Exploring the origin of a unique mutant allele in twin-tail goldfish using CRISPR/Cas9 mutants.

Artificial selection has been widely applied to genetically fix rare phenotypic features in ornamental domesticated animals. For many of these animals, the mutated loci and alleles underlying rare phenotypes are known. However, few studies have explored whether these rare genetic mutations might have been fixed due to competition among related mutated alleles or if the fixation occurred due to contingent stochastic events. Here, we performed genetic crossing with twin-tail ornamental goldfish and CRISPR/Cas9-mutated goldfish to investigate why only a single mutated allele-chdS with a E127X stop codon (also called chdAE127X)-gives rise to the twin-tail phenotype in the modern domesticated goldfish population. Two closely related chdS mutants were generated with CRISPR/Cas9 and compared with the E127X allele in F2 and F3 generations. Both of the CRISPR/Cas9-generated alleles were equivalent to the E127X allele in terms of penetrance/expressivity of the twin-tail phenotype and viability of carriers. These findings indicate that multiple truncating mutations could have produced viable twin-tail goldfish. Therefore, the absence of polymorphic alleles for the twin-tail phenotype in modern goldfish likely stems from stochastic elimination or a lack of competing alleles in the common ancestor. Our study is the first experimental comparison of a singular domestication-derived allele with CRISPR/Cas9-generated alleles to understand how genetic fixation of a unique genotype and phenotype may have occurred. Thus, our work may provide a conceptual framework for future investigations of rare evolutionary events in domesticated animals.

Genomic screening of fish-specific genes in gnathostomes and their functions in fin development.

In this study, we comprehensively searched for fish-specific genes in gnathostomes that contribute to development of the fin, a fish-specific trait. Many previous reports suggested that animal group-specific genes are often important for group-specific traits. Clarifying the roles of fish-specific genes in fin development of gnathostomes, for example, can help elucidate the mechanisms underlying the formation of this trait. We first identified 91 fish-specific genes in gnathostomes by comparing the gene repertoire in 16 fish and 35 tetrapod species. RNA-seq analysis narrowed down the 91 candidates to 33 genes that were expressed in the developing pectoral fin. We analyzed the functions of approximately half of the candidate genes by loss-of-function analysis in zebrafish. We found that some of the fish-specific and fin development-related genes, including fgf24 and and1/and2, play roles in fin development. In particular, the newly identified fish-specific gene qkia is expressed in the developing fin muscle and contributes to muscle morphogenesis in the pectoral fin as well as body trunk. These results indicate that the strategy of identifying animal group-specific genes is functional and useful. The methods applied here could be used in future studies to identify trait-associated genes in other animal groups.

The dwarf neon rainbowfish Melanotaenia praecox, a small spiny-rayed fish with potential as a new Acanthomorpha model fish: II. Establishment of a microinjection procedure for genetic engineering.

BACKGROUND: Rainbowfish is a clade of colorful freshwater fish. Melanotaenia praecox is a small rainbowfish species with biological characteristics that make it potentially useful as an experimental model species. We anticipate that M. praecox could become a new model used in various fields, such as ecology, evolution, and developmental biology. However, few previous studies have described experimental set-ups needed to understand the molecular and genetic mechanisms within this species. RESULTS: We describe detailed procedures for genetic engineering in the rainbowfish M. praecox. By using these procedures, we successfully demonstrated CRISPR/Cas-mediated knockout and Tol2 transposon-mediated transgenesis in this species. Regarding the CRISPR/Cas system, we disrupted the tyrosinase gene and then showed that injected embryos lacked pigmentation over much of their body. We also demonstrated that a Tol2 construct, including a GFP gene driven by a ubiquitous promoter, was efficiently integrated into the genome of M. praecox embryos. CONCLUSIONS: The establishment of procedures for genetic engineering in M. praecox enables investigation of the genetic mechanisms behind a broad range of biological phenomena in this species. Thus, we suggest that M. praecox can be used as a new model species in various experimental biology fields.

The dwarf neon rainbowfish Melanotaenia praecox, a small spiny-rayed fish with potential as a new Acanthomorpha model fish: I. Fin ray ontogeny and postembryonic staging.

BACKGROUND: Fish fins with highly variable color patterns and morphologies have many functions. In Actinopterygii, the free parts of fins are supported by "soft rays" and "spiny rays." Spiny rays have various functions and are extremely modified in some species, but they are lacking in popular model fish such as zebrafish and medaka. Additionally, some model fish with spiny rays are difficult to maintain in ordinary laboratory systems. RESULTS: Characteristics of the small, spiny-rayed rainbowfish Melanotaenia praecox render it useful as an experimental model species. Neither fish age nor body size correlate well with fin development during postembryonic development in this species. A four-stage developmental classification is proposed that is based on fin ray development. CONCLUSIONS: Melanotaenia praecox is an ideal species to rear in laboratories for developmental studies. Our classification allows for postembryonic staging of this species independent of individual age and body size. Development of each fin ray may be synchronized with dorsal fin development. We discuss the differences in mechanisms regulating soft, spiny, and procurrent ray development.

Morphological evolution and diversity of pectoral fin skeletons in teleosts.

The Teleostei class has the most species of the fishes. Members of this group have pectoral fins, enabling refined movements in the water. Although teleosts live in a diverse set of environments, the skeletal pattern of pectoral fins in teleosts is considered to show little morphological variability. Here, in order to elucidate variations in pectoral fin skeletons and to identify their evolutionary processes, we compared the pectoral fin skeletons from 27 species of teleosts. We identified several variations and a diversity of pectoral fin skeletal patterns within some teleost groups. Taken together with previous reports on teleost skeletons, our findings reveal that in the course of teleost evolution, there are a mixture of conserved and non-conserved components in the pectoral fin skeletons of teleosts, and that teleosts may have experienced the variation and conservation of the number and shape of the proximal radials, the loss of the mesocoracoid, and the change in the distal radial-fin ray relationship.

Pleiotropic functions of chordin gene causing drastic morphological changes in ornamental goldfish.

Breeders and fanciers have established many peculiar morphological phenotypes in ornamental goldfish. Among them, the twin-tail and dorsal-finless phenotypes have particularly intrigued early and recent researchers, as equivalent morphologies are extremely rare in nature. These two mutated phenotypes appeared almost simultaneously within a short time frame and were fixed in several strains. However, little is known about how these two different mutations could have co-occurred during such a short time period. Here, we demonstrate that the chordin gene, a key factor in dorsal-ventral patterning, is responsible not only for the twin-tail phenotype but also for the dorsal-finless phenotype. Our F2 backcrossing and functional analyses revealed that the penetrance/expressivity of the dorsal-finless phenotype can be suppressed by the wild-type allele of chdS. Based on these findings, we propose that chdSwt may have masked the expression of the dorsal-finless phenotype, acting as a capacitor buffering gene to allow accumulation of genetic mutations. Once this gene lost its original function in the twin-tail goldfish lineages, the dorsal-finless phenotype could be highly expressed. Thus, this study experimentally demonstrates that the rapid genetic fixation of morphological mutations during a short domestication time period may be related to the robustness of embryonic developmental mechanisms.

Developmental independence of median fins from the larval fin fold revises their evolutionary origin.

The median fins of modern fish that show discrete forms (dorsal, anal, and caudal fins) are derived from a continuous fold-like structure, both in ontogeny and phylogeny. The median fin fold (MFF) hypothesis assumes that the median fins evolved by reducing some positions in the continuous fin fold of basal chordates, based on the classical morphological observation of developmental reduction in the larval fin folds of living fish. However, the developmental processes of median fins are still unclear at the cellular and molecular levels. Here, we describe the transition from the larval fin fold into the median fins in zebrafish at the cellular and molecular developmental level. We demonstrate that reduction does not play a role in the emergence of the dorsal fin primordium. Instead, the reduction occurs along with body growth after primordium formation, rather than through actively scrapping the non-fin forming region by inducing cell death. We also report that the emergence of specific mesenchymal cells and their proliferation promote dorsal fin primordium formation. Based on these results, we propose a revised hypothesis for median fin evolution in which the acquisition of de novo developmental mechanisms is a crucial evolutionary component of the discrete forms of median fins.

Thalidomide and its metabolite 5-hydroxythalidomide induce teratogenicity via the cereblon neosubstrate PLZF.

Thalidomide causes teratogenic effects by inducing protein degradation via cereblon (CRBN)-containing ubiquitin ligase and modification of its substrate specificity. Human P450 cytochromes convert thalidomide into two monohydroxylated metabolites that are considered to contribute to thalidomide effects, through mechanisms that remain unclear. Here, we report that promyelocytic leukaemia zinc finger (PLZF)/ZBTB16 is a CRBN target protein whose degradation is involved in thalidomide- and 5-hydroxythalidomide-induced teratogenicity. Using a human transcription factor protein array produced in a wheat cell-free protein synthesis system, PLZF was identified as a thalidomide-dependent CRBN substrate. PLZF is degraded by the ubiquitin ligase CRL4CRBN in complex with thalidomide, its derivatives or 5-hydroxythalidomide in a manner dependent on the conserved first and third zinc finger domains of PLZF. Surprisingly, thalidomide and 5-hydroxythalidomide confer distinctly different substrate specificities to mouse and chicken CRBN, and both compounds cause teratogenic phenotypes in chicken embryos. Consistently, knockdown of Plzf induces short bone formation in chicken limbs. Most importantly, degradation of PLZF protein, but not of the known thalidomide-dependent CRBN substrate SALL4, was induced by thalidomide or 5-hydroxythalidomide treatment in chicken embryos. Furthermore, PLZF overexpression partially rescued the thalidomide-induced phenotypes. Our findings implicate PLZF as an important thalidomide-induced CRBN neosubstrate involved in thalidomide teratogenicity.

Zebrafish can regenerate endoskeleton in larval pectoral fin but the regenerative ability declines.

We show for the first time endoskeletal regeneration in the developing pectoral fin of zebrafish. The developing pectoral fin contains an aggregation plate of differentiated chondrocytes (endochondral disc; primordium for endoskeletal components, proximal radials). The endochondral disc can be regenerated after amputation in the middle of the disc. The regenerated disc sufficiently forms endoskeletal patterns. Early in the process of regenerating the endochondral disc, epithelium with apical ectodermal ridge (AER) marker expression rapidly covers the amputation plane, and mesenchymal cells start to actively proliferate. Taken together with re-expression of a blastema marker gene, msxb, and other developmental genes, it is likely that regeneration of the endochondral disc recaptures fin development as epimorphic limb regeneration does. The ability of endoskeletal regeneration declines during larval growth, and adult zebrafish eventually lose the ability to regenerate endoskeletal components such that amputated endoskeletons become enlarged. Endoskeletal regeneration in the zebrafish pectoral fin will serve as a new model system for successful appendage regeneration in mammals.

Regrowth of zebrafish caudal fin regeneration is determined by the amputated length.

Fish have a high ability to regenerate fins, including the caudal fin. After caudal fin amputation, original bi-lobed morphology is reconstructed during its rapid regrowth. It is still controversial whether positional memory in the blastema cells regulates reconstruction of fin morphology as in amphibian limb regeneration, in which limb blastema cells located at the same proximal-distal level have the same positional identity. We investigated growth period and growth rate in zebrafish caudal fin regeneration. We found that both the growth period and growth rate differed for fin rays that were amputated at the same proximal-distal level, indicating that it takes different periods of time for fin rays to restore their original lengths after straight amputation. We also show that more proximal amputation takes longer period to reconstruct the original morphology/size than more distal amputation. Statistical analysis suggested that both the growth period/rate are determined by amputated length (depth) regardless of the fin ray identity along dorsal-ventral axis. In addition, we suggest the possibility that the structural/physical condition such as width of the fin ray at the amputation site (niche at the stump) may determine the growth period/rate.

Insights regarding skin regeneration in non-amniote vertebrates: Skin regeneration without scar formation and potential step-up to a higher level of regeneration.

Skin wounds are among the most common injuries in animals and humans. Vertebrate skin is composed of an epidermis and dermis. After a deep skin injury in mammals, the wound heals, but the dermis cannot regenerate. Instead, collagenous scar tissue forms to fill the gap in the dermis, but the scar does not function like the dermis and often causes disfiguration. In contrast, in non-amniote vertebrates, including fish and amphibians, the dermis and skin derivatives are regenerated after a deep skin injury, without a recognizable scar remaining. Furthermore, skin regeneration can be compared with a higher level of organ regeneration represented by limb regeneration in these non-amniotes, as fish, anuran amphibians (frogs and toads), and urodele amphibians (newts and salamanders) have a high capacity for organ regeneration. Comparative studies of skin regeneration together with limb or other organ regeneration could reveal how skin regeneration is stepped up to a higher level of regeneration. The long history of regenerative biology research has revealed that fish, anurans, and urodeles have their own strengths as models for regeneration studies, and excellent model organisms of these non-amniote vertebrates that are suitable for molecular genetic studies are now available. Here, we summarize the advantages of fish, anurans, and urodeles for skin regeneration studies with special reference to three model organisms: zebrafish (Danio rerio), African clawed frog (Xenopus laevis), and Iberian ribbed newt (Pleurodele waltl). All three of these animals quickly cover skin wounds with the epidermis (wound epidermis formation) and regenerate the dermis and skin derivatives as adults. The availability of whole genome sequences, transgenesis, and genome editing with these models enables cell lineage tracing and the use of human disease models in skin regeneration phenomena, for example. Zebrafish present particular advantages in genetics research (e.g., human disease model and Cre-loxP system). Amphibians (X. laevis and P. waltl) have a skin structure (keratinized epidermis) common with humans, and skin regeneration in these animals can be stepped up to limb regeneration, a higher level of regeneration.

Pattern of fin rays along the antero-posterior axis based on their connection to distal radials.

BACKGROUND: Teleost paired fins are composed of two endoskeletal domains, proximal and distal radials, and an exoskeletal domain, the fin ray. The zebrafish pectoral fin displays elaborately patterned radials along the anteroposterior (AP) axis. Radials are considered homologous to tetrapod limb skeletons, and their patterning mechanisms in embryonic development are similar to those of limb development. Nevertheless, the pattern along the AP axis in fin rays has not been well described in the zebrafish pectoral fin, although several recent reports have revealed that fin ray development shares some cellular and genetic properties with fin/limb endoskeleton development. Thus, fin ray morphogenesis may involve developmental mechanisms for AP patterning in the fin/limb endoskeleton, and may have a specific pattern along the AP axis. RESULTS: We conducted detailed morphological observations on fin rays and their connection to distal radials by comparing intra- and inter-strain zebrafish specimens. Although the number of fin rays varied, pectoral fin rays could be categorized into three domains along the AP axis, according to the connection between the fin rays and distal radials; additionally, the number of fin rays varied in the posterior part of the three domains. This result was confirmed by observation of the morphogenesis process of fin rays and distal radials, which showed altered localization of distal radials in the middle domain. We also evaluated the expression pattern of lhx genes, which have AP patterning activity in limb development, in fin rays and during distal radial development and found these genes to be expressed during morphogenesis in both fin rays and distal radials. CONCLUSION: The fin ray and its connection to the endoskeleton are patterned along the AP axis, and the pattern along the AP axis in the fin ray and the radial connection is constructed by the developmental mechanism related to AP patterning in the limb/fin bud. Our results indicate the possibility that the developmental mechanisms of fin rays and their connection are comparable to those of the distal element of the limb skeleton.

Embryonic and postembryonic development of the ornamental twin-tail goldfish.

BACKGROUND: Twin-tail ornamental goldfish have "bifurcated median fins," a peculiar morphology known to be caused by a mutation in the chdA gene. However, several ambiguities regarding the development of the phenotype remain due to a paucity of detailed observations covering the entire developmental timeframe. RESULTS: Here, we report a detailed comparative description of embryonic and postembryonic development for two representative twin-tail ornamental goldfish strains and single-tail common goldfish. Our observations reveal a polymorphic developmental process for bifurcated median fins; disrupted axial skeletal development at early larval stages; and modified bilateral location of the pelvic fin. CONCLUSIONS: Variations in development of bifurcated median fins and disrupted axial skeletal patterns reflect how artificial selection for adult morphological features influenced molecular developmental mechanisms during the domestication of twin-tail ornamental goldfish. The polymorphic appearance of bifurcated median fins also implies that, unlike previously proposed hypotheses, the development of these structures is controlled by molecular mechanisms independent of those acting on the pelvic fin. Our present findings will facilitate further study of how modifications of preexisting developmental systems may contribute to novel morphological features. Developmental Dynamics 248:251-283, 2019. © 2019 The Authors. Developmental Dynamics published by Wiley Periodicals, Inc. on behalf of American Association of Anatomists.

A novel allele of the goldfish chdB gene: Functional evaluation and evolutionary considerations.

The twin tail of ornamental goldfish is known to be caused by a nonsense mutation in one chordin paralogue gene. Our previous molecular studies in goldfish revealed that the ancestral chordin gene was duplicated, creating the chdA and chdB genes, and the subsequent introduction of a stop codon allele in the chdA gene ( chdA E127X ) caused the twin-tail morphology. The chdA E127X allele was positively selected by breeders, and the allele was genetically fixed in the ornamental twin-tail goldfish population. However, little is known about the evolutionary history of the chdB paralogue, begging the question: are there the functionally distinct alleles at the chdB locus, and if so, how did they evolve? To address these questions, we conducted molecular sequencing of the chdB gene from five different goldfish strains and discovered two alleles at the chdB gene locus; the two alleles are designated chdB 1 and chdB 2 . The chdB 1 allele is the major allele and was found in all investigated goldfish strains, whereas the chdB 2 allele is minor, having only been found in one twin-tail strain. Genetic analyses further suggested that these two alleles are functionally different with regard to survivability ( chdB 1 > chdB 2 ). These results led us to presume that in contrast to the chdA locus, the chdB locus has tended to be eliminated from the population. We also discuss how the chdB 2 allele was retained in the goldfish population, despite its disadvantageous function. This study provides empirical evidence of the long-term retention of a disadvantageous allele under domesticated conditions.

Morphogenetic mechanism of the acquisition of the dinosaur-type acetabulum.

Understanding morphological evolution in dinosaurs from a mechanistic viewpoint requires the elucidation of the morphogenesis that gave rise to derived dinosaurian traits, such as the perforated acetabulum. In the current study, we used embryos of extant animals with ancestral- and dinosaur-type acetabula, namely, geckos and turtles (with unperforated acetabulum), and birds (with perforated acetabulum). We performed comparative and experimental analyses, focusing on inter-tissue interaction during embryogenesis, and found that the avian perforated acetabulum develops via a secondary loss of cartilaginous tissue in the acetabular region. This cartilage loss might be mediated by inter-tissue interaction with the hip interzone, a mesenchymal tissue that exists in the embryonic joint structure. Furthermore, the data indicate that avian pelvic anlagen is more susceptible to paracrine molecules, e.g. Wnt ligand, secreted by the hip interzone than 'reptilian' anlagen. We hypothesize that during the emergence of dinosaurs, the pelvic anlagen became susceptible to the Wnt ligand, which led to the loss of the cartilaginous tissue and to the perforation in the acetabular region. Thus, the current evolutionary-developmental biology study deepens our understanding of morphological evolution in dinosaurs and provides it with a novel perspective.

An alternative evolutionary pathway for the twin-tail goldfish via szl gene mutation.

The twin-tail of ornamental goldfish provides unique evolutionary evidence that the highly conserved midline localization of axial skeleton components can be changed by artificial selection. This morphological change is known to be caused by a nonsense mutation in one of the recently duplicated chordin genes, which are key players in dorsal-ventral (DV) patterning. Since all of the multiple twin-tail ornamental goldfish strains share the same mutation, it is reasonable to presume that this mutation occurred only once in domesticated goldfish. However, zebrafish with mutated szl gene (another DV patterning-related gene) also exhibit twin-tail morphology and higher viability than dino/chordin-mutant zebrafish. This observation raises the question of whether the szl gene mutation could also reproduce the twin-tail morphology in goldfish. Here we show that goldfish have at least two subfunctionalized szl genes, designated szlA and szlB, and depletion of these genes in single-fin goldfish was able to reproduce the bifurcated caudal fin found in twin-tail ornamental goldfish. Interestingly, several phenotypes were observed in szlA-depleted fish, while low expressivity of the twin-tail phenotype was observed in szlB-depleted goldfish. Thus, even though szl gene mutations may produce twin-tail goldfish, these szl gene mutations might not be favorable for selection in domestic breeding. These results highlight the uniqueness and rarity of mutations that are able to cause large-scale morphological changes, such as a bifurcated axial skeleton, with high viability and expressivity in natural and domesticated populations.

Flight feather development: its early specialization during embryogenesis.

BACKGROUND: Flight feathers, a type of feather that is unique to extant/extinct birds and some non-avian dinosaurs, are the most evolutionally advanced type of feather. In general, feather types are formed in the second or later generation of feathers at the first and following molting, and the first molting begins at around two weeks post hatching in chicken. However, it has been stated in some previous reports that the first molting from the natal down feathers to the flight feathers is much earlier than that for other feather types, suggesting that flight feather formation starts as an embryonic event. The aim of this study was to determine the inception of flight feather morphogenesis and to identify embryological processes specific to flight feathers in contrast to those of down feathers. RESULTS: We found that the second generation of feather that shows a flight feather-type arrangement has already started developing by chick embryonic day 18, deep in the skin of the flight feather-forming region. This was confirmed by shh gene expression that shows barb pattern, and the expression pattern revealed that the second generation of feather development in the flight feather-forming region seems to start by embryonic day 14. The first stage at which we detected a specific morphology of the feather bud in the flight feather-forming region was embryonic day 11, when internal invagination of the feather bud starts, while the external morphology of the feather bud is radial down-type. CONCLUSION: The morphogenesis for the flight feather, the most advanced type of feather, has been drastically modified from the beginning of feather morphogenesis, suggesting that early modification of the embryonic morphogenetic process may have played a crucial role in the morphological evolution of this key innovation. Co-optation of molecular cues for axial morphogenesis in limb skeletal development may be able to modify morphogenesis of the feather bud, giving rise to flight feather-specific morphogenesis of traits.

Activation of the hypothalamic feeding centre upon visual prey detection.

The visual system plays a major role in food/prey recognition in diurnal animals, and food intake is regulated by the hypothalamus. However, whether and how visual information about prey is conveyed to the hypothalamic feeding centre is largely unknown. Here we perform real-time imaging of neuronal activity in freely behaving or constrained zebrafish larvae and demonstrate that prey or prey-like visual stimuli activate the hypothalamic feeding centre. Furthermore, we identify prey detector neurons in the pretectal area that project to the hypothalamic feeding centre. Ablation of the pretectum completely abolishes prey capture behaviour and neurotoxin expression in the hypothalamic area also reduces feeding. Taken together, these results suggest that the pretecto-hypothalamic pathway plays a crucial role in conveying visual information to the feeding centre. Thus, this pathway possibly converts visual food detection into feeding motivation in zebrafish.

Transient inflammatory response mediated by interleukin-1ß is required for proper regeneration in zebrafish fin fold.

Cellular responses to injury are crucial for complete tissue regeneration, but their underlying processes remain incompletely elucidated. We have previously reported that myeloid-defective zebrafish mutants display apoptosis of regenerative cells during fin fold regeneration. Here, we found that the apoptosis phenotype is induced by prolonged expression of interleukin 1 beta (il1b). Myeloid cells are considered to be the principal source of Il1b, but we show that epithelial cells express il1b in response to tissue injury and initiate the inflammatory response, and that its resolution by macrophages is necessary for survival of regenerative cells. We further show that Il1b plays an essential role in normal fin fold regeneration by regulating expression of regeneration-induced genes. Our study reveals that proper levels of Il1b signaling and tissue inflammation, which are tuned by macrophages, play a crucial role in tissue regeneration.

Upstream regulation for initiation of restricted Shh expression in the chick limb bud.

BACKGROUND: The organizing center, which serves as a morphogen source, has crucial functions in morphogenesis in animal development. The center is necessarily located in a certain restricted area in the morphogenetic field, and there are several ways in which an organizing center can be restricted. The organizing center for limb morphogenesis, the ZPA (zone of polarizing activity), specifically expresses the Shh gene and is restricted to the posterior region of the developing limb bud. RESULTS: The pre-pattern along the limb anteroposterior axis, provided by anterior Gli3 expression and posterior Hand2 expression, seems insufficient for the initiation of Shh expression restricted to a narrow, small spot in the posterior limb field. Comparison of the spatiotemporal patterns of gene expression between Shh and some candidate genes (Fgf8, Hoxd10, Hoxd11, Tbx2, and Alx4) upstream of Shh expression suggested that a combination of these genes' expression provides the restricted initiation of Shh expression. CONCLUSIONS: Taken together with results of functional assays, we propose a model in which positive and negative transcriptional regulatory networks accumulate their functions in the intersection area of their expression regions to provide a restricted spot for the ZPA, the source of morphogen, Shh. Developmental Dynamics 246:417-430, 2017. © 2017 Wiley Periodicals, Inc.