The effects of exposure to endocrine disruptors during sex differentiation are of special concern for a number of reasons: this process is very sensitive to the effects of relatively low doses of endocrine disruptors, the effects are irreversible, i. e., the system is "imprinted" by the initial hormonal environment, functional alterations of the sex differentiation are often not apparent until after puberty or even later in life, and the abnormalities, which include malformations and infertility, cannot be predicted from the transient alterations in hormone levels produced by similar exposure in adult animals. It is important to understand the key role that animal models play in this research as developmental reproductive tox-icity data are often critical in the assessment of noncancer health effects of EDCs. Furthermore, when similar congenital reproductive abnormalities have been detected in laboratory species, they have dramatically facilitated our understanding of genetic errors of steroid metabolism and receptor function in humans. For this reason, rodent and other models have great utility for evaluating the potential of xenobiotics to alter human reproductive development. We do not have to wait until adverse effects appear in the human population, as we did with DES and other drugs, to take precautionary action.
The role of hormones in sex differentiation in mammals is well understood. Even the most severe alterations of this process are not lethal. The normal development of the sexual phenotype from an indifferent state entails a complex series of events . Genetic sex is determined at fertilization and this normally governs the expression of the "male factor" and the subsequent differentiation of gonadal sex. Prior to sex differentiation, the embryo has the potential to develop a male or female phenotype. Following gonadal sex differentiation, testicular secretions induce differentiation of the male duct system and external genitalia. The development of phenotypic sex includes persistence of either the Wolffian (male) or Mullerian (female) duct system, and differentiation of the external genitalia and the central nervous system (CNS). Other organ systems, like the liver, muscles, and brain, are "imprinted" as well. The male phenotype arises due to the action of testicular secretions, testosterone, and Mullerian inhibiting substance. Testosterone induces the differentiation of the Wolffian duct system into the epididymis, vas deferens, and seminal vesicles, while its metabolite DHT induces the development of the prostate and male external genitalia. It has been suggested that in the absence of these secretions, the female phenotype is expressed (whether or not an ovary is present). This hypothesis is supported by the observation that when estrogen receptor a and b genes are both knocked out (a/?ERKO mouse) the female mouse still develops a complete reproductive tract . However, there is some indication of intersex in the ovaries as Sertoli-like cells, normally found in the testis of the male, are found within the ovary. In the CNS, testosterone is aromatized (via the steroidogenic enzyme aromatase) to estradiol and reduced via 5a-reductase to DHT in a species-specific manner. It has been suggested for some species that all three hormones (testosterone, DHT, and estradiol) play a role in the masculinization of the CNS.
Reports of EDC-induced alterations of sex ratios and possible production of intersex in wildlife presents a new scientific challenge. A decline in the number of males born has been reported in the zone of Seveso contaminated with highest levels of TCDD , and the general decline in the percentage of males born in industrialized nations world-wide also has been linked to the EDC issue [78-82]. However, Jacobsen et al.  and Biggar et al.  suggested that this trend was more likely related to family size, not EDCs, while Vartiainen et al.  reported that the changes in male proportions did not correlate with industrialization or the use of pesticides or hormonal drugs, rendering a causal association unlikely. In addition, such a trend was not observed in Ireland, which industrialized at about the same time at the rest of Europe .
Data from animal studies suggest that alterations of the hormonal environment can result in changes in sex ratios in the offspring [87, 88]. Vandenbergh and Huggett  reported that in mice the mother's intrauterine position during fetal development affects the sex ratio of her offspring. Females adjacent to two males, presumably exposed to higher androgen levels, produced first litters that were 58% males vs females that were adjacent to two females in utero, which had fewer (42 %) males in the first litter. Likewise, the intrauterine position of gerbils appears to alter the sex ratio in a similar fashion . Asynchronous mating, in which the time of mating varies from the time of ovulation, has been shown to alter sex ratios in several mammalian species [90, 91]. In the field, mammalian populations can show significant changes in sex ratio that are related to environmental factors. Kruuk et al.  reported that dominant female red deer (Cervus elaphus) produce more males at low population levels, an effect that disappeared as the population increased to carrying capacity, implying a social influence on sex determination which could be mediated via endocrine mechanisms. In a human study, Sas and Szollosi  reported that the sex ratio of children born to fathers treated with methyltestosterone was 67 % male. While such observations appear in conflict with the concept of Mendelian genetics, mechanisms that could explain altered sex ratios have been identified. For example, hormones or other factors that regulate genes present on the Y chromo some that control sperm motility by modulating the activity of sperm motility kinases could alter the probability of Y sperm fertilizing the egg .
In most cases, mammalian hermaphrodites are pseudohermaphrodites (PHs) rather than true hermaphrodites (THs) [95 - 97]. In PHs, varying degrees of intersex of the hormone-dependent tissues is achieved but gonadal differentiation is consistent with the genetic sex of the individual. The degree of natural PHs among species can vary greatly with some species normally having masculinized females like the spotted hyaena (Crocuta crocuta)  or polled goats . Genetic mapping of the autosomal region in XX sex-reversal and horn development revealed that in goats, abnormalities in sex determination are intimately linked to a dominant Mendelian gene coding for the "polled" (hornless) character . This species provides an interesting animal model for cases of SRY-negative XX males. In the absence of information about what is "normal" it is difficult to interpret whether or not a female PH rate of 1.5% for the polar bear is abnormal .
In THs, the "intersex" individual has both ovarian and testicular tissue [96, 97]. Unilateral THs have an ovary on one side and a testis on the other or an ovotestis on one side and an ovary or a testis on the other. Bilateral THs have ovarian and testicular tissue on both sides, usually in the form of an ovotestis. The nature of the gonad influences the differentiation of the ipsilateral reproductive tract and only a very small percentage of TH humans are fertile. In addition, only a handful (reported as four)  of TH individuals, including all mammalian species, have been shown to have bilateral TH with separate ovary and testis on both sides and a complete male and female reproductive tract.
In humans, TH is generally related to one of several different errors of genetic sex determination: There are a number of genetic errors involving sex determining mechanisms which result in either TH or PH including complete and incomplete sex reversals (XX males and XY females); sex chromosome anomalies , single gene defects coding for a defective steroidogenic enzyme, which leads to reduced synthesis of sex steroids (20,22-desmolase; 17-ketosteroid reductase; and 5a-reductase deficiency), defective steroid receptor, resulting in abnormal handling of androgens in the target tissues (complete androgen insensitivity syndrome, Reifensten syndrome), and various other genetic defects (LH deficiency and lack of responsiveness to human chorionic gonadotropin) .
In wildlife there are some rather unusual examples of altered sex ratios in lemmings. Lyon  reviewed the sex ratios of different types of "female" wood lemmings (Myopus schisticolor),in which the percentage of females born ranges from 50% in one type (XX) to 100% in another type of female (XY). In the latter case, XY females develop carrying mutation on their single deviant X chromosome, which prevents development of the male phenotype. These XY females are fertile but they produce only daughters (50% XX and 50% XY). A similar condition exists in the varying lemming (Dicrostonyx torquatus).
TH is also common in some mammalian species. TH has been identified in four species of European moles (insectivora, mammalia). In Talpa occidentalis, T. europaea, T. romana,and T. stankovici all XX individuals are intersex, whereas XY individuals have a normal male phenotype. Intersex XX females had bilateral ovotestes with a small portion of histologically normal ovarian tissue and a vari ably sized large mass of dysgenetic testicular tissue, accompanied by a small epididymis. SRY gene was found to be present in males but not females .
TH has been identified in a St Lawrence beluga whale (Delphinapterus leu-cas) by De Guise et al. . This animal had two testes, two separate ovaries, and the complete ducts of each sex, but the cervix, vagina, and vulva were absent. Unilateral TH has been noted in an FVB/N mouse  with male and female gonads and reproductive tissues on opposite sides.
Our interest in the ability of EDCs to induce TH was stimulated by reports of possible TH appearing simultaneously in 1998 in 29/87 small mammals at two sites in California, including a former wildlife refuge, which was closed due to extensive selenium contamination. In the process of collecting animals for assessment of contaminant levels by overnight snap-trapping, it was noted that four species (house mouse - Mus musculus, deer mouse - Peromyscus manicu-latus, California vole - Microtus californicus, and the western harvest mouse -Reithrodontomys megalotis) appeared to be bilateral THs, displaying tissues which were tentatively identified as paired separate ovaries and scrotal testes and uterine tissue. Externally, the animals were phenotypic males. A collection of animals several years prior to the present sampling in 1995 revealed a lower incidence (3 of 105) of possible intersex small mammals . In this regard, we examined reproductive tissues from six animals (four Mus musculus and two Reithrodontomys megalotis) from one site (not Kesterson) identified as possible intersex based upon the tentative identification of the presence of ovarian, tes-ticular, and uterine tissues. The entire reproductive tract plus gonads were preserved as a unit in fixative in the laboratory after trapping . These tissues were trimmed, embedded in paraffin, sectioned, stained with hematoxilin and eosin, and evaluated by a veterinary pathologist (Pathco, Inc, Research Triangle Park) for the microscopic appearance of the tissues and tissue identification. In general, the tissues submitted were in various stages of advanced autolysis, which accounts for the difficulty in identification of organs during gross necropsy. However, the microscopic appearance of the tissues was adequate for identification purposes. Each animal was confirmed as a male, and no female tissues were identified from any of the animals. Male tissues confirmed were testis, epididymis, seminal vesicles, and other sex accessory tissues. In some cases the testes were mature and sperm was present in the epididymis. Serial sections through the enlarged cephalic portions of the seminal vesicles (initially identified as possible ovaries) confirmed only seminal vesicle tissue.
The fact that TH was not identified in these animals, which had been initially identified as possible "intersex", is consistent with our initial expectations. Not only is the occurrence of TH animals with paired, separate ovaries and testis, along with both a male and female duct system extremely rare, but also the simultaneous appearance of this form of TH in four species of rodents in less than a decade would seem to be of very low probability. However, as indicated above, marked deviations from the standard mammalian plan for sex determination have been noted in rodents (voles and moles). For this reason, it was critical that these gross observations be confirmed or refuted with a histological examination. Initially, the preliminary diagnosis of TH was linked by the press to contaminants at the former wildlife refuge, and had such a profound effect been confirmed, there was concern for the induction of similar alterations in other species, including humans.
Was this article helpful?