Evolutionary developmental biology of the mammalian middle ear: using virtual
reconstruction to integrate development and biomechanics

A project undertaken at the, School of Biological Sciences, The University of Queensland and supervised by Vera Weisbecker

The evolution of the mammalian secondary jaw joint  and mammalian middle ear (MME) is arguably the most stunning evolutionary transformation in the 320-million year journey from large reptile-like mammalian ancestors to the mammals of today. Over this time period, the jaw joint bones and associated rear lower jaw bones of the synapsid lineage (of which mammals are the only survivors) changed dramatically. Starting off as a fully functional jaw joint, these bones first assumed a dual role of hearing and chewing and ultimately specialized entirely to be part of the minute chain of ear ossicles that are the prerequisite for the mammalian superior hearing abilities.

The original synapsid jaw joint, which is essentially the same as that of all reptiles, consisted of the articular (turqoise in Fig.1) bone in the lower jaw and quadrate (dark purple in Fig.1) in the upper jaw, which turned into the hammer and anvil of the mammalian middle ear. The mammalian ear drum is also suspended on an  former jaw bone, the angular (pink in Fig.1). To replace the primary jaw joint, mammals have evolved a new, secondary jaw joint between the dentary (green in Fig.1) and squamosal bone (red in Fig.1).

The formation of the MME represents a developmental as well as biomechanical feat, which has included the miniaturization of the rear lower jaw bones to form fine-tuned hearing ossicles  instead of well-braced chewing implements. The driving forces behind this transformation, and the biomechanical adaptations that made it possible, have been mainly researched based on evidence from fossils. However, one of the striking properties of MME formation is that it is recapitulated during mammalian development and particularly accessible in developing marsupials. This project used a novel combination of virtual reconstructions and computed-tomography based large-scale developmental sampling of marsupials and monotremes to :

  1. Test the hypothesis that the growing mammalian brain played a deciding role in the detachment of the ear ossicles from the lower jaw both in development and evolution. This was thought to be observable through a developmental decrease in size, and movement away from the jaw , in middle ear ossicles(Fig. 2). However, our work showed that the middle ear bones of growing marsupials and monotremes do not move at all (with exception of the echidna, see below) and that there is no size decrease of the ossicles relative to the remainder of the skull. Rather, we see a co-occurrence between middle ear detachment and molar eruption, suggesting that perhaps a change in jaw function might be behind the developmental detachment in the middle ear. We are currently following this up through biomechanical (finite element) analysis of the impact that molar growth has on the developing lower jaw (Fig. 3)
  2. Look for evidence that the final detachment of middle ear bones occurred independently in monotremes and marsupials; we are in the final stages of developing a probability-based scenario of these processes using bayesian inference. While working on this, we found that some of the character states from the fossil record can be artificially created in virtual reconstruction by reducing the amount of outer bone layers (Fig. 4). This is leading us to suspect that some fossils with particular character states on mammalian ear evolution may need to be re-interpreted.
  3. Use virtual biomechanical simulations to understand how intermediate forms of the mammalian middle ear could work efficiently. This as allowed us to “warp” broken fossils to similar, extant skulls and thus enables us to analyze a complete skull’s biomechanics, rather than having to rely on a fragmented fossil (Fig. 5). This work is ongoing as part of a Discovery Project grant that was funded based on this Hermon Slade grant.

Ramirez-Chaves H.E., Wroe S.W., Selwood L., Hinds L.A., Leigh C., Koyabu D., Kardjilov N., and Weisbecker V. (2016). Mammalian development does not recapitulate suspected key transformations in the evolutionary detachment of the mammalian middle ear. Proc. R. Soc. B 283: 20152606. http://dx.doi.org/10.1098/rspb.2015.2606.

Figure captions:

Fig.1: Tracing the evolution of the mammalian middle ear using virtual 3D reconstructions of the skull bones. The middle ear of mammals consists of miniature versions of the bones that form the lower jaw joint in reptiles like Agamas (top left; scan source: www.Digimorph.org). Fossils of early mammalian relatives as old as 245 million years, like Thrinaxodon liorhinus (top right; scan source: www.digimorph.org) show that this jaw joint became increasingly specialized for hearing, but still retained its chewing function. A similar transitional arrangement is still visible in the early development of marsupials like newborn rat kangaroos (bottom left) before the bones assume their exclusive hearing function in grown animals (an adult rat kangaroo is depicted at bottom right). Green, Dentary; Pink, Angular/Ectotympanic; Turqoise, Articular/Hammer; Purple, Quadrate/Anvil.

Fig.2: To understand the Middle Ear development before and after  detachment from the lower jaw, we use the proxy of Meckel’s cartilage disappearance (bracketed arrows); this developmental series is from a rat kangaroo.

Fig. 3: If the pressure of an erupting and growing molar is simulated to act on the lower jaw of a marsupial (a brushtail possum joey, in this instance), it is possible to understand whether molar eruption pushes away the middle ear from the jaw. “hot” colours indicate areas of high stress on the bone, where developmental deformation is expected to take place; “cold” colours indicate no stresses.

Fig.4: Virtually changing the thickness of bone in the lower jaw of a mammal (top- before, bottom-after) makes it look very different (see arrow), and quite similar to some fossils that are thought to still have an attached middle ear.

Fig.5: Turning a broken fossil in to a complete specimen using virtual repair has allowed us to produce a complete skull (in this case of the Thrinaxodon skull from Fig. 1) for biomechanical analysis. Dark patches are repaired areas where no fossil information was available.

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.