Insights into the formation and diversification of a novel chiropteran wing membrane from embryonic development

BACKGROUND

Through the evolution of novel wing structures, bats (Order Chiroptera) became the only mammalian group to achieve powered flight. This achievement preceded the massive adaptive radiation of bats into diverse ecological niches. We investigate some of the developmental processes that underlie the origin and subsequent diversification of one of the novel membranes of the bat wing: the plagiopatagium, which connects the fore- and hind limb in all bat species.

RESULTS

Our results suggest that the plagiopatagium initially arises through novel outgrowths from the body flank that subsequently merge with the limbs to generate the wing airfoil. Our findings further suggest that this merging process, which is highly conserved across bats, occurs through modulation of the programs controlling the development of the periderm of the epidermal epithelium. Finally, our results suggest that the shape of the plagiopatagium begins to diversify in bats only after this merging has occurred.

CONCLUSIONS

This study demonstrates how focusing on the evolution of cellular processes can inform an understanding of the developmental factors shaping the evolution of novel, highly adaptive structures.

Convergent deployment of ancestral functions during the evolution of mammalian flight membranes

Lateral flight membranes, or patagia, have evolved repeatedly in diverse mammalian lineages. While little is known about patagium development, its recurrent evolution may suggest a shared molecular basis. By combining transcriptomics, developmental experiments, and mouse transgenics, we demonstrate that lateral Wnt5a expression in the marsupial sugar glider (Petaurus breviceps) promotes the differentiation of its patagium primordium. We further show that this function of Wnt5a reprises ancestral roles in skin morphogenesis predating mammalian flight and has been convergently used during patagium evolution in eutherian bats. Moreover, we find that many genes involved in limb development have been redeployed during patagium outgrowth in both the sugar glider and bat. Together, our findings reveal that deeply conserved genetic toolkits contribute to the evolutionary transition to flight in mammals.

Differential cellular proliferation underlies heterochronic generation of cranial diversity in phyllostomid bats

Background

Skull diversity in the neotropical leaf-nosed bats (Phyllostomidae) evolved through a heterochronic process called peramorphosis, with underlying causes varying by subfamily. The nectar-eating (subfamily Glossophaginae) and blood-eating (subfamily Desmondontinae) groups originate from insect-eating ancestors and generate their uniquely shaped faces and skulls by extending the ancestral ontogenetic program, appending new developmental stages and demonstrating peramorphosis by hypermorphosis. However, the fruit-eating phyllostomids (subfamilies Carollinae and Stenodermatinae) adjust their craniofacial development by speeding up certain developmental processes, displaying peramorphosis by acceleration. We hypothesized that these two forms of peramorphosis detected by our morphometric studies could be explained by differential growth and investigated cell proliferation during craniofacial morphogenesis.

Results

We obtained cranial tissues from four wild-caught bat species representing a range of facial diversity and labeled mitotic cells using immunohistochemistry. During craniofacial development, all bats display a conserved spatiotemporal distribution of proliferative cells with distinguishable zones of elevated mitosis. These areas were identified as modules by the spatial distribution analysis. Ancestral state reconstruction of proliferation rates and patterns in the facial module between species provided support, and a degree of explanation, for the developmental mechanisms underlying the two models of peramorphosis. In the long-faced species, Glossophaga soricina, whose facial shape evolved by hypermorphosis, cell proliferation rate is maintained at lower levels and for a longer period of time compared to the outgroup species Miniopterus natalensis. In both species of studied short-faced fruit bats, Carollia perspicillata and Artibeus jamaicensis, which evolved under the acceleration model, cell proliferation rate is increased compared to the outgroup.

Conclusions

This is the first study which links differential cellular proliferation and developmental modularity with heterochronic developmental changes, leading to the evolution of adaptive cranial diversity in an important group of mammals.

The Bat as a New Model of Cortical Development

The organization of the mammalian cerebral cortex shares fundamental features across species. However, while the radial thickness of grey matter varies within one order of magnitude, the tangential spread of the cortical sheet varies by orders of magnitude across species. A broader sample of model species may provide additional clues for understanding mechanisms that drive cortical expansion. Here, we introduce the bat Carollia perspicillata as a new model species. The brain of C. perspicillata is similar in size to that of mouse but has a cortical neurogenic period at least 5 times longer than mouse, and nearly as long as that of the rhesus macaque, whose brain is 100 times larger. We describe the development of laminar and regional structures, neural precursor cell identity and distribution, immune cell distribution, and a novel population of Tbr2+ cells in the caudal ganglionic eminence of the developing neocortex of C. perspicillata. Our data indicate that unique mechanisms guide bat cortical development, particularly concerning cell cycle length. The bat model provides new perspective on the evolution of developmental programs that regulate neurogenesis in mammalian cerebral cortex, and offers insight into mechanisms that contribute to tangential expansion and gyri formation in the cerebral cortex.

Transcriptomic insights into the genetic basis of mammalian limb diversity

Background

From bat wings to whale flippers, limb diversification has been crucial to the evolutionary success of mammals. We performed the first transcriptome-wide study of limb development in multiple species to explore the hypothesis that mammalian limb diversification has proceeded through the differential expression of conserved shared genes, rather than by major changes to limb patterning. Specifically, we investigated the manner in which the expression of shared genes has evolved within and among mammalian species.

Results

We assembled and compared transcriptomes of bat, mouse, opossum, and pig fore- and hind limbs at the ridge, bud, and paddle stages of development. Results suggest that gene expression patterns exhibit larger variation among species during later than earlier stages of limb development, while within species results are more mixed. Consistent with the former, results also suggest that genes expressed at later developmental stages tend to have a younger evolutionary age than genes expressed at earlier stages. A suite of key limb-patterning genes was identified as being differentially expressed among the homologous limbs of all species. However, only a small subset of shared genes is differentially expressed in the fore- and hind limbs of all examined species. Similarly, a small subset of shared genes is differentially expressed within the fore- and hind limb of a single species and among the forelimbs of different species.

Conclusions

Taken together, results of this study do not support the existence of a phylotypic period of limb development ending at chondrogenesis, but do support the hypothesis that the hierarchical nature of development translates into increasing variation among species as development progresses.

Electronic supplementary material

The online version of this article (doi:10.1186/s12862-017-0902-6) contains supplementary material, which is available to authorized users.

The Relationship between Gene Network Structure and Expression Variation among Individuals and Species

Variation among individuals is a prerequisite of evolution by natural selection. As such, identifying the origins of variation is a fundamental goal of biology. We investigated the link between gene interactions and variation in gene expression among individuals and species using the mammalian limb as a model system. We first built interaction networks for key genes regulating early (outgrowth; E9.5-11) and late (expansion and elongation; E11-13) limb development in mouse. This resulted in an Early (ESN) and Late (LSN) Stage Network. Computational perturbations of these networks suggest that the ESN is more robust. We then quantified levels of the same key genes among mouse individuals and found that they vary less at earlier limb stages and that variation in gene expression is heritable. Finally, we quantified variation in gene expression levels among four mammals with divergent limbs (bat, opossum, mouse and pig) and found that levels vary less among species at earlier limb stages. We also found that variation in gene expression levels among individuals and species are correlated for earlier and later limb development. In conclusion, results are consistent with the robustness of the ESN buffering among-individual variation in gene expression levels early in mammalian limb development, and constraining the evolution of early limb development among mammalian species.

Retinal projections in the short-tailed fruit bat, Carollia perspicillata, as studied using the axonal transport of cholera toxin B subunit: Comparison with mouse

To provide a modern description of the Chiropteran visual system, the subcortical retinal projections were studied in the short-tailed fruit bat, Carollia perspicillata, using the anterograde transport of eye-injected cholera toxin B subunit, supplemented by the silver-impregnation of anterograde degeneration following eye removal, and compared with the retinal projections of the mouse. The retinal projections were heavily labeled by the transported toxin in both species. Almost all components of the murine retinal projection are present in Carollia in varying degrees of prominence and laterality. The projections: to the superior colliculus, accessory optic nuclei, and nucleus of the optic tract are predominantly or exclusively contralateral; to the dorsal lateral geniculate nucleus and posterior pretectal nucleus are predominantly contralateral; to the ventral lateral geniculate nucleus, intergeniculate leaflet, and olivary pretectal nucleus have a substantial ipsilateral component; and to the suprachiasmatic nucleus are symmetrically bilateral. The retinal projection in Carollia is surprisingly reduced at the anterior end of the dorsal lateral geniculate and superior colliculus, suggestive of a paucity of the relevant ganglion cells in the ventrotemporal retina. In the superior colliculus, in which the superficial gray layer is very thin, the projection is patchy in places where the layer is locally absent. Except for a posteriorly located lateral terminal nucleus, the other accessory optic nuclei are diminutive in Carollia, as is the nucleus of the optic tract. In both species the cholera toxin labeled sparse groups of apparently terminating axons in numerous regions not listed above. A question of their significance is discussed.

Early oogenesis in the short-tailed fruit bat Carollia perspicillata: transient germ cell cysts and noncanonical intercellular bridges

The ovaries of early embryos (40 days post coitum/p.c.) of the bat Carollia perspicillata contain numerous germ-line cysts, which are composed of 10 to 12 sister germ cells (cystocytes). Variability in the number of cystocytes within the cyst and between the cysts (defying the Giardina rule) indicates that the mitotic divisions of the cystoblast are asynchronous in this bat species. Serial section analysis showed that the cystocytes are interconnected via intercellular bridges that are atypical, strongly elongated, short-lived, and rich in microtubule bundles and microfilaments. During slightly later stages of embryonic development (44-46 days p.c.), somatic cells penetrate the cyst, and their cytoplasmic projections separate individual oocytes. Separated oocytes surrounded by a single layer of somatic cells constitute the primordial ovarian follicles. The oocytes of C. perspicillata are similar to mouse oocytes and are asymmetric. In both species, this asymmetry is clearly recognizable in the localization of the Golgi complexes. The presence of germ-line cysts and intercellular bridges (although noncanonical) in the fetal ovaries of C. perspicillata suggest that the formation of germ-line cysts is an evolutionarily conserved phase in the development of the female gametes in a substantial part of the animal kingdom.

Endothelial cell hyperproliferation and stratification in uteroplacental blood vessels of the black mastiff bat, Molossus rufus

Placentation was studied histologically and immunocytochemically in black mastiff bats obtained at frequent intervals throughout pregnancy. These were bred in a captive colony or collected from a reproductively-synchronized wild population. During late pregnancy, the single fetus was largely sustained by a discoidal, hemochorial placenta located at the cranial end of the right uterine horn. This invariant positioning was determined by a vascular tuft that developed there both during early pregnancy and non-pregnant cycles. This provided a scaffold for early placental morphogenesis. As development proceeded, small arterioles and venules serving the tuft were converted to large uteroplacental vessels. Within the base of the placenta, these became lined by an unusual vascular epithelium composed of intermingled patches of multilayered endothelial cells and cytotrophoblast. Initially, the endothelium became multilayered by hypertrophy, proliferation, and infolding of its basal lamina. These created endothelial bilayers usually insinuated between basal laminae. The development of temporary gaps in the laminae then permitted further enlargement of the vessels and proliferation of the endothelial cells as monolayer sheets or chains. The latter were interconnected, forming a complex, stratified, cellular network associated with a prominent meshwork of basal laminae. Throughout much of pregnancy, these endothelial cells were cuboidal to columnar and possessed an abundance of basal glycoprotein granules presumably containing basal lamina precursors. The cells also expressed vimentin and frequently von Willebrand factor, but not cytokeratins or desmin. Pronounced thickening of the endothelia and amplification of their basal laminae likely evolved to greatly strengthen the walls of the uteroplacental vessels.

Ovulation, fertilization, and early embryonic development in the menstruating fruit bat, Carollia perspicillata

To characterize periovulatory events, reproductive tracts were collected at 12 hr intervals from captive-bred, short-tailed fruit bats, Carollia perspicillata, on days 1-3 post coitum and examined histologically. Most bats bred readily. Graafian follicles developed large antra and exhibited preovulatory expansion of the cumulus oophorus. Ovulation had occurred in some on the morning, and in most by the evening, of day 1. The single ovum was released as a secondary oocyte and fertilized in the oviductal ampulla. Ovulated secondary oocytes were loosely associated with their cumulus cells, which were lost around the initiation of fertilization. Supernumerary spermatozoa were occasionally noted attached to the zonae pellucidae of oviductal ova, but never within the perivitelline space. By day 2, most ova had reached the pronuclear stage and by day 3, early cleavage stages. Several lines of evidence indicate that C. perspicillata is a spontaneous ovulator with a functional luteal phase. Most newly mated females had recently formed, but regressing corpora lutea, and thickened (albeit menstrual) uteri despite having been housed with males only for brief periods (<23 days). Menstruation is usually periovulatory in this species. Furthermore, the interval between successive estrus periods in most mated females that failed to establish ongoing pregnancies at the first was 21-27 days. Menstruation involved substantial endometrial desquamation, plus associated bleeding, and generally extended to the evening of day 3, the last time point studied. In nearly all females with a recent corpus luteum (n = 24 of 25; 96%), the preovulatory or newly ruptured follicle was in the opposite ovary.

Selectivity in the transport of spermatozoa to oviductal reservoirs in the menstruating fruit bat, Carollia perspicillata

To better document the timing of ovulation and fertilization, female reproductive tracts were collected every 12 h from captive-bred fruit bats (Carollia perspicillata) on days 1–3 postcoitum and examined histologically. This also permitted observations on sperm transport, storage, and disposition. As the animals had previously been sexually segregated, most had been cycling and possessed menstrual uteri at the time of collection. Menstruation is periovulatory in this species. A widespread, headfirst orientation of spermatozoa to the uterine mucosa was observed in specimens apparently collected soon after insemination. Thereafter, however, this relationship was limited in most cases to the area around the entrance of each uterotubal junction (UTJ). A small number of spermatozoa also colonized the UTJs, which functioned as temporary sperm reservoirs on days 1–2. AlthoughC. perspicillatais monovular, no consistent differences were observed between the two oviducts in the pattern of sperm storage and release. Very few sperm were ever observed in the isthmus or ampulla (the site of fertilization). Menstrual debris (including fine particulate matter) and leukocytes present in the uterine cavity in most tracts did not gain access to the UTJ with the spermatozoa. Smooth muscle and abundant elastic fibers in the wall of the intramural UTJ, as well as receptors on its luminal epithelial cells, may play roles in the selective transport of spermatozoa to the fertilization site. While some spermatozoa are phagocytosed in the uterine lumen or by epithelial cells in the UTJ, the fate of most is probably expulsion into the vagina.

Generating timed pregnancies in fruit bats (Carollia perspicillata)

When female short-tailed fruit bats (Carollia perspicillata) are bred, they are first kept sexually segregated for some months to adapt to captivity (in the case of recently captured animals) and to ensure that none are already pregnant. In the case of captive-reared females that had previously been housed with males, 4 mo should be sufficient. In the case of wild-caught females, 7 or 8 mo may not be too long, because introduction into captivity can substantially prolong existing early pregnancies (i.e., at or earlier than the primitive streak stage). Wild-caught females are also much more prone than captive-reared females to take pregnancies into delay after captive breeding. It is important to note that females should not be housed in sexual isolation for prolonged periods, because this can eventually lead to the development of markedly hyperplastic uteri and reduced fertility. Although this is a very serious problem for zoos maintaining all-female colonies of Carollia, it is not the case with our captive colony because the adult females are regularly bred, both to replenish the colony and to inhibit the development of hyperplastic uteri. The generation of timed pregnancies is described in this protocol. For breeding purposes, a stud male with prominent testes is introduced into each cage of females. Vaginal aspirates are then checked each morning thereafter for the presence of spermatozoa.

Collection of short-tailed fruit bats (Carollia perspicillata) from the wild

Adult female short-tailed fruit bats (Carollia perspicillata) exhibit a relatively high degree of reproductive synchronization in populations sampled carefully in Central America and on Trinidad. On the basis of these studies and temporal data for pregnancies in captive-bred animals, one may be able to sample adult female Carollia from other populations and, during much of the year, predict when different embryonic stages might be prevalent. Most adult female Carollia on Trinidad seem to carry two pregnancies each year, based on the observation that more than 90% are typically pregnant when sampled around the middle of each pregnancy period. For many females, the first pregnancy appears to be established between September and early November, includes a period of post-implantational developmental delay at the primitive streak stage, and is completed in March or April. A peak in births has been observed around April 1. Most parous females then conceive again at a post-partum estrus. In captive animals, this estrus usually occurs between 3 and 6 d after parturition, but sometimes, it is several days later. Using the available temporal data on pregnancies in both captive and wild Carollia, it is possible to predict when embryos at particular stages of development are most likely to be carried by females in the wild population on Trinidad. A similar approach might be used to collect embryos from another population, as outlined in this protocol.

Genetic regulation of mammalian diversity

Mammals have evolved a variety of morphological adaptations that have allowed them to compete in their natural environments. The developmental genetic basis of this morphological diversity remains largely unknown. Bats are mammals that have the unique ability of powered flight. We have examined the molecular embryology of bats and investigated the developmental genetic basis for their highly derived limbs used for flight. Initially, we developed an embryo staging system for a model chiropteran, Carollia perspicillata, the short-tailed fruit bat that has subsequently been used for staging other bat species. Expression studies focusing on genes that regulate limb development indicate that there are similarities and differences between bats and mice. To determine the consequences of these expression differences, we have conducted an enhancer switch assay by gene targeting in mouse embryonic stem cells to create mice whose genes are regulated by bat sequences. Our studies indicate that cis-regulatory elements contribute to the morphological differences that have evolved among mammalian species.

Embryonic staging system for the Black Mastiff Bat, Molossus rufus (Molossidae), correlated with structure-function relationships in the adult

An embryonic staging system for Molossus rufus (also widely known as Molossus ater) was devised using 17 reference specimens obtained during the postimplantation period of pregnancy from wild-caught, captive-bred females. This was done in part by comparing the embryos to a developmental staging system that had been created for another, relatively unrelated bat, Carollia perspicillata (family Phyllostomidae). Particular attention was paid to the development of species-specific features, such as wing and ear morphology, and these are discussed in light of the adaptive significance of these structures in the adult. M. rufus can be maintained and bred in captivity and is relatively abundant in the wild. This embryonic staging system will facilitate further developmental studies of M. rufus, a model species for one of the largest and most successful chiropteran families, the Molossidae.

The short-tailed fruit bat Carollia perspicillata: a model for studies in reproduction and development

Carollia perspicillata has proven to be a valuable laboratory model for studies in reproduction and development. We present here an overview of the care and handling of Carollia in captivity and discuss some pertinent studies in reproductive biology. Finally, we describe various features of the genome and some of the genetic manipulations that are now possible.

Alcian blue/alizarin red staining of cartilage and bone of short-tailed fruit bat (Carollia perspicillata)

This protocol is used to stain embryo skeleton with alcian blue and alizarin red at later stages of development, when there is significant replacement of the cartilaginous early skeleton with ossified bone. It has been used with good results on the short-tailed fruit bat Carollia perspicillata and other bat species from CS 20 through neonatal stages.

Regulatory divergence modifies limb length between mammals

Natural selection acts on variation within populations, resulting in modified organ morphology, physiology, and ultimately the formation of new species. Although variation in orthologous proteins can contribute to these modifications, differences in DNA sequences regulating gene expression may be a primary source of variation. We replaced a limb-specific transcriptional enhancer of the mouse Prx1 locus with the orthologous sequence from a bat. Prx1 expression directed by the bat enhancer results in elevated transcript levels in developing forelimb bones and forelimbs that are significantly longer than controls because of endochondral bone formation alterations. Surprisingly, deletion of the mouse Prx1 limb enhancer results in normal forelimb length and Prx1 expression, revealing regulatory redundancy. These findings suggest that mutations accumulating in pre-existing noncoding regulatory sequences within a population are a source of variation for the evolution of morphological differences between species and that cis-regulatory redundancy may facilitate accumulation of such mutations.

Isolation, genomic structure and developmental expression of Fgf8 in the short-tailed fruit bat, Carollia perspicillata

Fibroblast growth factor-8 (Fgf8) encodes a secreted protein which was initially identified as the factor responsible for androgen-dependant growth of mouse mammary carcinoma cells (Tanaka et al., 1992). Fgf8 has been subsequently implicated in the patterning and growth of the gastrulating embryo, paraxial mesoderm (somites), limbs, craniofacial tissues, central nervous system and other organ systems during the development of several vertebrate model animals. Consistent with these findings, Fgf8 is expressed in a complex and dynamic pattern during vertebrate embryogenesis. Here we report the isolation and characterization of a bat (Carollia perspicillata) Fgf8 orthologue. Compared with those of other model vertebrates, Carollia Fgf8 is conserved with respect to genomic structure, sequence and many domains of developmental expression pattern. Interestingly, the expression domain marking the apical ectodermal ridge of the developing limb shows a striking difference compared to that of mouse, consistent with evolutionary diversification of bat limb morphology.

The evolutionary and developmental basis of parallel reduction in mammalian zeugopod elements

Understanding the mechanisms by which parallel evolution occurs has the potential to clarify the complex relationship between evolution and development. In this study, we examine the role of development in the repeated reduction of zeugopod elements during mammalian evolution, a functionally important phenomenon enabling locomotor specialization. By completing a morphometric study (incorporating both analyses of variation and phylogenetics) of mammalian limbs, we are able to demonstrate an evolutionary trend toward width reduction in posterior zeugopod elements of the forelimbs and hindlimbs, the ulna and fibula, respectively. We also examine the developmental basis of limb reduction in three test cases, the bat Carollia perspicillata ulna and fibula and the mouse Mus musculus fibula. The most common pattern of reduction, that of reduced element width, was achieved via the same developmental process in both bat and mouse limbs (i.e., by a slower growth rate relative to other skeletal elements), suggesting that the parallel reduction of the posterior zeugopod element within mammals could have occurred primarily by the repeated evolution of the same developmental mechanism. However, our findings also suggest that the developmental mechanisms behind the parallel evolution of other, more taxon-specific characteristics of limb reduction (i.e., element fusion) are not conserved.

Interdigital webbing retention in bat wings illustrates genetic changes underlying amniote limb diversification

Developmentally regulated programmed cell death sculpts the limbs and other embryonic organs in vertebrates. One intriguing example of species-specific differences in apoptotic extent is observed in the tissue between the digits. In chicks and mice, bone morphogenetic proteins (Bmps) trigger apoptosis of the interdigital mesenchyme, leading to freed digits, whereas in ducks, Bmp antagonists inhibit the apoptotic program, resulting in webbed feet. Here, we show that the phyllostomid bat Carollia perspicillata utilizes a distinct mechanism for maintaining interdigit tissue. We find that bat forelimb and hindlimb interdigital tissues express Bmp signaling components but that only bat hindlimbs undergo interdigital apoptosis. Strikingly, the retention of interdigital webbing in the bat forelimb is correlated with a unique pattern of Fgf8 expression in addition to the Bmp inhibitor Gremlin. By using a functional assay, we show that maintenance of interdigit tissue in the bat wing depends on the combined effects of high levels of Fgf signaling and inhibition of Bmp signaling. Our data also indicate that although there is not a conserved mechanism for maintaining interdigit tissue across amniotes, the expression in the bat forelimb interdigits of Gremlin and Fgf8 suggests that these key molecular changes contributed to the evolution of the bat wing.

Development of bat flight: morphologic and molecular evolution of bat wing digits

The earliest fossil bats resemble their modern counterparts in possessing greatly elongated digits to support the wing membrane, which is an anatomical hallmark of powered flight. To quantitatively confirm these similarities, we performed a morphometric analysis of wing bones from fossil and modern bats. We found that the lengths of the third, fourth, and fifth digits (the primary supportive elements of the wing) have remained constant relative to body size over the last 50 million years. This absence of transitional forms in the fossil record led us to look elsewhere to understand bat wing evolution. Investigating embryonic development, we found that the digits in bats (Carollia perspicillata) are initially similar in size to those of mice (Mus musculus) but that, subsequently, bat digits greatly lengthen. The developmental timing of the change in wing digit length points to a change in longitudinal cartilage growth, a process that depends on the relative proliferation and differentiation of chondrocytes. We found that bat forelimb digits exhibit relatively high rates of chondrocyte proliferation and differentiation. We show that bone morphogenetic protein 2 (Bmp2) can stimulate cartilage proliferation and differentiation and increase digit length in the bat embryonic forelimb. Also, we show that Bmp2 expression and Bmp signaling are increased in bat forelimb embryonic digits relative to mouse or bat hind limb digits. Together, our results suggest that an up-regulation of the Bmp pathway is one of the major factors in the developmental elongation of bat forelimb digits, and it is potentially a key mechanism in their evolutionary elongation as well.

Embryonic staging system for the short-tailed fruit bat,Carollia perspicillata, a model organism for the mammalian orderChiroptera, based upon timed pregnancies in captive-bred animals

There are approximately 4,800 extant species of mammals that exhibit tremendous morphological, physiological, and developmental diversity. Yet embryonic development has been studied in only a few mammalian species. Among mammals, bats are second only to rodents with regard to species number and habitat range and are the most abundant mammals in undisturbed tropical regions. Bat development, though, remains relatively unstudied. Here, we describe and illustrate a staging series of embryonic development for the short-tailed fruit bat, Carollia perspicillata, based on embryos collected at timed intervals after captive matings. As Carollia can be readily maintained and propagated in captivity and is extremely abundant in the wild, it offers an attractive choice as a chiropteran model organism. This staging system provides a framework for studying Carollia embryogenesis and should prove useful as a guide for embryological studies of other bat species and for comparisons with other orders of mammals.

Hoxd13 expression in the developing limbs of the short-tailed fruit bat, Carollia perspicillata

Bat forelimbs are highly specialized for sustained flight, providing a unique model to explore the genetic programs that regulate vertebrate limb diversity. Hoxd9-13 genes are important regulators of stylopodium, zeugopodium, and autopodium development and thus evolutionary changes in their expression profiles and biochemical activities may contribute to divergent limb morphologies in vertebrates. We have isolated the genomic region that includes Hoxd12 and Hoxd13 from Carollia perspicillata, the short-tailed fruit bat. The bat Hoxd13 gene encodes a protein that shares 95% identity with human and mouse HOXD13. The expression pattern of bat Hoxd13 mRNA during limb development was compared with that of mouse. In bat and mouse hindlimbs, the expression patterns of Hoxd13 are relatively similar. However, although the forelimb Hoxd13 expression patterns in both organisms during early limb bud stages are similar, at later stages they diverge; the anterior expression boundary of bat Hoxd13 is posterior-shifted relative to the mouse. These findings, compared with the Hoxd13 expression profiles of other vertebrates, suggest that divergent Hoxd13 expression patterns may contribute to limb morphological variation.

A morphological and immunohistochemical comparison of mammary tissues from the short-tailed fruit bat (Carollia perspicillata) and the mouse

In the present study, mammary tissues from the fruit bat (Carollia perspicillata) and mouse (Mus musculus) were compared using histological and immunohistochemical methods. Because the female bat exhibits greater reproductive similarities to humans, it might provide a useful animal model for studying mammary physiology and disease with relevance to our own species. In lactating and recently lactating specimens, bat tissue had significantly fewer adipocytes and more collagenous connective tissue compared to the mouse. The proteins Stat5a, keratin 5, Npt2b, and E-cadherin were all similarly localized in mouse and bat mammary tissues taken from lactating animals. The present study demonstrates that whereas the epithelial compartment and the presence of differentiation markers are conserved between the mouse and bat, differences exist in the stromal compartment.

An ultrastructural study of interstitial implantation in captive-bred, short-tailed fruit bats, Carollia perspicillata: trophoblastic adhesion and penetration of the uterine epithelium

Trophoblastic adhesion to, and penetration of, the uterine epithelium during implantation have been examined in captive-bred Carollia perspicillata at the light and electron microscopic levels. Initial adhesion is localized to marginal ridges immediately over the apical intercellular junctions of the epithelial cells. Penetration then involves the intrusion of trophoblastic processes between the epithelial cells and the formation of junctional complexes between the two cell types. As larger areas of adhesion develop, they still occur most often near the intercellular boundaries between the more flattened epithelial cells or on their lateral sides. This suggests that many (if not all) of these broad areas of adhesion to the trophoblast had actually formed along what had originally been part of the lateral (rather than the apical) surfaces of the epithelial cells. Portions of the apical cell surfaces further removed from their intercellular boundaries usually were not adherent to the trophoblast and still had microvilli. Upon reaching the basal lamina of the uterine epithelium, invasion of the endometrium is temporarily retarded, and trophoblast cells migrate instead between the basal lamina and the epithelial cells. This occurs extensively along both the luminal and glandular epithelia around the implantation site, but not significantly along their apical surfaces. This again indicates that adhesive interactions between the trophoblast and those apical surfaces are at most very limited in Carollia. The epithelial cells appear to be viable until separated from their basal laminae by the trophoblast. They are then phagocytized by the trophoblast. During initial penetration of the uterine luminal epithelium, the trophoblast is still entirely of the cellular variety. Syncytiotrophoblast does not begin to appear until later, when the trophoblast first comes into contact with endometrial capillaries.

Comparative studies on limb morphogenesis in mice and bats: a functional genetic approach towards a molecular understanding of diversity in organ formation

The basis of species-specific morphogenesis has been a topic of fascination and speculation for centuries. In 1828, Karl Ernst von Baer noted that at the pharyngula stage of development all vertebrate embryos are morphologically very similar. Most subsequent hypotheses have proposed that the vertebrate body plan develops by a conserved mechanism, and that divergent forms develop by differential elaboration on this basic plan. Gene cloning and expression studies have largely confirmed that the genetic pathways of embryonic patterning are highly conserved. The finding that the proteins encoded by paralogous and orthologous genes within and between species can functionally replace each another is no longer novel; in most cases this is the expected result. How, then, does divergent morphology arise between species? One hypothesis that fits well with comparative data is that divergent morphogenesis arises from genetic differences in the timing, level and pattern of orthologous gene expression during development. This idea is being tested using a functional genetic approach comparing limb morphogenesis between the mouse and bat.

Altered trophoblastic differentiation and increased trophoblastic invasiveness during delayed development in the short-tailed fruit bat, Carollia perspicillata

During pregnancy in the short-tailed fruit bat, lengthy post-implantational delays in conceptus development can occur in response to stress in captivity and seasonally in the wild. When comparisons were made between uteri carrying embryos in delay at the primitive streak stage and those growing more rapidly, many differences were noted. During delay the developing chorioallantoic placenta was generally smaller, contained a higher ratio of cytotrophoblast to syncytiotrophoblast, and had been invaded only to a limited extent on its embryonic side by mesoderm. Furthermore, much of the cytotrophoblast appeared relatively undifferentiated, randomly-oriented, linked primarily by primitive junctions, and lacked a basal lamina. In contrast, in placentae serving somite and limb-bud stage embryos, sizeable areas were noted that consisted only of more highly differentiated syncytiotrophoblast perforated by maternal vascular spaces (trophospongium). The first contact of the allantois with the developing placenta was also noted at the somite stage, and this initiated widespread invasion of the placenta by mesenchyme and allantoic blood vessels. Wherever this invasion had occurred, the cytotrophoblast between the mesenchyme and syncytiotrophoblast of the interhaemal barrier consisted of a single, polarized layer of more differentiated cells with an associated basal lamina. Eventually, all of the trophospongium was invaded by cytotrophoblast and vascularized fetal mesenchyme. These observations suggest that in addition to its germinal function, cytotrophoblast in this bat may play a major role in controlling mesenchymal invasion and angiogenesis on the embryonic side of the placenta. During the period of delay, highly invasive trophoblast is also released by the placenta. This invades the myometrium and sometimes extrauterine tissues via interstitial migration along maternal capillaries and veins.

Unusual patterns of intermediate filament protein expression by the trophoblast and decidual cells of the short-tailed fruit bat, Carollia perspicillata

In the short-tailed fruit bat (Carollia perspicillata) pregnancy can be prolonged by the occurrence of lengthy delays after implantation. This is associated with the development of highly invasive trophoblast that can penetrate the myometrium, mesenteries of the reproductive tract and the oviducts via perivascular (interstitial) routes. In order to confirm the identity and distribution of this trophoblast, intermediate filament protein immunocytochemistry was utilized. In some respects the expression of these proteins differed from what has been reported for more commonly-studied species. Cytotrophoblast in the placenta, its cytotrophoblastic shell and the highly invasive trophoblast strongly expressed cytokeratins. As pregnancy progressed, however, cytokeratin expression by syncytiotrophoblast lining much of the placental labyrinth became very weak. The cytotrophoblastic shell and highly invasive trophoblast also expressed vimentin. The highly invasive trophoblast was unusual in that it developed dendritic processes that sometimes extended out into adjacent tissues in great profusion. Decidual cells generally expressed desmin and vimentin; however, some also coexpressed cytokeratins. These observations indicate that some of the trophoblast in Carollia undergoes a significant epithelial-mesenchymal transformation. They also suggest that caution should be exercised in relying upon intermediate filament proteins as markers for cell identification purposes in exotic species, or when the patterns of protein expression by fetal and maternal cells might be altered in pathological or experimental situations.

Advanced oviductal development, transport to the preferred implantation site, and attachment of the blastocyst in captive-bred, short-tailed fruit bats, Carollia perspicillata

The final stages of embryonic development in the oviduct, transport of the embryo to the uterus, and the initial stages of implantation have been examined in captive-bred Carollia perspicillata at the light and electron microscopic levels. Development progressed to the expanded, zona pellucida-free, blastocyst stage in the oviduct. The abundance of microvilli on the exterior of the trophoblast varied with the degree of blastocyst expansion and cell shape, and may function in part as a membrane reservoir. Cells of the blastocyst also typically contained many lipid droplets and prominent areas of cytoplasm occupied by finely granular material (probably glycogen) instead of organelles. In most females, closure of the uterine lumen occurred prior to, or around the time of, transport of the blastocyst to the usual implantation site and appeared to play a role in preventing transport of the blastocyst too far distally in the uterus. This was associated with increased endometrial edema, particularly in the fundic region of the simplex uterus, and the extravasation of many erythrocytes into the endometrial stroma. Both of these changes began while the blastocyst was still being held in the oviduct and became pronounced during implantation. Engulfment of these erythrocytes by processes of the endometrial stromal cells and their phagocytosis by macrophages was also observed. Implantation was usually initiated within narrow tubular segments, lined by endometrium, that were located between the end of each oviduct and the main cavity of the uterus, or from immediately adjacent areas of the main cavity. During the early stages of implantation, the blastocyst was clasped by the endometrium at the implantation site, and this was associated with extensive interdigitation of the microvilli of the trophoblast and adjacent uterine epithelial cells. Initial adhesion of the trophoblast, which was still cellular rather than synctial, occurred over the apical intercellular junctions of the uterine epithelial cells.

Relationships between orientation of the blastocyst during implantation, position of the chorioallantoic placenta, and vascularization of the uterus in the noctilionoid bats Carollia perspicillata and Noctilio sp

In most eutherian mammals, the inner cell mass (ICM) of the blastocyst assumes an almost constantly specific orientation to the uterus at the time of implantation, and this is usually correlated with subsequent positioning of the fetal membranes and chorioallantoic placenta. Although these relationships tend to be conserved between closely related species, this is not the case in the noctilionoid bats. In Carollia perspicillata, which has a simplex uterus, the ICM of the single blastocyst becomes oriented towards the uterotubal junction on the side of ovulation, and the discoidal placenta develops in a fundic position. In Noctilio sp., which have partially bicornuate uteri, the ICM becomes oriented instead towards an endometrial ridge that runs along the antimesometrial to lateral side of the gravid horn. As development proceeds, however, the blastocyst rotates almost 180 degrees, and the discoidal placenta eventually assumes a mesometrial to lateral position. In these species, implantation and subsequent development of the discoidal placenta clearly seem to be targeting major maternal vessels supplying the uterus, rather than exhibiting a consistent pattern of orientation relative to its mesenteric attachments. This permits their chorioallantoic placentae to develop a dual maternal blood supply that may be essential for the development of relatively large, precocious infants.

Co-expression of cytokeratins and vimentin by highly invasive trophoblast in the white-winged vampire bat, Diaemus youngi, and the black mastiff bat, Molossus ater, with observations on intermediate filament proteins in the decidua and intraplacental trophoblast

Histological and immunocytochemical studies of gravid reproductive tracts obtained from the white-winged vampire bat (Diaemus youngi) and the black mastiff bat (Molossus ater) have established that both species develop unusually invasive trophoblast. This is released by the developing discoidal haemochorial placenta, expresses both cytokeratins and vimentin, and invades the myometrium and adjacent tissues (including the ovaries) via interstitial migration within the walls of maternal blood vessels. Hence, this trophoblast is noteworthy for the extent to which it undergoes an epithelial-mesenchymal transformation. In Molossus, it originates from the cytotrophoblastic shell running along the base of the placenta, is mononuclear, and preferentially invades maternal arterial vessels serving the discoidal placenta. This trophoblast may have a role in dilatation of these vessels when the discoidal placenta becomes functional. In Diaemus, the highly invasive trophoblast appears to originate instead from a layer of syncytiotrophoblast on the periphery of the placenta is multinucleated, and vigorously invades both arterial and venous vessels. During late pregnancy, it becomes extensively branched and sends attenuated processes around many of the myometrial smooth muscle fibres. In view of its distribution, this trophoblast could have important influences upon myometrial contractility and the function of blood vessels serving the gravid tract. Other aspects of intermediate filament expression in the uteri and placentae of these bats are also noteworthy. Many of the decidual giant cells in Molossus co-express cytokeratins and vimentin, while the syncytiotrophoblast lining the placental labyrinth in Diaemus late in pregnancy expresses little cytokeratin.

The interhaemal barrier in the chorioallantoic placenta of the greater mustache bat, Pteronotus parnellii, with observations on amplification of its intrasyncytial lamina

Chorioallantoic placentae were obtained from a reproductively synchronized wild population of greater mustache bats for ultrastructural and immunocytochemical examination. The single discoidal placenta was always located in a lateral to mesometrial position on the right side of the uterus, which in the non-pregnant state is partially bicornuate. The placenta was labyrinthine and haemodichorial in advanced pregnancy. The interhaemal barrier included syncytiotrophoblast that lined the maternal vascular spaces and an underlying, continuous layer of cytotrophoblast. The barrier also contained a discontinuous extracellular layer, the intrasyncytial lamina, that was usually completely surrounded by syncytiotrophoblast. Two lines of evidence suggest that the intrasyncytial lamina may serve, in part, to strengthen the interhaemal barrier: (1) the lamina became significantly thicker as the maternal vascular channels became larger; and (2) the syncytiotrophoblast in the walls of the smallest vascular tubules expressed only very limited amounts of cytokeratins, normally a major component of the cytoskeleton of cells of epithelial origin, while the cytokeratin-rich cytotrophoblast was often highly attenuated. It was not uncommon to see gaps between the ectoplasmic processes of syncytiotrophoblast which exposed portions of the intrasyncytial lamina to the maternal vascular space. As platelet adhesion was never observed in such areas, the intrasyncytial lamina may be augmented in part by material that is non-thrombogenic.

Giant, accordioned sperm acrosomes of the greater bulldog bat, Noctilio leporinus

Sperm of the greater bulldog bat Noctilio leporinus display an architecture that is totally unique among mammalian spermatozoa. The sperm head of Noctilio is extraordinarily large and flat and lies eccentrically with respect to the sperm tail. The major portion of the atypically large acrosome lies anterior to the nucleus and is shaped into a dozen accordionlike folds that run parallel to the long axis of the sperm. The ridge of each fold is shaped into approximately 60 minute, evenly spaced rises that extend along the entire length of the fold. We speculate that acrosome ridges may serve to strengthen the sperm head during transport.

Formation of reticulated endoderm, Reichert's membrane, and amniogenesis in blastocysts of captive-bred, short-tailed fruit bats, Carollia perspicillata

BACKGROUND AND METHODS

As part of an effort to develop the short-tailed fruit bat (Carollia perspicillata) as a new animal model for the study of interstitial implantation and trophoblast-uterine interactions, early embryogenesis was examined histologically and ultrastructurally in captive-bred females at different intervals after the first appearance of spermatozoa in daily vaginal smears (day 1 postcoitum [p.c.]).

RESULTS

In most of the early uterine embryos examined on days 16-18 p.c., much of the endoderm appeared as a reticulated meshwork; however, a unilocular yolk sac was formed prior to the development of any mesoderm. Early blastocysts of Carollia were also unusual in that endoderm surrounded much of the inner cell mass (ICM), Reichert's membrane continued over the dorsal side of the ICM, and basal laminalike material was observed around many of the endoderm and epiblast cells. A primordial amniotic cavity was formed between days 19 and 26 p.c. by cavitation. The first mesoderm appeared between days 23 and 26 p.c., concommitant with the development of an embryonic shield.

CONCLUSION

The unusual reticulated appearance of early endoderm in Carollia, which is reminiscent of that seen in early human blastocysts, may be attributable to constraints imposed on growth of the blastocyst by the site and mode of implantation, temporary retardation of trophoblastic invasion by the basal laminae of endometrial epithelial elements, and endodermal proliferation in anticipation of rapid yolk sac expansion. Reichert's membrane appears to play an important role in this species in tethering the ICM and embryonic shield to the developing placenta prior to the formation of significant amounts of mesoderm.

Delayed development in the short-tailed fruit bat, Carollia perspicillata

Pregnancy was studied in short-tailed fruit bats, Carollia perspicillata, both maintained in a captive breeding colony and collected from a reproductively synchronized wild population on the island of Trinidad. Gestation periods for captive females that successfully reared their young varied as follows: mated at a regular oestrus during their first year in captivity (105-178 days) (mean +/- SD: 145 +/- 19 days); mated at a postpartum oestrus during their first year in captivity (110-158 days) (133 +/- 16 days); mated during their second year in captivity (113-169 days) (127 +/- 12 days); females born and mated in captivity (113-159 days) (119 +/- 9 days). Most females in the last group had gestation periods of 113-119 days; this may represent the normal (nondelayed) gestation period for the species. Histological studies established that most of the observed variation in duration of gestation was due to delays occurring after the completion of implantation. It seems likely that stress, rather than age, was responsible for the prolongation of pregnancy in some animals, because this occurred less frequently in both younger and older females. There may be stressful situations in the wild (for example, lack of sufficient food or roosting sites) in which the ability to delay pregnancies would be of considerable adaptive value. Evidence was obtained that under some circumstances Carollia can extend gestation even further. Many wildcaught females successfully gave birth at 160-229 days after being isolated from breeding males in captivity. These had been captured at the time of year when, based upon subsequent histological studies of field collected specimens, most adult females should have been in early pregnancy. The field studies have also provided evidence that females in the wild population exhibit a seasonal prolongation of pregnancy.

Unusual aspects of inner cell mass formation, endoderm differentiation, Reichert's membrane development, and amniogenesis in the lesser bulldog bat, Noctilio albiventris

BACKGROUND AND METHODS

The early embryogenesis of the lesser bulldog bat, Noctilio albiventris (family Noctilionidae), was examined histologically in 59 pregnant females collected from a reproductively synchronized population in Colombia.

RESULTS

Early blastocysts of Noctilio are unusual in lacking a typical inner cell mass. Instead, cells inside of the trophoblast are dispersed for a period as a monolayer. A typical inner cell mass (ICM) only forms and becomes properly oriented after the initiation of implantation. Several features of Reichert's membrane in this species are also noteworthy: it develops between the ICM and trophoblast and between the parietal endoderm and trophoblast; it becomes linked to a meshwork of basal laminalike material that extends into the ICM; and it appears to be continuous, or fused, with prominent basal laminae that develop within the cytotrophoblastic villi that radiate throughout the preplacenta. Amniogenesis occurs by cavitation and converts the ICM into a hollow epiblastic vesicle. Gastrulation commences before this vesicle exhibits obvious differentiation into an embryonic shield and amniotic ectoderm.

CONCLUSIONS

Because development and proper orientation of a typical ICM in Noctilio occur after the initiation of implantation, these may involve the migration of cells on the interior of the blastocyst and/or an unusual method of early endoderm differentiation. The possibility exists that epiblast, endoderm, and cytotrophoblast may all contribute to the secretion of Reichert's membrane in this bat. Although the early embryogenesis of Noctilio exhibits many similarities to that in phyllostomid bats, substantial differences also exist between these closely related species.

Improved procedures for maintaining and breeding the short-tailed fruit bat (Carollia perspicillata) in a laboratory setting

Improved methods have been developed for maintaining and breeding the neotropical short-tailed fruit bat, Carollia perspicillata, in an easily-reproduced, laboratory setting. Bipartite, all-metal cages have been designed which permit efficient, non-injurious handling of the animals. Also, a fruit-based liquid diet, which is simple and inexpensive to prepare, has been formulated from readily-available canned and powdered components. When efforts were initiated to breed wild-caught animals at 3-6 months after capture, this progressed slowly in most (but not all) cages. Breeding took place more rapidly when the bats had been in captivity for 11-24 months. Most females (122/144, 84.7%) then bred within 30 days of the introduction of a stud male, and the overall conception rate was 94.3% in those females which bred. This pattern of breeding activity, evidence that this bat is a spontaneous ovulator, and observed intervals between successive breeding periods (apparently representing much or all of non-pregnant cycles) in some individuals suggest that the cycle length in many of these bats probably falls between 20 and 30 days. Analysis of the breeding data also showed that certain patterns of breeding activity were frequently associated with a failure of females to establish ongoing pregnancies. Most laboratory-bred females (69/81) that were permitted to carry their pregnancies to term did so and successfully reared their young. The laboratory-reared young have generally exhibited excellent body condition, and many have proved to be fertile. The bats were found to exhibit a post-partum oestrus, which in nearly all cases (35/36) was fertile. The interval between the discovery of a new baby and the detection of a sperm-positive vaginal aspirate varied between 3-10 days, but most frequently was 3-6 days. Females which aborted non-term fetuses also had a post-partum oestrus with similar timing. These findings indicate that the short-tailed fruit bat, which is widely abundant in the lowland tropics of the New World, can be successfully maintained under controlled conditions, in cages of modest dimensions, for research purposes.

Pregnancy in chiroptera

As with other aspects of their biology, bats exhibit considerable diversity in their reproductive characteristics. While early embryonic development in many species generally adheres to the pattern shown by more commonly studied eutherian mammals, others demonstrate advanced development in the oviduct (to the zona-free blastocyst stage) and a prolonged tubal journey of the embryo in a temporal sense. In some bats, the process for transporting ova through the oviducts can distinguish between living embryos and the remnants of dead ova. Implantation in bats is unusually diverse with respect to localization of the nidation sites (which in most species seem to be predetermined), orientation of the inner cell mass during implantation, depth of implantation, and trophoblastic invasiveness. Some species which can be bred in captivity would appear to be promising experimental models for studies of implantation because both this process and their endometrial biology in general exhibit many similarities to those of humans. Depending upon the species, bats develop either endotheliochorial or hemochorial placentas. Studies on black mastiff bats (Molossus ater) indicate that maternal endometrial endothelial cells probably play a very important morphogenetic role during placental development in this species. Similarities in placental structure suggest that this may also be the case in a number of other eutherian mammals.