Nature Genetics
25, 248 - 250 (2000)
doi:10.1038/76985
The labyrinthine placentaJulie Rinkenberger
& Zena WerbDepartment of Anatomy, University of California, San Francisco, California 94143, USA.
zena@itsa.ucsf.edu Fetal nutrition depends on the placenta, and specifically, the branching of a layer of placental trophoblast cells that sits between the maternal blood and fetal blood vessels. The discovery that expression of the transcription factor Gcm1 in trophoblast stem cells regulates branching of this specialized epithelium offers new insights into a process whose pivotal features have been generally obscure.The formation of a proper placenta is key to mammalian fetal development: without the placenta, fetuses die even in the absence of obvious defects1. On page 311 of this issue, Lynn Anson-Cartwright and colleagues2 report that the branching that generates the labyrinthine structure at the centre of the placenta is regulated by Gcm1, which encodes the transcription factor glial cells missing-1. Anson-Cartwright et al. and Jorg Schreiber et al.3 (who have recently reported a similar study in Molecular Cellular Biology) found that mice lacking Gcm1 fail to form the labyrinthine layer and succumb to placental insufficiency.
The placenta starts to form at a time when the metabolic requirements of the growing mouse embryo approach the capacity of the yolk sac. It is formed from the fusion of two tissues: the allantois, which derives from extra-embryonic mesoderm and eventually develops into the umbilical cord, and the extra-embryonic chorion (Fig. 1a), which derives from the polar trophectoderm overlying the inner cell mass of the blastula. Fetal blood vessels grow out of the allantoic mesoderm, invade the chorionic plate and instruct the trophoblast stem cells that reside within the chorionic plate to differentiate (Fig. 1b,c). At this point, the trophoblast cells fuse, forming syncytio-trophoblasts, and the three layers (allantoic mesoderm, interstitial syncytiotrophoblasts and chorion) come together to form the haemotrichorial labyrinth. Integration of fetal and maternal blood vessels within the labyrinth is critically dependent on the differentiation of the trophoblast stem cells to syncytiotrophoblasts. As the labyrinth forms, the chorion gives rise to trophoblast stem cells of the ectoplacental cone. Trophoblast giant cells differentiate from these stem cells and migrate outwards, embedding themselves in the uterine wall to anchor the forming placenta.
 | | Figure 1. Mutations affecting mouse placental development. | | |  | a, Mutations in genes encoding V-cam, a4 integrin and MJR (a co-chaperone) result in failure of the allantois and chorion to fuse. b, Lethal mutations affecting development at embryonic day (E) 9.5 to 10.5 (when the placenta supercedes the yolk sac) affect trophoblast (Gcm1, Hsp90b, Vhl, Dlx3) or fetal vessel function (Pparg, Tfeb, Mek1 and the gene encoding  V-integrin); all of these1,
3,
7,
8,
9,
10 confound formation of the labyrinth. c, Hgf and Esx1 mutations (which are not always lethal) affect later labyrinth development—from E11.5 to E16.5 (ref. 1).
 Full Figure and legend (32K) |
| |
Despite superficial differences, the labyrinth in the mouse and the floating chorionic villi in human are homologous structures, both characterized by extensive branching. And the similarities seem unlikely to stop there. The development of both mouse and human placentae (Fig. 2) may be partly regulated by the Gcm and basic helix-loop-helix (bHLH) gene families. In the mouse, two bHLH family members, Hand1 and Mash2, have antagonistic actions in determining trophoblast cell fate. Whereas Mash2 maintains trophoblast stem cells, Hand1 promotes the differentiation of trophoblast giant cells1,
4. The activity of each protein is regulated through its interaction with common dimerization partners; their success at competing for these factors determines whether differentiation takes place4. HAND1 is not expressed in the human placenta5, but other members of the bHLH family may have an equivalent role to mouse Hand1. HASH2, however, an orthologue of Mash2, is highly expressed in cytotrophoblast cells throughout pregnancy, as is GCM1 (which is not a bHLH protein).
| | Figure 2. Development of the mouse (a−c) and human (d−f) placenta. | | |  | a, At E10.5, the labyrinth forms. Maternal vessels are indicated in red, and fetal vessels, in light blue. b, The fetal vessels from allantoic mesoderm penetrate the chorion and invade the labyrinth. Syncytiotrophoblast (indicated in purple) forms around the developing fetal vessels. Maternal blood sinuses also enter the labyrinth and supply blood to the fetal vessels across the syncytiotrophoblast barrier. c, Trophoblast stem cells from the chorion express Gcm1 at the developing tips of the branching epithelium. Fetal mesenchyme containing vessel precursors undergoes branching morphogenesis and invasion behind the trophoblast layer. The invading trophoblasts will become the syncytiotrophoblast barrier between maternal blood sinuses and fetal blood vessels. d, Human placenta showing fetal villi and umbilical cord with fetal vessels in villi. Villi are bathed in maternal blood, which exits through the maternal veins. A syncytiotrophoblast sheath around the villi prevents direct contact between maternal and fetal blood. e, Two fetal villus trees illustrate the complex branching pattern. Each villus branch is also covered with branched microvilli (not shown) and surrounded by syncytiotrophoblast (indicated in purple). f, Each villus, depicted in cross section, is bound by a syncytiotrophoblast layer and bathed in maternal blood. Inside the villus is a layer of cytotrophoblast and the basement membrane. The central part of the villus contains fetal mesenchyme and vessels.
Full Figure and legend (83K) |
| |
Another bHLH family member at work in the placenta is ID2, whose expression is strong in cytotrophoblast stem cells but dissipates as they differentiate and invade the maternal blood vessels. Overexpression of ID2 in cytotrophoblast cultures increases the expression of GCM1, indicating that members of the GCM and bHLH families coordinately regulate cytotrophoblast differentiation5. The placentae of women with severe cases of pre-eclampsia, a disorder that sets in during the first trimester and involves hypertension, proteinuria and oedema, show reduced branching of the chorionic villi6 which may be caused by an abnormally low activity of a GCM or bHLH family member.
Mouse mutations
A vital characteristic of the haemotrichorial layer—that it facilitates and permits nutrient and gaseous exchange between mother and fetus—is consistent with runting or fetal death caused by mutations that affect it. In addition to the Gcm1 mutation, several other mouse mutations have been described that block distinct steps of placental development1,
3,
7,
8,
9,
10, in the absence of obvious embryonic defects. For example, in mice deficient in Hsp90 (a 'chaperone' protein that mediates protein folding), fetal vessels invade the chorion7 but trophoblast stem cells do not differentiate into syncytiotrophoblasts—and so a labyrinth never forms. Complementation with wild-type allantoic mesoderm, however, rescues labyrinth formation, indicating that Hsp90 is essential to syncytio-trophoblast differentiation.
In a similar manner to Gcm1 ablation, Ppar deficiency8 has a generally negative effect on the formation of a functional epithelial barrier by the syncytiotrophoblasts in the haemotrichorial layer. Gcm1 may be upstream of Pparg, as some branching and formation of some trilaminar layers is observed in the Pparg-/- embryos. Alternatively or concomitantly, mutant Ppar may affect syncytiotrophoblast differentiation independently, as trophoblasts at the haemotrichorial interface are unusually thick and accumulate lipids.
Branching out
When you mention the term "epithelial branching morphogenesis" in casual conversation (as one does), most people think: heart, breast, lung. Indeed, it is from these tissues that we have learned a great deal about the way in which interactions between the mesenchyme and epithelium direct and regulate branching morphogenesis. These tissues have provided us with several gene families that either signal or regulate branching morphogenesis11. For example, a transcription factor essential to branching morphogenesis of the embryonic kidney and lung is Pod1, another member of the bHLH family. Deletion of this gene results in perinatal lethality due to failure to form terminal air sacs, alveoli, and the air-ducts that supply them—all defects in terminal branching11. These animals also fail to develop fully differentiated glomeruli and have reduced numbers of glomeruli overall, another failure of terminal branching. In the rare cases where terminal branching does occur, the epithelium lining the branches is not fully differentiated or changes its fate.
In the placenta, the allantoic mesoderm instructs the chorionic stem cells, which then initiate the branching program through Gcm1. Unlike Gcm1, which regulates the earliest branching of the labyrinthine layer, Pod1 is expressed in mesenchyme and is dispensable for early branching in both kidney and lung. Neither Gcm1-/- trophoblasts nor Pod1-/- lung and kidney epithelia terminally differentiate but, in contrast with Pod1-/- lung epithelia, Gcm1-/- trophoblasts remain unaltered (that is, they fail to differentiate, instead of differentiating inappropriately). This indicates differences in the differentiation pathways regulated by these genes.
Invasion of the vasculature
Studies of bone formation12 clearly indicate that tissue morphogenesis is dependent on the development of, and coordination with, the vascular system. How this is regulated in organs like the kidney or lung remains obscure. From the work of Anson-Cartwright et al.2, however, it is clear that fetal vessel invasion and interdigitation with maternal blood sinuses require chorionic trophoblast branching morphogenesis and syncytiotrophoblast differentiation. Other mouse mutations result in lack of fetal vessel development or invasion of vessels into the labyrinth. For example, mutation of Esx1, which has a similar expression pattern to Gcm1, negatively regulates the branching of fetal vessels within the labyrinth13. Syncytiotrophoblast differentiation and the barrier function of the haemotrichorial layer are also affected in the Esx1-/- mice, which are runted at birth.
Whereas some pathways are likely to be different between mouse and human, a detailed understanding of the mechanisms that regulate branching morphogenesis during placental development will shed light on how these processes are altered in two severe problems of human pregnancy, pre-eclampsia and intrauterine growth retardation. Progress in understanding branching morphogenesis in the placenta also raises many new questions. What regulates the size of the surface area across which gas and nutrient exchange takes place? What mechanisms control the number and position of bifurcations? Does branching occur in response to hypoxic signals from the fetus? Why are there distinct trophoblast subtypes involved in invasion and the maternal-fetal barrier? The recent development of trophoblast stem cell lines14 offers promise not only for elucidating the differentiation of trophoblast lineages, but also for the rescue of fetuses with defective trophoblasts.
REFERENCES
- Rinkenberger, J.L., Cross, J.C. & Werb, Z. Dev. Genet. 21, 6–20 (1997). | Article | PubMed | ISI | ChemPort |
- Anson-Cartwright, L. et al. Nature Genet. 25, 311–314 (2000). | Article | PubMed | ISI | ChemPort |
- Schreiber, J. et al. Mol. Cell. Biol. 20, 2466–2474 (2000). | Article | PubMed | ISI | ChemPort |
- Scott, I.C., Anson-Cartwright, L., Riley, P., Reda, D. & Cross, J.C. Mol. Cell. Biol. 20, 530–541 (2000). | Article | PubMed | ISI | ChemPort |
- Janatpour, M.J. et al. Development 127, 549–558 (2000). | PubMed | ISI | ChemPort |
- Kingdom, J.C. & Kaufmann, P. Adv. Exp. Med. Biol. 474, 259–275 (1999). | PubMed | ISI | ChemPort |
- Voss, A.K., Thomas, T. & Gruss, P. Development 127, 1–11 (2000). | PubMed | ISI | ChemPort |
- Barak, Y. et al. Mol. Cell 4, 585–595 (1999). | Article | PubMed | ISI | ChemPort |
- Bader, B.L., Rayburn, H., Crowley, D. & Hynes, R.O. Cell 95, 507–519 (1998). | Article | PubMed | ISI | ChemPort |
- Giroux, S. et al. Curr. Biol. 9, 369–372 (1999). | Article | PubMed | ISI | ChemPort |
- Quaggin, S.E. et al. Development 126, 5771–5783 (1999). | PubMed | ISI | ChemPort |
- Vu, T.H. et al. Cell 93, 411–422 (1998). | Article | PubMed | ISI | ChemPort |
- Li, Y. & Behringer, R.R. Nature Genet. 20, 309–311 (1998). | Article | PubMed | ISI | ChemPort |
- Tanaka, S., Kunath, T., Hadjantonakis, A.K., Nagy, A. & Rossant, J. Science 282, 2072–2075 (1998). | Article | PubMed | ISI | ChemPort |
|
