport of oestrogens in semen to the female rat during mating and its effect on fertility. J. Reprod. Fertil. 12:381-384. Eroschenko, V. P. 1978. Alterations in the testes of the Japanese quail during and after the ingestion of the insecticide Kepone. Toxicol. Appl. Pharmacol. 43:535-545. Espey, L. L. 1978. Ovulation. Pp. 503-532 in R. E. Jones, ed. The Vertebrate Ovary. Plenum, New York. Fedorova, N. L., B. A. Markelov, A. V. Shafirkin, and G. Y. Plyukhina. 1985. The characteristic of male dog descendants during chronic combined ~y-irradiation and in the postirradiation period. Radiobiologiia 25(1):69-73. (In Russian; English summary) Felton, J. S., T. C. Kwan, B. J. Wuebbles, and R. L. Dobson. 1978. Genetic differences in polycyclic-aromatic-hydrocarbon metabolism and their effects on oocyte killing in developing mice. Pp. 15-26 in D. D. Mahlum, M. R. Sikov, P. L. Hackett, and F. D.

96 DRINKING WATER AND H"LTH Andrew, eds. Developmental Toxicology of Energy-Related Pollutants. Proceedings of the Seventeenth Annual Hanford Biology Symposium at Richland, Washington, October 17-19, 1977. Technical Information Center, U.S. Department of Energy, Oak Ridge, Tenn. (Available from the National Technical Information Service, Springfield, Va., as Publication No. CONF-771017.) Furuhjelm, M., B. Jonson, and C. C. Lagergren. 1962. The quality of human semen in spontaneous abortion. Int. J. Fertil. 7:17-21. Gangolli, S. D. 1982. Testicular effects of phthalate esters. Environ. Health Perspect. 45:77-84. Gaulden, E. C., D. C. Littlefield, O. E. Putoff, and A. L. Seivert. 1964. Menstrual abnormalities associated with heroin addiction. Am. J. Obstet. Gynecol. 90:155-160. Gaylor, D. W. 1978. Methods and concepts of biometrics applied to teratology. Pp. 429- 444 in J. G. Wilson and F. C. Fraser, eds. Handbook of Teratology. Vol. 4. Research Procedures and Data Analysis. Plenum, New York. Generoso, W. M., K. T. Cain, S. W. Huff, and D. G. Gosslee. 1978. Inducibility by chemical mutagens of heritable translocations in male and female germ cells of mice. Pp. 109-129 in W. G. Flamm and M. A. Mehlman, eds. Advances in Modern Toxicology. Vol. 5. Mutagenesis. Hemisphere, Washington, D.C. Generoso, W. M. 1982. A possible mechanism for chemical induction of chromosome aberrations in male meiotic and postmeiotic germ cells of mice. Cytogenet. Cell Genet. 33:74-80. Generoso, W. M., J. B. Bishop, D. G. Gosslee, G. W. Newell, C. Sheu, and E. von Halle. 1980. Heritable translocation test in mice. Mutat. Res. 76:191-215. Generoso, W. M., K. T. Cain, and A. J. Bandy. 1983. Some factors affecting the mutagenic response of mouse germ cells to chemicals. Pp. 227-239 in F. J. de Serres and W. Sheridan, eds. Utilization of Mammalian Specif~c Locus Studies in Hazard Evaluation and Estimation of Genetic Risk. Plenum, New York. Gerber, N., and R. K. Lynn. 1976. Excretion of methadone in semen from methadone addicts; comparison with blood levels. Life Sci. 19:787-792. Gondos, B. 1978. Oogonia and oocytes in mammals. Pp. 83-120 in R. E. Jones, ed. The Vertebrate Ovary. Plenum, New York. Gondos, B. 1980. Development and differentiation of the testis and male reproductive tract. Pp. 3-20 in A. Steinberger and E. Steinberger, eds. Testicular Development, Structure and Function. Raven Press, New York. Gordon, G. G., A. L. Southren, and C. S. Lieber. 1978. The effects of alcoholic liver disease and alcohol ingestion on sex hormone levels. Alcohol. Clin. Exp. Res. 2:259- 263. Green, S., A. Auletta, J. Fabricant, R. Kapp, M. Manandhgar, C.-J. Sheu, J. Springer, and B. Whitfield. 1985. Current status of bioassays in genetic toxicology the dominant lethal assay: A report of the U.S. Environmental Protection Agency Gene-Tox Program. Mutat. Res. 154:49-67. Greiner, R. 1982. Spermatogenesis after fractionated, low-dose irradiation of the gonads. Strahlentherapie 158:342-355. (In German; English abstract) Hacker, U., J. Schumann, and W. Gohde. 1980. Effects of acute gamma-irradiation on spermatogenesis as revealed by flow cytometry. Acta Radiol. Oncol. 19:361-368. Hacker, U., J. Schumann, W. Gohde, and K. Muller. 1981. Mammalian spermatogenesis as a biologic dosimeter for radiation. Acta Radiol. Oncol. 20:279-282. Hacker-Klom, U. 1985. Long term effects of ionizing radiation on mouse spermatogenesis. Acta Radiol. Oncol. 24:363-367.

Reproductive Toxicology 97 Hafez, E. S. E. 1977. Physio-anatomical parameters of andrology. Pp. 39-79 in E. S. E. Hafez, ed. Techniques of Human Andrology. North-Holland, New York. Haney, A. F. 1985. Effects of toxic agents on ovarian function. Pp. 181-210 in J. A. Thomas, K. S. Korach, and J. A. McLachlan, eds. Endocrine Toxicology. Raven Press, New York. Harclerode, J., L. Bird, H. Sawyer, V. Berger, R. Mooney, and R. Smith. 1984. Sex hormone levels in adult rats injected with delta-9-tetrahydrocannabinol and phencyclidine. Pp. 441-452 in S. Agurell, W. L. Dewey, and R. E. Willette, eds. The Cannabinoids: Chemical, Pharmacologic, and Therapeutic Aspects. Academic Press, Orlando, Fla. Hartman, C. G. 1959. The permeability of the vaginal mucosa. Ann. N.Y. Acad. Sci. 83:318-327. Heinrichs, W. L., and M. R. Juchau. 1980. Extrahepatic drug metabolism: The gonads. Pp. 319-332 in T. E. Gram, ed. Extrahepatic Metabolism of Drugs and Other Foreign Compounds. SP Medical and Scientific Books, New York. Hertig, A. T., and B. R. Barton. 1973. Fine structure of mammalian oocytes and ova. Pp. 317-348 in R. O. Greep and E. B. Astwood, eds. Handbook of Physiology. Section 7: Endocrinology. Vol. II. Female Reproductive System, Part 1. American Physiological Society, Washington, D.C. Holden, H. E. 1982. Comparison of somatic and germ ceil models for cytogenetic screening. J. Appl. Toxicol. 2:196-200. Hollister, L. E. 1973. Human pharmacology of drugs of abuse with emphasis on neu- roendocrine effects. Prog. Brain Res. 39:373-381. Hudson, P. M., K. Yoshikawa, S. F. Ali, J. C. Lamb IV, J. R. Reel, and J.-S. Hong. 1984. Estrogen-like activity of chlordecone (Kepone) on the hypothalamo-pituitary axis: Effects on the pituitary enkephalin system. Toxicol. Appl. Pharmacol. 74:383-389. IARC (International Agency for Research on Cancer). 1978. N-Nitrosodimethylamine. Pp. 125-175 in IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. Vol. 17. Some N-Nitroso Compounds. International Agency for Research on Cancer, Lyon, France. ICPEMC (International Commission for Protection against Environmental Mutagens and Carcinogens). 1983. Committee 1 Final Report: Screening strategy for chemicals that are potential germ-cell mutagens in mammals. Mutat. Res. 114:117-177. Infante, P. F., J. K. Wagoner, and R. J. Young. 1977. Chloroprene: Observations of carcinogenesis and mutagenesis. Pp. 205-217 in H. H. Hiatt, J. D. Watson, and J. A. Winsten, eds. Origins of Human Cancer. Book A. Incidence of Cancer in Humans. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Johnson, F. M., and S. E. Lewis. 1981. Mouse spermatogonia exposed to a high, multiply fractionated dose of a cancer chemotherapeutic drug: Mutation analysis by electropho- resis. Mutat. Res. 81: 197-202. Johnston, D. E., Y. Chiao, J. S. Gavaler, and D. H. Van Wiel. 1981. Inhibition of testosterone synthesis by ethanol and acetaldehyde. Biochem. Pharmacol. 30:1827-1831. Kalla, N. R., and M. P. Bansal. 1975. Effect of carbon tetrachloride on gonadal physiology in male rats. Acta Anat. 91:380-385. Kapp, R. W., Jr., and C. B. Jacobson. 1980. Analysis of human spermatozoa for Y chromosomal nondisjunction. Teratogen. Carcinogen. Mutagen. 1:193-211. Katz, D. F., and J. W. Overstreet. 1980. Mammalian sperm movement in the secretions of the male and female genital tracts. Pp. 481-489 in A. Steinberger and E. Steinberger, eds. Testicular Development, Structure, and Function. Raven Press, New Yo~k. Katz, D. F., and J. W. Overstreet. 1981. Sperm motility assessment by videomicrography. Fertil. Steril. 35:188-193.

98 DRINKING WATER AND HEALTH Knobil, E. 1980. The neuroendocrine control of the menstrual cycle. Recent Prog. Horm. Res. 36:53-88. Koeter, H. B. W. M. 1983. Relevance of parameters related to fertility and reproduction in toxicity testing. Am. J. Ind. Med. 4:81-86. Komatsu, H., T. Kakizoe, T. Niijima, T. Kawachi, and T. Sugimura. 1982. Increased sperm abnormalities due to dietary restriction. Mutat. Res. 93:439-446. Kraulis, I., H. Traikov, M. Sharpe, K. B. Ruf, and F. Naftolin. 1978. Steroid induction of gonadotropin surges in the immature rat. I. Priming effects of androgens. Endocri- nology 103: 1822-1828. Krause, W., K. Hamm, and J. Weissmuller. 1975. The effect of DDT on spermatogenesis of the juvenile rat. Bull. Environ. Contam. Toxicol. 14: 171-179. Krause, W., K. Hamm, and J. Weissmuller. 1976. Damage to spermatogenesis in juvenile rat treated with DDVP and malathion. Bull. Environ. Contam. Toxicol. 15:458-462. Kripke, B. J., A. D. Kelman, N. K. Shah, K. Balogh, and A. H. Handler. 1976. Testicular reaction to prolonged exposure to nitrous oxide. Anaesthesiology 44:105-113. Krzanowska, H. 1974. The passage of abnormal spermatozoa through the uterotubal junction of the mouse. J. Reprod. Fertil. 38:81-90. Kunitake, G., and V. Davajan. 1970. A new method of evaluating infertility due to cervical mucus-spermatozoa incompatibility. Fertil. Steril. 21 :706-714. Lamb, J. C., IV, and R. E. Chapin. 1985. Experimental models of male reproductive toxicology. Pp. 85-115 in J. A. Thomas, K. S. Korach, and J. A. McLachlan, eds. Endocrine Toxicology. Raven Press, New York. Lamb, J. C., IV, D. K. Gulati, V. S. Russell, L. Hommel, and P. S. Sabharwal. 1984. Reproductive toxicity of ethylene glycol monoethyl ether tested by continuous breeding of CD-1 mice. Environ. Health Perspect. 57:85-90. Leavitt, W. W., R. W. Evans, and W. J. Hendry, III. 1982. Etiology of DES-induced uterine tumors in the Syrian hamster. Adv. Exp. Med. Biol. 138:63-86. Leblond, C. P., and Y. Clermont. 1952a. Def~nition of the stages of the cycle of the seminiferous epithelium in the rat. Ann. N.Y. Acad. Sci. 55:548-573. Leblond, C. P., and Y. Clermont. 1952b. Spermiogenesis of rat, mouse, hamster and guinea pig as revealed by the "periodic acid-fuchsin sulfurous acid" technique. Am. J. Anat. 90:167-215. Lee, I. P., and R. L. Dixon. 1975. Effects of mercury on spermatogenesis studied by velocity sedimentation cell separation and serial mating. J. Pharmacol. Exp. Ther. 194:171- 181. Leenhouts, H. P., and K. H. Chadwick. 1981. An analytical approach to the induction of translocations in the spermatogonia of the mouse. Mutat. Res. 82:305-321. Leonard, A., G. DeKnudt, and G. Linden. 1975. Mutagenicity tests with aflatoxins in the mouse. Mutat. Res. 28:137-139. Lewin, L. M. 1977. Biochemical markers in human seminal plasma as a means of evaluating the functioning of the male reproductive tract. Pp. 505-511 in P. Troen and H. R. Nankin, eds. I~he Testis in Normal and Infertile Men. Raven Press, New York. Lin, Y. C., J. M. Loring, and C. A. Villee. 1982. Permissive role of the pituitary in the induction and growth of estrogen-dependent renal tumors. Cancer Res. 42:1015-1019. Lipshultz, L. I., and S. S. Howards, eds. 1983. Infertility in the Male. Churchill Living- stone, New York. 409 pp. Lutwak-Mann, C., K. Schmid, and H. Keberle. 1967. Thalidomide in rabbit semen. Nature 214: 1018-1020.

Reproductive Toxicology 99 Lyon, M. F., and B. D. Smith. 1971. Species comparisons concerning radiation-induced dominant lethals and chromosome aberrations. Mutat. Res. 11:45-58. Lyon, M. F., R. J. S. Phillips, and G. Fisher. 1979. Dose-response curves for radiation- induced gene mutations in mouse oocytes and their interpretation. Mutat. Res. 63:161- 173. MacLeod, J., and R. Z. Gold. 1951. The male factor in fertility and infertility. II. Sper- matozoon counts in 1000 men of known fertility and in 1000 cases of infertile marriage. J. Urol. 66:436-449. MacLeod, J., and Y. Wang. 1979. Male fertility potential in terms of semen auali~v A review of the past, a study of the present. Fertil. Steril. 31:103-116. ~ _ _ , . MacLusky, N. J., and F. Naftolin. 1981. Sexual differentiation of the central nervous system. Science 211:1294-1303. Malmborg, A., K. Dornbusch, R. Eliasson, and C. Lindholmer. 1976. Concentration of antibacterials in human seminal plasma. Pp. 53-59 in J. D. Williams and A. M. Geddes, eds. Chemotherapy. Vol. 4. Pharmacology of Antibiotics. Plenum, New York. Mandl, A. M., and H. M. Beaumont. 1964. The differential radiosensitivity of oogonia and oocytes at different developmental stages. Pp. 311-321 in W. D. Carlson and F. X. Gassner, eds. Effects of Ionizing Radiation on the Reproductive System. Pergamon, New York. Manikowska-Czerska, E., P. Czerski, and W. M. Leach. 1985. Effects of 2.45 GHz microwaves on meiotic chromosomes of male CBA/CAY mice. J. Hered. 76:71-73. Manson, J. M., and R. Simons. 1979. Influence of environmental agents on male repro- ductive failure. Pp. 155-179 in V. Hunt, ed. Work and the Health of Women. CRC Press, New York. Manson, J. M., H. Zenick, and R. D. Costlow. 1982. Teratology test methods for laboratory animals. Pp. 141-184 in A. W. Hayes, ed. Principles and Methods of Toxicology. Raven Press, New York. Martin, R. H., A. Rademaker, M. Barnes, K. Arthur, T. Ringrose, and G. Douglas. 1985. A prospective serial study of the effects of radiotherapy on semen parameters, and hamster egg penetration rates. Clin. Invest. Med. 8:239-243. Mattison, D. R. 1983. Ovarian toxicity: Effects on sexual maturation, reproduction and menopause. Pp. 317-342 in T. W. Clarkson, G. W. Nordberg, and P. R. Sager, eds. Reproductive and Developmental Toxicity of Metals. Plenum, New York. Mattison, D. R. 1985. Clinical manifestations of ovarian toxicity. Pp. 109-130 in R. L. Dixon, ed. Reproductive Toxicology. Raven Press, New York. Mattison, D. R., and S. S. Thorgeirsson. 1978. Gonadal aryl hydrocarbon hydroxylase in rats and mice. Cancer Res. 38:1368-1373. Mattison, D. R., and S. S. Thorgeirsson. 1979. Ovarian aryl hydrocarbon hydroxylase activity and primordial oocyte toxicity of polycyclic aromatic hydrocarbons in mice. Cancer Res. 39:3471-3475. Mattison, D. R., K. Shiromizu, and M. S. Nightingale. 1983. Oocyte destruction by polycyclic aromatic hydrocarbons. Am. J. Ind. Med. 4:191-202. McCann, S. M. 1982. Physiology, pharmacology and clinical application of LH-releasing hormone. Pp. 73-91 in T. G. Muldoon, V. B. Mahesh, and B. Perez-Ballester, eds. Recent Advances in Fertility Research. Part A. Developments in Reproductive Endo- crinology. Proceedings of The International Symposium on Recent Advances in Fertility Research held in Buenos Aires, Argentina, December 6-9, 1981. Alan R. Liss, New York. McCann, S. M., M. D. Lumpkin, H. Mizunuma, W. K. Samson, S. R. Ojeda, and A. Negro-Vilar. 1982. Control of gonadotropin secretion by brain amines and peptides. Pp.

]00 DRINKING WATER AND HEALTH 15-29 in T. G. Muldoon, V. B. Mahesh, and B. Perez-Ballester, eds. Recent Advances in Fertility Research. Part A. Developments in Reproductive Endocrinology. Proceedings of The International Symposium on Recent Advances in Fertility Research held in Buenos Aires, Argentina, December 6-9, 1981. Alan R. Liss, New York. McEwen, B. S. 1981. Neural gonadal steroid actions. Science 211:1303-1311. McLachlan, J. A., R. R. Newbold, K. S. Korach, J. C. Lamb IV, and Y. Suzuki. 1981. Transplacental toxicology: Prenatal factors influencing postnatal fertility. Pp. 213-232 in C. A. Kimmel and J. BueLke-Sam, eds. Developmental Toxicology. Raven Press, New York. Meistrich, M. L., and R. C. Samuels. 1985. Reduction in sperm levels after testicular irradiation of the mouse: A comparison with man. Radiat. Res. 102:138-147. Meistrich, M. L., N. R. Hunter, N. Suzuki, P. K. Trostle, and H. R. Withers. 1978. Gradual regeneration of mouse testicular stem cells after exposure to ionizing radiation. Radiat. Res. 74:349-362. Meistrich, M. L., J. L. Longtin, and W. A. Brock. 1979. Further purification of rat spermatogenic cells by density centrifugation. (Abstract G1126.) J. Cell Biol. 83:226 a. Mendelson, J. H., and N. K. Mello. 1984. Effects of marijuana on neuroendocrine hormones in human males and females. Pp. 97-114 in M. C. Braude and J. P. Ludford, eds. Marijuana Effects on the Endocrine and Reproductive Systems. NIDA Research Mon- ograph 44. National Institute on Drug Abuse, Department of Health and Human Services, Rockville, Md. Mian, T. A., N. Suzuki, H. J. Glenn, T. P. Haynie, and M. L. Meistrich. 1977. Radiation damage to mouse testis cells from [99mTc] pertechnetate. J. Nucl. Med. 18:1116-1122. Mintz, J., K. O'Hare, C. P. O'Brien, and J. Goldschmidt. 1974. Sexual problems of heroin addicts. Arch. Gen. Psychiatry 31:700-703. Miyaji, T., M. Miyamoto, and Y. Ueda. 1964. Inhibition of spermatogenesis and atrophy of the testis caused by nitrofuran compounds. Acta Pathol. Jpn. 14:261-273. Miyamoto, T. 1983. ~y-Ray-induced mutations in male germ cells of a recombination- defective strain (c3G) of Drosophila melanogaster. Mutat. Res. 120:27-36. Mohrenweiser H. W., and J. V. Neel. 1982. Models to man: Establishment of reference points for estimating genetic risk in man. Pp. 471-486 in B. A. Bridges, B. E. Butter- worth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Mosher, W. D. 1985. Reproductive impairments in the United States, 1965-1982. De- mography 22:415-430. Mukherjee, D. P., and S. P. Singh. 1967. Effect of increased carbon dioxide in inspired air on the morphology of spermatozoa and fertility of mice. J. Reprod. Fertil. 13:165- 167. Mukhtar, H., I. P. Lee, G. L. Foureman, and J. R. Bend. 1978. Epoxide metabolizing enzyme activities in rat testes: Postnatal development and relative activity in interstitial and spermatogenic cell compartments. Chem.-Biol. Interact. 22:153-165. Nansel, D. D., M. S. Aiyer, W. H. Meinzer, II, and E. M. Bogdanove. 1979. Rapid direct effects of castration and androgen treatment on luteinizing hormone-releasing hormone-induced luteinizing hormone release in the phenobarbital-treated male rat: Ex- amination of the roles direct and indirect androgen feedback mechanisms might play in the physiological control of luteinizing hormone release. Endocrinology 104:524-531. NCHS (National Center for Health Statistics). 1985. Births, Marriages, Divorces, and Deaths for July 1985. Monthly Vital Statistics Report, Vol. 34, No. 7, October 21, 1985. Public Health Service, U.S. Department of Health and Human Services, Hyatts- ville, Md. 12 pp.

Reproductive Toxicology 101 Neel, J. V., and E. Rothman. 1981. Is there a difference among human populations in the rate with which mutation produces electrophoretic variants? Proc. Natl. Acad. Sci. USA 78:3108-3112. Nelson, S. D., M. R. Boyd, and J. R. Mitchell. 1977. Role of metabolic activation in chemical-induced tissue injury. Pp. 155-185 in D. M. Jerina, ed. Drug Metabolism Concepts. ACS Symposium Series 44. American Chemical Society, Washington, D.C. Nicander, L. 1975. Changes produced in the male genital organs of rabbits and dogs by 2,6-cis-diphenylhexamethylcyclotetrasiloxane (KABI 1774). Acta Pharmacol. Toxicol. 36(Suppl. m):40-54. Niswander, K. R., and M. Gordon, eds. 1972. The Women and Their Pregnancies. The Collaborative Perinatal Study of the National Institute of Neurological Diseases and Strokes. W. B. Saunders, Philadelphia. 540 pp. [Originally DHEW Publication No. (NIH) 73-379.] Oakberg, E. F., and E. Clark. 1964. Species comparisons of radiation response of the gonads. Pp. 11-24 in W. D. Carlson and F. X. Gassner, eds. Effects of Ionizing Radiation on the Reproductive System. Pergamon, New York. Oesch, F., H. Glad, and H. Schmassmann. 1977. The apparent ubiquity of epoxide hy- dratase in rat organs. Biochem. Pharmacol. 26:603-607. Osterberg, R. E., G. W. Bierbower, and R. M. Hehir. 1977. Renal and testicular damage following dermal application of the flame retardant, tris(2,3-dibromopropyl) phosphate. J. Toxicol. Environ. Health 3:979-987. Overstreet, J. W. 1984. Assessment of disorders of spermatogenesis. Pp. 275-292 in J. E. Lockey, G. K. Lemasters, and W. R. Keye, Jr., eds. Reproduction: The New Frontier in Occupational and Environmental Health Research. Proceedings of the Fifth Annual RMCOEH Occupational and Environmental Health Conference held in Park City, Utah, April 5-8, 1983. Alan R. Liss, New York. Packman, P. M., and J. A. Rothchild. 1976. Morphine inhibition of ovulation: Reversal by naloxone. Endocrinology 99:7-10. Paladine, W. J., T. J. Cunningham, M. A. Donavan, and C. W. Dumper. 1975. Possible sensitivity to vinblastine in prostatic or seminal fluid. (Correspondence.) N. Engl. J. Med. 292:52. Palazzolo, R. J., J. A. McHard, E. J. Hobbs, O. E. Fancher, and J. C. Calandra. 1972. Investigation of the toxicologic properties of a phenylmethylcyclosiloxane. Toxicol. Appl. Pharmacol. 21: 15-28. Palmer, A. K. 1978. The design of subprimate animal studies. Pp. 215-253 in J. G. Wilson and F. C. Fraser, eds. Handbook of Teratology. Vol. 4. Research Procedures and Data Analysis. Plenum, New York. Palmer, A. K. 1981. Regulatory requirements for reproductive toxicology: Theory and practice. Pp. 259-287 in C. A. Kimmel and J. Buelke-Sam, eds. Developmental Tox- icology. Raven Press, New York. Pasi, A., J. W. Embree, Jr., G. H. Eisenlord, and C. H. Hine. 1974. Assessment of the mutagenic properties of diquat and paraquat in the murine dominant lethal test. Mutat. Res. 26:171-175. Pedersen, R. A., and B. Brandriff. 1980. Radiation- and drug-induced DNA repair in mammalian oocytes and embryos. Pp. 389-410 in W. M. Generoso, M. D. Shelby, and F. J. de Serres, eds. DNA Repair and Mutagenesis in Eukaryotes. Plenum, New York. Pedersen, T., and H. Peters. 1968. Proposal for a classif~cation of oocytes and follicles in the mouse ovary. J. Reprod. Fertil. 17:555-557.

]02 DRINKING WATER AND HEALTH Phillips, J. C., P. M. D. Foster, and S. D. Gangolli. 1985. Chemically-induced injury to the male reproductive tract. Pp. 117-134 in J. A. Thomas, K. S. Korach, and J. A. McLachlan, eds. Endocrine Toxicology. Raven Press, New York. Pomerantseva, M. D., and L. K. Ramaija. 1984. Chemical protection against genetic effect of radiation in male mice. Mutat. Res. 140:131-135. Preston, R. J. 1982. Chromosome aberrations in Recondensed sperm DNA. Pp. 515-526 in B. A. Bridges, B. E. Butterworth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Reznik, Y. B., and G. K. Sprinchan. 1975. Experimental data on the gonadotoxic effect of Nemagon. Gig. Sanit. 1975(6): 101-102. (In Russian) Robb, G. W., R. P. Amann, and G. J. Killian. 1978. Daily sperm production and epididymal sperm reserves of pubertal and adult rats. J. Reprod. Fertil. 54:103-107. Rosenfeld, D. L., and R. A. Bronson. 1980. Reproductive problems in the DES-exposed female. Obstet. Gynecol. 55:453-456. Russell, L. B., and M. D. Shelby. 1985. Tests for heritable genetic damage and for evidence of gonadal exposure in mammals. Mutat. Res. 154:69-84. Russell, L. B., P. B. Shelby, E. von Halle, W. Sheridan, and L. Valcovic. 1981. The mouse specific-locus test with agents other than radiations. Interpretation of data and recommendations for future work. Mutat. Res. 86:329-354. Russell, L. D. 1983. Normal testicular structure and methods of evaluation under exper- imental and disruptive conditions. Pp. 227-252 in T. W. Clarkson, G. F. Nordberg, and P. R. Sager, eds. Reproductive and Developmental Toxicity of Metals. Plenum, New York. Russell, W. L. 1951. X-ray-induced mutations in mice. Cold Spring Harbor Symp. Quant. Biol. 16:327-336. Russell, W. L., E. M. Kelly, P. R. Hunsicker, J. W. Bangham, S. C. Maddux, and E. L. Phipps. 1979. Specific-locus test shows ethylnitrosourea to be the most potent mutagen in the mouse. Proc. Natl. Acad. Sci. USA 76:5818-5819. Rutledge, J. C., K. T. Cain, N. L. A. Cacheiro, C. V. Cornett, C. G. Wright, and W. M. Generoso. 1986. A balanced translocation in mice with a neurological defect. Science 231 :395-397. Salisbury, G. W., R. G. Hart, and J. R. Lodge. 1977. The spermatozoan genome and fertility. Am. J. Obstet. Gynecol. 128:342-350. Sawyer, C. H. 1963. Discussion. Pp. 444-457 in A. V. Nalbandov, ed. Advances in Neuroendocrinology. University of Illinois Press, Urbana, Ill. Scher, P. M., C. G. Smith, and R. G. Almirez. 1983. The role of endogenous opioid peptides in the control of sex hormone levels in the male non-human primate. (Abstract 16.) J. Androl. 4:35. Sega, G. A. 1982. DNA repair in spermatocytes and spermatids of the mouse. Pp. 503- 513 in B. A. Bridges, B. E. Butterworth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Selby, P. B. 1983. Applications in genetic risk estimation of data on the induction of dominant skeletal mutations in mice. Pp. 191-210 in F. J. de Serres and W. Sheridan, eds. Utilization of Mammalian Specific Locus Studies in Hazard Evaluation and Esti- mation of Genetic Risk. Plenum, New York. Sheehan, D. M., W. S. Branham, K. L. Medlock, M. E. Olson, and D. Zehr. 1980. Estrogen plasma binding and regulation of development in the neonatal rat. (Abstract.) Teratology 21:68A.

Reproductive Toxicology 103 Sherins, R. J., D. Brightwell, and P. M. Sternthal. 1977. Longitudinal analysis of semen of fertile and infertile men. Pp. 473-488 in P. Troen and H. R. Nankin, eds. The Testis in Normal and Infertile Men. Raven Press, New York. Shtenberg, A. I., and M. N. Rybakova. 1968. Effect of carbaryl on the neuroendocrine system of rats. Food Cosmet. Toxicol. 6:461-467. Sinclair, D. A. R., and T. A. Grigliatti. 1985. Investigation of the nature of P-induced male recombination in Drosophila melanogaster. Genetics 110:257-279. Singh, A. R., W. H. Lawrence, and J. Autian. 1975. Dominant lethal mutations and antifertility effects of di-2-ethylhexyl adipate and diethyl adipate in male mice. Toxicol. Appl. Pharmacol. 32:566-576. Skare, J. A., and K. R. Schrotel. 1984. Alkaline elusion of rat testicular DNA: Detection of DNA strand breaks after in viva treatment with chemical mutagens. Mutat. Res. 130:283-294. Smith, C. G., and P. M. Gilbeau. 1985. Drug abuse effects on reproductive hormones. Pp. 249-267 in J. A. Thomas, K. S. Korach, and J. A. McLachlan, eds. Endocrine Toxicology. Raven Press, New York. Smith, C. G., M. T. Smith, N. F. Besch, R. G. Smith, and R. H. Asch. 1979. Effect of /19-tetrahydrocannabinol (THC) on female reproductive function. Pp. 449-467 in G. G. Nahas and W. D. M. Paton, eds. Marihuana: Biological Effects. Analysis, Metabolism, Cellular Responses, Reproduction and Brain. Proceedings of the Satellite Symposium of the 7th International Congress of Pharmacology, Paris, 22-23 July 1978. Pergamon, New York. Smith, C. G., N. F. Besch, and R. H. Asch. 1980. Effects of marihuana on the reproductive system. Pp. 273-294 in J. A. Thomas and R. L. Singhal, eds. Advances in Sex Hormone Research. Vol. 4. Urban and Schwarzenberg, Baltimore. Smith, K. D., and E. Steinberger. 1977. What is oligospermia? Pp. 489-503 in P. Troen and H. R. Nankin, eds. The Testis in Normal and Infertile Men. Raven Press, New York. Stenger, E.-G., L. Aeppli, D. Muller, E. Peheim, and P. Thomann. 1971. On the toxicology of ethyleneglycol-monoethylether. Arzneim.-Forsch. 21:880-885. (In German; English summary) Stott, W. T., and P. G. Watanabe. 1980. Kinetic interaction of chemical mutagens with mouse spe~lll in vivo as it relates to animal mutagenic effects. Toxicol. Appl. Pharmacol. 55:411-416. Strobino, B. R., J. Kline, and Z. Stein. 1978. Chemical and physical exposures of parents: Effects on human reproduction and offspring. Early Hum. Dev. 1:371-399. Surgeon General. 1981. The Health Consequences of Smoking. The Changing Cigarette. A Report of the Surgeon General. Office on Smoking and Health, Public Health Service, U.S. Department of Health and Human Services, Rockville, Md. 252 pp. Takasugi, N. 1976. Cytological basis for permanent vaginal changes in mice treated neo- natally with steroid hormones. Int. Rev. Cytol. 44:193-224. Takizawa, K., and D. R. Mattison. 1983. Female reproduction. Am. J. Ind. Med. 4:17- 30. Tokuhata, G. 1968. Smoking in relation to infertility and fetal loss. Arch. Environ. Health 17:353-359. Torkelson, T. R., S. E. Sadek, V. K. Rowe, J. K. Kodama, H. H. Anderson, G. S. Loquvam, and C. H. Hine. 1961. Toxicologic investigations of 1,2-dibromo-3-chloro- propane. Toxicol. Appl. Pharmacol. 3:545-559. Tsafriri, A. 1978. Oocyte maturation in mammals. Pp. 409-442 in R. E. Jones, ed. The Vertebrate Ovary. Plenum, New York.

|04 DRINKING WATER AND HEALTH Tugrul, S. 1965. Action teratogene de l'acide glutamique. Arch. Int. Pharmacodyn. 153:323- 333. (English summary) Ulstein, M. 1973. Fertility of donors at heterologous insemination. Acta Obstet. Gynecol. Scand. 52:97-101. Valcovic, L. R., and H. V. Mailing. 1975. Mutation detection in the biochemical specific locus system. (Abstract 58.) Mutat. Res. 31 :338-339. Van Thiel, D. H., J. S. Gavaler, and R. Lester. 1977. Ethanol: A gonadal toxin in the female. Drug Alcohol Depend. 2:373-380. Van Thiel, D. H., J. S. Gavaler, R. Lester, and R. J. Sherins. 1978. Alcohol-induced ovarian failure in the rat. J. Clin. Invest. 61:624-632. Waltschewa, W., M. Slatewa, and I. Michailow. 1972. Testicular changes due to long- term administration of nickel sulfate in rats. Exp. Pathol. 6:116-120. (In German; English abstract) Warburton, D., and F. C. Fraser. 1964. Spontaneous abortion risks in man: Data from reproductive histories collected in a medical genetics unit. Am. J. Hum. Genet. 16:1- 25. Wedig, J. H., and V. L. Gay. 1973. Potentiation of luteinizing hormone-releasing factor activities following pentobarbital anesthesia in the steroid-blocked castrated rat. Proc. Soc. Exp. Biol. Med. 144:993-998. Woodhead, A. D., R. B. Setlow, and V. Pond. 1984. The Amazon molly, Poeciliaformosa, as a test animal in carcinogenicity studies: Chronic exposures to physical agents. Natl. Cancer Inst. Monogr. 65:45-52. Working, P. K., and B. E. Butterworth. 1984. An assay to detect chemically induced DNA repair in rat spermatocytes. Environ. Mutagen. 6:273-286. Wyrobek, A. J. 1977. Sperm shape abnormalities in the mouse as an indicator of mutagenic damage. Pp. 519-528 in P. Troen and H. R. Nankin, eds. The Testis in Normal and Infertile Men. Raven Press, New York. Wyrobek, A. J., G. Watchmaker, J. Foote, and H. Singh. 1978. Effects of methyl meth- anesulfonate and dimethylnitrosamine on sperm production in Syrian hamsters. Pp. 68- 78 in D. D. Mahlum, M. R. Sikov, P. L. Hackett, and F. D. Andrew, eds. Developmental Toxicology of Energy-Related Pollutants. Proceedings of the Seventeenth Annual Han- ford Biology Symposium at Richland, Washington, October 17-19, 1977. Technical Information Center, U.S. Department of Energy, Oak Ridge, Tenn. (Available from the National Technical Information Service, Springfield, Va., as Publication No. CONF- 771017.) Wyrobek, A. J., L. A. Gordon, G. Watchmaker, and D. H. Moore II. 1982. Human sperm morphology testing: Description of a reliable method and its statistical power. Pp. 527- 541 in B. A. Bridges, B. E. Butterworth, and I. B. Weinstein, eds. Indicators of Genotoxic Exposure. Banbury Report 13. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Wyrobek, A. J., G. Watchmaker, and L. Gordon. 1984. An evaluation of sperm tests as indicators of germ-cell damage in men exposed to chemical or physical agents. Pp. 385- 405 in J. E. Lockey, G. K. Lemasters, and W. R. Keye, Jr., eds. Reproduction: The New Frontier in Occupational and Environmental Health Research. Proceedings of the Fifth Annual RMCOEH Occupational and Environmental Health Conference held in Park City, Utah, April 5-8, 1983. Alan R. Liss, New York. Yanagimachi, R., H. Yanagimachi, and B. J. Rogers. 1976. The use of zone-free animal ova as a test-system for the assessment of the fertilizing capacity of human spermatozoa. Biol. Reprod. 15:471-476.

Next: 4. Neurotoxic Effects »
Drinking Water and Health,: Volume 6 Get This Book
×
Buy Paperback | $85.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The most recent volume in the Drinking Water and Health series contains the results of a two-part study on the toxicity of drinking water contaminants. The first part examines current practices in risk assessment, identifies new noncancerous toxic responses to chemicals found in drinking water, and discusses the use of pharmacokinetic data to estimate the delivered dose and response. The second part of the book provides risk assessments for 14 specific compounds, 9 presented here for the first time.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!