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.
11 Introduction i: Less information is available concern- ng reproductive toxicology for females than for males. This lack of formal data applies to work with laboratory animals and humans and reflects substantial dif- ferences between males and females in the chronology of gametogenesis, the number and accessibility of germinal cells, and probable differences in neuroendocrine . - m1 lieu. This chapter identifies biologic proc- esses that may be sensitive to environmen- tal influences and hence may be particular- ly ripe for development and application of markers. It is arranged as an overview of reproductive biology and covers oogen- esis, development of the female reproduc- tive tract, maturation, maintenance of reproductive function, fertilization, implantation, and reproductive sene- scence. In utero assessments are limited to female reproductive tract development, and the use of human chorionic gonado- tropin (hCG) as an assessment of fecundity in women. The introductory material is followed by four chapters that discuss biologic markers of germ cell damage; development and aging; cyclic ovarian function; and conception, implantation, and early embryonic loss. This review is not comprehensive but focuses on major substantive aspects of female reproduc- tive toxicity. 149 The subcommittee's conclusions and recommendations regarding research oppor- tunities and directions for a program on biologic markers of female reproduction are presented in Chapter 16. A summary table lists and categorizes biologic mark- ers according to whether each marker may be used in large-scale human studies or only in studies of special populations and whether it needs further development or needs to be applied in animal studies. The physiologic processes and conse- quent landmarks of female reproductive function are discussed in this chapter. Identification of landmarks that describe normal functions is a necessary first step for identifying biologic assessments that might be used as markers of exposure or effect. Hypothetical and documented toxic effects are discussed for each stage of female reproductive function. Specific biologic markers are discussed in later chapters. OOGENESIS Primordial germ cells in humans are first detectable in the yolk sac at 3 weeks of development. These oogonia (approxi- mately 1,700) migrate to the gonadal ridge and undergo a period of active mitosis that peaks-with an increase to 5-7 x 106 cells- around month 5 of gestation. All the oogo-
150 nia enter the initial stage of meiosis by month 7, at which point they become known as primary oocytes. Approximately 60% of the store of germ cells is lost to physio- logic atresia between month 5 and birth; these cannot be replaced (Fig. 1 1-1~. Shortly after birth, the primary oocytes are arrested in late prophase of meiosis I. With the onset of puberty, the increas- ing production of gonadotropins stimu- lates the resumption of meiosis. The first meiotic division is completed just before ovulation (Fig. 11-2), and the second divi- sion is completed only when fertilization ensues. As the oogonia enter into meiosis, human cells contain a diploid number of chromo- somes and four copies of DNA. The two divi- sions of meiosis entail halving the number of chromosomes (meiosis I) and then halv- ing the amount of DNA (meiosis II). The female pronucleus is formed at fertiliza- tion and ultimately combines with the male pronucleus to reestablish the diploid state. Several aspects of oogenesis are note- worthy. First, exposure of female concep- tuses to germ cell toxicants is of particu- lar importance, because damaged oocytes never can be replaced. The stock of gametes is large, however, and some destruction undoubtedly can be tolerated without re- FEMALE REPRODUCTIVE TOXICOLOGY auctions in fertility or length of repro- ductive life. Second, mitosis in females ceases in utero, so the incidence of repli- cation-dependent mutations should be lower than in the male, whose stem cells divide continuously. Third, exposures or physiologic deterioration sustained during the long resting phase of oocyte maturation may account for the increased sensitivity of female germ cells to meiotic errors, such as nondisjunction. Whether a blood/follicle barrier exists to protect the oocyte from toxic substances is con- troversial, and more research on this ques- tion is needed. The mature oocyte appears to be capable of repairing its own damaged DNA (Pederson and Mangia, 1978) and, af- ter fertilization, damaged sperm DNA (Generoso et al., 1979~. DEVELOPMENT OF THE FEMALE REPRODUCTIVE TRACT The female reproductive tract develops on the urogenital ridge early in fetal life. The paired Mullerian ducts form the fallopian tubes, and fuse to form the uter- us and cervix. The vagina develops partly from the Mullerian duct and partly from the urogenital sinus, whereas the external genitalia evolve from the genital tuber- 7.0 FIGURE 11-1 Changes in total population of germ cells in human ovary with increasing age. Reprinted with permission from Baker, 1971. Copyright 1971, C.V. Mosby Co. E 5 0 - u' a) c: o' 3.0 At 1.0 0.6 0.3 Age Birth (months after conception) 30 50 Age (years)
INTRODUCTION DEVELOPMENTAL EVENTS Multiplication by mitosis Migration to genital ridge STATE OF GERM `3 CE' ~ -A ~ 6) PRIMORDIAL (if) GERM CEI I (I)' ~ |BIRTH-rabbit, ferret, mink, vole, hamster >a) (3_~ Final interphase 6) DNA synthesis ~ Meiotic prophase begins (3 PRIMARY ~ Mets BlRTH-N~st mammals > ~ ~ - . ,= Growth of oocyte and follicle PUBERTY Fo licular maturation tOVULATION- dog, fox > First Heists divisor begins Spend penetration- dog, fox First polar body emitted (may divide) |OVULATION-Most mammals > Sperm penetration- Most mammals Second meiosis division fertilization, and emiss on of second polar body LOGON ~; rid ~ - - ~ SECONDS \ J `:, OOCnE PRONUCLEATE () EGG (OOTID) cle, labioscrotal folds, and labioscrotal swellings. The hypothalamus and anterior pitui- tary, which are important in regulating reproduction, form from the developing central nervous system and the oral ecto- derm, respectively. Sex-based structural and functional differences in the hypo- thalamus of animals depend on the hormonal milieu early in development (Gorski, 1968; Brawer and Naftolin, 1979~. A similar sexual differentiation is believed by some to occur prenatally in the human hypothala- mus; this may influence later psychosexual orientation and reproductive function (Ehrhardt and Meyer-Bahlburg, 1979~. Prenatal exposure to the synthetic hor- mone diethylstilbestrol (DES) is the best-documented example of toxic influ- ence on the developing female reproductive tract. Anomalies of vagina, cervix, uter- us, fallopian tubes, and mesonephric duct remnants found in humans with hysterosal- pingography and usually detected after reproductive maturity seem to originate from disturbances in embryologic develop- ment(NewboldandMcLachlan, 1982~. Among women affected, these abnormalities of 151 FIGURE 11-2 Life cycle of a female germ cell. Re- pnnted with permission from Austin and Short, 1982. Copyright 1982, Cambridge University Press. structure may compromise reproductive performance (Herbst et al., 1980~. Whether prenatal exposure to DES also alters the hypothalamus or pituitary is being inves- tigated (for example, in a study of sexual activity and functioning in DES daughters (Meyer-Bahlburget al., 1985~. MATURATION Hormonal Changes Assumption of adult female reproductive function involves major increases in ovar- ian and adrenal production of steroids that induce maturation of cells that are responsive to sex steroids. The steroid production changes-at least those due to increased ovarian activity at puberty- are controlled by the hypothalamic- pituitary axis. During childhood, the hypothalamus is very sensitive to sex steroids, and low concentrations suppress gonadotropin- releasing hormone (GnRH) (Fig. 11-3~. This sensitivity decreases with the onset of puberty, and increasing concentrations of steroids are required for suppression.
152 o _ Cry X ,a ~ O 6) Q o) cS ° .> 0 A) Cr C FEl~4LE REPRODUCTIVE TOXICOLOGY Increase Decrease Maturation of negative feedback mechanism . i' , ~ ~ Increased / release of LRF 1 1 1 Increased pituitary responsiveness to LRF Rising gonadotropin levels Increased gonadal responsiveness to gonadotropins Rising sex steroid levels Sleep associated increase in LH secretion: episodic secretion of LH 1 '\1 Activation of positive feedback mechanisms I : I Fetus Infancy and childhood Puberty Adult FIGURE 11-3 Schematic illustration of development of hypothalami~pituita~y-gonadotropin-gonadal interrela- tionship In relation to onset of puberty. From Grumback et al., 1974. As a consequence, a new set point is reached, and pulsatile GnRH releases are sufficient to raise circulating luteiniz- ing hormone (LH) and follicle-stimulating hormone (FSH) to the point where follicular activity ensues. The resulting increased ovarian steroid production has negative and positive feedback effects on gonado- tropin release. The underlying hypothala- m~c mechanism that initiates these hor- monal changes at puberty is thought to be the augmentation of pulsatile GnRH secre- tion under the control of an oscillator in the arcuate nucleus (Knobil, 1980~. Late in puberty, positive feedback effects Menarche of estradiol (E^) lead to a DreovulatorY LH surge. Terasawa (1985) showed that these patterns usually cannot be evoked at earlier ages in the absence of the hor- monal milieu of a sexually mature adult. Maturational changes in the control of GnRH secretion occur in monkeys ovariec- tomized before puberty and thus are proba- bly independent of ovarian steroids. Simi- lar conclusions are drawn from the pattern of gonadotropin secretion in women with Turner's Syndrome (a form of gonadal dys- genesis). The nocturnal pattern of LH secretion ~ ,s~ also changes dramatically during puberty (Boyar et al., 1973) (Fig. 11-4~. The fre- quency of pulsatile LH release over 24 hours and the response to E2 are sensitive indicators of hypothalamic maturation, but cannot be readily applied as general assays to evaluate effects of toxicants on humans. In adult primates, the matura- tion of ovarian follicles depends on the frequency of LH pulses and is impaired by slight slowing of these pulses (Pohl et al., 1983~. - A major landmark of maturation is men- arche, the beginning of the menstrual func- tion. This usually occurs between the ages of 9 and 16 years, with a mean of 13 years (Tanner, 1981~. That precocious menarche can occur as early as the age of 1 month indicates that gonadal cells that respond to hormones have differentiated to the stage of hormonal competence in neonates . ~ . and can respond to endocrine signals at any age. Menarche and other characteris- tics of puberty can be delayed indefinite- ly, as in persons with Turner's syndrome and other types of gonadal dysgenesis.
INTRODUCTION 1O E 8 ~ 6 ~ 4 I 2 O 18 16 1a 1` 1( o _ ~ Sleep Prepuberty I J Start ~ For 1 1 1 1 1 1 1 22.00 02.00 06.00 10.00 14.00 18.00 22.00 CLOCK TIME Miril~t~ n~lh~rtv _ 1 1 1 1 1 1 1 22.00 02.00 06.00 10.00 14.00 18.00 22.00 CLOCK TIME 153 10 8 6 4 O 2t o 1 1 W\; 22.00 02.00 06.00 10.00 14.00 18.00 22.00 CLOCK TIME Sleep Adult I ~/~ - Start l l l l l l l 22.00 02.00 06.00 10.00 14.00 18.00 22.00 CLOCK TIME FIGURE 11~ Daily plasma LH patterns in a representative prepubertal girl (9 years), early pubertal (15 years), late pubertal (16 years), and young adult (23 years) males. Sleep stage pattern is depicted for each nocturnal sleep penod. Source: Repnnted with permission from Weitzman et al., 1975. The first menstrual cycle signals matur- ation of the hypothalamus and the reproduc- tive tract. Initial cycles often are ir- regular and can be anovulatory (Treloar et al., 1970; Tanner, 1981) (Fig. 11-5~. The length of the anovulatory phase varies and can be several years (Ashley-Montague, 1957; Ojeda et al., 1980~. Little is known about how particular toxicants influence menarche and the onset of fertility. But it is known that puberty can be delayed by exogenous factors such as stress, nutritional deficiencies, and emotional distress. (Girls with emo- tional distress also usually have defi- ciencies of gonadotropins that imply hypo- thalamic dysfunctions.) Environmental stress delays puberty in laboratory ro- dents. The onset of puberty is associated with achievement of a critical body size and fat content (Frisch, 1980; Tanner, 1981~. The restricted diets often chosen by athletes and dancers can delay menarche until the age of 20 years (Frisch, 1985~; similar effects occur during anorexia nervosa. Loss of more than one-third of body fat at any age causes reversible amen- orrhea(Frisch, 1985~. Although the neuroendocrine basis of the effects of stress and diet on puberty is unclear, some toxicants that interact with the stress-mediating hypothalamic- pituitary and sympathoadrenal systems might influence puberty. Such indirect effects could also result from toxicants that influence appetite or nutrient ab- sorption. An example of a toxicant that affects puberty in rodents is the insecticide DDT. Exposure of neonatal rats to DDT causes major changes in neuroendocrine func- tions, including early puberty and a syn- drome of delayed-onset, persistent estrus in association with a polyfollicu- lar ovarian status (Heinrichs et al., 1971~. This permanent reproductive im- pairment in female rodents resembles the neonatal masculinization of the hypo- thalamus and the polyfollicular, anovula- tory ovarian syndrome caused by exogenous steroids (Gorski, 1971; Mobbs et al., 1984~. Whether such effects of DDT are limited to a critical period during devel- opment is unknown. Exposure of rodents just before or after birth to estrogens and other steroids has profound effects
154 FIGURE 11-5 Comparisons of ag~related changes in ovulatory cycle length distribution from longitudinal studies of C57BL/6J nuce and humans. The ages are scaled to midlife. Sources: (Mice) Reprinted with permission from Nelson et al., 1982. Copyright 1982, Society for the Study of Reproduction. (Human) Reprinted with permission from Treloar et al., 1970. Copyright 1970, Allen Press, Inc. On adult reproductive functions that can be manifested at puberty or can induce precocious cessation of fertile cycles (delayed anovulatory syndrome). The mech- anism by which DDT causes persistent estrus may involve an estrogenic action, since DDT has a uterotrophic effect (Bitmap et al., 1968) and also can bind to cytosolic E2 receptors (Robison et al., 1985~. This example shows how environmental toxicants can interact with neuroendocrine matura- tion. No analogous phenomenon in humans has been found. Women exposed in utero to DES have normal age at menarche and appear to be fertile; menstrual cycles are report- ed to be regular in some studies (Barnes, 1979) and irregular in others (Herbst et al., 1980~. FEMALE REPRODUCTIVE TOXICOLOGY Con 9~ I z llJ 6 llJ c-) 5 50 C57BL/6J MICE 8 _ o l``o \ ,S . \, o \ o ~ to `b,, so, ~ °- -o_ trio- -O--°~ 4 _ 1 1 1 1 1~, 1 1 1 ~ 1 1y, 1 1 1 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 AGE (MONTHS) J - z C) 85 - , 90 80 - HUMAN 75 _ 70 _ 65 -50 z As ~ 1 ; 20 _ 15 _ 10 I I I . I, I . I . I . I . I, ~ I I . I . 1 . I . I . I .,, I, 1 . I . I . 0 4 20 24 28 32 36 40-8 -4 0 MENSTRUAL CHRONOLOGICAL YEAR PRE-MENOPAUSAL YEAR Adrenarche YEAR Adrenarche is the onset of menstruation and other physiologic changes of puberty induced by hyperactivity of the adrenal cortex. The concentration of the weak adrenal androgen dehydroepiandrosterone (DHEA) increases after age 6, several years before the increased secretion of estrogens, gonadotropins, and prolactin in pubertal girls (Hopper and Yen, 1975; Ojeda et al., 1980; Cutler and Loriaux, 1980; Reiter and Grumbach, 1982~. DHEA appears largely as a sulfate, DHEA-S, in human blood. Another androgen, ~-4- androstenedione, increases at slightly later ages. Accelerated growth and appear- ance of pubic and axillary hair are associ- ated with increased blood concentrations of adrenal steroids. The hormonal mechan- isms underlying adrenarche are not known,
INTRODUCTION but seem to be independent of mechanisms that influence gonadal development (CutlerandLoriaux, 1980; ReiterandGrum- bach, 1982~; children with gonadal dys- genesis can have normal adrenarche (Boyar etal.,1973~. Adrenal insufficiency in humans may delay puberty (Boyar et al., 1973), but this association is not found in all cases (Reiter and Grumbach, 1982~. Adrenalec- tomy of rats delays puberty, whereas cor- ticoid replacement restores the normal onset (Ramaley, 1978~. There is general agreement that adrenal steroids are not obligatory for maturation of hypothalamic controls over the gonad in either sex. Thelarche Thelarche (the beginning of breast de- velopment) is an estrogen-dependent pubertal event and can be evaluated by well-established criteria (Marshall and Tanner, 1969~. Its absence by the age of 13 years is diagnostic of delayed puberty (Ojeda et al., 1980~. A number of hormones are involved, including estrogens, pro- gesterone, corticosteroids, prolactin, and growth hormone. However, prolactin and E2 are considered pivotal for thelarche (Kleinberg, 1980~. If children of either sex are accidental- ly exposed to estrogens, early stages of thelarche occur. Premature and abnormal breast growth in children who were exposed to an estrogen-containing ointment has been documented (Halperin and Sizonenko, 1983~. An outbreak of precocious thelarche in Puerto Rican children is still unex- plained(Haddocketal., 1985;New,1985~. CYCLIC OVARIAN FUNCTION The ultimate purpose of cyclic ovarian function is to provide viable oocytes for fertilization by sperm ascending through the female genital tract. Ovarian hormones control the oviductal environment for fertilization and gamete transport and the endometrial environment for implanta- tion and embryonic development. Synthesis of the steroids responsible for this se- quence of events must be timely and re- quires maturation of the follicle, re- 155 lease of the oocyte, a normally function- ing corpus luteum, and ovarian responsive- ness to the presence of a conceptus. Unlike that in men, reproductive func- tion in women is characterized by cyclic fluctuations in pituitary gonadotropins and sex steroids (Fig. 11-6~. These hor- mones have a dynamic relationship with positive and negative steroid feedback on release of gonadotropins from the pitui- tary. Other nonsteroidal regulatory fac- tors of ovarian origin are also involved in this dynamic system. Cyclic ovulatory function starts with sexual maturity and continues until the fifth decade of life. This cyclicity de- pends on continual maturation of ovarian follicles. which is stimulated bv a normal- ly functioning hypothalamic - pituitary axis. The process begins with the recruit- ment of a cohort of competing follicles that are able to respond to gonadotrop- ins. Once a dominant follicle is selected, the remaining follicles in the cohort are suppressed. Approximately 65 days before ovulation, the primary oocyte begins maturing within a primordial follicle (Gougeon, 1982~. The early stages of this process appear to be independent of gonadotropin stimula- tion. The primary oocyte is surrounded by a single layer of cells-presumably forerunners of granulosa cells-and little is known of the physiology of early fol- licular development. Only the final 14 days of follicular maturation appear to be influenced cyc- lically by gonadotropins in primates. Women with surgically absent or irradiated pituitaries can be made to ovulate with a relatively short course of human meno- pausal gonadotropin (hMG), with hCG as a surrogate midcycle LH surge (Schwartz and Jewelewicz, 1981~. A typical course of hMG requires 10- 14 days to develop pre- ovulatory follicles, according to sono- graphic follicular size and estrogen pro- duction. The follicular phase in primates lasts approximately 14 days. Ablation experi- ments in nonhuman primates have suggested that, from the midpoint of the follicular phase through the late luteal phase, a dominant cyclic structure-a preovulatory
156 o 11 1 0 ~ 8 4 o 50 C O _ a) E 120 `,) _ a, I ~,~ 60 o 180t O - 6- a) ° oh _ ~ ~ 2 ct ~ ~ ~ E ~ 1 ~ ~ Menses Ovulation Menses ~1 0 \ Ovulation Menses (em ','0 e/ 1 ~ ~10 FEMALE REPRODUCTIVE TOXICOLOGY Ovulation Menses Ovulation ~~No:°~ ~ 1~0~°\i ~ = ·N ~°'o/.l,°~: r All an. ~ .. - ., ~;~ ~ ~1 1~el . .. .. .. .. .. .. .. .. 1' '' '' '' " I~ t~ ll ~~ ~ 1 ~ ~i~'.~l m . . . 4 ~ 0 ~ 0 . ~ .o x - - cervix: A /\ ~ , . .. .. .. .. .. .. .. .. .. .. .. .. a_ O ., ,, .. . .. .. .. .. .. .. .. .. l l l l l _ l l .1 ~ 36~e ~ · 'u~ ~ a) 36.2 1 5 10 15 20 25 1 5 10 15 20 25 1 s 10 15 20 25 Day of Cycle l I 1 ~ -85 -65 -45 -25 - 1 5 - 1 0 -5 0 Follicular Development: Days Before Ovulation FIGURE 11~ Selected events occurring during development of an ovulatory follicle. Note that development on cupies three full cycles. Parameters include size of developing follicles measured by ultrasound or during laparo- scopy, gonadotropin concentrations in serum; estrogen and progestogen concentrations in serum, saliva, and urine; ratio of urinary steroids; state of endometrial proliferation; relative cervical mucus volume; and basal body temperature. An average 2~29 day cycle was used as a model to which were fitted several results. Confidence integrals for salivary progesterone include ~ standard deviation of the mean; for urinary steroids, the 80% conf~- dence interval is shown. Source: Repnnted with permission from Campbell, 1985. an m 20 Q 1 C4)
INTRODUCTION follicle or corpus luteumis present. Removal of this structure at any time re- sults in recruitment of another dominant follicle and ovulation in approximately 14 days (Goodman et al., 1977~. The best explanation of this is that follicles are always maturing to a critical point at which either a dominant follicle is selec- ted or follicular atresia results. FSH appears to be the gonadotropin responsi- ble, in that LH does not rise early in the follicular phase coincidental with the process of follicle selection, as FSH does, and FSH can induce follicular devel- opment by itself. Asymmetrical steroid secretion of the ovary indicates the presence of a dominant cyclic structure by day 7 of the menstrual cycle, as determined by estrogen and pro- gesterone concentrations in the venous drainage of the ovaries (diZerega et al., 1980). How primates release only a single oocyte per cycle is unknown, and the mech- anism by which other developing follicles are suppressed remains controversial. Available data point to a local effect, inasmuch as removal of the corpus luteum results in ovulation from the contralater- al ovary. When the cyclic structure (eith- er a follicle or corpus luteum) is excised, FSH concentrations are 3 times higher than in a cycle in which the structure is not removed (Goodman and Hodgen,1977~. De- spite this relatively high FSH content, a single ovulation still occurs 14 days after ablation, with steroid production at appropriate magnitudes for a single follicle. Steroid production by preovulatory follicles parallels their health and states of differentiation, and estrogen production is the hallmark of follicular health (McNatty et al., 1976~. The cel- lular source of estrogen secretion is con- troversial and involves a complex biosyn- thetic process that requires interaction between follicular compartments. Some studies indicated that the theca (on the follicular sheath) is the major source of estrogen biosynthesis (Channing and Coudert, 1976; Younglai and Short, 1970), whereas others suggest that thecal and granulosa compartments are needed (Falck, 157 1959~. This "two-cell idea is strength- ened by the observation that granulosa cells are unable to synthesize androgens in appreciable amounts (Short, 1962; Bjer- sing and Carstensen, 1967), although the granulosa cell compartment processes greater aromatase activity than the thecal compartment (Schomberg, 1979~. Thecal cells have not been demonstrated to have FSH receptors (Richards et al., 1976), whereas granulosa cells have FSH-induci- ble aromatase activity (Dorrington et al., 1975~; therefore, FSH stimulation of aro- matase activity might be a key regulatory step in follicular estrogen production. Estrogen appears critical to follicular health. It is mitogenic to granulosa cells and inhibits premature biochemical lute- inization, and high concentrations are found in the fluid of healthy follicles (McNatty and Baird, 1978~. Development of FSH-induced aromatase activity could be critical in preventing atresia. That is probably the mechanism by which thera- peutic agents like clomiphene citrate (which raises FSH content) initiate fol- licular maturation that leads to ovula- tion. Attainment of granulosa cell aroma- tase activity appears to be critical in controlling the number of follicles matur- ing in each cycle. As preovulatory estrogen content rises, gonadotropin release is inhibited and FSH is suppressed differentially. Paradoxically, a sustained estrogen con- tent for approximately 24-36 hours has a positive feedback effect and results in a surge of both gonadotropins. That causes a series of events within the fol- licle, including intrafollicular pros- taglandin synthesis, terminal oocyte maturation, a shift in steroidogenesis from estrogen to progesterone by the granu- losa cells, morphologic luteinization, and, ultimately, rupture of the follicle and release of the oocyte with its invest- ment of cumulus cells into the peritoneal cavity (LeMaire et al., 1973~. Even before ovulation, preparation for luteinization by granulosa cells is well under way. The high estrogen content in follicular fluid is responsible for the dramatic increase in the number of granulosa cells, as well as for the appear-
158 ance of LH receptors. A decrease in serum E2 after the gonadotropin surge is consis- tent with blocked aromatase activity. Serum and follicular fluid progesterone increase concomitantly and mark the in- itiation of the luteinization process (McNatty et al., 1976~. Neovasculariza- tion of the granulosa compartment begins as vessels penetrate the basement membrane and invest the luteinized granulosa cells. Recent observations in which circulating steroids around the midcycle were careful- ly measured showed a rise in progesterone a few hours before follicular rupture (Horning et al., 1981~. Without sustaining factors secreted by a conceptus, a corpus luteum undergoes a programmed demise, with a life span of approximately 14 days. As noted earlier, peripheral progesterone begins to in- crease with the initiation of the LH surge and continues to increase until the mid- point of the luteal phase, when it begins a gradual decline that results in menses. In humans, hCG maintains the corpus lute- um during early pregnancy. Circulating hCG can be detected a few days after luteal progesterone peaks and prevents the pro- grammed involution of the corpus luteum (Saxena et al., 1974; Catt et al., 1975~. Excision of the corpus luteum or the ovaries at 7-8 weeks of gestation does not interrupt pregnancy; that suggests that the fetoplacental unit takes over the critical hormone production function necessary to maintain pregnancy (Csapo et al., 1972~. hCG content is high at this point in pregnancy and gradually declines after 13 weeks of gestation. No further role of the ovary is apparent in the devel- opment of human pregnancy. Screening for ovarian toxicity in viva is difficult, because of the interrela- tionships with the hypothalamic-pituitary axis and the fact that clinical problems arise only when there are major functional changes. The observation of disruption of the reproductive cycle does not permit discrimination of ovarian or central ner- vous system action. More detailed hormonal studies over some period are required. Effects of exposure to toxicants on cyc- lic ovarian function have not been inves- tigated widely, but several observations FEMALE REPRODUCTIVE TOXICOLOGY suggest that occurrence of menses is sensi- tive to environmental influences. Nutri- tional factors, stress, and vigorous exer- cise are well-established risk factors in anovulation (Warren, 1982, 1983; Green et al., 1986~. Cycle length was examined in women exposed to inorganic mercury vapor, with some indication that higher levels of exposure increased the risk of oligomenorrhea (DeRosis et al., 1985~. Hormones and central nervous system toxi- cants, such as metals, can disturb menstru- al function. Amenorrhea and altered men- strual hormone levels are parts of an ovar- ian failure pattern observed in women treated with antineoplastic agents, such as cyclophosphamide (Chapman, 1983~. Polymenorrhea has been found in women work- ing with synthetic hormones (Harrington et al., 1978), and menstrual dysfunction has been suggested by some studies of pre- natal exposure to DES (Bibbo et al., 1977; Barnes, 1979; Peress et al., 1982~. Sol- vents have been associated with menstrual dysfunction (WHO, 1986~. However, a re- cent study of women working with styrene found no increase in menstrual disorders (Lemasters et al., 1985~. The mechanisms of female reproductive toxicity vary. For example, a toxicant might mimic the action of naturally occur- ring reproductive hormones. Many chemi- cally unrelated compoundsincluding stilbene derivatives (e.g., DES), indus- trial chemicals (e.g., PCBs), and pesti- cides (e.g., DDT)exhibit estrogenic activity in bioassays, and thus have the potential to alter the normal estrogen feedback relationship between the gonad and the brain and so disrupt ovulation in a manner analogous to that of oral contra- ceptives. Environmental agents might alter hormone synthesis, storage, re- lease, transport, or metabolism. For ex- ample, compounds with estrogenic activity can cause luteolysis and inhibit the pro- duction of progesterone; as a consequence of these and other considerations (such as alterations in fallopian tube and en- dometrium functions), DES has been used as a morning-after pill to victims of sexu- al assault. Environmental agents also might cause gamete cytotoxicity. In a generally monotocous species, such as the
INTRODUCTION human, substantial loss of oocytes can be tolerated without disrupting the men- strual pattern. The injury is apparent only long after exposure, when premature ovarian senescence occurs, which makes identification of the responsible agent particularly difficult. FERTILIZATION Fertilization in the course of a repro- ductive cycle involves several processes, including: · Sperm-egg attachment. · The fertilized oocyte's blocking of penetration by additional sperm. · Completion of meiosis, with extrusion of the second polar body. · Development of the two pronuclei. Fertilization failure or delay might occur if exposure to toxicants interferes with timely ovulation or gamete transport. Toxicants also have the potential to alter oocyte biochemistry, affecting oocyte activation and pronuclear formation. During the pert-implantation period (weeks 1-3 in humans), the fertilized egg moves down the oviduct to the uterus, where it implants approximately 8 days after conception, and then blastulation and gastrulation occur. Insults that occur during this period can result in repair through compensatory hyperplasia, in embryo lethality, or in diverse malfor- mations that suggest genetic damage to the early zygote. Generoso et al. (1987) reported that early exposure of mouse zy- gotes to mammalian germ-cell mutagens induces a variety of congenital defects, as well as death. In this part of the re- port, one marker of pregnancy and early loss that has been used in epidemiologic studies is discussed. Biologic markers related to pregnancy are discussed in de- tail in Part III of this report. REPRODUCTIVE SENESCENCE Menopause is the cessation of menstrua- tion. The last menstrual cycle usually occurs at the age of 42-58 years, with a median of about 50 (MacMahon and Worcester, 159 1966; Gosden, 1985~. Menopause is preceded by gradual cycle lengthening and irregu- larity over 5 years or more (Fig. 11-4~. Perimenopausal cycles may be prolonged and anovulatory and are associated with decreased E2 and progesterone and in- creased LH and FSH (Sherman et al., 1976; Metcalf et al., 1981~. Foreshortened, possibly anovulatory, cycles also occur in association with abbreviated follicu- lar phases. Menopause is assumed to have occurred if amenorhea has persisted for at least 6 months. Decreasing fertility (ability to pro- duce offspring) is a hallmark of approach- ing menopause and follows remarkably simi- lar trends in modern populations of indus- trial countries, as well as in groups, such as the Hutterites, that try to maximize fertility and live offspring (Fig. 11- 7~. Marked decreases in fertility can occur 5 years before cycles become obvious- ly irregular. Therefore, menstrual cycle regularity is not a reliable assessment of fertility. Observed age-related de- crease of fertility in populations results from an increased number of women who have become sterile for many different reasons, as well as a reduced rate of con- ception among those who are still fertile (Federation CECOS, 1982; Menken et al., 1986~. Down syndrome and some other chromoso- mal abnormalities in offspring increase sharply after age 30 (Ferguson-Smith, 1983~. About 50% of all first-trimester spontaneous abortions in humans are aneu- ploid (Porter, 1986), so the doubling in spontaneous abortion from the age of 20 to the age of 40 could reflect an increase in chromosomal abnormalities. These ab- normalities often are lethal, and the evi- dence does not support weakened selection against abnormal fetuses with increasing maternal age (Golden, 1985~. A promising animal model, the rat, shows that delayed ovulation can increase embryo abnormali- ties (Page et al., 1983), thereby suggest- ing a role of age-related cycle lengthening to increased birth defects. However, no relation has been shown between cycle regu- larity, early menopause, and Down syndrome (Sigler et al., 1967~. The inadequate understanding of this major phenomenon
160 FEMALE REPRODUCTIVE TOXICOLOGY 8 600 lo ~ 500 LL 400 300 Or: LL ~ 200 lo: 6 100 , Hutterites (USA) 1920s marriages - ,';' ._ ~ -~_ ~ ~ 1 7th, 1 8th century French villages "> 17th, 18th century Europe "^ \ \ 'air,,: contemporary Third World \ ~ \ 'v\\ I I I 1 1 1 ~ - ~ 15-19 20-24 25-29 30-34 35-39 40-44 45-49 AGE GROUP FIGURE 11-7 Ag~ppecific fecundity in selected populations that do not practice contraception: rates per 1,000 mamed women. Source: Based on data In Leddon, 1977. also hampers studying the interactions of toxicants, age, and birth defects. The major cause of menopause is deple- tion of the ovarian stock of oocytes, which declines from birth onwards in all mammals (Fig. 11-8~. Individual variations in the onset of menopause are thought to re- sult from different rates of ovarian oocyte loss or different sizes of the ini- tial stock (Nelson and Felicio, 1987; Rich- ardson et al. 19871. Even within the same inbred strain, mice show extensive ~nd~- vidual differences in initial oocyte stocks and in the numbers of oocytes re- maining at the approach to acyclicity (Gosden et al., 1983~. Rodent models also show age-related ovarian depletion and lengthening of fertility cycles (Fig. 11- 5~. As ovarian oocytes and growing follicles become depleted, circulating concentra- tions of estrogens and progesterone de- crease to those found in ovariectomized women. Among the major changes associated with decreased steroids are onset of hot flushes, increases in blood LH and FSH, and atrophy of most organs that respond to sex-steroids. These changes vary wide- ly, but replacement of ovarian steroids usually prevents or attenuates them. Ovar- ian steroids have some effect on postmeno- pausal osteoporosis; oophorectomy in young women can precipitate premature bone loss, and steroid therapy appears to reduce the risk of age-related fractures. How- ever, inasmuch as bone loss begins about 10 years before measurable deficits in blood estrogens, other factors that do not depend directly on changes in blood estrogen concentrations must also be in- volved. DHEA-S decreases progressively after the age of 30 years and is not related to menopause (Orentreich et al., 1984~. Menopause might be influenced by toxi- cants and other environmental factors. For example, smokers of 14 or more cigar- ettes per day have menopause as much as 2 years earlier than nonsmokers and smokers of fewer than 14 cigarettes per day (Jick and Porter, 1977, Van Keep et al., 1979; Lindquist and Bengtsson, 1979; Kaufman et al., 1980~. Benzo~a~pyrene, a carcino- gen that occurs in tobacco smoke and urban air, can kill oocytes in mice (Mattison and Thorgeirsson, 1977) and may be an ac- tive causal agent in premature ovarian oocyte loss and early menopause. In addi- tion, cigarette smoke has been shown to be antiestrogenic (Michnovicz et al., 1986~.
INTRODUCTION in o o IL o 111 ~ 800,000 z 700,000 600,000 500,000 400,000 300,000 200,000 1 00,000 1 4,000 1 2,000 1 0,000 8,000 6,000 4,000 2,OOO MOUSE . - _ - }4 - of_ . . . O 100 200 300 400 500 600 AGE (DAYS) HUMAN : ~ rP~ . . 20 30 40 50 AGE ~EARS) . . O 10 Other environmental influences on meno- pause are less certain. Several reports have indicated a trend of increasingly later ages at menopause during the past century (Frommer, 1964), but other reports have not confirmed this trend (Van Keep et al., 1979; Gosden, 1985~. Economically underprivileged populations may also have slightly earlier menopause (MacMahon and Worcester, 1966; Soberon et al., 1960), but no general effect is established (Gosden,1985~. Toxic environmental estrogens are im- portant, because they resemble some as- pects of reproductive senescence. Sheep fed on phytoestrogen-containing clovers develop a permanent infertility syndrome 161 FIGURE 11-8 Loss of ovarian oopytes during aging In the mouse and human. Sources: (Mouse) Rev drawn from Jones and Krohn, 1961; (human) redrawn from Block, 1952. with hypothalamic histopathology (Adams, 1976, 1977~. These changes are intriguing- ly similar to the changes in aging acyclic rodents (Schipper et al., 1981; Finch et al., 1984) that can also be induced by chronic exposure to estrogens (Brewer et al., 1983; Mobbs et al., 1985~. In addition to neural damage, sheep with clover disease have impaired luteal function and sperm transport (Adams, 1 983; Adams et al., 1 98 1; Adams and Martin, 1983; Lightfoot et al., 1967~. Phytoestrogens interact with en- vironmental trace metals, cobalt inten- sifies the effects of phytoestrogen, and selenium blunts the effects of cobalt (Gardiner end Narin, 1969~.
