Analysis of human invasive cytotrophoblasts demonstrates mosaic aneuploidy.

A total of 24 chromosome-specific fluorescence in situ hybridization probes for interphase nucleus analysis were developed to determine the chromosomal content of individual human invasive cytotrophoblasts derived from in vitro cultured assays. At least 75% of invasive cytotrophoblasts were hyperdiploid and the total number of chromosomes ranged from 47 to 61. The results also demonstrated that these hyperdiploid invasive cytotrophoblasts showed significant heterogeneity. The most copy number gains were observed for chromosomes 13, 14, 15, 19, 21, and 22 with average copy number greater than 2.3. A parallel study using primary invasive cytotrophoblasts also showed a similar trend of copy number changes. Conclusively, 24-chromosome analysis of human non-proliferating cytotrophoblasts (interphase nuclei) was achieved. Hyperdiploidy and chromosomal heterogeneity without endoduplication in invasive cytotrophoblasts may suggest a selective advantage for invasion and short lifespan during normal placental development.

Somatic genomic variations in extra-embryonic tissues.

In the mature chorion, one of the membranes that exist during pregnancy between the developing fetus and mother, human placental cells form highly specialized tissues composed of mesenchyme and floating or anchoring villi. Using fluorescence in situ hybridization, we found that human invasive cytotrophoblasts isolated from anchoring villi or the uterine wall had gained individual chromosomes; however, chromosome losses were detected infrequently. With chromosomes gained in what appeared to be a chromosome-specific manner, more than half of the invasive cytotrophoblasts in normal pregnancies were found to be hyperdiploid. Interestingly, the rates of hyperdiploid cells depended not only on gestational age, but were strongly associated with the extraembryonic compartment at the fetal-maternal interface from which they were isolated. Since hyperdiploid cells showed drastically reduced DNA replication as measured by bromodeoxyuridine incorporation, we conclude that aneuploidy is a part of the normal process of placentation potentially limiting the proliferative capabilities of invasive cytotrophoblasts. Thus, under the special circumstances of human reproduction, somatic genomic variations may exert a beneficial, anti-neoplastic effect on the organism.

The human placenta is a hematopoietic organ during the embryonic and fetal periods of development.

We studied the potential role of the human placenta as a hematopoietic organ during embryonic and fetal development. Placental samples contained two cell populations-CD34(++)CD45(low) and CD34(+)CD45(low)-that were found in chorionic villi and in the chorioamniotic membrane. CD34(++)CD45(low) cells express many cell surface antigens found on multipotent primitive hematopoietic progenitors and hematopoietic stem cells. CD34(++)CD45(low) cells contained colony-forming units culture (CFU-C) with myeloid and erythroid potential in clonogenic in vitro assays, and they generated CD56(+) natural killer cells and CD19(+)CD20(+)sIgM(+) B cells in polyclonal liquid cultures. CD34(+)CD45(low) cells mostly comprised erythroid- and myeloid-committed progenitors, while CD34(-) cells lacked CFU-C. The placenta-derived precursors were fetal in origin, as demonstrated by FISH using repeat-sequence chromosome-specific probes for X and Y. The number of CD34(++)CD45(low) cells increased with gestational age, but their density (cells per gram of tissue) peaked at 5-8 wk, decreasing more than sevenfold at the onset of the fetal phase (9 wk of gestation). In addition to multipotent progenitors, the placenta contained myeloid- and erythroid-committed progenitors indicative of active in situ hematopoiesis. These data suggest that the human placenta is an important hematopoietic organ, raising the possibility of banking placental hematopoietic stem cells along with cord blood for transplantation.

Hypermethylation of RASSF1A in human and rhesus placentas.

The pseudomalignant nature of the placenta prompted us to search for tumor suppressor gene hypermethylation, a phenomenon widely reported in cancer, in the human placenta. Nine tumor suppressor genes were studied. Hypermethylation of the Ras association domain family 1 A (RASSF1A) gene was found in human placentas from all three trimesters of pregnancy but was absent in other fetal tissues. Hypermethylation of Rassf1 was similarly observed in placentas from the rhesus monkey but not the mouse. An inverse relationship between RASSF1A promoter methylation and gene expression was demonstrated by bisulfite sequencing of microdissected placental cells and immunohistochemical staining of placental tissue sections using an anti-RASSF1A antibody. Treatment of choriocarcinoma cell lines, JAR and JEG3, by 5-aza-2'-deoxycytidine and trichostatin A led to reduction in RASSF1A methylation but increased expression. These observations extend the analogy between the primate placenta and malignant tumors to the epigenetic level.