Lecture 13

The Gnathostome Limb

Early development

Connective tissue for the limb arises from the somatic lateral plate mesoderm (or hypomere). Remember, the hypomere has two parts: somatic mesoderm between the coelom and ectoderm (essentially the body wall) and the splanchnic mesoderm between the coelom and the endodermal gut. Muscles of the limbs are derived from the myotomes (part of the somites or epimeres).

Fins and limbs (Fig. 9.2)

1. Fin
1. basal pterygiophores (basals)
2. radial pterygiophores (radials)
3. dermal fin rays
   1. lepidotrichia - dermal bony rods (nerual crest derived???)
   2. ceratotrichia - keratinized dermal rods
2. limbs (see also Fig. 9.22)
   1. stylopodium
   2. zeugopodium
   3. autopodium
      1. medopodials
      2. metapodials
      3. digits

Girdles

1. pectoral
   1. fish (Fig. 9.12)
   2. tetrapod
2. pelvic
   1. fish
   2. tetrapod

topics

1. Origin of limbs
2. Origin of tetrapod limbs
3. homologies between tetrapod and actinopterygian limb
4. trends
5. digital homology between dinosaurs and birds

Origin of limbs (Kardong, pp. 319-323)

• modified gill arch model
  • This is a pectoral fin first model
• lateral fin fold model
  • This is a pectoral fin - pelvic fin appear at same time model

  Paleontological data:

  Many early, agnathan fishes had dermally supported keels and fin folds, some ventrolaterally as in the fin fold model, but these, as stated, are dermal and do not contain an endochondral skeleton. While some of the pectoral girdle is dermal, the basals, radials (and homologues in tetrapods) are endochondral. This suggests these lateral fin folds in early fishes, when present, are not homologous to today's fins.

  Paleontological evidence shows that some early fishes had lateral appendages in the region immediately posterior to the gills. Definite pectoral appendages with endochondral support (radials) do occur in some early vertebrates without pelvic appendages but pelvic appendages do not occur without the pectoral fins (at least in early vertebrates).

  Developmental data

  the developmental evidence of a continuous lateral fin fold is scanty although the fins develop from ventrolateral thickenings in sharks (Fig. 9.6.e).

  pharyngeal (gill) arch connective tissue is neural crest derived while connective tissue of the limb is from the lateral plate mesoderm.

  the pectoral girdle has both dermal and endochondral elements while the pharyngeal arches are all endochondral (except the dermal bones covering the mandibular arch).

  Conclusion

  Neither the fin fold or modified gill models are satisfactory. Michael Coates has suggested a modification of the modified gill arch model. That is, that the endochondral part of the pectoral girdle (scapulocoracoid) is an extrabranchial cartilage (the cartilages lateral to the gills) and that while it is not a modified gill arch, the gill arches and the pectoral girdle are both modifications of a serially homologous developmental program. The radials and derivatives are neomorphic in this scenario.

  This is all reviewed in the papers, which can be downloaded here.

  1. Coates, M.I. (1994). The origin of vertebrate limbs. Develop. Supplement, 169-180.
  2. Coates, M.I. (2003). The Evolution of Paired Fins. Theory in Bioscience 122: 266-287.

Origin of tetrapod limb

The typical chondrycthyian or actinopterygian fin is very different from the tetrapod limb. From what did the tetrapod limb evolve from and how is the organization homologous to that in actinopterygians. Look at the pictures of the chondrichthyian pectoral fin (Fig. 9.10) or the actinopterygian fin (Fig. 9.1 or Fig. 9.12). These look pretty different from the organization of the tetrapod limb (Fig. 9.22). The shark and actinopterygian have a row of proximal elements (basals) from anterior to posterior, each supporting one or more distal elements (radials). The tetrapod has a single basal element (the stylopodium), which seems to support two distal (zeugopodial) elements, each of which support one or two more distal elements.

Modern sarcopterygian fishes do not seem to help:

 

The coelacanth has a short midline axis of elements while the lungishes have a long midline axis. But if we look at the fossil sarcopterygian fishes that are closest to tetrapods, we see an organization in the fin that looks remarkably like that in the tetrapod limb.

The top panel is the pectoral fin (left) and pelvic fin (right) of Eusthenopteron. The bottom panel is the living coelacanth. Focussing on the Eustehnopteron, there are certainly many lipidotrichia but look at the support for these. There is a single stylopodial element, two zeugopodial elementsand a series of bifurcations and segmentations from the postaxial side autopodium.

To understand this better look at Fig. 9.7 and understand that the tetrapod limb elements develop from proximal to distal and start as a condensation of connective tissue which then later chondrifies and then later ossifies. As the connective tissue condensation grows with limb growth it can segment into a proximal segment and a distal segment the distal part can bifurcate into two distal segments.

developmental processes

1. mesenchymal condensation is the proximal to distal formation of discrete blocks (or condensations) of mesenchymal cells that form the basic pattern of the limb elements (the bones)
2. chondrification is the differentiation of chondrocytes (cartilage producing cells) within the condensations and the production of cartilage tissue.
3. ossification occurs when osteoblasts in the cartilage precursor begin to produce the bony tissue.

some new terms

• preaxial - the side toward the thumb/big toe (1st digit).
• postaxial - the side of the limb toward the pinky/little toe (5th digit)

Condensation occurs proximal to distal along an axis that goes through the single proximal element (humerus/femur) through the postaxial element (ulna/fibula) into the postaxial mesopodials (ulnare/fibulare), through the 4th metapodial, and finally through the 4th digit (exception: the axis in some or all salamanders is through the 2nd digit). Except for the 5th digit, all bifurcations (such as radius or tibia) are preaxial. The 5th digit seems to come from nowhere.

Important note 1: the digits are generally formed from 4th to 3rd to 2nd to 1st with the fifth developing late.

Important note 2: In animals that do not develop specific digits because of an evoluitonary loss, the trend is to lose them in the reverse order that they develop. Hence, the trend is to lose (evolutionarily) the 1st and 5th first, then 2nd, etc. Other than the possible exception of dinosaurs and birds, you don't lose the fourth digit without also having lost the first.

Some of this is reviewed in Shubin, N. H. and Alberch, P. (1986). A morphogenetic approach to the origin and basic organization of the tetrapod limb. In Evolutionary Biology, vol. 20 (ed. M. Hecht and B. Schaeffer), pp. 319-387.

So there is an axis of development that occurs posteriorly in the tetrapod limb. Now look at the pectoral fin of the Mississippi paddlefish, Polyodon, below (the left panel of (b) (this is an actinopterygian that retains many primitive features). There is a distinctly large posterior basal with a bifurcation distal to it. This posterior basal is the metapterygium and it and the endochondral bones distal to it form the metapterygial axis. It looks like the tetrapod limb is homologous to the metapterygial axis in Polyodon. So the tetrapod limb seems to have expanded the metapterygial axis and lost the pro- and meso- pterygium as well as the dermal lepidotrichia.

 

 

 

Trends (see text)

1. digit reduction in many tetrapods
2. limb loss in many tetrapods
3. frogs
4. flying vertebrates
5. swimming vertebrates
6. cursorial vertebrates

Digital homologies between birds and dinosaurs

Terms:

• Dx : short for the xth digit (ossified structure)
• Cx: short for the condensation of the xth digit (but see frame-shift hypothesis below).

Evidence

There is good evidence from paleontological record of increasing minimization and eventually, loss, of digits four and five. The primitive phalangeal formula for tetrapods, or at least amniotes, is 2-3-4-5-3 (these are the number of digits from D1 - D5). Birds have three digits with highly reduced phalangeal formula. However, Archaeopteryx, which everyone agrees is a bird, has three digits with 2, 3, and 4 phalanges. Most paleontologists would argues that birds are dinosaurs and that the Archaeopteryx formula is 2-3-4-0-0 just like the derived therapod dinosaurs.

Most developmental biologists believe that birds are not dinosaurs (but what are they) and that the formula for Archaeopteryx is 0-2-3-4-0. In other words, yes both derived therapods and Archaeopteryx have three digits with 2, 3, and 4 phalanges but these are D1-3 in dinos and D2-4 in birds. The reason that developmental biologists argue that archaeopteryx is D2-4 is that the primary axis goes through D4 and this is the most conservative digit (least likely to be lost). They therefore assign the initial condensation in the developing hand to be D4, and the two pre-axial condenstions would then be D3 and D2. in other words, they argue it would be unlikely for the axis to shift from D4 to D3. But what about 3 fingered dinosaurs? Did the axis shift for them and if it did, why couldn't birds have inherited this from therapods. Or, maybe the identification of the 3 fingers in the derived therapods is wrong and they really have D2-4 like in birds. Developmental biologists give no explanation for dinosaur digits and it would seem that either way, the developmental evidence is not inconsistant with a bird-dino sister group. More importantly, how do we form a consensus between the paleo data and the developmental data?

Click on image to get big picture

Prelude to Frame-Shift hypothesis:

The key: early morphogenesis and character identity may be decoupled

• e.g. Drosophila, there are segmentation genes that make segments and homeotic genes that identify segments
• In birds and mammals, corresponding regions (neck, trunk) express same combination of Hox genes regardless of which somites supply the mesenchyme.

Constraints to be explained:

• morphogenetic constraint: cannot lose d4 without loss of d1 (as in horses which have retained d3 only).
• functional constraint: need thumb which is necessary for grasping (common prey processing in dinos)

how to lose D1 without losing D1?

Frame-shift hypothesis of Wagner

Essentially, Wagner and Gauthier (Wagner, G. P. and Gauthier, J. A. (1999). 1,2,3 = 2,3,4: A solution to the problem of the homology of the digits in the avian hand. Proc. Natl. Acad. Sci. USA 96, 5111-5116) argued that there are condensation genes and digit identity genes and that condensations in both advanced therapods and aves are C2-C4 but that the genes coding for digit identity shifted to repattern the C2-C4 condensations into D1-3 -like digits in advanced therapods and birds. This maintained the grasping first digit (thumb) but it is made from C2 instead of C1.