The fetal-maternal interface of the nine-banded armadillo: endothelial cells of maternal sinus are partially replaced by trophoblast
© The Author(s). 2016
Received: 11 March 2016
Accepted: 3 June 2016
Published: 8 June 2016
The evolution of invasive placentation in the stem lineage of eutherian mammals entailed resolution of the incompatibility between a semi-allogenic fetus and the maternal immune system. The haemochorial placenta of nine-banded armadillo (Dasypus novemcinctus) is thought to conceal itself from the maternal immune system to some degree by developing inside a preformed blood sinus, with minimal contact with the uterine connective tissue. In the present study, we elucidate the micro-anatomical relationship between fetal and maternal tissue of the nine-banded armadillo using histochemical and immunohistochemical tools.
We conclude that the chorio-allantoic villi are separated from the myometrium by a vascular endothelial layer, as previously proposed. However, we also observe that the trophoblast cells establish direct contact with the endometrial stroma on the luminal side of the endometrium by partially replacing the endothelial lining of the sinus. Further, we demonstrate the presence of leukocytes, perhaps entrapped, in the placental fibrinoids at the interface between the intervillous space and the endometrial arcade.
The trophoblast of the armadillo invades the uterine tissue to a greater extent than was previously believed. We discuss the implications of this finding for the fetal-maternal immune tolerance.
KeywordsFetal-maternal interface Fetal-maternal tolerance Dasypus novemcinctus Nine-banded armadillo Placental fibrinoid
Viviparous mammals exhibit one of three types of placentation; in order of increasing invasiveness of the placenta, these are epitheliochorial, endotheliochorial, or haemochorial . Epitheliochorial placentation (e.g., in most marsupials, cattle, sheep) is non-invasive because the chorion does not establish sustained contact with uterine stroma. The other two types of placentation are invasive, meaning that the fetal chorion is in a sustained (>2 days) interaction with the endometrial stoma and/or maternal blood. Chorion invades only the luminal epithelium and contacts the maternal vascular endothelium in an endotheliochorial placenta (e.g., in carnivores, elephants), while it erodes both the luminal epithelium and the endothelium to directly contact the maternal blood in a haemochorial placenta (e.g., in human, rodents a. o. m.). Both types of invasive placentation lead to a direct and extended apposition of the fetal tissue and endometrial connective tissue.
Invasive placentation evolved in the stem lineage of eutherian mammals (also known as placental mammals) [2–4]. Invasive placentation faces two physiological challenges: (i) destruction of maternal tissue by the blastocyst induces an inflammatory reaction, and (ii) the invading blastocyst is semi-allogenic relative to the mother. That is, about half of the genes expressed by the fetus are of paternal origin, and thus are potential alloantigens for the maternal immune system . Evolution of invasive placentation in eutherian stem lineage necessitated the evolution of mechanisms to control inflammation , and to allow for the intrauterine growth of the fetus without undergoing rejection by the maternal immune system. Immunology at the fetal-maternal interface is an area of active research, and numerous mechanisms of immune tolerance have been discovered in mouse and human (reviewed in ).
Here we investigate the fetal-maternal interface of the nine-banded armadillo (Dasypus novemcinctus). Armadillo belongs to the eutherian clade Xenarthra. Xenarthra and the well-studied Euarchontoglires (e.g., mouse and human) bracket the entire diversity of eutherian mammals [8–10]. Its phylogenetic placement makes Xenarthra a critical lineage for better understanding the evolutionary origin of eutherian pregnancy.
Armadillo has a haemochorial placenta, but the manner in which the haemochorial arrangement is achieved is considerably different from that of other species with haemochorial placentation [11–14], as described below.
Enders  has divided armadillo placental development into three phases: (i) avillous phase of initial attachment, (ii) establishment and expansion of villous placenta, and (iii) phase of mature (fully developed or definitive) placenta. The initial site of attachment of the blastocyst is the most distal (i.e., anterior) tip of the fundus, at the intersection of two prominent mucosal folds perpendicular to each other . Endometrium contains preformed blood sinuses, even in the non-pregnant stage and during diapause, which are close to the luminal surface in the fundus . The blastocyst attaches to the fundic mucosa by the trophoblast layer on the embryonic pole (also known as the Rauber’s layer). Subsequently, the Rauber’s layer forms an annular thickening at the periphery of the initial site of attachment. Trophoblast on the abembryonic pole degenerates in the meantime, consequently giving rise to an inverted yolk-sac placenta towards the uterine corpus, in addition to the chorio-allantoic placenta being formed on the embryonic side.
In the second phase, luminal epithelium is penetrated by the trophoblast, and villous placenta is established in the preformed blood sinuses. Trophoblast invades luminal epithelium at limited areas, and sends fingerlike projections into the preformed blood sinuses that lie closely underneath the epithelium. These trophoblastic projections mature into chorio-allantoic villi upon development of mesodermal cores from the allantois. Further development and branching of the villi takes place within the blood sinuses, and is accommodated by expansion of the sinuses, rather than by destruction of maternal tissue. Maturation of the placenta is complete when fetuses are 6–7 cm in length, and is marked by the disappearance of the cytotrophoblast, leaving the villi without any cytotrophoblastic cell columns at their distal tips. From the disappearance of cytotrophoblast to parturition is the phase of mature or definitive placenta in which no more villous growth is thought to take place.
In the mature placenta, the myometrium and the blood sinuses are not separated by much endometrial stromal tissue, and as a consequence, armadillo lacks a decidua basalis between the placental villi and myometrium. Endometrial tissue is retained in small islands on the myometrial side, and on the luminal side of the villi in the form of a thin arch over the villous tree. This remnant of endometrium is not a decidua capsularis, as it encapsulates only the placenta, not the fetus. Enders and colleagues  aptly refer to it as the endometrial arcade. The maternal blood sinus can be considered as the intervillous space of the placenta since villi develop and lie within the sinus. Note that the haemochorial configuration is reached without extensive damage to maternal tissue except at the original sites of penetration.
In this study we describe the microanatomy of the interface between the placental villi and uterus. We show that the long-held idea that the armadillo trophoblast makes no physical contact with the endometrial connective tissue, by growing within a blood sinus lined by an uninterrupted vascular endothelium [12, 17], is valid for the myometrial side, but not for the arcadal side of the sinus. Partial invasion of vascular endothelium by trophoblast establishes contact between the trophoblast and endometrial stroma.
Animals and tissue harvesting
Animals used in this study
Stage of pregnancy
Crown-Rump length of fetuses
Post-implantation; before the establishment of vascular villi (early-gestation)
After the establishment of vascular villi, and before the loss of cell columns from the tips of villi (mid-gestation)
After the establishment of vascular villi, and before the loss of cell columns from the tips of villi (mid-gestation)
Harvested tissues were fixed in 4 % paraformaldehyde for 12–24 h, and partially dehydrated by successive washes in 50 % and 70 % ethanol for an hour each. At this point, tissues were shipped from Centerville, TX to our lab on ice–packs and stored at −20 °C until further processing.
Paraffin embedding and sectioning
Tissues were dehydrated with successive washes in ethanol, cleared in toluene, and embedded in paraffin blocks. Sections of thickness 2 μ or 5 μ were made on a microtome (Reichert-Jüng) and placed on poly-l-lysine coated glass slides.
Haematoxylin and eosin staining (H&E)
Haematoxylin and eosin staining was performed following the standard procedure with 1–2 min in haematoxylin and 1 min in eosin.
Masson’s trichrome staining
Protocol from  was followed. Slides were deparaffinized with xylene, brought to 70 % ethanol, and incubated overnight in zinc-formalin fixative at room temperature. Slides were dipped in deionized water (1 min), stained with Weigert’s iron-haematoxylin (3 min), and rinsed under running tap water (1 min). They were then stained with Biebrich Scarlet – Acid Fuchsin solution (4 min) and rinsed in acidified water to remove excess dye. The dye was further removed from collagen by immersing the slides in a solution of phosphomolybdic acid and phosphotungstic acid for about 10 min – this step was monitored under microscope to avoid excessive differentiation. Differentiation was followed by staining with 2 % Fast Green FCF (4 min), and two washes in acidified water (30 s each). Slides were dehydrated in ethanol, cleared in xylene, and coverslipped with Permount.
Periodic Acid – Schiff (PAS) and Periodic Acid – Schiff – Diastase (PAS-D) staining
PAS and PAS-D staining was performed on serial sections to make the two comparable. Protocol from  was followed. Slides were deparaffinized with xylene and brought to water. For PAS-D staining, the slides were incubated with 1 mg/ml of α-amylase (i.e., diastase; Sigma-Aldrich) for 30 min at 37 °C, rinsed in tap water, and then with distilled water. For PAS staining, this step was performed with water instead of α -amylase solution. Slides were then oxidized for 30 min with 1 % solution of periodic acid, washed under running tap water for 3 min, and immersed in Schiff’s reagent for 20 min. Slides were washed under copiously running tap water for 10 min. Counterstaining was performed with Gill 2 haematoxylin for 1–2 min. Slides were dehydrated in ethanol, cleared in xylene, and coverslipped with Permount.
Antibodies used in this study
Antibody product ID
1:50 (DAB); 1:200 (Alexa Fluor488)
Alexa Fluor 488
Alexa Fluor 555
Vascular endothelium of the maternal blood sinuses
In sum, the vascular endothelium of the preformed blood sinuses remains intact on the myometrial side, but is partially interrupted and replaced by trophoblast cells on the arcadal side. It is the latter that can be termed as the junctional zone of the definitive armadillo placenta, where fetal and endometrial stromal tissue lie in direct apposition.
Fibrinoids are not seen in H&E stained sections of the early-gestation stage of placentation studied, after implantation but before the establishment of vascular villi. Absence of fibrinoids at this stage was also confirmed by vWF immuno-staining and Masson’s trichrome staining (data not shown).
Endometrial extracellular matrix, especially collagens, undergoes substantial reorganization during pregnancy, which is necessary for accommodation of the growing fetus in the uterus, provision of a substrate for the placental structures and the immune cells at the fetal-maternal interface . We performed Masson’s trichrome staining to document the dynamics of the collagen in armadillo endometrium during pregnancy.
We tested for the presence of glycogen-containing cells in the armadillo fetal-maternal interface by Periodic Acid- Schiff (PAS) staining in conjunction with amylase (diastase) digestion (PAS-D). Briefly, PAS stains glycogen, glycoproteins, and glycolipids. Comparison of sections stained with PAS, with and without prior amylase-digestion, is used to detect glycogen .
Many cell-types bearing glycogen stores are present at the fetal-maternal interface in rodents and human . The precise function of glycogen stores in these cell-types is unclear, but they have been ascribed a nutritive function to the fetus , as well as have been proposed to be a necessary outcome of a hypoxic environment . Here we showed that the epithelia of the endometrium are glycogen storing cells in the armadillo during pregnancy, consistent with earlier findings by Enders and colleagues . It is possible that they provide the resources for the synthesis of mucopolysaccharides typically present on the luminal epithelia. Unlike in armadillo, in human  and rodent species , glycogen has also been described in endometrial stromal and decidual cells during pregnancy. It is not present in the armadillo endometrial stroma even in the early stage for which a decidual reaction has been described by Enders and colleagues . We did not detect glycogen in the syncytial trophoblast of armadillo (data not shown), similar to the situation in human  and rodents, with the potential exception of rat .
Placental fibrinoids have been described in species with haemochorial placentation [13, 24–26]. There are two types of placental fibrinoid: fibrin-type and matrix-type. The former is primarily a blood clot, while the latter is rich in extracellular matrix proteins e.g., collagen, fibronectin, laminin . Nelson and colleagues  have described fibrin-type fibrinoid in the term placenta of armadillo, masking the discontinuities within syncytial trophoblast layer surrounding the villi. We describe fibrinoids in our mid-gestation specimen of armadillo, even in specimens collected in January, two months before term. We also show that the fibrinoids are not limited to perivillous regions but are also seen in endometrial arcades. The remarkably high density of leukocytes present within the arcadal fibrinoids suggests a role in the regulation of immune reaction (see below).
A major enigma in reproductive biology is the question of how the maternal immune system tolerates the semi-allogenic fetus. Several mechanisms have been put forth to answer this question. In mouse these mechanisms include exclusion of cytotoxic T-cells from the fetal-maternal interface , regulatory T-cells , antigen presentation limited to the indirect pathway  or restriction of antigen-presentation altogether by the entrapment of dendritic cells in the endometrium . Armadillo, with its haemochorial placenta, also faces this immunological challenge. Enders and Welsh  have argued that the peculiar way in which armadillo placenta develops—with little destruction of maternal tissue and little exposure of the endometrial stroma to the trophoblast—may help explain fetal-maternal tolerance. They propose that armadillo trophoblast is protected from the maternal immune system because most of the placental development takes place within a confined maternal blood space lined by an intact endothelium. There is minimal damage to the maternal tissue, and there is minimal contact between the fetal tissue and maternal connective tissue, only at the original site of invasion. Immune cells are typically inactive in circulation; an alloantigen provokes the immune system more potently when it is present interstitially, in the connective tissue, and in the context of an inflammatory reaction, rather than intravascularly, as in the armadillo placenta.
It should be noted that Rezende and colleagues  studied three species of armadillo (Euphractus sexcinctus, Chaetophractus villosus and Chaetophractus vellerosus), that did not include the one studied here, nine-banded armadillo (Dasypus novemcinctus), and concluded that the endothelium of maternal blood sinuses was intact. They validated this interpretation by vimentin staining in Chaetophractus villosus. In the paper, the authors have documented immunohistochemical evidence for presence of an intact endothelium only on the myometrial side. It is not possible to determine from that report whether the endothelium is intact on the side of endometrial arcade, which is where we found it interrupted in Dasypus.
The revised understanding of the armadillo fetal-maternal interface presented here calls for extension to the model of immune tolerance proposed by Enders and Welsh . One possible mechanism for preventing an inflammatory response could be the placental fibrinoids. Fibrinoids have been hypothesized to suppress immune reaction by acting as an ‘immuno-absorbent sponge’ . Although their role in armadillo placenta remains to be studied in detail, our study provides evidence for the presence of numerous leukocytes, potentially trapped, in the placental fibrinoids. Which additional means of immune tolerance armadillo employs is unclear at the moment, but the biology of interaction between armadillo mother and fetus should be explored further to answer that question. In particular, one would expect additional mechanisms ensuring limited inflammation and immune tolerance to be deployed in the endometrial arcade, the junctional zone of the armadillo placenta.
We have documented the dynamics of collagen and glycogen at the fetal-maternal interface of nine-banded armadillo, and reported the presence of placental fibrinoids on the villi and in the endometrial arcade. We have also shown that the trophoblast of nine-banded armadillo invades the maternal tissue to a greater extent than was previously believed. It partially replaces the endothelial lining of the maternal blood sinuses and establishes a direct contact with endometrial stromal cells. This finding calls for further study of immunological tolerance and regulation of inflammation at the fetal-maternal interface in armadillo. Given its phylogenetic position, insights from armadillo, when combined with knowledge from other eutherian species, are critical for understanding the evolution of eutherian pregnancy.
Bovine serum albumin
Tyramide signal amplification
We would like to thank Dr. Harvey Kliman and Kristin Milano (Yale University) for their help in optimizing the immunohistochemistry protocol. We are grateful to Prof. Anthony Carter (University of Southern Denmark) and an anonymous reviewer for pointing us to papers by H. Strahl. GPW is grateful to Wes and Kendall Dunn, TX, for their help with collecting armadillos and for providing access to their property.
John Templeton Foundation grant #54860: “How new cell-types arise in evolution”. Opinions expressed in the article are not necessarily those of John Templeton Foundation.
Availability of data and materials
There are no additional data and material to be presented.
ARC and GPW conceived the study. GPW carried out the collection of animals. ARC performed the experiments. ARC and GPW interpreted the data and wrote the manuscript. Both authors read and approved the final manusript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Collection of animals was approved by the Yale University IACUC under protocol #2014-10906.
Consent to participate is not applicable because no human tissue or data were used in this study.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Grosser O. Frühentwicklung, Eihautbildung Und Placentation Des Menschen Und Der Säugetiere. München: JF Bergmann; 1927.
- Mess A, Carter AM. Evolutionary transformations of fetal membrane characters in Eutheria with special reference to Afrotheria. J Exp Zool B Mol Dev Evol. 2006;306(2):140–63.View ArticlePubMedGoogle Scholar
- Wildman DE, et al. Evolution of the mammalian placenta revealed by phylogenetic analysis. Proc Natl Acad Sci U S A. 2006;103(9):3203–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Elliot MG, Crespi BJ. Phylogenetic Evidence for Early Hemochorial Placentation in Eutheria. Placenta. 2009;30(11):949–67.View ArticlePubMedGoogle Scholar
- Medawar PB. Some immunological and endocrinological problems raised by the evolution of viviparity in vertebrates. In: Symp Soc Exp Biol. 1953.Google Scholar
- Chavan AR, Bhullar B-AS, Wagner GP. What was the ancestral function of decidual stromal cells? A model for the evolution of eutherian pregnancy. Placenta. 2016;40:40–51.View ArticlePubMedGoogle Scholar
- Erlebacher A. Immunology of the Maternal-Fetal Interface. Annu Rev Immunol. 2013;31:387–411.View ArticlePubMedGoogle Scholar
- dos Reis M, et al. Phylogenomic datasets provide both precision and accuracy in estimating the timescale of placental mammal phylogeny. Proc Biol Sci. 2012;279(1742):3491–500.View ArticlePubMedPubMed CentralGoogle Scholar
- O'Leary MA, et al. The Placental Mammal Ancestor and the Post-K-Pg Radiation of Placentals. Science. 2013;339(6120):662–7.View ArticlePubMedGoogle Scholar
- Tarver JE et al. The Interrelationships of Placental Mammals and the Limits of Phylogenetic Inference. Genome Biol Evol. 2016.
- Strahl H. Beiträge zur Entwicklungsgeschichte von Tatusia novemcincta L. In: Gasser E, editor. Festschrift Zum LXX. Geburtstage. Springer Berlin Heidelberg: Berlin; 1917. p. 1–40.View ArticleGoogle Scholar
- Enders AC. Development and structure of the villous haemochorial placenta of the nine-banded armadillo (Dasypus novemcinctus). J Anat. 1960;94(Pt 1):34–45.PubMedPubMed CentralGoogle Scholar
- Ramsey EM. The placenta: human and animal. New York: Praeger; 1982.Google Scholar
- Enders AC, Carter AM. Review: The evolving placenta: Different developmental paths to a hemochorial relationship. Placenta. 2012;33(Supplement):S92–8.View ArticlePubMedGoogle Scholar
- Patterson JT. Polyembryonic development in Tatusia novemcincta. J Morphol. 1913;24(4):559–683.View ArticleGoogle Scholar
- Enders AC, Buchanan GD, Talmage RV. Histological and Histochemical Observations on the Armadillo Uterus during the Delayed and Postimplantation Periods. Anat Rec. 1958;130(4):639–57.View ArticlePubMedGoogle Scholar
- Enders AC, Welsh AO. Structural Interactions of Trophoblast and Uterus during Hemochorial Placenta Formation. J Exp Zool. 1993;266(6):578–87.View ArticlePubMedGoogle Scholar
- Kiernan JA. Histological and histochemical methods: theory and practice. Bloxham: Scion; 2009.Google Scholar
- Aplin JD, et al. The Endometrium: Molecular, Cellular and Clinical Perspectives. London: Informa Healthcare; 2008.
- Wooding FBP, Burton G. Comparative placentation: structures, functions and evolution. Berlin: Springer; 2008. p. 301. xi.View ArticleGoogle Scholar
- Mossman HW. Comparative morphogenesis of the fetal membranes and accessory uterine structures. Contributions to Embryology. 1937;26(158):133–7.Google Scholar
- Dempsey EW, Wislocki GB. Observations on some histochemical reactions in the human placenta, with special reference to the significance of the lipoids, glycogen and iron. Endocrinology. 1944;35(6):409–29.View ArticleGoogle Scholar
- Wislocki GB, Deane HW, Dempsey EW. The histochemistry of the rodent's placenta. Am J Anat. 1946;78(3):281–345.View ArticlePubMedGoogle Scholar
- Kaufmann P, Huppertz B, Frank H-G. The fibrinoids of the human placenta: origin, composition and functional relevance. Ann Anat. 1996;178(6):485–501.View ArticlePubMedGoogle Scholar
- Bradbury S, Billington WD, Kirby DR. A histochemical and electron microscopical study of the fibrinoid of the mouse placenta. J R Microsc Soc. 1965;84:199–211.View ArticlePubMedGoogle Scholar
- Nelson DM, et al. Ontogenetic and phylogenetic evaluation of the presence of fibrin-type fibrinoid in the villous haemochorial placenta. Placenta. 1997;18(7):605–8.View ArticlePubMedGoogle Scholar
- Frank HG, et al. Immunohistochemistry of two different types of placental fibrinoid. Acta Anat (Basel). 1994;150(1):55–68.View ArticleGoogle Scholar
- Nancy P et al. Chemokine gene silencing in decidual stromal cells limits T cell access to the maternal-fetal interface. Science. 2012;336(6086):1317–21.View ArticlePubMedPubMed CentralGoogle Scholar
- Aluvihare VR, Kallikourdis M, Betz AG. Regulatory T cells mediate maternal tolerance to the fetus. Nat Immunol. 2004;5(3):266–71.View ArticlePubMedGoogle Scholar
- Erlebacher A, et al. Constraints in antigen presentation severely restrict T cell recognition of the allogeneic fetus. J Clin Invest. 2007;117(5):1399–411.View ArticlePubMedPubMed CentralGoogle Scholar
- Collins MK, Tay CS, Erlebacher A. Dendritic cell entrapment within the pregnant uterus inhibits immune surveillance of the maternal/fetal interface in mice. J Clin Invest. 2009;119(7):2062–73.PubMedPubMed CentralGoogle Scholar
- Rezende LC, et al. The fetomaternal interface in the placenta of three species of armadillos (Eutheria, Xenarthra, Dasypodidae). Reprod Biol Endocrinol. 2012;10:38.View ArticlePubMedPubMed CentralGoogle Scholar