Pharyngula

Bone is a sophisticated substance, much more than just a rock-like mineral in an interesting shape. It's a living tissue, invested with cells dedicated to continually remodeling the mineral matrix. That matrix is also an intricate material, threaded with fibers of a protein, type II collagen, that give it a much greater toughness—it's like fiberglass, a relatively brittle substance given resilience and strength with tough threads woven within it. Bone is also significantly linked to cartilage, both in development and evolution, with earlier forms having a cartilaginous skeleton that is replaced by bone. In us vertebrates, cartilage also contains threads of collagen running through it.

These three elements—collagen, cartilage, and bone—present an interesting evolutionary puzzle. Collagen is common to the matrices of both vertebrate cartilage and bone, yet the most primitive fishes, the jawless lampreys and hagfish, have been reported to lack that particular form of collagen, suggesting that the collagen fibers are a derived innovation in chordate history. New work, though, has shown that there's a mistaken assumption in there: lampreys do have type II collagen! This discovery clarifies our understanding of the evolution of the chordate skeleton.

Collagen in general is an ancient protein, thought to have evolved about 800 million years ago. It's such a useful substance, though, that variants have evolved repeatedly, and there are about 27 different types of collage in at least a dozen different classes, and it's the most common protein in your body, used in all of your connective tissues. One particular type, type II collagen, is an essential part of the matrix of your bones and cartilages.

Type II collagen was thought not to be present in lampreys, however, which led to the assumption that the lamprey skeleton evolved independently in that lineage. That led to some awkward problems in evolution, though. For instance, here's an anaspid ostracoderm from the late Devonian, a very primitive jawless fish that lived around 370 million years ago. This is an animal that is thought to be much more closely related to us gnathostomes (jawed vertebrates) than lampreys or hagfish, so you might expect it to have a skeleton more like ours.

a, Specimen 2a from the Miguasha Museum of Natural History in Quebec, Canada. Scale bar, 10 mm. b, Reconstruction of the specimen in a. The specimen shows tarry imprints of cartilage that became calcified later in life.

But no, this specimen is well enough preserved that the structure of its skeleton could be seen, and it looks remarkably similar to calcified lamprey cartilage. There isn't any way to tell if collagen was present in this animal, so it doesn't answer the question of the order of evolution of the matrix components: collagen, cartilage, or bone. It implies that a lamprey-like organization could have been an early stage in evolution, but if lampreys possess a different cellular method of building a skeleton, one that lacks collagen, this organization in Euphanerops would be only an example of convergence.

a,Euphanerops displays large, rounded spaces for inclusion of chondrocytes, which are lined with calicified cartilage. b, Lamprey chondrocyte spaces in groups of two or four ('cell nests'), as in Euphanerops. c, Lamprey cartilage calcified in vitro, showing the same calcification pattern as in Euphanerops. Images in b and c are reproduced from ref. 4. CN, cell nest; CS, chondrocyte space; IM, intervening matrix; TM, territorial matrix. Scale bars, 50 µm.

Zhang et al. did something interesting, though: they questioned the assumption that lampreys lack collagen II, and went looking for it. They screened a library of lamprey sequences and isolated two forms of collagen II, Col2α1a and Col2α1b. The presence of a collagen homolog related to our collagen II tells us that the gene arose before the lamprey-gnathostome split, but the real question is whether lamprey Col2α1 is used in constructing their skeleton…and yes, it is. It's all over the developing branchial cartilaginous skeleton.

Col2α1a expression during lamprey development. Col2α1a is expressed throughout the branchial skeleton (boxed) and posterior to the oral cavity.

Col2α1a expression during lamprey development. Whole mount in situ hybridization of lamprey embryos at stage 23. Col2α1a expression is evident in the somites and within the dermatome and sclerotome.

It's also present in skeletal elements of the trunk.

I like this paper, because it doesn't just stop there. Lamprey have collagen, OK, but as I've said before, the really interesting stuff is what goes on in the interactions between genes—regulatory networks are the key targets of change in evolution. In vertebrates, we know that an important upstream regulator of cartilage formation is a gene called Sox9, which is directly involved in turning on the collagen II gene. Is Sox9 also present in lampreys?

Sure is, and it's expressed in the same areas that turn on collagen II and form the cartilaginous skeleton. At least a couple of pieces of the gene regulatory network for skeleton formation are present in the lamprey, suggesting that this might be another of those conserved kernels that define a central element of vertebrate anatomy.

Sox9 expression during lamprey development. (A and B) Expression of lamprey Sox9 in embryos at stages 23 (A) and 24 (B). (B Inset) Sagittal section through hindbrain and pharyngeal arches are shown. Arrows indicate Sox9 expression in streams of neural crest cells invading pharyngeal arches.

With sequence data in hand, the authors could also do comparative analyses and put together cladograms for the collagens and Sox molecules they identified. As predicted from evolutionary theory, the usual nested hierarchy emerges, with the lamprey as the most different.

Minimum evolution phylogeny for fibril A collagen proteins as obtained with JTT plus Γ distances (α=0.906). Numbers indicate bootstrap scores for each node, based on 1,000 replicates, whereas branch lengths are proportional to expected replacements per site. This tree is rooted by sea urchin ColP2α. Equally and unequally weighted MP, ML, and BP also place the two lamprey sequences together at the base of the Col2α1 clade, with bootstrap scores or a posterior probability of 93%, 77%, 71%, and 87%, respectively. In contrast to this consistent support for a Col2α1 grouping, this ME tree is unique among the different phylogenies in its placement of the root along the Col5 clade.

It's a pretty result that simplifies and clarifies our view of the evolution of the skeleton—lampreys aren't just a weird and unique group, but representative of an earlier pattern of skeletal organization. It also suggests that collagen is a unifying substance that ties all vertebrates together—that what evolved first in the early history of chordates was a collagenous skeleton.