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.
14 Biologic Markers of Nonconceptive Menstrual Cycles Use of biologic markers to evaluate the effects of environmental toxicants on the normal menstrual cycle is a recent concept. Biologic monitoring can use a variety of body fluids (including fluids specific to reproduction) and tissues (Tables 14-1 and 14-2~. This section dis- cusses biologic markers that can be evalu- ated during the nonconceptive menstrual cycle. To produce menstrual dysfunction, en- vironmental agents have to have systemic toxicity that is manifested in the repro- ductive system or specificity for the re- productive system. Agents with activity peculiar to the reproductive tract might be difficult to identify because of the following: · The complex nature of the reproductive process. · The presence of spontaneously occur- ring disease with identical symptoms. · The act of procreation is largely vol- untary and, therefore, lack of conception might be a matter of choice rather than toxic effect. · The lack of a reliable epidemiologic data base on reproductive events or con- tinuing surveillance of reproductive with the reproductive cyclesuch as markers in the general population. basal body temperatures, cervical mucus The only reproductive tract fluid so far studied for the purpose of biologic markers is follicular fluid collected from patients undergoing in vitro fertiliza- tion/embryo transfer (IVF/ET). Several xenobiotics have been identified in fol- licular fluid (Trapp et al., 1984), but associations with adverse effects have not been investigated systemically. That is of particular concern, because follicu- lar fluid is in direct contact with the oocyte and the steroidally active granulo- sa cells. Toxic agents in the follicular fluid have the potential to alter granulo- sa-luteal cell function, as well as the developing early embryo. The menstrual cycle is a normal physio- logic event, and the fundamental markers of female reproductive function are the phenomena of the reproductive cycle, which are the following (in order of in- creasing complexity): · Characteristics of the normal men- strual cycle-such as interval, regular- ity, duration, and character of menses and the presence of premenstrual molimina (see below). · Bionhysical chances associated 179
180 FEMALE REPRODUCTIVE TOXICOLOGY TABLE 14-1 Human Body Fluids Potentially Useful In Measuring Biologic Markers F1uid Availabiliiva Comments Nonreproductive tract Buids Blood Unne Saliva Cerebrosp~nal fluid (CSF) Reproductive tract fluids Vaginal secretions Centrical secretions Utenne seminal fluid Tubal secretions Follicular fluid Peritoneal fluid Menstrual effluent + + + + + + + + + + + + + + + + + + + + + + Significant temporal fluctuations in hormone concentrations assessed in blood Provides good indication of cumulative expm sure and includes accumulated metabolites An ultrafiltrate of plasma Measurement of neurotransmitters limited by CSF/brain barrier Cycl~spec~fic, use hampered by bacterial con- tam~nation, often used in animal studies Cycle-specif~c, needs further development Poorly characterized secretory products, cycled specific, require further development Poorly characterized secretory products, cycled specific, need further development Only In stimulated cycles In IVF/ET,b requires further development Contains exudate of ovary, requires further den velopment Limited by autolysis, requires further develop ment aIncreasina number of +'s indicates more readily available. Some fluids available only from patient samples, such as from surgical patients. bIn vitro fertilizations/embryo transfer procedures. changes, vaginal cornification, and sexu- al behavior. · Endocrinologic characteristics of the ovulatory cycle. A normal menstrual cycle is defined by a pattern-every 26-30 days-(Treloar et al., 1970) of an ill-characterized con- stellation of symptoms termed premenstru- al molimina (e.g., breast tenderness, bloating, and mood swings followed by va- ginal bleeding). Ovulatory menses, typi- cally, is associated with some degree of lower abdominal cramping due to the effect of prostaglandins on the myometrium. The menstrual effluent comprises autolyzed endometrium, tissue fluid, and blood that has undergone clotting and lysis. Passage of excessive blood clots with the menstrual flow is unusual in ovulatory cycles, unless mechanical problems, such as endometrial polyps or leiomyomata, are present. Some biophysical changes can be used to determine whether ovulation has occur- red, such as basal body temperature shifts after ovulation and changes in the charac- ter and quality of cervical mucus. Other descriptors of the menstrual cycle, such as cyclic fluctuations in vaginal cytology and changes in sexual behavior, are subject to nonendocrine influences and are less useful for that purpose. Any agent that disrupts normal cyclic menstrual function may be described as causing reproductive toxicity. The repro- ductive tract is susceptible to disruption by environmental agents that affect the cerebral cortex, hypothalamus, pituitary, ovaries, fallopian tubes, or uterus. Aber- rant menstrual cycles can result from dis- ruptions of a diverse group of functions, including: · Synthesis, storage, transport, re- lease, and metabolism of neurotransmit- ters, gonadotropin-releasing hormone (GnRH), gonadotropins, and ovarian regu- latory peptides. · Gonadotropin responsiveness. · Ovarian steroidogenesis.
NONCONCEPTI~E MENSTRUAL CYCLES 181 TABLE 1~2 Human Tissues Available for Use in Measuring Biologic Markers of Reproductive Toxicity Hypothalamus Pituitary Uterus Endometrium Myometrium Fallopian tube + + + + + + + Ovarian cells Granulosa-luteal + + Thecal Stromal Adipose tissue Cytologic specimens + + + + + + Vaginal Cervical En d o met rial Peritoneal fluid + + + + + + + + + aIncreasing number of + 's indicates more readily available. Autopsy material Autopsy material Cycle specific, difficult to culture, heterogeneous cell pop- ulation Not endocrinologically active Cycle-specif~c secretory products, unclear physiologic signift- cance Large-culture methods possible, available in late follicular phase Organ cultures only, difficult to purify Minimal hormone secretion Useful as internal dose marker comparable with ovary, owing to lipid composition and steroid enzyme activity, available only from surgical procedures All composed primarily of exfoliated cells, most of which are degenerative; most require further development Available through Pap smears Available through Pap smears Requires invasive procedure Composed primarily of macrophages · Gametogenesis. End-organ response to sex steroids. Nonreproductive tract tissues, such as the liver and adrenal glands, can affect menstrual cyclicity by altering the pro- duction of sex-hormone-binding globulins (Vermeulen et al., 1969; Forest and Ber- trand, 1972; Anderson, 1974) or causing excessive synthesis of nongonadal andro- gens (Molinatti et al., 1964; Mahesh et al., 1968; Riddick and Hammond, 1975) Furthermore, the reproductive process is extremely sensitive to the general health of the woman, and metabolic stress (such as weight loss, hypothyroidism, excessive exercise, and glucocorticoid excess) could stop ovulation (Warren et al., 1975; Vigersky et al., 1977; Smith, 1980; Warren, 1980; Shangold et al., 1981). Severe psychologic stress can alter the reproductive process by suppressing gona- dotropin products and inducing amenorrhea (Fries et al., 1974; Rabkin and Struening, 1976; Sommer, 1978; Henry, 1980~. Healthy women occasionally experience anovulatory menstrual cycles that may be accompanied by altered menstrual cycle length and character but without long- term changes in fertility. Episodic toxic exposure might be analogous in effect to short-term use of oral contraceptives, which temporarily prevent ovulation but have no lasting anovulatory effect (Golditch, 1972; Jacobs et al., 1977; Tolis et al., 1979; Henzl, 1986~. Continuous exposure-e."., through chronic ingestion of contaminated food or water, occupation- al exposure, or long-term pharmacologic treatment-might have a profound influence on reproduction. Women undergoing anes- thesia is a case of short-term exposure (Soules et al., 1980~. Women working in operating rooms and chronically exposed to anesthesia have an increased number of miscarriages with increasing duration of exposure to anesthetic chemicals. Adverse effects might be overlooked if they are not life-threatening, rare, or specifically monitored in the popula- tion at risk. For example, the drug spiro- nolactone was used for many years to treat
182 hypertension, before it was noted that it had antiandrogenic activity and induced gynecomastia in males (Corvol et al., 1975; Boisselle and Tremblay, 1979; Shapiro and Evron, 1980; Cumming et al., 1982~. Infer- tility, however, is neither life-threat- ening nor rare and is likely to be undetect- ed in the general population, unless it is monitored. SPECIFIC MARKERS A sophisticated array of assays of bio- chemical events associated with ovulatory menstrual cycles is available, including assays of gonadal sex steroids (E2, es- trone, progesterone, 1 7-OH-progesterone, testosterone, and androstenedione), pitu- itary hormones (FSH, LH, and prolactin), FEAL4LE REPRODUCTIVE TOXICOLOaY 14-3 and 14-4~. Random measurement of these hormones to evaluate menstrual cycle normalcy in unselected women is not cost- effective; a better strategy would in- volve hormone measurements standardized for time in the cycle in an at-risk popula- tion with high exposures or symptoms, e.g., irregular menses. With this strategy, clinically inapparent hormonal changes in gonadal or pituitary hormones might be detected with cycle-specific hormone measurements. Alterations of the normal cycle are non- specll~lc responses and can occur against a background of spontaneous disease with an indistinguishable pattern of menstrual dysfunction. Therefore, alterations do not necessarily indicate that exposure to a toxicant has occurred. Once an abnor- hypothalamic neurotransmitters (GnRH and dopamine), and gonadal regulatory peptides (inhibin and activin) (Tables Marty of the reproductive cycle is detect- ed, characterization of a specific altera- tion in the hormonal profile is usually TABLE 1~3 Potential Biochemical Markers of Reproductive Toxicity for Evaluation In Viva Origin Biologic Markers Comments Central nervous system Pituitary Ovary Fallopian tube Uterus Cervix Vagina Neurotransmitters -Luteinizing hormone, follicle- stimu tat ing hormone, prolactin, adrenocorticotropic hormone, thyroid- stimulating hormone, growth hormone Steroids: estradiol, estrone, progesterone, testosterone, andro- stenedione Regulatory factors: relaxin, progestin- associated endometrial protein (PEP), prolac- tin, plasminogen activa- tor, inhibin, oocyte maturation inhibitor, luteinization inhibitor Secretory proteins Prolactin, other secretory proteins Prostaglandins Mucus Secretory proteins Measurement of CNS concentrations might not reflect circulating concentrations owing to blood-brain and brain-CSF barriers Pulsatile secretion makes adequate sampling difficult Cycle-specif~c Poorly characterized, cycle-specific, unclear physiologic significance Poorly characterized, specific, and the physim logic significance is unclear Poorly characterized, cycle-specific, of unclear physiologic significance Nonspecific, rapidly metabolized paracrine and autocrine hormones Cycle-speciB~c Poorly characterized, cycle-specific, easily con- taminated with bacteria
NONCONCEPTIVE MENSTRUAL CYCLES TABLE 14 4 Potential Biologic Markers of Reproductive To=aty for Evaluation In Vitro 183 Tissue Biologic Marker Comments Adipose Uterus Cervrx Endometr~um Myometrium Fallopian tube Ovary Granulosa Thecal cells Steroidogenesis Rate of mucus production Concentration of secretory proteins Steroid hormone response Concentrations of secretory proteins Lipid-soluble towns might accumulate, as in ovary Estrogen-dependent Cycl~specif~c, unclear physiologic significance Little relationship to menstrual cycle Cycl~specif~c, of unclear physiologic significance Rate of steroid- Excellent culture method ogenesis; gonadm tropin-receptor number or responsive ness; synthesis of regulatory factors Rate of steroid- ogenesis; gonadotrm pin receptor number or responsiveness Difficult to purifier possible. The underlying course, however, often is not well understood, and unrecog- nized reproductive toxicity is likely. Alterations of the normal menstrual cycle can occur either through disruption of follicular development or through onset of luteolysis. The exact mechanisms are unknown, but in most naturally occurring instances, altered gonadotropin or pro- gesterone secretion or action is suspect- ed. Reproductive toxicants might also alter the secretion or action of these hormones. Cyclicity An initial approach to exploring envi- ronmental toxicity is to evaluate whether menstrual cycles are regular. However, regardless of which organ is affected or the mechanism of toxicity, a substantial endocrinologic disruption of the menstru- al cycle must occur, if irregular menses is to be clinically detectable. For ex- ample, quantitative changes in the func- tion of the corpus luteum can occur without alterations in overall cycle length The hallmark of primate ovulation is (Murthy et al., 1970; Abraham et al., 1974; a spontaneous luteal phase whose dominant Jones et al., 1974; Radwanska et al., feature is the production of progesterone. 1976; Soules et al., 1977; Rosenfeld et That steroid has various biophysical ef- al., 1980; Radwanska et al., 1981; Gravanis et al., 1984~. Character of Menstrual Flow In a normal genital tract, menses is the inevitable result of a nonconceptive ovulatory cycle. Not all vaginal bleeding, however, is indicative of ovulation; anov- ulatory bleeding can occur often enough to be clinically indistinguishable from ovulatory bleeding. Typically, the flow lasts 3-7 days and is associated with some degree of dysmenorrhea. Few clots are passed with menstrual blood in an ovulatory cycle. Substantial changes in length from cycle to cycle or the presence of clots may indicate effects of external factors. When it is clinically important to distin- guish ovulatory from nonovulatorY bleed- . 1ng, a more accurate indicator than the character of the vaginal bleeding is re- quired. Detection of Corpus Luteum
184 facts; hence, luteal function is detec- table through a variety of clinical and laboratory measures, as noted briefly below. Progesterone in the Peripheral Serum Progesterone, a readily diffusible steroid, can be measured in various body fluids (Yoshimi et al., 1969; Mikhail, 1970; Yussman and Taymor, 1970; Lloyd et al., 1971; Rondell, 1974; Sherman and Kor- enman, 1975; Maathuis et al., 1978; Aedo et al., 1980; Donnez et al., 1982; Zorn et al., 1982; Crain and Luciano, 1983; Loumaye et al., 1985; Bouckaert et al., 1986; Kon- inckx et al., 1986) and reproductive tract fluids (Fowler et al., 1978; Botero-Ruiz et al., 1984), as can its glucuronidated metabolite in urine (Tietz et al., 1 97 1; Speroff et al., 1983; Rebar, 1986; Shack- leton, 1986~. Urine samples are easy to collect and simple to store and do not ex- hibit the rapid fluctuations in oroges- terone concentration observed in blood. Detailed comparisons of blood. salivas and urine must be performed for each assay to validate the relationships for that method. Each assay system might have spe- cific storage requirements, such as serum separation, freezing, and use of antibac- terial preservatives. Only one midluteal progesterone value is necessary to document that ovulation has occurred (Israel et al., 1972~. How- ever, evaluation of an inadequate luteal phase is controversial. No method is uni- versally accepted but several serum meas- urements of progesterone In the luteal phase would characterize the function of the corpus luteum. Basal Body-Temperature Shift Progesterone is a thermogenic hormone; when circulating concentrations rise above about 2-3 ng/ml, the basal body tem- perature rises by approximately 0.5°C, until the demise of the corpus luteum just before menses. Basal body temperature can be reliably measured orally or rectal- ly. New electronic thermometers might improve accuracy and, if coupled with a FEMALE REPRODUCTIVE TOXICOLOGY recorder, might prove applicable to large- scale monitoring. Cervical Mucus Changes E2, the dominant follicular-phase ster- oid, differs markedly from progesterone, the dominant luteal-phase steroid, in effect on cervical mucus quality and quan- tity. Changes in the character and quan- tity of cervical mucus can be useful clini- callv to predict hormonal status (Cliff, 1945; Pommerrenke, 1946; Birnberg, 1958; Moghissi, 1966; MacDonald, 1969; Moghissi et al., 1972; Moghissi, 1973~. Cervical mucus evaluation is qualitative and re- quires a pelvic examination. Vaginal Cytology The effects of E2 and progesterone on vaginal cytologic findings also differ (Papanicolaou, 1933; Riley et al., 1955; Rakoff, 1961; Frost, 1974~. Vaginal cytol- ogy is useful in monitoring the cyclicity of female rodents, but has less reliability in women because the changes from one day to the next are less evident. Self-col- lected vaginal specimens or measurements compatible with tampon use are feasible, but bacterial or fungal contamination mightcompromise individual measurements. Endometrial Histology Characteristic histologic changes ac- company follicular phase development in response to rising E2 concentrations. With the onset of a luteal phase, increasing progesterone concentration sequentially changes the histologic appearance in a well-defined progression to the menstrual endometrium (Noyes et al., 1950~. Histol- ogy is a reliable means of detecting ovula- tion and is semiquantitatively correlated progesterone production. Endometrial biopsy is extremely accurate for assessing the presence and competence of a corpus luteum, but has the serious disadvantage of being an invasive, painful, and expen- sive procedure that carries some risk of , _ genital tract infection.
NONCONCEPTIVE MENSTRUAL CYCLES Secretion of Progesterone-Stimulated Endometrial Proteins into Uterine Lumen The only two well-characterized proges- tin-dependent secretory proteins are pro- lactin (Maslar and Riddick, 1979; Daly et al., 1 983a) and progestin-associated endometrial protein (PEP) (Mazurkiewicz et al., 1981; Joshi, 1983~. Neither has been used clinically to evaluate the lu- teal phase, but efforts are under way to assess their usefulness. Cyclic changes of specific uterine luminal proteins have not been characterized, except for those of prolactin (Mater and Kuslis, 1987~. Aspiration of uterine luminal fluid is comparable in utility with endometrial biopsy and is an uncomfortable, invasive procedure with some hazard of infection; application as a marker must await a better understanding of uterine physiology. Reproductive Endocrinology In Vivo Measurements of other circulating re- productive hormones and regulatory fac- tors, such as GnRH, the gonadotropins, testosterone, and inhibin, have been used experimentally and clinically to study the physiology or pathophysiology of the menstrual cycle. An increasing catalog of paracrine hormones and growth factors is providing a new class of regulatory substances to consider. Measurements of reproductive hormones or regulatory factors, such as GnRH and gonadotropins, are accurate in detecting anovulation, but are of value mainly in identifying the site of the problem. Meas- urements of other ovarian hormones (e.g., testosterone and androstenedione) are of little value because their production is a consequence, not a cause, of anovula- tion. In Vitro Markers A variety of important biochemical events in the menstrual cycle can be evalu- ated with in vitro systems and used to de- termine the effects of putative toxicants (Table 14-4~. Any adverse influence of environmental agents needs to be confirmed with in viva testing, and in vitro systems 185 can be used as models to screen agents for reproductive toxicity, as well as to pro- vide details of the mechanisms. Examples of such models are: · Gonadotropin secretion by dispersed pituitary cell cultures. · Steroidogenesis, regulatory factor synthesis, and gonadotropin binding by ovarian cell organ or cell cultures (i.e., granulosa cells, thecal cells, and stromal cells). · Secretion of luteal-specific proteins by endometrial cell cultures. · Electrophysiology of myometrial cells in organ culture. · Mucus production by endocervical cell cultures. Follicular fluid, granulosa cells, oocytes, and other reproductive tract materials are readily available for rou- tine study from surgical specimens. Evaluating cellular physiology in vitro permits the effects of environmental agents on biochemical events of the normal menstrual cycle to be investigated. That approach has been applied to investigation of diseases that cause anovulatory infer- tility, such as polycystic ovary syndrome (Haney et al., 1986~. The in vitro approach should prove rewarding in estimating the potential for adverse biologic effects of environmental chemicals, and efforts in this area should be encouraged. BIOLOGIC RATIONALE Clinical validation of any biologic marker is critical to its application in detecting toxicity. Spontaneous anovu- lation in infertility patients can serve as an example of factor that is important in the consideration of environmentally related reproductive toxicity. Of the factors that are associated with infertilitysemen, cervical mucus, the endometrial cavity, oviductal function, ovulatory function, and endometriosis ovulation seems the most likely to be af- fected by environmental exposure. Defec- tive ovulation usually is manifested by irregular vaginal bleeding. Other signs and symptoms, such as galactorrhea, the
186 absence of premenstrual molimina, and a change in the character of menses, might be helpful, but their significance is di- minished without a disruption of menstrual cyclicity. The biologic markers previous- ly discussed are used clinically in infer- tility patients to detect ovulation when the vaginal bleeding pattern is not regular enough to permit accurate prediction of ovulation in nonconceptive menstrual cycles. Environmental agents whose pharmacolog- ic action is expected to be similar to that of contraceptives should be regarded as potential reproductive toxicants. Theor- etical validation of the markers of repro- ductive toxicity noted above also can be accomplished by considering therapeutic agents used for contraception. For in- stance, ovulation can be blocked by ster- oid-feedback inhibition of gonadotropin release. In addition, the end-organ re- sponses of the cervix, uterus, and oviducts to the natural ovarian steroids (Mishell, 1979) are altered by the three major classes of sex steroids-androgens, estrogens, and progestins. The develop- ment of birth control pills containing estrogens and progestins is based on con- traceptive efficacy and control of vaginal bleeding. Progestins alone can FEM4LE REPRODUCTIVE TOXICOLOGY alter gonadotropin release and endometri- al development; without estrogen, the endometrium is fragile, and vaginal bleed- ing is not well controlled. Estrogens alone are effective in preventing concep- tion and have been used clinically after sexual assaults (Kuchera, 1974~. A chemi- cally modified androgen, danazol, also has been used to suppress gonadotropins therapeutically and create an anovulatory state (Young and Blackmore, 1977; Barbieri et al., 1977; Guillebaud et al., 1977; Rannevik, 1979; Luciano et al., 1981), but its use is limited by the high frequency of undesirable androgenic side effects. The mechanisms of other hormones can also be demonstrated pharmacologically. Progesterone antagonists do not prevent ovulation, but alter decidualization, the endometrial response to progesterone production by the corpus luteum, enough to prevent implantation (Healy et al., 1983; Kovacs et al., 1984; Paris et al., 1984; Paris et al., 1986; Koering et al., 1986~. Inhibitors of progesterone syn- thesis induce biochemical luteolysis, which leads to failure of endometrial de- velopment in a fashion similar to that caused by progesterone antagonists (Bir- gersson and Johansson, 1983; van der Spuy et al., 1985; Webster et al., 1985~.