Identification of Plet-1 as a specific marker of early thymic epithelial progenitor cells.

The thymus is essential for a functional immune system, because the thymic stroma uniquely supports T lymphocyte development. We have previously identified the epithelial progenitor population from which the thymus arises and demonstrated its ability to generate an organized functional thymus upon transplantation. These thymic epithelial progenitor cells (TEPC) are defined by surface determinants recognized by the mAbs MTS20 and MTS24, which were also recently shown to identify keratinocyte progenitor cells in the skin. However, the biochemical nature of the MTS20 and MTS24 determinants has remained unknown. Here we show, via expression profiling of fetal mouse TEPC and their differentiated progeny and subsequent analyses, that both MTS20 and MTS24 specifically bind an orphan protein of unknown function, Placenta-expressed transcript (Plet)-1. In the postgastrulation embryo, Plet-1 expression is highly restricted to the developing pharyngeal endoderm and mesonephros until day 11.5 of embryogenesis, consistent with the MTS20 and MTS24 staining pattern; both MTS20 and MTS24 specifically bind cell lines transfected with Plet-1; and antibodies to Plet-1 recapitulate MTS20/24 staining. In adult tissues, we demonstrate expression in a number of sites, including mammary and prostate epithelia and in the pancreas, where Plet-1 is specifically expressed by the major duct epithelium, providing a specific cell surface marker for this putative reservoir of pancreatic progenitor/stem cells. Plet-1 will thus provide an invaluable tool for genetic analysis of the lineage relationships and molecular mechanisms operating in the development, homeostasis, and injury in several organ/tissue systems.

Differential expression of two estrogen receptor beta isoforms in the human fetal testis during the second trimester of pregnancy.

Testicular cancer is more common in individuals with disorders of the male reproductive tract. It has been suggested that inappropriate exposure to estrogens during fetal life may have an impact on maturation of testicular germ cells that are the cells of origin of the majority of testis cancers. The aim of the present study was to establish whether human fetal germ cells (gonocytes) are a potential target of estrogen action. To address this issue, we used RT-PCR and immunohistochemistry to examine the pattern of expression of estrogen receptors (ER alpha, ER beta, and ER beta 2 variant) in human fetal testes at 12-19 wk gestation. ER alpha, mRNA, and protein were not detected in any of the fetal testes. In contrast, using an antibody directed against the hinge domain of ER beta expression was detected in multiple testicular nuclei. RT-PCR with primers specific for full-length wild-type ER beta (ER beta 1) or the ER beta 2 variant formed by splicing of an alternative eighth exon, was performed on whole-tissue extracts and materials recovered by laser capture and revealed that mRNAs for both isoforms were expressed. Immunohistochemistry with isotype-specific monoclonal antibodies showed that ER beta 1 was low/undetectable in gonocytes, whereas these cells expressed the highest levels of ER beta 2, compared with other testicular cell types. Both ER beta 1 and ER beta 2 were detected in some but not all Sertoli cells, peritubular cells, and other interstitial cells including those tentatively identified as Leydig cells. Our immunohistochemical results demonstrate that during the second trimester, some but not all somatic cells within the human fetal testis express wild-type ER beta (ER beta 1) protein and/or the variant isoform of ER beta (ER beta 2) that lacks amino acids essential for binding of estradiol. ER beta 2 protein was readily detectable in fetal gonocytes, whereas ER beta 1 was not. We did not detect expression of ER alpha. The expression of ER beta 2, a variant proposed act as a dominant negative receptor, might prevent estrogen action in gonocytes. We suggest that during this period of fetal life, estrogenic ligands are most likely to act on somatic cells that contain ER beta 1 protein.

Immunohistochemical profiling of germ cells within the human fetal testis: identification of three subpopulations.

In the human fetal testis, germ cells that have migrated to the genital ridges become enclosed within testicular cords by 8 wk of gestation. Most papers refer to all types of germ cells as being "gonocytes" or "prespermatogonia," giving the impression that they are identical. Detailed morphological studies, however, have suggested a heterogeneous population. We have used single, double, and triple immunohistochemistry to evaluate the differentiation of cells within fetal testes recovered during the first (7-9 wk) and second (14-19 wk) trimesters. In the first trimester, differentiation of Sertoli cells preceded the formation of testicular cords and the differentiation of interstitial (Leydig, peritubular myoid) cells. Immunostaining for CHK2, C-KIT, placental alkaline phosphatase, PCTAIRE-1, and MAGE-A4 revealed that the proportion of germ cells expressing each of these proteins was correlated with gestational age. Expression of the pluripotency marker OCT4 was restricted to a population of small, round germ cells. Three types of germ cell were identified, and we propose that these should be known as gonocytes (OCT4pos/C-KITpos/MAGE-A4neg), intermediate germ cells (OCT4low/neg/C-KITneg/MAGE-A4neg), and prespermatogonia (OCT4neg/C-KITneg/MAGE-A4pos). In the first trimester, most germ cells had a gonocyte phenotype; however, from 18 wk of gestation, prespermatogonia were the most abundant cell type. These data provide evidence for the functional differentiation of human testicular germ cells during the second trimester of pregnancy, and they argue against these germ cells being considered as a homogeneous population, as in rodents.