Oliver W. Griffith

Ancestral state reconstructions require biological evidence to test evolutionary hypotheses: A case study examining the evolution of reproductive mode in squamate reptiles

To understand evolutionary transformations it is necessary to identify the character states of extinct ancestors. Ancestral character state reconstruction is inherently difficult because it requires an accurate phylogeny, character state data, and a statistical model of transition rates and is fundamentally constrained by missing data such as extinct taxa. We argue that model based ancestral character state reconstruction should be used to generate hypotheses but should not be considered an analytical endpoint. Using the evolution of viviparity and reversals to oviparity in squamates as a case study, we show how anatomical, physiological, and ecological data can be used to evaluate hypotheses about evolutionary transitions. The evolution of squamate viviparity requires changes to the timing of reproductive events and the successive loss of features responsible for building an eggshell. A reversal to oviparity requires that those lost traits re-evolve. We argue that the re-evolution of oviparity is inherently more difficult than the reverse. We outline how the inviability of intermediate phenotypes might present physiological barriers to reversals from viviparity to oviparity. Finally, we show that ecological data supports an oviparous ancestral state for squamates and multiple transitions to viviparity. In summary, we conclude that the first squamates were oviparous, that frequent transitions to viviparity have occurred, and that reversals to oviparity in viviparous lineages either have not occurred or are exceedingly rare. As this evidence supports conclusions that differ from previous ancestral state reconstructions, our paper highlights the importance of incorporating biological evidence to evaluate model-generated hypotheses.

Seahorse brood pouch transcriptome reveals common genes associated with vertebrate pregnancy

Viviparity (live birth) has evolved more than 150 times in vertebrates, and represents an excellent model system for studying the evolution of complex traits. There are at least 23 independent origins of viviparity in fishes, with syngnathid fishes (seahorses and pipefish) unique in exhibiting male pregnancy. Male seahorses and pipefish have evolved specialized brooding pouches that provide protection, gas exchange, osmoregulation, and limited nutrient provisioning to developing embryos. Pouch structures differ widely across the Syngnathidae, offering an ideal opportunity to study the evolution of reproductive complexity. However, the physiological and genetic changes facilitating male pregnancy are largely unknown. We used transcriptome profiling to examine pouch gene expression at successive gestational stages in a syngnathid with the most complex brood pouch morphology, the seahorse Hippocampus abdominalis. Using a unique time-calibrated RNA-seq data set including brood pouch at key stages of embryonic development, we identified transcriptional changes associated with brood pouch remodeling, nutrient and waste transport, gas exchange, osmoregulation, and immunological protection of developing embryos at conception, development and parturition. Key seahorse transcripts share homology with genes of reproductive function in pregnant mammals, reptiles, and other live-bearing fish, suggesting a common toolkit of genes regulating pregnancy in divergent evolutionary lineages.

Embryo implantation evolved from an ancestral inflammatory attachment reaction

Significance Our data suggest that implantation in eutherians is derived from an ancestral inflammatory reaction to embryo attachment in the therian ancestor. These results explain the paradoxical role of inflammation at the beginning and the end of pregnancy in humans: Inflammation is necessary for implantation and parturition, but for most of pregnancy, inflammation threatens the continuation of pregnancy. We argue that the role of inflammation during implantation is an ancestral response to the embryo as a foreign body. By changing the way investigators think about implantation, we expect this research to contribute to new ways to study and treat implantation disorders, the most vulnerable step of assisted reproductive technology, in women. The molecular changes that support implantation in eutherian mammals are necessary to establish pregnancy. In marsupials, pregnancy is relatively short, and although a placenta does form, it is present for only a few days before parturition. However, morphological changes in the uterus of marsupials at term mimic those that occur during implantation in humans and mice. We investigated the molecular similarity between term pregnancy in the marsupials and implantation in eutherian mammals using the gray short-tailed opossum (Monodelphis domestica) as a model. Transcriptomic analysis shows that term pregnancy in the opossum is characterized by an inflammatory response consistent with implantation in humans and mice. This immune response is temporally correlated with the loss of the eggshell, and we used immunohistochemistry to report that this reaction occurs at the materno–fetal interface. We demonstrate that key markers of implantation, including Heparin binding EGF-like growth factor and Mucin 1, exhibit expression and localization profiles consistent with the pattern observed during implantation in eutherian mammals. Finally, we show that there are transcriptome-wide similarities between the opossum attachment reaction and implantation in rabbits and humans. Our data suggest that the implantation reaction that occurs in eutherians is derived from an attachment reaction in the ancestral therian mammal which, in the opossum, leads directly to parturition. Finally, we argue that the ability to shift from an inflammatory attachment reaction to a noninflammatory period of pregnancy was a key innovation in eutherian mammals that allowed an extended period of intimate placentation.

High Food Abundance Permits the Evolution of Placentotrophy: Evidence from a Placental Lizard, Pseudemoia entrecasteauxii

Mechanisms of reproductive allocation are major determinants of fitness because embryos cannot complete development without receiving sufficient nutrition from their parents. The nourishment of offspring via placentas (placentotrophy) has evolved repeatedly in vertebrates, including multiple times in squamate reptiles (lizards and snakes). Placentotrophy has been suggested to evolve only if food is sufficiently abundant throughout gestation to allow successful embryogenesis. If scarcity of food prevents successful embryogenesis, females should recoup nutrients allocated to embryos via abortion, reabsorption, and/or cannibalism. We tested these hypotheses in the placentotrophic southern grass skink Pseudemoia entrecasteauxii. We fed females one of four diets (high constant, high variable, low constant, and low variable) during gestation and tested the effects of both food amount and schedule of feeding on developmental success, cannibalism rate, placental nutrient transport, offspring size, and maternal growth and body condition. Low food availability reduced developmental success, placental nutrient transport, offspring size, and maternal growth and body condition. Cannibalism of offspring also increased when food was scarce. Schedule of feeding did not affect offspring or mothers. We suggest that high food abundance and ability to abort and cannibalize poor-quality offspring are permissive factors necessary for placentotrophy to be a viable strategy of reproductive allocation.

Placental Lipoprotein Lipase (LPL) Gene Expression in a Placentotrophic Lizard, Pseudemoia entrecasteauxii

Viviparity (live birth) relies on a functional placenta, which is formed by cooperating maternal and embryonic tissues. In some viviparous lineages, mothers use this placenta to transport nutrients to feed developing embryos through pregnancy (placentotrophy). The Australian lizard, Pseudemoia entrecasteauxii, provides approximately 60% of the lipid for embryonic growth and metabolism to embryos across the placenta. Lipoprotein lipase (LPL) is an important enzyme in lipid transport in vertebrates. We examined patterns of LPL gene expression to identify its role in the uterus of pregnant P. entrecasteauxii. We used reverse transcription quantitative real time PCR to measure the expression of the LPL gene in the uterine tissue throughout reproduction and compared uterine LPL expression in chorioallantoic and yolk-sac placentae. Expression of the LPL gene is significantly higher in the uterus of late pregnant compared to non-pregnant and early pregnant P. entrecasteauxii, indicating a greater capacity for lipid transport towards the end of pregnancy. The period of high LPL gene expression correlates with the time that developing embryos are undergoing the greatest growth and have the highest metabolic rate. LPL gene expression is significantly higher in the uterine tissue of the yolk-sac placenta than the chorioallantoic placenta, providing the first molecular evidence that the yolk-sac placenta is the major site of lipid transport in pregnant P. entrecasteauxii.

Comparative genomics of hormonal signaling in the chorioallantoic membrane of oviparous and viviparous amniotes

In oviparous amniotes (reptiles, birds, and mammals) the chorioallantoic membrane (CAM) lines the inside of the egg and acts as the living point of contact between the embryo and the outside world. In livebearing (viviparous) amniotes, communication during embryonic development occurs across placental tissues, which form between the uterine tissue of the mother and the CAM of the embryo. In both oviparous and viviparous taxa, the CAM is at the interface of the embryo and the external environment and can transfer signals from there to the embryo proper. To understand the evolution of placental hormone production in amniotes, we examined the expression of genes involved in hormone synthesis, metabolism, and hormone receptivity in the CAM of species across the amniote phylogeny. We collected transcriptome data for the chorioallantoic membranes of the chicken (oviparous), the lizards Lerista bougainvillii (both oviparous and viviparous populations) and Pseudemoia entrecasteauxii (viviparous), and the horse Equus caballus (viviparous). The viviparous taxa differ in their mechanisms of nutrient provisioning: L. bougainvillii is lecithotrophic (embryonic nourishment is provided via the yolk only), but P. entrecasteauxii and the horse are placentotrophic (embryos are nourished via placental transport). Of the 423 hormone-related genes that we examined, 91 genes are expressed in all studied species, suggesting that the chorioallantoic membrane ancestrally had an endocrine function. Therefore, the chorioallantoic membrane appears to be a highly hormonally active organ in all amniotes. No genes are expressed only in viviparous species, suggesting that the evolution of viviparity has not required the recruitment of any specific hormone-related genes. Our data suggest that the endocrine function of the CAM as a placental tissue evolved in part through co-option of ancestral gene expression patterns.

Allelic expression of mammalian imprinted genes in a matrotrophic lizard, Pseudemoia entrecasteauxii

Genomic imprinting is a process that results in the differential expression of genes depending on their parent of origin. It occurs in both plants and live-bearing mammals, with imprinted genes typically regulating the ability of an embryo to manipulate the maternal provision of nutrients. Genomic imprinting increases the potential for selection to act separately on paternally and maternally expressed genes, which increases the number of opportunities that selection can facilitate embryonic control over maternal nutrient provision. By looking for imprinting in an independent matrotrophic lineage, the viviparous lizard Pseudemoia entrecasteauxii (Scincidae), we test the hypothesis that genomic imprinting facilitates the evolution of substantial placental nutrient transport to embryos (matrotrophy). We sequenced transcriptomes from the embryonic component of lizard placentae to determine whether there are parent-of-origin differences in expression of genes that are imprinted in mammals. Of these genes, 19 had sufficiently high expression in the lizard to identify polymorphisms in transcribed sequences. We identified bi-allelic expression in 17 genes (including insulin-like growth factor 2), indicating that neither allele was imprinted. These data suggest that either genomic imprinting has not evolved in this matrotrophic skink or, if it has, it has evolved in different genes to mammals. We outline how these hypotheses can be tested. This study highlights important differences between mammalian and reptile pregnancy and the absence of any shared imprinting genes reflects fundamental differences in the way that pregnancy has evolved in these two lineages.

The placenta as a model for understanding the origin and evolution of vertebrate organs

How organs originate and evolve is a question fundamental to understanding the evolution of complex multicellular life forms. Vertebrates have a relatively standard body plan with more or less the same conserved set of organs. The placenta is a comparatively more recently evolved organ, derived in many lineages independently. Using placentas as a model, we discuss the genetic basis for organ origins. We show that the evolution of placentas occurs by acquiring new functional attributes to existing tissues, changes in the patterning and development of tissues, and the evolution of novel cell types. We argue that a diversity of genomic changes facilitated these physiological transformations and that these changes are likely to have occurred during the evolution of organs more broadly. Finally, we argue that a key aspect to understanding the evolutionary origin of organs is that they are likely to result from novel interactions between distinct cell populations.

The mammalian decidual cell evolved from a cellular stress response

Among animal species, cell types vary greatly in terms of number and kind. The number of cell types found within an organism differs considerably between species, and cell type diversity is a significant contributor to differences in organismal structure and function. These observations suggest that cell type origination is a significant source of evolutionary novelty. The molecular mechanisms that result in the evolution of novel cell types, however, are poorly understood. Here, we show that a novel cell type of eutherians mammals, the decidual stromal cell (DSC), evolved by rewiring an ancestral cellular stress response. We isolated the precursor cell type of DSCs, endometrial stromal fibroblasts (ESFs), from the opossum Monodelphis domestica. We show that, in opossum ESFs, the majority of decidual core regulatory genes respond to decidualizing signals but do not regulate decidual effector genes. Rather, in opossum ESFs, decidual transcription factors function in apoptotic and oxidative stress response. We propose that rewiring of cellular stress responses was an important mechanism for the evolution of the eutherian decidual cell type.

Reply to Liu: Inflammation before implantation both in evolution and development

In a paper in PNAS (1) we show that, in opossums, endometrial inflammation follows attachment of the embryo to the uterine wall. We argue that this is similar to what occurred in the first live-bearing mammals, and that inflammatory signaling seen at implantation in eutherians evolved from this attachment-induced inflammation. In a letter on our paper, Liu (2) presents evidence of endometrial inflammatory gene expression in mice before embryonic attachment. Liu infers that inflammatory signaling at implantation cannot be a consequence of embryonic attachment because inflammation precedes implantation in mice and humans. Liu alternatively proposes that inflammation during implantation in eutherians has evolved because the embryo coopted L-selectin expression, allowing it to “act” like a leukocyte and facilitate endometrial infiltration. This is an interesting hypothesis, because it may explain why inflammatory … [↵][1]1To whom correspondence should be addressed. Email: oliver.griffith{at}yale.edu. [1]: #xref-corresp-1-1