| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Centre de Recherche en Reproduction Animale Université de Montréal St. Hyacinthe, Québec, Canada J2S 7C6
Address all correspondence and requests for reprints to: Bruce D. Murphy, Universite de Montreal, Centre de Recherche en Reproduction Animale, 3200 rue Sicotte, Saint Hyacinthe, Québec, Canada J2S 7C6. E-mail: bruce.d.murphy{at}umontreal.ca.
The evolution of mechanisms of implantation, embryo invasion, and placental formation was essential to the success of viviparity, the defining characteristic of Eutherian mammals. Numerous evolutionary variations on this theme resulted in different forms of placentation, with intimacy of the embryo-endometrial interaction ranging from a simple but firm attachment (epitheliochorial placentation) to erosion of the endometrial epithelium and maternal endothelium by the invading embryo (hemochorial placentation). It is not surprising that development of these specializations required cross talk between the embryonic passenger and the endometrium, the tissue that hosts it for the gestational journey. The developing embryo may demonstrate its presence even before arrival in the uterus because there is evidence that it can affect both the expression of oviductal genes and its own transport (1). The best-known signal from embryo to uterus is found in ruminants in which an immunoregulatory protein, interferon-
, has been conscripted for embryo signaling. In most other species and contexts, neither the message nor the response is well defined. Furthermore, the focus in studies of embryo implantation has been largely on the endometrial side of the conversation, i.e. how the uterine environment dictates the attachment and invasion of the trophoblast (2). The article in this issue by Kashiwagi et al. (3) is based on investigation of the other side of the dialogue, i.e. how the embryo directs the development of the placental specializations of the endometrium. They approached the problem by high-density microarray analysis, followed by confirmation of abundance of RNA and protein of selected genes.
In humans during the menstrual cycle and in rodents before implantation, the endometrial stroma undergoes a specialized terminal differentiation, resulting in formation of the so-called decidual cells. In rodent gestation, this response is provoked by the presence of the embryo (4). Decidualization has been ascribed to many functions in pregnancy; a short list would include the early nutrition of the embryo, the secretion of hormones and growth factors essential to embryogenesis, and the limitation of placental invasion of the endometrium (5). Differentiation of the stromal compartment appears essential to successful gestation because null mutation of genes that prevent decidualization in mice, such as the progesterone receptor gene, precludes implantation (6). Given its essential role in rodents, it is surprising that decidualization is very limited, or even nonexistent, in carnivores and some other mammals that display invasive placentation.
In rodents, the decidual response does not require a specific embryonic signal; rather it can be induced by mechanical stimulation of the uterus sensitized by an appropriate hormonal milieu. The resulting response has been given the name deciduoma (Fig. 1
). Kashiwagi et al. (3) exploited this model to compare the artificially induced deciduoma with the deciduum that is invoked by the attachment of the embryo. The mRNA of an enormous number, approximately 10% of the total genes expressed, or nearly 3000 genes, was found to be differentially abundant between the two conditions. From further analytical refinement, it was concluded that there was up-regulation of 753 and down-regulation of 686 genes, still a formidable number to investigate. Two categories of genes were the most common in the differentially expressed pool, those coding for products with a protein binding function (45%) and those involved in catalysis (23.5%). The remainder were distributed across the domains of signaling, transcriptional regulation, transport, etc., as in other studies of uterine gene expression (7). From a pool of candidates with the greatest differential expression, Kashiwagi et al. (3) selected some up-regulated and some down-regulated genes for in vivo evaluation of mRNA and protein abundance and for comparison of the deciduum-deciduoma model.
|
). Furthermore, the deciduum has fewer dividing cells than does the deciduoma, leading to the supposition that the embryo somehow attenuates decidual growth. For this reason, the patterns of expression of nine growth- and remodeling-related genes, including bone morphogenic protein (BMP)-2 and BMP7 as well as some proteases that degrade other growth factors were investigated. This analysis revealed that BMP2 and BMP7 were indeed more robustly expressed in the deciduoma, relative to the embryo-induced deciduum. Furthermore, genes that degrade or inhibit growth factors, such as the metalloproteinases and IGF binding proteins, consistently displayed stronger expression in the embryo-induced deciduum. These expression data, then, support the case for embryonic inhibition of decidual growth and the corollary of embryonic acceleration of stromal cell decidualization.
It was previously shown by another experimental paradigm that the largest percentage of the genes up-regulated by the trophoblast in human decidual cells was associated with immune responses (8). The mechanisms of this stimulation of immune-related genes by the uterus remain to be discovered. The article by Kashigawi et al. (3) implicates interferons as the embryonic signal because interferon-stimulated genes are expressed in greater abundance in embryo-induced deciduum, relative to deciduoma. Whereas no embryo-specific interferon has yet been shown to be expressed in mice, the interferon responses hearken back to interferon-, the trophoblast signal in ruminants that announces the presence of the embryo. Furthermore, interferons are well known to activate natural killer (NK) lymphocytes (9), and there is abundant evidence that NK cell populations increase with decidualization in both humans and the pregnant mouse model (10). Kashiwagi et al. (3) provide new insight by their demonstration that NK cell numbers are severalfold greater in the embryo-induced deciduum, relative to the deciduoma. Furthermore, a number of genes associated with NK cell activation are preferentially up-regulated when the embryo provokes decidualization, relative to the artificial deciduoma model. These intriguing findings are expected to provide further pieces to the persistent puzzle of maternal-fetal immunology.
The vascularity of the hemochorial placenta differs from virtually all other tissues because the cells in contact with blood are from the trophoblast, rather than endothelium. Endothelium is the usual gatekeeper that prevents thrombosis in nonplacental vascular beds. Previous studies have shown that uterine thrombosis is inhibited during gestation by the trophoblast (11). One of the genes essential for the process, as indicated by lethality due to placental thrombosis when knocked out, is the endothelial protein C receptor (EPCR) (12). Microarray and consequent RNA/protein confirmation in the article by Kashiwagi et al. (3) demonstrate that EPCR is an embryo-induced gene. The clinical significance of these findings may be great, given that unexplained problems of human gestation such as preeclampsia and interuterine growth restriction may be caused and are certainly exacerbated by placental thrombosis (13).
Thus, a strong case has been made for endometrial gene regulation by the embryo by Kashiwagi et al. (3), and they have raised several interrelated questions for further research. First, what are the embryonic signals that dictate the development and differentiation and the regulation growth in decidual tissue? Second, to what extent is the embryonic communication a monologue rather than a dialogue, i.e. does the differentiating deciduum talk back to the embryo, thereby provoking further embryonic signaling to the endometrium? Third, to what extent are these findings pertinent to human gestation and conditions of placental insufficiency, preeclampsia, and interuterine growth retardation? We hope that this work will stimulate further investigation to address these issues.
 |
Footnotes |
|---|
Abbreviations: BMP, Bone morphogenic protein; EPCR, endothelial protein C receptor; NK, natural killer.
Received May 30, 2007.
Accepted for publication June 4, 2007.
|
References |
|---|
 Top References |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
