Birds have dinosaur wings: The molecular evidence

Within developmental biology, the digits of the wing of birds are considered on embryological grounds to be digits 2, 3 and 4. In contrast, within paleontology, wing digits are named 1, 2, 3 as a result of phylogenetic analysis of fossil taxa indicating that birds descended from theropod dinosaurs that had lost digits 4 and 5. It has been argued that the development of the wing does not support the conclusion that birds are theropods, and that birds must have descended from ancestors that had lost digits 1 and 5. Here we use highly conserved gene expression patterns in the developing limbs of mouse and chicken, including the chicken talpid(2)mutant and polydactylous Silkie breed (Silkie mutant), to aid the assessment of digital identity in the wing. Digit 1 in developing limbs does not express Hoxd12, but expresses Hoxd13. All other digits express both Hoxd12and Hoxd13. We found this signature expression pattern identifies the anteriormost digit of the wing as digit 1, in accordance with the hypothesis these digits are 1, 2 and 3, as in theropod dinosaurs. Our evidence contradicts the long-standing argument that the development of the wing does not support the hypothesis that birds are living dinosaurs.     

Tbx18 and boundary formation in chick somite and wing development

The chicken Tbx gene, Tbx18, is expressed in lateral plate mesoderm, limb, and developing somites. Here we show that Tbx18 is expressed transiently in axial mesenchyme during somite segmentation. We present evidence from overexpression and transplantation experiments that Tbx18 controls fissure formation in the late stages of somite maturation. In presumptive wing lateral plate mesoderm, ectopic Tbx18 expression leads to anterior extension of the wing bud. These results suggest that Tbx18 is involved in producing mesodermal boundaries, generating in paraxial mesoderm morphological boundaries between somites and in lateral plate mesoderm a wing- or non-wing-forming boundary.     

Suppression of polydactyly of the Gli3 mutant (extra toes) by deltaEF1 homozygous mutation

Digit patterning is established through multiple genetic interactions. Delta-crystallin enhancer/E2-box factor (deltaEF1) is a zinc finger and homeodomain containing repressor protein, and is expressed in the posterior half of the forelimb bud and in the entire hindlimb bud during the early stage of limb development. The 6EF1-deficient mutant mice display various skeletal abnormalities, among which inferior ossification and abnormal patterning of autopodial bones are similar to those observed in Hox and Gli gene mutants. Gli3 mutant mice, extra toes (Xt), exhibit pre-axial polydactyly losing the identity of digit I. It is demonstrated here that deltaEF1null(lacZ) homozygosity suppressed formation of the extra digit, uniquely of the hindlimb, in both Gli3XtJ heterozygous and homozygous mutants, but with no restoration of digit I identity. In Gli3XtJ mutants, the Hoxd13 expression domain was expanded more dramatically in homozygotes. In Gli3XtJ;deltaEF1null(lacZ) double homozygous mutants, Hoxd13 expression once expanded in Gli3XtJ homozygous mutant was reduced, more conspicuously in the hindlimbs, which may account for hindlimb-restricted suppression of formation of the extra digit. The data suggest the possibility that the extent of Hoxd13 expression along the distal margin of the limb bud is determinative in defining the digit number.     

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.     

Hox genes, digit identities and the theropod/bird transition

Vargas and Fallon (2005. J Exp Zool (Mol Dev Evol) 304B:86-90) propose that Hox gene expression patterns indicate that the most anterior digit in bird wings is homologous to digit 1 rather than to digit 2 in other amniotes. This interpretation is based on the presence of Hoxd13 expression in combination with the absence of Hoxd12 expression in the second digit condensation from which this digit develops (the first condensation is transiently present). This is a pattern that is similar to that in the developing digit 1 of the chicken foot and the mouse hand and foot. They have tested this new hypothesis by analysing Hoxd12 and Hoxd13 expression patterns in two polydactylous chicken mutants, Silkie and talpid2. They conclude that the data support the notion that the most anterior remaining digit of the bird wing is homologous to digit 1 in other amniotes either in a standard phylogenetic sense, or alternatively in a (limited) developmental sense in agreement with the Frameshift Hypothesis of Wagner and Gautier (1999, i.e., that the developmental pathway is homologous to the one that leads to a digit 1 identity in other amniotes, although it occurs in the second instead of the first digit condensation). We argue that the Hoxd12 and Hoxd13 expression patterns found for these and other limb mutants do not allow distinguishing between the hypothesis of Vargas and Fallon (2005. J Exp Zool (Mol Dev Evol) 304B:86-90) and the alternative one, i.e., the most anterior digit in bird wings is homologous to digit 2 in other amniotes, in a phylogenetic or developmental sense. Therefore, at the moment the data on limb mutants does not present a challenge to the hypothesis, based on other developmental data (Holmgren, 1955. Acta Zool 36:243-328; Hinchliffe, 1984. In: Hecht M, Ostrom JH, Viohl G, Wellnhofer P, editors. The beginnings of birds. Eichst?tt: Freunde des Jura-Museum. p 141-147; Burke and Feduccia, 1997. Science 278:666-668; Kundrát et al., 2002. J Exp Zool (Mol Dev Evol) 294B:151-159; Larsson and Wagner, 2002. J Exp Zool (Mol Dev Evol) 294B:146-151; Feduccia and Nowicki, 2002. Naturwissenschaften 89:391-393), that the digits of bird wings are homologous to digits 2,3,4 in amniotes. We recommend further testing of the hypothesis by comparing Hoxd expression patterns in different taxa.     

Developmental genetic basis for the evolution of pelvic fin loss in the pufferfish Takifugu rubripes

Paired appendages were a key developmental innovation among vertebrates and they eventually evolved into limbs. Ancient developmental control systems for paired fins and limbs are broadly conserved among gnathostome vertebrates. Some lineages including whales, some salamanders, snakes, and many ray-fin fish, independently lost the pectoral, pelvic, or both appendages over evolutionary time. When different taxa independently evolve similar developmental morphologies, do they use the same molecular genetic mechanisms? To determine the developmental genetic basis for the evolution of pelvis loss in the pufferfish Takifugu rubripes (fugu), we isolated fugu orthologs of genes thought to be essential for limb development in tetrapods, including limb positioning (Hoxc6, Hoxd9), limb bud initiation (Pitx1, Tbx4, Tbx5), and limb bud outgrowth (Shh, Fgf10), and studied their expression patterns during fugu development. Results showed that bud outgrowth and initiation fail to occur in fugu, and that pelvis loss is associated with altered expression of Hoxd9a, which we show to be a marker for pelvic fin position in three-spine stickleback Gasterosteus aculeatus. These results rule out changes in appendage outgrowth and initiation genes as the earliest developmental defect in pufferfish pelvic fin loss and suggest that altered Hoxd9a expression in the lateral mesoderm may account for pelvis loss in fugu. This mechanism appears to be different from the mechanism for pelvic loss in stickleback, showing that different taxa can evolve similar phenotypes by different mechanisms.     

The chick oligozeugodactyly (ozd) mutant lacks sonic hedgehog function in the limb

We have analyzed a new limb mutant in the chicken that we name oligozeugodactyly (ozd). The limbs of this mutant have a longitudinal postaxial defect, lacking the posterior element in the zeugopod (ulna/fibula) and all digits except digit 1 in the leg. Classical recombination experiments show that the limb mesoderm is the defective tissue layer in ozd limb buds. Molecular analysis revealed that the ozd limbs develop in the absence of Shh expression, while all other organs express Shh and develop normally. Neither Ptc1 nor Gli1 are detectable in mutant limb buds. However, Bmp2 and dHAND are expressed in the posterior wing and leg bud mesoderm, although at lower levels than in normal embryos. Activation of Hoxd11-13 occurs normally in ozd limbs but progressively declines with time. Phase III of expression is more affected than phase II, and expression is more severely affected in the more 5' genes. Interestingly, re-expression of Hoxd13 occurs at late stages in the distal mesoderm of ozd leg buds, correlating with formation of digit 1. Fgf8 and Fgf4 expression are initiated normally in the mutant AER but their expression is progressively downregulated in the anterior AER. Recombinant Shh protein or ZPA grafts restore normal pattern to ozd limbs; however, retinoic acid fails to induce Shh in ozd limb mesoderm. We conclude that Shh function is required for limb development distal to the elbow/knee joints, similar to the Shh(-/-) mouse. Accordingly we classify the limb skeletal elements as Shh dependent or independent, with the ulna/fibula and digits other than digit 1 in the leg being Shh dependent. Finally we propose that the ozd mutation is most likely a defect in a regulatory element that controls limb-specific expression of Shh.     

Posterior apical ectodermal ridge removal in the chick wing bud triggers a series of events resulting in defective anterior pattern formation

The ability of the anterior apical ectodermal ridge to promote outgrowth in the chick wing bud when disconnected from posterior apical ridge was examined by rotating the posterior portion of the stage-19/20 to stage-21 wing bud around its anteroposterior axis. This permitted contact between the anterior and posterior mesoderm, without removing wing bud tissue. In a small but significant number of cases (10/54), anterior structures (digit 2) formed spatially isolated from posterior structures (digits 3 and 4). Thus, continuity with posterior ridge is not a prerequisite for anterior-ridge function in the wing bud. Nevertheless, posterior-ridge removal does result in anterior limb truncation. To investigate events leading to anterior truncation, we examined cell death patterns in the wing bud following posterior-ridge removal. We observed an abnormal area of necrosis along the posterior border of the wing bud at 6-12 h following posterior-ridge removal. This was followed by necrosis in the distal, anterior mesoderm at 48 h postoperatively and subsequent anterior truncation. Clearly, healthy posterior limb bud mesoderm is needed for anterior limb bud survival and development. We propose that anterior truncation is the direct result of anterior mesodermal cell death and that this may not be related to positional specification of anterior cells. In our view, cell death of anterior mesoderm, after posterior mesoderm removal, should not be used as evidence for a role in position specification by the polarizing zone during the limb bud stages of development. We suggest that the posterior mesoderm that maintains the anterior mesoderm need not be restricted to the mapped polarizing zone, but is more extensively distributed in the limb bud.     

How the pterosaur got its wings

Throughout the evolutionary history of life, only three vertebrate lineages took to the air by acquiring a body plan suitable for powered flight: birds, bats, and pterosaurs. Because pterosaurs were the earliest vertebrate lineage capable of powered flight and included the largest volant animal in the history of the earth, understanding how they evolved their flight apparatus, the wing, is an important issue in evolutionary biology. Herein, I speculate on the potential basis of pterosaur wing evolution using recent advances in the developmental biology of flying and non‐flying vertebrates. The most significant morphological features of pterosaur wings are: (i) a disproportionately elongated fourth finger, and (ii) a wing membrane called the brachiopatagium, which stretches from the posterior surface of the arm and elongated fourth finger to the anterior surface of the leg. At limb‐forming stages of pterosaur embryos, the zone of polarizing activity (ZPA) cells, from which the fourth finger eventually differentiates, could up‐regulate, restrict, and prolong expression of 5′‐located Homeobox D (Hoxd) genes (e.g. Hoxd11, Hoxd12, and Hoxd13) around the ZPA through pterosaur‐specific exploitation of sonic hedgehog (SHH) signalling. 5′Hoxd genes could then influence downstream bone morphogenetic protein (BMP) signalling to facilitate chondrocyte proliferation in long bones. Potential expression of Fgf10 and Tbx3 in the primordium of the brachiopatagium formed posterior to the forelimb bud might also facilitate elongation of the phalanges of the fourth finger. To establish the flight‐adapted musculoskeletal morphology shared by all volant vertebrates, pterosaurs probably underwent regulatory changes in the expression of genes controlling forelimb and pectoral girdle musculoskeletal development (e.g. Tbx5), as well as certain changes in the mode of cell–cell interactions between muscular and connective tissues in the early phase of their evolution. Developmental data now accumulating for extant vertebrate taxa could be helpful in understanding the cellular and molecular mechanisms of body‐plan evolution in extinct vertebrates as well as extant vertebrates with unique morphology whose embryonic materials are hard to obtain.     

GLI3 constrains digit number by controlling both progenitor proliferation and BMP-dependent exit to chondrogenesis

Inactivation of Gli3, a key component of Hedgehog signaling in vertebrates, results in formation of additional digits (polydactyly) during limb bud development. The analysis of mouse embryos constitutively lacking Gli3 has revealed the essential GLI3 functions in specifying the anteroposterior (AP) limb axis and digit identities. We conditionally inactivated Gli3 during mouse hand plate development, which uncoupled the resulting preaxial polydactyly from known GLI3 functions in establishing AP and digit identities. Our analysis revealed that GLI3 directly restricts the expression of regulators of the G(1)-S cell-cycle transition such as Cdk6 and constrains S phase entry of digit progenitors in the anterior hand plate. Furthermore, GLI3 promotes the exit of proliferating progenitors toward BMP-dependent chondrogenic differentiation by spatiotemporally restricting and terminating the expression of the BMP antagonist Gremlin1. Thus, Gli3 is a negative regulator of the proliferative expansion of digit progenitors and acts as a gatekeeper for the exit to chondrogenic differentiation.     

Level-specific role of paraxial mesoderm in regulation of Tbx5/Tbx4 expression and limb initiation

Tetrapod limbs, forelimbs and hindlimbs, emerge as limb buds during development from appropriate positions along the rostro-caudal axis of the main body. In this study, tissue interactions by which rostro-caudal level-specific limb initiation is established were analyzed. The limb bud originates from the lateral plate located laterally to the paraxial mesoderm, and we obtained evidence that level-specific tissue interactions between the paraxial mesoderm and the lateral plate mesoderm are important for the determination of the limb-type-specific gene expression and limb outgrowth. When the wing-level paraxial mesoderm was transplanted into the presumptive leg region, the wing-level paraxial mesoderm upregulated the expression of Tbx5, a wing marker gene, and down regulated the expression of Tbx4 and Pitx1, leg marker genes, in the leg-level lateral plate. The wing-level paraxial mesoderm relocated into the leg level also inhibited outgrowth of the hindlimb bud and down regulated Fgf10 and Fgf8 expression, demonstrating that the wing-level paraxial mesoderm cannot substitute for the function of the leg-level paraxial mesoderm in initiation and outgrowth of the hindlimb. The paraxial mesoderm taken from the neck- and flank-level regions also had effects on Tbx5/Tbx4 expression with different efficiencies. These findings suggest that the paraxial mesoderm has level-specific abilities along the rostro-caudal axis in the limb-type-specific mechanism for limb initiation.