Did climate change do for the tragulids?

Lesser Mouse-deer (Tragulus kanchil) at Singapore Zoo
Photo by Bjørn Christian Tørrisen (CC)

In a previous post, on giraffe and okapi placenta, I mentioned that tragulids (chevrotains or mouse-deer) were the most abundant ruminants in the Early Miocene. They were displaced by the pecoran ruminants and today are represented by a mere handful of species.

A new paper in PLoS One re-examines the European fossil fauna and shows tragulids already were on the way out in the Oligocene (full text here). The focus of the study is on a narrow time period called MP28 (MP stands for Mammal Palaeogene zone). This was a period of global warming that led to wooded environments being replaced by more open habitats and the appearance of seasonality including a dry season.

Pecorans have an additional forestomach, the omasum, and this may have given them the edge over tragulids in exploiting new resources. 

Placenta of Lesser Mouse-deer showing binucleate cells stained with
anti-bovine lactogen. From the Benirschke web site.

Although tragulids have the binucleate trophoblast cells that are the signature feature of ruminant placentation, they differ from pecorans in lacking cotyledons (reviewed here).

Selenka's gibbons

Bornean White-bearded Gibbon (Hylobates albibarbis)
Primate Info Net (University of Wisconsin) Photo Credit Marilyn Cole

Emil Selenka showed that the gibbon embryo, like that of humans and other apes, develops in the uterine wall beneath a decidua capsularis. But what species did he study?

Most of his figures are of a gibbon identified as Hylobates concolor (Harlan) from Borneo. The species name is still in use for Nomascus concolor, which is not found on Borneo. I now know, thanks to Dr. Thomas Geissman and his remarkable web site, that this reflects an extraordinary comedy of errors. Harlan described his ape as a hermaphrodite orangutan from Borneo; in fact it was a juvenile gibbon from Indochina!


Geographical distribution of gibbons.
(C) 2010 Thinh et al.

How then can we identify Selenka's gibbon? A study of mitochondrial genes (here) concluded that there were two species of gibbon on Borneo, one of them with three subspecies. Fortunately Selenka stated his specimens were collected on the left bank of the Kapuas River, in the territory occupied by the Bornean White-bearded Gibbon (Hylobates albibarbis) shown above.

Early stage of pregnancy in Hylobates albibarbis with amnion (A), yolk sac (D) and
exocoelom (Ex). The specimen had been flattened by contraction of the uterus but
the decidua capsularis (Dc) is clearly seen.

The embryo is depicted above. It had a primitive streak but no somites. Therefore it may correspond to Carnegie Stage 7 or early Stage 8 in the human. Selenka's paper can be found on the web (read only).

Evolution of the decidua

In preparation for pregnancy, the endometrium undergoes a process called decidualization (previous post). This involves a change in the size, shape and properties of the connective tissue cells (stromal fibroblasts). Decidualization is a necessary prerequisite for implantation of the blastocyst and often occurs in response to an embryonic signal. In women, decidualization happens in response to a maternal signal in the second half of the menstrual cycle.

 

Decidua was present in the most recent common ancestor of placental
mammals but was lost in some lineages. Data from A. M. Mess and A. M. Carter

Based on a phylogenetic analysis (here), Andrea Mess and I concluded that decidualization was present in the most recent common ancestor of placental mammals (extant Eutheria). It was lost in some lineages, especially in those that evolved a non-invasive epitheliochorial placenta.

Gray Four-eyed Opossum (Philander opossum)
Wikimedia Commons CC-BY-3.0 (André de Souza Pereira)

How about marsupials, all of which have a yolk sac placenta? In most placentation is non-invasive and none has been shown to have a decidua. In the Gray Four-eyed Opossum (Philander opposum), however, there is penetration of the endometrium by trophoblast and traces of a primitive decidual reaction (here). This ties in quite nicely with a recent study (here) of gene expression in the endometrium of the Gray Short-tailed Opossum. This identified a population of endometrial stromal fibroblasts that expressed progesterone receptor and some transcription factors associated with human decidual cells. On the other hand, the fibroblasts did not express the decidual marker desmin or other transcription factors required for decidualization.

The authors of the latter paper are part of a consortium that just published an extensive analysis of gene expression by the endometrium across mammals (here). The study included a frog, chicken, lizard, monotreme (Duck-billed Platypus), marsupial (Gray Short-tailed Opossum), and seven different placental mammals. It identified a huge number of genes that were recruited during the evolution of pregnancy in mammals, including many that are associated with the decidualization process.

The main thrust of the new paper is the central role played in evolution by transposable elements. These were co-opted into regulatory elements that coordinate the endometrial progesterone response.