Mice homozygous for the recessive mutation vestigial tail (v-t), which arose spontaneously on Chromosome 11, exhibit vertebral abnormalities, including loss of caudal vertebrae leading to shortening of the tail. Wnt-3a, a member of the wingless family of secreted glycoproteins, maps to the same chromosome. Embryos homozygous for a null mutation in Wnt-3a (Wnt-3a "e°) have a complete absence of tail bud development and are truncated rostral to the hindlimbs. Several lines of evidence reveal that vt is a hypomorphic allele of Wnt-3a. We show that Wnt-3a and vt eosegregate in a high-resolution backcross and fail to complement, suggesting that Wnt-3a "~° and vt are allelic. Embryos heterozygous for both alleles have a phenotype intermediate between that of Wnt-3a n~° and vt homozygotes, lacking a tail, but developing thoracic and a variable number of lumbar vertebrae. Although no gross alteration in the Wnt-3a gene was detected in vt mice and the Wnt-3a coding region was normal, Wnt-3a expression was markedly reduced in vt/rt embryos consistent with a regulatory mutation in Wnt-3a. Furthermore, the analysis of allelic combinations indicates that Wnt-3a is required throughout the period of tail bud development for caudal somitogenesis. Interestingly, increasing levels of Wnt-3a activity appear to be necessary for the formation of more posterior derivatives of the paraxial mesoderm.

It is well known that fetal androgens are required for male sexual differentiation, and it is thought that fetal ovaries are not steroidogenically active. However, molecular details, such as which steroidogenic enzymes are present in fetal testes and which enzymes are absent in fetal ovaries, have not been established. The pattern of expression of the genes that encode four of the steroidogenic enzymes necessary for androgen and estrogen production was examined during fetal development in mouse gonads. Messenger RNA (mRNA) expression for cholesterol side-chain cleavage (P450scc), 3 beta-hydroxysteroid dehydrogenase/delta 5-delta 4-isomerase (3 beta HSD), P450 17 alpha-hydroxylase/C17-20 lyase (P450c17), and P450 aromatase (P450arom) was determined before ovaries and testes were distinguishable (13 days postconception) and during sexual differentiation (15, 17, and 20 days postconception) using reverse transcriptase-polymerase chain reactions (RT-PCR). A PCR assay for Sry was used to determine gender on day 13. P450scc, 3 beta HSD, and P450c17 transcripts were detected at all ages in fetal testes, indicating that mRNAs for the steroidogenic enzymes that are required to convert cholesterol to androgens are present in the male gonad even before sexual differentiation. P450arom mRNA was detected in several fetal testes on day 17, but consistently observed on day 20. The expression of P450arom suggests the potential of fetal and neonatal testes to convert androgens to estrogens. In contrast, although 3 beta HSD mRNA was detected in several of the ovaries examined, the detection of P450scc, P450c17, and P450arom transcripts was rare. These data suggest that the absence of fetal ovarian steroid hormone production is the result of lack of expression of at least three of the steroidogenic enzymes, P450scc, P450c17, and P450arom.

An immunocytochemical assay for estrogen receptor (ER) was used to study the distribution of receptor in fetal and immature female mouse reproductive tracts. Immunoblots confirmed that a single band, the size of the ER, immunostained in extracts from day 15 and 17 fetal reproductive tracts. Staining was observed over nuclei of epithelial cells of the Mullerian duct and over nuclei of cells of the developing connective tissue (mesenchymal cells) of the reproductive tract on fetal day 15. By day 17 when a primitive uterus could be distinguished, ERs were detected in nuclei of mesenchymal cells, but in only a small portion of epithelial cells. A different pattern of immunocytochemical staining was observed in uteri from animals killed on the day of birth; cells of the connective tissue contained ER, but the epithelial cells did not. By 4 or 6 days after birth, more nuclei in the connective tissue stained for ER with a greater intensity compared to nuclear staining in epithelial cells. ERs were detectable in nuclei of both uterine epithelial cells and connective tissue cells on days 10 and 19 after birth.

To date, there is no conclusive evidence that ERs are present in preimplantation embryos. There are reports that estrogen is made by the rabbit blastocyst (61), and estrogens have been used to induce implantation in mice (62), but whether estrogens act through ERs in the embryo or in the maternal uterus is not known. ERs may be present in early embryos, but if so, levels are below the methods of detection used thus far. Perhaps with more sensitive immunodetection methods, it may be possible to detect ERs in embryos if they are present. Using PCR, messenger RNA for ER has been detected as early as the oocyte stage in mouse embryos (Q. Hou and J. Gorski, unpublished results). This was confirmed recently by Wu et al. (83a). Figure 7 shows a model for the pattern of ER expression in the developing mouse fetus based on the various reports discussed in this review. ERs are present in the 10-day mouse fetus, possibly in the developing ambisexual reproductive tract. Analysis of seven individual 10-day-old fetuses taken from the same litter showed similar levels of an immunostained protein the size of the ER in each fetus (57). The pattern of expression of ER between implantation and the development of the reproductive tract may be the same in male and female mice. Estrogen, acting through ERs, may be one factor (of many) that determines which cells are destined to be part of the indifferent reproductive tract. We were not able to isolate fetal mouse reproductive tracts at an indifferent stage (day 10) due to their very small size. One way to study ER in the indifferent reproductive tract would be to examine these tissues in a larger animal, such as the bovine, using similar immunodetection methods. The distribution of ER in the fetal mouse reproductive tract on fetal days 13 (before sexual differentiation) and 15 (initiation of sexual differentiation) is similar in males and females (71, 72). Thus, estrogen does not appear to be responsible for the initiation of sexual differentiation. Early experiments by Jost (41) showed that removal of the gonad from male or female rabbit fetuses resulted in the female phenotype, which lent weight to the hypothesis that ovarian hormones are not critical in the development of the female phenotype, whereas testicular hormones are essential for the development of the male phenotype.(ABSTRACT TRUNCATED AT 400 WORDS)

Immunodetection methods were used to detect estrogen receptors (ER) in male reproductive tracts on fetal days 13, 15, and 17 and on the day of birth. Immunocytochemistry revealed that most of the cells of the gonad and associated Wolffian duct stained for ER on fetal day 13. During the next 6 days, ER distribution changed, and by the day of birth, ERs were observed only in epithelial cells of the epididymis (derived from the Wolffian duct) and in a portion of cells from the testis. Immunoblots confirmed that a band the size of the ER stained in reproductive tracts for all ages studied. Similar to the fetal female, ERs are present throughout the early development of the fetal male reproductive tract. However, in contrast to the female, ERs appear to decrease in the male fetal reproductive tract at the time of birth.